APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Mar. 2004, p. 1475–1482 0099-2240/04/$08.00⫹0 DOI: 10.1128/AEM.70.3.1475–1482.2004 Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Vol. 70, No. 3
Impact of Agricultural Practices on the Zea mays L. Endophytic Community Dave Seghers, Lieven Wittebolle, Eva M. Top,† Willy Verstraete,* and Steven D. Siciliano‡ Laboratory of Microbial Ecology and Technology (LabMET), Ghent University, B-9000 Ghent, Belgium Received 11 September 2003/Accepted 17 November 2003
Agricultural practices are known to alter bulk soil microbial communities, but little is known about the effect of such practices on the plant endophytic community. We assessed the influence of long-term applications (20 years) of herbicides and different fertilizer types on the endophytic community of maize plants grown in different field experiments. Nested PCR-denaturing gradient gel electrophoresis (DGGE) analyses targeting general bacteria, type I or II methanotrophs, actinomycetes, and general fungi were used to fingerprint the endophytic community in the roots of Zea mays L. Low intraplant variability (reproducible DGGE patterns) was observed for the bacterial, type I methanotroph, and fungal communities, whereas the patterns for endophytic actinomycetes exhibited high intraplant variability. No endophytic amplification product was obtained for type II methanotrophs. Cluster and stability analysis of the endophytic type I methanotroph patterns differentiated maize plants cultivated by using mineral fertilizer from plants cultivated by using organic fertilizer with a 100% success rate. In addition, lower methanotroph richness was observed for mineral-fertilized plants than for organically fertilized plants. The use of herbicides could not be traced by fingerprinting the endophytic type I methanotrophs or by evaluating any other endophytic microbial group. Our results indicate that the effect of agrochemicals is not limited to the bulk microbial community but also includes the root endophytic community. It is not clear if this effect is due to a direct effect on the root endophytic community or is due to changes in the bulk community, which are then reflected in the root endophytic community. confirmed this view, and bacterial endophytic species have been implicated in the promotion of plant growth and protection against pathogens (3, 42). In the majority of studies investigating bacterial endophytic diversity in maize species the workers use cultivation-based techniques (isolation). The most frequently isolated members of the endophytic bacterial community in maize are Enterobacter spp. (members of the gamma subclass of the class Proteobacteria [gamma-proteobacteria]), followed by the betaproteobacterial Burkholderia spp. (29). Such studies provide an important glimpse into endophytic diversity, as well as microbial strains for further investigation. However, due to the unknown growth requirements of many bacteria and the presence of cells that are in a viable but noncultivable state (43), cultivation-dependent biodiversity studies of the endophytic community are somewhat limited. In recent work Chelius and Triplett (7) found that cultivation techniques captured 48% of the bacterial diversity retrieved by cultivation-independent clonal assessment. The 16S rRNA gene is a phylogenetic marker that is frequently used to describe the microbial community in natural environments without a need for cultivation (11, 45). Methods that rely on the 16S rRNA gene to characterize the microbial community structure include denaturing gradient gel electrophoresis (DGGE), terminal restriction fragment length polymorphism, and amplified ribosomal DNA restriction analysis. These methods are frequently used to study bacterial communities in soil ecosystems (17, 24) but have been used only rarely to evaluate endophytic microbial communities of agronomic crops. Recently, DGGE and terminal restriction fragment
Microbial endophytic species are present in a wide range of plant species and reside either within cells (23), in the intercellular space (33), or in the vascular system (2) of a plant. Microbial endophytes are typically defined as microorganisms that are detected after surface sterilization of a plant part (16) and are assumed to originate from the seed and/or the surrounding environment. It is not clear how important the seed reservoir is for the endophytic community. McInroy and Kloepper (29) found 103 to 105 CFU/g in cotton seeds and 10 CFU/g in maize seeds, whereas in an extensive study of citrus seeds (1) Araujo et al. could not detect any endophytic species. In contrast, the soil, particularly the rhizosphere, is widely accepted to be an important source of root endophytes, and most root endophyte species are also present in rhizosphere soil (13, 37). The root endophytes are thought to enter the plant by local cellulose degradation or fractures in the root system (14). Traditionally, bacterial endophytes were assumed to be latent pathogens that did no substantive harm and provided no benefit to the host plant (44), but recently, it has been proposed that much like their fungal brethren, some bacterial endophytes may also be beneficial (6). Recent reports have * Corresponding author. Mailing address: Laboratory of Microbial Ecology and Technology (LabMET), Faculty of Agricultural and Applied Biological Sciences, Ghent University, Coupure Links 653, B-9000 Ghent, Belgium. Phone: 32 (0)9 264 59 76. Fax: 32 (0)9 264 62 48. E-mail:
[email protected]. † Present address: Department of Biological Sciences, University of Idaho, Moscow, ID 83844-3051. ‡ Present address: Department of Soil Science, University of Saskatchewan, Saskatoon, Saskatchewan S7N 5A8, Canada. 1475
1476
SEGHERS ET AL.
