Zac Hanscom

The watercress glucosinolate-myrosinase system: a feeding deterrent to caddisflies, snails and amphipods

SummaryWatercress (Nasturtium officinale) possesses the glucosinolate-myrosinase system. This system is regarded as a classic example of chemical defense for terrestrial crucifers. Damage of watercress initiates myrosinase-mediated hydrolysis of phenylethyl glucosinolate to a toxic endproduct, phenylethyl isothiocyanate. In multiple choice tests, the amphipod Gammarus pseudolimnaeus, the limnephilid caddisflies Hesperophylax designatus and Limnephilus sp., and the physid snail Physella sp. all strongly preferred (10X) yellowed senescent watercress (FY) over fresh green watercress (FG), despite the 2X higher nitrogen content of green watercress (6.9% for FG vs 3.8% for FY). Green watercress contained 10–40 X more glucosinolate than FY watercress (6.4–8.5 mg/g wet for FG vs 0.2–0.7 mg/g wet for FY). However, when the watercress was heated (ca 70°C), to deactivate the myrosinase enzyme, multiple choice tests showed that these species shift their preferences to heated green watercress (HG). Heating deactivated the deterrent effect and overall preference (consumption) was HG ≥ HY > FY ≫ FG for Gammarus. HG > HY ≥ FY ≫ FG for Hesperophylax, HG > FY ≥HY ≥ FG for Limnephilus, and HG ≥ FY > HY ≥ FG for Physella. Thus heating resulted in a shift in preference from the low glucosinolate, but low nitrogen, unheated yellowed tissue to the high nitrogen green tissue. These results suggest that deactivation of the myrosinase enzyme, and hence isothiocyanate production, results in a shift in preference. Preliminary results with Hesperophylax indicate that addition of myrosinase to the test water, which resulted in the formation of isothiocyanate, results in a significant decrease in HG consumption from control levels (p < 0.001) and no change in preference for HY watercress. With Gammarus, myrosinase resulted in reduced consumption of both green and yellowed watercress, but no significant differential effect. These results provide evidence that the glucosinolate-myrosinase system, recognized as the principle deterrent system of terrestrial crucifers, is the feeding deterrent in watercress and also suggest that in the absence of a functioning deterrent system, nitrogen content may influence consumption.

WATERCRESS AND AMPHIPODS Potential Chemical Defense in a Spring Stream Macrophyte

We investigated the potential role of defensive chemicals in the avoidance of watercress (Nasturtium officinale) by the cooccurring amphipod,Gammarus pseudolimnaeus at two spring brooks: Carp Creek, Michigan and Squabble Brook, Connecticut. We conducted observations and laboratory experiments on the consumption of watercress, the toxicity of damaged (frozen) watercress, and the toxicity of damage-released secondary chemicals. Field-collected yellowed watercress typically lacked the bite and odor characteristic of green watercress and was consumed byG. pseudolimnaeus. G. pseudolimnaeus strongly preferred yellowed watercress to green watercress despite the higher nitrogen content of the latter (2.7 vs 5.4%), and usually consumed five times more yellowed watercress (>50% of yellowed leaf area vs. <8% of green leaf area presented). Fresh green watercress contained seven times more phenylethyl glucosinolate than yellowed watercress (8.9 mg/g wet vs. 1.2 mg/g). Cell-damaged (frozen) watercress was toxic toG. pseudolimnaeus (48-hr LC50s: ca. 1 g wet/liter), and the primary volatile secondary chemicals released by damage were highly toxic. The predominant glucosinolate hydrolysis product, 2-phenylethyl isothiocyanate had 48-hr LC50s between 0.96 and 3.62 mg/liter, whereas 3-phenylpropionitrile was less toxic, with 48-hr LC50s between 130 and 211 mg/liter. These results suggest that live watercress is chemically defended against consumption. The glucosinolate-myrosinase system, recognized as the principle deterrent system of terrestrial crucifers, is also possessed byN. officinale and may contribute to defense from herbivory by aquatic crustaceans. This system may be just one of many examples of the use of defensive chemicals by stream and lake macrophytes.

Isolation and characterization of glucocapparin in Isomeris arborea Nutt

Isomeris arborea (Capparaceae), is the only woody caper endemic to southern California and northern Baja. Methylglucosinolate, also known as glucocapparin, was the only glucosinolate found inI. arborea organs by paper chromatography of the thiourea derivatives and was quantitatively determined by gas chromatography by hydrolytic products. The concentration of glucocapparin ranged from an average of 4.6 mg/g wet weight in mature leaves to 5.2 mg/g wet weight in immature leaves. Buds averaged 6.2 mg/g wet weight and capsule walls 1.8 mg/g wet weight. Seeds contained an average of 14.3 mg/g wet weight of glucocapparin. Glucocapparin concentration was found to vary significantly among the mature leaves of individuals within a single population. This compound is known to be deleterious to nonadapted herbivores and may be implicated in the chemical defense mechanism ofI. arborea.

Glucocapparin variability among four populations ofIsomeris arborea Nutt

Glucocapparin (methylglucosinolate), a putative defense compound, was found to vary between desert and nondesert populations ofIsomeris arborea (Capparaceae): Plants from desert populations contained greater concentrations than nondesert plants in four of the five organs analyzed. Immature leaves at desert sites had average glucocapparin concentrations of 9.2 mg/g and 8.4 mg/g, while nondesert sites averaged 6.0 mg/g and 4.6 mg/g. Mature leaves from desert sites had average concentrations of 12.8 mg/g and 7.9 mg/g; leaves from plants at nondesert sites contained approximately one third to one half of those concentrations. A similar pattern was observed in capsule walls and seeds but not in flower buds; for these, non-desert plants contained a slightly higher concentration of glucocapparin. Our studies show that nitrogen and glucocapparin concentrations fluctuate throughout the year and contribute to the observed variability among populations during any particular season. Glucocapparin may fluctuate seasonally as much as 37% in immature leaves and 78% in mature leaves. In a controlled experiment, glucocapparin concentration varied inversely with nitrogen fertilizer treatment. The plants treated with fertilizer lacking nitrogen ranged from 10.1 mg/g to 10.9 mg/g glucapparin, which was roughly twice the concentration of those supplied with 20 mM nitrogen in the fertilizer.