https://orcid.org/0000-0001-6494-2097
Abstract
With significant surge of international trade in recent decades, increasingly more arthropod species have become established outside their natural range of distribution, causing enormous damage in their novel habitats. However, whether a species can colonize its new environment depends on its ability to overcome various barriers which may result in establishment failure, such as inbreeding depression and difficulty to find mates. Here, we used a haplodiploid pest, Tetranychus ludeni Zacher (Acari: Tetranychidae), which is native to Europe but now cosmopolitan, to investigate whether its reproductive strategies have facilitated its invasion success, providing knowledge to develop programs for prediction and management of biological invasions. We show that inbreeding had no negative influence on female reproductive outputs and longevity over 11 successive generations, allowing mother-son and brother-sister mating to occur at the invasion front without adverse consequences in fitness. Virgin females produced maximum number of sons in their early life to ensure subsequent mother-son mating but later saved resources to prolong longevity for potential future mating. Females maximized their resource allocation to egg production immediately after mating to secure production of maximum number of both daughters and sons as early as possible. Furthermore, mated females with mating delay increased proportion of daughters in offspring produced to compensate the loss of production of daughters during their virgin life. We suggest that the lack of inbreeding depression in successive generations and the ability to adjust resource allocations depending whether and when mating occurs may be the key features that have facilitated its invasion success.
Investigation into reproductive strategies that successful invaders may have employed to survive and colonize uncertain environment can provide knowledge for prediction and management of invasion risks. The European native haplodiploid mite, Tetranychus ludeni Zacher (Acari: Tetranychidae), has now invaded all continents except Antarctica (Bolland et al. 1998, CABI/EPPO 2020) and become an important pest of bean, eggplant, hibiscus, pumpkin, and other cucurbitaceous plants globally (Reddy 2001, Zhang 2003, Kaimal and Ramani 2011). To date, little is known about whether this mite has developed reproductive strategies that may have facilitated its invasion success, offering a good opportunity for the study of underlying mechanisms of biological invasions.
Biological invasion involves a series of stages such as introduction, establishment and spread, each of which has barriers that may result in invasion failure (Duncan et al. 2003, Heger and Trepl 2003, Blackburn et al. 2011). For example, in species with sexual reproduction, mates may be limited at the introduction front (Courchamp et al. 1999, South and Kenward 2001, Council 2002) where successful reproduction may be difficult (Heger and Trepl 2003, Blackburn et al. 2011), leading to population extinction. However, haplodiploid species may overcome the barrier of mate shortage at the invasion front (Baker 1955, Ward et al. 2012, Mazzolari et al. 2017), because females can produce haploid sons and then mate with their sons to produce both sons and daughters which can perform brother-sister mating thereafter (Cowan 1979, Adamson and Ludwig 1993, Peer and Taborsky 2005, McCulloch and Owen 2012, Schmidt et al. 2014). Mother-son mating could thus be potentially central to the invasion success of some haplodiploid species (Adamson and Ludwig 1993). Furthermore, mother-son and sibling mating may also reduce the cost and risk for mate search outside their natal habitat, increasing chances of successful colonization (Jordal et al. 2001).
Mother-son or sibling mating may result in inbreeding depression, reducing offspring fitness (Charlesworth and Willis 2009, Tien et al. 2015), and increasing extinction risks of small populations (Reed et al. 2003, O’Grady et al. 2006, Bozzuto et al. 2019, Nonaka et al. 2019). However, inbreeding depression does not appear to occur in some haplodiploid species after one (Peer and Taborsky 2005) or a few (Trevisan et al. 2016, Quaglietti et al. 2017) generations of inbreeding. Similarly, when we compare the reproductive fitness of the first generation with that of the 11th generation with different inbreeding levels in T. ludeni, we have not found any evidence of inbreeding depression (Zhou et al. 2020). Yet, it is not clear whether the lack of inbreeding depression remains consistent in successive generations, information of which is important for evaluation of whether inbred animals could have risk of population collapse in any generations.
