REPRODUCTION, SEX RATIO AND BACTERIAL COMMUNITIES OF THE COFFEE BERRY BORER Hypothenemus hampei F. (COLEOPTERA: CURCULIONIDAE)

Transcription

1 REPRODUCTION, SEX RATIO AND BACTERIAL COMMUNITIES OF THE COFFEE BERRY BORER Hypothenemus hampei F. (COLEOPTERA: CURCULIONIDAE) by Yobana Andrea Mariño Cárdenas A thesis submitted to the DEPARTMENT OF BIOLOGY FACULTY OF NATURAL SCIENCES UNIVERSITY OF PUERTO RICO RIO PIEDRAS CAMPUS In partial fulfillment of the requirements for the degree of DOCTOR IN PHILOSOPHY May 2015 San Juan, Puerto Rico Yobana Andrea Mariño-Cárdenas All rights reserved

2 ii

3 This thesis have been accepted by the faculty of the DEPARTMENT OF BIOLOGY FACULTY OF NATURAL SCIENCES UNIVERSITY OF PUERTO RICO RIO PIEDRAS CAMPUS In partial fulfillment of the requirements for the degree of DOCTOR IN PHILOSOPHY iii

4 TABLE OF CONTENTS LIST OF TABLES... vii LIST OF FIGURES... ix ABSTRACT... xi DEDICATION... xiv ACKNOWLEDMENTS... xv CHAPTER 1. GENERAL INTRODUCTION: Factors that affect the reproduction, sex ratio and bacterial communities in insects in general... 1 Introduction Factors that influence sex ratio and reproduction of insects Biotic factors Infection with endosymbiotic bacteria Population density and density of conspecifics Food resources (quality and amount) Abiotic factors Temperature Relative humidity Factors that shape bacterial communities in insects Biotic factors Food resources or diet Stage of development of insect hosts Abiotic factors Temperature Coffee and Coffee Berry Borers: an ideal system to study the effect of biotic and abiotic factors on reproduction, sex ratio and bacterial communities of insects The coffee berry borer (H. hampei) Dissertation thesis structure References CHAPTER 2: Sun vs. shade affects infestation, population sizes and sex ratio of the coffee berry borer (Hypothenemus hampei) in Puerto Rico Abstract Introduction Materials and Methods Study region and site descriptions Sampling methods Statistical analysis Results Environmental characteristics of plots Differences between sun vs. shade on infestation of CBB Differences between sun vs. shade on number of CBB per fruit Environmental variables and relationship to number of CBB per bored fruit Sex ratio Differences between sun vs. shade on the penetration positions of colonizing females of H. hampei Differences between sun vs. shade on trap capture rates Discussion CBB infestation levels compared to other coffee-growing countries Coffee fruit development and its association to CBB infestation Differences between sun vs. shade coffee on CBB infestation Coffee fruit development and number of CBB per fruit iv

5 4.5. Differences between sun vs. shade on number of CBB per fruit Differences in CBB infestation and population sizes between years Differences between sun vs. shade on sex ratio Use of traps with ethanol: methanol to monitor CBB populations Conclusions and recommendations Acknowledgments References Appendices CHAPTER 3: Wolbachia affects reproduction, sex ratio and population dynamics of the Coffee Berry Borer Hypothenemus hampei (Coleoptera: Curculionidae) Abstract Introduction Materials and Methods Insect Rearing Antibiotic Treatments Detection of Wolbachia DNA Extraction, PCR Amplification and DNA sequencing Phylogenetic classification in Wolbachia supergroups Effect of antibiotic treatments of CBB Reproduction and Sex Ratio Life Tables of H. hampei Statistical analysis Results Wolbachia detection and classification Effect of antibiotic treatment on Wolbachia infection and CBB reproduction Effect of antibiotic treatment of life table parameters of H. hampei Discussion Wolbachia in insects and status in the CBB Phylogenetic classification of Wolbachia supergroups Antibiotic treatments and their effect on CBB reproduction and fitness Conclusions Acknowledgments References Appendices CHAPTER 4: The bacterial microbiota of the coffee berry borer Hypothenemus hampei is influenced by host diet, development and environment...95 Abstract Introduction Materials and Methods Collection and laboratory rearing of coffee berry borers DNA extraction and sequencing Data processing Data analysis Alpha diversity Beta diversity Results General composition of the CBB microbiota and 16S rrna sequencing Antibiotic, environmental temperatures and host stage of development affect the frequency of Wolbachia Alpha diversity Beta diversity v

6 The composition of bacterial communities differs between field and laboratorycollected CBBs The composition of bacterial communities associated with CBBs differ between sun and shade Discussion Composition of bacterial communities and their possible role in CBB biology Antibiotic, environmental temperatures and host s stage of development affect the Wolbachia proportions Alpha and Beta diversity Conclusions Acknowledgments References Appendices CHAPTER 5: GENERAL CONCLUSIONS AND RECOMENDATIONS vi

7 LIST OF TABLES CHAPTER 2: Sun vs. shade affects infestation, reproduction and sex ratio of the coffee berry borer (Hypothenemus hampei) in Puerto Rico. Table 1. Coordinates and environmental characteristics of the study plots Table 2. Infestation and population densities of the coffee berry borer Hypothenemus hampei varies between sun vs. shade coffee, years and farms Table 3. Total number of females and males, percentage of fruits without males and sex ratio (mean ±se) of sex ratio of the coffee berry borer Hypothenemus hampei between sun vs. shade coffee, years and farms Table A.1. Species and family of trees associated with shade plots Table A.2. Results of generalized linear model evaluating the relationship between infestation of the coffee berry borer Hypothenemus hampei and type of coffee (sun vs. shade), year and farm Table A.3. Results of generalized linear model evaluating the relationship between the number of coffee berry borers per fruit and sun vs. shade, Year and Farm CHAPTER 3: Wolbachia affects reproduction, sex ratio and population dynamics of the Coffee Berry Borer Hypothenemus hampei (Coleoptera: Curculionidae). Table 1. Reference sequences of Wolbachia used to determine CBB Wolbachia supergroup classification Table 2. Number of females, males and sex ratio of Hypothenemus hampei reared in artificial diets with and without antibiotics; Mean ± SE and F:M ratio per vial for F1, F5 and F10 generations Table 3. Developmental time in days (Mean±SE) for life stages and life table parameters of Hypothenemus hampei reared in artificial diet with and without antibiotics Table A.1. Ingredients, amount and providers in the artificial diet to rear the coffee berry borer (H. hampei) Table A.2. Results of generalized linear model evaluating the relationship between CBB total population and type of diet (Control, Penicillin 0.1% and Tetracycline 0.1%) and generations Table A.3. Results of generalized linear model evaluating the relationship between CBB stages of development and type of diet (Control, Penicillin 0.1% and Tetracycline 0.1%) Table A.4. Stage transition frequencies and cumulative number of dead individuals in each stage of development of CBB in diet without antibiotics or control diet.92 vii

8 Table A.5. Stage transition frequencies and cumulative number of dead individuals in each stage of development of CBB in diet with penicillin 0.1% (w/v).93 Table A.6. Stage transition frequencies and cumulative number of dead individuals in each stage of development of CBB in diet with tetracycline 0.1%...94 CHAPTER 4: The bacterial microbiota of the coffee berry borer Hypothenemus hampei is influenced by host diet, development and environment. Table 1. Frequencies of the most abundant OTUs in the CBB Hypothenemus hampei. 120 Table 2. Alpha diversity indexes for each treatment of Hypothenemus hampei (Mean ± S.E. are shown) Table A.1. Ingredients, amount and providers in the artificial diet to rear the coffee berry borer (H. hampei) Table A.2. Predominant denovo OTUs assigned to family Pseudomonadaceae and BLASTed to compare and identify these sequences in GenBank viii

9 LIST OF FIGURES CHAPTER 1. GENERAL INTRODUCTION: Factors that affect the reproduction, sex ratio and bacterial communities in insects in general. Figure 1. Factors that affect population densities, sex ratio and bacterial communities of the coffee berry borer (Hypothenemus hampei) CHAPTER 2: Sun vs. shade affects infestation, population sizes and sex ratio of the coffee berry borer (Hypothenemus hampei) in Puerto Rico. Figure 1. Monthly changes in infestation (A and B) and number of CBB per fruit (C and D) of the coffee berry borer (Hypothenemus hampei F.) in shade vs. sun coffee. Means ± SE are shown Figure 2.Population structure of the coffee berry borer (Hypothenemus hampei F.). Means + SE are shown Figure 3. Proportion of the coffee berry borer (Hypothenemus hampei F.) colonizing females in positions A, B, C and D inside fruits during the growing season, and relation to sun vs. shade coffee and year Figure 4.Monthly averages of captured females of Hypothenemus hampei in traps in sun vs. shade coffee. Stages of coffee phenology in Puerto Rico are shown above the bars Figure 5.Total monthly precipitation in 2010 and 2011 for Adjuntas, Puerto Rico with the historical monthly average ( ) (Source: SERCC, Historical Climate Summaries for Puerto Rico and U.S. Virgin Islands) Figure A.1. Graphical representation for a coffee farm in Adjuntas, Puerto Rico. F1 in this study (A) Profile for shade, (B) Aerial view for shade (C) Profile for sun (D) Aerial view for sun 62 CHAPTER 3: Wolbachia affects reproduction, sex ratio and population dynamics of the Coffee Berry Borer Hypothenemus hampei (Coleoptera: Curculionidae). Figure 1.Basic life cycle of the coffee berry borer Hypothenemus hampei modeled as a population projection matrix Figure 2. Molecular phylogenetic analysis of the Wolbachia wsp sequences identified from Hypothenemus hampei with sequences of Wolbachia supergroups A and B Figure 3.Effects of antibiotics on Hypothenemus hampei populations (including eggs, larvae, pupae, juveniles and adults) for F1, F3, F5 and F10 generations. Means ± S.E. are shown Figure 4.Effects of antibiotics on the number of individuals (mean ± SE) per female for each stage of Hypothenemus hampei at F1, F3, F5 and F10 generations ix

10 Figure 5. Effects of antibiotics after five generations of treatment on Hypothenemus hampei. (A) Fecundity (number of eggs per female) and (B) % females ovipositing CHAPTER 4: The bacterial microbiota of the coffee berry borer Hypothenemus hampei is influenced by host diet, development and environment. Figure 1.Bacterial phyla in Hypothenemus hampei microbiota comparing eggs vs. adult females, Field samples vs. laboratory-reared on artificial diets Figure 2. Rarefaction curves for predicted OTUs of bacterial communities in Hypothenemus hampei Figure 3. Principal coordinate analyses for bacterial communities from Hypothenemus hampei Figure 4.Principal coordinate analyses for Hypothenemus hampei from field samples sun vs. shade x

11 ABSTRACT The coffee berry borer (CBB) Hypothenemus hampei is the world's most devastating coffee pest. Knowledge of the factors that affect its biology and ecology is an important tool to give effective management recommendations for this pest. Coffee plant management (e.g. sun vs. shade) can provide environments with contrasting abiotic conditions like temperature and relative humidity, which affect directly biotic factors such coffee quality and quantity, natural enemies and endosymbiotic bacteria. In this dissertation I combined field and laboratory experiments to determine the influence of abiotic factors as temperature and relative humidity and biotic factors such as infection with endosymbiotic bacteria (Wolbachia) on sex ratio, reproduction, population densities, diversity and composition of microbial communities of the CBB. Field experiments were conducted on three farms in the coffee-growing region of Adjuntas, Puerto Rico. In each farm two plots were chosen: sun coffee, in which plants are exposed to direct sunlight, and shade coffee, in which plants grow under the shade provided by up to nine tree species. I determined CBB infestation, population densities and sex ratio for the coffee-growing seasons of 2010 and 2011 (Chapter 1). In laboratory experiments I tested the effect of Wolbachia on population densities, sex ratio and population dynamics (Chapter 2), and I determined the effect of host stage of development, antibiotics and environmental temperature on bacterial communities of the CBB, with emphasis on Wolbachia (Chapter 3). Overall I found significant differences in temperature and relative humidity between sun and shade plots, and a significant effect of these abiotic factors on CBB infestation, which was higher in shade plots. The temperature appeared to be an important driver of CBB population densities and sex ratio: I found positive correlations between temperature and the xi

12 number of pupae, females and males. Larger populations and more biased sex ratios (towards females) were observed in sun plots which were characterized by higher temperatures. In addition to the lower populations in shade plots, I observed that shade over coffee also enhanced the presence of natural enemies such as ants and the entomopathogenic fungus Beauveria bassiana. Moreover, we found that the treatment with antibiotic tetracycline at 0.1% (w/v) and the exposition to higher temperatures caused a significant reduction in Wolbachia in adult females fed with the antibiotic and those collected from infested fruits growing in coffee sun plots, in both cases the proportion was reduced to 0.04%. The treatment with antibiotic significantly affected the fecundity of CBB females, which produced fewer eggs and lower population densities; the sex ratio was less skewed towards the females in the offspring from females that were fed with tetracycline; also, populations from females treated with tetracycline had reduced population growth rate (λ) and longer generation time (T). These results suggest that Wolbachia have positive effects on CBB fecundity and may be necessary to assure its normal and successful reproduction. Finally, I found that diet influenced structure and diversity of the bacterial community of the CBB. Microbiota of individuals reared on artificial diet was significantly different from those collected in coffee fruits, and samples from the field were more diverse. Antibiotic treatment reduced the diversity in samples from adult females but not in eggs. Environmental conditions do not affect the overall diversity; however, communities from eggs and adult females collected in fruits from sun plots were significantly different from those collected in fruits from shade plots. Higher temperatures observed in sun plots apparently affected endosymbiotic bacteria like Wolbachia. xii

13 In conclusion, reproduction and sex ratio of the CBB are affected by endosymbiotic bacteria and temperature. The combination of field and laboratory experiments showed a possible interaction between temperature and endosymbiotic bacteria, and higher temperatures reduced significantly the proportion of Wolbachia. The identification of the factors that contribute the reproductive success of the CBB and favor the presence of more females is necessary to develop more efficient management strategies for this important coffee pest. xiii

14 DEDICATION A mi más preciado tesoro mi mamá Luz Alba, a la memoria de mi padre Marco Antonio, a mi esposo Luis Antonio, a mis hermanos Patricia, Marina, William, Alicia y Claudia, a mi cuñados Guillermo, Nelson y Gregorio, a mis sobrinos y bisobrinos, por confiar en mi y ser la fuerza que me motiva a seguir adelante. xiv

15 ACKNOWLEDMENTS I am extremely grateful to my adviser Dr. Paul Bayman for his interest in my research topic, guidance, support and invaluable help with the English; I consider Paul more that my adviser a special friend. Also, I am extremely grateful to Dr. Jose Carlos Verle Rodrigues, for the support of my thesis work with research assistantships and for opening the doors of the facility of the Center for Excellence in Quarantine & Invasive Species for my laboratory experiments. Special thanks to Dr. Fernando Gallardo for supporting me with a fellowship, which was very important to cover field trips, and Dr. Elvira Cuevas and the NSF CREST- CATEC program for financial support of undergraduate students and assistance to present my research in several symposia. Finally, I am also grateful to Dr. Alberto Sabat for his guidance and advice with the analysis of population dynamic work and for helping me to obtain a thesis fellowship. Thanks to my family in Colombia they are the motor of my life, thanks for your help, love and to believe in me, for be proud of me, especially to my Mom Luz Alba for all her sacrifices and hard work for support our college. Thanks to my husband Tony for his love, patience, and specially for share his life with me and to be the support of my emotional and mental peace, to my fathers in love Mery and Ramon for their love and to care of us. To my kids of four paws for fill our home with love and life. To my friends Ana Maria and Victor that I have known for twenty years and I consider my brothers, for their unconditional friendship and emotional support when I needed it. To my newer friends who I appreciate so much: Eduar, Paola, Josean, Samir, Oscar, Fabiola, Patricia and Jose Miguel, thanks for everything and our get-together nights with a good cup of coffee. Thanks to Miguel Urdaneta and Gladys Ramos for their friendship and help during the xv

16 time I was a teaching assistant in microbiology, and special to Gladysita for always being worried about my Frida and for remembering all my birthdays. I am grateful to the growers (Melín Rullán, Anibal and Niulka, and others) for hospitality and permission to conduct experiments on their farms. I also thank Victor Vega, Fabiola Areces, Luis Ramírez, Marella Trifilio, Michelle Cruz, Ana Pamela Torres, Maria Dávila and Virmarie Díaz for their participation in fieldwork. We thank the personnel of the UPR Agricultural Experimental Substation in Adjuntas, especially Wigmar González and Alvaro Serrano. Sincere thanks to Dr. Maria-Eglée Pérez for her time, knowledge and patience for the analysis of the field data; also Dr. Carla Restrepo for her advice and for helping me to focus my thesis proposal and Dr. Maria Gloria Dominguez for the permission to use the computers of her lab to conduct analysis with QIIME. Special thanks to Michelle Cruz for her outstanding work in the laboratory; to Oscar Ospina for his invaluable help in the microbiota analysis and Dr. Filipa Godoy for her advice on data processing with QIIME. Thanks to staff of the facility of Sequencing and Genotyping at UPR, especially to MSc. Silvia Planas for her patience, dedication and careful handling of my samples. Many people were involved in the development of this thesis, and I apologize if I forgot some names. I also thank the University of Puerto Rico, especially the Department of Biology, the Graduate Program and Dean for Graduate Studies and Research (DEGI) for financial support to present my research in several symposia, and DEGI for supporting me with a Dissertation fellowship. Finally thanks to USDA-ARS and USDA-APHIS-PPQ cooperative agreements for the financial support. xvi

17 CHAPTER 1 Factors that affect the reproduction, sex ratio and bacterial communities in insects in general GENERAL INTRODUCTION 1

18 Introduction Reproduction, sex ratio and bacterial communities in insects are influenced by biotic factors and abiotic factors. Reproduction and sex ratio can be affected by biotic factors as food resources (quality and amount), population sizes, infection with endosymbiotic bacteria, and abiotic factors as temperature and relative humidity. Bacterial communities can be shaped according to food resources or diet, host stage of development and host environment. In some instances these factors may operate independently of each other, but in others they may interact (Figure 1). Reproductive performance of insects can be evaluated by assessing their egg load or number of mature oocytes in the ovaries, fecundity or number of offspring produced by an individual female and fertility or number of viable progeny (e.g. number of eggs that hatch to larvae) (Minkenberg et al., 1992; Awmack and Leather, 2002; Jordão et al., 2010). Sex ratio is defined as the proportion of females versus males in a population (Beukeboom, 2005). In the case of insects the primary signal for determining the sex of an individual is given by genetic factors, mainly sex determination systems. In species with sexual reproduction the expected sex ratio is 1:1 (female: male) (Traut, 1999; Schowalter, 2006); however, there are several species where this is not the case and there are more females that males in a population (Kirkendall et al., 1993; Kageyama et al., 1998; Vega et al., 2002; King and D'Souza, 2004). Differences in ploidy (e.g. diploid vs. haplodiploid organisms) have been associated with sex ratio distortion towards the females. The great majority of insect species have diploid systems, which are considered ancestral; however, there are other systems that have evolved from diploidy, for example, haplodiploid and functional haplodiploid or paternal genome elimination systems (Ross et al., 2010). 2

19 In diploid sex determination systems, sex is determined by the assortment of sex chromosomes, and if there is no effect of other factors like extrachromosomal infections, the sex ratio results in 1:1 female and male offspring (Hake and O'Connor, 2008). In haplodiploid systems, females develop from fertilized eggs and are diploids and males from unfertilized eggs and are haploid. Under this sex-determination system the progeny generally has highly female-biased sex ratios, which is product of parental influence, because this sex determination system provides the females physiological mechanisms to control the fertilization of their eggs; unfertilized eggs are less viable (Charnov et al., 1981; Werren and Beukeboom, 1998). True haplodiploidy is found in almost all species in the orders Hymenoptera (including honey, ants, and wasps), Thysanoptera (thrips) and some species of Heteroptera (bugs) and Coleoptera (beetles). In functional haplodiploidy or paternal genome elimination (PGE), both sexes develop from fertilized eggs, but paternally derived chromosomes are not transmitted to males because they are eliminated from their germ lines at different developmental stages (Werren and Beukeboom, 1998; Kuijper and Pen, 2010). Thus equal to true haplodiploidy females are diploids and male haploids. Functional haplodiploidy also leads to progenies with highly female-biased sex ratios, as a consequence of the high male mortality, which occurs due to the loss of paternal chromosomes (Engelstädter and Telschow, 2009). 1. Factors that influence sex ratio and reproduction of insects 1.1. Biotic factors Infection with endosymbiotic bacteria Extrachromosomal factors can affect the sex ratio and reproduction of insects. The term is used to refer to all heritable infections (bacteria, viruses and protozoans) not transmitted via chromosomes (Ebbert and Wrensch, 1993). The best-known of these is the infection with endosymbiotic bacteria that are maternally inherited via cytoplasm. It is 3

20 assumed both sexes are infected; however, the sperm do not carry enough cytoplasm to transmit the infection. For this reason these bacteria have evolved various mechanisms for manipulate the reproduction of their hosts, in order to increase the number of females in the offspring to enhance their own transmission. Reproductive manipulations include: 1) cytoplasmic incompatibility, 2) Parthenogenesis, 3) Feminization and 4) Male killing (Hurst and Jiggins, 2000; Weeks and Breeuwer, 2003; Zchori-Fein and Perlman, 2004; Werren et al., 2008; Cordaux et al., 2011). Wolbachia is the most widely studied group of endosymbiotic bacteria associated with biased sex ratios and has been demonstrated to induce all the mechanisms mentioned above (Werren, 1997; Stouthamer et al., 1999; Weeks et al., 2002; Werren et al., 2008). However, there are other, less-known bacteria that have the ability to induce some of these mechanisms For example, Cardinium causes cytoplasmic incompatibility, feminization and parthenogenesis (Zchori-Fein et al., 2001; Hunter et al., 2003; Weeks et al., 2003), Francisella can cause cytoplasmic incompatibility (Niebylski et al., 1997) and Rickettsia are known as male killers (Braig et al., 2008). At the same time, double infections have been observed with Cardinium and Wolbachia in several insect species (Hurst and Jiggins, 2000; Weeks et al., 2003) or Wolbachia and Rickettsia in members of family Curculionidae (beetles) (Zchori Fein et al., 2006). All reproductive manipulations affect the sex ratio. However, cytoplasmic incompatibility and male death also can affect reproduction, reducing significantly the number of offspring (Brelsfoard and Dobson, 2009). Wolbachia has been argued as one of the most important factors that influence the reproduction of insects (Werren, 1997; Stouthamer et al., 1999). In some cases Wolbachia have positive effects on host fecundity (Fry et al., 2004; Dong et al., 2006; Son et al., 2008; Jia et al., 2009) and fertility (Zchori Fein et al., 2006; Chen et al., 2012); they may be necessary to assure the normal reproduction of the 4

21 insects. For example, Wolbachia is necessary for the oogenesis of some wasps and beetles (Dedeine et al., 2001; Zchori Fein et al., 2006) Population density and density of conspecifics The effect of density of conspecifics in sex ratio can be explained through the local mate competition theory (LMC), which was designed to explain female-biased sex ratio in haplodiploids organisms (Charnov et al., 1981; Kirkendall, 1993; King and D'Souza, 2004). However, this theory also applies for diploid organisms. Mate competition assumes that some or all mating takes place at the natal site, resulting in competition for mates among males, including among brothers (Godfray et al., 1993; King and D'Souza, 2004). Females can adjust the sex ratio of their offspring in response to the number of females ovipositing in a same place, for example, in parasitoids a female can develops a greater proportion of males when she recognizes that its host has been previously parasitized by a conspecific female; the differences in sex ratio could be the result of the variation in local mate competition intensity (Flanagan et al., 1998; King and D'Souza, 2004). The Competition between individuals also affect the reproduction of insects, for example, under greater population densities competition for food and mates may result in a significant reduction of fecundity and fertility (Varley et al., 1974; Flanagan et al., 1998) Food resources (quality and amount) Food availability and quality are key factors of the reproduction of the insects. In the case of a herbivorous insect, its fecundity depends of the nutrients offered by the host plants: e.g. Nitrogen and Carbon (Awmack and Leather, 2002; Bauerfeind and Fischer, 2005; Jordão et al., 2010; Ross et al., 2011); a significant increase in females fecundity has been observed when they are feeding on host plants or diets rich in any of these two elements (Leatemia et al., 1995; Awmack and Leather, 2002; Jordão et al., 2010). Even more, in some species of 5

22 Lepidoptera, an adequate intake of carbohydrates and amino acids is necessary to complete the development of the reproductive systems of adults (Jordão et al., 2010). The effect of food resources on insect sex ratios is less studied. It can be explained by the sex allocation theory (Wild and West, 2009), which says the allocation of resources to males compared with females will depend of their reproductive function and the sex ratio is affected by the relative benefits of producing sons versus daughters. Under conditions with limited food resources the mothers should bias their sex ratio towards the sex that suffers less (Berndt and Wratten, 2005; Ross et al., 2011). For example, in the case of the mealybug Planococcus citri females with restrictions of food produce a more female-biased sex ratio, because the males are more reliant on maternal resources than females (Berndt and Wratten, 2005) Abiotic factors Temperature Temperature is a key factor in the ecology of insects; it directly affects their reproduction, development, survival, range and abundance (Ratte, 1985; Bale et al., 2002; Cui et al., 2008). Field and laboratory studies have demonstrated a positive correlation between temperature and insect reproduction (Naranjo and Sawyer, 1987; Wermelinger and Seifert, 1999; Keena, 2006; Teodoro et al., 2008; Jaramillo et al., 2009; Wang et al., 2009). For example, in the case of the coffee berry borer (H. hampei) higher populations and population growth rates (λ) have been observed at higher temperatures (Jaramillo et al., 2009; Teodoro et al., 2009). Moreover, the minimum and maximum temperature thresholds for the optimal reproduction have been established in laboratory experiments for several insect species. Extremely low and high temperatures may reduce the insect s fecundity through direct 6

23 interference with the normal development of oocytes, ovaries, testes, ejaculatory duct tracts and other reproductive organs (Barker and Herman, 1976; Goehring and Oberhauser, 2002; Wang et al., 2009). In insects (as I mentioned above) genetic factors are the primary signal for determining the sex of an individual; however, the sex ratio can be affected indirectly by environmental conditions as temperature. For example, in haplodiploid species males are more sensitive to environmental variation than diploid females, which can adapt more quickly to changing conditions through the heterozygosity of deleterious genes, and a greater mortality of haploid males is observed at high temperatures (Schowalter, 2006). Temperature also affects the mother s decisions to produce sons versus daughters in a determined environment: under extreme conditions of temperature a mother should bias her offspring sex ratio towards the sex that suffers less (Ross et al., 2011). Several studies have shown that females exposed to higher temperatures produce sex ratios more biased towards females (Nigro et al., 2007; Ross et al., 2011). Temperature can interact with other factors that are involved in reproduction and sex ratio, such as: sex determination systems, endosymbiotic bacteria, food quality and population sizes (Vaast et al., 2006; Gotoh et al., 2007; Traut et al., 2008; Teodoro et al., 2009). A classic example of the effect of temperature on sex determination systems was observed in the diploid moth Talaeporia tubulosa. In Lepidoptera the genetic notation of sex chromosome is WZ/ZZ, and in contrast to most other animal groups females are the heterogamic sex (WZ, and males are ZZ) (Traut, 1999; Traut et al., 2008). In absence of the W chromosome temperature determines the progeny's sex, if the temperatures are high (30-37 C) more females are produced, while at low temperatures (3-8 C) more males are produced (Traut et al., 2008). 7

