South African Journal of Botany 88 (2013) 373–379

Contents lists available at ScienceDirect

South African Journal of Botany

j ourna l homepage: www.e lsev ie r .com/ locate /sa jb

Effect of arbuscular mycorrhizal fungi inoculation on root traits androot volatile organic compound emissions of Sorghum bicolor

Xue-Guang Sun, Ming Tang ⁎College of Forestry, Northwest A&F University, Yangling, Shaanxi 712100, PR China

⁎ Corresponding author at: 3rd Taicheng Road, YanglTel./fax: +86 29 87080157.

E-mail addresses: sunxg@nwsuaf.edu.cn (X.-G. Sun), ta

0254-6299/$ – see front matter © 2013 SAAB. Published bhttp://dx.doi.org/10.1016/j.sajb.2013.09.007

a b s t r a c t

a r t i c l e i n f o

Article history:Received 21 March 2012Received in revised form 7 July 2013Accepted 11 September 2013Available online xxxx

Edited by M Gryzenhout

Keywords:Arbuscular mycorrhizal fungiRoot morphologyVolatile organic compoundHS-SPMESorghum

Volatile organic compounds (VOCs) emitted by plant roots have important functions that can influence therhizospheric environment. The aim of this study was to examine the effects of arbuscular mycorrhizal (AM)fungi on the profile of root VOCs. Sorghum (Sorghum bicolor) plants were grown in pots inoculated with eitherGlomus mosseae or Glomus intraradices, which formed mycorrhiza with the roots. Control plants were grown inpots inoculated with sterile inoculum and did not form mycorrhiza. Forty-four VOCs were determined usingheadspace solid-phase microextraction (HS-SPME) and gas chromatography–mass spectrometry (GC-MS).Alkanes were the most abundant type of VOCs emitted by both mycorrhizal and non-mycorrhizal plants. Boththe quantity and type of volatiles were dramatically altered by the presence of AM fungi, and these changeshad species specificity. Compared with non-mycorrhizal plants, mycorrhizal plants emitted more alcohols,alkenes, ethers and acids but fewer linear-alkanes. The AM fungi also influenced the morphological traits ofthe host roots. The total root length and specific root length of mycorrhizal plants were significantly greaterthan those of non-mycorrhizal plants; however, both the incidence and length of root-hair were dramaticallydecreased. Our findings confirm that AM fungi can alter the profile of VOCs emitted by roots as well as the rootmorphology of sorghum plants, indicating that AM fungi have the potential to help plants adapt to and altersoil environments.

© 2013 SAAB. Published by Elsevier B.V. All rights reserved.

1. Introduction

Many interactions between organisms are based on the emissionand perception of volatiles (Wenke et al., 2010). Individual scentcompounds or bouquets of odors are essential and suitable for inter-and intra-species recognition, attraction, and defense in their respectiveecological niche because they can be detected in small amounts by theorganisms and diffuse over long distances in the atmosphere. Untilrecently, most of the research on volatiles has focused on organismsthat live aboveground or the aerial parts of plants because it is easierto study volatiles that diffuse over long distances in the atmospherethan those that diffuse through the soil (Laothawornkitkul et al., 2009;Wenke et al., 2010).

However, in addition to the volatiles emitted by the aerial parts ofplants, roots normally release a complex mixture of chemicals as well.By emitting volatile metabolites underground, plants can directly orindirectly influence the community of soil-dwelling organisms byattempting to defend themselves against herbivores and plant pathogen-ic fungi and bacteria, supporting beneficial symbioses, and combatingcompetitive plant species (Nardi et al., 2000). A blend of unidentifiedroot volatiles emitted by Echinacea angustifolia have been shown to

ing, Shaanxi 712100, PR China.

ngm@nwsuaf.edu.cn (M. Tang).

y Elsevier B.V. All rights reserved.

have allelopathic effects on a range of different plant species such asLactuca sativa, Panicum virgatum and Sporobolus heterolepis (Viles andReese, 1996). Given that the microbial community in the rhizosphere isusually limited by carbon availability, carbon-containing root volatilescan be an important carbon and energy source (Kleinheinz et al., 1999;Owen et al., 2007). A direct correlation has been shown between thecomposition of volatiles and the choice of host plants, either carrotroots (Daucus carota ssp. sativus) or oak roots (Quercus sp.), by the larvaeof the forest cockchafer Melolontha hippocastani (Weissteiner andSchütz, 2006). These studies all indicate that root-emitted compoundscan play important and diverse roles in the rhizosphere (Lin et al.,2007; Wenke et al., 2010).

Changes in essential oils (Copetta et al., 2006; Gupta et al., 2002;Kapoor et al., 2004; Khaosaad et al., 2006; Prasad et al., 2011) and theemission of aboveground volatile organic compounds (VOCs) (Leitneret al., 2010; Rapparini et al., 2008; Strack et al., 2003) has been docu-mented in a few plants in response to mycorrhization. However, theeffect of AM on volatiles emitted by live roots has not been exploredto date. Leitner et al. (2010) studied the influence of mycorrhizationon herbivore-induced volatile emission in the aboveground parts ofMedicago truncatula. They found that the overall pattern of VOC emis-sion was quite similar in mycorrhizal and non-mycorrhizal plantswhen the plants were undamaged but that some differences could beobserved when the plants were damaged by herbivory. Similarly,Khaosaad et al. (2006) also found that although mycorrhization can

374 X.-G. Sun, M. Tang / South African Journal of Botany 88 (2013) 373–379

influence the content of secondary metabolites in aboveground plantparts, these changes depend on the genetic background of the plant.Changes in essential oils induced by arbuscular mycorrhizal (AM)fungi have been shown to depend not only on the temporal progressionof the interaction but also on the fungal species involved (Larose et al.,2002).

Fungi are also capable of emitting a diverse array of VOCs, includingmany that are not commonly produced by plants (Leff and Fierer, 2008).Larsen and Frisvad (1995) investigated the VOC production of 47different taxa within the genus Penicillium and identified 196 differentvolatile metabolites. Truffles can emit up to 200 volatile metabolites(Buzzini et al., 2005). Bäck et al. (2010) demonstrated that the VOCsemitted by eight fungi typical of the fungi occurring in boreal forestsoils comprised 27 compounds. The fungal VOCs identified in thesestudies comprised a variety of compounds, including alcohols, esters,ketones, alkenes, and aromatic compounds.

