Transgenerational effects of multiple mating in Spodoptera litura Fabricius (Lepidoptera: Noctuidae)

Abstract Polyandrous mating can result in sexual conflict and/or promote the evolution of mating patterns. Does multiple mating by females support the genetic benefits hypothesis and can it be validated as an evolutionary strategy? If we are to decipher the consequences of sexual interactions and understand the interplay of sexual conflict and multiple generational benefits, the transgenerational effects need to be followed over multiple generations. We investigated the effects of three mating patterns, single mating, repeated mating, and multiple mating, on parental Spodoptera litura copulation behavior, and then identified the impact on the development, survival, and fecundity of the F1 and F2 generations. Fecundity was not significantly affected in the F1 generation but was substantially enhanced in the F2 generation. There was a reversal of offspring fitness across the F2 generations from the F1 generations in progeny produced by multiple mating. In addition, the intrinsic rate of increase, finite rate of increase and net reproductive rate in the F1 generation the multiple mating treatment was significantly lower than in the single mating treatment, but there was no apparent effect on the F2 generation. Repeated mating had no significant effects on progeny fitness. We postulate that multiple mating imposes cross‐transgenerational effects and may ultimately influence multigenerational fitness in S. litura.

success (Egan et al., 2016;Omkar & Sahu, 2012). Mating frequencies beyond that required for female fertilization can lead to sexual conflict (Backhouse et al., 2012). Conflict and competition between individuals have been shown to promote the evolution of varied mating patterns (Zhang et al., 2017). Widespread instances of polyandry have led to interest into the evolution of the behavior, especially in determining what the ecological and evolutionary consequences of polyandry might be (Boulton & Shuker, 2015).
Evolutionary explanations have suggested that females may gain direct and/or indirect benefits from polyandry (Havens et al., 2011;Moore et al., 2003;Tregenza et al., 2003;Xu et al., 2019). Repeated mating, which is repetitious mating with the same male mate, is a simple form of mate selection for polyandry (Jennions et al., 2007).
In theory, females could obtain the same level of benefits via repeated mating with the same male as they derive from copulation with different males if sperm is not depleted (Sakaluk et al., 2002).

Multiple mating may benefit females in instances where males
transfer resources directly to the females during mating (Castrezana et al., 2017), such as replenishing sperm and nutrients in the ejaculate, which may result in an increased number of offspring (Egan et al., 2016;Hsu et al., 2014;Li et al., 2014;Xu et al., 2019). In addition, indirect benefits arise from the genetic benefit of increasing offspring genetic diversity (Dunn et al., 2009;Egan et al., 2016;Johnson & Brockmann, 2013;Tregenza & Wedell, 1998). Understanding the relative costs and benefits of repeated mating with a single male and multiple mating with several males is key to understanding the evolution of polyandry.
The process where mating patterns of the parental generation can greatly affect offspring developmental, and/or reproductive and evolutionary trajectories, is commonly referred to as transgenerational effects (Ducatez et al., 2012;Gao et al., 2020). The traditional view regarding polyandry states that there is a positive relationship between female mating frequency and the number of offspring produced. In addition, multiple mating by females is thought to increase the fitness of their offspring (McNamara et al., 2014), although there is a trade-off with longevity (Kawazu et al., 2014).
Spodoptera litura (Fabricius) (Lepidoptera: Noctuidae) is a serious worldwide agricultural and forest pest with high fecundity (Ahmad et al., 2007). Females of the species have multiple matings (Li et al., 2012;Wu et al., 2015), with paired insects mating as many as six times, with an average of 2.54 ± 0.33 matings in its lifetime (Di et al., 2020). In this study, we attempt to assess the benefit hypothesis of female polyandry in S. litura by measuring the effects resulting from single, multiple, and repeated matings of S. litura on the development and fitness of their offspring. We postulate that the mating pattern of females will affect the development and fecundity of their offspring. Studying the mating habits of this widespread species, while determining the plasticity of their offspring, is expected to provide a greater overall understanding of their ecology and evolution.

| Population maintenance and insect rearing
The colony of S. litura used in this study was established in June

| Mating treatments of parental generation (F 0 )
The females were assigned to either monandrous or polyandrous mating treatments. Three mating treatments were conducted. Treatment 1. Single mating: randomly assigned 1-day-old virgin females and males were paired. After a single mating, the male and female were separated. Treatment 2. Multiple matings: 1-day-old females were paired with a 1-day-old male. After mating once, the male was removed and replaced with a random virgin 1-day-old male and the female allowed to mate once with the second male and then separated. Treatment 3. Repeated matings: individual females were paired with the same male and allowed to mate twice and were then separated. Cotton balls soaked in a 10% honey solution were changed each day until adult death. Single mating, multiple mating, and repeated mating were mated 30, 20, and 28 times, respectively.
Egg masses were collected and recorded daily until the death of the female. Egg masses from different treatments were transferred to separate Petri dishes and used for the experiments outlined below.

