HDV Family of Self-Cleaving Ribozymes - Progress in Molecular Biology and Translational Science - page 139

search software on viral genomes. 3,111 With the advent of inexpensive DNA
sequencing, however, searchable genomic space has expanded enough that
the intricate HDV secondary structure can be detected in numerous loci.
The exponential increase in computer processing power has also facili-
tated the identification of complex RNA secondary structures on a broad
scale. In a functional RNA molecule, it is typically the structure rather than
the primary sequence that is conserved. In principle, a helix can form
between any two nucleotides capable of base pairing, and the helical regions
primarily dictate the overall structure that an RNA will assume. Conse-
quently, two ribozymes with identical secondary structures can be unre-
cognizable on a sequence level. This is particularly striking for HDV
ribozymes, where approximately 70 nucleotides are required to form the
motif, but only 6 are invariant between the two structures ( Fig. 4.2 ).
Although sequence-based searches have been highly refined over the past
few decades and have met with widespread success in protein-coding gene
identification, their utility is thus limited in identifying catalytic RNAs. Fur-
thermore, catalytic RNAs, unlike their protein counterparts, are not del-
imited by conserved translation initiation, termination, and splicing sites
that can aid in the identification of protein-coding regions with similar
function but different primary amino acid sequences. However, proteins
with similar domain arrangements will often catalyze similar reactions,
as do RNA molecules that share comparable folding patterns. Thus, a
structure-based approach is more applicable to identify ribozymes.
Structure-based searches for folded RNAmolecules were first developed
in the late 1980s. However, early programs required user-defined
helices, and as such they could not be used to identify unknown base-
pairing regions from a random chain of nucleotides. 112-114 The RNAMOT
program overcame many of the limitations associated with previous
motif-searching software by first allowing a helix to form from any string
of nucleotides and subsequently analyzing downstream portions for
complementarity. 115 RNAMOTwas used as a scaffold to develop additional
structure-based search programs, and it and its successors have been
repeatedly used to identify novel tRNAs, group I self-splicing introns, ham-
merhead ribozymes, and ligand-binding RNAs at new genomic
loci. 108,111,116,117 Comparable searches for HDV-like ribozymes were not
performed until after the first crystal structure was solved and the motif
was identified outside of the virus in the mammalian CPEB3 gene.
The combined effects of a complex secondary structure, limited search-
able genomic space, and a wealth of more readily detectable RNA motifs
have slowed the identification of novel HDV ribozymes. Nevertheless,