Home | Profile | Achievement | Programmes | Projects | Staffs | Publications | Journals |
Biotech Glossary | Bioinformatics | Lab Protocol | Notes | Malaysia University |

Use of QTLS for Witches’ Broom Resistance in Cocoa Breeding
Dario Ahnert
Universidade Estadual de Santa Cruz, Pro-Reitorlo de Pesquisa e Pos-Graduaçao. Km 16, Rodovia Tllieus-ltabuna, CEP 45650-000, Bahia. Brazil


Witches’ broom disease, caused by Crinipe!lis pemiciosa, is the most devastating cocoa pest in the American continent. The fungus attacks meristematic tissues such as young pods, flower cushions, vegetative shoots, and leaves causing reduced production and damage to the plant. Herein we report the results of a QTL study for resistance to witches’ broom and show how these findings can help to accelerate breeding for resistance. For the study, we used 62 F2 plants derived from the cross ICS I (susceptible parent) and SCA 6 (resistant parent), evaluated for witches’ broom resistance over two years at the CEPEC/CEPLAC research centre in Itabuna, Bahia, and 193 DNA markers (124 RAPO and 6g AFLP). Interval and composite interval mapping models identified one major genomic region (covering 27cM) associated with witches’ broom resistance on the SCA 6 linkage group. These results are in accordance with the pattern of segregation observed in the field for SCA 6 progenies and may be used in a marker-assisted selection process for gene pyramidisation,


Witches’ broom disease, caused by the fungus Crinipellis perniciosa, is one of the most devastating cocoa pests in the American continent threatening the cocoa industry in almost all producing countries in that region. The fungus attacks meristematic tissues such as young pods, flower cushions, vegetative shoots, and leaves causing reduced production and damage to the plant. The disease is responsible for 18% of the global cocoa losses (Van der Vossen 1999) and can reduce yield by more than 50% at the farm level. Disease control is based on the removal of infected tissues from the trees, application of copper fungicides to protect the young pods and use of resistant genotypes. Genetic resistance, the most efficient method of C. perniciosa control, has been used since the first outbreak of the fungus in Ecuador in 1918 (SteIl 1934). According to this author, the Ecuadorian farmers established new plantations with the resistant genotypes called ‘refractarios’. However this has not solved the problem in Ecuador, possibly due to the fact that farmers have used seeds derived from out-crossed ‘refractarios’ and these, according to the literature, are heterozygous genotypes.

The Scavina (SCA) selections are the most important sources of resistance, and have been widely used in the Brazilian cocoa breeding programme (Ahnert and Pires 2000). Other sources of resistance are available (Pires et al. 1999; Reyes 1999; Fonseca at a!. 1999), and these are also being incorporated into the breeding programme. Resistance to C. pemiciosa has a polygenic inheritance but in the case of SCA 6 it has been suggested that only a few genes may be responsible for disease control (Bartley 1977; Pires eta!. 1999). To study the inheritance of SCA 6 resistance for C. perniciosa, we used an F2 population of 82 plants, derived through self-pollination of the clone TSH 516, This clone is a selection derived from the cross SCA 6 X ICS 1. The F2 plants were grown at the CEPEC/CEPL.AC research centre in Itabuna, Bahia, and were six years old when field evaluation was initiated. Evaluation was carried out over two years, 1988 and 1999, by removing vegetative brooms four times per year and counting them. Each F2 plant, the parental clones SCA 6 and ICS 1, and the F1 clone TSH 516 were evaluated.

For these latter three clones, ten plants of each were evaluated in the germplasm collection. DNA was extracted from each F2 plant, from the F1 plant and from the parents. The following molecular markers were collected: Randomly Amplified Polymorphic DNA (RAPO) and Amplified Fragment Length Polymorphism (AFLP). A genetic linkage map was constructed using the software MAPMAKER 3.0 and QTLs were mapped using the software QTL Cartographer.

CEPLAC/CEPEC, Universidade Federal de Viçosa (UEV/BIOAGRO), Universidade Estadual de Santa Cruz (UESC) and Universidade Estadual Norte Fluminense (UENF) carried out the QTL mapping study. The data were used for a Master’s degree thesis by TV. Queiroz at UFV/BIOAGRO, Viçosa, Brazil. Herein, we will discuss the use of SCA 6 QTLs for resistance breeding. The complete manuscript showing the QTL mapping is under preparation.

