MAGNESIUM AND CATALYTIC RNA 247
The guanosine binding site of group I introns is located within P7 — a six - base -
pair helix with a single base bulge. The conserved bulge forms a base triple
(three bases connected by hydrogen - bonds in several specifi c, well - known
patterns, a very common motif in introns) that stacks with the terminal gua-
nosine. For instance, in the Tetrahymena group I intron, A263 forms a base
triple with the C262 – G312 base pair that stacks on top of the terminal guano-
sine ( ω G) in the PDB: 1X8W structure to be discussed below. (See Figures 6.5
and 6.6 .) The guanosine - binding pocket is complex, bringing together nucleo-
tides of the J6/7 and J8/7 joining regions within the major groove of P7.^32 The
binding pocket will be discussed for several group I intron crystal structures
in the following paragraphs.
The fi rst crystal structure of a group I ribozyme, a 160 nucleotide P4 – P6
domain of the self - splicing Tetrahymena thermophila intron, was published in
1996 at a resolution of 2.8 Å (PDB: 1GID).^13 The crystal structure was solved
using multiwavelength anomalous diffraction (MAD) and single isomorphous
replacement (SIR) using an osmium derivative to pinpoint metal - ion binding
sites. The fi nal model consisted of 154 nucleotides (nt), a total of 28 metal ions
and six waters. The complete secondary structure of the intron – exon complex
is shown in Figure 1 of reference 13 and reprinted as Figure 6.4. The Tetrahy-
mena P4 – P6 domain in the crystal structure PDB: 1GID contained about half
of the intron ’ s active site — the base - paired segments P4, P5, and P6 and joining
regions J3/4, J6/7, and J4/5 — conserved elements found in the catalytic cores
of all group I introns. The domain contained the P5abc extension (P5a, P5b,
and P5c), required for effi cient catalysis in the Tetrahymena thermophila intron.
It also featured noncanonically paired or “ bulge ” regions responsible for inter-
actions between helices (backbone – backbone interactions), metal - ion binding
sites, and interactions involving a GAAA tetraloop. In the crystal structure,
the P4 helix (PDB: 1GID residues C208 – C216, G112 – U107) interacts with the
A - rich bulge (PDB: 1GID residues G188 – U182) and the tetraloop receptor
helix J6a/6b (PDB: 1GID residues C222 – G227, G251 – U247) interacts with the
GAAA tetraloop (PDB: 1GID residues G149 – C154). Within the A - rich bulge,
A184 and A186 are invariant in group I intron subclasses IB and IC and A183
is conserved — deletion of the A - rich bulge or a point mutation of A186 to U
disrupts the domain ’ s global structure. In fact, the global P4 – P6 domain struc-
ture is sensitive to numerous nucleotide mutations in the A - rich bulge. The
A - rich bulge starts the folding of the domain ’ s substructure to include the
three - helix junction of P5a, P5b, and P5c. Phosphate oxygens in the A - rich
bulge are unusually close to each other, but charge density is alleviated by the
presence of several magnesium ions. The distances are collected in Table 6.1.
These interactions, plus others, stabilize the P4 – P6 domain ’ s conformation.
Site - directed mutagenesis and chemical protection studies indicate that G212
in the P4 helix interacts with residues in the A - rich bulge, and the crystal
structure shows interactions with A184 (O2 ′ G212 – N1 A184 = 2.8 Å , for
example). The same sorts of interactions are found between the GAAA