|
Possible biological role of the AGL15–SAP18 interaction
AGL15 accumulates in the nuclei of cells in the embryo beginning very early in development (by the eight-cell stage for Arabidopsis) and remains at relatively high levels into maturation, decreasing during desiccation (Perry et al., 1996). AGL15 is expressed after germination in Arabidopsis, in the vegetative shoot apical meristem, but not at the transition to reproductive development, although it is once again expressed at the bases of cauline leaves and floral organs (Fernandez et al., 2000). The level of expression in these tissues is generally at least 10-fold lower than that found in the embryo (Fernandez et al., 2000; Heck et al., 1995). One possible role for AGL15 is in regulating developmental transitions during the plant life cycle. The expression data supports this hypothesis. Ectopic overexpression of AGL15 via the 35S promoter causes delays at many points in the life cycle, including late flowering, a longer period for seeds to mature, and delayed senescence and abscission (Fernandez et al., 2000). Additionally, 35S:AGL15 transgenic lines show increased capability for cultured zygotic embryos, and for apical meristem tissue in a liquid culture system, to form somatic embryos (Harding et al., 2003). One possible explanation is that these tissues are delayed in exiting embryonic programs. Recently, a double mutant agl15 agl18 was documented to flower early under short-day conditions (Adamczyk et al., 2007). Epigenetic control is one means by which developmental transitions are regulated (for a review, see Berger and Gaudin, 2003Henderson and Dean, 2004; Boss et al., 2004). Therefore, interaction of AGL15 with chromatin-modifying proteins is very intriguing.
Phenotypes reported as being associated with HDA19 overexpression include decreased acetylation of histones, short siliques, abnormal cotyledons and leaves, late flowering and decreased fertility (Zhou et al., 2005). Some of these phenotypes, notably late flowering, have also been documented in plants that constitutively express AGL15 (Fernandez et al., 2000). However, both antisense HDA19 transgenic and athd-t1 homozygous lines have also been reported to cause a delay in flowering, in some cases observable only when plants are grown under long-day conditions (Tian and Chen, 2001; Tian et al., 2003; Wu et al., 2000). In agreement with the latter observation, treatment of Arabidopsis plants with the HDAC inhibitor trichostatin A also induces a late flowering effect (Wu et al., 2000). Also consistent with these results is the observation that a loss-of-function allele of fld, a component of an HDAC complex, results in hyperacetylation of histones associated with FLC, increased transcript abundance of FLC and a late flowering phenotype (He et al., 2003). However, loss-of-function of the histone acetyltransferase AtHAC1 also results in increased transcript abundance of FLC and a delay in flowering, perhaps caused by the regulation of components upstream of FLC (Deng et al., 2007).
Despite the fact that SAP18 is a single-copy gene, highly conserved even between plants and animals, plants homozygous for null alleles of sap18 are viable and do not exhibit any obvious phenotype until the plant is stressed (Song and Galbraith, 2006). Similarly Drosophila homozygous for null sap18 alleles are able to complete their life cycle (Singh et al., 2005). AGL15 is able to interact directly with at least one other member of the SIN3/HDAC1 complex (HDA19), suggesting that SAP18 is not essential in recruiting a functional HDAC1 complex to AGL15 bound promoters. SAP18 is broadly expressed, but transcript levels increase in response to NaCl, drought and cold stress (Song and Galbraith, 2006). Therefore regulation of targets involved in the stress response, such at CBF2, are intriguing. A number of sequence-specific DNA binding proteins are likely to serve as platforms, targeting the SIN3/HDAC1 complex to these various target genes. We propose that AGL15 functions as one of these platforms, thus recruiting SAP18 and ultimately other components of the SIN3/HDAC complex to a subset of target genes.
|
