About TSRI
Research & Faculty
News & Publications
Scientific Calendars
Scripps Florida
PhD Program
Campus Services
Work at TSRI
TSRI in the Community
Giving to TSRI
Site Map & Search

Scientific Report 2004


Design of Functional Synthetic Systems

M.R. Ghadiri, M. Al-Sayah, G. Ashkenasy, N. Ashkenasy, A. Chavochi, J. Fletcher, V. Haridas, W.S. Horne, Z.-Z. Huang, P. Imming, L. Leman, A. Loutchnikov, L. Motiei, Y. Norikane, N. Rahe, S. Rahimipour, J. Shin, J. van Maarseveen, D. Vodak, M. Yadav, R. Yamasaki

We are engaged in a multidisciplinary research effort to uncover new chemical and biochemical approaches for the design of functional molecular, supramolecular, and complex self-organized systems. Our endeavors span disciplines ranging from synthetic organic, bioorganic, and physical organic chemistry to nanotechnology, biophysics, enzymology, and molecular biology. Current research includes the design of synthetic peptide receptors and catalysts, antimicrobial self-assembling peptide nanotubes, semisynthetic allosteric enzymes, self-replicating molecular systems and emergent networks, single-molecule stochastic DNA sensing, and DNA-based logic gates and molecular computation.

Synthetic Peptide Catalysts

Rational design of biomolecular catalysts based on first principles is fundamentally valuable for assessing and advancing basic understanding of factors that contribute to the functioning of natural enzymes. Such endeavors also provide blueprints for fabrication of novel synthetic catalysts with tailored functional properties. In particular, we are interested in the aminoacyl transfer class of reactions because of their prominent roles in a number of chemical and biological processes. Our aim is to mimic the sequential aminoacyl transferase activity of nonribosomal peptide synthetases. Recently, we designed a modular peptide construct capable of catalyzing stoichiometric and site-specific aminoacyl transfer between noncovalently associated α-helical peptides with catalytic efficiencies on the order of 105 in neutral aqueous solutions (Fig. 1).

Fig. 1. Aminoacyl transferase reaction cycle of an engineered peptide catalyst. The process consists of the aminoacyl loading, rapid intersubunit acyl transfer, and aminoacyl reloading at the active site.

Antimicrobial Peptide Nanotubes

We showed that appropriately designed cyclic peptide subunits can self-assemble through hydrogen bond–directed ring stacking into open-ended hollow tubular structures (Fig. 2) that have marked antibacterial and antiviral activities in vitro. The effectiveness of this novel supramolecular class of bioactive species as selective antibacterial agents was recently highlighted by the high efficacy of one of these antimicrobials against lethal methicillin-resistant Staphylococcus aureus infections in mice. Currently, we are using rational design and selections from combinatorial cyclic peptide libraries to discover cell-specific targeting sequences and anticancer compounds.

Fig. 2. Self-assembling cyclic peptide nanotubes constitute a new class of a supramolecular approach to drug design. The illustration represents selective insertion of cyclic D,L-α-peptide nanotubes into the cell membranes of pathogens.

Design of Signal Self-Amplifying DNA Sensors

We constructed a novel sequence-specific DNA detection system (Fig. 3) based on rationally designed semisynthetic enzymes. The system is composed of covalently associated inhibitor-DNA-enzyme modules that function via DNA hybridization–triggered allosteric enzyme activation and signal amplification through substrate turnover. The functional capacity of the system is highlighted by the sequence-specific detection of approximately 10 fmol of DNA in less than 3 minutes under physiologic conditions. Our studies suggest that rationally designed intrasterically regulated enzymes may be a promising new class of reagents for highly sensitive, rapid 1-step detection of label-free DNA sequences that does not depend on polymerase chain reactions.

Fig. 3. Schematic representation of an intrasterically inactivated inhibitor-DNA-enzyme construct (left) and the DNA hybridization–triggered enzyme activation (right). The construct can be used to sense low concentrations of complementary DNA because of its built-in capacity for signal amplification via rapid substrate turnover.

Stochastic Analysis of Single-Molecule DNA Rotaxanes

We have a growing interest in the study of matter at the single-molecule level. For these studies, we use the transmembrane protein α-hemolysin as a rapid and highly sensitive sensor element for stochastic analysis of the molecules lodged or trapped inside the protein pore; the analysis relies on detecting the perturbations in conductance levels produced in the ion channel in the native protein. Using this technique, we recently developed an approach by which a single-stranded DNA molecule can be trapped in a specific configuration inside an α-hemolysin channel (Fig. 4), manipulated, and studied with high sensitivity at the single-molecule level. We are extending this approach to the design of rapid single-molecule DNA sensing and sequencing.

Fig. 4. Trapping a single DNA strand inside the pore of the transmembrane protein α-hemolysin in the form of a rotaxane. The identity of the nucleobase residing in the channel can be sensed by analyzing the interactions of the DNA with the protein channel residues near the constriction zone.

Dna-based Logic Gates and Molecular Computation

Semiconductor-based logic gates provide the fundamental computing devices of all microprocessors in use today. A logic gate functions by providing a specific output that depends on the inputs it receives. Various logic gates (or logic operations) can be coupled (wired) in a network to provide the higher order functions of a computer. Similarly, molecular analogs of logic gates may provide miniaturized and highly efficient parallel computing at the level of single molecules. However, to date, mimicking electronic computational processes by using purely molecular systems has been limited in most instances by the lack of design generality and/or potential addressability of the molecular logic gates.

We use the universal recognition properties of DNA to create addressable logic gates that are capable of AND, OR, XOR, NOR, NAND, and INHIBIT logic operations (Fig. 5). We used DNA-based logic gates to design computational circuits capable of performing simple arithmetic operations. Together these recent findings suggest the usefulness of DNA-based logic gates as a novel and effective supramolecular platform for the fabrication of molecular computational devices.

Fig. 5. Arithmetic operations can be conducted with molecules in solution. A half adder is shown using a network of AND and XOR DNA-based logic gates.


Horne, W.S., Stout, C.D., Ghadiri, M.R. A heterocyclic peptide nanotube. J. Am. Chem. Soc. 125:9372, 2003.


M. Reza Ghadiri, Ph.D.

Ghadiri Web Site