Determining the effects of viral mutations

Jesse Bloom

Fred Hutch Cancer Center / HHMI

@jbloom_lab

Disclosures

I am on the scientific advisory boards of Apriori Bio, Invivyd, Aerium Therapeutics, the Vaccine, Company, and Oncorus

I am an inventor on Fred Hutch licensed patents related to deep mutational scanning of viral proteins

Evolution was historically studied by examining changes in phenotype

Haeckel. Pedigree of Man (1866)

~55 years ago, it became possible to also study evolution in terms of genotype

Zuckerkandl and Pauling (1965)

How changes in genotype affect phenotype is complex

"There is no reason to expect that the extent of functional change in a polypeptide chain is proportional to the number of amino-acid substitutions... It is the type rather than the number of substitutions that is decisive."

Zuckerkandl and Pauling (1965)

We can now characterize evolution of genotype at high resolution

Phylogenetic tree of ~13 million full-length SARS-CoV-2 genomes provided by GISAID

We want to know how changes in genotype affect phenotype

Zuckerkandl and Pauling (1965)

"There is no reason to expect that the extent of functional change in a polypeptide chain is proportional to the number of amino-acid substitutions... It is the type rather than the number of substitutions that is decisive."

Yesterday I described yeast display deep mutational scanning of SARS-CoV-2 RBD

Current deep mutational scanning used by our group and that of Yunlong Cao et al are based on yeast displaying libraries of the spike receptor-binding domain (RBD) and selecting for antibody or ACE2 binding.

Limitations of RBD yeast-display deep mutational scanning

• Only examines mutations in RBD, which is just part of spike
• Measures antibody binding, not neutralization. They are unequal in polyclonal serum.
• Only works for viral entry proteins with domains amenable to yeast display.

Two alternatives

• Lentiviral deep mutational scanning of viral entry proteins
• Estimation of mutation effects from natural sequences

Lentiviral pseudotyping

Bernadeta Dadonaite

Kate Crawford

Caelan Radford

Lentiviral pseudotyping

• Many viruses have entry proteins amenable to lentiviral pseudotyping.
• However, traditional pseudotyping does not create genotype-phenotype link.

Two-step method to create genotype-phenotype linked spike-pseudotypes

Dadonaite et al (2022)

Sequencing only measures relative amounts, so include "absolute standard"

Dadonaite et al (2022)

Example: RBD antibody LY-CoV1404

Validates very well

Also works for non-RBD antbodies

We can also measure functional effects of mutations on viral entry

Method should be extensible to many other viral entry proteins

• Other viruses with pseudotype-able entry proteins: other coronaviruses, influenza viruses, Ebola virus, Nipah virus, Lassa virus
• System can provide safe way to do high-throughput mutational studies
• We need to be aware of information hazard
• Information hazard low for SARS-CoV-2 spike, which is already human adapted
• But maybe hazard of human-receptor adaptive mutations in potential pandemic viruses
• For such viruses, we propose to use natural host rather than human receptors

Estimating mutation effects from natural sequences

Estimating mutation effects from natural sequences

with Richard Neher

We now have a lot of SARS-CoV-2 sequences

Phylogenetic tree of ~13 million full-length SARS-CoV-2 genomes provided by GISAID

We first estimate mutation rates from four-fold degenerate sites

We count mutations at such sites along tree

Enough sequences that each neutral mutation expected to be observed many times

We then compare actual counts to expected counts (from four-fold degenerate sites) for each mutation to estimate effect

Conclusions

Zuckerkandl and Pauling (1965)

"There is no reason to expect that the extent of functional change in a polypeptide chain is proportional to the number of amino-acid substitutions... It is the type rather than the number of substitutions that is decisive."

Understanding effects of mutations is long-standing but important problem

Two new approaches

• Lentiviral pseudotype deep mutational scanning
• Analysis of natural sequences