Article | Open

Teosinte in Europe – Searching for the Origin of a Novel Weed

  • Scientific Reports 7, Article number: 1560 (2017)
  • doi:10.1038/s41598-017-01478-w
  • Download Citation
Published online:


A novel weed has recently emerged, causing serious agronomic damage in one of the most important maize-growing regions of Western Europe, the Northern Provinces of Spain. The weed has morphological similarities to a wild relative of maize and has generally been referred to as teosinte. However, the identity, origin or genetic composition of ‘Spanish teosinte’ was unknown. Here, we present a genome-wide analysis of single-nucleotide polymorphism (SNP) data for Spanish teosinte, sympatric populations of cultivated maize and samples of reference teosinte taxa. Our data are complemented with previously published SNP datasets of cultivated maize and two Mexican teosinte subspecies. Our analyses reveal that Spanish teosinte does not group with any of the currently recognized teosinte taxa. Based on Bayesian clustering analysis and hybridization simulations, we infer that Spanish teosinte is of admixed origin, most likely involving Zea mays ssp. mexicana as one parental taxon, and an unidentified cultivated maize variety as the other. Analyses of plants grown from seeds collected in Spanish maize fields and experimental crosses under controlled conditions reveal that hybridization does occur between Spanish teosinte and cultivated maize in Spain, and that current hybridization is asymmetric, favouring the introgression of Spanish teosinte into cultivated maize, rather than vice versa.


Increasing human transfer of plants across geographical regions leads to frequent introductions of non-native plant species around the world. If these introduced taxa meet close relatives, hybridization and introgression can occur1. The inadvertent introduction of species closely related to major crop plants may lead to hybridization and the formation of weedy lineages2. Alternatively, weedy taxa can also evolve directly from a domesticated ancestor3, as in the case of some weedy rice populations4, 5. Weedy lineages can incur massive yield losses and cause major costs6.

Maize is the third most important crop plant in Spain with production reaching almost 4.7 million tonnes7. In 2009, farmers in Northern Spain (Aragon) began to observe plants in their maize fields that resembled cultivated maize before the onset of flowering but then developed highly branching phenotypes with small cobs and shattering seeds8. These traits are typical for teosinte, wild relatives of cultivated maize9. Until 2014, this so-called ‘Spanish teosinte’ has spread in Aragon and has also been reported from a neighbouring region in Catalonia10, 11. About 750 ha of maize cultivation have been affected so far, mostly in Aragon11. Due to maize monocropping, density of Spanish teosinte can become high on affected fields and may cause severe maize yield losses and high economic costs12. In some regions of Aragon, this weed has become the prime agronomic problem for maize farmers10.

Maize (Zea mays mays) was domesticated from its wild relative teosinte about 9,000 years ago in southern Mexico13. Domesticated maize, including high-yielding hybrid maize varieties grown in Europe, is strictly monopodic, with non-shattering kernels that remain tightly attached to the cob, in contrast to the shattering kernels of the cob-less, and highly branched, teosinte. Although maize cobs, or parts of them, can remain in the field and do germinate and grow feral, it is considered unlikely that they successfully establish feral populations beyond arable fields and without human support9. Hence, cultivated maize is generally considered to have little to no risk of causing concerns as a volunteer or feral weed14.

Teosinte is the common name of a group of wild grasses (Poaceae) and includes highly variable species and subspecies that occur in scattered populations in many areas across Mexico and Central America (Mesoamerica)15. In many areas of Mesoamerica, teosinte populations have come under serious threat with the expansion of ranching and farming and are facing a massive decline in abundance to the point of extinction of some species16, which forced the Mexican government to install conservation measures for their protection17. Thus, in their centre of origin, many teosinte populations are endangered and require protection measures, although occasionally they can also act as local weeds18.

