Patent Application
20110212850
This application hereby incorporates the attached sequence listing in computer readable form and the attached Table 1 showing the sequences SEQ ID NOS:1-2805.
1. Field of the Invention
The present invention relates to methods of using probes and microarrays to measure multiple different stable isotopes in nucleic acids and identification and analysis of microbial communities.
2. Related Art
Identification of microorganisms responsible for specific metabolic processes remains a major challenge in environmental microbiology, one that requires the integration of multiple techniques.
Nucleic acid stable isotope probing (SIP) techniques (5, 6) are currently the most widely used means to directly connect specific substrate utilization to microbial identity, a grand challenge in the field of microbial ecology (7). For traditional SIP, natural microbial communities are incubated in the presence of a substrate enriched in a rare stable isotope (either 13C or 15N). The organisms, including their nucleic acids, incorporate the substrate and become isotopically enriched over time. Ultracentrifugation is used to separate isotopically enriched nucleic acids from lighter, unenriched nucleic acids for molecular analysis. In the past decade, these approaches have generated many advances in the understanding of microbial bioremediation, plant-microbe interactions and food web dynamics (8), yet they remain hindered by logistical drawbacks (9). These issues are intensified when working with density-gradient centrifugation of RNA, where the focus is on active organisms that are not necessarily replicating. Most notably, traditional DNA- and RNA-SIP isotope exposure risks fertilization effects by requiring high substrate concentrations in order to meet the sensitivity threshold of density gradient separation (in many systems >20% 13C DNA) (10) and is extremely difficult to perform with 15N labeled substrates (>40% 15N DNA required) (11). Other disadvantages include long exposure times (risking community cross-feeding), low-throughput (1-2 weeks lab processing time per sample batch), and incomplete quantification. Though related culture-independent approaches also have ideal qualities such as high sensitivity or in situ resolution (e.g. 13C-PLFA (12); EL FISH (13), FISH MAR (14), isotope arrays (15)), none combines high throughput, sensitivity, taxonomic resolution, and quantitative estimates of multiple stable isotope (15N and 13C) incorporation.
The present invention provides a method for quantification of stable isotope labeling using phylogenetic probes.
In another aspect, the present invention comprises community analysis using such phylogenetic probes.
The methods described have the ability to track the update of carbon, nitrogen and oxygen in ribonucleic acids and providing insight into how microorganisms metabolize these elements. The methods as described can track the uptake of carbon and nitrogen simultaneously and also be applied to oxygen. There is noother known method that can track the uptake of carbon and nitrogen simultaneously.
A method for determination of stable isotope incorporation in a organism or a community of organisms comprising the steps of: (a) supplying an organism or said community of organisms with a stable isotope labeled substrate for a defined period of time; (b) extracting RNA from the organisms; (c) fragmenting said RNA; (d) labeling a fraction the fragmented RNA with a detectable label; (e) hybridizing the labeled RNA to a set of oligonucleotide probes; (f) detecting hybridization signal strength of labeled RNA hybridized to any of the oligonucleotide probes and identifying and selecting the hybridized oligonucleotide probes as a responsive set of probes; (g) hybridizing a fraction of unlabeled RNA to a second set of oligonucleotide probes comprising the responsive set of probes; (h)detecting the unlabeled RNA hybridized to the responsive set of probes to determine the stable isotope incorporation into the organism using spectrometry or spectroscopy.
In one embodiment, the organism is a bacterium, archaea, fungi, plant, arthropod,or nematode, or other eukaryote. In a specific embodiment, the organism is a bacterium.
In one embodiment,the stable-isotope labeled substrate is 3H, 13C 15N, and/or 18O.
Extraction of RNA can be carried out by physical and/or chemical cell lysis and affinity column purification. Fragmentation is generally carried out by using either enzymes or chemicals or heat or a combination of these. A fraction or aliquot of the RNA is then labeled with a fluorescent molecule or a non-fluorescent molecule. Fragmentation and labeling can occur in some embodiments concurrently.
In one embodiment, the set of oligonucleotide probes comprising an array of oligonucleotide probes attached to a substrate such as a microarray or chip. The labeled fragmented RNA can then be added to a hybridization solution and the hybridization solution contacted to the array of oligonucleotide probes to allow the labeled RNA to hybridize to the probes.
In one embodiment, the set of oligonucleotide probes comprising 16S rRNA phylogenetic oligonucleotide probes. he set of 16S rRNA phylogenetic probes further comprising probes from the 16S rRNA gene, 23S rRNA gene, 5S rRNA gene, 5.8S rRNA gene, 12S rRNA gene, 18S rRNA gene, 28S rRNA gene, gyrB gene, rpoB gene, fusA gene, recA gene, cox1 gene, nif13 gene, RNA molecules derived therefrom, or a combination thereof.
The array with the hybridized labeled RNA is imaged with a fluorescence scanner and fluorescence intensity measured for each probe feature and the detection of hybridization signal strength provides a determination of the genes present in a organism or genes and/or organisms present in the community of organisms. The detection of hybridization signal strength also provides a means for normalization of the isotope signals detected.
In one embodiment, the probes that hybridized to the labeled RNA are synthesized onto a second array of oligonucleotide probes comprising down-selected probes or responsive probes. The unlabeled RNA is hybridized to the second array hybridized unlabeled RNA are imaged with a with a secondary ion mass spectrometer and isotope ratios are measured for each probe feature.