24 Endosymbiotic bacteria are sensitive to high temperatures; in species where biased sex ratios are associated with endosymbiotic bacteria heat treatments can eliminate or reduce the infection from their hosts (Hurst et al., 2001; Gotoh et al., 2007) such that the insect host begins to produce a progeny with sex ratios close to 1: Relative humidity Many studies have determined the effect of relative humidity (RH) as a factor driving the abundance, distribution and reproduction of insects (Human et al., 1998; Pinheiro et al., 2002; Boieiro et al., 2013). For instance, relative humidity was the more important abiotic factor explaining the population densities of the coffee berry borer H. hampei (Teodoro et al. (2008). Contrary to temperature, the relation between humidity and reproduction may be negative. For example, fecundity of females of Liriomyza sativae exposed to 30% of RH was three to four times higher than that observed in females maintained at levels above 50% (Costa-Lima et al., 2010). Similar to temperature there is a probably humidity threshold to ensure successful fertility in insects; humidity levels under 30% and above 70% caused a significant decrease in egg hatching of Triatoma brasiliensis, the main vector of Chagas disease (Guarneri et al., 2002). Low relative humidity can induce reproductive diapause in some species of insects. Diapause occurs in all stages of insect development; the effect in adults is observed in oogenesis, vitellogeneis, and mating behavior (Tatar and Yin, 2001). For example, relative humidity under 50% induced and maintained reproductive diapause in the braconid parasitoid Microplitis demolitor (Seymour and Jones, 2000) and the beetle Bruchidius atrolineatus (Lenga et al., 1993). 8

25 Relatively few studies have considered the effect of relative humidity on sex ratio. Relative humidity did not affect the sex ratio of emerging adults of Prostephanus truncates, an important pest of maize (Shires, 1979). Pappas et al. (2008) reared the predator Dichorchrysa prasina in a range of relative humidity from 12 to 95% and did not find a significant effect of relative humidity on sex ratio. 2. Factors that shape bacterial communities in insects 2.1 Biotic factors Food resources or diet Insects are associated with a range of obligate, facultative, pathogenic and parasitic symbionts (Wernegreen, 2002; Moran et al., 2008; Jones et al., 2013). The digestive system contains the vast majority of the insect s bacteria and most studies of bacterial communities have focused only on the gut (Dillon and Dillon, 2004; Kelley and Dobler, 2011); probably for these two reasons most studies have concluded that the main factor shaping bacterial communities of the insects is the substrate or diet (Chandler et al., 2011; Colman et al., 2012; Staubach et al., 2013; Aksoy et al., 2014; Montagna et al., 2015). For example, bacterial communities differed significantly between herbivorous ants and predatory ants, but the communities were similar among different species of ants with the same diet (Anderson et al., 2012). Alterations in the host s diet can change completely the community structure (Dillon and Dillon, 2004; Colman et al., 2012; Staubach et al., 2013). Bacterial communities in Drosophila species can change in response to diet changes and these did not differ significantly among species when the individuals fed on the same substrate or diet (Chandler et al., 2011; Staubach et al., 2013). 9

26 Stage of development of insect hosts Changes in diversity, abundance and structure of bacterial communities have been observed across life stages of insects (Wang et al., 2011; Martinson et al., 2012). For instance, bacterial communities of larvae of Nasonia species were simple, but composition and diversity increased with development into pupae and adult (Brucker and Bordenstein, 2012). Based on this observation we suggest a progressive colonization related with development: communities from eggs are probably formed mainly by vertically transmitted symbionts, while larvae can acquire other bacteria from the environment during feeding and bacterial communities from adults will be the most complex and assembled according to the diet. However, some studies have shown that some taxa are maintained throughout the life cycle (Delalibera et al., 2007; Vasanthakumar et al., 2008) Abiotic factors Temperature Few studies have measured the effect of temperature on total bacterial communities in insects; however, some have determined the effect of temperature on abundance, distribution and efficiency of transmission of some important symbionts. For example, Prado et al. (2009) showed that frequency of the most important symbiont in the pentatomid bug Nezara viridula decreased from 100% to less than 10% when the insects were reared at 20 C and 30 respectively. Temperature fluctuations also can change the abundance Wolbachia and can affect its efficiency of transmission to the offspring (Stouthamer et al., 1999; Hurst et al., 2001). Furthermore, heat treatments over 35 C have been used as an efficient method to remove Wolbachia infections (Werren, 1997; Kyei-Poku et al., 2003; Gotoh et al., 2007) in order to examine the effect of this endosymbiont on reproduction and fitness of their hosts. 10

27 3. Coffee and Coffee Berry Borers: an ideal system to study the effect of biotic and abiotic factors on reproduction, sex ratio and bacterial communities of insects The coffee coffee berry borer (CBB) system (Coffea arabica Hypothenemus hampei) is ideal to study the effects of temperature, relative humidity, food resources and their possible interactions on reproduction, sex ratio and bacterial communities of an insect under field and laboratory conditions. Because coffee can be planted under contrasting technologies (e.g. sun and dense shade), which provide habitats with different environmental conditions, for example, coffee shade plantations are characterized by high levels of relative humidity and temperatures ranging from 14 to 30 C, while in sun coffee plantations temperatures can vary from 16 to 38 C with lower relative humidity (Mariño et al., Submitted). Also the exposition of coffee plants to direct sunlight increases photosynthesis; sun favors the productivity of coffee (Perfecto et al., 1996; DaMatta and Rodríguez, 2007). Vaast et al. (2006) demonstrated that chlorogenic acid and trigonelline content are higher in fruits from sun plantations, important nitrogenous compounds in coffee fruits (Duarte et al., 2010). Nitrogen content is a determining factor in herbivorous insects fecundity (Awmack and Leather, 2002) The coffee berry borer (H. hampei) The coffee berry borer (CBB) Hypothenemus hampei (Coleoptera: Curculionidae: Scolytinae) is the most devastating pest of coffee worldwide (Damon, 2000; Jaramillo et al., 2006; Vega et al., 2009). The CBB has co-evolved with coffee plants and depends principally on coffee fruits for its food, reproduction, and development (Damon, 2000). The entire life cycle of CBB occurs inside of the berries and starts when adult females emerge from fallen or remaining fruits, about 60 to 100 days after flowering (DAF) (Muñoz, 1989; Monzón, 2004) in search of new berries to colonize. The females burrow through the exocarp, mesocarp and endocarp until they reach the endosperm and there they start to oviposite; the final damage is 11

28 consequence of feeding of larvae and adults (Jaramillo et al., 2006). The duration of development from egg to adult is positively correlated with temperature (Jaramillo et al., 2009), for example, this may take 28 to 34 days at 27 C (Damon, 2000). The CBB usually has a sex ratio skewed 10:1 towards females. This deviation has been ascribed to its sex determination system, which is functional haplodiploidy (Brun et al., 1995). However, it has also been suggested that the endosymbiotic bacteria Wolbachia could play a role in the biased sex ratio of the CBB. This bacterium has been detected in adult females from several countries and is thought to induce cytoplasmic incompatibility (Vega et al., 2002). The economic and social impacts of the CBB in coffee-producing areas have motivated research on its biology in order to mitigate damage. Most research has focused on the effect of environmental variables and coffee management on CBB infestation and population densities (Wrigley, 1988; Soto-Pinto et al., 2002; Teodoro et al., 2008; Teodoro et al., 2009), the effect of climate change on CBB damage and distribution (Jaramillo et al., 2009; Jaramillo et al., 2011), the use of natural enemies such as parasitoids (Murphy and Moore, 1990; Infante et al., 2001; Lauzière et al., 2001), ants (Armbrecht and Perfecto, 2003; Armbrecht and Gallego, 2007; Larsen and Philpott, 2010) and entomopathogenic fungi and bacteria (De la Rosa et al., 1997; Haraprasad et al., 2001; Méndez-López et al., 2003; Vega et al., 2008). Field and laboratory studies need to be combined to sort out interactions between the factors that influence CBB nutrition, protect against natural enemies and enhance its distribution and reproduction favoring the presence of more females. 12

29 4. Dissertation thesis structure The dissertation is composed of three chapters, focused on determining the effect of several factors and the interactions between them on the population densities, sex ratios and bacterial communities of the CBB. Population sizes and sex ratio can be influenced directly by environmental conditions, mainly temperatures, which are determined by the ecology of the coffee plants (Chapter 2). Population sizes and sex ratios also can be directly affected by endosymbionts that are insect s reproductive manipulators (e.g. Wolbachia) (Chapter 3). Environmental conditions like temperature also affect bacterial communities, specifically Wolbachia (Chapter 4) (Figure 1). Below I briefly describe the objectives and results of the three chapters: Chapter 2: Sun vs. shade affects infestation, population size per fruit and sex ratio of the coffee berry borer (Hypothenemus hampei) in Puerto Rico. In this chapter I compare the infestation, number of CBB per fruit and sex ratio between coffee plantations growing under total sunlight and shade provide by other trees. I sampled in three shade and three sun coffee plantations in Adjuntas, Puerto Rico during the coffee-growing seasons in 2010 and Using generalized lineal models I determined a significant effect of type of coffee on CBB infestation, population densities and sex ratio. Infestation was significantly higher in shade plots, while population sizes per fruit were higher in sun plots, and the sex ratio was more biased to females in sun. We found significant differences in temperature, relative humidity and light between these two coffee production systems: sun coffee plots were characterized by higher temperatures and light and lower relative humidity; environmental variables appeared to be an important driver of CBB population sizes and sex ratio. I found positive correlations between temperature and the number of pupae, females and males. 13

30 Chapter 3: Wolbachia affects reproduction, sex ratio and population dynamics of the Coffee Berry Borer Hypothenemus hampei (Coleoptera: Curculionidae). In the second chapter I amplify and sequence a segment of the wsp gene to determine the presence of Wolbachia in CBB adult females in Puerto Rico from field and laboratory samples. Borers were reared for ten generations in artificial diets Cenibroca with and without tetracycline 0.1% (w/v), which was added to eliminate or reduce the infection with Wolbachia in CBB populations. The treatment with antibiotic reduced significantly the proportion of Wolbachia from 0.49 to 0.04%; after three generations females fed with tetracycline produced significantly fewer total progeny and less skewed sex ratio, after fifth generation the proportion of females that oviposited and the amount of eggs per females were significantly lower in the treatments with the antibiotic. The CBB population dynamics also was affected by the antibiotic, CBB populations treated with tetracycline had less performance in population growth rate or λ= 1.07 compared with 1.11 for control. The mean generation time (T) increased by 6.2 days, so females reared with tetracycline would produce 8.1 generations per year compared with 9.4 for controls. These results suggest that Wolbachia are involved on CBB reproduction, sex ratio and population dynamics. Chapter 4: The bacterial microbiota of the coffee berry borer Hypothenemus hampei is influenced by host diet, development and environment. For this chapter I used MiSeq-based sequencing of the hypervariable region V4 of the 16S rrna gene to characterize the bacterial communities associated with wild and laboratory reared eggs and adult females. Wild samples were obtained from infested fruits from plants growing under sun and shade conditions, while samples from laboratory were composed by individuals from the eleven generation reared in Cenibroca diet with and without tetracycline. I used QIIME (Quantitative Insights Into Microbial Ecology) to process and analyze the data. In general, samples from infested fruits were more diverse than those from artificial diet. The treatment with tetracycline reduced 14

31 significantly the diversity in adult females but not in eggs. Environmental conditions did not affect the overall diversity of the bacterial communities. However, higher temperatures showed the same effect that antibiotic on Wolbachia. In adult females from fruits collected in sun plots the proportion were significantly reduced to 0.04%. Beta diversity (PCoA) showed that CBB-associated microbiota appears to be determined mainly by diet and stage of development. These results suggest that CBB bacterial communities are affected by temperature, diet and stage of development of the hosts. Ecology of host (e.g. sun vs. shade coffee) Environments with different characteristics (e.g. temperature) 1 INSECT : REPRODUCTION AND SEX RATIO 3 2 Different food resources Bacterial communities (e.g. Wolbachia) Figure 1. Factors that affect the reproduction, sex ratio and bacterial communities of the coffee berry borer (Hypothenemus hampei). (1) Chapter 2- Field environmental conditions: Temperature and relative humidity (blue line), (2) Chapter 3: Bacterial communities (mainly endosymbionts that are reproductive manipulators: Wolbachia) (green line) or (3) Chapter 4: Effect of temperature on Bacterial communities (red line 15

32 5. References Aksoy, E., Telleria, E.L., Echodu, R., Wu, Y., Okedi, L.M., Weiss, B.L., Aksoy, S., Caccone, A., Analysis of multiple tsetse fly populations in Uganda reveals limited diversity and species-specific gut microbiota. Appl. Environ. Microbiol. 80, Anderson, K.E., Russell, J.A., Moreau, C.S., Kautz, S., Sullam, K.E., Hu, Y., Basinger, U., Mott, B.M., Buck, N., Wheeler, D.E., Highly similar microbial communities are shared among related and trophically similar ant species. Mol. Ecol. 21, Armbrecht, I., Gallego, M.C., Testing ant predation on the coffee berry borer in shaded and sun coffee plantations in Colombia. Entomol. Exp. Appl. 124, Armbrecht, I., Perfecto, I., Litter-twig dwelling ant species richness and predation potential within a forest fragment and neighboring coffee plantations of contrasting habitat quality in Mexico. Agric. Ecosyst. Environ. 97, Awmack, C.S., Leather, S.R., Host plant quality and fecundity in herbivorous insects. Annu. Rev. Entomol. 47, Bale, J.S., Masters, G.J., Hodkinson, I.D., Awmack, C., Bezemer, T.M., Brown, V.K., Butterfield, J., Buse, A., Coulson, J.C., Farrar, J., Good, J.E.G., Harrington, R., Hartley, S., Jones, T.H., Lindroth, R.L., Press, M.C., Symrnioudis, I., Watt, A.D., Whittaker, J.B., Herbivory in global climate change research: direct effects of rising temperature on insect herbivores. Glob. Chang. Biol. 8, Barker, J.F., Herman, W.S., Effect of photoperiod and temperature on reproduction of the monarch butterfly, Danaus plexippus. J. Insect Physiol. 22, Bauerfeind, S.S., Fischer, K., Effects of adult-derived carbohydrates, amino acids and micronutrients on female reproduction in a fruit-feeding butterfly. J. Insect Physiol. 51, Berndt, L.A., Wratten, S.D., Effects of alyssum flowers on the longevity, fecundity, and sex ratio of the leafroller parasitoid Dolichogenidea tasmanica. Biol. Control 32, Beukeboom, L., Insects and sex. Proceedings of the Netherlands Entomological Society Meeting. University of Groningen, Centre for Ecological and Evolutionary Studies, pp Boieiro, M., Carvalho, J.C., Cardoso, P., Aguiar, C.A., Rego, C., e Silva, I.d.F., Amorim, I.R., Pereira, F., Azevedo, E.B., Borges, P.A., Spatial factors play a major role as determinants of endemic ground beetle beta diversity of Madeira Island Laurisilva. PloS one 8, e Braig, H., Turner, B., Perotti, M., Symbiotic Rickettsia. Insect Symbiosis 3. Brelsfoard, C.L., Dobson, S.L., Wolbachia-based strategies to control insect pests and disease vectors. Asia Pac. J. Mol. Biol. Biotechnol. 17,

33 Brucker, R.M., Bordenstein, S.R., The roles of host evolutionary relationships (genus: Nasonia) and development in structuring microbial communities. Evolution 66, Brun, L.O., Stuart, J., Gaudichon, V., Aronstein, K., French-Constant, R., Functional haplodiploidy: a mechanism for the spread of insecticide resistance in an important international insect pest. Proc. Natl. Acad. Sci. 92, Chandler, J.A., Lang, J.M., Bhatnagar, S., Eisen, J.A., Kopp, A., Bacterial communities of diverse Drosophila species: ecological context of a host microbe model system. PLoS Gene. 7, e Charnov, E., Los-den Hartogh, R., Jones, W., Van den Assem, J., Sex ratio evolution in a variable environment. Nature 289, Chen, S.-J., Lu, F., Cheng, J.-A., Jiang, M.-X., Way, M.O., Identification and biological role of the endosymbionts Wolbachia in rice water weevil (Coleoptera: Curculionidae). Environ. Entomol. 41, Colman, D., Toolson, E., Takacs Vesbach, C., Do diet and taxonomy influence insect gut bacterial communities? Mol. Ecol. 21, Cordaux, R., Bouchon, D., Grève, P., The impact of endosymbionts on the evolution of host sex-determination mechanisms. Trends Genet. 27, Costa-Lima, T., Geremias, L., Parra, J., Reproductive activity and survivorship of Liriomyza sativae (Diptera: Agromyzidae) at different temperatures and relative humidity levels. Environ. Entomol. 39, Cui, X., Fanghao, W., Ming, X., Tongxian, L., Effects of heat shock on survival and reproduction of two whitefly species, Trialeurodes vaporarioum and Bemisia tabaci biotype B. J. Insect Sci. 8, DaMatta, F., Rodríguez, N., Producción sostenible de cafetales en sistemas agroforestales del Neotrópico: una visión agronómica y eco fisiológica. Agron. Colomb. 25, Damon, A., A review of the biology and control of the coffee berry borer, Hypothenemus hampei (Coleoptera : Scolytidae). Bull. Entomol. Res. 90, De la Rosa, W., Alatorre, R., Trujillo, J., Barrera, J., Virulence of Beauveria bassiana (Deuteromycetes) strains against the coffee berry borer (Coleoptera: Scolytidae). J. Econ. Entomol. 90, Dedeine, F., Vavre, F., Fleury, F., Loppin, B., Hochberg, M.E., Boulétreau, M., Removing symbiotic Wolbachia bacteria specifically inhibits oogenesis in a parasitic wasp. Proc. Natl. Acad. Sci. U.S.A. 98, Delalibera, I., Vasanthakumar, A., Burwitz, B.J., Schloss, P.D., Klepzig, K.D., Handelsman, J., Raffa, K.F., Composition of the bacterial community in the gut of the pine engraver, Ips pini (Say)(Coleoptera) colonizing red pine. Symbiosis Rehovot 43,

34 Dillon, R., Dillon, V., The gut bacteria of insects: nonpathogenic interactions. Ann. Rev. Entomol. 49, Dong, P., Wang, J.-J., Zhao, Z.-M., Wolbachia infection and its influence on the reproduction of stored-product psocid, Liposcelis tricolor (Psocoptera: Liposcelididae). J. Insect Sci. 6. Duarte, G.S., Pereira, A.A., Farah, A., Chlorogenic acids and other relevant compounds in Brazilian coffees processed by semi-dry and wet post-harvesting methods. Food Chem. 118, Ebbert, M.A., Wrensch, D., Endosymbiotic sex ratio distorters in insects and mites. In: Wrensch, D.L., Ebbert, M.A. (Eds.), Evolution and diversity of sex ratio in insects and mites. Chapman Hall, New York, pp Engelstädter, J., Telschow, A., Cytoplasmic incompatibility and host population structure. Heredity 103, Flanagan, K., West, S., Godfray, H., Local mate competition, variable fecundity and information use in a parasitoid. Anim. Behav. 56, Fry, A., Palmer, M., Rand, D., Variable fitness effects of Wolbachia infection in Drosophila melanogaster. Heredity 93, Godfray, H., Hardy, I., Wrensch, D., Ebbert, M., Sex ratio and virginity in haplodiploid insects. Evolution and diversity of sex ratio in insects and mites., Goehring, L., Oberhauser, K.S., Effects of photoperiod, temperature, and host plant age on induction of reproductive diapause and development time in Danaus plexippus. Ecol. Entomol. 27, Gotoh, T., Noda, H., Ito, S., Cardinium symbionts cause cytoplasmic incompatibility in spider mites. Heredity 98, Guarneri, A.A., Lazzari, C., Diotaiuti, L., Lorenzo, M.G., The effect of relative humidity on the behaviour and development of Triatoma brasiliensis. Physiol. Entomol. 27, Hake, L., O'Connor, C., Genetic mechanisms of sex determination. Nature Education 1, 25. Haraprasad, N., Niranjana, S., Prakash, H., Shetty, H.S., Wahab, S., Beauveria bassiana a potential mycopesticide for the efficient control of coffee berry borer, Hypothenemus hampei (Ferrari) in India. Biocontrol Sci. Tech. 11, Human, K.G., Weiss, S., Weiss, A., Sandler, B., Gordon, D.M., Effects of abiotic factors on the distribution and activity of the invasive Argentine ant (Hymenoptera: Formicidae). Environ. Entomol. 27,

35 Hunter, M.S., Perlman, S.J., Kelly, S.E., A bacterial symbiont in the Bacteroidetes induces cytoplasmic incompatibility in the parasitoid wasp Encarsia pergandiella. P. Roy. Soc. Lond. B Biol. 270, Hurst, G., Jiggins, F.M., Male-killing bacteria in insects: mechanisms, incidence, and implications. Emerging Infect. Dis. 6, 329. Hurst, G.D., Jiggins, F.M., Robinson, S.J., What causes inefficient transmission of male-killing Wolbachia in Drosophila? Heredity 87, Infante, F., Mumford, J., Baker, P., Barrera, J., Fowler, S., Interspecific competition between Cephalonomia stephanoderis and Prorops nasuta (Hym., Bethylidae), parasitoids of the coffee berry borer, Hypothenemus hampei (Col., Scolytidae). J. Appl. Entomol. 125, Jaramillo, J., Borgemeister, C., Baker, P., Coffee berry borer Hypothenemus hampei (Coleoptera: Curculionidae): searching for sustainable control strategies. Bull. Entomol. Res. 96, Jaramillo, J., Chabi-Olaye, A., Kamonjo, C., Jaramillo, A., Vega, F.E., Poehling, H.M., Borgemeister, C., Thermal Tolerance of the Coffee Berry Borer Hypothenemus hampei: Predictions of Climate Change Impact on a Tropical Insect Pest. Plos One 4, e6487. Jaramillo, J., Muchugu, E., Vega, F.E., Davis, A., Borgemeister, C., Chabi-Olaye, A., Some like it hot: The influence and implications of climatic change on coffee berry borer (Hypothenemus hampei) and coffee production in east Africa. Plos One 4, Jia, F.-X., Yang, M.-S., Yang, W.-J., Wang, J.-J., Influence of continuous high temperature conditions on Wolbachia infection frequency and the fitness of Liposcelis tricolor (Psocoptera: Liposcelididae). Environ. Entomol. 38, Jones, R.T., Sanchez, L.G., Fierer, N., A cross-taxon analysis of insect-associated bacterial diversity. PloS one 8, e Jordão, A.L., Nakano, O., Janeiro, V., Adult carbohydrate feeding affects reproduction of Phthorimaea operculella (Zeller)(Lepidoptera: Gelechiidae). Neotrop. Entomol. 39, Kageyama, D., Hoshizaki, S., Ishikawa, Y., Female-biased sex ratio in the Asian corn borer, Ostrinia furnacalis: evidence for the occurrence of feminizing bacteria in an insect. Heredity 81, Keena, M.A., Effects of temperature on Anoplophora glabripennis (Coleoptera: Cerambycidae) adult survival, reproduction, and egg hatch. Environ. Entomol. 35, Kelley, S.T., Dobler, S., Comparative analysis of microbial diversity in Longitarsus flea beetles (Coleoptera: Chrysomelidae). Genetica 139, King, B., D'Souza, J., Effects of constrained females on offspring sex ratios of Nasonia vitripennis in relation to local mate competition theory. Can. J. Zool. 82,

36 Kirkendall, L., Ecology and evolution of biased sex ratios in bark and ambrosia beetles. In: Wrensch, D., Ebbert, M. (Eds.), Evolution and diversity of sex ratio: insects and mites. Chapman & Hall, New York, USA, pp Kirkendall, L., Wrensch, D., Ebbert, M., Ecology and evolution of biased sex ratios in bark and ambrosia beetles. In: Wrensch, D.L., Ebbert, M.A. (Eds.), Evolution and diversity of sex ratio in insects and mites. Chapman Hall, New York, pp Kuijper, B., Pen, I., The evolution of haplodiploidy by male-killing endosymbionts: importance of population structure and endosymbiont mutualisms. J. Evol. Biol. 23, Kyei-Poku, G., Floate, K., Benkel, B., Goettel, M., Elimination of Wolbachia from Urolepis rufipes (Hymenoptera: Pteromalidae) with heat and antibiotic treatments: implications for host reproduction. Biocontrol Sci. Tech. 13, Larsen, A., Philpott, S.M., Twig-Nesting Ants: The Hidden Predators of the Coffee Berry Borer in Chiapas, Mexico. Biotropica 42, Lauzière, I., Brodeur, J., Pérez-Lachaud, G., Host stage selection and suitability in Cephalonomia stephanoderis Betrem (Hymenoptera: Bethylidae), a parasitoid of the coffee berry borer. Biolog. Control 21, Leatemia, J., Laing, J., Corrigan, J., Effects of adult nutrition on longevity, fecundity, and offspring sex ratio of Trichogramma minutum Riley (Hymenoptera: Trichogrammatidae). Can. Entomol. 127, Lenga, A., Glitho, I., Huignard, J., Interactions between photoperiod, relative humidity and host-plant cues on the reproductive diapause termination in Bruchidius atrolineatus Pic (Coleoptera Bruchidae). Invertebr. Repr. Dev. 24, Mariño, Y., Perez, M.E., Gallardo, F., Trifilio, M., Cruz, M., Bayman, P., Submitted. Sun vs. shade affects infestation, reproduction and sex ratio of the coffee berry borer (Hypothenemus hampei) in Puerto Rico. Agric. Ecosyst. Environ. Martinson, V.G., Moy, J., Moran, N.A., Establishment of characteristic gut bacteria during development of the honeybee worker. Appl. Environ. Microbiol. 78, Méndez-López, I., Basurto-Ríos, R., Ibarra, J.E., Bacillus thuringiensis serovar israelensis is highly toxic to the coffee berry borer, Hypothenemus hampei Ferr.(Coleoptera: Scolytidae). FEMS Microbiol. Lett. 226, Minkenberg, O.P., Tatar, M., Rosenheim, J.A., Egg load as a major source of variability in insect foraging and oviposition behavior. Oikos, Montagna, M., Chouaia, B., Mazza, G., Prosdocimi, E.M., Crotti, E., Mereghetti, V., Vacchini, V., Giorgi, A., De Biase, A., Longo, S., Effects of the Diet on the Microbiota of the Red Palm Weevil (Coleoptera: Dryophthoridae). PloS one 10, e

