The Role of Root Exudates in Rhizosphere Interactions with Plants and Other Organisms Harsh P. Bais, 5 Tiffany L. Weir, 1,2 Laura G. Perry, 2,3 Simon Gilroy, 4 and Jorge M. Vivanco 1,2 1 Department of Horticulture and Landscape Architecture, 2 Center for Rhizosphere Biology, and 3 Department of Forest, Rangeland, and Watershed Stewardship, Colorado State University, Fort Collins, Colorado 80523-1173 4 Department of Biology, Pennsylvania State University, University Park, Pennsylvania 16802; email: [email protected]5 Department of Plant and Soil Sciences, Delaware Biotechnology Institute, Newark, Delaware 19711 Annu. Rev. Plant Biol. 2006. 57:233–66 The Annual Review of Plant Biology is online at plant.annualreviews.org doi: 10.1146/ annurev.arplant.57.032905.105159 Copyright c 2006 by Annual Reviews. All rights reserved First published online as a Review in Advance on January 30, 2006 1543-5008/06/0602- 0233$20.00 Key Words allelopathy, quorum-sensing, symbiosis, antimicrobial Abstract The rhizosphere encompasses the millimeters of soil surrounding a plant root where complex biological and ecological processes oc- cur. This review describes recent advances in elucidating the role of root exudates in interactions between plant roots and other plants, microbes, and nematodes present in the rhizosphere. Evidence in- dicating that root exudates may take part in the signaling events that initiate the execution of these interactions is also presented. Vari- ous positive and negative plant-plant and plant-microbe interactions are highlighted and described from the molecular to the ecosystem scale. Furthermore, methodologies to address these interactions un- der laboratory conditions are presented. 233 Annu. Rev. Plant Biol. 2006.57:233-266. Downloaded from arjournals.annualreviews.org by OREGON STATE UNIVERSITY on 10/30/07. For personal use only.
Transcript
ANRV274-PP57-10 ARI 5 April 2006 19:6
The Role of Root Exudatesin Rhizosphere Interactionswith Plants and OtherOrganismsHarsh P. Bais,5 Tiffany L. Weir,1,2
Laura G. Perry,2,3 Simon Gilroy,4 andJorge M. Vivanco1,2
1Department of Horticulture and Landscape Architecture, 2Center for RhizosphereBiology, and 3Department of Forest, Rangeland, and Watershed Stewardship,Colorado State University, Fort Collins, Colorado 80523-11734Department of Biology, Pennsylvania State University, University Park,Pennsylvania 16802; email: [email protected] of Plant and Soil Sciences, Delaware Biotechnology Institute, Newark,Delaware 19711
Annu. Rev. Plant Biol.2006. 57:233–66
The Annual Review ofPlant Biology is online atplant.annualreviews.org
AbstractThe rhizosphere encompasses the millimeters of soil surroundinga plant root where complex biological and ecological processes oc-cur. This review describes recent advances in elucidating the role ofroot exudates in interactions between plant roots and other plants,microbes, and nematodes present in the rhizosphere. Evidence in-dicating that root exudates may take part in the signaling events thatinitiate the execution of these interactions is also presented. Vari-ous positive and negative plant-plant and plant-microbe interactionsare highlighted and described from the molecular to the ecosystemscale. Furthermore, methodologies to address these interactions un-der laboratory conditions are presented.
233
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
Rhizosphere: thesoil zone thatsurrounds and isinfluenced by theroots of plants
Plant roots exude an enormous range ofpotentially valuable small molecular weightcompounds into the rhizosphere. Some ofthe most complex chemical, physical, and bi-ological interactions experienced by terres-trial plants are those that occur between rootsand their surrounding environment of soil(i.e., the rhizosphere). Interactions involvingplants roots in the rhizosphere include root-root, root-insect, and root-microbe interac-tions. Over the past decade, enormous stepshave been taken toward understanding thesedifferent types of interactions (79), and re-cently the field of plant biology has recognized
the importance of root exudates in mediatingthese biological interactions (9, 180, 187).
The rhizosphere represents a highly dy-namic front for interactions between rootsand pathogenic and beneficial soil microbes,invertebrates, and root systems of competi-tors (79). However, because plant roots arehidden belowground, many of the inter-esting phenomena in which they are in-volved have remained largely unnoticed. Inparticular, the role of chemical signals inmediating belowground interactions is onlybeginning to be understood. Chemical sig-naling between plant roots and other soil or-ganisms, including the roots of neighboringplants, is often based on root-derived chem-icals. The same chemical signals may elicitdissimilar responses from different recipients.Chemical components of root exudates maydeter one organism while attracting another,or two very different organisms may be at-tracted with differing consequences to theplant. A concrete example of diverse mean-ings for a chemical signal is the secretion ofisoflavones by soybean roots, which attracta mutualist (Bradyrhizobium japonicum) and apathogen (Phytopthora sojae) (122). The mech-anisms used by roots to interpret the innu-merable signals they receive from other roots,soil microbes, and invertebrates in the rhizo-sphere are largely unknown.
Root-root, root-microbe, and root-insectinteractions may be classified as either positiveor negative associations (Figure 1). A thirdcategory of neutral associations also exists, butis not addressed here. Positive interactions in-clude symbiotic associations with epiphytesand mycorrhizal fungi, and root coloniza-tion by bacterial biocontrol agents and plantgrowth–promoting bacteria (PGPB). Nega-tive interactions include competition or par-asitism among plants, pathogenesis by bacte-ria or fungi, and invertebrate herbivory. Thefactors that determine whether the chemicalsignature of a plant’s root exudates will be per-cieved as a negative or a positive signal stillrequire elucidation. However, accumulated
234 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
evidence suggests that root exudates have amajor role in determining outcomes of in-teractions in the rhizosphere and, ultimately,plant and soil community dynamics.
WHAT ARE ROOT EXUDATES?
In addition to accumulating biologically ac-tive chemicals, plant roots continuously pro-duce and secrete compounds into the rhi-zosphere (13, 60). Root exudation includesthe secretion of ions, free oxygen and wa-ter, enzymes, mucilage, and a diverse arrayof carbon-containing primary and secondarymetabolites (17, 172). Root exudation canbe broadly divided into two active processes.The first, root excretion, involves gradient-dependent output of waste materials with un-known functions, whereas the second, secre-tion, involves exudation of compounds withknown functions, such as lubrication and de-fense (8, 172). Roots release compounds viaat least two potential mechanisms. Root exu-dates are transported across the cellular mem-brane and secreted into the surrounding rhi-zosphere. Plant products are also releasedfrom root border cells and root border-likecells, which separate from roots as they grow(71, 175). Root exudates are often divided intotwo classes of compounds. Low-molecularweight compounds such as amino acids, or-ganic acids, sugars, phenolics, and other sec-ondary metabolites account for much of thediversity of root exudates, whereas high-molecular weight exudates, such as mucilage(polysaccharides) and proteins, are less di-verse but often compose a larger proportionof the root exudates by mass. Root exudationclearly represents a significant carbon cost tothe plant (117), and the magnitude of pho-tosynthates secreted as root exudates varieswith the type of soil, age, and physiologicalstate of the plant, and nutrient availability (21,23). Although the functions of most root exu-dates have not been determined, several com-pounds present in root exudates play impor-tant roles in biological processes (9, 10, 11, 98)(Figure 2). The following sections of this
Figure 1Plant-microbe positive and negative interactions. (a) Biocontrol of Bacillussubtilis (6051) on Arabidopsis thaliana roots by forming protective biofilmsagainst gram-negative bacteria Pseudomonas syringae pv. tomato DC3000.Panel a shows the formation of aggregates or biofilm by B. subtilis onArabidopsis root surface. Colonization of the root by B. subtilis biofilmslimits the root space available for P. syringae to infection. Additionally, B.subtilis, like other gram-positive biocontrols, produces an antibacterialcompound, surfactin, against P. syringae DC3000. (b) Pathogenic biofilmformation by a non-bonafide plant pathogen Pseudomonas aeruginosa on A.thaliana root surface. Panel b represents a crossover human pathogen, P.aeruginosa, which can infect plants under controlled conditions. P.aeruginosa forms pathogenic biofilm on Arabidopsis roots to exhibit fullpathogenesis in a plant model. (c–d ) Attachment of symbiont Sinorhizobiummeliloti on C. elegans outer cuticle. In this unique interaction C. elegans actsas a vector for S. meliloti to transfer rhizobial inoculum to legume roots.(e–f ) C. elegans feeding on the rhizobial lawn and nodule formation on hostMedicago roots. Panels c–f represent one of the first reports to show apositive tritrophic interaction. C. elegans, a soil nematode, uses S. meliloti asfood but does not digest all the bacteria. Instead the undigested S. melilotiand the attached bacteria on the C. elegans cuticle are transferred to the hostplant root to complete the vector-mediated symbiosis. Additionally, plantroots also trigger C. elegans behavioral response by emitting volatile signalsinviting nematodes to the root proximity. We kindly thank Dr. JunichiroHoriuchi for providing photos in panels c–f.
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 235
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
Allelopathy: theinhibition of growthin one species ofplants by chemicalsproduced by anotherspecies
review describe the importance of root exu-dates in positive and negative interactions thatdetermine plant and soil microbe growth andsurvival.
PLANT-PLANT INTERACTIONSMEDIATED BY ROOTEXUDATES
Resource competition, chemical interference,and/or parasitism lead to negative interactionsbetween plants (Figure 2). Root exudateshave the potential to influence all threemechanisms of interference. For a numberof plant species, root exudates play a directrole as phytotoxins in mediating chemicalinterference (i.e., allelopathy). In addition,root exudates are critical to the developmentof associations between some parasitic plantsand their hosts. Finally, root exudates mayplay important indirect roles in resourcecompetition by altering soil chemistry, soilprocesses, and microbial populations.
Positive interactions between plants arealso sometimes controlled by root exudates. Inparticular, some root exudates induce defenseresponses in neighboring plants. In somecases, the plant defenses induced by root exu-dates simply reduce susceptibility to pathogeninfection, whereas in other cases these de-fenses initiate production and release of leafyvolatiles that attract predators of plant ene-mies. In addition, effects of root exudates onsoil processes and microbial populations canlead to some positive effects on neighboringplants.
Negative Plant-Plant Interactions
Allelopathy. Chemical-mediated plant-plant interference, or allelopathy, is onemechanism by which plants may gain anadvantage over their competitors. Plants thatproduce and release potent phytotoxins canreduce the establishment, growth, or survivalof susceptible plant neighbors, thus reduc-ing competition and increasing resourceavailability. Plants release phytotoxins indecomposing leaf and root tissue, in leachatesfrom live tissue, in green leafy volatiles, andin root exudates (17, 187). Plant-producedphytotoxins vary considerably in chemicalstructure, mode of action, and effects onplants. Different phytotoxins in root exudatesaffect metabolite production, photosynthesis,respiration, membrane transport, germina-tion, root growth, shoot growth, and cellmortality in susceptible plants (47, 187).These effects on plant physiology, growth,and survival may in turn influence plant andsoil community composition and dynamics.
A number of phytotoxic compounds inplant root exudates have been identified,including but not limited to 7,8-benzoflavone(Acroptilon repens, Russian knapweed) (164),( ± )-catechin (Centaurea maculosa, spottedknapweed) (12), DIMBOA and DIBOA(Triticum aestivum, wheat) (190), juglone( Juglans nigra, black walnut) (89), 8-hydroxyquinoline (Centaurea diffusa, diffuseknapweed) (176), sorgoleone (Sorghumspp.) (133), and 5,7,4′-trihydroxy-3′,5′-dimethoxyflavone (Oryza sativa, rice) (101).These compounds share some structural
−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→Figure 2Rhizospheric chemical warfare: schematic representation of possible rhizospheric interactions mediatedby root exudates. Root-mediated rhizospheric interactions are broadly classified into two categories,positive and negative interactions. Positive interactions involve root exudate-mediated interactions withplant growth–promoting Rhizobacteria (PGPR). Roots produce chemical signals that attract bacteria andinduce chemotaxis. Positive interactions mediated by root exudates also include growth facilitators orgrowth regulator mimics that support growth of other plants and also perform cross-species signalingwith rhizospheric invertebrates. Contrastingly, negative interactions mediated by root exudates involvesecretion of antimicrobials, phytotoxins, nematicidal, and insecticidal compounds. The arrows in thepanels indicate chemical exchange. VAM, vesicular arbuscular mycorrhizas; SARs, systemic acquiredresistance.
236 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
Root
Root
Fungi
Bacteria
Bacteria
Nematode
Fungi
PGPRs; symbionts
Root herbivory
Nematicidal/ insecticidal compounds
Root Biocontrol, VAM, endophytes
Antifungal compounds
Phytotoxins
Type-III phytotoxins
Antibacterial compounds, QS mimics
Vectors for symbioses, SARs
Growth facilitators
Allelopathy
Positive Interactions
Negative Interactions
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 237
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
Autoinhibition:autotoxicity that isbeneficial to someplants within apopulation
components, such as aromaticity (with theexception of sorgoleone), and the presenceof hydroxyl and/or ketone groups. However,the structures of the compounds also varyconsiderably, and include flavonoids [7,8-benzoflavone, ( ± )-catechin, and 5,7,4′-tri-hydroxy-3′,5′-dimethoxyflavone], quinones(juglone and sorgoleone), quinolines (8-hydroxyquinoline), and hydroxamic acids(DIMBOA, DIBOA).
Phytotoxic root exudates can mediate neg-ative plant-plant interactions only if presentat sufficient concentrations to affect plantgrowth and survival. Centaurea maculosa, C.diffusa, and Sorghum spp. produce their phyto-toxins at high concentrations, whereas Juglansnigra appears to produce lower concentrationsof juglone. Young C. maculosa plants grown to-gether in liquid culture can produce >80 μgml−1 under standard conditions and >180 μgml−1 in the presence of fungal cell wall ma-terials (12). Soil ( ± )-catechin concentrationsaveraged 2.24 mg g−1 in one C. maculosa pop-ulation (9) and 1.55 mg g−1 in another popula-tion (144). 8-Hydroxyquinoline soil concen-trations in a C. diffusa population were lowerthan reported catechin concentrations, butstill relatively high: 0.25 mg g−1 (176). Varia-tion in ( ± )-catechin and 8-hydroxyquinolineconcentrations among seasons, years, and soiltypes has not yet been examined. Sorghum spp.rhizosecrete more sorgoleone than any othercompound in their root exudates (36). Agri-cultural species such as S. bicolor (sorghum)and S. sudanese (sudangrass) produce between1.3 and 1.9 mg g−1 of sorgoleone, whereas theinvasive weed S. halepense (johnsongrass) canrhizosecrete up to 14.8 mg g−1. Sorgoleoneconcentrations in Sorghum spp. soils have notyet been reported. Juglone concentrations insoil beneath J. nigra trees rarely exceed 3 ugg−1 of soil (89), suggesting that productionof juglone may be much lower than produc-tion of ( ± )-catechin, 8-hydroxyquinoline,and sorgoleone. However, both chemicalstability and production rates determine phy-totoxin concentrations in the rhizosphere. Ju-glone is relatively stable in soil and shows lit-
tle seasonal variation in concentration (89).In contrast, sorgoleone degrades quickly insoil (35), suggesting that continuously highproduction rates may be necessary to main-tain phytotoxic concentrations of sorgoleonein soil. Degradation rates of ( ± )-catechin and8-hydroxyquinoline in soil have not yet beendetermined.
