plant g proteins, phytohormones, and plasticity: three ...€¦ · the genomes of diploid...

www.stke.org/cgi/content/full/sigtrans;2004/264/re20

Heterotrimeric guanine nucleotide-binding proteins(G proteins) composed of Gα, Gβ, and Gγ subunitsare important transducers of hormonal signals in or-ganisms as evolutionarily distant as plants and humans. The genomes of diploid angiosperms, suchas that of the model species Arabidopsis thaliana,encode only single canonical Gα and Gβ subunits,only two identified Gγ subunits, and just one regula-tor of G protein signaling (RGS) protein. However, awide range of processes—including seed germina-tion, shoot and root growth, and stomatal regula-tion—are altered in Arabidopsis and rice plants withmutations in G protein components. Such mutantsexhibit altered responsiveness to a number of planthormones, including gibberellins, brassinosteroids,abscisic acid, and auxin. This review describes pos-sible mechanisms by which such pleiotropic effectsare generated and considers possible explanationsfor why G protein component mutations in plants failto be lethal. A possible role of G protein signaling inthe control of phenotypic plasticity, a hallmark ofplant growth, is also discussed.

IntroductionG proteins are found in organisms as diverse as slime molds,sponges, and humans. In mammals, G proteins mediate signalsfrom various sensory stimuli and neuroendocrine ligands; to do so,they draw upon a large stock of distinct G protein–coupled recep-tors (GPCRs) and regulatory proteins (1, 2). It is estimated thatabout half of today’s pharmaceuticals target G protein–based path-ways (3), and G protein component defects have been linked to vari-ous genetic diseases (4, 5).

New studies are also revealing roles of G proteins in hormonalsignaling in flowering plants. However, the textbook definition of ahormone as a molecule that is produced by one specific organ andconveyed to target tissues, where it elicits a physiological responseat low concentration, does not hold true for plant hormones. Mostplant hormones are synthesized by several different tissues or celltypes, and can act locally as well as at a distance. Like hormones ofmammals (2), plant hormones can be perceived by multiple types ofcells, and the nature of the responses that are thereby engenderedcan differ greatly between one cell type and another.

In this review, the roles of G proteins in plant hormonal signal-ing and response are discussed in the context of three questions:

1) What are the plant hormonal responses that are influenced byG proteins?2) How can only two plant heterotrimers mediate so manyplant hormone responses?

3) Given the paucity of plant G protein subunits, why aren’tplant G protein component knockout mutations lethal?

Some speculation is also offered regarding the importance of Gproteins in the context of phenotypic plasticity and evolutionary fitness. Discussion is limited to the two model plant species forwhich the full sequence of the genome is available, Arabidopsisthaliana and rice (Oryza sativa), and to the specific realm of hor-monal responses. Plants respond to a remarkable diversity of envi-ronmental cues and metabolites (Fig. 1), and G protein involvementin other aspects of plant signaling—notably responses to pathogens,pollutants, and drought—has also been demonstrated. For other as-pects of plant G protein function, including discussions of phyloge-ny and protein structure, the reader is referred to several recent re-views (6–8). In particular, Jones and Assmann (9) provide a com-parative evaluation of plant versus mammalian G protein signalingcomponents and downstream effectors.

Background and Overview: The Plant Heterotrimerand Associated Regulatory ProteinsIn the standard paradigm of G protein signaling, Gα-GDP, Gβ,and Gγ form an inactive complex. Ligand binding to an associ-ated GPCR elicits a conformational change in the Gα subunit,leading to exchange of GDP (guanosine diphosphate) for GTP(guanosine triphosphate) and a consequent conformationalchange that releases the Gβγ dimer. Free Gα and Gβγ are avail-able for interaction with downstream effectors until intrinsicGTPase activity of the Gα subunit allows reunion of the het-erotrimer. Experimentally, G protein activation also can be in-duced by mutagenesis of particular amino acids in the Gαsubunit, resulting in the creation of a permanently active Gαconformation, or by application of nonhydrolyzable GTP [forexample, guanosine 5´-O-(3´-thiotriphosphate) (GTP-γ-S)],which locks the G protein in the active state (1).

Recent pharmacological studies have revealed that otherstates of the GPCR beyond ligand-activated and inactive arepossible. Many GPCRs have basal activity that can be alteredby agonists and antagonists, as well as by inverse and proteanagonists. Protean agonists either stimulate or inhibit receptors,depending on the initial state of the receptor (10–12) (Fig. 2).Posttranslational modifications of GPCRs also modulate signal-ing, as does GPCR homo- or heterodimerization (1, 10). Thus,it may be appropriate to model GPCRs in the same way thatelectrophysiologists model ion channels, which can be open(active), closed (not active), or inactive (in a state not suscepti-ble to activation) (Fig. 2). Adding further layers of complexity,Gα activation by proteins that do not bear the archetypal seventransmembrane–spanning (7TMS) structure of the GPCRs oc-curs in some signaling pathways (13, 14), and signaling bysome GPCRs to non-Gα proteins also occurs (15).

Because the plant G protein field is relatively new, the com-plex regulation of GPCRs described above has yet to be

Plant G Proteins, Phytohormones, and Plasticity:Three Questions and a Speculation

Sarah M. Assmann(Published 21 December 2004)

R E V I E W

Biology Department, Pennsylvania State University, 208 Mueller Labora-tory, University Park, PA 16802–5301, USA. E-mail: [email protected]


experimentally evaluated in plants. However, basic elements ofG protein signaling appear to be conserved between plants andanimals, although plant genomes encode far fewer het-erotrimeric subunits than the 23 Gα (including splice variants),5 Gβ, and about 11 Gγ subunits present in mammaliangenomes (16). Indeed, in the fully sequenced Arabidopsis andrice genomes, the Gα and Gβ subunits are both encoded bysingle-copy genes, and there appear to be only two Gγ subunits(Table 1). Respectively, these genes are namedGPA1 in Arabidopsis (RGA1 in rice), AGB1(RGB1 in rice), and AGG1 and AGG2 (RGG1and RGG2 in rice) (17–23). All four of the Ara-bidopsis subunits exhibit ubiquitous expression[reviewed in (7)]. Existence in the plant plasmamembrane of a genuine heterotrimer has beendocumented by gel filtration analysis of proteincomplexes in rice; moreover, dissociation of thiscomplex by nonhydrolyzable GTP has been observed (23). Structural modeling of the Ara-bidopsis subunits suggests that an analogous het-erotrimer exists in Arabidopsis (24). Plant Gαsexhibit intrinsic GTPase activity, albeit at a lowerrate than that of mammalian Gαs (25–28), andconstitutively active mutants can be generated bythe same strategy as in mammalian systems (24,27). Downstream effectors of G protein signalingin plants include phospholipase C, K+ channels,anion channels, and Ca2+ channels (29–31), al-though the number of elements situated betweenthe G protein and each of these effectors remainsunknown (9). GPA1 has been shown biochemi-cally to interact with and thereby inhibit the ac-tivity of phospholipase D1α (PLD1α); GTP re-lieves this inhibition (32). A cupin-domain protein, AtPirin1, interacts with GPA1 in a yeasttwo-hybrid assay (33). It is expected that manymore effectors will be uncovered.