Teosinte and maize belong to the same genus, Zea, which consists of five species: 1) perennial diploid (2n = 20) Z. diploperennis, 2) perennial tetraploid (2n = 40) Z. perennis, 3) annual diploid (2n = 20) Z. luxurians, 4) annual diploid (2n = 20) Z. nicaraguensis, and 5) the annual species Zea mays. The latter encompasses four annual diploid (2n = 20) subspecies: (i) ssp. mays, the domesticated maize, (ii) ssp. mexicana, (iii) ssp. parviglumis, and (iv) ssp. huehuetenangensis19. Z.m. ssp. mexicana and Z.m. ssp. parviglumis are most closely related to domesticated maize, the latter subspecies being called ‘Balsas’ teosinte and considered the ancestor of cultivated maize9. All teosintes are believed to be endemic to Mesoamerica20 where cultivated maize and teosintes often grow in geographic proximity and flower synchronously. Overall, Z.m. ssp. mexicana grows in cooler, drier central highlands, mostly above 1800 m, while Z.m. ssp. parviglumis grows in warmer, wetter lower elevations in the river valleys of southern and western Mexico, mostly below 1800 m15.

Although it is known that all teosintes can hybridize with maize, this typically occurs at low rates even when teosinte is abundant21. Hybridization appears to be most common between domesticated maize and Z.m. ssp. parviglumis15, 22 but gene flow does not occur reciprocally, which may explain why teosintes continue to coexist even when growing in close vicinity to much larger maize populations21. When teosinte pollen is applied to maize silks, resulting hybrids are vigorous and highly fertile23. However, when teosinte is pollinated by maize pollen, plants of Z.m. ssp. mexicana set seed very inconsistently or not at all22. Similarly, Hufford et al. found evidence of adaptive introgression of Z.m. ssp. mexicana alleles into maize during its expansion to the highlands of central Mexico, but observed very little evidence for adaptive introgression in the other direction, from cultivated maize into Z.m. ssp. mexicana24.

This asymmetrical pollination success is under the control of a gene called the ‘Teosinte crossing barrier’ (Tcb)23 which may reproductively isolate at least some teosinte species from maize. Aylor et al. postulated that gene flow and subsequent introgression of maize alleles into teosinte populations most likely occur when teosinte first pollinates maize20. The resultant hybrids would then backcross with teosinte which could lead to the introgression of maize alleles into the teosinte background. They speculated that the pollination from teosinte to maize most likely represents the rate-limiting step in the introgression of maize alleles into teosinte20. However, this has not been investigated yet to any conclusive extent, even though teosinte and maize have intensively been studied from a population genetics perspective, including studies utilizing DNA samples recovered from archaeological specimens13, 24,25,26,27.

So far, the Spanish authorities have speculated that the introduced teosinte is Z.m. ssp. mexicana11. Spanish teosinte can produce long-lasting seed banks, and its control either by mechanical means, crop rotation or herbicide treatment has proven difficult10, 11. Knowing the origin of the novel weed in Spain may help to prevent introductions of further seed material, to monitor the spread of this weed, and to take targeted control measures in the future. The taxonomic identity of the Spanish teosinte, its introduction history, as well as its ecology and potential risks for local and neighbouring European farming systems remain largely unknown. To develop effective measures for monitoring, control and prevention, data on these aspects are fundamental. Here, we addressed the following research questions: (i) to what taxon can teosinte-like plants collected in Spain be assigned?, (ii) what is the potential origin of these Spanish teosinte lineages?, (iii) is there evidence for on-going hybridization between Spanish teosinte and commercially cultivated maize in Spain? To answer these questions, we collected Spanish teosinte and hybrid-like seeds (autumn 2014 and 2015) and leaf samples of Spanish teosinte and cultivated maize (summer 2015) in the region of Aragon, Spain, and genotyped these together with teosinte reference plants using the MaizeSNP50 BeadChip, a widely used resource for high-density genotyping of maize and its wild relatives28,29,30.


