DrinkwaterSnappRevised by HC111111033612


									Chapter 11: The rhizosphere in agricultural ecosystems.

Laurie E. Drinkwater and Sieglinde S. Snapp


       Agricultural systems represent the major form of land management, covering 5 billion ha

of the global terrestrial land area. The unintended consequences of agriculture extend well

beyond agricultural landscapes and include environmental degradation and social displacement

(Hambridge 1938; Vitousek et al.1997; Friedland et al. 1991). Many have advocated the adoption

of an ecosystem-based approach that would incorporate multifunctionality as an agricultural goal

and entail broad application of fundamental ecological principles to food production (Dale et al.

2000; Robertson and Swinton 2005; Drinkwater and Snapp, in review). This approach would

aim to reduce external inputs and environmental degradation by increasing the capacity for

internal, ecological processes to support crop production while contributing to other ecosystem

services (Dale et al 2000).

       Most efforts devoted to managing the rhizosphere in agricultural systems have

emphasized processes that contribute directly to maximizing yield within the context of resource-

intensive cropping system. Several excellent reviews are available covering the role of

rhizosphere biology in promoting crop growth under the nutrient rich conditions of high input

agriculture (cf. Pinton et al., 2001, Lynch, 1990). In particular, the biology of important root

pathogens and plant-microbial N-fixing symbioses have been extensively studied within this

context (ref-SSS). A smaller amount of rhizosphere research has focused on achieving modest

improvements in yields under severe nutrient or water limitations that are commonly found in
                                                                   Drinkwater and Snapp Page
low-input, subsistence agroecosystems of the developing countries where farmers do not have

access to purchased (Lynch 1990).

       In this chapter we will assess the current ecological understanding of the rhizosphere in

agroecosystems and broaden the scope of rhizosphere contributions to encompass a variety of

ecosystem functions beyond those directly related to maximizing crop growth and yields. Our

aim is to examine the potential for rhizosphere processes and plant-microbial interactions to

restore agroecosystem functions to reduce input dependancy and environmental degradation. We

begin with an inventory of how conventional, high input management has altered the soil

environment and biota in agroecosystems with particular emphasis on the consequences for the

rhizosphere. We then survey a range of rhizosphere processes and examine how current

management practices enhance or hinder the process and evaluate the potential for improved

functionality. Finally, we look ahead and discuss how management of the rhizosphere and plant-

microbial interactions could be approached within multifunctional, ecologically-sound

agricultural systems of the future.

Intensive agriculture: deliberate and inadvertent consequences for the rhizosphere

       The soil environment in agroecosystems reflects the legacy of the native ecosystem and

past management combined with current management practices. In early farming systems, human

modification was limited to altering plant species composition. As agriculture has evolved, the

degree of intervention has grown steadily, culminating with the current, resource intensive

“Green Revolution” production systems where management interventions are often the dominant

force shaping ecosystem structure and function. Intentional management of the rhizosphere has

focused mainly on biological control of root pathogens and enhancing obligate mutualisms and
                                                                     Drinkwater and Snapp Page
will be discussed later in this chapter. Here we briefly survey key modifications of the soil

environment and their unintentional consequences for the rhizosphere habitat. In practice,

farming systems fall along a continuum of intensity as depicted in Figure 1 and have varying

impacts on rhizosphere processes. Our discussion will emphasize the situation in conventional,

high input annual systems since these production systems supply a substantial portion of food on

a global basis, are continuing to expand in developing countries and also have the greatest impact

on the environment (Tilman 1999; Galloway 2000).

Tillage and soil structure

        Use of intensive tillage in agricultural production began with the development of the

plow (~ 3,000 BC in Mesopotamia and Egypt), which permitted large plots of soil to be

intensively mixed and planted to a monoculture (Tisdale et al., 1993). Tillage remains a

ubiquitous feature of nearly all annual cropping systems. A variety of tillage technologies have

been developed, however most primary tillage involves mixing the top 15-25 cm of soil in

preparation for planting. In addition to the periodic disruption of the soil environment, additional

consequences of tillage include radically altered pore volume and pore structure, reduced vertical

stratification, destruction of pore structure from past roots and hyphae, dispersal of microbial

communities and fungal hyphae networks and accelerated decomposition of soil organic matter

(SOM). Following tillage, the soil tends to settle so that porosity is reduced compared to the

original conditions of the native pre-tillage ecosystem. For example, the pore volume (% of soil

volume occupied by air or water) in native prairies of the Midwestern US is approximately 60-

70% compared to 45-50% in cultivated prairie soils (Baer et al. 2002, Motavalli and Miles 2002).

Thus, the reliance on tillage profoundly alters the soil environment in terms of atmosphere and
                                                                     Drinkwater and Snapp Page
water relations (drainage and water holding capacity) while also disrupting processes that are

influenced by contiguous networks formed by roots and fungal hyphae. No-tillage agricultural

systems have been developed for annual crops such as the Midwestern grain systems of the US,

however, very few are maintained in continuous no-till beyond several years (Evans et al., 2000).

As a result, only perennial agricultural systems such as pastures and some orchards maintain soil

environments that approximate the native state in terms of the degree of physical disturbance and

pore structure (Fig. 1).

Nutrient availability and soil chemistry

        Manipulation of soil chemistry began with the advent of liming to raise soil pH in early

Roman agriculture and has grown to be a major component of soil fertility management in

conventional agriculture so that soil pH is usually more neutral relative to native soils (Tisdall et

al. 1993). Other aspects of soil chemistry which are targeted through management relate to

optimizing the supply of nutrients to the crop. For the past 50 years, intensive agriculture has

focused on supplying soluble, plant available forms of major nutrients combined with

manipulation of soil pH and additions of micronutrients as indicated by soil tests (Drinkwater

and Snapp, in review). Compared to unmanaged terrestrial systems, the concentrations of

soluble, inorganic forms of major nutrients such as N and P are often several orders of magnitude

greater in agricultural soils (Booth et al. 2005).

Carbon flow and soil organic matter

        The use of tillage has major consequences for C distribution and turnover. Tillage

eliminates the O horizon and accelerates litter decomposition rates by mixing newly introduced

litter with soil (Buyanovsky et al. 1987). Furthermore, the advent of fertilizers and herbicides
                                                                    Drinkwater and Snapp Page
made it unnecessary to grow cover crops and forages in rotation with cash crops and permitted

the widespread adoption of the simplified crop sequences that are prevalent today (Auclair 1976,

Drinkwater and Snapp, in review). These rotations typically include bare fallow periods (when

land is maintained without any growing plants) in between cash crops (Tonitto et al., in press).

As a result, the time frame of actively growing plants in annual agriculture is commonly limited

to 4-8 months per year decreasing the rhizosphere habitat, C-fixation and inputs of labile C in

space and time (c.f. Drinkwater et al., 1995). Tillage, combined with soluble N additions and the

relatively labile composition of crop residues returned to the soil fosters rapid turnover of

particulate organic soil C pools (Buyanovsky et al., 1987; Cambradella) while decomposition of

the humified fraction may decrease (Neff et al., 2002) shifting the distribution of C pools so that

labile, particulate OM is proportionately reduced compared to humified OM (Wander et al. 2003-

LED). The reduction in total SOM (Campbell and Zentner 1993; Aref and Wander 1997)

combined with the disproportionate impact on labile C pools increases the severity of C-

limitation in bulk soil and exacerbates the tendency for nutrient saturation (Fenn et al. 1998;

Tonitto et al. 2004).

Consequences for the soil biota and the rhizosphere community

       The abiotic changes outlined above restructure the distribution and frequency of

microbial soil habitats in agroecosystems and lead to shifts in species abundance and richness.

Management interventions such as tillage, crop species composition and soil amendments (Frey

et al. 1999; Nakatsu et al. 2000; Bunemann et al. 2004 ) act in concert with the background soil

environment to alter the indigenous biota in bulk soil (Buckley et al., 2000; Steenwerth et al.

