Adaptive capacity of environmental stress resistance traits in Bactrocera dorsalis – expanded methods and materials

Anandi Bierman¹, Henriek Bosua¹, Minette Karsten¹, Carla M. Sgró², Kevin Malod³, Christopher Weldon,⁴ Nikos Papadopoulos⁵ & John S. Terblanche¹

1 – Centre for Invasion Biology, Department of Conservation Ecology & Entomology, Stellenbosch University, South Africa
2 – School of Biological Sciences, Monash University, Clayton, VIC, Australia
3 – Department of Conservation Ecology & Entomology, Stellenbosch University, South Africa
4 – Department of Zoology and Entomology, University of Pretoria, Pretoria, South Africa
5 – Laboratory of Entomology & Agricultural Zoology, Department of Agriculture Crop Production and Rural Environment, University of Thessaly, Volos, Greece

Corresponding author: John S. Terblanche –  jst@sun.ac.za

MATERIAL AND METHODS

Figure 1: Schematic diagram of study design depicting the separation of males and females from the starting population and the initiation of breeding cages or breeding lines at 20 days old. Breeding lines consist of 3 females and 1 male. Female offspring (1147 individuals) from each breeding line (102 in total) were used for measuring heat knock down time (HKDT), chill coma recovery time (CCRT) and desiccation resistance.

STUDY DESIGN
Pupae from a one year old laboratory colony (10 to 11 generations in laboratory culture) that is regularly supplemented with wild-caught individuals trapped in the region were obtained from Citrus Research International (CRI) (Nelspruit, Mpumalanga, South Africa). Adults were separated by sex within 24 hours of emergence to prevent mating. Virgin cohorts were housed in 30x30x30cm mesh cages (BugDorm) and were provided with a 50:50 mixture of granular sucrose (sugar cane) and brewer’s yeast (Sigma-Aldrich cat. 51475). Water was provided via damp cotton wool. The cages were housed inside climate controlled rooms at 26 °C and high relative humidity (RH).

Once sexual maturity (20 days after emergence) was attained, one virgin male and three virgin females were paired together in a mating cage. A 1:1 male:female ratio in the mating cages yielded very low egg production and a more optimum grouping of 1:3 male:female was used for this experiment. Mating cages consisted of 125mL transparent, plastic take-away cups with a mesh covering and food and water were supplied ad libitum. After a 2 day acclimation and mating period inside the cups at 26 °C and high relative humidity (40-60% RH), oviposition dishes were placed inside each mating cup for 48 hours before they were removed and replaced with a new oviposition dish (total of 2 egg collections per mating line). Egg dishes were prepared from 5mL lids filled with 2mL of diluted orange essence (1:10 v/v orange essence and distilled water) and covered with 2 sheets of parafilm (Lasec) that were pierced ~10 times with an insect pin. All eggs that were collected in the egg dishes were transferred via sterile 3mL Pasteur pipettes onto 90mL of prepared artificial larval diet supplied by CRI (ingredients: desiccated carrot, sucrose, yeast, prepared by mixing 125mL of dried powder with 200ml of boiling water). Larval diet cups with eggs were placed on top of 2cm of sterile sand in 1L plastic containers with a mesh lid for ventilation. Pupae were sifted from the sand of each container 14 days after egg dishes were sealed in the containers. The collected pupae from each cage were transferred to 3L clear plastic cages with a mesh sleeve and food and water supplied for the emerging adults. Upon emergence males were removed and discarded and the date of emergence recorded. A total of 1147 females consisting of 102 paternal lines were used for this study. Each of the 102 paternal lines (hereafter referred to as Lines) consisted of up to 20 female offspring, related as half-siblings by a shared sire and one of three females.

HEAT KNOCKDOWN
From at least 20 lines, 5-10 females were randomly selected at 8-10 days old. Selected individuals were transferred to pre-weighed and labelled 2.0mL microcentrifuge tubes and body mass of each fly was determined to 0.1mg ( Model: K25-cc-NR, huber, Germany, mass of the empty microbalance tube were subtracted from full tube). A programmable waterbath (Huber refrigerated bath circulator CC-410, USA) was set at 44°C and a floating platform of temperature conducting materials (aluminium foil and polystyrene) was floated on top of the water. The microcentrifuge tube containing flies were placed on top of the floating platform and covered with a transparent perspex sheet to limit heat loss while enabling the observer to have a clear line of sight. To monitor the precise temperature experienced by the flies inside the tube, the end of a fine gauge type T thermocouple was placed in an empty microcentrifuge tube, the tube sealed and the temperature inside the empty tube monitored and recorded with a thermocouple data logger (Pico Technologies, USB TC-08; UK). The microcentrifuge tubes containing flies and the empty tube with the thermocouple was placed on top of the floating platform at the same instant that a timer was started. The flies were observed inside the microcentrifuge tubes and the moment they lost all co-ordinated muscle movement, showed heat stupor and/or fell over and lost the ability to cling to surface of tubes the time was recorded in seconds as the heat knock down time (HKDT).