Using PCA, the SNP data allowed a clear separation between commercial maize varieties and all teosinte taxa included in this study (Fig. 1). Samples from Spanish maize varieties collected in the field or grown from seeds grouped with other commercial maize varieties previously analysed by Olukolu et al.29. Similarly, reference samples of Z.m. ssp. mexicana and ssp. parviglumis obtained from USDA grouped with samples of these species previously investigated by Pyhäjärvi et al.28 (Fig. 2). All other teosinte species obtained from CIMMYT and USDA formed separate groups (Figs 1 and 2). Teosinte from Brazil grouped with reference samples of Z. luxurians, confirming the results of Silva et al.31.

Figure 1
Figure 1

PCA of maize, teosinte and hybrid samples collected in Spain. Samples were collected from plants growing in the field (black labels), from seeds collected in the field and subsequently grown in a climate chamber (grey labels) or from reference material (grey labels). Labelled symbols represent own data that were compared to other publically available data (without labels) (27′476 SNPs and 662 individuals).

Figure 2
Figure 2

PCA of teosinte samples grown from the seeds obtained as a reference material. Labelled symbols represent own data that were compared to other publically available data (without labels) (27′476 SNPs and 280 individuals).

Samples collected in maize fields in Spain that did not represent commercial maize varieties formed two groups. One group, containing all individuals initially scored as Spanish teosinte plants, was clearly separated from commercial maize varieties but also from all currently recognized Mexican and Nicaraguan teosintes (Fig. 1). The second group of plants from Spain were intermediate between Spanish teosinte plants and commercial maize varieties (Fig. 1). Plants in this group were grown from hybrid-like seeds collected in the field that shared phenotypic features of maize and Spanish teosinte (represented by light green circles in Fig. 1). In addition, this group also encompassed our own experimental F1 hybrids between Spanish commercial maize and Spanish teosinte plants in which the teosinte served as pollen donors.

STRUCTURE analysis of samples from Spain, together with Z.m. ssp. mexicana and ssp. parviglumis, also clearly separated Spanish teosinte from the two subspecies of Z. mays and further revealed that Spanish teosinte shares some alleles with Z.m. ssp. mexicana. Further, individuals grown from the seeds that were collected in the field in Spain and that were intermediate between Spanish teosinte plants and Spanish commercial maize (represented by light green circles in Fig. 1) were found to be early generation hybrids between maize cultivated in Spain and Spanish teosinte (Fig. 3).

Figure 3
Figure 3

STRUCTURE analysis of Spanish maize varieties, hybrids, Spanish teosinte, Z.m. ssp. mexicana and ssp. parviglumis. Bar plot of assignment proportions at K = 6 combining own SNP data and other publically available data (27′476 SNPs and 360 individuals).

Because Spanish teosinte plants did not group with any of the known Mexican and Nicaraguan teosinte taxa, we evaluated a possible admixed origin by simulating hybridization between Spanish commercial maize genotypes and different teosinte species. Results suggest that commercial maize varieties currently planted in Spain were not involved in the formation of Spanish teosinte (Fig. 4), which is also consistent with the outcome of the STRUCTURE analysis (Fig. 3).

Figure 4
Figure 4

PCA of simulated hybridization between Spanish maize varieties and Z.m. ssp. mexicana. Coloured symbols represent empirical data (27′476 SNPs and 360 individuals) and black symbols represent simulated data.


The novel weed found in maize fields in Spain has been tentatively identified as a wild relative of maize, teosinte, but no systematic effort to establish its identity or ancestry has been undertaken until now. Knowing the identity of this weed is not only useful for monitoring and control, but fundamental to trace its origin and to assess its potential for interbreeding and future evolution towards a possibly even more damaging weed. We, therefore, genotyped the Spanish teosinte plants together with commercially grown maize and all currently recognized teosinte taxa from Mexico and Nicaragua. Surprisingly, the Spanish teosinte was not only clearly separated from cultivated maize but also from all teosinte species tested (Fig. 1). These results are not compatible with the published suggestion that teosinte plants observed in Spain represent Z.m. ssp. mexicana11.

