Experimental Procedures and Protocols
For growing large numbers of worms and synchronizing them, see http://cmgm.stanford.edu/~kimlab/wormliquidculture.html
For the germ line experiments, 5-8 150mm plates were harvested for each timepoint, and provided sufficient RNA for at least one, and usually many more, hybridizations. To these plates, 8 ml M9 (6g Na2HPO4, 3g KH2PO4, 5g NaCl, 0.25g MgSO4 per liter) was added to each plate. To avoid temperature shock, the M9 was pre-incubated at 25°C. The plates were slowly rotated on a shaker for 2 minutes, and then the M9 and worms were pipetted into 15ml Falcon tubes and spun at the fastest setting in a clinical centrifuge for ~1 minute. During this spin, the plates were washed a second time in 5ml of M9 per plate to collect the residual worms. After removal of the supernatant from Falcon tubes, the second wash was added and the tubes spun again. The number of tubes was consolidated by washing the pellets with fresh M9 once or twice more, and combining the pellets after each wash. The final pellet size in a tube was not more than 2.5ml in a 15ml conical. At this point, the first few steps of the RNA isolation were completed.
Total RNA Isolation:
To each worm pellet, 4 ml Gibco BRL Trizol was added per ml of packed worms, and vortexed vigorously until completely resuspended (longer than 1 min). This slurry was flash frozen in liquid N2, thawed at 37°C, and then vortexed. The freeze/thaw cycle was repeated, and the mixture stored at -80°C until all samples were collected. After thawing, 2ml Trizol/ml starting packed worms was added, and vortexed vigorously. Per ml of starting packed worms, 2 ml CHCl3 was added, the tubes were shaken by hand for 15 seconds, and then allowed to sit at RT for 3 minutes. The tubes were then centrifuged at 12,000g, for 15 minutes at 4°C. The top aqueous layer was removed to a fresh 15-ml conical tube, taking care not to disturb the interphase. An equal volume of isopropanol was added, and the tubes mixed well, and then allowed to sit at RT for10 minutes. The RNA pellets were then washed in10 ml 75% EtOH, spun at 7500g, for 5 minutes at 4°C. The pellets were briefly air-dried, and then dissolved in DEPC-ddH2O (0.5-1 ml per ml of starting packed worms). The concentration of the RNA was determined by spectrophotometer, and the quality of the RNA was checked by running 1ul on a standard 1% TAE gel.
mRNA Isolation (adapted from Joe DeRisi)
1. The oligo-dT cellulose was prepared by resuspending 1g (1 vial of Ambion oligo-dT cellulose) in 10 ml 1x NETS (100mM NaCl, 10mM Tris-Cl pH7.4, 10mM EDTA, 0.2% SDS), and then washing the cellulose 3 times with 10 ml 1x NETS in a 15-ml conical tube. The washes were spun down in a clinical centrifuge. The resin was suspended in 10 ml 2x NETS, for a final volume of ~13 ml.
To run a column, 1 mg total RNA was diluted to 1 ml with 10mM Tris-Cl pH 7.4, and then mixed with 1 ml resin as prepared above in a 2 ml Biorad minicolumn that is sealed on the bottom. The column was capped and the RNA was allowed to bind to the resin by placing on a rotator or nutator for 1 hour at RT. The bottom of the column was broken off after removing the cap, and mounted on a 15 ml conical tube. The contents were then allowed to flow through. After all the liquid had drained from the column, the column was removed to a new tube, and the resin was washed twice with 0.7 ml of 1x NETS. The column was removed to a new tube, and the resin washed three additional times with 0.7 ml of 1x NETS. To elute, the column was removed to a new 15 ml conical tube, and 0.7ml elution buffer (10mM Tris-Cl pH 7.4) that was preheated to 70°C was added twice in succession. To verify the enrichment, a gel was run of 5-7 ul of the contents of each of the conical tubes. Generally, the flowthrough and first two washes had very bright ribosomal RNA bands. Washes 3-5 generally had very little RNA. The eluate usually had much weaker rRNA bands and a brighter smear signifying mRNA. Usually, about 30-50% of the eluate was rRNA, representing about a 30-60 fold enrichment of mRNA.