The ecological relevance of phytotoxicroot exudates also depends on the suscep-tibility of the plants with which the al-lelopathic plants coexist. ( ± )-Catechin and8-hydroxyquinoline inhibit the growth of na-tive North American plants in communi-ties invaded by Centaurea maculosa (9, 186)and C. diffusa (176), respectively. In partic-ular, ( ± )-catechin inhibits root growth ofmore than 20 North American grasslandspecies (143). Likewise, sorgoleone, DIBOA,and 5,7,4′-trihydroxy-3′,5′-dimethoxyflavonelimit the growth of weeds that coexist in agri-cultural systems with Sorghum bicolor (133),Triticum aestivum (114), and Oryza sativa(101), respectively. However, most of these ex-periments were conducted under laboratory,and not field, conditions. Tests applying typi-cal soil phytotoxin concentrations under real-istic conditions are necessary to evaluate withmore certainty the importance of phytotoxinproduction to outcomes of plant-plant inter-ference (83). An even more informative ap-proach would involve comparisons with mu-tants or transgenic plants that do not producephytotoxins. Recently, a gene involved in sor-goleone production was identified in Sorghumbicolor (192), perhaps providing an opportu-nity for a clear test of the importance of al-lelopathy in one species.
Many plants also produce secondarymetabolites that inhibit the growth of con-specific plants (i.e., autotoxicity). Autotoxic-ity has been widely observed in agriculturalcrops and weeds, as well as in some plantsthat inhabit natural systems (160). Phytotoxicroot exudates appear to mediate autoinhibi-tion in at least some of these species, includingAsparagus officinalis (garden asparagus) (131),Cucumis sativa (garden cucumber) (195),
238 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
and Centaurea maculosa (spotted knapweed)(144). In many cases, plants that are al-lelopathic also exhibit signs of autotoxicity(160). However, only one study has identifiedthat the same root exudate is responsible forboth allelopathy and autotoxicity in a plantspecies. Perry et al. (144) demonstrated that( ± )-catechin, the phytotoxin produced byC. maculosa, also inhibits C. maculosa seedlingestablishment at high concentrations. Auto-toxicity may be a simple consequence of pro-ducing an allelochemical for which completeresistance is energetically expensive. Alterna-tively, autotoxicity may be beneficial to someplants within the population, a phenomenontermed autoinhibition. Autoinhibition maybenefit adult plants or seedlings that produceautoinhibitors by reducing the establishmentof intraspecific competitors in dense popula-tions (44), or may benefit ungerminated seedsby delaying germination in areas with intenseintraspecific competition, if the autoinhibitorinduces seed dormancy (146).
Many allelopathic plants, however, appearto be relatively resistant to the phytotoxinsthey produce. Furthermore, some nonallelo-pathic plants are also relatively resistant tophytotoxins produced by other plants. For ex-ample, in a study of grassland species resis-tant to Centaurea maculosa‘s phytotoxin, 8 of23 species examined were more resistant to( ± )-catechin than C. maculosa (143). Plantsemploy various methods to resist phytotoxinsin the rhizosphere. Some plants may avoid ef-fects of phytotoxins by sequestering the toxinsin vacuoles or specialized tissues, or by secret-ing the phytotoxins as they are taken up (189).Other plants avoid inhibition from phytotox-ins by altering the chemical structure of thetoxins. For example, Polygonella myriophylla(Small’s jointweed) avoids the effects of itsown phytotoxins, hydroquinone and benzo-quinone, by instead producing and releasingarbutin, a glycoside of hydroquinone (185).Microbial degradation of the glycoside allowsthe phytotoxins to be produced in the rhizo-sphere rather than in the plant. Similarly, Zeamays (corn) relies on N-glucosylation to avoid
Autotoxicity: aform of allelopathythat refers to a plant’sability to ward offcompetition fromnew growth withinits own species
the effects of DIMBOA, DIBOA, and BOA,phytotoxins secreted into the rhizosphere byTriticum aestivum (wheat) and several othergrasses. BOA glucosylation occurs in incu-bations with Z. mays, forming a substantiallyless toxic compound (158). Zea mays possessestwo glucosyltransferases, BX8 and BX9, thatact specifically on DIBOA and DIMBOA, andconfer resistance to DIBOA and DIMBOA intransgenic Arabidopsis thaliana plants, demon-strating the importance of BX8 and BX9 toZ. mays phytotoxin resistance (179).
Nevertheless, the sensitivity of manyplants to a range of plant-produced phyto-toxins suggests that resistance may be ener-getically expensive and limited to a subset ofspecies. Thus, negative biochemical interac-tions among plants may be an important fac-tor shaping plant community structure.
Community-scale interactions: biologicalinvasions. Over evolutionary time, plantsfrequently encountering allelopathic speciesare likely to acquire resistance to root-secreted phytotoxins. However, because phy-totoxin resistance probably involves someenergetic cost, plants that do not frequentlyencounter a phytotoxin may be unlikely topossess resistance to the toxin. Thus, tran-sient plant species might be more sensitiveto phytotoxins produced by other plants. Bythe same logic, phytotoxins produced by tran-sient plants might be expected to affect a widerarray of plant species than those that per-sist for long periods in particular plant com-munities. Among species that frequently as-sociate with one another, coevolution mightlead to an arms race of increasingly sophisti-cated allelochemicals with increasingly expen-sive requirements for resistance. Alternatively,coevolution in plant communities might de-crease the ecological importance of directchemical interference.
Biological invasions by exotic allelopathicplants present a unique case, in which na-tive species in the invaded range have mostlikely never encountered the phytotoxins pro-duced by the invader. As a result, these
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 239
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
“novel weapons” (29) would have much largernegative effects on “naıve” native species in aninvaded range than the “experienced” speciesin the invader’s native range. The greatersuccess of some exotic plants in their in-vaded ranges may be partially explained bythe sensitivity of native species to the phy-totoxins of the invader (9, 29, 77, 176). Todate, the novel weapons hypothesis for inva-sion has been tested for only a few species,although the available evidence suggests thatnumerous other exotic invaders may alsobe allelopathic (36, 61, 164). The strongestevidence for the novel weapons hypothesiscomes from experiments on two invaders ofNorth American grasslands, Centaurea dif-fusa (diffuse knapweed) and C. maculosa (spot-ted knapweed). Callaway & Aschehoug (29)found that adding activated carbon to ad-sorb organic compounds (i.e., root exudates)in C. diffusa soils alleviated phytotoxic effectson neighboring grass species. Their experi-ments indicated that North American grass-land species were significantly more inhib-ited by C. diffusa root exudates than theEuropean grassland species with which C. dif-fusa naturally coexists. Vivanco et al. (176) ap-plied 8-hydroxyquinoline, a phytotoxin iden-tified in C. diffusa root exudates, to NorthAmerican and European grassland speciesand also found that the North Americanspecies were significantly more susceptibleto the phytotoxin, suggesting an importantrole of 8-hydroxyquinoline in C. diffusa in-vasions in North America. In a similar ex-periment, Bais et al. (9) found that NorthAmerican grassland species are also moresensitive than European congeners to ( ± )-catechin, the phytotoxin identified in C. mac-ulosa root exudates, suggesting a similar roleof ( ± )-catechin in C. maculosa invasions. Fi-nally, Prati & Bossdorf (148) found that rootexudates from Allium petiolata, another in-vasive species in North America, had a sig-nificantly greater negative effect on a NorthAmerican species, Geum laciniatum, than on aEuropean congener, Geum urbanum, support-ing the novel weapons hypothesis for A. peti-
olata invasion. However, root exudates fromEuropean A. petiolata populations had similarnegative effects on the two congeners, indi-cating that A. petiolata phytotoxins may haveimportant ecological effects in both the nativeand the invaded range.
Biological invasions may result from phy-tochemical effects on soil chemistry and soilmicrobial communities as well as from di-rect chemical interference (184). For exam-ple, secondary metabolites from one invasiveplant, Carduus nutans (musk thistle), appearto inhibit nodulation and nitrogen fixation inleguminous species such as Trifolium repens(white clover) (182). Perhaps as a result,T. repens growth and survival is strongly re-duced in field patches invaded by C. nutans(183). C. nutans appears to tolerate the result-ing low-nitrogen conditions and benefit fromthe absence of competitors, re-establishingin previously invaded patches (182). In an-other example, secondary metabolites fromEmpetrum hermaphroditum (crowberry) in-hibit symbiotic associations between Pinussylvestris (Scots pine) trees and mycorrhizalfungi, thus reducing P. sylvestris nitrogen up-take (132). Moreover, secondary metabolitesin E. hermaphroditum litter inhibit soil mi-crobial and macrofaunal activity, thus reduc-ing decomposition rates and further reducingsoil nutrient availability (181). The effects ofE. hermaphroditum secondary metabolites onsoil processes, perhaps in conjunction withphytotoxic effects on forest plants (196), ap-pear to facilitate E. hermaphroditum domi-nance and reduce tree productivity (184). De-spite the strong evidence that plant secondarymetabolites can affect soil processes that inturn alter plant-plant interactions, the ef-fects of invasive plants’ root exudates on soilprocesses in their native and invaded rangeshave received little attention. More research isneeded to evaluate the importance of interac-tions between root exudates and soil processesas mechanisms of biological invasion.
Parasitic plant-host interactions. Root ex-udates are essential in the development of
240 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
associations between parasitic plants and theirplant hosts, an association that is negative forthe host and positive for the parasite. Morethan 4000 facultative and obligate parasiticplants have been identified to date (194). Thechemical cross talk that controls the locationof parasite germination and the developmentof physical connections between the parasiteand the host is well understood for several ob-ligate parasites, including Striga spp. (witch-weed) and Orobanche spp. (broomrape) (137).Most current knowledge of the role of rootexudates in parastic plants has been obtainedfrom research on Striga asiatica and S. hermon-thica (hereafter Striga) infestations of Sorghumspp.
Striga have very small seeds that cansurvive for only a few days after germinationbefore forming an association with a host(137). The limited carbohydrate reserves inStriga seeds restrict seedling root elongationbefore host attachment. Thus, arranging forgermination to coincide with proximity ofan appropriate host root is critical to Strigaseedling survival. To ensure that germinationoccurs near host roots, Striga seeds germinateonly in the presence of sustained (10–12 h)high concentrations of germination inducersexuded into the soil by host roots (31).Germination inducers vary between differentStriga hosts. To date, the only plant-producedStriga germination inducer that has beenidentified and characterized is sorghumxenognosin (SXSg). SXSg is highly unstablein aqueous solution (49), a useful trait for aStriga germination inducer because it is un-likely to persist in the soil and falsely indicatethe presence of a host. However, SXSg is sounstable that it initially seemed difficult toexplain how SXSg persisted and traveled inthe soil in quantities sufficient to affect nearbyStriga seeds (49). Fate & Lynn (49) providedan explanation for SXSg activity in soil bydemonstrating that a compound structurallysimilar to SXSg, recorcinol, is released insmall quantities with SXSg in sorghum rootexudates and stabilizes SXSg enough to allowit to induce Striga germination.
SXSg: sorghumxenognosin
Haustorium: aspecialized absorbingstructure of aparasitic plant, suchas the rootlikeoutgrowth of thedodder, that obtainsfood from a hostplant
Root exudates also play an integral rolein Striga haustorial formation. Haustoria arespecialized root structures in plant parasitesthat allow the parasites to infect host roots andform connections with host vascular tissue.The most recent evidence suggests that thechemical cross talk between Striga seedlingsand host roots that results in haustorial for-mation begins with the constitutive release ofhydrogen peroxide from Striga seedling roottips into the rhizosphere (94). Hydrogen per-oxide activates host, and perhaps parasite, per-oxidases that degrade host cell wall pectins,oxidatively releasing benzoquinones into therhizosphere (92). The host benzoquinones aredetected by the Striga seedling root, perhapsby redox activation of a receptor, and initi-ate haustorial formation (161). The mecha-nisms through which host benzoquinones in-duce haustorial development are not yet fullyunderstood, but involve downregulation of agene for one Striga expansin protein, and up-regulation of genes for two unusual expansins,saExp1 and saExp2 (135). Expansins enablecell expansion by disrupting hydrogen bondsin cell walls (120). saExp1 and saExp2 maybe important factors in the development andexpansion of the unusual root cells in Strigahaustoria.
Positive Plant-Plant Interactions
Induced herbivore resistance. Root exu-dates can also have positive effects in plant-plant interactions, although these have beenless frequently reported. In particular, someroot exudates increase herbivore resistancein neighboring plants. For example, Elytri-gia repens (couch-grass) produces several phy-totoxic compounds in its root exudates, ofwhich one, carboline, has been identified (61).Hordeum vulgare (barley) treated with either E.repens root exudates or with carboline alonewas significantly less likely to be chosen as ahost by aphids than control H. vulgare plants.Carboline in the absence of H. vulgare didnot repel aphids, indicating that H. vulgare
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 241
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
responses to E. repens root exudates are nec-essary for aphid repulsion. The induction ofH. vulgare defense responses by E. repens ex-udates may be a consequence of secondarymetabolite production resulting from expo-sure to E. repens phytotoxins. Alternatively, E.repens may produce carboline in part for in-duction of its own defense responses, and hasunintended effects on neighboring plants suchas H. vulgare.
Induced herbivore defense via predatorattraction. In addition to having directeffects on herbivore behavior, some rootexudates induce defense responses inneighboring plants that reduce herbivorepopulations indirectly by attracting predatorsand parasites of the offending herbivore(30). For example, V. faba plants under attackrelease root exudates that induce greenleafy volatile production in undamagedV. faba plants, which in turn attracts aphidparasitoids (42a). Similarly, Phaseolus lunatus(lima bean) plants under attack by spidermites produce root exudates that inducevolatile production in undamaged P. lunatusplants, attracting predatory mites (66). Greenleafy volatiles produced by plants underherbivore attack have also been shown toinduce volatile production in neighboringplants, increasing the predator attractionsignal (24). Thus, both root exudates andleafy volatiles can serve as signals to informplants of herbivores nearby. Plants that havedeveloped the ability to “eavesdrop” on thechemical status of their neighbors are morelikely to be prepared for herbivore attacks,and can participate in coordinated biocontrolefforts that may substantially reduce herbi-vore populations. Most research on inducedherbivore defense responses within plantcommunities has focused on volatile signalsand predator behavior aboveground. Furtherresearch is needed to identify and charac-terize the root exudates that initiate volatileproduction in neighboring, undamagedplants.
