Assessment of Coffee Berrydisease

ASSESSMENT OF COFFEE BERRY DISEASE, CHARACTERIZATION OF Colletotrichum kahawae ISOLATES AND EVALUATION OF RESISTANCE IN Coffea arabica COLLECTIONS FROM GIDAME, WESTERN ETHIOPIA

MSc. THESIS

ZENEBE WUBSHET HORDOFA

JANUARY 2019 JIMMA, ETHIOPIA

Assessment of Coffee Berry Disease, Characterization of Colletotrichum kahawae Isolates and Evaluation of Resistance in Coffea arabica Collections from Gidame, Western Ethiopia

By:

Zenebe Wubshet Hordofa

A Thesis

Submitted to the Department of Horticulture and Plant Sciences, School of Post Graduate Studies, College of Agriculture and Veterinary Medicine, Jimma University, for the Partial fulfillment of Degree of Master of Science in Agriculture (Plant Pathology)

Advisors:

1. Daniel Teshome Lopisso (Ph. D) 2. Weyessa Garedew Terefe (Ph. D)

January 2019 Jimma, Ethiopia

III

DEDICATION

I dedicate this thesis manuscript to my father Wubshet Hordofa.

STATEMENT OF THE AUTHOR

The thesis has been submitted for partial fulfillment of the requirement for MSc. Degree in plant pathology at Jimma University. The content of the thesis is the result of my original work except due references found in the text, no material is found which is previously published or written by another person. All sources of materials obtained from other sources have been suitably acknowledged in this thesis. I accept that the copy of my thesis must be lodge with the University Library and subject to the policy and procedure of the Jimma University. The thesis is made available as sources of research and study with the rule of copyright unless period of prohibition has been approved by the Dean of Graduate School. Briefs from the thesis are allowable without special permission provided the accurate manuscript in whole or part may be granted by the Dean of Graduate studies or by the Head of the Department of Plant Science of the Jimma University when in his judgment the proposed use of material is in the interest of the student. In all other occasions and permissions must be obtained from the author

______

Name Signature Date

BIBLOGRAPHY

The author Zenebe Wubshet Hordofa was born on May 25, 1990 from his father Wubshet Hordofa and his Mother Zewdnesh Alemu in Amhara Regional State, North Shoa zone, in Deneba Woreda. He completed his elementary, secondary and preparatory school at Anchikorer, Deneba and Debire Birhan schools respectively. He joined Debire Birhan University in 2011 and graduated in BSc. Degree in plant science on June 2013.Then, he was employed by the Ethiopian Institute of Agricultural Research Institute (EIAR) in May 2014 as junior crop protection researcher at Jimma Agricultural Research Center (JARC) and served for 2 years until he joined Jimma University for his MSc study in 2016.

ACKNOWLEDGEMENTS

First of all, I would like to thank almighty God with his mother for all things made possible through my life. My gratitude goes to Jimma University for accepting me to the MSc. program and Horticulture and Plant Science Department for providing me adequate knowledge in the basics of plant pathology which had been my long time desire. My deepest appreciation and gratitude also pies to my major advisor Dr. Daniel Teshome who accepted me as his advisehappily and supervised my work very closely with very critical comments, patience, positive and constructive remarks, I am very thankful for his great and unlimited support, I am also deeply indebted to Dr. Weyessa Garedew to welcome me as co-advisor and his critical support and encouragement throughout my study.

I would like to thank both of my advisors for their assistance, guidance and comments since the inception of this research project until the final thesis write up, supported me to acquire skill and knowledge. Thank you both of you for giving me the opportunity to learn from you and work with you.

I acknowledge Ethiopian Institute of Agricultural Research for giving me this chance and for the financial support. My thanks also qoes to Jimma Agricultural Research Center staff members especially the plant pathology laboratory staff for their unnosoued help during laboratory works. My special thanks also go to Mr. Dereje Amare, Wondmagegn G/Tsadk and Negasa Dechassa for their strong collaboration and encouragement throughout the whole thesis work.

Finally, I can’t leave without acknowledging the assistance of my parents who lifted me to go up in my enlightenment and development. Finally, the efforts of all other individuals who weredirectly or indirectly involved in my thesis work are really acknowledged.

ACRONYMS AND ABBRIVATIONS

ABT Attached Berry Test ANOVA Analysis of Variance CPS Coffee Production System CRD Complete Random Design CSA Central Statistical Agency DAI Days After Inoculation DBT Detached Berry Test DI Disease Incidence DMRT Duncan’s Multiple Range Test DPI Days Post Inoculation ECX Ethiopian Commodity Exchange FAO Food and Agricultural Organization GARSc Gera Agricultural Research Sub Center GPS Geographic Positioning System HAI Hours After Inoculation ICC International Coffee Council ICO International Coffee Organization ITC International Trade Center JARC Jimma Agricultural Research Center PDA Potato Dextrose Agar PSI Percent Severity Index RGB Red Green Blue SAS Statistical Analysis System SDW Sterilized Distilled Water USDA United State Department of Agriculture

TABLE OF CONTENTS Pages DEDICATION ...... I STATEMENT OF THE AUTHOR...... II BIBLOGRAPHY ...... IV ACKNOWLEDGEMENTS...... V ACRONYMS AND ABBRIVATIONS ...... VI TABLE OF CONTENTS ...... VII LIST OF TABLES ...... IX LIST OF FIGURES ...... X LIST OF TABLES IN THE APPENDIX ...... XI ABSTRACT ...... XII 1. INTRODUCTION ...... 1 2. LITERATURE REVIEW ...... 4 2.1. Importance of coffee in Ethiopian economy ...... 4 2.2. Productivity and production systems of coffee in Ethiopia ...... 4 2.3. Major constraints of coffee production in Ethiopia ...... 6 2.3.1. Occurrence and distribution of CBD...... 6 2.3.2. Taxonomic classification of C. kahawae ...... 8 2.3.3. Epidemiology and Disease symptoms ...... 8 2.3.4. Biology of C. kahawae ...... 10 2.3.5. Cultural and morphological characteristics ...... 11 2.3.6. Variation in virulence of C. kahawae isolate...... 12 2.3.7. Selection for CBD resistance in C. arabica ...... 13 2.3.8. CBD management options ...... 15 3. MATERIALS AND METHODS ...... 20 3.1. Description of the study area ...... 20 3.2. Assessment of CBD in farmers’ field ...... 21 3.3. Sample collection and fungal isolation ...... 22 3.4. Isolation and identification of C. kahawae isolates...... 23 3.5. Morphological characterization of C. kahawae isolates ...... 24 3.5.1. Macroscopic characterization ...... 24 3.5.2. Microscopic characterization ...... 24

VII

3.6. Determination of virulence in C. kahawae isolates ...... 25 3.6.1. Inoculum preparation and inoculation ...... 25 3.6.2. Data collection ...... 26 3.7. Evaluation of Gidame coffee accessions for resistance against CBD under field conditions, attached berry test ...... 27 3.8. Evaluation of Gidame coffee accessions for resistance against C. kahawae under laboratory conditions, detached berry test...... 31 3.9. Statistical analysis ...... 31 4. RESULTS AND DISCUSSION ...... 33 4.1. Assessment of CBD in farmers’ field ...... 33 4.1.1. CBD incidence and severity ...... 33 4.1.2. Relationship between CBD intensity and altitude ...... 34 4.1.3. Relation between coffee production system and disease intensity ...... 37 4.2. Macro and microscopic characterization of C. kahawae isolates ...... 38 4.2.1. Macroscopic characterization ...... 38 4.2.2. Microscopic characterization ...... 43 4.3. Determination of virulence in C. kahawae isolates ...... 44 4.3.1. Relationship between morphological characteristics and pathogenicity of C. kahawae isolates ...... 46 4.4. Evaluation of Gidame coffee accessions for resistance to CBD, attached berry test ...... 47 4.5. Evaluation of Gidame coffee accessions for resistance to C. kahawae under laboratory conditions, detached berry test...... 50 5. SUMMARY AND CONCLUSION ...... 54 6. REFERENCES ...... 56 7. APPENDICES ...... 69

VIII

LIST OF TABLES Contents Pages

Table 1. Descriptions of coffee farm study areas (Kebeles) in Gidame district ...... 20 Table 2. Assessment key for evaluation of coffee berry disease severity in Coffea arabica ...... 22 Table 3. List of Coffea arabica accessions/varieties used in the field trial...... 29 Table 4. List of Coffea arabica accessions/varieties used in the Attached Berry Test...... 31 Table 5. Disease incidence and percent severity index of coffee berry disease at different kebeles of Gidame district ...... 34 Table 6. Pearson correlation analysis between major factors and intensity (incidence and severity) of coffee berry disease in Gidame district ...... 35 Table 7. Macroscopic characteristics of Colletotrichum kahawae isolates collected from Gidame district ...... 40 Table 8. Microscopic characteristics of Colletotrichum kahawae isolates collected from Gidame district ...... 44 Table 9. Virulence of Colletotrichum kahawae isolates on susceptible C. arabica variety 370 as determined by detached berry test ...... 45 Table 10. Relationship between morphological characteristics and pathogenicity/ aggressiveness of Colletotrichum kahawae isolates ...... 47 Table 11. Percent infection of Coffea arabica accessions/varieties inoculated with Colletotrichum kahawae in attached berry test ...... 48 Table 12. Response of Coffee arabica accessions/varieties to Colletotrichum kahawae infection in detached berry test ...... 51

LIST OF FIGURES

Contents Pages

Figure 1. The systemic infection process of Colletotrichum kahawae ...... 10 Figure 2. General life cycle of Colletotrichum species ...... 11 Figure 3. Map of the study area ...... 20 Figure 4. Typical symptoms (black sunken lesions) on coffee berry disease of Coffeaarabica berries...... 23 Figure 5. Detached green Coffea arabica berries inoculated with Colletotrichum kahawaeisolates ...... 26 Figure 6. Relationship between severity (%) of coffee berry disease and across altitude range ...... 35 Figure 7. Disease incidence (DI) and percent severity index (PSI) of Coffee berry ...... 36 Figure 8. Intensity of coffee berry disease among production systems across Gidame district...... 37 Figure 9 Cultural morphology of representative ten days old culture of Colletotrichum kahawae isolates ...... 42 Figure 10. Conidial features of Colletotrichum kahawae isolates ...... 43 Figure 11. Coffea arabica accessions response for Colletotrichum kahawae ...... 53

LIST OF TABLES IN THE APPENDIX

Pages

Appendix Table 1. List of C. arabica accessions with mean CBD score under visual assessment at field conditions ...... 70 Appendix Table 2. ANOVA table for coffee berry disease incidence in nine kebeles of Gidame district ...... 71 Appendix Table 3. ANOVA table for coffee berry disease incidence in nine kebeles of Gidame district ...... 71 Appendix Table 4. ANOVA table for cultural growth rates of Colletotrichum Kahawae collected across different kebeles of Gidame district ...... 72 Appendix Table 5. ANOVA table for conidial size of Colletotrichum kahawae isolates (spore length)...... 72 Appendix Table 6. ANOVA for conidial size of Colletotrichum kahawae isolates (width) 72 Appendix Table 7. ANOVA table of sporulation capacity of Colletotrichum kahawae isolates ...... 73 Appendix Table 8. ANOVA for isolates on virulence determination in detached berry test ...... 73 Appendix Table 9. ANOVA table for resistance evaluation of Coffea arabica accessions against coffee berry disease with attached berry test ...... 73 Appendix Table 10. ANOVA table for resistance evaluation of Coffea arabica accessions against coffee berry disease under laboratory conditions detached berry test ...... 73

ASSESSMENT OF COFFEE BERRY DISEASE, CHARACTERIZATION OF Colletotrichum kahawae ISOLATES AND EVALUATION OF RESISTANCE IN Coffea arabica COLLECTIONS FROM GIDAME, WESTERN ETHIOPIA.

ABSTRACT Ethiopia is the center of origin and diversity for Arabica coffee which acts as an engine for the country’s economy. Coffee berry disease (CBD) caused by the fungus Colletotrichum kahawae is one of the major constraints for coffee production in the country. However, little information is available regarding the extent of the CBD and related factors in Gidame district. The objectives of this study were to assess the magnitude of CBD in coffee fields at Gidame district, to characterize and study the pathogenicity of C. kahawae isolates and further evaluate disease resistance in local C. arabica collections. A survey to assess the distribution and magnitude of CBD was conducted in nine selected kebeles during July to August 2017. Disease assessment and collection of CBD infected coffee berry samples were collected in a total of 45 farms. Furthermore, evaluation of CBD resistance was conducted under field (Attached berry test) and laboratory (detached berry test)conditions. The survey results showed the prevalence of CBD in all assessed areas with varied intensity that ranged from 66 - 86% and 16 to 50% of disease incidence and percent severity index, respectively. More importantly, increased CBD intensity was observed in higher altitudes resulting in a significant positive correlation with disease incidence (r=0.61) and severity (r=0.55). Furthermore, macroscopic and microscopic investigations revealed that C. kahawae isolates in the study area are highly diverse in terms of cultural and conidial characteristics (colony color and density, mycelial growth rate and conidial production). Mycelial growth rate differs significantly (p<0.001) with the range between 2.2 to 4.3 mm/24hrs. Similarly, sporulation capacity varied between 186.1- 572.3 spores/ml. Those observations were in strong agreement with the results of the virulence test that revealed significant variation (p<0.001) among C. kahawae isolates and infection percentage ranged between 34.8 and 88.7%. It was also observed that Gidame coffee accessions significantly differ (P<0.001) in resistance to CBD at field and laboratory conditions. Accessions G65, G15, G63, G66 and G72 showed the lowest CBD severity score in field (<10%) hence relatively resistant to CBD. On the other hand, coffee accessions G50, G89, G92 and G67 were found to be the most susceptible with >25% of berry infection. Overall, the present study not only showed that CBD is one of the most important diseases of coffee in the study area but also demonstrates the variation of virulence among C. kahawae isolates and the role of host resistance in combating CBD. Future research should focus on evaluating the promising coffee accessions in mutilocation field trials across several years. Besides, the diversity/identity of C. kahawae isolates should be verified using more reliable methods and further studies to understand mechanism of CBD resistance should be the priority research topic to fully understand the C. arabica - C. kahawae pathosystem.

Key Words: Incidence, Severity, Prevalence, Pathogenicity Test, Disease Resistance

XII 1. INTRODUCTION

For millions of communities in Africa, Asia and Latin America, coffee is not only the major sources of hard currency but also a means of livelihood and has a significant economic, social and spiritual impact on many communities with diverse cultural and/or psychological backgrounds (Labouisse et al., 2008; ITC, 2011; Chauhan et al., 2015). Coffee is the world’s second most traded commodity next to petroleum and serves as a direct source of income for the growers in different parts of the world (Kiwanian, 2013; FAO, 2015). In fact, this crop is the backbone for Ethiopian economy and contributes about 27% of the foreign exchange earnings and more than 25% of rural and urban employment (ICO, 2018).

Today, coffee is grown in more than 80 countries with more than 10 million hectares of land (Etana, 2018). Ethiopia is the primary center of origin and diversity of Arabica coffee and ranks 1st and 6th in Africa and the world respectively (ICO, 2018). The existence of ideal and diverse agroecologies together with forest, semi forest, garden, and plantations production systems in Ethiopia made coffee to be the leading agricultural produce in the country (Workafes and Kasahun, 2000). Coffee is largely cultivated in the Southern, South Western, Western and Eastern parts of Ethiopia (Emana, 2015) with 700,447 ha of total land coverage and 469, 091 tons of production per annum (CSA, 2017).

The Oromia Regional State is the homeland of C. arabica evidenced by the existence of wild coffee trees in the region with 464,426 ha (66%) of total area coverage and 317,316 tons of production per annum (CSA, 2017). Following to Jimma zone, West Wollega covers the highest coffee land (about 90,626 ha) with 67,074 tons of production andhas great contributionfor the national export market from the region (Mehari et al., 2016; CSA, 2017). Due to the presence of the highest heterogeneous germplasm, Gidame in the Western part of Oromia has been given the highest priority for improving local landrace development program that encourage coffee production. According to Gidame Agricultural Office report of 2017/2018, the district is the place where the highest germplasm with the largest coffee farm coverage (around 47,500 ha) exists. Despite the high production potential and economic importance of coffee in Gidame, its production has been affected by several biotic and abiotic constraints.Colletotrichumspecies that cause anthracnose symptoms have

1 been reported in several African countries (Rakotonirianaet al., 2013; Cristobel et al., 2017). Coffee berry disease (Colletotrichum kahawae), coffee wilt disease (Gibberella xylarioides) and coffee leaf rust (Hemileia vastatrix) attacking fruits, leaves, stems and roots, respectively have been reported as serious challenges of coffee production in tropical and subtropical areas (Derso and Waller, 2003; Adugna et al., 2009a).

C. kahawae (Waller and Bridge, 1993) is an aggressive and specialized fungal pathogen which becomes major concern in Africa due to the great damage on the economic parts of green coffee berries (Van der Vossen et al., 2015; Alemu et al., 2016). The first record of this disease in Africa was from Kenya in 1922 which later distributed to Angola, Cameroon, Malawi, Ethiopia, Tanzania and Uganda in short time (Hindorf and Omondi, 2011; Weir et al., 2012; Alemu, 2013). The losses due to CBD on individual farms vary considerably. Under conditions where no control measures are undertaken the losses can be reaching up to 100% in high rainfall and high altitude areas (Van der Graff, 1981) While, using resistant varieties play a significant role in combating the disease (Adugna et al., 2008).

The significance of CBD in C. arabica growing regions of Ethiopia has been reported in different research reports in the different parts of the country. In Oromia and Southern Nation Nationality and People (SNNP) has reported 38.8 and 17.2% mean incidence respectively (Zeru et al., 2012). Recently, Alemu et al. (2016) reported that all coffee producing ecosystems of Ethiopia are suffered with CBD more importantly in Borena (10- 80%), Gedio and Hararghe (40-100%), Illubabor (10-90%) Jimma and Sidama (30-90%). In Western Wollega, Jirata and Asefa (2000) and quiet recently Alemu et al. (2016) have reported a disease incidence of 22% and 30-80% respectively, indicating the very significance of the disease which may lead to a complete yield losses whenever susceptible landraces are cultivated (Belachew and Teferi, 2015). The same Author, Alemu et al. (2016), have reported a national average CBD incidence and severity of 52.5% and 29.9%, respectively. Regarding Gidame however, the extent of CBD intensity within different ecologies of Gidame district and the aggressiveness nature of the pathogen isolates from the area is not known specifically indicating, the need to assess the distributions and extent of the disease in small scale production conditions of this area.

