Structural and functional features of the isolated cardenolides used in this study, comprising the dominant compounds in Asclepias curassavica (70% of the total leaf cardenolides).

The structural complexity values are based on dox-g (Rubiano-Buitrago et al. 2025). We also provide a non-chromatography dependent metric of polarity, WLOGP (Daina et al. 2017). IC-50 represents μM needed to inhibit the monarch Na/K-ATPase by 50% in vitro, data from Agrawal et al. (2021) except for 15-Hydroxy calotropin which was generated for this study (Fig. S4, Table S1). The proportion of total cardenolides in A. curassavica leaves is based on past work (see methods). Proportion in “real mix” was scaled up from the proportion found in leaves to sum to 1.

A proposed biosynthesis pathway for cardenolides of A. curassavica built on coroglaucigenin (Rubiano-Buitrago et al. 2025).

Modifications are indicated by purple highlighting and each arrow indicates a hypothesized step; multiple arrows indicate multiple concerted reactions without displaying all the intermediates. All compounds, except the genin, are known to occur in A. curassavica, however gofruside is rarely found in high quantities in the foliage (Roy et al. 2005, Rubiano-Buitrago et al. 2022).

Monarch caterpillars are impacted by feeding on A. incarnata leaves painted with isolated cardenolides from Asclepias curassavica.

A) growth after nine days (most caterpillars in the 3rd instar), B) efficiency of conversion of digested matter, and C) total cardenolides sequestered. Shown are means ± SEs and different letters indicate a significant difference (p<0.05, Fisher’s LSD). The green symbol has no cardenolides added (A. incarnata) and the two orange symbols represent cardenolides with N,S-ring moiety.

Monarch caterpillars differentially sequester and excrete cardenolides when eating leaves painted with isolated compounds from Asclepias curassavica.

Shown are means ± SEs of the proportion of total cardenolides ingested that are sequestered or excreted (bars, left axis); shown also is the amount of cardenolide ingested (green dots, right axis). 15-Hydroxy calotropin, frugoside, and calactin are sequestered intact, while uscharin and voruscharin are stored after conversion to calotropin and calactin. Orange shading indicates cardenolides with N,S-ring moiety. Data on cardenolide concentrations sequestered and excreted on a mass basis are given in Fig. S2.

Monarch caterpillars differentially grow and sequester cardenolides when feeding on single isolated compounds from Asclepias curassavica compared to mixtures.

Shown are means ± SEs for several response variables with the units shown below the X axis. Sequestration efficiency was calculated by the mg of cardenolide ingested / mg cardenolide sequestered. Single effects are the average of the five compounds administered individually, whereas the mixture is the average of an equal mixture treatment and realistic mixture treatment. In all cases, total cardenolide concentrations were presented to caterpillars on an equimolar basis. Significance (p<0.01) is shown by *, while † indicates p=0.07. All treatment means are provided in Fig. S3).

Monarch caterpillars are impacted by feeding on leaf discs painted with isolated cardenolides from Asclepias curassavica.

A) growth is dependent on leaf mass consumed, with the same slope for all treatments, and B) excretion is dependent on leaf mass consumed, but with different slopes depending on treatments.

Monarch caterpillars differentially sequester (F6,102=102.22, p<0.001) and excrete (F6,100=37.12, p<0.001) cardenolides when feeding on leaf discs painted with isolated compounds from Asclepias curassavica.

Shown are means +/-SEs of cardenolide concentration (on a per body mass basis) and different letters indicate a significant difference (p<0.05, Fisher’s LSD, letters are only comparable within a tissue type). Orange shading indicates nitrogen-containing cardenolides. Data on absolute cardenolides sequestered are given in the main text. The negative control is A. incarnata, the very low cardenolide plant species that was used to paint on individual cardenolides. The positive control was A. curassavica leaves, the species from which cardenolides were isolated what we were simulating in our experimental treatments.

Mean ± SE of nine treatments growing monarch caterpillars on two controls (green: positive on A. curassavica and negative on A. incarnata), three non-N,S-cardenolides (gray), two N,S-cardenolides (orange), and 2 types of mixtures (striped).

Dashed line indicates the response of the realistic mixture, that which approximates the composition of cardenolides in A. curassavica. For all non-control treatments the total cardenolides applied to leaf discs was of equal concentration.

Inhibition curves estimated from six concentrations of 15-hydroxy calotropin on a sensitive (porcine) and adapted (monarch) sodium-potassium ATPase in vitro.

Shown are Means ± SEs.

1D NMR assignment of key positions for the identification of 15β-hydroxy-calotropin.

Chemical shifts with an asterisk correspond to values in CDCl3, whereas non-designated ones correspond to values in CD3OD. Data was compared with published records (El-Askary et al. 1993, Rubiano-Buitrago et al. 2022).