Mechanisms That Influence SoilResources
Some effects of root exudates on both posi-tive and negative plant-plant interactions mayalso be mediated by indirect effects on soilresources (84, 184). Root exudation can in-crease or decrease soil nutrient availability byaltering soil chemistry and soil biological pro-cesses. These effects can in turn influenceoutcomes of resource competition betweenplants, particularly if the root exudates alterthe limiting resources. Effects of root exudateson soil resource availability may most oftenbe strongest in the rhizosphere of the plantsthat produce them, providing a competitiveadvantage over neighboring plants that lackthe same abilities. However, in some systems,root exudates may influence soil propertieson a larger scale, with the potential for posi-tive or negative effects on soil resource avail-ability to neighboring plants. Here, we dis-cuss two of the mechanisms through whichroot exudation of plant secondary metabo-lites can influence soil resource availability:phytosiderophore secretion and organic acidsecretion.
Phytosiderophores and micronutrientavailability. Some root exudates that actas metal chelators in the rhizosphere canincrease the availability of metallic soilmicronutrients, including iron, manganese,copper, and zinc (37). Metal chelators formcomplexes with soil metals, thus releasingmetals that are bound to soil particles andincreasing metal solubility and mobility.The best evidence that plants use chelatorsin root exudates to increase micronutrientavailability comes from research on theeffects of graminoid phytosiderophores oniron (Fe) availability. Although Fe is oftenrelatively abundant in soil, it is also oftenpresent as insoluble Fe(III) precipitates,particularly in soils with high or neutral pH.Graminoid-secreted phytosiderophores bindto Fe(III) to form Fe(III)-phytosiderophores,
242 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
which grasses can take up with substantiallygreater efficiency than other chelated formsof Fe (153). Phytosiderophores that have beenidentified include nonproteinogenic aminoacids such as mugenoic and avenic acid (165).Graminoid secretion of phytosiderophoresis markedly greater in Fe-deficient thanFe-sufficient plants, indicating an importantrole of the compounds in mitigating Festress. Different grasses efficiently take upFe(III) bound to phytosiderophores pro-duced by other species (153), suggesting thatphytosiderophore secretion may increase Feavailability across graminoid communities.The evidence that rhizosecreted chelatorsplay a similarly important role in micronutri-ent availability to dicots is less strong than forgraminoids (88). However, many phenolicsproduced by dicots have the potential to formcomplexes with metallic micronutrients andmay also increase metal availability (37).
Organic acids and phosphorus availabil-ity. Organic acids can also act as metal chela-tors in the rhizosphere, but are thought tohave more important effects on phosphorusavailability than on micronutrient availability(37). Phosphorus, like iron, is often relativelyabundant in soils, but in unavailable forms. Inparticular, phosphorus is often bound in in-soluble ferric, aluminum, and calcium phos-phates, especially in soils with high pH. Or-ganic acids such as citric, malic, and oxalicacid can form complexes with the iron or alu-minum in ferric and aluminum phosphates,thus releasing plant-available phosphates intothe soil (37, 118). Organic acids may alsoincrease phosphorus availability by blockingphosphorus absorption sites on soil particlesor by forming complexes with cations on soilmineral surfaces (88). Several plants increaseorganic acid rhizosecretion substantially in re-sponse to phosphorus deficiencies, includingLupinus alba (white lupine) (87, 129), Bras-sica napus (rape) (80), and Medicago sativa(alfalfa) (108). Among species examined fororganic acid production in response to phos-phorus stress, lupines exhibit the strongest
trends (37). Lupines form clusters of special-ized root structures, termed proteoid roots,in response to phosphorus deficiency. Matureproteoid roots appear to both increase organicacid production and decrease organic acidmetabolism compared to nonproteoid roots,resulting in much higher levels of organicacid exudation (1.16 compared to 0.09 μmolh−1 g−1 in one study) (91, 129, 171). Per-haps as a result, phosphorus uptake can beas much as 50% greater in proteoid thannonproteoid lupine roots (129). However, todate, research on effects of organic acids onphosphorus availability and uptake has beenconducted mainly under relatively unrealis-tic laboratory conditions. Further studies todetermine rhizosphere concentrations of or-ganic acids in live soil and to examine the ef-fects of those concentrations on phosphorussolubility and uptake are needed to confirmthe role of organic acids in plant responses tophosphorus stress (88). In addition, it has notyet been determined whether the high ratesof organic acid secretion by lupines also in-crease phosphorus availability to neighboringplants.
Plant-Plant Molecular Interactions
The molecular targets of root exudates re-main poorly defined. For allelochemicals, arange of cellular effects have been reported,from loss of plasma membrane integrity andion leakage (54) to inhibition of photosyn-thetic and respiratory electron transport (1,54, 141) and inhibition of cell division (3).There are very few cases where the effects ofallelochemicals are proposed to be more orless direct. For example, sorgoloene likely in-terferes with mitochondrial electron transportby inhibiting the reduction of cytochrome c1
by cytochrome b, a site inhibited by several hy-droxyquinone analogus (178), and the photo-synthetic electron transport chain by blockingoxidation of the PSII-reduced primary elec-tron acceptor, by binding to QB (62). Simi-larly, juglone and sorgoleone inhibit plasmamembrane proton pumping (72, 73), likely
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 243
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
ROS: reactiveoxygen species
contributing directly to loss of membrane in-tegrity and ion leakage. One other poten-tial direct allelochemical effect contributingto cell death is through generation of reactiveoxygen species (ROS) and subsequent oxida-tive damage to the target plant. Bais et al. (9)reported that catechin produced from Cen-taurea maculosa could elicit root toxicity as-sociated with an increase in ROS produc-tion by the susceptible root. Scavenging theROS change reduced catechin’s toxicity, lead-ing to the idea that ROS might be part ofthe phytoxic cascade elicited by this root exu-date. Indeed, an increase in oxidative stress hasbeen proposed as a widespread phenomenonin such allelopathic responses (9, 34, 152). En-vironmental stress is often linked to oxidativestress, which is countered by a plant antiox-idant system including ascorbate, superoxidedismutase, catalase, and the glutathione sys-tem (4). ROS can have wide-ranging damag-ing effects on biology through directly mod-ifying cellular components. One such actionthat may be highly relevant to allelochemical-induced toxicity is ROS-related effects on thelipid bilayer, such as lipid peroxiadation. Lipidperoxidation leads to the destruction of thepolyunsaturated fatty acids that are integralto membrane integrity and transport activitiesacross the plasma membrane. Increase in lipidperoxidation accompanies addition of aque-ous allelochemical in tomato and cucumberroots (34, 147) and, as noted above, electrolyteleakage from cells is often associated with al-lelopathic response. It is interesting that arange of antioxidant system-related genes areinduced in Arabidopsis treated with catechin(9). Thus, although a major pathway to plantresistance to allelochemical action is thoughtto be through chemical detoxification and se-questration (187), the relationship betweenantioxidant system and allelochemical resis-tance is worthy of a more in-depth study.
However, in addition to a direct role incell mortality, ROS is also well character-ized as a signaling molecule (4). For example,ROS gates signal-related ion channels (e.g.,52, 100, 140) and has critical roles in mediat-
ing hormone responses (105). These observa-tions highlight the possibility that root exu-dates could act via triggering a host of signal-ing events within the susceptible plant. Thus,flavanoids are well characterized in animalcells as being signaling molecules (188), andin plants they act in many signaling and regu-latory pathways. For example, they modulateauxin transport either directly through inter-actions with the transport system (26, 124) orpossibly indirectly via regulating the vesiculartrafficking responsible for targeting this sys-tem to the correct membrane surface (139).Similarly, flavanoids play roles in pollen ger-mination (121), perhaps via a protein kinasesignaling cascade (67). However, the molecu-lar sites of action and signaling cascades trig-gered by flavanoids in general, and especiallyby the varied components of root exudates, areunknown. Defining potential receptors andthe associated signaling systems for these ex-udates is an area with great potential to helpelucidate how exudates yield such highly spe-cific and yet varied responses in susceptibleplants.
Plant-microbe interactions can positively in-fluence plant growth through a variety ofmechanisms, including fixation of atmo-spheric nitrogen by different classes of pro-teobacteria (123), increased biotic and abioticstress tolerance imparted by the presence ofendophytic microbes (157), and direct and in-direct advantages imparted by plant growth–promoting rhizobacteria (63) (Figure 2). Bac-teria can also positively interact with plantsby producing protective biofilms or antibi-otics operating as biocontrols against poten-tial pathogens (7), or by degrading plant- andmicrobe-produced compounds in the soil thatwould otherwise be allelopathic or even au-totoxic. However, rhizosphere bacteria canalso have detrimental effects on plant healthand survival through pathogen or parasite
244 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
infection. Secreted chemical signals from bothplants and microbes mediate these complexexchanges and determine whether an interac-tion will be malevolent or benign.
Root colonization is important as the firststep in infection by soil-borne pathogens andbeneficial associations with microorganisms.The “rhizosphere effect,” first described byHiltner in 1904 (78), assumes that many mi-croorganisms are attracted to nutrients ex-uded by plant roots. Hiltner observed thatthe number and activity of microorganismsincreased in the vicinity of plant roots. How-ever, in addition to providing a carbon-richenvironment, plant roots initiate cross talkwith soil microbes by producing signals thatare recognized by the microbes, which inturn produce signals that initiate colonization.Motility is an important trait for competitivepathogens and beneficial microbes and en-ables participation in this cross talk (39, 112,113). Chemical attraction of soil microbes toplant roots, or chemotaxis, is a well under-stood mechanism involved in initiating crosstalk between plant roots and microbes (8). An-other recently discovered mechanism involvesthe use of electric potentials in plant roots,produced by electrogenic ion transport at theroot surface, to attract swimming zoospores ofoomycete plant pathogens to plant root sur-faces (174). These data also suggest that elec-trical signals may mask the chemical signals inmediating short-range responses of oomycetezoospores to root surfaces. It is not knownwhether the perception of chemotaxis or elec-trotaxis signals may affect the likelihood thatsoil microbes will act as pathogens or sym-bionts. Below, we describe in depth the directand indirect positive and negative roles of rootexudates in mediating plant-microbe interac-tions in the rhizosphere.
Positive Plant-Microbe Interactions
Nodulation of legumes by rhizobia. Rhi-zobia form symbiotic associations with legu-minous plants by fixing atmospheric nitrogenin root nodules. Scientists have always won-
dered whether plants outside the Fabaceaefamily might be manipulated to form associa-tions with rhizobia (109). However, rhizobia-legume interactions are very specific, allow-ing specific rhizobial strains to nodulate withspecific host legumes. Sinorhizobium melilotieffectively nodulates species of the Medicago,Melilotus, and Trigonella genera, whereas Rhi-zobium leguminosarum bv viciae induces nod-ules in the Pisum, Vicia, Lens, and Lathyrusgenera. However, not all rhizobia-legumeassociations are this limited. For example,Rhizobium strain NGR234 nodulates with232 species of legumes from 112 generatested and even nodulates with the nonlegumeParasponia andersonii, a member of the elmfamily (149). Conversely, not all membersof the legume family form nodules. Of thethree subfamilies of legumes, Caesalpinoideae,Mimosoideae, and Papilionoideae, members ofthe basal subfamily Caesalpinoideae are mainlynon-nodulating. Thus, nodulation and pre-sumably nitrogen fixation are not ubiquitouswithin the legume family.
The signal components largely respon-sible for these specific host-microbe rela-tionships belong to a class of compoundstermed flavonoids (145). More than 4000different flavonoids have been identified invascular plants, and a particular subset ofthem is involved in mediating host speci-ficity in legumes (142). Isoflavonoids are onlyfound in members of the legume family.Daidzein and genistein, isoflavonoids pro-duced by soybean (Glycine max), effectivelyinduce Bradyrhizobium japonicum nod genes,but inhibit S. meliloti nod gene expression.S. meliloti nod genes are instead induced byluteolin (145). This specificity enables rhi-zobia to distinguish their hosts from otherlegumes. The specific flavonoid not only in-duces nod gene expression, but also rhizo-bial chemotaxis. Nevertheless, other than theisoflavones, most flavonoids are not unique tolegumes. How do soil rhizobia recognize theirhost and initiate the symbiosis when non-legume plant species growing in the same areaare also sources of flavonoids? Apparently,
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 245
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
AMF: arbuscularmycorrhizal fungi
GUS:β-glucuronidase
CHS: chalconesynthase
once the flavonoids are perceived, anotherlevel of specificity comes into play. Flavonoidsare perceived as aglycones, which induce rhi-zobial nod genes by interacting with the geneproduct of nodD, a LysR-type regulator. Thisinteraction results in a conformational changein the NodD protein that allows it to bindto nod box elements in the promoters ofthe nod genes (142). The concerted expres-sion of these genes leads to the synthesisof Nod factor molecules, lipochitooligosac-charides, that usually consist of four or fiveβ-1,4 N-acetylglucosamines, with the termi-nal nonreducing sugar N-acylated by a 16–18carbon fatty acid. Nod factors can be chem-ically modified with acetate, sulfate, or car-bamoyl groups, or can have different sugars,such as arabinose, fructose, and substitutedfructose. The degree of saturation of the acyltail may also vary (142). The assemblage ofthese substitutions results in a specific Nodfactor that is recognized by the host legume.
Mycorrhizal associations. Unlike the selec-tive legume-rhizobial associations, arbuscu-lar mycorrhizal fungi (AMF) and plant rootsform associations in more than 80% of ter-restrial plants. This symbiotic relationship in-creases nutrient uptake, improving plant fit-ness, and in turn, the associated fungi extractlipids and carbohydrates from the host root(5, 130). Both AMF and rhizobial associationswith plants derive from a common ancestralplant-microbe interaction, likely of fungal ori-gin. This position is supported by the fact thatAMF and rhizobia share conserved proteinsthat regulate both AMF and rhizobial associ-ations with plants (107). AMF may recognizethe presence of a compatible host throughroot exudates, similar to recognition by rhi-zobia (125, 166). Evidence for a fungal sig-naling molecule that induces plant gene ac-tivation was obtained from experiments byKosuta et al. (102), in which fungal hyphaeand host roots were grown in close proxim-ity but physically separated by impenetrablemembranes. In this system, a Medicago EARLYNODULATION11 (ENOD11)-promoter::β-
glucuronidase (GUS) fusion, which is respon-sive to both AMF and a rhizobial Nod-factor(90), was activated at a distance from the fun-gal hyphae (102). This was the first experi-mental evidence for a postulated fungally de-rived, diffusible signaling molecule.
