Limits...
Comprehensive computational design of mCreI homing endonuclease cleavage specificity for genome engineering.

Ulge UY, Baker DA, Monnat RJ - Nucleic Acids Res. (2011)

Bottom Line: Homing endonucleases (HEs) cleave long (∼ 20 bp) DNA target sites with high site specificity to catalyze the lateral transfer of parasitic DNA elements.Experimental verification of a range of these designs demonstrated that over 2/3 (24 of 35 designs, 69%) had the intended new site specificity, and that 14 of the 15 attempted specificity shifts (93%) were achieved.These results demonstrate the feasibility of using structure-based computational design to engineer HE variants with novel target site specificities to facilitate genome engineering.

View Article: PubMed Central - PubMed

Affiliation: Department of Biochemistry, Howard Hughes Medical InstituteUniversity of Washington, Box 357705, Seattle, WA 98195, USA.

ABSTRACT
Homing endonucleases (HEs) cleave long (∼ 20 bp) DNA target sites with high site specificity to catalyze the lateral transfer of parasitic DNA elements. In order to determine whether comprehensive computational design could be used as a general strategy to engineer new HE target site specificities, we used RosettaDesign (RD) to generate 3200 different variants of the mCreI LAGLIDADG HE towards 16 different base pair positions in the 22 bp mCreI target site. Experimental verification of a range of these designs demonstrated that over 2/3 (24 of 35 designs, 69%) had the intended new site specificity, and that 14 of the 15 attempted specificity shifts (93%) were achieved. These results demonstrate the feasibility of using structure-based computational design to engineer HE variants with novel target site specificities to facilitate genome engineering.

Show MeSH
Graphical representation of mCreI computational design output. A library of 3200 mCreI designs was generated using RD against all 4 base pair possibilities at each target site position from ±3 to ±11 (see text). The RD-predicted specificities and energies of 117 designs are plotted that represent the most energetically stable or the most specific of the 50 designs generated for each design target. Only a single design is plotted for instances in which the most stable and most specific design were the same. Experimentally validated design specificities are represented by squares labeled with the design base pair and position. Useful designs for target site positions ±11 did not emerge and are not represented, nor is a design that cleaved −8C that was an unanticipated—albeit a sequence-specific—outcome of an attempt to design for −8G (Table 1, Design 12).
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3105429&req=5

Figure 3: Graphical representation of mCreI computational design output. A library of 3200 mCreI designs was generated using RD against all 4 base pair possibilities at each target site position from ±3 to ±11 (see text). The RD-predicted specificities and energies of 117 designs are plotted that represent the most energetically stable or the most specific of the 50 designs generated for each design target. Only a single design is plotted for instances in which the most stable and most specific design were the same. Experimentally validated design specificities are represented by squares labeled with the design base pair and position. Useful designs for target site positions ±11 did not emerge and are not represented, nor is a design that cleaved −8C that was an unanticipated—albeit a sequence-specific—outcome of an attempt to design for −8G (Table 1, Design 12).

Mentions: We based our choice of designs for further analysis on a combination of RD-predicted specificity, favorable energies and structural plausibility at all design base pair positions and across a range of predicted specificities (Figure 3). We experimentally characterized the cleavage specificity and activity of 35 mCreI design variants representing 13 different single base pair-variant target sites, as well as four different designs against the native mCreI target site (Table 1). Some of the included designs had small amino acid variations at the same residue because in silico calculations did not provide a single best solution despite iterative design attempts (see, e.g. Designs 5–8 against target −9C).Figure 3.


Comprehensive computational design of mCreI homing endonuclease cleavage specificity for genome engineering.

Ulge UY, Baker DA, Monnat RJ - Nucleic Acids Res. (2011)

Graphical representation of mCreI computational design output. A library of 3200 mCreI designs was generated using RD against all 4 base pair possibilities at each target site position from ±3 to ±11 (see text). The RD-predicted specificities and energies of 117 designs are plotted that represent the most energetically stable or the most specific of the 50 designs generated for each design target. Only a single design is plotted for instances in which the most stable and most specific design were the same. Experimentally validated design specificities are represented by squares labeled with the design base pair and position. Useful designs for target site positions ±11 did not emerge and are not represented, nor is a design that cleaved −8C that was an unanticipated—albeit a sequence-specific—outcome of an attempt to design for −8G (Table 1, Design 12).
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3105429&req=5

Figure 3: Graphical representation of mCreI computational design output. A library of 3200 mCreI designs was generated using RD against all 4 base pair possibilities at each target site position from ±3 to ±11 (see text). The RD-predicted specificities and energies of 117 designs are plotted that represent the most energetically stable or the most specific of the 50 designs generated for each design target. Only a single design is plotted for instances in which the most stable and most specific design were the same. Experimentally validated design specificities are represented by squares labeled with the design base pair and position. Useful designs for target site positions ±11 did not emerge and are not represented, nor is a design that cleaved −8C that was an unanticipated—albeit a sequence-specific—outcome of an attempt to design for −8G (Table 1, Design 12).
Mentions: We based our choice of designs for further analysis on a combination of RD-predicted specificity, favorable energies and structural plausibility at all design base pair positions and across a range of predicted specificities (Figure 3). We experimentally characterized the cleavage specificity and activity of 35 mCreI design variants representing 13 different single base pair-variant target sites, as well as four different designs against the native mCreI target site (Table 1). Some of the included designs had small amino acid variations at the same residue because in silico calculations did not provide a single best solution despite iterative design attempts (see, e.g. Designs 5–8 against target −9C).Figure 3.

Bottom Line: Homing endonucleases (HEs) cleave long (∼ 20 bp) DNA target sites with high site specificity to catalyze the lateral transfer of parasitic DNA elements.Experimental verification of a range of these designs demonstrated that over 2/3 (24 of 35 designs, 69%) had the intended new site specificity, and that 14 of the 15 attempted specificity shifts (93%) were achieved.These results demonstrate the feasibility of using structure-based computational design to engineer HE variants with novel target site specificities to facilitate genome engineering.

View Article: PubMed Central - PubMed

Affiliation: Department of Biochemistry, Howard Hughes Medical InstituteUniversity of Washington, Box 357705, Seattle, WA 98195, USA.

ABSTRACT
Homing endonucleases (HEs) cleave long (∼ 20 bp) DNA target sites with high site specificity to catalyze the lateral transfer of parasitic DNA elements. In order to determine whether comprehensive computational design could be used as a general strategy to engineer new HE target site specificities, we used RosettaDesign (RD) to generate 3200 different variants of the mCreI LAGLIDADG HE towards 16 different base pair positions in the 22 bp mCreI target site. Experimental verification of a range of these designs demonstrated that over 2/3 (24 of 35 designs, 69%) had the intended new site specificity, and that 14 of the 15 attempted specificity shifts (93%) were achieved. These results demonstrate the feasibility of using structure-based computational design to engineer HE variants with novel target site specificities to facilitate genome engineering.

Show MeSH