Plants also appear to have far fewer GPCRsthan the 800+ complement found in mammals(34). In fact, only one Arabidopsis protein,GCR1, bears prominent sequence similarity(within a limited region) to metazoan GPCRs(35–37). GCR1 can be co-immunoprecipitatedwith GPA1 from plant tissue (36). This result iscertainly suggestive of GPCR function for GCR1,but until a GCR1 ligand is found, its classifica-tion as a GPCR remains putative. Plants may haveadditional 7TM proteins that function as GPCRsbut are difficult to discover via searches for se-quence similarity, just as mammalian GPCRswithin the different GPCR families (38) can have little or no se-quence similarity to one another.

Proteins that modulate signaling through GPCRs and G pro-teins in mammals include (i) GPCR kinases (GRKs) andnon–GRK-type receptor kinases, which desensitize GPCRs toactivation by ligand; (ii) GPCR phosphatases, including proteinphosphatase 2A enzymes (PP2As), which resensitize GPCRs;(iii) arrestins, which terminate signaling by promoting GPCRinternalization (39) and also function as adaptor proteins thatregulate mitogen-activated protein kinase cascades (40); (iv)phosducins, which interfere with signaling in certain tissues by

scavenging Gβγ dimers (41); and (v) regulators of G proteinsignaling (RGS), which promote Gα GTPase activity and thustypically accelerate return to the heterotrimeric state (42). Ofthese regulatory proteins, PP2As (43) and one RGS protein(RGS1) (28) are present in Arabidopsis, as confirmed by bothsequence and functional analysis. Arabidopsis RGS1 is unusualin that it has both an RGS domain and a 7TM domain; thus, itmay function as both a GPCR and an RGS (28).

What Are the Plant Hormonal Responses Influencedby G Proteins?Application of each of the major plant hormones leads to character-istic responses of various plant organs at particular stages of develop-ment. Although responses to exogenous hormones do not alwaysfaithfully reflect responses to endogenous hormone levels (44), thesestereotypical responses have nonetheless become useful hallmarksfor evaluation of hormone sensitivity or responsiveness, or both, ingenetically altered experimental plants. Analysis of Gα, Gβ, GCR1,and RGS1 mutants for hormonal responses has been initiated in Arabidopsis, whereas in rice only mutants of the rice Gα subunit

R E V I E W

Hormones and metabolitesAbscisic acid (ABA)

Auxin

Brassinosteroids (BR)

Cytokinin

Ethylene

Gibberellins (GA)

Glutamate

Jasmonic acid

Sugars

Cotyledons

Hypocotyl

Primary root

Stomatal pore

Guard cell

Fig. 1. Stages of development of the model plant Arabidopsis thaliana.Hormones and metabolites that affect physiology, growth, and development arelisted. Figure influenced by figure 1 of (77) and figure 1 of (80).


gene RGA1 have been studied todate. For both species, only a sub-set of plant hormones—primarilygibberellic acid, brassinosteroids,abscisic acid, and auxin—and asubset of the attendant responseshave been characterized thus far.

Gibberellins. Classical roles ofgibberellic acid (GA) include pro-motion of seed germination, elon-gation of the hypocotyl [the por-tion of the seedling stem locatedbelow the cotyledons (Fig. 1)] andstem, leaf expansion, flowering,and fruit expansion. Indeed, grow-ers spray some varieties of tablegrapes with GA to increase grapeberry size (45). Growth-promotiveeffects of GA in the vegetativeplant are typically attributed to in-duction of cell expansion ratherthan stimulation of cell division.

Arabidopsis mutants in the Gprotein α or β subunit (gpa1,agb1), the putative GPCR (gcr1),and the RGS protein (rgs1) allshow reduced seed germination in response to exogenous GA appli-cation as well as increased sensitivity to inhibition of germinationby the GA synthesis inhibitor paclobutrazol, whereas cell lines over-expressing the Gα subunit GPA1 are hypersensitive to GA in seedgermination assays (37, 46). These results suggest that the het-erotrimer either transduces or modulates the GA response of seeds;these possibilities are discussed in more detail below.

Arabidopsis seedlings expressing a constitutively active versionof GPA1 (GPA1QL) have longer hypocotyls when grown in darknessas a result of increased cell elongation (28), which is also consistentwith a GA-hypersensitive phenotype, although this has yet to bespecifically tested. rgs1 mutants also show this long hypocotyl phenotype, consistent with the idea that RGS1 reduces the GPA1activation state in wild-type plants (28).

During the final, reproductive phase of growth, Arabidopsisplants harboring G protein component mutations also show phe-notypes suggestive of, but not yet definitely linked to, alteredGA signaling. Thus, agb1 plants exhibit shorter floral parts andthicker, blunted fruits, a response that is somewhat phenocopiedby overexpression of GPA1 (24, 47). Such similarity may occurbecause overexpression of GPA1 results in sequestration ofAGB1 by GPA1, resulting in an AGB1-deficient phenotype.

In rice, the d1 mutant, harboring a null mutation of the RGA1Gα subunit gene (23), also shows reduced GA sensitivity (48–50).d1 mutants require higher concentrations of GA to induce expres-sion of several GA-specific genes in the seed (50). In d1 seeds thealeurone layer, a tissue rich in hydrolytic enzymes, exhibits loweractivity of α-amylase, a GA-induced enzyme that is crucial for mo-bilization of the carbohydrate reserves that nurture the embryo dur-ing germination (50). The rice mutants of RGA1 also show GA-related alterations in vegetative morphology; these mutants aredwarves as a result of decreased sensitivity to GA promotion of in-ternode elongation (49, 50). However, other organs of rice, specifi-cally the lower part or sheath of the second leaf, show normal sensi-tivity to GA (50), indicating tissue specificity of response.

Brassinosteroids. Brassinosteroids (BRs) have only relative-ly recently received recognition as genuine plant steroid hor-mones (51). Exogenous BRs can stimulate cell expansion andhave general growth-promotive effects on above-ground plantparts, although root growth can be inhibited (51). Brassino-steroid mutants exhibit a dark green phenotype when light-grown and exhibit aspects of light-grown morphology whengrown in the dark. Application of the synthetic BR brassinolide(BL) promotes Arabidopsis seed germination in plants with mu-tations in genes related to GA biosynthesis or perception.

