Coastal seawater was collected from the Woods Hole, MA, town pier using a rinsed bucket and transported to MIT in a 50L carbuoy; Prochlorococcus was undetected in this water by flow cytometry (per Moore et al., 1998), and total cell density was 4.15 x 10^6 by Sybr-stained flow cytometric counts. Prochlorococcus strains MED4, MIT9515, MIT9312 and MIT9313 were separately spiked into the natural samples, each to final cell concentrations ranging from ~10^1 to 10^6 cells/ml. Culture cell concentrations were measured using flow cytometry and necessary dilutions of cultures were made with 0.2μm-filtered Sargasso Sea water. Each aliquot of Woods Hole water with its spiked-in Prochlorococcus was filtered through a GF-A prefilter, then collected on a Supor-200 (#60300, Pall Corporation, Ann Arbor, Michigan) 25mm 0.2μm filter, using a MasterFlex peristaltic pump system (Cole-Parmer Instrument Company, Vernon Hills, Illinois). Filtered volumes ranged from 250ml to 1L. Filters were immediately frozen. All additions and filtrations were made within 24 hours of water collection.
Extracted molecule
genomic DNA
Extraction protocol
Extractions were a modification of a filter extraction protocol described previously (Suzuki et al, 2001). Filters were transferred to 2.0ml screw-top microcentrifuge tubes and 242μl of lysis buffer was added to each (lysis buffer: 40mM EDTA, 50mM Tris ph8.3, 0.73M sucrose, 1.15mg/ml lysozyme (Sigma, #L-6876), 200μg/ml RNase (Qiagen, Valencia, California, #1018048), 0.2μm-filter-sterilized). Samples were incubated at 37 deg C for 30 minutes, rotating. 13.5μl of a Proteinase K solution (10mg/ml (EMD, #24568-2) in 40mM EDTA, 50mM Tris ph8.3, 0.73M sucrose) was added, and SDS was added to a final concentration of 1%. Each sample was incubated at 55 deg C, rotating, overnight. The samples were then extracted with the DNeasy 96 Tissue kit (Qiagen, Valencia, CA), by a modification of the manufacturer’s protocol. Each tube received 300μl of Buffer AL (buffer AL/E without ethanol added), was vortexed, and incubated for 70 deg C for 10 minutes. Then 300μl of 99% ethanol was added to each, they were vortexed, and pipetted onto the 96-well spin plate. The plate was sealed with the Airpore sheets (supplied with kit) and spun. All spins were carried out at 40 deg C, 4612 x g in a Sorvall Legend RT centrifuge (Kendro Laboratory Products, Newtown, Connecticut). The plate was spun 10 minutes, 500μl Buffer AW1 was added to each well, the plate re-sealed and spun 5 minutes. 500μl Buffer AW2 was then added to each well, the plate resealed and spun 5 minutes. To dry the plate, the column portion was then transferred to a new rack of elution microtubes RS (supplied with kit) and incubated for 15 minutes at 70 deg C. To elute, 200μl Buffer AE pre-heated to 70 deg C was then added to each well, the plate was re-sealed, incubated 1 minute and spun for 2 minutes. The elution was repeated with an additional 200μl. The eluted DNA was then concentrated using Excela-Pure 96-well PCR purification kits (Edge BioSystems, Gaithersburg, Maryland), following the manufacturer’s protocol. Each well was rinsed once with 100μl nuclease-free water (#9937, Ambion, Austin, Texas), then resuspended in 20μl dilute TE (1mM Tris pH 8, 0.1mM EDTA pH 8), transferred to a clean 96-well plate, and stored at -20 deg C. Concentrations were measured by Nanodrop.
Label
Cy3
Label protocol
Target DNA was amplified and labeled using A/B/C random-amplification (Wang et al., 2003), with the modification that the initial reverse transcription step was omitted. Briefly, random-primed amplification was carried out in three reactions: Round A used Sequenase to extend Primer A (GTT TCC CAG TCA CGA TCN NNN NNN NN); Round B used 20 rounds of PCR to amplify the resulting fragments, using Primer B (GTT TCC CAG TCA CGA TC); and Round C used 10 rounds of PCR to incorporate amino-allyl-deoxyuridine triphosphates (aa-dUTP). For the environmental samples, the amount of DNA into each reaction was normalized to represent 70ml of filtered seawater. All A/B/C reactions were performed in triplicate and pooled. Amplification products were cleaned using a Microcon YM-30 and concentrated to 9μl in nuclease-free water, and labeled with Cy3 by combining 8μl aa-DNA, 2μl 0.5M NaHCO3 and 5μl Cy3 dye (33μg in DMSO), and incubating at room temperature in the dark for one hour. Samples were cleaned in a Microcon YM-30, and concentrated to 19μl in TE.
Hybridization protocol
17.33μl of labeled DNA was added to hybridization buffer for final concentrations of 3X SSC, 0.2% SDS, 0.4 mg/ml poly A, 0.02M HEPES, pH7, in a final volume of 25μl. Samples were denatured 4 minutes at 100 deg C, then pipetted onto the arrays. Arrays were hybridized overnight in a heating oven (Model 2000 Micro Hybridization Incubator, Robbins Scientific, Sunnyvale, California), then washed, first vigorously for 30 seconds in 0.6X SSC, 0.03% SDS, and second in 0.06X SSC vigorously for 30 seconds then gently for five minutes. For the data shown in this paper, hybridizations were done at 65 deg C and washes were done at room temperature.
Scan protocol
Hybridized arrays were scanned using an Axon Instruments 4000B scanner (Foster City, CA) and the data was normalized and filtered using perl scripts written for the purpose, by the steps described below.
Description
Environmental DNA with added Prochlorococcus strain 9515, 10^5 cells/ml
Data processing
(1) Signal intensities for each spot were calculated by subtracting the local background (mean F532 – median B532, as calculated by GenePix Pro 5.1 software, Axon Instruments). (2) The median value across replicates was calculated for each probe. (3) For each probe set, the number of probes greater than twice the mean negative control signal was calculated, before further processing. (4) Filter I: Arrays with less than half their positive control probes exceeding twice the mean negative control signal were considered poor quality, low dynamic range, arrays and were excluded from further analysis. (5) Each probe signal was corrected for non-specific binding by subtracting the mean negative control spot signal. (6) The data was then normalized for array-to-array variations in brightness by dividing each probe signal by the mean positive control signal. This positive control signal was the mean signal across the Halobacterium salinarum probes in each hybridization, with identical amounts of H. salinarum DNA having been added to each reaction prior to amplification and labeling. (7) Filter II: In order for a genotype to be considered “present”, at least 45% of its probes had to exceed twice the mean negative control signal. (8) Finally, each genotype signal was calculated as either the MEAN or TUKEY BIWEIGHT across its probe set. Finally, outlier arrays were identified as having normalized mean positive control signal less than 25% of the average across the experimental series, and were excluded from further analyses. Pre-processed data was imported into Excel for visualization.
