A novel, integrated method for management of witches' broom disease in Cacao in Bahia, Brazil

Abstract

A three-year field study was conducted in Bahia, Brazil to validate several strategies for management of witches' broom disease in cacao caused by Moniliophthora perniciosa. Treatments which were applied alone or in combination included applications of biological control fungus Trichoderma stromaticum, fungicide copper hydroxide and phytosanitary broom removal. When compared with untreated control treatments, higher pod yields and consistently lower pod losses were obtained by alternating fungicide with biocontrol application. Pod losses caused by witches' broom were also reduced by fungicide treatment or by phytosanitary broom removal when applied alone or in combination, however total pod production per tree was consistently low whenever broom removal was used as a management strategy. While application of biocontrol fungus alone was not able to reduce witches' broom on pods, it reduced vegetative broom formation and also increased the number of pod-forming flower cushions. The present study indicates that alternating fungicide copper hydroxide with biocontrol fungus T. stromaticum without expensive phytosanitary broom removal is not only a better disease management strategy in Bahia but also results in better yields and thereby better net economic returns.

Introduction

Cacao production in Brazil has been in decline for a number of years since the outbreak in the late 1980s (Pereira et al., 1990) of witches' broom disease caused by the basidiomycete fungus Moniliophthora perniciosa (=Crinipellis perniciosa Aime and Phillips-Mora, 2005). Since the outbreak, several thousand hectares of once productive cacao farms have been abandoned in Brazil and also in other parts of Latin America (Evans and Prior, 1987, Pereira et al., 1996, Griffith et al., 2003, Hebbar, 2007). The state of Bahia accounts for nearly 60% of the total cocoa production in Brazil. Cacao is planted there under the shade of forest species and under highly humid conditions with year round rainfall (Midlej and Santos, 2007). This environment is ideal for pathogen infection, persistence and rapid spread (Rudgard and Butler, 1987, Andebrhan, 1988;). The basidiospores released by the pathogen infect all meristematic tissues of cacao, including young shoots, flower cushions and pods (Wheeler, 1985). Infected tissues also lose apical dominance and, suffer hypertrophy, resulting in broom formation. Brooms are green and photosynthesize at an early stage, but eventually die and produce basidiocarps containing infective badisiospores on the dry brooms (Wheeler, 1985).

Previous efforts to contain the rapid spread of the disease in Bahia state failed for several reasons, most importantly, the lack of genetic resistance during the initial outbreak of the disease (Pereira et al., 1996). Witches' broom has been equally destructive in neighboring countries, such as Peru, Ecuador and Colombia, where losses range from 30 to 60%, and are often compounded by the presence of another equally-destructive disease, frosty pod rot caused by Moniliophthora roreri (Evans and Prior, 1987, Purdy and Schmidt, 1996, Griffith et al., 2003). Recently in Brazil, there has been a slight improvement in the total production due to deployment of disease tolerant varieties and better understanding of the disease epidemiology (Lopes et al., 2003). Grafting with resistant cultivars is accepted as the quickest method to rehabilitate older farms. Current estimates are that 30% (∼150,000 ha) of the 450,000 ha under cacao production in Bahia has been grafted with new tolerant varieties. However, yields in most farms are still low (350 kg/ha) (IBGE, 2007). Reduction in disease pressure using basic good farming practices such as phytosanitation, fertilization or application of fungicides have been practiced in Brazil for several years (Pereira, 1985a, Pereira et al., 1996, Aitken, 1997). Fungicide application on cacao in Brazil was developed mainly for managing black pod disease caused by Phytophthora spp. (Pereira, 1985b, Laker and Ram, 1991, Aitken, 1997). Increase in costs of copper-based fungicides over the past few years and low yield (300 kg/ha) has made this expensive intervention uneconomical for the vast majority of farmers.