Only a little research has been done concerning BR respons-es in G protein component mutants. The rounded-leaf or “rotun-difolia” phenotype of gpa1 and agb1 Gα and Gβ subunit mu-tants is similar to that observed in a cytochrome P450 mutantposited to be defective in plant steroid synthesis, suggesting apossible link (52). In Arabidopsis seeds pretreated with the GAbiosynthesis inhibitor paclobutrazol, BL rescue of seed germi-nation is reduced in gpa1, agb1, and gcr1 single mutants rela-tive to the response of wild-type seeds (37, 46). Likewise, in the

R E V I E W

I2 C1 O1

Constitutive (basal)activity state of GPCR

Agonist orprotean agonist

Antagonist

Agonist orinverse agonistplus antagonist

Inverse agonistor protean agonist

Degradation,internalization,posttranslationalmodification,protein-proteininteraction

Posttranslationalmodification,protein-proteininteraction

Degradation,internalization,posttranslationalmodification,protein-proteininteraction

Posttranslationalmodification,protein-proteininteraction

O2 I1

Fig. 2. Known modes of GPCR modulation in mammals, by analogy to ion channel models. Ionchannel models commonly consider states of the ion channel as open (O, active), closed (C, notactive), and inactive (I, a state in which the channel cannot be activated, even in the presence ofan activating stimulus). Similar reasoning can be applied to GPCRs. In the absence of antagonists, agonists increase activity beyond the basal level (O1); inverse agonists decrease ac-tivity beneath the basal level. Protean agonists can have either agonist or inverse agonist effects,depending on the level of constitutive activity. Antagonists bring activity back to the basal level. Lo-calization, modification, and interaction with other proteins also affect GPCR status.

Protein Arabidopsis Rice

Putative G protein–coupledreceptor (GPCR)

GCR1 –

Gα GPA1 RGA1

Gβ AGB1 RGB1

Gγ AGG1 RGG1

Gγ AGG2 RGG2

Regulator of G proteinsignaling (RGS)

RGS1 –

Table 1. Plant G protein components discussed in this review.


rice d1 mutant, BL stimulation of gene expression is less effec-tive than in wild-type plants (53). These results suggest that dis-ruption of G protein signaling reduces sensitivity to BR.

Abscisic Acid. In contrast to GA and BR, abscisic acid(ABA) has a general growth-retarding role. ABA is importantfor embryo development within the seed and for subsequent de-velopment of seed dormancy; indeed, some ABA-deficient mu-tants of maize show viviparous seed germination on the ear(54). In young seedlings under conditions where sufficient wa-ter is available, application of ABA delays overall growth andinhibits elongation of the primary (main) root. ABA also hasimportant acclimatory roles in plant stress responses, particular-ly responses to drought, cold, and salinity (55). Throughout theplant body, elevated concentrations of ABA increase transcriptabundance of a number of genes whose products are importantfor stress tolerance. ABA also contributes to plant water reten-tion during drought by inhibiting the opening and promoting theclosure of stomata, microscopic pores on the leaf surfacethrough which plants simultaneously take up CO2 and lose wa-ter vapor (Fig. 1) (56).

ABA responses have been evaluated in Arabidopsis G pro-tein component mutants but not in corresponding mutants inrice. Arabidopsis gcr1 mutants exhibit ABA hypersensitivity ininhibition of primary root elongation and in expression of someABA-related genes in whole seedlings (36). Arabidopsis gcr1mutants are also hypersensitive to ABA inhibition of germina-tion (36), a phenotype that is shared by single mutants of gpa1and agb1 (33, 46) as well as by double and triple mutant combi-nations among these three genes (57).

However, with respect to stomatal aperture regulation, aquite different story is emerging. gpa1 and agb1 mutants showwild-type ABA induction of stomatal closure but actually ex-hibit insensitivity to ABA inhibition of stomatal opening, ratherthan the hypersensitivity displayed by other tissues (30, 58, 59).The similar phenotypes of the Gα and Gβ mutants suggest thateither Gα or the concerted action of Gα and Gβγ mediate theABA response in wild-type cells. Stomatal opening is driven inpart by an influx of K+ ions into the two guard cells that borderand define the stomatal pore. As a result of this influx, the cellstake up water, swell, and separate from one another, thus widen-ing the stomate. K+ influx occurs through inwardly rectifyingK+ channels that are inhibited by ABA in wild-type plants butnot in gpa1 [or agb1 (59)] mutants (30, 58). K+ influx occursupon membrane hyperpolarization, and this driving force for K+

uptake is lessened by membrane depolarization. In wild-typeguard cells, two processes that contribute to ABA promotion ofdepolarization are activation of anion channels that mediate an-ion efflux and activation of Ca2+-permeable channels that medi-ate Ca2+influx. Regulation of slow anion channels by ABA is al-tered in gpa1 mutants (30) and ABA-activation of Ca2+-perme-able channels is abrogated (31), thereby contributing to ABA in-sensitivity of stomatal opening.

In guard cells, ABA activates sphingosine kinase activity, whichleads to production of the lipid metabolite sphingosine-1-phosphate(S1P). In mammalian systems, S1P signals through specific GPCRs(60). S1P mimics ABA effects on K+ and anion channel regulationin wild-type plants but fails to do so in gpa1 mutants, which sug-gests that an S1P signal (generated by ABA) signals via Gprotein–based cascades in plants as well as in mammals (58).

Given the reduced ABA sensitivity of guard cells in gpa1 andagb1 mutants, one might then predict that mutation of the putative

GPCR, GCR1, would also lead to ABA hyposensitivity, but this isnot the case. gcr1 mutant guard cells are similar to gcr1 seeds andprimary roots in showing hypersensitivity both to ABA inhibition ofstomatal opening and to ABA promotion of stomatal closure. gcr1guard cells also exhibit hypersensitivity to S1P in these two process-es. These results imply that GCR1 is a negative regulator of ABAsignaling, because elimination of gcr1 in mutant plants results inABA and S1P hypersensitivity (36).

Although the electrophysiology of ABA-regulated ion chan-nels has yet to be investigated in cell types other than guardcells, the fact that gpa1 (and agb1) mutants are hypersensitiveto ABA in seed and root tissue, but hyposensitive to the hor-mone in stomatal opening, clearly points to cell specificity inthe G protein dependence of ABA signaling pathways. Such cellspecificity is also illustrated in auxin responses.

Auxin. Auxin regulates both cell division and cell expansion;which process is enhanced depends on both the auxin concen-tration and the cell type. Endogenous auxin appears to promotestem and primary root elongation by enhancing responsivenessto GA (61). Endogenous auxin indirectly promotes dormancy ofaxillary buds (62), thereby reducing branching in above-groundtissue. However, application of exogenous auxin to intact wild-type plants inhibits primary root growth while simultaneouslypromoting the formation and elongation of lateral roots, whichare initiated from a different stem cell population than the pri-mary root (63). Clearly, the role of auxin is highly cell-specificeven in wild-type plants.

