Common mycelium network of mycorrhizas alters plant biomass and soil properties between trifoliate orange seedlings

doi: 10.9755/ejfa.2015-09-731


INTRODUCTION
Arbuscular mycorrhizal fungi (AMF), one of soil inhabited microorganisms, can colonize roots of ~80% land's plants to form a beneficial association, viz., arbuscular mycorrhizas (AMs) (Smith and Read, 2008).The AMs accelerate metabolism of host plants (Schweiger and Muller, 2015;Wu et al., 2013bWu et al., , 2015b) ) and also change physical and chemical properties in plants and soil (Wang et al., 2014;Wu et al., 2014bWu et al., , 2015a)).Extraradical mycorrhizal mycelium as a part of mycorrhizas show fast and random growth, and can colonize the neighbor plants for the establishment of a common mycelium network (CMN) (Walder et al., 2012).The CMN can exchange nutrients and signaling between plants (Southworth et al., 2005;Meding and Zasoki, 2008;Bainard et al., 2011;Buscardo et al., 2014).The study conducted by Cruz et al. (2003) revealed that CMN induced by Gigaspora margarita could be established between papaya (Carica papaya) and bahiagrass (Paspalum notatum) plants.Various nutrients, such as P, C, N and As, were communicated by the CMN between plants (Tuffen et al., 2002;Nakano-Hylander and Olsson, 2007;Meding and Zasoki, 2008).As a result, soil CMN had the functioning for healthy growth of the receptor plants (Barto et al., 2012).
Soil aggregation, as an ecological variable, strongly affects the global climate and soil degradation, gas exchanging, and nutrient cycling (Piotrowski et al., 2004).Soil aggregate stability depends on a number of biological factors, Common mycelium networks (CMNs) of arbuscular mycorrhizas link neighbour plants and thus exhibit important roles in underground communication of substances between plants.In this study, a two-compartmented rootbox separated by 37-μm (mycelium, but not root, can pass through the size mesh) or 0.45-μm (both mycelium and root can't pass through the size mesh) mesh was used, where one compartment was inoculated with Paraglomus occultum.We confirmed whether CMNs establish between trifoliate orange seedlings and have the roles in improving both plant growth and soil properties in receptor plant (the plant inoculated without mycorrhizal fungi but infected by mycorrhizal mycelium of another inoculated plant).A CMN was formed between trifoliate orange seedlings under separation of 37-μm but not 0.45-μm mesh, resulting in a moderate root colonization of receptor plant.The mycorrhizal inoculation significantly increased leaf, stem, and root fresh weight and rhizospheric three glomalin-related soil protein (GRSP) concentrations, soil organic carbon, and mean weight diameter in the donor plant (the inoculated plant with mycorrhizal fungi).The CMN under 37-μm mesh condition had significantly positive effects on the above growth and soil properties in the receptor plant.Under 0.45-μm mesh, the AMF inoculation in donor plant considerably inhibited biomass production of receptor plant, but increased easily-extractable GRSP, total GRSP, soil organic carbon, and mean weight diameter in receptor plant.It suggested that AMF inoculation and the subsequent CMN establishment would benefit improvement of plant growth and soil aggregation and fertility in donor and receptor plant.
including fungal hyphae, microbial biomass, root systems, input of C and N from fresh matrix, and aromatic humic substances (García-Orenes et al., 2012 ).Mycorrhizal hyphae can enwind soil aggregates and thus play a vital role in forming and stabilizing soil aggregates (García-Orenes et al., 2012;Peng et al., 2013).In addition, mycorrhizal hyphae secrete a special glyprotein, glomalin, into soils defined as glomalin-related soil protein (GRSP) (Wang et al., 2014;Wu et al., 2012Wu et al., , 2014a)).GRSP improved soil structure, reduced degradation of soil organic matter, and increased carbon sequestration in soil (Rillig, 2004;Wu et al., 2008).AMF inoculation not only improved soil aggregate distribution at the size of 1-2 mm, but also considerably increased soil organic carbon (SOC) and GRSP level (Wu et al., 2014b).Wang et al. (2014) observed that different AMF species had different effects on aggregate stability.
Trifoliate orange (Poncirus trifoliata L. Raf.), a close Citrus species, is widely used as a citrus rootstock and strongly dependent on AMs.In orchard, lots of citrus trees are neighbors, whereas it is not known whether CMN formed between citrus plants.In this study, we made a twochambered rootbox to simulate a CMN and thus confirmed whether CMN could establish between trifoliate orange seedlings and CMN had the roles in increasing plant growth and improving soil properties in receptor plant.

