Aptamers to Natural Products

Jenison et al. (1994) selected an RNA aptamer that binds to the alkaloid theophyl-line with high sensitivity and high specificity. Theophylline is used as bronchodi-lator in the treatment of asthma, bronchitis, and emphysema, but because of its narrow therapeutic index, serum levels must be monitored carefully to avoid serious poisoning. Theophylline is also chemically similar to theobromine and caffeine (Fig. 4.5), which may be present in serum samples due to consumption of coffee or chocolate. Thus, diagnostic methods must discriminate efficiently among these compounds. A counterselection with caffeine was performed during the SELEX process. The dissociation constant of the selected aptamer was 100nmol/L and its affinity to theophylline was remarkably 10 000 times higher than to caffeine, which differs only from theophylline by a single methyl group at nitrogen atom N7. Antibodies against theophylline show a discrimination factor of only 1000. This demonstrates that RNA molecules can exhibit an extremely high degree of ligand recognition and discrimination.

Solving the structure of the theophylline aptamer by NMR spectroscopy (Zimmermann et al., 1997) yielded an explanation of the high specificity and demonstrated an unusual folding motif, including an S-turn of the backbone, a novel base zipper, and a 1-3-2 stacking motif. The theophylline is nearly completely enclosed within the aptamer and the many contacts between the ligand and the RNA direct the specificity. In particular, the presence of hydrogen bonds between the N7 position of the theophylline and the aptamer account for the high discrimination factor compared with caffeine. The methyl group at position N7 of caffeine disrupts the hydrogen bonds and steric repulsion may lead to a loss of further hydrogen bonds and therefore a loss of binding affinity. The structure is deposited at the RCSB protein databank under the code 1EHT.

Caffeine Aptamer Interaction With Target

Fig. 4.5 Theophylline and its closely related compounds caffeine, theobromine and 3-methylxanthine. The selected aptamer recognizes theophyl-lin 10000 times better than caffeine with a Kd of 100 nmol/L for theophyl-line. A single C to A mutation in position 27 switches the specificity of the aptamer from theophylline to 3-methylxanthine.

Fig. 4.5 Theophylline and its closely related compounds caffeine, theobromine and 3-methylxanthine. The selected aptamer recognizes theophyl-lin 10000 times better than caffeine with a Kd of 100 nmol/L for theophyl-line. A single C to A mutation in position 27 switches the specificity of the aptamer from theophylline to 3-methylxanthine.

Recently Anderson and Mecozzi (2005) have reduced the theophylline aptamer from an original 33-mer to a remarkable 13-mer by using iterative computational deletion and molecular dynamics simulations. The resulting 13-mer still shows good binding affinity and specificity to theophylline by discriminating the structurally close caffeine.

The theophylline aptamer has been used to allosterically regulate ribozymes. The allosteric ribozyme can either be activated or inhibited by effectors binding to the appended aptamer domain. Soukup et al. (2000) selected allosteric hammerhead ribozymes that were activated more than 3000-fold upon theophylline binding. In addition they isolated an allosteric ribozyme variant that exhibited an effector specificity change from theophylline to 3-methylxanthine by carrying only a single C to A replacement in position 27 (C27A). Robertson and Ellington (2000) designed a ribozyme ligase whose activity was enhanced up to 1600-fold in the presence of theophylline.

Frauendorf and Jaschke (2001) used the allosteric hammerhead ribozyme in combination with a molecular beacon for the detection and quantification of theophylline. The allosteric hammerhead ribozyme cleaves only in the presence of the effector theophylline with a regulation factor of 100. Here they used an intermolecular system by combining the ribozyme with a substrate oligonucleotide, which carried the hammerhead cleavage site, a fluorophore, and a quenching dye as terminal labels. With this system the fluorescence signal of the ribozyme cleavage rate could be correlated to the theophylline concentration in a dynamic range of 0.01-2 mmol/L theophylline. The presence of caffeine showed no influence on the cleavage reaction (Frauendorf and Jaschke, 2001).

The theophylline aptamer has also been used to regulate gene expression. Harvey et al. (2002) showed inhibition of translation in the presence of theophylline by having introduced the theophylline aptamer into the 5'-untranslated region (UTR) of the target mRNA. The translation was inhibited efficiently, when the aptamer was introduced in triplicate within the 5'-UTR. Introduction within the coding region or in the 3'-UTR did not show any inhibition (Harvey et al., 2002). Additionally they demonstrated that the translation inhibition was due to the inhibition of the interaction of the small ribosomal subunit with the mRNA, thus inhibiting the formation of the 80S initiation complex.

Another example of the use of the theophylline-responsive riboswitch to control gene expression was given by Suess et al. (2004) and Desai and Gallivan (2004). Suess et al. introduced the riboswitch by using the theophylline aptamer as a receptor domain for specific ligand binding and a communication module proposed to perform helix slipping. This was inserted near the ribosome-binding site (RBS) in such a way that the aptamer's non-bound conformation interfered with the ri-bosome assembly. Binding of the ligand theophylline induced a structural transition and allowed ribosome binding and thus control of gene expression in Bacillus subtilis (Suess et al., 2004). Desai and Gallivan (2004) used the theophylline apta-mer to show that synthetic riboswitches can also operate in the Gram-negative bacterium E. coli. They cloned the theophylline aptamer in the 5'-UTR close to the RBS of a p-galactosidase reporter gene. They could show a theophylline

4.6 Aptamers to Natural Products

dose-dependent increase of p-galactosidase activity in E. coli, while the structurally similar caffeine showed no effect. They also approved the change in selectivity for 3-methylxanthine for the previously published point mutation C27A of the theophylline aptamer, showing gene regulation upon addition of 3 -methylxanthine, but not theophylline. They further determined that gene regulation occurs on the translation rather than the transcription level (Desai and Gallivan, 2004).

