The evolutionary origins of the cat attractant nepetalactone in catnip

Abstract

Catnip or catmint (Nepeta spp.) is a flowering plant in the mint family (Lamiaceae) famed for its ability to attract cats. This phenomenon is caused by the compound nepetalactone, a volatile iridoid that also repels insects. Iridoids are present in many Lamiaceae species but were lost in the ancestor of the Nepetoideae, the subfamily containing Nepeta. Using comparative genomics, ancestral sequence reconstructions, and phylogenetic analyses, we probed the re-emergence of iridoid biosynthesis in Nepeta. The results of these investigations revealed mechanisms for the loss and subsequent re-evolution of iridoid biosynthesis in the Nepeta lineage. We present evidence for a chronology of events that led to the formation of nepetalactone biosynthesis and its metabolic gene cluster. This study provides insights into the interplay between enzyme and genome evolution in the origins, loss, and re-emergence of plant chemical diversity.

Fig. 1. Iridoid and nepetalactone biosynthesis.

(A) Volatile nepetalactone stereoisomers found in Nepeta. Harpagide is a typical nonvolatile iridoid glucoside found in other iridoid-producing Lamiaceae. (B) Early iridoid biosynthetic pathway in plants: geraniol synthase (GES), geraniol 8-hydroxylase (G8H), 8-hydroxygeraniol oxidoreductase (HGO), and iridoid synthase (ISY). (C) Knowledge of nepetalactone biosynthesis in N. mussinii as reported in Lichman et al. (9). The crossed double bond depiction in the ISY product refers to unknown or undefined stereochemistry.

Fig. 2. Lamiaceae and ISY.

(A) Phylogenetic tree of Lamiaceae species as reported in (10); black circles, bootstrap support ≥99%. (B) Cladogram of the PRISE gene family. See fig. S15 for annotated phylogram.

Fig. 3. Metabolic gene clusters in Nepeta.

(A) Genomic organization and syntenic relationships of nepetalactone biosynthesis gene cluster in Nepeta. Gray polygons show syntenic relationships. (B) Biosynthetic pathway of nepetalactone stereoisomer formation based on combination of in vitro assay data and in planta gene expression. The crossed double bond depiction in the ISY product refers to unknown or undefined stereochemistry. (C) In vitro multienzyme cascade assays demonstrating selective formation of nepetalactone stereoisomers (see also figs. S16 to S20). (D) Differential expression of nepetalactone biosynthetic genes in N. mussinii individuals with distinct nepetalactone stereo-chemotypes. Significant differences determined by analysis of variation with Tukey’s post hoc test (n = 4 to 10; ****P < 0.0001, ***P < 0.001, **P < 0.01, *P < 0.05) (see also fig. S21).

Fig. 4. Evolution of ISY in Nepeta.

(A) Genomic location of three PRISE homologs in N. mussinii. Similar genome structures are present in N. cataria (table S7 and Fig. 3A). (B) Phylogram of PRISEs from Nepetinae (subtribe containing Nepeta). The highly diverged pseudogenes NcSISYA and NcSISYB were excluded from the phylogenetic inference to improve model support. Pie charts show in vitro enzyme activities of extant PRISEs from hyssop and Nepeta and reconstructed ancestral PRISEs. For enzyme activities, the area of each chart sector is proportional to relative activity; total chart area is proportional to the sum of relative activities. HoP5βRA failed to express in sufficient quantity to assay. All branches have >99% support unless noted, see fig. S22 for alignment of sequences and fig. S23 for annotated phylogram. Ho, H. officinalis; Nm, N. mussinii; Nc, N. cataria; AGFO, Agastache foeniculum; GLHE, G. hederacea.

Fig. 5. Enzyme and genome evolution in Nepeta iridoid biosynthesis.

(A) Selected Nepeta clades from chronograms of NEPS and PRISE genes. Blue bars are 95% confidence intervals. Colored branches represent genomic location. Complete chronograms can be found in figs. S27 and S28. (B) Proposed chronology of events in Nepeta nepetalactone biosynthesis evolution.