Get documents about "Copy of Poster_Barnaud_V2003
Document Sample
The dynamics of sorghum genetic diversity in
a traditional farming system
Contact: adelinebarnaud@hotmail.com
H. I. JOLY (CIRAD, UMR CEFE, France); E. GARINE (Laboratoire d’Ethnologie et de Sociologie Comparative, France); M.
DEU (CIRAD, UMR PIA, France); D. MCKEY (Université Montpellier II, UMR CEFE, France); C. KASHAH (PO Box Poli,
Cameroon); S. MONNE (PO Box Poli, Cameroon); J. BOLTEU (Sodecoton, Cameroon); E. OUIN KOÏDA, (Sodecoton,
Cameroon); G. TRIGUEROS (CIRAD, UMR CEFE,France); A. BARNAUD (CIRAD, UMR CEFE, France; Centre for Invasion
Biology, University of Stellenbosch, South Africa).
In traditional agroecosystems, farmers grow a large diversity of species and landraces. This diversity is often culturally important, and it may
also lower the risk of crop failure owing to vagaries of climate, diseases, pests, and soil limitations.
Understanding landrace diversity is of interest in the study of evolutionary forces under domestication, and has applications in the design of
programs for the conservation, management and use in breeding programs of genetic resources.
Sorghum, Sorghum bicolor spp. bicolor, is a Our study was conducted among Duupa
main crop throughout semi-arid regions of farmers in Northern Cameroon.
Africa and Asia. Five basic races (bicolor, Cultivation of sorghum is central to
guinea, durra, caudatum, kafir) and 10 Duupa agriculture and society. We
intermediates are recognized. Sorghum is found more than 40 sorghum landraces
a wind-pollinated annual crop that is in a single village, Fields are composed
considered to be predominantly selfing, of several landraces, from 4 to 18.
although outcrossing can reach 0.30. Seeds from different landraces are
mixed in a common bowl before sowing.
The Duupa practice of planting numerous sorghum landraces closely mixed in a field should
lead to extensive gene flow among landraces: how are landraces maintained ?
1 Landraces may not be strongly 2 Unidentified biological barriers 3 Selection by farmers for
differentiated. Individuals of a given (mating system, phenological particular combinations of
landrace might share a few major differences, incompatibility) may morphological traits in the
genes responsible for morphological minimize gene flow among choice may preserve the
distinctiveness, while showing little landraces. identity of landraces.
differentiation at other loci.
We developed a multidisciplinary approach, involving biologists and social scientists, to investigate the dynamics of
genetic diversity of sorghum at the scale of a single village.
Cluster D : 2 landraces
(Fst between them is 0.39) 100
Taux d’allofécondation (%)
Taux d’allofécondation (%)
100
Progenies identication
90
Progenies identication
50
90
by farmers (%)
50 80 80
by farmers (%)
80 80
70 70
70 70 41
50
5050
41
41 50
50 50 60 60
60 60 41
41
41
50 50
50 50
See gooriya 40 40
41 41
41
41 41 50
50
Yatta 40
30
36 40
30
30 30 41
20 20 20
2 3
20 20
8 10 10
10 10
5 0 0
0 0 50 2
3 50
50
51
48
51 48
Cluster C
3
48 2
35
27
27
27 39
48 51
51
intermediates races
51
20
35 48
35 36
20
35
42
42 100
Taux d’allofécondation (%)
53
53 36
Progenies identication
53 27 36 36
53 36 35 35 42
48
90
35 3936
by farmers (%)
20
19
48 80 80
51
51 39 42 20
35 70 70
Taux d’allofécondation (%)
35
100 53 53 51
51 35
51
51 60
2 2 36 20 60
Progenies identication
90 53
53 53
53 50 50
53
80 39 48 42
by farmers (%)
73 80 14
6
40 40
35
70 70 6 35 30 28 30
60 60 35 35 2
35
35
14
Cluster B guinea 20
16
20
50 50 10 10
14 14
landraces
1
40 40 40 14
39 0 0
39
39
48 42 0
30 30 14 9
20
39
39 3939
42
42
5 36 2039
39 4242
48
2 36
20
3+2
20 2 6 20
35 35 39
36 39 42
5 36 48 4242
20
20
10 10 9
9
36
35
35 3
39
3948
12 99 48 3 19 42
36 36
0 0 9
315
3 3 3 19 19 19
20 36
36 20
29 19 3
19 20 36 36 48
20
35 48
15 35 36 36
9 19
3 19
19 19
15 3
19
15
15
36
Methods:
15 36
20 20 19
20
15
19
3 19 15
• Neighbor-joining tree based on 14 microsatellites and 21 landraces
15 3
19 19
15
15
15 3 • Estimates of multilocus outcrossing rate based on progenies analysis (1050 plants
20
15
and 6 microsatellites)
15
Cluster A : guinea landraces
15
• Identification by six farmers of the progeny of several landraces. Plants were sown
15
in a complete random block design in an experimental field. Farmers identified each
20
panicle as belonging to same landrace as the maternal plant, an other landrace or a
100
morphotype unsuitable for sowing (against selection).
Taux d’allofécondation (%)
Results:
Progenies identication
90
80 80
by farmers (%)
70 70 Sorghum landraces have overlapping flowering times.
60 60
50 50
In this study, different factors explain how sorghum landraces are maintained:
40 40
• Cluster A have a common genetic background because of high outcrossing rate
30 30
1+2 20
10
0
15
0
7
20
10
0
• Cluster B have a common genetic background with a lower outcrossing rate
• Cluster C: pollen competition appear explain the pattern of genetic diversity
• Cluster D: see gooriya is cleistogame, floral traits explain thus the pattern of genetic diversity ; selection seems the
primary factors explaining the maintenance of yatta.
Our study demonstrates the unsuspected impact of farmers’ practices on the evolution of
crop populations. Only when farmers’ behaviour is adequately taken into account will we
achieve accurate understanding of the dynamics and the maintenance of diversity in their
crop populations.
Barnaud et al. (2009) A weed-crop complex in sorghum: The dynamics of genetic diversity in a traditional farming system. American Journal of Botany 96: 1-11.
Barnaud et al. (2008) High outcrossing rates in fields with mixed sorghum landraces: how are landraces maintained? Heredity 101: 445-452.
Barnaud et al. (2007) Local genetic diversity of sorghum in a village in northern Cameroon: structure and dynamics of landraces. Theoretical and Applied Genetics 114:
237-248.
Related docs
