Assessing the plasticity of DNA target site recognition of the PI-SceI homing endonuclease using a bacterial two-hybrid selection system

The PI-SceI protein from Saccharomyces cerevisiae is a member of the LAGLIDADG family of homing endonucleases that have been used in genomic engineering. To assess the flexibility of the PI-SceI-binding interaction and to make progress towards the directed evolution of homing endonucleases that cleave specified DNA targets, we applied a two-hybrid method to select PI-SceI variants from a randomized expression library that bind to different DNA substrates. In particular, the codon for Arg94, which is located in the protein splicing domain and makes essential contacts to two adjacent base-pairs, and the codons for four proximal residues were randomized. There is little conservation of the wild-type amino acid residues at the five randomized positions in the variants that were selected to bind to the wild-type site, yet one of the purified derivatives displays DNA-binding specificity and DNA endonuclease activity that is similar to that of the wild-type enzyme. A spectrum of DNA-binding behaviors ranging from partial relaxation of specificity to marked shifts in target site recognition are present in variants selected to bind to sites containing mutations at the two base-pairs. Our results illustrate the inherent plasticity of the PI-SceI/DNA interface and demonstrate that selection based on DNA binding is an effective means of altering the DNA cleavage specificity of homing endonucleases. Furthermore, it is apparent that homing endonuclease target specificity derives, in part, from constraints on the flexibility of DNA contacts imposed by hydrogen bonds to proximal residues.     

Isolation and characterization of new homing endonuclease specificities at individual target site positions

Homing endonucleases are highly specific DNA endonucleases, encoded within mobile introns or inteins, that induce targeted recombination, double-strand repair and gene conversion of their cognate target sites. Due to their biological function and high level of target specificity, these enzymes are under intense investigation as tools for gene targeting. These studies require that naturally occurring enzymes be redesigned to recognize novel target sites. Here, we report studies in which the homodimeric LAGLIDADG homing endonuclease I-CreI is altered at individual side-chains corresponding to contact points to distinct base-pairs in its target site. The resulting enzyme constructs drive specific elimination of selected DNA targets in vivo and display shifted specificities of DNA binding and cleavage in vitro. Crystal structures of two of these constructs demonstrate that substitution of individual side-chain/DNA contact patterns can occur with almost no structural deformation or rearrangement of the surrounding complex, facilitating an isolated, modular redesign strategy for homing endonuclease activity and specificity.     

DNA binding and cleavage by the HNH homing endonuclease I-HmuI

The structure of I-HmuI, which represents the last family of homing endonucleases without a defining crystallographic structure, has been determined in complex with its DNA target. A series of diverse protein structural domains and motifs, contacting sequential stretches of nucleotide bases, are distributed along the DNA target. I-HmuI contains an N-terminal domain with a DNA-binding surface found in the I-PpoI homing endonuclease and an associated HNH/N active site found in the bacterial colicins, and a C-terminal DNA-binding domain previously observed in the I-TevI homing endonuclease. The combination and exchange of these features between protein families indicates that the genetic mobility associated with homing endonucleases extends to the level of independent structural domains. I-HmuI provides an unambiguous structural connection between the His-Cys box endonucleases and the bacterial colicins, supporting the hypothesis that these enzymes diverged from a common ancestral nuclease.     

The crystal structure of the gene targeting homing endonuclease I-SceI reveals the origins of its target site specificity

The I-SceI homing endonuclease enhances gene targeting by introducing double-strand breaks at specific chromosomal loci, thereby increasing the recombination frequency. Here, we report the crystal structure of the enzyme complexed to its DNA substrate and Ca(2+) determined at 2.25A resolution. The structure shows the prototypical beta-saddle of LAGLIDADG homing endonucleases that is contributed by two pseudo-symmetric domains. The high specificity of I-SceI is explained by the large number of protein-DNA contacts, many that are made by a long beta-hairpin loop that reaches into the major groove of the DNA. The DNA minor groove is compressed at the catalytic center, bringing the two scissile phosphodiester bonds into close proximity. The protein-Ca(2+)-DNA structure shows the protein bound to its DNA substrate in a pre-reactive state that is defined by the presence of two asymmetric active sites, one of which appears poised to first cleave the DNA bottom strand.     