length polymorphism analyses were used to study the endophytic community in potato plants (12, 37), and amplified ribosomal DNA restriction analysis was used to study the root endophyte diversity in Zea mays L. (7). The authors found a wide diversity (six bacterial phylogenetic divisions and 74 distinct phylotypes) of bacteria associated with the roots of Z. mays L. It was suggested that the bacteria associated with maize roots were a subset of the larger soil microbial community. This viewpoint postulates that a certain subpopulation of the wider soil community prospers in the root endophytic zone, suggesting that the root endophytic population composition is due to interactions of plant-specific and soil-specific factors. We hypothesized that if rhizosphere and endophytic communities are closely linked, the endophytic communities may reflect differences in agronomic practices. In previous work the researchers found that application of different types of fertilizer (compost versus mineral fertilizer) resulted in altered bulk soil methanotrophic diversity and activity (35). In the present study, we used group-specific DGGE to fingerprint the endophytic community in the roots and kernels of Z. mays L. Initially, we tried to confirm the finding of Chelius and Triplette (7) that the structure of the endophytic community in maize roots was related to the structure of the microbial community in the bulk soil and the rhizosphere soil. In the second part of this study, we evaluated the structure of the root endophytic community in maize plants grown under different agricultural conditions to investigate the potential effects of herbicide use and the use of mineral fertilizer versus organic fertilizer on this structure. Finally, we also assessed the structure of the endophytic community in maize kernels originating from plants subjected to the different agricultural treatments. MATERIALS AND METHODS Experimental sites and sampling. Since 1982, maize plants (variety LG21.85) were cultivated on two experimental fields located near Melle, Belgium. Each experimental field was divided into eight replicate blocks, and there were four blocks (6 by 6 m2) per agricultural treatment. Experimental field 1 contained four blocks which received both atrazine (750 g ha⫺1) and metalochlor (2,000 g ha⫺1) (treatment A) and four blocks which received no herbicide (weeds were removed manually) (treatment K). Both treatments in experimental field 1 received mineral fertilizer (153 kg of N ha⫺1). The herbicide regimen was thus the only parameter that was different for treatment A and treatment K. The blocks located in experimental field 2 received either mineral fertilizer (treatment M) or vegetable, fruit, and garden compost (GFT compost) (treatment G). The weeds in experimental field 2 were removed mechanically, and so the fertilizer regimen was the only parameter that was different for treatment M and treatment G. Thus, this experimental setup allowed separate evaluation of the effects of herbicides and different fertilizer regimens on the endophytic community. We acknowledge that this experimental design resulted in pseudoreplication and that we did not have multiple fields for each treatment, but this was the result of the logistical constraints of long-term agricultural experiments. Samples of four maize plants per treatment were obtained in September 2002, 2 weeks before harvesting of all plants. In addition to the maize plants, samples of bulk and rhizosphere soil were also obtained from the treatment G blocks. Surface sterilization of maize roots and kernels. Roots and kernels were washed with tap water and distilled water to remove attached soil. Subsequently, the plant parts were immersed in ethanol and shaken manually for 1 min. The ethanol was replaced by sodium hypochlorite (amended with 0.1% Tween 20), and the preparations were shaken manually for 1 min. The solution was replaced by fresh sodium hypochlorite, and the plant parts were shaken on a rotary shaker (140 rpm) for an additional 20 min. Finally, the plant parts were washed three times with sterile distilled water. To ensure that the surfaces were sterile, maize roots and kernels were imprinted on tryptic soy agar plates and the water from the final washing step was spread on tryptic soy agar plates (29). Since in this study we focused mainly on molecular techniques, the final wash water was also
APPL. ENVIRON. MICROBIOL. subjected to PCR analysis (after boiling to release possible DNA) with general bacterial primers P338F and P518r to evaluate the surface sterility (4). Root and kernel samples that were not contaminated as determined by both the culturedependent and culture-independent sterility tests were kept and used for further analysis. Plate counts of endophytes. Surface-sterilized plant pieces (roots and kernels separately) were first placed in a sterilized physiological solution (8.5 g of NaCl liter⫺1; dilution, 10⫺1) and subsequently homogenized with a Stomacher LabBlender (Seward Medical, London, United Kingdom). The number of cultivable endophytes in maize roots or kernels, expressed in CFU per gram (fresh weight), was determined by spreading 0.1 ml of homogenized surface-sterilized plant material onto R2A (Difco, Detroit, Mich.) agar medium. Four replicates for each agricultural treatment were spread on the agar plates and incubated for 7 days at 28°C. DNA isolation and PCR amplification. DNA was extracted from the surfacedisinfected plant material by using a protocol previously described for soil samples (8). First, the minimum amount of plant material necessary to generate reproducible patterns for one plant was determined (analysis of intraplant variability). Therefore, DNA was extracted three times from plant material (1 or 5 g) originating