37 Monzón, A., Control biológico de la broca (Hypothenmus hampei) en el cultivo de café. Universidad Nacional Agraria (UNA). Dirección de Investigación, Extensión y Postgrado (DIEP). Serie Técnica, Moran, N.A., McCutcheon, J.P., Nakabachi, A., Genomics and evolution of heritable bacterial symbionts. Annu. Rev. Genet. 42, Muñoz, R., Ciclo biológico y reproducción partenogenética de la broca del fruto de cafeto Hypothenemus hampei (Ferr.). Turrialba 39, Murphy, S., Moore, D., Biological control of the coffee berry borer, Hypothenemus hampei (Ferrari)(Coleoptera, Scolytidae): previous programmes and possibilities for the future. Biocontrol News and information 11, Naranjo, S.E., Sawyer, A.J., Reproductive biology and survival of Diabrotica barberi (Coleoptera: Chrysomelidae): effect of temperature, food, and seasonal time of emergence. Ann. Entomol. Soc. Am 80, Niebylski, M.L., Peacock, M.G., Fischer, E.R., Porcella, S.F., Schwan, T.G., Characterization of an endosymbiont infecting wood ticks, Dermacentor andersoni, as a member of the genus Francisella. Appl. Environ. Microbiol. 63, Nigro, R.G., Campos, M.C.C., Perondini, A.L.P., Temperature and the progeny sexratio in Sciara ocellaris (Diptera, Sciaridae). Genet Mol. Biol. 30, Pappas, M., Broufas, G., Koveos, D., Effect of relative humidity on development, survival and reproduction of the predatory lacewing Dichochrysa prasina (Neuroptera: Chrysopidae). Biolog. Control 46, Perfecto, I., Rice, R.A., Greenberg, R., Van der Voort, M.E., Shade Coffee: A Disappearing Refuge for Biodiversity. BioScience 46, Pinheiro, F., Diniz, I., Coelho, D., Bandeira, M., Seasonal pattern of insect abundance in the Brazilian cerrado. Austral Ecol. 27, Prado, S.S., Golden, M., Follett, P.A., Daugherty, M.P., Almeida, R.P., Demography of gut symbiotic and aposymbiotic Nezara viridula L.(Hemiptera: Pentatomidae). Environ. Entomol. 38, Ratte, H.T., Temperature and insect development. Environmental physiology and biochemistry of insects. Springer, pp Ross, L., Dealey, E.J., Beukeboom, L.W., Shuker, D.M., Temperature, age of mating and starvation determine the role of maternal effects on sex allocation in the mealybug Planococcus citri. Behav. Ecol. sociobiol. 65, Ross, L., Pen, I., Shuker, D.M., Genomic conflict in scale insects: the causes and consequences of bizarre genetic systems. Biological Reviews 85,

38 Schowalter, T., Population ecology. Insect ecology: an ecosystem approach. Elsevier Inc Burlington, MA, USA, pp Seymour, J.E., Jones, R.E., Humidity-terminated diapause in the tropical braconid parasitoid Microplitis demolitor. Ecol. Entomol. 25, Shires, S., Influence of temperature and humidity on survival, development period and adult sex ratio in Prostephanus truncatus (Horn)(Coleoptera, Bostrichidae). J. Stored Prod. Res. 15, Son, Y., Luckhart, S., Zhang, X., Lieber, M.J., Lewis, E.E., Effects and implications of antibiotic treatment on Wolbachia-infected vine weevil (Coleoptera: Curculionidae). Agric. Forest Entomol. 10, Soto-Pinto, L., Perfecto, I., Caballero-Nieto, J., Shade over coffee: its effects on berry borer, leaf rust and spontaneous herbs in Chiapas, Mexico. Agroforest. Syst. 55, Staubach, F., Baines, J.F., Künzel, S., Bik, E.M., Petrov, D.A., Host species and environmental effects on bacterial communities associated with Drosophila in the laboratory and in the natural environment. PloS one 8, e Stouthamer, R., Breeuwer, J.A., Hurst, G.D., Wolbachia pipientis: microbial manipulator of arthropod reproduction. Ann. Rev. Microbiol. 53, Tatar, M., Yin, C.-M., Slow aging during insect reproductive diapause: why butterflies, grasshoppers and flies are like worms. Exp. Geront. 36, Teodoro, A., Klein, A.-M., Tscharntke, T., Environmentally mediated coffee pest densities in relation to agroforestry management, using hierarchical partitioning analyses. Agric. Ecosyst. Environ. 125, Teodoro, A., Klein, A., Reist, P., Tscharntke, T., Agroforestry management affects coffee pests contingent on season and developmental stage. Agric. For. Entomol. 11, Traut, W., The evolution of sex chromosomes in insects: Differentiation of sex chromosomes in flies and moths. Eur J Entomol 96, Traut, W., Sahara, K., Marec, F., Sex chromosomes and sex determination in Lepidoptera. Sexual Development 1, Vaast, P., Bertrand, B., Perriot, J.J., Guyot, B., Genard, M., Fruit thinning and shade improve bean characteristics and beverage quality of coffee (Coffea arabica L.) under optimal conditions. J. Sci. Food Agric. 86, Varley, G.C., Gradwell, G.R., Hassell, M.P., Insect population ecology: an analytical approach. Univ of California Press. Vasanthakumar, A., Handelsman, J., Schloss, P.D., Bauer, L.S., Raffa, K.F., Gut microbiota of an invasive subcortical beetle, Agrilus planipennis Fairmaire, across various life stages. Environ. Entomol. 37,

39 Vega, F.E., Benavides, P., Stuart, J.A., O'Neill, S.L., Wolbachia infection in the coffee berry borer (Coleoptera: Scolytidae). Ann. Entomol. Soc. Am 95, Vega, F.E., Infante, F., Castillo, A., Jaramillo, J., The coffee berry borer, Hypothenemus hampei (Ferrari)(Coleoptera: Curculionidae): a short review, with recent findings and future research directions. Terr. Arthropod Rev. 2, Vega, F.E., Posada, F., Aime, M.C., Pava-Ripoll, M., Infante, F., Rehner, S.A., Entomopathogenic fungal endophytes. Biol. Control 46, Wang, X.-G., Johnson, M.W., Daane, K.M., Nadel, H., High summer temperatures affect the survival and reproduction of olive fruit fly (Diptera: Tephritidae). Environ. Entomol. 38, Wang, Y., Gilbreath III, T.M., Kukutla, P., Yan, G., Xu, J., Dynamic gut microbiome across life history of the malaria mosquito Anopheles gambiae in Kenya. PloS one 6, e Weeks, A.R., Breeuwer, J.A., A new bacterium from the Cytophaga-Flavobacterium- Bacteroides phylum that causes sex ratio distortion. Insect symbiosis, Weeks, A.R., Reynolds, K.T., Hoffmann, A.A., Wolbachia dynamics and host effects: what has (and has not) been demonstrated? Trends Ecol. Evol. 17, Weeks, A.R., Velten, R., Stouthamer, R., Incidence of a New Sex ratio distorting Endosymbiotic Bacterium among arthropods. P. Roy. Soc. Lond. B Biol. 270, Wermelinger, B., Seifert, M., Temperature-dependent reproduction of the spruce bark beetle Ips typographus, and analysis of the potential population growth. Ecol. Entomol. 24, Wernegreen, J.J., Genome evolution in bacterial endosymbionts of insects. Nature Rev. Genet. 3, Werren, J.H., Biology of Wolbachia. Ann. Rev. Entomol. 42, Werren, J.H., Baldo, L., Clark, M.E., Wolbachia: master manipulators of invertebrate biology. Nat. Rev. Microbiol. 6, Werren, J.H., Beukeboom, L.W., Sex determination, sex ratios, and genetic conflict. Annu. Rev. Ecol. Syst., Wild, G., West, S.A., Genomic imprinting and sex allocation. Am. Naturalist 173, E1- E14. Wrigley, G., Coffee. Tropical Agriculture Series. Longman Scientific & Technical and John Wiley & Sons, New York, pp Zchori-Fein, E., Gottlieb, Y., Kelly, S., Brown, J., Wilson, J., Karr, T., Hunter, M., A newly discovered bacterium associated with parthenogenesis and a change in host selection behavior in parasitoid wasps. Proc. Natl. Acad. Sci. U.S.A. 98,

40 Zchori-Fein, E., Perlman, S.J., Distribution of the bacterial symbiont Cardinium in arthropods. Mol. Ecol. 13, Zchori Fein, E., Borad, C., Harari, A.R., Oogenesis in the date stone beetle, Coccotrypes dactyliperda, depends on symbiotic bacteria. Physiol. Entomol. 31,

41 CHAPTER 2 Sun vs. shade affects infestation, population size per fruit and sex ratio of the coffee berry borer (Hypothenemus hampei) in Puerto Rico Mariño-Cárdenas, Y., Pérez, M.E., Gallardo, F., Trifilio, M., Cruz, M. and Bayman. P. Sun vs. shade affects infestation, reproduction and sex ratio of the coffee berry borer (Hypothenemus hampei) in Puerto Rico. Submitted in March 10, 2015 to Agriculture, Ecosystem & Environment. 25

42 Abstract Sun and shade coffee (Coffea arabica L.), provide different biotic and abiotic environments which can affect the damage, distribution and reproduction of coffee pests. However, the effect of shade on damage by the coffee berry borer (CBB) Hypothenemus hampei (Ferrari), the most important coffee pest worldwide, is controversial and unclear. We compared infestation, population size per fruit and sex ratio of the CBB in three shade and three sun coffee plantations in Adjuntas, Puerto Rico during the coffee-growing season for two years. According to generalized linear models, CBB infestation was significantly higher in shade plots, and ranged from 2-68% compared with 3-29% in sun. However, the population sizes were higher in sun coffee fruits than in shade coffee, and the sex ratio was more biased to females in sun coffee. Sun coffee was characterized by higher temperatures and lower relative humidity. Environmental variables appeared to be important drivers of CBB population sizes in field, with positive correlations between temperature and the number of pupae, juveniles, females and males. This tendency of larger populations in sun plots could represent a serious threat for coffee. In addition to fewer CBB in infested fruits in shade plots, we observed that shade coffee also had more natural enemies of CBB such as ants and entomopathogenic fungi. 1. Introduction The coffee berry borer (CBB) Hypothenemus hampei Ferrari (Coleoptera: Curculionidae) is the most devastating pest of coffee worldwide (Damon, 2000; Soto-Pinto et al., 2002; Jaramillo et al., 2006; Vega et al., 2009). The CBB has co-evolved with coffee plants depends principally on coffee fruits for food, reproduction, and development (Damon, 2000). Economic losses are consequences of: (1) Drilled young fruits that abscise and fall to 26

43 the ground, (2) Drilled green and red fruits that do not abscise, but because of CBB damage have low weight and quality at harvest (Damon, 2000; Webge et al., 2003). In the last few decades, intensification of coffee production has changed the coffee agroecosystem. Coffee (Coffea arabica L.) evolved in Ethiopian forests as an understory tree, and is thus shade-dependent (Soto-Pinto et al., 2002; DaMatta et al., 2007). When coffee was introduced to Asia and Latin America, initially it was grown under shade to retain the physiological attributes of shade plants. In the 1980s the use of shade was abandoned or reduced to increase production (Perfecto et al., 1996; DaMatta and Rodríguez, 2007), to combat fungal diseases, specifically the coffee leaf rust Hemileia vastratix (Perfecto et al., 1996; DaMatta et al., 2002; Soto-Pinto et al., 2002) and pests, especially CBB (Beer et al., 1998; DaMatta et al., 2002). This modernization changed abiotic characteristics such as temperature and relative humidity (Teodoro et al., 2008), and biotic characteristics such as crop density, presence of herbivores, natural enemies of CBB (e.g., parasitoids, ants and pathogens, (Letourneau et al., 2009), and resource quality and quantity (Hunter and Price, 1992). These changes affected directly and indirectly the CBB's distribution, survival, abundance and reproduction (Bale et al., 2002; Silva et al., 2011). Several studies have investigated the effect of shade vs. sun on infestation of CBB, but the conclusions are unclear. Some authors have reported higher infestation of CBB in shade coffee than in sun coffee (Wrigley, 1988; Larsen and Philpott, 2010), presumably as a consequence of higher humidity, protection from the wind and optimal temperatures (21-27 C) (Damon, 2000; Vega et al., 2009). Others have found no significant effect of shade vs. sun on CBB infestation (Soto-Pinto et al., 2002). Even less clear is the effect of sun vs. shade on population sizes and sex ratio of CBB. Recently it has been argued that the CBB is heattolerant and population growth increases with temperature (Jaramillo et al., 2009; Burgiel and Muir, 2010). And modeling and climatic data suggest the CBB is favored by rising 27

44 temperatures as a consequence of climate change, increasing its distribution and damage to coffee (Jaramillo et al., 2011). These diverse results may partly reflect the fact that the studies have been done in diverse countries, with varying conditions. The objective of this study was to compare coffee planted under shade vs. sun in terms of CBB infestation, population densities and sex ratio over the course of the coffee-growing seasons of 2010 and 2011 in Puerto Rico. We asked the following questions: 1) Does the coffee agroecosystem (shade vs. sun) affect the infestation rate of CBB? We predicted that CBB would attain higher infestation in shade coffee than in sun coffee as a consequence of higher humidity and protection from the wind. 2) Does sun vs. shade coffee affect the population size per fruit and sex ratio of CBB within infested fruits? We hypothesized that fruits exposed to direct sunlight and higher temperatures will have larger CBB populations and more biased sex ratios (higher female: male) per fruit because more resources are available. 2. Materials and Methods 2.1. Study region and site descriptions The study was carried out in three farms in Adjuntas, part of the coffee-growing region in the central mountains of Puerto Rico. Adjuntas has an altitude of ~550 m above sea level, temperatures of C and mean annual precipitation of ~1870 mm (Harmsen et al., 2009). In each site two plots were chosen: 1) sun coffee (managed as a monoculture and exposed to direct sunlight,), and 2) shade coffee (with shade provided by up to nine tree species) (Table A.1 and Figure A.1). The characteristics of these plots are summarized in Table 1. During 2010, we sampled farms F1 and F2, but in 2011 we had to replace farm F2 with F3, because in F2 the grower had pruned all sun coffee plants. 28

45 The management of the three farms was similar: glyphosate was applied 3-4 x per year to control weeds. No chemicals were applied to control the CBB; however, in farms F1 and F3 the growers applied the entomopathogenic fungus Beauveria bassiana (commercial name Mycotrol) once a year. Also, all growers conducted a rigorous collection of all fruits after harvest to remove reservoirs of CBB between crops (a technique called 're-re' in Spanish, (Benavides et al., 2003; Jaramillo et al., 2006). From September 2011 to February 2012, temperature and relative humidity were recorded hourly using Onset HOBO dataloggers. Canopy cover was measured by taking photographs with a Canon EOS 100D equipped with a 4.5 mm F2.8 EXDC circular fisheye lens. The camera was mounted with the top north on a tripod at 100 cm above the ground. Photographs were taken in the morning ( Atlantic Standard Time). Ten photographs were taken in each plot and analyzed using Gap Light Analyzer (GLA Version 2.0) (Frazer et al., 1999). GLA gives the percentage of canopy openness; we used these values to determine canopy cover of shade plots as 1 % canopy openness Sampling methods Infestation of CBB was surveyed from June to November 2010 and This period included 120 DAF (days after main flowering), harvest and approximately 30 days after harvest. Twenty coffee plants were randomly selected in each plot and the central branch was tagged; the number of fruits and CBB-bored fruits on this branch was counted. The same branches were censused every two weeks, and newly-bored fruits were recorded. Population sizes per fruit were surveyed from June to December; this was a month longer than the sampling for infestation, extended to evaluate the number of CBBs in fruits that remained on the branches after the main harvest. Three bored fruits were collected from ten tagged plants every two weeks, for a total of 30 fruits per plot (180 totals per sampling). These coffee fruits were dissected with the aid of a dissecting microscope at 30-50X; we 29

46 counted the number of all developmental stages of CBB (i.e. eggs, larvae, pupae, juveniles and adults). All adults and juveniles were sexed, and the sex ratio was expressed as the proportion of females in the population of juveniles plus adults. The position of colonizing females inside the fruit was classified following Bustillo et al. (1998): (A) when the colonizing female has started to drill the fruit but has not yet reached the endocarp, and was not present (B) when the female has penetrated the endocarp, but not the endosperm; (C) when the female has reached the endosperm but has not oviposited yet; (D) when the female has made her gallery in the endosperm, and immature stages of the CBB are found. Artesanal traps were used to monitor CBB females flying in search of new fruits from July 2010 to March We placed six traps per plot, four at the ends and two in the center; all traps were hung 1.5 m above the ground. Traps were made from 2 L plastic soft drink bottles, the outer surface painted red with five windows (12 X 3.0 cm) cut 10 cm above the bottom (Uemura-Lima et al., 2010). A mix of ethanol-methanol (1:1) was placed in 50 ml plastic centrifuge tubes, and these were hung from a wire inside the bottle 8 cm below the top. The bottom of the bottle contained 200 ml of water with 1% liquid detergent, added to trap CBB females. The mix of water and detergent was replaced monthly, and CBB were counted by direct counts or by weight, depending on the density of capture Statistical analysis One-way ANOVAs were performed to determine differences in temperature, relative humidity, percentage of canopy openness between types of coffee (sun vs. shade) and canopy cover between shade plots. The probability of CBB infestation was estimated at branch level using a logistic model with the binomial variable CBB attack/non-attack; time (i.e., sampling date) was considered as a continuous variable to adjust curves of infestation. The abundance of 30

47 individuals of CBB was estimated at fruit level using a log-lineal model with a Poisson distribution. For population sizes we analyzed data beginning in August, when the first eggs were observed. Also, the abundance of females in each position (A- D) and females caught with traps were estimated using log-lineal models with a Poisson distribution. For all models, we considered as independent variables sun vs. shade, farm and year. To evaluate CBB sex ratio, we used only the data from fruits in which we observed juveniles and adults (the CBB stages where individuals can be sexed by visual inspection). The sex ratio was calculated as the proportion of CBBs that were females, counting only fruits with males. ANOVAs followed by post-hoc Tukey tests were used to test differences between sun vs. shade and years on the sex ratio; before the analyses the data were transformed to arsine(x) to fulfill assumptions of a normal distribution. Linear regressions were used to determine the relationship between number of individuals in different stages of development and sex ratio with maximum, minimum and mean temperature ( C) and relatively humidity (%), based on biweekly averages. All data analyses were performed in R (R, 2013). 3. Results 3.1. Environmental characteristics of plots Temperature was significantly higher in sun plots than shade plots (F = 11.78, df=1, P <0.001); the maximum temperatures registered were 38.2 C and 30.7 C for sun and shade plots, respectively. Humidity was significantly lower in sun plots than shade plots (F=191.7, df=1, P <0.001); minimum relative humidity was 28.3% for sun plots and 60.3% for shade plots (Table 1). The GLA results showed than sun plots differed significantly from shade plots in canopy openess (sun average, 69%; shade average, 17%; F=431.6, df=1, P <0.001). No 31

48 differences were observed in percentages of canopy cover in shade plots (F = 2.10, df=2, P =0.1420) Differences between sun vs. shade on infestation of CBB 2 The logistic model (including interactions between variables) fit these data well (χ 15 = 42451, P <0.0001). In general, the estimated infestation was significantly greater in shade plots than in sun plots (generalized linear model (glm), binomial error, Z= 18.46, P <0.001); this behavior was consistent in farms F1 and F3, but in farm F2 infestation was slightly higher in sun than shade (Table A.2 and 2). Infestation levels ranged from 2 to 68 % in shade and 3 to 29% in sun (Table 2). The estimated CBB infestation was higher in 2011 than 2010 (glm, binomial error, Z= 1.67, P< 0.001), to the extent that it was higher in sun plots in 2011 than in shade plots in Estimated infestation through time increased similarly in both years and was highest in November, just before harvest (Fig. 1a and 1b; glm, binomial error, Z= 1.36, P=0.171, Table A.2). Comparing among farms, CBB infestation was highest in F3 and lower in F2 than F1 (Table A.2); the shade plot of F3 had highest infestation levels of all (Figure 1a and 1b) Differences between sun vs. shade on number of CBB per fruit The log-lineal model (including interactions between variables) gave a good fit for this data (χ 2 7 = 636, P <0.0001). In general, the estimated population was significantly higher in sun than shade (Table A.3, glm, Poisson errors, Z=12.36, P < 0.001), and in 2011 compared with 2010 (Table A.3, glm, Poisson errors, Z=9.74, P< 0.001). The estimated population per farm was higher in F2 compared with F1 and lower in F3 compared with F1 (Table A.3). More than half of bored fruits contained no CBB offspring. Percent of bored fruits without CBB offspring was significantly higher in shade (79.4 %) than sun (55.8 %) (F = 16.80, df= 1, P < ) (Table 2). In both sun and shade, the percent fruits with CBB 32

49 offspring and the number of individuals per fruit increased significantly through the coffeegrowing season; in sun plots, highest populations were observed in September and October, whereas in shade, where fruits develop more slowly, peak of population occurred in November (Figure 1c and 1d). The estimated abundances of eggs, larvae and adults were significantly higher in sun plots (glm, Poisson errors, eggs χ 2 3 = 227, Z=3.59, P = ; larvae χ 2 3 = 68, Z=4.18, P = < and adults χ 2 3 = 8, Z=2.63, P= 0.008). However, the abundance of juveniles and pupae did not differ significantly between sun and shade (pupae χ 2 3 = 214, Z=1.57, P = and juveniles χ 2 3 = 51.5, Z= 1.73, P = 0.08) (Figure 2). Our sampling strategy did not permit detection of adult progeny who had already left the fruits Environmental variables and relationship to number of CBB per bored fruit Sun coffee plots had significantly higher temperatures and lower relative humidity than shade plots (Table 1). In sun plots, the total number of some stages was positively and significantly correlated with mean temperature (pupae: P=0.002, R 2 = 0.48; juveniles: P=0.039, R 2 = 0.40; females: P=0.031, R 2 = 0.43; males: P=0.037, R 2 = 0.47) and the median temperature (females: P=0.037, R 2 = 0.41). In shade plots, the total number of juveniles was positively correlated with minimum temperature (P=0.03, R 2 = 0.25), mean temperature (P=0.007, R 2 = 0.39) and median temperature (P=0.001, R 2 = 0.54). Also the number of females was positively correlated with the median temperature (P=0.031, R 2 = 0.26). No stage of development of CBB was correlated with relative humidity in either sun or shade Sex ratio We found significant differences in CBB sex ratio between sun vs. shade (F= 4.47, df=1, P = 0.03) and years (F= 5.41, df=1, P = 0.02); the sex ratio was more skewed towards 33

50 females in sun than shade plots (0.83 and 0.76 respectively) (Table 3). No males were found in 77.5% of shade fruits and 80.1% of sun fruits (counting only fruits with progeny of juveniles and adults) but this difference was not significant (F= 0.17, df=1, P = 0.67) (Table 3). No correlation was found between CBB sex ratio and any environmental variables Differences between sun vs. shade on the penetration positions of colonizing females of H. hampei The proportions of colonizing H. hampei females in the four positions inside the fruits as a function of sun vs. shade and year are presented in Figure 3. The proportion of colonizing H. hampei females in each positions varied during the season; the highest proportion of females found in position A was in June (28%), when penetration was just starting, followed by July (20%) and August (20%). The highest percentages of females in position B were in June (64%) and July (52%) and position C in July (25%) and August (28%).The highest proportion of females in position D was in November (61%) and December (53%), when fruits were ripest, followed by October (51%) and September (45%). The number of females in positions B and D were affected by sun vs. shade (glm, Poisson errors, position B: Z= 3.78, P < ; position D: Z= 3.81, P < 0.001). Colonizing females in position B were higher in shade plots (33%) than in sun plots (25%); whereas position D was higher in sun plots (43%) than in shade plots (34%) (Fig 3); in other words, more colonizing females reached the endosperm and reproduced in sun than in shade Differences between sun vs. shade on trap capture rates The estimated abundance of females caught in traps varied significantly between sun vs. shade coffee and during the growing season (glm, Poisson errors, sun vs. shade χ 2 1 = 3794, Z=61.66, P <0.0001; months χ 2 10 = , Z=778.02, P <0.0001). Capture was significantly higher in sun than shade coffee. Highest capture rates were obtained from 34

51 February to June and November and December with a marked peak in February 2010, when few coffee fruits were available for attack (Figure 4). 4. Discussion 4.1. CBB infestation levels compared to other coffee-growing countries This study shows that sun vs. shade coffee has a significant effect on CBB infestation, number of individuals per fruit, and sex ratio. Infestation of CBB was significantly greater in shade coffee plots, while CBB individuals per bored fruit were higher in sun coffee, and the sex ratio was more biased to females in sun coffee. Overall, the infestation of CBB in Adjuntas, Puerto Rico in this study was high, ranging from 2 to 67% of fruits surveyed (Table 2). This infestation is higher than in previous studies: for example, in Brazil in 1992-'93 infestation was 21 to 32% (Cure et al., 1998); in Mexico in 1997 infestation was 0.1 to 19% (Soto-Pinto et al., 2002), and from 5-35% in 2008 (Larsen and Philpott, 2010). In Colombia infestation was <2 to 25% in 1995 and < 2 to 13% in 1996 (Benavides et al., 2003). Perhaps the most relevant comparison is with Mexico in 1983, about five years after CBB was first reported there, when infestation was 10-15% (Barrera, 2005); CBB arrived in Puerto Rico in 2007 (NAPPO, 2007); four years later, in our second sampling season, far more fruits were infested than in any of the above-cited studies. CBB is considered the most important pest of coffee worldwide, but if the study sites in Adjuntas are representative, the level of damage in Puerto Rico appears to be unprecedented Coffee fruit development and its association to CBB infestation In both sun and shade coffee the infestation and population densities of CBB increased during the season; several studies have reported a close relation between coffee crop maturity and CBB damage (Cure et al., 1998; Mathieu et al., 1999; Damon, 2000; Camilo et al., 2003; Valdés and Ravelo, 2006; Teodoro et al., 2009; Vázquez et al., 2012). Coffee fruit 35

52 development can be divided into four stages: 1) blooming (February-May), 2) fruit growth and ripening (June-August) 3) ripening (September- November) and 4) senescence (December-January) (Arcila et al., 2002; Gay et al., 2006). The infestation of CBB increased through time for both years and in both sun and shade, and peaked in November, when most fruits were at stage three (Figure 1a and 1b). A similar pattern was seen in Cuba: CBB infestation started to increase in September and peaked in November (Vázquez et al., 2012). The gradual increase in infestation through the season is mainly a consequence of fruit maturation and growth of CBB populations through several generations Differences between sun vs. shade coffee on CBB infestation CBB infestation was significantly higher in shade than in sun: from 2-67% in shade vs. 3-29% in sun (Table 2). Previous studies also found that shade favored infestation (Wrigley, 1988; Larsen and Philpott, 2010). Our shade plots had higher and more stable humidity levels and lower temperatures than the sun plots (Table 1); these environmental conditions are conducive to CBB damage (Damon, 2000). However, shade coffee also had higher percentages of fruits in which the CBB attack only reached position A; an average of 79% of bored fruits had no reproduction inside them, compared to 56% in sun (Table 2). Position A is initial perforation; this means that many shade coffee fruits are drilled but the seed is not damaged. Fruits from sun plots reach optimum conditions earlier than shade plots and there are more fruits with optimum conditions for CBB penetration and reproduction, because in coffee plots with high levels of shade the development and maturation of coffee fruits are delayed (Geromel et al., 2008) Coffee fruit development and number of CBB per fruit In both sun and shade coffee CBB population sizes per fruit increased during fruit maturation (Fig 1c and 1d); we observed the first endosperm penetration in early July and the 36

