• benzoxazinoids;
  • community assembly;
  • fungal colonization;
  • fungal endophytes;
  • Fusarium;
  • plant defense compounds;
  • selective breeding;
  • Zea mays (maize)


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • • 
    Plant defense compounds are common stressors encountered by endophytes. Fungi readily evolve tolerance to these compounds, yet few studies have addressed the influence of intraspecific variation in defense compound production on endophyte colonization. We compared the influence of defense compound production on the composition of fungal endophyte communities in replicated field experiments.
  • • 
    Maize (Zea mays) produces benzoxazinoids (BXs), compounds with antifungal byproducts persistent in the environment. Fungi were isolated from leaf and root tissue of two maize genotypes that produce BXs, and a natural mutant that does not. Isolates representing the species recovered were tested for tolerance to 2-benzoxazolinone (BOA), a toxic BX byproduct.
  • • 
    In seedling roots and mature leaves, the community proportion with low BOA tolerance was significantly greater in BX nonproducers than producers. Mean isolation frequency of Fusarium species was up to 35 times higher in mature leaves of BX producers than nonproducers.
  • • 
    Fungal species with relatively high tolerance to BOA are more abundant in BX producing than BX nonproducing maize. Production of BXs may increase colonization by Fusarium species in maize, including agents of animal toxicosis and yield-reducing disease in maize. Overall, results indicate that production of defense compounds can significantly alter endophyte community assembly.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Communities of nonmycorrhizal fungal endophytes occupy most plants (Schulz & Boyle, 2005; Arnold, 2007). They can directly affect plant fitness and indirectly influence surrounding plant and arthropod communities (Arnold et al., 2003; Finkes et al., 2006; Rudgers et al., 2007). Fungal endophytes are transmitted in seed (vertically), or from plant to plant via fungal propagules (horizontally). An excellent example of obligate vertical transmission is the well-studied ascomycete, Neotyphodium, which forms associations with temperate grasses mediated by molecular signaling dependent on the compatibility of endophyte and host species (reviewed in Schardl et al., 2004). Phylogenetic and molecular genetic evidence indicate co-evolution of some Neotyphodium species with their plant hosts (for reviews see Clay & Schardl, 2002; Schardl et al., 2004). By contrast, the majority of fungal endophytes are transmitted horizontally or facultatively by seed. These fungi can form diverse species assemblages within plants, often at high density (Arnold, 2007). There are still many open questions about the mechanisms that influence colonization and community structure of horizontally transmitted endophytes, but certain abiotic environmental factors and plant defense compounds are known to be important.

Abiotic factors such as fertilizer application can influence the community structure of fungal endophytes in maize (Zea mays) (Seghers et al., 2004), and water activity and temperature can significantly influence the outcome of interactions between maize endophytes (Marin et al., 1998). Plant defense compounds can also influence interactions between the plant and its surrounding community of fungi, bacteria, insects, and plants (Niemeyer & Perez, 1995). All plant species studied to date produce defense compounds (Hashimoto & Shudo, 1996; Dixon, 2001). Common tolerance strategies adopted by fungi include activation of membrane transporters that pump toxicants out of cells, and enzymatic detoxification (VanEtten et al., 2001). Detoxification of host compounds can be a virulence factor among pathogens (VanEtten et al., 2001).

Ability to detoxify may also increase competitiveness of root endophytes (Carter et al., 1999). Increase in endophyte competitiveness among tolerant species is likely the result of higher growth rates in the presence of host toxins than less well-adapted species (Arnold et al., 2003; Nicol et al., 2003; Saunders & Kohn, 2008). This fitness benefit would be expected to influence endophyte community assembly. In a comparison of fungal communities in Avena sativa (oat) and Triticum sp. (wheat) roots, Carter et al. (1999) found that the majority of oat-derived isolates tolerated the oat defense compound, avenacin A-1. The spatial scale of influence may increase when plant defense compounds are secreted in soil, affecting colonization by mycorrhizal and soil fungi (Stinson et al., 2006; Broeckling et al., 2008).

Commercial maize has been selectively bred to produce high quantities of defense compounds, the benzoxazinoids (BXs). Several BX byproducts are toxic to microbes, insects and plants (Barry & Darrah, 1991; Niemeyer & Perez, 1995; Hashimoto & Shudo, 1996). The primary BXs found in maize are 2,4-dihydroxy-7-methoxy-2H-1,4-benzoxazin-3-one (DIMBOA) and 2,4-dihydroxy-2H-1,4-benzoxazin-3-one (DIBOA). These compounds reside in the cell vacuole as biologically inactive beta-glucosides. They are enzymatically converted to toxic benzoxazinoids upon cell disruption (Hashimoto & Shudo, 1996), ultimately degrading to the biologically active and stable benzoxazolinones, 6-methoxy-2-benzoxazolinone (MBOA) and 2-benzoxazolinone (BOA). These are formed systemically and secreted from root tissue (Park et al., 2004). They can be produced both constitutively and in response to tissue damage (Cambier et al., 2000; Oikawa et al., 2004).

All commercial maize genotypes produce BXs. Concentrations of DIMBOA have been recorded ranging from 2.86 to 12.90 mm kg−1 FW (Zuniga et al., 1983; Cambier et al., 2000). The concentrations of BXs and their toxic byproducts in plant tissue can vary with plant age, tissue, genotype and environment (Zuniga et al., 1983; Richardson & Bacon, 1993; Cambier et al., 2000). Uptake of BX byproducts from soil by plants has been reported (Chiapusio et al., 2004). Accumulation of BX byproducts in soil is expected to depend on all of these factors, as well as the interactions between members of the microbial community (Bacon et al., 2007).

