Description
Plant secondary metabolites are a key defence against herbivores, and their evolutionary origin is likely from primary metabolites. Yet for this to occur, an intermediate step of overexpression of primary metabolites would need to confer some advantage to the plant. Here, we examine the evolution of overexpression of the essential amino acid, L‐tyrosine and its role as a defence against herbivores. We examined overexpression of tyrosine in 97 species of Inga (Fabaceae), a genus of tropical trees, at five sites throughout the Neotropics. We predicted that tyrosine could act as an anti‐herbivore defence because concentrations of 4% tyrosine in artificial diets halved larval growth rates. We also collected insect herbivores to determine if tyrosine and its derivatives influenced host associations. Overexpression of tyrosine was only present in a single lineage comprising 21 species, with concentrations ranging from 5% to 20% of the leaf dry weight. Overexpression was pronounced in expanding but not in mature leaves. Despite laboratory studies showing toxicity of L‐tyrosine, Inga species with tyrosine suffered higher levels of herbivory. We therefore hypothesize that overexpression is only favoured in species with less effective secondary metabolites. Some tyrosine‐producing species also contained secondary metabolites that are derived from tyrosine: tyrosine‐gallates, tyramine‐gallates and DOPA‐gallates. Elevated levels of transcripts of prephenate dehydrogenase, an enzyme in the tyrosine biosynthetic pathway that is insensitive to negative feedback from tyrosine, were found only in species that overexpress tyrosine or related gallates. Different lineages of herbivores showed contrasting responses to the overexpression of tyrosine and its derived secondary metabolites in their host plants. Synthesis. We propose that overexpression of some primary metabolites can serve as a chemical defence against herbivores, and are most likely to be selected for in species suffering high herbivory due to less effective secondary metabolites. Overexpression may be the first evolutionary step in the transition to the production of more derived secondary metabolites. Presumably, derived compounds would be more effective and less costly than free tyrosine as anti‐herbivore defences.
Retention time and abundance of tyrosine and derivatives in Inga
Tyrosine content and the retention times and abundances of tyrosine, tyramine and DOPA gallate ions for each species and site. The ‘Species code’ is our identifier for both species and site. Codes beginning in 'Ing' are from Barro Colorado Island, Panama; 'LA' are from Los Amigos, Peru, 'M' are from Manaus (Km41), Brazil; 'N' are from Nouragues, French Guiana; and 'T' are from Tiputini, Ecuador. Exp and M are expanding and mature leaves, respectively, with ‘n’ indicating sample size. RT is retention time on the UPLC column. TIC is total ion count or ion abundance. Most species/site combinations had five independent UPLC-MS analyses; the exceptions are noted. All UPLC-MS analyses were as described in ‘Analysis of gallic acid depsides of tyrosine, tyramine and DOPA’. For this table, one sample with the highest signal was analyzed per species/site combination by searching for the m/z values given in Table S2. Since the ion abundances of these amines were consistently higher by factors of 6x to 60x for positive mode relative to negative mode, this analysis exclusively used positive mode data in order to maximize sensitivity.
Table-S4.xls
Transcriptomic data for Inga species
Quality metrics for transcriptomes. Quantity and quality of RNA, library size, Trinity assembly metrics and the percentage of contigs annotated through the Trinnotate pipeline.
Table-S5.csv
Recorded damage to Inga by insect herbivores separated by site and tyrosine expression
Herbivore damage to expanding leaves for species with and without tyrosine overexpression. Species that contained tyrosine-derivatives in addition to free tyrosine were included in the tyrosine-plus category. Data are from French Guiana (FG), Brazil (B), Peru (Pe) and Ecuador (E). Panama is excluded from the analysis as only one species contained tyrosine. An ANOVA exploring the effect of tyrosine status (T+ or T-) on herbivory showed lower damage for species without tyrosine (p < 0.001).
