Abstract Many lady beetles expel an autogenously produced alkaloid-rich ‘reflex blood’ as an antipredator defense. We conducted an experiment to determine whether there was a measurable fitness cost associated with the daily induction of this defensive behavior, and whether costs differed between native (Coccinella novemnotata Herbst (Coleoptera: Coccinellidae)) and invasive species (Coccinella septempunctata L. (Coleoptera: Coccinellidae) and Harmonia axyridis (Pallas) (Coleoptera: Coccinellidae)). Newly mated females were provided a restricted or unrestricted amount of aphids and were bled for 10 d. We measured the mass of reflex blood produced and the total number and viability of eggs laid per day. The amount of reflex blood released per day increased for C. septempunctata at the restricted level and did not change for any other species-diet level combination. We did not detect a significant cost of reflex bleeding on the quantity or viability of eggs laid by any species, even at the restricted aphid level. Remarkably, bled individuals at the ad libitum level laid significantly more viable eggs compared to controls. All species laid significantly fewer total eggs (49–69% fewer) at the low versus high aphid level. These results demonstrate that while resource scarcity has a negative impact on fecundity, repeated use of the reflex bleeding defense system does not. These results support the findings of other reports and strongly suggest that adult lady beetles incur no measurable physiological costs related to the induction of the reflex-bleeding defense. Coccinella novemnotata, Harmonia axyridis, cost, reflex bleeding, defense A multitude of plant and animal taxa have evolved chemical defense systems to protect against natural enemies (Feeny 1976, Ruxton et al. 2004, Eisner et al. 2007). Plants are known to create a diverse array of chemical compounds that are not directly related to primary metabolism, and it has been posited that the production of secondary metabolites evolved to defend against would-be phytophages, and many secondary metabolites are known to negatively affect herbivores (Bennett and Wallsgrove 1994). Animals, on the other hand, are able to acquire chemical defenses against predators following two major pathways; they can sequester toxic compounds from their food or synthesize defensive chemicals de novo (Blum 1981, Pasteels et al. 1983, Bowers 1992). The interaction of plants and insects, as mediated by host plant chemistry, has long fascinated ecologists and has led to the generation of a multitude of hypotheses surrounding the evolutionary forces that drive the speciation of both groups of organisms (Ruxton et al. 2004, Eisner et al. 2007, Agrawal et al. 2012, Hare 2012). One of the most obvious factors on which natural selection can act are the costs associated with the production and maintenance of chemical defensive systems. Numerous case studies and review articles have been devoted to identifying and assessing the potential costs of de novo synthesis and sequestration of chemical defenses in plants and their herbivores (Zvereva and Kozlov 2016 and references therein), however, comparatively few studies have been conducted on the fitness costs of de novo synthesis of chemical defenses in predators (Zvereva and Kozlov 2016), and none directly link fitness costs to a competitive advantage of invasive over native species. Aphidophagous coccinellids are ubiquitous, and both native and introduced species have been used successfully in classical and augmentative biological control programs in North America for over a century (Gordon 1985, Obrycki and Kring 1998). Native North America coccinellids like the ninespotted lady beetle, Coccinella novemnotata Herbst (Coleoptera: Coccinellidae), the transverse lady beetle, Coccinella transversoguttata Falderman, and the two-spotted lady beetle, Adalia bipunctata L., experienced large declines in their geographic distribution and abundance shortly after the introduction and establishment of the European native Coccinella septempunctata L. (Coleoptera: Coccinellidae) (Harmon et al. 2006, Losey et al. 2012). Since their decline, an additional highly competitive lady beetle species, Harmonia axyridis (Pallas) (Coleoptera: Coccinellidae), has also become established in North America (Koch et al. 2008) and has expanded its range across much of the continent (Snyder and Evans 2006, Leppanen et al. 2011). Numerous hypotheses exist as to how non-native lady beetle species have displaced natives, ranging from widespread habitat loss (Harmon et al. 2006) to intraguild predation, and competition (Gardiner et al. 2011). Laboratory studies have found that C. novemnotata develops more slowly than its invasive congener C. septempunctata (Ugine and Losey 2014), and is susceptible to intraguild predation by C. septempunctata and H. axyridis (Turnipseed et al. 2014, Tumminello et al. 2015, Ducatti et al. 2017). C. septempunctata has also been shown to have a significantly higher rate of aphid consumption than C. novemnotata, indicating an advantage in scramble competition for limited aphid resources (Hoki et al. 2014). While most of the current hypotheses regarding interactions between native and invasive species have focused on competition, intraguild predation and population dynamics, differences in the costs associated with defensive behaviors and de novo synthesis of defensive chemistry have not been thoroughly investigated. Assuming that beetles have a finite amount of resources to divide between defense and reproduction, differences in the partitioning of resources as a function of species (i.e., invasive species devoting more to reproduction) could help to explain native species’ decline. Many species of coccinellids have developed an antipredator defense system composed of the classic red/black warning coloration combined with defensive chemistry. When attacked, lady beetles engage in an antipredator defensive behavior known as reflex bleeding whereupon they expel a