The presently described methods provide high throughput, sensitivity, taxonomic resolution, and quantitative estimates of multiple stable isotope (15N and 13C) incorporation. In one embodiment, microbial identity and function are connected by isolating rRNA from individual taxa through hybridization to phylogenetic probes. In one embodiment, the probes are displayed on a substrate surface, such as a custom glass microarray. After hybridization, these probe features are then analyzed for isotope enrichment. In some embodiments, the probes are analyzed using analysis techniques including but not limited to, spectrometry, spectroscopy, and quantitative secondary ion mass spectrometry imaging.
Direct NanoSIMS analysis is made possible by implementing a new surface chemistry for synthesis of DNA on conductive material. With this approach, thousands of unique phylogenetic probes assaying hundreds of taxa can be quickly analyzed from a single sample.
The present methods may be used in applications such as the evaluation of how certain organisms metabolize cellulose and what enzymes they use to do this; evaluation of what organisms have the ability to degrade pollutants in an environmental sample such as oil using water samples from the recent Gulf oil spill; or a study of carbon sequestration.
Initial experiments utilized a single bacterial strain (Pseudomonas stutzeri) grown on 13C glucose as the sole carbon source to determine the feasibility of successful hybridization of extracted RNA on the microarray surface, and detection of 13C from the hybridized RNA.
Referring now to
Little is known about organic carbon incorporation patterns in marine and estuarine environments, partly because the dominant organisms are uncultured and cannot be directly interrogated in the laboratory. We used the Chip-SIP method to test whether different taxa incorporate amino acids, fatty acids, and starch for their carbon growth requirements.
Thus, in one embodiment, the present invention provides methods for quantification of stable isotope labeling to observe and measure resource partitioning in microbial communities using phylogenetic probes. In one embodiment, the phylogenetic probes can be designed. In another embodiment, phylogenetic probes previously designed and provided in the previous applications hereby incorporated by reference can be used.
In one embodiment, such a method involves labeling microbial nucleic acids with stable isotope-labeled substrates (e.g, 13C-amino acids, cellulose or 15NH4). Current methods for stable-isotope probing require large quantities of label to be incorporated into nucleic acids prior to density gradient separation (e.g. refs. Radajewski S, Meson P, Parekh N R & Murrell J C 2000. Nature 403: 646-649; Manefield M., Whiteley, A. S., Griffiths, R. I. and Bailey, M. J. 2002. Appl. Environ. Microbiol. 68:5367-73), however the necessary quantities of labeled substrate often impose a significant disturbance on system energy and metabolite flux. The presently described approach is to capture ribosomal RNA using sequence specific probes targeting 16S rRNA (Brodie, E. L., T. Z. DeSantis, D. C. Joyner, S. M. Baek, J. T. Larsen, G. L. Andersen, T. C. Hazen, P. M. Richardson, D. J. Herman, T. K. Tokunaga, J. M. M. Wan, and M. K. Firestone. 2006. Appl. Environ. Microbiol. 72:6288-6298), and the captured RNA is then analyzed for isotope ratios. Microarrays represent the highest-throughput approach for RNA capture; combining this with analysis methods allows isotope ratios to be determined for potentially hundreds of species within complex communities.
In some embodiments, the methods provides for a method comprising steps as the following described process. An organism or multiple organisms, such as a community of organisms, are supplied with a stable-isotope (e.g., 3H, 13C, 15N, 18O) labeled substrate for a defined period of time. RNA is extracted from the organisms or community organisms using any number of established procedures as is known in the art.
The organism RNA is fragmented using known fragmentation methods including use of enzymes, chemicals or heat or a combination of these. A first fraction or an aliquot of fragmented RNA is labeled with a fluorescent molecule or a non-fluorescently labeled molecule such as biotin. This can occur concurrently with fragmentation in some embodiments.
The labeled fraction of fragmented RNA is added to a hybridization solution and hybridized to a microarray slide. Weakly bound RNA can be removed from the microarray surface by washing in solutions of varying stringency. The RNA that is hybridized to the probes are then imaged to detect hybridization signal strength and thereby quantify the labeled RNA to determine the community organism composition and also to correct and normalize the isotope signals in the RNA bound to each probe.
Currently the organism composition and normalization of isotope signal occurs on a different device than the fluorescent detection of hybridization signal strength and measurement of isotope ratio or isotope incorporation. In such a case, the fluorescent detection provides a subset of responsive probes that correlate to the presence of a specific gene and/or an organism in the sample or the community. After this detection, the organisms are identified and a down-selected probe analysis is carried out. New probes to identify an organism can be designed, or the same probes from the larger set of oligonucleotide probes can be used. For example, in some instances, sequence information generated from reverse-transcribed RNA (cDNA) from the same samples is used to select unique regions for probe design. The down-selected set of new or responsive probes is then synthesized and arrayed onto a separate substrate. A reserved fraction of RNA is then hybridized to the down-selected set of probes and imaged whereby the determination of the isotope incorporation into the organism using spectrometry or spectroscopy.
If a separate device to determine the isotope incorporation into the organism is not required, then a separate set of down-selected probes does not need to be made, but the determination made directly on the RNA hybridized to the larger set of probes.
These steps are meant to provide a basic process and one having skill the art should understand that optimizations and variations to the method are contemplated.
Examples of organisms that can be used in the present methods include but are not limited to, prokaryotic and eukaryotic organisms such as bacteria, archaea, fungi, plants, arthropods, nematodes, avians, mammals, and other eukaryotes, or viruses and phage. In one embodiment, the organism, multiple organisms or a community of organisms is bacteria, archaea, fungi, plants, arthropods, or nematodes. For larger organisms, a cell or tissue sample may be obtained and the RNA extracted from the sample.
The RNA extracted from the organisms may be the total RNA including ribosomal, messenger, and transfer RNA or it may be a subset of the total RNA.