Evolution of I-SceI homing endonucleases with increased DNA recognition site specificity

Elucidating how homing endonucleases undergo changes in recognition site specificity will facilitate efforts to engineer proteins for gene therapy applications. I-SceI is a monomeric homing endonuclease that recognizes and cleaves within an 18-bp target. It tolerates limited degeneracy in its target sequence, including substitution of a C:G₊₄ base pair for the wild-type A:T₊₄ base pair. Libraries encoding randomized amino acids at I-SceI residue positions that contact or are proximal to A:T₊₄ were used in conjunction with a bacterial one-hybrid system to select I-SceI derivatives that bind to recognition sites containing either the A:T₊₄ or the C:G₊₄ base pairs. As expected, isolates encoding wild-type residues at the randomized positions were selected using either target sequence. All I-SceI proteins isolated using the C:G₊₄ recognition site included small side-chain substitutions at G100 and either contained (K86R/G100T, K86R/G100S and K86R/G100C) or lacked (G100A, G100T) a K86R substitution. Interestingly, the binding affinities of the selected variants for the wild-type A:T₊₄ target are 4- to 11-fold lower than that of wild-type I-SceI, whereas those for the C:G₊₄ target are similar. The increased specificity of the mutant proteins is also evident in binding experiments in vivo. These differences in binding affinities account for the observed ∼36-fold difference in target preference between the K86R/G100T and wild-type proteins in DNA cleavage assays. An X-ray crystal structure of the K86R/G100T mutant protein bound to a DNA duplex containing the C:G₊₄ substitution suggests how sequence specificity of a homing enzyme can increase. This biochemical and structural analysis defines one pathway by which site specificity is augmented for a homing endonuclease.     

The C-terminal loop of the homing endonuclease I-CreI is essential for site recognition, DNA binding and cleavage

Meganucleases are sequence-specific endonucleases with large cleavage sites that can be used to induce efficient homologous gene targeting in cultured cells and plants. These enzymes open novel perspectives for genome engineering in a wide range of fields, including gene therapy. A new crystal structure of the I-CreI dimer without DNA has allowed the comparison with the DNA-bound protein. The C-terminal loop displays a different conformation, which suggests its implication in DNA binding. A site-directed mutagenesis study in this region demonstrates that whereas the C-terminal helix is negligible for DNA binding, the final C-terminal loop is essential in DNA binding and cleavage. We have identified two regions that comprise the Ser138–Lys139 and Lys142–Thr143 pairs whose double mutation affect DNA binding in vitro and abolish cleavage in vivo. However, the mutation of only one residue in these sites allows DNA binding in vitro and cleavage in vivo. These findings demonstrate that the C-terminal loop of I-CreI endonuclease plays a fundamental role in its catalytic mechanism and suggest this novel site as a region to take into account for engineering new endonucleases with tailored specificity.     

Thermodynamics of DNA target site recognition by homing endonucleases

The thermodynamic profiles of target site recognition have been surveyed for homing endonucleases from various structural families. Similar to DNA-binding proteins that recognize shorter target sites, homing endonucleases display a narrow range of binding free energies and affinities, mediated by structural interactions that balance the magnitude of enthalpic and entropic forces. While the balance of ΔH and TΔS are not strongly correlated with the overall extent of DNA bending, unfavorable ΔH_binding is associated with unstacking of individual base steps in the target site. The effects of deleterious basepair substitutions in the optimal target sites of two LAGLIDADG homing endonucleases, and the subsequent effect of redesigning one of those endonucleases to accommodate that DNA sequence change, were also measured. The substitution of base-specific hydrogen bonds in a wild-type endonuclease/DNA complex with hydrophobic van der Waals contacts in a redesigned complex reduced the ability to discriminate between sites, due to nonspecific ΔS_binding.     

