Impacts of Climate Variability on Stemborers-Natural Enemies Trophic Level Interaction: Implications for Biological Control Under Global Change

IMPACTS OF CLIMATE VARIABILITY ON STEMBORERS-NATURAL ENEMIES TROPHIC LEVEL INTERACTION: IMPLICATIONS FOR BIOLOGICAL CONTROL UNDER GLOBAL CHANGE

REYARD MUTAMISWA Reg. No: 13100028 BSc Natural Resources Management & Agriculture (Hons) Crop Production & Horticulture (MSU), MSc Tropical Entomology (UZ)

Department of Biological Sciences and Biotechnology, Faculty of Science, Botswana International University of Science and Technology [email protected]

A Thesis Submitted to the Faculty of Science in Partial Fulfilment of the Requirements for the Doctor of Philosophy in Biological Sciences of BIUST

Supervisor: Dr Casper Nyamukondiwa Department of Biological Sciences and Biotechnology, Faculty of Science, BIUST [email protected]

December, 2017

DECLARATION AND COPYRIGHT

I, REYARD MUTAMISWA, declare that this thesis is my own original work and that it has not been presented and will not be presented to any other university for a similar or any other degree award.

Signature:

This thesis is copyright material protected under the Berne Convention, the Copyright Act of 1999 and other international and national enactments, in that behalf, on intellectual property. It must not be reproduced by any means, in full or in part, except for short extracts in fair dealing; for researcher private study, critical scholarly review or discourse with an acknowledgement, without the written permission of the office of the Provost, on behalf of both the author and BIUST

CERTIFICATION

The undersigned certifies that he has read and hereby recommends for acceptance by the Faculty of Science a thesis titled: Impacts of Climate Variability on Stemborers-Natural Enemies Trophic Level Interaction: Implications for Biological Control Under Global Change, in fulfilment of the requirements for the degree of Doctor of Philosophy in Biological Sciences (Applied Entomology) of BIUST.

………………………………… Dr. Casper Nyamukondiwa (Supervisor)

Date: 03 February 2018

iii

DEDICATION

This thesis is dedicated to my family and my mom. Your prayers, support and love during the course of this study are highly appreciated.

ACKNOWLEDGEMENTS

Firstly I would like to thank Almighty God for giving me strength and wisdom throughout this study. Secondly, my sincere gratitude goes to my supervisor Dr. Casper Nyamukondiwa for his unwavering support, guidance and mentorship throughout the study period. This research would not have been possible without your patience, excellent supervision and encouragement. I sincerely thank Botswana International University of Science and Technology (BIUST) for granting me this opportunity to pursue my studies and financial support. Special thanks also go to International Centre for Insect Physiology and Ecology (ICIPE), South African Sugarcane Research Institute (SASRI) for provision of insects that I used for the entire duration of this study. I would also like to extend my gratitude to Ministry of Agriculture, Botswana for provision of import permits for the insects that I used in this research. I also express my sincere gratitude to Dr. Frank Chidawanyika for editing and advice in the write up and several anonymous referees for the constructive comments on earlier versions of different chapters of this thesis. Many thanks to my laboratory colleagues Eva Moeng, Honest Machekano, Mphoeng Ofithile, Mmabaledi Buxton, Nonofo Gotcha, Rebaone Motswagole and Biological Sciences department technical team for their support throughout my study. Finally, I also extend my sincere appreciation to my family and my mom for their unconditional support in my life. May God richly bless you.

TABLE OF CONTENTS

Page

DECLARATION AND COPYRIGHT...... ii CERTIFICATION ...... iii DEDICATION ...... iv ACKNOWLEDGEMENTS ...... v LIST OF TABLES ...... xii LIST OF FIGURES ...... xv ABSTRACT ...... xx

CHAPTER 1 General Introduction ...... 1 1.0 Climate change and insect phenology ...... 2 1.1 Insect Thermal Biology and Physiology ...... 3 1.2 Insect freeze strategies and supercooling points (SCPs) ...... 9 1.3 Phenotypic plasticity in insects as a response to changes in temperature ...... 11 1.4 Economic Importance of Cereal Stemborers ...... 12 1.4.1 Biology and Ecology of Busseola fusca ...... 13 1.4.2 Biology and Ecology of Sesamia calamistis ...... 13 1.4.3 Biology and Ecology of Chilo partellus ...... 14 1.5 Control of Cereal Stemborers ...... 15 1.5.1 Chemical Control ...... 15 1.5.2 Cultural Control ...... 16 1.5.3 Push - Pull Technology...... 17 1.5.4 Biological control ...... 18 1.6 Biology and ecology of Cotesia flavipes and Cotesia sesamiae ...... 18

1.7 Problem statement and justification of the study ...... 19 1.8 Objectives ...... 22 1.8.1 Overall Objective ...... 22 1.8.2 Specific Objectives ...... 23 1.8.3 Hypotheses...... 23 References ...... 23

CHAPTER 2 Dominance of spotted stemborer, Chilo partellus Swinhoe (Lepidoptera: Crambidae) over indigenous stemborer species in Africa’s changing climates: ecological and thermal biology perspectives ...... 39 2.1 Introduction ...... 40 2.2 Africa’s Changing Climate...... 45 2.3 Competitive Displacement of Indigenous Stemborer Species ...... 46 2.3.1 Extension of C. partellus geographical range ...... 48 2.3.2 Increase in numbers of C. partellus generations ...... 50 2.3.3 Overwintering of C. partellus ...... 53 2.3.4 Impact of Variability in Temperature and Relative Humidity on the Ecology of Stemborers ...... 55 2.3.5 Physiological Tolerances in C. partellus ...... 58 2.3.6 Host range of C. partellus ...... 65 2.3.7 Asynchrony between C. partellus and its natural enemies ...... 66 2.4 Future Directions and Conclusions ...... 67 References ...... 70

CHAPTER 3 Comparative assessment of the thermal tolerance of spotted stemborer, Chilo partellus Swinhoe (Lepidoptera: Crambidae) and its larval parasitoid, Cotesia sesamiae Cameron (Hymenoptera: Braconidae)…… ...... 81 3.1 Introduction ...... 82 3.2 Materials and Methods ...... 85 3.2.1 Insect Culture ...... 85

vii

3.2.2 Upper and Lower Lethal Temperature Assays ...... 85 3.2.3 Effect of Ramping rate on Critical Thermal Limits (CTLs)...... 86 3.2.4 Supercooling points (SCPs) ...... 87 3.2.5 Heat knock-down time (HKDT) and Chill-coma recovery time (CCRT) ...... 88 3.2.6 Statistical analyses ...... 88 3.3 Results ...... 89 3.3.1 Upper and Lower Lethal Temperature Assays ...... 89 3.3.2 Effect of Ramping rate on Critical Thermal Limits (CTLs)...... 95 3.3.3 Supercooling Points (SCPs) ...... 97 3.3.4 Heat knock-down time and Chill-coma recovery time ...... 97 3.4 Discussion ...... 99 References ...... 104

CHAPTER 4 Thermal plasticity mediates the interaction between host Chilo partellus Swinhoe (Lepidoptera: Crambidae) and endoparasitoid Cotesia flavipes Cameron (Hymenoptera: Braconidae) under rapidly changing environments ...... 113 4.1 Introduction ...... 114 4.2 Material and Methods...... 117 4.2.1 Study insects and rearing conditions ...... 117 4.2.2 Phenotypic Plasticity of Low Temperature Tolerance ...... 118 4.2.3 Phenotypic Plasticity of High Temperature Tolerance ...... 121 4.2.4 Microclimate data...... 122 4.2.5 Statistical analyses ...... 122 4.3 Results ...... 123 4.3.1 Phenotypic Plasticity of Low Temperature Tolerance ...... 123 4.3.2 Phenotypic Plasticity of High Temperature Tolerance ...... 133 4.3.3 Microclimate data...... 139 4.4 Discussion ...... 141 4.4.1 Phenotypic Plasticity of Low Temperature Tolerance ...... 141 4.4.2 Phenotypic Plasticity of High Temperature Tolerance ...... 144

viii

References ...... 147

CHAPTER 5 Thermal resilience may shape population abundance of two sympatric congeneric Cotesia species (Hymenoptera: Braconidae) ...... 156 5.1 Introduction ...... 157 5.2 Materials and Methods ...... 159 5.2.1 Study Organisms and Rearing Conditions ...... 159 5.2.2 Lower and Upper Lethal Temperature Assays ...... 160 5.2.3 Critical Thermal Limits (CTLs) ...... 161 5.2.4 Supercooling Points (SCPs) ...... 161 5.2.5 Chill Coma Recovery Time (CCRT) ...... 162 5.2.6 Heat Knock down Time (HKDT) ...... 162 5.2.7 Statistical Analyses ...... 163 5.3 Results ...... 163 5.3.1 Lower and Upper Lethal Temperature Assays ...... 163 5.3.2 Critical Thermal Limits (CTLs) ...... 168 5.3.3 Supercooling Points (SCPs) ...... 170 5.3.4 Chill Coma Recovery Time (CCRT) ...... 170 5.3.5 Heat Knockdown Time (HKDT) ...... 171 5.4 Discussion ...... 172 References ...... 176

CHAPTER 6 Superior basal and plastic thermal responses to environmental heterogeneity in invasive exotic Chilo partellus Swinhoe over indigenous Busseola fusca (Fuller) and Sesamia calamistis Hampson...... 185 6.1 Introduction ...... 186 6.2 Materials and Methods ...... 189

6.2.1 Insect Culture and Rearing Conditions ...... 189 6.2.3 Effects of Thermal Acclimation on CTLs ...... 190 6.2.4 Effects of Starvation Acclimation on CTLs...... 191 6.2.5 Effects of Hardening on CTLs ...... 191 6.2.7 Statistical Analyses ...... 192 6.3 Results ...... 193 6.3.1 Thermal Tolerance Assays ...... 193 6.3.2 Effects of Thermal Acclimation on CTLs ...... 195 6.3.3 Effects of Starvation Acclimation on CTLs...... 197 6.3.4 Effects of Hardening on CTLs ...... 199 6.3.5 Effects of Desiccation Acclimation on CTLs ...... 203 6.4 Discussion ...... 205 References ...... 209

CHAPTER 7 General Discussion ...... 220 References ...... 226

APPENDIX 1 Diversity and abundance of lepidopteran stem borer natural enemies in natural and cultivated habitats in Botswana ...... 230 1.1 Introduction ...... 231 1.2 Materials and methods ...... 233 1.2.1. Field surveys ...... 233 1.2.2. Laboratory incubation ...... 236 1.2.3. Data analysis ...... 236 1.3 Results ...... 237 1.3.1. Natural enemy diversity, stemborer hosts and host plants ...... 237 1.3.2. Stemborer parasitism in cultivated and wild host plants ...... 249 1.3.3 Relative abundance of parasitoid species ...... 251 1.4 Discussion ...... 253

1.4.1. Natural enemy diversity, stemborer hosts and host plants ...... 253 1.4.2 Parasitism levels in cultivated and wild host plants...... 254 1.4.3. Relative abundance of parasitoid species ...... 255 References ...... 256

APPENDIX 2 First Report of Tomato Leaf miner, Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae) in Botswana ...... 265 2.1 Introduction ...... 266 2.2 Materials and Methods ...... 268 2.2.1 Morphological identification ...... 269 2.2.2 Plant damage symptoms ...... 269 2.2.3 Sex Pheromone Trapping ...... 269 2.2.4. Molecular analysis ...... 272 2.2.5. Data analysis ...... 272 2.3. Results ...... 273 2.3.1 Morphological features ...... 273 2.3.2 Crop damage symptoms...... 274 2.3.3 Pheromone trapping ...... 275 2.3.4 Molecular analysis ...... 278 2.4 Discussion ...... 279 2.5 Conclusion ...... 282 References ...... 282

LIST OF TABLES

Page Table 2.1: A compilation of generation time of four major cereal stemborer life cycles under optimum environmental conditions. Generation time was measured as the amount of time needed to complete development of one stemborer life cycle (from egg to adult)………………………51 Table 2.2: Average fecundity of different species of cereal stemborers (measured as the average total number of eggs produced per life time) under optimum environmental conditions of temperature, humidity and photoperiod………………………………………………………….53 Table 2.3: Mean diapause termination time for cereal stemborers. Termination time was defined as the amount of time needed to restore activity of a ‘sleeping’ stemborer following optimum temperature conditions…………………………………………………………………………………..54 Table 2.4: Lower and upper developmental temperature thresholds, developmental temperature range and mean developmental time at different temperatures for three major stemborers, (Chilo partellus, Sesamia calamistis and Busseola fusca) (modified from Khadioli et al., 2014a;b; Glatz et al., 2016)………………………………………………………………………………………56 Table 2.5: Results of basal thermal tolerance ±SD in three major stemborers, (B. fusca, S. calamistis and C. partellus) measured as (A) critical thermal maxima and (B) critical thermal minima. Means with the same superscript letter are not significantly different…………………….... ………………………………………………………………….62 Table 2.6: Summary results from a factorial ANOVA showing the effects of species (Chilo partellus, Busseola fusca and Sesamia calamistis) and temperature acclimation (low, optimum and high) on critical thermal maxima and minima (CTmax and CTmin, respectively). SS = sums of squares, d.f.=degrees of freedom………………………………………………………………...63 Table 3.1: Summary statistical results of the effects of temperature, duration of exposure and their interactions on the survival of Chilo partellus eggs following lower and upper lethal temperature treatments. Analysis were done using generalized linear models (GLZ) assuming binomial distribution with a logit link function in R version 3.3.0………………..…………….90 Table 3.2: Summary statistical results of the effects of temperature, duration of exposure and their interactions on the survival of Chilo partellus larvae following lower and upper lethal temperature treatments. Analysis were done using generalized linear models (GLZ) assuming binomial distribution with a logit link function in R version 3.3.0……………………………..91

xii

Table 4.1: Summary table showing species and developmental stage specific hardening and discriminating temperatures (2 h duration) for host, C. partellus and parasitoid, C. flavipes…118 Table 4.2: Summary results from a full factorial ANOVA showing the effects of species (C. partellus and C. flavipes) and temperature acclimation (low, optimum and high) on low temperature activity measured as CTmin, CCRT and SCP. SS = sums of squares, d.f.=degrees of freedom…………………………………………………………………………………………………131 Table 4.3: Summary results from a full factorial ANOVA showing the effects of C. partellus developmental stage (larvae and adults) and temperature acclimation (low, optimum and high) on low temperature activity measured as CTmin, CCRT and SCP. SS = sums of squares, d.f.= degrees of freedom……………………………………………………………………………………132 Table 4.4: Summary results from a full factorial ANOVA showing the effects of species (C. partellus and C. flavipes) and temperature acclimation (low, optimum and high) on high temperature activity measured as CTmax and HKDT. SS = sums of squares, d.f.=degrees of freedom…………………………………………………………………………………………………138 Table 4.5: Summary results from a full factorial ANOVA showing the effects of C. partellus developmental stage (larvae and adults) and temperature acclimation (low, optimum and high) on high temperature activity measured as CTmax and HKDT. SS = sums of squares, d.f.= degrees of freedom……………………………………………………………………………………………..139 Table 5.1: Summary statistical results of the effects of temperature, duration of exposure and their interactions on the survival of Cotesia sesamiae adults following lower and upper lethal temperature treatments. Analysis were done using generalized linear models (GLZ) assuming binomial distribution with a logit link function in R version 3.3.0…………………………….164 Table 5.2: Summary statistical results of the effects of temperature, duration of exposure and their interactions on the survival of Cotesia flavipes adults following lower and upper lethal temperature treatments. Analysis were done using generalized linear models (GLZ) assuming binomial distribution with a logit link function in R version 3.3.0…………………………….165 Table 5.3: Summary statistical results from full factorial ANOVA showing effects of species on critical thermal limits (CTmax and CTmin)……………………………………………………….168 Table 6.1: Summary table showing cold and heat hardening temperatures (2 h duration) for B. fusca, S. calamistis and C. partellus larvae……………………………………………….……192

xiii

Table 6.2: Summary statistical results from one-way ANOVA showing effects of species (B. fusca, S. calamistis and C. partellus) on CTmin and CTmax ………………………………..…..192 Table 6.3: Summary results from a two-way ANOVA showing the effects of species (B. fusca,

S. calamistis and C. partellus) and thermal acclimation (low, optimum and high) on CTmin and

CTmax. SS = sums of squares, d.f.=degrees of freedom……………………………………..…….195 Table 6.4: Summary statistical results from full factorial ANOVA showing effects of starvation acclimation on CTmin and CTmax for B. fusca, S. calamistis and C. partellus. SS = sums of squares, d.f.= degrees of freedom……………………………………………………………………197 Table 6.5: Summary statistical results from full factorial ANOVA showing effects of (A) RCH and (B) RHH on CTmin and CTmax for B. fusca, S. calamistis and C. partellus. SS = sums of squares, d.f.=degrees of freedom…………………………………………………………………….200 Table 6.6: Summary statistical results from full factorial ANOVA showing effects of desiccation acclimation on CTmin and CTmax for B. fusca, S. calamistis and C. partellus. SS = sums of squares, d.f.=degrees of freedom…………………………………………………………………….203

APPENDIX Table 1.1: Details of natural enemy species recovered from stemborers on cultivated and wild host plants in Botswana……………………………………………………………………..….239 Table 1.2: Natural enemies and their associated cultivated host plants in Botswana…….…....243 Table 1.3: Natural enemies and their associated wild host plants in Botswana………….….....245 Table 1.4: Larval and pupal parasitism in cultivated host plants in Botswana…………..….....250 Table 1.5: Larval and pupal parasitism in wild host plants in Botswana……………….…...…251 Table 2.1: Delimited detection zones for Tuta absoluta detection in Matshelagabedi village, Northern District of Botswana. Data was collected 7 days (03 December to 10 December 2016) and trapped moths were counted using dyed pointers and tally counters……………….…..….271

xiv

LIST OF FIGURES

Page Figure 1.1: Responses to temperature in a hypothetical organism (Adapted from Wharton, 2002;

Piyaphongkul, 2013). CTmin= critical thermal minima; CTmax = critical thermal maxima; CCT = chill coma temperature; HCT = heat coma temperature; LLT = lower lethal temperature; ULT = upper lethal temperature…………………………………………………………………………..4 Figure 1.2: The thermobiological scale, redrawn from Vannier (1994)………………………….6 Figure 1.3: Responses of insects cooled to low temperature, redrawn from (Denlinger & Lee, 2010)……………………………………………………………………………………………..10 Figure 2.1: Global distribution of cereal stemborers (A) Busseola fusca (B) Sesamia calamistis (C) Chilo partellus (D) Eldana saccharina (redrawn from Plantwise Technical Factsheet 2013; Yonow et al., 2016). Colours indicate presence of stemborer species……………………………..43 Figure 2.2: Results of plasticity of thermal tolerance in three major stemborers, (B. fusca, S. calamistis and C. partellus) measured as (A) critical thermal maxima and (B) critical thermal minima following acclimation for 3 days at low optimum and high temperature (23; 28 and 33°C respectively). Vertical bars denote ±95% CL. Means with the same letter are not significantly different from each other. Post hoc tests were done separately for each species…………...…...64 Figure 3.1: Mean (±95% confidence interval) survival of Chilo partellus (A) eggs and (B) larvae at different low temperatures, and (C) eggs and (D) larvae at different high temperatures each applied over four different exposure times………………………………………………………93 Figure 3.2: Mean (±95% confidence interval) survival of Cotesia sesamiae adults at different low (A) and high (B) temperatures applied for two hours……………………………………….94

Figure 3.3: Effects of experimental heating and cooling rates on Chilo partellus adult (A) CTmax and (B) CTmin, larval (C) CTmax and (D) CTmin, and Cotesia sesamiae adult (E) CTmax and (F)

CTmin. (mean± 95% confidence limits). n = 20 per group. Means with the same letter are not significantly different from each other……………………………………………………….….96 Figure 3.4: The effect of developmental stage on supercooling points in Chilo partellus larvae, pupae and adults. Error bars represent 95% CLs (N = 16 per group). Means with the same letter are not significantly different from each other……………………………………………..……97

Figure 3.5: The effect of developmental stage (larvae vs adults) on Chilo partellus (A) heat knock-down time and (B) chill-coma recovery time. Error bars represent 95% CLs (N = 30). Means with the same letter are not significantly different from each other………………..……97 Figure 3.6: Microclimatic temperature data in austral summer (December 2015-April 2016) and winter (May-July 2016) were recorded using Thermocron iButtons (Dallas Semiconductors, Model DS1920) (0.5°C accuracy; 1 h sampling frequency) from a maize field, Veggies For All Farm, Glen Valley, Gaborone, Botswana (S24.61224; E25.97326; 977 m.a.s.l), where both Chilo partellus and C. sesamiae are found……………………………………………………………..99 Figure 4.1: Schematic diagram for acclimation experimental protocols. For full description see sections 4.2.2 and 4.2.3…………………………………………………………………………120 Figure 4.2: Summary low temperature plasticity results showing (A) effects of 2h pretreatments/hardening on survival at discriminating low temperatures of Chilo partellus larvae (pretreatment [2°C 2h] on low temperature survival [-8°C 2 h]), (B) effects of (pretreatment [3°C 2 h] on low temperature survival [-7°C 2 h]) on Cotesia flavipes adults, (C) acclimation effects to ° ° low optimum and high temperature (23; 28 and 33 C respectively) on CTmin, (D) CCRT (at 0 C chill coma temperature) and (E) SCP. Error bars represent 95% CLs (N = 50 per group). Means with the same letter are not significantly different from each other. Post hoc tests were done separately for each species…………………………………………………………….………..127 Figure 4.3: Summary low temperature plasticity results showing (A) effects of 2h pretreatments/hardening on survival at discriminating low temperatures of Chilo partellus pupae (pretreatment [2°C 2h] on low temperature survival [-8°C 2 h]), (B) effects of (pretreatment [1°C 2 h] on low temperature survival [-9°C 2 h]) on Chilo partellus adults, (C) acclimation effects to ° ° low optimum and high temperature (23; 28 and 33 C respectively) on CTmin, (D) CCRT (at 0 C chill coma temperature) and (E) SCP. Error bars represent 95% CLs (N = 50 per group). Means with the same letter are not significantly different from each other. Post hoc tests were done separately for each developmental stage………………………………………………….……130 Figure 4.4: Summary low temperature plasticity results showing (A) effects of 2h pretreatments/hardening on survival at discriminating high temperatures of Chilo partellus larvae (pretreatment [41°C 2h] on high temperature survival [46°C 2 h]), (B) effects of (pretreatment [35°C 2 h] on high temperature survival [40°C 2 h]) on Cotesia flavipes adults, (C) acclimation ° effects to low optimum and high temperature (23; 28 and 33 C respectively) on CTmax and (D)

xvi

HKDT (at 45 and 49°C knockdown temperature for C. flavipes adults and C. partellus larvae respectively). Error bars represent 95% CLs (N = 50 per group). Means with the same letter are not significantly different from each other. Post hoc tests were done separately for each species…………………………………………………………………………………………..135 Figure 4.5: Summary low temperature plasticity results showing (A) effects of 2h pretreatments/hardening on survival at discriminating high temperatures of Chilo partellus pupae (pretreatment [42°C 2h] on high temperature survival [47°C 2 h]), (B) effects of (pretreatment [39°C 2 h] on high temperature survival [44°C 2 h]) on Cotesia flavipes adults, (C) acclimation ° effects to low optimum and high temperature (23; 28 and 33 C respectively) on CTmax and (D) HKDT (at 49°C knockdown temperature). Error bars represent 95% CLs (N = 50 per group). Means with the same letter are not significantly different from each other. Post hoc tests were done separately for each developmental stage…………………………………………………137 Figure 4.6: Microclimatic temperature data in austral summer and winter (January 2016 – February 2017) from a maize field, BCA Farm, Gaborone, Botswana (S24.56449; E25.94883; 994 m.a.s.l), which hosts both Chilo partellus and parasitoid C. flavipes. Temperature data was recorded using Thermocron iButtons (Dallas Semiconductors, Model DS1920) (0.5°C accuracy; 1 h sampling frequency). The number of events likely eliciting RCH and RHH within the same timeframe was calculated based on temperature-time interactions. Red and blue dashed lines represents RCH and RHH temperatures for Cotesia flavipes and Chilo partellus respectively……………………………………………………………………………………..140 Figure 5.1: Mean (±95% confidence interval) survival of (A) C. sesamiae and (B) C. flavipes at different low temperatures, and (C) C. sesamiae and (D) C. flavipes at different high temperatures applied over four different durations……………………………………..………167

Figure 5.2: Effects of species on (A) CTmin and (B) CTmax. Error bars represent 95% CLs (N = 20). Means with the same letter are not significantly different from each other………….……169 Figure 5.3: The effect of species on supercooling points. Error bars represent 95% CLs (N = 16 per group). Means with the same letter are not significantly different from each other……..…170 Figure 5.4: Effects of species on chill-coma recovery time. Error bars represent 95% CLs (N = 30). Means with the same letter are not significantly different from each other………………171 Figure 5.5: Effects of species on heat knock-down time. Error bar represent 95% CLs (N = 30). Means with the same letter are not significantly different from each other…………………....172

xvii

Figure 6.1: Effects of species on (A) CTmin and (B) CTmax. Error bars represent 95% CLs. Means with the same letter are not significantly different from each other……………………………194 Figure 6.2: Thermal acclimation effects for three major stemborers, (B. fusca, S. calamistis and

C. partellus) on (A) CTmin and (B) CTmax following acclimation for 3 days at low optimum and high temperature. Vertical bars denote ±95% CL. Means with the same letter are not significantly different from each other. Post hoc tests were done separately for each species…………………………………………………………………………………………..196

Figure 6.3: Effects of starvation acclimation on (A) CTmin and (B) CTmax for Busseola fusca, Sesamia calamistis and Chilo partellus larvae. Error bars represent 95% CLs. Means with the same letter are not significantly different from each other. Post hoc tests were done separately for both treatments and across species……………………………………………….…………….198

Figure 6.4: Effects of RCH on (A) CTmin and (B) CTmax and RHH on (C) CTmin and (D) CTmax for Busseola fusca, Sesamia calamistis and Chilo partellus larvae. Error bars represent 95% CLs. Means with the same letter are not significantly different from each other. Post hoc tests were done separately for both treatments and across species……………………………………….202

Figure 6.5: Effects of desiccation acclimation on (A) CTmin and (B) CTmax for Busseola fusca, Sesamia calamistis and Chilo partellus. Error bars represent 95% CLs. Means with the same letter are not significantly different from each other. Post hoc tests were done separately for both treatments and across species…………………………………………………………………..204

APPENDIX Figure 1.1: Botswana maps showing (A) areas surveyed for stemborers in cultivated and natural habitats during 2014/15 and 2015/16 austral summer (B) average annual temperatures and (C) average annual rainfall for the sampled districts………………………………………………234 Figure 1.2: Quantitative bipartite interaction networks for (A) stemborer - parasitoid and (B) host plant - parasitoid in cultivated and natural habitats in Botswana……………………….…248 Figure 1.3: Relative abundance of cereal stemborer parasitoids in (A) cultivated and (B) natural habitats in Botswana…………………………………………………………………………....252 Figure 2.1: Current distribution of Tuta absoluta in Africa, redrawn from EPPO, 2016)…….267 Figure 2.2: Yellow delta pheromone trap placed at (A) Genesis farm (S21.14776; E27.64744; 987 m.a.s.l) (tunnel agroecosystem) and (B) open forest (S21.16684; E27.57120; 1013 m.a.s.l)

xviii wild habitat) with predominant Colophospermum mopane, North East District of Botswana.

Photos by H. Machekano and R. Mutamiswa………………………………………………………………270 Figure 2.3: Tuta absoluta PH-937-OPTI female sex pheromone (Russell IPM®) used in the delta traps……………………………………………………………………………………………..271

Figure 2.4: Tuta absoluta: male genitalia. Photograph by R. Ramukhesa…………………………….273 Figure 2.5: Typical Tuta absoluta damage symptoms on (A) tomato plants (B) wild host Solanum lichtensteinii (C) distorted and damaged fruit following larval fruit infestation, and (D) advanced damage signs with exit holes, secondary infection, hanging skins with consumed internal contents and accelerated senescence at Genesis farm (S21.14776; E27.64744; 987 m.a.s.l). Photos by H. Machekano and R. Mutamiswa…………………………………………….……….275 Figure 2.6: Sex pheromone trap catches for T. absoluta male moths (A) set up inside the yellow delta trap and (B) sticky pad retrieved from the yellow delta trap following one week trapping period at Genesis Farm (S21.14776; E27.64744; 987 m.a.s.l). Photos by H. Machekano and R.

Mutamiswa………………………………………………………………………………….……..276 Figure 2.7: A map of Botswana showing detection sites for Tuta absoluta…………….…….277 Figure 2.8: Mean number of Tuta absoluta male moths/trap over a period of seven days from different sites distant from core detection point in north east Botswana. Traps were baited using Tuta absoluta PH-937-OPTI (Russel IPM)……………………………………………………278 Figure 2.9: A comparison of the sequences of Botswana specimens with existing sequences on GenBank; KJ657881, KJ657680, KC852871, KT452897, KP793742, KC852872, KP324753 and an outgroup KX862248. The evolutionary history was inferred using the Neighbor-Joining method (Saitou & Nei, 1987). The optimal tree with the sum of branch length = 0.13050185 is shown. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (100 replicates) are shown next to the branches (Felsenstein, 1985). The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the Kimura 2- parameter method (Kimura, 1980) and are in the units of the number of base substitutions per site. The analysis involved 9 nucleotide sequences. Codon positions included were 1st+2nd+3rd+Noncoding. All positions containing gaps and missing data were eliminated. There were a total of 510 positions in the final dataset. Evolutionary analyses were conducted in MEGA6 (Tamura et al., 2013)………………………………………………………………….279

xix

ABSTRACT

Lepidopteran stemborers are key destructive insect pests of cereal crops in sub-Saharan Africa with the indigenous noctuids Busseola fusca (Fuller) and Sesamia calamistis Hampson and the exotic crambid Chilo partellus Swinhoe accounting for ~5-75% of potential cereal crop yield losses. The main larval endoparasitoids comprising, indigenous Cotesia sesamiae and exotic Cotesia flavipes Cameron contribute to a significant extent in managing these economic herbivorous insect pests. However, efficacy of biological control under global change is currently under threat, and factors limiting trophic level abundance, distribution and population phenologies of these insect species under climate change remain unclear. Against this background, this study provides an understanding of the thermal ecophysiology of laboratory reared B. fusca, S. calamistis and C. partellus and their larval parasitoids, C. sesamiae and C. flavipes under variable climatic conditions and how their population phenologies are likely to be shaped by variable thermal regimes under global change scenarios. This study reports implications of these results thereof for biological control efficacy under climate change. Firstly, a comparative assessment of C. partellus and C. sesamiae thermal tolerance using dynamic and static protocols showed developmental stage differences in C. partellus thermal tolerance (with respect to lethal temperatures and critical thermal limits) and a compromised temperature tolerance of C. sesamiae relative to its host. This indicates potential asynchrony between host- parasitoid population phenology and consequently biocontrol efficacy under global change. Secondly, I investigated the short to medium term phenotypic plasticity of thermal tolerance of C. partellus developmental stages (larvae, pupae and adults) and its larval parasitoid C. flavipes (adults). Rapid cold hardening (RCH) and rapid heat hardening (RHH) effects in C. partellus larvae, pupae and adults and C. flavipes adults were highly significant (P ˂ 0.001). In addition, high temperature acclimation (33°C) improved critical thermal limits [CTLs (CTmin and CTmax)] and heat knock-down time (HKDT) for C. partellus larvae and C. flavipes adults (P ˂ 0.0001) respectively while low temperature (23°C) acclimation enhanced supercooling point (SCP) for C. flavipes and chill-coma recovery time (CCRT) for both C. partellus larvae and C. flavipes adults (P ˂ 0.0001) suggesting that the host is more plastic than the parasitoid. Thirdly, the study explored basal thermal tolerance of larval endoparasitoids, C. sesamiae and C. flavipes and how this may influence their geographical distribution and biological conservation. Cotesia flavipes showed a higher basal thermal tolerance (with respect to traits of temperature tolerance, CTLs,

HKDT and CCRT) than C. sesamiae indicating future non overlapping thermal environments of these species under climate change. Furthermore, since C. flavipes is exotic, its physiological advantage over indigenous C. sesamiae may likely mean exotic species may likely drive local species out of their native ranges, with significant consequences on biodiversity conservation and synergistic biocontrol efficacy. Lastly, the study investigated the effects of heterogeneous stressful environments on thermal tolerance vis a vis hardening (RCH and RHH), thermal (low, optimum and high), starvation and desiccation acclimation on thermal tolerance traits, measured as critical thermal limits on laboratory reared stemborers B. fusca, S. calamistis and C. partellus using dynamic (ramping) protocols. Thermal acclimation to low, optimum and high temperatures was highly significant for CTmin (P ˂ 0.001) across all species. Similarly, across all species,

RCH, RHH, starvation and desiccation acclimation were significant (P ˂ 0.001) for CTmin.

However, RHH, starvation and desiccation were not significant (P > 0.05) for CTmax across all species. These findings show differential thermal tolerances following exposure to heterogeneous environmental stress habitats. However, Chilo partellus was more plastic and more heat tolerant following starvation acclimation while B. fusca was more cold tolerant following RCH, starvation and desiccation acclimation. In summary, this work revealed asymmetrical variation in thermal ecophysiology across test species, appearing to be enhanced in C. partellus relative to S. calamistis and B. fusca and in C. flavipes relative to C. sesamiae. Therefore with projected climate change, C. partellus is highly likely to extend its geographical distribution relative to S. calamistis and B. fusca ultimately displacing these indigenous stemborer species. In addition, the larval parasitoids may not cope with the geographical expansion of their hosts due to variability in thermal tolerance, albeit indigenous C. sesamiae looks more vulnerable than exotic C. flavipes. This may have broad implications on biological control efficacy leading to more pest outbreaks. These results are useful in designing pest management options for invasive species and conducting pest risks assessments.

xxi

1 2

3 CHAPTER 1

10 General Introduction

11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

26 1.0 Climate change and insect phenology 27 Global climate change is a transformation in the long term weather patterns and is evident 28 from increase in global average temperature and its variability, changes in the rainfall pattern and 29 extreme climatic events (Selvaraj et al., 2013). The global mean surface temperature is predicted 30 to increase between 1.4 and 5.8°C by 2100 (Karuppaiah & Sujavanad, 2012). Climate changes 31 have major impacts on the agroecosystem, and insects being ectotherms may be the most 32 affected (Jaworski &Hilszczański, 2013). Such changes in climate could profoundly affect the 33 population dynamics of insect pests, change their ‘status’ and consequently upset timing of 34 control measures. Researchers have shown that global temperature increase may potentially 35 accelerate development rate, survival, reproduction and population size of tropical and 36 subtropical insects (Selvaraj et al., 2013) as well as altering their biogeography (Nyamukondiwa 37 et al., 2011). Consequently, insects may be capable of completing higher numbers of generations 38 per year (Deka et al., 2009) ultimately resulting in more crop damage (Yamamura & Kiritani, 39 1998), asynchrony in the timing of control strategies and as a result, affecting efficacy of control 40 measures (Mutamiswa et al., 2017a). Future direction and rate of climate change as well as its 41 direct and indirect effects are complex and may be difficult to predict (Gutierrez et al., 2008). 42 Therefore, it is essential to understand the mechanisms behind climate influence on the 43 interaction of ecosystem components and agricultural environment especially on insects of 44 economic importance to agroecosystems (Jaworski & Hilszczański, 2013). Some insect pests 45 which are already present but occurring at low densities in small areas, may be able to exploit the 46 changing climatic conditions by spreading more widely and reaching damaging population 47 densities (Karuppaiah & Sujavanad, 2012). As a consequence, there is a likelihood of farmers 48 experiencing extensive impacts on insect management strategies under climate change. 49 Abiotic factors such as temperature and humidity have been reported to directly influence 50 insect population dynamics through modulation of development rates, survival, fecundity, 51 parasitism and dispersal (Selvaraj et al., 2013). Exposure to temperature extremes induces lethal 52 and sub lethal damage to predators and parasitoids (Hance et al., 2007), whose ecological 53 function is to keep the insects below economic injury level. Consequently, temperature extremes 54 associated with global change may also likely offset the efficacy of biological control 55 programmes (Selvaraj et al., 2013). Natural enemies and host insect populations may respond 56 differently to changes in temperature. Apart from surviving thermal extremes, natural enemies

57 need to counter climate change by mating and then locating hosts effectively across a wider 58 range of thermal and humidity conditions (Thomson et al., 2010). Parasitism may be reduced if 59 host populations emerge and pass through vulnerable life stages before parasitoids emergence. 60 Hosts may pass through vulnerable life stages more quickly at higher temperatures, reducing the 61 window of opportunity for parasitism which may give great set back to the survival and 62 multiplication of parasitoids (Fand et al., 2012). While natural enemies have some potential to 63 deal with changes in climatic variability through acclimation and diapause, and by having some 64 innate resistance of thermal extremes, climatic variability may influence their effectiveness in 65 controlling pests (Thomson et al., 2010). Field data suggest that parasitoids are more sensitive to 66 climatic variability than their hosts, possibly related to the greater intrinsic rate of population 67 increase shown by hosts following mortality of both groups (Thomson et al., 2010). 68

69 1.1 Insect Thermal Biology and Physiology 70 Temperature can impact insect physiology and development directly or indirectly through 71 the physiology or existence of hosts. Depending on the insect development “strategy”, 72 temperature has been reported to exert various effects (Bale et al., 2002; Petzoldt & Seaman, 73 2006). Some insect species such as cicadas and arctic moths take several years to complete one 74 life cycle hence they tend to moderate variabilities in temperature over their entire life history 75 (Bale et al., 2002). In addition, some insect pests develop more rapidly during time periods 76 when temperatures are favourable (Petzoldt & Seaman, 2006). Day-Degree (DD) or phenology 77 based models are therefore used to predict the emergence of these insects and their potential to 78 damage crops. 79 Insect responses to temperature may be summarised using the performance curve (Figure 80 1.1) (Wharton, 2002; Piyaphongkul, 2013) and thermobiological scale (Figure 1.2) (Vannier, 81 1994). Performance curves are essential when elucidating capacity responses of insects (Chown 82 & Terblanche, 2007). The lower and upper lethal temperature is the temperature required to kill a 83 certain proportion of the population in a specified time period (Bale & Walters, 2001). At 84 temperatures below and above the optimum, metabolism slows and eventually ceases due to the 85 damaging and lethal effects of low and high temperatures. The upper lethal temperature (ULT) is

86 normally higher than the critical thermal maxima (CTmax) and heat coma temperature (HCT)

87 whilst lower lethal temperature (LLT) will be lower than the critical thermal minima (CTmin) and

88 chill coma temperature (CCT) (Figure 1.1). However, the difference between the heat coma 89 temperature and ULT is sometimes small e.g. around 1˚C or less, and insect species experiencing 90 heat coma may not be able to recover, so the heat coma temperature may effectively be the ULT 91 (Hazell et al., 2010, Piyaphongkul, 2013). There are a range of methods used to measure the 92 ULT and LLT including ‘direct plunge’ where insects are transferred directly to a potentially 93 high and low temperature, and ‘dynamic methods’ where the temperature is increased or 94 decreased gradually from a set point benign temperature and mortality is assessed at a sequence 95 of increasingly higher and lower temperatures (Nyamukondiwa & Terblanche 2009; 96 Piyaphongkul, 2013). 97 98 99 100 101 102 103 104 105 106 107 108 109 Figure 1.1: Responses to temperature in a hypothetical organism (Adapted from Wharton, 2002;

110 Piyaphongkul, 2013). CTmin= critical thermal minima; CTmax = critical thermal maxima; CCT = 111 chill coma temperature; HCT = heat coma temperature; LLT = lower lethal temperature; ULT = 112 upper lethal temperature 113 114 The thermobiological scale is critical when considering resistance responses (Chown & 115 Terblanche, 2007). In the middle of the thermobiological scale (Figure 1.2) is the optimum 116 temperature for an insect’s survival, growth and evolutionary fitness. At both ends of the scale, 117 continuation of the change in temperature results first in knockdown of the insect (Chown & 118 Nicholson, 2004), then in prolonged coma, and finally irreversible trauma and death (Figure 1.2).

119 At low temperatures, insects show a wider variety of responses to sub-lethal and potentially 120 lethal temperatures than they do at high temperatures, including responses to non-freezing 121 temperatures and those made in preparation for the decline of temperatures below the freezing 122 point of water (Chown & Nicholson, 2004).

123 124 125 Figure 1.2: The thermobiological scale, redrawn from Vannier (1994). 126 127 128

129

130 The temperature at which heat induces injury and/or death in insects varies both spatially 131 and temporally (Chown & Terblanche, 2007). This is because insect thermal tolerance is 132 dynamic and organisms often shift their fitness traits in relation to thermal environment related to 133 timing (seasons) and space. Furthermore, differences in methods of measurement usually assess 134 different traits resulting in dissimilar outcomes hence it is difficult to reach general conclusions 135 regarding upper thermotolerance limits (Chown & Nicolson, 2004). However, in insects they 136 generally do not exceed 53˚C, and are usually not much lower than 30˚C, although these values 137 depend on the trait being measured (Chown & Terblanche, 2007) and methodological context 138 (Chown et al., 2009). It has been suggested that heat shock proteins (Hsps) might be responsible 139 for both survival of potentially lethal temperatures and for improved knockdown resistance in 140 insects (Nielsen et al., 2005; Chown & Terblanche, 2007). Indeed, expression of Hsps has been 141 observed in response to both heat and cold shock (Chown & Nicolson, 2004), symbolizing some 142 degree of overlap in mechanisms for tolerance to both thermal extremes. Nevertheless, tolerance 143 to low temperatures has been more associated with carbohydrate metabolism (Sinclair et al., 144 2013; Sinclair, 2015). 145 Knowledge of thermal tolerance is crucial to understanding of ecology, evolution, and 146 physiology of organisms as these traits play an important role in determining species range 147 distributions (Hazell et al., 2008) and how they may be altered by climate change (Walther et al., 148 2002). Consequently, thermal tolerance traits are also important determinants of the likelihood 149 that a species will become established upon introduction into a novel environment and likely 150 become invasive (Slabber et al., 2007; Ward & Masters, 2007; Nyamukondiwa et al., 2010), and 151 the magnitude of the potential threat due to climate change faced by species with low dispersal 152 ability (Thomas et al., 2004). Thermal tolerance traits include ULT and LLT (Bale et al., 1994; 153 Sinclair et al., 2006) and a range of non-lethal measures including the high and low CTLs, knock 154 down and chill coma temperatures (Gaston & Chown, 1999; Hazell et al., 2008 reviewed in 155 Chown & Nicholson, 2004) and measures of the time taken to knockdown an organism at high 156 temperature and to recover from cold-induced coma (heat knockdown time [HKDT] and chill 157 coma recovery time [CCRT] respectively) (Ayrinhac et al., 2004; Castaneda et al., 2005). Of the 158 non-lethal measures of thermal tolerance described above, the indexes most commonly employed

159 by researchers are the CTLs (CTmin and CTmax) or coma temperatures (Hazell et al., 2008).

160 The ability of an organism to remain active under extreme conditions is a significant 161 component of fitness (Loeschcke & Hoffmann, 2007). Therefore determining the limits to 162 activity is an important first step in understanding the ways in which environmental variation 163 affects fitness and the dynamics of a given population. Thermal limits have received 164 considerable attention because their investigation provides insight into the manner in which 165 climate shapes variation in the ecology, distribution and evolution of species (Ghalambor et al., 166 2006; Chown & Terblanche, 2007). Furthermore, upper temperature limits are positively related 167 to optimal performance temperatures and these limits are relatively simple to measure (Chown & 168 Nicolson, 2004). In consequence, factors that affect the assessment of thermal tolerance are 169 significant (Chown & Nicolson, 2004). Typically, limits are assessed using either dynamic or 170 static protocols (Hoffmann et al., 2003). Briefly, the dynamic method involves changing 171 temperature (either up or down) at a constant rate, starting from a setup standard benign 172 temperature for the experimental organism and assessing individual knockdown. One dynamic 173 technique that is widely used to resolve both high and low temperature thresholds is the 174 determination of CTLs (Lutterschmidt & Hutchison, 1997; Beitinger et al., 2000; Chown & 175 Nicolson, 2004). Critical thermal limits are considered ecologically relevant because they 176 provide an indication of the activity range for a population under acute exposure conditions 177 (Somero 2005; Terblanche et al., 2007), although animals may mediate effects of environmental 178 temperature variation through behavioral adjustments (Huey et al., 2003). Critical thermal limits

179 are defined as the lowest (CTmin) and maximum (CTmax) temperatures that allow organismal 180 activity (Chown & Nicholson, 2004) and are manifested in insects as the temperature at which 181 each individual organism loses coordinated muscle function, consequently losing the ability to

182 respond to mild stimuli (e.g. prodding). In the case of CTmax, this loss of muscle function

183 coincides with death such that recovery will not be possible, while in the case of CTmin, recovery 184 occurs, and hence, it’s not immediately lethal. 185 HKDT is the time required for an individual to lose motor function during exposure to a 186 high temperature (Huey et al., 1992; Angilletta, 2009) while CCRT is the time required for an 187 individual to recover from loss of motor function following exposure to a low temperature 188 (Castaneda et al., 2005; Angilletta, 2009; MacMillan et al., 2016). Many different methods have 189 been used to test for heat resistance, the time it takes an insect to be knocked down by higher 190 temperatures and knockdown temperature, the upper temperature at which insects lose mobility

191 or are unable to remain upright (Gilchrist & Huey, 1999; Milton & Partridge, 2008). Heat 192 knockdown is induced when insects gradually become paralysed and unresponsive to stimulus 193 during heat stress exposure. Changes in low temperature tolerance have often been measured by 194 survival rates after exposure to low temperatures or CCRT (Gibert et al., 2001a; Milton & 195 Partridge, 2008). A chill coma is induced when insects become immobilised upon exposure to 196 low, non-freezing temperatures (approximately 0˚C). Similar to heat knockdown, the insects lose 197 motor activity and remain paralysed until they are returned to normal temperatures where they 198 progressively regain mobility and activity (David et al., 1998; Milton & Partridge, 2008). Such 199 indices represent vulnerable temperatures ecologically as they represent thermal windows where 200 insects are vulnerable to predation and compromise different life history fitness traits. 201

202 1.2 Insect freeze strategies and supercooling points (SCPs) 203 SCP is the temperature at which ice begins to form within the body fluids after exposure 204 to low temperature (also called the temperature of crystallization) (Hance et al., 2007). At this 205 point water and aqueous solutions remain unfrozen at temperatures below their melting point. In 206 laboratory experiments, SCP for each individual organism is determined as the lowest 207 temperature recorded prior to a spike in temperature associated with the latent heat of 208 crystallization (Nyamukondiwa et al., 2013). Consequently, following determination of the SCP 209 of an insect, it is possible to establish whether the insect is freeze tolerant, freeze intolerant or 210 chill susceptible. Insect freeze tolerance refers to the ability of some insect species surviving ice 211 formation within their tissues (Lee & Costanzo, 1998). Freeze-intolerant species which cannot 212 survive ice formation within their tissues are killed once they reach their SCP hence can also 213 sustain lethal damage at temperatures above that point (Hance et al., 2007). Freeze intolerant 214 insects avoid ice formation in body fluids through developing a high supercooling capacity, 215 whereas some dehydrate, thereby reducing their melting point to the ambient temperature 216 (Denlinger & Lee, 2010). In that respect, insects that have evolved freeze-tolerance strategies 217 avoid tissue damage through controlling where, when and to what extent ice forms in their body 218 fluids (Lee & Costanzo, 1998). In contrast to freeze intolerant insects that are able to exist in 219 cold conditions by supercooling, freeze tolerant insects tolerate the formation of ice in their body 220 tissues and fluids (Sinclair et al., 2015). Physiologically this is accomplished through inoculate 221 freezing, the production of ice nucleating proteins, crystalloid proteins, crystalloid compounds

222 and microbes (Lee & Constanzo, 1998). An exposure to low temperature can result in physical 223 damage caused by either ice crystals or metabolic injuries such as osmotic stress or anoxia 224 (Turnock & Fields, 2005; Hance et al., 2007), both of which can be lethal to insect pests and 225 associated parasitoids. Chill susceptible insects are killed by cold in the absence of internal ice 226 formation (Sinclair et al., 2015). Generally, non-freezing injury is cumulative and can be lethal if 227 the intensity or duration of that injury surpasses a certain threshold (Turnock & Fields, 2005; 228 Hance et al., 2007). Such damage can be permanent, leading to death, or it can be repaired if the 229 insect is temporarily exposed to higher temperature. As an insect is cooled to sub-zero 230 temperatures ice does not form at 0°C until the temperature falls below the melting point of the 231 insect’s body fluids (Figure 1.3) Insects only begin to supercool when they are cooled to 232 temperatures below their melting point (Denlinger & Lee, 2010). Determination of SCPs is 233 critical under global change, since it is expected that climate variability will increase (Deutsch et 234 al., 2008), with consequent increase in likely freezing temperatures. Thus, SCPs may provide a 235 clue on how insects may cope upon exposure to sub-zero temperatures. 236

237 238 Figure 1.3: Responses of insects cooled to low temperature, redrawn from (Denlinger & Lee, 239 2010) 240 241

242 1.3 Phenotypic plasticity in insects as a response to changes in temperature 243 Phenotypic plasticity is the range of phenotypic traits (physiological, morphological, 244 behavioural, and life-history traits) expressed by a single genotype in response to variable 245 environmental conditions (Schlichting & Smith, 2002; Whitman & Agrawal, 2009; Nicotra et al., 246 2010). Insects are able to survive extreme temperature variation at daily to weekly time scales 247 using a host of physiological and biochemical mechanisms. One such mechanism is rapid cold 248 hardening (RCH), defined as a rapid improvement in survival at a low lethal temperature after 249 brief pre-treatment to a prior sub-lethal temperature shock (Stotter & Terblanche, 2009). In 250 certain life-stages of a particular species as much as 80-100% improvement in survival has been 251 reported, for example Bactrocera oleae (Koveos, 2001), Drosophila melanogaster (Jensen et al., 252 2007; Rajamohan & Sinclair, 2008), in Tephritids, Ceratitis capitata and C. rosa 253 (Nyamukondiwa et al., 2010), Zaprionus vittiger (Nyamukondiwa & Terblanche, 2010a) and an 254 array of other drosophilids (Nyamukondiwa et al., 2011). In addition, both C. rosa and C. 255 capitata show the ability to alter thermal tolerance to new thermal conditions (e.g. simulated 256 summer or winter conditions in the laboratory) within a single generation (Nyamukondiwa et al., 257 2013). Low temperature responses in Lepidoptera, Thaumatotibia leucotreta (Stotter & 258 Terblanche, 2009) proved its ability to adjust thermal tolerance at longer time scales. 259 Furthermore, of the insect species that show acute physiological protection mechanisms such as 260 RCH, considerable variation in the magnitude of the protection afforded exists both within 261 (Jensen et al., 2007) and among species (Chown & Nicolson, 2004; Nyamukondiwa & 262 Terblanche, 2010a) and the time course of responses (Weldon et al., 2011). Similarly, at warm 263 temperatures, brief exposure to sub-lethal high temperatures can result in greater survival at a 264 lethal high temperature (Stotter & Terblanche, 2009) a phenomenon called rapid heat hardening 265 (RHH). Indeed, this mechanism has been shown to improve survival in several insects including 266 Trichogramma carverae (Scott et al., 1997), Ceratitis capitata (Kalosaka et al., 2009; 267 Nyamukondiwa et al., 2010), Cydia pomonella (Chidawanyika & Terblanche, 2011a), and 268 several Drosophila species (Nyamukondiwa et al., 2011). Another way that insects can use 269 phenotypic plasticity to adjust their physiology to their current environment is through 270 acclimatization (Angilletta, 2009; Austin, 2013). Acclimatization is the physiological change that 271 is associated with a chronic change in the natural environment of an organism, where many 272 environmental factors are changing at once (Hochachka & Somero, 2002; Tattersall et al., 2012).

273 The effect of an individual component of the environment (such as temperature) on an organism 274 is called acclimation (Austin, 2013). Acclimation can be observed in a controlled laboratory 275 environment where various environmental factors are held constant to isolate the physiological 276 effects of a single factor of interest on performance and survival of an organism (Hochachka & 277 Somero, 2002; Tattersall et al., 2012). There has been increasing interest by physiological and 278 behavioural ecologists in acclimation effects on ectotherms to varying temperature regimes 279 (Jumbam et al., 2008; Gray, 2013; Hemmati et al., 2014). It is thought that these plasticity 280 mechanisms are employed by insects to quickly alter their physiological tolerances at short 281 timescales, for example over the course of a day to optimise key features like mating and feeding 282 (Shreve et al., 2004) consequently critical to the survival of sudden, unpredictable temperature 283 extremes typical in the wild (Chown & Nicolson, 2004). 284 Many researchers assume phenotypic plasticity has evolved as a mechanism to improve 285 fitness in an environment in which the organism has been acclimated. This is known as the 286 “beneficial acclimation hypothesis” (BAH) (Kristensen et al., 2008; Angilletta, 2009; Cooper et 287 al., 2010). However, this BAH has been debated in literature, as organisms often adjust 288 incorrectly to their environment and as a result, suffer fitness consequences (Wilson & Franklin, 289 2002; Angilletta, 2009; Cooper et al., 2010). For example, D. melanogaster reared at a high 290 temperature did not have better fitness traits compared to flies reared at intermediate and low 291 temperatures (Gibert et al., 2001a; Austin, 2013). Therefore, phenotypic plasticity may in some 292 instances, not necessarily always increase the fitness of organisms in their natural environment 293 (Gibert et al., 2001b; Austin, 2013). Indeed, the optimal developmental temperature hypothesis 294 suggests otherwise, that organisms developed optimally have a fitness advantage at both thermal 295 extremes (Bahrndorff et al., 2016; Slotsbo et al., 2016), and the phenomenon has been 296 demonstrated in Drosophila. 297

298 1.4 Economic Importance of Cereal Stemborers 299 The cereal stemborer species present in sub-Saharan Africa belong to the Noctuidae, 300 Crambidae and Pyralidae families. From these families, the major cereal stemborer species 301 include the indigenous pyralid Eldana saccharina Walker, the crambids Chilo orichalcociliellus 302 (Strand) and exotic C. partellus Swinhoe and the noctuids Busseola fusca (Fuller) and Sesamia 303 calamistis Hampson (Kfir et al., 2002; Obonyo et al., 2010; Addo-Bediako and Thanguane,

304 2012). In most cases, these borers occur as a complex of species with overlapping spatial 305 distributions (Chabi-Olaye et al., 2001). In sub-Saharan Africa, cereal crops are mainly grown at 306 subsistence level by resource-poor farmers and these stemborers are the major biotic constraints 307 to cereal production, causing losses ranging between 5% and 75% of potential crop yield (De 308 Groote, 2002; Kipkoech et al., 2006). Severity and nature of stemborer damage is dependent on 309 the season, variety (Overholt et al., 1994; Kfir et al., 2002; Tefera et al., 2010), borer species, the 310 plant growth stage, the number of larvae feeding on the plant, and the plant's reaction to borer 311 feeding (DeGroote, 2002). Almost all cereal plant parts, leaves, stems, tassels and ears are 312 attacked. Infestations by these stemborers increase the incidence and severity of stalk rots and 313 may increase the contamination of the grains with toxin-producing fungi such as Aspergillus 314 flavus (Kfir et al., 2002). 315

316 1.4.1 Biology and Ecology of Busseola fusca 317 The African maize stemborer, Busseola fusca (Fuller) (Lepidoptera: Noctuidae) is 318 indigenous to Africa and is a common pest in many sub-Saharan African countries (Tefera et al., 319 2010). Its distribution and pest status varies with region. In eastern and southern Africa, it is a 320 pest at higher altitudes (above 600 m), but in central Africa it occurs from sea level to over 2000 321 m, while in west Africa, it is primarily in the dry savannah zone (Kfir et al., 2002; Tefera, 2004). 322 The life cycle of B. fusca takes 7 to 8 weeks (Polaszek, 1998) and an adult moth lays about 400- 323 500 eggs (Sharma & Nwanze, 1997) between the leaf sheaths and stalk and on the outer ear 324 husks (Tefera et al., 2010). The eggs hatch after 10 days (at benign temperatures) and larvae 325 move up the plant and start feeding on the tender leaves in the funnel creating ‘window’ like 326 structures (Tefera et al., 2010). The latter stages of larvae bore into the stem and feed on central 327 tissue. Larval stage takes ≥ 35 days undergoing 6 instars before pupating. The pupal stage takes 9 328 to 14 days (at benign temperatures) before adult eclosion (Tefera et al., 2010). 329

330 1.4.2 Biology and Ecology of Sesamia calamistis 331 The pink stemborer, Sesamia calamistis (Hampson) (Lepidoptera: Noctuidae) is one of 332 the indigenous stemborer pests associated with maize and sorghum in Africa (Ong’amo et al., 333 2008). However, its economic importance varies across the continent. In West Africa, it is a

334 major cereal insect pest, but is a minor pest in eastern and southern Africa (Bosque-Pérez & 335 Schulthess, 1998; Ong’amo et al., 2008). Differences in its pest status may be attributed to 336 variations in host range and ecological preferences among populations (Seshu Reddy, 1998). 337 Sesamia calamistis is reported to have originally colonised non-cultivated hosts belonging to the 338 Poaceae and Cyperaceae (Ong’amo et al., 2008) and presumably switched to cultivated crops 339 after the domestication of sorghum and the introduction of maize in Africa (Polaszek & Khan, 340 1998; Ong’amo et al., 2008). Adult female can lay up to 400 eggs between the lower leaf sheaths 341 and stems (Sharma & Nwanze, 1997). Eggs hatch in 5 to 7 days and first instar larvae penetrate 342 the stems where they start feeding. Larval stage also undergoes 6 instars in 4 to 6 weeks before 343 pupation. Pupation takes between 9 and 13 days before adult eclosion (Sharma & Nwanze, 344 1997). 345

346 1.4.3 Biology and Ecology of Chilo partellus 347 The spotted stemborer, C. partellus, exotic to Africa, is one of the most important 348 stemborers in East and Southern Africa (Tefera et al., 2010). Since its invasion of the African 349 continent, it was limited to warm, low-altitude regions of eastern and southern Africa (Kfir, 350 1997) but recent studies have reported it extending its geographical distribution to cooler and 351 higher altitude areas (Khadioli et al., 2014; Mutamiswa et al., 2017a). The life cycle of spotted 352 stemborer takes about 3 to 4 weeks, sometimes longer at lower temperatures, and shorter in 353 warm climates. The emerging larvae move to the whorls where they start feeding on tender 354 leaves of the plants and later extending into the stem. Some may move to nearby host plants 355 (Berger, 1992). Severely attacked plants usually dry up partly or entirely, resulting in 'dead heart' 356 symptom’ (Tefera et al., 2010). Early attacked plants show stunted growth and poor ear 357 development. Stem tunneling by older larvae has been reported to interfere with nutrients 358 transfer to the ear (Polaszek, 1998; Kfir et al., 2002; Tefera et al., 2010). Larvae spin silken 359 threads which are launched by wind into the air to infest neighboring plants (Sithole, 1989). This 360 is an instinctive dispersal mechanism which serves in reducing competition between larvae that 361 hatch from the same egg batch thereby increasing the chances of their survival. Up to six 362 generations of C. partellus can occur per year (Chinwada, 2002). 363 Chilo partellus moths sexually mature in 3 days at benign temperatures (Lwande et al., 364 1993) after which they lay flat and oval, creamy white eggs approximately 0.8mm long. Females

365 are capable of laying between 600-800 eggs in their entire lifetime (Sharma & Davies, 1988). 366 These eggs hatch into larvae within 2-5 days (Sharma & Davies, 1988). Larvae are up to 25 mm 367 long when fully grown, with a prominent reddish-brown head (Tefera et al., 2010). The body is 368 creamy-white to yellowish-brown, with four purple-brown longitudinal stripes and usually with 369 very conspicuous dark-brown spots along the back, which give them a spotted appearance. 370 Young larvae initially feed in the leaf whorl whereas older larvae tunnel into stems. In warm 371 conditions larval development is usually completed in about 15 to 20 days after which they 372 pupate in damaged stems (Tefera et al., 2010). Pupae are up to 15mm long, slender, shiny and 373 light yellow-brown to dark red-brown in colour. Adults eclose within 7 to 10 days depending on 374 temperature (Ofomata et al., 2000; Kfir et al., 2002; Muhammad & Underwood, 2004; Tefera et 375 al., 2010). Adults are relatively small moths with wing lengths ranging from 7 to 17mm and a 376 wingspan of 20 to 25mm (Tefera et al., 2010). Adults emerge from pupae in the late afternoon or 377 early evening. They are active at night and rest on plants and plant debris during the day (Tefera 378 et al., 2010). 379

380 1.5 Control of Cereal Stemborers

381 1.5.1 Chemical Control 382 Under severe infestation, chemical control can provide an effective means of managing 383 stemborers. However, chemical application is only effective economically, if applied based on 384 economic threshold levels, thus pest scouting and monitoring may be successful prior to crop 385 damage (Bosque-Perez, 1995). All stemborer control strategies involving the use of chemicals 386 generally target the larval stage. Dusts (Malathion and Carbaryl) and sprays are applied down the 387 funnel of young plants to kill the emerging and feeding first instar larva and are effective against 388 these developmental stages (Van den Berg & Nur, 1998). Deltamethrin also gives an effective 389 control against stemborers in maize and sorghum when applied 10-14 days after crop emergence. 390 In commercial farming systems, foliar sprays of Monocrotophos and Pyrethroids are common for 391 the control of stemborers. Granular application of insecticides in the whorls of maize and 392 sorghum plants is the most effective and economic way of controlling stemborer species which 393 feed in the whorls (Van den Berg & Nur 1998). In the smallholder sector, whorl-applied granular 394 insecticides such as Trichloforn (Dipterex) are the most widely used because of their ease of

395 application and lack of the requirement for special application equipment (Chinwada, 2002). 396 They are easy and safe to handle, can be applied by hand, and can be applied accurately in a 397 controlled manner (Van den Berg & Nur 1998). Soil applied granular systemic pesticides such as 398 Carbofuran applied at planting have also been shown to be effective in controlling stemborers 399 (Van den Berg & Nur, 1998). 400 Although chemicals are effective in controlling stemborers, they have a danger of 401 resistance development (Hill, 1987). In addition, they may not be available or affordable to 402 smallholder resource poor subsistence farmers and may come at a cost to environmental 403 biodiversity. The boring characteristic of the larvae protects them against the sprays and hence 404 regular sprays may be required which subsistence farmers in developing countries cannot afford

405 (Sithole, 1989). Because of the low profit margin of maize and sorghum, smallholder farmers 406 cannot afford the cost of chemical control against stemborers (Kfir, 1998). Against this 407 background, it is necessary to involve as many ecologically sound and sustainable control 408 methods as possible in integrated pest management (IPM) systems. 409

410 1.5.2 Cultural Control 411 Cultural control is a long-established method of modifying the habitat to make the 412 environment unfavourable for the survival and reproduction of pests (Calatayud et al., 2014). 413 Indeed, it is the most relevant and economic method of stemborer control available for resource- 414 poor farmers in Africa (Van den Berg et al., 1998). This management strategy is considered the 415 first line of defense against pests and among the oldest traditional practices. 416 Practicing good crop hygiene, which includes the destruction of crop residues (stems and 417 stubbles), removing volunteer crop plants and/or alternative hosts can be used as a cultural 418 control strategy against stemborer infestations. This reduces carryover of stemborers from one 419 growing season to the next, and will help limit the most damaging attacks on young crops early 420 in the growing season. Burning of crop debris breaks the life cycle by destroying larvae that 421 hibernate inside stalks and plant material. It has been suggested that the partial burning of stalks 422 will result in the killing of 95% of larvae (Adesiyum & Ajay, 1980; Plant Health Australia, 423 2009). Damage avoidance by manipulation of sowing dates may also be used to avoid 424 synchronisation of the pest and the most vulnerable plant stages. However, this is not practical in 425 situations where lack of water is a major constraint as farmers usually plant after first rains

426 (Adesiyum & Ajay, 1980). Studies on several stemborers in Africa showed that soil nutrient 427 levels, such as nitrogen, greatly influenced nutritional status of the plant and the plant’s tolerance 428 to stemborer attack (Mgoo et al., 2006). Although an increase in nitrogen is related to higher pest 429 level loads and tunnel damage, there is also an increase in plant vigour with a net benefit to the 430 plant as reflected in lower yield losses (Sétamou et al., 1995; Mgoo et al., 2006). Intercropping 431 maize with cowpea is an effective way of reducing damage by the spotted stemborer larvae 432 migrating from neighbouring plants, where the cowpeas serve as a physical barrier to the 433 movement of borers across plants. 434

435 1.5.3 Push - Pull Technology 436 The push-pull technology is a strategy for controlling agricultural pests by using repellant 437 “push”- and attractant “trap” plants (Cook et al., 2007). The pull approach relies on a 438 combination of companion crops to be planted around and among maize and sorghum crops. 439 Both domestic and wild grasses can help to protect the crops by attracting and trapping 440 stemborers. The grasses are planted in the border around the maize and sorghum fields where 441 invading adult moths become attracted to chemicals emitted by the grasses. Instead of landing on 442 the maize or sorghum plants, the insects head to the attractive trap plant (Cook et al., 2007). 443 These grasses also serve as a haven for the borers’ natural enemies which ultimately attack 444 stemborers species on cultivated cereal crops. Napier grass (Pennisetum purpureum) and Sudan 445 grass (Sorghum vulgare sudanese) act as trap crops with Napier more attractive to stemborers 446 than maize. It ‘pulls’ the moths to lay eggs on it but does sustain stemborer larvae to develop on 447 it due to the sticky substance it secretes which physically traps the insect pest and limits its 448 damage (Cook et al., 2007). In addition, Sudan grass attracts maize female stemborers for 449 oviposition resulting in maize yield increase (Khan et al., 1997). The “push” approach is 450 provided by the plants that emit chemicals (kairomones) which repel stemborer moths and drive 451 them away from the main crop (maize or sorghum). Desmodium uncinatum planted between 452 rows of maize or sorghum is well known for producing a smell or odour that repels stemborer 453 moths (Khan et al., 1997). Since it is a legume, it has the advantage of improving soil fertility 454 through enhanced soil organic matter content and nitrogen fixation (Khan et al., 1997). In 455 addition, it can also serve as an animal feed and can also effectively suppress deleterious weeds, 456 for example Striga asiatica (Teka, 2014). Intercropping maize with molasses grass (Melinis

457 minutiflora), a non-host for stemborers, will also reduce stemborer infestation on maize. This 458 grass produces volatile agents which repel stemborers but attract the parasitic wasp Cotesia 459 sesamiae (Khan et al., 2005). 460

461 1.5.4 Biological control 462 Biological control is a method of managing pests that relies on predation, parasitism, 463 herbivory and other natural mechanisms and can be an important component of IPM programs 464 (Hoy, 1994; Chidawanyika et al., 2012). The main attraction of biological control is that it 465 lowers the need for using chemicals and there is therefore no environmental pollution, which 466 affect non-targeted flora and fauna. If pests reach economically damaging levels, biological 467 control agents may not imminently eliminate pest pressure (Linker et al., 2009). Thus, under 468 these scenarios, biological controls should be integrated with other control measures because 469 different methods are effective at different times and locations under varying conditions (Anchal 470 et al., 2013) 471 There are a number of known natural enemies of cereal stemborers in Southern Africa 472 including parasitoids of eggs, larvae and pupae (see Mutamiswa et al., 2017b). Predators like 473 ants, spiders, earwigs, nematodes and microbial pathogens have been reported to attack different 474 life stages under various natural conditions. In general, evidence has shown that indigenous 475 natural enemies are unable to keep stemborer populations below economic injury levels (Bonhof 476 et al., 1997). Natural enemies successful in controlling C. partellus include the pathogen Bacillus 477 thuringiensis kurstaki, egg parasitoids Trichogramma spp. pupal parasitoids Xanthopimpla 478 stemmator, Dentichasmias busseolae, Pediobus furvus, Lepidoscelio sp. (Sithole, 1990) and larval 479 parasitoids Cotesia sesamiae, Cotesia flavipes. Bacillus thuringiensis kurstaki attacks stemborer 480 larvae and Cotesia flavipes has been successful in attacking stemborer larvae in Kenya, Pakistan, 481 South Africa, Uganda and Tanzania (Kfir et al., 2002). 482

483 1.6 Biology and ecology of Cotesia flavipes and Cotesia sesamiae 484 Cotesia flavipes and C. sesamiae have similar biology and ecology (Overholt et al., 1997; 485 Niyibigira, 2003). Cotesia flavipes, native to Indo-Australian region, is a gregarious koinobiont, 486 larval endo-parasitoid of lepidopteran stemborers (Emana, 2007). Cotesia sesamiae (Cameron)

487 (Hymenoptera: Braconidae), indigenous to Africa belongs to the C. flavipes monophyletic 488 complex (Kimani-Njogu & Overholt, 1997; Muirhead et al., 2012) and is also a gregarious 489 endoparasitoid of lepidopteran stemborers of Crambidae, Pyralidae, and Noctuidae families. 490 Cotesia flavipes was imported, mass reared in the 90's, and subsequently released in East 491 and Southern Africa. Adult females of C. flavipes and C. sesamiae parasitise medium and large 492 sized (4-6th instar) stemborer larvae while the stemborers are feeding inside the plant stems 493 (Overholt et al., 1994; Ngi-Song et al., 1997). Each female wasp lays about 40-60 eggs in a host 494 larva (Mochiah et al., 2001). At 25˚C, 50-80% RH, and 12L:12D photocycle, the egg/larval 495 period lasts 9-13 days, after which the larvae exit the host and spin cocoons (Overholt et al., 496 1994). Adults emerge from cocoons 5-6 days later (Overholt et al., 1994) and the sex ratio is 497 usually female biased (60-70%) (Poting, 1996). Adult longevity ranges from ˂ 24 hours if the 498 parasitoid is unfed (Wiedenmann et al., 1992) to as long as 10 days if honey and water are 499 provided (Overholt et al., 1994). Following field releases, C. flavipes is now established in 500 several countries (Kenya, Tanzania, Mozambique, Uganda, Ethiopia, Zambia, Zimbabwe, 501 Zanzibar, Malawi, and Somalia) (Kfir et al., 2002). When C. partellus is the host, C. flavipes has 502 a higher host searching capacity than C. sesamiae (Sallam et al., 1999). The distribution of these 503 two Cotesia species is influenced by climate, with C. sesamiae common in wetter regions while 504 C. flavipes is common in warm and dry regions (Songa et al., 2001; Niyibigira, 2003). 505

506 1.7 Problem statement and justification of the study 507 Abiotic factors such as temperature and relative humidity are critical in insect pest 508 management and successful control with temperature being the most predominant (Aneni et al., 509 2013). Most importantly, temperature determines the likelihood of reproductive sterility or 510 mortality, especially at extremes and is therefore of critical importance to insect population 511 dynamics (Wang et al., 2014). However, temperature tolerance of an insect is not a fixed but 512 rather plastic trait (Nyamukondiwa & Terblanche, 2009; Stotter & Terblanche, 2009). 513 Temperature tolerance may vary significantly depending on the ecological and evolutionary 514 history of a species (e.g. temperate vs. tropical environment) (Kleynhans et al., 2014), or its life- 515 stage (e.g. adult, larvae) or age and sex (Nyamukondiwa & Terblanche, 2009). Ultimately, it is 516 these factors, combined with complex interactions between duration and severity of exposure 517 (longer or more severe exposures are typically more lethal) that determine an insect’s thermal

518 tolerance. Therefore, a comprehensive understanding of the effects of temperature, rainfall and 519 relative humidity on interactions between insect pests and parasitoids is important in developing 520 improved pest management strategies (Aneni et al., 2013). However, this information is 521 currently unknown for lepidopteran stemborers and their antagonistic parasitoids. 522 Some insect pests have been reported to increase their invasion potential in relation to 523 their ability to deal with changing climates through phenotypic plasticity and variation in basal 524 tolerance (Nyamukondiwa et al., 2010; Chidawanyika et al., 2012). In addition, diapause which 525 is a form of suspended growth and development (Alford, 2010), behavioural adaptation such as 526 habitat selection, basking intensity, restriction of activity periods and selective exploitation of 527 environmental thermal fluxes (Yang et al., 2008; Piyaphongkul, 2013) also increase invasion 528 potential. A classic example is the invasive spotted stemborer C. partellus, which was first 529 introduced in Africa accidentally in Malawi around 1930 (Tams, 1932) but managed to establish 530 itself in several African countries, becoming more destructive than the major indigenous species 531 namely B. fusca, S. calamistis and E. saccharina (Sithole, 1990; Chidawanyika et al, 2012; 532 Mutamiswa et al., 2017a). In addition, this stemborer species has been reported extending 533 geographical distribution from lowland areas to high elevation tropical areas and humid 534 transitional regions (Khadioli et al., 2014) thereby displacing the indigenous species 535 (Mutamiswa et al., 2017a). Whilst the first introduction of C. partellus into Africa may have not 536 been due to climate change, it has become apparent that its further establishment in several 537 African regions follows a distinct pattern, which may be partly influenced by both climate and 538 altitude (Ong’amo et al., 2006; Chidawanyika et al., 2012) as well as repeated propagule 539 pressure (Hayes & Barry, 2008; Mainali et al., 2015). However, information on plasticity of 540 thermal tolerance for C. partellus, B. fusca, S. calamistis and their larval parasitoids C. sesamiae 541 and C. flavipes is currently unknown. 542 In this regard, the ability of insects to undergo such dramatic and rapid changes in 543 temperature tolerance deserves thorough investigation for several reasons of which four are most 544 significant to biological control. First, since the thermal tolerance of a species forms an integral 545 component of bioclimatic and phenology modelling (e.g. Kuhrt et al., 2006), without knowledge 546 of a species’ temperature tolerance any prediction of a pest’s geographic distribution and 547 abundance under future climate change scenarios is severely limited. This has great implications 548 on pest/parasitoid interactions and consequently biological control. Second, the ability of an

549 insect species to undergo rapid thermal hardening, or improve its temperature tolerance quickly, 550 could severely affect interactions between pest and its natural enemy at the short and long 551 timescales. This is especially important since terrestrial organisms are likely to experience 552 increasing frequency and magnitude of extreme temperatures with predicted climate change 553 (Easterling et al., 2000; Tomozeiu et al. 2006; Im et al., 2011). Third, C. flavipes and C. 554 sesamiae have overlapping ecological niches, and often have synergistic effects on biological 555 control efficacy of stemborers. However, how climate change will affect the two congeners, and 556 its effects on biological control is not well understood. Fourth, insects encounter heterogeneous 557 stressors periodically and simultaneously under natural environments (Bubliy et al., 2012) and 558 these have a bearing on their thermal tolerance ultimately influencing trophic level interactions 559 in the agroecosystem, biogeography and biological control efficacy in the face of climate change. 560 Against this background, an understanding of cereal stemborers (B. fusca, S. calamistis 561 and C. partellus) and their larval endoparasitoids (C. flavipes and C. sesamiae) population 562 dynamics under variable climate conditions is essential for predicting seasonal phenology and 563 accurate forecasting of moth abundance which has downstream impacts on basic applications of 564 any biological control programme (timing and parasitoid). Fluctuating patterns of stemborer 565 abundance suggest that DD models are perhaps inadequate or incomplete for predicting pest 566 numbers (Worner, 1992; Bonhomme, 2000) because insect microhabitats temperatures are 567 different from the environmental temperature data obtained from weather stations used in DD 568 calculations (Herms, 2004). Moreover, many insects employ physiological and behavioral 569 mechanisms to counter extreme conditions. For example, insects may move to dark surfaces in 570 the sun when they are too cool, and to light surfaces in the shade when they are too warm 571 (Herms, 2004). Key aspects of variation in population dynamics are still poorly predicted and 572 problematic, resulting at least in some cases repeated and unpredicted outbreaks or infestations 573 even in areas under strict pest control. The reason for this mismatch between predictions and 574 observed moth catches are unclear but could include a lack of trait variability incorporated into 575 DD models or that abiotic factors interact in a non-linear fashion to affect population dynamics 576 and performance. 577 While previous studies have assayed thermal tolerance in Ceratitis sp. Bactrocera sp. 578 Drosophila sp. plant hoppers and codling moths (e.g. Koveos, 2001; Grout & Stoltz, 2007; 579 Rajamohan & Sinclair, 2008; Kalosaka et al., 2009; Nyamukondiwa & Terblanche, 2009;

580 Chidawanyika & Terblanche, 2011a; Piyaphongkul, 2013), little has been done for stemborers 581 and their parasitoids, especially in the context of climate change. Indeed, dramatic changes in 582 high and low temperature tolerance can occur between life-stages of various insect species 583 (Bowler & Terblanche, 2008) but this has not been investigated for B. fusca, S. calamistis C. 584 partellus and their parasitoids C. flavipes and C. sesamiae. Moreover, to my knowledge, no 585 study to date, has looked at trophic level thermal biology, with a view of predicting how 586 changing thermal regimes may offset trophic level coevolved relationships (though see Schreven 587 et al., 2017). Consequently, the kind of data that is required to understand how natural diurnal 588 temperature and humidity fluctuations impact population dynamics through direct temperature 589 and moisture-dependent mortality is currently not available. To better elucidate the link between 590 cereal stemborers (B. fusca, S. calamistis and C. partellus) and their parasitoids (C. flavipes and 591 C. sesamiae) thermal biology, population persistence and species expansion, this study aims at 592 developing a mechanistic physiological model approach (and one which considers B. fusca, 593 S.calamistis, C. partellus, C. flavipes and C. sesamiae developmental stages) to understanding 594 thermal biology and population dynamics of these insect species. This specifically incorporates 595 the thermal parameters such as lower and upper lethal temperatures, critical limits to activity, 596 SCPs, CCRT and HKDT measured from laboratory experiments. The data generated from this 597 study will fill a substantial knowledge gap for a range of environmental conditions with 598 significant management implications. Importantly, these results will directly improve the 599 predictive ability of which environmental factors determine population fluctuations in B. fusca, 600 S. calamistis, C. partellus, C. flavipes and C. sesamiae with consequent improvement in the 601 efficacy of biological control programmes. 602

603 1.8 Objectives

604 1.8.1 Overall Objective 605 The broad objective of this project was to improve understanding of climate variability on 606 stemborers-natural enemies trophic level interaction and its implications on biological control 607 under global change 608

609 1.8.2 Specific Objectives 610 1. To investigate the dominance of exotic C. partellus relative to indigenous stemborer 611 species (Chapter 2) 612 2. To determine basal thermal tolerance for C. partellus and C. sesamiae (Chapter 3) 613 3. To assess phenotypic plasticity of temperature tolerance for C. partellus and C. flavipes 614 (Chapter 4) 615 4. To determine thermal tolerance for larval endoparasitoids C. sesamiae and C. flavipes 616 (Chapter 5) 617 5. To assess the effects of environmental heterogeneity on B. fusca, S. calamistis and C. 618 partellus thermal fitness (Chapter 6) 619

620 1.8.3 Hypotheses 621 1. Exotic C. partellus and indigenous stemborer species are equally codominant 622 2. Thermal tolerance of C. partellus and C. sesamiae are in synchrony and their interaction 623 may evolve together with thermal variability 624 3. Chilo partellus is equally thermally plastic compared to its biological control agent, C. 625 flavipes 626 4. Cotesia sesamiae and C. flavipes have similar basal thermal tolerance. 627 5. Heterogeneous environmental stressors have no effects on B. fusca, S. calamistis and C. 628 partellus thermal fitness 629

630 References 631 Adesiyum, A.A. & Ajay, O. (1980) Control of sorghum stem borer Busseola fusca by partial 632 burning of stalk. International Journal of Pest Management, 26, 113-117. 633 Addo-Bediako, A. & Thanguane, N. (2012) Stemborer distribution in different sorghum cultivars 634 as influenced by soil fertility. Agricultural Science Research Journal, 2, 189- 194. 635 Alford, L. (2010) The thermal macrophysiology of core and marginal populations of the aphid 636 Myzus persicae in Europe. PhD Thesis. University of Birmingham.

