Katia Koelle Sarah Cobey Bryan Grenfell Mercedes Pascual Epochal evolution shapes the phylodynamics of interpandemic influenza (H3N2) DIMACS, 9-10 October.

Katia Koelle Sarah Cobey Bryan Grenfell Mercedes Pascual

Epochal evolution shapes the phylodynamics of interpandemic influenza (H3N2)

DIMACS, 9-10 October 2006

HK68

EN72

VI75

TX77

BK79

SI87?

?

Pathogen diversity and cross-immunity

Modeling Cross-Immunity

• Strains with high sequence similarity must have high cross-immunity• Strains with low sequence similarity must have low cross-immunity

e.g. Gog & Grenfell, PNAS (2002)

Explaining limited diversity of hemagglutinin

Ferguson, Galvani, Bush, Nature (2003)

Explosive diversity

Strain-specific cross-immunity

Actual HA1phylogeny

Simulatedphylogeny

Explaining limited diversity

Strain-specific cross-immunity + generalized immunity

Years since infection

Imm

un

ity

Limited diversity

Ferguson, Galvani, Bush, Nature (2003)

• Can sequence evolution be used as a proxy for antigenic evolution when modeling influenza’s hemagglutinin?

(i.e. does genotype approximate phenotype?)

• Propose alternative to this genotype-phenotype map for influenza’s hemagglutinin evolution

Modeling cross-immunity between flu strains

• Consider the effect of this new mapping on the phylogenetics and dynamics (i.e. phylodynamics) of influenza H3N2

Unrooted ML trees of sequences in the HK68 and EN72 clusters

= 60-80%

= >90% = >90%

Influenza clusters

Cluster designations as in Smith et al. 2004

• Strains with high sequence similarity can have low cross-immunity• Strains with low sequence similarity can have almost complete cross-immunity

Topology of influenza clusters

Genotype cannot serve as a proxy for antigenic phenotype

STRAIN 1

Sequence (genotype)

…ATGATGTGCCGGAT…

…FLIMFYNKSR…

STRAIN 2

Sequence (genotype)

…ATGATCTGCCGGAT…

…FLIDFYNKSR…

Tertiary HA structure (phenotype)

Tertiary HA structure (phenotype)

Cross-immunity

Genotype-phenotypemapping?

Genotype-phenotype mapping for RNA 2o structures

More genotypes than phenotypes

genotype(sequence)

phenotype(shape)

Fontana & Schuster, JTB (1998)

Neutral networks

Fontana & Schuster, JTB (1998)

• “A neutral mutation does not change the phenotype but it does change the potential for change… What appears to be a sudden and abrupt

change at the phenotypic level has been the result of neutral genetic drift.” -Fontana

Evolutionary dynamics on neutral networks

Avera

ge s

truct

ure

dis

tance

to t

arg

et

Fontana & Schuster, JTB (1998)

Neutral network mapping for proteins

• Single sequence changes can result in large changes in protein conformation.

• Changing a sequence by a large number of mutations may have no appreciable effect on protein conformation.

Lau and Dill

Bornberg-Bauer & Chan, PNAS (1999)

Implications for modeling cross-immunity

…FLIMFYNKSR…

Traditionalcross-immunity

models

Neutral network topology Bornberg-Bauer

Modeling influenza’s hemagglutinin

15 a.a.

(45 nucs.) 5 epitopes

Changing the shape of an epitope

15 a.a.

5 epitopes

• Adaptation of Kauffman’s NK model that generates neutral networks in genotype space (Newman and Engelhardt)

• Framework assumes epistatic or context-dependent interaction between amino acids located in the same epitope

3

Neutrality and sequence evolution:subbasins, portals, and epochal evolution

HK68

EN72

VI75

TX77

BK79

SI87?

?