The organisms are supplied with amino acids, cellulose or other labeled substrate containing a stable-isotope. Examples of such stable isotopes include but are limited to 3H, 13C, 15N, and/or 18O. Examples of such labeled substrate include 13C-amino acids, cellulose or 15NH4 labeled substrate.
The organisms are supplied the labeled substrate for a defined period of time, such as for several minutes, hours or days. In one embodiment, a microbial community is supplied a labeled substrate for a period of 12, 18, or 24 hours.
Extraction of RNA from the organisms are generally carried out using methods known in the art. Examples of RNA extraction methods for microbial communities are provided in the Examples. In one embodiment, physical and/or chemical cell lysis and affinity column purification is used to extract RNA from the organisms or the cell or tissue sample from the organisms.
Fragmentation of the RNA is often carried out using enzymes, chemicals or heat or any combination of these. A fraction or aliquot of the fragmented RNA is labeled with a fluorescent label for suitable detection or with a label having a known binding partner to which a detectable label can be attached. In another embodiment, the fragmented RNA is labeled with a fluorescent molecule such as Alexafluor 546. In some embodiments, the fragmented RNA is labeled with biotin to which a fluorescently labeled streptavidin can be bound.
After labeling a fraction of the RNA, hybridization of the fragmented labeled RNA to a set of oligonucleotide probes is carried out. The set of oligonucleotide probes is typically attached to a solid planar substrate or on a microarray slide. However, it is contemplated that the probes may be attached to spheres, or other beads or other types of substrates. The substrates often made of materials including but not limited to, silicon, glass, metals or semiconductor materials, polymers and plastics. The substrates may be coated with other metals or materials for specific properties. In one embodiment, the substrate is coated with indium tin oxide (ITO) to provide a conductive surface for NanoSIMS analysis. The oligonucleotide probes may be present in other analysis systems, including but not limited to bead or solution multiplex reaction platforms, or across multiple platforms, for example, Affymetrix GeneChip® Arrays, Illumina BeadChip® Arrays, Luminex xMAP® Technology, Agilent Two-Channel Arrays, MAGIChips (Analysis systems of Gel-immobilized Compounds) or the NanoString nCounter Analysis System. The Affymetrix (Santa Clara, Calif., USA) platform DNA arrays can have the oligonucleotide probes (approximately 25 mer) synthesized directly on the glass surface by a photolithography method at an approximate density of 10,000 molecules per μm2 (Chee et al., Science (1996) 274:610-614). Spotted DNA arrays use oligonucleotides that are synthesized individually at a predefined concentration and are applied to a chemically activated glass surface.
The oligonucleotide probes are probes generally of lengths that range from a few nucleotides to hundreds of bases in length, but are typically from about 10 mer to 50 mer, about 15 mer to 40 mer, or about 20 mer to about 30 mer in length.
In one embodiment, the oligonucleotide probes is a set of phylogenetic probes. In another embodiment, the phylogenetic probes comprising 16S rRNA phylogenetic probes. In one embodiment, the set of 16S rRNA phylogenetic probes further comprising probes from the 16S rRNA gene, 23S rRNA gene, 5S rRNA gene, 5.8S rRNA gene, 12S rRNA gene, 18S rRNA gene, 28S rRNA gene, gyrB gene, rpoB gene, fusA gene, recA gene, cox1 gene, nif13 gene, RNA molecules derived therefrom, or a combination thereof.
Features of phylogenetic microarrays of the invention include the use of multiple oligonucleotide probes for every known category of prokaryotic organisms for high-confidence detection, and the pairing of at least one mismatch probe for every perfectly matched probe to minimize the effect of nonspecific hybridization. In some embodiments, each perfect match probe corresponds to 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or more mismatch probes. These and other features, alone or in combination as described herein, make arrays of the invention extremely sensitive, allowing identification of very low levels of microorganisms.
Methods to design and select suitable probes and arrays for Chip-SIP analysis are described in detail in co-pending U.S. patent application Ser. No. 12/474,204, filed on May 28, 2009 published as US-2009-0291858-A1, and co-pending international application having application number PCT/US2010/040106, filed on Jun. 25, 2010, both of which are incorporated by reference in their entirety for all purposes.
In one embodiment, the 16s rRNA phylogenetic probes are provided on a microarray chip, such as the G2 Phylochip or the G3 Phylochip available from Phylotech, Inc. (Second Genome, Inc., San Francisco, Calif.) and Affymetrix (Santa Clara, Calif.).
Again, the RNA that is hybridized to the probes are then imaged to detect hybridization signal strength and thereby quantify the labeled RNA to determine the community organism composition and also to correct and normalize the isotope signals in the RNA bound to each probe.
In one embodiment, for analysis for microbial composition and normalization of isotope signals, microarrays hybridized with fluorescent/biotin labeled RNA are imaged with a fluorescence scanner and fluorescence intensity measured for each probe feature or “spot”. Arrays can be scanned using any suitable scanning device. Non-limiting examples of conventional microarray scanners include GeneChip Scanner 3000 or GeneArray Scanner, (Affymetrix, Santa Clara, Calif.); and ProScan Array (Perkin Elmer, Boston, Mass.); and can be equipped with lasers having resolutions of 10 pm or finer. The scanned image displays can be captured as a pixel image, saved, and analyzed by quantifying the pixel density (intensity) of each spot on the array using image quantification software (e.g., GeneChip Analysis system Analysis Suite, version 5.1 Affymetrix, Santa Clara, Calif.; and ImaGene 6.0, Biodiscovery Inc. Los Angeles, Calif., USA). For each probe, an individual signal value can be obtained through imaging parsing and conversion to xy-coordinates. Intensity summaries for each feature can be created and variance estimations among the pixels comprising a feature can be calculated.
