Irvingia, a Forest Host of the Cocoa Black-Pod Pathogen, Phytophthora Megakarya, in Cameroon

Plant Pathology (2003) 52, 486–490

IrvingiaBlackwell Publishing Ltd. , a forest host of the cocoa black-pod pathogen, Phytophthora megakarya, in Cameroon

K. A. Holmesa*†, H. C. Evansa, S. Wayneb and J. Smithb aCABI Bioscience, Silwood Park, Ascot, Berkshire SL5 7TA; and bCABI Bioscience, Bakeham Lane, Egham, Surrey TW20 9TY, UK

Phytophthora megakarya is an increasingly important invasive pathogen of cocoa in West Africa, where it forms part of the black-pod disease complex together with the more widespread, but less aggressive, P. palmivora. Its purported centre of diversity, based on evidence from molecular studies and mating types, lies on the Cameroon–Nigeria border, from where it has spread throughout both countries and more recently into Ghana and the Ivory Coast. A survey undertaken in an ancient primary forest (Korup National Park) in western Cameroon, adjacent to the Nigerian border, identified Irvingia sp., close to I. gabonensis (Irvingiaceae), as a wild host of P. megakarya, using both morphological and molecular characterization. ITS fingerprinting of two additional Phytophthora isolates, obtained during a supplementary survey in central Cameroon, showed identical banding patterns for the isolate from a cocoa pod, whilst the isolate from an unknown forest fruit proved to belong to P. heveae. This is the first authenticated record of P. heveae from the African continent. Subsequent AFLP analysis grouped the two P. megakarya isolates with other Cameroonian isolates held in the CABI Genetic Resource Collection, but separated them from the Ghanaian isolates.

Keywords: black pod, characterization, cocoa, forest host, Irvingia, Phytophthora megakarya

Gabon (Brasier et al., 1981). It has been suggested that Introduction P. megakarya has moved from an indigenous forest host Cocoa (Theobroma cacao, Sterculiaceae) is an important or hosts onto cocoa and that this was probably a relatively cash crop for smallholder farmers in Central and West recent event, since P. megakarya is still in an invasive Africa. It was introduced into Africa from South America phase, having arrived in Ghana within the last 15 years, in the 19th century, and West Africa now accounts for from where it now poses a threat to the neighbouring 60% of world cocoa production. As an introduced Ivory Coast (Kebe et al., 2002). Crop losses caused by species, it became subject to ‘new-encounter’ attack from P. palmivora, prior to the appearance of P. megakarya indigenous insect pests and pathogens, particularly mirids in Ghana, were estimated at 4·9–13·5% (Dakwa, 1984). The or capsids (Miridae) and Cocoa swollen shoot virus (CSSV), subsequent arrival of the more aggressive P. megakarya which moved from sterculiaceous forest hosts (Johnson, has led to crop losses of 60–100% (Dakwa, 1987; Opoku 1962; Posnette, 1981). One of the most economically et al., 2000). The continued progress of P. megakarya important diseases of cocoa in Central and West Africa is through the cocoa-producing countries of West Africa has black pod caused by Phytophthora megakarya, which is the potential to significantly reduce the world’s cocoa also a new-encounter pathogen of cocoa. production and to impact on resource-poor farmers, leading Phytophthora megakarya was originally identified to socioeconomic, and possibly political as well as ecolog- in Nigeria in 1979 (Brasier & Griffin, 1979). Prior to this, ical, instability (Rice & Greenberg, 2000). It is postulated it was considered to be a local variant of P. palmivora, that the centre of origin of P. megakarya may lie in the which is a ubiquitous pathogen of cocoa globally. P. meg- primary forests of Central and West Africa, since a molecular akarya, in contrast, is known to be restricted in distribu- study identified areas of diversity in P. megakarya which tion to Central and West Africa (Ortiz-Garcia et al., appear to correspond to ancient, Ice-Age forest refugia 1994). In 1981, it was the predominant species in Nigeria, (Nyasse et al., 1999). Both mating types (A1 and A2) but was also present in Cameroon, Fernando Po and were found near the Nigeria–Cameroon border (Nyasse et al., 1999). Korup National Park in western Cameroon, *To whom correspondence should be addressed. which is contiguous with Cross River National Park in Nigeria, is an ancient primary forest which is a centre of †E-mail: [email protected] diversity for Sterculiaceae, and in particular the genus Accepted 3 March 2003 Cola (closely related to Theobroma), one of the dominant

