Figure 1.
The glucosinolate-myrosinase system and proposed pathways of aromatic nitrile metabolism in P. rapae larvae.
A. Myrosinase-catalyzed hydrolysis of glucosinolates upon plant tissue disruption yields an unstable aglucone which most commonly rearranges to a toxic isothiocyanate. Larvae of P. rapae redirect glucosinolate breakdown to the formation of simple nitriles by the gut nitrile-specifier protein (NSP). R, variable side chain. B. Examples of glucosinolates with aromatic (i.e. benzene ring-containing) side chains. C. Upon ingestion of plant material by P. rapae larvae, 1 and 2 are converted to phenylacetonitrile (3) and 3-phenylpropionitrile (4), respectively. These undergo further metabolism to the glycine conjugates 5–7 which are excreted with the feces. The major metabolite of 1 is hippuric acid (N-benzoylglycine, 5; 23, 24), the major metabolite of 2 is N-(3-phenylpropionyl)glycine (7, this study). N-phenylacetylglycine (6) is formed as a minor metabolite from both glucosinolates. This study establishes the pathways from 3 and 4 to 5–7. While the conversion of 3 to 5 involves a C1-loss through HCN release (route a), the side chain of 4 is maintained throughout its major metabolic pathway (route e). Reactions a, b and c are catalyzed by an NADPH-dependent microsomal enzyme activity. Reaction d, and likely, reaction e involve nitrilase activity from the ingested plant material. Compounds 9, 11b and 12 were detected as intermediates in this study. Bold and thin arrows indicate major and minor metabolic pathways, respectively.
Figure 2.
Differential metabolism of aromatic glucosinolates with different side chain lengths in P. rapae larvae.
Feces were collected from P. rapae larvae that had fed on leaves of A. thaliana Col-0, 35S:CYP79A2, Tropaeolum majus, or Nasturtium officinale. Glycine conjugates 5 (white), 6 (light-gray), and 7 (dark-gray) were quantified in feces extracts by HPLC-MS using 13C-labeled 5, 6, and 7 as standards. Means ± SD are given with N (number of biological replicates). Each replicate represents a pair of larvae.
Table 1.
Plant nitrilase activity contributes only to the minor pathway of benzylglucosinolate metabolism in P. rapae larvae.
Figure 3.
Organic phase metabolites of 2-phenylethylglucosinolate in plant homogenates and P. rapae larvae.
Dichloromethane extracts of N. officinale leaf autolysates (A) and the organic phase of dichloromethane/water extracts of feces from P. rapae larvae that had fed on N. officinale leaves (B) were analyzed by GC-MS. Shown are total ion current traces. IS, internal standard.
Figure 4.
NADPH-dependent hydroxylation of aromatic nitriles by P. rapae gut microsomes.
Larval gut microsomes were incubated with 2.5 mM phenylacetonitrile 3 (A–D) or 2.5 mM 3-phenylpropionitrile 4 (E–H) for 45 min at 31°C in the presence (A, C–E, G, H) or absence (B, F) of NADPH. In C and G, microsomes were flushed with CO prior to addition of NADPH. In D and H, microsomes were heated (95°C, 5 min) prior to the assay. Assays were extracted with dichloromethane, and the organic phases analyzed by GC-MS. Shown are total ion current traces. IS, internal standard.
Figure 5.
Cyanide is released as a consequence of aromatic nitrile hydroxylation by P. rapae gut microsomes.
Larval gut microsomes were incubated with 2.5 mM phenylacetonitrile 3 (A–D) or 2.5 mM 3-phenylpropionitrile 4 (E–H) in the presence (A, C–E, G, H) or absence (B, F) of NADPH. Cyanide was captured by derivatization. Shown are HPLC-MS/MRM traces of the derivatization product (X). In C and G, microsomes were flushed with CO prior to addition of NADPH. In D and H, microsomes were heated (95°C, 5 min) prior to the assay.
Figure 6.
Performance of P. rapae larvae on cyanogenic plants.
P. rapae and S. littoralis larvae were allowed to feed on either A. thaliana Col-0 wildtype (gray) or cyanogenic A. thaliana 3x/dhurrin plants (dark grey). After 10 d, surviving larvae were counted and weighted. Larval mortality is given as means ± SEM of three independent experiments. Larval weights are given as means ± SEM from one out of three independent experiments. Results of all experiments are shown in Table S1. Numbers in the bars indicate N (number of surviving individuals).
Figure 7.
Detoxification of cyanide in P. rapae.
A. Scheme of the reaction catalyzed by rhodanese. B. Scheme of the reaction catalyzed by β-cyanoalanine synthase. C. β-Cyanoalanine content in P. rapae larvae after nine days of feeding on wildtype, benzylglucosinolate-rich (35S:CYP79A2) and cyanogenic (3x/dhurrin) A. thaliana plants. Larvae were extracted with dichloromethane and water. The aqueous phase was analyzed by HPLC-MS. Data are means ± SD. N (number of larvae analyzed) is given in the bars, p values (t-test) for the comparison with Col-0 above the bars. D. Quantitative analysis of β-cyanoalanine and SCN− with M+1 after 24 h [15N]HCN fumigation of the larvae. Each bar represents the mean ± SD of N = 16 individual larvae. P values (t-test) are given above the bars for the comparison of fumigated (dark-grey bars) to non-fumigated larvae (light-grey bars). Data in C and D are each from one out of at least three independent experiments that all showed significant differences (p<0.05).