Mediator has been first discovered in yeast (Saccharomyces cerevisiae), which contains 25 subunits, while approximately 31 and 34 subunits are found in mammals and plants, respectively (reviewed in Mathur et al., 2011; Allen and Taatjes, 2015; Samanta and Thakur, 2015). The Mediator complex is organized into four modules; the Head, Middle, Tail, together known as the core Mediator, and CYCLIN-DEPENDENT KINASE 8 (CDK8), which is reversibly associated with Mediator. In yeast and Metazoan species, CDK8 modulates transcription by phosphorylating TFs and influencing RNA Pol II-Mediator interactions (Poss et al., 2013; Tsai et al., 2013; Nemet et al., 2014). The Tail module of the Mediator complex interacts with the DNA-bound TFs while the Head module interacts with RNA Pol II and may also be involved in basal or activator-independent transcription. The Middle module provides the flexibility required for this large protein complex to exhibit necessary conformational changes in response to RNA Pol II binding (Karijolich and Hampsey, 2012).

Mediator can modulate RNA Pol II transcription by influencing the assembly of the pre-initiation complex (PIC), Pol II pausing, elongation and re-initiation and chromatin architecture (Allen and Taatjes, 2015). It is now well-established that the Mediator complex is involved in both transcriptional activation and repression. Indeed, Mediator can interact with both transcriptional activators and also repressors (Dolan and Chapple, 2017). In addition to its activator roles, the CDK8 kinase module is involved in repressing transcription by binding to the Tail module and blocking the interaction of Mediator with RNA Pol II (Conaway and Conaway, 2011).

In this review, diverse functions performed by MED25, one of the relatively well-conserved subunits of the Mediator complex, will be reviewed. Please see recent reviews for additional information on the plant Mediator complex (Buendía-Monreal and Gillmor, 2016; Yang et al., 2016; Dolan and Chapple, 2017).

The discovery of MED25 in plants was somewhat serendipitous. First, screening of mutagenized Arabidopsis populations identified the recessive pft1 (phytochrome and flowering time 1) mutant that showed a late flowering phenotype under suboptimal light conditions. This suggested that the mutated gene regulates shade-avoidance (rapid extension of stem and accelerated flowering in plants grown under dense canopies) (Cerdán and Chory, 2003). Subsequent genetic analyses placed PFT1 downstream from phyB and upstream from FLOWERING TIME (FT), a positive regulator of flowering. The cloning of the mutated gene in the pft1 mutant has revealed that PFT1 is a nuclear protein with similarity to some transcriptional activators from animals (Cerdán and Chory, 2003). However, it was not realized until the biochemical purification of the Mediator complex from plants that PFT1 encodes MED25 (Bäckström et al., 2007). As explained in detail below, since then, the med25 mutant has been independently identified in different forward screens (Zheng et al., 2006; Seguela-Arnaud et al., 2015) and a number of functions regulated by PFT1/MED25 characterized.

The yeast does not contain a MED25 homolog while a single gene encoding MED25 is found in plants (Kidd et al., 2011). The location of MED25 within the Mediator complex is unknown. However, the finding that MED25 interacts with a number of TFs (see below) and also with MED16, a Tail module subunit, suggests that MED25 is most likely associated with the Tail module of the Mediator complex (reviewed by Kidd et al., 2011; Yang et al., 2014).

The MED25 protein contains multiple domains that perform specific functions (Figure 1). The conserved von Willebrand Factor Type A (vWF-A) domain located at the amino terminus of MED25 is required for its interaction with the Mediator complex via the Tail subunit MED16 (Yang et al., 2014). The ACID (Activator Interacting Domain) region is required for the interaction of MED25 with transcriptional activators such as AP2/ERF and MYC TFs as well as suppressors such as JAZ proteins (Çevik et al., 2012; Chen et al., 2012; Zhang et al., 2015). VP16, a TF from the Herpes simplex virus with strong transcription activation activity in plants also binds to MED25-ACID through its TAD (Yang et al., 2004). MED25 also contains a conserved glutamine (Q) rich poly-track in its C-terminus, possibly involved in transcriptional activation (Cerdán and Chory, 2003; Bäckström et al., 2007; Elfving et al., 2011). The presence or absence of this Q-rich region or its length influences flowering-associated functions of MED25 (Rival et al., 2014). The alternative splicing of the Q-rich region of MED25, which generates two different isoforms, is evolutionary conserved and potentially contributes to diverse functions performed by this protein (Rival et al., 2014). As explained in detail below, MED25 does not directly interact with RNA pol II but activates or represses transcription by interacting with activators or repressors (Kidd et al., 2009; Yang et al., 2016; Dolan and Chapple, 2017).

