Introduction
Intraspecific variation occurs when a species' anatomy, physiology, behavior, and social organization change as it adjusts to changing environmental conditions by modifying its dietary behaviors (Wcislo, Reference Wcislo1989). Such variation may be directly related to genetic differences between individuals within a species (Honěk, Reference Honěk1993). The development of individuals within a population plays an important role in regulating the population (Sih et al., Reference Sih, Cote, Evans, Fogarty and Pruitt2012). But regardless of the suitable abiotic and biotic conditions, variations in the development rate have been found in many insect predators (Pandey et al., Reference Pandey, Mishra and Omkar2013). This variation in development rate within an egg batch is termed developmental rate polymorphism (DRP) and has been a source of fascination for researchers. The existence of DRP under optimal conditions in an aphidophagous ladybirds, Menochilus sexmaculatus (Fabricius) (Singh et al., Reference Singh, Mishra and Omkar2016) and Propylea dissecta (Mulsant) (Siddiqui et al., Reference Siddiqui, Omkar and Mishra2017) and Parthenium beetle, Zygogramma bicolorata (Pallister) (Pandey et al., Reference Pandey, Mishra and Omkar2013; Afaq et al., Reference Afaq, Kumar and Omkar2021) reveals its genetic regulation under the influence of abiotic factors on developmental variants but biotic factors are still unexplored, especially to relative prey abundance on selected lines. The differential developmental rates under differing conditions exist commonly but the existence of two rates of development in a cohort under each environmental condition is still a mystery. This is possibly a way for a particular species to escape from unfavorable environmental conditions, or it could be the emerging individual's genetic or predetermined perspective. Studies have also revealed that such populations are examined across a number of generations under defined conditions that may be repeated, whether in a laboratory settings or in the nature (Rajpurohit et al., Reference Rajpurohit, Peterson, Orr and Marlon2016; Bono et al., Reference Bono, Smith, Pfennig and Burch2017).
In ladybirds, developmental variants (slow and fast developers) are found in each cohort (Singh et al., Reference Singh, Mishra and Omkar2016), and are governed by genetic factors (Bailey and Bataillon, Reference Bailey and Bataillon2016). The development is species-specific and strongly dependent on the ambient temperature (Afaq et al., Reference Afaq, Kumar and Omkar2021), photoperiod (Bono et al., Reference Bono, Smith, Pfennig and Burch2017), population density (Ungerová et al., Reference Ungerová, Kalushko and Nedvĕd2010), and quality and quantity of food (Singh et al., Reference Singh, Mishra and Omkar2016). However, the ability of a predator to survive, develop, and reproduce in prey-scarce conditions is the most important aspect that determines the fitness of their own immature stages during development (Singh et al., Reference Singh, Mishra and Omkar2016) and its biocontrol potential (Siddiqui et al., Reference Siddiqui, Omkar, Paul and Mishra2015, Reference Siddiqui, Omkar and Mishra2017). The prey consumption seemingly decreases with the increase in developmental duration (Siddiqui et al., Reference Siddiqui, Omkar and Mishra2017; Pervez and Sharma, Reference Pervez and Sharma2021). The fast-growing individuals are more vulnerable to starvation owing to their need to sustain higher metabolic rates (Sundström and Devlin, Reference Sundström and Devlin2011). The optimal growth rate is often lower than the maximum rate achievable indicating that rapid growth is costly (Metcalfe and Monaghan, Reference Metcalfe and Monaghan2001). Also, rapid growth may be associated with reduced developmental control and an increased developmental error (Nylin and Gottard, Reference Nylin and Gottard1998).
If the costs of maintaining conversions are significant (Edelaar et al., Reference Edelaar, Piersma and Postma2005), selection will quickly remove it from an environment with a little relevance, however, there are substantial demands on resource allocation other than maintaining plasticity levels. In spite of having been so many works on ladybirds, the effect of abiotic and biotic factors on selection is still unexplored. Therefore, the present study was conducted to evaluate the food allocation strategy on scarce and abundant prey supplies among the intraspecific control (pre-selected/F 1) and (post-selected/F 15) selected individuals of P. dissecta whether the control variants (slow/fast) exploit more food than that of post-selected one or their consumption remains constant after a selection process. The results will help understand the strategies of control and selected slow and fast developers and helping in the mass multiplication of predatory ladybirds for their use in biocontrol of various pest species.