AM fungi are an integral part of terrestrial ecosystems given thatthey form symbiotic associations with more than 90% of plant species.The most prominent contribution of these fungi to plant growth ismainly the uptake of phosphorus and other elements by extraradicalmycorrhizal hyphae, and transference to the root tissues (Giri andMukerji, 2004; Wu et al., 2010). The effect of ectomycorrhiza on theemission of root VOCs has been studied previously by Krupa and Fries(1971): roots of Scots pine (Pinus sylvestris) infected by themycorrhizalfungus Boletus variegates led to an increased production and/or accumu-lation of volatiles compared with the control roots.

Given that VOCs play a fundamental role in the life of a plant, variousanalytical techniques have been developed to determine their identity(Dib et al., 2010). Classical methods such as solvent extraction andsteam distillation extraction are routinely used, followed by gas chro-matography–mass spectrometry (GC-MS) analyses. These methodsusually require a large biomass and relevant solvent selection so thatother constituents of the cell do not interfere with the GC-MS analysis(Bais et al., 2003). By contrast, headspace solid phase microextraction(HS-SPME) enables rapid fingerprinting of the headspace of a plant(Bicchi et al., 2000; Bicchi et al., 2007; Muselli et al., 2009; Paoliniet al., 2007) and is recommended as a sensitive and reproducibletechnique for identifying VOCs. An attractive feature of this method isits near in-situ collection of VOCs, and it has been successfully used inprevious analyses of root VOCs (Bais et al., 2003; Mucciarelli et al.,2007). Until recently, the method most commonly used to collect theVOCs emitted from plants was still the solvent extraction method.However, even when methods have involved the HS-SPME technique,the plant materials have still been dried and ground, or cut into smallpieces (Dib et al., 2010) before collecting the VOCs. Therefore, the VOCscollected in previous studies may not have been those emitted by plantsduring their normal metabolism. By contrast, in this study, we collectedand analyzed the VOCs emitted directly from fresh roots to representemissions that occur under normal physiological conditions.

Sorghum (Sorghum bicolor) is the fifth most important cereal crop(ICRISAT, 2004) and is grown worldwide. The influence of AM fungi onsorghum performance has been investigated previously (Bagayokoet al., 2000; Isopi et al., 1995); however, few studies on root trait compar-isons have been reported to date andnone comparing the VOC emissionsof mycorrhizal and non-mycorrhizal roots. We hypothesize that the for-mation of AM may influence the profile of VOCs in the rhizosphere andmay show species specificity. In this study, we compared the root traitsand VOC emissions of mycorrhizal and non-mycorrhizal sorghum rootsand examined the influence of AM fungal species on VOC emissions.

2. Materials and methods

2.1. Experiment design and AM fungal materials

The experiments were laid out in a randomized complete blockdesign with three treatments (plants inoculated with Glomus mosseae,

Glomus intraradices, or without inoculation), and each treatmentcontains 14 plants (five for plant growth responses determination, fivefor the root traits measurement and four for the VOC analysis). For eas-ier collection and least detriment to the roots, the sorghum plantletswere planted in river sand. The sand was first passed through a 2 mmsieve and washed with tap water and then sterilized with hot air at180 °C for 3 h before placing into pots with a 20-cm diameter top(maximum volume 4 L).

The AM fungal inoculants G. mosseae (BGC NM01A) and G.intraradices (BGC BJ09), which were bought from Bank of Glomales inChina, comprised spores, mycorrhizal fragments, and infested soil. Atablespoon of inoculant was added to each pot (approximately 1 g ofinoculant per seedling). Each gram of G. mosseae inoculant contained23 spores and each gram of G. intraradices inoculant contained 37spores. The number of spores per gram of soil was determined using awet sieving and decanting method (Gerdemann and Nicolson, 1963).Pots containing the non-mycorrhizal control plants (CK) were mock-inoculatedwith a tablespoon of sterile inoculum that had been sterilizedby autoclaving.

2.2. Plant and growth conditions

Sorghum bicolor seeds, which were bought from a local supplier(cultivar, TaiGu LvBao® F1 hybrid), were surface sterilized with 5%H2O2-solution for 5 min and then rinsed with sterilized distilledwater. The seeds were then left to germinate and grow in sterilizedriver sand supplied with tap water only. After the seedlings haddeveloped two leaves each, four uniform seedlings were transplantedto each pot containing the sterilized sand and the live or sterile AMfungal inoculum. Two weeks after transplanting, the seedlings werethinned to one seedling per pot.

Plants were grown in a glasshouse under natural day/night condi-tions and each pot of plants was watered with 100 mL of tap waterevery three days. Twoweeks after transplanting, 50 mL of Hoagland so-lution (containing only half the concentration of phosphorus normallypresent in Hoagland solution, i.e. to 0.5 mmol L−1 PO4

−3) was addedto each pot; this procedure was repeated every two weeks. This lowconcentration of P improves plant growthwhilemaintaining high levelsof AM colonization (Gao, 2002). Harvesting was carried out 12 weeksafter transplanting. The mean temperature during the growth periodwas 28 °C.

2.3. Root extraction and measurements

At harvest, the pot was soaked in water. The whole root system ofeach plant was harvested by carefully washing away the sand with agentle water spray to minimize disturbance to the root system. Theplant height, diameter of stem, and the number of leaf were determinedwith five replications. After that the dry weights (80 °C for 48 h) ofshoots and roots were determined separately, and then the root/shootratio was calculated.

For root morphology determination, roots from another five plantswere separated from shoots with scissor individually and the rootmorphological parameters were determined separately for each plant.Total root length was determined using the gridline intersection meth-od (Tennant, 1975). The live fine roots (b2-mm diameter) were thenseparated from coarse roots (N2-mmdiameter), dead roots, and organicmatter fragments. Only live fine roots were used for further analysis.Live fine roots were distinguished from the dead roots under a stereo-microscope based on color, root elasticity, and the degree of cohesionof cortex, periderm, and stele (Röderstein et al., 2005).

Fine root diameter was determined by examining 100 randomlyselected root segments (1 cm in length) for each plant. Root hair lengthwas determined by measuring up to 200 root hairs on 50 fine rootsegments (1 cm in length) for each sample. Root hair incidence wasassessed by the presence or absence of root hairs in 200 intersections

Table 1Effects of AM inoculation on growth responses of sorghum plants.

Measurement Treatments⁎

CK G. m. G. i.

