External Chemical Defenses; a natural history view

After a long hiatus (due to qualifying exams, a dissertation proposal, hundreds of Trichostema in the greenhouse and a couple manuscripts in progress), here is the long-awaited (by likely my parents, grandparents, and maybe 3 other people) blog post that outlines THE TOPIC OF MY DISSERTATION (in a rather digressive way). I took a long and circuitous route to find this topic, but its very exciting to me and when I've presented it publicly (at Ecological Society of America meeting last August and informally to many colleagues), it has been greeted with interest. This post will mostly deal with how I got to this topic and why it is a great thing for insect-plant biology, a later one will detail the implications (a paper on which was recently accepted!).

The stem of Trichostema laxum (Lamiaceae). The little bubbles on the tips of the glandular trichomes are full of an oily, wonderfully vinegary-smelling fluid, of unknown utility.

Coming from the northeast and botanizing and entomologizing in Massachusetts and Rhode Island, I knew my fair share of plants - herbs and trees, shrubs and aquatic plants - and many of their associated insects. Coming to California, I was exposed to more. A whole lot more! I met a great deal of new taxa - from aromatic plains of sagebrush, to strange long-horned moths (Adela) to smooth barked manzanitas and madrones to the most spectacular diversity of wildflowers I've ever seen to albatrosses wheeling offshore. As a natural historian, I had a lot to learn and I continue to learn new plants, insects and even occasionally birds, each time I venture out to a new spot, and often when I revisit old spots.

A Mormon cricket, Anabrus simplex. White Mountain, CA, Sept. 2012. A mind-blowingly large katydid.

In the first year, I bounced from project to project, trying to raise parasitoids from caterpillars, looking at galls and leaf miners for inspiration and tractable systems to look at the effects of plants and their chemistry on predators and parasitoids of the herbivores feeding on those plants. None of which really panned out individually, but I spent a lot of time wandering around with a net, hand lens and notebook. And I learned a lot of natural history of California, the coast, the coast range, the valley, and the west side of the Sierras, in the process.Then I happened upon the chenopod Blitum californicum and ended up spending a summer playing with various chenopod species and getting a paper out of it. This, coupled with work done with Billy Krimmel and Ian Pearse (see their research here), got me thinking about plants really hard. And I realized a major difference between plants back home and here: plants back home are generally rather glabrous (smooth-surfaced) and here they have a myriad of glandular trichomes, general stickiness, oiliness, resins, etc. and they often have strong smells (the latter point is an oft-mentioned feature of plants of Mediterranean climates and why many of our kitchen spices - e.g. sage, rosemary, oregano - hail from these areas).

The glandular trichomes on Aquilegia eximia (serpentine columbine: Ranunculaceae)
are extremely sticky and entrap enormous quantities of small flies and wasps.

So I spent more time seeking out these plants and examining them closely. I also spent time reading the literature on secretory tissues in plants. And I began to think, abstractly, what might the differences between compounds put onto plant surfaces and those inside a plant be? I made lists, I woke up in the middle of the night with ideas and eventually, I distilled these many ideas into five broad differences between the potentially-defensive chemical secretions and those sequested inside plant tissues. These form the meat of the accepted paper, to be detailed later. Instead I'll briefly touch on how I experiment with external chemicals and why this is important and exciting.

The first approach I took to playing with (e.g. experimenting) external chemicals was testing their efficacy at preventing damage to plants. I did, and continue to do, this in two ways (in the chenopod paper as well as many small unpublished tests). The first way is to remove or reduce the defense, gently, using a paintbrush or a sponge, leaving the leaf surface intact. I then run choice or no choice (palatability) assays on these plants, usually using a the wonderfully generalist spotted cucumber beetle, which rarely fails to eat at least a little bit of a plant (but also will eat a LOT of a plant it likes). The second is to take the external chemical (either from the plant or the known chemical) and place it onto another plant, either in natural concentrations, or varying the concentrations, looking for changes in herbivory.

Lab-grown, highly glandular Antirrhinum californicum.

I also do this in the field - something that is really hard to mimic with internal chemicals. Removal of exudates from plants can be difficult (e.g. Yerba Santa, with really tough leaf resins), but can also be really easy, as in the case of Antirrhinum californicum, the California snapdragon. In the field, removing the exudates of this snapdragon caused a really interesting response: insects - mostly a heliothine noctuid, caused more damage to the exudate-removed plants,as expected; but mammals (deer and/or black-tailed jackrabbits) ate preferentially the plants with exudates intact (highly significant result). This suggests that the exudates may be both a defense and a liability in nature, an interesting result that wouldn't have been possible in a lab - and wouldn't have been easy to find with an internal defense, as manipulation would have been more difficult.

Deer or jackrabbit eaten A. californicum from the experiment. 

Why is this exciting? Well, a manipulation like this wouldn't really be possible with internal defenses. There are two general ways to look at the effect of internal chemical(s) on other organisms: (1) find or create lines that differ in concentrations of chemicals, or look comparatively across species that differ, or (2) create, via genetic techniques, knockout lines that lack a compound. The problem with both approaches is that there is pleitropy (one gene doing multiple things). If you have lines that differ in a compound, they also differ in other aspects. Similarly, if you have a gene knockout (or duplication), that will almost certainly have effects on other aspects of the plant. Of course, there are positives to both of those approaches: the comparative approach allows investigations on evolution of a trait and a genetic manipulation allows a level of integration and detail that no field study on a natural population will ever approach. In my systems, it seemed like a good way to get at defense mechanisms. But was I the first to do it? Absolutely not, in the marine world (Mark Hay's lab) have been doing these sorts of exudate removals on seaweeds for decades. Yet terrestrial folks don't cite these papers or think in quite the same way (the marine folks cite terrestrial chemical ecology all the time!). And even they weren't the first! Thomas Hartmann pointed out in a 2007 paper on the history of plant-insect science, Ernst Stahl, in ~1900, removed the acid droplets secreted by evening primroses (Oenothera spp.) and found that the plants became far more palatable to herbivorous snails and slugs. However, despite the ease and history of these experiments, they are rare. And I'm not quite sure why.

The most satisfying part of these investigations so far (more to be detailed soon) has been combining a variety of approaches to think about a problem. I've made lots of natural history observations (examined lots of plants with secretions), thought about problems in creative ways (can I think of fundamental differences between internal and external chemicals, ecologically?), read the literature (from Darwin and earlier, to the present), planned and ran experiments, interpreted data (with such strange results as in the snapdragons) and am working on integrating it all into a dissertation, which I hope - in a few years - will be a cohesive body of work that other people will add to, build upon and apply to new systems and problems.