Material and methods
Stock maintenance
The wild P. dissecta population was taken from agricultural regions near Lucknow, Uttar Pradesh, India (16.8470°N, 80.9470°E) to be used as laboratory stock. Under standard laboratory settings [27 ± 1°C temperature; 65 ± 5% relative humidity and 14L:10D photoperiod in a BOD Incubator (YORCO; York Scientific Industries Pvt. Ltd., India)], they were raised in clear plastic Petri dishes (9.0 × 2.0 cm) and fed ad libitum pea aphid Acyrthosiphon pisum (Harris) (Hemiptera: Aphididae) (reared under standard greenhouse condition on their mentioned host plant) bred on the broad bean. After an initial mass collection, the stock was continually replenished with wild captured individuals throughout the season to minimize inbreeding. In the BOD incubator, mature males and females were coupled in Petri dishes (size and conditions as above). The females mated and laid eggs, which were separated every 24 h and the prey were replenished. Under aforementioned lab conditions, the newly hatched larvae were individually reared for further experimental setups.
Separation of lines of slow and fast developers
Into individual Petri dishes (size as above) 100 virgin males and females, each were taken from the outbred laboratory stock, were placed along with ad libitum of A. pisum. To prevent egg cannibalism, a single batch of eggs was gathered from each mate. They were observed every 12 h for hatchings. The newly hatched first instars were separated and reared individually till adults emerged (under the aforesaid laboratory settings). Based on their total developmental duration (from egg to adult), the emerging adults were divided into two categories: slow developers and fast developers. Furthermore, ‘slow developers’ have long total developmental periods and ‘fast developers’ have a short developmental period. Using a random breeding method to minimize pairing between siblings or close relatives, two distinct developmental lines were formed by mating slow–slow developers and fast–fast developers for up to ten generations (Swallow and Garland, Reference Swallow and Garland2005). Following that, mixed matings were performed among individuals of either developmental variation to minimize inbreeding depression (slow or fast developers; Swallow and Garland, Reference Swallow and Garland2005), for an additional five generations. Newly emerged from slow and fast developing adults in the F 15 generation (experimental generation) and the F 1 generation (control) were then used to evaluate the following mentioned parameters. After 6 h, the weight of the newly emerged adults was recorded. The percent of immature survival (number of surviving out of a total number of eggs), slow–fast emergence ratio (number of slow or fast developers/total number of individuals emerged), and fecundity with percent egg viability (10 days old) were calculated in F 1 and F 15 generation of both slow and fast developers.
Experimental design
For the study, we followed the standardized prey quantity and experimental regime for prey-scarce and prey-abundant conditions for P. dissecta by Singh et al. (Reference Singh, Mishra and Omkar2016). Ten-day-old unmated adults (ten pairs) from each of F 1 (control) and F 15 (selected) generation of each developmental variant were paired in separate plastic Petri dishes (9.0 × 2.0 cm) and placed on different prey quantities, viz. prey-scarce (3–5 second and third instars of A. pisum per day) and prey-abundant (25–30 second and third instars of A. pisum per day) conditions. A total of 250 eggs from the first 5 days of oviposition of both generations (F 1 and F 15) of each developmental variant on each prey quantity were selected. Hatched instars were individually reared in Petri dishes (9.0 × 2.0 cm) on the same prey quantity as provided to their parents till adult emergence. They were observed twice a day for survival and molting. After 6 h, the weight of the newly emerged adults weighed using electronic balance (Sartorius CP225-D; 0.01 mg precision) was recorded. The percent of immature survival (number of surviving out of a total number of eggs) and slow-fast emergence ratio was calculated for both F 1 and F 15 generations of each variant. The newly emerged adults (10 days old) of each type, i.e. slow and fast developers, were paired in Petri dishes (size as above) and provided with the same prey supply on which they had completed development. Daily oviposition was recorded for the next 20 days and egg viability was recorded in mating pairs from each type (i.e. slow and fast) for both control and selected lines, respectively.
