1. In screw cap microfuge tubes mix:
6 M glyoxal 5.4 ul
DMSO 16 ul
0.1 M Na phosphate pH 7.0 3 ul
RNA (up to 10ug) 5.4 ul
incubate at 50deg. for 1 hour.
Set up a tube or two of RNA markers as well; these can be purchased from Promega (catalog # G3151). Use 2 ug of makers per lane. These markers have sizes of 9488, 6225, 3911, 2800, 1898, 872, 562, and 363 bases.
2. Pour the gel: use 1% agarose in 10 mM Na phosphate pH 7.0. After heating in the microwave, cool to 70deg., add solid sodium iodoacetate to 10 mM (to inactivate RNAses), cool to 50deg., and pour the gel.
3. Set up the gel: submerge the gel in running buffer (10 mM Na phosphate pH 7.0). This low strength buffer requires recirculation during the run, so put a stir bar in each buffer tank, set up a peristaltic pump to recirculate buffer between the two tanks. Put a stir plate under each buffer tank.
4. Chill the RNA tubes on ice, spin them down briefly in the microfuge. Add 4 ul loading buffer to each tube, and immediately load the samples. You may want to add some loading buffer containing xylene cyanol and bromophenol blue to an extra lane on the gel to monitor progress of the electrophoresis. An extra blank lane should be left between the marker lane and your sample lanes. That way, it is easy to cut the marker lane off the blot and stain it. If a marker lane is left on the blot during probing, be aware that some of the markers (9.4, 6.2, 0.87 kb) hybridize with pUC vector sequences that will likely contaminate your probe; this is all the more reason to leave a space between the markers and your samples.
5. Run the gel at 3-4 V/cm. Wait about 10 minutes after turning on the power before starting the stir bars and peristaltic pump; this allows the RNA to run into the gel first so that the buffer circulation won't disturb it in the wells. It takes 2.5 hours to run the gel about 11 cm; the gel and buffer will become warm during the run.
6. After the gel is run, no further treatment of the gel is necessary before blotting. It is immediately blotted to Nytran in 20X SSPE, exactly as you would for a Southern blot (see Michael Koelle's Southern blot protocol). After blotting overnight carefully mark the wells on the blot, UV crosslink just as you would a Southern, and dry the filter in a vacuum oven (takes about 20 minutes).
7. Visualizing the RNA markers and/or samples: We found that post-staining glyoxal or formaldehyde gels with ethidium bromide doesn't work well. It's not possible to run glyoxal gels in ethidium, since glyoxal reacts with the ethidium. If you just want a quick check of what an RNA sample looks like, it is possible to mix ethidium bromide with RNA sample and then load the mixture on a formaldehyde gel (see Sambrook et al. for this method). This gives a good stain, but it's not nearly as nice as what you get if you blot the gel and methylene blue stain, as described below.
When making Northern blots, however, it is preferable to visualize the RNA markers directly on the blot, since this allows for the most accurate alignment with the autorad. Cut the marker lanes (and any extra sample lanes you want to stain) off the blot. Soak in 5% acetic acid for 15 minutes, and then transfer to 0.5 M sodium phosphate (pH 5.2), 0.04% methylene blue for 5-10 minutes at room temperature. Rinse the blot in water for 5-10 minutes, until the background is white. The markers should appear as sharp blue bands on a white background. Total RNA from C. elegans should have two heavy ribosomal RNA bands at about 3.5 kb and 1.7 kb, and a lighter smear of mRNA centered around 2 kb. Contaminating E. coli can be detected by the presence of 3.0 and 1.5 kb E. coli ribosomal RNAs.
8. Before probing the blot, soak it in 20 mM Tris pH 8.0 at 65deg. for 30 minutes. This removes the glyoxal.
9. Probe the filter exactly as you would a Southern blot. Northerns can be reused many times, so make sure to always keep the blot moist, and freeze it in a seal-a-meal bag between uses. It's best to just let the signal from previous probings decay over time before reusing the blot; repeatedly stripping the filter may decrease the signal on subsequent probings.