Plant height (cm) 35.80 ± 5.26 a 48.60 ± 4.22 b 46.20 ± 2.60 bStem diameter (mm) 2.76 ± 0.38 a 4.46 ± 0.31 b 4.70 ± 0.52 bLeaf number 5.80 ± 0.84 a 6.60 ± 0.55 ab 7.40 ± 1.52 bDry weight (g) Shoot 0.48 ± 0.07 a 1.00 ± 0.27 b 1.25 ± 0.22 b

Root 0.21 ± 0.10 a 0.46 ± 0.21 a 0.33 ± 0.10 aTotal 0.69 ± 0.17 a 1.45 ± 0.46 b 1.58 ± 0.23 b

Root:shoot ratio 0.41 ± 0.16 a 0.45 ± 0.12 a 0.27 ± 0.10 a

⁎ Treatments: plants were grown in pots inoculated with sterile inoculum (CK),Glomusmosseae (G. m.), or Glomus intraradices (G. i.); values are presented as the mean ± thestandard deviation (S.D.), n = 5. Data were compared using Tukey's HSD test: significantdifferences (P b 0.05) are identified by different letters within a row.

375X.-G. Sun, M. Tang / South African Journal of Botany 88 (2013) 373–379

of roots using a gridline method (Siqueira and Saggin-Júnior, 2001;Zangaro et al., 2005) for each sample. Fine root diameter and root hairlength were determined using a microscope at ×100 magnificationwith an ocular micrometer (Manjunath and Habte, 1991; Schweigeret al., 1995). All the fresh root fractions examined above were placedin a drying chamber at 80 °C until constant weight to obtain the drymass. Specific root length (root length per unit of root mass) was de-rived from calculating the root length and root dry mass of each plant.

2.4. HS-SPME technique and VOC collection

VOCs emitted by sorghum roots were collected using the HS-SPMEtechnique. First, the water-flushed roots (described in Section 2.3)were separated from the shoots using scissors (four independent plantswere used, i.e. n = 4). Approximately 2 g of roots (fresh weight) wereplaced in a 25 mL vial (Supelco Inc., Bellefonte, PA, USA) for micro-distillation (the roots were sheared from the root-crown if necessary).Samplings were performed using 25 mL vials to reduce headspacevolumes. Two drops of distilled water were added to the surface ofroots to keep them moist. All the above procedures were performedquickly and gently to minimize disturbance to the roots. The HS-SPMEtechnique involves adsorption of analytes by a thinfilm of the extractivephase covering the surface of a silica-fused fiber. The fiber is inserteddirectly into the headspace containing the analytes. For all experiments,a SPME holder 57330-U (Supelco) with a fiber coated withdivinylbenzene/carboxen/polydimethylsiloxane (DVD/CAR/PDMS) 50/30 μm stationary phase/film thickness was used. The fiber was intro-duced into the vial while still protected inside the needle of the holderand then exposed to root headspace for 45 min at 30 °C. It wasextracted and desorbed for at least 10 min into the gas chromatograph(GC). The fiber was reconditioned by inserting it into the GC injector at220 °C for at least 10 min.

2.5. Gas chromatography–mass spectrometry (GC-MS)

After inserting the SPME needle into the GC injection port, the fiberwas pushed out and thermally desorbed for 10 min at 220 °C. Analyseswere carried out using an Agilent 6890 N GC at 230 °C and held insplitless mode for 1 min. A 30 m × 0.25 mm i.d. DB-Wax capillary col-umn (Agilent)with a 0.25 μmfilm (PEG-20 M) thicknesswas usedwiththe following temperature program: initial temperature 40 °C, temper-ature rise to 60 °C at the rate of 4 °C/min and held for 10 min, thenramp of 5 °C/min up to 200 °C and held for 2 min (split/splitless ratio1:20). The carrier gas was helium at 1 ml/min. GC-MS analysis ofSPME extracts was performed with the GC coupled to a quadrupolemass spectrometer 5973 MSD (mass selective detector, Agilent). Ionsource temperature, 150 °C; energy ionization, 70 eV; electron ioniza-tion mass spectra was acquired with a mass range of 35–350 Da. Head-space components were identified by comparison with the NIST/EPA/NIH Mass Spectral Library.

2.6. AM colonization

After the VOC analysis, the root samples (n = 4) were examined todetermine root colonization by AM fungi. The proportion of plant rootscolonized with AM fungi was estimated by staining the roots accordingto Phillips and Hayman (1970) and then examining them microscopi-cally using a gridline intersect method (Giovannetti and Mosse, 1980).At least 150 randomly selected 1-cm root segments (i.e. 150 cm ofroot) were examined per plant.

2.7. Data analysis

Data were checked for normality and homogeneity of variancebefore further analysis. Data were statistically analyzed using one-wayanalysis of variance (ANOVA), and the means were compared by

Tukey's HSD (P ≤ 0.05) test using the statistical software SPSS 17.0.0(Statistical Product and Service Solutions, SPSS Inc. Chicago, IL). AMcolonization data were arcsine square root transformed before theanalysis and presented in the original scale of measurement. Data arepresented as means with their standard deviation.

3. Results

3.1. Plant growth

After 12 weeks growth, both AM fungi formedmycorrhiza with sor-ghum roots: 65.85% ± 1.97% of roots examined had been colonized byG. mosseae and 72.25% ± 5.26% of roots examined had been colonizedby G. intraradices. Plants grown in pots inoculated with sterile inoculum(CK) did not form mycorrhiza.

AM fungi improved the growth of sorghumplants and the results forboth AM fungal treatments were similar. The plant height, stem diame-ter, leaf number, and shoot mass (including leaf blade) of mycorrhizalplantswere all significantly higher than those of non-mycorrhizal plants(Table 1). Furthermore, the total masses of the mycorrhizal plants weresignificantly higher than those of the non-mycorrhizal plants.Compared with the CK treatment, total biomass was increased by110% and 129% in the G. mosseae and G. intraradices treatments, respec-tively. By contrast, the divergences between root masses were notsignificant even though the mycorrhizal roots showed apparentincreases compared with the non-mycorrhizal roots. Similarly, no sig-nificant differences between treatments were found for the root:shootratio, although the plants grown in pots inoculated with G. intraradicesdid show a decreasing trend.

