Symposium
e -Xtra*
Cacao Diseases: Important Threats to Chocolate Production Worldwide
Development of a Marker Assisted Selection Program for Cacao R. J. Schnell, D. N. Kuhn, J. S. Brown, C. T. Olano, W. Phillips-Mora, F. M. Amores, and J. C. Motamayor First, second, third, and fourth authors: USDA-ARS Subtropical Horticulture Research Station (SHRS), 13601 Old Cutler Road, Miami, FL 33158; fifth author: Centro Agronómico Tropical de Investigación y Enseñanza (CATIE), Turrialba, Costa Rica; sixth author: Instituto Nacional Autónomo de Investigaciones Agropecuarias (INIAP), Pichilingue, Ecuador; and seventh author: Mars, Inc. c/o USDA-ARS SHRS, 13601 Old Cutler Road, Miami, FL 33158.
ABSTRACT Schnell, R. J., Kuhn, D. N., Brown, J. S., Olano, C. T., Phillips-Mora, W., Amores, F. M., and Motamayor, J. C. 2007. Development of a marker assisted selection program for cacao. Phytopathology 97:1664-1669. Production of cacao in tropical America has been severely affected by fungal pathogens causing diseases known as witches’ broom (WB, caused by Moniliophthora perniciosa), frosty pod (FP, caused by M. roreri) and black pod (BP, caused by Phytophthora spp.). BP is pan-tropical and causes losses in all producing areas. WB is found in South America and parts of the Caribbean, while FP is found in Central America and parts of South America. Together, these diseases were responsible for over 700
Theobroma cacao L. is an understory tree from the Amazon basin whose beans are the basic component of cocoa and chocolate. Cacao is an out-crossing diploid (n = × = 10) with a relatively small genome (17). Much genetic diversity exists within T. cacao; however, current commercial cultivars have a narrow genetic base. The primary types of cacao grown have traditionally been referred to as Forastero, Trinitario, and Criollo. Most of the world’s production is based on Forastero-derived types (12). The Forastero material is usually further divided into upper Amazon and lower Amazon types. The Trinitario types are hybrids between Forastero and Criollo types (20). Three of the major fungal diseases of cacao are witches’ broom (WB) caused by Moniliophthora perniciosa (Stahel) Aime and Phillips-Mora (formerly Crinipellis perniciosa), frosty pod (FP) caused by M. roreri (Cif. and Par.) Evans, and black pod (BP) caused by five species of Phytophthora: P. palmivora (Butl.) Butl., P. citrophthora (R. E. Smith and E. H. Smith) Leonian (1925), P. capsici Leonian (1922) emend. A. Alizadeh and P. H. Tsao (1988); Tsao (1991); Mchau and Coffey (1995), P. megasperma (Drechsler) and P. megakarya (Brasier and Griffin). Breeding for resistance to WB began in the early 1950s using the Scavina (SCA) cultivars SCA6 and SCA12, as donor parents for resistance genes (14). Field breeding experience with these cultivars and their progeny, have revealed several types of resistance mechanisms: canopy, cushion, flower, and pod resistance Corresponding author: R. J. Schnell; E-mail address:
[email protected] * The e-Xtra logo stands for “electronic extra” and indicates that the online version contains supplemental material not included in the print edition. Figure 1 appears in color online. doi:10.1094 / PHYTO-97-12-1664 This article is in the public domain and not copyrightable. It may be freely reprinted with customary crediting of the source. The American Phytopathological Society, 2007.
1664
PHYTOPATHOLOGY
million US dollars in losses in 2001 (4). Commercial cacao production in West Africa and South Asia are not yet affected by WB and FP, but cacao grown in these regions is susceptible to both. With the goal of providing new disease resistant cultivars the USDA-ARS and Mars, Inc. have developed a marker assisted selection (MAS) program. Quantitative trait loci have been identified for resistance to WB, FP, and BP. The potential usefulness of these markers in identifying resistant individuals has been confirmed in an experimental F1 family in Ecuador. Additional keywords: molecular breeding, Theobroma cacao.