foul-smelling, alkaloid-rich substance from the joints between their femora and tibiae (Pasteels et al. 1973, Holloway et al. 1991, de Jong et al. 1991), which can be an effective deterrent against predation by ants and birds (Marples et al. 1994, Majerus et al. 2007). Female coccinellids also add these same alkaloids to their eggs, which have been shown to significantly reduce intraguild egg predation (Kajita et al. 2010) and sibling cannibalism (Sato et al. 2009). Additionally, lady beetle defensive alkaloids are synthesized via the fatty acid pathway (Haulotte et al. 2012), and as suggested by Nazareth and Machado (2015) there is a high demand for fatty acids when provisioning eggs with yolk (Trougakos and Margaritis 2002). It follows that there could be competition for fatty acids between these two allocations and that the degree of that competition could vary among invasive and native species. This could manifest in differential rates of population growth and the eventual displacement of ‘high cost of defense’ species, especially when food resources are limited. A few studies have attempted to determine whether there are costs associated with the induction of reflex bleeding response in coccinellids. Grill and Moore (1998) showed that H. axyridis developed more slowly and had a decreased mean adult body weight when larvae were induced to reflex bleed. Holloway et al. (1991) and de Jong et al. (1991) showed that the concentration of alkaloids in the reflex blood produced by adult C. septempunctata and A. bipunctata decreased significantly over time, suggesting that its production may be slow and by extension, expensive to synthesize. However, these studies only focused on intraspecific and sexual variation in alkaloid concentrations and had no direct measure of fitness costs. Knowledge of the relationship between the deployment of defensive behaviors and fitness costs among competing species may provide addition insight into the success of invasive over native coccinellids. To determine whether the decline of the North American native C. novemnotata may be due in part to a higher cost of producing defensive compounds relative to the introduced species C. septempunctata and H. axyridis, we conducted an experiment in which we induced reflex bleeding in mated adult female C. novemnotata, C. septempunctata, and H. axyridis for 10 consecutive days and determined the number of viable offspring produced. Because we believed that the daily replenishment of reflex blood and alkaloids may be energetically expensive and limited by nutrient availability, we included a diet-level treatment in which we provided beetles with a reduced number of aphids to maximize our opportunity to observe fitness costs, and to simulate a poor resource environment as occurs when aphids disperse from a patch or populations are otherwise depleted. Materials and Methods Insect Colonies The C. novemnotata and C. septempunctata used in these experiments came from laboratory colonies originating from Long Island, NY, and our H. axyridis colony originated from adults collected in Ithaca, NY. Beetles were reared in 44 ml clear plastic-lidded portion cups containing a small piece of paper towel (2.5 × 7 cm2) and fed an ad libitum diet of freshly collected pea aphids (Acyrthosiphon pisum Harris). Eggs laid by mated female beetles were collected daily for use in this experiment. Lady beetle colonies were maintained at 23 ± 2°C and a photoperiod of 16:8 (L:D) h. Colonies of pea aphids and green peach aphids (Myzus persicae (Sulzer)) were maintained in separate environmental growth chambers on fava bean plants (Vicia faba L., var. Windsor) at 22 ± 2°C and a photoperiod of 16:8 (L:D) h. Experiment Plants to maintain aphids were generated by placing two fava bean seeds at the bottom of 10.2-cm-diameter pots. The seeds were then covered with LM-series professional growing media (Lambert, Quebec City, QC, Canada), and maintained in a greenhouse at 25 ± 2°C and a photoperiod of 16:8 (L:D) h for 10 d. Several 6–8 cm clippings of fava plants that were infested with 100–200 mixed-aged green peach aphids were placed upon the soil of each pot, and pots were individually covered with micro-perforated plastic bread bags (160 holes per 6.45 cm2; Webstaurant.com, Lancaster, PA). Bagged plants were placed into non-perforated trays for sub-irrigation with tap water and returned to the greenhouse for 10 d. Newly-hatched (<24 h) C. septempunctata, C. novemnotata, and H. axyridis larvae were reared to the second instar in the afore-mentioned 44-ml portion cups containing a small piece of paper towel and an ad libitum diet of pea aphids. Once larvae molted to the second instar, they were transferred singly to the bagged fava bean plants (one larva per pot). We chose to rear and transfer second instar larvae because this greatly increased the number of larvae surviving to adulthood compared to the transfer of first instars. Pots containing larvae were maintained in a greenhouse at 25 ± 2°C and a photoperiod of 16:8 (L:D) h and watered daily. Seven days post-eclosion, adult beetles were removed from the bagged plants, sexed, and paired for mating (one male + one female) within empty portion cups. Once copulation had ceased, male and female beetles were separated and the males were discarded. Newly-mated females were placed in 44-ml portion cups with a piece of paper towel and provided an excess of mixed-aged pea aphids for 24 h prior to the start of the experiment. We switched adult lady beetles to a diet of pea aphids, as it is difficult to produce enough green peach aphids to feed large groups of mated females, and it has been shown that a mixed diet of prey can be beneficial to predator fitness (Marques et al. 2015). The initial mass of each beetle was determined 24 h post-mating and recorded. Females were then arbitrarily assigned to one of two pea aphid diet levels, ad libitum versus a restricted amount (40 mg/d). Half of the beetles within each diet treatment were induced to reflex bleed daily for 