Physiologia Plantarum 2012 Copyright © Physiologia Plantarum 2012, ISSN 0031-9317

Root proteases: reinforced links between nitrogen uptakeand mobilization and drought toleranceAjay Kohlia,∗, Joan Onate Narcisoa, Berta Mirob and Manish Raoranea

aPlant Breeding, Genetics, and Biotechnology Division, International Rice Research Institute, DAPO, Metro Manila, PhilippinesbCrop and Environmental Sciences Division, International Rice Research Institute, DAPO Box 7777, Metro Manila, Philippines

Correspondence*Corresponding author,e-mail: a.kohli@cgiar.org

Received 22 June 2011;revised 15 November 2011

doi:10.1111/j.1399-3054.2012.01573.x

Integral subcellular and cellular functions ranging from gene expression,protein targeting and nutrient supply to cell differentiation and cell deathrequire proteases. Plants have unique organelles such as chloroplastscomposed of unique proteins that carry out the unique process ofphotosynthesis. Hence, along with proteases common across kingdoms,plants contain unique proteases. Improved knowledge on proteases canlead to a better understanding of plant development, differentiation anddeath. Because of their importance in multiple processes, plant proteasesare actively studied. However, root proteases specifically are not as wellstudied. The associated rhizosphere, organic matter and/or inorganic mattermake roots a difficult system. Yet recent research conclusively demonstratedthe occurrence of endocytosis of proteins, peptides and even microbes byroot cells, which, hitherto known for specialized pathogenesis or symbiosis,was unsuspected for nutrient uptake. These results reinforced the importanceof root proteases in endocytosis or root exudate-mediated nutrient uptake.Rhizoplane, rhizosphere or in planta protease action on proteins, peptidesand microbes generates sources of nitrogen, especially during abiotic stressessuch as drought. This article highlights the recent research on root proteasesfor nitrogen uptake and the connection of the two to drought-tolerancemechanisms. Drought-induced proteases in rice roots, as known from riceexpression databases, are discussed for future research on certain M50,Deg, FtsH, AMSH and deubiquitination proteases. The recent emphasis onlinking drought and plant hydraulics to nutrient metabolism is illustratedand connected to the value of a systematic study of root proteases in cropimprovement.

Abbreviations – AA, amino acid; ABA, abscissic acid; AMSH, associated molecule with the SH3 domain of STAM;ATP, adenosine triphosphate; BNF, biological nitrogen fixation; C, carbon; ClpP1, caseinolytic protease P1; CP, cysteineprotease; Deg, degradation of periplasmic protein; ER, endoplasmic reticulum; FC, field capacity; FtsH, filamentationtemperature-sensitive H; GFP, green fluorescent protein; IPT, isopentenyltransferase; LHT, lysine histidine transporter; MT,membrane-tethered; N, nitrogen; OAA, oxaloacetate; OTU, ovarian tumor; PAGE, polyacrylamide gel electrophoresis; PDZ,post synaptic density protein (PSD95), drosophila disc large tumor suppressor (DlgA) and zonula occludens-1 protein (zo-1);PSII, photosystem II; PTR, peptide transporter; RIP, regulated intramembrane proteolysis; SREBP, sterol-regulated elementbinding protein; SUMO, small ubiquitin-related modifier; TF, transcription factor; Ulp, ubiquitin-like protease; UPR, unfoldedprotein response.

Physiol. Plant. 2012

Introduction

Proteases cleave peptide bonds, leading to protein break-down. Thus, the general perception on biological func-tions of proteases was largely restricted to degradationof toxic, undesirable and damaged proteins, releasingamino acids (AAs) that could be reused in protein syn-thesis. However, proteases are now known to regulatevarious biological processes spanning the life cycle ofplants, starting with cell division to specialized tissue,organ or organelle formation and development such asgametes, embryo, seed coat, stomata and chloroplast(Van der Hoorn 2008). Their activity in response toexternal stimuli such as pest and pathogen attack (Xia2004, Baek and Choi 2008) or in response to internaldevelopmental stimuli such as senescence has also beencharacterized (Woltering 2004).

Sequenced genomes of Arabidopsis and rice revealclose to 700 and 800 proteases, respectively, which areclassified into clans and families, while members withinthe families also exhibit functional variation (Van derHoorn 2008). Plants with larger genomes are likely tohave a higher number of proteases. However, the bio-logical function of only around 40 proteases is known.Considering the involvement of proteases in variousaspects of plant growth, development, survival andsenescence, a better understanding of their phenotypiceffects and molecular mechanisms is imperative. Thisis especially true for plant-specific members of somefamilies of proteases such as the plastidial proteases. Forexample, chloroplast biogenesis and function involvethe chloroplast genome-encoded caseinolytic proteaseP1 (ClpP1) whose mutation arrests etiolated shoot devel-opment in tobacco plants (Kuroda and Maliga 2003).Nuclear-encoded Clp proteases could not complementClpP1 mutation, thus suggesting that ClpP1 performsmore than the housekeeping protein breakdown func-tion. Kuroda and Maliga (2003) speculated that ClpP1may act on a specific regulatory protein. Similarly, theplastidial filamentation temperature-sensitive H (FtsH)protease complex is involved in the critical function ofturnover of the photosystem II (PSII) reaction center D1protein. It is also involved in other processes required forthe development and maintenance of the photosyntheticapparatus (Liu et al. 2010). Studies on plant proteaseshave advanced substantially over the past few years andan ever-increasing understanding of the role of proteasesin the life cycle of plants is emerging. The regulatoryrole of plant proteases in post-translational modifica-tion of proteins by limited proteolysis at specific sites isbeing recognized as a key route to involvement in differ-ent functions. Limited and site-specific proteolysis givesrise to functional enzymes and regulatory proteins and

peptides, facilitates subcellular targeting, and is requiredfor protein assembly (Schaller 2004). In a recent review,Van der Hoorn (2008) collated information on 40 plantproteases with known phenotypes and described someof the well-characterized proteases representing differentfunctional aspects in different cellular processes. Onlytwo of the plant proteases described were concernedwith roots. One affected root meristem maintenance inArabidopsis (Casamitjana-Martinez et al. 2003) and theother had a role in Medicago truncatula root infectionby Sinorhizobium meliloti (Combier et al. 2007). In thelast 3 years, only one more protease in roots has sincebeen assigned a phenotype, a cysteine protease (CP)that is a negative regulator of nodules and bacteroidsenescence in Astragalus sinicus (Li et al. 2008). A rel-ative lack of detailed characterization of root proteasesis perhaps indicative of the difficulties associated withworking with roots under natural environments.

Molecular and genetic studies on roots have largelyconcentrated on root development, morphology andarchitecture (reviewed by Rost 2011). The other popu-larly explored aspect of roots is the root–microbe sym-biosis for nitrogen fixation. Murray (2011) and Yokotaand Hayashi (2011) reviewed the genes involved inroot–microbe symbiosis in legumes. Traditionally, rootshave been considered little more than a conduit for waterand dissolved minerals. Genetic controls and molecularbiology of mineral acquisition by roots, including bothmacronutrients such as nitrogen, phosphorus, potas-sium, sulfur and calcium and micronutrients such asiron, molybdenum, boron, copper, manganese, zinc,chloride, silicon, sodium and nickel, etc., are nowbeing elucidated. Kraiser et al. (2011) presented an inte-grated view on different mechanisms that help plantsto acquire and maintain an adequate nitrogen sup-ply. These included root-specific gene expression andfunction modulation, root growth and developmentmodulation and root–microbe association modulation.Similarly, the state of knowledge on root uptake anduse of phosphorus (Panigrahy et al. 2009), potassium(Szczerba et al. 2009) and sulfate (Rouached et al.2009) has also been reviewed, while Chen et al. (2008)reviewed membrane transporters involved in the uptakeof the three macronutrients, i.e. nitrogen, phosphate andpotassium.

The three main aspects of root research – root develop-ment and symbiotic and non-symbiotic nutrient acqui-sition – were reviewed for an integrated understandingof root adaptations to biotic and abiotic constraints onroot function and plant health (Hodge et al. 2009). Suchintegrative studies on nutrient acquisition are increasingdue to an imperative need to optimize nutrient applica-tion, uptake and use as an important aspect of precision

Physiol. Plant. 2012

farming in the future so as to increase land and plantyield potential under normal and/or stress conditions.Improving the yields of major cereal crops is a key tofuture food security under population and climatic pres-sure. As cereals do not depend on nutrient availabilitythrough symbiosis, an improved understanding of non-symbiotic nutrient uptake and mobilization is desirable.Interestingly, recent reports on two modes of nutrientuptake have added to our knowledge on the mech-anisms that roots employ for the purpose. One is rootexudate-mediated nutrient uptake, which was known forsome time, but more details of exudate components andtheir roles are emerging now. The second is endocytosisof proteins and intact bacteria by root cells. Endocyto-sis by roots was known in connection with symbiotic orpathogenic relationships, but clear demonstrations of theprocess being used for nutrient acquisition have beenprovided recently. Both mechanisms invoke a crucialrole for root proteases, exuded or internal, in order tobreak down, use and remobilize the scavenged AAs,peptides and proteins and thus maintain the nitrogenstatus of plants.