from one maize plant (plant G3). To check the interplant variability, DNA was extracted from four different plants that were cultivated under the same circumstances (plants G1, G2, G3, and G4). In addition, DNA was also extracted from bulk soil samples and rhizosphere samples (soil attached to the roots after sampling of the maize plant) from the treatment G block by using a previously described method (8). This made it possible to compare the community structure within the roots (endophytes) and the community structure outside the roots (rhizosphere) for agricultural treatment G. Finally, the endophytic communities of maize plants cultivated by using different agricultural practices were compared. To do this, samples of four maize plants per treatment were obtained and surface sterilized as described above. After extraction of the DNA of the soil and plant samples, a 100-l aliquot of each crude extract was purified with Wizard PCR preps (Promega, Madison, Wis.), and the DNA concentration was measured spectrophotometrically. Subsequently, a nested PCR was performed to obtain DNA amplification products suitable for DGGE analysis (4). General primers targeting all bacteria (P388f and P518r) (32), actinomycetes (P243f and P518r) (19), and fungi (EF4f and NS3r) (40) were used. In addition, group-specific 16S rRNA gene primers targeting type I (MB10 gamma) and type II (MB9 alfa) methanotrophs were used in this study (17). All PCRs were performed with a 9600 thermal cycler (Perkin-Elmer, Norwalk, Conn.). The final concentrations of the different components in the master mixture (in DNase- and RNase-free filter-sterilized water [Sigma-Aldrich Chemie, Steinheim, Germany]) were as follows: 0.2 M for each primer, 200 M for each deoxynucleoside triphosphate, 1.5 mM MgCl2, 1⫻ thermophilic DNA polymerase, 10⫻ reaction buffer (MgCl2 free), 1.25 U of Taq DNA polymerase (Promega) 50 l⫺1, and 400 ng of bovine serum albumin (Hoffman-La Roche, Basel, Switzerland) l⫺1. In the first PCR round, 1 l of purified DNA was added to 24 l of the PCR master mixture, and in the second round, 1 l of amplified product from the first round was added to 24 l of the PCR mixture. After each PCR, the size of the amplification product was verified on a 1% agarose gel. DGGE analysis. DGGE analysis was performed by using a DCode system (Bio-Rad, Hercules, Calif.) as described previously (4). In brief, PCR amplification samples were loaded onto 8% (wt/vol) polyacrylamide gels in 1⫻ TAE (20 mM Tris, 10 mM acetate, 0.5 mM EDTA [pH 7.4]). The polyacrylamide gels were made with a denaturant gradient ranging from 45 to 60%. Electrophoresis was performed overnight for 17 h at 60°C and 38 V. After electrophoresis, the gels were soaked for 30 min in SYBR Green I nucleic acid gel stain (dilution, 1:10,000; FMC BioProducts, Rockland, Maine). Each stained gel was immediately photographed on a UV transillumination table with a video camera module (Vilbert Lourmat, Marne-la Valle´, France). Statistical processing of DGGE patterns. The DGGE patterns were clustered by using Bionumerics software (Applied Maths, Kortrijk, Belgium). A matrix of similarities for the densiometric curves of the band patterns was calculated based on the Pearson product-moment correlation coefficient, and dendrograms were created by using Ward linkage (46). Relevant and nonrelevant clusters were separated by the statistical cluster cutoff method (Bionumerics manual 2.5). In addition, the stability of the endophytic patterns originating from plants cultivated under different agricultural circumstances was calculated. For the different primer sets used in this study, the percentage of DGGE patterns assigned to the correct agricultural treatment was calculated. The group violation method was used to calculate these percentages (Bionumerics manual 2.5). Finally, the richness of each of the different bacterial communities was esti-
Z. MAYS ENDOPHYTIC COMMUNITY
VOL. 70, 2004
FIG. 1. Assessment of intraplant variability of the structure of the endophytic communities in maize roots (a) and in maize kernels (b). General bacterial primers were used. R1, R2, R3, and R4, root samples (1 g) originating from one plant; K1, K2, K3, and K4, kernel samples (1 g) originating from one plant. Bands 1, 2, and 3 from kernels were cut out and sequenced.
mated. The bands on the DGGE gels were divided into band classes, and these classes were exported to EstimateS (version 6.0; R. K. Colwell; http://viceroy.eeb .uconn.edu/estimates). This program allows statistical analysis of species richness from samples by calculating the Chao1 index (19): Chao1 ⫽ Sobs ⫹ (L2/2M), where Sobs is the total number of species observed, L is the number of species that occur in only one sample (unique species), and M is the number of species that occur in exactly two samples. This is an incidence-based nonparametric estimator that uses presence-absence data and can be used with 16S rRNA DGGE patterns to obtain a first estimate of community richness, making it a suitable index for PCR-based analysis (22). Identification of dominant bands in DGGE patterns. To identify the most dominant members of the maize endophytic community, several bands were cut out from a DGGE gel and placed into 20 l of sterile water. These bands were reamplified with bacterial primers P338f and P518r, and the amplification product was loaded onto a new DGGE gel. The steps were repeated until a pure amplification product (one band on a DGGE gel) was obtained. The amplification products were subsequently sequenced by ITT Biotech (Bielefeld, Germany). The partial sequences (approximately 180 bp) were aligned with 16S rRNA gene sequences obtained from the National Center for Biotechnology Information database by using the BLAST search program, version 2.0.