Results and discussion

Phenotypic data and OTt. mapping
Most of the F2 plants showed the same level of field resistance as SCA 6, with approximately one vegetative broom/plant/removal. A few plants had more than ten brooms/plant/removal, and thus they resembled the susceptible genotype ICS1, which had more than 50 brooms/plant/removal. These findings suggest a trend towards a segregation rate of 3:1, expected for a single dominant gene. They are in accordance with earlier findings that a single dominant allele controls SCA 6 resistance (Bartley 1977).

A total of 193 DNA markers (124 RAPO and 69 AFLP) were mapped along 25 linkage groups covering 1,713cM. One major QTL was mapped on linkage group 11, linked to a SCA 6 marker, and explained approximately 36% of the phenotypic variation for resistance (Queiroz et a!. 2000). A more detailed study of this region using a larger number of F2 plants, candidate resistance genes as markers, and cloning and sequencing may help to reveal further details of the genetics behind SCA 6 resistance. This major SCA 6 QTL, with a dominant pattern of inheritance, can be used for rapid genetic gain in a breeding programme. Due to this feature and the good general combining ability of SCA 6, this clone is already widely used in the Brazilian breeding populations in Bahia and in the Amazon region.

Use of SCA 6 genes for resistance breeding
When SCA 6 and SCA 12 progenies were first tested for resistance to C. perniciosa in Trinidad and Ecuador, SCA’s resistance was not considered to be durable (Bartley 1977). According to this author, initially the progenies had low levels of infection but as the experiment progressed, infection increased. Similar results were observed for the SCA progenies in hybrid trials in the Amazon (State of Rondonia). We believe that one of the major reasons for this is the increase in inoculum pressure in the field resulting from infections on susceptible plants from other crosses in the experiment and in the neighbourhood. It seems that as the level of inoculum gets higher, infection levels increase as well. Another reason may be the evolution of races of the fungus that can overcome the SCA resistance genes.

Among and within SCA progenies, some plants are usually more resistant than others, indicating that more than one gene is involved in resistance. For example, the clone TSR 1188 is one of the most witches’ broom resistant clones in Bahia. It may have the SCA QTL plus some minor genes from CS 1 and IMC 67 that increase resistance.

Studies of witches’ broom resistance in Brazil were initiated in the 1960’s when CEPLAC started field trials in the Amazon to evaluate crosses between clones (F1 hybrids). Soon it became evident that the different hybrid combinations reacted differently to the attack of the fungus. It was observed that progenies derived from SCA 6 and SCA 12, PA 150, MC 67, Cruzeiro do Sul and RB suffered less pod loss and infection than all the other hybrids (Ahnert et a!. 1991). Based on this information, we decided to eliminate all highly susceptible clones from the Amazon cocoa breeding programme, leaving only the more resistant clones for hybrid seed production. So, the cultivars released to farmers from the 1990’s onwards in that region have been a mixture of medium to highly witches’ broom resistant hybrids. This degree of genetic control, associated with phytosanitary removal of brooms in the dry season, has helped to ensure relatively low yield losses caused by witches’ broom in the Amazon.

C. pemiciosa reached the cocoa plantations in Bahia in 1989. This provided the first opportunity for a very good field evaluation for witches’ broom resistance since there were many germplasrn and hybrid trials at CEPEC that were uniformly established in the field with plants being at least eight years old. Initially, with a low to medium level of inoculum pressure, genotypes with different patterns of resistance could be identified (Pires et a!. 1999).

According to Van der Vossen (1999), a cultivar is considered to have durable resistance when it is grown for a long time in an environment considered favourable to the disease and the resistance remains effective. Having this concept in mind, 15 selected resistant clones (mostly SCA 6 derived ones) have been released recently by CEPLAC to growers in Bahia. Local farmers have often selected other resistant plants in their own farms (many apparently also derived from SCA clones), and are currently using these as well as the 15 officially recommended clones to renew their plantations.

