1

1Title: Microbial diversity of landslide soils assessed by RFLP and SSCP fingerprints.
2
3Key words: landslide soils/RFLP/SSCP
4
5Authors: Marco Guida1, Paolo Losanno Cannavacciuolo1, Mara Cesarano2, Marco Borra3, Elio Biffali3,

6Giovanna De Mieri4, Raffaella DAlessandro4 and Bruna De Felice4.

7
81 Department of Biology, University Federico II of Naples, Italy, via Cinthia ed. 7 I-80134,

9Naples, Italy
10
112 DiSTAR - Department of Earth Sciences, Environment and Resources, University Federico II of Naples,
12Italy, Via Mezzocannone, 8 I-80134, Naples, Italy
133 Zoological Station Anton Dohrn, Villa Comunale, 80121 Napoli, Italy
14
154 DISTABIF- Department of Environmental, Biological and Pharmaceutical Science and Technologies,
16University of Naples II, Via Vivaldi 43, 81100 Caserta, Italy
17
18
19
20
21
22
23
24
25
26Corresponding author
27Prof. Bruna De Felice, PhD
28Department of Environmental, Biological and Pharmaceutical Science and Technologies,
29University of Naples II,
30Via Vivaldi 43,
3181100 Caserta, Italy
32Tel: ++39-823-274543
33Fax: ++39-823-274571
34e-mail: bruna.defelice@unina2.it
35
36

3
37
38

Abstract

39Landslides are a significant component of natural disasters in most countries of the world. Understanding

40these destructive phenomena through the analysis of possible correlation between microbial communities
41and the alteration of the soil responsible for landslides is important in order to reduce their negative
42consequences. To address this issue, bacterial and fungal communities in soils triggering landslides in
43Termini-Nerano and Massa Lubrense-Nerano (Naples, Italy) were analysed by genetic profiling techniques.
44Fingerprints were generated by single strand conformation polymorphisms (SSCP) and random amplified
45polymorphic DNA (RAPD). The microbial community, in both soil types, was enriched in species which
46could contribute to degradation process occurring during landslides, forming biofilms and leading to the
47transformation or the formation of minerals. Indeed, some of the identified bacteria, were found to favor the
48transformation of clay minerals. These findings suggest a possible relationship between bacterial and fungal
49community colonizing soils and taking place landslides.
50
51
52
53
54
55
56Keywords: microbial community; molecular analysis; biofilm; soil transformation.

5
57
58

Introduction

59
60Landslides are caused by movements of earth secondary to altered stability of the soil and underlying

61materials. Landslides occur as a consequence of several interacting factors, such as man-induced, climatic,
62earthquakes or other natural events (Keefer 2002; Baioni 2011; Huggel et al. 2012).
63Despite the progresses reached in the development of landslide risk assessment in recent years (Xie et al.

642003; Jamaludin et al. 2006; Pardeshi et al. 2013), landslides still remain a major geological hazard in
65several areas of the world, leading to human lives loss and to economical burden both for governments and
66for resident in affected areas.
67The complexity of phenomena possibly leading to landslide occurrence is often focused on the

68investigation of macroscopic triggering factors, such as the ones mentioned above. However, the role of
69microbial community resident in soils in determining soil properties and soil mechanical deterioration has
70been previously supposed (Radina 1973; Futagami et al. 2010).
71One main purpose in microbial ecology is the understanding of microbial diversity. Culturable methods

72contain significantly less genetic information than the genomes of the total microbiota (Rondon et al.
732000). To rise above this problem, molecular techniques using DNA or RNA extracted directly from the
74soil have been used (Ward et al. 1990; Torsvik et al. 1998).
75Single-strand conformation polymorphism (SSCP) and random amplified polymorphic DNA (RAPD)

76represent valid alternatives to culturable methods, offering a sensitive strategy to find out microbial
77community from soil (Nair et al. 2002; Lim et al. 2005). SSCP is performed by PCR amplifying a double78stranded DNA fragment that is subsequently denatured to single-stranded DNA and subjected to
79nondenaturing polyacrylamide gel. The mobility of the single-stranded DNA in the gel is related not only to
80its length but also to its nucleotide sequence (Sunnucks et al. 2000).
81RAPD assay has several advantages when compared to other microbial community analyses, since it

82requests no prior knowledge about the genome under investigation, a single random oligonucleotide primer
83and small amount of material is needed (Atienzar and Jha 2006).
84Up to date, only few studies have focused on the correlation between the bacterial community in soils and

85the alteration of the land responsible for landslides. However, evidence of the presence of some kinds of
86bacteria providing the transformation of clay minerals have been reported (Alekseeva et al. 2009).