53 first eggs at the end of July in sun plots, corresponding to DAF (days after flowering). In shade plots, oviposition started in early August corresponding to 163 DAF. Field data from sun plots in Colombia agree with this: in sun coffee, oviposition started DAF (Ruiz-Cárdenas and Baker, 2010). Also, sun and shade coffee differed in the time of maximum number of CBB per fruit, which was in October for sun plots and November for shade plots; as mentioned above shade delays the maturation of coffee fruits by at least a month (Geromel et al., 2008). Several researchers have showed a direct relation between fruit ripening and population sizes. For example, Mathieu et al. (1999) reported a significant correlation between these two variables; Camilo et al. (2003) found an increase in the number of eggs, larvae, pupae and juveniles of CBB as fruit development proceeded Differences between sun vs. shade on number of CBB per fruit Relatively few studies have compared number of CBB per infested fruit between sun and shade coffee; here we found significantly higher numbers of CBB in sun than in shade fruits. The insect s reproduction can be influenced by biotic and abiotic factors (Bale et al., 2002; Silva et al., 2011). Among abiotic factors, several studies have demonstrated a significant correlation between temperature and CBB population sizes: Jaramillo et al. (2009) reared the CBB in the laboratory at eight temperatures from 15 to 35 ºC and found that development was faster and the population growth rate (λ) was greater at higher temperatures. Field experiments also showed a positive correlation between temperature and CBB population densities (Teodoro et al., 2008). Also, Teodoro et al. (2009) observed higher populations in 'intensively managed' plots, which like our sun plots had higher temperatures and lower relative humidity than shade (Table 1). Although we did not find a significant correlation between temperature and total population, as Teodoro et al. (2008) did, we found that the numbers of individuals of some 37

54 stages of development were correlated with temperature; in both sun and shade coffee, the number of females and juveniles was positively correlated with the mean and median temperature. Number of juveniles was also positively correlated with minimum temperature in shade plots. In sun plots, the number of males and pupae were positively correlated with mean temperature; contrary to Teodoro et al. (2009) we did not observe any influence of relative humidity on CBB populations. CBB populations can also be influenced by biotic factors such as natural enemies. It has been argued that shade plots with higher plant diversity and structural complexity have higher diversity and abundance of natural enemies of CBB: parasitoids, ants, pathogens and birds (Perfecto et al., 1996; Greenberg et al., 1997; Armbrecht and Gallego, 2007; Philpott et al., 2008; Letourneau et al., 2009; Johnson et al., 2010; Larsen and Philpott, 2010). This higher abundance and diversity of natural enemies in shade may partly explain our higher population densities of CBB in sun coffee, and the higher percentage of fruits that were bored but with no CBB individuals inside in shade plots. Armbrecht and Gallego (2007) found that shade coffee in Colombia is more diverse in species of ants than sun coffee; also CBB adults put into spiral traps were more frequently removed from the traps installed in shade coffee than sun coffee. The ant genus Solenopsis was the most abundant in both sun and shade; species of Solenopsis and Azteca are predators of CBB; small Solenopsis can penetrate infested fruits through CBB entry holes and may remove CBB females and their progeny (Armbrecht and Perfecto, 2003; Armbrecht and Gallego, 2007; Pardee and Philpott, 2011). We occasionally observed larvae, pupae and adults of Myrmelachista ants inside the fruits, and in these fruits we did not observe any CBB individuals; these ants may penetrate the fruits to predate the borers and/or use the fruits to reproduce. Like Armbrecht and Gallego (2007) we found Myrmelachista only in shade coffee. 38

55 Another important natural enemy of CBB is the fungus Beauveria bassiana (De la Rosa et al., 1997; Haraprasad et al., 2001; Vega et al., 2008). In shade plot F3 infested fruits with Beauveria were common, and this plot had high CBB infestation but low population size per fruit (Table 2). In shade coffee microbial insecticides are more effective than in sun coffee; in sun plots direct sunlight, higher temperatures and lower humidity impede fungal germination, establishment and survival (Inglis et al., 2001; Pucheta et al., 2006). Fruits from sun plots are more exposed to direct sunlight and the spores' photoinactivation by UV has been described as the most limiting environmental factor (Edgington et al., 2000). Another important biotic factor that influences the CBB population size per fruit is food availability and quality (Awmack and Leather, 2002). In herbivorous insects like the CBB fecundity depends mainly on nutrients offered by the host plant (Awmack and Leather, 2002; Bauerfeind and Fischer, 2005; Jordão et al., 2010; Ross et al., 2011). In coffee, sunlight directly influences food quality and quantity; for example, Vaast et al. (2006) demonstrated that sucrose, chlorogenic acid and trigonelline content are higher in fruits from sun plants. Trigonelline and chlorogenic acids are important nitrogenous compounds in coffee fruits (Duarte et al., 2010). Nitrogen content of food is a determining factor in herbivorous insects' fecundity; for example, when the sycamore aphid (Drepanosiphum platanoidis) feeds on sycamore with increased amino acid content, its fecundity is higher (Awmack and Leather, 2002). The availability of food in sun coffee fruits is higher than in shade coffee fruits: we observed that in our sun plots the fruits were larger, and they had more endosperm which provides more resources for the development of CBB Differences in CBB infestation and population size per fruit between years CBB infestation and population sizes per fruit were higher in 2011 than 2010 (Tables 2, A.2 and A.3). These differences can be explained in the context of the close link between 39

56 the life cycle of CBB and the stages of coffee fruit development, at least in places like Puerto Rico where the blooming period includes a dry season before fruit set (see above). During this time CBB females can survive in rotting fruits on the ground or remaining on the plant until rains trigger their emergence (Baker et al., 1992; Damon, 2000; Barrera, 2005; Barrera et al., 2006); when the rainy season starts, the new fruits are sufficiently developed to be attacked by CBB. In Puerto Rico these rains generally start in March, but in 2010 there was sufficient rain in January to stimulate emergence of CBB females before fruits were available (Figure 6), causing a significant reduction in CBB populations. Also, 2010 was wetter than Cumulative rainfall was 3,096 mm in 2010 and 2,279 mm in 2011, compared with the annual historic average from of 1,972 mm (SERCC, 2007). Previous studies suggested a significant effect of rainfall on CBB dynamics (Constantino et al., 2010; Constantino et al., 2011a; Rodríguez et al., 2013). Constantino et al. (2011a) showed a higher CBB infestation (> 40%) during a dry year in Colombia, while in a wetter year the infestation was 1 12 %. Rainfall also affects the survival of adult and immature CBB in remaining fruits after harvest; higher rainfall induces faster decomposition of the fruits and enhances the proliferation of fungi and bacteria which can infect CBB (Baker et al., 1992; Damon, 2000; Ruiz-Cárdenas and Baker, 2010; Rodríguez et al., 2013) Differences between sun vs. shade on penetration position of colonizing females Position of colonizing females is affected by temperature: at high temperatures (35ºC) in laboratory experiments, females do not reach the endosperm and remain in position B (Jaramillo et al., 2009). However, females from our sun plots were exposed to temperatures up to 38.2 ºC and we found a higher proportion of females in position D (43%). Females reared in the laboratory were exposed to constant temperature, while in the field temperatures 40

57 fluctuate. In sun plots, nighttime temperature ranged from ºC, which may have allowed CBB to start the colonization and reach the endosperm; Sponagel (1994, cited by Damon, 2000) suggested females can reach the endosperm in approximately 8 hours. The stages of coffee fruit development also affect the position of colonizing females (Camilo et al., 2003); we observed highest percentages of females in position B in June (64%) and July (52%), before fruits were mature. (Position B was assigned to females who have penetrated the exocarp, but penetration of the endosperm has not taken place, and the female was present in the fruit). Females in position B are most susceptible to management practices, which indicate that June and July are the best months to increase the presence of natural enemies such as Beauveria bassiana Differences between sun vs. shade on sex ratio We found a significant effect of sun vs. shade on CBB sex ratio, which was more skewed towards females in sun than shade (Table 3). This is the first report that environmental conditions affect CBB sex ratios in the field; in lab experiments CBB sex ratio was not affected by temperatures from 15 to 35 ºC (Jaramillo et al., 2009). Insect sex ratio can be influenced by genetic factors such as sex determination systems (Normark, 2003), extra chromosomal factors including endosymbiotic bacteria (Ebbert and Wrensch, 1993; Weeks et al., 2003; Werren et al., 2008; Kageyama et al., 2012) and environmental factors such as temperature (Nigro et al., 2007; Ross et al., 2011), population densities (Charnov et al., 1981; Kirkendall et al., 1993; King and D'Souza, 2004) and food resources (Rosenfeld and Roberts, 2004; Berndt and Wratten, 2005). The CBB usually has a sex ratio skewed 10:1 towards females. This deviation has been ascribed to its sex determination system, which is functional haplodiploidy (Brun et al., 1995); female-biased sex ratios are common in haplodiploid organisms (Vega et al., 2002; Jaramillo et al., 2006). It has also been suggested that extrachromosomal factors, mainly 41

58 Wolbachia, could play a role in the biased sex ratio of CBB; Vega et al. (2002) reported the presence of this bacterium in CBB from several countries. In our case sun plots were characterized by higher temperatures (Table 1), under sex allocation theory, under which a female who is exposed to extreme environmental factors (such as temperature) should bias her offspring sex ratio towards the sex that suffers less (Ross et al., 2011). Several studies have shown that insects exposed to higher temperatures produce more daughters (Nigro et al., 2007; Ross et al., 2011). In our case, we observed highest number of females in sun plots (Table 3) and a positive correlation between densities of females with mean and median temperature. In addition to female-biased sex ratios, 35-74% of fruits with CBB progeny did not contain males (Table 3). There are two possible explanations; the first is parthenogenesis. It is assumed that CBB reproduction is exclusively sexual (Barrera et al., 1995; Gingerich et al., 1996; Constantino et al., 2011b); however, Muñoz (1989) found that the CBB can be parthenogenetic, since isolated females in the laboratory can produce 55% fertile eggs without fertilization. We suggest that parthenogenesis may occur in the field. Also, in some of these fruits without males, we found eight females or more with other CBB stages: eggs, larvae, pupae. These females might be progeny of the original colonizing female, producing fertile eggs without fecundation. The second possible explanation is associated with male behavior. According to the literature, males are smaller than females; have reduced, degenerate wings, are incapable of flight and never leave the fruit (Brun et al., 1995; Gingerich et al., 1996; Damon, 2000; Jaramillo et al., 2006); However, it is possible that under certain conditions males also can leave the fruit after having fertilized their sisters, which would affect observed sex ratios. 42

59 4.9. Use of traps with ethanol: methanol to monitor CBB populations The traps were successful at catching CBB females; about 375, 000 were caught in 36 traps over 20 months. The capture density increased after the main harvest, at which time the availability of coffee fruits on the trees declines rapidly; we also observed a pronounced peak during March with a capture density of 1700 to 7400 females per trap. March had high rainfall and humidity, which stimulate females emergence and dispersion (Baker et al., 1992; Mathieu et al., 1999; Damon, 2000; Barrera et al., 2006). Previous studies likewise observed a relation between trap capture densities and coffee crop development (Mathieu et al., 1999; Barrera et al., 2006; Vázquez et al., 2012). 5. Conclusions This study provides evidence that the type of coffee management can affect the infestation, population size per fruit and sex ratio of the coffee berry borer H. hampei. We observed higher populations of CBB and a stronger sex ratio skew towards females in sun plots, which were characterized by higher temperatures. Jaramillo et al. (2009) recommended the re-introduction of shade trees into sun coffee plantations to mitigate the effect of climate change on coffee plants and the CBB; we observed that shade reduced the maximum temperatures about 8ºC and increase the humidity in a 20%. Our shade plots were characterized by levels of shade over 78%, which were selected in order to have a good contrast between sun and shade coffee. However, is important maintain a balance between shade and productivity; for commercial production shade must be 30-40% maximum. In this respect our values for shade coffee may not represent most commercial farms. CBB infestation was higher in shade plots, but up to 75% of fruits were only perforated and no CBB offspring were observed inside them. Moreover, shade coffee fruits had fewer individuals than fruits from sun, probably because of resource limitations. 43

60 Additionally, we observed that shade favored the presence of natural enemies of the CBB, such as the fungus Beauveria bassiana; our results suggest the best time to apply the fungus is June and July, when the majority of colonizing females are in susceptible positions. Also, our results suggest that artesanal traps are an efficient method for management of the CBB; these traps were most effective from November to June, a period that included post-harvest, interharvest and early fruit growth. 6. Acknowledgments We are grateful to the growers (Melín Rullán, Anibal and Niulka) for hospitality and permission to conduct experiments on their farms We thank Victor Vega, Fabiola Areces, Luis Ramírez, Ana Pamela Torres, Maria Dávila and Virmarie Díaz for their participation in fieldwork. We thank the personnel of the UPR Agricultural Experimental Substation in Adjuntas, especially Wigmar González and Alvaro Serrano. We also thank the NSF CREST- CATEC program and its director Elvira Cuevas for support of undergraduate students. This project was supported by USDA ARS Specific Cooperative Agreement References Arcila, J., Buhr, L., Bleiholder, H., Hack, H., Meier, U., Wicke, H., Application of the extended BBCH scale for the description of the growth stages of coffee (Coffea spp.). Ann Appl. Biol. 141, Armbrecht, I., Gallego, M.C., Testing ant predation on the coffee berry borer in shaded and sun coffee plantations in Colombia. Entomol. Exp. Appl. 124, Armbrecht, I., Perfecto, I., Litter-twig dwelling ant species richness and predation potential within a forest fragment and neighboring coffee plantations of contrasting habitat quality in Mexico. Agric. Ecosyst. Environ. 97, Awmack, C.S., Leather, S.R., Host plant quality and fecundity in herbivorous insects. Annu. Rev. Entomol. 47, Baker, P., Barrera, J., Rivas, A., Life-history studies of the coffee berry borer (Hypothenemus hampei, Scolytidae) on coffee trees in southern Mexico. J. Appl. Ecol.,

61 Bale, J.S., Masters, G.J., Hodkinson, I.D., Awmack, C., Bezemer, T.M., Brown, V.K., Butterfield, J., Buse, A., Coulson, J.C., Farrar, J., Good, J.E.G., Harrington, R., Hartley, S., Jones, T.H., Lindroth, R.L., Press, M.C., Symrnioudis, I., Watt, A.D., Whittaker, J.B., Herbivory in global climate change research: direct effects of rising temperature on insect herbivores. Global Change Biol. 8, Barrera, J., Gómez, J., Alauzet, C., Can the coffee berry borer (Hypothenemus hampei) reproduce by parthenogenesis?. Entomol. Exp. Appl. 77, Barrera, J.F., Investigación sobre la broca del café en México: logros, retos y perspectivas. Sociedad Mexicana de Entomología y El Colegio de la Frontera Sur. Chiapas, México, Barrera, J.F., Herrera, J., Villacorta, A., García, H., Cruz, L., Barrera, J., Montoya, P., Trampas de metanol-etanol para detección, monitoreo y control de la broca del café Hypothenemus hampei. Sociedad Mexicana de Entomología y El Colegio de la Frontera Sur Manzanillo, Colima, México, Bauerfeind, S.S., Fischer, K., Effects of adult-derived carbohydrates, amino acids and micronutrients on female reproduction in a fruit-feeding butterfly. J. Insect. Physiol. 51, Beer, J., Muschler, D., Kass, D., Somarriba, E., Shade management in coffee and cacao plantations. Agrofor. Syst. 38, Benavides, P., Bustillo, A., Cárdenas, R., Montoya, E., Análisis biológico y económico del manejo integrado de la broca del café en Colombia. Cenicafé 54, Berndt, L.A., Wratten, S.D., Effects of alyssum flowers on the longevity, fecundity, and sex ratio of the leafroller parasitoid Dolichogenidea tasmanica. Biol. Control 32, Brun, L., Stuart, J., Gaudichon, V., Aronstein, K., French-Constant, R., Functional haplodiploidy: a mechanism for the spread of insecticide resistance in an important international insect pest. Proc. Natl. Acad. Sci. 92, Burgiel, S.W., Muir, A.A., Invasive species, climate change and ecosystem-based adaptation: Addresing multiple drivers of global change. Global Invasive Species Programme, Washington, DC, US and Nairobi, Kenya, p. 55. Bustillo, A., Cárdenas, R., Villalba, D., Benavides, P., Orozco, J., Posada, F., Manejo integrado de la broca del café Hypothenemus hampei (Ferrari) en Colombia. Cenicafé, 134. Camilo, J.E., Olivares, F.F., Jiménez, H.A., Fenología y reproducción de la broca del café (Hypothenemus hampei Ferrari) durante el desarrollo del fruto. Agron. Mesoam. 14, Charnov, E., Los-den Hartogh, R., Jones, W., Van den Assem, J., Sex ratio evolution in a variable environment. Nature 289,

62 Constantino, L., Gil, Z., Jaramillo, A., Benavides, P., Bustillo, A., Efecto del fenómeno El Niño y La Niña en la dinámica de infestación de la Broca del Café Hypothenemus hampei (Coleoptera: Curculionidae) en un gradiente altitudinal en la cordillera central Colombiana. Sociedad Colombiana de Entomología, Bogotá, Colombia, 124. Constantino, L., Gil, Z., Jaramillo, A., Benavides, P., Bustillo, A., 2011a. Efecto del cambio y la variabilidad climática en la dinámica de infestación de la broca del café, Hypothenemus hampei en la zona cafetera de Colombia. Sociedad Colombiana de Entomología, Manizales, Colombia, Constantino, L.M., Navarro, L., Berrio, A., Acevedo, F.E., Rubio, D., Benavides, P., 2011b. Aspectos biológicos, morfológicos y genéticos de Hypothenemus obscurus e Hypothenemus hampei (Coleoptera: Curculionidae: Scolytinae). Rev. Colomb. Entomo. l 37, Cure, J.R., Santos, R.H., Moraes, J.C., Vilela, E.F., Gutierrez, A.P., Fenologia e dinâmica populacional da broca do café Hypothenemus hampei (Ferr.) relacionadas às fases de desenvolvimento do fruto. An. Soc. Entomol. Brasil 27, DaMatta, F., Rena, A., Zambolim, L., Ecofisiologia de cafezais sombreados ea pleno sol. O estado da arte de tecnologias na produção de café. Universidade Federal de Viçosa, Viçosa, DaMatta, F., Rodríguez, N., Producción sostenible de cafetales en sistemas agroforestales del Neotrópico: una visión agronómica y eco fisiológica. Agron. Colomb. 25, DaMatta, F., Ronchi, C., Maestri, M., Barros, R., Ecophysiology of coffee growth and production. Braz. J. Plant. Physiol. 19, Damon, A., A review of the biology and control of the coffee berry borer, Hypothenemus hampei (Coleoptera : Scolytidae). B. Entomol. Res. 90, De la Rosa, W., Alatorre, R., Trujillo, J., Barrera, J., Virulence of Beauveria bassiana (Deuteromycetes) strains against the coffee berry borer (Coleoptera: Scolytidae). J. Econ. Entomol. 90, Duarte, G.S., Pereira, A.A., Farah, A., Chlorogenic acids and other relevant compounds in Brazilian coffees processed by semi-dry and wet post-harvesting methods. Food Chem. 118, Ebbert, M.A., Wrensch, D., Endosymbiotic sex ratio distorters in insects and mites. In: Wrensch, D.L., Ebbert, M.A. (Eds.), Evolution and diversity of sex ratio in insects and mites. Chapman Hall, New York, pp Edgington, S., Segura, H., De la Rosa, W., Williams, T., Photoprotection of Beauveria bassiana: testing simple formulations for control of the coffee berry borer. Int. J. Pest Manage 46, Frazer, G.W., Canham, C., Lertzman, K., Gap Light Analyzer (GLA), Version 2.0: Imaging software to extract canopy structure and gap light transmission indices from true- 46

63 colour fisheye photographs, users manual and program documentation. Simon Fraser University, Burnaby, British Columbia, and the Institute of Ecosystem Studies, Millbrook, New York 36. Gay, C., Estrada, F., Conde, C., Eakin, H., Villers, L., Potential impacts of climate change on agriculture: A case of study of coffee production in Veracruz, Mexico. Clim. Change 79, Geromel, C., Ferreira, L.P., Davrieux, F., Guyot, B., Ribeyre, F., Brígida dos Santos Scholz, M., Protasio Pereira, L.F., Vaast, P., Pot, D., Leroy, T., Effects of shade on the development and sugar metabolism of coffee (Coffea arabica L.) fruits. Plant Physiol. and Biochem. 46, Gingerich, D.P., Borsa, P., Suckling, D.M., Brun, L.-O., Inbreeding in the coffee berry borer, Hypothenemus hampei (Coleoptera: Scolytidae) estimated from endosulfan resistance phenotype frequencies. B. Entomol. Res. 86, Greenberg, R., Bichier, P., Angon, A.C., Reitsma, R., Bird populations in shade and sun coffee plantations in central Guatemala. Conserv. Biol. 11, Haraprasad, N., Niranjana, S., Prakash, H., Shetty, H.S., Wahab, S., Beauveria bassiana a potential mycopesticide for the efficient control of coffee berry borer, Hypothenemus hampei (Ferrari) in India. Biocontrol Sci. Tech. 11, Harmsen, E.W., Miller, N.L., Schlegel, N.J., Gonzalez, J., Seasonal climate change impacts on evapotranspiration, precipitation deficit and crop yield in Puerto Rico. Agric Water Manag. 96, Hunter, M.D., Price, P.W., Playing chutes and ladders: heterogeneity and the relative roles of bottom-up and top-down forces in natural communities. Ecology 73, Inglis, G., Goettel, M., Butt, T., Strasser, H., Use of hyphomycetous fungi for managing insect pests. In: Butt, T., Jackson, C., and N. Magan (Ed.), Fungi as biocontrol agents: Progress, Problems and Potential CABI Publishing, Oxon, UK, pp Jaramillo, J., Borgemeister, C., Baker, P., Coffee berry borer Hypothenemus hampei (Coleoptera: Curculionidae): searching for sustainable control strategies. B. Entomol. Res. 96, Jaramillo, J., Chabi-Olaye, A., Kamonjo, C., Jaramillo, A., Vega, F.E., Poehling, H.M., Borgemeister, C., Thermal Tolerance of the Coffee Berry Borer Hypothenemus hampei: Predictions of Climate Change Impact on a Tropical Insect Pest. Plos One 4, e6487. Jaramillo, J., Muchugu, E., Vega, F.E., Davis, A., Borgemeister, C., Chabi-Olaye, A., Some like it hot: The influence and implications of climatic change on coffee berry borer (Hypothenemus hampei) and coffee production in east Africa. Plos One 4, Johnson, M., Kellermann, J., Stercho, A., Pest reduction services by birds in shade and sun coffee in Jamaica. Anim. Conserv. 13,

64 Jordão, A.L., Nakano, O., Janeiro, V., Adult carbohydrate feeding affects reproduction of Phthorimaea operculella (Zeller)(Lepidoptera: Gelechiidae). Neotrop. Entomol. 39, Kageyama, D., Narita, S., Watanabe, M., Insect sex determination manipulated by their endosymbionts: incidences, mechanisms and implications. Insects 3, King, B., D'Souza, J., Effects of constrained females on offspring sex ratios of Nasonia vitripennis in relation to local mate competition theory. Can. J. Zool. 82, Kirkendall, L., Wrensch, D., Ebbert, M., Ecology and evolution of biased sex ratios in bark and ambrosia beetles. In: Wrensch, D.L., Ebbert, M.A. (Eds.), Evolution and diversity of sex ratio in insects and mites. Chapman Hall, New York, pp Larsen, A., Philpott, S.M., Twig Nesting Ants: The Hidden Predators of the Coffee Berry Borer in Chiapas, Mexico. Biotropica 42, Letourneau, D.K., Jedlicka, J.A., Bothwell, S.G., Moreno, C.R., Effects of natural enemy biodiversity on the suppression of arthropod herbivores in terrestrial ecosystems. Annu Rev. Ecol. Evol. Syst. 40, Mathieu, F., Brun, L.-O., Frerot, B., Suckling, D., Frampton, C., Progression in field infestation is linked with trapping of coffee berry borer, Hypothenemus hampei (Col., Scolytidae). J. Appl. Entomol. 123, Muñoz, R., Ciclo biológico y reproducción partenogenética de la broca del fruto de cafeto Hypothenemus hampei (Ferr.). Turrialba 39, NAPPO, Detection of Coffee Berry Borer, Hypothenemus hampei, in Puerto Rico - United States. Official Pest Reports ( Nigro, R.G., Campos, M.C.C., Perondini, A.L.P., Temperature and the progeny sexratio in Sciara ocellaris (Diptera, Sciaridae). Genet. Mol. Biol. 30, Normark, B.B., The evolution of alternative genetic systems in insects. Annu. Rev. Entomol. 48, Pardee, G.L., Philpott, S.M., Cascading indirect effects in a coffee agroecosystem: effects of parasitic phorid flies on ants and the coffee berry borer in a high-shade and lowshade habitat. Environ. Entomol. 40, Perfecto, I., Rice, R.A., Greenberg, R., Van der Voort, M.E., Shade Coffee: A Disappearing Refuge for Biodiversity. BioScience 46, Philpott, S.M., Arendt, W.J., Armbrecht, I., Bichier, P., Diestch, T.V., Gordon, C., Greenberg, R., Perfecto, I., Reynoso, R., Soto-Pinto, L., Biodiversity loss in Latin American coffee landscapes: review of the evidence on ants, birds, and trees. Conserv. Biol. 22, Pucheta, M., Flores Macías, A., Rodríguez Navarro, S., De La Torre, M., Mecanismo de acción de los hongos entomopatógenos. Interciencia 31,

