Previous Article | Next Article ![]()
Applied and Environmental Microbiology, October 2005, p. 6319-6324, Vol. 71, No. 10
0099-2240/05/$08.00+0 doi:10.1128/AEM.71.10.6319-6324.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
and
Jillian F. Banfield1,2
Department of Environmental Science, Policy, and Management,1 Department of Earth and Planetary Sciences, University of California, Berkeley, California 947202
Received 7 December 2004/ Accepted 10 May 2005
|
|
|---|
106 cells/ml, oxidizing ferrous iron as the sole energy source. 16S rRNA-internal transcribed spacer region clone library analysis confirmed that the isolate is a member of Leptospirillum group III and that the culture is axenic. We propose the name Leptospirillum ferrodiazotrophum sp. nov. for this iron-oxidizing, free-living diazotroph. This study highlights how environmental sequence data can provide insights for culturing previously uncultured microorganisms. |
|
|---|
In a previous study, we assembled large genomic fragments of five microbial species from a simple subsurface biofilm growing in pH 0.8, 37°C, metal rich acid mine drainage (AMD) (34). These acidophilic biofilms are essentially self-sustaining chemolithoautotrophic communities that obtain their energy by oxidizing ferrous iron and reduced sulfur compounds. They receive limited fixed carbon and nitrogen from external sources and appear to obtain most of their carbon and nitrogen requirements by fixation of atmospheric CO2 and N2.
To date, the only acidophilic microorganisms demonstrated to be capable of nitrogen fixation are Acidithiobacillus ferrooxidans and Leptospirillum ferrooxidans (21). However, these bacteria have never been found in AMD communities growing at a pH of <1 (1, 27). It was anticipated that other Leptospirillum species present in the AMD biofilm (34) would be responsible for nitrogen fixation. The genus Leptospirillum has been divided into three groupsI, II, and IIIon the basis of 16S rRNA gene phylogeny (6). Representatives of groups II and III were identified in the biofilm analyzed by community genomics (34). L. ferrooxidans is a representative of group I, L. ferriphilum is a representative of group II (10). However, no cultured representatives of Leptospirillum group III have been described.
The single nif operon detected in the AMD community genomic data was assigned to a genome fragment from the Leptospirillum group III population, a relatively minor member of the biofilm (<10%). Since a near-complete genome for the Leptospirillum group II community member was obtained, we inferred that this organism lacks nitrogen-fixing genes. This suggested a clear strategy for isolating the first cultivated representative of Leptospirillum group III based on the ability to carry out nitrogen fixation. The result highlights the potential of environmental genomic data to provide insights needed to bypass the cultivation bottleneck.
|
|
|---|
Sample collection and processing.
Samples of pink biofilms growing at pH 0.7 to 1.3 and at 37 to 42°C AMD were collected from Ultra-back A and C drifts of the Richmond Mine at Iron Mountain, Calif., in November 2003. For characterization by fluorescence in situ hybridization (FISH), samples were fixed in paraformaldehyde, as previously described (6), and stored at 20°C. Samples for nucleic acid extraction were stored at 80°C.
Nucleic acid extraction.
Nucleic acids were extracted from samples with lysozyme (2 mg/ml), and cellular proteins were digested by using a STEP buffer (sodium dodecyl sulfate [0.5%], 50 mM Tris-HCl [pH 7.5], 400 mM EDTA, proteinase K [1 mg/ml], and Sarkosyl [0.5%]) at 50°C for 30 min. After two phenol-chloroform extractions, the DNA was precipitated with isopropanol and sodium acetate, and resuspended in Tris-EDTA buffer.
FISH and bright-field microscopy.
FISH was used to screen biofilm samples for abundance of Leptospirillum group III. Probes were commercially synthesized and 5' labeled with the fluorochrome fluorescein isothiocyanate, Cy3, or Cy5 (Interactiva, Ulm, Germany). Paraformaldehyde-fixed samples were probed with Leptospirillum group III-specific probes (LF1252) used in combination with general Leptospirillum (LF655), bacterial (EUB338mix), and archaeal (ARC915) probes under standard hybridization and wash conditions (6). Gram stains were also performed on cultures and environmental AMD biofilms. A Leica DM RX microscope was used for visualization of FISH preparations. Images were captured by using the Wasabi software package (Hamamatsu).
Isolation, growth conditions, and microscopic examination.