The nitrogen status of plants under stress is a criticalfactor in stress tolerance. Under water-deficit conditions,for example, nitrogen content can decline drastically(Gonzalez-Dugo et al. 2010). A corollary to this is thatincreased root proteases would have a critical role instress tolerance because they can mobilize nitrogenfrom external or internal sources. This is somewhat adifferent view on the role of plant proteases under stress,specifically wheat CPs, whereby an increase in theirstress-associated expression in leaves was proposed as amarker of susceptibility to water-deficit (Simova-Stoilovaet al. 2010). A similar study conducted on the roots ofred clover did not show an increase in CPs till after21 days of stress, after which the roots were terminallyconsigned to senescence (Webb et al. 2010). Neverthe-less, preliminary results from our laboratory combinedwith extensive information gleaned from rice expressiondatabases clearly implicate a role for rice root proteasesin compensating for the nitrogen stress concomitant todrought. It is argued that proteases that are seen to bespecifically upregulated in roots under drought belongto two categories, cell death and nutrient mobiliza-tion/regulation. It is further suggested that during droughtthe balance may shift more toward the latter categoryin tolerant accessions of rice, conferring a better abilityto uptake and remobilize nitrogen under stress. Such ashift from vacuolar degradative protease activity to non-vacuolar energy-dependent regulatory protease activitywas shown in wheat leaves in relation to developmentalstage and water saturation deficit (Wisniewski and Zag-danska 2001), although total protease activity was not

substantially different. The importance of this shift understress was not established and such observations havenot been made in roots yet.

We suggest that further research on stress-responsiveproteases in roots and their specific roles would addvaluable information on integrating the root–leaf and/orroot–tiller response in terms of hydraulics and nutrientre-mobilization. This would be particularly useful forrice, whose roots seem to survive long after the aerialparts have wilted. The return of favorable conditionssupports a fast-tracked growth of new tillers that couldbear seeds. This is reminiscent of red clover roots surviv-ing for a long time after defoliation or grazing (Binghamand Rees 2008, Webb et al. 2010).

Root proteases and nitrogen uptake

The root–soil interface is an active site for the exchangeof organic and inorganic C and N, the two majorelements of plant nutrition, whereas other macro-and micronutrients are also sourced from the rhizo-sphere–soil continuum. The C and N exchange at theroots driven by variables of soil, microbial biomassand plants is highly complex and is spatiotemporallyregulated. Jones et al. (2009) presented four possibleutilities for such exchanges, two of which centeredon root–root and root–microbe interactions and theother two on nutrient uptake and distribution by plants.For root–microbe interaction between Medicago andSinorhizobiummeliloti, peptidases were found to be animportant component of the root exudates (De-la-Penaand Vivanco 2010). Interestingly, however, Godlewskiand Adamczyk (2007) found that proteases were secretedby the roots of 15 different agricultural and wild plantspecies without any microbial or other artificial meansof inducing protease secretion. Their studies with wheat(Adamczyk et al. 2008) and Allium (Adamczyk et al.2009) suggested that such protease secretion was part ofthe plants’ strategy to increase free AAs as a source of N.Thus, AAs in soil could result from the activity of suchexuded root proteases on proteins and peptides outsideof the roots. Alternately, increased root protease activitywithin the roots under stress could be the source of AAsin the soil, again as part of the root exudate. Indeed, itwas established that increased root exudates of Festucarubra under stress were richer in AAs (Paterson et al.2003), while Lesuffleur et al. (2007) reported that rootexudates of different plants under stress were rich indifferent AAs. These and other studies (Schobert andKomor 1987, Persson and Nasholm 2002) lend credi-bility to the proposition by Jones et al. (2009) that onepossible utility of exudates is re-absorption of some oftheir components by roots to maintain C and N balance.

Physiol. Plant. 2012

That roots do uptake AAs from soil and use them invivo was demonstrated through a combined proteomicand isotope tracer approach, whereby labeled glycinesupplied as an N source was traced in plant metabo-lites in Lolium perenne (Thornton et al. 2007). In thisstudy, root proteome analysis with and without glycineexhibited upregulation of a glycine metabolism enzyme,among others, in the former case.

The proteases in the rhizosphere may equally comefrom microbes. Vagnerova and Macura (1974) demon-strated that protease activity because of microbes washigher in the wheat rhizoplane than in the rhizosphereand surrounding soil. However, plant root exudatesexercise a selective influence so that only a small andspecific proportion of microbial diversity can inhabit therhizosphere and/or the rhizoplane, and the rhizosphereinhabitants may vary depending also on environmen-tal and plant growth variables (Hartmann et al. 2009).Thus, proteases in root exudates or within the roots orthose from rhizoplane non-symbiotic bacteria may beresponsible for protein breakdown, facilitating N uptakein the form of AAs. Indeed, a root–microbe symbioticassociation is another mechanism of organic N avail-ability to plants. Although there was evidence that plantscould use AA from soil or culture media, these wereartificially supplied. Paungfoo-Lonhienne et al. (2008)demonstrated that axenically cultivated non-symbioticHakeaactites (a woody plant) as well as the herbaceousArabidopsis could use protein as a source of nitrogenin the absence of any other symbiotic or non-symbioticorganism. Their results validated a role for root proteasesin the absence of any other microbe, when the root sur-face fluoresced from proteolysis of an externally suppliedprotein–chromophore complex capable of fluorescenceonly upon proteolysis. More importantly, it was alsoshown with the use of green fluorescent protein (GFP)and through the lack of protein degradation productsin the incubation media that intact protein was takenup by the roots as long as root hair were present. Thesame group further elegantly demonstrated that rootsof tomato and Arabidopsis were capable of taking upintact Escherichia coli and yeast cells, which were thendegraded within the root cells and were at least a sourceof nitrogen, as detected by an increase in 15N concentra-tions in the leaves after incubation of tomato roots with15N-labeled E. coli (Paungfoo-Lonhienne et al. 2010).Such consumption of E. coli N clearly implicated rootproteases. There was evidence through induction of spe-cific genes and structures that the process of taking upintact microbes by roots involves root outgrowths thatsurround E. coli cells stuck to the root surface throughmucilage secreted either from the roots or from the bacte-ria. Endocytosis was also implicated as a possible process

that could facilitate microbe uptake because inductionof genes for cytoskeleton structure and re-organizationwas noticed. The importance of endocytosis in differentaspects of plant–microbe interactions, microbe recogni-tion and colonization of plant cells by endosymbiontshas been recently reviewed with a special empha-sis on clathrin-mediated endocytosis (Leborgne-Castelet al. 2010).

These results clearly linking N uptake and use toroot proteases, apart from the uptake of inorganic N orsymbiotic N in plants, are from studies over the past4 years only on a limited number of species. Althoughnot much information is available on any such parallelsin crop plants in field conditions, the stage is set for sucha search.

Drought and nitrogen uptake

The importance of the relationship between drought andN uptake by plants is becoming steadily clear. Droughtinvariably reduces biological nitrogen fixation (BNF) andconsequently the N content, biomass and yield of plants.In peanut, BNF and biomass increasingly declined under2/3 of field capacity (FC) water and 1/3 of FC for 12different lines (Pimratch et al. 2008). However, therewas a positive correlation between BNF and biomassunder both water-deficit conditions and the relationshipwas stronger under severe water-deficit. The resultssuggested that maintenance of high BNF under droughtwould be useful in maintaining high yield under water-deficit conditions. Indeed, this approach was earlierused to select Glycine max genotypes with high abilityto fix N under drought, which led to the release ofdrought-tolerant varieties (Chen et al. 2007). However,it is not only the symbiosis-dependent plants that copebetter with drought under improved N content, but alsothe symbiosis-independent plants. For example, in rice,various physiological parameters such as photosynthesis,relative chlorophyll content, water use efficiency,transpiration and overall growth were significantlyhigher with the application of N during drought thanunder drought with no N application (Suralta 2010).Similarly, under drought stress applied to barley at threegrowth stages, tillering, shooting and earing, the negativeeffect on grain yield could be relatively relieved with Nfertilizer application (Krcek et al. 2008).