10 consecutive days. Our technique for collecting reflex blood was modeled after that described by Holloway et al. (1991). A piece of 4-mm-diameter silicon tubing that was attached to a vacuum pump with a draw equal to 20 psi held individual beetles in place by creating a gentle vacuum with the elytra. Reflex bleeding was induced by gently probing the legs and abdomen of beetles with the back end of a tared 5-µl glass capillary tube. The front end of the capillary tube was used to collect droplets of reflex blood, and the mass of reflex blood produced was measured within 30 s of its collection. The beetles were considered fully bled once agitation of all six individual joints failed to produce additional reflex blood. This technique provided a relative measure of the amount of reflex blood exuded across the three species and two diet levels tested. The mass of the reflex blood was recorded to the nearest 0.1 mg on a Mettler-Toledo analytical balance. No-bleed control beetles were briefly subjected to the vacuum (10–15 s) to control for the general stress of handling; no control beetles exuded reflex blood during the course of this study. Female beetles were transferred daily to a clean cup containing a piece of paper towel and aphids at their assigned level, and the number of eggs laid by each beetle was recorded daily. Eggs were maintained for 3–4 d at 25°C to determine egg viability, which was the sum of the number of emerged larvae and eggs with a visible larva within the chorion. Beetles were defined as having successfully mated if they laid at least one viable egg. In total, we analyzed data from 53 C. septempunctata, 61 C. novemnotata, and 48 H. axyridis females over three experimental replications of the study. Statistical Analysis All analyses were conducted using the statistical packages JMP (Version 11.0; SAS Institute, 2013) and SAS Pro software (Version 9.3; SAS Institute, 2011, Cary, NC). The total number of eggs laid during the 10-d test period and the number of eggs laid per day were analyzed separately using mixed model analysis of variance (ANOVA). These data were transformed using the natural log to correct for skew and to normalize the mean-variance ratio. Our first analysis investigated the effect of species origin (native vs. introduced) on the total number of eggs laid, and our model included the fixed effect ‘origin’ and the random effect experimental block (date). The same analysis was conducted the native species excluded from the data set to determine whether the total egg production of the two introduced species differed from one another. Next, to investigate the effects of our treatments on total number of eggs laid over the test period (i.e., pooled across days), our fixed effects included species, diet level, bleed treatment, and all two- and three-way interactions, with experimental block (date) and insect (replicate) treated as random variables. Analysis of the total number of eggs laid over time included the factors time (days), species, diet level, bleed treatment, and all two-, three-, and four-way interactions. Post-hoc analyses were performed using Tukey’s HSD and Student’s T-test (α = 0.05). The overall viability of eggs pooled across days, and the egg viability over time (effect of time on egg viability), were analyzed using PROC GLIMMIX. Analysis of the overall egg viability included the fixed effects species, diet level, bleed treatment, and all two-way interactions, with the random effects of experimental block and insect. The model used to analyze egg viability over time included the factors time (day), species, diet level, bleed treatment, and all two-, three-, and four-way interactions. Post-hoc analyses were conducted on the egg viability data collected on days 3 and 8 post mating using the lsmeans/pdiff statement in SAS. This time period (day 3) allowed female beetles that received a reduced amount of aphids to lay any eggs that they had developed before the start of the experiment, and day 8 was tested as it represented a time point equidistant to the end of the test as day 3 was to the beginning. The total mass of reflex blood exuded by beetles during the 10-d test period and per day was analyzed using mixed-model ANOVA. Our model included the fixed effects species, diet level, female mass as a covariate, and experimental block as a random effect. Analysis of total reflex blood over time included the factors species, diet, and day. Results Total Eggs Laid and Eggs Laid Over Time Overall, across diet and bleed treatments, the total number of eggs laid over the 10-d test period did not differ as a function of species being native versus introduced (F(1,156.9) =0.64, P = 0.42), and there was also no difference in the total number of eggs laid by the two introduced species (F(1,42.6) = 0.67, P = 0.42). There was not a significant effect of reflex bleeding for 10 consecutive days on total number of eggs laid during the 10-d test period (F(1,152.5) = 2.89, P = 0.09, Fig. 1); bled insect laid an average of 323 ± 19 eggs and insects in the no-bleed treatment laid an average of 294 ± 19 eggs. The total number of eggs laid was significantly affected by the lady beetle species × diet level interaction (F(2, 151.9) = 6.12, P = 0.003, Fig. 1). Student’s t-test showed a significant difference within the low diet treatment and also revealed a between-diet-level difference; the introduced species C. septempunctata laid 26.7% fewer eggs compared to the native species C. novemnotata on the restricted aphid diet. There was also a main effect of diet level (F(1,151.9) = 90.17, P < 0.0001), with beetles provided pea aphids ad libitum laying a mean of 430 ± 17 eggs, and those provided a low diet laying 187 ± 9 eggs during the test period. The total number of eggs laid was not significantly different for the three lady beetle species tested (F(2,147.8) =0.38, P = 0.68); C. septempunctata, C. novemnotata and H. axyridis laid 323 ± 28, 312 ± 19 and 287 ± 24 eggs, respectively. Similarly, there was no effect