Previous Article | Next Article ![]()
Applied and Environmental Microbiology, January 2006, p. 346-352, Vol. 72, No. 1
0099-2240/06/$08.00+0 doi:10.1128/AEM.72.1.346-352.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Centro de Investigación Básica de España, Merck Research Laboratories, Merck Sharp & Dohme de España S.A., Josefa Valcárcel 38, 28027 Madrid, Spain
Received 27 May 2005/ Accepted 3 October 2005
|
|
|---|
|
|
|---|
The family Geodermatophilaceae (18) includes the three genera Geodermatophilus (10), Blastococcus (1), and Modestobacter (12). Some strains of the genus Geodermatophilus have been isolated from stone and monument surfaces in the Mediterranean basin, although many of them originate from extreme cryptolithic environments (3, 24). The presence and diversity of Geodermatophilaceae in the environment was previously described after a 16S rRNA gene restriction analysis confirmed the presence of three well-separated clusters of the genera Geodermatophilus, Blastococcus, and Modestobacter (24, 25).
In recent years, 16S rRNA gene PCR primers have been extensively used for the molecular identification of microorganisms at different taxonomic levels (11, 15, 19, 20) and more specifically of different families and genera of actinomycetes (13, 16, 17). Two probes were recently described for the specific detection of Geodermatophilaceae and Modestobacter in rock surfaces by using fluorescence in situ hybridization (22). Alternatively, the development of a real-time PCR (RT-PCR) with specific primers could provide a quantitative method to detect the presence of members of these taxa in any given environmental sample.
We have focused here on the development of family-specific primers to detect by RT-PCR the presence of members of the family Geodermatophilaceae in stones and associated saxicolous lichen samples. Two different geographical origins in Spain were sampled to evaluate the presence of this microbial group: the Central Mountain System in Madrid, with plutonic and metamorphic rocks covered with sedimentary rocks and the limestone outcrops from San Vicente beach in Mallorca island from Balearic islands. We discuss the occurrence of members of Geodermatophilaceae in both geological systems.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Geographical origin of stone samples used in this study, detection of Geodermatophilaceae using primers Geosp2 and Geosp1, and concentration of total DNA
|
Genomic DNA extraction.
Total genomic DNAs from the type strains used in the present study were recovered and purified as previously described (8).
Stone DNA extraction.
Total DNA extraction from lichen-colonized and lichen-free stone surfaces was adapted from a previously described method (26). Stone particles were (250 mg) frozen in liquid nitrogen, ground, and suspended in 0.5 ml of 0.4% (wt/vol) skim milk solution (Difco). Stone particles were sedimented by centrifugation for 10 min at 17,500 x g. Supernatants (300 µl) were mixed with 200 µl of extraction buffer (0.66% sodium dodecyl sulfate, 0.31 M NaCl, 110 mM potassium acetate [pH 5.1]), vortexed, and added to 500 µl of water-saturated phenol. After extraction, supernatants (400 µl) were recovered by centrifugation (15 min, 17,500 x g), and total DNA was precipitated overnight at 20°C by addition of 1 ml of ethanol. After a 15-min centrifugation at 17,500 x g, DNA was washed with 75% cold ethanol and finally dissolved in 100 µl of sterile distilled water. DNA was quantified by spectrometry at 260 nm. The initial yields of total stone DNA is indicated in Table 1. The concentrations of total stone DNA from all of the samples were normalized to 1 ng/µ land serially diluted from 1 to 105 ng/µl.
Design of oligonucleotide primers.
