Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/springer-journals/genetic-diversity-and-salt-tolerance-of-bacterial-communities-from-two-OitjjnUoX2
- Publisher
- Springer Journals
- Copyright
- Copyright © 2009 by University of Milan and Springer
- Subject
- Life Sciences; Microbiology; Microbial Genetics and Genomics; Microbial Ecology; Fungus Genetics; Medical Microbiology; Applied Microbiology
- ISSN
- 1590-4261
- eISSN
- 1869-2044
- DOI
- 10.1007/BF03175594
- Publisher site
- See Article on Publisher Site
Abstract
Annals of Microbiology, 59 (1) 25-32 (2009) Genetic diversity and salt tolerance of bacterial communities from two Tunisian soils 1,2 2 1 1 Darine TRABELSI *, Alessio MENGONI , Mohammed ELARBI AOUANI , Ridha MHAMDI , Marco BAZZICALUPO Laboratoire des Interactions Légumineuses-Microorganismes, Centre de Biotechnologie de Borj-Cedria, BP 901, Hammam-Lif 2050, Tunisia Department of Evolutionary Biology, University of Firenze, via Romana 17, I-50125 Firenze, Italy Received 22 September 2008 / Accepted 10 February 2009 ABSTRACT - Microbial ecology studies on arid soils are particularly important for the analysis of biological functions during desertifica- tion. Although much is known about the arid saline flora, few researches have directly compared the bacterial communities of saline arid soils with cultivated soils in Northern Africa. Bacterial communities present in two soils from Soliman (north of Tunisia), one salty and neglected, and the other cultivated, were investigated by using both cultivation dependent and independent approaches. The first approach was used to assess the presence of salt tolerant bacteria and the relationships among salt (NaCl) resistance phenotype, soil characteristics and phylogenetic assignment of strains. Total community analysis, performed by T-RFLP on total DNA, was carried out to investigate the relationships between total community fingerprinting with cultivated isolates diversity. The cultivated isolates from salty soil were more genetically diverse, harbouring strains that can grow at high salt concentration. Moreover, the salt resistance of isolates was found not to be related to any particular phylogenetic group, being widespread among isolates belonging to different bacterial subdivisions. Ribotype richness, evaluated as number of different T-RFLP bands (TRFs), was shown to be higher in the agricultural soil than in the salty soil and several agricultural soil-specific TRFs were detected. Key words: bacterial communities, T-RFLP, salt tolerance, arid soil. INTRODUCTION an important research tool for investigating the relationships and interactions between environmental factors and microbial evolu- Soil is a very complex system that comprises a variety of micro- tion at metabolic and gene levels (Gould and Corry, 1980). habitats with different physicochemical gradients and discon- Current analyses of soil communities are often performed tinuous environmental conditions where microorganisms adapt through culture-independent DNA based methods. Among them, to microhabitats and live together in communities interacting Terminal-Restriction Fragment Length Polymorphism (T-RFLP) is with each other and with other components of the soil biota. an extensively exploited tool used to analyze the genetic diversity Determining factors, such as temperature, pH, or geographic of amplified 16S rRNA genes of a microbial community (Liu et location, that correlate with differences between diverse micro- al., 1997; Marsh, 1999; Dunbar et al., 2001; Berg et al., 2005; bial communities revealed how easily microbes tolerate different Mengoni et al., 2004, 2007). However, cultivation-independent kinds of environmental changes (Girvan et al., 2003; Lozupone approaches cannot provide insights about the phenotypes of the and Knight, 2007). Moreover, a recent meta-analysis found that bacteria within the community, which are the most directly linked the major environmental determinant of microbial community to community function and fitness with respect to environmental composition is salinity rather than extremes of temperature, pH, stressful conditions. In particular, salt resistance of bacteria, is or other physical and chemical factors (Lozupone and Knight, an important parameter for assessing functionality of soil eco- 2007). In several studies, the investigation of microbial com- systems in arid regions. The selection of salt tolerant strains munities in soils undergoing salinization has stirred much of and their subsequent inoculation was a strategy suggested for the attention because of the relevance for agriculture and as improvement of soil fertility especially in relation to plant-associ- an example of “extreme” environment (Liu et al., 1997; Corpet, ated bacteria (Zahran, 1991; Lal and Khanna, 1994; Bouhmouch 1988; Zahran, 1999; Smit et al., 2001; Caton et al., 2004). et al., 2001) and for crop cultivation of the saline lands in arid Bacteria living in unusual extreme conditions might in fact offer regions and of the salt-affected soils. Moreover, interest in microbes living in extreme environments, as arid soils