Forest host of Phytophthora megakarya 487 genera in the nearby Mt Cameroon reserve (Cable & mini kit (Qiagen Ltd, Crawley, West Sussex, UK). The Cheek, 1998). DNA was quantified and the concentration standardized An exploratory survey was undertaken in an attempt to 500 ng µL−1, using a GeneQuant pro DNA/RNA to determine if P. megakarya occurs within this primary calculator (Amersham Biosciences UK Ltd, Little Chalfont, forest ecosystem and, if so, to identify its original forest Buckinghamshire, UK). host(s). If successful, the results of this preliminary study may provide the basis for future management strategies ITS-RFLP analysis for confirmation of species based on the principles of classical biological control, in The ITS region was amplified using the primers which coevolved natural enemies from the centre of origin ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′- or diversity of the target invasive pest (weed, arthropod or TCCTCCGCTTATTGATATGC-3′). PCR components pathogen) are considered to offer the most potential as comprised 0·2 mm dNTP (Gibco Life Sciences, Life Tech- biocontrol agents (Greathead, 1995). nologies Ltd, Paisley, UK), 50 pmol each of ITS primer ITS1 and ITS4 (Amersham Biosciences), 5 U of Tth enzyme (from Thermus aquaticus) (HT Biotechnology Materials and methods Ltd, Cambridge, UK) with manufacturer’s buffer (1·5 mm MgCl ); 5–50 ng of DNA was added to a total PCR reac- Collection of isolates 2 tion volume of 50 µL. The PCR reaction was performed A preliminary survey was undertaken in July 2001 in the in a Hybaid Express Thermal Cycler (Hybaid Ltd, Ted- ancient primary forest of Korup National Park, in the dington, Middlesex, UK) with an initial denaturing step western region of Cameroon bordering Nigeria, as well at 94°C for 4 min, followed by 30 cycles at 94°C for 1 min, as in Mbalmayo forest reserve and neighbouring cocoa 55°C for 1 min and 72°C for 1·5 min. The reaction was farms in the central part of the country near Yaoundé. completed by a final stage at 72°C for 10 min. Samples of soil and fruit (fallen and attached) were col- Restriction enzyme digestions were performed with lected from members of the Sterculiaceae, and opportun- Hinf1, Taq1, Msp1 and Alu1 using buffers supplied by istically from the fallen fruit of other forest trees, as well the manufacturer (Gibco Life Sciences). The products of as from cultivated cocoa. Isolations were made from plant digestion were separated by electrophoresis for 3 h in a material directly onto PARPNH-agar selective media 2% LE agarose gel (Flowgen, Ashby de la Zouch, Leices- (Jung et al., 2000) and soil samples from around tree tershire, UK), stained with ethidium bromide and observed bases were seeded onto the selective media. These were under UV light. The sizes of the resulting products were subsequently purified in the UK. estimated against a 50-kb ladder (Gibco Life Sciences) and Additional cultures from the CABI Bioscience Genetic the data entered into www.phytid.org, a web-based Resource Collection (IMI, Egham, Surrey, UK), originat- Phytophthora identification database managed by CABI ing on cocoa, were included in the study for comparative Bioscience and the Scottish Crop Research Institute (SCRI, purposes (see Table 1). All isolates were maintained on Invergowrie, Dundee, UK). 10% V8 agar at 27°C. AFLP analysis AFLP analysis was undertaken on isolates of interest Genetic analysis resulting from the survey and a representative cross- DNA extraction section of isolates held in the CABI Genetic Resource DNA was extracted from cultures grown for 3–4 days Collection (see Table 1). in V8 broth at 27°C with orbital shaking at 200 rpm. The method used was adapted from Vos et al. (1995). Harvested mycelium was pelleted by centrifugation at Restriction digests using 5 U EcoRI and 5 U HpaII 13 800 g for 5 min, washed in TE buffer and then pelleted (Promega UK Ltd, Southampton, Hampshire, UK) were per- again, prior to DNA extraction using the QIAGEN DNA formed on 500 ng of genomic DNA using manufacturer’s

