Limits...
Expanding LAGLIDADG endonuclease scaffold diversity by rapidly surveying evolutionary sequence space.

Jacoby K, Metzger M, Shen BW, Certo MT, Jarjour J, Stoddard BL, Scharenberg AM - Nucleic Acids Res. (2012)

Bottom Line: One strategy to address such limitations is to identify close homologs of existing LHEs possessing superior biophysical or catalytic properties.Of these enzymes, I-HjeMI demonstrated the greatest activity in vivo and was readily crystallizable, allowing a comparative structural analysis.Taken together, our results suggest that even highly homologous LHEs offer a readily accessible resource of related scaffolds that display diverse biochemical properties for biotechnological applications.

View Article: PubMed Central - PubMed

Affiliation: Program in Molecular and Cellular Biology, University of Washington, Box 357275, Seattle, WA 98195, USA.

ABSTRACT
LAGLIDADG homing endonucleases (LHEs) are a family of highly specific DNA endonucleases capable of recognizing target sequences ≈ 20 bp in length, thus drawing intense interest for their potential academic, biotechnological and clinical applications. Methods for rational design of LHEs to cleave desired target sites are presently limited by a small number of high-quality native LHEs to serve as scaffolds for protein engineering-many are unsatisfactory for gene targeting applications. One strategy to address such limitations is to identify close homologs of existing LHEs possessing superior biophysical or catalytic properties. To test this concept, we searched public sequence databases to identify putative LHE open reading frames homologous to the LHE I-AniI and used a DNA binding and cleavage assay using yeast surface display to rapidly survey a subset of the predicted proteins. These proteins exhibited a range of capacities for surface expression and also displayed locally altered binding and cleavage specificities with a range of in vivo cleavage activities. Of these enzymes, I-HjeMI demonstrated the greatest activity in vivo and was readily crystallizable, allowing a comparative structural analysis. Taken together, our results suggest that even highly homologous LHEs offer a readily accessible resource of related scaffolds that display diverse biochemical properties for biotechnological applications.

Show MeSH

Related in: MedlinePlus

LHE characterization by flow cytometry. (A) Expression of full-length protein, determined by flow cytometry in a representative experiment. Staining against a C-terminal epitope tag and associated measured fluorescence intensity (Y-axis) and N-terminal epitope tag (X-axis) allows qualitative and quantitative assessment of surface expression. I-TasMIP, I-TinMIP and I-VinIP show minimal full-length protein (quadrant I and dual-positive), indicating reduced thermostability and/or poor folding. (B) Percentage of expressing (dual-positive) cells, as in panel (A), is summarized for five replicates (three for I-TasMIP, I-TinMIP and I-VinIP) with standard deviation plotted. (C) A western blot using antibodies against the N-terminal epitope tag allowed visualization of full length and truncated protein. Much of I-AchMI was expressed as ∼33 kDa protein fragment. Only minimal full-length protein and primarily heterogeneously truncated I-TasMIP, I-TinMIP and I-VinIP products were expressed, while I-Hje, I-PnoMI and I-AniI were primarily full length and in great abundance. (D) Demonstration of the gating strategy used to normalize substrate for the flow cleavage assay. These displayed populations are already normalized for enzyme concentration by a uniform, narrow FITC (C-terminal epitope) gate (data not shown). Equivalent amounts of tethered dsOligo across samples was selected by finding a streptavidin–PE level (rectangle) for each sample for which all DNA-A647 median fluorescence intensities (dashed horizontal line) were equal in the Ca++ sample (blue population). This gate was held constant for the matched pair Mg++ sample (red population), allowing quantification of magnesium-dependent loss of the DNA-conjugated fluorophore. The left half of the plot shows the population in the rectangular PE gate from the right plot (follow arrow). (E) Dividing the median Alexa647 fluorescence intensity of the calcium-containing sample (blue) by that of the magnesium-containing sample (red) yields a ratio proportional to the amount of enzymatic activity for a given LHE.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3367166&req=5

gkr1303-F2: LHE characterization by flow cytometry. (A) Expression of full-length protein, determined by flow cytometry in a representative experiment. Staining against a C-terminal epitope tag and associated measured fluorescence intensity (Y-axis) and N-terminal epitope tag (X-axis) allows qualitative and quantitative assessment of surface expression. I-TasMIP, I-TinMIP and I-VinIP show minimal full-length protein (quadrant I and dual-positive), indicating reduced thermostability and/or poor folding. (B) Percentage of expressing (dual-positive) cells, as in panel (A), is summarized for five replicates (three for I-TasMIP, I-TinMIP and I-VinIP) with standard deviation plotted. (C) A western blot using antibodies against the N-terminal epitope tag allowed visualization of full length and truncated protein. Much of I-AchMI was expressed as ∼33 kDa protein fragment. Only minimal full-length protein and primarily heterogeneously truncated I-TasMIP, I-TinMIP and I-VinIP products were expressed, while I-Hje, I-PnoMI and I-AniI were primarily full length and in great abundance. (D) Demonstration of the gating strategy used to normalize substrate for the flow cleavage assay. These displayed populations are already normalized for enzyme concentration by a uniform, narrow FITC (C-terminal epitope) gate (data not shown). Equivalent amounts of tethered dsOligo across samples was selected by finding a streptavidin–PE level (rectangle) for each sample for which all DNA-A647 median fluorescence intensities (dashed horizontal line) were equal in the Ca++ sample (blue population). This gate was held constant for the matched pair Mg++ sample (red population), allowing quantification of magnesium-dependent loss of the DNA-conjugated fluorophore. The left half of the plot shows the population in the rectangular PE gate from the right plot (follow arrow). (E) Dividing the median Alexa647 fluorescence intensity of the calcium-containing sample (blue) by that of the magnesium-containing sample (red) yields a ratio proportional to the amount of enzymatic activity for a given LHE.