637 Anchal, S., Diwevidi, V.D., Singh, S., Pawar, K.K., Jerman, M., Singh, L.B., Singh, S. & 638 Srivastawa, D. (2013) Biological control and its importance in agriculture. International 639 Journal of Biotechnology and Bioengineering Research, 3, 175-180. 640 Aneni, T.I., Aisagbonhi, C.I., Iloba, B.N., Adaigbe, V.C. & Ogbebor, C.O. (2013) Influence of 641 weather factors on seasonal population dynamics of Coelaenomenodera elaeidis (Coleoptera: 642 Chrysomelidae) and its natural enemies in NIFOR, Nigeria. Proceedings of the International 643 Academy of Ecology and Environmental Sciences, 3(4), 344-352. 644 Angilletta, M.J. (2009) Thermal adaptation: A theoretical and empirical synthesis. New York, 645 New York: Oxford University Press. 646 Austin, C.J. (2013) Thermal adaptation of life history traits in the Drosophila melanogaster 647 group. University of Western Ontario - Electronic Thesis and Dissertation Repository. Paper 648 1685. 649 Ayrinhac, A., Debat, V., Gibert, P., Kister, A.G., Legout, H., Moreteau, B., Vergilino, R. & 650 David J.R. (2004) Cold adaptation in geographical populations of Drosophila melanogaster: 651 phenotypic plasticity is more important than genetic variability. Functional Ecology, 18, 700 652 – 706. 653 Bahrndorff, S., Gertsen, S., Pertoldi, C. & Kristensen, T.N. (2016) Investigating thermal 654 acclimation effects before and after a cold shock in Drosophila melanogaster using 655 behavioural assays. Biological Journal of the Linnean Society, 117(2), 241-251. 656 Bale, J.S., Masters, G.J., Hodkinson, I.D., Awmack, C., Bezemer, T.M., Brown, V.K., 657 Butterfield, J., Buse, A., Coulson, J.C., Farrar, J., Good, J.E.G., Harrington, R., Hartley, S., 658 Jones, T.H., Lindroth, R.L., Press, M.C., Symrnioudis, I., Watt, A.D. & Whittaker, J.B. 659 (2002) Herbivory in global climate change research: direct effects of rising temperature on 660 insect herbivores. Global Change Biology, 8, 1–16 661 Bale, J.S., Strathdee, A.T. & Strathdee, F.C. (1994) Effects of low temperature on the arctic 662 aphid Acrythosiphon brevicorne . Functional Ecology, 8, 621 – 626. 663 Bale, J.S. & Walters, K.F.A. (2001) Overwintering biology as a guide to the establishment 664 potential of non-native arthropods in the UK. In Environment and Animal Development. 665 Genes, Life Histories and Plasticity. BIOS Scientific Publishers Ltd, Oxford, UK. p 358.

666 Bate, R., Van Rensburg, G.D.J. & Pringle, K.L. (1990) Oviposition site preference by the grain 667 sorghum stem borer Chilo partellus (Swinhoe) (Lepidoptera: Pyrallidae) on maize. Journal 668 of the Entomological Society of Southern Africa, 53, 75 - 79. 669 Beitinger, T.L., Bennett, W.A. & McCauley, R.W. (2000) Temperature tolerance of North 670 American freshwater fishes exposed to dynamic changes in temperature. Environmental 671 Biology of Fishes, 58, 237–275. 672 Berger, A. (1992) Larval movement of Chilo partellus (Lepidoptera: Pyrallidae) within and 673 between plants: timing, density responses and survival. Bulletin of Entomological Research, 674 82, 441 - 448. 675 Bonhof, M.J., Overholt, W.A., Huis, A. & Polaszek, A. (1997) Natural enemies of cereal stem 676 borers in East Africa: a review. Insect Science and its Application, 17, 19-35. 677 Bonhomme, R. (2000) Bases and limits to using degree day units. European Journal of 678 Agronomy, 13(1), 1–10. 679 Bosque-Pérez, N.A. & Schulthess, F. (1998) Maize: West and central Africa. African Cereal 680 Stem Borers: Economic Importance, Taxonomy, Natural Enemies and Control. CAB 681 International, Wallingford. Pp 11-24. 682 Bosque-Perez, N.A. (1995) Major Insect Pests of Maize in Africa: Biology and Control. IITA 683 Research Guide 30. Training Program; International Institute of Tropical Agriculture (IITA), 684 Ibadan. Nigeria. Pp 30. 685 Bowler, K. & Terblanche, J.S. (2008) Insect thermal tolerance: what is the role of ontogeny, 686 ageing and senescence. Biological Reviews, 83, 339-355. 687 Bubliy, O.A., Kristensen, T.N., Kellermann, V. & Loeschcke, V. (2012) Plastic responses to four 688 environmental stresses and cross-resistance in a laboratory population of Drosophila 689 melanogaster. Functional Ecology, 26, 245–253. 690 Calatayud, P.-A., Okuku, G., Musyoka, B., Khadioli, N., Ong’amo, G., & Le Ru, B. (2014). 691 Busseola segeta, a potential new pest of maize in western Kenya. Entomology Ornithology & 692 Herpetology: Current Research, 3, 132. 693 Castaneda, L.E., Lardies, M.A. & Bozinovic, F. (2005) Interpopulational variation in recovery 694 time from chill coma along a geographic gradient: a study in the common woodlouse, 695 Porcellio laevis. Journal of Insect Physiology, 51, 1346 – 1351.

696 Chabi-Olaye, A., Schulthess, A.F., Poehling, H.M. & Borgemeister, C. (2001) Factors affecting 697 biology of Telenomous isis (Polaszek) (Hymenoptera: Scelionidae), an egg parasitoid of 698 cereal stem borers in West Africa. Biological Control, 21, 44 - 54. 699 Chidawanyika, F., Mudavanhu, P. & Nyamukondiwa, C. (2012) Biologically based methods for 700 pest management in agriculture under changing climates: Challenges and future directions. 701 Insects, 3, 1171-1189. 702 Chidawanyika, F. & Terblanche, J.S. (2011a) Rapid thermal responses and thermal tolerance in 703 adult codling moth Cydia pomonella (Lepidoptera:Totricidae). Journal of Insect Physiology, 704 57, 108-117. 705 Chidawanyika, F. & Terblanche, J.S. (2011b) Costs and benefits of thermal acclimation for 706 codling moth, Cydia pomonella (Lepidoptera: Tortricidae): implications for pest control and 707 the sterile insect release programme. Evolutionary Applications, 4, 534-544. 708 Chinwada, P. (2002) Stem borer parasitism by Cotesia sesamiae and Sturmiopsis parasitica and 709 an assessment of the need to introduce Cotesia flavipes in Zimbabwe. Ph.D. Thesis, Kenyatta 710 University. Kenya. 711 Chown, S.L., Jumbam, K.R., Sørensen, J.G. & Terblanche, J.S. (2009) Phenotypic variance, 712 plasticity and heritability estimates of critical thermal limits depend on methodological 713 context. Functional Ecology, 23, 133-140. 714 Chown, S.L. & Terblanche, J.S. (2007) Physiological diversity in insects: Ecological and 715 evolutionary contexts. Advances in Insect Physiology, 33, 50-153. 716 Chown, S.L. & Nicolson, S.W. (2004) Insect Physiological Ecology: Mechanisms and Patterns. 717 Oxford University Press, Oxford. 718 Cook, S. M., Khan, Z.R. & Pickett. J.A. (2007) The use of push-pull strategies in integrated 719 pest management. Annual Review of Entomology, 52, 375-400. 720 Cooper, B.S., Czarnoleski, M. & Angilletta, M.J. (2010) Acclimation of thermal physiology in 721 natural populations of Drosophila melanogaster: a test of an optimality model. Journal of 722 Evolutionary Biology, 23, 2346–2355. 723 Conlong, D.E. (1994) A review and perspectives for the biological control of the African 724 sugarcane stalk borer Eldana saccharina Walker (Lepidoptera: Pyralidae). Agriculture, 725 Ecosystem and Environment, 48, 9-17.

726 David, R.J., Gibert, P., Pla, E., Petavy, G., Karan, D. & Moreteau, B. (1998) Cold stress 727 tolerance in Drosophila-analysis of chill coma recovery in D. melanogaster. Journal of 728 Thermal Biology, 23, 291–299. 729 De Groote, H. (2002) Maize yield loss from stem borers in Kenya. Journal of Insect Science, 22, 730 89–96. 731 Deka, S., Barthakur, S. Pandey, R. (2009) Potential Effects of Climate Change on Insect Pest 732 Dynamics. Indian Agricultural Research Institute. New Delhi. pp 301-312. 733 Denlinger, L.D. & Lee, R.E. (2010) Low Temperature Biology of Insects. Cambridge University 734 Press. Cambridge. 735 Deutsch, C.A., Tewksbury, J.J., Huey, R.B., Sheldon, K.S., Ghalambor, C.K., Haak, D.C. 736 Martin, P.R. (2008) Impacts of climate warming on terrestrial ectotherms across latitude. 737 Proceedings of the National Academy of Science USA, 105, 6668–6672. 738 (doi:10.1073/pnas.0709472105). 739 Easterling, D.R., Meehl, G.A., Parmesan, C., Changnon, S.A., Karl, T.R. & Mearns, L.O. (2000) 740 Climate extremes: observations, modelling, and impacts. Science, 289, 2068-2074. 741 Emana, G.D. (2007) Comparative studies of the influence of relative humidity and temperature 742 on the longevity and fecundity of the parasitoid, Cotesia flavipes. Journal of Insect Science, 743 7(19), 1-7. 744 Fand, B.B., Kamble, A.L. & Kumar, M. (2012) Will climate change pose serious threat to crop 745 pest management: A critical review? International Journal of Scientific and Research 746 Publications, 2(11), 2250-3153. 747 Gaston, K.J. & Chown , S.L. (1999) Elevation and climatic tolerance: a test using dung beetles. 748 Oikos, 86, 584 – 590. 749 Ghalambor, C.K., Huey, R.B., Martin, P.R., Tewksbury, J.J. & Wang, G. (2006) Are mountain 750 passes higher in the tropics? Janzen’s hypothesis revisited. Integrative and Comparative 751 Biology, 46, 5-17. 752 Gibert, P., Moreteau, B., Petavy, G., Karan, D. & David, J.R. (2001a) Chill coma tolerance, a 753 major climatic adaptation among Drosophila species. Evolution, 55, 1063–1068 754 Gibert, P., Huey, R.B. & Gilchrist, G.W. (2001b) Locomotor performance of Drosophila 755 melanogaster: interactions among developmental and adult temperatures, age, and 756 geography. Evolution, 55, 205–209.

757 Gilchrist, G.W. & Huey, R.B. (1999) The direct response of Drosophila melanogaster to 758 selection on knockdown temperature. Heredity, 83, 15–29. 759 Gray, E.M. (2013) Thermal acclimation in a complex life cycle: The effects of larval and adult 760 thermal conditions on metabolic rate and heat resistance in Culex pipiens (Diptera: ulicidae). 761 Journal of Insect Physiology, 59, 1001-1007. 762 Grout, T.G. & Stoltz, K.C. (2007) Developmental rates at constant temperatures of three 763 economically important Ceratitis species (Diptera: Tephritidae) from Southern Africa. 764 Environmental Entomology, 36, 1310-1317. 765 Gutierrez, A.P., Ponti, L., Oultremont, T. & Ellis, C.K. (2008) Climate change effects on 766 poikilotherm tritrophic interactions. Climate Change, 87, 167-192. 767 Hance, T., van Baaren, J., Vernon, P. & Boivin, G. (2007) Impact of extreme temperatures on 768 parasitoids in a climate change perspective. Annual Review of Entomology, 52, 107-126. 769 Hayes, K.R. & Barry, S.C. (2008) Are there any consistent predictors of invasion success? 770 Biological Invasions, 10, 483-506. 771 Hazell, S.P., Neve, B.P., Groutides, C., Douglas, A.E., Blackburn, T.M. & Bale, J.S. (2010) 772 Hyperthermic aphids: insights into behaviour and mortality. Journal of Insect Physiology, 56, 773 123-131. 774 Hazell, S.P., Perdesen, B.P., Worland, M.R., Blackburn, T.M. & Bale, J.S. (2008) A method for 775 the rapid measurement of thermal tolerance traits in studies of small insects. Physiological 776 Entomology, 33, 389-394. 777 Hemmati, C., Moharramipour, S. & Talebi, A.A. (2014) Effects of cold acclimation, cooling rate 778 and heat stress on cold tolerance of the potato tuber moth Phthorimaea operculella 779 (Lepidoptera: Gelechiidae). European Journal of Entomology 111, 487-494. 780 Herms, D.A. (2004) Using degree-days and plant phenology to predict pest activity, pp. 49-59. In 781 Krischik., V. & Davidson, J. (eds) Integrated Pest Management of Midwest landscapes. 782 Minnesota Agricultural Experiment Station Publication SB-07645. St Paul. Minnesota. 783 Hill, D.S. (1987) Agricultural insect pests of the tropics and their control. 2nd Edition. Cambridge 784 University Press. Cambridge. UK 785 Hochachka, P. W. & Somero, G.N. (2002) Biochemical Adaptation. New York, NY: Oxford 786 University Press.

787 Hoffmann, A.A., Sørensen, J.G. & Loeschcke, V. (2003) Adaptation of Drosophila to 788 temperature extremes: bringing together quantitative and molecular approaches. Journal of 789 Thermal Biology, 28, 175–216. 790 Hoy, M.A. (1994) Parasitoids and Predators in Management of Arthropod Pests. In Introduction 791 to Insect Pest Management; Metcalf, R.L., Luckman, W.H., Eds.; John Wiley Sons: New 792 York, NY, USA. 793 Huey, R.B., Crill, W.D., Kingsolver, J.G. & Weber, K.E. (1992) A method for rapid 794 measurement of heat or cold resistance of small insects. Functional Ecology, 6, 489-494 795 Huey, R.B., Hertz, P.E. & Sinervo, B. (2003) Behavioural drive versus behavioural inertia in 796 evolution: a null model approach. The American Naturalist, 161, 357–366. 797 Im, E.S., Jung, I.W. & Bae, D.H. (2011) The temporal and spatial structures of recent and future 798 trends in extreme indices over Korea from a regional climate projection. International 799 Journal of Climatology, 31, 72-86. 800 Jaworski, T. & Hilszczański, J. (2013) The effect of temperature and humidity changes on 801 insects development and their impact on forest ecosystems in the context of expected climate 802 change. Leśne Prace Badawcze. Forest Research Papers, 74 (4), 345-355 803 Jensen, D., Overgaard, J. & Sorensen, J.G. (2007) The influence of developmental stage on cold 804 shock resistance and ability to cold-harden in Drosophila melanogaster. Journal of Insect 805 Physiology, 53, 179-186. 806 Jumbam, K.R., Jackson, S., Terblanche, J.S., McGeoch, M.A. & Chown, S.L. (2008) 807 Acclimation effects on critical and lethal thermal limits of workers of the Argentine ant, 808 Linepithema humile. Journal of Insect Physiology, 54, 1008–1014. 809 Kalosaka, K., Soumaka, E., Politis, N. & Mintzas, A.C. (2009) Thermotolerance and Hsp70 810 expression in Mediterranean fruit fly Ceratitis capitata. Journal of Insect Physiology, 55, 811 568-573. 812 Karuppaiah, V. & Sujayanad, G.K. 2012. Impacts of climate change on population dynamics of 813 insect pests. World Journal of Agricultural Sciences, 8(3), 240-246 814 Kfir, R., Overholt, W.A., Khan, Z.R. & Polaszek, A. (2002) Biology and management of 815 economically important lepidopteran cereal stem borers in Africa. Annual Review of 816 Entomology, 47, 701-731.

817 Kfir, R. (1998) Maize and Sorghum: Southern Africa, African Cereal Stem borers, Economic 818 Importance, Taxonomy, Natural Enemies and Control. CAB International. Wallingford, 819 Oxon, UK. Pp. 29-37. 820 Kfir, R. (1997) Competitive displacement of Busseola fusca (Lepidoptera: Noctuidae) by Chilo 821 partellus (Lepdoptera: Pyralidae). Annals of the Entomological Society of America, 90, 619- 822 624. 823 Khadioli, N., Tonnang, Z.E.H., Muchugu, E., Ong’amo, G., Achia, T., Kipchirchir, I., Kroschel, 824 J. & Le Ru, B. (2014) Effect of temperature on the phenology of Chilo partellus (Swinhoe) 825 (Lepidoptera: Crambidae), simulation and visualization of the potential future distribution of 826 C. partellus in Africa under warmer temperatures through the development of life-table 827 parameters. Bulletin of Entomological Research, 104, 809–822. 828 Khan, Z.R., Chiliswa, P., Ampong-Nyarko, K., Smart, L.E., Polaszek, A., Wandera, J. & Mulaa, 829 M.A. (1997) Utilisation of wild gramineous plants for management of cereal stemborers in 830 Africa. Insect Science and its Application, 17, 143 - 150. 831 Khan, Z.R., Muyekho, F.N., Njuguna, E., Pickett, J.A., Wadhams, L.J., Dibogo, N., Ndiege, A., 832 Genga, G. & Luswetti, C. (2005) A Primer on Planting and Managing 'Push-Pull' Fields for 833 Stemborer and Striga Control in Maize - A Step-by-Step Guide for Farmers. ICIPE. Nairobi 834 Kimani-Njogu, S.W. & Overholt, W.A. (1997) Biosystematics of the Cotesia flavipes species 835 complex (Hymenoptera: Braconidae), parasitoids of the gramineous stemborers. Insect 836 Science and its Application, 17, 119–130. 837 Kipkoech, A.K., Schulthess, F., Yabann, W.K., Maritim, H.K. & Mithöfer, D. (2006) Biological 838 control of cereal stem borers in Kenya: A cost benefit approach, Annales de la Société 839 entomologique de France (N.S.): International Journal of Entomology, 42, 519-528. 840 Kleynhans, E., Conlong, D. & Terblanche, J.S. (2014) Host plant-related variation in thermal 841 tolerance if Eldana saccharina. Entomologia Experimentalis et Applicata, 150, 113-122. 842 Koveos, D.S. (2001) Rapid cold hardening in the olive fruit fly Bactrocera oleae under 843 laboratory conditions. Entomologia Experimentalis et Applicata, 101, 257-263 844 Kristensen, T.N., Hoffmann, A.A., Overgaard, J., Sørensen, J.G. & Hallas, R. (2008) Costs and 845 benefits of cold acclimation in field released Drosophila. Proceedings of the National 846 Academy of Science of the United States of America, 105, 216-221.

847 Kuhrt, U., Samietz, J., Hohn, H. & Dorn, S. (2006) Modelling the phenology of codling moth: 848 Influence of habitat and thermoregulation. Agriculture, Ecosystems and Environment, 117, 849 29-38. 850 Lee, R.E. & Costanzo, J.P. (1998) Biological ice nucleation and ice distribution in cold-hardy 851 ectothermic animals. Annual Review of Physiology, 60, 55-72. 852 Linker, H.M., Orr, D.B. & Barbercheck, M.E. (2009) Insect Management on Organic Farms. 853 North Carolina Cooperative Extension Service. North Carolina. 854 Loeschcke, V. & Hoffmann, A.A. (2007) Consequences of heat hardening on a field fitness 855 component in Drosophila depend on environmental temperature. The American Naturalist, 856 169, 175–183. 857 Lutterschmidt, W.I. & Hutchison, V.H. (1997) The critical thermal maximum: data to support 858 the onset of spasms as the definitive end point. Canadian Journal of Zoology, 75, 1553- 859 1560. 860 Lwande, W., Gikonyo, N., Lux, S., Hassanali, A. & Lofqvist, J. (1993) Periodicity in the 861 frequency of calling and quality of pheromone in volatile emissions of the spotted stalk 862 borer, Chilo partellus. Bulletin OILB/SROP, 16(10), 174. 863 MacMillan, H.A., Schou, M.F., Kristensen, T.N. & Overgaard, J. (2016) Preservation of 864 potassium balance is strongly associated with insect cold tolerance in the field: a seasonal 865 study of Drosophila subobscura. Biology Letters, 12, 20160123. DOI: 866 10.1098/rsbl.2016.0123. 867 Mailafiya, D.M., Le Ru, B.P., Kairu, E.W., Dupas, S. & Calatayud, P.A. (2011) Parasitism of 868 lepidopterous stemborers in cultivated and natural habitats. Journal of Insect Science, 11, 1- 869 19. 870 Mainali, K. P., Warren, D.L., Dhileepan, K., McConnachie, A., Strathie, L., Hassan, G., Karki, 871 D., Shrestha, B.B. & Parmesan, C. (2015) Projecting future expansion of invasive species: 872 comparing and improving methodologies for species distribution modeling. Global Change 873 Biology, 21, 4464-4480. 874 Milton, C.C. & Partridge, L. (2008) Brief carbon dioxide exposure blocks heat hardening but not 875 cold acclimation in Drosophila melanogaster. Journal of Insect Physiology, 54, 32-40.

876 Mgoo, V.H., Makundi, R.H., Pallangyo, B., Schulthess, F. & Jiang, N.O. (2006) Yield loss due 877 to the stem borer Chilo partellus (Swinhoe) (Lepidoptera: Crambidae) at different nitrogen 878 application rates to maize. Annales de la Société Entomologique de France, 42, 487-494. 879 Mochiah, M.B., Ngi - Song, J.A., Overholt, W.A. & Botchey, M. (2001) Host suitability of four 880 cereal stemborers (Lepidoptera: Crambidae, Noctuidae) for different geographic populations 881 of Cotesia sesamiae (Cameron) (Hymenoptera: Braconidae) in Kenya. Biological Control, 882 21, 285 - 292. 883 Muhammad, L. & Underwood, E. (2004) The maize agricultural context in Kenya, 884 Environmental risk assessment of genetically modified organisms: A case study of Bt maize 885 in Kenya. Vol 1. CABI Publishing. Wallingford. 886 Muirhead, K.A., Murphy, N.P., Sallam, N., Donnellan, S.C. & Austin, A.D. (2012) 887 Phylogenetics and genetic diversity of the Cotesia flavipes complex of parasitoid wasps 888 (Hymenoptera: Braconidae), biological control agents of lepidopteran stemborers. Molecular 889 Phylogenetics and Evolution, 63, 904–914. 890 Mutamiswa, R., Chidawanyika, F. & Nyamukondiwa, C. (2017a) Dominance of spotted 891 stemborer Chilo partellus Swinhoe (Lepidoptera: Crambidae) over indigenous stemborer 892 species in Africa’s changing climates: ecological and thermal biology perspectives. 893 Agricultural and Forest Entomology, DOI: 10.1111/afe.12217. 894 Mutamiswa, R., Moeng, E., Le Ru, B.P., Conlong, D.E., Assefa, Y., Goftishu, M. & 895 Nyamukondiwa, C. (2017b) Diversity and abundance of lepidopteran stem borer natural 896 enemies in natural and cultivated habitats in Botswana. Biological Control, 115, 1-11. 897 Nicotra, A.B., Atkin, O, K., Bonser, S.P., Davidson, A, M., Finnegan, E, J., Mathesius, U., Poot, 898 P., Purugganan, M.D., Richards, C.L., Valladares, F. & Kleunen, M. (2010) Plant phenotypic 899 plasticity in a changing climate. Trends in Plant Science, 15, 684-692 900 Nielsen, M., Overgaard, J., Sorensen, J. G., Holmstrup, M., Justesen, J. & Loeschcke, V. (2005) 901 Role of HSF activation for resistance to heat, cold and high-temperature knock-down. 902 Journal of Insect Physiology, 51, 1320–1329. 903 Niyibigira, E.I. (2003) Genetic variability in Cotesia flavipes and its importance in biological 904 control of lepidopteran stem borers. Ph.D. dissertation, Wageningen University, The 905 Netherlands.

906 Ngi-song, A.J. & Overholt, W.A. (1997) Host location and acceptance by Cotesia flavipes 907 Cameron and C. sesamiae (Cameron) (Hymenoptera: Braconidae), parasitoids of African 908 gramineous stem borers: role of frass and other host cues. Biological Control, 6, 136-142 909 Nyamukondiwa, C., Weldon, C.W., Chown, S.L., le Roux, P.C. & Terblanche, J.S. (2013) 910 Thermal biology, population fluctuations and implications of temperature extremes for the 911 management of two globally significant insect pests. Journal of Insect Physiology, 59, 1199- 912 1211. 913 Nyamukondiwa, C., Terblanche, J.S., Marshall, K.E. & Sinclair, B.J. (2011) Basal but not heat 914 tolerance constraints plasticity among Drosophila species (Diptera: Drosophilidae). Journal 915 of Evolutionary Biology, 24, 1927-1938. 916 Nyamukondiwa, C., Kleynhans, E. & Terblanche, J.S. (2010) Phenotypic plasticity of thermal 917 tolerance contributes to the invasion potential of Mediterranean fruit flies (Ceratitis 918 capitata). Ecological Entomology, 35, 565-575. 919 Nyamukondiwa, C. & Terblanche, J.S. (2010) Rapid cold hardening in Zapprionus vittiger 920 (Coquillett) (Diptera: Drosophilidae). CryoLetters, 31(6), 504-512. 921 Nyamukondiwa, C. & Terblanche, J.S. (2009) Thermal tolerance in adult Mediterranean and 922 Natal fruit flies (Ceratitis capitata and Ceratitis rosa): effects of age, gender and feeding 923 status. Journal of Thermal Biology, 34, 406-414. 924 Obonyo, M., Schulthess, F., Le Ru, B.P., Van den Berg, J. & Calatayud, P.A. (2010) Host 925 recognition and acceptance behaviour in Cotesia sesamiae and C. flavipes (Hymenoptera: 926 Braconidae), parasitoids of gramineous stemborers in Africa. European Journal of 927 Entomology, 107, 169-176. 928 Ofomata, V.C., Overholt, W.A., Lux, S.A., Van Huis, A. & Egwuatu, R.I. (2000) Comparative 929 studies on the fecundity, egg survival, larval feeding and development of Chilo partellus 930 (Swinhoe) and Chilo orichalcociliellus Strand (Lepidoptera: Crambidae) on five grasses. 931 Annals of the Entomological Society of America, 93, 492-499. 932 Ong’amo, G.O., Le Rü, B.P., Moyal, P.A., Calatayud, P., Le Gall, P., Ogol, C.K.P.O., Kokwaro, 933 E.D., Capdevielle-Dulac, C. & Silvain, J.F. (2008) Host plant diversity of Sesamia 934 calamistis: cytochrome b gene sequences reveal local genetic differentiation. Entomologia 935 Experimentalis et Applicata, 128, 154-161.

936 Ong’amo, G.O., Le Rü, B.P., Dupas, S., Moyal, P.A. & Silvain, J.F. (2006) Distribution, pest 937 status and agro-climatic preferences of lepidopteran stem borers of maize in Kenya. Annales 938 de la Société Entomologique de France, 42, 171-177. 939 Overholt, W.A., Ngi-Song, A.J., Omwega, C.O., Kimani-Njogu, S.W., Mbapila, J., Sallam, M.N. 940 and Ofomata, V. (1997) A review of the introduction and establishment of Cotesia flavipes 941 Cameron in East Africa for the biological control cereal stemborers. Insect Science and its 942 Application, 17, 79-88. 943 Overholt, W.A., Ochieng, J.O., Lammers, P. & Ogedah, K. (1994) Rearing and field release 944 methods for Cotesia flavipes Cameron (Hymenoptera: Braconidae), a parasitoid of tropical 945 gramineous stemborers. Insect Science and its Application, 15, 253-259. 946 Petzoldt, C. & Seaman, A. (2006) Climate Change Effects on Insects and Pathogens. New York 947 State Agricultural Extension Station. Geneva. NY 14456: 9-12. 948 Piyaphongkul, J. (2013) Effects of thermal stress on brown plant hopper. Phd Thesis. University 949 of Birmingham. Birmingham. England. 950 Plant Health Australia and Kalang Consultancy Services. (2009) Industry biosecurity plan for the 951 grains industry, Threat specific contingency plan: Spotted stem borer Chilo partellus. Plant 952 Health Australia. Deakin Act 2600. 953 Polaszek, A. (1998) African cereal stem borers: Economic importance, taxonomy, natural 954 enemies and control. CABI Publishing. Wallingford. 955 Polaszek, A. & Khan, Z.R. (1998) Host plants. African Cereal Stem Borers: Economic 956 Importance, Taxonomy, Natural Enemies and Control. CAB International, Wallingford. Pp 957 4–10. 958 Poting, R.P.J. 1996. Hunting for hiding hosts, the behavioral ecology of stem borer parasitoid 959 Cotesia flavipes. Netherlands Foundation of Scientific Research in the Tropics (WOTRO - 960 W84-325). Wageningen. 961 Rajamohan, A. & Sinclair B.J. (2008) Short term hardening effects on survival of acute and 962 chronic cold exposure by Drosophila melanogaster larvae. Journal of Insect Physiology, 54, 963 708-718. 964 Sallam, N.M., Overholt, W.A. & Kairu, E. (1999) Comparative evaluation of Cotesia flavipes 965 and C. sesamiae (Hymenoptera: Braconidae) for the management of Chilo partellus 966 (Lepidoptera: Pyralidae) in Kenya. Bulletin of Entomological Research, 89, 185-191.

967 Schlichting, C.D. & Smith, H. (2002) Phenotypic plasticity: linking molecular mechanisms with 968 evolutionary outcomes. Evolutionary Ecology, 16, 189–211. 969 Schreven, S.J.J., Frago, E., Stens, A., de Jong, P.W. & van Loon, J.J.A. (2017) Contrasting 970 effects of heat pulses on different trophic levels, an experiment with a herbivore-parasitoid 971 model system. PLoS ONE, 12(4), e0176704. 972 Scott, M., Berrigan, D. & Hoffmann, A.A. (1997) Costs and benefits of acclimation to elevated 973 temperature in Trichogramma carveraee. Entomologia Experimentalis et Applicata, 85, 211- 974 219. 975 Selvaraj, S., Ganeshamoorthi, P. & Pandiaraj, T. (2013). Potential impacts on recent climate 976 change on biological control agents in agroecosystem: A review. International Journal of 977 Biodiversity and Conservation, 5, 845-852. 978 Seshu Reddy, K.V. (1998) Maize and sorghum: East Africa. African Cereal Stem Borers: 979 Economic Importance, Taxonomy, Natural Enemies and Control. CAB International, 980 Wallingford. Pp 25-27. 981 Sétamou. M., Schulthess, F., Bosque-Pérez, N.A. & Thomas-Odjo, A. (1995) The effect of stem 982 and cob borers on maize subjected to different nitrogen treatments. Entomologia 983 Experimentalis et Applicata, 77, 205-210. 984 Sharma, H. C. & Davies, J.C. (1988) Insect and other animal pests of millets. International Crops 985 Research Institute for the Semi-Arid Tropics. Patancheru. India. 986 Sharma, H.C. & Nwanze, K.F. (1997) Insect pests of sorghum: biology, extent of losses, and 987 economic thresholds. Pages 9-23 in Plant resistance to insects in sorghum (Sharma, H.C., 988 Faujdar Singh, and Nwanze. K.E., eds.). Patancheru 502 324, Andhra Pradesh, India: 989 International Crops Research Institute for the Semi- Arid Tropics. 990 Shreve, S.M., Kelty, J.D. & Lee, R.E. (2004) Preservation of reproductive behaviours during 991 modest cooling: rapid cold hardening fine tunes organismal response. Journal of 992 Experimental Biology, 207, 1797-1802. 993 Sinclair, B.J. (2015) Linking energetics and overwintering in temperate insects. Journal of 994 Thermal Biology, 54, 5-11. 995 Sinclair, B.J., Coello Alvarado, L.E. & Ferguson, L.V. (2015) An invitation to measure insect 996 cold tolerance: Methods, approaches, and workflow. Journal of Thermal Biology, 53, 180- 997 197.

998 Sinclair, B.J., Ferguon, L.V., Salehipour-shirazi, G. & MacMillan, H.A. (2013) Cross-tolerance 999 and cross-talk in the cold: Relating low temperatures to desiccation and immune stress in 1000 insects. Integrative and Comparative Biology, 53, 545-56. 1001 Sinclair, B.J., Terblanche , J.S., Scott , M.B., Blatch, G.L., Jaco-Klok, C. & Chown, S.L. (2006) 1002 Environmental physiology of three species of Collembola at Cape Hallett, North Victoria 1003 Land, Antarctica. Journal of Insect Physiology, 52, 29-50. 1004 Sithole, S.Z. (1990) Status and control of the stem borer, Chilo partellus Swinhoe (Lepidoptera: 1005 Pyralidae) in Southern Africa. International Journal of Tropical Insect Science, 11, 479-488. 1006 Sithole, S.Z. (1989) Sorghum Stem Borers in Southern Africa. International Workshop on 1007 Sorghum Stemborers, 17-20 Nov 1987, ICRISAT Center, India. 1008 Slabber, S., Worland, M.R., Leinaas, H.P. & Chown, S.L. (2007) Acclimation effects on thermal 1009 tolerances of springtails from sub-Antarctic Marion Island: indigenous and invasive species. 1010 Journal of Insect Physiology, 53, 113 – 125. 1011 Slotsbo, S., Schou, M.F., Kristensen, T.N., Loeschcke, V. & Sørensen, J.G. (2016) Reversibility 1012 of developmental heat and cold plasticity is asymmetric and has long lasting consequences 1013 for adult thermal tolerance. Journal of Experimental Biology, doi: 10.1242/jeb.143750. 1014 Somero, G. N. (2005) Linking biogeography to physiology: evolutionary and acclimatory 1015 adjustments of thermal limits. Frontiers in Zoology, 2, 1, DOI: 10.1186/1742-9994-2-1. 1016 Songa, J.M., Overholt, W.A., Mueke, J.M. Okell, R.O. (2001) Colonization of Cotesia flavipes 1017 (Hymenoptera: Braconidae) in stem borers in the semi-arid Eastern province of Kenya. Insect 1018 Science and its Application, 21, 289–295. 1019 Stotter, R. L. & Terblanche, J.S. (2009) Low-temperature tolerance of false codling moth 1020 Thaumatotibia leucotreta (Meyrick) (Lepidoptera: Tortricidae) in South Africa. Journal of 1021 Thermal Biology, 34, 320-325. 1022 Tams, W.H.T. (1932) New species of African Heterocera. Entomologist, 65,1241–1249. 1023 Tattersall, G.J., Sinclair, B.J., Withers, P.C., Fields, P.A., Seebacher, F., Cooper, C.E. & Maloney, 1024 S.K. (2012) Coping with thermal challenges: physiological adaptations to environmental 1025 temperatures. Comprehensive Physiology, 2, 2151–2202. 1026 Tefera, T., Mugo, S., Tende, R. & Likhayo, P. (2010) Mass Rearing of Stem borers, Maize 1027 weevil and Larger grain borer Insect Pests of Maize. CIMMYT. Nairobi.

1028 Tefera, T. (2004) Lepidopterous stemborers of sorghum and their natural enemies in eastern 1029 Ethiopia. Tropical Science, 44, 128-130 1030 Teka, H.B. (2014) Advance research on Striga control: A review. African Journal of Plant 1031 Science, 8(11), 492-506. 1032 Terblanche, J.S., Deere, J.A., Clusella-Trullas, S., Janion, C. & Chown, S.L. (2007) Critical 1033 thermal limits depend on methodological context. Proceedings of the Royal Society of 1034 London B Biological Sciences, 274, 2935–2942. 1035 Thomas, C.D., Cameron, A. Green, R.E., Bakkenes, M. & Beaumont, L.J. (2004) Extinction risk 1036 from climate change. Nature, 427, 145–48. 1037 Thomson, L.J., Macfadyen, S. & Hoffmann, A.A. (2010) Predicting the effects of climate 1038 change on natural enemies of agricultural pests. Biological Control, 52, 296-306. 1039 Tomozeiu, R., Pavan, V., Cacciamani, C. & Amici, M. (2006).Observed temperature changes in 1040 Emilia-Romagna: mean values and extremes. Climate Research, 31, 217-225. 1041 Turnock, W.J. & Fields, P.G. (2005) Winter climates and coldhardiness in terrestrial insects. 1042 European Journal of Entomology, 102, 561–76. 1043 Van den Berg, J., Nur, A.F. & Polaszek, A. (1998) Cultural control. In Cereal Stem Borers in 1044 Africa: Economic importance, Taxonomy, Natural enemies and Control. International 1045 Institute of Entomology, CAB International: Wallingford. Pp 333-347. 1046 Van den Berg, J. & Nur, A.F. (1998) Chemical control, African cereal stemborers, economic 1047 importance, taxonomy, natural enemies and control. CAB International. Wallingford. Pp 1048 319- 332. 1049 Vannier, G. (1994) The thermobiological limits of some freezing tolerant insects: the 1050 supercooling and thermo-stupor points. Acta Oecologica, 15, 31-42. 1051 Yamamura, K. & Kiritani. K. (1998) A simple method to estimate the potential increase in the 1052 number of generations under global warming in temperate zones. Applied Entomology and 1053 Zoology, 33, 289-298. 1054 Yang, J., Sun, Y.Y. An, H. & Ji, X. (2008) Northern grass lizards (Takydromus septentrionalis) 1055 from different populations do not differ in thermal preference and thermal tolerance when 1056 acclimated under identical thermal conditions. Journal of Comparative Physiology: 1057 Biochemical, Systemic, and Environmental Physiology, 178, 343-349.

1058 Walther, G.R., Post, E., Convey, P., Menzel, A., Parmesan, C., Beebee, T.J.C., Fromentin, J.M., 1059 Hoegh-Guldberg, O. & Bairlein, F. (2002) Ecological responses to recent climate change. 1060 Nature, 416, 389 – 395. 1061 Wang, J.C., Zhang, B., Li, H.G., Wang, J.P. & Zheng, C.Y. (2014) Effects of exposure to high 1062 temperature on Frankliniella occidentalis (Thysanoptera: Thripidae), under Arrhenotoky and 1063 sexual reproduction conditions. Florida Entomologist, 97(2), 504-510. 1064 Ward, N.L. & Masters, G.J. (2007) Linking climate change and species invasion: an illustration 1065 using insect herbivores. Global Change Biology, 13, 1605 – 1615. 1066 Weldon, C.W., Terblanche, J.S. & Chown, S.L. (2011) Time-course for attainment and reversal 1067 of acclimation to constant temperature in two Ceratitis species, Journal of Thermal Biology, 1068 36, 479-485. 1069 Wharton, D.A. (2002) Life at the limits: Organisms in Extreme Environments. Cambridge 1070 University Press. Cambridge. UK. 1071 Whitman, D.W. & Ananthakrishnan, T.N. (2009) Phenotypic Plasticity of Insects: Mechanisms 1072 and Consequences. Science Publishers. Enfield, NH, USA. 1073 Wiedenmann, R.N., Smith, J.W. & Darnell, P.O. (1992) Laboratory rearing and biology of the 1074 parasite Cotesia flavipes (Hymenoptera: Braconidae) using Diatraea saccharalis 1075 (Lepidoptera: Pyralidae) as a host. Environmental Entomology, 21, 1160-1167. 1076 Wilson, R. S. & Franklin, C.E. (2002) Testing the beneficial acclimation hypothesis. Trends in 1077 Ecology & Evolution, 17, 66–70. 1078 Worner, S.P. (1992) Performance of phenological models under variable temperatures regimes: 1079 Consequences of the Kaufmann or rate summation effect. Environmental Entomology, 21, 1080 689-699. 1081

1082

1083

1084

1085

1086

1087

1088

1089 CHAPTER 2

1090

1091

1092

1093

1094

1095

1096 Dominance of spotted stemborer, Chilo partellus Swinhoe (Lepidoptera: Crambidae) over 1097 indigenous stemborer species in Africa’s changing climates: ecological and thermal biology 1098 perspectives*

1099 1100 1101 1102 1103 1104 1105 1106 * Published as “Mutamiswa, R., Chidawanyika, F. & Nyamukondiwa, C. (2017) Dominance of 1107 spotted stemborer, Chilo partellus Swinhoe (Lepidoptera: Crambidae) over indigenous stemborer 1108 species in Africa’s changing climates: ecological and thermal biology perspectives. Agricultural 1109 and Forest Entomology, 19, 344-356. DOI: 10.1111/afe.12217

1110 2.1 Introduction 1111 The evidence of anthropogenic global climate change is increasingly becoming apparent 1112 (IPCC, 2014). Climatic changes such as shifts in rainfall and seasonal thermal patterns are 1113 already being witnessed through extreme weather events such as droughts and floods, heat waves 1114 and cold snaps at increased magnitude and frequency (IPCC, 2014). With global mean surface 1115 temperature predicted to increase by 1.4 to 5.8°C by 2100 (Karuppaiah & Sujayanad, 2012), 1116 much concern has been duly placed on the potential impact of such changes on biodiversity 1117 including the potential invasion pathways of invasive fauna and flora (Sutherst et al., 2007; Ward 1118 & Masters, 2007). For example, several works have shown that different plant and animal 1119 species are vulnerable to climate change (Parmesan & Yohe, 2003; Thomas et al., 2004; Hance 1120 et al., 2007). 1121 In Southern Africa, agriculture is very important for food security and livelihoods, yet 1122 vulnerability to climate risks is considered very high because of heavy reliance on natural 1123 resources, particularly in rain-fed traditional agricultural systems (Kutywayo et al., 2013). For 1124 crop production systems, the threat of undesirable biota such as weeds, insect pests and disease 1125 pathogens is projected to increase under climate change scenarios (Juroszek & von Tiedmann, 1126 2013). For example, it has been postulated that rising temperatures may increase geographical 1127 ranges of insects (Parmesan et al., 1999) resulting in exotic insects, including destructive crop 1128 pests realizing new hosts among indigenous plants and crops. In addition, temperature changes 1129 may alter the phenology of indigenous pests resulting in decreased duration of over-wintering 1130 period and rapid population growth, thereby increasing the pest pressure on the production 1131 system (Jaramillo et al., 2011). The potential reduction in diversity of beneficial insects 1132 including natural enemies of pests and crop pollinators coupled with phenological asynchrony 1133 between insect pests and their natural enemies will further compound problems already 1134 bedeviling agriculture (Mwalusepo et al., 2015). The minimization of crop losses and reduction 1135 of pesticide use to meet regulatory environmental obligations are two key areas in agriculture 1136 which will become more challenging (Chidawanyika et al., 2012). The magnitude of climate 1137 change is expected to be greater in the tropics, whose countries are mostly developing and hence 1138 have low capacity for mitigation and adaptive measures (IPCC, 2014; Mwalusepo et al., 2015). 1139 The understanding of potential impacts of climate change on this pivotal sector is therefore 1140 important towards building resilience.

1141 Cereal crops, in particular maize (Zea mays L.) and sorghum (Sorghum bicolor L. 1142 Moench), are strategic crops in achieving food security. Despite significant strides in maize 1143 varietal development to suit diverse conditions, many problems still impede optimal production. 1144 Among them, a complex of lepidopteran stemborer insect pests damage the crop at different 1145 developmental stages (Addo-Bediako & Thanguane, 2012) with losses ranging between 5% and 1146 73% of potential crop yield (De Groote, 2002; Mailafiya et al., 2009). The major cereal 1147 stemborers of economic importance present in Africa include spotted stemborer (Chilo partellus 1148 Swinhoe), African maize stemborer (Busseola fusca) (Fuller), pink stemborer (Sesamia 1149 calamistis Hampson), sugarcane stemborer (Eldana saccharina Walker) and coastal stemborer 1150 (Chilo orichalcociliellus) (Strand) (Kfir et al., 2002; Addo-Bediako & Thanguane, 2012). 1151 Distribution of these stemborers species in Africa varies spatially (Figure 2.1A-D), with most of 1152 them occurring as a complex of species with overlapping spatial and temporal distributions 1153 (Chabi-Olaye et al., 2001). 1154 Several questions using diverse approaches have aptly been asked in an attempt to 1155 understand future insect pest-maize interactions under climate change scenarios (e.g. Aragòn et 1156 al., 2010; Mwalusepo et al., 2015). Bioclimatic models based on species distribution records and 1157 phenology data have successfully explained current distribution of the stemborers with their 1158 predictive power justifiably accepted with high degree of confidence (see e.g. Mwalusepo et al., 1159 2015). These correlative habitat models describe the relationship between the distributions of a 1160 species across the landscape with physical or biotic factors (Herborg et al., 2007). These models 1161 have therefore been useful in identifying potential hotspots for biodiversity in general (Herborg 1162 et al., 2007; Mika et al., 2008) and specifically maize pests distribution (Aragòn et al., 2010; 1163 Mwalusepo et al., 2015). The distribution may be as a result of not only the abiotic factors, but 1164 also the consequence of other historical, unique and contingent factors such as certain dispersal 1165 constraints, biotic interactions, anthropogenic effects and stochastic events (Lo´ pez-Darias et al., 1166 2008; Aragòn et al., 2010). 1167 1168 1169 1170 1171

1172 1173 A 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 B 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205

1206 1207 C 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 D 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 Figure 2.1: Global distribution of cereal stemborers (A) Busseola fusca (B) Sesamia calamistis 1236 (C) Chilo partellus (D) Eldana saccharina (redrawn from Plantwise Technical Factsheet 2013; 1237 Yonow et al., 2016). Colours indicate presence of stemborer species. 1238

1239 Insects are ectotherms by nature and therefore their body temperature is always in 1240 equilibrium with environmental temperature (Chown & Nicolson, 2004). Temperature thus plays 1241 a key role in defining survival ranges together with performance of biological activities such as 1242 locomotion (Nyamukondiwa et al., 2010), reproduction, feeding and developmental rates 1243 (Nyamukondiwa et al., 2013). As such, insects have evolved a diverse range of morphological, 1244 physiological, and behavioral adaptations to counter adverse climate conditions (Chown & 1245 Terblanche, 2007; Nyamukondiwa et al., 2013). Several studies have already demonstrated how 1246 insect distribution, in particular Lepidoptera, can be influenced by temperature (Parmesan et al., 1247 1999; Thomson et al., 2010). Temperature effects on life history traits have been investigated in 1248 C. partellus and B. fusca in Africa (Khadioli et al., 2014a; Mwalusepo et al., 2015; Glatz et al., 1249 2016). Other studies have also looked at the impact of drought on the performance (crop damage, 1250 fecundity, development) of maize pests or their natural enemies (Chidawanyika et al., 2014). 1251 While lethal/sub-lethal temperature effects largely affect insect phenology, several physiological 1252 mechanisms have been reported to be key for survival and preservation of crucial life history 1253 traits during typical thermal stress. For example, rapid heat hardening through accumulation of 1254 heat shock proteins has been shown to improve survival in otherwise lethal temperatures (Chown 1255 & Nicolson, 2004; Nyamukondiwa et al., 2010). 1256 Dominance is the degree of ecological influence of a particular species within a 1257 community (Freedman, 1995). It can be manifested in the form of individual numerical 1258 advantage, a greater degree of damage (for pest insects), and higher growth and reproductive 1259 rates relative to competitors. Dominance can lead to success of a species and coupled with 1260 ecological plasticity, these may be the most demographic traits related to invasion and 1261 dominance success of some species. In addition, climate matching between the new and native 1262 environment of the invasive species coupled with the absence of natural enemies are the 1263 ecological characteristics aiding invasion success (Farji-Brener & Coley, 1998). Lepidopterans 1264 are known to respond to climate change through changes in phenology i.e. the timing of annually 1265 recurring events (e.g. emergence from overwintering, voltinism) and also changes in 1266 geographical distribution, such as altitudinal range expansion (Parmesan, 2007). The distribution 1267 and abundance of the stemborers may be influenced by altitude and environmental conditions 1268 especially temperature, precipitation, relative humidity (Getu et al., 2002) and biotic factors such 1269 as natural enemies and alternative host plants (Addo-Bediako & Thanguane, 2012). Chilo

1270 partellus, which is indigenous to Asia was first recorded in Malawi around 1930 (Tams, 1932). 1271 In 1953 it was recorded in Tanzania (Duerdon, 1953) and in South Africa in 1958 (Van 1272 Hamburg, 1979). It later spread to most of the lowland areas of the countries in eastern and 1273 southern Africa dominating the indigenous stemborer species (Kfir, 1997). Recent studies have 1274 reported C. partellus in cooler and higher altitude areas, conversely to its known optimal 1275 ecological conditions (Kfir, 1997; Ong’amo et al., 2006; Mwalusepo et al., 2015) indicating an 1276 expansion of its geographic ranges. Indeed, C. partellus has been observed to displace 1277 indigenous stemborers in some of the locations it has invaded (Kfir, 1997; Ofomata et al., 1278 1999a). The reasons behind C. partellus competitive advantage and hence dominance are largely 1279 unexplained. In addition, it is largely unknown how temperature tolerance, its interactions with 1280 relative humidity and plasticity thereof are likely to shape C. partellus and its larval parasitoid 1281 Cotesia flavipes, especially in the face of climate change. 1282 This paper therefore reviewed the factors that may contribute to the dominance of C. 1283 partellus over indigenous stemborer species in Africa, with emphasis on thermal physiology 1284 under changing African climates. I hypothesized that, C. partellus may possess a high basal 1285 thermal tolerance and plasticity of temperature tolerance, allowing it to invade and thrive in new 1286 environments. 1287

1288 2.2 Africa’s Changing Climate 1289 The climatic zones in Africa are shaped by sea surface temperature changes, land use and 1290 vegetation patterns, which play significant roles in both intra-seasonal variability and onset of 1291 rain season (Stathers et al., 2013). The average temperatures in Africa are projected to increase 1292 by 1.5 to 3°C, by 2050 and will gradually continue upwards beyond this time (IPCC, 2014). 1293 Climate change in Africa will especially be associated with a rise in temperatures and its 1294 variability thereof, erratic rainfall patterns, increased extreme weather events (flooding, 1295 droughts, heat waves and cold snaps) (Stathers et al., 2013; IPCC, 2014). Temperatures are 1296 projected to rise by about 1˚C in East, Central, Southern and West Africa and slightly less in the 1297 Sahel by 2030 (Lobell et al., 2008; Stathers et al., 2013). Furthermore, for the southern African 1298 region during the period 2036 – 2065, small areas along western and central Zimbabwe, eastern 1299 Zambia and north-eastern Mozambique, might experience a decline in rainfall. By contrast, 1300 western and northern Angola, northern Mozambique, and eastern South Africa (including

1301 Lesotho) are simulated to experience higher annual rainfall totals than at present (IPCC, 2014). 1302 During the same period, an increase in minimum and maximum temperature of between 0.8 and 1303 3.6˚C per annum is expected across the region with the interior regions expected to experience 1304 more intense warming than the coastal areas. Spring is expected to experience the greatest 1305 increase in temperature compared to the other seasons (IPCC, 2014). The fourth 1306 Intergovernmental Panel on Climate Change (IPCC) assessment suggests that warming of the 1307 African continent (~3.4°C by 2080-2099) is most likely to be higher than the global mean annual 1308 warming all over the continent and in all seasons, with drier subtropical regions warming more 1309 than the humid tropics (Wingqvist & Dahlberg, 2008; Stathers et al., 2013). Therefore, during 1310 the same period southern Africa is likely to experience a 10% rainfall decrease and more 1311 frequent droughts, while east Africa is likely to experience increased rainfall (Lobell et al., 2008; 1312 Stathers et al., 2013 but see Williams & Funk, 2011; Mwalusepo et al., 2015). In recent decades, 1313 eastern Africa has experienced an increased rainfall over the northern sector and declining 1314 amounts over the southern sector (Schreck & Semazzi, 2004; Stathers et al., 2013). It is 1315 predicted that in equatorial and subtropical eastern Africa, warm sea surface temperatures will 1316 lead to increased droughts as the number of rain events during the dry season will decrease as 1317 well as overall annual precipitation (Masters & Norgrove, 2010). Such changes in precipitation 1318 are likely to reduce the annual flow of rivers. For African agroecosystems, shifts in climate are 1319 significant as they generate novel structures in original insect pest abundance, emergence of 1320 secondary to primary pests (Newton et al., 2011), colonization of novel habitats previously 1321 unfavorable (Parmesan et al. 1999) and, importantly, modifications of habitats (Webb et al., 1322 2005), which may compromise the maintenance of biodiversity and efficacy of biological control 1323 through natural enemies (Chidawanyika et al., 2012). By 2055, such complex changes in climate 1324 are predicted to treble cereal yield losses, especially associated with stemborers. And these losses 1325 will be more pronounced in higher than lower altitudes (Mwalusepo et al., 2015). 1326

1327 2.3 Competitive Displacement of Indigenous Stemborer Species 1328 Competitive displacement is the removal of a formerly established species from a habitat 1329 through superior use, acquisition, or defense of resources by another species (Reitz & Trumble, 1330 2002). Most displacements that have been identified were triggered by the introduction or 1331 invasion of an exotic species although other environmental factors may predispose a species to

1332 being displaced (Reitz & Trumble, 2002). Temperature has been reported as one of the factors 1333 influencing intra and interspecific interactions as well as limiting the geographical distribution of 1334 insects. For instance the interaction between Nicrophorus orbicollis Say and N. defodiens 1335 Mannerheim (Coleoptera: Silphidae) when feeding on the same carcass has been reported to be 1336 dependent on temperature (Wilson et al., 1984). The displacement of one insect species by 1337 another on crops of economic importance is of great significance to pest management (Kfir, 1338 1997). Intra and inter-specific competition among exotic and native stemborer species can 1339 influence their performance as well as spatio-temporal population dynamics. As for C. partellus, 1340 crawling and ballooning were reported as the mechanisms employed by larvae in dispersing 1341 (Berger, 1992). The process of ballooning includes the spinning of thin silk-like threads which 1342 are launched by wind into the air to infest neighboring plants (Sithole, 1990; Berger, 1992). This 1343 mechanism serves to reduce intraspecific competition amongst larvae that hatch from the same 1344 egg batch thereby increasing survival through resource use optimization (Chapman et al., 1983; 1345 Berger, 1992). However, this survival mechanism is not very common in indigenous stemborer 1346 species and may confer a relative survival advantage to C. partellus. Chilo partellus has a high 1347 colonization potential and this may facilitate its invasiveness and interspecific competition 1348 success. In an experiment involving C. partellus, B. fusca and S. calamistis, interspecific 1349 competition was more apparent than intraspecific competition with C. partellus being highly 1350 competitive in resource utilization indicating its high fitness potential as compared to S. 1351 calamistis and B. fusca (Ntriri et al., 2016). Therefore, in favorable environments where C. 1352 partellus coexists with either of these noctuid species, the crambid will have high survival 1353 advantage (Ntiri et al., 2016). In most of the low-lying (lowveld) areas of eastern and southern 1354 Africa, C. partellus has become the most significant and dominant borer species (Kfir, 1997; 1355 Mwalusepo et al., 2015). Similarly, in other African landscapes, C. partellus has managed to 1356 successfully displace other indigenous stemborer species (Kfir, 1997; Kfir et al., 2002; Yonow et 1357 al., 2016). For example, C. partellus was first detected in Madagascar in 1972, and three years 1358 thereafter became an economically significant insect pest of maize and sorghum compared to the 1359 native stemborer Chilo orichalcociliellus (Delobel, 1975). The displacement of C. 1360 orichalcociliellus by C. partellus was reported in Kenya (Ofomata et al., 1999a; Dejen et al., 1361 2014), with C. partellus abundance increasing from 28.5% to 87% while C. orichalcociliellus 1362 reduced from 66% to 14% (Zhou et al., 2001; Jiang et al., 2006). In Kenya’s eastern province, B.

1363 fusca was more abundant than C. partellus around 1980 (Seshu, 1983; Dejen et al., 2014). 1364 However, between 1996 and 1998, B. fusca became less pronounced with C. partellus 1365 dominating (Dejen et al., 2014). Similarly, there was a partial displacement of B. fusca by C. 1366 partellus over a seven year period in the South African eastern highveld region (Kfir, 1997; 1367 Sylvain et al., 2015). The displacement was more pronounced in sorghum where C. partellus 1368 proportion increased 664 fold between 1986 and 1992 (Kfir, 1997; Dejen et al., 2014). Due to its 1369 superior invasion potential to indigenous stemborer species following introduction to novel 1370 habitats, (Kfir, 1997; Ofomata et al., 2000; Niyibigira et al., 2001), C. partellus occupies 1371 appropriate feeding habitats earlier than the native borer species, thereby outcompeting the 1372 population of competitive stemborers that successfully colonize the same habitats (Niyibigira et 1373 al., 2001). Busseola fusca has been reported to avoid ovipositing on host plants with previous C. 1374 partellus infestation (Kfir, 1997). However, C. partellus can oviposit on host plants which are 1375 infested with and/or have previous B. fusca infestation (Kfir, 1997). This may be because C. 1376 partellus females, upon oviposition emit chemical odours that deter B. fusca females from 1377 ovipositing on the same plants or that plant volatiles produced by the host deter B. fusca 1378 oviposition (Kfir, 1997). Moreover, neonate C. partellus larvae can disperse longer distances as 1379 compared to C. orichalcociliellus, and this may enable C. partellus to utilise more host plants 1380 than the relatively less mobile indigenous stemborer (Ofomata, 1997; Overholt et al., 2000). 1381 Limited studies have been conducted in assessing the effects of global temperature increases on 1382 the competitive interactions of insect species utilising the same resource and habitat (Ntiri et al., 1383 2016). Although competitive displacement of indigenous stemborers by C. partellus has been 1384 reported in Africa, mechanisms behind the borer species displacement remain relatively 1385 unknown (Ntiri et al., 2016). 1386

1387 2.3.1 Extension of C. partellus geographical range 1388 As global climatic temperatures increase, threats from invasive tropical and subtropical 1389 insect species may increase as they expand their geographical distribution to temperate regions 1390 (Chown et al., 2007; Weldon et al., 2016). Therefore, the ability of these insect species to endure 1391 dry conditions may contribute to their invasion potential since migration from the moist 1392 subtropical and tropical regions exposes them to low rainfall and increasingly drier environments 1393 (Chown et al. 2007; Weldon et al., 2016). Following its first record in Africa (Malawi) (Tams,

1394 1932), C. partellus has become a widespread pest for cereal crops in eastern and southern 1395 Africa, becoming more destructive than the indigenous stemborer species including C. 1396 orichalcociliellus (Sithole, 1990; Kfir, 1997; Ofomata et al., 1999a; Overholt et al., 2000; 1397 Chidawanyika et al., 2012; Dejen et al., 2014). The interaction between elevation and 1398 temperature was thought to be the most important factor determining the distribution of C. 1399 partellus (Overholt et al., 1997; Tamiru et al., 2007). However, some researchers documented a 1400 combination of climatic factors such as temperature, rainfall and humidity influencing its 1401 distribution, with temperature being the most important (Getu et al., 2002; Tamiru et al. 2007). 1402 Since C. partellus is predominantly found in lowvelds where average temperatures are high, I 1403 therefore hypothesize that, it may possess an enhanced high temperature tolerance and a 1404 compromised low temperature tolerance. The former allows it to thrive in warmer environments 1405 while the latter limits its colonization to new high altitude areas. However, high basal tolerance 1406 may indicate greater plasticity for high temperature tolerance for C. partellus (see 1407 Nyamukondiwa et al., 2011), which may confer a survival advantage for the stemborer under 1408 novel environments. Furthermore, insects also trade-off their basal low temperature tolerance for 1409 plasticity (Nyamukondiwa et al., 2011), a scenario which may mean that C. partellus may also 1410 be able to cope with low temperature variations typical of climate change through phenotypic 1411 plasticity. Since the appearance of C. partellus in African continent, it has continuously 1412 expanded its distribution to nearly all countries in the warm, low-altitude regions of eastern and 1413 southern Africa (Kfir, 1997; Dejen et al., 2014; Mwalusepo et al., 2015). Previous research in 1414 Ethiopia, reported that C. partellus was a predominant species at lower altitude of less than 1700 1415 m a.s.l. and B. fusca was dominant at high altitude of 1160 - 2600 m a.s.l. and in cooler areas 1416 (Assefa, 1985). Until the year 1990, the distribution of C. partellus was limited to low and mid- 1417 altitude areas (<1500 m a.s.l.), and was attributed to favorable abiotic factors, mainly 1418 temperature and humidity (Zhou et al., 2001; Khadioli et al., 2014a). From 1999 to 2000, a 1419 survey in Ethiopia reported that C. partellus had widened its distribution from 500-1700 to 1420 between 500 and 1900 m a.s.l. whereas B. fusca was reported 1030-2320 m a.s.l. (Emana et al., 1421 2001; Dejen et al., 2014). This work revealed the extension of C. partellus into landscapes where 1422 only B. fusca and S. calamistis would ‘naturally inhabit’. Similarly, Dejen et al. (2014) reported 1423 C. partellus extending its distribution from 1400 to 2084 m a.s.l. This therefore shows that C. 1424 partellus distribution may not solely be limited by elevation, but rather a complex of factors

1425 which likely include relative humidity/rainfall, temperature and topographic features (Dejen et 1426 al. 2014; Mwalusepo et al., 2015). In South Africa the only stemborer that used to survive at an 1427 elevation of ≥1600 m a.s.l. was B. fusca. Following C. partellus invasion in the region, it rapidly 1428 increased its share of the total stemborer population every year. Within two years, it became the 1429 predominant stemborer, constituting 90% of the total population (Dejen et al., 2014). In addition, 1430 Tamiru et al. (2007) recorded C. partellus at an elevation of 2088 m a.s.l. in Ethiopia suggesting 1431 gradual expansion of the insect to new ecological niches. Furthermore, in Kenya, Ong’amo et al. 1432 (2006) reported the presence of C. partellus in relatively high altitude zones where the insect is 1433 least expected (Tamiru et al., 2012). Therefore, considering its rate of expansion, C. partellus is 1434 most likely to continue expanding to high altitude areas and occupying new niches. The 1435 establishment of C. partellus in several African regions follows a distinct pattern, which may be 1436 partly influenced by the close interaction across temperature, relative humidity and altitude 1437 (Ong’amo et al., 2006; reviewed in Chidawanyika et al., 2012). This therefore means any 1438 experiments which sought to predict the likely distribution and abundance of C. partellus in 1439 future should incorporate all these parameters as factors likely dictating its population 1440 phenology/dynamics and hence geographic distribution. Similarly, models predicting the likely 1441 effects of climate change on stemborer population dynamics should incorporate all factors of 1442 temperature, humidity and the ability of organisms to compensate for such changes in 1443 environmental conditions through plasticity (within generation) or evolutionary potential (across 1444 generations). 1445

1446 2.3.2 Increase in numbers of C. partellus generations 1447 For multivoltine insects, the number of generations per year is significantly affected by 1448 climatic conditions, chief amongst them, temperature (Bale, 2010). Higher temperatures 1449 associated with climate change result in faster accumulation of degree days hence shorter 1450 generation times, milder winters thus longer growing seasons with probable shifts from 1451 univoltine to bivoltine/multivoltine life cycles in some insect species (Bale, 2010). It has been 1452 estimated that with a 2°C temperature increase, insects might experience one to five additional 1453 life cycles per season (Yamamura & Kiritani, 1998). Model predictions indicate a potential 1454 increase in the number of C. partellus generations in the year 2050, which indicates that higher 1455 insect pest infestations and yield losses may occur (Kroschel et al., 2013). This success may be

1456 attributed to differences in its thermal biology as compared to that of other related stemborers, C. 1457 orichalcociliellus, B. fusca, S. calamistis and E. saccharina. The life cycle of C. partellus is 1458 highly temperature dependent (Ofomata et al., 2000; Mbapila et al., 2002), taking about 3 to 4 1459 weeks at an optimum temperature of 30°C (Kfir, 1997; Khadioli et al., 2014a) (Table 2.1). 1460 However, for some indigenous stemborers, life cycles range from 6 to 12 weeks with E. 1461 saccharina having the longest life cycle of 8-12 weeks at 27±2°C (see Table 2.1). 1462 1463 Table 2.1: A compilation of generation time of four major cereal stemborer life cycles under 1464 optimum environmental conditions. Generation time was measured as the amount of time needed 1465 to complete development of one stemborer life cycle (from egg to adult). 1466 Stemborer Average Optimum Environmental Conditions Reference Species Life Cycle (Weeks) T°C RH(%) (L:H) B. fusca 7-8 25 65± 5 12:12 Harris & Nwanze, 1992; Polaszek 1998; Khadioli et al., 2014b S. calamistis 6-8 28 65± 5 12:12 Khadioli et al., 2014b

E. saccharina 8-12 27±2 75± 5 10:14 Sharma & Nwanze, 1997; Polaszek, 1998

C. partellus 3-4 30 65± 5 12:12 Kfir, 1997; Khadioli et al., 2014a

1467 T°C: Temperature; RH: Relative Humidity; L:H: Photoperiod 1468 1469 Due to its short life cycle, C. partellus may have more generations per year as well as rapid 1470 population growth, suggesting its likely dominance over other stemborer species. Furthermore, 1471 C. partellus’ short generation time may confer the species a superior competitive advantage over 1472 its competitors, through its higher potential rate of increase (Kfir, 1997; Mbapila et al., 2002; 1473 Dejen et al., 2014) and rapid development (Reitz & Trumble, 2002) which may possibly give it a 1474 numerical advantage (Sujay et al., 2010). Up to six generations of C. partellus can occur per year

1475 (Panwar, 1995) and this may likely be higher given the increase in mean temperatures with 1476 global change. In addition, it is known that when equal numbers of C. partellus and C. 1477 orichalcociliellus infest the same maize, sorghum, or wild-sorghum host plant, C. partellus can 1478 successfully complete development, suggesting superiority during interspecific competition 1479 (Ofomata, 1997). 1480 The average fecundity of stemborers varies in relation to species (see Table 2.2). Chilo 1481 partellus has the highest fecundity per unit time compared to the other indigenous stemborer 1482 species. In an experiment involving C. partellus and C. orichalcociliellus, a comparison of life 1483 traits revealed that the invasive C. partellus laid more viable eggs, with its larvae consuming 1484 more food and had a higher survival and shorter developmental time as compared to C. 1485 orichalcociliellus (Ofomata et al., 2000; Ntiri et al., 2016). Therefore, with its short life cycle 1486 coupled with high fecundity, C. partellus may have a numerical advantage thereby dominating 1487 other stemborer species which are associated with longer life cycles and low fecundity. 1488 Similarly, with climate change coupled with shorter generation times, Khadioli et al. (2014a) 1489 predicted an increase in the number of C. partellus generations in most countries in east and 1490 southern Africa suggesting that economic losses might increase and also encroach to new agro- 1491 ecosystems/habitats. Hence I suggest both intrinsic factors (e.g. rapid generation times, 1492 polyphagy and interspecific competitiveness), and extrinsic factors (e.g. increased reproductive 1493 output), likely caused by high fecundity may be contributory factors to C. partellus dominance 1494 when introduced into novel environments. This may explain why it has actively displaced 1495 indigenous stemborers upon introduction to new habitats. Nevertheless, in addition to 1496 temperature, many other interacting ecological factors need to be considered. For example, 1497 selection of a good host plant permitted multivoltism while selection of poorer quality hosts 1498 enhanced ‘escape’ from natural enemies in Papilio spp. larvae (Scriber, 2002). 1499 1500 1501 1502 1503

1504 Table 2.2: Average fecundity of different species of cereal stemborers (measured as the average 1505 total number of eggs produced per life time) under optimum environmental conditions of 1506 temperature, humidity and photoperiod. 1507 Stemborer Average Duration Optimum Reference species Fecundity (Days) Environmental Conditions

T°C RH L:H (%)

S. calamistis 400 5 28 65± 5 12:12 Sharma & Nwanze, 1997; Khadioli et al., 2014

B. fusca 400-500 5-6 26±1 70± 12:12/ Sharma & Nwanze 1997; Khadioli et 10 14:10 al., 2014; Glatz et al., 2016 E. 400-600 14 27±2 75± 5 10:14 Sharma & Nwanze, 1997 saccharina C. partellus 500-722 4 30 65± 5 12:12 Taneja & Nwanze, 1990; Sharma & Nwanze, 1997; Khadioli et al., 2014

1508 ToC : Temperature; RH: Relative Humidity; L:H: Photoperiod 1509

1510 2.3.3 Overwintering of C. partellus 1511 Diapause is a physiological state of arrested growth and development neurohormonally 1512 mediated and frequently elicited by environmental cues (Hance et al., 2007). It is important in 1513 facilitating an insect to survive unfavorable environmental conditions which include, amongst 1514 others, extremes of temperature, unavailability of food as well as drought. Initiation of diapause, 1515 maintenance and termination are highly characterized by an array of morphological, 1516 physiological, biochemical and behavioral features and the process synchronises insect 1517 phenologies with seasonal changes in its environment, especially temperature. Towards the end 1518 of the cropping season (coincident with winter in southern Africa or the dry season in eastern 1519 Africa), most of the cereal stemborers undergo a resting phase in which they spend in crop 1520 residues (Kfir et al., 2002). In laboratory experiments (26°C; 16:8h daylength), Kfir (1993)

1521 recorded B. fusca mean diapause termination time of 34 days (Table 2.3). In addition, Ofomata et 1522 al. (1999) showed that mean diapause termination time of C. partellus and C. orichalcociliellus 1523 were 9.7 and 14.4 days respectively at 28°C and 12:12 (L:D) photoperiod (Table 2.3). 1524 1525 Table 2.3: Mean diapause termination time for cereal stemborers. Termination time was defined 1526 as the amount of time needed to restore activity of a ‘sleeping’ stemborer following optimum 1527 temperature conditions. Stemborer Temperature Photoperiod Mean Diapause Reference Species (°C) (L:D) Termination Time (days) Chilo partellus 28±0.5°C 12:12 9.7 Ofomata et al., 1999b Busseola fusca 26°C 16:8 34 Kfir, 1993 Chilo 28±0.5°C 12:12 14.4 Ofomata et al., 1999b orichalcociliellus 1528

1529 Therefore, C. partellus terminates diapause earlier than C. orichalcociliellus and B. fusca thus it 1530 completes its life cycle earlier and utilizes habitat resources faster hence optimising growth, 1531 increasing propagules hence dominance over other species. Busseola fusca, S. calamistis, and C. 1532 orichalcociliellus are known to obligatorily diapause for a number of months during the dry 1533 season (Kfir, 1991; Ofomata et al., 1999b; Kfir et al., 2002). However, C. partellus only 1534 diapause facultatively under unfavorable climatic conditions (Ofomata et al., 1999b). It 1535 overwinters in the lower and seldom upper parts of dry cereal stems or below the ground where it 1536 is protected from environmental extremes and natural enemies (Kfir, 1991). Availability of 1537 favorable conditions like humid conditions are known to induce a quick reduction in the juvenile 1538 hormone of diapausing larvae and this initiates diapause termination. Overwintering C. partellus 1539 populations terminate diapause emerging as moths about one month earlier in the season than 1540 other indigenous stemborer species (Kfir, 1997; Ofomata et al., 1999b). Chilo partellus 1541 facultative diapause implies that it may survive predictable, unfavorable environmental 1542 conditions, such as temperature extremes, drought or reduced food availability consequently 1543 having a survival advantage over relatively obligatory diapausing indigenous stemborers. 1544