7
87The samples analyzed in our research were collected from two drillings made in the area of Nerano-Termini

88and Massa Lubrense-Nerano (Naples, Italy).

89From the geological point of view, the studied area is characterized by the presence of rudist limestones of
90the Upper Cretaceous sediments, on which they transgress the Miocene limestone consisting of pecten and
91a clayey sandstone formation between the ages of Late Cretaceous and early Oligocene. The materials
92involved in the landslide of 1963 have been ascribed to their training arenaceous clay Lower Miocene
93(Cotecchia and Melidoro 1966).
94In our research, we studied microorganism communities in the two soils triggering landslides by culture-

95independent analyses, such as RAPD and SSCP.

96Genetic analyses were performed to verify a possible connection between soils weathering involved in the

97landslides and microorganisms possibly affecting the soil structure alterations. For this reason we used
98SSCP and RAPD strategy as useful typing methods to determine microbial soil community, and
99comparing the two different fingerprinting methods.

9
100
101

Methods

102
103Research sites and soil sampling
104Soil samples were collected from two areas involved in landslides, Nerano-Termini and Massa Lubrense-

105Nerano (Naples, Italy) climate zones.

106Samples were drilled out from the ground (in situ sampling), as undisturbed samples, by using the Mazier

107sampler. To avoid a possible contamination of these soil materials, we collected the two samples using
108sterilized steel dies.
109Three samples were collected from each site: A1, Nerano-Termini soil taken to a depth of 21,55m; A2,

110Massa Lubrense Nerano soil taken to a depth of 8,10m; B1, Nerano-Termini soil taken to a depth of
11121,42m; B2, Massa Lubrense-Nerano soil taken to a depth of 7,97m; C1, Nerano-Termini soil taken to a
112depth of 21,55m; C2 Massa Lubrense-Nerano soil, taken to a depth of 7,85m.
113
114
115Total DNA extraction from bulk soil
116Total DNA was extracted from 0.25 g of soil sample using the PowerSoil DNA Isolation Kit (MO BIO,

117Carlsbad, CA). The yield and quality of purified DNA was checked by agarose gel electrophoresis (0.8%
118w/v agarose) and UV visualization of the ethidium bromide stained gels.
119
120RAPD analysis
121Genomic profiles of microbial community were studied using RAPD-PCR with 10 different 10-mer

122random primers (data not shown). Only one primer OPU18 (Operon Technologies, Alameda, CA), provided
123consistent and reproducible band patterns and therefore it was selected for further analysis.
124Briefly, the RAPD reaction was performed with 20 ng DNA soil in a total volume of 25 L containing 2.5

125L of 10x enzyme assay buffer, 2nM of random (10 bp) primer, 100 M each of dATP, dCTP, dGTP, and
126dTTP, and 2.5U of AmpliTaq DNA polymerase (Life Technologies, NY,USA). The amplification was
127performed in a Perkin-Elmer 9600 thermocycler programmed for 45 cycles as follows: 1st cycle of 1min at
12894C, 1min at 38C, 1 min at 72C, followed by a final extension cycle of 15min at 72C. PCR products
129were resolved on 2.5% agarose gel and stained with ethidium bromide.

10

11
130
131SSCP analysis of PCR-amplified 16S rRNA and ITS gene fragments
132Single Strand Conformational Polymorphism Analysis (SSCP) of microbial communities was performed as

133described by Schwieger and Tebbe (Schwieger and Tebbe 1998). Bacterial 16S rRNA gene sequences were
134amplified by PCR using the primer pair Unibac-II-515f (5-GTG CCA GCA GCC GC-3) and Unibac-II135927rP (5-CCC GTC AAT TYM TTT GAG TT-3) (Zachow et al. 2008). PCR was started with an initial
136denaturation step at 95C for 5min, followed by 32 cycles of 95C, 20 s; 54C, 15 s; 72C, 30 s; and
137elongation at 72C, 10 min. The PCR was performed by using a total volume of 50 l containing 1U
138AmpliTaq DNA polymerase, 1.5 mM MgCl2, 0.2 M of each primer and 1 l of template DNA.
139Fingerprinting of fungal communities by SSCP was carried out as described by Schwieger and Tebbe
140(Schwieger and Tebbe 1998). A nested PCR was applied to obtain genetic fingerprints of fungal
141communities. In a first PCR the fungus-specific primer pair ITS1f/ITS4rP (White et al. 1990) was used,
142whereas the primer pair ITS1f/ITS2rP (White et al. 1990) was used for the second PCR. The first PCR (20
143l) contained of 1 U AmpliTaq DNA polymerase, 2.25 mM MgCl2, 0.5 mg/ml BSA, 1.5% DMSO, 0.2 M
144of each primer and 1 l of template DNA. PCR was started with an initial denaturation step at 95 C for
1457min, followed by 30 cycles of 94C, 45 s; 56C, 2 min; 72C, 2 min; and elongation at 72C, 10 min.
146Samples served as templates for the second PCR. For bacterial DNA, the amplicons were separated at 400
147V and 26C in 8 % acrylamide gels and for fungal DNA 9% acrylamide gels. The electrophoretic
148separation was conducted under non-denaturing conditions using the DCode Universal Mutation Detection
149System (Bio-Rad, Hercules, CA). Silver staining was used for the routine detection of DNA bands in SSCP
150gels (Bassam et al. 1991).
151Extraction, re-amplification and sequencing of DNA from silver-stained SSCP profile
152Selected bands of the SSCP community profiles were cut out with a sterile razor blade, and the single-

