Important!!! It is possible to compare directly the results of hybridizations on the membranes from the SAME LOT. Be careful, the same hybridization pattern does not guaranty the identity of lots.
Handling of membranes
never touch membranes by the naked fingers (only forceps or gloves).
try to take membranes only by edges (out of the spotting area). It is rather easy to damage surface of membranes by the forceps.
after the first hybridization membranes should be always wet.
it is good idea to have a kind of "history" for each membrane (the list of the used probes and overnight autographs of the stripped membranes after each probe).
Before the first hybridization
wash filters two-three times for 10 min in (1xSSC, 0.1%SDS) at normal temperature;
wash three times for 20 min in (0.1xSSC, 0.1%SDS) at about 80oC (we do it on rocking platform in air thermostat; we have only 65oC thermostat, so we pour SSC/SDS solution at about 95oC and leave it to shake for 20 min);
air dry membranes on the 3MM filter paper; store at normal temperature between two 3MM sheets;
when we hybridize membranes first time we did not change pre-hybridization solution (dry membrane can't change the composition of buffer);
it is extremely important to use enough volume of hybridization buffer and to check, that there is no air bubbles under membrane. Otherwise pale zones will appear on membrane.
the dynamic range of signals from the complex hybridization is larger, then the dynamic range of the phosphorimager, so we perform at least two expositions. One is short (about 0.5-2 hours), another - long (at least two days). Then, data from the quantification of both pictures combine in one table (weak signals - from the long exposition; strong - recalculation from short exposition);
50µm resolution on the image is absolutely enough for membranes with 5x5 pattern (0.25mm pins). The only effect of higher resolutions is slowing of processing by the image analyzing programs.
When compare two mRNA samples you should normalize pictures to compensate difference related with labeling procedure (or different exposition time for the sequential experiments). It is not a good idea to use particular "housekeeping gene" as a general rule for normalization. All housekeeping genes change their expression levels in some particular situations. It was suggested to use a number of housekeeping genes, but just now we prefer to compare the pictures by the "general view".
The procedure extremely good for the projects related with the search of differentially expressed genes. The task itself guaranties, that the majority of genes do not change expression levels significantly. So, if to ask "Excel" program to plot signals as a graph (one axe - one sample; another axe - another sample), the optimal linear relation between data will be find readily.