The gi mutants were described as late flowering mutants for the first time (Rédei, 1962). There are several gi mutants described in literature such as gi-1, gi-2, gi-3, gi-4, gi-5, gi-6, gi-11, gi-12, gi-100, gi-200, gi-201, gi-596, and gi-611 (summarized in Table 1; Figure 2). Some of the gi mutants were shown to influence the activity of the circadian clock, while others alter diverse responses (Park et al., 1999). The gi-1 allele, lacking the C-terminal part of GI, was responsible for shortening the period of the clock, while the gi-2 allele, lacking both the C-terminal and the central region of GI, lengthened the period. While the gi-1 mutation shortened the period of CAB2 expression, the gi-2 mutation lengthened the period of CAB2 expression (Park et al., 1999). This suggests that the central region of the protein or the terminal half of the protein most probably fine-tunes the period length of the circadian clock.

The gi-2 mutant at higher temperature of about 28°C showed longer hypocotyl and flowered earlier in comparison to the plants grown at temperatures of 18 and 22°C (Araki and Komeda, 1993). Even though higher temperature were shown to regulate flowering (at 18, 22, 28°C) and hypocotyl elongation (at 22, 28°C) in gi-2 mutant, it was almost equally sensitive toward vernalization as in WT. Vernalization is the exposure of plants to prolonged cold temperature that leads to earlier flowering cue in Arabidopsis. This implies that probably GI regulates flowering using a vernalization-independent pathway (Martinez-Zapater and Somerville, 1990; Koornneef et al., 1991; Araki and Komeda, 1993).

The alleles of GI are the result of random mutagenesis or T-DNA insertion which have aided in understanding its various functions. Alleles such as gi-1, gi-2, gi-3, and gi-6 introduce premature stop codon whereas gi-4 and gi-5 most probably alter the C-terminus of the protein due to frame-shift mutations (Fowler et al., 1999). No GI expression was detected in the gi-11 and gi-201 alleles carrying T-DNA insertion (Richardson et al., 1998; Martin-Tryon et al., 2007). The gi-100 mutation, originally identified in a red light screen, also contained a T-DNA insertion, but produced a truncated transcript of about 2 kb due to the absence of the 3′ end of GI (Huq et al., 2000). The transcript level in gi-1, gi-2, and gi-3 is lower compared to that of gi-4, gi-5, gi-6, and gi-100, which show similar or higher levels compared to their respective WT (Fowler et al., 1999; Huq et al., 2000). The role of GI in blue light dependent hypocotyl elongation was revealed using the gi-200 allele, consisting of a substitution of the serine 932 (Martin-Tryon et al., 2007).

Various deletions in GI sequences and its phenotypes are summarized in Table 1. After analyzing the data depicted in Table 1, it is evident that any deletion in GI mostly causes defects in the flowering time, circadian clock, and control of hypocotyl elongation. In the gi-4 mutant, improper splicing leads to a loss of 90 amino acids from the C-terminus causing late flowering. This deletion also causes the over-expression of its own transcript suggesting that the C-terminal 90 amino acids are required for its auto-regulation and flowering time (Fowler et al., 1999). The abundance of the gi-4 transcript could be due to increased stability or decreased decay which needs to be verified. Since GI stimulates CO transcription, this C-terminal domain of GI might be acting as an enhancer of CO transcription or involved in the recruitment of activators to the CO promoter.

The seeds of Wassilewskija (Ws) ecotype expressing CAB:LUC were mutagenized and screened for altered period length. Two novel alleles gi-596 and gi-611 were identified in this screen (Gould et al., 2006). In the gi-596 allele, mutation caused by the substitution of the serine residue at 191 position to phenylalanine (S191F) did not affect the flowering-time although the period length of the circadian clock is lengthened and longer hypocotyl was observed under both red and blue light conditions. This suggests that the serine 191 residue might have an important role in photoreceptor signaling. On the contrary, the mutation in gi-611 allele was mapped to the lysine 281. This allele showed significantly early flowering in SDs suggesting that this lysine in WT is involved in decelerating the flowering time (Gould et al., 2006). Since the Ws ecotype is a natural null for the high light sensor Phytochrome D, the phenotype observed could be a combinatorial effect of the lack of this photoreceptor and the respective mutations in GI allele (Aukerman et al., 1997). It would be interesting to evaluate if these alleles in Col background would show the similar light dependent effect to rule out the involvement and interaction of PHYD in this process. Both the positions, Lys281 and Ser191 are conserved in the Col-0 and Ler-0 ecotypes and thus, the role of these residues could be confirmed by the expression of the respective GI alleles containing the substitutions in these ecotypes to determine the importance of these mutated residues.