153stranded DNA of these bands was eluted for 3 h at 37C and 500 rpm in 50 l crush and soak buffer (0.5
154M ammonium acetate, 10 mM Mg2+-acetate, 1 mM EDTA [pH 8.0] and 0.1% sodium dodecyl sulfate)
155(Sambrook et al. 1989). The eluted DNA was precipitated with ethanol and finally resuspended in 12l of
15610 mM TrisHCl, pH 8.0. Extracted DNA fragments were re-amplified by PCR, cloned using TA Cloning
157kit (Invitrogen) and sequenced.
158
159DNA sequence analysis

12

13
160Sequencing of the amplicons was performed using the BigDye Terminator v3.1 Cycle Sequencing Kit

161(Applied Biosystems), with the same primers described before, in an ABI 3100 automatic DNA sequencer
162(Applied Biosystems). The sequences were matched in BLAST (Altschul et al. 1990) and phylogenetic
163analysis was performed using MEGA version 5.0 (Tamura et al. 2011) after multiple alignment of data by
164ClustalW (Thompson et al. 1994). Distance matrix and neighbour-joining methods (Saitou and Nei 1987)
165were applied for tree construction.
166The bootstrap consensus tree was inferred from 1000 replicates (Felsenstein 1985).
167The evolutionary history was inferred using the Neighbor-Joining method and the bootstrap consensus tree

168inferred from 1000 replicates. The percentage of replicate trees in which the associated taxa clustered
169together in the bootstrap test (1000 replicates) are shown next to the branches.

14

15
170

Results

171RAPD Fingerprint Profiles

172In order to analyze the microbial community, fingerprints were performed by RAPD analysis using DNA

173extracted from the soil samples. The selected RAPD primer produced a number comprised between from
17410 to 16 fragments, 2001,500 bp long (Fig.1A).
17528 RAPD fragments were excised, cloned and sequenced. Table 1 shows the best sequence homology

176BLAST matching obtained from both soil types clones. Two clone sequences were not attributable to a
177specific taxonomic order (uncultured alpha Proteobacterium and uncultured Gram-positive bacterium),
178while the remaining 26 clones were found to belong to 24 different genera.
179Fig. 1B shows the phylogenetic relationships between clones and microorganisms identified with BLAST

180analysis.
181The analysis of the microorganisms found in each soil sample has allowed the identification of different

182orders.
183Burkholderiales (46%), Enterobacteriales (27%) and Pseudomonadales (8%) were the main orders

184identified in Nerano-Termini soil, while Enterobacteriales (22%) and Pseudomonadales (22%) were
185mainly represented in Massa Lubrense-Nerano soil. Moreover, microorganisms belonging to
186Mycoplasmatales, also if less represented, were found in both soils.
187Bacteria belonging to Lactobacillales and Neisseriales orders were specifically found in Nerano-Termini

188soils, while Actinomycetales and Bacillales were found only in Massa Lubrense-Nerano soil.
189In details, ten species belonging to Enterobacteriales order (Cronobacter turicensis, Citrobacter koseri

190ATCC, Erwinia billingiae strain Eb661, Serratia marcescens FGI94, Uncultured alpha Proteobacterium,
191Uncultured Gram-positive bacterium, Mycobacterium sp., Bacillus sp., Leuconostoc mesenteroides,
192Uncultured Aquamonas sp.) and twelve microorganisms belonging to Burkholderiales (Leptothrix
193cholodnii SP-6, Methylibium petroleiphilum, Thiomonas sp., Thiomonas intermedia K12, Delftia sp.,
194Polaromonas naphthalenivorans, Delftia acidovorans, Alicycliphilus denitrificans BC, Ramlibacter
195tataouinensis, Alicycliphilus denitrificans K601, Acidovorax citrulli, Variovorax paradox) have been
196identified only in Nerano-Termini samples (Table 1).
197No specific species belonging to Enterobacteriales order was found in Massa Lubrense-Nerano soil and

198Pseudomonadales (uncultured Pseudomonas sp. and Pseudomonas aeruginosa) bacteria were found in both

16

17
199soils. Moreover, both the uncultured Gram-positive bacterium and the uncultured alpha Proteobacterium

200were present in the two soil types.

201
202SSCP profiles of bacterial communities
203To establish the diversity of microbial isolates from the two studied soils, fingerprints were performed also

204by SSCP analysis of 16S rRNA genes (Fig.2A).

205In order to assign single-stranded PCR products generated to specific microorganisms, each band was

206extracted from polyacrylamide gels, subjected to PCR with the same primers as in the first amplification
207and then cloned and sequenced.
208Eleven SSCP fragments were excised, cloned and sequenced. Table 2 shows results obtained from sequence

209analysis obtained from both soil types and the corresponding phylogenetic tree is showed in Fig. 2B. The
210nucleotide sequences retrieved from the SSCP profile showed similarities to 16S rRNA gene sequences
211deposited in public databases with a maximum identity ranging from 92 to 99%.
212Five clone sequences were not attributable to a specific taxonomic order (three sequences related to

213uncultured bacteria, one sequence related to uncultured archaeon and one sequence related to uncultured
214alpha Proteobacterium), while the remaining 6 clones were found to belong to 5 different genera.
215It is interesting that, as well as the RAPD analysis, most sequences were related to bacteria belonging to

216the Pseudomonadales order (18% in Nerano-Termini and 33% in Massa Lubrense-Nerano).

217Actinomycetales, Enterobacteriales, Mycoplasmatales and Alphaproteobacteria microorganisms were
218common to the two soil types studied. Instead, uncultured bacteria, uncultured archea and
219Desulfuromonadales (uncultured Geobacter sp.) were specifically found in Nerano-Termini soil sample.
220However, the profiles generated with SSCP analysis showed a reduced number of identified bacteria when
221compared to RAPD analysis.
222
223Fingerprinting of fungal communities by SSCP
224In order to characterize fungal communities, two different soil types were analysed employing SSCP

225analysis. SSCP patterns were obtained with a nested PCR using the fungi-specific primer pair ITS.
226Amplicons were separated on acrylamide gels for fungal DNA. In total, 18 different bands, showed in Fig.
2273, were isolated from SSCP profiles.

18

19
228Dominant bands were excised from SSCP gels and re-amplified by PCR, cloned and sequenced.

229Bioinformatic analysis of sequences cloned from SSCP bands showed he presence of fungal DNA in our
230samples. The identity of single-stranded PCR products isolated are shown in Table 3. On the basis of
231BLASTn results, six clones could not be attributed to any taxonomic order (two sequences were associated
232to uncultured marine fungi, one to uncultured Basidiomycota, two to uncultured fungus and one to
233Sordariomycetes sp.). The remaining 12 clones were found to be related to sequences from fungi belonging
234to three orders (Malasseziales, Eurotiales and Hypocreales). No overlap between the fungi orders found for
235the two soil samples was showed after clone analysis.
236In details, in Nerano-Termini soil Malasseziales fungi were mainly represented (43%), but also Eurotiales,

237uncultured Basidiomycota and uncultured fungi were identified. In Massa Lubrense-Nerano,

238Sordariomycetes and uncultured fungi were identified, and Hypocreales represented the most frequent
239order (82%).

20

10

21
240
241

Discussion

242Landslides are one of the most severe natural disturbances characterized by the rapid movements of the

243earths or other solid material. They can cause extensive damage to life having important landscape and
244ecosystem-wide effects on nutrient availability. Studies on landslides are often focused on landslide
245distribution, nutrient availability and plant successions (Guariguata and Larsen 1990; Myster and Walker
2461997). Previously, it has been reported that the processes of transformation of clay minerals such as
247intensification of removal of exchange bases and dissolution of silicates and iron oxides occurred in the
248presence of the alkaliphilic cyanobacterial community (Alekseeva et al.2009). However, little has been
249studied to determine the relationships between microbial community composition and the alteration of the
250land responsible for landslides.
251In the present work we analyzed the bacterial and fungal communities in soils triggering landslides in

252Nerano-Termini and Massa Lubrense-Nerano by culture-independent method, based on DNA analysis, such
253as RAPD and SSCP.
254RAPD fingerprinting results and phylogenetic analysis showed that Nerano-Termini and Massa Lubrense-

255Nerano soils were characterized by a prevalence of microorganisms belonging to Enterobacteriales and

256Pseudomonadales orders.
257RAPD analysis allowed also the identification of differences in microbial communities in the two soils.

