While there are various techniques out there, we have found that the most informative/rigorous is northern blotting using probes hybridising to ETS/ITS sequences within the 45S pre-rRNA.
While it is slightly more work than other techniques that are available (such as qPCR or run-ons), we find it gives us the greatest range of information and most reliable data. While Northerns against mature rRNAs can be another option, we tend to find that we simply get so much signal from the 28S/18S that we cannot resolve the full range of less stable/abundant processing intermediates - hence using ETS/ITS probes.
If you're interested in more detailed protocols, you can check out our most recent paper on bioRxiv (http://www.biorxiv.org/content/early/2016/03/16/043935). Alternatively, if you are working with humans you may find that your gene of interest is included in the recent high-throughput study from Denis LaFontaine's lab - see their database at http://www.ribogenesis.com/.
To measure ribosome biogenesis (which is controlled by Pol I) in human cancer cell line, you can use qRT-PCR to get a first-hand data. Note that you have to use specific primer sets to target the pre-rRNA transcripts. Normally people use the ITS (Internal transcribed spacer). Here is the primer info (h means human):
hB2M
ATGGAGGTTTGAAGATGCC
CTAAGTTGCCAGCCCTCCT
h47S rRNA ITS1
GACCCCTTGGGGGGATCG
CGCGGACACCACCCCACA
h47S rRNA ITS2
CCCGCCCCGCGGCCCGC
CGACGCGGAAGCTCGGGA
Note that B2M is the internal control gene whose expression is not much dependent on Pol I activity. Also, you can use Pol I specific inhibitor CX-5461 to test your hypothesis. Finally, you can read my recent publication to get more info (2016, Nat Commun, Stem cell and neurogenic gene-expression profiles link prostate basal cells to aggressive prostate cancer).
While qPCR such as the above does technically work, it gives you only very limited information about ribosome biogenesis - so much so that you can often miss important alterations of/defects in biogenesis, or misinterpret ones that are present.
The chief reason for this is that, using qPCR, you cannot tell which of the numerous different rRNA precursors your qPCR amplicons have come from. For just one example, the complete or near-complete ITS1 sequence is present in 47S, 45S, 43S, 41S, 30S, 26S and 21S precursors, and an ITS1-targeted qPCR amplicon will only ever be able to distinguish some, but not all, of these. Furthermore, several of these components (e.g. 41S vs. 30S/26S) reside in different, complementary and partially redundant maturation pathways; if any treatment/mutant you make alters the balance of these pathways (as several mutants are known to do), you are very unlikely to resolve this alteration by qPCR.
While qPCR is technically easier/faster, the key advantage of Northerns over qPCR is that you are able to size-discriminate your rRNA precursors; for example, if you were to use the one ITS1 qPCR amplicon above as a Northern probe, you would be able to observe all 7 different precursors above on a gel and measure any alterations in levels densitometrically. This is the reason why we avoid using qPCR for the analysis of any pre-rRNA species, and use Northerns exclusively.
Hope that helps clear up the relative merits of the different methods!
Cheers,
Dan
P.S. Note that while RNA Pol I transcribes three of the four mature rRNAs (28S, 18S, 5.8S), the fourth (5S) is transcribed instead by RNA Pol III.
P.P.S. As a final addendum, using B2M - an mRNA - as a control in conditions of possible ribosome biogenesis defects could be dangerous, as such defects can frequently feed back to regulate mRNA levels. A potentially better control may be a separate abundant, non-coding RNA such as 7SK.
Dingxiao, thank you for your great suggestion to use ITS1 and ITS2 as a proxy to measure ribosome biogenesis. We bought the three pair of primers (ITS1, ITS2 and B2M) you wrote down here and in your paper, and we would like to ask you some questions. What is the primer concentrations you used in your experiments? We have tested the primers efficiency with a standard curve and we get double peaks in the melting curves (potential indication of primer dimers). We used primers at 500 nM concentration. In addition we have observed that ITS2 primer seems to be somehow more specific than ITS1 (this was the result of several independent experiments); have you observed results similar to this?