Xanthomonas wilt of banana - better known as BXW, and to a lesser extent as BBW (for banana bacterial wilt) - is a lethal bacterial disease caused by the bacterium Xanthomonas campestris pv. musacearum (Xcm). Before 2001, it was found only in Ethiopia, where it affects bananas and its close relative enset (Ensete ventricosum)1. It has since spread to the Great Lakes region of East Africa, where it has caused considerable losses when left uncontrolled. Although no resistant cultivars have been identified, the disease can be controlled by using cultural practices.
Removing the male bud as soon as the fruits have set will prevent transmission of the disease by insects carrying the bacteria-laden exudates produced by infected plants. Male bud removal coupled with using bacteria-free planting material and desinfecting cutting tools before using them on healthy plants will prevent the disease from getting established. In places where the disease is present, removing the diseased stem has been shown to reduce the incidence of new infections to negligible levels when used in combination with the practices that prevent transmission of the bacteria.
Distribution and spread of the disease
Xanthomonas wilt is currently found only in eastern Africa and the northeastern corner of the Democratic Republic of Congo. It was first reported on enset and banana in Ethiopia in 19682, although earlier records report a disease consistent with these symptoms as present in the 1930s3. Spread beyond Ethiopia was not reported until the disease was found in Uganda in 20014. Subsequent spread to other countries in eastern Africa has proceeded rapidly. In many instances, however, the exact time of introduction is not known.
Although it is said to have been first observed in 2001 by farmers in the North Kivu Province, the disease was not confirmed in the Democratic Republic of Congo until 20045. In Rwanda, Xanthomonas wilt was first observed in the northern part of the country in 20056. It was also reported to have spread to Tanzania in 2005, and to Kenya7 and Burundi in 200689.
An analysis of data collected between 2001 and 2006 in Uganda and in six countries of the Great Lakes region in 2007 suggest that the number of newly affected areas declined between 2004 and 2005, and that the spread of the disease was changing from being more or less continuous to more isolated outbreaks10.
Cereals, especially maize, may serve as a reservoir to the bacteria11.
So far, no resistant cultivars have yet been identified and field observations in areas where the bacterium is present suggest that all commonly grown cultivars succumb to the disease. Laboratory experiments on 7 cultivars and a wild species observed differences in susceptibility, with the cultivar 'Kayinja' being the most susceptible, whereas 10% of the potted Musa balbisiana plants that had showed early signs of necrosis, eventually recovered and did not wilt12. Similarly, all the three-month plants from 42 genotypes tested in a pot trial eventually developed the disease, except for the Musa balbisiana plants, from which no bacterial cells were recovered13. In later screenhouse and field trials, Musa balbisiana plants were symptom-free 6 weeks after a single inoculation at dosages that caused disease on the Pisang Awak controls14. Disease symptoms were observed six weeks after a second inoculation, but only at the highest dosages.
Some cultivars, however, possess characteristics (such as persistent male bracts and flowers or no male bud) that make it harder for the bacteria to infect the plants under natural conditions. While these cultivars escape transmission by insect vectors, they are still vulnerable to transmission through cutting tools.
The disease causes loss both through death of the plant and rotting of the fruit. The leaves gradually turn yellow and start looking lifeless as if they were melting under intense heat. They eventually turn brown and die.
In flowering plants, the first symptoms of insect transmission are a drying rot and blackening of the male bud that start with the outer bracts and eventually extend to the rachis. The fruits ripen unevenly and prematurely, turning from green to yellow and black rapidly. The pulp of the rotting fruits shows rusty brown stains.
Internal symptoms revealed by doing a cross-section of an infected pseudostem are yellow-orange streaking of the vascular tissues and the presence of a yellow bacterial ooze, which can also be seen from any other infected plant part.
In the absence of other symptoms, the leaf symptoms of Xanthomonas wilt can be confused with those of Fusarium wilt. In plants affected by Fusarium, yellowing and wilting of the leaves typically progresses from the older to the younger leaves. The wilted leaves may also snap at the petiole and hang down the pseudostem. In plants affected by Xanthomonas, the wilting can begin with any leaf and the infected leaves tend to snap along the leaf blade.
The flower and fruit symptoms are similar to those observed in other bacterial wilts, but no other bacterial wilt of bananas are known to occur in Africa.
Modes of transmission
Insects, farm tools and infected planting material are the main agents of transmission. Their relative importance in spreading the disease depends on the management practices being applied, the type of cultivar and agroecological conditions.
Leaf wetness has also been implicated in disease establishment. Inoculated leaves of 3-month-old tissue-culture plants that were kept wet for 72 hours developed symptoms within 14 days, contrary to the control plants whose leaves were allowed to dry15. The results may explain why a higher percentage of non-flowering 'Pisang Awak' was infected the high altitude region of Masisi in DR Congo (26%) than in the mid-altitude region of Central Uganda (6.1 to 9.3%)15. In the high altitudes, rainfall is evenly distributed throughout the year, as opposed to two rainy seasons interspersed with a dry season at mid-altitudes. High plant densities could also contribute to disease spread when inoculum splashes down from the taller infected plants to the suckers growing below.