Sequence comparison and analysis were carried out by using programs from the University of Wisconsin GCG package (version 7.2, 1994). GenBank 16S rRNA/DNA sequences were used to design the primers Geosp2 (5'-TCCAAGAAATTGGTGCTA-3') and Geosp1 (5'-CAGTTGTKGCCCAGAGAC-3'; reverse primer) shown in Fig. 1. Alignments of the 16S region were performed by using the multiple alignment software CLUSTAL W (21) to determine the regions only conserved among Geodermatophilaceae species from which the family-specific primers were derived. The genus specificity of oligonucleotides was tested against all DNA sequences available in GenBank with the FASTA program. The melting temperature (Tm) was estimated by using the formulae of Thomas and Dancis and the Lathe formulae (20). Relative Tm values obtained using 0.3 M as a standard salt concentration helped to design pairs of primers with similar high melting temperatures. The presence of primer dimers was discarded by analysis of the melting curve in RT-PCR using iCycler IQ software (Bio-Rad Laboratories, Hercules, CA). The designed oligonucleotides were supplied by Eurogentec (Liege, Belgium).
![]() View larger version (23K): [in a new window] |
FIG. 1. Standard curve for the log starting concentration of serial dilutions of DNA from Geodermatophilus obscurus DSM43160T (A) and Blastococcus aggregatus DSM4725T (B) versus the CT value.
|
Quantification of template DNA was performed by interpolation in a standard curve of threshold cycle (CT) values generated by amplification of known concentrations of genomic DNA of Geodermatophilus obscurus DSM43160T and Blastococcus aggregatus DSM4725T. Amplification efficiencies (E) were calculated by the formula E = 10(1/m) 1, where m is the slope of the standard curve (7).
Calculation of 16S rRNA gene molecules.
To determine the number of molecules of 16S rRNA gene amplified by RT-PCR, we estimated the weight of the 587-bp Geodermatophilaceae-specific amplification fragment in 3.99 x 105 Da (587 bp x 680 Da/bp). To calculate the number of nanomoles per microgram, we used the following calculation: 103 ng/µg x [1 nmol/(3.99 x 105 ng)] = 2.5 x 103 nmol/µg. The number of molecules per microgram was determined as follows: [(6.022 x 1014 molecules)/nmol] x 2.5 x 103 nmol/µg = 1.505 x 1012 molecules/µg = 1.505 x 109 molecules/ng).
Cloning of PCR fragments from stone samples.
Amplification products with the primers Geosp2 and Geosp1 from stone samples and lichens were cloned by using the TOPO cloning Kit (Invitrogen Life Technologies).
DNA sequencing.
PCR primers 27f and 1525r (9) were used for the amplification of the 16S rRNA gene genes of the wild-type isolates. The 1,500 bp. PCR products were purified and used as a template in sequencing reactions with the primers 27f, 357f, 530f, 926f, 1392f, 1525r, 1110r, 685r, and 357r (9). Geodermatophilaceae amplification fragments from stone samples cloned into pCRTOPO vector were sequenced by using M13 forward and reverse primers (TOPO Cloning Kit; Invitrogen Life Technologies). Amplified and cloned DNA fragments were sequenced by using an ABI Prism Dye terminator cycle sequencing kit (Amersham Biosciences).
Sequence analysis.
Sequences were assembled by using the GCG Fragment Assembly System (Program Manual for the Wisconsin Package, version 8). Alignments of the 16S sequences were performed by using the multiple alignment program CLUSTAL W (22). The phylogenetic analysis was completed with 16S rRNA gene sequences of type species of actinomycetes available in GenBank. The data were resampled with 1,000 bootstrap replicates (4) by using the heuristic search option of PAUP (21). The percentage of bootstrap replicates that yielded each grouping was used as a measure of statistical confidence. A grouping found on 95% of the bootstrap replicates was considered statistically significant.
The GenBank accession numbers for the 16S rRNA gene sequences of Blastococcus sp. isolates CIBE-G1, CIBE-G2, CIBE-G3, CIBE-G4, CIBE-G5, CIBE-G6, CIBE-G7, CIBE-G8, CIBE-G9, CIBE-G10, CIBE-G11, CIBE-G12, CIBE-G13, and CIBE-G14 are AY903267 to AY903280, respectively. The accession numbers for the amplified partial 16S region from stone samples GEO1A-10, GEO1A-9, GEO1A-4, GEO1A-1, GEO1A-3, GEO1A-2, GEO1A-7, GEO1A-6, GEO1A-8, and GEO1A-5 are AY903281 to AY903290, respectively.
|
|
|---|
Detection and quantification of Geodermatophilaceae DNA in stone surfaces by specific amplification.