In one attempt at testing the hypothesis that there could be some relation between Spanish teosinte and expatriated populations of teosintes sensu stricto, we included material from a Brazilian population of teosinte that was introduced there as a forage species. Recently, it was reported that this teosinte from southern Brazil belonged in the species Z. luxurians31, based on morphological traits and the location of chromosomal knobs. However, no genetic analysis had been performed to date. Our results confirmed that this Brazilian teosinte is Z. luxurians, as suggested by the study of Silva et al.31, but it did not associate with Spanish teosinte.

The fact that Spanish teosinte did not group with any currently recognized teosinte taxa from Mexico and Nicaragua, or with known populations of expatriated teosintes, led us to hypothesize that it might be of admixed origin. Indeed, our STRUCTURE analysis revealed that Spanish teosinte shares alleles with Z.m. ssp. mexicana. Very likely, Spanish teosinte could originate from hybridization and subsequent backcrossing between cultivated maize and Z.m. ssp. mexicana. Such a process would seem necessary to explain our data, given the current state of knowledge on the asymmetric crossing behaviour between maize and teosintes in general20,21,22,23. Simulated hybridization suggests that the hybridization leading to Spanish teosinte likely did not occur in very recent years. Whether such a weed had originally formed in Europe or elsewhere is open to further research.

The key hybridization event(s) leading to the emergence of Spanish teosinte could have taken place inadvertently in any of the Mesoamerican regions where native Z.m. ssp. mexicana and domesticated maize coexist, or indeed through human intervention elsewhere, including Europe, as Z.m. ssp. mexicana has been used in various breeding programs in order to improve agronomic traits of cultivated maize32 or to evaluate the suitability of the hybrids as a forage grass33, 34. Against such a single-introduction hypothesis, the origin of Spanish teosinte could also be complex, resembling the case of weedy rice in the USA, where some populations seem to have arisen via hybridization between cultivated rice and weedy rice35, whereas other seem to have evolved directly from cultivated rice5, 36.

The risk of further hybridization events between commercial maize and Spanish teosinte is worth considering in light of the possibility of further development of even more invasive weed than the population currently found in Northern Spain. The plants grown from hybrid-like seeds, which we collected in the field in Spain, proved indeed to be intermediate between cultivated maize and teosinte, having elongated lateral branches and producing bunches of ears, often combined with tassels. These samples grouped together (light green circles in Fig. 1) with our experimental F1 hybrids derived from crosses in which Spanish commercial maize plants acted as pollen recipients and Spanish teosinte as pollen donors. Thus, we suggest that further hybridization is not only possible, but is already happening in the field, albeit so far we have only encountered teosinte-to-maize hybrid seed, and not maize-to-teosinte seed.

Surprisingly, we have not detected any hybrids among the plants that we sampled directly in the field. A plausible explanation is that when maize is pollinated by teosinte, the seeds do not easily shatter, with many of them being harvested together with the commercially grown maize. Pollination occurring in the other direction, from cultivated maize to teosinte, is known to be significantly less viable, due to the control of Teosinte crossing barrier (Tcb)23. Lu et al. observed that in silk carrying the Tcb, pollen tubes had clustered callose plugs and their growth was slower in comparison to pollen tubes of compatible crosses37. Such crossing incompatibility may also exist between commercial maize and Spanish teosinte. From our exploratory crossing experiments, we can confirm that hand-pollination of Spanish teosinte with maize pollen results only rarely in viable seeds (unpublished results). Thus, while the crossing barrier may reduce the frequency with which such hybrids are formed in the field, it is important to note that it does not preclude their formation, even though at low rates. Therefore, we expect that more intensive sampling, over a longer period of time, may likely reveal the existence of maize-to-teosinte introgressants, especially if on-going hybridization between teosinte and maize leads to the formation of hybrids that are more compatible with a teosinte mother plant, as postulated by Aylor et al.20.