Defects in the circadian clock components have been found to affect the GI transcription. CIRCADIAN CLOCK ASSOCIATED 1 (CCA1), a core component of the circadian clock, reduces the GI expression by binding to CCA1-binding site on GI promoter (Lu et al., 2012). GI transcript, thus accumulates toward the middle of the day, when CCA1 expression is repressed by TIMING OF CAB EXPRESSION 1 (TOC1). The rhythmicity of GI transcript level is lost in elf3 mutant in continuous light (LL) suggesting that ELF3 might also regulate the GI mRNA abundance (Fowler et al., 1999). Since CCA1 and ELF3 have been proposed to physically interact to control flowering time and hypocotyl elongation, it would be interesting to investigate the coordinated involvement of these two proteins in the regulation of GI transcription. Clock proteins, such as, LIGHT-REGULATED WD 1 and 2 (LWD1 and LWD2) also affect the GI expression pattern, since in lwd1lwd2 double mutant GI transcript is most abundant at ZT 6 instead of ZT 10 (Wu et al., 2008). The two proteins being very similar (∼90% identity) possess functional redundancy, evident from single mutants being phenotypically similar to WT. Another clock associated gene, TIME FOR COFFEE (TIC) is also known to regulate the rhythmicity of GI in Arabidopsis. In tic mutants, GI transcript level is lower and the peak is shifted ∼4 h earlier than in WT plants (Hall et al., 2003). Since in both the lwd1lwd2 and tic mutants the GI expression is shifted to ZT6, it suggests that the activities of both the proteins might be required for the repression of the GI transcription in the morning. pseudo-response regulators (PRRs), namely, PRR5, PRR7, and PRR9 also have been proposed to regulate GI expression and therefore, flowering time via the CO-FT module (Nakamichi et al., 2007; Kawamura et al., 2008). Epistatic analysis and mutant combinations between LWD1/2, PRRs, and TIC would be beneficial to explain the additive roles of these genes products and the genetic hierarchy of the genes regulating the inhibition of GI expression. The expression of GI at the wrong time of the diurnal cycle is known to cause flowering time defects in At (Fornara et al., 2009). These mutants might behave as late flowering due to the untimely expression of GI. Although a lot is known from the transcript analysis, the work at the protein level is far from being understood due to the unavailability of a GI anti-serum that could detect the endogenous GI protein. The detailed post-translational regulation of GI is explained in the Section “Post-Translational Regulation of GIGANTEA.”

Several studies have demonstrated that light quality and quantity influence GI transcription, although systematic studies involving changes in the light fluence and wavelength to evaluate GI expression is yet to be carried out. In Arabidopsis, upon transition to night, GI mRNA level decreases with a half-life of about 1 h irrespective of the photoperiod (Fowler et al., 1999). A significant light dependent down-regulation is also detected in the legume Medicago truncatula suggesting a similar mechanism might coordinate light sensing with transcriptional activity (Paltiel et al., 2006). GI mRNA accumulation pattern in both Arabidopsis and M. truncatula showed a secondary peak at ZT 2 under SDs as well as LDs (Paltiel et al., 2006). This peak could be the result of an acute response to light. A similar peak of GI mRNA at ZT 2 has also been documented in plants grown under blue light. The role of blue light in the regulation of this early secondary peak of GI needs to be thoroughly examined using mutants that are affected in blue light signaling. This would clarify if the peak at ZT 2 is due to photoreceptors or secondary signaling components involved in blue light signaling. The peak expression of GI is delayed by approximately 4 h in plants grown in low red:far-red (R:FR) light conditions in comparison to plants grown in white light condition (Wollenberg et al., 2008). This indicated that the photoreceptors and their activity are fine-tuning the timing and quantity of the GI transcript. The accumulation of the GI protein in the morning around ZT 3–4 and its consequence in plant development has not been studied yet, that needs to be evaluated in depth.