1545 2.3.4 Impact of Variability in Temperature and Relative Humidity on the Ecology of 1546 Stemborers 1547 Temperature is considered as the most important environmental factor that directly 1548 influences insect development, survival, reproduction and distribution (Bale, 2010; Khadioli et 1549 al., 2014b). In a temperature related experiment, lower temperatures conferred a competitive 1550 advantage on smaller males of seed beetle Stator limbatus (Horn) (Coleoptera: Chrysomelidae), 1551 which out-competed larger ones in reaching a potential mate (Moya-Laraño et al., 2007). 1552 Individual insect populations have specific optimum temperatures at which they develop 1553 optimally, and lower and upper threshold temperatures beyond which they cannot develop 1554 (Khadioli et al., 2014b). Khadioli et al. (2014b) reported that B. fusca and S. calamistis 1555 populations develop within the temperature range of 15°C–30°C with an optimum at 25 and 28°C 1556 respectively. However, C. partellus populations develop within the range of 18 - 35°C (Khadioli 1557 et al., 2014a). This translates into a greater thermal window (across all developmental stages) for 1558 C. partellus activity relative to the other two species B. fusca and S. calamistis (see Table 2.4). 1559 1560

1561 Table 2.4: Lower and upper developmental temperature thresholds, developmental temperature 1562 range and mean developmental time at different temperatures for three major stemborers, (Chilo 1563 partellus, Sesamia calamistis and Busseola fusca) (modified from Khadioli et al., 2014a;b; Glatz 1564 et al., 2016) 1565 Development Lower Upper Thermal Mean Development Time (Days) Stage Development Development activity

Temperature Temperature window (°C) (°C) (°C) 15°C 25°C 30°C 35°C

C. partellus

Eggs 18 35 17 - 6.4 4.9 4.8

Larvae 18 35 17 - 32.8 21.9 28.3

Pupae 18 35 17 - 9.3 7.0 7.7

Adults 18 35 17 - 10.05 7.6 5.5

B. fusca

Eggs 15 30 15 22.47 7.2 6.54 -

Larvae 15 30 15 90.2 50.69 58.12 -

Pupae 15 30 15 40.0 14.67 16.6 -

Adults 15 30 15 12.02 9.33 4.57 -

S. calamistis 15

Eggs 15 30 15 19.7 7.09 6.74 -

Larvae 15 30 15 130.41 46.52 35.46 -

Pupae 15 30 15 35.19 11.2 9.75 -

Adults 15 30 15 8.4 8.89 7.07 -

1566

1567 Greater thermal window optimizes key behaviors/processes such as development, mating, 1568 oviposition and dispersal during environmental conditions where the same traits are limited for S. 1569 calamistis and B. fusca. This confers a survival and comparative advantage of C. partellus on 1570 key behavioral traits relative to other indigenous stemborer species. Previous studies reported 1571 mean development time of ~48 days (25°C) for C. partellus reared on artificial diet (Mbapila et 1572 al., 2002; also see Khadioli et al., 2014a). However, the total developmental time of B. fusca at 1573 25°C and 26-30°C were 72.56 and 52.6 days respectively (Khadioli et al., 2014b; Glatz et al., 1574 2016) and 64.81 days for S. calamistis at 25°C (see table 2.2) (Khadioli et al., 2014b). This 1575 confers developmental advantage to C. partellus relative to B. fusca and S. calamistis. Moreover, 1576 basing on developmental temperatures (see table 2.4), C. partellus developmental stages can 1577 survive higher basal temperatures as compared to B. fusca and S. calamistis. However, its 1578 developmental stages cannot sustain low temperatures as compared to those of B. fusca and S. 1579 calamistis (Ntiri et al., 2016). With global temperatures expected to continue increasing, C. 1580 partellus is most likely to have a numerical advantage over other species ultimately out 1581 competing them. High basal high temperature tolerance and low basal low temperature may 1582 mean C. partellus can survive both thermal extremes through plasticity, since insects tend to 1583 trade-off basal low but not high temperature tolerance for plasticity (see discussions in 1584 Nyamukondiwa et al., 2011). Hence C. partellus may have a thermal tolerance advantage and 1585 may likely cope with the imminent climate change effects, relative to other stemborer species. 1586 Under future high temperature scenarios, C. partellus may expand further from most dry lowland 1587 areas to higher elevated areas in east and south eastern African countries (Kroschel et al., 2013) 1588 since temperatures are gradually increasing extending into high veld areas. It is most likely to 1589 expand to higher elevation in eastern Africa (Burundi, Ethiopia, Kenya, Uganda, Tanzania, 1590 Rwanda), and to mid elevation in southern Africa (Mozambique, Zambia, Zimbabwe, South 1591 Africa) (Khadioli et al., 2014a). This may likely increase the level of damage to cereal crops in 1592 these high elevation regions, given that C. partellus causes more injury than B. fusca on maize in 1593 some regions (Ong’amo et al., 2006). With southern Africa expected to experience a 10% 1594 rainfall decrease and more frequent droughts and east Africa expected to receive high rainfalls 1595 (Lobell et al., 2008; Stathers et al., 2013), C. partellus may probably enter long periods of 1596 diapause in southern Africa thereby surviving harsh climatic conditions. Nevertheless, under 1597 favorable conditions, C. partellus is known to terminate diapause quicker than the identified

1598 indigenous species (Kfir, 1997; Ofomata et al., 1999b; Overholt et al., 2000). Laboratory 1599 experiments examining the displacement of stemborer C. orichalcociliellus showed that C. 1600 partellus had significantly higher fecundity compared to indigenous C. orichalcociliellus 1601 (Ofomata et al., 2000; Tamiru et al., 2012) hence may successfully outcompete the indigenous 1602 stemborer. I hypothesize that the expected global change, especially temperature increase and its 1603 variability may directly influence C. partellus geographical distribution range as well as pest 1604 status. Therefore, there is a potential of C. partellus invading and establishing in several new 1605 habitats through its dominance over other stemborer species (Kfir, 1997). 1606

1607 2.3.5 Physiological Tolerances in C. partellus 1608 Insects generally encounter different stressful environmental extremes and usually 1609 survive through morphological and physiological changes that are mediated by a range of 1610 behavioral, hormonal, cellular, and biochemical mechanisms (Chown & Nicolson, 2004). Some 1611 insect pests may increase their invasion success in relation to their ability to deal with changing 1612 climates either through phenotypic plasticity or variation in basal tolerance (Nyamukondiwa et 1613 al., 2010; 2011; Chidawanyika et al., 2012), behavioral adaptation (over short time scales) e.g. 1614 habitat selection, basking intensity, restriction of activity periods and selective exploitation of 1615 environmental thermal fluxes (Yang et al., 2008), diapause and evolution in the longer 1616 timescales (reviewed by Bale, 2010). I hypothesize that, the invasion success of C. partellus may 1617 also stem from physiological adaptations that enable it to have competitive advantage over other 1618 indigenous stemborer species and survive in different habitats. The major physiological factors 1619 contributing to its invasion success and hence dominance may typically comprise either 1620 increased environmental tolerance and/or increased ability to compensate the short and long 1621 timescales through plasticity, especially to temperature and humidity. Following introduction to 1622 a new habitat, C. partellus may be able to thrive in the short term by having higher basal or 1623 inherent resistance to prevailing climate, or by mounting short term responses to these extremes 1624 thereby avoiding potentially lethal effects. Though not yet reported for C. partellus, this form of 1625 short-term phenotypic plasticity may be one important mechanism enhancing survival of this 1626 stemborer upon introduction to new habitats (reviewed in Whitman & Ananthakrishnan, 2009). 1627 Indeed, acclimation, a form of phenotypic plasticity, has been reported to be a crucial 1628 determinant of field fitness in insects (Kristensen et al., 2008) and may enhance survival of

1629 ‘alien’ species upon introduction to novel thermal habitats (Chown et al., 2007). To my 1630 knowledge, such physiological mechanisms have not yet been reported for C. partellus but I 1631 strongly hypothesise it is unlikely the species will invade, establish and be dominant without 1632 coping to environmental stressors through plasticity. Following establishment, over the long 1633 time-scales, rapid evolutionary adaptation, favouring the selection of novel phenotypes may also 1634 aid in the quick establishment and hence dominance of C. partellus relative to other stemborer 1635 species (Lee et al., 2007), unless genetic constraints play a significant role (Gilchrist & Lee 1636 2007). 1637 To determine whether C. partellus invasion success over other indigenous stemborer 1638 species is aided by basal and plasticity of thermal tolerance, I conducted standard rapid 1639 laboratory experiments (see Nyamukondiwa et al., 2010; Weldon et al., 2011). The initial 1640 colonies of C. partellus, B. fusca and S. calamistis were obtained from International Centre for 1641 Insect Physiology and Ecology (ICIPE), Kenya. The colonies have been in the laboratory for >20 1642 generations with regular augmentation with wild populations to increase genetic diversity. These 1643 stemborers were optimally reared on artificial diet in trays for developing larvae following 1644 standardized protocols (Ochieng et al., 1985; Onyango & Ochieng’-Odero, 1994; Tefera et al., 1645 2010). using 12L:12D photoperiod, 25 ± 1°C and 75 ± 5% relative humidity (RH) for B. fusca 1646 and 12L:12D, 28 ± 1°C and 75 ± 5% RH for C. partellus and S. calamistis in Memmert climate 1647 chambers (Memmert HPP 260, Memmert GmbH + Co.KG, Germany). Critical thermal limits

1648 (CTLs), that is CTmin and CTmax were measured for adult stemborers (B. fusca, S. calamistis and 1649 C. partellus) following rearing at optimum temperatures to test for basal physiological 1650 temperature tolerance (see materials and methods). In brief, critical thermal limits started at ° 1651 optimum temperature, ramping up (for CTmax) and down (for CTmin) at 0.25 C per minute until 1652 endpoint. The endpoint or CTLs were regarded as the temperature where insects lost ability to 1653 respond to gentle stimuli like mild prodding. Following this, I compared the plasticity of thermal 1654 tolerance across the three species using a species specific methodological approach (see e.g. 1655 Nyamukondiwa et al., 2011) (see materials and methods). 1656 1657

1658 Materials and Methods 1659 Study Animals and Rearing Conditions 1660 Stemborers used in these experiments were obtained from the International Centre for 1661 Insect Physiology and Ecology (ICIPE), Kenya as pupae. The insects had been in culture for 1662 >150 generations with regular supplementation with wild caught stemborers to reduce inbreeding 1663 depression. Upon arrival in Palapye, Botswana, stemborers were reared on artificial diet (see 1664 Ochieng et al., 1985; Onyango & Ochieng’-Odero, 1994; Tefera et al., 2010) in Bugdorm cages 1665 (32.5cm3) (Bugdorm-BD43030F, Megaview Science Co., Ltd, Taiwan), placed in Memmert 1666 climate chambers (Memmert HPP 260, Memmert GmbH + Co.KG, Germany), at 25°C; 75% RH; 1667 12L:12D for B. fusca and 28°C; 75% RH; 12L:12D photoperiod for C. partellus and S. 1668 calamistis. Upon eclosion, only ‘lively’ adults of mixed sex were used for thermal biology 1669 experiments, since sex seem to play no significant effect in thermal tolerance (see 1670 Nyamukondiwa & Terblanche, 2009). 1671 1672 Critical Thermal Limits (CTLs) 1673 To determine temperature tolerance in C. partellus, S. calamistis and B. fusca, critical 1674 thermal limits were assessed using species specific methodological approach as outlined by 1675 Nyamukondiwa & Terblanche (2009). Adults (n =10) of C. partellus, S. calamistis and B. fusca 1676 were individually placed in insulated double-jacketed series of chambers (‘organ pipes’) 1677 connected to a programmable water bath (Lauda Eco Gold, Lauda DR.R. Wobser GMBH and 1678 Co. KG, Germany) filled with 1:1 water: propylene glycol to allow for subzero temperatures 1679 (Chown & Nicholson, 2004). A thermocouple (type K 36 SWG) connected to a digital 1680 thermometer (53/54IIB, Fluke Cooperation, USA) was inserted into the control chamber to

1681 record chamber temperature. Both critical thermal maximum (CTmax) and critical thermal ° ° 1682 minimum (CTmin) experiments started at a set point temperature of 25 C for B. fusca and 28 C for 1683 C. partellus and S. calamistis. Once placed in the ‘organ pipes’, organisms were given 10 1684 minutes to equilibrate at benign temperature before ramping up or down at a rate of 0.25°C/min

1685 until their CTmax or CTmin were recorded. This was repeated twice to yield sample sizes of n =

1686 20. The CTmax and CTmin were defined as the temperature at which each individual insect lost 1687 coordinated muscle function, consequently losing the ability to respond to mild stimuli (e.g.

1688 prodding) (e.g. Nyamukondiwa & Terblanche, 2009). In the case of CTmax, this loss of muscle

1689 function coincided with death such that recovery was not possible, while in the case of CTmin, 1690 recovery occured. 1691 1692 Acclimation Effects on CTLs 1693 To determine longer-term acclimation effects on thermal tolerance for C. partellus, B. 1694 fusca and S. calamistis, CTLs were measured on moths acclimated as adults to low, optimum or 1695 high temperature as outlined by Nyamukondiwa & Terblanche (2010b). Low temperature 1696 acclimation was defined for each species as 5°C less rearing/benign temperature while high 1697 temperature acclimation was defined as +5°C rearing temperature. Here, flies acclimated to 1698 optimum temperature were considered as a control group. Two day-old C. partellus, B. fusca 1699 and S. calamistis were acclimated for 3 days at (23, 28 and 33°C), (20, 25 and 30°C) and (23, 28 1700 and 33°C) respectively. Acclimation for this duration is suffice to elicit some plastic responses in 1701 like invertebrates e.g. Ceratitis capitata (see Weldon et al., 2011). Ten individual moths from 1702 each acclimation were randomly loaded into the ‘organ pipes’ and were given 10 minutes to 1703 equilibrate at benign temperature before ramping up or down at a rate of 0.25°C/min until their

1704 CTmax or CTmin were measured. This was repeated twice to yield sample sizes of n = 20. The

1705 CTmax and CTmin were defined as the temperature at which each individual insect lost 1706 coordinated muscle function, consequently losing the ability to respond to mild stimuli (e.g. 1707 prodding) (e.g. Nyamukondiwa & Terblanche, 2010b). Both basal thermal tolerance and 1708 acclimation/plasticity data were checked for normality and equality of variances using the 1709 Shapiro-Wilk and Hartley-Bartlett tests, respectively, and in all cases assumptions were met. To 1710 determine the effect of species on CTLs, and acclimation effects on CTLs, results were analysed 1711 using one-way ANOVA and factorial ANOVA respectively followed by Tukey-Kramer’s post- 1712 hoc tests to identify homogenous groups.

1713 Basal CTmax results showed that all species differed significantly with respect to high 1714 temperature tolerance. C. partellus had significantly higher heat tolerance than S. calamistis and

1715 B. fusca (F(2,177) = 294.32, p<0.0001; Table 2.5). Busseola fusca and S. calamistis had

1716 significantly higher cold tolerance than C. partellus (F(2,57) = 235.38, p<0.0001; Table 2.5). 1717 However, there was no significant difference in cold tolerance between S. calamistis and B. fusca

1718 (Table 2.5). We found a significant species x acclimation temperature interaction for both CTmin

1719 and CTmax following low and high temperature acclimation (Table 2.6). Acclimation at high

1720 temperature generally increased CTmax for C. partellus relative to B. fusca and S. calamistis 1721 (Figure 2.2A). However, acclimation at both low and high temperature compromised high 1722 temperature tolerance for both C. partellus and S. calamistis indicating some compromised 1723 potential for plasticity. Similarly, C. partellus also showed high plastic responses to low 1724 temperature tolerance compared to S. calamistis and B. fusca (Figure 2.2B). There was no

1725 significant difference across all acclimation treatments for B. fusca in respect of CTmin. 1726 Correspondingly, S. calamistis acclimated at low temperature performed worse off than the 1727 controls indicating some inability to compensate for the cold. Overall, this showed that C. 1728 partellus has a high basal temperature tolerance and plasticity thereof. In addition, its 1729 compromised basal low temperature tolerance may be a cost to its high plasticity to low 1730 temperature (Figure 2.2B). These results are in keeping with related research to date, indicating 1731 insects may trade off their basal low but not high temperature tolerance for plasticity (as in 1732 Nyamukondiwa et al., 2011). This may help explain that ecophysiology may aid the current 1733 observed invasion potential of C. partellus relative to B. fusca and S. calamistis. 1734 1735 Table 2.5: Results of basal thermal tolerance ±SD in three major stemborers, (B. fusca, S. 1736 calamistis and C. partellus) measured as (A) critical thermal maxima and (B) critical thermal 1737 minima. Means with the same superscript letter are not significantly different. 1738 Stemborer species Critical thermal maxima Critical thermal minima

Busseola fusca 44.69±0.78a 1.47±0.55 a

Sesamia calamistis 46.24±1.21b 1.78±0.47a

Chilo partellus 47.81±0.45c 4.83±0.60 b

1739 1740

1741 Table 2.6: Summary results from a factorial ANOVA showing the effects of species (Chilo 1742 partellus, Busseola fusca and Sesamia calamistis) and temperature acclimation (low, optimum

1743 and high) on critical thermal maxima and minima (CTmax and CTmin, respectively). SS = sums of 1744 squares, d.f.=degrees of freedom. 1745 Effect SS DF MS F P

Species 416.1 2 208 521,6 0.000

CT max Acclimation 26.9 2 13.5 33.8 0.000

Species*Acclimation 30 4 7.5 18.8 <0.0001

Species 286 2 143 674.3 0.000

CTmin Acclimation 2 2 1 4.6 0.011

Species*Acclimation 37.6 4 9.4 44.3 <0.0001

1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761

1762 49 Busseola fusca a 1763A Sesamia calamistis a a 48 Chilo partellus 1764

°C ) 1765 47 b 1766 1767 46 c 1768 a 45 ab 1769 a b 1770 44 Critical thermal ( maxima 1771

43 1772 Low Optimum High

Acclimation temperature (°C) 1773 1774

6 Busseola fusca B Sesamia calamistis b Chilo partellus b

C) 5

a 4

b b a 2 a a

Critical minima thermal (

1 Low Optimum High Acclimation temperature (°C ) 1775 1776 1777 Figure 2.2: Results of plasticity of thermal tolerance in three major stemborers, (B. fusca, S. 1778 calamistis and C. partellus) measured as (A) critical thermal maxima and (B) critical thermal 1779 minima following acclimation for 3 days at low optimum and high temperature (23; 28 and 33°C 1780 respectively). Vertical bars denote ±95% CL. Means with the same letter are not significantly 1781 different from each other. Post hoc tests were done separately for each species

1782 Beneficial acclimation hypothesis states that acclimation to a particular environment 1783 gives an organism a performance advantage in that environment over another organism that has 1784 not been exposed to that particular environment (Chown & Terblanche, 2007). Basing on this 1785 hypothesis, C. partellus may be migrating from the relatively warm low to middle altitudes to 1786 new, relatively cooler high altitude habitats. Since mean global temperatures are gradually 1787 increasing, causing a ‘warming’ in the rather cool middle and high altitude areas, C. partellus 1788 may have a performance advantage over other indigenous stemborer species through beneficial 1789 acclimation and hence may be fitter than the latter. Furthermore, C. partellus may have adopted 1790 behavioral mechanisms of coping with high temperature through its thermal history in low 1791 altitudes as has been reported in other insects (e.g. Nyamukondiwa et al., 2010; 2011). However, 1792 indigenous stemborers e.g. S. calamistis and B. fusca may not be able to sufficiently acclimatize 1793 to rapidly changing thermal conditions, hence may have performance and survival disadvantages 1794 in the field (see discussions in Khadioli et al., 2014b). However, little is known on the thermal 1795 requirements for the development of C. partellus life stages, which would be useful in the 1796 development of pest invasion risk maps (Khadioli et al., 2014a) such as those developed by 1797 Yonow et al. (2016). 1798

1799 2.3.6 Host range of C. partellus 1800 Stemborers are among the phytophagous insects that occurred in wild hosts before 1801 domestication of sorghum and introduction of maize in Africa (Polaszek & Khan, 1998; 1802 Ristanovic, 2001). They have been associated with a wide range of wild hosts belonging to 1803 Poaceae, Cyperaceae and Typhaceae families for millions of years (Le Ru et al., 2006). 1804 However, their pest status has been elevated by the provision of abundant resources that occur in 1805 monocultural ecosystems in which maize and sorghum crops create continuous availability of 1806 resources (Overholt et al., 1997; Mwalusepo et al., 2015). Besides surviving in cultivated 1807 habitats, previous studies have also revealed that C. partellus can survive in wild grasses more 1808 than other indigenous stemborer species (Songa et al., 2002 see Ofomata et al., 2000). A survey 1809 that was conducted in Kenya, Tanzania and Uganda indicated that B. fusca was not common in 1810 wild hosts, showing its relative preference for cultivated hosts e.g. maize and sorghum crops (Le 1811 Ru et al., 2006). However, a related study revealed C. partellus as the most common borer 1812 species in wild host plants (Songa et al., 2002). This may be due, in part, to its high potential in

1813 colonizing novel environments (Overholt et al., 1997; Kfir, 1997) and ability to utilize diverse 1814 range of hosts. It is therefore likely that C. partellus adjusts faster to the wild grass hosts and that 1815 it colonizes the suitable feeding niches in these habitats much earlier than the other stemborers, 1816 hence decreasing the number of other stemborer species that successfully colonize these grasses 1817 (Songa et al., 2002). In line with this study, the success of C. partellus is probably centered on 1818 switching from wild to cultivated hosts and vice versa. Furthermore, in areas where host plants 1819 are abundantly distributed and climate conducive for development, C. partellus is known to be 1820 active all year round (Khadioli et al., 2014a). Therefore, with its short life cycle, host plant 1821 switching/polyphagy coupled with quick diapause termination, it can rapidly increase its 1822 population thus ensuring numerical advantage over other indigenous stemborer species. 1823

1824 2.3.7 Asynchrony between C. partellus and its natural enemies 1825 Changes in the synchronization of the host and its parasitoid through variations in their 1826 thermal preferences, or by changes in the geographical distribution of either species can disrupt 1827 the equilibrium between hosts and parasitoids (Hance et al., 2007). If a parasitoid species has a 1828 lower base temperature than its host, the parasitoid could emerge earlier in the seasonal 1829 development of its host. When the parasitoids emerge early, host stemborer population exposed 1830 to adult parasitoids is considerably reduced. Mwalusepo et al. (2015) predicted future 1831 asynchrony between C. partellus and its larval parasitoid Cotesia flavipes Cameron, with 1832 efficacy expected to be better at low than high altitude. Therefore, with C. partellus gradually 1833 expanding to moist-mid and high altitude areas, there are high chances of disruption of the 1834 biological control due to a mismatch between the host and its main larval parasitoid C. flavipes. 1835 This implies that C. partellus may be passing through vulnerable life stages more quickly at 1836 higher temperatures, reducing the window of opportunity for parasitism which may give great set 1837 back to the survival and multiplication of C. flavipes. Future failure of biological control may 1838 thus explain the reasons for its population increase and displacement of other indigenous 1839 stemborer species. Mailafiya et al. (2010) showed that the distribution of C. flavipes and Cotesia 1840 sesamiae Cameron was influenced by geographic range of their respective stemborer hosts. 1841 Considering the predicted expansion of C. partellus to moist mid-altitude and moist high altitude 1842 areas, C. flavipes is also expected to shift to these areas. However, predicted climate change 1843 indicate 10% rainfall decrease and more frequent droughts in southern Africa (Stathers et al.

1844 2013) and a reduction in rainfall during the long rainy season in east Africa (Mwalusepo et al., 1845 2015). These dry periods may significantly affect Cotesia species (Mwalusepo et al., 2015) 1846 leading to a disruption in biological control and ultimately promoting a sudden increase in C. 1847 partellus populations which may eventually displace other indigenous stemborer species, whose 1848 biological control may not be compromised as such. The two major parasitoid species of C. 1849 partellus, C. flavipes and C. sesamiae are known to have a short generation time (18 days at 25– 1850 30°C) (Mochiah et al., 2001) compared to that of their stemborer hosts (30-58 days at 25–30oC) 1851 (Polaszek, 1998). These parasitoids’ high reproductive rates with a female biased sex ratio 1852 (Omwega & Overholt, 1997) demonstrate a high potential for stemborer pest regulation. 1853 However, Cotesia species lack the ability to locate and parasitize diapause larvae in dried maize 1854 stems (Overholt et al., 1997). Since the warming of African continent is gradually increasing, 1855 synchronization between C. partellus and these parasitoids may be affected. This implies that the 1856 occurrence of extreme temperatures or higher fluctuations in temperature may strongly affect the 1857 host-parasitoid population dynamics through altering the capacity of parasitoids to locate hosts 1858 and host habitats. Chilo partellus might be escaping parasitism due to asynchrony leading to 1859 increase in the number of generations as well as expansion to other regions. Increase in 1860 population is known to create intraspecific competition amongst the host species and this may 1861 facilitate the migration of C. partellus to other regions consequently displacing other indigenous 1862 stemborer species. 1863

1864 2.4 Future Directions and Conclusions 1865 Global warming and climate change may have negative consequences on diversity and 1866 abundance of stemborers hence increasing the extent of crop losses that will impact negatively 1867 profitability of crop production, food security and livelihoods (Gregory et al., 2009; Kumar, 1868 2016; Yonow et al., 2016). Through a number of inter-related processes, including range 1869 extensions and phenological changes, as well as increased rates of population development, 1870 growth, migration, and overwintering (Hance et al., 2007), most studies have concluded that 1871 insect pests generally become more abundant as temperature increases (Cannon, 1998; 1872 Yamamura & Kiritani, 1998; Lin, 2011). Few studies have included parasitoids in their models 1873 and the effect of these parasitoids in shaping the insect pest. In moist tropical forests, parasitoids 1874 may suffer more than their hosts from a decrease in precipitation, which will then cause an

1875 increase in the host population because of a reduced level of parasitism (Coley, 1998; Hance et 1876 al., 2007). 1877 Understanding stemborers and their parasitoids population dynamics under variable climate 1878 conditions is essential for predicting seasonal phenology and accurate forecasting of moth 1879 abundance which has downstream impacts on basic applications of any pest management 1880 program. Indeed, dramatic changes in high and low temperature tolerance can occur between 1881 life-stages of various insect species (Bowler & Terblanche, 2008). Adaptations to extreme 1882 temperatures are often different among stemborers and parasitoids developmental stages (Hance 1883 et al., 2007), but this has not been investigated. Adverse effects of climate change on the activity 1884 and effectiveness of stemborer/natural enemy systems need to be considered in future pest 1885 management programs. This involves understanding physiological tolerances of these stemborers 1886 and their natural enemies in order to devise methods for management. Model predictions indicate 1887 a potential increase in the number of spotted stemborer generations in the year 2050, which 1888 indicates that higher insect pest infestations and yield losses may occur (Kroschel et al., 2013). 1889 Most bioclimatic models to date were purely based on temperature and could not take into 1890 consideration other climatic factors such as rainfall and relative humidity and, non-climatic 1891 factors such as planting patterns and frequency, diapause or other physiological factors that 1892 enable insects to cope with climate changes in the short and long timescales e.g. plasticity (see 1893 Khadioli et al., 2014a; Mwalusepo et al., 2015). Despite overwhelming evidence that species are 1894 evolving with global change (Bradshaw & Holzapfel, 2006), to my knowledge, no models to 1895 date, have incorporated the ability of stemborers to evolve. I suggest future predictions of climate 1896 change effects on insect abundance should incorporate evolutionary capacity (e.g. Kearney et al., 1897 2009), since it’s one of the ways insects use to cope to environmental variation, upon 1898 introduction to novel habitats. 1899 Therefore there is need to: 1900  Predict and map trends of potential changes in geographical distribution, and study how 1901 climatic changes will affect development, incidence, performance and population 1902 dynamics of stemborer species and associated natural enemies. Furthermore, when 1903 predicting direct effects of climate change, on insect hosts, natural enemies and their 1904 interactions, species genotypic and phenotypic plasticity should be considered (e.g. 1905 Thomson et al., 2010).

1906  It is a major challenge predicting future impacts of climate variability without a 1907 comprehensive understanding of the species’ thermal tolerance and performance under 1908 current climatic conditions. For many insect species, including some economic pests, this 1909 information is still lacking. An understanding of stemborers’ thermal biology mainly 1910 focusing on basal thermal tolerance and plasticity thereof may thus provide valuable 1911 information for predictive models for forecasting economic insect pest outbreaks, 1912 conducting pest risk assessments, establishing invasive species, biogeographical shifts, 1913 and the potential implications of global climate change. 1914  Evidence from diverse insect taxa suggests that, in a new habitat, species may often 1915 undergo rapid and observable evolutionary change (Gilchrist & Lee, 2007). This may 1916 mean that genetic heterogeneity in phenotypic traits may be inherently present amongst 1917 early invader and propagule generations, following colonization for adaptive evolution to 1918 occur in response to novel habitats. In addition, phenotypic plasticity, and particularly 1919 cross-generational developmental plasticity may be playing a key role in producing 1920 phenotypes capable of competing favorably with native species. With C. partellus 1921 quickly evolving to new habitats, thriving and dominating other stemborer species, there 1922 is need to understand its selection response and genetic variation as well as factors that 1923 have facilitated it to evolve and outcompete other indigenous stemborer species. 1924 The invasive stemborer, C. partellus is a highly competitive colonizer and with Africa’s 1925 changing climate, it is most likely to expand and establish in novel habitats. In addition to other 1926 biophysical factors likely aiding its abundance, knowledge of thermal tolerance, plasticity and 1927 evolutionary potential for C. partellus and its larval parasitoid is vital to the understanding of 1928 how they may react to potential climate change with implications for the prediction of pest 1929 outbreaks. Furthermore, a careful analysis on how stemborers-parasitoids systems react to 1930 changes in temperature and relative humidity is needed to assist prediction models for pest 1931 management under global change. In ecosystems, the tritrophic interactions between plants, 1932 herbivorous insects, and their natural enemies (predators, parasitoids, and pathogens) are affected 1933 by climatic changes in diverse ways. Therefore, research should also focus on tritrophic 1934 interactions rather than host-parasitoid level dynamics only. The invasiveness of C. partellus and 1935 its geographic spread are very significant to crop protection. Therefore, a comprehensive total 1936 ecology of this pest and other indigenous stemborers should also be studied.

1937 References 1938 Addo-Bediako, A. & Thanguane, N. (2012) Stemborer distribution in different sorghum cultivars 1939 as influenced by soil fertility. Agricultural Science Research Journal, 2, 189- 194. 1940 Aragòn, P., Baselga, A. & Lobo, J.M. (2010) Global estimation of invasion risk zones for the 1941 western corn rootworm Diabrotica virgifera virgifera: integrating distribution models and 1942 physiological thresholds to assess climatic favourability. Journal of Applied Ecology, 47, 1943 1026-1035. 1944 Assefa, G.A. (1985) Survey of lepidopterous stemborers attacking maize and sorghum in 1945 Ethiopia. Ethiopian. Journal of Agricultural Science, 7, 55-59. 1946 Bale, J.S. (2010) Implications of Cold-tolerance for Pest Management. Low Temperature 1947 Biology of Insects (ed. by D.L. Denlinger & R.E. Lee), pp. 342-372. Cambridge University 1948 Press, UK. 1949 Berger, A. (1992) Larval movement of Chilo partellus (Lepidoptera: Pyrallidae) within and 1950 between plants: timing, density responses and survival. Bulletin of Entomological Research, 1951 82, 441 - 448. 1952 Bradshaw, W.E. & Holzapfel, C.M. (2006) Evolutionary response to rapid climate change. 1953 Science, 312, 1477-1478. 1954 Bowler, K. & Terblanche, J.S. (2008) Insect thermal tolerance: what is the role of ontogeny, 1955 ageing and senescence? Biological Reviews, 83, 339-355. 1956 Cannon, R.J.C. (1998) The implications of predicted climate change for insect pests in the UK, 1957 with emphasis on non-indigenous species. Global Change Biology, 4, 785-796. 1958 Chabi - Olaye, Schulthess, A.F., Poehling, H.M. & Borgemeister, C. (2001) Factors affecting 1959 biology of Telenomous isis (Polaszek) (Hymenoptera: Scelionidae), an egg parasitoid of 1960 cereal stemborers in West Africa. Biological Control, 21, 44 - 54. 1961 Chapman, R.F., Woodhead, S. & Bernays, E.A. (1983) Survival and dispersal of young larvae 1962 of Chilo partellus (Swinhoe) (Lepidoptera: Pyralidae) in two cultivars of sorghum. Bulletin 1963 of Entomological Research, 73, 65-74. 1964 Chidawanyika, F., Mudavanhu, P. & Nyamukondiwa, C. (2012) Biologically based methods for 1965 pest management in agriculture under changing climates: Challenges and future directions. 1966 Insects, 3, 1171-1189.

1967 Chidawanyika, F., Midega, C.A.O., Bruce, T.J.A., Duncan, F., Pickett, J.A. & Khan, Z.R.K. 1968 (2014) Oviposition acceptance and larval development of Chilo partellus stemborers in 1969 drought-stressed wild and cultivated grasses of East Africa. Entomologia Experimentalis et 1970 Applicata, 151, 209-217. 1971 Chown, S.L. & Nicolson, S.W. (2004) Insect Physiological Ecology: Mechanisms and Patterns. 1972 Oxford University Press, Oxford. 1973 Chown, S.L., Slabber, S., McGeoch, M.A., Janion, C. & Leinaas, H.P. (2007) Phenotypic 1974 plasticity mediates climate change responses among invasive and indigenous arthropods. 1975 Proceedings of the Royal Society of London. Series B, 274, 2661-2667. 1976 Chown, S.L. & Terblanche, J.S. (2007) Physiological Diversity in Insects: Ecological and 1977 Evolutionary Contexts. Advances in Insect Physiology, 33, 50-152. 1978 Coley, P.D. (1998) Possible effects of climate change on plant/herbivore interactions in moist 1979 tropical forests. Climate Change, 39, 455–472. 1980 Delobel, A. (1975) Chilo orichalcociliellus Strand (Lepidoptera: Pyralidae), fôreur des tiges du 1981 sorgho et du mais à Madagascar II. Premières données biologiques. ORSTOM Sér Biol, 1982 10, 11-16. 1983 De Groote H, (2002) Maize yield loss from stemborers in Kenya. Journal of Insect Science, 22: 1984 89–96. 1985 Dejen, A., Getu, E., Azerefegne, F. & Ayalew, A. (2014) Distribution and impact of Busseola 1986 fusca (Fuller) (Lepidoptera: Noctuidae) and Chilo partellus (Swinhoe) (Lepidoptera: 1987 Crambidae) in Northeastern Ethiopia. Journal of Entomology and Nematology, 6, 1-13. 1988 Duerdon, J.C. (1953) Stem borers of cereal crops at Kongwa, Tanganyika, 1950-52. East African 1989 Agricultural and Forestry Journal, 19, 105-119. 1990 Emana, G., Overholt, W.A. & Kairu, E. (2001) Ecological management of cereal stemborers in 1991 Ethiopia. pp. 55-99. In Seventh Eastern and Southern Africa Regional Maize conference, 11- 1992 15th February 2001. Nairobi Kenya. 1993 Farji-Brener, A.G. & Corley, J.C. (1998) Successful invasions of hymenopteran insects into NW 1994 Patagonia. Ecologia Austral, 8, 237-249. 1995 Freedman, B. (1995) Environmental Ecology: The Ecological effects of Pollution, Disturbance 1996 and Other Stresses. Second Edition. Academic Press. London. Pp 550.

1997 Getu, E., Overholt, W.A., Kairu, E., Macopiyo, L. & Zhou, G. (2002) Predicting the distribution 1998 of Chilo partellus (Swinhoe) (Lepidoptera: Crambidae) and Cotesia flavipes (Cameron) 1999 (Hymenoptera: Braconidae) in Ethiopia using correlation, step-wise regression and 2000 geographic information system. Insect Science and its Application, 22, 123-129. 2001 Gilchrist, G.W. & Lee, C.E. (2007) All stressed out and nowhere to go: does evolvability limit 2002 adaptation in invasive species? Genetica, 129, 127–132. 2003 Glatz, J., du Plessis, H. & Van den Berg, J. (2016) The effect of temperature on the development 2004 and reproduction of Busseola fusca (Lepidoptera: Noctuidae). Bulletin of Entomological 2005 Research, doi:10.1017/S0007485316000572. 2006 Gregory, P.J., Johnson, S.N., Newton, A.C. & Ingram, J.S.I. (2009) Integrating pests and 2007 pathogens into the climate change/ food security debate. Journal of Experimental Botany, 60, 2008 2827–2838. 2009 Hance, T., Van Baaren, J., Vernon, P. & Boivin, G. (2007) Impact of extreme temperatures on 2010 parasitoids in a climate change perspective. Annual Review of Entomology, 52, 107–126. 2011 Harris, K.M. & Nwanze, K.F. (1992) Busseola fusca (Fuller), the African maize stemborer: a 2012 handbook of information. ICRISAT. CABI. Oxon. UK pp 84. 2013 Herborg, L.M., Jerde, C.L., Lodge, D.M., Ruiz, G.M. & MacIsaac, H.J. (2007) Predicting 2014 invasion risk using measures of introduction effort and environmental niche models. 2015 Ecological Applications, 17, 663-674. 2016 IPCC. (2014) Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and 2017 III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Core 2018 Writing Team, R.K. Pachauri and L.A. Meyer (eds.)]. IPCC, Geneva, Switzerland, pp 151. 2019 Jaramillo, J., Muchugu, E., Vega, F.E., Davis, A., Borgemeister, C. & Chabi-Olaye, A. (2011) 2020 Some like it hot: the influence and implications of climate change on coffee berry borer 2021 (Hypothenemus hampei) and coffee production in East Africa. Plos One, 6, e24528. doi: 2022 10.1371/journal.pone.0024528 PMID: 21935419. 2023 Jiang, N., Zhou, G., Overholt, W.A., Muchugu, E. & Schulthess, F. (2006) The temporal 2024 correlation and spatial synchrony in the stemborer and parasitoid system of Coast Kenya with 2025 climate effects, Annales de la Société entomologique de France (N.S.): International Journal 2026 of Entomology, 42, 3-4, 381-387.

2027 Juroszek, P. and von Tiedemann, A. (2013) Plant pathogens, insect pests and weeds in a 2028 changing global climate: a review of approaches, challenges, research gaps, key studies and 2029 concepts, Journal of Agricultural Science, doi:10.1017/S0021859612000500. 2030 Karuppaiah, V. & Sujayanad, G.K. (2012) Impacts of climate change on population dynamics of 2031 insect pests. World Journal of Agricultural Sciences, 8, 240-246. 2032 Kearney, M., Porter, W.P., Williams, C., Ritchie, S. & Hoffmann, A.A. (2009) Integrating 2033 biophysical and evolutionary theory to predict climatic impacts on species’ ranges: the 2034 dengue mosquito Aedes aegypti in Australia. Functional Ecology, 23, 528-538. 2035 Kfir R. (1991) Duration of diapause in the stem borers, Busseola fusca and Chilo partellus. 2036 Entomologia Experimentalis et Applicata, 61, 265–270. 2037 Kfir, R. (1993) Diapause termination in Busseola fusca (Lepidoptera: Noctuidae) in the 2038 laboratory. Annals of the Entomological Society of America, 86, 273-277. 2039 Kfir, R. (1997) Competitive displacement of Busseola fusca (Lepidoptera: Noctuidae) by Chilo 2040 partellus (Lepdoptera: Pyralidae). Annals of the Entomological Society of America, 90, 619- 2041 624. 2042 Kfir, R., Overholt, W.A., Khan, Z.R. & Polaszek, A. (2002) Biology and management of 2043 economically important lepidopteran cereal stemborers in Africa. Annual Review of 2044 Entomology, 47, 701–713. PMID: 11729089. 2045 Khadioli, N., Tonnang, Z.E.H., Muchugu, E., Ong’amo, G., Achia, T., Kipchirchir, I., Kroschel, 2046 J. & Le Ru, B. (2014a) Effect of temperature on the phenology of Chilo partellus (Swinhoe) 2047 (Lepidoptera: Crambidae), simulation and visualization of the potential future distribution of 2048 C. partellus in Africa under warmer temperatures through the development of life-table 2049 parameters. Bulletin of Entomological Research, 104, 809–822. 2050 Khadioli, N., Tonnang, Z.E.H. Ong’amo, G., Achia, T., Kipchirchir, I., Kroschel, J. & Le Ru, B. 2051 (2014b) Effect of temperature on the life history parameters of noctuid lepidopteran 2052 stemborers, Busseola fusca and Sesamia calamistis. Annals of Applied Biology, 2053 doi:10.1111/aab.12157. 2054 Kristensen, T.N., Hoffmann, A.A., Overgaard, J., Sørensen, J.G. & Hallas, R. (2008) Costs and 2055 benefits of cold acclimation in field released Drosophila. Proceedings of the National 2056 Academy of Science of the United States of America, 105, 216-221.

2057 Kroschel, J., Tonnang, H.E.Z., Le Ru, B. & Hanna, R. (2013) Analysing climate impacts on 2058 insect pests using phenology modeling and geographic information system implemented in 2059 the insect life cycle modeling software. ICIPE. Nairobi. Kenya. 2060 Kumar, M. (2016) Impact of climate change on crop yield and role of model for achieving food 2061 security. Environmental Monitoring and Assessment, 188, 465. DOI 10.1007/s10661-016- 2062 5472-3. 2063 Kutywayo, D., Chemura, A., Kusena, W., Chidoko, P. & Mahoya, C. (2013) The Impact of 2064 Climate Change on the Potential Distribution of Agricultural Pests: The Case of the Coffee 2065 White Stem Borer (Monochamus leuconotus P.) in Zimbabwe. PLoS ONE, 8, e73432. 2066 doi:10.1371/journal.pone. 0073432. 2067 Lee, C.E., Remfert, J.L. & Chang, Y. (2007) Response to selection and evolvability of invasive 2068 populations. Genetica, 129, 179-192. 2069 Le Ru, B.P., Ong’amo, G.O., Moyal, P., Ngala, L., Musyoka, B., Abdullah, Z., Cugala, D. et al. 2070 (2006) Diversity of lepidopteran stem borers on monocotyledonous plants in eastern Africa 2071 and the islands of Madagascar and Zanzibar revisited. Bulletin of Entomological Research, 2072 96, 555-563. 2073 Lin, B.B. (2011) Resilience in agriculture through crop diversification: Adaptive management 2074 for environmental change. BioScience, 61, 183-193. 2075 Lobell, D.B., Burke, M.B., Tebaldi, C., Mastrandrea, M.D., Falcon, W.P & Naylor, R.I. (2008) 2076 Prioritising climate change adaptation needs for food security in 2030. Science, 319, 607- 2077 610. 2078 Lo´ pez-Darias, M., Lobo, J.M. & Gouat, P. (2008) Predicting invasive species potential 2079 distribution: the exotic Barbary ground squirrel in the Canarian Archipelago and west 2080 Mediterranean region. Biological Invasions, 10, 1027–1040. 2081 Mailafiya, D.M., Le Ru, B.P., Kairu, E.W., Calatayud, P-A. & Dupas, S. (2010) Geographic 2082 distribution, host range and perennation of Cotesia sesamiae and Cotesia flavipes Cameron 2083 in cultivated and natural habitats in Kenya. Biological Control, 54, 1–8. 2084 Mailafiya. D.M., Le Ru, B.P., Kairu. E.W., Calatayud, P.A. and Dupas, S. (2009) Species 2085 diversity of lepidopteran stem borer parasitoids in cultivated and natural habitats in Kenya. 2086 Journal of Applied Entomology, 133, 416-429.

2087 Masters, G. & Norgrove, L. (2010) Climate change and invasive alien species. CABI Working 2088 Paper 1, Switzerland, pp. 30. 2089 Mbapila, J.C., Overholt, W.A. & Kayumbo, H.Y. (2002) Comparative development and 2090 population growth of an exotic stemborer, Chilo partellus (Swinhoe), and an ecologically 2091 similar congener, Chilo orichalcociliellus (Strand) (Lepidoptera: Crambidae). Insect Science 2092 Application, 22, 21–27. 2093 Mika, A.M.., Weiss, R.M., Olfert, O., Hallett, R.H. & Newman, J.A. (2008) Will climate change 2094 be beneficial or detrimental to the invasive swede midge in North America? Contrasting 2095 predictions using climate projections from different general circulation models. Global 2096 Change Biology, 14, 1721-1733. 2097 Mochiah, M.B., Ngi - Song, J.A., Overholt, W.A. & Botchey, M. (2001) Host suitability of four 2098 cereal stemborers (Lepidoptera: Crambidae, Noctuidae) for different geographic populations 2099 of Cotesia sesamiae (Cameron) (Hymenoptera: Braconidae) in Kenya. Biological Control, 2100 21, 285 - 292. 2101 Moya-Laraño, J., El-Sayyid, M.E.T. & Fox, C.W. (2007) Smaller beetles are better scramble 2102 competitors at cooler temperatures. Biology Letters, 3, 475–478. 2103 Mwalusepo, S., Tonnang, H.E.Z., Massawe, E.S., Okuku, G.O., Khadioli, N., Johansson, T., 2104 Calatayud, P. & Le Ru, B.P. (2015) Predicting the Impact of Temperature Change on the 2105 Future Distribution of Maize StemBorers and Their Natural Enemies along East African 2106 Mountain Gradients Using Phenology Models. PLoS ONE, 10, e0130427. 2107 doi:10.1371/journal.pone.0130427. 2108 Newton, A.C., Johnson, S.N. & Gregory, P.J. (2011) Implications of climate change for diseases,

2109 crop yields and food security. Euphytica, 179, 3-18.

2110 Niyibigira. E.I., Abdallah. Z.S., Overholt, W.A., Lada. V.Y. & Van Huis. A. (2001) Distribution 2111 and abundance in maize and sorghum of lepidopteran stemborers and associated indigenous 2112 parasitoids in Zanzibar. Insect Science and its Application, 21, 335-346. 2113 Ntiri, E.S., Calatayud, P.A., Van Den Berg, J., Schulthess, F. & Le Ru, B. P. (2016) Influence of 2114 temperature on intra- and interspecific resource utilization within a community of 2115 lepidopteran maize stemborers, PLoS ONE, 11, e0148735. 2116 doi:10.1371/journal.pone.0148735.

2117 Nyamukondiwa, C., Kleynhans, E. & Terblanche, J.S. (2010) Phenotypic plasticity of thermal 2118 tolerance contributes to the invasion potential of Mediterranean fruit flies (Ceratitis 2119 capitata). Ecological Entomology, 35, 565-575. 2120 Nyamukondiwa, C. & Terblanche, J.S. (2009) Thermal tolerance in adult Mediterranean and 2121 Natal fruit flies (Ceratitis capitata and Ceratitis rosa): effects of age, gender and feeding 2122 status. Journal of Thermal Biology, 34, 406-414. 2123 Nyamukondiwa, C. & Terblanche, J.S. (2010) Within-generation variation of critical thermal 2124 limits in adult Mediterranean and Natal fruit flies Ceratitis capitata and Ceratitis rosa: 2125 thermal history affects short-term responses to temperature. Physiological Entomology, 35, 2126 255-264. 2127 Nyamukondiwa, C., Terblanche, J.S., Marshall, K.E. & Sinclair, B.J. (2011) Basal but not heat 2128 tolerance constraints plasticity among Drosophila species (Diptera: Drosophilidae). Journal 2129 of Evolutionary Biology, 24, 1927-1938. 2130 Nyamukondiwa, C., Weldon, C.W., Chown, S.L., le Roux, P.C. & Terblanche, J.S. (2013) 2131 Thermal biology, population fluctuations and implications of temperature extremes for the 2132 management of two globally significant insect pests. Journal of Insect Physiology, 59, 1199- 2133 1211. 2134 Ochieng, R.S., Onyango, F.O. & Bungu, M.D.O. (1985) Improvement of techniques for mass 2135 culture of Chilo partellus (Swinhoe). Insect Science and its Application, 6, 425-428 2136 Ofomata, V. C. (1997) Ecological interactions between Chilo orichalcociliellus (Strand) and 2137 Chilo partellus (Swinhoe) (Lepidoptera: Pyralidae) on the Kenya Coast. Ph.D. thesis. 2138 Nnarndi Azikiwe University, Anambra State, Nigeria. 2139 Ofomata, V. C., Overholt, W.A. & Egwuatu, R.I. (1999) Diapause termination of Chilo partellus 2140 Swinhoe and Chilo orichalcociliellus Strand (Lepidoptera: Crambidae). Insect Science and 2141 its Application, 19, 187-191. 2142 Ofomata, V.C., Overholt, W.A., Lux, S.A., Van Huis, A. & Egwuatu, R.I. (2000) Comparative 2143 studies on the fecundity, egg survival, larval feeding and development of Chilo partellus 2144 (Swinhoe) and Chilo orichalcociliellus strand (Lepidoptera: Crambidae) on five grasses. 2145 Annals of the Entomological Society of America, 93, 492-499.

2146 Ofomata, V.C., Overholt, W.A., Van Huis, A., Egwuatu, R.I. & Ngi-Song, A.J. (1999) Niche 2147 overlap and interspecific association between Chilo partellus and Chilo orichalcociliellus on 2148 the Kenya coast. Entomologia Experimentalis et Applicata, 93, 141-148. 2149 Omwega, C.O. & Overholt, W.A. (1997) Progeny production and sex ratios of field populations 2150 of the parasitoids Cotesia flavipes and Cotesia sesamiae reared from gramineous stemborers 2151 in coastal Kenya. Insect Science and its Application, 17, 137 - 142. 2152 Ong’amo, G.O., Le Rü, B.P., Dupas, S., Moyal, P., Calatayud, P. A. & Silvain, J.F. (2006) 2153 Distribution, pest status and agroclimatic preferences of lepidopteran stemborers of maize in 2154 Kenya. Annales de la Société Entomologique de France, 42, 171–177. 2155 Onyango, F.O & Ochieng’-Odero, J.P.R. (1994) Continuous rearing of the maize stem borer 2156 Busseola fusca on an artificial diet. Entomologia Experimentalis et Applicata, 73, 139-144. 2157 Overholt, W.A., Ngi-Song, A.J., Omwega, C.O., Kimani-Njogu, S.W., Mbapila, J., Sallam, M.N. 2158 and Ofomata, V. (1997) A review of the introduction and establishment of Cotesia flavipes 2159 Cameron in East Africa for the biological control cereal stemborers. Insect Science and its 2160 Application, 17, 79-88. 2161 Overholt, W. A., Songa, J.M., Ofomata, V. & Jeske, R. (2000) The spread and ecological 2162 consequences of the invasion of Chilo partellus (Swinhoe) (Lepidoptera: Crambidae) in 2163 Africa. In E. E. Lyons and S. E. Miller Invasive species in eastern Africa: Proceedings of a 2164 Workshop held at ICIPE, July 5-6, 1999. ICIPE Science Press, Nairobi. 2165 Panwar, V.P. (1995) Agricultural Insect Pests of Crops and their Control. New Delhi, Kalyani 2166 Publishers. 2167 Parmesan, C., Ryrholm, N., Stefanescu, C., Hill, J.K., Thomas, C.D., Descimon, H., Huntley, B., 2168 Kaila, L., Kullberg, J. & Tammaru, T., et al. (1999) Pole ward shifts in geographical ranges 2169 of butterfly species associated with regional warming. Nature, 399, 579–583. 2170 Parmesan, C. (2007) Influences of species, latitudes and methodologies on estimates of 2171 phenological response to global warming. Global Change Biology, 13, 1860-1872. 2172 Parmesan, C. & Yohe, G. (2003) A globally coherent fingerprint of climate change impacts 2173 across natural systems. Nature, 421, 37–42. 2174 Polaszek, A. (1998) African cereal stemborers: Economic importance, taxonomy, natural 2175 enemies and control. CABI Publishing. Wallingford.Polaszek, A. & Khan, Z.R. (1998) Host

2176 plants. African Cereal Stem Borers: Economic Importance, Taxonomy, Natural Enemies and 2177 Control (ed. by A Polaszek), pp. 4–10. CAB International, Wallingford, UK. 2178 Reitz, S.R. & Trumble, J.T. (2002) Competitive Displacement among Insects and Arachnids. 2179 Annual Review of Entomology, 47, 435-65. 2180 Ristanovic, D. (2001) Maize. In: R.H. Raemaekers (Editor) Crop Production in Tropical Africa. 2181 Goekint Graphics nv. Belgium. pp. 23 – 45. 2182 Scriber, J.M. (2002) Latitudinal and local geographic mosaics in host plant preferences as shaped 2183 by thermal units and voltinism in Papilio spp. (Lepidoptera). European Journal of 2184 Entomology, 99, 225-239. 2185 Seshu Reddy, K.V. (1983) Sorghum stemborers in eastern Africa. Insect Science and its 2186 Applications, 4, 33-39. 2187 Sharma, H.C. (2010) Global Warming and Climate Change: Impact on Arthropod Biodiversity, 2188 Pest Management, and Food Security. International Crops Research Institute for the Semi- 2189 Arid Tropics (ICRISAT), Andhra Pradesh, India. 2190 Sharma, H.C. & Nwanze, K.F. (1997) Insect pests of sorghum: biology, extent of losses, and 2191 economic thresholds. Pages 9-23 in Plant resistance to insects in sorghum (Sharma, H.C., 2192 Faujdar Singh, and Nwanze. K.E., eds.). Patancheru 502 324, Andhra Pradesh, India: 2193 International Crops Research Institute for the Semi- Arid Tropics. 2194 Schreck, C.J. & Semazzi, F.H.M. (2004) Variability of the climate of eastern Africa. 2195 International Journal of Climatology, 24, 681-701. 2196 Sithole, S.Z. (1990) Status and control of the stemborer, Chilo partellus Swinhoe (Lepidoptera: 2197 Pyralidae) in Southern Africa. International Journal of Tropical Science, 11, 479-488. 2198 Songa, J. M., Overholt, W.A., Mueke, J.M. & Okello, R.O. (2002) Regional distribution of 2199 lepidopteran stemborers and their parasitoids among wild grasses in the semi-arid eastern 2200 Kenya. African Crop Science Journal, 10, 183-194. 2201 Stathers, T., Lamboll, R. & Mvumi, B.M. (2013) Postharvest agriculture in changing climates: 2202 its importance to African smallholder farmers. Food Security, 5, 361-392. 2203 Sujay, H.Y., Sattagi, H.N. & Patil K.R. (2010) Invasive alien insects and their impact on 2204 agroecosystem. Karnataka Journal of Agricultural Sciences, 23, 26-34.

2205 Sutherst, R., Baker, R.H.A., Coakley, S.M., Harrington, R., Kriticos, D.J. & Scherm, H. (2007) 2206 Pests under global change—meeting your future landlords? In: Canadell JG (ed) Terrestrial 2207 ecosystems in a changing world. Springer, Berlin, Germany, pp 211–225. 2208 Sylvain, N.M., Manyangarirwa, W., Tuarira, M. & Onesime, M.K. (2015) Effects of 2209 lepidopterous stemborers, Busseola fusca (Fuller) and Chilo partellus (Swinhoe) on maize 2210 (Zea mays L) yield: A review. International Journal of Innovative Research and 2211 Development, 4, 181-188. 2212 Tamiru, A., Getu, E. & Jembere, B. (2007) Role of some ecological factors for an altitudinal 2213 expansion of spotted stemborer Chilo partellus (Swinhoe) (Lepidoptera: Crambidae). 2214 Ethiopian Journal of Science, 30, 71-76. 2215 Tamiru, A., Getu, E., Jembere, B. & Bruce, T. (2012) Effect of temperature and relative 2216 humidity on the development and fecundity of Chilo partellus (Swinhoe) (Lepidoptera: 2217 Crambidae). Bulletin of Entomological Research, 102, 9–15. 2218 Tams, W.H.T. (1932) New species of African Heterocera. Entomologist, 65, 1241–1249. 2219 Taneja, S.L. & Nwanze K.F. (1990) Mass rearing of Chilo spp. on artificial diets and its use in 2220 resistance screening. Insect Science and its Application, 11, 605-616. 2221 Thomas, C.D., Cameron, A. Green, R.E., Bakkenes, M. & Beaumont, L.J. (2004) Extinction risk 2222 from climate change. Nature, 427, 145–48. 2223 Tefera, T., Mugo, S., Tende, R. & Likhayo, P. (2010) Mass Rearing of Stem borers, Maize 2224 weevil and Larger grain borer Insect Pests of Maize. CIMMYT. Nairobi. 2225 Thomson, L. J., Macfadyen, S. & Hoffmann, A.A. (2010) Predicting the effects of climate 2226 change on natural enemies of agricultural pests. Biological Control, 52, 296–306. 2227 Van Hamburg, H. (1979) The grain-sorghum stalk-borer, Chilo partellus (Swinhoe) 2228 (Lepidoptera: Pyralidae): seasonal changes in adult populations in grain-sorghum in the 2229 Transvaal. Journal of the Entomological Society of South Africa, 42, 1-9. 2230 Ward, N.L. & Masters, G.J. (2007) Linking climate change and species invasion: an illustration 2231 using insect herbivores. Global Change Biology, 13, 1605-1615. 2232 Weldon, C.W., Terblanche, J.S. & Chown, S.L. (2011) Time-course for attainment and reversal 2233 of acclimation to constant temperature in two Ceratitis species. Journal of Thermal Biology, 2234 36, 479-485.

2235 Weldon, C.W., Boardman, L., Marlin, D. & Terblanche, J.S. (2016) Physiological mechanisms 2236 of dehydration tolerance contribute to the invasion potential of Ceratitis capitata 2237 (Wiedemann) (Diptera: Tephritidae) relative to its less widely distributed congeners. 2238 Frontiers in Zoology, 13,15. DOI 10.1186/s12983-016-0147-z. 2239 Whitman, D.W. & Ananthakrishnan, T.N. (2009) Phenotypic Plasticity of Insects: Mechanisms 2240 and Consequences. Science Publishers, Enfield, NH, USA. 2241 Wilson, D., Knollenberg, W. & Fudge, J. (1984) Species packing and temperature dependent 2242 competition among burying beetles (Silphidae, Nicrophorus). Ecological Entomology, 9(2), 2243 205–16. 2244 Yamamura, K. & Kiritani, K. (1998) A simple method to estimate the potential increase in the 2245 number of generations under global warming in temperate zones. Applied Entomology and 2246 Zoology, 33, 289-298. 2247 Yang, J., Sun, Y.Y. & An, H. (2008) Northern grass lizards (Takydromus septentrionalis) from 2248 different populations do not differ in thermal preference and thermal tolerance when 2249 acclimated under identical thermal conditions. Journal of Comparative Physiology: 2250 Biochemical, Systemic, and Environmental Physiology, 178, 343-349. 2251 Yonow, T., Kriticos, D.J., Ota, N., Van Den Berg, J. & Hutchison, W.D. (2016) The potential 2252 global distribution of Chilo partellus, including consideration of irrigation and cropping 2253 patterns. Journal of Pest Science, DOI 10.1007/s10340-016-0801-4 2254 Zhou G., Overholt W.A. & Mochiah M.B. (2001) Changes in the distribution of lepidopteran 2255 maize stemborers in Kenya from the1950s to 1990s. Insect Science and its Application, 21, 2256 395-402. 2257 2258 2259 2260 2261 2262 2263 2264 2265

2266 2267

2268 CHAPTER 3

2269

2270

2271

2272

2273

2274

2275 Comparative assessment of the thermal tolerance of spotted stemborer, Chilo partellus 2276 Swinhoe (Lepidoptera: Crambidae) and its larval parasitoid, Cotesia sesamiae Cameron 2277 (Hymenoptera: Braconidae) *

2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 * Published as “Mutamiswa, R., Chidawanyika, F. & Nyamukondiwa, C. (2017) Comparative 2290 assessment of the thermal tolerance of spotted stemborer, Chilo partellus Swinhoe (Lepidoptera: 2291 Crambidae) and its larval parasitoid, Cotesia sesamiae Cameron (Hymenoptera: Braconidae). 2292 Insect Science. DOI 10.1111/1744-7917.12466”

2293 3.1 Introduction 2294 Most insect species are ectotherms, having body temperature always in equilibrium with 2295 environmental temperature (Chown & Nicolson, 2004). Moreover, most physiological and 2296 biochemical processes are maximized at thermal optima and outside this range, insect 2297 performance and fitness may be reduced (Martin & Huey, 2008; Angilletta, 2009). At extreme 2298 temperatures, the rate of acquisition and consumption of resources is significantly affected, 2299 thereby impacting on growth, development, reproduction and mortality (Chown & Nicolson, 2300 2004; Angilletta, 2009). Abiotic factors such as temperature and humidity can affect survival and 2301 fitness traits and, over longer timescales, may affect seasonal phenology and population 2302 dynamics (Wallner, 1987; Selvaraj et al., 2013). Insects consequently employ various 2303 physiological adjustments to enhance fitness and survival rates in stressful environments. 2304 Societal concern on the impacts of anthropogenic climate change on biodiversity 2305 increases the need to understand how organisms respond to thermal variability, including their 2306 potential for geographic range expansion and invasion (Parmesan et al., 1999; Andrew, 2013; 2307 Andrew & Terblanche, 2013; Karsten et al., 2015). Temperature is one key factor that is highly 2308 labile under changing climates (IPCC, 2014), and in all likelihood, will result in population shifts 2309 in geographic distribution and seasonal phenology of insects over time (Parmesan et al., 1999; 2310 Mwalusepo et al., 2015; Sinclair et al., 2015). Under acute adverse thermal exposure, insects 2311 employ a range of plastic responses including behavioural avoidance of extreme environments 2312 by changes in periods of activity within a day or mere shifts in feeding positions on a plant 2313 (Kührt et al., 2006; Huey et al., 2002; Dillon et al., 2009). Other forms of plasticity involve 2314 physiological adjustments that are chiefly induced by sub-lethal temperatures (Overgaard & 2315 Sorenson, 2008; Chidawanyika & Terblanche, 2011) and sometimes other forms of stress such as 2316 starvation (Bubliy et al., 2012) and desiccation (Gray & Bradley, 2005; Bazinet et al., 2010). 2317 However, in the event that a species fails to mount any compensatory response under persistent 2318 stressful environments, extinction may likely occur (Chown & Terblanche, 2007). The 2319 determination of thermal tolerance is therefore a significant step in understanding the ways in 2320 which environmental variation may affect the population dynamics and fitness of an insect 2321 species (Ma et al., 2014), and its potential establishment in new environments (Nyamukondiwa 2322 et al., 2010; Buckley & Huey 2016). The thermal tolerance traits commonly measured in 2323 ectotherms vary and may include both lethal and sub-lethal tests (see Sinclair et al., 2015).

2324 Lethal tests that result in the death of the insects include lethal temperature determination, time 2325 to death at set temperatures and supercooling points, dependent on insect freezing strategy 2326 (Formby et al., 2013; Alford et al., 2016). Sub-lethal tests include determination of critical 2327 thermal limits (CTLs) (i.e. the onset of extreme conditions, beyond which an insect’s activity is 2328 compromised), time to heat knock-down and/or recovery after chill coma (Andersen et al., 2015; 2329 Alford et al., 2016). CTLs are considered more ecologically relevant as temperature can be 2330 raised or cooled using rates that mimic field conditions (Rezende et al., 2011; Alford et al., 2016; 2331 Verberk et al., 2016). 2332 The thermal tolerance of insects is not static. Instead, it is plastic and depends on a range 2333 of factors. For example, a prolonged prior exposure to sub-lethal thermal environments 2334 commonly referred to as acclimation, can improve performance in otherwise lethal or stressful 2335 conditions (Kristensen et al., 2008; Nyamukondiwa & Terblanche, 2010; Chidawanyika & 2336 Terblanche, 2011; Andrew et al., 2013). Under even more acute prior exposure (hardening), 2337 similar plastic responses that result in improved survival can also be induced (Yi et al., 2007). 2338 The rate of heating or cooling has also been shown to influence thermal tolerance. In general, 2339 faster rates of temperature change result in poor thermal tolerance due to the limited period for 2340 physiological adjustments by insects (Nyamukondiwa & Terblanche, 2010; Chidawanyika & 2341 Terblanche 2011). Other studies have also shown population differences in insect thermal 2342 tolerance to be influenced by ontogeny (Klockmann et al., 2016) and body size (Cavieres et al., 2343 2016). 2344 Across species, temperature tolerance may differ, depending on the environments in 2345 which they have evolved (Chown & Terblanche, 2007). Even when the species have evolved the 2346 same adaptations to the environment, within generation variations in thermal tolerance may 2347 occur and in all likelihood shape geographic distributions (reviewed in Sgrò et al., 2016). At 2348 much finer geographic scales, changes in fitness among interacting species may disrupt co- 2349 evolved interactions, be they mutual or antagonistic (Parmesan, 2006). As the climate changes, 2350 some species may suffer detrimental effects, while others may emerge as ‘winners’, at least in 2351 the short term. For antagonistic relationships such as host-parasitoid interactions in agriculture, 2352 this is of concern as pest abundance may rise due to reductions in parasitoid abundance. 2353 Understanding the thermal tolerance of pest-parasitoid complexes is therefore important in order

2354 to understand and better predict agricultural pest pressure under changing climates (Kleynhans et 2355 al., 2014). 2356 The spotted stemborer, Chilo partellus (Swinhoe) (Lepidoptera: Crambidae) is one of the 2357 most important insect pests of cereal crops in eastern and southern Africa (Khadioli et al., 2014). 2358 It was accidentally introduced to Africa (Malawi) from Asia (Tams, 1932). Its range has 2359 expanded to successfully establish in high elevation areas, highland tropical areas and humid 2360 transitional regions (Khadioli et al., 2014) and threatens to displace indigenous stemborer species 2361 (Kfir, 1997). Cotesia sesamiae Cameron (Hymenoptera: Braconidae) which is native to Africa, is 2362 a gregarious larval endoparasitoid that attacks a wide range of stemborer hosts, which include the 2363 noctuids Busseola fusca Fuller and Sesamia calamistis Hampson, and crambids Chilo 2364 orichalcocilielus Strand, and C. partellus (Branca et al., 2011). Like other insects, the 2365 distribution of C. sesamiae is inevitably influenced by changes in climate (Mwalusepo et al., 2366 2015). For most studies on thermal tolerance of arthropods of agricultural importance, 2367 investigators mainly focus on the insect pests, without considering their natural enemies (e.g. 2368 Koveos, 2001; Kalosaka et al., 2009; Nyamukondiwa & Terblanche, 2009; Chidawanyika & 2369 Terblanche, 2011; Chanthy et al., 2012). Determining the thermal biology of C. partellus and its 2370 endoparasitoid, C. sesamiae, is therefore an important step in documenting the potential impacts 2371 of a changing climate on host-parasitoid population dynamics. Here, I investigate the thermal 2372 tolerance of C. partellus and its parasitoid, C. sesamiae, and present preliminary host-natural 2373 enemy comparisons. To date, thermal tolerance comparisons between insect pests and their 2374 biocontrol agents are generally lacking. I hypothesize that thermal tolerance of C. partellus and 2375 C. sesamiae were in synchrony and that this host host-parasitoid interaction may evolve together 2376 with thermal variability. Here, I investigated the thermal tolerance of different developmental 2377 stages of C. partellus and C. sesamiae by determining ranges of temperature-time interactions in 2378 terms of lower and upper lethal temperatures, critical limits to activity, supercooling points, chill- 2379 coma recovery time and heat knock-down time. Information on all these is important in 2380 developing pest risk assessments, designing pest management programs and mass-rearing natural 2381 enemies for applied biological control programs. 2382 2383 2384

2385 3.2 Materials and Methods 2386

2387 3.2.1 Insect Culture 2388 The initial colonies of C. partellus and C. sesamiae were obtained from the International 2389 Centre for Insect Physiology and Ecology (ICIPE), Kenya and South African Sugar Research 2390 Institute (SASRI), South Africa, respectively. Chilo partellus was reared on artificial diet 2391 (Ochieng et al., 1985; Tefera et al., 2010), under 12L:12D photoperiod, 28 ± 1°C and 65 ± 10% 2392 relative humidity (RH) in a climate chamber ( HPP 260, Memmert GmbH + Co.KG, Germany). 2393 Pupae were maintained in open Petri dishes in Bugdorm rearing cages (240mm3; Bugdorm- 2394 BD43030F, Megaview Science Co., Ltd, Taiwan) under the same rearing conditions as that of 2395 larvae until adult eclosion. Upon emergence, adults had access to water-soaked cotton wool. 2396 Wax paper was placed in the cages as an oviposition surface for gravid females. To maintain 2397 uniformity among test insects, eggs were harvested after every 12 h, and transferred from wax 2398 paper to artificial diet and allowed to hatch into larvae, subsequently molting five times before 2399 pupation. Pupae were harvested every 24 h and placed in rearing cages for eclosion. Once every 2400 month during summer, wild-caught moths were added to the colony to promote heterozygosity 2401 among traits of interest (e.g. Nyamukondiwa & Terblanche, 2009). For the C. sesamiae colony, 2402 parasitised S. calamistis from SASRI were maintained in the climate chambers under 12L:12D 2403 photoperiod, 28 ± 1°C and 65 ± 10% RH until parasitoid emergence. Emerged parasitoids had 2404 access to food ad libitum (25% honey water from a cotton wick) until they were used in thermal 2405 tolerance assays as 24–48-h-old adults. 2406

2407 3.2.2 Upper and Lower Lethal Temperature Assays 2408 Upper and lower lethal temperatures (ULT and LLT) were assayed as outlined by 2409 Chidawanyika & Terblanche (2011), using a direct plunge protocol (e.g. Terblanche et al., 2008) 2410 in programmable water baths (Systronix, Scientific Engineering (Pty) Ltd, South Africa). ULT 2411 assays ranged 37-48°C and 41-49°C for C. partellus eggs and larvae respectively (for durations 2412 0.5; 1; 2 and 4 h) while that for C. sesamiae adults ranged 36-39°C (for 2 h duration) until 0- 2413 100% mortality was determined. A mixture of propylene glycol and water (1:1 ratio) was used to 2414 enable water baths to operate at sub-zero temperatures without freezing. Age was strictly

2415 controlled in all assays as it has been shown to significantly affect thermal tolerance (Bowler & 2416 Terblanche, 2008; Nyamukondiwa & Terblanche, 2009; Mudavanhu et al., 2014). Ten C. 2417 partellus eggs (black head stage) or larvae (6th instar), as well as C. sesamiae adults (24–48 h 2418 old), were placed in 60 ml polypropylene vials (n = 50) and placed in the water bath for each 2419 temperature/time treatment. Relative humidity in the ULT experiments was maintained using a 2420 wet cotton wick in the vials to avoid desiccation-related mortality. Water bath temperatures were 2421 verified using digital thermometers (Fluke 53/54IIB, Fluke Cooperation, USA) for the duration 2422 of each treatment. Following ULT treatment, vials containing the assayed C. partellus eggs, 2423 larvae and C. sesamiae adults were placed in a climate chamber (28 ± 1°C, 65% ± 10% RH and 2424 12:12 L:D photoperiod) until survival was recorded. Chilo partellus larvae and C. sesamiae 2425 adults had access to food during the ‘recovery’ period. For the purposes of this study, survival 2426 was defined as hatching success (for eggs) and ability to pupate (for larvae). For adults, a 2427 coordinated muscle response to stimuli such as gentle prodding, or normal behaviors such as 2428 feeding, flying or mating (24 h post treatment) symbolized survival. For LLT assays, the same 2429 plunge protocols were followed, as ULTs, with low temperature x time treatment interactions, 2430 before recording survival. Assays ranged -9 to 6°C and -14 to -2°C for C. partellus eggs and 2431 larvae respectively (for durations 0.5; 1; 2 and 4 h) while that for C. sesamiae adults ranged -1 to 2432 4°C (for 2 h duration) until 0-100% mortality was determined. 2433 To determine the ecological relevance of LLTs and ULTs for C partellus and C. sesamiae 2434 recorded above, microclimatic temperatures were taken from a cereal field at Veggies For All 2435 Farm, Glen Valley, Gaborone, Botswana (S24.61224; E25.97326; 977 m.a.s.l) where both host 2436 and parasitoid are present. 2437

2438 3.2.3 Effect of Ramping rate on Critical Thermal Limits (CTLs) 2439 To assess whether heating or cooling rates significantly affected thermal tolerance in C.