Agro ecological based local landraces development is very important to utilize the available genetic resources found in different coffee growing areas (Van der Vossen, 2015;

2 Emana, 2015). It is widely accepted that the use of resistant varieties provides the opportunity of controlling diseases in sustainable and environmentally compatible manner. Promising sources of disease resistance in coffee germplasm have been developed in different countries like Ethiopia, Kenya and Tanzania (Mtenga et al., 2012). Rather than variety adoption problem, the challenges of long breeding cycle associated with the long growth period of coffee have slowed down the progress of further varietal improvement works (Zeru, 2006).

Jimma Agricultural Research Center (JARC) has been playing a crucial role by developing and releasing improved coffee varieties adapted to different agroecologies (Belachew, 2001). With the focus of local landrace development to alleviate CBD problem, the center has developed 31 CBD resistant varieties for different coffee producing regions of Ethiopia (Benti, 2017; Teferi and Belachew, 2018). From those, only four varieties have been developed for Western coffee growing areas since 2010. However, to achieve the apparent economic development program of the producers as a whole and the region in particular, such a very small number of improved varieties are not sufficient as compared to the diverse agroechological niches, enormous available coffee genetic resources and the high coffee production potential of the zone.

In general, it is important to increase the genetic base of improved coffee varieties preferred by farmers. Assessing and developing varieties which bring tangible change on farmer’s economy by increasing productivity through effective control of CBD while minimizing production costs and reducing potential consequences on human health and the environment is crucial. One of the primary steps to achieve this goal is to assess the extent of CBD and further understand the potentials of local plant materials in combating this disease. With this background, the present study was conducted with the following objectives:

 To assess the status of CBD in coffee fields of Gidame district in Western Wollega, Ethiopia

 To isolate, characterize and study the pathogenicity of C. kahawae isolates

collected from diseased plants farms of Gidame district

 To evaluate local C. arabica germplasm collections from Gidame for resistance

toC. kahawae

3 2. LITERATURE REVIEW

2.1. Importance of coffee in Ethiopian economy

Commercial production of coffee depends on two major species, Coffea arabica and Coffea canephora Pierre which account for 65% and 35% of production respectively (Chauhan et al., 2015). Coffee production by exporting countries has been on a gradual rise. The coffee market is dominated by the three top producing countries Brazil, Vietnam and Colombia with 30%, 15% and 12% respectively (USDA, 2018). Brazil is the largest Arabica coffee producer in the world, while Vietnam, which is a relatively new candidate and now the world’s largest Robusta producer (ICO, 2016).

Apart from being the birth place of C. arabica, Ethiopia is also a major producer and consumers of high value coffee from Africa and ranks 6th in the world (ICO, 2018). Coffee industry is the back bone of the country economy in the agricultural sectors and has contributed vital role in the national economy (Grabs, 2017). Coffee in Ethiopia accounts for 40% of total export and 10% of total government revenue (Temesgen and Getachew, 2016).

According to CSA (2017) report, coffee has been the most essential Ethiopian's green gold which is being as largest export commodity and shares 4.94% of the area under all crops in the country. It has significant impact on the socio-economic life of the people and economic development of the country. Coffee is source of revenue for more than 15 million people (16% population) whom directly or indirectly engaged in the production, processing and trading (CSA, 2017). It contributes as immeasurable amount of Ethiopian export earnings and plays an important role in social gathering and local consumptions, more than half of the Ethiopian’s coffee produce is consumed locally (Tadesse et al., 2014).

2.2. Productivity and production systems of coffee in Ethiopia

C. arabica is Ethiopian’s indigenous crop which serves as the major source of hard currency and largely cultivated in the Southern, South Western and Eastern parts of the country (Amano, 2014). Having a wider agro-ecological zones and genetic diversity in Ethiopia, offers the greatest opportunity to produce superior C. arabica which allowed Ethiopia to be competitive in the world market. About 700,447 ha of the land have

4 cropped with coffee (CSA, 2017; Temesgen and Tufa, 2017). Kaffa, Illubabor, Jimma, Wollega, Sidamo, Gedeo, Yirgachefe and Hararghe are the chief major coffee producing areas in the country (Chauhan et al., 2015; Garuma et al., 2015).

There is a variation in coffee production system in Ethiopia. Mainly four (forest, semi- forest, garden and plantation coffee) production systems are found in the country (Workafes and Kasahun, 2000) and the contribution of each production system is 10%, 35%, 50% and 8% of the total coffee production respectively.

Ethiopia is endowed with enormous genetic diversity and different coffee types with unique taste and flavor. The country in general and the study area in particular has favorable agroecological and socio-cultural conditions for coffee production. Wollega coffee is produced in highly diversified garden and semi-forest production systems adapted to different ecological conditions (Gurmessa et al., 2012). In the area, coffee is grown as main cash crop for the producers starting from immemorial time (ECX, 2008). The presence of the highest indigenous coffee germplasm which is promising for the improved local landraces development in the Western Wollega (Gidame localities) is given to the highest priority by the researchers. Eventhough, these germplasms are found, most of the farmers have grown local landraces having poor performance with garden and semi- forest type of production system in the areas. Likewise, their management practices are still traditional which result in low yield compared to the world standard. For instance, most of the farmers use slashing to remove weeds and shrubs from their coffee farms and filling the open spaces with local seedlings (Habte, 2005; Zeru, 2006). Hence, there must be considerable attention to develop improved local landraces from the available germplasm for the area.

Belachew (1997) reported that, indigenous coffee cultivars are location specific and management practices also vary slightly from region to region. Yilma and Kufa (2002) also reported that 70 to 80% of Wollega coffee producers depend on the production of the local coffee types. This leads variation in the intensity of the major coffee diseases (coffee berry disease, coffee wilt disease and coffee leaf rust) from area to area and even among genotypes in the country. With this concern, Jimma Agricultural Research Center has released only four coffee cultivars for Wollega zones since 2010. These coffee cultivars have several attributes like disease resistance, high yielding performance, good vigor, good quality and attractive bean size (Gidisa, 2016; Benti, 2017).

2.3. Major constraints of coffee production in Ethiopia

Even if coffee is one of the leading export commodity for Ethiopian economy, its average yield is very low (about 748 kg/ha) in contrast to the achievement in Latin America and Asia. This is due to different factors like limited use of improved technologies and best cultural practices by producers, the occurrence of insect pests, diseases and coffee weeds (Adugna et al, 2009b; Teferi and Ayano, 2017). Coffee producers mainly challenged by coffee berry disease (Colletotrichum kahawae), Coffee wilt disease (Gibberella xylarioides) and coffee leaf rust (Hemileia vastatrix) which leads to higher losses in coffee yield by attacking the green berries, vascular tissues (Xylem and phloem) and photosynthetic part (the leaf) respectively. These are the major challenges of coffee production in the coffee ecology in Africa especially in Ethiopia starting from the earlier time (Belachew et al., 2016; Teferi and Belachew, 2018).

2.3.1. Occurrence and distribution of CBD

In African continent, the main cause of yield loss in Arabica coffee is due to coffee berry disease (CBD) caused by C. kahawae (Pires et al., 2015). Due to its economic damage on expanding green coffee berries that induce premature fruit drop and/or fruit mummification, it has considered to be coffee anthracnose which result in the great damage (Mohammed and Jambo, 2015; Emana, 2015).

The disease was first noticed and identified by McDonald in Kenya 1922 and up to 75% losses were reported soon after first appearance. Despite the little attention received at the early stage of it emergence, African coffee growers soon observed a rapid dissemination of CBD throughout important Arabica coffee growing areas. It was reported in Angola in 1930, Zaire in 1937, Cameroon 1955-1957, Uganda 1959, Tanzania in 1964 and Zimbabwe, Malawi and Zambia 1985 (Weir et al., 2012; Belachew and Teferi, 2015). The disease was occurred and reported in Ethiopia much later than Kenya (after five decades). It was first appeared in Sidamo then, spread very quickly in all coffee growing areas until 1978 and caused remarkable losses (Hindorf and Omondi, 2011). Its impact is more important at high altitude plantations above 1500 m in tropics (Fino, 2014). Occasionally, it will have occurred in the lower altitude having cold and damp weather conditions similar to climatic conditions in the higher altitudes (Zeru et al., 2009).

Different research outputs pointed out that, the impact of CBD is varied from place to place and season to season (Belachew and Teferi, 2015). Ejetta (1986) reported 51 and 81% of yield losses due to CBD at Melko and Wondogent respectively. According to the assessment conducted in 1994 in Oromia and Southern Nation Nationality and Peoples (SNNP) of the regions, the result indicated 38.8 and 17.2% mean percent prevalent of the disease respectively (Zeru et al., 2012). In addition, the survey was repeated in 1997 and 1998 in six major coffee growing zones of Oromia region and showed an average of 31 and 32 % of CBD severity in the respective years (Jirata and Asefa, 2000), in 10 zones and 31 woredas of the SNNP reported 40 and 22.8% of incidence and severity, respectively (Negash and Abate, 2000).

The impact of CBD is still more prevalent in the African continent but has not given critical focus as the major problem elsewhere. Since its emergence, has been associated with sever yield losses and could be a factor preventing cultivation of Arabica coffee in the highland regions. In African countries that grown C. arabica, crop losses of 20-30% are common but can exceed 80% in extremely wet years, if no control measures are applied (Silva, 2010).

Depending on host susceptibility, pathogen aggressiveness and favorable weather conditions, the occurrence and intensity of CBD varies from place to place and season to season (Alemu et al., 2016). The disease is very severe and causes appreciable yield loss in the area received low relative temperature and high relative humidity mainly in the rainy seasons. Prevalence of low temperature with high rainfall for extended period of time favors CBD development and increases disease occurrence that can results to rots and fruit falling with post-harvest losses ranging from 40 to 80% (Hindorf and Omondi, 2011). Waller et al. (2007) reported more than 80% losses from total production due to CBD in Cameroon.

CBD has distributed to many parts of Ethiopia within short time. This is due to the free movement of coffee planting materials from CBD infected areas that facilitate frequent distribution of the disease (Zeru, 2006). Alemu et al. (2016) reported that all coffee producing ecosystems of Ethiopia are suffered with the occurrence of CBD more importantly in Borena (10-80%), Gedio and Hararghe (40-100%), Illubabor (10-90%) Jimma and Sidama (30-90%) and West Wollega (30-80%) of CBD incidence in each assessed zones during 2014 cropping seasons. Derso (1997) reported that, CBD results 24

7 to 30% national average losses on the susceptible landraces. When environmental conditions favored disease development, the loss due to CBD may reach 100% (Adugna et al., 2009a).

2.3.2. Taxonomic classification of C. kahawae

The former taxonomic description and position ofC. kahawae was subject to confusion from the range of Colletotrichum spp. which are isolated from coffee plants, four groups (C. coffeanum mycelia, C. coffeanum acervuli. coffeanum pink and coffee berry disease (CBD) strains) are described initially based on their morphological traits (Gibbs, 1969). Of which, the fourth group was the only species which able to infected the green berries. C. gloesporioides and C. acutatum are the former groups which are non-pathogenic for green coffee berries (Hindorf, 1970). However, C. coffeanum was described in 1901 in Brazil from diseased coffee berries (Hyd et al., 2009). Gichuru (2007) and Prihastuti et al. (2009) found that, C. gloesporioides occurred as saprophyte (weak pathogen) synonym with CBD pathogen which more aggressive on ripe coffee berries and coffee tissues with the symptoms of whitish to dark grey discoloration. In 1993 Waller and Bridge described C. kahawae as the causal agent of CBD and as a distinct species based on morphological, cultural and biochemical character (Batista et al., 2017). The current taxonomy of the fungus is under Kingdom: Fungi, Phylum: Ascomycota, Class: coelomocytes,Order: Melanconiales, Family: melanconiaceous, Genus: Colletotrichum, Species: Colletotrichu kahawae (Owaka, 2011).

2.3.3. Epidemiology and Disease symptoms

Plant infecting fungi comprises diverse groups of organisms with different life and methods by which they infect and colonize their hosts (Silva et al., 2017). The only current host for C. kahawae is coffee (Batista et al., 2017). The pathogen is able to colonize and infect coffee trees at different stages (Berries, flowers, leaves and maturing bark of the branches). Host susceptibility, climatic conditions and pathogen virulence are the key factors that determine the occurrence and intensity of CBD from time to time and place to place (Gichimu et al., 2014).

The development of CBD pathogen (C. kahawae) is heavily dependent on rain /water for conidial production, dispersion, germination and infection. The asexual spore (conidia) from the mummified berries and twig barks which disseminated by rain splash are the primary inoculum sources (Griffith et al., 1991; Waller and Bridge, 1993). Movement of

8 the spore is down ward in tree canopies that guided with the movement of film of water (moisture) while long distance movement of CBD inoculum has been occurred through passive vectors like man, insects, vehicles and birds which can carry viable spores or via the movement of infected plant materials (Biratu, 1995). This is the reason why coffee crowns CBD spores are laterally dispersed between trees and branches by wind slashing yet localized; downward movement is the primary inoculum movement (Silva et al., 2006).

In the infection process, C. kahawae uses hemibiotrophic strategy that includes post penetrative asymptomatic bio trophic phase (pathogen attacks the host cell without killing them) followed by destructive necrotrophic phases (involves the increased activities of wall degrading enzyme to function in C. kahawae pathogenicity) which culminate the appearance of disease symptoms and the reproduction of the fungus (Loureiro et al., 2012). Tissue colonization is associated with sever cell wall alterations and death of the host protoplast (MouneBedimo et al, 2010). Its infection starts with conidial germination (asexual spore) which can be happening 24 hrs. after contact with the host plant tissue and follows elongation of the germ tube in the apical selection is differentiates in to melanized apresorium which functions to penetrate the plant cell cuticle (Fig 1) directly via turgor pressure (mechanical pressure), secretion of cutin degrading enzymes or the combination of the two processes (Silva et al., 2006; MouneBedimo et al., 2010).

The major and the most economically important infection of C. kahawae and its damages are on the expanding green mature berries (Gichuru, 2007; Gidisa, 2016). The first disease symptom is dark brown slightly sunken spots under suitable environmental conditions. The spots enlarge to cover the whole berries and masses of conidia may be visible and will form black and shriveled lesions on the beans, whereas, infection on ripe berries will be seen as dark sunken patches that spread rapidly and may cover whole berries. Under favorable conditions, the pathogen rapidly sporulates by forming mass of conidia and enters inside the berries via penetrating and infection takes place in expanding green berries, leading to premature dropping and mummification of berries (Waller et al., 2007).

Based on resistant or susceptible response of coffee genotypes, two types of symptoms (scab lesions and active lesions) can be occurred. Scab lesions are plants resistance response that restricts fungal development which allows only the formation of small black spots. As well, the deeper layers of the fruit are not invaded and the lesion appears

9 stationary not affecting the normal development of the green berry. While, in susceptible plants active lesions development is observed starting as little black spots, in the presence of good conditions can form dark, sunken, active lesions that rapidly expand and destroy the entire fruit (Fino, 2014; Diniz, 2018).

Figure 1.The systemic infection process of Colletotrichum kahawae (Silva,2010)

2.3.4. Biology of C. kahawae

C. kahawae can infect all stages of the crop from flowers to the ripe fruits and occasionally leaves but maximum crop losses occur following infection of green berries with the formation of dark sunken lesions with sporulation (Gichimu et al., 2014; Kamau, 2015). The perfect state for some species of Colletotrichum happening on coffee has been proved to be the Ascomycetes Glomerella cingulata which is generally polyphagous. In 1901 Noack detected for the first time C. acutatum in Brazil causing leaf spots and dieback (branches) of C. arabica (Loureiro et al., 2011). Unparalleled adaptation of infecting green coffee berries enables the CBD pathogen to occupy a unique ecological niche which separate it on a functional level from all other Colletotrichum species.

Understanding the complex life style patterns of Colletotrichum spp. (Fig 2) and the dynamic state of their interaction with their host has important implication for management. The biology and slow growth form of CBD pathogen can be distinguished metabolically by its inability to use either citrate or titrate as a sole carbon source (Loureiro et al., 2012). The name C. kahawae has a strong practical meaning and an

10 indissoluble connection with the coffee disease. Bringing a volatile group of fungi able to infect a wide range of organs and hosts worldwide under the same specific epithet (which means coffee) is likely to disrupt the accuracy required in the nomenclature for pathogen identification and cause a great deal of confusion, even if distinction at the sub specific level is provided (Batista et al., 2017). Pathogen diversity can be limited by lack of perfect state which promotes sexual recombination (Omondi, 1998).

Figure 2. General life cycle of Colletotrichum species (Silva et al., 2017)

2.3.5. Cultural and morphological characteristics

Cultural characteristics of C. kahawae were first described by McDonald in 1926.The main distinguishing characteristics of the pathogen include slow growing, cottony and dark grey to greenish mycelium mainly in the young culture. These features are become variable in old cultures (Waller et al., 1993; Omondi, 1998). They are common cultural characteristics that distinct it from other forms of Colletotrichum species associated with ripening coffee berries, branches and leaf symptoms. In 1969 Gibbs described four main strains of Colletotrichum species collected from coffee with the base of their colony characters and able to separate between the saprophytic strains (mycelia with and without acervuli and pinkish form) in such the conidia of C. kahawae borne directly on the hyphae

11 (Gibbs, 1969). The detailed morphological study of Colletotrichum spp. collected from different ecosystem was made by Hindorf in 1970 in Kenya.

Variation among C. kahawae isolates can be existed due to aggressiveness and cultural characteristics like sporulation, conidial size and shape, appearance on culture, growth rate of the isolates on different media (Rodrigues et al., 1991). Omondi (1998) reported that, these morphological descriptive criteria are used as accurate classification of Colletotrichum spp. that invading coffee. He also reported 2-4 mm/day of average growth rate while, Zeru (2006) reported 4.4 mm/d of average growth rate, 14.1 and 4.2µm of the conidial length and width respectively when it cultured in different media. The conidia of C. kahawae can be distributed by water splashes because it requires the presence of water or 100% relative humidity and optimum temperature of about 22 0C for germination.

2.3.6. Variation in virulence of C. kahawae isolate.

CBD has distinctive features which enables it to occupy a unique ecological niche and separates it on a functional basis from all other Colletotrichum species. One is its pathogenicity towards developing green coffee berries (attached and detached berries) and seedling hypocotyls. In places where C. arabica is grown most extensively in the highlands, the pathogen C. kahawae did reach the wild population in which much of it was susceptible to the disease (Batista et al., 2017).

