The origins and spread of the opium poppy (Papaver somniferum L.) revealed by genomics and seed morphometrics

Abstract

The opium poppy (Papaver somniferum L.) is one of the most important plants in human history. It is the main source of opiates used as analgesic medicines or psychotropic drugs, the latter related to addiction problems, illegal trafficking and geopolitical issues. Poppyseed is also used in cooking. The prehistoric origins, domestication and cultivation spread of the opium poppy remain unresolved. Traditionally, Papaver setigerum has been considered the wild ancestor with early cultivation presumed to have occurred in the Western Mediterranean region, where setigerum is autochthonous. Other theories suggest that somniferum may have been introduced by Southwest Asian early farmers as a weed. To investigate these hypotheses, we analysed 190 accessions from 15 Papaver species using genotype-by-sequencing and geometric morphometric (GMM) techniques. Our analysis revealed that setigerum is the only taxa genetically close to somniferum and can be better described as a subspecies. The domesticated plants are, however, distinct from setigerum. Additionally, GMM analysis of seeds also revealed morphological differences between setigerum and somniferum. Some phenotypically wild setigerum accessions exhibited intermediate genetic features, suggesting introgression events. Two major populations were found in somniferum and, to some extent, these correspond to differences in seed form. These two populations may reflect recent attempts to breed varieties rich in opiates, as opposed to varieties used for poppyseed production. This study supports the idea that opium poppy cultivation began in the Western Mediterranean, with setigerum as the wild progenitor, although some wild varieties are likely to be feral forms, which can confound domestication studies.This article is part of the theme issue 'Unravelling domestication: multi-disciplinary perspectives on human and non-human relationships in the past, present and future'.

Keywords: agrobiodiversity; domestication; opiates; plant genome; seed morphology.

Figure 1.

Population structure of 186 accessions of wild and cultivated Papaver based on 4102 SNPs (Dataset1) discovered by GBS. (A) PCA: wild accessions are shown as open points and cultivated somniferum as full points. (B) STRUCTURE K = 2 and K = 5 models.

Figure 2.

Population structure of 158 accessions of Papaver somniferum and Papaver setigerum based on 7048 SNPs (Dataset2) discovered by GBS. (A) PCA: accessions described as wild in the genebank passport data are shown as open points and those referred to as domesticated somniferum as full points; names of accessions referred to in the text are noted. (B) STRUCTURE K = 3 and K = 5 models.

Figure 3.

Geographical distribution of the three clusters identified in the STRUCTURE model K = 3 for setigerum and somniferum accessions. Each pie chart represents an individual accession and the proportional memberships to each of the three clusters.

Figure 4.

Association between environmental variables and allelic frequencies as well as between geographical and genetic distances, for the K = 3 model defined Pop1 and Pop2 (setigerum and somniferum accessions). (A,B) Random forest test for association between environmental variables and allele frequencies for Pop1 and Pop2, respectively.

Figure 5.

(A) GMM PC1 × PC2 factorial plane of the PCA on Papaver sp. shape variables. (B) Reconstructed mean shapes for the two poppy taxa shown with the intensity of differences between the two theoretical mean outlines.

Figure 6.

Scores on the four linear discriminant axes on a single and balanced (n = 150) LDA on the K = 5 grouping provided by STRUCTURE.