Faculty of Bioscience Engineering 

Academic year 2012 – 2013 

 

Suppression of Fusarium oxysporum f. sp. cubense on banana in an agroforestry system in Brazil 

 

 

 

 

 

 

Velkeneers Ellen 

Promotor: Prof. dr. ir. Monica Höfte 

Co‐promotor: Dr. ir. Soraya de Carvalho França 

Tutor: Ir. Pauline Deltour 

 

Master’s dissertation submitted in partial fulfillment of the requirements for the degree of  

Master of Science of Nutrition and Rural Development:  Tropical Agriculture

 

  iii 

The author and the promoter give the permission to use this thesis for consultation and to copy parts of it for personal use. Every other use is subject to the copyright laws; more specifically the source must be extensively specified when using results from this thesis.

Ghent, June 2013

The promotor:

Prof. dr. ir. Monica Höfte.

The author:

Ellen Velkeneers

 

  iv 

 

  i 

In Minas Gerais, Brazil, agroforestry systems were designed within an agroecological framework as a reaction to socio-economic and ecological difficulties faced by family farmers. In one system, variation was observed in the incidence of Fusarium wilt on banana ‘Maçã’, a very susceptible cultivar. The disease is caused by Fusarium oxysporum f.sp. cubense (Foc). This soil-borne fungus devastates ‘Maçã’ in monoculture. Therefore the hypothesis had been formulated that various levels of soil suppressiveness are present in the agroforestry system. The pathogen was isolated and characterized by performing molecular analysis and a pathogenicity test. Except for one, all isolates were identified as F. oxysporum and two isolates obtained from plant material acted as Foc race 1. The viability of chlamydospores buried in soils sampled around healthy and diseased banana plants was tested. Also, the differences in soil microbial functional diversity between the soils were investigated by community physiological profiling (Biolog). Although no significant differences between the soil samples from healthy and diseased banana plants could be observed, the initial hypothesis was not rejected. Soil suppressiveness consists of a complex and dynamic web of interactions, with the viability of chlamydospores and community physiological profiles being only a part of the puzzle to investigate.

In Minas Gerais, Brazilië, werden ‘agroforestry’ systemen geïmplementeerd vanuit een agro-ecologische gedachte, als reactie op de socio-economische en ecologische moeilijkheden van de kleinschalige landbouwers. In een van de systemen werd variatie geobserveerd bij het uitbreken van Fusarium wilt op ‘Maçã’ bananen, een heel gevoelige cultivar. De ziekte wordt veroorzaakt door Fusarium oxysporum f.sp. cubense (Foc), deze bodempathogeen vernietigt ‘Maçã’ in monocultuur. Dit leidde tot de hypothese dat er verschillende graden van bodemweerbaarheid aanwezig zijn in het systeem, die de ziekte kunnen onderdrukken. Het pathogeen werd geïsoleerd en gekarakteriseerd, zowel op moleculair niveau als door een pathogeniciteitstest. Alle isolaten werden geïdentificeerd als F. oxysporum, uitgezonderd een. Twee isolaten bekomen van plantmateriaal gedroegen zich als Foc race 1. De levensvatbaarheid van de chlamydosporen, begraven in bodemstalen bemonsterd rond gezonde en zieke bananenbomen werd getest. Ook het verschil in functionele diversiteit van de microbiële bodemgemeenschap tussen de bodems werd onderzocht via ‘community physiological profiling’ (Biolog). Ondanks dat er geen significante verschillen tussen de bodems van gezonde en zieke bomen kon worden aangetoond, wil dit niet zeggen dat de initiële hypothese volledig wordt verworpen. Bodemweerbaarheid is een complex en dynamisch web van interacties, met de levensvatbaarheid van chlamydosporen en de fysiologische profielen van de microbiële bodemgemeenschap slechts een deel van de te onderzoeken puzzel.

Em Minas Gerais, Brasil, sistemas agroflorestais foram delineados dentro de uma estrutura agroecológica como a resposta às dificuldades socio-econômicas e ecológicas enfrentadas pelos membros das famílias agricultoras. No sistema de um destas famílias, foi observada uma variação na incidência da doença Mal do Panamá em banana ‘Maçã’, uma cultivar altamente suscetível. A doença é causada por Fusarium oxysporum f sp. cubense (Foc). Este fungo de solo devasta ‘Maçã’ quando cultivada em monocultura. Portanto, foi formulada a hipótese que vários níveis de supressividade do solo estão presentes nos sistemas agroflorestais. Isolados foram obtidos e caracterizados através de testes de patogenicidade e análise molecular. Com uma exceção, todos os isolados foram identificados como F. oxysporum e dois isolamentos obtidos das plantas agiram como Foc raça 1. Foi testada a viabilidade dos clamidosporos, enterrados nas amostras de solos coletadas ao redor de bananeiras saudáveis e doentes. As diferenças da diversidade funcional microbiológica dos solos também foi investigada pelo ‘community physiological profiling’ (Biolog). Embora nenhuma diferença significante tenha sido observada, a hipótese inicial não foi rejeitada. A supressividade dos solos consiste em um complexo e dinâmico emaranhado de interações, sendo o caráter da viabilidade dos clamidosporos e da diversidade funcional microbiológica parte de todo um quebra-cabeça a ser investigado.

  ii 

Preface

“Every accomplishment starts with the decision to try, thereafter success depends on a combination of insistence from the self and support from the other”

Life is definitely a journey and writing a thesis to graduate, an important destination to reach. I would never have been able to arrive on this beautiful spot without the support of many people that I would love to thank.

Mams and Paps, you always gave me the freedom to choose the direction. The compass often starts spinning, but you are there to support the trek no matter how bumpy the roads are. Roel and Julie, you provided the weekly energy to recharge my batteries. Ook aan mijn grootouders, bedankt voor jullie steun. Friends, I am so grateful that you joined me even though a working-friendship balance is a difficult exercise. First, I thought about listing your names but throughout the years so many people who inspired and supported me found a place in my heart. Nevertheless, a special “Thank you!” to Jean-Pierre, Silke, Maarten and Karolien, all my roommates of the past five years, the Bios-group from Leuven, the salsa dancers in Troya and the interesting friends I met at Ghent University.

While I was searching for an agroecological subject, Prof. Monica Höfte brought me in contact with Soraya, whose enthusiasm convinced me to write this thesis about a specific phytopathological problem of family farmers in Minas Gerais, Brazil. Universidade Federal de Viçosa was the next stop on this adventure. Prof. Irene M. Cardoso and Arne welcomed me like family. Prof. Olinto Liparini Pereira gave me the opportunity to work in the “Laboratório de Patologia de Sementes e Pós-colheita”, while Alexandre Reis Machado took care of the daily personal guidance. I also would love to thank Deiziane, Camila, Stefânia, André & André of the lab for their support. Two months in Viçosa would not have been a life experience without the friendship of Rodrigo, Fausto, Bela, Amana, Jaque, Larissa, Matheus, Arthur, Livia, Fabricio, Nina & Pepe. Gente, estou tão feliz de conhecer vocês! I was welcomed by Dadinho & Cida to do the sampling on their field. Fiquei surpresa com o entusiasmo de vocês. Este tipo de pesquisa pode ser conduzida graças a vocês. É meu maior desejo compartilhar os resultados desta colaboração com vocês! Apart from taking samples home, I also received a lot of agroecological experiences from CTA-ZM, GAO etc. Vou lembrar sempre!

The environment then changed into the “Laboratory of Phytopathology” at Ghent University. The whole academic year Pauline reached me both hands and feet to get me through the methodological jungle of the laboratory experiments. In addition, her optimism and work ethic taught me another perspective at the world and I was happy to be the first member of her banana team. I also would love to thank the other hard working people of the laboratory and my fellow thesis students, especially Thomas and Lien, for their nice companionship. Nadia and Ilse, thanks for all the assistance! Thanks to Prof. Marie-Christine Van Labeke, in vitro breeding of the banana plants was possible. Soraya sended me on the journey, but she also made sure I reached the destination as she assisted me in the writing process. Writing in academic English appeared difficult, but her patience and corrections resulted in a growth process that will bear its fruits in the future. Indeed the future! One destination reached, a new upcoming wind for the next trip. Hence, enjoy the result of one year full of fungi, banana plantlets, scientific articles and interesting discussions.  

   

  iii 

List of Abbreviations

AFLP Amplified Length Polymorphism

BCA Biological Control Agent

BAP 6-Benzylaminopurine

Cirad Centre de coopération Internationale en Recherche Agronomique pour le Développement

CLPP Community Level Physiological Profiling

CTA-ZM NGO Centre of Alternative technologies, Zona da Mata

Embrapa Empresa Brasileira de Pesquisa Agropecuária

Foc Fusarium oxysporum f.sp. cubense

GN Grande Naine cultivar

IAA Indole-3-acetic acid

IBGE Instituto Brasileiro de Geografia e Estatística

ITC Bioversity’s International Transit Centre

NGO Non Governmental Organisation

PCA Principal Component Analysis

PCR Polymerase Chain Reaction

PDA Potato Dextrose Agar

PDB Potato Dextrose Broth

PGPR Plant Growth Promoting Rhizobacteria

PSUE Proportional Substrate Utilization Efficiency

RAPD Random Amplified Polymorphic DNA

RFLP Restriction Fragment Length Polymophism

SB Silver Bluggoe cultivar

ST4 Subtropical race 4

SSR Simple Sequence Repeats

TR4 Tropical race 4

UFV Federal University of Viçosa

VCG Vegetative Compatibility Group

Table of Contents 

PREFACE  II 

LIST OF ABBREVIATIONS  III 

1. INTRODUCTION  1 

2. LITERATURE REVIEW  3 

2.1 FRAMEWORK  3 2.2 FUSARIUM WILT ON BANANA  6 2.3 CONTROL STRATEGIES OF FUSARIUM WILT ON BANANA  10 2.4 SOIL SUPPRESSIVENESS  13 2.5 BIODIVERSITY AND DISEASE SUPPRESSION  14 

3. MATERIAL & METHODS  17 

3.A PATHOGEN ISOLATION AND CHARACTERIZATION  17 3.A.1 PATHOGEN ISOLATION  17 3.A.2 CHARACTERIZATION  18 3.B SOIL SUPPRESSIVENESS  21 3.B.1 EFFECT ON SURVIVING STRUCTURES  21 3.B.2 COMMUNITY LEVEL PHYSIOLOGICAL PROFILING (BIOLOG)  23 

4. RESULTS & DISCUSSION  24 

4.A PATHOGEN ISOLATION AND CHARACTERIZATION  24 4.A.1 PATHOGEN ISOLATION  24 4.A.2 CHARACTERIZATION  24 4 B. SOIL SUPPRESSIVENESS  36 4.B.1 EFFECT ON SURVIVING STRUCTURES  36 4.B.2 COMMUNITY LEVEL PHYSIOLOGICAL PROFILING (BIOLOG)  41 

5. CONCLUSION  47 

REFERENCES  48 

 

 

  1 

1. Introduction Small-scale farming or Family Farming is nowadays considered the engine of rural development (Altieri, 2009). Nevertheless, family farmers have also been facing socio-economic and ecological challenges. As a reaction on these difficulties, an agroecological approach was proposed. Agroecology consists of three pillars: a science, a movement and a practice with the goal to develop sustainable agricultural and food systems. This approach contributes to food security (De Schutter, 2010; Altieri et al., 2012) and to independent viable farms (Altieri, 2009). To use the complex ecosystem as the foundation of the adopted agricultural system is highly knowledge intensive. Therefore, cooperation between different actors is a necessity (Francis et al., 2003).

In Minas Gerais, Brazil, agroforestry systems containing coffee, banana and trees were designed through a participatory approach between farmers, The Federal University of Viçosa (UFV) and a non-governmental organization Centre of Alternative Technologies of Zona da Mata (CTA-ZM) (Cardoso et al., 2001). The project started in 1993 to achieve solutions to following problems: soil degradation, decreased production and declining biodiversity. Nowadays, the partnership still exists with the objective of finding answers to questions and problems of the farmers (Souza et al., 2012).

One of these problems is the presence of Fusarium oxysporum f.sp. cubense (Foc) causing Panama disease on banana. In Brazil, it was first observed in 1930 causing the elimination of large commercial Maçã (Silk subgroup, group AAB) cultivations in the São Paulo state, Espirito Santo, the Minas Gerais triangle and South Goias (Ploetz, 1990). This soil-borne pathogen is difficult to control because of its hidden status and because fungicides do not work. The only successful control measures are avoiding presence of Foc in the soil by using clean planting material and resistance breeding (Moore et al., 1995). While nowadays, industrial agriculture provides resistant cultivars as the Cavendish for the international market, local consumers prefer susceptible varieties like ‘Maçã’. In addition, new host-pathogen interactions and the incorporation of new genes for resistance to Fusarium wilt into widely planted cultivars will cause the development of new races, which will again challenge banana breeding (Gordon & Martyn, 1997). Although it is assumed that Foc race 1 is prevalent in the country, Cavendish cultivars are also found diseased, which questions the presence of race 4 (Ploetz, 1990).

One of the involved farmers in the participative process is Geraldo Cândido da Silva (Dadinho), who observed that the disease intensity of Fusarium wilt was high in his monocultural field of Maçã, but remarkably lower in his mixed agroforestry system. This observation led to the formulated hypothesis and objectives of this thesis. To understand the system, a multi-disciplinary approach is needed. Former research to investigate the physical and chemical characteristics of the soils had already been carried out by other students (De Weerd, 2012; Vacas, 2012 and Chavassieux, 2012). Now, the PhD project of Pauline Deltour aims to contribute by studying the phytopathological aspects. It is in this framework that this thesis was conducted.

 

  2 

The first part describes the isolation and characterization of the pathogen from the investigated field. By using PCR with specific primers and by checking the DNA in the Genbank, the isolates were characterized on molecular level. To identify the different races, a pathogenicity test was conducted with Grande Naine, Silver Bluggoe and Silk cultivars. In the second part the difference in soil suppressiveness between soil samples taken around diseased and healthy plants was investigated. Chlamydospores, the survival structures of the fungus, were buried on microscopic slides in the soil samples and after an incubation period, their germination capacity was observed. Also, community level physiological profiles of the samples were obtained by the usage of Biolog Ecoplates.

Hypothesis

Our hypothesis states that various levels of soil suppressiveness (conducive and suppressive spots) are present in the agroforestry system.

1. Foc race 1 is the causal agent of the observed disease symptoms

2. Chlamydospore viability is lower in healthy soils

3. There are differences in soil microbial functional diversity between soils around healthy and diseased plants.

Objectives

1. To isolate & characterize Foc present in the investigated soils

2. To test the viability of the chlamydospores in soils around healthy and diseased plants

3. Investigate differences in soil microbial functional diversity between soils around healthy and diseased plants

  3 

2. Literature review  

2.1 Framework

Small-scale farming, the engine of rural development

The United Nations declared 2014 to be “The International Year of Family Farming” (General Assembly United Nations, 2012). This declaration is part of a broader changing paradigm in rural development as described by Ellis & Biggs (2001). They provide a commendable in-depth view on the dynamics of rural development concepts throughout time; like the shift from industrial up scaling and modernization in the 50s towards the recognition of the importance of small-scale farming within rural development. Although theoretically the shift towards the idea of small-scale farming as the engine of rural development already arose during the 60s, the accomplishment in practice needed time (Ellis & Biggs, 2001). Given this information, “The International Year of Family Farming” in 2014 can be seen as a sign of recognition towards the importance of small-scale farming.

According to De Schutter (2009), the Brazilian law (n° 11.326) describes a family farm as an agricultural establishment that occupies up to four fiscal modules 1, employs predominantly family labour and is run by family members. During an investigation of 2006 by IBGE (Instituto Brasileiro de Geografia e Estatística), 4 367 902 sites were identified as family-agriculture representing 84.4% of the total establishments in Brazil. They occupy only 24.3% of the total agricultural area, as the average size of the farm is 18.37 ha compared with an average size of 309.18 ha for the non-family farms. On national production level, family-farms provide for example 87% cassava, 70% beans, 60% bananas, 58% milk and 59% pigs (Embrapa, 2004; IBGE, 2006).

Small-scale farming is known to have following advantages compared to larger farms. Contradicting the economies of scale, as the farm size decreases, productivity per measure of area has the possibility to increase. This inverse relationship is caused by a more optimal use of the available resources and by a more diversified production typically for small-scale farming. In addition, this diversity in production assures food security when price volatility causes too low selling prices for the grown crop or too high prices to purchase food on the market. The fact that small-scale farming is more labour intensive can be considered a disadvantage. However, to mitigate unemployment and massive rural-urban migration in search for jobs this aspect can rather be turned into an opportunity. Because job creation combined with multifunctional linkages towards other non-farm activities can lead to economic development. A focus on small-scale farmers will lead to more purchasing power and the creation of new job opportunities linked with the initial development on the farms. Compared with large-scale farming, the development occurs with more social equity as a larger part of society shares in the endogenous growth (Rosset, 2000; Ellis & Biggs, 2001; Altieri, 2009).

Although small-scale farming is considered of major importance in rural development, family farmers have also been facing many challenges. Small-scale farming is mainly associated with poverty and lack of access to productive assets, medical services, education, markets, credit and insurance. Also the obligatory competition in the                                                              1 A fiscal module is a municipal unit of measurement, expressed in hectares, that reflects the predominant patterns of land occupation in the region, its profitability and the estimated agricultural needs of a family farm. In practice, a module can vary from 5 ha in fertile, market-connected areas, to 110 ha in remote Amazonian areas. The national median of a module is currently 30 ha.

  4 

liberalized, international economy leads to major dependencies and vulnerabilities among family farmers (Schneider & Niederle, 2010; Broad & Cavanagh, 2012). Next to the socio-economic constraints, there are also ecological challenges to overcome like the degradation of land and water resources, the impact of pests and diseases, the loss of ecosystem services and the dependency on the climatic conditions (Tsharntke et al., 2012).

Family-farmers of Zona da Mata face socio-ecological constraints

The Zona da Mata (Minas Gerais), situated in a biodiversity hotspot, was once fully covered by Atlantic Coastal Rainforest. Nowadays, only 7% of the natural forest remains, mainly in small fragments or protected areas like the Serra do Brigadeiro Natural Park (Jackson et al., 2012). The replacement of the originally Atlantic Coastal rainforest had an impact on the nutrient cycle in the forest ecosystem and within a few decades the soil fertility was drastically reduced, mainly due to erosion caused by management practices of the farmers. As a consequence new deforestation took place by farmers in search for fertile land (Cardoso et al., 2001). However, deforestation became difficult as untouched land became scarce. Also, the foundation of the “Serra do Brigadeiro National Park” (1993) prevented more land to be cleared (Cardoso & Ferrari, 2006). Improvement of the soil became a tangible necessity. Although the adaptation of the Green Revolution methods in the 1970s was meant to improve production on the already cleared land, it also caused significant environmental deterioration (like biodiversity loss and agrochemical pollution) and a weakening of the ‘family farm’ as an economic enterprise (indebtedness and dependencies, competition with large commercial enterprises). In summary, practices not well adapted to the environment led to ecological and social problems (Cardoso et al., 2001).

Agroecology, a positive approach for the challenges

Clearly, handling the major problems described before will depend on the political and economical decisions made on different levels. But on farm level, the choices made may influence the direction in which the society evolves. In this way, pressure can be built up to change the framework and improve the situation of farmers. An example of such an approach is the shift to agroecological practices.

