Abstract

The present study investigated bacterial diversity associated with the roots of maize through the use of culture-dependent and culture-independent methods. Bacterial 16S–23S rDNA internal transcribed spacer sequences (ITS) primers were used to amplify sequences obtained directly from the root matrix by Percoll gradient separation. This assay showed that γ-Proteobacteria within Enterobacter, Erwinia, Klebsiella, Pseudomonas, and Stenotrophomonas genera were predominant groups. The culturable component of the bacterial community was also assessed, revealing that the predominant group was Firmicutes, mainly of Bacillus genus, while Achromobacter, Lysinibacillus, and Paenibacillus genera were rarely found in association with the roots. Only two genera within γ-Proteobacteria, Enterobacter and Pseudomonas, were found in the culture collection. Differences in richness and diversity between the rhizospheric and endophytic bacterial communities were also evidenced. The spectrum of bacteria naturally associated with maize roots is wide and the magnitude of such diversity will depend on the methods chosen for analysis. The knowledge of this spectrum will facilitate the search of microorganisms capable of exerting antagonism to diverse pathogens or detecting plant growth enhancers.

1. Introduction

Maize (Zea mays L.) is one of the three most important agronomic crops in terms of world production, together with rice and wheat. As world cereal consumption tends to increase due to a constantly growing population, productivity should be significantly improved through different strategies that allow an optimization of yields without implicating an increased sown area [1]. Despite the partial success of synthetic chemical pesticides and fertilizers in achieving this goal, a change to environmentally friendly and conservative alternatives is required to protect biodiversity and sustainability of agroecosystems and natural systems all over the world.

Soil microbial communities play an integral and often unique role in ecosystem functions and are among the most complex, diverse, and important assemblages in the biosphere [2]. The study of plant-associated microorganisms is of great importance for biotechnological applications, for example, biological control of plant pathogens, plant growth promotion, or isolation of active compounds [3, 4].

Most studies on rhizospheric and endophytic bacteria and their community structure have been performed by using culture-dependent approaches. Isolation of culturable bacteria is appropriate for functional analysis; however, as a high percentage of naturally occurring bacteria remains in a nonculturable state [5, 6] the use of culture-independent methods provides additional information on the diversity of bacterial communities [7]. Previous studies have analyzed bacterial taxa associated with maize. Some of these studies have been focused on the culturable fraction [810] while others assessed bacterial diversity independently of culture [1114]. However, none of these studies provided an analysis of rhizospheric and endophytic bacterial diversity assessed by both culture-dependent and culture-independent methods. The present work was carried out to gain insight into bacterial diversity associated with the roots of maize seedlings by culture-dependent and culture-independent approaches.

2. Materials and Methods

2.1. Plant Material and Sampling

Grains of maize cv. Monsanto DK684RR2 were sown in plastic pots, 10 cm in diameter ×15 cm high, filled with sandy loam soil obtained from an Argentinean maize field (pH 6.1 in water 1 : 1 w/v, 1.4% organic matter, 86 ppm of nitrates).

All pots were irrigated with sterile distilled water once immediately after sowing to half full soil water holding capacity. Pots with seeds were then incubated with 12 h photoperiod and at 28±2°C until sampling.

Twenty days after sowing, ten seedlings were harvested and their root systems were aseptically separated from the shoots for further analysis. Roots were processed immediately after sampling.

2.2. Analysis of Bacterial Diversity by Culture-Dependent Methods
2.2.1. Isolation and Number Estimation of Bacteria from Maize Roots

Roots of harvested plants (average weight of 3 g) were individually submerged in a volume of sterile phosphate-buffered saline (PBS: NaCl 8 g, KCl 0.2 g, Na2HPO4 1.15 g, KH2PO4 0.2 g, deionized water 1000 ml, pH 7.3) obtaining a 1/10 dilution of the sample. Roots were vigorously vortexed for 5 min to separate adherent soil and serial 10-fold dilutions were prepared in sterile PBS up to 10−8. Aliquots of 0.1 ml were taken from the different dilutions and plated in duplicate on nutrient agar (NA: meat extract 3 g, soy peptone 5 g, NaCl 8 g, agar-agar 15 g) to count colony forming units (CFUs) of the rhizoplane. Afterwards, roots were washed with sterile distilled water and surface-sterilized by immersion in 1.3% sodium hypochlorite (15 min) followed by four rinses with sterile distilled water, an immersion in 2% sodium thiosulfate to remove the residual sodium hypochlorite [15] and, a final rinse with sterile distilled water. Surface sterilization was considered to be achieved by the absence of CFUs in NA plates. Roots were then placed into a chilled sterile mortar under sterile conditions and macerated with a saline buffer (50 mM KH2PO4, 150 mM NaCl, pH: 7.6) cooled at 4°C to isolate bacteria from the root inner tissues. Serial dilutions and plating were performed as previously described for the analysis of bacteria isolated from the rhizoplane.