The plant hormone JA regulates responses to biotic and abiotic stresses (Kazan, 2015). Under low hormone levels (i.e., no stress), JA responses are suppressed by JAZ proteins through their interactions with MYC TFs and the co-repressor TPL recruited either directly through JAZ repressor EAR (Ethylene-responsive element binding factor-associated amphiphilic repression) motifs (Kazan, 2006) or an accessory protein called NINJA, which also contains an EAR motif (Kazan and Manners, 2012, 2013; Wasternack and Song, 2016). JAs produced under stress conditions are sensed by the COI1-JAZ co-receptor complex. This leads to the destruction of JAZ repressors and the relief of repression on MYC TFs, which then regulate diverse JA responses (Kazan and Manners, 2013). However, until the discovery of MED25, how MYC TFs and other JA responsive TFs recruit the Mediator complex, had remained enigmatic (Kidd et al., 2009).

The roles of PFT1/MED25 in JA signaling was first identified in a genome-wide analysis of Arabidopsis TFs as a putative TF responsive to the fungal pathogen Alternaria brassicicola and methyl JA (McGrath et al., 2005). The Arabidopsis pft1/med25 mutant shows attenuated expression of JA responsive defense genes and altered resistance to necrotrophic pathogens, suggesting that MED25 controls an essential point within the JA pathway (Kidd et al., 2009). In addition, the bestatin resistant 6 (ber6) mutant, which shows sensitivity to root inhibition by bestatin, a strong chemical activator of the JA signaling pathway, was found to be allelic to med25 (Zheng et al., 2006).

The discovery of physical interactions between MED25 and various JA-associated TFs has provided further evidence about the role of MED25 and JA signaling. First, a high throughput yeast two hybrid (Y2H) screening of 1589 Arabidopsis TFs from 62 different families using the conserved ACID region of MED25 as bait has identified eight MED25-interacting TFs, including four AP2/ERFs belonging to Group IX of the AP2/ERF family (Ou et al., 2011) implicated in JA signaling and pathogen resistance (McGrath et al., 2005). Five of these TFs could bind to the GCC box promoter element (Ou et al., 2011), which confers JA responsiveness (Brown et al., 2003). Subsequently, other JA responsive TFs, including ORA59 (OCTADECANOID-RESPONSIVE ARABIDOPSIS AP2/ERF59) and ERF1 (ETHYLENE RESPONSE FACTOR1) as well as the master regulator MYC2 and related MYC TFs MYC3 and MYC4 have been shown to be interacting with MED25 (Çevik et al., 2012; Chen et al., 2012). The conserved EDLL motif (Tiwari et al., 2012) present in the carboxyl-terminal region of AP2/ERFs such as ERF1, ERF15, and TDR1/ERF98, seems to be important for the interaction of these AP2/ERFs with the ACID region of MED25 (Çevik et al., 2012). MED25 seems to genetically act downstream from ORA59 and ERF1 and is required for JA-mediated activation of the pathogen defense genes such as PDF1.2. In addition, MED25 is required for MYC2-mediated activation of insect defense gene VSP and MYC2-dependent suppression of PDF1.2 during JA signaling (Kidd et al., 2009, 2010; Çevik et al., 2012).

Similarly to MYC TFs, JAZ repressors such as JAZ9 interact with MED25 through their conserved JAS (JA Associated) domains (Zhang et al., 2015). So, it appears that both repressors (e.g., JAZs) and activators (e.g., MYCs) compete for binding to MED25 during JA signaling. Therefore, interfering with the MYC-MED25 interaction seems to be one of the mechanisms employed by JAZ proteins to suppress JA-responsive gene expression (Zhang et al., 2015, 2017).

MED25 also interacts with CDK8 (Zhu et al., 2014; Figure 2 and Table 1). Interestingly, similarly to MED25, CDK8 positively regulates JA-responsive defense gene expression (e.g., PDF1.2) and resistance to A. brassicicola. However, in contrast to MED25, which positively regulates resistance to Botrytis cinerea (Kidd et al., 2009), CDK8 negatively regulates resistance to this pathogen (Zhu et al., 2014). Furthermore, both MED25 and CDK8 interact with WIN1 (WAX INDUCER 1) (Zhu et al., 2014), an AP2/ERF TF that regulates cuticle development and plant defense (Sela et al., 2013). Together, these findings suggest that MED25 and CDK8 perform overlapping functions within the JA signaling.

