Anderson, 1999), or you could switch to using RNA probes in the hybridization, which are 3- to 5-fold more sensitive than DNA probes in typical hybridization buffers (Ambion Technical Bulletin 168, and references therein). Dramatic differences in the sensitivity of northern blots can also be seen from using different hybridization buffers. If you remain dissatisfied with the Northern data, and you are not interested in determining the size of the target, switching to a more sensitive technique such as nuclease protection or RT-PCR might help. Nuclease protection assays, which are 5-to 10-fold more sensitive than traditional membrane hybridizations, can accommodate 80 to 100 ug of nucleic acid in a single experiment RT-PCR can detect extremely rare messages, for example, 400 copies of a message in a 1 ug sample as described by Sun et al ( 1998). RT-PCR is currently the most sensitive of the RNa analy- sis techniques, enabling detection and quantitation of the rarest of targets. Quantitative approaches have become increasingly reliable with introduction of internal standards such as in com- petitive PCR strategies(Totzke et aL., 1996; Riedy et al., 1995) Dot slot blots In this procedure, RNA samples are directly applied to a mem- brane, either manually or under vacuum through a filtration manifold. Hybridization of probe to serial dilutions of sample can quickly generate quantitative data about the expression level of a target. Total RNA or poly (A)RNA can be used in this assay. Since the RNa is not size-fractionated on an agarose gel, a potential drawback to using total RNA in dot/slot blots is that signal of interest cannot be distinguished from cross-hybridization t rRNA. Switching to poly(A)RNA as the target source might alle- viate this problem. However. it is crucial that relevant positive and negative controls are run with every dot/slot blot, whether the source of target nucleic acid is total RNA or poly(A)RNA Hybridization to Gene Arrays and Reverse Dot blots Gene arrays consist of cDNA clones(sometimes in the form of PCR products, sometimes as oligonucleotides)or the correspond- ing oligos spotted at high density on a nylon membrane, glass slide, or other solid support. By hybridizing labeled cDNA probes reverse transcribed from mRNA, the expression of potentially hundreds of genes can be simultaneously analyzed. This procedure requires that the labeled cDNa be present in excess of the target spotted on the array. This is difficult to achieve unless poly(a) RNA is used as template in the labeling reaction 200 Martin et al(Anderson, 1999), or you could switch to using RNA probes in the hybridization, which are 3- to 5-fold more sensitive than DNA probes in typical hybridization buffers (Ambion Technical Bulletin 168, and references therein). Dramatic differences in the sensitivity of Northern blots can also be seen from using different hybridization buffers. If you remain dissatisfied with the Northern data, and you are not interested in determining the size of the target, switching to a more sensitive technique such as nuclease protection or RT-PCR might help. Nuclease protection assays, which are 5- to 10-fold more sensitive than traditional membrane hybridizations, can accommodate 80 to 100mg of nucleic acid in a single experiment. RT-PCR can detect extremely rare messages, for example, 400 copies of a message in a 1mg sample as described by Sun et al. (1998). RT-PCR is currently the most sensitive of the RNA analy￾sis techniques, enabling detection and quantitation of the rarest of targets. Quantitative approaches have become increasingly reliable with introduction of internal standards such as in com￾petitive PCR strategies (Totzke et al., 1996; Riedy et al., 1995). Dot/Slot Blots In this procedure, RNA samples are directly applied to a mem￾brane, either manually or under vacuum through a filtration manifold. Hybridization of probe to serial dilutions of sample can quickly generate quantitative data about the expression level of a target.Total RNA or poly(A) RNA can be used in this assay. Since the RNA is not size-fractionated on an agarose gel, a potential drawback to using total RNA in dot/slot blots is that signal of interest cannot be distinguished from cross-hybridization to rRNA. Switching to poly(A) RNA as the target source might alle￾viate this problem. However, it is crucial that relevant positive and negative controls are run with every dot/slot blot, whether the source of target nucleic acid is total RNA or poly(A) RNA. Hybridization to Gene Arrays and Reverse Dot Blots Gene arrays consist of cDNA clones (sometimes in the form of PCR products, sometimes as oligonucleotides) or the correspond￾ing oligos spotted at high density on a nylon membrane, glass slide, or other solid support. By hybridizing labeled cDNA probes reverse transcribed from mRNA, the expression of potentially hundreds of genes can be simultaneously analyzed.This procedure requires that the labeled cDNA be present in excess of the target spotted on the array. This is difficult to achieve unless poly(A) RNA is used as template in the labeling reaction. 200 Martin et al