258Burkholderiales, Lactobacillales and Neisseriales bacteria were specifically found in Nerano-Termini soils,
259while Actinomycetales and Bacillales were found only in Massa Lubrense-Nerano soil.
260The comparison of RAPD results with bacterial SSCP results showed that the first method allows the

261identification of a larger population of microorganisms. Both analysis provided evidence of a prevalence of

262Pseudomonadales and the presence of Actinomycetales and Enterobacteriales in studied soils.
263It is known that most soil bacteria are organized in biofilms on roots, litter, or soil particles. A biofilm can

264be broadly defined as an organised system of microbial cells associated with surfaces, often with unique
265structural phenomena, and are characterized by distinct physical and chemical gradients that affect
266microbial metabolic processes, that are heavily influenced by the substratum. Bacteria attach to surfaces,
267form biofilms on them, and often alter the surfaces as a result of this interaction, leading to the destruction
268(dissolution), transformation, or even the formation of minerals. Moreover, it has been suggested that the

22

11

23
269biofilm microenvironment may stimulate bacterial production of specific extracellular enzymes involved in

270degradation of organic material (Jass et al. 2002).

271Interestingly, the orders and genera mainly identified in soil types in this study, Enterobacteriales,

272Burkholderiales and Pseudomonadales, are well-characterised biofilm-associated microorganism groups.

273Among the Enterobacteriales, E. coli O157 has been demonstrated to survive in cold water for up to

274twelve weeks in mud, sludge and other sediments, including those occurring naturally in lakes, seas and
275along riverbeds (Hall-Stoodley and Stoodley 2005). Molecular typing analyses has determined that the
276bacterium is capable of surviving for months, and perhaps years, heavily embedded within biofilm
277matrices, persisting outside a host organism in animal faeces or faecally-derived material, on inorganic
278substrata (such as wood or metal), and in both treated and untreated waters (Cooper et al. 2007).
279Beside, the presence of species of Burkholderiales, Burkholderia sp. specifically, as members of anode

280biofilm communities surfaces has also been reported (Chung and Okabe 2009).
281Additionally, the typical Pseudomonas bacterium might be found in biofilm in nature, attached to some

282surface or substrate, or in a planktonic form, as a unicellular organism, actively swimming by means of its
283flagellum. Among the identified species belonging to the genus Pseudomonas, Pseudomonas aeruginosa is
284ubiquitous in soil and water has forming an antibiotic-resistant biofilm (Drenkard and Ausubel 2002).
285Previously, it has been reported that frequent occurrence in soil structure of the ions with variable valence

286(e.g Fe, Mn, etc.) makes minerals extremely sensitive to environmental conditions (Stucki 1988), beside
287bacteria and fungi can be absorbed on the surface of clay minerals and the products of their metabolism
288interact with the minerals (Sokolova et al. 2005). Among the bacterial species identified in Nerano-Termini
289soil, Leptothrix cholodnii is an aerobic, sheath forming, filamentous bacteria, able to oxidize Mn2+ and Fe2+
290and is usually found in oligotrophic, slowly running, iron- and manganese-rich water (Siering and Ghiorse
2911996). Several organic or inorganic electron donors, such as reduced inorganic sulphur compounds, could
292be used in carbon dioxide fixation during anoxygenic photosynthetic growth of Thiomonas sp., which has
293showed particular carbon and energy metabolic capacities (Frigaard and Dahl 2009).
294Also other bacteria found in our soil samples have been previously found to be able to metabolize soil-

295related compounds. Delftia acidovorans is known to degrade a number of organic compounds such as 2-(4296sulfophenyl)butyrate (SPB) and may be useful for the degradation of linear alkylbenzenesulfonate (LAS)
297surfactant in wastewater treatment (Schulz et al. 2000). Ramlibacter tataouinensis Strain TTB310 is able to
298use only acetate, pyruvate, beta-hydroxybutyrate, gamma-hydroxybutyrate, DL-lactate or propionate of

24

12

25
299tested carbon sources and it is able to reduce nitrate to nitrite; moreover, a beta-glucosidase activity has

300been reported (Heulin et al. 2003; Gommeaux et al. 2005).

301Reported data suggest a role of biofilm-forming bacteria in chemical composition of the soil. The

302association of genetic analysis to identify microorganism present in soil samples with compositional
303analysis would be a valid approach to increase current knowledge about features leading to landslides.
304Other than bacteria, our genetic analyses revealed the presence of fungi in soil samples. Our results about

305the fungal community showed differences among species composition in two soil types.
306Some of them are known to be involved in biochemical processes possibly related to soil properties. In

307Nerano-Termini soil we identified uncultured Trichocomaceae. Trichocomaceous are saprophytic fungi

308whose role in soil aggregation remains uncertain. However, recently, it has been reported that
309Trichocomaceous fungi are able to create soil aggregates, most probably as a result of hyphal enmeshment.
310Once formed, the aggregates are not stable, likely due to the loss of hyphal wall integrity during autolysis,
311leading to disintegration of macroaggregates into microaggregates or reversion to the original size
312distribution (Daynes et al. 2012).
313It is interesting that in Massa Lubrense Nerano soil we identified saprophytic fungi, Fusarium and Nectria.

314This finding supports a possible role of fungal community in determining physical properties of soils.
315Our research, other than provide evidence of microbial community differences between two landslide-

316subjected soils, allows the comparison of two molecular analyses in finding bacterial and fungal species.
317Our results showed that RAPD analysis allowed to obtain more detailed results about the bacterial
318community present in the studied soil samples, while SSCP analysis with specific primers for fungi
319identification resulted in a more detailed profile of the fungal community.
320Moreover, the identification of still uncultured microbial species underlies the limitation of standard

321culture-based techniques, highlighting the relevance of molecular studies to analyze microbial communities
322from complex samples, such as the soil. Previous studies support the importance to identify uncultured
323microbial species in soil samples, as they can have specific metabolic capabilities not expressed by cultured
324species. For example, antibiotic resistance genes knowledge (Riesenfeld et al. 2004), and enzyme discovery
325(Kim et al. 2008) can be extended by the discovery of uncultured microorganisms, allowing the research of
326compounds which inhibit resistance mechanisms.
327In conclusion, the present study showed that bacterial and fungal communities found in Nerano-Termini

328and Massa Lubrense-Nerano soils were enriched in species which could contribute to degradation processes

26

13

27
329leading to soil properties change possibly involved in landslides occurring in the studied sites. Furthermore,

330the results of this study can be considered as an initial support for evaluating the relationship among
331bacterial and fungal community colonizing soils and the occurring of landslides. However, further studies
332are necessary to confirm the direct involvement of bacterial and fungal species to determine soil properties.
333

28

14

29

334
335

References

336Alekseeva TV, Sapova EV, Gerasimenko LM, Alekseev AO (2009) Transformation of clay minerals caused

337by an alkaliphilic cyanobacterial community. Microbiology 78: 776-784.

338Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment search tool. J

339Mol Biol 215:403-410.

340
341Atienzar FA, Jha AN (2006) The random amplified polymorphic DNA (RAPD) assay and related

342techniques applied to genotoxicity and carcinogenesis studies: a critical review. Mutat Res 613(2-3):76343102.
344Baioni D (2011) Human activity and damaging landslides and floods on Madeira Island. Nat Hazards Earth

345Syst Sci 11:3035-3046.

346Bassam BJ, Caetano-Anolles G, Gresshoff PM (1991) Fast and sensitive silver staining of DNA in

347polyacrylamid gels. Anal Biochem 196:8083.

348Cotecchia V, Melidoro G (1966) Geologia e Frana di Termini Nerano (Penisola Sorrentina). Geologia

349Applicata e Idrogeologia 1:93-126.

350
351Cooper IR, Taylor HD, Hanlon GW (2007) Virulence traits associated with verocytotoxigenic Escherichia

352coli O157 recovered from freshwater biofilms. Journal of Applied Microbiology 102:1293-1299.
353
354Chung K, Okabe S (2009) Characterization of electrochemical activity of a strain ISO2-3 phylogenetically

355related to Aeromonas sp. isolated from a glucose-fed microbial fuel cell. Biotechnol Bioeng 104:901-10.
356
357Daynes CN, Zhang N, Saleeba JA, McGee PA (2012) Soil aggregates formed in vitro by saprotrophic

358Trichocomaceae have transient water-stability. Soil Biology & Biochemistry 48: 151-161.
359
360Drenkard E, Ausubel FM (2002) Pseudomonas biofilm formation and antibiotic resistance are linked to

361phenotypic variation. Nature 416:740-3.

362
363 Felsenstein J (1985) Confidence limits on phylogenies: An approach using the bootstrap. Evolution

36439:783-791.

30

15

31
365Frigaard NU, Dahl C (2009) Sulfur metabolism in phototrophic sulfur bacteria. Adv Microb Physiol

366

54:103200.

367Futagami T, Sakai H, Kakugawa K, Fujimoto A, Fukuhara T, Fujiwara Y, Sakurai S (2010) Debris flows

368promoted by mechanical deterioration of the ground due to eutrophication of hillside ecosystems. WIT
369Transactions on ecology and the environment 141:703-714.
370Guariguata MR, Larsen MC (1990) Preliminary map showing landslides El Yunque quadrangle, Puerto

371Rico. U.S. Geological Survey Open-file Report, 89-257, scale 1:20,000, 1 sheet.
372Gommeaux M, Barakat M, Lesourd M, Thiry J, Heulin T (2005) A morphological transition in the

373

pleomorphic bacterium Ramlibacter tataouinensis TTB310. Res Microbiol 156:1026-1030.

374Hall-Stoodley L, Stoodley P (2005) Biofilm formation and dispersal and the transmission of human

375pathogens. Trends in Microbiology 13:7-10.

376Heulin T, Barakat M, Christen R, Lesourd M, Sutra L, De Luca G, Achouak W (2003) Ramlibacter

377
378

tataouinensis gen. nov., sp. nov., and Ramlibacter henchirensis sp. nov., cyst-producing bacteria
isolated from subdesert soil in Tunisia. Int J Syst Evol Micr 53: 589-594.

379Huggel C, Clague JJ, Korup O (2012) Is climate change responsible for changing landslide activity in high

380mountains? Earth Surf Process Landforms 37:7791.

381Jass J, Roberts SK, Lappin-Scott HM (2002) Microbes and enzymes in biofilms. In: Burns RG, Dick RD

382(eds) Enzymes in the Environment. Activity, Ecology and Applications, Marcel Dekker Inc., New York,
383USA, pp 307326.
384Jamaludin S, Haut BBK, Umar H (2006) Evaluation of slope assessment systems for predicting landslides

385of cut slopes in granitic and meta-sediment formations. Am J Environ Sci 2:135-141.
386Kim SJ, Lee CM, Han BR, Kim MY, Yeo YS, Yoon SH, Koo BS, Jun HK (2008) Characterization of a gene

387encoding cellulase from uncultured soil bacteria. FEMS Microbiol Lett 282(1): 44-51.
388Keefer DK (2002) Investigating Landslides Caused by Earthquakes A Historical Review. Surv Geophys

38923:473-510.
390Lim H, Lee KH, Hong CH, Bahk GJ, Choi WS (2005) Comparison of four molecular typing methods for

391the differentiation of Salmonella spp. Int J Food Microbiol 105:411-418.

392Myster RW, Walker LR (1997) Plant successional pathways on landslides in Puerto Rico landslides. J Trop

393Ecol 13:163-173.
394
32

16

33
395Nair S, Lin TK, Pang T, Altwegg M (2002) Characterization of Salmonella serovars by PCR-single-strand

396conformation polymorphism analysis. J Clin Microbiol 40:2346-2351.

397Pardeshi SD, Autade SE, Pardeshi SS (2013) Landslide hazard assessment: recent trends and techniques.

398Springerplus 2:523.
399Radina VV (1973) Role of microorganisms in the formation of properties of soils and their stress state

400Hydrotechnical Construction 7:847-850.

401Riesenfeld CS, Goodman RM, Handelsman J (2004) Uncultured soil bacteria are a reservoir of new

402antibiotic resistance genes. Environ Microbiol. 6(9):981-9.

403Rondon MR, August PR, Bettermann AD, Brady SF, Grossman TH, Liles MR, Loiacono KA, Lynch BA,

404MacNeil IA, Minor C, Tiong CL, Gilman M, Osburne MS, Clardy J, Handelsman J, Goodman RM (2000)
405Cloning the soil metagenome: a strategy for accessing the genetic and functional diversity of uncultured
406microorganisms. Appl Environ Microb 66:2541-2547.
407Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: a laboratory manual, second edition. Cold

408Spring Harbor Laboratory Press, Cold Spring Harbor, New York, USA.
409
410Saitou N & Nei M (1987) The neighbor-joining method: a new method for reconstructing phylogenetic

411

trees. Mol Biol Evol 4: 406-425.

412Schulz S, Dong W, Groth U, Cook AM (2000) Enantiomeric degradation of 2-(4-sulfophenyl)butyrate via

413

4-sulfocatechol in Delftia acidovorans SPB1. Appl Environ Microb 66:1905-1910.