2440 partellus and C. sesamiae, CTLs [critical thermal maximum, (CTmax) and critical thermal

2441 minimum, (CTmin)] were assessed following standardized protocols (see Nyamukondiwa & 2442 Terblanche, 2010; Mudavanhu et al., 2014). Only C. partellus larvae (6th instar) and adult moths 2443 (24-48 h old), and C. sesamiae adults (24-48 h old) were used in the experiments. Ten 2444 individuals for each species and life-stage were placed in a series of insulated double-jacketed 2445 chambers (‘organ pipes’) connected to a programmable water bath (Lauda Eco Gold, Lauda

2446 DR.R. Wobser GMBH and Co. KG, Germany) filled with 1:1 water: propylene glycol to allow 2447 for subzero temperatures (Chown & Nicholson, 2004) and subjected to different constant heating 2448 or cooling rates. In all cases, assays started at a set point 28°C (equivalent to the rearing 2449 temperature) for 10 mins (to allow for insects temperature equilibration) before ramping the ° −1 2450 temperature up (CTmax) or down (CTmin) using 3 independent rates (0.12, 0.25 and 0.5 C min ). 2451 This was repeated twice to yield sample sizes of n = 20 per ramping rate treatment (20 2452 replications) for each life-stage and species. A thermocouple (type K 36 SWG) connected to a 2453 digital thermometer (53/54IIB, Fluke Cooperation, USA) was inserted into a control chamber to 2454 record chamber temperature. It was assumed that individual organism body temperature was in 2455 equilibrium with the chamber temperature under the experimental conditions employed, as has 2456 been observed for other similar insect species (Terblanche et al., 2007a; Mudavanhu et al., 2457 2014). CTLs were defined as the temperature at which each individual insect lost coordinated 2458 muscle function which was regarded as a lack of response to mild prodding (e.g. Nyamukondiwa 2459 & Terblanche, 2010). 2460

2461 3.2.4 Supercooling points (SCPs) 2462 The supercooling points of three C. partellus life-stages stages, larvae (6th instar), pupae 2463 and adults (24-48 h old) were assayed as outlined by Nyamukondiwa et al. (2013). To determine 2464 SCPs, sixteen individuals of each developmental stage were individually loaded into 0.65 ml 2465 microcentrifuge tubes. Each insect was placed in contact with the tip of a type-T copper- 2466 constantan thermocouple (762-1121, Cambridge, UK), inserted through the lid of the tube and 2467 both the insect and thermocouple were held in place using cotton wool. Thermocouples were 2468 connected to one of two 8-channel Picotech TC-08 (Pico Technology, Cambridge, UK) 2469 thermocouple interfaces and temperatures were recorded at 1s intervals using PicoLog software 2470 for windows (Pico Technology, Cambridge, UK). Experiments began by holding individual 2471 insects at 15°C for 10 mins (to allow equilibration of insect body temperatures) before ramping 2472 down at 0.5°C min-1 until SCPs were recorded. SCP for each individual was determined as the 2473 lowest temperature recorded prior to a spike in temperature associated with the latent heat of 2474 crystallization (Nyamukondiwa et al., 2013). 2475 2476

2477 3.2.5 Heat knock-down time (HKDT) and Chill-coma recovery time (CCRT) 2478 HKDT and CCRT were assayed on C. partellus larvae (6th instar) and adults (24-48 h 2479 old) as outlined in Weldon et al. (2011). For HKDT, 10 replicate organisms were individually 2480 placed in 0.65 ml microcentrifuge tubes and were placed in a climate chamber connected to a 2481 camera (HD Covert Network Camera, DS-2CD6412FWD-20, Hikvision Digital Technology Co., 2482 Ltd, China) that was linked to a computer. The organisms were then exposed to a test 2483 temperature of 49.0±0.3°C and 65% RH in the climate chamber. This knockdown temperature 2484 (49.0°C) was chosen based on preliminary investigations of upper critical temperatures to 2485 activity, which ranged 47.88±0.76°C for adults and 48.5±0.74°C for larvae. This was repeated 2486 three times to yield sample sizes of n = 30 for each developmental stage. All observations were 2487 recorded from the climate chamber video recording. HKDT was defined as the time (in minutes) 2488 at which activity was lost by the organisms following exposure to 49.0°C in a climate chamber. 2489 For CCRT, 10 replicate insects (6th instar larvae) and (24-48 h old moths) were individually 2490 placed in 0.65 ml microcentrifuge tubes and then loaded into a large zip-lock bag which was 2491 submerged into a water bath (Systronix, Scientific, South Africa) filled with 1:1 water: propylene 2492 glycol set at 0°C for 1 hour. This temperature x time treatment has been shown to elicit chill- 2493 coma in other insect taxa (see Weldon et al., 2011; Sinclair et al., 2015). Following 1 h at chill- 2494 coma temperature, the tubes were removed from the water bath and immediately placed in a 2495 climate chamber (Memmert HPP 260, Memmert GmbH + Co.KG, Germany) set at 28°C, 65% 2496 RH for recovery. The chamber was connected to a camera (as in HKDT) that was linked to a 2497 computer where observations of CCRT were recorded. This was also repeated three times to 2498 yield sample sizes of n = 30 per treatment. CCRT was defined as the time (in minutes) required 2499 to regain coordinated movement ability (in larva) to stand upright on its legs (for adult moths) 2500 (Milton & Partridge, 2008). 2501

2502 3.2.6 Statistical analyses 2503 Data analyses were carried out in STATISTICA, version 13.0 (Statsoft Inc., Tulsa, 2504 Oklahoma) and R version 3.3.0 (R Development Core Team, 2016). Lethal temperature assays 2505 (upper and lower lethal limits), SCPs, HKDT and CCRT were analysed using generalized linear 2506 models (GLZ) assuming a binomial (for lethal temperature assays) and Gaussian (SCPs, HKDT 2507 and CCRT) distribution and a logit link function in R statistical software. CTLs met the linear

2508 model assumptions of constant variance and normal errors, so the effects of ramping rate on 2509 CTLs were analysed using one-way ANOVA in STATISTICA. Tukey-Kramer’s post-hoc tests 2510 were used to separate statistically heterogeneous groups. 2511

2512 3.3 Results 2513

2514 3.3.1 Upper and Lower Lethal Temperature Assays 2515 An increase in the severity of temperature exposure at both low and high temperatures 2516 resulted in increased egg (Table 3.1; Figure 3.1) and larval (Table 3.2; Figure 3.1) mortalities. In 2517 addition, an increase in duration of exposure at any lethal temperature resulted in increased 2518 mortality of the eggs (Figure 3.1A; 3.1D) and larvae (Figure 3.1B; 3.1D). The interaction of 2519 temperature and duration of exposure for C. partellus eggs was highly significant at both low and 2520 high temperature (Table 3.1). Similarly, temperature by duration of exposure interaction effects 2521 for C. partellus larvae were highly significant for LLT and marginally non-significant for ULT 2522 (Table 3.2). 2523 Temperature significantly affected the survival of C. sesamiae adults at both low and 2524 high temperature (LLT: χ2 =180.05, d.f =4, P ˂ 0.0001; ULT: χ2 =162.78, d.f =3, P ˂ 0.0001). 2525 An increase in severity and duration of both low and high thermal exposure generally resulted in 2526 increased C. sesamiae adult mortality (Figure 3.2). By comparison of the 2 h stress exposure, 2527 host C. partellus LLTs varied from -9 to 3°C (Figure 3.1A; B) while ULTs ranged 37 to 46°C 2528 (Figure 3.1C; D), compared with C. sesamiae LLT range of -1 to 4°C (Figure 3.2A) and ULT 2529 range of 36 to 39°C (Figure 3.2B). 2530 2531 2532 2533 2534 2535 2536

2537 Table 3.1: Summary statistical results of the effects of temperature, duration of exposure and 2538 their interactions on the survival of Chilo partellus eggs following lower and upper lethal 2539 temperature treatments. Analysis were done using generalized linear models (GLZ) assuming 2540 binomial distribution with a logit link function in R version 3.3.0. 2541 Parameter d.f χ2 P-Value Lower lethal temperature Duration 3 154.57 ˂0.0001 Temperature 8 1348.59 ˂0.0001 Temperature * Duration 24 74.27 ˂0.0001

Upper lethal temperature Duration 3 143.79 ˂0.0001 Temperature 7 1252.86 ˂0.0001 Temperature * Duration 21 58.69 ˂0.0001 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556

2557 Table 3.2: Summary statistical results of the effects of temperature, duration of exposure and 2558 their interactions on the survival of Chilo partellus larvae following lower and upper lethal 2559 temperature treatments. Analysis were done using generalized linear models (GLZ) assuming 2560 binomial distribution with a logit link function in R version 3.3.0. 2561 Parameter d.f χ2 P-Value Lower lethal temperature Duration 3 169.16 ˂0.0001 Temperature 6 1143.82 ˂0.0001 Temperature * Duration 18 33.73 0.0135

Upper lethal temperature Duration 3 361.18 ˂0.001 Temperature 6 1217.68 ˂0.001 Temperature * Duration 18 28.10 0.0605 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576

120 A 100

80 0.5hr 1hr 2hr 60 4hr

Survival (%) Survival

-9 -8 -7 -5 -4 -1 2 5 6 Temperature (oC) 2577 2578 2579

120 B 100

80 0.5hr 1hr 2hr 60 4hr

Survival (%) Survival

-14 -13 -11 -9 -7 -4 -2 Temperature (oC) 2580 2581 2582

120 C

100

Survival (%) Survival 0.5hr 20 1hr 2hr 4hr 0

37 38 40 42 43 46 47 48 Temperature (oC) 2583 2584

120 D

100

% Survival % 40 0.5hr 1hr 20 2hr 4hr 0

41 44 45 46 47 48 49 Temperature (oC) 2585 2586 2587 Figure 3.1: Mean (±95% confidence interval) survival of Chilo partellus (A) eggs and (B) larvae 2588 at different low temperatures, and (C) eggs and (D) larvae at different high temperatures each 2589 applied over four different exposure times

A 100

Survival (%) Survival

-1 0 1 3 4 Temperature (oC) 2590

2591

B 100

Survival (%) Survival

36 37 38 39 Temperature (oC) 2592

2593 Figure 3.2: Mean (±95% confidence interval) survival of Cotesia sesamiae adults at different 2594 low (A) and high (B) temperatures applied for two hours. 2595

2596 3.3.2 Effect of Ramping rate on Critical Thermal Limits (CTLs)

2597 Ramping rate significantly affected CTmin in C. partellus adults (F2.57 =45.79, P ˂ 0.0001)

2598 (Figure 3.3B), larvae (F2.57 =95.72, P ˂ 0.0001) (Figure 3.3D) and C. sesamiae adults (F2.57 2599 =232.81, P ˂ 0.0001) (Figure 3.3F). Faster cooling rates compromised adult C. partellus and C.

2600 sesamiae CTmin (Figure 3.3A; F) but enhanced CTmin in C. partellus larvae (Figure 3.3D).

2601 Similarly, ramping rate also affected CTmax. Faster ramping rates significantly improved CTmax

2602 in C. partellus adults (F2.57 =61.7, P ˂ 0.0001) (Figure 3.3A) and larvae (F2.57 =59.73 P ˂ 0.0001)

2603 (Figure 3.3C), and adult C. sesamiae (F2.57 =533.04, P ˂ 0.0001) (Figure 3.3E). 2604 2605

50 6 A B a

o a

o 49 5 a

48 b 4

b 47 3 c

Critical thermal thermal minimaCritical (

Critical thermal maxima ( thermal maxima Critical

46 0.12 0.25 0.5 2 0.12 0.25 0.5 o o 2606 Ramping rate ( C/minute) Ramping rate ( C/minute)

50 9.0 C D c

C) a 49 o 8.5

b b 48 8.0

7.5 47 a c

Critical thermal minima (

Critical thermal maximaCritical ( 7.0 46 0.12 0.25 0.5 0.12 0.25 0.5 o Ramping rate ( C/minute) o 2607 Ramping rate ( C/minute)

E 3.6 F 46 c

C) c

b o C)

44 3.0

b 42 2.4

40 b 1.8 a

Critical thermal maxima ( a minimathermal ( Critical 38 1.2 0.12 0.25 0.5 0.12 0.25 0.5 o o a 2608 Ramping rate ( C/minute) Ramping rate ( C/minute) a 2609

2610 Figure 3.3: Effects of experimental heating and cooling rates on Chilo partellus adult (A) CTmax

2611 and (B) CTmin, larval (C) CTmax and (D) CTmin, and Cotesia sesamiae adult (E) CTmax and (F)

2612 CTmin. (mean± 95% confidence limits). n = 20 per group. Means with the same letter are not 2613 significantly different from each other.

2614 3.3.3 Supercooling Points (SCPs) 2615 The mean supercooling points for larvae, pupae and adults were -11.82±1.78, - 2616 10.43±1.73 and -15.75±2.47 respectively. There was no significant difference in C. partellus 2617 SCPs for larvae and pupae (χ2 =59.34, d.f =2, P > 0.05) (Figure 3.4) However, adult C. partellus 2618 had significantly enhanced supercooling points compared with its larvae and pupae (χ2 =59.34, 2619 d.f =2, P ˂ 0.0001) (Figure 3.4).

-6 a a -8

-10 b -12

-14

-16

Supercooling point Supercooling (°C) -18

-20 Larvae Pupae Adults Developmental stage 2620 2621 2622 Figure 3.4: The effect of developmental stage on supercooling points in Chilo partellus larvae, 2623 pupae and adults. Error bars represent 95% CLs (N = 16 per group). Means with the same letter 2624 are not significantly different from each other. 2625

2626 3.3.4 Heat knock-down time and Chill-coma recovery time 2627 Heat knock-down time varied significantly between C. partellus larvae and adults (χ2 2628 =352.67, d.f =1, P ˂ 0.0001) (Figure 3.5A). Larvae showed a significantly longer time to knock- 2629 down compared to the adults (Figure 3.5A). Similarly, there was a significant difference in

2630 CCRT between C. partellus larvae and adults (χ2 =303.02, d.f =1, P ˂ 0.0001) (Figure 3.5B). 2631 Contrary to HKDT, adults recovered faster than larvae following chill-coma (0°C; 1 h) (Figure 2632 3.5B).

18 a A

14 b

Heat knock-down time (minutes)

6 Larv ae Adults Developmental stage 2633 2634

6 B a

3 b

Chill coma recovery time (minutes) 0 Larvae Adults Developmental stage 2635 2636 2637 Figure 3.5: The effect of developmental stage (larvae vs adults) on Chilo partellus (A) heat 2638 knock-down time and (B) chill-coma recovery time. Error bars represent 95% CLs (N = 30). 2639 Means with the same letter are not significantly different from each other.

C) o

Temperature ( Temperature 10

-10

1/1/2016

2/1/2016

3/1/2016

4/1/2016

5/1/2016

6/1/2016

7/1/2016 12/1/2015 Time (Days) 2640 2641 Figure 3.6: Microclimatic temperature data in austral summer (December 2015-April 2016) and 2642 winter (May-July 2016) were recorded using Thermocron iButtons (Dallas Semiconductors, 2643 Model DS1920) (0.5°C accuracy; 1 h sampling frequency) from a maize field, Veggies For All 2644 Farm, Glen Valley, Gaborone, Botswana (S24.61224; E25.97326; 977 m.a.s.l), where both Chilo 2645 partellus and C. sesamiae are found. 2646

2647 3.4 Discussion 2648 Enhanced basal and plastic responses by invasive species to abiotic stressors are key 2649 physiological factors contributing to their invasion success (Slabber et al., 2007; Nyamukondiwa 2650 & Terblanche, 2010). The interactions between temperature and RH tolerance are likely to shape 2651 the survival of insects, and indeed C. partellus and its larval parasitoid, C. sesamiae. To my 2652 knowledge, this is the first report detailing comparative thermal tolerance of C. partellus and C. 2653 sesamiae, and how this may shape future biological control in the face of climate change.

2654 Consequently, the thermal tolerance of C. partellus and C. sesamiae are in asynchrony revealing 2655 that the host and the parasitoid may not evolve together with thermal variability. 2656 As expected, the current results showed that the severity and duration of exposure were 2657 important in determining the survival of C. partellus and C. sesamiae, in keeping with previous 2658 studies (Chown & Terblanche, 2007; Chidawanyika & Terblanche, 2011). LLTs and ULTs for 2659 C. partellus eggs and larvae ranged from -9 to 6°C; 37 to 48°C and -14 to -2°C; 41 to 49°C 2660 respectively for 0.5–4 h treatments, revealing that larvae were more temperature tolerant than the 2661 eggs. This is contrary to previous studies, highlighting that mobile insect developmental stages 2662 may require less physiological protection against environmental stressors, as their mobility, may 2663 employ behavioral compensatory mechanisms in avoiding extreme conditions (Jensen et al., 2664 2007; Blanckenhorn et al., 2014; Abbar et al., 2016). Nevertheless, the current results are also in 2665 keeping with other previous studies documenting developmental stage related differences in 2666 environmental stress tolerance (Cavieres et al., 2016; Klockmann et al., 2016). Indeed, Zhang et 2667 al. (2015) showed that diamondback moth (Plutella xylostella) eggs had compromised high 2668 temperature tolerance, relative to its larval stages. For C. sesamiae, LLTs ranged from -1 to 4°C 2669 while ULTs ranged from 36 to 39°C (2 h treatments). A comparison of LLTs and ULTs for C. 2670 partellus and C. sesamiae (for the 2 h duration) indicated that the parasitoid is more vulnerable 2671 to temperature stress at both temperature extremes (see Figure 3.1; 3.2). Larval C. partellus is 2672 known to be the economically destructive stage, and is also the target by endoparasitoid C. 2673 sesamiae. Differential high and low temperature tolerance therefore suggests a mismatch 2674 between the host and its parasitoid in the face of climate change. LLT and ULT data gathered 2675 here is important in determining thermal windows for the performance, fitness and survival of 2676 these species. High and low temperatures are known to limit flight, mating, reproduction, 2677 development (measured as accumulated degree days) and other biological processes that affect 2678 insect phenology (Chidawanyika & Terblanche, 2011; Nyamukondiwa et al., 2013). With 2679 African temperatures expected to increase by 1.5 - 3°C by 2050 (Gemeda & Sima, 2015), 2680 coupled with increased likelihood of prolonged heat waves, and cold snaps (IPCC, 2014), it may 2681 be expected that host stemborer C. partellus may likely survive both prolonged high and low 2682 temperatures better than its parasitoid. In Africa, this represents a potential risk in C. partellus 2683 pest outbreaks, which may negatively affect cereal crop productivity, food and nutrition security, 2684 and ultimately household incomes. Microclimate data recorded here showed that both low and

100

2685 high temperatures causing C. partellus and C. sesamiae mortality are frequently encountered 2686 under field conditions (Figure 3.6). 2687 This may imply that, in the absence of compensatory mechanisms, temperature associated 2688 mortality may be a major factor shaping the population dynamics of the pest-parasitoid complex 2689 in nature. 2690 The present study also demonstrates the significant effects of ramping rates on CTLs. 2691 This is in agreement with Terblanche et al. (2007b) who indicated that CTLs highly depend on 2692 methodological context. CTLs significantly vary with the experimental protocol vis-à-vis starting 2693 temperature, ramping rate and experimental duration. Although CTLs are widely accepted as 2694 important indices of thermal tolerance (Wang et al., 2002; Nyamukondiwa & Terblanche, 2010), 2695 few studies specifically examined these responses in an attempt to gain insight into ecologically 2696 relevant estimates of thermal tolerance under laboratory conditions (Slabber, 2007; 2697 Nyamukondiwa & Terblanche, 2010). Here, I demonstrate that slower ramping rates

2698 significantly reduced CTmax in C. partellus adults, larvae and C. sesamiae adults. This suggests 2699 that a prolonged heating period, even under dynamic thermal variability, may compromise 2700 fitness of organisms due to the longer period under stress. In line with the notion that mortality 2701 at a stressful temperature depends on temperature severity and the duration of exposure (Rezende 2702 et al., 2011; 2014), my results suggest that C. partellus may not be able to adjust its phenotypic 2703 plasticity of high temperature tolerance in the short term (see discussions in Nyamukondiwa & 2704 Terblanche, 2010; Sgrò et al., 2016). Furthermore, slower cooling rates significantly improved

2705 CTmin in adult C. partellus and C. sesamiae but not in C. partellus larvae. This suggests that 2706 larval C. partellus may not have the capacity to adjust its lower temperature tolerance through 2707 phenotypic plasticity, similar to the rapid cold hardening (RCH) response in other insects (see

2708 Bale et al., 1989; Kelty & Lee, 1999). However, improved CTmin following slower ramping rates 2709 in adult C. partellus and C. sesamiae suggests that prolonged duration at low temperatures may 2710 allow the insects to develop some low temperature protection, which suggest RCH. Such

2711 ecologically relevant cooling rates have been observed to improve CTmin in many Diptera 2712 [Bactrocera tryoni (Meats, 1973), Musca domestica (Coulson & Bale, 1990), Eurosta solidaginis 2713 (Lee et al., 1993), D. melanogaster (Kelty & Lee, 1999; Jensen et al., 2007), Sarcophaga 2714 crassipalpis (Rinehart et al., 2000), Bactrocera oleae (Koveos, 2001), Ceratitis (Nyamukondiwa 2715 & Terblanche, 2010)] and the lepidopteran, Cydia pomonella (Chidawanyika & Terblanche,

101

2716 2011). A comparison of upper temperature limits to C. partellus and C. sesamiae activities 2717 (Figure 3.3A; C; E) suggest that C. sesamiae is likely to cease activity at mild temperatures 2718 relative to the host. This disadvantage may compromise the efficacy of the parasitoid as a 2719 biological control agent in the face of climate change. However, the current comparative 2720 assessment of thermal tolerance in C. partellus vs C. sesamiae is based on evaluation of LLTs, 2721 ULTs and CTLs. While these are reliable means of measuring thermal tolerance (Chown & 2722 Nicolson, 2004; Andersen et al., 2015), I suggest further investigations on other thermal biology 2723 life history traits including whether the parasitoid exhibits improved performance under stressful 2724 conditions through phenotypic plasticity and laboratory adaptation. In addition, an understanding 2725 of diurnal thermal activity of the parasitoid in relation to host should also be investigated in the 2726 face of climate change. 2727 The mean supercooling points for larvae, pupae and adults of C. partellus were - 2728 11.82±1.78, -10.43±1.73 and -15.75±2.47 respectively. Previous studies on Ceratitis sp revealed 2729 immobile pupae had the lowest SCP as compared to the mobile developmental stages, larvae and 2730 adults (Nyamukondiwa et al., 2013). Indeed, from an evolutionary perspective, ‘immobile’ 2731 developmental stages (eggs, and pupae) are almost always confined to a specific habitat, and are 2732 less likely capable of compensating to environmental stress using behaviour, e.g. escape. As a 2733 result, these ‘immobile’ stages often have evolved enhanced inherent basal temperature tolerance 2734 (see Bowler & Terblanche, 2008; Lansing et al., 2000). The SCP results are, however, in 2735 disagreement with that since adult C partellus moths had the lowest SCPs (most enhanced). 2736 Overwintering insects are divided into two main categories basing on their ability to survive the 2737 freezing of their body water. Insects that can survive ice formation within their tissues are called 2738 freeze tolerant whereas those that cannot survive ice formation within their tissues are called 2739 freeze susceptible or freeze intolerant (Andreadis et al., 2005). By contrast, chill-susceptible 2740 insects are killed by the cold in the absence of internal ice formation (Sinclair et al., 2015). In my 2741 study, there were no insects that recovered from chill-coma following SCP measurement, 2742 regardless of developmental stage. This therefore suggests that C. partellus may be chill- 2743 susceptible. Lower SCPs of adult C. partellus may thus be ecologically not that significant since 2744 the species may not be freeze tolerant, and moreover, LLTs were much higher than the freezing 2745 temperature (Figure 3.1A; B) (Sinclair et al., 2015). Furthermore, microclimate data recorded 2746 here were above the SCPs recorded for the species (Figure 3.6). Generally, for many insects,

102

2747 cold tolerance strategy is derived from separate measurement of SCP and LLT; where LT50 ˂

2748 LLT = freeze tolerant, LT50 = LLT = freeze intolerant/avoidant and LT50 > LLT = chill- 2749 susceptible (Bemani et al., 2012; Sinclair et al., 2015). This study therefore shows that all C. 2750 partellus developmental stages are likely killed by a prolonged low temperature exposure and 2751 thus chill susceptible. 2752 HKDT and CCRT measures the time required for enzyme activity and/or membrane 2753 permeability to be lost or return to levels sufficient for neuromuscular function, respectively 2754 (MacMillan & Sinclair, 2011a). Since chill-coma is reversible and locomotion easy to detect 2755 visually in mobile insects, chill-coma recovery is often used as an ecologically sound measure of 2756 cold tolerance (Sinclair et al., 2015). As in previous studies (Cavieres et al., 2016; Klockmann et 2757 al., 2016), the current results show that C. partellus HKDT and CCRT were significantly 2758 affected by developmental stage. Larvae took more time to be knocked down by high 2759 temperature relative to adults, suggesting an enhanced temperature tolerance, for the trait of 2760 HKDT. The reason why larvae were more heat tolerant in this case is unknown but may be 2761 related to body water content. Larvae generally have higher water content than adults and thus 2762 may have high dehydration tolerance and low rates of water loss (Yoder et al., 2003). This is in 2763 agreement with comparison of water loss rate for larvae and adults of Belgica antarctica in dry 2764 air which revealed that larvae had a higher dehydration tolerance than adults (Benoit et al., 2765 2007). These suggest that organisms with less body water may be more susceptible to 2766 temperature and desiccation related mortality than organisms with high body water content (Gray 2767 & Bradley, 2005). In most insect species, high levels of internal body water usually extend the 2768 time that is required for dehydration to levels that elicit mortality. In chill-susceptible insects, 2769 exposure to chill-coma temperature causes the bulk movement of hemolymph sodium, 2770 magnesium, calcium and water into the alimentary canal (MacMillan & Sinclair, 2011b) and 2771 chill-coma recovery usually involves the re-establishment of hemolymph ion homeostasis 2772 (MacMillan et al., 2012). Chilo partellus larvae took more time to recover from chill-coma as 2773 compared to the adults indicating a higher cold tolerance for the adults for traits of CCRT. This 2774 suggests developmental stage related differences in maintenance and re-establishment of ion 2775 homeostasis, essential for chill-coma recovery. Indeed, similar studies have indicated 2776 interspecific variation in maintenance of water and ion homeostasis accounts for differences in 2777 Drosophila species cold tolerance in a laboratory (MacMillan et al., 2015) and field setting

103

2778 (MacMillan et al., 2016). This therefore shows that the ability to maintain ion homeostasis is 2779 significant for insects in the wild to cope with the diurnal temperature fluctuations involving 2780 chilling (see e.g. Figure 3.6). Nevertheless, developmental stage related differences in C. 2781 partellus CCRT and ion homeostasis recorded here still warrants further investigation. 2782 In conclusion, results of the present study document thermal tolerance in C. partellus and 2783 its larval endoparasitoid, C. sesamiae, and implications for biological control in the face of 2784 climate change. To my knowledge, this is the first comparative study detailing the thermal 2785 tolerance between this host and parasitoid. The current results show that C. partellus has a high 2786 basal thermal tolerance as compared to C. sesamiae. This implies that for traits of thermal 2787 tolerance, the two species may not be capable of co-evolving, which is an undesirable 2788 characteristic for interacting species. With global temperatures expected to gradually increase, 2789 this suggests that C. partellus can inhabit slightly warmer, and cooler climates where C. 2790 sesamiae may not thrive. This represents significant potential burden to biological control 2791 programmes of the economic pest, with implications on food security. Furthermore, I also 2792 document developmental stage differences in high and low temperature tolerance for C. 2793 partellus. Overall, this work provides insight into how thermal biology affects the population 2794 dynamics of pest-parasitoid interactions and is thus useful for model-based predictions of pest 2795 distribution and population dynamics, pest risk analysis and pest management. 2796

2797 References 2798 Abbar, S., Schilling, M.W. & Phillips, T.W. (2016) Time–mortality relationships to 2799 control Tyrophagus putrescentiae (Sarcoptiformes: Acaridae) exposed to high and low 2800 temperatures. Journal of Economic Entomology, 109, 2215-2220. 2801 Alford, L., Burel, F. & van Baaren, J. (2016) Improving methods to measure critical thermal 2802 limits in phloem-feeding pest insects. Entomologia Experimentalis et Applicata, 159, 61-69. 2803 DOI: 10.1111/eea.12410. 2804 Andreadis, S.S., Milonas, P.G. & Savopoulou-Soultani, M. (2005) Cold hardiness of diapausing 2805 and non-diapausing pupae of the European grapevine moth, Lobesia botrana. Entomologia 2806 Experimentalis et Applicata, 117, 113-118.

104

2807 Andrew, N.R. (2013) Population dynamics of insects: impacts of a changing climate. In: Rohde 2808 K (ed) The Balance of Nature and Human Impact, Cambridge Cambridge University Press, 2809 pp 311-323. 2810 Andrew, N.R., Hart, R.A., Jung, M.P., Hemmings, Z. & Terblanche, J.S. (2013) Can temperate 2811 insects take the heat? A case study of the physiological and behavioural responses in a 2812 common ant, Iridomyrmex purpureus (Formicidae), with potential climate change. Journal of 2813 Insect Physiology, 59, 870-880. doi: 10.1016/j.jinsphys.2013.06.003. 2814 Andrew, N.R. & Terblanche, J.S. (2013) The response of insects to climate change. In: Salinger J 2815 (ed) Living in a Warmer World: How a changing climate will affect our lives. David 2816 Bateman Ltd Auckland, pp 38-50. 2817 Andersen, J.L., Manenti, T., Sørensen, J.G., MacMillan, H.A., Loeschcke, V. & Overgaard, J. 2818 (2015) How to assess Drosophila cold tolerance: chill coma temperature and lower lethal 2819 temperature are the best predictors of cold distribution limits. Functional Ecology, 29, 55–65. 2820 Angilletta Jr, M.J. (2009) Thermal Adaptation. A Theorelatical and Empirical Synthesis. Oxford 2821 University Press, New York. 2822 Bale, J.S., Hansen, T.N., Nishino, M. & Baust, J.G. (1989) Effect of cooling rate on the survival 2823 of larvae, pupariation and adult emergence of the gallfly Eurosta solidaginis. Cryobiology, 2824 26, 285-289. 2825 Bazinet, A.L., Marshall, K.E., Macmillan, H.A., Williams, C.M. & Sinclair, B.J. (2010) Rapid 2826 changes in desiccation resistance in Drosophila melanogaster are facilitated by changes in 2827 cuticular permeability. Journal of Insect Physiology, 56, 2006–2012. 2828 Bemani, M., Izadi, H., Mandian, K., Khani, A. & Samih, M.A. (2012) Study on the physiology 2829 of diapause, cold hardiness and supercooling point of overwintering of the pistachio fruit hull 2830 borer, Arimania comaroffi. Journal of Insect Physiology, 58, 897-902. 2831 Benoit, J.B., Lopez-Martinez, G., Michaud, M.R., Elnitsky, M.A., Lee Jr, R.E. & Denlinger, D. 2832 (2007) Mechanisms to reduce dehydration stress in larvae of the Antarctic midge, Belgica 2833 antarctica. Journal of Insect Physiology, 53, 656–667. 2834 Blanckenhorn, W.U., Gautier, R., Nick, M., Puniamoorthy, N. & Schäfer, M.A. (2014) Stage- 2835 and sex-specific heat tolerance in the yellow dung fly Scathophaga stercoraria. Journal of 2836 Thermal Biology, 46, 1-9. doi: 10.1016/j.jtherbio.2014.09.007.

105

2837 Bowler, K. & Terblanche, J.S. (2008) Insect thermal tolerance: what is the role of ontogeny, 2838 ageing and senescence? Biological Reviews, 83, 339–355. 2839 Branca, A., Le Ru, B.P., Vavre, F., Silvain, J.F. & Dupas, F. (2011) Intraspecific specialization 2840 of the generalist parasitoid Cotesia sesamiae revealed by polyDNAvirus polymorphism and 2841 associated with different Wolbachia infection. Molecular Ecology, 20, 959-971. doi: 2842 10.1111/j.1365-294X.2010.04977.x. 2843 Bubliy, O.A., Kristensen, T.N., Kellermann, V. & Loeschcke, V. (2012) Plastic responses to four 2844 environmental stresses and cross-resistance in a laboratory population of Drosophila 2845 melanogaster. Functional Ecology, 26, 245–253. 2846 Buckley, L.B. & Huey, R.B. (2016) How extreme temperatures impact organisms and the 2847 evolution of their thermal tolerance. Integrative and Comparative Biology, 56, 98-109. 2848 doi:10.1093/icb/icw004. 2849 Cavieres, G., Bogdanovich, J.M. & Bozinovic, F. (2016) Ontogenetic thermal tolerance and 2850 performance of ectotherms at variable temperatures. Journal of Evolutionary Biology, 29, 2851 1462-1468. 2852 Chanthy, P., Martin, B., Gunning, R. & Andrew, N.R. (2012) The effects of thermal acclimation 2853 on lethal temperatures and critical thermal limits in the green vegetable bug, Nezara viridula 2854 (L) (Hemiptera: Pentatomidae). Frontiers in Physiology, 3, 465: doi: 2855 10.3389/fphys.2012.00465. 2856 Chidawanyika, F. & Terblanche, J.S. (2011) Rapid thermal responses and thermal tolerance in 2857 adult codling moth Cydia pomonella (Lepidoptera: Totricidae). Journal of Insect Physiology, 2858 57, 108- 117. 2859 Chown, S.L. & Nicolson, S.W. (2004) Insect Physiological Ecology. Mechanisms and Patterns. 2860 Oxford University Press, Oxford. 2861 Chown, S.L. & Terblanche, J.S. (2007) Physiological Diversity in Insects: Ecological and 2862 Evolutionary Contexts. Advances in Insect Physiology, 33, 50-152. 2863 Coulson, S.J. & Bale, J.S. (1990) Characterisation and limitations of the rapid cold-hardening 2864 response in the housefly Musca domestica (Diptera: Muscidae). Journal of Insect Physiology, 2865 36, 207-211. 2866 Dillon, M.E., Wang, G., Garitty, P.A. & Huey, R.B. (2009) Thermal preference in Drosophila. 2867 Journal of Thermal Biology, 34, 109-119.

106

2868 Formby, J.P., Krishnan, N. & Riggins, J.J. (2013) Supercooling in the redbay ambrosia beetle 2869 (Coleoptera: Curculionidae). Florida Entomologist, 96, 1530–1540. 2870 Gemeda, D.O. & Sima, A.D. (2015) The impacts of climate change on African continent and the 2871 way forward. Journal of Ecology and the Natural Environment, 7(10), 256-262. 2872 Gray, E.M. & Bradley, T.J. (2005) "Physiology of desiccation resistance in Anopheles gambiae 2873 and Anopheles arabiensis". American Journal of Tropical Medicine and Hygiene, 73(3), 2874 553–559. 2875 Huey, R., Carlson, M., Crozier, L., Frazier, M., Hamilton, H., Harley, C., Hoang, A. & 2876 Kingsolver, J. (2002) Plant versus animals: Do they deal with stress in different ways? 2877 Integrative and Comparative Biology, 42(3), 415-423. 2878 IPCC. (2014) Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and 2879 III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Core 2880 Writing Team, R.K. Pachauri and L.A. Meyer (eds.)]. IPCC, Geneva, Switzerland, pp 151. 2881 Jensen, D., Overgaard, J. & Sørensen, J.G. (2007) The influence of developmental stage on cold 2882 shock resistance and the ability to cold-harden in Drosophila melanogaster. Journal of Insect 2883 Physiology, 53, 179-186. 2884 Kalosaka, K., Soumaka, E., Politis, N. & Mintzas, A.C. (2009) Thermotolerance and Hsp70 2885 expression in Mediterranean fruit fly Ceratitis capitata. Journal of Insect Physiology, 55, 2886 568-573. 2887 Karsten, M., van Vuuren, B.J., Addison, P. & Terblanche, J.S. (2015) Deconstructing 2888 intercontinental invasion pathway hypotheses of the Mediterranean fruit fly (Ceratitis 2889 capitata) using a Bayesian inference approach: are port interceptions and quarantine 2890 protocols successfully preventing new invasions? Diversity and Distributions, 21, 813-825. 2891 Kelty, J.D. & Lee, R.E. Jr. (1999) Induction of rapid cold hardening by cooling at ecologically 2892 relevant rates in Drosophila melanogaster. Journal of Insect Physiology, 45, 719-726. 2893 Kfir, R. (1997) Competitive displacement of Busseola fusca (Lepidoptera: Noctuidae) by Chilo 2894 partellus (Lepdoptera: Pyralidae). Annals of the Entomological Society of America, 90, 619- 2895 624. 2896 Khadioli, N., Tonnang, Z.E.H., Muchugu, E., Ong’amo, G., Achia, T., Kipchirchir, I., Kroschel, 2897 J. & Le Ru, B. (2014) Effect of temperature on the phenology of Chilo partellus (Swinhoe) 2898 (Lepidoptera: Crambidae), simulation and visualization of the potential future distribution of

107

2899 C. partellus in Africa under warmer temperatures through the development of life-table 2900 parameters. Bulletin of Entomological Research, 104, 809–822. 2901 Kleynhans, E., Conlong, D.E. & Terblanche, J.S. (2014) Host plant-related variation in thermal 2902 tolerance of Eldana saccharina. Entomologia Experimentalis et Applicata, 150, 113-122. 2903 DOI: 10.1111/eea.12144. 2904 Klockmann, M., Gunter, F. & Fischer K. (2016) Heat resistance throughout ontogeny: body size 2905 constraints thermal tolerance. Global Change Biology, 23, 686-696. DOI: 10.1111/gcb.13407 2906 Kristensen, T.N., Hoffmann, A.A., Overgaard, J., Sørensen, J.G. & Hallas, R. (2008) Costs and 2907 benefits of cold acclimation in field released Drosophila. Proceedings of the National 2908 Academy of Science of the United States of America, 105, 216-221. 2909 Kührt, U., Samietz, J., Höhn, H. & Dorn, S. (2006) Modelling the phenology of codling moth: 2910 influence of habitat and thermoregulation. Agriculture, Ecosystems & Environment, 117, 29- 2911 38. 2912 Lansing, E., Justesen, J. & Loeschcke, V. (2000) Variation in the expression of HSP70, the 2913 major heat-shock protein, and thermotolerance in larval and adult selection lines of 2914 Drosophila melanogaster. Journal of Thermal Biology, 25, 443-450. 2915 Lee, R.E., McGrath, J.J., Morason, R.T. & Taddeo, R.M. (1993) Survival of intracellular 2916 freezing, lipid coalescence and osmotic fragility in fat body cells of the freeze-tolerant gallfly 2917 Eurosta solidaginis. Journal of Insect Physiology, 39, 445-450. 2918 Koveos, D.S. (2001) Rapid cold hardening in the olive fruit fly Bactrocera oleae under 2919 laboratory conditions. Entomologia Experimentalis et Applicata, 101, 257-263. 2920 MacMillan, H.A. & Sinclair, B.J. (2011a) Mechanisms underlying insect chill-coma. Journal of 2921 Insect Physiology, 57, 12-20. 2922 MacMillan, H.A. & Sinclair, B.J. (2011b) The role of the gut in insect chilling injury: cold- 2923 induced disruption of osmoregulation in the fall field cricket, Gryllus pennsylvanicus. 2924 Journal of Experimental Biology, 214, 726-734; doi: 10.1242/jeb.051540. 2925 MacMillan, H.A., Williams, C.M., Staples, J.F. & Sinclair, B.J. (2012) Reestablishment of ion 2926 homeostasis during chill-coma recovery in the cricket Gryllus pennsylvanicus. Proceedings 2927 of the National Academy of Sciences of the United States of America, 109, 20750 - 20755

108

2928 MacMillan, H.A., Andersen, J.L., Davies, S.A. & Overgaard, J. (2015) The capacity to maintain 2929 ion and water homeostasis underlies interspecific variation in Drosophila cold tolerance. 2930 Scientific Reports, 5, 18607 doi:10.1038/srep18607. 2931 MacMillan, H.A., Schou, M.F., Kristensen, T.N. & Overgaard, J. (2016) Preservation of 2932 potassium balance is strongly associated with insect cold tolerance in the field: a seasonal 2933 study of Drosophila subobscura. Biology Letters, 12, 20160123. DOI: 2934 10.1098/rsbl.2016.0123. 2935 Ma, F.Z., Lu,¨ Z.C., Wang, R. & Wan, F.H. (2014) Heritability and evolutionary potential in 2936 thermal tolerance traits in the invasive Mediterranean cryptic species of Bemisia tabaci 2937 (Hemiptera: Aleyrodidae). PLoS ONE, 9, e103279. doi:10.1371/journal.pone.0103279 2938 Martin, T.L. & Huey, R.B. (2008) Why “suboptimal” is optimal: Jensen’s inequality and 2939 ectotherm thermal preferences. The American Naturalist, 171, 102–118. 2940 Meats, A. (1973) Rapid acclimatization to low temperature in the Queensland fruit fly, Dacus 2941 tryoni. Journal of Insect Physiology, 19, 1903-1911. 2942 Milton, C.C. & Partridge, L. (2008) Brief carbon dioxide exposure blocks heat hardening but not 2943 cold acclimation in Drosophila melanogaster. Journal of Insect Physiology, 54, 32-40 2944 Yi, S.X., Moore, C.W. & Lee, R.E. Jr. (2007) Rapid cold-hardening protects Drosophila 2945 melanogaster from cold-induced apoptosis. Apoptosis, 12, 1183-1193. 2946 Mudavanhu, P., Addison, P. & Conlong, D.E. (2014) Impact of mass rearing and gamma 2947 irradiation on thermal tolerance of Eldana saccharina. Entomologia Experimentalis et 2948 Applicata, 153, 55-63. 2949 Mwalusepo, S., Tonnang, H.E.Z., Massawe, E.S., Okuku, G.O., Khadioli, N., Johansson, T., 2950 Calatayud, P. & Le Ru, B.P. (2015) Predicting the impact of temperature change on the 2951 future distribution of maize stemborers and their natural enemies along east African 2952 mountain gradients using phenology models. PLoS ONE, 10, e0130427. 2953 doi:10.1371/journal.pone.0130427. 2954 Nyamukondiwa, C. & Terblanche, J.S. (2009) Thermal tolerance in adult Mediterranean and 2955 Natal fruit flies (Ceratitis capitata and Ceratitis rosa): effects of age, gender and feeding 2956 status. Journal of Thermal Biology, 34, 406-414.

109

2957 Nyamukondiwa, C., Kleynhans, E. & Terblanche, J.S. (2010) Phenotypic plasticity of thermal 2958 tolerance contributes to the invasion potential of Mediterranean fruit flies (Ceratitis 2959 capitata). Ecological Entomology, 35, 565-575. 2960 Nyamukondiwa, C. & Terblanche, J.S. (2010) Within-generation variation of critical thermal 2961 limits in adult Mediterranean and Natal fruit flies Ceratitis capitata and Ceratitis rosa: 2962 thermal history affects short-term responses to temperature. Physiological Entomology, 35, 2963 255-264. 2964 Nyamukondiwa, C., Weldon, C.W., Chown, S.L., le Roux, P.C. & Terblanche, J.S. (2013) 2965 Thermal biology, population fluctuations and implications of temperature extremes for the 2966 management of two globally significant insect pests. Journal of Insect Physiology, 59, 1199- 2967 1211. 2968 Ochieng, R.S., Onyango, F.O. & Bungu, M.D.O. (1985) Improvement of techniques for mass 2969 culture of Chilo partellus (Swinhoe). Insect Science and its Application, 6, 425-428. 2970 Overgaard, J. & Sorensen, J.G. (2008) Rapid thermal adaptation during field temperature 2971 variations in Drosophila melanogaster. Cryobiology, 56, 159–162. 2972 Parmesan, C. (2006) Ecological and evolutionary responses to recent climate change. Annual 2973 Review of Ecology, Evolution, and Systematics, 37, 637-669. 2974 Parmesan, C., Ryrholm, N., Stefanescu, C., Hill, J.K., Thomas, C.D., Descimon, H., Huntley, B., 2975 Kaila, L., Kullberg, J., Tammaru, T., Tennent, W.J., Thomas, J.A. & Warren, M. (1999) Pole 2976 ward shifts in geographical ranges of butterfly species associated with regional warming. 2977 Nature, 399, 579–583. 2978 R Development Core Team. (2016) R: A Language and Environment for Statistical Computing. 2979 R Foundation for Statistical Computing, Vienna, Austria. 2980 Rezende, E.L., Tejedo, M. & Santos, M. (2011) Estimating the adaptive potential of critical 2981 thermal limits: methodological problems and evolutionary implications. Functional Ecology, 2982 25, 111-121. 2983 Rezende, E.L., Castaneda, L.E. & Santos, M. (2014) Tolerance landscapes in thermal ecology. 2984 Functional Ecology, 28, 799-809 2985 Rinehart, J.P., Yocum, G.D. & Denlinger, D.L. (2000) Thermotolerance and rapid cold 2986 hardening ameliorate the negative effects of brief exposures to high or low temperatures on 2987 fecundity in the flesh fly, Sarcophaga crassipalpis. Physiological Entomology, 25, 330-336.

110

2988 Selvaraj, S., Ganeshamoorthi, P. & Pandiaraj, T. (2013) Potential impacts of recent climate 2989 change on biological control agents in agro-ecosystem: A review. International Journal of 2990 Biodiversity and Conservation, 5(12), 845-852. 2991 Sgrò, C.M., Terblanche, J.S. & Hoffmann, A.A. (2016) What can plasticity contribute to insect 2992 responses to climate change? Annual Review of Entomology, 61,433-451. DOI: 2993 10.1146/annurev-ento-010715-023859. 2994 Slabber, S., Worland, M.R., Leinaas, H.P. & Chown, S.L. (2007) Acclimation effects on thermal 2995 tolerances of springtails from sub-Antarctic Marion Island: indigenous and invasive species. 2996 Journal of Insect Physiology, 53, 113 – 125. 2997 Sinclair, B.J., Coello Alvarado, L.E. & Ferguson, L.V. (2015) An invitation to measure insect 2998 cold tolerance: Methods, approaches, and workflow. Journal of Thermal Biology, 53, 180- 2999 197. 3000 Tams, W.H.T. (1932) New species of African Heterocera. Entomologist, 65, 1241–1249. 3001 Tefera, T., Mugo, S., Tende, R. & Likhayo, P. (2010) Mass Rearing of Stem borers, Maize 3002 weevil and Larger grain borer Insect Pests of Maize. CIMMYT. Nairobi. 3003 Terblanche, J.S., Marais, E. & Chown, S.L. (2007a) Stage-related variation in rapid cold 3004 hardening as a test of the environmental predictability hypothesis. Journal of Insect 3005 Physiology, 53, 455–462. 3006 Terblanche, J.S., Deere, J.A., Clusella-Trullas, S., Janion, C. & Chown, S.L. (2007b) Critical 3007 thermal limits depend on methodological context. Proceedings of the Royal Society of 3008 London B, 274, 2935-2942. 3009 Terblanche, J.S., Clusella-Trullas, S., Deere, J.A. & Chown, S.L. (2008) Thermal tolerance in a 3010 south-east African population of tsetse fly Glossina pallidipes (Diptera:Glossinidae): 3011 implications for forecasting climate change impacts. Journal of Insect Physiology, 54, 114– 3012 127. 3013 Verberk, W.C.E.P., Overgaard, J., Ern, R., Bayley, M., Wang, T., Boardman, L & Terblanche, S. 3014 (2016) Does oxygen limit thermal tolerance in arthropods? A critical review of current 3015 evidence. Comparative Biochemistry and Physiology A, 192, 64-78. 3016 Wallner, W.E. (1987) Factors affecting insect population dynamics: differences between 3017 outbreak and non-outbreak species. Annual Review of Entomology, 32, 317-340.

111

3018 Wang, S., Ikediala, J.N., Tang, J. & Hansen, J.D. (2002) Thermal death kinetics and heating rate 3019 effects of fifth-instar Cydia pomonella (L.) (Lepidoptera: Tortricidae). Stored Products 3020 Research, 38, 441–553. 3021 Weldon, C.W., Terblanche, J.S. & Chown, S.L. (2011) Time-course for attainment and reversal 3022 of acclimation to constant temperature in two Ceratitis species, Journal of Thermal Biology, 3023 36, 479-485. 3024 Yoder, A., Pollock, D.A. & Benoit, J.B. (2003) Moisture requirements of the ladybird beetle 3025 Stethorus nigripesin relation to habitat preference and biological control. Entomologia 3026 Experimentalis et Applicata, 109, 83–87. 3027 Zhang, W., Qian Chang, X.Q., Hoffmann, A., Zhang, S. & Sen Ma, C. (2015) Impact of hot 3028 events at different developmental stages of a moth: the closer to adult stage, the less 3029 reproductive output. Scientific Reports, 5, 10436 DOI: 10.1038/srep10436. 3030 3031 3032 3033 3034 3035 3036

112

3037 3038

3039 CHAPTER 4

3040

3041

3042

3043

3044

3045

3046 Thermal plasticity mediates the interaction between host Chilo partellus Swinhoe 3047 (Lepidoptera: Crambidae) and endoparasitoid Cotesia flavipes Cameron (Hymenoptera: 3048 Braconidae) under rapidly changing environments*

3049

3050

3051

3052 3053 3054 3055 3056 3057 3058 * Published as “Mutamiswa, R., Chidawanyika, F. & Nyamukondiwa, C. (2017) Thermal 3059 plasticity potentially mediates the interaction between host Chilo partellus Swinhoe 3060 (Lepidoptera: Crambidae) and endoparasitoid Cotesia flavipes Cameron (Hymenoptera: 3061 Braconidae) under rapidly changing environments. Pest Management Science. 3062 DOI:10.1002/ps.4807

113

3063 4.1 Introduction 3064 Temperature is arguably the most important abiotic factor influencing insect physiology, 3065 ecology and consequently overall environmental fitness (Dillon et al., 2009). As ectotherms, 3066 insect exposure to extreme temperatures may cause irreversible loss of fitness and ultimately 3067 mortality (Dillon et al., 2009). Thus, temperature influences seasonality and evolutionary 3068 responses over short and long time scales in the wild (Nguyen et al., 2014). Due to global 3069 change, resistance to temperature stress in insects may be compromised (Fischer et al., 2010), 3070 with some organisms emerging ‘winners’ while others suffer the harmful effects (Mutamiswa et 3071 al., 2017a). Indeed, this is true for different trophic levels, and more especially herbivore 3072 parasitoid systems that show differential thermal sensitivity at different trophic levels (Romo & 3073 Tylianakis, 2013; Schreven et al., 2017). In consequence, insects employ a diverse range of 3074 morphological, physiological, and behavioral adaptations, or a combination of the mechanisms to 3075 counter adverse climate conditions (Chown & Terblanche, 2007; Nyamukondiwa et al., 2013). 3076 Physiologically, insects are known to survive extreme temperature variation through phenotypic 3077 plasticity, the ability of a same genotype to produce variable phenotypes under heterogeneous 3078 environments (Whitman & Ananthakrishnan, 2009). This may be in the form of either hardening 3079 (in the short-term) or acclimation under controlled laboratory conditions (e.g. Chidawanyika & 3080 Terblanche 2011a) and acclimatization in the field (longer-term) (Whitman & Ananthakrishnan, 3081 2009; Chidawanyika & Terblanche 2011b; Colinet & Hoffman, 2012; Nyamukondiwa et al., 3082 2013; Sgrò et al., 2016). At short time scales, hardening allows individuals to adjust their 3083 phenotype temporarily to the environment in which they currently inhabit (David et al., 2004; 3084 Austin & Moehring, 2013), and this may be significant in preserving key life-history traits (see 3085 Shreve et al., 2004). One such mechanism is rapid cold hardening (RCH), defined as a rapid 3086 improvement in survival at a low lethal temperature after brief pre-treatment to a prior sub-lethal 3087 temperature shock (Lee et al., 1987). Indeed, RCH has been reported to improve survival in 3088 several insect taxa including Lepidoptera, Thaumatotibia leucotreta (Stotter & Terblanche, 2009; 3089 Chidawanyika & Terblanche 2011a), Tephritids, Ceratitis capitata and Ceratitis rosa 3090 (Nyamukondiwa et al., 2010), Zaprionus vittiger (Nyamukondiwa & Terblanche, 2010a) and 3091 several Drosophila species (Kelty & Lee, 2001; Overgaard & Sørensen, 2008; Nyamukondiwa et 3092 al., 2011; Overgaard et al., 2014a). Similarly, at high temperatures, brief exposure to sub-lethal 3093 high temperatures can result in improved survival at a lethal high temperature (Chen et al.,

114

3094 1991), a phenomenon called rapid heat hardening (RHH). This mechanism has been shown to 3095 improve survival in several insects including C. capitata (Kalosaka et al., 2009; Nyamukondiwa 3096 et al., 2010), Cydia pomonella (Chidawanyika & Terblanche 2011a), Drosophila species 3097 (Nyamukondiwa et al., 2011) and Zygogramma bicolorata (Chidawanyika et al., 2017). 3098 Acclimation, on the other hand, improves survival through a rapid reversible change in 3099 phenotype in response to a relatively chronic pre-exposure period (days, seasons) to conditions 3100 similar to future potentially stressful novel thermal conditions (Woods & Harrison, 2002; 3101 Terblanche et al., 2006). This is sometimes referred to as beneficial acclimation (Woods & 3102 Harrison, 2002) and only a match in pre-exposure and future stressful conditions will result in 3103 improved survival whilst mismatches may result in reduced fitness (Kristensen et al., 2008; 3104 Chidawanyika & Terblanche, 2011b). These plastic responses are the first in a suite of 3105 mechanisms that an organism may use to survive when faced with novel environments 3106 (Nyamukondiwa et al., 2010), and are significant for maintaining fitness under field conditions 3107 (Kristensen et al., 2008; Chidawanyika & Terblanche, 2011b). Indeed, similar beneficial 3108 acclimatory responses have been reported in D. melanogaster (Colinet & Hoffman, 2012), C. 3109 capitata (Basson et al., 2012), Nezara viridula (Chanthy et al., 2012) and Phthorimaea 3110 operculella (Hemmati et al., 2014). 3111 Temperature tolerance across species has been reported to vary, depending on ecological 3112 and evolutionary history (Chown & Terblanche, 2007; Stotter & Terblanche, 2009; 3113 Nyamukondiwa & Terblanche, 2010b), developmental stage or age and sex (Nyamukondiwa & 3114 Terblanche, 2009). Hence even for species with related evolutionary adaptation strategies, 3115 variations in thermal tolerance may occur ultimately shaping species abundance and distributions 3116 (Chen et al., 2011; Sorte et al., 2011; Hoffmann et al., 2012; Nguyen et al., 2014; Overgaard et 3117 al., 2014b; Mutamiswa et al., 2017a; reviewed in Sgrò et al., 2016). For interacting species, 3118 differential changes in fitness may disrupt long evolved mutual or antagonistic ecological 3119 interactions (Parmesan, 2006; Yang & Rudolf, 2010), ultimately changing ecosystem community 3120 structures and function (Grimm et al., 2013) For example, some species may have limited 3121 compensatory abilities and thus are highly vulnerable to heterogeneous thermal regimes while 3122 others may compensate through plasticity thereby avoiding potentially lethal effects (Sgrò et al., 3123 2016). In agriculture, the efficacy of biological control programs will be poor if important 3124 interactions between crop pests and their natural enemies are disrupted due to asynchrony

115

3125 between the host-natural enemy complex, should there be differential thermal sensitivity at the 3126 two trophic levels (Dyer et al., 2013; Romo & Tylianakis, 2013). Consequently, understanding 3127 the physiological responses of pest-parasitoid complexes to thermal stress is essential for 3128 predicting how climate change may affect their distribution, population phenology and ultimately 3129 the efficacy of biological control thereof (Hance et al., 2007; Bale, 2010; Amarasekare & 3130 Savage, 2010; Schreven et al., 2017; Kleynhans et al., 2014). 3131 The spotted stemborer, C. partellus (Swinhoe) (Lepidoptera: Crambidae) is one of the 3132 most damaging lepidopteran stemborers of sorghum and maize in eastern and southern Africa 3133 (Khadioli et al., 2014). It is native to Asia and was first recorded in Africa (Malawi) around 1930 3134 (Tams, 1932). Chilo partellus has a history of invasion success, expanding to high elevation 3135 areas, highland tropical areas and humid transitional areas of east and southern African regions 3136 (Khadioli et al., 2014; Mutamiswa et al., 2017b). The successful establishment of C. partellus in 3137 novel environments and expansion to highland areas (Khadioli et al., 2014) probably stems from 3138 basal physiological tolerances (Mutamiswa et al., 2017b), morphological and behavioural 3139 adaptations that enabled them to displace indigenous stemborer species (Kfir, 1997). Cotesia 3140 flavipes Cameron (Hymenoptera: Braconidae), native to Asia is a gregarious koinobiont larval 3141 endoparasitoid of C. partellus (Dejen et al., 2013). The parasitoid prefers warm and dry areas 3142 (Songa et al., 2001) and, like other insects, variation from this optimum may affect fitness. While 3143 previous research investigating phenotypic plasticity of arthropods of agricultural importance 3144 mainly focused on insect pests (e.g Rajamohan & Sinclair, 2008; Nyamukondiwa et al., 2010; 3145 Nyamukondiwa & Terblanche, 2010a; Chidawanyika & Terblanche, 2011a) but see (Schreven et 3146 al., 2017), to my knowledge, no studies to date have focused on plastic thermal responses among 3147 host-parasitoid complexes, and their implications on pest management. Hence, investigating 3148 phenotypic plasticity of insect pest and its parasitoid gives insights into how host-parasitoid 3149 population dynamics will be affected by changes in climate. 3150 Here, I investigate the phenotypic plasticity of thermal tolerance of C. partellus 3151 developmental stages (larvae, pupae and adults) and its larval parasitoid C. flavipes (adults) and 3152 present implications on biological control efficacy in the face of climate change. Specifically, I 3153 determined short (RCH and RHH) and long (acclimation) term plastic responses of survival at 3154 extreme temperatures for these two species. I hypothesize that C. partellus is equally thermally 3155 plastic compared to its biological control agent, C. flavipes, as set by their long host-parasitoid

116

3156 co-evolutionary history. The results of plastic responses for the two interacting species will be 3157 essential in developing pest risk assessments and designing pest management programs under 3158 global change. 3159

3160 4.2 Material and Methods 3161

3162 4.2.1 Study insects and rearing conditions 3163 Chilo partellus and C. flavipes colonies were obtained from International Centre for 3164 Insect Physiology and Ecology (ICIPE), Kenya. These colonies had been in the laboratory for 3165 more than 20 generations with regular augmentation with wild populations to maintain genetic 3166 diversity. Chilo partellus larvae were reared in climate chambers (HPP 260, Memmert GmbH + 3167 Co.KG, Germany) on artificial diet (Ochieng et al., 1985; Tefera et al., 2010), in trays at 28 ± 3168 1°C, 65 ± 10% relative humidity (RH) and 12L: 12D photocycle. Pupae were held in open Petri 3169 dishes in Bugdorm rearing cages (240mm3; Bugdorm-BD43030F, Megaview Science Co., Ltd, 3170 Taiwan) under the same larval rearing conditions until adult eclosion. Upon emergence, all 3171 adults had access to water ad libtum from soaked cotton wool while gravid females had access to 3172 wax paper placed in the cages as an oviposition surface. For uniformity among test insects, eggs 3173 were harvested after every 12 h and transferred from wax paper to artificial diet where they 3174 hatched into larvae. The larvae then went through five instar stages before pupation. Thereafter, 3175 pupae were collected daily and placed in Petri dishes in rearing cages for adult eclosion. To 3176 maintain genetic similarity to wild populations, wild-caught moths were added to the colony 3177 once every month during summer (e.g. Nyamukondiwa & Terblanche, 2009). For C. flavipes 3178 colony, parasitised C. partellus larvae from ICIPE were maintained in the climate chambers 3179 under 12L: 12D photocycle, 28 ± 1°C and 65 ± 10% RH until adult parasitoid eclosion. Emerged 3180 parasitoids had access to food ad libitum (25% honey: water mixture from a cotton wick) until 3181 they were used in thermal tolerance assays. In the hardening and acclimation experiments, C. 3182 partellus larvae (6th instar), pupae, adults (24-48 h old) and C. flavipes adults (24-48 h old) were 3183 used. 3184 3185

117

3186 4.2.2 Phenotypic Plasticity of Low Temperature Tolerance 3187 Rapid cold hardening experiments were performed using established protocols. (e.g. 3188 Terblanche et al., 2008; Nyamukondiwa et al., 2010; Chidawanyika & Terblanche, 2011a). I first 3189 determined in the preliminary trials the lower discriminating temperatures i.e. temperatures 3190 causing 80-100% mortality upon 2h exposure to a stressful low temperature. A pre-treatment of 3191 10°C above the lower discriminating temperature is generally suffice to elicit a RCH response. 3192 (reviewed in Lee & Denlinger, 2010). Since discriminating temperatures differed across species 3193 and developmental stage, I thus used a standardized, species-developmental stage-specific 3194 approach for eliciting hardening responses (see e.g. Nyamukondiwa et al., 2011) (Table 4.1). 3195 3196 Table 4.1: Summary table showing species and developmental stage specific hardening and 3197 discriminating temperatures (2 h duration) for host, C. partellus and parasitoid, C. flavipes. 3198 Species Developmental Hardening Hardening Discriminating Discriminating stage low high low high temperature temperature temperature temperature (°C) (°C) (°C) (°C)

Chilo Larvae 2 41 -8 46 partellus

Pupae 2 42 -8 47

Adults 1 39 -9 44

Cotesia Adults 3 35 -7 40 flavipes

3199 3200 Five replicates of 60 ml vials consisting 10 stemborers of each life-stage (larvae, pupae and 3201 adults) and C. flavipes (adults) were placed in a growth chamber at 28°C, 65 ± 10% RH for 30 3202 min to allow for equilibration. These insects were later hardened for 2h at a non-lethal stressful 3203 low temperature and then immediately subjected to the discriminating temperature (Table 4.1). 3204 Control organisms were held at optimum temperature before subjecting them directly into the

118

3205 discriminating temperature so that any difference in mortality was a result of the hardening 3206 treatment. Following treatment, insects were returned to 28°C, 65% RH and 12L: 12D in climate 3207 chambers before scoring survival. Treatment and control insects had access to food and water ad 3208 libtum during the recovery period. Survival was defined as ability to pupate (for larva), eclose 3209 (for pupae) and coordinated muscle response to stimuli such as gentle prodding, or normal 3210 behaviours such as feeding, flying or mating for adults (as in e.g. Nyamukondiwa et al., 2010). 3211 Because of the close interaction between insect physiology and ecology, CTLs ramping 3212 assays remain ecologically relevant measures of understanding how species respond to 3213 heterogeneous stressful environments (Terblanche et al., 2011). Longer-term acclimation effects 3214 on thermal tolerance for C. partellus and C. flavipes were assayed as outlined by Nyamukondiwa 3215 & Terblanche (2010b). High temperature acclimation was defined for each species as 5°C above 3216 rearing/benign temperature while low temperature acclimation was defined as 5°C less rearing 3217 temperature. The organisms acclimated to optimum temperature were considered as a control

3218 group (e.g. Terblanche et al., 2006). Critical thermal minima (CTmin) were measured following 3219 acclimation of insects to 23 (low), 28 (optimum) and 33°C (high) for duration of 3 days (Figure 3220 4.1). 3221 Acclimation for this duration is generally suffice to elicit some plastic responses in other similar 3222 insect taxa eg. Bicyclus anynana, C. capitata (Fischer et al., 2010; Weldon et al., 2011). Ten 3223 individual organisms (for each species and developmental stage) from each acclimation were 3224 randomly placed in the ‘organ pipes’ connected to a programmable water bath (Lauda Eco Gold, 3225 Lauda DR.R. Wobser GMBH and Co. KG, Germany) filled with 1:1 water: propylene glycol to 3226 allow for subzero temperatures (Chown & Nicolson, 2004). Assays started at a set point of 3227 benign/rearing temperature of 28°C, 65 ± 10% RH for 10 mins (to allow equilibration of insect 3228 body temperatures) before ramping the temperature down at a rate of 0.25°C/min until their

3229 CTmin were recorded. A thermocouple (type K 36 SWG) connected to a digital thermometer 3230 (53/54IIB, Fluke Cooperation, USA) was inserted into a control chamber to record organ pipe 3231 chamber temperature. This was repeated twice to yield sample sizes of n = 20. Critical thermal 3232 minima was defined as the temperature at which each individual insect lost coordinated muscle 3233 function, consequently losing the ability to respond to mild stimuli (e.g. prodding) (e.g. 3234 Nyamukondiwa & Terblanche, 2010b). 3235

119

3236 3237 Acclimation 3238 (3 days)

3239 Measure CTLs,SCP, HKDTand CCRT o 3240 23 C

RearingTemperature (28 65% RH 3241 3242 3243

3244 o 28 C 3245 65% RH 3246 3247 3248

3249 C)

3250 o 33 C 3251

65% RH 3252 3253 3254 3255 Figure 4.1: Schematic diagram for acclimation experimental protocols. For full description see 3256 sections 4.2.2 and 4.2.3 3257 3258 CCRT experiments were performed following 3 days acclimation (as in e.g. Weldon et 3259 al., 2011) (Figure 4.1). To determine CCRT, 10 replicate organisms from each developmental 3260 stage were individually placed in 0.65ml microcentrifuge tubes and then loaded into a large zip- 3261 lock bag that was submerged into a water bath (Systronix, Scientific, South Africa) filled with 3262 1:1 water: propylene glycol set at 0°C for 1 hour. This temperature by time treatment has also 3263 been reported to elicit chill-coma in other insect taxa (see Weldon et al., 2011; Sinclair et al., 3264 2013). Following 1 h at this chill-coma temperature, the tubes were immediately transferred from 3265 the water bath to the climate chamber (Memmert HPP 260, Memmert GmbH + Co.KG, 3266 Germany) set at 28°C, 65% RH for recovery. This was repeated thrice to yield sample sizes of n

120

3267 = 30 (30 replications) per each treatment. The chamber was connected to a video recording 3268 camera (HD Covert Network Camera, DS-2CD6412FWD-20, Hikvision Digital Technology Co., 3269 Ltd, China) which was linked to a computer. All observations and data were recorded from this 3270 climate chamber video recording system. CCRT was defined as the time (in minutes) required to 3271 regain coordinated movement ability (in larva) to stand upright on its legs (for adults) (Milton & 3272 Partridge, 2008; Chown & Nicolson, 2004). 3273 SCPs were assayed as outlined by Nyamukondiwa et al. (2013) following acclimation as 3274 previously outlined. A total of 16 acclimated individuals of each life stage were individually 3275 placed in 0.65 ml microcentrifuge tubes. Each insect was attached to a tip of a type-T copper- 3276 constantan thermocouple (762-1121, Cambridge, UK) that was inserted through the lid of the 3277 tube and both the insect and thermocouple were secured in contact using a cotton wool. The 3278 thermocouples were connected to one of two 8-channel Picotech TC-08 (Pico Technology, 3279 Cambridge, UK) thermocouple interfaces and temperatures were recorded at 1s intervals using 3280 PicoLog software for windows (Pico Technology, Cambridge, UK). Experiments started at a set 3281 point temperature of 15°C for 10 mins (for insects’ temperature equilibration) before ramping 3282 down at a rate of 0.5°C min-1 until the observation of SCPs. In this study, SCP for each 3283 individual was defined as the lowest temperature recorded prior to a spike in temperature 3284 associated with the latent heat of crystallization (e.g. Nyamukondiwa et al., 2013). 3285

3286 4.2.3 Phenotypic Plasticity of High Temperature Tolerance 3287 Rapid heat hardening experiments were also performed using established protocols (e.g. 3288 Terblanche et al., 2008; Nyamukondiwa et al., 2010; Chidawanyika & Terblanche, 2011a). Like 3289 in RCH experiments, discriminating high temperatures were first determined during preliminary 3290 trials. A pre-exposure of 5°C below the discriminating upper temperature is generally suffice to 3291 elicit a RHH response (reviewed in Hoffmann et al., 2003; Nyamukondiwa et al., 2011). 3292 Organisms were allowed to equilibrate in a growth chamber at 28°C, 65 ± 10% RH for 30 min 3293 before being hardened for 2h at a stressful high temperature (see Table 4.1). Following 3294 hardening, organisms were given 30 minutes recovery time to allow the de novo synthesis of heat 3295 shock proteins (HSPs) (see Nyamukondiwa et al., 2010) before subjecting them for 2h at 3296 discriminating high temperature (Table 4.1). Control organisms were also held at optimum 3297 temperature before subjecting them directly into the discriminating temperature so that any

121

3298 difference in mortality was a result of the hardening treatment. Following treatment, insects were 3299 later returned to 28°C, 65% RH and 12L: 12D in climate chambers before scoring survival.

3300 Following high temperature acclimation CTmax was measured following generally the same

3301 methodology as in CTmin above. However, for CTmax, temperature was ramped up at a rate of 3302 0.25°C/min until the temperature at which each individual insect lost coordinated muscle 3303 function, consequently losing the ability to respond to mild stimuli (e.g. prodding) (e.g. 3304 Nyamukondiwa & Terblanche, 2010b). 3305 For HKDT following acclimation, 10 replicate insects were loaded individually into 3306 0.65ml microcentrifuge tubes and placed in a climate chamber that was connected to a camera 3307 linked to a computer. In the climate chamber, C. partellus larvae and adults were subjected to a 3308 test temperature of 49.0±0.3°C, 65% RH whist C. flavipes adults were exposed to 45.0±0.3°C, 3309 65% RH. These knockdown temperatures (45.0 and 49.0°C) were selected based on preliminary 3310 trials of upper critical temperatures to activity, which were 44.6±0.37°C; 47.88±0.76°C and 3311 48.5±0.74°C for C. flavipes adults, C. partellus adults and larvae, respectively. This was repeated 3312 thrice to yield sample sizes of n = 30 (30 replications) per each experiment and developmental 3313 stage. All observations were recorded from the climate chamber video recording system. HKDT 3314 was defined as the time (in minutes) at which activity was lost by the organisms following 3315 exposure to heat knockdown temperature. 3316

3317 4.2.4 Microclimate data 3318 To determine the thermal environment experienced by C. partellus and C. flavipes in the 3319 field, shaded microclimatic temperatures were recorded using Thermocron iButtons (Dallas 3320 Semiconductors, Model DS1920) (0.5°C accuracy; 1 h sampling frequency) from a maize field, 3321 BCA Farm, Gaborone, Botswana (S24.56449; E25.94883; 994 m.a.s.l) which hosts these two 3322 species. Recordings were done during the period January 2016 to February 2017. 3323

3324 4.2.5 Statistical analyses 3325 Data analyses were carried out in STATISTICA, version 13.0 (Statsoft Inc., Tulsa, 3326 Oklahoma). Before analysing the results for hardening and acclimation responses, data were 3327 checked for normality and equality of variances using the Shapiro-Wilk test and Hartley-Bartlett

122

3328 tests, respectively. In the hardening assays, key assumptions of ANOVA were not met, while for 3329 the acclimation assays (and subsequent CTLs, CCRT, HKDT and SCPs), key assumptions of 3330 ANOVA were satisfied. To investigate the effects of hardening on the survival of C. partellus 3331 and C. flavipes, treatment groups were analysed using Generalised linear models (GLZ) 3332 assuming a binomial distribution and a logit link function. Acclimation effects on CTLs, SCPs, 3333 HKDT and CCRT were assessed using full factorial analysis of variance (ANOVA). Tukey- 3334 Kramer’s post-hoc tests were used to separate statistically heterogeneous groups. Overlap in 95% 3335 confidence limits was used to identify statistically homogeneous groups. 3336 Using microclimatic data, I calculated the number of likely natural RCH and RHH events 3337 occurring during the time I recorded my microclimate data using Matlab software, R2017a 3338 (MathWorks). The RCH events were taken from the following range of temperatures: 0.5≤1˂1.5; 3339 1.5≤2˂2.5 and 2.5≤3˂3.5 whilst RHH events were generated from the temperature ranges: 3340 34.5≤35˂35; 38.5≤39˂39.5; 40.5≤41˂41.5 and 41.5≤42˂42.5 for a period ≥2 h, enough to elicit 3341 hardening responses. 3342

3343 4.3 Results 3344

3345 4.3.1 Phenotypic Plasticity of Low Temperature Tolerance 3346 A 2 h low temperature pretreatment (RCH) of 2 and 3°C significantly enhanced survival 3347 during a 2 h exposure at discriminating low temperature of -8 and -7°C for C. partellus larvae 3348 (χ2 =39.97, d.f=1, P ˂ 0.0001) (Figure 4.2A), and C. flavipes adults (χ2 =60.01, d.f=1, P ˂ 3349 0.0001) respectively (Figure 4.2B). Similarly, RCH (1 and 2°C) significantly improved low 3350 temperature survival (-9 and -8°C) for C. partellus pupae (χ2 =6.91, d.f=1, P = 0.008 (Figure 3351 4.3A) and adults) (χ2 =83.72, d.f=1, P ˂ 0.0001) respectively (Figure 4.3B). 3352 There was a significant acclimation, species and acclimation by species interaction effect

3353 for low temperature tolerance vis a vis CTmin, CCRT and SCP (Table 4.2). Acclimation to low ° 3354 temperature (23 C for 3 days) significantly improved CTmin (Figure 4.2C) and CCRT (Figure 3355 4.2D) for both C. partellus larvae and C. flavipes adults. Nevertheless, acclimation to high 3356 temperature (33°C for 3 days) generally compromised C. partellus larvae and adult C. flavipes

3357 CTmin and CCRT (Figure 4.2C; D). Furthermore, high temperature acclimation enhanced CCRT

123

3358 for partellus larvae relative to the control (Figure 4.2D). A comparison of C. partellus 3359 developmental stages for low temperature tolerance shows that low temperature acclimation

3360 improved CTmin and CCRT for both adults and larvae (Table 4.3; Figure 4.3C; D). The effect of

3361 developmental stage by acclimation interaction was highly significant for CTmin, CCRT and SCP 3362 (Table 4.3). 3363 Acclimation to low temperature (23°C for 3 days) increased SCP (more positive SCP) in 3364 C. partellus larvae while marginally enhancing it (more negative SCP) in C. flavipes adults 3365 (Figure 4.2E). However, acclimation to high temperature (33°C for 3 days) enhanced SCP in C. 3366 partellus while compromising it in C. flavipes (Figure 4.2E). There was a significant difference 3367 (P ˂ 0.05) in SCP for C. partellus and C. flavipes acclimated at low temperature while no 3368 significant differences (P > 0.05) were recorded for the same species acclimated at high 3369 temperature (Figure 4.2E). In addition, low temperature acclimation resulted in significantly 3370 compromised SCPs for C. partellus developmental stages (larvae, pupae and adults) (Figure 3371 4.3E). 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385

124

100 A

80 b

Survival (%) Survival a

0 Control Treatment 3386 3387

100 B b

Survival (%)

a 20

0 3388 Control Treatment

125

10 Chilo partellus c C 9 Cotesia flavipes b

o 8

7 a

3 f 2 e

Critical thermal minima ( Criticalminima thermal 1 d

0 Low Optimum High Acclimation temperature (oC) 3389 3390

5 Chilo partellus D Cotesia flavipes

b 4 c

d c 3 a

Chill coma recovery time (minutes) recoverytime coma Chill 1 Low Optimum High Acclimation temperature (oC) 3391

126

-6 Chilo partellus E -8 Cotesia flavipes

b -10

C) b

o -12

-14

-16 c c

ac -18 a

-20

Supercooling point ( point Supercooling

-22

-24 Low Optimum High Acclimation temperature (oC) 3392 3393 3394 Figure 4.2: Summary low temperature plasticity results showing (A) effects of 2h 3395 pretreatments/hardening on survival at discriminating low temperatures of Chilo partellus larvae 3396 (pretreatment [2°C 2h] on low temperature survival [-8°C 2 h]), (B) effects of (pretreatment [3°C 3397 2 h] on low temperature survival [-7°C 2 h]) on Cotesia flavipes adults, (C) acclimation effects to ° ° 3398 low optimum and high temperature (23; 28 and 33 C respectively) on CTmin, (D) CCRT (at 0 C 3399 chill coma temperature) and (E) SCP. Error bars represent 95% CLs (N = 50 per group). Means 3400 with the same letter are not significantly different from each other. Post hoc tests were done 3401 separately for each species 3402

127

100 A

80 b

a 60

Survival (%)

0 3403 Control Treatment 3404

100

b B

a 40

Survival (%) Survival

0 3405 Control Treatment

128

10 Larvae c C 9 Adults

C) b

7 a

6 e e 5

Critical thermal minima ( Criticalminima thermal 4

3 Low Optimum High Acclimation temperature (oC) 3406 3407

5 Larvae b D Adults

4 a

e 3 e

Chill coma recovery time (minutes) recovery Chill coma 1 Low Optimum High Acclimation temperature (oC) 3408

129

-8 Larvae d -9 E Pupae -10 d Adults C) -11

o -12 -13 a -14 a a a -15 ac -16 a -17

Supercooling point ( -18 bc -19 -20 Low Optimum High Acclimation temperature (oC) 3409 3410 3411 Figure 4.3: Summary low temperature plasticity results showing (A) effects of 2h 3412 pretreatments/hardening on survival at discriminating low temperatures of Chilo partellus pupae 3413 (pretreatment [2°C 2h] on low temperature survival [-8°C 2 h]), (B) effects of (pretreatment [1°C 3414 2 h] on low temperature survival [-9°C 2 h]) on Chilo partellus adults, (C) acclimation effects to ° ° 3415 low optimum and high temperature (23; 28 and 33 C respectively) on CTmin, (D) CCRT (at 0 C 3416 chill coma temperature) and (E) SCP. Error bars represent 95% CLs (N = 50 per group). Means 3417 with the same letter are not significantly different from each other. Post hoc tests were done 3418 separately for each developmental stage 3419 3420

130

3421 Table 4.2: Summary results from a full factorial ANOVA showing the effects of species (C. 3422 partellus and C. flavipes) and temperature acclimation (low, optimum and high) on low

3423 temperature activity measured as CTmin, CCRT and SCP. SS = sums of squares, d.f.=degrees of 3424 freedom. Trait Effect SS DF MS F P

CTmin Intercept 2368.297 1 2368.297 36991.96 ˂ 0.001

Acclimation 76.166 2 38.083 594.84 ˂ 0.001

Species 1348.711 1 1348.711 21066.39 ˂ 0.001

Species*Acclimation 11.258 2 5.629 87.92 ˂ 0.001

Error 7.299 114 0.064

CCRT Intercept 2133.243 1 2133.243 3850.196 ˂ 0.0001

Acclimation 44.028 2 22.014 39.732 ˂ 0.0001

Species 8.667 1 8.667 15.642 ˂ 0.0001

Species*Acclimation 20.827 2 10.413 18.795 ˂ 0.0001

Error 146.272 264 0.554

SCP Intercept 24541.45 1 24541.45 5883.550 ˂ 0.0001

Acclimation 105.76 2 52.88 12.677 ˂ 0.0001

Species 811.54 1 811.54 194.558 ˂ 0.0001

Species*Acclimation 427.18 2 213.59 51.206 ˂ 0.0001

Error 375.41 90 4.17

3425 3426 3427

131

3428 Table 4.3: Summary results from a full factorial ANOVA showing the effects of C. partellus 3429 developmental stage (larvae and adults) and temperature acclimation (low, optimum and high)

3430 on low temperature activity measured as CTmin, CCRT and SCP. SS = sums of squares, d.f.= 3431 degrees of freedom. Trait Effect SS DF MS F P

CTmin Intercept 4583.088 1 4583.088 25327.07 ˂ 0.0001

Acclimation 81.369 2 40.685 224.83 ˂ 0.0001

Developmental stage 312.987 1 312.987 1729.63 ˂ 0.0001

Developmental ˂ 0.0001 9.666 2 4.833 26.71 Stage*Acclimation

Error 20.629 114 0.181

CCRT Intercept 1810.387 1 1810.387 7510.453 ˂ 0.0001

Acclimation 36.831 2 18.415 76.397 ˂ 0.0001

Developmental stage 77.710 1 77.710 322.382 ˂ 0.0001

Developmental ˂ 0.0001 6.713 2 3.357 13.925 Stage*Acclimation

Error 41.943 174 0.241

SCP Intercept 31693.49 1 31693.49 6733.752 ˂ 0.0001

Acclimation 96.63 2 48.31 10.265 ˂ 0.0001

Developmental stage 245.59 2 122.79 26.089 ˂ 0.0001

Developmental ˂ 0.0001 409.42 4 102.36 21.747 Stage*Acclimation

Error 635.40 135 4.71

132

3432 4.3.2 Phenotypic Plasticity of High Temperature Tolerance 3433 As in low temperature hardening, 2 h high temperature pretreatments (35 and 41°C) 3434 significantly improved survival during 2 h exposure at discriminating high temperatures (40 and 3435 46°C) in C. partellus larvae (χ2 =105.54, d.f =1, P ˂ 0.001) (Figure 4.4A) and C. flavipes adults 3436 (χ2 =96.41, d.f =1, P ˂ 0.001) (Figure 4.4B) respectively. Furthermore, RHH at 42 and 39°C 3437 significantly improved high temperature survival at 47 and 44°C for C. partellus pupae (χ2 3438 =27.12, d.f =1, P ˂ 0.0001) (Figure 4.5A) and adults (χ2 =23.42, d.f =1, P ˂ 0.0001) (Figure 3439 4.5B) respectively. 3440 There was a significant interaction effect among the factors acclimation, species and

3441 acclimation by species for high temperature tolerance vis a vis CTmax and HKDT (Table 4.4). A

3442 comparison of C. partellus developmental stages for CTmax and HKDT showed a significant

3443 acclimation and developmental effect for both CTmax and HKDT, but interaction effects for the 3444 two (developmental stage x acclimation) were only significant for HKDT (Table 4.5). ° 3445 Acclimation to high temperature (33 C) significantly enhanced CTmax for both C. flavipes

3446 and C. partellus larvae (Figure 4.4C). High temperature acclimation marginally increased CTmax 3447 for C. partellus larvae and adults whilst low temperature acclimation showed no significant 3448 differences from the control (Figure 4.4C). High temperature (33°C for 3 days) acclimation 3449 increased HKDT for C. flavipes but had no effect on its larval C. partellus host (Figure 4.4D). 3450 Similarly, low temperature (23°C for 3 days) acclimation compromised HKDT for larval host C. 3451 partellus while enhancing the trait for parasitoid C. flavipes (Figure 4.4D). A comparison of C.

