domain was attached to stem II of hammerhead ... Second messengers such as cAMP or cGMP are feasible ... PCR to generate double stranded DNA templates.
© 7999 Oxford University Press
Nucleic Acids Symposium Series No. 42
275-276
Allosteric ribozymes sensitive to the second messengers cAMP and cGMP Makoto Koizumi, Jo Nita Q. Kerr, Garrett A. Soukup and Ronald R. Breaker
Department of Molecular, Cellular and Developmental Biology, Yale University, 219 Prospect Street, New Haven, CT 06520, USA
ABSTRACT We have engineered allosteric ribozymes by combining modular rational design with combinatorial strategies. This new procedure was used to create allosteric ribozymes that are activated by specific nucleoside 3',5'-cyclic monophosphates (cNMPs). A random-sequence domain was attached to stem II of hammerhead ribozymes via a communication module that serves as an interface between ribozyme and the effector binding site. Subjecting this initial random pool to in vitro selection methods produced populations that respond, or cleave, only in the presence of specific effector molecules. From generation 18, 20 and 23, cGMP, cCMP and cAMP-specific responsive ribozymes, respectively, were isolated and characterized. These methods show great promise for engineering allosteric ribozymes and for creating new ligand-specific aptamers.
INTRODUCTION Both rational and combinatorial design strategies can be used to create ribozymes that function in response to small organic compounds. For example, previous ribozyme engineering efforts in our laboratory have yielded a series of "allosteric ribozymes" that can be controlled by effectors such as ATP and flavin mononucleotide (FMN).1'3 In each case, a ligandspecific aptamer domain was fused to stem II of the hammerhead ribozyme. When the aptamer binds its corresponding ligand, the resulting conformational change influences the activity of the adjoining catalytic domain of the construct. Such allosteric ribozymes have significant potential for use as precision biosensor elements or for novel genetic control elements for use in genetic engineering protocols. Second messengers such as cAMP or cGMP are feasible effectors for allosteric ribozymes for intracellular gene regulation because concentrations of these nucleotides and their analogues are controlled by cells, and their concentrations also can be intentionally influenced by external factors. To create allosteric ribozymes with selective sensitivity to
cAMP or cGMP, it is critical that the complementary aptamer domain be made highly specific for its respective effector. Conventional methods for the generation of aptamers employ an affinity column to which the desired ligand is attached. However, steric hindrance by the matrix precludes full access of the RNA to the target ligand. We have employed a new procedure (Fig. 1) that favors the isolation of allosteric ribozymes sensitive to any effector molecule of choice; in this case the four natural nucleoside 3',5'cyclic monophosphates (cNMPs). A random-sequence domain was attached to stem II of hammerhead ribozymes via a "communication module" that serves as an interface between the ribozyme and the effectorbinding site.3 A pool of 1015 RNA sequence variants based on this tripartite design was used for in vitro selection, which yielded populations of allosteric ribozymes that cleave only in the presence of specific cNMP effectors. These methods should be applicable for the development of new allosteric ribozymes that respond to nearly any effector molecule. MATERIAL AND METHODS In vitro selection for allosteric ribozymes sensitive to cAMP, cGMP, cCMP, and cUMP was carried out using repeated rounds of negative and positive selection. For negative selection, an initial randomsequence RNA pool (9.3 nmol, 5.6 x 1015 molecules) was prepared by in vitro transcription and incubated in a reaction mixture containing 50 mM Tris-HCl (pH 7.5) and 20 mM MgCl2 in the absence of the four cNMPs, and uncleaved products were isolated by denaturing 10% polyacrylamide gel electrophoresis (PAGE). Positive selection was subsequently performed in the same buffer containing 500 uM each of the four cNMPs. The resulting 5'-cleavage products were purified by PAGE and were amplified by reverse transcription and PCR amplification (RT-PCR). Additional rounds of selective amplification were repeated in a similar fashion until effector-sensitive ribozyme function was detected. The cNMP-sensitive RNA populations were cloned, sequenced and further analyzed by establishing the effector-mediated modulation of ribozyme kinetics.