Agroecology consists of three interacting pillars; it is considered as a science, a movement and a practice. The definition evolved over the world and depending on the country the concept originated more in one or another pillar, resulting in different views on the concept (Wezel et al., 2009). In Brazil, agroecology was founded in the social response on the adverse effects of agricultural modernization and trade liberalization on family-farmers in the 1970s (Altieri et al., 2012). First, it was a movement for rural development and environmental aspects in agriculture, which stimulated a search for alternative practices. In recent years, agroecology has also been considered a scientific discipline with an integration of social dimensions. Agroecology as a science is based on an adaptation of the scientific interpretation of agroecology as the fusion of ecology and sustainable agriculture. While the social dimensions are integrated by using systemic, interdisciplinary and participatory approaches combined with the integration of indigenous knowledge (Embrapa, 2006). Instead of only focussing on agricultural production and the immediate environmental impacts, agroecology has also been defined as the integrative study of the ecology of the entire food system, with ecological, economic and social dimensions. By properties that resemble those of natural ecosystems, systems can be designed that close nutrient cycles, depend more on renewable energy, reduce inefficiencies in production and promote environmental health. To accomplish such goals, it is essential to build bridges among the disciplines in production agriculture by mixing

  5 

fields of sociology, anthropology, environmental sciences, ethics, and economics. Equally important is to look beyond the farm gate into the rural landscape and community (Francis et al., 2003).

A clear link between food security, food sovereignty and agroecology has been formulated. Agroecology raises productivity at field level when total output is considered, thereby contributing to food availability. Agroecology reduces the dependency on expensive inputs as usage of the natural resources is more efficiently. It also creates jobs in the rural areas, which contributes to increasing purchasing power, leading to better food accessibility. As the production is diverse, the adequacy of the available food is improved as well. By its environmental friendly practices, it contributes to sustainability. Also, the farm’s resilience towards impacts of climate change increases due to the on-farm biodiversity, which results in an improved recovery (Altieri, 2009; De Schutter, 2010; Altieri et al., 2012).

Cooperation as a necessity

Agroecology is very knowledge-intensive because of the necessity to profoundly understand every local ecosystem to build productive agricultural systems adapted to the environment. This knowledge can only be fully gathered when different actors work together. In Viçosa (Minas Gerais, Brazil) a partnership was established, consisting of small-scale farmes, the Federal University of Viçosa (UFV) and the NGO Centre of Alternative technologies, Zona da Mata (CTA-ZM). The UFV contributed with theoretical insights and methodologies, NGO staff offered general insights into agricultural processes, facilitation skills to engage the three parties and a strategic socio-political analysis to ensure the research was relevant. As a third but equal part, the farmers added their in-depth insights into local ecological conditions and dynamics, their clearly expressed economic needs and their fields (Cardoso et al., 2001).

As a result of this partnership, agroforestry systems were established. Agroforestry has developed as an interface between agriculture and forestry in response to the special needs and conditions of tropical developing countries. The potential of agroforestry for soil improvement and conservation is generally accepted. Indeed, it is recognized as a land use system, which is capable of yielding both wood and food while at the same time conserving and rehabilitating ecosystems (Nair, 1993). The selection of the appropriate trees and the implementation was conducted with participative rural appraisal (Souza et al., 2010). Through this process the value of the indigenous knowledge of farmers was recognized by their participation in the design, monitoring, evaluation and adjustment of the agroforestry systems. This resulted in more appropriate solutions, supported by the farmers with a more sustainable outcome. The participatory process enabled the farmers to continue with the agroforestry systems even when encountering difficulties, allowing agroforestry to show its potential (Souza et al., 2012).

 

  6 

2.2 Fusarium wilt on banana

The host

Musa spp. are among the most consumed fruits of the world. The species are grouped according to the number of chromosome sets they contain and the relative proportion of Musa acuminata (A) and Musa balbisiana (B) in their genome. Most seedless cultivars of banana are triploid hybrids, producing parthenocarpic fruits (Nelson et al., 2006). In 2003, global banana production was 68.2 million tons with the largest producers: India (24.11%), Brazil (9.48%) and China (8.54%). From the 6.5 million tons bananas produced in Brazil, only 3.41% was sold on the international market. Within Brazil, banana is the second most important fruit after the orange based on quantity produced, value of the production and the amount consumed. The most popular banana varieties are Prata, Pacovan, Prata Anã, Maçã, Mysore, Terra and D'Angola (the AAB group) for the domestic market. Prata, Pacovan and Prata Anã account for 60% of the cultivated area of banana in Brazil. Nanica, Nanicao and Grande Naine (AAA) are produced for the export sector. On smaller scale Ouro (AA), Figo Cinza and Figo Vermelho (ABB), Caru Verde and Caru Roxa (AAA) are also produced. Brazilians prefer the Maçã cultivar because of its apple taste, but it is susceptible to Fusarium Wilt. Banana in Brazil is harvested throughout the year with a peak in February and March and lowest production in October and November (Embrapa, 2004).

Symptoms Fusarium Wilt

Following external and internal symptoms indicate the presence of Fusarium wilt (or Panama disease) on banana (Nelson, 1981; Moore et al., 1995; Ploetz, 2006).

A. External symptoms

- Light green to pale yellow streaks or patches at the base of the petiole of the oldest leaves, discoloured vascular strands immediately below the epidermis, yellowing of the leaf margin, which starts with the older leaves and progresses to the younger ones. -A difference can be made between the leaf-yellowing syndrome whereby the leaves turn yellow and the non-yellowing syndrome whereby they stay green except for a streak or patch in the petiole and collapse. -Formation of a “skirt” of dead leaves around the pseudostem, due to collapsing at the petiole or towards the base of the midrib (Figure 1). -“Spiky” appearance of the plant due to late wilting of young leaves that stay erect and green. -Development of longitudinal splits in the pseudostem (Figure 1). -Growth continues in an infected plant and therefore emerging leaves are usually paler with a reduced and distorted lamina. Many infected suckers may be produced before the clump finally dies.

B. Internal symptoms

A reddish brown discolouration of the xylem which is first visible in the roots, continues then in the rhizome, is most prominent where the stele joins the cortex and finally includes large portions of the pseudostem (Figure 2). The colour is pale yellow in the early stages to dark red or black in the late stages. Young infected planting material may not exhibit symptoms because the vascular system needs to be fully developed. Symptoms are not visible in the fruit.

  7 

Figure 1 External symptoms Panama disease (Embrapa, 2004). Left picture: Typical skirt of dead leaves (A.P. Matos). Right picture: Development of longitudinal splits in the pseudostem (Z.J.M. Cordeiro) 

Figure 2 Internal symptom Panama disease (Embrapa, 2004). Vascular discolouration from the oldest xylem 

vessels (outside) towards the younger ones (inside) (Z.P. Matos). 

Taxonomy

Leslie and Summerell (2006) gave a brief history of the Fusarium taxonomy in their Fusarium laboratory manual. The taxonomy debate started in 1809 with the recognition of the typical canoe- or banana-shaped conidia by Link and after numerous revisions the debate is far from closed. The species Fusarium oxysporum Schlechtendahl emend. Snyder and Hansen was described in 1940. But later it appeared a complex of morphologically similar filamentous fungi (Kistler, 1997). It consists of mainly saprophytic strains, but also contains plant pathogens that cause vascular wilts, rots and damping-off of hundreds of host species. Agriculturally and economically, it is the most important taxon in the genus Fusarium (Ploetz, 2006).

  8 

Formae speciales

Pathogenic strains of F. oxysporum cause wilt diseases on a number of agronomically important crops. Isolates that attack the same crop are grouped in the same forma specialis. To date more than 150 formae speciales have been described (Baayen et al., 2000). Fusarium oxysporum f. sp. cubense (E.F. Smith) Snyder & Hansen (Foc) causes Panama disease or Fusarium wilt on banana (Ploetz, 2006).

Races

Since the mid-1900s these formae speciales are further subdivided in races according to the susceptible cultivar (Table 1). Four races of Foc have been described (Moore, 1995; Embrapa, 2004; Pereira et al., 2005; Ploetz, 2006).

The race 4 isolates have been divided into subtropical race 4 (ST4) and tropical race 4 (TR4). ST4 attacks Cavendish bananas previously exposed to cold winter temperature in South Africa, Australia, Taiwan and the Canary Islands and TR4 infects Cavendish bananas without the previous exposure to cold conditions in the tropical regions of Southeast Asia and Australia. The disease has already caused huge economic losses in banana production. Because of this, the banana export is again in danger (Li et al., 2011a).

For Maçã, the disease can lead to losses of 100%. The varieties of the Terra group (like Thap Maeo, Pacovan Ken, Preciosa e Maravilha) are resistant2. It is important to mention that stress can break resistance (Embrapa, 2004).

Table 1 The susceptible cultivars and their genetic group, grouped per race of Foc 

Race Susceptible Cultivar

Group

1 Gros Michel Pome -Silk (eg. Maçã) -Pisang Awak

AAA AAB AAB ABB

2 Bluggoe Figo Cinza Figo Vermelho Figo Marmelo Pelipita & Other plantains

ABB

3 Heliconia spp. 4 Cavendish

Caipira + All Cultivars susceptible for race 1 & 2

AAA

Races can only be distinguished from each other by pathogenicity tests. Care is needed because the tests can be influenced by temperature, host age, method of inoculation and other variables, which causes inconsistencies in the results of different laboratories (Fourie et al., 2011).

                                                               2 As the presence of Race 4 is not yet confirmed in Brazil, they are resistant for race 1 & 2.

  9 

VCG

In an attempt to understand the population biology, the concept of a Vegetative Compatibility Group (VCG) had been developed. Two individuals are grouped in one VCG when they can form a stable heterokaryon by hyphal fusion. Twenty-four different VCGs have been described for Foc (Katan, 1999).

Genetic diversity

DNA sequencing, restriction fragment length polymorphisms (RFLPs), random amplified polymorphic DNAs (RAPDs), amplified fragment length polymorphisms (AFLPs) and micro- satellites or simple sequence repeats (SSRs) are methods used to study genetic diversity in Foc (Nayaka et al., 2011). As a result of the use of these techniques many new facts were discovered:

The forma specialis concept is pathogenetically useful, but not phylogenetically informative. Some isolates were more closely related to another forma specialis, than within one group of pathogens for a specific host (Koenig et al., 1997). The division in different races has no phylogenetical link either. It is shown that Foc subtropical race 4 isolates are more closely related to individuals associated with other formae speciales of F. oxysporum than to race 1 or 2 isolates of Foc (Groenewald et al., 2006; Baayen et al., 2000).

VCGs represent good phenotypic characters for assessing diversity within population. The formation of a stable heterokaryon however, depends on a few genes. A mutation in one of these genes can lead to the classification of two individuals in another VCG, notwithstanding that they can be closely related. Furthermore, the relationship between VCGs and races is complex, with a single race being associated with several VCGs (Bentley et al., 1998).

Biology

The life cycle starts when chlamydospores, the asexual resting cells of the dormant fungus, are stimulated to germinate. The stimulus may be host or nonhost plant roots, or contact with pieces of fresh noncolonized plant debris. The fungus enters the plant by penetration through the tips and elongation zone of lateral roots or through wounded main roots. If young roots are the place of infection the fungus moves either inter- or intracellularly through the cortex to the developing xylem vessels and invades them before they are fully mature. However, if a wound is the place of ingress, the fungus enters the xylem vessels directly. The intercellularly movement occurs along the junctions of the root epidermal cells, while intracellulary movement occurs by direct penetration. The fungus enters the epidermal cells through what appeared to be a narrow penetration pore. The hyphae swell and form a constriction that returns to its normal size once inside the epidermal cell. However, neither true appressoria nor penetration pegs were yet observed during the penetration process. The pathogen is spread throughout the plant by microconidia that are detached and carried upwards in the sap stream. They infect neighbouring cells through pores in cell end plates. Also within the banana cells chlamydospores are formed from thickened hyphae. The apical meristem and undifferentiated vascular tissue are not invaded. The fungus is confined to the xylem until normal plant functions cease. Then browning of tissue increases and fungal growth occurs throughout the adjacent parenchyma tissue with a continue production of conidia and chlamydospores, which return to the soil as the plant decays. The fungus is also able to colonize and persist in the roots of alternative hosts even though these plants remain symptomless under field conditions (Nelson, 1981; Li et al., 2011b). Especially close relatives of the banana and several species of weeds and grasses serve as alternative hosts, like in

  10 

Brazil the invasive species Paspalum fasciculatum, Panicum purpurascens, Lxophorus unisetus, Commelina diffusa, the roots of Paspalum spp. and Amaranthus spp. that grow often near the banana plants (Embrapa, 2004).

No sexual stage is known for F. oxysporum, which does not produce a teleomorph. However, the two mating type idiomorph alleles, needed for sexual reproduction, are present. Probably mutations in these genes or a little error in the formation of pheromones and receptors occurred (Fourie et al., 2011).

Spread of the pathogen occurs most commonly by infected rhizomes or suckers, translocation of the soil by humans or animals or the use of contaminated tools. When the fungus is only introduced once in a disease-free field, it will spread slowly from plant to plant. However, if spores are carried through water (flood or irrigation water from drained infested soil), the disease can spread fast and ruin a plantation within months if the conditions are optimal (Moore et al., 1995; Embrapa, 2004).

2.3 Control strategies of Fusarium Wilt on banana

Following control methods are being investigated to control Fusarium wilt: Resistance breeding, biological control, chemical control, cultural technics and soil improvement. This last one being the initialization and optimization of soil suppressiveness, will be discussed separately.

A. Resistance

According to different authors, resistance breeding is the most appropriate solution against Fusarium wilt. Other ways of controlling the disease are being investigated and are discussed below, but none of them has been as effective as the presence of a resistant cultivar (Ploetz, 2006). The major example is the replacement of Gros Michel by Cavendish in the early 1960s against Foc race 1. In this way, the banana export production could be rescued from a total destruction (Ploetz, 1990).

As the pathogen evolves constantly, new resistant cultivars need to be bred. To obtain the resistant cultivars by conventional breeding is challenging, as cultivated banana varieties are triploid and sterile, making cross-pollination difficult. Furthermore, it is time consuming and the cultivars suffer often from reluctance when presented to consumers. A second way to breed bananas is the use of somaclonal variation, which has been more successful than expected, but this method is time consuming as well and has a low efficiency in the early selection stage. Also, genetic transformation has been suggested as a solution. Yet, genetic transformation has been hindered by the lack of resistance mechanisms, resistance markers, available resistance genes and the poor regeneration rate during transformation (Li et al., 2011c). However recently, the first report on the generation of transgenic banana with resistance to Fusarium wilt has been published (Paul et al., 2011). Animal genes that negatively regulate apoptosis were incorporated in plants, which induced resistance. Although the exact mechanisms are unknown, it is stated that Foc race 1 needs a necrotrophic phase in the life cycle. Also proven successful against Fusarium wilt was the incorporation and expression of floral defensin genes in banana (Ghag et al., 2012).

As shown, many efforts are done to find or make resistant cultivars for the Panama disease. However, resistance can be broken, as shown by the appearance of race 4, a threat for the Cavendish plantations (Koeppel, 2011). In addition, cultivars need to be bred for different goals: consumers and farmers preference, taste, size, other diseases etc. (Van den Berg et al., 2007). Furthermore, small farmers interchange

  11 

plant material, because they often do not have the means to buy the improved cultivars. Consequently, other ways to control the Panama disease are investigated.

B. Biological control

Biological control can be accomplished by direct interaction, when the control agent reduces the population of the pathogen through parasitism, antibiosis or competition or by indirect interaction, when the control agent interacts with the pathogen through the host. The indirect interaction is also known as induced resistance, because the agent induces a response by the host plant resulting in the host being more resistant to the disease (Marois, 1990). Biological control agents (BCA) are organisms (insects, bacteria and fungi) that are used to control pests and diseases through natural interactions, which is mostly promoted by an anthropogenic intervention. Different BCA to control Fusarium Wilt on banana are described. The agents can interact with the pathogen through the microbiological web in the soil or within the host.

Several studies have investigated the ability of Pseudomonas fluorescens to suppress Fusarium wilt disease of banana. Also a high degree of genetic and functional diversity among isolates of fluorescent pseudomonads associated with rhizospheric soil of banana has been found (Naik et al., 2008). P. fluorescens reduced the disease incidence in Cavendish bananas in greenhouse trials (Nel et al. 2006a). After in vitro assays, greenhouse and field experiments it also showed maximum growth inhibition of Foc on the susceptible cultivar Rasthali (Saravanan et al., 2003). Akila et al. (2011) tested P. fluorescens and Bacillus subtilis in combination with botanical fungicides to control Foc and concluded that it was effective. Indeed, Bacillus spp. were isolated from the roots of banana and identified to have an antagonistic effect against Foc (Sun et al., 2011). Both genera are known to contain plant growth promoting rhizobacteria (PGPR) (Akila et al. 2011). These soil bacteria colonize the roots of the plants and secrete antibiotics, lytic enzymes and phytohormones that react antagonistic, induce resistance and promote nutrient absorption and plant growth (Lugtenberg & Kamilova, 2009). Nevertheless, P. fluorescens and B. subtilis are not the only Foc-antagonistic bacteria (Thangavelu & Mustaffa, 2012). Also Streptomyces spp. showed antagonistic activity (Getha et al., 2005). Recently, Li et al. (2012a) added 16 new chemotactic Foc-antagonistic bacterial species, which brings the total on approximately 26 species. The fact that the amount of known antagonistic bacteria against Foc has more than doubled after one investigation, shows that more species could probably be added and that the biodiversity limit has not yet been reached. Also fungi are known to suppress Fusarium wilt on banana like Trichoderma spp. (Saravanan et al., 2003; Nel et al., 2006; Thangavelu et al., 2004) and non-pathogenic Fusarium oxysporum (Nel et al. 2006b).

Most of the described antagonists are not only obtained from the rhizosphere, but can also be isolated from the plant material itself. When microorganisms live within the plant tissue at some time during the life cycle, yet do not induce disease symptoms they are defined endophytes. The relation is mostly mutualistic, with the endophyte helping in the defence against pest and diseases. They contribute to the control through induction of resistance by activation of the biochemical and structural defence mechanisms. As an extra advantage, they promote plant growth through facilitation of increased nutrient uptake or through synthesis of plant hormones (Petrini, 1991). To strengthen in vitro banana plantlets against pest and diseases, the use of endophytes was demonstrated to be a promising technique (Dubois et al, 2006). Lian et al. (2009) used an indigenous mix of endophytes obtained from native banana plants, which resulted in a substantial reduction in the infection and severity of Fusarium wilt disease (race 4), as well as an increment of the plant growth parameters.

  12 

Endophytic bacteria are shown to act as effective elicitors that activate the host defense mechanisms. Fishal et al. (2010) concluded that Pseudomonas spp. could be promising biological control agents that can trigger resistance against Foc race 4 in susceptible Berangan banana. In Brazil, diazotrophic endophytic bacteria Burkholderia and Herbaspirillum showed the potential to reduce Foc attack and to promote the growth of banana ‘Maçã’ plants (Weber et al., 2007). Also fungi, like Penicillium citrinum, are known to induce the host defence mechanisms although it did not provide complete resistance for Foc race 4 (Ting et al., 2012). As well, endophytic Streptomycete antagonists of Fusarium Wilt were isolated (Cao et al., 2005). But most focus is put on the use of non-pathogenic Fusarium oxysporum. Forsyth et al. (2006) isolated non-pathogenic Fusarium oxysporum but apart from reducing the disease symptoms, one strain also enhanced the disease incidence. Therefore the concern arises whether it is truly non-pathogenic or that it can develop in a new race (Gordon & Martyn, 1997). Belgrove et al. (2011) isolated Fusarium oxysporum strains from disease-suppressed fields. A selection gave positive results in the greenhouse, but the field trail appeared not successful.