The detection limits of the primers Geosp2 and Geosp1 was established in standard curves representing the threshold cycles (CT) versus the log of 10-fold serial dilutions (1 to 105 ng/µl) of DNA from Geodermatophilus obscurus DSM43160T and Blastococcus aggregatus DSM4725T (Fig. 1). A strong linear relationship between CT and the log of the starting copy number was demonstrated with both G. obscurus DSM43160T DNA and B. aggregatus DSM4725T DNA standard curves (correlation coefficients r2
0.917 and r2
0.985; amplification efficiencies [E] = 162.7% and [E] = 117.6%, respectively). The lowest estimated amount of DNA detected by PCR is in both cases in the range of 105 ng/µl. However, the CT values generated by amplification of the minimal level of DNA using both standard curves shows differences with lower values for G. obscurus DNA (CT = 30) than for B. aggregatus DNA (CT = 36) (Fig. 1).
The family-specific primers were applied to detect the presence of Geodermatophilaceae in DNA from stone samples obtained from two different geographic origins in Spain. Total DNAs were purified from the surface of five stone samples, including three lichen colonized and five lichen-free surfaces and diluted to a final concentration of 1 ng/µl. A specific amplification fragment was detected in all of the samples with the exception of the limestone sample three from Mallorca Island (Table 1). Serial dilutions of the template DNA samples (101 to 1011 ng/µl) were tested in RT-PCR experiments to evaluate the concentration range of Geodermatophilaceae DNA in crude stone DNA extracts by extrapolation in both standard curves (Table 2). In all cases, the minimal concentration of total stone DNA required to obtain an amplification product was ca. 102 ng/µl, and no significant differences were observed between the stone DNAs and stone with lichen DNA samples (Table 2). The levels of Geodermatophilaceae DNA detected in total stone DNA samples as deduced from the quantitative amplification of serial dilutions show similar concentrations and range between 102 and 103 ng/µl.
|
View this table: [in a new window] |
TABLE 2. RT-PCR quantification of Geodermatophilaceae DNA in serial dilutions of stone DNA samples using the primers Geosp2 and Geosp1a
|
|
View this table: [in a new window] |
TABLE 3. Calculation of total 16S rRNA content and total cell counts per milligram of stone
|
To evaluate the effect of nonspecific DNA on the detection level of our primers in amplification reactions with crude stone DNA preparations, we performed a competition experiment where Blastococcus aggregatus DSM4725T DNA was added to a DNA preparation of stone sample 3 where no amplification products were detected with primers Geosp1 and Geosp2. The presence of endogenous bacterial DNA in the stone sample was confirmed by RT-PCR experiments with the universal primers 1497f and 115r directed to conserved sequences at the end and beginning of the 16S and 23S rRNA genes, respectively (9). Serial dilutions of total stone DNA (101 ng/µl) containing B. aggregatus DNA (102 ng/µl) were tested in real-time PCR experiments to evaluate the concentration range of B. aggregatus DNA detected in these conditions. The minimal concentration of B. aggregatus DNA is 1.17 x 104 ng/µl, whereas this stone sample dilution contains 103 ng/µl B. aggregatus DNA showing the 10-fold reduction in the detection levels in crude stone samples (Table 4).
|
View this table: [in a new window] |
TABLE 4. RT-PCR quantification of Geodermatophilaceae DNA in serial dilutions of stone DNA from sample 3A using the primer pairs 1497f-115r and Geosp2-Geosp1
|
The number of 16S rRNA genes per genome reported in Actinomycetales is still limited to a small number of genera, and it has been shown to vary from one to six copies as described in Streptomyces and Thermomonospora species (14, 27; rRNA Operon Copy Number Database [http://rrndb.cme.msu.edu]). Given the lack of information regarding the number of 16S rRNA genes per genome in Geodermatophilaceae, we have considered a range of one to six genes per genome to calculate the number of cells per milligram of stone (Table 3). Total cell counts in stone samples ranged from a high of 5.01 x 107 to 3.06 x 108 (sample 2A) to a low of 4.04 x 106 to 2.42 x 107 (sample 5B) cells/mg of stone using DNA from G. obscurus in the standard curve (dilution of 101 ng/µl) (Table 3).