Reviewing water-deficit and nitrogen nutrition ofcrops, Gonzalez-Dugo et al (2010) elaborated thatmineral N from the rhizosphere is largely made availablethrough the transpiration stream and that under water-deficit mineral N fluxes decrease in the xylem. Theconcomitant N assimilation is also more sensitive towater scarcity. Why does increased N, organic or

Physiol. Plant. 2012

inorganic, through transpiration pull or through exudatesand/or endocytosis, help to relieve the effects of drought?Plants under stress tend to maintain the internal nutrientstatus of the C–N ratio either by modulating thepartitioning of nutrients or by nutrient re-mobilization.Vegetative growth may be controlled to limit the tissuesthat require balanced nutrient status. For example,Norway spruce under drought but with an N sourceavailable exhibited less new shoot development thanthis species under drought, but with no N sourceavailable (Rosengren-Brinck and Nihlgard 1995). Thissuggested preferential allocation of available N tomaintain the nutrient status of a limited body of the plant.In rice, reproductive tiller formation may be favoredover new leaf development under drought and with Navailability.

Under progressive drought, an imbalance in the inter-nal nutrient status of aerial parts is more likely to occurbefore terminal water-deficit per se. This is because theperception of water-deficit initiates hydraulic and hor-monal changes that restrict stomatal opening, in turnrestricting transpiration pull and carbon dioxide entry,both of which initiate changes in C and N metabolismand assimilation. In tomato, root restriction stress, whichmimics water stress, affected xylem sap ABA concentra-tion, which in turn directly affected aerial characteristicssuch as stomatal conductance, leaf water potential,intercellular CO2 concentration, accumulation of car-bohydrates, maximum carboxylation rate of Rubisco,ribulose-1,5-bisphosphate regeneration and quantumyield of PSII electron transport (Shi et al. 2008). Hence,under drought, along with setting in motion the phys-iological and molecular machinery for efficient wateruptake and use, the plant’s effort is also toward acti-vating mechanisms that ensure efficient nutrient uptake,re-mobilization and use. For nutrient uptake and mobi-lization by roots the activity of exuded or internal rootmay be critical. Proteases in the aerial parts may alsobe activated for the purpose of leaf N re-mobilization.However, root proteases are likely to play a more impor-tant role as being at the site of N uptake. Our preliminaryresults discussed below demonstrate much higher pro-tease activity in the roots of different rice genotypes thanin the shoots, both with and without drought.

Relation between root proteases, nutrientstatus and drought

Between ammonium and nitrate as an N source underdrought, rice showed a preference for the former. Adirect link with increased photosynthesis in the presenceof ammonium nutrition was noticed (Guo et al. 2007).Rice is special in that flooded field and drought situation

change the main source of N from ammonium tonitrate, as also the degree of non-symbiotic BNF, inturn influencing the relative importance of ammoniumand nitrate (Buresh et al. 2008). However, uptake ofAAs by plant roots as additional or alternate source ofN through AA transporters has been known (Nasholmet al. 2009). Interestingly, it was shown by carbon andnitrogen isotope labelling that wheat roots took upAAs in the form of tripeptides in preference to nitratesand single AAs. Uptake of nitrates was inferior to thatof ammonium, L-trialanine and even the biologicallyunutilized D-trialanine (Hill et al. 2011). Also, asillustrated above, peptides, proteins and/or microbesof the rhizosphere/rhizoplane could be used as a sourceof N under drought and root proteases may have animportant role in hydrolyzing soil proteins or in digestingthose taken up by roots.

Drought-induced proteases are known from the aerialparts of plants such as Arabidopsis (Koizumi et al. 1993),pea (Jones and Mullet 1995), Phaseolus (Hieng et al.2004) and wheat (Zagdanska and Wisniewski 1996,1998, Demirevska et al. 2008, Zang et al. 2010). Suchinduction correlates positively with drought-sensitiveor drought-tolerant cultivars. This confusion exists ifdetailed characterization of the proteases involved isnot available because senescence-related proteases aswell as nutrient re-mobilization and regulatory proteasesboth can be induced under drought. It is possible that theformer class of proteases is induced more in drought-sensitive genotypes while the latter class is inducedmore in tolerant genotypes. For example, increasedprotease induction in drought-sensitive genotypes wasnoticed in Phaseolus (Roy-Macauley et al. 1992, Hienget al. 2004), while in wheat the drought-tolerant vari-eties exhibited increased induction (Zagdanska andWisnievski 1998, Demirevska et al. 2008). Importantly,the proteases studied in wheat were ATP-dependentregulatory proteases while in Phaseolus they were vac-uolar serine and/or CPs. For wheat under water-deficit,total protease activity increased and CPs largely con-tributed to the increase (Zagdanska and Wisnievski1996). While Simova-Stoilova et al. (2006) associatedincreased CP activity with sensitive wheat genotypes,Zang et al. (2010) recently associated another wheatCP (TaCP) with drought tolerance through transgenicvalidation in Arabidopsis, whereby constitutive expres-sion of TaCP enhanced the drought tolerance of thetransgenic Arabidopsis. Indeed, different CPs are asso-ciated with senescence and drought (Khanna-Chopraet al. 1999) and both categories, degradative and pro-cessive CPs, can be upregulated. Depending on whichCPs were characterized, the association with tolerant orsusceptible genotypes could vary.

Physiol. Plant. 2012

Despite evidence for the importance of intracellularproteolysis during drought, most likely in re-organizingcellular metabolism, the role of specific proteases understress remains uncertain. This is even more the casefor root proteases. The limited information available onroot proteases is mostly from work on isolated root tipsand largely in relation to C starvation. For example,a root protease upregulated in maize root tips underglucose starvation was described (Chevalier et al. 1995)and it was shown that protein catabolism in maize roottips followed fatty acid catabolism under C starvation(Dieuaide-Noubhani et al. 1997). In general, endopepti-dase and protease expression increased in excised maizeroot tips under C-starvation conditions (Brouquisse et al.2001, 2007). Sugar starvation in tomato roots led to areduction in root carbohydrate as well as in root proteincontent (Devaux et al. 2003) implicating increased pro-tease activity in the latter case. In the case of intact rootsof Phaseolus vulgaris under stress, protease activity didnot increase in the first 3 days of stress and AA levelsdecreased, including that of asparagine (Bathellier et al.2009). However, intact roots within 3 days of stress mayhave still not reached a point of critical stress more easilyachieved in excised roots as exemplified by the increasein AAs after excision of red clover roots from shoots(Bingham and Rees 2008). Interestingly, asparagine syn-thetase – the enzyme for synthesis of asparagine, themajor N transport molecule in plants during stress – washighly upregulated in Arabidopsis under sugar stressimposed by darkness (Gaufichon et al. 2010). This sug-gests that during C starvation the plant prepares for Nre-mobilization.

What is the relationship between drought, C starvationand protease upregulation? The water deficit-mediatedhydraulic signal from roots to shoots results in changes inleaf cell turgor leading to hormonal signaling-mediatedstomatal closure (Christmann et al. 2007). Stomatal clo-sure is useful in limiting transpiration under drought, butit also limits carbon dioxide intake, which in turn affectsnet photosynthesis (McDowell et al. 2008, Pinheiro andChaves 2011). A change in photosynthesis leads tothe initiation of drought-induced C starvation, whichresults in a sugar-mediated signaling cascade for there-organization of cellular metabolism. Overall, drought-induced C starvation is seen to be the major threat toplant survival concomitant to changes in C–N ratioand in plant hydraulics (McDowell, 2011). Recent find-ings suggest that a number of late molecular responsesare more to changes in carbohydrate availability than toenvironmental cues (Smith and Stitt 2007) and that plantsmay survive some period of complete loss of conductivity(McDowell 2011). Additionally, during drought, nutri-ent and mineral uptake decline (Sardans and Penuelas

2005), transport to the aerial parts decreases (Huand Schmidhalter 2005) and energy costs of convert-ing nitrogen into usable forms increase (Farooq et al.2009, Pinheiro and Chaves 2011). Under such nitrogen-deficient conditions, the action of proteases may sup-ply the AAs, such as asparagine and glutamine forN re-mobilization. Additionally, root exudate-mediatedand/or root endocytosis-mediated uptake of water, nutri-ents, organic molecules and microbes can provide alifeline to the plant, depending on the timing and sever-ity of drought. It is to be noted that the organic moleculessourced through the activity of proteases are not just Nsources, but also C sources as observed with asparaginebreakdown into oxaloacetate (OAA) through the activ-ity of the amidohydrolase asparaginase. The OAA canbe used in the Krebs cycle, which produces the reduc-ing power as hydrogen atoms for the electron transportchain and intermediates that can be further anabolized(Simpson 2011). Additional root growth observed undernutrient deficiency was proposed to be important insourcing additional soil area for nutrient uptake (Lopez-Bucio et al. 2003). Hence, drought-induced root growthis opposite to shoot growth retardation under drought.Nutrients for root growth are sourced as organic matterfrom the soil, thus saving energy in converting inorganicN into an AA (Nasholm et al. 2009). Another source ofwater and nutrients is the process of autophagy. Cellsrespond to stress with a burst of autophagic activitythrough co-ordinated upregulation of autophagy genesat the onset of starvation in Arabidopsis and proteasesare part of the autophagic gene complement (Rose et al.2006). A lack of these proteases makes Arabidopsismutants hypersensitive to C and N starvation (Thompsonet al. 2005). Indeed, root tips of Arabidopsis undergoingwater stress exhibited autophagy (Duan et al. 2010).