Further studies are needed to fully understand the evolutionary origin and demographic processes involved in the formation of Spanish teosinte. In particular, much more intensive sampling, both spatially and temporally, could reveal the amount of genetic and phenotypic diversity, as well as the extent to which the population of Spanish teosinte is changing, including through hybridization with cultivated maize. However, we now have a basis of understanding that enables necessary practical measures to confront the challenge of this serious, possibly invasive, weed. With this understanding, it should be possible to establish monitoring and mapping efforts to track, and hopefully contain, the spread of the weed. Whether this weed remains a serious problem only in the Spanish region where it is now confined, or whether it expands to other maize-growing areas in Europe and beyond may be determined by the degree with which these questions are pursued.


Plant material

Teosinte and hybrid-like seeds were collected from two different sites in the region of Zaragoza, Spain in autumn 2014 and 2015. In summer 2015 we surveyed the maize producing region around Zaragoza and collected leaf samples from three fallow and five standing maize fields. We sampled 3–20 plants per field (Supplementary Table S1). The leaf samples were dried and stored on silica gel for further molecular analysis.

Seeds collected in the field were germinated and grown under controlled conditions (20–25 °C, 50–65% rh, 16/8 h L/D) in climate chambers at ETH Zurich. Once matured, leaf samples were collected from 34 plants (Supplementary Table S1), dried and stored on silica gel for further molecular analysis.

Reference material of the different teosinte taxa was obtained from CIMMYT and USDA (Supplementary Table S1). Seeds of teosinte grown in Brazil were obtained from local markets in the state of Santa Catarina, Brazil. All teosinte seeds were shelled and germinated on filter paper under controlled conditions in the climate chamber (20–25 °C, 50–65% rh, 16/8 h L/D). Also, the seeds of the following maize varieties grown in Spain were germinated on filter paper: LG30490YG, ES TORQUAZ, ROJO, PR33D48, DKC66-66. The first leaves were sampled, lyophilized and stored in silica gel for further molecular analysis.

Experimental crosses were performed with six Spanish insect resistant Bt maize (LG30490YG) and 14 teosinte plants grown from the seeds collected in Spain under controlled conditions in a climate chamber (20–25 °C, 50–65% rh, 12/12 h L/D). The adventitious roots were sampled, lyophilized and stored on silica gel for further molecular analysis. Three seeds from one cob from a maize plant (mother) pollinated by teosinte were germinated and sampled in the same way as the reference material.

SNP genotyping

DNA was extracted from dried leaf and root material with CTAB buffer, following a slightly modified protocol described in Doyle and Doyle38, and quantified using NanoDrop (Thermo Science). For the experimental crosses, DNA for the mother plant (Bt maize) was pooled from 6 different plant samples using the same concentration levels, similarly, for the father plant (teosinte) DNA from 14 different plant samples was pooled together. Genotyping was conducted at the UC Davis Genome Center using the MaizeSNP50 BeadChip and Infinium HD Assay (Illumina, San Diego, CA, USA).

Data analysis

SNPs were called using GenomeStudio V2009.1 (Illumina). Gen Train score had to be larger than 0.7 to retain a SNP leading to 41,784 SNPs. SNP data were deposited in the Dryad repository ( To the dataset we added published and publically available data on Z.m. ssp. mexicana and ssp. parviglumis individuals28 and commercial maize varieties29 (Supplementary Table S1).

Genotype data was then imported into R39 and individuals and loci with more than 12% and 5% missing data, respectively, were excluded from the analysis, leading to 27′476 SNPs and 662 individuals. From that dataset we used subsets of data for the different analyses. The packages adegenet40 and FactoMineR41 were used to conduct principal component analysis (PCA). Our samples from Spain were assigned into three different categories: Spanish maize, Spanish teosinte, and hybrid, based on the morphological traits shown in Fig. 5.

Figure 5
Figure 5

Basic morphological differences between Spanish commercial maize, hybrid and Spanish teosinte plants. Plants were grown in a climate chamber from seeds collected in the field in Spain.

STRUCTURE analysis42 was conducted with different K’s and 10 replications using the admixture model with correlated allele frequencies. The analysis was performed separately for all Spanish maize varieties, hybrid, Spanish teosinte genotypes, Z.m. ssp. mexicana and ssp. parviglumis (Fig. 3) and for all commercial maize varieties, Spanish teosinte, Z.m. ssp. mexicana, ssp. parviglumis, Z. luxurians, Z. diploperennis, Z. perennis and Z. nicaraguensis (Supplementary Fig. S1). The best K was then selected using STRUCTURE harvester43.

