In this research, cabbage white butterflies (Pieris rapae) were raised on an artificial diet to examine differences in heavy metal toxicity. In doing so, this study provides general methods for rearing and laboratory studies of this easy-to-manipulate butterfly system. This discussion first considers more general questions about the methods reviewed here, then reviews our scientific findings before concluding with reflections on the components of the artificial diet.
The protocol reviewed here gives steps of a general rearing method for cabbage white butterflies, but there are many points within this protocol that can be tweaked. For instance, while the case study presented here uses sponges for feeding, other researchers have had luck with dental wicks and silk flowers filled with honey water5. While the present study uses honey water as food, other researchers have used sugar solutions and even Gatorade. If pupae need to be weighed, or moved to other conditions for emergence (e.g., inducing diapause and needing to cold store for 1 month), the researcher can easily remove them from the cups by spritzing them with water to moisten their silk attachments and grab them with feather forceps, then re-hanging them using double-sided tape. If researchers need more flexibility in terms of when adult butterflies are moved into cages for adult behavior, they can be held in the refrigerator for several weeks, but they need to be fed. Every several days, the butterflies should be taken out to be fed a dilute honey water solution. Under indoor lighting, this can be done by using a pin to unroll their proboscis into the food. On the adult performance end, a wide range of fitness measures can be taken on cabbage white butterflies. Body size can be measured as the wet or dry mass of larvae at certain stages, pupae, or adults (sacrificed, or held in glassine envelopes), or through the measurement of wing length in the program ImageJ (see12,24,25,28). The lifetime fecundity of females can be measured through daily egg collections on host plants25,69,70, and the size of specific traits can be measured as a metric of performance; for instance, the mass or volume of the brain or individual brain regions62,71,72, or the mass or protein content of the thorax or flight muscle62,70. Finally, adults can be used in behavioral studies to test any number of questions examining the effect of diet manipulation on foraging or oviposition choice27,73.
If the rearing protocol is not working as expected, there are a few aspects to troubleshoot. First, one can ask whether the light levels are high enough to elicit normal adult behavior. While lab-adapted lines of Pieris will lay eggs under fluorescent light, the only artificial light that works for wild-type lines are powerful broad-spectrum greenhouse lights. Natural light in greenhouses, windowsills, or outdoors works best to elicit mating and egg-laying behavior. Second, if eggs are not hatching or if larvae are dying early in development, there are a few things to consider. The host plant material must be organic, noting that "organic" plants from stores are sometimes treated with chemicals that can kill larvae, so raising one's own host plants is often best. If the host acceptance rate is lower, younger leaves with higher nitrogen content can be attempted, presenting potted plants instead of individual leaves and ensuring females are mated. Females will accept seeding Brassica, even small sprouts that are 2 weeks of age. The paraffin method works well to transfer eggs to different conditions, but it should be noted that the acceptance rate tends to be lower than whole plants. Third, all the components of the diet must be of high quality and not expired. Flaxseed oil should be replaced annually and stored in the fridge24,25. Wheat germ, the vitamin mix, and antibiotics should also be kept cool. Fourth, one can consider tweaking the diet cup setup. Any number of disposable plastic cup types can be used for rearing, from 1 oz to 15 oz. We have found that 4 oz is a good size to allow for adult emergence and packs nicely into our climate chambers. Holes poked in the lids allow for airflow, but too many holes can dry the diet in low humidity conditions, so this number may need to be adjusted. Fifth, the conditions in the climate chamber may need to be adjusted in combination with the cup conditions. If the conditions are too dry, host plants with eggs may dry out before larvae can be transferred, and cups with diet may dry out before butterflies emerge. On the other hand, if the conditions are too wet, the cups can harbor mold and disease. Researchers may need to adjust the airflow in cups through the use of mesh lids, or more or less holes in the lids. Another common issue is chamber lights that are bright enough to cause temperature swings in the cups and a build-up of condensation; using dimmer lights is an easy option for larval rearing.
With respect to the research questions in this paper, this study found that cabbage whites were relatively more sensitive to copper than to nickel or zinc. Copper had significant negative impacts on development time at concentrations as low as 50 ppm (Figure 3 and Table 3) and on survival at 500 ppm (Figure 4, Table 4). In contrast, there were no negative effects of nickel on survival (up to 500 ppm; Figure 3) or negative effects on development time at 100 ppm (Figure 4). Cabbage whites were fairly tolerant of zinc, with survival effects seen only at 1,000 ppm (Figure 3) and negative effects on development time starting at 100 ppm (Figure 4). Based on the relatively greater concentrations of zinc in butterfly tissue and mustards (their host plant; Figure 1), it was expected that a relatively greater tolerance to zinc would be seen. However, the sensitivity to copper and the tolerance of nickel were somewhat unexpected given the very low levels of nickel in butterfly tissue (Figure 1) and the necessity of copper as a micronutrient. These unexpected findings are discussed below after considering the tolerance of these metals in other butterflies and moths.