The critical developmental step in the lifecycle of mycorrhizal fungi is hyphal branch-ing, which ensures contact with the hostroot and establishment of symbiosis (38). Thebranch-inducing factor is a plant signalingmolecule that triggers hyphal morphogene-sis preceding successful root colonization (25,58). The development of an in vitro bioassayfor hyphal branching in germinating sporesfrom the genus Gigaspora (126) facilitated theanalysis of the chemical characteristics anddistribution of branching factor in the plantkingdom. Branch-inducing factor was presentin root exudates of all the mycotrophic plantstested, but absent in those of nonhost plants.Flavonoids have been ruled out as branch-ing factor candidates because root exudates ofmaize mutants deficient in chalcone synthase(CHS) show comparable activity to those ofthe wild type (25). Root exudates from phos-phate (P)-limited plants are more active thanthose from plants with sufficient P, suggest-ing that the production and/or exudation ofbranching factor in roots is regulated by Pavailability (126). Recently, a sesquiterpene,which triggers hyphal branching in dormantmycorrhizal fungi, was identified from Lo-tus japonicus root exudates (2), establishing anovel role for root exudates in plant root-mycorrhizal cross talk.
As described above, mycorrhizal fungi ex-tensively invade host root tissues upon per-ceiving a chemical response from the hostroots. However, the spread of mycorrhizalmycelium occurs only in the root cortex, sug-gesting that host plants exert control overfungal proliferation, confining it to specificroot tissues. Defense processes, which are-triggered in response to microbial invasion,are modulated in mycorrhizal roots (56).Most host plants show remarkably little cy-tological reaction to appressorium formation
246 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
or the first steps of root colonization (57).Some elements of plant defense response suchas phenylpropanoid biosynthesis, oxidativestress-induced enzymes, and pathogenesis-related (PR) genes are activated in mycor-rhizal roots. In most cases, however, thesedefense responses are weak, transient, orstrictly localized, differing from those inplant-pathogen interactions (57). Transcriptsencoding enzymes of the flavonoid biosyn-thetic pathway, phenylalanine ammonia lyase(PAL), and chalcone synthase (CHS), butnot the defense-specific enzyme isoflavonereductase (IFR), are induced specifically incells containing arbuscules in M. truncat-ula. This induction may reflect biosynthe-sis of flavonoid compounds that stimulatethe growth of mycorrhizal fungi rather thanproduction of antimicrobial phytoalexins (68,69). Changes in the profiles of antioxida-tive enzymes such as superoxide dismutase(SOD), catalases, and peroxidases have alsobeen observed in mycorrhizal roots (19, 136).A recent study by Lanfranco et al. (106) de-scribes the cloning and characterization ofa CuZnSOD gene from Gigaspora margaritaand presents evidence that this gene is dif-ferentially expressed during the fungal lifecycle. The study also showed that the ex-pression levels of G. margarita CuZnSOD areenhanced following exposure to plant rootexudates.
Plant growth–promoting bacteria. Bacte-ria thrive on abundant nutrients in the rhizo-sphere and some of these rhizobacteria pro-vide benefits to the plant, resulting in plantgrowth stimulation (63). Bacteria are likelyto locate plant roots through cues exudedfrom the root, and root exudates such as car-bohydrates and amino acids stimulate PGPBchemotaxis on root surfaces (162). Root exu-dates also influence flagellar motility in somerhizospheric bacteria (39). To test the hypoth-esis that motility was induced by chemotaxistoward exudate components, cheA mutants,motile but defective in flagella-driven chemo-taxis, were constructed in four strains of Pseu-
PAL: phenylalanineammonia lyase
IFR: isoflavonereductase
Phytoalexins: toxiccompoundsproduced by higherplants in response toattack by pathogensand to other stresses;sometimes referredto as plantantibiotics, butrather nonspecific,having a generalfungicidal andbacteriocidal action
SOD: superoxidedismutase
MOMP: MajorOuter MembraneProtein
Biotransform: thetransformation of amaterial by microbialaction
ISR: inducedsystemic resistance
domonas fluorescens, a known PGPB (112, 113).Relative to wild-type bacteria, mutants had astrongly reduced ability to competitively col-onize roots (39). Thus, chemotaxis appearsto be important for competitive colonizationby extracellular PGPB. The bacterial MajorOuter Membrane Protein (MOMP) also playsan important role in early host recognition.MOMPs from Azospirillum brasilense bind tomembrane-immobilized root extracts fromseveral plant species with differing affinities.The A. brasilense MOMP showed stronger ad-hesion to extracts of cereals than extracts oflegumes and tomatoes, and may act as an ad-hesin involved in root adsorption and cell ag-gregation of the bacterium (27).
Some PGPB produce phytostimulators,which directly enhance plant growth. In addi-tion to fixing atmospheric nitrogen, Azospiril-lum spp. secrete phytohormones such as aux-ins, cytokinins, and gibberellins (163). Thereis the exciting possibility that most PGPB arecapable of producing growth regulators con-tinuously, provided that precursors of phy-tohormones are available in the rhizosphere.Root exudates could supply the pool of pre-cursors for PGPBs to biotransform. An inter-esting report describes the mapping of sugarand amino acid availability in the root exu-dates of Avena barbata (85). The study showedthe availability of tryptophan mainly near theroot tip region. Tryptophan is the precursorfor a major auxin, indole 3-acetic acid (33),suggesting that PGPB could exploit root ex-udate pools for various precursors of growthregulators.
Other rhizobacteria create “suppressivesoils” by controlling plant diseases caused bysoil fungi and bacteria. The mechanisms re-sponsible for this biocontrol activity includecompetition for nutrients, niche exclusion, in-duced systemic resistance (ISR), and the pro-duction of antifungal metabolites. The bio-control agents that are best characterized atthe molecular level belong to the genus Pseu-domonas. Most of the identified Pseudomonasbiocontrol strains produce antifungal metabo-lites, of which phenazines, pyrrolnitrin,
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 247
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
DAPG:2,4-diacetylphloro-glucinol
CMV: Cucumbermosaic virus
RA: rosmarinic acid
2,4-diacetylphloroglucinol (DAPG), and py-oluteorin are most frequently detected. How-ever, antifungal metabolites belonging to theclass of cyclic lipopeptides, such as viscosi-namide (127) and tensin (128), have also beendiscovered. Viscosinamide prevents infectionof Beta vulgaris L. (sugarbeet) by Pythiumultimum (170). Arabidopsis thaliana ecotypeColumbia plants (Col-0) treated with thePGPBs Serattia marcescens strain 90–166 andBacillus pumilus strain SE34 developed mi-nor disease symptoms upon infection with theCucumber mosaic virus (CMV) (156). Thestudy also showed that the acquired resistancein Arabidopsis plants to CMV by B. pumilusstrain 90–166 is caused by adapting a signal-ing pathway for virus protection that is in-dependent of salicylic acid (156). Finally, itwas reported that some of the known gram-positive biocontrol PGPBs (such as B. subtilis6051 strain) assist plants in evading a gram-negative plant pathogen, Pseudomonas syringaepv. tomato DC3000, by forming a protectivebiofilm on A. thaliana roots limiting pathogenaccess to the root surface and by producing anantimicrobial cyclic lipopeptide surfactin (7).
Negative Plant-Microbe Interactions
Antimicrobial effects. Plant root exudatessubstantially increase microbial activity in therhizosphere (134). The role root exudatesplay in pathogenesis of root-infecting bac-teria and fungi, however, has not been fullyappreciated, in part because of inadequatemethods available for analysis. Just as sym-biotic root-microbe interactions depend onsecondary metabolites in root exudates forinitiation and development of beneficial as-sociations, the survival of physically vulnera-ble root cells under continuous attack frompathogenic microorganisms depends on “un-derground chemical warfare” mediated byplant secretion of phytoalexins, defense pro-teins, and other as yet unknown chemicals (8,9, 50). Arabidopsis, rice, corn, soybean, and themodel legume Medicago truncatula, which havebeen subject to intensive sequencing efforts,
are, collectively, rich sources of antimicro-bial indole, terpenoid, benzoxazinone, andflavonoid/isoflavonoid natural products. Theunexplored chemodiversity of root exudatesin all these genetically tractable species is anobvious place to search for novel biologicallyactive compounds, including antimicrobials.
Bais et al. (11) identified rosmarinic acid(RA), a caffeic acid ester, in the root exudatesof hairy root cultures of sweet basil (Ocimumbasilicum) elicited using fungal cell wall ex-tracts from Phytophthora cinnamoni. Basil rootsalso exuded RA by fungal in situ challengewith Pythium ultimum, and RA demonstratedpotent antimicrobial activity against an ar-ray of soil-borne microorganisms, includingan opportunistic plant pathogen Pseudomonasaeruginosa (11). Brigham et al. (22) reportedthat Lithospermum erythrorhizon hairy rootsshowed elicited, cell-specific production ofpigmented naphthoquinones that had biolog-ical activity against soil-borne bacteria andfungi. These findings strongly suggest the im-portance of root exudates in defending the rhi-zosphere against pathogenic microorganisms.
Distinguishing between phytoalexins,which are produced in response to pathogenattack, and phytoanticipins, which are pro-duced constitutively and prior to attack, canbe difficult, because the terms describe invivo antimicrobial activity. In most cases,local concentrations of phytoalexins havenot been measured in cells that are in directcontact with invading microorganisms. Oneexception is a careful study of the cellular-and organ-level concentrations of differentclasses of phenylpropanoids in the rootexudates of A. thaliana. Phenylpropanoidlevels were significantly higher in rootsthat were challenged by nonhost bacterialpathogens (nonhost Pseudomonas syringaestrains) compared to host bacterial pathogens(P. syringae pv. tomato DC3000). Bacterialpathogens capable of infecting roots andcausing disease were resistant to thesecompounds, suggesting an important role ofthese compounds in defense against nonhostpathogens (6). In contrast, a recent study
248 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
revealed that concentrations of indolic andphenylpropanoid secondary metabolites inA. thaliana roots increased upon infectionwith the root-pathogenic oomycete Pythiumsylvaticum (15, 167). These results indicatethat roots differ greatly from root exudateswith regard to the nature and relative abun-dance of major soluble phenylpropanoidconstituents and with regard to responsesto applied biological stress. To date, only afew studies have been undertaken to gaininsights into the diverse metabolic realm ofantimicrobial root exudates. These recentfindings outline the current direction of thisfield, which may lead to the discovery of novelantimicrobial compounds and to unravelingas yet unknown root-microbe interactions inthe rhizosphere.
Quorum-sensing inhibitors and signalmimics. A number of studies have shownoverlap in the virulence factors that are re-quired for bacterial pathogenesis in bothmammalian and plant systems (86, 150, 151).In a large number of pathogenic bacteria,initiation of the production and secretion ofthese virulence factors is controlled by a phe-nomenon described as quorum-sensing (QS).Briefly, QS is a density-dependent regula-tory mechanism that was first described in theaquatic bacteria Vibrio fischeri as the signal-mediated induction of the lux genes respon-sible for bioluminescence (45). QS activa-tion is mediated by small autoinducer (AI)molecules, which are responsible for cell-cellcommunication, and the coordinated actionof many bacteria, including plant-associatedbacteria. The most commonly reported typeof autoinducer signals are N-acyl homoserinelactones (AHLs) (177), although half a dozenother molecules, including diketopiperazinesin several gram-negative bacteria (81), a fura-nosyl borate diester in Vibrio harveyi (32), andγ-butyrolactone in Streptomyces (191), havealso been implicated in density-dependentsignaling. Typically, a basal level of AHLs areconstitutively synthesized until a thresholdpopulation of bacteria has been achieved, at
Quorum-sensing(QS): thedensity-dependentmechanism used bymany bacteria toregulate geneexpression in acoordinated manner
AI: autoinducermolecules
AHL: N-acylhomoserine lactones
which point these molecules serve as ligandsto a global transcription regulator (LuxR orLuxR-like proteins) that activates many QS-controlled genes, including virulence factors.The rhizosphere contains a higher proportionof AHL-producing bacteria as compared tobulk soil, suggesting that they play a role incolonization (48). This leads to the specula-tion that plants could be using root-exudedcompounds in the rhizosphere to take advan-tage of this bacterial communication systemand influence colonizing communities. Dis-covery and characterization of these plant-secreted compounds could have important bi-ological implications in both agriculture andmedicine.
Since the discovery of penicillin, only alimited selection of new antibiotics have beendiscovered or synthesized for treating bacte-rial infections. These antibiotics work by in-terfering with specific metabolic events thatultimately culminate in the death of the bac-teria. However, selective pressure exerted bythis approach has resulted in the survival ofantibiotic-resistant bacterial strains. This hascreated an urgent need for new strategiesto control bacterial infections (74). A recenttrend in drug discovery has been to search forcompounds that are capable of inhibiting orinterfering with QS in pathogenic bacteria.QS inhibitors may prove to be valuable treat-ments for bacterial infections because they de-crease selective pressure by having little ef-fect on bacterial growth and survival, whiledownregulating the production of antibiotic-resistant biofilms and bacterial toxins (75, 76).Because most bacteria are naturally present insoil, yet only a handful of these bacteria havebecome successful plant pathogens, it standsto reason that plants, using their staggeringarray of root-secreted phytochemicals, mayhave evolved the ability to interfere with bac-teria via their QS systems.
Indeed, a fair body of evidence suggeststhat cross talk between plants and bacteriamay occur through QS signal mimics. QSmediates several plant-microbe interactions,both pathogenic and beneficial. The first
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 249
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
described examples of plant-secreted QSmimics were halogenated furanones producedby the marine red algae, Delisea pulchra (59).These compounds are structurally similarto bacterial AHLs and are capable of in-terfering with QS-controlled processes suchas swarming and bioluminescence (59), aswell as production of virulence factors andbiofilm formation in Pseudomonas aeruginosa(76). These particular compounds displacedtritiated AHLs from E. coli cells engineeredto overproduce LuxR receptors (116), lead-ing to reduced LuxR activity by destabiliz-ing this protein, and resulting in acceleratedproteolytic degradation (115). Furthermore,halogenated furanone concentrations foundon the algal surface were sufficient to preventgram-negative bacteria from colonizing algalthalli (43, 97).