(24). Resistance from the Scavina clones is known to be controlled by few genes with major effects (6). Some success has been obtained in selecting for resistance in Trinidad and in Brazil (1). Two different seedling stage screening methods have been developed. The belt inoculation technique developed by Purdy et al. (26) requires sophisticated equipment, a specialized inoculation chamber, and is cumbersome for use in a breeding program. More recently, Surjecto-Maharaj et al. (31) demonstrated that the simple agar block method developed by Sreenivasan (30) was highly effective and reproducible. According to the International Cacao Germplasm Database (ICGD; http://www.icgd.reading.ac.uk/ background.php) 427 clones in various collections contain resistance/tolerance to WB. Breeding for FP resistance is currently underway at the Centro Agronómico Tropical de Investigación y Enseñanza (CATIE), Costa Rica, where ‘UF 273’ and ‘UF 712’, both resistant to FP, have been used as parents, and families are currently under evaluation. The pod inoculation technique developed by PhillipsMora (23) is being used for screening for FP resistance, although to date only four germplasm accessions have been identified in the ICGD as having tolerance to FP. Breeding for resistance to BP has been underway for many years in most cacao improvement programs. Resistance is heritable with narrow sense and broad sense heritability estimates of 0.33 and 0.51, respectively (13). Many sources of resistance to BP have been found and 208 accessions have been identified and listed in the ICGD as having some degree of resistance. A leaf disk inoculation method has been developed and used to screen large numbers of genotypes at the seedling stage (10). A pod inoculation technique is used to evaluate advanced selections. Selection for BP resistance has been successful, with the exception of P. megakarya, which is causing severe losses in West Africa. In 1999, the USDA-ARS initiated a project to apply modern molecular genetic techniques to cacao breeding in collaboration with Mars, Inc. The primary goals of this project are to develop a
marker assisted selection (MAS) program and to disseminate new, productive, disease resistant cultivars of cacao. Development of the MAS breeding program for disease resistance and enhanced productivity. In order to facilitate the development of an MAS program, several steps were necessary. (i) Cooperative projects had to be developed with national agricultural research services (NARS) and international organizations actively working on disease resistance and increased productivity. (ii) Molecular markers had to be developed for DNA fingerprinting, genetic diversity analysis, the production of linkage maps, and quantitative trait loci (QTL) mapping. (iii) Suitable, existing mapping populations had to be identified, allowing the location of QTL for disease resistance, quality, and productivity. (iv) Molecular techniques needed to be applied in the current breeding programs. Specific cooperative agreements with NARS and international institutes. Specific cooperative agreements have secured cooperation with national and international research organizations in Central and South America, West Africa, and South Asia. This cooperation involves screening experimental families for disease reactions, phenotypic evaluation for productivity and quality characteristics, and the release of new improved cultivars. The number of genotypes currently under evaluation in the breeding program is approximately 23,000 (Table 1). Development of molecular markers. Approximately 320 microsatellite markers have been developed by the Centre de Coopération Internationale en Recherche Agronomique pour le Développement (CIRAD), France, and USDA-ARS SHRS Miami (6,19,25). The marker names, primer sequences, annealing temperatures, and repeat type, are listed at (http://iapreview.ars.usda. gov/Main/docs.htm?docid=13040). We have also developed 27 markers from microsatellite-containing expressed sequence tags (ESTs) (2). We estimate that approximately 120 additional useful EST-microsatellite markers could be developed from the current T. cacao unigene set. Candidate gene markers were identified using degenerate polymerase chain reaction (PCR) primers. Resistance gene homologues (RGH) of the nucleotide binding site/leucine rich repeat class (NBS/LRR and Toll/interleukin-1 receptor NBS/LLR) and WRKY transcription factor genes were isolated from cacao by using