It has been proposed that the ability to detoxify benzoxazolinones enhances colonization success in maize (Glenn et al., 2001). Some detoxifying species, particularly Fusarium verticillioides, Fusarium subglutinans, Fusarium proliferatum and Fusarium graminearum cause disease in corn, but are also common endophytes. As endophytes, they can lead to the asymptomatic contamination of grain, in some cases producing toxins that cause mycotoxicosis in animals, and are suspected risk factors for cancers and other human health problems (Ueno et al., 1997; Marasas, 2001; Marasas et al., 2004). Contamination of maize grain by Fusarium is estimated to cause millions of dollars of economic loss annually in the USA (Wu, 2007; Wu & Munkvold, 2008). Breeding programs aimed at deterring infection of maize by Fusarium using native resistance mechanisms have been largely unsuccessful (Munkvold, 2003).

Here, we investigated the influence of BX production on the assembly and composition of fungal endophyte communities in maize. Specifically, three hypotheses were tested: (1) BX production increases the proportion of fungi tolerant to the toxic BX byproduct, BOA, in endophyte communities; (2) BX production increases the incidence of Fusarium in maize; and (3) BX producing genotypes harbor endophyte communities that are less diverse than genotypes that do not produce BXs. To test these hypotheses, fungal endophyte communities from three maize genotypes were compared: one was a natural mutant deficient in BXs (BX–) and the other two were commercial genotypes that produce BXs (BX+). The objective was to observe communities in BX– versus BX+ maize. Common species were tested for BOA tolerance in vitro. Community structure was consistent with expectations for BX influence in seedling roots and mature leaves. BX production significantly increased the frequency of Fusarium in leaves of mature plants.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Maize genotypes

Each location was planted with three maize (Z. mays L.) lines W22 and B37, two genotypes both producing BXs and commonly used to produce commercial hybrids, and bxbx, the only recorded natural mutant lacking the ability to produce BXs (Hamilton, 1962). The three genotypes are Yellow Dent maize, characterized by a genetic background of flint and floury maize and common kernel phenotypes (Smith et al., 2004). Because of the nonlinear dynamics of BX concentration in plants and soil, we chose to assess the influence of BX production, rather than concentration, on endophyte communities.

Study site and collection times

Maize was planted in two Ontario locations approx. 123 km apart: Ridgetown, with a history of soybean crops from 2001 to 2004, and Harrow, with a history of maize from 1999 to 2004. Within the plot, each genotype was planted in 12 rows, one genotype per row, as a row intercropping design. Assignment of rows within the field was random. The Ridgetown plot was surrounded by soybean, which does not produce BXs, and the Harrow plot by BX-producing maize. Planting was on June 12, 2005, with sampling carried out 2 wk and 9 wk subsequently. Whole plants were collected in paper bags and stored at 4°C.

Isolation of fungi from plant tissue

Plants were rinsed with distilled water and surface dried at room temperature. From each plant, eight healthy 1.0 × 2.0 cm segments were taken 0.5 cm from the midrib and above the leaf collar of the second and third leaf blades (2-wk-old plants) or fourth and fifth blades (9-wk-old plants). Plants collected at 2 wk had three emergent leaves (V3 growth stage). Plants collected at 9 wk were entering the R3 stage of growth, midway through kernel development and approx. 3 wk before physiological maturity (Ritchie et al., 1993). Leaf senescence begins at physiological maturity. Our goal was to analyse healthy plant material; tissue was therefore collected before maturity. Healthy root segments 2.0 cm long and 0.2–0.3 cm diameter were taken from the radicle and lateral seminal roots. Tissue segments were surface-sterilized first in 70% ethanol (2 min), then in 0.53% NaOCl (2 min) and finally in sterile double-distilled water (2 min).

Tissue segments were incubated on two growth media. A general, neutral medium, potato dextrose agar (PDA; Difco, Detroit, MI, USA) was used to capture a relatively broad snapshot of the fungal community. A selective medium amended with 1.00 mg ml−1 BOA (Glenn et al., 2001) was used to determine the mean number of Fusarium colonies per plant (subsequently termed, abundance), isolated from 9-wk-old plants; this time-point was initially selected as an indicator of potential infestation of crop debris. Potato dextrose agar is expected to be more favorable than BOA medium to fungi that are BOA-sensitive.

Four leaf and four root segments were incubated on PDA amended with antibiotics (1.00 g l−1 streptomycin sulfate and 0.25 g l−1 neomycin sulfate), and four segments of each tissue were incubated on BOA medium. Segments from a total of 144 plants were plated on PDA (12 plants, 1 plant per row × 2 locations × 2 times: 2 wk and 9 wk post-planting × 3 maize genotypes); 9–12 plants per treatment were analyzed. A total of 144 plants were plated on BOA medium (24 plants, 2 plants per row × 2 locations × 1 time: 9 wk post planting × 3 maize genotypes); 16–24 plants (8–12 rows) per treatment were analysed. All tissues were surface sterilized and plated within 96 h of collection. Plates were incubated at room temperature under a 12-h light–12-h dark regime.

Effectiveness of the surface sterilization procedure was tested on PDA by plating out 500 µl of the rinse water from the sterilization procedure, and independently by using the tissue imprint method described by Schulz & Boyle (2005). Approximately 25% of tissue samples processed were tested, and no surface contaminants were detected.

Identification of fungal isolates

Fungi emerging from plant tissue were isolated and established in axenic culture. Sporulating cultures were identified morphologically. Each Fusarium isolate was established in axenic culture from a single spore and identified morphologically when diagnostic characters were evident, or using DNA sequence data when such characters were ambiguous or absent. Isolates of Fusarium were grown on Carnation Leaf Agar (Leslie et al., 2006) and PDA for morphological identification (Summerell et al., 2003; Leslie et al., 2006). For each Fusarium species that was identified morphologically, the identity of a subset of isolates was verified with DNA sequence data. Nonsporulating fungi were grouped into morphotypes. To maintain consistency, all morphotyping was done by one of the authors (M.S.). Morphotypes were then confirmed in blind tests by the other author (L.M.K.).