2002), which in turn influences rhizosphere community composition. The few studies comparing
                                                                Drinkwater and Snapp Page
rhizosphere community composition across agoecosystems suggest that soil type (particularly

texture) is the most important factor, followed by management history. The influence of plant-

related attributes (crop species and cultivar) appear to have less influence on rhizosphere

community composition compared to these environmental characteristics (Araujo da Silva et al.

2003, Salles et al. 2004; Bassio et al. 2005). The effects of cultivation on microbial community

structure in bulk soil appear to be long-lasting and can still be detected years after agricultural

management has ended (Buckley et al. 2000, 2003).

       It is clear that under intensified agriculture, the rhizosphere community is faced with a

unique soil environment that differs substantially from the one in which plant-microbial

interactions originally evolved. Ecosystem services that were once supplied by plants and

associated soil organisms are now largely provided through a variety of inputs. Plant functional

diversity has been replaced by increased inputs such as tillage, soluble fertilizers and pesticides

(Drinkwater and Snapp, in review). In essence, it is the management system that has created the

dependancy for many of the agricultural inputs that target the belowground system in agriculture,

similar to the pesticide treadmill that was first proposed in the 1960's to describe the increased

dependancy on insecticides created by chemical control of aboveground herbivorous arthorpods

(Smith and van der Bosch 1967). The use of tillage necessitates the need for continued tillage

due to diminished SOM and degraded soil structure (Topp et al. 1995). The high concentrations

of plant-available nutrients in space and time may reduce the role of mutualist rhizosphere

organisms since energetics favor plant acquisition of these soluble nutrients which are supplied

in quantities surpassing crop needs. Finally, alterations in the soil environment combined with

simplified rotations often increases the frequency and severity of pathogen infections leading to
                                                                Drinkwater and Snapp Page
dependance on broad spectrum fungicides such as methyl bromide (Larson and Shaw 1995;

Gilreath et al. 2004).

Rhizosphere processes and agrecosystem function

       It is against this backdrop of a highly modified soil environment and the cascading effects

on soil biota that we examine rhizosphere function in agriculture and consider how to redirect

management to restore rhizosphere processes and agroecosystem functions. Rhizosphere

microorganisms and their associated primary producers contribute both directly and indirectly to

a wide range of ecosystem functions (Figure 2). Processes such as aggregation, nutrient cycling,

hydrology and C storage are jointly mediated by plants and soil organisms through interactions in

the rhizosphere although the significance of rhizosphere contributions has generally been

diminished in agroecosystems by inputs and other interventions (Figure 1).

Rhizosphere-mediated aggregation

       Soil aggregation determines the pore structure and dispersion resistance of soil and is

central to soil and ecosystem functioning. The proportion of soil particles sequestered in

aggregates contributes to the movement and storage of water, soil aeration, and species

composition and distribution of soil organisms. These factors interact with one another and

influence a numerous ecosystem processes including: 1) water holding capacity, infiltration and

erosion, 2) temporal and spatial distribution of anaerobic conditions 3) growth of plant roots and

fungal hyphae, and 4) the cycling and storage of nutrients and carbon (Angers 2002). The process

of aggregate formation is particularly important in agroecosystems where tillage periodically

disrupts soil structure and accelerates the breakdown of macroaggregates and physically-

protected labile SOM. While parent material and clay surface chemistry regulate micro-aggregate
                                                                    Drinkwater and Snapp Page
formation Kay (1990), plants and soil organisms are the major drivers of macro-aggregate

formation (Tisdall and Oades 1982; Oades 1984,). As a result, the feedbacks between plants, soil

organisms and soil structure are a key regulator of productivity and biogeochemical functioning

of agroecosystems. In soils with a high proportion of fine silt and clay particles, aggregation is a

prerequisite for life because most plants require some level of aggregate structure to create pores

for aeration and water flow in order to grow in these soils.

       The potential for agricultural plants and their associated rhizosphere organisms to foster

aggregate formation has been studied since the early 1980's (Reid and Goss 1981, Tisdall and

Oades, 1982). Most of this research has focused on the effect of different plant species on water

stable aggregation in bulk soil. In general, perennial forage crops and annual legumes and

grasses tend to promote aggregation while cash crops either reduce or have no impact on water-

stable aggregation in bulk soils (Tisdall and Oades 1979, Reid and Goss 1981). These are the

agricultural plants that have been removed from most rotations in favor of simplified rotations

enabled by modern agriculture. Maize, tomato and wheat actually decreased aggregate stability

while growth of perennial ryegrass and alfalfa tended to increase it (Reid and Goss 1981). This

increased aggregate stability in bulk soil accrues over time and has been related to plant

parameters such as total root biomass or root length (Rillig et al. 2002), microbial

polysaccarhides produced in the rhizosphere (Reid and Goss, 1981), and more recently, fungal

populations associated with the rhizosphere (Haynes and Beare 1997, Rillig et al. 2002) or fungal

products such as glomalin (Wright and Upadhyaya, 1996, 1998).

       The role of plant-microbe symbiosis in aggregate formation is most extensively

documented for arbuscular mycorrhizal fungi (AMF) which are emerging to be an important
                                                                      Drinkwater and Snapp Page
biotic regulator of water stable aggregation in bulk soil. The discovery of glomalin, a collection

of iron-containing glycoproteins produced by AMF (Wright et al., 1996) has provided an

unprecedented opportunity to study the ecology of biotic aggregate formation. To date, glomalin

has been found in virtually all soils tested for the glycoprotein although the quantity can vary

from up to 100 mg g soil-1 in tropical forest soils to 3-4 mg g soil-1 in temperate agricultural soils

(Wright and Anderson, 2000, Rillig et al., 2001). Glomalins are moderately stable component of

the SOM, with a mean turnover time reported to range from 6-40 years. The structure of glomalin

has not been characterized so the true function of the glycoprotein remains unresolved and is an

active area of research.

       More recently, the role of rhizosphere bacteria in promoting aggregation of soil within

close proximity to roots, i.e. rhizosphere or root-adhering soil has been investigated (Gouzou et

al. 1993, Bezzate et al. 2000). Several rhizosphere organisms that foster aggregate formation

through production of exopolysaccharides (EPS) have been identified and linked to improved

soil structure of root-associated soil (Gouzou et al. 1993, Amellal et al., 1999, Bezzate et al.

2000, Alami et al. 2000). The EPS-producing bacteria Paenibacillus polymyxa (strain CF43), an

N-fixing bacteria endemic to the wheat (Triticum aestivum L.) rhizosphere fosters significant

increases in the aggregation and water-holding capacity of soil adjacent to roots (Bezzate et al,

2000) through production of a fructosyl polymer (Gouzou et al. 1993, Bezzate et al. 2000). In a

study of sunflower (Helianthus annuus L.) and an EPS-producing Rhizobium sp (Strain YAS34)

isolated from the sunflower rhizosphere, inoculation with this organism resulted in increased

abundance in the rhizosphere, modified soil structure and water holding capacity around the root

system, and a corresponding improvement in the drought resistance of the plant (Alami et al.
                                                                      Drinkwater and Snapp Page
2000). This type of localized modification of soil structure through the production of EPS

appears to be important for non-irrigated agricultural systems at the scale of individual plants. It

is likely to be most significant for short term modifications of water movement and storage since

these polysaccharides are readily decomposed and probably have a much shorter MRT than

glomalin-type substances.