Indeed, most research to find appropriate biological control agents against Fusarium wilt is conducted in greenhouses. However, difficulties occur for field application as a high inoculum density of the control agent is necessary and as little information is available about the interactions with other organisms in the soil. This leads to a lack of consistency of biological control performance, which is the most important reason for lack of adoption (Fravel et al., 2003). Therefore a better understanding of the microbial soil ecology is a critical aspect. By investigating the chemical composition of banana root exudates and by screening for chemotactic bacterial strains, Li et al. (2012a) tried to obtain more insight in the possibilities to control the beneficial bacteria. Also the research investigating the abiotic and biotic interactions of soil suppressiveness as described below contributes to our global understanding. Another aspect to take into consideration is that the success of biological control on crop plants not only depends on effective antagonists, but also on the costs involved and the way of application (Fravel et al., 2003). Therefore, research has been conducted on Trichoderma harzianum to find a sustainable way for mass production (Thangavelu et al., 2004). Furthermore, combining different biological control applications can be very effective. Akila et al. (2011) conclude that soil born pathogens like Foc can only be kept under control when an integrated disease management program is followed.

C. Chemical control

Chemical control of Foc is difficult due to the production of thick-walled, long living chlamydospores in the soil (Nelson, 1981). Recently, biodegradable chitosan was found to inhibit Foc race 4 in vitro (Al-Hetar et al., 2010). Botanical fungicides are extracts from plants used to control fungi, considered as an alternative for synthetic chemicals. Among the 22 plants screened by Akila et al. (2011), the aqueous leaf extract of Datura metel completely inhibited the mycelial growth of Foc race 1 and reduced the disease incidence. They also tested the effect of carbendazim and the commercial botanical fungicides Wanis 20EC and Damet 50 EC, which appeared effective. Even better results where obtained when they were combined with biological control agents. Both researches were conducted in vitro and were effective in the control of the mycelial growth, which is still not a solution for the persistent chlamydospores.

D. Other cultural methods

Farmers are recommended by Embrapa (2004) to avoid areas where the pathogen is present, to use clean tools, free from nematodes because they can break the

  13 

resistance by facilitating the penetration through the wounds and to keep the pH close to neutral, with optimal levels of Mg and Ca, which leads to a less favourable environment for the pathogen. Soils with a high level of organic matter are considered to increase the competition between the microorganisms. The banana plants have to receive sufficient nutrition with a good relation between K, Ca and Mg. Infected banana plants should be removed and the place should be treated with herbicides and calcium.

In China, banana plants are rotated with Chinese leek (Allium tuberosum) for 2 to 3 years to control Fusarium wilt. Field trials and the local farmers’ experience show that the adoption of this system can be an efficient way to control Fusarium wilt in areas heavily infested with the pathogen. Adoption of the rotation system has greatly reduced the Fusarium wilt incidence and has brought economic benefit to the farmers. Chinese leek imposes no harm on humans or the environment. Hence, the use of Chinese leek to control Fusarium wilt has the potential to be an efficient, environmentally friendly way of disease management (Huang et al., 2012).

2.4 Soil Suppressiveness

Already in the 1930s, Fusarium wilt suppressive soils were described in Central America. They supported production of susceptible Gros Michel for more than 30 years, while susceptible soils were not productive after 5 to 10 years. Reports of suppressive soils around the world; West-Africa, Ecuador, Canary Islands, Philippines Jamaica and South Queensland indicate that this is a natural phenomenon. Nevertheless, it is not static and it can be destroyed by flood fallowing or intensifying cultural practices (Stover, 1990).

Given a susceptible host, disease suppression is the suppression of the pathogen (Termorshuizen & Jeger, 2008). Pathogen-suppressive soils were defined as soils in which the pathogen does not establish or persist, establishes but causes little or no damage, or establishes and causes disease for a while but thereafter the disease is less important, although the pathogen persists in the soil. Consequently the disease incidence or severity was lower than expected for the prevailing environment despite ample opportunity for the pathogen to have been introduced. A differentiation can be made between general and specific suppression. General suppression of a pathogen is directly related to the total amount of microbiological activity at a time critical to the pathogen like propagule germination or prepenetration growth in the host rhizosphere. Not one specific microorganism is responsible for general suppression, but the total active microbial biomass competes with the pathogen and possibly causes inhibition. Specific suppression operates against a background of general suppression but is more qualitative, owing to more specific antagonistic effects of select microorganisms to the pathogen. Undoubtedly, most soils involve a combination of general and specific suppression (Cook & Baker, 1983).

Enhanced suppression is typically correlated with enrichment of antagonistic or competitive activities in one or more components of the soil microbial community. However, research may be hampered by an overemphasis on pathogen-antagonist interactions with little consideration of the bulk of microbial interactions in soil. Nevertheless, a dynamic co-evolutionary approach may be vital for realizing long-term, stable suppressiveness (Kinkel et al., 2011). Suppressiveness is indeed based on microbiological interactions, however these are dependent on abiotic characteristics of the soil (Alabouvette, 1999). From this perspective, it is clear that understanding the different interactions in the soil is the first step towards a sustainable disease control by enhancing of the suppressiveness that naturally exists in every soil (Alabouvette, 1999). Nevertheless, the complexity of the system may not be underestimated.

  14 

While in general much research is conducted in an attempt to understand the driving factors of soil suppressiveness, the literature about suppression of Foc is less extensive. Despite the high interest for the isolation of BCAs as described before, little is known about the ecological dynamics in the soil web. In the samples, taken by Peng et al. (1999), counts of bacteria and actinomycetes appeared significantly higher in the suppressive soil. In contrast, the estimated populations of filamentous fungi and yeasts were lower in the suppressive soil than in the conducive soil. There are few reports describing the effect of soil chemical and physical characteristics on the symptoms of Fusarium wilt in banana plantations. Peng et al. (1999) reported that chlamydospore germination of Foc, under controlled experimental conditions, are decreased by extreme soil pH values and temperature and by increased CaCo3, Ca(OH)2 and Fe-EDDHA supply. A more alkaline environment, increasing temperature and reduced soil water content increased the disease severity, while CaCo3, Ca(OH)2 and Fe-EDDHA supply reduced the disease severity. The suppressive soil was characterised by a higher concentration of calcium, magnesium, sodium and especially potassium in comparison with the conducive soil. Field investigation by Domínguez et al. (2001) showed that soil solution pH, electrical conductivity and soluble sodium were higher for samples from suppressive spots than from conducive ones. The mass of water stable aggregates was higher in the conducive soils. They hypothesize that a greater stability of the aggregates might increase the available iron content, which might cause more severe disease incidence, because iron is needed for germination of the spores. Also the effect of organic amendments and solarisation on Foc was investigated (Nasir et al. 2003). A difference was made between disease incidence as the frequency of invasion and disease activity, measured by the presence of Foc in the pseudostem. They found an increase in disease incidence recorded in soil amended with chicken manure. Increases in the mineral nitrogen content of the soil treatment were associated with a higher frequency of invasion, but this was not the case for urea. The activity of Foc in the soil was significantly reduced by the addition of composed sawdust and unexpectedly the activity was enhanced in the solarised soil.

Nevertheless, several studies have been conducted on soil suppressiveness of Fusarium wilt on other plants. Weller et al. (2002) reviewed the microbial interactions responsible for specific suppressiveness to Fusarium Wilt. Amir & Alabouvette (1993) conclude that the level of suppressiveness is not independent of soil texture as changing the texture of a sandy soil by addition of an active clay enhances the suppressiveness of the soil against F. oxysporum f.sp. lini. Also compost can induce suppressiveness against soil borne pathogens (Noble & Coventry, 2005). Adding compost could reduce the disease severity induced by Fusarium oxysporum f.sp. melonis, basilici, radicis-lycopersici, radicis-cucumerinum (Yogev et al., 2006; Yogev et al., 2011), f.sp. lini (Serra-Wittling et al., 1996) and f.sp. lycopersici (Cotxarrera et al., 2002).

2.5 Biodiversity and disease suppression

Diversification at different scales is one of the two pillars of the agroecological approach, alongside soil quality enhancement. Biodiversity is considered the engine of the ecosystem and copying the dynamics of the natural ecosystem should lead to sustainable agricultural systems (Nicholls & Altieri, 2004). Agrobiodiversity has a positive effect on stability, yield and quality on farm level. Furthermore, it leads to multifunctional agriculture with the associated socio-economic advantages like ecosystem services and recreation (Gurr et al., 2003; Jackson et al., 2007; Malezieux et al., 2009).

  15 

A complex agroforestry system has the highest degree of biodiversity within the known agricultural systems (Malézieux et al., 2009). Many evidence was found that agrobiodiversity contributes to pest and disease control (Letourneau et al., 2011; Ratnadass et al., 2012). Because the development of therapeutic tools to control pests is a circle which keeps on creating new problems whenever a target is reached, Lewis et al. (1997) expressed the need to develop farming practices that are compatible with ecological systems and to design cropping systems that naturally limit the elevation of an organism to pest status. Ratnadass et al. (2012) reviewed all known pathways through which vegetational diversification reduces the impact of pests and diseases (Figure 3). Here, we will focus on the potential impact of a diversified system on soil borne diseases.

Figure  3  Different  pathways  through  which  vegetational  diversification  reduces  the  impact  of  pest  and diseases (Ratnadass et al., 2012). 

Plants differ in microbiological community associated with the rhizosphere, therefore the presence of plants influences the diversity in the soil and a link can be observed between the above- and below-ground food web. Introducing selected plant diversity may turn out to be a better option for building up beneficial populations than directly inoculating the soil with beneficial microorganisms (Ratnadass et al., 2012). Latz et al. (2012) showed that plant diversity contributes to plant community resistance against pathogens by fostering beneficial bacterial communities. They investigated the effect of plant diversity on the bacterial excretion of two antifungal compounds, as well as their effect on soil suppressiveness towards R. solani. Certain plants influence the soil community by exudates or through releasing chemicals after litter decomposition. Insunza et al. (2002) found that nematicidal plants benefitted bacteria that release chitinase. As described in section 2.3, chitinase showed also in vitro adverse effects on Foc (Al-Hetar et al., 2011).

Agroforestry systems have a high amount of organic matter in the soils. This leads to more microbiological activity because organic amendments are rich in labile carbon fractions, which are a source of energy for microorganisms (Janvier et al., 2007). The more microorganisms present, the higher the chance for antagonism to occur (Altieri, 1999) and the more difficult it will be for any given portion of the microbial population to obtain the nutrition and energy necessary for their growth (Cook & Baker, 1983).

  16 

Sequeira (1962) already found that adding organic amendments induced chlamydospore germination of Foc, which made them susceptible for bacterial antagonism. This bacterial activity was also increased by the supply of organic amendments and the communities differed depending on the type of amendment. However, the increase in microbial activity does not always results in disease suppressiveness (Grantina et al., 2011). Nevertheless, more diversity is not only causing more soil suppressiveness due to an increased amount of antagonists, it also contributes to co-evolution, leading to co-evolutionary hotspots, which can contribute to disease suppression in the long term (Kinkel et al., 2011). Results of compost on disease suppression vary a lot depending on the type of compost, the rate of application, the degree of decomposition and the target pathogen (Janvier et al., 2007). The inconsistency in the results with organic matter reveals the need to identify specific parameters that can predict the suppressiveness of each organic matter type in combination with different pathogen species (Bonanomi et al., 2010).

Still, increasing plant species diversity as such is insufficient to reduce pest and disease incidence, as augmenting biodiversity can also induce new pest and disease problems (Schroth et al., 2000). Therefore, not maximization but optimization of the vegetational diversification is necessary to develop sustainable agro-ecosystems. Improvement of the understanding of the mechanisms involved should be a first step in this development (Ratnadass et al., 2012).

 

  17 

3. Material & Methods  

3.A Pathogen isolation and characterization

To investigate a disease, the identification of the causal agent is necessary. Pathogen isolation was carried out at the Federal University of Viçosa, Minas Gerais, Brazil at the Laboratory of Seed- and Post-harvest Pathology under supervision of Prof. Olinto Liparini Pereira. The characterization was performed at the Laboratory of Phytopathology, Ghent University.

3.A.1 Pathogen Isolation

The samples were taken during the dry season (August) at the farm of Geraldo Cândido da Silva (Dadinho) and Aparecida Silva (Cida), which is located about 2.5 km southeast of Pedra Dourada, Minas Gerais, Southeast Brazil (20°50’ S, 42°07’ W), on a north-facing slope along a path with a downhill slope from east to west. Previous work selected ten diseased banana plants (1D-10D) and eight healthy banana plants (1S-8S) based on external symptoms of Fusarium Wilt (De Weerd, 2012). Six of these were sampled: Five “diseased” banana plants (1D, 2D, 5D, 6D, 7D) and one “healthy” appearing banana plant (2S). This last one did not show any exterior symptoms and was therefore considered as healthy, but after longitudinally cutting the pseudostem vascular discoloration was observed. Plant tissue was taken by cutting a vertical strip in the pseudostem until the front (the edge between healthy and discoloured tissue) was visible. This section was separated, transported in paper bags and stored in the refrigerator (Moore et al., 1995). Based on previous work that investigated the soil characteristics (Vacas, 2012; Chavassieux, 2012), ten soil samples were also taken. For each sample one sampling spot was chosen within a radius of 30 cm from the banana plant in the upper 15-20 cm of the soil. In this way, samples were taken around five “diseased” (1D, 2D, 5D, 6D, 7D) and five “healthy” banana plants (1S, 4S, 5S, 6S, 7S). The soil samples were transported in plastic bags and stored in the refrigerator.

The section of plant tissue was cut in pieces of approximately 0.5 cm2. The pieces were first submerged in 70% ethanol for 1 min, surface sterilized in 1% NaOCl for 3 min, rinsed in sterile water and dried in sterilized filter paper. The pieces were then plated on potato dextrose agar (PDA) with streptomycin. Fungi with morphological characteristics of Fusarium spp. were isolated. To isolate the fungi from the soil samples, a serial dilution method was used. The dilutions (10-2 to 10-5 g soil ml-1) were plated on Martin’s rose bengal agar. The isolates that were identified as Fusarium, based on microscopic and cultural features (Nelson et al., 1983; Leslie & Summerell, 2006) were stored.

At Ghent University, the isolates were preserved in four different ways: (1) Dried toothpicks at room temperature (Larkin & Fravel, 1999). (2) Dried filter paper technique, stored in autoclaved small envelopes (Larkin & Fravel, 1999). (3) Grown on slants with PDA of half strength (1/2 PDA) at room temperature for two weeks, thereafter placed in the refrigerator (4°C) (4) Cryopreservation (-80°C) after submerging a plug of one week old in a sterilized solution of 50% glycerol and 50% distilled water.

  18 

3.A.2 Characterization

Molecular Identification

Isolates were characterized based on sequencing of the rDNA-ITS region and by using specific primers (Table 2). In addition to our isolates (P2D, P7D, P2S from plant material and S2, S3, S6, S9, S11, S16 from soil), one isolate (E) of Foc race 1 from Brazil provided by Empresa Brasileira de Pesquisa Agropecuária (EMBRAPA) and four isolates of Foc tropical race 4 provided by CIRAD France (MIAE01225-01228), were included in the analyses as control.

Table 2 Molecular identification of the strains based on following primers. Target: Level of identification given by the primer. Primer: Name of the primer. Sequence: DNA sequence of the primer. Amplicon size: The size of the expected amplican in basepairs. T= Tropical; ST = Subtropical. 

Target Primer Sequence Amplicon size (bp)

Reference

Fusarium oxysporum

FOF1/FOR2 5’-ACATACCACTTGTTGCCTCG-3’/ 5’-CGCCAATCAATTTGAGGAACG-3’

340 Mishra et al., 2003

F. oxysporum f.sp. cubense Race 4 (T&ST)

Foc1/Foc2 5’-CAGGGGATGTA TGAGGAGGCT-3’/ 5’-GTGACAGCGTCGTCTAGTTC C-3’

242 Lin et al., 2009

F. oxysporum f.sp. cubense Race 4 (T)

FocTR4-F /FocTR4-R

5’-CACGTTTAAGGTGCCATGAGAG-3’/ 5’-CGCAC GCCAGGACTGCCTCGTGA-3’

463 Dita et al., 2010

A direct colony PCR strategy (Van Buyten et al., 2012) was conducted to check the presence of Fusarium oxysporum. In addition, the ITS 4-5 region was multiplied to obtain a confirmation on species level. After purification (QIAquick PCR purification Kit Protocol), the DNA could be sequenced and compared with sequences in Genbank using the blast algoritm.

To check the presence of F. oxysporum race 4 (tropical and subtropical), a direct colony PCR strategy was followed (Van Buyten et al., 2012). Because the positive control of race 4 gave a negative result, the PCR was repeated with DNA obtained from the DNA extraction method of ILVO (Debode Jan & Cremelie Pieter, Personal communication, January 2013). Mycelium was added in a 0.5 ml microtube containing 40 µl of 0.25 M NaOH and placed in a water bath for 30 s. Then 40 µl of 0.25 M HCl, 20 µl of 0.5 M Tris–HCl (pH 8) and 20 µl of 0.25% (w/v) Nonidet P-40 were added and the tubes were replaced in the boiling water for 2 min. DNA in the samples was quantified using a Nanodrop® spectrophotometer, diluted to 10 ng µl−1 and stored at −20°C. Also pure DNA of the positive controls, the Embrapa strain (E) and two strains from CIRAD (MIAE01225 and MIAE01226) were extracted and included in the PCR for verification. The PCR program was conducted as described by the developers of the primers (Mishra et al., 2003; Lin et al., 2009; Dita et al., 2010).

 

  19 

Pathogenicity test

Silk, Silver Bluggoe and Grande Naine cultivars were used to differ between race 1,2 and 4 (Table 3). The in vitro cultures of these cultivars were obtained from the Bioversity's International Transit Centre (ITC), Laboratory of Tropical Crop Improvement, KU Leuven.

Table 3 Three different cultivars, belonging to different genetic groups were used to identify the races.  ST= Subtropical, T= Tropical. 

Race Silk AAB

Silver Bluggoe ABB

Grande Naine AAA

1 Susceptible Resistant Resistant 2 Resistant Susceptible Resistant 4 ST

Susceptible Susceptible Susceptible after cold stress

4 T Susceptible Susceptible Susceptible

Before the pathogenicity tests could be performed, the banana plantlets were multiplied and reared in vitro (Israeli et al., 1995). To obtain enough banana plantlets to perform the pathogenicity test, different in vitro multiplication cycles of 4-6 weeks on Murashige and Skoog multiplication medium with vitamins were completed. The used medium consisted of Murashige and Skoog medium with vitamins (4.3 g), ascorbic acid (10 mg), sucrose (30 g), 6-Benzylaminopurine (BAP 2.25 mg/l) and indole-3-acetic acid (IAA 0.175 mg/l). The pH was adjusted to 5.7 and gelrite (2 g/l) was added. Subsequently, the medium was cooked and autoclaved. To stimulate shoot elongation and rooting, the shoots were placed on proliferation medium with activated charcoal (1.75 g/l) during 5 weeks. The proliferation medium is similar to the multiplication medium with the only difference that 0.225 mg/l BAP was added.

