Discussion
The results of our study show that both macronutrient intake and a simulated infection threat influence the expression of life history traits in decorated crickets, G. sigillatus . Specifically, both diet and an immune challenge with heat-killed bacteria: 1) affected male reproductive effort, quantified here as male calling effort and 2) greatly altered male survival. The two diets used in the current study were deliberately selected because they had been shown previously to either maximize calling effort (P:C = 1:8, high carbohydrate diet) or to maximize immune function, at least with respect to encapsulation ability (P:C = 5:1, high protein diet) (Rapkin et al., 2018), thereby making it more likely that the effects of our immune challenge would be contingent on the diet on which males had been maintained. However, we found no or, at best, weak evidence of an interaction between the diet on which males were maintained and the level of the immune challenge that they experienced in the expression of life-history traits. Moreover, although our results aligned with the anticipated effects of these two diets on male calling, surprisingly, they were in apparent contradiction of the anticipated effect on male immunity. We elaborate on these differences below and seek to identify potential mechanisms mediating observed treatment effects and the factors potentially constraining an interaction between diet and the magnitude of an immune challenge on male life history strategy.
Males maintained on a high protein diet exhibited significantly higher mortality than those held on the high carbohydrate diet. Nearly 30% (68 out of 249) of males maintained on the high protein diet died prior to end to the experiment compared with less than 14% (33 out of 245) of males held on the high carbohydrate diet. This is consistent with numerous findings across taxa showing that lifespan is truncated in animals fed a high protein, low carbohydrate diet (reviewed in Le Couteur et al., 2016 and Simpson et al., 2017), including G. sigillatus (J. Hunt, unpublished data). These recent discoveries, which contrast with conventional Y resource allocation models (van Noordwijk and de Jong, 1986; Zera and Harshman, 2001), have inspired the lethal protein hypothesis (Lee et al., 2008; Fanson et al., 2009; Simpson and Raubenheimer, 2009). Proposed mechanisms mediating the lethal protein hypothesis include increased mitochondrial generation of radical oxygen species (Sanz et al., 2004), changes in the relationship between insulin/IGF-1 and amino acid signaling (e.g., TOR) pathways (Kapahi et al., 2004), and damage to organs from nitrogenous waste products. Although further studies are needed to conclusively determine whether the diet-induced mortality in the present study is due to an excess of protein or a deficit of carbohydrates, it is unlikely due to the latter as males were always provided more food than they could eat, suggesting that they were not carbohydrate limited. Males held on the high protein diet consumed a greater amount of food than those held on the high carbohydrate diet across the entire range of body sizes, which could have been due to a greater nutritional demand for carbohydrates in the former. Indeed, Rapkin et al. (2018) showed that when male G. sigillatus are given a dietary choice, they regulate their intake of protein and carbohydrate to a P:C ratio of 1:2.
As was the case with diet, infection cue dose had a significant effect on male survival through the end of the two-day calling period. Specifically, the highest infection cue dose resulted in higher male mortality than any of the other infection cue levels. This suggests that our simulated infection threat imposed a physiological cost on males, and that at the most severe infection threat level, this cost was manifest in increased mortality. This is consistent with a cost of using the immune system (Sadd and Schmid-Hempel, 2009), which has been shown to have consequences for survival in other insect taxa (Moret and Schmid-Hempel, 2000; Armitage et al., 2003). This has also previously been shown to be independent of nutritional condition in another cricket species (Jacot et al., 2004).
Both diet and infection cue treatment had significant effects on male reproductive effort, measured here as the time spent calling to attract a mate (termed calling effort). Calling effort was significantly higher in males held on the high carbohydrate diet than in those held on the high protein diet, consistent with previous work (Rapkin et al., 2018). We also found that infection cue, specifically at a low dose, impacts male reproductive effort. If they called, males injected with a low infection cue dose called more relative to naive controls. This increased calling effort is consistent with a pattern of terminal investment (Duffield et al., 2017), but, surprisingly, neither of the two higher infection cue treatments differed from the naive control group with respect to calling effort. However, at least in the case of males injected with the high infection cue dose, males experienced at least a two-fold higher mortality during the calling period compared with males in the other infection cue treatments, and thus, males in this group were, to some extent, self-selected. If the males that died were those that were more inclined to terminally invest, whereas those that survived prioritized maintenance over reproduction, then this would attenuate any effect of infection cue treatment on calling effort at the more severe infection threat levels.