Considering first the above-ground plant organs, hypocotylsin gpa1 and agb1 mutants are shorter than in wild-type plantsbecause they contain fewer cells (24, 64). If this effect is relatedto auxin, it must reflect altered auxin regulation of cell division,because assays of auxin-related hypocotyl cell elongation showno substantial differences between gpa1 and agb1 mutants ver-sus wild-type plants (24). gpa1 leaves similarly show reducedcell division, whereas GPA1 overexpressors show ectopic celldivision (30). These results imply a role for G proteins in stimu-lating cell division in above-ground tissues.

In roots, the G protein component mutants show clear alter-ation in auxin responsiveness of lateral root formation (24). Inthe presence of auxin, agb1 mutants and GPA1 overexpressorsform many more lateral roots than do wild-type plants. In theabsence of exogenous auxin, gpa1 mutants form fewer lateralroots than do wild-type plants. These results are consistent witha model whereby free βγ subunits (postulated to be more abun-dant in the gpa1 genotype and less abundant in the agb1 andGPA1 overexpressor genotypes) negatively regulate the auxin-induced cell division that leads to the production of lateral rootprimordia. According to this scenario, the status of the activeheterotrimer in auxin response would be signaled not by Gα,but rather by interaction of Gβγ with downstream effectors.Auxin may act in part by reducing the abundance of the nega-tive regulator AGB1, because in wild-type seedlings, theamount of AGB1 transcript is decreased by auxin treatmentwhile that of GPA1 is increased (24).

How Can Only Two Plant Heterotrimers Mediate SoMany Plant Hormone Responses?Even though several of the important plant hormones listed inFig. 1 have yet to be evaluated for altered responsiveness in theG protein component mutants, and responses to the other hor-mones have been only incompletely studied, it is evident that G

R E V I E W


protein signaling modulates a plethora of hormonal effects,thereby influencing numerous developmental processes andphysiological responses. The information summarized abovesuggests at least two explanations for the versatility of plant Gprotein signaling.

A Limited Number of Heterotrimeric Combinations Trans-duce Signals from Multiple Hormones. In seeds, germinationresponses to GA, BR, and ABA are all altered in G proteincomponent mutants (46). Not all of these responses have beenevaluated in the rice rga1 mutants, but responses to GA and BRseem similarly altered in rice seeds.

Two models have been put forth to explain such results (50).On the basis of the results observed in rice, it has been suggest-ed that there may be a high-sensitivity GA pathwaythat is transduced through the rice Gα subunitRGA1 and (presumably) a GPCR, as well as a low-sensitivity pathway, mediated by a non–GPCR-typeGA receptor, that functions independently of G pro-tein signaling (Fig. 3A). Thus, in the rice d1 mutant,high sensitivity to GA is lost but low sensitivity toGA is retained. According to this model, RGA1might actually respond to only one hormone, GA,but with lack of input from the high-sensitivity GApathway, the ultimate effect would be an apparentincreased sensitivity to signals that are inhibitory forgermination.

An alternative model (46, 50) is that GA doesnot signal via a GPCR and G protein, but rather Gαmodulates the sensitivity of the seed to GA, perhapsby increasing the effectiveness of BR in promotionof GA sensitivity (Fig. 3B). Thus, when Gα is elimi-nated, sensitivity to GA is reduced even though Gαdoes not function directly in the GA signaling path-way. Consistent with this possibility, overexpressionof GPA1, while causing GA hypersensitivity, doesnot eliminate the GA requirement for Arabidopsisseed germination (46).

A third model would add another level of com-plexity, namely, that GPA1 influences not only sensi-tivity to GA but also sensitivity to all the stimuli thatregulate seed germination (Fig. 3C). In both the sec-ond and third models, the G protein does not in facttransduce many hormonal signals; rather, it acts as amodulator of many hormonal responses. Thus, G pro-teins might play an important role as “integrators ofcell status,” perhaps signaling something as general asmetabolic status or “homeostasis.” Global identifica-tion of proteins that physically interact with plant Gprotein components could help substantially in differ-entiating among these three models.

G protein mutants are not the only mutants to show alter-ations in responsiveness to several different hormones (65). In-deed, such “cross-talk” in hormone signaling may be the normrather than the exception in plants. For example, ABA-deficientmutants have been isolated on the basis of the phenotype of ger-mination insensitivity to the GA biosynthesis inhibitor paclobu-trazol (66), whereas some ethylene-insensitive mutants showenhanced sensitivity to inhibition of seed germination by ABA(67). Mutants in a PP2A regulatory subunit show phenotypesrelated to auxin, ABA, and ethylene signaling (43, 68, 69). Boththe second and third models can also apply to any proteins in-

volved in hormonal cross-talk, unless such proteins function di-rectly in hormone biosynthesis.

Limited Heterotrimeric Combinations Yield Different Outputsin Different Cell Types. From the information summarized above,it is clear that one hormone can have drastically different effectson different cell types. For example, GA opposes ABA action inseeds, but there is little evidence for GA responsiveness of guardcells; auxin inhibits lateral branching in shoots but promotes for-mation and growth of lateral roots. Cell-specific roles of G pro-teins apparently confer some of this specificity, as exemplified bythe fact that certain organs of the rice Gα mutant exhibit drastical-ly reduced GA sensitivity, whereas in other organs, sensitivity tothis hormone is essentially wild-type (50).

One way to achieve cellular specificity in G protein signal-ing would be by cell-specific expression of GPCRs. Only twocandidate GPCRs have been identified in plants to date, GCR1and RGS1, but there may be more GPCRs that are not identifi-able by sequence similarity searches. There also could be differ-ent levels of basal activation of a given GPCR in a given celltype, leading to differential susceptibility to endogenous ligandsof the various classes summarized in Fig. 2. In addition, protein-protein interactions and posttranslational modificationswill also affect GPCR activity status; as yet, these are essential-ly unexplored topics in plants. Finally, non–GPCR-type regula-tors of the heterotrimer are beginning to be uncovered in ani-

R E V I E W

A

GA

GA

GA

BBR

Heterotrimer activation

GPCR Heterotrimer activation Germination

Germination

High sensitivity

Low sensitivity

Non-GPCR

Non-GPCR

C

Heterotrimer activation

Germination

BRGA ABAGlucose

Fig. 3. Models consistent with G protein component mutant phenotypes in hor-monal regulation of seed germination. (A) A GPCR directly couples perceptionof GA to activation of the G protein heterotrimer in a high-sensitivity GA-sensing pathway, whereas a low-sensitivity GA-sensing pathway uses a differ-ent, non–GPCR-based pathway. (B) The heterotrimer does not couple percep-tion of GA but strengthens the promotive effects of BR on GA-stimulated ger-mination. (C) The heterotrimer does not directly couple to any hormone per-ception pathway, but its status modulates the strength of both stimulatory (forexample, GA and BR) and inhibitory (for example, glucose and ABA) inputs in-to seed germination. In all panels, dotted lines represent multistep pathways.


mals (13). These may exist in plants as well, but they would bedifficult to identify by bioinformatic approaches, because pro-teins with this function may share neither sequence nor struc-tural homology.