276
Nucleic Acids Symposium Series No. 42 communication module
cNMP
*obs + /*obs (nn/nff ratiol
log frobs
—1
-
-4 -3 -2 -1 0 -+o—i—•—•—•-
5 1 0
70
•1
59
1. transcription 2. PAGE
Figure 1. Scheme for the "allosteric selection" ofribozymes that are activated by specific effector molecules. Precursor RNAs are (I) subjected to negative selection in the absence of effector. Uncleaved RNAs are isolated by PAGE, then subjected to positive selection in the presence of a mixture of the four cNMPs. Cleaved RNAs are (II) amplified by RTPCR to generate double stranded DNA templates. The resulting DNAs are (III) transcribed using bacteriophage T7 RNA polymerase (T7 RNAP) to generate a new population of RNA molecules that are (IV) subjected to the next round of negative and positive selections. (V) Double-stranded DNAs from the desired rounds of selection are cloned and sequenced for further analysis. The boxed T7 represents a double-stranded promoter sequence for T7 RNAP.
RESULTS AND DISSCUSSION Using the selection scheme depicted in Fig. 1, we conducted repeated rounds of negative and positive for in vitro selection for allosteric ribozymes sensitive to cAMP, cGMP, cCMP, and cUMP. In generation 18, 20, and 23, specific responses for cGMP, cCMP, and cAMP, respectively, were observed. However, specific cleavage with cUMP was not observed even after 28 rounds of in vitro selection. The effector specificities of the representative clones cGMP-1, cCMP-1, and cAMP-1 were determined as described in Fig. 2. Isolated clones exhibit a specific allosteric response only when presented with their corresponding cNMP effector. Specifically, the ratio of the observed rate constants in the presence (kobs+) versus the absence (kobs) of the corresponding effector for the cGMP-1, cCMP-1, and cAMP-1 clones are 510, 70, and 59, respectively. These values reflect the rate enhancements that are induced by the cNMP effectors.
Figure 2. Selective activation of representative allosteric ribozymes by cNMPs. Internally 32P-labeled RNAs that are sensitive to cGMP, cCMP and cAMP (as identified) were incubated for 15 min in the absence of effector (-) or in the presence of 500 uM of the nucleoside 3',5'-cyclic monophosphates A, G, C and U as indicated under the reaction conditions used for in vitro selection. Reaction products were separated by denaturing 10% PAGE and the bands were visualized and quantified using a Phosphorlmager and ImageQuant software (Molecular Dynamics). Open and filled arrowheads identify the precursor and 5' cleavage products, respectively. The 3' cleavage products have greater electrophoretic mobility than the significantly larger precursor RNAs and 5'-cleavage fragments, and therefore are not present on the images.
Using various analogues for the cNMPs, we investigated the requirements for the specific recognition the allosteric effectors. For example, each clone remains inactive in the presence of its corresponding nucleoside 5'-monophosphate or nucleoside 3'-monophosphate analogue of cNMP, indicating that each effector-binding site can distinguish small differences in the chemical structures of their ligand. Also, when cNMP was added into ribozyme reaction, rapid effector-dependent activation was observed instantaneously. These results indicate that allosteric ribozymes can rapidly convert to an active state upon specific recognition of their cognate effector molecule. Further optimization of these allosteric ribozymes could lead to their use as precision biosensor elements or as novel genetic control elements. REFFERENCES 1. Tang,J. and Breaker.R.R. (1997) Chem. Biol. 4,453-459. 2. Tang,J. and Breaker.R.R. (1997) RNA 3, 914-925. 3. Soukup.GA. and Breaker,R.R. (1999) Proc. Natl. Acad. Sci. USA 96, 3584-3589.
