Artifactual bands associated with alkaline transfer.

Alkaline transfer to nylon membranes is a widely used modification of the Southern blot procedure (1). However, several reports describe decreased hybridization efficiency with alkaline transfer (2,3,4). Shorter exposure time to alkali, or lower alkali concentration have been suggested to increase hybridization efficiency. We report an example of altered DNA banding following alkali transfer compared to standard Southern transfer. Human genomic DNA (40 /tg) was cleaved to completion with BamHI (Promega); 10 ng were loaded into four adjacent wells of a 0.6% agarose gel and elecrrophoresed 18 h at 1 V/cm. The lanes were separated, and each section subjected to a different method of transfer. Lanes 1 and 2 were alkali transferred for 18 or 4 h, respectively, onto Zeta-probe nylon membrane (Biorad) according to the manufacturer's directions. Lane 3 (Zetaprobe) and lane 4 (nitrocellulose) were transferred for 18 h using 20XSSPE (5). All membranes were rinsed in 2xSSPE and vacuum dried. Membranes were prehybridized, hybridized, and washed in parallel according to the Bio-Rad protocol. Specifically, hybridization was carried out in 0.5 M NaH2PO4, pH 7.2, 7% SDS, 1 mM EDTA at 65°C for 18 h. The probe was human glutathione peroxidase cDNA (6) labelled with P by random priming (Dupont-NEN) to a specific activity of 5 x lO^pm/jig. The membranes were washed at 65°C, first in 40 mM NaH2PO4, pH 7.2, 1 mM EDTA, 5% SDS, then followed by 40 mM NaH2PO4, pH 7.2, 1 mM EDTA, 1% SDS. Three non-variable BamHI restriction fragments (12.1, 9.2, and 0.7 kbp) were observed in each of the lanes 1 through 4. These findings agree with previous results (6, 7). However, three additional bands (7.9, 5.4, 1.4 kbp) were observed in those lanes (1 and 2) subjected to alkaline transfer. These artifactual bands were selectively eliminated by washing at higher temperature, as shown by the identical alkaline blots washed at 65 °C (lane 5) and 70°C Oane 6). This result was reproducible with other preparations of DNA, reagents, restriction enzyme, and probe. Howeer, artifactual bands were not observed when the identical blots were hybridized with a different probe (human glutathione reductase cDNA; not shown). We conclude that, with certain probes, artifactual bands may appear under conditions currently recommended for alkaline transfer, hybridization, and washing. We speculate that these bands are due to partially complementary sequences which are unavailable for hybridization in intact DNA (e.g., because of secondary structure) but are exposed in short DNA fragments likely to result from prolonged alkaline treatment.