Bacteriophage T4 gene 32 encodes a singlestranded DNA binding protein required for T4 DNA replication, recombination, and repair. Previous attempts at cloning gene 32 have failed due to a presumably deleterious effect on host cell viability. In addition, overexpression of gene 32 would be expected to be limited by the autoregulatory ability of the gene 32 product g32P. A repetitive A+T-rich sequence flanking the ribosome binding site of gene 32 has been implicated in this translational regulation. To circumvent these problems, the wild-type gene for g32P has been reconstructed in M13 using restriction fragments from T4 g32am453 and synthetic oligodeoxynucleotides so that it no longer includes its native promoter and putative autoregulatory region. The g32am453 codon TAG was changed back to TGG as in wild-type gene 32 using site-directed oligodeoxynucleotide mutagenesis. In vectors containing the X leftward promoter PL, gene 32 is overexpressed with the resulting transcripts being derepressed at g32P concentrations that repress the wild-type gene 32 transcripts.The Mr 34,387 bacteriophage T4 gene 32 product (g32P) has served as a prototype for a class of proteins that bind tightly and cooperatively to single-stranded (ss) DNA and that in vitro stimulate the activity of their cognate DNA polymerase. In vivo, g32P has been shown to be required in stoichiometric amounts for DNA replication, repair, and recombination (for review see ref. 1). Physicochemical studies on g32P and its complexes with ss DNA are beginning to provide insight into general mechanisms for the binding of proteins to singlestranded nucleic acids. In the case of g32P, both nitration (2) and 1H-NMR (3, 4) studies indicate that five tyrosine side chains are involved in binding to ss DNA. Since one phenylalanine and three tyrosine residues have been implicated (5) in the binding of the bacteriophage fd gene 5 protein to ss DNA, it appears that hydrophobic interactions between the side chains of aromatic amino acids and the bases of a polynucleotide may represent a common binding mechanism for several ss DNA binding proteins (4).To identify which tyrosine residues in the known g32P sequence (6) are at the interface of the g32P-ss DNA complex, we have cloned gene 32 so that it could be subjected to in vitro oligodeoxynucleotide site-directed mutagenesis. Site-specific mutagenesis provides the opportunity to sequentially substitute nonaromatic amino acids for each tyrosine residue in g32P. 1H-NMR and nucleic acid binding experiments on the resulting mutant proteins can then be used to determine the contribution of each of the tyrosine residues that had been replaced or, for that matter, any other g32P amino acids to the overall free energy of binding of the protein to ss DNA.Previous attempts to clone gene 32 have failed presumably due to a deleterious effect of g32P on host cell viability (7). Even if this problem were avoided by placing gene 32 under the control of an inducible promoter, in vitro studies indicate that overexpression of g32P w...