Hindorf et al. (1997) stated that for correct identification and characterization of disease causing fungal pathogens there must be a need to detect the morphological features such as mycelial color, growth rate, conidial production in the invitro conditions and testing the pathogenicity on the berries or hypocotyls.

Coffee genotypes were tested using local C. kahawae strains in Kenya but aggressiveness difference was not observed (Van der Vossen et al.,1976). Furthermore, differential interactions between host and pathogen population were never found in Ethiopia and it was improbable and were caused by gene-for–gene specificity (Van der Graff, 1981). Aggressiveness can be considered as quantitative measure of the level of disease reached over time. Most of the aggressive pathogen reached at the specific disease level faster than the less aggressive and can be measured by spore production, infection size, latent periods and its severity on the host (Luzolo et al., 2010; Pires et al., 2016).

Manga et al. (1998) reported that factors like geographical origins have attributed for aggressiveness of C. kahawae populations. The variation in both sporulation and rate of growth of the pathogen can lead to variation in aggressiveness of C. kahawae isolates (Kilambo, 2008). Beynon et al. (1995) reported that C. Kahawae strains collected from Cameroon, Zimbabwe, Kenya, Burundi, Tanzania and Ethiopia, the strains from Malawi were the most virulence and cause severe disease in comparison to others. According to Margaret (2011), variation within pathogen population can be appeared as a result of migration or gene flow followed by mutation or recombination. Derso and Waller (2003) found that C. kahawae isolates (collected from Yirgachefe, Gore and Gera garden coffee areas) were pathogenic to the hypocotyls of susceptible coffee cultivars and varied in their aggressiveness. Isolates collected from different geographical locations may showed similar reaction but varies in their aggressiveness on the inoculated genotypes (Fredrik et al., 2015). This indicates that the smallest variability of C. kahawae isolates can leads to genetic uniformity among isolates (Bridge et al., 2008)

According to Omondi et al. (2001) report isolates can be varying (differences in the virulence) in the reaction of pathogen populations and C. arabica genotypes. Such pathogenic variations have been determined using coffee detached green berries or seedling hypocotyls (Beynon et al., 1995). Virulence tests with isolates of C. kahawae also confirmed that the CBD pathogen exhibited variation for aggressiveness but no races were detected (Fredrick et al., 2015). Susceptible genotypes are clear indicators of aggressiveness difference among pathogen isolates and consequently difference in aggressiveness should always be performed on susceptible genotypes (Castiblanco et al., 2018). Those all findings have important implications for development of resistant cultivars and being base for breeding programs.

2.3.7. Selection for CBD resistance in C. arabica

CBD is known to be a devastating pathogen in Arabica coffee production in Africa. Control of this disease by frequent fungicide sprays is expensive (30 to 40% of the total production costs) and not always effective and usually beyond the means of smallholder coffee growers (Van der Vossen, 2001). Unlike other diseases, such as coffee leaf rust, breeding for CBD resistances of Arabica coffee has been more encouraging because it becomes the first threat of coffee production in different cultivation areas.

Rather than natural CBD infection in the field, several artificial inoculation methods have been developed for screening coffee plants for disease resistance. Inoculation on detached green berries and on seedling hypocotyls is the most common methods (Van der Vossen et al., 1976). Detached green berry inoculation test was used for resistance screening of C. arabica accessions against CBD under artificial conditions. This offers the advantage of direct assessment of C. arabica and C. kahawae interaction at fruit level and allowed easy way of resistant genotypic characterization under field conditions (Pinard et al., 2012).

Defense responses of the genotypes are considered to be dependent on the pathogen lifestyle and the genetic constituent of the host (Adolf and Ercolano, 2015; Ma and Ma, 2016).Pre-formed and induced mechanisms are among the most mechanisms in the host response for pathogen. In coffee resistance for CBD can be characterized by restricted fungal growth associated with host resistance like hypersensitive reaction (HR) which is the rapid localized cell death, formation of structural barriers and antifungal compounds like accumulation of callose, lignin-like and phenolic compounds (Silva et al., 2006; Loureiro et al., 2012). Some of these mechanisms are likely to be pathogen nonspecific and could also be induced by mechanical injury. It may be possible to develop some of the observed biochemical and structural changes into methods of screening for CBD resistance.

There is a wide publication on the availability of CBD resistance in some Ethiopian coffee genotypes. Biratu (1995) found that the presence of variations in aggressiveness and absence of races within C. kahawae isolates, existence of significant differences among coffee selections/isolates and their infection effects. More often, traits of interest in perennial crops such as disease resistance can be observed and screened more importantly at late stages of development and require assessment over a number of years at different location. Besides, the susceptibility of green berries for C. kahawae is varied with the development stages (Vieira et al., 2018).

In Ethiopia, CBD resistance selection program was designed soon after the outbreak of CBD in 1971. High genetic variability among coffee genotypes were observed through massive selection programs on indigenous resistance coffee populations basically with four critical steps, selection and testing of coffee mother trees without or with very low CBD infection via visual assessment and berry count in the natural infested fields and attached or detached berry tests, screening of their progeny and multiplication in large

14 blocks (Ejetta, 1986). Robinson (1976) found that resistant trees occurred at frequencies of 0.1% to 1% and one tree in every 10,000 possessed both resistance and high yield and quality. Genetic resistance appears to be partial in C. arabica and complete in C. canephora (Van der Vossen et al., 2015).

Breeding strategies for resistance selection provides sustainable and long term management of the disease in different coffee producing countries (Alemu et al., 2018) However, the application of the methods may vary from country to country depending on the genetic variability, ecological conditions and production problems (Belachew, 2001). As Ethiopia is the center of origin and diversity of Arabica coffee, initial emphasis of breeding was agroechological based CBD resistant varieties development by subsequent evaluation of the mother trees at their place of origin and the progenies at laboratory and experimental plots. In Ethiopia, pure line selection and intra specific hybridization methods were used commonly (Benti, 2017).

Efforts have been devoted during the past four decades towards genetic based resistant varieties development and identification. For first time, from 218 promising selections 13 CBD resistant varieties were identified and released within five years (1972-1978) interval in the country (Van der Graff, 1981). The release of these varieties have played significant role in the coffee industry in the country and the varieties are still under production at diverse agroecologies (Belachew, 2001; Benti, 2017). Since 1985, great efforts have been made with adopting similar selection scheme and testing procedures in search for more resistant cultivars within the heterogeneous populations in various coffee growing areas of the country, now about 31 pure lines of CBD resistant cultivars have been released and are in production in coffee growing areas of the country (Belachew et al., 2015; Benti, 2017).

2.3.8. CBD management options

In order to improve the income gained from coffee, there must be a need to manage coffee berry disease in all countries which produces it because the disease has become a serious threat today (Van der Vossen et al., 2015; Alemu et al., 2016). Some of the methods are:

Cultural practices: Cultural practices are likely to reduce losses due to CBD. There are some reports those that promote good aeration and rapid drying of the canopy, such as adequate pruning and wide spacing can reduce disease incidence. Alemu and Sokar (2000) reported that the use of irrigation in coffee farm areas in the areas can shifted the

15 susceptible stage of berry development which attributed to reduce CBD incidence (Gidisa, 2016). Pruning is an agricultural operation commonly practiced in tree crops. It is used to shape tree architecture, renew the assimilating system and stimulate new reproductive organs. It is also recommended for controlling numerous diseases. Pruning during the vegetative growth period can leads to the effective means of removing diseased branches, berries, susceptible and old trees. This leads reduce the initial inoculum sources through maintenance pruning and mummified berry removal (Holb, 2005; Bedimo et al., 2007).

As coffee trees are shade lovers’, proper shading of coffee trees considerably reduced CBD development. This is because shade reduces sunlight and plays an important role in modification of the micro environmental conditions (reduce temperature fluctuations, air movements and could limit the rain intensity) and can work as a barrier and limits the dispersal and development of the pathogen (Jha et al., 2014). Bedimo et al. (2008) reported 30 and 50% of CBD incidence with and without shade in the coffee plantation respectively, which indicates the significance of the shade.

Host plant resistance: Variations in host susceptibility for C. kahawae are identified. As well, there is a strong agreement that host resistance (HR) is the most preferred, appropriate, cost effective, biologically save and sustainable way of controlling diseases and pests in crop ecology (Margaret, 2011). Growing resistance varieties need strong work in identifying the sources of genetic resistance within coffee tree collections established in different producing country.

Resistant variety development in Ethiopia started after the outbreak of CBD in 1971 and is still in use as the best disease management option with little progress this is due weak linkage between the research and extension system and limited knowledge of the technology. Founding of stable resistance (durable resistance) could guarantee stable production for the growers. Availability of varieties with durable resistance to CBD could provide economically and environmentally attractive and alternative for costly fungicide spraying (Van der Vossen and Walyaro. 2009). It also avoids residual effects of fungicides to the sprayers and saves the time spent for spraying. In particular, it is a useful resource for poor smallholder farmers who are the main coffee producers in Ethiopia. Development of resistant varieties increases the productivity and production of coffee in the country. It also saves the nation a considerable amount of foreign exchange that could be spent for purchasing fungicides (Habte, 2005).

For many years, HR approach was applied in different African countries such as Tanzania, Cameroon, Kenya, Ethiopia (Bedimo et al., 2008). The exhaustive testing of materials through selection for resistance to CBD on locations (the mother trees) and in laboratory (their progenies) where the epidemic is not only severe but also regularly present is very vital (Derso et al., 2000). Masaba and Van der Vossen (1982) reported that, formation of cork barriers as the mechanism of resistance to CBD based on the stability against changes in pathogen populations. With this focus Jimma Agricultural Research Center has developed 40 coffee varieties characterized by high yielder, with the better quality for different Ethiopian agroecollogies (Benti, 2017). Among these varieties, 31 of them are reported as CBD resistant selections. Resistance to CBD in Coffea arabica most probably is horizontal/ quantitative in nature and controlled by three to five recessive genes (Van der Graff, 1984).

Biological control: implies the involvement and the use of antagonistic microorganisms like fungi and bacteria which are specialized and have modified genes or gene products to attack and control plant pathogens and the diseases they cause. Plant pathogenic microbes have an enormous impact on agricultural productivity and will cause total crop loss. Pathogenic fungi in general and Fusarium spp. in particular are highly destructive pathogens of both greenhouse and field-grown major crops under favorable conditions for disease development (Mulatu, 2012). Fungicide spray can limit the development of CBD but inadequate fungicide sprays or improper timing of sprays also result in higher levels of disease than where the sprays have not been applied. This is due to the direct substantial quantitative and qualitative effect of fungicides on the non-target micro flora some of which are antagonistic to the CBD pathogen (Margaret, 2011). This indicates that, there must be integration of biological CBD control methods with the other components like HR.

Biological elements or biological control of plant pathogens are the primary factors in disease suppression but the possibility in the sustainable context is becomes the major contradiction issues among researchers (Weller et al., 2002). The rich in diversity of the microbial world provides a seemingly endless resource for this purpose. Biological control is also likely to be more robust than disease control with synthetic chemicals. The complexity of the organism interactions, the involvement of numerous mechanisms of disease suppression by a single microorganisms and the addictiveness of most biocontrol

17 agents to the environment in which they are used all contribute to the belief that biocontrol will be more durable than synthetic chemicals (Serani et al., 2007; Rodriguez, 2016).

The groups of soil microorganisms with antagonistic properties towards plant pathogens are diverse, including plant associated prokaryotes and eukaryotes. Bacillus/Paenibacillus and Pseudomonas spp. are the most widely used biocontrol agents which have receiving much attention and tested on a wide variety of plant species for their ability to control diseases. The worldwide interest in these groups of bacteria was sparked by studies initiated for sustainable production systems. They are the main candidates for the biological control of diseases induced by fungal pathogens and they have been applied successfully to suppress Fusarium wilts of various plant species (Mulatu, 2012; Rodriguez, 2016).

Asfaw et al. (2018) collected 348 specimens (from leafs, twigs and berries) from Oromia and SNNPs regions to test as the bio-agents for the management of CBD pathogen under laboratory conditions. After stepwise testing of these isolates, they reported 33 bio agents (10 anti-biotic bacteria, 10 antibiotic yeast, 10 antibiotic filamentous fungi and 3 lytic bacteria) as effective bio agents of CBD control under laboratory conditions. This indicates that, if we give great concern in the biocontrol agents for CBD management there will be effective result also at field conditions.

Induction of systemic resistance: this is mechanisms associated with gene induction, the activation of a wide range of resistance mechanisms and the production of a wide range of defense compounds which result in metabolic and structural changes inside plants against pathogens (Sillero. 2012). This is a widespread phenomenon that has been intensively investigated with respect to the underlying signaling pathways as well as to its potential use in plant protection (Saikia et al., 2003). Different authors reported that, application of plant defense inducing chemicals are used to triggering host resistance against pathogens (Ata et al., 2008; Aleandri et al., 2010). These plant inducing chemicals are also important in triggering resistance in C. arabica against CBD.

Recently in Ethiopia, Alemu et al. (2018) conducted an experiment on the effects of inducing chemicals (Jasmonic acid, Monopotassium phosphate, Dipotassium phosphate and Salicylic acid) against the development of CBD via artificially inoculated coffee hypocotyls and have reported that all the tested chemicals are effective for the reduction of

CBD index. Finally, they indicated that application of these compounds in different concentration can give the best result in CBD control, indicating they could serve as the alternative potential tools for future CBD management in our country.

Chemical control: The use of chemicals are the most common practices which can give immediate control of anthracnose diseases like CBD but cultivation methods and yield levels limits application. It is often difficult to sprays trees which are unpruned and irregularly spaced. Many areas are inaccessible during the rainy season. Therefore, areas which have an above average yield prospect, suffer due to severe losses and the disease remains unchecked. The proper uses of chemicals are heavily subsidized to make treatment attractive in the suitable areas (Vander Graff, 1981).

The high cost of fungicides (45% of production cost including the labors), appearance of resistant pathogen biotype and other social and health related problems of the conventional agriculture on the environment is the immense topics when thinking to increase the interest of sustainable agriculture and biodiversity conservation (Swami and Alane, 2013). Furthermore, millions of coffee farmers are facing problem not only with low coffee prices but also a growing interest in organically grown coffee across the globe. Despite such elaborate control measures, losses of the potential crop becoming high under unfavorable weather conditions. Moreover, it is costly in a context of smallholdings struggling to cope with an unprecedented economic crisis (Alwora and Gichuru, 2014; Etana, 2018). These problems make it essential to look for alternative strategies that can ensure competitive and save habits of coffee production systems (Abera et al., 2011; Ngouegn et al., 2017).

3. MATERIALS AND METHODS

3.1. Description of the study area

The present study consisted of field survey, laboratoryand greenhouse experiments. The field survey for assessment of CBD was carried out at farmers’ field of Gidame district (Table 1). Gidame district is found in Western Wollega zone of Oromia region which is located around 08. 696oN latitude and 34.782oE longitude with an elevation of 1844 m.a.s.l. It receives 1750 mm of mean annual rainfall and has mean annual temperature of 25 oC (Gidame Agricultural Office report, unpublished).

Table 1. Descriptions of coffee farm study areas (Kebeles) in Gidame district Latitude Longitude Altitude range Kebeles (oN) (oE) (m.a.s.l) Gray Sonka 09.089 34.478 1645-1805 Gray Bisha 9.076 34.531 1590-1725 Gray Horo 9.044 34.537 1676-1710 Geba Fechasa 9.029 34.565 1645-1695 Buri 08.989 34.496 1830-1969 Abote 09.100 34.591 1545-1615 Bata 08.961 34.555 1892-1967 Chomen kela 8.951 34.577 1825-1989 GidameTwon 08.988 34.603 1815-1875

Figure 3. Map of the study area

20 Laboratory and Greenhouse experiments (isolation and characterization of C. kahawae, pathogenicity test and resistance evaluation) were conducted at Jimma Agricultural Research Center (JARC). The field experiment to evaluate disease resistance in C. arabica germplasm collections was conducted at Gera Agricultural Research Sub-Center (GARSc) by super imposing on already planted coffee accessions from Gidame district in 2014/2015 cropping season. GARSc is located in Jimma zone, South Western Ethiopia (Latitude: 7.1170 N; Longitude: 36.00 E) with an average elevation of 1900 m.a.s.l. The area represents cool sub humid, low to high altitudes of coffee growing agroecologies and receives an average annual rainfall of 1877.8 mm. The minimum and maximum temperature of the area is 10.4 and 24 0C respectively (Netsere and Kufa, 2015).

3.2. Assessment of CBD in farmers’ field

Assessment of CBD in farmer’s field was carried out in Gidame district of Western Wollega Zone during the 2017/18 cropping season. A total of 9 Kebeles were selected purposively based on the altitude and potentials of coffee production. From each kebele, 5 coffee fields were selected purposively and visual assessment of CBD on the whole plant and on berries were conducted following the procedure Mohammed and Jambo (2015). Ten trees were randomly selected from each farm by diagonal fashion within 6-8 m intervals. Then, each tree was monitored for the absence or presence of CBD symptoms (like scab and dark sunken lesion on the berry, berry rot, depressed or dried berry and fruit fall before harvest). Percent disease incidence was computed using the formula:

Disease incidnce (DI) = x100

On-farm CBD severity scoring was carried out following similar procedures on the same plants selected for DI assessment. Briefly, 3 branches from 3 strata (top, middle and bottom) were selected from each tree and CBD severity was estimated using a disease assessment key 0-6 disease score scale (Table 2) via critical observation of the lesion size and its extent (spread) on the diseased berry parts using the field guide (Kamau, 2015).Accordingly, percent severity index (PSI) was computed as

PSI = x100 .

Finally, CBD prevalence was determined as the number of infected fields over total number of fields assessed during the survey.

Furthermore, basic information such as production system and shade type was recorded and the record of GPS coordinates and altitude (m.a.s.l.) were taken using smart GPS instrument. The secondary information like age of the tree, types of varieties cultivated in the farm and disease management practices was obtained by simple questionnaires on the households of each assessed farm (Appendix 11).

Table 2. Assessment key for evaluation of coffee berry disease severity in Coffea arabica (adopted from Mohammed and Jambo, 2015) with sight modification

Disease index Descriptions

0 Healthy green berries without symptoms 1 Black sunken lesions cover < 2% of the green berries surface 2 Black sunken lesions cover 2-5% of the berries surface; approximately3mm in diameter 3 Black sunken lesions cover 6-10% of the berries surface shows black lesions approximately 5 mm in diameter 4 Black sunken lesions cover 11-50% the berries surface; approximately 7mm in diameter 5 Black sunken lesions cover 51-99% of the berries surface; approximately15 mm in diameter 6 >99% or the whole surface of berries covered with black sunken lesions; mummified berries

3.3. Sample collection and fungal isolation

To verify the identity of the pathogen causing CBD, a total of 45 farms were surveyed and samples of 30 green coffee berries (per farm) with active and black sunken lesion (Fig. 4) were collected from 21 farms (where CBD recorded). Immediately after sampling, the diseased green berries were placed in a plastic bag and reserved in cool box at 18-20 oC (Kilambo, 2008) at JARC Phytopathology Laboratory until further isolation.