414Stucki JW (1988) Structural Iron in Smectites. In: Stucki JW, Goodman BA, and Schwertmann U (eds) Iron

415in Soils and Clay Minerals. NATO ASI Series. Vol. 217, pp 625676.
416
417Sokolova TA, Dronova TYa, Tolpeshta II (2005) Glinistye mineraly v pochvakh (Clay Minerals in Soils).

418Grif and K, Tula.

419
420Siering PL, Ghiorse WC (1996) Phylogeny of the Sphaerotilus-Leptothrix group inferred from

421morphological comparisons, genomic fingerprinting, and 16S ribosomal DNA sequence analyses. Int J Syst
422Bacteriol 46:173-182
423
424Sunnucks P, Wilson AC, Beheregaray LB, Zenger K, French J, Taylor AC (2000) SSCP is not so difficult:

425the application and utility of single-stranded conformation polymorphism in evolutionary biology and
426molecular ecology. Mol Ecol 9:16991710.
427

34

17

35
428Schwieger F, Tebbe CC (1998) A new approach to utilize PCR-single-strand-conformation polymorphism

429for 16S rRNA gene-based microbial community analysis. Appl Environ Microb 64:48704876.
430
431Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S (2011) MEGA5: Molecular Evolutionary

432Genetics Analysis using Maximum Likelihood, Evolutionary Distance, and Maximum Parsimony Methods.
433Molecular Biology and Evolution 28:2731-2739.
434Thompson JD, Higgins DG, Gibson TJ (1994) Clustal W: improving the sensitivity of progressive multiple

435sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice.
436Nucleic Acids Res 22:46734680.
437
438Torsvik V, Daae FL, Sandaa RA, Ovres L (1998) Novel techniques for analysing microbial diversity in

439natural and perturbed environments. J Biotechnol 64:53-62.

440
441Xie M, Zhou G, Esaki T (2003) GIS component based 3D landslide hazard assessment system:

4423DSlopeGIS. Chin Geogra Sci 13:66-72.

443Ward DM, Weller R, Bateson MM (1990) 16S rRNA sequences reveal numerous uncultured

444microorganisms in a natural community. Nature 345:63-65.

445
446White TJ, Bruns TD, Lee S, Taylor J (1990) Analysis of phylogenetic relationship by amplification and

447direct sequencing of ribosomal RNA genes. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ (eds) PCR
448Protocols: A Guide to Methods and Applications. Academic Press, San Diego, pp 315322.
449
450Zachow C, Tilcher R, Berg G (2008) Sugar beet-associated bacterial and fungal communities show a high

451indigenous antagonistic potential against plant pathogens. Microb Ecol 55:119129.

452
453
454
455
456
457
458

36

18

37
459
460Figure Legends
461
462Fig. 1 Results from RAPD fingerprints.
463A RAPD fingerprints of DNA from soil microbial communities. M: the DNA ladder consisting of 15 blunt
464fragments ranging in length from 100 to 1,500 bp, at 100-bp increments, and an additional fragment at
4652,072 bp (Roche). Lane 1: sample A1; lane 2: sample A2; lane 3: sample B1; lane 4: sample B2; lane 5:
466sample C1; lane 6: sample C2; lane 7: negative control.
467B Phylogenetic tree of bacterial community profiles generated from RAPD analysis was obtained using the
468Neighbor-Joining method and the bootstrap consensus tree inferred from 1000 replicates. The percentage of
469replicate trees in which the associated taxa clustered together in the bootstrap test are shown next to the
470branches.
471
472
473Fig. 2 Results from SSCP analysis of bacterial community.
474A SSCP analyses of PCR-amplified 16S rRNA genes of soil samples. Lane 1: sample A1; lane 2: sample
475A2; lane 3: sample B1; lane 4: sample B2; lane 5: sample C1; lane 6: sample C2.
476B Phylogenetic tree of 16S rRNA SSCP clones was obtained using the Neighbor-Joining method and the
477bootstrap consensus tree inferred from 1000 replicates. The percentage of replicate trees in which the
478associated taxa clustered together in the bootstrap test are shown next to the branches.
479
480Fig. 3 Results from SSCP analysis of fungal community
481A SSCP fingerprint patterns of fungal communities obtained of soil samples. Lane 1: sample A1; lane 2:
482sample A2; lane 3: sample B1; lane 4: sample B2; lane 5: sample C1; lane 6: sample C2.
483B Phylogenetic tree of fungal community profiles generated by SSCP analysis was obtained using the
484Neighbor-Joining method and the bootstrap consensus tree inferred from 1000 replicates. The percentage of
485replicate trees in which the associated taxa clustered together in the bootstrap test are shown next to the
486branches.
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508

38

19