Specialization as a joint evolution of host choice and local adaptation

Specialization as a joint evolution

of host choice and local

adaptation

V. Ravigné (CIRAD Montpellier)

I. Olivieri (ISEM Montpellier)

U. Dieckmann (IIASA Laxenburg)

Speciation and specialization in parasites

Host diversity may be an important source of disruptive selection in parasites

Speciation requires a source of disruptive selection

When does host diversity favor specialization ?

When does specialization associates with parasite diversification ?

Host 1 Host 2

?

Specialist Specialist

Coalition of

specialists Generalist

Host choice and local adaptation

Specialization Host 1 Host 2

?

Specialist Specialist Coalition of specialists Generalist Host choice Preference Behavioral adaptation Picky Opportunist Local adaptation Performance Physiological adaptation Specialist Generalist

Ingredients of a general but simple model of specialisation

• A local adaptation trait p that determine local fitness (fecundity or survival within each host) : w₁(p) and w₂(p)

• Two hosts (1 and 2) and a (clonal) parasite population

• How individuals distribute among hosts is controlled by a host choice trait h (proportion of individuals in host 2 after dispersal stage)

Æ may be the result of complex choice mechanisms • Population density regulation

Three simple life cycles combining dispersal, selection and regulation

Regulation Dispersal Mixing Selection Dempster 1955

Hard selection

Dispersal Regulation Mixing Selection Levene 1953

Soft selection

Dispersal Selection Mixing Regulation Model 3 Common pool Host 1 Host 2 C₁ C₂ ×(1-h) ×h (p) × w × w_{1 2}(p)

Model analysis

Resident parasite population : p, h, N

Mutant parasite population : p_m, h_m, N_m(t)

Regulation Dispersal Mixing Selection Hard selection Dispersal Regulation Mixing Selection Soft selection Dispersal Selection Mixing Regulation Model 3 C₁ C₂ m 1 m m 2 m , m m 1 2 (1 ) ( ) ( ) ( , ) ln (1 ) ( ) ( ) p h h w p h w p s p h h w p hw p ⎛ − + ⎞ = ⎜ _{− +} ⎟ ⎝ ⎠ 1 1 m m 2 2 m m , m m 1 1 2 2 1 1 2 2 ( ) 1 ( ) ( , ) ln ( ) ( ) 1 ( ) ( ) p h c w p h c w p h s p h c w p c w p h c w p c w p h ⎛ − ⎞ = _⎜ + _⎟ + − + ⎝ ⎠

(

)

(

m

)

1 m m 2 m , m m 1 2 1 2 1 ( ) _{( )} ( , ) ln 1 ( ) ( ) p h h w p _{h w p} s p h c c h w p hw p ⎛ − ⎞ = _⎜⎜ + _⎟⎟ − ⎝ ⎠

Evolution of local adaptation - Something trivial

Local adaptation trait p

Local fitness w₁(p) Optimal phenotype in host 1 Optimal phenotype in both hosts w₂(p)

Stabilizing selection toward optimum – no selection for

specialization

Local adaptation trait p

Local fitness

w₂(p)

Optimal phenotype in

host 1

Trade-off in local adaptation to both hosts – there exists a

singular strategy

w₁(p)

Optimal phenotype in

Evolution of local adaptation – Soft selection – Constant host output W₁ W₂ p Local fitness Time Mean phenotype p W₁ W₂ p Local fitness Time Mean phenotype p

p

* Symmetric 0.25 0.5 0.75 1 W₁ W₂ p Local fitness Time Mean phenotype p 0

Decreasing trade-off strength Evolutionary attractor

Branching point

Evolution of local adaptation – Hard selection – Variable host output W₁ W₂ p Local fitness Time Mean phenotype p

p

* Symmetric case 0.25 0.5 0.75 1 W₁ W₂ p Local fitness Time Mean phenotype p 0

Decreasing trade-off strength Evolutionary attractor Branching point Evolutionary repellor W₁ W₂ p Local fitness Time Mean phenotype p

Decreasing trade-off strength

Local adaptation trait,

p

* C onstant output Variable output Symmetric distribution of individuals among habitats

Biased distribution of individuals among habitats 0 0.5 1 1.5 0 0.5 1 1.5 0.25 0.5 0.75 1 0.25 0.5 0.75 1 0 Evolutionary attractor Branching point Evolutionary repellor Convex trade-off Concave trade-off Convex trade-off Concave trade-off

w₂( p ) w₁( p ) w₂( p ) w₁( p ) w₂( p ) w₁( p ) Local adaptation trade-off

Concave trade -off

Slightly convex trade-off

Very convex trade-off

Evolution of local adaptation under fixed and unconditional habitat choice Constant host output

(Soft selection)

Branching

_{2 specialists} Variable host output

(Hard selection and Model 3)

Evolutionary

attractor

1 intermediate local adaptation phenotype

Bistability

1 specialist

Evolution of host choice alone 2

h

∗

=

c

2 2 1 1 2 2

( )

( )

( )

c w p

h

c w p

c w p

∗

₌

+

Local regulation Global regulation

Constant output Variable output

Soft selection Model 3 Hard selection

0

*

=

h

or 1

Joint evolution of host choice and local adaptation

h

1

p

1 0

p

1 0 0

Local regulation Global regulation

Constant output Variable output

Soft selection Model 3 Hard selection

p

• Only two possible outcomes :

• branching and emergence of a coalition of two extreme specialists • bistability and emergence of a single specialist

Joint evolution of local adaptation and host choice Constant

host outputs Local

Branching

_{2 specialists} regulation

(Soft sel.

and Model 3)

Bistability

1 specialist in ideal free distribution

Variable host outputs (Hard selection and Model 3) Global regulation (Hard sel.)

Bistability

1 specialist leaving an empty niche

w₂( p ) w₁( p ) w₂( p ) w₁( p ) w₂( p ) w₁( p ) Local adaptation trade-off

Concave trade -off

Slightly convex trade-off

Very convex trade-off

Evolution of local adaptation under fixed and unconditional host choice Constant host outputs

(Soft Selection)

Branching

_{2 specialists} Variable host outputs

(Hard sel. + Model 3)

Evolutionary

attractor

1 intermediate local adaptation phenotype

Bistability

1 specialist (Soft selection)

D

2

( )

w p

1

( )

w p

h

0 0.5 1 0.5 1 0 0.5 1

Joint evolution of host choice and local adaptation

Evolution through small mutations steps

Æ discrepancy with population genetics analysis of these life cycles - no temporal variability

- no cost of choosiness

- no constraints (e.g., time limitation)

Host choice evolution Æ no generalist

More realistic population dynamics

- are suprisingly well captured by these qualitative

considerations on local vs. global regulation and constant vs. variable output

s

c

₁ 0.5 1

s

0.5 1 0.5 1

c

0.5₁ 1

c

0.5₁ 1

c

0.5₁ 1 Soft selection Model 3 Maintenance of local adaptation polymorphism Emergence of polymorphism through small mutations steps

Random distribution of individuals among habitats Random distribution of individuals among habitats 0 0 0 0 0 0 Matching host choice Evolving host choice 0.2 γ = γ =0.5 γ = 0.9 γ =1.2 Any trade-off convexity w₁(p) p Local fitness w₂(p) s

Next…

Larger mutation steps

₌

Lesser importance of trade-off shape

Importance of the genetic architecture of local adaptation

and host choice

Evolution of the genetic architecture of traits under selection

in heterogeneous environments