Isolates obtained on PDA were used to characterize 24 fungal endophyte communities (2 locations × 2 times × 2 tissue types × 3 maize genotypes). All isolates obtained from PDA were morphotyped, and isolates recovered more than three times were identified taxonomically. Of the isolates obtained on BOA medium, 57 isolates, 18 from leaf tissue and 39 from root tissue, were identified using molecular sequence data.

DNA isolation, polymerase chain reaction (PCR) amplification and sequencing

Total genomic DNA was isolated using the DNeasy Plant Minikit (Qiagen, Mississauga, ON, Canada). For identification of morphotypes, the nuclear ribosomal internal transcribed spacer region (ITS) was amplified by PCR using primers ITS-1F and NLB-3 (c. 700 bp) (Gardes & Bruns, 1993; Martin & Rygiewicz, 2005). For identification of Fusarium isolates, the translation elongation factor 1-alpha (TEF) gene was amplified using primers TEF-1 and TEF-2 (c. 700 bp) (Geiser et al., 2004).

The final volumes of the PCR mixture (50-µl volume) were 9.75 µl glass-distilled H2O, 5.00 µl 10× PCR buffer, 5.00 µl deoxynucleoside triphosphates, 0.50 µl of each primer (50 mm), 4.00 µl of magnesium chloride, 0.25 µl Amplitaq DNA polymerase (Perkin-Elmer, Norwalk, CT, USA), and 25.00 µl of a 50-fold dilution of genomic DNA. A Perkin-Elmer GeneAmp System 9600 or 9700 thermocycler was used to amplify PCR product. The PCR program used was: 1 cycle of 95°C for 8 min, 35 cycles of 95°C for 1 min, 55°C for 30 s and 72°C for 1 min, and 1 cycle of 72°C for 10 min. Sequencing was performed at the Genetic Analysis Core Facility (USDA-ARS-ERRC, Wyndmoor, PA, USA) using an ABI 3730 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA).

Contigs were assembled and edited in sequencher 4.6 (Gene Codes Corporation, Ann Arbor, MI, USA), and blast searches of the NCBI GenBank database were conducted for tentative identification. A 99% sequence match to a sequence of known origin in the database was counted as a correct species identification. DNA sequences have been deposited in GenBank as FJ496215-FJ496332.

Assignment of isolates to BOA tolerance class

Isolates were assigned to a tolerance class corresponding to the highest concentration of BOA that supported growth (Table 1). The PDA was amended with BOA (stock solution of 100 mg ml−1 in anhydrous ethanol) in each of the following concentrations: 0.00, 0.25, 0.50, 0.75, 1.00, 1.10, 1.20 mg ml−1. From a total of 24 species/morphotypes, 61 isolates were assigned to tolerance classes. Two isolates per species/morphotype were assigned when available. All Fusarium species and species/morphotypes recovered more than three times were included. Strains were incubated in quadrant Petri dishes for 14 d in the dark at 22°C, and scored for growth or no growth.

Table 1.  2-Benzoxazolinone (BOA) tolerance thresholds of fungal endophyte species/morphotypes isolated from maize (Zea mays) (bxbx, B37, W22 genotypes)
Tolerance classBOA tolerance thresholdSpecies/morphotypeOrderNumber of isolates in tolerance classTotal number of isolates tested
  1. Names in bold type indicate species with isolates in two tolerance classes.

I0.25 mg ml−1Alternaria alternataPleosporales4 4
Morphotype no. 14Sordariales4 4
Periconia macrospinosaAnamorphic ascomycete3 3
II0.50 mg ml−1Cladosporium sp.Capnodiales2 2
Epicoccum nigrumAnamorphic ascomycete2 2
Fusarium acuminatumHypocreales1 1
Fusarium incarnatum-equiseti species complexHypocreales5 5
Fusarium oxysporum species complexHypocreales613
Fusarium proliferatumHypocreales1 1
Fusarium redolensHypocreales1 1
Fusarium sporotrichioidesHypocreales1 1
Fusarium tricinctumHypocreales1 1
Penicillium sp.Eurotiales3 3
Periconia circinatumAnamorphic ascomycete3 3
III0.75 mg ml−1Fusarium oxysporum species complexHypocreales713
Fusarium solani species complexHypocreales2 2
Fusarium verticillioidesHypocreales1 1
Morphotype no. 22Anamorphic ascomycete2 2
Morphotype no. 51Anamorphic ascomycete1 1
Nigrospora oryzaeTrichosphaeriales2 4
Trichocladium sp.Sordariales2 4
Trichoderma sp.Hypocreales2 2
IV1.00 mg ml−1Fusarium culmorumHypocreales4 4
Fusarium graminearumHypocreales1 1
Trichocladium sp.Sordariales2 4
Nigrospora oryzaeTrichosphaeriales2 4
V.1.10 mg ml−1Rhizopus sp.Mucorales3 3

Statistical analyses

For all statistical analyses, each row was considered the unit of replication. For the PDA assays, one plant per row was sampled. For the BOA assay, two plants per row were sampled. The average number of isolates per plant, averaged across each row, was considered a replicate.

Abundance of Fusarium in 9-wk-old plants  One-way anovas were followed by Tukey–Kramer Honestly Significant Difference (HSD) tests for pairwise comparisons between the BX– and BX+ genotypes. The threshold for statistical significance was P ≤ 0.05. Analyses were conducted using JMP in (version 5.1; SAS Institute Inc., Cary, NC, USA).

Diversity and similarity of fungal endophyte communities  As described earlier in the section Identification of Fungal Isolates, 24 fungal endophyte communities were characterized from isolates obtained on PDA. Diversity was measured using the Simpson's inverse diversity index (D) and Fisher's alpha (α) (Magurran, 2004). Similarity was determined using the Jaccard's (Magurran, 2004) and Morisita–Horn indices (Chao et al., 2005). All were calculated using estimates (Colwell, 2000).