More recently, additional roles for the Mediator complex and MED25 in auxin signaling has been identified. It appears that MED13/MAB2/GCT (MACCHI-BOU2/GRAND CENTRAL), which is part of the CDK8 kinase domain of the Mediator complex, regulates auxin signaling through the association with ARF7 and ARF19 and Aux/IAA14 (Ito et al., 2016). Under low auxin levels, Aux/IAA14 represses ARF7 and ARF9 by forming a complex with the co-repressor TPL and the Mediator subunit MED13 to inhibit the interaction between the core Mediator complex and RNA Pol II (Figure 2). Under high auxin levels, Aux/IAA14 is degraded and this results in the dissociation of TPL and MED13 from the complex. ARF7 and ARF9, by binding to the auxin responsive element (AuxE) found in the LBD16 (LOB DOMAIN 16) promoter, interact with MED25 to recruit the core Mediator complex and RNA Pol II (Ito et al., 2016).

Jasmonate and auxin signaling share many mechanistic similarities (reviewed by Pérez and Goossens, 2013). Firstly, similarly to COI1, the F-box protein TIR1 (TRANSPORT INHIBITOR RESPONSE 1) acts as an auxin receptor. Secondly, similarly to the role of JAZ proteins acting as repressors of JA signaling, under low hormone levels, auxin responses are suppressed by Aux/IAA repressors, which recruit the co-repressor TOPLESS (TPL) through their EAR domains. Thirdly, similarly to the activation of JA signaling through the COI1-mediated degradation of JAZ repressors, the reception of auxin leads to the TIR1-mediated degradation of repressors and the activation of the downstream signaling pathway by ARFs (AUXIN RESPONSE FACTORS) acting as positive regulators of auxin signaling through (Kazan and Manners, 2009).

As stated above, MED25 interacts with both activators (MYCs and ERFs) and repressors (JAZs) of JA signaling (Çevik et al., 2012). Therefore, the discovery that MED25 interacts with both activators and repressors of IAA signaling (Ito et al., 2016; Figure 2), suggests that MED25 performs mechanistically generic roles in both signaling pathways. Furthermore, as stated above, MED25 interacts with CDK8 during the regulation of JA signaling (Zhu et al., 2014), and with MED13, a component of the CDK8 kinase, during the regulation of auxin signaling (Ito et al., 2016).

Ethylene signaling is involved in the regulation of biotic and abiotic stresses (Kazan, 2015), including iron (Fe) deficiency in plants (Lucena et al., 2015). The link between MED25 and ethylene signaling became evident when the involvement of MED25 in Fe deficiency responses was discovered (Yang et al., 2014). The med25 mutant shows increased sensitivity to Fe deficiency, suggesting that MED25 is a positive regulator of these responses. MED25 interacts with both MED16, another Mediator subunit involved in regulating Fe deficiency responses, and EIN3 (ETHYLENE INSENSITIVE 3) and EIL1 (ETHYLENE INSENSITIVE 3 LIKE), known regulators of both ethylene signaling and Fe deficiency responses (Brumbarova et al., 2015). EIN3 and EIL1 regulate the expression of bHLH TF FIT1 (Fe DEFICIENCY INDUCED TF 1), which in turn regulates the expression of Fe-stress responsive genes IRT1 (IRON-REGULATED TRANSPORTER 1) and FRO2 (FERRIC REDUCTION OXIDASE 2) (Yang et al., 2014; Zhang et al., 2014).

The med25 mutant shows increased sensitivity to the inhibition of germination and seedling growth by ABA, suggesting a role for MED25 in regulating ABA signaling. In addition, Em6, and RAB-RELATED GENE18 (RAB18) show increased ABA responsiveness in the med25 mutant, suggesting that MED25 is a negative regulator of ABA responses during seed germination and early seedling growth (Chen et al., 2012). Supporting this notion, MED25, physically interacts with ABI5 (ABA-INSENSITIVE5), a basic Leu zipper TF that regulates ABA signaling during seed maturation and germination (Chen et al., 2012). The positive and negative regulation of JA and ABA signaling through interaction with MYC and ERF TFs (Çevik et al., 2012; Chen et al., 2012) and ABI5 (Chen et al., 2012), respectively, also suggests that MED25 may have a role in modulating the JA-ABA crosstalk (Anderson et al., 2004; Liu and Avramova, 2016).

Given its role as a positive regulator of JA signaling, a role for MED25 in plant disease resistance is also expected. The JA signaling pathway provides susceptibility to the root infecting pathogen Fusarium oxysporum and the bacterial pathogen Pseudomonas syringae DC3000 but is required for resistance to leaf infecting necrotrophic pathogens (Thatcher et al., 2009 reviewed by Kazan and Lyons, 2014). Mutant analyses show MED25 differentially regulates resistance to different pathogens. The med25 mutant shows increased resistance F. oxysporum (Kidd et al., 2009) and P. syringae (Chen et al., 2012) but increased susceptibility to leaf infecting necrotrophs A. brassicicola, B. cinerea, and Sclerotinia sclerotiorum (Kidd et al., 2009; Wang et al., 2015). This finding is consistent with the known function of MED25 in JA signaling. There is also evidence that MED25 regulates the interaction of plant roots with soil microbes (Carvalhais et al., 2015).