The explants with developed roots were transplanted into small plastic greenhouses with potting soil (Figure 4). It is important to wash off the medium as fungi can grow on the left parts. The small greenhouses were placed in a growth room with assimilation light (16/8) at 25°C. One week later, hardening off was started and from this stage on the plants were regularly watered. When three weeks old, the plantlets were replanted into pots. Hoagland solution was provided every two weeks to assure sufficient nutrition (Israeli et al., 1995).

Figure 4 Banana shoots with roots (left) ready to be planted in small plastic greenhouses (right) 

  20 

The pathogenicity test could be performed when the plants reached 10-15 cm in height (after 6-8 weeks), as the vascular system needs to be fully developed before inoculation (Mak et al., 2004). Following isolates were tested on the available varieties (Table 4) with three replicates each.

Table 4 Overview of the tested isolates and the used cultivars. X = Indicating the used combination. 

Isolate Grande Naine Silver Bluggoe Silk Embrapa X X X P2S X X P7D X X MIAE01225 X MIAE01226 X MIAE01227 X MIAE01228 X

To prepare the inoculum, the isolates were grown on ½ PDA medium for one week. A spore suspension was made, which was then mixed with Fusarium Armstrong medium (Booth, 1971) and shaken for five days at 25°C. After filtering through cheesecloth, the inoculum density was adjusted to 106 spores ml-1. For inoculation the cleaned roots were immersed in the spore suspension during two hours and then replanted in potting soil (Nel et al., 2006b).

The plants were evaluated every two days. Each leaf was scored according to following scale: 0= no symptoms, 1= less than 50% yellow or brown discolouration, 2= more than 50% yellow or brown discolouration, 3= 100% wilted leaf. The disease index was calculated as (Σ (number of leaves x score) / (3 x total number of leafs)) x 100. Final evaluation took place four weeks after the first symptoms appeared. Again the disease index was calculated, the whole plant was scored according to general appearance: pseudostem colour, wilting, rotting. The plant was longitudinally cut in two pieces and the rhizome discolouration index (Mak et al., 2004) and relative lesion length (Fortunato et al., 2012) were calculated. Mann-Whitney U tests were performed in SPSS (p=0.05). To confirm Koch’s postulate, pieces of the pseudostem and the rhizomes were taken and the pathogen was isolated as described before.

Table 5 Indicators final evaluation, with formula and scores. A) External Symptoms, B) Internal Symptoms 

Indicator A) EXTERNAL SYMPTOMS

Formula Scores Significance

Disease Index (DI%)

Σ (Score * N° leaves) _________________ *100 3 * Total leaves

0 1 2 3

Green Leaf <50% yellow or brown >50% yellow or brown Dead leaf

External discolouration pseudostem

0 1 2

Green Light Brown Brown

Wither Score 0 1 2

Erect Flaccid Totally withered

Rot 0 0.5 1

No Yes, but not collapsed Collapsed

  21 

Indicator B) INTERNAL SYMPTOMS

Formula Scores Significance

Internal Discolouration Pseudostem =Relative Lesion Length

Length internal discolouration ___________ Total Length

Rhizome Discolouration Index (RDI)

1

No discolouration of tissue of stellar region rhizome or surrounding tissue

2 No discolouration of stellar region rhizome. But discolouration at junction of root and rhizome

3 A trace to 5% of stellar region discoloured 4 6-20% of stellar region discoloured 5 21-50% of stellar region discoloured 6 More than 50% of stellar region discoloured 7 Discolouration of the entire rhizome 8 Dead plant

3.B Soil Suppressiveness

To investigate the hypothesis that various levels of soil suppressiveness are present in the agroforestry system, the effect of the soil on the surviving structures was studied. Also, the functional diversity on bacterial community level of the different soil samples was investigated. The used soil samples were the same as described for the isolation of Foc. The Embrapa isolate was used to investigate chlamydospore production and the effect of the soil on the chlamydospores.

3.B.1 Effect on surviving structures

3.B.1.1 Chlamydospore Production

To be able to study the effect on chlamydospores, different methods to produce them were compared.

Corn flour medium

To prepare corn flour medium, 100 g of sand and 200 ml of distilled water were added to 1 kg of corn flour. Nine autoclavable polypropylene bags were filled with 110 g of the mixture and autoclaved. A spore suspension (9.7 x 107 spores ml-1) from a 4-week-old culture was prepared. The bags were inoculated with 1 ml, 3 ml or 5 ml of the spore suspension with each three replicates. They were then daily kneaded to prevent formation of hardened structures in the medium (Alexandre Reis Machado, Laboratory of Seed- and Post-harvest Pathology UFV, Personal communication, August 2012). After one month, fungal growth was observed and the method was evaluated. The medium (5-10 ml) was dissolved in water and filtered through cheesecloth. After centrifugation (5 min at 2000 rpm), the water was removed and the pellet was resuspended in 20 ml of sucrose solution (50%). After shaking, the solution was centrifuged for 1 min at 2000 rpm. The upper 5 ml was resuspended in water and centrifuged again for 2 min on 6000 rpm. The lowest 5 ml was kept and the chlamydospore concentration was determined by a Bürker counting chamber.

  22 

Soil agar medium

Soil agar medium was prepared by placing 500 g of dry soil with 15 g of agar into a flask and the volume was brought to 1 l by adding water. The medium was autoclaved and poured while frequently shaking (Klotz et al., 1988). Potato Dextrose Agar (PDA) plugs with Foc of 10–day-old colonies were placed on the soil agar medium. After 10 days of incubation at room temperature the experiment was evaluated. The mycelium was harvested by extracting the loose mycelium from the edge of the plugs with a tweezer. Thereafter, it was submerged in 5 ml of water and mixed. A Bürker counting chamber was used to determine the chlamydospore concentration.

Plug in salt solution

A salt solution was prepared (0.007M KH2PO4, 0.002M MgSO4.7H2O) (Stevenson & Becker, 1972). The solution was divided over three falcons (5 ml each) and a PDA plug of a one-week-old colony was added. During two weeks, the falcons were being shaken at 28°C in the dark. An incubation period of two additional weeks without shaking was needed before evaluation. After mixing the plug, the chlamydospore concentration was counted with the Bürker counting chamber. The same experiment was performed with a spore suspension (107 spores ml-1) obtained of a one-week–old colony.

Spore suspension in soil

A spore suspension (107 spores ml-1) was grown overnight in Potato Dextrose Broth (PDB). The PDB was removed by centrifugation (6000 rpm) combined with twice a washing step with sterile water. The suspension (2 ml of 107 spores ml-1) was then added to 50 ml of dry fine sand (Couteaudier & Alabouvette, 1990). The chlamydospore formation was evaluated as described before in the section “corn flour medium”.

3.B.1.2 Effect on chlamydospores

To investigate the effect of the soil on the viability of the chlamydospores, they were placed on microscope slides, which were buried in jars filled with the soil samples (Sequeira, 1962). First, water agar was prepared, autoclaved and kept on 45°C in a water bath. A chlamydospore suspension was obtained from a plug in salt solution as described before. The water agar (5 ml) was mixed with the spore suspension (5 ml) resulting in a concentration of 2.5 x 104 chlamydospores ml-1, 4.8 x 106 microconidia ml-1 and 3.2 x 104 macroconidia ml-1. Three drops (100 µl) of this suspension were spread over the microscope slides and they were left to solidify. The same procedure was conducted with a spore suspension (105 microconidia ml-1) of a two weeks old culture. The slides (three per sample) were then buried in soil (Figure 5), which had already been incubated during six days after the soil moisture content calculated on dry weight was corrected to 28% to activate microbial life (Steiner et al., 2008).

After 60 days of incubation, the slides were recovered. Overnight, germination of the chlamydospores was provoked by placing the slides in potato dextrose broth. After a staining step with vinegar and ink, the slides were evaluated under the microscope (Olympus BX51) at 40x magnification (Figure 5). The relative amount of germinated chlamydospores and the relative amount of formed chlamydospores were calculated. Statistical analyses followed the binary logistic regression analyses and Mann-Whitney U test in SPSS, both with significance level of 5%.

  23 

3.B.2 Community Level Physiological Profiling (Biolog)

Functional diversity of the soil samples was analysed by a community-level assay of the microbial community structure based on the usage of carbon sources from Biolog EcoplatesTM (Garland & Mills, 1991; Choi & Dobbs, 1999). The functional diversity of a community can be defined as its potential activity, which is the capability to adapt its metabolism, relative composition and/or size to varying conditions (Preston-Mafham et al., 2002). In the case of Biolog, the number, types, activities and rates at which a collection of carbon sources are utilised by a bacterial community are investigated (Harch et al., 1997). This is known as the microbial community function, implying the actual catabolic activity expressed, which can provide a view on the functional diversity (Preston-Mafham et al., 2002). The obtained community-level physiological profiles can then be compared to look for differences in soil microbial community between diseased and healthy considered soils.

Ecoplates contain 31 different carbon sources and a water control with each three replicates, resulting in 96 wells. Every well includes a tetrazolium dye, an indicator for respiration, which is reduced in the purple coloured formazan after consumption of the carbon source (Choi & Dobbs, 1999). A soil extract was prepared by suspending 5 g of soil in 50 ml of phosphate buffer (0.05M, pH 7). The suspension was shaken for 10 min at 200 rpm, followed by decantation during 30 min. The supernatant was diluted to the final concentration of 10-3 g soil ml-1 and every well was filled with 150 µl of the extraction. Daily during one week, the Biolog plates were read for absorbance changes with a microplate reader at 590 nm (Deltour, 2012). To study the difference in functional diversity, three major components were analysed (Garland, 1997). (1) Calculation of the average well colour development (AWCD) provided a view on the microbial density and activity of the samples. (2) The variation between the samples could be visualised by performing principal component analyses (PCA). (3) While Lorenz-curves were drawn and Gini-coefficients were calculated to analyse the diversity within the samples (Harch et al., 1997).

Generally, the data used for the PCA are limited to a few selected time points (Garland & Mills, 1991; Garland et al., 2001; Xu et al., 2008). However, the absorbance vs. incubation time curves contain additional information not available from any single time point analysis; such as lag time, rate of colour development and maximum absorbance (Guckert, 1996). To include this information, the area under the curve was calculated. Since increases in density alone will decrease the lag time, increase the slope and increase the integral of colour development, the multivariate profile of this kinetic approach can be influenced just as greatly by inoculum density as the multivariate profile of the absorbance from a single point plate reading (Garland, 1997). Therefore, normalisation was obtained by dividing the areas under the curves by the total area for each sample, resulting in the “Proportional substrate utilisation efficiency” (PSUE) for each carbon source, which is independent of inoculum density (Pietikäinen et al., 2000).

Figure 5 Left: Microscopic slides with chlamydospores buried in the soil. Right: Germinated (red arrow) and not germinated (black arrow) chlamydospores, stained with vinager‐ink (blue).        

  24 

4. Results & Discussion  

4.A Pathogen isolation and characterization

4.A.1 Pathogen Isolation

Fourteen isolates have been used during the present study (Table 6). Three isolates were obtained from plant material: two from banana plants showing disease symptoms (P2D and P7D) and one isolate from a plant without external symptoms (P2S). Two weeks after plating the collected soil samples, six Fusarium-like isolates were selected (S2, S3, S6, S9, S11 and S16). In addition to our isolates, one isolate of Foc race 1 (E) from Brazil provided by Empresa Brasileira de Pesquisa Agropecuária (EMBRAPA) and four isolates of Foc tropical race 4 from Taiwan (MIAE01225 to 01228) provided by Centre de coopération internationale en recherche agronomique pour le développement (Cirad, France), were included in the following analyses as positive controls.

Table 6 Isolates and their origin. D = Plants with external symptoms of Fusarium Wilt. S= Plants without external symptoms of Fusarium Wilt.

Isolate Origin Country Source P2D Plant 2D Brazil Own isolation P7D Plant 7D Brazil Own isolation P2S Plant 2S Brazil Own isolation S2 Soil 1S Brazil Own isolation S3 Soil 5S Brazil Own isolation S6 Soil 7S Brazil Own isolation S9 Soil 6D Brazil Own isolation S11 Soil 6D Brazil Own isolation S16 Soil 1D Brazil Own isolation E / Brazil Embrapa MIAE01225 / Taiwan Cirad MIAE01226 / Taiwan Cirad MIAE01227 / Taiwan Cirad MIAE01228 / Taiwan Cirad

4.A.2 Characterization

Molecular identification and pathogenicity tests were conducted to characterize the different isolates.

4.A.2.1 Molecular identification

Universal primers for fungi and specific primers for F. oxysporum and Foc race 4 were used for the molecular identification (Table 7). Specific primers for F. oxysporum amplified a 340-bp fragment of all isolates but one (P2D). Sequencing of rDNA ITS region (amplified with universal primers) confirmed the identity of all our isolates, except P2D, as F. oxysporum. The isolate P2D was identified as Gibberella sp., the sexual stage of Fusarium. However, no sexual stage is known for F. oxysporum (Fourie et al., 2011). Recently, Gibberella musae, has been proposed as a new species (Anamorph: Fusarium musae) (Van Hove et al., 2011). As it morphologically shares a number of characteristics, the species was previously

  25 

identified as F. verticillioides (Sacc.), a species of the Gibberella fujikuroi species complex (GFSC). Nevertheless, F. musae can be differentiated from F. verticillioides on the basis of the production of pseudochlamydospores and characteristics of microconidia and phialides. In addition, it produces moniliformin but no fumonisins. Multilocus (EF-1 a, b-tubulin, calmodulin and RPB2) phylogenetic analysis for species of the GFSC, clearly resolved 11 F. musae isolates as a novel phylogenetically distinct species, which can be seen as sister of F. Verticillioides (Van Hove et al., 2011). The race specific primer for subtropical and tropical race 4 (Foc1/Foc2), as well as the race specific primer for tropical race 4 (FocTR4-F /FocTR4-R) failed to generate amplicons. Surprisingly, even the four strains from Taiwan (included as positive controls for tropical race 4) did not show any amplification. Two explanations are suggested: either the provided strains do not belong to Foc race 4 or what has been called race 4 comprises a genetically heterogeneous group and not all strains of this group are detected by the available race specific primers. This last explanation seems plausible as the molecular background of Foc and Fusarium oxysporum is still largely unknown (Baayen et al., 2000), which can explain a discrepancy between recently developed markers and the used strains. Also a mutation in the target DNA sequence could occur, as Fusarium oxysporum is known to mutate easily on full strength PDA (Nelson, 1981).

Table 7 Molecular identification. "-" = No PCR product formed. "+" = PCR product formed with the expected amplification size. “Foc” = Fusarium oxysporum f.sp. cubense, “ST” = subtropical, “T”= tropical.

Primer FOF1/FOR2 ITS4 and ITS5 Foc1/Foc2

FocTR4-F /FocTR4-R

Target species

Fusarium oxysporum

Fungi Foc ST&T Race 4

Foc T Race 4

P2D - Gibberella sp. (99%)* - - P7D + F. oxysporum (100%) - - P2S + F. oxysporum (100%) - - S2 + F. oxysporum (100%) - - S3 + F. oxysporum (99%) - - S6 + F. oxysporum (100%) - - S9 + F. oxysporum (100%) - - S11 + F. oxysporum (100%) - - S16 + F. oxysporum (100%) - - E + F. oxysporum (100%) - - MIAE01225 + / - - MIAE01226 + / - - MIAE01227  + / - - MIAE01228 + / - -

*identity

4.A.2.2 Pathogenicity test

The usage of specific primers is fast and preferred above pathogenicity tests because especially race 4 shows different aggressiveness under varying environmental conditions (Li et al., 2011a). However, no other race specific primers have been developed yet, as there is a difference between the artificial classification based on the pathogenic behaviour in the field and the phylogenetic classification (Ploetz et al., 2006). Because of this absence, time-consuming pathogenicity tests are still a necessity.

  26 

The pathogenicity of the Embrapa isolate was determined on Grande Naine, Silk and Silver Bluggoe. Due to a shortage of plantlets, we could not characterize all our isolates concerning their pathogenicity on the three cultivars. We selected two isolates (P7D and P2S) to be tested on Grande Naine and Silk. The isolates provided by Cirad were tested only on Grande Naine. The disease index, based on the amount of discolouration of leaves per plant, was followed through time giving an indication of the infection and disease process. Final evaluation was conducted four weeks after inoculation. At that moment, the vascular discolouration was calculated and the external and rhizomal discolouration were scored. Also the withering and rotting of the plantlets were observed.

Silk was clearly susceptible to the Embrapa (E) isolate (Figure 6), resulting in dead plantlets (100% disease index) after 22 days post inoculation. Remarkably, Grande Naine plantlets showed external disease symptoms as well. However, the disease severity was lower on this cultivar than on Silk, reaching the highest value of about 60% at 17 dpi. Silver Bluggoe was also affected by inoculation, but the disease index remained stable and lower than 30% after 15 dpi.  

Figure 6 Disease index (%) through time (dpi= days post inoculation) for the Embrapa isolate (E) tested on Grande Naine (GN), Silk and Silver Bluggoe (SB) cultivars. Error bars indicate the standard error.  

The behaviour of the P2S isolate (Figure 7) was similar to the Embrapa isolate regarding the interaction with Grande Naine (disease index varied around 60%). The P2S isolate did not cause disease on Silk in the first 13 days after inoculation, but thereafter a strong increase in the disease index was noticed, which at the final stage was higher than the disease index of the Grande Naine plantlets.

0 

20 

40 

60 

80 

100 

7  9  11  13  15  17  19  21  23  25  27 

Disease index (%

) 

dpi 

Embrapa

Ctrl GN 

Ctrl Silk 

Ctrl SB 

E GN 

E Silk 

E SB 

  27 

Figure 7 Disease index (%) through time (dpi= days post inoculation) for the P2S isolate tested on Grande Naine (GN) and Silk cultivars. Error bars indicate the standard error.  

The P7D isolate (Figure 8) affected Grande Naine, but in general, the disease index was slightly lower compared with the Embrapa isolate. Similar as the P2S isolate, no external symptoms were observed on inoculated Silk plantlets during the first 13 days after inoculation. From that time point, the disease severity steadily increased, reaching the disease index of 100% at 28 dpi.

Figure 8 Disease index (%) through time (dpi= days post inoculation) for the P7D isolate tested on Grande Naine (GN) and Silk cultivars. Error bars indicate the standard error.  

The isolates provided by Cirad were tested only on Grande Naine (Figure 9). Only after 15 days, the amount of external symptoms became prominent. Clear differences in pathogenicity were observed among the isolates: MIAE01227 was the most aggressive one, followed by MIAE01228 and MIAE01226, while MIAE01225 showed few external disease symptoms.

0 

20 

40 

60 

80 

100 

7  9  11  13  15  17  19  21  23  25  27 

Disease index (%

) 

dpi 

P2S

Ctrl GN 

Ctrl Silk 

P2S GN 

P2S Silk 

0 

20 

40 

60 

80 

100 

7  9  11  13  15  17  19  21  23  25  27 

Disease index (%

) 

dpi 

P7D

Ctrl GN 

Ctrl Silk 

P7D GN 

P7D Silk 

  28 

Figure 9 Disease index (%) through time (dpi= days post inoculation) for the MIAE01225, MIAE01226, MIAE01227 and MIAE01228 isolates from Cirad tested on Grande Naine (GN) cultivar. Error bars indicate the standard error.  