BOA tolerance of fungal endophyte communities  The distribution of isolates in BOA tolerance classes was determined for each community. To test for a difference between the distribution of isolates in BOA tolerance classes in BX+ and BX– plants, a likelihood ratio test was conducted using the JMPin statistical analysis software package. When necessary, results were then adjusted for small expected values using the formula (Gotelli & Ellison, 2004):

  • χ2adjusted = χ2/qmin
  • image(Eqn 1)

(n is the number of rows, m is the number of columns, N is the total sample size, v is the degrees of freedom and Yi,j is the frequency of observations in row i, column j).

When global likelihood ratio tests detected a significant difference between the distribution of isolates in BX+ and BX– plants, further tests for a difference between the proportion of fungi in tolerance classes I and II in each community were conducted. Follow-up likelihood ratio tests were conducted on tolerance classes I and II, because a mean of 83% of isolates in each community were in these tolerance classes.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Relative abundance and diversity of endophytes from BX– and BX+ maize genotypes in 2- and 9-wk-old plants

Infection density, measured as the proportion of tissue segments yielding isolates, ranged from 54 to 100%, and was higher in August than in June in leaf and root tissue of all maize genotypes (see the Supporting Information, Fig. S1). Of 1495 fungal isolates obtained, 1018 were isolated on PDA, and 477 on BOA. The genus Fusarium includes several species complexes. Consequently, the Fusarium oxysporum species complex, the Fusarium incarnatum-equiseti species complex, and the Fusarium solani species complex were each treated as single species in the analyses. A total of 43 species/morphotypes were identified. All of the 13 species BOA were also isolated on PDA. On BOA medium, roots yielded approximately twice the number of isolates as leaves.

Species richness and presence/absence differed between communities from leaf and root tissue on PDA. Leaf and root tissue had 31% (13/42) species/morphotypes in common, root communities had 60% (25/42) species/morphotypes present that were not isolated from leaves, and leaf communities had 9% (4/42) species/morphotypes that were not isolated from roots.

The diversity of endophyte communities was assessed using isolates obtained on PDA. Values of Simpson's D (inverse) and Fisher's α were similar across communities from all three maize genotypes, with one exception: in communities isolated from seedling roots in Harrow, BX– plants had higher diversity than BX+ plants (Table S1). In all treatments (2 locations × 2 collection times × 2 tissue types), diversity of communities was lower in August than in June.

Based on the similarity indices, the expectation that endophyte communities from BX+ plants would be more similar to one another than either were to communities from BX– plants was observed in four of eight treatments (Fig. S2). Overall, diversity and similarity indices detected a difference between BX+ and BX− communities in Harrow more often than Ridgetown.

BOA tolerance levels of endophytic fungi

The BOA tolerance threshold of isolates ranged from 0.25 to 1.10% concentration. No isolates were able to grow at 1.20%. Isolates were categorized in five BOA tolerance classes (Table 1). Twenty-one of 24 species/morphotypes had no intraspecific variation in tolerance level. The remaining three species, Trichocladium sp., Nigrospora oryzae, and members of the Fusarium oxysporum species complex had isolates in two adjacent tolerance classes.

Partitioning of endophyte communities by BOA tolerance level

The distribution of isolates in BOA tolerance classes was determined for each community characterized from PDA (Fig. 1). Isolates of the species/morphotypes that had intraspecific variation in tolerance level were equally distributed in the adjacent tolerance classes for the analyses.


Figure 1. Distribution of fungal endophyte community members isolated on potato dextrose agar (PDA) into Tolerance classes I–V. Tolerance class (TC) I = 0.25% 2-benzoxazolinone (BOA) tolerance threshold, TC II = 0.50% threshold, TC III = 0.75% threshold, TC IV = 1.00% threshold, TC V = 1.10% threshold. Leaf and root endophyte communities from bxbx (BX–, circles), W22 (BX+, squares) and B37 (BX+, triangles) were characterized in 2-wk-old and 9-wk-old maize (Zea mays) plants. Bold type indicates treatments indicated by global χ2 tests to have communities distributed differently among TCs in BX+ and BX– plants. Asterisks show where follow-up χ2 square tests indicate that the proportion of isolates was significantly different between BX– and both BX+ genotypes. The dashed lines emphasize trends; they do not indicate a series of measurements over time.

A significant difference in the tolerance class distribution was seen in four of the eight treatments, in which BX– plants had a significantly greater proportion of isolates in tolerance class I than did BX+ plants (Table 2, Fig. 1). These four treatments were communities from seedling roots in Ridgetown (bxbx by W22: χ2 = 7.64, P = 0.0219; bxbx by B37: χ2 = 23.71, P = <0.00001) and in Harrow (bxbx by W22: χ2 = 13.64, P = 0.0041; bxbx by B37: χ2 = 11.83, P = 0.0027), and communities from 9-wk-old leaves in Ridgetown (bxbx by W22: χ2 = 10.95, P = 0.012; bxbx by B37: χ2 = 10.21, P = 0.0169) and in Harrow (bxbx by W22: χ2 = 12.96, P = 0.0047; bxbx by B37: χ2 = 5.94, P = 0.0148). In communities from seedling roots in Harrow, BX+ plants had a significantly higher proportion of isolates in tolerance class II compared with BX− plants (Table 2). In seedling roots, BX− plants had dominant species in tolerance class I (Alternaria alternata in Harrow and Periconia macrospinosa in Ridgetown), while BX+ plants had dominant species in classes II and III (members of the Fusarium equiseti-incarnatum species complex and the F. oxysporum species complex in both locations). In communities from 9-wk-old leaves, A. alternata was dominant in the three plant genotypes, but was isolated less frequently in BX– than in BX+ plants.