During drought, nutrients and even water sourcedthrough different mechanisms by roots, either from out-side or within, need to be transported to the aerialparts. The uptake of AAs and their transfer into theroot xylem have been described in Ricinus communis(Schobert and Komor 1990), and the corresponding AAtransporters in this plant have been characterized (Bicket al. 1998). Overall importance of organic N (AAs,peptides and proteins) uptake by roots and associatedtransporters were reviewed by Nasholm et al. (2009)and Tegeder and Rentsch (2010). Nitrogenous nutri-ents in xylem can flow to phloem and to the shoots(Atkins 2000, Zhang et al. 2010) through specific trans-porters, including peptide transporters (PTRs) (Lubkowitz2011). Proton-coupled, independent or pH-gradient-driven AA and sugar influx, efflux and bidirectionaltransporters at the plasma membrane and tonoplasthave been described (Yang et al. 2010, Okumoto and

Physiol. Plant. 2012

Pilot 2011). The lysine histidine transporter (LHT1) is ahigh affinity broad-spectrum AA transporter expressedin the leaf and root epidermis and it directly affectsAA uptake and distribution in plants (Hirner et al.2006). Upregulation of LHT1 was recently shown indrought-tolerant rice plants generated by overexpressionof the cytokinin synthesis gene, isopentenyltransferase(IPT), under the stress- and maturation-inducedpromoter PSARK (Peleg et al. 2011). Root-specific expres-sion of the Arabidopsis LHT1, amino acid permease(AtAAP) type, proline and glycine betaine (ProT) andPTR among others, under abiotic and biotic stresses, wasrecently reviewed by Tegeder and Rentsch (2010).

These studies on the relationship between changingnutrient status under drought suggest that maintaining thenutrient status of plants may improve drought tolerance.Indeed, survival of maize plants under drought byaddition of exogenous sugars revealed that the maincause for tissue mortality was not low water potential,but low carbon availability (Boyle et al. 1991). Ourcontention is that root proteases play an importantrole in nutrient uptake and mobilization through theprocesses discussed above, to maintain the criticalnutrient balance as long as possible and thus contributeto drought tolerance. We support this hypothesis throughanalyzing the rice expression database for the expressionof proteases in roots under drought at the criticaldevelopmental stages, as described below.

Expression variation in rice root proteasesunder drought

For total soluble protein extracts following differentextraction protocols, total protease activity was appar-ently higher in rice roots than in shoots as seen bythe extent of degradation of high-molecular-weight pro-teins after polyacrylamide gel electrophoresis (PAGE)-mediated separation (Raorane et al. 2011). Fig. 1 showsmore degradation of proteins if leaf and root tissue (L/R)proteins from one genotype were extracted togetherthan when only leaf tissue (L) was used. The L/R sampleswere far more degraded than L samples to begin with andbecame progressively more degraded with time. That theroot extract per se was responsible for the degradation ofleaf proteins was demonstrated by leaf extract degrada-tion on the addition of root extract (Raorane and Narciso,unpublished data). The addition of a protease inhibitorcocktail in the extraction buffer did not substantiallyalter the scenario (Raorane et al. 2011). This was truefor plants at the 15-day-old seedling stage as well as attillering stage for 12 different rice genotypes under nor-mal non-stress conditions in hydroponics and soil-grownplants (Raorane and Narciso, unpublished data). Under

drought stress imposed on 15-day-old seedlings of the 12rice genotypes varying for drought tolerance, preliminaryresults suggested a trend toward higher total root proteaseactivity in the tolerant genotypes. This was also noticedfor salinity treatment whereby a tolerant genotype suchas Pokkali exhibited stronger upregulation of proteasescompared to the intolerant IR29 (Miro and Raorane,unpublished data). It was not clear if this was becauseof more proteases in the roots of more tolerant plants.To obtain an idea on whether or not that could be thecase, the rice expression database (www.ricearray.org/)was queried for changes in expression pattern of 672proteases in roots and leaves at tillering and panicleelongation stages under control and drought conditions(Experiment GSE26280: Genome-wide temporal–spatialgene expression profiling of drought responsiveness inrice; Wang et al. 2011). The two stages were chosenfor comparison because of recent trends in screeningfor reproductive stage drought tolerance (Serraj et al.2009). Fig. 2 shows that in leaves and roots at thetillering or panicle development stage under droughtstress, 32 proteases were upregulated twofold or morein comparison to the control plants. Of these, 11 wereuniquely upregulated in roots, 5 at tillering stage and6 at panicle elongation stage. Similarly, 15 proteaseswere unique to leaves, 2 at tillering stage and 13 atpanicle elongation stage. Upregulation of one proteasewas common to roots at both stages of development

M L1 L/R1 L2 L//R2 L3 L/RR3

Fig. 1. PAGE of total soluble protein of leaves (L) or leaves and roots(L/R) after incubation at 37◦C for 10, 20 and 30 min (1, 2 and 3),respectively. The leaf extracts (L1, L2 and L3) are not affected whilea progressive degradation of proteins with time is obvious in leaf androot extracts (L/R1, L/R2 and L/R3), especially in the region between thearrows.

Physiol. Plant. 2012

2

LT: 6

To

2

otal: 21

1

5

1

1

RT: 8

6

1

LE: 1

RE: 1

8

Total:

1315

12

: 20

Fig. 2. Rice proteases showing more than twofold induction underdrought. RT, root at tillering stage; RE, root at panicle elongation stage;LT, leaves at tillering stage; LE, leaves at panicle elongation stage. Intotal, 21 proteases (15 + 6) were upregulated in leaves and 20 (12 +8) in roots. Six proteases were common to more than one box and, ofthose, only one was common to leaves and roots at both developmentalstages.

and the other 5 were upregulated in roots and leavesboth at the same or different stages of development(Fig. 2). Only one protease was more than twofoldupregulated in all four tissues under drought. Thiswas an ATP-dependent La protease. The importanceof such regulatory proteases in wheat under droughtwas described earlier (Zagdanska and Wisnievski1998, Demirevska et al. 2008) and a distinction wasmade between the common vacuolar degradativeproteases and the regulatory processive proteases. Inour analysis, rice leaves exhibited a larger numberof uniquely upregulated proteases. These comprisedseveral aspartic proteases, including nepenthesin and thedeubiquitination CPs such as a small ubiquitin-relatedmodifier (SUMO) protease, an ovarian tumor (OTU)-likeprotease and four ubiquitin-like protease (Ulp) proteases.These proteases serve a regulatory function by activatingthe proteins that are deubiquitinated, including histones(H2B), which affects the transcriptional activation ofgenes such as those involved in flowering (Schmitzet al. 2009). Similarly, two serine carboxypeptidaseswere uniquely upregulated in leaves, which alsogenerally help in protein maturation through posttranslational modifications. Other proteases were non-specific subtilisins and CPs most likely involvedin general protein degradation. In roots also, acouple of unique nepenthesins, subtilisins, serinecarboxypeptidases, a Ulp protease and a CP (EP-B1) wereupregulated. However, more than twofold upregulation

of an M50 protease, a degradation of periplasmic protein(Deg) protease and an FtsH protease was unique to roots.In fact, these proteases were among the most highlyupregulated ones in roots.

The M50 proteases are a class of membrane-boundmetalloproteases. Their active sites lie in transmembranedomains that participate in regulated intramembraneproteolysis (RIP). They perform site-specific proteolysiswithin the membranes, of other membrane-tethered (MT)proteins, to liberate the cytosolic fragment for signalingor transcription (Brown et al. 2000). RIP proteases arespecial in that they apparently use water to hydrolyze thepeptide bond despite the hydrophobic intramembraneenvironment (Wolfe and Kopan 2004). RIP can driveprocesses ranging from differentiation, development,sterol/lipid metabolism and response to unfoldedproteins. Drought conditions do induce an unfoldedprotein response (UPR) pathway (Zhang et al. 2008). RIPparticipates in this induction by liberating the cytosolicfragment of the MT transcription factor (TF) bZIP60and bZIP28, which induce the transcription of protein-folding chaperones, called binding proteins [(BiP);Iwata and Koizumi (2005), Liu et al. (2007)]. Also, theconnection between drought, sterol levels in roots andstomatal responses has been established in Arabidopsis(Pose et al. 2009). Changes in root sterol levels can affectthe sterol-regulated element binding protein (SREBP) TF.Interestingly, RIP proteases such as M50 are implicated inSREBP site 2 proteolysis (S2P) to activate the transcriptionfactor (Rawson et al. 1997). It is not difficult to furtherimagine a relationship between sterol level and the UPRdue to changes in sterols affecting changes in glycans,in turn affecting protein glycosylation, and thus leadingto ER stress-mediated induction of UPR during drought.The upregulation of an M50 RIP protease in roots underdrought thus seems logical and important.