Creation of an artificial bifunctional intein by grafting a homing endonuclease into a mini-intein

The majority of inteins are comprised of a protein splicing domain and a homing endonuclease domain. Experimental evidence has demonstrated that the splicing domain and the endonuclease domain in a bifunctional intein are largely independent of each other with respect to both structure and activity. Here, an artificial bifunctional intein has been created through the insertion of an existing homing endonuclease into a mini-intein that is naturally lacking this functionality. The gene for I-CreI, an intron-encoded homing endonuclease, was grafted into the monofunctional Mycobacterium xenopi GyrA intein at the putative site of the missing endonuclease. The resulting fusion protein was found to be capable of protein splicing similar to that of the parent intein. In addition, the protein demonstrated site-specific endonuclease activity that is characteristic of the I-CreI homing endonuclease. The function of each domain therefore remained unaffected by the presence of the other domain. This artificial fusion of the two domains is a potential novel mobile genetic element.     

I-BasI and I-HmuI: two phage intron-encoded endonucleases with homologous DNA recognition sequences but distinct DNA specificities

I-HmuI and I-BasI are two highly similar nicking DNA endonucleases, which are each encoded by a group I intron inserted into homologous sites within the DNA polymerase genes of Bacillus phages SPO1 and Bastille, respectively. Here, we present a comparison of the DNA specificities and cleavage activities of these enconucleases with homologous target sites. I-BasI has properties that are typical of homing endonucleases, nicking the intron-minus polymerase genes in either host genome, three nucleotides downstream of the intron insertion site. In contrast, I-HmuI nicks both the intron-plus and intron-minus site in its own host genome, but does not act on the target from Bastille phage. Although the enzymes have distinct DNA substrate specificities, both bind to an identical 25bp region of their respective intron-minus DNA polymerase genes surrounding the intron insertion site. The endonucleases appear to interact with the DNA substrates in the downstream exon 2 in a similar manner. However, whereas I-HmuI is known to make its only base-specific contacts within this exon region, structural modeling analyses predict that I-BasI might make specific base contacts both upstream and downstream of the site of intron insertion. The predicted requirement for base-specific contacts in exon 1 for cleavage by I-BasI was confirmed experimentally. This explains the difference in substrate specificities between the two enzymes, including the observation that the former enzyme is relatively insensitive to the presence of an intron upstream of exon 2. These differences are likely a consequence of divergent evolutionary constraints.     

SegG endonuclease promotes marker exclusion and mediates co-conversion from a distant cleavage site

Bacteriophages T2 and T4 are closely related T-even phages. However, T4 genetic markers predominate in the progeny of mixed infections, a phenomenon termed marker exclusion. One region previously mapped where the frequency of T2 markers in the progeny is extremely low is located around gene 32. Here, we describe SegG, a GIY-YIG family endonuclease adjacent to gene 32 of phage T4 that is absent from phage T2. In co-infections with T2 and T4, cleavage in T2 gene 32 by T4-encoded SegG initiates a gene conversion event that results in replacement of T2 gene 32 markers with the corresponding T4 sequence. Interestingly, segG inheritance is limited, apparently because of the physical separation of its cleavage and insertion sites, which are 332 base-pairs apart. This contrasts with efficient inheritance of the phage T4 td group I intron and its endonuclease, I-TevI, for which the distance separating the I-TevI cleavage site and td insertion site is 23 base-pairs. Furthermore, we show that co-conversion tracts generated by repair of SegG and I-TevI double-strand breaks contribute to the localized exclusion of T2 markers. Our results demonstrate that the endonuclease activities of SegG and I-TevI promote the spread of these two endonucleases to progeny phage, consistent with their role as selfish genetic elements, and also provide a mechanism by which the genetic contribution of T2 markers to progeny phage is reduced.     