| First generation (F 1 ) life table after different mating treatments
In order to determine the effects of multiple and repeated matings on the offspring of S. litura, 100 randomly selected eggs were allocated to treatments the day after egg laying began obtained from each of the previously discussed mating treatments, when all mating for different mating treatments have been completed (single mating mated once, multiple mating, and repeated mating mated twice). Each group of 100 eggs were placed on moistened filter paper in plastic Petri dishes (2 cm high, 9 cm diameter). The newly hatched larvae were carefully transferred onto fresh tobacco leaves in individual 200 mL plastic vials (one per vial) (4 cm high, 8 cm diameter). The tobacco leaves were replaced as necessary, increasing in frequency with age until the larvae ceased feeding at the prepupal stage. Once offspring emerged as adults, females and males were paired at random and allowed to mate once. The females and males were then placed together in 1000 mL plastic containers and fed a 10% honey solution.
Eggs were collected daily and placed on moistened filter paper in plastic Petri dishes as described above. Fecundity was evaluated by counting the number of eggs laid per female per day. The duration of each developmental stage and survival was assessed daily until the death of each individual. All insects in this and subsequent experiments were maintained in a climate chamber at 27 ± 1°C, 60 ± 5% RH and 14:10 h (L:D) photoperiod.

| Second generation (F 2 ) life table after different mating treatments
Based on first-generation data obtained from the different mating systems, egg masses were collected from the three treatments after the F 1 generation adults had mated (single mating, multiple mating, and repeated mating adults of the F 1 generation all mate once). A total of 100 randomly collected eggs from each of the F 1 generation mating treatments were collected and transferred to Petri dishes (2 cm high, 9 cm diameter) containing moistened filter paper, and allowed to hatch. All newly hatched larvae were then individually transferred into 200 mL plastic vials (4 cm high, 8 cm diameter) and provided with tobacco leaves. The tobacco leaves were changed daily to maintain adequate nutrition until pupation. After emergence, adults from the same mating treatment were paired and allowed to mate once. Eggs were collected each day. The developmental, survival, longevity, and reproductive data were recorded as described above until the death of all individuals.

| Data analysis
The raw data for copulation duration, number of eggs, and longevity of the F 0 , F 1, and F 2 generation individuals were analyzed using SPSS 21 (SPSS 21.0, IBM). All data were tested for normality using the Kolmogorov-Smirnov test and met the assumption of ANOVA.
The number of eggs' data was transformed into a logarithmic transformation to approximate a normal distribution. The number of eggs and longevity were subsequently analyzed by one-way analysis of variance (ANOVA) followed by Tukey's HSD test (p < .05). The distribution of the copulation duration data was skewed and was analyzed by nonparametric Kruskal-Wallis analysis of variance. All data are presented as the mean ± standard error (SE). Survival curve analysis was evaluated by log-rank test for the different mating treatments.
The parameters of intrinsic rates of increase, finite rate of increase, net reproductive rates, and mean generation time were calculated using the bootstrap method included in the computer program TIMING-MSChart (Chi, 2020). Because bootstrap analysis uses random resampling, a small number of replications will generate variable means and standard errors. To generate fewer variable results, 100,000 replications were used in this study.

| Effects of mating treatment on parental generation (F 0 )
Multiple mating did not significantly affect the number of eggs (F 2,61 = 0.273, p = .762) or the longevity (F 2,84 = 0.619, p = .541) of the females, but did significantly affect the copulation duration (df = 2, χ 2 = 11.849, p = .003; Figure 1). There was no significant difference in the copulation duration between single mating and repeated mating of different mating treatments, but there was a significant difference in the copulation duration between single mating and multiple mating of different mating treatments. The single mating treatment was noticeably shorter than the multiple mating treatment.

| Transgenerational effect on immature developmental time of offspring
The immature development time in different mating treatments and different generations was markedly different (Figure 2). In the F 1 generation, the development time of immatures from single matings was noticeably longer than that of the multiple mating and repeated mating treatments. In the F 2 generation, however, the development time of immatures in the repeated mating treatment was significantly shorter than in the single mating treatment and multiple mating treatment. The development time of immatures in the single mating F 1 generation was significantly longer than it was in the F 2 generation, but multiple mating and repeated mating in the F 1 generation was significantly shorter than in the F 2 generation.