AHL signal mimics have also been found insecretions of higher plants and in the unicel-lular green algae, Chlamydomonas reinhardtii,but their exact chemical nature has not beenidentified (55, 168, 169). Limited studies haveshown that several higher plants, includingPisum sativum (pea), Coronilla varia (crownvetch), Medicago truncatula, Oryza sativa (rice),Glycine max (soybean), and Lycopersicum lycop-ersicon (tomato), all contain components intheir exudates that are capable of activating bi-oluminescence in several QS reporter strains(169). These compounds partitioned into po-lar solvents, suggesting that they are not struc-turally similar to the AHLs, and probably in-teract with bacterial QS systems differentlythan structural analogues such as the halo-genated furanones. These signal mimics ap-pear to stimulate QS-controlled processes inmost cases. For instance, swarming in Ser-ratia liquefaciens appeared to be specificallyinduced by P. sativum exudates as well asseveral other plant compounds, as indicatedby parallel induction of both swarming andswrA gene expression and synthesis of ser-rawettin, a lipopeptide surfactant required forsurface swimming (46). In addition, exudatesof several other plants activated biolumines-cence in LuxRI’, AhyRI’, and LasRI’ plasmid
reporters. On the other hand, pea seedling ex-udates inhibited AHL-controlled behaviors inChromobacterium violaceum (169), and a puri-fied AHL mimic from C. reinhardtii specifi-cally stimulated the LasR receptor in P. aerug-inosa; however, the effect on Sinorhizobiummeliloti was ambiguous, with some QS-relatedproteins being stimulated and others beingsuppressed (168). These data hint that QS sig-nal mimics may be widespread in the plantkingdom, and suggest that these mimic com-pounds interact specifically with different QSreceptors from bacteria, leading to either theactivation of transcription of QS-controlledgenes or the destabilization and degradationof the receptor protein (14).
Although QS signal mimics have beenfound in a range of plant species, they appearto be particularly prevalent among nodulat-ing plants, such as P. sativum, C. varia, andM. truncatula. As previously discussed, an in-tricate two-way signaling between nitrogen-fixing rhizobia and leguminous host plantsis required to form a symbiotic relationship.The nodulating plant M. truncatula has theability to detect and respond to nanomolarconcentrations of bacterial AHLs from bothS. meliloti and P. aeruginosa (119). Proteomeanalysis revealed significant changes in the ac-cumulation of more than 150 proteins in re-sponse to these bacterial AHLs, with aboutone third of those proteins showing distinctdifferences in terms of direction or magni-tude of change in accumulation, or timing ofthe response to the different AHLs. This sug-gests that a general set of genes is activatedin response to bacterial AHLs, but that theplant can also differentiate between AHLs toactivate specific genes. Exposure to C6-HL,the principal AHL produced by several bacte-rial species, including some Rhizobium strains,also led to increased secretion of AHL mimicsin exudates of M. truncatula (119). Althoughdirect proof remains elusive, indirect lines ofevidence suggest that leguminous plants mayhave evolved the ability to secrete AHL mim-ics as a means of increasing the efficiency oftheir nitrogen-fixing symbionts while possibly
250 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
confusing would-be pathogens by causingthem to activate QS-controlled genes beforethere is a sufficiently large number of bacteriato overcome host defenses.
Ecological Plant-MicrobeInteractions
Plant-microbe interactions in the rhizosphereare responsible for a number of intrinsic pro-cesses such as carbon sequestration, ecosys-tem functioning, and nutrient cycling (159).The composition and quantity of microbesin the soil influence the ability of a plant toobtain nitrogen and other nutrients. Plantscan influence these net ecosystem changesthrough deposition of secondary metabolitesinto the rhizosphere that attract or inhibitthe growth of specific microorganisms. Thisrhizodeposition, made up of small-molecularweight metabolites, amino acids, secreted en-zymes, mucilage, and cell lysates, can rangefrom less than 10% of the net carbon as-similation by a plant to as much as 44% ofa nutrient-stressed plant’s total carbon (64,138). Soil microbes utilize this abundant car-bon source, thereby implying that selectivesecretion of specific compounds may encour-age beneficial symbiotic and protective re-lationships whereas secretion of other com-pounds inhibit pathogenic associations (6,80, 81).
Fons et al. (51) demonstrated that theycould change the microbial population dy-namics in the rhizosphere of Trifolium subter-raneum (clover) by adding 1% saponin fromGypsophila paniculata. Aquaspirrillum spp., typ-ically found in G. paniculata rhizospheres,became the dominant microbe in the T. sub-terraneum rhizosphere. Furthermore, Chry-seomonas spp. and Acinetobacter spp., the twopreviously dominant bacteria found in theT. subterraneum rhizosphere, were signifi-cantly decreased (51). Although there were noapparent negative effects on T. subterraneumcolonized by Aquaspirillium spp., other studieshave shown that changes in the microbial pop-ulations colonizing a plant’s rhizosphere can
Rhizoremediation:the contribution ofrhizospheremicrobes to thedegradation ofenvironmentalpollutants
have detrimental or beneficial effects. Call-away et al. (28) showed that fungicide treat-ments affected the interactions between theinvasive weed Centaurea maculosa and neigh-boring plant species. For instance, C. macu-losa biomass was increased in untreated soilswhen growing with two native grass species,Festuca idahoensis and Koelaria cristata; how-ever, this effect was not seen when C. mac-ulosa was grown alone or with these twograsses in Benomyl-treated soils. This indi-rectly suggests that mycorrhizal fungi associ-ated with these grasses favor the growth ofC. maculosa. However, when the same exper-iment was conducted using C. maculosa andthe forb, Gaillardia aristata, the opposite ef-fect was observed, with G. aristata-associatedfungi apparently having detrimental effects onC. maculosa growth. None of the beneficial ordetrimental effects were seen when C. mac-ulosa was grown in the presence of differentsoil microbial communities when competingplants were absent, indicating that these ef-fects are not direct, but part of more complexecosystem-level interactions.
Plant root exudates also affect the level ofcontamination found in soil and ground waterfrom various environmental pollutants. Thisrhizoremediation results from root exudate-mediated stimulation of bacterial growth andsurvival, resulting in more efficient degra-dation of environmental pollutants (103).In addition, root colonization of pollutant-degrading bacteria allows penetration andspread of these beneficial bacteria to other ar-eas of the soil. This naturally occurring pro-cess is effective for degradation of a variety ofenvironmental pollutants. For example, Pseu-domonas putida strains associated with root sys-tems of Zea mays (corn) and Triticum aestivum(wheat) effectively rhizoremediate soils con-taining 3-methyl benzoate and 2,4-D, respec-tively (95, 154). To enhance this process, selectpairings of specific plant species and bacterialspecies or communities that would allow evenmore efficient and targeted degradation of en-vironmental contaminants are being sought(103, 104).
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 251
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
Direct and Indirect Effects of RootExudates on Rhizosphere Nematodes
As described above, root exudates provide asource of organic carbon to soil microbes,leading to abundant microbial populations inthe rhizosphere (53). Microbial-feeding ne-matodes take advantage of these dense mi-crobial populations as a food source and in-crease microbial turnover, and thus nutrientsupply, to the plant when digesting microbes(65). Plant species and environmental condi-tions greatly affect the quality and quantity ofcarbon and nutrient sources secreted into therhizosphere and the structure of the microbialcommunity around roots, but the influence ofthese factors on microbe-nematode interac-tions is still unknown.
Root-feeding nematodes may participatein complex interactions with roots and soilmicrobes. Rovira et al. (155) estimated that,despite the large microbial populations in therhizosphere, bacteria occupy <10% of theroot surface and that fungal hyphal densitiesare only 12–14 mm m−2 root. At such densi-ties, mobile nematodes may readily avoid ne-matode microbial pathogens and select uncol-onized sections of root on which to feed. Inaddition, the accumulation of root-secretednematicidal compounds may be avoided byparasitic nematodes. Until recently, there waslittle work on the impact of root exudates onrhizosphere interactions between plant roots,microbes, and nematodes. Using a 14C pulse-labeling technique, Yeates et al. (193) demon-strated that infection of white clover (Tri-folium repens) roots by Heterodera trifolii andvarious other nematodes leads to a significantincrease in photosynthetically fixed 14C in soilmicrobial biomass. These results indicate thatwhite clover plants infected by plant-parasiticnematodes generally release more organiccompounds into the rhizosphere. Thus, in-creasing carbon translocation to the soil mi-crobial biomass as a consequence of the activ-ity of root-feeding nematodes may be anothermechanism by which microfaunal grazing en-hances microbial turnover. In addition, the ef-
fects of rhizosphere nematodes on the qualityand quantity of root exudates in turn influencethe activity of both plant pathogenic and ben-eficial microorganisms in the rhizosphere (20,93). Roots infected with Meloidogyne incognitaact as metabolic sinks, and symplastic trans-port of nutrients from the phloem to thefeeding cell, and ultimately the nematode, re-sults in increased leakage into the rhizospherecompared to healthy plants (42). Exudatesfrom tomato roots infected with M. incognitacontain more water-soluble 14C and largerconcentrations of several metal ions thanthose from healthy roots (173). Associatedchanges in the rhizospheric carbon:nitrogen(C:N) ratio alter the trophic state of Rhizoc-tonia solani, making the fungus a pathogen.The importance of nematode-associated in-creases in root exudate concentrations and al-tered nutrient ratios to interactions betweennematodes and microbial pathogens are notyet known.
Most knowledge of microbe-nematode in-teractions in the rhizosphere has been de-rived from research with rhizobia, mycor-rhizal fungi, and plant pathogens (93). Suchresearch has clearly demonstrated complextritrophic webs, in which nematodes and mi-croorganisms act in competitive, additive, orsynergistic associations to affect the planthost. In addition, a recent study has redefinedthe beneficial association of the tritrophicinteractions between plant roots, microbes,and nematodes. This new study shows thatsoil-dwelling nematodes, such as Caenorhab-ditis elegans, may mediate interactions be-tween roots and rhizobia in a positive way,leading to nodulation (82). Horiuchi et al.(82) found that C. elegans transfers the rhi-zobium species Sinorhizobium meliloti to theroots of the legume Medicago truncatula in re-sponse to plant root-released volatiles thatattract the nematode. Thus, root-secretedvolatiles, in addition to other root-secretedchemicals, may also play an important rolein multitrophic interactions. Research onthe tritrophic interactions between plants,
252 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
nematodes, and microbial pathogens will con-tribute much to our understanding of the sig-naling systems mediated by root exudates inthe rhizosphere.
METHODS TO STUDYINTERACTIONS MEDIATED BYROOT EXUDATES
The biggest hurdle to the study of plant-plantand plant-microbe interactions mediated byroot exudates is the underground nature ofthe roots. The study of root exudation re-quires knowledge of both the structure andfunction of a root system, as well as a mean-ingful assessment of the rhizospheric commu-nity. One must consider the abundance anddistribution of plant species and the func-tional diversity and redundancy present in mi-crobial communities. Some striking studieshave used the exudation of fluorescent com-pounds as a marker for such interactions (10).The majority of conventional methods usedin studying plant-plant and plant-microbe in-teractions involve in vitro tissue culture tech-niques (7, 10). Briefly, to study plant-plant in-teractions mediated by root exudates, plantsare regrown in vitro in an aerated liquid me-dia, root exudates are harvested, and concen-trated exudates are tested for phytotoxicity onseedlings that share the rhizosphere of thetested plant (Figure 3). This methodologicalpartitioning of root exudates has led to the iso-lation of a number of phytotoxins secreted byinvasive plants. However, the full complexityof interactions occurring in a natural rhizo-sphere is eliminated in this system, and thusresults should be viewed with caution (12, 16,164, 176). There are two different ways to ex-tract phytochemicals from root exudates: Oneinvolves extraction specifically for polar com-pounds, usually with methanol, whereas thesecond method targets nonpolar compoundsusing nonpolar solvents. This differential par-titioning of root exudates results in isolation ofvarious classes of chemical compounds, suchas flavonoids, quinalones, carbolines, and ter-penes (12, 16, 164, 176).
CLSM: confocallaser scanningmicroscopy
Identifying plant-produced antimicro-bials, profiling rhizosphere microbes, andstudying microbial colonization requiresseveral methodologies. The diversity ofmetabolic functions possessed by micro-bial communities is often examined usingBIOLOG GN substrate utilization assays(41), which assess the ability of the com-munity as a whole to utilize select carbonsubstrates. A DNA microarray technique forthe simultaneous identification of ecologicalfunction and phylogenetic affiliation of mi-crobial populations has also been developed(96). This approach permits the assessmentof growth rate and substrate utilization ofindividual microbial populations within acommunity. Advances in microscopy havealso greatly facilitated study of root-microbeinteractions. Confocal laser scanning mi-croscopy (CLSM), in combination withvarious other fluorescent markers and re-porter gene systems, is used to observe andmonitor rhizosphere bacterial populations onthe root surface. Most of these studies havebeen conducted with biocontrol microbes,specifically gram-negative Pseudomonas spp.(112). Using a combination of immunoflu-orescence and an rRNA-targeting probethat monitors the presence and metabolicactivity of P. fluorecens DR54, Lubeck et al.(111) showed that bacteria at the root tipare the most metabolically active and thatendogenous bacteria enter the rhizospheretwo days after inoculation. Visualization ofinteractions among carrot roots, mycorrhizalmycelium, and P. fluorescens CHA0 showedthat mucoid mutant strains of CHA0 adheremuch better to the root, indicating that acidicextracellular polysaccharides can enhanceroot colonization (18). Also by using mi-croscopy, it was shown that a gram-positivebiocontrol bacteria B. subtilis competes forspace against a pathogenic gram-negativebacteria P. syringae on Arabidopsis root surfaces(7).
The screening and functional identifica-tion of the diverse array of natural com-pounds present in root exudates that affect
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 253
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
O
OH
OH
In vitro plant cultures on
rotatory shaker
In vitro plantcultures on
rotatory shaker
Collection of root
exudates
Collection of root
exudates
Analyticalseparation
Bioassay on the susceptible plant species
Positive hits
Bioassay with the individualfractions
Bioassay on the susceptible plant species
Organic extraction
of the crude exudates
Compound characterization
Bioassay on the susceptible plant species
Positive hits
Antimicrobial assay
Figure 3A flow chart representation of methods involved in collection, separation, bioassay, and candidcompound characterization from plant root exudates.
rhizospheric microbes is a daunting task. Un-til recently, only traditional in planta extrac-tion and subsequent testing of crude extractsdirectly on microbes was available. A caveatof this method is the inability to observe
direct interactions between plant roots andmicrobes. To bypass this shortfall, one couldgrow plants and microbes together under invitro conditions and observe the effect eithercomponent exerts on the other. This method
254 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
could also be used to identify antimicrobialsor QS mimics from plant root exudates. Stud-ies to observe global gene-expression levels inrhizospheric microbes upon interacting withroots and root exudates are also possible usinga number of microbes whose genomes havebeen sequenced. These studies would high-light the physiological functioning of micro-bial cells in a specific environment.