3.2. Effects of AM inoculation on root traits

The diameter of the fine roots (b2 mm) of sorghum grown in potsinoculatedwith liveG.mosseae or sterile inoculumwere similar, where-as the diameter of the fine roots of sorghum grown in pots inoculatedwith live G. intraradices were significantly greater (Fig. 1a). After12 weeks growth, the roots of sorghum were extensive and more than20 m in length (Fig. 1b). In general, mycorrhizal plants had significantlyhigher values for both total root length and specific root length thannon-mycorrhizal plants (Fig. 1b, c). Plants that formed mycorrhizawith G. intraradices had the longest roots (more than twice the lengthof non-mycorrhizal plants), whereas plants that formed mycorrhizawith G. mosseae had the longest specific roots.

Root-hair incidence for plants grown in pots inoculated with G.mosseae (15.72%) or G. intraradices (18.30%) was half the level of thatrecorded for plants grown in pots inoculated with sterile inoculum(44.67%) (Fig. 1d, e). A similar situation was also recorded for root-hair length. Furthermore, the root-hair incidence and root-hair lengthof plants grown in pots inoculated with either of the AM fungi weresimilar.

Fig. 1. Fine root diameter (a), root length (b), specific root length (c), root-hair incidence (d), and root-hair length (e) of 12-week-old sorghum plants. Treatments: plants were grown inpots inoculatedwith sterile inoculum (CK), Glomusmosseae (G.m.), orGlomus intraradices (G. i.). Error bars represent ± the standard deviation (S.D.) (n = 5). Data were compared usingTukey's HSD test, and significant differences (P b0.05) are identified by different letters within one chart.

376 X.-G. Sun, M. Tang / South African Journal of Botany 88 (2013) 373–379

3.3. Effects of AM inoculation on the emission of root VOCs

From headspace collections of root emissions, 44 VOCs were identi-fied by GC-MS in total: 18 from the roots of plants that received the CKtreatment, 21 from the roots of plants grown in pots inoculated withG. mosseae and 17 from the roots of plants grown in pots inoculatedwith G. intraradices (Table 2). Both the components and contentsvaried dramatically among the three treatments. Only ninecompounds were detected in more than one treatment: namely, aceticacid, dianhydromannitol, tetratetracontane, octacosane, heptacosane,n-decanoic acid, pentaethylene glycol, octadecane, 3-ethyl-5-(2-ethylbutyl)-, and tetratriacontane.

The linear-alkanes were the most abundant type of VOCs emitted(Fig. 2), accounting for 73.55% ± 7.93% (CK treatment), 56.75 ± 8.62%(G. mosseae treatment), and 34.93% ± 1.48% (G. intraradices treatment)of the VOCs emitted. The single compounds with the highestemissions for the CK, G. mosseae and G. intraradices treatmentswere tetratriacontane (25.95% ± 8.19%), tetratriacontane (32.47% ±10.98%), and acetic acid (23.98% ± 2.17%), respectively. Compared withnon-mycorrhizal plants, mycorrhizal plants emitted significantly morealcohols, alkenes, ethers and acids but fewer linear-alkanes (P ≤0.05).Non-mycorrhizal plants did not emit branched-alkanes and mycorrhi-zal plants did not emit aldehydes. There were no obvious differencesin the quantities of esters emitted by mycorrhizal or non-mycorrhizalplants (P ≤0.05).

It is noticeable that the VOC profiles that were induced by AM fungihad high species specificity. The roots of plants grown in pots inoculatedwith G. mosseae emitted more linear alkanes and alcohols but fewerbranched alkanes, ketones, acids and phenyl compounds comparedwith the roots of plants grown in pots inoculated with G. intraradices.

4. Discussion

In this study, we found that AM fungi can alter the profile of VOCsemitted by roots. Alkaneswere themost abundant type of VOCs emittedby both mycorrhizal and non-mycorrhizal plants. The VOC emissionprofiles induced by the AM fungi had species specificity. Comparedwith non-mycorrhizal plants,mycorrhizal plants emittedmore alcohols,alkenes, ethers and acids but fewer linear alkanes. We also found thatAM fungi enhanced the growth of roots while suppressing root-hair de-velopment, which supports similar findings reported in earlier studies.

Mycorrhization of sorghum plants had a positive effect on plantgrowth, which may be due to the more limited supply of nutrients toplants grown in pots inoculated with sterile inoculum (particularlyphosphorus) (Smith and Read, 1997). Similar results have been demon-strated in several other research studies (Bagayoko et al., 2000; Isopiet al., 1995). Although not significantly different, the slightly smallerroot:shoot ratio recorded for the G. intraradices treatment comparedwith the other two treatments may reflect that the roots of mycorrhizalplants had to supply a relatively larger shoot not only with mineralnutrients but also with water (Kothari et al., 1990).

Roots function dually as a support system and as a nutrient uptakeorgan of plants, and their morphology is a reflection of their functions.Here, we found that mycorrhization with AM fungi can greatly changethe morphological traits of sorghum roots. The diameter of fine roots,total root length and specific root length of mycorrhizal plants wereall significantly greater than those of non-mycorrhizal plants, whereasthe root-hair incidence was significantly lower and root-hair lengthwas significantly shorter. Previous studies (Bagayoko et al., 2000; Isopiet al., 1995) have also reported that mycorrhizal sorghum plants havea significantly greater total root length, specific root length and more

Table 2VOCs (mean ± S.D. of the relative peak area, %) emitted from sorghum roots after 30 min HS-SPME.

RT (min) Compounds Relative peak area (%)

Treatments⁎

CK G. m. G. i.