The Integrated Pest Management (IPM) method currently recommended to the cacao farmers by the Brazilian cacao research and extension agency CEPLAC (Comissao Executiva do Plano da Lavoura Cacaueira) include four to six applications of copper fungicides plus two to four phytosanitary broom removals per year and also application of biocontrol fungus Trichoderma stromaticum (Rudgard and Butler, 1987, Aitken, 1997, Almeida et al., 2003). However, the integrated approach has not been adopted widely due to high costs of materials and labor (Aitken, 1997, Midlej and Santos, 2007). Phytosanitation, unless practiced over a wide geographic area, was not effective in controlling the disease (Rudgard and Butler, 1987). Even though 95% of the brooms were removed at an experimental site in Rondonia State in Brazil, 50% of the pods were still infected (Rudgard and Butler, 1987). Therefore, there is a need, especially in low input small farms, to find alternatives, to rationalize the use of expensive chemical fungicides and to reduce the frequency of expensive, labor-intensive phytosanitary broom removal.

Recently, numerous reports have been published on the potential use of biocontrol fungi against witches' broom disease (Costa et al., 1996, Bastos, 1996a, Bastos, 1996b, Krauss and Soberanis, 2001, Bateman et al., 2005, Holmes et al., 2004, Pomella et al., 2007). Biocontrol fungi have been reported to reduce growth, sporulation and spread of cacao pathogens through direct mycoparasitism or through production of active metabolites (Bastos, 1996a, Samuels et al., 2000, Sanogo et al., 2002, Aneja et al., 2005, Degenkolb et al., 2006). Most of these studies have usually consisted of small-scale trials with T. stromaticum strain TVC as the active ingredient, as a part of the IPM strategy (Bastos, 1996b, Krauss and Soberanis, 2001, Hjorth et al., 2003, Hebbar, 2007, Pomella et al., 2007). Based on a recent study on diversity and spread of T. stromaticum (De Souza et al., 2006), two distinct genetic groups of the biocontrol fungus have been isolated from cacao farms in Bahia, one fast growing and similar to the recently introduced commercial (Tricovab) strain TVC (Group II), and the other (Group I) with no previous record of introduction but fastidious and slow growing. The presence of both the non-introduced or naturally-occurring strain (Group I) and the introduced (Group II) strain in a large area in Bahia associated specifically with the diseased witches' broom tissue (pods and brooms) has characteristics of “classical biocontrol agents”. This has been defined by Holmes et al. (2004) as the introduction of a biocontrol agent to a new locale where they did not originate nor occur naturally. This is atypical or unusual for biocontrol agents of fungal pathogens. The naturally-occurring biocontrol fungus can be detected now more frequently than before in cacao farms in Bahia, and not only in areas that were sprayed but also in areas that had not been treated, indicating a possible previous introduction and natural spread of the fungus (De Souza et al., 2006). The Group I strains of T.stromaticum, being fastidious, are yet to be mass produced for field application. At present, the population of the naturally-occurring strains in cacao farms is considered not high enough to reduce the current disease pressure. Therefore, larger-scale production and field application has progressed only with the Group II strains of T. stromaticum. T. stromaticum isolates obtained from the recent survey from Bahia are currently being screened for their biocontrol activity on the brooms (Pomella, unpublished) and strains superior to the original type strain TVC have been obtained (Hjorth et al., 2003, De Souza et al., 2006).

This paper describes results from a three-year field study in a previously abandoned cacao field using current and new approaches for witches' broom control. The objective of this study was to identify a cost-effective and integrated method that could be recommended to cacao farmers in Bahia, Brazil. The following variables were measured (i) the effectiveness of biocontrol fungus T. stromaticum, fungicide copper hydroxide or their combinations to increase total pod yields (number of pods per tree) and reducing pod losses (% of pods infected) due to witches' broom disease, (ii) the effectiveness of the current practice of phytosanitary broom removal on pod losses and yields, (iii) to estimate the treatment effects on total and commercial dry bean yields, (iv) test the effectiveness of the above treatments in reducing broom formation on the branches and on the flower cushions and also stimulate flower cushion formation, the pod forming site; and finally (v) to conduct a cost analysis of the above management practices and estimate the economic returns.