The Deg proteases are ATP-independent serineendopeptidases, which contain one or more [post synap-tic density protein (PSD95), drosophila disc large tumorsuppressor (DlgA) and zonula occludens-1 protein (zo-1)] (PDZ) protein–protein interaction domains (Kiesel-bach and Funk 2003). The PDZ domain regulates theprotease activity and oligomerization of Deg proteases.Deg proteases display dual function, as general proteasesand as specific regulatory proteases in signal transduc-tion or protein maturation (chaperones). This variationmay be attributed to the monomer and oligomer con-figuration respectively. Also, Deg proteases can eitherhave a broad substrate range or highly specific sub-strates. An Arabidopsis glyoxysomal Deg protease wasspecifically involved in the maturation of the glyoxyso-mal malate dehydrogenase (Helm et al. 2007). Degproteases have not been characterized in rice and

Physiol. Plant. 2012

root-specific or drought-inducible Deg proteases are notknown. However, a Deg protease was earlier shownto be relevant to protein quality control in chloroplasts(Huesgen et al. 2009) through a role in repairing UV-Bdamage of PSII (Nixon et al. 2010). In this particularfunction the Deg proteases had a complementary rolewith the FtsH proteases, which are ATP-dependent, Zn-metalloproteases. Both Deg and FtsH proteases are ratherknown as plastidial proteases in leaves and not in root.An FtsH protease was characterized to be responsible forthe variegated phenotype through affecting chloroplastdevelopment (Pogson and Albrecht 2011). Therefore,root-specific upregulation of rice Deg and FtsH pro-teases under drought warrants attention. It is likely thattheir function in degrading unassembled membrane pro-teins and/or oxidatively damaged proteins makes themrelevant in drought as part of the UPR pathway similar tothe M50 and rhomboid proteases. It is noteworthy thatproteases highly and uniquely upregulated in rice rootsunder drought, although belonging to different classes,may be concerned with the RIP and UPR pathways bothof which can either affect the C–N pool under stressby internal proteolysis and/or by facilitating externalnutrient uptake as discussed above or affect protein pro-cessing and maturation for signaling and transcription.

Additionally, our in silico analysis revealed someproteases which, although not twofold upregulatedunder drought, were specifically induced in responseto drought in both leaves and roots at all develop-mental stages assessed by Wang et al. (2011) (Fig. 3,Table 1). Most of these were similar to proteases dis-cussed above for example, a SUMO protease, a Degprotease, an SCP, an OTU-like CP, and an M16 domaincontaining Zinc peptidase similar to the M50 peptidase.However, there were four unique proteases, a rhomboid(serine) protease, an AMSH-like protease, a signal pep-tide peptidase and a mitochondrial processing peptidase.Rhomboid proteases are also a part of the RIP arsenaland are substrate-specific (Kanaoka et al. 2005). Thepotential role of these proteases under drought assumesmore importance with the knowledge that plants canendocytose large organic molecules and microbes(Paungfoo-Lonhienne et al. 2010), which invokesmucilage-mediated attraction and receptor-mediatedendocytosis of molecules/microbes. Both of these canbe imagined to be affected by root RIP proteases upreg-ulated under drought. Interestingly, the AMSH proteaseis closely associated with clathrin-mediated endosomaltransport of nutrients through symplastic or apoplas-tic movements from one part of the plant to another.The other two peptidases are also processing peptidasesthat enable protein trafficking to the correct subcellularcompartments.

Fig. 3. Drought-induced upregulation of a rhomboid protease desig-nated by the oligo-array element, in leaves and roots at differentdevelopmental stages.

An arbitrary comparative estimate of an increasein protease activity contributed by the twofold ormore upregulated proteases in roots and leaves (Fig. 2)was obtained by the sum of fold increases in roots(20 proteases upregulated 52.57-fold) and leaves (21proteases upregulated 54.69-fold). The result that ahigher number of proteases were upregulated and toa larger extent in leaves than in roots was contrary tothe observation of higher and more potent root proteaseactivity. This suggests that root proteases may be highlydegradative and/or processive. The FtsH protease fromSynechocystis is indeed highly processive (Komendaet al. 2007) and such an attribute is likely retained inrice root proteases.

Root proteases and crop improvement

Increasing biomass and seed yield is mostly the ultimategoal of a large body of agricultural research. However,plant growth and productivity are directly and stronglyaffected by edaphic and other environmental factors,most of which are perceived by, responded to andeffected on roots. Yet, root responses at the physiologicaland/or molecular levels are poorly understood. Althoughit is becoming increasingly appreciated that optimalplant growth and productivity can best be achieved by

Physiol. Plant. 2012

Table 1. Proteases specifically induced in response to drought in both leaves and roots at all developmental stages assessed under experimentGSE26280 by Wang et al. (2011)

Array element ID Gene ID Gene name

Os.10283.1.S1_at LOC_Os04g01240 Serine-type peptidaseOs.11236.1.S1_at LOC_Os02g12650 Puromycin-sensitive aminopeptidaseOs.13615.1.S1_at LOC_Os01g47262 AMSH-like protease, putative, expressedOs.4124.1.S1_a_at LOC_Os01g51390 Mitochondrial-processing peptidaseOs.4146.1.S1_at LOC_Os01g25370 SUMO proteaseOs.46745.1.S1_at LOC_Os02g57710 Signal peptide peptidase-like 2BOs.53632.1.S1_at LOC_Os05g13370 OsRhmbd12–putative rhomboid proteaseOs.6594.1.S1_at LOC_Os11g24510 OsSCP56–serine carboxypeptidaseOs.8776.1.S1_at LOC_Os03g64219 OTU-like cysteine protease family proteinOsAffx.24819.1.S1_at LOC_Os02g50880 OsDegp3–putative Deg proteaseOsAffx.24834.1.S1_x_at LOC_Os02g52390 M16 domain containing zinc peptidase

better understanding and leveraging the role of roots,several issues are emerging in how best to strengthen thedesirable traits of roots while avoiding their undesirabletraits for overall crop improvement (Herder et al. 2010).

Recent studies have highlighted unconventionalsources and routes of nutrient uptake by roots, while

strong links between drought tolerance and nutrientstatus have also been highlighted above. These studiessuggest a central role for root proteases not only innutrient uptake, but also in water uptake throughprocessing aquaporins for their role in maintainingplant water potential (Sade et al. 2009, Maurel et al.

Fig. 4. A schematic model for the role of proteases in proteinaceous nutrition uptake through different modes by root cells at the soil interface.Left side (beige) represents the cell cytoplasm, cell membrane ad cell wall, while the right side represents the soil. Plant proteases (brown globules)and/or microbial proteases (yellow globules) act on bacteria (red rods) and proteins (green zigzags) in the soil to generate peptides (small green rods)and/or amino acids (green AA). Similarly such peptides and AAs are generated within the plant cell by the action of endogenous plant proteaseson endogenous proteins. Additionally, root exudates (evaginated pustule at cell–soil interface) may comprise plant proteins and peptides furtherdegraded at the rhizoplane or in the soil into peptides and AAs by exuded plant proteases or microbial proteases. It is known that peptides andAAs serve as sources of carbon and nitrogen through further breakdown within the plant cell. Peptide and AA uptake is well known. However,under drought conditions, apparently intact proteins and complete bacterial and yeast cells are also taken up by the root cells through the process ofendocytosis (invaginated pustule). Under well-watered conditions root exudates and other process may facilitate most part of the breakdown throughmicrobial action in the soil, thus enabling re-uptake of simpler peptides and AAs. However, under drought conditions most breakdown and uptakemay happen at the rhizoplane or within the cells, facilitated by endocytosis of larger proteins and cells.

Physiol. Plant. 2012

2010). Yet, very limited knowledge exists on theidentity and function of root proteases under normalor stress conditions. A better understanding of rootproteases will lead to designing robust strategies forcrop improvement by using proteases as gene-basedmarkers for genotyping and population surveys orthrough transgenic approaches. Regulatory proteasescan play an important part in crop improvementbecause they, like transcription factors and kinases,can affect multiple cascades through spatio-temporal-specific protein processing, leading perhaps to a burst ofresponse without necessarily involving transcription fora part of the response.