3452 partellus developmental stages shows high temperature acclimation increasing CTmax for larvae 3453 and no effect on adults (Figure 4.5C). Moreover, acclimation to high temperature also increased 3454 HKDT for adults, but not for larvae (Figure 4.5D). 3455

133

3456 3457

100 A b

Survival (%) Survival

a 20

0 Control Treatment 3458 3459 3460

100 B b

Survival (%) Survival

0 Control Treatment 3461

134

51 Chilo partellus C 50 b Cotesia flavipes

o a a 49

46 c 45 c

b 44

Critical thermal( maxima 43

42 Low Optimum High Acclimation temperature (oC) 3462

23 Chilo partellus e D 22 Cotesia flavipes 21 a 20

18 d b

16 b

15 c

Heat knock-down time (minutes) 13 Low Optimum High Acclimation temperature (oC) 3463 3464 3465 Figure 4.4: Summary low temperature plasticity results showing (A) effects of 2h 3466 pretreatments/hardening on survival at discriminating high temperatures of Chilo partellus larvae 3467 (pretreatment [41°C 2h] on high temperature survival [46°C 2 h]), (B) effects of (pretreatment 3468 [35°C 2 h] on high temperature survival [40°C 2 h]) on Cotesia flavipes adults, (C) acclimation ° 3469 effects to low optimum and high temperature (23; 28 and 33 C respectively) on CTmax and (D) 3470 HKDT (at 45 and 49°C knockdown temperature for C. flavipes adults and C. partellus larvae 3471 respectively). Error bars represent 95% CLs (N = 50 per group). Means with the same letter are 3472 not significantly different from each other. Post hoc tests were done separately for each species

135

100 b A

a 40

Survival (%) Survival

0 Control Treatment 3473 3474

100 B b 80

a 60

Survival (%) Survival

0 Control Treatment 3475

136

50 Larvae C Adults

C) b

49 a a

c 48 c

Critical thermal maximaCritical (

47 Low Optimum High Acclimation temperature (oC) 3476 3477

22 Larvae a D 20 Adults

18 b b bc

16 d 14

10 e

Heat knock-down time (minutes) knock-down Heat

8 Low Optimum High Acclimation temperature (oC) 3478 3479 3480 Figure 4.5: Summary low temperature plasticity results showing (A) effects of 2h 3481 pretreatments/hardening on survival at discriminating high temperatures of Chilo partellus pupae 3482 (pretreatment [42°C 2h] on high temperature survival [47°C 2 h]), (B) effects of (pretreatment 3483 [39°C 2 h] on high temperature survival [44°C 2 h]) on Cotesia flavipes adults, (C) acclimation ° 3484 effects to low optimum and high temperature (23; 28 and 33 C respectively) on CTmax and (D) 3485 HKDT (at 49°C knockdown temperature). Error bars represent 95% CLs (N = 50 per group). 3486 Means with the same letter are not significantly different from each other. Post hoc tests were 3487 done separately for each developmental stage

137

3488 Table 4.4: Summary results from a full factorial ANOVA showing the effects of species (C. 3489 partellus and C. flavipes) and temperature acclimation (low, optimum and high) on high

3490 temperature activity measured as CTmax and HKDT. SS = sums of squares, d.f.=degrees of 3491 freedom. 3492 Trait Effect SS DF MS F P

CTmax Intercept 259821.7 1 259821.7 520710.5 ˂ 0.0001

Acclimation 18.7 2 9.4 18.8 ˂ 0.0001

Species 590.6 1 590.6 1183.6 ˂ 0.0001

Species*Acclimation 5.3 2 2.6 5.3 ˂ 0.01

Error 56.9 114 0.5

HKDT Intercept 67121.44 1 67121.44 8642.910 ˂ 0.0001

Acclimation 876.76 2 438.38 56.448 ˂ 0.0001

Species 263.38 1 263.38 33.915 ˂ 0.0001

Species*Acclimation 261.25 2 130.62 16.820 ˂ 0.0001

Error 2050.24 264 7.77

3493 3494 3495 3496 3497 3498 3499 3500 3501 3502

138

3503 Table 4.5: Summary results from a full factorial ANOVA showing the effects of C. partellus 3504 developmental stage (larvae and adults) and temperature acclimation (low, optimum and high)

3505 on high temperature activity measured as CTmax and HKDT. SS = sums of squares, d.f.= degrees 3506 of freedom. Trait Effect SS DF MS F P

CTmax Intercept 279908.2 1 279908.2 1067923 ˂ 0.0001

Acclimation 5.9 2 3.0 11 ˂ 0.0001

Developmental stage 24.7 1 24.7 94 ˂ 0.0001

Developmental 0.7 2 0.4 1 0.262871 Stage*Acclimation

Error 29.9 114 0.3

HKDT Intercept 44231.72 1 44231.72 26834.95 ˂ 0.0001

Acclimation 141.82 2 70.91 43.02 ˂ 0.0001

Developmental stage 746.85 1 746.85 453.10 ˂ 0.0001

Developmental 919.36 2 459.68 278.88 ˂ 0.0001 Stage*Acclimation

Error 286.80 174 1.65

3507

3508 4.3.3 Microclimate data 3509 Microclimate data revealed several periods where conditions were suitable for eliciting 3510 RCH and RHH (Figure 4.6). For C. partellus, only one incidence was suitable for RCH whilst 3511 the larval and pupal stages had 11. Cotesia flavipes had 15 potential RCH events with 3512 temperatures and sufficient duration to elicit such responses as determined under laboratory 3513 conditions (see table 4.1). In addition, a total of 51, 42, 18 and 3 potential RHH events

139

3514 corresponding to C. flavipes adults, C. partellus adults, larvae and pupae respectively were 3515 recorded during the same period (Figure 4.6). 3516

Temperature 10

-5

1/1/2016 1/1/2017 2/1/2016 3/1/2016 4/1/2016 5/1/2016 6/1/2016 7/1/2016 8/1/2016 9/1/2016 2/1/2017

10/1/2016 11/1/2016 12/1/2016 Time (Days) 3517 3518 3519 Figure 4.6: Microclimatic temperature data in austral summer and winter (January 2016 – 3520 February 2017) from a maize field, BCA Farm, Gaborone, Botswana (S24.56449; E25.94883; 3521 994 m.a.s.l), which hosts both Chilo partellus and parasitoid C. flavipes. Temperature data was 3522 recorded using Thermocron iButtons (Dallas Semiconductors, Model DS1920) (0.5°C accuracy; 3523 1 h sampling frequency). The number of events likely eliciting RCH and RHH within the same 3524 timeframe was calculated based on temperature-time interactions. Red and blue dashed lines 3525 represents RCH and RHH temperatures for Cotesia flavipes and Chilo partellus respectively. 3526

140

3527 4.4 Discussion 3528 Global change exerts a major pressure on insect relative fitness, phenology and 3529 demographic patterns with subsequent impact on trophic interactions and ecosystem function 3530 (Bale, 2010; Bannerman et al., 2011; Ma et al., 2014). Indeed, temperature has a critical impact 3531 on specialist predator-prey systems such as parasitoid-host relationships that are important in 3532 agriculture (Bannerman et al., 2011; Romo & Tylianakis, 2013). As a result, thermal limits to 3533 activity, symbolizing the physiological range of host versus parasitoid across thermal gradients 3534 ought to be known. Parasitoid thermal windows and thresholds have direct consequences on 3535 parasitism and hence efficacy of biological control (Romo & Tylianakis, 2013; Schreven et al., 3536 2017). A comprehensive understanding of host and parasitoid basal thermal tolerance coupled 3537 with plasticity thereof is significant in determining potential net agroecosystem responses among 3538 the interacting trophic levels under global climate change (Romo & Tylianakis, 2013; 3539 Mutamiswa et al., 2017b). To my knowledge, this is the first report detailing plasticity of thermal 3540 tolerance of C. partellus and C. flavipes, two key antagonistic insects in staple cereal production 3541 systems of sub-Saharan Africa. 3542

3543 4.4.1 Phenotypic Plasticity of Low Temperature Tolerance 3544 Investigating plasticity of cold tolerance provides a guideline on how natural enemies 3545 released for biocontrol will establish under natural conditions (Hatherly et al., 2009). The current 3546 study reported positive RCH effects in C. partellus and C. flavipes in keeping with previous 3547 studies in D. melanogaster (Overgaard & Sørensen, 2008), Thaumotibia leucotreta (Stotter & 3548 Terblanche, 2009), C. rosa (Nyamukondiwa et al., 2010), Z. vittiger (Nyamukondiwa & 3549 Terblanche, 2010a) and C. pomonella (Chidawanyika & Terblanche, 2011a). This mechanism 3550 improves key fitness traits (Shreve et al., 2004) and may likely help both trophic levels in 3551 surviving sudden cold snaps, typical under global change. By improving cold tolerance through 3552 RCH, C. partellus will modify its relationship with parasitoid C. flavipes. Hence, for the later to 3553 thrive, it should also adaptively respond with phenotypic plasticity matching that of its host in 3554 order to keep their phenologies in sync. Indeed, plasticity follows such patterns, among 3555 interacting trophic levels, and in all cases where plastic responses offer higher fitness benefits 3556 than canalization (Whitman & Ananthakrishnan, 2009). Mechanisms behind RCH responses are 3557 not clear but may include the accumulation of carbohydrate cryoprotectants, membrane

141

3558 restructuring and inhibition of apoptotic cell death (Lee & Denlinger, 2010; reviewed in Teets & 3559 Denlinger, 2013). In the ecosystem, RCH response permits insects to adjust their physiological 3560 condition so that it matches environmental conditions (Yi et al., 2017). Under natural conditions, 3561 insects often face concurrent exposure to several stressors. For instance, cold fronts have been 3562 reported to reduce relative humidity and temperature (Moeller et al., 1993), suggesting that 3563 insects may experience brief chilling and desiccation (Yi et al., 2017) under these environmental 3564 conditions. Given that C. partellus and C. flavipes can cold harden, it likely explains their 3565 survival merit during winter seasons in tropical regions. Furthermore, this implies that RCH may 3566 allow these two species to track diurnal or seasonal environmental temperature fluctuations 3567 (Terblanche et al., 2010) thus enhancing development, courtship behaviours, mating and feeding 3568 during their lifetime (Kelty & Lee, 2001; Shreve et al., 2004). Nevertheless, plastic responses to 3569 low temperature may also be maladaptive (Whitman & Ananthakrishnan, 2009). A comparison 3570 of the host and its parasitoid shows that both species have almost similar levels of survival and 3571 magnitude of plasticity to acute low temperature exposures following pretreatment to sub-lethal 3572 low temperatures. However, these species differ in hardening and discriminating temperatures, 3573 with C. partellus cold hardening at lower temperatures than C. flavipes. Microclimatic 3574 temperature data recorded in an agro-ecosystem where both C. partellus and C. flavipes coexist 3575 showed that temperatures eliciting RCH responses at both trophic levels occur naturally (Figure 3576 4.6). This may imply that under current climate variability (IPCC, 2014), RCH may be an 3577 important mechanism aiding fitness (Yi et al., 2007; Kristensen et al., 2008). This has been true 3578 for some insects, which can rapidly cold harden under ecological ramping rates (Lee et al., 1987; 3579 Nyamukondiwa et al., 2013).

3580 Similar to RCH responses, I also document significant effects of acclimation on CTmin. 3581 Acclimation effects on thermal tolerance may depend on severity of stress exposure and has been 3582 reported to vary across species (Kellett et al., 2005), developmental stages (Terblanche et al., 3583 2007: Marais et al., 2009) and also thermal history (Nyamukondiwa & Terblanche, 2010b). The

3584 current results show that acclimation to low temperature significantly decreased CTmin (thus, 3585 improved cold tolerance) in both host C. partellus and parasitoid C. flavipes, in keeping with 3586 previous studies to date e.g. Glossina pallidipes (Terblanche et al., 2006) and Drosophila species

3587 (Nyamukondiwa et al., 2011). In addition, acclimation to high temperature improved CTmin for 3588 both species, suggesting significant cross-tolerance in corroboration with a previous study on

142

3589 Linepithema humile (Jumbam et al., 2008) and D. melanogaster (Bubliy et al., 2012). This cross- 3590 tolerance suggests the presence of overlapping thermal stress resistance mechanisms (Bubliy et 3591 al., 2012) or signal pathways (‘cross talk’) (see Sinclair et al., 2013). Species comparison of

3592 CTmin between the two trophic levels (Figure 4.2C), and across all acclimation treatments 3593 showed that C. flavipes is more cold tolerant than C. partellus. This may have a positive effect 3594 on C. flavipes survival under variable low temperatures, relative to its herbivore host C. 3595 partellus. Similarly, a comparison of C. partellus pupae and adults showed that indeed they can 3596 also adjust their low temperature tolerance at short timescales (Figure 4.3A; B). Critical thermal 3597 minima and CCRT also showed the same positive acclimation trends for C. partellus pupae and 3598 adults (Figure 4.3C; D). 3599 I also recorded an enhanced CCRT in C. partellus following both low and high 3600 temperature acclimation. However, an improved CCRT in C flavipes only occurred following 3601 low temperature acclimation (Figure 4.2D). The results corroborate findings in D. melanogaster 3602 (Hoffmann et al., 2005) and Acheta domesticus (Lachenicht et al., 2010) where significant 3603 plastic responses for CCRT were recorded following acclimation to summer and winter 3604 conditions. Furthermore, C. flavipes generally recovered faster from chill coma than host C. 3605 partellus suggesting that when faced with chill coma temperatures, the parasitoid has better 3606 survival chances than the host. These experimentally investigated chill-coma temperatures were 3607 frequently detected in nature (Figure 4.6), implying that chill coma recovery may be a key aspect 3608 for the parasitoid to preserve key activities. It has been reported that insects are more prone to 3609 attack by predators during chill-coma (see Alford et al., 2014). I therefore speculate, with 3610 caveats, that the superior chill-coma recovery times prevalent in C. flavipes may improve 3611 predation rates of C. partellus in nature, under variable stressful chill coma temperatures. 3612 The superior enhanced cold tolerance in C. flavipes following acclimation to low 3613 temperatures was also confirmed by much lower SCPs contrary to what was the case with C. 3614 partellus (Figure 4.2E). Even though acclimation to high temperatures significantly improved 3615 cold tolerance in C. partellus, there was no significant difference between the two species after 3616 pretreatments at this thermal regime. Insect species with very low SCPs (~-20°C to -30°C) and 3617 high cold tolerance levels are regarded as freeze intolerant (Bale, 1996). Cotesia flavipes showed 3618 SCP of approximately ~-20°C and high levels of cold tolerance following acclimation to low 3619 temperatures relative to C. partellus. The mechanisms underlying such supercooling capacity

143

3620 have been reported to include body osmolyte accumulation, clearing of gut contents, specific 3621 antifreeze compounds and use of thermal hysteresis proteins (Jones et al., 2008; Denlinger & 3622 Lee, 2010) suggesting that it may be freeze avoiding. Hence, the parasitoid may survive more 3623 acute lower temperatures, a fitness benefit over its host. A comparison of SCPs for C. partellus 3624 developmental stages showed that pupae had the lowest SCPs relative to larvae and adults 3625 (Figure 4.3E). This is consistent with the notion that immobile life-stages with limited capacity 3626 to behaviorally thermoregulate, display more phenotypically plasticity with regards to tolerance 3627 of thermal extremes (Chown & Nicolson, 2004). Consequently, they have evolved an inherent 3628 basal low temperature tolerance (see Lansing et al., 2000; Bowler & Terblanche, 2008). This is 3629 in agreement with Nyamukondiwa et al. (2013), who reported that immobile pupae have inherent 3630 low SCPs. Overall, low temperature plasticity results may confer activity and thus fitness 3631 advantage for C. flavipes relative to its larval host C. partellus. In the absence of reciprocal low 3632 temperature phenotypic plasticity on the part of the host, this represents a positive trend on 3633 biological control under global change (see Agrawal, 2001). 3634

3635 4.4.2 Phenotypic Plasticity of High Temperature Tolerance 3636 Climate scenarios are hypothesized to involve both rising temperatures and frequent rapid 3637 changes in diurnal thermal fluctuations as evidenced by the recent occurrence of unusual heat 3638 waves in many countries (IPCC, 2014). As such, survival of high temperature variability is 3639 critical for survival and interaction of different trophic level species. The current results 3640 document a positive RHH response in the host C. partellus` larvae and adult C. flavipes 3641 parasitoids (Figure 4.4A; B). The temperatures eliciting such hardening responses were 3642 occasionally recorded in the field. Hence such plasticity, or lack thereof, may mediate the host- 3643 parasitoid interactions in tropical agroecosystems in current and future climate change scenarios. 3644 Several studies investigating different insect taxa have also reported such RHH and demonstrated 3645 its potential to improve survival under rapidly changing thermal environments (Kalosaka et al., 3646 2009; Nyamukondiwa et al., 2010; Chidawanyika & Terblanche, 2011a; Chidawanyika et al., 3647 2017). Similarly, RHH also improved survival of C. partellus pupae and adults (Figure 4.5A; B). 3648 Mechanisms eliciting RHH responses include heat shock proteins (HSPs) which prevent the 3649 accumulation of structurally denatured proteins (Hoffmann et al., 2003). In nature, these results 3650 imply that these two interacting species may survive variable thermal environments using RHH

144

3651 responses. This allows them to track diurnal or seasonal environmental temperature fluctuations 3652 (Terblanche et al., 2010; Nyamukondiwa et al., 2013). Although larval C. partellus and C. 3653 flavipes generally have similar magnitude of RHH responses they, nevertheless, vary 3654 significantly in hardening and discriminating temperatures (see Table 4.1). Chilo partellus 3655 discriminating and hardening temperature is higher than that of its endoparasitoid C. flavipes. 3656 This implies that, while both species may acutely adjust their high temperature plasticity, C. 3657 partellus does so, at significantly higher basal temperatures than its parasitoid. This means under 3658 more severe high temperature fluctuations, plasticity alone may not be able to cushion the 3659 parasitoid, since its thermal windows of activity are more benign as compared to those of C. 3660 partellus. This gives the insect pest (host) a survival advantage over the biocontrol agent 3661 (parasitoid), under intense and variable high temperature conditions. Moreover, a look at 3662 microclimate data (Figure 4.6) shows that the parasitoid is more likely to survive under stressful 3663 high temperature conditions (≥35°C), than its larval host (≥41°C). Hence the parasitoid may 3664 suffer more fitness and mortality as a consequence of repeated stressful conditions under natural 3665 conditions, relative to its host C. partellus (see Chown SL and Nicolson, 2004; Schreven et al., 3666 2017).

3667 Acclimation to high temperature generally increased CTmax in C. partellus and C. flavipes 3668 in keeping with previous studies on Ceratitis species (Nyamukondiwa & Terblanche, 2010b; 3669 Weldon et al., 2011) Drosophilids (Nyamukondiwa et al., 2011) and Nezara species (Chanthy et

3670 al., 2012). Mechanisms for improving this CTmax may comprise up-regulation of HSPs 3671 consequently minimising aggregation of denatured proteins following heat stress (Edgerly et al.,

3672 2005). Significant differences in CTmax between C. partellus and C. flavipes across all 3673 acclimation treatments indicate that C. partellus is more heat tolerant than C. flavipes (Figure 3674 4.4D). This means under more drastic thermal variability (e.g. ~45°C), parasitoid C. flavipes may 3675 be more vulnerable as opposed to the insect pest C. partellus, and this has implications on the 3676 ecological balance and pest management using the parasitoid. While C. flavipes is more cold 3677 tolerant than C. partellus and African temperatures expected to increase by 1.5 to 3°C by 2050 3678 (IPCC, 2014), the host stemborer pest is highly likely to survive high temperatures compared to 3679 its parasitoid, and thus emerge ‘winner’ relative to its biocontrol parasitoid. Unless the parasitoid 3680 mounts constitutive and plastic counter-responses to its host, this may create asynchronous 3681 phenologies with negative implications on pest management using the parasitoid.

145

3682 Acclimation to high temperatures increased HKDT in C. flavipes but not in C. partellus, 3683 suggesting adaptive phenotypic plasticity for the parasitoid and not the larval insect pest for

3684 HKDT. This is contrary to CTmax results, which documented adaptive phenotypic plasticity for 3685 both parasitoid and host, although the host has a merit for having an inherent basal high 3686 temperature tolerance. The reason for this trend for HKDT is unknown. However, this may mean 3687 that establishing trophic level differences in thermal sensitivity may be complex and most likely 3688 trait dependent. Nevertheless, acclimation to low temperature significantly improved HKDT in 3689 the insect pest, indicating the presence of cross-tolerance (see Bubliy et al., 2012). This was not 3690 the case with the parasitoid as low temperature acclimation actually resulted in shorter HKDT. 3691 According to theoretical models (Gabriel et al., 2005), insects surviving in more thermal 3692 heterogeneous environments are highly likely to have greater magnitude of plasticity compared 3693 to those from more homogeneous environments. Since climate change is accompanied by 3694 increased climate variability (IPCC, 2014), this implies that acclimation may be a significant 3695 physiological mechanism allowing host-parasitoid associations adjust their phenotype and 3696 develop compensatory mechanisms to survive thermally heterogeneous environments. Basing on 3697 the measurements of CTLs, HKDT and CCRT in larval C. partellus host and parasitoid C. 3698 flavipes, my results generally support the beneficial acclimation hypothesis (Leori et al., 1994). 3699 In conclusion, the current results document plasticity of thermal tolerance in C. partellus and its 3700 larval parasitoid, C. flavipes, and implications for pest management using biological control 3701 under climate change. My results generally suggest that plasticity may enhance survival of these 3702 two species to extreme low and high temperatures. However, parasitoid C. flavipes may have 3703 fitness advantages at low temperatures than the larval C. partellus host, while the latter becomes 3704 fitter at high temperatures, (though different for HKDT). In both cases, differential plastic 3705 responses may symbolize loss of coevolved relationships between the two trophic levels, and 3706 have negative consequences on biological control. These results have two major implications 3707 that are of broader importance: 1) with global climate change, C. partellus may inhabit slightly 3708 warmer environments than its parasitoid C. flavipes (performs better in cooler environments), 2) 3709 that host-parasitoid relationships are complex and likely trait dependent, 3) host- parasitoid 3710 differential plastic responses indicate loss of long coevolved interactions which may offset 3711 efficacy of biocontrol under rapidly changing environments. 3712

146

3713 References 3714 Agrawal, A.A. (2001) Phenotypic plasticity in the interactions and evolution of species. Science, 3715 294, 321-325. 3716 Alford, L., Andrade, T.O., Georges, R., Burel, F. & van Baaren, J. (2014) Could Behaviour and 3717 Not Physiological Thermal Tolerance Determine Winter Survival of Aphids in Cereal 3718 Fields? PLoS ONE, 9(12), e114982. doi:10.1371/journal.pone. 0114982. 3719 Amarasekare, P. & Savage, V. (2012) A framework for elucidating the temperature dependence 3720 of fitness. The American Naturalist, 179,178–191. 3721 Andersen, J.L., Manenti, T., Sørensen, J.G., MacMillan, H.A., Loeschcke, V. & Overgaard, J. 3722 (2014) How to assess Drosophila cold tolerance: chill coma temperature and lower lethal 3723 temperature are the best predictors of cold distribution limits. Functional Ecology, 29, 55– 3724 65. 3725 Austin, C.J. & Moehring, A.J. (2013) Optimal temperature range of a plastic species, Drosophila 3726 simulans. Journal of Animal Ecology, 82, 663–672. 3727 Bale, J.S. (2010) Implications of cold-tolerance for pest management. In: Denlinger, D.L. & Lee, 3728 R.E., Jr. (Eds) Low Temperature Biology of Insects. 342–372. Cambridge University Press, 3729 Cambridge, U.K. 3730 Bale, J.S. (1996) Insect cold hardiness: a matter of life and death. European Journal of 3731 Entomology, 93, 369–382. 3732 Bannerman, J.A., Gillespie, D.R. & Roitberg, B.D. (2011) The impacts of extreme and 3733 fluctuating temperatures on trait-mediated indirect aphid-parasitoid interactions. Ecological 3734 Entomology, 36, 490-498. 3735 Basson, C.H., Nyamukondiwa, C. & Terblanche, J.S. (2012) Fitness costs of rapid cold- 3736 hardening in Ceratitis capitata. Evolution, 66, 296–304. 3737 Bowler, K. &Terblanche, J.S. (2008) Insect thermal tolerance: what is the role of ontogeny, 3738 ageing and senescence? Biological Reviews, 83, 339–355. 3739 Bubliy, O.A., Kristensen, T.N., Kellermann, V. & Loeschcke, V. (2012) Plastic responses to four 3740 environmental stresses and cross-resistance in a laboratory population of Drosophila 3741 melanogaster. Functional Ecology, 26, 245–253. 3742 Chanthy, P., Martin, R.J., Gunning, R.V. & Andrew, N.R. (2012) The effects of thermal 3743 acclimation on lethal temperatures and critical thermal limits in the green vegetable bug,

147

3744 Nezara viridula (L.) (Hemiptera: Pentatomidae). Frontiers in Physiology, 3, 465. doi: 3745 10.3389/fphys.2012.00465. 3746 Chen, I.C., Hill, J.K., Ohlemuller, R., Roy, D.B. & Thomas, C.D. (2011) Rapid range shifts of 3747 species associated with high levels of climate warming. Science, 333, 1024–1026. 3748 Chen, C.P., Lee, R.E. & Denlinger, D.L. (1991) Cold shock and heat shock: a comparison of the 3749 protection generated by brief pre-treatment at less severe temperatures. Physiological 3750 Entomology, 16, 19-26. 3751 Chidawanyika, F., Nyamukondiwa, C., Strathie, L. & Fischer, K. (2017) Effects of thermal 3752 regimes, starvation and age on heat tolerance of the Parthenium Beetle Zygogramma 3753 bicolorata (Coleoptera: Chrysomelidae) following dynamic and static protocols. PLoS ONE, 3754 12, e0169371. doi:10.1371/journal.pone.0169371. 3755 Chidawanyika, F. & Terblanche, J.S. (2011a) Rapid thermal responses and thermal tolerance in 3756 adult codling moth Cydia pomonella (Lepidoptera: Totricidae). Journal of Insect 3757 Physiology, 57, 108-117. 3758 Chidawanyika, F. & Terblanche, J.S. (2011b) Costs and benefits of thermal acclimation for 3759 codling moth, Cydia pomonella (Lepidoptera: Tortricidae): implications for pest control and 3760 the sterile insect release programme. Evolutionary Applications, 4, 534-544. 3761 Chown, S.L. & Nicolson, S.W. (2004) Insect Physiological Ecology: Mechanisms and Patterns. 3762 Oxford University Press, Oxford. 3763 Chown, S.L. & Terblanche, J.S. (2007) Physiological Diversity in Insects: Ecological and 3764 Evolutionary Contexts. Advances in Insect Physiology, 33, 50-152. 3765 Colinet, H. & Hoffman, A.A. (2012) Comparing phenotypic effects and molecular correlates of 3766 developmental, gradual and rapid cold acclimation responses in Drosophila melanogaster. 3767 Functional Ecology, 26, 84-93. 3768 David, J.R., Allemand, R., Capy, P., Chakir, M., Gibert, P.., Pétavy, G. & Moreteau, B. (2004) 3769 Comparative life histories and ecophysiology of Drosophila melanogaster and D. simulans. 3770 Genetica, 120,151–163. 3771 Dejen, A., Getu, E., Azerefegne, F. & Ayalew, A. (2013) Distribution and extent of Cotesia 3772 flavipes Cameron (Hymenoptera: Braconidae) parasitism in Northeastern Ethiopia. 3773 International Journal of Tropical Insect Science, 5, 9–19.

148

3774 Denlinger, D.L. & Lee, R.E. (2010) Low Temperature Biology of Insects. Cambridge University 3775 Press, Cambridge. 3776 Dillon, M.E., Wang, G., Garitty, P.A. & Huey, R.B. (2009) Thermal preference in Drosophila. 3777 Journal of Thermal Biology, 34, 109-119.

3778 Dyer, L.A., Richards, L.A., Short, S.A. & Dodson, C.G. (2013) Effects of CO2 and temperature 3779 on tritrophic interactions. PLoS ONE, 8(4), e62528. Doi:10.1371/journal.pone.0062528. 3780 Edgerly, J.S., Tadimalla, A. & Dahlhoff, E.P. (2005) Adaptation to thermal stress in lichen- 3781 eating webspinners (Embioptera): habitat choice, domicile construction and the potential 3782 role of heat shock proteins. Functional Ecology, 19, 255-262. 3783 Fischer, K., Dierks, A., Franke, K., Geister, T.L., Liszka, M., Winter, S. & Pflicke, C. (2010), 3784 Environmental effects on temperature stress resistance in the tropical butterfly Bicyclus 3785 anynana. PLoS ONE, 5(12), e15284. doi:10.1371/journal.pone.0015284. 3786 Gabriel, W., Luttbeg, B., Sih, A. & Tollrian, R. (2005) Environmental tolerance, heterogeneity, 3787 and the evolution of reversible plastic responses. The American Naturalist, 166, 339–353. 3788 Grimm, N.B., Chapin III, F.S., Bierwagen, B., Gonzalez, P., Groffman, P.M., Luo, Y., Melton, 3789 F., Nadelhoffer, K., Pairis, A., Raymond, P.A., Schimel, J. & Williamson, C.E. (2013) The 3790 impacts of climate change on ecosystem structure and function. Frontiers in Ecology and 3791 the Environment, 11, 474-482. 3792 Hatherly, I.S., Pedersen, B.P. & Bale, J.S. (2009) Effect of host plant, prey species and 3793 intergenerational changes on the prey preferences of the predatory mirid Macrolophus 3794 caliginosus. Biological Control, 54, 35-45. 3795 Hemmati, C., Moharramipour, S. & Talebi, A.A. (2014) Effects of cold acclimation, cooling rate 3796 and heat stress on cold tolerance of the potato tuber moth Phthorimaea operculella 3797 (Lepidoptera: Gelechiidae). European Journal of Entomology 111, 487-494. 3798 Hoffmann, A.A., Chown, S.L. & Clusella-Trullas, S. (2012) Upper thermal limits in terrestrial 3799 ectotherms: how constrained are they? Functional Ecology, 27, 934–949. 3800 Hoffmann, A.A., Shirriffs, J. & Scott, M. (2005) Relative importance of plastic versus genetic 3801 factors in adaptive differentiation: geographical variation for stress resistance in Drosophila 3802 melanogaster from eastern Australia. Functional Ecology, 19, 222–227.

149

3803 Hoffmann, A.A., Sørensen, J.G. & Loeschcke, V. (2003) Adaptation of Drosophila to 3804 temperature extremes: bringing together quantitative and molecular approaches. Journal of 3805 Thermal Biology, 28, 175–216. 3806 Hance, T., Van Baaren, J., Vernon, P. & Boivin, G. (2007) Impact of extreme temperatures on 3807 parasitoids in a climate change perspective. Annual Review of Entomology, 52,107–126. 3808 IPCC. (2014) Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and 3809 III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Core 3810 Writing Team, R.K. Pachauri and L.A. Meyer (eds.)]. IPCC, Geneva, Switzerland, pp 151. 3811 Jones, D.B., Giles, K.L. & Elliot, N.C. (2008) Supercooling points of Lysiphlebus testaceipes 3812 and its host Schizaphis graminum. Environmental Entomology, 37, 1063-1068. 3813 Jumbam, K.R., Jackson, S., Terblanche, J.S., McGeoch, M.A. & Chown, S.L. (2008) 3814 Acclimation effects on critical and lethal thermal limits of workers of the Argentine ant, 3815 Linepithema humile. Journal of Insect Physiology, 54, 1008–1014. 3816 Kalosaka, K., Soumaka, E., Politis, N. & Mintzas, A.C. (2009) Thermotolerance and Hsp70 3817 expression in Mediterranean fruit fly Ceratitis capitata. Journal of Insect Physiology, 55, 3818 568-573. 3819 Kellett, M., Hoffmann, A.A. & McKechnie, S.W. (2005) Hardening capacity in the Drosophila 3820 melanogaster species group is constrained by basal thermotolerance. Functional Ecology, 3821 19, 853–858. 3822 Kelty, J.D. & Lee, R.E.Jr. (2001) Rapid cold-hardening of Drosophila melanogaster (Diptera: 3823 Drosophiladae) during ecologically based thermoperiodic cycles. Journal of Experimental 3824 Biology, 204, 1659-1666. 3825 Kfir, R. (1997) Competitive displacement of Busseola fusca (Lepidoptera: Noctuidae) by Chilo 3826 partellus (Lepdoptera: Pyralidae). Annals of the Entomological Society of America 90, 619- 3827 624. 3828 Khadioli, N., Tonnang, Z.E.H., Muchugu, E., Ong’amo, G., Achia, T., Kipchirchir, I., Kroschel, 3829 J. & Le Ru, B. (2014) Effect of temperature on the phenology of Chilo partellus (Swinhoe) 3830 (Lepidoptera: Crambidae), simulation and visualization of the potential future distribution of 3831 C. partellus in Africa under warmer temperatures through the development of life-table 3832 parameters. Bulletin of Entomological Research, 104, 809–822.

150

3833 Kleynhans, E., Conlong, D.E. & Terblanche, J.S. (2014) Host plant-related variation in thermal 3834 tolerance of Eldana saccharina. Entomologia Experimentalis et Applicata, 150(2), 113-122. 3835 DOI: 10.1111/eea.12144. 3836 Kristensen, T.N., Hoffmann, A.A., Overgaard, J., Sørensen, J.G. & Hallas, R. (2008) Costs and 3837 benefits of cold acclimation in field released Drosophila. Proceedings of National Academy 3838 of Sciences of the United States of America, 105, 216-221. 3839 Lachenicht, M.W., Clusella-Trullas, S., Boardman, L., Le Roux, C. & Terblanche, J.S. (2010) 3840 Effects of acclimation temperature on thermal tolerance, locomotion performance and 3841 respiratory metabolism in Acheta domestica L. (Orthoptera: Gryllidae). Journal of Insect 3842 Physiology, 56, 822–830. 3843 Lansing, E., Justesen, J. & Loeschcke, V. (2000) Variation in the expression of HSP70, the 3844 major heat-shock protein, and thermotolerance in larval and adult selection lines of 3845 Drosophila melanogaster. Journal of Thermal Biology, 25, 443-450. 3846 Lee, R.E., Chen, C.P. & Denlinger, D.L. (1987) A rapid cold-hardening in insects. Science 238, 3847 1415-1417. 3848 Lee, R.E. & Denlinger, D.L. (2010) Rapid cold-hardening: ecological significance and 3849 underpinning mechanisms. In: Low Temperature Biology of Insects (D.L. Denlinger & R.E. 3850 Lee, eds), Cambridge University Press, UK. pp. 35–58. 3851 Leori, A.M., Bennett, A.F. & Lenski, R.E. (1994) Temperature acclimation and competitive 3852 fitness: an experimental test of the beneficial acclimation assumption. Proceedings of 3853 National Academy of Sciences of the United States of America, 91, 1917-1921. 3854 Ma, F.Z., Lu, Z.C., Wang, R. & Wan, F.H. (2014) Heritability and evolutionary potential in 3855 thermal tolerance traits in the invasive Mediterranean cryptic species of Bemisia tabaci 3856 (Hemiptera: Aleyrodidae). PLoS ONE, 9, e103279. doi:10.1371/journal.pone.0103279. 3857 Marais, E., Terblanche, J.S. & Chown, S.L. (2009) Life-stage related differences in hardening 3858 and acclimation of thermal tolerance traits in the kelp fly, Paractora dreuxi (Diptera: 3859 Helcomyzidae). Journal of Insect Physiology, 55, 36-343. 3860 Milton, C.C. & Partridge, L. (2008) Brief carbon dioxide exposure blocks heat hardening but not 3861 cold acclimation in Drosophila melanogaster. Journal of Insect Physiology, 54,32-40.

151

3862 Moeller, C.C., Huh, O.K., Roberts, H.H., Gumley, L.E. & Menzel, W.P. (1993) Response of 3863 Louisiana coastal environments to a cold front passage. Journal of Coastal Research, 9(2), 3864 434–447. 3865 Mutamiswa, R., Chidawanyika, F. & Nyamukondiwa, C. (2017a) Comparative assessment of the 3866 thermal tolerance of spotted stemborer, Chilo partellus Swinhoe (Lepidoptera: Crambidae) 3867 and its larval parasitoid, Cotesia sesamiae Cameron (Hymenoptera: Braconidae). Insect 3868 Science. DOI 10.1111/1744-7917.12466. 3869 Mutamiswa, R., Chidawanyika, F. & Nyamukondiwa, C. (2017b) Dominance of spotted 3870 stemborer Chilo partellus Swinhoe (Lepidoptera: Crambidae) over indigenous stemborer 3871 species in Africa’s changing climates: ecological and thermal biology perspectives. 3872 Agricultural and Forest Entomology, DOI: 10.1111/afe.12217. 3873 Nguyen, C., Bahar, M.H., Baker, G. & Andrew, N.R. (2014) Thermal tolerance limits of 3874 diamondback moth in ramping and plunging assays. PLoS ONE, 9, e87535. 3875 doi:10.1371/journal.pone.0087535. 3876 Nyamukondiwa, C., Kleynhans, E. & Terblanche, J.S. (2010) Phenotypic plasticity of thermal 3877 tolerance contributes to the invasion potential of Mediterranean fruit flies (Ceratitis 3878 capitata). Ecological Entomology, 35, 565-575. 3879 Nyamukondiwa, C. & Terblanche, J.S. (2010a) Rapid cold hardening in Zapprionus vittiger 3880 (Coquillett) (Diptera: Drosophilidae). CryoLetters, 31, 504-512. 3881 Nyamukondiwa, C. & Terblanche, J.S. (2010b) Within-generation variation of critical thermal 3882 limits in adult Mediterranean and Natal fruit flies Ceratitis capitata and Ceratitis rosa: 3883 thermal history affects short-term responses to temperature. Physiological Entomology, 35, 3884 255-264. 3885 Nyamukondiwa, C. & Terblanche, J.S. (2009) Thermal tolerance in adult Mediterranean and 3886 Natal fruit flies (Ceratitis capitata and Ceratitis rosa): effects of age, gender and feeding 3887 status. Journal of Thermal Biology, 34, 406-414. 3888 Nyamukondiwa, C., Terblanche, J.S., Marshall, K.E. & Sinclair, B.J. (2011) Basal but not heat 3889 tolerance constraints plasticity among Drosophila species (Diptera: Drosophilidae). Journal 3890 of Evolutionary Biology, 24, 1927-1938. 3891 Nyamukondiwa, C., Weldon, C.W., Chown, S.L., le Roux, P.C. & Terblanche, J.S. (2013) 3892 Thermal biology, population fluctuations and implications of temperature extremes for the

152

3893 management of two globally significant insect pests. Journal of Insect Physiology, 59,1199- 3894 1211. 3895 Ochieng, R.S., Onyango, F.O. & Bungu, M.D.O. (1985) Improvement of techniques for mass 3896 culture of Chilo partellus (Swinhoe). Insect Science and its Application, 6, 425-428. 3897 Overgaard, J., Kearney, M.R. & Hoffmann, A.A. (2014a) Sensitivity to thermal extremes in 3898 Australian Drosophila implies similar impacts of climate change on the distribution of 3899 widespread and tropical species. Global Change Biology, 20, 1738–1750. 3900 Overgaard, J. & Sørensen, J.G. (2008) Rapid thermal adaptation during field temperature 3901 variations in Drosophila melanogaster. Cryobiology, 56, 159–162. 3902 Overgaard, J., Sørensen, J.G., Com, E. & Colinet, H. (2014b) The rapid cold hardening response 3903 of Drosophila melanogaster: complex regulation across different levels of biological 3904 organization. Journal of Insect Physiology, 62, 46-53. 3905 Parmesan, C. (2006) Ecological and evolutionary responses to recent climate change. Annual 3906 Review of Ecology, Evolution and Systematics, 37, 637-669. 3907 Rajamohan, A. & Sinclair, B.J. (2008) Short term hardening effects on survival of acute and 3908 chronic cold exposure by Drosophila melanogaster larvae. Journal of Insect Physiology, 54, 3909 708-718. 3910 Romo, C.M. & Tylianakis, J.M. (2013) Elevated temperatures and drought interact to reduce 3911 parasitoid effectiveness in suppressing hosts. PLoS ONE, 8(3), e58136. doi: 3912 10.1371/journal.pone.0058136. 3913 Shreve, S.M., Kelty, J.D. & Lee, R.E. (2004) Preservation of reproductive behaviors during 3914 modest cooling: rapid cold-hardening fine-tunes organismal response. Journal of 3915 Experimental Biology, 207, 1797-1802. 3916 Schreven, S.J.J., Frago, E., Stens, A., de Jong, P.W. & van Loon, J.J.A. (2017) Contrasting 3917 effects of heat pulses on different trophic levels, an experiment with a herbivore-parasitoid

3918 model system. PLoS ONE, 12(4), e0176704. 3919 Sgrò, C.M., Terblanche, J.S. & Hoffmann, A.A. (2016) What can plasticity contribute to insect 3920 responses to climate change? Annual Review of Entomology. DOI: 10.1146/annurev-ento- 3921 010715-023859.

153

3922 Sinclair, B.J., Ferguon, L.V., Salehipour-shirazi, G. & MacMillan, H.A. (2013) Cross-tolerance 3923 and cross-talk in the cold: Relating low temperatures to Desiccation and immune stress in 3924 insects. Integrative and Comparative Biology, 53, 545-56. 3925 Songa, J.M., Overholt, W.A., Mueke, J.M. & Okell, R.O. (2001) Colonization of Cotesia 3926 flavipes (Hymenoptera: Braconidae) in stem borers in the semi-arid Eastern province of 3927 Kenya. Insect Science and its Application, 21, 289–295. 3928 Sorte, C.J.B., Jones, S.J. & Miller, L.P. (2011) Geographic variation in temperature tolerance as 3929 an indicator of potential population responses to climate change. Journal of Experimental 3930 Marine Biology and Ecology, 400, 209-217. 3931 Stotter, R.L. & Terblanche, J.S. (2009) Low temperature tolerance of false codling moth 3932 Thaumotibia leucotreta (Meyrick) (Lepidoptera: Tortricidae) in South Africa. Journal of 3933 Thermal Biology, 34, 320–325. 3934 Tams, W.H.T. (1932) New species of African Heterocera. Entomologist, 65, 1241–1249. 3935 Teets, N.M. & Denlinger, D.L. (2013) Physiological mechanisms of seasonal and rapid cold- 3936 hardening in insects. Physiological Entomology, 38,105-116. 3937 Tefera, T., Mugo, S., Tende, R. & Likhayo, P. (2010) Mass Rearing of Stem borers, Maize 3938 weevil and Larger grain borer Insect Pests of Maize. CIMMYT. Nairobi. 3939 Terblanche, J.S., Clusella-Trullas, S., Deere, J.A. & Chown, S.L. (2008) Thermal tolerance in a 3940 south-east African population of tsetse fly Glossina pallidipes (Diptera: Glossinidae): 3941 implications for forecasting climate change impacts. Journal of Insect Physiology, 54, 14– 3942 127. 3943 Terblanche, J.S., Deere, J.A., Clusella-Trullas, S., Janion, C. & Chown, S.L. (2007) Critical 3944 thermal limits depend on methodological context. Proceedings of the Royal Society of 3945 London B, 274, 2935–2943. 3946 Terblanche, J.S., Hoffmann, A.A., Mitchell, K.A., Rako, L., le Roux, P.C. & Chown, S.L. (2011) 3947 Ecologically relevant measures of tolerance to potentially lethal temperatures. Journal of 3948 Experimental Biology, 214, 3713-3725. 3949 Terblanche, J.S., Klok, C.J., Krafsur, E.S. & Chown, S.L. (2006) Phenotypic plasticity and 3950 geographic variation in thermal tolerance and water loss of the tsetse Glossina pallidipes 3951 (Diptera: Glossinidae): implications for distribution modelling. TheAmerican Journal of 3952 Tropical Medicine and Hygiene, 74, 786–794.

154

3953 Terblanche, J.S., Nyamukondiwa, C. & Kleynhans, E. (2010) Thermal variability alters climatic 3954 stress resistance and plastic responses in a globally invasive pest, the Mediterranean fruit fly 3955 (Ceratitis capitata). Entomologia Experimentalis et Applicata, 137, 304–315. 3956 Weldon, C.W., Terblanche, J.S. & Chown, S.L. (2011) Time-course for attainment and reversal 3957 of acclimation to constant temperature in two Ceratitis species, Journal of Thermal Biology, 3958 36, 479-485. 3959 Whitman, D.W. & Ananthakrishnan, T.N. (2009) Phenotypic plasticity of insects: mechanisms 3960 and consequences. Science Publishers. Enfield, NH. 3961 Woods, H.A. & Harrison, J.F. (2002) Interpreting rejections of the beneficial acclimation 3962 hypothesis: when is physiological plasticity adaptive? Evolution, 56, 1863–1866. 3963 Yang, L.H. & Rudolf, V.H.W. (2010) Phenology, ontogeny and the effects of climate change on 3964 the timing of species interactions. Ecology Letters, 13, 1-10. 3965 Yi, S.X., Gantz, J.D. & Lee, R.E. Jr. (2017) Desiccation enhances rapid cold-hardening in the 3966 flesh fly Sarcophaga bullata: evidence for cross tolerance between rapid physiological 3967 responses. Journal of Comparative Physiology B, 187, 79–86. 3968 Yi, S.X., Moore, C.W. & Lee, R. (2007) Rapid cold-hardening protects Drosophila 3969 melanogaster from cold-induced apoptosis. Apoptosis, 12, 1183-1193. 3970 3971 3972

3973

3974

155

3975

3976

3977 CHAPTER 5

3978

3979

3980

3981

3982

3983

3984 Thermal resilience may shape population abundance of two sympatric congeneric Cotesia

3985 species (Hymenoptera: Braconidae) *

3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 * Published as “Mutamiswa, R., Machekano, H., Chidawanyika, F. & Nyamukondiwa, C. 3998 (2018) Thermal resilience may shape population abundance of two sympatric congeneric Cotesia 3999 species (Hymenoptera: Braconidae). PLOS ONE. doi.org/10.1371/journal.pone.0191840

156

4000 5.1 Introduction 4001 Abiotic factors such as temperature and relative humidity have a direct effects on 4002 development, reproduction, abundance, biogeography (Hoffman et al., 2012; Nguyen et al., 4003 2014) and survival of ectotherms (Loeschcke & Hoffmann, 2007; Ma et al., 2014), including 4004 parasitoids (Thomas & Blanford, 2003; Hance et al., 2007). Of these, temperature is considered 4005 the predominant abiotic factor affecting both herbivorous insects and their antagonistic 4006 biological control agents (Bale et al., 2002; Romo & Tylianakis, 2013). Given that most insects 4007 are ectotherms, and their population growth is temperature driven, extreme temperatures have a 4008 bearing on their rate of development as well as biochemical and physiological processes 4009 (Nyamukondiwa & Terblanche, 2009). Since most insects have a limited ability to control body 4010 temperature, they have developed a range of mechanisms for survival under stressful thermal 4011 environments (Bale & Hayward, 2009). Some of the mechanisms involve behavioral avoidance 4012 (Bale & Hayward, 2009), morphological adaptations (Nyamukondiwa et al., 2010) as well as 4013 daily (Overgaard & Sorenson, 2008; Nyamukondiwa et al., 2013) and seasonal (Khani & 4014 Moharramipour, 2010; Chidawanyika & Terblanche, 2011a) thermal tolerance adjustment. 4015 However, failure to employ some compensatory mechanisms to survive these extreme 4016 temperatures may offset fitness traits, ultimately leading to limited activity and poor performance 4017 of life-history traits leading to population decline and seldom species extinction (Chown & 4018 Terblanche, 2007). Insect temperature tolerance is typically not static (Nyamukondiwa & 4019 Terblanche, 2009; Stotter & Terblanche, 2009), and may be influenced by a range of factors 4020 including age (Nyamukondiwa & Terblanche, 2009), developmental stage (Bowler & 4021 Terblanche, 2008; Nyamukondiwa & Terblanche, 2009), thermal history (Nyamukondiwa & 4022 Terblanche, 2010; Chidawanyika et al., 2017) and species ecological and evolutionary history 4023 (e.g. temperate vs. tropical environment) (Kleynhans et al., 2014). These factors, combined with 4024 complex interactions between duration and severity of exposure may determine an insect’s 4025 thermal tolerance with longer or more severe exposures typically lethal (Chown & Nicholson, 4026 2004; Stotter & Terblanche, 2009). Thermal life history and physiological tolerances vary in 4027 insects with some species emerging ‘winners’ whilst others succumb to detrimental effects 4028 (Parmesan, 2006; Mutamiswa et al., 2017a) thus disrupting their phenological and biogeographic 4029 patterns (Bale et al., 2002). Parasitoids such as Cotesia sesamiae and Cotesia flavipes Cameron 4030 (Hymenoptera: Braconidae) have different evolutionary and thermal history patterns, even

157

4031 though they have been reported to coexist occupying the same ecological niche (Overholt et al., 4032 2000; Zhou et al., 2003; Mailafiya et al., 2011). This competitive interaction has been beneficial 4033 in African agriculture as the combined effect of the two species has led to increased pest 4034 suppression (Zhou et al., 2001; Kfir et al., 2002). However of concern, is how population 4035 decreases in one species following environmental perturbations e.g. thermal stress may affect 4036 host stemborer abundance outbreaks, and the likely costs of increased pest pressure. 4037 Cotesia flavipes is a gregarious larval endoparasitoid of lepidopteran stemborers (Manjoo 4038 & Bajpai, 2011). It is native to Asia and is the closest coevolved larval parasitoid of Chilo 4039 partellus Swinhoe (Padmaja & Prabhakar, 2004; Divyal et al., 2009; Dejen et al., 2013). 4040 Following its first release in Kenya in 1993 (Dejen et al., 2013), it was subsequently introduced 4041 in various east and southern African countries where it successfully controlled C. partellus in 4042 maize and sorghum (Assefa et al., 2008). On the other hand, C. sesamiae is indigenous to Africa 4043 and endemic to the sub-Saharan region. It belongs to the C. flavipes monophyletic complex 4044 (Kimani-Njogu & Overholt 1997; Muirhead et al., 2012) and is a generalist gregarious 4045 endoparasitoid of the entire lepidopteran cereal stemborer complex (Mailafiya et al., 2010). 4046 Cotesia flavipes and C. sesamiae have been reported to reduce lepidopteran stemborer densities 4047 between 32 and 55% in agroecosystems and are of importance in cereal farming in sub-Saharan 4048 Africa (Zhou et al., 2001; Kfir et al., 2002), in particular, maize (Zea mays L.) and sorghum 4049 Sorghum bicolor (L.) Moench which are the staple crops in large parts of this region. The 4050 distribution of these two Cotesia species is influenced by climate, with C. sesamiae common in 4051 wetter regions and C. flavipes common in warm and dry regions (Songa et al., 2001; Niyibigira, 4052 2003). 4053 Due to the discrepancies in both evolutionary origin and habitat preferences among these 4054 parasitoids, I sought to investigate survival and functional activity limits under acute thermal 4055 variability, typical in African agroecosystems. With global mean temperatures projected to 4056 increase by 1.4 to 5.8°C by 2100 (Walther et al., 2002; Karuppaiah & Sujayanad, 2012), coupled 4057 with increased frequencies of heat waves and cold snaps (IPCC, 2014), a comprehensive 4058 understanding of the thermal tolerance and performance of parasitoid species (Hance et al. 2002; 4059 Mutamiswa et al., 2017b) is of paramount importance for effective prediction of climate change 4060 impacts on insect population dynamics (Ward & Masters, 2007), extinction (Thomas et al., 2004; 4061 Amarasekare & Savage, 2012) and efficacy of biological control (Alford et al., 2016). Previous

158

4062 studies reported that climate change may interrupt host-parasitoid relationships due to 4063 differences in thermal preferences thus increasing the risk of host outbreaks (Voigt et al., 2003; 4064 reviewed in Hance et al., 2007; de Sassi & Tylianakis, 2012, Mutamiswa et al., 2017a). While 4065 several studies have assayed thermal tolerance in insect pests (e.g. Koveos, 2001; Grout & Stoltz, 4066 2007; Nyamukondiwa & Terblanche, 2009; Chidawanyika & Terblanche, 2011a; Piyaphongkul, 4067 2013; Nguyen et al., 2014; Overgaard et al., 2014) and recently C. partellus and its parasitoid C. 4068 sesamiae (Mutamiswa et al., 2017a), to my knowledge, no studies have investigated the thermal 4069 tolerance of coexisting parasitoids of importance in agriculture. Here, I investigated basal 4070 thermal tolerance of indigenous and exotic larval endoparasitoids (C. sesamiae and C. flavipes, 4071 respectively) of lepidopteran cereal stemborers and implications on geographic distribution and 4072 biological control under climate change. I investigated ability to tolerate extreme temperatures 4073 (lower and upper) following exposure to different ranges of temperature-time interactions, lower 4074 and upper critical limits to activity, supercooling points, time to heat knock-down and recovery 4075 following chill coma. Since C. sesamiae is predominant in warm wetter regions (Songa et al., 4076 2001; Niyibigira, 2003), indigenous to Africa and attacks a wide range of stemborer species 4077 (Mailafiya et al., 2009; 2010), I hypothesised that it has a higher basal thermal tolerance than the 4078 exotic C. flavipes. An in-depth understanding of the species` basal thermal tolerance is of 4079 importance in pest management including site selection for initial parasitoid release (based on 4080 thermal preferences), timing of augmentative releases and general designing of efficacious 4081 biological control programs for lepidopteran stemborers in changing climate. The study is also 4082 significant in explaining how introduced exotic parasitoids in classical biological control may 4083 shape ecosystem function and competitive ability of congeneric species occupying the same 4084 niche under dynamic environmental conditions. 4085

4086 5.2 Materials and Methods 4087

4088 5.2.1 Study Organisms and Rearing Conditions 4089 Initial colonies of C. sesamiae and C. flavipes were obtained from South African 4090 Sugarcane Research Institute (SASRI), South Africa and International Centre for Insect 4091 Physiology and Ecology (ICIPE), Kenya, respectively. These insects had been in the laboratory

159

4092 for more than 20 generations with regular supplementation with wild populations to maintain 4093 heterozygosity. Cotesia sesamiae colony was created using parasitized S. calamistis larvae whilst 4094 parasitised C. partellus larvae where used for C. sesamiae colony. Both colonies were 4095 maintained in the climate chambers (HPP 260, Memmert GmbH + Co.KG, Germany) under 4096 12L:12D photoperiod, 28 ± 1°C and 65 ± 10% RH on artificial diet (Ochieng et al., 1985; Tefera 4097 et al., 2010) in 30 ml plastic vials with perforated screw-cap lids. Emerging parasitoid cocoons 4098 were separated and held according to species, under similar conditions in open Petri dishes 4099 placed in Bugdorm rearing cages (240mm3; Bugdorm-BD43030F, Megaview Science Co., Ltd, 4100 Taiwan) until eclosion. Eclosed parasitoids had access to food ad libitum (25% honey: water 4101 from a cotton wick) until they were used in thermal tolerance assays as 24–48 h-old adults. 4102

4103 5.2.2 Lower and Upper Lethal Temperature Assays 4104 Using a direct plunge protocol in programmable water baths (Systronix, Scientific, South 4105 Africa), containing a mixture of propylene glycol and water (1:1 ratio to enable sub-zero 4106 temperature operation without freezing), LLT and ULT were assayed using established protocols 4107 by (Terblanche et al., 2008; Chidawanyika & Terblanche, 2011a; Mutamiswa et al., 2017a). In 4108 brief, ten 24-48 h old adults from each species (C. sesamiae and C. flavipes) (n = 50) were 4109 placed in 30ml polypropylene vials with perforated screw-cap lids and placed in water tight zip- 4110 lock bag which was submerged in a programmable water bath for each temperature/time 4111 treatment for either ULT or LLT assays. Digital thermometers (Fluke 53/54IIB, Fluke 4112 Cooperation, USA) were used to monitor the temperature of the water bath for the duration of 4113 each treatment. Post treatment, propylene vials containing assayed insects were placed in a 4114 climate chamber (28 ± 1°C, 65 ± 10% RH and 12:12 L:D photocycle, and supplied with food 4115 (25% honey: water from a cotton wick) during the whole recovery period. Survival was then 4116 recorded 24 hours post treatment. For the purposes of this study, survival was defined as 4117 coordinated muscle response to stimuli such as gentle prodding, or normal behaviors such as 4118 feeding, flying or mating (24 h post treatment). ULT assays for both species ranged from 35 to 4119 44°C at 0.5, 1, 2 and 4 h duration of exposure or until 0-100% mortality was recorded. LLT 4120 assays also used a range of temperature × time combinations that realized 0-100% mortality from 4121 -15 to 5°C for 0.5, 1, 2 and 4 h. In all the assays 24–48 h-old adult parasitoids of mixed sex were

160

4122 used, since sex seem to play no significant role in thermal tolerance traits (see discussions in 4123 Nyamukondiwa & Terblanche, 2009). 4124

4125 5.2.3 Critical Thermal Limits (CTLs)

4126 Critical thermal limits (CTmin and CTmax) were measured using a dynamic protocol as 4127 outlined by Nyamukondiwa & Terblanche (2009). Ten individual adults of C. sesamiae or C. 4128 flavipes (24-48 h old) were randomly placed in a series of insulated double-jacketed chambers 4129 (‘organ pipes’) connected to a programmable water bath (Lauda Eco Gold, Lauda DR.R. Wobser 4130 GMBH and Co. KG, Germany) filled with 1:1 water: propylene glycol and subjected to a 4131 constant cooling or heating rate. In the ‘organ pipes’, insects were first given 10 minutes to 4132 equilibrate at 28°C (equivalent to the benign rearing temperature) before ramping temperature up ° −1 4133 (CTmax) or down (CTmin) at a rate of 0.25 C min . This was repeated twice to yield sample sizes 4134 of n = 20 per treatment. A thermocouple (type T 36 SWG) connected to a digital thermometer 4135 (53/54IIB, Fluke Cooperation, USA) was inserted into the control chamber to monitor chamber 4136 temperature. Individual insect body temperature was assumed to be in equilibrium with the 4137 chamber temperature under experimental conditions as employed in other similar sized insect

4138 taxa (Terblanche et al. 2007). The CTmax and CTmin were defined as the temperature at which 4139 each individual insect lost coordinated muscle function, consequently losing the ability to 4140 respond to mild prodding (e.g. Nyamukondiwa & Terblanche, 2010). 4141

4142 5.2.4 Supercooling Points (SCPs) 4143 Adult parasitoids (24-48 h old) SCPs were assayed as outlined by Nyamukondiwa et al. 4144 (2013). For each species, sixteen insects were individually placed in 0.65 ml microcentrifuge 4145 tubes where each insect was placed in contact with the tip of a type-T copper-constantan 4146 thermocouple (762-1121, Cambridge, UK). The thermocouples were inserted through the lid of 4147 each tube with both insect and thermocouple secured in place by a cotton wool. All 4148 thermocouple readings were taken via two 8-channel Picotech TC-08 (Pico Technology, 4149 Cambridge, UK) that relayed information to a computer equipped with PicoLog software for 4150 windows (Pico Technology, Cambridge, UK). Temperatures were continuously monitored and 4151 recorded at 1s intervals. In all cases, experiments commenced at a set point temperature of 15°C

161

4152 for 10 mins (to allow equilibration of insect body temperatures) before ramping down at 0.5°C 4153 min-1 until SCPs were recorded. In the current study, SCP for each individual was defined as the 4154 lowest temperature recorded prior to a spike in temperature associated with the latent heat of 4155 crystallization (see Nyamukondiwa et al., 2013). 4156

4157 5.2.5 Chill Coma Recovery Time (CCRT) 4158 For both species, CCRTs were assayed as outlined by Weldon et al. (2011). A total of 10 4159 replicate adults (24-48 h old) were placed individually in 0.65ml microcentrifuge tubes and then 4160 loaded into a large zip-lock bag which was subsequently submerged into a water bath (Systronix, 4161 Scientific, South Africa). The water bath, filled with 1:1 water: propylene glycol mixture, was set 4162 at 0°C for 1 hour. This temperature × time treatment has been shown to elicit chill-coma in other 4163 like insect taxa (see Weldon et al., 2011; Sinclair et al., 2015). After 1 hour at chill-coma 4164 temperature, the tubes were removed from the water bath and transferred to a Memmert climate 4165 chamber set at 28°C, 65% RH for recovery. The chamber was connected to a camera (HD Covert 4166 Network Camera, DS-2CD6412FWD-20, Hikvision Digital Technology Co., Ltd, China) that 4167 was linked to a computer where observations were recorded. CCRT was defined as the time (in 4168 minutes) required for an adult to stand upright on its legs (Milton & Partridge, 2008). This was 4169 repeated three times to yield sample sizes of n = 30 per treatment. 4170

4171 5.2.6 Heat Knock down Time (HKDT) 4172 HKDTs for both species were assayed following Weldon et al. (2011). Ten replicate 4173 adult parasitoids (24 to 48 hrs) were individually placed in 0.65ml microcentrifuge tubes and 4174 placed in a climate chamber connected to a camera linked to a computer as in CCRT. The tubes 4175 carrying the parasitoids were then exposed to a test temperature of 45±0.3°C, 65% RH in the 4176 climate chamber. This knockdown temperature (45°C) was selected basing on preliminary 4177 investigations of upper critical temperatures to activity ranging 39.5±0.99°C and 44.6±0.63°C for 4178 C. sesamiae and C. flavipes, respectively. This was repeated three times to yield sample sizes of 4179 n = 30 for each species. All observations were made from the climate chamber video recording 4180 system. HKDT was defined as the time (in minutes) at which organisms lost activity following 4181 exposure to 45°C in the climate chamber.

162

4182 5.2.7 Statistical Analyses 4183 Data analyses were carried out in STATISTICA, version 13.2 (Statsoft Inc., Tulsa, 4184 Oklahoma) and R version 3.3.0 (R Development Core Team, 2016). LLT and ULT assays, SCPs, 4185 HKDT and CCRT results did not meet the assumptions of ANOVA, thus were analysed using 4186 generalized linear models (GLZ) assuming a binomial (LLT and ULT) and Gaussian (SCPs, 4187 HKDT and CCRT) distribution and a logit link function in R statistical software. CTLs met the 4188 linear model assumptions of constant variance and normal errors, therefore, they were analysed 4189 using one-way ANOVA in STATISTICA. Tukey-Kramer’s post-hoc tests were used to separate 4190 statistically heterogeneous groups. 4191

4192 5.3 Results 4193

4194 5.3.1 Lower and Upper Lethal Temperature Assays 4195 Temperature and duration of exposure significantly influenced survival of C. sesamiae and C. 4196 flavipes adults at both low and high temperatures (P ˂ 0.0001) (Table 5.1; 5.2). Duration of 4197 exposure × temperature interaction effects were highly significant for C. flavipes ULT (P ˂ 4198 0.001) (Table 5.2) whilst the same interactions were not significant (P > 0.05) for both species’ 4199 (LLT) and C. sesamiae (ULT) (Table 5.1; 5.2). A comparison of 0.5, 1, 2 and 4 h durations for 4200 C. sesamiae and C. flavipes following exposure to lethal low temperatures ranged from -5 to 5°C 4201 and -15 to -1°C for 0 – 100% survival respectively (Figure 5.1A; B). Upper lethal temperatures 4202 for the same durations of exposure ranged from 35 to 42°C for C. sesamiae and 37 to 44°C for C. 4203 flavipes (Figure 5.1C; D). An increase in severity of low and high temperature exposure resulted 4204 in increased mortalities for both species (Figure 5.1A-D). Similarly, an increase in duration of 4205 exposure to lethal low and high temperatures generally resulted in survival decrease for both C. 4206 sesamiae and C. flavipes adults (Figure 5.1A-D). 4207 4208 4209 4210 4211

163

4212 Table 5.1: Summary statistical results of the effects of temperature, duration of exposure and 4213 their interactions on the survival of Cotesia sesamiae adults following lower and upper lethal 4214 temperature treatments. Analysis were done using generalized linear models (GLZ) assuming 4215 binomial distribution with a logit link function in R version 3.3.0. 4216 Parameter d.f χ2 P-Value Lower lethal temperature Duration 3 337.48 ˂0.0001 Temperature 7 1306.56 ˂0.0001 Temperature * Duration 21 26.47 0.1892

Upper lethal temperature Duration 3 317.98 ˂0.0001 Temperature 6 1211.61 ˂0.0001 Temperature * Duration 18 23.15 0.185

4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231

164

4232 Table 5.2: Summary statistical results of the effects of temperature, duration of exposure and 4233 their interactions on the survival of Cotesia flavipes adults following lower and upper lethal 4234 temperature treatments. Analysis were done using generalized linear models (GLZ) assuming 4235 binomial distribution with a logit link function in R version 3.3.0. 4236 Parameter d.f χ2 P-Value Lower lethal temperature Duration 3 358.91 ˂0.0001 Temperature 7 1061.1 ˂0.0001 Temperature * Duration 21 24.19 0.2838

Upper lethal temperature Duration 3 296.14 ˂0.0001 Temperature 5 977.14 ˂0.0001 Temperature * Duration 15 32.52 ˂0.001

4237 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247

165

A 100 0.5 hr 1 hr 80 2 hr 4 hr

Survival (%)

-15 -13 -11 -9 -7 -5 -3 -1 0 1 3 4 5 Temperature (oC) 4248 4249

B 100

0.5 hr 80 1 hr 2 hr 4 hr 60

Survival (%) Survival

-15 -13 -11 -9 -7 -5 -3 -1 0 1 3 4 5 Temperature (oC) 4250 4251

166

C 100

0.5 hr 60 1 hr 2 hr 40 4 hr

Survival (%) Survival

35 36 37 38 39 41 42 44 Temperature (oC) 4252 4253

D 100

60 0.5 hr 1 hr 2 hr 40 4 hr

Survival (%) Survival

35 37 38 39 40 42 44 Temperature (oC) 4254 4255 Figure 5.1: Mean (±95% confidence interval) survival of (A) C. sesamiae and (B) C. flavipes at 4256 different low temperatures, and (C) C. sesamiae and (D) C. flavipes at different high 4257 temperatures applied over four different durations. 4258

167

4259 5.3.2 Critical Thermal Limits (CTLs) 4260 Critical thermal minima and maxima varied significantly between C. sesamiae and C. 4261 flavipes adults (Table 5.3; Figure 5.2A; B). Cotesia flavipes recorded a significantly lower (P ˂ ° ° 4262 0.001) CTmin (1.34±0.59 C) than C. sesamiae (2.57±0.41 C). Similarly, C. flavipes also recorded ° 4263 a significantly higher (P ˂ 0.001) CTmax (44.63±0.63 C) compared to congeneric C. sesamiae 4264 (39.51±0.99°C). 4265 4266 Table 5.3: Summary statistical results from full factorial ANOVA showing effects of species on

4267 critical thermal limits (CTmax and CTmin). 4268 Trait Effect SS DF MS F P

Intercept 152.49 1 152.49 2006.79 <0.0001

CT min Species 15.25 1 15.25 200.72 <0.0001 Error 2.89 38 0.076

Intercept 70778.57 1 70778.57 122885.1 ˂ 0.001

CT max Species 262.14 1 262.14 455.1 ˂ 0.001 Error 21.89 38 0.58

4269 4270

168

3.0 a 2.8 A 2.6

o 2.4 2.2 b 2.0 1.8 1.6 1.4 1.2 1.0

Critical thermal minima ( minima thermal Critical 0.8 0.6 0.4 Cotesia sesamiae Cotesia flavipes Species 4271 4272

46 b B 45

o 44

42 a

Critical thermal maxima ( maxima thermal Critical 38

37 Cotesia sesamiae Cotesia flavipes Species 4273 4274

4275 Figure 5.2: Effects of species on (A) CTmin and (B) CTmax. Error bars represent 95% CLs (N = 4276 20). Means with the same letter are not significantly different from each other. 4277

169

4278 5.3.3 Supercooling Points (SCPs) 4279 There was no significant difference in the supercooling points (χ2 =0.0368, d.f =1, P = 4280 0.848) between C. sesamiae and C. flavipes (Figure 5.3). The mean supercooling points for C. 4281 sesamiae and C. flavipes adults were -20.26±1.07°C and -20.34±1.19°C respectively (Figure 5.3). 4282 SCPs were lower (Figure 5.3) than the LLTs (Figure 5.1A; B) for both parasitoid species 4283 indicating that mortality temperatures were above those of supercooling.

-17 a a -18

o -19

-20

-21

-22

Supercooling points ( -23

-24 Cotesia sesamiae Cotesia flavipes Species 4284

4285 Figure 5.3: The effect of species on supercooling points. Error bars represent 95% CLs (N = 16 4286 per group). Means with the same letter are not significantly different from each other 4287

4288 5.3.4 Chill Coma Recovery Time (CCRT) 4289 There was a significant difference in CCRT between C. sesamiae and C. flavipes adults 4290 (χ2 =23.41, d.f =1, P ˂ 0.0001) (Figure 5.4). The average CCRTs were 3.31±0.69 and 4.01±0.82 4291 minutes for C. flavipes and C. sesamiae adults respectively (Figure 5.4). Cotesia flavipes 4292 recovered significantly faster than C. sesamiae following exposure to chill-coma temperature

170

4293 (0°C; 1 h) (Figure 5.4), symbolising a higher tolerance to low temperature for C. flavipes relative 4294 to C. sesamiae.