At the level of the heterotrimer, although there are only twoknown heterotrimeric combinations, it is possible that addition-al Gγ subunits will be uncovered, because sequence homologyis not a strong predictor of Gγ subunits (21). There is also diver-sity in the ways that the heterotrimer can couple with down-stream signals. In mammalian systems, some signals are trans-duced primarily through Gα and others through the Gβγ dimer,whereas some signal via both of these elements (1). Evidence isaccumulating that such diversity also holds true for plants; forexample, the modulation of auxin-regulated lateral root produc-tion appears to be mediated primarily by AGB1. Finally, patterns of expression of downstream effectors [of which only ahandful have been identified in plants to date (9)] can also con-fer specificity of response.

Thus, it may turn out that cell specificity of G protein–basedhormonal signaling in plants can be readily accounted for bycombinatorial specificity among all of the proteins that make upthe G protein signaling “super-pathway.” Clearly, much researchis required in this realm to uncover this postulated richness of Gprotein signaling in plants.

Why Aren’t Plant G Protein Component Knockout Mutations Lethal?Intuitively, one would expect that a gene knockout is less likelyto be lethal if there exist in the genome multiple copies of thegene, or closely related genes, or if the gene has a very restrict-ed pattern of expression. When multiple, closely related iso-forms are encoded, one expects that there may be true biochem-ical or “proximate” redundancy of function, whereas if gene ex-pression is restricted, especially to a nonessential part of theplant (trichomes, for example), one expects that the deficiencymay be surmountable. However, in the case of G protein com-ponent genes, the situation is exactly the opposite: In both Ara-bidopsis and rice, Gα and Gβ are each represented by singlegenes and only two Gγ genes have been identified. Moreover,these genes are widely expressed throughout the organism (7).Yet gene knockouts of plant G protein components do not leadto lethality, in contrast to some Gα mutants of yeast and mouse(70, 71) or to clear “genetic diseases” as seen in animal models(72, 73).

For a possible explanation to this puzzle, we may need notlook any further than the phenomenon of hormonal cross-talk inplants, already discussed above. Because input from many dif-ferent hormones (and other signals) is integrated into a givenoutcome, such as hypocotyl cell size or the decision to germi-nate, functional or ultimate redundancy may take the place ofbiochemical or proximate redundancy. In other words, there areprobably many paths that plant cell metabolism can take toachieve the same final outcome of a particular seed status orcell size. Therefore, when G protein signaling is genetically al-tered in the mutants, the presence of functional redundancy pro-vides at least a partial compensation that is sufficient to main-tain plant viability. This explanation is especially likely if Gproteins function according to the third model mentioned above(Fig. 3), where their input into such phenomena is not as prima-ry, directly coupled transducers of a specific hormonal ligand,but rather as modulators of a multihormone-based response.

Phenotypic Plasticity and G Proteins: A SpeculativeConclusionFrom the observation that none of the effects of plant G proteincomponent mutation as yet uncovered appear to be “life-threaten-ing,” one might reach the conclusion that G proteins do not play ascentral a role in plant physiology as they do in animal physiology.However, such a conclusion may be premature. The hallmark ofplant growth is phenotypic plasticity, defined as the ability of anygiven genotype to yield different phenotypes in response to differentenvironments (74, 75). It is argued that because plants are sessile,they have evolved increased plasticity over their mobile animalcounterparts, thus allowing improved acclimation to a changing butinescapable environment (76, 77). Because G protein componentmutation may eliminate inputs that contribute to the achievement offunctional redundancy, such mutations may reduce the possibilitythat the mutants will be able to alter their phenotype and acutely ac-climate to varying environmental conditions (Fig. 4A). Paradoxical-ly, a divergence from the wild-type phenotype under any one condi-tion could potentially signify a decrease in overall plasticity, if themutant phenotype is “locked in” over a range of conditions in whichthe wild type plastically responds. One potential example is therange of stature of the rice Gα subunit mutant d1: The dwarf d1 mu-tant shows a much smaller range of absolute plant height than donondwarf varieties. An extreme hypothetical example is depicted inFig. 4A, where the function describing the relationship betweenphenotype and environment (the “reaction norm”) of the mutant is ahorizontal line, indicating a complete lack of plasticity.

Alternatively, G protein mutation could affect phenotypicvariation [here defined as the range of phenotypes observedwithin any one environment, and sometimes referred to as “de-velopmental noise” (74)] but not the amount of plasticity (Fig.4, B and C). Whether a reduction in variation would ultimatelybe beneficial or detrimental would depend on whether the mu-tant phenotype converged on (Fig. 4B) or diverged from (Fig.4C) the “ideal” phenotype. A third possibility is that G proteinmutation does not affect the amount of either plasticity or phe-notypic variation, but simply shifts the reaction norm (Fig. 4D).Again, whether such a shift would be helpful or harmful woulddepend on whether the shift is toward or away from some idealphenotype.

In the absence of data to address the hypotheses of Fig. 4,why speculate that plant G proteins may be important for plasticity? The reason is inherent in the extremely pleiotropicnature of the G protein mutations themselves. With G proteinsaffecting so many diverse plant developmental and physiologi-cal responses, it seems reasonable to hypothesize that plasticityunder a range of environmental conditions may be affected,with possible attendant consequences for plant fitness.

Indeed, recent plant ecophysiological studies on other Ara-bidopsis mutants are revealing that the extent of plasticity of agiven genotype is an important component of the fitness equa-tion. For example, Arabidopsis mutants that lack photoreceptorsfor perception of specific light signals are at a disadvantage incertain environments (78). Therefore, it will be of interest toevaluate the plasticity, and ultimately the reproductive success,of G protein mutants under a range of environments, includingnonoptimal conditions. Such experiments may reveal vital con-tributions of plant G proteins to fitness after all.

Given the rate of anthropogenically based changes in the globalenvironment, plasticity may be one of the more important traits forplant species survival. The extent of plasticity of human phenotypic

R E V I E W


response, although much more limited than that of plants, has im-portant implications for human health (79). Thus, despite theirscarcity in “green genomes,” studies on plant G proteins and theirregulators may prove informative to a wide variety of disciplines.