RESULTS Intraplant variability of the endophytic community. We first determined the minimal amount of plant material needed to generate reproducible 16S rRNA gene DGGE patterns for a single plant sample. Initially, 1-g samples of surface-disinfected roots and kernels (four replicates) originating from plant G3 were subjected to PCR-DGGE analysis conducted with general bacterial primers (Fig. 1a). The DGGE banding patterns for the root samples were dissimilar, with an average similarity of only 10%. Thus, a single maize plant could not be reproducibly assessed with 1 g of roots, and we increased the sample size to 5 g of surface-sterilized root material (Fig. 2a). This increased the level of similarity of replicate samples from the same plant to 90% (as determined with general bacterial primers). In contrast to the surface-disinfected root samples, 1 g of surface-disinfected kernels resulted in highly similar (90%) DGGE patterns (Fig. 1b). Relationship of the root endophytic community to the soil microbial community. Our goal was to compare the structures
1477
of the endophytic communities in four different maize plants (5 g of root material), all grown under the same circumstances (agricultural treatment G), with the structures of the communities of the bulk and rhizosphere soil in which the maize plants were cultivated. Before PCR-DGGE analysis, the DNA yields after extraction were examined, and no significant difference between the different samples was observed. Cluster analysis of the DGGE patterns (as determined with general bacterial primers) of the endophytic community indicated that this community was different than the bulk and rhizosphere community (Fig. 2a). The PCRs for the bulk and rhizosphere soil samples were probably inhibited as low PCR amplification yields were observed. As a result, only a few faint bands were visible on the DGGE gels. In contrast, the patterns for the endophytic community contained approximately 15 distinct bands. Thus, it was difficult to determine if bands present in the root interior (endophytes) were also present in the soil community. To identify the numerically dominant populations in the root endophytic community, a number of dominant bacterial bands were excised from the gel (Fig. 2a), reamplified, and sequenced. The names and accession numbers for most closely related organisms and their percentages of similarity are shown in Table 1. All sequences fell into the gamma subdivision of the Proteobacteria; four of the six sequences were pseudomonad sequences, one was an Enterobacter sequence, and one was a Rahnella sequence. Group-specific primers targeting type I and type II methanotrophs were also used to evaluate the presence of methanotrophs in the roots of a maize plant. The root endophytic DNA samples yielded an amplification product for type I methanotrophs, whereas type II methanotrophs could not be amplified (Table 2). DGGE analysis of the type I methanotroph endophytes (Fig. 2b) confirmed the different community structures of the bacteria living inside the maize roots (endophytes) and the bacteria living just outside the roots (rhizosphere). Most endophyte bands were also visible in the rhizosphere samples. Reproducible patterns resulting in low intraplant variability (100% similarity) and low interplant variability (80% similarity) were obtained. Primers targeting actinomycetes and fungi also yielded amplification products (Fig. 2c and d). The actinomycete intraplant reproducibility was low (50% similarity), and the interplant reproducibility was even lower (40% similarity). Therefore, the actinomycete primer set was not used in further analyses. The reproducibility for the endophytic fungal community was slightly better than that for the actinomycetes, with an intraplant similarity of 80% and an interplant similarity of 60%. Based on the number of bands on the DGGE gels, richness estimates (Chao1 indices) were calculated for the different parts of the soil ecosystem (Table 3). No index could be calculated for the total bacterial community in bulk and rhizosphere soil because DGGE could not adequately separate different bands. The use of group-specific primers for methanotrophs showed that there was a decreasing trend in richness from the bulk soil to the rhizosphere soil to the root endophytic community for plants cultivated in the soil. In contrast, the fungal richness appeared to be higher for the endophytic community than for the soil communities.
1478
SEGHERS ET AL.
APPL. ENVIRON. MICROBIOL.
FIG. 2. Structure of the maize root endophytic community and cluster analysis obtained with different primer sets, including general bacterial primers (a), type I methanotrophic primers (b), primers specific for actinomycetes (c), and fungal primers (d). Bu1 and Bu2, bulk soil samples; Rh1 and Rh2, rhizosphere samples; G3, G3⬘, and G3⬙, endophytic samples originating from one plant (intraplant variability); G1, G2, and G4, endophytic samples originating from different plants cultivated in the same field (interplant variability). Bands 1 to 6 (obtained with general bacterial primers) were cut out and sequenced.
Z. MAYS ENDOPHYTIC COMMUNITY
VOL. 70, 2004
1479
TABLE 1. Sequence analysis of bands retrieved from endophytic DGGE patterns (obtained with general bacterial primers) for maize roots and kernels Closest match in GenBank database
Band
Taxon
Maize rootsa 1 2 3 4 5 6 Maize kernelsb 1 2 3 a b
Accession no.
% Similarity
Putative phylum
Rahnella aquatilis Enterobacter sp. Pseudomonas sp. Pseudomonas syringae Pseudomonas sp. Pseudomonas putida
X79938 AF500319 AY275482 AY275476 AY275484 AJ308312
99 97 96 98 98 97
Gamma-proteobacteria Gamma-proteobacteria Gamma-proteobacteria Gamma-proteobacteria Gamma-proteobacteria Gamma-proteobacteria
Zea mays chloroplast Enterobacter sp. Enterobacter intermedius
X86563 AF500319 AF310217
99 98 97
Gamma-proteobacteria Gamma-proteobacteria
See Fig. 2a. See Fig. 1b.