5.5 a

5.0

4.5 b

4.0

3.5

3.0

2.5

Chill coma recovery time (minutes) 2.0 Cotesia sesamiae Cotesia flavipes Species 4295 4296 Figure 5.4: Effects of species on chill-coma recovery time. Error bars represent 95% CLs (N = 4297 30). Means with the same letter are not significantly different from each other. 4298

4299 5.3.5 Heat Knockdown Time (HKDT) 4300 Like in CCRT, there was a significant difference in HKDT between C. sesamiae and C. 4301 flavipes (χ2 =207.77, d.f =1, P ˂ 0.0001) (Figure 5.5). The mean HKDTs for C. sesamiae and C. 4302 flavipes adults were 19.91±1.06 and 25.67±1.37 minutes respectively (Figure 5.5). Cotesia 4303 flavipes showed a significantly longer knock-down time than C. sesamiae following exposure to 4304 knock-down temperature (45.0±0.3°C) (Figure 5.5), symbolising a significantly higher resistance 4305 to heat shock for the exotic C. flavipes relative to indigenous C. sesamiae. 4306

171

30 b

a 22

Heat knock-down time (minutes)

16 Cotesia sesamiae Cotesia flavipes Species 4307 4308 4309 Figure 5.5: Effects of species on heat knock-down time. Error bars represent 95% CLs (N = 30). 4310 Means with the same letter are not significantly different from each other. 4311

4312 5.4 Discussion 4313 Evolutionary history, basal thermal tolerance and phenotypic plasticity mediates insect 4314 population phenologies and dynamics including biogeographic patterns and fitness traits under 4315 climate variability associated with anthropogenic climate change (Helmuth et al., 2005; 4316 Ghalambor et al., 2007; Santos et al., 2012). In the current study, survival of C. sesamiae and C. 4317 flavipes was dependent on both severity and duration of temperature exposure as in other insect 4318 taxa (Chidawanyika & Terblanche, 2011a; Nguyen et al., 2014; Mutamiswa et al., 2017a). I 4319 report C. sesamiae LLTs and ULTs ranging -5 to 5°C and 35 to 42°C respectively (for 0.5-4h 4320 exposures), in keeping with related studies (see Mutamiswa et al., 2017a). This result affirms the 4321 significance of magnitude of temperature and duration of exposure on overall physiological 4322 fitness. At more extreme temperatures, duration of exposure was less important than more 4323 benign temperatures (see Figure 5.1A-D), perhaps due to the irreversible damage on cell and 4324 protein function caused by extreme acute temperatures (Chown & Nicholson, 2004; Harrison et 4325 al., 2012). For 0.5–4 h treatments, LLTs for C. sesamiae and C. flavipes ranged from -5 to 5°C 4326 and -15 to -1°C whilst ULTs ranged from 35 to 42°C and 37 to 44°C respectively, suggesting that

172

4327 C. flavipes is more cold and heat tolerant than C. sesamiae. Given these responses, an important 4328 question will be whether these two species may continue coexisting in agroecosystems 4329 efficiently regulating lepidopteran stemborer populations under the current and projected climate 4330 change. With African temperatures projected to increase (IPCC, 2007; Gemeda & Sima, 2015), 4331 at the same time associated with high frequency of heat waves, and cold snaps (IPCC, 2014), it is 4332 highly likely that C. flavipes will survive more prolonged low and high temperature extremes 4333 than its congener C. sesamiae. When C. partellus is the host, C. flavipes has been reported to 4334 have a higher host searching efficiency even at low stemborer host densities than C. sesamiae 4335 (Wiedenmann & Smith, 1993; Sallam et al., 1999). Given these survival advantages, C. flavipes 4336 may, in all likelihood, have a higher competitive resource utilization and greater fitness than C. 4337 sesamiae in both cooler and warmer environments. Moreover, ULT and LLT temperatures 4338 reported here are often experienced under natural setups (see Mutamiswa et al., 2017a), 4339 symbolizing the ecological significance of these findings. Furthermore, sudden heat waves and 4340 cold snaps are expected to rise with global change (IPCC, 2014) and this likely has a 4341 compounding effect on fitness traits and survival of the two congeneric parasitoid species. 4342 The present study also demonstrated how temperature can limit the functional activity of both 4343 parasitoid species. CTLs are regarded as ecologically relevant measures of determining the 4344 impact of thermal variability on insect functional activities, e.g. mating, host searching, 4345 migration, feeding and others (Terblanche et al., 2011; Overgaard et al., 2012; Andersen et al. 4346 2014). These activities represent key fitness traits for insects and in particular parasitoids, whose 4347 ecological functions depend on host searching ability (Mailafiya et al., 2009), an activity trait

4348 highly dependent on temperature. Cotesia flavipes had lower CTmin and higher CTmax compared 4349 to C. sesamiae (Figure 5.2), indicating a fitness advantage for the exotic C. flavipes relative to 4350 congeneric C. sesamiae at both thermal extremes. This result also translates into a broader 4351 thermal activity window for C. flavipes relative to its congener. As such, C. flavipes may likely 4352 optimise key life history activities better off than congeneric C. sesamiae under stressful and 4353 variable thermal regimes. Hence, under changing climates, biogeographic patterns and 4354 abundance of C. sesamiae will be more constrained than C. flavipes due to differential thermal 4355 activity windows. In other insect taxa, such declines in abundance and changes in biogeographic 4356 patterns have been attributed to changes in population dynamics due to the influence of 4357 temperature (see Terblanche et al., 2008; Piyaphongkul et al., 2012; Nooten et al., 2014). Hence,

173

4358 thermal tolerance superiority of C. flavipes reported here coupled with superior host searching 4359 ability (Mailafiya et al., 2009), may likely make it a better competitor thus likely dominant over 4360 C. sesamiae in sub-Saharan Africa, and where the two share the same niche. 4361 I also document, for the first time, SCPs for adult C. flavipes and C. sesamiae. The 4362 average SCPs for C. sesamiae and C. flavipes were -20.26±1.14 and -20.34±1.41 respectively. 4363 Overwintering insects can be regarded as freeze tolerant (surviving ice formation within tissues), 4364 freeze intolerant (not surviving ice formation within tissues) (Andreadis et al., 2005) and chill 4365 susceptible (killed by chilling) (Sinclair et al., 2015). Generally, insect species with depressed 4366 SCPs ranging between ~ -20°C to -30°C are regarded as freeze intolerant (Bale, 1996). While my 4367 findings show both C. sesamiae and C. flavipes SCPs ~ -20°C, the failure to recover following 4368 supercooling suggest both species may be chill susceptible. Moreover, chill susceptibility is 4369 further affirmed by LLT results (Figure 5.1A; B) which indicated mortality temperatures for both 4370 species were way above those of supercooling. Nevertheless, supercooling ability remains a 4371 significant survival strategy for freeze intolerant species, likely facilitated by rapid accumulation 4372 of cryoprotectants and extracellular agents such as sugars, polyols and sorbitol, to avoid 4373 membrane rupture (Jones et al., 2008; Sinclair et al., 2015). 4374 Chill coma recovery time is another important indicator of cold stress resistance in terrestrial 4375 arthropods (David et al., 2003; Findsen et al., 2014). In the present study, C. flavipes recovered 4376 significantly faster than C. sesamiae following exposure to chill coma temperature (0°C for 1 h), 4377 suggesting cold tolerance and survival advantage over its congener when faced with cold shock. 4378 Chill coma recovery has been reported to be mediated by rapid resumption of ion homeostasis 4379 (Findsen et al. 2013; MacMillan et al., 2016). Therefore, it is likely that the faster chill coma 4380 recovery for C. flavipes may be through enhanced ability for ion homeostasis relative to C. 4381 sesamiae suggesting differential time course of ion homeostasis even for closely related taxa. 4382 This is despite the two species having shown similar SCPs. My study therefore also demonstrates 4383 the complexity of rapid thermal responses among insects and that different metrics of thermal 4384 tolerance may suggest divergent conclusions. 4385 Cotesia flavipes took more time (25.67±1.37 minutes) to be knocked down at high 4386 temperature (45°C) than C. sesamiae (19.91±1.06), further confirming its enhanced fitness 4387 advantage when faced with heat shock compared to C. sesamiae. These differential responses 4388 may be a result of the differences in the ability to rapidly express transient genetic responses

174

4389 responsible for upregulation of chaperone proteins that enable survival and activity at high 4390 temperatures (Feder & Hofmann, 1999). In nature, the ability to rapidly shorten or avoid heat 4391 knockdown and recover from chill coma temperatures is of high ecological relevance as any 4392 limited locomotion due to temperature stress, attributed to cessation of neuromuscular activity 4393 due to disruption of ion homeostasis (Findsen et al., 2014; Pujol-Lereis et al., 2016), may lead to 4394 opportunistic predation by natural enemies. It is therefore highly likely that C. sesamiae may be 4395 more prone to such predation than C. flavipes under stressful thermal environments. 4396 Furthermore, prolonged inactivity following chill coma and heat knockdown may result in 4397 compromised realization of life history traits including reproductive fitness and ultimately 4398 delivery of the ecosystem services like host pest parasitisation. Thus, the current results also 4399 report a fitness and survival advantage of exotic C. flavipes relative to indigenous C. sesamiae 4400 following heat shock. 4401 Overall, the current results reveal C. flavipes has superior basal thermal tolerance than 4402 congeneric C. sesamiae at both thermal extremes. Even though C. sesamiae and C. flavipes can 4403 coexist with interactive synergistic effects that maximize biocontrol efficacy against lepidopteran 4404 stemborer species (Overholt et al., 2000; Mailafiya et al., 2011), I conclude that C. flavipes will 4405 become dominant under climate change scenarios due to its superior thermal physiology. 4406 However, I make this conclusion with caveats as I only determined basal responses and not 4407 phenotypic plasticity of the various matrices that I determined. Other studies have shown how 4408 several factors such as within- and transgenerational -thermal history (Kristensen et al., 2008; 4409 Nyamukondiwa & Terblanche, 2010; Chidawanyika et al., 2011b), age and feeding status 4410 (Nyamukondiwa & Terblanche, 2009; Chidawanyika et al., 2017) and ontogeny (Andersen et al., 4411 2014) can influence responses to thermal exposure with significant potential for mitigating 4412 adverse effects of climate change (reviewed in Stillman, 2003; Sgrò et al., 2016). Future studies 4413 should therefore endeavor to explore such plasticity since the knowledge can aid enhancement of 4414 quality of mass-reared parasitoid insects (Chidawanyika et al., 2012; Sørensen et al., 2012; 4415 Terblanche, 2014). For example, developmental and adult acclimation in moths resulted in 4416 improved field performance of mass- reared Cydia pomonella (measured as flight response to 4417 pheromone traps) under conditions similar to acclimation conditions (Chidawanyika & 4418 Terblanche, 2011a). This evidence is in addition to earlier work on the Queensland fruit fly 4419 where manipulation of thermal plasticity through preconditioning resulted in improved field

175

4420 performance (Meats, 1973; 1976). Pest managers can therefore utilize such preconditioning to 4421 improve the survival and performance of the parasitoids and augment natural populations, in 4422 particular for C. sesamiae which had poorer basal thermal tolerance. This will not only aid 4423 performance of C. sesamiae, but also ensure that introduced alien C. flavipes may not become 4424 over dominant to ensure maintenance of native biodiversity in agroecosystems. 4425

4426 References 4427 Alford, L., Burel, F. & van Baaren, J. (2016) Improving methods to measure critical thermal 4428 limits in phloem-feeding pest insects. Entomologia Experimentalis et Applicata, 159, 61-69. 4429 DOI: 10.1111/eea.12410. 4430 Amarasekare, P. & Savage, V. (2012) A framework for elucidating the temperature dependence 4431 of fitness. American Naturalist, 179, 178–191. 4432 Andersen, J.L., Manenti, T., Sørensen, J.G., MacMillan, H.A., Loeschcke, V. & Overgaard, J. 4433 (2014) How to assess Drosophila cold tolerance: chill coma temperature and lower lethal 4434 temperature are the best predictors of cold distribution limits. Functional Ecology, 29, 55–65. 4435 Andreadis, S.S., Milonas, P.G. & Savopoulou-Soultani, M. (2005) Cold hardiness of diapausing 4436 and non-diapausing pupae of the European grapevine moth, Lobesia botrana. Entomologia 4437 Experimentalis et Applicata, 117, 113-118. 4438 Assefa, Y., Mitchell, A., Conlong, D.E. & Muirhead, K.A. (2008) Establishment of Cotesia 4439 flavipes (Hymenoptera: Braconidae) in sugarcane fields of Ethiopia and origin of founding 4440 population. Journal of Economic Entomology, 101(3), 686-691. 4441 Bale, J.S. & Hayward, S.A.L. (2009) Insect overwintering in a changing climate. The Journal of 4442 Experimental Biology, 213, 980-994. 4443 Bale, J.S. (1996) Insect cold hardiness: a matter of life and death. European Journal of 4444 Entomology, 93, 369–382. 4445 Bale, J., Masters, G.J., Hodkins, I.D., Awmack, C., Bezemer, T.M., Brown, V.K., Buterfield, J., 4446 Buse, A., Coulson, J.C., Farrar, J., Good, J.E.G., Harrigton, R., Hartley, S., Jones, T.H., 4447 Lindroth, R.L., Press, M.C., Symrnioudis, I., Watt, A.D. & Whittaker, J.B. (2002) Herbivory 4448 in global climate change research: direct effects of rising temperature on insect herbivores. 4449 Global Change Biology, 8, 1-16.

176

4450 Bowler, K. & Terblanche, J.S. (2008) Insect thermal tolerance: what is the role of ontogeny, 4451 ageing and senescence? Biological Reviews, 83, 339–355. 4452 Chidawanyika, F., Nyamukondiwa, C., Strathie, L. & Fischer, K. (2017) Effects of thermal 4453 regimes, starvation and age on heat tolerance of the Parthenium Beetle Zygogramma 4454 bicolorata (Coleoptera: Chrysomelidae) following dynamic and static protocols. PLoS ONE 4455 12:e0169371. doi:10.1371/journal.pone.0169371. 4456 Chidawanyika, F. & Terblanche, J.S. (2011a) Rapid thermal responses and thermal tolerance in 4457 adult codling moth Cydia pomonella (Lepidoptera: Totricidae). Journal of Insect Physiology, 4458 57, 108- 117. 4459 Chidawanyika, F. & Terblanche, J.S. (2011b) Costs and benefits of thermal acclimation for 4460 codling moth, Cydia pomonella (Lepidoptera: Tortricidae): implications for pest control and 4461 the sterile insect release programme. Evolutionary Applications, 4, 534-544. 4462 Chown, S.L. & Nicolson, S.W. (2004) Insect Physiological Ecology: Mechanisms and Patterns. 4463 Oxford University Press, Oxford. 4464 Chown, S.L. & Terblanche, J.S. (2007) Physiological Diversity in Insects: Ecological and 4465 Evolutionary Contexts. Advances in Insect Physiology, 33, 50-152. 4466 David, J.R., Gibert, P., Moreteau, B., Gilchrist, G.W. & Huey, R.B. (2003) The fly that came in 4467 from the cold: recovery time from cold temperature in Drosophila subobscura. Functional 4468 Ecology, 17, 425–430. 4469 Dejen, A., Getu, E., Azerefegne, F. & Ayalew, A. (2013) Distribution and extent of Cotesia 4470 flavipes Cameron (Hymenoptera: Braconidae) parasitism in Northeastern Ethiopia. 4471 International Journal of Insect Science, 5, 9–19. 4472 de Sassi, C. & Tylianakis, J.M. (2012) Climate change disproportionately increases herbivore 4473 over plant or parasitoid biomass. PLoS ONE, 7(7), e40557. 4474 Divyal, K., Marulasiddesa, K.N., Karupanidhi, K. & Sankar, M. (2009) Population dynamics of 4475 stem borer, Chilo partellus (Swinhoe) and its interaction with natural enemies in sorghum. 4476 Indian Journal of Science and Technology, 3, 1–2. 4477 Feder, M.E. & Hofmann, G.E. (1999) Heat-shock proteins, molecular chaperones, and the stress 4478 response. Annual Review of Physiology, 61, 243–282.

177

4479 Findsen, A., Andersen, J.L., Calderon, S. & Overgaard, J. (2013) Rapid cold hardening improves 4480 recovery of ion homeostasis and chill coma recovery time in the migratory locust, Locusta 4481 migratoria. Journal of Experimental Biology, 216, 1630-1637. 4482 Findsen, A., Pedersen, T.H., Petersen, A.G., Nielsen, O.B. & Overgaard, J. (2014) Why do 4483 insects enter and recover from chill coma? Low temperature and high extracellular potassium 4484 compromise muscle function in Locusta migratoria. Journal of Experimental Biology, 217, 4485 1297-1306 doi:10.1242/jeb.098442. 4486 Gemeda, D.O. & Sima, A.D. (2015) The impacts of climate change on African continent and the 4487 way forward. Journal of Ecology and the Natural Environment, 7(10), 256-262. 4488 Ghalambor, C.K., McKay, J.K., Carroll, S.P. & Reznick, D.N. (2007) Adaptive versus non- 4489 adaptive phenotypic plasticity and the potential for contemporary adaptation in new 4490 environments. Functional Ecology, 21, 394–407. 4491 Grout, T.G. & Stoltz, K.C. (2007) Developmental rates at constant temperatures of three 4492 economically important Ceratitis species (Diptera: Tephritidae) from Southern Africa. 4493 Environmental Entomology, 36, 1310-1317. 4494 Hance, T., Van Baaren, J., Vernon, P. & Boivin, G. (2007) Impact of extreme temperatures on 4495 parasitoids in a climate change perspective. Annual Review of Entomology, 52, 107–126. 4496 Harrison, J.F., Woods, H.A. & Roberts, S.P. (2012) Ecological and Environmental Physiology of 4497 Insects. Oxford University Press, UK. 4498 Helmuth, B., Kingsolver, J.G. & Carrington, E. (2005) Biophysics, physiological ecology, and 4499 climate change: does mechanism matter? Annual Review of Physiology, 67, 177–201. 4500 Hoffmann, A.A., Chown, S.L. & Clusella-Trullas, S. (2012) Upper thermal limits in terrestrial 4501 ectotherms: how constrained are they? Functional Ecology, 27, 934–949. 4502 IPCC. (2007) In: Parry, M.L., Canziani, O.F., Palutikof, J.P., van der Linden, P.J. & Hanson C.E. 4503 (Eds), Climate change 2007: Impacts, adaptation and vulnerability, Contribution of Working 4504 Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. 4505 Cambridge University Press. Cambridge. UK. 4506 IPCC. (2014) Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and 4507 III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Core 4508 Writing Team, R.K. Pachauri and L.A. Meyer (eds.)]. IPCC, Geneva, Switzerland, pp 151.

178

4509 Jones, D.B., Giles, K.L. & Elliot, N.C. (2008) Supercooling points of Lysiphlebus testaceipes 4510 and its host Schizaphis graminum. Environmental Entomology, 37, 1063-1068. 4511 Karuppaiah, V. & Sujayanad, G.K. (2012) Impacts of climate change on population dynamics of 4512 insect pests. World Journal of Agricultural Sciences, 8, 240-246. 4513 Kfir, R., Overholt, W.A., Khan, Z.R. & Polaszek, A. (2002) Biology and management of 4514 economically important lepidopteran cereal stem borers in Africa. Annual Review of 4515 Entomology, 47, 701–713. 4516 Khani, A. & Moharramipour, S. (2010) Cold hardiness and supercooling capacity in the 4517 overwintering larvae of the codling moth, Cydia pomonella. Journal of Insect Science, 10, 83. 4518 Kimani-Njogu, S.W. & Overholt, W.A. (1997) Biosystematics of the Cotesia flavipes species 4519 complex (Hymenoptera: Braconidae), parasitoids of the gramineous stemborers. Insect 4520 Science and its Application, 17, 119–130. 4521 Kleynhans, E., Conlong, D. & Terblanche, J.S. (2014) Host plant-related variation in thermal 4522 tolerance if Eldana saccharina. Entomologia Experimentalis et Applicata, 150, 113-122. 4523 Koveos, D.S. (2001) Rapid cold hardening in the olive fruit fly Bactrocera oleae under 4524 laboratory conditions. Entomologia Experimentalis et Applicata, 101, 257-263. 4525 Loeschcke, V. & Hoffmann, A.A. (2007) Consequences of heat hardening on a field fitness 4526 component in Drosophila depend on environmental temperature. The American Naturalist, 4527 169, 175–183. 4528 Ma, F.Z., Lu¨, Z.C., Wang, R. & Wan, F.H. (2014) Heritability and evolutionary potential in 4529 thermal tolerance traits in the invasive Mediterranean cryptic species of Bemisia tabaci 4530 (Hemiptera: Aleyrodidae). PLoS ONE, 9, e103279. doi:10.1371/journal.pone.0103279. 4531 Mailafiya, D.M., Le Ru, B.P., Kairu, E.W., Calatayud, P-A. & Dupas, S. (2010) Geographic 4532 distribution, host range and perennation of Cotesia sesamiae and Cotesia flavipes Cameron in 4533 cultivated and natural habitats in Kenya. Biological Control, 54, 1–8. 4534 Mailafiya, D.M., Le Ru, B.P., Kairu, E.W., Calatayud ,PA. & Dupas, S. (2009) Species diversity 4535 of lepidopteran stem borer parasitoids in cultivated and natural habitats in Kenya. Journal of 4536 Applied Entomology, 133, 416-429. 4537 Mailafiya, D.M., Le Ru, B.P., Kairu, E.W., Dupas, S. & Calatayud, P.A. (2011) Parasitism of 4538 lepidopterous stemborers in cultivated and natural habitats. Journal of Insect Science, 11, 1- 4539 19.

179

4540 Manjoo S, Bajpai NK (2011) Cotesia flavipes Cameron parasitizing Chilo partellus (Swinhoe): 4541 Host-age dependent parasitism, cocoon formation and sex ratio. Journal of Biological 4542 Control, 25(4), 323–325. 4543 Meats, A. (1973) Rapid acclimatization to low temperature in the Queensland fruit fly, Dacus 4544 tryoni. Journal of Insect Physiology, 19, 1903-1911. 4545 Meats, A. (1976) Thresholds for cold torpor and cold survival in the Queensland fruit fly, and 4546 predictability of rates of change in survival threshold. Journal of Insect Physiology, 22, 1505- 4547 1509. 4548 Milton, C.C. & Partridge, L. (2008) Brief carbon dioxide exposure blocks heat hardening but not 4549 cold acclimation in Drosophila melanogaster. Journal of Insect Physiology, 54, 32-40. 4550 Muirhead, K.A., Murphy, N.P., Sallam. N., Donnellan, S.C. & Austin, A.D. (2012) 4551 Phylogenetics and genetic diversity of the Cotesia flavipes complex of parasitoid wasps 4552 (Hymenoptera: Braconidae), biological control agents of lepidopteran stemborers. Molecular 4553 Phylogenetics and Evolution, 63, 904–914. 4554 Mutamiswa, R., Chidawanyika, F. & Nyamukondiwa, C. (2017a) Comparative assessment of the 4555 thermal tolerance of spotted stemborer, Chilo partellus Swinhoe (Lepidoptera: Crambidae) 4556 and its larval parasitoid, Cotesia sesamiae Cameron (Hymenoptera: Braconidae). Insect 4557 Science, DOI 10.1111/1744-7917.12466. 4558 Mutamiswa, R., Chidawanyika, F. & Nyamukondiwa, C. (2017b) Dominance of spotted 4559 stemborer Chilo partellus Swinhoe (Lepidoptera: Crambidae) over indigenous stemborer 4560 species in Africa’s changing climates: ecological and thermal biology perspectives. 4561 Agricultural and Forest Entomology, DOI: 10.1111/afe.12217. 4562 Nguyen, C., Bahar, M.H., Baker, G. & Andrew, N.R. (2014) Thermal tolerance limits of 4563 diamondback moth in ramping and plunging assays. PLoS ONE, 9, e87535. 4564 doi:10.1371/journal.pone.0087535. 4565 Nooten, S.S., Andrew, N.R. & Hughes, L. (2014) Potential impacts of climate change on insect 4566 communities: A transplant experiment. PLoS ONE, 9(1), e85987. 4567 Nyamukondiwa, C., Kleynhans, E. & Terblanche, J.S. (2010) Phenotypic plasticity of thermal 4568 tolerance contributes to the invasion potential of Mediterranean fruit flies (Ceratitis capitata). 4569 Ecological Entomology, 35, 565-575.

180

4570 Nyamukondiwa, C. & Terblanche, J.S. (2009) Thermal tolerance in adult Mediterranean and 4571 Natal fruit flies (Ceratitis capitata and Ceratitis rosa): effects of age, gender and feeding 4572 status. Journal of Thermal Biology, 34, 406-414. 4573 Nyamukondiwa, C. & Terblanche, J.S. (2010) Within-generation variation of critical thermal 4574 limits in adult Mediterranean and Natal fruit flies Ceratitis capitata and Ceratitis rosa: 4575 thermal history affects short-term responses to temperature. Physiological Entomology, 35, 4576 255-264. 4577 Nyamukondiwa, C., Weldon, C.W., Chown, S.L., le Roux, P.C. & Terblanche, J.S. (2013) 4578 Thermal biology, population fluctuations and implications of temperature extremes for the 4579 management of two globally significant insect pests. Journal of Insect Physiology, 59,1199- 4580 1211. 4581 Ochieng, R.S., Onyango, F.O. & Bungu, M.D.O. (1985) Improvement of techniques for mass 4582 culture of Chilo partellus (Swinhoe). Insect Science and its Application, 6, 425-428. 4583 Overgaard, J., Kearney, M.R. & Hoffmann, A.A. (2014) Sensitivity to thermal extremes in 4584 Australian Drosophila implies similar impacts of climate change on the distribution of 4585 widespread and tropical species. Global Change Biology, 20, 1738–1750. 4586 Overgaard, J., Kristensen, T.N. & Sørensen, J.G. (2012) Validity of thermal ramping assays used 4587 to assess thermal tolerance in arthropods. PLoS ONE, 7(3), e32758. 4588 Overgaard, J. & Sorensen, J.G. (2008) Rapid thermal adaptation during field temperature 4589 variations in Drosophila melanogaster. Cryobiology, 56, 159–162. 4590 Overholt, W.A., Songa, J.M., Ofomata, V. & Jeske, R. (2000) The spread and ecological 4591 consequences of the invasion of Chilo partellus (Swinhoe) (Lepidoptera: Crambidae) in 4592 Africa. In E. E. Lyons and S. E. Miller Invasive species in eastern Africa: Proceedings of a 4593 Workshop held at ICIPE, July 5-6, 1999. ICIPE Science Press, Nairobi. 4594 Padmaja, P.G. & Prabhakar, M. (2004) Natural parasitization of spotted stem borer, Chilo 4595 partellus (Swinhoe) on sweet sorghum in Andhra Pradesh. Indian Journal of Entomology, 66, 4596 285–286. 4597 Parmesan, C. (2006) Ecological and evolutionary responses to recent climate change. Annual 4598 Review of Ecology, Evolution, and Systematics, 37, 637-669.

181

4599 Piyaphongkul, J., Pritchard, J. & Bale, J. (2013) Can tropical insects stand the heat? A case study 4600 with the Brown Planthopper Nilaparvata lugens (Sta°l). PLoS ONE, 7(1), 4601 e29409.doi:10.1371/journal. Pone.0029409. 4602 Pujol-Lereis, L.M., Fagali, N.S., Rabossi, A., Catala, A. & Quesada-Allue, L.A. (2016) Chill- 4603 coma recovery time, age and sex determine lipid profiles in Ceratitis capitata tissues. Journal 4604 of Physiology, 87, 53-62. 4605 R Development Core Team (2016) R: A Language and Environment for Statistical Computing. R 4606 Foundation for Statistical Computing, Vienna, Austria. 4607 Romo, C.M. & Tylianakis, J.M. (2013) Elevated temperatures and drought interact to reduce 4608 parasitoid effectiveness in suppressing hosts. PLoS ONE, 8(3), e58136. doi: 4609 10.1371/journal.pone.0058136. 4610 Sallam, N.M., Overholt, W.A. & Kairu, E. (1999) Comparative evaluation of Cotesia flavipes 4611 and C sesamiae (Hymenoptera: Braconidae) for the management of Chilo partellus 4612 (Lepidoptera: Pyralidae) in Kenya. Bulletin of Entomological Research, 89, 185-191. 4613 Santos, M., Castaneda, L.E. & Rezende, E.L. (2012) Keeping pace with climate change: what is 4614 wrong with the evolutionary potential of upper thermal limits? Ecology and Evolution, 2(11), 4615 2866–2880. 4616 Sgrò, C.M., Terblanche, J.S. & Hoffmann, A.A. (2016) What can plasticity contribute to insect 4617 responses to climate change? Annual Review of Entomology, 61, 433-451. DOI: 4618 10.1146/annurev-ento-010715-023859. 4619 Sinclair, B.J., Coello Alvarado, L.E. & Ferguson, L.V. (2015) An invitation to measure insect 4620 cold tolerance: Methods, approaches, and workflow. Journal of Thermal Biology, 53, 180- 4621 197. 4622 Songa, J.M., Overholt, W.A., Mueke, J.M. & Okell, R.O. (2001) Colonization of Cotesia 4623 flavipes (Hymenoptera: Braconidae) in stem borers in the semi-arid Eastern province of 4624 Kenya. Insect Science and its Application, 21, 289–295. 4625 Sørensen, J.G. Addison, M.F. & Terblanche, J.S. (2014) Mass-rearing of insects for pest 4626 management: Challenges, synergies and advances from evolutionary physiology. Crop 4627 Protection, 38, 87-94. 4628 Stillman, J.H. (2003) Acclimation capacity underlies susceptibility to climate change. Science, 4629 301, 65. DOI: 10.1126/science.1083073.

182

4630 Stotter, R.L. & Terblanche, J.S. (2009) Low temperature tolerance of false codling moth 4631 Thaumotibia leucotreta (Meyrick) (Lepidoptera: Tortricidae) in South Africa. Journal of 4632 Thermal Biology, 34, 320–325. 4633 Tefera, T., Mugo, S., Tende, R. & Likhayo, P. (2010) Mass Rearing of Stem borers, Maize 4634 weevil and Larger grain borer Insect Pests of Maize. CIMMYT. Nairobi. 4635 Terblanche, J.S. (2014) Physiological performance of field-released insects. Current Opinion in 4636 Insect Science, 4, 60-66. 4637 Terblanche, J.S., Clusella-Trullas, S., Deere, J.A. & Chown, S.L. (2008) Thermal tolerance in a 4638 south-east African population of tsetse fly Glossina pallidipes (Diptera: Glossinidae): 4639 implications for forecasting climate change impacts. Journal of Insect Physiology, 54, 114– 4640 127. 4641 Terblanche, J.S., Hoffmann, A.A., Mitchell, K.A., Rako, L., le Roux, PC. & Chown, S.L. (2011) 4642 Ecologically relevant measures of tolerance to potentially lethal temperatures. Journal of 4643 Experimental Biology, 214, 3713-3725. 4644 Terblanche, J.S., Marais, E. & Chown, S.L. (2007) Stage-related variation in rapid cold 4645 hardening as a test of the environmental predictability hypothesis. Journal of Insect 4646 Physiology, 53, 455–462. 4647 Thomas, C.D., Cameron, A, Green, R.E., Bakkenes, M., Beaumont, L.J., Collingham, Y.C., 4648 Erasmus, B.F.N., Ferreira de Siqueira, M., Grainger, A., Hannah, L., Hughes, L., Huntley, B., 4649 van Jaarsveld, A.S., Midgley, G.F., Miles, L., Ortega-Huerta, M.A., Peterson, A.T., Phillips, 4650 O.L. &Williams, S.E. (2004) Extinction risk from climate change. Nature, 427, 145-148. 4651 Thomas, M.B. & Blanford, S. (2003) Thermal biology in insect-parasite interactions. Trends in 4652 Ecology and Evolution, 18, 344-350. 4653 Voigt, W., Perner, J., Davis, A.J., Eggers, T., Schumacher, J., Bährmann, R., Fabian, B., 4654 Heinrich, W., Köhler, G., Lichter, D., Marstaller, R. & Sander, F.W. (2003) Trophic levels are 4655 differentially sensitive to climate. Ecology, 84, 2444-2453. 4656 Ward, N.L. & Masters, G.J. (2007) Linking climate change and species invasion: an illustration 4657 using insect herbivores. Global Change Biology, 13, 1605–1615. 4658 Walther, G.R., Post, E., Convey, P., Menzel, A., Parmersan, C., Beebee, T.J.C., Fromentin, J.M., 4659 Gulberg, O.V. & Bairlein, F. (2002) Ecological responses to recent climate change. Nature, 4660 416, 389-395.

183

4661 Weldon, CW., Terblanche, J.S. & Chown, S.L. (2011) Time-course for attainment and reversal 4662 of acclimation to constant temperature in two Ceratitis species. Journal of Thermal Biology, 4663 36, 479-485. 4664 Wiedenmann, R.N. & Smith, J.W.Jr. (1993) Functional response of the parasite Cotesia flavipes 4665 (Hymenoptera: Braconidae) at low densities of the host Diatraea saccharalis (Lepidoptera: 4666 Pyralidae). Environmental Entomology, 22, 849–858. 4667 Zhou, G., Baumgartner, J. & Overholt, W.A. (2001) Impact of an exotic parasitoid on stemborer 4668 (Lepidoptera) populations dynamics in Kenya. Ecological Applications, 11, 1554–1562. 4669 Zhou, G, Overholt, W.A. & Kimani-Njogu, S.W. (2003) Species richness and parasitism in 4670 assemblage of parasitoids attacking maize stem borer in coastal Kenya. Ecological 4671 Entomology, 28, 109–118. 4672

184

4673 4674 4675

4676 CHAPTER 6

4677

4678

4679 Superior basal and plastic thermal responses to environmental heterogeneity in invasive 4680 exotic stemborer Chilo partellus Swinhoe over indigenous Busseola fusca (Fuller) and 4681 Sesamia calamistis Hampson.

4682

4683

4684 4685 4686 4687 4688 4689 4690 4691 4692 4693

4694 *Published as “Mutamiswa, R., Chidawanyika, F. & Nyamukondiwa, C. (2018) Superior basal 4695 and plastic thermal responses to environmental heterogeneity in invasive exotic stemborer Chilo 4696 partellus Swinhoe over indigenous Busseola fusca (Fuller) and Sesamia calamistis Hampson.” 4697 Physiological Entomology. DOI: 10.1111/phen.12235

185

4698 6.1 Introduction 4699 Global increase in temperature coupled with erratic and limited rainfall exerts potent 4700 environmental stressors such as heat, dehydration and starvation that determine insect population 4701 dynamics (Bale et al., 2002; Clusella-Trullas et al., 2011; Bauerfeind & Fischer, 2013; 4702 Hoffmann et al., 2013). While climate change brings multiple overlapping stressors, temperature 4703 tolerance remains a critical attribute for physiological adaptation for invasive species and 4704 organisms living in different bioclimatic regions (Nyamukondiwa et al., 2010). Temperature 4705 stress directly influences diurnal and seasonal activity and the outcome of several life-history 4706 traits including, reproduction resource acquisition, dispersal and ultimately survival (Hoffman et 4707 al., 2012; Ma et al., 2014). Under natural environments, insect responses to temperature stress 4708 may be influenced by transient events including repeated exposure to sublethal temperatures that 4709 can induce plasticity thereby improving performance and survival in otherwise lethal 4710 environments (Marshall & Sinclair, 2012; Karl et al., 2014; Kingsolver et al., 2015). Indeed, 4711 thermal suitability of novel environments for invasive insect species is dependent on both basal 4712 and phenotypic plasticity to tolerance traits. Failure to acquire such enhanced basal and plasticity 4713 may result in even more deleterious consequences under lethal environments (Chidawanyika & 4714 Terblanche, 2011). 4715 The mechanisms inducing such thermal plasticity are temperature-time dependent, hence 4716 duration and magnitude are key determinants of insect-temperature interactions (Chown & 4717 Nicolson, 2004). For example, acute exposure to sub-lethal temperatures may induce hardening 4718 which involves physiological adjustments such as expression of heat shock proteins in order to 4719 preserve key life history traits (Sinclair & Roberts, 2005). On the other hand, acclimation is 4720 employed under relatively chronic exposure ranging from diurnal to seasonal scale (Overgaard & 4721 Sørensen, 2008; Khani & Moharramipour, 2010). In nature, however, thermal stress is 4722 experienced in concert with other environmental stressors making it difficult to predict outcomes 4723 of thermal events. For example, heat stress often coincides with long periods of poor 4724 precipitation (drought) as predicted under climate change scenarios (IPCC, 2014). For insect 4725 herbivores, such drought events directly influence both quality and availability of host plants 4726 (Bale et al., 2002; Buerfiend & Fischer, 2013) resulting in additional stressors that mediate 4727 thermal responses such as starvation (Chidawanyika et al., 2017). Furthermore, periods of water 4728 deficit are known to variably exert desiccation stress among different insect life stages with

186

4729 subsequent impact on thermal sensitivity and ultimately survival (Bubliy et al., 2012). Resistance 4730 to most of these stressors is energetically costly for most ectotherms (Fusi et al., 2016; Verbek et 4731 al., 2016), hence the quality and availability of host plant material mediates the outcome of 4732 environmental stress. For example, in order to reduce desiccation stress, insects convert stored 4733 fat to metabolic water, accumulate water (e.g. from water-based food, nectar) in tissues (Hadley, 4734 1994), close their spiracles and increase reabsorption of water in the rectum (Harrison et al., 4735 2012). Hence prolonged periods of desiccation stress without replenishment of body fat reserves 4736 due to resource unavailability can have dire fitness consequences. To cope with starvation, 4737 insects have been reported to improve survival ability through suppression of metabolic rate, 4738 seeking cooler habitats and also exploitation of body reserves such as lipids (Rion & Kawecki, 4739 2007; Scharf et al., 2016). Hence, resistance to multiple stressors among insects is interlinked 4740 and several studies have confirmed the presence of cross-resistance among different stressors 4741 (e.g. MacMillan et al., 2009, Chidawanyika & Terblanche, 2011; Sinclair et al., 2007, 2013; 4742 Renault et al., 2015; Kalra et al., 2017). For example, starvation reduced cold tolerance in 4743 Tribolium castaneum (Coleoptera: Tenebrionidae) (Scharf et al., 2016) and heat tolerance in 4744 Zygogramma bicolorata (Coleoptera: Chrysomeliade) (Chidawanyika et al., 2017). In addition, 4745 desiccation reduced critical thermal maxima in Drosophila melanogaster exposed to a range of 4746 sub-lethal temperatures (Da Lage et al., 1989). Furthermore, in spring-tails (Folsomia candida), 4747 moderate desiccation improved cold tolerance (Bayley et al., 2001). This interplay among 4748 different stressors suggests shared physiological mechanisms (cross tolerance) or signal 4749 pathways (cross talk) (Sinclair et al., 2013) and likely coevolution of resistance mechanisms to 4750 related environmental stress traits. 4751 It is also widely accepted that ontogeny (developmental effects) influences sensitivity to 4752 environmental stress (Spicer & Gaston, 1999; Chown & Nicolson, 2004; Bowler & Terblanche, 4753 2008). Generally, less mobile life-stages (usually juveniles) are thought to be more tolerant and 4754 exhibiting greater phenotypic plasticity to environmental stress as compensation to their limited 4755 capacity to maintain optimal conditions through behavioural mechanisms. Several studies have 4756 supported this idea in high (Gilchrist et al., 1997; Klok & Chown, 2001) and low (Vernon & 4757 Vanier, 1996, Klok & Chown, 2001, Jensen et al., 2007) temperature stress and desiccation 4758 resistance (Woods & Singer, 2001). Given the above background, it is apparent that 4759 environmental stress resistance among arthropods is mediated by both biotic and abiotic factors.

187

4760 For both invasive and indigenous insects, including crop insect pests of economic importance, 4761 such basal and phenotypic plasticity (including cross tolerance and cross talk) mediates climate 4762 change responses and ultimately invasion potential (Chown et al., 2007; Renault et al., 2015). 4763 Indeed, modeling studies have revealed that, even in scenarios where there is repeated and high 4764 propagule pressure, the thermal suitability and adaptation to novel environment remains critical 4765 for invasive species establishment (Hayes & Barry, 2008; Mainali et al., 2015). It is therefore of 4766 importance to elucidate physiological responses of invasive agricultural pests facing multiple 4767 stressors as it gives insights into potential population dynamics under climate change. 4768 Here I investigated the effects of multiple environmental stresses (cold and heat 4769 hardening, thermal, desiccation and starvation acclimation) on the thermal tolerance of two 4770 indigenous stemborer species Busseola fusca (Lepidoptera: Noctuidae) (Fuller) and Sesamia 4771 calamistis Hampson (Lepidoptera: Noctuidae) and the exotic Chilo partellus Swinhoe 4772 (Lepidoptera: Crambidae). The African maize stemborer, B. fusca and pink stemborer, S. 4773 calamistis are regarded as the most destructive maize and sorghum insect pests in many 4774 countries of sub-Saharan Africa (Ong’amo et al., 2008; Khadioli et al., 2014a). The distribution 4775 of B. fusca varies with region with the pest common at higher altitudes (above 600 m a.s.l) in 4776 east and southern Africa, ≥ 2000 m.a.s.l in central Africa and dry lowland savanna areas in west 4777 Africa (Ajayi, 1998; Kfir et al., 2002). In east Africa, S. calamistis is available at all elevations, 4778 though in low populations (Ntiri et al., 2016a). The spotted stemborer, C. partellus is native to 4779 Asia and is the predominant insect pest of cereals in east and southern Africa (Kfir, 1997; 4780 Overholt et al., 2000). It was previously reported limited to low and mid-elevation areas (Zhou 4781 et al., 2001) and has latter expanded to high altitude and humid transitional areas (Khadioli et al., 4782 2014b) displacing indigenous stemborer species (Kfir, 1997; Mutamiswa et al., 2017a). 4783 Nevertheless, it remains unknown how invasion potential of invasive species is likely shaped in 4784 the face of multiple stresses.

4785 Critical thermal limits (CTLs, critical thermal minima [CTmin] and maxima [CTmax]) have 4786 received increasing attention in the study of thermal tolerance of insects because they provide an 4787 understanding on species distribution and their response to global change (Chown & Terblanche, 4788 2007; Deutsch et al., 2008). They are ecologically relevant (Alford et al., 2016), and represent 4789 essential functional thermal limits of organisms (e.g. Nyamukondiwa & Terblanche, 2010). 4790 Thus, their determination is the first step in understanding the activity range for a given

188

4791 population (reviewed in Somero, 2005). However, previous studies have indicated that CTLs 4792 depend on methodological context (Terblanche et al., 2007a), ontogeny (Cavieres et al., 2016) 4793 and may vary with trait and across species (Kelty & Lee, 1999). 4794 Despite the importance of invasive lepidopteran stemborers to African agriculture, no 4795 studies have investigated their physiological responses to multiple stressors. Related studies have 4796 specifically focused on non-crop pest insect model organisms (e.g. Sinclair et al., 2007; 4797 Nyamukondiwa et al., 2011; Bubily et al., 2013; Chidawanyika et al., 2017; Scharf et al., 2017). 4798 In line with this background, I determined the CTLs to activity for B. fusca, S. calamistis and C. 4799 partellus following rearing at benign temperatures, post thermal, starvation and desiccation 4800 acclimation as well as cold and heat hardening. These plastic responses to heterogeneous 4801 stressors may help explain invasion potential for invasive species upon facing novel 4802 environments. 4803

4804 6.2 Materials and Methods 4805

4806 6.2.1 Insect Culture and Rearing Conditions 4807 The initial colonies of B. fusca, S. calamistis and C. partellus were obtained as pupae 4808 from International Centre for Insect Physiology and Ecology (ICIPE), Kenya. The insects have 4809 been in culture for more than 20 generations with regular additions of wild populations to 4810 maintain heterozygosity. Pupae were maintained in open Petri dishes in Bugdorm rearing cages 4811 (240mm3; Bugdorm-BD43030F, Megaview Science Co., Ltd, Taiwan) under optimum 4812 conditions of 28 ± 1°C, 65 ± 10% relative humidity (RH), 12L: 12D photoperiod for S. calamistis 4813 and C. partellus and 25 ± 1°C, 75 ± 10% RH, 12L: 12D photoperiod for B. fusca in climate 4814 chambers (HPP 260, Memmert GmbH + Co.KG, Germany) until adult eclosion. Following 4815 emergence, adults had access to water provided as moistened cotton wads. Wax paper was also 4816 placed in the cages as an oviposition surface for gravid females. To maintain uniformity among 4817 test insects, eggs were harvested after every 12 h, and transferred to artificial diet (Ochieng et al., 4818 1985; Tefera et al., 2010) where they were allowed to hatch into larvae that were later used in 4819 subsequent experiments. In all cases 6th instar larvae (last instar developmental stage) were used 4820 in assays for all species.

189

4821 6.2.2 Thermal Tolerance Assays 4822 Critical thermal limits were assayed using established dynamic protocols 4823 (Nyamukondiwa & Terblanche, 2009; Mutamiswa et al., 2017b). For each species, ten larvae 4824 were individually placed in an insulated double jacketed chamber (‘organ pipes’) that was 4825 connected to a programmable water bath (Lauda Eco Gold, Lauda DR.R. Wobser GMBH and 4826 Co. KG, Germany) filled with 1:1 water: propylene glycol to accommodate subzero temperatures 4827 (Chown & Nicholson, 2004). Critical thermal minima and maxima experiments began at set 4828 point rearing temperature of 25°C for B. fusca and 28°C for C. partellus and S. calamistis where 4829 10 mins was given to allow equilibration of insect body and ambient temperatures. Thereafter, ° −1 4830 temperature was ramped up (CTmax) or down (CTmin) at a rate of 0.25 C min until all the insects 4831 reach their CTLs. The CTLs were defined as the temperature at which each individual insect lost 4832 coordinated muscle function which was regarded as a lack of response to mild stimuli such as 4833 prodding (e.g. Nyamukondiwa & Terblanche, 2010). While this ramping rate is faster than 4834 natural diurnal temperature fluctuations, it represents a compromise between experimental 4835 throughput and ecological relevance relative to CTLs work undertaken to date (see Chown et al., 4836 2009). A thermocouple (type K 36 SWG) connected to a digital thermometer (53/54IIB, Fluke 4837 Cooperation, USA) was inserted into the control chamber to verify chamber temperature. In all 4838 cases, insect body temperature was assumed to be in equilibrium with chamber temperature 4839 (Terblanche et al., 2007b; Mudavanhu et al., 2014; Mutamiswa et al., 2017b).These experiments 4840 were repeated twice to yield sample sizes of n = 20 for each species. 4841

4842 6.2.3 Effects of Thermal Acclimation on CTLs 4843 To induce potential plasticity, B. fusca larvae (40 per each acclimation temperature) were 4844 acclimated in artificial diet in Memmert climate chambers at 20 (low), 25 (optimum) and 30°C 4845 (high) for duration of 3 days as outlined by Nyamukondiwa & Terblanche, 2010. Similar aged S. 4846 calamistis and C. partellus were acclimated at 23 (low), 28 (optimum) and 33°C (high) for the 4847 same duration as that of B. fusca. These acclimation temperatures (5°C above and below 4848 optimum) and duration (3 days) are suffice to elicit plastic responses to temperature (Fischer et 4849 al., 2010; Nyamukondiwa & Terblanche, 2010; Weldon et al., 2011). For each species, larvae 4850 that remained under their rearing temperature during the acclimation period were treated as the

190

4851 control groups (e.g. Terblanche et al., 2006). Following acclimation, CTLs were measured for 4852 each species as outlined in thermal tolerance assays. 4853

4854 6.2.4 Effects of Starvation Acclimation on CTLs 4855 To assess the impact of starvation acclimation on CTLs, a total of 40 larvae from each 4856 species were loaded in 8, 30ml plastic vials with perforated screw-cap lids. Larvae were then 4857 starved for 4 days in climate chambers under respective rearing conditions, but with access to 4858 water ad libtum. Control larvae (40 from each species) had access to artificial diet for the same 4859 duration at abovementioned conditions. After 4 days, CTLs for both control and starved larvae 4860 were measured for each species as previously outlined. 4861

4862 6.2.5 Effects of Hardening on CTLs 4863 To test the impact of hardening (RCH and RHH) on CTLs, I first determined the 4864 hardening temperatures for each species. For all species, cold and heat hardening temperatures

4865 were derived from CTmin and CTmax for each stemborer species (Hoffmann et al., 2003; Lee &

4866 Denlinger, 2010) and these were defined as 6°C below CTmin and 7°C below CTmax respectively 4867 (Table 6.1). A total of 40 larvae from each species were cold and heat hardened for 2 hours. 4868 Control larvae (40 from each species) were maintained in climate chambers at respective 4869 benign/rearing conditions before measuring their CTLs. Since cold hardening temperatures 4870 induce chill coma in test stemborer species, insects were allowed to recover from chill coma for 4871 30 minutes following treatment at respective benign temperature before measuring CTLs. 4872 However, for heat hardened insects, CTLs were measured 2 hours post hardening as previously 4873 outlined. 4874 4875

191

4876 Table 6.1: Summary table showing cold and heat hardening temperatures (2 h duration) for B. 4877 fusca, S. calamistis and C. partellus larvae. Species Low temperature High temperature hardening (°C) hardening (°C) Busseola fusca -3 40 Sesamia calamistis -2 41 Chilo partellus 2 41 4878 4879 4880 6.2.6 Effects of Desiccation Acclimation on CTLs 4881 Desiccation assays were undertaken using standard established protocols in the laboratory 4882 modified from Bauerfeind et al. (2014). For all species, five last instar larvae were placed in 8 4883 (N=40) perforated 30ml empty plastic vials with aerated screw-cap lids. The vials were placed 4884 on top of wire gauze in a glass sealed desiccator containing 200g silica gel for 24 hours. 4885 Preliminary assays had shown this duration eliciting at least 40% body water loss for all species. 4886 The desiccator was placed in a climate chamber under rearing conditions for each respective 4887 stemborer species. Thermocron iButtons (Dallas Semiconductors, Model DS1920) were used to 4888 measure temperature and RH inside the desiccator at 30-min intervals. The average temperature 4889 and RH in the desiccators were 25.4 ± 0.38°C and 20.04 ± 2.03% for B. fusca, 28 ± 0.44°C and 4890 20.96 ± 1.14 % for S. calamistis and 28 ± 0.53°C and 24.11 ± 1.12% for C. partellus larvae 4891 respectively. After 24 hours, the insects were transferred to artificial diet for a further 24 hours 4892 under their respective rearing conditions before measuring CTLs. Controls (40 larvae from each 4893 species) were maintained in climate chambers at abovementioned rearing conditions until their 4894 CTLs were measured as previously outlined. 4895

4896 6.2.7 Statistical Analyses 4897 All analyses were carried out in STATISTICA, version 13.2 (Statsoft Inc., Tulsa, 4898 Oklahoma). The Shapiro-Wilk and Hartley-Bartlett tests were used to check data for normality 4899 and equality of variances respectively. All data met the linear model assumptions of constant 4900 variance and normal errors. Therefore, thermal tolerance assays and thermal acclimation effects 4901 on CTLs were analysed using one-way and two-way ANOVA respectively. In addition,

192

4902 hardening, starvation and desiccation acclimation effects were analysed using full factorial 4903 analysis of variance (ANOVA). To separate statistically heterogeneous groups, Tukey-Kramer’s 4904 post-hoc tests were used. Overlap in 95% confidence limits was used to identify statistically 4905 homogeneous groups. 4906

4907 6.3 Results 4908

4909 6.3.1 Thermal Tolerance Assays 4910 Critical thermal limits significantly varied across species of B. fusca, S. calamistis and C. 4911 partellus larvae (Table 6.2; Figure 6.1A; B). Critical thermal minima were different for all 4912 species, with B. fusca having the most enhanced cold tolerance and C. partellus the worst (Figure

4913 6.1A). Busseola fusca had the lowest CTmax compared to S. calamistis and C. partellus (Figure 4914 6.1B). Nevertheless, there was no significant difference between S. calamistis and C. partellus

4915 CTmax. 4916 4917 Table 6.2: Summary statistical results from one-way ANOVA showing effects of species (B.

4918 fusca, S. calamistis and C. partellus) on CTmin and CTmax 4919 Trait Effect SS DF MS F P

CTmin Intercept 1505 1 1505 17799.61 ˂ 0.001 Species 296.63 2 148.31 1754.09 ˂ 0.001 Error 4.82 57 0.085

CTmax Intercept 137004.4 1 137004.4 348106.6 ˂ 0.0001 Species 30.3 2 15.2 38.5 ˂ 0.0001 Error 22.4 57 0.4 4920

193

c 8 A

o 7

Critical thermal minima ( minima thermal Critical a 3

Busseola fusca Sesamia calamistis Chilo partellus Species 4921 4922

49 b B

a 47

Critical thermal maxima ( maxima thermal Critical

46 Busseola fusca Sesamia calamistis Chilo partellus Species 4923 4924

4925 Figure 6.1: Effects of species on (A) CTmin and (B) CTmax. Error bars represent 95% CLs. Means 4926 with the same letter are not significantly different from each other.

194

4927 6.3.2 Effects of Thermal Acclimation on CTLs

4928 Acclimation and species had highly significant effects on CTmin. Furthermore,

4929 acclimation * species interaction effects were significant for low temperature tolerance (CTmin)

4930 (Table 6.3; Figure 6.2A). Acclimation to low temperature enhanced CTmin for both B. fusca and 4931 C. partellus larvae while compromising it for S. calamistis larvae (Figure 6.2A). In addition,

4932 acclimation to high temperature improved CTmin for B. fusca and S. calamistis and reduced it for

4933 C. partellus larvae (Figure 6.2A). Overall, CTmin values were enhanced in B. fusca, and worst in

4934 C. partellus. Acclimation to low temperature marginally increased CTmax for C. partellus and 4935 compromised it for B. fusca and S. calamistis larvae (Figure 6.2B). Similarly, acclimation to high

4936 temperature compromised CTmax for both B. fusca and S. calamistis and enhanced it for C.

4937 partellus larvae (Figure 6.2B). Generally, C. partellus had an enhanced CTmax at all acclimation 4938 levels, relative to S. calamistis and B. fusca. 4939

4940 Table 6.3: Summary results from a two-way ANOVA showing the effects of species (B. fusca,

4941 S. calamistis and C. partellus) and thermal acclimation (low, optimum and high) on CTmin and

4942 CTmax. SS = sums of squares, d.f.=degrees of freedom. 4943 Trait Effect SS DF MS F P

CTmin Intercept 3917.2 1 3917.2 42707.21 ˂ 0.001 Acclimation 18.31 2 9.16 99.81 ˂ 0.001 Species 933.47 2 466.73 5088.55 ˂ 0.001 Species*Acclimation 70.75 4 17.69 192.83 ˂ 0.001 Error 15.68 171 0.092

CTmax Intercept 411395.5 1 411395.5 1330911 ˂ 0.0001 Acclimation 1.6 2 0.8 3 0.0743 Species 112.3 2 56.2 182 ˂ 0.0001 Species*Acclimation 4 4 1 3 0.0145 Error 52.9 171 0.3 4944 4945

195

10 c Busseola fusca A Sesamia calamistis b Chilo partellus

o 8

a a 4 b b a c 2

Critical thermal minima ( minima thermal Critical

0 Low Optimum High Acclimation temperature (oC) 4946 4947

50 Busseola fusca B Sesamia calamistis b Chilo partellus

o 49 a a

a a a 48

a a a 47

Critical thermal maxima ( maxima thermal Critical

46 Low Optimum High Acclimation temperature (oC) 4948 4949 4950 Figure 6.2: Thermal acclimation effects for three major stemborers, (B. fusca, S. calamistis and

4951 C. partellus) on (A) CTmin and (B) CTmax following acclimation for 3 days at low optimum and 4952 high temperature. Vertical bars denote ±95% CL. Means with the same letter are not 4953 significantly different from each other. Post hoc tests were done separately for each species

196

4954 6.3.3 Effects of Starvation Acclimation on CTLs

4955 Starvation acclimation significantly improved CTmin across all species (Table 6.4; Figure

4956 6.3A), while only improving CTmax for C. partellus (Figure 6.3B). However, this acclimation

4957 compromised CTmax for S. calamistis and had no significant effect relative to the control in B.

4958 fusca (Figure 6.3B). Busseola fusca recorded the lowest CTmin while C. partellus recorded the

4959 highest CTmax (Figure 6.3A; B). The magnitude of CTLs plasticity (difference between control 4960 and treatment survival) following starvation acclimation was generally highest in C. partellus 4961 (Figure 6.3A; B). In addition, there was a significant species * acclimation interaction effect for

4962 CTmin and non-significant effect for CTmax following starvation acclimation (Table 6.4). 4963 4964 Table 6.4: Summary statistical results from full factorial ANOVA showing effects of starvation

4965 acclimation on CTmin and CTmax for B. fusca, S. calamistis and C. partellus. SS = sums of 4966 squares, d.f.= degrees of freedom. 4967 Trait Effect SS DF MS F P

CTmin Intercept 2020.48 1 2020.48 18238.57 ˂ 0.001 Species 313.4 2 156.7 1414.51 ˂ 0.001 Starvation Acclimation 98.28 1 98.28 887.19 ˂ 0.001 Species*Acclimation 53.61 2 26.8 241.94 ˂ 0.001 Error 12.63 114 0.111

CTmax Intercept 274620.7 1 274620.7 785240.8 ˂ 0.0001 Species 73.9 2 36.9 105.6 ˂ 0.0001 Starvation Acclimation 0.3 1 0.3 1 0.325 Species*Acclimation 3.6 2 1.8 5.1 0.00739 Error 39.9 114 0.3 4968 4969

197

9 Control a A 8 Starvation Acclimation

o 7

5 b c

4 d

e 3

f Critical thermal minima ( minima thermal Critical 2

1 Busseola fusca Chilo partellus Sesamia calamistis Species 4970 4971

49.5 a Control B Starvation Acclimation

C) 49.0

o b

48.5 bc

c 48.0

47.5 d d 47.0

Critical thermal maxima ( maxima thermal Critical

46.5 Busseola fusca Chilo partellus Sesamia calamistis Species 4972

4973 Figure 6.3: Effects of starvation acclimation on (A) CTmin and (B) CTmax for Busseola fusca, 4974 Sesamia calamistis and Chilo partellus larvae. Error bars represent 95% CLs. Means with the 4975 same letter are not significantly different from each other. Post hoc tests were done separately for 4976 both treatments and across species

198

4977 6.3.4 Effects of Hardening on CTLs 4978 RCH significantly affected critical thermal minima and maxima across all species tested 4979 (Table 6.5A; Figure 6.4A; B). Short term low temperature pretreatment (RCH) significantly

4980 enhanced cold tolerance measured as CTmin, for B. fusca and C. partellus while reducing it for S.

4981 calamistis (Figure 6.4A). The magnitude of plastic responses for CTmin was highest in C.

4982 partellus relative to B. fusca and S. calamistis. Similarly, RCH decreased CTmax for all species, 4983 marginally so, but significant for B. fusca and non significant for S. calamistis and C. partellus 4984 (Figure 6.4B). Furthermore, there was a significant species * acclimation interaction effect for

4985 CTmin and a non-significant interaction effect for CTmax (Table 6.5A). On the other hand, RHH

4986 significantly affected CTmin and not CTmax (Table 6.5B). RHH compromised CTmin for B. fusca, 4987 while enhancing it for C. partellus (Figure 6.4C). Nevertheless, RHH did not have an effect on S.

4988 calamistis CTmin (Figure 6.4C). Similarly, pretreatment to high temperature (RHH) did not have

4989 a significant effect on CTmax across all species (Figure 6.4D). Nevertheless, there was a

4990 significant species effect for both CTmin and CTmax following RHH (Table 6.5B). In addition,

4991 there was a significant species * acclimation interaction effect for CTmin and a non-significant

4992 interaction effect for CTmax (Table 6.5B). 4993 4994

199

4995 Table 6.5: Summary statistical results from full factorial ANOVA showing effects of (A) RCH

4996 and (B) RHH on CTmin and CTmax for B. fusca, S. calamistis and C. partellus. SS = sums of 4997 squares, d.f.=degrees of freedom. 4998 A Trait Effect SS DF MS F P

CTmin Intercept 2632.03 1 2632.03 27924.78 ˂ 0.001 Species 360.92 2 180.46 1914.59 ˂ 0.001 RCH 12.68 1 12.68 134.48 ˂ 0.001 Species*Acclimation 76.87 2 38.45 407.78 ˂ 0.001 Error 10.75 114 0.094

CTmax Intercept 271463 1 271463 630788.1 ˂ 0.0001 Species 79.6 2 39.8 92.5 ˂ 0.0001 RCH 5.9 1 5.9 13.8 ˂ 0.001 Species*Acclimation 1.3 2 0.7 1.5 0.222 Error 49.1 114 0.4 B

CTmin Intercept 2836.3 1 2836.3 21842.72 ˂ 0.0001 Species 204.55 2 102.27 787.62 ˂ 0.0001 RHH 2.58 1 2.58 19.88 ˂ 0.0001 Species*Acclimation 113.81 2 56.91 438.25 ˂ 0.0001 Error 14.8 114 0.13

CTmax Intercept 273741.2 1 273741.2 895580.4 ˂ 0.0001 Species 60.5 2 30.2 98.9 ˂ 0.0001 RHH 0.1 1 0.1 0.2 0.645 Species*Acclimation 0.1 2 0.1 0.3 0.998 Error 34.8 114 0.3 4999

200

9 Control a A 8 RCH

o 7

6 b b

5 c 4

d 3 e

Critical thermal minima ( Criticalminima thermal 2

1 Busseola fusca Chilo partellus Sesamia calamistis Species 5000 5001

49 a Control ab B RCH ab

o b 48

c 47

Critical thermal maxima ( Criticalmaxima thermal

45 Busseola fusca Chilo partellus Sesamia calamistis Species 5002

201

9 Control a RHH C 8

6 b c 5

d d 4

e 3

Critical thermal minima ( minima thermal Critical

2 Busseola fusca Chilo partellus Sesamia calamistis

Species 5003

49 a Control a D RHH

C) o a a

b b 47

Critical thermal maxima ( maxima thermal Critical

46 Busseola fusca Chilo partellus Sesamia calamistis

Species 5004 5005

5006 Figure 6.4: Effects of RCH on (A) CTmin and (B) CTmax and RHH on (C) CTmin and (D) CTmax 5007 for Busseola fusca, Sesamia calamistis and Chilo partellus larvae. Error bars represent 95% CLs. 5008 Means with the same letter are not significantly different from each other. Post hoc tests were 5009 done separately for both treatments and across species

202

5010 6.3.5 Effects of Desiccation Acclimation on CTLs

5011 Like in starvation acclimation, desiccation acclimation significantly affected CTmin but

5012 not CTmax across all species (Table 6.6; Figure 6.5A; B). Desiccation acclimation improved

5013 CTmin across all species relative to their controls, with C. partellus recording the highest

5014 magnitude of plasticity (Figure 6.5A). Contrary to CTmin results, desiccation acclimation

5015 increased, marginally so, but not significant for B. fusca CTmax. Nevertheless, this acclimation

5016 did not have any significant effect on C. partellus and S. calamistis CTmax (Figure 6.5B).

5017 Furthermore, there was a significant species * acclimation interaction effect for CTmin and non-

5018 significant interaction effect for CTmax following desiccation acclimation (Table 6.6). 5019 5020 Table 6.6: Summary statistical results from full factorial ANOVA showing effects of desiccation

5021 acclimation on CTmin and CTmax for B. fusca, S. calamistis and C. partellus. SS = sums of 5022 squares, d.f.=degrees of freedom. Trait Effect SS DF MS F P

CTmin Intercept 1847.11 1 1847.11 23167.57 ˂ 0.001 Species 368.21 2 184.11 2309.18 ˂ 0.001 Desiccation Acclimation 141.27 1 141.27 1771.86 ˂ 0.001 Species*Acclimation 33.77 2 16.88 211.75 ˂ 0.001 Error 9.09 114 0.08

CTmax Intercept 274372 1 274372 718862.2 ˂ 0.0001 Species 39.3 2 19.7 51.5 ˂ 0.0001 Desiccation Acclimation 0.1 1 0.1 0.3 0.576 Species*Acclimation 2.4 2 1.2 3.2 0.0451 Error 43.5 114 0.4 5023

203

9 Control a A 8 Desiccation Acclimation

C) o 7

5 b c 4 d d 3

2 e

Critical thermal minima ( minima thermal Critical 1

0 Busseola fusca Chilo partellus Sesamia calamistis Species 5024

49 Control a ab B Desiccation Acclimation

o ab b

c 47

Critical thermal maxima ( maxima thermal Critical

46 Busseola fusca Chilo partellus Sesamia calamistis

Species 5025 5026

5027 Figure 6.5: Effects of desiccation acclimation on (A) CTmin and (B) CTmax for Busseola fusca, 5028 Sesamia calamistis and Chilo partellus. Error bars represent 95% CLs. Means with the same 5029 letter are not significantly different from each other. Post hoc tests were done separately for both 5030 treatments and across species

204

5031 6.4 Discussion 5032 Insect thermal sensitivity is mediated by environmental conditions (Santos et al., 2011). 5033 The current study revealed that C. partellus is generally more heat tolerant compared to B. fusca 5034 and S. calamistis. Furthermore, B. fusca was more cold resistant relative to C. partellus and S.

5035 calamistis. While my study reported a higher CTmax for C. partellus adults and lower CTmin for 5036 B. fusca adults relative to S. calamistis adults (Mutamiswa et al., 2017b), the current study shows 5037 the same trend for larvae across all species (Figure 6.1). Nevertheless, the current study reveals 5038 novel differences in magnitude of plasticity of CTLs across the three species following 5039 acclimation to heterogeneous stressors. Chilo partellus showed a higher magnitude of plasticity

5040 of low temperature tolerance (CTmin) following thermal, starvation, desiccation, short term low 5041 and high temperature acclimation relative to S. calamistis and B. fusca. The same trend was true

5042 for C. partellus following acclimation to temperature and starvation for CTmax. Such 5043 physiological attributes may confer survival merits for C. partellus upon introduction to novel 5044 habitats and may help explain the increased dominance of the stemborer over indigenous species 5045 (Mutamiswa et al., 2017a). Similarly, Ntiri et al. (2016b) reported that performance of these 5046 three stemborer species is dependent on thermal tolerance limits for development. Generally, B. 5047 fusca thrives under low temperatures while C. partellus favors warmer habitats. 5048 Effects of acclimation on insect thermal tolerance have been reported to vary across 5049 species (Kellet et al., 2005), developmental stages (Terblanche et al., 2007a; Marais et al., 2009)

5050 and also thermal history (Nyamukondiwa & Terblanche, 2010). As for CTmin, B. fusca and C. 5051 partellus larvae showed improved cold tolerance compared to S. calamistis larvae following low 5052 temperature acclimation (Figure 6.2A) indicating some improved potential for plasticity. 5053 Similarly, S. calamistis and B. fusca larvae also showed an enhanced cold tolerance for the same 5054 trait following acclimation to high temperature. These results are in keeping with previous study 5055 by Mutamiswa et al., (2017a) that reported improved performance of S. calamistis and B. fusca 5056 adults following acclimation to high temperature. Acclimation to high temperature improved

5057 CTmax for C. partellus relative to S. calamistis and B. fusca (Figure 6.2B) suggesting higher level 5058 of plasticity for C. partellus compared to other stemborer species. In consequence, the results 5059 reveal that C. partellus has a compromised basal low temperature tolerance and an improved 5060 basal high temperature tolerance, which comes with improved plasticity at both ends of 5061 temperature relative to other two borer species (see discussions in Nyamukondiwa et al., 2011).

205

5062 This is in agreement with previous study on Drosophila species which reported that insects may 5063 trade-off basal low but not high temperature tolerance for plasticity (Nyamukondiwa et al., 5064 2011). Given that global mean temperatures are projected to increase (Walther et al., 2002; 5065 Karuppaiah & Sujayanad, 2012), coupled with increased frequencies of heat waves and cold 5066 snaps (IPCC, 2014), C. partellus is highly likely to have a survival advantage at high 5067 temperatures relative to B. fusca and S. calamistis with lower thermal regimes. However, while 5068 B. fusca and S. calamistis may likely have a survival basal merit at low temperatures, it also 5069 appears, C. partellus may compensate for this through improved plasticity. This result affirms 5070 the notion that thermal trait characteristics among invasive species vary across taxonomic groups 5071 (Sakai et al., 2001). 5072 Starvation has been reported to influence thermal tolerance in mosquitoes (Culex 5073 fatigans), head lice (Pediculus humanus) and bed bugs (Cimex lectularius) (DeVries et al.,

5074 2016). The current results showed an enhanced high temperature tolerance (CTmax) for C. 5075 partellus relative to other stemborer species following starvation acclimation (Figure 6.3B), in 5076 keeping with results on C. lectularius (De Vries et al., 2016). However, starvation acclimation

5077 did not improve CTmax for B. fusca and S. calamistis in agreement with Scharf et al. (2016), who 5078 reported lack of improved heat tolerance (measured as heat knock down time) for Tribolium 5079 castaneum. In addition, B. fusca, S. calamistis and C. partellus showed an improved cold

5080 tolerance (measured as CTmin) following starvation acclimation (Figure 6.3A) in agreement with 5081 previous study on Locusta migratoria that reported improved cold tolerance (measured as chill 5082 coma recovery time) after 24 hour starvation period (Andersen et al., 2013). Similarly, starvation 5083 also improved cold tolerance for vinegar fly D. melanogaster (Le Bourg, 2015). The increased

5084 CTmax following starvation acclimation likely suggests shared physiological mechanisms for the 5085 traits for C. partellus and a lack thereof, for S. calamistis and B. fusca. Moreover, this may likely 5086 suggest that mechanisms underlying cross tolerance to environmental stressors are different

5087 across different taxonomic groups. Similarly, the improvement in CTmin following starvation 5088 acclimation indicate cross tolerance for the traits involved, and likely suggest coevolution of 5089 resistance mechanisms e.g. through lipid metabolism (Sinclair et al., 2015). Differential heat and 5090 cold tolerance responses for these stemborer species following starvation gives an insight on the 5091 difficulties these insects may face when they encounter multiple but simultaneous interacting 5092 environmental stressors in their natural habitats. In the current study, I found supporting

206

5093 evidence for cross-susceptibility (marginal decrease in tolerance) of upper thermal limits to 5094 starvation for S. calamistis in keeping with previous study on forest ant, A. picea (Nguyen et al., 5095 2017). DeVries et al. (2016) reported a correlation between lower metabolic rate and lower upper 5096 thermal limits for C. lectularius starved for 21 days. Low metabolism leads to low respiratory

5097 water loss hence the inability to cool through active evaporative cooling thus low CTmax (DeVries

5098 et al., 2016). Given that B. fusca recorded the lowest CTmax relative to S. calamistis and C. 5099 partellus, it’s highly likely that its metabolic rate may have been lowered during starvation. 5100 The current study also demonstrates insects’ response to CTLs following short term 5101 pretreatment (RCH and RHH). Busseola fusca and C. partellus recorded an improved cold stress

5102 resistance (CTmin) following RCH, in agreement with previous studies on D. melanogaster and 5103 F. candida which reported a significant cold stress resistance following cold hardening 5104 (Sejerkilde et al., 2003; Waagner et al., 2013). Nevertheless, magnitude of this improvement was 5105 asymmetrical, being higher for C. partellus relative to other two stemborer species (Figure 5106 6.4A). This cold tolerance may give survival advantage for these two borer species B. fusca and 5107 C. partellus when they experience cold winters in sub Saharan Africa, albeit C. partellus may 5108 have a fitness edge over B. fusca through greater plasticity magnitude. The results are also in 5109 keeping with previous studies to date (Assefa, 1985; Mwalusepo et al., 2015) that showed B. 5110 fusca preferring cooler climates compared to S. calamistis and C. partellus. However, cold 5111 hardening did not improve heat tolerance for all the borer species (Figure 6.4B) indicating

5112 evidence of marginal cross susceptibility of CTmax to RCH. The results also showed that RHH

5113 did not improve cold tolerance (CTmin) for S. calamistis while significantly compromising for B.

5114 fusca. However, RHH significantly improved CTmin for C. partellus (Figure 6.4C). This result 5115 suggests a survival demerit for S. calamistis and B. fusca when faced with these multiple 5116 stressors relative to C. partellus. Such cross tolerance has been reported in other studies (e.g. 5117 Bublily et al., 2012), and may be linked with overlap in mechanisms e.g. heat shock proteins. 5118 Although, the stemborer species were hardened at temperatures ranging from 40-41°C, RHH

5119 results showed negligible and non significant change in CTmax (Figure 6.4D) in keeping with

5120 Chidawanyika et al. (2017) who reported poor heat tolerance (measured as CTmax) using 5121 dynamic protocol for Z. bicolorata following 2 h hardening at 35°C. However, the reason for this 5122 trend is unknown. The beneficial acclimation hypothesis nevertheless states that acclimation 5123 response to changing environmental conditions would be beneficial for survival under the new

207

5124 set of conditions (Leori et al., 1994). Heat shock proteins have been reported to enhance heat 5125 resistance in insects (Sørensen et al., 2003). For example, Krebs (1999) reported that short term 5126 exposure of D. simulans adults to temperatures between 33 and 37°C resulted in an increased 5127 expression of Hsp70. Similarly, Tungjitwitayakul et al. (2015) reported production of heat shock 5128 proteins (Hsp70 and Hsp90) by Sitophilus zeamais following heat shock at 30–50°C for 1 h. 5129 However, in the current study S. calamistis and C. partellus were exposed to a sub-lethal 5130 temperature of 41°C for 2 hours, thus it is highly likely that enhanced heat tolerance for these 5131 species may have been facilitated by the production of heat shock proteins and other metabolic 5132 regulation pathways following RHH. However, the mechanisms behind cold and heat tolerance 5133 following RCH and RHH in these stemborer species still warrants further investigation.

5134 Like in starvation acclimation, desiccation acclimation also resulted in improved CTmin 5135 for all stemborer species (Figure 6.5A) indicating cross tolerance effects against cold stress. 5136 Insects have been reported to upregulate their heat shock proteins synthesis following desiccation 5137 (e.g Feder & Hofmann, 1999) and this response may have led to improved cold tolerance for the

5138 three borer species. This improved CTmin is in agreement with previous study on the soil 5139 collembolan, F. candida which reported improved cold tolerance following desiccation stress 5140 (Bayley et al., 2001). Ring & Danks (1994) reported that there is a linkage between desiccation 5141 tolerance and cold tolerance in insects with some species utilising desiccation as a mechanism 5142 for cold tolerance strategy (Holmstrup et al., 2002). Contrary to the relationship between

5143 starvation acclimation and CTmin, I found no significant association between starvation

5144 acclimation and CTmax. This result suggests no cross tolerance between traits of starvation and 5145 high temperature tolerance for the three stemborer species. A previous study on D. melanogaster

5146 has shown negligible impact of desiccation and starvation stress on CTmax (Kellermann et al., 5147 2012), in agreement with the current study. The current results also show marginal negative

5148 impacts of desiccation stress on high temperature tolerance (CTmax) for S. calamistis and C. 5149 partellus. Similarly, Tammarielo et al. (1999) reported a decrease in high temperature tolerance 5150 for flesh fly, Sarcophaga crassipalpis following 24 h desiccation stress. In tropical regions the 5151 frequencies and severity of droughts are expected to increase under climate change (IPCC, 5152 2014), hence small insects will be at risk of desiccation stress due to their high surface area to 5153 volume ratio (Bujan et al., 2016). Moreover, in southern African future climate change scenarios, 5154 extreme heat is expected to occur simultaneously with reduced rainfall (Lobell et al., 2008;

208

5155 Stathers et al., 2013) resulting in insects experiencing simultaneous heat and desiccation stress 5156 (Mueller & Seneviratne, 2012; Nguyen et al., 2016). This result suggests vulnerability of the 5157 three borer species when faced with multiple stressors of desiccation and high temperature. 5158 The current study demonstrates significant impacts of heterogeneous environmental 5159 stressors on stemborer species thermal tolerance and how that likely shapes their fitness traits

5160 under climate change. I document significant improvement in CTmin following acclimation to

5161 high temperature for B. fusca and S. calamistis and improvement in CTmax following high

5162 temperature acclimation for C. partellus. Similarly, I record costs of CTmin following acclimation 5163 to high temperature for C. partellus. Nevertheless, one thing is apparent for my study, that C. 5164 partellus has a higher magnitude of plasticity for CTLs following acclimation to temperature, 5165 starvation, RCH, RHH and desiccation. This increased plasticity may likely aid its invasion 5166 potential and establishment when faced with novel heterogeneous stressors ultimately displacing 5167 indigenous stemborer species in sub Saharan Africa. Indeed, invasive species should possess 5168 inherent high potential for rapid evolutionary response to changing environmental conditions 5169 (Lee, 2002; Whitney & Gabler, 2008). The results of this study have major implications on 5170 mechanisms allowing insects survival under diverse heterogeneous stressors and may likely 5171 explain invasive potential of economic insects. 5172

5173 References 5174 Ajayi, O. (1998) Sorghum: West Africa. In African Cereal Stem Borers: Economic Importance, 5175 Taxonomy, Natural Enemies and Control, pp. 39–45. Ed A. Polaszek.Wallingford, UK: 5176 CABI. van Asch M., Visser M.E. (2007) Phenology of forest caterpillars and their host 5177 trees: the importance of synchrony. Annual Review of Entomology, 52, 37–55. 5178 Alford, L., Burel, F. & van Baaren, J. (2016) Improving methods to measure critical thermal 5179 limits in phloem-feeding pest insects. Entomologia Experimentalis et Applicata, 159, 61-69. 5180 DOI: 10.1111/eea.12410. 5181 Andersen, J.L., Findsen, A. & Overgaard, J. (2013) Feeding impairs chill coma recovery in the 5182 migratory locust (Locusta migratoria). Journal of Insect Physiology, 59, 1041–1048. 5183 Assefa, G.A. (1985) Survey of lepidopterous stemborers attacking maize and sorghum in 5184 Ethiopia. Ethiopian Journal of Agricultural Science, 7, 55–59.