Table 1 Additional Phytophthora isolates IMI accession number Identiﬁcation Origin of Phytophthora isolate recorded from Theobroma cacao in the CABI Genetic Resource Collection, used for AFLP 386150 P. citrophthora Indonesia (North Sulawesi) analysis 386154 P. megakarya Cameroon (Manyu, Ayoke) 386155 P. megakarya Cameroon (Nde, Sangwa) 386158 P. megakarya Cameroon (Haute, Nyone) 386160 P. megakarya Cameroon (Meme) 386162 P. megakarya Cameroon (Meme) 386163 P. megakarya Equatorial Guinea (Bioko) 386165 P. megakarya Gabon (Oyem-Est) 386172 P. megakarya Ghana (Ashanti, Akomodan) 386173 P. megakarya Ghana (Ashanti, Akomodan) 386209 P. megakarya Ghana (Western Region, Essam)

488 K. A. Holmes et al. buffers at 37°C for 4 h, prior to the addition of adaptors to EcoRI (5 pmol µL−1; EcoRAD1, 5′-CTCG- TAGACTGCGTACC-3′; EcoRAD2, 5′-AATTGGTACG- CAGTCTAC-3′) and HpaII (50 pmol µL−1; HpaII AD 1, 5′-GACGATGAGTCCTGAG-3′; HpaII AD 2, 5′-CGCT- CAGGACTCATCGT-3′), 1 U T4 Ligase containing 100 mm ATP and appropriate buffer (Sigma-Genosys Ltd, Pamp- isford, Cambridgeshire, UK) in a reaction volume of 50 µL. Ligation proceeded overnight at 37°C. Preamplifications were performed using EcoAD1 and HpaAD1. The restriction ligation product (5–50 ng) was added to 50 pmol EcoAD1 and 50 pmol HpaAD1, 20 mm dNTP, 1 U Taq polymerase (Sigma) buffered with 1·5 mm µ MgCl2, with a final volume of 25 L. PCR was performed using a Hybaid Express Thermal Cycler operating the following programme: 94°C for 4 min, followed by 15 cycles at 94°C for 0·5 min, 56°C for 1 min and 72°C for 1·5 min. The reaction was ended by a final step at 72°C for 10 min. Preamplified product (5 µL) was then used for selective amplification using Eco-GC (2 pmol) and either Hpa-AC (100 pmol) or Hpa-AG (100 pmol) primers. In each case, primer Eco-GC was labelled with IRD 700 fluorescent dye (MWG-Biotech UK Ltd, Milton Keynes, Buckingham- shire, UK). Selective amplification proceeded in 25-µL reaction volumes comprising one of the above primer combinations, 20 mm dNTPs, 1 U Taq (Sigma) and 1·5 mm Figure 1 Composite gel comparing the ITS fingerprints of Phytophthora MgCl2, using a Hybaid Express Thermal Cycler under the isolates (P. heveae and P. megakarya) from Cameroon. Lanes 1 and 8, ° following parameters: 94 C for 2 min; followed by 13 50-bp ladder; lanes 2 and 3, Alu1 digests; lanes 4 and 5, Taq1 digests; cycles of 94°C for 0·5 min, 65°C for 0·5 min (temperature and lanes 6 and 7, Hinf1 digests of P. heveae and P. megakarya (forest decreasing by 0·7°C in each cycle to a final temperature isolate), respectively. of 56·6°C) and 72°C for 1·5 min; then 23 cycles at 94°C for 0·5 min, 56°C for 0·5 min and 72°C for 1·5 min. The reaction concluded with a final extension