Remarks on the chemoecology and phylogeny
of Danainae *
Milkweed butterflies are usually considered to belong to the best investigated lepidopterans. For some 20 years now, a 'comprehensive' book ("Milkweed butterflies" by Phil Ackery & Dick Vane-Wright) has been available.
It is difficult to find an entomological text book which does not mention danaines.
Danaines are flagships in the discussion of mimicry.
And, indeed, we know quite a bit on the morphology and biology of this taxon, most of which is related to chemoecology.
If one does not concern oneself with all details, a really neat story can be constructed:
- Danainae feed on milkweeds (Asclepiadaceae).
- Danainae are protected by cardenolides obtained from larval hostplants and stored to the adult stage.
- As a result of cardenolide sequestration, Danainae became Batesian mimicry models as well as Müllerian co-mimics.
- Sequestration of cardenolides is not a trait of species but one of an individual because a given species can exploit hostplants with varying amounts of cardenolides. Thus, there is automimicry, too.
- As a result of mimicry (i.e. visual similarity), elaborate chemcial courtship communication appears necessary. It involves dual androconial organs (alar organs and abdominal hairpencils) between which mechanical contacts are established, also the use of pheromone-transfer-particles and pheromone bouquets of high complexity having as many as 80 components.
- Male danaines exhibit pharmacophagy with respect to pyrrolizidine alkaloids, i.e. independent of nutritional requirements, they search for dry parts of PA-containing plants, apply a solvent via their proboscides and reimbibe it with dissolved PAs.
These chemicals are sequestered and provide protection against antagonists (invertebrates as well as vertrebrates), they are used as precursors for male pheromone components, and they are transferred to the female with the spermatophore as a nuptial gift – to provide protection to the female and her eggs.
Now, what can be said about the phylogeny of the high functional diversity of Danainae? Unfortunately: not all that much yet.
My introductory 'synthesis' is artificial and by no means valid for all danaines, not even for a representative set of species. On some species we have very detailed knowledge on various aspects (I think of excellent contributions by Lincoln Brower, Meron Zalucki, Steve Malcolm, and Keith Brown – to name only but a few colleagues).
Strikingly, the best investigated species, Danaus plexippus, is the most outstanding danaine. Moreover, there is no typical species within the subfamily, not even can we name a genus showing all the interesting features in combination.
On the majority of species we have hardly any biological knowledge and we see the challenges of functional diversity in danaines but detailed facts remain a mystery. Necessarily, generalisations made at this point cannot be correct.
Let me briefly discuss only some gaps in knowlege and facts which spoil the seemingly clear picture:
- Many Danainae feed on dogbanes and figs rather than on milkweeds.
- Not all danaines feeding on cardenolide-containing plants do sequester these chemicals. Many never use cardenolide-containing hostplants. Mind you, storage has only be proven for three species – most have not been checked yet.
- Discussing the primary hostplants with respect to presence or absence of cardenolides cannot be correct. Not only are there significant differences within the chemicals that we call cardenolides, there are many other secondary chemicals in milkweeds, dogbanes and figs. They have hardly been considered but may be they play a bigger role in hostplant associations than cardenolids.
Some evidence: Kenji Honda has demonstrated for Ideopsis similis feeding on Tylophora tanatekae that phenanthroindolizidines stimulate egg laying.
Haribal & Renwick have shown that flavonol glycosides from Asclepias act as oviposition stimuli for the monarch.
What about other danaines utilizing Tylophora or Asclepias, respectively? Comparative approaches are required and likely will to add a new dimension to the story. And there are more candidate secondary chemicals in the huge group of Asclepiadoidea.
- Sequestration of PAs as protective chemicals is also an individual trait. In contrast to cardenolides which are kind of fixed when an adult butterfly hatches, there is much more variation and dynamics. There is not only individual variation, but intra-individual temporal variation.
Imagine a danaine population. There are individuals with cardenolides in varying quantities, others lack them completely. The very same individuals can be divided into those lacking PAs and others with PAs in varying quantities. Within the latter the PAs also vary over time according to availability of PAs and mating status.
- Storage of PAs has been shown for numerous species in the course of 'feeding' experiments, but rarely in an ecological context. The quantitative dynamics in relation to environmental conditions is almost unknown. Batesian, Müllerian and auto mimicry gets quite bizar and calls for detailed studies.
- Gathering PAs appears unspecific; from our experimental studies it seems that any 1,2-dehydropyrrolizidine is being detected and sequestered. But which are the natural sources? Which are the natural qualities and quantities of sequestered PAs???
Most information on PAs come from experiments and baiting in the field and extremely few on observations under natural conditions.
- The dogma that only males gather PAs is correct for some species only. Of others also females visit PA sources. However, on most species data on PA visitation have not been differenciated according to sexes attracted.
- Do all danaines really use dihydropyrrolizines? Only a few have been checked – and these show differences with respect to the molecules synthesized from PAs, and different biochemical pathways are involved. Recently, we have reported on new molecules derived from PAs in ithomiines – perhaps danaines also exhibit a greater biosynthetic capacity.
- Much more gaps and questions could be mentioned, including those on mechanical contacts between alar and abdominal organs and on their different roles, on general differences in courtship strategies, on pheromone components and on the production and function of pheromone-transfer-particles etc. etc.
