The pupal stage of insects is relatively immobile, leaving the insect with no possibility to escape from predators; however, it is not defenseless. Indeed, insect pupas have evolved a wide range of passive antipredator strategies (Lindstedt et al., 2019), such as hiding in vegetation, cryptic coloration, camouflage, and mimicry (Wiklund and Sillén-Tullberg, 1985; Skelhorn et al., 2010; Gaitonde et al., 2018). Many pupas have also evolved secondary defense mechanisms, which may be chemical, physical, or behavioral (Lindstedt et al., 2019). The production of a cocoon by the larva before the pupal stage is a common form of mechanical protection in holometabolan insects. Such a cocoon can protect them from predators, parasitoids, and unsuitable climatic conditions (Craig, 1997). The pupas of some insects are capable of limited movements that may have a potentially deimatic (frightening) function against predators. For example, Cole (1959) noted a lower rate of parasitoid oviposition on butterfly hosts whose pupas are capable of intensive wriggling movements. Pupas of other insects also utilize protection by other organisms (Pierce et al., 2002; Hammer and Moran, 2019). Caterpillars of the moth Platyperpia virginalis (Boisduval) utilize spines of plants for pupation. The survival of pupas was higher in spiny plants than when spines were removed experimentally (Grof-Tisza et al., 2015). The most unique way in which pupas gain protection is by manipulating the behavior of host species (Mokkonen and Lindstedt, 2016). Pupas of lycaenid butterflies have evolved acoustic mechanisms to manipulate ants’ guarding behavior (Pierce et al., 2002; Barbero et al., 2009). In addition to a pair of stridulatory organs, Lemonias caliginea (Butler) pupas possess a pair of glands on the thorax that may produce a chemical attractive to ants (Ross, 1964). Behavioral manipulation of the host in order to obtain protection for the pupa is also typical for ichneumonid wasps (Ichneumonidae) from the Polysphincta group of genera (polysphinctines), which are ectoparasitoids and koinobionts exclusively associated with spiders as hosts (Eberhard and Gonzaga, 2019). Koinobiont parasitoids attack hosts that continue feeding and growing during parasitism. Their larvae live on the surface of a host, from which they suck the haemolymph, and upon which they gradually mature to the final instar. During the larval stage, the ichneumonid wasp is protected by an active spider host, which is not paralyzed and uses various antipredatory strategies. This unique protection provided by the live spider host is lost when the final instar larva kills and consumes the spider to reach the next stage of its ontogeny, the pupal stage. Shortly before pupation, the larva may induce an alteration in the spider’s web building behavior, resulting in the production of a “cocoon web,” a silk structure which serves as protection for the parasitoid pupa (Eberhard, 2000; Korenko and Pekár, 2011; Korenko et al., 2014, 2018, 2022; Gonzaga et al., 2015; Kloss et al., 2016; Takasuka et al., 2017; Eberhard and Gonzaga, 2019). The architecture of cocoon webs often resembles that of molting webs of non-parasitized spiders (Eberhard and Gonzaga, 2019). Web-building spiders actively protect themselves against enemies and adverse environmental conditions by seeking a safe place and/or by building their own shelter (the production of silk structures). For example, spiders often build their own shelter by combining silk and surrounding material (Cloudsley-Thompson, 1995). Shelter construction is an important antipredatory behavior not only for spiders (Manicom et al., 2008), but also for their koinobiont parasitoids. Some spiders use leaves as shelters, and the pupas of some polysphinctines remain concealed and protected in these structures (Gonzaga and Sobczak, 2007; Sobczak et al., 2014; Gonzaga et al., 2016).