has been fuelled by the discovery of novel taxa in these ecosystems and * Corresponding Author. Phone: +21696786697; by the potential for discovery of valuable enzymes, polymeric E-mail: trabelsi_darine2@yahoo.fr 26 D. TRABELSI et al. materials, and bioactive compounds (Gould and Corry, 1980). heterotrophic bacteria were evaluated as colony-forming units For instances, in recent studies, salt-tolerant actinomycetes have (CFUs). Each CFU determination was performed in triplicate and been explored for their capacity to produce numerous bioactive an average value of bacterial titre was determined. Colonies hav- molecules, particularly antibiotics (Vasavada et al., 2006), and ing different morphology (colour, shape, etc.) were picked up and enzymes as alkaline protease (Thumar and Singh, 2007). Finally, isolated from each sample. Each colony was purified on R2A and bacteria from saline environments may be considered models stored at –20 °C in R2A broth medium containing 50% glycerol. for biological salt tolerance (Lanyi, 1979). These organisms have evolved in saline environments and are able to overcome the Salt tolerance of bacterial isolates. Isolates were incubated deleterious effects of salts up to saturating concentrations. in liquid Luria Broth (LB) medium with increasing concentration In the present work we investigated the bacterial communi- of NaCl (no salt, 100, 300, 600 mM and 1 M). In each tube, ties of two soils, one salty and abandoned, the other one cultivat- 200 μl from 24 h preculture (OD = 1.0) was added to 5 ml ed, which are in close proximity in Soliman (north of Tunisia) by of fresh growth medium. Growth was measured by turbidity using both cultivation dependent and independent approaches. at 620 nm after 24 h incubation at 27 °C in a rotary shaker at The aim was: i) to isolate and characterize salt resistant strains 150 rpm. Growths were performed in triplicate for each isolate. and the cultivable communities in the two soils, and ii) to deter- Bacterial isolates were classified by their growth at different NaCl mine the presence and extent of taxonomic differences between concentration in liquid medium. To confirm result, salt tolerance the total bacterial communities of agricultural and saline soil. was also estimated on solid LB medium with the same NaCl con- centrations and scored for growth after 24 and 48 h incubation at 27 °C. MATERIALS AND METHODS PCR-amplification, ARDRA sequencing and analysis of Soil sampling and characterization. Two soils, one saline and bacterial 16S rDNA. DNA was extracted from each isolate after one agricultural (1 km far from each other) from Soliman (lon- growing in liquid LB medium at 27 °C for 24 h by using the cetylt- gitude, 10°29’30’’E; latitude, 36°41’47’’N), North Tunisia were rimethylammonium bromide (CTAB) method (Ellis et al., 1999). studied. On each site (at about 10 cm depth from “naked” soil) Extracted DNA was quantified after agarose gel electrophoresis soil samples of 500 g were taken in March 2006 with a clean and staining with ethidium bromide (1 μg/ml). Amplification of steel spatula sterilized with ethanol. Three replicates were 16S rDNA was performed as previously described (Mhamdi et al., sampled across the site for both soils, at a distance of about 2002) using primer (27f, 5’ GAGAGTTTGATCCTGGCTCAG, and 35 m between each sampling point. The agricultural soil was 1495r, 5’ CTACGGCTACCTTGTTACGA). For the ARDRA (Amplified a fallow field with sandy soil type previously used for intense Ribosomal DNA Restriction Analysis) (Vaneechoutte et al., 1992), production of onion (Allium cepa), olive (Olea europea) and 1 μg of the amplified 16S rDNA was digested with 2 units of barley (Hordeum vulgare). The saline soil was clay, without the restriction enzyme AluI (Invitrogen, San Giuliano Milanese, agricultural activity, with a few vegetation (Medicago spp.) and Italy) in a total volume of 15 μl at 37 °C for 3 h. The reaction some halophytes Arthrocnemum indicum (Willd.) Moq., Suaeda mixture was then incubated at 65 °C for 10 min to inactivate the fruticosa Forsk., and Sesuvium portulacastrum L. Soil samples restriction enzyme. Reaction products were separated by agarose -1 were transported to the laboratory in a sealed polyethylene bag gel (2.5% w/v) electrophoresis in TAE buffer run at 10 V cm -1 at ambient temperature and stored at 4 °C in sterile containers, for 1.5 h and stained with 1 μg ml ethidium bromide. For the until analysis the next day. sequencing reaction of isolates from selected ARDRA groups the Moisture content, pH, organic matter, texture and electrical amplified 16S rDNA was purified from salts and primers with the conductivity of the soil samples were determined. Soil texture MinElute PCR purification kit (Qiagen Inc., Chatsworth, CA, USA) was determined by wet-sieving and sedimentation using the according to the manufacturer’s instructions. Direct sequenc- Kohn-Pipette method after organic C destruction with H O and ing of amplified 16S rDNA, of three isolates per ARDRA group, 2 2 chemical dispersion using Na P O (Hartge and Horn, 1992). was performed with the 27f primer