Liquid serial dilution was used in isolation attempts due to difficulties in preparing solid media and obtaining growth on plates at a pH of
1.2. Serial dilution has been successfully applied in the isolation of several other organisms (24, 29). One gram of biofilm (ca. 107 to 108 cells) from Ultra-back A was homogenized by using a 16G1 needle and syringe, filtered through a 0.7-µm GD/X syringe filter (Whatman, Inc.) to remove cell aggregates, and inoculated into a modified 9K media (31). Adaptations of this medium have been used to grow other Leptospirillum species, e.g., L. ferrooxidans (2) and L. ferriphilum (10). The medium was modified to be nitrogen-free by removal of ammonium sulfate [(NH4)2SO4] and calcium nitrate [Ca(NO3)2], the only sources of fixed nitrogen in the medium. In order to mimic the pH of the natural AMD biofilm, the nitrogen-free 9K medium was prepared at pH 1.2. The sample was serially diluted to 1010. The modified medium contains ferrous sulfate (FeSO4 · 7H2O) as the primary energy source. All serial dilutions were grown in the dark at 37°C and shaken at 150 rpm for 21 days. Cultures were periodically examined microscopically using a 2x Live/Dead BacLight staining solution (Molecular Probes, Inc.). DNA extractions and FISH were performed on terminal cultures in SD1 and SD3. After an axenic population of cells was obtained, subcultures were grown both without a nitrogen source and in the presence of fixed nitrogen [23 mM (NH4)2SO4].
16S rRNA-ITS and nifH libraries.
The universal 16S rRNA gene primer 27F (18) was used in combination with the 23S rRNA gene primer 23S-115R (18) in order to amplify the 16S internal transcribed spacer (ITS) region from DNA extracted from cultures. General nested nifH (37) primers were used to amplify nifH genes from bulk DNA extracted from cultures and environmental biofilm samples. Each 50-µl reaction mixture contained (as the final concentration) 1x reaction buffer (Invitrogen), 1.5 mM MgCl2, 200 µM concentrations of each deoxynucleoside triphosphate, 0.3 U of Platinum Taq DNA polymerase (Invitrogen),
50 ng of genomic DNA, and 200 ng of forward and reverse primers for each gene. Thermal cycling was carried out on a Hybaid PCR Express Thermal Cycler with an initial denaturation step of 96°C for 10 min, followed by 30 cycles at 94°C for 1 min, 50°C for 1 min, and 72°C for 2 min, followed finally by an extension step of 72°C for 5 min. The PCR products were purified with a QIAGEN QiaQuick PCR purification kit and were cloned by using the TOPO TA Cloning System (Invitrogen). Complete 16S rRNA-ITS sequences were obtained from 10 clones from each library using broadly conserved sequencing primers (27F, 530F, 1100F, 1100R, 1492R, and 115R [18]). Partial nifH sequences were obtained from 20 clones from both environmental and isolate nifH gene libraries. All sequencing was performed by the University of California Berkeley DNA Sequencing Facility on Applied Biosystems 48 capillary 3730 or Applied Biosystems 16 capillary 3100 Genetic Analyzer Systems.
Phylogenetic analysis.
The compiled sequences were aligned by using the ARB software package (19), and alignments were refined manually. Phylogenetic trees based on comparative analysis of the 16S rRNA gene and NifH protein were constructed by using neighbor joining and maximum-likelihood methods. Bootstrap values were inferred by using parsimony in PAUP* 4.0 with default settings. Bayesian inferences were generated using MRBAYES (15). A consensus tree was constructed in PAUP* 4.0. Posterior probabilities of branching points were estimated by Bayesian inference using MRBAYES v2.01 (14) with the default settings and general time-reversible model (six substitution types [n = 6]) and a gamma rate correction (rates = gamma) suggested for RNA data. A total of 100,000 generations were calculated for the data set, from which a tree was saved every 100 generations. The likelihoods of the trees had converged on a stable value after 20,000 generations, and a consensus tree was constructed from the final 80,000 generations (800 trees) in PAUP* 4.0.
|
|
|---|
![]() View larger version (9K): [in a new window] |
FIG. 1. Comparison of the nif operon identified in the AMD biofilm community genomic data and the nif operon of L. ferrooxidans L3.2 (22). Dashed lines indicate missing sequence in L. ferrooxidans. CHP, conserved hypothetical protein.
|
Although no nif operon genes were detected in the Leptospirillum group II environmental genome, some structural and regulatory genes known to be involved in nitrogen fixation are present, including a complete nifSU-hesB-hscBA operon, glnK (PII protein), nifA, and nifL. This observation suggests that Leptospirillum group II lost the nif operon after speciation.