Mentions: Yeast surface display represents a convenient method for characterizing putative LHEs in high throughput, as it provides facile access to quantitative information on protein folding and stability, DNA binding and cleavage, without the need for large scale enzymatic purification (17,28). In this approach, the enzyme is fused to an inducible surface displayed protein, Aga2p, which is anchored to the yeast's exterior by two disulfide bonds (29). Transit through the ER quality control and secretory pathways helps ensure that only LHEs which are stably folded at the induction temperature (20–30°C) are expressed on the yeast surface; dysfunctional variants which do not fold correctly are retained in direct proportion to their thermal stability (30). To compare the properties of six of the closest I-AniI homologs identified in Figure 1, yeast codon optimized ORFs were synthesized (Supplementary Table S1) and subcloned to the pCTCON2 vector. Relative expression levels were assessed using staining for N-terminal hemaggluTinMIn and C-terminal myc epitope tags (28,30) (Figure 2A). Three of the six homologs, I-AchMI, I-HjeMI and I-PnoMI, expressed full-length proteins on the yeast surface; the latter two very well, as determined by the level of C-terminal epitope tag expression (Figure 2B). I-TasMIP, I-TinMIP and I-VinIP surface expressed poorly, presumably because they were insufficiently stable at the 30°C induction temperature. Consistent with this interpretation, poor surface expression correlated with the accumulation of heterogeneously truncated proteins containing only the N-terminal tag, a pattern confirmed by western blot of the surface released protein (Figure 2C) and congruent with previous observations of surface-expressed proteins of low thermostability (31–34). Notably, the level of surface expression correlated with the level of amino acid sequence homology to I-AniI (Supplementary Figure S3).Figure 2.


Expanding LAGLIDADG endonuclease scaffold diversity by rapidly surveying evolutionary sequence space.

Jacoby K, Metzger M, Shen BW, Certo MT, Jarjour J, Stoddard BL, Scharenberg AM - Nucleic Acids Res. (2012)

LHE characterization by flow cytometry. (A) Expression of full-length protein, determined by flow cytometry in a representative experiment. Staining against a C-terminal epitope tag and associated measured fluorescence intensity (Y-axis) and N-terminal epitope tag (X-axis) allows qualitative and quantitative assessment of surface expression. I-TasMIP, I-TinMIP and I-VinIP show minimal full-length protein (quadrant I and dual-positive), indicating reduced thermostability and/or poor folding. (B) Percentage of expressing (dual-positive) cells, as in panel (A), is summarized for five replicates (three for I-TasMIP, I-TinMIP and I-VinIP) with standard deviation plotted. (C) A western blot using antibodies against the N-terminal epitope tag allowed visualization of full length and truncated protein. Much of I-AchMI was expressed as ∼33 kDa protein fragment. Only minimal full-length protein and primarily heterogeneously truncated I-TasMIP, I-TinMIP and I-VinIP products were expressed, while I-Hje, I-PnoMI and I-AniI were primarily full length and in great abundance. (D) Demonstration of the gating strategy used to normalize substrate for the flow cleavage assay. These displayed populations are already normalized for enzyme concentration by a uniform, narrow FITC (C-terminal epitope) gate (data not shown). Equivalent amounts of tethered dsOligo across samples was selected by finding a streptavidin–PE level (rectangle) for each sample for which all DNA-A647 median fluorescence intensities (dashed horizontal line) were equal in the Ca++ sample (blue population). This gate was held constant for the matched pair Mg++ sample (red population), allowing quantification of magnesium-dependent loss of the DNA-conjugated fluorophore. The left half of the plot shows the population in the rectangular PE gate from the right plot (follow arrow). (E) Dividing the median Alexa647 fluorescence intensity of the calcium-containing sample (blue) by that of the magnesium-containing sample (red) yields a ratio proportional to the amount of enzymatic activity for a given LHE.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3367166&req=5

