Processing of the Escherichia coli leuX tRNA transcript, encoding tRNALeu5, requires either the 3′→5′ exoribonuclease polynucleotide phosphorylase or RNase P to remove the Rho-independent transcription terminator

Bijoy K. Mohanty 0 Sidney R. Kushner 0 0 Department of Genetics, University of Georgia , Athens, GA 30605, USA Here we report a unique processing pathway in Escherichia coli for tRNALeu5 in which the exoribonuclease polynucleotide phosphorylase (PNPase) removes the Rho-independent transcription terminator from the leuX transcript without requiring the RhlB RNA helicase. Our data demonstrate for the first time that PNPase can efficiently degrade an RNA substrate containing secondary structures in vivo. Furthermore, RNase P, an endoribonuclease that normally generates the mature 50-ends of tRNAs, removes the leuX terminator inefficiently independent of PNPase activity. RNase P cleaves 4-7 nt downstream of the CCA determinant generating a substrate for RNase II, which removes an additional 3-4 nt. Subsequently, RNase T completes the 30 maturation process by removing the remaining 1-3 nt downstream of the CCA determinant. RNase E, G and Z are not involved in terminator removal. These results provide further evidence that the E. coli tRNA processing machinery is far more diverse than previously envisioned. - The 86 transfer RNA (tRNA) genes in Escherichia coli are transcribed either as part of complex operons containing multiple tRNAs, ribosomal RNAs (rRNAs) or messenger RNAs (mRNAs) or as monocistronic transcripts. Irrespective of their physical organization, each precursor tRNA contains extra sequences at both the 50- and 30-ends that are removed to generate a mature, functional tRNA. Either RNase E or RNase P are required for the separation of tRNAs that are transcribed as part of polycistronic transcripts to generate pre-tRNAs that become substrates for RNase P, which cleaves them endonucleolytically to generate the mature 50 termini (1). While 50-end processing of all tRNAs has been shown to be catalyzed by only endonucleases, 30-end maturation is believed to be a multistep process involving both endoand exoribonucleases (2). For example, for tRNA transcripts terminating with a Rho-independent transcription terminator, an initial endonucleolytic cleavage can remove the stem-loop followed by exonucleolytic trimming to generate the mature 30 terminus (35). In contrast, 30-end processing can be all exonucleolytic for tRNA transcripts that are terminated in a Rho-dependent fashion (6). Of the eight known 30!50 exoribonucleases in E. coli, RNase T and RNase PH are the most important for final 30-end maturation (7). Interestingly, polynucleotide phosphorylase (PNPase), a major 30!50 exoribonuclease encoded by pnp gene, has only been shown to be involved in shortening of long 30 trailer sequences (>15 nt) generated after intergenic endonucleolytic cleavages of polycistronic tRNA operons (3,8) or Rho-dependent transcription termination (6). Thus, although PNPase plays a significant role in the processing and degradation of a variety of RNA species (7), its role in tRNA processing and maturation has been thought to be minimal. Of the eight leucine tRNA genes encoded in E. coli, seven are part of five different polycistronic operons that depend on either RNase E or RNase P or both for initial processing and maturation (36). Here we describe a unique processing pathway for the eighth leucine tRNA, leuX that encodes tRNALeu5. Specifically, PNPase initiates processing by removing the the Rho-independent transcription terminator and stops 13 nt downstream of the CCA determinant. This is the first in vivo demonstration of processing of an RNA substrate containing a stem-loop structure by PNPase, even though the enzyme has been shown to be inhibited by such structures in vitro (9). Endonucleases such as RNase E, RNase G, RNase LS or RNase Z are not involved. Our data further show that RNase P, the endonuclease responsible for generating mature 50 termini, also removes the terminator from 10% of primary transcripts by cleaving 47 nt downstream of the CCA determinant, generating substrates for RNase II, which removes an additional 34 nt. Subsequently, RNase T completes the 30 maturation process by removing the last 13 nt downstream of the CCA determinant left by either PNPase or RNase II. MATERIALS AND METHODS The E. coli strains used in this study were all derived from MG1693 (thyA715) (E. coli Genetic Stock Center, Yale University). The rne-1 and rnpA49 alleles encode temperature sensitive RNase E and RNase P proteins, respectively, that do not support cell viability at 44 C (1012). SK5665 (rne-1) (11), SK2525 (rnpA49 rbsD296::Tn10) (4), SK2534 (rne-1 rnpA49 rbsD296::Tn10) (4), SK10019 (pnpD683) (13), CMA201 (Drnb) (14), SK5726 (pnp-7 rnb-500) (15), SK10443 (rnpA49 rbsD296::Tn10 pnpD683) (6), SK7988 (DpcnB) and SK10148 (Drnt) (4) have been previously described. A P1 lysate grown on SVK53 (MC1061/DrhlB) was used to transduce both MG1693 and SK10019 (pnpD683) to construct SK10553 (DrhlB) and SK10554 (pnpD683 DrhlB), respectively. SVK53 is similar to SVK1 (MC1061/DrhlB) (16) but has a nearby mini Tn10 that was used for tetracycline selection. A P1 lysate grown on JW2798 (Keio Collection, Japan) was used to transduce MG1693 and SK2525 (rnpA49) to construct SK4390 (DrppH::kan) and SK4395 (DrppH:kan rnpA49), respectively. SK2059 (rnr::kan) was constructed by P1 mediated transduction using CA265R (rnr::kan) as the donor strain. Growth of bacterial strains, isolation of total RNA and northern analysis Bacterial strains were grown (6) and total RNA was extracted as described previously (17). The RNAs were quantified by measuring the OD260 using a Nanodrop (ND1000) apparatus. Total RNA was separated in 6% polyacrylamide gels containing 8 M urea in TrisBorate EDTA (TBE) buffer as described previously (6). Primer extension Primer extension analysis of the leuX transcripts was carried out as described previously (6) with the following modifications. The leuX nucleotide sequence was obtained from a PCR DNA product (containing leuX genomic sequences) using the Promega fmol sequencing kit and the primer LEUX (primer b, Figure 1A) that was also used for the reverse transcription. The sequences were analyzed on a 6% PAGE containing 8M urea. RTPCR cloning and sequencing of 5030 ligated transcripts The 50- and 30-ends of leuX transcripts were identified by cloning and sequencing the RTPCR products obtained from 50!30 end-ligated circular RNAs following the methods described previously (17). The 5030 junctions of the cDNAs were amplified with a pair of gene-specific primers using GoTaq Green Master Mix (Promega). In vitro transcription and RNase E digestion of leuX transcripts The full-length leuX runoff transcript was obtained by in vitro transcription using a PCR amplified DNA template that contained a modified T7 RNA polymerase promoter. The leuX DNA template was amplified using a 50-primer (T7-LEUX, 50GGATCCTAATACGACT CACTATAGTTTTCCGCATACCTCTTCA30) and a 30-primer (LEUX-442, 50AACACTGGATTTCAGGCA TAA30) that was designed to generate the leuX transcript with identical in vivo 50 (+1) end and 30 (...truncated)