In summary: there more questions than answers. But the answers we already have, make the questions extremely exciting.
Obiviously danaine mimicry cannot be explained by one plant association only – but by two, the primary (nutrional) hostplants and the supplementary (PA) hostplants. But these two types of chemicals are not only obtained by the insects in quite different ways – they also have very different modes of action. Are they thus supplementing against a given antagonist or acting against different types of antagonists? In this context, we should not just consider predators but also parasitoids!
Also: there is a direct correlation between intraspecific and interspecific chemical communication. But questions such as "was protection via larvae or via adults first"? "were cardenolides utilized first or rather pyrrolizidines?" in the current knowledge find no basis to be answered, and we also know too little to understand the phylogeny of the mimicry associations, of the pheromone systems, and also of the relation between chemical defense and intraspecific chemical communication.
Do we, at least, see a phylogeny of relationships to primary hostplants? There are hundreds of records on Danainae and hostplants. What can they tell us?
The systematics of the Apocynaceae – Asclepiadoideae has greatly improved recently due to many detailed molecular studies. We collaborate with Sigried Liede, a recognized asclepiad systematist, and tried to relate danaines' hostplant records with the new system of the Apocynaceae. There is no time to discuss the cladogram, but I can summarize that we are faced with a picture which is good for everything and nothing at the same time. If you choose the right host references, you can 'prove' almost any hypothesis.
This is apparently because:
- many records are solitary,
- host records published are full of misidentifications,
- a large amount of records is just copied from unspecified sources,
- a record from a given area cannot necessarily be generalized to the species level, etc.
Also: many names have recently been changed; e.g., 6 of 7 species of Cynanchum recorded as hostplants for danaines are now placed in the genera Vincetoxicum and Tylophora, respectively. Consequently, many records giving genus names only are of no use at all. But interestingly, in the pattern we see that Cynanchum species are utilized by Amauris only and not also by Danaus!
If we 'correct' the all data according to the above comments, not many are left – too few for conclusive statements.
What remains without any doubt is:
- Amauris, the African genus which apparently has the most complex mimicry associations feeds exclusively on hostplants not containing cardenolides.
- Larvae of the SE-Asian genus Idea exclusively feed on Parsonsia, an apocynad rich in PAs. The adults do not show pharmacophagy.
But these perceptions do not provide answers but rather more puzzles.
What are the conclusions?
We apparently have to wait another while to gain proper insight into the phylogeny of Danainae.
1. Definitely, comprehensive molecular studies are urgently needed – only they can resolve the many uncertainties in the cladograms of Ackery & Vane-Wright of 1984 which are solely based on morphological data.
2. We need experimental approaches, biological and chemical ones. They are nowadays difficult to be funded but will be done eventually by someone.
But in particular, we
thirdly are in need for many more basic (= old-fashioned) ecological (field) data, because molecular information alone does not permit to understand the evolution of the various life-histories and life-styles, and experimental approaches are necessarily restricted to selected species.
The data required cannot be gathered straightforwardly by a PhD student, not even by ten or a hundred. One highly depends on opportunities such as being at the right time at the right spot, and one cannot do this job systematically.
I am sure that many observations are in the minds of people already but are not published for a variety of reasons. But accidental observations are of equal value to systematic searching, and a single record can be of high valuable in the large context. Since we are not talking about missing data from the US or Europe or Australia but of Africa, SE Asia and South America, the knowledge and abilities of amateurs (and bachelor and master students) living in these regions need to be involved.
Perhaps, we need an interactive platform for specialists and non-specialists, for amateurs and professionals. And with the world wide web several options are available.
Apart from hostplant records, we need field records on parasitoids, and the very same arguments apply. Understanding of the phylogeny of danaines cannot be obtained without consideration of parasitoids, and – in particular – on parasitoids in relation to hostplant chemistry.
Published records deal only with 30 species of parasitoids on 15 taxa of danaines. Thus for most species we do not know parasitoids at all. And for the ones we know some parasitoids, they cannot be evaluated in relation to the hostplant chemistry
Phil Ackery, Dick Vane-Wright and myself have almost finished to put the available biological information on Danainae into a database – those from their book but including information which came up since the book was published.
We are using CoenoSys®, a database system which has been developed at my Institute as a tool for databasing and evaluating biotic relationships and which is used for other purposes, too. I cannot go into details now, but CoenoSys® carries the information in a much higher resultion than the book could do, e.g. we add geodata and include plants' and insects' chemistry on a detailed and interlinked level.
The system needs to be made fit for the internet that everybody will eventually have access to it. In addition, a platform should be provided for students of danaidology worldwide which offers assistance for ID of insects and plants to ensure accuracy of records from those field-workers who might not have the necessary background. Again: amateurs are likely to contribute more on that kind of data than any so-called professional could possibly do.
I have highlighted facts and fiction in research on danaines and by doing that I have 'put some water into the nice wine' danaines donate us. Please, accept my apologies! But I think all the details make danaines even more fascinating and instead of reducing interest for these great creatures I hope to have increased interest!
Finally, let me state that by looking at arctiids and ithomiines, we see most intriguing parallels – and differences ! – to the discussed situation in the danaines – and my plea for old-fashioned field-work also applies.
* This text – a short summary of facts and questions – is, in part, based on a paper by Michael Boppré presented at the XXII International Congress of Entomology, Brisbane, 2004