With flow cytometry based detection systems, a representative fraction of microparticles in each sublot of microparticles can be examined. The individual sublots, also known as subsets, can be prepared so that microparticles within a sublot are relatively homogeneous, but differ in at least one distinguishing characteristic from microparticles in any other sublot. Therefore, the sublot to which a microparticle belongs can readily be determined from different sublots using conventional flow cytometry techniques as described in U.S. Pat. No. 6,449,562. Typically, a laser is shined on individual microparticles and at least three known classification parameter values measured: forward light scatter (C1) which generally correlates with size and refractive index; side light scatter (C2) which generally correlates with size; and fluorescent emission in at least one wavelength (C3) which generally results from the presence of fluorochrome incorporated into the labeled target sequence. Because microparticles from different subsets differ in at least one of the above listed classification parameters, and the classification parameters for each subset are known, a microparticle's sublot identity can be verified during flow cytometric analysis of the pool of microparticles in a single assay step and in real-time. For each sublot of microparticles representing a particular probe, the intensity of the hybridization signal can be calculated along with signal variance estimations after performing background subtraction.
In one embodiment, responsive probe-sets are then identified based on a set criteria. See
Various methods of mass spectrometry may be used in addition to detection using the present phylogenetic probes, such as nanoSIMS (nanoscale secondary ion mass spectrometry) or time-of-flight secondary ion mass spectrometry or other methods or means of spectrometry or spectroscopy. In other embodiments, the use of spectroscopic methods that may be employed include Raman spectroscopy or reflectance or absorbance spectroscopy. In one preferred embodiment, for analysis of isotope incorporation into organisms, microarrays hybridized with non-fluorescently labeled RNA are imaged with a secondary ion mass spectrometer, such as a SIMS or NanoSIMS device. In a specific embodiment, the NanoSIMS device is a NimbleGen MAS and the probe array is synthesized onto ITO-coated slides suitable for NanoSIMS analysis.
In some embodiments, sequence information generated from reverse-transcribed RNA (cDNA) from the same samples is used to select unique regions for probe design.
In another embodiment, the array of probes is synthesized on a substrate coated with Indium Tin Oxide (ITO) to provide a conductive surface for NanoSIMS analysis. For example, ranked PM probes plus corresponding MM probes are synthesized using the NimbleGen MAS on ITO-coated slides suitable for NanoSIMS analysis.
Current and future research will focus on the cellulose-degrading and N-fixing microorganisms found in the guts of the passalid beetle Odontotaenius disjunctus. This microbial community represents a naturally- selected highly-efficient lignocellulose degrading consortium, including Pichia stipitis, a yeast with high capacity for xylose fermentation (Nardi, J. B., C. M. Bee, L. A. Miller, N. H. Nguyen, S.-O. Suh, and M. Blackwell. 2006. Arthropod Struct. Devel. 35:57-68; Suh, S.-O., J. V. McHugh, D. Pollock, and M. Blackwell. 2005. Mycolog. Res. 109:261-265). RNA from beetles have been analyzed with LBNL's Phylochip (Brodie, E. L., T. Z. DeSantis, D. C. Joyner, S. M. Baek, J. T. Larsen, G. L. Andersen, T. C. Hazen, P. M. Richardson, D. J. Herman, T. K. Tokunaga, J. M. M. Wan, and M. K. Firestone. 2006. Appl. Environ. Microbiol. 72:6288-6298) and probes are being chosen for analysis based on signal intensity relative to background.
We have demonstrated the capability of the Chip-SIP method to link phylogenetic identity and biogeochemical function. We have achieved this by incubating natural microbial communities in the presence of isotope-enriched substrates and analyzing rRNA from those communities for isotopic enrichment in a taxon-specific manner using phylogenetic microarrays. This method can be applied to all microbial systems to advance our understanding of the microorganisms involved in the sequestration of soil and marine carbon, the deconstruction of biofuel feedstocks, biodegradation of organic pollutants and bioimmobilization of radionuclides and heavy metals.
In another embodiment, the phylogenetic probes and the present methods can be used by detecting how the labeled isotope is incorporated or expressed in an organism for resource partitioning. Observing what organisms are actively consuming of a labeled substrate can provide for identifying contaminant degraders, organisms metabolizing biofuel feed stocks and soil/marine organic matter, and optimizing or monitoring biostimulation of microbes for bioremediation as further examples.
To test the chip-SIP approach, we grew a single bacterial strain (Pseudomonas stutzeri) in a minimal medium with 13C-glucose as the sole carbon source and extracted its RNA. After fluorescent labeling, the RNA was hybridized to a microarray probe set consisting of >100 sequences targeting different regions of the P. stutzeri 16S rRNA gene. Measured isotopic enrichment of these probe spots strongly depended on the efficiency of target RNA hybridization, as quantified by fluorescence (
Before applying chip-SIP to natural communities, we sought to test its sensitivity and ability to discriminate a mixture of differentially labeled bacterial taxa. Two bacterial strains, Vibrio cholerae and Bacillus cereus, were grown separately to different 15N and 13C isotopic enrichments, then their combined RNA was hybridized to an array consisting of probe sets targeting each organism. Both 13C (
These present example describes the materials and methods used in the Examples.