Although this article concentrated on establishing thelinks between root proteases and water and nutrientuptake and mobilization during drought stress, proteasesfunctional in root growth, branching and surface mayhave an effect on aerial tissues. Root apical and lateralmeristem activity and root vascular development arecontrolled by Cle peptides, which are the result ofspecific processing proteases (Ni et al. 2010). Processingof proteins facilitating symplastic movement or xylem tophloem transporters may also be achieved by proteases.

Studies on proteases in general and root proteases inparticular are also important to differentiate betweendesirable and undesirable proteases under specificconditions. An understanding of root function for waterand nutrient uptake and mobilization, which is inclusiveof studies on root proteases, would fast-track ourphysiological and molecular understanding of stresstolerance. This would lead to informed strategies forcrop improvement under conditions of adverse climateand shrinking agricultural lands and help meet thegoals of adequate food and nutrition for the increasingpopulation.

Conclusions and perspectives

Root proteases may have a role in water and nutrientuptake and mobilization, the quality and quantity of rhi-zodeposition, microbial population modulation and rootfunction as a source and sink for nutrients. Increasing evi-dence suggests that root proteases may be an importantaspect of root development, differentiation and stresssignaling. Various structural and functional classes ofproteases make it difficult to generalize the role of pro-teases in a process, for example, drought-tolerant plantsmay exhibit upregulation of some proteases and down-regulation of others. Tissue-specific protease isozymesalso exhibit functional diversity, for example, chloroplastproteases such as Deg, Clp, FtsH, etc. are upregulated inroots, where they may not at all be connected with pro-cessing photosynthesis-related proteins. Such inherent

diversity and the difficulty of working with roots undernatural conditions pose a challenge in characterizingthis important class of proteins. It is certain, however,that making substantial headway in understanding andaffecting above-ground tissues toward crop improve-ment will critically rest on our understanding of roots.In that regard, root proteases will be only a part ofthe sum, but a very important part, since their capac-ity to regulate cascades of reactions places them at parwith other protein modification systems such as kinases.Recent literature clearly makes a case for the importanceof root proteases in drought tolerance. Fig. 4 presentsa schematic version of the role of proteases in nutrientuptake especially under drought conditions. Drought isa major bottleneck in food security at present and evenmore so in the climate change-driven future scenariosof agriculture. It is clear that addressing drought tol-erance only in terms of water uptake, movement anduse efficiency will be only half the understanding ondrought tolerance because nutrient status imbalancesunder drought apparently affect the plant sooner thanthe lack of water. Root proteases can positively affectnutrient and water status more than any other class ofproteins under drought where anabolic processes in theabove-ground source tissues start suffering even beforecritically low levels of water are reached. Research ondrought tolerance must therefore put equal if not moreemphasis on the physiology and molecular biology of thenutrient status of plants under drought and as a corollaryon proteases in general and root proteases in particular.

Acknowledgements – We thank Drs Akshaya KumarBiswal, Abdelbagi Ismail, Roland Buresh and StephanHaefele of IRRI for critical comments and suggestions onimproving the manuscript. We thank Dr Bill Hardy of IRRIfor improving the manuscript through meticulous editorialcorrections.

References

Adamczyk B, Godlewski M, Smolandera A, Kitunen V(2009) Degradation of proteins by enzymes exuded byAllium porrum roots – a potentially important strategyfor acquiring organic nitrogen by plants. Plant PhysiolBiochem 47: 919–925

Adamczyk B, Godlewski M, Zimny J, Zimny A (2008)Wheat (Triticumaestivum) seedlings secrete proteasesfrom the roots and, after protein addition, grow well onmedium without inorganic nitrogen. Plant Biol 10:718–724

Atkins CA (2000) Biochemical aspects of assimilatetransfers along the phloem path: N-solutes in lupins.Aust J Plant Physiol 27: 531–537

Physiol. Plant. 2012

Baek KH, Choi D (2008) Roles of plant proteases inpathogen defense. Plant Pathol J 24: 367–374

Bathellier C, Tcherkez G, Mauve C, Bligny R, Gout E,Ghashghaie J (2009) On the resilience of nitrogenassimilation by intact roots under starvation, as revealedby isotopic and metabolomic techniques. RapidCommun Mass Spectrom 23: 2847–2856

Bick JA, Neelam A, Hall JL, Williams LE (1998) Amino acidcarriers of Ricinus communis expressed during seedlingdevelopment: molecular cloning and expressionanalysis of two putative amino acid transporters,RcAAP1 and RcAAP2. Plant Mol Biol 36: 377–385

Bingham IJ, Rees RM (2008) Senescence and N releasefrom clover roots following permanent excision of theshoot. Plant Soil 303: 229–240

Boyle MG, Boyer JS, Morgan PW (1991) Stem infusion ofliquid culture medium prevents reproductive failure ofmaize at low water potential. Crop Sci 31: 1246–1252

Brouquisse R, Evrard A, Rolin D, Raymond P, Roby C(2001) Regulation of protein degradation and proteaseexpression by mannose in maize root tips: Pisequestration by mannose may hinder the study of itssignaling properties. Plant Physiol 125: 1485–1498

Brouquisse R, Rolin D, Cortes S, Gaudillere M, Evrard A,Roby C (2007) A metabolic study of the regulation ofproteolysis by sugars in maize root tips: effects ofglycerol and dihydroxyaceton. Planta 225: 693–709

Brown MS, Ye J, Rawson RB, Goldstein JL (2000)Regulated intramembrane proteolysis: a controlmechanism conserved from bacteria to humans. Cell100: 391–398

Buresh RJ, Reddy KR, van Kessel C (2008) Nitrogentransformations in submerged soils. In: Schepers JS,Raun WR (eds) Nitrogen in Agricultural Systems.Agronomy Monograph 49. ASA, CSSA, and SSSA,Madison, WI, pp 401–436

Casamitjana-Martinez E, Hofhuis HF, Xu J, Liu CM,Heidstra R, Scheres B (2003) Root-specific CLE19overexpression and the sol1/2 suppressors implicatea CLV-like pathway in the control of Arabidopsisroot meristem maintenance. Curr Biol 13:1435–1441

Chen P, Sneller CH, Purcell LC, Sinclair TR, King TR,Ishibashi CAT (2007) Registration of soybean germplasmlines R01-416F and R01-581F for improved yield andnitrogen fixation under drought stress. J Plant Regist 1:167–168

Chen YF, Wang Y, Wu WH (2008) Membrane transportersfor nitrogen, phosphate and potassium uptake in plants.J Integr Plant Biol 50: 835–848

Chevalier C, Bourgeois E, Pradet A, Raymond P (1995)Molecular cloning and characterization of six cDNAsexpressed during glucose starvation in excised maize(Zea mays L.) root tips. Plant Mol Biol 28: 473–485

Christmann A, Weiler EW, Steudle E, Grill E (2007) Ahydraulic signal in root-to-shoot signalling of watershortage. Plant J 52: 167–174

Combier JP, Vernie T, de Billy F, El Yahyaoui F, Mathis R,Gamas P (2007) The MtMMPL1 early nodulin is a novelmember of the matrix metalloendoproteinase familywith a role in Medicago truncatula infection bySinorhizobium meliloti. Plant Physiol 144: 703–716

De-la-Pena C, Vivanco JM (2010) Root-microbeinteractions: the importance of protein secretion. CurrProteomics 7: 265–274

Demirevska K, Simova-Stoilova L, Vassileva V, VasevaIGrigorova B, Feller U (2008) Drought-induced leafprotein alterations in sensitive and tolerant wheatvarieties. Gen Appl Plant Physiol 34: 79–102

Devaux C, Baldet P, Joubes J, Dieuaide-Noubhani M,Just D, Chevalier C, Raymond P (2003) Physiological,biochemical and molecular analysis of sugar-starvationresponses in tomato roots. J Exp Bot 54: 1143–1151

Dieuaide-Noubhani M, Canioni P, Raymond P (1997)Sugar-starvation-induced changes of carbon metabolismin excised maize root tips. Plant Physiol 115:1505–1513

Duan YF, Zhang WS, Li B, Wang YN, Li K, SodmergenHan C, Zhang Y, Li X (2010) An endoplasmic reticulumresponse pathway mediates programmed cell death ofroot tip induced by water stress in Arabidopsis. NewPhytol 186: 681–695

Farooq M, Wahid A, Kobayashi N, Fujita D, Basra SMA(2009) Plant drought stress: effects, mechanisms andmanagement. Agron Sustain Dev 29: 185–212