Marker assisted breeding
The identification of the major QTL of SCA 6 for resistance and the knowledge that other genes for resistance contribute to increased resistance, opens the possibility of gene pyramidisation via Marker Assisted Selection (MAS). Since the SCA 6 major QTL has such a large effect, the presence of any other resistance genes in the same plant can easily be masked in the phenotypic evaluation. However, if markers can be developed which are closely linked to these other genes, MAS can be used to detect those plants which possess these other resistance genes in addition to the SCA 6 QTL. If suitable markers can be developed, MAS can be carried out at the seedling stage, thus accelerating the breeding process. Seedlings derived from double, triple or higher level crosses from different sources of resistance could be screened for the presence of markers linked to QTL and grafted onto old rootstocks for phenotypic selection. Grafting onto rootstocks would accelerate evaluation and, hence the breeding cycle. MAS would eliminate the need for phenotypic screening in the greenhouse and facilitate gene pyramidisation~

Even though resistance provided by this major SCA 6 QTL may have lost some of its efficiency in Ecuador and Rondonia (Brazil), it is still very effective in Trinidad and in Bahia. In combination with other genes, it could provide durable resistance to local strains of C. perniciosa.


The financial support received for this work from BNB, BIOAGRO, CNPq, CAPES, CEPLAC/CEPEC and UESC is gratefully acknowledged.


Ahnert 0. and J.L. Pires. 2000. Use of the available genetic variability of cocoa in Brazil . Pages 104-113 in Proceedings of Technical Meeting: State of the Knowledge on Mass Production of Genetically Improved Propagules of Cocoa. Ilheus, Bahia, Brazil, October, 1998.
Ahnert 0., P.F,ft Machado, B.G.D. Bartley, R.G. Kobayashi and F,C.O. Silva. 1991.Desenvolvimento e avaliaçao de cacaueiros hibridos para a AmazOnia. In lnforme Tecnico, Belem, PA, Brasil, CEPLAC/CORAM, 1989-90.
Bartley B.G.D. 1977. The status of genetic resistance in cocoa to Crinipellis perniciosa (Stahel) Singer. Pages 57-69 in Proceedings of The International Cocoa Research Conference, Caracas, Venezuela. Cocoa Producers Alliance, Lagos, Nigeria.
Fonseca SEA.. E.F. Neto, P.S.B. Albuquerque and C.M.V.C. Almeida. 1999. A contribuição das pesquisas na AmazOnia brasileira para resistencia as doenças. Pages 103—105 in Proceedings of the International Workshop on the Contribution of Disease Resistance to Cocoa Variety Improvement. November, 1996, Salvador, Bahia, Brazil. INGENIC, Reading, U.K.
Pires J.L., W.R. Monteiro, E.D.M.N. Luz, S.D.V.M. Silva, L.R.M. Pinto, A. Figueira, K.P.Gramacho, U.V. Lopes, P.S.B. Albuquerque, MM. Yamada, D. Ahnert and M.IB. Brugnerotto. 1999. Cocoa breeding for witches broom resistance at CEPEC, Bahia, Brazil. Pages 91—101 in Proceedings of the International Workshop on the Contribution of Disease Resistance to Cocoa Variety Improvement. November 1996. Salvador, Bahia, Brazil. INGENIC, Reading, UK.
Queiroz VT., CT. Guimaraes, D. Ahnert, J.w.a Mota, MG. Pereira, R.F. Daher, V,R.M. Miranda, E.G. Barros and MA. Moreira. 2000. Mapping of major OTLs associated with witches broom resistance in cocoa. Page 227 in Genetics and Molecular Biology, Volume 23 No. 3, Supplement. Program and Abstracts - 46~ Congresso Nacional de Genetica. Aguas de LindOia, São Paulo. Brazil.
Reyes L.C. de. 1999. Reacciones de cultivares de cacao a los principales patogenos presentes en Venezuela. Pages 75-81 in Proceedings of the International Workshop on the Contribution of Disease Resistance to Cocoa Variety Improvement. November, 1996, Salvador, Bahia, Brazil. INGENIC, Reading, UK.
Stell F. 1934. Report on his visit to Equador to study witches’ broom disease (Marasmious pemiciosa ) of cocoa: Trinidad and Tobago. Council paper No. 137. Government Printery, Port-of-Spain, Trinidad. Van der Vossen HAM. 1999. Strategies of variety improvement in cocoa with emphasis on durable resistance. Pages 26-31 in Proceedings of the International Workshop on the Contribution of Disease Resistance to Cocoa Variety Improvement. November, 1996, Salvador, Bahia, Brazil. INGENIC, Reading. UK.