| Transgenerational effect on fecundity of offspring
Significant differences occurred in the fecundity in different generations and in different mating treatments (Figure 3). There was no significant difference in the number of eggs among the different mating treatments (F 0 : F 2,61 = 0.084, p = .920; F 1 : F 2,85 = 1.942, p = .150). The number of offspring in the F 2 generation of the multiple matings, however, was significantly higher than in the single and repeated matings (F 2,55 = 6.351, p = .003). In multiple matings, egg production was significantly greater in the F 2 generation than in the F I G U R E 3 Number of eggs produced by Spodoptera litura F 0 , F 1 and F 2 generations after different mating treatments. Different lowercase letters on the top of the bars indicate the significant differences among different mating treatments at p < .05 by Tukey's HSD multiple range test. Different uppercase letters on the top of the bars indicate the significant differences among different generations at p < .05 by Tukey's HSD multiple range test.

| Transgenerational effect on survival rate
The survival rates were significantly different in the different generations within different mating treatments (Figure 4). The survival rates of the F 2 generation, overall, were higher than those of the F 1 generation.

| Transgenerational effect on longevity
There was a significant difference in the longevity from the different mating treatments ( Figure 5). There was no significant difference between the longevity of the F 0 and F 2 generations (F 0 generation: F 2,125 = 2.897, p = .059; F 2 generation: F 2,92 = 0.684, p = .507). In the F 1 generation the longevity of multiple mating was significantly lower than it was in individuals from single mating and repeated mating (F 2,60 = 5.220, p = .008). There was no significant difference in longevity between single mating and repeated mating in the F 0 , F 1, and F 2 generations (Single mating: F 2,87 = 2.534, p = .085; Multiple mating: F 2,74 = 0.939, p = .398). The longevity of F 1 generation from multiple matings was significantly shorter than F 0 and F 2 generation (F 2,67 = 3.267, p = .044).

| Transgenerational effect on population parameters
The intrinsic rate of increase (r), finite rate of increase (λ), net reproductive rate (R 0 ), and mean generation time (T) for the different treatments and generations are listed in Table 1. Significant differences were observed in the population parameters of the offspring from different maternal mating treatments in the F 1 and F 2 generations. In the F 1 generation, the values for the intrinsic rate of increase and finite rate of increase in the single mating treatment (r = 0.138 d −1 , λ = 1.149 d −1 ) were higher than those in the multiple mating treatment (r = 0.087 d −1 , λ = 1.091 d −1 ). These values, however, were significantly higher in the repeated mating treatment (r = 0.176 d −1 , λ = 1.168 d −1 ) of the F 2 generation. There were significant differences in the net reproductive rate of the F 1 generation and mean generation time of the F 2 generation.

| DISCUSS ION
According to Arnqvist and Nilsson (2000), understanding mating behavior in insects is at the heart of comprehending their ecology and evolution. The sexy-sperm and bet-hedging hypothesis put forth by several researchers suggest that polyandry is likely to be an evolutionary strategy (Egan et al., 2016;McNamara et al., 2014;Sarhan & Kokko, 2007). Polyandry is a common phenomenon in many lepidopteran species (Jennions & Petrie, 2000;Kawazu et al., 2014;Royer & McNeil, 1993;Taylor et al., 2014;Wiklund et al., 2001). In order to thoroughly understand the effectiveness of this mating behavior, it is essential to follow the fitness gained (or lost) through successive generations (Zajitschek et al., 2018). In this study, transgenerational fitness was observed after different mating treatments of S. litura.
Maternal mating had no significant effect on the F 0 and F 1 generations, but clearly increased fecundity in the F 2 generation.
Polyandry is not only an evident mechanism for supporting genetic heterogeneity but also a behavior that affects the adaptation of females and their progeny (Ryazanova, 2011;Taylor et al., 2014).
Its benefits may manifest more in the subsequent mating success of offspring than in their immediate viability (Sakaluk et al., 2002).
Several previous studies have noted that in some species, multiple mating could increase offspring fecundity, fitness, and reproductive success (Bernasconi & Keller, 2001 F I G U R E 4 Survival probability of Spodoptera litura F 1 and F 2 generations after different mating treatments. Omkar & Sahu, 2012). In Anegleis cardoni (Weise) (Coleoptera: Coccinellidae) and Cnaphalocrocis medinalis (Guenée) (Lepidoptera: Crambidae), polyandry may provide direct benefits to females, generating mating gifts and increasing fecundity (Kawazu et al., 2014;Omkar & Sahu, 2012;Worthington & Kelly, 2016). In others, however, offspring fitness was not markedly different (House et al., 2011;Klemme et al., 2014;Zajitschek et al., 2018). The S. litura is polygynandrous (Di et al., 2020). However, previous studies failed to indicate whether multiple mating by female S. litura affected reproduction and longevity in their maternal generation (Bezzerides et al., 2008;Klemme et al., 2014). No evidence was found for any short-term parental benefits resulting from S. litura polyandry. Our results demonstrated that multiple mating increases offspring fecundity. Multiple mating had no significant effect on the fecundity and longevity in F 0 generation, but significantly increase fecundity of F 2 generation. Therefore, multiple mating may bring fitness benefit on fecundity of S. litura.