Gaufichon L, Reisdorf-Cren M, Rothstein SJ, Chardon F,Suzuki A (2010) Biological functions of asparaginesynthetase in plants. Plant Sci 179: 141–153

Godlewski M, Adamczyk B (2007) The ability of plants tosecrete proteases by roots. Plant Physiol Biochem 45:657–664

Gonzalez-Dugo V, Durand JL, Gastal F (2010) Waterdeficit and nitrogen nutrition of crops. Agron SustainDev 30: 529–544

Guo SW, Chen G, Zhou Y, Shen QR (2007) Ammoniumnutrition increases photosynthesis rate under waterstress at early development stage of rice (Oryza sativaL.). Plant Soil 296: 115–124

Hartmann A, Schmid M, van Tuinen D, Berg G (2009)Plant-driven selection of microbes. Plant Soil 321:235–257

Helm M, Luck C, Prestele J, Hierl G, Huesgen PF, FrohlichT, Arnold GJ, Adamska I, Gorg A, Lottspeich F, Gietl C(2007) Dual specificities of the glyoxysomal/peroxisomalprocessing protease Deg15 in higher plants. Proc NatlAcad Sci USA 104: 11501–11506

Herder GD, Isterdael GV, Beeckman T, Smet ID (2010)The roots of a new green revolution. Trends Plant Sci15: 600–607

Physiol. Plant. 2012

Hieng B, Ugrinovic K, Sustar-Vozlic J, Kidric M (2004)Different classes of proteases are involved in theresponse to drought of Phaseolus vulgaris L. cultivarsdiffering in sensitivity. J Plant Physiol 161: 519–530

Hill PW, Quilliam RS, DeLuca TH, Farrar J, Farrell M,et al. (2011) Acquisition and Assimilation of Nitrogen asPeptide-Bound and D-Enantiomers of Amino Acids byWheat. PLoS ONE 6(4): e19220. doi:10.1371/journal.pone.0019220

Hirner A, Ladwig F, Stransky H, Okumoto S, Keinath M,Harms A, Frommer WB, Koch W (2006) ArabidopsisLHT1 is a high-affinity transporter for cellular aminoacid uptake in both root epidermis and leaf mesophyll.Plant Cell 18: 1931–1946

Hodge A, Berta G, Doussan C, Merchan F, Crespi M(2009) Plant root growth, architecture and function.Plant Soil 321: 153–187

Hu Y, Schmidhalter U (2005) Drought and salinity: acomparison of their effects on the mineral nutrition ofplants. J Plant Nutr Soil Sci 168: 541–549

Huesgen PF, Schuhmann H, Adamsk I (2009) Deg/HtrAproteases as components of a network for photosystem IIquality control in chloroplasts and cyanobacteria. ResMicrobiol 160: 726–732

Iwata Y, Koizumi N (2005) An Arabidopsis transcript ionfactor, AtbZIP60, regulates the endoplasmic reticulumstress response in a manner unique to plants. Proc NatlAcad Sci USA 102: 5280–5285

Jones DL, Nguyen C, Finlay RD (2009) Carbon flow in therhizosphere: carbon trading at the soil-root interface.Plant Soil 321: 5–33

Jones JT, Mullet JE (1995) A salt-inducible anddehydrationinducible pea gene, Cyp15a, encodes acell-wall protein with sequence similarity to cysteineproteases. Plant Mol Biol 28: 1055–1065

Kanaoka M, Urban S, Freeman M, Okada K (2005) AnArabidopsis rhomboid homolog is an intramembraneprotease in plants. FEBS Lett 579: 5723–5728

Khanna-Chopra R, Srivalli B, Ahlawat YS (1999) Droughtinduces many forms of cysteine proteases not observedduring natural senescence. Biochem Biophys ResCommun 255: 324–327

Kieselbach T, Funk C (2003) The family of Deg/HtrAproteases: from Escherichia coli to Arabidopsis. PhysiolPlant 119: 337–346

Koizumi M, Yamaguchi-Shinozaki K, Tsuji H, Shinozaki K(1993) Structure and expression of two genes thatencode distinct drought-inducible cysteine proteinasesin Arabidopsis thaliana. Gene 129: 175–182

Komenda J, Tichy M, Prasil O, Knoppova J, Kuvikova S, deVries R, Nixon PJ (2007) The exposed N-terminal tail ofthe D1 subunit is required for rapid D1 degradationduring photosystem II repair in Synechocystissp PCC6803. Plant Cell 19: 2839–2854

Kraiser T, Gras DE, Gutierrez AG, Gonzalez B,Gutierrez RA (2011) A holistic view of nitrogenacquisition in plants. J Exp Bot 62: 1455–1466

Krcek M, Slamka P, Olsovska K, Brestic M, Bencikova M(2008) Reduction of drought stress effect in spring barley(Hordeumvulgare L.) by nitrogen fertilization. Plant SoilEnviron 54: 7–13

Kuroda H, Maliga P (2003) The plastid clpP1 protease geneis essential for plant development. Nature 425: 86–89

Leborgne-Castel N, Adam T, Bouhidel K (2010)Endocytosis in plant-microbe interactions. Protoplasma247: 177–193

Lesuffleur F, Paynel F, Bataille MP, Le Deunff E, Cliquet JB(2007) Root amino acid exudation: measurement of highefflux rates of glycine and serine from six different plantspecies. Plant Soil 294: 235–246

Li YX, Zhou L, Li YG, Chen DS, Tan XJ, Lei L, Zhou JC(2008) A nodule-specific plant cysteine proteinase,AsNODF32, is involved in nodule senescence andnitrogen fixation activity of the green manure legumeAstragalus sinicus. New Phytol 180: 185–192

Liu JX, Srivastava R, Che P, Howell SH (2007) Anendoplasmic reticulum stress response in Arabidopsis ismediated by proteolytic processing and nuclearrelocation of a membrane associated transcriptionfactor, bZIP28. Plant Cell 19: 4111–4119

Liu X, Yu F, Rodermel S (2010) Arabidopsis chloroplastFtsH, var2 and suppressors of var2 leaf variegation: areview. J Integr Plant Biol 52: 750–761

Lopez-Bucio J, Cruz-Ramirez A, Herrera-Estrella L (2003)The role of nutrient availability in regulating rootarchitecture. Curr Opin Plant Biol 6: 280–287

Lubkowitz M (2011) The oligopeptide transporters: a smallgene family with a diverse group of substrates andfunctions? Mol Plant 4: 407–415.

Maurel C, Simonneau T, Sutka M (2010) The significanceof roots as hydraulic rheostats. J Exp Bot 61: 3191–3198

McDowell NG (2011) Mechanisms linking drought,hydraulics, carbon metabolism, and vegetationmortality. Plant Physiol 155: 1051–1059

McDowell NG, White S, Pockman WT (2008)Transpiration and stomatal conductance across a steepclimate gradient in the southern Rocky Mountains.Ecohydrology 1: 193–204

Murray JD (2011) Invasion by invitation: rhizobialinfection in legumes. Mol Plant Microbe Interact 24:631–639

Nasholm T, Kielland K, Ganeteg U (2009) Uptake oforganic nitrogen by plants. New Phytol 182: 31–48

Ni J, Guo Y, Jin H, Hartsell J, Clark SE (2010)Characterization of a CLE processing activity. Plant MolBiol 75: 67–75

Nixon PJ, Michoux F, Yu J, Boehm M, Komenda J (2010)Recent advances in understanding the assembly andrepair of photosystem II. Ann Bot 106: 1–16

Physiol. Plant. 2012

Okumoto S, Pilot G (2011) Amino acid export in plants: amissing link in nitrogen cycling. Mol Plant 4: 453–463.