For mother-son mating to occur in haplodiploid animals, virgin mothers would have to wait until their sons become sexually mature, by which time, even if they eventually mate with their sons, their reproductive fitness could decrease to the minimum due to mating delay, leading to extinction of the population. Therefore, females of successive haplodiploid invaders may have developed strategies to allow mother-son mating to take place with limited impact on their reproductive fitness. For example, females may reduce their reproductive rate when virgin to extend their longevity (Bonato and Gutierrez 1996, 1999) and increase resource allocations to reproduction after mating (Schmidt et al. 2014). However, it is still unclear how T. ludeni females adjust their reproductive strategies including resource allocation in response to mating delay, knowledge of which is vital to the understanding of invasion success in this mite and prediction of invasive potential of other haplodiploid animals.
In the present study, we carried out a series of experiments to determine whether inbreeding could increase risk of population collapse in successive generations and whether females could adjust their resource allocation in response to mating delay that facilitates mother-son mating in T. ludeni. First, we set up three breeding lines, namely mother-son mating, brother-sister mating and outbreeding, allowed mating to occur according to treatments for 11 successive generations, and recorded female reproductive outputs and longevity in each generation to evaluate the effect of inbreeding over generations. Because females are about 10-d old when their sons mature, all females were 10-d old when mated to make the data comparable. Second, we tested how females adjusted their resource allocation independent of inbreeding by comparing lifetime fecundity, daughter production, and longevity between females of different mating status, namely virgin, mated when 1-d old and mated when 10-d old.
Materials and Methods
Experimental Mites
We established a laboratory colony of T. ludeni from field-collected mites on Passiflora mollissima (Kunth) (Malpighiales: Passifloraceae) in Palmerston North, New Zealand, and maintained it on 3- to 5-wk-old common bean [Phaseolus vulgaris L. (Fabales: Fabaceae)] plants. We divided the colony into two (A and B) and kept them in two separate rooms for about eight generations before experiments. We used the first expanded leaves of P. vulgaris for all experiments. All colonies were maintained, and experiments carried out under the environmental conditions of 25 ± 1°C, 40 ± 10% RH, and photoperiod of 14:10 (L:D) h.
We randomly selected 40 male and 40 female deutonymphs from Colony A and individually transferred mites onto a clean leaf square (5.0 × 5.0 cm) placed on wet cotton wool in a Petri dish (9.5-cm diameter × 1.5-cm height) to ensure virginity. We then individually paired an 1-d-old female with an 1-d-old male on a clean leaf square (2.0 × 2.0 cm) to allow them to mate once and then transferred the mated female onto a new leaf square (2.0 × 2.0 cm) for oviposition for 5 d. We randomly selected three female deutonymphs that developed from the above eggs laid by each mated female for the following experiments.
Effect of Inbreeding Over Generations on Female Reproductive Fitness and Survival
To examine whether and how inbreeding affected reproductive fitness and survival over generations, we performed three treatments: 1) MSM—mothers and their sons mated for 11 successive generations, 2) BSM—brothers and sisters mated for 11 successive generations, and 3) OBM (outbreeding) —females from Colony A and males from Colony B mated for 11 successive generations. Because females in treatment MSM were about 10-d old when their sons became adults, for all three treatments we allowed females in each generation to mate when they were 10-d old. We individually transferred female deutonymphs prepared as described above to leaf squares (2.5 × 2.5 cm) and allowed virgin female adults to lay eggs for 10 d and then individually paired them with 1-d-old virgin males according to treatments until death. We replaced the leaf squares once every 5 d and randomly selected one to three female deutonymphs produced by each mated female within the first 5 d after females mated to start the next generation. We recorded the number of eggs laid before and after mating and the number and percentage of daughters produced after mating by each mated female, once every 5 d until she died. We used the following number of mated females for data recording from the first to the 11th generations, respectively: 30, 24, 18, 27, 34, 30, 22, 33, 26, 30, and 28 for MSM; 29, 20, 15, 26, 27, 31, 24, 34, 27, 32, and 27 for BSM; and 31, 25, 18, 35, 34, 37, 25, 36, 26, 31, and 29 for OBM.