The strategy of pupa protection in ichneumonid wasps associated with spiders is composed of two lines of defense, these combining the defenses of both the parasitoid and the spider host. The parasitoid larva produces a protective cocoon and the spider host produces a web. Both should serve as protection for the parasitoid during the pupal stage. A spider under manipulation by the penultimate and final instar larva provides a normal or modified web, which is used as mechanical protection against potential enemies. The parasitoid manipulates the web building behavior of the spider host, which often builds a unique silk structure. Web manipulation by the parasitoid larva has been reported in a number of spider species (e.g., Kloss et al., 2017; Korenko et al., 2018, 2022; Eberhard and Gonzaga, 2019; Gonzaga et al., 2022). All studies on the manipulation of spider web building behavior work with the hypothesis that the purpose of the modified spider web is to protect the parasitoid during the pupal stage against natural elements and predators. This, however, has been supported by only three studies so far. Matsumoto (2009) reported that Agelena limbata Thorell (Agelenidae) individuals, when parasitized by Brachyzapus nikkoensis (Uchida), produced a silk-specific structure that blocked entry to crawling predators, such as ants, and thus reduced the predation risk faced by the parasitoid pupas. Gonzaga et al. (2010) documented that the reinforced cocoon webs produced by the spider Trichonephila clavipes (Linnaeus) increased protection for Hymenoepimecis wasps by preventing the cocoon from being knocked to the substrate. Kloss et al. (2016) observed that web modifications by Cyclosa spiders reduced the frequency of web ruptures and increased the chance of parasitoid adult emergence. Rain events were considered as the major cause of web ruptures and subsequent parasitoid mortality. In nature, parasitoid pupas can be predated by a range of arthropod and vertebrate predators (mainly birds). However, very limited evidence concerning the predation of the pupal stages of wasps from the Polysphincta group is available. We assume that predators that attack spiders and their cocoons (Pekár and Raspotnig, 2022) are potential predators of parasitoid pupas.

The protective efficiency against predators of the cocoon produced by the parasitoid pupa in conjunction with silk web structures provided by the spider host has not previously been investigated. We hypothesized that each of these two lines of defense provides a certain level of protection to the parasitoid pupa. We used several potential predators with various foraging strategies to test the effectiveness of these two defense lines. Ants, crickets, cockroaches and true bugs (Heteroptera) were used as model predators possessing the two most common types of mouthparts (chewing and sucking).

We found that the pupas of ichneumonid wasps are accepted as food by various potential predators, including insects with both chewing and sucking mouthparts. Of the six potential predators tested, from various foraging guilds, surprisingly, only the true predator, the Platymeris bug, showed no interest in the bare pupa. This lack of interest can be explained by the fact that many true predators hunt actively moving prey (Schmitz, 2010). Strangely, the granivorous Pyrrhocoris bug was an effective predator of ichneumonid wasp pupas. They presumably used pupas as an alternative source of nourishment and accepted this food probably due to its lack of movement, which is similar to seeds (Kristenová et al., 2011). The fact that pupal age, predator type, and predator size did not significantly affect the feeding latency with respect to bare pupas or the mortality of bare pupas shows that pupal protection is important for parasitoid survival.

Pupal mortality decreased considerably when the pupa was protected by a cocoon. Lasius ants successfully attacked bare pupas by carrying them away in their mandibles. Yet, when the pupa was protected by a cocoon, the ants lost interest. Insects are equipped with various highly sensitive food detectors (Hallem et al., 2006). It seems that the loss of direct contact between the ant and the pupa due to the barrier created by the cocoon probably resulted in a failure to detect potential prey. Predators with chewing mouthparts, namely the Gryllus cricket and the Shelfordella cocroach, caused high mortality to pupas protected by a cocoon. Protection by a cocoon significantly decreased pupa mortality only when Acheta crickets were used. A strong cocoon wall protected pupas against predators with sucking mouthparts, but was not effective against predators with chewing mouths. Cocroaches and crickets consumed both pupas and their protective cocoons. The protection provided by the cocoon was not sufficient, though it decreased the success of predation considerably in the case of suckers. A strong cocoon wall prolonged the latency to attack and theoretically increased pupa survival, though not against predators with chewing mouth parts. For example, the Pyrrhocoris bug was not able to suck pupas protected by a cocoon with a strong cocoon wall, though it was able to suck pupas in cocoons with a soft wall. True bugs possess a stylet which can pierce the prey’s cuticule and suck hemolymph (Alford, 1999). Our study focused on the effect of cocoon wall type, but the limited number of experiments with a particular predator did not allow us to sufficiently test predator responses to the cocoons of two different species. This is a challenge for future study.