Figure 4. Typical symptoms (black sunken lesions) on coffee berry disease of Coffea arabica berries. Photo taken by the Author

3.4. Isolation and identification of C. kahawae isolates

Studies on morphological characteristics aimed at identifying or differentiating C.kahawaeisolates collected from the study areas. For this study, 14 pure cultures of representative C. kahawae isolates from Gidame kebeles and one reference isolate from Gera (GC) were used. The cultural characteristics like mycelia growth (vigor, aerial growth, colony color, density and radial growth) and the microscopic characteristics such as conidial size and sporulation were studied with the method used by Zeru (2006) and Mohammed and Jambo (2015).

Isolation and identification of the causative agent of CBD from infected berries were carried out as described in Beynon et al. (1995) and Photita et al. (2005). Briefly, 5 × 5 mm pieces of outer cover tissues were taken from the margin of infected green berries (half healthy and half infected) were surface sterilized by dipping into a 5% sodium hypochlorite solution (0.5%W/V) and rinsed three times by immersing in sterilized distilled waterfor one minute. After removing the excess moisture by placing under a laminar hood for 1hr, samples were transferred on the prepared Petri plates containing potato dextrose agar (PDA). Then, plates were sealed with parafilm and incubated at 25oC for 7-10 days. For further purification and identification, the mycelia growth from the marginal growth was taken and transferred to a new PDA petri plates under aseptic conditions. Two antibiotics streptomycin (50 µg/ml) and chloramphenicol (25µg/ml) were used to avoid bacterial contamination. In case of contaminated plates, sub-culturing was made to get pure cultures. Finally, pure cultures were preserved in 50% PDA slant method and stored at 4 oC for later.

23 3.5. Morphological characterization of C. kahawae isolates

3.5.1. Macroscopic characterization

Cultural (Macroscopic) and morphological (Microscopic) characteristics of C. kahawae isolates were studied following the procedures of Biratu and Hulluka (1995) and Zeru (2006). Cultural variations of the isolates were examined by comparing the cultures of the isolates grown on PDA plates in three replications. For comparison, the reference isolate (GC) originally isolated from Gera was used as a check. Accordingly, the mycelial tip of 10 days old culture of the isolates were aseptically transferred to the center of PDA plates and incubated at 25 oC in three replications. Cultural features including colony color, texture and mode of growth were examined 7 days after incubation. In addition, the radial growth of each isolate was measured using a ruler from two directions on the reverse side of the petrideshes in 3 days’ interval. From these measurements, mean diameter of the fungal colony growth and the rate of radial growth were calculated. Moreover, mycelia color and appearance were examined from obverse and reverse sides of the culture by using Red Green Blue (RGB) color codes chart (Rayner, 1970). Vigor of aerial mycelium growth (dense, medium and irregular or scarce type) was examined from the obverse sides of the 10 days old plates.

3.5.2. Microscopic characterization

For microscopic characterization, 10 days old cultures of the isolates which showed difference in cultural characteristics (section 3.5.1) a total of 15 representative isolates including the standard Gera isolate were used for this analysis. Conidial size and sporulation capacity were studied under microscope (compound microscope adjusted at 40x magnification). To determine sporulation capacity, 10 days old fungal cultures on PDA plates were washed by flooding with 10 ml of distilled water, rubbed with sterilized scaple and transferred to 50 ml sterilized beaker, thoroughly stirred for 15 minutes with magnetic stirrer and then filtered through double layers of sterile cheesecloth. Adrop of conidial suspension were placed on microscopic slides and examined under light microscope. The number of conidia per milliliter was counted using haemocytometer and the average number of conidia per milliliter was calculated after taking 9 haemocytometer counts of each isolate. Afterwards, the conidial size (length and width) was measured by randomly taking 90 conidia per isolate using ocular micrometer fitted into compound microscope adjusted at 40x magnification (Biratu and Hulluka, 1995; Zeru, 2006).

3.6. Determination of virulence in C. kahawae isolates

This study was carried out to determine the level ofvirulence/ aggressiveness of 15 C. kahawae isolates (14 isolates collected from Gidame district, selected out of 21 isolates based on morphological differences) plus the reference Gera isolate GCon C. arabica detached green berries of the well-known susceptible variety 370 (Belachew, 2001)obtained from JARC was used. Accordingly, level of virulence was assessed using the detached berry test (DBT) following the procedures used in Zeru (2006) and Kamau (2015). Prior to inoculation, fully expanded healthy green coffee berries were collected from matured coffee trees selected on the basis of green berry maturity 14 to 15 weeks after flowering (Kilambo et al., 2013, Kamau, 2015).

Briefly, the green berries were surfaced sterilized with 5% sodium hypochlorite solution and rinsed three times with SDW and dried using sterile cotton cloth. The berries were arranged into a plastic box lined with tissue paper and all the representative isolates (14 isolates) that are identified morphologically and the well-known aggressive C. kahawae isolate from Gera (GC isolate) was used as the source of inoculum. The experiment was laid out in CRD design in 3 replications containing 6 berries per replication.

3.6.1. Inoculum preparation and inoculation

Conidial suspension of each isolate was prepared from 10 days old culture. The C. kahawae isolates cultured on the PDA plate was washed by flooding with 10 ml of SDW, rubbed with sterilized scaple and transferred to 50 ml sterilized beaker, thoroughly stirred for 15 minutes with magnetic stirrer and then filtered through double layers of sterile cheesecloth. Spore concentration was determined with haemocytometer and adjusted to 2x106/ml (Pinard et al., 2012; Kamau, 2015). Then, a drop of (≈25µl) the spore suspension was placed at the center of each berry (Fig .5) using pipette. Un inoculated boxes were sprayed with SDW were used as control. After hermetically closing the boxes to maintain high relative humidity needed for infection process and symptom development, boxes were incubated at 25 oC for 14 day.

A B

Figure 5. Detached green Coffea arabica berries inoculated with Colletotrichum kahawae spore suspension(A) and sterile distilled water (B)

3.6.2. Data collection

Data collection was carried out at three time points starting from7thdays post inoculation (DPI) when the 1st symptoms of CBD appeared and (Pinard et al., 2012). Disease incidence was computed after counting the number of infected and healthy berries. Whereas, disease severity was determined as the percentage of necrotic area using the visual scale from 0 to 100% of the total berry surface affected on the scale of 0-6 described under section 3.2. Finally, average infection percentage (AIP) was calculated as

AIP = [Ir1 + Ir2 + Ir3]/ Where, I sum of disease score, r replication and N the total number of berries in the replication (Kamau, 2015)

3.7. Evaluation of Gidame coffee accessions for resistance against CBD under field conditions, attached berry test

This study was conducted at (Gera Agricultural Research Sub Center (GARSc) coffee planting trial site. The experiment was superimposed during the 2017/18 cropping year on four years old coffee trees which have been planted since July 2014/2015. The experiment was laid out in a 10x10 simple lattice design with two replications consisting of 92 C. arabica accessions and 8 CBD resistant known varieties W92, W76, W66, 8136, W78, 7514, 7416 and 7576 (Adugna and Jefuka, 2008) were used as reference for screening of Gidame C. arabica accessions in the field trial (Table 3). Each replication contains 10 incomplete blocks with 10 germplasm. Each plot in the incomplete blocks consisted of 6 coffee trees with 2m spacing between rows and plants. All the agronomic management practices were applied according to standard procedures. For early discrimination of the susceptible accessions, the overall disease pressure (CBD incidence) was assessed on each individual accession following similar methods described in section 3.2. The promising accessions that showed lower CBD infestation in visual scoring (Appendix Table 1) were used in the next step resistance screening, attached berry test (ABT) and detached berry test (DBT).

The ABT was conducted by applying C. kahawae inoculum at Gera on the branches of growing green berries following the procedures of Van der Graff (1981) and Kilambo et al. (2013). The aim of this study was to estimate the difference in natural infestation and further verify the resistance level of coffee accessions with artificial inoculation. The Gidame coffee accessions that showed better resistance to CBD under field conditions were used in this study (Table 4). For inoculation, the virulent C. kahawae isolate originally isolated from Gera (Derso and Waller, 2003) was used. Inoculation was performed by random sampling of 3 trees/plot and then 3 strata/tree followed by 1 branch/stata (from top, middle and bottom layers), resulting in a total of 9 branches per plot.

For inoculum preparation, green berries (from infested fields) with black active lesions were collected in plastic boxes from Gera, slightly wetted with sterile distilled water (SDW) and stored at room temperature (RT) for 48 hrs. After sporulation, the berries were rinsed and the suspension was filtered using sterile cheesecloth. Then, conidial density was adjusted to 2x106 conidia/ml using haemocytometer and the marked strata were

27 sprayed with (≈25 µl per berry) of C. kahawae spore suspension using hand sprayer (Van der Graff, 1981). Immediately after inoculation, branches were covered with a paper bag to favor disease development. The bags were removed 24 hrs. after inoculation (HAI). Three weeks after inoculation severity data was scored and computed based on the methods as described under section 3.2.

Table 3. List of Coffea arabica accessions/varieties used in the field trial. Collection Accession Collection Accessions Collection No Accession name No No Site Name Site name site 1 G1 Geba Fechasa 35 G35 Gray Horo 69 G69 Gidame Town 2 G2 Geba Fechasa 36 G36 DhaboteRabo 70 G70 Gidame Town 3 G3 Geba Fechasa 37 G37 DhaboteRabo 71 G71 Gidame Town 4 G4 Geba Fechasa 38 G38 DhaboteRabo 72 G72 Gidame Town 5 G5 Geba Fechasa 39 G39 DhaboteRabo 73 G73 Gidame Town 6 G6 Geba Fechasa 40 G40 DhaboteRabo 74 G74 Horo Kundi 7 G7 Geba Fechasa 41 G41 DhaboteRabo 75 G75 Horo Kundi 8 G8 Geba Fechasa 42 G42 DhaboteRabo 76 G76 Horo Kundi 9 G9 Geba Fechasa 43 G43 DhaboteRabo 77 G77 Horo Kundi 10 G10 Geba Fechasa 44 G44 Gray Sonka 78 G78 Horo Jilo 11 G11 Geba Fechasa 45 G45 Gray Sonka 79 G79 Horo Jilo 12 G12 Geba Fechasa 46 G46 Gray Sonka 80 G80 Horo kayo 13 G13 Buri 47 G47 Gray Sonka 81 G81 Horo Jilo 14 G14 Buri 48 G48 Gray Sonka 82 G82 Horo Jilo 15 G15 Buri 49 G49 Gray Sonka 83 G83 Horo Jilo 16 G16 Buri 50 G50 Gray Sonka 84 G84 Dale Gere 17 G17 Buri 51 G51 Kelem 85 G85 Dale Gere 18 G18 Buri 52 G52 Kelem 86 G86 Dale Gere 19 G19 Buri 53 G53 Kelem 87 G87 Dale Gere 20 G20 Buri 54 G54 Kuri 88 G88 Dale Gere 21 G21 Buri 55 G55 Kuri 89 G89 Dale Gere

Table 3. List of Coffea arabica accessions/varieties used in the field trial…continued

No Accession Collection No Accession Collection No Accessions Collection name Site name Site name site 22 G22 Buri 56 G56 Kuri 90 G90 Dale Gere 23 G23 Abote 57 G57 Kuri 91 G91 Lalo Gere 24 G24 Abote 58 G58 Kuri 92 G92 Lalo Gere 25 G25 Abote 59 G59 Kuri 93 W-78 RV 26 G26 Abote 60 G60 Kuri 94 W-66 RV 27 G27 Abote 61 G61 Bata 95 W-92 RV 28 G28 Abote 62 G62 Bata 96 W-76 RV 29 G29 Abote 63 G63 Chomen kela 97 7514 RV 30 G30 Abote 64 G64 Chomen kela 98 7576 RV 31 G31 Gray Horo 65 G65 Chomen kela 99 7416 RV 32 G32 Gray Horo 66 G66 Chomen kela 100 8136 RV 33 G33 Gray Horo 67 G67 Chomen kela 34 G34 Gray Horo 68 G68 Chomen kela

RV = Reference varieties. All accessions have been collected in 2013.

Table 4. List of Coffea arabica accessions/varieties used in the Attached Berry Test.

Accession No. Accession Accession Accession No. No. No. name name name name

1 G-42 13 G-19 25 G-50 37 G-66 2 G-47 14 G-13 26 G-31 38 G-52 R 3 G-48 15 7514R 27 G-85 39 W-92 4 G-49 16 G-16 28 G-87 40 G-57 5 G-92 17 G-82 29 G-89 41 G-51 6 G-72 18 G-91 30 G-67 42 G-56 7 G-73 19 G-10 31 G-65 43 G-55 R 8 G-71 20 W-78R 32 G-69 44 7416 9 G-77 21 G-54. 33 G-68 45 G-37 R 10 G-83 22 G-84 34 8136 46 G-21 11 G-15 23 G-40 35 G-70 47 G-63 R R 12 W-76 24 W-66R 36 G-64 48 7576

All accessions have been collected in 2013. (R) resistant reference varieties

3.8. Evaluation of Gidame coffee accessions for resistance against C. kahawae under laboratory conditions, detached berry test

For this study, 30 best performed accessions/varieties from ABT study (see section 3.5; Table 4), 4 highly susceptible C. arabica accessions identified under visual disease score in the field as well as 2 well known reference varieties (the resistant 741 and susceptible 370) were used. The experiment was laid out in CRD design in three replications containing 20 berries per replication. Similar methods of inoculum preparation, inoculation and data collection procedures were used as described under section 3.6.1.

3.9. Statistical analysis The survey data for disease incidence and severity were analyzed using one-way ANOVA. Differences in macroscopic and microscopic characteristics of the isolates, disease data from resistance evaluation and virulence determination studies were analyzed using one- way ANOVA (SAS version 9.3 software packages). Treatment means were compared using Duncan Multiple Range Test (DMRT). Before analysis of variance, all data sets were tested for normal distribution using the normality test, the data from field evaluation

31 was transformed with Arcsin (score data) and ANOVA was performed for field and greenhouse data (Gomez and Gomez, 1984). The relationships among disease variables, altitudes of surveyed farms and pathogen characteristics were determined by Pearson (product-moment) correlation analysis using the SAS software (Proc procedure).

4. RESULTS AND DISCUSSION

4.1. Assessment of CBD in farmers’ field

4.1.1. CBD incidence and severity

The survey result showed that CBD was prevalent in all assessed areas (i.e.100%) and there was significant difference (p<0.05) among kebeles (Appendix Table 2). The magnitude of CBD incidence was relatively highest (86.0%) at Chomen Kela kebele which was not different significantly from Bata, Buri, Geba Fechasa, Gidame Town and Gray Sonka kebeles (Table 5). On the other hand, relatively the lowest (66.0 and 68.0%) CBD incidence was recorded from Gray Horo and Gray Bisha kebeles but the values were not statistically different from Abote, Gray Sonka and Geba Fechasa Kebeles. The overall DI ranged between 66.0 to 86.0% with an average incidence of 76.0% (Table 5).

Moreover, the result of the analysis of variance showed highly significant difference (p<0.001) in percent CBD severity index among assessed kebeles (Appendix Table 3). CBD PSI ranged between 19.7 to 49.7%. The highest PSI was recorded in Buri (49.7%) and Bata (49.0%) kebeles which was not differ significantly from Chomen Kela, Gray Bisha, Geba Fechasa and Abote kebeles (Table 5). Interestingly, Gray Horo Kebele had the lowest (19.7%) CBD PSI. The overall average CBD PSI was 38.2% and relatively higher PSI (>39%) was recorded in Buri, Bata, Chomen Kela, Gray Bisha and Geba Fechasa kebeles. On the other hand, the PSI in Gray Sonka, Gray Horo, Abote and Gidame Town was lower than the overall average (Table 5). Unlike DI, which tells the presence or absence of the disease in the assessed areas, PSI indicates the percentage of relevant host tissue covered by disease symptoms (Aswan et al., 2016) which is directly correlated with the damage caused on coffee farms due to the intensity of CBD.

Table 5. Disease incidence and percent severity index of coffee berry disease at different kebeles of Gidame district Kebele DI PSI Gray Horo 66.0c 19.7C Gray Sonka 74.0abc 32.0b Gidame Town 74.0abc 34.3b Abote 70.0bc 36.3ab Geba Fechasa abc ab 78.0 39.7 Gray Bisha 68.0c 41.0ab Chomen Kela 86.0a 42.0ab Buri 84.0ab 48.9a Bata 84.0ab 49.7a CV (%) 14.1 24.7 Means with the same letters are not differing significantly (DMRT; 14.2-16.1 of DI and 12.9-14.7 of PSI at p<0.05 and P<0.01). The mean values are obtained from the sum of five farms. The CV value is gained after data transformation.

In this study, all assessed farms were covered with local coffee cultivars which are known for low yield and genetic susceptibility to CBD. The farmers managed their coffee farm using hand ploughing and slashing. This factor could favor disease development in the areas. As compared to report made by Alemu et al. (2016) of 30-80% CBD incidence in the Western Wollega Zone (Gimbi and Haru districts), the result 66-86% of CBD incidence at Gidame district is very important. In Ethiopia, it has reported that under favorable conditions CBD can cause up to 100% yield losses unless management options are applied (Adugna et al., 2009a; Gidisa, 2016). The presence of susceptible coffee landraces isconducive for the development of CBD epidemic which influence coffee production and reduce the income generation to farmers (Benti, 2017; Castiblanco et al., 2018). Moreover, climate change could also be contribute to affect the potential of some of the previously released good yielding and resistant varieties (Belachew and Teferi, 2015; Alemu et al., 2016).

4.1.2. Relationship between CBD intensity and altitude

Results of the Pearson correlation analysis between disease intensity (DI and PSI) and altitude showed highly significant (P<0.001) and strong positive correlation (r=0.61) and PSI (r=0.55) respectively (Table 6).