To simulate the putative hybrids between the Spanish maize and Z.m. ssp. mexicana we use the hybrid function of adegenet40 with 10 individuals each.

Additional Information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


  1. 1.

    Ellstrand, N. C. et al. Introgression of crop alleles into wild or weedy populations. Annu. Rev. Ecol. Evol. Syst. 44, 325–345, doi:10.1146/annurev-ecolsys-110512-135840 (2013).

  2. 2.

    Ellstrand, N. C. et al. Gene Flow and Introgression from Domesticated Plants into Their Wild Relatives. Source Annu. Rev. Ecol. Syst. 30, 539–563, doi:10.1146/annurev.ecolsys.30.1.539 (1999).

  3. 3.

    Ellstrand, N. C. et al. Crops gone wild: Evolution of weeds and invasives from domesticated ancestors. Evol. Appl. 3, 494–504, doi:10.1111/j.1752-4571.2010.00140.x (2010).

  4. 4.

    Cao, Q. et al. Genetic diversity and origin of weedy rice (Oryza sativa f. spontanea) populations found in North-eastern China revealed by simple sequence repeat (SSR) markers. Ann. Bot. 98, 1241–1252, doi:10.1093/aob/mcl210 (2006).

  5. 5.

    Reagon, M. et al. Genomic patterns of nucleotide diversity in divergent populations of U.S. weedy rice. BMC Evol. Biol. 10, 180, doi:10.1186/1471-2148-10-180 (2010).

  6. 6.

    Valverde, B. E. In Crop Ferality and Volunteerism (ed. Gressel, J.) 279–294 (2005).

  7. 7.

    FAO. FAOSTAT. (2014).

  8. 8.

    Pardo, G. et al. El Teosinte (Zea mays, spp.). Cent. Sanid. y Certificación Veg. DGA (2014).

  9. 9.

    Doebley, J. The genetics of maize evolution. Annu. Rev. Genet. 38, 37–59, doi:10.1146/annurev.genet.38.072902.092425 (2004).

  10. 10.

    Pardo, G. et al. Presencia de teosinte (Zea spp.) como mala hierba en los regadíos del valle del Ebro. XV Congr. la Soc. Española Malherbología La Malherbología y la Transf. tecnológica Sevilla, 19–22 Oct. 2015 417–424 (2015).

  11. 11.

    Pardo, G. et al. El teosinte: descripción, situación actual en el valle del Ebro y resultados de los primeros ensayos. Vida Rural 408, 42–47 (2016).

  12. 12.

    Pardo, G., Cirujeda, A. & Martínez, Y. Evaluación del impacto económico de una especie invasora en el regadío de Aragón: el teosinte. Rev. Española Estud. Agrosociales y Pesq. 3, 67–96 (2016).

  13. 13.

    Matsuoka, Y. et al. A single domestication for maize shown by multilocus microsatellite genotyping. Proc. Natl. Acad. Sci. USA 99, 6080–6084, doi:10.1073/pnas.052125199 (2002).

  14. 14.

    Andersson, M. S. & de Vicente, M. C. Gene Flow between Crops and Their Wild Relatives. (The Johns Hopkins University Press, 2010).

  15. 15.

    Fukunaga, K. et al. Genetic diversity and population structure of teosinte. Genetics 169, 2241–2254, doi:10.1534/genetics.104.031393 (2005).

  16. 16.

    Wilkes, G. Urgent notice to all maize researchers: Disappearance and extinction of the last wild teosinte population is more than half completed. A modest proposal for teosinte evolution and conservation in situ: The Balsas, Guerrero, Mexico. Maydica 52, 49–58 (2007).

  17. 17.

    Benz, B. F. In Recent advances in the conservation and utilization of genetic resources: Proceedings of the Global Maize Germplasm Workshop (ed. CIMMYT) 59–69 (1988).

