Beauveria bassiana as an Artificial Endophyte in Tissue-cultured Banana Plants: a Novel Way to Combat the Banana Weevil Cosmopolites sordidus
J. Akello, T. Dubois and D. Coyne
International Institute of Tropical Agriculture, Kampala, Uganda
Soil Ecosystem, Phytopathology and Nematology, University of Bonn, Bonn, Germany
Keywords: colonisation, endophytic, entomopathogen, inoculation, Musa, tissue culture
Beauveria bassiana, which is effective against a variety of insect pests, is the most
researched and commercialised fungal biopesticide. Laboratory and screenhouse studies have revealed great potential of this entomopathogenic fungus for use against the banana weevil, Cosmopolites sordidus, in banana. However, impractical field
delivery methods and high costs associated with the application of B. bassiana against C.
sordidus prevent its use and commercialisation in banana fields. Our research has revealed that B. bassiana can colonise internal banana tissues for at least four months after tissue-cultured plantlets are dipped in a spore suspension. The type of banana cultivar did not affect colonisation by Beauveria bassiana and, even when elevated B.
bassiana doses were used, plant growth was not reduced. In a set of three screenhouse experiments, larval mycosis rates in B. bassiana-treated plants were 23.5-88.9% and the
presence of the fungus inside treated plants led to a reduction in larval damage of up to >50%. Application of B. bassiana as an artificial endophyte inside banana plants
could circumvent bottlenecks associated with its application as a conventional biopesticide, because i) it kills the damaging larval stages inside the plant, ii) it is protected from adverse biotic and abiotic factors, iii) little inoculum is required, drastically reducing its cost, and iv) farmers do not need to apply the biological control organism themselves, as the technology is easily transferable to a commercial tissue culture producer.
Bananas and plantains (Musa spp.) are the key components in food security and
agricultural sustainability, and a source of income to the resource poor farmers around the East African highland region (Karamura, 1993). However, the production of bananas and plantains in East Africa has been steadily declining due to several constraints, including pests and diseases. The banana weevil Cosmopolites sordidus (Coleoptera: Curculionidae) remains
the primary arthropod pest of bananas and plantains, causing yield losses of up to 100% (Gold et al., 2004; Koppenhöfer et al., 1994). Unfortunately, because the adult weevils are mostly concealed in soil, and the larvae are protected within the banana rhizome and pseudostem, control by conventional insecticides, cultural practices or classical biological control methods has proven challenging and impractical (Treverrow et al., 1993). Whereas factors such as resistance, high cost and environmental pollution have hampered the use of insecticides (Collins et al., 1991; Gold et al., 1999) cultural practices, including trapping, good crop husbandry and clean planting materials are labour-intensive (Gold et al., 2001; Masanza et al., 2005). Furthermore, attempts for biological control using exotic natural enemies have had only limited impact (Koppenhöfer and Schmutterer, 1993), while few banana or plantain cultivars provide tolerance or resistance against C. sordidus (Kiggundu et
MANAGING THE BANANA WEEVIL WITH Beauveria bassiana AS A
Beauveria bassiana is among the most widely used and studied fungal biopesticides against insect pests. It is an entomopathogenic fungus, which is known to be virulent against >200 species of insects (Feng et al., 1994). As a biopesticide, B. bassiana offers an
environmentally safe control measure against insect pests and has been widely used as a biopesticide for their management. Several commercial products based on B. bassiana are
available for managing various species of Curculionidae (Adane et al., 1996; De La Rosa et al., 1997; Rice and Cogburn, 1999). In laboratory bioassays, the fungus was highly pathogenic to C. sordidus, causing more than 90% mortality within two weeks, and this
success prompted scientists to test its performance as a conventional biopesticide under field conditions.
The first attempts for field application of B. bassiana against C. sordidus were by
means of using conidial powders or suspensions. Spraying a fungal suspension at the base of banana mats (Delattre and Jean-Bart, 1978) or planting suckers in soil treated with B.
bassiana (Nankinga, 1994) had no effects on banana weevil density. Schoeman and Botha (2003) performed a stem-baiting technique in which B. bassiana powder was introduced into
two holes that were dug into the rhizomes of flowered plants. However, banana weevil mortality was only evident after the second application, indicating that inoculum of the pathogen had to build up to a critical level in order to have any effect on banana weevil populations.
