4. RESULTS AND DISCUSSION

The autoradiograph in Figure 1 shows an example of differential display results obtained using Display Systems Differential Display kit (PGC Scientific, Minneapolis, MN). In this experiment, total RNA was obtained from K562 human erythroleukemia cells (ATCC, Rockville, MD) grown in the absence (condition 1) or presence (condition 2) of chronic nitric oxide exposure (5). Duplicate cDNA synthesis reactions were performed using downstream primer 9 (T11GG) as the anchoring primer and PCR performed as described. Amplified fragments from the two cDNA fractions were separated on a 6% denaturing acrylamide gel, transferred and dried onto Whatmann 3M paper and exposed to film for 10 days. Two differentially expressed sequences were identified, D9U10 and D9U12. Both appear to be expressed under condition 1, but not condition 2 as seen highlighted in Figure 1. The two highlighted differentially expressed bands were cut from the gel and eluted into 100 microliters of water by boiling for 15 minutes. The eluted DNA samples were then used as templates for PCR amplification. Initial attempts to reamplify from these eluted samples, using 12.5 microliters as template in a new 50 microliters PCR reaction consistently failed. For the trial shown in Figure 2, 2.5 to 10 microliters of the eluted product was used in a 50 microliters PCR reaction. As seen in Figure 2, reamplification was accomplished for both products only when 2.5 microliters of template was used, while larger amounts of eluted sample consistently failed to re-amplify. Limiting the volume of eluted DNA sample in subsequent PCR reactions was the only modification that reliably and consistently led to reamplification in 52 separate instances (unpublished results). The biochemical basis for our observation is unknown, but the presence of inhibitors of Taq polymerase in the eluted DNA preparation, such as urea (1) cannot be excluded. This method of reamplification avoids the ethanol precipitation step of the eluted DNA fragment used by some investigators (2, 4).

Differential expression of the two sequences reported here was confirmed by Northern analysis. An mRNA Northern blot probed with the radiolabelled D9U10 sequence is shown in Figure 3. Only cells grown in the presence of nitric oxide trans-gene expression express a 4 kilobase mRNA identified using this probe. Beta actin mRNA signal is shown as a control.

Differential display is a challenging technique to master. The reamplification of DNA sequences eluted from denaturing acrylamide gels must be consistent and reliable. When attempting to reamplify these valuable sequences, it should be kept in mind that including more of the eluted DNA sample in the reamplification PCR mixture is not necessarily better.

Figure 1. Autoradiograph of a differential display gel. Radiolabeled differential display PCR products from downstream primer T11GG and 4 random upstream primers, U10-13, using cDNA derived from cells grown in the absence (condition 1) or presence (condition 2) of chronic nitric oxide exposure. Boxes highlight two PCR products found to be differentially expressed.

Figure 2. Reamplification of the differentially expressed DNA fragments. DNA fragments identified in Figure 1 were cut from the gel, and eluted in 100 microliters water by boiling for 15 minutes. PCR reactions were performed as described using 2.5 to 10 microliters volumes of these eluents; the products were separated on a 2% agarose gel and stained with ethidium bromide. D9 indicates the anchoring primer, and U10 and U12 are the two upstream primers, respectively.

Figure 3. Northern blot prepared with mRNA isolated from K562 cells in the presence and absence of nitric oxide expression probed with radiolabelled D9U10. A 4 kilobase mRNA is identified in the K562 lane, but absent in the KNOS-F lane. Beta actin signals from the same Northern blot are shown as a control for loading.