array scanning protocol
1) Scanner settings: a) Pixel resolution: 5um. b) Laser power: 100% c) PMT Gain: Varriable. The PMT gain setting was adjusted for each array to appreciate the maximum dynamic range of signal (0 to 65,000). This was accomplished by increasing the PMT gain (range is 86 to 862) until some spots had some saturated pixels. 2) MYcroarray slide orientation: MYcroarray glass slides have a notch (chamfor) in one corner. When the slide is held such that the notch is in the upper right hand corner, the array is on the facing surface and is oriented with the top of the array toward the notched end of the slide. MYcroarray slides were inserted into the Axon 4000B scanner cassette with the notched corner away from the operator and to the left. After scanning, the image orientation will match the .gal file feature extraction grid orientation. For the present study, all arrays were scanned at a PMT gain of 400. At a PMT gain setting of 400, the average number of spots per array with greater than 10% saturated pixels was 50.9±14 (out of a possible 38,946 Af spots). 3) Data Extraction: Data were extracted from scanned images with GenePix Pro Software (version 184.108.40.206). The gal file grid (called the “Array List” by GenePix Pro software) was manually positioned over the array image. Within Block Properties, Column Spacing, Row Spacing and Rotation (degrees) were micro-adjusted as necessary to align the block and feature indicators. Automatic spot finding algorithms (e.g., “Find Array”, “Find All Blocks” and “Align Features”) were not used. Compared to spot diameters (~60um), smaller feature indicator diameters (35um) were used to extract data from the center of the spot where the sequence fidelity is exceptional. Signal data at the periphery of the spot is not trustworthy because this is where there may be rogue sequences due to MYcroarray’s manufacturing process. Before data extraction, the images were carefully reviewed to assess individual feature quality and any feature observed to have a technical issue (i.e., occlusion by dust, bubble or deposit, or other issue) was manually flagged “bad” (-100 in the “flags” field in the results file). In the present study, 56 out of 350,514 spots were designated as “bad”. Following data extraction, the information is stored in a tab delimited GenePix Pro Results file (.gpr). Median feature pixel intensity data were used for further analysis.
nucleic acid hybridization to array protocol
The appropriate amount of labeled target was added to 180?L of hybridization solution master mix. An appropriate amount of nuclease free water was added to a final volume of 250?L. The final hybridization solution was composed of 6X SSPE (0.9M NaCl, 36mM NaH2PO4, 24mM Na2HPO4, 6mM EDTA), 10% Formamide (Sigma, #F9037), 0.01% Tween-20 (Sigma, #P9416) and 2.5?l (1 ?g/?l) fluorescently labeled control oligos (proprietary sequences for hybridization to MYcroarray QC probes). The hybridization solution was then heated to 65oC for 10 minutes followed by cooling on ice for 5 minutes. The hybridization chamber and gasket slide (2 gaskets per slide; Agilent Technologies) were pre-heated to 60oC. A gasket slide was fitted snugly against one end of a hybridization chamber and then the hybridization solution was pipette within one gasket. While held at approximately a 45 degree angle, carefully, one MYcroarray slide was inverted, fitted snugly against the same end of the hybridization chamber (as the gasket slide) and slowly lowered on top of the gasket slide creating a chamber for the hybridization solution with a mixing bubble contained within the chamber. The labeled end of the gasket slide and the bar coded end of the MYcroarray slide were oriented to opposite ends of the sandwich. The hybridization chamber was sealed tightly and placed in an Agilent Hybridization Oven (manufactured by Shel Lab). The rotational control was set to 6. Following hybridization, the arrays were washed before scanning. Each MYcroarray/gasket slide sandwich was quickly removed from the hybridization cassette and submerged in 1X SSPE (24C). The sandwich was carefully pried open with flat plastic forceps and the liberated MYcroarray slide was transferred to fresh 1X SSPE (24C). The present study utilized nine 40K MYcroarray slides hybridized independently with nine samples from three experimental conditions (three biologic replicates per experimental condition). The washing was done in three batches of three MYcroarray slides. Each wash batch contained MYcroarrays that had been hybridized to one sample from each of the three experimental conditions. MYcroarrays were stored briefly in 1X SSPE while sister MYcorarrays from one wash batch were liberated from their hybridization sandwich (estimated storage time was <2min). The wash sequence was as follows. Wash# [SSPE] Temp Time 1 1X 24C 3min 2 1X 24C 3min 3 1X 50C 5min 4 1X 24C 3min 5 0.5X 24C 30sec All washes were performed in baths while the wash solution was gently circulated by stir bar. Following the final wash, arrays were spun dry in a microarray mini-centrifuge (Arrayit).