Adapted (for flu ) from Crutchfield, 2002

Coupling to an epidemiological modelC

lust

ers

InfectedSusceptible Recovered

Adapted for clusters, from Gog & Grenfell, PNAS (2002)

Dynamic Consequences of Neutral Network Model

• Cluster transitions

• Peaks in incidence during

cluster transition years

• Refractory year

Years

Greene et al. (2006)

Comparison with observed influenza dynamics

0.1

Simulated tree

Observed HA tree(from Smith et al.

sequences)0.1

• Explosion of diversity within clusters• Cluster transitions cause selective sweeps• No need for generalized immunity to limit HA diversity

Phylogenetic Consequences

Expected pattern in genetic diversity arising from epochal evolution

Supporting empirical evidence

Influential sites modelOnly changes at very few sites can precipitate a cluster jump, and their ability to do so does not depend on the genetic background in which they occur.

Genetic diversification within clusters does not facilitate adaptive change, and can be safely ignored.

Context-dependent modelChanges at most sites can precipitate a cluster jump if those changes occur in the right genetic background.

Cluster innovations are guided by the process of neutral diffusion, via changing the genetic background of sequences.

Notions of neutrality

See also Wagner, 2005 for a discussion on types of neutrality in non-flu systems

Epitope Residue Substitution Associated transition Clusters with neutral polymorphism at residueO 25 LI SY97-FU02

C 50 KR VI75-TX77 HK68 (K/R), BE89 (K/R), WU95 (I/K/R), SY97 (G/R)

RG SY97-FU02

C 53 ND EN72-VI75 TX77 (D/N), SI87 (D/N), BE92 (D/G), FU02 (D/N)

DN VI75-TX77

ND TX77-BK79

C 54 NS TX77-BK79 HK68 (N/S), TX77 (N/S), WU95 (G/S), SY97 (I/S)

E 62 IK TX77-BK79

KE WU95-SY97

E 75 HQ SY97-FU02 BE92 (H/N), WU95 (H/N)

E 82 EK VI75-TX77 TX77 (E/K), SI87 (E/K)

KE TX77-BK79

E 83 EK SY97-FU02 EN72 (K/T), SI87 (E/K)

A 122 TN HK68-EN72 SI87 (K/N), BE92 (K/N), WU95 (K/N)

A 124 GD BK79-SI87 BK79 (D/G/S), SI87 (D/E), BE92 (A/D/G/N), WU95 (D/G/S)

A 131 AT SY97-FU02 SI87 (A/T), SY97 (A/D)

A 133 NS TX77-BK79

SD BE89-BE92

A 137 NS EN72-VI75 HK68 (N/S), VI75 (G/S), BK79 (S/Y), SY97 (S/Y)

SY VI75-TX77

A 143 PS TX77-BK79 EN72 (P/T)

A 144 GD HK68-EN72 HK68 (D/G), BK79 (D/N/V), SI87 (I/V), SY97 (D/I/N/V)

A 145 SN EN72-VI75

NK SI87-BE89

KN BE89-BE92

NK BE92-WU95

A 146 GS TX77-BK79 HK68 (G/R), WU95 (G/S)

B 155 TY HK68-EN72

YH BK79-SI87

HT SY97-FU02

B 156 KE TX77-BK79 BK79 (E/K), SI87 (E/K)

EK BE89-BE92

KQ WU95-SY97

QH SY97-FU02

B 158 GE VI75-TX77 BE89 (D/E), BE92 (D/E)

EK WU95-SY97

B 160 TK TX77-BK79 EN72 (A/T), WU95 (K/R)

B 164 LQ EN72-VI75

QL VI75-TX77

D 172 DG TX77-BK79 TX77 (D/G), BE89 (D/G), BE92 (D/G), WU95 (D/G), SY97 (D/E)

D 174 FS EN72-VI75 BE89 (F/V)

SF VI75-TX77

B 188 ND HK68-EN72 HK68 (D/N), EN72 (D/N), BK79 (D/E), BE89 (D/E), WU95 (D/N)

B 189 QK HK68-EN72 BE92 (R/S)