Statistical analysis
To check for normal distribution, data on total developmental periods of variations (egg to adult) were subjected to the Kolmogorov–Smirnov test of normality. The total developmental duration of slow and fast developers of P. dissecta of F 1 and F 15 generation when fed on scarce and abundant prey quantity was not normally distributed (table 1). The frequency data of the developmental durations were then graphed to show distribution patterns, which was found to be bimodal (fig. 1).
Frequency distribution of total developmental duration (TDD) of (a) control (F 1) and (b) selected line (F 15) of slow and fast developers of P. dissecta at scarce and abundant prey supply. Bars indicate number of individuals emerging at each developmental duration. ‘F’ indicates fast developers while ‘I’ and ‘S’ indicate intermediate and slow developers, respectively.
Normality of developmental duration of variants of P. dissecta on scarce and abundant prey supply
All percent data were subjected to arcsine square root transformation before further analysis. General linear MANOVA was conducted with generation (F 1 and F 15), prey quantity (abundant and scarce), developmental variant (slow/fast), acting as independent factors and developmental duration, adults body mass as a dependent factor. The χ 2 ‘goodness of fit’ analysis was used for the comparison of emergence ratio of slow and fast developers, survival for both control, and selected line of slow and fast developers. Means were compared using post hoc Tukey's honest significance test at 5% levels. All statistical analyses were performed using MINITAB 15.0. Mortality value or ‘k’ value was calculated from life table attributes following Morris and Miller (Reference Morris and Miller1954) and Southwood (Reference Southwood1978).
Results
The fecundity of selected fast developer females was higher than those of control fast developer females on abundant prey supply. Further analysis showed maximum fecundity by selected fast developers on abundant prey supply and minimum by selected slow developers on scarce prey supply (table 2). The interactions between developmental variants and generation and prey quantity and developmental variants were significant but the interaction between prey quantity and generation along with the interaction of all three independent factors were insignificant (table 2).
Reproductive attributes of developmental variants of control and selected line of P. dissecta on scarce and abundant prey supply
Values are mean ± SE.
Small letters represent comparison of means between slow–slow and fast–fast developers of both generations on each prey species.
Capital letters represent comparison of means between slow and fast developers of control and selected line within a prey species.
Percent egg viability of control and selected developmental variants was higher in selected fast developers on abundant prey supply and lower in control slow developers on scarce prey supply. The MANOVA results also revealed a significant influence of prey quantity generation, and developmental variants on percent egg viability while the interactions between developmental variants and generation and prey quantity and developmental variants were significant but the interaction between prey quantity and generation along with the interaction of all three independent factors were insignificant (table 2).
The maximum selected slow developers were recorded when the larvae were fed on scarce and minimum in slow developers from control on abundant prey supply. Further, maximum selected fast developers were recorded on abundant diet and minimum unselected fast developers on scarce prey (fig. 2).
Effect of scarce and abundant prey supply on slow:fast emergence ratio of control (F 1) and selected line (F 15) of P. dissecta.
The percent immature survival was highest in selected fast developers than unselected fast developers on both prey supplies while the immature survival was suppressed after selection in slow developers. The data revealed maximum immature survival of selected fast developers on abundant while minimum by control fast developers on scarce prey supply (fig. 3). The χ 2 revealed a significant difference between slow/fast developers of both control and selected line on scarce and abundant diets, respectively.
Effect of scarce and abundant prey supply on immature survival of control (F 1) and selected line (F 15) of slow and fast developers of P. dissecta.
Tukey's post hoc comparison of individual means showed a statistically significant difference between the total developmental duration of control and selected developmental variants. The longest duration was of selected slow developers on a scarce diet while shortest was of selected fast developers on an abundant diet. General linear MANOVA revealed a statistically significant effect of prey quantity, generations (F 1 and F 15), and developmental variants (slow and fast developers) on the total developmental duration (days). Interactions between prey quantity and generation, developmental variants and generations, prey quantity and developmental variants all three independent factors were significant (table 3).
