News and Publications

Freezing Molecules in Space and Time: Assembly and Encapsulation

When several copies of a molecule interact and assemble, new behaviors emerge that are not apparent in the individual molecules but are unique to their assemblies. Predicting these behaviors is often as difficult as predicting the shape of a honeycomb by studying a single bee. Our research focuses on these multimolecular assemblies: how to make them, what forces hold them together, how fast they come together, what sequence of events leads to them, and what are the assemblies good for. These assemblies are dynamic, constantly forming and dissipating on timescales that range from milliseconds to hours, long enough for many types of molecular encounters and even chemical reactions to take place within the assemblies. Getting good pictures of these assemblies is like hitting a moving target.

Encapsulation

The assemblies we study have another feature: they form hollow, capsulelike structures in which smaller molecules are temporarily captured. The result is a single molecule that is temporarily isolated in space and time, completely surrounded by the capsule. This arrangement allows us to measure the behavior of individual molecules or single-molecule events. The capsules can be used as sensors, as reaction chambers, and as vehicles for delivery across membranes.

Synthesis of the capsular assemblies is constantly improving. The latest methods are inspired by biology in which a single structure can be used repeatedly in different settings to give a diverse set of structures. In pursuing this modular approach, we identified a glycoluril unit as this multipurpose module (Fig. 1). The unit has a wealth of hydrogen-bonding sites, and the hydrogen bonds hold the transitory assembly together. Because of its rigid skeleton, the module maintains a single, predictable shape. This shape has a gentle curvature similar to a part (roughly a sixth) of an eggshell.

We connect the various molecular scaffolds that position the modules for assembly and by piecing these scaffolds together, form the spherical shell of a capsule. In the capsule shown in Figure 1, a large cavity is created when the molecules assemble. The volume of the cavity is several hundred cubic angstroms, and large holes in the structure allow the rapid entrance and exit of small molecules. Large molecules enter and exit slowly, and we can observe them inside by using nuclear magnetic resonance spectroscopy. Overall, the capsules sort molecules according to size; that is, the capsules act as "molecular sieves."

In a second type of synthesis, the capsule is formed by folding a single very large molecule into the appropriate shape (Fig. 2). In this case, a bowl-shaped molecule, a calixarene, was used to provide molecular curvature, and 2 of these subunits were permanently linked. By coupling the 2 molecules, we reduced their independence and their entropies and improved their chances of folding properly. The capsule is formed when the 2 bowls come together, rim to rim. Urea groups arranged along the rim provided the hydrogen bonds in this example. Assembly of the capsule was enhanced, and the expected improvements in trapping small molecules inside were detected. However, some unexpected features emerged: certain of these molecules were able to form liquid crystals. Because of their ability to encapsulate small molecules in this physical state, the assembles may have unique applications as "smart" materials.

How much information or instruction is required to bring, for example, 4 individual molecules together into a capsule? This question is a different way of looking at entropy. We found that ureas and sulfonamides prefer hydrogen bonding to each other rather than to themselves, and we used this information as a form of instruction for assembly. A molecule was prepared with a urea at one end and a sulfonamide at the other (Fig. 3). As before, the curvature of the molecule, corresponding to a quarter of a sphere, was built into the glycoluril groups. The result was a predictable arrangement of 4 subunits, head to tail. Even so, the capsule emerged from its pieces only when the proper occupant was present to be trapped inside. These occupants fill about 55% of the space in the capsule. We discovered that this level of filling space is a powerful and perhaps universal driving force for molecular recognition and assembly.

In another study, we are exploring the synthesis and properties of "open-ended" molecular containers. We are preparing these containers with ever deepening cavities, and we found that the molecule shown in Figure 4 can temporarily trap smaller fullerenes such as C₆₀. We are beginning to find practical applications for these containers in molecular separations. Basic questions of how molecules get in and out of the container have been answered: the 4 sides must fold down before the resident occupant is displaced by the incoming one.

Molecular Diversity and Combinatorial Chemistry

We have synthesized structures in which a diversity of functional groups--acids and bases, hydrophobic and hydrophilic surfaces, large and small groups, and groups of all shapes--protrude from a single face of a molecular platform. One example of such a platform is the triphenylene structure shown in Figure 5. The ability of the platform to interact with small molecules comes from its 3-fold symmetry and rigidity. The triphenylene forms complexes with flat structures that just fit within it and have complementary chemical surfaces. Caffeine and nitrated aromatic molecules are complexed.

