RNA Editing
One of the basic assumptions of the central dogma of molecular biology (DNA makes RNA makes protein) is that the primary sequence of the messenger RNA (mRNA) is a faithful copy of the DNA sequence of the gene. This idea was modified slightly with the discovery of interven ing sequences (introns). The sequence of the mRNA coded in the genome could be "interrupted," and the intron was removed during RNA processing. Nevertheless, the sequence of the mRNA could be found within the gene sequence.
Since about 1987, however, evidence has accumulated for cases in which the sequence of the mRNA is NOT found in its gene. The mRNA sequence was somehow altered to provide the codons that did not exist in the gene; the RNA was "edited."
Evidence for RNA editing came from several systems when comparisons of protein, cDNA (mRNA), and genomic DNA sequences were made. RNA editing can be divided into two broad classes: single nucleotide changes, and addition of one or more nucleotides. Please see the reviews of (1-9) for general reviews on the phenomena.
I. Evidence of the first type of editing comes from mammalian and plant systems.
Apolipoprotein B is expressed in two forms in mammals. In humans, apo B100 (512kd) is made in the liver, and is involved in transport of endogenously synthesized triglycerides, as well as being the ligand that mediates the clearance of LDL-cholesterol from circulation by LDL receptor. A second form, apo B48, (242kd), about 1/2 the size of apo B100, is made in the intestine, and is essential for chylomi cron formation and transport of dietary cholesterol and triglycerides. In the rat liver, both forms are made in the liver, and the ratios of the two forms is regulated by thyroid hormone.
Sequencing of cDNAs made from mRNAs from these tissues showed two forms of mRNA. One form had an open reading frame consistent with the size of the high MW protein. The other, however, had a stop codon (UAA) at a position occupied by a CAA codon (glu) in the other form. The length of the shorter open reading frame corresponded to the low MW protein. Analysis of genomic DNA, however, indicated that only a single gene existed, and the gene had a CAA codon (10-13).
Similar C to U changes were found in studies of plant mitochondrial DNA expression (14,15). It had been thought that the CGG codon, normally arginine, coded for tryptophan in plant mitochondria. This was because CGG codons were often found in plant mitochondrial DNA sequences at positions that are trp in other organisms (the protein sequences of many mitochondrially encoded respiratory chain subunits are highly conserved).
However, when the mRNAs were sequenced directly using specific
primers, it was found that CGG's coding for conserved arg
residues were CGG, whereas CGG's coding for conserved trp
residues were UGG.
Sequencing of other mitochondrial specific mRNAs found changes at
other codons: CAC to UAC (arg to tryp), CAT/CAC to UAU/UAC (his
to tyr), CTC to UUC (leu to pro), TCT/TCG to UUU/UUG (ser to
phe/leu), and CGG/CCA to CUG/CUA (proline to leu). These changes
occured only at the position where the change was to a conserved
amino acid. No differences between DNA and mRNA were observed in
non-coding regions of the mRNA.
A more recent example has been found in the expression of certain subunits of glutamate-gated channels in the brain (16). In this case, A in the genome becomes "I" in the mRNA (I, inosine, would behave like a G residue).
In all cases, a single deamination reaction could account for the observed change (C to U, A to I), although base exchange, as occurs during modification of G to Q in the anticodon of certain tRNAs, has not been rigorously ruled out.
Recently, an in vitro system to study the RNA editing of ApoB mRNA has been developed. Scott and colleagues (17-19), have used primer extension analysis of the products of incubating synthetic mRNA templates with cell-free extracts of McArdle 777 cells, a rat hepatoma cell line that produces both forms of ApoB. They can demonstrate about 2-3% editing. Editing is not dependent upon nucleotide triphosphates or creatine kinase, suggestion that an exchange reaction is unlikely. The activity is not found in extracts of HeLa cells, or Hep3b cells, neither of which make Apo B48. The proteins responsible for specific recognition, as well as the deaminase may assemble into a complex, called the editosome (20).
Analysis of mRNAs carrying deletions of various segments of the mRNA localized the minimally required sequence to be a 55 nt region spanning the editing site (i.e., 55 nt out of 14,121 nt for the mRNA!). Further deletion and mutational analysis demonstrate a key role of the 12 nucleotides immediately downstream of the editing site. Appearance of this sequence in other genes may indicate an RNA editing site.
Addition of nucleotides to mRNA
An even stranger phenomena was discovered in the kinetoplastid mitochondrial system. Kinetoplstid flagellates are parasitic protozoa whose life cycle passes between two organisms: mainly insect and vertebrate. The vertebrate bloodstream form has a "repressed" mitochondrial genome expression: no cyotochromes are present; an incomplete TCA system; energy is derived from glycolysis only. The procyclic form, in the insect, has "derepressed mitochondrial genome expression: complete set of TCA enzymes and cytochromes.