Plates were incubated at 28°C during 48 h and after this time total count and count per colony type were performed and expressed as CFUs g−1 root. All different colony types isolated were purified on NA tubes and incubated at 28°C for 48 h.

2.2.2. PCR Amplification of 16S rRNA and 16S–23S rRNA Spacer Sequences

Purified bacterial colonies were placed in Eppendorf tubes with sterile distilled water and 1 μl of proteinase K 10 mg ml−1. Tubes were incubated 1 h at 80°C, 1 h at 60°C and finally, 15 min at 95°C. PCR amplications were carried out in 50 μL containing 5 μl of each cell lysate sample, 5 μl of 10x PCR buffer (500 mM KCl; 100 mM Tris-HCl, pH 8.3), 75 mM of MgCl2, 10 mM of each dNTP, 15 pmol of each primer, and 1.25 U of Taq polymerase. The primer sets used for the amplification of 16S rRNA and 16S–23S rRNA internal transcribed spacer sequences (ITS) are shown in Table 1 and the position of analyzed sequences within the rRNA operon is presented in Figure 1. Conditions for the amplification of 16S rRNA sequences consisted of an initial denaturation at 94°C (3 min), 33 cycles of amplification at 94°C (45 s), 56°C (1 min), and 72°C (2 min), plus one final extension step 5 min at 72°C. ITS sequence amplification consisted of an initial denaturation at 94°C (5 min), 25 cycles of amplification at 94°C (1 min), 52°C (2 min), and 72°C (2 min), plus one final extension step 7 min at 72°C. Amplifications were performed in a GeneAmp PCR System 9700 (Perkin Elmer).

2.2.3. Electrophoresis and Imaging

Aliquots (5 μl) of the different amplified samples were combined with loading buffer (Ficoll 15%, glycerol 30%, and 0.25% xylene cyanol) and the preparations were electrophoresed on 1% agarose gels in TAE buffer at 100 V for 1 h. The gels were stained with a solution of ethidium bromide 0.5 mg L−1 [19] and visualized and photographed on a Stratagene Eagle Eye Imaging System. A DNA ladder (0.1–10 Kb; New England BioLabs) was used as the molecular size marker.

2.3. Analysis of Bacterial Diversity by Culture-Independent Methods

The root macerate, obtained after surface-sterilization as previously described, in a sterile chilled mortar was filtered through four sheets of cheesecloth under sterile conditions. A volume of 1 ml of the filtrate was softly poured into chilled centrifuge tubes with Percoll as described for bacteroid isolation from root nodules by Reibach et al. [20]. The following minor modifications were performed to prepare the gradient: 3.5 ml of KH2PO4 0.5 M + 1.5 M NaCl were added to chilled centrifuge tubes, followed by 24.5 ml of Percoll and finally 7 ml of sterile distilled water were added to the tubes (final volume: 35 ml). Tubes were centrifuged at 14000 rpm for 110 min (Sorvall RC-5B Refrigerated High Speed Centrifuge) and after centrifugation the upper fractions of the gradient were carefully discarded. The bacterial fraction was transferred to clean centrifuge tubes and washed twice with the saline buffer to remove rests of Percoll (4000 rpm for 30 min). The bacterial pellet obtained was finally resuspended in the saline buffer, stored at 4°C overnight and then used for DNA extraction.

2.4. DNA Extraction and PCR Conditions

Total DNA was extracted directly from the bacterial fraction using GenomicPrep Cells and Tissue DNA Isolation Kit (Amersham), according to the manufacturer’s procedure and stored at −20°C until use. PCR amplification of the 16S–23S spacer regions was performed using G1 and L1 primers as described for the culture-dependent approach. PCR products were cloned using TOPO-TA Cloning Kit (Invitrogen) and E. coli DH5α chemically competent cells. White colonies were screened for the insertion of ITS sequences using M13 PCR primers. The amplification program consisted of an initial denaturing step at 94°C (3 min), 33 cycles of amplification at 94°C (45 s), 48°C (1 min), and 72°C (2 min), plus one final 5 min extension step at 72°C. All different size bands were sequenced at Macrogen Inc. (Seoul, Korea).

