Biological parameters, life table and thermal requirements of Thaumastocoris peregrinus (Heteroptera: Thaumastocoridae) at different temperatures

Temperature affects the development, population dynamics, reproduction and population size of insects. Thaumastocoris peregrinus Carpintero et Dellape (Heteroptera: Thaumastocoridae) is a eucalyptus pest. The objective of this study was to determine biological and life table parameters of T. peregrinus on Eucalyptus benthamii at five temperatures (18 °C; 22 °C; 25 °C; 27 °C and 30 °C) with a relative humidity (RH) of 70 ± 10% and photoperiod of 12 hours. The duration of each instar and the longevity of this insect were inversely proportional to the temperature, regardless of sex. The nymph stage of T. peregrinus was 36.4 days at 18 °C and 16.1 days at 30 °C. The pre-oviposition period was 5.1 days at 30 °C and 13.1 days at 18 °C and that of oviposition was 7.6 days at 30 °C and 51.2 days at 22 °C. The generation time (T) of T. peregrinus was 27.11 days at 22 °C and 8.22 days at 30 °C. Lower temperatures reduced the development and increased the life stage duration of T. peregrinus. Optimum temperatures for T. peregrinus development and reproduction were 18 and 25 °C, respectively.

www.nature.com/scientificreports www.nature.com/scientificreports/ and fecundity 27 . Thus, the objective of this study was to evaluate the effect of different temperatures on biological parameters of T. peregrinus.
Survival analysis using the Cox's Proportional Hazards model showed a higher death risk (hazard ratio; HR) for nymphs and adults (females and males) of T. peregrinus as temperature increased (Table 3) and (Fig. 2).

Discussion
Temperature strongly influences insect development in both single generation progeny and in organisms that are established and successfully continued for multiple generations 28 . Thaumastocoris peregrinus development and reproduction reinforces the temperature effect on insects 29 , with the duration of its juvenile stage decreasing as temperature increases, as found for Corythucha ciliate (Say) (Hemiptera: Tingidae) and Loxostege sticticalis (L.) (Lepidoptera: Crambidae) 30 .
The shorter duration of each instar and of the adult period of T. peregrinus at higher temperatures is due to increased metabolism, food intake and energy, allowing the insect to reach the next stage 31,32 . Other factors, such as poor food quality 9,33,34 , decreased the survival and/or insect growth rate 35 . The ladybird Harmonia axyridis (Pallas) (Coleoptera: Coccinellidae) 36,37 , the dragonfly Ischnura verticalis (Odonata: Coenagrionidae) 38 , and the locust Romalea microptera (Orthoptera: Romaleidae) 39 had shorter juvenile stages at increased temperatures. www.nature.com/scientificreports www.nature.com/scientificreports/ Thaumastocoris peregrinus had a shorter pre-oviposition period with increased temperature, indicating the effect of this parameter on this organism. This is also reflected in the mating and egg laying of T. peregrinus as reported for Phenacoccus madeirensis Green (Hemiptera: Pseudococcidae) 40 and Leptocoris achinensis (Dallas) (Hemiptera: Alydidae) 41 and food/temperature and bioecology interactions 42 as reported for Cimex lectularius (Linnaeus 1758; Hemiptera: Cimicidae) 43 . The number of eggs per T. peregrinus female at 26 °C on Eucalyptus urophylla x Eucalyptus camaldulensis 5 and Eucalyptus scoparia at different temperatures 4 and with E. tereticornis at 25 °C 44 varied within certain limits 45 . The longer pre-oviposition period, at least for some T. peregrinus females at lower temperatures could be due to the longer time required for this predator to develop its ovary 46 .
Thaumastocoris peregrinus female and male longevity was increased at temperatures between 18 to 22 °C, which could be due to reduced metabolic processes at lower temperatures, affecting development and life history 47 . At low metabolic rates, certain physiological processes are suppressed, for example reproduction 48  The optimal temperature range for T. peregrinus development and reproduction between 25 and 30 °C was similar to those reported for egg, nymph and egg-adult periods, respectively, for this bug 44,51 , as well as for Nezara viridula (L.) (Hemiptera: Pentatomidae) collected in soybean fields at climatically different locations 28 . The linear increase in the ratio between instars and of the adult stage duration of T. peregrinus (1/D) confirms the energy gain for its physiological processes 52 .
The low survival at high temperatures indicates a phenotypic plasticity for T. peregrinus in different environments 53 .
The higher thermal constant of T. peregrinus nymph development, 338.50 DD with a minimum of 9 °C shows the impact of low temperatures on this insect 51,54 . This result was also observed for Axinoscymnus cardilobus (Ren and Pang) (Coleoptera: Coccinellidae), with 204 DD; it took 67 days at a minimum of 9.07 °C to complete one generation, while this was 120 days 55 at 17 °C. However, the accumulated temperature for the nymph-to-adult period of T. peregrinus, with 395.43 DD with a temperature limit of 9.93 °C shows its high adaptive potential. This species needed 905.65 DD in Canberra, Australia, to complete a generation and survived at temperatures below 1.5 °C, with adults recovering at higher temperatures 51 .
The e R 0 , rm, T and λ of T. peregrinus showed shorter development periods and higher growth rates with increased temperature, similar to that reported for Megacopta cribraria (F.) (Hemiptera: Plastaspidae) 56 and Jakowleffia setulosa (Jakovley, 1874) (Hemiptera: Lygaeidae) 57 . These characteristics are important to understand the impact of temperature on insect growth, survival, reproduction and population increase 58,59 . This is necessary because the energy generated by the anabolism and catabolism metabolic processes for insect growth and reproduction depends on the environmental temperature 60 .
The environmental temperature affected the development, fertility, longevity and mortality of T. peregrinus. Thus, the definition of thermal requirements for T. peregrinus can assist traditional techniques in managing this pest. As well, this important data can be used in simulating population dynamics, monitoring, population peaks, occurrence, ecological zoning and modeling in order to manage this pest.°C  www.nature.com/scientificreports www.nature.com/scientificreports/