degenerate PCR primers for a highly conserved region of these candidate genes (3,15,16). Although few of these genes have been correlated with resistance to a specific pathogen, they comprise an important group of candidate genes that can be developed as markers for disease resistance. All markers cannot be mapped in all populations. Each mapping population will segregate for different subsets of these markers. To date, a total of 282 microsatellites, 11 RGH, and 4 WRKY markers have been mapped in our program (5). Linkage mapping and QTL discovery. Lanaud et al. (18) published the first linkage map of cacao using an F1 population produced from mating ‘UPA 402’ with ‘UF 676’. The initial map was produced using 100 F1 progeny trees and data from five isozyme loci, four functional genes, 55 restriction fragment length polymorphisms (RFLP) produced from genomic DNA, and 28 random amplified polymorphic DNA (RAPD) markers. This cross was further saturated by Risterucci et al. (27) with more markers (424) and marker classes as well as 81 additional F1 progeny trees, to produce the first high-density linkage map of cacao (885.4 cM). A more recent map has been reported from this population by Pugh et al. (25), using 135 progeny and codominant markers from past maps, 201 new microsatellite markers and 16 RGH markers for a total of 465 markers. Total genome length of the final integrated map was 783 cM and good colinearity was seen between the two parental maps of this cross, constructed separately, and the integrated map produced by Risterucci et al. (27). Recombination-based map distances between markers were also similar between parental maps of Pugh et al. (25) and the
map by Risterucci et al. (27). QTL have been identified for yield, yield-related agronomic traits, and resistance to black pod disease by Crouzillat et al. (9,10), Flament et al. (11), and Clement et al. (7,8). The first F2 map in cacao was created from 146 cacao trees produced by selfing an F1 progeny (‘TSH516’) of the cross ‘ICS1’ × ‘SCA6’ from the Comissão Executiva do Plano da Lavoura Cacaueira (CEPLAC), Brazil (6). One hundred seventy microsatellite markers were used to generate this map, along with eight RGH and four stress-related WRKY genes, for a total of 182 markers. Joinmap version 3 software (34) was used to create the map, with 10 linkage groups (6). Good colinearity was seen between this map and the codominant map of Pugh et al. (25), with only four small pair-wise inversions of single marker positions. The mean Pearson correlation coefficient between the two maps for all markers in common over all linkage groups was 0.993. Vegetative broom resistance was measured by counting the total number of brooms for 79 trees within the population. Data were taken twice a year for 5 years, from 1997 to 2002. Two QTL for resistance to WB disease were found, one producing a major effect (51.1) and one a minor effect (6.7). Both of these showed important dominance effects, as well as one QTL for trunk diameter (Table 2). Two RGH have been identified as flanking markers: RGH 11 for a QTL for WB resistance on linkage group (LG) 1, and RGH 2 for a QTL for trunk diameter on LG 9. Another cacao map was produced from the cross ‘Pound7’ × ‘UF273’ at CATIE to identify QTL for disease resistance and horticultural traits (5). One hundred eighty loci were analyzed for the entire population of 260 trees, resulting in 10 linkage groups that cover a total of 889.8 cM. Horticultural traits were measured beginning in 1998 and concluding in 2005 and included height of the first jorquette (first point at which trunk develops lateral branches), average trunk diameter growth (measured semi-annually), months from planting to flowering and fruiting, and pod color. Groups of trees (25 to 35 at a time) were inoculated by the method of Phillips-Mora (23) scoring for both internal and external infection of pods. Each tree was inoculated 6 to 10 times using as many fruits as were available at the time (usually one to TABLE 1. Genotypes under evaluation in the USDA-ARS International breeding program Location
Population
Number of genotypes
CATIEw in Costa Rica
INIAPx in Ecuador
CCIy in PNG Almirantez in Brazil USDA in USA All locations w x y z
F1 mapping population ‘Pound7’ × ‘UF273’ The 34-hybrid family trial Two segregating populations for QTL identification Parents of the 34-hybrid family trial Selected clone trial Subtotal Witches’ broom disease plot Old populations under selection Chalmers and Allen collection Subtotal VSD and BP mapping populations Almirante (59 families) Germplasm collection Subtotal Miami (quarantine and nursery) Mayaguez (germplasm and nursery) Subtotal Total
