sample: Pool of all coral samples included in the experiment.
Extracted molecule
total RNA
Extraction protocol
Qiazol used to extract total RNA. RNeasy Mini kit used to clean RNA. Details: Total RNA from all frozen coral fragments was isolated using Qiazol lysis reagent (QIAGEN). Live tissue was chiseled off each coral fragment and homogenized using a pre-chilled mortar and pestle embedded in dry ice. Frozen coral powder was transferred directly to Qiazol. Two chloroform extractions were performed, followed by isopropanol precipitation and 2 washes in 70% ethanol. RNA pellets were re-dissolved in nuclease-free water and cleaned further over RNeasy Mini columns (QIAGEN). RNA quality and quantity were assessed with a NanoDrop ND-1000 spectrophotometer and an Agilent 2100 Bioanalyzer. Microarray protocols followed those established by the Center for Advanced Technology at the University of California, San Francisco (http://cat.ucsf.edu/).
Label
Cy3
Label protocol
Labeled using Amersham CyDye Post-labeling kit. http://cat.ucsf.edu/pdfs/hybCourseProtocols_v2.7.pdf Specifically, 15ug of total RNA were primed with 10ug/uL of Oligo-dT primer for 10min at 70oC. Reverse transcription (RT) lasted for 2hr at 42oC using a master mix containing a 4:1 ratio of aminoallyl-dUTP to TTP. Following RT, single-stranded RNA was hydrolyzed by incubating the RT reactions in 10uL 0.5M EDTA and 10uL 1M NaOH for 15min at 65oC. After hydrolysis, RT reactions were cleaned using QIAGEN MinElute Reaction Purification columns. Cy3 and Cy5 dyes (GE Healthcare) were dissolved in 20uL DMSO, and the coupling reactions lasted for 2hr at room temperature in the dark.
Qiazol used to extract total RNA. RNeasy Mini kit used to clean RNA. Details: Total RNA from all frozen coral fragments was isolated using Qiazol lysis reagent (QIAGEN). Live tissue was chiseled off each coral fragment and homogenized using a pre-chilled mortar and pestle embedded in dry ice. Frozen coral powder was transferred directly to Qiazol. Two chloroform extractions were performed, followed by isopropanol precipitation and 2 washes in 70% ethanol. RNA pellets were re-dissolved in nuclease-free water and cleaned further over RNeasy Mini columns (QIAGEN). RNA quality and quantity were assessed with a NanoDrop ND-1000 spectrophotometer and an Agilent 2100 Bioanalyzer. Microarray protocols followed those established by the Center for Advanced Technology at the University of California, San Francisco (http://cat.ucsf.edu/).
Label
Cy5
Label protocol
Labeled using Amersham CyDye Post-labeling kit. http://cat.ucsf.edu/pdfs/hybCourseProtocols_v2.7.pdf Specifically, 15ug of total RNA were primed with 10ug/uL of Oligo-dT primer for 10min at 70oC. Reverse transcription (RT) lasted for 2hr at 42oC using a master mix containing a 4:1 ratio of aminoallyl-dUTP to TTP. Following RT, single-stranded RNA was hydrolyzed by incubating the RT reactions in 10uL 0.5M EDTA and 10uL 1M NaOH for 15min at 65oC. After hydrolysis, RT reactions were cleaned using QIAGEN MinElute Reaction Purification columns. Cy3 and Cy5 dyes (GE Healthcare) were dissolved in 20uL DMSO, and the coupling reactions lasted for 2hr at room temperature in the dark.
Hybridization protocol
Prior to hybridization, microarrays were post-processed by: 1) UV crosslinking at 60 mJ; 2) a “shampoo” treatment (3x SSC, 0.2% SDS at 65oC); 3) blocking with 5.5g succinic anhydride dissolved in 335mL 1-methyl-2-pyrrilidinone and 15mL sodium borate; and 4) drying via centrifugation. Dye-coupled cDNAs were cleaned (QIAGEN MinElute), and appropriate Cy3 and Cy5 labeled cDNAs were mixed together in a hybridization buffer containing 0.25% SDS, 25mM HEPES, and 3x SSC. The hybridization mixtures were boiled for 2min at 99oC then allowed to cool at room temperature for 5min. The cooled hybridization mixtures were pipetted under an mSeries Lifterslip (Erie Scientific), and hybridization took place in Corning hybridization chambers overnight at 63oC. Microarrays were washed twice in 0.6x SSC and 0.01% SDS followed by a rinse in 0.06x SSC and dried via centrifugation. Slides were immediately scanned using an Axon 4000B scanner.
Scan protocol
Scanned using Axon 4000B scanner with GenePix 6.0 software.
Description
Six fragments (9.5 ± 3.5cm2) from the top (2.7m), middle (3.7m), and bottom (5.2m) of one massive colony of Montastraea faveolata were collected with a hammer and chisel at “La Bocana” reef near Puerto Morelos, Quintana Roo, Mexico on 31 July 2007. The fragments were divided evenly between two aquaria (50L) that received a constant flow of seawater (~0.64L/min). Each aquarium was fit with a water pump connected to a spray-bar to provide constant water movement and aeration. Both aquaria were placed in a common pond with flowing water to buffer diurnal temperature fluctuations, and both aquaria were exposed to shaded ambient light. All coral fragments were mounted on plasticene and kept at a depth of ~7cm. On the night of 1 September, one 200-Watt aquarium heater was turned on in the treatment aquarium, and a second heater was turned on 3 days later. During the thermal stress experiment, the control aquarium received an average water temperature of 28.8 ± 1.2oC; the heated aquarium, 31.5 ± 1.1oC. PAR present during the thermal stress experiment averaged 420 ± 152 uE. On the night of 7 September, all fragments were frozen in liquid nitrogen.
Data processing
All microarrays were scanned using an Axon 4000B scanner (Molecular Devices), and TIFF images were generated with GenePix Pro 6.0 software. Gridding was performed using Spotfinder 3.1.1 (TIGR) with the Otsu segmentation method and background correction (the top 25% of background pixels were discarded prior to local background estimation). Using MIDAS 2.19 (TIGR), background-corrected data were LOWESS normalized, and in-slide duplicates were averaged. Log2 ratios were manually calculated, and data from all 18 hybridizations were compiled into a single file. Genes were included in statistical analyses only if there were data for two out of three hybridizations for a given category (i.e. top-control, top-treatment, middle-control, middle-treatment, etc…). All statistical analyses and clustering were performed in TMEV 4.0 (TIGR). A two-factor ANOVA was performed with “location” as factor one (3 groups, n=6 per group) and “treatment” as factor two (2 groups, n=9 per group). Differentially expressed genes were chosen at α = 0.05. To look for relatedness in gene expression, array trees were generated with complete linkage clustering according to Manhattan distance. Support values for tree nodes were generating by bootstrapping the gene expression data over 1000 replicates. Additionally, t-tests between control (n=3) and heat-stressed (n=3) samples were performed to identify differentially expressed genes within each region of the colony (top, middle, and bottom). One-way ANOVAs were also performed to look for differences between control top, middle, and bottom fragments, and again for heat-treated top, middle, and bottom fragments.