Besides light, temperature too has been found to regulate GI expression. Warmer temperature of 28°C up-regulates GI mRNA level as compared to the cooler temperatures of 12°C at dawn (Paltiel et al., 2006). The night time repression of GI transcription was shown to be temperature dependent and regulated by evening complex (EC) night time repressor constituted of ELF3, ELF4, and LUX ARRHYTHMO (LUX; Nusinow et al., 2011; Mizuno et al., 2014). The EC night time repressor was revealed to bind to the GI promoter through LUX binding site (LBS).

GIGANTEA has been proposed to regulate its own expression, since the mutants, gi-1 and gi-2, have lower expression of the GI alleles, approximately 40 and 20% of the WT transcript, respectively (Fowler et al., 1999; Park et al., 1999). But this auto-regulatory role of GI transcription is contradictory, since, gi-4 and gi-6 lines show ∼30% higher expression of GI compared to its WT (Fowler et al., 1999). This effect could be either due to the difference in the ecotypes or differential regulation of the transcript stability. Another question worth investigating would be the abundance of the mutant proteins produced in each mutant, which would require a functional GI antiserum. The positive or negative auto-regulatory role suggests that mutations at different residues in the coding sequence can influence the abundance of transcriptional enhancers or repressors, affecting GI expression.

Over-expression of GI leads to the constitutive accumulation of the GI transcript throughout the photoperiod. Despite its constant expression level, GI protein follows a cyclic pattern of accumulation in both LDs and SDs. This is suggestive of the degradation of the GI protein (David et al., 2006). GI was found to be ubiquitinated upon dusk, a pre-requisite for its degradation via the 26S proteasome mechanism (David et al., 2006). In the dark phase, nuclear GI abundance has been shown to be regulated by the E3 ubiquitin ligase activity of COP1 and ELF3 (Yu et al., 2008). The interaction between COP1 and GI is ELF3 dependent, where ELF3 serves as an adaptor protein (Yu et al., 2008). The shuttling of COP1 between the nucleus and the cytoplasm is regulated by light (von Arnim and Deng, 1994). COP1 being nuclear localized in the night phase makes it competent for COP1–ELF3 mediated degradation of GI through 26S proteasome.

Upon heat shock GI is SUMOylated (López-Torrejón et al., 2013). It has been proposed that SUMOylation prevents the degradation of GI, thus enhancing its abundance. GI accumulation has been correlated with earlier flowering under heat stress. The identification of SUMOylation and ubiquitination sites in GI that alter its stability and degradation could be of pivotal importance in manipulating flowering time of crop plants. Current knowledge on the transcriptional and post-translational regulation of GI is presented schematically in Figure 3.

Photoreceptors such as phytochromes, cryptochromes, UV-light receptor, and phototropins help plants to sense variations in the light quality, quantity, and direction. The red and far-red light photoreceptors, phytochromes, are encoded by a multigene family, PhyA–E in Arabidopsis. While PhyA is the far-red light receptor, PhyB–E function as red light receptors with PhyB playing a predominant role. They mediate very-low-fluence responses (VLFRs), low-fluence responses (LFRs), and the high-irradiance responses (HIRs), with reference to the photon flux density (Casal et al., 1998). Like phyB-9 mutant, gi-100 also shows elongated hypocotyl when grown under saturated red light (Huq et al., 2000). Neither the genes nor the proteins abundance of PhyA and PhyB are influenced in gi-100 (Huq et al., 2000). Therefore, GI was suggested to function downstream of PhyA and PhyB. Mutation in GI leads to decreased VLFR under FR light suggesting its role in PhyA signaling (Oliverio et al., 2007). The gi mutants showed reduced seed germination and cotyledon unfolding in FR light conditions. These phenotypes are rescued by over-expression of GI. This suggested that GI might have a positive role specifically in PhyA mediated VLFR. GI also has a role in regulating flowering in low R:FR ratio which might be attributed to PhyA signaling (Wollenberg et al., 2008). Both PhyA and PhyB form NBs like GI. It would be interesting to determine if Phys and GI are present in the same sub-nuclear complexes and the localization of GI in the NBs alters the Phy-mediated functions.

