The integrase family of site-specific recombinases: regional similarities and global diversity.

A combination of two methods for detecting distant relationships in protein primary sequences was used to compare the site-specific recombination proteins encoded by bacteriophage lambda, phi 80, P22, P2, 186, P4 and P1. This group of proteins exhibits an unexpectedly large diversity of sequences. Despite this diversity, all of the recombinases can be aligned in their C-terminal halves. A 40-residue region near the C terminus is particularly well conserved in all the proteins and is homologous to a region near the C terminus of the yeast 2 mu plasmid Flp protein. This family of recombinases does not appear to be related to any other site-specific recombinases. Three positions are perfectly conserved within this family: histidine, arginine and tyrosine are found at respective alignment positions 396, 399 and 433 within the well-conserved C-terminal region. We speculate that these residues contribute to the active site of this family of recombinases, and suggest that tyrosine-433 forms a transient covalent linkage to DNA during strand cleavage and rejoining.     

Sequence analysis of an artificial family of RNA-binding peptides

Diverse peptide sequences recognizing the lambda boxB RNA hairpin were previously isolated from a library encoding the 22-residue lambda N peptide with random amino acids at positions 13-22 using mRNA display. We have statistically analyzed amino acid distributions in 65 unique sequences from rounds 11 and 12 of this selection and evaluated the resulting structural and functional predictions by alanine-scanning mutagenesis and circular dichroism spectrometry. This artificial sequence family has a consensus structure that continues the bent alpha helix of lambda N up to position 17 when bound to lambda boxB. A charge pair (E(14)R(15)) and hydrophobic patch (A(21)L(22) or V(21)L(22)) have important functional roles in this context. Notably, amino acid covariance reveals six specific pairs of random region positions with >95% significant linkage and strong overall helical (i+1, i+3, and i+4) couplings. The covariance analysis suggests that (1) the sequence context of every residue in each insert has been optimized, (2) selected sequences are local optima on a rugged fitness landscape, and (3) it is possible to detect more subtle structural features with artificial protein sequence families than natural homologs. Our results provide a framework for investigating the structures of in vitro selected proteins by functional minimization, reselection, and covariance analysis.     

The phi 80 and P22 attachment sites. Primary structure and interaction with Escherichia coli integration host factor

Although the lambdoid bacteriophage phi 80 and P22 possess site-specific recombination systems analogous to bacteriophage lambda, they have different attachment (att) site specificities. We have identified and determined the nucleotide sequences of the att sites of phi 80 and P22 and have examined the interaction of these sites with purified Escherichia coli integration host factor (IHF). The sizes of the homologous core regions of the att sites vary greatly: P22 has a 46-base pair core, while phi 80 and lambda have 17- and 15-base pair cores, respectively. The core sequences of the three phage show no significant homology, although dispersed regions of homology in arm sequences indicate that the three phage att sites are related. All three att sites have a high A + T composition, and restriction fragments carrying these sites migrate anomalously upon polyacrylamide gel electrophoresis. IHF binds to a site to the left of the common core in the phi 80 and P22 phage att sites (attP) and to a site to the right of the core in P22 attP and attB (the bacterial att site). In the lambda system, IHF interacts with three regions on attP (designated H1, H2, and H') and none on attB (Craig N., and Nash, H.A. (1984) Cell 39, 707-716). Alignment of the IHF sites of all three phage results in a consensus sequence for IHF binding, Pyr-AANNNNTTGATAT. Among the three phage, the number of IHF sites differs; however, the location and orientation of the binding sites in relation to the respective core regions are well conserved. An IHF site analogous to lambda H2 is present in both phi 80 and P22 attP, while a site analogous to lambda H' is present in P22 attP. This conservation suggests that IHF plays a very similar role in the site-specific recombination pathways of all three phage, and that the flanking arm sequences are necessary for phi 80 and P22 attP function, as is the case for lambda attP function. These structural similarities presumably reflect a conservation of the mechanism of site-specific recombination for the three phage.     

The carboxy-terminal 14 amino acids of phage lambda N protein are dispensable for transcription antitermination.

The analogous N proteins encoded by lambdoid bacteriophages lambda, 21, and 22 are very different in amino acid sequence, except at their carboxy-terminal ends. Since N lambda remains functional despite the deletion of most of its terminal region of homology to N21, that region of homology cannot represent a region of conserved function.     

Cleavage of bacteriophage phi 80 CI repressor by RecA protein

We have purified the CI repressor protein of bacteriophage phi 80. Its N-terminal amino acid sequence and its amino acid composition agree with those predicted from the nucleotide sequence of the cI gene. The phi 80 CI repressor was cleaved at a Cys-Gly bond by the wildtype RecA protein in the presence of single-stranded DNA and ATP or its analogues. This cleavage site is different from other repressors such as LexA, lambda CI and P22 C2, which were cleaved at an Ala-Gly bond. The phi 80 CI repressor was cleaved at the same site by the RecA430 protein, but was not cleaved by the RecA1 protein. This effect of the bacterial recA mutations on cleavage is consistent with the fact that prophage phi 80 in recA430 cells can be induced by irradiation with ultraviolet light, while the prophage in recA1 cells cannot.     