Growth of single strains and incubation of field samples. Strains of Pseudomonas stutzeri ATTC 11607, Vibro cholerae ATCC 14104, and Bacillus cereus D17 were grown from −80° C. frozen stock in Luria-Bertani (LB) broth at 37° C. until late log phase, and transferred into 12C glucose-amended M9 minimal medium until late log phase. Then, a 10 μl aliquot was inoculated into 10 ml of M9 enriched in 13C glucose and/or 15N ammonium and the culture was again grown until late log phase. An enrichment of 10% 13C indicates 10% of the glucose in the medium was 99% enriched in 13C, and 90% of the glucose had natural carbon (1.1% 13C and 98.9% 12C). Cells were centrifuged, washed, and frozen at −80° C. Bulk measurements (by
Isotope Ratio Mass Spectrometry) showed that Pseudomonas cells grown in full 13C glucose were enriched between 680,000 and 900,000 permil, equivalent to 90 atm %.
For field experiments, surface water was collected at the public pier in Berkeley, Calif. USA (37°51′46.67″N, 122°19′3.23″W) and immediately brought back to the laboratory i cooler. Glass bottles (500 ml) were filled without air space and dark incubated at 14° C. For the first set of experiments, samples were simultaneously incubated with 50 μM 99 atm % 13C glucose and 200 μM 99 atm % 15N ammonium, and subsamples harvested after 2, 6, and 24 hrs by filtration through a 0.22 polycarbonate filter which was then immediately frozen at −80° C. Background concentrations of ammonium in San Francisco bay range from 1-14 μM (1); typically estuarine glucose concentrations are 5-100 nM (2). For the second set of experiments, water samples were incubated as described above with 8 μM mixed amino acids (99 atm % 13C and 99 atm % 15N labeled; Omicron), 500 μg L−1 algal fatty acids (98 atm % 13C; Omicron), or 50 μg L−1 nucleic acids (90 atm % 13C; RNeasy extract from 13C Pseudomonas stutzeri), collected by filtration after 12 hrs and frozen at −80° C. These substrate additions were designed to result in concentrations at the high end of what is typically measured in estuarine environments: 2-7 μM amino acids (3), 25 μg L−1 fatty acids (4) and 10 μg L−1 DNA (5).
RNA extraction and labeling. RNA from pelleted cells (laboratory strains) and filters (field samples) was extracted with the Qiagen RNEasy kit according to manufacturer's instructions, with slight modifications for the field samples. Filters were incubated in 200 μL TE buffer with 5 mg mL−1 lysozyme and vortexed for 10 min at RT. RLT buffer (800 μL, Qiagen) was added, vortexed, centrifuged, and the supernatant was transferred to a new tube. Ethanol (560 μl) was added, mixed gently, and the sample was applied to the provided minicolumn. The remaining manufacturer's protocol was subsequently followed. At this point, RNA samples were split: one fraction saved for fluorescent labeling (see below), the other was kept unlabeled for NanoSIMS analysis. This procedure was used because the fluorescent labeling protocol introduces background carbon (mostly 12C) that dilutes the 13C signal (data not shown). Alexafluor 546 labeling was done with the Ulysis kit (Invitrogen) for 10 min at 90° C. (2 μL RNA, 10 μL labeling buffer, 2 μL Alexafluor reagent), followed by fragmentation. All RNA (fluorescently labeled or not) was fragmented using 5× fragmentation buffer (Affymetrix) for 10 min at 90° C. before hybridization. Labeled RNA was purified using a Spin-OUT™minicolumn (Millipore), and RNA was concentrated by ethanol precipitation to a final concentration of 500 ng μL−1.
Target taxa selection by PhyloC hip analysis and de novo probe design. RNA extracts from SF Bay SIP experiment samples were treated with DNAse I and reverse-transcribed to produce cDNA using the Genechip Expression 3′ amplification one-cycle cDNA synthesis kit (Affymetrix). The cDNA was PCR amplified with bacterial and archaeal primers, fragmented, fluorescently labeled, and hybridized to the G2 PhyloChip which is described by E. L. Brodie et al., in “Application of a high-density oligonucleotide microarray approach to study bacterial population dynamics during uranium reduction and reoxidation.” Appl. Environ. Microbiol. 72, 6288 (2006) hereby incorporated by reference, and commercially available from Affymetrix (Santa Clara, Calif.) through Second Genome (San Francisco, Calif.).
Taxa (16S operating taxonomic units, OTU) considered to be present in the samples were identified based on 90% of the probes for that taxon being responsive, defined as the signal of the perfect match probe >1.3 times the signal from the mismatch probe. From approximately 1500 positively identified taxa, we chose a subset of 100 taxa commonly found in marine environments to target with chip-SIP. We also did not target OTUs previously identified from soil, sewage, and bioreactors as our goal was to characterize the activity of marine microorganisms. Using the Greengenes database (7) implemented in ARB (8), we designed 25 probes (25 by long), to create a ‘probe set’ for each taxon (Table 1; SEQ ID NOS: 1-2805), as well as general probes for the three domains of life. Probes for single laboratory strains (Pseudomonas stutzeri, Bacillus cereus, and Vibrio cholerae) were also designed with ARB (Table 1).