Table 6. Pearson correlation analysis between major factors and intensity (incidence and severity) of coffee berry disease in Gidame district

Variables DI PSI Alt CPS

PSI 0.78**

Alt 0.61** 0.55**

CPS 0.31* 0.29* -0.35

Age 0.46* 0.36* - 0.23 -0.24* DI= Disease incidence, PSI= percent severity index, CPS= coffee production system. * and ** significant level at P<0.05 and P<0.01 respectively.

70 y = 0.0566x - 61.531 60 50 40 30 20 10

Disease severity (DS) severity Disease 0 1200 1300 1400 1500 1600 1700 1800 1900 2000 Altitude (m)

Figure 6. Relationship between severity (%) of coffee berry disease and across altitude range

Compilation of DI and PSI data across altitudes also revealed increased disease intensity with increase in altitude (Fig. 7). The highest coffee tree damage was recorded between the altitude range of 1879-1989 m.a.s.l. with highest mean in DI and PSI of 90 and 50.3%, respectively. The survey result showed that CBD damage increased with increase in altitude. Similarly, Bedimo et al. (2007), Zeru et al. (2009), Emana (2015) and Kagezi et al. (2018) pronounced the impact of CBD is high in the altitude areas where favorable environmental conditions are found, especially temperature and moisture conditions (i.e. relatively low temperature and high moisture together with susceptible host) exist.

Moreover, similar results on the positive relationship between altitude and level of CBD

35 have been reported by Zeru et al. (2009) who found that CBD was more common in the higher elevations in Illubabor zone. Alemu et al. (2016) also reported that CBD intensity was more prevalent in higher altitude areas of Oromia and SNNPs region of Ethiopia. Furthermore, Davis et al. (2012) and Mtenga and Reuben (2016) supported that CBD damage is high in the areas where susceptible genotypes and favorable environmental conditions (relatively low temperature, high altitude and high moisture) occur.

The information regarding the existence of the high CBD infestation (severity and incidence) in Gidame district as much important and gives deep insight for the CBD importance particularly in the district. Though, Jirata and Asefa (2000) and Alemu et al. (2016) have reported the intensity of 22% and 30-80% of severity and incidence respectively in Western Wollega zone (Gimbi and Haru districts), indicating that the disease can lead to a complete yield loss whenever susceptible landraces are cultivated without adequate control measures (Belachew and Teferi, 2015).

DI PSI 120 100 80 60 40 90.0 70.0 71.4 75.5 60.0 50.3 20 32.8 34.8 37.9 16.7 0 Mean of DI and of DI PSI(%) Mean 1434-1545 1546-1656 1557-1767 1768-1878 1879-1989 Altitude range Figure 7. Disease incidence (DI) and percent severity index (PSI) of Coffee berry disease across different altitude ranges in Gidame district. For variable mean data obtained from 9 kebeles is shown. Bras indicate standard deviations.

Since Ethiopia is the center of origin and diversity for C. arabica (Derso and Waller, 2003; Benti, 2017), variations in susceptibility of coffee cultivars to CBD were observed since the first outbreak in 1971. The dissemination potential of the pathogenfrom one coffee cultivation area to new areas has been great threat at high altitude areas (Kebati et al., 2016; Batista et al., 2017). The disease survey data obtained from the present study can give a useful insight (preliminary information) on the extent of CBD at Gidame district and could be used as valuable input for designing short and/or long term management

36 strategies such as selection of resistant lines that can adapt to various ecological conditions. Despite the great effort that has been undertaken to replace the CBD susceptible coffee cultivars with resistant and better yielding ones, the old coffee trees (which are mainly CBD susceptible) still represent the largest share (>80%) of Ethiopian coffee production (Alemu et al., 2016; Etana, 2018). This situation could offer a good opportunity for early dissemination of the pathogen to new coffee production areas in many African countries especially in Ethiopia (Hindorf and Omondi, 2011; Weir et al., 2012). Consequently, CBD epidemic development is today’s great challenge of Ethiopian’s coffee production especially for small scale farms like the Gidame coffee growing farmers. Therefore, reducing the impact of CBD should be given a priority in order to boost coffee production in the country in general and the study area in particular. To this end, creating a strong collaboration among the different stake holders in the coffee production system may play a significant role in improving the sector from the establishment of coffee plantations towards commercialization and export of the commodity.

4.1.3. Relation between coffee production system and disease intensity

The survey result showed that there was weak positive correlation between coffee production systems and incidence (r = 0.31) and severity (r = 0.29) of CBD (Table 6). The relationship revealed that the disease intensity (DI and PSI) became slightly strong in the garden followed by semi-forest production systems (Fig 8).

DI PSI

100

40 73.1 80.0

isease intensity (%) intensity isease 20 36.5 40.4 D

0 Semi forest Garden Coffee production system Figure 8. Intensity of coffee berry disease among production systems across Gidame district. The bars in dictate the standard deviation

In the study area 80.0% of DI and 40.4% PSI was recorded from garden production system whereas 73.1% DI and 36.5% was from the semifrorest (Fig. 8). CBD intensity variation between garden and semifrorest production systems could be associated with different factors. In garden production system the farms are composed of homogenous coffee populations and has human interference (management strategies, cropping type, etc.) which could attribute for the increament of disease level. While, heterogeneous nature of the coffee populations coupled with the low inputs and low human inference in the semi- forest coffee production systems that attribute to the relatively reduced disease levels.

Similarly, Bieysse et al. (2009) reported that frequent human interference hastened the spatial and temporal dispersal of CBD pathogen. On the other hand, CBD intensity variation among production systemscould be existed due to the shade difference. In the garden system, majority of the coffee trees looked with or without shade that attribute to high disease intensity. While, in the semi-forest production systems coffee trees are self- grown with little human interference containing natural forest cover where different species of trees provide shade to the coffee trees (Cerda et al., 2017). Vaast et al. (2006), Bedimo et al. (2008) and Kebati et al. (2016) have reported that shading creates microclimatic conditions that help to delay fruit ripening which could lead to a shift in the period of berry susceptibility in relation to high disease pressure. Bedimo et al. (2007) also suggested that shade can affect and modify certain rainfall characteristics that could influence conidial dispersal. The canopy of these plants might intercept certain raindrops, divert others from their route or reduce the speed of those droplets that might reach the coffee trees (Jha et al.,2014). Similarly, Alemu et al. (2016) have reported variation in CBD intensity which was high in the garden production followed by semifrorest and forest production systems conducted in different coffee growing areas of Ethiopia.

4.2. Macro and microscopic characterization of C. kahawae isolates

4.2.1. Macroscopic characterization

Based on visual observation dense, scarce and very scarce types of colony (texture) were identified at the front side of most culture plats. From these, 46.6%, scarce, 33.3% dense and 20% very scarce types of colony appearance (mycelial growth) were identified from

38 each culture plates of the isolates described in table 7.

Moreover, difference in colony color of the isolates was observed from both sides the cultured plates (Fig 9). From front side, 20% of dark gray, 20% light gray and 20% white gray, 13.3% floral white, 13.3% lavender and 13.3% old lace colony colors were observed. From the back side of the plates, most of the isolates had reflected Navajo white and light golden rod colony color (26.6% of each) followed by 20.0%, 13.1 and 13.1% with light gray, medium violet and dark olive green colony colors, respectively (Table 7). In general, up to the first 4-6 incubation days most of C. kahawae isolates had whitish mycelial color and week after incubation colony colors have been changed into light gray on PDA. Similarly, Biratu (1995), Zeru et al. (2008) and Emana (2015) reported the existence of cultural variation of C. kahawae isolates from the samples collected from Kaffa, Illubabor and Hararghe, respectively.

The result in mycelial growth showed a highly significant difference (p< 0.001) among the isolates (Appendix Table 4). Overall mean radial mycelial growth rate ranged from 2.2 to 4.3 mm/day. Isolate GC was the fastest growing with the mean radial growth rate of 4.3 mm/day but did not differ significantly from isolate CK1, BA, AB1 and GB1. On the other hand, the slowest mean radial growth rate 2.2 mm/day was recorded from isolate GS followed by GF3 and AB2 isolates which did not significantly differ from GS (Table 7). Agreed with this result, Biratu and Hulluka (1989) have reported variations in radial mycelial growth of C. kahawae isolates with an average growth rate of 6.5 and 6.7 mm/24 hrs. for dark and grayish mycelia of CBD pathogen on PDA respectively. But the present result was more close with that of Zeru (2006) which was 4.4 mm/ 24hrs.

Table 7. Macroscopic characteristics of Colletotrichum kahawae isolates collected from Gidame district

Isolate Colony aspects Colony color Colony Growth Code Front Back (mm/day) GF1 Scarce White gray Navajo white 3.5cde BA Dense Dark gray Light Gray 4.1ab ghi AB2 Very scarce Lavender Light gay 2.6 GC Dense Dark gray Light gray 4.3a GS Very scarce Lavender Medium violet red 2.2i fgh BU2 Dense Old lace Light golden rod 3.8 CK2 Scarce Light gay Light golden rod 3.2efg GB1 Scarce white Gray Navajo white 3.8a-d GF3 Very scarce Floral white Navajo white 2.5hi def GH Scarce Dark gray Dark olive green 3.3 b-e GT Scarce Old lace Medium violet red 3.5 def GB4 Scarce Light gray Light golden rod 3.3 AB5 Scarce Light gray Light golden rod 3.3def abc AB1 Dense Floral white Navajo white 4.0 a Ck1 Dense White gray Dark olive-green 4.15 CV (%) . 9.1 Means followed with the same letters are not significantly different (DMRT; 0.5-0.6 at p <0.05). The reference Gera isolate shown in bold. The letters in the isolates code refers to the first two letters of the name of the kebele and the number indicates the farm’s number where the isolates are collected.

Cultural characters have a good taxonomic criterion in distinguishing fungalspecies like Colletotrichum (Zakaria and Bailey, 2000). Colletotrichum isolates associated with coffee differences in mycelial growth, colony characters and growth rates among the Colletotrichum isolates allowed them grouping into separate class by visual observation on the culture (Mnuel et al., 2010). Moreover, conidial morphology has always been emphasized over other taxonomic criteria in taxonomic investigations of the genus Colletotrichum (Kebati et al., 2016). However, distinguishing Colletotrichum species culturally can be difficult and needs experience and acritical observation or identification tools because some species found associated in coffee like C. gloesporioides, C. acutatum and C. kahawae closely related that makes confusion of cultural identification (Prihastuti et al., 2009). Kebati et al. (2016) also noticed that Colletotrichum isolates associated to coffee could show homogenous growth on cultural growth and makes confusion in cultural identification.

As C. kahawae species are slow growing nature in mycelial growth rate in culture medium, this may use as distinguishing criterion of CBD pathogen from other Colletotrichum species and could serves as indicator of variability within the species (Abra et al., 2016). C. kahawae has only been found in association with coffee and its characteristics are consistent among isolates across its range of distribution (Silvia et al., 2012; Batista et al., 2017). Nature of infection and its uniqueness to occupy ecological niche towards developing green berries is the key distinguishing characteristics of CBD pathogen from all other Colletotrichum species. Likewise, CBD pathogen appears mainly with close relation to C. gloesporioides but it can be separated easily by its distinct pathogenicity on green coffee berries (Emana, 2015; Pires et al., 2016).

B Figure 9 Cultural morphology of representative ten days old culture of

Colletotrichum kahawae isolates (A) Obverse side (B) Back side

4.2.2. Microscopic characterization

The result in microscopic characterization indicated considerable variations in conidial size among the 15 representative isolates (Fig 10). ANOVA also revealed highly significant difference (P<0.001) in conidial sizes among the isolates (Appendix Table 5 and 6). Accordingly, the largest and smallest conidial length and width were recorded from isolate GB2 and GC, GF3, Ck2 and GH respectively (Table 8). Compared to conidial width, more variability was observed in conidial length of the isolates. The ratio(indicating conidial variability of each isolate counted per ml) of the shortest isolates GH, GF3, Ck2 and GC of the length conidial to the longest isolate GB2 was 1: 6. GB2 isolate showed the largest conidial length (5.9 µm) and width (3.0 µm). The ratio of the narrowest (isolate BA and GB4) conidial width to the broadest isolate GB2) were about 1:5. In fact, conidial size of most isolates was in the range of 3.8-5.9 µm length and 2.0-3.0 µm width (Table 8).

Isolate CK1 Isolate GC

Figure 10. Conidial features of Colletotrichum kahawae isolates

The other microscopic characterization parameter considered in this study was sporulation capacity. Results showed that except GS, all isolates produced abundant conidia with highly significant differences (P<0.001) among each other (Appendix Table 7). The variation in number of conidia production ranged between 186.1 to 572.3 conidia per ml with the highest number produced by the reference isolate from Gera (isolate GC) followed by isolate Ck1which was 459.1 conidia/ml (Table 8). Whereas, least and the intermediate numbers of conidia was produced by GS (186.1 conidia/ml) and GB4 (340.3 conidia/ml) respectively. Derso and Waller (2003), Zeru (2006) and Prihastuti et al. (2009) reported conidial morphology as important distinguishing features of C. kahawae isolates from other Colletotrichum species and its sporulation potential can vary from isolate to

43 isolates.

Table 8. Microscopic characteristics of Colletotrichum kahawae isolates collected from Gidame district

Isolate Collection site Sporulation Capacity Conidial size(µm) Code (Kebeles) No of conidia/ml Length Width GF1 Geba Fechasa 344.9e 3.9def 2.1def BA Bata 440.7cb 4.3b 2.0g AB2 Abote 219.6gh 4.2cb 2.2cde GC Gera 572.3a 3.8ef 2.2cde GS Gray Sonka 186.1h 4.1b-e 2.2cde BU Buri 382.5ced 4.0b-e 2.1c-f CK2 Chomen kela 261.3fg 3.8ef 2.1c-f GB2 Gray Bisha 373.0de 5.9a 3.0a GF3 Geba Fechasa 200.0gh 3.8ef 2.3b GH Gray Horo 327.7ef 3.8ef 2.1def GT Gidame Town 364.6de 3.9def 2.2cde GB4 Gray Bisha 340.3 e 4.0b-f 2.0g AB5 Abote 330.3e 3.9 def 2.2cde AB1 Abote 430.0bcd 4.1bcd 2.1def Ck1 Chomen kela 459.3b 3.9def 2.3b CV (%) 11.5 3.5 2.5 Data shown mean values of 15 measurements. Means with the same letter are not statistically different (DMRT; 67.6 - 80.3 of sporulation, 0.25-.29 of conidial length and 0.09-0.12 of conidial width at P<0.05). Results of the standard isolate Gera is shown in bold.

4.3. Determination of virulence in C. kahawae isolates

The result in the virulent test revealed that all isolates were pathogenic to variety 370 and showed district and significant (p<0.001) variations in the level of aggressiveness (Table 9; Appendix Table 8). Interestingly, the highest level of berry infection (88.7%) was observed in plants infected with the reference isolate GC followed by isolates CK1, AB1, BA, CK2 and BU whose aggressiveness did not significantly differ from GC. On the other hand, the least infection percentage (34.8%) was caused by isolate GS followed by GF3, AB2, GH, AB5, GB4, GF1, GT and GB2 which caused significantly lower level of infection percentage compared to the reference isolate GC (Table 9).

Table 9. Virulence of Colletotrichum kahawae isolates on susceptible C. arabica variety 370 as determined by detached berry test

Isolates Berry infection percentage d GS 34.8 cd GF3 37.5 cd AB2 42.1 cd GH 44.9 cd AB5 45.7 cd GB4 48.4 cd GF1 50.3 bcd GT 58.7 bcd GB2 60.4 a-d BU 61.4 abc CK2 63.9 abc BA 65.0 ab AB1 78.6 ab CK1 83.1 a GC 88.7 Control 0.0e CV (%) 16.9 Isolates are sorted according to increasing level of aggressiveness. Means with the same letter are not statistically different (DMRT; 24.8-29.6 at P<0.05). The most aggressive reference isolate from Gera is marked in bold.

In the present study, the symptoms produced by the pathogen on artificial inoculation on the detached berries showed similar symptomsof those observed under natural infection at field. It is true that susceptible genotypes allow the expression of clear differences in aggressiveness when exposed to different pathogen isolates (Castiblanco et al., 2018). Chen et al. (2005) described that unlike other Colletotrichum species such as C. gloesporioides who do not often infect coffee berries except at berry maturity stage (red berries), C. kahawae causes distinct symptoms (like deep sunken lesions on the green berry surface) which subsequently invade the whole berry surface and destroy it completely. The occurrence and intensity of CBD varies from place to place and season to season, depending on mutual effect of host susceptibility, pathogen aggressiveness and favorable climatic conditions (Biratu, 1995). From the pathogen perspective, there might be variations in growth rate and production of physical structures and chemical secretions involved in host-attachment and penetration. Kamau (2015) have shown that conidia of C.

45 kahawae germinate and differentiate into melanized aspersoria both “in vitro” and “in vivo” and penetrate different coffee organs (hypocotyls, leaves and young green berries).

Pathogenicity tests on detached green coffee berries enables to separate C. kahawae from other Colletotrichum species. These conditions should constitute as the basis for further studies to confirm the ability of the detached berry technique to mimic field interaction and therefore it’s potential to contribute to a better understanding of the CBD epidemiology. The aggressiveness of the pathogen can be considered as quantitative measure of the level of the disease reached over time. This indicates that the most aggressive pathogen reached at a specific disease level faster than the less aggressive one. The situation can be measured via latent period, spore production, infection, lesion size and disease severity (Bedimo et al., 2012, Pires et al., 2016). Derso and Waller (2003) stated that the DBT technique can be used to compare the aggressiveness of several strains of C. kahawae from the same or different geographic origins. Despite the fact that isolates collected from different geographical locations may show similar symptoms on the inoculated genotypes, significant variations might exist in their pathogenicity, aggressiveness and other characteristics such as mycelial growth rate and conidial production (Fredric et al., 2015). Agreed with this result, Kamau (2015) reported considerable pathogenic variabilityamong C. kahawae isolates that are collected from Kenya on detached green berries.

4.3.1. Relationship between morphological characteristics and pathogenicity of C. kahawae isolates

Pearson correlation analysis revealed highly significant (P<0.001) and strong positive correlation of isolates virulence with the mycelial growth rate (r=0.96) and sporulation capacity (r=0.95) of C. kahawae isolates. In contrast, conidial size (conidial length and width) showed non-significant correlation with the virulence of the isolates (Table 10), suggesting that unlike the size of the fungal fruiting structures, faster mycelial growth rate and high sporulation capacity might contribute to the aggressiveness of C. kahawae isolates in C. arabica. Kilambo et al. (2013) recently studied the effect of C. kahawae strains on arabica coffee varieties in Tanzanian and stated that conidial size of C. kahawae spp. has no contribution for pathogenicity of the isolates.