Stojanovic et al. (2000) selected a DNA aptamer to cocaine as a fluorescent sensor. The secondary structure of the aptamer was predicted to be a three-way junction with a lipophilic binding pocket at the junction. The authors split the cocaine aptamer at a loop region into two halves, which were allowed to reassemble to form the ligand-binding pocket as shown in Fig. 4.6. By tuning the complementarity of the two strands the equilibrium could be adjusted so that the two individual subunits would be favored over the assembled heterodimer in the absence of the ligand, but shifted to the assembled form upon ligand binding. One subunit was labeled with a fluorophore and the other with a quenching dye. Upon ligand binding, the labeled ends come into close proximity resulting in signal quenching. The authors demonstrated that the sensor could reliably report cocaine concentrations in a range from 10 to 1000 mmol/L. The sensor showed excellent selectivity, as cocaine metabolites such as benzoyl ecgonine or ecgonine methyl ester were not recognized. Figure 4.7 shows the structures of cocaine and its metabolites.

Caffeine Aptamer Sensing

472 nm 518 nm

Fig. 4.6 Secondary structure of the cocaine sensor. Self-assembly of the two aptamer halves in the presence of cocaine results in fluorescence quenching. F, fluorescence dye; Q, quenching dye; cocaine, cocaine-binding site. Figure according to Stojanovic et al. (2000).

472 nm 518 nm

Fig. 4.6 Secondary structure of the cocaine sensor. Self-assembly of the two aptamer halves in the presence of cocaine results in fluorescence quenching. F, fluorescence dye; Q, quenching dye; cocaine, cocaine-binding site. Figure according to Stojanovic et al. (2000).




Benzoyl ecgonine o


o oh

Fig. 4.7 Cocaine and its metabolites. H3C V—/

The aptamer recognizes only the cocaine, but not its metabolites. Ecgonine methyl ester

A year later, the same group presented an alternative sensoring system by shortening and thereby destabilizing one of the three stems. Now the stem and the three-way junction can only be formed upon ligand binding. Thus double labeling the ends of the stem with a fluorophore and a quencher would result in a quenching of the fluorescence signal upon ligand binding. The new aptameric sensor had a detection range for cocaine concentrations from 10 mmol/L to 2.5mmol/L, similar to the previously presented one. Additional features of the new aptameric sensor are its function in magnesium-free buffers and even in serum media. The selectivity of the sensor is still the same in discriminating the cocaine metabolites. The only drawback is that for clinical or forensic applications the sensor must have a sensitivity in the picomolar range (Stojanovic et al., 2001).

Another strategy was followed in using the original cocaine aptamer as a colori-metric sensor. For this the authors screened about 35 dyes that would bind to the three-way helical junction of the aptamer, but binding of the ligand cocaine would change the microenvironment of the dye chromophore by displacing it from the aptamer and thus generate a signal. It turned out that the cyanine dye diethylthio-tricarbocyanine iodide displayed a significant attenuation of absorbance dependent on the cocaine concentration. By this an aptamer-based colorimetric cocaine sensor was constructed with a detection range from 0.5 to 64 mmol/L (Stojanovic and Landry, 2002). Again the cocaine metabolites were not detected by the sensor.

Kato et al. (2000a) have isolated several single-stranded DNA aptamers that bind to the steroid cholic acid. Cholic acid is a metabolite of cholesterol, which belongs to the bile acids. After minimization the selected aptamers all showed a consensus secondary structure of a three-way junction, which probably forms a hydrophobic cavity in which the cholic acid could bind. The dissociation constants of the aptamers ranged from 6.4 to 67.5 mmol/L. In a follow-up publication the same group reported that the cholic acid aptamer binds even stronger to other

4.7 Aptamers to Organic or Fluorescent Dyes | 109

bile acids like chenodeoxycholic acid, lithocholic acid, cholic acid methyl ester, or even trans-androsterone, indicating that the binding action is governed merely by hydrophobic interaction (Kato et al., 2000b). It was suggested that binding depends only on the size or shape of the steroids and not on specific interactions between polar functional groups. This was further confirmed by mutagenesis analyses of the aptamers, suggesting that the binding does not require a sequence-specific tertiary structure but proper folding of the secondary structure.

Driven by the publication of Kato et al. (2000b), Stojanovic et al. (2003) used their cocaine aptamer with its three-way junction-binding site with various small modifications to sense other hydrophobic compounds like metabolic steroids characteristic for particular diseases in human urine (Stojanovic et al., 2003). For this they selected a variety of four different aptamers, all based on the cocaine aptamer. A phosphorthioate group was introduced at the three-way junction and functionalized with a thiol-reactive fluorophore. In the absence of the ligand, the fluorophore enters the three-way junction of the aptamer and gets quenched by the guanosine residues. In the present of the ligand, the fluorophore gets displaced by ligand binding, resulting in a threefold increase of fluorescence. The constructed variants of the aptamer have introduced either a mismatch in different positions on the stem and/or the position of the fluorophore was varied. The different aptamers had small differences in response to different metabolites and Stojanovic et al. could demonstrate by using a panel of four different aptamers that unspiked human urine could be distinguished from urine spiked with deoxycorticosterone 21-glucose or with dehydroisoandrosterone-3-sulfate, both metabolites characteristic for diseases (Stojanovic et al., 2003).

Dealing With Drugs

Dealing With Drugs

Get All The Support And Guidance You Need To Be A Success At Dealing With Drugs. This Book Is One Of The Most Valuable Resources In The World When It Comes To A Parents Guide To The Drug Talk.

Get My Free Ebook

Post a comment