       Questions about the function of these compounds, their evolutionary significance and the

mechanisms that control their production remain unanswered. Furthermore, the extent to which

cultivar selection, tillage and irrigation may have inadvertently influenced crop-microbial

interactions that promote or enhance aggregate formation is not known. The promotion of

aggregation through production of complex, extracellular polymers has often been viewed as a

secondary consequence of release of these substances in the environment. Clearly, bacterial

production of EPS is a widespread phenomena occurring across microbial habitats associated with

the formation of biofilms (Morris and Monie, 2003). However, given the prevalence of organisms

that are able to release copious amounts of these compounds into the soil and the benefits that

accrue through improvements in soil structure it is possible that under certain conditions soil

structure modification is the major function of these compounds. Some evidence supports this

possibility. First, although a systematic study of the abundance of EPS-producing species across

ecosystems that vary in terms of clay content has not been conducted, some evidence suggests that

at least within bacterial species, strains that are present in high-clay soils tend to produce

significant amounts of EPS (Achouak et al., 1999) and promote aggregate formation (Bezzate et

al. 2000). One in vivo experiment demonstrated that glomalin production is extremely plastic and

may be responsive to environmental conditions such as pore structure (Rillig and Steinberg 2002).
                                                                     Drinkwater and Snapp Page
This study suggested that a primary function of glomalin may be related to fostering aggregate

formation and soil structure improvement (Rillig and Steinberg, 2002).

Decomposition and net mineralization of nutrients

       Plant-mediated decomposition and corresponding mineralization of nutrients via the

rhizosphere (“priming effect”, see Cheng, Chapter xx) is not considered to be important in

conventional agriculture and hence, deliberate management of this process has not been

attempted, although it is considered to be of central importance in organically-managed systems

(Drinkwater 2004). In wealthier, industrialized countries major nutrients are generally supplied in

surplus quantities as soluble, inorganic fertilizers resulting in cropping systems which are

maintained in a state of nutrient saturation, particularly when the cash crop is present. Despite

application of luxurious amounts of N and use of refined best management practices (BMP’s),

crops still acquire 40-80% of their N from endogenous soil reserves and an average of 50% of the

N applied is lost from agricultural landscapes (Galloway 2000). In contrast, agroecosystems of

poorer developing countries do not have access to manufactured fertilizers and are often

producing crops in soils that have depleted nutrient pools from long histories of farming without

adequate nutrient return to fields.

       Microbial production of extracellular enzymes that can attack polymers and release small,

soluble molecules is an important mechanism contributing to the internal cycling of N, P and S

(McGill and Cole 1981). A priming effect of increased mineralization in response to additions of

labile C intended to simulate root exudates has been demonstrated for N (xx), P (xx) and S (xx).

While some plants are able to produce and secret enzymes required for P mineralization (Vance et
                                                                        Drinkwater and Snapp Page
al. 2000) and possibly S mineralization (ref-Sieg), release of nutrients from organic compounds is

largely carried out by heterotrophic microorganisms (McGill and Cole 1981).

       We expect that during millions of years of coevolution plant-microbial feedbacks have

evolved to regulate this co-dependancy (Kiers et al.-LED). In split root studies, rhizodeposition is

increased by roots exposed to greater concentrations of inorganic N compared to roots from the

same plant that are under low inorganic N conditions (Paterson et al., 2003). Until the advent of

Haber-Bosch N, the presence of inorganic N was indicative of net mineralization (with the

exception of soils where NH 4+ is present in clays). Thus, plants could increase their access to N

through root proliferation and exudation of labile C to support decomposition when inorganic N

patches were encountered. Some have speculated that plant-mediated decomposition would be

most beneficial to plants in soils where NPP is limited by N shortage (Paterson 2003). However,

in soils where N is truly limiting, N-fixing capabilities are advantageous (Vitousek et al., 2002)

whereas plant-mediated decomposition is more important in ecosystems where organic N

reservoirs are present and net N mineralization from these sources can be enhanced by increased

energy supply.

       Clearly, greater reliance on plant-mediated mineralization for nutrient acquisition in

agroecosystems would reduce the potential for nutrient losses due to the tight coupling between

the release of soluble, potentially mobile nutrient forms and plant uptake in the rhizosphere. This

could be particularly advantageous in the case of nitrogen. Inorganic nutrient pools can be

extremely small in ecosystems while high rates of net primary productivity (NPP) are maintained

if N-mineralization and plant assimilation are spatially and temporally connected in this manner

(cf. Jackson et al., 1988). We believe it is safe to say that the feedback mechanisms regulating this
                                                                   Drinkwater and Snapp Page
exchange are not fully understood and many questions remain to be answered it we are to

effectively manage this process in agroecosystems. In particular, it will be important to confirm

the identity of the SOM pools accessed by plant-mediated decomposition and whether or not

agroecosystems can be managed to increase these reservoirs without fostering increases in net N

mineralization in the absence of plants. Decomposition of chemically recalcitrant substrates is

accelerated in the rhizosphere (Siciliano et al. 2003). Recently, Ferris et al. (2004) have proposed

that food-web structure could also be influenced by management to optimize this process in

cropping systems where organic N sources predominate. Finally, it is unclear how plant breeding,

which has occurred primarily under nutrient saturated conditions, has affected the ability of crops

to access these recalcitrant pools of organic N via this mechanism. While crop roots certainly

stimulate SOM decomposition through this mechanism, selection in soil environments where

inorganic N and P are supplied in surplus quantities would tend to favor crops that did not

squander fixed C to obtain N or P.

Biologially-mediated weathering

       Weathering-induced release of cations and phosphorus from primary minerals is generally

considered of marginal interest in agricultural nutrient management, as the time-frame often

involves decades for significant changes. The significance of weathering may need to be re-

thought in light of the evidence that chemical dissolution processes are markedly promoted (10 to

20 fold) by plants and associated microorganisms through release of organic and carbonic acids,

mineral tunneling and related mineral-dissolving exudation (Hoffland et al., 2003; Schwartzman

and Volk, 1989). Few agricultural studies to date have encompassed the long-term view that is

addressed in the chapter by Richter et al., which documents the profound effect of rhizosphere
                                                                     Drinkwater and Snapp Page
processes on mineral weathering and soil formation. In this chapter we focus on shorter time

frames. Studies with perennial species have shown that measurable plant-induced weathering of

calcium and magnesium(Borman et al., 1998), and phosphate (Hinsinger and Gilkes, 1997), can

occur over a couple years. These processes can no longer be ignored in agricultural nutrient

management, particularly for lower-input farming systems.

       The interaction of plants, associated rhizo-organisms and soil minerals determine site-

specific nutritional benefits. Feldspar tunneling by ectomyrcorrhizal hyphae has been shown to be

associated with oxalate and citrate exudation and enhanced plant uptake of calcium and potassium

(Hoffland et al. 2003). Similarly, organic acid secretion at moderate levels is associated with

complexation of Fe3+ adsorbed to the surface of soil minerals, and improved iron nutritional

status (Jones et al., 1996). Rhizosphere-mediated acidification processes are multi-fold and play a

central role in modification of the edaphic environment, both short and long-term. The soil

formation consequences of organic acid exudation and other biotically induced acidification

processes are discussed indepth in Chapter X; Richter et al.

       Phosphorus is generally the most growth-limiting factor in agriculture that is legume-based

or nitrogen enriched. Biogeochemical processes control the species of phosphorus present and

bio-availability within cropping systems. A wide range of plant-mediated strategies have evolved

to enhance phosphorus supply, primarily through enhancing surface areas and access to sparingly

soluble phosphorus pools. Root morphological traits and mycorrhizal symbiosis are well known

means to enhance surface area and optimize rhizosphere activity of phosphatase enzymes and

organic acids (Illmer et al 1995; Oberson et al 2001). Organic acids are excreted in significant

quantities by many dicotyledonous plant species. This phenomenon is induced by low-P supply,
                                                                   Drinkwater and Snapp Page
and in some cases, Fe deficiency (ref-SSS). Organic acids enhance nutrient availability through

direct effects, as organic anions complex with Ca, Al3+ or block the sorption of P to other

charged sites and indirect effects through alteration of rhizosphere pH and stimulation of

microbial activity that enhances nutrient availability (Fox et al., 1999; Erich et al. 2002). Ligand

exchange can also occur in which the P bound to Fe or Al oxyhydroxides is replaced by the

organic anions (Lunstrom, 1995).