At the final evaluation, 28 days after the inoculation, the Embrapa isolate demonstrated the highest virulence towards the Silk bananas, resulting in dead plantlets (Figure 10). It also affected Grande Naine but to a significantly smaller extent. The disease index of Silver Bluggoe (25%) did not significantly differ from the control. P2S affected both Grande Naine and Silk plantlets. Also P7D affected Silk, and Grande Naine to a significantly smaller extent. From the Cirad strains, MIAE01227 behaved the most virulent followed by MIAE01228. The external leaf symptoms of MIAE01225 (17%) were comparable with the control (12%). After isolation from the infected plants, the isolates showed the expected morphological characteristics, confirming Koch’s postulate.

Figure 10 Disease index (%) at final evaluation (28 days post inoculation) grouped per isolate. Mann‐Whitney U test (p=0.05). Black  letters  indicate significant differences between different cultivars of one  isolate, while red  letters  indicate  the  significant  differences  between  the  different  isolates  of  CIRAD  on  Grande  Naine cultivars. Asterix= Significantly different from the control. Ctrl= Control, GN= Grande Naine, SB= Silver Bluggoe, M25= MIAEO1225, M26= MIAEO1226, M27= MIAE01227, M28= MIAE01228. 

0 

20 

40 

60 

80 

100 

7  9  11  13  15  17  19  21  23  25  27 

Disease index (%

) 

dpi 

Cirad

CtrlGN

MIAE01225GN

MIAE01226GN

MIAE01227GN

MIAE01228GN

b 

a ab 

b 

c 

a 

a 

a 

a 

b 

a 

a 

b ab 

0 10 20 30 40 50 60 70 80 90 100 

GN  Silk  SB  GN  Silk  SB  GN  Silk  GN  Silk  GN  GN  GN  GN 

Ctrl  Embrapa  P2S  P7D  M25  M26  M27  M28 

 Disease index (%

) 

 *   *    *    *   *  * * *   * 

  29 

Leaf increment, the amount of new leaves developed after inoculation, may influence the disease index as the calculation is based on the total amount of leaves. In our test, the leaf increment of all cultivars was not significantly different from the control, and within one isolate group no significant differences could be observed as well (Mann-Whitney U tests, p=0.05; data not shown). Nevertheless, the stunted leaf growth of Grande Naine plantlets infected with MIAE01226 and MIAE01227 was remarkable. The Silk banana plantlets inoculated with P7D even stopped producing new leaves.

The external colour of the pseudostem was scored (Figure 11). Inoculation with MIAE01227 caused significant browning of the pseudostem of Grande Naine plantlets. Also Silk plantlets infected with P7D showed a brown pseudostem. Light discolouration could be seen on Grande Naine plantlets infected with MIAE01226 and MIAE01228 and on the pseudostem of the Silk plantlets infected with P2S. Remarkably, the pseudostem of the Silk banana plantlets infected with the Embrapa isolate stayed completely green.

 

Figure  11  Scoring  of  the  external  colour  of  the  pseudostem:  0=  Green,  1=  Light  Brown,  2=  Brown.  Mann‐Whitney U test  (p = 0.05). Black  letters  indicate significant differences between different cultivars of one isolate, while red letters indicate the significant differences between the different isolates of CIRAD on Grande Naine cultivars. Asterix= Significantly different from the control. Ctrl= Control, GN= Grande Naine, SB= Silver Bluggoe, M25= MIAEO1225, M26= MIAEO1226, M27= MIAE01227, M28= MIAE01228. 

 Surprisingly, no or very little vascular discolouration (Figure 12) was observed from the Embrapa isolate. Also plants inoculated with P2S did not show significant vascular discolouration, although a trend could be remarked with the Silk plantlets showing a higher internal discolouration of 51% compared with the Grande Naine plantlets (12%). This trend was demonstrated to be significant for the P7D isolate. MIAE01226 and MIAE01227 demonstrated significant vascular discolouration compared with the control. MIAE01228 showed a high variability, which might explain the fact that it is not significantly different from the control although an internal discolouration of 67% could be observed.

a  a  a  a  a  a 

a 

a 

a 

b 

a 

a 

a a 

0 0,2 0,4 0,6 0,8 1 

1,2 1,4 1,6 1,8 2 

GN  Silk  SB  GN  Silk  SB  GN  Silk  GN  Silk  GN  GN  GN  GN 

Ctrl  Embrapa  P2S  P7D  M25  M26  M27  M28 

Score 

External colour pseudostem 

  *  * 

  30 

 

Figure  12 Vascular  discolouration:  the  relative  length of  the discoloured  internal  lesion divided by  the  total length  of  the  pseudostem.  Mann‐Whitney  U  test  (p=0.05). Black  letters  indicate  significant  differences between different  cultivars of one  isolate, while  red  letters  indicate  the  significant differences between  the different  isolates of CIRAD on Grande Naine  cultivars. Asterix= Significantly different  from  the  control. Ctrl= Control,  GN= Grande Naine,  SB  =  Silver  Bluggoe, M25  = MIAEO1225, M26= MIAEO1226, M27= MIAE01227, M28= MIAE01228.

Although no internal vascular discolouration of the pseudostem could be observed for the Embrapa isolate, a discoloured trace of 10% of the stellar region of the rhizome was visible in Grande Naine, Silk and Silver Bluggoe plantlets (Figure 13). Furthermore, the roots did clearly not survive the infection with the Embrapa isolate. Grande Naine and Silver Bluggoe succeeded to produce new rootlets above the infection while Silk did not recover. Although P2S and P7D did not cause a lot of external symptoms on the Grande Naine plantlets, 21-50% of the stellar region was discoloured. The same is true for the stellar region of the plantlets treated with the MIAE01225 isolate, which showed discolouration between 6-20%, although it appeared non-pathogenic according to the external symptoms. Nevertheless, no significant differences within the isolate groups could be observed, probably because the differences between the values of the scoring scale were too little to obtain a significant difference with a non-parametric test.

a  a  a  a  a  a 

a 

a 

a 

b 

a 

ab 

b ab 

0 10 20 30 40 50 60 70 80 90 100 

GN  Silk  SB  GN  Silk  SB  GN  Silk  GN  Silk  GN  GN  GN  GN 

Ctrl  Embrapa  P2S  P7D  M25  M26  M27  M28 

Relative lesion length (%

) Vascular discolouration  

*  *   * 

  31 

Figure 13 Scoring of the rhizomal discolouration: 1= No discolouration of tissue of stellar region of rhizome or surrounding tissue, 2= No discolouration of stellar region of rhizome. But discolouration at junction of root and rhizome, 3= A trace to 5% of stellar region discoloured, 4= 6‐20% of stellar region discoloured. 5= 21‐50% of stellar  region  discoloured,  6= More  than  50%  of  stellar  region  discoloured,  7=  Discolouration  of  the  entire rhizome,  8=  Dead  plant.  Mann‐Whitney  U  test  (p=0.05). Black  letters  indicate  significant  differences between different  cultivars of one  isolate, while  red  letters  indicate  the  significant differences between  the different  isolates of CIRAD on Grande Naine  cultivars. Asterix= Significantly different  from  the  control. Ctrl= Control,  GN= Grande Naine,  SB  =  Silver  Bluggoe, M25  = MIAEO1225, M26= MIAEO1226, M27= MIAE01227, M28= MIAE01228. 

It could be observed that the external discolouration of the pseudostem, the internal vascular discolouration and the rhizome discolouration are correlated. Indeed, after performing a Spearman correlation test, following correlation coefficients could be confirmed (Table 8).

Table 8 Results of a Spearman correlation test (p=0.01), indicating a strong correlation between the different discolouration indicators. * indicates significant correlation 

Indicators Discolouration

Correlation Coefficient rS

External-Vascular 0.937* Vascular-Rhizome 0.850* Rhizome-External 0.840*  

Additional information could be observed externally (pictures in attachment 1). Given the withering of the plant as a whole, Grande Naine plantlets stayed erect after infection with Embrapa, P2S and P7D isolates. However, they caused withering of the Silk banana plantlets. Again MIAE01227 caused withering of the Grande Naine plantlets, while MIAE01225 seemed to be non-pathogenic. The isolates MIAE01227 and MIAE01228 caused rotting of the Grande Naine plantlets, while P2S and P7D caused rotting of the Silk bananas.

 

a  a  a 

a a 

a 

a 

a 

a 

a 

a a 

a  a 

1 

2 

3 

4 

5 

6 

7 

8 

GN  Silk  SB  GN  Silk  SB  GN  Silk  GN  Silk  GN  GN  GN  GN 

Ctrl  Embrapa  P2S  P7D  M25  M26  M27  M28 

Score 

Rhizome discolouration  

*      *      *    *     *      *     *     *      *     *    * 

  32 

Embrapa

The Embrapa isolate clearly attacked the rootlets, causing rot. This resulted in immediate withering of the leaves the first days after inoculation. The Grande Naine and Silver Bluggoe plantlets could recover afterwards, because of newly produced roots at the rhizome. Although no remarkable vascular discolouration could be observed, rotting of the roots appeared sufficient to cause death. A possible explanation for the lack of vascular discolouration could be that the plant transpiration was stopped as a result of the attacked roots and the fungus was no longer transported through the vascular system.

P2S

The isolate caused disease in Silk plantlets with clear vascular and rhizomal discolouration, together with rotting of the pseudostem. The disease symptoms of the Silk banana plantlets appeared only after two weeks. However, from the Silk plantlets one stayed healthy explaining the high variability. Remarkably, the isolate was obtained from a banana plant without external symptoms as well. Therefore repetition would be interesting. Although threatened immediately after inoculation and the observation of discolouration of the stellar region of the rhizome, the Grande Naine plantlets showed less severe disease symptoms that stayed stable through time. Also, vascular discolouration was present but not extended.

P7D

The P7D isolate behaved similar to the P2S isolate through time. However, it demonstrated to be even more virulent towards the Silk banana plantlets than towards the Grande Naine plantlets. The Grande Naine plantlets were slightly affected as well, but repetition is necessary as the degree of susceptibility was not too clear. P7D could cause disease symptoms in Grande Naine plantlets, but to a little extent and they seemed to recover.

Race 4 isolates

The first symptoms only appeared after two weeks, but thereafter the disease developed fast. MIAE01227 is the most aggressive isolate; withering, almost no leaf increment, dark browning of the pseudostem and subsequent rotting, internal and external discolouration were clearly shown significant. Nevertheless, MIAE01228 showed the same virulence but one plant stayed healthy which influenced the mean. An error during inoculation could be the cause and therefore repetition will be necessary. Grande Naine plantlets were also susceptible to MIAE01226, but it was less virulent. MIAE01225 caused little disease symptoms; an internal discolouration was visible, although not significantly different from the control and the rhizome showed significant discolouration, but as a whole the plantlets appeared healthy.

The typical Foc wilting symptoms can be explained by the plugging and the toxin theory. According to the plugging theory the vessels of the infected plant are plugged by fungal hyphae, callose, tylose and gel. These substances can impair water transport, finally resulting in water deficiency and wilting. More recently, it was found that Foc produces fusaric acid (5-n-butyl-pyridine-2-carboxylic acid), which attacks the cell membranes of the leaves, resulting in leaf membrane injury. Subsequently, the leaves lose water despite stomatal closure and absence of transpiration rate (Dong et al., 2012). Foc also produces moniliformin, but at very low levels (O’Donnell et al., 2009).

According to the described reactions of the different cultivars, it is possible to differentiate between following disease processes: 1) plantlets that showed disease

  33 

symptoms from the beginning but remained stable or recovered 2) plantlets that showed disease symptoms from the beginning and did not recover, 3) plantlets that only demonstrated the disease after two weeks and deteriorated (Figure 14).

Although the resistance mechanisms of the banana plants against Fusarium wilt are not yet fully understood, research has revealed part of the process. It is known that the pathogen behaves non-specific as it also enters non-host plants (Nelson, 1981). However, Li et al. (2011b) tested the effect of root exudates from resistant banana varieties on the spore germination and hyphal growth of Foc. They observed the absence of the attachment and the penetration of Foc on resistant banana rootlets and therefore concluded that chemicals released in the root environment could contribute to the conductive resistance. Nevertheless, during this experiment penetration of the root system probably happened. By the inoculation treatment small wounds were certainly created and as it is known that wounding of the roots favours penetration (Nelson, 1981), it was most possible for Foc to enter. This explains the discolouration of the rhizomes in all plantlets except the control. Thus, the defence mechanism responsible for the distinction between susceptibility and resistance were activated after the penetration.

After penetration, the host reacts with a non-specific vascular occlusion. Vascular occluding gels are formed that at least temporarily immobilize spores above the trapping site. Thereafter the vascular parenchyma cells form tyloses that wall off the lumena of the infected vessels completely above the trapping sites. By synthesis and release of phenolic substances these structures are then lignified. This reaction is also present in susceptible hosts, but because of a time gap between weakening of the gels and occlusion by tyloses the vessels remain open for a certain period. In this way successive generations of spores can be carried for long distances from trapping site to trapping site. Thus, because of this time gap, rapid systemic colonization can be achieved in the susceptible host (Beckman et al., 1990).

This can possibly explain the in this thesis observed differences in disease symptoms through time. Primarily foliar symptoms might have been caused by the total occlusion of the vascular system. If this occlusion had the capacity to fully trap the disease, new vascular tissue could be produced and the plant recovered. This process could possibly explain the first observed disease process. Plants that reacted in this way were considered resistant although initial disease symptoms could be visible. Former pathogenicity tests also considered plants resistant when they initially showed symptoms but remained stable through time or recovered completely (Paul et al., 2011; Ghag et al., 2012). The attack of the Embrapa isolate on the Silk banana plantlets was very aggressive on the roots, therefore immediately disease symptoms were shown and as the plant was not able to stop the disease and produce new rootlets, it was considered susceptible. The delay of the disease symptoms in the third observed disease process may be explained by the initial continued transpiration as the host failed to create the tyloses in time. However, after the incubation period the pathogen had probably colonized the vascular system causing fatal occlusion. Therefore these plants were defined as susceptible.

  34 

Figure 14 Decision tree as a basis for the conclusion about the outcome of the interactions between isolates and  cultivars.  E=  Embrapa, M26= MIAE01226, M27= MIAE01227, M28= MIAE01228, M29= MIAE01229,  GN= Grande Naine and SB= Silver Bluggoe plantlets. 

When we follow the described reasoning (Figure 14), the pathogen-host interactions could be divided in resistant and susceptible interactions (Table 9). The pathogenicity test revealed that Embrapa could indeed be considered as a race 1. The isolates that were said to be race 4, indeed acted pathogenic towards Grande Naine, but with MIAE01225 behaving far less virulent. The P2S and P7D isolates behaved like race 1. The fact that the P2S isolate could be obtained from a banana plant without external disease symptoms, confirms that a time lag exists between infection and disease progress. It is therefore of main importance to follow the dynamics in disease progress of the sampling spots.

Table 9 Conclusion of the pathogenicity test 

Isolate Cultivar Susceptibility Embrapa Grande Naine Resistant Silk Susceptible Silver Bluggoe Resistant P2S Grande Naine Resistant Silk Susceptible P7D Grande Naine Resistant Silk Susceptible MIAE01225 Grande Naine Not defined* MIAE01226 Grande Naine Susceptible MIAE01227 Grande Naine Susceptible MIAE01228 Grande Naine Susceptible

* No external symptoms visible, but significant discolouration of the rhizome.

Symptoms 

Immediately Visible   

Stable 

Resistant E GN E SB

P2S GN P7D GN 

Increase 

Susceptible E Silk 

Only visible after 

incubation period 

Increase 

Susceptible P2S Silk P7D Silk

M26, M27, M28 GN 

  35 

Although a general disease process was described that allowed us to differentiate between susceptible and resistant varieties, the defence mechanisms differ of course between the cultivars as a consequence of breeding and (co)evolution. To understand these host-pathogen mechanisms is the subject of different investigations (Aguilar et al., 2000; De Ascencao & Dubery, 2000; Li et al., 2012b; Van den Berg et al., 2007) that mainly focus on the commercial Cavendish cultivar with the goal to understand the interaction with race 4. This knowledge can result in potential applications, like the use of silicon to protect the susceptible Maça cultivar against Foc (Fortunato et al., 2012). However, it would lead us beyond the scope of this thesis to revise them in detail. Second, the pathogenicity test is based on three replications per test and not all obtained isolates could be tested due to the time consuming replication and cultivation of the banana plantlets. Repetition is therefore essential. A second pathogenicity test cycle had been started, but final evaluation could not be done within the time-scope of this thesis.

 

  36 

4 B. Soil suppressiveness

4.B.1 Effect on surviving structures

To test the hypothesis that the soils around healthy banana plants are more suppressive than those around diseased banana plants, the effect of the soils on the surviving structures was investigated. In addition, the bacterial metabolic profiles of the soil samples were studied, using Biolog ecoplates.

Chlamydospores are asexual reproductive resting cells that play an important role in the long-term survival of the pathogen in the soil. They can be recognized as dark-rounded structures with a thick and multi-layered cell wall, which makes them resistant to chemical fumigation and other stress factors. Therefore, these survival structures are also formed as a reaction on harmful circumstances (Ohara & Tsuge, 2004; Goh et al., 2009). To investigate if factors in the soil contribute to suppressing Fusarium wilt, the effect of the soil on the chlamydospore survival was investigated.

4.B.1.1 Chlamydospore production

Corn flour medium Unfortunately, this method could not be properly evaluated. This was the first method tested and at that moment it was difficult to distinguish chlamydospores from other fungal spores and particles of the medium.

Soil agar medium After ten days of incubation, the chlamydospore production on the soil agar plates was evaluated. The loose mycelium from the edge of 21 plugs of 0.5 cm2 was taken with a tweezer, submerged in 5 ml water and mixed. This resulted in a concentration of 8.27 x 104 chlamydospores ml-1.

Plug in salt solution After two weeks shaking followed by two weeks resting, 4.6 x 104 chlamydospores ml-1 were obtained. A high variability in size and colour of chlamydospores could be observed.

Spore suspension in soil Of the added microconidia, 69% became chlamydospores (6.9 x 106 chlamydospores ml-1). Couteaudier & Alabouvette (1990) found that 39% of a low density of microconidia (3 x 103 cfu/g soil) of Fusarium oxysporum f.sp. lini turned into chlamydospores after three days of incubation in the soil. Therefore, this method is useful for the inoculation of soil. Moreover, the chlamydospores were checked after 3.5 months and they were still viable. This means that this method is interesting to be used for preservation of Foc isolates.

Chlamydospores function as a survival strategy in between new infection cycles. Therefore, it is not surprising that chlamydospore production is found to be higher in stress situations caused by biotic and abiotic factors (Hebbar et al., 1996; Goh et al., 2009). In literature, the formation of chlamydospores in Fusarium spp. has been well documented. In general, an environment low in carbon (energy) amended with salt (Nagao & Hattori, 1983), root exudates (Mondal et al., 1998) or bacterial exudates (Li et al., 2012a) provokes chlamydospore formation.

Salt solution and soil agar medium were both successful and practical methods. The salt solution method produced chlamydospores in a practical way, but a higher variability could be observed in the morphology of the chlamydospores. Therefore the question could be posed if they are fully representative or if the salt solution itself had an adverse effect on the chlamydospores. On the other hand, natural circumstances are mimicked with the soil agar method, possibly resulting in more realistic chlamydospores. However, a disadvantage of the soil agar method is the need for a high amount of plates to obtain sufficient production.