Goats and other livestock can carry the bacteria in their mouth and as such spread the disease to healthy plants. Pests such as weevils and nematodes can also help the bacteria in the soil gain entry into the plant through the injuries they make on the root system.
Transmission through insects
Fresh wounds offer bacteria a point through which they can enter or leave the plant. Bacteria have been isolated from the sap and ooze collected from the cushions to which the male flowers were attached and the scars made by the fallen bracts, and to a lesser extent from the nectar of flowers16.
An insect visiting the male bud of an infected plant can get bacteria on its body through a wound that exudes bacteria-laden ooze. The bacteria on the insect's body can then infect healthy plants when the insect visits healthy plants that have similar wounds.
Even though female flowers are more visited than male flowers and the loss of their bract also leaves a scar, experiments suggest that infection only occurs through the cushions of male flowers as no flower infection occurs when the male bud is removed17. It could be because the female bract scars are less numerous and less accessible than the male bract scars.
In Uganda, the bacterium was isolated from stingless bees (Plebeina denoiti and undetermined species of the Apidae family), honeybees (Apis mellifera), fruit flies (Drosphilidae family) and grass flies (Chloropidae family)16.
Cultivars that have persistent bracts and flowers are less likely to contract the disease from flying insects visiting the male bud, although some cultivars whose bracts and flowers do fall off also seem to escape infection. In general, East African highland bananas (EAHB) seem less prone to insect transmission, maybe because their inflorescence is less attractive to insects. The cultivar 'Kayinja', on the other hand, is very prone to floral infection18.
Insect tramsmission is also less frequent at high altitudes, probably because the lower temperatures are not favourable to insect vectors. Male bud infections were rare at altitudes of 1,700m and abov in Ethiopia19 and the Masisi district (1700 m) of DR Congo, where the disease was first observed5.
In 2012, a survey of the insect species visiting male buds in fields located at low, medium and high altitudes was conducted during each of Rwanda's cropping season (short dry, short rainy, long dry and long rainy)20. Seventeen insect species belonging to 12 families were collected and the Xcm bacterium was isolated from the external body parts of representatives of each species. The highest number of insects was recorded in low altitude areas during the long rainy season. The incidence of floral infections declined with increasing altitude, in parallel with insect activity. Insect activity also varied with the type of cultivar, with dessert and beer types attracting more insects than the cooking and multipurpose cultivars. 'Kayinja' (a beer cultivar), 'Kamarasenge' (an AAB dessert cultivar) and 'Injagi' (an EAHB cooking cultivar) attracted the largest number of insects in their respective use group, regardless of altitude. At the other end of the spectrum, not only did 'Nkazikamwe' (an EAHB cooking cultivar), 'Impura' (an EAHB beer cultivar) and 'Ikinyangurube' (an AAA dessert cultivar) attract less insects, they also possess persistent bracts and neutral flowers that interfere with insect transmission.
Transmission through contaminated tools
Cutting tools used by farmers, such as a machete, can be a key mechanism by which the disease is spread in intensively managed cropping systems. Transmission occurs when a farmer uses a cutting tool on an infected plant, where it comes in contact with the bacteria in the sap or ooze, and then on a healthy plant.
The regular harvesting of leaves for sale or other domestic use, which is common in Uganda, can contribute to the spread of the disease through contaminated tools.
Transmission through infected planting materials
The spread, over short and long distances, has been associated with the movement of symptomless but infected suckers for replanting. Latent infections could explain the spread of Xanthomonas wilt in previously disease-free areas by farmers who unwittingly exchanged what they thought were healthy suckers21. Banana plant residues can also spread the disease22.
Even though no resistant cultivars have been identified, experience and research, as well as an understanding of how the bacteria move through the plant, have shown that the impact of Xanthomonas wilt can be mitigated by the adoption of crop management strategies that have proved successful against other banana bacterial wilts, in particular, the timely removal of the male bud to prevent transmission by insects and strict sanitation on the farm to avoid transmission through contaminated tools. In wet and low-altitude areas, it has also been proposed to grow cultivars that have persistent bracts that protect the male bud from insect-transmitted infections23.
The risk of infection through the inflorescence has been shown to be markedly reduced by the timely removal of the male bud (after the last hand has formed) by means of a forked stick (not cutting it off with a knife)17. In disease-free and infected areas, the recommendations is to remove male bud of healthy plants as soon as the last hand of fruit has formed. Using a forked stick not only avoids the risk of moving bacteria around on knives if the plant is infected, but also enables farmers to remove out-of-reach male buds.