step of 72°C obtained in this study, plus representative P. megakarya for 5 min. isolates from West African countries and, as an outgroup, Labelled PCR products were heated at 90°C for 3 min, an isolate of P. citrophthora from Indonesia (see Table 1). prior to loading 1 µL onto a 20% acrylamide gel for This analysis identified a primary cluster at 78% similar- separation of the AFLP fragments using a LI-COR DNA ity for all the P. megakarya isolates, with the additional analyser, GENE REDIR 4200, according to the recommen- Phytophthora spp. forming outlying lineages. Additional ded AFLP protocol supplied by MWG-Biotech UK Ltd. groupings were evident within the P. megakarya cluster AFLP DNA fingerprints were analysed using GelCom- that identified country-specific origins. The isolates pare II (Applied Maths, Kortrij, Belgium). The results obtained from Cameroon in this survey formed a distinct from the two sets of primers were combined, creating a grouping, separated from the other P. megakarya isolates binary matrix for analysis by Dice’s similarity coefficient at the 80% similarity level (Fig. 2). (Dice, 1945) using unweighted pair-grouping by mathe- matical averaging (upgma) (Sokal & Michener, 1958). Discussion The primary objective of this study was to identify the Results original forest host or hosts of the causal agent of cocoa Three Phytophthora isolates were identified from the black-pod disease, P. megakarya. Cola spp. were consid- survey by ITS fingerprinting (Fig. 1): P. heveae (IMI 387088), ered as potential forest hosts, as P. megakarya had been isolated from an unknown fallen fruit in a forest reserve, previously identified on cultivated C. nitida in Cameroon Mbalmayo; P. megakarya (IMI 387089), isolated from a (Nyasse et al., 1999). Although a number of Cola spp. cocoa pod obtained from a cocoa farm, Mbalmayo; and were sampled, P. megakarya could not be isolated from a further P. megakarya isolate (IMI 387090), isolated these in this study. However, Phytophthora spp. were iso- from necrotic areas of fallen, but still green, fruit of lated from additional forest samples, notably from fallen Irvingia sp. from the Korup National Park. fruit of a member of the Irvingiaceae, probably I. gabon- Based on these identifications, genetic diversity was ensis (elephant mango). Molecular analysis of the ITS examined using AFLPs from a selected group of 14 Phy- region confirmed the Phytophthora isolates from Irvingia tophthora isolates comprising the above three isolates and cocoa as being P. megakarya, and that from an

Figure 2 Dendrogram relating AFLP genomic diversity among Phytophthora isolates based on UPGMA and Dice’s coefﬁcient (forest isolate of P. megakarya is indicated by an asterisk).