The ability of the triphenylene to interact with large molecular surfaces such as proteins is due to the large surface area spanned by the groups indicated as the spheres in Figure 5. We place these groups on the platform randomly and then screen them for a specific behavior. These new platforms represent uncharted areas in the landscape of molecular diversity.

Publications

Brody, M.S., Schalley, C.A., Rudkevich, D.M., Rebek, J., Jr. Synthesis and characterization of an intramolecularly self-assembled capsule. Angew. Chem. Int. Ed., in press.

Martín, T., Obst, U., Rebek, J., Jr. Hydrogen-bonding preferences and the filling of space provide information for molecular assembly and encapsulation. Science 281:1842, 1998.

Mink, D., Mecozzi, S., Rebek, J., Jr. Natural products analogs as scaffolds for supramolecular and combinatorial chemistry. Tetrahedron Lett. 39:5709, 1998.

Pryor, K.E., Rebek, J., Jr. Multifunctionalized glycolurils. Org. Lett. 1:39, 1999.

Schalley, C.A., Martín, T., Obst, U., Rebek, J., Jr. Characterization of encapsulation complexes in the gas phase and solution. J. Am. Chem. Soc. 121:2133, 1999.

Schalley, C.A., Rivera, J.M., Martín, T., Santamaria, J., Siuzdak, G., Rebek, J., Jr. Structural examination of supramolecular architectures by electrospray ionization mass spectrometry. Eur. J. Org. Chem. 6:1325, 1999.

Szabo, T., O'Leary, B., Rebek, J., Jr. Self-assembling sieves. Angew. Chem. Int. Ed. 37:3410, 1998.

Tucci, F.C., Rudkevich, D.M., Rebek, J., Jr. Deeper cavitands. J. Org. Chem. 64:4555, 1999.

Waldvogel, S.R., Wartini, A.R., Rasmussen, P.H., Rebek, J., Jr. A triphenylene scaffold with C_3v-symmetry and nanoscale dimensions. Tetrahedron Lett. 40:3515, 1999.

Evolution of Novel Enzymes

F.E. Romesberg, A.K. Ogawa, R. Jimenez, D.L. McMinn, Y. Wu, S. Roy, T. Sera, G. Xia, D. Hannay, E. Meggers, L. Chen, M. Berger, O. Abou-Zied, J. Chin, A. Henry

The focus of our laboratory is the development of novel methods to understand protein function and evolution. We are interested in developing new approaches for the study of function (protein dynamics) and new methods for expanding these functions to include new activities (in vitro directed evolution).

EVOLUTION OF NOVEL CATALYTIC FUNCTION

Information storage and replication play a central role in life. The "genetic alphabet" is composed of the letters G, C, A, and T, which designate the bases guanine, cytosine, adenine, and thymine, respectively. These letters (or bases) combine to form 2 base pairs as a result of specific patterns of hydrogen bonding. We are interested in expanding the genetic alphabet by developing a third base pair composed of unnatural nucleobases. We have been exploring hydrophobicity, as opposed to hydrogen bonding, as a driving force for base pairing.

We are developing phage display methods, as well as general high-throughput screening methods, for the in vitro evolution of DNA polymerases capable of efficient and high-fidelity synthesis of DNA that contains unnatural nucleobases. We are also developing phage display methods to evolve proteins with interesting dynamic properties. For example, the enzyme photolyase harvests the energy of visible light to catalyze repair of thymine dimers in DNA. We are developing methods to evolve photolyase proteins with novel properties, such as increased rates of energy transfer and electron transfer.

PROTEIN FUNCTION AND DYNAMICS

To understand protein dynamics, we are exploring the role of atomic motion, in addition to domain motion and conformational changes, in protein function and catalysis. Selective deuteration of proteins, at nonexchangeable positions, results in a carbon-deuterium oscillator that absorbs light in an otherwise transparent region of the infrared spectrum. This modification allows the selective deposition or measurement of energy in a specific and localized protein vibration. Time-resolved spectroscopy can then be used to determine the role of vibrational energy in protein function and catalysis.

The first step in either natural or in vitro evolution is the generation of a pool of diversity, which must be sufficient to drive evolution. We are interested in using photon echo spectroscopy to characterize the diversity of protein libraries. This technique allows the quantitative determination of the static inhomogeneity that contributes to an electronic or vibrational linewidth. This inhomogeneity may be related to functional binding-site diversity. We are characterizing natural immunologic libraries of antibodies and in vitro libraries of protein with bound cofactors.