The two major membrane skeletal proteins (articulins) of Euglena gracilis define a novel class of cytoskeletal proteins.

60% of the peripheral membrane skeleton of Euglena gracilis consists of equimolar amounts of two proteins (articulins) with M(r)s in SDS gels of 80 and 86 kD. To understand eventually how these proteins assemble and function in maintaining cell form and membrane integrity we have undertaken a molecular characterization of articulins. A lambda gt11 expression library constructed from Euglena gracilis mRNAs was screened with antibodies against both articulins. Two sets of cDNAs were recovered, and evidence from three independent assays confirmed that both sets encoded articulins: (a) Anti-articulin antibodies recognized a high molecular weight beta-galactosidase (beta-gal) fusion protein expressed in bacteria infected with lambda gt11 cDNA clones. (b) Antibodies generated against the bacterially expressed beta-gal fusion protein identified one or the other articulin in Western blots of Euglena proteins. These antibodies also localized to the membrane skeletal region in thin sections of Euglena. (c) Peptide maps of the beta-gal fusion protein were similar to peptide maps of Euglena articulins. From the nucleotide sequence of the two sets of cDNAs an open reading frame for each articulin was deduced. In addition to 37% amino acid identity and overall structural similarity, both articulins exhibited a long core domain consisting of over 30 12-amino acid repeats with the consensus VPVPV--V--. Homology plots comparing the same or different articulins revealed larger, less regular repeats in the core domain that coincided with predicted turns in extended beta-sheets. Outside the core domain a short hydrophobic region containing four seven-amino acid repeats (consensus: APVTYGA) was identified near the carboxy terminus of the 80-kD articulin, but near the amino terminus of the 86-kD articulin. No extensive sequence similarities were found between articulins and other protein sequences in various databanks. We conclude that the two articulins are related members of a new class of membrane cytoskeletal proteins.     

Achieving specificity in selected and wild-type N peptide-RNA complexes: the importance of discrimination against noncognate RNA targets

The boxB RNA pentaloops from the P22 and lambda phages each adopt a GNRA tetraloop fold upon binding their cognate arginine-rich N peptides. The third loop base in P22 boxB (3-out) and the fourth in lambda boxB (4-out) are excluded to accommodate this structure. Previously, we selected a pool of lambda N sequences with random amino acids at loop contacting positions 13-22 for binding to either of these two GNRA-folded pentaloops or a canonical GNRA tetraloop and isolated a class of peptides with a new conserved arginine (R15). Here, we characterize the binding of lambda N and these R15 peptides using fluorescent titrations with 2-aminopurine labeled versions of the three GNRA-folded loops and circular dichroism spectrometry. All peptides preferentially bind the lambda boxB RNA loop. lambda N and R15 peptide specificity against the P22 loop arises from the cost of rearranging its loop into the 4-out GNRA structure. Modeling indicates that the interaction of R8 with an additional loop phosphate in the 4-out GNRA pentaloop selectively stabilizes this complex relative to the tetraloop. R15 peptides gain additional discrimination against the tetraloop because their arginine also preferentially interacts with the 4-out GNRA pentaloop phosphate backbone, whereas K14 and W18 of lambda N contribute equal affinity when binding the tetraloop. Nonspecific electrostatic interactions by basic residues near the C-termini of these peptides create significantly steeper salt dependencies in association constants for noncognate loops, aiding discrimination at high salt concentrations. Our results emphasize the importance of considering specificity against noncognate as well as nonspecific targets in the combinatorial and rational design of biopolymers capable of macromolecular recognition.     

Organization of the early region of bacteriophage phi 80. Genes and proteins

An EcoRI segment containing the early region of bacteriophage phi 80 DNA that controls immunity and lytic growth was identified as a segment whose presence on a plasmid prevented growth of infecting phi 80cI phage. The nucleotide sequence of the segment (EcoRI-F) and adjacent regions was determined. Based on the positions of amber mutations and the sizes of some gene products, the reading frames for five genes were identified. From the relative locations of these genes in the genome, the properties of some isolated gene products, and the analysis of the structures of predicted proteins, the following phi 80 to lambda analogies are deduced: genes cI and cII to their lambda namesakes; gene 30 to cro; gene 15 to O; and gene 14 to P. An amber mutation by which gene 16 was defined is a nonsense mutation in the frame for gene 15 protein, excluding the presence of gene 16. An amber mutation in gene 14 or 15 inhibits phage DNA synthesis, as is the case with their lambda analogues, gene O or P. Some characteristics of proteins from the early region predicted from their primary structures and their possible functions are discussed.