The eluate was then divided into two 1.5ml eppendorf tubes, and extracted twice with CHCl3 to remove cellulose. The RNA was then precipitated with 0.3M NaOAc, and one volume of isopropanol at -20°C for 1 hour, then spun 30 minutes in microfuge at 4°C. The mRNA pellets were then washed with 75% EtOH, and spun for 15 min at 4°C. The pellets were allowed to air dry for about 10-15 minutes, then resuspended in 10-20 µl DEPC-treated ddH2O, depending on pellet size. A spectrophotometer reading was taken using 1 ul in a microcuvette. The mRNA was then stored at -80°C in aliquots of 5-10ug to avoid thawing and re-freezing.
Reverse Transcription/cDNA labelling and Hybridization
To each 5ug aliquot of mRNA, 1.5µl of 2µg/µl anchored dT primer (TTTTTTTTTTTTXN, where T=thymidine, X=any base except T, N=any base), and 3.5µl DEPC-ddH2O were added, and the tube heated at 70°C for 10 minutes, then placed on ice for 8 minutes. To each tube, 9.6µl Trimix (0.6µl 25mM dA,C,G/10mM dT, 0.6µl 5x Gibco RT buffer, 3µl 0.1M DTT, 5.4µl H2O), 4.4µl DEPC-ddH2O, 3µl Cy-3 or Cy-5 dUTP (or dCTP), 1µl RNAse inhibitor, and 2µl Gibco BRL Superscript II reverse transcriptase were added. In the germ line experiments, fem-3 and reference RNA were labeled with Cy-5, and fem-1 and wild type and glp-4 were labeled with Cy-3. The tubes were incubated at 42°C for 90 minutes. After cDNA synthesis, the RNA was degraded by adding 1.5µl of 1M NaOH and putting the tube at 65°C for 10 minutes, then the reaction was neutralized with 1.5µl of 1M HCl and diluted with TE to 500µl. The tubes were spun in Micron-30 (Amicon Bioseparations) for 8 minutes at setting 10 in a microcentrifuge, and the spin-through was discarded. An additional 500µl TE was added to the retentate, and respun through same column. The retentate in the column was then inverted, and collected in a fresh tube by spinning for 2 minutes at setting 10. The appropriate combinations of Cy-3 labelled cDNA and Cy-5 labelled cDNA were mixed until a purple color was achieved and add 1.5µl of 5µg/µl tRNA was added. This mixed sample was diluted again to 500ul with TE and spun through a new Micron-30 as before. The retentate was concentrated until a volume of about 10-12ul was achieved, and then collected by inversion and centrifugation into a fresh column. To this probe, 4µl 20x SSC, 0.5µl 10% SDS, and TE to a total volume of 27µl was added. The probe was heated at 100°C for 2 minutes, spun in at RT in a microfuge at full speed for 1-2 minutes. The probe was placed carefully on a coverslip, and then the microarray was lowered, DNA spot side down, until the coverslip adhered to the slide. The microarray was then placed in a waterproof hybridization chamber, and hydrated with a drop of 3xSSC on one end of the slide, away from the coverslip. The chamber was sealed, and the slide immersed in a 65°C waterbath for 20h.
After hybridization, the hybridization chamber was removed from the 65°C waterbath, dried off, and opened. The microarray slides were placed in glass slide racks, and washed in glass slide chambers as follows. The slides were immersed in 200 ml of 3x SSC/0.2% SDS and gently agitated until coverslip slid off, before dunking up and down for about 1 minute. The slides were then placed in a new slide rack, and immersed in 200 ml of 0.2x SSC and dunked up and down for about 1 minute. The slides in the same rack were then immersed in 200 ml of 0.1x SSC, and dunked up and down for about 1 minute. The slides were dried by centrifuging the glass rack in a table top centrifuge designed to hold 96-well plates for 3-4 minutes at setting 600-700 rpm. The slides were stored dry in a dark box until scanning.
The slides were placed face down in a Genepix 4000A scanner by Axon Instruments, and preview scanned so that the PMTs (photomultiplier tube values) for each channel (Cy-3 or Cy-5) could be adjusted accordingly. For the germ line microarrays, PMTs ranged from 500-900 (on a scale of 1-1000). The slides were scanned at a 10 micron resolution.