5.81 Acetone – 1.96 ± 1.95 –

6.14 2,5-Dimethoxy-4-(methylsulfonyl)amphetamine 3.52 ± 0.94 – –

9.33 Cyclohexene,1-(2-methylpropyl)- – 4.20 ± 1.40 –

14.10 Ethanol, 2-(hexadecyloxy)- – – 2.83 ± 0.3419.99 Acetic acid 4.37 ± 0.55a 5.95 ± 2.14a 23.98 ± 2.17b20.49 Furfural 1.29 ± 0.82 – –

25.15 2-Pyrrolidinone, 1-methyl- – – 2.39 ± 0.6325.16 Isosorbide 0.96 ± 0.59 – –

25.48 Dianhydromannitol 1.55 ± 0.92a 2.24 ± 1.01a –

30.38 1,2-Benzisothiazole – – 2.68 ± 0.7930.40 Benzothiazole 2.69 ± 1.51 – –

31.25 Octadecane, 3-ethyl-5-(2-ethylbutyl)- – – 2.29 ± 0.4031.30 Tetratetracontane – 5.91 ± 3.28a 10.09 ± 0.94a31.38 Octacosane 17.91 ± 2.30b 9.02 ± 1.08a 6.92 ± 2.32a31.53 Nonacosane 19.06 ± 1.91 – –

32.14 1b,5,5,6a-Tetramethyl-octahydro-1-oxa-cyclopropa[a]inden-6-one – – 2.38 ± 0.7032.33 7-Methoxy-2,2,4,8-tetramethyltricyclo[5.3.1.0(4,11)]undecane 1.58 ± 1.29 – –

33.03 Azulene, 1,4-dimethyl-7-(1-methylethyl)- 1.46 ± 0.87 – –

33.04 Naphthalene, 1,6-dimethyl-4-(1-methylethyl)- – 2.08 ± 0.29 –

33.42 Heptacosane 4.83 ± 1.21b 5.14 ± 0.57b 2.13 ± 0.54a34.08 9H-Fluorene, 9-methylene- 1.78 ± 0.39 – –

34.09 á-N-Normethadol – 1.73 ± 0.80 –

35.03 n-Decanoic acid 2.03 ± 0.50a 2.47 ± 1.15a –

35.16 Octaethylene glycol monododecyl ether – – 1.30 ± 0.0236.03 Pentaethylene glycol – 0.90 ± 0.27a 1.77 ± 0.14b36.68 Butyrophenone, 2′,3,4′,6′-tetramethyl- 2.66 ± 0.78 – –a36.93 Hexagol – 3.51 ± 1.14 –

37.03 Geranyl isovalerate 1.30 ± 0.99 – –

37.04 Methyl 2,3,4-tri-O-acetyl-á-D-xylopyranoside – – 1.66 ± 0.5537.36 Octadecane, 3-ethyl-5-(2-ethylbutyl)- – 2.79 ± 0.69a 6.23 ± 0.31b37.46 Tetratriacontane 25.95 ± 8.19a 32.47 ± 10.98a –

37.96 Tetracosane 5.79 ± 1.65 – –

38.39 Hentriacontane – – 15.79 ± 1.5238.44 Octadecanoic acid – 3.66 ± 0.92 –

38.50 Octadecanoic acid, 2-hydroxy-1,3-propanediyl ester – 1.29 ± 0.33 –

39.28 Pentacosane – 4.21 ± 1.27 –

39.35 Benzene, 1,1′-(1-butenylidene)bis- – – 12.35 ± 0.5539.38 Benzene, 1,1′-(3-methyl-1-propene-1,3-diyl)bis- – 3.50 ± 1.74 –

39.76 Hexadecanoic acid, 2-(octadecyloxy)ethyl ester – 2.38 ± 1.00 –

40.12 Octaethylene glycol – – 2.07 ± 0.6740.57 Methyl á-d-ribofuranoside – 1.90 ± 0.31 –

40.61 à-d-Lyxofuranoside, methyl – 2.65 ± 0.37 –

40.82 1,2-Benzenedicarboxylic acid, bis(2-methylpropyl) ester – – 3.15 ± 0.3640.97 1,2-Benzenedicarboxylic acid, butyl octyl ester 1.25 ± 0.36 – –

RT, retention time; ‘–’ indicates that no compounds were detected.⁎ Treatments: plants were grown in pots inoculatedwith sterile inoculum (CK), Glomusmosseae (G. m.), or Glomus intraradices (G. i.); values are presented as themean ± S.D. (n = 3–4).

Values in the same row followed by the same letters are not significantly different from each other (P b0.05).

Fig. 2. Categories of VOC profiles identified from the sorghum roots of plants grown in pots inoculatedwith G.mosseae (G. m.), G. intraradices (G. i.) or sterile inoculum (CK). Relative peakareas are used to indicate the relative content of compounds. The total areas exceed 100% because some VOCs belong to more than one category: for example, furfural belongs to thealdehydes and alkenes, and dianhydromannitol belongs to the alcohols and alkenes.

377X.-G. Sun, M. Tang / South African Journal of Botany 88 (2013) 373–379

378 X.-G. Sun, M. Tang / South African Journal of Botany 88 (2013) 373–379

root branching compared with non-mycorrhizal plants. Price et al.(1989) have suggested that the hyphae of AM fungi may perform anabsorbing role at a lower metabolic cost than root-hairs. Therefore, thedevelopment of fewer root hairs on the roots ofmycorrhizal plants com-pared with the control treatment was not surprising. Foehse and Jungk(1983) demonstrated that both the density and length of root hairs arenegatively correlated with the plant internal P concentration, whichmay explain, at least in part, why the roots of mycorrhizal plants had alower root-hair density and were shorter in length compared with theroots of non-mycorrhizal plants given that AM fungi are known toincrease shoot and root P concentrations.

Plants emit a wide range of VOCs from roots into the surroundingrhizosphere and many of these compounds mediate important interac-tions between plants and their environment (Arimura et al., 2005;Louarn et al., 2012). Although volatile metabolites involved in theinteraction of plant roots and the establishment of infection byectomycorrhizal fungi have been documented previously (Krupa andFries, 1971; Krupa et al., 1973), the effects of various other microorgan-isms, particularly the effect of AM fungi on the emission of root VOCs,have been long neglected. Here, we found that the three treatmentstested resulted in the emission of different root VOC profiles. It isbelieved that fungal elicitors can stimulate the synthesis of existingplant metabolites or induce the transient accumulation of new second-ary metabolite compounds (Bais et al., 2003; DiCosmo and Misawa,1985). The presence of an endophytic fungus other than AM fungi hasbeen found to dramatically alter the emission of root volatiles, boththe components and the quantities of VOCs detected (Mucciarelliet al., 2007). Furthermore, given that root morphology can determineroot function, the significant changes induced by AM fungi inoculationobserved in this study may contribute to the differences in the VOCsemitted. Moreover, the VOCs collected in our study may include somecompounds emitted by the AM fungi given that the roots cannotcompletely exclude the fungal tissues. This may be the main reasonwhy divergences among treatments were so large. Like plants, fungialso produce a multitude of volatiles, and volatile production and emis-sion is highly species dependent (Kuske et al., 2005; Larose et al., 2002;Schnürer et al., 1999).