The function of MED25 is not restricted to biotic stress signaling. MED25 physically interacts with abiotic-stress associated TFs, DREB2A, ZFHD1, and MYB-Like (Elfving et al., 2011; Table 1). Upon binding of DREB2A to MED25, conformational alterations occur in the overall Mediator structure (Blomberg et al., 2012). The Arabidopsis mutants for these three TF genes as well as the med25 mutant show increased sensitivity to salt stress, suggesting that MED25 is a positive regulator of salt tolerance. In contrast, the med25 mutant show increased drought tolerance (Elfving et al., 2011). Therefore, MED25 seems to differentially regulate different abiotic stress responses.

So far, the function of MED25 has been mainly investigated in Arabidopsis. However, emerging evidence indicates that at least some MED25-mediated functions are conserved between dicots and monocots. For instance, TaMED25, the wheat ortholog of MED25, complements the defense and developmental phenotypes of the Arabidopsis med25 mutant (Kidd et al., 2009), suggesting that TaMED25 performs functionally similar roles to MED25. Wheat plants with one or two subgenome copies or homoeologs of deleted TaMED25 were successfully isolated from a heavy ion bombarded hexaploid wheat population (Fitzgerald et al., 2010). However, despite extensive efforts, complete knockouts with deleted A, B, and D homoeologs of TaMED25 could not be isolated. This seems to suggest an essential function for TaMED25 in wheat (Fitzgerald et al., 2015). However, wheat and barley plants with reduced expression of TaMED25 and HvMED25 by virus-induced gene silencing show increased resistance to the powdery mildew pathogen, Blumeria graminis f. sp. tritici (Liu et al., 2016). Therefore, similarly to the Arabidopsis MED25 conferring susceptibility to F. oxysporum and P. syringae, TaMED25 acts as a powdery mildew susceptibility gene in wheat and barley.

Moreover, similarly to Arabidopsis MED25, which interacts with EIN/EIL1, TaMED25 interacts with TaEIL1, the wheat homolog of Arabidopsis EIL1 (Liu et al., 2016). Although, if TaMED25 and TaEIL1 are involved in regulating Fe-deficiency responses is unknown, the interaction between TaMED25 and TaEIL1 certainly supports a role for TaMED25 in ethylene signaling in wheat. Future studies may reveal additional MED25-regulated functions that are conserved in plants.

As reviewed here, it is becoming evident that MED25 is master regulator of transcriptional responses to multiple endogenous and exogenous signals (Balderas-Hernández et al., 2013). Given that there are 1000s of TFs in plants but only a limited number of Mediator subunits, particularly those that reside in the TF-interacting Tail module of the Mediator complex, it is probably not surprising that many different TFs may interact with the same Mediator subunit to initiate gene expression. For this reason, different Mediator subunits seem to perform both diverse and overlapping roles. For instance, MED16, a Tail module subunit, regulates SA and JA signaling (Wathugala et al., 2012), Fe homeostasis (Yang et al., 2014), cold-acclimation (Hemsley et al., 2014) and regulation of transcriptional responses to cell wall defects (Sorek et al., 2015). The Middle module subunit MED8 regulates flowering time (Kidd et al., 2009), cell wall development (Seguela-Arnaud et al., 2015) as well as defense (Kidd et al., 2009; An and Mou, 2013). MED18, located in the Head module of the Mediator complex, regulates flower development (Zheng et al., 2013) and plant defense (Lai et al., 2014; Liao et al., 2016; Fallath et al., 2017).

In addition, a few other Mediator complex subunits seem to regulate overlapping functions as MED25. For instance, similarly to MED25, MED16 recruits RNA polymerase II to the promoters ORA59 and PDF1.2. MED16 interacts with WRKY33, a defense associated TF and is required for pathogen (B. cinerea)-mediated induction of PDF1.2 (Wang et al., 2015, 2016). However, in contrast to MED25, which is required for both insect/wound and pathogen response branches of the JA signaling pathway, MED16 seems to be specific for pathogen defense (Wang et al., 2015). Given that the expression of ORA59, ERF1 and PDF1.2, are influenced by both JA and ethylene pathways, it is tempting to speculate that MED25 is mostly involved in JA whereas MED16 is in the ET pathway. Similarly to MED25, MED8 regulates disease resistance, flowering time (Kidd et al., 2009) and organ size (Xu and Li, 2012).