Microarray synthesis and hybridization. A custom conductive surface for the microarrays was used to eliminate charging during SIMS analysis. Glass slides coated with indium-tin oxide (ITO; Sigma) were treated with an alkyl phosphonate hydroxy-linker (patent pending) to provide a starting substrate for DNA synthesis. Custom-designed microarrays (spot size=17 μm) were synthesized using a photolabile deprotection strategy (9) on the LLNL Maskless Array Synthesizer (Roche Nimblegen, Madison, Wis.). Reagents for synthesis (Roche Nimblegen) were delivered through the Expedite (PerSeptive Biosystems) system. For quality control (to determine that DNA synthesis was successful), slides were hybridized with complimentary Arabidopsis calmodulin protein kinase 6 (CPK6) labeled with Cy3 (Integrated DNA Technologies), which hybridized to fiducial marks, probe spots with the complementary sequence synthesized throughout the array area. Probes targeting microbial taxa were arranged in a densely packed formation to decrease the total area analyzed by imaging secondary ion mass spectrometry (NanoSIMS). For array hybridization, RNA samples in 1× Hybridization buffer (NimbleGen) were placed in Nimblegen X4 mixer slides and incubated inside a Maui hybridization system (BioMicro® Systems) for 18 hrs at 42° C. and subsequently washed according to manufacturer's instructions (NimbleGen). Arrays with fluorescently labeled RNA were imaged with a Genepix 4000B fluorescence scanner at pmt=650 units. Arrays with non-fluorescently labeled RNA were marked with a diamond pen and also imaged with the fluorescence scanner to subsequently navigate to the analysis spots in the NanoSIMS. These spots were observable in the fluorescence image because fiducial probe spots were synthesized around the outline of the area to be analyzed by NanoSIMS. Prior to NanoSIMS analysis, samples were not metal coated to avoid further dilution of the RNA's isotope ratio or loss of material. Slides were trimmed and mounted in custom-built stainless steel holders.
NanoSIMS analyses. Secondary ion mass spectrometry analysis of microarrays hybridized with 13C and/or 15N rRNA was performed at LLNL with a Cameca NanoSIMS 50 (Cameca, Gennevilliers, France). A Cs+ primary ion beam was used to enhance the generation of negative secondary ions. Carbon and nitrogen isotopic ratios were determined by electrostatic peak switching on electron multipliers in pulse counting mode, alternately measuring 12C14N− and 12C15N− simultaneously for the 15N/14N ratio, and then simultaneously measuring 12C14N− and 13C14N− for the 13C/12C ratio. We used this peak switching strategy because the secondary ion count rate for the CN− species in these samples is 5-10 times higher than any of the other carbon species (e.g., C−, CH−, C2−), and therefore higher precision was achieved even though total analytical time was split between the two CN− species at mass 27. If only one isotopic ratio was needed, peak switching was not performed. Mass resolution was set to ˜10,000 mass resolving power to minimize the contribution of isobaric interferences to the species of interest (e.g., 11B16O− contribution to 13C14N−< 1/100; 13C2− contribution to 12C14N−< 1/1000). Analyses were performed in imaging mode to generate digital ion images of the sample for each ion species. Analytical conditions were optimized for speed of analysis, ability to spatially resolve adjacent hybridization locations, and analytical stability. The primary beam current was 5 to 7 pA Cs+, which yielded a spatial resolution of 200-400 nm and a maximum count rate on the detectors of ˜300,000 cps 12C14N. Analysis area was 50×50 μm2 with a pixel density of 256×256 with 0.5 or 1 ms/pixel dwell time. For peak switching, one scan of the analysis area was made per species set, resulting in two scans per analytical cycle. With these conditions, reproducible secondary ion ratios could be measured for a maximum of 4 cycles through the two sets of measurements before the sample was largely consumed. Data were collected for 2 to 4 cycles. Based on total counts for the analyzed cycles, we achieved precision of 2-3% for 13C14N and 1-4% for 15N12C, depending on the enrichment and hybridization intensity. A single microarray analysis of approximately 2500 probes, with an area of 0.75 mm2 and the acquisition of 300 images, was carried out using the Cameca software automated chain analysis in 16 hours. Ion images were stitched together and processed to generate isotopic ratios with custom software (LIMAGE, L. Nittler, Carnegie Institution of Washington). Ion counts were corrected for detector dead time on a pixel by pixel basis. Hybridization locations were selected by hand or with the auto-ROI function, and ratios were calculated for the selected regions over all cycles to produce the location isotopic ratios. Isotopic ratios were converted to delta values using δ=[(Rmeas/Rstandard)−1]×1000, where Rmeas is the measured ratio and Rstandard is the standard ratio (0.00367 for 15N/14N and 0.011237 for 13C/12C). Data were corrected for natural abundance ratios measured in unhybridized locations of the sample.
Data analyses. For each taxon, isotopic enrichment of individual probe spots was plotted against fluorescence and the linear regression slope was calculated with the y-intercept constrained to natural isotope abundances (zero permil for 15N data and −20 permil for 13C data). This calculated slope (permil/fluorescence), which we refer to as the ‘hybridization-corrected enrichment (HCE), is a metric that can be used to compare the relative incorporation of a given substrate by different taxa. It should be noted that due to the different natural concentrations of 13C and 15N, and more importantly, different background contributions from the microarray, HCEs for 15N substrates and 13C substrates are not comparable. To construct a network diagram (e.g.
In a second set of experiments, we tested the viability of chip-SIP for a diverse natural community, using a sample from the San Francisco (SF) Bay, a eutrophic estuary. The bay water was incubated in the dark with micromolar concentrations of 15N ammonium and 13C glucose for 24 hrs, a timescale long enough to ensure detectable isotopic labeling of the dominant active community. We expected the most active taxa to incorporate these substrates, as they are of small molecular weight, do not require extracellular breakdown before uptake, and directly feed into central carbon and nitrogen metabolic pathways. This chip-SIP array consisted of 2500 probes targeting 100 microbial taxa selected from a PhyloChip analysis of the same sample (Table 1; 16). Based on RNA fluorescence, we positively detected 73 taxa. As in the experiments with laboratory cultures, the relationship between fluorescence and isotopic incorporation for each taxon was positive and linear for both 15N and 13C (e.g.