Data was gathered from the scanned image using Genepix 2.0, software that accompanies the scanner. A grid containing a series of circles corresponding to the spots of the array was laid over the image, and the circles manually adjusted to each spot, a process known as "masking" or "gridding". The intensity of each pixel within the circle was calculated for each channel, termed "foreground" or "feature". Background for each spot was measured by increasing the circle size by two pixels and measuring pixel intensity along that outer circle. For each spot on the microarray, we took the mean pixel intensity of both foreground and background, and multiplied each number by the number of pixels in the spot. The background was then subtracted from the foreground. For each microarray hybridization, the Cy-3 and Cy-5 channels were normalized, a process necessary to avoid inaccurate measurements that would occur if one channel were consistently brighter than the other. For each channel, the mean spot intensity was determined by summing all the spot intensities, and dividing by the total number of spots. Then, the spot intensity of each gene was divided by the mean spot intensity, giving a ratio in units of mean gene expression (MGE). This ratio reflects the relative intensity of that gene's hybridization, compared to all other genes hybridized by the same probe, rather than the absolute value of the spot intensity, thus removing inherent differences in signal strength in each channel. Somewhere in here the natural log (ln) was taken, and the data were flattened. The ln value of a gene in one channel was then subtracted from the ln value of the same gene in the second channel to get the ln difference.
Some of the experiments were indirect; that is, each hybridization consisted of an experimental sample compared to a common reference mRNA. All experiments with wild type and glp-4 RNAs (L2, L3, L4, adult timepoints of each genotype, three-four repeats of each) were indirect. For these 30 experiments, the ln difference between sample and reference was calculated for every gene in each hybridization, and then all repeats of the same comparison were averaged together. For example, the ln differences for each gene in all four adult wild type vs reference experiments were averaged together. To compare wild type to glp-4 samples, the average glp-4/ref ln difference was subtracted from the average wt/ref ln difference.
To determine whether the ln difference value was statistically significant, we calculated a standard error unique to each gene, using a wide variety of microarray experiments, not just germ line arrays included in this study. Any microarray experiment that was repeated two or more times total was included. Because multiple repeats were performed for each timepoint and each genotype, we divided the gene-specific error by the square root of the number of repeats, then used this error in a Fishers' t-test to determine the likelihood that the two means (wt/ref and glp-4/ref) were different.
Making the 2/3 array
Gene-specific primers were designed to amplify a portion of each gene from genomic DNA. Each product was designed to include at least 70% coding region (where gene structure made this possible), avoid repeat regions of the genome, and be about 1kb in length. Primers were synthesized at either the PAN facility at Stanford University or by the Stanford Genome Technology Center. Standard PCR conditions were as follows: 1x Perkin Elmer standard PCR buffer, 1% formamide, 50ng genomic DNA, 2 mM MgCl2, 0.5 mM each left and right primers, 0.2 mM dNTPs. The thermal cycler was set at 94°C 2 min, then 30 cycles of 94°C 30sec, 54-56°C 45sec, 72°C 1 min, followed by an extension time of 72°C 5 min. 5ul of PCR product was removed for gel analysis, where each product was scored as a "no band" (nb), "faint" (f), "doublet" (d), or successful (+). Successful and faint products were included in subsequent analysis, while doublets and no bands were excluded. The PCR products were transferred to Corning plates prior to precipitation in some cases, otherwise they were left in the PCR plates. 5µl of 3M NaOAc and 150µl of 100% ethanol were added to the PCR reactions (in some cases isopropanol was added), held at -20°C at least one hour, and then spun in a Sorvall RC-3B centrifuge at 4000rpm for 30 minutes at 4°C. 150ul of 70% ethanol was then added, and the plates were chilled and spun as before. The ethanol was aspirated off, and the plates allowed to dry, covered by a Kimwipe, in a hood. The pellets were resuspended in 24µl of 3x SSC and aliquotted into 4-384 plates, containing 5µl each.
Microarray slides were printed from one set of these 384 well plates by Rosetta, Inc.