To compare the present data with metal sensitivity measured in other Lepidoptera, data from existing studies were compiled on the minimum concentration, where heavy metals negatively impacted survival49,50,51,56,63,64,65,66,67,68; these studies focused on moths, especially pest species (Galleria mellonella, Lymantria dispar, Plutella xylostella, Spodoptera sp.). All of the measured sensitivity values in this study fall close to the range measured for these other species (Figure 5). However, the measure of nickel tolerance in this study does seem to be higher than expected-while there was not a significant effect of survival at 500 ppm, the previous study on Pieris rapae also found a very high tolerance for nickel (significant effects starting at 1,000 ppm56), despite low levels in their tissue naturally (Figure 1). The measure of copper sensitivity in this study also seems to be at the low end for studies of Lepidoptera. While the use of an artificial diet allows a convenient and controlled comparison of relative metal sensitivity, it is important to note that components of the diet could alter the measurement of absolute metal sensitivity. For instance, vitamin C in the diet could offset metal-induced oxidative stress74, or antibiotics in the diet could alter any effects of microbes on the processing of metals75. An interesting line of future research would be to systematically manipulate such diet components to test effects on metal toxicity, especially given questions about the functional role of lepidopteran gut microbes76,77 and nectar components that may have antioxidant properties78. In addition, variation in dietary requirements across species can make interspecific comparisons challenging, and artificial diet-based methods should be complemented with manipulations of host plants.
These butterflies are particularly tolerant of nickel and sensitive to copper. Previous research has noted that many plants in the mustard family, which includes plants favored by Pieridae, hyper-accumulate nickel as a defensive mechanism against herbivores55,56,63,79,80,81. This hyper-accumulation is over 1,000 ppm in plant tissue, which is orders of magnitude greater than what is seen in most plants (Figure 1). It is possible that Pieris have a particularly high tolerance for nickel due to past selection by such nickel accumulators, as previously speculated26. While copper has been less frequently studied as a micronutrient in insect diets, there is some evidence that it plays a small role in reproduction and immunity, although primarily in blood-feeding insects (e.g.,82,83). It is possible that copper plays a less important physiological role in butterflies than in other animals84,85,86, consistent with recent work highlighting how copper may be as concerning of a pollutant for insects as lead, cadmium, and mercury (e.g.,87,88,89). While Pieris have been shown to avoid copper contamination at low levels90, the mobility of copper in plants (e.g., moving into leaves and flowers) has also flagged it as a metal contaminant of concern91.
While these results provide interesting data on the relative toxicity of these metals to cabbage white butterflies, this paper also aims to be of general use as a detailed visual illustration of methods for rearing this powerful system. Cabbage whites are easy to rear and manipulate in controlled lab experiments4,5 facilitating studies of host searching6,7,8, foraging9,10,11, and sexual selection12,13,14. The ability to rear these butterflies on an artificial diet is key in creating common garden conditions for comparisons and to manipulate nutrients, toxins, and even novel host plants. However, it is important to note that this artificial diet is not necessarily the optimal artificial diet for this species, and could likely be improved with future manipulations. For instance, the salt mix in this diet (and other lepidopteran diets) was originally developed for vertebrates and has higher calcium levels than what most insects need92,93. Thus, some of our rearing efforts have made custom salt mixes with lower calcium levels (e.g.,62), and others make use of "Beck's salt mix", which may be more appropriate for many insect species94. In our own manipulations, we also found that butterflies performed better with relatively less wheat germ and relatively more cellulose compared to original concentrations4. One area in need of further attention is the lipid source and concentration in the diet. For instance, past work has shown that shifting from linseed oil (used in this study) to phospholipids increased the mating rates and growth rates of Pieris on artificial diets95. Supplementation of specific fatty acids in artificial diets may have additional positive effects96,97. Optimizing the artificial diet of Pieris98,99 creates opportunities for addressing interesting questions about nutritional ecology100,101,102, evolutionary ecology, and ecotoxicology. These artificial diet approaches allow researchers to address questions about the role of specific lipids in cognitive evolution103, pre-adaptation to toxins28, dietary components that reduce the toxicity of pollutants104, or stoichiometric interactions between nutrients105.