Material and Methods
Insect rearing and temperature conditions. The experiments were conducted at the Forest Entomology Laboratory of Embrapa Florestas in Colombo, Paraná state, Brazil. Thaumastocoris peregrinus was reared in the laboratory at 24 ± 2 °C, 60 ± 10% RH, and a photoperiod of 12:12 h L:D on bouquets of Eucalyptus benthamii Maiden & Cambage (Myrtales: Myrtaceae) branches. The branches were fixed in a piece of foam to prevent drowning the insects in a 500-mL glass flask filled with water 61,62 . The effect of temperature on various biological parameters of T. peregrinus was evaluated at five constant temperatures (18,22,25,27  Nymph development. Newly hatched T. peregrinus nymphs were individually placed in acrylic plates (2.8 cm diameter × 1.5 cm) with a Eucalyptus benthamii fresh leaf disk (2.1 cm diameter) with its petiole introduced in a hydrogel layer (hydroplan-EB/HyC, SNF SA Floger) to maintain the leaf turgor. The eucalyptus leaf discs were replaced every two days. The duration and viability of T. peregrinus instars were assessed daily. Instar changes were evaluated based on the exuvia presence. Survival was evaluated in relation to the number of live individuals beginning each instar.
Adult reproduction and longevity. Thaumastocoris peregrinus adults (<24 h old) were sexed based on its morphological characteristics 6 . A couple of this insect was placed per Petri dish (5.0 cm in diameter) with a fresh E. benthamii leaf disc (4.9 cm diameter). The pre-oviposition (female emergence to the first egg laying) and oviposition period, fecundity (number of eggs per female per day), longevity and mortality of T. peregrinus were evaluated. The males were not replaced. Mortality data were used to calculate longevity. Females were maintained until death, and egg numbers were use in the analysis. Development thresholds and thermal constants. The temperature development threshold (Tt) and thermal constant (K) of T. peregrinus were estimated using the hyperbole method 63 , based on the duration of the different instars, the nymph stage and the egg-adult period at 18, 22, 25, 27 and 30 ± 2 °C. The T. peregrinus instar development rate and nymph-to-adult period was regressed against temperature using a linear equation given by  www.nature.com/scientificreports www.nature.com/scientificreports/ the formula: 1/D = a + bT, where, 1/D is the insect development time (D) in days, and T is the temperature (°C). The intercept ratio over the slope of the regression line corresponds to the threshold temperature (Tt) and the thermal constant (K) was estimated by taking the inverse of the slope (1/b) 64 . T. peregrinus fertility life table at each temperature was built with specific survival at age x (lx), specific fertility (m x ) and number of offspring reaching the age x in the next generation (l x m x ). These data were used to calculate the net reproductive rate (R 0 ), time between generations (T), innate ability to increase (r m ) and finite rate of increase (λ) of this insect 65 . Biological parameter analysis. All data were first analyzed using the Shapiro-Wilk and Bartlett tests to determine data normality and homogeneity. The data related to each instar duration and of the nymph-to-adult period did not conform to normality, even after log transformation. Therefore, the comparisons were made using the non-parametric Kruskal-Wallis test. Pre-oviposition, oviposition, fecundity, oviposition rate and female and male longevity data were normally distributed, and thus they were analyzed using a linear model followed by a post hoc pairwise comparisons performed using Tukey HSD test (function glht, package multcomp) 66 . Sex ratio was analyzed using a generalized linear model (GLM) assuming binomial distribution 67,68 . The analyses were performed with the software R, version 3.3.2. The fertility life table was analyzed by Jackknife and the averages compared by Student's t-test using the software SAS version 9.1 69 .

Life table analysis. The
Survival analysis. Survival curves were fitted and analysed using Kaplan-Meier survival probabilities (R version 3.3.2, "survival", "survminer" packages) 70,71 followed by a pairwise comparisons Mantel-Haenszel Test (Log-Rank test) and Cox Proportional-Hazard Model (PH Model). The data evaluated at 30 °C was used as the reference for the other treatments (temperatures) on Multivariate Cox regression. Individuals who did not die by the end of the nymph period were censored (0 = death event did not occur; 1 = death event occurred). The adult individuals were not censored, because the experiment finished with the death of all insects (females and males).