of the species × diet level × bleed treatment, species × bleed treatment, or diet level × bleed treatment interactions (P > 0.22). Fig. 1. View largeDownload slide Mean ± SE number of eggs laid during the 10-d test period by C. septempunctata, C. novemnotata and H. axyridis as a function of (A) bleed treatment, and (B) aphid level. Bars with the same lowercase letter above them are not significantly different from one another (Students t-test). Fig. 1. View largeDownload slide Mean ± SE number of eggs laid during the 10-d test period by C. septempunctata, C. novemnotata and H. axyridis as a function of (A) bleed treatment, and (B) aphid level. Bars with the same lowercase letter above them are not significantly different from one another (Students t-test). The number of eggs laid over time varied significantly as a function of the day × species × diet level interaction (F(18,792) = 2.15, P = 0.004), but not by the main effect of bleeding (F(1,41) = 1.4, P = 0.24). To examine the significant effect of time (day) further we ran an independent model for each species using the fixed effects diet level and day (Fig. 2) and determined that the interaction of diet and day was significant for C. septempunctata (F(9,180) = 9.01, P < 0.0001) and C. novemnotata (F(9,297) = 4.61, P < 0.0001), but not for H. axyridis (F(9,297) = 0.46, P = 0.90). There were significant two-way interactions for the day × species (F(18,792) =1.72, P = 0.03) and day × diet level (F(9,792) = 11.1, P < 0.0001). Inspection of Fig. 2 shows that the daily egg production by the two Coccinella species at the restricted diet amount started higher compared to the ad libitum treatment, and drop below the ad libitum treatment by day 3. This is as compared to H. axyridis at the restricted aphid level, which both started and stayed lower then the ad libitum treatment. The significant day × diet level interaction is driven by higher egg production by Harmonia compared to C. novemnotata on days 3 and 4 (P < 0.05). Fig. 2. View largeDownload slide The mean ± SE number of eggs laid per day by C. septempunctata (A), C. novemnotata (B), and H. axyridis (C), as a function of aphid diet level. Fig. 2. View largeDownload slide The mean ± SE number of eggs laid per day by C. septempunctata (A), C. novemnotata (B), and H. axyridis (C), as a function of aphid diet level. Egg Viability There was a significant effect of the bleed treatment × diet level interaction on the total number of viable eggs produced during the 10-d test period (i.e., pooled across days; F(1, 153) = 4.3, P = 0.04), and there was no main effect of lady beetle species (F(2, 154.8) = 1.9, P = 0.15). Lady beetles that were bled daily and provided an ad libitum diet of pea aphids laid significantly more viable eggs (386 ± 22/10 d; 1.35 times more) than the no-bleed/ad libitum control group and the daily bleed and control groups provided a restricted number of pea aphids (3.02 times more), which did not differ from each other. There was not a significant main effect of reflex bleeding or diet level on the proportion of viable eggs laid during the 10d test period across the species we tested (F(1, 141.1) = 2.24, P = 0.14; F(1, 140.9) = 0.18, P = 0.67, respectively). There was, however, a significant main effect of species (F(2, 141.2) = 5.44, P = 0.005). A larger proportion of C. septempunctata eggs were viable (0.836 ± 0.026, n = 53) compared to C. novemnotata (0.715 ± 0.035, n = 61) and H. axyridis (0.678 ± 0.052, n = 48), which did not differ significantly from each other. We also found a significant two-way diet level × bleed treatment interaction (F(1,141.1) = 5.95, P = 0.016, Fig. 3). Individuals that were bled daily and provided an ad libitum diet laid significantly more viable eggs compared to the ad libitum—and restricted diet—no bleed treatments, which did not differ from each other (Fig. 3). Fig. 3. View largeDownload slide Box plot of the proportion of viable eggs produced during the 10-d test period as a function of diet level and bleeding treatment. The lower and upper boxes represent the first and third quartile, respectively, their intersection is the median, and the whiskers represent the minimum and maximum viabilities. Fig. 3. View largeDownload slide Box plot of the proportion of viable eggs produced during the 10-d test period as a function of diet level and bleeding treatment. The lower and upper boxes represent the first and third quartile, respectively, their intersection is the median, and the whiskers represent the minimum and maximum viabilities. Analysis of the proportion of eggs that were viable over time revealed a significant three-way day × diet level × bleed treatment interaction (F(1,1230) = 90.9, P < 0.0001) as shown in Fig. 4, and a significant two-way day × species interaction (F(2,1230) = 39.79, P < 0.0001). C. septempunctata egg viability was significantly higher than both C. novemnotata and H. axyridis on day 3 (P < 0.013) and day 8 (P < 0.004) after a Bonferroni correction (α = 0.016), lending support to the significant difference in egg viability over time as a function of species. Fig. 4. View largeDownload slide The mean ± SE viable proportion of eggs laid by C. septempunctata (A), C. novemnotata (B), and H. axyridis (C) as a function of aphid diet level and bleed treatment over time. Fig. 4. View largeDownload slide The mean ± SE viable proportion of eggs laid by C. septempunctata (A), C. novemnotata (B), and H. axyridis (C) as a function of aphid diet level and bleed treatment over time. Reflex Blood Mass There was a significant main effect of species (F(2,43.2) = 4.66, P = 0.01) on the mean mass of reflex blood pooled across days, and no significant effects of female mass (F(1,73.8) = 3.34, P = 0.62) or diet level (F(1,73.0) = 0.24, P = 0.62). Post-hoc Tukey’s testing showed that the total mean mass of blood exuded by H. axyridis (13.28 ± 1.07 mg/10 d) was significantly lower than that from both C. novemnotata and C. septempunctata (17.81 ± 0.96 mg/10 d and 17.80 ± 1.19 mg/10 