The gi mutants showed longer hypocotyl in comparison to WT under blue light (Martin-Tryon et al., 2007). Earlier, it had been suggested that GI may be either a positive regulator of TOC1 or act parallel to it for the regulation of hypocotyl elongation. Since only gi not toc1 mutants show the longer hypocotyl in blue light, it can be inferred that GI does not regulate TOC1 for hypocotyl elongation (Martin-Tryon et al., 2007).

The circadian clock controls many processes depending on the length of the day and night cycle in an organism. In plants, the rhythmic expressions of various genes are influenced by the circadian clock, thereby regulating functions such as elongation of hypocotyl, petioles and inflorescence stem, movement of cotyledon and leaf, and flowering time. CCA1, LATE ELONGATED HYPOCOTYL (LHY), and TOC1 are the core components of circadian oscillator in plants (Somers, 1999). In 2005, the clock was proposed to be an interlocking network of proteins working in a feedback loop (Locke et al., 2005). According to the new model of clock, while the morning elements LHY and CCA1 repress TOC1 transcription, the evening element TOC1 down-regulates LHY/CCA1 accumulation, differing with the earlier observations (Alabadí et al., 2001; Gendron et al., 2012; Huang et al., 2012).

To understand the circadian rhythm in Arabidopsis, the ESPRESSO Quantitative Trait Loci (QTL) was generated from the cross between Ler and Cvi ecotypes (Swarup et al., 1999). Ler and Cvi ecotypes were suggested to comprise of an even distribution of alleles involved in the shortening and lengthening of period, since the progeny of their cross generated lines which had period length both longer and shorter than the parents. GI was identified as one of the genes that could be responsible for regulating the rhythms of cotyledon movement (Park et al., 1999; Swarup et al., 1999). The gi mutants have diverse circadian periods than WT concluding that GI has a role in period length regulation. Mutation in GI affects CHLOROPHYLL A/B-BINDING PROTEIN 2 (CAB2) gene expression which is also under the control of circadian clock (Park et al., 1999).

Soon after a day of imbibition of seeds, GI is required for initiating the rhythmicity of the circadian clock (Salomé et al., 2008). Mutations in the GI locus affect the CCA1 and LHY gene expression in both LDs and SDs conditions (Fowler et al., 1999). A recent study proposed that both the nuclear and cytosolic GI are required to positively and negatively regulate LHY expression, respectively, that fine-tunes the clock function (Kim et al., 2013b). Over-expression or mutations of CCA1 and LHY disrupted the GI expression (Fowler et al., 1999). Accordingly, the double mutant of LHY and CCA1 showed early abundance of GI transcript (Mizoguchi et al., 2002, 2005). It suggests that GI operates in a feedback loop as a component to maintain the rhythmicity and period length of the clock.

The established LHY/CCA1-TOC1 module of the clock could not explain the experimental data like the time difference of about 12 h between LHY/CCA1 abundance in morning and TOC1 accumulation in evening (Alabadí et al., 2001; Locke et al., 2005). It was therefore proposed that LHY/CCA1-TOC1 module comprises of other components. One of the components was predicted to be GI, whose expression followed the same pattern as predicted by the in silico analysis and was subsequently experimentally confirmed (Locke et al., 2006). Further work suggested that GI alone would not be able to regulate the observed time difference (Kawamura et al., 2008). TOC1 in turn is regulated by GI along with ZEITLUPE (ZTL), an F-box protein (Kim et al., 2011). ZTL is a blue light photoreceptor which is stabilized by its interaction with GI and Heat Shock Protein 90 (HSP90). Together the ZTL-GI complex control TOC1 level (Kim et al., 2007).

Temperature compensation is an important characteristic of the circadian clock to maintain the rhythm over a range of environmental temperature. GI was recognized as a candidate regulating temperature compensation effect, especially at higher temperatures (Edwards et al., 2005; Gould et al., 2006). Since fluctuations in the temperature regulate the abundance of GI transcript, it could be plausible that GI and temperature sensing mechanism crosstalk and feedback each other.