on an ABI310 automated 4 2 7 The oven dry method was used for moisture content, weigh- sequencer (Applied Biosystems, Foster City, CA, USA) using the ing samples before and after drying. The percent of weight Big Dye Terminator kit, version 2 (Applied Biosystems). Obtained loss was used as the percent moisture content of the sample. sequences were matched against the GenBank-EMBL-DDBJ data- The pH of each sample was determined by using a pH meter base with the BLAST program (Altschul et al., 1997) to provide method: ten grams of soil were shaken in 25 ml of de-ionized taxonomic interpretation (that is to identify the bacterial groups) water for 10 min and then allowed to settle for 30 min. A pH of the selected ARDRA groups. meter was then used to determine the pH of the supernatant Shannon index of diversity was calculated by using the water without disturbing the settled soil. The amount of car- on-line calculator present at http://www.changbioscience. bon or organic matter was determined using the weight loss com/genetics/shannon.html. on ignition test (Dean 1974). For the electrical conductivity, direct fields measurements were taken with the Veris EC sensor T-RFLP analysis. Total soil DNA from the same samples used (Veris, Inc., Salina, Kan). for the isolation of bacteria was extracted as previously described (Yeates et al., 1998). DNA was quantified by agarose gel eletro- Isolation of bacteria. Triplicate subsamples (1 g) were taken phoresis and by spectrophotometric reading. PCR reactions from each soil sample, kept in sterile plastic boxes at 4 °C for a were performed in 50 μL volume containing 20 ng of template few hours before using. Then, homogenized in 10 ml of physi- DNA and 2 U of Taq DNA polymerase (Invitrogen) using primer ological solution and 100 μl of 10-fold serial dilutions of the sus- 27f labeled with 6-carboxyfluorescein (FAM) and primer 1495r pension were spread in triplicate on low nutrient R2A medium as previously reported (Mengoni et al., 2002). Amplified 16S for heterotrophs (Difco) (Øvreås and Torsvik, 1998). In order to rDNA was digested separately with 20 U of MspI, RsaI and TaqI -1 inhibit the growth of fungi, 300 μg ml cycloheximide was added. (Promega, New England Biolabs, Beverly, MA, USA) restriction Plates were incubated at 27 °C for 2 days; after titres of aerobic enzymes for 3 h at 37 °C (or 65 °C for TaqI). A 200 ng aliquot of Ann. Microbiol., 59 (1), 25-32 (2009) 27 the digested product was resolved by capillary electrophoresis on and higher, 8 x 10 CFU/g for the saline soil. Plates were visually an ABI310 Genetic Analyzer (Applied Biosystems) using ROX 500 inspected and, on plates of the same soil site, colonies with dif- (Applied Biosystems) as a size standard. Fragment sizes from 35 ferent morphology were picked up and re-isolated on R2A agar to 500 bp were considered for profile analysis. plates. A total of 208 colonies were selected. In general, colonies TRFs, with appropriate size, derived from MspI restric- from agricultural soil appeared more similar to each other, and tion digestion were cloned by using an adaptor-based method this is why we selected more colonies from saline than from agri- (Mengoni et al., 2002) and MiCA3 web tool (http://mica.ibest. cultural soil (140/68). uidaho.edu/) to allow their taxonomic interpretation. Three clones, for each selected TRF size, were selected for sequencing Salt tolerance of bacterial isolates (performed as previously described). Growth of triplicate culture of bacterial isolates at various salt concentrations was evaluated (Table 2). As there was no differ- Statistical analysis of T-RFLP profiles. Analysis of T-RFLP ence in the results obtained on solid and liquid medium, data profiles was performed with the software GelComparII 3.0 in Table 2 refer only to liquid cultures. Among 208 isolates, (Applied Maths, Kortrijk, Belgium) directly from chromatogram five main phenotypic classes were scored. Phenotypic class P5 files. Only fragments with fluorescence intensity >50 arbitrary was the most abundant and comprised isolates growing up to units of fluorescence were considered. Alignment of the profiles 1 M NaCl. However, no correlation was found between OTUs was performed and a binary matrix, in which the presence or (see below and Table 3) and salt-tolerant phenotype, each OTU absence of peaks was scored as string of ones or zeros. Matrices including different phenotypes, that is each phenotypic class from the three different restriction digestions were linearly com- for salt tolerance included isolates from different taxonomic bined to obtain a unique binary vector for each sample. The groups. However, OTU 6, 35 and 36 isolated from the saline soil matrix of binary vectors was then used to compute the communi- samples, were included in the most salt resistant phenotype ty similarity values based on the Jaccard coefficient of similarity, (P5). The distribution of salt tolerant phenotypes between the which only takes into account band sharing between vectors. The two soil types showed that a higher proportion of isolates from matrix of mean Jaccard similarity was then used for Unweighted saline soil was found to be resistant to high NaCl concentrations Pair Group Method with Arithmetic mean (UPGMA) clustering and compared with those from agricultural soil (one-way ANOVA, P Principal Component Analysis (PCA) using the modules present in < 0.0001). The distribution of isolates resistant to 600 mM and NTSYSpc 2.02 (Rohlf, 1990). 