Table 10. Relationship between morphological characteristics and pathogenicity/ aggressiveness of Colletotrichum kahawae isolates

Variables Pathogenicity P-value Growth rate 0.96** 0.001 Sporulation capacity 0.95** 0.001 Conidial length 0.21 0.15 Conidial width -0.03 0.12

The correlation result in this study revealed that differences in sporulation capacity among isolates could be a useful parameter for characterizing fungal isolates due to the fact that variation in spore density among the colletotrichum isolates was significant. Fungal isolates with high spore density have been shown to increase the rate of invasion of cell walls of the host plant by Colletotrichum spp. that leads the isolate to be more pathogenic to host (Jeffries et al., 1990: Bedimo et al.,2008)

Furthermore, the sporulation capacity and growth rate of colletotrichum isolates may have an important implication in the study of the epidemiology of the disease. The variability in fungal pathogenicity and the close relationship between sporulation and virulence could provide a useful information base for screening coffee germplasm collections for resistance to the pathogen and subsequent breeding programs for durable resistance through the selection of highly sporulated and virulent fungal isolates Kilambo et al. (2009),whichimplies that culture morphology and pathogenic variability could be preliminary parameters for characterizing the fungal isolates (Fokunang et al., 2000).

4.4. Evaluation of Gidame coffee accessions for resistance to CBD, attached berry test

Results obtained from the analysis of variance showed a highly significant difference (P<0.001) in the level of CBD resistance among the accessions (Table 11; Appendix Table 9). The highest CBD severity was recorded from accession G50 which showed highly significant CBD severity (33.4%) compared to all other accessions /varieties. Interestingly, the lowest (2.3%) disease severity (DS) was recorded from G63 followed by W76, G65, G72, 7416; G66, G57, G15 and G70 which did not statistically differ from G63 (Table 11). On the other hand, G84 and G85 showed the intermediate resistance similar with the known reference varieties (W66, 8136 and 7576) which were not significantly different from each other. The mean ranged between 2.3 to 33.4% for the attached berries (Table

11) and the result in this study clearly showed the presence of certain accessions that have been better or comparable CBD resistantto the reference varieties.

Table 11. Percent infection of Coffea arabica accessions/varieties inoculated with Colletotrichum kahawae in attached berry test

Coffee accessions Disease severity Coffee accessions Disease severity G63 2.3v G51 15.8h-o W76R 2.6v G71 16.1 g-o G65 3.7uv G21 16.8 f-n G72 4.7t-v G77 17.0 e-m 7416R 6.34s-v G16 17.7e-m G66R 7.3r-v G47 17.8d-m G57 8.7q-v G37 17.8d-m G15 8.6q-v G54 17.9d-k G70 9.3o-t G69 18.4c-k W78R 9.4o-t G48 19.3c-j W66R 10.3 n-t 7514R 19.3c-j 8136R 10.4n-t G49 19.4 c-j G84 10.9n-t G42 20.6 b-i G85 11.0 m-t G55 20.6b-i 7576R 11.0l-t G19 20.7 b-i G10 11.9d-k G73 21.6 b-g G91 12.7k-s G88 22.7 b-g W92R 12.7 j-s G83 23.2b-f G82 12.9j-s G31 23.7 b-e G56 13.2j-s G13 24.6bc G40 13.3j-r G92 24.8bcd G64 13.4-r G67 26.3b G87 14.4i-q G89 27.0B G52 15.1h-p G50 33.4A CV (%) 22.48 Accessions/varieties are sorted in the increasing level of severity. Means followed with the same letters are not significantly different (DMRT; 5.6 - 7.1 at p<0.05). (R) the Reference verities

The nature resistance in coffee was considered as horizontal in which the resistance is controlled by recessive genes (Van der Graff, 1981). In this study, the variations among C. arabica accessions for the pathogen response can be associated with the genetic makeup

48 of the accessions. For long time C. arabica is known to be a self-fertile crop but in the recent study by Yadesa (2014), C. arabica yields 76% of out crossing from neighboring plot of coffee accessions in the unmanaged forest trees this can leads to change in genetic make of the accessions and variations in host pathogen reaction.

On the other hand, Silva et al. (2006) and Van der Vossen and Walyaro (2008) found that host resistance appears to be largely based on the rapid formation of a cork barrier in the pericarp of the developing fruit distal (point of attachment) from the initial infection site that effectively prevents the pathogen from further invading of healthy tissue which totally absent or incomplete in susceptible host plants (Belachew, 2001).

In the resistant coffee genotypes, fungal growth can be restrict associated with a seriesof hypersensitive reaction (HR) responses. HR indicates the rapid and efficient plant resistance mechanism that leads to a localized plant cell death in response to invasion by a pathogen and is characterized by a rapid loss of membrane integrity in the infected host cells (Hoglund et al., 2005; Singh and Upadhyay, 2013). Previous studies showed that Arabica coffee resistance to C. kahawae also results from constitutive and induced mechanisms operating at different stages of pathogenesis (Loureiro et al., 2012). The difference response of the genotypes will depend on the lifestyle of the pathogen and genetic constituent of the host (Andolf and Ercolano, 2015; Ma and Ma, 2016). Likewise, variation could be existed within individual coffee selections of each coffee population (locality) in reaction to CBD. In this study, those considerable variations have also beenobserved among Gidame coffee accessions in percent infection of attached berries in ABT conducted at Gera. Similar findings were reported by Zeru (2006), Gichuru (2007) and Kilambo et al. (2013) they underlined that genotypic variation in pathogen infection under filed conditions could be examined by ABT via artificial inoculation but better knowledge of both the pathogens and the coffee plant diversity allowed to identify durable resistant varieties which are novel and economical approaches against CBD.

Generally, Plants have different ways of defense mechanisms that recognize potentially dangerous pathogens and rapidly respond, before pathogen cause serious damage (Sanabria et al., 2010). Once the pathogen attacks plant tissue, the host plant challenges the advancement of the infection in a series of defense reactions. Basal resistance is the first line of pre-formed and inducible defense response that protects plants against various

49 groups of pathogens (Thomas et al., 2011). Successful pathogens deploy effectors that would deceive basal defense for further infection and colonization. This chain of effector- resistance gene co-evolution is at least attributed to mutation and horizontal transfer of genes of the pathogen and selection pressure on plant for resistance (Anderson et al., 2010). variations in these coffee accessions could be an opportunity for the next better resistant varietal development through breeding. As resistance in perennial crops like coffee is observed and screened during the late stage of development (Van der Vossen et al., 2015), multi locational repeated evaluation of these accessions over time could be the future line of work for pathologists and breeders.

4.5. Evaluation of Gidame coffee accessions for resistance toC. kahawae under laboratory conditions, detached berry test

The result in this study indicated that there was highly significant difference (p<0.001) among C. arabica accessions (Table 12; Appendix Table 10). The result in the resistant reference variety 741 showed the lowest mean of CBD infection (6.7%) which significantly varied from all other accessions and/or varieties. In addition to this, relatively the lowest percent CBD infection (24.0%) was recorded from G65 followed by G63, W76, G15, G72, 7416 and G66 which did not differ significantly from G65. On the other hand, the intermediate infection percentage was recorded from W78, 8136, G70, G57, G85. While, the highest CBD percent infection (87.3%) was recorded from accession G78 which was also susceptible at field natural infestation followed by G04, G92, G71 370, 71 and G91 which did not differ statistically from G78. Interestingly, those three Gidame coffee accessions (G78, G04 and G71) showed higher infection percentage more than the well-known reference variety 370 in magnitude. On the other hand, from 4 susceptible accessions G78, G04, G03 and G02 that were taken from early discarded accessions during visual evaluation under natural field infestation, G78 and G04 correspondingly showed the highest susceptibility under the laboratory DBT more than the susceptible reference variety 370 (Table 12).

Table 12. Response of Coffee arabica accessions/varieties to Colletotrichum kahawae infection in detached berry test

Infection Infection Accessions percentage Accessions Percentage 741 6.7k G10 55.5fg G65 24.0j G82 56.8efg G63 25.5j G31 62.2d-g W76 28.2j G51 63.3d-g G15 31.9j G56 63.7c-g G72 34.0j G02 64.4c-g 7516 36.6ij W92 65.1c-g G66 38.1hij W66 66.1c-g 8136 51.3ghi G03 67.8c-g W78 51.3ghi G89 67.8c-g G84 51.5ghi G67 69.6b-f G57 51.6ghi G91 72.4a-e G85 52.4f-i G70 75.1a-d 7514 52.7f-i 370 76.5a-d G47 54.0fgh G71 80.5abc G83 54.1fgh G92 84.9ab G55 55.0fg G-04 85.1ab 7576 55.2fg G78 87.2a CV (%) 15.5 Accessions /varieties are sorted in the increasing order of infection percentage. Means followed the same letters are not significantly different (DMRT; 14.3-17.9 at p<0.05). The reference varieties had shown in bold. (R) Resistant variety. (S) Susceptible variety.

The result revealed that those highly susceptible accessions G78 and G04 under filed conditions with natural infestation become susceptible under controlled conditions in laboratory. This implies that genotypes which are susceptible under field conditions could be susceptible under controlled conditions if no change in pathogen strains (Biratu, 1995; Belachew, 2001).

In this study, clear variations between susceptible and resistant accessions were observed with detached berry inoculation test. There were clear observations that unlike resistant accessions in the areas of successful pathogen infection showed continuous and entirely

51 covered the berry surface with black lesion on the susceptible accessions (Fig. 11C) while restricted scab lesions that hindered further penetration of the pathogen into intercellular parts of the berries were observed on the resistant accessions (Fig 11B). and Waller et al. (2007) reported that scab lesion formation is resistant host response which is more common on the coffee cultivars possessing resistance.

In any pathosystem, there can be various strategies of infection and resistance in different combinations and magnitudes depending on genotypes and environmental (external and internal) conditions. In coffee and C. kahawae association different factors may induce susceptibility and operate at all stages of pathogenesis from conidial germination to sporulation of the fungus (Gichuru, 1997). By reducing conidial germination and apersoria formation, CBD resistant coffee types can be able to reduce infection sites. This gives the resistant accessions further advantage in that mobile resistance factors that can be concentrated into fewer areas, resulting in higher concentrations per infection site, even when initial quantities are similar to those in susceptible accessions (Pinard et al., 2013).

The existence of fungi toxic compounds (could be performed and induced by infection) in coffee could be another resistance mechanism. The brown crust formed on the berry surface could restrict further infection and leads to starvation of the pathogen (Gichuru, 1997). This could also be achieved by observed cell death at the periphery of infected area and the formation of cork barriers. The scab lesions are common expression of CBD resistance at distinct stages of pathogenesis (Gichuru et al. 2007). These mechanisms could eliminate bio trophic associations with pathogen and block nutrient transfer to the infected area (Waller, 2007). The occurrence of responses to infection ahead of the hyphae demonstrated the existence of elicitors which itis desirable to investigate further (Silva et al., 2006).

On the other hand, the finding by Chen et al. (2004) revealed that green coffee berries possess inherent antifungal compounds counter infection of coffee by C. kahawae strains. Such antifungal compounds may exist in these Gidame coffee accessions/ varieties that contributed to CBD resistance. However, more studies are needed to confirm the involvement of antifungal compounds in resistance of Gidame coffee accessions to CBD.

A B C

Figure 11. Coffea arabica accessions response for Colletotrichum kahawae, (A) the control (B), relatively resistant, (C) susceptible accession after 14th days of inoculation

The disease is initiated mainly from diseased berries (green, ripen and mummified) and infected plant parts (flowers, barks, twigs and leaves) and appears every year again on previously infected coffee trees. Detached berry technique is the possible means relative ranking of cultivar resistance stating from early time which is still useful for a differential interaction analysis and varietal characterization (Pinard et al., 2012). Generally, resistant varieties have the potential to reduce cost of production and are the save ways of disease management approach and needs great focus in the sustainable use (Newton, 2016). This promising accessions in this study which showed better result can be the base line of breeding programs in the future work.

5. SUMMARY AND CONCLUSION

The magnitude of CBD was assessed in nine representative kebeles in Gidame district of the Western Wollega zone. The survey result showed that CBD intensity was prevalent in all assessed fields of Gidame district that varied among kebeles with the range between 66 to 86% and 16.7- 49.7% of incidence and PSI, respectively.Besides, the extent of CBD intensity was increased with altitude range that revealed positive and strong relationship with incidence (r=0.61) and PSI (r=0.51) and the intensity was also varied among production systems. Relatively, CBD was more important in the garden than the semifrorest production systems. For this variation, human interference for pathogen dissemination and shade variation considered to be key attributes. However, more variability of CBD extent was observed in altitude range (in the highlands). As CBD information regarding to Gidame district was limited, this reportcan give important perceptionon CBD damage particularly in the study area that leads to set management and reduce the impact in the small scale farms.

Furthermore, total of 15 representative isolates, 14 from Gidame district and one reference isolate GC from Gera were used to study cultural, morphological and pathogenic variation among isolates. All these isolates studied for aerial mycelial growth on PDA and showed 33.3% dense, 46.6% scarce and 20% very scarce type. There were no isolates that exceeded the reference isolate (GC) in the mycelial radial growth within the range between 2.2 to 4.3 mm/day of all isolates. Morphologically, C. kahawae isolates were varied in conidial sizes (ranged 3.8-5.9 µm L± 2.0-3.0 µm) and sporulation capacity. Significant Variations was observed between isolates in conidial production and the highest number of conidia was produced by the reference isolate Which ranged from 186.1 to 572.3/ ml.

The result in the virulence determination test indicated that all the collected isolates were pathogenic to the detached green coffee berries green berries with varied aggressiveness among isolates in percent infection ranged ranging from 34.8 - 88.7%. The result confirmed that, fastest radial growth together with high sporulation capacity of the isolate have the direct contribution and indicating significant positive correlation with pathogenicity (r =0.96) and (r = 0.95), respectively. As well, no isolates showed higher aggressiveness than the reference isolate GC on the DBT. More importantly, in virulence study the control (green berries sprayed SDW) showed free of infection indicating the

54 absence of quicent infection (latent infection) on the berries implies all the inoculated green coffee berries were disease free.

The result in the resistance evaluation test under field (ABT)and laboratory (DBT), considerable variations were detected among coffee accessions. The mean percent of berry infection ranged from 2.31 to 33.4% in field ABT and 6.7 to 87.3% in the laboratory DBT. In comparison to all accessions/varieties, C. arabica accessions G65, G66, G63, G72 and G15 showed low CBD infection under both conditions. The genetic makeup of the accessions, out crossing, the production of fungi toxic compounds and physical barriers were considered to be factors contributing for resistance existed variation among coffee accessions.Likewise, the result in this study showed that CBD susceptible accessions G78 and G04 under filed conditions also showed the highest CBD infection in the laboratory conditions indicating resistance is genetically inherited and cannot be reversed unless changing the nature of the pathogen virulence.

Generally, the present study revealed the importance of CBD in Gidame district and demonstrates the role of host resistance in combating the disease. Those C. arabica accessions that showed low level of infection in CBD resistance evaluation test under filed and laboratory conditions in this study is an opportunity for further breading work and could be the best alternatives to CBD management particularly for the study area.However, for full recommendation of these accessions future research on searchingof more aggressive isolates via surveying various agroecologies, seedling hypocotyl test, promoting these resistant accessions across multi locations over multi yearsand molecular identification should be conducted.

6. REFERENCES

Abera, A., Lemessa, F. and Muleta, D., 2011. The antifungal activity of some medicinal plants against coffee berry disease caused by Colletotrichum kahawae. International Journal of Agricultural Research,6: 268-279.

Abera, A., Lemessa F and Adugna G., 2016. Morphological Characteristics of Colletotrichum Species Associated with Mango (Mangifera indica L.) in Southwest Ethiopia. Journal of Food Science and Quality Management, 48: 106-115.

Adugna, G. and Jefuka C., 2008. Resistance levels of Arabica coffee cultivars to coffee berry disease, coffee wilt and leaf rust diseases in Ethiopia. Proceedings of the 12th Crop Science Society of Ethiopia (CSSE), 22 – 24 May 2006, Addis Ababa, Ethiopia, pp. 92 – 103.

Adugna, G., Teferi D, Jafuka C, Tesfaye S, Zeru A. 2008. Research findings in coffee wilt disease. In: Adugna G, Bellachew B, Shimbir T, Taye E, Kufa T, editors. Coffee diversity and knowledge. Proceeding of a national workshop four decades of coffee research and development in Ethiopia, p. 205–260.

Adugna, A., Chemeda J., Arega Z., Tesfaye. 2009a. Advances in coﬀee diseases research in Ethiopia. Increasing crop production through improved plant protection. Volume ii proceeding of the 14th annual conference of the plant protection society of Ethiopia, 540: 275–303.

Adugna, A., Million, A., Hindorf, H., Arega, Z., Teferi, D. and Jefuka, C., 2009b. Coffee wilt disease in Ethiopia. Coffee Wilt Disease, pp.50-68.

Aleandri, M.P., Reda, R., Tagliavento, V., Magro, P. and Chilosi, G., 2010. Effect of chemical resistance inducers on the control of Monosporascus root rot and vine decline of melon. MediterraneanPhytopathology, 49: 18-26.

Alemu, F., 2013. Assessment of the current status of coffee diseases at Gedeo and Sidama zone, Ethiopia. International Journal of advanced research, 1: 192-202.

Alemu, K., Adugna, G., Lemessa, F. and Muleta, D., 2016. Current status of coffee berry disease (Colletotrichum kahawae Waller & Bridge) in Ethiopia. Archives of Phytopathology and Plant Protection, 49: 421-433.

Alemu, K., Adugna, G., Lemessa, F. and Muleta, D., 2018. Induction of systemic resistance in Arabica coffee (Coffea arabica L.) against coffee berry disease (Colletotrichum kahawae Waller and Bridge) mediated through plant defense activator. International Journal of Pest Management, pp.1-11.

Alemu, T. and Sokar, I., 2000. The status of coffee berry disease in minor coffee growing regions, pp. 29-34

Alwora, G.O. and Gichuru, E.K., 2014. Advances in the management of coffee berry disease and coffee leaf rust in Kenya. Journal of Renewable Agriculture,47:777-780

Amano, A.A., 2014. Coffee production and marketing in Ethiopia. European Journal of Bus Management, 6: 109-22.

Anderson, J. P., Gleason, C. A., Foley, R. C., Thrall, P. H., Burdon, J. B., & Singh, K. B., 2010. Plants versus pathogens evolutionary arms race. Functional Plant Biology, 37:499

Andolfo, G. and Ercolano, M.R., 2015. Plant innate immunity multicomponent model. Frontiers in plant science, 6: 987.