An advantage of chip-SIP's ability to detect 13C and 15N on the same array is its potential to uncover physiological diversity, based on the relative incorporation of two substrates incubated simultaneously. Our ability to measure taxon-specific substrate incorporation allowed us to reveal that the relationship between ammonium and glucose incorporation was linear: organisms with high ammonium incorporation (high 15N HCEs) also exhibited high glucose incorporation (high 13C HCEs), and vice versa (
Marine microorganisms, most of which remain uncultivated, control the release, transformation, and remineralization of ˜50 Gigatons of fixed carbon annually, resulting in biological carbon sequestration to the deep sea (P. Falkowski et al., The global carbon cycle: a test of our knowledge of earth as a system. Science 290, 291 (2000)). Identifying the microbes responsible for C cycling processes in the marine microbial loop and the factors affecting C cycling rates in marine ecosystems is a critical precursor to the development of predictive models of microbial responses to environmental perturbations (e.g., pollution, nutrient inputs or global change). Currently, the ecological niches of marine microorganisms, heterotrophic bacteria in particular, are often categorized as “copiotrophic” or “oligotrophic” depending on their predominant location, for example high-nutrient and particle-rich coasts versus low-nutrient open oceans, or warm, well-lit, productive surface waters versus the cold, dark deep (S. J. Giovannoni, U. Stingl, Molecular diversity and ecology of microbial plankton. Nature 437, 343 (2005)). The advent of 16S rRNA sequencing and environmental genomics have revolutionized marine microbial ecology by assembling a “parts list” of genetic diversity (M. S. Rappé, P. F. Kemp, S. J. Giovannoni, Phylogenetic diversity of marine coastal picoplankton 16S rRNA genes cloned from the continental shalf off Cape Hatteras, N.C. Limnol. Oceanogr. 42, 811 (1997)) and functional capability (E. F. DeLong et al., Community genomics among stratified microbial assemblages in the ocean's Interior. Science 311, 496 (2006)), but the goal of linking phylogenetic identity and in situ functional roles of uncultivated microorganisms remains largely unattained. In addition, while the comparative ‘omics strategy to gain ecosystem functional information has been fruitful, it relies on sequence comparison rather than direct measurements of biogeochemical activity. To gain a mechanistic understanding of microbial control of biogeochemical cycles in the ocean and elsewhere, it is necessary to move beyond microbial diversity or metagenomic surveys towards trait-based functional studies that directly and simultaneously measure the biogeochemical activities of hundreds of microbial taxa in their native environment.
In a third set of experiments, we compared predicted and actual substrate use of three organic substrates by a diverse natural community, an example of the type of experiment that can eventually lead to more realistic models of marine food web structure (20). In this case, we applied chip-SIP to another set of SF Bay samples incubated separately with isotopically-labeled amino acids, nucleic acids, and fatty acids. These substrates make up a significant proportion of photoautotrophic biomass (21) that provide the majority of fixed carbon substrates for the marine microbial food web.
We detected isotopic enrichment of at least one of the three added substrates in 52 out of the 81 taxa with positive RNA hybridization (
To compare genome-predicted potential biogeochemical activity to our measured substrate incorporation data, we examined the presence of genes involved in the extracellular degradation or transport of these substrates in the sequenced genomes of marine bacterial isolates (Table 2). Table 2 is shown below:
Agreia sp. PHSC20C1
Algoriphagus sp. PR1
Aurantimonas sp. SI85-9A1
Bacillus sp. B14905
Bacillus sp. NRRL B-14911
Bacillus sp. SG-1
Beggiatoa sp. PS
Bermanella marisrubri
Blastopirellula marina DSM 3645
Caminibacter mediatlanticus TB-2
Candidatus Blochmannia
pennsylvanicus BPEN
Candidatus Pelagibacter ubique
Carnobacterium sp. AT7
Congregibacter litoralis KT71
Croceibacter atlanticus HTCC2559
Cyanothece sp. CCY 0110
Dokdonia donghaensis MED134
Erythrobacter litoralis HTCC2594
Erythrobacter sp. NAP1
Erythrobacter sp. SD-21
Finegoldia magna ATCC 29328
Flavobacteria bacterium BAL38
Flavobacteria bacterium BBFL7
Flavobacteriales bacterium ALC-1
Flavobacteriales bacterium HTCC2170
Fulvimarina pelagi HTCC2506
Hoeflea phototrophica DFL-43
Hydrogenivirga sp. 128-5-R1-1
Idiomarina baltica OS145
Janibacter sp. HTCC2649
Kordia algicida OT-1
Labrenzia aggregata IAM 12614
Leeuwenhoekiella blandensis MED217
Lentisphaera araneosa HTCC2155
Limnobacter sp. MED105
Loktanella vestfoldensis SKA53
Lyngbya sp. PCC 8106
Marinobacter algicola DG893
Marinobacter sp. ELB17
Marinomonas sp. MED121
Mariprofundus ferrooxydans PV-1
Methylophilales bacterium HTCC2181
Microscilla marina ATCC 23134
Moritella sp. PE36
Neptuniibacter caesariensis
Nisaea sp. BAL199
Nitrobacter sp. Nb-311A
Nitrococcus mobilis Nb-231
Nodularia spumigena CCY9414
Oceanibulbus indolifex HEL-45
Oceanicaulis alexandrii HTCC2633
Oceanicola batsensis HTCC2597
Oceanicola granulosus HTCC2516
Parvularcula bermudensis HTCC2503
Pedobacter sp. BAL39
Pelotomaculum thermopropionicum SI
Phaeobacter gallaeciensis 2.10
Phaeobacter gallaeciensis BS107
Photobacterium angustum S14
Photobacterium profundum 3TCK
Photobacterium sp. SKA34
Planctomyces maris DSM 8797
Plesiocystis pacifica SIR-1
Polaribacter irgensii 23-P
Polaribacter sp. MED152
Prochlorococcus marinus AS9601
Prochlorococcus marinus MIT 9211
Prochlorococcus marinus MIT 9301
Prochlorococcus marinus MIT 9303