Arabidopsis thaliana dawn and dusk, GI regulates the clock rhythm along with ELF4 (Kim et al., 2012). GI was also required for iron-deficiency induced long circadian clock rhythm (Chen et al., 2013). Reduced depolymerization of actin filament caused the period of the circadian clock to shorten, as evident from the shortened period of GI expression (Tóth et al., 2012). Since GI expression is under the control of the circadian clock, GI accumulation pattern has been exploited to screen for novel clock mutants (Onai et al., 2004). Many components that mediate between GI and the clock are still to be unraveled. The role of GI in the regulation of the clock documented till date is summarized in Figure 4.

GIGANTEA is a major mediator between the circadian clock and the master regulator of photoperiodic flowering time control, CO. GI upregulates CO transcription, thereby accelerating time required to flower. Koornneef et al. (1998) showed that a novel mutant, gi-3, is epistatic to CO and FLOWERING LOCUS T (FT) way back. Mutation in GI led to a decrease in the accumulation of CO mRNA without affecting its cycling phase compared to its WT that led to delayed flowering (Suárez-López et al., 2001). Mutants in the GI locus or over-expressors of GI did not discriminate day-length for flowering. Accordingly, gi mutants were later flowering and over-expressors were earlier in both LDs and SDs (Rédei, 1962; Araki and Komeda, 1993; Mizoguchi et al., 2005).

During dawn, CO transcription is repressed by the combinatorial activity of the DOF transcription repressors bound to the CO promoter. In LDs, the expression of FKF1 and GI coincide at ZT10. Therefore, toward the middle of the day the accumulation of GI along with FKF1 forms a complex competent to degrade the DOF factors. This elevates the CO transcription, thereby leading to FT expression (Imaizumi et al., 2003, 2005; Sawa et al., 2007). While in SDs, since FKF1 accumulates 3 h after GI peaks, it does not allow the formation of the degradation complex, therefore leading to a low abundance of CO transcript. This photoperiod pathway where GI regulates FT expression in a CO-dependent manner is schematically depicted in Figure 5. GI regulates the abundance of FKF1, which is involved in the proteasomal degradation of proteins (Fornara et al., 2009). Post-transcriptionally, GI also controls the sub-cellular level of CYCLING DOF FACTOR 2 (CDF2; Fornara et al., 2009). FKF1 belongs to a family of F-Box proteins containing two other candidates – LOV KELCH Protein 2 (LKP2) and ZTL. The blue light dependent interaction between GI and FKF1 is mediated by the LOV (Light, Oxygen, or Voltage) domain of FKF1 and the amino-terminal of GI in vivo (Sawa et al., 2007). The gi-100 mutant is later flowering than the F-Box triple mutant fkf1 ztl-4 lkp2-1. This might be due to the presence of GI in fkf1 ztl-4 lkp2-1, which down-regulates the abundance of CDF transcripts, or the presence of an additional layer of control by GI bypassing the triple F-Box module.

There are at least two independent mechanisms through which GI regulates FT expression independent of CO. While the first mechanism involves microRNA, the second mechanism is through the binding of GI to the FT promoter. The microRNA based control involves miRNA172, which is positively regulated in the presence of GI. The miR172 inhibits the expression of TARGET OF EAT1 (TOE1), an APETALA 2 (AP2)-related transcriptional repressor of FT (Jung et al., 2007). In the recent past, expression of GI specifically in the mesophyll or vascular tissue was carried out. This rescued the late-flowering phenotype of gi-2 under both day length conditions and two different temperatures of 16 and 23°C (Sawa and Kay, 2011). The expression of GI in mesophyll and vascular tissue was done using tissue specific promoters LIGHT-HARVESTING COMPLEX B2.1 (pLhCB2.1) and SUCROSE TRANSPORTER 2 (pSUC2), respectively. While expression pattern of GI under the control of pLhCB2.1 is altered and peaked at ZT 0, GI expressed under the phloem specific promoter led to the over-expression of the transcript with peak at ZT 10. The FT transcript level was up-regulated without an increase in CO mRNA in both day-length conditions. GI was shown to binds to the FT transcriptional repressors such as SHORT VEGETATIVE PHASE (SVP), TEMPRANILLO 1 (TEM1), and TEMPRANILLO 2 (TEM2), and their specific target regions within the FT promoter in the mesophyll, thereby relieving the repression and promoting FT transcription (Sawa and Kay, 2011). The degradation of the FT transcriptional repressors or the unavailability of their binding sites on the FT promoters due to the presence of GI could lead to the abundance of the FT transcript. FT expressed in the vascular tissue normally triggers flowering. Since GI expressed in mesophyll accelerated flowering, elevating the FT level in vasculature, the signal from mesophyll GI most likely induces CO transcription in vasculature. Alternatively, the GI could be transported to the vascular tissues and induce the photoperiod module which needs to be investigated.