       There can be high costs associated with biologically-mediated enhancement of nutrient

availability. An extreme case is Lupinus albus, where organic acid excretion under P-deficient

conditions accounts for over one-third of carbon assimilated (Johnson et al., 1994). Root system

architecture and support for mycorrhizal fungi to enhance surface area and organic acid secretion

activity require plant investment below ground that has been estimated at 20 to 40% higher under

low-phosphorus conditions compared to high-phosphorus (Nielsen et al., 1998). The benefits to

plant adaptations and enhanced rhizosphere activity is considerable; access to sparingly soluble

nutrient pools allows plant establishment and growth under very low nutrient conditions. This was

shown in the case of unfertilized wheat: the soil fungus Penicillium radicum isolated from a low-

phosphorus rhizosphere altered pH locally and enhanced phosphorus solubility from complexes

with calcium, colloidal aluminum and iron (Whitelaw et al., 1999).

       In addition to enhancing plant nutritional status in the short-term, available nutrient pools

over time and nutrient influx in agroecosystems are influenced by assimilation of inorganic

phosphorus by rhizosphere mediated processes that protect phosphorus from physio-chemical

adsorption reactions with soil particles. Chief among these processes are microbial turnover and

organic matter mineralization processes which are synchronized with plant and microbial uptake.

Indirect evidence for this is the enhanced levels of microbial P and cycling of P from inorganic to
                                                                   Drinkwater and Snapp Page
organic and plant forms associated with managed systems that had enhanced soil biological

activity and legume presence (Oberson et al. 2001). Labeled glucose and residue studies have

recently shown that biomass P turnover is rapid, approximately twice as fast as C (Kouno et al.,

2002). This indicates that the potential for microbial P pools to support plant P requirements may

have been markedly underestimated, and could be exploited through deliberate selection as

discussed later in this chapter.

Rhizobia and mycorhizal associations

       A case can be made that the evolutionary process of plant-microsymbiont relationships

have been mediated by agricultural practices, many of which favor parasitism over mutualism

(e.g., Kiers et al., 2002). A key example is the suppression of the N2-fixation process by nitrate.

There is genetic variation in both plant host and Rhizobium bacteria for tolerance of the N2-

fixation process to the presence of nitrate, but in the vast majority of cases the presence of nitrate

is highly suppressive to symbiotic N2-fixation (e.g., in soybean, Neo et al., 1996 and in chickpea,

Nour et al., 1994). Further, there is evidence that frequent fertilization with soluble inorganic N

and P has increased the presence of parasitic rhizobia and mycorrhizae relative to beneficial

symbionts (Johnson et al., 2003). Numerous studies have shown that nodulation and nodule

activity are both suppressed by N fertilization in grain and forage legumes (Kiers et al., 2002;

Vargas et al., 2000). Indirect consequences of the suppression of N2-fixation by soil nitrogen may

include suppression of mycorrhizal function since flavonoids that induce nodulation also stimulate

hyphal growth of the AM fungi (Rengel, 2002).

       Application of inorganic phosphorus has been widely shown to directly suppress

mycorrhizal infection of roots (Jasper et al., 1979; Smith et al., 1994), and to suppress function of

the plant-mycorrhizal symbiosis in maize and soybean (McGonigle et al., 1999). Disturbance from
                                                                       Drinkwater and Snapp Page
tillage is another major factor that reduces the presence of mycorrhizal symbiosis (Galez et al.,

2001), although a recent review presents evidence that nutrient input level rather than intensity of

disturbance is a regulator of mycorrhizal-mediated nutrient acquisition (Harrier and Watson,

2003). In grassland studies application of nitrogen fertilizer has been shown to alter mycorrhizal

growth, community structure and function (Johnson et al., 2003). Interestingly, arbuscular

mycorrhiza from unfertilized grass sites enhance plant growth whereas mycorrhiza inoculum from

fertilized site had no growth-enhancing effect, possibly because fertilization alters the benefits and

costs of mycorrhizal symbioses and tends to foster a parasitic rather than mutualist function

(Corkidi et al. 2002). Reliance on soluble nutrients markedly alters community dynamics in the

rhizosphere and may have inadvertently selected for ineffective mycorrhizal and legume-

rhizobium symbioses in modern agricultural systems.

       Inundation through inoculation or bacterization of seedlings are direct means to enhance

beneficial organisms in the rhizosphere. This has been a successful approach for symbiotic

associations with a high degree of specificity, most notably the legume-Rhizobia symbiosis. For

less specific interations, indigenous rhizosphere populations generally resist colonization by

inoculated organisms, and are rarely altered by inoculation in the absence of plant-organism

specificity. However, bacterization of tissues with limited colonization and the potential to

concentrate benefical organisms, such as seeds is another approach that has met with success. In

the case of Pseudomonas fluorescens applied to tomato seeds reduced development of the

pathogen Pythium ultimum was observed, and suppressive capacity was linked to sidephore

production in a strain comparison of mutant and wildtype P. fluorescens (Hultberg et al., 2000).

     The interaction of plant and species inoculated, and the outcome in terms of colonization

and development of a symbiotic organ such as nodules are highly dependent on space and time.
                                                                       Drinkwater and Snapp Page
This is illustrated by the demonstration that the origin of nodule inhabitants is markedly

influenced by rhizosystem architecture in inoculated soybeans (Graham et al 2004). Nodules

located near the central root system are developed through plant symbiotic interactions with

inoculated Rhizobium species, while external nodules far from the central axis are likely to be

inhabited by indigenous, and often ineffective, Rhizobium.

     The availability of soluble nutrients in conventional agriculture has implications for the

nature of root symbiont mutalism, both over the short-term and the long-term. The energetics of

substantial doses of fertilizer will favor rhizobial and mycorrhizal associations that tend towards

parasitic rather than mutalistic symbioses with crop species (Kiers et al 2002). Mycorrhizal

species composition in high nutrient input corn systems has been shown to favor ineffective

strains (Douds Johnson? SSS- refs). Interestingly, Kiers and colleagues (2002) have suggested

that energetics have favored evolutionary selection of less efficient strains of Rhizobia in mixed

strain nodules and mycorrhizal species present in conventional agriculture. Research is needed in

this fundamentally important area for agricultural management. The paradox is how to maintain

mutalistic root symbionts within a high fertilizer environment that fosters parasitic relationships.

Community ecology of the rhizosphere and biological control

       The selective pressures exerted by intensive management practices combined with crop

breeding appear to have altered the structure and function of rhizosphere communities in

agricultural systems in unexpected ways. The edaphic environment is modified as shown in Figure

1, with high inorganic nutrient availability and low diversity carbon inputs associated with

conventional agricultural systems, which profoundly influences substrate, habitat availability and

microbial community dynamics (Hoitink and Boehm, 1999). The complex interactions mediated
                                                                      Drinkwater and Snapp Page
by nutrient, carbon and microbial interactions in the rhizosphere were elucidated in an on-farm

California study where multivariate analysis demonstrated that corky root severity was positively

associated with inorganic nitrogen status in conventional versus organic fields (Workneh et al.,

1994). The research over two seasons found a consistently negative association between disease

and microbial activity in soil, which was temporally and spatially related to the incorporation of

green manures.

       In general, conventional agricultural practices stimulate facultative saprophytic pathogens

and increase crop susceptibility to disease. This is evident in intensively managed, high value

vegetable crops where reliance on fumigation, multiple tillage operations and high rates of

fertilizer is often associated with compacted soils, low levels of soil microbial activity and

recurring root health problems (Abawi and Widmer, 2000). Root senescence, proliferation of

soil-borne plant parasitic organisms and invasion of root tissue are all enhanced to varying degrees

by salinity, poorly aggregated soils combined with changes in concentrations of soluble nutrients

(Snapp and Shennan, 1992; Workneh and van Bruggen, 1994).