Effects of agricultural treatments on the root endophytic community. The composition of the root endophytic community was clearly influenced by different fertilizer treatments (Fig. 3). Cluster analysis of the DGGE patterns obtained with type I methanotroph primers separated the samples subjected to treatment M (mineral fertilizer) and the samples subjected to treatment G (GFT compost), and there was 50% similarity between the patterns obtained for these two treatments. This separation was limited to the type I methanotrophs as the bacterial community as a whole did not generate separate clusters according to the fertilizer type. Analysis of the fungal community also did not differentiate between the different fertilizer treatments. The herbicide applications did not result in separate clusters for treatment K and treatment A for all primer combinations investigated (total bacterial, methanotrophic, and fungal primers). In addition, the different agricultural practices had little effect on the number of cultivable root endophytes; the sizes of the endophytic populations for all four treatments were approximately 7 log10 CFU per g of roots (data not shown). Stability analysis of the type I methanotroph patterns yielded an average rate of correct classification (ARCC) of 76%, and the accuracy of the assignments for the compost treatment and the mineral fertilizer treatment was 100% (Table 4). Differentiation of endophytic communities exposed to herbicides from nonexposed communities was not as successful; only 24% of the herbicide-exposed samples were assigned to the correct group, whereas 80% of the nonexposed communities were assigned to the correct group. Despite this limitation, the methanotroph patterns were more informative than the data
obtained with the general bacterial primers, which yielded an ARCC of 68%, and the data obtained with the fungal primers, which yielded an ARCC of 40%. Supporting the ARCC analysis, richness estimates for the root endophytic communities originating from the different agricultural treatments indicated that the richness for the type I methanotrophs in the organically (GFT compost) fertilized soil was significantly higher than the richness for the other treatments (Table 3). The estimates of the total bacterial and fungal richness did not reveal significant differences between the different agricultural treatments. Endophytic communities within maize kernels. Both culture-dependent and culture-independent methods revealed the presence of endophytic species in the maize kernels obtained from plants subjected to the different agricultural treatments. The number of cultivable endophytes was approximately 4 log units lower in the kernels than in the roots, and the values did not differ between treatments (data not shown). PCR analysis of the endophytic community in the maize kernels revealed amplification with general bacterial primers, while other primer combinations did not yield amplification products suitable for DGGE (Table 2). The general bacterial patterns were dominated by one very intense band that was not bacterial in origin. Sequence analysis (Table 1) revealed that band 1 showed the highest similarity with the Z. mays chloroplast sequence (accession no. X86563). The less dominant
TABLE 3. Estimation of richness based on DGGE patterns obtained with different primer combinations Chao1 indexa
TABLE 2. Amplification of soil and endophytic samples with different primer combinations Target
Soil samples
Root endophytes
Kernel endophytes
Bacteria (general primers) Type I methanotrophs Type II methanotrophs Fungi Actinomycetes
⫹a ⫹ ⫹ ⫹ ⫹
⫹ ⫹ ⫺ ⫹ ⫹
⫹ ⫺ ⫺ ⫺ ⫺
a
⫹, amplification (positive signal on agarose gel); ⫺, no amplification.
Sample
Bulk soil (treatment G) Rhizosphere (treatment G) Endophytic community Treatment G Treatment M Treatment K Treatment A
Bacteria
17 a 16 a 17 a 18 a
Type 1 methanotrophs
Fungi
24.0 a 20.0 b
5.0 a 6.0 b
16 c 11 d 11 d 12 d
30 c 28 c 28 c 26 c
a Means followed by the same letter are not statistically significant as determined by the Duncan multiple-range test (P ⬍ 0.05).
1480
SEGHERS ET AL.
APPL. ENVIRON. MICROBIOL.
FIG. 3. Cluster analysis of endophytic root fingerprints obtained from plants cultivated under different agricultural conditions by using general bacterial primers (a), type I methanotrophic primers (b), and fungal primers (c). K1 to K4, maize plants cultivated without herbicides (field 1); A1 to A4, maize plants cultivated with herbicides (field 1); G1 to G4, maize plants cultivated with GFT compost (field 2); M1 to M4, maize plants cultivated with mineral fertilizer (field 2).
bands 2 and 3 exhibited the highest similarity with sequences of Enterobacter sp. (accession no. AF500319) and Enterobacter intermedius (AF310217), respectively. Subsequently, kernels obtained from maize plants grown under the different agricultural conditions were subjected to a PCR-DGGE analysis. Cluster analysis of the endophytic communities in the kernels did not reveal a significant separation of groups based on fertilizer application or herbicide application (data not shown). DISCUSSION Analysis of the root endophytic community with group-specific primers differentiated between organic and mineral agriTABLE 4. Assignment of DGGE patterns obtained with different primer sets to the different agricultural treatments Primers
General bacterial
Type I methanotrophs
General fungal
a
Treatment
A K M G A K M G A K M G
% of DGGE patterns assigned to the following groups: Treatment Treatment Treatment Treatment A K M G
26a 74
13 87a
24 76
a
20 80a
36 64
a
49 51a
Percentage of correctly assigned DGGE patterns.