260 2,720 800 68 160 4,008 7,320 4,800 300 12,420 1,600 4,113 379 4,492 250 300 550 23,070
Centro Agronómico Tropical de Investigación y Ensenañza, Turrialba, Costa Rica. Instituto Nacional Autónomo de Investigaciones Agropecuarias, Estación Experimental Tropical Pichilingue (EET), Quevedo, Ecuador. Cocoa and Coconut Institute, Rabaul, Papua New Guinea. Mars, Inc., experimental farm in Itabuna, Bahia, Brazil. Vol. 97, No. 12, 2007
1665
three). For BP, data were obtained using the method described in Crouzillat et al. (10), scoring 10 days postinoculation. Five QTL for FP resistance were found using MapQTL version 5.0 (32). Also found were three QTL for BP resistance with exceptionally high logarithm of odds (LOD) scores compared to other BP QTL found in previous studies (8), two QTL for height of first jorquette formation and one QTL each for average trunk diameter growth and pod color (5). A composite map was created by combining data from the Brazilian F2 population (6), the F1 data from the population grown at CATIE (5), and the original reference population from CIRAD (18), to which additional trees and microsatellite markers had been added (Fig. 1). Original maps were reformed, the information was combined, and the weighted regression technique was used to form this map, carefully removing markers that tended not to link well in the group map, so as to get the most accurate placement of microsatellite and RGH markers over multiple backgrounds. JoinMap version 4 (33) was used for making this map, with additional measures of map quality not available in earlier versions. The homology of microsatellite or RGH markers is generally stable across different clones, and thus QTL linked to specific microsatellite markers or RGH markers are likely to be linked in a relatively stable manner, at least within the current, rather narrow base of commercial clones represented by the parents in these three crosses. The total length of this map was 804.6 cM compared with 885.4 cM in the high-density map of Risterucci et al. (27). The map of Risterucci et al. (27) contains functional genes, RFLP, microsatellite, known telomeric genes, isozyme markers, RAPDs, and amplified fragment length polymorphisms (AFLPs). It is known that different types of markers tend to map to different areas of the genome (18,25). Therefore, our composite map likely leaves small portions of the genome unmarked. However, microsatellite markers are becoming preferred in applied marker projects due to their repeatability and relative ease of use. Our current composite map gives the best estimate of the map position of each microsatellite on it by combining the linkage information from three maps. To date, our composite map contains 282 microsatellite markers, 7 RGH, and 4 WRKY markers, for a total of 293 markers. Linkage groups ranged from 50.8 to 105.1 cM, with an average distance between markers of 2.7
cM. One advantage of the composite map is the ability to identify genes mapped in one population relative to QTL mapped in different populations. An example of this is the position of RGH 7 and RGH 8 that were mapped in the Brazilian F2 and flank a strong QTL (LOD = 14.6) for BP on LG 10 that was identified in the F1 population from CATIE. Future research will apply maximum likelihood and other methods to improve marker placement in dense areas. Use of markers in the breeding program. A major problem in cacao breeding is the inconsistency in the performance of known clones when used as parents in field trials, which is due in part to incorrect genetic identification. It has been estimated that the misidentification of cacao accessions could be as high as 15 to 40% in some major germplasm collections (21). Schnell et al. (29) demonstrated with microsatellite markers that many seedlings from the key cross of ‘SCA6’ × ‘ICS1’ made in Trinidad to select for WB resistance were misidentified. Parentage analysis of 186 trees, assigned 134 as true hybrids and 52 as offtype trees. Correct identification of full-sib families is important in the estimation of heritability, prediction of genetic gain, and identification of superior parents. Pollen contamination and mislabeling of plant families has been a serious problem in cacao improvement programs and in germplasm