       Management that prioritizes living cover and diversity of carbon inputs is associated with

enhanced activity and presence of soil microorganisms. If rhizophere organisms are well-

established, this will tends to suppress soil-borne disease organisms through mechanisms such as

competition for resources and habitat (OSullivan and OGara, 1992), antagonistic compounds

(Robleto et al., 1998), degradation of pathogenicity factors or pathogen cell walls, promotion of

vigorous, healthy roots (Snapp et al., 1991; 2003) and induction of resistance in the target plant

against the pathogen (van Wees et al., 1999). Soilborne phytopathogens encounter antagonism

from rhizosphere microorganisms before, during and after primary infection and secondary spread
                                                                      Drinkwater and Snapp Page
within the root. Readers are refereed to recent reviews which focus on the mechanisms of

suppressive soils (Gamliel et al. 2000; Sturz and Christie 2003).

       A well-studied example of rhizosphere occupants and consequences for soil-borne disease

is Take-All (Gaeumannomyces graminis var. tritici) in wheat. This is one of the most important,

devastating fungal diseases in cereal production around the world (Cook et al.1998). After initially

severe outbreaks, the disease is generally suppressed through a phenomena known as Take-all

decline. Altered population dynamics of rhizosphere bacteria have been shown to be consistently

associated with suppression of take-all, and most recently a role has been shown for the presence

of pseudomonads that secrete compounds that directly inhibit G. Graminis var. tritici (Gardener et

al., 2001), as well as competition within the rhizosphere. Inoculation with the Pseudomonas

species has not been consistently successful, rather the consistent environment of a no-till,

temporal wheat monoculture appears to provide an environment within which the Pseudomonas

species thrive that suppress G. Graminis var. tritici through multiple avenues. This is an example

of the complex, community level interactions in the rhizosphere that require more study to

elucidate underlying processes. Use of these principles is essential to the design of management

practices that foster crop root health through established, thriving communities of microorganisms

in the rhizosphere.

Plant species and cultivar effects on rhizosphere processes

       While it is well-known that rhizosphere community composition varies across plant

species and cultivars (Hawkins et al, Chapter xx), there is limited information about the functional

significance of these differences in plant-associated microbial communities. Plant selection in the

last half-century has occurred almost entirely under management regimes that include fumigated

soils with luxurious additions of nutrients and sufficient water (Boyer et al. 1982). This strategy
                                                                    Drinkwater and Snapp Page
of reducing environmental variation by providing ample resources reduces gene by environment

interaction, and enhances the power of selection for specific traits (ref-LED). In many cases, the

underlying genotypic alterations that have contributed to improving yield potential have not

generally been identified (Boyer et al., 1982). Plant breeding approaches have altered root

architecture (Jackson 1995; Jackson and Koch 1997) the ability or root branching to respond to

environmental conditions, and root exudates. For example, in wheat, allelopathic abilities

resulting from root exudates are reduced in modern hybrids compared to older varieties

(Bertholdsson 2003). Growing evidence indicates that this strategy has altered plant-microbial

interactions in the rhizosphere and may have selected against traits that allow plants to maintain

beneficial rhizosphere associations that influence some of the processes we have discussed.

       Studies comparing rhizosphere microbial community composition across cultivars have

reported varying results in the degree of overlap in terms of the species present in the various

habitats of the rhizosphere. Many of these studies do not include information about the

characteristics of the different plant cultivars included in the study making it difficult to interpret

the role of cultivar in shaping rhizosphere community composition. Since the development of

genetically-modified crops, there have been numerous studies investigating the effect of these new

genotypes on plant-associated soil microbes. Studies comparing the original non-GMO hybrid to

the GMO version in terms of root associated microbes often find no significant differences in

rhizosphere community composition (ref-LED). This is not surprising since these cultivars differ

only in by a single gene and frequently the gene in question is not expressed in root function. As

more sophisticated molecular techniques become widely available, a picture is emerging which

suggests that more closely related cultivars will have greater similarities in rhizosphere

community compositino compared to more distantly related cultivars (ref-LED). Comparisons of
                                                                        Drinkwater and Snapp Page
traditional or pre-industrial cultivars to modern, highly-selected varieties often find differences in

rhizosphere community composition but rarely link these differences to microbial community

function or rhizosphere processes.

       One of the few cases where physiological differences in cultivars has been linked to

changes in rhizosphere community composition and function is the case of ammonia-oxidizing

bacteria (AOB) populations in the rhizosphere of traditional versus modern rice cultivars

(Briones et al., 2002, 2003). Efficient management of N in rice paddies is particularly challenging

because rice paddies are typically maintained under flooded conditions and are essentially anoxic

below the soil-water interface (Buresh 1991). Briones et al. (2002) used fluorescence in situ

hybridization (FISH) with rRNA-targeted probes specific for the AOB to characterize AOB

populations on the rice root surface. The abundance of active AOB in the rhizoplane of three rice

cultivars was 2-3 orders of magnitude greater than is generally found in rice paddy soils. They

also reported differences in the total abundance and species composition of the rhizoplane

microbial communities across cultivars (Table 1). The greater abundance of the faster-growing

Nitrosomonas spp. can be partially attributed to the greater secretion of O2 by cv IR63087-1-17

compared to cv. Mahsuri (Briones 2002). However, differences in NH4+ would also be needed in

order to support Nitrosomonas. The greater abundance of heterotrophic bacteria in the rhizoplane

of cv. Mashuri suggests that other factors such as differences in root exudates and N dynamics

(i.e. NH4+ assimilation by the plants and heterotrophs in the rhizosphere) may also contribute to

the observed distributions of AOB species and nitrification rates (Briones 2003). This case

illustrates the complexities involved in intentionally managing biogeochemical processes through

cultivar-microbe interactions.
                                                                      Drinkwater and Snapp Page
       Finally, another area of recent study addresses the question of whether or not there are

non-obligate, root-associated microorganisms that are endemic to the rhizosphere of particular

plant species. Resolving this question could contribute to the development of breeding strategies

that target crops and their associated microorganisms. Wheat offers a particularly interesting

system for this question since it is one of the oldest agricultural plants and is currently cultivated

across a wide variety of climates and soil types. Initial studies comparing the species present in

wheat rhizospheres suggest that there are root-associated microorganisms that could be considered

endemic to the wheat rhizosphere and that 2000 years of wheat cultivation has altered the genetic

structure of these species (Guemouri-Athmani et al., 2000). Furthermore, the longevity of wheat

cultivation in Algerian soils was correlated with decreased phenotypic and genetic diversity and

higher frequency of N-fixing strains of Paenibacillu polymyxa abundance in rhizosphere-

associated soil.

The rhizosphere in ecological agriculture

       We recognize that ultimately the transition to ecologically-sound, sustainable food

production systems that meet human needs will be complex and will require fundamental changes

in cultural values and human behavior (Boyden, 2004) as well as the application of ecological

knowledge to agricultural management. The management of biocomplexity to promote ecological

processes and restoration of agroecosystem function will require the development and application

of ecosystem-based management strategies ( Dale et al., 2000). Ecosystem management is a land-

management approach that 1) takes into account the full suite of organisms and ecosystem

processes, 2) applies the concept that ecosystem function depends on ecosystem structure and

diversity, 3) recognizes that ecosystems are spatially and temporally dynamic and 4) includes
                                                                        Drinkwater and Snapp Page
sustainability as a primary goal (Dale et al., 2000). Application of this approach will require that

we redirect management practices to create a soil environment that is more conducive to

supporting potential contributions from rhizosphere processes. Opportunities to redirect

management include repopulating annual systems with plants in space and time, integration of

rotation, tillage and soil amendment practices and changes in the way we approach cultivar


Using increased plant species biodiversity in space and time

       The benefits of management practices that expand plant presence in space and time have

long been recognized (Hambridge, 1938). Use of accessory crops to fill niches in time and space

can greatly enhance rhizosphere activity, but have been dismissed as not economically viable by

many observers. Cover crop research often involves reduced harvestable productivity because

cash crops are not grown as frequently, and many rotation sequence and cover crop integration

studies have been short-term where the feedbacks from coupling C and N have not been fully

considered (Torbet et al., 1996; Young-Stivers 1999). This has led to the perception that these

practices have a high opportunity cost that is not offset by the benefits, which have only been

narrowly considered (Ghaffarzadeh, 1997; Labarta et al., 2002).