Primers used to amplify 16S rRNA sequences were not used in the culture-independent assay since they presented nonspecific binding to chloroplast sequences. Bacterial identities were confirmed by using both different primer pairs (16S and 16S–23S).

2.5. Analysis of Obtained Sequences

Bacterial taxonomic affiliations were assigned based on the closest match to sequences available at the National Center for Biotechnology Information (NCBI) database (http://www.ncbi.nlm.nih.gov/) using the BLAST algorithm [21]. Sequences of bacteria obtained dependently and independently of culture were deposited in the GenBank nucleotide sequence database (HQ336294-HQ336333).

2.6. Richness and Diversity Indexes

Richness (d) and diversity (H) indexes were calculated based on data corresponding to the total number of bacterial isolates per identified species after sequence analysis. Indexes were also evaluated for bacteria obtained independently of culture by evaluating the total number of clones and the numbers per identified species after sequence analysis, as previously described in Pereira et al. [22], where d=𝑆1𝐶log𝑁,H=𝑁𝑁log𝑁Σ𝑛𝑖log𝑛𝑖.(1)𝑆 = number of different species or groups; 𝑁 = total number of individuals; 𝐶=2.3; 𝑛𝑖 = number of individuals within the 𝑖 species or group.

3. Results

3.1. Total Numbers of Bacterial Isolates

Rhizospheric bacterial populations recovered on NA medium ranged from 9.5×106 to 3.1×107 CFU g−1 root, while populations recovered from the root inner tissues ranged from 5.0×104 to 2.8×106 CFU g−1 root (data not shown). Bacterial numbers for each colony type are shown in Table 2.

3.2. Rhizoplane versus Endophytic Bacterial Isolates

According to the culture-dependant approach some bacteria belonging to the Firmicutes, Lysinibacillus and Paenibacillus genera were located exclusively on the rhizoplane of maize seedlings while others such as Achromobacter isolates (β-Proteobacteria) were recovered from the roots only as endophytes. On the other hand, isolates from Bacillus genus (Firmicutes group) were equally distributed in the external and inner root tissues of maize seedlings. Richness and diversity of culturable bacteria were higher at the rhizoplane (d index: 6.84, H index: 2.10, resp.) than at the root inner tissues (d index: 3.49, H index: 1.48, resp.).

3.3. Culture-Independent versus Culture-Dependent Approach

ITS sequences of representative bacteria isolated from the roots of maize seedlings corresponded to the same taxa as defined by 16S rRNA sequences (data not shown), thus validating a comparative analysis. The analysis of ITS sequences obtained by the culture independent approach showed that the largest fraction of the clones of root endophytic bacteria belonged to γ-Proteobacteria (79.5%), with Enterobacter, Klebsiella, Erwinia and Pseudomonas as prevalent bacterial genera identified. The remaining 20.5 percent of the clones belonged to Bacillus (Table 3). Richness index (d) calculated for the clones was 3.56 while diversity index (H) was 1.81.

According to the bacterial community composition associated with the roots of maize seedlings assessed by culture, 68% of total CFUs belonged to β-Proteobacteria (Achromobacter), 30% to Firmicutes (Bacillus), and only 2% were identified as γ-Proteobacteria (Pseudomonas) (Table 2). Within the group of endophytic bacteria 81.5% of the isolates belonged to Bacillus genus, 11.1% to Achromobacter and the remaining 7.4% was identified as Pseudomonas. Bacterial richness (d) and diversity (H) indexes calculated for endophytic culturable bacteria were slightly lower than the ones obtained in the culture-independent approach (3.49 and 1.48, resp. versus 3.56 and 1.81).

4. Discussion

In the present work culture dependent and culture independent methods were applied to analyze bacterial diversity associated with the roots of maize seedlings. Most frequently reported bacteria linked to the maize crop (Zea mays L.) are Firmicutes (Bacillus), γ-Proteobacteria (Pseudomonas, Enterobacter, Pantoea, and Klebsiella), β-Proteobacteria (Burkholderia), α-Proteobacteria (Rhizobium), and Actinobacteria (Arthrobacter). Less frequent are Paenibacillus (Firmicutes), Achromobacter, Herbaspirillum (β-Proteobacteria), Erwinia (γ-Proteobacteria), Sphingomonas (α-Proteobacteria), other Actinobacteria, and Bacteroidetes [18, 2325]. In agreement with these previous surveys, within the Firmicutes group we identified Bacillus species from both the rhizoplane and root inner tissues of maize seedlings and Enterobacter and Pseudomonas were found as main genera within the γ-Proteobacteria. Bacteria from Bacillus genus have been reported as maize kernel endophytes [10] and have been isolated not only from maize [10, 18, 26] but also from many different plants such as peas, potatoes, conifers, bananas, and bean [2731]. Their activities as biocontrol agents of various fungi [32, 33], as phosphorous solubilizers [34] and as auxin producers [35] have also been reported.