management. Currently, most genotypes used in the breeding program are verified using microsatellite markers and phenotypic traits. Additionally, a subset of the full-sib families are sampled to determine the error rate in our progeny trials. The major QTL on LG 9 has four loci with alleles significantly associated with resistance to WB. The minor QTL on LG 1 has three loci with alleles significantly associated with resistance to WB. In each case, the ‘SCA6’ parent of ‘TSH516’ contributed the allele associated with the resistant phenotypes (Table 3). To confirm the ability of these markers to differentiate resistant seedlings, we tested the association of the resistant haplotype with resistant progenies in a family from the F1 cross ‘SCA6’ × ‘SIL1’ at INIAP, Ecuador. Preliminary results indicate that plants with the favorable haplotype have a significantly lower mean level of total brooms. Only two markers were segregating in this cross, mTcCIR35 and mTcCIR160, and both of the favorable alleles were from the ‘SCA6’ parent. Using this family of 124 mature plants, brooms were counted twice a year during 2002. Marker
TABLE 2. Quantitative trait loci (QTL) found in each linkage group (LG) and descriptive information for use in selecting progeny
QTL name
First flanking markers
Second flanking marker
Mapping population
More desirable parent
4.5 9.4 9.6 8.6 9.8 8.7 7.3 23.0 13.1
mTcCIR240 mTcCIR240 mTcCIR55 mTcCIR236 mTcCIR26 mTcCIR168 mTcCIR200 mTcCIR61 mTcCIR168
mTcCIR100 mTcCIR100 mTcCIR46 mTcCIR26 mTcCIR26 mTcCIR168 mTcCIR225 mTcCIR37 mTcCIR222
F1 CATIE F1 CATIE F1 CATIE F1 CATIE F1 CATIE F1 CATIE F1 CATIE F1 CATIE F1 CATIE
UF273 UF273 UF273 UF273 UF273 Pound7 UF273 Pound7 Pound7
10.7
19.8
mTcCIR158
mTcCIR18
F1 CATIE
Bothy
3.9
4.5
3.5
mTcCIR9
SHRS2
F1 CATIE
Bothy
4 9
15.08 4.0
253.08z 8.15
N/A 51.1
mTcCIR158 mTcCIR160
mTcCIR107 mTcCIR35
F1 CATIE F2 CEPLAC
N/A SCA6
1
2.6
3.38
6.7
mTcCIR264
RGH11
F2 CEPLAC
SCA6
9
2.6
2.93
15.6
RGH2
mTcCIR178
F2 CEPLAC
SCA6
LG (reference mapw number)
Significance threshold
LODx peak
2 2 7 8 8 4 8 10 4
3.9 3.9 3.9 3.9 3.9 4.1 4.1 4.1 3.9
3.9 7.0 5.1 5.2 7.2 6.7 6.1 14.6 9.6
4
4.0
6
Frosty pod exterior 1 Frosty pod interior 1 Frosty pod exterior 2 Frosty pod exterior 3 Frosty pod interior 3 Black pod 1 Black pod 2 Black pod 3 Average trunk diameter growth Height of first jorquette formation 1 Height of first jorquette formation 2 Pod color Witches’ broom number major Witches’ broom number minor Trunk diameter w Pugh
et al. 2004 (25). Log of the odds of significance of the specific QTL. y Both parents had desirable alleles; ‘UF273’ could have slightly better effect. z Chi-squared value rather than logarithm of odds. x
1666
PHYTOPATHOLOGY
Percent variance explained
Fig. 1. Composite map generated from the F1 family produced by crossing ‘Pound7’ × ‘UF273’ and evaluated for phenotypic characters at CATIE, the F2 family produced by selfing TSH 516 and evaluated for phenotypic characters at CEPLAC, and the F1 family produced by crossing ‘UPA 402’ × ‘UF 676’. Disease resistance and productivity QTL locations indicated by vertical bars on the left of the linkage groups are as follows: frosty pod interior and exterior resistance (FrPod_Int and FrPod_Ext), witches’ broom resistance (WB), black pod resistance (Phtoph), height to first jorquette formation (JOR), fruit number (Fruit), average trunk diameter growth (Ave_gr_rt_trk), trunk diameter (Trunk-Dia) and pod color (Pod_Color). Markers designated CIR were developed at CIRAD. Markers developed at the USDA-ARS are designated RGH, W and SHRSTc for resistance gene homologue, WRKY and microsatellite markers, respectively. Vol. 97, No. 12, 2007
1667
TABLE 3. Use of quantitative trait loci (QTL) information in marker-assisted selectionz QTL
Marker
‘ICS1’
‘SCA6’
‘TSH516’
Resistanceassociated allele
Major QTL on LG 9 flanking markers
mTcCIR 24 mTcCIR 35 mTcCIR 157 mTcCIR 160
184, 184 235, 235 149, 149 288, 288
184, 192 230, 230 149, 153 288, 293
184, 192 230, 235 149, 153 288, 293
192 230 153 293
Minor QTL on LG 1 flanking markers
RGH11 mTcCIR 22 mTcCIR 264
626, 626 296, 286 196, 196
626, 649 278, 282 208, 210
626, 649 282, 286 196, 210
649 282 210
z
QTL were identified in an F2 population grown in CEPLAC, Brazil, and evaluated for vegetative brooms from 1997 to 2002 (6). ‘TSH516’ is an F1 seedling of the cross ‘ICS1’ × ‘SCA6’. For each of the seven loci linked to the witches’ broom QTL, the favorable allele (indicated in bold) is from the ‘SCA6’ parent.