       A return to traditional rotations with extended cover from forages and winter annual

cereals may not be economically feasible given agricultural policies that favor simplified corn

and soybean cropping patterns. Targeted diversification strategies could, however, be employed

to capitalize on particular plant species and associated microbes that contribute to ecosystem

functions. The potential for a single plant species to influence rhizosphere dynamics is large in

managed ecosystems where spatial and temporal diversity is limited, particularly if plant species
                                                                      Drinkwater and Snapp Page
can be classified in terms of their various roles in belowground processes (Tilman, 1999).

Studies investigating the impacts of plant functional groups on longer-term processes are needed.

Identifying a wider variety of species as well as modifying current cover crop species to fill

particular niches could greatly increase the potential for cover crop adoption (Snapp et al., 2005).

As we discuss in this paper, breeding of cash crops and auxiliary plants could take into account

rhizosphere effects and consequences for agroecosystem function as primary selection traits.

       Intercrops where plant root systems occupy a larger volume of the soil could improve the

efficiency of row crop systems (Willey, 1990). Cereals and legumes are complementary species

in terms of below-ground niches, with the deeper taproot system of legumes exploring soil

resources unaccessable to shallow cereal-root systems. The presence of active root systems over

space, and for a greater part of the year improves synchronization of nutrient mineralization and

uptake leading to significantly enhanced efficiency (McKracken et al. 1994). The extended period

of root activity in rotations including covercrops may be the most significant factor contributing

to increased net C sequestration due to the role of roots in fostering aggregate formation (Wander

et al. 1994). Continuous C inputs from roots also support an active soil microbial community for

a greater part of the year and may increase net C retention. In microcosm studies, the proportion

of C retained from small incremental C inputs was greater than for a single large pulse of C (Jans

Hammermeister et al. 1998).

Integrated plant and amendment strategies

       Strategies need to be developed that combine P application with assimilation in biological

sinks, through management and integration of species that augment levels of soil organic acids

and phosphates. Application of sparingly-soluble sources of P to crops (e.g., most legumes) that
                                                                         Drinkwater and Snapp Page
can assimilate P into biological pools is an efficient strategy that has been underappreciated, and

could be used to bypass desorption, precipitation and occlusion of P (Vance et al., 2003).

Legumes are important vehicles to enhance P availability through diverse mechanisms, including

modified roots, secretion of organic acids and enhanced P-solubilizing activity through

microorganisms (Oberson et al., 1999). Similarly, targeted use of animal manures can facilitate

plant and microbial uptake of P into available forms (Erich et al 2002; Laboski and Lamb, 2003).

Where manure is utilized at sustainable, moderate levels and livestock are distributed extensively

across the landscape, organic-P sources appear to be inherently less vulnerable than inorganic

fertilizer sources to loss from occlusion, erosion or leaching (Powell and Saleem, 1994).

Manipulation of mycorrhizal populations to develop more efficient plant-symbiont combinations

is in its infancy, but strategies that can be pursued include use of sparingly-soluble rock P, reduced

tillage and integration of auxiliary plants that are highly mycorrhizal.

Selection for plant-microbe consortia

        We see many opportunities for plant breeding to enhance plant-microbial interactions in

ways that contribute to restored ecosystem functions. The traditional breeding framework views

plants as single organisms, and as a result, selection of crops and mutualists is often conducted

separately. This book supports the notion that plants are more accurately viewed as a consortium

consisting of a primary producer and many species of associated microbes (or depending on your

bias, microorganisms and their associated plants!). Agricultural breeding programs should select

for well-adapted consortia that can achieve necessary levels of NPP through optimization of plant-

microbial collaborations.

       This approach is illustrated by novel plant breeding efforts to enhance N-fixation capacity,

through research on root system architecture, root hairs, nodulation, N assimilation and
                                                                  Drinkwater and Snapp Page
Rhizobium traits (Herridge and Danso 1995; Vance paper). Variable success has been achieved,

as might be expected due to the complexity of the symbiosis which involves over 100 genes

(Schultze and Kondorosi, 1998). A pioneering effort over several decades of a multidisciplinary

team to enhance BNF in Alfalfa found that nitrogen assimilation and sink capacity as well as

dormancy regulation were unexpected drivers of plant-mediated interaction with Rhizobium, and

the BNF capacity of the symbiosis (-SSS).

       A successful example of selection that markedly altered rhizosphere inhabitants and the

Rhizobium infection process is the soybean breeding program in Brazil. Nitrogen-derived from

BNF and yield adaptation to nutrient limited cropping systems has been significantly enhanced

through a long-term collaboration of soil microbiolgists and plant breeders (Alves et al., 2003).

More efficient, Brady-Rhizobium strains that support higher levels of BNF through the symbiosis

were introduced and at first failed to compete against indigenous strains for nodule space (Nishi et

al., 1996). Over ten years adaptation occurred and a few of the introduced strains developed

improved competitive ability against indigenous strains for nodule space. This ability to adapt was

inadvertently selected for in this breeding effort conducted under low N conditions.


Literature Cited
                                                                        Drinkwater and Snapp Page
       Abawi, G. S. and T. L. Widmer. 2000. Impact of soil health management practices on soilborne

                     pathogens, nematodes and root diseases of vegetable crops. Applied Soil Ecology


       Achouak, W. R., Christen, M. Barakat, M.-H. Martel, and T. Heulin. 1999. Int. J. Syst. Bacteriol.


       Amellal, N., R. Bartoli, G. Villemin, A. Talouizte and T. Heulin. 1999. Effects of inoculation of

              EPS-production Pantoea agglomerans on wheat rhizosphere aggregation. Plant and Soil


       Angers, D. A. & Mehuys, G. R. 1989. Effects of cropping on carbohydrate content and water

              stable aggregation of a clay soil.Canadian Journal of Soil Science. 69:373-380.

       Angers, D. A. & Mehuys, G. R. 1988. Effects of cropping on macro-aggregation of a marine-clay

              soil. Canadian Journal of Soil Science 68:723-732.

       Alves BJR, Boddey RM and Urquiaga S. 2003. The success of BNF in soybean in Brazil. Plant

              and Soil 252:1-9.

Banziger M and Cooper M. 2001. Breeding for low input conditions and consequences for participatory

              plant breeding: examples from tropical maize and wheat. Euphytica 122:503-519.

       Banziger M and Lafitte HR. 1997. Efficiency of secondary traits for improving maize for low-

              nitrogen target environments. Crop Science 37:1110-1117.

       Barnes DK, Heichel GH, Vance CP and Ellis WR. 1984. A multiple-trait breeding program for

                     improving the symbiosis for N2 fixation between Medicago sativa and Rhizobium

              meliloti. Plant and Soil 82:303-314.
                                                                 Drinkwater and Snapp Page
Berman FI and Dubinsky Z. 1999. Balanced growth and aquatic plants: Myth or reality?

       Phytoplankton use the imbalance between carbon assimilation and biomass production to

       their strategic advantage. Bioscience 49:29-37.

Bezzate, S. S. Aymerich, R. Chambert, S. Czarnes, O. Berge and T. Heulin. 2000. Disruption of

       the Paenibacillus polymyxa levansucrase gene impairs its ability to aggregate soil in the

       wheat rhizosphere. Env. Microbiol. 2:333-342.

