Interpreting SARS-CoV-2 evolution

 

Jesse Bloom

Fred Hutch Cancer Center / HHMI

 

These slides at https://slides.com/jbloom/awaji-cov

 

The Faroe Islands

"Measles had not prevailed on the Faroes since 1781, then it broke out early in April 1846."

"Of the 7782 inhabitants, about 6000 were taken with measles."

"Of the many aged people still living in the Faroes who had measles in 1781, not one was attacked the second time."

Panum was describing immune memory, which provides lifelong protection from measles

What was reason for thinking coronavirus immunity would not be affected by viral mutations?

Coronaviruses have lower mutation rate than other RNA viruses

Coronaviruses are only RNA viruses with proofreading activity in their polymerase, and so have ~5- to 10-fold lower mutation rate than influenza virus

The average single-nucleotide mutation to SARS-CoV-2 had occurred >10,000 independent times in human-transmitted SARS-CoV-2 by third year of the pandemic.

But evolution of SARS-CoV-2 is not mutation-limited

Evolution is not just mutation: also involves selection on phenotypic effects of mutations

SARS-CoV-2 spike has acquired mutations in regions bound by neutralizing antibodies

Main regions where neutralizing antibodies bind

Sites of mutations in recent (BA.2.86) SARS-CoV-2 strain relative to early 2020 strain

These mutations have eroded antibody neutralization of newer variants

neutralization from original COVID-19 vaccine

Original vaccine induced high neutralizing antibody titers against early viral strains

These mutations have eroded antibody neutralization of newer variants

newer viral variants

neutralization from original COVID-19 vaccine

Deep mutational scanning to measure effects of mutations to viral proteins

RBD

fluorescent ACE2

yeast

fluorescent tag on RBD

Yeast display of spike receptor-binding domain (RBD) to measure ACE2 binding

cell sorting

Deep mutational scanning: selections on libraries of mutants, read out by sequencing

Experiments identified N501Y as a mutation that increases ACE2 binding

RBD

fluorescently labeled antibody

yeast

fluorescent tag on RBD

Yeast display to measure how all RBD mutations escape antibody binding

site in RBD

antibody escape

Experiments identified mutations at E484 as causing most antibody escape

484

These anecdotes suggest experiments can identify mutations subsequently selected by evolution

Viral infection is more biologically relevant measure of mutation effects

However, there are legitimate biosafety concerns about mutagenizing actual virus

actual SARS-CoV-2 virion: pathogen capable of spread in humans

pseudotyped lentiviral particle: not a pathogen, cannot spread in humans

actual SARS-CoV-2 virion: pathogen capable of spread in humans

pseudotyped lentiviral particle: not a pathogen, cannot spread in humans

Therefore we use pseudoviruses rather than actual replicative SARS-CoV-2

Pseudoviruses are modified viruses that can only undergo single round of infection

We need to link phenotype (spike mutant) and genotype (sequence encoded by virus)

To link genotype to phenotype, we encode spike in viral genome with barcode

We then create genotype-phenotype linked libraries by two-step process

Result is library of spike mutant pseudoviruses with identifying barcodes

  • Captures full cell-entry function of spike
  • Each mutant identified by short barcode
  • Pseudoviruses safe at biosafety-level 2
  • Broadly applicable to viral entry proteins

Measured how spike mutations affect three molecular phenotypes

Measured how spike mutations affect three molecular phenotypes

Measured how spike mutations affect three molecular phenotypes

Neutralization by soluble ACE2 is proportional to ACE2 binding affinity

Measured how spike mutations affect three molecular phenotypes

Full workflow to measure effects of mutations on antibody neutralization

Deep mutational scanning correlates well with traditional neutralization assays

How do these experimental measurements relate to actual SARS-CoV-2 evolution?

Estimating variant fitness (multinomial logistic regression)

Estimating variant fitness (multinomial logistic regression)

We analyze changes in clade growth rather than raw clade growth

change in clade growth

clade growth

Deep mutational scanning measurements correlate with SARS-CoV-2 variant fitness

Combining phenotypes offers best predictions (mutations can affect multiple phenotypes)

We can explain ~55% of the variance in growth of different clades, with largest fraction of variance uniquely explained by sera escape.

But human population in which SARS-CoV-2 evolves is changing...

Initially, most people were exposed to early strain, imprinting antibody response (     )

First exposed to early strain in original vaccine

First infected by an early strain (pre-Omicron)

Imprinting to early strain shapes antibody response even after later exposures

But antibody response of infants is imprinted by more recent strains

Imprinted by recent strain

We measured which spike mutations erode neutralization by antibodies from people imprinted with different strains

Adults vaccinated with original strain followed by various exposures

Adult sera from Helen Chu's HAARVI study; infant sera from Mary Staat's IMPRINT cohort

6-12 month infants first infected by XBB*

Dramatic differences in escape between adult vs infant sera

escape caused by mutations at site

site in spike

Average of sera from six infants with only a single infection by XBB* in 2023

Average of sera from ten adults imprinted by original vaccine in 2021

Adult sera have similar escape (mostly in RBD)

site in spike

escape caused by mutations at site

Sites of escape from sera of adults imprinted with original vaccine, then exposed to various infections and vaccinations.

adult 1

adult 2

adult 3

adult 4

adult 5

adult 6

2021

site in spike

escape caused by mutations at site

Sites of escape from infants with only single infection with XBB*

infant 1

infant 2

infant 3

infant 4

infant 5

infant 6

2023

Infant sera have similar escape (mostly in NTD)

Implications of population immune heterogeneity

How does it impact viral evolution when an antibody-escape mutation only affects the immunity of some individuals in the population?

 

How do different age groups contribute to driving viral evolution?

 

How much does person-to-person variation in impacts of viral mutations shape disease susceptibility?

Important questions related to population immune heterogeneity

How does it impact viral evolution when an antibody-escape mutation only affects the immunity of some individuals in the population?

 

Which are the key age groups for driving viral evolution?

 

How much does person-to-person variation in impacts of viral mutations shape disease susceptibility?

We recently developed new method to measure neutralization of ~100 of influenza strains by 100s of human sera

Sequencing-based neutralization assays

A single 96-well plate can measure ~3,000 curves (~1,000 titers in triplicate)

Unpublished data available here

We measured neutralization of many influenza strains by large set of human sera

sera from 95 individuals of different ages

Measured neutralization by each individual serum

Unpublished data available here

We measured neutralization of many influenza strains by large set of human sera

sera from 95 individuals of different ages

Measured neutralization by each individual serum

Measured neutralization by pool of all sera

Success of influenza strains correlates to fraction sera with low titers, not titer of sera pool

This result underscores importance of population immune heterogeneity in viral evolution

Bloom lab

Thanks

 

Fred Hutch Cancer Center

Trevor Bedford

 

University of Washington

Helen Chu and HAARVI cohort

Neil King

David Veesler

 

Cincinnati Children's Hospital

Mary Staat

David Haslam

Allie Burrell

Tyler Starr (now at Utah)

Allie Greaney (now at UCSF)

Kate Crawford

William Hannon

Caelan Radford

Brendan Larsen

Bernadeta Dadonaite

Rachel Eguia