d, respectively), which did not differ significantly from each other. Analysis of the mass of reflex blood exuded by females over time revealed a significant three-way species × diet level × day interaction (F(2,701.7) = 8.98, P < 0.0001). We reran this analysis as a function of species to further understand the three-way interaction, and found that the mean mass of reflex blood expelled by C. septempunctata increased significantly over time at the low aphid level, but not the ad libitum level (F(1, 241) = 26.1, P < 0.0001), and did not change for C. novemnotata (F(1, 283.1) = 2.9, P = 0.09) or H. axyridis (F(1, 178) = 2.4, P = 0.12) as a function of aphid level (Fig. 5). Fig. 5. View largeDownload slide The mean ± SE mass of reflex blood exuded by (A) C. septempunctata, (B) C. novemnotata, and (C) H. axyridis females over time and as a function of aphid level. Open boxes represent responses of beetles on the restricted aphid diet and filled dots represent responses at the ad libitum aphid level. Fig. 5. View largeDownload slide The mean ± SE mass of reflex blood exuded by (A) C. septempunctata, (B) C. novemnotata, and (C) H. axyridis females over time and as a function of aphid level. Open boxes represent responses of beetles on the restricted aphid diet and filled dots represent responses at the ad libitum aphid level. Discussion According to resource allocation theory, organisms have a finite amount of time and nutritional resources to partition among their various life functions (Cody 1966). Therefore, investments in antipredator chemical defenses, specifically de novo synthesis of defensive alkaloids and carrier fluid in our case, should presumably be debited from investments in growth and reproduction. In this study, we did not detect significant fitness costs as measured in terms of the total number of eggs laid, or their viability for any of the native and invasive species tested, even when food amount was restricted. On the contrary, those individuals that were bled for 10 consecutive days and provided an ad libitum diet of pea aphids obtained a fitness benefit from the daily induction of their bleeding defense and laid significantly more viable eggs compared to control insects. While several studies have documented costs associated with sequestered and autogenous defenses in herbivorous insects (Rowell-Rahier and Pasteels 1986, Codella and Raffa 1995, Camara 1997, Zvereva et al. 2017) a recent meta-analysis revealed that across systems there is a general lack of physiological costs associated with chemical defenses (Zvereva and Kozlov 2016). Far fewer studies have been conducted on the physiological costs of de novo defenses in predators, and they had mixed results. Nazareth, Sudatti and Machado (2016) and Nazareth and Machado (2015) showed that female harvestmen (Opiliones) bearing mature eggs produced lower levels of defensive secretions compared to non- and post-reproductive females, suggesting a trade-off between egg production and defense. Grill and Moore (1998) showed that when H. axyridis larvae were induced to reflex bleed during development, the resulting adults were smaller compared to control beetles, which they implied was a significant fitness cost. On the other hand, Holloway et al. (1991) and de Jong et al. (1991) showed that C. septempunctata, and A. bipunctata L. that were induced to reflex bleed, did not lay different numbers of eggs compared to control insects, suggesting that lady beetles enjoy a ‘no-physiological cost defensive system’. No studies that we are aware of have reported an increase in the fitness (viable egg production) of an arthropod when its defensive chemicals are depleted over time, as we observed here. Our beetles laid fewer eggs that hatched at the restricted diet amount, a result similar to Perry and Roitberg (2005), who reported that adult H. axyridis laid significantly more inviable ‘trophic’ eggs when switched to a restricted pea aphid diet. They concluded that female H. axyridis were influencing egg viability as an oviposition strategy to provision neonate larvae with a nutritional resource (trophic eggs) when aphids were scarce. Additional studies that quantify the number of aphids consumed as a function of reflex bleeding are needed to determine if bled individuals engage in compensatory aphid feeding to replenish depleted resources needed for defense. The results of such a study could 1) confirm a cost of the use of defensive systems in terms of increased foraging and handling time to obtain the resources needed to ‘recharge’ the full defensive system, and 2) determine whether lady beetles lay trophic eggs as a strategy to provision larvae during lean times, or if nutrient limitation restricts viable egg production, especially in laboratory-reared individuals with single-prey species diets. In this study, we found that the wet mass of reflex blood expelled over the 10-d test period remained relatively stable for each species tested and that the introduced species H. axyrids produced less reflex fluid compared to C. septempunctata (introduced) and C. novemnotata (native). The relative consistency of reflex blood weights over time suggests that production of the carrier fluid is not costly. Holloway et al. (1991) and de Jong et al. (1991) showed that the amount (mg) of reflex blood exuded by field-collected adult C. septempunctata, and A. bipunctata emerging from diapause increased for the first 5 d post-collection when induced to reflex bleed, and that the amount of reflex blood expelled dropped significantly by about one-third in each study and tailed off between days 7 and 15. The amount of reflex blood collected from C. septempunctata by Holloway et al. (1991) ranged from an average of about 2.9 mg on the first collection to a maximum of approximately 4.6 mg on day 8. C. septempunctata and C. novemnotata in our study produced on average 1.8 mg of reflex blood per day, which is considerably less than that reported by Holloway et al. (1991). This is surprising because Holloway