The two AM fungi tested here induced the emission of distinct VOCprofiles from plant roots. Although no direct evidence is available atpresent to support this phenomenon, many studies have found similarresults while testing the effects of AM on plant performance. The effectof mycorrhization on plant growth, mineral nutrition and secondarymetabolites varies depending on the specific AM fungus involved(Cavagnaro et al., 2004; Larose et al., 2002; Morandi et al., 1984; Rajuet al., 1990).

Sorghum has a strong phytotoxic impact on weeds, and this toxicityresults from the production and release of phenolics (Zakir et al., 2008and references therein). Here, we detected significant changes inphenolic compounds among the three treatments, which may indicatethe potential and different roles that distinct AM fungal species play inplant community determination. Although sorgoleone is a major com-ponent of the root exudates of sorghum, and is one of the most studiedallelochemicals (Czarnota et al., 2003; Dayan et al., 2010), we did notdetect it in our study. This may be because this compound has low vol-atility. Indeed, sorgoleone is more likely to be found in root extracts andexudates.

Low-molecular-weight carbon compounds, such as organic acids,are readily assimilated by microorganisms and are proposed to play aprimary role in regulating microbial community dynamics in therhizosphere (Shi et al., 2011 and references therein). In this study, wedetected three kinds of organic acids (acetic acid, n-decanoic acid andoctadecanoic acid) in the VOCs emitted by sorghum roots. Plantsgrown in pots inoculated with G. mosseae induced a greater number ofdifferent types of acids and G. intraradices induced greater quantitiesof acids compared with plants grown in pots inoculated with sterile in-oculum. Fungi are believed to emit a substantial amount of alcohols

(Kuske et al., 2005). Here,we also found thatmycorrhized roots emittedmore alcohols than the non-mycorrhized, which may reflect alcoholemissions by the AM fungi rather than the roots.

The function of biogenic VOCs emitted from flowers or fruits arefairly well-defined; however, the functions of compounds emittedfrom roots are less well understood (Kesselmeier and Staudt, 1999).The functions of most of the VOCs detected in our study are unknownowing to the limited information available. Further studies to investi-gate the function of specific compounds would be useful. However,based on the limited literature available at present, we can hypothesizethat the VOCs emitted by roots may have important functions indetermining the rhizosphere environment, such as involvement inplant–fungus interactions (Mucciarelli et al., 2007; Wenke et al.,2010), the regulation of microbial activity (Lin et al., 2007; Wenkeet al., 2010), and allelopathy (Czarnota et al., 2003; Dayan et al., 2010;Louarn et al., 2012), as well as contributing to the belowground carboncycle (especially alkanes) (Kleinheinz et al., 1999; Wenke et al., 2010).The different root VOC profiles induced by the three different treat-ments suggest that AM fungi have the ability to change the rhizosphere,and that those changes have species specificity.

It is surprising to find that AM fungi elicited some novel compoundssuch as pentaethylene glycol, hexagol, and pentacosane. Research fromBais et al. (2003)may add some clues to our findings, and they believedthat plant cell-wall receptors have been activated by a molecule(s) of afungal elicitor to elicit a response(s) leading to the synthesis of volatilecompounds. These novel compounds might have eco-physiologicalroles in plant defense or rhizosphere environment determination.

Considering the broad distribution and the importance of AM fungalassociations in terrestrial ecosystems in general andmore specifically inagricultural systems (Sawers et al., 2008), there is a need for furtherstudies to analyze the impact of mycorrhizal associations on root VOCsemitted by a broad range of species and to elucidate the functionalproperties of the volatile fractions from an environmental angle.

Acknowledgements

This research was supported by the National Natural ScienceFoundation of China (31270639, 31170607, 31170567), the Programfor Changjiang Scholars and Innovative Research Team in University ofChina (IRT1035) and the Ph. D. Programs Foundation of EducationMinistry of China (20100204110033, 20110204130001). We alsothank anonymous reviewers for their valuable suggestions to enhancethe manuscript.

References

Arimura, G., Kost, C., Boland, W., 2005. Herbivore-induced, indirect plant defences.Biochimica et Biophysica Acta-molecular and Cell Biology of Lipids 1734, 91–111.

Bäck, J., Aaltonen, H., Hellén, H., Kajos, M.K., Patokoski, J., Risto, T., Pumpanen, J., Heinonsalo,J., 2010. Variable emissions of microbial volatile organic compounds (MVOCs)from root-associated fungi isolated from Scots pine. Atmospheric Environment 44,3651–3659.

Bagayoko, M., George, E., Römheld, V., Buerkert, A., 2000. Effects ofmycorrhizae and phos-phorus on growth and nutrient uptake of millet, cowpea and sorghum on a WestAfrican soil. Journal of Agricultural Science 135, 399–407.

Bais, H.P., Dattatreya, B.S., Ravishankar, G.A., 2003. Production of volatile compounds byhairy root cultures of Cichorium intybus L. under the influence of fungal elicitorsand their analysis using solid-phase micro extraction gas chromatography–massspectrometry. Journal of the Science of Food and Agriculture 83, 769–774.

Bicchi, C., Drigo, S., Rubiolo, P., 2000. Influence of fibre coating in headspace solid phasemicroextraction-gas chromatographic analysis of aromatic and medicinal plants.Journal of Chromatography. A 892, 469–485.

Bicchi, C., Cordero, C., Liberto, E., Sgorbini, B., Rubiolo, P., 2007. Reliability of fibres insolid-phase microextraction for routine analysis of the headspace of aromatic andmedicinal plants. Journal of Chromatography. A 1152, 138–149.

Buzzini, P., Gasparetti, C., Turchetti, B., Cramarossa, M.R., Vaughan-Martini, A., Martini, A.,Pagnoni, U.M., Forti, L., 2005. Production of volatile organic compounds (VOCs) byyeasts isolated from the ascocarps of black (Tuber melanosporum Vitt.) and white(Tuber magnatum Pico) truffles. Archives of Microbiology 184, 187–193.

Cavagnaro, T.R., Smith, F.A., Smith, S.E., 2004. Interactions between arbuscular mycorrhi-zal fungi and amycorrhiza-defectivemutant tomato: does a noninfective fungus alter

379X.-G. Sun, M. Tang / South African Journal of Botany 88 (2013) 373–379

the ability of an infective fungus to colonise the roots — and vice versa? NewPhytologist 164, 485–491.