1 M NaCl (76 isolates), revealed that they were represented by To assess the amount and the significance of the genetic 17 major OTUs. differences between saline and agricultural soil community pro- files, the Analysis of Molecular Variance (AMOVA, Excoffier et 16S rDNA profiling al., 1992) was applied to the T-RFLP matrix described above. 208 bacterial isolates were taxonomically typed by using ARDRA This test, originally developed to infer the genetic structure of and 16S rDNA sequencing; these isolates were also phenotypi- populations (Mengoni and Bazzicalupo, 2002), was here applied cally characterized for NaCl tolerance. as an alternative to the classical ANOVA (Miller, 1997) in deriving Fifty different ARDRA restriction patterns (Operational a measure of genetic differentiation between bacterial commu- Taxonomic Units, OTUs or haplotypes) were found after diges- nities. The software ARLEQUIN 3.1 (Excoffier et al., 2006) was tion of amplified 16SrDNA with AluI. Thirty of the 50 OTUs were used for AMOVA computation and statistical significance was represented by less than four isolates. The remaining twenty assessed after 100000 permutations. OTUs accounted for 160 (more than 74%) of the total number of isolates, and are referred to as “major OTU”, being represented by a minimum of 4 isolates. RESULTS In Fig. 1 the number of OTUs of each soil samples is reported. Twenty-nine OTUs were found among 68 isolates from agricul- Soil characteristics and bacterial titres tural soil samples and 39 OTUs were scored for 140 isolates from Physical and chemical characteristics of the soil samples are saline soil samples. Within the 20 major OTUs, six were exclu- reported in Table 1. As expected, a major difference between soil sively recovered from saline soil (OTU6, OTU7, OTU8, OTU15, types was found in the electrical conductivity (EC) with the saline OTU31, and OTU35) and just one was recovered from agriculture soil having far higher values than the agricultural one. Bacterial soil (OTU26). Among these 20 major OTU, 3 isolates per OTU titres were on average 1.5 x 10 CFU/g for the agriculture soil were chosen for a partial 16SrDNA sequencing of the first 400- TABLE 1 - Soil physical and chemical characteristics and bacterial titres Samples Agricultural soil Saline soil pH 8.46 ± 0.13 8.93 ± 0.13 Moisture content (%) 32.33 ± 0.65 54 ± 4 Electrical conductivity (EC) (mmho/cm)* 0.7 ± 0.11 4.66 ± 1.72 Organic matter (%) 0.75 ± 0.4 1.1 ± 0.16 Carbon (%) 0,53 ± 0.23 0.66 ± 0.06 Texture Sandy Clay 7 7 Bacterial titres (CFU/g of soil) 1.5 ± 0.97 x 10 8.13 ± 1.36 x 10 * 1 dS/m = 1 mmho/cm (Bauder et al., 2004). Values are means ± Standard deviation of triplicate measures. 28 D. TRABELSI et al. TABLE 2 - Salt resistance phenotypes of the 208 bacterial isolates Phenotype code Number of isolates NaCl* 100 mM 300 mM 600 mM 1 M P1 17 - - - - P2 28 + - - - P3 37 + + - - P4 50 + + + - P5 76 + + + + * Salt tolerance in liquid cultures is designated are as follows: + = good growth, as in the absence of salt; - = no growth. Growth was measured by means of OD at 620 nm after 24 h incubation at 27 °C in a rotary shaker at 150 rpm. Growths were performed in triplicate for each isolate. Bacterial isolates were classified by their growth at different NaCl concen- tration. The OD at 620 nm ≥ 1.0 was considered as good growth. TABLE 3 - Shannon index estimates of bacterial diversity and number of ARDRA OTUs for isolates from agricultural and saline soil Sample Number of Shannon Number of major Shannon index Major OTUs index OTUs* OTUs Total agricultural 29 2.95 13 2.27 Total saline 39 3.09 20 2.52 * Major OTUs are defined as those represented by a minimum of 4 isolates. 500 bases from 27f primer annealing site in order to phyloge- gave the lowest (17). The largest number of TRFs was observed netically assigning them to bacterial taxonomic groups. Results in the agricultural soil samples (70 TRFs). Among the 86 TRFs, obtained (Table 4) allowed to recognize 7 bacterial genera, the nine were ubiquitous in all soil types, six TRFs (produced after most represented being the genus Bacillus (10 out of 20 OTUs). restriction digestion with MspI) were exclusively present in the agriculture soils, whereas no TRFs were exclusive to the saline T-RFLP community profiles soils. The six TRFs, identified as specific of agricultural soil type T-RFLP analysis of 16SrRNA gene fragments amplified from total were taxonomically interpreted after cloning and sequencing and DNA was used to evaluate and compare total bacterial commu- were assigned to Bacillus (EU122184), Rubrobacter (EU122187), nity diversity in the above described soils. The three restriction Betaproteobacteria (EU122185 and EU122186), Acidobacteria endonucleases applied to 16S