65 Rodríguez, D., Cure, J.R., Gutierrez, A.P., Cotes, J.M., Cantor, F., A coffee agroecosystem model: II. Dynamics of coffee berry borer. Ecol. Modell. 248, Rosenfeld, C.S., Roberts, R.M., Maternal diet and other factors affecting offspring sex ratio: a review. Biol. Reprod. 71, Ross, L., Dealey, E.J., Beukeboom, L.W., Shuker, D.M., Temperature, age of mating and starvation determine the role of maternal effects on sex allocation in the mealybug Planococcus citri. Behav. Ecol. Sociobiol. 65, Ruiz-Cárdenas, R., Baker, P., Life table of Hypothenemus hampei (Ferrari) in relation to coffee berry phenology under Colombian field conditions. Sci. Agr. 67, SERCC, Historical Climate Summaries for Puerto Rico and the US Virgin Island. University of North Carolina at Chapel Hill. Available for download at Silva, N., Frizzas, M., Oliveira, C., Seasonality in insect abundance in the "Cerrado" of Goiás State, Brazil. Rev. Bras. Entomol. 55, Soto-Pinto, L., Perfecto, I., Caballero-Nieto, J., Shade over coffee: its effects on berry borer, leaf rust and spontaneous herbs in Chiapas, Mexico. Agroforest Syst. 55, Sponagel, K., La broca del café Hypothenemus hampei en plantaciones de café robusta en la Amazonía Ecuatoriana. Wissenschaftlicher, Fachverlag, Giessen, Ger, p Team, R., Develoment Core. "R Foundation for Statistical Computing". Vienna, Austria 482. Teodoro, A., Klein, A.-M., Tscharntke, T., Environmentally mediated coffee pest densities in relation to agroforestry management, using hierarchical partitioning analyses. Agric. Ecosyst. Environ. 125, Teodoro, A., Klein, A., Reist, P., Tscharntke, T., Agroforestry management affects coffee pests contingent on season and developmental stage. Agric. For. Entomol. 11, Uemura-Lima, D.H., Ventura, M.U., Mikami, A.Y., Da Silva, F.C., Morales, L., Responses of Coffee berry borer, Hypothenemus hampei (Ferrari)(Coleoptera: Scolytidae), to vertical distribution of methanol: ethanol traps. Neotrop. Entomol. 39, Vaast, P., Bertrand, B., Perriot, J.J., Guyot, B., Genard, M., Fruit thinning and shade improve bean characteristics and beverage quality of coffee (Coffea arabica L.) under optimal conditions. J. Sci. Food Agric. 86, Valdés, B.C., Ravelo, H.G., Caracterización de la dinámica poblacional de la broca del café (Hypothenemus hampei (Ferrari)(Curculionidae: Scolytinae) durante el desarrollo de los frutos. Centro Agrícola 33,

66 Vázquez, L.L., Alfonso, J., Ramos, Y., Martínez, A., Moreno, D., Matienzo, Y., Relaciones de Hypothenemus hampei Ferrari (Coleoptera: Curculionidae: Scolytinae) con el suelo del cafetal como base para su manejo agroecológico. Agroecología 7, Vega, F.E., Benavides, P., Stuart, J.A., O'Neill, S.L., Wolbachia infection in the coffee berry borer (Coleoptera: Scolytidae). Ann. Entomol. Soc. Am. 95, Vega, F.E., Infante, F., Castillo, A., Jaramillo, J., The coffee berry borer, Hypothenemus hampei (Ferrari)(Coleoptera: Curculionidae): a short review, with recent findings and future research directions. Terr. Arthropod. Rev. 2, Vega, F.E., Posada, F., Aime, M.C., Pava-Ripoll, M., Infante, F., Rehner, S.A., Entomopathogenic fungal endophytes. Biol. Control 46, Webge, K., Cilas, C., Decazy, B., Alauzet, C., Dufour, B., Estimation of Production Losses Caused by the Coffee Berry Borer (Coleoptera: Scolytidae) and Calculation of an Economic Damage Threshold in Togolese Coffee Plots. J. Econ. Entomol. 96, Weeks, A.R., Velten, R., Stouthamer, R., Incidence of a New Sex ratio distorting Endosymbiotic Bacterium among arthropods. Proc. R. Soc. Lond. [Biol] 270, Werren, J.H., Baldo, L., Clark, M.E., Wolbachia: master manipulators of invertebrate biology. Nat. Rev. Microbiol. 6, Wrigley, G., Coffee. Tropical Agriculture Series. Longman Scientific & Technical and John Wiley & Sons, New York, pp

67 Table 1. Coordinates and environmental characteristics of the study plots Farm Coordinates Temperature ( C) Min - Max Humidity (%) Min - Max Canopy cover (%) Number of coffee plants in 20m transect Sun Shade Sun Shade Shade Sun Shade F1 18º N % º W F2 18º N % º W F3 17º N % º23 15 W

68 Table 2. Infestation and population size per fruit of the coffee berry borer Hypothenemus hampei varies between sun vs. shade coffee, years and farms Year 2010 Shade coffee Sun coffee Variable F1 F2 F1 F2 Statistic values Mean ± s.e. of % infestation per plot 11.98± ± ± ±0.57 Z= 28.27/ P < b Mean ± s.e. of total # CBB per fruit 4.81± ± ± ±0.43 Z= 9.95/ P < b % infected fruits without CBB offspring a 66.9 (221/330) 54.2(179/330) 48.1(159/330) 45.8 (151/330) F = 2.78/ P < c Year 2011 F1 F3 F1 F3 Statistic values Mean ± s.e. of % infestation per plot 35.67± ± ± ±1.74 Z= 29.04/ P < b Mean ± s.e. of total # CBB per fruit 6.78± ± ± ±0.49 Z= 5.07/ P < b % infected fruits without CBB offspring a 63.3 (228/360) 79.4 (285/360) 49.7 (179/360) 55.8 (201/360) F = 8.81 / P = c a Proportion of bored fruits with no CBB offspring (no eggs, larvae, pupae, juveniles or adults). b Values based on generalized linear models c Values based on ANOVA tests 52

69 Table 3. Total number of females and males, percentage of fruits without males and sex ratio (mean ±se) of sex ratio of the coffee berry borer Hypothenemus hampei between sun vs. shade coffee, years and farms Year 2010 Shade coffee Sun coffee Statistic values Variable F1 F2 F1 F2 Total number of females d Z= 2.82 / P = b Number of males d Z= 0.82 / P = b Percentage of fruits without males e F=0.49 / P = c Sex ratio 1 f 0.79± ± ± ±0.02 F=7.07 / P = c Year 2011 F1 F3 F1 F3 Statistic values Total number of females d Z= 1.42 / P = b Total number of males d Z= 2.28 / P = b Percentage of fruits without males e F=0.31 / P = c Sex ratio 1 f 0.81± ± ± ±0.06 F= 4.30/ P =0.04 c b Values based on generalized linear models c Values based on ANOVA tests d Total number of individuals counted; in the case of females includes all females (colonizing females and females from progeny). e (# fruits without males/ # fruits with progeny of juveniles and adults) x100 f Calculated as the proportion of females in fruits with progeny and only includes those with newly emerged and adults, we did not include the fruits in which we observed only the colonizing females and females with immature progeny (eggs, larvae and pupae). 53

70 Shade Sun A B Mean % Infestation Mean % Infestation June July August September October November 0 June July August September October November 25 Shade Sun 2010 C D Mean CBB per fruit ab abc abcd c bcd abcd cd abc Mean CBB per fruit a abc abcd abcd cd bcd abcd abcd August September October November 0 August September October November Figure 1. Monthly changes in infestation (A and B) and number of CBB per fruit (C and D) of the coffee berry borer (Hypothenemus hampei F.) in shade vs. sun coffee. Means ± SE are shown. Different letters indicate significant differences in the number of CBB individuals per infested fruit between sun and shade coffee based on ANOVAs and post-hoc Tukey tests (P<0.05); analyses included only data from fruits with reproduction and data were transformed to log (x+1). Note that scale for % infestation differs; infestation reached 4x higher levels in 2011 than in

71 10 Shade-2010 Sun-2010 Shade-2011 Sun Mean CBB per fruit 6 4 *** *** *** *** * 2 *** 0 Eggs Larvae Pupae Juveniles Adults Figure 2.Population structure of the coffee berry borer (Hypothenemus hampei F.). Means + SE are shown. Asterisks indicate significant differences in between sun vs. shade and years (generalized linear model significance codes ***= P < 0.001, **= P < 0.01, * =P < 0.05, =P < 0.1) 55

72 PA PB PC PD Shade coffee Sun coffee Proportion of females Proportion of females July August September October November December 0.0 July August September October November December Shade coffee Sun coffee Proportion of females June July August September October November December Proportion of females June July August September October November December Figure 3. Proportion of the coffee berry borer (Hypothenemus hampei F.) colonizing females in positions A, B, C and D inside fruits during the growing season, and relation to sun vs. shade coffee and year. 56

73 Interharvest/ Blooming Shade - coffee Fruit growth and maturation Sun - coffee Harvest Post-harvest/ senescence Average females captured Febreaury March April May June July August September October November December Month Figure 4. Monthly averages of captured females of Hypothenemus hampei in traps in sun vs. shade coffee. Stages of coffee phenology in Puerto Rico are shown above the bars. 57

74 Historical ( ) 600 Total Montlhy Precipitation (mm) Jan Figure 5.Total monthly precipitation in 2010 and 2011 for Adjuntas, Puerto Rico with the historical monthly average ( ) (Source: SERCC, Historical Climate Summaries for Puerto Rico and U.S. Virgin Islands) 0 Feb March April May June July August September October November December 58

75 8. Appendices Table A.1. Species and family of trees associated with shade plots Farm Species and Family Family F1 F2 F3 Guarea guidonia Cyathea arborea Inga vera Thespesia grandiflora Spathodea campanulata Miconia prasina Cordia sulcata Cecropia schreberiana Musa x paradisiaca Tabebuia heterophylla Inga vera Inga laurina Roystonea borinquena Spathodea campanulata Artocarpus altilis Musa paradisiaca Citrus paradisi Inga vera Inga laurina Tabebuia heterophylla Artocarpus altilis Citrus sinensis Spathodea campanulata Tectona grandis Andira inermis Meliaceae Cyatheaceae Fabaceae Malvaceae Bignoniaceae Melastomataceae Boraginaceae Urticaceae Musaceae Bignoniaceae Fabaceae Fabaceae Arecaceae Bignoniaceae Moraceae Musaceae Rutaceae Fabaceae Fabaceae Bignoniaceae Moraceae Rutaceae Bignoniaceae Lamiaceae Fabaceae 59

76 Table A.1. Results of generalized linear model evaluating the relationship between infestation of the coffee berry borer Hypothenemus hampei and type of coffee (sun vs. shade), year and farm Effects Estimate Standard error Z - value P Intercept <0.0001*** Type of coffee (shade vs. sun) <0.0001*** Year (2010 vs. 2011) <0.0001*** Farm (F1 vs. F2) <0.0001*** Farm (F1 vs. F3) <0.0001*** Increase in infestation through time vs Year (2010 vs. 2011): Type of coffee <0.0001*** (shade vs. sun) Type of coffee (shade vs. sun) : Farm <0.0001*** (F1 vs. F2) Type of coffee (shade vs. sun) : Farm <0.0001*** (F1 vs. F3) 60

77 Table A.2. Results of generalized linear model evaluating the relationship between the number of coffee berry borers per fruit and sun vs. shade, Year and Farm Effects Estimate Standard error Z - value P Intercept < *** Type of coffee (Shade vs. Sun) < *** Year (2010 vs. 2011) < *** Farm (F1 vs. F2) < *** Farm (F1 vs. F3) < *** Year (2010 vs. 2011): Type of < *** coffee (shade vs. sun) Type of coffee (shade vs. sun) : < *** Farm (F1 vs. F2) Type of coffee (shade vs. sun) : < *** Farm (F1 vs. F3) 61

78 A B C D Figure A.1. Graphical representation for a coffee farm in Adjuntas, Puerto Rico. F1 in this study (A) Profile for shade, (B) Aerial view for shade (C) Profile for sun (D) Aerial view for sun. Here, we are showing the number of coffee plants in a 20m transect and the location of Onset HOBO dataloggers. 62

79 CHAPTER 3 Wolbachia affects reproduction, sex ratio and population dynamics of the Coffee Berry Borer Hypothenemus hampei (Coleoptera: Curculionidae) Mariño-Cárdenas, Y., Verle-Rodrigues, J.C. and Bayman, P. In Revision. Wolbachia affects reproduction, sex ratio and population dynamics of the Coffee Berry Borer Hypothenemus hampei (Coleoptera: Curculionidae) 63

80 Abstract Wolbachia are widely distributed endosymbiotic bacteria that influence the reproduction and fitness of many insects. In the case of the coffee berry borer (CBB, Hypothenemus hampei, the most devastating coffee pest worldwide) Wolbachia has been reported in CBB from several countries. In this study, the infection of Wolbachia in CBB from Puerto Rico was detected by sequencing the wsp gene. Tetracycline (0.1% w/v) reduced significantly the proportion of Wolbachia in the CBB metagenome from 0.49% to 0.04%. This reduction was accompanied by changes in CBB reproduction, sex ratio and population dynamics. Feeding with tetracycline for three generations females produced significantly fewer total progeny, fewer ovipositing females and fewer eggs per female. The sex ratio was less skewed towards females in progeny of females treated with tetracycline. However, there was no effect of antibiotic treatment on the abundance of males. This does not support the idea that Wolbachia is involved in sex determination of CBB. CBB populations treated with tetracycline had lower population growth rate or λ= 1.07 compared with 1.11 for control. The mean generation time (T) increased by 6.2 days, so females reared with tetracycline would produce 8.1 generations per year compared with 9.4 for controls. Our results suggest that Wolbachia or other endosymbionts contribute to the reproductive success of the CBB, which opens the door to the possibility of methods of CBB control mediated by Wolbachia management. 1. Introduction The coffee berry borer (CBB) Hypothenemus hampei Ferrari (Coleoptera: Curculionidae) is the most devastating coffee pest worldwide (Damon, 2000; Soto-Pinto et al., 2002; Jaramillo et al., 2006; Vega et al., 2009). The CBB has co-existed with coffee plants for a very long time and depends principally on coffee fruits for food, reproduction, 64

81 and development (Damon, 2000). The entire life cycle occurs inside the fruit; young females are fertilized by sibling males, and then exit the fruit to find new fruits to colonize and start oviposition (Damon, 2000; Vega et al., 2002; Vega et al., 2006; Vega et al., 2009). The CBB usually has a sex ratio skewed 10:1 towards females. This deviation has been ascribed to its sex determination system, which is functional haplodiploidy (Brun et al., 1995). In this system both sexes develop from fertilized eggs, but paternally derived chromosomes are not transmitted to males because they are eliminated from their germline at various stages (Fry et al., 2004; Jia et al., 2009). The female-biased sex ratio in organisms with functional haplodiploidy is a consequence of the disruption and death of the embryos when paternal chromosomes are lost (Dong et al., 2006). It has also been suggested that endosymbiotic bacteria could play a role in the biased sex ratio of CBB; some endosymbionts are known as reproductive parasites, able to induce reproductive abnormalities in their hosts, including cytoplasmic incompatibility, parthenogenesis, feminization and male killing (Zhou et al., 1998; Jeyaprakash and Hoy, 2000; Lo et al., 2002; Werren et al., 2008). Wolbachia is the most extensively studied genus of endosymbiotic bacteria, implicated in all types of described reproductive abnormalities. Wolbachia has been detected in CBB adult females from several countries and is thought to induce cytoplasmic incompatibility (Vega et al., 2002). Wolbachia distort sex ratios to favor the presence of females to enhance its own transmission. It may also be necessary for normal reproduction of various insects (Dedeine et al., 2001; Fry et al., 2004; Dong et al., 2006; Zchori Fein et al., 2006; Son et al., 2008; Jia et al., 2009; Chen et al., 2012). For example, in some species of beetles (Family Curculionidae) the elimination of Wolbachia with antibiotic treatments significantly reduced fecundity, and in some cases eggs laid by cured females were not viable (Zchori Fein et al., 2006; Chen et al., 2012). 65

82 In the case of CBB the effect of possible elimination of Wolbachia on sex ratio and other reproductive traits has been not tested. The objectives of this study were to determine the infection status of Wolbachia in females from Puerto Rico, and test the effect of removal of Wolbachia on CBB reproduction. We asked the following questions: 1) Does Wolbachia affect the sex ratio of CBB? We hypothesized that the elimination of Wolbachia with antibiotics will change sex ratio in CBB, from 10:1 female: male to closer to 1:1, during several generations. 2) Does Wolbachia affect reproduction of CBB? Based on data from other Curculionidae, we predicted that the elimination of Wolbachia with antibiotics will reduce reproductive success of CBB. 2. Materials and Methods 2.1. Insect Rearing. Coffee berry borers were reared for ten consecutive generations in the Cenibroca artificial diet (Portilla, 1999). Ingredients, quantities and names of suppliers are listed in Table A ml of diet was placed in sterile, clear plastic (23 x 92 mm, with plug of bonded dense weave cellulose acetate, Genessee Scientific, San Diego, CA). Insects were reared in dark in a growth chamber (Model 818, Thermo Scientific, Dubuque, IA) set at 25ºC and 80 96% relative humidity. All experiments were done in the Center for Excellence in Quarantine and Invasive Species at University of Puerto Rico Agricultural Station in Río Piedras, PR. Adult CBB females were collected from infested fruits from coffee plantations in Adjuntas, Puerto Rico (18º N, 66º W, 527 m a.s.l). Fruits were carefully dissected with aid of a stereoscope, and females were removed and surface-disinfected by 66

83 submersion in 1% sodium hypochlorite for 1 min, followed by two rinses with sterile distilled water. For all generations three females were placed in each vial. After each generation adult females were removed from the diet and mixed (randomized) before disinfection and transferal to fresh diet to begin the next generation Antibiotic Treatments For direct and continuous feeding of larvae and adult females, tetracycline hydrochloride (Fisher Bioreagents) and penicillin G (Fisher Bioreagents) 0.1% (w/v) were added to the Cenibroca diet. Tetracycline is efficient at removing Wolbachia (Dedeine et al., 2001; Van Lenteren, 2003; Gotoh et al., 2007); penicillin is not active against Wolbachia (Taylor et al., 2000; Fenollar et al., 2003) but is sometimes is used to control bacterial growth in insect diets (Van Lenteren, 2003); it was used as a parallel control Detection of Wolbachia DNA Extraction, PCR Amplification and DNA sequencing DNA was extracted from entire bodies of adult females obtained from infested fruits in Adjuntas, Puerto Rico, and from the Cenibroca diet with and without antibiotics. Insects were conserved dry and stored in the freezer -20ºC; total genomic DNA was extracted using the DNeasy Plant Mini Kit (Qiagen Sciences). The presence of Wolbachia in the CBB females was determined by amplification of part of the wsp gene (primers wsp-f: 5 GGGTCCAATAAGTGATGAAGAAAC 3 and wsp- R: 5 TTAAAACGCTACTCCAGCTTCTGC 3 ; (Kondo et al., 2002) with annealing at 58 ºC. Positive controls for PCRs were total genomic DNA of wild Drosophila melanogaster and negative controls were dh 2 O. 67

84 PCR products were cleaned on QIAquick columns (Qiagen, Inc.) and sequenced in both directions in the University of Puerto Rico Sequencing and Genotyping Facility (UPR SGF). The resulting sequences were assembled, manually examined for errors using CodonCode Aligner (version 5.1.5) and BLASTed to compare the CBB sequences with other Wolbachia sequences in GenBank Phylogenetic classification in Wolbachia supergroups Phylogenetic analysis was used to assign sequences strains from CBB to Wolbachia supergroups. Three Wolbachia wsp sequences from this study were compared to GenBank reference sequences from 41 species of insects, including CBB (Vega et al., 2002), (Table 1). Sequences were aligned with ClustalW in Mesquite (Maddison and Maddison, 2001). A phylogenetic tree was constructed with Mega 6.0 (Tamura et al., 2013) using neighbor joining, a maximum composite likelihood model and bootstrap analysis with 1000 replications Effect of antibiotic treatments of CBB Reproduction and Sex Ratio The number of eggs, larvae, pupae, juveniles and adults in 20 vials per treatment were counted by carefully dissecting the diet from each tube. The sex ratio was calculated as the number of CBB females divided by the total number of juveniles and adults (the CBB stages where individuals can be sexed by visual inspection). To determine the effect of the elimination or reduction of Wolbachia on CBB fecundity, females from the fifth generation of the three treatments were transferred individually to vials with artificial diet (n=50 per treatment), and oviposition and number of eggs per female was registered daily for two months. 68

85 2.5. Evaluation of Wolbachia infection Proportions of Wolbachia in the microbiota of adult females from control and tetracycline diets (after eleven generations) were determined using metagenome data. Briefly the methodology used was: pooled DNA from five females, with four replicates per treatment, was extracted using DNeasy Plant Mini Kit (Qiagen Sciences). Bacterial 16S rrna gene was amplified using the hypervariable region V4 with primers 515-F and 806-R (Caporaso et al., 2012) with the addition of barcoding sequences and Illumina adapters. Sequencing was done on the Illumina MiSeq platform. Sequences were assigned to Wolbachia using OTU tables generated with QIIME (Caporaso et al., 2010). Other aspects of the metagenomes will be described elsewhere Life Tables of H. hampei Five cohorts of 20 eggs each were established for each type of diet. Eggs were maintained in dark at 25ºC and a relative humidity between 80 96%. Each cohort was evaluated daily for 50 days at 24-h intervals to determine CBB s survival and the transitions to the next stages of development. Also, the development time from egg to adult and the duration of each stage were determined. For population dynamics analysis the basic life cycle for the CBB was defined as egg, larva, pre-pupa, pupa, juvenile and adult (Figure 1). The probabilities of the projection matrix were calculated at intervals of six days: (Gi) probability of growth or molt between stages, (Pi) probability to surviving and remaining in the same stage, and (Fi) is the fertility of adult females, these values were based on the data obtained in this study for oviposition of fifty females (Figure 4). The probability of adult survival was estimated as 0.99, because observed mortality of adults was very rare once they emerged; several studies also reported low mortality rates in field and laboratory experiments (Bergamin, 1943; Romero and Cortina, 2007; Ruiz-Cárdenas and Baker, 2010). 69

86 Figure 1.Basic life cycle of the coffee berry borer Hypothenemus hampei modeled as a population projection matrix. Gi: probability of transition between stages, Pi: probability of surviving and remaining in the same stage, Fi: fertility of adult females. (1) Eggs, (2) Larvae, (3) Pre-pupae, (4) Pupae, (5) Juveniles, (6) Adults. For the projection matrix, variations in the size and age structure of a population N through time t and t+1 were computed from the equation: N t+1 = AN t Where A is a population matrix and N is a vector describing the age of structured population. Thus the projection matrix was: Based on the projection matrix for each cohort in the three types of diet we calculated: the population growth rate (λ) and mean generation time (T) (Caswell, 2001). All calculations were carried out with PopTools (version 3.2) Statistical analysis 70

87 The abundance of individuals in various stages of development of the coffee berry borer and the number of eggs laid per female were estimated using a log-lineal model with a Poisson distribution; type of diet was defined as an independent variable. Proportion tests were used to compare differences in the numbers of females that oviposited among types of diet and to determine differences in the number of reads of Wolbachia found in adult females between tetracycline vs. control diets. One way ANOVAs followed by post-hoc Tukey tests (significance level P=0.05) were used to test differences among types of diet on the sex ratio, population parameters and developmental time from egg to adult and the duration of each stage of development. All analyses were performed in R (Team, 2013). 3. Results 3.1. Wolbachia detection and classification A 750-bp fragment was amplified and sequenced from the Wolbachia wsp gene using the total DNA from CBB field-collected females from shade coffee, field-collected females from sun coffee, and females reared in the lab on the Cenibroca diet. The positive control (Drosophila) amplified a band of similar size; no product amplified in any of the negative controls. Sequences of seven wsp fragments exhibited >99% sequence similarity to Wolbachia from Drosophila melanogaster (JF505383). The wsp sequences of H. hampei from Puerto Rico grouped with those of Wolbachia supergroup A. The three CBB sequences used in the analysis formed a separate subclade, closely related to that from the parasitoid Tachinaephagus zealandicus (DQ380884) (Kyei- Poku et al., 2006) (Figure 2). In contrast, a wsp consensus sequence from H. hampei from India and Brazil (AF389084) grouped with supergroup B (Vega et al. (2002), and with others from adzuki bean beetle Callasobruchus chinensis (AB038339) (Kondo et al., 2002) 71

88 3.3. Effect of antibiotic treatment on Wolbachia infection and CBB reproduction After ten generations the proportion of Wolbachia in the CBB microbiota was significantly lower in females from the diet with tetracycline than from the control diet (0.04% vs. 0.49%, χ 2 = , P <0.001). The log-lineal model gave a good fit for the effects of diet and generation on CBB total population (χ 2 7 = 4907, P <0.0001). In general, the estimated population was significantly higher in control and penicillin diets than in tetracycline (Table A.2). The estimated population per generation increased independently of type of diet through the generations (Table A.2). It was not until the third generation that the antibiotic affected the abundance of some developmental stages and the total population of CBB. After the third generation females reared with tetracycline produced significantly fewer progeny, and in the F 10 generation less than half as many individuals were observed compared with control and penicillin diets (Figure 3 and 4). The estimated abundance of individuals at all stages of development varied significantly among diets (glm, Poisson errors, eggs χ 2 2 = 36.2, Z= 232.2, P <0.0001; larvae χ 2 2 = 114.9, Z= 194.7, P < , pupae χ 2 2 = 92, Z= 107.9, juveniles χ 2 2 = 4.5, Z= 52.20, P = < and adults χ 2 2 = 263.9, Z=295.4, P< ). In general, the numbers of individuals in all stages of development was significantly lower in diet with tetracycline than in control (Figure 4 and Table A.3). The proportion of females that oviposited did not differ significantly among types of diet (Fig. 5; 68% control, 66% penicillin and 38% tetracycline, χ 2 = 3.90, df=2, P =0.14). However, the number of eggs per female was significantly lower in females treated with Tetracycline: on average each female laid 3.45 eggs compared with in control and in penicillin (glm, Poisson errors, eggs χ 2 2 = 832.9, Z= 86.79, P <0.0001) (Figure 5). 72

89 Overall significant differences were found in sex ratio between types of diet (F= 12.61, df=2, P < ). The sex ratio was more skewed towards females in control and penicillin diets than tetracycline (0.88 for control and penicillin and 0.84 for tetracycline). These differences in sex ratio were significant from the F 3 generation; no differences were observed in F 1 (F= 1.05, df=2, P = 0.35) or F 2 (F= 1.11, df=2, P =0.33). The tendency of sex ratios more skewed towards females was observed for all generations in diets control and penicillin diets (Table 2). However, no differences were observed in the number of males among diets (control vs. penicillin: χ = 2.88, Z= 1.32, P = 0.25; control vs. tetracycline: χ 2 = 2.88, Z= , P = 0.59) Effect of antibiotic treatment of life table parameters of H. hampei The duration of the juvenile stage was significantly longer in tetracycline that in control and penicillin (F= 3.34, df=2, P = 0.03). The duration of remaining stages did not differ between diets (eggs: F= 1.91, df=2, P = 0.15; larva: F= 0.89, df=2, P =0.41; pre-pupa: F= 0.86, df=2, P =0.42 and pupa: F= 1.59, df=2, P =0.20. Egg-to-adult developmental time did not differ significantly between diets (F= 0.95, df=2, P = 0.38), but was slightly higher for individuals reared with tetracycline (Table 3). Life table parameters for the three diets are presented in Table 3; no differences were observed between parameters: poulation growth rate λ (F= 3.13, df=2, P = 0.08) and mean generation time T (F= 0.77, df=2, P = 0.48) among diets. For all diets the population growth was >1. Despite, we do not found significant differences in the mean generation (T), we observed that tetracycline treatment increased the time by 6.2 days, which means that females reared with tetracycline will produce fewer generations per year (Table 3). 73