Mortality during the pupal stage caused by predators has also been documented in various groups of holometabolan insects (Battisti et al., 2000; Heisswolf et al., 2009; Kollberg et al., 2014). Pupas use different methods of protection against predators, and their effectiveness often varies according to the predator taxa (Lindstedt et al., 2019). Pupas may avoid a predator attack through cryptic coloration (Wiklund and Sillén-Tullberg, 1985), chemical camouflage (Mizuno et al., 2018; Wang et al., 2018), or by pupating in the ground to avoid predators foraging on the surface (East, 1974). Alternatively, pupas can be aposematic in color, sound, or odor to announce their unsuitability as prey (Gaitonde et al., 2018; Lindstedt et al., 2019). Apart from the cocoon built by the parasitoid before pupation, physical defense in some insect pupas is also provided by hairs or spines originating at the larval stage. These can also play an important role in pupal defensive structures (Murphy et al., 2010; Sugiura and Yamazaki, 2014).

Different types of pupal defenses are combined to provide safety for the pupal development of various insects (Lindstedt et al., 2019). Ichneumonid wasp pupas use simple physical cocoon protection, which is common in other insects, but they have also evolved a unique technique to increase pupa survival – specifically, the manipulation of spider web building behavior to produce the so-called ‘cocoon web’ (Eberhard and Gonzaga, 2019). The phenomenon of host manipulation in parasitoids associated with spiders is manifested by the alteration of web building behavior (Eberhard and Gonzaga, 2019). Shortly before killing the spider, the penultimate instar larva of the parasitoid changes the web-building behavior of the spider host to produce a special silk (Eberhard and Gonzaga, 2019). Many parasitoids that associate with spiders building a dense 3D tangle web do not induce modification of the normal web (Fritzén and Fjellberg, 2014; Korenko, 2017). We tested here wasp species that mostly do not modify the web architecture. We found that the 3D silk structrures provided for the parasitoid pupas were highly protective against the tested predators. When the cocoon with the pupa was also protected by a spider web, pupa mortality decreased for all predators. Only predators with strong chewing mouthparts, namely crickets, were able to chew through the spider silk and attack the cocoon with the pupa. When the cocoon with the pupa was located in the air, the mortality decreased to only 5%. The cricket, A. domestica, was presumably able to climb the installed twig where the 3D web was located, damage it, and reach the pupa of Z. albicoxa, probably due to its smaller body size (16 mm).We did not consider here hyperparasitoids as predators. The protection of the pupa by the two lines of defense does not appear to be effective against the highly specialized hyperparasitoids attacking spider parasitoids. There are 27 documented cases of hyperparasitoids from the families Chalcididae, Eulophidae and Ichneumonidae attacking the pupas of ichneumonid wasps associated with spiders (Pádua et al., 2022). However, future observations in the field are needed to obtain insights into the impacts of predators on populations of ichneumonidwasp parasitoids associated with spiders.

We confirmed under laboratory conditions that ichneumonid wasps associated with spiders are accepted as prey by various predators. We found that insects with different foraging strategies and with different types of mouthparts are potential predators; however, further work needs to be done in the field in order to identify the true range of natural predators. This is because such observations have so far been scattered and very rare, and because targeted data collection from nature requires systematic long-term research. Future research should focus on spider predators, especially egg predators, because the parasitoid pupa is immobile like an egg sac and, indeed, can also resemble an egg sac to an egg predator or scavenger. The pupal stage is a risky period in the development of polysphincta ichneumonid wasps, and survival during this period depends on successful pupation and on remaining in (or on) the spider web, where the pupa is well protected not only by the cocoon created by the parasitoid final instar larva, but also by the web architecture created by the spider host. We provide the first evidence of the roles of these two lines of defense in the survival of ichneumonid wasp pupas. We found support for the hypothesis that the spider web is an important component of parasitoid defense during the pupal stage. This hypothesis is cardinal in the study of the behavioral manipulation of spiders induced by larvas of parasitoid ichneumonid wasps.

In brief, we found the two lines of pupa defense described above to have a significant function under laboratory conditions. Although conditions in nature differ from the artificial environment of the laboratory, predators from the tested foraging guilds are common in nature and therefore represent a potential danger for pupas. Thus, the existence of these defensive lines likely confirms the need to protect against predators in the pupal stage.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

SK, JS, and AK-A collected model organisms and performed experiments. SK designed the experiments and responsible for the integrity of this work. SK and SP analyzed the data and wrote the paper. All authors contributed to the article and approved the submitted version.

The study was financially supported by grant LTAUSA19084 of the Ministry of Education Youth and Sports of the Czech Republic. JS was supported by Czech University of Life Sciences Prague, Internal Faculty of Agrobiology, Food and Natural Resources, FAPPZ grant – 21150/1312/3152.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fevo.2022.1061700/full#supplementary-material