More Realistic Projections of Population Responses to Stressors: Incorporating Genetics (SP2 MYP)

Diane Nacci, Jeffrey Markert (ORISE/EPA), Denise Champlin, Jason Grear, Anne Kuhn, Ronglin Wang (EPA ORD-NERL), Mark Bagley (EPA ORD-NERL)

Current population risk assessment methods do not take into account population genetics. However, it is widely acknowledged that chemical stressors can alter the genetic structure of wildlife populations, and that the genetic structure of wildlife populations can affect stressor responses.

• Guidance is needed on how to take advantage of increasingly accessible molecular genetic information to improve projections of wildlife population responses to chemical and non-chemical stressors.

• Major ecological research questions as to how to integrate population genetics and dynamics need advancement.

• Finer scale issues specific to Agency needs, such applicabil ity and logistics, are also required for risk assessments supporting the regulation of chemicals by the US EPA’s Office of Pesticide Programs.

A research program integrating ecologists, geneticists, and toxicologists across US EPA’s Office of Research and Development research laboratories has been developed to advance understanding of population genetic-dynamic interactions in populations under stress. We designed research to take advantage of complementary expertise and capabilit ies, and used theoretical as well as experimental approaches, which include aquatic species of ecological importance and regulatory usefulness.

Agency Problem Conclusions and Future Directions

Conclusions representing Office of Research and Development research laboratory collaborative research and graduate and post-doctoral training are reported in key publications (including several that received the Agency’s Science and Technology Achievement Awards), and include:

• Genetic diversity reduction exacerbates stress responses. Addit ional research is needed to evaluate molecular genetic markers for their usefulness, and validate and generalize relationships predicting stress responses in low diversity populations.

• For some species, responses to mult i-generational chemical exposures include genetic changes that mitigate chemical toxicity. The molecular basis for chemical tolerance and its applicability for predicting variation among species in chemical sensit ivity is currently under investigation.

References

Burnett, K.G., Bain, L.J., Baldwin, W.S., Callard, G.V., Cohen, S., Di Guilio, Evans, D.H., Gomez-Chiarri, M., Hahn, M.E., Hoover, C.A., Karchner, S.I., Katoh, F., MacLatchy , D.L., Marshall, W.S., Meyer, J.N., Nacci, D.E.,Oleksiak, M.F., Feew, B.B., Singer, T.P., Stegeman, J.J., Towle, D.W., Van Veld, P.A., Vogelbein, W.K., Whitehead, A., Winn, R., and Crawford, D.L. 2007. Fundulusas the premier teleostmodel in environmental biology : opportunities for new insights using genomics. Comparative Biochemistry and Physiology : http://dx.doi.org/10.1016/j . cbd.2007.09.001.

Cohen, C.S. (EPA/NRC post-doctoral fellow), Tirindelli, J., Gomez-Chiarri, M., and Nacci, D. 2006. Functional implications of Major histocompatibilityvariation using estuarine fish populations. Integrative & Comparative Biology (in press).

McMillan, A.M. (EPA/NRC post-doctoral fellow), Bagley, M.J., Jackson, S.A., Nacci, D.E. 2006. Genetic diversity and structure of an estuarine fish (Fundulusheteroclitus) indigenous to sites associated with a highly contaminated urban harbor. Ecotoxicology 15: 539 –548.

Nacci, D., Coiro, L., Champlin, D., Jayaraman, S., McKinney, R., Gleason, T., Munns, W.R., Jr., Specker, J., and Cooper, K. 1999. Adaptation of wild fish populations to dioxin-like environmental contamination. Marine Biology 134: 9-17 EPA STAA III.

Nacci, D., Coiro, L., Champlin, D., Jayaraman, S., and McKinney, R. 2002a. Predicting the Occurrence of Genetic Adaptation to Dioxin Like Compounds in Populations of the Estuarine Fish Fundulusheteroclitus. Environmental Toxicology and Chemistry . 21(7): 1525-1532 EPA STAA Honorable Mention.

Nacci, D., Gleason, T., Gutjahr-Gobell, R., Huber, M., and Munns, W.R., Jr. 2002b. Effects of Environmental Stressors on Wildlife Populations: In Coastal and Estuarine Risk Assessment: Risk on the Edge. Editor: Newman, M.C. CRC Press/Lewis Publishers, Washington, DC EPA STAA Honorable Mention.

Nacci, D., Kohan, M., Coiro, L., and George, E. 2002c. Effects of Benzo(a)pyrene Exposure on a PCB-adapted Fish Population. Aquatic Toxicology 57:203-215.

Roark, S.A. (EPA/NNEMS graduate fellow), Nacci, D., Coiro, L., Champlin, D., and Guttman, S.I. 2004. Population genetic structure of a non-migratory marine fish Fundulusheteroclitus across a strong gradient of PCB contamination. Environmental Toxicology and Chemistry 24(3): 717 – 725.

Van Veld, P. A., and Nacci, D. 2007. Chemical Tolerance: acclimation and adaptations to chemical stress. In: The Toxicology of Fishes. Editors: Di Giulio, R.T., and Hinton, D.E. Tay lor and Francis, Washington, DC.

Research Goals

The goal of this program is to demonstrate and test techniques to better understand how genetic alterations modify risks to wildlife populations from chemical and non-chemical stressors. We have focused on two areas, one addressing generalized stress responses and the other specific to our concerns with chemical stressors:

(1) Characterizing predictive relationships between reduced genetic diversity and population sustainability, and evaluating tools (markers and models) for incorporating genetics into population dynamic projections.

(2) Characterizing genetic responses to multi-generational exposures to toxic chemicals, identifying the genetic basis for chemical tolerance, and evaluating its broader applicability in understanding species variation in chemical sensitivities.

Methods, Models, Approach, and Results-to-Date

2) Characterizing genetic responses to multi-generational exposures to toxic chemicals, identifying the genetic basis for chemical tolerance and evaluating its broader applicability in understanding variation among species in chemical sensitivities

Fundulus heteroclitus

• Evolutionary and population theory predicts that given sufficient genetic diversity, species/populations may adapt to mult igenerational stressors, including chemical stressors.

• Although infrequently observed in the wild, ‘evolved chemical tolerance’ can reveal biochemical and genetic adaptive strategies, and suggest mechanisms underlying inter-species variation in chemical sensit ivit ies.

• Wild populations of a non-migratory marine fish species (Fundulus heteroclitus, Atlantic killifish) vary extremely in their sensit ivity to PCBs and other dioxin-like compounds (e.g., Van Veld and Nacci 2007). Some chronically-exposed Fundulus populations display profound adaptive variation in chemical sensit ivity, or rapidly evolved chemical tolerance (Nacciet al. 1999, 2002a).

• A Fundulus population resident to an estuarine PCB superfund site, New Bedford Harbor, MA (USA) , provides a useful model to study genetic changes and ecological implications associated with mult i-generational chemical exposures.

Nacci_SP2GeneticsPeer07FINAL.ppt 9/17/07

1) Characterizing predictive relationships between reduced genetic diversity and population sustainability, and evaluating tools (markers and models) for incorporating genetics into population dynamic projections

• Evolutionary and population genetic theory predict that populations with low levels of genetic diversity are more likely to lack the allelic resources necessary to adapt and persist in novel environments.

• Defining the strength and shape of the relationship between genetic diversity and population fitness will improve predictions of adaptive capacity, i.e., persistence in response to stressors.

• A marine toxicity test species (Americamysis bahia) provides a tractable biological model to construct populations that vary in their genetic diversity.

• Genetically constructed populations provide opportunity to test genetic tools and statist ical methods for estimating genetic risk, and determine empirically the relationship between genetic diversity and adaptive responses to stress.

Americamysis bahia

Counts grouped by salinity and diversity

top row = seawater ,bottom row = low salinity,orang e=diversity level

week

cou

nt

50

100

150

0 5 10

1 2 3 4 5sal

1 2 3 4 5sal

0 5 10

1 2 3 4 5sal

1 2 3 4 5sal

0 5 10

1 2 3 4 5sal

1 2 3 4 5sal

0 5 10

1 2 3 4 5sal

1 2 3 4 5sal

0 5 10

1 2 3 4 5sal

50

100

150

1 2 3 4 5sal

Experimental System

Stress Response x Genetic Diversity

Genetic Diversity Affects Population Growth Rates and Extinction Probabilities

Genetic Basis: Molecular approaches to characterize chemical tolerance

?Variation among fish species

in chemical sensitivity

Normal developmentIn PCB-exposed tolerant fish embryo

Abnormal developmentIn PCB-exposed sensit ive embryo

Genetic Characterization of Embryonic DNA in F. heteroclitus: Quantitative Trait Locus molecu lar/statistical approach

associating molecular genetic markers with tolerance trait

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 2 4 6 8

Lethal [PCB], log scale

Sensit ive(reference)Fundulus

Tolerant (NBH)Fundulus

Fre

quen

cy

Laboratory-Constructed Sensit ive x Tolerant Hybrids

Breeding strategy segregates tolerance traits among progeny, which vary phenotypically upon exposure to PCBs

Genetic Basis

Empirical Relation ship b etwee n Indige nous Contamination and Se nsitivi ty to PCBs for Fun dulu s heterocli tus Pop ulations

0

1

2

3

4

5

6

0 1 2 3 4 5 6

Sediment PCBs, log (ng/g dry+1)

Se

nsit

ivit

y to

PC

B1

26, l

oc

(LC

20+

1)

0

1

2

3

4

5

6

1930 1940 1950 1960 1970 1980 1990 2000 2010

Sedim

ent PCBs

, log10

(ng/g

dry

)

(Nacci et al. 2002a, b)

Adaptive variation in PCB tolerance among resident Fundulus populations

New Bedford Harbor (NBH), MAPCB Superfund Site

Magnitude and duration of NBH contamination,including empirical benchmark (red dash) for adverse ecological effects

Tolerance is adaptive to levels of site contamination Intra- specific tolerance is large relative to inter-specific tolerance

Tolerance is inherited through two generations of lab-rearing, F1 in bold, F2 in light line

NBH

WI (Reference) Measured in NBH

F. heteroclitus Eggs

(Nacci et al. 1999)

(Nacciet al.2002b)

0

20

40

60

80

100

120

0 2 4 6PCB126, log (ng/L)

Em

bryo

nic

and

Larv

al S

urvi

val,

%

WI (Reference)

10

1000

100000

LH BT FH - S RT FM CC LH JM WS NP ZF FH - R

Fish sp ecies te st ed

TC

DD

LC

egg

50 (

pg

/g w

et

we

igh

t)

(Adapted from Elonen et al. 1999 in Nacciet al. 2002a)

WI (Reference): white bar

NB: red bar

Adaptive, profound inherited chemical tolerance in NBH Fundulus

Neither AFLPs (above right, adapted from McMillan et al. 2006) nor allozymes (Roark et al. 2005) nor MHB snps (Cohen 2002) show reduced genetic diversity despite strong phenotypic divergence.

Chemically-adapted Fundulus are relatively fit under unstressed conditions (e.g., normal reproductive output, Nacciet al. 2002b) and stressed conditions (low oxygen or pathogen exposures, Nacciet al. in preparation).

(Cohen 2002)

Few general but some s pecific, predicted trade-offs in NBH Fundulus

Genetic Responses: Rapid adaptation and its ecological implications

• Altered biochemical pathways in NBH Fundulus correlates of chemical tolerance (e.g., Van Veldand Nacci 2007). However, target genes have not been identified.

• A quantitative trait locus (QTL) approach uses classic breeding strategies, taking advantage of variation among Fundulus populations in PCB tolerance, and provides unique opportunity tocharacterize biochemical, genetic and genomic mechanisms for PCB tolerance in this (and other?) fish species.

• Fundulus is becoming an important model species for ecological genomics (Burnett et al., 2007)

Model Species: amenable to laboratory and field studies

Wild Populations

Lab-Constructed Populations

Mono-hybrids1x diversity

Di-hybrids2x diversity

Tri-hybrids3x diversity

Molecular genetic markers

Development and evaluation of molecular markers for A. bahia

Box-whisker plots of log-stochastic growth rate estimated separatelyfor each population (tank) using the discrete-t ime stochastic-logisticpopulation growth model of Dennis and Taper (1994).

1x 2x 6x 8x adm

0.0

0.4

0.8

Pr(θ) 5

2

SeaWater

1x 2x 6x 8x adm

0.0

0.4

0.8

Pr(θ) 5

2

Low Salinity

Box-whisker plots of the probability of Nt passing below N0/100 within 52 weeks (1 year). Probabilit ies are separately estimated by simulation for each population from separately estimated parameters of the discrete-t ime stochastic-logistic model of Dennis and Taper (1994).

Impact and Outcomes

This research expands and refines current population risk assessment capabilit ies for aquatic and wildlife species, including (but not limited to) threatened and endangered species, as needed by US EPA’s Offices of Pesticides Programs and Water. Research reduces uncertainty in ecological risk assessment by providing tools and methods to estimate the magnitude and outcome of genetic and evolutionary changes produced by anthropogenic stressors with respect to population viability, including:

• Molecular genetic markers to acquire genetic information for aquatic species of interest; molecular approaches (i.e., quantitative trait loci, QTL) to identify genetic loci related to fitness under condit ions including chronic chemical exposure.

• Population genetic methods that provide complementary information on the effects of chemical stressors, including long-term exposures, on biological integrity and ecological condit ion

• Improved predictive and diagnostic population modeling approaches that incorporate genetic alterations as a key determinant of population viability.

0

5

10

15

20

25

age-1 age-2 age-3Age class

egg

s pe

r da

y NBHWI

0

0.2

0.4

0.6

0 1 2 3 4 5Sediment PCB concentration,

log ng/g dry weight

Div

ersi

ty in

dex

0

4

8

12

16

Pol

ymor

phic

Loc

i

Nei's h Shannon's I #Polymorphicloci

Adult F1 DO Test: 0.5 mg/L DO

0

20

40

60

80

100

120

0 2 4 6 8 10 12Exposure time, hours

Sur

viva

l, %

NBH WAN LT50SCO

Male (only) Survi val, 10^7 cfu/fish

-200

20406080

100120

JUNW

IJU

NNBJU

LWI

JULN

B

SEPW

ISEP

NB

OCTWI

OCTNB

Month Population

Su

rviv

al, %

NB

0

20

40

60

80

100

120

0 2 4 6PCB126, log ng/L

Em

bryo

nic

and

Larv

al S

urvi

val,

%

0

20

40

60

80

100

0 1 2 3 4 5

Estimated Egg Burden: PCB126, log10(ng/g)

Em

bry

on

ic a

nd

Lar

val

Su

rviv

al, %

0

20

40

60

80

100

120

0 2 4 6 8B(a)P, log (ng/L + 1)

Em

bryo

nic

and

Larv

al S

urvi

val,

%

10

1000

100000

LT BT FH RT FM CC LH J M WS NP ZF FH - R

Fish species te sted

TCDD

LC

egg5

0 (p

g/g

wet

wei

ght)

(Adapted from Elonenet al. 1999, in Van Veldand Nacci 2007)

Some chemical exposures are more toxic to PCB-tolerant fish, (Nacciet al. 2002c)

Linking theoretical and empirical approaches

1x 2x 6x 8x adm

-0.5

1.5

logλ

S

SeaWater

1x 2x 6x 8x adm

-0.5

0.5

1.5

logλ

S

Low Salinity