Adult male Swiss Webster mice (30-35 g) (Harlan, Indianapolis, IN) were maintained in a controlled environment with a 12:12-hrs light-dark cycle and food and water administered ad libitum. Experiments were performed on mice anesthetized to a deep plane of surgical anesthesia with an intraperitoneal injection of pentobarbital sodium (50 mg/kg). Additional doses were administered as necessary by monitoring respiration and heart rate in each animal. Core temperature of the animals was maintained at 37ºC using a temperature controlled environmental chamber and monitored via a rectal temperature probe. Surgical preparations comprised a small incision (0.5 cm) proximal to the iliac crest to expose the sciatic nerve as previously described (65). In brief, a pair of bipolar hook electrodes was implanted adjacent to the nerve and the wound was sealed with cyanoacrylate glue. The hindlimb was subsequently immobilized by positioning the animal prone on a plexiglass platform fitted with a patellar brace to fix the knee. The animal was positioned with the chest wall over an aperture designed to fit physiologic sensors built into the platform. A length of 2.0 silk suture was attached to the Achilles tendon proximal to the calcaneous process and this ligature was then connected to a force transducer (Grass FT10; Astromed Inc, West Warwick, RI) to record the performance of the gastrocnemius-plantarus-soleus muscle group throughout the stimulation time course. Prior to each experiment, the voltage-force relation was determined to establish supramaximal stimulation conditions and the length-tension relation was determined to set the resting length for maximum twitch tension development. Contractions were induced by sciatic nerve stimulation using a Grass S48 stimulator (0.5 msec duration, 2-5 volts). The muscles were allowed to rest 15 minutes after these preliminary studies to allow for full metabolic recovery at physiologic temperatures (82). Supramaximal stimulation was applied at a rate of 10 Hz, the standard frequency employed in previous studies but for a shorter duration (4 hours) (34, 35, 67, 68). Force was digitally captured using an AT MIO 16E A/D board and software (National Instruments, Austin, TX) and analyzed using custom MatLab software (MathWorks, Natick, MA) (31). Following cessation of the stimulation period, test twitches were evaluated to confirm muscle recovery in each animal prior to tissue harvest. At the end of each experiment the EDL and soleus muscles were carefully dissected and either flash frozen in liquid nitrogen and stored at -80ºC for later analysis of mRNA expression or fixed at resting length in 10% neutral buffered formalin for histological processing.
Extracted molecule
total RNA
Extraction protocol
Frozen tissue specimens were transferred to 15 ml tubes (Sarstedt, Inc. Newton, NC) containing a ten-fold volume of TRIzol based on tissue weight (1ml/0.1mg; Life Technologies/Gibco, Gaithersburg,MD) and subjected to RNA purification as previously described (38). Homogenization was performed in these tubes followed by incubation at room temperature to permit complete dissociation of nucleoprotein complexes. Addition of 0.2 ml of chloroform per 1 ml of TRIzol reagent and centrifugation was performed to allow phase separation of the RNA. Total RNA was precipitated by mixing the aqueous phase with isopropanol. Samples were subsequently incubated, centrifuged and the supernatant removed. The pellet was washed with 75% ethanol, vortexed, centrifuged and air-dried followed by reconstitution in 100 ul of nuclease free water. For cultured cells, TRIzol was added directly to culture plates (5mls/100mm plate) after they were rinsed with ice cold PBS. The cells were harvested with a cell scraper into 15 ml tubes followed by homogenization. The aqueous layer was collected after organic phase separation as previously described and RNA was precipitated, washed with ethanol and resuspended in nuclease-free water as for the whole muscles. Criteria for inclusion in microarray studies was a spectrophotometric absorption ratio 260/280>1.8 (NanoDrop, Wilmington, DE) and RIN value >0.8 via electrophoretic analysis (Agilent Bioanalyzer 2100; Agilent Technologies, Santa Clara, CA).