Phylogeny of the nif operon found in the community genomic data.
Comparative analysis of NifH and other conserved Nif proteins (not shown) indicate that the Leptospirillum species form a monophyletic group clustering within the proteobacterial radiation of the group I NifH proteins (Fig. 2). This suggests that Leptospirillum species may have obtained their nif operon via an ancient lateral gene transfer event involving a proteobacterial donor, possibly an ancestor of the acidophile Acidithiobacillus ferrooxidans. In addition, there is evidence of lateral gene transfer within proteobacterial classes (e.g., Bradyrhizobium, a member of the Alphaproteobacteria, clustering with Gammaproteobacteria [25]), suggesting that lateral transfer may have played a significant role in the evolution of nitrogen fixation (7).
![]() View larger version (18K): [in a new window] |
FIG. 2. Phylogenetic tree inferred from NifH protein identified in the AMD biofilm community (boldface) and reference isolate microbial genomes. Branch points supported (bootstrap values, >74%) by all inference methods used are indicated by solid circles, and those supported by most inference methods are indicated by open circles. Branch points without circles were not resolved (bootstrap values, <75%) as specific groups in different analyses. The bar represents 10% estimated sequence divergence. Outgroups are Chlorobium tepidum TLS (group II) and Methanothermobacter thermoautotrophicum (group III).
|
![]() View larger version (63K): [in a new window] |
FIG. 3. FISH images of the isolate. (A) Cells hybridized with Cy5-labeled EUB338mix; (B) FITC-labeled LF655 (Leptospirillum group I, II, and III); (C) Cy3-labeled LF1252 (only Leptospirillum group III); (D) combination of all three probes.
|
![]() View larger version (26K): [in a new window] |
FIG. 4. Maximum-likelihood dendrogram of Leptospirillum genera based on comparative analyses of 16S rRNA gene data. Group designations are bracketed on the right. Branch points supported (bootstrap values, >74%) by all inference methods used are indicated by solid circles, and those supported by most inference methods are indicated by open circles. Branch points without circles were not resolved (bootstrap values, <75%) as specific groups in different analyses. Shown in blue are the 16S rRNA gene sequence for each of the genomes recovered in the community genomics analysis (34) and in red is the 16S rRNA sequence of the Leptospirillum group III isolate. The outgroup used was Nitrospira marina. The bar represents 10% estimated sequence divergence.
|
Molecular characterization of isolate.
The 16S rRNA gene was directly PCR amplified from bulk DNA extracted from the terminal dilution in SD1 by using broad specificity primers and sequenced. Comparative analysis confirmed that the sequence represents a member of Leptospirillum group III (Fig. 4).
In order to check the purity of the serial dilution cultures, 16S rRNA-ITS clone libraries were prepared from the terminal dilutions in SD1 and SD3 and analyzed. Up to three nucleotide polymorphisms were detected in both the 16S rRNA and ITS regions in the SD1 library, suggesting the presence of multiple strains of a single species. By the third serial dilution (SD3), nearly all 16S-ITS sequences were identical. Only a single unduplicated polymorphism was found in the ITS region of a few clones. Since the incidence of polymorphisms is close to the expected error rate, we conclude that a pure culture has been obtained, referred to here as strain UBA1. The 16S rRNA gene sequence of the Leptospirillum group III organism identified in the AMD biofilm shotgun sequence data is very closely related but not identical to strain UBA1 (Fig. 4). We propose the name Leptospirillum ferrodiazotrophum sp. nov. (fer.ro.di.a.zo.tro' phum. L. neut. n. ferrum iron, N.L. n. ferro- referring to ferrous ions, N.L. diazo- referring to dinitrogen, N.L. neut. adj. suffix -trophum living on, N.L. neut. adj. ferrodiazotrophum living on ferrous ions and dinitrogen.) for this isolate. This strain has been deposited in the ATCC culture collection (accession number ATCC BAA-1181).
