Ribonuclease mimics

a mixture of the AC and AD positional isomers, as the mono-protonated species. 

We had invoked a simultaneous two proton transfer mechanism rather than a sequential mechanism - in which one catalytic group followed the other in the overall process - and were able to test this with a technique called proton inventory. We examined the original cyclodextrin fcfv-imidazole in mixtures of water and D2O and saw that the rate constant as a function of deuterium concentration followed a curved line, indicating that the isotope effect involved two different protons rather than a single one. To validate this, we also examined the same kind of plot with the cyclodextrin mono-imidazole, in which only one proton would be expected to be moving in the transition state, and this indeed followed a linear plot supporting a single proton motion in the isotope effect. 

Some of this work is described in reviews that include the cyclodextrin work. We also examined a comparison of catalysts with the AB placement of imidazoles but based on either a-cyclodextrin, j8-cyclodextrin or y-cyclodextrin. And these were also compared with substrates carrying various groups other than t-butyl. The conclusion from this work was that by far the best rates and selectivities were seen in cases where there was a very 

Helical structures are important in proteins, and the extent of heUx formation in a given polypeptide can be influenced by external factors. In a study related to this question we saw that with appropriate cyclodextrin dimers we could induce helix formation in some oligopeptides if the helix structure presented hydrophobic side chains in a geometry such that our dimer could bind to them, stabilizing the helix.  

In our simplest study, we examined the cleavage/cyclization of uridyluridine 16, abbreviated UpU, with a 3 -5 phosphate diester link (Fig. 1.9). We used a concentrated buffer consisting of imidazole and imidazolium cation, mimicking the state of the two imidazoles in the enzyme. Indeed we saw that the buffer catalyzed the cyclization/cleavage reaction, forming uridine 2 -3 cyclic phosphate and liberating uridine. However, we also saw that there was some isomerization of the starting material, from the 3 -5 phosphate diester to the 

2 -5 phosphate diester. The reaction showed a bell-shaped pH vs rate profile, indicating that both the basic imidazole and the acidic imidazolium ion were catalytic, as with the enzyme. In the enzyme the two catalytic groups operate simultaneously, while in this imidazole buffer system they operated sequentially (the rate was first order in buffer concentration). There was a first step catalyzed by imidazolium ion, and then a second step catalyzed by imidazole. 

In a two-step mechanism there must be an intermediate, a five-coordinate phosphorane 17. We confirmed this from the observation that this intermediate could branch, either to the cleavage product 18 or to the isomer 19 of the starting material. Detailed kinetics showed how each step was catalyzed.We did a related study for the cleavage by a combination of imidazole and Zn + in water.  

Even before this work was started, we had prepared a mimic 20 of ribonuclease consisting of j8-cyclodextrin with two imidazole rings attached to the C-6 primary methylenes of two different glucose residues of the cyclodextrin.The substrate was not an RNA, but a cyclic phosphate 21 whose cleavage mimicked the cleavage of the 2 -3 cyclic phosphate of RNA. This substrate, 4-t-butylcatechol cyclic phosphate, bound well into the cyclodextrin in water, and the catalyst hydrolyzed the cyclic phosphate with a bell-shaped pH vs rate profile, showing that both the imidazole and the imidazolium ion were playing a catalytic role. 

The first catalyst had imidazoles on the C-6 carbons of the farthest apart glucose residues A and D, contaminated to some extent by the A,C isomer. We saw that the hydrolysis was quite regiospecific (Fig. 1.10), cleaving the bond between the phosphorus and the oxygen on carbon 1 of the substrate to form product 22. This was as expected from a mechanism in which a water molecule is delivered perpendicular to the cyclodextrin axis as models predicted. In a later catalyst the imidazoles were mounted further from the ring and significantly cleaved the P—O bond to carbon 2, again consistent with models.  

---

Breslow R, Dong SD, Webb Y, Xu R. Further studies on the buffer-catalyzed cleavage and isomerization of uridyluridine. Medium and ionic strength effects on catalysis by morpholine, imidazole, and acetate buffers help clarify the mechanisms involved and their relationship to the mechanism used by the enzyme ribonuclease and by a ribonuclease mimic. J. Am. Chem. Soc. 1996 118 6588-6600. 

Phosphate esters can be cleaved by template catalysts, especially those with cyclodextrin binding groups and linked catalytic groups. Catalysis of the hydrolysis of a bound cyclic phosphate by ribonuclease mimics has been extensively studied [92-98], as has catalysis by enzyme mimics carrying bound metal ions [99-102]. 

R. Breslow, P. Bovy, C. Lipsey Hersh, Reversing the selectivity of cyclodextrin bisimidazole ribonuclease mimics by changing the catalyst geometry, J. Am, Chem. Soc., 1980, 102, 2115-2117. 

R. Breslow, C. Schmuck, Goodness of ht in complexes between substrates and ribonuclease mimics effects on binding, catalytic rate constants, and regiochemistry, J. Am. Chem. Soc., 1996, 118, 6601-6605. 

The mechanistic work also guided the synthesis of a ribonuclease mimic that is more effective as an artificial enzyme than are related compounds whose geometries were based on the classical ideas about the enzyme mechanism. The artificial enzyme shares many of the properties of the natural enzyme. 

The availibility of three geometric isomers of our artificial enzyme lets us examine other reactions that can show bifunctional catalysis. Enolization of a ketone— and its addition to an aldehyde group in an aldol condensation—are two cases examined so far in which an isomer of our catalyst is preferred that is not the one that was best in the ribonuclease mimic. The geometric preference indicates something novel about the geometry of enolization reactions. 

We applied this test to our ribonuclease mimic, the 6A,6B isomer of cyclodextrin bisimidazole, cleaving the bound cyclic phosphate 21. We found that there was indeed a square dependence of the kinetic isotope effect on the mole fraction of deuterium in the water solvent, and interestingly the values of the isotope effect for the two protons in flight were almost identical with those that had been seen with the enzyme itself and its normal substrate.As described above, the protonation in the model system involves an imidazolium ion putting a proton on the substrate phosphate anion as the imidazole delivers a water molecule to the phosphorus. 

---