Booth, Mary S., Stark, John M., Rastetter, Edward. 2005. Controls on gross nitrogen cycling rates

              in terrestrial ecosystems: A synthesis and analysis of data from the literature

       Ecological Monographs: Vol. 75, No. 2, pp. 139–157.

Boyer JS. 1982. Plant productivity and environment. Science 218:443-448.

Briones AM, Jr., Okabe S, Umemiya Y, Ramsing NB, Reichardt W and Okuyama H. 2002.

       Influence of different cultivars on populations of ammonia-oxidizing bacteria in the root

       environment of rice. Applied and Environmental Microbiology 68:3067-3075.

Briones AM, Jr., Okabe S, Umemiya Y, Ramsing NB, Reichardt W and Okuyama H. 2003.

       Ammonia-oxidizing bacteria on root biofilms and their possible contribution to N use

       efficiency of different rice cultivars. Plant and Soil 250:335-348.

Buyanovsky et al., 1987. Ecology. 68:2023

Chen J and Ferris H. 1999. The effects of nematode grazing on nitrogen mineralization during

       fungal decomposition of organic matter. Soil Biology and Biochemistry 31:1265-1279.
                                                                   Drinkwater and Snapp Page
       Cheng W, Zhang Q, Coleman DC, Carroll CR and Hoffman CA. 1996. Is available carbon

                      limiting microbial respiration in the rhizosphere? Soil Biology and Biochemistry


       Clarholm M. 1985. Interactions of bacteria, protozoa and plants leading to mineralization of soil

              nitrogen. Soil Biology Biochemistry 17:181-187.

       Dalmastri C, Chiarini L, Cantale C, Bevivino A and Tabacchioni S. 1999. Soil type and maize

              cultivar affect the genetic diversity of maize root-associated Burkholderia cepacia

              populations. Microbial Ecology 38:273-284.

       Elser JJ and Urabe J. 1999. The stoichiometry of consumer-driven nutrient recycling: theory,

              observations, and consequences. Ecology 80:735-751.

       Ferris H, Venette RC, Van Der Meulen HR and Lau SS. 1998. Nitrogen mineralization by

              bacteria-feeding nematodes: verification and measurement. Plant and Soil 203:159-171.

       Galez, L., D.D. Douds, Jr., L.E. Drinkwater and P. Wagoner. 2001. Effect of tillage and farming

                      system upon VAM fungus populations and mycorrhizas and nutrient uptake of

              maize. Plant Soil 228:299-308.

       Gamliel, A., M. Austerweil and G. Kritzman. 2000. Non-chemical approach to soilborne pest

                      management B organic amendments. Crop Protection. 19:847-853.

       Germida JJ, Siciliano SD (2001) Taxonomic diversity of bacteria associated with the roots of

              modern, recent and ancient wheat cultivars. Biology and Fertility of Soils 33:410-415.

Gilreath, JP, Noling JW, Santos BM (2004) Methyl bromide alternatives for bell pepper (Capsicum

              annuum) and cucumber (Cucumis sativus) rotations. Crop Protection. 23:347-351
                                                                     Drinkwater and Snapp Page
Girvan, M.S., J. Bullimore, A.S. Ball, J.N. Pretty and A.M. Osborn. 2004. Responses of active

               bacterial and fungal communities in soils under winter wheat to different fertilizer

       and pesticide regimens. App. And Environ Microbiology 70:2692-2701.

Gouzou, L., Burtin, G., Phillippy, R., Bartoli, F. and T. Heulin. 1993. Effect of inoculation with

       Bacillus polymyxa on soil aggregation in the wheat rhizosphere: preliminary examination.

       Geoderma. 56:479-490.

Hamilton EW, III and Frank DA. 2001. Can plants stimulate soil microbes and their own nutrient

       supply? Evidence from a grazing tolerant grass. Ecology 82:2397-2402.

Harrier, L.A. and C.A. Watson. 2003. The role of arbuscular mycorrhizal fungi in sustainable

               cropping systems. Advances in Agronomy.79: 185-225

Hayes, R.J. & Beare, M.H. (1997). Influence of six crop species on aggregate stability and some

       labile organic matter fractions. Soil Biology and Biochemistry. 29:1647-1653.

Hultberg, M., B. Alsanius and P. Sundin. 2000. In vivo and in Vitro interactions by Pseudomonas

               fluorescens and Pythium ultimum in the suppression of damping off in tomato

       seedlings. Biological Control 19:1-8.

Jackson LE and Koch GW. 1997. The ecophysiology of crops and their wild relatives. Pages 3-37

       in Jackson LE, editor. Ecology in agriculture. Academic Press Inc., San Diego, CA.

Jackson LE. 1995. Root architecture in cultivated and wild lettuce (Lactuca spp.). Plant Cell and

       Environment 18:885-894.

Jackson LE, Strauss RB, Firestone MK and Bartolome JW. 1988. Plant and soil nitrogen

       dynamics in California annual grassland. Plant and Soil 110:9-17.
                                                                   Drinkwater and Snapp Page
Jasper, D.A., Robson, A.D. and Abbott, L.K., 1979. Phosphorus and the formation of vesicular-

               arbuscular mycorrhizas. Soil Biol. Biochem. 11, pp. 501B505

Jessen DL, Barnes DK and Vance CP. 1988. Bidirectional selection in alfalfa for activity of

       nodule nitrogen and carbon-assimilating enzymes. Crop Science 28:18-22.

Johnson, N.C., D.L. Rowland, L. Corkidi, L.M. Egerton-Warburton and E.B. Allen. 2003.

               Nitrogen enrichment alters mycorrhizal allocation at five mesic to semi-arid

       grasslands. Ecology 84:1895-1908.

Kerry, B.R. 2000. Rhizosphere interactions and the exploitation of microbial agents for the

               biological control of plant-parasites. Ann. Rev. Phytopathol. 38:423-441.

Kiers, E.T., S.A. West and R.F. Denison. 2002. Mediating mutalisms: farm management practices

               and evolutionary changes in symbiont co-operation. J. of Applied Ecology 39:745-


Knops, J.M.H., D. Tilman, N.M. Haddad, S. Naeem, J. Haarstad, M.E. Ritchie, K.M. Howe, P.B.

               Reich, E. Siemann and J. Groth. (1999) Effects of plant species richness on

       invasion dynamics, disease outbreaks, insect abundances and diversity. Ecol Lett 2:286-


Larson, KD, Shaw DV (1995) Relative performance of strawberry genotypes on fumigated and

       non-fumigated soils. J Am Soc Hortic Sci 120:274-277

Liljeroth E, Kuikman P and Van Veen JA. 1994. Carbon translocation to the rhizosphere of maize

       and wheat and influence on the turnover of native soil organic matter at different soil

       nitrogen levels. Plant and Soil 161:233-240.
                                                                 Drinkwater and Snapp Page
McGonigle, T.P., M.H. Miller and D. Young. 1999. Mycorrhizae, crop growth and crop

               phosphorus nutrition in maize-soybean rotations given various tillage treatments.

       Plant and Soil 210:33-42.

Morris, C. E. and J. M. Monier. 2003. The ecological significance of biofilm formation by plant-

       associated bacteria. Annu. Rev. Phytopathol. 41:429-453.

Naeem S, Hahn DR and Schuurman G. 2000. Producer-decomposer co-dependency influences

       biodiversity effects. Nature 403:762-764.

Neff, J.C., A.R. Townsend, G. Gleixner, S.J. Lehman, J. Turnbull, and W.D. Bowman. 2002.

       Variable effects of nitrogen additions on the stability and turnover of soil carbon. Nature

       London 419:915-917.

Neo et al., 1996

Nour et al., 1994

Olsson, S. and S. Alstrom. 2000. Characterization of bacteria in soils under barley monoculture

               and rotation. Soil Biol. Bichem. 32:1443-1451.