CONCLUSION
We presented a partial picture of the inter-actions that occur in the rhizosphere and therole of root exudates in mediating some ofthese processes. However, our understand-ing of these interactions is incomplete due tothe difficulty of studying underground pro-cesses under controlled yet realistic condi-tions. Thus, developing novel methodologiesto study rhizosphere ecology under naturalconditions is needed and will require collab-oration between plant biologists, ecologists,and soil scientists to develop rhizotron sys-tems where biochemical and molecular biol-ogy studies could be performed on site. It is
clear that our understanding of root-mediatedprocesses has moved beyond the classical be-lief that the sole functions of roots are an-chorage and uptake of water and nutrients.It is now understood that roots are rhizo-sphere ambassadors, facilitating communica-tion between the plant and other organismsin the soil. Ecological knowledge indicatesthat aboveground interactions could poten-tially be translated to belowground responsesin plants. What does this mean at the rhi-zosphere level? What is the effect of above-ground herbivory on the ability of roots toinitiate microbial symbiosis or to fight mi-crobial attack? A clear understanding of themolecular process involved in the actual se-cretion of phytochemicals by roots is alsoneeded in order to develop molecular mark-ers for this process. Finally, synthesis of theknowledge of root exudation from the molec-ular to the ecosystem scale will potentiallylead to the development of better plants ca-pable of absorbing more nutrients, detoxify-ing soils more efficiently, or more effectivelywarding off invasive weeds and pathogenicmicrobes.
ACKNOWLEDGMENTS
This work was supported by grants from the U.S. Department of Defense-SERDP (SI-1388to J.M.V.) and the National Science Foundation (NSF-IBN 0335203, J.M.V. and S.G.).
LITERATURE CITED
1. Abrahim D, Braguini WL, Kelmer-Bracht AM, Ishii-Iwamoto EL. 2000. Effects of fourmonoterpenes on germination, primary root growth, and mitochondrial respiration ofmaize. J. Chem. Ecol. 26:611–24
The firstidentification of acompound thatinduces branchingin mycorrhizalfungi.
2. Akiyama K, Matsuzaki K, Hayashi H. 2005. Plant sesquiterpenes induce hyphalbranching in arbuscular mycorrhizal fungi. Nature 435:824–27
3. Anaya AL, Pelayo-Benavides HR. 1997. Allelopathic potential of Mirabilis jalapa L.,(Nyctaginaceae): effects on germination, growth and cell division of some plants. Al-lelopathy 4:57–68
4. Apel K, Hirt H. 2004. Reactive oxygen species: metabolism, oxidative stress, and signaltransduction. Annu. Rev. Plant Biol. 55:373–99
5. Bago B, Pfeffer PE, Abubaker J, Jun J, Allen JW, et al. 2003. Carbon export from arbus-cular mycorrhizal roots involves the translocation of carbohydrate as well as lipid. PlantPhysiol. 131:1496–507
Rhizotron: a deviceused to view andmanipulateorganisms in anatural rhizosphere
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 255
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
Evidence that theability of a bacterialstrain to colonize aplant’s root systemand subsequentlyto become apathogen is basedon its ability toovercome theplantsantimicrobial rootexudates.
6. Bais HP, Prithiviraj B, Jha AK, Ausubel FM, Vivanco JM. 2005. Mediation ofpathogen resistance by exudation of antimicrobials from roots. Nature 434:217–21
7. Bais HP, Fall R, Vivanco JM. 2004. Biocontrol of Bacillus subtilis against infection ofArabidopsis roots by Pseudomonas syringae is facilitated by biofilm formation and surfactinproduction. Plant Physiol. 134:307–19
8. Bais HP, Park SW, Weir TL, Callaway RM, Vivanco JM. 2004. How plants communicateusing the underground information superhighway. Trends Plant Sci. 9:26–32
9. Bais HP, Vepachedu R, Gilroy S, Callaway RM, Vivanco JM. 2003. Allelopathy and exoticplant invasion: from molecules and genes to species interactions. Science 301:1377–80
10. Bais HP, Park SW, Stermitz FR, Halligan KM, Vivanco JM. 2003. Exudation of fluores-cent beta-carbolines from Oxalis tuberosa L. roots. Phytochemistry 61:539–43
11. Bais HP, Walker TS, Schweizer HP, Vivanco JM. 2002. Root specific elicitation andantimicrobial activity of rosmarinic acid in hairy root cultures of sweet basil (Ocimumbasilicum L.). Plant Physiol. Biochem. 40:9837
12. Bais HP, Walker TS, Stermitz FR, Hufbauer RA, Vivanco JM. 2002. Enantiomeric-dependent phytotoxic and antimicrobial activity of ( ± )-catechin. A rhizosecreted racemicmixture from spotted knapweed. Plant Physiol. 128:1173–79
13. Bais HP, Loyola-Vargas VM, Flores HE, Vivanco JM. 2001. Root-specific metabolism:the biology and biochemistry of underground organs. In Vitro Plant. 37:730–41
14. Bauer WD, Mathesius U. 2004. Plant responses to bacterial quorum-sensing signals.Curr. Opin. Plant Biol. 7:429–33
15. Bednarek P, Schneider B, Svatos A, Oldham NJ, Hahlbrock K. 2005. Structural complex-ity, differential response to infection, and tissue specificity of indolic and phenylpropanoidsecondary metabolism in Arabidopsis roots. Plant Physiol. 138:1058–70
16. Belz RG, Hurle K. 2005. Differential exudation of two benzoxazinoids–one of the deter-mining factors for seedling allelopathy of Triticeae species. J. Agric. Food Chem. 53:250–61
17. Bertin C, Yang XH, Weston LA. 2003. The role of root exudates and allelochemicals inthe rhizosphere. Plant Soil 256:67–83
18. Bianciotto V, Andreotti S, Balestrini R, Bonfante P, Perotto S. 2001. Mucoid mutantsof the biocontrol strain Pseudomonas fluorescens CHA0 show increased ability in biofilmformation on mycorrhizal and nonmycorrhizal carrot roots. Mol. Plant- Microbe Interact.14:255–60
19. Blilou I, Ocampo JA, Garcia-Garrido JM. 2000. Induction of Ltp (lipid transfer protein)and Pal (phenylalanine ammonia-lyase) gene expression in rice roots colonized by thearbuscular mycorrhizal fungus Glomus mosseae. J. Exp. Bot. 51:1969–77
20. Bowers JH, Nameth ST, Riedel RM, Rowe RC. 1996. Infection and colonization ofpotato roots by Verticillium dahliae as affected by Pratylenchus penetrans and P. crenatus.Phytopathology 86:614–21
21. Brady NC, Weil RR. 1999. The Nature and Property of Soils. Upper Saddle Hall, NJ:Prentice Hall
22. Brigham LA, Michaels PJ, Flores HE. 1999. Cell-specific production and antimicrobialactivity of naphthoquinones in roots of Lithospermum erythrorhizon. Plant Physiol. 119:417–28
23. Brimecombe MJ, De Leij Frans AAM, Lynch JM. 2001. Nematode community struc-ture as a sensitive indicator of microbial perturbations induced by a genetically modifiedPseudomonas fluorescens strain. Biol. Fertil. Soils 34:270–75
24. Bruin J, Sabelis MW. 2001. Meta-analysis of laboratory experiments on plant-plant in-formation transfer. Biochem. Sys. Ecol. 29:1089–102
256 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
25. Buee M, Rossignol M, Jauneau A, Ranjeva R, Becard G. 2000. The pre-symbiotic growthof arbuscular mycorrhizal fungi is induced by a branching factor partially purified fromplant root exudates. Mol. Plant-Microbe Interact. 13:693–98
26. Buer CS, Muday GK. 2004. The transparent testa4 mutation prevents flavonoid synthesisand alters auxin transport and the response of Arabidopsis roots to gravity and light. PlantCell 16:1191–205
27. Burdman S, Dulguerova G, Okon Y, Jurkevitch E. 2001. Purification of the major outermembrane protein of Azospirillum brasilense, its affinity to plant roots, and its involvementin cell aggregation. Mol. Plant- Microbe Interact. 14:555–58
28. Callaway RM, Thelen GC, Barth S, Ramsey, Gannon JE. 2004. Soil fungi alter inter-actions between the invader Centaurea maculosa and North American natives. Ecology85:1062–71
29. Callaway RM, Aschehoug ET. 2000. Invasive plants versus their new and old neighbors:a mechanism for exotic invasion. Science 290:521–23
30. Chamberlain K, Guerrieri E, Pennacchio F, Pettersson J, Pickett JA, et al. 2001. Canaphid-induced plant signals be transmitted aerially and through the rhizosphere? Biochem.Syst. Ecol. 29:1063–74
31. Chang M, Netzly DH, Butler LG, Lynn DG. 1986. Chemical-regulation of distance -characterization of the 1st natural host germination stimulant for Striga asiatica. J. Am.Chem. Soc. 108:7858–60
32. Chen X, Schauder S, Potier N, van Drosselaer A, Pelczer I, et al. 2002. Structural iden-tification of a bacterial quorum-sensing signal containing boron. Nature 415:545–49
33. Cooke TJ, Poli D, Sztein AE, Cohen JD. 2002. Evolutionary patterns in auxin action.Plant Mol. Biol. 49:319–38
34. Cruz-Ortega R, Ayala-Cordero G, Anaya AL. 2002. Allelochemical stress produced bythe aqueous leachate of Callicarpa acuminata: effects on roots of bean, maize, and tomato.Physiol. Plant. 116:20–27
35. Czarnota MA, Paul RN, Dayan FE, Nimbal CI, Weston LA. 2001. Mode of action,localization of production, chemical nature, and activity of sorgoleone: a potent PSIIinhibitor in Sorghum spp. root exudates. Weed Technol. 15:813–25
36. Czarnota MA, Rimando AM, Weston LA. 2003. Evaluation of root exudates of sevensorghum accessions. J. Chem. Ecol. 29:2073–83
37. Dakora FD, Phillips DA. 2002. Root exudates as mediators of mineral acquisition inlow-nutrient environments. Plant Soil 245:35–47
38. De Carvalho-Niebel F, Timmers AC, Chabaud M, Defaux-Petras A, Barker DG. 2002.The Nod factor-elicited annexin MtAnn1 is preferentially localized at the nuclear pe-riphery in symbiotically activated root tissues of Medicago truncatula. Plant J. 32:343–52
39. de Weert S, Vermeiren H, Mulders IH, Kuiper I, Hendrickx N, Bloemberg GV, et al.2002. Flagella-driven chemotaxis towards exudate components is an important trait fortomato root colonization by Pseudomonas fluorescens. Mol. Plant-Microbe Interact. 15:1173–80
40. Di Giovanni GD, Watrud LS, Seidler RJ, Widmer F. 1999. Fingerprinting of mixedbacterial strains and BIOLOG gram-negative (GN) substrate communities by enter-obacterial repetitive intergenic consensus sequence-PCR (ERIC-PCR). Curr. Microbiol.38:217–23
41. Dorhout RC, Gommers FJ, Kolloffel C. 1993. Phloem transport of carboxyfluoresceinthrough tomato roots infected with Meloidogyne incognita. Physiol. Mol. Plant Pathol. 43:1–10
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 257
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
42. Dicke M, Dijkman H. 2001. Within-plant circulation of systemic elicitor of induceddefence and release from roots of elicitor that affects neighbouring plants. Biochem. Syst.Ecol. 29:1075–87
42a. Du YJ, Poppy GM, Powell W, Pickett JA, Wadhams LJ, Woodcock CM. 1998. Identi-fication of semiochemicals released during aphid feeding that attract parasitoid Aphidiuservi. J. Chem. Ecol. 24:1355–68
43. Dworjanyn SA, de Nys R, Steinberg PD. 1999. Localization and surface quantificationof secondary metabolites in the red algae Delisea pulchra. Mar. Biol. 133:727–36
44. Dyer AR. 2004. Maternal and sibling factors induce dormancy in dimorphic seed pairsof Aegilops triuncialis. Plant Ecol. 172:211–18
45. Eberhardt A, Burlingame AL, Eberhardt C, Kenyon GL, Nealson KH, OppenheimerNJ. 1981. Structural identification of autoinducer of Phytobacterium fischeri luciferase.Biochemistry 20:2444–49
46. Eberl L, Molin S, Givscov M. 1999. Surface motility of Serratia liquefaciens MG1. J.Bacteriol. 181:1703–12
47. Einhellig FA. 1995. Mechanisms of action of allelochemicals in allelopathy. In Allelopathy:Organisms, Processes, and Applications, ed. Inderjit, KMM Dakshini, FA Einhellig, pp. 96.Washington, DC: Am. Chem. Soc.
48. Elasri M, Delorme S, Lemanceau P, Stewart G, Laue B, et al. 2001. Acyl-homoserine lac-tone production is more common among plant-associated Pseudomonas spp. than amongsoilborne Pseudomonas spp. Appl. Environ. Microbiol. 67:1198–209
49. Fate GD, Lynn DG. 1996. Xenognosin methylation is critical in defining the chemicalpotential gradient that regulates the spatial distribution in Striga pathogenesis. J. Am. Chem.Soc. 118:11369–76
51. Fons F, Amellal N, Leyval C, Saint-Martin N, Henry M. 2003. Effects of gypsophilasaponins on bacterial growth kinetics and on selection of subterranean clover rhizospherebacteria. Can. J. Microbiol. 49:367–73
52. Foreman J, Demidchik V, Bothwell JH, Mylona P, Miedema H, et al. 2003. Reactiveoxygen species produced by NADPH oxidase regulate plant cell growth. Nature 422:442–46
53. Foster RC. 1986. The ultrastructure of the rhizoplane and rhizosphere. Annu. Rev. Phy-topathol. 24:211–34
54. Galindo JCG, Hernandez A, Dayan FE, Tellez FA, Macias RN, Paul SO. 1999. Duke,Dehydrozaluzanin C, a natural sesquiterpenolide, causes rapid plasma membrane leakage.Phytochemistry 52:805–13
55. Gao M, Teplitski M, Robinson JB, Bauer WD. 2003. Production of substances by Med-icago truncatula that affect bacterial quorum sensing. Mol. Plant-Microbe Interact. 16:827–34
56. Garcia-Garrido JM, Ocampo JA. 2002. Regulation of the plant defence response in ar-buscular mycorrhizal symbiosis. J. Exp. Bot. 53:1377–86
57. Gianinazzi-Pearson V. 1996. Plant cell responses to arbuscular mycorrhizal fungi: gettingto the roots of the symbiosis. Plant Cell. 8:1871–83
58. Giovannetti M, Sbrana C, Silvia A, Avio L. 1996. Analysis of factors involved in fungalrecognition response to host-derived signals by arbuscular mycorrhizal fungi. New Phytol.133:65–71
258 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
The first exampleof a eukaryoticquorum-sensingsignal mimiccapable ofinterfering withbacterialquorum-sensingsystems.