The kinetoplast referes to the very large mitochondrrion which supplies energy required for flagellar function. It has two types of genomic DNA: about 10,000 or more heterogeneous 1kb minicircles, and about 50 homogeneous 22kb maxicircles. The maxicircle is equivalent to mitochondrial (mt) DNA of other organisms. It has genes for 2 mt rDNAs (1150 and 600nt), cytochrome oxidase (COX) subunits I and II, NAD dehydrogenease subunits I, IV and V, ATPase VI, and several other open reading frames. The function of the minicircles was unknown until recently (see below).
It was noted that the DNA sequence of several maxicircle genes were unusual. Several genes lacked an AUG initiation codon; several had internal frameshifts in the protein coding region. Analysis of cDNAs made from mRNA transcripts showed that the functional mRNA was created by the insertion of one or more U residues throughout the sequences (4,21,22). In the case of T.brucei COXIII, over 50% of the sequence is generated in this manner (23).
Examination of many clones obtained from PCR cDNA gene banks showed that in addition to unedited and fully edited transcripts, the RNA population also contained many forms of "partially edited" transcripts, which were taken to be intermediates in the process (22). Among these were some forms that had additional U residues which would later have to be removed if they went on to fully edited forms. It was also noted that, in general, editing seemed to proceed from the 3' to 5' direction. This suggested that the editing process was a post- rather than a co- transcriptional event. This is in contrast to the situation found for certain mammalian viruses (see end).
The biggest question, however, is where is the information coming from to specify how many U's, and where they should go in the primary transcript. Small RNAs were detected that possessed a sequence that had complimentarity to regions of edited RNAs, if one allowed for G:U paring as well as G:C pairing. The 5'-end of these RNAs had a region that had complimentarity to a portion of the mRNA transcripts 3' to the site of editing. This region would be present in both edited and non-edited transcripts. Finally, these small RNAs had 3'-oligo U tracts (24). These RNA transcripts are coded in both maxi- and mini circle DNA.
A model was proposeed in which the small RNAs would serve as a "guide" to fill in the missing information (25). The 5'-end of the guide RNA (gRNA) could serve as an "anchor" sequence, pairing with a region 3' to a site to be edited. The 3'-OH of the oligo-U tail attacks the phosphodiester bond of the mRNA at the first mismatch. The results in a chimeric RNA, whose 5'-end is the guide RNA, and the 3'- end in the 3'-end of the mRNA transcript, linked by the oligo-U sequence. Evidence for the existence of the chimeric RNAs were ob tained by PCR amplification and sequence analysis (26). The oligo-U's then pair with A or G of the guide RNA/mRNA duplex starting after the mismatch. This continues until no further pairing can occur (C's or U's). The 3'-OH of the 5'-half of the mRNA then attacks the phospho diester bond of the chimeric RNA at this mismatch, resulting in the "religation" of the 5- and 3'- halfs of the mRNA, but now separated by some U residues. Note that the complimentarity to the guide mRNA has now been extended. Further pairing beyond the editing point might occur for a single guide RNA until the next occurence of a G or A in the guide RNA not present in the mRNA. The process can then begin again. This process is similar to the mechanisms proposed for mRNA intron splicing.
Recently, an in vitro system has been able to form the chimeric RNAs (27,28). Formation of the chimera is ATP dependent. Such a system should aid in the isolation and purifcation of the enzymes responsible.
Three other types of nucleotide insertion have been documented:
The first type is the insertion of one or two G residues in certain mammalian viral mRNAs (29,30). Because this appears to occur within a run of G residues, it is thought to occur during transcrip tion, possibly by "slippage" of the RNA polymerase.
The second case is that of insertion of C residues with certain mRNAs of the slime mold, where a single C residue is inserted at 54 different locations (31).
Finally, sequence analysis of gene and RNA suggest that certain tRNA transcripts are also edited. In one case, a UC dinucleotide in the anticodon stem of a rat aspartate tRNA is actually a CT in it's gene (32). A second example was found in the amino acid acceptor stem of mitochondrial tRNAs of a protozoan (33). In this case, RNA editing occurs to establish proper base-pairing within the stem, not present in the gene sequence; the changes seem localized to the 5' side of the stem, and involve T to A, T to G, and A to G changes. Thus, structural, as well as informational RNAs can be edited.