Although diet and infection cue treatment independently influenced male calling effort, we found no evidence of an interaction between these two factors. The absence of such an interaction contrasts with an early study demonstrating an interaction between male age and infection cue dose, in which older male G. sigillatus increased their calling effort in response to the same graded increase in infection threat used here, whereas younger males did not (Duffield et al., 2018). Why do males retain this level of plasticity in calling effort with respect to age, but not diet? One possibility is that age overrides any influence of diet on the propensity to shift life history strategy, given that it has been previously shown that males at the age used in this study exhibit terminal investment, whereas younger males do not (Duffield et al., 2018). A standard lab diet was provided in this previous study, but it is possible that terminal investment in younger males could be elicited by certain combinations of diet and infection cue dose, in line with the dynamic terminal investment threshold model (Duffield et al., 2017). An alternative is that males were terminally investing in other important components of reproductive effort that we did not measure in the present study. For example, the spermatophore transferred by males at copulation includes a large gelatinous mass, the spermatophylax, that the female consumes after mating as a nuptial food gift, which is critical to male fertilization success (Sakaluk, 1984; Sakaluk, 1986; Sakaluk and Eggert, 1996; Eggert et al., 2003). Duffield et al. (2015) showed that immune-challenged males terminally invest by altering the free amino acid profile of the spermatophylax, enhancing its gustatory appeal to females (see also Gershman et al., 2012). Moreover, male chemical cues, in the form of cuticular hydrocarbons (CHCs), greatly influence whether a female mounts a male, a necessary antecedent to copulation (Weddle et al., 2013; Capodeanu-Nägler et al., 2014). The CHC profile of a male, and by extension, his attractiveness, can be significantly influenced by his nutritional environment (Weddle et al. 2012). Both CHC expression (Rapkin et al., 2016) and the amino acid composition of the spermatophylax (Rapkin et al., 2017a) are optimized at a P:C ratio of 1:1.3 and 1:1.5, respectively, ratios far removed from the two diets offered in the present study.
Despite our initial expectation, we found no evidence that males held on the high protein diet exhibit enhanced immune function. Specifically, there was no positive effect of diet on the number of circulating hemocytes, presence of microaggregations, or antibacterial activity of the hemolymph. We did, however, find a significant interaction between diet and infection cue dose on the incidence of circulating microaggregations of hemocytes, suggesting that both factors could affect this immune response. This contrasts with the results of a previous study showing that among 24 diets differing in P∶C ratio and distributed along six nutritional rails, a P:C = 5:1 ratio optimizes immune function as assessed using an encapsulation assay (Rapkin et al., 2018). In the present study, however, we did not measure the encapsulation ability of males, and the difference between the two studies may be a function of the different components of immunity that were measured. Different facets of immunity can be triggered by different types of threats or regulated independently, resulting in positive, negative, or no associations between immune components (Adamo, 2004; Forsman et al., 2008). Indeed, although Rapkin et al. (2018) demonstrated a significant effect of macronutrient intake on male encapsulation ability, they found no effect of diet on the activity of phenoloxidase, an important enzyme in the melanization cascade. Ultimately, we are cautious to draw any firm conclusions about the influence of diet and infection cue on immune parameters, due to the interesting but confounding differential effects of these factors on survival prior to the assaying of immunity.
In conclusion, both macronutrient intake and a simulated infection threat independently influenced the survival and reproductive effort of male G. sigillatus . There was evidence for terminal investment, as males increased calling effort at the low infection cue dose, but interpretation of responses at the higher threat levels was hampered by the differential mortality of males across diet and infection cue treatments, the latter demonstrating a cost of immune activation for survival. There was, however, no evidence of an interaction between diet and infection cue dose in their influence on calling effort, suggesting that the threshold for terminal investment was not contingent on diet, in contrast to earlier work documenting a shifting terminal investment threshold contingent on male age (Duffield et al., 2018). The absence of a dynamic terminal investment threshold may have been due to the previously documented influence of age masking other intrinsic and extrinsic factors, or, alternatively, males prioritizing investment in other components of reproductive effort. Regardless, how reproductive effort changes in accordance with the various intrinsic and extrinsic factors that alter an individual’s residual reproductive value remains a fertile area of inquiry (Duffield et al., 2017), but any generalizations must await additional comparative studies that preferably incorporate multiple facets of male reproductive effort.