rDNAs amplicons yielded a total (EU122188) and Actinobacteria (EU122189) (Table 5). A match- of 86 different peaks or TRFs (Terminal Restriction Fragments). ing of TRFs with RDP release 9.60 sequences with MiCA3 web tool MspI produced the highest number of peaks (45), while TaqI gave similar results for most the TRFs. FIG. 1 - Distribution of different bacteria isolates among the major OTUs in the two soil types, and putative genus assignment of the major ARDRA types. B, Bacillus; St, Stenotrophomas; P, Pseudomonas; M, Microbacterium; S, Sphinogomonas; Br, Bordetella; R, Rhizobium. Ann. Microbiol., 59 (1), 25-32 (2009) 29 TABLE 4 - Taxonomic interpretation of the major OTUs OTU GenBank accession Best matched sequence Number of isolates ** number number (accession number, % of similarity)* OTU1 EU122164 Bacillus sp. (AF539671, 89%) 48 OTU3 EU122165 Stenotrophomonas sp. (AJ011332, 92%) 10 OTU4 EU122166 Paenibacillus sp. (AY336562, 91%) 4 OTU6 EU122167 Bacillus sp. (AF157696, 96%) 4 OTU7 EU122168 Bacillus sp. MK03 (AB062678, 83%) 6 OTU8 EU122169 Bacillus sp. MP2 (DQ985283, 92%) 9 OTU9 EU122170 Bacillus sp. (AF295302, 99%) 6 OTU10 EU122171 Pseudomonas (DQ862549, 94%) 12 OTU12 EU122172 Bacillus sp. (DQ275178, 94%) 4 OTU14 EU122173 Sphingomonas sp. (AB033944, 99%) 4 OTU15 EU122174 Bacillus sp. (EF152359, 95%) 4 OTU16 EU122175 Pseudomonas sp. (EF528294, 94%) 6 OTU21 EU122176 Microbacterium sp. (AB271048, 94%) 13 OTU22 EU122177 Pseudomonas sp. (EF424136, 96%) 4 OTU25 EU122178 Bacillus sp. (EF656456, 98%) 4 OTU26 EU122179 Bacillus sp. (DQ993678, 92%) 4 OTU28 EU122180 Bordetella sp. (EF442019, 96%) 4 OTU31 EU122181 Microbacterium sp.(AY974047, 94%) 4 OTU35 EU122182 Bacillus sp. ( EF638801, 97%) 4 OTU36 EU122183 Rhizobium sp. (EF070127, 96%) 6 * The bacterial group shown is that comprising the best matched sequences after BLAST search. In brackets the acces- sion numbers and percentage of similarity of the best matched sequences are reported. ** Number of isolates with ARDRA pattern identical to that of the sequenced strain. From each major OTU, three isolates were sequenced. To assess the distribution of ribotypic diversity between the 45% of similarity, respectively). The first principal component of two soil types, Principal Component Analysis (PCA), UPGMA clus- PCA, in agreement with UPGMA clustering, separated saline soil tering (Fig. 2) and AMOVA were used. Both analyses recognized a samples (in particular SS1 and SS3) from the agricultural soil separation between the community profiles of the two soil types samples. In the second principal component, the heterogeneity and correctly clustered T-RFLP profiles of soil samples according within soil types was shown. To quantitatively assess the degree to their origin. In particular UPGMA showed a large separation of heterogeneity within soil samples and between soil types, an of saline vs. agricultural soil (32% of similarity between the two AMOVA was performed in order to detect an uneven distribution clusters). The most similar samples were SS1 and SS3 (63% of ribotypic T-RFLP variance with respect to the different soil of similarity). Triplicate samples from saline soil type appeared type. Obtained results showed that soil heterogeneity within soil slightly more heterogeneous than agricultural ones (42.5 and type was the most important factor affecting variance (within soil TABLE 5 - Taxonomic interpretation of TRFs specific of agricultural soil type* Size (nt) Best matched sequence MiCA 3 matching*** ** (accession number, % of similarity) 167 Bacillus sp. (AY211124, 99%) Bacillus, Chloroflexi, Helicobacter, Finegoldia, Brevibacter, Staphylococcus, Corynebacterium 494 Uncultured betaproteobacterium (EF219646, 95%) Uncultured betaproteobacteria, Acinetobacter, Enterobacter, Pantoea, Escherichia coli 488 Uncultured betaproteobacterium (AM691112, 97%) Acidovorax, Comamonas, Uncultured betaproteobacteria 286 Uncultured Rubrobacter (AY571811, 99%) Uncultured bacteria 196 Uncultured Acidobacteria (EF664827, 98%) Halomonas, Streptomyces 297 Uncultured actinobacterium (EF651774, 100%) Uncultured bacteria * The 6 TRFs, due to MspI digestion of amplified 16SrDNA, were taxonomically interpreted after cloning and MiCA3 web tool. ** The bacterial group shown is that comprising the best matched sequences after BLAST search. In brackets the accession numbers and percentage of similarity of the best matched sequences are reported. *** For the MiCA3 matching, the names of the matched organisms with Ribosomal Database release 9.60 (comprising 511847 sequences) is reported. 