Total development duration and body mass of slow and fast developers of control and selected line of P. dissecta on scarce and abundant prey supply
General linear MANOVA shows effect of prey quantity, generation and developmental variants and their interactions on these parameters.
Values are mean ± SE.
Small letters represent comparison of means between slow–slow and fast-fast developers of both generations on each prey species.
Capital letters represent comparison of means between slow and fast developers of control and selected line within a prey species.
The fast developers showed the heaviest adult body mass when fed on abundant diet and lowest by slow developers on scarce diet irrespective of generation (table 3). Comparison of means revealed that selected fast developers were heavier than other developmental variants of both F 1 and F 15 generations. The interactions between prey quantity and generation; developmental variants and generations; prey quantity and developmental variants; amid interactions between all three independent factors were significant (table 3).
Comparison of kappa (k) value revealed the highest value for F 1 on scarce diet (0.200) > F 1 on abundant diet > F 15 on scarce diet > F 15 on abundant diet (fig. 4). However, the mortality values or ‘k’ values were lowest for the selected individuals on abundant diet and highest for control individuals on a scarce diet.
Kappa value of immature stages of control (F 1) and selected lines (F 15) of P. dissecta on scarce and abundant prey supply.
Discussion
The results revealed significant differences between the control individuals of P. dissecta than those selected after 15 generations. The results indicate the presence of developmental variants in F 1 and F 15 generation on a scarce and abundant diet. Prey quantity significantly influenced the developmental duration, reproduction, and mortality value of both generations’ developmental variants.
Divergence in a number of emerged individuals was significant but not in the emergence ratio, indicating major differences in control and selected variants. During the development of P. dissecta, some larvae develop fast and some develop slowly. A high metabolic rate is linked with a short developmental period and high fecundity (Hoffmann and Parsons, Reference Hoffmann and Parsons1989). The study reveals that minimum biomass is required to completely develop earlier than slow developers (Huges, Reference Huges1980).
From the ecological point of view, the reason for fast emergence might be to minimize local extinction by catastrophic events (Thomas et al., Reference Thomas, Elmes and Wardlaw1998). On a physiological basis this might be due to hatching asynchronization (Osawa, Reference Osawa1992), eggs with different metabolic rates due to allelic differences (Sloggett and Lorenz, Reference Sloggett and Lorenz2008), and/or mother laying eggs with different sizes and nutritional content (Hodek et al., Reference Hodek, Van Emden and Honek2012). Thus, seemingly some unknown mechanisms operate during the development of P. dissecta that enhance or inhibit the pace of development so it continues throughout the generational rearing. This shows that an egg batch of P. dissecta possesses selectable genetic variation for the developmental duration. This might be the product of differential environmental induction of genomic programs that guide trade-offs allocated toward different traits during development in response to intrinsic and extrinsic cues (Snell, Reference Snell-Rood2013).
Previous studies in other insects like bean weevil, Acanthoscelides obtectus (Say) (Darka and Nikola, Reference Darka and Nikola2013) and flour beetles, Tribolium castaneum Herbst, and T. confusum (Giraldeau and Caraco, Reference Giraldeau and Caraco2018) also confirm the effect of relative prey abundance on various parameters. Similar findings were also reported in the European cabbage butterfly, Pieris rapae (Linnaeus) in which fast-developing larvae were often less parasitized than slow-developing ones, and were no evidence for a positive relationship between development time and the incidence of parasitism (Benrey and Denno, Reference Benrey and Denno1997), myrmecophilous butterfly, Maculinea rebeli (Hirchke) (Thomas et al., Reference Thomas, Elmes and Wardlaw1998), Indian meal moth, Plodia interpunctella (Hubner) (Naeemullah and Takeda, Reference Naeemullah and Takeda1998), maturing worms (Skorping, Reference Skorping2007), and ladybirds, M. sexmaculatus (Singh et al., Reference Singh, Mishra and Omkar2016) and P. dissecta. Thus, the present study revealed a significant effect of prey supply on slow and fast developers in the F 1 and F 15 generations. Hoffmann and Parsons (Reference Hoffmann and Parsons1989) reported in Drosophila melanogaster Meigen that lines selected for increased resistance to many environmental stresses have lowered metabolic rate and behavioral activity levels.