Table 2.  Results of χ2 tests for a difference between proportion of isolates in Tolerance Classes I (0.25% 2-benzoxazolinone (BOA) tolerance threshold) and II (0.50% BOA tolerance threshold) in endophyte communities from maize (Zea mays) bxbx (BX– genotype), W22 and B37 (BX+ genotypes)
TissuePlant ageLocationTolerance classGenotype comparisonχ2P-value
  • *

    Tests were conducted following global χ2 tests that detected significant difference between communities from BX+ and BX– plants in 2-wk-old roots and 9-wk-old leaves. P-values < 0.05 are indicated by bold type.

Leaf9 wkHarrowIbxbx and W2212.720.0004
bxbx and B3711.420.0007
IIbxbx and W225.170.023
bxbx and B375.040.0248
RidgetownIbxbx and W228.380.0038
bxbx and B3723.43< 0.0001
IIbxbx and W220.430.513
bxbx and B3710.780.001
Root2 wkHarrowIbxbx and W229.290.0023
bxbx and B375.240.022
IIbxbx and W226.640.01
bxbx and B370.840.3584
RidgetownIbxbx and W225.250.022
bxbx and B378.520.0035
IIbxbx and W221.990.1586
bxbx and B376.710.0096

Abundance of Fusarium in 9-wk-old plants

To compare the difference in relative abundance of Fusarium isolates per plant, the mean number of colonies isolated on BOA medium from the BX+ genotypes was compared with that from the BX– plants. A significant difference was detected in leaves (Harrow F = 7.03, P = 0.0046; Ridgetown F = 9.12, P = 0.0014), but not in roots (Harrow F = 0.29, P = 0.7501; Ridgetown F = 0.08, P = 0.921). In leaf tissue, averaging values for the two BX+ genotypes, BOA-tolerant fungi were 14 times (Harrow) to 35 times (Ridgetown) more abundant in BX+ genotypes than in the BX− genotype (Fig. 2). From the isolates obtained on BOA medium that were identified, 96.5% were Fusarium species (Table 3. A total of nine Fusarium species, one isolate of Nigrospora oryzae and one isolate of Trichocladium sp. were identified. Of these species, six were isolated from leaf and root tissue, four from roots but not from leaves, and one only from leaf tissue. The most abundant species were members of the F. oxysporum species complex and the F. equiseti-incarnatum species complex.


Figure 2. Mean number of isolates per plant obtained on 2-benzoxazolinone (BOA)-medium from leaf (a) and root (b) tissue of 9-wk-old maize (Zea mays) plants in Harrow and Ridgetown, Ontario, Canada. Based on identification of a sample of these isolates, approx. 96.5% of isolates obtained on BOA medium were Fusarium species. Tukey–Kramer HSD tests were conducted to compare mean number of isolates per plant obtained from bxbx (BX–), W22 (BX+), and B37 (BX+) genotypes. Analysis of leaf and root tissue was conducted separately. The same letter above two columns indicates no significant difference between means. Vertical bars, ± SE.

Table 3.  Identity of a subset of isolates obtained from 9-wk-old maize (Zea mays) tissue plated on 2-benzoxazolinone (BOA) medium
TaxonTotal identifiedTotal from leavesTotal from roots
Fusarium oxysporum species complex27522
Fusarium incarnatum-equiseti species complex 84 4
Fusarium subglutinans 43 1
Fusarium graminearum 41 3
Fusarium culmorum 30 3
Fusarium sporotrichioides 43 1
Fusarium proliferatum 20 2
Fusarium verticillioides 21 1
Fusarium solani 10 1
Nigrospora oryzae 11 0
Trichocladium sp. 10 1


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Presence of BXs influences endophyte community structure

These results suggest that BX production contributes to the structuring of endophyte communities in seedling roots and 9-wk-old leaves (Fig. 1). The differences over time in these tissues could arise from temporal changes in BX byproduct concentrations and allocation within tissue. Concentrations of constitutive DIMBOA and related compounds have been demonstrated to decrease with shoot and root age (Cambier et al., 2000). This would explain why there were more BOA-tolerant fungi in BX+ than in BX− roots in the seedling, but not in 9-wk-old plants.

Fungal-mediated changes in MBOA concentration may be a factor in the relatively high abundance of BOA sensitive isolates in leaves of 9-wk-old BX− plants compared with BX+ plants. Oikawa et al. (2004) observed that inoculation of A. alternata in mature maize leaves induced production of HDMBOA-Glc, a proposed precursor to MBOA. Tolerance of fungi to MBOA is positively correlated with tolerance to BOA (Glenn et al., 2001). In the present study, A. alternata (Tolerance class I) was isolated more frequently from BX– leaves than BX+ leaves in 9-wk-old plants. Induction of HDMBOA-Glc production by A. alternata could result in release and accumulation of MBOA, which would be self-inhibitory, allowing colonization by other species with higher BOA/MBOA tolerance.

Genotypes of a host species can vary significantly in their influence on community organization (Whitham et al., 2006). Assessing the influence of Ustilago maydis resistance on endophyte community assembly in maize, Pan et al. (2008) found that community structure was not correlated with resistance, but rather with host genotype. Consistent with this, our results demonstrate differences in endophyte communities among all three genotypes, but with striking differences in tolerance of host toxins among community members between the BX− mutant and the BX+ genotypes.

McGill et al. (2006) propose that the most direct route to understanding the mechanisms underlying assembly of speciose communities is through study of functional trait variation across environmental gradients. Here, we have compared the extremes of the BX concentration gradient. Traditional species diversity measures did not detect a difference between endophyte communities from BX+ and BX− genotypes. When the distribution of functional traits in communities was analysed, clear differences between the communities of BX+ and BX– genotypes were apparent. Such a shift in functional diversity, despite no change in species diversity, has frequently been noted in plant communities (for review of plant functional types in ecology see Duckworth et al., 2000).