CHILL COMA RECOVERY TIME
From at least 20 lines, 5-10 females were randomly selected at 8-10 days old. Selected individuals were transferred to pre-weighed and labelled 2.0mL microcentrifuge tubes and body mass of each fly was determined to 0.1mg ( Model: K25-cc-NR, Huber, Germany, mass of the empty microbalance tube were subtracted from full tube). An empty microcentrifuge tube containing a type T thermocouple recorded the temperature via a thermocouple data logger (Pico Technologies, USB TC-08; UK). An ice-slurry (0°C) was prepared inside an airtight ice box and the microcentrifuge tubes with flies and tube with thermocouple was placed inside sealed waterproof bags (Ziploc bags) and placed inside the ice slurry for 1 h. After 1 h all the tubes were removed from the ice slurry and placed on a counter in a climate controlled room at 25 °C and a timer started. The flies were positioned so that they were on their backs inside the tubes and were carefully observed. As soon as a fly regained full muscle use and could right itself, the time was recorded in seconds and taken as the chill coma recovery time (CCRT).

DESICCATION
From at least 20 lines, 20 females aged 8-10 days were selected and the initial mass determined to 0.1mg.
Flies were placed in 2.0mL microcentrifuge tubes with 8 small air-holes pierced in the sides. These labelled, fly containing tubes were then transferred to airtight containers with a layer of silica gel at the bottom that absorbed moisture and created a desiccating environment (< 10% RH). The relative humidity of the containers was monitored with Hygrochron iButtons® (model no. 189-7734 DS1923-F5#; Maxim Integrated). Once placed inside the desiccation chambers flies were checked every 3 h, dead individuals were removed and the time of death recorded in hours from initial onset of experiment. Dead individuals were immediately weighed to 0.1mg to obtain the body mass at death and then transferred to a drying oven set at 65°C for 72 h. The dried flies were re-weighed to obtain dry body mass. Initial body water could be calculated as Initial body mass – Dry body mass. The water loss at death was determined as the Initial body mass – Body mass at death.

DATA ANALYSIS
All statistical analyses were conducted using R (R version 4.1.3 (2022-03-10)). To estimate additive genetic variance, underlying thermal performance and desiccation traits we utilised a restricted maximum-likelihood based ‘animal model’ through the ASReml-R package (Butler et al. 2017). Analyses were run on thermal performance traits (time to knock down upon heat exposure (HKDT) and righting response time upon cold exposure (CCRT)) as well as desiccation traits (Initial Mass, Mass at death, Dry Mass, Time to death, Mass loss at death, Percentage Mass loss, Initial body water and Water loss rate). All trait values were log-transformed to conform to the assumption of normality except for the desiccation traits of Mass loss at death and Water loss rate. Estimates for additive variance (VA) and maternal variance (VM) were obtained by including ‘Individual’ and ‘Line’ as random effects in the animal model, respectively. The model was run a second time with the addition of ‘initial mass’ as fixed effect for both additive variance (VA) and maternal variance (VM). In ‘animal models’ failure to control for fixed effects such as specific environmental conditions or variability in physical characteristics between test organisms (eg. age, sex, mass) can result in bias in the residual variance (VR) (Wilson et al. 2010). In this instance all test animals were female and the variability in mass was selected as fixed effect since sex and age were not variable.

Narrow sense heritability (h2) and standard error was calculated using the vpredict function in ASReml-R. The significance of VA (with individual as random effect) and VM (with ‘Line’ as random effect) was assessed using the z-ratio as calculated in ASReml-R. The z-ratio is defined as the ratio of VA or VM to its standard error and when this value is larger than 2, the parameter estimate is more than 2 standard errors from 0 and is therefore significant. Residual diagnostic plots were generated using the plot function in ASReml-R, to detect departures from the residual assumptions of normality. Pair-wise genetic covariances and correlations were estimated by running separate bivariate models for each pairwise trait combination of desiccation traits.

REFERENCES

Butler, D.G., Cullis, B.R., Gilmour, A.R., Gogel, B.J., Thompson, R. 2017. ASReml-R reference manual version 4. VSN International Ltd, Hemel Hempstead, HP1 1ES, UK.