McLain
Although several researchers have suggested that multiple mating could increase genetic benefits in the form of genetic diversification and increase genetic compatibility (Bilde et al., 2009;Boulton & Shuker, 2015;Johnson & Brockmann, 2013), others have found that repeated mating seems to be beneficial solely in terms of fecundity (Ivy & Sakaluk, 2005;Omkar & Mishra, 2005;Walker & Allen, 2010).
Interestingly, the females of many insect species appear to discriminate against previous mates (Archer & Elgar, 1999;Bateman, 1998;Xu & Wang, 2009;Zeh et al., 1998). Male mating history may also have an effect on the mating choices in some females (Michaud et al., 2013). S. litura females have been shown to have a preference for their previous mates over novel males (Li et al., 2014). Moreover, in Nicrophorus vespilloides (Herbst) (Coleoptera: Silphidae) and the cricket, G. bimaculatus De Geer repeated mating apparently confers no known indirect benefits to their offspring (House et al., 2008;Tregenza & Wedell, 1998). Our results demonstrated that repeated mating had no obvious effects on fecundity compared with single mating. A possible explanation for this would be that repeated disturbance could affect fecundity (Li et al., 2015). Moreover, repeated mating could be costly to female S. litura in terms of time and energy.

Sexual interactions experienced by females in one generation
can potentially permeate through subsequent generations, affecting the reproductive success and survival of future generations (Zajitschek et al., 2018), that is, transgenerational effects (Heard & Martienssen, 2014). It is necessary to evaluate fitness consequences over successive generations if we want to understand the consequences of sexual interactions. The results of our study indicate that as a consequence of multiple mating there was a reversal of offspring fitness in the F 2 generations, where fecundity in the F 1 generations was not significantly increased, but was substantially enhanced in the F 2 generations. We suspect that multiple mating of S. litura not only induces cross-generational fitness, but, in general, is a behavior that is beneficial to the fecundity of S. litura. The multiple mating confers transgenerational benefits, which is in line with Zajitschek et al. (2018). Reversal effects in progeny fitness across different generations have also been found F I G U R E 5 Longevity of Spodoptera litura F 0 , F 1, and F 2 generations in different mating treatments. Different lowercase letters on the top of bars indicate significant differences among different mating treatments at p < .05 by Tukey's HSD multiple range test. Different uppercase letters on the top of bars indicate significant differences among different generations at p < .05 by Tukey's HSD multiple range test.
TA B L E 1 Population parameters (mean ± SE) of the F 1 and F 2 generations of Spodoptera litura after different mating treatments. in Drosophila melanogaster, where the fitness of the sons increased but grandsons decreased with increasing maternal sexual interactions (Brommer et al., 2012). Therefore, even in the absence of immediate benefits due to multiple mating, it will be necessary to determine whether multiple mating behavior has a potential reversing effect that is retained in succeeding generations as well, because transgenerational effects resulting from polyandry may potentially influence the rate and extent of evolutionary change (Bloch Qazi et al., 2017;Bonduriansky & Day, 2009). However, it is still unclear whether polyandry influences only the F 2 generations or if it enables effects that may extend into ensuing generations.
This will need further study.

| CON CLUS ION
Bet-hedging theory explains how a group of individuals should optimize fitness in unpredictable and varying environments by sacrificing the mean fitness in order to decrease variation in fitness (Olofesson et al., 2009). Life history and mating behavior can be modulated by conditions experienced by a parental generation, and such transgenerational effects may arise from epigenetic mechanisms or "bet-hedging" (Henshaw & Holman, 2015;Tougeron et al., 2020;Zajitschek et al., 2018). This study demonstrates that multiple mating by S. litura females caused cross-transgenerational effects. Although there were no discernable effects on the F 1 generations, significantly improvements were seen in the fecundity of the F 2 generations.
Indeed, maternal polyandry had effects that carried over for at least two generations. Our study highlights the importance of continuing observations of the effects across multiple generations to fully comprehend the net transgenerational consequences of sexual interactions. Even when there are no observable immediate costs or benefits due to sexual conflict, potential reversal effects in subsequent generations need to be considered to reveal the stability and mechanisms of transgenerational effects and long-term consequences that may result from the evolution of diverse mating patterns.

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors declare that they have no conflicts of interest.