Limits...
Mapping the Pareto optimal design space for a functionally deimmunized biotherapeutic candidate.

Salvat RS, Parker AS, Choi Y, Bailey-Kellogg C, Griswold KE - PLoS Comput. Biol. (2015)

Bottom Line: As a result, there is a growing need for improved deimmunization technologies.Various measures of protein performance were found to map a functional sequence space that correlated well with computational predictions.These results represent the first systematic and rigorous assessment of the functional penalty that must be paid for pursuing progressively more deimmunized biotherapeutic candidates.

View Article: PubMed Central - PubMed

Affiliation: Thayer School of Engineering, Dartmouth, Hanover, New Hampshire, United States of America.

ABSTRACT
The immunogenicity of biotherapeutics can bottleneck development pipelines and poses a barrier to widespread clinical application. As a result, there is a growing need for improved deimmunization technologies. We have recently described algorithms that simultaneously optimize proteins for both reduced T cell epitope content and high-level function. In silico analysis of this dual objective design space reveals that there is no single global optimum with respect to protein deimmunization. Instead, mutagenic epitope deletion yields a spectrum of designs that exhibit tradeoffs between immunogenic potential and molecular function. The leading edge of this design space is the Pareto frontier, i.e. the undominated variants for which no other single design exhibits better performance in both criteria. Here, the Pareto frontier of a therapeutic enzyme has been designed, constructed, and evaluated experimentally. Various measures of protein performance were found to map a functional sequence space that correlated well with computational predictions. These results represent the first systematic and rigorous assessment of the functional penalty that must be paid for pursuing progressively more deimmunized biotherapeutic candidates. Given this capacity to rapidly assess and design for tradeoffs between protein immunogenicity and functionality, these algorithms may prove useful in augmenting, accelerating, and de-risking experimental deimmunization efforts.

Show MeSH
Comparison of computed and experimental Pareto optimal plots.A Pareto plot of computed design parameters Sseq vs. Sepi. B Experimental analog of the computed Pareto plot. An integrated score for experimentally measured molecular function (equation 3) is plotted vs. a global score for experimentally determined immunoreactivity (equation 4). C Experimental Pareto plot of normalized, reciprocal Tm vs. Global Quantitative Immunoreactivity. D Experimental Pareto plot of normalized Km vs. Global Quantitative Immunoreactivity. E Experimental Pareto plot of normalized, reciprocal kcat vs. Global Quantitative Immunoreactivity. F Experimental Pareto plot of normalized, reciprocal kcat/Km vs. Global Quantitative Immunoreactivity. G Experimental Pareto plot of averaged, normalized, reciprocal kcat and Tm vs. Global Quantitative Immunoreactivity. H Experimental Pareto plot of averaged, normalized, reciprocal kcat/Km and Tm vs. Global Quantitative Immunoreactivity. Pareto optimal enzymes are shown as blue circular markers, sub-optimal 4-mutation variants are shown as orange circular markers, and wild type P99βL is shown as a red square. Note that the computed Pareto plot best captures overall molecular performance, as represented by integrated performance values (e.g. averaging kinetic parameters with thermostability parameters).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4288714&req=5

pcbi-1003988-g008: Comparison of computed and experimental Pareto optimal plots.A Pareto plot of computed design parameters Sseq vs. Sepi. B Experimental analog of the computed Pareto plot. An integrated score for experimentally measured molecular function (equation 3) is plotted vs. a global score for experimentally determined immunoreactivity (equation 4). C Experimental Pareto plot of normalized, reciprocal Tm vs. Global Quantitative Immunoreactivity. D Experimental Pareto plot of normalized Km vs. Global Quantitative Immunoreactivity. E Experimental Pareto plot of normalized, reciprocal kcat vs. Global Quantitative Immunoreactivity. F Experimental Pareto plot of normalized, reciprocal kcat/Km vs. Global Quantitative Immunoreactivity. G Experimental Pareto plot of averaged, normalized, reciprocal kcat and Tm vs. Global Quantitative Immunoreactivity. H Experimental Pareto plot of averaged, normalized, reciprocal kcat/Km and Tm vs. Global Quantitative Immunoreactivity. Pareto optimal enzymes are shown as blue circular markers, sub-optimal 4-mutation variants are shown as orange circular markers, and wild type P99βL is shown as a red square. Note that the computed Pareto plot best captures overall molecular performance, as represented by integrated performance values (e.g. averaging kinetic parameters with thermostability parameters).

Mentions: The studies described here were enabled by an advanced deimmunization algorithm that seamlessly integrates immunogenic epitope prediction with in silico analysis of the functional consequences associated with deimmunizing mutations. We combined the IP2 deimmunization formulation with the Pepfr optimization algorithm [28], [32] to design a suite of 18 Pareto optimal P99βL variants. Each of these designs optimally balances two objective functions – one modeling immunogenicity and the other functionality – such that no other single variant is predicted to outperform with respect to both design objectives. Inspection of the Pareto optimal designs reveals that, in the context of the mathematical model, there is an inverse relationship wherein ever greater deimmunization is achieved at the expense of progressively reduced function (Fig. 8A). To assess the practical implications of these predicted tradeoffs, we have recombinantly produced all 18 designed enzymes and rigorously characterized their stability, activity, and immunoreactivity with human MHC II proteins. The results of this analysis represent the first systematic assessment of the functional penalty that is paid for pursuing progressively more deimmunized drug candidates.