  37 

4.B.1.2 Soil effect on chlamydospores

The effect of the different soil samples on the surviving structures of Foc was investigated by burying three microscope slides with chlamydospores in the soil for 60 days. After germination was provoked overnight on potato dextrose broth, the chlamydospores were stained and the proportion of germinated chlamydospores was evaluated. To perform the binary logistic regression test, the obtained values of the three slides were summed.

Chlamydospore germination varied between 10 and 60% (Figure 15). Interestingly, the lowest and highest percentages of germination were found in the soils collected around plants considered as ‘healthy’ (S) and ‘diseased’ (D), respectively. However, when statistical analysis was performed, comparing the soils from healthy plants with those from diseased plants, no significant difference was detected (Mann-Whitney U test, p=0.05). After excluding sample 6S, which was also found as an outlier in the subsequent Biolog experiment, the germination percentage of chlamydospores buried in soils from the diseased banana plants was significantly higher than the chlamydospore germination after burying in the soils from the healthy plants (Mann-Whitney U test, p=0.05).

Figure 15 Chlamydospore germination on the slides after an incubation period of 60 days in the different soil samples and subsequent provocation of germination on potato dextrose broth. Different letters indicate significant differences by a Binary logistic regression test (p=0.05). The numbers of chlamydospores counted per sample (total sum of the three repetitions) and the soil pH are shown under the x-axis. The red columns demonstrate the average chlamydospore germination for the soils sampled around diseased banana plants (D) and for the soils sampled around the healthy banana plants (S).

Also microscope slides with microconidia were buried for 60 days and after staining, the proportion of formed chlamydospores was evaluated (Figure 16). The formation of chlamydospores out of microconidia was rather low. Remarkably, 6S and 5D showed higher relative chlamydospore formation. After performing a Mann-Whitney U test (p=0.05), no significant difference in formation of chlamydospores out of microconidia suspension could be found between the soils considered as healthy and diseased.

  38 

Figure 16 Chlamydospores formed out of microconidia on the slides after 60 days of incubation in the different soil samples. Different letters indicate significant differences by Binary logistic regression test (p=0.05). The numbers of (chlamydo)spores counted per sample are shown under the x-axis.

In banana production, soils suppressive to Fusarium wilt have always played an important role in coping with the disease. Nevertheless, the mechanisms responsible for disease suppression are still not known (Stover, 1990). Soil suppressiveness can be investigated by inoculating the soil with the pathogen and observing the effect of the soil on the disease symptoms of the plants. However, another approach is to look at the effect of the soil on the pathogen propagules in the soil (Sequeira, 1962). In this thesis, the chlamydospores of Foc were buried in the different soil samples and after incubation the viability was investigated through the germination capacity after induction on PDB. Because it was not possible to separate chlamydospores from micro- and macrocondidia, also microconidia were buried separately to investigate the new formation of chlamydospores. However, the direct formation of chlamydospores out of microconidia was low and therefore this did not influence the results of the germination capacity of the buried chlamydospores.

The average percentage of chlamydospore germination was low (maximum 60%) compared with preliminary experiments in our laboratory. Two factors could have influenced the germination rate. Firstly, the used protocol to provoke germination of the chlamydospores and secondly, the effect of the soil on the viability of the chlamydospores and subsequent on the germination capacity. The first factor could be rejected as preliminary experiments conducted on different soil to test the procedure provoked abundant germination of the chlamydospores of the same isolate. Therefore, the results suggest that the soils themselves influenced the observed germination of the chlamydospores.

As soil samples were taken around healthy banana plants, we expected to find a higher level of suppressiveness within this soil compared with the ‘diseased’ soil. Soil suppressiveness can be due to the effect of soil factors on different stages of the life cycle of a fungus, like the viability of the chlamydospores. The differences in observed germination are attributed to viability because we provoked germination artificially and, therefore, every chlamydospore is assumed to have the same chance to germinate. Viability of chlamydospores can be affected by abiotic stress situations like heat, which makes them more vulnerable to parasitism by microbial antagonists resulting in a lower germination capacity (Arora et al., 1996). Also changes in pH, temperature and matric potentials appeared to influence the viability of

  39 

chlamydospores (Mondal & Hyakumachi, 1998). However, all these abiotic factors were constant during our experiment.

Although direct penetration of ungerminated chlamydospores of F. oxysporum f. sp. raphani by antagonists has been observed (Toyota & Kimura, 1993), Sequeira (1962) has shown that the presence of unidentified gram-negative rod bacteria could cause lysis of germinated chlamydospores, while ungerminated chlamydospores were resistant. The presence of organic matter in combination with a high relative humidity increased the germination of the chlamydospores as carbohydrates became readily available. In addition, the activity of the specific saprophytic community related with the added organic amendment increased, which resulted in antagonism against Foc (with sugarcane being the most effective). Former research conducted around the same banana plants studied in the present thesis, revealed that the total soil organic matter, the labile organic matter, the total nitrogen content and the soil moisture content were significantly higher for soil samples taken around the healthy banana plants (Chavassieux, 2012). Although those sampling spots were not exactly the same spots sampled in this thesis, the banana plants were the same. Therefore, it was hypothesized that indeed the circumstances in the soil around healthy banana plants were optimal for chlamydospore germination and subsequent antagonism by the present soil bacterial community.

During the experiment, all soil samples had the same soil moisture content (28%), but differences in the soil organic matter could have led to germination during the incubation period. This would have made them more sensitive for antagonism, resulting in lysis. It was then expected that the ‘diseased’ soils had less favourable circumstances for germination and therefore the chlamydospores would have probably survived, until germination was provoked by PDB. Indeed, when we exclude 6S from the statistical analysis, the ‘diseased’ soils show a significantly higher germination capacity and thus viability compared with the ‘healthy’ soils. Also less chlamydospores could be found in the healthy soils (except for 7S).

In opposite to the findings of Sequeira (1962), Peng et al. (1999) found that for most of the investigated factors, chlamydospore germination was positively correlated with disease incidence as factors that negatively influenced the germination, also reduced the disease severity. Also Dominguez et al. (2001) suggested that more available iron in conducive soils, leads to higher chlamydospore germination and consequently to more disease. In addition, germination of (chlamydo)spores is density dependent (Couteaudier & Alabouvette, 1990). An explanation for the contradiction with Sequeira (1962) might be that when antagonistic activity is low, higher germination rate will lead to more disease incidence, but when antagonists are present the germinated chlamydospores are an easier target.

In summary, it was expected that the chlamydospore viability and subsequently the germination capacity would differ between the ‘diseased’ and ‘healthy’ soils. The viability was expected to be lower in the suppressive soil due to the presence of a microbiological community attacking the chlamydospores. Two hypotheses were described; or the chlamydospores germinated before induction on PDB and were subsequently harmed by the bacterial community, or they lost the viability due to direct adverse environmental factors. Nevertheless, no significant difference in chlamydospore viability could be observed between the two groups. It also should be noted that a variability within the soil samples of the ‘healthy’ and ‘diseased’ considered soils exists as the environmental circumstances varied and that the differences observed might be caused by different processes. Despite this, when 6S was excluded from the analysis, indeed the ‘diseased’ soils had a significantly higher proportion of viable chlamydospores.

  40 

However, the used method still needs optimization. A consequence of the usage of slides is an uneven contact of the spores with the soil as some spores are directly in contact with the glass and others are placed on the surface of the agar. Although good results were obtained with other soil, in this experiment the agar was clearly degraded, which influenced the obtained results. The destruction of the agar could be a consequence of the pressure used to bury the slides in the compact and heavy clay soil. However, the chlamydospore slides that were buried in ‘healthy’ soils, except for 7S, showed other patterns of degradation, as the agar seemed to have disappeared. Degradation was also observed on the slides with microconidia in the soils 5D and 6S, which had the highest chlamydospore formation as well. The negative impact of the soil on the agar could result in formation of survival structures of the fungus, because these are produced under stressed circumstances. The significant difference could also be caused by the error made as a consequence of a lower countable surface. The pH of all samples varied around 7 indicating that the soil was not remarkably acid or basic. Therefore, the hypothesis was formulated that biotic factors might attack the agar. Different agarase producing bacteria have been described. They can cause enzymatic hydrolysis of the agarose in the agar resulting in agar breakdown. Most of these bacteria are found in (marine) aquatic environment, as agar is produced by algae. However, Hosoda et al. (2003) isolated different agar-degrading Paenibacillus strains from the rhizosphere of spinach. In agriculture, Paenibacillus spp. are known for nitrogen fixation and plant growth promotion (Lal & Tabacchioni al., 2009). Klein et al. (2013) found the interesting fact that in soils suppressive for Fusarium oxysporum f.sp. radicis-cucumerinum, there was a shift in microbial community with among other bacteria more Paenibacillus spp. More specific, Paenibacillus polymyxa E681 showed in vitro antifungal activity against F. oxysporum by the production of Fusaricidin (Choi et al., 2009). Paenibacillus spp. had also been encountered in the roots of in vitro bananas as an endophyte (Thomas & Soly, 2009). In Brazil, different Paenibacillus spp. had already been extracted from the soil (Lal & Tabacchioni, 2009). After reviewing this information, it is suggested that probably Paenibacillus spp. could be present in the soils and might have caused the degradation of the agar. However, it is obvious that the experiment should be repeated to check whether the agar is degraded again. If this were the case, it would be interesting to investigate the presence of these bacteria in the soils.

Repetition after optimization of the method is necessary and still useful to underpin the hypothesis that there exists a difference in soil suppressiveness between the ‘healthy’ and ‘diseased’ soils, which can be related to viability of the chlamydospores. Nevertheless, it might also be that other factors caused the observed variety in disease incidence in the field. Larkin et al. (1993) could not observe a significant difference in chlamydospore germination or pathogen propagule destruction of Fusarium oxysporum f.sp. niveum between conducive and suppressive soils. They postulate that even if the soils did not appear pathogen suppressive, disease suppressiveness could be explained by specific isolates, capable of reducing the disease incidence. Indeed, as Nasir et al. (2003) stated, disease severity does not conform to a dose-response model, because they did not find a relation between the amount of disease propagules and disease incidence. Therefore they suggest that the management of plant nutrition and genotypic selection to strengthen the plants is more likely to be effective. However as said before, suppressiveness is complex and depending on several factors (Janvier et al., 2007). Disease suppressive soils have been recognized for over 100 years and the mechanisms have been the subject of study for nearly four decades (Mazzola, 2002). Focusing on one aspect of the complex web, like the viability of chlamydospores, has to be considered a way to simplify the investigation of the system. Nevertheless, by simplifying, the interaction of the different factors is neglected. As stated before, suppressiveness can only be fully understood when all parameters are combined (Janvier et al., 2007).

  41 

4.B.2 Community Level Physiological Profiling (Biolog)

To analyse the functional diversity on community level in the soil samples, Biolog ecoplates were used.

First, the average well colour development (AWCD) was calculated, which indicates the activity of the microbial community per sample. The activity increased with time as expected, but the rate differed between the samples. Especially, the community of 6S showed lower activity compared with the others (Figure 17).

Figure 17 Average Well Colour Development (AWCD) in ecoplates (Biolog) measured during 7 days for the soil samples taken around diseased (A) and healthy (B) banana plants. Error bars indicate the standard error. 

The proportional substrate utilization efficiency (PSUE) was calculated as described in material and methods. Further analyses were conducted on the obtained PSUE. A PCA was conducted to investigate the carbon source consumption (Table 10) patterns of the bacterial communities of the soils. Sample 6S behaved clearly different from the other samples (Figure 18). As only 46.7% of the variability could be explained by the first two principal components, another PCA was performed, considering the groups of carbon sources (Preston-Mafham et al., 2002). When the sources were grouped, the first two components explained already 81% of the variability (Figure 19), but still a high variation between the subsamples could be

0 0,2 0,4 0,6 0,8 1 

1,2 1,4 1,6 1,8 2 

0  24  48  72  96  120  144  168 

Absorbance (590nm

) 

Time (h) 

A 

1D 

2D 

5D 

6D 

7D 

0 0,2 0,4 0,6 0,8 1 

1,2 1,4 1,6 1,8 2 

0  24  48  72  96  120  144  168 

Absorbance (590nm

) 

Time(h) 

B 1S 

4S 

5S 

6S 

7S 

  42 

observed. The most important sources of variation in the analysis were the groups Amides (V7) and Esters (V8).

Table 10 Variables used in the PCA indicating the different carbon sources 

 

Figure 18 Principal Component Analysis (PCA) conducted on the Proportional Substrate Utilization Efficiencies (PSUE).  The  variables  indicate  the  different  carbon  sources  of  the  Ecoplates.  The  three  repetitions  per  soil sample are shown separately. D= Soil samples from ‘diseased’ soil, S= Soil samples from ‘healthy’ soil. 

Comp.1

Comp.2

-0.8 -0.6 -0.4 -0.2 0.0 0.2

-0.8

-0.6

-0.4

-0.2

0.0

0.2

1D

1D

1D

2D2D

2D5D5D5D

6D6D

6D

7D

7D

7D1S

1S

1S

4S

4S4S5S

5S

5S

6S

6S

6S

7S7S7S

-20 -15 -10 -5 0 5

-20

-15

-10

-50

5

V1

V2

V3 V4

V5

V6

V7V8

V9

V10

V11V12

V13V14V15

V16

V17

V18

V19

V20V21

V22

V23

V24V25

V26

V27

V28

V29

V30V31

V1 Pyruvic acid methyl ester  V17  D‐Galacturonic Acid V2 Tween 40  V18  2‐Hydroxy Benzoic Acid V3 Tween 80  V19  4‐Hydroxy Benzoic Acid V4 α‐Cyclodextrin  V20  γ‐Hydroxybutyric Acid V5 Glycogen  V21  Itaconic Acid V6 D‐Cellobiose  V22  α‐Ketobutyric Acid V7 α‐D‐Lactose  V23  D‐Malic Acid V8 β‐Methyl‐D‐Glucoside  V24  L‐Arginine V9 D‐Xylose  V25  L‐Asparagine 

V10 i‐Erythritol  V26  L‐Phenylalanine V11 D‐Mannitol  V27  L‐Serine V12 N‐Acetyl‐D‐Glucosamine  V28  L‐Threonine V13 D‐Glucosaminic Acid  V29  Glycyl‐L‐Glutamic Acid V14 Glucose‐1‐Phosphate  V30  Phenylethylamine V15 D,L‐α‐Glycerol Phosphate  V31  Putrescine V16 D‐Galactonic Acid γ‐Lactone 

21.8% 

24.9% 

  43 

 

 

Figure 19 Principal Component Analysis (PCA) conducted on the Proportional Substrate Utilization Efficiencies (PSUE). The variables indicate different groups of carbon sources as shown in (A). The colours indicate which carbon sources of Table 10 are grouped together. The three repetitions per soil sample are shown separately. The legend of the symbols is given in (B).  

The diversity within the soil samples could be visualised by the Lorenz curve (Figure 20) and by calculating the Gini coefficient (Table 11). The larger the area between the perfect equality line and the Lorenz curve, the higher the Gini coefficient is and the lower the functional diversity. While the Lorenz curves for all the diseased

V1 Polymeric

V2 Carbohydrates

V3 Miscellaneous

V4 Carboxilic Acids

V5 Phenolic Acids

V6 Amino Acids

V7 Amides

V8 Esters

A B

  44 

considered soils coincide, there is more difference in diversity shown for the healthy considered soils. Especially 6S shows the highest Gini coefficient and therefore the lowest functional diversity, meaning there was a higher usage of certain carbon sources and a less even distribution in carbon source consumption. After performing a Mann-Whitney U test (p=0.05, data not shown) we could conclude that based on these samples, there is no significant difference between the Gini coefficient of healthy and diseased considered soils. Therefore the functional diversity within samples is not significantly different between healthy and diseased considered soils.

 

Figure 20 Lorenz curves of the Proportional Substrate Utilization Efficiency (PSUE) of the carbon sources of the Biolog Ecoplates. A) Samples from 'diseased' soil B) Samples from 'healthy' soil. The larger the area between the 45° curve and the Lorenz curve, the less diversity in used carbon sources.

Table 11 Gini coefficients indicating the diversity in carbon source consumption of the Biolog Ecoplates of the bacterial community extracted from soil samples taken around diseased (D) and healthy (S) banana plants.  The higer the Gini coefficient, the less diversity in used carbon sources. Letters indicate significant differences by Mann‐Whitney U test (p=0.05) within the D and the S group. 

Samples Gini Samples Gini 1D 0.22 a 1S 0.20 a 2D 0.21 a 4S 0.23 b 5D 0.21 a 5S 0.20 a 6D 0.23 a 6S 0.31 c 7D 0.24 a 7S 0.24 b

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 

,032 ,097 ,161 ,226 ,290 ,355 ,419 ,484 ,548 ,613 ,677 ,742 ,806 ,871 ,935 1,00 

Cumulative  PSUE 

Cumulative number of carbon sources  

1D 

2D 

5D 

6D 

7D 

45° 

A

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 

,032 ,097 ,161 ,226 ,290 ,355 ,419 ,484 ,548 ,613 ,677 ,742 ,806 ,871 ,935 1,00 

Cumulative PSUE 

Cumulative number of carbon sources 

1S 

4S 

5S 

6S 

7S 

45° 

B

  45 

As 6S showed a distinct pattern, another PC analysis was done, considering only the soils of healthy plants. From this analysis, 63.1% of the variability was explained by the first two components. Figure 21 shows the five carbon sources that demonstrated the highest variation between the samples. It is clear that the bacterial communities present in sample 6S used more pyruvic acid, tween 40 and D-glucosaminic acid and less L-arginine and phenylethylamine. However, only a significant difference could be demonstrated for L-arginine, while 1S demonstrated similar consumption of pyruvic acid and phenylethylamine. The high variability in the consumption on Tween 40 and D-Glucosaminic acid can explain the absence of a significant difference.

 

Figure 21 Proportional Substrate Utilization Efficiency (PSUE) of the carbon sources: V1= Pyruvic Acid, V2= Tween 40, V13= D‐Glucosaminic acid, V24= L‐arginine, V30= Phenylethylamine. Only the samples from the ‘healthy’ soil are shown. Letters indicate significant differences within the consumption of one carbon source (Mann‐Whitney U test, p=0.05). Error bars indicate the standard error.

As Alabouvette (1999) stated, suppressiveness is fundamentally microbial in nature. It results from complex microbial interactions between the pathogen and the saprophytic microflora. Possible direct interactions contributing to suppressiveness are competition, antibiosis and parasitism. Also indirect interactions can occur through induced resistance (Marois, 1990). In addition, the microbial community is dynamic and changes due to co-evolution, which can contribute to suppressiveness (Kinkel et al., 2011). Therefore significant differences in functional microbial community between soil samples collected around healthy and diseased banana plants were expected. However, no significant differences were observed. The fact that ten different sampling spots were chosen in a highly diversified system can possibly explain why no differences were detected between ‘healthy’ and ‘diseased’ soils. Also no profoundly differences between the samples as such could be observed (except for 6S). This does not mean that the soil spots contain the same microbial communities, but from a functional diversity approach no clear differences could be observed in the consumption of the investigated carbon sources.