Prochlorococcus marinus MIT 9515
Prochlorococcus marinus NATL1A
Pseudoalteromonas sp. TW-7
Pseudoalteromonas tunicata D2
Psychroflexus torquis ATCC 700755
Psychromonas sp. CNPT3
Reinekea sp. MED297
Rhodobacterales bacterium HTCC2150
Rhodobacterales bacterium HTCC2654
Rhodobacterales sp. HTCC2255
Roseobacter litoralis Och 149
Roseobacter sp. AzwK-3b
Roseobacter sp. CCS2
Roseobacter sp. MED193
Roseobacter sp. SK209-2-6
Roseovarius nubinhibens ISM
Roseovarius sp. 217
Roseovarius sp. HTCC2601
Roseovarius sp. TM1035
Sagittula stellata E-37
Shewanella benthica KT99
Sphingomonas sp. SKA58
Sulfitobacter sp. EE-36
Sulfitobacter sp. NAS-14.1
Synechococcus sp. BL107
Synechococcus sp. RS9916
Synechococcus sp. RS9917
Synechococcus sp. WH 5701
Synechococcus sp. WH 7805
Ulvibacter sp. SCB49
Vibrio alginolyticus 12G01
Vibrio campbellii AND4
Vibrio harveyi HY01
Vibrio shilonii AK1
Vibrio sp. MED222
Vibrio splendidus 12B01
Vibrionales bacterium SWAT-3
Incorporation of leucine and other amino acids is routinely used as a proxy for bacterial production in aquatic systems (22) and metatranscriptomic evidence suggests most marine bacterial taxa incorporate amino acids directly (23). As nearly all genomes of marine bacteria (106/110) possess annotated putative amino acid transporters, we expected most of the active microbes in the SF Bay system would incorporate amino acids. Bacterial uptake of single nucleosides (e.g. thymidine) is ubiquitous and used to measure rates of growth (24), but only a few studies have examined longer nucleic acid molecules as a source of carbon or nitrogen for microbial metabolism (see ref. 25 as a recent example). Considering that half (55/110) of fully sequenced marine bacterial genomes contain at least one identified nucleoside transporter or extracellular nuclease, we expected nucleic acid incorporation could be a common phenomenon in the environment. Finally, we also chose to examine fatty acid incorporation because marine bacterial isolates commonly reveal high lipase activity (26), although only 38/110 sequenced bacterial genomes contained identified lipid transporters. In addition, comparative genomics has shown that oligotrophic marine bacterial genomes contain significantly more genes for lipid metabolism and fatty acid degradation than copiotrophic genomes (27). If oligotrophs favor fatty acid incorporation, we hypothesized that it would be less common than amino acid incorporation in our samples since a eutrophic estuary should favor copiotrophs.
In general terms, our results agree with predictions made from available marine genomic data: amino acids were the most commonly incorporated (46 taxa), followed by nucleic acids (32 taxa) and then fatty acids (18 taxa). However, the chip-SIP and genomic data did not always concur. For example, all the Vibrio genomes we examined contain putative enzymes for the utilization of the three substrates tested (Table 2), yet chip-SIP indicates the Vibrio taxa we detected incorporated only amino acids (
A frequently accepted, although increasingly controversial view in microbial ecology (29), maintains that 16S phylogeny is closely related to functional role. It is widely assumed that taxa that are closely related by 16S phylogeny are more likely to be functionally similar than to taxa more phylogenetically distant. This concept has been a major assumption of microbial ecology research, without which 16S diversity surveys lose their functional context. Chip-SIP allowed us to test this assumption by matching functional in situ resource use to 16S phylogenetic relationships.
As an example, we mapped substrate utilization data across a subset of the Gammaproteobacterial phylogeny (
Based on the success of these initial experiments, chip-SIP may facilitate great strides in our understanding of the functional mechanisms that underlie patterns of microbial diversity. Using this high resolution, high-sensitivity approach, we have revealed patterns of resource utilization in an estuarine community with critical implications for our understanding of carbon cycling in marine environments. These data considerably expand upon previous studies that have identified marine bacterial resource partitioning based on seasonal and small-scale spatial habitat use (30) by adding relative rates of substrate utilization as a critical component of the bacterial niche.
The above examples are provided to illustrate the invention but not to limit its scope. Other variants of the invention will be readily apparent to one of ordinary skill in the art and are encompassed by the appended claims. All publications, databases, and patents cited herein are hereby incorporated by reference for all purposes.
This application claims priority to U.S. Provisional Patent Application No. 61/302,535, filed on Feb. 8, 2010 and to U.S. Provisional Patent Application No. 61/302,827 filed on Feb. 9, 2010, both of which are hereby incorporated by reference. This application is related to concurrently filed U.S. patent application No. ______, filed today on Feb. 8, 2011, entitled “Devices, Methods and Systems for Targeted Detection (Attorney Docket No. IL-12105),” which claims priority to these same U.S. Provisional Patent applications and is hereby incorporated by reference in its entirety.
This invention was made in part by the US DOE Office of Biological and Environmental Research Genomic Sciences research program and the LLNL Laboratory Directed Research and Development (LDRD) program with government support under Contract No. DE-AC02-05CH11231 and under Contract DE-AC52-07NA27344 awarded by the U.S. Department of Energy. The government has certain rights in the invention.
| Number | Date | Country | |
|---|---|---|---|
| 61302535 | Feb 2010 | US | |
| 61302827 | Feb 2010 | US |