gkr1303-F2: LHE characterization by flow cytometry. (A) Expression of full-length protein, determined by flow cytometry in a representative experiment. Staining against a C-terminal epitope tag and associated measured fluorescence intensity (Y-axis) and N-terminal epitope tag (X-axis) allows qualitative and quantitative assessment of surface expression. I-TasMIP, I-TinMIP and I-VinIP show minimal full-length protein (quadrant I and dual-positive), indicating reduced thermostability and/or poor folding. (B) Percentage of expressing (dual-positive) cells, as in panel (A), is summarized for five replicates (three for I-TasMIP, I-TinMIP and I-VinIP) with standard deviation plotted. (C) A western blot using antibodies against the N-terminal epitope tag allowed visualization of full length and truncated protein. Much of I-AchMI was expressed as ∼33 kDa protein fragment. Only minimal full-length protein and primarily heterogeneously truncated I-TasMIP, I-TinMIP and I-VinIP products were expressed, while I-Hje, I-PnoMI and I-AniI were primarily full length and in great abundance. (D) Demonstration of the gating strategy used to normalize substrate for the flow cleavage assay. These displayed populations are already normalized for enzyme concentration by a uniform, narrow FITC (C-terminal epitope) gate (data not shown). Equivalent amounts of tethered dsOligo across samples was selected by finding a streptavidin–PE level (rectangle) for each sample for which all DNA-A647 median fluorescence intensities (dashed horizontal line) were equal in the Ca++ sample (blue population). This gate was held constant for the matched pair Mg++ sample (red population), allowing quantification of magnesium-dependent loss of the DNA-conjugated fluorophore. The left half of the plot shows the population in the rectangular PE gate from the right plot (follow arrow). (E) Dividing the median Alexa647 fluorescence intensity of the calcium-containing sample (blue) by that of the magnesium-containing sample (red) yields a ratio proportional to the amount of enzymatic activity for a given LHE.
Mentions: Yeast surface display represents a convenient method for characterizing putative LHEs in high throughput, as it provides facile access to quantitative information on protein folding and stability, DNA binding and cleavage, without the need for large scale enzymatic purification (17,28). In this approach, the enzyme is fused to an inducible surface displayed protein, Aga2p, which is anchored to the yeast's exterior by two disulfide bonds (29). Transit through the ER quality control and secretory pathways helps ensure that only LHEs which are stably folded at the induction temperature (20–30°C) are expressed on the yeast surface; dysfunctional variants which do not fold correctly are retained in direct proportion to their thermal stability (30). To compare the properties of six of the closest I-AniI homologs identified in Figure 1, yeast codon optimized ORFs were synthesized (Supplementary Table S1) and subcloned to the pCTCON2 vector. Relative expression levels were assessed using staining for N-terminal hemaggluTinMIn and C-terminal myc epitope tags (28,30) (Figure 2A). Three of the six homologs, I-AchMI, I-HjeMI and I-PnoMI, expressed full-length proteins on the yeast surface; the latter two very well, as determined by the level of C-terminal epitope tag expression (Figure 2B). I-TasMIP, I-TinMIP and I-VinIP surface expressed poorly, presumably because they were insufficiently stable at the 30°C induction temperature. Consistent with this interpretation, poor surface expression correlated with the accumulation of heterogeneously truncated proteins containing only the N-terminal tag, a pattern confirmed by western blot of the surface released protein (Figure 2C) and congruent with previous observations of surface-expressed proteins of low thermostability (31–34). Notably, the level of surface expression correlated with the level of amino acid sequence homology to I-AniI (Supplementary Figure S3).Figure 2.

Bottom Line: One strategy to address such limitations is to identify close homologs of existing LHEs possessing superior biophysical or catalytic properties.Of these enzymes, I-HjeMI demonstrated the greatest activity in vivo and was readily crystallizable, allowing a comparative structural analysis.Taken together, our results suggest that even highly homologous LHEs offer a readily accessible resource of related scaffolds that display diverse biochemical properties for biotechnological applications.

View Article: PubMed Central - PubMed

Affiliation: Program in Molecular and Cellular Biology, University of Washington, Box 357275, Seattle, WA 98195, USA.

ABSTRACT
LAGLIDADG homing endonucleases (LHEs) are a family of highly specific DNA endonucleases capable of recognizing target sequences ≈ 20 bp in length, thus drawing intense interest for their potential academic, biotechnological and clinical applications. Methods for rational design of LHEs to cleave desired target sites are presently limited by a small number of high-quality native LHEs to serve as scaffolds for protein engineering-many are unsatisfactory for gene targeting applications. One strategy to address such limitations is to identify close homologs of existing LHEs possessing superior biophysical or catalytic properties. To test this concept, we searched public sequence databases to identify putative LHE open reading frames homologous to the LHE I-AniI and used a DNA binding and cleavage assay using yeast surface display to rapidly survey a subset of the predicted proteins. These proteins exhibited a range of capacities for surface expression and also displayed locally altered binding and cleavage specificities with a range of in vivo cleavage activities. Of these enzymes, I-HjeMI demonstrated the greatest activity in vivo and was readily crystallizable, allowing a comparative structural analysis. Taken together, our results suggest that even highly homologous LHEs offer a readily accessible resource of related scaffolds that display diverse biochemical properties for biotechnological applications.

Show MeSH
Related in: MedlinePlus