nucleic acid labeling protocol
1) Production of amino allyl conjugated cRNA (aa-cRNA). Total RNA (1000ng) was converted to amino allyl conjugated cRNA using Life Technologies (Ambion) Amino Allyl MessageAmp™ II aRNA Amplification Kit (AM1753). Amino-allyl-UTP was incorporated into the cRNA during the IVT reaction. The manufacturer’s protocol was followed. The expected yield from 1000ng is about 200?g. 2) Fluorescent Labeling of aa-cRNA. One vial of amine reactive fluorescent dye (Alexa Fluor 555; Life Technologies A32756) was coupled to 55ug of aa-cRNA following the manufacturer’s instructions. Unincorporated dye was removed using RNeasy Mini Columns (Qiagen 74104). The Qiagen’s quick clean-up protocol was followed. 3) Fragmentation of fluor-labeled cRNA. Fluorescent dye labeled cRNA was fragmented using heat and Zn++ cations. The expected mean fragment length is 125-150 nucleotides. Fragmented cRNA is considered sufficient for use on MYcroarrays if the mean fragment length less than 200 nucleotides.
nucleic acid extraction protocol
Total RNA was isolated from Aspergillus fumigatus Af293. Frozen fungal pellets were physically disrupted by grinding with a mortar and pestle on dry ice (held at -80 degrees C prior to extraction) and the resulting powdered cell extracts were used for total RNA extraction using the RNeasy plant minikit (Qiagen,GmbH, Hilden, Germany) according to the manufacturers’ instructions. The presence of intact 28S and 18S ribosomal RNA bands was confirmed by gel electrophoresis on a 1 % agarose gel. rRNA ratio was determined using a Bioanalyzer (Agilent 2100 Bioanalyzer).
Lactobacillus plantarum 16 is grown for 48 h in MRS broth (oxoid) anaerobically at 30 degrees to obtain antifungal supernatant. Aspergillus fumigatus strain Af293 is grown on Sabouraud dextrose agar for approximately 4-5 days (or until sporulation occurs) at 30 degrees aerobically.
Aspergillus fumigatus Af293 spores (105 spores ml-1) were inoculated into 50 ml Sabouraud dextrose broth and incubated with agitation at 30 degrees C for 20 h. 2X concentrated cell-free supernatant (from Lb. plantarum 16)(v/v) was added to A. fumigatus Af293 culture. Samples were re-incubated at 30 degrees C for 10 min. 10 ml aliquots were withdrawn and centrifuged at 4,400 g for 10 min at 4 degrees C. The resulting fungal pellets were frozen immediately and stored at -80 degrees C prior to RNA extraction. A. fumigatus Af293 without the addition of cCFS served as an untreated control. Samples were prepared in triplicate.
normalization data transformation protocol
I) Background Correction II) Scaling III) Quantile normalization IV) Trimmed mean Background correction: Due to MYcroarray’s manufacturing process, the area immediately surrounding a spot is not suitable to estimate background levels. When MYcroarray probes do not bind target, the residual signal is almost always lower compared with the surrounding area. In most instances, local background values would over-estimate the real contribution of background signal in the extracted signal intensity of each probe. Therefore, a global background correction factor was used instead of a local background correction factor. The background value, within each array, was determined by the 5th percentile darkest spot from the distribution of median signal intensities for all “good” spots (i.e., control spots, empty spots, “bad” spots and spots with >10% saturated pixels were ignored for determining the background value).  Background corrected signal = (Median pixel signal – background) +C In other words, the median pixel signal value of the 5th percentile darkest spot (from the distribution of “good” spots for each array) was subtracted from the median pixel signal for each probe and then a constant was added. The constant was added to shift the background corrected signal distribution (of each array) such that the minimum background corrected value in the entire study was 2. In this study, C = 52. Scaling: To adjust for differences in signal intensity (i.e., the overall signal for each array), a scale factor was created. Only “good” probes that met the same criteria for the global background estimation (above) were used for scaling. The average signal for “good” spots was calculated for each array. A scale factor (SF) for arrayi was calculated as follows:  SF(Arrayi) = average “good” spot signal (Arrayi) / median of average “good” spot signal from all arrays The scaled for each probe on arrayi was calculated by multiplying the background corrected signal  by the scale factor  for arrayi.  Scaled signal = Background corrected signal x SF Quantile normalization The scaled signal data were quantile normalized using the ranked median quantiles. Briefly, the scaled signal for each probe on arrayi was ranked. The ranked signal value was replaced with the median quantile value of the same rank. To determine the median quantile values, the scaled signal data for each array was independently sorted (smallest value to highest value) without regard to ID. The sorted scaled signal data were merged and the median was calculated for each quantile and the resulting median quantiles were ranked. Trimmed mean: For the series of 4 identical probe replicates that survey each gene, a trimmed mean was calculated to generate the final normalized signal. Only normalized signals from “good” spots were used. The trimmed mean was calculated by discarding the maximum and minimum quantile normalized signal followed by averaging the remaining signal values. If a gene had three or less remaining “good” probe replicates, than the average of the remaining replicate signal was used.