64a 36
3 97a
100a 0
0 100a
13a 87
38 62a
cultural practices. In contrast, the use of herbicides was not reflected in the structure of the maize endophytic community. The effects of different agricultural practices on the bulk soil microbial community have been extensively studied (8, 10, 41), but to the best of our knowledge, this is the first report of the impact of agricultural practices on the endophytic community. With the recent recognition that endophytic bacteria (3, 42) play an important role in growth promotion and disease resistance, this research raises the possibility that it may be possible to optimize bacterial control agents in crop plants by considering the agricultural practices when bacterial inoculants are used. In previous research the workers found that intensive use of mineral fertilizers substantially changes the composition of the bulk soil microbial community (28). In particular, the methane-oxidizing community is negatively affected by intensive use of mineral fertilizers, whereas organic fertilizers stimulate the soil methanotrophic community (35). The results of the present investigation indicate not only that the bulk soil methanotrophic community is dependent of the fertilizer type but also that the endophytic community is dependent on this factor as well. In contrast, the methanotroph DGGE analysis did not differentiate between the herbicide-treated endophytic community and the nontreated community, although a previous study indicated that the structure of the bulk soil methanotrophic community was altered due to herbicide treatment (36). We postulated that since the herbicides were applied preemergence, there was little direct effect on the maize plants and thus on the endophytic community. The observed fertilizer effect was limited to the root endophytic community, and little dif-
VOL. 70, 2004
ference between kernels was observed. The reason for the fertilizer dependence in the roots and not in the kernels can probably be explained by the fact that the methane generated in the soil immediately surrounding the roots provides an important C source for the root methanotrophic bacteria. In contrast, the kernels do not have such a ready source of methane. In the present study, use of primers targeting type I methanotrophs resulted in amplification for the root endophytic samples, whereas use of primers targeting type II methanotrophs did not, although both primer combinations resulted in amplification for the soil samples (Table 2). This observation supports the hypothesis that the plant selects which endophytes can colonize the interior of the roots. The results of this study indicate that maize roots are preferentially colonized by gamma-proteobacteria since the most dominant bands obtained with general bacterial primers were classified in this group. Type I methanotrophs also belong to the gammaproteobacteria, explaining why they were detected in the root interior whereas type II methanotrophs (alpha-proteobacteria) were not (15). The role of the endophytic methanotroph community in plant health and productivity in upland agricultural settings is currently unknown; all previous studies investigating methanotroph diversity of the root systems were limited to rice or submerged macrophytes (5, 9, 18, 20, 21). The results of this study support the concept that the endophytic bacterial community is a subset of the soil community (13, 29, 38). Richness analysis revealed that the bulk soil had the highest bacterial richness and the endophytic community had the lowest bacterial richness. In a previous study the workers obtained similar results and found that the bacterial diversity decreased from the bulk soil to the rhizosphere to the endophytic community of Lolium perenne and Trifolium repens (27). In a DGGE study Smalla et al. also found a rhizosphere effect; i.e., there was increased abundance of certain populations in the vicinity of the plant roots, which resulted in a lower number of visible bands in the rhizosphere DGGE patterns (39). In the present study, all methanotroph endophytic DGGE bands were also visible in the gels obtained with rhizosphere soil, suggesting that the soil was the major source of the endophytic species. Similarly, most endophytes recovered from barley (31), sugar beet (25), cucumber (26), and wheat and canola plants (13) were also detected in the samples of the soil in which the plants were cultivated. The trend for fungal richness was opposite to the trend displayed by bacteria, with the endophytic community having greater taxonomic richness than the bulk soil community. DGGE detects only the most dominant species. There may be bulk soil fungal species that comprise less than 1% of the community which, due to the limitation of DGGE technology, are not detectable (30). Such an artifact may explain the opposite trend observed. In the present study we found that using 5 g of plant material resulted in reproducible DGGE patterns for the root endophytic community in a single plant. This is in contrast to molecularly based studies of the endophytic root community, in which typically 0.2 to 0.5 g of surface-sterilized material is used (34, 37). The increase in sample size did not reduce the interplant variability for the actinomycete community, which may have been related to the different morphology of these organisms compared to that of methanotrophs or the general bac-
Z. MAYS ENDOPHYTIC COMMUNITY
1481