30 D. TRABELSI et al. FIG. 2 - A: UPGMA dendrogram computed from Jaccard similarity matrix among T-RFLP profiles of amplified 16S rDNA genes from the different soils. Scale bar indicates Jaccard similarity values. B: Principal component analysis of the T-RFLP profiles of agricul- tural (SA) and saline (SS) soil types. Dim-1: first principal component, Dim-2: second principal component. The percentage of variance due to the displayed component is reported. type 68.57%) (Table 6). However, in agreement with previous in saline habitats (Rosenberg, 1983; Rodriguez, 1991). Among analyses, a clear separation of T-RFLP profiles with respect to soil the 50 ARDRA groups, 6 of them were found in saline soil only type was found, estimating the among soil variance component and were assigned to the genera Bacillus and Microbacterium. to more than 30% (P < 0.0001) of the total genetic variance. Previous reports indicated that Gram-positive bacteria are well represented in saline habitats, and members of the genera Bacillus and Micrococcus are dominant among other Gram- DISCUSSION positive bacteria in saline soils (Zahran et al., 1992). Concerning the salt resistant phenotype of isolates, the high- Salinity is one of the main factors shaping bacterial community er proportion and extent of salt resistance was retrieved within composition. Here we isolated heterotrophic bacteria from two saline soil. Interestingly, salt tolerance phenotypes were not soil types in Northern Tunisia, one undergoing salinity and deser- related to any particular ARDRA group. This fact may suggest the tification, the other one cultivated, aiming to determine their presence of a relatively high number of molecular/genetic ways genetic diversity and salt resistant phenotypes. Moreover we salinity tolerances can evolve and spread into the community. investigated total community composition aiming at character- Even though salt-tolerant strains occurred in both saline and non izing the differences between the bacterial communities present saline sites, the isolates showing the highest salt tolerance were in the two soils. found in saline soil. It is worth of note that, among the isolates Obtained results on cultivated bacteria showed that, in our identified as belonging to the genus Rhizobium, all were shown conditions, saline soil had the highest titres and the highest diver- to be resistant to 600 mM and 1 M NaCl. A number of studies sity (as Shannon index) for heterotrophic cultivable bacteria. have shown that osmoadaptation in Rhizobium appears to be On the 208 bacterial isolates, 50 ARDRA restriction patterns atypical compared to that found in many enteric bacteria (for a were found after digestion of amplified 16SrDNA. 16SrDNA review see Miller and Wood, 1996) and rhizobia have been previ- sequencing of main ARDRA groups showed that bacterial iso- ously reported to colonize the saline soil and establish effective lates, for both saline and agricultural soil, consisted mainly of symbiosis with plants (Zahran, 1999). organisms taxonomically close to the genera Bacillus, by far The total community analysis, performed by T-RFLP, showed the most frequent, Pseudomonas, Microbacterium, Rhizobium, a general clustering of bacterial communities profiles accordingly Bordetella, Sphingomonas and Stenothrophomonas. These gen- with soil types. However, differences between profiles, espe- era were found to dominate the cultivated bacterial communi- cially for those of the agricultural soil, were found. Concerning ties in several other arable and forest soils (Smit et al., 2001; the ribotypic diversity, the agricultural soil showed the high- Axelrood et al., 2002; Zhou et al., 2004) and were reported also est values with 70 TRFs against 45 found in saline soil T-RFLP TABLE 6 - Analysis of molecular variance (AMOVA) for 6 soil samples combined in two different soil types (saline and agricultural) using 86 T-RFLP bands* Source of variation d.f. Sum of squares Variance Total (%) P Among soil types 1 31.667 6.11111 31.43 < 0.0001 Within soil types 4 53.333 13.33333 68.57 < 0.0001 Total 5 85.000 19.44444 * The 6 samples were distributed between 2 soil types. AMOVA was performed attributing the grouping of samples accord- ing to their site of sampling that is AS1, AS2, AS3 for agricultural soil type and SS1, SS2, SS3 for saline soil type. The total variance derived from T-RFLP profiles was attributed to the three hierarchical partitions: first lane, among soil types; second lane, within soil types. Data show the degrees of freedom (d.f.), the sum of squared deviation, the variance com- ponent estimate, the percentage of total variance contributed by each component and the probability of obtaining a more extreme component estimate by chance alone (P-value). The P-values were estimated computing 100000 permutations. Ann. Microbiol., 59 (1), 25-32 (2009) 31 profiles. Agricultural activities, such as tillage, intercropping, against plant pathogenic fungi. FEMS. Microbiol. Ecol., 51 rotation, drainage, and fertilizers, can have significant implica- (2): 215-229. tions for the microorganisms present in the soil (Øvreås and Bouhmouch I., Brhada F., Fillali M.A., Aurag J. (2001). Selection Torsvik, 1998). In the saline soil type, though some TRFs are of osmotolerant and effective strains of Rhizobiaceae for inoc- specific for a particular sample, no TRFs were found common ulation of common bean (Phaseolus vulgaris) in Moroccan for all saline soil samples and exclusive of the saline soil type. saline soil. Agronomie, 21 (6): 591-599. On the contrary, six TRFs exclusive of the agricultural soil were Buckley D.H., Schmidt T.M. (2003). Diversity and dynamics detected and identified as belonging to bacterial domains like of microbial communities in soils from agro-ecosystems. Acidobacteria, Betaproteobacteria and Actinobacteria. Such phy- Environ. Microbiol., 5 (6): 441-452. logenetic groups were found to dominate the bacterial commu- Caton T.M., Witte L.R., Ngyuen H.D., Buchheim J.A., Buchheim nity structures in several agricultural and forest soils (Galinski M.A., Schneegurt M.A. (2004). Halotolerant aerobic het- and Tindall, 1982). Although members of Acidobacteria were erotrophic bacteria from the great salt plains of Oklahoma. especially abundant in arable soils (Buckley and Schmidt, 2003), Microb. Ecol., 48 (4): 449-462. Smit et al. (2001) suggested that the ratio between the number Corpet F. (1988). Multiple sequence alignment with hierarchical of Proteobacteria and Acidobacteria might be indicative of the clustering. Nucleic Acids Res., 16 (22): 10881-10890. trophic level of the soil. Consequently, future analyses on those soils could specifically address this point by clone library and Dunbar J., Ticknor L.O., Kuske C.R. (2001). Phylogenetic spe- FISH analyses. Nevertheless, the functional role of the members cificity and reproducibility and new method for analysis of of Acidobacterium division in soil is still unknown (Galinski and terminal restriction fragment profiles of 16S rRNA genes Tindall, 1982; Felske et al., 2000). Within agricultural soil sam- from bacterial communities. Appl. Environ. Microbiol., 67 ples, the largest number of TRFs was found for SA3 and could (1): 190-197. be related with the lower organic matter and low quantities of Dean W.E. (1974). Determination of carbonate and organic mat- humic acids present there (Paul and Clark, 1989), in fact, it has ter in calcareous sediments and sedimentary rocks by loss on been reported that humic acids may introduce a bias toward low ignition; comparison with other methods. J. Sediment Petr., diversity estimates (LaMontagne et al., 2002). However, to fully 44: 242-248. address the relationships between soil parameters and bacterial Ellis R.J., Thompson I. P., Bailey M.J. (1999). Temporal fluctua- community profiles, more samples distributed along a gradient tions in the pseudomonad population associated with sugar are needed to perform statistical analyses of association and beet leaves. FEMS Microb. Ecol., 28 (4): 345-356. determine the quantitative effect of soil chemical parameters Excoffier L., Smouse P.E., Quattro M. (1992). Analysis of molecu- over community composition in these soils. lar variance inferred from metric distances among DNA hap- In conclusion, this paper reports an initial analysis showing lotypes: Application to human mitochondrial DNA restriction possible association between the presence of some bacterial data. Genetics, 131 (2): 479-491. flora and soil salinity. In particular, genetic diversity on T-RFLP Excoffier L., Guillaume L., Schneider S. (2006). An Integrated pattern seems to be related to soil parameters. Moreover, in this work, a number of bacterial isolates resistant to high NaCl Software Package for Population Genetics Data Analysis. concentrations (1 M) has been isolated which can have a great Manual Arlequin ver 3.1. impact on the performances of vegetation of saline arid soils. Felske A., de Vos W.M., Akkermans A.D.L. (2000). Spatial distri- In particular plant-associated bacteria as those belonging to the bution of 16S rRNA levels from uncultured acidobacteria in genus Rhizobium have been found which could be useful for land soil. Lett. iAppl. Microbiol., 31: 118-122. restoration managements. Galinski E.A., Tindall B.J. (1982). Biotechnological prospects for halophiles and halotolerant micro-organisms. In: Herbert Acknowledgments R.H., Sharp R.J., Eds, Molecular Biology and Biotechnology of This work was partially supported by University of Firenze (Italy) Extremophiles, Blackie & Son, Glasgow, pp. 76-114. “Scambi Culturali e Cooperazione Interuniversitaria Internazionale Girvan M.S., Bullimore J., Pretty J.N., Osborn A.M., Ball A.S. (cap.f.s.2.16.04)”. (2003). Soil type is the primary determinant of the composi- tion of the total and active communities in arable soils. Appl. Environ. Microbiol., 69 (3): 1809-1900. REFERENCES Gould G.W., Corry J.E.L., Eds (1980). Microbial Growth and Survival in Extremes of Environments, Academic Press, New Altschul S.F., Madden T.L., Schäffer A.A., Zhang J., Zhang Z., York, pp.215-226. Miller W., Lipman D.J. (1997). Gapped BLAST and PSI-BLAST: A new generation of protein database search programs. Hartge K.H., Horn R. (1992). Die physikalische Untersuchung von rd Nucleic Acids Res., 25 (17): 3389-3402. Boden. 