These results confirm the ability of these pairs of primers to detect the presence of Geodermatophilaceae directly from crude stone DNA extracts prior to the microbial isolation, and this is the first study detecting and quantifying this taxon in the environment using quantitative PCR.
Cloning and sequence analysis of PCR products from stone DNA.
To confirm the specificity of the PCR primers, we cloned and sequenced the amplification fragments obtained by using the primers Geosp1 and Geosp2 from the stone sample 1A from La Cabrera Mountains. Ten transformants were randomly selected, and the sequences obtained from the 587-bp amplification fragments (positions 133 to 720 of 16S rRNA gene region) showed a high conservancy among the sequences, with nucleotide similarity ranging between 99.0 to 99.87%. FastA analysis of cloned sequences shows as closest match sequences of type species of the genera Geodermatophilus and Blastococcus (99% similarity), therefore validating the specificity of the primers for the detection of members of Geodermatophilaceae in stone samples.
Identification of wild-type isolates and evaluation of their diversity.
One of the purposes of the design of the primers was the rapid detection of members of the family among the strains that are isolated from the environment and that share morphological traits with the family Geodermatophilaceae. We have applied these pairs of primers to the rapid PCR identification of 14 wild-type strains isolated from stone sample 1A (La Cabrera Mountain, Madrid, Spain) that exhibited morphological characteristics of Geodermatophilaceae. All of the isolates were dark-brown pigmented and exhibited a sparse growth as irregularly shaped coccoid cell aggregates. The tentative taxonomic identification of the wild-type isolates was confirmed in all cases by real-time PCR with Geodermatophilaceae-specific primers (data not shown). To confirm the genus assignment of our isolates, we determined the almost complete nucleotide sequence of the 16S rRNA gene of the 14 strains. In all of the cases we found that the annealing sequence of each pair of primers was highly conserved among the strains, supporting the specificity of the primers. The alignment shows high homology with the sequence of the forward primer Geosp2 (one mismatch) and reverse primer Geosp1 (two mismatches) (see supplemental material). Among the strains assigned to the family Geodermatophilaceae the level of sequence similarity for the complete 16S rRNA gene sequence ranges from 98.9 to 99.86%, indicating the high relatedness of this group of isolates obtained from the same environment. For all of the isolates the closest sequence match was observed with species of the genus Blastococcus (sequence similarity levels 97.1 to 99.2%), and lower sequence homologies were observed with other members of the family, such as species of the genus Geodermatophilus (96.7 to 96.4%) and the genus Modestobacter (97.%).
The sequences from cloned amplification fragments of stone sample 1A DNA and the homologous region of the 16S rRNA gene of the 14 sequenced wild-type strains present a high sequence similarity that ranged between 98 and 96.5%. The high sequence conservation of the amplified region makes it impossible to distinguish at the genus level among members of the family, and therefore these results do not exclude the presence of members of other genera of the family in the stone sample.