Once the disease has been detected in a field,disinfecting tools (pangas, knives, leaf removers, hoes etc.) before using them on other plants would help prevent the spread of the bacteria to healthy plants. The tools can be sterilized by flaming until just too hot to touch (not red-hot) or by soaking in a sodium hypochlorite solution (one cup of household bleach in 5 cups of water).
Removing infected plants
From the early days of the epidemics, farmers were advised to uproot diseased mats, and to dispose of the plant debris, before replanting using clean planting material. (Studies on the survival of bacteria in the soil suggest that it should be safe to replant bananas after six months24.) The practice, however, was not widely adopted, especially among resource-poor households. Since one person can only remove about two mats per day25, removing a large number of diseased mats turned out to be particularly demanding. Moreover, even after destruction of the mat, suckers might emerge, requiring farmers to frequently return to the field to ensure complete destruction of the mats. A 2010 household survey targeting Uganda’s main production systems – the intensively managed EAHB production systems and the less intensively managed ‘Kayinja’ ones – showed that 53% of EAHB farmers and 30% of ‘Kayinja’ farmers uprooted the entire mat26.
Mats can also be destroyed by injecting an herbicide into the pseudostem of the mother plant27. Notwithstanding health and environmental issues (the most effecitve herbicide turns out to be more persistent in the environment28), adoption by farmers has been poor. Some of the possible reasons that have been cited are: inaccessibility of herbicides in rural areas, perceived high cost of herbicides and reluctance to inject infected mother plants as asymptomatic suckers will also die29.
The elimination of an infected mat can also be done gradually by cutting down all the plants at soil level and destroying emerging suckers. The rhizomes are left to rot, a process that can take up to 24 months if no measure is taken to accelerate decomposion, during which time the soil will remain infective for bananas. Ugandan farmers have used the practice to clear heavily infected banana plots for planting annual crops29.
Single diseased stem removal
An alternative to uprooting the entire mat is cutting off the pseudostem at soil level. In the early days of the epidemic, studies on the systemicity of the bacteria suggested that the practice would be effective in preventing the entry of the bacteria in the rhizome, but only if the diseased plant was at the first stage of a flower infection (shrivelled bract) and only for 'Kayinja'30. However, long-term studies have since shown that the bacteria do not go on to systematically colonize all the suckers attached to the rhizome. Moreover, even when the bacteria are present in a sucker, the disease may not develop or develop much later (latent infections), making the single diseased stem removal (SDSR) technique a recommended alternative to uprooting the entire mat3129.
Efforts to address the outbreak
Poster used to raise awareness.
Beyond what individual farmers can do on their plots, a number of measures, such as intensive surveillance and reporting of new outbreaks (with prompt action to investigate them and take action) and strict control of the movement of plant material from infected areas to unaffected ones, have been proposed34. However, in plots of 'Kayinja', which are frequently neglected and for which ownership and responsibility for maintenance is often obscure, the likelihood of stopping the progress of the disease through coordinated actions is judged low34.
Early on in the outbreak, affected and threatened countries were urged to develop national and regional strategies35. In Uganda, the Ministry of Agriculture Animal Industry and Fisheries (MAAIF) and the National Agricultural Research Organization (NARO) embarked on an intensive programme to raise awareness of the problem and enable farmers to identify the symptoms of the disease and to implement control measures36. The country was divided into endemic areas (where the disease was present), frontline (bordering the endemic areas) and threatened (disease-free) in the hope of isolating, and eventually eradicating, the disease from the endemic areas.
A number of approaches were used to reach out to farmers: the extension system, media (local radio, television, newspapers and brochures) civic, cultural and local leaders, and a participatory development communication approach which consisted in organizing, at the local level, farmers, local leaders, extension staff, researchers and communication specialists373839. An analysis of different approaches used in Uganda to mobilise farmers suggests that farmer field schools have been more effective in reducing the incidence of Xanthomonas wilt than the traditional approach of using extension services and mass media40.
In 2006, a regional Crop Crisis Control project (C3P) was initiated to reduce the spread and impact of the disease through education and training in East and Central Africa41. The project adopted a cascade training model, that is those who participated in a regional training were expected to conduct further training in their country. The people trained in-country would then take the training to the community level. In less than a year, the regional trainings had reached more than 1,000 extension and research staff while more than 30,000 farmers had been trained at the community level. However, considerable differences in capacities and institutional structures between countries affected implementation.
Coupling data on the incidence of the disease with other types of information, such as the share of bananas in daily food intake and the level food insecurity, was suggested to identify areas that should be targeted in priority42.
Field trials of the SDSR technique suggest that it could help keep the disease in check. In Uganda, Kenya and the Democratic Republic of Congo (DR Congo) the proportion of diseased plants had declined to 3% to 20% 9 months after the start of a field trial44. One year after the start of separate field trials in the North and South Kivu provinces of DR Congo, the proportion of diseased plants had been reduced to less than 0.5%45.