Effect of Mating Status on Female Lifetime Reproductive Fitness and Survival
To determine whether mating status affected reproductive fitness and survival independent of inbreeding, we set up three treatments: 1) WMD (with mating delay)—virgin females from Colony A were allowed to lay eggs for 10 d and then mate with 1-d-old virgin males from Colony B, 2) NMD (without mating delay)—1-d-old virgin females from Colony A were allowed to mate with 1-d-old virgin males from Colony B, and 3) VF (virgin females)—virgin females from Colony A were allowed to lay eggs for lifetime without mating. We recorded lifetime number of eggs and daughters and the percentage of daughters after mating every 5 d until females died as described above for one generation. For fair comparisons, data from females that lived over 10 d were used for analysis. We used 31, 26, and 21 females for WMD, NMD, and VF, respectively.
Statistical Analysis
We used a logistic linear model to determine the effect of inbreeding on female reproductive fitness and longevity over 11 generations in three treatments, mother-son mating, brother-sister mating, and outbreeding: y = exp(a + bx), where y is the number of eggs and daughters produced, percentage of daughters, or female longevity, x is generation, and a and b are constant parameters of the model. We used a Negative Binomial distribution with a log link function for count data (number of eggs and daughters, and longevity), and a Gamma distribution with a log link function for percentage of daughters (GLIMMIX Procedure).
We applied a simple linear regression model to determine the egg-laying patterns over lifetime in virgin females, females with mating delay and females without mating delay (REG Procedure): y = a + bx, where y is the number of eggs, x is female age (days), and a is a constant and b the slope of regression. Because we recorded the number of eggs laid once every 5 d, virgin females and females with mating delay had the same status in the first 10 d of life and mating had a major impact on egg laying, we used three regression lines to fit each treatment when x ≤10, = 10–15, and ≥15 d, respectively, and compared oviposition patterns between treatments. If 95% confidence limits (CLs) of slopes overlap, then there is no significant difference (Julious 2004). We compared the total number of eggs laid and female survival in all three treatments and the number and proportion of daughters produced by females with mating delay and without mating delay. Data on the total number and percentage of daughters were normally distributed (Shapiro–Wilk test, UNIVARIATE Procedure) and analyzed by an analysis of variance (ANOVA, GLM Procedure). The total number of eggs laid in all three treatments were not normally distributed and analyzed using nonparametric ANOVA (GLM Procedure). Female survival was analyzed using a Lifetest (LIFETEST Procedure). We performed all analyses using SAS software (SAS 9.4, SAS Institute Inc., Cary, NC).
Results
Effect of Inbreeding Over Generations on Female Reproductive Fitness and Survival
Our results indicate that inbreeding had little effect on reproductive output and female longevity of each of 11 successive generations (eggs: F1,300 = 1.60, P = 0.2063 for MSM; F1,290 = 0.04, P = 0.8494 for BSM; F1,325 = 0.02, P = 0.8975 for OBM; daughters: F1,300 = 1.15, P = 0.2844 for MSM; F1,290 = 0.03, P = 0.8581 for BSM; F1,325 = 0.25, P = 0.6209 for OBM; percentage of daughters: F1,300 = 0.51, P = 0.4767 for MSM; F1,290 = 3.79, P = 0.0524 for BSM; F1,322 = 0.03, P = 0.8550 for OBM; longevity: F1,286 = 2.16, P = 0.1430 for MSM; F1,284 = 3.12, P = 0.0786 for BSM; F1,315 = 0.09, P = 0.7664 for OBM; Fig. 1). There was no significant difference in regression slopes between treatments for all parameters recorded (overlapping 95% CLs).
Number of eggs (A), number of daughters (B), percentage of female offspring after mating (C), and female longevity (D) over 11 generations in mother-son, brother-sister, and outbreeding treatments in Tetranychus ludeni.