Panigrahy M, Rao DN, Sarla N (2009) Molecularmechanisms in response to phosphate starvation in rice.Biotechnol Adv 27: 389–397

Paterson E, Thornton B, Sim A, Pratt S (2003) Effects ofdefoliation and atmospheric CO2 depletion on nitrateacquisition, and exudation of organic compounds byroots of Festuca rubra. Plant Soil 250: 293–305

Paungfoo-Lonhienne C, Lonhienne TGA, Rentsch D,Robinson N, Christie M, Webb RI, Gamage HK,Carroll BJ, Schenk PM, Schmidt S (2008) Plants can useprotein as a nitrogen source without assistance fromother organisms. Proc Natl Acad Sci USA 105:4524–4529

Paungfoo-Lonhienne C, Rentsch D, Robatzek S, Webb RI,Sagulenko E, Nasholm T, Schmidt S, Lonhienne TGA(2010) Turning the table: plants consume microbes as asource of nutrients. PLoS One 5: e11915

Peleg Z, Reguera M, Tumimbang E, Walia H, Blumwald E(2011) Cytokinin-mediated source/sink modificationsimprove drought tolerance and increase grain yield inrice under water-stress. Plant Biotechnol J 9: 747–758

Persson J, Nasholm T (2002) Regulation of amino aciduptake in conifers by exogenous and endogenousnitrogen. Planta 215: 639–644

Pimratch S, Jogloy S, Vorasoot N, Toomsan B,Patanothai A, Holbrook CC (2008) Relationshipbetween biomass production and nitrogen fixationunder drought-stress conditions in peanut genotypeswith different levels of drought resistance. J Agron CropSci 194: 15–25

Pinheiro C, Chaves MM (2011) Photosynthesis anddrought: can we make metabolic connections fromavailable data? J Exp Bot 62: 869–882

Pogson BJ, Albrecht V (2011) Genetic dissection ofchloroplast biogenesis and development: an overview.Plant Physiol 155: 1545–1551

Pose D, Castanedo I, Borsani O, Nieto B, Rosado A,Taconnat L, Ferrer A, Dolan L, Valpuesta V, Botella MA(2009) Identification of the Arabidopsis dry2/sqe1-5mutant reveals a central role for sterols in droughttolerance and regulation of reactive oxygen species.Plant J 59: 63–76

Raorane M, Narciso JO, Kohli A (2011) Total solubleprotein extraction for improved proteomic analysis oftransgenic plants. In: Twyman RM, Christou P (eds)Methods in Molecular Biology: Multigene Engineeringin Plants. Springer-Verlag, New York. (in press)

Rawson R, Zelenski N, Nijhawan D, Ye J, Sakai J,Hasan M, Chang T, Brown M, Goldstein J (1997)Complementation cloning of SP2, a gene encoding aputative metalloprotease required for intramembranecleavage of SREBPs. Mol Cell 1: 47–57

Rose TL, Bonneau L, Der C, Marty-Mazars D, Marty F(2006) Starvation-induced expression ofautophagy-related genes in Arabidopsis. Biol Cell 98:53–67

Rosengren-Brinck U, Nihlgard B (1995) Effects ofnutritional status on the drought resistance in Norwayspruce. Water Air Soil Pollut 85: 1739–1744

Rost TL (2011) The organization of roots of dicotyledonousplants and the positions of control points. Ann Bot 107:1213–1222

Rouached H, Secco D, Arpat AB (2009) Getting the mostsulfate from soil: regulation of sulfate uptake transportersin Arabidopsis. J Plant Physiol 166: 893–902

Roy-Macauley H, Zuily-Fodil Y, Kidric M, Pham Thi AT,Vieira da Silva J (1992) Effect of drought stress onproteolytic activities in Phaseolus and Vigna leaves fromsensitive and resistant plants. Physiol Plant 85: 90–96

Sade N, Vinocur BJ, Diber A, Shatil A, Ronen G, Nissan H,Wallach R, Karchi H, Moshelion M (2009) Improvingplant stress tolerance and yield production: is thetonoplast aquaporin SlTIP2;2 a key to isohydric toanisohydric conversion? New Phytol 181: 651–661

Sardans J, Penuelas J (2005) Drought decreases soilenzyme activity in a Mediterranean Quercus ilex L.forest. Soil Biol Biochem 37: 455–461

Schaller A (2004) A cut above the rest: the regulatoryfunction of plant proteases. Planta 220: 183–197

Schmitz RJ, Tamada Y, Doyle MR, Zhang X, Amasino RM(2009) Histone H2B deubiquitination is required fortranscriptional activation of FLOWERING LOCUS C andfor proper control of flowering in Arabidopsis. PlantPhysiol 149: 1196–1204

Schobert C, Komor E (1987) Amino acid uptake by Ricinuscommunis roots: characterization and physiologicalsignificance. Plant Cell Environ 10: 493–500

Schobert C, Komor E (1990) Transfer of amino acids andnitrate from the roots into the xylem of Ricinuscommunis seedlings. Planta 181: 85–90

Serraj R, Kumar A, McNally KL, Slamet-Loedin I,Bruskiewich R, Mauleon R, Cairns J, Hijmans RJ (2009)Improvement of drought resistance in rice. Adv Agron103: 41–99

Shi K, Ding XT, Dong DK, Zhou YH, Yu JQ (2008) Rootrestriction-induced limitation to photosynthesis intomato (Lycopersicon esculentum Mill.) leaves. SciHortic 17: 197–202

Simova-Stoilova L, Vaseva I, Grigorova B, Demirevska K,Feller U (2010) Proteolytic activity and cysteine proteaseexpression in wheat leaves under severe soil droughtand recovery. Plant Physiol Biochem 48: 200–206

Simova-Stoilova L, Vassileva V, Petrova T, Tsenov N,Demirevska K, Feller U (2006) Proteolytic activity inwheat leaves during drought stress and recovery. GenAppl Plant Physiol 32: 91–100

Physiol. Plant. 2012

Simpson K (2011) Transport of nitrogen in plants. PhDDissertation, University of Sheffield, Sheffield.www.shef.ac.uk/aps/mbiolsci/kelly/dissertation.pdf(Accessed 28 July 2011)

Smith AM, Stitt M (2007) Coordination of carbon supplyand plant growth. Plant Cell Environ 30: 1126–1149

Suralta RR (2010) Plastic root system developmentresponses to drought-enhanced nitrogen uptake duringprogressive soil drying conditions in rice. Philipp AgricSci 93: 458–462

Szczerba MW, Britto DT, Kronzucker HJ (2009) K+

transport in plants: physiology and molecular biology.J Plant Physiol 166: 447–466

Tegeder M, Rentsch D (2010) Uptake and partitioning ofamino acids and peptides. Mol Plant 3: 997–1011

Thompson AR, Doelling JH, Suttangkakul A, Vierstra RD(2005) Autophagic nutrient recycling in Arabidopsisdirected by the ATG8 and ATG12 conjugationpathways. Plant Physiol 138: 2097–2110

Thornton B, Osborne SM, Paterson E, Cash P (2007) Aproteomic and targeted metabolomic approach toinvestigate change in Lolium perenne roots whenchallenged with glycine. J Exp Bot 58: 1581–1590

Vagnerova K, Macura J (1974) Relationships between plantroots, proteolytic organisms and activity of protease.Folia Microbiol 19: 525–535

Van der Hoorn R (2008) Plant proteases: from phenotypesto molecular mechanisms. Annu Rev Plant Biol 59:191–223

Wang D, Pan Y, Zhao X, Zhu L, Fu B, Li Z (2011)Genome-wide temporal-spatial gene expressionprofiling of drought responsiveness in rice. BMC Genom12: 149

Webb KJ, Jensen EF, Heywood S, Morris SM, Linton PE,Hooker JE (2010) Gene expression and nitrogen loss insenescing root systems of red clover (Trifoliumpratense). J Agric Sci 148: 579–591

Wisniewski K, Zagdanska B (2001) Genotype-dependentproteolytic response of spring wheat to water deficiency.J Exp Bot 52: 1455–1463

Wolfe MS, Kopan R (2004) Intramembrane proteolysis:theme and variations. Science 305: 1119–1123

Woltering EJ (2004) Death proteases come alive. TrendsPlant Sci 9: 469–472

Xia YJ (2004) Proteases in pathogenesis and plant defence.Cell Microbiol 6: 905–913

Yang H, Bogner M, Stierhof YD, Ludewig U (2010)H+-Independent glutamine transport in plant root tips.PLoS One 5: e8917

Yokota K, Hayashi M (2011) Function and evolution ofnodulation genes in legumes. Cell Mol Life Sci 68:1341–1351

Zagdanska B, Wisniewski K (1996) Endoproteinaseactivities in wheat leaves upon water deficit. ActaBiochim Pol 43: 515–520

Zagdanska B, Wisniewski K (1998) ATP-dependentproteolysis contributes to the acclimation-induceddrought resistance in spring wheat. Acta Physiol Plant20: 41–48

Zang QW, Wang CX, Li XY, Guo ZA, Jjing RL, Zhao J,Chang XP (2010) Isolation and characterization of agene encoding a polyethylene glycol-induced cysteineprotease in common wheat. J Biosci 35: 379–388

Zhang H, Ohyama K, Boudet J, Chen Z, Yang J, Zhang M,Muranaka T, Maurel C, Zhu JK, Gong Z (2008) Dolicholbiosynthesis and its effects on the unfolded proteinresponse and abiotic stress resistance in Arabidopsis.Plant Cell 20: 1879–1898

Zhang L, Tan Q, Lee R, Trethewy A, Lee YH, Tegeder M(2010) Altered xylem-phloem transfer of amino acidsaffects metabolism and leads to increased seed yield andoil content in Arabidopsis. Plant Cell 22: 3603–3620

Edited by E. Pesquet

Physiol. Plant. 2012