Expression of 35S::GI:GFP in gi-3 plants complemented the late flowering phenotype of gi-3. On the contrary, expressing the 35S::GFP:GI in gi-3 caused later flowering compared to the background lines indicating that the N-terminal fusion of GI might be either non-functional or might not be imported into the nucleus. In the transgenic line expressing C-terminal fusion, the fusion protein was localized to the nucleus and formed NBs (Mizoguchi et al., 2005). In an independent study, transgenic plants expressing glucocorticoid receptor (GR) fusion of GI flowered with ∼20 leaves less when treated with dexamethasone, compared to its untreated control which flowered with ∼55 leaves under LDs (Günl et al., 2009). In 15 day old seedlings, the induction of flowering time genes like CO and FT took place ∼28 h after dexamethasone treatment causing early flowering. This indicates that cytoplasmic retention of GI most probably delays time to flower. Mutation in GI is epistatic to mutation in ELF4 and together regulate CO expression (Kim et al., 2012). Recent studies showed that ELF4 sequesters GI into nuclear bodies, thereby preventing GI to associate with the CO promoter (Kim et al., 2013b). It would be interesting to know the nature of the GI nuclear bodies and the components there in, using biochemical approach followed by mass spectrometric analysis.

GIGANTEA interacts with N-terminal tetracopeptide domains of SPINDLY (SPY), a plant O-linked β-N-acetylglucosamine transferase, and antagonizes its activity, thereby, promoting flowering (Tseng et al., 2004). Acetylglucosamine transferases have role in the addition of acetylglucosamine residues to proteins, which often competes with phosphorylation. This suggests that sugar modification may function as an important event in flowering time regulation. The known pathways through which GI regulates flowering are summarized in the Figure 6.

Besides flowering time, circadian clock, and light signaling regulation, GI has been implicated in other processes such as, sucrose signaling (Dalchau et al., 2011), starch accumulation (Eimert et al., 1995), and stress tolerance (Kurepa et al., 1998a; Fowler and Thomashow, 2002; Kim et al., 2013a; Riboni et al., 2013). The control of cotyledon movement, transpiration, and hypocotyl elongation responses have been shown to be attributed to the concerted activity of SPY and GI (Sothern et al., 2002; Tseng et al., 2004). The precise nature of this interaction is still unclear. However, GI functions antagonistically to SPY. The interaction of GI with SPY and ELF4 independently regulates hypocotyl length, where mutation in ELF4 and SPY are epistatic to gi-2.

GIGANTEA has been demonstrated to play a role between sucrose signaling and the circadian clock while grown in DD (Dalchau et al., 2011). Plants entrained in LD when shifted to DD, maintained the rhythmic GI expression exclusively in the presence of sucrose suggesting light independent control of GI rhythmicity. Although contradictory evidence on role of sucrose on GI expression has been reported, sucrose seems to affect the GI expression through SENSITIVE TO FREEZING6 (SFR6) locus (Knight et al., 2008; Usadel et al., 2008). More precise experiments are required to unravel this mechanism. In the leaves of Arabidopsis, starch accumulation is elevated in the gi mutants (Eimert et al., 1995). On the contrary, presence of multiple copies of GI led to starch accumulation in the progeny of a cross between A. thaliana and A. arenosa, suggesting the antagonistic role of GI in these plants (Ni et al., 2009).