Two-chambered rootbox preparation
The studied rootbox was made of plexiglass, whose length, width and height were 18.5, 12, and 16 cm, respectively.The rootbox was divided into two equal compartments by 37-μm or 0.45-μm nylon mesh.Meanwhile, 37-μm nylon mesh can allow mycorrhizal extraradical hyphae, but not plant roots, to enter another compartment, and 0.45-μm nylon mesh can't allow both mycorrhizal hyphae and plant roots to enter another compartment.In the rootbox, an air gap (1.5 cm width) was created in the center of rootbox using two layers of nylon mesh to reduce diffusion of substances between the two compartments.

Plant culture
Seeds of trifoliate orange were disinfected with 70% alcohol for 5 min, washed with distilled water, and germinated in autoclaved (121°C, 0.11 Mpa, 1.5 h) sands in 26°C and 16:8 photoperiod.Two four-leaf-old trifoliate orange seedlings without mycorrhization were transplanted into each compartment of the two-chambered rootbox.The 1,500 g soil (Xanthi-Udic Ferralsol, FAO system) was supplied into each chamber.The soil collected from a citrus orchard of Yangtze University campus (30°36′N, 112°14′E) was air-dried, sieved (4 mm), and autoclaved with 121°C and 0.11 Mpa for 1.5 h.
Approximately 1,500 spores of Paraglomus occultum (C.Walker) J.B. Morton & D. Redecker were applied into the designed chamber of the rootbox, where the inoculated trifoliate orange was regarded as the donor, and the seedlings from the other chamber of the same rootbox was designed as the receptor.Non-AMF inoculated treatment received same amount of autoclaved (121°C, 0.11 Mpa, 1.5 h) inocula plus 2 mL inoculum filtrate (25 μm filter) to keep similar microbial communities except the AM fungus.The spores of this strain were obtained from the Bank of Glomeromycota in China and propagated with white clover as a host plant for 12 weeks under potted conditions.
The experiment consisted of three treatments: (1) Each chamber of the rootbox under separation 37-μm mesh condition did not receive the AMF inoculation (TO - /37μm/TO -); (2) One chamber of the rootbox only received AMF inoculation as donor plant, and the two chambers of the rootbox were separated by 37-μm mesh (TO + /37μm/TO -); (3) One chamber of the rootbox only received AMF inoculation as donor plant, and the two chambers of the rootbox were separated by 0.45-μm mesh (TO + /0.45μm/TO -).Each treatment had four replicates with the completely randomized arrangement, resulting in a total of 12 rootboxes.
The experiment was performed during March 28 -August 16, 2014 in a glass house, where photon flux density is 880 μmol m -2 s -1 , day/night temperature 28/21°C, and relative air humidity 85%.

Variable determinations
At harvest, the seedlings from each chamber of the rootbox were divided into leaf, stem, and root, whose fresh weight was measured.
Root mycorrhizas were stained according to the protocol of Phillips and Hayman (1970).Root AM colonization was calculated as the percentage of AM colonized root length against total root length.Soil hyphal length was measured according to the procedure of Bethlenfalvay and Ames (1987).Glomalin-related soil protein (GRSP) fractions, including easily-extractable glomalin-related soil protein (EE-GRSP) and difficultly-extractable glomalin-related soil protein (DE-GRSP), were extracted and measured following the protocol as described by Wu et al. (2014a;2015a).Total glomalin-related soil protein (T-GRSP) is the sum of EE-GRSP and DE-GRSP.SOC content was determined by the dichromate oxidation spectrophotometric method (Rowell, 1994).Mean weight diameter (MWD) was expressed as soil aggregate stability (Kemper and Rosenau, 1986).
Distribution of water-stable aggregates at the size of 0.25-0.5,0.5-1, 1-2, and 2-4 mm was determined using wet-sieving procedure, and then MWD was calculated by the following formula: MWD = WiXi , where n, Wi, and Xi stand for number of size fractions, proportion of the i size fraction, and mean diameter of the i sieve opening (mm), respectively (Kemper and Rosenau, 1986).