et al. (1991) and de Jong et al. (1991) also show that larger beetles expel greater amounts of reflex blood, and our C. septempunctata and C. novemnotata were as large or larger than those used by Holloway et al. (1991). This may be expected given that we fed our beetles a minimum of 40 mg wet weight of pea aphids, all of which were consumed by each species, and the maximum wet weight of reflex blood produced by any beetle in this study was 3.5 mg. It is unclear why the wet weight of reflex blood collected between these two studies is so different, although it may be related to the physiological state of the two groups of animals; our beetles were laboratory-reared from egg to adult and those in Holloway et al. were collected from overwintering populations. Additionally, both Holloway et al. (1991) and de Jong et al. (1991) showed via gas chromatography-mass spectroscopy that the amount of alkaloid contained in reflex blood decreased steadily with each successive induction of the bleed defense system, indicating that the biosynthesis of defensive alkaloids is slow. Therefore, it may not be surprising that we did not observe a significant physiological fitness cost associated with repeated reflex bleedings as alkaloid production would siphon off only a small amount of metabolic energy per day, although the loss of body fluids may also incur costs (Higginson et al. 2011). It has been argued that the combination of aposematic coloration and costly defensive chemistry should lead to the development of automimics within a population (Holloway et al. 1991). The driving force of evolution should act on the cost associated with the synthesis of chemical defenses to produce defensive ‘cheaters’, however, Holloway et al. (1991) and de Jong et al. (1991 found no evidence of this strategy in C. septempunctata or A. bipunctata. Our findings support the conclusions of Holloway and de Jong, that there is no physiological cost associated with the production of defensive secretions for adult lady beetles on which natural selection can act to produce automimics. H. axyridis females expelled significantly less reflex blood compared to the two Coccinella species. This may be related to the fact that the primary defensive alkaloid in the reflex blood of both C. septempunctata and C. novemnotata is N-oxide coccinelline, versus harmonine in H. axyridis (Haulotte et al. 2012), and harmonine has been shown to be more toxic in bioassays than coccinelline (Nedvěd et al. 2010). As previous experiments have shown that different defensive alkaloids synthesized by coccinellids can vary substantially in their toxicity (Marples et al. 1989, Daloze et al. 1994), it is not unreasonable to assume that the lower quantity of reflex blood exuded by H. axyridis may be related to harmonine’s biochemical properties like toxicity. C. septempunctata was the only species to show a slight, although significant, decrease in the mass of reflex blood exuded over time when provided the low diet treatment. The influence of prey limitation on C. septempunctata’s ability to produce eggs and maintain supplies of defensive carrier fluid over time suggests that this species has an apparent intolerance to prey scarcity over the time frame of the experiment. This may help to explain the success of H. axyridis over C. septempunctata in North America in recent decades. The low diet treatment’s significant effect on reducing the total number of eggs laid and egg viability over time in C. septempunctata and C. novemnotata confirms the notion that Coccinellids have decreased reproductive output when prey is scarce. Therefore the advantages that C. septempunctata has in scramble competition as evidenced by Hoki et al. (2014) could significantly affect the size and reproductive success of the native C. novemnotata (Losey et al. 2012). However, C. novemnotata was able to lay 26.7% more eggs than C. septempunctata in the low diet treatment, maintaining higher reproductive output in a prey-limited environment. This may help to explain C. novemnotata’s ability to survive in the low-quality marginal environments to which they have been displaced (Losey et al. 2014). For C. septempunctata and C. novemnotata, the lack of significance of diet levels on the total number of eggs laid on day 1 and day 2 but significance on day 3 indicates a lag in the effects of diet treatment (Fig. 1). As all beetles were provided aphids ad libitum up to the day prior to the start of the experiment, evidence of the restricted diet did not reveal itself immediately. This supports both species’ ability to tolerate short-term prey scarcity and produce a substantial number of eggs post-mating. The lack of a significant effect of diet restriction on the total number of eggs laid by H. axyridis over time is evidence of their greater ability to tolerate prey scarcity than C. septempunctata and C. novemnotata. Previous experiments in our laboratory have also demonstrated H. axyridis’s competitive advantage over C. novemnotata in intraguild predation, with H. axyridis surviving significantly more often than C. novemnotata when paired in a prey-limited environment (Ducatti et al. 2017). H. axyridis combined advantages in intraguild predation and tolerance of prey scarcity as evidenced by our results, in addition to their aggressive behavior and high predation efficiency (Leppanen et al. 2011), may help to explain their explosive success across North America (Koch et al. 2008). This study is one of only a handful that has attempted to determine whether there are significant allocation costs associated with the de novo synthesis of chemical defenses in a predatory arthropod; none were found. We were able to highlight several differences in the amount of reflex blood expelled and tolerance to prey scarcity in C. septempunctata, C. novemnotata, and H. axyridis, which may aid in explaining the decline of C. novemnotata and the rise of H. axyridis in North America. We have also identified a possible mechanism to explain how lady beetles