Monitoring banana weevil populations with pseudostem traps in plots that received several different formulations and delivery systems of B. bassiana revealed low insect
mortality (e.g. often <1%) (Nankinga, 1999). Nevertheless, carrier substrates were identified that could control banana weevil populations in the field. When Nankinga (1999) applied corn (Zea mais) bran containing B. bassiana conidia to the topsoil around banana mats, she
attained infectivity rates of 48% and 20% one month and five months after application, respectively. Godonou et al. (2000) applied an oil palm (Elaeis guineensis) kernel cake-based
formulation of conidial powder to planting holes and suckers and obtained 41% mortality among adults two months after application. Although these infection rates are acceptable, application rates of B. bassiana formulations were estimated at 250-500 kg/ha, which are not logistically or economically feasible for resource-poor farmers.
Combinations of B. bassiana formulations with cultural control practices, such as
pheromone and pseudostem traps, have also been tested. Field control of C. sordidus using B.
bassiana-baited pheromone traps was <13% (Tinzaara, 2005). Immersion of pseudostem traps in a B. bassiana suspension also resulted in low infection rates of 5% (Mesquita, 1988). For farmers, combining B. bassiana formulations with pheromones or pseudostem traps can
be complicated and expensive, and therefore not an attractive option.
The use of B. bassiana as a conventional biopesticide is costly, complicated and
labour-intensive. In the case of C. sordidus control in banana fields, farmers would more
often need to produce and apply the biopesticide formulation themselves, in great quantities, using great manpower and at high financial cost. Moreover, assessment periods during the experiments documented above were short and often spanned only a couple of weeks, further demonstrating the inadequacy of using B. bassiana as a conventional biopesticide for control
of C. sordidus. The lack of success in translating high pathogenicity percentages obtained in the laboratory into high field virulence against C. sordidus may be caused by the impact of
abiotic pressures, such as rainfall, sunlight, relative humidity and soil characteristics, on conidial viability (Hallsworth and Magan, 1999; O’Callaghan et al., 2001; Bruck and Lewis,
2002), necessitating repeated applications. More importantly, all application methods thus far only target the exposed adults and not the damaging larvae, which exist within the rhizome. Hence, a more cost-effective and practical, and a less labour-intensive method for field
application of B. bassiana needs to be developed that targets the damaging stage inside the banana plant.
FUNGAL ENDOPHYTES: NATURAL BIOLOGICAL CONTROL ORGANISMS
INSIDE THE PLANT
Studies conducted at the International Institute of Tropical Agriculture (IITA) indicate that banana plants, as with most other crops, harbor several species of non-pathogenic fungi, called endophytes. Endophytes are organisms that, at some time during their life-cycle, live within plant tissues, yet do not cause any disease symptoms to its host (Petrini, 1991). In many cases, endophytes promote plant fitness, mainly by conferring resistance to biotic and abiotic stresses (Dubois et al., 2006). At IITA, naturally occurring Fusarium spp. strains have
been identified that, when inoculated into tissue culture banana plants, provide protection against banana weevils and other pests (Griesbach, 2000; Dubois et al., 2004; Paparu, 2004). From a commercial point of view, use of endophytes in banana plants is very promising. Research has focused on re-introducing these naturally occurring endophytes into otherwise sterile banana tissue-cultured plants, at low doses and before plants are sold to farmers. Hence, natural equilibrium is restored, providing farmers with endophyte-enhanced clean planting material, which are protected from pests and diseases at the time of field planting. Successful research into use of natural endophytes has spurred parallel research into use of conventional entomopathogens, such as B. bassiana, as artificial endophytes.