Previous studies indicated a similar ratio of slow and fast developers in coccinellids, P. dissecta and M. sexmaculatus (Singh et al., Reference Singh, Mishra and Omkar2016), and Chrysomelidae, Z. bicolorata Pallister (Afaq et al., Reference Afaq, Kumar and Omkar2021) on standard and variable abiotic conditions. Our study reveals a maximum number of selected slow developers on scarce diets and selected fast developers on abundant diets in both F 1 and F 15 generations. This shift probably allows individuals to maximize fitness by allocating resources differentially among phenotypic traits (Kasumovic and Hall, Reference Kasumovic, Hall, Try and Brooks2011). The variation in the number of slow and fast developers within diets possibly indicates the increased mortality of a particular development stage on each diet but the ratio remains constant in F 1 and F 15 generation, which might be a kind of individual strategy of resource allocation in accord to the recent situation. This may probably be due to the increased efficiency of slow developers in attaining the minimum weight required for successful completion of development than fast developers (Huges, Reference Huges1980). The reduced efficiency of selected slow developers might be due to attaining the threshold weight for achieving the next developmental stage and/or the higher sensitivity to starvation, which may result in their higher mortality (Rotkopf et al., Reference Rotkopf, Alcalay, Bar-Hanin, Barkae and Ovadia2013) and low body mass in a selected line. Previous studies also showed increased conversion efficiencies at low levels of food consumption were also reported in aphidophagous mirids and chrysopids (Zheng et al., Reference Zheng, Hagen, Daane and Miller1993).
A strong positive correlation between fecundity and egg viability in both abundant and scarce prey conditions was recorded in this study, i.e. when prey quantity was reduced, and then the fecundity and percent egg viability were also reduced. This negative influence of reduced prey quantity can be attributed to the availability of decreased nutrient resources, which restrict the development and reproduction of the predator (O'Brien et al., Reference O'Brien, Boggs and Fogel2005). Overall fecundity of developmental variants of F 1 and F 15 generation was low under scarce prey supply, which can be imputed to reduced nutrient resources hindering the development as well as the reproduction of the ladybirds (Majerus, Reference Majerus1994; Moczek, Reference Moczek1998) but as a generational effect, this was minimum in selected slow developers and unselected fast developers. This might be due to the gradual slow process of ovariole development (Evans, Reference Evans2003) and resorption of eggs (Cope and Fox, Reference Cope and Fox2003) in selected slow developers or might be due to low consumption rate than that of fast developers. Besides, Reznik and Vaghina (Reference Reznik and Vaghina2013) reported that nutrients (quality and quantity of prey) affect the rate of reproductive maturation and fecundity in Harmonia axyridis Pallas. Maximum fecundity in selected fast developers might be due to improved strain due to the selection process. Since food quantity greatly influences the intrinsic growth and reproductive rates of ladybirds (Lawo and Lawo, Reference Lawo and Lawo2011), selected individual of fast developers of P. dissecta showed the highest oviposition and egg viability under abundant prey conditions. Thus, it may be inferred that the life history traits change in response to nutrient stress. However, the successful development of both larvae and adults under food-stressed conditions suggest the occurrence of strong selection pressure in the natural population of ladybird beetle for survival and reproduction even under adverse conditions.
