Isolation and some properties of DNA coding for tRNA1met from Xenopus laevis

DNA containing the reiterated genes for tRNA1met has been partially purified from Xenopus laevis by centrifugation in actinomycin C1-CsCl and Ag+-Cs2SO4 gradients. These gradients separate the tRNA1met genes from those coding for tRNA2met and tRNAval, thus confirming our earlier suggestion that these genes are not intermingled with each other (Clarkson, Birnstiel, and Purdom, 1973a). The gradients also demonstrate the existence of a minor 5S DNA fraction which appears to differ from that previously isolated by Brown, Wensink, and Jordon (1971). When the enriched tDNA1met is digested to completion with either of the restriction endoncucleases EcoRl or Hpa l, the tRNA1met genes are predominantly found within DNA fragments that are about 3100 base pairs long. A partial digestion with EcoRl shows that these fragments arise from the regular spacing of the enzyme restriction sites. The 3100 base pair EcoRl fragments are cleaved by Hpa l into fragments to two size classes, one of which is about 2200 base pairs long and contains the tRNA1met genes. The shorter fragments are about 700 base pairs long, and they appear to contain genes coding for at least one other kind of tRNA species. X. laevis tDNA1met thus comprises tandemly repeated DNA whose component parts show little if any length heterogeneity.