Effect of Mating Status on Female Lifetime Reproductive Fitness and Survival
Lifetime oviposition patterns are shown in Fig. 2. Within the first 10 d of female life, the number of eggs laid by WMD and VF significantly decreased over time (F1,60 = 56.21, P < 0.0001 for WMD; F1,40 = 59.56, P < 0.0001 for VF) with similar decline rate (overlapping 95% CLs) but that by NMD significantly increased during the same period (F1,50 = 22.67, P < 0.0001). When female age = 10–15 d, the number of eggs laid significantly increased in WMD (F1,60 = 98.60, P < 0.0001), significantly decreased in NMD (F1,45 = 40.80, P < 0.0001) and remained similar in VF (F1,40 = 0.77, P = 0.3859). Although oviposition significantly increased after mating in both WMD and NMD, it increased significantly faster in WMD than in NMD (nonoverlapping 95% CLs). After female aged ≥15 d, the number of eggs laid decreased over time in all treatments (F1,58 = 43.15, P < 0.0001 for WMD; F1,33 = 1.64, P = 0.2092 for NMD; F1,66 = 17.52, P < 0.0001 for VF), with the decrease in WMD being significantly faster than in NMD and VF (nonoverlapping 95% CLs).
Lifetime oviposition patterns in Tetranychus ludeni females of different mating status: with mating delay (A), without mating delay (B), and virgin females (C).
Virgin females (VF) laid significantly fewer eggs than mated females (WMD and NMD) in their lifetime, and WMD females laid significantly fewer eggs than NMD ones (F2,75 = 18.50, P < 0.0001; Fig. 3A). In mated treatments, WMD females produced significantly fewer number of daughters (F1,55 = 7.57, P = 0.0080; Fig. 3B) but significantly higher proportion of daughters (F1,55 = 7.15, P = 0.0098; Fig. 3C) in their after-mating life than NMD ones. The survival probability of females in different treatments was significantly different, with an order of VF > WMD > NMD (, P < 0.0001; Fig. 4).
Lifetime number of eggs (A) and daughters (B) and percentage of daughters after mating (C) of Tetranychus ludeni females with different mating status. Treatments with different letters are significantly different (P < 0.05).
Survival probability of Tetranychus ludeni females of different mating status. Lines with different letters are significantly different (P < 0.05).
Discussion
Inbreeding depression may reduce offspring fitness (Charlesworth and Willis 2009, Tien et al. 2015), leading to extinction of small populations (Reed et al. 2003, O’Grady et al. 2006, Bozzuto et al. 2019, Nonaka et al. 2019). Previous studies show that the inbreeding depression can occur in any generation of inbred animals, which could cause invasion failure. For example, in the haplodiploid Stigmaeopsis miscanthi (Saito) (Acari: Tetranychidae) (Mori et al. 2005) and Phytoseiulus persimilis (Athias-Henriot) (Acari: Phytoseiidae) (Çekin and Schausberger 2019), inbreeding depression does not occur in the first few generations but takes place in later generations whereas in T. urticae Koch (Acari: Tetranychidae), it is only detected in the first few generations (Tien et al. 2015). In the present study, we found no evidence for inbreeding depression in any of the 11 inbred generations in T. ludeni (Fig. 1), suggesting that inbreeding has no negative impact on its invasion success at any points or generations. We attribute this phenomenon to the notion that the expression of deleterious recessive alleles from inbreeding may be selected out in small populations of some haplodiploid animals (e.g., Quaglietti et al. 2017, Eyer et al. 2018). The mechanism allows them to undertake mother-son and brother-sister mating at the invasion front without adverse consequences in fitness (Adamson and Ludwig 1993, Kronauer et al. 2012, McCulloch and Owen 2012, Schmidt et al. 2014, Lantschner et al. 2020, Queffelec et al. 2020).