TABLE 4. Number of total brooms among seedlings from a ‘SCA6’ × ‘SIL1’ family segregated by allelic configuration at mTcCIR 35 and mTcCIR 160 n
Witches’ broomy
mTcCIR 35 first allele 228
mTcCIR 35 second allele 230z
mTcCIR 160 first allele 288
mTcCIR 160 second allele 293z
32 23 44 25 124
156 d 108 c 94 b 78 a
0.5 0.5 1 0
0.5 0.5 0 1
0.5 0 0.5 0
0.5 1 0.5 1
Genetic group Group 1 Group 2 Group 3 Group 4 Total y z
Mean separation by least significant difference test, P = 0.05. Favorable allele derived from ‘SCA6’.
analysis was performed using the two microsatellite markers. The plants within the family could be divided into four groups based on the allelic configuration of the two markers, as given in Table 4. Of the 124 plants, 32 were heterozygous for the favorable allele at both loci and they have the highest disease rating (Group 1, 156 brooms). Twenty-three of the plants were heterozygous at mTcCIR37 for the favorable allele but homozygous for the favorable allele at mTcCIR160 (Group 2, 108 brooms) and 44 plants had the reciprocal configuration (Group 3, 94 brooms). Both of these groups were intermediate in their disease reactions. Twenty five of the plants were homozygous for the favorable allele at both loci and these had the lowest total number of brooms (Group 4, 78 brooms). All groups were significantly different based on the analysis of variance using least square means (Table 4). CONCLUSIONS Over 23,000 genotypes are currently under evaluation for disease resistance and enhanced productivity. Many of these families have been verified using a subset of the microsatellite markers. Insuring the integrity of our breeding populations has significantly reduced errors in our phenotypic data. Currently over 100 QTL have been identified and are listed in the CocoaGenDB (http://cocoagendb.cirad.fr). Stability of many QTL has not been well tested over genetic backgrounds, nor over multiple environments. However, one BP QTL identified in the CATIE population co-located with a BP QTL identified by Clement et al. (8), and Ndoumbe et al. (22). Based on the LOD scores, number of years tested, and number of individuals in the families, we speculate that approximately 45 of the QTL estimated could be stable and therefore useful over genetic backgrounds, environments, and years. The major and minor QTL for WB resistance are currently being used for selection. One of the major objectives of our project over the next few years will be the confirmation of QTL stability for productivity, quality, and disease resistance. The current breeding project has been underway for 5 years and significant progress has been made in developing an MAS program (28). The expected outcome from the accelerated cacao MAS is new, precocious, high yielding, disease-resistant cultivars that increase income for farmers while maintaining quality attributes required by the confectionary industry. 1668
PHYTOPATHOLOGY