unknown fruit as being P. heveae. According to the most plantations for the dispersal of P. megakarya, via either recent database (CAB International, 2001), this is the first the forest elephant or humans. These dispersal mechanisms authenticated record of P. heveae from the African could have been the original means by which the cocoa continent. AFLP analysis of the current Cameroonian introduced into West Africa became exposed to P. meg- P. megakarya isolates clustered them with P. megakarya akarya, which was then able to adapt to become an on- isolates from West African cocoa, but showed these iso- tree pathogen because of the cauliflorous habit of cocoa, lates to be somewhat distinct. It was also observed that with those pods on the lower trunk being particularly the P. megakarya isolate from Korup National Park was vulnerable to rain-splashed soil inoculum. distinct from the P. megakarya isolate from a cocoa farm As the purported centre of origin of the pathogen, it in Mbalmayo. From this preliminary survey in Korup would be expected that coevolved natural enemies of P. National Park, it seems likely that P. megakarya is a megakarya would be present in the West African primary soilborne pathogen within the forest with the ability to forest ecosystem. During the exploratory survey reported rapidly colonize recently fallen fruit. In this sense, it is here, a number of antagonistic fungi, such as Clonos- a primary invader or opportunistic pathogen of healthy tachys and Lecanicillium spp., were found associated with dehiscent fruit. P. megakarya in Korup National Park, which severely Cola spp. and Irvingia spp. are utilized by people in hampered the original isolation and subsequent subcul- West Africa for trade and domestic purposes (Ndoye turing of P. megakarya. This raises the possibility of the et al., 1998). The transplanting of Irvingia spp. from the use of classical biological control, through the introduc- forest to cocoa plantations, where they are used for shade tion of coevolved natural enemies, either inoculatively (Shiembo et al., 1996), is a common practice (Ayuk et al., or inundatively, into the invaded or exotic areas (Evans, 1999) which could serve to transfer soil inoculum. The 2002) as a means to halt the advance of P. megakarya. collection of fallen fruit as food could also provide a Phytophthora megakarya was recently reported to have method for dispersal of the pathogen from the forest to been isolated, at very low levels, from the roots of a cocoa plantations. In Korup, the elusive forest elephant, number of shade-tree species in cocoa farms in Ghana which has recently been identified as a distinct species (Opoku et al., 2002). This indicates that P. megakarya has (Roca et al., 2001), eats the fallen Irvingia fruit (H. C. a potentially disparate host range. Future studies will Evans and K. A. Holmes, personal observation), and this focus on determining the wild host range of P. megakarya behaviour has also been reported in Santchou Reserve, and collecting and screening microbial antagonists from Cameroon (Tchamba & Seme, 1993). This means that these forest populations in both Korup National Park and a potential route exists between the forest and cocoa the contiguous Cross River National Park in Nigeria.