A phylogenetic tree based on complete 16S rRNA gene sequences of the wild-type isolates was built by using the maximum-parsimony method (Fig. 2) showing the inter- and intraspecific relationships of the wild-type strains to reference strains of the family Geodermatophilaceae, such as the genera Geodermatophilus, Blastococcus, and Modestobacter and other members of the order Actinomycetales. The topology of the tree confirms previous works where the monophyletic lineage of the family Geodermatophilaceae is supported by 100% of bootstrap value (24). All of the wild-type strains in our study cluster with 70% of bootstrap value and are closely related within the same group with species of the genus Blastococcus. This close relationship to members of this genus is highly supported by the bootstrapping values (100%). Nevertheless, the presence of this cluster clearly distinct from the cluster grouping the remaining species of Blastococcus highly supported by a 98% of bootstrap could suggest the existence of a new species within the genus Blastococcus that will have to be investigated further given the different morphological features exhibited in culture. Previous works on stones biodiversity have noted that, whereas strains of the genus Geodermatophilus are very rare, strains of Modestobacter are very common on stone surfaces from different climates and strains of the genus Blastococcus could be found on rock surfaces in the Mediterranean basin (25). Our data support the view that the interspecific diversity of the genus Blastococcus is higher than the number of species described at the moment, being the surface of rock samples and the microenvironment that is associated valuable sources for isolation of new members of the genus.
![]() View larger version (36K): [in a new window] |
FIG. 2. Phylogenetic tree showing the position of the wild-type strains of the family Geodermatophilaceae derived from 16S rRNA gene analysis by using the maximum-parsimony method in PAUP 4.0. The numbers above the branches indicate the bootstrap percentage from 1,000 replicates that was used as a measure of statistical confidence. GenBank accession numbers of sequences from all of the species included in the analysis are as follows: X92359 (Geodermatophilus obscurus G16), X92357 (Geodermatophilus obscurus DSM 43162), X92355 (Geodermatophilus obscurus DSM 43161), X92356 (Geodermatophilus obscurus DSM 43160), L46061 (Geodermatophilus obscurus dictyosporus 33S), L46020 (Geodermatophilus obscurus G15), Y18646 (Modestobaster multiseptatus DSM 44406), AJ315674 (Blastococcus saxobsidens BC412), AJ316573 (Blastococcus saxobsidens BC521), AJ316572 (Blastococcus saxobsidens BC517), AJ316571 (Blastococcus saxobsidens BC448/DSM 44509T), AJ316570 (Blastococcus saxobsidens BC512), X97888 (Nocardiopsis lucentesis U0297), D85497 (Streptoalloteichus hindustans IFO15115), X53191 (Kibdelosporangium aridum 2030), Z38017 (Saccharomonospora azurea K161), AF139830 (Saccharomonospora glauca KCTC 3673), Z38021 (Saccharomonospora viridis 1852), AF051342 (Amycolatopsis rugosa DSM 43194T), U9334 (Saccharopolyspora hirsuta ATCC 27865), X53197 (Saccharopolyspora hordei A54), AF061976 (Saccharopolyspora rectivirgula 28042), Z14111 (Pseudonocardia halophobica 294045), X55609 (Pseudonocardia nitrificans IFAM 379), AJ252826 (Pseudonocardia hidrocarboxidans DSM 43281), X76959 (Pseudonocardia compacta DSM 43592), X93187 (Actinoplanes philippinensis DSM 43019), D85479 (Couchioplanes caeruleus IFO 13939), D85476 (Catenuloplanes japonicus IFO 14176), X92597 (Micromonospora echinospora DSM 43141), U58528 (Dactylosporangium aurantiacum ATCC 23491), D85477 (Catellatospora citrea IFO 14495), D86946 (Pilimelia terevasa IFO 15934), X97888 (Nocardiopsis lucentesis DSM 44048T 14626), U48973 (Microtetraspora fusca IFO 13915), D85494 (Planobispora longispora IFO 13918), U83912 (Microbispora bispora ATCC 19993), U58257 (Actinomadura madurae JCM 7436), AB041132 (Streptomyces acidiscabies ATCC 49003), D63862 (Streptomyces scabies ATCC 49173), Y15501 (Streptomyces griseus ATCC 10137), D85116 (Streptomyces lavendulae IFO 12789), and AB018487 (Bacillus subtilis ATCC 21331).
|
Supplemental material for this article may be found at http://aem.asm.org/. ![]()
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»