90 4. Discussion The main results of the present study can be summarized as follows: (1) CBB females from Puerto Rico are infected with Wolbachia, at least some strains of which belong to supergroup A, unlike the previously consensus sequenced CBB strain from India and Brazil. (2) Tetracycline affects the reproduction of CBB: females reared with this antibiotic produced less than half the progeny compared with females reared with penicillin and controls without antibiotics. Also, in females treated with tetracycline fertility was reduced to a quarter part, and fewer females oviposited. (3) Wolbachia probably is not involved in sex determination of CBB; we observed no significant differences of antibiotic treatments on the number of males produced by females treated with tetracycline vs. controls, even though the sex ratio was slightly less skewed towards females. (4) The treatment with tetracycline affected the life table parameters evaluated; CBB populations treated with tetracycline had lower population growth rate or λ= 1.07 compared with 1.11 for control. The mean generation time (T) increased by 6.2 days, so females reared with tetracycline would produce 8.1 generations per year compared with 9.4 for controls Wolbachia in insects and status in the CBB Wolbachia are widespread and common insect endosymbionts. Up to 66% of insect species may have relations with Wolbachia (Hilgenboecker et al., 2008). Wolbachia was detected by PCR in adult CBB females from Benin, Brazil, Colombia, Ecuador, El Salvador, Honduras, India, Kenya, Mexico, Nicaragua, and Uguanda (Vega et al., 2002), and in this study Wolbachia was detected in adult females from Puerto Rico. Data from CBB microbiota suggest a very low proportion of Wolbachia in the CBB, these proportions were 0.49% and 0.16% of 16S sequences in adult females reared in artificial diets and collected from field respectively (Mariño et al., In revision). This low proportion of Wolbachia may partly explain why we were unable to amplify Wolbachia in 90% of trials; 74

91 Vega et al. (2002) also suggested low titer of Wolbachia may explain the lack of amplification in some samples Phylogenetic classification of Wolbachia supergroups For Wolbachia supergroup classification the wsp gene was preferred by Zhou et al. (1998); the wsp gene is evolving much faster than 16S rdna and ftsz genes. However, phylogenies using these three genes concur in the four major supergroups of Wolbachia A and B are found exclusively in arthropods, C and D in nematodes (Werren and Windsor, 2000; Lo et al., 2002). Wolbachia detected in CBB females from Puerto Rico belonged to supergroup A, while the consensus sequence from India and Brazil belonged to supergroup B (Vega et al., 2002) (Figure 2). Wolbachia in Coleoptera has been described for A and B supergroups; in some cases two individuals of a single species are infected with different supergroups (Lachowska et al., 2010) Antibiotic treatments and their effect on CBB reproduction and fitness Tetracycline treatment is effective curing endosymbiotic Wolbachia infections in insects (Werren, 1997; Stouthamer et al., 1999). Tetracycline has been used to determine the role of Wolbachia in the reproduction and fitness of its insect hosts (Dedeine et al., 2001; Dong et al., 2006; Zchori Fein et al., 2006; Dong et al., 2007; Son et al., 2008; Chen et al., 2012). The effect of tetracycline on reproduction and fitness of Wolbachia -infected insects can result from either: (i) indirect effects on the insect by curing or reducing Wolbachia infection or (ii) direct effects on host physiology by toxicity or production of toxic metabolites, independent of its effect on Wolbachia (Dedeine et al., 2001). To reduce the possibility that observed effects on CBB reproduction was a result of direct antibiotic toxicity, we included a parallel treatment with penicillin G; penicillin G is not effective against Wolbachia (Stouthamer et al., 1999; Taylor et al., 2000; Fenollar et al., 2003).Tetracycline 75

92 and penicillin differ in their modes of action and ranges of activity, however, both antibiotics can affect directly the fecundity and survival of insects (Srivastava and Auclair, 1976; Büyükgüzel and Yazgan, 2001; Dale and Welburn, 2001; Zchori Fein et al., 2006). Our results showed that females reared in both penicillin and control diets produced a similar number of individuals in all stages of development, and there were no significant effects of penicillin on reproduction and life table parameters (Figure 3-5 and Tables A.2 and A.3). Therefore, the reduced reproduction with tetracycline can probably be attributed to the effect of reduction of Wolbachia (or perhaps other bacteria). 0.1% (w/v) tetracycline significantly reduced Wolbachia infection, but did not remove it completely, even after ten generations of continuous treatment. However, in other studies 0.1% or less tetracycline was enough to remove Wolbachia from infected hosts (Breeuwer, 1997; Kageyama and Traut, 2004; Charlat et al., 2007; Gotoh et al., 2007). Even though the infection was not completely cured, our results strongly suggest a relationship between Wolbachia and the successful reproduction of CBB females. After F 3 the reproduction by females treated with tetracycline started to decrease significantly; in F 10 the progeny was less than half compared to females reared in penicillin and control diets (Figure 3 and 4). Also, females treated with tetracycline produced significantly fewer eggs (Figure 5). This fertility reduction and interpretation are in agreement with data from other beetles in the family Curculionidae. In the date stone beetle (Coccotrypes dactyliperda) mated females fed with 3% tetracycline produced significantly fewer eggs than control females (Zchori Fein et al., 2006). Similarly, in the rice water weevil (Lissorhoptrus oryzophilus) females treated with % tetracycline produced fewer eggs and none of these eggs were viable (Chen et al., 2012). The authors of both studies attributed the reduction in fecundity in these species to elimination of endosymbiotic bacteria, mainly Wolbachia. 76

93 Infection by endosymbiotic bacteria has been shown to be associated with femalebiased sex ratios (Werren, 1997; Jiggins et al., 2000; Weeks et al., 2003; Dyson and Hurst, 2004; Gotoh et al., 2007). Several studies have suggested that Wolbachia is directly related to the female-biased sex ratio observed in its hosts; in some cases, females cured with antibiotics produce more males or only males (Stouthamer et al., 1990; Pijls et al., 1996; Majerus et al., 1998; Hurst et al., 2000; Giorgini, 2001; Pannebakker et al., 2004). Vega et al. (2002) suggested that Wolbachia plays a role in the female-biased sex ratio observed in the CBB. Our results showed that Wolbachia s affects on the sex ratio are at best indirect: females treated with tetracycline produced significantly fewer daughters than females from controls, but tetracycline did not affect significantly the number of males produced in each type of diet (Table 2 and A.3). The less skewed sex ratio observed with tetracycline was more a consequence of the reduction in the number of females than the production of more males. Contrary to Vega et al. (2002), we suggest that Wolbachia is not involved in the sex determination of CBB; a significant increase in the number of males in progenies from females treated with tetracycline would be the expected result if Wolbachia were involved in sex determination of CBB. However, this result needs to be confirmed through the production of Wolbachia-cured lines. The best method to assess the impact of Wolbachia on sex determination in its hosts is to compare an infected line with a Wolbachia- cured line (Stouthamer and Mark, 2002). It is important to determine the minimum dose of tetracycline needed to completely cure Wolbachia infections; 3% (w/v) tetracycline eliminated endosymbiotic bacteria including Wolbachia from Coccotrypes dactyliperda (Coleoptera: Curculionidae) (Zchori Fein et al., 2006). However, such high concentrations also exacerbate toxicity issues. Alternatively, another antibiotic like rifampicin could be used; rifampicin is also effective against Wolbachia (Dedeine et al., 2001; Fenollar et al., 2003). 77

94 Relatively few studies have measured the effect of Wolbachia in arthropod population dynamics. The estimation of life parameters allows determination of the growth potential for a population. Our data showed that the reduction of Wolbachia infection with tetracycline reduced the values of both evaluated parameters; the population growth rate (λ) was reduced in populations treated with tetracycline, while the mean generation time (T) increased, so females reared in tetracycline produce less progeny and fewer generations per year 8.1 generations compared with 9.4 for control. Similarly, Wolbachia-free populations of Liposcelis tricolor (Psocoptera: Liposcelididae) had lower performance in all life parameters evaluated: intrinsic rate of increase rm, net reproductive value R 0, and mean generation time T (Dong et al., 2007). Taken together, these data suggest that Wolbachia are important for successful reproduction of the CBB, although it is possible that the effect is mediated by another endosymbiotic bacterium also reduced by tetracycline. 5. Conclusions In summary, in this study we detected Wolbachia in adult females from Puerto Rico; phylogenetic classification analysis showed that CBB can be infected with Wolbachia belonging to the supergroups A and B, as has been found in other Coleoptera (Lachowska et al., 2010). The Wolbachia phylogeny may be useful in tracing the origin of the CBB population that invaded Puerto Rico in 2007 (NAPPO, 2007); this origin is a matter of debate. Based on microbiota analysis, we observed that the proportion of Wolbachia in CBB is very low less < 1%. Wolbachia appears to contribute to reproductive success in the CBB: antibiotic treatment significantly reduced the proportion of Wolbachia, which had negative effects on reproduction and population dynamic of CBB, though not on sex ratio as expected. This result opens the door to the possibility of CBB management mediated by Wolbachia infection, 78

95 either by reducing Wolbachia infection or encouraging growth of competing microorganisms in the CBB microbiota. Future research might focus on treatments with higher concentrations of tetracycline or other antibiotics like rifampicin in order to produce Wolbachia-free CBB lines; it is also important to elucidate the mechanism by which Wolbachia influences CBB reproduction, which is not clear. To test whether the mechanism is cytoplasmic incompatibility crosses between Wolbachia-infected and free individuals should be done. The effects and mechanisms of Wolbachia on CBB appear to be complex, but our results show they have pronounced effects on the reproductive biology of this important pest. 6. Acknowledgments Special thanks to Michelle Cruz for outstanding work in the laboratory and to Oscar Ospina for help with microbiota analysis. We also thank the NSF CREST-CATEC program and its director Elvira Cuevas for support of undergraduate students. Thanks to UPR Sequencing and Genotyping facility, supported by NCRR AABRE grant #P20 RR16470.This research was supported by USDA Cooperative Agreements (PI Stephen A. Rehner, USDA- ARS Systematic Mycology and Microbiology Laboratory, Beltsville MD), and USDA- APHIS-PPQ Z-258, Z-265 and Z-273 ( , ) (PI José Carlos Verle Rodrigues, Professor of Department of Crops and AgroEnvironmental Sciences at University of Puerto Rico, Mayagüez Campus). 7. References Bergamin, J., Contribuição para o conhecimento da biologia da broca do café Hypothenemus hampei (Ferrari, 1867)(Col. Ipidae). Arq. Inst. Biol 14, 31Ð72. Breeuwer, J., Wolbachia and cytoplasmic incompatibility in the spider mites Tetranychus urticae and T. turkestani. Heredity 79,

96 Brun, L.O., Stuart, J., Gaudichon, V., Aronstein, K., French-Constant, R., Functional haplodiploidy: a mechanism for the spread of insecticide resistance in an important international insect pest. Proc Natl Acad Sci 92, Büyükgüzel, K., Yazgan, Ş., Effects of Antimicrobial Agents on the Survival and Development of Larvae of Pimpla turionellae L.(Hymenoptera: Ichneumonidae) reared on an artificial diet. Turk. J Zool. 26, Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., Fierer, N., Pena, A.G., Goodrich, J.K., Gordon, J.I., QIIME allows analysis of highthroughput community sequencing data. Nat. methods 7, Caporaso, J.G., Lauber, C.L., Walters, W.A., Berg-Lyons, D., Huntley, J., Fierer, N., Owens, S.M., Betley, J., Fraser, L., Bauer, M., Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. ISME J 6, Caswell, H., Matrix population models. Wiley Online Library. Charlat, S., Davies, N., Roderick, G.K., Hurst, G.D., Disrupting the timing of Wolbachia-induced male-killing. Biol. Lett. 3, Chen, S.-J., Lu, F., Cheng, J.-A., Jiang, M.-X., Way, M.O., Identification and biological role of the endosymbionts Wolbachia in rice water weevil (Coleoptera: Curculionidae). Environ. Entomol. 41, Dale, C., Welburn, S., The endosymbionts of tsetse flies: manipulating host parasite interactions. Int. J. Parasitol. 31, Damon, A., A review of the biology and control of the coffee berry borer, Hypothenemus hampei (Coleoptera : Scolytidae). B Entomol Res 90, Dedeine, F., Vavre, F., Fleury, F., Loppin, B., Hochberg, M.E., Boulétreau, M., Removing symbiotic Wolbachia bacteria specifically inhibits oogenesis in a parasitic wasp. Proc. Nat. Acad. Sci. U.S.A. 98, Dong, P., Wang, J.-J., Hu, F., Jia, F.-X., Influence of Wolbachia infection on the fitness of the stored-product pest Liposcelis tricolor (Psocoptera: Liposcelididae). J. Econ. Entomol. 100, Dong, P., Wang, J.-J., Zhao, Z.-M., Wolbachia infection and its influence on the reproduction of stored-product psocid, Liposcelis tricolor (Psocoptera: Liposcelididae). J. Insect. Sci. Dyson, E.A., Hurst, G.D., Persistence of an extreme sex-ratio bias in a natural population. Proc. Nat. Acad. Sci. U.S.A. 101, Fenollar, F., Maurin, M., Raoult, D., Wolbachia pipientis growth kinetics and susceptibilities to 13 antibiotics determined by immunofluorescence staining and real-time PCR. Antimicrob. Agents Chemother. 47,

97 Fry, A., Palmer, M., Rand, D., Variable fitness effects of Wolbachia infection in Drosophila melanogaster. Heredity 93, Giorgini, M., Induction of males in thelytokous populations of Encarsia meritoria and Encarsia protransvena: a systematic tool. Biocontrol 46, Gotoh, T., Noda, H., Ito, S., Cardinium symbionts cause cytoplasmic incompatibility in spider mites. Heredity 98, Hilgenboecker, K., Hammerstein, P., Schlattmann, P., Telschow, A., Werren, J.H., How many species are infected with Wolbachia? a statistical analysis of current data. FEMS Microbiol. Lett. 281, Hurst, G.D., Johnson, A.P., vd Schulenburg, J.H.G., Fuyama, Y., Male-killing Wolbachia in Drosophila: a temperature-sensitive trait with a threshold bacterial density. Genetics 156, Jaramillo, J., Borgemeister, C., Baker, P., Coffee berry borer Hypothenemus hampei (Coleoptera: Curculionidae):searching for sustainable control strategies. B Entomol Res 96, Jeyaprakash, A., Hoy, M., Long PCR improves Wolbachia DNA amplification: wsp sequences found in 76% of sixty-three arthropod species. Insect Mol. Biol. 9, Jia, F.-X., Yang, M.-S., Yang, W.-J., Wang, J.-J., Influence of continuous high temperature conditions on Wolbachia infection frequency and the fitness of Liposcelis tricolor (Psocoptera: Liposcelididae). Environ. Entomol. 38, Jiggins, F.M., Hurst, G.D., Majerus, M.E., Sex-ratio-distorting Wolbachia causes sexrole reversal in its butterfly host. Proc. R. Soc. Lond. B Biol. Sci. 267, Kageyama, D., Traut, W., Opposite sex specific effects of Wolbachia and interference with the sex determination of its host Ostrinia scapulalis. Proc. R. Soc. Lond. B Biol. Sci. 271, Kondo, N., Ijichi, N., Shimada, M., Fukatsu, T., Prevailing triple infection with Wolbachia in Callosobruchus chinensis (Coleoptera: Bruchidae). Mol. Ecol. 11, Kyei-Poku, G., Giladi, M., Coghlin, P., Mokady, O., Zchori-Fein, E., Floate, K., Wolbachia in wasps parasitic on filth flies with emphasis on Spalangia cameroni. Entomol. Exp. Appl. 121, Lachowska, D., Kajtoch, Ł., Knutelski, S., Occurrence of Wolbachia in central European weevils: correlations with host systematics, ecology, and biology. Entomol. Exp. Appl. 135, Lo, N., Casiraghi, M., Salati, E., Bazzocchi, C., Bandi, C., How many Wolbachia supergroups exist? Mol. Biol. Evol. 19, Maddison, W.P., Maddison, D., Mesquite: a modular system for evolutionary analysis. 81

98 Majerus, M., Knowles, B., Bertrand, D., Hurst, G.D., Extreme variation in the prevalence of inherited male-killing microorganisms between three populations of Harmonia axyridis (Coleoptera: Coccinellidae). Heredity 81, Mariño, Y., Ospina, O., Bayman, P., In revision. Host s diet, stage of development and environment in which it develops affect the bacterial communities of the coffee berry borer Hypothenemus hampei (Coleoptera: Curculionidae). NAPPO, Detection of Coffee Berry Borer, Hypothenemus hampei, in Puerto Rico- United States. Official Pest Reports. Pannebakker, B.A., Beukeboom, L.W., van Alphen, J.J., Brakefield, P.M., Zwaan, B.J., The genetic basis of male fertility in relation to haplodiploid reproduction in Leptopilina clavipes (Hymenoptera: Figitidae). Genetics 168, Pijls, J., Steenbergen, H.J.v., Alphen, J.J.v., Asexuality cured: the relations and differences between sexual and asexual Apoanagyrus diversicornis. Heredity 76, Portilla, M., Mass rearing technique for Cephalonomia stephanoderis (Hymenoptera: Bethylidae) on Hypothenemus hampei (Coleoptera: Scolytidae) developed using Cenibroca artificial diet. Rev. Col. Entomol. 25, Romero, J., Cortina, H., Tablas de vida de Hypothenemus hampei (Coleoptera: Curculionidae: Scolytinae) sobre tres introducciones de café. Rev. Col. Entomol. 33, Ruiz-Cárdenas, R., Baker, P., Life table of Hypothenemus hampei (Ferrari) in relation to coffee berry phenology under Colombian field conditions. Sci Agr 67, Son, Y., Luckhart, S., Zhang, X., Lieber, M.J., Lewis, E.E., Effects and implications of antibiotic treatment on Wolbachia-infected vine weevil (Coleoptera: Curculionidae). Agric. Forest Entomol. 10, Soto-Pinto, L., Perfecto, I., Caballero-Nieto, J., Shade over coffee: its effects on berry borer, leaf rust and spontaneous herbs in Chiapas, Mexico. Agroforest Syst 55, Srivastava, P., Auclair, J., Effects of antibiotics on feeding and development of the pea aphid, Acyrthosiphon pisum (Harris)(Homoptera: Aphididae). Can. J Zool. 54, Stouthamer, R., Breeuwer, J.A., Hurst, G.D., Wolbachia pipientis: microbial manipulator of arthropod reproduction. Annu. Rev. Microbiol. 53, Stouthamer, R., Luck, R.F., Hamilton, W., Antibiotics cause parthenogenetic Trichogramma (Hymenoptera/Trichogrammatidae) to revert to sex. Proc. Nat. Acad. Sci. U.S.A. 87, Stouthamer, R., Mark, F., Influence of antibiotics on the offspring production of the Wolbachia- infested parthenogenetic parasitoid Encarsia formosa. J. Invertebr. Pathol. 80, Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S., MEGA6: molecular evolutionary genetics analysis version 6.0. Mol. Biol. Evol. 30,

99 Taylor, M., Bandi, C., Hoerauf, A., Lazdins, J., Wolbachia bacteria of filarial nematodes: a target for control? Parasitol. Today 16, Team, R., Develoment Core. "R Foundation for Statistical Computing". Vienna, Austria 482. Van Lenteren, J.C., Quality control and production of biological control agents: theory and testing procedures. CABI. Vega, F.E., Benavides, P., Stuart, J.A., O'Neill, S.L., Wolbachia infection in the coffee berry borer (Coleoptera: Scolytidae). Ann Entomol Soc Am 95, Vega, F.E., Infante, F., Castillo, A., Jaramillo, J., The coffee berry borer, Hypothenemus hampei (Ferrari)(Coleoptera: Curculionidae): a short review, with recent findings and future research directions. Terr Arthropod Rev 2, Vega, F.E., Posada, F., Infante, F., Coffee insects: ecology and control. Encyclopedia of pest management, 1-4. Weeks, A.R., Velten, R., Stouthamer, R., Incidence of a New Sex ratio distorting Endosymbiotic Bacterium among arthropods. Proc R Soc Lond [Biol] 270, Werren, J.H., Biology of Wolbachia. Ann. Rev. Entomol. 42, Werren, J.H., Baldo, L., Clark, M.E., Wolbachia: master manipulators of invertebrate biology. Nat. Rev. Microbiol. 6, Werren, J.H., Windsor, D.M., Wolbachia infection frequencies in insects: evidence of a global equilibrium? Proc. R. Soc. Lond. B Biol. Sci. 267, Zchori Fein, E., Borad, C., Harari, A.R., Oogenesis in the date stone beetle, Coccotrypes dactyliperda, depends on symbiotic bacteria. Physiol. Entomol. 31, Zhou, W., Rousset, F., O'Neill, S., Proc. R. Soc. Lond. B Biol. Sci. 265,

100 Table 1. Reference sequences of Wolbachia used to determine CBB Wolbachia supergroup classification Host insect species Order Wolbachia Supergroup GenBank Accession No. Aphthona nigriscutis Coleoptera A AF Bruchid sp. Coleoptera A KC Callosobruchus chinensis Coleoptera A AB Ceutorhynchus subpubescens Coleoptera A HQ Conotrachelus nenuphar Coleoptera A GU Cynotrachelus coloratus Coleoptera A GU Diabrotica barberi Coleoptera A AY Drosophila sturtevanti Diptera A AY Lutzomia shannoni Diptera A AF Elasmucha signoretii Heteroptera A AB Formica excecta Hymenoptera A AY Nasonia vitripennis Hymenoptera A AF Spalangia cameroni Hymenoptera A DQ Tachinaephagus zealandicus Hymenoptera A DQ Trichogramma kaykai Hymenoptera A AF Caudra cautella Lepidoptera A AF Ephesia kuehniella Lepidoptera A AB Nymphalis xanthomelas Lepidoptera A AB Polyplax serrata Anaplura B AY Callosobruchus chinensis Coleoptera B AB Conotrachelus nenuphar Coleoptera B GU Conotrachelus nenuphar Coleoptera B GU Hypothenemus hampei Coleoptera B AF Lissorhotrus oryzophilus Coleoptera B HM Naupactus cervinus Coleoptera B GQ Tribolium confusum Coleoptera B AF Culex sitiens Diptera B AF Coquilletidia crassipes Diptera B AF Drosophila fundita Diptera B JX Drosophila simulans Diptera B AF Phlebotomus perniciosus Diptera B AF Bermisia tabaci Hemiptera B AF Nesophrosyne silvicola Hemiptera B JX Orius nagaii Hemiptera B AB Tagosedes orizocolus Hemiptera B AF Mysius plebeius Heteroptera B AB Spilostethus hospes Heteroptera B AB Acromyrmex echinatior Hymenoptera B AF Torymus bedeguaris Hymenoptera B AF Acraea encedon Lepidoptera B AJ Mamestra brassicae Lepidoptera B AB

101 Table 2. Number of females, males and sex ratio of Hypothenemus hampei reared in artificial diets with and without antibiotics; Mean ± SE and F:M ratio per vial for F1, F5 and F10 generations Variable F 1 F 5 F 10 Control T 0.1% P 0.1% Statistics Control T 0.1% P 0.1% Statistics Control T 0.1% P 0.1% Statistics Females 20.8± ± ±2.53 Z=61.0/P=0.04 * 62.0± ± ±4.79 Z=1.05/P<0.001 * 84.2± ± ±8.64 Z=182.0/P<0.001 * Males 2.4± ± ±0.38 Z=7.29/P<0.77 * 9.35± ± ±0.71 Z=30.48/P<0.001 * 11.3± ± ±0.82 Z=36.45/P=0.74 * Sex Ratio 0.89±0.01a 0.86±0.02a 0.84±0.02a F=1.05/P=0.35 ** 0.86±0.01b 0.80±0.01a 0.89±0.01b F=11.45/P<0.001 ** 0.88±0.01b 0.83±0.01a 0.89±0.01b F=6.39/P=0.003 ** * =Values based on generalized linear models ** = Values based on ANOVA tests. Means followed by the same letter within rows are not significantly different (P= 0.05, Post-hoc Tukey s test) Table 3. Developmental time in days (Mean±SE) for life stages and life table parameters of Hypothenemus hampei reared in artificial diet with and without antibiotics Diet Egg Larva Pre-pupa Pupa Juvenile Egg to adult Λ T control 6.08±0.22a 15.22±0.75a 1.94±0.11a 6.37±0.15a 4.05±0.12a 34.41±0.75a 1.11±0.01a 39.01±3.13a penicillin 6.07±0.31a 15.45±0.59 a 1.87±0.07a 6.78±0.18a 4.09±0.19a 34.26±0.61a 1.10±0.01a 39.19±5.39a tetracycline 6.10±0.21a 15.52±0.74a 2.03±0.09a 6.82±0.26a 4.67±0.24b 35.13±0.94a 1.07 ±0.01a 45.19±3.05a λ = population growth rate; T= mean generation time. Statistical significance based on ANOVA tests. Means followed by the same letter within rows are not significantly different (P= 0.05, Post-hoc Tukey s test) 85 85

102 Figure 2. Molecular phylogenetic analysis of the Wolbachia wsp sequences identified from Hypothenemus hampei with sequences of Wolbachia supergroups A and B. Reference sequences from 41 species of insects including the CBB from (Vega et al., 2002). The tree was constructed by Neighbor-Joining (NJ) using the Maximum Composite Likelihood model and midpoint rooted. Bootstrap probabilities > 50% are shown at the nodes. Accession numbers for sequences from GenBank are shown. 86

103 300 Control Tetracycline Penicillin 250 Mean Population First Third Fifth Tenth Generation Figure 3. Effects of antibiotics on Hypothenemus hampei populations (including eggs, larvae, pupae, juveniles and adults) for F1, F3, F5 and F10 generations. Means ± S.E. are shown. 87

104 50 First Generation Control Penicillin Tetracycline 50 Third Generation aaa Mean CBB per female a a a ab b aaa aaa aaa a a a Mean CBB per female a a b aaa a a b aab b a a b Eggs Larvae Fifth Generation Pupae Juveniles Adults Population Eggs Larvae Pupae Tenth Generation Juveniles Adults Population a a Mean CBB per female a a Eggs b aaa Larvae aaa Pupae aaa Juveniles aaa Adults a a Population b Mean CBB per female a a Eggs b a b Larvae c a a b Pupae aaa Juveniles a a Adults b Population b Figure 4. Effects of antibiotics on the number of individuals (mean ± SE) per female for each stage of Hypothenemus hampei at F1, F3, F5 and F10 generations. Different letters indicate significant differences in the number of individuals per female between the types of diet based on ANOVAs and post-hoc Tukey test (P<0.05). Note that the Y-axis scale differs; reproduction was almost 2x higher in the F 10 generation than F 1 and F 3. 88

105 Control Penicillin Tetracycline 25 A 80 B Mean eggs per female *** % females with oviposition Control Penicillin Tetracycline Diet 0 Control Penicillin Tetracycline Diet Figure 5. Effects of antibiotics after five generations of treatment on Hypothenemus hampei. (A) Fecundity (number of eggs per female) and (B) % females ovipositing. Asterisks show significance in the generalized linear model: ***, P < 0.001; **,P < 0.01; *,P <0.05; P <0.1). 89