209

5185 Bale, J., Masters, G.J., Hodkins, I.D., Awmack, C., Bezemer, T.M., Brown, V.K., Buterfield, J., 5186 Buse, A., Coulson, J.C., Farrar, J., Good, J.E.G., Harrigton, R., Hartley, S., Jones, T.H., 5187 Lindroth, R.L., Press, M.C., Symrnioudis, I., Watt, A.D. & Whittaker, J.B. (2002) Herbivory 5188 in global climate change research: direct effects of rising temperature on insect herbivores. 5189 Global Change Biology, 8, 1-16. 5190 Bauerfeind, S.S. & Fischer, K. (2013) Increased temperature reduces herbivore host-plant 5191 quality. Global Change Biology, 19(11), 3272-3282. 5192 Bauerfeind, S.S., Kellermann, V., Moghadam, N.N., Loeschcke, V. & Fischer, K. (2014) 5193 Temperature and photoperiod affect stress resistance traits in Drosophila melanogaster. 5194 Physiological Entomology, 39, 237-246. 5195 Bayley, M., Petersen, S.O., Knigge, T., Kohler, H.R. & Holmstrup, M. (2001) Drought 5196 acclimation confers cold tolerance in the soil collembolan Folsomia candida. Journal of 5197 Insect Physiology, 47, 1197–1204. 5198 Bowler, K. & Terblanche, J.S. (2008) Insect thermal tolerance: what is the role of ontogeny, 5199 ageing and senescence? Biological Reviews, 83, 339–355. 5200 Bubliy, O.A., Kristensen, T.N., Kellermann, V. & Loeschcke, V. (2012) Plastic responses to four 5201 environmental stresses and cross-resistance in a laboratory population of Drosophila 5202 melanogaster. Functional Ecology, 26, 245–253. 5203 Bubliy, O.A., Kristensen, T.N. & Loeschcke, V. (2013) Stress –induced plastic responses in 5204 Drosophila simulans following exposure to combination of temperature and humidity 5205 levels. Journal of Experimental Biology, 216, 4601-4607. 5206 Bujan, J., Yanoviak, S.P. & Kaspari, M. (2016) Desiccation resistance in tropical insects: causes 5207 and mechanisms underlying variability in a Panama ant community. Ecology and Evolution, 5208 6, 6282-6291. 5209 Cavieres, G., Bogdanovich, J.M. & Bozinovic, F. (2016) Ontogenetic thermal tolerance and 5210 performance of ectotherms at variable temperatures. Journal of Evolutionary Biology, 29, 5211 1462-1468. 5212 Chidawanyika, F., Nyamukondiwa, C., Strathie, L. & Fischer, K. (2017) Effects of thermal 5213 regimes, starvation and age on heat tolerance of the Parthenium Beetle Zygogramma 5214 bicolorata (Coleoptera: Chrysomelidae) following dynamic and static protocols. PLoS ONE, 5215 12, e0169371. doi:10.1371/journal.pone.0169371.

210

5216 Chidawanyika, F. & Terblanche, J.S. (2011) Rapid thermal responses and thermal tolerance in 5217 adult codling moth Cydia pomonella (Lepidoptera: Totricidae). Journal of Insect Physiology, 5218 57, 108- 117. 5219 Chown, S. L., Jumbam, K.R., Sørensen, J.G. & Terblanche, J.S. (2009) Phenotypic variance, 5220 plasticity and heritability estimates of critical thermal limits depend on methodological 5221 context. Functional Ecology, 23, 133-140. 5222 Chown, S.L. & Nicolson, S.W. (2004) Insect Physiological Ecology: Mechanisms and Patterns. 5223 Oxford University Press, Oxford. 5224 Chown, S.L., Slabber, S., McGeoch, M.A., Janion, C. & Leinaas, H.P. (2007) Phenotypic 5225 plasticity mediates climate change responses among invasive and indigenous arthropods. 5226 Proceedings of the Royal Society B: Biological Sciences, 274, 2531–2537. 5227 Chown, S.L. & Terblanche, J.S. (2007) Physiological Diversity in Insects: Ecological and 5228 Evolutionary Contexts. Advances in Insect Physiology, 33, 50-152. 5229 Clusella-Trullas, S., Blackburn, T.M., Chown, S.L. (2011) Climatic predictors of temperature 5230 performance curve parameters in ectotherms imply complex responses to climate change. 5231 The American Naturalist, 177(6), 738-75. 5232 Da Lage, J.L., Capy, P. & David J.R. (1989) Starvation and desiccation tolerance in Drosophila 5233 melanogaster adults: Effects of environmental temperature. Journal of Insect Physiology, 5234 35, 453–457. 5235 Deutsch, C.A., Tewksbury, J.J., Huey, R.B., Sheldon, K.S., Ghalambor, C.K., Haak, D.C. & 5236 Martin, P.R. (2008) Impacts of climate warming on terrestrial ectotherms across latitude. 5237 Proceedings of the National Academy of Science of the United States of America, 105, 5238 6668–6672. 5239 DeVries, Z.C., Kells, S.A. & Appel, A.G. (2016) Estimating the critical thermal maximum 5240 (CTmax) of bed bugs, Cimex lectularius: Comparing thermolimit respirometry with 5241 traditional visual methods. Comparative Biochemistry and Physiology, Part A, 197, 52-57. 5242 Feder, M.E. & Hofmann, G.E. (1999) Heat-shock proteins, molecular chaperones, and the stress 5243 response: Evolutionary and ecological physiology. Annual Review of Physiology, 61, 243– 5244 282.

211

5245 Fischer, K., Dierks, A., Franke, K., Geister, T.L., Liszka, M., Winter, S. & Pflicke, C. (2010) 5246 Environmental effects on temperature stress resistance in the tropical butterfly Bicyclus 5247 anynana. PLoS ONE, 5(12), e15284. doi:10.1371/journal.pone.0015284. 5248 Fusi, M., Cannicci, S., Daffonchio, D., Mostert, B., Pörtner, H.-O. & Giomi, F. (2016) The trade- 5249 off between heat tolerance and metabolic cost drives the bimodal life strategy at the air- 5250 water interface. Scientific Reports, 6, 19158 10.1038/srep19158. 5251 Gilchrist, G.W., Huey, R.B. & Partridge, L. (1997) Thermal sensitivity of Drosophila 5252 melanogaster: Evolutionary responses of adults and eggs to laboratory natural selection at 5253 different temperatures. Physiological Zoology, 70, 403–414. 5254 Goto, S.G. & Kimura, M.T. (1998) Heat and shock responses and temperature adaptations in 5255 subtropical and temperate species of Drosophila. Journal of Insect Physiology, 44, 1233- 5256 1239. 5257 Hadley, N.F. (1994) Water relations of terrestrial arthropods. Academic Press, San Diego. 5258 Harrison, J.F., Woods, H.A. & Roberts, S.P. (2012) Ecological and environmental physiology of 5259 insects. Oxford University Press, New York. 5260 Hayes, K. R. & Barry, S.C. (2008) Are there any consistent predictors of invasion success? 5261 Biological Invasions, 10, 483-506. 5262 Hoffmann, A.A., Chown, S.L. & Clusella-Trullas, S. (2013) Upper thermal limits in terrestrial 5263 ectotherms: how constrained are they? Functional Ecology, 27, 934–949. 5264 Hoffmann, A.A., Sørensen, J.G. & Loeschcke, V. (2003) Adaptation of Drosophila to 5265 temperature extremes: bringing together quantitative and molecular approaches. Journal of 5266 Thermal Biology, 28, 175–216. 5267 Holmstrup, M., Bayley, M. & Ramløv, H. (2002) Supercool or dehydrate? An experimental 5268 analysis of overwintering strategies in small permeable Arctic invertebrates. Proceedings of 5269 the National Academy of Sciences of the United States of America, 99, 5716–5720. 5270 IPCC. (2014) Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and 5271 III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Core 5272 Writing Team, R.K. Pachauri and L.A. Meyer (eds.)]. IPCC, Geneva, Switzerland, pp 151. 5273 Jensen, D., Overgaard, J. & Sørensen, J.G. (2007) The influence of developmental stage on cold 5274 shock resistance and ability to cold-harden in Drosophila melanogaster. Journal of Insect 5275 Physiology, 53, 179–186.

212

5276 Kalra, B., Tamang, A.M. & Parkash, R. (2017) Cross tolerance effects due to adult heat 5277 hardening, desiccation and starvation acclimation of tropical Drosophilid Zaprionus 5278 indianus. Comparative Biochemistry and Physiology A. doi:10.1016/j.cbpa.2017.04.014. 5279 Karl, I., Becker, M., Hinzke, T., Mielke, M., Schiffler, M. & Fischer, K. (2014) Interactive 5280 effects of acclimation temperature and short-term stress exposure on resistance traits in the 5281 butterfly Bicyclus anynana. Physiological Entomology, 39, 222-228. 5282 Kellermann, V., Overgaard, J., Hoffmann, A.A., Fløjgaard, C., Svenning, J.C. & Loeschcke, V. 5283 (2012) Upper thermal limits of Drosophila are linked to species distributions and strongly 5284 constrained phylogenetically. Proceedings of the National Academy of Sciences, 109, 5285 16228-16233. 5286 Kellet, M., Hoffmann, A.A. & Mckechnie, S.W. (2005) Hardening capacity in the Drosophila 5287 melanogaster species group is constrained by basal thermotolerance. Functional Ecology, 5288 19, 853-858. 5289 Kelty, J.D. and Lee, R.E. Jr. (1999) Induction of rapid cold hardening by cooling at ecologically 5290 relevant rates in Drosophila melanogaster. Journal of Insect Physiology, 45, 719-726. 5291 Kfir, R. (1997) Competitive displacement of Busseola fusca (Lepidoptera: Noctuidae) by Chilo 5292 partellus (Lepdoptera: Pyralidae). Annals of the Entomological Society of America, 90, 619- 5293 624. 5294 Kfir, R., Overholt, W.A., Khan, Z.R. & Polaszek, A. (2002) Biology and Management of 5295 Economically Important Lepidopteran Cereal Stem Borers in Africa. Annual Review of 5296 Entomology, 47, 701-731. 5297 Khadioli, N., Tonnang, Z.E.H. Ong’amo, G., Achia, T., Kipchirchir, I., Kroschel, J. & Le Ru, B. 5298 (2014a) Effect of temperature on the life history parameters of noctuid lepidopteran 5299 stemborers, Busseola fusca and Sesamia calamistis. Annals of Applied Biology, 165, 373- 5300 386.doi:10.1111/aab.12157. 5301 Khadioli, N., Tonnang, Z.E.H., Muchugu, E., Ong’amo, G., Achia, T., Kipchirchir, I., Kroschel, 5302 J. & Le Ru, B. (2014b) Effect of temperature on the phenology of Chilo partellus (Swinhoe) 5303 (Lepidoptera: Crambidae), simulation and visualization of the potential future distribution of 5304 C. partellus in Africa under warmer temperatures through the development of life-table 5305 parameters. Bulletin of Entomological Research, 104, 809–822.

213

5306 Khani, A. & Moharramipour, S. (2010) Cold hardiness and supercooling capacity in the 5307 overwintering larvae of the codling moth, Cydia pomonella. Journal of Insect Science, 10, 83. 5308 Kingsolver, J.G., MacLean, H.J., Goddin, S.B. & Augustine, K.E. (2015) Plasticity of upper 5309 thermal limits to acute and chronic temperature variation in Manduca sexta larvae. Journal 5310 of Experimental Biology, 219, 1290-1294. 5311 Klok, C.J. & Chown, S.L. (2001) Critical thermal limits, temperature tolerance and water 5312 balance of a sub-Antarctic kelp fly, Paractora druexi (Diptera: Helcomyzidae). Journal of 5313 Insect Physiology, 47, 95–109. 5314 Krebs, R.A. (1999) A comparison of Hsp70 expression and thermotolerance in adults and larvae 5315 of three Drosophila species. Cell Stress Chaperones, 4, 243-249. 5316 Le Bourg, E. (2015) Fasting and other mild stresses with hormetic effects in Drosophila 5317 melanogaster can additively increase resistance to cold. Biogerontology, 16, 517–527. 5318 Lee, C.E. (2002) Evolutionary genetics of invasive species. Trends in Ecology & Evolution, 17, 5319 386-391. 5320 Lee, R.E. & Denlinger, D.L. (2010) Rapid cold-hardening: Ecological significance and 5321 underpinning mechanisms. Low Temperature Biology of Insects (ed. by D.L. Denlinger & 5322 R.E. Lee), Cambridge University Press, U.K. pp 35-58. 5323 Leori, A.M., Bennett, A.F. & Lenski, R.E. (1994) Temperature acclimation and competitive 5324 fitness: an experimental test of the beneficial acclimation assumption. Proceedings of 5325 National Academy of Sciences of the United States of America, 91, 1917-1921. 5326 Lobell, D.B., Burke, M.B., Tebaldi, C., Mastrandrea, M.D., Falcon, W.P & Naylor, R.I. (2008) 5327 Prioritising climate change adaptation needs for food security in 2030. Science, 319, 607- 5328 610. 5329 Ma, F.Z., Lu,¨ Z.C., Wang, R. and Wan, F.H. (2014) Heritability and evolutionary potential in 5330 thermal tolerance traits in the invasive Mediterranean cryptic species of Bemisia tabaci 5331 (Hemiptera: Aleyrodidae). PLoS ONE, 9, e103279. doi:10.1371/journal.pone.0103279. 5332 MacMillan, H.A., Walsh, J.P. & Sinclair, B.J. (2009) The effects of selection for cold tolerance 5333 on cross‐tolerance to other environmental stressors in Drosophila melanogaster. Insect 5334 Science, 16(3), 263-276. 5335 Mainali, K. P., Warren, D.L., Dhileepan, K., McConnachie, A., Strathie, L., Hassan, G., Karki, 5336 D., Shrestha, B.B. & Parmesan, C. (2015) Projecting future expansion of invasive species:

214

5337 comparing and improving methodologies for species distribution modeling. Global Change 5338 Biology, 21, 4464-4480. 5339 Marais, E., Terblanche, J.S. & Chown, S.L. (2009) Life-stage related differences in hardening 5340 and acclimation of thermal tolerance traits in the kelp fly, Paractora dreuxi (Diptera: 5341 Helcomyzidae). Journal of Insect Physiology, 55, 36-343. 5342 Marshal, K.E. & Sinclair, B.J. (2012) The impacts of repeated cold exposure on insects. Journal 5343 of Experimental Biology, 215, 1607-1613. 5344 Mudavanhu, P., Addison, P. & Conlong, D.E. (2014) Impact of mass rearing and gamma 5345 irradiation on thermal tolerance of Eldana saccharina. Entomologia Experimentalis et 5346 Applicata, 153, 55-63. 5347 Mueller, B. & Seneviratne, S.I. (2012) Hot days induced by precipitation deficits at the global 5348 scale. Proceedings of the National Academy of Sciences, 109, 12398–12403. 5349 Mutamiswa, R., Chidawanyika, F. & Nyamukondiwa, C. (2017a) Dominance of spotted 5350 stemborer Chilo partellus Swinhoe (Lepidoptera: Crambidae) over indigenous stemborer 5351 species in Africa’s changing climates: ecological and thermal biology perspectives. 5352 Agricultural and Forest Entomology, DOI: 10.1111/afe.12217. 5353 Mutamiswa, R., Chidawanyika, F. & Nyamukondiwa, C. (2017b) Comparative assessment of the 5354 thermal tolerance of spotted stemborer, Chilo partellus Swinhoe (Lepidoptera: Crambidae) 5355 and its larval parasitoid, Cotesia sesamiae Cameron (Hymenoptera: Braconidae). Insect 5356 Science, DOI 10.1111/1744-7917.12466. 5357 Mwalusepo, S., Tonnang, H.E.Z., Massawe, E.S., Okuku, G.O., Khadioli, N., Johansson, T., 5358 Calatayud, P. & Le Ru, B.P. (2015) Predicting the impact of temperature change on the 5359 future distribution of maize stemborers and their natural enemies along east African 5360 mountain gradients using phenology models. PLoS ONE, 10, e0130427. 5361 doi:10.1371/journal.pone.0130427. 5362 Nguyen, A.D., DeNovellis, K., Resendez, S., Pustilnik, J.D., Gotelli, N.J., Parker, J.D. & Cahan, 5363 S.H. (2017) Effects of desiccation and starvation on thermal tolerance and the heat-shock 5364 response in forest ants. Journal of Comparative Physiology B, DOI 10.1007/s00360-017- 5365 1101-x.

215

5366 Nyamukondiwa, C., Kleynhans, E. & Terblanche, J.S. (2010) Phenotypic plasticity of thermal 5367 tolerance contributes to the invasion potential of Mediterranean fruit flies (Ceratitis 5368 capitata). Ecological Entomology, 35, 565-575. 5369 Nyamukondiwa, C. & Terblanche, J.S. (2010) Within-generation variation of critical thermal 5370 limits in adult Mediterranean and Natal fruit flies Ceratitis capitata and Ceratitis rosa: 5371 thermal history affects short-term responses to temperature. Physiological Entomology, 35, 5372 255-264. 5373 Nyamukondiwa, C. & Terblanche, J.S. (2009) Thermal tolerance in adult Mediterranean and 5374 Natal fruit flies (Ceratitis capitata and Ceratitis rosa): effects of age, gender and feeding 5375 status. Journal of Thermal Biology, 34, 406-414. 5376 Nyamukondiwa, C., Terblanche, J.S., Marshall, K.E. & Sinclair, B.J. (2011) Basal but not heat 5377 tolerance constraints plasticity among Drosophila species (Diptera: Drosophilidae). Journal 5378 of Evolutionary Biology, 24, 1927-1938. 5379 Ntiri, E.S, Calatayud, P.A., Van den Berg, J. & Le Ru, B.P. (2016a) Density dependence and 5380 temporal plasticity of competitive interactions during utilisation of resources by a 5381 community of lepidopteran stemborer species. Entomologia Experimentalis et Applicata, 5382 162, 272-283. 5383 Ntiri, E.S., Calatayud, P.A., Van den Berg, J., Schulthess, F. & Le Ru, B.P. (2016b) Influence of 5384 temperature on intra- and interspecific resource utilization within a community of 5385 lepidopteran maize stemborers. PLoS ONE, 11, e0148735. 5386 Ochieng, R.S., Onyango, F.O. & Bungu, M.D.O. (1985) Improvement of techniques for mass 5387 culture of Chilo partellus (Swinhoe). Insect Science and its Application, 6, 425-428. 5388 Ong’amo, G.O., Le Rü, B.P., Moyal, P.A., Calatayud, P., Le Gall, P., Ogol, C.K.P.O., Kokwaro, 5389 E.D., Capdevielle-Dulac, C. & Silvain, J.F. (2008) Host plant diversity of Sesamia 5390 calamistis: cytochrome b gene sequences reveal local genetic differentiation. Entomologia 5391 Experimentalis et Applicata, 128, 154-161. 5392 Overgaard, J. & Sorensen, J.G. (2008) Rapid thermal adaptation during field temperature 5393 variations in Drosophila melanogaster. Cryobiology, 56,159–162. 5394 Overholt, W. A., Songa, J.M., Ofomata, V. & Jeske, R. (2000) The spread and ecological 5395 consequences of the invasion of Chilo partellus (Swinhoe) (Lepidoptera: Crambidae) in

216

5396 Africa. In E. E. Lyons and S. E. Miller Invasive species in eastern Africa: Proceedings of a 5397 Workshop held at ICIPE, July 5-6, 1999. ICIPE Science Press, Nairobi. 5398 Renault, D., Henry, Y. & Colinet, H. (2015) Exposure to desiccating conditions and cross- 5399 tolerance with thermal stress in the Lesser mealworm Alphitobius diaperinus (Coleoptera: 5400 Tenebrionidae). Revue d’Ecologie (Terre et Vie), 70, 33-41. 5401 Ring, R.A. & Danks, H.V. (1994) Desiccation and cryoprotection: Overlapping adaptations. 5402 CryoLetters, 15, 181–190. 5403 Rion, S. & Kawecki, T.J. (2007) Evolutionary biology of starvation resistance: What we have 5404 from Drosophila. Journal of Evolutionary Biology, 20, 1655-1664. 5405 Sakai, A. K., Allendorf, F.W., Holt, J.S., Lodge, D.M., Molofsky, J., With, K.A., Baughman, S., 5406 Cabin, R.J., Cohen, J.E., Ellstrand, N.C., McCauley, D.E., O'Neil, P., Parker, I.M., 5407 Thompson, J.N. & Weller, S.G. (2001) The population biology of invasive species. Annual 5408 Review of Ecology and Systematics, 32, 305-332. 5409 Santos, M., Castaneda, L.E. & Rezende, E.L. (2011) Making sense of heat tolerance estimates in 5410 ectotherms: lessons from Drosophila. Functional Ecology, 25, 1169-1180. 5411 Scharf, I., Daniel, A., MacMillan, H.A. & Katz, N. (2017) The effect of fasting and body 5412 reserves on cold tolerance in 2 pit-building insect predators. Current Zoology, 63(3), 287- 5413 294. 5414 Scharf, I. Galkin, N. & Halle, S. (2015) Disentangling the consequences of growth temperature 5415 and adult acclimation temperature on starvation and thermal tolerance in the red flour beetle. 5416 Evolutionary Biology, 42, 54–62. 5417 Scharf, I., Wexler, Y., MacMillan, H.A., Presman, S., Simson, E. & Rosenstein, S. (2016) The 5418 negative effect of starvation and the positive effect of mild thermal stress on thermal tolerance of 5419 the red flour beetle, Tribolium castaneum. Science of Nature, 103, 20. DOI 10.1007/s00114- 5420 016-1344-5. 5421 Sejerkilde, M., Sorensen, J.G. & Loeschcke, V. (2003) Effects of cold and heat hardening on 5422 thermal resistance in Drosophila melanogaster. Journal of Insect Physiology, 49, 719-726 5423 Sinclair, B.J., Coello Alvarado, L.E. & Ferguson, L.V. (2015) An invitation to measure insect 5424 cold tolerance: Methods, approaches, and workflow. Journal of Thermal Biology, 53, 180- 5425 197.

217

5426 Sinclair, B.J., Ferguson, L.V., Salehipour-shirazi, G. & MacMillan, H.A. (2013) Cross- tolerance 5427 and cross-talk in the cold: relating low temperatures to desiccation and immune stress in 5428 insects. Integrative and Comparative Biology, 53, 545-556. 5429 Sinclair, B.J., Nelson, S., Nilson, T.L., Roberts, S.P. & Gibbs, A.G. (2007) The effect of 5430 selection for desiccation resistance on cold tolerance of Drosophila melanogaster. 5431 Physiological Entomology, 32(4), 322-327. 5432 Sinclair, B.J. & Roberts, S.P. (2005) Acclimation, shock and hardening in the cold. Journal of 5433 Thermal Biology, 30, 557–562. 5434 Somero, G.N. (2005) Linking biogeography to physiology: evolutionary and acclimatory 5435 adjustments of thermal limits. Frontiers in Zoology, 2, 1. doi:10.1186/1742-9994-2-1. 5436 Sørensen, J.G., Kristensen, T.N. & Loeschcke, V. (2003) The evolutionary and ecological role of 5437 heat shock proteins. Ecology Letters, 6, 1025–1037. 5438 Spicer, J.I. & Gaston, K.J. (1999) Physiological Diversity and its Ecological Implications. 5439 Blackwell Science Ltd. Oxford. 5440 Stathers, T., Lamboll. R. & Mvumi, B.M. (2013) Postharvest agriculture in changing climates: 5441 its importance to African smallholder farmers. Food Security, 5, 361-392. 5442 Tammarielo, S.T., Rinehart, J.P., & Denlinger, D. L. (1999) Desiccation elicits heat shock 5443 protein transcription in the flesh fly, Sacophaga crassipalpis, but does not enhance tolerance 5444 to high or low temperatures. Journal of Insect Physiology, 45, 933-938. 5445 Tefera, T., Mugo, S., Tende, R. & Likhayo, P. (2010) Mass Rearing of Stem borers, Maize 5446 weevil and Larger grain borer Insect Pests of Maize. CIMMYT. Nairobi. 5447 Terblanche, J.S., Deere, J.A., Clusella-Trullas, S., Janion, C. & Chown, S.L. (2007a) Critical 5448 thermal limits depend on methodological context. Proceedings of the Royal Society of 5449 London B, 274, 2935-2942. 5450 Terblanche, J.S., Marais, E. & Chown, S.L. (2007b) Stage-related variation in rapid cold 5451 hardening as a test of the environmental predictability hypothesis. Journal of Insect 5452 Physiology, 53, 455–462. 5453 Terblanche, J.S., Klok, C.J., Krafsur, E.S. & Chown, S.L. (2006) Phenotypic plasticity and 5454 geographic variation in thermal tolerance and water loss of the tsetse Glossina pallidipes 5455 (Diptera: Glossinidae): implications for distribution modelling. The American Society of 5456 Tropical Medicine and Hygiene, 74, 786–794.

218

5457 Tungjitwitayakul, J., Tatun, N., Vajarasathira, B. & Sakurai, S. (2015) Expression of heat shock 5458 protein genes in different developmental stages and after temperature stress in the maize 5459 weevil (Coleoptera: Curculionidae). Journal Economic Entomology, 108(3), 1313–1323. 5460 Verberk, W.C.E.P., Overgaard, J., Ern, R., Bayley, M., Wang, T., Boardman, L. & Terblanche, 5461 J.S. (2016) Does oxygen limit thermal tolerance in arthropods? A critical review of current 5462 evidence. Comparative biochemistry and physiology. Part A, Molecular & integrative 5463 physiology, 192, 64–78. 5464 Vernon, P. & Vannier, G. (1996) Developmental patterns of supercooling capacity in a 5465 subantarctic wingless fly. Experientia, 52, 155–158. 5466 Waagner, D., Holmstrup, M., Bayley, M. & Sørensen, J.G. (2013) Induced cold-tolerance 5467 mechanisms depend on duration of acclimation in the chill sensitive Folsomia candida 5468 (Collembola). The Journal of Experimental Biology, 216, 1991-2000. 5469 Weldon, CW., Terblanche, J.S. & Chown, S.L. (2011) Time-course for attainment and reversal 5470 of acclimation to constant temperature in two Ceratitis species. Journal of Thermal Biology, 5471 36, 479-485. 5472 Whitney, K.D. & Gabler, C.A. (2008) Rapid evolution in introduced species, 'invasive traits' and 5473 recipient communities: challenges for predicting invasive potential. Diversity and 5474 Distributions, 14, 569-580. 5475 Woods, H.A. & Singer, M.S. (2001) Contrasting responses to desiccation and starvation by eggs 5476 and neonates of two Lepidoptera. Physiological and Biochemical Zoology, 74, 594–606 5477 Zhou, G., Overholt, W.A. & Mochiah, M.B. (2001) Change in the distribution of lepidopteran 5478 maize stem borer in Kenya from the 1950s to 1990s. Insect Science and its Application, 21, 5479 395–402. 5480 5481 5482 5483 5484 5485 5486 5487

219

5488 5489

5490 CHAPTER 7

5491

5492

5493

5494

5495

5496 General Discussion

5497 5498 5499 5500 5501 5502 5503 5504 5505 5506 5507 5508 5509 5510 5511 5512 5513

220

5514 Climate change effects on agro-ecosystems are becoming apparent with seasonal and 5515 long term variations influencing insect pests population dynamics, abundance and activities of 5516 natural enemies and efficacy of pest management (Selvaraj et al., 2013). In the agroecosystem, 5517 higher trophic levels, usually associated with natural enemies are highly sensitive to climate 5518 variability compared to the hosts (Voigt et al., 2003). In this respect, divergences in the thermal 5519 tolerances of the pest and its natural enemy may ultimately lead to a disruption of their temporal 5520 or geographical synchronization, increasing the risk of host outbreaks (Hance et al., 2007). 5521 Although there have been extensive studies on thermal tolerance of various insect pests (e.g. 5522 Koveos, 2001; Grout & Stoltz, 2007; Rajamohan & Sinclair, 2008; Kalosaka et al., 2009; 5523 Nyamukondiwa & Terblanche, 2009; Chidawanyika & Terblanche, 2011; Piyaphongkul, 2013), 5524 to the best of my knowledge, this is the first study detailing climate variability effects on 5525 stemborers-natural enemies trophic level interaction and how this may impact on biological 5526 control under global change. 5527 The present study reviewed the ecological and thermal physiological factors contributing 5528 towards C. partellus invasion potential and dominance over indigenous stemborer species in 5529 Africa (Chapter 2; Mutamiswa et al., 2017a). Laboratory experiments showed that the exotic C. 5530 partellus had a higher basal thermal tolerance and plasticity thereof, compared to indigenous B. 5531 fusca and S. calamistis. This indicated that C. partellus trades off its basal low but not high 5532 temperature tolerance for plasticity as in like insect taxa (e.g Nyamukondiwa et al., 2011), 5533 further explaining that ecophysiology may aid its invasion potential relative to B. fusca and S. 5534 calamistis. Apart from its improved physiological tolerances, the study also showed that short 5535 generation time, overwintering physiology, temperature and relative humidity resilience, wide 5536 host preferences and asynchrony with biocontrol agents may have facilitated C. partellus 5537 invasion potential and dominance over indigenous stemborer species. These attributes revealed 5538 that C. partellus is highly competitive and with Africa’s changing climate, it has a high 5539 likelihood of expanding and establishing in novel habitats. 5540 Basal and plasticity of thermal tolerance of stemborers and their coexisting stemborer 5541 parasitoids determine their abundance, biogeographical patterns and activity over spatial and 5542 temporal scales ultimatly impacting biocontrol programs under climate variability. The current 5543 study also explored the comparative thermal tolerance of different developmental stages of Chilo 5544 partellus and its larval parasitoid, Cotesia sesamiae (Chapter 3; Mutamiswa et al., 2017b) and

221

5545 coexisting parasitoid species (C. sesamiae and C. flavipes) (Chapter 5) through measuring their 5546 lower and upper lethal temperatures, critical limits to activity, supercooling points, chill-coma 5547 recovery time and heat knock-down time. The severity and duration of exposure influenced the 5548 survival of C. partellus, C. sesamiae and C. flavipes in agreement with previous studies (Chown 5549 & Terblanche, 2007; Chidawanyika & Terblanche, 2011). LLTs and ULTs comparisons for C. 5550 partellus and C. sesamiae (2 h duration) showed the parasitoid being less cold and heat tolerant 5551 relative to the host (Figure 3.1B; 3.1D; 3.2A; 3.2B; Chapter 3). On the other hand, for 0.5–4 h 5552 treatments, LLTs for C. sesamiae and C. flavipes ranged from -5 to 5°C and -15 to -1°C whereas 5553 ULTs ranged from 35 to 42°C and 37 to 44°C respectively (Figure 5.1A-D; Chapter 5), 5554 suggesting that C. flavipes is more cold and heat tolerant than C. sesamiae. Critical thermal 5555 limits are dependent on methodological context (Terblanche et al. (2007) and experiments 5556 conducted using ecologically relevant ramping rates showed slower ramping rates significantly

5557 reducing CTmax in C. partellus adults, larvae and C. sesamiae adults (Figure 3.3A; C; E; Chapter 5558 3) indicating that in the short term these species may not be able to adjust their phenotypic 5559 plasticities of high temperature tolerance (see discussions in Nyamukondiwa & Terblanche,

5560 2010; Sgrò et al., 2016). Similarly, slower cooling rates significantly improved CTmin in adult C. 5561 partellus and C. sesamiae indicating that a prolonged low temperature exposure may allow the 5562 insects to develop some low temperature protection, suggesting RCH. Cotesia flavipes had lower

5563 CTmin and higher CTmax than C. sesamiae (Figure 5.2; Chapter 5), indicating a fitness advantage 5564 for the exotic C. flavipes relative to indigenous C. sesamiae at both thermal extremes. This result 5565 translates into a broader thermal safety margin for C. flavipes which ultimately confers better 5566 optimization of key life history activities such as host searching ability (Mailafiya et al., 2009) 5567 more than congeneric C. sesamiae under stressful and variable thermal regimes. Cotesia flavipes 5568 recovered faster from chill coma and took more time to be knockdown by high temperature than 5569 C. sesamiae (Figure 5.4; 5.5; Chapter 5) signifying its survival and fitness advantage to cold and 5570 heat exposure relative to its congener. Furthermore, SCPs were recorded for the first time for 5571 these two Cotesia species and ranged ~ -20°C. In the current study the two Cotesia species failed 5572 to recover after supecooling suggesting that they were chill susceptible (killed by chilling) 5573 (Sinclair et al., 2015). With the host larvae being the most targeted developmental stage by the 5574 parasitoids, the differential thermal tolerance responses indicate a mismatch between C. partellus 5575 and C. sesamiae in the face of climate change. Consequently, this may negatively affect future

222

5576 biological control efficacy. Furthermore, with African temperatures projected to increase 5577 (Gemeda & Sima, 2015), coupled with increased likelihood of prolonged heat waves, and cold 5578 snaps (IPCC, 2014), it is highly likely that C. partellus may survive both prolonged high and low 5579 temperatures better than C. sesamiae. This shows potential risks associated with C. partellus 5580 outbreaks, which may negatively affect cereal crop productivity, household and national food 5581 security. Similarly, C. flavipes may survive more prolonged low and high temperature extremes 5582 than its congener C. sesamiae due to its superior thermal physiology ultimately displacing the 5583 generalist indigenous parasitoid thus more oubreaks of other stemborer species. 5584 An understanding of plasticity of cold and heat tolerance provides an insight for survival 5585 and interaction of different trophic level species and how natural enemies released for biocontrol 5586 will establish under natural conditions (Chapter 4). The current study showed evidence of RCH 5587 and RHH responses in C. partellus and C. flavipes in corroboration with previous studies (e.g 5588 Overgaard & Sorenson, 2008; Kalosaka et al., 2009; Stotter & Terblanche, 2009; 5589 Nyamukondiwa et al., 2010; Nyamukondiwa & Terblanche, 2010; Chidawanyika & Terblanche, 5590 2011; Chidawanyika et al., 2017). Microclimatic temperature data recorded in an agro-ecosystem 5591 where both C. partellus and C. flavipes coexist showed that temperatures eliciting RCH and 5592 RHH responses at both trophic levels occur naturally (Figure 4.4; Chapter 4). This may imply 5593 that under projected climate change (IPCC, 2014), RCH and RHH may enhance fitness for both 5594 species albeit C. partellus recorded a lower cold hardening and discriminating temperature and 5595 higher heat hardening and discriminating temperature relative to C. flavipes (Table 4.1; Chapter 5596 4). Species comparison of CCRT and SCP (following acclimation to low and optimum

5597 temperatures) and CTmin (across all acclimation treatments) between the two trophic levels 5598 (Figure 4.2C; 4.2D; 4.2E; Chapter 4), showed that C. flavipes is more cold tolerant than C. 5599 partellus. In consequence, this may have a fitness and survival advantage for C. flavipes relative 5600 to C.partellus under variable low temperatures hence positive trend on biological control under

5601 global change (see Agrawal, 2001). On the other hand, C. partellus showed a higher CTmax than 5602 C. flavipes across all acclimation treatments indicating that the host is more heat tolerant than the 5603 parasitoid. While C. flavipes was more cold tolerant than C. partellus and Africa projected to 5604 experience rise in temperatures, erratic rainfall patterns and increased extreme weather events 5605 (Stathers et al., 2013; IPCC, 2014), the host is highly likely to survive high temperatures

223

5606 compared to its parasitoid, and this may create asynchronous phenologies with negative 5607 implications on biological control efficacy. 5608 In nature, insects are often exposed to episodic multiple environmental stressors that may 5609 influence their activity, survival abundance and geographic range expansion. In this study, I 5610 investigated the effects of acclimation to temperature, starvation and desiccation on thermal 5611 tolerance, measured as critical thermal limits on laboratory reared B. fusca, S. calamistis and 5612 C.partellus using dynamic (ramping) protocols (Chapter 6). Chilo partellus had a higher 5613 magnitude of plasticity of low temperature tolerance following thermal, starvation, desiccation, 5614 short term low and high temperature acclimation relative to B. fusca and S. calamistis. In 5615 addition, C. partellus also showed the same trend following thermal and starvation acclimation 5616 for high temperature tolerance. This indicates fitness and survival advantage for C. partellus 5617 upon introduction to novel habitats and may reflect its increased dominance over indigenous 5618 species (Mutamiswa et al., 2017a). Busseola fusca and C. partellus larvae showed improved cold 5619 tolerance following low temperature acclimation whereas S. calamistis and B. fusca larvae also 5620 showed an enhanced cold tolerance following acclimation to high temperature in keeping with 5621 previous study (Mutamiswa et al., 2017a). On the other hand, high temperature acclimation

5622 improved CTmax for C. partellus relative to S. calamistis and B. fusca suggesting higher plasticity 5623 levels for C. partellus relative to other stemborer species. Overall, the results show that C. 5624 partellus has a compromised basal low temperature tolerance and an improved basal high 5625 temperature tolerance, which is generally associated with improved plasticity relative to other 5626 two borer species (see discussions in Nyamukondiwa et al. 2011). Starvation acclimation 5627 improved high temperature tolerance for C. partellus relative to other stemborer species (Figure 5628 6.3B; Chapter 6), in keeping with C. lectularius results (De Vries et al., 2016). Furthermore, 5629 starvation also improved cold tolerance (Figure 6.3A; Chapter 6) for B. fusca, S. calamistis and 5630 C. partellus in agreement with previous study on Locusta migratoria (Andersen et al., 2013) and 5631 D. melanogaster (Le Bourg, 2015). Overall, this improvement in cold and heat tolerance 5632 following starvation acclimation indicates cross tolerance effects for the involved traits. Busseola 5633 fusca and C. partellus recorded an improved cold stress resistance following RCH, in agreement 5634 with previous studies on D. melanogaster and F. candida (Sejerkilde et al., 2003; Waagner et al., 5635 2013) although C. partellus had a higher magnitude of improvement (Figure 6.4A; Chapter 6). 5636 This cold tolerance may give these two borer species survival advantage when they encounter

224

5637 cold winters in sub Saharan Africa, albeit C. partellus may have fitness superiority over B. fusca 5638 through greater magnitude of plasticity. On the other hand, RHH significantly improved cold 5639 tolerance for C. partellus (Figure 6.4C; Chapter 6) indicating cross tolerance as reported in 5640 previous studies (e.g. Bublily et al., 2012). This implies that S. calamistis and B. fusca may have 5641 survival disadvantage when faced with these multiple stressors in their natural habitats relative to 5642 C. partellus. Desiccation acclimation also resulted in improved cold tolerance for all stemborer 5643 species (Figure 6.5A; Chapter 6) indicating cross tolerance effects against cold stress. Insects 5644 have been reported to upregulate their heat shock proteins synthesis following desiccation (e.g. 5645 Feder & Hofmann, 1999) and this response may have led to improved cold tolerance for the

5646 three borer species. Desiccation stress marginally impaired high temperature tolerance (CTmax) 5647 for S. calamistis and C. partellus in keeping with previous study by Tammarielo et al. (1999). 5648 Given these heat and cold tolerance variabilities for these stemborer species following hardening, 5649 thermal, starvation and desiccation acclimation, it therefore gives an insight on the difficulties 5650 these insects may face when they experience multiple environmental stressors in their natural 5651 habitats. 5652 While the current study investigated basal thermal tolerance and plasticity for laboratory 5653 reared stemborer and parasitoid species, future studies should consider comparative assessment 5654 of laboratory and wild populations for the same traits to determine their fitness and survival 5655 under climate change. Although the current study explored the effects of environmental stressors 5656 on thermal tolerance (measured as CTLs) on stemborer species only, future studies should focus 5657 on determining how these stressors may influence thermal tolerance (incorporating more life 5658 history traits) of these stemborer species and their biological control agents under climate 5659 change. The current study revealed evidence of hardening (RCH and RHH) in C. partellus and 5660 C. flavipes which led to improved cold and heat tolerance in these species, thus future studies 5661 should investigate mechanisms behind these hardening responses. Furthermore, a combined 5662 temperature and relative humidity variability effect on life history traits of stemborers and their 5663 parasitoids should also be considered in future studies. 5664 In conclusion, the results of this thesis indicate that climate variability may have dire 5665 consequences on stemborers-natural enemies trophic level interactions in sub Saharan Africa. 5666 Variations in thermal preferences for stemborer species and their associated parasitoid species 5667 recorded in this project may lead to changes in temporal and geographical synchronization

225

5668 ultimately increasing the risks of host outbreaks. Because of its superior basal thermal tolerance 5669 and plasticity thereof, exotic C. partellus has a high likelihood of displacing indigenous 5670 stemborer species, B. fusca and S. calamistis as well as escaping parasitism from C. sesamiae 5671 and C. flavipes with low thermal regimes under global change. This may negatively influence 5672 future classical biological control programs since the host may have a comparative advantage 5673 following exposure to extreme conditions relative to the parasitoid. Overall the results of this 5674 project provide valuable information for predictive models for forecasting economic insect pest 5675 outbreaks, conducting physosanitary risk assessments, establishing invasive species, 5676 biogeographical shifts, and the potential implications of global climate change. 5677

5678 References 5679 Agrawal, A.A. (2001) Phenotypic plasticity in the interactions and evolution of species. Science, 5680 294, 321-325. 5681 Andersen, J.L., Findsen, A. & Overgaard, J. (2013) Feeding impairs chill coma recovery in the 5682 migratory locust (Locusta migratoria). Journal of Insect Physiology, 59, 1041–1048. 5683 Bubliy, O.A., Kristensen, T.N., Kellermann, V. & Loeschcke, V. (2012) Plastic responses to four 5684 environmental stresses and cross-resistance in a laboratory population of Drosophila 5685 melanogaster. Functional Ecology, 26, 245–253. 5686 Chidawanyika, F., Nyamukondiwa, C., Strathie, L. & Fischer, K. (2017) Effects of thermal 5687 regimes, starvation and age on heat tolerance of the Parthenium Beetle Zygogramma 5688 bicolorata (Coleoptera: Chrysomelidae) following dynamic and static protocols. PLoS ONE, 5689 12, e0169371. doi:10.1371/journal.pone.0169371. 5690 Chidawanyika, F. & Terblanche, J.S. (2011) Rapid thermal responses and thermal tolerance in 5691 adult codling moth Cydia pomonella (Lepidoptera: Totricidae). Journal of Insect Physiology, 5692 57, 108-117. 5693 Chown, S.L. & Terblanche, J.S. (2007) Physiological Diversity in Insects: Ecological and 5694 Evolutionary Contexts. Advances in Insect Physiology, 33, 50-152. 5695 DeVries, Z.C., Kells, S.A. & Appel, A.G. (2016) Estimating the critical thermal maximum 5696 (CTmax) of bed bugs, Cimex lectularius: Comparing thermolimit respirometry with 5697 traditional visual methods. Comparative Biochemistry and Physiology, Part A, 197, 52-57.

226

5698 Feder, M.E. & Hofmann, G.E. (1999) Heat-shock proteins, molecular chaperones, and the stress 5699 response: Evolutionary and ecological physiology. Annual Review of Physiology, 61, 243– 5700 282. 5701 Gemeda, D.O. & Sima, A.D. (2015) The impacts of climate change on African continent and the

5702 way forward. Journal of Ecology and the Natural Environment, 7(10), 256-262.

5703 Grout, T.G. & Stoltz, K.C. (2007) Developmental rates at constant temperatures of three 5704 economically important Ceratitis species (Diptera: Tephritidae) from Southern Africa. 5705 Environmental Entomology, 36, 1310-1317. 5706 Hance, T., Van Baaren, J., Vernon, P. & Boivin, G. (2007) Impact of extreme temperatures on 5707 parasitoids in a climate change perspective. Annual Review of Entomology, 52, 107–126. 5708 IPCC. (2014) Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and 5709 III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Core 5710 Writing Team, R.K. Pachauri and L.A. Meyer (eds.)]. IPCC, Geneva, Switzerland, pp 151. 5711 IPCC (2007) In: Parry, M.L., Canziani, O.F., Palutikof, J.P., van der Linden, P.J. and Hanson 5712 C.E. (Eds), Climate change 2007: Impacts, adaptation and vulnerability, Contribution of 5713 Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on 5714 Climate Change. Cambridge University Press. Cambridge. UK. 5715 Kalosaka, K., Soumaka, E., Politis, N. & Mintzas, A.C. (2009) Thermotolerance and Hsp70 5716 expression in Mediterranean fruit fly Ceratitis capitata. Journal of Insect Physiology, 55, 5717 568-573. 5718 Koveos, D.S. (2001) Rapid cold hardening in the olive fruit fly Bactrocera oleae under 5719 laboratory conditions. Entomologia Experimentalis et Applicata, 101, 257-263. 5720 Le Bourg, E. (2015) Fasting and other mild stresses with hormetic effects in Drosophila 5721 melanogaster can additively increase resistance to cold. Biogerontology, 16, 517–527. 5722 Mailafiya, D.M., Le Ru, B..P, Kairu, E.W., Calatayud, PA. & Dupas, S. (2009) Species diversity 5723 of lepidopteran stem borer parasitoids in cultivated and natural habitats in Kenya. Journal of 5724 Applied Entomology, 133, 416-429. 5725 Mutamiswa, R., Chidawanyika, F. & Nyamukondiwa, C. (2017a) Dominance of spotted 5726 stemborer Chilo partellus Swinhoe (Lepidoptera: Crambidae) over indigenous stemborer

227

5727 species in Africa’s changing climates: ecological and thermal biology perspectives. 5728 Agricultural and Forest Entomology, DOI: 10.1111/afe.12217. 5729 Mutamiswa, R., Chidawanyika, F. & Nyamukondiwa, C. (2017b) Comparative assessment of the 5730 thermal tolerance of spotted stemborer, Chilo partellus Swinhoe (Lepidoptera: Crambidae) 5731 and its larval parasitoid, Cotesia sesamiae Cameron (Hymenoptera: Braconidae). Insect 5732 Science, DOI 10.1111/1744-7917.12466. 5733 Nyamukondiwa, C., Kleynhans, E. & Terblanche, J.S. (2010) Phenotypic plasticity of thermal 5734 tolerance contributes to the invasion potential of Mediterranean fruit flies (Ceratitis 5735 capitata). Ecological Entomology, 35, 565-575. 5736 Nyamukondiwa, C. & Terblanche, J.S. (2010) Within-generation variation of critical thermal 5737 limits in adult Mediterranean and Natal fruit flies Ceratitis capitata and Ceratitis rosa: 5738 thermal history affects short-term responses to temperature. Physiological Entomology, 35, 5739 255-264. 5740 Nyamukondiwa, C. & Terblanche, J.S. (2009) Thermal tolerance in adult Mediterranean and 5741 Natal fruit flies (Ceratitis capitata and Ceratitis rosa): effects of age, gender and feeding 5742 status. Journal of Thermal Biology, 34, 406-414. 5743 Nyamukondiwa, C., Terblanche, J.S., Marshall, K.E. & Sinclair, B.J. (2011) Basal but not heat 5744 tolerance constraints plasticity among Drosophila species (Diptera: Drosophilidae). Journal 5745 of Evolutionary Biology, 24,1927-1938. 5746 Overgaard, J. & Sorenson, J.G. (2008) Rapid thermal adaptation during field temperature 5747 variations in Drosophila melanogaster. Cryobiology, 56, 159–162. 5748 Piyaphongkul, J. (2013) Effects of thermal stress on brown plant hopper. Phd Thesis. University 5749 of Birmingham. Birmingham. 5750 Rajamohan, A. & Sinclair B.J. (2008) Short term hardening effects on survival of acute and 5751 chronic cold exposure by Drosophila melanogaster larvae. Journal of Insect Physiology, 54, 5752 708-718. 5753 Sejerkilde, M., Sorensen, J.G. & Loeschcke, V. (2003) Effects of cold and heat hardening on 5754 thermal resistance in Drosophila melanogaster. Journal of Insect Physiology, 49, 719-726. 5755 Selvaraj, S., Ganeshamoorthi, P. & Pandiaraj, T. (2013). Potential Impacts on recent climate 5756 change on biological control agents in agroecosystem: A review. International Journal on 5757 Biodiversity and Conservation, 5, 845-852.

228

5758 Sgrò, C.M., Terblanche, J.S. and Hoffmann, A.A. (2016) What can plasticity contribute to insect 5759 responses to climate change? Annual Review of Entomology, 61, 433-451. 5760 Sinclair, B.J., Coello Alvarado, L.E. & Ferguson, L.V. (2015) An invitation to measure insect 5761 cold tolerance: Methods, approaches, and workflow. Journal of Thermal Biology, 53, 180- 5762 197. 5763 Stathers, T., Lamboll. R. & Mvumi, B.M. (2013) Postharvest agriculture in changing climates: 5764 its importance to African smallholder farmers. Food Security, 5, 361-392. 5765 Stotter, R.L. & Terblanche, J.S. (2009) Low temperature tolerance of false codling moth 5766 Thaumotibia leucotreta (Meyrick) (Lepidoptera: Tortricidae) in South Africa. Journal of 5767 Thermal Biology, 34, 320–325. 5768 Tammarielo, S.T., Rinehart, J.P., & Denlinger, D. L. (1999) Desiccation elicits heat shock 5769 protein transcription in the flesh fly, Sacophaga crassipalpis, but does not enhance tolerance 5770 to high or low temperatures. Journal of Insect Physiology, 45, 933-938. 5771 Terblanche, J.S., Deere, J.A., Clusella-Trullas, S., Janion, C. and Chown, S.L. (2007) Critical 5772 thermal limits depend on methodological context. Proceedings of the Royal Society of 5773 London B, 274, 2935-2942. 5774 Terblanche, J.S., Klok, C.J., Krafsur, E.S. & Chown, S.L. (2006) Phenotypic plasticity and 5775 geographic variation in thermal tolerance and water loss of the tsetse Glossina pallidipes 5776 (Diptera: Glossinidae): implications for distribution modelling. The American Society of 5777 Tropical Medicine and Hygiene, 74, 786–794. 5778 Voigt, W., Perner, J., Davis, A.J., Eggers, T., Schumacher, J., Bährmann, R., Fabian, B., 5779 Heinrich, W., Köhler, G., Lichter, D., Marstaller, R. & Sander, F.W. (2003) Trophic levels are 5780 differentially sensitive to climate. Ecology, 84, 2444-2453. 5781 Waagner, D., Holmstrup, M., Bayley, M. & Sørensen, J.G. (2013) Induced cold-tolerance 5782 mechanisms depend on duration of acclimation in the chill sensitive Folsomia candida 5783 (Collembola). The Journal of Experimental Biology, 216, 1991-2000. 5784 5785

5786

5787

229

5788

5789

5790 APPENDIX 1

5791

5792

5793

5794

5795

5796

5797 Diversity and abundance of lepidopteran stem borer natural enemies in natural and 5798 cultivated habitats in Botswana*

5799

5800

5801

5802

5803

5804

5805

5806 * Published as “Mutamiswa, R., Moeng, E., Le Ru, B.P., Conlong, D.E., Assefa, Y., Goftishu, 5807 M. & Nyamukondiwa, C. (2017) Diversity and abundance of lepidopteran stem borer natural 5808 enemies in natural and cultivated habitats in Botswana. Biological Control, 115, 1-11.

230

5809 1.1 Introduction 5810 Cereal crops such as maize, sorghum and millet are mostly grown by small-scale farmers 5811 at subsistence level in sub-Saharan Africa, with the majority of cereal fields usually surrounded 5812 by patches of natural habitats harbouring stemborer wild host plants (e.g. grasses and sedges) 5813 (Mailafiya et al., 2009). These natural habitats often serve as refugia for stemborers and their 5814 parasitoids, sustaining a diversity of stemborer natural enemies within the agroecosystem 5815 (Mailafiya et al., 2009; Moolman et al., 2013). Lepidopteran stem borers are among the most 5816 destructive insect pests of cereal crops in sub-Saharan Africa, accounting for 5-75% of yield 5817 losses (De Groote, 2002; Kfir et al., 2002; Kipkoech et al., 2006; Moolman et al., 2013). In 5818 Africa, this represents a significant household food insecurity burden since ~70-80% of the 5819 population depends on subsistence agriculture (FAO, 2002). The major lepidopteran cereal 5820 stemborer species accounting for damage in Africa include the indigenous pyralid Eldana 5821 saccharina Walker, the crambid Chilo orichalcociliellus (Strand), the noctuids Busseola fusca 5822 (Fuller) and Sesamia calamistis Hampson and the exotic crambid Chilo partellus Swinhoe (Kfir 5823 et al., 2002; Obonyo et al., 2010; Addo-Bediako & Thanguane, 2012). 5824 The abundance and distribution of cereal stemborers may be influenced by abiotic factors 5825 such as temperature and precipitation, and biotic factors such as natural enemies and alternative 5826 host plants (Mailafiya et al., 2009; Addo-Bediako & Thanguane, 2012). Both parasitoid diversity 5827 and parasitism levels have been reported as higher in arable fields intermixed with natural 5828 habitats than in arable fields only (Thies et al., 2003), indicating the role of natural habitats in 5829 hosting parasitoid diversity. Although natural habitats have been reported to provide refuges for 5830 some parasitoid species, low levels of stemborer parasitism (˂8%) have been recorded in East 5831 and Austral Africa in wild host plants (Mailafiya et al., 2011; Moolman et al., 2013). 5832 Gramineous plants are known to produce secondary metabolites that recruit stemborer 5833 parasitoids (Potting et al., 1995; Arab & Bento, 2006). Indeed, this is the case for maize and 5834 sorghum plants infested with B. fusca and C. partellus (Mutyambai et al., 2015). Isolated 5835 volatiles from these plants are effective in recruiting the braconid parasitoids Cotesia sesamiae 5836 Cameron and Cotesia flavipes Cameron, and the ichneumonid Dentichasmias busseolae Heinrich 5837 (Arab & Bento, 2006). 5838 An understanding of the diversity, spatial distribution, abundance and ecological 5839 contributions of stemborer natural enemies in both natural and managed ecosystems is required

231

5840 to implement an integrated approach to stemborer pest management (Conlong & Rutherford, 5841 2009). Natural enemies play an important role in the control of lepidopteran stemborers (Sylvain 5842 et al., 2015; Midingoyi et al., 2016) and help prevent pest outbreaks (Muyekho et al., 2005; 5843 Mailafiya et al., 2010a, 2013; Kodjo et al., 2013). In eastern and southern Africa, cereal 5844 stemborer populations are regulated by a wide range of native and exotic parasitoid species 5845 (Conlong, 2000; Zhou et al., 2003; Moolman et al., 2013). Most of these parasitoids belong to 5846 the orders Hymenoptera and Diptera (Polaszek, 1998; Conlong, 2000; Zhou et al., 2003; 5847 Moolman et al., 2013). There have been numerous attempts over the past decades to introduce 5848 exotic parasitoids into Africa for biological control of indigenous and exotic stemborers, but only 5849 a few species were established successfully (Conlong, 1994, 1997, 2001; Kfir et al., 2002). 5850 Nevertheless, C. flavipes, a C. partellus larval endoparasitoid, was introduced from Pakistan into 5851 Kenya in 1993 where it became established (Overholt et al., 1997; Kfir et al., 2002; Kipkoech et 5852 al., 2006; Dejen et al., 2013), and releases have subsequently been successful in nine other 5853 countries in east and southern Africa (Kipkoech et al., 2006; Omwega et al., 2006; Chinwada et 5854 al., 2008). Different life stages of stemborers are attacked by predators such as ants, spiders, 5855 earwigs, nematodes and microbial pathogens [Beauveria bassiana (Balsamo), Bacillus 5856 thuringiensis Berliner, Serratia marcescens Bizio] in various landscapes (Oloo, 1989; Bonhof et 5857 al., 1997; Assefa et al., 2006). However, data also have shown that indigenous natural enemies 5858 may be unable to keep stemborer populations below economic levels (Overholt et al., 1994; 5859 Bonhof et al., 1997), unless their potential efficacy can be improved through conservation. 5860 The diversity and abundance of stemborer natural enemies both vary across the landscape 5861 and the distribution of some may be threatened by global climate change (Batalden et al., 2007; 5862 Kodjo et al., 2013). Previous studies have documented lepidopteran stemborers, their parasitoids, 5863 and associated host plants in many African countries, including Ethiopia (Assefa et al., 2006), 5864 Uganda (Conlong and Mugalula, 2001; Matama-Kauma et al., 2007; 2008), Kenya (Mailafiya et 5865 al., 2009), DR Congo (Kankonda et al., 2017), Cameroon (Ndemah et al., 2007), Togo (Kodjo et 5866 al., 2013), Zimbabwe (Chinwada & Overholt, 2001; Mazodze & Conlong, 2003), Mozambique 5867 (Conlong & Goebel, 2002, 2006; Moolman et al., 2013) and South Africa (Moolman et al., 5868 2013). Apart from a survey of Cyperus papyrus L. for parasitoids of E. saccharina (Conlong et 5869 al., 1988), similar information has been lacking for Botswana. Through extensive surveys, I 5870 assessed the diversity and abundance of stemborer natural enemies and their associated hosts in

232

5871 both cultivated and natural habitats in Botswana. Such information is crucial in understanding 5872 the roles of natural enemies in regulating insect pest populations, and may help develop 5873 improved pest management systems. Following the results of Mailafiya et al. (2009), I 5874 hypothesized that natural enemy species diversity would be higher in natural compared to 5875 cultivated habitats in Botswana. 5876

5877 1.2 Materials and methods 5878

5879 1.2.1. Field surveys 5880 Field surveys in cereal crops and wild host plants were conducted in all ten districts of 5881 Botswana during the austral summers of 2014/15 and 2015/16 (Figure 1.1). Sampling was 5882 conducted once in each district during both seasons, for a total of 84 sites sampled. 5883

233

5884 A

5885 5886

B C A A

5887 5888 5889 Figure 1.1: Botswana maps showing (A) areas surveyed for stemborers in cultivated and natural 5890 habitats during 2014/15 and 2015/16 austral summer (B) average annual temperatures and (C) 5891 average annual rainfall for the sampled districts. 5892 5893

234

5894 1.2.1.1. Sampling in natural habitats 5895 Wild lepidopteran stemborer host plants belonging to the families Poaceae, Cyperaceae 5896 and Typhaceae were sampled (i) around cereal crops, (ii) in open patches along forest roads, (iii) 5897 on banks of streams or rivers (iv) in swamps and, (v) in national parks. A biased sampling 5898 procedure was employed to increase the chances of finding stemborers, as previous studies 5899 reported considerably lower densities of borers in wild host plants than in cultivated crops 5900 (Gounou & Schulthess, 2004; Le Ru et al., 2006; Mailafiya et al., 2009; Moolman et al., 2013; 5901 2014). Potential host plants were inspected for stemborer damage symptoms such as dry leaves 5902 and shoots (dead hearts), scarified leaves (window panes and pin/shot holes), larval entry and 5903 exit holes in stems, and the presence of frass. Stems that exhibited damage symptoms were 5904 dissected in the field and any borer life stages found within were placed individually in 30 ml 5905 plastic vials (South African Sugarcane Research Institute, South Africa) containing artificial diet 5906 and sealed with aerated screw-caps. The diet for noctuids was that formulated by Onyango and 5907 Ochieng’-Odero (1994), and that for pyralids and crambids, that of Walton & Conlong (2016). 5908 Predatory insects, parasitoid cocoons or dead larvae were placed individually in labelled plastic 5909 vials without diet, while pupae were placed in the same vials with a piece of wet cotton wool to 5910 prevent desiccation. 5911

5912 1.2.1.2. Sampling in cultivated habitats 5913 To survey cereal crops, fields of sweet sorghum, maize, millet and sorghum were selected 5914 every 10–15 km along minor and major roads in each district 10-15 fields per village in each 5915 district. Each field was divided into four quadrats and 10 plants per quadrat were randomly 5916 sampled along a transect. Each plant was dissected and searched for stemborer and parasitoid life 5917 stages which were collected as described above for wild host plants. 5918 5919 For each vial, I recorded (1) the date of collection, (2) name of host plant (if it could be identified 5920 in the field), (3) stemborer developmental stage and (4) borer and/or parasitoid species (if it 5921 could be identified in the field). Latitude, longitude and altitude were also recorded for each field 5922 using a portable Geographic Positioning System (GPS) (Garmin GPSMAP 62st). Following 5923 collection, all vials were placed in coolers and taken to the laboratory for subsequent incubation. 5924

235

5925 1.2.2. Laboratory incubation 5926 In the laboratory, all insects were incubated in climate-controlled chambers (HPP 260, 5927 Memmert GmbH + Co.KG, Germany) at 25 ± 1 oC, 65 ± 10% RH, and a 12:12 (L:D) 5928 photoperiod until adult moth or parasitoid emergence. Predatory insects and emerged adult 5929 parasitoids were preserved in 70% alcohol for subsequent identification. Adult stemborers and 5930 parasitoids were identified to family by gross morphology, and to genus or species level by the 5931 third author in the biosystematics unit of African Insect Science for Food and Health (ICIPE), 5932 Kenya. This was done by examining external morphology and genitalia as described in Polaszek 5933 (1998). When identification to species level was not possible, genus and family levels were used. 5934 Wild host plant species were identified at the department of National Museum and Monuments, 5935 Herbarium unit, Botswana. 5936

5937 1.2.3. Data analysis 5938 Host plant - parasitoid and stemborer - parasitoid interaction networks were analyzed in 5939 R using ‘BIPARTITE’ package (Dormann et al., 2009). In all interaction network analyses, the 5940 number of parasitized hosts against host plants and stemborer species in all habitats were used. 5941 Parasitoid species diversity in natural and cultivated habitats was computed using Simpson’s 5942 diversity index (Simpson, 1949):

∑ n(n − 1) 퐷 = N(N − 1)

5943 Where D is the Simpson diversity index, N is the total number of organisms found and n is the 5944 number of individuals of each species. Percentage parasitism of (1) larvae and (2) pupae and, (3) 5945 the relative abundance of each parasitoid species were calculated as outlined by Zhou et al 5946 (2003), Dejen et al. (2013), Rahaman et al. (2014) and Baskaran et al. (2017) as follows:

Number of parasitized larvae X 100 5947 (1) 퐿푎푟푣푎푙 푝푎푟푎푠𝑖푡𝑖푠푚 = Number of healthy larvae + Number of parasitized larvae 5948 Number of parasitized pupae X 100 5949 (2) 푃푢푝푎푙 푝푎푟푎푠𝑖푡𝑖푠푚 = Number of healthy pupae + Number of parasitized pupae 5950

236

Total of individuals of each sp X 100 5951 (3) 푅푒푙푎푡𝑖푣푒 푎푏푢푛푑푎푛푐푒 표푓 푒푎푐ℎ 푝푎푟푎푠𝑖푡표𝑖푑 푠푝 = Total number of all parasitoids 5952

5953 1.3 Results 5954

5955 1.3.1. Natural enemy diversity, stemborer hosts and host plants 5956 A total of 1497 stemborer larvae and 228 pupae were collected from three different 5957 cultivated plants and 11 different wild host plants belonging to the families Poaceae, Cyperaceae 5958 and Typhaceae (Table 1.2; 1.3; Figure 1.2B). From these collections, 30 natural enemy species 5959 (27 parasitoids and 3 predators) belonging to 13 insect families and two orders were recovered. 5960 Parasitoid diversity was generally higher in cultivated than in natural habitats (D = 0.23 versus 5961 0.14; Table 1.1). Nineteen of the 30 parasitoid species (16 larval, three pupal) were recovered 5962 from three stemborer species in cultivated hosts, whereas 11 species (nine larval, two pupal) 5963 were recovered from seven stemborer species in wild hosts. Furthermore, three predator species 5964 were recovered from both cultivated and natural habitats. The hymenopteran parasitoids 5965 appeared dominant to the dipteran parasitoids in all the sampled districts, with Braconidae and 5966 Tachinidae the most common families represented in the two orders, respectively. The stemborer 5967 – parasitoid interaction network showed a moderate Shannon interaction diversity (H' = 2.56)

5968 and a low generality (Gqw =1.23). Similarly, the host plant – parasitoid interaction showed

5969 moderate Shannon interaction diversity (H' = 2.93) and generality (Gqw= 1.81). A total of 19 5970 different parasitoid species were recovered from C. partellus in both cultivated and natural 5971 habitats (Figure 1.2A). Eleven and 10 parasitoid species were recovered from maize and sweet 5972 sorghum, respectively, with four species parasitizing stemborers on S. corymbosus in natural 5973 habitats (Figure 1.2B). 5974 The most common larval parasitoids were the gregarious species C. flavipes and C. 5975 sesamiae, plus the solitary C. curvimaculatus, all recovered from C. partellus (Figure 1.2A). The 5976 main pupal parasitoids included the gregarious P. furvus and the solitary Gambroides 5977 nimbipennis Seyrig (Hymenoptera: Ichneumonidae), also recovered from C. partellus. The larval 5978 parasitoids G. indicus, C. curvimaculatus and the pupal parasitoid G. nimbipennis were 5979 recovered from both cultivated and wild host plants infested with C. partellus, Sesamia

237

5980 nonagrioides Lefèbvre, and S. jansei (Table 1.1). Nevertheless, the exotic C. flavipes and the 5981 indigenous C. sesamiae were more specific to C. partellus in cultivated habitats (Table 1.1; 5982 Figure 1.2A). The stemborers, C. partellus and S. calamistis were most common on the 5983 cultivated host plants, whereas E. saccharina, S. jansei and S. nonagrioides were more common 5984 on wild host plants. As for predators, L. humile, C. peringueyi and Aenictus sp were common on 5985 both cultivated and wild host plants where they fed on larvae and pupae of E. saccharina, C. 5986 partellus S. jansei and S. calamistis within their larval galleries (Table 1.1). Hymenopteran 5987 parasitoids were recovered from borers on Poaceae, Cyperaceae and Typhaceae, whereas 5988 Dipterans were recovered from borers on plants of the Poaceae and Cyperaceae only (Table 1.2; 5989 1.3). In cultivated habitats, maize and sorghum were the most common host plants, whereas S. 5990 corymbosus was the most common stemborer host in natural habitats (Table 1.2; 1.3; Figure 5991 1.2B). 5992

238

5993 Table 1.1: Details of natural enemy species recovered from stemborers on cultivated and wild host plants in Botswana. 5994 Location Coordinates Parasitoid species Family Habitat Host Host Parasitism stage Strategy Hymenoptera Chanoga S20.17703; E23.67496; 939m Pediobius furvus Gahan Eulophidae C Cp P G Mogobane S24.94659; E25.73672; 1048m Modipane S24.61832; E26.15096; 1020m Tsootsha S22.11460; E20.88564; 1190m Oodi S24.54710; E26.00949; 966m Euvipio rufa Szepligeti Braconidae C Cp L S Mabele S17.98124; E24.64453; 935m C Con L S Maun S19.99686; E23.43182; 937m Gambroides nimbipennis Ichneumonidae N Tot P G Seyrig Mabele S17.98124; E24.64453; 935m C Cp P G Goniozus indicus Ashmead Bethylidae C Cp L G Maun S19.99686; E23.43182; 937m N Sj L G Kazungula S17.80556; E25.24944; 937m N Sn L G Oodi S24.54710; E26.00949; 966m Cotesia flavipes Cameron Braconidae C Cp L G Seleka S22.95112; E27.96191; 783m Bobonong S21.98344; E28.44528; 707m Ramatlabama S25.60480; E25.56388; 1301m Mabele S17.98124; E24.64453; 935m Glen valley S24.59256; E25.97819; 960m Ramotswa S24.82314; E25.86776; 1003m Cotesia sesamiae Cameron Braconidae C Cp L G

239

Tlokweng S24.62219; E25.97400; 945m Hatsalatladi S24.04468; E25.62869; 1107m Dithojane S22.45031; E26.76353; 1076m Chelonus curvimaculatus Braconidae C Cp L S Cameron Moshupa S24.78787; E25.41477; 1096m N Cp L S Maun S20.00484; E23.42598; 924m N Sj L S Gaborone S24.70752; E25.90632; 992m N Sj, Cp L S Dam Glen valley S24.59256; E25.97819; 960m Dolichogenidea polaszeki Braconidae C Cp L S Walker Qabo S21.05622; E21.74792; 1082m Stenobracon rufus Szepligeti Braconidae C Cp P S Lobatse S25.25092; E25.70840; 1200m Temelucha picta Holmgren Ichneumonidae C Cp L S Chanoga S20.17411; E23.69703; 935m Chalcidoidea sp. Encrytidae C Cp L S Bookane S23.06528; E27.53493; 881m Ceraphronidae sp. Ceraphronidae C Cp L S Ditladi S21.47007; E27.52998; 952m Eichneumonidea sp. Braconidae C Sc, Cp L S Mabutsane S24.44993; E23.58053; 1091m Rhaconotus sp. Braconidae C Cp L G Norbanus sp. Pteromalidae C Sc, Cp L G Kumakwane S24.72081; E25.51619; 1059m Linepithema humile Mayr Formicidae C Cp L G Ndumba S17.97402; E24.44734; 932m Parakarungu S18.03961; E24.29545; 938m N Es P G Maun S20.00484; E23.42598; 924m N Es L G Nxamaseri S18.60587; E22.08834; 1008m N Es L G Kasane, S24.61832; E26.15096; 1020m Aenictus sp. Formicidae C Sc L G Mochudi S17.98124; E24.64452; 935m C Cp P G

240

Matlapana S 19.94073; E 23.50739; 937m Martins Drift S 22.99681; E 27.94028; 783m N Es L G Hatsalatladi S24.04468; E25.62869; 1107m Crematogaster peringueyi Formicidae C Sc L G Emery Ramotswa S24.82314; E25.86776; 1003m Maun S20.00484; E23.42598; 924m N Sj L G Kasane S17.793325; E25.18863; 944m Bracon testaceorufatus Braconidae N Sc L G Granger Kanye S24.94807; E25.34476; 1239m Dentichasmias busseolae Ichneumonidae N Tot P S Heinrich Kasane S17.79203; E25.19751; 930m Cremastinae sp. Ichneumonidae N Sc L S Maun S20.00484; E23.42598; 924m Cercerinae sp. Philantenidae N Sj L S Nxamaseri S18.60587; E22.08834; 1008m Formicidae sp. Formicidae N Es L S Samedupi S 20.11207; E 23.52979; 935m Syzeuctus sp. Ichneumonidae N Tot, Sj L S Diptera Hatsalatladi S24.04468; E25.62869; 1107m Descampsina sesamiae Tachinidae C Cp L G Mesnil Qabo S21.05622; E21.74792; 1082m Mabutsane S24.44993; E23.58053; 1091m Sturmiopsis parasitica Tachinidae C Cp L S/G Curran Oodi S24.54710; E26.00949: 966m Sepsidae sp. Sepsidae C Cp L G Hatsalatladi S24.04468; E25.62869; 1107m Chloropidae sp. Chloropidae C Cp L S Gatsatsa S25.26881; E25.27763; 1254m Satau S18.09805; E24.43839; 944m Siphona murina Mesnil Tachinidae N Sn, Sp L G

241

Shakawe S 18.35927; E 21.84229; 956m Phoridae sp Phoridae N Es L G 5995 5996 C, cultivated; N, natural; Cp, Chilo partellus; Con, Concofrontia sp.; Sj, Sesamia jansei; Sn, Sesamia nonagrioides, Tot, Totricidae sp.; Sc, 5997 Sesamia calamistis; Es, Eldana saccharina; Sp, Sesamia perplexa; L, larva; P, pupa; G, gregarious; S, solitary

5998

242

5999 Table 1.2: Natural enemies and their associated cultivated host plants in Botswana 6000 Natural Enemy Species Crop plant (Poaceae)

Maize Sorghum Sweet sorghum

Bethylidae Goniozus indicus X Braconidae Euvipio rufa X X Cotesia flavipes X X X Cotesia sesamiae X X X Chelonus curvimaculatus X Dolichogenidea polaszeki X Stenobracon rufus X Eichneumonidea sp. X Bracon testaceorufatus Rhaconotus sp. X Ceraphronidae Ceraphronidae X Chloropidae Chloropidae sp. X Encrytidae Chalcidoidea sp. X Eulophidae Pediobius furvus X X

243

Formicidae Aenictus sp. X X Linepithema humile X X Crematogaster peringueyi X X Formicidae sp. Ichneumonidae Telemucha picta X Gambroides nimbipennis X Cremastinae sp. Dentichasmias busseolae Syzeuctus sp. Philantenidae Cercerinae sp. Phoridae Phoridae sp. Pteromalidae Norbanus sp. X Sepsidae Sepsidae sp. X Tachnidae Descampsina sesamiae X X Sturmiopsis parasitica X Siphona murina 6001

6002 244

6003 Table 1.3: Natural enemies and their associated wild host plants in Botswana 6004 Natural Enemy Species Poaceae Cyperaceae Typhaceae

auricomus

subsp

Oryza Oryza

nigritanus

pyramidalis corymbosus

Echinochloa Echinochloa

Chrysopogon

Cyperus dives Cyperus

arundinaceum

Typha latifolia

longistaminata

Schoenoplectus Schoenoplectus

Hyparrhenia

subsp

Sorghum bicolor

Vossia cuspidata Vossia papyrus Cyperus

Cyperus digitatus digitatus Cyperus Bethylidae Goniozus indicus X X Braconidae Euvipio rufa Cotesia flavipes Cotesia sesamiae Chelonus curvimaculatus X Dolichogenidea polaszeki Stenobracon rufus Eichneumonidea sp Bracon testaceorufatus X Rhaconotus sp. Ceraphronidae Ceraphronidae Chloropidae Chloropidae sp.