LITERATURE CITED 1. Ahnert, D., and J. L. Pires. 2000. Use of the available genetic variability of cocoa in Brazil: State of the knowledge on mass production of genetically improved propagules of cocoa. Proc. Tech. Mtg. 1:104-113. 2. Borrone, J. W., Brown, J. S., Kuhn, D. N., Motamayor, J. C., and Schnell, R. J. 2007. Microsatellite markers developed from Theobroma cacao L. expressed sequence tags. Mol. Ecol. Notes 7:236-239. 3. Borrone, J. W., Kuhn, D. N., and Schnell, R. J. 2004. Isolation, characterization, and development of WRKY genes as useful genetic markers in Theobroma cacao. Theor. Appl. Genet. 109:495-507. 4. Bowers, J. H., Bailey, B. A., Hebbar, P. K, Sanogo, S., and Lumsden, R. D. 2001. The impact of plant diseases on the world chocolate production. Online. Plant Health Progress doi:10:1094/PHP-2001-0709-01-RV. 5. Brown, J. S., Phillips-Mora, W., Power, E. J., Krol, C., CervantesMartinez, C., Motamayor, J. C., and Schnell, R. J. 2007. Mapping QTL for resistance to frosty pod and black pod diseases and horticultural traits in Theobroma cacao. Crop Sci. (In press). 6. Brown, J. S., Schnell, R. J., Motamayor, J. C., Lopes, U., Kuhn, D. N., and Borrone, J. W. 2005. Resistance gene mapping for witches’ broom disease in Theobroma cacao L. in an F2 population using microsatellite markers and candidate genes. J. Am. Soc. Hortic. Sci. 130:366-373. 7. Clement, D., Risterucci, A. M., Grivet, L., Motamayor, J. C., N’Goran, J., and Lanaud, C. 2003. Mapping quantitative trait loci for bean traits and ovule number in Theobroma cacao L. Genome 46:103-111. 8. Clement, D., Risterucci, A. M., Grivet, L., Motamayor, J. C., N’Goran, J., and Lanaud, C. 2003. Mapping QTL for yield components, vigor and resistance to Phytophthora palmivora in Theobroma cacao L. Genome 46:204-212. 9. Crouzillat, D., Menard, B., Mora, A., Phillip, W., and Petiard, V. 2000. Quantitative trait analysis in Theobroma cacao using molecular markers. Yield QTL detection and stability over 15 years. Euphytica 114:13-23. 10. Crouzillat, D., Phillips, W., Fritz, P., and Petiard, V. 2000. Quantitative trait loci analysis in Theobroma cacao using molecular markers. Inheritance of polygenic resistance to Phytophthora palmivora in two related cacao populations. Euphytica 114:23-36. 11. Flament, M. H., Kebe, I., Clement, D., Piereti, I., Risterucci, A. M., N’Goran, J. A. K., Cilas, C., Despreaux, D., and Lanuad, C. 2001. Genetic mapping of resistance factors to Phytophthora palmivori in cacao. Genome 44:79-85. 12. Hunter, R. J. 1990. The status of Cacao (Theobroma cacao, Sterculiaceae) in the western hemisphere. Econ. Bot. 44:425-439. 13. Iwaro, A. D., Thevenin, J. M., Butler, D. R., and Eskes, A. B. 2005. Usefulness of the detached pod test for assessment of cacao resistance to Phytophthora pod rot. Eur. J. Plant Pathol. 113:173-182. 14. Johnson, E. S., O’Conner, C. B., Sreenivasan, T. N., and Schnell, R. J. 2003. Population structure of the witches’ broom pathogen of cacao in Trinidad and Tobago. Abstract no. 126 E in: Proc. Intl. Cocoa Research Conf. 14. Cocoa Producers’ Alliance, Lagos, Nigeria.