Label
Biotin
Label protocol
In vitro transcription (IVT) of total RNA utilized the CodeLink™ Target Preparation Protocol (GE CodeLink™ Microarray System, Piscataway, NJ) as previously described (38). Briefly, 5 micrograms of purified RNA underwent cDNA synthesis using Invitrogen First and Second strand cDNA kits (Invitrogen, Austin,TX). Six bacterial control mRNAs were amplified in parallel to serve as fiducial markers and for intensity normalization during microarray scanning. Purification of double-stranded cDNA was performed using the QIAquick purification kit (Qiagen, Germantown, MD) and cRNA was synthesized via reverse transcriptase (Ambion, Austin, TX) in biotin-11-UTP (Perkin Elmer; Boston, MA) by incubation for 14 hours in a water bath at 37C. Biotin-labeled cRNA was recovered using an RNeasy Mini kit (Qiagen, Germantown, MD) and quantitation of the cRNA was performed by UV spectrophotometry at 260 nm. Confirmation of cRNA diversity was obtained using the Bioanalyzer 2100 to generate an electrophoretogram for each IVT reaction substrate which was characterized regarding sample yield, integrity and size diversity against a calibrated laboratory RNA standard. Based on evaluation of bacterial sequences incorporated as controls in the IVT assay, this step resulted in amplification of mRNA by only 50 to 100-fold thereby reducing the potential for errors associated with primer specific exponential amplification (77).
Hybridization protocol
CodeLink™ fragmentation buffer was added to the cRNA (5ul buffer/10ug cRNA) and incubated 20 minutes at 94C. Fragmented cRNA was suspended in hybridization solution containing nuclease-free water such that 250ul loading volume per array contained 10ug of cRNA. The solution was vortexed, heated (5 minutes at 90C) and transferred to ice prior to loading. Each slide contained a removable plastic hybridization chamber with an infusion port that was sealed immediately after loading. Microarrays were placed in a mixing incubator (Innova 4080 Shaking incubator, New Brunswick Scientific Co., Edison, NJ) for 18 hours, 300 rpm at 37C. CodeLink arrays were selected because of their high sensitivity and low error rate compared to other platforms especially regarding transcripts expressed at low copy numbers (73). After hybridization, arrays were individually taken from the incubator and their hybridization chambers were removed. The slides were placed in a reservoir containing 0.75X TNT (0.1M Tris-HCl, pH 7.6, 0.15 M NaCl, 0.05% Tween-20) for a 1 hour high-stringency wash at 46C. The arrays were incubated in Streptavidin Alexa Fluor 647 (Molecular Probes, Eugene, OR) for 30 minutes (0.2% Alexafluor-647 in TNB=0.1 M Tris-HCl, 0.15 M NaCl, 0.5% NEN Blocking Reagent, pH 7.6 (Perkin Elmer, Boston, MA). Slides were subsequently washed (1XTNT-5 min. each of 4 washes at RT; 0.1% SSC and 0.05% Tween20, 30 sec. at RT) and dried by low speed centrifugation.
Scan protocol
Slides were scanned using a 4000B GenePix scanner (Axon Instruments, Foster City, CA) that was calibrated prior to each use. Laser scanning parameters were set at 635 nm, PMT voltage at 600, resolution of 10 microns and the scan area adjusted for the entire array. Analysis of the array image was configured using the CodeLink™ Expression Analysis software (version 5.0). The program generates raw intensity outputs for each spot by segmentation and detection of spot intensity versus surrounding background. Data for each spot was determined as intensity per pixel within the probe zone after local background subtraction. The data for the array were then generated as both raw intensity values and normalized for the large dynamic range by dividing each spot by the overall median intensity value for that array. Values derived from poor or questionable spot profiles were removed from further computation as were all manufacturing errors designated by the CodeLink™ manufacturing spot report.
Description
Biotinylated cRNA probe is prepared and hybridization is carried out following the CodeLink protocols and reagents
Data processing
Raw intensities were normalized by CodeLink software based on global median for each array. CodeLink also calculated a threshold value based on the negative controls. The raw threshold value for this array is 55.558 and the normalized threshold value for this array is 1.058. To obtain the same data set we used for analysis, the normalized threshold value must be subtracted from each probe in the "value" column.
