Fig. 1.

Isolation of genes that are misexpressed in sciatic nerve in the absence of pou3f1. A, Photograph of an ethidium bromide-stained acrylamide gel containing driver, DP1, and DP2 cDNA from P0 pou3f1 +/+ and −/− sciatic nerves. The sizes of the cDNAs can be estimated from the φX174HaeIII digest (lane M). RDA was performed in parallel using cDNAs that were digested with eitherDpnII or NlaIII. Faint bands, presumably attributable to ribosomal RNA or other highly repetitive RNAs, can be seen in the lanes containing the driver cDNA. Note the absence of any band derived from carrier RNA (584 bp). After one round of RDA (DP1 lanes), specific bands, many of similar sizes, begin to appear in both the +/+ and −/−lanes, with either DpnII- orNlaIII-digested cDNA. After two rounds of RDA, the +/+ and −/− lanes contain distinct banding patterns (DP2 lanes). DP2 DNAs were cloned into pBKSII⁺ for analysis. In DP3 cDNA, only a few bands that correspond to Neo and LacZ increase in intensity relative to DP2 cDNA (data not shown). B, Southern blots of PCR-amplified cDNA (Snorthern blots) from pou3f1 +/+ and −/− sciatic nerves after hybridization to different radiolabeled cDNA fragments generated by RDA. The sizes of bands are shown to theright of each panel and include the 24 bp linkers on the ends of the fragments. All of the misexpressed genes that were confirmed by this analysis are shown, except for Neo. The lacZ probe was made from JBSN11, a plasmid with a double insert. In addition to lacZ, it contains an unknown sequence, which is not differentially expressed, and thus serves as an internal control. JBSN18, JBSN21, JBSN61B, JBSN64, JBSN57, JBSN70, and JBSN125 are activated by pou3f1, because they are expressed at lower levels in cDNA derived from pou3f1 −/− nerves. The pou3f1 probe was derived from a separate RDA experiment and is used here as a control to confirm its absence in −/− cDNA. The names of the genes containing these fragments are given below each pair oflanes.