Search

Single-stranded DNA or RNA libraries used in SELEX experiments usually include primer-annealing sequences for PCR amplification. In genomic SELEX, these fixed sequences may form base pairs with the central genomic fragments and interfere with the binding of target molecules to the genomic sequences. In this study, a method has been developed to circumvent these artificial effects. Primer-annealing sequences are removed from the genomic library before selection with the target protein and are then regenerated to allow amplification of the selected genomic fragments. A key step in the regeneration of primer-annealing sequences is to employ thermal cycles of hybridization-extension, using the sequences from unselected pools as templates. The genomic library was derived from the bacteriophage fd, and the gene 5 protein (g5p) from the phage was used as a target protein. After four rounds of primer-free genomic SELEX, most cloned sequences overlapped at a segment within gene 6 of the viral genome. This sequence segment was pyrimidine-rich and contained no stable secondary structures. Compared with a neighboring genomic fragment, a representative sequence from the family of selected sequences had about 23-fold higher g5p-binding affinity. Results from primer-free genomic SELEX were compared with the results from two other genomic SELEX protocols.

The gene 5 protein (g5p) encoded by filamentous Ff phages is an ssDNA-binding protein, which binds to and sequesters the nascent ssDNA phage genome in the process of phage morphogenesis. The g5p also binds with high affinity to DNA and RNA sequences that form G-quadruplex structures. However, sequences that would form G-quadruplexes are absent in single copies of the phage genome. Using SELEX (systematic evolution of ligands by exponential enrichment), we have now identified a family of DNA hairpin structures to which g5p binds with high affinity. After eight rounds of selection from a library of 58-mers, 26 of 35 sequences of this family contained two regions of complete or partial complementarity. This family of DNA hairpins is represented by the sequence: 5'-d(CGGGATCCAACGTTTTCACCAGATCTACCTCCTCGGGATCCCAAGAGGCAGAATTCGC)-3' (named U-4), where complementary regions are italicized or underlined. Diethyl pyrocarbonate modification, UV-melting profiles, and BamH I digestion experiments revealed that the italicized sequences form an intramolecular hairpin, and the underlined sequences form intermolecular base pairs so that a dimer exists at higher oligomer concentrations. Gel shift assays and end boundary experiments demonstrated that g5p assembles on the hairpin of U-4 to give a discrete, intermediate complex prior to saturation of the oligomer at high g5p concentrations. Thus, biologically relevant sequences at which g5p initiates assembly might be typified better by DNA hairpins than by G-quadruplexes. Moreover, the finding that hairpins of U-4 can dimerize emphasizes the unexpected nature of sequence-dependent structures that can be recognized by the g5p ssDNA-binding protein.

We review CD studies of a single-stranded DNA binding protein, g5p, of the Ff group of bacterial viruses. The CD spectrum of the g5p is dominated by a positive tyrosine La band at 229 nm, to which all five of the protein tyrosines contribute. The La band becomes much less positive upon binding of g5p to nucleic acids. CD spectra of mutant proteins identified a single tyrosine, Y34, that is largely responsible for this CD perturbation. At >250 nm, CD perturbations of nucleic acids can be monitored during g5p binding, and CD titrations have identified two distinct modes of binding of the g5p at physiological ionic strength (0.2 M NaCl). SELEX selection of sequences bound preferentially by g5p yielded a G-rich sequence that is closely related to telomere sequences and has CD properties of a G-tetraplex. CD spectroscopy showed that the presumed G-quadruplex form is maintained within saturated g5p x DNA complexes.