BX+ plants have a higher incidence of Fusarium than BX− plants

Fusarium was isolated 14 to 35 times more frequently from BX+ leaves than from BX− leaves in 9-wk-old plants. This suggests that Fusarium species are more competitive in the presence of BXs. Successive cropping of maize or other BX-producing plants in the same locality may increase the dominance of Fusarium in the endophyte community. Carry-over across seasons would be expected through crop residue, which is considered to be the most important source of inoculum for endophyte infection (Sutton, 1992). Our data suggest that crop residue of BX+ genotypes will have more Fusarium inoculum than the BX– genotype. If BX concentrations are maintained in crop residue and soil, our data suggest that Fusarium is likely to accumulate. Contamination of maize with Fusarium species can lead to human disease, yield loss and livestock toxicosis (Ueno et al., 1997; Marasas, 2001; Marasas et al., 2004; Wu, 2007). Reducing inoculum is the most direct approach to reducing fungal infection (Jouany, 2007).

Commercial maize has been selectively bred to produce elevated levels of BXs. Results from our study suggest that presence of BXs can significantly enhance the competitive ability of Fusarium species. Given that BXs are general phytoprotectants, commercial cultivation of BX− maize is not realistic. Investigation of a relationship between BX concentration and abundance of Fusarium in maize tissue could inform crop management strategies. It is possible that there is a threshold concentration of BXs that is high enough to provide insecticidal benefits to the plant, but low enough to obviate any colonization benefit to Fusarium species.

Commonalities between endophyte community ecology in agricultural and naturally occurring plants

There is an assumption that endophyte communities of agricultural plants and plants in their natural habitat are fundamentally different. Likely this stems from emphasis on individual fungal species as pathogens in agricultural crops, versus a more holistic approach to understanding endophyte communities of wild plants. However, we see four major commonalities between endophyte communities in cultivated and naturally occurring plants. First, the present study found that frequency of endophyte infection increases with plant age, with infection density of leaves reaching 100% in 9-wk-old leaves. Second, we found that in all treatments, diversity decreased over time (Table S1). Both of these trends have also been observed in endophyte communities of tropical plants (Herre et al., 2007). Third, results presented here indicate that above- and below-ground tissue harbor distinct endophyte assemblages, a pattern observed in a diversity of plants (e.g. Kumar & Hyde, 2004). Finally, plant residue is considered the most important source of fungal inoculum in the life histories of both agricultural and wild endophytes (Sutton, 1992; Herre et al., 2007).

Another potential similarity between endophyte communities of plants in agricultural and natural environments is the role of interspecific interactions between fungi in mediating community structure. Arnold et al. (2003) proposed that interspecific competition mediated by leaf chemistry is a common mechanism shaping endophyte communities. Our results are consistent with this hypothesis. Fusarium was significantly more abundant in BX+ plants compared with BX– plants, indicating that BXs provide a competitive advantage to Fusarium. Observational data from the present study indicate that interspecific facilitation may also influence endophyte communities. When isolating fungi from leaf tissue on BOA medium, colonies of A. alternaria and P. macrospinosa occasionally emerged from the colony center of BOA detoxifying Fusarium species. All of the A. alternaria and P. macrospinosa tested, including isolates obtained on BOA medium (1.00% concentration), could not grow above a 0.25% concentration of BOA. This may indicate that BOA detoxifying Fusarium species can facilitate the growth of less tolerant species. A previous in vitro study found that maize endophytes able to tolerate 1.00% and 1.10% concentrations of BOA facilitated growth of maize endophytes unable to grow above a 0.25% and 0.50% concentration (Saunders & Kohn, 2008). These commonalities between agricultural and wild endophyte communities may represent general mechanisms in endophyte community assembly, and would therefore be constructive areas for future research.


We found that plant defense compounds are a significant factor in structuring fungal endophyte communities of maize. Glenn et al. (2001) proposed that BX production by maize enhances the ecological success of Fusarium species in maize; our results support this hypothesis. Further, we found that non-Fusarium species with intermediate BOA tolerance levels had a colonization advantage over BOA-sensitive fungi in BX-producing plants, indicating that a large proportion of the fungal community is influenced by defense compound production.

Our data point to the possibility that breeding for elevated concentrations of BXs in maize has unintentionally allowed for increased colonization by Fusarium and the possibility of increased inoculum load in plant residue and in soil. These results are preliminary to the next logical step of breeding isogenic corn lines that only differ in quantitative production of BXs. The caveat to this approach will be that concentrations of BX are not static in planta, and are also likely to be dynamic in the surrounding soil and crop residue.