3 edn., Ferdinand Enke, Stuttgart. Axelrood P.E., Chow M.L., Radomski C.C., McDermott J.M., Davies Lal B., Khanna. S. (1994). Selection of salt-tolerant Rhizobium J. (2002). Molecular characterization of bacterial diversity isolates of Acacia nilotica. World J. Microbiol. Biotechnol.,10 from British Columbia forest soils subjected to disturbance. (6): 637-639. Can. J. Microbiol., 48 (7): 655-674. LaMontagne M.G., Michel F.C., Holden P.A., Reddy C.A. (2002). Bauder T.A., Waskom R.M., Davis J.G. (2004). Irrigation Water Evaluation of extraction and purification methods for obtain- Quality Criteria Irrigation Crop Sciences No. 0.506. ing PCR-amplifiable DNA from compost for microbial commu- nity analysis. J. Microbiol. Methods, 49 (3): 225-264. Berg G., Krechel A., Ditz M., Sikora R.A., Ulrich A., Hallmann J. (2005). Endophytic and ectophytic potato-associated bacte- Lanyi J.K. (1979). Salt tolerance in microorganisms. In: Hollander rial communities differ in structure and antagonistic function A., Aller J.C., Epstein.E., San Pietri A., Zaborsky O.R., Eds, 32 D. TRABELSI et al. The Biosphere Concept: An Approach to the Utilization of Paul E.A., Clark F.E. (1989). Soil Microbiology and Biochemistry. Under Exploited Resources. Plenum Press, New York, pp. Academic Press Inc, San Diego. 217-232. Rodriguez V.F. (1991). Biotechnological potential of halobac- Liu W.T., Marsh T.L., Cheng H., Forney L.J. (1997). Characterization teria. In: Danson M.J., Hough D.W., Lunt G.G., Eds, The of microbial diversity by determining terminal restriction Archaebacteria: Biochemistry and Biotechnology, Portland fragment length polymorphisms of genes encoding 16S Press, London and Chapel Hill. rRNA. Appl. Environ. Microbiol., 63 (11): 4516-4522. Rohlf F.J. (1990) Numerical Taxonomy and Multivariate Analysis Lozupone A.C., Knight R. (2007). Global patterns in bacterial System. Version 2.02. Exeter Software, New York. diversity. PNAS, 104 (27): 11436-11440. Rosenberg A. (1983). Pseudomonas halodurans sp. nov., a halo- Marsh T.L. (1999). Terminal restriction fragment length poly- tolerant bacterium. Arch. Microbiol., 136 (2): 117-123. morphism (T-RFLP): an emerging method for characterizing Smit E., Leeflang P., Gommans S., Broek V.D.J., Mil V.S., Wernars diversity among homologous populations of amplification K. (2001). Diversity and seasonal fluctuations of the domi- products. Curr. Opin. Microbiol., 2 (3): 323-327. nant members of the bacterial soil community in a wheat Mengoni A., Bazzicalupo M. (2002). The statistical treatment field as determined by cultivation and molecular methods. of data and the analysis of molecular variance (AMOVA) in Appl. Environ. Microbiol., 67 (5): 2284-2291. molecular microbial ecology. Ann. Microbiol., 52: 95-101. Thumar J.T., Singh P.S. (2007). Secretion of an alkaline pro- Mengoni A., Grassi E., Bazzicalupo M. (2002). A cloning meth- tease from a salt tolerant and alkaliphilic, Streptomyces cla- od for the taxonomic interpretation of T-RFLP patterns. vuligerus strain mit-1. Br. J. Microbiol., 38: 766-772. Biotechniques, 33 (5): 990-992. Vaneechoutte M., Rossau R., De Vos P., Gillis M., Janssens D., Mengoni A., Grassi E., Barzanti R., Biondi E.G., Gonnelli C., Kim Paepe N. (1992). Rapid identification of the bacteria of the C.K., Bazzicalupo M. (2004). Genetic diversity of bacterial Comamonadaceae with amplified ribosomal DNA-restriction communities of serpentine soil and of rhizosphere of the analysis (ARDRA). FEMS Microbiol. Lett., 93 (3): 227-234. nickel-hyperaccumulator plant Alyssum bertolonii. Microb. Vasavada S.H., Thumar J.T., Singh S.P. (2006). Secretion of a Ecol., 48 (2): 209-217. potent antibiotic by salt-tolerant and alkaliphilic actinomycete Mengoni A., Giuntini E., Bazzicalupo M. (2007). Application of ter- Streptomyces sannanensis strain RJT-1. Curr. Sci., 91 (10): minal-restriction fragment length polymorphism for molecu- 1393-1397. lar analysis of soil bacterial communities. In: Varma A., Yeates C., Gillings M.R., Davison N., Altavilla N., Veal D.A. (1998). Oelmuller R., Eds, Soil Biology, Vol. 11, Advanced Techniques Methods for microbial DNA extraction from soil for PCR ampli- in Soil Microbiology, Springer-Verlag, Berlin, Heidelberg, pp. fication. Biol. Proced., 1 (1): 40-47. 295-305. Zahran H.H. (1991). Conditions for successful Rhizobium- Mhamdi R., Laguerre G., Aouani M.E., Mars M., Amarger legume symbiosis in saline environments. Biol. Fertil. Soils, N. (2002). Different species and symbiotic genotypes of 12 (1): 73-80. Weld rhizobia can nodulate Phaseolus vulgaris in Tunisian soils. FEMS Microb. Ecol., 41: 77-84, DOI 10.1016/S0168- Zahran H.H., Moharram A.M., Mohammad H.A. (1992). Some eco- 6496(02)00264-7 logical and physiological studies on bacteria isolated from salt- affected soils of Egypt. J. Basic Microbiol., 32 (6): 405-413. Miller K.J., Wood J.M. (1996). Osmoadaptation by rhizosphere bacteria. Annu. Rev. Microbiol., 50: 101-136. Zahran H.H. (1999). Microbiol Rhizobium-legume symbiosis and nitrogen fixation under severe conditions and in an arid cli- Miller R.G.Jr. (1997). Beyond ANOVA. Basics of Applied Statistics. Chapman & Hall, London, UK. mate. Mol. Biol. Rev., 63 (4): 968-989. Øvreås L., Torsvik V. (1998). Microbial diversity and commu- Zhou J., Xia B., Huang H., Palumbo A.V., Tiedje J.M. (2004). nity structure in two different agricultural soil communities. Microbial diversity and heterogeneity in sandy subsurface Microbial. Ecol., 36 (3): 303-315. soils. Appl. Environ. Microbiol., 70 (3): 1723-1734.
Journal
Annals of Microbiology – Springer Journals
Published: Nov 24, 2009