Our results demonstrate that T. ludeni had developed strategies to adjust resource allocations in response to uncertain environment. For example, virgin females produced highest number of sons in their early life (Fig. 2A and C) to secure subsequent mother-son mating. This feature can ensure continuous population growth in uncertain situations, such as no males available at the invasion front. However, if there were still no males to mate after this reproductive episode, virgin females saved resources by reducing egg production (Fig. 2A and C) to prolong longevity (Fig. 4) for potential future mating. Some other spider mites, such as T. marianae McGregor (Acari: Tetranychidae), Mononychellus progresivus Doreste (Acari: Tetranychidae) (Bonato and Gutierrez 1996, 1999), and T. urticae (Li and Zhang 2020), appear to share similar reproductive strategies with T. ludeni, that may contribute to their invasion success.
Immediately after receiving ejaculates, females sharply increased their resource investment in egg production regardless of whether mating delay occurred (Fig. 2A and B) to ensure production of maximum number of both daughters and sons as early as possible. This resource allocation strategy may reduce risks of potential reproductive failure in a new and unstable environment (Stearns 1992, Stearns et al. 2000), facilitating establishment of a newly invaded population (Dangremond and Feller 2016, Fetters and McGlothlin 2017, James et al. 2017). The very reproductive strategy also occurs in haplodiploid mite Schizotetranychus celarius (Banks) (Acari: Tetranychidae) (Saitō 1987) and ant Cardiocondyla argyrotricha (Hymenoptera: Formicidae) (Schmidt et al. 2014). Our results show that both after-mating increase and after-peak decline of egg laying were faster in females with mating delay than in those without mating delay (Fig. 2A and B). This suggests that T. ludeni females can perform a clear resource allocation trade-off (Waelti and Reyer 2007, Billman and Belk 2014), i.e., higher resource investment in eggs by delay-mated females as compared to undelay-mated ones results in faster decrease of resource available for future egg production. The ability to quickly adjust resource allocation depending on when mating occurs may provide more flexibility for the species to establish in a new environment.
Although virgin females laid fewer eggs in their lifetime than mated ones (Fig. 3A), their sons are larger which produce more daughters after mating than those of mated ones (Zhou et al. 2018). This life history strategy should also contribute to future population growth and invasion success (Wiernasz et al. 2001, De Jesus and Reiskind 2016, Zhou et al. 2018). Our results demonstrate that females with mating delay produced fewer number of eggs and daughters as compared to those without mating delay. However, the reduction was less than 25% (Fig. 3A and B), which may not be enough to lead to establishment failure. Furthermore, females with mating delay produced higher proportion of daughters than those without mating delay (Fig. 3C), which may help compensate the loss of production of daughters in their earlier life and catch up population growth.
In the present study, we show that inbreeding has no negative influence on female reproductive outputs and longevity throughout 11 inbred generations, which allows mother-son and brother-sister mating to occur at the invasion front without adverse consequences in fitness. Virgin females lay maximum number of sons in their early life to ensure subsequent mother-son mating but later save resources to prolong longevity for potential future mating. Females maximize their resource allocation to egg production immediately after mating to secure production of maximum number of both daughters and sons as early as possible, reducing risks of potential reproductive failure in a new environment. Finally, mated females with mating delay quickly increase proportion of daughters in offspring produced to compensate the loss of production of daughters during their virgin life. These reproductive strategies of T. ludeni coupled with its adaptation to wide ranges of hosts (Gotoh et al. 2015) and temperatures (Gotoh et al. 2015, Ristyadi et al. 2019) may have facilitated its invasion success. Our findings may be also useful for prediction of invasive potential of other haplodiploid pests.
Acknowledgments
We thank Professor Z.-Q. Zhang for identification of this spider mite to species and K. Sinclair for technical assistance. We also thank two anonymous reviewers for their constructive comments and suggestions. This work was supported by the New Zealand-China Doctoral Research Scholarships Programme to PZ, the China Scholarship Council-Massey University Joint Scholarship Program to CC and Massey University Research Funds to XZH and QW, respectively.