106 8. Appendices Table A.1. Ingredients, amount and providers in the artificial diet to rear the coffee berry borer (H. hampei) Ingredients Amount per liter Providers Distilled water 750 ml Laboratory Agar 10 g Bio-serv Casein, pure 30 g Fisher scientific Yeast (Brewers Hydrolysate) 15 g Bio-serv Sugar 10 g Supermarket Coffee bean powder (12% of humidity) 150 g Benzoic acid 1 g VWR International Benomyl (Benlate) 3 g Dupont Vanderzant vitamins mixture for insects 1 g Bio-serv Wesson mix salts 1.6 g Bio-serv Formaldehyde 2 ml Fisher scientific Ethanol 70% 20 ml Ricca chemical company Table A.2. Results of generalized linear model evaluating the relationship between CBB total population and type of diet (Control, Penicillin 0.1% and Tetracycline 0.1%) and generations Effects Estimate Standard error Z - value P Intercept <0.0001*** Type of diet (Control vs. Penicillin) *** Type of diet (Control vs. Tetracycline) <0.0001*** Generation (F 1 vs. F 2 ) <0.0001*** Generation (F 1 vs. F 3 ) <0.0001*** Generation (F 1 vs. F 4 ) <0.0001*** Generation (F 1 vs. F 5 ) <0.0001*** Generation (F 1 vs. F 10 ) <0.0001*** 90

107 Table A.3. Results of generalized linear model evaluating the relationship between CBB stages of development and type of diet (Control, Penicillin 0.1% and Tetracycline 0.1%) Effects Estimate Standard error Z - value P Eggs (Intercept) <0.0001*** Type of diet (Control vs. Penicillin) *** Type of diet (Control vs. Tetracycline) <0.0001*** Larvae (Intercept) <0.0001*** Type of diet (Control vs. Penicillin) <0.0001*** Type of diet (Control vs. Tetracycline) <0.0001*** Pupae (Intercept) <0.0001*** Type of diet (Control vs. Penicillin) <0.0001*** Type of diet (Control vs. Tetracycline) <0.0001*** Juveniles (Intercept) <0.0001*** Type of diet (Control vs. Penicillin) Type of diet (Control vs. Tetracycline) * Adults (Intercept) <0.0001*** Type of diet (Control vs. Penicillin) <0.0001*** Type of diet (Control vs. Tetracycline) <0.0001*** Females (Intercept) <0.0001*** Type of diet (Control vs. Penicillin) <0.0001*** Type of diet (Control vs. Tetracycline) <0.0001*** Males(Intercept) <0.0001*** Type of diet (Control vs. Penicillin) Type of diet (Control vs. Tetracycline)

108 Table A.4. Stage transition frequencies and cumulative number of dead individuals in each stage of development of CBB in diet without antibiotics or control diet. Measure Eggs Larvae Pre-pupae Pupae Juveniles Adults Total Day 6 Number of individuals Death Total Day 12 Number of individuals Cumulative death Total Day 18 Number of individuals Cumulative death Total Day 24 Number of individuals Cumulative death Total Day 30 Number of individuals Cumulative death Total Day 36 Number of individuals Cumulative death Total Day 42 Number of individuals Cumulative death Total Day 50 Number of individuals Cumulative death Total Note: Five cohorts of 20 eggs were evaluated. Here I present the data of the total number of individuals evaluated for the five cohorts. 92

109 Table A.5. Stage transition frequencies and cumulative number of dead individuals in each stage of development of CBB in diet with penicillin 0.1% (w/v). Measure Eggs Larvae Pre-pupae Pupae Juveniles Adults Total Day 6 Number of individuals Death Total Day 12 Number of individuals Cumulative death Total Day 18 Number of individuals Cumulative death Total Day 24 Number of individuals Cumulative death Total Day 30 Number of individuals Cumulative death Total Day 36 Number of individuals Cumulative death Total Day 42 Number of individuals Cumulative death Total Day 50 Number of individuals Cumulative death Total Note: Five cohorts of 20 eggs were evaluated. Here I present the data of the total number of individuals evaluated for the five cohorts. 93

110 Table A.6. Stage transition frequencies and cumulative number of dead individuals in each stage of development of CBB in diet with tetracycline 0.1%. Measure Eggs Larvae Pre-pupae Pupae Juveniles Adults Total Day 6 Number of individuals Death Total Day 12 Number of individuals Cumulative death Total Day 18 Number of individuals Cumulative death Total Day 24 Number of individuals Cumulative death Total Day 30 Number of individuals Cumulative death Total Day 36 Number of individuals Cumulative death Total Day 42 Number of individuals Cumulative death Total Day 50 Number of individuals Cumulative death Total Note: Five cohorts of 20 eggs were evaluated. Here I present the data of the total number of individuals evaluated for the five cohorts. 94

111 CHAPTER 4 The bacterial microbiota of the coffee berry borer Hypothenemus hampei is influenced by host diet, development and environment Mariño-Cárdenas, Y., Ospina, O., and Bayman, P. In Revision. The bacterial microbiota of the coffee berry borer Hypothenemus hampei is influenced by host diet, development and environment. 95

112 Abstract Hypothenemus hampei, also known as the coffee berry borer (CBB), is the most important pest of coffee. Studies of the microbiota associated with this species are relatively few, and have focused on the cultivable gut microbiota and its role in caffeine detoxification; little attention has been given to the diversity and composition of the microbiota and environmental factors that shape it. We used Illumina-based sequencing of the hypervariable region V4 of the bacterial 16S rrna gene to characterize the communities associated with wild and laboratory-reared eggs and adult females. Composition of the CBB-associated microbiota appears to be influenced by host diet, stage of development and environment. The microbiota of the CBB was dominated by Proteobacteria, mainly Pseudomonadaceae. The genera Acinetobacter, Ochrobactrum, Pedobacter, Pseudomonas, Sphingobacterium, Stenotrophomonas and Wolbachia were common in both wild and laboratory reared CBBs, and may represent a core microbiota of this species. Some of these genera have been previously reported to be involved in caffeine detoxification. However, they might also participate in other important functions in their hosts, such as nutrition, defense against entomopathogens and reproduction. This study represents the first use of high-throughput sequencing to characterize microbiota diversity and composition in CBB, and the first extrapolate its functional importance in CBB control. 1. Introduction Insects are associated with range symbionts, including obligate, facultative, pathogenic and parasitic relationships (Wernegreen, 2002; Baumann, 2005; Moran et al., 2008; Brownlie and Johnson, 2009). These symbionts play many important and complex roles in the biology of their hosts, including nutritional adaptations (Baumann, 2005; Douglas, 2009), heat tolerance (Russell and Moran, 2006; Moran et al., 2008), protection against parasites and entomopathogens (Haine, 2008; Brownlie and Johnson, 2009; Indiragandhi et 96

113 al., 2011) and reproductive manipulations (Zchori-Fein and Perlman, 2004; Perotti et al., 2006; Werren et al., 2008). The coffee berry borer Hypothenemus hampei Ferrari (Coleoptera: Curculionidae) is the most devastating pest of coffee worldwide (Damon, 2000; Soto-Pinto et al., 2002; Jaramillo et al., 2006; Vega et al., 2009). The CBB and depends principally on coffee fruits for food and reproduction (Damon, 2000). The entire life cycle occurs inside the fruit; only fertilized females exit to find new fruits to colonize and start oviposition (Damon, 2000; Vega et al., 2002; Vega et al., 2009). The economic and social impacts of the CBB in coffee-producing areas have motivated research on its distribution, natural enemies and strategies for its control (De la Rosa et al., 1997; Armbrecht and Perfecto, 2003; Silva et al., 2006; Armbrecht and Gallego, 2007; Teodoro et al., 2008; Jaramillo et al., 2011). Conversely, little attention has been paid to the microbial community associated with the CBB, and its role in nutrition, defense and reproduction. Preliminary studies have surveyed the culturable bacterial community in the CBB gut, in order to determine its role in caffeine degradation (Ceja-Navarro et al. cited by Vega et al., 2015) and to detect the presence of reproductive manipulators such as Wolbachia in adult females (Vega et al., 2002). Novel insect management strategies could be based on disruption of obligate symbioses that are required for nutrition and reproduction of the insect or resistance to natural enemies (Douglas, 2007). However, any possible application of CBB control based on bacterial symbiosis requires an extensive study of its microbiota, in order to identify a stable core microbiota and its functions in CBB biology. Here we analyze and compare eggs and adult females collected in infested fruits from coffee plantations growing under full sun and shade, and reared in artificial diets with and without antibiotics; in order to answer the following questions: 97

114 1) Is there a core microbiota of the CBB that changes during development? We hypothesized that there is a succession of microbiota related with the host s stage of development, in which microbiotas of eggs are formed mainly by heritable symbionts and those that adhere to the outer membrane during the oviposition, while communities of adult females will be more complex, but with some taxa maintained throughout life history stages. 2) How do antibiotics affect the bacterial community composition and diversity of CBB? We hypothesized that treatment with antibiotics can reduce the diversity and affect the composition of bacterial communities from eggs and adult females of CBB. 3) How do environmental conditions affect the bacterial community composition and diversity of CBB? Based on laboratory studies of the effect of temperature on Wolbachia density and its transmission efficiency, we predicted that the higher temperatures in sun coffee plantations can reduce Wolbachia proportion compared to shade coffee. 2. Materials and Methods 2.1. Collection and laboratory rearing of coffee berry borers We compared CBB microbiota from: (i) eggs vs. adult females, (ii) wild individuals collected in infested fruits from coffee plantations growing under full sun vs. shade, and (iii) individuals reared for eleven generations in the artificial diet Cenibroca with tetracycline 0.1% vs. control without antibiotics. We included only adult females and eggs because one of our main goals was to determine the proportion of Wolbachia in the CBB and the presence of other endosymbionts that are reproductive manipulators of insects like Rickettsia, which has been reported in other Curculionidae, and these endosymbionts are maternally inherited (Zchori Fein et al., 2006). Wild CBBs were collected from infested coffee fruits (Coffea arabica) from plantations growing under full sun and shade, in two farms in Adjuntas, Puerto Rico: Farm 1 98

115 (18º N, 66º W, 527 m a.s.l) and Farm 2 (18º N, 66º W, 624 m a.s.l). Fruits were dissected to extract adult females and eggs. CBBs were reared for eleven consecutive generations in the artificial diet Cenibroca (Portilla, 1999). For antibiotic treatment tetracycline hydrochloride 0.1% (w/v) (Fisher Bioreagents) was added to the Cenibroca diet for direct and continuous feeding of adult females. For all generations, insects were reared in the dark in a growth chamber at 25ºC and 80 96% relative humidity. After F 11 vials with diets were carefully dissected to collect adult females and eggs. Mean generation time was approx days for control and 35.2 days for the diet with tetracycline (Mariño et al., In Revision) DNA extraction and sequencing Insects were stored dry at -20ºC until DNA was extracted. Total DNA from 20 pooled eggs and the entire bodies of five pooled females, with four replicates per treatment, were extracted using DNeasy Plant Mini Kit (Qiagen Sciences). The hypervariable V4 region of the bacterial 16S rrna gene was amplified using standard primers (515-F: 5 TGCCAGCMGCCGCGGTAA 3 and 806-R: 5 GGACTACHVGGGTWTCTAAT 3 ) (Caporaso et al., 2012) with the addition of Illumina adapters sequences to amplify 250 bp. The PCR amplification program was an initial denaturation at 94 ºC for three minutes followed by 35 cycles of denaturation at 94 ºC for 30 seconds, annealing at 54 ºC for 45 seconds, extension at 72 ºC for 1 minute and final extension at 72 ºC for 10 minutes. All PCRs included negative controls, from which no products were obtained. Amplicons were sequenced on the Illumina MiSeq platform at the University of Puerto Rico Sequencing and Genotyping Facility (UPR-SGF). 99

116 2.3. Data processing Illumina sequence data was processed using the scripts included in MacQIIME version (Caporaso et al., 2010). The initial step joined the paired end reads using QIIME s join_paired_ends.py script. The assembled reads were divided into fasta and qual files with the script convert_fastaqual_fastaq.py. The resulting files were used as input for the split_libraries.py script, forward and reverse primers were removed using the truncate only argument, and sequences with average Quality Score < 30 and a length 200 bp < x > 300 bp were discarded. Operational taxonomic units (OTUs) were selected using the pick_otus.py script, which made clusters based on 97% similarity of the filtered sequence set. From each cluster a representative sequence was selected and taxonomy was assigned using the Ribosomal Database Project (RDP) Classifier (Wang et al., 2007). Chimeras were identified using the ChimeraSlayer algorithm and the template alignment provided by PyNAST, and eliminated from otu_table.biom using the filter_otus_from_otu_table.py script. Sequences from representative Pseudomonadaceae were extracted from the OTU table and BLASTed to identify the most similar sequences in GenBank. These taxa composed almost the half of the sequences in the CBB microbiota, and could not be identified to genus using the Greengenes database Data analysis Alpha diversity For α diversity calculation, we rarefied the sequences using the QIIME s default parameters. Based on rarefaction curves we calculated the indices of Shannon or community diversity (Shannon, 1948) and Chao 1 or species richness (Chao, 1984). One way ANOVAs followed 100

117 by post-hoc Tukey tests (significance level P=0.05) were used to test significance of differences between samples Beta diversity Based on the rarefied dataset, we calculated beta diversity (community variation) for microbiota using Bray-Curtis dissimilarity matrices (Bray and Curtis, 1957) and Unifrac metrics (Lozupone and Knight, 2005). We used jackknife principal coordinate analysis (PCoA) to visualize differences between communities. Proportion tests were used to test differences in the number of sequences assigned to Wolbachia among samples. ANOVAs and proportion tests were done in R (R, 2013). 3. Results 3.1. General composition of the CBB microbiota and 16S rrna sequencing 16,116,312 V4 region 16S rrna gene sequences were obtained from 32 samples. No significant difference was observed in the number of sequences per sample (F= 2.14, df=7, P = 0.079), with a mean of 503,634 ± 28,200 sequences. Sequences were clustered into 65,533 OTUs. Unknown bacteria represented only 0.1% of all sequences. In general, microbiota was mostly composed of the phyla Proteobacteria (90.1%), Bacterioidetes (5.5%), Firmicutes (3.1%), Actinobacteria (0.6%) and Cyanobacteria (0.3%). Adult females from control diets contained a high proportion of Firmicutes (21.6%) (Figure 1). Microbiota of both adults and eggs, independent of diet, were dominated by families Pseudomonadaceae (48%) and Enterobacteriaceae (16.6%). Pseudomonadaceae were most abundant in samples from artificial diets; while Enterobacteriaceae were relatively more abundant in samples collected from infested fruits in the field (Table 1). The 97% of reads assigned to Pseudomonadaceae had a 98 to 100% of similarity to Pseudomonas sp. sequences 101

118 deposited in GenBank and detected in other invasive beetle (Dendroctonus valens) and soil (Table A.2). Fifteen genera represented more that 0.1% of CBB microbiota. Acinetobacter, Ochrobactrum, Pedobacter, Pseudomonas, Sphingobacterium, Stenotrophomonas and Wolbachia were common in both eggs and females from field and laboratory. Enterococcus was only associated with adults; however, it was absent from adults that were fed tetracycline (Table 1) Antibiotic, environmental temperatures and host stage of development affect the frequency of Wolbachia Wolbachia was the only endosymbiont found that is known to manipulate host reproduction. Wolbachia was significantly more frequent in eggs than in adult females, and in samples from artificial diets than samples from the field (field: eggs 0.14 vs. females 0.11%, χ 2 = 113.8, df=1, P <0.001; artificial diets: eggs 0.29% vs. females 0.26%, χ 2 = 52.80, df=1, P <0.001). Wolbachia was significantly more frequent in adults from control diets than in adults from diets with tetracycline, as expected (Females control 0.39% vs. tetracycline 0.04%, χ 2 = , df=1, P <0.001). However, contrary to expectations, in eggs from control diets Wolbachia was lower than in eggs from diets with tetracycline (Eggs: control 0.19% vs. tetracycline 0.47%, χ 2 = , df=1, P <0.001). Wolbachia was significantly more frequent in CBB females from shade coffee than sun coffee, as expected, probably higher temperatures that prevalence in sun coffee plantations reduced the Wolbachia proportion (Females: shade coffee 0.16% vs. sun coffee 0.04, χ 2 = 1271, df=1, P <0.001). Overall, the total proportion of Wolbachia in the CBB microbiota was 0.2% (Table 1). 102

119 3.2. Alpha diversity We observed significant differences among samples in both indices for α diversity (Shannon: F= 11.60, df=7, P < and Chao 1: F= 4.14, df=7, P = 0.004). Rarefaction curves are plotted in Figure 2. Overall, bacterial communities associated with samples from coffee fruits in field were more diverse than those from artificial diets; antibiotic treatment reduced significantly the diversity of adult females but not of the eggs (Table 2). Differences in diversity between sun and shade were not significant (Shannon: F=0.33, df=3, P= 0.81, and Chao 1: F=1.13, df=3, P= 0.38). However, diversity and species richness was slightly lower in bacterial communities from sun than from shade (Table 2) Beta diversity The composition of bacterial communities differs between field and laboratorycollected CBBs Principal coordinates analysis (PcoA) including both field and laboratory samples showed that PC1 explained 74.1 % of variation and separated the bacterial communities from field and laboratory individuals. PC2 explained 15.5% of the variation and separated eggs from adult females in field samples (Figure 3). Adult females and eggs from artificial diets harbored similar communities and these communities were dissimilar from those from fieldcollected CBBs. Contrary to samples from artificial diets, communities from eggs and adult females from the field had distinct communities (Figure 3) The composition of bacterial communities associated with CBBs differ between sun and shade Principal coordinate analysis (PCoA) including only field samples showed that PC1 explained 71% of variation and separated the bacterial communities from eggs and adults. 103

120 PC2 explained 19% of the variation and separated the communities according to type of coffee (sun vs. shade) (Figure 4). 4. Discussion The main results of this study can be summarized as follows: (1) CBB bacterial communities were dominated by Proteobacteria; in this phylum families Pseudomonadaceae and Enterobacteriaceae were the most abundant. (2) The genera Acinetobacter, Ochrobactrum, Pedobacter, Pseudomonas, Sphingobacterium, Stenotrophomonas and Wolbachia were common in both eggs and adults from wild and laboratory reared CBBs, and may represent a core microbiota of this species. (3) Antibiotic treatment with tetracycline affected the diversity of communities from adults but not in eggs, and reduced significantly the proportion of Wolbachia from 0.49% to 0.04%. (4) Bacterial communities from fieldcollected individuals from sun coffee were significantly different than those from shade coffee. Sun plots are hotter than shade plots (Mariño et al., Submitted); exposition to these higher temperatures may be responsible for the reduction in Wolbachia from 0.16 to 0.04%. (5) Bacterial communities were shaped according to the CBB s diet, stage of development and the environment in which it develops Composition of bacterial communities and their possible role in CBB biology Bacterial communities from eggs and adult females of the coffee berry borer (CBB) Hypothenemus hampei were represented by Proteobacteria, Bacteriodetes, Firmicutes, Actinobacteria and Cyanobacteria. The four first phyla represent a core microbiome associated with animals ((Jones et al., 2013) and dominate the microbiota of other insects (Egert et al., 2005; Vasanthakumar et al., 2008; Moran et al., 2012; Jones et al., 2013; Wong et al., 2013).. 104

121 Consistent with previous studies, we found that CBB s bacterial communities were dominated by Proteobacteria (90.1%) (Vasanthakumar et al., 2008; Nachappa et al., 2011; Osei Poku et al., 2012; Staubach et al., 2013; Wong et al., 2013; Aksoy et al., 2014; Chandler et al., 2014). Proteobacteria contains the most important insect symbionts (Jones et al., 2013) including those that influence fundamental processes in their hosts such as nutrition (Douglas, 2009; Aksoy et al., 2014), immune response (Oliver et al., 2003; Oliver et al., 2005; Teixeira et al., 2008) and reproduction (Dedeine et al., 2001; Dong et al., 2006; Zchori Fein et al., 2006; Son et al., 2008). At the family level, the microbiota of both eggs and adult females were composed of at least 56.9% Pseudomonadaceae and 17% Enterobacteriaceae. The family Pseudomonadaceae has been detected in invertebrate guts including termites, and is a cosmopolitan group due its capacity to grow and subsist in invertebrate hosts (Ley et al., 2008). Enterobacteriaceae contains various maternally transmitted symbiont of arthropods (Moran et al., 2005), including important primary symbionts like Buchnera in aphids, Blochmannia in ants and Wigglesworthia in tsetse flies (Moran et al., 2008; Weiss and Aksoy, 2011). Family Pseudomonadaceae was present in all samples, but was most abundant in individuals reared in artificial diets; in adult females fed with tetracycline this represented 90.7% of the microbiota (Table 1). Probably the majority of these bacteria are tetracycline resistant (Li et al., 1994; Kieboom and de Bont, 2001; Dang et al., 2008). GenBank comparisons showed that 97% of the sequence assigned to Pseudomonadaceae had % of identity with sequences of the genus Pseudomonas detected in other invasive beetle species and soil (Table A.2). The genera Pseudomonas, Sphingobacterium and Stenotrophomonas have been previously isolated as endophytic bacteria of coffee (Vega et al., 2005; Mariño and Zapata, 105

122 2009). Pseudomonas, Stenotrophomonas and Ochrobactrum were also isolated from the gut of the CBB and can detoxify caffeine (Ceja-Navarro et al. cited by Vega et al. 2015). Pseudomonas strains isolated from the gut of insects showed antagonistic activity against entomopathogenic fungi including Beauveria bassiana and Metarhizium anisopliae in vitro (Indiragandhi et al., 2007; Blackburn et al., 2008) and the bacterium Bacillus thuringiensis (Indiragandhi et al., 2008). These three important entomopathogens have been previously reported as enemies of the CBB (De la Rosa et al., 1997; Haraprasad et al., 2001; Méndez-López et al., 2003; Vega et al., 2008). The genera Acinetobacter, Sphingobacterium and Ochrobactrum produce siderophores (Indiragandhi et al., 2008; Gross and Loper, 2009; Gauglitz et al., 2012), known for their antifungal and antibacterial activity (Dillon and Dillon, 2004; Indiragandhi et al., 2011); which may help their host in caffeine detoxification and may also contribute to host defenses against pathogens. Most studies of microbiota of insects have focused only on the gut; relatively few studies have included the whole body or different tissues of the insect. We used whole insects to include reproductive manipulators like Wolbachia, which are found in other organ systems (Jones et al., 2013). A study of the microbiota from hemolymph, alimentary canal and whole body of the sharpshooter Homalodisca vitripennis, detected Wolbachia only in samples from whole insects (Hail et al., 2011) Antibiotic, environmental temperatures and host s stage of development affect the Wolbachia proportions Wolbachia is the best known reproductive manipulator, but Rickettsia, Cardinium and Spiroplasma have also been shown to have similar effects in some cases (Hurst et al., 2000; Zchori-Fein et al., 2001; Hunter et al., 2003; Weeks et al., 2003; Goto et al., 2006; Braig et al., 2008); double infections with Wolbachia and Rickettsia have been reported in the date stone beetle Coccotrypes dactyliperda (Coleoptera: Curculionidae) (Zchori Fein et al., 106

123 2006). In this study, however, Wolbachia was the only known reproductive parasite detected in the CBB, and this was present in proportions of <1% (Table 1). Maternally transmitted reproductive parasites like Wolbachia can manipulate their hosts sex ratios in order to favor females, thus enhancing their own transmission (Stouthamer et al., 2002). Wolbachia has been detected in CBB females from several countries, including Puerto Rico (Vega et al., 2002; Mariño et al., In Revision); its mode of action is thought to be cytoplasmic incompatibility (Vega et al., 2002). In addition to affecting sex ratio, Wolbachia can influence the reproduction and fitness of its hosts (Stouthamer et al., 1999; Dedeine et al., 2001; Zchori Fein et al., 2006). In some beetles in the family Curculionidae the elimination of Wolbachia with antibiotics significantly reduced fecundity, and in some cases eggs from cured females were not viable (Zchori Fein et al., 2006; Chen et al., 2012). The proportion of Wolbachia was significantly higher in artificial diets; individuals reared in laboratory are exposed to constant conditions (for example, stable temperature and food quality), whereas in the field fluctuations in environmental factors can reduce bacterial density (Stouthamer et al., 1999). Furthermore, Hurst et al. (2001) found that temperature play a major role in Wolbachia density and in its transmission, in Drosophila reared in laboratory vertical transmission was inefficient at 25ºC. Temperatures of coffee plots where wild CBBs were collected ranged from ºC for shade plots and ºC for sun plots (Mariño et al., Submitted), these fluctuations in temperature may reduce the Wolbachia density and cause inefficient transmission in wild individuals. The Cenibroca diet also contains antimicrobial preservatives, which we mention in detail below; probably these preservatives can reduce the density of other bacteria but not Wolbachia, which require antibiotics like tetracycline and rifampicin for efficient elimination from their hosts (Dedeine et al., 2001; Fenollar et al., 2003). 107

124 The treatment with tetracycline and the exposition to high environmental temperatures reduced frequency of Wolbachia (Table 1); the use of antibiotics and exposition to high temperatures are two established methods to cure or reduce Wolbachia infections (Dedeine et al., 2001; Gotoh et al., 2007). Temperatures in coffee leaves exposed to direct sunlight can be over 40 C (Maestri and Barros, 1977), and exposed fruits probably reach a similar temperature; in sun coffee plots where CBB were collected, as we mentioned before maximum temperatures were 38.2 C. These temperatures might be sufficiently high to reduce the Wolbachia infection, a possible advantage of sun coffee over shade coffee. Finally, the proportion of Wolbachia was significantly higher in eggs than females; obligate endosymbionts like Wolbachia pass vertically to the next generation via transovarial transmission at early stages of oogenesis. Alternatively, newly-hatched insects can acquire the symbionts by consuming or probing the eggshell or symbiont capsules, both structures are provided with symbiont-filled particles (Fukatsu and Hosokawa, 2002; Kikuchi et al., 2007; Crotti et al., 2010). However, most gut and surface bacteria are not transmitted to the eggs. Thus it is not surprising that Wolbachia composes a greater share of the microbiota of eggs than of adult females. Moreover, tetracycline reduced significantly the bacterial diversity of adult females but not that of eggs. A possible explanation is that consumption of the antibiotic affected the digestive tract more than the reproductive system; as we mentioned before bacterial communities from eggs probably are mainly formed for heritable symbionts, which are localized in a highly specialized organ, the bacteriome (Moran et al., 2008). This organ may insulate the eggs against the direct effect of antibiotics. 108