that deploy their chemical defenses are able to lay more viable eggs compared to no-bleed controls and how aphid abundance may affect this. It is important to note that this study only investigated a subset of potential physiological costs, and others, like the impact of bleeding on larval development and survival, and ecological costs like increased susceptibility to predation, pathogens, and parasitism, may exist. A deeper more thorough evaluation of the costs of de novo synthesis of chemical defense is needed in both herbivores and predators to better understand how autogenously produced defenses evolve. Acknowledgments We would like to thank our undergraduate student assistants for technical assistance, the National Science Foundation for project and salary support, the Cornell University Agricultural Experiment Station for a Student Hatch Grant, and the Cornell Statistical Consulting Unit for assisting with analyses. We would also like to thank EE Zvereva for providing constructive comments on an earlier version of the manuscript. References Cited Agrawal , A. A. , A. P. Hastings , M. T. Johnson , J. L. Maron , and J. P. Salminen . 2012 . Insect herbivores drive real-time ecological and evolutionary change in plant populations . Science . 338 : 113 – 116 . Google Scholar CrossRef Search ADS PubMed Bennett , R. N. , and R. M. Wallsgrove . 1994 . Secondary metabolites in plant defense mechanisms . New Phytol . 127 : 617 – 633 . Google Scholar CrossRef Search ADS Blum , M. S . 1981 . Chemical defenses of arthropods . Academic Press , London, United Kingdom . Bowers , M. D . 1992 . The evolution of unpalatability and the cost of chemical defense in insects , pp. 216 – 244 . In B. D. Roitberg and M. B. Isman (eds.), Insect chemical ecology: an evolutionary approach . Chapman & Hall , New York, NY . Camara , M. D . 1997 . Physiological mechanisms underlying the costs of chemical defense in Junonia coenia Hübner (Nymphalidae): a gravimetric and quantitative genetic analysis . Evol. Ecol . 11 : 451 – 469 . Google Scholar CrossRef Search ADS Codella , S. G. , Jr , and K. F. Raffa . 1995 . Host plant influence on chemical defense in conifer sawflies (Hymenoptera: Diprionidae) . Oecologia . 104 : 1 – 11 . Google Scholar CrossRef Search ADS PubMed Cody , M. L . 1966 . A general theory of clutch size . Evol . 20 : 174 – 184 . Google Scholar CrossRef Search ADS Daloze , D. , J. C. Braekman , and J. M. Pasteels . 1994 . Ladybird defense alkaloids: structural, chemotaxonomic and biosynthetic aspects (Col.: Coccinellidae) . Chemoecol . 5 : 173 – 183 . Google Scholar CrossRef Search ADS Ducatti , R. D. , T. A. Ugine , and J. Losey . 2017 . Interactions of the Asian lady beetle, Harmonia axyridis (Coleoptera: Coccinellidae), and the North American Native Lady Beetle, Coccinella novemnotata (Coleoptera: Coccinellidae): prospects for recovery post-decline . Environ. Entomol . 46 : 21 – 29 . Google Scholar PubMed Eisner , T. , M. Eisner , and M. Siegler . 2007 . Secret weapons: defenses of insects, spiders, scorpions, and other many-legged creatures . Belknap Press , Cambridge, United Kingdom . Feeny , P . 1976 . Plant apparency and chemical defense , pp. 1 – 40 . In J. Wallace and R. Mansell (eds.), Biochemical interaction between plants and insects . Springer , NY, US . Gardiner , M. M. , M. E. O’Neal , and D. A. Landis . 2011 . Intraguild predation and native lady beetle decline . PLoS One . 6 : e23576 . Google Scholar CrossRef Search ADS PubMed Gordon , R. D . 1985 . The coccinellidae (Coleoptera) of America North of Mexico . NY Entomol. Soc . 93 : 1 – 912 . Grill , C. P. , and A. J. Moore . 1998 . Effects of a larval antipredator response and larval diet on adult phenotype in an aposematic ladybird beetle . Oecologia . 114 : 274 – 282 . Google Scholar CrossRef Search ADS PubMed Hare , J. D . 2012 . How insect herbivores drive the evolution of plants . Science . 338 : 50 – 51 . Google Scholar CrossRef Search ADS PubMed Harmon , J. P. , E. Stephens , and J. Losey . 2006 . The decline of native coccinellids (Coleoptera: Coccinellidae) in the United States and Canada , pp. 85 – 94 . In Beetle conservation . Springer , Netherlands . Haulotte , E. , P. Laurent , and J. C. Braekman . 2012 . Biosynthesis of defensive Coccinellidae alkaloids: incorporation of fatty acids in adaline, coccinelline, and harmonine . Europ J. Org. Chem . 2012 : 1907 – 1912 . Google Scholar CrossRef Search ADS Higginson , A. D. , J. Delf , G. D. Ruxton , and M. P. Speed . 2011 . Growth and reproductive costs of larval defence in the aposematic lepidopteran Pieris brassicae . J. Anim. Ecol . 80 : 384 – 392 . Google Scholar CrossRef Search ADS PubMed Hoki , E. , J. Losey , and T. A. Ugine . 2014 . Comparing the consumptive and non-consumptive effects of a native and introduced lady beetle on pea aphids (Acyrthosiphon pisum) . Biol. Control 70 : 78 – 84 . Google Scholar CrossRef Search ADS Holloway , G. J. , P. W. de Jong , P. M. Brakefield , and H. de Vos . 1991 . Chemical defense in ladybird beetles (Coccinellidae). I. Distribution of coccinelline and individual variation in defence in 7-spot ladybirds (Coccinella septempunctata) . Chemoecol . 2 : 7 – 14 . Google Scholar CrossRef Search ADS de Jong , P. W. , G. J. Holloway , P. M. Brakefield , and H. de Vos . 1991 . Chemical defense in ladybird beetles (Coccinellidae). II. Amount of reflex fluid, the alkaloid adaline and individual variation in defense in 2-spot ladybirds (Adalia bipunctata) . Chemoecol . 2 : 15 – 19 . Google Scholar CrossRef Search ADS Kajita , Y. , J. J. Obrycki , J. J. Sloggett , and K. F. Haynes . 2010 . Intraspecific alkaloid variation in ladybird eggs and its effects on con- and hetero-specific intraguild predators . Oecologia . 163 : 313 – 322 . Google Scholar CrossRef Search ADS PubMed Koch , R. L. , and T. L. Galvan . 2008 . Bad side of a good beetle: the North American experience with Harmonia axyridis . BioControl . 53 : 23 – 35 . Google Scholar CrossRef Search ADS Leppanen , C. , A. Alyokhin , and S. Gross . 