Role of adenine functional groups in the recognition of the 3′-splice-site AG during the second step of pre-mRNA splicing

The AG dinucleotide at the 3′ splice sites of metazoan nuclear pre-mRNAs plays a critical role in catalytic step II of the splicing reaction. Previous studies have shown that replacement of the guanine by adenine in the AG (AG → GG) inhibits this step. We find that the second step was even more severely inhibited by cytosine (AG → CG) or uracil (AG → UG) substitutions at this position. By contrast, a relatively moderate inhibition was observed with a hypoxanthine substitution (AG → HG). When adenine was replaced by a purine base (AG → PG) or by 7-deazaadenine (AG → c7AG), little effect on the second step was observed, suggesting that the 6-NH2 and N7 groups do not play a critical role in adenine recognition. Finally, replacement of adenine by 2-aminopurine (AG → 2-APG) had no effect on the second step. Taken together, our results suggest that the N1 group of adenine functions as an essential determinant in adenine recognition during the second step of pre-mRNA splicing.

Stabilisation of TG- and AG-containing antiparallel DNA triplexes by triplex-binding ligands

We have used DNase I footprinting to examine the interaction of several triplex-binding ligands with antiparallel TG- and AG-containing triplexes. We find that although a 17mer TG-containing oligonucleotide on its own fails to produce a footprint at concentrations as high as 30 µM, this interaction can be stabilised by several ligands. Within a series of disubstituted amidoanthraquinones we find that the 2,7- regioisomer affords the best stabilisation of this TG triplex, though the 1,8- isomer also stabilises this interaction to some extent. By contrast the 1,5- and 2,6- regioisomers show no interaction with TG triplexes. Similar studies with a 13mer AG-containing oligonucleotide show the opposite pattern of stabilisation: the 2,6- and 1,5- isomers stabilise this triplex, but the 2,7- and 1,8-compounds do not. The polycyclic compound BePI strongly stabilises TG- but not AG-containing triplexes, while a substituted naphthylquinoline interacts with both antiparallel triplex motifs.

Structure-activity relationships derived by machine learning: the use of atoms and their bond connectivities to predict mutagenicity by inductive logic programming.

We present a general approach to forming structure-activity relationships (SARs). This approach is based on representing chemical structure by atoms and their bond connectivities in combination with the inductive logic programming (ILP) algorithm PROGOL. Existing SAR methods describe chemical structure by using attributes which are general properties of an object. It is not possible to map chemical structure directly to attribute-based descriptions, as such descriptions have no internal organization. A more natural and general way to describe chemical structure is to use a relational description, where the internal construction of the description maps that of the object described. Our atom and bond connectivities representation is a relational description. ILP algorithms can form SARs with relational descriptions. We have tested the relational approach by investigating the SARs of 230 aromatic and heteroaromatic nitro compounds. These compounds had been split previously into two subsets, 188 compounds that were amenable to regression and 42 that were not. For the 188 compounds, a SAR was found that was as accurate as the best statistical or neural network-generated SARs. The PROGOL SAR has the advantages that it did not need the use of any indicator variables handcrafted by an expert, and the generated rules were easily comprehensible. For the 42 compounds, PROGOL formed a SAR that was significantly (P < 0.025) more accurate than linear regression, quadratic regression, and back-propagation. This SAR is based on an automatically generated structural alert for mutagenicity.