Statistical analysis
The data were statistically analyzed with one-factor of ANOVA in SAS software, and the Duncan's multiple range test was used to compare the significant differences between the treatments at P<0.05.

Mycorrhizal status
There was no root AM colonization and soil hyphal length in trifoliate orange subjected to the TO -/37μm/TO -treatment.However, when the trifoliate orange seedlings as the donor were inoculated with P. occultum, root AM colonization of the donor plant was 37.66±1.84and 58.66±1.79%under the separation of 37 and 0.45 μm conditions, and root AM colonization of the corresponding receptor plant was 30.36±2.22 and 0%, respectively.Our study also indicated that the soil hyphal length ranged from 10.3±0.9 to 30.6±2.0 cm/g soil in donor plant inoculated with occultum under the separation of 37 and 0.45 μm conditions, and extraradical mycelium could pass through the 37 μm mesh but not the 0.45 μm mesh under the TO + /37μm/TO -conditions (Fig. 1), resulting in 10.3±0.9 cm/g soil in rhizosphere of receptor plant under separation of 37 μm mesh.Moreover, the root of the receptor plant under the TO + /37μm/TO - condition was colonized by the extraradical mycelium of the donor plant, further indicating that a CMN established between the seedlings from two compartments of a rootbox.

Biomass production
The present results showed that inoculation with P. occultum significantly increased the production of leaf, stem, and root fresh weight in donor plant, irrespective of separation with 37 or 0.45 µm mesh (Table 1), which agrees with the results of Wu et al. (2013b) in different citrus genotypes.Our results also indicated that the extraradical mycelium of the donor plant could pass through 37 µm but not 0.45 µm mesh to increase leaf, stem, and root fresh biomass of the receptor.Interestingly, when rice as a donor was inoculated with F. mosseae, the CMN only improved plant dry weight of the receptor rice, but not the receptor watermelon (Ren et al., 2013).This growth improvement in receptor plant caused by CMN may be due to P transfer from donor to receptor (Li et al., 2004).Tuffen et al. (2002) found that CMN established between leek plants, while did not affect plant growth of the receptor.This may ascribe to the differences in CMN types induced by AMF, the receptor plant genotypes used, and growth substrates (Drew et al., 2006), after all there is an existence of compatibility between donor and receptor plants caused by CMN (Xu et al., 2012).Our study also indicated that leaf, stem, and root biomass of the receptor plant under 0.45 µm conditions was profoundly decreased, compared with the receptor plant treated by TO -/37μm/ TO -or TO + /37μm/TO -.Possibly, the 0.45 µm mesh prevents the part underground substrate communications between donor and receptor plants, resulting in the growth decrease of the receptor plant.

GRSP production
In the rhizosphere of the donor plant, the AMF inoculation significantly increased EE-GRSP, DE-GRSP, and T-GRSP   2), which is in agreement with the findings of Wu et al. (2014a) and Wang et al. (2014), who reported the positive contribution of AMF inoculation to the production of GRSP fractions in rhizosphere of trifoliate orange seedlings grown in pots or two-chambered rootbox.In the rhizosphere of the receptor plant, both TO + /37μm/TO -and TO + /0.45μm/TO -treatments significantly increased EE-GRSP and T-GRSP level, but did not alter DE-GRSP concentration, as compared with TO -/37μm/TO -treatment.This result indicated that CMN under 37 µm mesh condition also released GRSP fractions into the receptor rhizosphere, and GRSP might pass through 0.45 µm mesh from donor to receptor  rhizospheres.EE-GRSP is new produced glomalin and relatively active, and DE-GRSP is older glomalin and insensitive (Wu et al., 2015a).Therefore, the new and active EE-GRSP, but not DE-GRSP, was significantly positively correlated with soil hyphal length, irrespective of the donor or receptor rhizosphere (Fig. 2A, 2B).