Mapping the Pareto optimal design space for a functionally deimmunized biotherapeutic candidate.

Salvat RS, Parker AS, Choi Y, Bailey-Kellogg C, Griswold KE - PLoS Comput. Biol. (2015)

Comparison of computed and experimental Pareto optimal plots.A Pareto plot of computed design parameters Sseq vs. Sepi. B Experimental analog of the computed Pareto plot. An integrated score for experimentally measured molecular function (equation 3) is plotted vs. a global score for experimentally determined immunoreactivity (equation 4). C Experimental Pareto plot of normalized, reciprocal Tm vs. Global Quantitative Immunoreactivity. D Experimental Pareto plot of normalized Km vs. Global Quantitative Immunoreactivity. E Experimental Pareto plot of normalized, reciprocal kcat vs. Global Quantitative Immunoreactivity. F Experimental Pareto plot of normalized, reciprocal kcat/Km vs. Global Quantitative Immunoreactivity. G Experimental Pareto plot of averaged, normalized, reciprocal kcat and Tm vs. Global Quantitative Immunoreactivity. H Experimental Pareto plot of averaged, normalized, reciprocal kcat/Km and Tm vs. Global Quantitative Immunoreactivity. Pareto optimal enzymes are shown as blue circular markers, sub-optimal 4-mutation variants are shown as orange circular markers, and wild type P99βL is shown as a red square. Note that the computed Pareto plot best captures overall molecular performance, as represented by integrated performance values (e.g. averaging kinetic parameters with thermostability parameters).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4288714&req=5

pcbi-1003988-g008: Comparison of computed and experimental Pareto optimal plots.A Pareto plot of computed design parameters Sseq vs. Sepi. B Experimental analog of the computed Pareto plot. An integrated score for experimentally measured molecular function (equation 3) is plotted vs. a global score for experimentally determined immunoreactivity (equation 4). C Experimental Pareto plot of normalized, reciprocal Tm vs. Global Quantitative Immunoreactivity. D Experimental Pareto plot of normalized Km vs. Global Quantitative Immunoreactivity. E Experimental Pareto plot of normalized, reciprocal kcat vs. Global Quantitative Immunoreactivity. F Experimental Pareto plot of normalized, reciprocal kcat/Km vs. Global Quantitative Immunoreactivity. G Experimental Pareto plot of averaged, normalized, reciprocal kcat and Tm vs. Global Quantitative Immunoreactivity. H Experimental Pareto plot of averaged, normalized, reciprocal kcat/Km and Tm vs. Global Quantitative Immunoreactivity. Pareto optimal enzymes are shown as blue circular markers, sub-optimal 4-mutation variants are shown as orange circular markers, and wild type P99βL is shown as a red square. Note that the computed Pareto plot best captures overall molecular performance, as represented by integrated performance values (e.g. averaging kinetic parameters with thermostability parameters).
Mentions: The studies described here were enabled by an advanced deimmunization algorithm that seamlessly integrates immunogenic epitope prediction with in silico analysis of the functional consequences associated with deimmunizing mutations. We combined the IP2 deimmunization formulation with the Pepfr optimization algorithm [28], [32] to design a suite of 18 Pareto optimal P99βL variants. Each of these designs optimally balances two objective functions – one modeling immunogenicity and the other functionality – such that no other single variant is predicted to outperform with respect to both design objectives. Inspection of the Pareto optimal designs reveals that, in the context of the mathematical model, there is an inverse relationship wherein ever greater deimmunization is achieved at the expense of progressively reduced function (Fig. 8A). To assess the practical implications of these predicted tradeoffs, we have recombinantly produced all 18 designed enzymes and rigorously characterized their stability, activity, and immunoreactivity with human MHC II proteins. The results of this analysis represent the first systematic assessment of the functional penalty that is paid for pursuing progressively more deimmunized drug candidates.

Bottom Line: As a result, there is a growing need for improved deimmunization technologies.Various measures of protein performance were found to map a functional sequence space that correlated well with computational predictions.These results represent the first systematic and rigorous assessment of the functional penalty that must be paid for pursuing progressively more deimmunized biotherapeutic candidates.

View Article: PubMed Central - PubMed

Affiliation: Thayer School of Engineering, Dartmouth, Hanover, New Hampshire, United States of America.

ABSTRACT
The immunogenicity of biotherapeutics can bottleneck development pipelines and poses a barrier to widespread clinical application. As a result, there is a growing need for improved deimmunization technologies. We have recently described algorithms that simultaneously optimize proteins for both reduced T cell epitope content and high-level function. In silico analysis of this dual objective design space reveals that there is no single global optimum with respect to protein deimmunization. Instead, mutagenic epitope deletion yields a spectrum of designs that exhibit tradeoffs between immunogenic potential and molecular function. The leading edge of this design space is the Pareto frontier, i.e. the undominated variants for which no other single design exhibits better performance in both criteria. Here, the Pareto frontier of a therapeutic enzyme has been designed, constructed, and evaluated experimentally. Various measures of protein performance were found to map a functional sequence space that correlated well with computational predictions. These results represent the first systematic and rigorous assessment of the functional penalty that must be paid for pursuing progressively more deimmunized biotherapeutic candidates. Given this capacity to rapidly assess and design for tradeoffs between protein immunogenicity and functionality, these algorithms may prove useful in augmenting, accelerating, and de-risking experimental deimmunization efforts.

Show MeSH