125 4.3. Alpha and Beta diversity Alpha diversity varied strongly across samples; bacterial communities associated with wild individuals were more diverse (Shannon indexes) and rich in species (Chao 1 indexes) than those from individuals reared in artificial diets (Table 2). Similar results have been found with other insect species (Geib et al., 2009; Rani et al., 2009; Staubach et al., 2013). Insects in the field acquire new bacterial taxa from the environment, while individuals on an artificial diet are exposed to fewer. The Cenibroca diet also contains antimicrobial preservatives: benzoic acid, one of the most common preservatives in the food industry, which inhibits the growth of bacteria and fungi (Brul and Coote, 1999; Davidson et al., 2007), and benomyl, used to avoid fungal contamination. However, benomyl inhibits growth of some bacteria (Rousk et al., 2009). Comparing the diversity observed here with previous studies of microbiota in Coleoptera using 16S deep sequencing is difficult. Diversity estimates are affected by sample preparation, the use of entire body or gut only, amplicon length, choice of region to sequence and quality control procedures (Contreras et al., 2010; Schloss et al., 2011; Bokulich et al., 2013; He et al., 2013; Chandler et al., 2014). For example, diversity in our field samples was dissimilar from those of other Coleoptera (Jones et al., 2013). Both studies used the entire body of insects, but the preservation method was different: we preserved our samples dry at -20ºC until DNA was extracted; Jones et al. (2013) conserved the samples in 70% ethanol at -20ºC and we used hypervariable region V4; they used regions V1 and V3. Beta diversity analyses showed that bacterial communities of the CBB can be affected by host s diet, stage of development and environment which it develops. When we combined the samples from wild and laboratory reared individuals, we observed that diet was the most important factor that explained the variation between communities. Diet has been previously 109

126 suggested as a major determinant of bacterial community composition in insects (Chandler et al., 2011; Kelley and Dobler, 2011; Colman et al., 2012; Staubach et al., 2013; Aksoy et al., 2014; Montagna et al., 2015). Alterations in host s diet can change completely the community composition and structure (Dillon and Dillon, 2004; Colman et al., 2012; Staubach et al., 2013). In a second PCoA we analyzed only the bacterial communities from wild individuals from coffee fruits in sun vs. shade plots. The results from these analyses showed that the host s stage of development explained the 71% of variation between communities and separates the bacterial communities from wild individuals according to host s stage of development (Figure 4). Changes in diversity, composition and structure of bacterial communities have observed across life stages of several insect species (Vasanthakumar et al., 2008; Wang et al., 2011; Brucker and Bordenstein, 2012; Martinson et al., 2012). For example, bacterial communities of larvae of Nasonia species were simple and less diverse than those from pupae and adults (Brucker and Bordenstein, 2012). This suggests a progressive colonization during CBB development; as mentioned above, communities from eggs are formed mainly by heritable symbionts and those that adhere to the outer membrane during the oviposition; while larvae can acquire other bacteria from the environment during alimentation and bacterial communities from adults will be most complex and shaped by diet, environmental conditions and age of adults. For example, Wong et al. (2011) found changes in bacterial composition according with the increase in the age of Drosophila adult hosts. In our case the adult females that we obtained from artificial diets were the same age, however, those collected in field were of unknown age, although we collected only the colonizing females or females that were starting to drill the coffee fruits. In the second PCoA analysis, PC2 explained 19% of variation and separated bacterial communities of eggs and adults fruits of sun coffee vs. those from shade coffee (Figure 4). 110

127 Environment can exert selective pressures in which only those microorganisms that are active and can grow in a particular environment will be present and increase in number (Fuhrman, 2009). We observed that the differences in environmental conditions between sun and shade (e.g. temperature, relative humidity and food quality) influenced bacterial community composition and structure but not overall diversity, as have been suggested before by Jones et al. (2013)(Table 2). 5. Conclusions Novel CBB management strategies could be based on disruption of symbionts that are required for nutrition and reproduction or those that may confer protection against natural enemies. However, more research is necessary in order to clarify the role in the biology of the CBB of these bacteria. 6. Acknowledgments Special thanks to Dr. José Carlos Verle Rodrigues for access to the Center for Excellence in Quarantine and Invasive Species at University of Puerto Rico in San Juan, Puerto Rico to rear CBBs. Thanks to Michelle Cruz for outstanding work in the laboratory and Dr. Filipa Godoy for invaluable advice on metagenomics and QIIME. This work was supported by USDA Cooperative Agreement (PI Stephen A. Rehner, USDA- ARS Systematic Mycology and Microbiology Laboratory, Beltsville MD). Thanks to UPR Sequencing and Genotyping facility, supported by NCRR AABRE grant #P20 RR16470; especially to Msc. Silvia Planas for her patience, dedication and careful handling of the samples. 7. References Aksoy, E., Telleria, E.L., Echodu, R., Wu, Y., Okedi, L.M., Weiss, B.L., Aksoy, S., Caccone, A., Analysis of multiple tsetse fly populations in Uganda reveals limited diversity and species-specific gut microbiota. Appl. Environ. Microbiol. 80,

128 Armbrecht, I., Gallego, M.C., Testing ant predation on the coffee berry borer in shaded and sun coffee plantations in Colombia. Entomol. Exp. Appl. 124, Armbrecht, I., Perfecto, I., Litter-twig dwelling ant species richness and predation potential within a forest fragment and neighboring coffee plantations of contrasting habitat quality in Mexico. Agric. Ecosyst. Environ. 97, Baumann, P., Biology of bacteriocyte-associated endosymbionts of plant sap-sucking insects. Annu. Rev. Microbiol. 59, Blackburn, M.B., Gundersen-Rindal, D.E., Weber, D.C., Martin, P.A., Farrar, R.R., Enteric bacteria of field-collected Colorado potato beetle larvae inhibit growth of the entomopathogens Photorhabdus temperata and Beauveria bassiana. Biolog. Control 46, Bokulich, N.A., Subramanian, S., Faith, J.J., Gevers, D., Gordon, J.I., Knight, R., Mills, D.A., Caporaso, J.G., Quality-filtering vastly improves diversity estimates from Illumina amplicon sequencing. Nat. Methods 10, Braig, H., Turner, B., Perotti, M., Symbiotic Rickettsia. Insect Symbiosis 3. Bray, J.R., Curtis, J.T., An ordination of the upland forest communities of southern Wisconsin. Ecol. Monogr. 27, Brownlie, J.C., Johnson, K.N., Symbiont-mediated protection in insect hosts. Trends Microbiol. 17, Brucker, R.M., Bordenstein, S.R., The roles of host evolutionary relationships (genus: Nasonia) and development in structuring microbial communities. Evolution 66, Brul, S., Coote, P., Preservative agents in foods: mode of action and microbial resistance mechanisms. Int. J. Food Microbiol. 50, Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., Fierer, N., Pena, A.G., Goodrich, J.K., Gordon, J.I., QIIME allows analysis of highthroughput community sequencing data. Nature methods 7, Caporaso, J.G., Lauber, C.L., Walters, W.A., Berg-Lyons, D., Huntley, J., Fierer, N., Owens, S.M., Betley, J., Fraser, L., Bauer, M., Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. ISME J 6, Ceja-Navarro, J.A., Vega, F.E., Hao, Z., Lim, H., Kosina, P., Infante, F., Brodie, E.L., Gut microbiota mediated caffeine dexotification in the primary insect pest of coffee. Submitted. Chandler, J.A., James, P.M., Jospin, G., Lang, J.M., The bacterial communities of Drosophila suzukii collected from undamaged cherries. PeerJ 2, e474. Chandler, J.A., Lang, J.M., Bhatnagar, S., Eisen, J.A., Kopp, A., Bacterial communities of diverse Drosophila species: ecological context of a host microbe model system. PLoS Genet. 7, e

129 Chao, A., Nonparametric estimation of the number of classes in a population. Scand. J. Stat., Chen, S.-J., Lu, F., Cheng, J.-A., Jiang, M.-X., Way, M.O., Identification and biological role of the endosymbionts Wolbachia in rice water weevil (Coleoptera: Curculionidae). Environ. Entomol. 41, Colman, D., Toolson, E., Takacs-Vesbach, C., Do diet and taxonomy influence insect gut bacterial communities?. Mol. Ecol. 21, Contreras, M., Costello, E.K., Hidalgo, G., Magris, M., Knight, R., Dominguez-Bello, M.G., The bacterial microbiota in the oral mucosa of rural Amerindians. Microbiol. 156, Crotti, E., Rizzi, A., Chouaia, B., Ricci, I., Favia, G., Alma, A., Sacchi, L., Bourtzis, K., Mandrioli, M., Cherif, A., Acetic acid bacteria, newly emerging symbionts of insects. Appl. Environ. Microbiol. 76, Damon, A., A review of the biology and control of the coffee berry borer, Hypothenemus hampei (Coleoptera : Scolytidae). B Entomol Res 90, Dang, H., Ren, J., Song, L., Sun, S., An, L., Diverse tetracycline resistant bacteria and resistance genes from coastal waters of Jiaozhou Bay. Microb. Ecol. 55, Davidson, P., Taylor, T., Doyle, M., Beuchat, L., Chemical peservatives and natural antimicrobial compounds. Food Microbiol., De la Rosa, W., Alatorre, R., Trujillo, J., Barrera, J., Virulence of Beauveria bassiana (Deuteromycetes) strains against the coffee berry borer (Coleoptera: Scolytidae). J Econ Entomol 90, Dedeine, F., Vavre, F., Fleury, F., Loppin, B., Hochberg, M.E., Boulétreau, M., Removing symbiotic Wolbachia bacteria specifically inhibits oogenesis in a parasitic wasp. Proc. Natl. Acad. Sci. U.S.A. 98, Dillon, R., Dillon, V., The gut bacteria of insects: nonpathogenic interactions. Ann. Rev. Entomol. 49, Dong, P., Wang, J.-J., Zhao, Z.-M., Wolbachia infection and its influence on the reproduction of stored-product psocid, Liposcelis tricolor (Psocoptera: Liposcelididae). J. Insect. Sci. 6. Douglas, A., The microbial dimension in insect nutritional ecology. Funct. Ecol. 23, Douglas, A.E., Symbiotic microorganisms: untapped resources for insect pest control. Trends Biotechnol. 25, Egert, M., Stingl, U., Bruun, L.D., Pommerenke, B., Brune, A., Friedrich, M.W., Structure and topology of microbial communities in the major gut compartments of 113

130 Melolontha melolontha larvae (Coleoptera: Scarabaeidae). Appl. Environ. Microbiol. 71, Fenollar, F., Maurin, M., Raoult, D., Wolbachia pipientis growth kinetics and susceptibilities to 13 antibiotics determined by immunofluorescence staining and real-time PCR. Antimicrob. Agents Chemother. 47, Fuhrman, J.A., Microbial community structure and its functional implications. Nature 459, Fukatsu, T., Hosokawa, T., Capsule-transmitted gut symbiotic bacterium of the Japanese common plataspid stinkbug, Megacopta punctatissima. Appl. Environ. Microbiol. 68, Gauglitz, J.M., Zhou, H., Butler, A., A suite of citrate-derived siderophores from a marine Vibrio species isolated following the Deepwater Horizon oil spill. J. Inorg. Biochem. 107, Geib, S.M., del Mar Jimenez-Gasco, M., Carlson, J.E., Tien, M., Hoover, K., Effect of host tree species on cellulase activity and bacterial community composition in the gut of larval Asian longhorned beetle. Environ. Entomol. 38, Goto, S., Anbutsu, H., Fukatsu, T., Asymmetrical interactions between Wolbachia and Spiroplasma endosymbionts coexisting in the same insect host. Appl. Environ. Microbiol. 72, Gotoh, T., Noda, H., Ito, S., Cardinium symbionts cause cytoplasmic incompatibility in spider mites. Heredity 98, Gross, H., Loper, J.E., Genomics of secondary metabolite production by Pseudomonas spp. Nat. Prod. Rep. 26, Hail, D., Lauzìere, I., Dowd, S.E., Bextine, B., Culture independent survey of the microbiota of the glassy-winged sharpshooter (Homalodisca vitripennis) using 454 pyrosequencing. Environ. Entomol. 40, Haine, E.R., Symbiont-mediated protection. Proc. R. Soc. Lond. B Biol. Sci. 275, Haraprasad, N., Niranjana, S., Prakash, H., Shetty, H.S., Wahab, S., Beauveria bassiana a potential mycopesticide for the efficient control of coffee berry borer, Hypothenemus hampei (Ferrari) in India. Biocontrol Sci Tech 11, He, Y., Zhou, B.-J., Deng, G.-H., Jiang, X.-T., Zhang, H., Zhou, H.-W., Comparison of microbial diversity determined with the same variable tag sequence extracted from two different PCR amplicons. BMC Microbiol. 13, 208. Hunter, M.S., Perlman, S.J., Kelly, S.E., A bacterial symbiont in the Bacteroidetes induces cytoplasmic incompatibility in the parasitoid wasp Encarsia pergandiella. Proc. R. Soc. Lond. B Biol. Sci. 270,

131 Hurst, G.D., Jiggins, F.M., Robinson, S.J., What causes inefficient transmission of male-killing Wolbachia in Drosophila?. Heredity 87, Hurst, G.D., Johnson, A.P., vd Schulenburg, J.H.G., Fuyama, Y., Male-killing Wolbachia in Drosophila: a temperature-sensitive trait with a threshold bacterial density. Genetics 156, Indiragandhi, P., Anandham, R., Madhaiyan, M., Kim, G.-H., Sa, T., Cross-utilization and expression of outer membrane receptor proteins for siderophore uptake by Diamondback moth Plutella xylostella (Lepidoptera: Plutellidae) gut bacteria. FEMS Microbiol. Lett. 289, Indiragandhi, P., Anandham, R., Madhaiyan, M., Poonguzhali, S., Kim, G., Saravanan, V., Sa, T., Cultivable bacteria associated with larval gut of prothiofos-resistant, prothiofossusceptible and field-caught populations of diamondback moth, Plutella xylostella and their potential for, antagonism towards entomopathogenic fungi and host insect nutrition. J. Appl. Microbiol. 103, Indiragandhi, P., Anandham, R., Sa, T.M., Functional Significance of Insect Gut Bacteria and Their Role in Host Insect Processes, Development, and Crop Production. Bacteria in Agrobiology: Plant Growth Responses. Springer, pp Jaramillo, J., Borgemeister, C., Baker, P., Coffee berry borer Hypothenemus hampei (Coleoptera: Curculionidae): searching for sustainable control strategies. B Entomol Res 96, Jaramillo, J., Muchugu, E., Vega, F.E., Davis, A., Borgemeister, C., Chabi-Olaye, A., Some like it hot: The influence and implications of climatic change on coffee berry borer (Hypothenemus hampei) and coffee production in east Africa. Plos One 4, Jones, R.T., Sanchez, L.G., Fierer, N., A cross-taxon analysis of insect-associated bacterial diversity. PloS one 8, e Kelley, S.T., Dobler, S., Comparative analysis of microbial diversity in Longitarsus flea beetles (Coleoptera: Chrysomelidae). Genetica 139, Kieboom, J., de Bont, J.A., Identification and molecular characterization of an efflux system involved in Pseudomonas putida S12 multidrug resistance. Microbiol. 147, Kikuchi, Y., Hosokawa, T., Fukatsu, T., Insect-microbe mutualism without vertical transmission: a stinkbug acquires a beneficial gut symbiont from the environment every generation. Appl. Environ. Microbiol. 73, Ley, R.E., Lozupone, C.A., Hamady, M., Knight, R., Gordon, J.I., Worlds within worlds: evolution of the vertebrate gut microbiota. Nat. Rev. Microbiol. 6, Li, X.-Z., Livermore, D.M., Nikaido, H., Role of efflux pump (s) in intrinsic resistance of Pseudomonas aeruginosa: resistance to tetracycline, chloramphenicol, and norfloxacin. Antimicrob. Agents Chemother. 38,

132 Lozupone, C., Knight, R., UniFrac: a new phylogenetic method for comparing microbial communities. Appl. Environ. Microbiol. 71, Maestri, M., Barros, R.S., Coffee. In: Alvim, P., Koslowski, T. (Eds.), Ecophysiology of Tropical Crops. American Press, London, pp Mariño, Y., Perez, M.E., Gallardo, F., Trifilio, M., Cruz, M., Bayman, P., Submitted. Sun vs. shade affects infestation, reproduction and sex ratio of the coffee berry borer (Hypothenemus hampei) in Puerto Rico. Agric. Ecosyst. Environ. Mariño, Y., Verle, J.C., Bayman, P., In Revision. Wolbachia affects the reproduction, sex ratio and population dynamics in the Coffee Berry Borer Hypothenemus hampei (Coleoptera: Curculionidae). Mariño, Y., Zapata, M., Diversidad de bacterias en cicadélidos vectores potenciales de Xylella fastidiosa Wells et al. en cafetos de Puerto Rico. J. Agric. Univ. PR 93, Martinson, V.G., Moy, J., Moran, N.A., Establishment of characteristic gut bacteria during development of the honeybee worker. Appl. Environ. Microbiol. 78, Méndez-López, I., Basurto-Ríos, R., Ibarra, J.E., Bacillus thuringiensis serovar israelensis is highly toxic to the coffee berry borer, Hypothenemus hampei Ferr.(Coleoptera: Scolytidae). FEMS Microbiol. Lett. 226, Montagna, M., Chouaia, B., Mazza, G., Prosdocimi, E.M., Crotti, E., Mereghetti, V., Vacchini, V., Giorgi, A., De Biase, A., Longo, S., Effects of the Diet on the Microbiota of the Red Palm Weevil (Coleoptera: Dryophthoridae). PloS one 10, e Moran, N.A., Hansen, A.K., Powell, J.E., Sabree, Z.L., Distinctive gut microbiota of honey bees assessed using deep sampling from individual worker bees. PloS one 7, e Moran, N.A., McCutcheon, J.P., Nakabachi, A., Genomics and evolution of heritable bacterial symbionts. Ann. Rev. Genet. 42, Moran, N.A., Russell, J.A., Koga, R., Fukatsu, T., Evolutionary relationships of three new species of Enterobacteriaceae living as symbionts of aphids and other insects. Appl. Environ. Microbiol. 71, Nachappa, P., Levy, J., Pierson, E., Tamborindeguy, C., Diversity of endosymbionts in the potato psyllid, Bactericera cockerelli (Hemiptera: Triozidae), vector of zebra chip disease of potato. Curr. Microbiol. 62, Oliver, K.M., Moran, N.A., Hunter, M.S., Variation in resistance to parasitism in aphids is due to symbionts not host genotype. Proc. Natl. Acad. Sci. U.S.A. 102, Oliver, K.M., Russell, J.A., Moran, N.A., Hunter, M.S., Facultative bacterial symbionts in aphids confer resistance to parasitic wasps. Proc. Natl. Acad. Sci. U.S.A. 100, Osei Poku, J., Mbogo, C., Palmer, W., Jiggins, F., Deep sequencing reveals extensive variation in the gut microbiota of wild mosquitoes from Kenya. Mol. Ecol. 21,

133 Perotti, M.A., Clarke, H.K., Turner, B.D., Braig, H.R., Rickettsia as obligate and mycetomic bacteria. FASEB J. 20, Portilla, M., Mass rearing technique for Cephalonomia stephanoderis (Hymenoptera: Bethylidae) on Hypothenemus hampei (Coleoptera: Scolytidae) developed using Cenibroca artificial diet. Revista colombiana de Entomologia 25, Rani, A., Sharma, A., Rajagopal, R., Adak, T., Bhatnagar, R.K., Bacterial diversity analysis of larvae and adult midgut microflora using culture-dependent and cultureindependent methods in lab-reared and field-collected Anopheles stephensi-an Asian malarial vector. BMC Microbiol. 9, 96. Rousk, J., Demoling, L.A., Bååth, E., Contrasting short-term antibiotic effects on respiration and bacterial growth compromises the validity of the selective respiratory inhibition technique to distinguish fungi and bacteria. Microb. Ecol. 58, Russell, J.A., Moran, N.A., Costs and benefits of symbiont infection in aphids: variation among symbionts and across temperatures. Proc. R. Soc. Lond. B Biol. Sci. 273, Schloss, P.D., Gevers, D., Westcott, S.L., Reducing the effects of PCR amplification and sequencing artifacts on 16S rrna-based studies. PloS one 6, e Shannon, C.E., A mathematical theory of communication. Bell Syst. Tech. J. 27, Silva, F.C.d., Ventura, M.U., Morales, L., Capture of Hypothenemus hampei Ferrari (Coleoptera, Scolytidae) in response to trap characteristics. Sci. Agric. 63, Son, Y., Luckhart, S., Zhang, X., Lieber, M.J., Lewis, E.E., Effects and implications of antibiotic treatment on Wolbachia-infected vine weevil (Coleoptera: Curculionidae). Agric. Forest Entomol. 10, Soto-Pinto, L., Perfecto, I., Caballero-Nieto, J., Shade over coffee: its effects on berry borer, leaf rust and spontaneous herbs in Chiapas, Mexico. Agroforest Syst 55, Staubach, F., Baines, J.F., Künzel, S., Bik, E.M., Petrov, D.A., Host species and environmental effects on bacterial communities associated with Drosophila in the laboratory and in the natural environment. PloS one 8, e Stouthamer, R., Breeuwer, J.A., Hurst, G.D., Wolbachia pipientis: microbial manipulator of arthropod reproduction. Ann. Rev. Microbiol. 53, Stouthamer, R., Hurst, G.D., Breeuwer, J.A., Sex ratio distorters and their detection. Sex ratios, concept and research methods. Cambridge Univ. Press, Cambridge, Team, R., Develoment Core. "R Foundation for Statistical Computing". Vienna, Austria

134 Teixeira, L., Ferreira, Á., Ashburner, M., The bacterial symbiont Wolbachia induces resistance to RNA viral infections in Drosophila melanogaster. PLoS Biol. 6, e Teodoro, A., Klein, A.-M., Tscharntke, T., Environmentally mediated coffee pest densities in relation to agroforestry management, using hierarchical partitioning analyses. Agric Ecosyst Environ 125, Vasanthakumar, A., Handelsman, J., Schloss, P.D., Bauer, L.S., Raffa, K.F., Gut microbiota of an invasive subcortical beetle, Agrilus planipennis Fairmaire, across various life stages. Environ. Entomol. 37, Vega, F.E., Benavides, P., Stuart, J.A., O'Neill, S.L., Wolbachia infection in the coffee berry borer (Coleoptera: Scolytidae). Ann Entomol Soc Am 95, Vega, F.E., Infante, F., Castillo, A., Jaramillo, J., The coffee berry borer, Hypothenemus hampei (Ferrari)(Coleoptera: Curculionidae): a short review, with recent findings and future research directions. Terr Arthropod Rev 2, Vega, F.E., Infante, F., Johnson, A.J., The Genus Hypothenemus, with emphasis on H. hampei, the coffee berry borer. In: Vega, F.E., Hofstetter, R. (Eds.), Bark bettles: Biology and Ecology of native and invasive species. Elsevier, San Diego, California, pp Vega, F.E., Pava Ripoll, M., Posada, F., Buyer, J.S., Endophytic bacteria in Coffea arabica L. J. Basic Microbiol. 45, Vega, F.E., Posada, F., Aime, M.C., Pava-Ripoll, M., Infante, F., Rehner, S.A., Entomopathogenic fungal endophytes. Biol Control 46, Wang, Q., Garrity, G.M., Tiedje, J.M., Cole, J.R., Naive Bayesian classifier for rapid assignment of rrna sequences into the new bacterial taxonomy. Appl. Environ. Microbiol. 73, Wang, Y., Gilbreath III, T.M., Kukutla, P., Yan, G., Xu, J., Dynamic gut microbiome across life history of the malaria mosquito Anopheles gambiae in Kenya. PloS one 6, e Weeks, A.R., Velten, R., Stouthamer, R., Incidence of a New Sex ratio distorting Endosymbiotic Bacterium among arthropods. Proc. R. Soc. Lond. B Biol. Sci. 270, Weiss, B., Aksoy, S., Microbiome influences on insect host vector competence. Trends Parasitol. 27, Wernegreen, J.J., Genome evolution in bacterial endosymbionts of insects. Nature Rev. Genet. 3, Werren, J.H., Baldo, L., Clark, M.E., Wolbachia: master manipulators of invertebrate biology. Nat. Rev. Microbiol. 6, Wong, A.C., Ng, P. and Douglas, A.E Low-diversity bacterial community in the gut of the fruitfly Drosophila melanogaster. Environ. Microbiol. 13 (7):

135 Wong, A.C., Chaston, J.M., Douglas, A.E., The inconstant gut microbiota of Drosophila species revealed by 16S rrna gene analysis. ISME J. 7, Zchori-Fein, E., Gottlieb, Y., Kelly, S., Brown, J., Wilson, J., Karr, T., Hunter, M., A newly discovered bacterium associated with parthenogenesis and a change in host selection behavior in parasitoid wasps. Proc. Natl. Acad. Sci. U.S.A. 98, Zchori-Fein, E., Perlman, S.J., Distribution of the bacterial symbiont Cardinium in arthropods. Mol. Ecol. 13, Zchori Fein, E., Borad, C., Harari, A.R., Oogenesis in the date stone beetle, Coccotrypes dactyliperda, depends on symbiotic bacteria. Physiol. Entomol. 31,

136 Table 1. Frequencies of the most abundant OTUs in the CBB Hypothenemus hampei. Treatments show differences between CBBs from the field vs. artificial diets, from sun vs. shade coffee, eggs vs. adult females, and control diets vs. diets with tetracycline. OTUs are identified by the closet hit in the Ribosomal Database Project (RDP) Classifier. Number of sequences is after all quality control steps. Taxonomy Total Eggs sun a Eggs shade a Females sun a Females shade a Eggs control b Eggs tetracycline b Females control b Females tetracycline b Actinobacteria Brachybacterium Rhodococcus Bacteriodetes Sphingobacterium Pedobacter Firmicutes Enterococcus Turicibacteraceae Proteobacteria Alphaproteobacteria Agrobacterium Ochrobactrum Rhizobium Gammaproteobacteria Pectobacterium Trabulsiella Enterobacteriaceae other Acinetobacter Pseudomonas Pseudomonadaceae- other Rhodanobacter Stenotrophomonas Xanthamonadaceae other Wolbachia Rickettsiaceae other All other taxa Total number of sequences in sample a Stages collected from infested coffee fruits in field b Stages from F 11 generation reared in artificial diet Cenibroca; control: diet without antibiotic; T 0.1% : diet with 0.1% tetracycline (w/v)

137 Table 2. Alpha diversity indexes for each treatment of Hypothenemus hampei (Mean ± S.E. are shown) Treatment Shannon Chao 1 Field, Eggs, sun 3.96 ± 0.12 a ± ab Field, Eggs, shade 3.92 ± 0.33 a ± ab Field, Females, sun 3.47 ± 0.28 ab ± ab Field, Females, shade 3.85 ± 0.65 a ± a Laboratory, Eggs, control 1.42 ± 0.09 cd ± b Laboratory, Eggs, tetracycline 2.60 ± 0.46 abc ± ab Laboratory, Adults, control 1.94 ± 0.19 bcd ± b Laboratory, Adults, tetracycline 0.97 ± 0.07 d ± b Field samples collected from infested coffee fruits, sun: coffee plants are growing exposed to direct sunlight and shade: coffee plants are growing under shade provided by other trees. 121

138 Figure 1. Bacterial phyla in Hypothenemus hampei microbiota comparing eggs vs. adult females, Field samples vs. laboratory-reared on artificial diets, and shade vs. sun coffee. Laboratory samples were the eleventh generation reared in artificial diet Cenibroca; control: diet without antibiotic and tetracycline: diet with 0.1% tetracycline (w/v). 122