245

Encrytidae Chalcidoidea sp. Eulophidae Pediobius furvus Formicidae Aenictus sp. X Linepithema humile X X Crematogaster peringueyi X X Formicidae sp. X Ichneumonidae Telemucha picta Gambroides nimbipennis X X Cremastinae sp. X Dentichasmias busseolae X Syzeuctus sp. X X Philantenidae Cercerinae sp. X Phoridae Phoridae sp. X Pteromalidae Norbanus sp. Sepsidae Sepsidae sp.

246

Tachnidae Descampsina sesamiae Sturmiopsis parasitica Siphona murina X X 6005

247

6006

6007 6008 Figure 1.2: Quantitative bipartite interaction networks for (A) stemborer - parasitoid and (B) 6009 host plant - parasitoid in cultivated and natural habitats in Botswana

248

6010 1.3.2. Stemborer parasitism in cultivated and wild host plants 6011 A higher level of parasitism was recorded in cultivated habitats compared to natural 6012 habitats (Table 1.4; 1.5). In cultivated habitats, the parasitoids C. flavipes and C. sesamiae had 6013 the highest levels of larval parasitism, whereas G. nimbipennis had the highest levels of pupal 6014 parasitism. In natural habitats, C. curvimaculatus had the highest larval parasitism, whereas D. 6015 busseolae had highest pupal parasitism. 6016 6017

249

6018 Table 1.4: Larval and pupal parasitism in cultivated host plants in Botswana. 6019 Host Plant Species Host species Parasitoid species % Parasitism Maize Chilo partellus Cotesia flavipes 34.7 (33) Cotesia sesamiae 18.8 (13) Euvipio rufa 4.3 (2) Sepsidae sp. 2.1 (1) Pediobius furvus 8.5 (5) Descampsina sesamiae 5.9 (1) Chalcidoidea sp. 7.7 (1) Telemucha picta 2.9 (1) Dolichogenidea polaszeki 6.3 (1) Stenobracon rufus 6.1 (2) Sorghum Chilo partellus Cotesia flavipes 5.7 (2) Cotesia sesamiae 12.5 (2) Chloropidae sp. 4.5 (2) Descampsina sesamiae 6.3 (1) Ceraphonidae sp. 7.1 (1) Sweet sorghum Concofrontia sp Euvipio rufa 5.9 (1) Sesamia calamistis Echneumonidea sp. 2.8 (1) Chilo partellus Cotesia flavipes 17.6 (3) Cotesia sesamiae 29 (9) Goniozus indicus 5.9 (1) Rhaconotus sp. 4.3 (1) Norbanus sp. 4.3 (1) Sturmiopsis parasitica 4.3 (1) Chelonus curvimaculatus 9.1 (1) Echneumonidea sp. 3.2 (1) Gambroides nimbipennis 10.6 (3) 6020 6021 ( ) number of parasitized hosts 6022

250

6023 Table 1.5: Larval and pupal parasitism in wild host plants in Botswana. 6024 Host Plant Species Host species Parasitoid species % Parasitism Sorghum bicolor subsp Sesamia calamistis Bracon testaceorufatus 4 (1) arundinaceum Sesamia nonagrioides Siphona murina 3.3 (1) Schoenoplectus Sesamia jansei Goniozus indicus 4.8 (1) corymbosus Cercerinae sp. 9.5 (2) Tortricidae sp Chelonus curvimaculatus 14.3 (3) Dentichasmias busseolae 9.1 (2) Cyperus dives Chilo partellus Chelonus curvimaculatus 7.8 (5) Oryza longistaminata Sesamia nonagrioides Goniozus indicus 3.5 (1) Typha latifolia Sesamia jansei Gambroides nimbipennis 6.7 (1) Tortricidae sp Syzeuctus sp. 4.2 (1) Chrysopogon Sesamia calamistis Cremastinae sp. 4 (1) nigritanus Cyperus papyrus Eldana saccharina Formicidae sp. 3.8 (1) Vossia cuspidata Sesamia perplexa Siphona murina 4.5 (1) Echinochloa Sesamia jansei Syzeuctus sp. 5.2 (1) pyramidalis Cyperus papyrus Eldana saccharina Phoridae sp. 3.7 (1) 6025 6026 ( ) number of parasitized hosts 6027

6028 1.3.3 Relative abundance of parasitoid species 6029 Hymenopteran parasitoids were more abundant than Dipteran parasitoids in both kinds of 6030 habitats (Figure 1.3A; B). The larval parasitoids C. flavipes and C. sesamiae were the most 6031 abundant in cultivated habitats (Figure 1.3A), whereas C. curvimaculatus dominated in natural 6032 habitats (Figure 1.3B). Similarly, among pupal parasitoids, the hymenopteran P. furvus was the 6033 most abundant in cultivated habitats, whereas D. busseolae dominated in natural habitats.

251

6034

6035 6036 6037 Figure 1.3: Relative abundance of cereal stemborer parasitoids in (A) cultivated and (B) natural 6038 habitats in Botswana.

252

6039 1.4 Discussion 6040

6041 1.4.1. Natural enemy diversity, stemborer hosts and host plants 6042 I collected 30 natural enemy species attacking immature stages of lepidopteran 6043 stemborers (27 parasitoids and three predators) in both cultivated and natural habitats in 6044 Botswana. Previously, only nine parasitoid species had been reported on sorghum, maize and 6045 Cyperus papyrus L. (Conlong et al., 1988; Obopile & Mosinkie, 2003) and no predatory species 6046 had been reported attacking stemborers in the country. In the current study, the predators L. 6047 humile, C. peringueyi and Aenictus sp. were all observed on both cultivated and wild host plants. 6048 Conlong et al. (1988) reported the presence of Goniozus natalensis Gordh (= G. indicus) on C. 6049 papyrus along the Chobe river in Botswana, which is consistent with my collection of this 6050 parasitoid in natural habitats. Similarly, Obopile & Mosinkie (2003) reported the presence of C. 6051 sesamiae, P. furvus and S. rufus, which is in agreement with my observations (Table 1.1). 6052 My results showed a higher parasitoid diversity in cultivated than natural habitats, which is 6053 contrary to previous surveys in Cameroon (Ndemah et al., 2007), Uganda (Matama-Kauma et 6054 al., 2008) and Kenya (Mailafiya et al., 2009) that all reported a higher parasitoid diversity in 6055 natural habitats. However, my results are consistent with a previous study by Zhou et al. (2003) 6056 that reported a higher parasitoid diversity in cultivated habitats in the coastal region of Kenya. 6057 The current study also recorded a moderate Shannon interaction diversity and generality for both 6058 stemborer–parasitoid and host plant– parasitoid interaction, indicating that, on average, each 6059 stemborer species was attacked by one to two parasitoid species (Helena et al., 2016). This also 6060 implies that every host plant hosted stemborers that were attacked by one to two parasitoid 6061 species. This niche overlap among parasitoids may serve to improve biological control efficacy, 6062 assuming that increased interspecific competition does not compromise the impact of key 6063 species. 6064 Chilo partellus was the stemborer most commonly attacked by parasitoids and predators 6065 in cultivated habitats, whereas Sesamia species were most commonly attacked in natural habitats, 6066 which may reflect the high abundance of these species in these respective habitats. The 6067 sugarcane stemborer, E. saccharina, was the least attacked by natural enemies in natural habitats, 6068 possibly because the larvae spin a lot of silk around themselves and the walls of their galleries 6069 and secrete exudates upon disturbance and may be repellent to natural enemies (Sampson &

253

6070 Kumar, 1986; Hordzi & Botchey, 2012). The larval parasitoids C. flavipes and C. sesamiae and 6071 pupal parasitoid P. furvus were the most prevalent species, in keeping with previous results from 6072 eastern Africa (Zhou et al., 2003; Mailafiya et al., 2009). Whereas C. sesamiae is considered a 6073 generalist parasitoid (Branca et al., 2011), C. flavipes is more specific to C. partellus (Mailafiya 6074 et al., 2010b), but the abundance of these two parasitoids on C. partellus in cultivated habitats 6075 may reflect the high availability of C. partellus as a host. Although C. partellus was recovered 6076 from Cyperus dives, C. flavipes was not recorded in natural habitats in the current study, similar 6077 to the results by Moolman et al. (2013). The gregarious larval parasitoid Bracon testaceorufatus 6078 Granger was recorded parasitizing S. calamistis in the natural habitats and is known to attack rice 6079 stemborers (van Achterberg & Walker, 1998). In South Africa, Moolman et al. (2013) recovered 6080 this parasitoid species from S. jansei suggesting that the parasitoid may have a broader stemborer 6081 host range. The Cyperaceous host plant S. corymbosus had the highest parasitoid diversity in 6082 natural habitats, in keeping with results by Moolman et al. (2013), whereas maize and sorghum 6083 had the highest parasitoid diversity in cultivated habitats. This may be because cultivated plants 6084 have higher levels of nutrients for stemborers, and reduced allelochemical toxicity, relative to 6085 wild host plants (Ofomata et al., 2000; Mailafiya et al., 2009). These attributes could promote 6086 higher stemborer survival and thus greater parasitoid richness in cultivated relative to wild hosts. 6087 To my knowledge, this is the first report of C. curvimaculatus parasitizing C. partellus on C. 6088 dives. 6089

6090 1.4.2 Parasitism levels in cultivated and wild host plants 6091 Larval and pupal parasitism in cultivated habitats were in keeping with previous reports 6092 from East Africa for both life stages (Matama-Kauma et al., 2008; Mailafiya et al., 2009; Zhou 6093 et al., 2003), and parasitism levels were higher in cultivated than in natural habitats, in 6094 agreement with the results of Mailafiya (2011). I did not measure stemborer densities across 6095 habitats, but low levels of parasitism in natural habitats probably reflect lower densities of 6096 stemborers relative to cultivated habitats (Mailafiya et al., 2009; Moolman et al., 2014) which 6097 would have negative effects on parasitoid searching efficiency (Udayagiri & Welter, 2000). 6098 Indeed, the observed parasitoid diversity may depend on complex interactions between the host 6099 plants and related stemborer species. Obopile & Mosinkie (2003) reported C. partellus 6100 parasitism levels reaching 13.3% for C. sesamiae, 10.3% for P. furvus and 3.2% for S. rufus on

254

6101 maize and sorghum in Botswana, whereas the current study found generally higher C. partellus 6102 parasitism levels in cultivated habitats. Again, this may be due to higher densities of C. partellus 6103 relative to other stemborer species in these habitats. Cotesia flavipes showed the highest levels of 6104 larval parasitism in cultivated habitats, which corroborated the results of Sétamou et al. (2005). 6105 The gregarious C. flavipes has demonstrated high searching capacity that permits it to exploit 6106 habitats with low stemborer densities (Wiedenmann & Smith, 1993; Mailafiya et al., 2009). 6107 Although the exotic C. flavipes has not been released in Botswana, it has been released in South 6108 Africa, (Skoroszewski & Van Hamburg, 1987), Mozambique, (Cugala & Omwega, 2001), 6109 Malawi, Zambia (Sohati et al., 2001) and Zimbabwe (Chinwada et al., 2001). Since these 6110 countries border Botswana (except Malawi), releases in these neighboring countries probably 6111 account for the presence of this exotic species in Botswana. The parasitoid has been reported to 6112 disperse >2000km away from release sites (Assefa et al., 2008), suggesting high potential for 6113 migration, spread and establishment in novel environments. Gambroides nimbipennis showed 6114 higher pupal parasitism levels in cultivated habitats relative to natural habitats, which is probably 6115 also indicative of high host density and resource availability in cultivated habitats. 6116

6117 1.4.3. Relative abundance of parasitoid species 6118 Compared to other parasitoid species, the larval parasitoids C. flavipes and C. sesamiae 6119 and the pupal parasitoid P. furvus were higher in relative abundance in cultivated habitats, 6120 whereas the larval parasitoid C. curvimaculatus and the pupal parasitoid D. busseolae had higher 6121 relative abundance in natural habitats. Previous studies have reported ecological similarities 6122 between C. sesamiae and C. flavipes that suggest niche overlaps between these two parasitoids. 6123 Indeed, competition between the two congeners has resulted in the displacement of C. sesamiae 6124 in some regions (see Ngi-song & Overholt, 1997; Sallam et al., 1999; Zhou et al., 2003). 6125 However, in the current study, the two parasitoid species dominated in both abundance and 6126 parasitism of the same host, C. partellus. This may indicate that C. flavipes and C. sesamiae tend 6127 to share the same habitat, regardless of whether they attack the same host or not (Overholt et al., 6128 2000). Consistent with this inference, Mailafiya et al. (2011) reported a significant interaction 6129 between C. flavipes and C. sesamiae in regulating cereal stemborer populations in cultivated 6130 habitats. The presence and dominance of C. partellus in most districts in Botswana makes C. 6131 flavipes a suitable exotic larval parasitoid for augmentative field releases (Overholt et al., 2000).

255

6132 Chelonus curvimaculatus was recovered in both habitat types, suggesting that this species is a 6133 generalist that can switch habitats, and could potentially be used in augmentative biological 6134 control with C. flavipes and C. sesamiae, assuming they have a synergistic effect collectively. 6135 Since P. furvus is indigenous, and also a dominant species in Botswana, this pupal parasitoid 6136 may supplement the actions of larval parasitoids in managing cereal stemborers (Conlong and 6137 Cugala, 2008). Although some studies have reported the presence of hyperparasitoids in the 6138 stemborer food web (Polaszek & LaSalle, 1995; Rwomushana et al., 2005; Wale et al., 2006; 6139 Moolman et al., 2013), their absence in the present study suggests they will not be an interfering 6140 factor in Botswana. 6141 In conclusion, the current study documented the diversity of stem borer natural enemies 6142 in Botswana, and their associations with stemborers and particular host plants in both natural and 6143 cultivated habitats. The 30 species of natural enemies collected included 24 species of parasitoids 6144 and three species of predators that were recorded for the first time in Botswana. Natural enemy 6145 diversity and stemborer parasitism were both higher in cultivated than natural habitats, probably 6146 due to higher host availability in the former. Overall, Botswana landscapes host a diverse 6147 community of natural enemies that contribute to biological control of lepidopteran stemborers, 6148 and an equally diverse flora that hosts stemborers and their guild of parasitoids. 6149

6150 References 6151 Addo-Bediako, A. & Thanguane, N. (2012) Stemborer distribution in different sorghum cultivars 6152 as influenced by soil fertility. Agricultural Science Research Journal, 2, 189- 194. 6153 Arab, A. & Bento, J.M.S. (2006) Plant Volatiles: New Perspectives for Research in Brazil. 6154 Neotropical Entomology, 35, 151-158. 6155 Assefa, Y., Conlong, D.E. & Mitchell, A. (2006) First records of the stemborer complex 6156 (Lepidoptera: Noctuida; Crambidae; Pyralidae) in commercial sugarcane estates of Ethiopia, 6157 their host plants and natural enemies. Proceedings of the South African Sugar Technologists’ 6158 Association, 80, 202-213 6159 Assefa, Y., Mitchell, A., Conlong, D.E. & Muirhead, K.A. (2008) Establishment of Cotesia 6160 flavipes (Hymenoptera: Braconidae) in sugarcane fields of Ethiopia and origin of founding 6161 population. Journal of Economic Entomology, 101, 686-691.

256

6162 Baskaran, R.K.M., Sharma, K.C. & Kumar, J. (2017) Seasonal and relative abundance of stem- 6163 borer and leaf-folder in wet land rice eco-system. Journal of Entomology and Zoology 6164 Studies, 5, 879-884. 6165 Batalden, R.V., Oberhauser, K. & Peterson, A.A. (2007) Ecological niches in sequential 6166 generations of eastern North American Monarch butterflies (Lepidoptera: Danaidae): the 6167 ecology of migration and likely climate change implications. Environmental Entomology, 36, 6168 1365–1373. 6169 Bonhof, M.J., Overholt, W.A., Van Huis, A. & Polaszek, A. (1997) Natural enemies of cereal 6170 stem borers in east Africa: a review. Insect Science and its Application, 17, 18–35. 6171 Branca, A., Le Ru, B.P., Vavre, F., Silvain, J.F. & Dupas, F. (2011) Intraspecific specialization 6172 of the generalist parasitoid Cotesia sesamiae revealed by polyDNAvirus polymorphism and 6173 associated with different Wolbachia infection. Molecular Ecology, doi: 10.1111/j.1365- 6174 294X.2010.04977.x 6175 Chinwada, P., Schulthess, A.F., Overholt, W.A., Jowah, P. & Omwega, C.O. (2008) Release and 6176 establishment of Cotesia flavipes for biological control of maize stemborers in Zimbabwe. 6177 Phytoparasitica, 36, 160-167 6178 Chinwada, P. & Overholt, W.A. (2001) Natural enemies of maize stemborers on the highveld of 6179 Zimbabwe. African Entomology, 9, 67-75 6180 Conlong, D.E. (2001) Biological control of indigenous African stem borers, what do we know? 6181 Insect Science and its Application, 21, 267-274. 6182 Conlong, D.E. (2000) Indigenous African parasitoids of Eldana saccharina (Lepidoptera: 6183 Pyralidae). Proceedings of the South African Sugar Technologists Association 74, 201-211” 6184 Conlong, D.E., 1997. Biological control of Eldana saccharina Walker (Lepidoptera: Pyralidae) 6185 in sugarcane in South Africa: Constraints identified from 15 years of research. Insect Science 6186 and its Application, 17, 69-78 6187 Conlong, D.E. (1994) A review and perspectives for the biological control of the African 6188 sugarcane stalkborer Eldana saccharina Walker (Lepidoptera: Pyralidae). Agriculture, 6189 Ecosystems and Environment, 48, 9-17 6190 Conlong, D.E. & Cugala, D. (2008) The use of classical and augmentation biological control for 6191 the south–east Asian borer Chilo sacchariphagus Bojer (Lepidoptera: Crambidae) in 6192 Mozambican sugarcane. In MASON, P.G., GILLESPIE, D.R. AND VINCENT, C. (EDS)

257

6193 Proceedings of the Third International Symposium on Biological Control of Arthropods. 6194 Christchurch, New Zealand. February 8-13, 2009. pp. 116-122. USDA Forest Service, 6195 Morgantown, WV. Publication FHTET-2008-06, December 2008. 636p. 6196 Conlong, D.E. & Goebel, F.R. (2006) Trichogramma bournieri Pintureau and Babault 6197 (Hymenoptera: Trichogrammatidae) and Chilo sacchariphagus Bojer (Lepidoptera: 6198 Crambidae) in sugarcane in Mozambique- A new association. Annales de la Société 6199 entomologique de France, 42, 417-422. 6200 Conlong, D.E. & Goebel, F.R. (2002) Biological control of Chilo sacchariphagus (Lepidoptera: 6201 Crambidae) in Mozambique: The first steps. Proceedings of the South African Sugar 6202 Technologists Association, 76, 310-320. 6203 Conlong, D.E., Graham, D.Y. & Hastings, H. (1988) Notes on the natural host surveys and 6204 laboratory rearing of Goniozus natalensis Gordh (Hymenoptera: Bethylidae), a parasitoid of 6205 Eldana saccharina Walker (Lepidoptera: Pyralidae) larvae from Cyperus papyrus L. in 6206 Southern Africa. Journal of the Entomological Society of Southern Africa, 51, 115-127 6207 Conlong, D.E. & Mugalula, A. (2001) Eldana saccharina (Lepidoptera: Pyralidae) and its 6208 parasitoids at Kinyara Sugar Works, Uganda. Proceedings of the South African Sugar 6209 Technologists Association, 75, 183-185. 6210 Conlong, D.E. & Rutherford, S.R. (2009) Conventional and New Biological and Habitat 6211 Interventions for Integrated Pest Management Systems: Review and Case Studies using 6212 Eldana saccharina Walker (Lepidoptera: Pyralidae). Chapter 10, pp 241-261. In Peshin, R 6213 and Dhawan, A.K. (Eds). Integrated Pest Management: Innovation-Development Process, 6214 Vol.1. Ch. 10. Springer, Dordrecht, The Netherlands. ISBN 978-1-4020-8991-6 (Print); 978- 6215 1-4020-8992-3 (online). DOI 10.1007/978-1-4020-8992-3_10. 6216 Cugala, D.R. & Omwega, C.O. (2001) Cereal stem borer distribution and abundance and the 6217 introduction and establishment of Cotesia flavipes Cameron (Hymenoptera: Braconidae) in 6218 Mozambique. Insect Science and its Application, 21, 281–287. 6219 Dejen, A., Getu, E., Azerefegne, F. & Ayalew, A. (2013) Distribution and Extent of Cotesia 6220 flavipes Cameron (Hymenoptera: Braconidae) Parasitism in Northeastern Ethiopia. 6221 International Journal of Insect Science, 5, 9-19 6222 De Groote, H. (2002) Maize yield loss from stemborers in Kenya. Journal of Insect Science, 22, 6223 89–96.

258

6224 Dormann, C.F., Fruend, J., Bluethgen, N. & Gruber, B. (2009) Indices, graphs and null models: 6225 analyzing bipartite ecological networks. The Open Ecology Journal, 2, 7–24. 6226 FAO. (2002) Comprehensive Africa Agriculture Development Programme (CAADP). NEPAD. 6227 Rome. Italy 6228 Gounou, S. & Schulthess, F. (2004) Spatial distribution of lepidopterous stemborers on 6229 indigenous host plants in West Africa and its implications for sampling schemes. African 6230 Entomology, 12, 171-178 6231 Helena, M., Peixoto, P., Pufal, G., Staab, M., Martins, C.F. & Kleins, A.M. (2016) Diversity and 6232 specificity of host-natural enemy interactions in an urban-rural interface. Ecological 6233 Entomology, 41, 241-252. 6234 Hordzi, W.H.K. & Botchey, M.A. (2012) Some parasitoids of lepidopterous stem borer pests on 6235 maize in southern Ghana. Bulletin of Environment, Pharmacology and Life Sciences, 1, 77- 6236 83 6237 Kankonda, O.M., Akaibe, B.D., Ong’amo, G.O. & Le Ru, B.P. (2017) Diversity of lepidopteran 6238 stemborers and their parasitoids on maize and wild host plants in the rain forest of Kisangani, 6239 DR Congo. Phytoparasitica, 42, 57-69. DOI 10.1007/s12600-017-0561-6 6240 Kfir, R., Overholt, W.A., Khan, Z.R. & Polaszek, A. (2002) Biology and management of 6241 economically important lepidopteran cereal stemborers in Africa. Annual Review of 6242 Entomology, 47, 701–731. PMID: 11729089. 6243 Kipkoech, A.K., Schulthess, F., Yabann, W.K., Maritim, H.K. & Mithöfer, D. (2006) Biological 6244 control of cereal stem borers in Kenya: A cost benefit approach, Annales de la Société 6245 entomologique de France (N.S.): International Journal of Entomology, 42, 519-528 6246 Kodjo, T.A., Komi, A., Mawufe, A.K. & Komlan, W. (2013) Maize stemborers distribution, 6247 their natural enemies and farmers’ perception on climate change and stemborers in southern 6248 Togo. Journal of Applied Biosciences, 64, 4773-4786 6249 Le Ru, B.P., Ong’amo, G.O., Moyal, P., Ngala, L., Musyoka, B., Abdullah, Z., Cugala, D., 6250 Defabachew, B., Haile, T.A., Matama-Kauma, T., Lada, V.Y., Negassi, B., Pallangyo, K., 6251 Ravolonandrianina, J., Sidumo, A., Omwega, C.O., Schulthess, F., Calatayud, P.A. & 6252 Silvain, J.F. (2006) Diversity of lepidopteran stem borers on monocotyledonous plants in 6253 eastern Africa and the islands of Madagascar and Zanzibar revisited. Bulletin of 6254 Entomological Research, 96, 555–563.

259

6255 Mailafiya, D.M., Le Ru, B.P., Kairu, E.W., Dupas, S. & Calatayud, P.A. (2011) Parasitism of 6256 lepidopterous stemborers in cultivated and natural habitats. Journal of Insect Science, 11, 1- 6257 19 6258 Mailafiya, M., Le Ru, B.P., Kairu, E.W., Calatayud, P.A. & Dupas, S. (2010a) Factors affecting 6259 stem borer parasitoid species diversity and parasitism in cultivated and natural habitats. 6260 Environmental Entomology, 39, 57-67 6261 Mailafiya, D.M., Le Ru, B.P., Kairu, E.W., Calatayud, PA. & Dupas, S. (2010b) Geographic 6262 distribution, host range and perennation of Cotesia sesamiae and Cotesia flavipes Cameron 6263 in cultivated and natural habitats in Kenya. Biological Control, 54, 1–8. 6264 Mailafiya, D.M., Le Ru, B.P., Kairu, E.W., Calatayud, P.A. & Dupas, S. (2009) Species diversity 6265 of lepidopteran stem borer parasitoids in cultivated and natural habitats in Kenya. Journal of 6266 Applied Entomology, 133, 416-429 6267 Matama-Kauma, T., Schulthess, F., Le Ru, B.P., Mueke, J.M., Ogwang, J.A. & Omwega, C.E. 6268 (2008) Abundance and diversity of lepidopteran stemborers and their parasitoids on selected 6269 wild grasses in Uganda. Crop Protection, 27, 505-513 6270 Matama-Kauma ,T., Schulthess, F., Ogwang, J.A.,, Mueke, J.M. & Omwega, C.E. (2007) 6271 Distribution and relative importance of lepidopteran cereal stemborers and their parasitoids 6272 in Uganda. Phytoparasitica, 35, 27-36 6273 Mazodze, R. & Conlong, D.E. (2003) Eldana saccharina (Lepidoptera: Pyralidae) in sugarcane 6274 (Saccharum hybrids), sedge (Cyperus digitatus) and bulrush (Typha latifolia) in south- 6275 eastern Zimbabwe. Proceedings of the South African Sugar Technologists Association, 77, 6276 256-274. 6277 Midingoyi, S., Affognon, H., Macharia, I., Ong’amo, G., Abonyo, E., Ogola, G., De Groote, H. 6278 & Le Ru, B. (2016) Assessing the long-term welfare effects of the biological control of 6279 cereal stemborer pests in East and Southern Africa: Evidence from Kenya, Mozambique and 6280 Zambia Research Paper. Agriculture, Ecosystems and Environment, 230, 10–23. 6281 Moolman, J., Van den Berg, J., Conlong, D., Cugala, D., Siebert, S. & Le Ru, B.P. (2014) 6282 Species diversity and distribution of lepidopteran stem borers in South Africa and 6283 Mozambique. Journal of Applied Entomology, 138, 52-66

260

6284 Moolman, H.J., Van den Berg, J., Conlong, D., Cugala, D., Siebert, S.J. & Le Ru, B.P. (2013) 6285 Diversity of stemborer parasitoids and their associated wild host plants in South Africa and 6286 Mozambique. Phytoparasitica, 41, 89-104 6287 Muyekho, F.N., Barrion, A.T. & Khan, Z.R. (2005) Host range for stemborers and associated 6288 natural enemies in different farming systems of Kenya. African Crop Science Journal, 13, 6289 173-183 6290 Mutyambai, D.M., Bruce, T.J.A., Midega, C.A.O., Christine, M., Woodcock, C.M., Caulfield, 6291 J.C., Van Den Berg, J., Pickett, J.A. & Khan, Z.R. (2015) Responses of parasitoids to 6292 volatiles induced by Chilo partellus oviposition on teosinte, a wild ancestor of maize. 6293 Journal of Chemical Ecology, 41, 323–329. 6294 Ndemah, R., Schulthess, F., Leru, B.P. & Bame, I. (2007) Lepidopteran cereal stemborers and 6295 associated natural enemies on maize and wild grass hosts in Cameroon. Journal of Applied 6296 Entomology, 131, 658-668 6297 Ngi-song, A.J. & Overholt, W.A. (1997) Host location and acceptance by Cotesia flavipes 6298 Cameron and C. sesamiae (Cameron) (Hymenoptera: Braconidae), parasitoids of African 6299 gramineous stem borers: role of frass and other host cues. Biological Control, 6, 136-142 6300 Obonyo, M., Schulthess, F., Le Ru, B.P., Van den berg, J. & Calatayud, P.A. (2010) Host 6301 recognition and acceptance behaviour in Cotesia sesamiae and C. flavipes (Hymenoptera: 6302 Braconidae), parasitoids of gramineous stemborers in Africa. European Journal of 6303 Entomology, 107, 169-176 6304 Obopile, M. & Mosinkie, K.T. (2003) Natural enemies of lepidopterous stemborers on maize and 6305 sorghum in Botswana. UNISWA Journal of Agriculture, 11, 64-69 6306 Ofomata, V.C., Overholt, W.A., Lux, S.A., Van Huis, A. & Egwuatu, R.I. (2000) Comparative 6307 studies on the fecundity, egg survival, larval feeding and development of Chilo partellus 6308 (Swinhoe) and Chilo orichalcociliellus strand (Lepidoptera: Crambidae) on five grasses. 6309 Annals of the Entomological Society of America, 93, 492-499. 6310 Oloo, G.W. (1989) The role of local natural enemies in population dynamics of Chilo partellus 6311 (Swinhoe) (Pyralidae) under subsistence farming systems in Kenya. Insect Science and its 6312 Application, 10, 243-251. 6313 Omwega, C.O., Muchugu, E., Overholt, W.A. & Schulthess, F. (2006) Release and establishment 6314 of Cotesia flavipes (Hymenoptera: Braconidae) an exotic parasitoid of Chilo partellus (Lep.,

261

6315 Cambridae) in East and Southern Africa. Annales de la Société Entomologique de France, 6316 42, 511-517. 6317 Onyango, F.O. & Ochieng’-Odero, J.P.R. (1994) Continous rearing of the maize stem borer 6318 Busseola fusca on an artificial diet. Entomologia Experimentalis et Applicata, 73, 139-144 6319 Overholt, W.A., Ngi-Song, A.J., Omwega, C.O., Kimani-Njogu, S.W., Mbapila, J., Sallam, M.N. 6320 & Ofomata, V. (1997) A review of the introduction and establishment of Cotesia flavipes 6321 Cameron in East Africa for the biological control cereal stemborers. Insect Science and its 6322 Application, 17, 79-88. 6323 Overholt, W.A., Ogedah, K. & Lammers, P.M. (1994) Distribution and sampling of C. partellus 6324 (Swinhoe) (Lepidoptera: Pyralidae) in maize and sorghum at the Kenyan Coast. Bulletin of 6325 Entomological Research, 84, 367-378. 6326 Overholt, W.A., Songa, J.M., Ofomata, V. & Jeske, R. (2000) The spread and ecological 6327 consequences of the invasion of Chilo partellus (Swinhoe) (Lepidoptera: Crambidae) in 6328 Africa. In E. E. Lyons and S. E. Miller Invasive species in eastern Africa: Proceedings of a 6329 Workshop held at ICIPE, July 5-6, 1999. ICIPE Science Press, Nairobi 6330 Polaszek, A. (1998) African cereal stem borers: Economic importance, taxonomy, natural 6331 enemies and control. CABI Publishing. Wallingford 6332 Polaszek, A. & LaSalle, J. (1995) The hyperparasitoids (Hymenoptera: Ceraphronidae, 6333 Encyriidae, Eulophidae, Eurytomidae) of cereal stem borers (Lepidoptera: Noctuidae, 6334 Pyralidae) in Africa. African Entomology, 3, 131-146. 6335 Potting, R.P.J., Vet L.E.M., Dicke, M., 1995. Host microhabitat location by stem-borer 6336 parasitoid Cotesia flavipes: The role of herbivore volatiles and locally and systemically 6337 induced plant volatiles. Journal of Chemical Ecology, 21, 525-539. 6338 Rahaman, M.M., Islam, K.S., Jahan, M. & Mamun M.A.A. (2014) Relative abundance of stem 6339 borer species and natural enemies in rice ecosystem at Madhupur, Tangail, Bangladesh. 6340 Journal of the Bangladesh Agricultural University, 12, 267–272, ISSN 1810-3030 6341 Rwomushana, I., Kyamanywa, S., Ngi-Song, A.J. & Omwega, C.O. (2005) Incidence of larval 6342 and pupal parasitoids of Lepidoptera in Uganda with special reference to new associations. 6343 International Journal of Tropical Insect Science, 25, 274–280

262

6344 Sallam, M.N., Overholt, W.A. & Kairu, E. (1999) Comparative evaluation of Cotesia flavipes 6345 and C. sesamiae (Hymenoptera: Braconidae) for the management of Chilo partellus 6346 (Lepidoptera: Pyralidae) in Kenya. Bulletin of Entomological Research, 89, 185-191 6347 Sampson, M.A. & Kumar, R. (1986b) Parasitism of Descanpsina Sesamiae ((Mensil) on Sesamia 6348 species in sugarcane in southern Ghana. Insect Science and Its Application, 7, 543- 546. 6349 Sétamou, M., Jiang, N. & Schulthess, F. (2005) Effect of the host plant on the survivorship of 6350 parasitized Chilo partellus Swinhoe (Lepidoptera: Crambidae) larvae and performance of its 6351 larval parasitoid Cotesia flavipes Cameron (Hymenoptera: Braconidae). Biological Control, 6352 32, 183-190. 6353 Simpson, E.H. (1949) Measurement of diversity. Nature, 163, 688. 6354 Skoroszewski, R.W. & Van Hamburg, H. (1987) The release of Apanteles flavipes (Cameron) 6355 (Hymenoptera: Braconidae) against stalk borers in maize and grain sorghum in South Africa. 6356 Journal of the Entomological Society of Southern Africa, 50, 249-255 6357 Sohati, P.H., Musonda, E.M. & Mukanga, M. (2001) Distribution of cereal stem borers in 6358 Zambia and release of Cotesia flavipes Cameron, an exotic natural enemy of Chilo partellus 6359 (Swinhoe). Insect Science and its Applications, 21, 311-316. 6360 Sylvain, N.M., Manyangarirwa, W., Tuarira, M. & Onesime, M.K. (2015) Effects of 6361 lepidopterous stemborers, Busseola fusca (Fuller) and Chilo partellus (Swinhoe) on maize 6362 (Zea mays L) yield: A review. International Journal of Innovative Research and 6363 Development, 4, 181-188 6364 Thies, C., Steffan-Dewenter, I. & Tscharntke, T. (2003) Effects of landscape context on 6365 herbivory and parasitism at different spatial scales. Oikos, 101, 18-25 6366 Udayagiri, S. & Welter, S.C. (2000) Escape of Lygus hesperus (Heteroptera: Miridae) eggs from 6367 parasitism by Anaphesiole (Hymenoptera: Mymaridae) in strawberries: plant structure 6368 effects. Biological Control, 17, 234-242. 6369 Van Achterberg, C. & Walker, A.K. (1998) Braconidae. In A.Polaszek (Ed.), African stem 6370 borers; economic importance, taxonomy, natural enemies and control. CTA/CABI. 6371 Wallingford, UK, pp. 137-185. 6372 Wale, M., Schulthess, F., Kairu, E.W. & Omwega, C.O. (2006) Distribution and relative 6373 importance of cereal stem borers and their natural enemies in the semi-arid and cool-wet

263

6374 ecozones of the Amhara State of Ethiopia, Annales de la Société entomologique de France 6375 (N.S.), 42, 389-402, DOI:10.1080/00379271.2006.10697471 6376 Walton, A.J. & Conlong, D.E. (2016) Radiation biology of Eldana saccharina (Lepidoptera: 6377 Pyralidae) Florida Entomologist, 99, 36-42. 6378 Wiedenmann, R.N. & Smith, J.W.Jr. (1993) Functional response of the parasite Cotesia flavipes 6379 (Cam.) (Hym.: Braconidae) at low densities of the host Diatraea saccharalis (Lepidoptera: 6380 Crambidae). Environmental Entomology, 22, 848-858 6381 Zhou, G., Overholt, W.A. & Kimani-Njogu, S.W. (2003) Species richness and parasitism in 6382 assemblage of parasitoids attacking maize stem borer in coastal Kenya. Ecological 6383 Entomology, 28, 109–118. 6384

6385

6386

6387

6388

6389

6390

6391

6392

6393

6394

6395

6396

6397

6398

6399

6400

264

6401

6402

6403

6404 APPENDIX 2

6405

6406

6407

6408

6409

6410

6411 First Report of Tomato Leaf miner, Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae) in 6412 Botswana*

6413

6414

6415

6416

6417

6418

6419

6420

6421 * Published as “Mutamiswa, R., Machekano, H. & Nyamukondiwa, C. (2017) First Report of 6422 Tomato Leaf miner, Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae) in Botswana. 6423 Agriculture and Food Security, 6, 49

265

6424 2.1 Introduction 6425 Insect pest invasions have been rapidly increasing worldwide, and with increased 6426 movement of people and goods from one country to another, there are high chances for increased 6427 numbers of invasive species conquering many regions (Pimentel et al., 2001). Furthermore, 6428 climate is changing (IPCC, 2014), and in consequence, impact on the rate of biological 6429 invasions, insect distribution, abundance, and impacts of such invasions on a global scale (Hill et 6430 al., 2016). In Africa, the risk and rate of invasion has dramatically increased (Pimentel et al., 6431 2001), with destructive insect pests like Chilo partellus Swinhoe (Tams, 1932), Prostephanus 6432 truncatus (Horn) (Dunstan et al., 1981), Phenacoccus manihoti Matile-Ferrero 6433 (Neuenschwander, 2001) and Bactrocera dorsalis (Hendel) (Lux et al., 2003) invading the 6434 continent over the past decades. These invasions pose a significant risk in a continent where ~70- 6435 80% of the population depends on agriculture for household food security and sustenance (FAO, 6436 2002). Exports of vegetables from developing countries have immensely increased in the past 6437 decades and horticultural exports have been reported to contribute to improvement in food 6438 security and livelihoods in these developing countries (Van den Broecke & Maertens, 2016). 6439 Tomato, Solanum lycopersicum L. (Solanales: Solanaceae) is highly ranked as a food and cash 6440 crop (Varela et al., 2003; Tumuhaise et al., 2016) in Africa and indeed Botswana (Madisa et al., 6441 2012). However, biotic factors such as insect pests hamper production (Tumuhaise et al., 2016), 6442 which may compromise household and national food security. 6443 The tomato leafminer, Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae) also known 6444 as South American tomato pinworm is one of the key insect pests of tomato (Brévault et al., 6445 2014). This may be because of the extensive damage it elicits, lack of ecologically relevant 6446 methods for management (Ferrara et al., 2001) and pesticide resistance (Roditakis et al., 2015). 6447 It is very destructive to tomato plants and fruits causing between 80-100% yield losses if left 6448 uncontrolled (Desneux et al., 2010). Furthermore, it has a wide host range, and has been 6449 reported on cultivated solanaceous plants such as eggplant (Solanum melongena L.), potato 6450 (Solanum tuberosum L.), pepper (Capsicum annuum L.), tobacco (Nicotiana tabacum L.), wild 6451 solanaceous weeds like black nightshade Solanum nigrum, Solanum elaeagnifolium, Solanum 6452 puberulum, Datura stramonium, Datura ferox, Nicotiana glauca and garden bean (Phaseolus 6453 vulgaris L.) (Ferracini et al., 2012). It is of South American origin, Andes region, Peru, and was 6454 first detected in eastern Spain in 2006 (Desneux et al., 2010; Brévault et al., 2014). Thereafter, it

266

6455 invaded the Mediterranean Basin, Europe, Middle East, South Asia (India), north, east and west 6456 Africa (Desneux et al., 2010; Desneux et al., 2011; Brévault et al., 2014). In north Africa, T. 6457 absoluta was detected north of the Sahel, Tunisia and Morocco in 2008 (Desneux et al., 2010; 6458 Abbes et al., 2012; Ouardi et al., 2012), west Africa; Niger and Nigeria in 2010, Senegal in 2012 6459 (Pfeiffer et al., 2013) east Africa; Sudan and Ethiopia in 2011 (Anon, 2012; Pfeiffer et al., 2013), 6460 Kenya in 2013 (Tumuhaise et al., 2016), Tanzania in 2014 (Chidege et al., 2016) and Uganda in 6461 2015 (Tumuhaise et al., 2016), southern Africa; Zambia (IPPC, 2016a) and South Africa in 2016 6462 (IPPC, 2016b; Visser et al., 2017) (see distribution map, Figure 2.1). 6463

T. absoluta reported

T. absoluta not reported

6464

6465 Figure 2.1: Current distribution of Tuta absoluta in Africa, redrawn from EPPO, 2016).

267

6466 Tuta absoluta is nocturnal and adults spend the day hiding between leaves (Desneux et 6467 al., 2010). It is multivoltine, producing between 10-12 generations per year (Barrientos et al., 6468 1998; EPPO, 2005). It is highly tolerant to high temperature, with a developmental optimum of 6469 30ºC and a wide developmental thermal window (Krechemer et al., 2015; Martins et al., 2016). 6470 This implies that during its geographic range expansion, T. absoluta has a high probability of 6471 thriving in novel variable thermal environments, and even with global climate change (IPCC, 6472 2014). The biological cycle takes an average of 24-38 days at 27 ºC and females can lay 250-300 6473 cylindrical creamy yellow eggs, mostly singly on aerial parts and young fruits of the host plant 6474 (Barrientos et al., 1998; EPPO, 2005; Desneux et al., 2010). After 4-6 days, the eggs develop 6475 into 0.5mm yellow or green larvae (EPPO, 2005; Krechemer et al., 2015). The larval stage takes 6476 12-15 days and goes through four developmental instars (EPPO, 2005; Harizanova et al., 2009). 6477 The first 2 instars have been reported to mine between the epidermal layers of the leaf leading to 6478 a reduction in the photosynthetic area and premature senescence (Pfeiffer et al., 2013). 6479 Thereafter, larvae leave the mines as 3rd and 4th instars, boring into stalks, apical buds and fruits 6480 (Ferracini et al., 2012). Larval damage also promotes the entry of secondary pathogens causing 6481 fruit rot (EPPO, 2005) and forms the most destructive developmental stage (Harizanova et al., 6482 2009; Desneux et al., 2010; Retta & Berhe, 2015). Pupation takes place in the soil, within the 6483 mines, on the leaves or in packaging materials (Pfeiffer et al., 2013; Retta & Berhe, 2015) and 6484 can last 9-11 days at benign climatic conditions (EPPO, 2005). With the rapid north to south 6485 movement and high invasion potential of T. absoluta over wide geographic areas in Africa (Fig 6486 1), it still poses a biosecurity threat to countries where it has not yet been detected. Here, I make 6487 the first report of T. absoluta in the north eastern part of Botswana, and recommend management 6488 strategies following detection. Detection and description of new species upon introduction in 6489 novel environments is of paramount importance in crafting management plans for invasive 6490 species (Saccaggi et al., 2016). I then explore preventative and control measures to be 6491 undertaken by various stakeholders to reduce its spread into pest free areas or further invasions 6492 into neighboring countries. 6493

6494 2.2 Materials and Methods 6495 The first economic damage of T. absoluta was reported at Genesis farm (S21.14776; 6496 E27.64744), Matshelagabedi village in North East District of Botswana. A report was made to

268

6497 the Department of Plant Protection in the Botswana Ministry of Agriculture. The second 6498 detection point was Noka Farm (S21.12860; E27.48830), Francistown, in the same district. 6499 Following detection, a physical assessment of crop damage symptoms and presents of the pest 6500 was conducted and infested tomato fruit samples were collected and incubated in climate 6501 chambers (HPP 260, Memmert GmbH + Co.KG, Germany) at 24 ± 1 °C, 70±5% RH (Bawin et 6502 al., 2014) in insect cages (35cm3) until adult eclosion. Emerged adult moths were collected and 6503 examined for confirmation. 6504

6505 2.2.1 Morphological identification 6506 Morphological characters were confirmed at Botswana International University of 6507 Science and Technology (BIUST), Botswana and Stellenbosch University, Cape Town, South 6508 Africa. Adult moths were collected and knocked down in killing jars containing Chloroform 6509 absorbed in cotton wool. Using a stereo microscope, Bestscope (model BS3060BT) connected to 6510 a computer, morphological features were examined and photographed using a microscope 6511 mounted 5.0 MP digital camera (DCM-510) (Hangzhou Scopetek® Opto Electric Co, Hangzhou, 6512 China). Tuta absoluta can be reliably identified morphologically using male genital features, by 6513 examining the valvae and gnathos (Hayden et al., 2013). Males have broad, horseshoe-shaped 6514 gnathos and a digitate valva, with a medial hump and constriction (Hejazi et al., 2016).

6515 2.2.2 Plant damage symptoms 6516 Damaged tomato plants, plant parts and fruits from tunnels, open fields and net shades were 6517 collected and examined visually using illuminated bench magnascopes (RBM 101 model, 6518 Radical Instruments, India). Tunnelling and feeding behaviour in fruits, stems and leaves was 6519 observed. 6520

6521 2.2.3 Sex Pheromone Trapping 6522 Yellow delta traps (Chempac- Progressive Agricare®) equipped with sticky pads were 6523 placed in tomato tunnels, open fields on and around the core detection point (Figure 2.2). A 6524 synthetic sex pheromone, Tuta absoluta-optima PH-937-OPTI (Russel IPM) was used as the lure 6525 (see Figure 2.3). The traps were set following a survey protocol developed by (Manrakhan et al., 6526 2012) with necessary modifications (Table 2.1)

269

6527 6528 6529 6530 6531 6532 6533 6534 6535 6536 6537

6538 A 6539 6540 6541 6542 6543 6544 6545 6546 6547 6548 6549 6550 B 6551 6552 6553 Figure 2.2: Yellow delta pheromone trap placed at (A) Genesis farm (S21.14776; E27.64744; 6554 987 m.a.s.l) (tunnel agroecosystem) and (B) open forest (S21.16684; E27.57120; 1013 m.a.s.l) 6555 wild habitat) with predominant Colophospermum mopane, North East District of Botswana. 6556 Photos by H. Machekano and R. Mutamiswa 6557

270

6558 6559 6560 Figure 2.3: Tuta absoluta PH-937-OPTI female sex pheromone (Russell IPM®) used in the delta 6561 traps. 6562 6563 Table 2.1: Delimited detection zones for Tuta absoluta detection in Matshelagabedi village, 6564 Northern District of Botswana. Data was collected 7 days (03 December to 10 December 2016) 6565 and trapped moths were counted using dyed pointers and tally counters. 6566 Location Distance from the core Number of traps/zone detection site (m) Core detection point (Genesis farm) 0 6 1st (Open forest) 500 6 2nd (Open forest) 8000 2 3rd (Open forest) 16000 2 4th (Tomato fields) 24000 2

271

6567 2.2.4. Molecular analysis 6568 I extracted total genomic DNA from moths and larvae from the field collected and 6569 incubated samples. DNA extractions were performed using the QIAmp®DNA Micro Kit 6570 (Qiagen GmbH, Hilden, Germany) according to the manufacturer’s protocol. Extracted DNA 6571 concentrations were measured using a NanoDrop® ND-1000 Spectrophotometer (NanoDrop 6572 Technologies, Inc.). One nanogram of DNA was used in subsequent PCR amplifications. PCR 6573 products of ~685 bp in length were amplified for the COI gene using the primer set LCO1490 6574 (5'-GGTCAACAAATCATAAAGATATTGG-3') and HCO2198 (5'- 6575 TAAACTTCAGGGTGACCAAAAAATCA-3'), which amplifies a 710 bp fragment of the COI 6576 gene in a wide range of invertebrate taxa (Folmet et al., 1994). The TopTaq MasterMix kit 6577 (QIAGEN) was used in all reactions. Thermocycling conditions consisted of denaturation at 95 6578 °C for 1 minute, followed by 35 cycles of 95 °C for 45 seconds, 51 °C for 45 seconds and 72 °C 6579 for 1 minute; with a final extension at 72 °C for 3 minutes. The PCR product was visualized on a 6580 1.2 % agarose gel and purified using the Wizard® Genomic DNA Purification Kit (Promega 6581 Corporation). Purified products were sequenced both ways using BigDye® Terminator v3.1 6582 chemistry (Applied Biosystems) with the same primer pair used for the PCR 6583 reactions. Sequences were edited and aligned using the CLC main workbench 6.9. 6584

6585 2.2.5. Data analysis 6586 Insect morphological features were identified under microscope (1:10 zoom ratio, 0.8 × ~ 6587 0.8 obj. mag, 50~75mm binocular head) (BestScope®, China). Qualitative data on crop damage 6588 symptoms were collected by observation and sample incubation. Moth trap count data were 6589 collected from the sticky pads and counted. Detected adult moth counts were presented as 6590 graphs. The COI sequence data from molecular analysis was compared to the standards for 6591 confirmation. Sequence data was compared with existing sequences in the Genebank 6592 (KX443111, KX443108, KP793741, KP814057 and JQ749676) and (KJ657881, KJ657680, 6593 KC852871, KT452897, KP793742, KC852872, KP324753 and an outgroup KX862248) in 6594 MEGA6 (Tamura et al. 2013), the latter of which were used to draw the phylogenetic tree. 6595

272

6596 2.3. Results 6597

6598 2.3.1 Morphological features 6599 Morphological features on the collected moths show head vertex covered with appressed 6600 scales that appear flattened against the head. The labial palps were also scally and had the same 6601 colour as the head with a distinguishingly forward projecting, up-curved shape with a relatively 6602 long pointed apical segment. Other general features like body size and colour also concurred 6603 with those of T. absoluta. Observed male genitalia conformed to that of T. absoluta as described 6604 by (Hayden et al., 2013) and also in agreement with (Visser et al., 2017) (Figure 2.4). The 6605 aedygus/ phallus was characterised by a broader, basal prominent caecum. The uncus was hood- 6606 shaped and quite broad at the apex. The uncus was attached to a tegument basally broadened 6607 with an ovate gnathos. The valvae were digitate within inner margin covex shape and each was 6608 covered with lightly dense setae. The vinculum was broad and well developed, with an elongated 6609 and broad saccus. 6610

6611 6612

6613 Figure 2.4: Tuta absoluta: male genitalia. Photograph by R. Ramukhesa 6614

273

6615 2.3.2 Crop damage symptoms 6616 Survey of the tomato fields and surrounding habitats showed extensive damage inflicted 6617 by T. absoluta. The damage was characterised by extensive wilting of whole plants associated 6618 with severe leaf and stem. The shoots had distorted shoots with signs of die back, dead hearts 6619 and wilting. The leaves showed lesions of different sizes, large necrotic areas, wilting and 6620 chlorosis (Figure 2.5A). Frass was largely visible on all damaged parts of the plants. Tuta 6621 absoluta damage was also observed on wild hosts e.g. Solanum lichtensteinii (Willd), a wild 6622 solanaceous plant native to Southern Africa (Figure 2.5B). The damage was also characterised 6623 by lesions and necrotic damage. However, the damage on S. lichtensteinii was less severe than 6624 that on tomato plants (see Figure 2.5A; B). Tomato fruit damage symptoms were characterised 6625 by internal feeding with distinct exit holes (Figure 2.5C; D) and substantial frass. Fruits attacked 6626 in their early developmental stages had distorted shapes and relatively smaller size (Figure 2.5C). 6627 Most damaged mature fruits showed signs of secondary infection, subsequent decomposition and 6628 loss of internal fruit contents (Figure 2.5D). 6629

274

A B

C D 6630 6631 6632 Figure 2.5: Typical Tuta absoluta damage symptoms on (A) tomato plants (B) wild host 6633 Solanum lichtensteinii (C) distorted and damaged fruit following larval fruit infestation, and (D) 6634 advanced damage signs with exit holes, secondary infection, hanging skins with consumed 6635 internal contents and accelerated senescence at Genesis farm (S21.14776; E27.64744; 987

6636 m.a.s.l). Photos by H. Machekano and R. Mutamiswa 6637

6638 2.3.3 Pheromone trapping 6639 Sex pheromone traps caught varying numbers of T. absoluta male moths (Figure 2.6) 6640 depending on location and whether the traps were in the open tomato fields, in tunnels or open 6641 natural/forest habitat. I recorded T. absoluta male moths on all pheromone baited traps and sites 6642 in the north eastern part of Botswana (see Figure 2.7 for detection sites). The core detection site 6643 however, showed generally higher mean number of moths per trap than the rest of the sites. More 6644 moths were caught inside the tunnels than outside (see Figure 2.8). The capture of male moths

275

6645 using a species synthetic equivalent of female emitted species specific sex pheromone (Ferrara et 6646 al., 2001) confirms the moths reported here were indeed T. absoluta. 6647 6648 6649 6650 6651 6652 6653 6654 6655

6656 A 6657 6658 6659 6660 6661 6662 6663 6664 6665

6666 B 6667 6668 Figure 2.6: Sex pheromone trap catches for T. absoluta male moths (A) set up inside the yellow 6669 delta trap and (B) sticky pad retrieved from the yellow delta trap following one week trapping

6670 period at Genesis Farm (S21.14776; E27.64744; 987 m.a.s.l). Photos by H. Machekano and R. Mutamiswa 6671

276

6672 6673 6674 Figure 2.7: A map of Botswana showing detection sites for Tuta absoluta 6675 6676 6677

277

400 371

350 315 300

250

200 188

150 106 100

Mean number moths/ of number Mean trap 50 50 26

0 Kitso Joyce Genesis Genesis Moji Noka Selamang Lordic Farm Farm Farm Farm farm Farm (Net (Open (Inside (Open (Open (Open shade forest tomato tomato tomato tomato tomato) 500m) tunnels) field) field) field) Detection site 6678 6679 6680 Figure 2.8: Mean number of Tuta absoluta male moths/trap over a period of seven days from 6681 different sites distant from core detection point in north east Botswana. Traps were baited using 6682 Tuta absoluta PH-937-OPTI (Russel IPM). 6683

6684 2.3.4 Molecular analysis 6685 The consensus sequence was used in a BLASTN (basic local alignment search tool) 6686 search (Altschul et al., 1990) to find matching sequences (Zhang et al., 2000). The COI sequence 6687 and phylogenetic tree matched the T. absoluta (100% identity) sequences KJ657881, KJ657680, 6688 KC852871, KT452897, KP793742, KC852872 (see Figure 2.9), as well as KX443111 and 6689 KX443108 (Sint et al., 2016); KP793741 (first report from India, Asokan et al., 2015, 6690 unpublished data) KP814057 (Shashank et al., 2015) and JQ749676 (Bettaïbi et al., 6691 2012). Phylogenetic analysis showed genetic distances between 0.00-0.59 between the 6692 Botswana T. absoluta sample and other samples deposited in the Genebank, and between 12.67- 6693 12.84 between T. absoluta samples and the outgroup Ephysteris promptella. The molecular data 6694 confirmed the present specimens as T. absoluta.

278

KJ657681 KJ657680 KC852871 87 KT452897 KP793742. Tuta_absoluta_Botswana KC852872 KP324753 KX862284.1 Ephysteris promptella

0.01 6695 6696 6697 Figure 2.9: A comparison of the sequences of Botswana specimens with existing sequences on 6698 GenBank; KJ657881, KJ657680, KC852871, KT452897, KP793742, KC852872, KP324753 and 6699 an outgroup KX862248. The evolutionary history was inferred using the Neighbor-Joining 6700 method (Saitou & Nei, 1987). The optimal tree with the sum of branch length = 0.13050185 is 6701 shown. The percentage of replicate trees in which the associated taxa clustered together in the 6702 bootstrap test (100 replicates) are shown next to the branches (Felsenstein, 1985). The tree is 6703 drawn to scale, with branch lengths in the same units as those of the evolutionary distances used 6704 to infer the phylogenetic tree. The evolutionary distances were computed using the Kimura 2- 6705 parameter method (Kimura, 1980) and are in the units of the number of base substitutions per 6706 site. The analysis involved 9 nucleotide sequences. Codon positions included were 6707 1st+2nd+3rd+Noncoding. All positions containing gaps and missing data were eliminated. There 6708 were a total of 510 positions in the final dataset. Evolutionary analyses were conducted in 6709 MEGA6 (Tamura et al., 2013). 6710

6711 2.4 Discussion 6712 The recent invasion and rapid geographic spread of T. absoluta poses a major threat to 6713 both natural and agroecosystems (Desneux et al., 2010) in the African region. While T. absoluta 6714 has been reported in east, north and west Africa (Desneux et al., 2010; Abbes et al., 2012; 6715 Ouardi et al., 2012), to my knowledge, this is the first scientific report, detailing the presence of 6716 this pest in Botswana and generally Southern Africa.

279

6717 6718 Tuta absoluta was caught at all the sites where the traps were set in the two districts of 6719 Botswana suggesting that there are high chances of widespread distribution of the insect pest in 6720 these and other districts. This is also supported by the detection of moths at distant places from 6721 the core detection site (Genesis farm) and in natural habitats. Indeed, this finding agrees with 6722 previous study by (Desneux et al., 2010) who detected adult T. absoluta at ~10 kilometers from 6723 the tomato fields. Presence of T. absoluta in the wild may explain its host polyphagy and also 6724 means it may be capable of migrating distances beyond 10km. Indeed, the adult moths have 6725 been reported to fly distances of up to 100km (CFIA, 2016), and given their small size, they are 6726 easily blown by wind currents. This characteristic further improves spread of propagules upon 6727 introduction in new habitats and thus invasion success. 6728 The mitochondrial COI (cytochrome c oxidase subunit I) sequencing through DNA 6729 Barcoding linked the Botswana trap catches to T. absoluta samples from Tunisia (Accession 6730 number JQ749676) and India (Accession number KP793741), the characteristic host damage 6731 symptoms and sex specific pheromone trapping consequently supporting morphological 6732 identification that was done in Botswana and South Africa. This confirmed moth identity 6733 as T. absoluta, suggesting that this alien pest is highly invasive and is expanding its geographic 6734 distribution. The insect pest has been reported being on a downward incursion from north Africa 6735 towards the south, and this may have been promoted by its high reproductive capacity (Tonnang 6736 et al., 2015), wide host range (Retta & Berhe, 2015), wide developmental thermal windows 6737 (Krechemer et al., 2015; Martins et al., 2016), continuous vegetable production across political 6738 borders, absence of effective surveillance and monitoring systems, lack of effective sanitary and 6739 phytosanitary measures and increase in intra-continental trade (Mohamed, 2014). Moreover, 6740 resistance to conventional and new chemicals (Gontijo et al., 2013; Campos et al., 2014) has 6741 been reported as a contributing factor towards its invasion success (Lietti et al., 2005; Roditakis 6742 et al., 2015). In addition, African ecological and climatic conditions are similar to those of South 6743 American countries (Desneux et al., 2010), suggesting that T. absoluta may establish and invade 6744 other countries in the southern African region. 6745 Most of the farmers in Botswana who are into tomato production rely on importing 6746 seedlings from neighbouring countries. In addition, retailers are known to import tomato fruits 6747 from Zambia, Zimbabwe and South Africa for supplying domestic market (personal

280

6748 observation). Zambia reported its first detection of T. absoluta in May 2016 (IPPC, 2016a) while 6749 South Africa reported hers in September 2016 (IPPC, 2016b). Therefore, there are high chances 6750 that T. absoluta may have found its way into Botswana through tomato fruits, seedlings from its 6751 trade partners through e.g. packing materials, boxes, crates, pallets and to some extent outdoor 6752 markets selling the fruits from infested areas. Furthermore, the insect may have migrated across 6753 borders as flying adult moths. Given its high invasion potential, the moth has been reported to 6754 drift with wind currents (Desneux et al., 2010), fly up to 100 kilometres and move between non 6755 screened greenhouses and outdoor crops (CFIA, 2016), suggesting that they can move long 6756 distances in a country or across borders colonizing novel environments. The recent invasions by 6757 T. absoluta in Africa were reported in Tanzania (Chidege et al., 2016), Uganda (Tumuhaise et 6758 al., 2016), Zambia and South Africa (IPPC, 2016a, 2016b; Visser et al., 2017). Botswana 6759 therefore, records the latest detection of the invasive species in the southern African region. This 6760 implies that all the countries bordering Botswana e.g. Namibia and Zimbabwe, and others e.g. 6761 Mozambique, Malawi, Angola may be at significant risk. It may also imply that, the pest may 6762 have already invaded these countries but not yet reported hence the need for strict surveillance 6763 and quarantine regulations in these countries. Although tomato is the preferred host by T. 6764 absoluta, this invasion poses a threat to other solanaceaous crops in the country and region such 6765 as Solanum tuberosum, Capsicum annuum, Nicotiana tabacum and leguminous Phaseolus 6766 vulgaris. Given the socioeconomic values of these commodities to the African communities, the 6767 current T. absoluta invasion may have negative consequences on the agricultural export market, 6768 household and national food security and thus livelihoods. 6769 The detection of T. absoluta in Botswana specifically on tomatoes poses a threat to 6770 tomato production in the country and region thereby affecting food and nutritional security. This 6771 is because it prefers tomatoes to other solanaceous crops (Karadjova et al., 2013; Tumuhaise et 6772 al., 2016). Given the rate at which this pest is spreading across the continent (Pimentel et al., 6773 2001; Pfeiffer et al., 2013; Tumuhaise et al., 2016) and its potential damage on tomato plants, 6774 there is need to come up with effective management options to avoid further invasions. In this 6775 regard, I recommend effective monitoring of its spread, conducting pest risk assessments for the 6776 country and region, developing effective sanitary and phytosanitary measures, awareness 6777 campaigns, farmer trainings, countrywide surveys to determine pest free zones and introduction 6778 of biological control agents. In addition, coordinated efforts amongst stakeholders, research

281

6779 specialists and extension officers in Botswana and across the Southern African region should be 6780 employed to implement effective monitoring systems and area-wide pest management. 6781

6782 2.5 Conclusion 6783 I conclude that morphology of the male genitalia, and confirmation through molecular 6784 data, positive sex specific pheromone lure trapping and host plant damage symptoms all 6785 confirmed association with T. absoluta. Therefore based on these findings I confirm the first 6786 record of Tuta absoluta in Botswana. 6787

6788 References 6789 Abbes, K., Harbi, A. & Chermiti, B. (2012) The tomato leafminer Tuta absoluta, (Meyrick) in 6790 Tunisia: current status and management strategies. Bulletin OEPP/EPPO Bulletin, 42, 226- 6791 233 6792 Altschul, S.F., Gish, W., Miller, W., Myers, E.W. & Lipman, D.J. (1990 "Basic local alignment 6793 search tool." Journal of Molecular Biology, 215, 403-410. 6794 Anon. (2012) Tuta absoluta information network. http://www.tutaabsoluta.com/ 6795 Barrientos, Z.R., Apablaza, H.J., Norero, S.A. & Estay, P.P. (1998) Threshold temperature and 6796 thermal constant for development of the South American tomato moth, Tuta absoluta 6797 (Lepidoptera, Gelechiidae). Ciencia e investigación agrarian, 25, 133–7. 6798 Bawin, T., DeBacker, L., Dujeu, D., Legrand, P., Megido, R.C., Francis, F. & Verheggen, F.J. 6799 (2014) Infestation level influences oviposition site selection in the tomato leafminer, Tuta 6800 absoluta (Lepidoptera: Gelechiidae). Insects, 5, 877-884 6801 Brévault, T., Sylla, S., Diatte, M., Bernadas, G. & Diarra, K. (2014) Tuta absoluta Meyrick 6802 (Lepidoptera: Gelechiidae): A new threat to tomato production in Sub-Saharan Africa. 6803 African Entomology, 22, 441-444. 6804 Campos, M.R., Rodrigues, A.R.S., Silva, W.M., Silva, T.B.M., Silva, V.R.F., Guedes, R.N.C. & 6805 Siqueira, H.A.A. (2014) Spinosad and the tomato borer Tuta absoluta: a bioinsecticide, an 6806 invasive pest threat, and high insecticide resistance. PLoS ONE, 9, e103235 6807 CFIA. (2016) Tuta absoluta (Tomato Leafminer) - Fact Sheet. Government of Canada 6808 Publications. Ontario

282

6809 Chidege, M., Al-zaidi, S., Hassan, N., Julie, A., Kaaya, E. & Mrogoro, S. (2016) First record of 6810 tomato leaf miner Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae) in Tanzania. 6811 Agriculture and Food Security, 5, 17. DOI:10.1186/s40066-016-0066-4 6812 Desneux, N., Luna, M.G., Guillemaud, T. & Urbaneja, A. (2011) The invasive South American 6813 tomato pinworm, Tuta absoluta continues to spread in Afro-Eurasia and beyond: the new 6814 threat to tomato world production. Journal of Pest Science, 84, 403-408 6815 Desneux, N., Wajnberg, E., Wyckhuys, K.A.G., Burgio, G., Arpaia, S., Narva´ez-Vasquez, C.A., 6816 Gonza´lez-Cabrera, J., Catala´n Ruescas, D., Tabone, E., Frandon, J., Pizzol, J., Poncet, C., 6817 Cabello, T. & Urbaneja, A. (2010) Biological invasion of European tomato crops by Tuta 6818 absoluta: ecology, geographic expansion and prospects for biological control. Journal of 6819 Pest Science, 83, 197-215 6820 Dunstan, W.R. & Magazini, I.A. (1981) Outbreaks and new records, United Republic of 6821 Tanzania. The larger grain borer on stored products. FAO Plant Protection Bulletin, 29, 80- 6822 81. 6823 EPPO. (2016) EPPO Global Database. https://gd.eppo.int 6824 EPPO. (2005) Data sheets on quarantine pests: Tuta absoluta. OEPP/EPPO Bulletin, 35, 434– 6825 435. 6826 FAO. (2002) Comprehensive Africa Agriculture Development Programme (CAADP). NEPAD. 6827 Rome. Italy 6828 Felsenstein, J. (1985) Confidence limits on phylogenies: An approach using the 6829 bootstrap. Evolution, 39, 783-791.

6830 Ferracini, C., Ingegno, B.L., Navone, P., Ferrari, E., Mosti, M., Tavella, L. & Alma, A. (2012) 6831 Adaptation of indigenous larval parasitoids to Tuta absoluta (Lepidoptera: Gelechiidae) in 6832 Italy. Journal of Economic Entomology, 105, 1311–1319. 6833 Ferrara, F., Vilela, E.F., Jham, G.N., Eiras, A.E., Picanco, M.C., Attygale, A.B., Svatos, A., 6834 Frighetto, R.T.S. & Meinwald, J. (2001) Evaluation of synthetic major component of the sex 6835 pheromone of Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae). Journal of Chemical 6836 Ecology, 275, 907-917 6837 Folmer, O., Black, M., Hoeh, W., Lutz, R. & Vrijenhoek, R. (1994) DNA primers for 6838 amplification of mitochondrial cytochrome coxidase subunit I from diverse metazoan 6839 invertebrates. Molecular Marine Biology and Biotechnology, 3, 294–299.

283

6840 Gontijo, P.C., Picanco, M.C., Pereira, E.J.G., Martins, J.C., Chediak, M. & Guedes, R.N.C. 6841 (2013) Spatial and temporal variation in the control failure likelihood of the tomato leaf 6842 miner, Tuta absoluta. Annals of Applied Biology, 162, 50–59 6843 Harizanova, V., Stoeva, A. & Mohamedova, M. (2009) Tomato Leaf Miner, Tuta absoluta 6844 (Povolny) (Lepidoptera: gelechiidae) – First Record in Bulgaria. Agricultural Science and 6845 Technology, 1, 95 – 98. 6846 Hayden, J. E., Lee, S., Passoa S.C., Young, J., Landry, J.-F., Nazari, V., Mally, R., Somma, L.A. 6847 & Ahlmark. K.M. (2013) Digital Identification of Microlepidoptera on Solanaceae. USDA- 6848 APHIS-PPQ 6849 Hill, M. P., Bertelsmeier, C., Clusella-Trullas, S., Garnas, J. R., Robertson, M.P. & Terblanche J. 6850 S. (2016) Predicted decrease in global climate suitability masks regional complexity of 6851 invasive fruit fly species response to climate change. Biological Invasions, 18, 1105-1119. 6852 IPPC. (2016) Reporting pest presence: Preliminary surveillance reports on Tuta absoluta in 6853 Zambia. International Plant Protection Convention (IPPC). September 14, 2016. 6854 https://www.ippc.int/en/countries/zambia/pestreports/2016/09/reporting-pest-presence- 6855 preliminary-surveillance-reports-on-tuta-absoluta-in-zambia/. 6856 IPPC. (2016) First detection of Tuta absoluta in South Africa. International Plant Protection 6857 Convention (IPPC). September 1, 2016. https://www.ippc.int/en/countries/south- 6858 africa/pestreports/2016/09/first-detection-of-tuta-absoluta-in-south-africa 6859 IPCC. (2014) Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and 6860 III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Core 6861 Writing Team, R.K. Pachauri and L.A. Meyer (eds.)]. IPCC, Geneva, Switzerland, pp 151. 6862 Karadjova, O., Ilieva, Z., Krumov, V., Petrova, E. & Ventsislavov, V. (2013) Tuta absoluta 6863 (Meyrick) (Lepidoptera: Gelechiidae): Potential for entry, establishment and spread in 6864 Bulgaria. Bulgarian Journal of Agricultural Science, 19, 563-571. 6865 Kimura M. (1980) A simple method for estimating evolutionary rate of base substitutions 6866 through comparative studies of nucleotide sequences. Journal of Molecular Evolution, 16, 6867 111-120.

284

6868 Krechemer, F.D. & Foerster, L.A. (2015) Tuta absoluta (lepidoptera: Gelechiidae): thermal 6869 requirements and effect of temperature on development, survival, reproduction and 6870 longevity. European Journal of Entomology, 112, 658-663 6871 Lietti, M.M.M., Botto, E. & Alzogaray, R.A. (2005) Insecticide resistance in Argentine 6872 populations of Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae). Neotropical 6873 Entomology, 34, 113-119. 6874 Lux, S.A., Copeland, R.S., White, I.M., Manrakhan, A. & Billah, M.K. (2003) A new invasive 6875 fruit fly species from the Bactrocera dorsalis (Hendel) group detected in East Africa. Insect 6876 Science and its Application, 23, 355-361. 6877 Madisa, M.E., Obopile, M. & Assefa, Y. (2012) Analysis of horticultural production trends in 6878 Botswana. Journal of Plant studies, 1, doi:10.5539/jps.v1n1p25 6879 Manrakhan, A., Venter, J.H. & Hattingh, V. (2012) Action plan for the control of African 6880 invader fruitfly, Bactocera invadens Drew, Tsuruta and White, Citrus Research 6881 International. Department of Agriculture, Forestry and fisheries, South Africa 6882 Martins, J.C., Picanc, M.C., Bacci, L., Guedes, R.N.C., Santana Jr, P.A., Ferreira, D.O. & 6883 Chediak, M. (2016) Life table determination of thermal requirements of the tomato borer 6884 Tuta absoluta. Journal of Pest Science, 89, 897-908 6885 Mohamed, S.A. (2014) Development and implementation of a sustainable IPM and surveillance 6886 program for the invasive tomato leafminer, Tuta absoluta (Meyrick), in North and sub- 6887 Saharan Africa. Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH, 6888 Eschborn, Germany 6889 Neuenschwander, P. (2001) Biological control of the Cassava Mealybug in Africa: A Review. 6890 Biological Control, 21, 214–229 6891 Pfeiffer, D., Muniappan, R., Sall, D., Diatta, P., Diongue, A. & Dieng, E.O. (2013) First record 6892 of Tuta absoluta (Lepidoptera Gelechiidae) in Senegal. Florida Entomologist, 96, 661-662 6893 Pimentel, D., McNair, S., Janecka, J., Wightman, J., Simmonds, C., O’Connell, C., Wong, E., 6894 Russel, L., Zern, J., Aquino, T. & Tsomondo, T. (2001) Economic and environmental threats 6895 of alien plant, animal, and microbe invasions. Agriculture, Ecosystems and Environment, 84, 6896 1–20 6897 Quardi, K., Chouibani, M., Rahel, M.A., Andel Akel, M., 2012. Stratégie Nationale de lutte 6898 contre la mineuse de la tomate Tuta absoluta Meyrick. EPPO Bulletin 42, 281-290.

285

6899 Retta A. N. & Berhe, D.H. (2015) Tomato leaf miner –Tuta absoluta (Meyrick), a devastating 6900 pest of tomatoes in the highlands of Northern Ethiopia: a call for attention and action. 6901 Research Journal of Agriculture and Environmental Management, 4, 264–269. 6902 Roditakis, E., Vasakis, E., Grispou, M., Stavrakaki, M., Nauen, R., Gravouil, M. & Bassi, A. 6903 (2015) First report of Tuta absoluta resistance to diamide insecticides. Journal of Pest 6904 Science, 88, 9, doi:10.1007/s10340-015-0643-5 6905 Saccaggi, D.L., Karsten, M., Robertson, M.P., Kumschick, S., Somers, M.J., Wilson, J.R.U. & 6906 Terblanche, J.S. (2016) Methods and approaches for the management of arthropod border 6907 incursions. Biological Invasions. doi:10.1007/s10530-016-1085-6. 6908 Saitou, N. & Nei M (1987) The neighbor-joining method: A new method for reconstructing 6909 phylogenetic trees. Molecular Biology and Evolution, 4, 406-425. 6910 Shashank, P.R., Chandrashekar, K., Meshram, N.M. & Sreedevi, K. (2015) Occurrence of Tuta 6911 absoluta (Lepidoptera: Gelechiidae) an invasive pest from India. Indian Journal of 6912 Entomology, 77(4), 323-329 6913 Sint, D., Sporleder, M., Wallinger, C., Zegarra, O., Oehm, J., Dangi, N., Giri, Y.P., Kroschel, J. 6914 & Traugott, M. (2016) A two dimensional pooling approach towards efficient detection of 6915 parasitoid and pathogen DNA at low infestation rates. Methods in Ecology and Evolution, 6916 DOI: 10.1111/2041-210X.12621. 6917 Zhang, Z., Schwartz, S., Wagner, L. & Miller, W. (2000) A greedy algorithm for aligning DNA 6918 sequences. Journal of Computational Biology, 7, 203–214. 6919 Tams, W.H.T. (1932) New species of African Heterocera. Entomologist, 65, 1241–1249. 6920 Tonnang, H.E.Z., Samira, F. M., Khamis, F. & Ekesi, S. (2015) Identification and risk 6921 assessment for worldwide invasion and spread of Tuta absoluta with a focus on Sub-Saharan 6922 Africa: Implications for phytosanitary measures and management. PLoS ONE, 10, 6923 e0135283. doi:10.1371/journal. 6924 Tumuhaise, V., Khamis, F.M., Agona, A., Sseruwu, G. & Mohamed, S.A. (2016) First record of 6925 Tuta absoluta (Lepidoptera: Gelechiidae) in Uganda. International Journal of Tropical 6926 Insect Science, 36, 135–139, doi:10.1017/S1742758416000035 6927 Van den Broecke, G. & Maertens, M. (2016) Horticultural exports and food security in 6928 developing countries. Global Food Security, 10, 11-20

286

6929 Varela A. M., Seif, A. & Löhr, B. (2003) A Guide to IPM in Tomato Production in Eastern and 6930 Southern Africa. icipe. Science Press, Nairobi, Kenya. 144 pp. 6931 Visser D, Uys VM, Nieuwenhuis RJ, Pieterse W. First records of the tomato leaf miner Tuta 6932 absoluta (Meyrick, 1917) (Lepidoptera: Gelechiidae) in South Africa. BioInvasions, 2017;6 6933 6934 6935 6936 6937 6938 6939 6940 6941 6942 6943 6944 6945 6946 6947 6948

287