Structure of a hyperthermophilic archaeal homing endonuclease, I-Tsp061I: contribution of cross-domain polar networks to thermostability

A novel LAGLIDADG-type homing endonuclease (HEase), I-Tsp061I, from the hyperthermophilic archaeon Thermoproteus sp. IC-061 16 S rRNA gene (rDNA) intron was characterized with respect to its structure, catalytic properties and thermostability. It was found that I-Tsp061I is a HEase isoschizomer of the previously described I-PogI and exhibits the highest thermostability among the known LAGLIDADG-type HEases. Determination of the crystal structure of I-Tsp061I at 2.1 A resolution using the multiple isomorphous replacement and anomalous scattering method revealed that the overall fold is similar to that of other known LAGLIDADG-type HEases, despite little sequence similarity between I-Tsp061I and those HEases. However, I-Tsp061I contains important cross-domain polar networks, unlike its mesophilic counterparts. Notably, the polar network Tyr6-Asp104-His180-107O-HOH12-104O-Asn177 exists across the two packed alpha-helices containing both the LAGLIDADG catalytic motif and the GxxxG hydrophobic helix bundle motif. Another important structural feature is the salt-bridge network Asp29-Arg31-Glu182 across N and C-terminal domain interface, which appears to contribute to the stability of the domain/domain packing. On the basis of these structural analyses and extensive mutational studies, we conclude that such cross-domain polar networks play key roles in stabilizing the catalytic center and domain packing, and underlie the hyperthermostability of I-Tsp061I.     

Profile of the DNA recognition site of the archaeal homing endonuclease I-DmoI.

I- Dmo I is a homing enzyme of the LAGLI-DADG type that recognizes up to 20 bp of DNA and is encoded by an archaeal intron of the hyperthermophilic archaeon Desulfurococcus mobilis . A combined mutational and DNA footprinting approach was employed to investigate the specificity of the I- Dmo I-substrate interaction. The results indicate that the enzyme binds primarily to short base paired regions that border the sites of DNA cleavage and intron insertion. The minimal substrate spans no more than 15 bp and while sequence degeneracy is tolerated in the DNA binding regions, the sequence and size of the cleavage region is highly conserved. The enzyme has a slow turnover rate and cuts the coding strand with a slight preference over the non-coding strand. Complex formation produces some distortion of the DNA double helix within the cleavage region. The data are compatible with the two DNA-binding domains of I- Dmo I bridging the minor groove, where cleavage occurs, and interacting within the major groove on either side, thereby stabilizing a distorted DNA double helix. This may provide a general mode of DNA interaction at least for the LAGLIDADG-type homing enzymes.     

Crystallization and preliminary X-ray studies of I-PpoI: a nuclear, intron-encoded homing endonuclease from Physarum polycephalum.

The homing endonuclease I-PpoI is encoded by an optional third intron, Pp LSU 3, found in nuclear, extrachromosomal copies of the Physarum polycephalum 26S rRNA gene. This endonuclease promotes the lateral transfer or "homing" of its encoding intron by recognizing and cleaving a partially symmetric, 15 bp homing site in 26S rDNA alleles that lack the Pp LSU 3 intron. The open reading frame encoding I-PpoI has been subcloned, and the endonuclease has been overproduced in E. coli. Purified recombinant I-PpoI has been co-crystallized with a 21 bp homing site DNA duplex. The crystals belong to space group P3(1)21, with unit cell dimensions a = b = 114 A, c = 89 A. The results of initial X-ray diffraction experiments indicate that the asymmetric unit contains an enzyme homodimer and one duplex DNA molecule, and that the unit cell has a specific volume of 3.4 A3/dalton. These experiments also provide strong evidence that I-PpoI contains several bound zinc ions as part of its structure.     