Figure-3.pdf
Assembled transcriptomes
Trinity assemblies of Inga transcriptomes
Inga_Trinity_assembelies.tar.gz
Salmon counts with I_sapindoides as reference
Salmon counts for each transcriptome. Generated using I. sapindoides BCI-97 as the reference transcriptome
To_sap_97_Salmon.tar.gz
Annotation of Inga transcriptomes
Trinnotate annotation for transcriptomes assembled from RNASeq of wild collected expanding leaves
trinnnotates.tar.gz
Retention time and abundance of tyrosine and derivatives in Inga
Tyrosine content and the retention times and abundances of tyrosine, tyramine and DOPA gallate ions for each species and site. The ‘Species code’ is our identifier for both species and site. Codes beginning in 'Ing' are from Barro Colorado Island, Panama; 'LA' are from Los Amigos, Peru, 'M' are from Manaus (Km41), Brazil; 'N' are from Nouragues, French Guiana; and 'T' are from Tiputini, Ecuador. Exp and M are expanding and mature leaves, respectively, with ‘n’ indicating sample size. RT is retention time on the UPLC column. TIC is total ion count or ion abundance. Most species/site combinations had five independent UPLC-MS analyses; the exceptions are noted. All UPLC-MS analyses were as described in ‘Analysis of gallic acid depsides of tyrosine, tyramine and DOPA’. For this table, one sample with the highest signal was analyzed per species/site combination by searching for the m/z values given in Table S2. Since the ion abundances of these amines were consistently higher by factors of 6x to 60x for positive mode relative to negative mode, this analysis exclusively used positive mode data in order to maximize sensitivity.
Table-S4.xls
Transcriptomic data for Inga species
Quality metrics for transcriptomes. Quantity and quality of RNA, library size, Trinity assembly metrics and the percentage of contigs annotated through the Trinnotate pipeline.
Table-S5.csv
Recorded damage to Inga by insect herbivores separated by site and tyrosine expression
Herbivore damage to expanding leaves for species with and without tyrosine overexpression. Species that contained tyrosine-derivatives in addition to free tyrosine were included in the tyrosine-plus category. Data are from French Guiana (FG), Brazil (B), Peru (Pe) and Ecuador (E). Panama is excluded from the analysis as only one species contained tyrosine. An ANOVA exploring the effect of tyrosine status (T+ or T-) on herbivory showed lower damage for species without tyrosine (p < 0.001).
Figure-3.pdf
Assembled transcriptomes
Trinity assemblies of Inga transcriptomes
Inga_Trinity_assembelies.tar.gz
Salmon counts with I_sapindoides as reference
Salmon counts for each transcriptome. Generated using I. sapindoides BCI-97 as the reference transcriptome
To_sap_97_Salmon.tar.gz
Annotation of Inga transcriptomes
Trinnotate annotation for transcriptomes assembled from RNASeq of wild collected expanding leaves
trinnnotates.tar.gz
Data Citation
Coley, Phyllis D.; Endara, María‐José; Ghabash, Gabrielle et al. (2019). Data from: Macroevolutionary patterns in overexpression of tyrosine: an anti-herbivore defense in a speciose tropical tree genus, Inga (Fabaceae) [Dataset]. Dryad. https://doi.org/10.5061/dryad.687m14t
| Date made available | 6 Jun 2019 |
|---|---|
| Publisher | Dryad |
| Geographical coverage | South America |
Research output
- 1 Article
-
Macroevolutionary patterns in overexpression of tyrosine: An anti‐herbivore defence in a speciose tropical tree genus, Inga (Fabaceae)
Coley, P. D., Endara, M., Ghabash, G., Kidner, C. A., Nicholls, J. A., Pennington, R. T., Mills, A. G., Soule, A. J., Lemes, M. R., Stone, G. N., Kursar, T. A. & Shefferson, R. (Editor), 13 May 2019, (E-pub ahead of print) In: Journal of Ecology. 107, 4, p. 1620-1632Research output: Contribution to journal › Article › peer-review
Open AccessFile
Cite this
- DataSetCite