Reid, J. B. & Gross, M. J. 1981. Suppression of decomposition of 14C-labeled plant roots in the

       presence of living roots of maize and perennial ryegrass. Journal of Soil Science 33:387-


Reid, J. B. & Gross, M. J. 1981. Effect of living roots of different plant species on the aggregate

       stability of two arable soils. Journal of Soil Science 32:521-541.

Rengel, Z. 2002. Breeding for better symbiosis. Plant Soil 245:147-162.

Robleto, E.A., J. Borneman, E.W. Triplett. 1998. Effects of bacterial antibiotic production on

               rhizosphere microbial communities. Appl. Environ. Microbiol. 64:5020-5030.
                                                                Drinkwater and Snapp Page
Siciliano SD, Germida JJ, Banks K and Greer CW. 2003. Changes in microbial community

       composition and function during a polyaromatic hydrocarbon phytoremediation field trial.

       Applied and Environmental Microbiology 69:483-489.

Siciliano SD, Theoret CM, De FJR, Hucl PJ and Germida JJ. 1998. Differences in the microbial

       communities associated with the roots of different cultivars of canola and wheat. Canadian

       Journal of Microbiology 44:844-851.

Silver WL, Herman DJ and Firestone MK. 2001. Dissimilatory nitrate reduction to ammonium in

       upland tropical forest soils. Ecology 82:2410-2416.

Smith MS and Tiedje JM. 1979. The Effect of Roots on Soil Denitrification. Soil Science Society

       of America Journal 43:xx-xx.

Smith RF, van der Bosch R (1967) Integrated control. In: Kilgore WW, Doutt RL, (eds), Pest

               control: Biological, physical, and selected chemical methods. Academic Press,

       New York, pp 295-340.

Snapp, S.S. W. Kirk, B. Roman and J.D. Kelly. 2003. Root traits play a role in integrated

               management of Fusarium root rot in snap beans. HortScience38:187-191.

Snapp, S. S. and C. Shennan 1992 Salinity effects on root growth and death dynamics of tomato,

               Lycopersicon esculentum Mill. New Phytologist 121: 71-79

Sturz, A.V. and B.R. Christie. 2003. Beneficial microbial allelopathies in the root zone: the

               management of soil quality and plant disease with rhizobacteria. Soil and Tillage

       Research 72:107-123.
                                                                  Drinkwater and Snapp Page
Snapp, S. S., C. Shennan and A.H.C. van Bruggen 1991 Salinity effects on severity of

               Phytophthora parasitica infection, inorganic ion status and growth of Lycopersicon

       esculentum Mill. New Phytologist 119:275-284

Sullivan, D.J. and F. Gara. 1992. Traits of fluorescent Pseudomonas spp. Involved in suppression

               of plant root pathogens. Microbiol. Rev. 56:662-676.

Tate III RL, Parmelee RW, Ehrenfeld JG and O'Reilly L. 1991. Nitrogen mineralization: root and

       microbial interactions in pitch pine microcosms. Soil Science Society of America Journal


Tisdall, J. M. & Oades, J. M. 1979. Stabilization of soil aggregates by the root system of ryegrass.

       Australian Journal of Soil Research 17: 429-441.

VanWees, S.C.M. M. Luijendijk, I Smoorenburg, L.C. van Loon and C.M.J. Pieterse. 1999.

               Rhizobacteria-mediated induced systemic resistance (SIR) in Arabidopsis. Plant

       Mol. Biol. 41:537-549.

Vargas, M.T.A. 2000. Response of field-grown bean (Phaseolus vulgaris L.) to Rhizobium

               inoculation and nitrogen fertilizer in two Cerrados soils. Biology and Fertility of

       Soils. 32:228-233.

Wall DH and Moore JC. 1999. Interactions underground: sil biodiversity, mutualism, and

       ecosystem processes. Bioscience 49:109-117

Widmer TL, Mitkowski NA, Abawi GS 2002. Soil organic matter and management of plant-

               parasitic nematodes. JOURNAL OF NEMATOLOGY 34: 289-295
                                                                Drinkwater and Snapp Page
Workneh, F. and A. H. C. van Bruggen. 1994. Suppression of corky root of tomatoes in soil from

              organic farms associated with soil microbial activity and nitrogen status of soil and

       tomato tissue. Phytopathology. 84:688-694

Zhu et al., 2001. Phosphorus (P) efficiencies and mycorrhizal responsiveness of old and modern

              wheat cultivars. Plant and Soil 237:249-256
                                                                       Drinkwater and Snapp Page

Figure Legends

Figure 1. Variation in management intensity and consequences for the rhizosphere. Agricultural

management systems are distinguished by the reliance on agrochemical inputs (fertilizers,

pesticides, other agrochemicals) and tillage intensity. As the level of input dependance is reduced,

the reliance on plant functional diversity generally increases as does the extent of the rhizosphere

in space and time. Reduced dependence on tillage for annual crops (no-till) and perennial systems

(pastures, agroforestry) are closest to unmanaged ecosystems in terms of soil physical

environment and the presence of rhizosphere remnants as such as intact networks root and hyphal

pores (Williams et al. 2005). Conventional production systems rely on high inputs and intensive

tillage. Integrated pest management aims to substitute cultural practices and managed biodiversity

for pesticides and has become a standard approach to managing insecticides efficiently in many

cropping systems (Lewis et al. 1997). Pesticide-free agriculture is a grower-initiated approach

largely oriented toward filling consumer demand for foods that are free of pesticides (Ott 1990).

Integrated or low input systems combine agrochemical use with ecological or organic practices

with the goal of reducing environmental impacts while achieving high yields (Reganold et al.

2002). Organic agriculture is the dominant from of ecologically-based food production in the US

and seeks to minimize external inputs while avoiding all synthetic agrochemicals (NOP, 2004).

Ecological and biodynamic systems originated in Europe and have an increased requirement for

internalized N-fixation compared to US certified organic (IFOAM 2005). Two examples of

biologically-based systems with minimal tillage are 1) natural farming, an extremely low-intensity
                                                                   Drinkwater and Snapp Page
system for annual crops (Japan; Fukuoka 1985) and 2) permaculture (Australia; Mollison 1990)

integrates annual and perennial production and has spread to the US and other countries.

Figure 2. Rhizosphere processes contribute to a variety of ecosystem functions and services and

are the outcome of plant-microbial collaborations. Some, such as and soil retention are strongly

governed by plant species and others such as nitrification are largely controlled by

                                                                     Drinkwater and Snapp Page
Table 1. Comparison of rice cultivars, their rhizosphere composition and biogeochemical function

(from Briones et al. 2002, 2003).

                              Improved traditional                Modern hybrid

 Cultivar                     Mahsuri                             IR63087-1-17

 Fertilizer use efficiency    Able to use either NH4 or NO3       Greater efficiency with NH4


 Rhizosphere environment      Roots are less permeable to O2      Roots leak more O2

 Rhizosphere community        Rhizoplane is dominated by          Heterotrophs may be less abundant

 composition(1)               heterotrophs                        compared to Mahsuri rhizoplane

 Most abundant ammonia        Nitrosopira sp. (Able to grow at    Nitrosomonas sp. (Fast growing, R

 oxidizing bacteria           lower substrate concentrations, K   strategist)


 Nitrogen cycling(2)          Nitrification is not detectable     Nitrification rate: 1.2 ug N g soil


  Detection and characterization of ammonia-oxidizing bacteria by PCR-DGGE targeting the

amoA gene. Quantification of AOB in rhizosphere and rhizoplane was based on FISH.

  Based on in situ field experiments using the 15N pool dilution method.
Drinkwater and Snapp Page
            Drinkwater and Snapp Page
Figure 1.
            Drinkwater and Snapp Page
Figure 2.

To top