Bacillus and Pseudomonas were previously identified as major genera isolated from 20-day-old maize seedlings using the same cultivar (DK864RR2) and identified based on morphophysiological analysis [8], thus indicating correspondence between morphophysiological and genetic approaches (Table 4).

Achromobacter xylosoxidans, the most abundant species from the root inner tissues of DK684RR2 maize seedlings in the culture dependent approach, has been found previously associated with maize, promoting mycotoxin synthesis inhibition [36] and has also been reported as a growth promoting bacteria of Brassica juncea [37]. The fact that A. xylosoxidans was isolated from the inner tissues of maize seedlings suggest that this bacterium is kernel borne.

Paenibacillus was a minor isolate in our work. It has been previously isolated from the rhizosphere of maize [38] and wheat [39] and has been also found as an endophyte of Eucalyptus, soybean, and Phaseolus vulgaris beans [31, 33, 40].

Estimated bacterial numbers here are within the ranges with the ones reported for maize endophytes in previous surveys [23, 24, 26].

Not all bacterial genera identified by the culture dependent approach were recovered by the culture independent method. Culture-independent methods may be affected by the heterogeneous lysis of different bacterial species and may suffer from primer bias however they supposedly provide more complete data of the bacterial community including viable but not culturable bacteria (VNC) and also nonviable bacteria (NV). Underrepresentation of VNC and NV in cultures is expected and could explain the differences observed with both approaches. Maybe many γ-Proteobacteria remain non culturable. It has been previously reported that some endophytic bacteria can attain a VNC state [41]. On the other hand, it is known that some Bacillus isolates produce a wide spectrum of antibiotics such as bacillomicin, megacine, fengicine, iturin, mycosubtilin and zwittermicin that may possibly inhibit growth of other bacteria in plates. In addition, some Pseudomonas isolates may produce 2,4-diacetilfluoroglucinol, fenacine and pirrolnitrine that may restrict even more the spectrum of microorganisms able to grow in culture.

Gomes et al. [42] also found Proteobacteria as the major group when studying maize rhizospheric bacteria independently of culture. Schmalenberger and Tebbe [14] have also stated that most bacterial sequences obtained from the maize rhizoplane by direct DNA extraction from roots belonged to Proteobacteria. Sun et al. [43] found β-Proteobacteria as the dominant group in root inner tissues when studying rice-associated bacteria, as we did by the culture dependent approach.

Enterobacter has been found previously as an endophyte of maize [10, 26] and other plants such as rice, cotton, papaya and poplar [4447] and its ability to antagonize maize fungal pathogens within Fusarium genus has also been described [48]. Stenotrophomonas has not been reported before in association with maize, but it was found as the most abundant genus present in rice [43]. Endophytic Stenotrophomonas were also obtained from potato, coffee and poplar [28, 46, 49].

We detected here a different bacterial community profile than the one reported by Chelius and Triplett [18] when studying culturable and not culturable bacteria associated with the roots of physiologically mature maize plants (Figure 2). Differences may be explained by variables such as the plant age (102 instead of 20-day-old plants in this study), other conditions related to the growth of plants such as fertilization used in comparison to no fertilization in this work, and the plant material used (cv. Pioneer 3751 instead of cv. Monsanto DK684RR2). It has been reported that the plant genotype and age markedly influence the profile of associated bacteria [42, 5054].

In a recent review on endophytes, it is considered that culture dependent and independent approaches are complementary [55]. Current methods available for the analysis of the effects of different microorganisms on a particular crop rely on isolation and consequently on culture. The knowledge of the wide spectrum of maize associated bacteria will facilitate the search of bacteria capable of exerting antagonism to pathogenic infections, or the detection of biological plant growth enhancers. A new view of maize associated bacteria could be obtained in the future by metagenomic and functional metagenomic analyses and results derived from culture independent approaches like the one performed in the present work could be a basis for such studies.

Acknowledgment

The authors thank J. Quiroz-López for technical help, DGAPA for PAPIIT Grant IN200709-3, UNAM, México, and CONICET PIP 5822/05 from Argentina.