terial community. In this study we only investigated maize plants, and it is not known if different amounts of material must be analyzed for other plant species. However, our results clearly indicate that molecular investigations of the root endophytic community require that investigators first assess the relationship between within-plant variability and sample size. In conclusion, our results demonstrate that agricultural practices significantly influence certain populations of the root endophytic community. At this time, it is not known how changing methanotroph populations influence plant survival, but given the close relationship between endophytes and plants, this is an area that warrants further research. In this study, plant development (plant age) and the water potential of the soil were similar for all of the plants studied. Consequently, these parameters did not interfere with the results obtained. To extrapolate our results to other plant ages and other water potentials, more research is necessary. REFERENCES 1. Araujo, W. L., W. Maccheroni, C. I. Aguilar-Vildoso, P. A. V. Barroso, H. O. Saridakis, and J. L. Azevedo. 2001. Variability and interactions between endophytic bacteria and fungi isolated from leaf tissues of citrus rootstocks. Can. J. Microbiol. 47:229–236. 2. Bell, C. R., G. A. Dickie, W. L. G. Harvey, and J. W. Y. F. Chan. 1995. Endophytic bacteria in grapevine. Can. J. Microbiol. 41:46–53. 3. Benhamou, N., S. Gagne, D. Le Quere, and L. Dehbi. 2000. Bacterialmediated induced resistance in cucumber: beneficial effect of the endophytic bacterium Serratia plymuthica on the protection against infection by Pythium ultimum. Phytopathology 90:45–56. 4. Boon, N., W. De Windt, W. Verstraete, and E. M. Top. 2002. Evaluation of nested PCR-DGGE (denaturing gradient gel electrophoresis) with groupspecific 16S rRNA primers for the analysis of bacterial communities from different wastewater treatment plants. FEMS Microbiol. Ecol. 39:101–112. 5. Calhoun, A., and G. M. King. 1998. Characterization of root-associated methanotrophs from three freshwater macrophytes: Pontederia cordata, Sparganium eurycarpum, and Sagittaria latifolia. Appl. Environ. Microbiol. 64:1099–1105. 6. Chanway, C. P. 1996. Endophytes: they’re not just fungi! Can. J. Bot. 74: 321–322. 7. Chelius, M. K., and E. W. Triplett. 2001. The diversity of archaea and bacteria in association with the roots of Zea mays L. Microb. Ecol. 41:252– 263. 8. El Fantroussi, S., L. Verschuere, W. Verstraete, and E. M. Top. 1999. Effect of phenylurea herbicides on soil microbial communities estimated by analysis of 16S rRNA gene fingerprints and community-level physiological profiles. Appl. Environ. Microbiol. 65:982–988. 9. Eller, G., and P. Frenzel. 2001. Changes in activity and community structure of methane-oxidizing bacteria over the growth period of rice. Appl. Environ. Microbiol. 67:2395–2403. 10. Engelen, B., K. Meinken, F. von Wintzintgerode, H. Heuer, H. P. Malkomes, and H. Backhaus. 1998. Monitoring impact of a pesticide treatment on bacterial soil communities by metabolic and genetic fingerprinting in addition to conventional testing procedures. Appl. Environ. Microbiol. 64:2814– 2821. 11. Felske, A., A. Wolterink, R. van Lis, W. M. de Vos, and A. D. L. Akkermans. 1999. Searching for predominant soil bacteria: 16S rDNA cloning versus strain cultivation. FEMS Microbiol. Ecol. 30:137–145. 12. Garbeva, P., L. S. van Overbeek, J. W. L. van Vuurde, and J. D. van Elsas. 2001. Analysis of endophytic bacterial communities of potato by plating and denaturing gradient gel electrophoresis (DGGE) of 16S rDNA based PCR fragments. Microb. Ecol. 41:369–383. 13. Germida, J. J., S. D. Siciliano, J. R. de Freitas, and A. M. Seib. 1998. Diversity of root-associated bacteria associated with held-grown canola (Brassica napus L.) and wheat (Triticum aestivum L.). FEMS Microbiol. Ecol. 26:43–50. 14. Gough, C., C. Galera, J. Vasse, G. Webster, E. C. Cocking, and J. Denarie. 1997. Specific flavonoids promote intercellular root colonization of Arabidopsis thaliana by Azorhizobium caulinodans ORS571. Mol. Plant-Microbe Interact. 10:560–570. 15. Graham, D. W., J. A. Chaudhary, R. S. Hanson, and R. G. Arnold. 1993. Factors affecting competition between type-I and type-II methanotrophs in 2-organism, continuous-flow reactors. Microb. Ecol. 25:1–17. 16. Hallmann, J., A. QuadtHallmann, W. F. Mahaffee, and J. W. Kloepper. 1997. Bacterial endophytes in agricultural crops. Can. J. Microbiol. 43:895–914. 17. Henckel, T., M. Friedrich, and R. Conrad. 1999. Molecular analyses of the methane-oxidizing microbial community in rice field soil by targeting the
1482
18. 19.
20. 21.
22. 23. 24.
25. 26. 27. 28. 29. 30.
31. 32.
SEGHERS ET AL.
genes of the 16S rRNA, particulate methane monooxygenase, and methanol dehydrogenase. Appl. Environ. Microbiol. 65:1980–1990. Henckel, T., U. Jackel, S. Schnell, and R. Conrad. 2000. Molecular analyses of novel methanotrophic communities in forest soil that oxidize atmospheric methane. Appl. Environ. Microbiol. 66:1801–1808. Heuer, H., M. Krsek, P. Baker, K. Smalla, and E. M. H. Wellington. 1997. Analysis of actinomycete communities by specific amplification of genes encoding 16S rRNA and gel-electrophoretic separation in denaturing gradients. Appl. Environ. Microbiol. 63:3233–3241. Horz, H. P., A. S. Raghubanshi, E. Heyer, C. Kammann, R. Conrad, and P. F. Dunfield. 2002. Activity and community structure of methane-oxidising bacteria in a wet meadow soil. FEMS Microbiol. Ecol. 41:247–257. Horz, H. P., M. T. Yimga, and W. Liesack. 