15. Kuhn, D. N., Heath, M., Wisser, R. J., Meerow, A., Brown, J. S., Lopes, U., and Schnell, R. J. 2003. Resistance gene homologues in Theobroma cacao as useful genetic markers. Theor. Appl. Genet. 107:191-202. 16. Kuhn, D. N., Narasimham, G., Nakamura, K., Brown, J. S., Schnell, R. J., and Meerow, A. W. 2006. Identification of cacao TIR-NBS-LRR resistance gene homologues and their use as genetic markers. J. Am. Soc. Hortic. Sci. 131:806-813. 17. Lanaud, C., Hammon, C. P., and Duperray, C. 1992. Estimation of the nuclear DNA content of Theobroma cacao by flow cytometry. Cafe Cacao 36:3-8. 18. Lanaud, C., Risterucci, A. M., N’Goran, J. A. K., Clement, D., Flament, M. H., Lauent, V., and Flaque, M. 1995. A genetic linkage map of Theobroma cacao L. Theor. Appl. Genet. 107:191-202. 19. Lanaud, C., Risterucci, A. M., Pieretti, I., Falque, M., Bouet, A., and Lagoda, P. J. L. 1999. Isolation and characterization of microsatellites in Theobroma cacao L. Mol. Ecol. 8:2141-2152. 20. Motamayor, J. C., Risterucci, A. M., Heath, M., and Lanaud, C. 2003. Cacao domestication II: Progenitor germplasm of the Trinitario cacao cultivar. Heredity 91:322-330. 21. Motilal, L. and Butler, D. 2003. Verification in global cacao germplasm collections. Genet. Res. Crop Evol. 50:799-807. 22. Ndoumbe, M., Bieysse, D., and Cilas, C. 2001. Multi-trait selection in a diallel crossing scheme of cocoa. Plant Breeding 120:365-367. 23. Phillips-Mora, W. 1996. Studies at CATIE on moniliasis resistance (Moniliophthora roreri [Cif. & Par.] Evans et al.). Pages 111-117 in: Proc. International Workshop on the Contribution of Disease Resistance to Cocoa Variety Improvement. INGENIC, Bahia, Brazil. 24. Pires, J. L., Monteiro, W. R., Luz, E. D. M. N., Silva, S. D. V. M., Pinto, L. R. M., Figueria, A., Ahnert, D. E., and Brugnerotto, M. I. B. 1999. Cocoa Breeding for Witches’ Broom Resistance at CEPEC, Bahia, Brazil. Pages 91-101 in: Proc. International Workshop on the Contribution of Disease Resistance to Cocoa Variety Improvement. INGENIC, Bahia, Brazil. 25. Pugh, T., Fount, A. M., Brottier, P., Abouladze, M., Deletrez, C., Courtois, B., Clement, D., Larmande, P., N’Goran, J. A. K., and Lanaud, C. 2004. A
26. 27. 28.
29.
30.
31.
32. 33. 34.
new cacao linkage map based on codominant markers: Development and integration of 201 new microsatellite markers. Theor. Appl. Genet. 108:1151-1161. Purdy, L. H., Schmidt, R. A., Dickstein, E. R., and Frias, G. A. 1997. An automated system for screening Theobroma cacao for resistance to witches’ broom. Agrotropica 9:119-126. Risterucci, A. M., Grivet, L., N'Goran, J. A. K., Pieretti, I., Flament, M. H., and Lanaud, C. 2000. A high-density linkage map of Theobroma cacao. Theor. Appl. Genet. 101:948-955. Schnell, R. J., Brown, J. S., Kuhn, D. N., Cervantes-Martinez, C., Borrone, J. W., Olano, C. T., Motamayor, J. C., Phillips, W., Johnson, E., Monteverde-Penso, E. J., Amores, F., and Lopes, U. 2007. Current challenges of tropical tree crop improvement: Integrating genomics into an applied cacao breeding program. Acta Hortic. 738:129-144. Schnell, R. J., Heath, M. A., Johnson, E. S., Brown, J. S., Olano, C. T., and Motamayor, J. C. 2004. Frequency of off-type progeny among the original ICS1 × SCA6 reciprocal families made for selection for disease resistance in Trinidad. INGENIC Newsl. 9:34-39. Sreenivasan, T. N. 1986. A new procedure for selection of cocoa progeny resistant to witches’ broom disease caused by Crinipellis periciosa. Cocoa Research Unit Report for 1984–1986. Cocoa Research Unit, 40-1, St. Augustine, Trinidad. Surujdeo-Maharaj, S., Umaharan, P., Butler, D. R., and Sreenivasan, T. N. 2003. An optimized screening method for identifying levels of resistance to Crinipellis perniciosa in cacao (Theobroma cacao). Plant Pathol. 52:464-475. Van Ooijen, J. W. 2004. MapQTL 5, Software for the Mapping of quantitative trait loci in experimental populations. Kyazma B.V., Wageningen, the Netherlands. Van Ooijen, J. W. 2006. JoinMap Version 4.0, Software for the calculation of genetic linkage maps. Kyazma B.V., Wageningen, the Netherlands. Van Ooijen, J. W., and Voorrips, R. E. 2001. JoinMap Version 3.0, Software for the calculation of genetic linkage maps. Plant Res. Int., Wageningen, the Netherlands.
Vol. 97, No. 12, 2007
1669
