One of the primary pathways in the both the humoral and cellular immune responses is the prophenol-oxidase (PPO) cascade, which results in the oxidation of tyrosine derivatives such as L-DOPA by activated phenol oxidase (PO), and deposition of melanin on the invasive body [12,62,63]. The product of the PPO pathway is measured using absorbance in spectrophotometric assays. Hemolymph samples are incubated in buffer solution with serine proteases such as chymotrypsin [64] added to activate PO. A substrate such as L-DOPA is added to allow the enzymatic oxidation to proceed. The change in absorbance over time in the product is measured using spectrophotometry, and correlated with enzyme activity and thus immune response. The measurement of PO activity is strongly associated with susceptibility to certain pathogens and arthropod endoparasitoids [59,65], but is not correlated with susceptibility to other pathogens (such as viruses [57,66]). Measurement of PO activity is less parasite-specific than the two following assays, both of which simulate endoparasitoids. Assays of PO activity assays require only hemolymph samples rather than destroying the insect to find and remove an implant, which provides four advantages to the remaining implant-based techniques: (1) it is more easily performed in natural systems; (2) there is no wait for melanization in the hemocoel, and the sample isn’t lost if the insect dies during implantation; (3) one insect can be tested multiple times; (4) a parasitized insect can be tested, then kept alive to allow continued parasite development or emergence.

Solitary endoparasitoids are those in which single offspring emerge successfully from a single host; the successful endoparasitoid offspring may emerge alone because it was the sole invader, or it killed competing larvae inside its host [67,68]. Solitary development is considered the ancestral trait in Hymenopteran endoparasitoids [69], which more commonly exhibit this lifestyle compared to other endoparasitoid orders [68].

Solitary endoparasitoids have been simulated using 2.0 × 0.20 mm nylon microfilament implants placed in the hemocoel of the target insects for a pre-determined amount of time (as short as 1 hr [65,70]) during which encapsulation and melanization of the implant both occur. Implants are photographed upon removal, and the size of the hemocyte capsule and the color of the melanized implant can both be measured using various imaging software packages such as ImageJ, Image Pro, and Adobe Photoshop. Color is scored using grey color caused by melanization, and a non-implanted control microfilament is used as a comparison for both color and size after encapsulation and melanization.

Gregarious development, the successful development of at least two endoparasitoid larvae in a single host, is considered a derived character in the Hymenoptera [69] and a general feature of endoparasitoids belonging to other orders such as Coleoptera and Diptera [68]. Gregarious development may be the result of superparasitism (several adult female parasitoids insert eggs into the same individual host, or a single female repeatedly oviposits into the same individual host) or large clutches of eggs inserted during one oviposition event.

Gregarious endoparasitoids have been simulated by injecting a group of glass beads in physiological saline into the hemocoel. Sephadex chromatography beads, such as A-25 beads with a diameter of 40–120 µm, are commonly used [71,72,73]. A fixed volume of 10–15 µL saline + beads is injected into the hemocoel, and beads are individually recovered from caterpillars dissected after a pre-determined amount of time has passed for melanization to occur. As with filament implants, the size and color of the capsule can both be measured using images of each bead. Beads are often dyed red to facilitate retrieval; in this case the red color of melanized and control beads is compared to calculate melanization (Figure 1).

Generalist and specialist herbivores both encounter variation in host plant quality traits while foraging. This variation may be the result of the variation provided by multiple plant species, and variation within a plant species, population, or individual. Generalist herbivores exhibit two foraging strategies; individuals of herbivore species considered generalists at the individual level may sample and consume many plant species [24,79], while those that are generalists at the population level may remain on a single plant individual or members of a single plant species [80,81]. Individual specialist herbivores generally follow a similar strategy to the latter, remaining on a single plant species or even a single plant individual throughout their lifespan [20].

Foraging on several plants species can provide herbivores with wide qualitative variation in nutritional and defensive compounds. A mixed diet throughout an individual’s lifespan provides many benefits from a multitrophic perspective, such as improving fecundity, and growth rate [82,83], and pharmacophagy (see [48], Section 2). However, a mixed diet may not necessary improve immune response. For instance, generalist Parasemia plantaginis (Lepidoptera: Arctiidae) larvae fed a 3-species mixed diet had a weaker immune response than those feeding on single plant species [84]. In this case, the mixed diets may have diluted the intake of compounds beneficial to growth and the immune response.