Characterization of intein homing endonuclease encoded in the DNA polymerase gene of Thermococcus marinus

The DNA polymerase gene of Thermococcus marinus ( Tma ) contains an intein inserted at the pol-b site that possesses a 1611-bp ORF encoding a 537-amino acid residue. The LAGLIDADG motif, often found in site-specific DNA endonucleases, was detected within the amino acid sequence of the intein. The intein endonuclease, denoted as PI- Tma , was purified as a naturally spliced product from the expression of the complete DNA polymerase gene in Escherichia coli . PI- Tma cleaved intein-less DNA sequences, leaving four-base-long, 3'-hydroxyl overhangs with 5'-phosphate. Nonpalindromic recognition sequences 19 bp long were also identified using partially complementary oligonucleotide pair sequences inserted into the plasmid pET-22b(+). Cleavage by PI- Tma was optimal when present in 50 mM glycine–NaOH (pH 10.5), 150 mM KCl and 12 mM MgCl₂ at 70 °C.     

Microscopic insight to specificity of metal ion cofactor in DNA cleavage by restriction endonuclease EcoRV

Restriction endonucleases protect bacterial cells against bacteriophage infection by cleaving the incoming foreign DNA into fragments. In presence of Mg²⁺ ions, EcoRV is able to cleave the DNA but not in presence of Ca²⁺, although the protein binds to DNA in presence of both metal ions. We make an attempt to understand this difference using conformational thermodynamics. We calculate the changes in conformational free energy and entropy of conformational degrees of freedom, like DNA base pair steps and dihedral angles of protein residues in Mg²⁺(A)-EcoRV-DNA complex compared to Ca²⁺(S)-EcoRV-DNA complex using all-atom molecular dynamics (MD) trajectories of the complexes. We find that despite conformational stability and order in both complexes, the individual degrees of freedom behave differently in the presence of two different metal ions. The base pairs in cleavage region are highly disordered in Ca²⁺(S)-EcoRV-DNA compared to Mg²⁺(A)-EcoRV-DNA. One of the acidic residues ASP90, coordinating to the metal ion in the vicinity of the cleavage site, is conformationally destabilized and disordered, while basic residue LYS92 gets conformational stability and order in Ca²⁺(S) bound complex than in Mg²⁺(A) bound complex. The enhanced fluctuations hinder placement of the metal ion in the vicinity of the scissile phosphate of DNA. Similar loss of conformational stability and order in the cleavage region is observed by the replacement of the metal ion. Considering the placement of the metal ion near scissile phosphate as requirement for cleavage action, our results suggest that the changes in conformational stability and order of the base pair steps and the protein residues lead to cofactor sensitivity of the enzyme. Our method based on fluctuations of microscopic conformational variables can be applied to understand enzyme activities in other protein-DNA systems.     

Isolation and characterization of a type II restriction endonuclease from Streptococcus thermophilus

The alpha-toxin (phospholipase C) of Clostridium perfringens has been reported to contain catalytically essential zinc ions. We report here that histidine residues are essential for the co-ordination of these ion(s). Incubation of alpha toxin with diethylpyrocarbonate, a histidine modifying reagent, did not result in the loss of phospholipase C activity unless the protein was first incubated with EDTA, suggesting that zinc ions normally protect the susceptible histidine residues. When the amino acid sequences of three phospholipase C's were aligned, essential zinc binding histidine residues in the non-toxic B. cereus phospholipase C were found in similar positions in the toxic C. perfringens enzyme and the weakly toxic C. bifermentans phospholipase C.     

Restriction barrier composed of an extracellular nuclease and restriction endonuclease in the unicellular cyanobacterium Microcystis sp.

Abstract The unicellular cyanobacterium Microcystis aeruginosa K-81 has two types of restriction barrier, an extracellular nuclease and sequence-specific endonuclease. The nuclease was detected in the culture supernatant and it was easily released from the cells by washing with water or buffer containing Triton X-100. This nuclease was identified as a polypeptide of about 28 kDa that digested covalently closed circular and linear double-stranded DNAs, including chromosomal DNA from M. aeruginosa K-81. Among another 13 Microcystis strains examined, 3 produced an extracellular nuclease. Furthermore, M. aeruginosa K-81 contained two sequence-specific endonucleases, Mae K81I and Mae K81II, which were isoschizomers of Sp /I and Sau 96I, respectively.