Limits...
The inherent mutational tolerance and antigenic evolvability of influenza hemagglutinin.

Thyagarajan B, Bloom JD - Elife (2014)

Bottom Line: We used deep mutational scanning to examine the extent to which a high inherent mutational tolerance contributes to this antigenic evolvability.These data enable us to infer the preference for each amino acid at each site in hemagglutinin.These inferences are consistent with existing knowledge about the protein's structure and function, and can be used to create a model that describes hemagglutinin's evolution far better than existing phylogenetic models.

View Article: PubMed Central - PubMed

Affiliation: Division of Basic Sciences, Fred Hutchinson Cancer Research Center, Seattle, United States.

ABSTRACT
Influenza is notable for its evolutionary capacity to escape immunity targeting the viral hemagglutinin. We used deep mutational scanning to examine the extent to which a high inherent mutational tolerance contributes to this antigenic evolvability. We created mutant viruses that incorporate most of the ≈10(4) amino-acid mutations to hemagglutinin from A/WSN/1933 (H1N1) influenza. After passaging these viruses in tissue culture to select for functional variants, we used deep sequencing to quantify mutation frequencies before and after selection. These data enable us to infer the preference for each amino acid at each site in hemagglutinin. These inferences are consistent with existing knowledge about the protein's structure and function, and can be used to create a model that describes hemagglutinin's evolution far better than existing phylogenetic models. We show that hemagglutinin has a high inherent tolerance for mutations at antigenic sites, suggesting that this is one factor contributing to influenza's antigenic evolution.

Show MeSH

Related in: MedlinePlus

Correlation of the site-specific amino-acid preferences determined in our study with the “relative fitness” (RF) values reported by Wu et al. (2014). Wu et al. (2014) report RF values for 2350 of the 564×19 = 10716 possible amino-acid mutations to the WSN HA examined in our study (they only examine single-nucleotide changes and disregard certain types of mutations due to oxidative damage of their DNA).To compare across the data sets, we have normalized their RF values by the RF value for the wildtype amino-acid (which they provide for only 2264 of the 2350 mutations). We then correlate on a logarithmic scale these normalized RF values with the ratio of our measurement of the preference for the mutant amino acid divided by the preference for the wildtype amino acid, using the preferences from our combined replicates. For mutations for which Wu et al. (2014) report an RF of zero, we assign a normalized RF equal to the smallest value for their entire data set. There is a significant Pearson correlation of 0.48 between the data sets, indicating that both our experiments and those of Wu et al. (2014) are capturing many of the same constraints on HA. The data and code used to create this figure are available via http://jbloom.github.io/mapmuts/example_WSN_HA_2014Analysis.html; this plot is the file correlation_with_Wu_et_al.pdf described therein.DOI:http://dx.doi.org/10.7554/eLife.03300.015
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4109307&req=5

fig7: Correlation of the site-specific amino-acid preferences determined in our study with the “relative fitness” (RF) values reported by Wu et al. (2014). Wu et al. (2014) report RF values for 2350 of the 564×19 = 10716 possible amino-acid mutations to the WSN HA examined in our study (they only examine single-nucleotide changes and disregard certain types of mutations due to oxidative damage of their DNA).To compare across the data sets, we have normalized their RF values by the RF value for the wildtype amino-acid (which they provide for only 2264 of the 2350 mutations). We then correlate on a logarithmic scale these normalized RF values with the ratio of our measurement of the preference for the mutant amino acid divided by the preference for the wildtype amino acid, using the preferences from our combined replicates. For mutations for which Wu et al. (2014) report an RF of zero, we assign a normalized RF equal to the smallest value for their entire data set. There is a significant Pearson correlation of 0.48 between the data sets, indicating that both our experiments and those of Wu et al. (2014) are capturing many of the same constraints on HA. The data and code used to create this figure are available via http://jbloom.github.io/mapmuts/example_WSN_HA_2014Analysis.html; this plot is the file correlation_with_Wu_et_al.pdf described therein.DOI:http://dx.doi.org/10.7554/eLife.03300.015

Mentions: Figure 7 compares the mutational effects determined in our study to those from Wu et al. (2014). There is a highly significant correlation between the results of the two studies—but the inferred mutational effects are certainly not identical. Because Wu et al. (2014) do not provide the data for replicates of their experiment, we are unable to assess whether the variability between the two different studies exceeds the variability between experimental replicates within each study. So one can imagine both biologically interesting and uninteresting explanations for the imperfect correlation between the results of the two studies. The interesting explanation is that differences in experimental methodology could lead to different selection pressures on specific mutations: for instance, Wu et al. (2014) use A549 cells while we use MDCK-SIAT1 cells, and perhaps the impact of certain mutations is dependent on the cell line. The uninteresting explanation is that the imperfect correlation is simply due to noise in the experimental measurements. Unfortunately, it is not straightforward to distinguish between these two explanations. This difficulty in pinpointing reasons for inter-study variation highlights a limitation of the high-throughput experimental methodology employed by ourselves and Wu et al. (2014): while such experiments provide a wealth of data, numerous factors can create noise in these data (sequencing errors, population bottlenecks, epistasis among mutations, etc). Realizing the full potential of such studies will therefore require extensive experimental controls and biological replicates to quantify errors and noise to enable comparisons across data sets.10.7554/eLife.03300.015Figure 7.Correlation of the site-specific amino-acid preferences determined in our study with the “relative fitness” (RF) values reported by Wu et al. (2014). Wu et al. (2014) report RF values for 2350 of the 564×19 = 10716 possible amino-acid mutations to the WSN HA examined in our study (they only examine single-nucleotide changes and disregard certain types of mutations due to oxidative damage of their DNA).