Plant defensive and nutritive traits can vary dramatically at the population level and the individual plant level (through ontogenetic changes and herbivore-induced changes as well as variation among tissues). This variability exposes herbivores consuming a single plant species to a wide variety of biotic conditions, even under similar environmental conditions. Epirrita autumnata (Lepidoptera: Geometridae) a generalist that feeds on several deciduous tree species [85], with cyclic population outbreaks, has provided several insights in this regard. Over several seasons and outbreaks, E. autumnata immune response has been found to be higher on high-quality trees [61], low-quality trees [65,86], induced trees [70], and alternative tree species [87]. The generally enhanced immune response on low-quality diets in this species (with the exception of [61]) may reflect a tradeoff between growth and immune response, as larvae stressed by diet quality may allocate higher proportions of resources to survival.

Plant quality is often correlated with immune responses of both specialists and generalists. A survey of published research examining 10 generalist and specialist Lepidoptera species found that cellular immune response (measured as PO activity, implants, and bead injections (Table 1)) was higher in 5 species (counting E. autumnata [61]) when reared on or collected from a “high quality” or preferred plant, as determined by measuring other fitness correlates such as growth rate and larval or pupal mass. Only the generalist E. autumnata showed a reduced response and the others showed no change in immune response when reared on high-quality plants compared to low-quality plants (Table 1). The general association between high-quality diets and strong immune responses in this limited sample broadly fits a pattern showing that invertebrates experiencing favorable conditions also show strengthened immune responses [88,89]. Plant identity was associated with significantly strengthened or weakened immune responses in all of the specialist species and all of the generalist species except Grammia incorrupta (which showed a consistent immune response regardless of host plant species) (Table 1), supporting generally the hypothesis that lower trophic levels influence higher trophic interactions.

Melanization and encapsulation responses of insects presented with foreign implants have been clearly linked to nutritional status. Starvation has been shown to reduce immune responses [4,57,94], as have low-quality diets that induce poor nutritional efficiencies [73,90,95]. Nutritional macromolecules such as amino acids and proteins provided by plants have important effects on herbivore growth and fitness [96], the role of these molecules in enhancing immune responses has received recent attention as well. The quality and quantity of nutritional resources are both important to immune response; for example, high-quality protein may contribute more to the nitrogen pool that the immune response draws from [95].

The PPO cascade that leads to melanization of foreign objects generates abundant reactive oxygen species that can harm the insect [97,98,99]. Plant-provided antioxidant compounds such as flavonoids, phenolics, and carotenoids remove these reactive oxygen species and have been shown to enhance immune response [100,101]. The effects of antioxidants on the immune response are of particular interest in evolution of host plant choices of generalist species. For instance, encapsulation ability of the generalist arctiid Parasemia plantaginis fed 3 plant species and an artificial diet was highest when fed plant species contributing compounds high in antioxidant activity such as flavonoids and carotenoids [84]. Moreover, encapsulation responses of other generalist caterpillars reared on Taraxacum officinale and Malva parviflora have been shown to be consistently high [73,75], and both plants are known to produce high levels of antioxidant flavonoids [102,103]. As mentioned above (Section 2), Grammia incorrupta larvae parasitized by dipteran endoparasitoids preferentially select flavonoid-rich M. parviflora over plants providing defensive compounds that can be sequestered [73].

Further evidence that plant antioxidants and other secondary compounds increase encapsulation and melanization comes from comparisons between herbivores reared on whole plants and artificial diets; recipes for the latter include primary nutrients, preservatives, and antibiotics but lack plant secondary compounds unless added separately. Both generalist and specialist herbivores benefit from diets enriched with these compounds. Weakened immune responses have been found in both generalist [73,84] and specialist herbivores [90] reared on artificial diets compared to plant diets. Thus, although a possible link between antioxidants and a strengthened immune response has been observed in a limited number of studies, further research is needed to determine the extent to which parasitism status induces a preference for plants high in antioxidants compared to other compounds.