Large microbial community diversity may offer the significant means for the creation of coevolutionary hotspots, contributing to the creation of disease-suppressive soil communities in agricultural fields (Kinkel et al., 2011). Nevertheless, the Lorenz curves show no significant difference in functional diversity within the samples. This means that there was no significant difference in the number of different carbon sources used (substrate richness) and the usage of the carbon sources (substrate evenness) between the soil samples.

ab  a 

a 

c 

ab a 

a  a c  b 

a a 

a b  b 

b  a a 

a  a 

a  a  a 

c 

b 

0 

2 

4 

6 

8 

10 

12 

14 

V1  V2  V13  V24  V30 

PSUE 

1S 

4S 

5S 

6S 

7S 

  46 

Although no differences in microbial community function and diversity between the groups of soils collected from ‘healthy’ and ‘diseased’ plants could be observed, the soil sample 6S showed a completely different pattern in the Biolog analysis compared with the other samples. It showed a lower microbial activity and a lower functional diversity with a higher consumption of pyruvic acid, Tween 40 and D-glucosaminic acid and a lower consumption of L-arginine and phenylethylamine compared with the other soil samples. Little information is available to explain this behaviour, but a hypothesis is formulated below. De Weerd (2012) carried out her studies in the same agroforestry system that we are studying in the present thesis. She described the surrounding trees of the same banana plants that we sampled and remarked that 6S was surrounded by banana Prata (Musa spp.), Ipê amarelo (Tabebuia spp.), coffee (Coffea spp.) and Embaúba (Cecropia glaziovi). The roots of Angico Vermelho (Piptadenia rigida) and Fedegoso (Senna macranthera) were also found in the rhizosphere of the investigated banana plant. Although the trees were found around other banana plants as well, the last one was only found close to the sampling spot 6S. Senna macranthera was found to inhibit the in vitro growth of Staphylococcus aureus strains isolated from bovine mastitis (Diaz et al., 2010). Also chrysophanol, chrysophanol-8-methyl ether and physcion, which are known to have antimicrobial characteristics, were extracted from its bark (Branco et al., 2011). This information prompts us to speculate that the tree Fedegoso may cause the spread of antimicrobial compounds, which would explain the low microbial activity in the soil samples around 6S. If indeed this tree or another one would have an adverse effect on the microbiological life of this spot than this would also explain the extremely high hydraulic conductivity of soil 6S compared with the other samples found by Vacas (2012). Microbiological clogging is found to lower the hydraulic conductivity of the soil (Seki et al., 1998). In other words, a lower quantity of certain microorganisms could possibly enhance the hydraulic conductivity of the soil.

Although the Biolog Ecoplates are a fast and easy technique to get a first image about the soil community, the provided information should not be overestimated. Since microbes are located on and within soil particles, even after shaking the sample with buffer a significant amount of the community will be left in the sediment. The minimal density necessary to obtain a positive reaction in the wells is strain-dependent (Verschuere et al., 1997) and this density can depend on the culturability of the strain or the lack of suitable carbon sources in the plates (Haack et al., 1995; Degens & Harris, 1997), while fast growing bacteria can influence the obtained profile as well (Verschuere et al., 1997). However, for comparison of different sites, it provides a potential insight into the functional ability of the community. Although it is only from a limited portion of the community, all samples are treated in the same way, which makes comparison possible (Preston-Mafham et al., 2002). Nevertheless, the fact that no significant differences in community level physiological profile between the soils samples could be observed does not mean that the microbial communities of the soil samples are equal. Furthermore, through this method we could not obtain information about the identification of the microbial diversity, therefore the possibility that antagonistic interactions occur in the ‘healthy’ soils could not be rejected.

 

  47 

5. Conclusion  

Starting a new research topic is a time-consuming and difficult process especially when a complex system is the subject of investigation. This thesis demonstrates the results of the first steps to get insight in the complexity of disease suppression of Fusarium wilt on banana in an agroforestry system.

So far the conclusions of our research are:

• The presence of Foc in the investigated agroforestry system was confirmed by isolation and characterization of the pathogen.

• Two isolates were characterized as Foc race 1 and six isolates still need to be tested.

• Neither the chlamydospore germination test nor ‘Community Level Physiological Profiles’ (Biolog) could clearly distinguish ‘healthy’ and ‘diseased’ soils.

• The methods to produce chlamysdospores and test their viability in soil still need to be optimized.

Although the results of the soil suppressiveness experiments demonstrate that the specific hypotheses are not confirmed, still the general hypothesis that a certain degree of suppressiveness is present in the investigated agroforestry system should not be rejected. The conducted experiments are a first step in analysing the system and plenty of other interactions can still be researched.

Some specific recommendations for future work are:

• The unexpected degradation of the agar on the microscope slides influenced the results of the test on chlamydospore viability. Therefore, the experiment should be repeated with more caution during the burying step and with less soil per jar. In this way, the pressure on the slides would be diminished. If agar degradation occurs again, it could be interesting to investigate the presence of Paenibacillus spp. in the soils.

• Remarkably, the 6S soil sample demonstrated a different profile and a lower functional diversity. It would be interesting to obtain more samples from this spot and to see if there is a correlation with the litter of Senna macranthera.

• More samples should be taken in the field and other experiments should be performed to study the contribution of biological factors to soil suppressiveness. For example, natural and sterilized soils can be inoculated with Foc and the disease on banana plantlets can be investigated.

Specific research on local scale is a first step in understanding the agroforestry system. Given the number of biotic and abiotic parameters potentially related to soil suppressiveness, it seems obvious that a holistic approach, involving several areas of science has to be followed (Janvier et al., 2007).

  48 

References  

Aguilar E.A., Turner D.W. & Sivasithamparam K. (2000) Fusarium oxysporum f. sp. cubense inoculation and hypoxia alter peroxidase and phenylalanine ammonia lyase activities in nodal roots of banana cultivars (Musa sp.) differing in their susceptibility to Fusarium wilt. Australian Journal of Botany 48:589-596

Akila R., Rajendran L., Harish S., Saveetha K., Raguchander T. & Samiyappan R. (2011) Combined application of botanical formulations and biocontrol agents for the management of Fusarium oxysporum f.sp. cubense (Foc) causing Fusarium wilt in banana. Biological Control 57:175-183

Alabouvette C. (1999) Fusarium wilt suppressive soils: an example of disease-suppressive soils. Australasian Plant Pathology 28:57-64

Al-Hetar M.Y., Abidin M.A.Z., Sariah M. & Wong M.Y. (2011) Antifungal activity of chitosan against Fusarium oxysporum f. sp. cubense. Journal of Applied Polymer Science 120:2434-2439

Altieri M.A. (1999) The ecological role of biodiversity in agroecosystems. Agriculture, Ecosystems and Environment 74:19-31

Altieri M.A. (2009) Agroecology, Small Farms, and Food Sovereignty. ProQuest Social Sciences Premium Collection 61:102-112

Altieri M.A., Funes-Monzote F.R. & Petersen P. (2012) Agroecologically efficient agricultural systems for smallholder farmers: contribution to food sovereignty. Agronomy for Sustainable Development 32:1-13

Amir H. & Alabouvette C. (1993) Involvement of soil abiotic factors in the mechanisms of soil suppressiveness to fusarium wilts. Soil Biology & Biochemistry 25:157-164

Arora D.K., Pandey A.K. & Srivastva A.K. (1996) Effects of heat stress on loss of C; germination and pathogenicity from chlamydospores of Fusarium oxysporum f.sp. ciceri. Soil Biology & Biochemistry 28:399-407

Baayen R.P., O'Donnell K., Bonants J.M., Cigelnik, Kroon L.P.N.M., Roebroeck E.J.A. & Waalwijk C. (2000) Gene genealogies and AFLP analyses in the Fusarium oxysporum complex identify monophyletic and nonmonoplyletic formae speciales causing wilt and rot disease. Phytopathology 90:891-900

Beckman C.H. (1990) Host responses to the pathogen In: Ploetz R.C. Fusarium Wilt of Banana. American Phytopathological Society Press, St.Paul, Minnesota, USA 140p.

Belgrove A., Steinberg C. & Viljoen, A. (2011) Evaluation of nonpathogenic Fusarium oxysporum and Pseudomonas fluorescens for Panama disease control. Plant Disease 95:951-959

  49 

Bentley, S., Pegg, K. G., Moore, N. Y., Davis, R. D. & Buddenhagen, I. W. (1998) Genetic variation among vegetative compatibility groups of Fusarium oxysporum f. sp. cubense analyzed by DNA fingerprinting. Phytopathology 88:1283-1293.

Bonanomi G., Antignani V., Capodilupo M. & Scala F. (2010) Identifying the characteristics of organic soil amendments that suppress soilborne plant diseases. Soil Biology & Biochemistry 42:136-144

Booth C. (1971) The genus Fusarium. Commonwealth Mycological Institute, Kew, England 237p.

Branco A., Pinto A.C., Schripsema J. & Braz-Filho R. (2011) Anthraquinones from the bark of Senna macranthera. Anais da Academia Brasileira de Ciências 83:1159-1163

Broad R. & Cavanagh J. (2012) The development and agriculture paradigms transformed: Reflections from the small-scale organic rice fields of the Philippines. The Journal of Peasant Studies 39:1181-1193

Cao L., Qiu Z., You J., Tan H. & Zhou S. (2005) Isolation and characterization of endophytic streptomycete antagonists of fusarium wilt pathogen from surface-sterilized banana roots. FEMS Microbiology Letters 247:147-152

Cardoso I.M., Guijt I., Franco F.S., Carvalho A.F., & Ferreira Neto P.S. (2001) Continual learning for agroforestry system design: university, NGO and farmer partnership in Minas Gerais, Brazil. Agricultural Systems 69:235-257

Cardoso I.M. & Ferrari E.A. (2006) Construindo o conhecimento agroecológico: trajetoria de interação entre ONG, universidade e organizações de agricultores. Agriculturas 3:28-32

Chavassieux D.C.M. (2012) Control of Fusarium wilt in an agroforestry system with banana Maçã (Musa acuminata) in Pedra Dourada, Minas Gerais, Brazil. Internship report Wageningen University, The Netherlands and Universidade Federal de Viçosa, Minas Gerais, Brazil 20p.

Choi K. & Dobbs F.C. (1999) Comparison of two kinds of Biolog microplates (GN and ECO) in their ability to distinguish among aquatic microbial communities. Journal of Microbiological Methods 36:203-213

Choi S., Park S., Kim R., Lee C., Kim J.F. & Park S. (2008) Identification and functional analysis of the fusaricidin biosynthetic gene of Paenibacillus polymyxa E68. Biochemical and Biophysical Research Communications 365:89-95

Cook R.J. & Baker K.F. (1983) The nature and practice of biological control of plant pathogens. The American Phytopathological Society Press St. Paul, Minnesota, USA 539p.

Cotxarrera L., Trillas-Gay M.I., Steinberg C. & Alabouvette C. (2002) Use of sewage sludge compost and Trichoderma asperellum isolates to suppress Fusarium wilt of tomato. Soil Biology & Biochemistry 34:467-476

  50 

Couteaudier Y. & Alabouvette C. (1990) Survival and inoculum potential of conidia and chlamydospores of Fusarium oxysporum f.sp. lini in soil. Canadian Journal of Microbiology 36:551-556

De Ascensao A.R.D.C.F. & Dubery I.A. (2000) Panama disease: Cell wall reinforcement in banana roots in response to elicitors from Fusarium oxysporum f. sp. cubense race four. Phytopathology 90:1173-1180.

Degens B.P. & Harris J.A. (1997) Development of a physiological approach to measuring the catabolic diversity of soil microbial communities. Soil Biology & Biochemistry 29:1309-1320

Deltour P. (2012) Beheersing van Rhizoctonia solani op sla door het bevorderen van de bodemweerbaarheid. Masterproef voorgedragen tot het behalen van de graad van Master in de bio-ingenieurswetenschappen: Landbouwkunde 76p.

De Schutter O. (2009) Mission to Brazil. United Nations A/HRC/13/33/Add.6 23p.

De Schutter O. (2010) Report submitted by the Special Rapporteur on the right to food, Olivier De Schutter. United Nations A/HRC/16/49 21p.

De Weerd N. (2012) Fusarium wilt in banana Maçã. Internship report Wageningen University, The Netherlands and Universidade Federal de Viçosa, Minas Gerais, Brazil 15p.

Diaz M.A.N., Rossi C.C., Mendonça V.R., Silva D.M., Ribon A.O.B., Aguilar A.P. & Munoz G.D. (2010) Screening of medicinal plants for antibacterial activities on Staphylococcus aureus strains isolated from bovine mastitis. Revista Brasileira de Farmacognosia 20:724-728

Dita M.A., Waalwijk C., Buddenhagen I.W., Souza M.T.J. & Kema G.H.J. (2010) A molecular diagnostic for tropical race 4 of the banana fusarium wilt pathogen. Plant Pathology 59:248-357

Domínguez J., Negrín M.A. & Rodríguez C.M. (2001) Aggregate water-stability, particle-size and soil solution properties in conducive and suppressive soils to Fusarium wilt of banana from Canary Islands (Spain). Soil Biology & Biochemistry 33:449-455

Dong X., Ling N., Wang M., Shen Q. & Guo S. (2012) Fusaric acid is a crucial factor in the disturbance of leaf water imbalance in Fusarium-infected banana plants. Plant Physiology and Biochemistry 60:171-179

Dubois T., Coyne D., Kahangi E., Turoop L. & Nsubuga E. N. (2006) Endophyte-enhanced banana tissue culture: technology transfer through public-private partnerships in Kenya and Uganda. African Technology Development Forum Journal 3:18-23

Ellis F. & Biggs S. (2001) Evolving Themes in Rural Development 1950s-2000s. Development Policy Review 19:437-448

  51 

Embrapa ed. Borges A. L. & Souza L.S. (2004) O cultivo da bananeira. Embrapa Mandioca e Fruticultura, Cruz das Almas, Bahia, Brazil 279p.

Embrapa ed. do Amaral Pereira F., Carneiro M.R. & de Andrade L.M. (2006) Marco referencial em agroecologia. Embrapa Informação Tecnológica, Brasilia, Federal District, Brazil 66p.

Fishal E.M.M., Meon S. & Yun W.M. (2010) Induction of tolerance to Fusarium wilt and defense-related mechanisms in the plantlets of Susceptible Berangan Banana Pre-Inoculated with Pseudomonas sp. (UPMP3) and Burkholderia sp. (UPMB3). Agricultural Sciences in China 9:1140-1149

Forsyth L.M., Smith L.J. & Aitken E.A.B. (2006) Identification and characterization of non-pathogenic Fusarium oxysporum capable of increasing and decreasing Fusarium wilt severity. Mycological Research 1:929-935

Fortunato A.A., Rodrigues F.A., Baroni J.C.P., Césat G., Soares B., Rodriguez M.A. & Pereira O.L. (2012) Silicon suppresses Fusarium wilt development in banana plants. Journal of Phytopathology 160:674-679

Fourie G., Steenkamp E.T., Ploetz R.C., Gordon T.R. & Viljoen A. (2011) Current status of the taxonomic position of Fusarium oxysporum formae specialis cubense within the Fusarium oxysporum complex. Infection, Genetics and Evolution 11:533-542

Francis C., Lieblein G., Gliessman S., Breland T.A., Creamer N., Harwood R., Salomonsson L., Helenius J., Rickerl D., Salvador R., Wiedenhoeft M., Simmons S., Allen P., Altieri M., Flora C. & Poincelot R. (2003) Agroecology: The Ecology of Food Systems. Journal of Sustainable Agriculture 22:99-118

Fravel D., Olivain C. & Alabouvette C. (2003) Fusarium oxysporum and its biocontrol. New Phytologist 157:493-502

Garland J.L. (1997) Analysis and interpretation of community-level physiological profiles in microbial ecology. FEMS Microbiology Ecology 24:289-300

Garland J.L. & Mills A.L. (1991) Classification and characterization of heterotrophic microbial communities on the basis of patterns of community-level sole-carbon-source utilization. Applied and Environmental Microbiology 57:2351-2359

Garland J.L., Mills A.L. & Young J.S. (2001) Relative effectiveness of kinetic analysis vs single point readings for classifying environmental samples based on community-level physiological profiles (CLPP). Soil Biology & Biochemistry 33:1059-1066

General Assembly United Nations (2012) International Year of Family Farming, 2014 A/RES/66/222 2p.

Getha K., Vikineswary S., Wong W.H., Seki T., Ward A. & Goodfellow M. (2005) Evaluation of Streptomyces sp. strain g10 for suppression of Fusarium wilt and rhizosphere colonization in pot-grown banana plantlets. Journal of Industrial Microbiology and Biotechnology 32:24-32

  52 

Ghag S.B., Shekhawat U.K.S., Ganapathi T.R. (2012) Petunia floral defensins with unique prodomains as novel candidates for development of Fusarium wilt resistance in transgenic banana plants. PLoS ONE 7: e39557

Goh Y.K., Daida P. & Vujanovic V. (2009) Effects of abiotic factors and biocontrol agents on chlamydospore formation in Fusarium graminearum and Fusarium sporotrichioides. Biocontrol Science and Technology 19:151-167

Gordon T.R. & Martyn R.D. (1997) The evolutionary biology of Fusarium oxysporum. Annual Review of Phytopathology 35:111-28

Grantina L., Kenigsvalde K., Eze D., Petrina Z., Skrabule I., Rostoks N. & Nikolajeva V. (2011) Impact of six-year-long organic cropping on soil microorganisms and crop disease suppressiveness. Žemdirbystė=Agriculture 98:399-408

Groenewald S., Van Den Berg N., Marasas W.F.O., Viljoen A. (2006) The application of high-throughput AFLP's in assessing genetic diversity in Fusarium oxysporum f. sp. cubense. Mycological research 1:297-305

Guckert J.B., Carr G.J., Johnson T.D., Hamm B.G., Davidson D.H. & Kumagai Y. (1996) Community analysis by Biolog: curve integration for statistical analysis of activated sludge microbial habitats. Journal of Microbiological Methods 27:183-197

Gurr G.M., Wratten S.D. & Luna J.M. (2003) Multi-function agricultural biodiversity: pest management and other benefits. Basic and Applied Ecology 4:107-116

Haack S.K., Garchow H., Klug M.J. & Forney L.J. (1995) Analysis of factors affecting the accuracy reproducability, and interpretation of microbial community carbon source utilization patterns. Applied and Environmental Microbiology 61:1458-1468

Harch B.D., Correll R.L., Meech W., Kirkby C.A. & Pankhurst C.E. (1997) Using the Gini coefficient with BIOLOG substrate utilisation data to provide an alternative quantitative measure for comparing bacterial soil communities. Journal of Microbiological Methods 30:91-101

Hebbar K.P., Lewis J.A., Poch S.M. & Lumsden R.D. (1996) Agricultural by-products as substrates for growth, conidiation and chlamydospore formation by a potential mycoherbicide, Fusarium oxysporum strain EN4. Biocontrol Science and Technology 6:263-276

Hosoda A., Sakai M. & Kanazawa S. (2003) Isolation and Characterization of Agar-degrading Paenibacillus spp. associated with the rhizosphere of spinach. Bioscience Biotechnology & Biochemistry 67:1048-1055

Huang Y.H., Wang R.C., Li C.H., Zuo C.W., Wei Y.R., Zhang L., Yi G.J. (2012) Control of Fusarium wilt in banana with Chinese leek. European Journal of Plant Pathology 134:87-95

IBGE (2006) Censo Agropecuário 2006. Agricultura familiar. Censo Agropecuário, Rio de Janeiro 267p.