Aggregate stability
As shown in Table 2, MWD in donor and receptor rhizosphere was significantly higher in the seedlings exposed to the TO + /37μm/TO -and the TO + /0.45μm/ TO -treatment than to the TO -/37μm/TO -treatment.It is known that mycorrhizal hyphae played an important role in soil aggregation and stabilization (Leifheit et al., 2014).
And a byproduct of AM hyphae, GRSP, can cement soil aggregates into macro-aggregates as a glue agent in soil (Rillig, 2004;Wu et al., 2013a).Correlationship analysis showed that EE-GRSP and T-GRSP were significantly positively correlated with MWD in rhizosphere of donor and receptor plant (Fig. 3A, 3B), and soil hyphal length was significantly positively correlated with MWD only rhizosphere of receptor plant (Fig. 4).Wang et al. (2015) also confirmed that exogenous EE-GRSP strongly mediated soil aggregate stability in citrus rhizosphere.As stated by Wu et al. (2014a), aggregate stability was involved in many factors, including roots, root exudates, soil hyphae, and GRSPs.In these factors, GRSPs exhibited the primary functioning in aggregate stability.However, under P stress conditions, there was not a significant correlation of EE-GRSP and T-GRSP with soil hyphal length in F. mosseaecolonized trifoliate orange grown in sand (Wu et al., 2015a).This may be due to the difference of growth substrate and AMF species used.

SOC
The present study showed that SOC content in rhizosphere of trifoliate orange was significantly higher under the TO + /37μm/TO -and the TO + /0.45μm/TO -conditions than under the TO -/37μm/TO -condition, irrespective of the donor or receptor plant.GRSPs not only contained aromatic (42-49%) and carboxyl (24-30%) C but also low aliphatic (4-11%) and carbohydrate-type (4-16%) C contents (Schindler et al., 2007).Moreover, GRSPs in tropical areas accounted for about 4%c of total C in soil.In this study, EE-GRSP and T-GRSP represented significantly positive correlation with SOC (Fig. 5), suggesting that  GRSP positively contributed soil C pools.Moreover, soil hyphal length was significantly positively correlated with SOC concentration in rhizosphere of donor and receptor (Fig. 6).Wang et al. (2014) also showed that AMF-induced SOC alteration was dependent on AMF species.SOC as an important component of the soil fertility, is the core of soil quality and function (He et al., 2011).Therefore, AMF inoculation and subsequent CMN formation may take part in the contribution to soil fertility.

CONCLUSIONS
This study firstly confirmed the CMN presence between trifoliate orange seedlings under P. occultum inoculation condition.The CMN also colonized the roots of the receptor plant under separation of 37-µm but not 0.45-µm mesh, inducing a moderate root AM colonized level in the receptor plant, which significantly increase rhizospheric EE-GRSP, T-GRSP, SOC, and MWD in receptor plant, but considerably inhibited biomass production of receptor plant.Our results suggested that AMF inoculation and the subsequent CMN establishment benefited to improvement of plant growth and rhizospheric soil aggregation and fertility in donor and receptor plant.

Fig 2 .
Fig 2. Linear regression between soil hyphal length and GRSP concentration in rhizosphere of donor (A, the inoculated plant) and receptor (B, the infected plant by mycorrhizal mycelium from another inoculated plant) trifoliate orange (Poncirus trifoliata) seedlings inoculated with Paraglomus occultum and grown in a twocompartmented rootbox s eparated by 37-mm or 0.45-mm mesh (n = 12)

Fig 3 .
Fig 3. Linear regression between GRSP concentration and MWD in rhizosphere of donor (A, the inoculated plant) and receptor (B, the infected plant by mycorrhizal mycelium from another inoculated plant) trifoliate orange (Poncirus trifoliata) seedlings inoculated with Paraglomus occultum and grown in a two-compartmented rootbox separated by 37-mm or 0.45-mm mesh (n = 12).

Fig 4 .Fig 5 .Fig 6 .
Fig 4. Linear regression between soil hyphal length and MWD in rhizosphere of donor (the inoculated plant) and receptor (the infected plant by mycorrhizal mycelium from another inoculated plant) trifoliate orange (Poncirus trifoliata) seedlings inoculated with Paraglomus occultum and grown in a two-compartmented rootbox separated by 37-mm or 0.45-mm mesh (n = 12)

Table 2 : Effect of a CMN on rhizospheric GRSP concentration, SOC concentration and MWD between trifoliate orange seedlings inoculated with Paraglomus occultum and grown in a two-compartmented rootbox separated by 37-mm or 0.45-mm mesh Plant status
Note: Different letters between treatment indicate significant differences (Duncan test, P<0.05).Donor: The plant inoculated with mycorrhizal fungi.Receptor: The plant inoculated without mycorrhizal fungi but infected by mycorrhizal mycelium of another inoculated plant