Acknowledgements Jung T, Blaschke H, Obwald W, 2000. Involvement of soilborne Phytophthora species in Central European oak decline and These studies were supported with the aid of a grant from the effect of site factors on the disease. Plant Pathology 49, the USDA-ARS and Mars Inc., USA. We thank Mr Moke, 706–18. park ranger, for expert guidance in the Korup National Kebe IB, N’Guessan FK, Keli JZ, Bekon AK, 2002. Cocoa IPM Park and colleagues at IRAD (Institute of Agricultural research and implementation in Cote D’Ivoire. In: Vos J, Research for Development), Yaoundé, Cameroon, for their Neuenschwander P, eds. Proceedings of the West African assistance in facilitating the survey. Regional Cocoa IPM Workshop, November 13–15, 2001, Material was imported into the UK under DEFRA Cotonou, Benin. Newbury, UK: CPL Press, 32–7. licence no. PHL 182/3854 (8/2001). Ndoye O, Perez MR, Eyebe A, 1998. The Markets of Non-Timber Forest Products in the Humid Forest Zone of References Cameroon. Rural Development Forestry Network Paper No. 22c. London, UK: Rural Development Forestry Network, Ayuk ET, Duguma B, Franzel S, Kengue J, Mollet M, Overseas Development Institute. Tiki-Manga T, Zenkeng P, 1999. Use, management and Nyasse S, Grivet L, Risterucci AM, Blaha G, Berry D, Lanaud C, economic potential of Irvingia gabonensis in the humid Despreaux D, 1999. Diversity of Phytophthora megakarya in lowlands of Cameroon. Forest Ecology and Management Central and West Africa revealed by isozyme and RAPD 113, 1–9. markers. Mycological Research 103, 1225–34. Brasier CM, Griffin MJ, 1979. Taxonomy of Phytophthora Opoku IY, Akrofi AY, Appiah AA, 2002. Shade trees are palmivora on cocoa. Transactions of the British Mycological alternative hosts of the cocoa pathogen Phytophthora Society 72, 111–43. megakarya. Crop Protection 21, 629–34. Brasier CM, Griffin MJ, Maddison AC, 1981. The cocoa black Opoku IY, Appiah AA, Akrofi AY, 2000. Phytophthora pod Phytophthoras. In: Gregory PH, Maddison AC, eds. megakarya: a potential threat to the cocoa industry in Epidemiology of Phytophthora on Cocoa in Nigeria: Final Ghana. Ghana Journal of Agricultural Science 33, Report of the International Cocoa Black Pod Research 135–42. Project. Phytopathological Paper No. 25. Kew, UK: Ortiz-Garcia CF, Herail C, Blaha G, 1994. Utilisation des Commonwealth Mycological Institute, 18–30. isozymes entant que marquers pour l’identification specifique CAB International, 2001. Crop Protection Compendium, des Phytophthora responsables de la pourriture brune des Global Module, 3rd edn. Wallingford, UK: CAB cabosses dans les pays producteurs de cacao. Proceedings International. of the XIth International Cocoa Research Conference, Cable S, Cheek M, 1998. The Plants of Mount Cameroon. Kew, Yamoussoukro, Ivory Coast, 1993. Lagos, Nigeria: UK: Royal Botanic Gardens. Cocoa Producers’ Alliance, 135–43. Dakwa JT, 1984. Nationwide black pod survey: joint Posnette AF, 1981. The role of wild hosts in cocoa swollen shoot CRIG/Cocoa Production Division Project. Annual Report, disease. In: Thresh JM, ed. Pests, Pathogens and Vegetation. 1976/77–1978/79. Tafo, Ghana: Cocoa Research Insitute London, UK: Pitman Ltd, 75–8. of Ghana, 263. Rice RA, Greenberg R, 2000. Cacao cultivation and the Dakwa JT, 1987. A serious outbreak of black pod disease in a conservation of biological diversity. Ambio 29, marginal area of Ghana. Proceedings of the Xth International 167–73. Cocoa Research Conference, Santo Domingo, Dominican Roca AL, Georgiadis N, Pecon-Slattery J, O’Brien SJ, 2001. Republic. Lagos, Nigeria: Cocoa Producers’ Alliance, Genetic evidence for two species of elephants in Africa. 447–52. Science 293, 1473–6. Dice LR, 1945. Measurement of the amount of ecologic Shiembo PN, Newton AC, Leakey RRB, 1996. Vegetative association between species. Ecology 26, 297–302. propagation of Irvingia gabonensis, a West African fruit tree. Evans HC, 2002. Invasive neotropical pathogens of tree crops. Forest Ecology and Management 87, 185–92. In: Watling R, Frankland TC, Ainsworth AM, Isaac S, Sokal RR, Michener CD, 1958. A statistical method for Robinson CH, eds. Tropical Mycology 2: Micromycetes. evaluating systematic relationships. University of Kansas Wallingford, UK: CABI Publishing, 135–52. Science Bulletin 28, 1409–38. Greathead DJ, 1995. Benefits and risks of classical biological Tchamba MN, Seme PM, 1993. Diet and feeding behaviour control. In: Hokkanen HMT, Lynch JM, eds. Biological of the forest elephant in the Santchou Reserve, Cameroon. Control: Benefits and Risks. Cambridge, UK: Cambridge African Journal of Ecology 31, 165–71. University Press, 53–63. Vos P, Hogers R, Bluker M, Reijens M, Van der Lee T, Hornes M, Johnson CG, 1962. Capsids: a review of current knowledge. Frijters A, Pot J, Peleman J, Kleiper M, Zabeau M, 1995. In: Wills JB, ed. Agriculture and Land Use in Ghana. Oxford, AFLP: a new technique for DNA fingerprinting. Nucleic UK: Oxford University Press, 316–31. Acids Research 23, 4407–14.