One of the key differences between specialist and generalist herbivores is response to consuming plant defensive compounds [20]. Many specialist herbivores possess enzyme systems highly specialized to metabolize specific substrates. Substrate-specialized enzyme systems may be ineffective at metabolizing compounds structurally dissimilar to the substrate, causing specialists to respond poorly to novel compounds. Generalist herbivores are able to consume many more plant species and their various associated compounds due to broad-substrate enzyme systems [104,105]. Although substrate-specific enzymes are present in generalist herbivores [106], the prevalence in generalists of broad-substrate enzyme systems often makes them more susceptible to deleterious effects of defensive compounds compared to specialists. The reduced performance of specialists consuming novel compounds, and both specialists and generalists consuming large amounts of defensive compounds can include a weakened immune response.

Metabolizing large amounts of consumed plant compounds may be energetically expensive regardless of enzyme system specificity. These costs may divert needed resources away from immune responses. Increased dietary levels of glucosinolates and iridoid glycosides, two plant compounds associated with reduced digestive efficiencies [73,90,107,108] have been shown to have detrimental effects on the immune response of several generalist and specialist Lepidopteran species (Table 2).

Performance measures such as growth rate, size, nutritional indices were also reduced along with immune response as higher doses of these compounds were consumed [75,90,93]. One specialist, Ceratomia undulosa (Lepidoptera: Sphingidae), was fed leaf discs supplemented with a novel compound (catalpol), with toxic effects of this compounds potentially contributing to its weakened immune response at a low dose [75]. A negative correlation between hydrolyzable tannins in host tree leaves and Epirrita autumnata immune response (with a positive correlation between hydrolyzable tannins and growth) was the only exception found to the general pattern of immune response mirroring other performance measures as a function of host plant chemistry [86], likely due to a strong tradeoff in this species between investments in growth versus immune response.

Physiological and behavioral differences between generalists and specialists may explain instances in which the immune response of the former is not influenced by secondary compound consumption. Immune response of the grazing generalist Grammia incorrupta is not affected when consuming high doses of either iridoid glycosides or pyrrolizidine alkaloids (references in Table 2). Unlike many other generalists and most specialists, individual G. incorrupta sample several plants providing potentially several distinct chemical classes. The consistently high immune responses of G. incorrupta measured regardless of diet may be explained by broad-substrate enzyme systems that do not compete for resources with the immune response, unique enzyme systems in the immune response itself, or some other unique physiological trait of this species. More research is needed to determine if other generalists with a similar feeding style (e.g., other Arctiidae and several groups of Orthoptera) are similarly able to maintain a high immune response regardless of type and dose of dietary defensive compounds.

Sequestration of plant secondary compounds, storing them in the hemocoel or other tissues in concentrations higher than the plant producing them, is common among both specialist and generalist herbivores [110,111], although specialist herbivores may sequester higher concentrations of plant compounds compared to generalists [43,111,112]. Sequestration is noted as an effective form of chemical defense against predators, pathogens, and endoparasitoids [111]. However, several cases have been found in which endoparasitoids are as or more successful in hosts sequestering higher levels of plant compounds compared to those in hosts that do not sequester [113,114,115]. Smilanich et al. [90] have proposed the “vulnerable host hypothesis” to explain this phenomenon, positing that defensive compound sequestration and the immune response can be antagonistic metabolic processes.

Competing metabolic demands between sequestration and immune responses that show a tradeoff between the two processes are central to supporting the “vulnerable host hypothesis.” Plant compounds sequestered in the hemocoel are often modified by enzymes before storage (e.g., [116]) and cross gut epithelia with the help of transport proteins to reach their destination, both energetically expensive processes that may compete with immune response as well as a variety of other life functions [117,118,119]. Evidence that sequestration is energetically expensive was provided by path analysis models revealing that the specialist Junonia coenia (Lepidoptera: Nymphalidae) experienced reduced respiration rates as the amount of the iridoid glycoside catalpol sequestered increased [90]. Increasing amounts and concentrations of catalpol sequestered by J. coenia has been associated repeatedly with reduced nutritional efficiency and weakened immune response ([75,90,108], Figure 2,), while simultaneously reducing susceptibility to invertebrate predators [120,121]. The use of catalpol sequestration to test the “vulnerable host hypothesis” has been extended to two other herbivores as well; melanization response was negatively associated with sequestration in the specialist Ceratomia catalpae (Lepidoptera: Sphingidae),but not the generalist Spilosoma congrua (Lepidoptera: Arctiidae) (Figure 2). Specialized C. catalpae enzyme and carrier systems may require more energy for accumulating high catalpol concentrations (5x the concentrations accumulated by S. congrua), while S. congrua may lack these energy-intensive transport systems.