  18. 18.

    Wilkes, H. G. Hybridization of Maize and Teosinte, in Mexico and Guatemala and the Improvement of Maize. Econ. Bot. 31, 254–293, doi:10.1007/BF02866877 (1977).

  19. 19.

    Hufford, M. B., Bilinski, P., Pyhäjärvi, T. & Ross-Ibarra, J. Teosinte as a model system for population and ecological genomics. Trends Genet. 28, 606–615, doi:10.1016/j.tig.2012.08.004 (2012).

  20. 20.

    Aylor, D. E., Baltazar, B. M. & Schoper, J. B. Some physical properties of teosinte (Zea mays subsp. parviglumis) pollen. J. Exp. Bot. 56, 2401–2407, doi:10.1093/jxb/eri232 (2005).

  21. 21.

    Baltazar, B. M., de Jesus Sanchez-Gonzalez, J., de la Cruz-Larios, L. & Schoper, J. B. Pollination between maize and teosinte: an important determinant of gene flow in Mexico. Theor. Appl. Genet. 110, 519–526, doi:10.1007/s00122-004-1859-6 (2005).

  22. 22.

    Ellstrand, N. C., Garner, L. C., Hegde, S., Guadagnuolo, R. & Blancas, L. Spontaneous hybridization between maize and teosinte. J. Hered. 98, 183–187, doi:10.1093/jhered/esm002 (2007).

  23. 23.

    Evans, M. M. S. & Kermicle, J. L. Teosinte crossing barrier1, a locus governing hybridization of teosinte with maize. Theor. Appl. Genet. 103, 259–265, doi:10.1007/s001220100549 (2001).

  24. 24.

    Hufford, M. B. et al. The Genomic Signature of Crop-Wild Introgression in Maize. PLoS Genet. 9(5), e1003477, doi:10.1371/journal.pgen.1003477 (2013).

  25. 25.

    Jaenicke-Després, V. et al. Early allelic selection in maize as revealed by ancient DNA. Science 302, 1206–8, doi:10.1126/science.1089056 (2003).

  26. 26.

    Ross-Ibarra, J., Tenaillon, M. & Gaut, B. S. Historical divergence and gene flow in the genus Zea. Genetics 181, 1399–1413, doi:10.1534/genetics.108.097238 (2009).

  27. 27.

    van Heerwaarden, J. et al. Genetic signals of origin, spread, and introgression in a large sample of maize landraces. Proc. Natl. Acad. Sci. 108, 1088–1092, doi:10.1073/pnas.1013011108 (2011).

  28. 28.

    Pyhäjärvi, T., Hufford, M. B., Mezmouk, S. & Ross-Ibarra, J. Complex patterns of local adaptation in teosinte. Genome Biol. Evol. 5, 1594–1609, doi:10.1093/gbe/evt109 (2013).

  29. 29.

    Olukolu, B. A. et al. A connected set of genes associated with programmed cell death implicated in controlling the hypersensitive response in maize. Genetics 193, 609–620, doi:10.1534/genetics.112.147595 (2013).

  30. 30.

    Arteaga, M. C. et al. Genomic variation in recently collected maize landraces from Mexico. Genomics Data 7, 38–45, doi:10.1016/j.gdata.2015.11.002 (2016).

  31. 31.

    Silva, N. C. D. A., Vidal, R., Costa, F. M., Vaio, M. & Ogliari, J. B. Presence of Zea luxurians (Durieu and Ascherson) Bird in Southern Brazil: Implications for the conservation of wild relatives of maize. PLoS One 10(10), e0139034, doi:10.1371/journal.pone.0139034 (2015).

  32. 32.

    Wang, L., Yang, A., He, C., Qu, M. & Zhang, J. Creation of new maize germplasm using alien introgression from Zea mays ssp. mexicana. Euphytica 164, 789–801, doi:10.1007/s10681-008-9730-5 (2008).

  33. 33.