1. Hajduk, S. L.; Harris, M. E.; Pollard, V. W. (1993) RNA editing in kinetoplastid mitochondria. FASEB J. 7, 54-63
2. Gray, M. W.; Covello, P. S. (1993) RNA editing in plant mitochon dria and chloroplasts. FASEB J.7, 64-71
3, Benne, R. (1989) RNA editing in trypanosome mitochondria. Biochem. Biophys. Acta1007, 131-139
4.Simpson, L.; Shaw, J. (1989) RNA editing and the mitochondrial cryptogenes of kinetoplastid protozoa. Cell 57, 355-366
5. Eisen, H. (1988) RNA editing: who's on first? Cell 53, 331-332
6. Stuart, K. (1991) RNA editing in mitochondrial mRNA of trypanosomatids. TIBS 16, 68-72
7. Czech, T. R. (1991) RNA editing: world's smallest introns? Cell 64, 667-669
8. Cattaneo, R. (1992) RNA editing in chloroplast and brain. TIBS 17, 4-5
9. Chan, L. (1993) RNA editing - exploring one mode with apolipopro tein B messenger RNA. Bioessays 15, 33-41
10. Chen et al. (1987) Apolipoprotein B-48 is the product of a messenger RNA with an organ-specific in-frame stop codon. Science 238, 363-366
11. Powell et al. (1987) A novel form of tissue-specific RNA processing produces apolipoprotein B48 in intestine. Cell 50, 831-840
12. Tennyson et al. (1989) Expression of apolipoprotein B mRNAs encoding higher and lower molecular weight isoproteins in rat liver and intestine. Proc. Natl. Acad. Sci. USA 86, 500-504
13. Tennyson et al. (1989) Characterization of single base substitution in edited apolipoprotein B transcripts. Nucl. Acids Res. 17, 691-698
14. Gualberto et al. (1989) RNA editing in wheat mitochondria results in the conservation of protein sequences. Nature 341, 660-662
15. Covello; Gray (1989) RNA editing in plant mitochondria. Nature 341, 552-666
16. Sommer, B.; K" hler, M.; Sprengel, R.; Seeburg, P. H. (1991) RNA editing in brain controls a determinant of ion flow in glutamate-gated channels. Cell 67, 11-19
17. Driscoll, D. M.; Wynne, J. K.; Wallis, S. C.; Scott, J. (1989) An in vitro system for the editing of apolipoprotein B mRNA. Cell 58, 519-525
18. Davies, M. S.; Wallis, S. C.; Driscoll, D. M.; Wynee, J. K.; Williams, G. W.; Powel, L. M.; Scott, J. (1989) Sequence requirements for apolipoprotein B RNA editing in transfected rat hepatoma cells. J. Biol. Chem. 264, 13395-13398
19. Shah, R. R.; Knott, T. J.; Legros, J. E.; Navaratnam, N. A.; Greeve, J. C.; Scott, J. (1991) Sequence requirements for the editing of apolipoprotein B mRNA. J. Biol. Chem. 266, 16301-16304
20. Backus, J. W.; Smith, H. C. (1991) Apolipoprotein B mRNA sequenc es 3' of the editing site are necessary and sufficient for editing and editosome assembly. Nucl. Acids Res. 19, 6781-6786
21. Shaw, J. M.; Feagin, J. E.; Stuart, K.; Simpson, L. (1988) Edit ing of kinetoplastid mitochondrial mRNAs by uridine addition and deletion generates conserved amino acid sequences and AUG initiation codons. Cell 53, 401-411
22. Sturm, N. R.; Simpson, L. (1990) Partially edited mRNAs for cytochrome b and subunit III of cytochrome oxidase from Leishmania tarentolae mitochondria: RNA editing intermediates. Cell 61, 871-878
23. Feagin; Abraham; Stuart (1988) Extensive editing of the cyto chrome c oxidase III transcript in iTrypanosome brucei. Cell 53, 413-422
24. Blum, B.; Simpson, L. (1990) Guide RNAs in kinetoplastid mitochondria have a nonencoded 3'-oligo(U) tail involved in recognition of the preedited region. Cell 62, 391-397
25. Blum, B.; Bakalara, N.; Simpson, L. (1990) A model for RNA edit ing in kinetoplastid mitochondria: "Guide" RNA molecules transcribed from maxicircle DNA provide the edited information. Cell 60, 189-198
26. Blum, B.; Sturm, N. R.; Simpson, A. M.; Simpson, L. (1991) Chi meric gRNA-mRNA molecules with oligo(U) tails covalently linked at sites of RNA editing suggest that U addition occurs by transesterifi cation. Cell 65, 543-550
27 Blum, B.; Simpson, L. (1992) Formation of guide RNA/messenger RNA chimeric molecules in vitro, the initial step of RNA editing, is dependent on an anchor sequence. Proc. Natl. Acad. Sci. USA 89, 11944-11948
28. Koslowsky, D. J.; G" ringer, H. U.; Morales, T. H.; Stuart, K. (1992) In vitro guide RNA/mRNA chimaera formation in Trypanosoma brucei RNA editing. Nature 356, 807-809
29. Thomas; Lamb; Peterson (1988) Two mRNAs that differ by two non templated nucleotides encode the amino coterminal proteins P and V of the paramyxovirus SV5. Cell 54, 891-902
30. Cattaneo et al. (1989) Measles virus editing provides an addi tional cysteine-rich protein. Cell 56, 759-764
31. Mahendran, R.; Spottswood, M. R.; Miller, D. L. (1991) RNA edit ing by cytidine insertion in mitochondria of Physarum polycephalum. Nature 349, 434-438
32. Beier, H.; Lee, M. C.; Sekiya, T.; Kuchino, Y.; Nishimura, S. (1992) Two nucldotides next to the anticodon cytoplasmic rat tRNA+asp are likely generated by RNA editing. Nucl. Acids Res. 20, 2679-2683
33. Lonergan, K. M.; Gray, M. W. (1993) Editing of transfer RNAs in Acanthamoeba castellanii mitochondria. Science 259, 812-816