BEAUVERIA BASSIANA AS AN ARTIFICIAL ENDOPHYTE
Beauveria bassiana is known to occur naturally in soil and plant residues (Feng et al., 1994). Beauveria bassiana colonisation of cotton (Gossypium spp.), potato (Solanum
tuberosum), jimsonweed (Datura stramonium), and common cocklebur (Xanthium
strumarium) has been documented (Jones, 1994). Recently, research has demonstrated that B.
bassiana can be induced to form an artificial endophytic relationship through inoculation in a variety of crops, such as corn (Bing and Lewis, 1991; Wagner and Lewis, 2000; Cherry et al., 2004), tomato (Lycopersicon esculentum) (Ownley et al., 2004), cocoa (Theobroma cacao)
(Posada and Vega, 2005), coffee (Posada and Vega, 2006), date palm (Phoenix dactylifera)
(Gomez-Vidal et al., 2006) and opium poppy (Papaver somniferum) (Quesada-Moraga et al.,
2006). Some studies investigating endophytic B. bassiana persistence for prolonged periods
of time found that the fungus remained in the plant during the entire growing season (Bing and Lewis, 1991). In many instances, endophytic colonisation was associated with insect and even disease control (Bing and Lewis, 1993; Cherry et al., 2004; Ownley et al., 2004). Bing and Lewis (1991), and Cherry et al. (2004), for instance, reported a significant reduction in Ostrinia nubilalis populations and Sesamia calamistis damage after introducing B. bassiana
in corn plants. In another study, treatment of tomato seeds with B. bassiana provided
seedling protection against damping-off disease (Ownley et al., 2004).
BEAUVERIA BASSIANA AS AN ENDOPHYTE AGAINST THE BANANA WEEVIL
Inoculation of Tissue-cultured Banana Plants with Beauveria bassiana
At IITA, research into endophytic use of B. bassiana began with the assessment of
various potential inoculation techniques. In Africa, banana is increasingly produced through micro-propagation in private, specialized tissue culture laboratories. After removal of the plantlets from the laboratory, these tissue-cultured plants are sold to farmers following a hardening period in a screenhouse. Our efforts have focused on colonising plantlets prior to or during the hardening stage. The highly popular East African highland banana cultivars (cv.) Kibuzi and Mpologoma (genomic group AAA-EA) were used throughout. Initial inoculation methods tested on cv. Kibuzi included root and rhizome dipping in a B. bassiana conidial
10 suspension (1.5 h in 300 mL of 1.5 × 10conidia/mL), the use of a solid substrate colonized
by B. bassiana (1% (w/v)) and plant injection with a B. bassiana conidial suspension (1 mL 8of 10 conidia/mL). Three Ugandan B. bassiana strains (G41, S204 and WA) were selected
on the basis of high virulence against C. sordidus, high sporulation and origin of isolation.
One month after inoculation, the highest colonisation was achieved with the dip method (33.3-78.7% in the rhizome) and the injection method (31.5-67.4% in the rhizome). However, the injection method induced relatively high plant mortality and reduced plant growth, whereas the dip method did not negatively affect plant fitness. Beauveria bassiana strain G41
achieved the highest percentage colonisation, especially when the dip method was used (Akello et al., 2007). Based on this study, we adopted the dip method using B. bassiana strain
G41 for highest colonisation of tissue-cultured banana plants.
Assuming that in planta virulence against C. sordidus is related to B. bassiana
colonisation levels, we worked towards optimising inoculum dose and dipping duration. On the other hand, for higher inoculum doses and longer dipping durations, a negative fitness cost to the host plant may occur. In an attempt to optimise plant colonisation by this fungus, 811the effects of B. bassiana doses (from 1.5 × 10 to 1.5 × 10 conidia/mL) and dipping
durations (1, 2 and 4 h) on colonisation and growth of tissue-cultured banana plantlets were investigated. The optimal dose and dipping duration for effective colonisation of tissue-910 cultured banana plants varied depending on plant part, but ranged from 1.5 × 10-1.5 × 10
conidia/mL for 2-4 h. Interestingly, endophytic colonisation of tissue-cultured banana plants by B. bassiana after dipping created no negative impact on plant growth, even when plants were inoculated at the highest dose (Akello, 2007).
In all of the above experiments, percentage root and rhizome colonisation was consistently higher than pseudostem base colonisation. In persistence studies in the screenhouse, B. bassiana strain G41 successfully colonised banana plant tissues and persisted for at least 16 weeks after inoculation. After 16 weeks, colonisation averaged 43.5-49.6%, 0.0-34.4% and 0.0-13.3% in the roots, rhizomes and pseudostems, respectively (Akello, 2007). The prolonged presence of B. bassiana in banana tissues may be crucial, especially if
antagonism of endophytic B. bassiana is mediated through direct parasitism. Bing and Lewis