References1. A. J. Morris, C. C. Malbon, Physiological regulation of G protein–linked

signaling. Physiol. Rev. 79, 1373–1430 (1999).2. M. N. P. C. H. Davies, Ed., Understanding G Protein–Coupled Receptors

and Their Role in the CNS (Oxford Univ. Press, Oxford, 2002).3. G protein–coupled receptors: The devil is in the detail [editorial]. Nat. Rev.

Drug Discov. 3, 539 (2004).4. A. Lania, G. Mantovani, A. Spada, G protein mutations in endocrine

diseases. Eur. J. Endocrinol. 145, 543–559 (2001).

5. A. M. Spiegel, Defects in G protein–coupled signal transduction in humandisease. Annu. Rev. Physiol. 58, 143–170 (1996).

6. H. Ma, GTP-binding proteins in plants: New members of an old family.Plant Mol. Biol. 26, 1611–1636 (1994).

7. S. M. Assmann, Heterotrimeric and unconventional GTP binding proteins inplant cell signaling. Plant Cell 14 (suppl.), S355–S373 (2002).

8. A. M. Jones, G protein–coupled signaling in Arabidopsis. Curr. Opin. PlantBiol. 5, 402–407 (2002).

9. A. M. Jones, S. M. Assmann, Plants: The latest model system for G proteinresearch. EMBO Rep. 5, 572–578 (2004).

10. T. Kenakin, Inverse, protean, and ligand-selective agonism: Matters of re-ceptor conformation. FASEB J. 15, 598–611 (2001).

11. F. Gbahou, A. Rouleau, S. Morisset, R. Parmentier, S. Crochet, J. S. Lin, X.Ligneau, J. Tardivel-Lacombe, H. Stark, W. Schunack, C. R. Ganellin, J. C.Schwartz, J. M. Arrang, Protean agonism at histamine H3 receptors in vitroand in vivo. Proc. Natl. Acad. Sci. U.S.A. 100, 11086–11091 (2003).

R E V I E W

A

Drought

Phen

oty

pe

Well-watered Waterlogged

B

Drought

Phen

oty

pe


C

Drought

Phen

oty

pe


D

Drought

Phen

oty

pe


Fig. 4. Hypothetical examples of how phenotypes ofwild-type plants (blue boxes) and G protein componentmutant plants (green boxes) might differ in terms of ei-ther plasticity (represented by shifts in the vertical posi-tion of the box across environments) or phenotypic varia-tion (represented by the dimensions of the box within anenvironment). The “ideal” phenotype is arbitrarily as-sumed to be the center point of the wild-type phenotypicresponse in each environment. (A) Wild-type and mutantplants exhibit the same amount of phenotypic variationwithin an environment, but the mutant has no phenotypicplasticity. (B) Wild-type and mutant plants exhibit thesame amount of phenotypic plasticity, but the mutant ex-hibits reduced phenotypic var iation within each environment. As illustrated, the mutant is converging onthe ideal phenotype and may thus be predicted to exhibitgreater fitness. (C) Wild-type and mutant plants exhibitthe same amount of phenotypic plasticity, but the mutantexhibits increased phenotypic variation within each envi-ronment as well as divergence from the “ideal” pheno-type. (D) Wild-type and mutant plants exhibit the sameamount of phenotypic plasticity as well as the sameamount of variation, but the mutant’s phenotypic re-sponse curve (reaction norm) is shifted away from the“ideal” phenotype.


12. The state of GPCR research in 2004 [editorial]. Nat. Rev. Drug Discov. 3,577–624 (2004).

13. M. J. Cismowski, A. Takesono, M. L. Bernard, E. Duzic, S. M. Lanier, Receptor-independent activators of heterotrimeric G proteins. Life Sci. 68,2301–2308 (2001).

14. D. R. Manning, Evidence mounts for receptor-independent activation ofheterotrimeric G proteins normally in vivo: Positioning of the mitotic spindlein C. elegans. Sci. STKE 2003, pe35 (2003).

15. J. A. Brzostowski, A. R. Kimmel, Signaling at zero G: G protein–independentfunctions for 7-TM receptors. Trends Biochem. Sci. 26, 291–297 (2001).

16. B. Vanderbeld, G. M. Kelly, New thoughts on the role of the βγ subunit in Gprotein signal transduction. Biochem. Cell Biol. 78, 537–550 (2000).

17. A. Ishikawa, H. Tsubouchi, Y. Iwasaki, T. Asahi, Molecular cloning and char-acterization of a cDNA for the α subunit of a G protein from rice. Plant CellPhysiol. 36, 353–359 (1995).

18. H. Ma, M. F. Yanofsky, E. M. Meyerowitz, Molecular cloning and characteri-zation of GPA1, a G protein α subunit gene from Arabidopsis thaliana.Proc. Natl. Acad. Sci. U.S.A. 87, 3821–3825 (1990).

19. C. A. Weiss, C. W. Garnaat, K. Mukai, Y. Hu, H. Ma, Isolation of cDNAs en-coding guanine nucleotide-binding protein β-subunit homologues frommaize (ZGB1) and Arabidopsis (AGB1). Proc. Natl. Acad. Sci. U.S.A. 91,9554–9558 (1994).

20. A. Ishikawa, Y. Iwasaki, T. Asahi, Molecular cloning and characterization ofa cDNA for the β subunit of a G protein from rice. Plant Cell Physiol. 37,223–228 (1996).

21. M. G. Mason, J. R. Botella, Completing the heterotrimer: Isolation and char-acterization of an Arabidopsis thaliana G protein γ-subunit cDNA. Proc.Natl. Acad. Sci. U.S.A. 97, 14784–14788 (2000).

22. M. G. Mason, J. R. Botella, Isolation of a novel G protein γ-subunit fromArabidopsis thaliana and its interaction with Gβ. Biochim. Biophys. Acta1520, 147–153 (2001).

23. C. Kato, T. Mizutani, H. Tamaki, H. Kumagai, T. Kamiya, A. Hirobe, Y. Fuji-sawa, H. Kato, Y. Iwasaki, Characterization of heterotrimeric G proteincomplexes in rice plasma membrane. Plant J. 38, 320–331 (2004).

24. H. Ullah, J. G. Chen, B. Temple, D. C. Boyes, J. M. Alonso, K. R. Davis, J. R.Ecker, A. M. Jones, The β-subunit of the Arabidopsis G protein negativelyregulates auxin-induced cell division and affects multiple developmentalprocesses. Plant Cell 15, 393–409 (2003).

25. Y. Iwasaki, T. Kato, T. Kaidoh, A. Ishikawa, T. Asahi, Characterization of theputative α subunit of a heterotrimeric G protein in rice. Plant Mol. Biol. 34,563–572 (1997).

26. H. S. Seo, C. H. Choi, S. Y. Lee, M. J. Cho, J. D. Bahk, Biochemical charac-teristics of a rice (Oryza sativa L., IR36) G protein α-subunit expressed inEscherichia coli. Biochem. J. 324, 273–281 (1997).

27. G. S. Aharon, A. Gelli, W. A. Snedden, E. Blumwald, Activation of a plantplasma membrane Ca2+ channel by TGα1, a heterotrimeric G protein αsubunit homologue. FEBS Lett. 424, 17–21 (1998).

28. J. G. Chen, F. S. Willard, J. Huang, J. Liang, S. A. Chasse, A. M. Jones, D. P.Siderovski, A seven-transmembrane RGS protein that modulates plant cellproliferation. Science 301, 1728–1731 (2003).

29. F. Apone, N. Alyeshmerni, K. Wiens, D. Chalmers, M. J. Chrispeels, G.Colucci, The G protein–coupled receptor GCR1 regulates DNA synthesisthrough activation of phosphatidylinositol-specific phospholipase C. PlantPhysiol. 133, 571–579 (2003).

30. X. Q. Wang, H. Ullah, A. M. Jones, S. M. Assmann, G protein regulation ofion channels and abscisic acid signaling in Arabidopsis guard cells.Science 292, 2070–2072 (2001).

31. L. Perfus-Barbeoch, S. M. Assmann, Reactive oxygen-species-dependentabscisic acid signaling in Arabidopsis guard-cells lacking the heterotrimericG protein α subunit, GPA1. Abstract, 13th International Workshop on PlantMembrane Biology. July 6-10, 2004, Montpellier, France.

32. J. Zhao, X. Wang, Arabidopsis phospholipase Dα1 interacts with the het-erotrimeric G protein α-subunit through a motif analogous to the DRY motifin G protein–coupled receptors. J. Biol. Chem. 279, 1794–1800 (2004).

33. Y. R. Lapik, L. S. Kaufman, The Arabidopsis cupin domain protein AtPirin1interacts with the G protein α-subunit GPA1 and regulates seed germina-tion and early seedling development. Plant Cell 15, 1578–1590 (2003).

34. K. L. Pierce, R. T. Premont, R. J. Lefkowitz, Seven-transmembrane receptors. Nat. Rev. Mol. Cell Biol. 3, 639–650 (2002).

35. L. G. Josefsson, L. Rask, Cloning of a putative G protein–coupled receptorfrom Arabidopsis thaliana. Eur. J. Biochem. 249, 415–420 (1997).

36. S. Pandey, S. M. Assmann, The Arabidopsis putative G protein–coupled re-ceptor GCR1 interacts with the G protein α subunit GPA1 and regulatesabscisic acid signaling. Plant Cell 16, 1616–1632 (2004).

37. J. G. Chen, S. Pandey, J. Huang, J. M. Alonso, J. R. Ecker, S. M. Assmann,A. M. Jones, GCR1 can act independently of heterotrimeric G protein in re-sponse to brassinosteroids and gibberellins in Arabidopsis seed germination. Plant Physiol. 135, 907–915 (2004).

38. J. Bockaert, J. P. Pin, Molecular tinkering of G protein–coupled receptors:An evolutionary success. EMBO J. 18, 1723–1729 (1999).

39. H. A. Rockman, W. J. Koch, R. J. Lefkowitz, Seven-transmembrane-spanning receptors and heart function. Nature 415, 206–212 (2002).

40. K. L. Pierce, R. J. Lefkowitz, Classical and new roles of β-arrestins in theregulation of G protein–coupled receptors. Nat. Rev. Neurosci. 2, 727–733(2001).

41. R. Schulz, The pharmacology of phosducin. Pharmacol. Res. 43, 1–10(2001).

42. E. M. Ross, T. M. Wilkie, GTPase-activating proteins for heterotrimeric Gproteins: Regulators of G protein signaling (RGS) and RGS-like proteins.Annu. Rev. Biochem. 69, 795–827 (2000).

43. A. DeLong, K. Mockaitis, S. Christensen, Protein phosphorylation in the deliveryof and response to auxin signals. Plant Mol. Biol. 49, 285–303 (2002).

44. B. J. Dekkers, J. A. Schuurmans, S. C. Smeekens, Glucose delays seedgermination in Arabidopsis thaliana. Planta 218, 579–588 (2004).

45. J. Christadoulou, R. J. Weaver, R. M. Pool, Relation of gibberellin treatmentto fruit-set, berry development and cluster compactness in Vitis viniferagrapes. Proc. Am. Soc. Hort. Sci. 92, 301–310 (1968).

46. H. Ullah, J. G. Chen, S. Wang, A. M. Jones, Role of a heterotrimeric G pro-tein in regulation of Arabidopsis seed germination. Plant Physiol. 129,897–907 (2002).

47. K. A. Lease, J. Wen, J. Li, J. T. Doke, E. Liscum, J. C. Walker, A mutant Arabidopsis heterotrimeric G protein β subunit affects leaf, flower, and fruitdevelopment. Plant Cell 13, 2631–2641 (2001).

48. M. Ashikari, J. Wu, M. Yano, T. Sasaki, A. Yoshimura, Rice gibberellin-insen-sitive dwarf mutant gene Dwarf 1 encodes the α-subunit of GTP-bindingprotein. Proc. Natl. Acad. Sci. U.S.A. 96, 10284–10289 (1999).

49. Y. Fujisawa, T. Kato, S. Ohki, A. Ishikawa, H. Kitano, T. Sasaki, T. Asahi, Y.Iwasaki, Suppression of the heterotrimeric G protein causes abnormalmorphology, including dwarfism, in rice. Proc. Natl. Acad. Sci. U.S.A. 96,7575–7580 (1999).

50. M. Ueguchi-Tanaka, Y. Fujisawa, M. Kobayashi, M. Ashikari, Y. Iwasaki, H.Kitano, M. Matsuoka, Rice dwarf mutant d1, which is defective in the αsubunit of the heterotrimeric G protein, affects gibberellin signal transduction. Proc. Natl. Acad. Sci. U.S.A. 97, 11638–11643 (2000).

51. G. J. Bishop, C. Koncz, Brassinosteroids and plant steroid hormone signaling. Plant Cell 14 (suppl.), S97–S110 (2002).

52. H. Ma, Plant G proteins: The different faces of GPA1. Curr. Biol. 11,R869–R871 (2001).

53. G. Yang, M. Matsuoka, Y. Iwasaki, S. Komatsu, A novel brassinolide-enhanced gene identified by cDNA microarray is involved in the growth ofrice. Plant Mol. Biol. 52, 843–854 (2003).

54. B. C. Tan, S. H. Schwartz, J. A. Zeevaart, D. R. McCarty, Genetic control ofabscisic acid biosynthesis in maize. Proc. Natl. Acad. Sci. U.S.A. 94,12235–12240 (1997).

55. C. D. Rock, Pathways to abscisic acid-regulated gene expression. NewPhytol. 148, 357–396 (2000).

56. S. M. Assmann, in Plant Hormones: Biosynthesis, Signal Transduction, Action!, P. J. Davies, Ed. (Kluwer Academic, Dordrecht, Netherlands,2005), pp. 391–412.

57. S. Pandey, J.-G. Chen, A. M. Jones, S. M. Assmann, unpublished data.58. S. Coursol, L. M. Fan, H. Le Stunff, S. Spiegel, S. Gilroy, S. M. Assmann,

Sphingolipid signalling in Arabidopsis guard cells involves heterotrimeric Gproteins. Nature 423, 651–654 (2003).

59. L.-M. Fan, J.-G. Chen, A. M. Jones, S. M. Assmann, unpublished data.60. S. Spiegel, S. Milstien, Sphingosine-1-phosphate: An enigmatic signalling

lipid. Nat. Rev. Mol. Cell Biol. 4, 397–407 (2003).61. X. Fu, N. P. Harberd, Auxin promotes Arabidopsis root growth by modulat-

ing gibberellin response. Nature 421, 740–743 (2003).62. O. Leyser, Regulation of shoot branching by auxin. Trends Plant Sci. 8,

541–545 (2003).63. I. Casimiro, T. Beeckman, N. Graham, R. Bhalerao, H. Zhang, P. Casero, G.

Sandberg, M. J. Bennett, Dissecting Arabidopsis lateral root development.Trends Plant Sci. 8, 165–171 (2003).

64. H. Ullah, J. G. Chen, J. C. Young, K. H. Im, M. R. Sussman, A. M. Jones,Modulation of cell proliferation by heterotrimeric G protein in Arabidopsis.Science 292, 2066–2069 (2001).

65. S. Gazzarrini, P. McCourt, Cross-talk in plant hormone signalling: WhatArabidopsis mutants are telling us. Ann. Bot. 91, 605–612 (2003).

66. K. M. Leon-Kloosterziel, M. A. Gil, G. J. Ruijs, S. E. Jacobsen, N. E.Olszewski, S. H. Schwartz, J. A. Zeevaart, M. Koornneef, Isolation andcharacterization of abscisic acid-deficient Arabidopsis mutants at two newloci. Plant J. 10, 655–661 (1996).

67. N. Beaudoin, C. Serizet, F. Gosti, J. Giraudat, Interactions between abscisicacid and ethylene signaling cascades. Plant Cell 12, 1103–1115 (2000).

68. J. M. Kwak, J. H. Moon, Y. Murata, K. Kuchitsu, N. Leonhardt, A. DeLong, J.I. Schroeder, Disruption of a guard cell-expressed protein phosphatase 2Aregulatory subunit, RCN1, confers abscisic acid insensit ivity inArabidopsis. Plant Cell 14, 2849–2861 (2002).

69. P. B. Larsen, J. D. Cancel, Enhanced ethylene responsiveness in the Ara-bidopsis eer1 mutant results from a loss-of-function mutation in the protein

R E V I E W


phosphatase 2A regulatory subunit, RCN1. Plant J. 34, 709–718 (2003).70. B. H. Spain, D. Koo, M. Ramakrishnan, B. Dzudzor, J. Colicelli, Truncated

forms of a novel yeast protein suppress the lethality of a G protein α sub-unit deficiency by interacting with the β subunit. J. Biol. Chem. 270,25435–25444 (1995).

71. M. Jiang, G. Boulay, K. Spicher, M. J. Peyton, P. Brabet, L. Birnbaumer, U.Rudolph, Inactivation of the Gα i2 and Gαo genes by homologous recombination. Receptors Channels 5, 187–192 (1997).

72. K. Hosoda, R. E. Hammer, J. A. Richardson, A. G. Baynash, J. C. Cheung,A. Giaid, M. Yanagisawa, Targeted and natural (piebald-lethal) mutations ofendothelin-B receptor gene produce megacolon associated with spottedcoat color in mice. Cell 79, 1267–1276 (1994).

73. R. M. Chemelli, J.T.Willie, C. M. Sinton, J. K. Elmquist, T. Scammell, C. Lee, J. A.Richardson, S. C. Williams, Y. Xiong, Y. Kisanuki, T. E. Fitch, M. Nakazato, R. E.Hammer, C. B. Saper, M. Yanagisawa, Narcolepsy in orexin knockout mice:Molecular genetics of sleep regulation. Cell 98, 437–451 (1999).

74. M. Pigliucci, Phenotypic Plasticity: Beyond Nature and Nurture (Tech. Rep.0-8018-6788-6, Johns Hopkins Univ. Press, Baltimore, MD, 2001).

75. S. E. Sultan, Phenotypic plasticity for plant development, function and lifehistory. Trends Plant Sci. 5, 537–542 (2000).

76. A. Trewavas, Aspects of plant intelligence. Ann. Bot. 92, 1–20 (2003).77. S. Gilroy, A. Trewavas, Signal processing and transduction in plant cells:

The end of the beginning? Nat. Rev. Mol. Cell Biol. 2, 307–314 (2001).78. M. Pigliucci, J. Schmidt, Genes affecting phenotypic plasticity in

Arabidopsis: Pleiotropic effects and reproductive fitness of photomor-phogenic mutants. J. Evol. Biol. 12, 551–562 (1999).

79. P. Bateson, D. Barker, T. Clutton-Brock, D. Deb, B. D’Udine, R. A. Foley, P.Gluckman, K. Godfrey, T. Kirkwood, M. M. Lahr, J. McNamara, N. B. Met-calfe, P. Monaghan, H. G. Spencer, S. E. Sultan, Developmental plasticityand human health. Nature 430, 419–421 (2004).

80. L. Perfus-Barbeoch, A. M. Jones, S. M. Assmann, Plant heterotrimeric Gprotein function: Insights from Arabidopsis and rice mutants. Curr. Opin.Plant Biol. 7, 719–731 (2004).

81. G protein research in my laboratory is supported by NSF grant MCB-0209711 and U.S. Department of Agriculture grant 2003-35304-13924. Iacknowledge S. Sultan (Wesleyan University) and the publications of T.Trewavas for providing me with insight into plant plasticity. This review isdedicated to Fakhri A. Bazzaz, an ecologist who always helped me to seethe “big picture,” on the occasion of his retirement from the Department ofOrganismic and Evolutionary Biology at Harvard University.

Citation: S. Assmann, Plant G proteins, phytohormones, and plasticity:Three questions and a speculation. Sci. STKE 2004, re20 (2004).

R E V I E W

plant g proteins, phytohormones, and plasticity: three ...€¦ · the genomes of diploid...

Documents