Copetta, A., Lingua, G., Graziella, B., 2006. Effects of three AM fungi on growth, distributionof glandular hairs, and essential oil production in O. basilicum L. var. Genovese.Mycorrhiza 16, 485–494.

Czarnota, M.A., Rimando, A.M., Weston, L.A., 2003. Evaluation of root exudates of sevensorghum accessions. Journal of Chemical Ecology 29, 2073–2083.

Dayan, F.E., Rimando, A.M., Pan, Z., Baerson, S.R., Gimsing, A.L., Duke, S.O., 2010.Sorgoleone. Phytochemistry 71, 1032–1039.

Dib, M.E.A., Djabou, N., Desjobert, J.M., Allali, H., Tabti, B., Muselli, A., Costa, J., 2010.Characterization of volatile compounds of Daucus crinitus Desf. Headspace solidphase microextraction as alternative technique to hydrodistillation. ChemistryCentral Journal 4, 16.

DiCosmo, F., Misawa, M., 1985. Eliciting secondary metabolites in plant cell cultures.Trends in Biotechnology 3, 318–322.

Foehse, D., Jungk, A., 1983. Influence of phosphate and nitrate supply on root hair forma-tion of rape, spinach and tomato plants. Plant and Soil 74, 359–368.

Gao, L.L., 2002. Control of arbuscular mycorrhizal colonization: studies of a mycorrhiza-defective tomato mutant. PhD thesis The University of Adelaide, Adelaide, Australia.

Gerdemann, J.W., Nicolson, T.H., 1963. Spores of mycorrhizal Endogone species extractedfrom soil by wet sieving and decanting. Transactions of the British MycologicalSociety 46, 235–244.

Giovannetti, M., Mosse, B., 1980. An evaluation of techiques for measuring vesiculararbuscular mycorrhizal infection in roots. New Phytologist 84, 489–500.

Giri, B., Mukerji, K.G., 2004. Mycorrhizal inoculant alleviates salt stressing Sesbaniaaegyptiaca and Sesbania grandiflora under field conditions: evidence for reduced sodi-um and improved magnesium uptake. Mycorrhiza 14, 307–312.

Gupta, M.L., Prasad, A., Ram, M., Kumar, S., 2002. Effect of the vesicular-arbuscular mycor-rhizal (VAM) fungus Glomus fasciculatum on the essential oil yield related charactersand nutrient acquisition in the crops of different cultivars of menthol mint (Menthaarvensis) under field conditions. Bioresource Technology 81, 77–79.

ICRISAT, 2004. Sorghum, a crop of substance. International Crops Research Institute forthe Semi-arid Tropics. Patancheru, Andhra Pradesh, India.

Isopi, R., Fabbri, P., Puppi, D.M.G., 1995. Dual inoculation of Sorghum bicolor (L.) Moenchssp. bicolor with vesicular arbuscular mycorrhizas and Acetobacter diazotrophicus.Symbiosis 18, 43–55.

Kapoor, R., Giri, B., Mukerji, K.G., 2004. Improved growth and essential oil yield andquality in Foeniculum vulgare Mill. on mycorrhizal inoculation supplemented withP-fertilizer. Bioresource Technology 93, 307–311.

Kesselmeier, J., Staudt, M., 1999. Biogenic volatile organic compounds (VOC): an overviewon emission, physiology and ecology. Journal of Atmospheric Chemistry 33, 23–88.

Khaosaad, T., Vierheilig, H., Nell, M., Zitterl-Eglseer, K., Novak, J., 2006. Arbuscular mycor-rhiza alter the concentration of essential oils in oregano (Origanum sp., Lamiaceae).Mycorrhiza 16, 443–446.

Kleinheinz, G.T., Bagley, S.T., St. John, W.P., Rughani, J.R., McGinnis, G.D., 1999. Character-ization of alpha-pinene-degrading microorganisms and application to a bench-scalebiofiltration system for VOC degradation. Archives of Environmental Contaminationand Toxicology 37, 151–157.

Kothari, S.K., Marschner, H., George, E., 1990. Effect of VA mycorrhizal fungi and rhizo-sphere microorganisms on root and shoot morphology, growth and water relationsin maize. New Phytologist 116, 303–311.

Krupa, S., Andersson, J., Marx, D.H., 1973. Studies on ectomycorrhizae of pine. IV. Volatileorganic compounds in mycorrhizal and nonmycorrhizal root systems of Pinusechinata Mill. European Journal of Forest Pathology (4), 194–200.

Krupa, S., Fries, N., 1971. Studies on ectomycorrhizae of pine. I. Production of volatileorganic compounds. Canadian Journal of Botany-Revue Canadienne de Botanique49, 1425–1431.

Kuske, M., Romain, A., Nicolas, J., 2005. Can an electronic nose detect fungi in indoorenvironments? Building and Environment 40, 824–831.

Laothawornkitkul, J., Taylor, J.E., Paul, N.D., Hewitt, C.N., 2009. Biogenic volatile organiccompounds in the Earth system. New Phytologist 183, 27–51.

Larose, G., Chênevert, R., Moutoglis, P., Gagné, S., Piché, Y., Vierheilig, H., 2002. Flavonoidlevels in roots of Medicago sativa are modulated by the developmental stage of thesymbiosis and the root colonizing arbuscular mycorrhizal fungus. Journal of PlantPhysiology 159, 1329–1339.

Larsen, T.O., Frisvad, J.C., 1995. Chemosystematics of Penicillium based on profiles ofvolatile metabolites. Mycological Research 99, 1167–1174.

Leff, J.W., Fierer, N., 2008. Volatile organic compound (VOC) emissions from soil and littersamples. Soil Biology and Biochemistry 40, 1629–1636.

Leitner, M., Kaiser, R., Hause, B., Boland, W., Mithöfer, A., 2010. Does mycorrhization influenceherbivore-induced volatile emission in Medicago truncatula. Mycorrhiza 20, 89–101.

Lin, C., Owen, S.M., Peñuelas, J., 2007. Volatile organic compounds in the roots andrhizosphere of Pinus spp. Soil Biology and Biochemistry 39, 951–960.

Louarn, J., Carbonne, F., Delavault, P., Bécard, G., Rochange, S., 2012. Reduced germinationof Orobanche cumana seeds in the presence of arbuscular mycorrhizal fungi or theirexudates. PLoS One 7 (11), e49273. http://dx.doi.org/10.1371/journal.pone.0049273.