59. Givskov M, Nys RD, Manefield M, Gram L, Maximilien R, Eberl L, et al. 1996.Eukaryotic interference with homoserine lactone-mediated prokaryotic signaling.J. Bacteriol. 178:6618–22
60. Gleba D, Borisjuk NV, Borisjuk LG, Kneer R, Poulev A, Skarzhinskaya M, et al. 1999.Use of plant roots for phytoremediation and molecular farming. Proc. Natl. Acad. Sci.USA 25:5973–77
61. Glinwood R, Pettersson J, Ahmed E, Ninkovic V, Birkett M, Pickett J. 2003. Change inacceptability of barley plants to aphids after exposure to allelochemicals from couch-grass(Elytrigia repens). J. Chem. Ecol. 29:261–74
62. Gonzalez VM, Kazimir J, Nimbal C, Weston LA, Cheniae GM. 1997. Inhibition of aphotosystem II electron transfer reaction by the natural product sorgoleone. J. Agric.Food Chem. 45:1415–21
63. Gray EJ, Smith DL. 2005. Intracellular and extracellular PGPR: commonalities anddistinctions in the plant-bacterium signaling processes. Soil Biol. Biochem. 37:395–410
64. Grayston SJ, Wang SQ, Campbell CD, Edwards AC. 1998. Selective influence of plantspecies on microbial diversity in the rhizosphere. Soil Biol. Biochem. 30:369–78
65. Griffiths BS. 1989. The role of bacterial feeding nematodes and protozoa in rhizospherenutrient cycling. Asp. Appl. Biol. 22:141–45
66. Guerrieri E, Poppy GM, Powell W, Rao R, Pennacchio F. 2002. Plant-to-plant commu-nication mediating in-flight orientation of Aphidius ervi. J. Chem. Ecol. 28:1703–15
67. Guyon V, Tang WH, Monti MM, Raiola A, Lorenzo GD, et al. 2004. Antisense pheno-types reveal a role for SHY, a pollen-specific leucine-rich repeat protein, in pollen tubegrowth. Plant J. 39:643–54
68. Harrison MJ. 2005. Signaling in the arbuscular mycorrhizal symbiosis. Annu. Rev. Mi-crobiol. 59:19–42
69. Harrison MJ. 1999. Molecular and cellular aspects of the arbuscular mycorrhizal sym-biosis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50:361–89
70. Deleted in proof71. Hawes MC, Gunawardena U, Miyasaka S, Zhao X. 2000. The role of root border cells
in plant defense. Trends Plant Sci. 5:128–3372. Hejl AM, Koster KL. 2004. The allelochemical sorgoleone inhibits root H+-ATPase and
water uptake. J. Chem. Ecol. 30:2181–9173. Hejl AM, Koster KL. 2004. Juglone disrupts root plasma membrane H+-ATPase activity
and impairs water uptake, root respiration, and growth in soybean (Glycine max) and corn(Zea mays). J. Chem. Ecol. 30:453–71
74. Hentzer M, Givskov M. 2003. Pharmacological inhibition of quorum sensing for thetreatment of chronic bacterial infections. J. Clin. Investig. 112:1300–7
75. Hentzer M, Wu H, Andersen JB, Riedel K, Rasmussen TB, Bagge N, et al. 2003. At-tenuation of Pseudomonas aeruginosa virulence by quorum-sensing inhibitors. EMBO J.22:3803–15
76. Hentzer M, Riedel K, Rasmussen TB, Heydorn A, Andersen JB, Parsek MR, et al. 2002.Inhibition of quorum sensing in Pseudomonas aeruginosa biofilm bacteria by a halogenatedfuranone compound. Microbiology 148:87–102
77. Hierro JL, Callaway RM. 2003. Allelopathy and exotic plant invasion. Plant Soil 256:29–3978. Hiltner L. 1904. Uber neure Erfahrungen und probleme auf dem gebeit der bodenback-
teriologie und unter besonderer berucksichtigung der grundungung und brache. Arb.Deut. Landwirsch Ges. 98:59–78
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 259
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
79. Hirsch AM, Bauer WD, Bird DM, Cullimore J, Tyler B, Yoder JI. 2003. Molecular signalsand receptors: controlling rhizosphere interactions between plants and other organisms.Ecology 84:858–68
80. Hoffland E, Findenegg G, Nelemans J, van den Boogaard R. 1992. Biosynthesis androot exudation of citric and malic acids in phosphate-starved rape plants. New Phytol.122:675–80
81. Holden MTG, Chhabra SR, de Nys R, Stead P, Bainton NJ, et al. 1999. Quorum-sensingcross-talk: isolation and chemical characterization of cyclic dipeptides from Pseudomonasaeruginosa and other gram-negative bacteria. Mol. Microbiol. 33:1254–66
82. Horiuchi JI, Prithiviraj B, Bais HP, Kimball BA, Vivanco JM. 2005. Soil nema-todes mediate positive interactions between legume plants and rhizobium bacteria.Planta 15:1–10
83. Inderjit, Callaway RM. 2003. Experimental designs for the study of allelopathy. Plant Soil256:1–11
84. Inderjit, Weiner J. 2001. Plant allelochemical interference or soil chemical ecology? Per-spect. Plant Ecol. Evol. Syst. 4:3–12
85. Jaeger JH, Lindow SE, Miller S, Clark E, Firestone, MK. 1999. Mapping of sugar andamino acid availability in soil around roots with bacterial sensors of sucrose and trypto-phan. Appl. Environ. Microbiol. 65:2685–90
86. Jha AK, Bais HP, Vivanco JM. 2005. Enterococcus faecalis uses mammalian virulence-related factors to exhibit potent pathogenicity in the Arabidopsis thaliana plant model.Infect. Immun. 73:464–75
87. Johnson JF, Vance CP, Allan DL. 1994. Phosphorus stress-induced proteoid roots showaltered metabolism in Lupinus albus. Plant Physiol. 104:657–65
88. Jones DL, Kuzyakov Y, Hodge A. 2004. Plant and mycorrhizal regulation of rhizodepo-sition. New Phytol. 163:459–80
89. Jose S, Gillespie AR. 1998. Allelopathy in black walnut ( Juglans nigra L.) alley cropping.I. Spatio-temporal variation in soil juglone in a black walnut-corn (Zea mays L.) alleycropping system in the midwestern USA. Plant Soil 203:191–97
90. Journet EP, van Tuinen D, Gouzy J, Crespeau H, Carreau V, et al. 2002. Exploring rootsymbiotic programs in the model legume Medicago truncatula using EST analysis. NucleicAcids Res. 30:5579–92
91. Kania A, Neumann G, Martinoia E, Langlade N. 2003. Phosphorus deficiency-inducedmodifications in citrate catabolism and in cytosolic pH as related to citrate exudation incluster roots of white lupin. Plant Soil 248:117–27
92. Keyes WJ, O’Malley RC, Kim D, Lynn DG. 2000. Signaling organogenesis in parasiticangiosperms: xenognosin generation, perception, and response. J. Plant Growth Regul.19:217–31
93. Khan MW, ed. 1993. Nematode Interactions. London: Chapman & Hall. 377 pp.94. Kim DJ, Kocz R, Boone L, Keyes WJ, Lynn DG. 1998. On becoming a parasite: evaluating
the role of wall oxidases in parasitic plant development. Chem. Biol. 5:103–1795. Kingsley MT, Fredrickson JK, Meting FB, Seidler RJ, 1994. Environmental restoration
using plant-microbe. In Bioremediation of Chlorinated and Polyaromatic Hydrocarbon Com-pounds, ed. RE Hinchee, A Leeson, L Semprini, S Kong. pp 287–92. Boca Raton, FL:Lewis
96. Kirk JL, Beaudette LA, Hart M, Moutoglis P, Klironomos JN, et al. 2004. Methods ofstudying soil microbial diversity. J. Microbiol. Methods 58:169–88
260 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
97. Kjelleberg S, Steinberg P, Givskov M, Gram L, Manefield M, de Nys R. 1997. Do marinenatural products interfere with prokaryotic AHL regulatory systems? Aquat. Microbial.Ecol. 13:85–93
98. Kneer R, Poulev AA, Olesinski A, Raskin I. 1999. Characterization of the elicitor-inducedbiosysnthesis and secretion of genistin from roots of Lupinus luteus. J. Exp. Bot. 50:1553–59
99. Deleted in proof.100. Kohler B, Hills A, Blatt MR. 2003. Control of guard cell ion channels by hydrogen
peroxide and abscisic acid indicates their action through alternate signaling pathways.Plant Physiol. 131:385–88
101. Kong CH, Liang WJ, Xu XH, Hu F, Wang P, Jiang Y. 2004. Release and activity ofallelochemicals from allelopathic rice seedlings. J Agric. Food Chem. 52:2861–65
102. Kosuta S, Chabaud M, Lougnon G, Gough C, Denarie J, Barker DG, Becard G.2003. A diffusible factor from arbuscular mycorrhizal fungi induces symbiosis-specificMtENOD11 expression in roots of Medicago truncatula. Plant Physiol. 131:952–62
103. Kuiper I, Lagendijk EL, Bloemberg GV, Lugtenberg BJJ. 2004. Rhizoremediation: abeneficial plant-microbe interaction. Mol. Plant-Microbe Interact. 17:6–15
104. Kuiper I, Bloemberg GV, Lugtenberg BJJ. 2001. Selection of a plant-bacterium pair anovel tool for rhizostimulation of polycyclic aromatic hydrocarbon degrading bacteria.Mol. Plant-Microbe Interact. 14:1197–205
105. Kwak JM, Mori IC, Pei ZM, Leonhardt N, Torres MA, et al. 2003. NADPH oxidaseAtrbohD and AtrbohF genes function in ROS-dependent ABA signaling in Arabidopsis.EMBO J. 22:2623–33
106. Lanfranco L, Novero M, Bonfante P. 2005. The mycorrhizal fungus Gigaspora margaritapossesses a CuZn superoxide dismutase that is up-regulated during symbiosis with legumehosts. Plant Physiol. 137:1319–30
107. Levy J, Bres C, Geurts R, Chalhoub B, Kulikova O, Duc G, et al. 2004. A putative Ca2+
and calmodulin-dependent protein kinase required for bacterial and fungal symbioses.Science 303:1361–64
108. Lipton DS, Blevins DG, Blanchar RW. 1987. Citrate, malate and succinate concentrationin exudates from P-sufficient and P-stressed Medicago sativa L. seedlings. Plant Physiol.85:315–17
109. Long SR. 2001. Genes and signals in the rhizobium-legume symbiosis. Plant Physiol.125:69–72
110. Deleted in proof111. Lubeck PS, Hansen M, Sorensen J. 2000. Simultaneous detection of the establishment of
seed-inoculated Pseudomonas fluorescens strain DR54 and native soil bacteria on sugar beetroot surfaces using fluorescence antibody and in situ hybridization techniques. FEMSMicrobiol. Ecol. 33:11–19
112. Lugtenberg BJ, Dekkers L, Bloemberg GV. 2001. Molecular determinants of rhizospherecolonization by Pseudomonas. Annu. Rev. Phytopathol. 39:461–90
113. Lugtenberg BJ, Chin-A-Woeng TF, Bloemberg GV. 2002. Microbe-plant interactions:principles and mechanisms. Antonie Van Leeuwenhoek. 81:373–83
114. Macias FA, Simonet AM, Molinillo JMG, Castellano D, Marin D, Oliveros-Bastidas A.2005. Structure-activity relationships (SAR) studies of benzoxazinones, their degradationproducts and analogues. Phytotoxicity on standard target species (STS). J. Agric. FoodChem. 53:538–48
115. Manefield M, Rasmussen TB, Hentzer M, Andersen JB, Steinberg P, Kjellberg S,et al. 2002. Halogenated furanones inhibit quorum-sensing through accelerated LuxRturnover. Microbiology 148:1119–27
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 261
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
116. Manefield M, de Nys R, Kumar N, Read R, Givskov M, et al. 1999. Evidence thathalogenated furanones from Delisea pulchra inhibit acylated homoserine lactone (AHL)-mediated gene expression by displacing the AHL signal from its receptor protein. Micro-biology 145:283–91
117. Marschner H. 1995. Mineral Nutrition of Higher Plants, Second Edition. London: Academic118. Masaoka Y, Kojima M, Sugihara S, Yoshihara T, Koshina M, Ichihara A. 1993. Dissolution
of ferric phosphate by alfalfa (Medicago sativa L.) root exudates. Plant Soil 155/156:75–78
The first evidencethat plants candifferentially detectand respond tobacterialquorum-sensingsignalingmolecules.
119. Mathesius U, Mulders S, Gao M, Teplitski M, Caetano-Anolles G, Rolfe BG, et al.2003. Extensive and specific responses of a eukaryote to bacterial quorum-sensingsignals. Proc. Natl. Acad. Sci. USA 100:1444–49
120. McQueen-Mason S, Cosgrove DJ. 1994. Disruption of hydrogen-bonding betweenplant-cell wall polymers by proteins that induce wall extension. Proc. Natl. Acad. Sci.USA 91:6574–78
121. Mo Y, Nagel C, Taylor LP. 1992. Biochemical complementation of chalcone synthase mu-tants defines a role for flavonols in functional pollen. Proc. Natl. Acad. Sci. USA 89:7213–17
122. Morris PF, Bone E, Tyler BM. 1998. Chemotropic and contact responses of Phytophthorasojae hyphae to soybean isoflavonoids and artificial substrates. Plant Physiol. 117:1171–78
123. Moulin L, Munive A, Dreyfus B, Boivin-Masson C. 2001. Nodulation of legumes bymembers of the beta-subclass of Proteobacteria. Nature 411:948–50
124. Murphy A, Peer WA, Taiz L. 2000. Regulation of auxin transport by aminopeptidasesand endogenous flavonoids. Planta 211:315–24
125. Nagahashi G, Douds DD Jr. 2003. Action spectrum for the induction of hyphal branchesof an arbuscular mycorrhizal fungus: exposure sites versus branching sites. Mycol. Res.107:1075–82
126. Nagahashi G, Douds DD. 1999. A rapid and sensitive bioassay with practical applica-tion for studies on interactions between root exudates and arbuscular mycorrhizal fungi.Biotechnol. Tech. 13:893–97
127. Nielsen TH, Christophersen C, Anthoni U, Sørensen J. 1999. Viscosinamide, a newcyclic depsipeptide with surfactant and antifungal properties produced by Pseudomonasfluorescens DR54. J. Appl. Microbiol. 87:80–86
128. Nielsen TH, Thrane C, Christophersen C, Anthoni U, Sorensen J. 2000. Structure, pro-duction characteristics and fungal antagonism of tensin - a new antifungal cyclic lipopep-tide from Pseudomonas fluorescens strain 96.578. J. Appl. Microbiol. 89:992–1001
129. Neumann G, Romheld V. 1999. Root excretion of carboxylic acids and protons inphosphorus-deficient plants. Plant Soil 211:121–30
130. Newman EI, Reddell P. 1987. The distribution of mycorrhizas among families of vascularplants. New Phytol. 106:745–51
131. Nigh EL Jr . 1990. Stress factors influencing Fusarium infection in asparagus. Acta Hort.271:315–22
132. Nilsson M-C, Hogberg P, Zackrisson O, Fengyou W. 1993. Allelopathic effects by Em-petrum hermaphroditum Hagerup on development and nitrogen uptake by roots and my-corrhiza of Pinus sylvestris L. Can. J. Bot. 71:620–28
133. Nimbal CI, Pedersen JF, Yerkes CN, Weston LA, Weller SC. 1996. Phytotoxicity anddistribution of sorgoleone in grain sorghum germplasm. J. Agric. Food Chem. 44:1343–47
134. Oger PM, Mansouri H, Nesme X, Dessaux Y. 2004. Engineering root exudation of Lotustoward the production of two novel carbon compounds leads to the selection of distinctmicrobial populations in the rhizosphere. Microb. Ecol. 47:96–103
135. O’Malley RC, Lynn DG. 2000. Expansin message regulation in parasitic angiosperms:marking time in development. Plant Cell 12:1455–65
262 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
136. Palma JM, Longa MA, del Rio LA, Arines J. 1993. Superoxide dismutase in vesiculararbuscular mycorrhizal red clover plants. Physiol Plant 87:77–83
137. Palmer AG, Gao R, Maresh J, Erbil WK, Lynn DG. 2004. Chemical biology of multi-host/pathogen interactions: chemical perception and metabolic complementation. Annu.Rev. Phytopath. 42:439–64
138. Patterson E, Sims A. 2000. Effect of nitrogen supply and defoliation on loss of organiccompounds from roots of Festuca rubra. J. Exp. Bot. 51:1449–57
139. Peer WA, Bandyopadhyay A, Blakeslee JJ, Makam SN, Chen RJ, et al. 2004. Variation inexpression and protein localization of the PIN family of auxin efflux facilitator proteins inflavonoid mutants with altered auxin transport in Arabidopsis thaliana. Plant Cell 16:1898–911
140. Pei ZM, Murata Y, Benning G, Thomine S, Klusener B, et al. 2000. Calcium channelsactivated by hydrogen peroxide mediate abscisic acid signalling in guard cells. Nature406:731–4
141. Penuelas J, Ribas-Carbo M, Giles L. 1996. Effects of allelochemicals on plant respirationand oxygen isotope fractionation by the alternative oxidase. J. Chem. Ecol. 22:801–5
143. Perry LG, Johnson C, Alford ER, Vivanco JM, Paschke MW. 2006. Screening of grasslandplants for restoration after spotted knapweed invasion. Rest. Ecol. In press
( ± )-catechin maybe produced by C.