2001. Detection of methanotroph diversity on roots of submerged rice plants by molecular retrieval of pmoA, mmoX, mxaF, and 16S rRNA and ribosomal DNA, including pmoA-based terminal restriction fragment length polymorphism profiling. Appl. Environ. Microbiol. 67:4177–4185. Hughes, J. B., J. J. Hellmann, T. H. Ricketts, and B. J. M. Bohannan. 2001. Counting the uncountable: statistical approaches to estimating microbial diversity. Appl. Environ. Microbiol. 67:4399–4406. Jacobs, M. J., W. M. Bugbee, and D. A. Gabrielson. 1985. Enumeration, location, and characterization of endophytic bacteria within sugar-beet roots. Can. J. Bot. 63:1262–1265. Kowalchuk, G. A., Z. S. Naoumenko, P. J. L. Derikx, A. Felske, J. R. Stephen, and I. A. Arkhipchenko. 1999. Molecular analysis of ammonia-oxidizing bacteria of the beta subdivision of the class Proteobacteria in compost and composted materials. Appl. Environ. Microbiol. 65:396–403. Lilley, A. K., J. C. Fry, M. J. Bailey, and M. J. Day. 1996. Comparison of aerobic heterotrophic taxa isolated from four root domains of mature sugar beet (Beta vulgaris). FEMS Microbiol. Ecol. 21:231–242. Mahaffee, W. F., and J. W. Kloepper. 1997. Temporal changes in the bacterial communities of soil, rhizosphere, and endorhiza associated with fieldgrown cucumber (Cucumis sativus L.). Microb. Ecol. 34:210–223. Marilley, L., and M. Aragno. 1999. Phylogenetic diversity of bacterial communities differing in degree of proximity of Lolium perenne and Trifolium repens roots. Appl. Soil Ecol. 13:127–136. Marschner, P., E. Kandeler, and B. Marschner. 2003. Structure and function of the soil microbial community in a long-term fertilizer experiment. Soil Biol. Biochem. 35:453–461. McInroy, J. A., and J. W. Kloepper. 1995. Survey of indigenous bacterial endophytes from cotton and sweet corn. Plant Soil 173:337–342. Muyzer, G., E. C. de Waal, and A. Uitterlinden. 1993. Profiling of complex microbial populations using denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl. Environ. Microbiol. 59:695–700. Normander, B., and J. I. Prosser. 2000. Bacterial origin and community composition in the barley phytosphere as a function of habitat and presowing conditions. Appl. Environ. Microbiol. 66:4372–4377. Ovreas, L., L. Forney, F. L. Daae, and V. Torsvik. 1997. Distribution of
APPL. ENVIRON. MICROBIOL.
33. 34. 35.
36.
37.
38.
39.
40.
41. 42. 43. 44. 45. 46.
bacterioplankton in meromictic Lake Saelenvannet, as determined by denaturing gradient gel electrophoresis of PCR-amplified gene fragments coding for 16S rRNA. Appl. Environ. Microbiol. 63:3367–3373. Patriquin, D. G., and J. Dobereiner. 1978. Light-microscopy observations of tetrazolium-reducing bacteria in endorhizosphere of maize and other grasses in Brazil. Can. J. Microbiol. 24:734–742. Reiter, B., U. Pfeifer, H. Schwab, and A. Sessitsch. 2002. Response of endophytic bacterial communities in potato plants to infection with Erwinia carotovora subsp. atroseptica. Appl. Environ. Microbiol. 68:2261–2268. Seghers, D., E. M. Top, D. Reheul, R. Bulcke, P. Boeckx, W. Verstraete, and S. D. Siciliano. 2003. Long-term effects of mineral versus organic fertilizers on activity and structure of the methanotrophic community in agricultural soils. Environ. Microbiol. 10:867–877. Seghers, D., K. Verthe´, D. Reheul, R. Bulcke, S. D. Siciliano, W. Verstraete, and E. M. Top. 2003. Effect of long-term herbicide applications on the bacterial community structure and function in an agricultural soil. FEMS Microbiol. Ecol. 46:139–146. Sessitsch, A., B. Reiter, U. Pfeifer, and E. Wilhelm. 2002. Cultivation-independent population analysis of bacterial endophytes in three potato varieties based on eubacterial and actinomycetes-specific PCR of 16S rRNA genes. FEMS Microbiol. Ecol. 39:23–32. Siciliano, S. D., and J. J. Germida. 1999. Taxonomic diversity of bacteria associated with the roots of field-grown transgenic Brassica napus cv. Quest, compared to the non-transgenic B. napus cv. Excel and B. rapa cv. Parkland. FEMS Microbiol. Ecol. 29:263–272. Smalla, K., G. Wieland, A. Buchner, A. Zock, J. Parzy, S. Kaiser, N. Roskot, H. Heuer, and G. Berg. 2001. Bulk and rhizosphere soil bacterial communities studied by denaturing gradient gel electrophoresis: plant-dependent enrichment and seasonal shifts revealed. Appl. Environ. Microbiol. 67:4742– 4751. Smit, E., P. Leeflang, B. Glandorf, J. D. van Elsas, and K. Wernars. 1999. Analysis of fungal diversity in the wheat rhizosphere by sequencing of cloned PCR-amplified genes encoding 18S rRNA and temperature gradient gel electrophoresis. Appl. Environ. Microbiol. 65:2614–2621. Somerville, L., and M. P. Greaves. 1987. Pesticide effects on soil microflora. Taylor and Francis, London, England. Sturz, A. V., and J. Nowak. 2000. Endophytic communities of rhizobacteria and the strategies required to create yield enhancing associations with crops. Appl. Soil Ecol. 15:183–190. Tholozan, J. L., J. M. Cappelier, J. P. Tissier, G. Delattre, and M. Federighi. 1999. Physiological characterization of viable-but-nonculturable Campylobacter jejuni cells. Appl. Environ. Microbiol. 65:1110–1116. Thomas, W. D., and R. W. Graham. 1952. Bacteria in apparently healthy pinto beans. Phytopathology 42:214. Torsvik, V., R. Sorheim, and J. Goksoyr. 1996. Total bacterial diversity in soil and sediment communities—a review. J. Ind. Microbiol. 17:170–178. Ward, J. H. 1963. Hierarchial grouping to optimize an objective function. J. Am. Stat. Assoc. 58:236–244.