Manjunath, A., Habte, M., 1991. Root morphological characteristics of host species havingdistinct mycorrhizal dependency. Canadian Journal of Botany-Revue Canadienne deBotanique 69, 671–676.

Morandi, D., Bailey, J.A., Gianinazzi-Pearson, V., 1984. Isoflavonoid accumulation in soy-bean roots infected with vesicular-arbuscular mycorrhizal fungi. Physiological PlantPathology 24, 357–364.

Mucciarelli, M., Camusso, W., Maffei, M., Panicco, P., Bicchi, C., 2007. Volatile terpenoids ofendophyte-free and infected peppermint (Mentha piperita L.): chemical partitioningof a symbiosis. Microbial Ecology 54, 685–696.

Muselli, A., Pau, M., Desjobert, J.M., Foddai, M., Usai, M., Costa, J., 2009. Volatile constitu-ents of Achillea ligustica All. by HS-SPME/GC/GC-MS. Comparison with essential oilsobtained by hydrodistillation from Corsica and Sardinia. Chromatographia 69,575–585.

Nardi, S., Concheri, G., Pizzeghello, D., Sturaro, A., Rella, R., Parvoli, G., 2000. Soil organicmatter mobilization by root exudates. Chemosphere 41, 653–658.

Owen, S.M., Clark, S., Pompe, M., Semple, K.T., 2007. Biogenic volatile organic compoundsas potential carbon sources for microbial communities in soil from the rhizosphere ofPopulus tremula. FEMS Microbiology Letters 268, 34–39.

Paolini, J., Nasica, E., Desjobert, J.M., Muselli, A., Bernardini, A.F., Costa, J., 2007. Analysis ofvolatile constituents isolated by hydrodistillation and headspace solid-phasemicroextraction from Adenostyle briquetii Gamisans. Phytochemical Analysis 19,266–276.

Phillips, J.M., Hayman, D.S., 1970. Improved procedures for clearing roots and stainingparasitic and vesicular-arbuscular mycorrhizal fungi for rapid assessment of infection.Transactions of the British Mycological Society 55, 158–161.

Prasad, A., Kumar, S., Khaliq, A., Pandey, A., 2011. Heavy metals and arbuscular mycorrhi-zal (AM) fungi can alter the yield and chemical composition of volatile oil of sweetbasil (Ocimum basilicum L.). Biology and Fertility of Soils 47, 853–861.

Price, N.S., Roncadori, R.W., Hussey, R.S., 1989. Cotton root growth as influenced by phos-phorus nutrition and vesicular-arbuscular mycorrhizas. New Phytologist 111, 61–66.

Raju, P.S., Clark, R.B., Ellis, J.R., Maranville, J.W., 1990. Effects of species of VA-mycorrhizalfungi on growth and mineral uptake of sorghum at different temperatures. Plant andSoil 121, 165–170.

Rapparini, F., Llusià, J., Peñuelas, J., 2008. Effects of arbuscular mycorrhizal (AM) coloniza-tion on terpene emission and content of Artemisia annua L. Plant Biology 10, 108–122.

Röderstein, M., Hertel, D., Leuschner, C., 2005. Above- and below-ground litter productionin three tropical montane forests in southern Ecuador. Journal of Tropical Ecology 21,483–492.

Sawers, R.J.H., Gutjahr, C., Paszkowski, U., 2008. Cereal mycorrhiza: an ancient symbiosisin modern agriculture. Trends in Plant Science 13, 93–97.

Schnürer, J., Olsson, J., Börjesson, T., 1999. Fungal volatiles as indicators of food and feedsspoilage. Fungal Genetics and Biology 27, 209–217.

Schweiger, P.F., Robson, A.D., Barrow, N., 1995. Root hair length determines beneficial ef-fect of a Glomus species on shoot growth of some pasture species. New Phytologist131, 247–254.

Shi, S., Richardson, A.E., O'Callaghan, M., DeAngelis, K.M., Jones, E.E., Stewart, A., Firestone,M.K., Condron, L.M., 2011. Effects of selected root exudate components on soil bacte-rial communities. FEMS Microbiology Ecology 77, 600–610.

Siqueira, J.O., Saggin-Júnior, O.J., 2001. Dependency on arbuscular mycorrhizalfungi and responsiveness of some Brazilian native woody species. Mycorrhiza11, 245–255.

Smith, S.E., Read, D.J., 1997. Mycorrhizal Symbiosis, 2nd ed. Academic Press, San Diego.Strack, D., Fester, T., Hause, B., Schliemann, W., Walter, M.H., 2003. Arbuscular mycorrhi-

za: biological, chemical, and molecular aspects. Journal of Chemical Ecology 29,1955–1979.

Tennant, D., 1975. A test of modified line intersect method estimating root length. Journalof Ecology 63, 995–1001.

Viles, A.L., Reese, R.N., 1996. Allelopathic potential of Echinacea angustifolia D.C. Environ-mental and Experimental Botany 36, 39–43.

Weissteiner, S., Schütz, S., 2006. Are different volatile pattern influencing host plantchoice of belowground living insects. Mitteilungen der Deutschen Gesellschaft fürAllgemeine und Angewandte Entomologie 15, 51–55.

Wenke, K., Kai, M., Piechulla, B., 2010. Belowground volatiles facilitate interactionsbetween plant roots and soil organisms. Planta 231, 499–506.

Wu, Q.S., Zou, Y.N., Liu, W., Ye, X.E., Zai, H.F., Zhao, L.J., 2010. Alleviation of salt stress incitrus seedlings inoculated with mycorrhiza: changes in leaf antioxidant defensesystems. Plant Soil and Environment 56, 470–475.

Zangaro, W., Nishidate, F.R., Camargo, F.R.S., Romagnoli, G.G., Vandresen, J., 2005. Rela-tionships among arbuscular mycorrhizas, root morphology and seedling growth oftropical native woody species in southern Brazil. Journal of Tropical Ecology 21,529–540.

Zakir, H.A.K.M., Subbarao, G.V., Pearse, S.J., Gopalakrishnan, S., Ito, O., Ishikawa, T.,Kawano, N., Nakahara, K., Yoshihashi, T., Ono, H., Yoshida, M., 2008. Detection, isola-tion and characterization of a root-exuded compound, methyl 3-(4-hydroxyphenyl)propionate, responsible for biological nitrification inhibition by sorghum (Sorghumbicolor). New Phytologist 180, 442–451.