    Hall, A. D. The evaluation of teosinte-maize hybrid for forage. Proc. Agron. Soc. New Zeal. 7, 37–39 (1977).

  34. 34.

    Barrière, Y., Montalant, Y. & Boyat, A. Etudes des teneurs en protéines et des valeurs agronomiques de descendances de croisement maïs téosinte. Agron. EDP Sci. 4, 417–422, doi:10.1051/agro:19840502 (1984).

  35. 35.

    Londo, J. P. & Schaal, B. A. Origins and population genetics of weedy red rice in the USA. Mol. Ecol. 16, 4523–4535, doi:10.1111/j.1365-294X.2007.03489.x(2007).

  36. 36.

    Kanapeckas, K. L. et al. Escape to Ferality: The Endoferal Origin of Weedy Rice from Crop Rice through De-Domestication. PLoS One 11(9), 0162676, doi:10.1371/journal.pone.0162676 (2016).

  37. 37.

    Lu, Y., Kermicle, J. L. & Evans, M. M. S. Genetic and cellular analysis of cross-incompatibility in Zea mays. Plant Reprod. 27, 19–29, doi:10.1007/s00497-013-0236-5 (2014).

  38. 38.

    Doyle, J. & Doyle, J. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem. Bull. 19, 11–15 (1987).

  39. 39.

    R Development Core Team. R: A language and environment for statistical computing, reference index version 2.15.0. (R Foundation for Statistical Computing: Vienna, Austria, 2015).

  40. 40.

    Jombart, T. & Ahmed, I. adegenet 1.3-1: new tools for the analysis of genome-wide SNP data. Bioinformatics 27, 3070–3071, doi:10.1093/bioinformatics/btr521 (2011).

  41. 41.

    Lê, S., Josse, J. & Husson, F. FactoMineR: An R Package for Multivariate Analysis. J. Stat. Softw. 25, 1–18 (2008).

  42. 42.

    Pritchard, J. K., Stephens, M. & Donnelly, P. Inference of population structure using multilocus genotype data. Genetics 155, 945–959 (2000).

  43. 43.

    Earl, D. A. & vonHoldt, B. M. STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv. Genet. Resour. 4, 359–361, doi:10.1007/s12686-011-9548-7 (2012).

Download references


We thank Stiftung GEKKO, Stiftung Rübel, Zukunftsstiftung Landwirtschaft and Software AG Stiftung for financial support.

Author information


  1. ETH Zurich, Institute of Integrative Biology (IBZ), Universitätstrasse 16, 8092, Zurich, Switzerland

    • Miluse Trtikova
    • , Andre Lohn
    • , Bernadette Oehen
    • , Alex Widmer
    •  & Angelika Hilbeck
  2. Agroecology and Food Systems Chair, Universitat de Vic – Universitat Central de Catalunya, c/de la Laura 13, 08500, Vic, Spain

    • Rosa Binimelis
  3. University of California Berkeley, Department of Environmental Science, Policy and Management, 108 Hilgard Hall, 94720, Berkeley, USA

    • Ignacio Chapela
  4. ETH Zurich, Genetic Diversity Centre (GDC), Universitätstrasse 16, 8092, Zurich, Switzerland

    • Niklaus Zemp


  1. Search for Miluse Trtikova in:

  2. Search for Andre Lohn in:

  3. Search for Rosa Binimelis in:

  4. Search for Ignacio Chapela in:

  5. Search for Bernadette Oehen in:

  6. Search for Niklaus Zemp in:

  7. Search for Alex Widmer in:

  8. Search for Angelika Hilbeck in:


A.H., A.W., I.C. and M.T. conceived and designed the study. A.H., R.B., B.O. and M.T. conducted the field work. A.L. and M.T. performed the experiments and the laboratory work. N.Z. analysed the data and produced the figures. M.T., A.W. and A.H. wrote the main manuscript text. All authors reviewed the manuscript.

Competing Interests

The authors declare that they have no competing interests.

Corresponding author

Correspondence to Miluse Trtikova.

Supplementary information


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Creative Commons BY

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit