Cullin-RING Ubiquitin Ligase Family in Plant Abiotic Stress Pathways

Authors

  • Liquan Guo,

    1. Department of Molecular, Cellular and Developmental Biology, Yale University, New Haven 06520, Connecticut, USA
    2. State Environmental Protection Key Laboratory of Wetland Ecology and Vegetation Restoration, Northeast Normal University, Changchun 130024, China
    3. Department of Biological Engineering, Jilin Business and Technology College, Changchun 130062, China
    Search for more papers by this author
  • Cynthia D. Nezames,

    1. Department of Molecular, Cellular and Developmental Biology, Yale University, New Haven 06520, Connecticut, USA
    Search for more papers by this author
  • Lianxi Sheng,

    1. State Environmental Protection Key Laboratory of Wetland Ecology and Vegetation Restoration, Northeast Normal University, Changchun 130024, China
    Search for more papers by this author
  • Xingwang Deng,

    1. Department of Molecular, Cellular and Developmental Biology, Yale University, New Haven 06520, Connecticut, USA
    Search for more papers by this author
  • Ning Wei

    Corresponding author
    1. Department of Molecular, Cellular and Developmental Biology, Yale University, New Haven 06520, Connecticut, USA
    Search for more papers by this author

Tel: +1 203 432 3897; Fax: +1 203 432 5726; E-mail: ning.wei@yale.edu

Abstract

The ubiquitin-proteasome system is a key mechanism that plants use to generate adaptive responses in coping with various environmental stresses. Cullin-RING (CRL) complexes represent a predominant group of ubiquitin E3 ligases in this system. In this review, we focus on the CRL E3s that have been implicated in abiotic stress signaling pathways in Arabidopsis. By comparing and analyzing these cases, we hope to gain a better understanding on how CRL complexes work under various settings in an attempt to decipher the clues about the regulatory mechanism of CRL E3s.

  • image

[ Ning Wei (Corresponding author)]

Introduction

The ubiquitin-proteasome system (UPS) is a widely used mechanism in essentially all developmental processes and physiological responses of plants, including acclimation to abiotic environmental factors. The UPS is initiated with a conserved enzymatic cascade, consisting of E1, E2, and E3 enzymes, that are dedicated to attaching ubiquitin (Ub) to selected proteins in eukaryotic cells (Hershko and Ciechanover 1998). Proteins tagged with ubiquitin chains are typically destined for degradation by the 26S proteasome, although the ubiquitin modification may also have other signaling functions (Hershko and Ciechanover 1998; Hua and Vierstra 2011). The substrate selectivity of ubiquitin attachment is determined by the E3 ligase, which binds to the specific substrate and catalyzes the transfer and the covalent attachment of the ubiquitin chain to the target protein. Among the over 1,500 UbE3 ligases in Arabidopsis, the Cullin-RING ubiquitin ligases (CRLs) represent the largest E3 ligase family (Hotton and Callis 2008; Hua and Vierstra 2011).

Abiotic stresses such as high and low temperature, drought, flooding, salinity, UV and strong light irradiation, mineral deficiency, acidic soil, air pollutants, chemical toxins, and mechanical stress can adversely affect plant growth and crop production (Hirayama and Shinozaki 2010; Walley and Dehesh 2010; Huang et al. 2012). In contrast to biotic signals, which are perceived by specific receptors in plants, most abiotic stress signals are often initially sensed via perturbation at the plasma membrane. Signals from the plasma membrane then generate second messengers such as calcium, reactive oxygen species (ROS), and inositol phosphates. In the case of DNA damage, the DNA lesions induced by UV or other genotoxic insults can also be the sites of signal initiation. In addition, abiotic stresses invoke signaling activities of plant hormones to modify the growth and potentiate adaptive responses. For example, abiotic stressors often induce the production of abscisic acid (ABA) and ethylene, and cross-talk with other hormonal pathways such as those of salicylic acid (SA), jasmonic acid (JA) and auxin, to bring about a gene expression profile that underlies stress-specific as well as general adaptive responses in plants (Hirayama and Shinozaki 2010; Miller et al. 2010; Huang et al. 2012). Owing to the sessile nature of plants, adequate responses to their environment are imperative for the survival and success of the plant species.

The UPS plays an active role in both biotic and abiotic responses of plants (Dreher and Callis 2007; Trujillo and Shirasu 2010; Hua and Vierstra 2011; Stratmann and Gusmaroli 2012). In this review, we focus on the involvement of CRL families of ubiquitin E3s in the abiotic responses in plants from the perspective of functionality and mechanisms of CRL complexes. Recent reviews on regulation by general ubiquitin E3 ligases including RING or U-box E3s in biotic (Trujillo and Shirasu 2010; Stratmann and Gusmaroli 2012) and abiotic (Lee and Kim 2011; Lyzenga and Stone 2012) stresses can be found elsewhere.

Cullin-RING Ubiquitin Ligases, Organization and Diversity

CRLs are composed of a cullin protein as a scaffold, a small RING protein RBX1 that stably binds to cullin's carboxyl terminus, and a substrate-recognizing module that dynamically binds to cullin's amino terminal region (Figure 1, Zheng et al. 2002; Hua and Vierstra 2011). In Arabidopsis, three main cullin types, CUL1/CUL2, CUL3a/b, and CUL4 have been studied in depth. Each of the cullins assembles a distinct CRL ligase that ubiquitinates a specific group of targets (Figure 1, Hotton and Callis 2008; Hua and Vierstra 2011). While RBX1 can form a stable complex with each of the three different cullins, a specific set of substrate receptors binds to each cullin system. As illustrated in Figure 1, cullin 1 (CUL1) assembles into the SCF (skp1-cullin-F-box), or CRL1 complex, with ASK proteins as its adaptor and F-box protein as the substrate receptor. Cullin 3 is encoded by a two-member gene family CUL3a and CUL3b in Arabidopsis, and it directly binds its substrate receptors, which are BTB/POZ-domain proteins, to form CRL3. Cullin 4 (CUL4) uses DDB1 (a/b) and DWD proteins as adaptors and substrate receptors, respectively, to assemble CRL4 E3 complexes (Hotton and Callis 2008; Hua and Vierstra 2011). For example, The F-box protein TIR1 is a substrate receptor for SCFTIR1 (or CRL1TIR1) (Gray et al. 2001). In the presence of auxin, SCFTIR1binds to Aux/IAA transcriptional repressors via TIR1 and mediates their degradation, leading to activation of auxin-responsive gene expression (Gray et al. 2001; Mockaitis and Estelle 2008).

Figure 1.

Diagram of Cullin-RING ligase (CRL) subunit organization and its population in Arabidopsis.

The substrate receptor (F-box proteins, BTB proteins, and DWD proteins) is responsible for substrate specificity and defines the functional identity of each CRL E3s. The large collection of these substrate receptors is the reason that CRL E3s is the largest family of ubiquitin ligases in Arabidopsis or human cells (Hua and Vierstra 2011). In Arabidopsis, there are about 700 different F-box substrate receptors that assemble through one of 19 ASK adaptor proteins with CUL1-RBX1 into CRL1 (or SCF) (Gagne et al. 2002; Hua and Vierstra 2011). Similarly, there are around 80 BTB/POZ-domain substrate receptor proteins that can assemble directly with CUL3-RBX1 to form CRL3 complexes (Figueroa et al. 2005; Gingerich et al. 2005). There are more than 80 DWD domain proteins that can assemble with DDB1a or DDB1b with CUL4-RBX1 to form CRL4 ligase complexes (Lee et al. 2008). In some cases, the WDxR motif itself, without the entire 16-amino acid DWD box in which the WDxR motif is embedded, is sufficient to mediate interaction with DDB1. This would potentially add additional substrate receptors to the number of total CRL4 complexes (Zhang et al. 2008a). Together, the CRL family of ubiquitin ligases constitutes well above 800 individual E3s in Arabidopsis, which likely regulate thousands of targets in the cell (Hua and Vierstra 2011, see Figure 1).

Several Arabidopsis CRLs, as identified by their substrate receptors (F-box proteins, BTB proteins and DWD proteins), have been shown to participate in various abiotic responses (Table 1). In several cases, the target of the CRL has been identified. For example, CRL4DWA1/2 has been found to target the transcription factor ABA Insensitive 5 (ABI5) for ubiquitination and degradation (Lee et al. 2010a). However, in many more cases, specific targets of the E3 have not been clearly identified at this time. Here, we summarize these case studies in an attempt to comprehend the variations and workings of CRL E3 complexes in the abiotic response pathways in plants.

Table 1.  Cases of Arabidopsis Cullin-RING ligases (CRLs) in abiotic stress
CRL coreAdaptorSubstrate receptorFunctional remarks and putative targetsReferences
CUL1-RBX1ASK (1–19) F-box proteins   
CUL1ASK14 DOR Inhibitor for drought- or ABA-induced stomatal closure Zhang et al. 2008b
 ASK1 AtTLP9 Required for ABA-dependent inhibition of seed germination Lai et al. 2004
   AtTLP3 Subcellular translocation responsive to salt, osmolarity, H2O2 Reitz et al. 2012
 ASK1,2,4,11,18 EDL3 Positive regulator of ABA, affect sensitivity to osmolarity Koops et al. 2011
 ASK1,2,4,11,14 AtFBP7 Required for protein synthesis during temperature stress Irina et al. 2007
   AFB4 Auxin receptor. Required for high-temperature-induced hypocotyl elongation Greenham et al. 2011
   atFBX2 Negatively regulate P(i) starvation response. Interact with transcription factor BHLH32 Chen et al. 2008
   TIR1 Auxin receptor. Required for P(i) starvation induced root expansion Perez-Torres et al. 2008
CUL3-RBX1 BTB/POZ    
(CUL3a, CUL3b) BPMs  Expression induced by stress. Regulate ABA response and stomatal behavior. Target ATHB6 transcription factor Lechner et al. 2011; Weber & Hellmann 2009; Weber et al. 2005
  ARIA  Bind transcription factor ABF2 and ADAP. Positively regulate response to ABA and osmolarity Kim et al. 2004; Lee et al. 2009
  BT2  Activator of telomerase in response to nutrients, stress and hormones Mandadi et al. 2009
CUL4-RBX1DDB1 DWD/WDxR   
CUL4  DWA1, DWA2 Regulate sensitivity to ABA and salt. Target transcription factor ABI5. Mutant drought tolerantLee JH et al. 2010a
   DWA3 Negative regulator of ABA Lee JH et al. 2011
   PRL1 Integrator of stress, glucose, and hormonal response. Targeting Akin10 Lee JH et al. 2008
   DRS1 Required for ABA-dependent drought responseLee I et al. 2010b
   MSI1 Maintaining genome integrity Dumbliauskas et al. 2011
   TANMEI/ALT2 Mediate aluminum-induced growth arrest via the ATR cell cycle checkpoint mechanism Nezames et al. 2012
   DDB2 Repair UV-induced DNA lesions. Tolerance to UV-B and UV-C Molinier et al. 2008
   CSA1 Repair UV-induced DNA lesions Biedermann and Hellmann 2010; Zhang et al. 2010

CRL4, a Key Player in DNA Repair and DNA Damage Responses to Genotoxic Stress

Ultraviolet light (UV) irradiation can cause covalent crosslinks between neighboring nucleotides of the DNA, resulting in cyclobutane pyrimidine dimers (CPD) and (6–4) pyrimidine-pyrimidone photoproducts (6–4PP). These DNA lesions negatively affect both transcription and DNA replication, and can also have mutagenic consequences. DDB1 and DDB2 (Damaged DNA Binding protein 1 and 2) were initially isolated from mammalian cells as a heterodimeric protein complex that binds to the UV-induced DNA lesions (Chu and Chang 1988). Later, it was found that the DDB1-DDB2 complex assembles with CUL4-RBX1 to form the ubiquitin E3 complex, CRL4DDB2, in animals as well as in plants (Jackson and Xiong 2009). As in animals, the plant CRL4DDB2 complex is critical in repairing UV-induced DNA lesions, and genetic mutations in CUL4, DDB1a, and DDB2 all led to hypersensitivity to UV-C (Koga et al. 2006; Jean et al. 2008). Jean et al. (2008) further revealed some interesting mechanistic parallels between plant and human CRL4DDB2. First, UV irradiation stimulates rapid translocation of DDB1 from the cytoplasm to the nucleus in both plant and human cells. Second, DDB2, a DWD-containing protein, undergoes degradation after UV exposure in a CRL4DDB2-dependent manner in both plant and animal cells. In addition to DDB2, CSA1 (Cockayne syndrome factor A) has also been implicated in repairing UV-damaged DNA in conjunction with CUL4. CSA1 is another DWD protein, and it functions as a CRL4CSA1 E3 complex in mediating DNA repair, similar to its mammalian counterpart (Zhang et al. 2010; Biedermann and Hellmann 2010).

DNA damage or genotoxic stress, resulting either from radiation or from toxic chemicals, can also cause cell cycle arrest, and CRL4 plays a key role in this response. Studies have shown that in this response, CRL4 appears to work closely with Ataxia-telangiectasia and Rad3-related (ATR), a critical cell cycle checkpoint protein kinase. In the case of the UV damage pathway, ATR is required for UV-induced nuclear translocation of DDB1 as well as for UV-induced turnover of DDB2 (Molinier et al. 2008; Jean et al. 2011). ATR also plays a crucial function in the aluminum (Al)-induced stress pathway (Rounds and Larsen 2008). In acidic soils, aluminum assumes a highly-phytotoxic trivalent cation, Al3+, which inhibits root growth. Although Al toxicity has been speculated to be complex, increasing evidence suggests that high concentrations of Al3+ cause an as-yet undefined mechanism of Al-dependent DNA damage. Accumulation of Al-dependent DNA damage triggers an active process that arrests the cell cycle at the root tip, resulting in root growth inhibition (Rounds and Larsen 2008; Nezames et al. 2012). Nezames et al. (2012) reported that TANMEI/ALT2 (Aluminum tolerance 2), a DWD protein with high homology to CSA, mediates this process in Arabidopsis, likely as the CRL4ALT2 complex. Moreover, loss of either ATR or ALT2 results in hypersensitivity to DNA crosslinking agents but tolerance to Al. Again, Al-induced growth arrest by TANMEI/ALT2 requires ATR, suggesting that an ATR-mediated checkpoint underlies the cell cycle arrest induced by high concentrations of aluminum (Nezames et al. 2012).

DDB1 (a/b), the adaptor protein that recruits DWD substrate receptor proteins to CUL4-RBX1, has broad pleiotropic functions, in contrast to the individual DWD proteins such as DDB2, CSA1, or ALT2 which specifically participate in DNA repair or Al tolerance, respectively. For example, DDB1 is also linked to ABA signaling and drought response by recruiting DWA proteins to CRL4 (described in the following section), and is involved in photomorphogenesis via associating with the WDxR motif-containing COP1-SPA complex (Chen et al. 2010). Besides its function in recruiting DWD/WDxR proteins, DDB1 also forms a stable complex called CDD that contains, in addition to DDB1, COP10 E2 variant and DET1, which does not contain a DWD domain (Yanagawa et al. 2004). As an interesting note, the animal counterpart of the CDD complex is the DDD-E2 complex, which consists of DDB1, DET1, DDA1, and a UBE2E family of E2 enzyme (Pick et al. 2007). DET1 is another multi-functional protein that is involved in photomorphogenesis and circadian clock regulation of flowering (Lau et al. 2011). Most relevant to this review is the report by Castells et al. (2011), in which it was shown that DET1 and DDB2 appear to work closely in DNA repair, because an appropriate dosage of DET1 protein is necessary for the efficient removal of UV photoproducts and UV-induced degradation of DDB2. Also, the DET1 protein is degraded concomitantly with DDB2 upon UV irradiation in a CUL4-dependent manner (Castells et al. 2011).

Similar to animal and fungal systems, the conserved functions of plant CRL4s appear to be in maintaining genome stability, chromatin dynamics, gene expression and parental imprinting during normal plant development, as well as during genotoxic stress (Molinier et al. 2008; Zhang et al. 2008a; Dumbliauskas et al. 2011; Pazhouhandeh et al. 2011).

CRLs are Part of ABA Signaling in Drought and Osmotic Stresses

In response to abiotic stresses, plants often invoke their hormonal signaling network to cope with and adapt to the changing environment. Among various plant hormones, abscisic acid (ABA) has a central role in the plant abiotic stress response, as ABA signaling appears to have been evolved for organisms (including but not limited to plants) to cope with environmental insults such as drought or dehydration, high salinity, and cold temperatures (Chinnusamy et al. 2008; Hirayama and Shinozaki 2010; Hauser et al. 2011). Abiotic stress generally induces the production of ABA, which also crosstalks with other plant hormonal pathways such as SA, JA, ethylene, and auxin to regulate various aspects of plant growth and development from seed germination to flowering and leaf senescence (Antoni et al. 2011; Huang et al. 2012). These signaling activities enable the plants to activate ABA-dependent and ABA-independent gene expression profiles that bring about both specific responses such as drought-induced stomatal closure (Zhang et al. 2008b; Lechner et al. 2011), and general responses such as stress-induced increase of reactive oxygen species (ROS) activities (Miller et al. 2010). Numerous abiotic responsive transcription factors (Hirayama and Shinozaki 2010; Walley and Dehesh 2010; Huang et al. 2011) and their ubiquitin E3 regulars (Lyzenga and Stone 2012) have been identified, although information specifically regarding CRL type E3s is still emerging. All three CRL families, SCF (CRL1), CRL3, and CRL4, are involved in regulating ABA signaling (Table 1), probably owing to the complicated nature of the network and the variation of the responses. Here, we will not attempt to describe all the signaling cascades, but will instead focus on those case studies involving CRLs in the responses.

SCF (CRL1)

A few F-box proteins have been shown to be involved in ABA signaling, one of which is DOR, a regulator of stomatal aperture (Zhang et al. 2008b). Drought stress stimulates the production of ABA, which induces a rapid closing of stomata to prevent water loss by transpiration. DOR was isolated as a negative regulator of ABA signaling, and it functions to inhibit stomatal closure. DOR is strongly expressed in guard cells, but its expression is suppressed by ABA treatment (Zhang et al. 2008b). The dor null mutant is hypersensitive to ABA in stomatal closing, and is substantially more tolerant to drought, while plants overexpressing DOR are more susceptible to the drought stress. Zhang et al. (2008b) further demonstrated that DOR binds to ASK1 and CUL1, forming a bona fide SCFDOR (or CRL1DOR) complex in Arabidopsis (Zhang et al. 2008b). Affymetrix GeneChip analysis suggested that DOR likely regulates ABA biosynthesis under drought stress.

A family of F-box proteins in Arabidopsis is known as Tubby-like proteins (AtTLPs) (Lai et al. 2004).The Tubby domain is highly conserved between mammals and plants, but the 11AtTLP proteins in Arabidopsis distinguish themselves from those of mouse in having an F-box at the N-terminal region (except for AtTLP8) (Lai et al. 2004). Among them, AtTLP9 has been shown to interact with the ASK1 CRL1 adaptor, and to positively regulate ABA-mediated inhibition of seed germination (Lai et al. 2004). In addition, AtTLP2 and AtTLP7 exhibit altered expression in the abi1 mutant (Lai et al. 2004), and AtTLP3 displays subcellular relocation in response to ABA treatment, indicating that these AtTLP genes are most likely involved in the ABA pathway.

Another multi-functional F-box protein is EDL3. EDL3 gene expression is rapidly induced under osmotic stress, high salinity, and upon application of ABA (Koops et al. 2011). Functionally, EDL3 has been shown to positively regulate ABA signaling in the regulation of seed germination, root growth, seedling greening, and transition to flowering (Koops et al. 2011). EDL3 was shown to interact with ASK proteins, thus presumably forming a functional SCFEDL3 ubiquitin E3 complex (Koops et al. 2011).

CRL3

Stomatal closure in response to drought is one of the ABA-mediated events in which multiple CRLs are involved. In addition to SCFDOR, Arabidopsis encodes half a dozen BTB/POZ-MATH proteins (BPM) 1–6 proteins, or MATH-BTB type of substrate receptors, that can be assembled with CUL3 and CUL3b to form CRL3BPME3 complexes (Weber et al. 2005; Weber and Hellmann 2009). In contrast to CRL1DOR, CRL3BPM positively regulates ABA-induced stomatal closure in guard cells. The BPMs interact with members of the ERF/AP2 family of transcription factors through the MATH domain (Weber and Hellmann 2009). They also bind and trigger ubiquitination and degradation of HD-ZIP class of transcription factors such as ATHB6, a negative regulator of the ABA signaling pathway (Lechner et al. 2011). In short, the overexpression of BPM proteins promotes ATHB6 protein degradation. Silencing BPM or overexpressing ATHB6 inhibits ABA- (or drought-) induced stomatal closure, which can result in greater transpirational water loss and a decline in plant growth (Lechner et al. 2011). Lechner et al. (2011) convincingly showed that CRL3BPM are critical positive regulators of ABA signaling by modulating the activity and turnover of transcription factors.

Another positive regulator of ABA signaling is Aria, which contains a BTB domain as well as an Armadillo (arm) repeat (Kim et al. 2004). ABA sensitivity in seed germination is enhanced by Aria overexpression and impaired by the knockout mutation. During seedling growth, Aria overexpression enhanced salt tolerance. Aria interacts with abZIP transcription factor ABF2 (Kim et al. 2004) and an AP-2 class transcription factor ADAP (Lee et al. 2009), both of which are positive regulators of ABA signaling. It is unclear whether Aria affects ABF2 and ADAP protein stability. However, based on phenotype and gene expression analyses of overexpression lines and loss-of-function mutants, it appears that Aria functions to promote the ABF2 and ADAP activities rather than inhibiting them (Kim et al. 2004; Lee et al. 2009). This function is different from the mechanism of BPMs, which target the transcription factor ATHB6 for degradation (Lechner et al. 2011).

CRL4

Although CRL4 is best known for its role in maintaining genome stability, it is also involved in ABA signaling as indicated by characterization of DWA (DWD hypersensitive to ABA) proteins. Lee et al. (2010a) reported that dwa1 and dwa2 mutants exhibit hypersensitivity to ABA and NaCl in root growth and seed germination, but show increased tolerance to drought in adult plants. DWA1 and DWA2 interact with each other and form a CRL4DWA E3 complex with CUL4. Both DWA1 and DWA2 directly interact with transcription factor ABI5, an important component of ABA signaling, and mark it for ubiquitination and degradation (Lee et al. 2010a). DWA3 was reported to be another negative regulator in the ABA response, but in contrast to DWA1 or DWA2, it does not directly bind to ABI5 (Lee et al. 2011).

In addition, a DWD gene known as DRS1 (Drought Sensitive 1) was identified by genomic functional network modeling, and was confirmed genetically (Lee et al. 2010b). The Arabidopsis drs1 mutant retained significantly less water than the wild-type (WT) under drought stress, and was insensitive to ABA-mediated water transpiration, but ABA treatment had no impact on drs1 seed germination (Lee et al. 2010b). This phenotype suggests that DRS1 plays an important and specific role in ABA-mediated water retention, though the mechanism of DRS1 has yet to be determined.

CRL and Auxin Receptors in Adaptive Responses to Temperature Stress and Nutrient Deprivation

SCF and phosphate (Pi) starvation

Environmental acclimation often requires plants to adjust the growth pattern of the roots. This process is primarily controlled by auxin, an important phytohormone perceived by a family of F-box proteins (AFBs), which assemble to SCF complexes (Gray et al. 2001; Mockaitis and Estelle 2008). In response to limited inorganic phosphate (Pi), an essential nutrient that plants absorb from soil, plants can adjust their gene expression pattern to alter their root architecture and growth pattern. This includes increases in the number and length of root hairs, increases in lateral roots, an increase in the expression of Pi transporters, and increases in the induction of endogenous and secreted phosphatases and RNases, among other adaptive responses in plants (Hermans et al. 2006). Induction of lateral root formation under Pi starvation is mediated through the auxin pathway. Perez-Torres et al. (2008) showed that Pi deprivation promotes the expression of the auxin receptor TIR1 F-box protein, which upon assembly to SCFTIR1, targets the degradation of AUX/IAA auxin response repressors, consequently enabling transcription factors such as ARF19 to mediate gene expression involved in lateral root formation. Mutation in tir1 drastically diminishes lateral root formation in low Pi conditions, while overexpression of TIR1 phenocopies the root structure observed in Pi-deprived seedlings (Perez-Torres et al. 2008).

FBX2 is also involved in the Pi starvation response, and contains a WD40 domain and an F-box (Chen et al. 2008). FBX2 was isolated as a negative regulator in root hair formation during the Pi starvation response (Chen et al. 2008). The fbx2 null mutant displays a constitutive Pi starvation phenotype and a hypersensitivity to Pi starvation with regard to root hair formation (Chen et al. 2008). In addition, FBX2 binds to a bZIP transcription factor BHLH32, which is also a negative regulator of the pathway whose mutant phenotype is similar to that of fbx2 (Chen et al. 2007). It is hypothesized that FBX2, or SCFFBX2, works with BHLH32 to inhibit root hair formation in Pi-sufficient conditions (Chen et al. 2008), but the mechanism of FBX2 remains unclear.

Temperature stresses

Temperature stress responses have been extensively studied in plants, revealing elaborated transcriptional cascades and the interplay of arrays of transcription factors (Hirayama and Shinozaki 2010; Huang et al. 2012). However, little information is available at the present time in regards to participating CRLs in these pathways. Still, an F-box protein AtFBP7 has been shown to be required for efficient protein translation during temperature stress (Irina et al. 2007). AtFBP7 expression is induced after cold and heat stress treatment, and it interacts with SCF adaptor protein ASK proteins. As of this writing, it is unclear how AtFBP7 works, but this finding paves the way for a new area of stress research in plants.

Auxin is known to be involved in increased hypocotyl elongation in high temperatures in Arabidopsis (Gray et al. 1998). It was recently found that this response of auxin is mediated by AFB4, a member of a unique clade of auxin receptors that is primarily responsible for hypocotyl elongation induced by picloram herbicide or high temperatures (Greenham et al. 2011). The mechanism of AFB4 has yet to be resolved. Cold stresses also affect auxin-mediated root growth, but auxin transport rather than auxin signaling and perception is involved in cold stresses (Shibasaki et al. 2009).

Regulation of CRL E3s and the COP9 Signalosome

Our understanding of how the CRL family of ubiquitin E3 complexes is involved in abiotic stress responses in plants is still at a basic level. Nonetheless, the versatility of CRL complexes and the diversity of the CRL target substrates can be easily glimpsed through the above case studies. With increasing clarity on the structural details of the CRL complexes (Zheng et al. 2002; Fischer et al. 2011), which share common organizational features, is it possible that there is a common regulatory mechanism(s) that governs this large family of E3s?

A common regulatory mechanism among the three major groups of CRLs is covalent modification by an ubiquitin-like protein named Rub1 (related to Ubiquitin, also known as Nedd8 in animals). Attachment of Rub1 to cullin proteins (rubylation) is accomplished through an enzymatic cascade similar to ubiquitin conjugation, but by a set of E1, E2, and E3 enzymes dedicated specifically to Rub1 instead of ubiquitin (del Pozo and Estelle 1999; del Pozo et al. 2002). Removal of Rub1, called derubylation, is carried out by the metallo-isopeptidase activity of CSN (COP9 signalosome), a conserved eukaryotic protein complex consisting of eight subunits (CSN1 to CSN8) (Lyapina et al. 2001; Schwechheimer et al. 2001). Since rubylation has been shown to stimulate the CRL assembly and its ligase activity, this modification thus represents an important mechanism by which CRL activity can be regulated (Hotton and Callis 2008; Hua and Vierstra 2011).

In addition to its protease activity, CSN can also inhibit CRLs by non-enzymatic means and by recruiting deubiquitinating enzymes and/or protein kinases (Wei et al. 2008; Fischer et al. 2011; Enchev et al. 2012). However, genetic studies have shown that a deficiency in CSN activity is detrimental to CRL function, indicating that negative regulation by CSN is necessary for sustained and optimized CRL function in vivo (Schwechheimer et al. 2001; Wang et al. 2003). Furthermore, null mutants of CSN survive embryo development, but die at the seedling stage (Wei and Deng 1992; Esther et al. 2008), while null mutants in each of the cullin genes in Arabidopsis: cul1, cul3a;cul3b double mutant, and cul4, cause embryonic lethality (Shen et al. 2002; Figueroa et al. 2005; Dumbliauskas et al. 2011). This is not surprising, since the CRL class of ubiquitin E3 ligases is used by plants to regulate diverse developmental processes and physiological responses, including the response pathways to abiotic stresses. This phenotype seems to fit the role of CSN as a modulator of CRL E3 ligases.

It is clear from the above case analyses that, even in the same signaling pathway, different CRL E3s may act as inhibitors or as activators. For example, SCFDOR acts negatively, while CRL3BMP acts positively, to regulate drought and ABA-induced stomatal closure (Zhang et al. 2008b; Lechner et al. 2011). Under Pi starvation, SCFTIR1 facilitates, while SCFFBX2 deters, auxin-mediated expansion of the root system (Chen et al. 2008; Pérez-Torres et al. 2008). This suggests that the general level of CRL activity is not up- or downregulated by certain stimuli. Instead, each specific CRL must be regulated individually, most likely in a manner dependent on its substrate receptor and substrate availability. One layer of regulation is done by controlling the expression level or localization of the specific substrate receptor. Indeed, the expression of many F-box and BTB proteins is induced by different stresses. Another layer of regulation could be mediated by CSN, the principal regulator of the entire family of CRLs. In a recent study, Fischer et al. (2011) showed that the substrate-dependent activity of CRL4DDB2 is conferred by CSN, because CSN inhibits CRL4DDB2 only in the absence of UV-damaged DNA. As a result, CRL4DDB2 is active when UV-damaged DNA is bound to DDB2, but inactive in the absence of damaged DNA (Fischer et al. 2011). How CSN accomplishes such substrate-dependent regulation of CRL4DDB2 would be a challenging problem to be resolved by future studies.

(Co-Editor: Giovanna Serino)

Acknowledgements

L.G. was supported by the “New Century Excellent Talents” program of the Ministry of Education, China [NCET-10-0153], the China Postdoctoral Visiting-Scholar Fellowship (201104519; 20100481042) and the Ministry of Environmental Protection of the PRC (2011467032). C.D.N. is a Yale University Brown Postdoctoral Fellow. N.W. and X.W.D. are supported by a National Institutes of Health grant (GM47850), the National Science Foundation (NSF) Plant Genome Program (DBI0922604), and the NSF 2010 program (MCB-0929100).

Ancillary