Future experiments on endophyte communities of other host species are needed to test the possibility that host defense compounds are a general mechanism in structuring communities. Studies incorporating measurements of multiple factors, such as competition and facilitation between microbes, species composition in the soil-borne ‘spore-bank’ and microbial niche overlap will help to identify key fungal traits in colonization. From this, we will be better able to form hypotheses about evolution of host colonization strategies within a community context.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank L. Hopcroft, S. Saunders-Caradonna and A. Snare, for laboratory assistance, T. Anderson (Agriculture and Agri-Food Canada, Harrow, ON, Canada) and A. Tenuta (OMAFRA, Ridgetown, ON, Canada) for assistance in the field, and J. Brotherton (Dept Crop Science, University of Illinois, Urbana, IL, USA) for providing seed. We greatly appreciate advice from J. B. Anderson (Cell and Systems Biology, University of Toronto, Mississauga, ON, Canada), A. E. Glenn (USDA-ARS Toxicology and Mycotoxin Unit, Athens, GA, USA), P. M. Kotanen (Ecology and Evolutionary Biology, University of Toronto, Mississauga, ON, Canada) and C. Sirjusingh (Cell and Systems Biology, University of Toronto, Mississauga, ON, Canada). This research was supported by a Discovery Grant to L.M.K. from the Natural Sciences and Engineering Research Council of Canada.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Arnold AE. 2007. Understanding the diversity of foliar fungal endophytes: progress, challenges and frontiers. Fungal Biology Reviews 21: 5166.
  • Arnold AE, Mejia LC, Kyllo D, Rojas EI, Maynard Z, Robbins N, Herre EA. 2003. Fungal endophytes limit pathogen damage in a tropical tree. Proceedings of the National Academy of Sciences, USA 100: 1564915654.
  • Bacon CW, Hinton DM, Glenn AE, Macias FA, Marin D. 2007. Interactions of Bacillus mojavensis and Fusarium verticillioides with a benzoxazolinone (BOA) and its transformation product, APO. Journal of Chemical Ecology 33: 18851897.
  • Barry D, Darrah LL. 1991. Effect of research on commercial hybrid maize resistance to European corn-borer (Lepidoptera, Pyralidae). Journal of Economic Entomology 84: 10531059.
  • Broeckling CD, Broz AK, Bergelson J, Manter DK, Vivanco JM. 2008. Root exudates regulate soil fungal community composition and diversity. Applied and Environmental Microbiology 74: 738744.
  • Cambier V, Hance T, de Hoffmann E. 2000. Variation of DIMBOA and related compounds content in relation to the age and plant organ in maize. Phytochemistry 53: 223229.
  • Carter JP, Spink J, Cannon PF, Daniels MJ, Osbourn AE. 1999. Isolation, characterization, and avenacin sensitivity of a diverse collection of cereal-root-colonizing fungi. Applied and Environmental Microbiology 65: 33643372.
  • Chao A, Chazdon RL, Colwell RK, Shen T-J. 2005. A new statistical approach for assessing compositional similarity based on incidence and abundance data. Ecology Letters 8: 148159.
  • Chiapusio G, Pellissier F, Gallet C. 2004. Uptake and translocation of phytochemical 2-benzoxazolinone (BOA) in radish seeds and seedlings. Journal of Experimental Botany 55: 15871592.
  • Clay K, Schardl C. 2002. Evolutionary origins and ecological consequences of endophyte symbiosis with grasses. American Naturalist 160: S99S127.
  • Colwell RK. 2000. Estimates: statistical estimation of species richness.
  • Dixon RA. 2001. Natural products and plant disease resistance. Nature 411: 843847.
  • Duckworth JC, Kent M, Ramsay PN. 2000. Plant functional types: an alternative to taxonomic plant community description in biogeography? Progress in Physical Geography 24: 515542.
  • Finkes LK, Cady AB, Mulroy JC, Clay K, Rudgers JA. 2006. Plant–fungus mutualism affects spider composition in successional fields. Ecology Letters 9: 344353.
  • Gardes M, Bruns TD. 1993. ITS primers with enhanced specificity for basidiomycetes − application to the identification of mycorrhizae and rusts. Molecular Ecology 2: 113118.
  • Geiser DM, Jimenez-Gasco MD, Kang SC, Makalowska I, Veeraraghavan N, Ward TJ, Zhang N, Kuldau GA, O'Donnell K. 2004. FUSARIUM-ID v. 1.0: A DNA sequence database for identifying Fusarium. European Journal of Plant Pathology 110: 473479.
  • Glenn AE, Hinton DM, Yates IE, Bacon CW. 2001. Detoxification of corn antimicrobial compounds as the basis for isolating Fusarium verticillioides and some other Fusarium species from corn. Applied and Environmental Microbiology 67: 29732981.
  • Gotelli NJ, Ellison AM. 2004. The analysis of categorical data. In: Gotelli NJ, Ellison AM, eds. A primer of ecological statistics. Sunderland, MA, USA: Sinauer Associates, Inc., 349382.
  • Hamilton RH. 1962. Segregation for a cyclic hydroxamate in maize seedlings. Maize Genetics Cooperation Newsletter 36: 7172.
  • Hashimoto Y, Shudo K. 1996. Chemistry of biologically active benzoxazinoids. Phytochemistry 43: 551559.
  • Herre EA, Mejia LC, Kyllo DA, Rojas E, Maynard Z, Butler A, van Bael SA. 2007. Ecological implications of anti-pathogen effects of tropical fungal endophytes and mycorrhizae. Ecology 88: 550558.
  • Jouany JP. 2007. Methods for preventing, decontaminating and minimizing the toxicity of mycotoxins in feeds. Animal Feed Science and Technology 137: 342362.
  • Kumar DSS, Hyde KD. 2004. Biodiversity and tissue-recurrence of endophytic fungi in Tripterygium wilfordii. Fungal Diversity 17: 6990.
  • Leslie JF, Summerell BA, Bullock S. 2006. The Fusarium laboratory manual. Oxford, UK: Blackwell Science.
  • Magurran AE. 2004. Measuring biological diversity. Oxford, UK: Blackwell Science.
  • Marasas WFO. 2001. Discovery and occurrence of the fumonisins: a historical perspective. Environmental Health Perspectives 109: 239243.