maculosa as a meansof regulatingintraspecificcompetition bylimiting conspecificseedlingestablishment.
144. Perry LG, Thelen GC, Ridenour WM, Weir TL, Callaway RM, et al. 2005. Dualrole for an allelochemical: ( ± )-catechin from Centaurea maculosa root exudatesregulates conspecific seedling establishment. J. Ecol. 93:1125–36
145. Peters NK, Frost JW, Long SR. 1986. A plant flavone, luteolin, induces expression ofRhizobium meliloti nodulation genes. Science 233:977–80
146. Picman J, Picman AK. 1984. Autotoxicity in Parthenium hysterophorus and its possible rolein control of germination. Biochem. Syst. Ecol. 12:287–92
147. Politycka I. 1996. Peroxidase activity and lipid peroxidation in roots of cucmber seedlingsinfluenced by derivatives of cinnamic and benzoic acids. Acta Physiol. Plant 18:365–70
148. Prati D, Bossdorf O. 2004. Allelopathic inhibition of germination by Alliaria petiolata(Brassicaceae). Am. J. Bot. 91:285–88
150. Rahme LG, Stevens EJ, Wolfort SF, Shao J, Tompkins RG, Ausubel FM. 1995. Commonvirulence factors for bacterial pathogenicity in plants and animals. Science 286:1899–902
151. Rahme LG, Tan MW, Le L, Wong SM, Tompkins RG, et al. 1997. Use of modelplant hosts to identify Pseudomonas aeruginosa virulence factors. Proc. Natl. Acad. Sci. USA94:13245–50
152. Romero-Romero T, Sanchez-Nieto S, San Juan-Badillo A, Anaya AL, Cruz-Ortega R.2005. Comparative effects of allelochemical and water stress in roots of Lycopersicumesculentum Mill. (Solonaceae). Plant Sci. 168:1059–66
153. Romheld V, Marschner H. 1985. Evidence for a specific uptake system for iron phy-tosiderophores in roots of grasses. Plant Physiol. 80:175–80
154. Ronchel MC, Ramos JL. 2001. Dual system to reinforce biological containment of re-combinant bacteria designed for rhizoremediation. Appl. Environ. Micro. 67:2649–56
155. Rovira AD, Newman EI, Bowen HJ, Campbell R. 1974. Quantitative assessment of therhizosphere microflora by direct microscopy. Soil Biol. Biochem. 6:211–16
156. Ryu CM, Farag MA, Hu CH, Reddy MS, Kloepper JW, Pare PW. 2004. Bacterial volatilesinduce systemic resistance in Arabidopsis. Plant Physiol. 134:1017–26
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 263
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
157. Schardl CL, Leuchtmann A, Spiering MJ. 2004. Symbioses of grasses with seed-bornefungal endophyte. Annu. Rev. Plant Biol. 55:315–40
158. Sicker D, Schneider B, Hennig L, Knop M, Schulz M. 2001. Glycoside carbamates frombenzoxazolin-2(3H)-one detoxification in extracts and exudates of corn roots. Phytochem-istry 58:819–25
161. Smith CE, Ruttledge T, Zeng ZX, Omalley RC, Lynn DG. 1996. A mechanism forinducing plant development: the genesis of a specific inhibitor. Proc. Natl. Acad. Sci. USA93:6986–91
162. Somers E, Vanderleyden J, Srinivasan M. 2004. Rhizosphere bacterial signalling: a loveparade beneath our feet. Crit. Rev. Microbiol. 30:205–235
163. Steenhoudt O, Vanderleyden J. 2000. Azospirillum, a free-living nitrogen-fixing bac-terium closely associated with grasses: genetic, biochemical and ecological aspects. FEMSMicrobiol. Rev. 24:487–506
164. Stermitz FR, Bais HP, Foderaro TA, Vivanco JM. 2003. 7,8-Benzoflavone: a phytotoxinfrom root exudates of invasive Russian knapweed. Phytochemistry 64:493–97
165. Sugiura Y, Nomoto K. 1984. Phytosiderophores: structures and properties of mugineicacids and their metal complexes. Struct. Bond. 58:107–35
166. Tamasloukht M, Sejalon-Delmas N, Kluever A, Jauneau A, Roux C, Becard G, et al. 2003.Root factors induce mitochondrial-related gene expression and fungal respiration duringthe developmental switch from asymbiosis to pre-symbiosis in the arbuscular mycorrhizalfungus Gigaspora rosea. Plant Physiol. 131:1468–78
167. Tan J, Bednarek P, Liu J, Schneider B, Svatos A, Hahlbrock K. 2004. Universally occur-ring phenylpropanoid and species-specific indolic metabolites in infected and uninfectedArabidopsis thaliana roots and leaves. Phytochemistry 65:691–99
168. Teplitski M, Chen H, Rajamani S, Gao M, Merighi M, Sayre RT, et al. 2004. Chlamy-domonas reinhardtii secretes compounds that mimic bacterial signals and interfere withquorum sensing regulation in bacteria. Plant Physiol. 134:137–46
169. Teplitski M, Robinson JB, Bauer WD. 2000. Plants secrete substances that mimic bacte-rial N-acyl homoserine lactone signal activities and affect population density dependentbehaviors in associated bacteria. Mol. Plant-Microbe Interact. 13:637–648
170. Thrane C, Harder Nielsen T, Neiendam Nielsen M, Sorensen J, Olsson S. 2000.Viscosinamide-producing Pseudomonas fluorescens DR54 exerts a biocontrol effect onPythium ultimum in sugar beet rhizosphere. FEMS Microbiol. Ecol. 33:139–46
171. Uhde-Stone C, Temple SJ, Vance CP, Allan DL, Zinn KE, et al. 2003. Acclimation ofwhite lupin to phosphorus deficiency involves enhanced expression of genes related toorganic acid metabolism. Plant Soil 248:99–116
172. Uren NC. 2000. Types, amounts and possible functions of compounds released into therhizosphere by soil grown plants. In The Rhizosphere: Biochemistry and Organic Substancesat the Soil Interface, ed. R Pinton, Z Varanini, P Nannipieri. pp. 19–40. New York: MarcelDekker
173. Van Gundy SD, Kirkpatrick JD, Golden J. 1977. The nature and role of metabolic leakagefrom root-knot nematode galls and infection by Rhizoctonia solani. J. Nematol. 9:113–21
174. van West P, Morris BM, Reid B, Appiah AA, Osborne MC, et al. 2002. Oomycete plantpathogens use electric fields to target roots. Mol. Plant-Microbe Interact. 15:790–98
264 Bais et al.
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
175. Vicre M, Santaella C, Blanchet S, Gateau A, Driouich A. 2005. Root border-like cells ofArabidopsis. Microscopical characterization and role in the interaction with rhizobacteria.Plant Physiol. 138:998–1008
176. Vivanco JM, Bais HP, Stermitz FR, Thelen GC, Callaway RM. 2004. Biogeographi-cal variation in community response to root allelochemistry: novel weapons and exoticinvasion. Ecol. Lett. 7:285–92
177. von Bodman SB, Bauer WD, Coplin DL. 2003. Quorum-sensing in plant pathogenicbacteria. Annu. Rev. Phytopathol. 41:455–82
178. Von Jagow G, Link TA. 1986. Use of specific inhibitors on the mitochondrial bc1 complex.Meth. Enzymol. 126:253–71
179. von Rad U, Huttl R, Lottspeich F, Gierl A, Frey M. 2001. Two glucosyltransferases areinvolved in detoxification of benzoxazinoids in maize. Plant J. 28:633–42
180. Walker TS, Bais HP, Grotewold E, Vivanco JM. 2003. Root exudation and rhizospherebiology. Plant Physiol. 132:44–51
181. Wardle DA, Lavelle P. 1997. Linkages between soil biota, plant litter quality, and de-composition. In Driven by Nature, Plant Litter Quality and Decomposition, ed. G Cadisch,KE Giller, pp. 107–24. Wallingford: CAB Intl.
182. Wardle DA, Nicholson KS, Ahmed M, Rahman A. 1994. Interference effects of theinvasive plant Carduus nutans L. against the nitrogen fixation ability of Trifolium repens L.Plant Soil 163:287–97
183. Wardle DA, Nicholson KS, Rahman A. 1993. Influence of plant age on the allelopathicpotential of nodding thistle (Carduus nutans L.) against pasture grasses and legumes. WeedRes. 33:69–78
184. Wardle DA, Nilsson M-C, Gallet C, Zackrisson O. 1998. An ecosystem-level perspectiveof allelopathy. Biol. Rev. Camb Phil. Soc. 73:305–19
185. Weidenhamer JD, Romeo JT. 2004. Allelochemicals of Polygonella myriophylla: chemistryand soil degradation. J. Chem. Ecol. 30:1067–82
186. Weir TL, Bais HP, Vivanco JM. 2003. Intraspecific and interspecific interactions medi-ated by a phytotoxin, (-)-catechin, secreted by the roots of Centaurea maculosa (spottedknapweed). J. Chem. Ecol. 29:2397–412
187. Weir TL, Park SW, Vivanco JM. 2004. Biochemical and physiological mechanisms me-diated by allelochemicals. Curr. Opin. Plant Biol. 7:472–79
188. Williams RJ, Spencer JP, Rice-Evans C. 2004. Flavonoids: antioxidants or signallingmolecules? Free Radic. Biol. Med. 36:838–49
189. Williamson GB. 1990. Allelopathy, Koch’s postulates and neck riddles. In Perspectives inPlant Competition, ed. JB Grace, D Tilman, pp. 143–62. London: Academic
190. Wu HW, Haig T, Pratley J, Lemerle D, An M. 2000. Allelochemicals in wheat (Triticumaestivum L.): variation of phenolic acids in root tissues. J. Agric. Food Chem. 48:5321–25
191. Yamada Y, Nihira T. 1998. Microbial hormones and microbial chemical ecology. InComprehensive Natural Products Chemistry. Vol. 8, ed. DHR Barton, K Nakanishi, pp. 377–413. Oxford: Elsevier
192. Yang XH, Scheffler BE, Weston LA. 2004. SOR1, a gene associated with bioherbicideproduction in sorghum root hairs. J. Exp. Bot. 55:2251–59
193. Yeates GW. 1999. Effects of plants on nematode community structure. Annu. Rev. Phy-topath. 37:127–49
194. Yoder JI. 1999. Parasitic plant responses to host plant signals: a model for subterraneanplant-plant interactions. Curr. Opin. Plant Biol. 2:65–70
www.annualreviews.org • Root Exudates in Rhizosphere Interactions 265
Ann
u. R
ev. P
lant
Bio
l. 20
06.5
7:23
3-26
6. D
ownl
oade
d fr
om a
rjou
rnal
s.an
nual
revi
ews.
org
by O
RE
GO
N S
TA
TE
UN
IVE
RSI
TY
on
10/3
0/07
. For
per
sona
l use
onl
y.
ANRV274-PP57-10 ARI 5 April 2006 19:6
195. Yu JQ, Ye SF, Zhang MF, Hu WH. 2003. Effects of root exudates and aqueous rootextracts of cucumber (Cucumis sativus) and allelochemicals, on photosynthesis and antiox-idant enzymes in cucumber. Biochem. Sys. Ecol. 31:129–39
196. Zackrisson O, Nilsson M-C. 1992. Allelopathic effects by Empetrum hermaphroditum onseed germination of two boreal tree species. Can. J. Forest Res. 22:1310–19
Laser Microdissection of Plant Tissue: What You See Is What You GetTimothy Nelson, S. Lori Tausta, Neeru Gandotra, and Tie Liu � � � � � � � � � � � � � � � � � � � � � � � � � � 181
Integrative Plant Biology: Role of Phloem Long-DistanceMacromolecular TraffickingTony J. Lough and William J. Lucas � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 203
The Role of Root Exudates in Rhizosphere Interactions with Plantsand Other OrganismsHarsh P. Bais, Tiffany L. Weir, Laura G. Perry, Simon Gilroy,
Mosses as Model Systems for the Study of Metabolism andDevelopmentDavid Cove, Magdalena Bezanilla, Phillip Harries, and Ralph Quatrano � � � � � � � � � � � � � � 497
Structure and Function of Photosystems I and IINathan Nelson and Charles F. Yocum � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 521
Glycosyltransferases of Lipophilic Small MoleculesDianna Bowles, Eng-Kiat Lim, Brigitte Poppenberger, and Fabian E. Vaistij � � � � � � � � � � � 567
![The preceding root system drives the composition and ......and amino acid root exudates compared with wheat [33]. In the present study, we compare the function and taxonomic structure](https://static.documents.pub/doc/80x56/607b8b1609f25816c86cf034/the-preceding-root-system-drives-the-composition-and-and-amino-acid-root.jpg)
