DNA display II. Genetic manipulation of combinatorial chemistry libraries for small-molecule evolution

Abstract

Biological in vitro selection techniques, such as RNA aptamer methods and mRNA display, have proven to be powerful approaches for engineering molecules with novel functions. These techniques are based on iterative amplification of biopolymer libraries, interposed by selection for a desired functional property. Rare, promising compounds are enriched over multiple generations of a constantly replicating molecular population, and subsequently identified. The restriction of such methods to DNA, RNA, and polypeptides precludes their use for small-molecule discovery. To overcome this limitation, we have directed the synthesis of combinatorial chemistry libraries with DNA "genes," making possible iterative amplification of a nonbiological molecular species. By differential hybridization during the course of a traditional split-and-pool combinatorial synthesis, the DNA sequence of each gene is read out and translated into a unique small-molecule structure. This "chemical translation" provides practical access to synthetic compound populations 1 million-fold more complex than state-of-the-art combinatorial libraries. We carried out an in vitro selection experiment (iterated chemical translation, selection, and amplification) on a library of 10(6) nonnatural peptides. The library converged over three generations to a high-affinity protein ligand. The ability to genetically encode diverse classes of synthetic transformations enables the in vitro selection and potential evolution of an essentially limitless collection of compound families, opening new avenues to drug discovery, catalyst design, and the development of a materials science "biology."

Figure 1. The In Vitro Selection Cycle

Experiments are initiated with a nucleic acid library (colored DNA). The sequence of each gene directs the synthesis of a corresponding gene product (colored ball) that is physically linked to its encoding nucleic acid. The gene products are subjected to selection, for example, through binding to an immobilized macromolecule (cyan widget at bottom). The nucleic acid encoding selected gene products is amplified and used as input for a subsequent cycle.

Figure 2. Split-and-Pool Synthesis of a Combinatorial Chemistry Library

A mixture of solid supports (balls with rotated “L” at top) is randomly split into subpools. A distinct chemical building block (red, green, or blue ball) is coupled to the supports in each subpool. The supports are repooled and mixed. This process of splitting, chemistry, and pooling is iterated until the library synthesis is complete. The small molecules ultimately synthesized are combinations of the different building blocks (colored circles, squares, and diamonds). As highlighted by the black bead, the path taken by a support through the split-and-pool synthesis (right, middle, left) determines the small molecule synthesized on it (blue ball, green square, red diamond). The number of reactions performed is the sum of the number of subpools in each split (3 + 3 + 3 = 9). The number of unique small molecules generated is the product of the number of subpools in each split (3 × 3 × 3 = 27).

Figure 3. Chemical Translation

(A) Schematic showing the structure of the DNA support library. Small molecules are synthesized at the 5′ end of 340-base ssDNA genes. The ssDNA consists of 20-base noncoding regions (black lines labeled Z₁–Z₇) and 20-base coding positions (colored bars labeled [a–j]_1–6). All library members contain the same seven DNA sequences at the seven noncoding regions. At each of the six coding positions, ten mutually exclusive DNA codons, (a–j)_n, are present, for a total of 60 different sequences. Each coding region specifies the addition of a single subunit to a growing small molecule. A unique reactive site (in this case a primary amine) for small-molecule synthesis is attached to the 5′ end of the ssDNA through a polyethylene glycol linker (squiggly line). Resin beads coated with an oligonucleotide complementary to one codon (anticodon beads, gray ball at right) capture by hybridization ssDNAs containing the corresponding codon. (B) Chemical translation is a split-and-pool synthesis, with splitting directed by DNA hybridization. A ssDNA library is hybridized to a set of anticodon columns (gray balls) corresponding to the set of codons present at a single coding position. The ssDNA genes partition into subpools based on sequence identity. Distinct chemical subunits (colored balls) are coupled to the DNA in each subpool. Finally, the DNA is repooled, completing the encoded addition of one subunit to the growing small molecule. The process of hybridization splitting, chemistry, and pooling is repeated for all subsequent coding regions. (C) Schematic product of chemical translation. The sequence of the small-molecule subunits (colored balls) corresponds to the sequence of codons (colored bars) in the ssDNA gene.

Figure 4. Sequence-Directed Splitting

Seven serially truncated ssDNAs differing in sequence at one coding position (illustrated at left of gel, number of bases indicated) were hybridized to seven anticodon columns (cylinders at top of gel). The load (lane 1), flow through (lane 2), and column elutes (lanes 3–9) were analyzed by denaturing polyacrylamide gel electrophoresis.

Figure 5. Peptide Synthesis on DNA

(A) Structure of the [Leu]enkephalin–DNA conjugate. (B) High performance liquid chromatography chromatogram of the [Leu]enkephalin peptide synthesized using succinimidyl ester chemistry on a 20-base oligonucleotide modified with a 5′ primary amine (20mer). A 10-base oligonucleotide without the 5′ primary amine (10mer) was included in the reactions as a control for nonspecific DNA modification. The red and blue traces are the DNA before and after chemistry, respectively. The mass of the major product peak (42-min retention time) matches the expected mass of the [Leu]enkephalin–DNA conjugate. (C) Electromobility shift assay of peptides synthesized on 340-base ssDNA. Conjugates were eletrophoresed on a native agarose gel in the absence (lanes 1, 3, 5, and 7) or presence (lanes 2, 4, 6, 8, and 9) of the [Leu]enkephalin-binding antibody 3-E7. [Leu]enkephalin (L) or a scrambled sequence (S) was synthesized on a 5′ amino-modified 20-base oligonucleotide, which was subsequently used as a primer for PCR (lanes 1–4), or directly on 5′ amino-modified 340-base ssDNA, which was subsequently converted to dsDNA (lanes 5–9). Addition of free [Leu]enkephalin peptide (lane 9) competes away binding.

Figure 6. Reduction to Practice

Chemical translation requires iteration of a chemistry step and two column-transfer steps. ssDNA is transferred from anticodon columns to DEAE Sepharose columns by cyclically pumping 50% DMF through a pair of columns (one hybridization, one DEAE) attached in series for 1 h at 45 °C. Chemistry is performed on ssDNA bound to each DEAE column. ssDNA is transferred from DEAE columns to anticodon columns by cyclically pumping a 1.5-M NaCl buffer through all DEAE columns and all anticodon columns associated with the next coding position for 1 h at 70 °C and 1 h at 46 °C. Efficiencies for each step are indicated in red.

Figure 7. In Vitro Selection of a Nonnatural Peptide Library

(A) Library building blocks. Proteinogenic building blocks are shown in green. (B) Approximately 70 DNA genes from each round of selection were sequenced, and the results are summarized as a histogram plot. The x-axis indicates the number of amino acid residue matches to [Leu]enkephalin encoded by a library sequence. The y-axis indicates the library generation (0, starting material; 1, after round one selection; 2, after round two selection). The z-axis indicates the number of sequences encoding a particular number of matches (x-axis) in a particular round (y-axis). (C) The top row reports the round two library consensus sequence, which matches [Leu]enkephalin. The second row reports the percentage of round two library clones that encode the [Leu]enkephalin amino acid at each residue position. The third row reports the identity and frequency of the most commonly occurring non-[Leu]enkephalin subunit at each position.