2011 . Competition for aphid prey between different lady beetle species in a laboratory arena . Psyche: J. Entomol . 2012 . Losey , J. , J. Perlman , J. Kopco , S. Ramsey , L. Hesler , E. Evans , L. Allee , and R. Smyth . 2012 . Potential causes and consequences of decreased body size in field populations of Coccinella novemnotata . Biol. Control 61 : 98 – 103 . Google Scholar CrossRef Search ADS Losey , J. E. , L. L. Allee , E. Stephens , R. R. Smyth , P. Priolo , L. Tyrrell , S. Chaskey , and L. Stellwag . 2014 . Lady beetles in New York: insidious invasions, erstwhile extirpations, and recent rediscoveries . North East. Nat . 21 : 271 – 284 . Google Scholar CrossRef Search ADS Majerus , M. E. , J. J. Sloggett , J. F. Godeau , and J. L. Hemptinne . 2007 . Interactions between ants and aphidophagous and coccidophagous ladybirds . Pop. Ecol . 49 : 15 – 27 . Google Scholar CrossRef Search ADS Marples , N. M. , P. M. Brakefield , and R. J. Cowie . 1989 . Differences between the 7-spot and 2-spot ladybird beetles (Coccinellidae) in their toxic effects on a bird predator . Ecol. Entomol . 14 : 79 – 84 . Google Scholar CrossRef Search ADS Marples , N. M. , W. van Veelen , and P. M. Brakefield . 1994 . The relative importance of colour, taste and smell in the protection of an aposematic insect Coccinella septempunctata . Anim. Behav . 48 : 967 – 974 . Google Scholar CrossRef Search ADS Marques , R. V. , R. A. Sarmento , F. Lemos , M. Pedro-Neto , M. W. Sabelis , M. Venzon , A. Pallini , and A. Janssen . 2015 . Active prey mixing as an explanation for polyphagy in predatory arthropods: synergistic dietary effects on egg production despite a behavioural cost . Funct. Ecol . 29 : 1317 – 1324 . Google Scholar CrossRef Search ADS Nazareth , T. M. , and G. Machado . 2015 . Egg production constrains chemical defenses in a neotropical arachnid . PLoS One . 10 : e0134908 . Google Scholar CrossRef Search ADS PubMed Nazareth , T. M. , D. B. Sudatti , and G. Machado . 2016 . Chemical defense as a condition-dependent trait in harvestmen . J. Chem. Ecol . 42 : 1047 – 1051 . Google Scholar CrossRef Search ADS PubMed Nedvěd O , P. Kalushkov , X. Fois , D. Ungerová , A. Rozsypalová . 2010 . Harmonia axyridis: six-legged alligator or lethal fugu? IOBC/WPRS Bulletin 58: 65–68 . Obrycki , J. J. , and T. J. Kring . 1998 . Predaceous Coccinellidae in biological control . Annu. Rev. Entomol . 43 : 295 – 321 . Google Scholar CrossRef Search ADS PubMed Pasteels , J. M. , C. Deroe , B. Tursch , J. C. Braekman , D. Daloze , and C. Hootele . 1973 . Distribution et activites des alcaloides defensifs des Coccinellidae . J. Insect. Phys . 19 : 1771 – 1784 . Google Scholar CrossRef Search ADS Pasteels , J. M. , J. C. Gregoire , and M. Rowell-Rahier . 1983 . The chemical ecology of defense in arthropods . Annu. Rev. Entomol . 28 : 263 – 289 . Google Scholar CrossRef Search ADS Perry , J. C. , and B. D. Roitberg . 2005 . Ladybird mothers mitigate offspring starvation risk by laying trophic eggs . Behav. Ecol. Sociobiol . 58 : 578 – 586 . Google Scholar CrossRef Search ADS Rowell-Rahier , M. , and J. M. Pasteels . 1986 . Economics of chemical defense in chrysomelinae . J. Chem. Ecol . 12 : 1189 – 1203 . Google Scholar CrossRef Search ADS PubMed Ruxton , G. D. , T. N. Sherratt , and M. P. Speed . 2004 . Avoiding attack: the evolutionary ecology of crypsis, warning signals, and mimicry . Oxford University Press , Oxford, United Kingdom . SAS Institute . 2011 . SAS system for personal computers, release 9.3 . SAS Institute , Cary, NC . SAS Institute Inc . 2013 . JMP, version 11 . SAS Institute Inc ., Cary, NC . Sato , S. , K. Kushibuchi , and H. Yasuda . 2009 . Effect of reflex bleeding of a predatory ladybird beetle, Harmonia axyridis (Pallas)(Coleoptera: Coccinellidae), as a means of avoiding intraguild predation and its cost . App. Entomol. Zool . 44 : 203 – 206 . Google Scholar CrossRef Search ADS Snyder , W. E. , and E. W. Evans . 2006 . Ecological effects of invasive arthropod generalist predators . Annu. Rev. Ecol. Evol. Syst . 37 : 95 – 122 . Google Scholar CrossRef Search ADS Trougakos , I. P. , and L. H. Margaritis . 2002 . Novel morphological and physiological aspects of insect eggs , pp. 3 – 31 . In M. Hilker , T. Meiners (eds.), Chemoecology of insect eggs and egg deposition . Blackwell Publishing , Oxford, United Kingdom . Tumminello , G. , T. A. Ugine , and J. E. Losey . 2015 . Intraguild interactions of native and introduced coccinellids: the decline of a flagship species . Environ. Entomol . 44 : 64 – 72 . Google Scholar CrossRef Search ADS PubMed Turnipseed , R. K. , T. A. Ugine , and J. E. Losey . 2014 . Effect of prey limitation on competitive interactions between a native lady beetle, Coccinella novemnotata, and an invasive lady beetle, Coccinella septempunctata (Coleoptera: Coccinellidae) . Environ. Entomol . 43 : 969 – 976 . Google Scholar CrossRef Search ADS PubMed Ugine , T. A. , and J. E. Losey . 2014 . Development times and age-specific life table parameters of the native lady beetle species Coccinella novemnotata (Coleoptera: Coccinellidae) and its invasive congener Coccinella septempunctata (Coleoptera: Coccinellidae) . Environ. Entomol . 43 : 1067 – 1075 . Google Scholar CrossRef Search ADS PubMed Zvereva , E. L. , and M. V. Kozlov . 2016 . The costs and effectiveness of chemical defenses in herbivorous insects: a meta-analysis . Ecol. Monogr . 86 : 107 – 124 . Zvereva , E. L. , V. Zverev , O. Y. Kruglova , and M. V. Kozlov . 2017 . Strategies of chemical anti-predator defences in leaf beetles: is sequestration of plant toxins less costly than de novo synthesis ? Oecologia . 183 : 93 – 106 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of Entomological Society of America. All rights reserved. For permissions, please e-mail: firstname.lastname@example.org. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Environmental Entomology – Oxford University Press
Published: Aug 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera