Genetic Diversity of Striga hermonthicaPopulations in ... · REFERENCES • Belay G, A Mori 2006. Biochem Syst Ecol 34:554-561. • Bharathalakshmi, CR Werth, LJ Musselman 1990. Plant

Genetic Diversity of Striga hermonthica Populations in Ethiopia: Evaluating the Role of Geography and

Host Specificity in Shaping Population Structure

Kamal I. Mohamed & Amy B. Welsh

Department of Biological Sciences

State University of New York – Oswego

BACKGROUND

�S. hermonthic is native to Ethiopia & Sudan.

�Obligate outbreeder.

�Widely distributed with host specific strains.

�Most damaging species.

�Represents a potential invasive threat to cereal production worldwide.

RIFT VALEY

�The Rift Valley (2.5 myo) is a potential geographic barrier to the dispersal of S. hermonthica.

�Previous studies have demonstrated its significance in determining population structure in Ethiopia (e.g., Belay and Mori 2006,

Silvestrini et al. 2007, Kebede et al. 2007).

OBJECTIVES

Identify which evolutionary force, host specificity or geography, is playing the greatest role in shaping genetic diversity within S. hermonthica in Ethiopia.

HYPOTHESIS

Geographic barriers to dispersal, specifically the Rift Valley, played a major evolutionary role in genetic differentiation in S. hermonthica populations.

A portion of the Rift valley in the

southern study area

MATERIALS & METHODS

�Sample Collection: 25 samples were randomly

selected from 12 populations in Ethiopia.

� Laboratory Procedures: DNA extracted from 10

samples using Standard CTAP extraction

procedure (Doyle & Doyle 1987, Cullings 1992).

�AFLPs were then analyzed.

�Primer combinations were: EcoACT/MseCTC;

EcoAGC/MseCTC; EcoACC/MseCTC; and

EcoACC/MseCAT (Gethi et al. 2005).

Figure 1. Locations of sampling sites (represented by stars) and hosts from which S. hermonthica were collected in Ethiopia.

STATISTICAL ANALYSIS

� Levels of genetic diversity within a population were measured by calculating % of polymorphic loci and

expected heterozygosity.

� Genetic differences between populations were

measured by FST (Wright 1978).

� The proportion of variance attributable to differences

in host is determined by Analysis of Molecular

Variance (AMOVA).

� Mantel test was used to determine correlation , if any, between geographic and genetic distances.

RESULTS & DISCUSSION

Table 1. Genetic diversity of each population, as measured

by % of polymorphic loci & expected heterozygosity.

60.9 .204

Table 2. Genetic differentiation between the populations, as measured by FST.

All values are significant at p<0.05.

Average = .146

.107

.121

.111

.167

Genetic diversity due to

geography

Geography appears to play the greatest role in determining genetic differences between S. hermonthica populations

Figure 3. Neighbor-joining tree based on Nei’s genetic distance (after Lynch and Milligan 1994). Numbers on the tree represent bootstrap

percentages, based on 1000 replicates.

Figure 4. Results of spatial autocorrelation analysis showing sampling sites and potential genetic barrier (bold line). Numbers along the line represent bootstrap percentages, based on 1000 replicates.

Sorg

hum

Effect of geographic distance

�There appears to be a small isolation-by-

distance effect (R2=0.025, p=0.001).

� (R2=.61) was observed by Botanga et al. (2002)

for S. asiatica.

�This would indicate that the most likely mode

of dispersal is a stepping-stone model, with

those populations geographically proximate

providing the source for colonizers.

Genetic diversity due to host

specificity� Specificity does not appear to be the primary factor

shaping the population structure of S. hermonthica.

� This is consistent with the results for S. hermonthica

by Bharathalakshmi et al. (1990) and Koyama 2000.

� It suggests that specialization of S. hermonthica to its host may be a recent phenomenon with

insufficient time for genetic differences to arise.

� This is consistent with Olivier et al. 1998 observation of host specificity break down in the field.

Table 4. Results of AMOVA analyses.

Rift Valley groups consist of 2 groups: populations east and west.

Host groups consist of 4 groups: sorghum, tef, maize, and millet.

Probability values are based on 999 permutations.

Significant ΦST values (p<0.05) are in bold.

Management implications

�The high genetic diversity of S. hermonthica presents a challenge for the development of resistance.

�Broad genetic background enable S. hermonthica to parasitize multiple hosts, highly invasive.

�Hybridization with non-weedy Strigaspecies provides a gene reservoir.

Management implications

�The abundance and highly variable seed bank complicate management.

�Highly contaminated soils preclude cereal cultivation in some areas.

�Improper practices that promote Strigaseed dispersal must be addressed.

�Effective control of Striga is based on containment and eradication.

Acknowledgement

This study is supported by SUNY - Oswego

REFERENCES• Belay G, A Mori 2006. Biochem Syst Ecol 34:554-561.

• Bharathalakshmi, CR Werth, LJ Musselman 1990. Plant Syst Evol 172:1-12.

• Cullings KW 1992. Mol Ecol 1:233-240.

• Doyle JJ, JL Doyle 1987. Phytochemistry Bulletin 19:11-15.

• Gethi JG, ME Smith, SE Mitchell, S Kresovich 2005. Weed Res 45:64-73.

• Kebede M, D Ehrich, P Taberlet, S Nemomissa, C Brochmann 2007. Mol

Ecol 16:1233-1243.

• Koyama ML 2000. Pages 247–260 in BIG Haussmann et al. (ed.) Breeding for striga resistance in cereals. Margraf Verlag, Weikersheim, Germany.

• Lynch M, BG Milligan 1994. Mol Ecol 3:91-99.

• Olivier A, J-C Glaszmann, C Lanaud, GD Leroux 1998. Plant Syst Evol

209:33-45.

• Silvestrini M, MG Junqueira, AC Favarin, O Guerreiro-Filho, MP Maluf, MB Silvarolla, CA Colombo 2007. Genet Resour Crop Evol 54:1367-1379.

• Wright S 1978 Evolution and the genetics of populations: variability within and among natural populations. University of Chicago Press, Chicago.