The gi-3 mutants showed higher tolerance capacity to redox cycling agent, paraquat, and H₂O₂ (Kurepa et al., 1998a). Tolerance against paraquat is counteracted by the exogenously applied polyamines such as spermidine, spermine, and putrescine (Kurepa et al., 1998b). Paraquat treatment upregulated endogenous levels of putrescine in gi-3 and WT. Since exogenous application of polyamines is effective for the resistance, the mechanism of the transporters during this stress needs attention. Oxidative stress due to herbicide imazethapyr has been shown to increase GI abundance and cause earlier flowering by ∼4 days (Qian et al., 2014). The mechanism behind higher tolerance to oxidative stress mediated by GI is still unclear.

Kurepa et al. (1998a) showed that gi mutants, gi-3, gi-4, gi-5, and gi-6, have more chlorophyll accumulation in comparison to WT in presence of paraquat. Even treatment with nitric oxide (NO) reduces the GI mRNA abundance and increases the chlorophyll content (He et al., 2004). In both the cases above, lower abundance of functional GI can be correlated to higher accumulation of chlorophyll. The role of GI in regulating the chlorophyll content needs to be studied in mutants and over-expressors of GI. Chlorophyll accumulation in allotetraploid, obtained by a cross between A. thaliana and A. arenosa, is higher than the WT individuals (Ni et al., 2009). The starch and chlorophyll accumulation in allotetraploids is exactly opposite in comparison to that seen in A. thaliana. The reverse trend might be due to post-transcriptional silencing posed by the presence of multiple homologous sequences of GI transcript, essentially a co-suppression phenomenon.

Dynamin, a GTPase having role in vesicle recycling during endocytosis, was found to interact with TAP tagged GI in rice (Abe et al., 2008). Although mutation in dynamin gene did not have any effect on the flowering time, it showed aerial rosette phenotype in Arabidopsis. In Arabidopsis, GI has been found to be involved in setting of fruits (Brock et al., 2007). No significant association of the GI haplogroup was detected with days to flower, petiole length, and inflorescence height. A significant association was observed between one haplogroup with fruit set, producing 14% more fruit than other haplogroups. Such studies in the crop plants could help in increasing the yield.

GIGANTEA mRNA levels increases about five- to eightfold in the cold treated Arabidopsis plants suggesting that GI is a cold-responsive gene (Fowler and Thomashow, 2002). The flowering time of gi mutants was further delayed when exposed to low temperature compared to WT (Cao et al., 2005). C-repeat Binding proteins (CBFs) have been known to regulate various genes responsive to cold and are implicated in cold stress tolerance. On the contrary, Cao et al. (2005), it was revealed that GI regulates cold acclimation through CBF-independent pathway. The ability to tolerate and acclimatize toward cold is reduced in gi mutants suggesting the protective role of GI in cold tolerance.

Recently, the role of GI under salt stress was documented (Kim et al., 2013a). Although, salt stress did not affect the GI expression, it affected the GI protein stability in pGI::GI-HA transgenics (Cao et al., 2005; Kim et al., 2013a). It seems plausible that there is a mechanism at the post-translational level that regulates GI abundance. GI also regulated the activities of the proteins involved in the salt stress tolerance. It interacts with Salt Overly Sensitive 2 (SOS2) directly and inhibits the activity of SOS1, a Na⁺/H⁺ anti-porter. Therefore, GI is a negative regulator of salt tolerance and is degraded during salt stress. According to a recent model, plants under salt stress would flower later than when grown in normal growth conditions reasoned for the degradation of GI (Park et al., 2013).

In At and other plants, the tolerance to higher salinity, enhanced cold, and sustained drought were manifested by the increase of sub-cellular level of abscisic acid (ABA). Recent reports indicated that GI has role in ABA-dependent drought escape tolerance. It suggests that the GI regulation of salt and cold stress tolerance could very likely be ABA-mediated (Riboni et al., 2013). Drought stress up-regulates GI transcription and in turn, increases the abundance of miR172E variant (Han et al., 2013). WRKY DNA binding protein 44 (WRKY44) was found to be suppressed by GI in drought stress and interact with TOE1. GI-miR172-WRKY44 were proposed to be in the same pathway possibly associated with drought stress tolerance. On the same line of thinking, the light dependent GI-mediated stomatal opening response could be ABA mediated (Ando et al., 2013). GI also has a role in wall in-growth deposition in phloem parenchyma transfer cells in A. thaliana in response to high light and cold stress (Edwards et al., 2010; Chinnappa et al., 2013).