Asfaw, K., Husein T., Ayalew A., Dejene M. and Sahile S., 2018. In vitro Screening and characterizing the most promising antagonistic microorganism as Biocontrol Agent(s) against Colletotrichum kahawae. European Journal of Experimental Biology, 8: 1-13

Aswan, D., Dr. Kamala N. and Dr. Joy. M., 2016. Studies on morphological characteristics of Colletotrichumsp the causal organism of snake gourd anthracnose, 2: 160-166.

Ata, A.A., El-Samman, M.G., Moursy, M.A. and Mostafa, M.H., 2008. Inducing resistance against rust disease of sugar beet by certain chemical compounds. Egyptian Journal of Phytopathology, 36:113-132.

Bailey, J.A., O´Connell, R.J., Pring, R.J. and Nash, C., 1992. Infection strategies of Colletotrichum species. In: Bailey JA, Jeger MJ (eds), Colletotrichum: Biology, Pathology and Control. Wallingford, UK, pp. 88-120.

Batista, D., Silva, D.N., Vieira, A., Cabral, A., Pires, A.S., Loureiro, A.P., Azinheira, H., Talhinhas, P. and Silva, M.D.C., 2017. Legitimacy and implications of reducing Colletotrichum kahawae to subspecies in plant pathology. Frontiers in plant science, 7: 2051.

Bedimo, J.A.M., Bieysse, D., Njiayouom, I., Deumeni, J.P., Cilas, C. and Nottéghem, J.L., 2007. Effect of cultural practices on the development of arabica coffee berry disease, caused by Colletotrichum kahawae. European Journal of Plant Pathology, 119: 391.

Bedimo, J.M., Njiayouom, I., Bieysse, D., Nkeng, M.N., Cilas, C. and Nottéghem, J.L., 2008. Effect of shade on Arabica coffee berry disease development: toward an agroforestry system to reduce disease impact. Phytopathology, 98: 1320-1325.

Belachew, B., 1997. Arabica coffee breeding in Ethiopia: International scientific colloquium on coffee A review. A review. In: (ASIC’97). 22 - 25 July 1997, Nairobi, Kenya. 17: 406-414.

Belachew, B., 2001. Arabica coffee breeding for yield and resistance to coffee berry disease (Colletotrichum kahawae sp. Nov). Dissertation, Imperial College at Wye, University of London, pp. 232

Belachew, K. and Teferi D., 2015. Climatic variables and impact of coffee berry diseases (Colletotrichum Kahawae) in Ethiopian coffee production, 5: 55-64

Belachew, K., Teferi, D. and Gidisa, G., 2015. Screening of Some Coffea arabica genotypes against coffee wilt diseases (Gibberella xylarioides Heim and Saccus) at Jimma, Southwest Ethiopia. International Journal of Sustainable Agricultural Research, 2: 66-76.

Belachew, k., Demelash T., Niguse H., and Sisay T., 2016. The Statue and Management of coffee wilt disease (Gibberella xylarioides) in Ethiopian coffee production,6: 16-25.

Benti, T., 2017. Progress in Arabica coffee breeding in Ethiopia: Achievements, challenges and Prospects. Ethiopian Institute of Agricultural Research (EIAR), Jima Agricultural Research Center (JARC), Jimma, Ethiopia, 33:15-25.

Beynon, S.M., Coddington, A., Lewis, B.G. and Varzea, V., 1995. Genetic variation in the coffee berry disease pathogen, Colletotrichum kahawae. Physiological and molecular plant pathology, 46: 457-470.

Bieysse, D., Cilas C, Mouen J, Musoli P, Avelino J. 2009. Coffee disease risk analysis: how epidemiology knowledge could help in assessing and preventing disease invasion. In: The proceedings of 22nd International Conference on Coffee Science, ASIC 2008; Sep 14–19; Campinas, SP, Brazil; p. 1422–1423.

Biratu, T. and Hulluka M.,1989. Colletotrichum species associated with coffee berry disease in Hararghe. Ethiopian Journal of. Agricultural Science. 11: 1-6.

Biratu, T., 1995. Studies on Colletotrichum population of Coffea arabica L. in Ethiopia and evaluation of reactions of coffee germplasm. Ph. D thesis, Bonn University, Germany, pp. 231.

Bridge, P.D., Waller, J.M., Davies, D. and Buddies, A.G., 2008. Variability of Colletotrichum kahawae in relation to other Colletotrichum species from tropical perennial crops and the development of diagnostic techniques. Journal of Phytopathology, 156: 274-280.

Castiblanco, V., Castillo, H.E. and Miedaner, T., 2018. Candidate Genes for aggressiveness in a Natural Fusarium culmorum Population Greatly Differ between Wheat and Rye Head Blight. Journal of Fungi, 4: 14.

Cerda, R., Allinne, C., Gary, C., Tixier, P., Harvey, C.A., Krolczyk, L., Mathiot, C., Clément, E., Aubertot, J.N. and Avelino, J., 2017. Effects of shade, altitude and management on multiple ecosystem services in coffee agroecosystems. European Journal of Agronomy, 82: 308-319.

Chauhan, R., M. S. Hooda and Agena Anjulo Tanga, 2015. Coffee: the backbone of

Ethiopian economy, 1: 082-086.

Chen, C. and Dickman, M.B., 2005. Proline suppresses apoptosis in the fungal pathogen Colletotrichum trifolin. Proceedings of the National Academy of Sciences, 102:3459- 3464.

Chen, Z., Nunes M, Silva MC, Rodrigues CJ, 2004. Apresorium turgor pressure of Colletotrichum kahawae might have a role in coffee cuticle penetration. Mycology, 96: 1199–1208

Cristobel, M., A.L., de Jesus Yanez-Morales, M., Solano-Vidal, R., Segura-León, O. and Hernandez-Anguiano, A.M., 2017. Diversity of Colletotrichum species in coffee (Coffea arabica) plantations in Mexico. European journal of plant pathology, 147: 605-614.

CSA (Central Statistical Agency), 2017. Area and production of major crops in the Federal democratic republic of Ethiopia central statistical agency agricultural sample survey, 1: 1- 117.

Davis, A.P., Gole, T.W., Baena, S. and Moat, J., 2012. The impact of climate change on indigenous arabica coffee (Coffea arabica): predicting future trends and identifying priorities. PLoS One, 7: 47981.

Derso, E., 1997.Coffee diseases and their significance in Ethiopia. ASIC,17: 723-726.

Derso, E. and Waller, J.M., 2003. Variation among Colletotrichum isolates from diseased coffee berries in Ethiopia. Crop protection, 22:561-565.

Derso, E., Girma A. and Gebrezgi H., 2000. Control of coffee berry disease by fungicide in Ethiopia. In processing of the workshop on control coffee berry disease in Ethiopia, pp. 35-47.

Diniz, I.I.P.D.S., 2018. An innovative integrated approach to characterize coffee resistance mechanisms to Colletotrichum kahawae, pp. 1-84.

ECX (Ethiopian Commodity Exchange), 2008. Analysis of coffee supply, production, utilization and marketing issues and Challenges in Ethiopia, 61: 1.

Ejetta, M., 1986. A review of coffee diseases and their control in Ethiopia. In Tsedeke Abate (ed.) Proceedings of the first Ethiopian crop protection symposium, 4-7 February 1985, Addis Ababa, Ethiopia, Pp. 179-195.

Emana, B.T., 2015. Distribution assessment and pathogenicity test of coffee berry disease (Colletotrichum kahawae) in Hararghe, Ethiopia, 2: 038-042

Etana, M.B., 2018. Review on the Management of Coffee Berry Disease (Colletotrichum kahawae) in Ethiopia. Journal of Food Science and Quality Management, 76:73-76

FAO (Food and Agricultural Organization), 2015. Food and Agricultural Organization of the United Nation statistical pocketbook of coffee, pp. 1-194.

Fino, J.R.V., 2014. Analysis of RNA-seq data from the interaction of Coffea spp. Colletotrichum kahawae (Doctoral dissertation), pp. 1-74.

Fokunang, C. N., Dixon, A.G.O., Akem, C.N. and Ikotun, T., 2000. Cutural, morphological and pathogenic variability in Colletotrichum gloesporioides f. sp. from cassava (Mani hot esculent) in Nigeria. Pakistan Journal of Biological Sciences, 3: 542- 546.

Fredrick, M. N., Owaka M., Chrispine O., Elijah K., 2015. Pathogenicity of Colletotrichum kahawae in Kenya. International Journal of Science, 6: 2319-7064.

Garuma, H., Berecha, G. and Abedeta, C., 2015. Influence of Coffee Production Systems on the Occurrence of Coffee Beans Abnormality: Implication on Coffee Quality. Asian Journal of Plant Sciences, 14: 40-44.

Gibbs, J.N., 1969. Inoculum sources for coffee berry disease. Annals of Applied Biology, 64: 515-522.

Gichimu, B.M., Gichuru, E.K., Mamati, G.E. and Nyende, A.B., 2014. Occurrence of Ck- 1 gene conferring resistance to coffee berry disease in Coffea arabica cv.55 Ruiru 11 and its parental genotypes. Journal of Agricultural and Crop Research, 2:51-61.

Gichuru, E.K., Mwang’ombe, A.W. and Gupta, V.K., 1997. Peroxidase activity in Coffea arabica cultivars resistant and susceptible to coffee berry disease. Journal of African Crop Science. l 2: 223-232.

Gichuru, E.K., 2007. Sensitive response and resistance to berry disease (Colletotrichum kahawae) of two coffee varieties (Coffea arabica and C. canephora): histological comparisons of interactions. Agronomies Africanize, 19: 233-240.

Gidisa, G., 2016. A review on the status of coffee berry disease (Colletotrichum kahawae) in Ethiopia, .6:140-151.

Gomez, K. A. and Gomez, A. A., 1984. Statistical procedures for agricultural research, second edition John Wiley and Sons Inc., New York, pp. 680.

Grabs, J., 2017. "The rise of buyer-driven sustainability governance: Emerging trends in the global coffee sector. ZenTra Working Paper in Transnational Studies No. 73 / 2017. Available at SSRN: https://ssrn.com.

Griffiths, E., Gibbs, J.N. and Waller, J.M., 1991. Control of coffee berry disease. Annals of Applied Biology, 67: 45-74.

Gurmessa, F., Soromessa, T. and Kielbasa, E., 2012. Structure and regeneration status of

Komto Afromontane moist forest, East Wollega Zone, west Ethiopia. Journal of Forestry Research, 23:205-216.

Habte E., 2005. Evaluation of Wollega coffee germplasm for yield, yield component and resistant to coffee berry disease at early bearing stage. An MSc thesis submitted to school of graduate studies of Alemaya University, pp. 1-58.

Hindorf, H., 1970. Colletotrichum spp. isolated from Coffea arabica L. in Kenya for phytoplankton Journal of Plant Diseases and Protection, pp. 328-331.

Hindorf, H., Biratu, T. and Omondi, C.O., 1997. Correct identification of the pathogen/Colletotrichum kahawae/causing coffee berry disease (CBD) in Kenya Nairobi, 17: 599-603.

Hindorf, H. and Omondi, C.O., 2011. A review of three major fungal diseases of Coffea arabica L. in the rainforests of Ethiopia and progress in breeding for resistance in Kenya. Journal of advanced research, 2: 109-120.

Hoglund, S., Larsson S. and Wingsle, G., 2005. Both hypersensitive and non- hypersensitive responses are associated with resistance in Salix viminalis against the gall midge Dasineura marginemtorquens. Journal of Experimental Botany, 56: 3215-3222.

Holb, I.J., 2005. Effect of pruning on apple scab in organic apple production. Plant Disease, 89: 611-618.

Hyde, K.D., Cai, L., Cannon, P.F., Crouch, J.A., Crous, P.W., Dam, U., Goodwin, P.H., Chen, H., Johnston, P.R., Jones, E.B.G. and 'Liu, Z.Y., 2009. Colletotrichum names in current use. Fungal Diversity, 39:147-183.

ICO (International Coffee Organization), 2016. International coffee organization associated meetings for sustainable credit guarantee scheme and promotes scaling up of enhanced processing practices in Ethiopia and Rwanda (CFC/ICO/48), United Kingdom, pp. 1-46

ICO (International coffee organization), 2018. Coffee market report in the international trade, challenges and opportunities facing the sector, pp.1-8

International Trade Center (ITC), 2011.The Coffee Exporter's Guide, Third ed., Geneva, (Doctoral dissertation), pp. 267.

Jeffries, P., Dodd, J.C., Jeger, M.J. and Plumbley, R.A., 1990. The biology and control of Colletotrichum species on tropical fruit crops. Plant pathology, 39: 343-366.

Jha, S., Bacon, C.M., Philpott, S.M., Ernesto Mendez, V., Läderach, P. and Rice, R.A., 2014.Shade coffee: update on a disappearing refuge for biodiversity. Bioscience, 64: 416- 428.

Jirata and Assefa, 2000. Status of coffee berry disease in Oromia region. In: proceedings of the workshop on control of coffee berry Disease (CBD) in Ethiopia, August 13-15 1999, Addis Ababa, Ethiopia, pp. 9-17

Kagezi, H., Patrick, K., Judith, K., Nicholas, O.D., Lilian, N. and Wagoire, W., 2018. Predicting the Response of Insect Pests and Diseases of Arabica Coffee to Climate Change along an Altitudinal Gradient in Mt. Elgon Region, Uganda. Journal of Agriculture and Environmental Sciences, 7: 134-140.

Kamau, A.P.K., 2015. Characterization of coffee genotypes derived from crossing Rume Sudan and SL 28 coffee varieties against coffee berry disease (CBD) causal pathogen (Colletotrichum kahawae), pp.1-61.

Kebati, R.K., Nyangeri, J., Omondi, C.O. and Kubochi, J.M., 2016. Effect of artificial shading on severity of coffee berry disease in Kiambu County, Kenya. Annual Research & Review in Biology, 9: 1-11

Kilambo, D.L., 2008. Virulence of Colletotrichum kahawae strains and their effect on resistant Arabica coffee varieties in Tanzania (Doctoral dissertation, Sokoine University of Agriculture), pp,1-79.

Kilambo, D.L., Guerra-Guimaraes, L., Varzea, V., Loureiro, A., Silva, M., Mabagala, Mtenga R.B., D.J., Masumbuko, L. and Teri, J.M., 2009. Pathogenicity of Colletotrichum kahawae strains and their effect on resistant Arabica coffee varieties in Tanzania, pp. 1430-1438.

Kilambo, D.L., Silva, M., Matagalpa, R.B., Varzea, V.M.P. and Teri, J.M., 2013. Nature of Resistance of Lyamungu Hybrids to Colletotrichum kahawae Strains. International Journal of Biotechnology Applications, 5: 137.

Kiwanian, R., 2013. Essays on global coffee supply chains: improving small-scale producers' income (Doctoral dissertation, Rutgers University-Graduate School-Newark), pp. 1-157.

Labouisse, J.P., Bellachew, B., Kotecha, S. and Bertrand, B., 2008. Current status of coffee (Coffea arabica L.) genetic resources in Ethiopia: implications for conservation. Genetic Resources and Crop Evolution, 55: 1079.

Loureiro, A., Guerra-Guimarães, L., Lidon, F.C., Bertrand, B., Silva, M.C. and Várzea, V., 2011. Enzymatic characterization of Colletotrichum kahawae isolates with different levels of aggressiveness. Tropical Plant Pathology, 36:287-293.

Loureiro, A., Nicole, M.R., Varzea, V., Moncada, P., Bertrand, B. and Silva, M.C., 2012. Coffee resistance to Colletotrichum kahawae is associated with lignification’s, accumulation of phenols and cell death at infection sites. Physiological and molecular plant pathology,77:23-32.

Luzolo M, Talhinhas P, Várzea V, Neves-Martins J, 2010. Characterization of Colletotrichum Kahawae isolates causing coffee berry disease in Angola. Journal of Phytopathology,158:310–313.

Ma K-W, Ma W., 2016. Phytohormone pathways as targets of pathogens to facilitate infection. Plant Molecular Biology, 91: 713–725.

Manuel, L., Talhinhas, P., Várzea, V. and Neves‐Martins, J., 2010. Characterization of Colletotrichum kahawae isolates causing coffee berry disease in Angola. Journal of phytopathology, 158: 310-313.

Margaret, O., 2011. Characterization of the genetic diversity and pathogenicity of Colletotrichum kahawae using random amplified polymorphic DNA (RAPD) analysis (Doctoral dissertation, School of Pure and Applied Sciences, Kenyatta University), pp.1- 56. Masaba, D.M. and Vander Vossen, H.A.H., 1982. Evidence of cork barrier formation as a resistance mechanism to berry disease (Colletotrichum coffeanum) in Arabica coffee. European Journal of Plant Pathology, 88: 19-32.

McDonald, J.,1926. A preliminary account of a disease of green coffee berries in Kenya colony. Trans. Brazilian Mycological Society. 11:145-154.

Mehari, B., Redi-Abshiro, M., Chandravanshi, B.S., Combrinck, S. and McCrindle, R., 2016. Characterization of the cultivation region of Ethiopian coffee by elemental analysis. Analytical Letters, 49: 2474-2489.

Mohammed, A. and Jambo, A., 2015. Importance and characterization of coffee berry disease (Colletotrichum kahawae) in Borena and Guji Zones, Southern Ethiopia. Journal of Plant Pathology and Microbiology, 6: 1-6.

Mouen Bedimo, J.A., Bieysse, D., Nyassé, S., Nottéghem, J.L. and Cilas, C., 2010. Role of rainfall in the development of coffee berry disease in Coffea arabica caused by Colletotrichum kahawae, in Cameroon. Plant pathology, 59 :324-329.

Mtenga, D.J.I. and Reuben S.O.W.M., 2012. Variation in resistance to coffee berry disease (colletotrichum kahawae) among germplasm progenitors at Tanzania Coffee Research Institute (TCRI), 2:198-203.

Mtenga, D.J. and ReubenS.O.W.M., 2016. Diversity, combining ability and coffee berry disease (colletotrichum kahawae) resistance among Ethiopian and Tanzanian Arabica coffee genotypes (Doctoral dissertation) Sokoine University of Agriculture, pp. 1-76.

Mulatu, A., 2012. Characterization and Testing of Antifungal Extracts from Trichoderma isolates against Fusarium xylarioides, the Causative agent of Coffee Wilt disease, pp. 1-70.

Negash, T. and Abate, S.,2000. August. Status of CBD in SNNP. In proceedings of the

63 workshop on control of coffee berry Disease (CBD) in Ethiopia, pp. 18-28.