  53 

Insunza V., Alström S. & Eriksson K.B. (2002) Root bacteria from nematicidal plants and their biocontrol potential against trichodorid nematodes in potato. Plant and Soil 241:271-27

Israeli Y., Lahav E. & Renveni O. (1995) In vitro culture of bananas p.155-167 In: Gowen S. (1995) Bananas & Plantains. Chapman & Hall, 2-6 Boundary Row, London

Jackson L.E., Pascual U. & Hodgking T. (2007) Utilizing and conserving agrobiodiversity in agricultural landscapes. Agriculture, Ecosystems and Environment 121:196-210

Jackson L.E., Pulleman M.M., Brussaard L., Bawa K.S., Brown G.G., Cardoso I.M., de Ruiter P.C., García-Barrios L., Hollander A.D., Lavelle P., Ouédraogo, Pascual U., Setty S., Smukler S.M., Tscharntke T. & Van Noordwijk (2012) Social-ecological and regional adaptation of agrobiodiversity management across a global set of research regions. Global Environmental Change 22:623-639

Janvier C., Villeneuve F., Alabouvette C., Edel-Hermann V., Mateille T. & Steinberg C. (2007) Soil health through soil disease suppression: which strategy from descriptors to indicators? Soil Biology & Biochemistry 39:1-23

Katan T. (1999) Current status of vegetative compatibility groups in Fusarium oxysporum. Phytoparasitica 27:51-64

Kinkel L.L., Bakker M.G. & Schlatter D.C. (2011) A coevolutionary framework for managing disease-suppressive soils. Annual review of phytopathology 49:47-67.

Kistler H.C. (1997) Genetic diversity in the plant-pathogenic fungus Fusarium oxysporum. Phytopathology 87:474-479

Klein E., Ofek M., Katan J., Minz D. & Gamliel A. (2013) Soil suppressiveness to Fusarium disease: shifts in root microbiome associated with reduction of pathogen root colonization. Phytopathology 103:23-33

Klotz L.V., Nelson P.E. & Toussoun T.A. (1988) A medium for enhancement of chlamydospore formation in Fusarium species. Mycologia 80:108-109

Koenig, R. L., Ploetz, R. C., and Kistler, H. C. (1997) Fusarium oxysporum f. sp. cubense consists of a small number of divergent and globally distributed clonal lineages. Phytopathology 87:915-923.

Koeppel D. (2011) The Beginning of the End for Bananas ? The Scientist 3p.

Lal S. & Tabacchioni S. (2009) Ecology and biotechnological potential of Paenibacillus polymyxa: a minireview. Indian Journal of Microbiology 49:2-10

Larkin R.P. & Fravel D.R. (1999) Mechanisms of action and dose-response relationships governing biological control of Fusarium wilt of tomato by nonpathogenic Fusarium spp. Phytopathology 89:1152-1161

  54 

Larkin R.P., Hopkins D.L. & Martin F.N. (1993) Ecology of Fusarium oxysporum f.sp. niveum in soils suppressive and conducive Fusarium wilt of watermelon. Phytopathology 83:1105-1116 Latz E., Eisenhauer N., Rall B.C., Allan E., Roscher C., Scheu S. & Jousset A. (2012) Plant diversity improves protection against soil-borne pathogens by fostering antagonistic bacterial communities. Journal of Ecology 100:597-604 Leslie J.F. & Summerell B.A. (2006) The Fusarium laboratory manual. Blackwell Publishing 388p.

Letourneau D.K., Armbrecht I., Rivera B.S., Lerma J.M, Carmona E.J., Daza M.C., Escobar S., Galindo V., Gutiérrez, Lopze S.D., Mejia J.L., Rangel A.M.A., Rangel J.H., Rivera L., Saavedra C.A., Torres A.M. & Trujillo A.R. (2011) Does plant diversity benefit agroecosystems? Ecological Applications 21:9-21

Lewis W.J., van Lenteren J.C., Phatak S.C. & Tumlinson J.H. (1997) A total system approach to sustainable pest management. Proceedings of the National Academy of Sciences of the United States of America 94:12243-12248

Li M.H., Yang B., Leng Y., Chao C.P., Liu J.M., He Z.F., Jiang Z.D. & Zhong S. (2011a) Molecular characterization of Fusarium oxysporum f.sp. cubense race 1 and 4 isolates from Taiwan and southern China. Canadian Journal of Plant Pathology 33:168-178

Li C., Chen S., Zuo C., Sun Q., Ye Q., Yi G. & Huang B. (2011b) The use of GFP-transformed isolates to study infection of banana with Fusarium oxysporum f.sp. cubense race 4. European Journal of Plant Pathology 131:327-340

Li W., Qian C., Mo Y., Hu Y. & Xie J. (2011c) Tolerance of banana for fusarium wilt is associated with early H202 accumulation in the roots. African Journal of Biotechnology 10:11378-11387

Li P., Ma L., Feng Y.L., Mo M.H., Yang F.X., Dai H.F., Zhao Y.X. (2012a) Diversity and chemotaxis of soil bacteria with antifungal activity against Fusarium wilt of banana. Journal of Industrial Microbiology and Biotechnology 39:1495-1505

Li C., Deng G., Yang J., Viljoen A., Jin Y., Kuang R., Zuo C., Lv Z., Yang Q., Sheng O, Wei Y., Hu C., Dong T. & Yi G. (2012b) Transcriptome profiling of resistant and susceptible Cavendish banana roots following inoculation with Fusarium oxysporum f. sp. cubense tropical race 4. BMC Genomics 13:374

Lian J., Wang Z., Cao L., Tan H., Inderbitzin P., Jiang Z. & Zhou S. (2009) Artificial inoculation of banana tissue culture plantlets with indigenous endophytes originally derived from native banana plants. Biological Control 51:427-434

Lin Y., Chang J., Liu E., Chao C., Huang J. & Chang P.L. (2009) Development of a molecular marker for specific detection of Fusarium oxysporum f. sp. cubense race 4. European Journal of Plant Pathology 123:353-365

Lugtenberg B. & Kamilova F. (2009) Plant-growth-promoting rhizobacteria. Annual Review of Microbiology 63:541-556

  55 

Mak C., Mohamed A.A., Liew K.W. & Ho Y.W. (2004) Early screening technique for Fusarium wilt resistance in banana micropropagated plants. In: Jain S.M. & Swennen R. Banana improvement: Cellular, Molecular Biology and Induced Mutations. Science Publishers, Inc. 400p.

Malézieux E., Crozat Y., Dupraz C., Laurans M., Makowski D., Ozier-Lafontaine H., Rapidel B., de Tourdonnet S., Valantin-Morison M. (2009) Mixing plant species in cropping: concepts, tools and models. Agronomy for Sustainable Development 29:43-62

Marois J.J. (1990) Biological control of diseases caused by Fusarium oxysporum In: Ploetz R.C. Fusarium Wilt of Banana. American Phytopathological Society Press, St.Paul, Minnesota, USA 140p.

Mazzola M. (2002) Mechanisms of natural soil suppressiveness to soilborne diseases. Antonie van Leeuwenhoek 81:557-564

Mishra P.K., Fox R.T.V. & Culham A. (2003) Development of a PCR-based assay for rapid and reliable identification of pathogenic Fusaria. FEMS Microbiology Letters 218:329-332

Mondal, S. N. & Hyakumachi, M. (1998) Carbon loss and germinability, viability, and virulence of chlamydospores of Fusarium solani f. sp. phaseoli after exposure to soil at different pH levels, temperatures, and matric potentials. Phytopathology 88:148-155

Moore N.Y., Bentley S., Pegg K.G. & Jones D.R. (1995) Musa disease fact sheet N°5 Fusarium wilt of banana. Inibap Montpellier 4p.

Nagao H. & Hattori T. (1983) Chlamydospore formation in hyphae of Fusarium oxysporum f.sp. raphani. Journal of General and Applied Microbiology 29:187-193

Naik P.R., Sahoo N., Goswami D., Ayyadurai N. & Sakthivel N. (2008) Genetic and functional diversity among fluorescent pseudomonads isolated from the rhizosphere of banana. Microbial Ecology 56:492-504

Nair P.K.R. (1993) An introduction to agroforestry. Kluwer Academic Publishers 500p.

Nasir N., Pittaway P.A. & Pegg K.G. (2003) Effect of organic amendments and solarisation on Fusarium wilt in susceptible banana plantlets, transplanted into naturally infested soil. Australian Journal of Agricultural Research 54:251-257

Nayaka S.C., Wulff E.G., Udayashankar A.C., Nandini B.P., Niranjana S.R., Mortensen C.N. & Prakash H.S. (2011) Prospects of molecular markers in Fusarium species diversity. Applied Microbiology and Biotechnology 90:1625-1639

Nel B., Steinberg C., Labuschagne N. & Viljoen A. (2006a) The potential of nonpathogenic Fusarium oxysporum and other biological control organisms for suppressing fusarium wilt of banana. Plant Pathology 55:217-223

  56 

Nel B., Steinberg C., Labuschagne N. & Viljoen A. (2006b) Isolation and characterization of nonpathogenic Fusarium oxysporum isolates from the rhizosphere of healthy banana plants. Plant Pathology 55:207-216

Nelson P.E. (1981) Life cycle and epidemiology of Fusarium oxysporum In: Mace M.E., Bell A.A. & Beckman C.H. (1981) Fungal wilt diseases of plants. Academic Press, Inc. Texas, USA 640p.

Nelson P.E., Toussoun T.A. & Marasas W.F.O. (1983) Fusarium species. An illustrated manual for identification. The Pennsylvania state University Press. University Park and London, 193p.

Nelson S.C., Ploetz R.C. & Kepler A.K. (2006) Musa species (banana and plantain) Species Profiles for Pacific Island Agroforestry. Permanent Agriculture Resources Hawai 33p.

Nicholls C.I. & Altieri M.A. (2004) Designing species-rich, pest-suppressive agroecosystems through habitat management. Agroecosystems Analysis, Agronomy Monograph 43:49-61

Noble R. & Coventry E. (2005) Suppression of soil-borne plant diseases with composts. Biocontrol Science and Technology 15:3-20

O'Donnell K., Gueidan C., Sink S., Johnston P.R., Crous P.W., Glenn A., Riley R., Zitomer N.C., Colyer P., Waalwijk C., van der Lee T., Moretti A., Khang S., Kim H., Geiser D.M., Juba J.H., Baayen R.P., Cromey M.G., Bithell S., Sutton D.A., Skovgaard K., Ploetz R., Kistler H.C., Elliott M., Davis M. & Sarver B.A.J. (2009) A two-locus DNA sequence database for typing plant and human pathogens within the Fusarium oxysporum species complex. Fungal Genetics and Biology 46:936-948

Ohara T. & Tsuge T. (2004) FoSTUA, encoding a basic helix-loop-helix protein, differentially regulates development of three kinds of asexual spores, macroconidia, microconidia, and chlamydospores, in the fungal plant pathogen Fusarium oxysporum. Eukaryotic Cell 3:1412-1422

Paul J., Becker D.K., Dickman M.B., Harding R.M., Khanna H.K. & Dale J.L. (2011) Apoptosis-related genes confer resistance to Fusarium wilt in transgenic 'Lady Finger' bananas. Plant Biotechnology Journal 9:1141-1148

Peng H.X., Sivasithamparam K., Turner D.W. (1999) Chlamydospore germination and Fusarium wilt of banana plantlets in suppressive and conducive soils are affected by physical and chemical factors. Soil Biology & Biochemistry 31:1363-1374

Pereira J.C.R., Pereira J.R., Castro M.E.A. & Gasparotto L. (2005) Occorência do Mal-do-Panama em bananeira do subgupo Figo em Piau, Minas Gerais. Fitopatologia Brasileira 30:554

Petrini O. (1991) Fungal endophytes in tree leaves. In: Andrews J.H. & Hirano S.S. Microbial Ecology of leaves. Springer, New York 499p.

  57 

Pietikäinen J., Hiukka R. & Fritze H. (2000) Does short-term heating of forest humus change its properties as a substrate for microbes? Soil Biology & Biochemistry 32:277-288

Ploetz R.C. (1990) Population biology of Fusarium oxysporum f.sp. cubense In: Ploetz R.C. Fusarium Wilt of Banana. American Phytopathological Society Press, St.Paul, Minnesota, USA 140p.

Ploetz, R.C. (2006) Fusarium wilt of banana is caused by several pathogens referred to as Fusarium oxysporum f. sp. cubense. Phytopathology 96:653-656.

Preston-Mafham J., Boddy L. & Randerson P.F. (2002) Analysis of microbial community functional diversity using sole-carbon-source utilisation profiles - a critique. FEMS Microbiology Ecology 42:1–14

Ratnadass A., Fernandes P., Avelino J. & Habib R. (2012) Plant species diversity for sustainable management of crop pests and diseases in agroecosystems. Agronomy for Sustainable Development 32:273-303

Rosset P. (2000) The Multiple Functions and Benefits of Small Farm Agriculture in the Context of Global Trade Negotiations. The Society for International Development. SAGE Publications 43:77–82

Saravanan T., Muthusamy M. & Marimuthu T. (2003) Development of integrated approach to manage the fusarial wilt of banana. Crop Protection 22:1117-1123

Schneider S. & Niederle P.A. (2010) Resistance strategies and diversification of rural livelihoods: the construction of autonomy among Brazilian family farmers. The Journal of Peasant Studies 37:379-405

Schroth G., Krauss U., Gasparotto L., Aguilar J.A.D. & Vohland K. (2000) Pests and diseases in agroforestry systems of the humid tropics. Agroforestry Systems 50:199-241

Seki K., Miyazaki T. & Nakano M. (1998) Effects of microorganisms on hydraulic conductivity decrease in infiltration. European Journal of Soil Science 49:231-236

Sequeira L. (1962) Influence of organic amendments on survival of Fusarium oxysporum f.sp; cubense in the soil. Phytopathology 52:976-982

Serra-Wittling C., Houot S. & Alabouvette C. (1996) Increased soil suppressiveness to Fusarium wilt of flax after addition of municipal solid waste compost. Soil Biology & Biochemistry 28:1207-1214

Souza H.N., Cardoso I.M., Fernandes J. M., Garcia F.C.P., Bonfim V.R., Santos A.C., Carvalho A.F. & Mendonça E.S. (2010) Selection of native trees for intercropping with coffee in the Atlantic Rainforest biome. Agroforestry Systems 80:1-16

  58 

Souza H.N., Cardoso I.M., de Sa' Mendonça E., Carvalho A.F., de Oliveira G.B., Gjorup D.F. & Bonfim V.R. (2012) Learning by doing: a participatory methodology for systematization of experiments with agroforestry systems, with an example of its application. Agroforestry Systems 85:247–262

Steiner C., Das K.C., Garcia M., Förster B. & Zech W. (2008) Charcoal and smoke extract stimulate the soil microbial community in a highly weathered xanthic Ferralsol. Pedobiologia 51:359-366

Stevenson I.L. & Becker S.A.W.E. (1972) The fine structure and development of chlamydospores of Fusarium oxysporum. Canadian Journal of Microbiology 18:997-1002

Stover R.H. (1990) Fusarium wilt of banana: Some history and current status of the disease In: Ploetz R.C. Fusarium Wilt of Banana. American Phytopathological Society Press, St.Paul, Minnesota, USA 140p.

Sun J.B., Peng M., Wang Y.G., Zhao P.J. & Xia Q.Y. (2011) Isolation and characterization of antagonistic bacteria against fusarium wilt and induction of defense related enzymes in banana. African Journal of Microbiology Research 5:509-515

Termorshuizen A.J. & Jeger M.J. (2008) Strategies of soilborne plant pathogenic fungi in relation to disease suppression. Fungal Ecology 1:108-114

Thangavelu R. & Mustaffa M.M. (2012) Current Advances in the Fusarium Wilt Disease Management in Banana with Emphasis on Biological Control In: Cumagun C. J. Plant Pathology InTech 362p.

Thangavelu R., Palaniswami A. & Velazhahan R. (2004) Mass production of Trichoderma harzianum for managing fusarium wilt of banana. Agriculture, Ecosystems and Environment 103:259-263

Thomas P. & Soly T.A. (2009) Endophytic bacteria associated with growing shoot tips of banana (Musa sp.) cv. Grand Naine and the affinity of endophytes to the host. Microbial Ecology 58:952-964

Ting A.S.Y., Mah S.W. & Tee C.S. (2012) Evaluating the feasibility of induced host resistance by endophytic isolate Penicillium citrinum BTF08 as a control mechanism for Fusarium wilt in banana plantlets. Biological Control 61:155-159

Toyota K. & Kimura M. (1993) Colonization of chlamydospores of Fusarium oxysporum f.sp. raphani by soil bacteria and their effects on germination. Soil Biology & Biochemistry 25:193-197

Tscharntke T., Clough Y., Wanger T.C., Jackson L., Motzke I., Perfecto I., Vandermeer J., Whitbread A. (2012) Global food security, biodiversity conservation and the future of agricultural intensification. Biological Conservation 151:53-59

  59 

Vacas R. L. (2012) Relação entre incidência de Fusarium oxysporum f.sp. cubense em bananeira e a qualidade do solo em propriedade rual manejo agroecológico. Internship report Universidade de Córdoba, Espanha and Universidade Federal de Viçosa, Minas Gerais, Brazil. 49p.

Van Buyten E., Banaay C.G.B., Cruz C.V. & Höfte M. (2012) Identity and variability of Pythium species associated with yield decline in aerobic rice cultivation in the Philippines. Plant Pathology 62:139-153

Van den Berg N., Berger D.K., Hein I., Birch P.R.J., Wingfield M.J. & Viljoen A. (2007) Tolerance in banana to Fusarium wilt is associated with early up-regulation of cell wall-strengthening genes in the roots. Molecular Plant Pathology 8:333-341

Van Hove F., Waalwijk C., Logrieco A., Munaut F. & Moretti A. (2011) Gibberella musae (Fusarium musae) sp. nov., a recently discovered species from banana is sister to F. verticillioides. Mycologia 103:570-585

Verschuere L., Fievez V., Van Vooren L. & Verstraete W. (1997) The contribution of individual populations to the Biolog pattern of model microbial communities. FEMS Microbiology Ecology 24:353-362

Weber O.B., Muniz C.R., Vitor A.O., Freire F.C.O. & Oliveira V.M. (2007) Interaction of endophytic diazotrophic bacteria and Fusarium oxysporum f. sp. cubense on plantlets of banana ‘Maçã’. Plant and Soil 298:47-56

Weller D.M., Raaijmakers J.M. & McSpadden G. B. B. (2002) Microbial populations responsible for specific soil suppressiveness to plant pathogens. Annual Review of Phytopathology 40:309-48

Wezel A., Bellon S., Doré T., Francis C., Vallod D. & David C. (2009) Agroecology as a science, a movement and a practice. Agronomy for Sustainable Development 29:503-515

Xu Q., Jiang P. & Xu Z. (2008) Soil microbial functional diversity under intensively managed bamboo plantations in southern China. Journal of Soils and Sediments 8:177-183

Yogev A., Raviv M., Hadar Y., Cohen R. & Katan J. (2006) Plant waste-based composts suppressive to diseases caused by pathogenic Fusarium oxysporum. European Journal of Plant Pathology 116:267-278

Yogev A., Laor Y., Katan J., Hadar Y., Cohen R., Medina S. & Raviv M. (2011) Does organic farming increase soil suppression against Fusarium wilt of melon? Organic Agriculture 1:203-216

  60 

Attachment I  

Final evaluation pathogenicity test.

1st Row: The combination of the isolate with the cultivar

2nd Row: Picture of the plant four weeks post inoculation

3rd Row: Longitudinal section of the plant in row 2