Using broad specificity nifH primers, PCR clone libraries were constructed from both the UBA1 culture and Ultra-back A biofilm sample. The partial nifH gene (
350 bp) amplified from the isolate was identical at the nucleotide level to the gene encoded on the Leptospirillum group III genome fragment. The Ultra-back A biofilm nifH genes were slightly divergent at the nucleotide level but identical at the amino acid level. This result suggests that Leptospirillum group III is the only detectable nitrogen fixer in this extremely acidic AMD environment. The absence of nifH in Leptospirillum group II was also confirmed by using general nifH primers on several group II enrichments (data not shown). However, it is possible that that some Leptospirillum group II members at other sites retain nitrogen fixation genes.
|
|
|---|
Leptospirillum group III has only recently been recognized via culture-independent techniques and appears to be limited to acidic environments (1). Currently, there are nine full-length 16S rRNA gene sequences representing this group, with sequence divergence of up to 5%. This level of divergence suggests that Leptospirillum group III likely comprises multiple species of the same genus (32). Based on 16S rRNA gene sequence (Fig. 4), L. ferrodiazotrophum is closely related to organisms detected previously within the Richmond Mine at Iron Mountain (4, 5, 12), as well as a sequence reported from coal refuse (3).
The basic characteristics of L. ferrodiazotrophum strain UBA1 are consistent with its membership in the genus Leptospirillum (20), including morphology, motility, Gram stain, aerobic metabolism, acidophily, growth temperature, iron oxidation, and GC content. Growth of UBA1 on nitrogen-free media confirms the inference made from the environmental genomic data that members of Leptospirillum group III are nitrogen fixing chemolithoautotrophs (34). We attempted several acetylene reduction assays on UBA1 to directly demonstrate nitrogen fixation but were only able to detect picomolar amounts of ethylene production under low oxygen conditions (data not shown). This level of acetylene reduction is comparable to that reported previously for L. ferrooxidans (21). Expression of L. ferrodiazotrophum nif genes in natural AMD biofilms from the Richmond Mine has been confirmed in a recent shotgun proteome study (23a). Since nitrogenases are extremely oxygen sensitive and L. ferrodiazotrophum grows via iron oxidation coupled to reduction of molecular oxygen, the nitrogenase must be protected from oxygen. Fast respiration rates in combination with a high oxygen affinity cytochrome cbb3-type heme-copper oxidase (34) may be responsible for protecting the nitrogenase in UBA1, as has been reported for Azotobacter vinelandii (23).
Biological nitrogen fixation is an energetically expensive process, and growth in many natural environments is limited by the availability of fixed nitrogen. However, AMD microbial communities grow at the air-solution interface, in the presence of an unlimited supply of ferrous iron, their primary energy source. So long as the community contains an organism with genes necessary for nitrogen fixation, nitrogen supply may not be limiting. Leptospirillum group III is a minor component (<10%) of most AMD biofilms within the Richmond Mine. Due to this low abundance and the vital resource of fixed nitrogen that it provides, Leptospirillum group III can be considered a keystone species in this environment (34). The inferred loss of nitrogen fixation in the typically dominant Leptospirillum group II population supports this hypothesis. Removal of Leptospirillum group III would likely result in the collapse of the entire community. This is a potentially important observation, given that microbial communities in AMD systems control the rate of pyrite oxidation and thus AMD formation (30).
It is interesting to speculate as to why nitrogen fixation is strongly partitioned within the AMD biofilm communities. It is possible that the fitness of the community is enhanced by synergistic interactions among the dominant populations. For example, by analogy with rhizosphere communities, fixed nitrogen may be exported in return for certain metabolites (e.g., vitamins) or molecules involved in energy production or in return for aiding in the control of oxygen levels (11, 36). Synergistic interactions may be tested using future cultivation-b ased community assembly studies.
The isolation of members of lineages without cultivated representatives is a major challenge in microbiology (38). Directed isolation approaches have only used 16S rRNA-directed probes to track the progress of isolation attempts (14, 24) without prior knowledge of the targeted organisms physiology. Environmental genome data can provide detailed insight into the metabolic potential of uncultivated organisms. To date, genome sequences have only been used to optimize growth conditions of cultivated organisms such as Escherichia coli (13, 16). Although nitrogen fixation was an obvious basis for selective isolation, the cultivation of L. ferrodiazotrophum amply demonstrates the utility of environmental genome data for formulating isolation strategies for uncultivated organisms.
This research was funded by the U.S. Department of Energy Microbial Genome Program and the National Science Foundation Biocomplexity Program.
Present address: Microbial Ecology Program, Department of Energy Joint Genome Institute, Walnut Creek, CA 94598. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»