New Phytologist Supporting Information

Article title:

Pollen sterol diversity is associated with phylogenetics and environment but not pollinators

Authors:

Pengjuan Zu, Hauke Koch, Orlando Schwery, Samuel Pironon, Charlotte Phillips, Ian Ondo, Iain W. Farrell, W. David Nes, Elynor Moore, Geraldine A. Wright, Dudley I. Farman, Philip C. Stevenson

Article acceptance date: TBD

The following Supporting Information is available for this article:

Supplementary figures:

Fig. S1. 2D-NMDS (non-metric multudimensional scaling) plot of pollen sterol profile similarities (Bray-Curtis) between plant species with major plant families highlighted with different symbols (see legend). Distances between points correspond to sterol profile dissimilarities.

Fig. S2. 24-methylenecholesterol content (µg/mg pollen) of plants without pollen as bee reward (no bee pollination/collection of pollen by bees), or with pollen as reward for bees (based on evidence of bee pollination and pollen collection by bees).

Fig. S3. Total sterol content (µg/mg pollen) of plants without pollen as bee reward, or with pollen as reward for bees. Indentations represent 95% confidence intervals.

Fig. S4. GC-MS spectra of the 25 phytosterols identified in our study (after Tri-sil derivatisation, extraction details see Materials and methods section).

Supplementary tables (submitted separately: multiple sheets in Excel):

Table S1. Data table (plant species, scores for different environmental variables/principal components, pollination modes, sterol composition (relative & absolute amounts).

Table S2. Scientific name and family for all sampled species, along with suggested OTL synonyms (which were subsequently used) and taxon IDs; species excluded from the phylogeny are highlighted in grey; reason for exclusion due to issues in the data and/or the OTL taxonomy are indicated.

Table S3. Variable contributions to axes of PCA of 13 environmental variables.

Table S4. Results of linear model tests for phylogenetic independent contrasts (PICs) of total sterol amount/diversity against PICs of environmental variables and niche volume.

Plant family Amaryllidaceae

Scilla madeirensis Asparagaceae Hyacinthoides non-scripta Asphodelaceae 0.150 Asteraceae Bryonia dioica Lachenalia ensifolia Boraginaceae Cactaceae Campanulaceae Symphytum uplandicum Gesneria ventricosa Caprifoliaceae Silene dioica Colchicaceae Nymphoides indica Buddleja davidii Ericaceae Verbascum arcturus Hedera helix 0.075 Symphytum officinale Polemonium caeruleum Euphorbiaceae Taraxacum officinale Fabaceae Iris sibirica Hieracium umbellatum 2

Schlumbergera truncata Campanula isophylla Lamiaceae e t Camellia sinensis Malvaceae a Drosera regia Campanula fragilis

n Paeonia peregrina Parodia schumanniana Cirsium vulgare i Nymphaeaceae Echinopsis spachiana d r Cleistocactus samaipatanus Paeonia delavayi Helminthotheca echioides Onagraceae o Centaurea nigra o Prunus avium Papaveraceae

C Malus domestica 0.000 Jacobaea vulgaris Primulaceae Matucana paucicostata Prunus spinosa Tanacetum vulgare Rosaceae Scrophulariaceae Achillea ptarmica Achillea millefolium Cistus heterophyllus Paeoniaceae Aloe perdita Gesneriaceae Aloe lateritia Other families Pinus ayacahuite

-0.075 Aloe vera Anthriscus sylvestris Catharanthus trichophyllus Philadelphus

Eucalyptus perriniana

-0.150 -0.300 -0.225 -0.150 -0.075 0.000 0.075 0.150 Coordinate 1

Fig. S1. 2D-NMDS (non-metric multudimensional scaling) plot of pollen sterol profile similarities (Bray-Curtis) between plant species with major plant families highlighted with different symbols (see legend). Distances between points correspond to sterol profile dissimilarities.

) n e l l o p

g 20 m / g µ (

t n e t n o c

l o r e t s e l

o 10 h c e n e l y h t e M - 4 2

0

No Yes Pollen as reward for bees

Fig. S2. 24-methylenecholesterol content (µg/mg pollen) of plants without pollen as bee reward (no bee pollination/collection of pollen by bees; n = 22), or with pollen as reward for bees (based on evidence of bee pollination and pollen collection by bees; n = 54).

40

) 30 n e l l o p

g m / g µ (

t

n 20 e t n o c

l o r e t s

l a t

o 10 T

0

No Yes Pollen as reward for bees

Fig. S3. Total sterol content (µg/mg pollen) of plants without pollen as bee reward (n = 22), or with pollen as reward for bees (n = 54). Indentations represent 95% confidence intervals.

Fig. S4. GC-MS spectra of the 25 phytosterols identified in our study (after Tri-sil derivatisation, extraction details see Materials and methods section).