Netsre A. and Kufa T., 2015. Determining Suitable Shade Trees, Panting Pattern and Spacing for Arabica coffee production in South and South Western Ethiopia, 5:9-15.

Newton, A.C., 2016. Exploitation of Diversity within Crops—the Key to Disease Tolerance, Frontiers in plant science, 7: 665.

Nguyen, P., 2010. Colletotrichum spp. associated with anthracnose disease on Coffee in Vietnam and on some other major tropical crops, 39:1-58.

Nguyen Y., G. R. Tombeng Noumbo, E. Keute Kajdoum and C. G. Nchongboh, 2017. Antifungal Activities of Plant E, extracts against coffee berry disease caused by Colletotrichum kahawae L, 4: 60-66.

Omondi, 1998. Genetic diversity among isolates of Colletotrichum kahawae causing coffee berry disease and their interactions with varieties and breeding populations of Arabica coffee (Doctoral dissertation), pp. 1-138.

Omondi, C.O., Ayiecho, P.O., Mwang'ombe, A.W. and Hindorf, H., 2001. Reaction of some Coffea arabica genotypes to strains of Colletotrichum kahawae, the cause of coffee berry disease. Journal of Phytopathology, 148:61-63.

Owaka, M., 2011. Characterization of the genetic diversity and pathogenicity of Colletotrichum kahawae, using Random Amplified Polymorphic DNA (RAPD) analysis (doctoral dissertation school of pure and applied Science Kenyatta university, 7:225-263.

Photita, W., Taylor, P.W., Ford, R., Hyde, K.D. and Lumyong, S., 2005. Morphological and molecular characterization of Colletotrichum species from herbaceous plants in Thailand. Fungal Diversity, 18: 117-133.

Pinard, F., Omondi, C.O. and Cilas, C., 2012. Detached berries inoculation for characterization of coffee resistance to coffee berry disease. Journal of Plant Pathology, 94:517-523.

Prihastuti, H., CAI, L., Chen, H., McKenzie, E.H.C. and Hyde, K.D., 2009. Characterization of Colletotrichum species associated with coffee berries in northern Thailand. Fungal Diversity, 39:89-109.

Rakotoniriana, E.F., Scauflaire, J., Rabemanantsoa, C., Urveg-Ratsimamanga, S., Corbusier, A.M., Quetin-Leclercq, J., Declerck, S. and Munaut, F., 2013. Colletotrichum gigasporum sp. Nov., a new species of Colletotrichum producing long straight conidia. Mycological Progress, 12:403-412.

Rayner, R.W.,1970. A Mycological Color Chart: In Common Wealth Mycological Institute, Kew, Surrey, UK, pp. 20.

Robinson, R.A., 1976. Plant pathosystem. In Plant Pathosystem Springer, Berlin, Heidelberg, pp. 15-31. Rodrigues, Jr., C.J., Várzea, V.M.P., Hindorf, H. and Medeiros, E.F., 1991. Strains of Colletotrichum coffeanum Noack causing coffee berry disease in Angola and Malawi with characteristics different to the Kenya strain. Journal of Phytopathology, 131: 205-209.

Rodriguez, G.A.A., 2016. Mode of action of Phialomyces macrospores as a biocontrol agent in coffee infected with Colletotrichum spp., pp. 1-55.

Saikia, R., Singh, T., Kumar, R., Srivastava, J., Srivastava, A.K., Singh, K. and Arora, D.K., 2003. Role of salicylic acid in systemic resistance induced by Pseudomonas fluorescens against Fusarium oxysporum f. sp. Ciceri in Chickpea. Microbiological Research, 158: 203-213.

Sanabria, N.M., Huang, J.C. and Dubery, I.A., 2010. Self/non-self-perception in plants in innate immunity and defense. Self/nonself, 1:40-54.

Serani, S., Taligoola, H.K. and Hakiza, G.J., 2007. An investigation into Fusarium spp. associated with coffee and banana plants as potential pathogens of Robusta coffee. African journal of ecology, 45: 91-95.

Sillero, J.C., Rojas-Molina, M.M., Ávila, C.M. and Rubiales, D., 2012. Induction of systemic acquired resistance against rust, ascochyta blight and broomrape in faba bean by exogenous application of salicylic acid and benzothiadiazide. Crop Protection, 34: 65-69.

Silva, M.D.C., Várzea, V., Guerra-Guimarães, L., Azinheira, H.G., Fernandez, D., Petitot, A.S., Bertrand, B., Lashermes, P. and Nicole, M., 2006. Coffee resistance to the main diseases: leaf rust and coffee berry disease. Brazilian Journal of Plant Physiology, 18:119- 147.

Silva, D.N.P.R., 2010. Analysis of the neutral and adaptive genetic variation of Colletotrichum kahawae and its relationship with the C. gloesporioides complex (Doctoral dissertation), pp.1-88.

Silva, D.N., Talhinhas, P., Varzea V., Cai, L., Paulo O.S., Batista, D., 2012. Application of the Apn2 locus to improve the systematics of the Colletotrichum gloesporioides complex: from (Coffea species) hosts. Mycologia, 104:396-409.

Silva, M.C., Abranches, R., Loureiro, J. and Talhinhas, P., 2016. Cytogenetic characterization of Colletotrichum kahawae, the causal agent of coffee berry disease, reveals diversity in mini chromosome profiles and genome size expansion. Plant Pathology, 65:968-977.

Silva, D.D., Crous, P.W., Ades, P.K., Hyde, K.D. and Taylor, P.W., 2017. Life styles of Colletotrichum species and implications for plant biosecurity. Fungal Biology Reviews, 31: 155-168

Singh, V.K. and Upadhyay, R.S., 2013. The hypersensitive response: a case of cell death induction in plants. International Journal of Engineering, 2: 1-16.

Swami, C.S. and Alane, S.K., 2013. Efficacy of some botanicals against seed–borne fungi of green gram (Phaseolus aureus Roxb.). Bioscience Discovery, 4: 107-110.

Tadesse, G., Zavaleta, E., Shennan, C. and Fitzsimons, M., 2014. Local ecosystem service use and assessment vary with socio-ecological conditions: A case of native coffee-forests in southwestern Ethiopia. Human Ecology, 42:873-883.

Tadesse, W. G., 2014. Conservation and use of coffee genetic resources in Ethiopia: challenges and opportunities in the context current global situations, pp. 1-24.

Teferi, D. and Ayano A., 2017. Assessment of commercially grown coffee types for their tolerance to bacterial blight of coffee (BBC) at Sidama and Gedeo Zones, Southern Ethiopia. 6: 0.001-0.005

Teferi, D. and Belachew K., 2018. A review of coffee diseases research in Ethiopia. International Journal of Agriculture and Biosciences, 7: 65-70.

Temesgen, A. and Tufa. A., 2017. Analysis of Coffee Farm Productivity in Darolabu District, West Hararghe Zone, Ethiopia, 2: 158-161.

Temesgen, F. And Getachew, B., 2016. Analysis of technical efficiency of coffee production on small holder farmers in case of Sasiga & Limu Districts of East Wollega Zone, pp, 1-36.

Thomas, B. P. H. J., Nurnberg, T., & Jostens, M. H. A. J.,2011. Of Pathogen Association Molecular Patterns (PAMPs) and Effectors: The Blurred PAMP Triggered Immunity (PTI) Dichotomy in the. of the Plant Cell, 23:4–15.

USDA (United states Department of Agriculture), 2018. Coffee world markets and trade foreign agricultural service in USA, pp.1-9.

Vaast P, Bertrand B, Perriot JJ, Guyot B, Genard M. 2006. Fruit thinning and shade improve bean characteristics and beverage quality of coffee (Coffea arabica L.) under optimal conditions. Journal of Science and Food Agriculture, 86:197–204.

Van der Graff, 1981. Selection for Arabica coffee types resistant to coffee berry disease in Ethiopia. Mededelingen land Bouwhoge School, Wageningen, Nederland. Pp. 110.

Van der Graff, N. A., 1984. Resistance to coffee berry disease in Ethiopia, the CBD program from 1972 to 1979. In the Proceedings of the first regional workshop on coffee berry disease. 19-23 July 1982, Addis Ababa, Ethiopia, pp. 145-166.

Van der Vossen, H.A.M., Cook, R.T.A. and Murakaru, G.N.W., 1976. Breeding for resistance to coffee berry disease caused by Colletotrichum coffeanum Noack (Sensu

Hindorf) in Coffea arabica LI Methods of preselection for resistance. Euphytica, 25: 733- 745.

Van der Vossen, H.A.M. and Walyaro, D.J., 2009. Additional evidence for oligo genic inheritance of durable host resistance to coffee berry disease (Colletotrichum kahawae) in arabica coffee (Coffea arabica L.). Euphytica, 165: 105.

Van der Vossen, H., Bertrand, B. and Charrier, A., 2015. Next generation variety development for sustainable production of Arabica coffee (Coffea arabica L.): a review. Euphytica, 204: 243-256.

Vieira, A., Diniz, I., Loureiro, A., Pereira, A.P., Silva, M.C., Várzea, V. and Batista, D., 2018.Aggressiveness profiling of the coffee pathogen Colletotrichum kahawae. Journal of Plant Pathology, pp.1-32

Vossen, H.A.V.D., 2001. Agronomy and coffee breeding practices. Coffee: recent developments, pp.184-201.

Waller, J.W., M. Bigger and R.J. hillocks, 2007. Coffee pests, disease and their managements, pp.211-215.

Waller, S.A., Simpkins, S.A., Stead, D.E., Kurdziel, A., Hird, H. and Weekes, R.J., 2002. Identification of Agrobacterium spp. present within Brassica napus seed by TaqMan PCR implications for GM screening procedures. Archives of microbiology, 178: 338-343.

Waller, J.M., Bridge, P.D., Black, R. and Hakiza, G., 1993. Characterization of the coffee berry disease pathogen, Colletotrichum kahawae sp. Nov. Mycological Research, 97: 989- 994.

Weir, B.S., Johnston, P.R. and Dam, U., 2012. The Colletotrichum gloesporioides species complex. Studies in mycology, 73:115-180.

Workafes, W., Kassu K. 2000. Coffee production systems in Ethiopia. In: Proceedings of the workshop on control of coffee berry disease in Ethiopia; 1999 Aug 13–15, Addis Ababa; p. 99–107.

Yadessa, G., 2014. Genetic Diversity, Pollination Ecology and Organoleptic characteristics of Coffea arabica L. in Ethiopian moist forests of different management Intensity. (Doctoral dissertation in the degree of Doctor Biology), pp. 1-200

Yilma, T. and Kufa, T., 2002. Characterization of the farming systems of Haru woreda west Wollega zone, Oromia region. Research Report (Ethiopia), 44: 38.

Zakaria, M. and Bailey, J.A., 2000. Morphology and cultural variation among Colletotrichum isolates obtained from tropical forest nurseries. Journal of Tropical Forest Science, pp.1-20.

Zeru, A., 2006. Diversity of Arabica coffee populations in Afromontane rainforests of Ethiopia in relation to Colletotrichum kahawae and Gibberella xylarioides, pp.1-68.

Zeru, A., Teferi, D., Jafuka, C., Tesfaye, S., Siyoum, M. and Adugna, G., 2008. Success stories in managing coffee berry disease in Ethiopia. Coffee diversity and knowledge, EIAR, Addis Ababa, Ethiopia, pp. 239-249.

Zeru, A., Asefa F, Adugna G, Hindorf H., 2009. Occurrence of fungal diseases of Coffea arabica. In montane rainforests of Ethiopia. Journal of Application Botany and Food Quality. 28: 148–151.

Zeru, A., Assefa, F., Adugna, G. and Hindorf, H., 2012. Occurrence of fungal diseases of Coffea arabica L. in montane rain forests of Ethiopia. Journal of Applied Botany and Food Quality,82: 148-151.

7. APPENDICES

Appendix Table 1. List of C. arabica accessions with mean CBD score under visual assessment at field conditions Sr mean Sr. Mean Sr Mean no Accessions score no Accessions score no Accessions score 1 G-63/13 0.00 35 G-47/13 7.55 68 G-12/13 32.75

2 G-65/13 0.04 36 G-89/13 7.63 69 G-38/13 33.04

3 G-85/13 0.07 37 G-28/13 7.63 70 W-92/98 37.80

4 G-87/13 0.07 38 G-88/13 8.25 71 G-60/13 38.17

5 7416' 0.09 39 G-75/13 8.60 72 G-32/13 41.34

6 G-10/13 0.25 40 G-57/13 8.75 73 G-25/13 42.29

7 G-91/131 0.35 41 G-54/13 9.25 74 G-45/13 43.42

8 W-76/98 0.50 42 G-34/13 10.84 75 G-22/13 45.00

9 G-16/13 0.55 43 G-48/13 11.00 76 G-20/13 46.25

10 G-68/13 0.86 44 G-40/13 11.43 77 G-39/13 46.25

11 W-78/84 1.00 45 G-90/13 11.55 78 G-41/13 46.88

12 G-77/13 1.53 46 G-21/13 12.50 79 G-79/13 50.00

13 G-67/131 1.75 47 G-08/13 12.75 80 G-43/13 55.84

14 G-72/13 2.60 48 W-66/98 12.90 81 G-30/13 56.67

15 G-76/13 2.67 49 G-27/13 12.94 82 G-80/13 58.13

16 G-15/13 2.75 50 G-73/13 13.13 83 G-44/13 65.00

17 G-51/13 2.80 51 G-53/13 13.17 84 G-14/13 65.84

18 G-37/13 2.88 52 G-19/13 13.50 85 G-35/13 66.04

19 G-82/13 3.00 53 G-52/13 13.52 86 G-83/13 68.34

20 G-66/13 3.48 54 G-62/13 15.29 87 G-06/13 70.63

21 G-49/13 3.80 55 G-92/13 15.54 88 G-84/13 72.50

22 G-74/13 3.85 56 8136' 16.25 89 G-05/13 73.75

23 G-55/13 4.00 57 G-17/13 16.82 90 G-02/13 75.00

24 G-64/13 4.00 58 G-29/13 17.32 91 G-03/13 76.88

25 G-69/13 4.07 59 G-11/13 19.14 92 G-01/13 78.75

26 G-71/13 4.19 60 G-23/13 21.80 93 G-78/13 80.63

27 G-13/13 5.21 61 G-36/13 22.87 94 G-61/13 83.75

28 G-50/13 6.04 62 7514' 23.00 95 G-09/13 85.38

29 G-31/13 6.38 63 G-26/13 26.71 96 G-07/13 85.63

Appendix Table 1. List of C. arabica accessions with mean CBD score under visual assessment at field conditions ...... Continued

Sr Mean Sr. Mean Sr Mean no Accessions score no Accessions score no Accessions score 30 G-70/13 6.50 64 G-86/13 27.76 97 G-81/13 90.00 31 G-56/13 6.50 65 G-46/13 28.57 98 G-04/13 90.00 32 7576' 7.35 66 G-18/13 29.15 99 G-33/13 97.50 33 G-59/13 7.50 67 G-42/13 31.57 100 G-24/13 97.50 34 G-58/13 7.50 Mean 28.31 CV (%) 31.4

Appendix Table 2. ANOVA table for coffee berry disease incidence in nine kebeles of Gidame district

Source DF SS MS F- value P-value Rep 4 1168.888 292.222 2.53 0.067 Trt 8 2200.000 275.000 2.38 0.04 Error 24 2768.657 115.361 Total 44 8280.000

Appendix Table 3. ANOVA table for coffee berry disease incidence in nine kebeles of Gidame district

Source DF SS MS F Value P- value Rep 4 458.931 114.733 1.29 0.301 Trt 8 3361.288 420.161 4.72 0.001 Error 24 2134.215 88.926 Total 44 7594.424

Appendix Table 4. ANOVA table for cultural growth rates of Colletotrichum Kahawae collected across different kebeles of Gidame district

Source DF SS MS F-value P-value

Replication 2 0.8534 0.4226 4.48 0.020

Isolates 14 16.3541 1.1681 12.25 0.0001

Error 28 2.6710 0.0955

Total 44 23.4372

Appendix Table 5. ANOVA table for conidial size of Colletotrichum kahawae isolates (spore length) Source DF SS MS F-value P-value Replication 2 0.0274 0.0137 0.65 0.52 Isolates 14 11.8420 0.8458 40.32 0.0001 Error 28 0.8573 0.0209 Total 44 12.4568

Appendix Table 6. ANOVA for conidial size of Colletotrichum kahawae isolates (width)

Source DF SS MS F-value P-value Replication 2 0.0016 0.0008 0.27 0.7676

Isolates 14 2.2942 0.1638 52.78 0.0001

Error 28 0.0869 0.0031

Total 44 2.3828

Appendix Table 7. ANOVA table of sporulation capacity of Colletotrichum kahawae isolates Source DF SS MS F-value P-value Replication 2 3160.8821 1580.4410 0.97 0.3926 Isolates 14 45382.6446 32813.0460 20.07 0.0001 Error 28 45766.7729 1634.5276 Total 44 508310.2996

Appendix Table 8. ANOVA for isolates on virulence determination in detached berry test Source DF SS MS F-value P-value Replication 2 2403.5640 1201.7820 4.58 0.0136

Isolates 15 112028.5169 3200.8148 12.19 0.0001

Error 30 18382.1283 262.6018

Total 47 132814.2091

Appendix Table 9. ANOVA table for resistance evaluation of Coffea arabica accessions against coffee berry disease with attached berry test Source DF SS MS F Value P- value Replication 2 678.052 339.026 4.48 0.0148 Accessions 35 36631.2 1046.6 13.83 .0001 Error 70 5299.23 75.7034 Total 107 42608.5

Appendix Table 10. ANOVA table for resistance evaluation of Coffea arabica accessions against coffee berry disease under laboratory conditions detached berry test Source DF SS MS F-value P-value Replication 2 670.8545 335.4272 4.43 0.0155 Isolates 35 38753.6397 1107.2472 14.61 0.0001 Error 70 5303.9756 75.7710 Total 107 44728.4678

Appendix 11. Survey questionnaires for assessment of coffee berry disease at Gidame district of Western Wollega zone

Woreda ______Kebele______

Altitude______latitude ______

Owners (farmers) name ______

Age of the plant______

Production system (forest, semi forest, garden or plantation)

The cultivars used (local, improved or mixed)

Shade type

Management practices (slashing, ploughing, chemical spray………………)

Field data recording sheet

Severity (with 0-6 scale of berries from CBD incidence Tree no selected strata In tree based Remark

Top Middle Bottom Total Mean 1