  • Marasas WFO, Riley RT, Hendricks KA, Stevens VL, Sadler TW, Gelineau-van Waes J, Missmer SA, Cabrera J, Torres O, Gelderblom WCA et al . 2004. Fumonisins disrupt sphingolipid metabolism, folate transport, and neural tube development in embryo culture in vivo: a potential risk factor for human neural tube defects among populations consuming fumonisin-contaminated maize. Journal of Nutrition 134: 711716.
  • Marin S, Companys E, Sanchis V, Ramos AJ, Magan N. 1998. Effect of water activity and temperature on competing abilities of common maize fungi. Mycological Research 102: 959964.
  • Martin KJ, Rygiewicz PT. 2005. Fungal-specific PCR primers developed for analysis of the ITS region of environmental DNA extracts. BMC Microbiology 5: 110.
  • McGill BJ, Enquist BJ, Weiher E, Westoby M. 2006. Rebuilding community ecology from functional traits. Trends in Ecology and Evolution 21: 178185.
  • Munkvold GP. 2003. Cultural and genetic approaches to managing mycotoxins in maize. Annual Review of Phytopathology 41: 99116.
  • Nicol RW, Yousef L, Traquair JA, Bernards MA. 2003. Ginsenosides stimulate the growth of soilborne pathogens of American ginseng. Phytochemistry 64: 257264.
  • Niemeyer HM, Perez FJ. 1995. Potential of hydroxamic acids in the control of cereal pests, diseases and weeds. In: Inderjit, Daksini KMM, Einhellig FA, eds. Allelopathy: organisms, processes and applications. Washington DC, WA, USA: ACS Symposium Series 582, American Chemical Society, 260270.
  • Oikawa A, Ishihara A, Tanaka C, Mori N, Tsuda M, Iwamura H. 2004. Accumulation of HDMBOA-Glc is induced by biotic stresses prior to the release of MBOA in maize leaves. Phytochemistry 65: 29953001.
  • Pan JJ, Baumgarten A, May G. 2008. Effects of host plant environment and Ustilago maydis infection on the fungal endophyte community of maize (Zea mays). New Phytologist 178: 147156.
  • Park WJ, Hochholdinger F, Gierl M. 2004. Release of the benzoxazinoids defense molecules during lateral and crown root emergence in Zea mays. Journal of Plant Physiology 161: 981985.
  • Richardson MD, Bacon CW. 1993. Cyclic hydroxamic acid accumulation in corn seedlings exposed to reduced water potentials before, during, and after germination. Journal of Chemical Ecology 19: 16131624.
  • Ritchie SW, Hanway JJ, Benson GO, Herman JC, Lupkes ST. 1993. How a corn plant develops, Special Report No. 48. Ames, IA, USA: Iowa State University of Science and Technology, Cooperative Extensive Service.
  • Rudgers JA, Holah J, Orr SP, Clay K. 2007. Forest succession suppressed by an introduced plant–fungal symbiosis. Ecology 88: 1825.
  • Saunders M, Kohn LM. 2008. Host-synthesized secondary compounds influence the in vitro interactions between fungal endophytes of maize. Applied and Environmental Microbiology 74: 136142.
  • Schardl CL, Leuchtmann A, Spiering MJ. 2004. Symbioses of grasses with seedborne fungal endophytes. Annual Review of Plant Biology 55: 315340.
  • Schulz B, Boyle C. 2005. The endophytic continuum. Mycological Research 109: 661686.
  • Seghers D, Wittebolle L, Top EM, Verstraete W, Siciliano SD. 2004. Impact of agricultural practices on the Zea mays L. endophytic community. Applied and Environmental Microbiology 70: 14751482.
  • Smith CW, Betran J, Runge ECA. 2004. Corn: origin, history, technology and production. Hoboken, NJ, USA: John Wiley and Sons, Inc.
  • Stinson KA, Campbell SA, Powell JR, Wolfe BE, Callaway RM, Thelen GC, Hallett SG, Prati D, Klironomos JN. 2006. Invasive plant suppresses the growth of native tree seedlings by disrupting belowground mutualisms. PLoS Biology 4: 727731.
  • Summerell BA, Salleh B, Leslie JF. 2003. A utilitarian approach to Fusarium identification. Plant Disease 87: 117128.
  • Sutton JC. 1992. Epidemiology of wheat head blight and maize ear rot caused by Fusarium graminearum. Canadian Journal of Plant Pathology 4: 195209.
  • Ueno Y, Iijima K, Wang SD, Sugiura Y, Sekijima M, Tanaka T, Chen C, Yu SZ. 1997. Fumonisins as a possible contributory risk factor for primary liver cancer: a 3-year study of corn harvested in Haimen, China, by HPLC and ELISA. Food Chemistry and Toxicology 35: 11431150.
  • VanEtten H, Temporini E, Wasmann C. 2001. Phytoalexin (and phytoanticipin) tolerance as a virulence trait: why is it not required by all pathogens? Physiological and Molecular Plant Pathology 59: 8393.
  • Whitham TG, Bailey JK, Schweitzer JA, Shuster SM, Bangert RK, Leroy CJ, Lonsdorf EV, Allan GJ, DiFazio SP, Potts BM et al . 2006. A framework for community and ecosystem genetics: from genes to ecosystems. Nature Reviews Genetics 7: 510523.
  • Wu F. 2007. Measuring the economic impacts of Fusarium toxins in animal feeds. Animal Feed Science and Technology 137: 363374.
  • Wu F, Munkvold GP. 2008. Mycotoxins in ethanol co-products: modeling economic impacts on the livestock industry and management strategies. Journal of Agricultural and Food Chemistry 56: 39003911.
  • Zuniga GE, Argandona VH, Niemeyer HM, Corcuera LJ. 1983. Hydroxamic acid content in wild and cultivated Gramineae. Phytochemistry 22: 26652668.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Fig. S1 Collection data and community bar graphs of isolates obtained from root (S1a) and leaf (S1b) tissue of maize (Zea mays) on potato dextrose agar (PDA).

Fig. S2 Similarity coefficients of fungal endophyte communities from leaf and root tissue of maize (Zea mays) grown in Harrow and Ridgetown, Ontario, Canada.

Table S1 Diversity of fungal endophyte communities in maize (Zea mays)

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

NPH_2746_sm_TableS1.doc68KSupporting info item
NPH_2746_sm_Figs_legends.doc56KSupporting info item
NPH_2746_sm_FigS1a.pdf282KSupporting info item
NPH_2746_sm_FigS1b.pdf280KSupporting info item
NPH_2746_sm_FigS2.pdf306KSupporting info item