The inherent mutational tolerance and antigenic evolvability of influenza hemagglutinin.

Thyagarajan B, Bloom JD - Elife (2014)

Correlation of the site-specific amino-acid preferences determined in our study with the “relative fitness” (RF) values reported by Wu et al. (2014). Wu et al. (2014) report RF values for 2350 of the 564×19 = 10716 possible amino-acid mutations to the WSN HA examined in our study (they only examine single-nucleotide changes and disregard certain types of mutations due to oxidative damage of their DNA).To compare across the data sets, we have normalized their RF values by the RF value for the wildtype amino-acid (which they provide for only 2264 of the 2350 mutations). We then correlate on a logarithmic scale these normalized RF values with the ratio of our measurement of the preference for the mutant amino acid divided by the preference for the wildtype amino acid, using the preferences from our combined replicates. For mutations for which Wu et al. (2014) report an RF of zero, we assign a normalized RF equal to the smallest value for their entire data set. There is a significant Pearson correlation of 0.48 between the data sets, indicating that both our experiments and those of Wu et al. (2014) are capturing many of the same constraints on HA. The data and code used to create this figure are available via http://jbloom.github.io/mapmuts/example_WSN_HA_2014Analysis.html; this plot is the file correlation_with_Wu_et_al.pdf described therein.DOI:http://dx.doi.org/10.7554/eLife.03300.015
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4109307&req=5

fig7: Correlation of the site-specific amino-acid preferences determined in our study with the “relative fitness” (RF) values reported by Wu et al. (2014). Wu et al. (2014) report RF values for 2350 of the 564×19 = 10716 possible amino-acid mutations to the WSN HA examined in our study (they only examine single-nucleotide changes and disregard certain types of mutations due to oxidative damage of their DNA).To compare across the data sets, we have normalized their RF values by the RF value for the wildtype amino-acid (which they provide for only 2264 of the 2350 mutations). We then correlate on a logarithmic scale these normalized RF values with the ratio of our measurement of the preference for the mutant amino acid divided by the preference for the wildtype amino acid, using the preferences from our combined replicates. For mutations for which Wu et al. (2014) report an RF of zero, we assign a normalized RF equal to the smallest value for their entire data set. There is a significant Pearson correlation of 0.48 between the data sets, indicating that both our experiments and those of Wu et al. (2014) are capturing many of the same constraints on HA. The data and code used to create this figure are available via http://jbloom.github.io/mapmuts/example_WSN_HA_2014Analysis.html; this plot is the file correlation_with_Wu_et_al.pdf described therein.DOI:http://dx.doi.org/10.7554/eLife.03300.015
Mentions: Figure 7 compares the mutational effects determined in our study to those from Wu et al. (2014). There is a highly significant correlation between the results of the two studies—but the inferred mutational effects are certainly not identical. Because Wu et al. (2014) do not provide the data for replicates of their experiment, we are unable to assess whether the variability between the two different studies exceeds the variability between experimental replicates within each study. So one can imagine both biologically interesting and uninteresting explanations for the imperfect correlation between the results of the two studies. The interesting explanation is that differences in experimental methodology could lead to different selection pressures on specific mutations: for instance, Wu et al. (2014) use A549 cells while we use MDCK-SIAT1 cells, and perhaps the impact of certain mutations is dependent on the cell line. The uninteresting explanation is that the imperfect correlation is simply due to noise in the experimental measurements. Unfortunately, it is not straightforward to distinguish between these two explanations. This difficulty in pinpointing reasons for inter-study variation highlights a limitation of the high-throughput experimental methodology employed by ourselves and Wu et al. (2014): while such experiments provide a wealth of data, numerous factors can create noise in these data (sequencing errors, population bottlenecks, epistasis among mutations, etc). Realizing the full potential of such studies will therefore require extensive experimental controls and biological replicates to quantify errors and noise to enable comparisons across data sets.10.7554/eLife.03300.015Figure 7.Correlation of the site-specific amino-acid preferences determined in our study with the “relative fitness” (RF) values reported by Wu et al. (2014). Wu et al. (2014) report RF values for 2350 of the 564×19 = 10716 possible amino-acid mutations to the WSN HA examined in our study (they only examine single-nucleotide changes and disregard certain types of mutations due to oxidative damage of their DNA).

Bottom Line: We used deep mutational scanning to examine the extent to which a high inherent mutational tolerance contributes to this antigenic evolvability.These data enable us to infer the preference for each amino acid at each site in hemagglutinin.These inferences are consistent with existing knowledge about the protein's structure and function, and can be used to create a model that describes hemagglutinin's evolution far better than existing phylogenetic models.

View Article: PubMed Central - PubMed

Affiliation: Division of Basic Sciences, Fred Hutchinson Cancer Research Center, Seattle, United States.

ABSTRACT
Influenza is notable for its evolutionary capacity to escape immunity targeting the viral hemagglutinin. We used deep mutational scanning to examine the extent to which a high inherent mutational tolerance contributes to this antigenic evolvability. We created mutant viruses that incorporate most of the ≈10(4) amino-acid mutations to hemagglutinin from A/WSN/1933 (H1N1) influenza. After passaging these viruses in tissue culture to select for functional variants, we used deep sequencing to quantify mutation frequencies before and after selection. These data enable us to infer the preference for each amino acid at each site in hemagglutinin. These inferences are consistent with existing knowledge about the protein's structure and function, and can be used to create a model that describes hemagglutinin's evolution far better than existing phylogenetic models. We show that hemagglutinin has a high inherent tolerance for mutations at antigenic sites, suggesting that this is one factor contributing to influenza's antigenic evolution.

Show MeSH
Related in: MedlinePlus