KR BK79-SI87

B 190 ED BE89-BE92 EN72 (D/E), SI87 (D/E), BE89 (D/E), WU95 (D/V), SY97 (D/X)

B 193 SD EN72-VI75 EN72 (N/S), BK79 (K/N), SI87 (N/S), BE89 (K/N/S)

DN VI75-TX77

B 196 VA WU95-SY97 BK79 (I/V), BE89 (I/V), BE92 (I/V), SY97 (A/X)

B 197 QR TX77-BK79 BE92 (Q/R), WU95 (H/Q/R)

D 201 RK EN72-VI75 BE92 (G/K/R)

KR VI75-TX77

O 202 VI SY97-FU02

D 207 RK HK68-EN72 HK68 (K/R), WU95 (K/R), SY97 (K/R)

D 213 IV EN72-VI75 EN72 (I/V), BK79 (I/R/V)

VI VI75-TX77

D 217 IV EN72-VI75

VI TX77-BK79

O 222 WR SY97-FU02

O 225 GD SY97-FU02

D 230 IV EN72-VI75

VI VI75-TX77

D 244 VL TX77-BK79 HK68 (I/V), EN72 (L/V), TX77 (L/V)

E 260 MI VI75-TX77 BE92 (I/L)

E 262 TN BE89-BE92 TX77 (K/N), BK79 (K/N), SI87 (K/N), BE89 (I/T), BE92 (K/N), WU95 (N/S)C 276 NK WU95-SY97 HK68 (A/T), BE92 (I/N/T), WU95 (N/T)

C 278 IS EN72-VI75 HK68 (I/V), BE89 (N/S), BE92 (K/N/S), WU95 (N/S)

HK68 (I/K/R/S), EN72 (I/N/R/S), VI75 (K/N), BE92 (K/N), WU95 (K/N), SY97 (K/N)

HK68 (I/K/V), EN72 (I/M), VI75 (I/K), TX77 (I/K), SI87 (E/K), BE92 (E/K)

HK68 (D/N), BK79 (N/S), BE89 (N/S), BE92 (D/N), WU95 (D/N)

Importance of genetic background, i.e.

context-dependency

Influential sites

Pair

wis

e n

ucle

oti

de

diff

ere

nces in

HA

1

Boom-and-bust of genetic diversity empirically supported

Observed pattern in genetic diversity

Diversification within clusters cannot be rejectedunder the null, neutral model of random speciation.

Observations of tree balance

0.1

Conclusions

• An alternative, empirically-supported model of influenza’s hemagglutinin evolution can account for both H3N2’s dynamic and the phylogenetic patterns of its HA1.

• Incorporating appropriate genotype-phenotype maps for the effect of mutations at the phenotypic level may be important for understanding pathogen evolution.

Acknowledgments

David Alonso, Stefano Allesina, Luis Chaves, Diego Moreno, Aaron King

Center for the Study of Complex Systems

NSF graduate student fellowship (S.C.)McDonnell Foundation (Centennial Fellowship to M.P.)

Derek Smith, Ron Fouchier, Sharon Greene, Cecile Viboud, Maciej Boni

Jamie Lloyd-Smith, Igor Volkov, Mary PossCIDD postdoctoral fellowship (K.K.)

Patterns of influenza phylodynamics (H3N2)

1. Annual outbreaks

2. Genetic drift

3. Genetic change Antigenic change

Fitch et al. (1997)

Greene et al. (2006)

Smith et al. (2004)

Patterns of genetic diversity

Characteristics of Influenza Evolution

Sequential replacement of clusters

Season

Clu

ste

r #

Antigenic clusters

Smith et al., Science (2004)

Characteristics of Influenza Evolution

Gradual genetic change Punctuated antigenic change

Smith et al., Science (2004)

Ge

net

ic d

ista

nce

fro

m 1

968

str

ain

An

tigen

ic d

ista

nce

fro

m 1

968

str

ain