Despite the advent of microarrays, Northern analysis remains a popular technique for analyzing gene expression. It is one of the best methods available for evaluating an RNA sample for both quality and quantity. Northerns can also reveal message size and the presence of alternatively spliced transcripts. However, a significant drawback of the technique is that it is relatively insensitive, especially when trying to detect rare targets with standard hybridization buffers, as typically only 1-5% of target molecules on a blot hybridize to probe (Vernier et al. (1996) Anal. Biochem. 235: 11–19). Here we reproduce this standard hybridization buffer data and show that by selecting a more sensitive hybridization buffer, the number of target molecules that hybridize to probe on a blot can be brought closer to 100%.

The inherent limitation in sensitivity associated with blot hybridization has long been recognized. The extent of this limitation is demonstrated by determining the percentage of target nucleic acid bound by probe using standard Northern protocols. The data in Figure 1 is the result of an experiment comparing the percent hybridization in standard hybridization buffer vs. ULTRAhyb® Ultrasensitive Hybridization Solution. An RNA target was synthesized with a trace amount of radiolabel to facilitate gel purification and subsequent quantitation. High specific activity DNA probe having complementarity to the RNA target were synthesized and quantitated. Known quantities of the target and probe were spotted onto separate membranes as indicated in the figure. The membranes with RNA target were incubated overnight with DNA probe in ULTRAhyb buffer or in a standard formamide hybridization buffer. Following hybridization, the blots were washed and exposed to film alongside the blot spotted with dilutions of known amounts of probe.


Figure 1. Hybridization Efficiency of ULTRAhyb® Buffer and Standard Hybridization Buffer. High specific activity random-primed DNA probes were synthesized using Ambion’s Strip-EZ™ DNA Kit. The indicated number of these probe molecules were spotted on the BrightStar® Plus Membrane on the left as a reference. The indicated quantities of sense strand RNA target molecules were spotted onto the BrightStar® Plus Membranes on the right and hybridized overnight to an excess of DNA probe in either standard hybridization buffer or in ULTRAhyb buffer. Because known amounts of probe and target were spotted, percent hybridization can be determined by comparing the signal from the blots of probe alone to those of the target blots that had been incubated with probe. Note that the hybridization signals generated with ULTRAhyb buffer are difficult to distinguish from the probe molecules spotted on the left.



The blot on the left reveals how much signal results from 1–30x105 randomly-primed DNA probe molecules alone. The blot on the right displays signal resulting from the hybridization of probe to a known amount of target. Comparing the two blots shows what percent of target molecules are bound to probe. In this example, the signal for 3x106 copies of target is about one-third as intense as 105 copes of probe suggesting that only 1% of the available target is bound by probe in standard hybridization buffer. In ULTRAhyb buffer the signal from 105 copies of target is almost as intense as 105 copies of probe, indicating that most of the target is bound by probe. In fact analyses indicate that as few as 10,000 molecules can be detected on a blot using either RNA or DNA probes (data not shown).