W. Tushemereirwe¹, A. Kangire¹, F. Ssekiwoko¹, L.C. Offord², J. Crozier², E. Boa², M. Rutherford² and J.J. Smith²*

¹ National Agricultural Research Organisation, Kawanda, Uganda
² CABI Bioscience, Egham, Surrey, TW20 9TY

*j.smith@cabi.org

Accepted for publication: 05/07/04

In October 2001 a wilting disorder of new aetiology was reported as affecting banana (Musa sp.) within the Mukono district of Uganda. The disorder was characterised by a rapid yellowing and wilting of the younger leaves (Fig. 1), a discolouration of the internal vascular vessels (Fig. 2), occasionally a dieback initiating from the male floral parts (Fig. 3) with internal rotting of banana fruits (Fig. 4). These symptoms were notably distinct from Fusarium wilt (Fusarium oxysporum) and ‘Matooke wilt’ (a wilt-like disorder of unknown aetiology), but strongly resembled Moko disease caused by Ralstonia solanacearum; although this particular pathogen had not previously been recorded on banana in Africa.

Figure 1: Wilted plant showing yellowing of leaves	Figure 2: Pseudostem cross-section showing vascular discolouration

A bacterium was isolated on nutrient agar that was identified by fatty acid (Microbial ID Inc. [MIDI]) and metabolic (Biolog, Inc, Hayward, CA, USA) analyses as Xanthomonas axonopodis [Probability score less than 0.4] and Xanthomonas campestris [ID probability score approx. 0.9], respectively. The presence of the Xanthomonas specific fatty acids 11:0 ISO, 11:0 ISO 3OH and 13:0 ISO 3OH was recorded. Pathogenicity tests on disease-free tissue culture derived banana plantlets by stem inoculation with a bacterial suspension, induced wilt symptoms consistent with field observations after 3 weeks. Re-isolation and identification, as outlined above, confirmed Koch’s postulates. Reference to the literature suggested the bacterium was Xanthomonas campestris pv. musasearum (Yirgou & Bradbury, 1968). However, this bacterium is relatively poorly described and not contained within either the MIDI or Biolog databases. To support the identification, rep-PCR (Louws et al., 1994) using ERIC and BOX primers was performed on the Ugandan banana isolate, cultures of X. campestris pv. musasearum from Ensete and Musa in Ethiopia [IMI 349461, IMI 349986, IMI350025] and other cultures of X. species from Africa. These analyses revealed an identical DNA fingerprint for all isolates from Musa and Ensete, but distinct fingerprints for the isolates from other hosts.

Figure 3: Dieback associated with infection of male flora parts	Figure 4: Rotting of banana bunch

This is the first report of X. campestris pv. musacearum outside of Ethiopia, where it is recorded as a pathogen of ensete and, to a lesser extent, banana. Accordingly, this pathogen has been given the common name of Ensete Bacterial Wilt, though the aptness of this now looks questionable. The risk posed by this new disease record to the contiguous banana plantation of Uganda is undetermined, but significant spread is already being observed. The causative organism for this new disease record of has been deposited within the CABI Genetic Resource Collection as IMI 386970.

Yirgou D, Bradbury JF, 1968. Bacterial wilt of ensete (Ensete ventricosum) incited by Xanthomonas musacearum sp.n. Phytopathology 58, 111-112.

Louws FJ, Fulbright DW, Stephens CT, de Bruijn FJ, 1994. Specific genomic fingerprinting of phytopathogenic Xanthomonas and Pseudomonas pathovars and strains generated with repetitive sequences and PCR. Applied and Environmental Microbiology 60, 2286-95.

The British Society for Plant Pathology