The higher egg viability by selected fast developers may be due to the large size of males (unpublished data) that possibly supply higher ejaculate, the better quality of genes in addition to accessory gland proteins (Helinski and Harrington, Reference Helinski and Harrington2011). This might be due to the effect of continuous rearing of identical lines. Lower fecundity in both F 1 and F 15 generation but enhanced egg viability in selected slow developers might be due to accelerated fitness shift for healthier parts and survivorship of slow developers. Studies also revealed that reduced egg viability under prey-scarce conditions as recorded in the present investigation may be attributed to (i) reduction in sperm or ancillary fluid production, which in turn might limit female reproductive output (Droney, Reference Droney1996), and (ii) reduced sperm production due to slow spermatogenesis and a lower rate of sperm survival in the males (Ponsonby and Copland, Reference Ponsonby and Copland1998).
According to Schuder et al. (Reference Schuder, Hommes and Larink2004), slowing down larval development during food-scarce conditions is one of the several mechanisms (like increase in conversion and exploitation efficiencies, etc.) displayed by larvae to compensate for a lack of food. Studies have shown that if food scarcity occurs before attaining the critical weight, several species extend their last larval duration beyond normal lengths (Nijhout et al., Reference Nijhout, Davidowitz and Roff2006). Thus, the longer developmental durations on scarce prey in selected slow developers in the present study may be due to the following reasons: (i) lengthening of developmental duration as a mechanism that allows the individuals to extend their feeding activity and acquire more opportunities to find and consume the necessary amount of prey or non-prey food to easily reach the critical weight (Shafiei et al., Reference Shafiei, Moczek and Nijhout2001) and (ii) if the fluctuation in food quantity occurs after attaining the critical weight, then the insects cease their growth without change in developmental duration and form resting stage, a phenomenon known as ‘determinate development’ (Nijhout et al., Reference Nijhout, Davidowitz and Roff2006). In the present study adults resulting from a scarce prey had smaller body mass while those fed on abundant diet result in larger adults. It is supported by many workers (Agarwala et al., Reference Agarwala, Yasuda and Sato2008). But body mass was enhanced in selected fast developers and depressed in selected slow developers which might be due to continuous cross-mating between slow–slow developers which results in low-weight offspring. Thus, more slow developers with a smaller adult's body mass that emerge under prey-scarce conditions probably have a lower reproductive success based on their smaller size, as suggested by many earlier researchers (Omkar and Afaq, Reference Omkar and Afaq2013).
Minimum mortality on an abundant diet is probably due to availability of enough prey resources to exploit and reach the minimum threshold weight necessary for changing into the next developmental stage. Also, evolutionary theory illustrates that fast development occurs under suitable conditions and slow development occurs under adverse conditions (e.g. Chown and Gaston, Reference Chown and Gaston2010). On scare diet, the mortality value of selected individuals was higher than that of selected individuals on abundant diet which was probably due to gradual acclimatization or transfer of unfavorable conditions in terms of food supply from grandparents to their grand progeny through some unknown genetic cues.
Conclusions
The study revealed that slow and fast developers are present at both scarce and abundant diets in F 1 and F 15 generation with the discrepancy in emerger's number, fecundity, percent egg viability, and mortality. On a scarce diet, selected slow developer promotes the survival of ladybirds. The slow developers were higher in number on a scarce diet in F 15 generation. However, on the abundant diet, the selected fast developers were higher in number, and developmental duration of immature stages on scare diet was high as compared to abundant diet (personal observation). This knowledge will help to understand that within an egg batch different rate polymorphisms were also found after being selected for 15 generations on both diets. Our results demonstrate that both the nutrient and selected line of variants affect the expression of the development rate, fecundity, and mortality of selected traits, but more importantly, that the environmental effects interact in complex ways with evolution experiments. Consequently, the food exploitation strategy was modified accordingly to generational rearing and intraspecific allocation for survival. Our study will provide a source of fascination for a number of evolutionary biologists. The selection of slow and fast developers of P. dissecta acts as a possible genetic tool for overall quality improvement and survival strategy for their offspring.
Acknowledgements
Siddiqui and Mishra would like to express their gratitude to the Department of Science and Technology in New Delhi, India, for the financial support provided under the Fast Track Young Scientist Scheme. Omkar is grateful to the Department of Higher Education, Government of U.P., Lucknow, India, for funding this project under the Centre of Excellence initiative.
Conflict of interest
None.
Ethical standards
None.
