The pFOXC2 and pFOXC3 ORF is predicted to encode a 62 kDa protein, and Western analysis using an antibody against a synthetic peptide derived from the pFOXC3-RT polypeptide sequence confirmed the presence of a ~ 60 kDa polypeptide in mt RNP particles (unpublished data). The pFOXC3-RT is concentrated in mt RNP particles and has reverse transcrip-tase activity. When mt RNPs from pFOXC-containing strains are incubated in appropriate buffers having [32P]-dNTPs, high molecular weight labeled DNAs are produced. These reactions are insensitive to the DNA-dependent DNA polymerase inhibitor actinomycin D, and sensitive to pretreatment with RNase A (Walther and Kennell 1999). The products are cDNAs of approximately 1.9 kb in length and, when used as probes in Southern hybridizations, the products hybridize with plasmid (+) strand sequences indicating that the plasmid transcript serves as template.
The length of the (-) strand cDNA products matches the size of the plasmid transcript, suggesting the cDNA synthesis begins at the 3' end of RNA templates. Posttreatment of (-) strand cDNA products with RNase, protease, or alkali did not affect the migration of the products in denaturing gels, indicating that the cDNAs were not bound by large RNA or protein primers (Simpson et al. 2004). The nature of the primer used in vivo has not been determined and is currently under investigation.
To gain a better understanding of the mechanistic properties of the pFOXC-RT, an in vitro reverse transcription system was developed that uses exogenously added RNA templates (i.e., exogenous reactions; Simpson et al. 2004). When liberated from endogenous RNAs via micrococcal nuclease digestion, the pFOXC-RT is able to copy in vitro synthesized RNAs added to reactions. To initially assess the template specificity of the pFOXC-RT, total mitochondrial RNA was added to reactions having MN-treated pFOXC-RT. The resulting cDNA products were found to hybridize to plasmid sequences rather than mtDNA fragments, indicating that the RT shows specificity for the plasmid transcript. Interestingly, unlike the circular retroplasmid RNAs, the 3' ends of pFOXC RNAs are not predicted to be extensively base-paired and a recognition site has not been identified.
In exogenous reverse transcription assays having in vitro synthesized RNAs corresponding to the 3' end of the plasmid transcript, the pFOXC-RT is capable of initiating cDNA synthesis using a variety of primers. In reactions containing the MN-treated pFOXC-RT and in vitro RNA templates, cDNA synthesis initiates via the extension of the 3' hydroxyl of snapped-back RNAs. When DNA oligonucleotides complementary to regions of the RNA templates are included in the reactions, the pFOXC-RT can readily use the oligonucleotides as primers. In both instances, the pFOXC-RT shows little specificity for RNA templates. While the ability to use DNA oligonucleotides and snapped-back RNAs as primers is not uncommon for retroviral RTs such as MMLV or AMV, comparative studies showed that the pFOXC-RT can use primers in an unconventional manner. The pFOXC-RT was found to require substantially less base-pairing between the DNA primers and RNA templates, and was able to extend snapped-back RNAs that had very minimal base-pairing interactions. DNA primers with as little as 5 bp of complementarity to sequences at the end of the RNA templates could be extended, and the pFOXC-RT showed a preference for primers bound at the 3' terminus (Simpson et al. 2004).
In other reactions, the pFOXC-RT was also found to be able to copy DNA templates. It was observed that in reactions containing DNA oligonucleotides, small, highly abundant labeled DNAs were produced. These products derived from the extension of DNA primers that had self-annealed (Simpson et al. 2004). Experiments using a series of single oligonucleotides that had varying degrees of self-complementarity confirmed that the products derived from the copying of DNA templates and demonstrated that the pFOXC-RT could extend DNA primers that contained 3' mismatches. Remarkably, primers having up to three terminal mismatches could be extended (Simpson et al. 2004). Taken together, these studies indicate that the pFOXC-RT has a remarkably loose specificity for primers and that it is capable of extending primers that have minimal base-pairing interactions, including those that have mismatched 3' termini.
While there are some similarities to the mechanism of reverse transcription associated with circular retroplasmids of Neurospora spp., the activity of the pFOXC-RT differs in many regards. This is somewhat surprising considering that retroplasmid RTs are considered to be members of the same monophyletic family of RTs. The differences detected thus far include:
1. The pFOXC-RT readily uses the 3' OH of RNA templates to prime cDNA synthesis, while the pMaur-RT rarely uses RNA primers and appears to depend on a specific RNA sequence, rather than base-pairing of the 3' end of the RNA primer (Wang and Lambowitz 1993a)
2. The pFOXC-RT is able to use DNA primers that anneal to internal regions of the transcript (Simpson et al. 2004), whereas the pMaur-RT cannot (Wang et al. 1992; Chen and Lambowitz 1997)
3. The pFOXC-RT is able to copy DNA templates, while the pMaur-RT cannot
4. Treatment of pFOXC-containing mt RNPs with micrococcal nuclease results in RT preparations free of endogenous nucleic acids, whereas MN-treated pMaur-RT preparations contain endogenous cDNA products that are used as primers for reverse transcription (Wang et al. 1992)
5. The pFOXC-RT has low specificity for RNAs, whereas the pMaur-RT highly prefers RNAs having a 3' terminal CCA sequence (Chen and Lambowitz 1997)
The enzymatic differences between RTs encoded by circular and linear retroplasmids likely reflect specific requirements involved in cDNA initiation and plasmid replication. For example, the flexibility the pFOXC-RT exhibits in using different types of primers and its ability to utilize minimally base-paired primer-template combinations could be properties that are essential for maintaining terminal repeat structures. In addition, its ability to extend mismatched 3'-terminal nucleotides would ensure that RNAs having heterogeneous terminal sequences could be copied. To our knowledge, the only other DNA polymerase known to extend primers having 3' mismatches is the Tetrahymena TERT, and this property is associated with the "de novo" repeat addition that adds telomeres to non-telomeric DNAs (Wang and Blackburn 1997; Wang et al. 1998; Ware et al. 2000).
Was this article helpful?