Traits in California weedy rice plants that mimic both medium-grain and gourmet specialty cultivars include erect leaf angles, cleft ligule shape, culm strength , green node color, panicle exertion , and ligule pubescence .Crop traits in weedy rice specific to co-occurring medium-grain cultivars include basal leaf sheath color and hull color . Crop traits grouping weedy rice with gourmet specialty varieties include intermediate tillering or spreading growth habit . These crop-like traits are presumably carried by the immediate progenitors of the weedy rice, the crop, and have not been lost by the crop. When shared between weedy and cultivated rice, some of these traits could reinforce weed persistence and adaptation to cultivation by visually disguising the weed, thus preventing detection.Wild-like traits in California weedy rice traits exhibit high variance and differentiate this nascent feral population from both medium-grain cultivated varieties M-104, M-202, M-204, and M-205 as well as gourmet varieties in several ways. Compared with medium-grain cultivars, California weedy rice has a purple pericarp, long fully developed awns, lower seed set but with more grains per panicle, open panicles with scabrous texture, more tillers and panicles, spreading growth habit, long culm with gold internodes and delayed and extended flowering period . The medium-grain cultivars in California have brown pericarp; short awns, , high seed set, indoor weed growing accessories less tillers and panicles than weedy rice, have compact panicles, are less than 100 cm tall, flower earlier than weedy rice, and have erect culms.
Distinct clusters within the California weedy rice population can be resolved by a multivariate analysis of variance . UPGMA cluster analysis confirmed morpho groups based on the key partitioned traits identified through PCA . The number of clusters N was given as three for several clustering methods. More clusters could be resolved amongst the weedy rice, but an additional split did not offer more information when co-occurring rice cultivars were included .Weedy rice in California is a newly established and distinct group in the USA. Isolation with migration modeling suggests that California SHA weedy rice diverged from California rice cultivars approximately 118 generations ago. The relatively recent divergence, distinct morphology, and small genetic relatedness with other US weedy rice indicate that this unique population has evolved separately from a cultivated ancestor. The recent origin of California weedy rice suggests that the population has differentiated since the establishment of rice cultivation in California and is in the early stages of segregating weedy traits, such as more tillers, extended flowering, and pigmented pericarp. Across all loci, we find no haplotypes in California cultivated rice that are not present in other japonicas . Further, there are no additional shared polymorphisms between California weedy rice and other japonicas that are not shared between California weedy and California crop rice. California weedy rice either diverged from japonicas outside of the US and was brought in to California after becoming weedy or diverged from California japonica cultivated in California.
The former argument that this weed was brought in is highly unlikely due to strict laws in rice seed import into California. Regardless, California weedy rice is distinct from the other US weedy rices and our coalescent IM estimates point to a recent de-domestication from the California japonica line. Indeed, while the generation values may be inflated , this does not necessarily mean that the estimate is greater than it should be. The relative values are indicative of the relatively recent divergence, which was our objective with this analysis. This result does not demonstrate definitively that time since divergence between California cultivated and weedy rice is different from that between BHA and California weedy rice; however, it is clear that California cultivated rice and California weedy rice have different origins, and more importantly, that the divergence of California weedy rice is more recently from cultivated rice in the same area than that from all other Oryza groups investigated. Moreover, while California weedy rice shares some similarity to weedy rice from the southern US at several loci as expected with a similar genetic background, divergence at few specific loci highlights changes required for California weedy rice to thrive in this novel agricultural system. If weedy rice in California is newly derived from a temperate japonica crop ancestor in current commercial fields, this suggests that new feral ecotypes can evolve anywhere rice is under domestication.
Our population genetic diversity analysis is consistent with recent establishment of the weed from a few individuals: low genetic diversity within and between California rice cultivars where weedy rice infestations occurred. Pairwise divergence estimates indicate high genetic divergence between California weedy rice and other groups. We examined the possibility that gourmetrice varieties could have been the source of California weedy rice because it may be an established cultivated or wild-weedy rice originating outside of North America that recently colonized California. Thus, we sequenced 12 informative STS loci in red colored pericarp gourmet rice varieties grown in California and found no shared haplotypes with California weedy rice . In addition, California weedy rice is distinct morphologically from gourmet varieties scored in this study . While both California weedy rice and its putative domestic progenitor share straw hull color, phenotypic traits that differentiate Californiaweedy red rice from cultivars include purple pericarp, increased plant height, longer leaves, greater tillering capacity, protracted flowering time, more grains per panicle, long awns, and seed shattering. The de-domestication origin of weedy rice direct from rice cultivars is not unique to this study, as at least two cases of O. sativa f. spontanea have been documented in the Guangdongand Liaoning provinces of China. Weedy rice in California is derived from rice cultivar in a smaller-scale rice growing region isolated from wild relatives and compliant with very strict guidelines for seed purity as well as tracking and reporting infestations. An earlier US weedy rice population genetics study by Londo and Schaal included microsatellite data for a single California weedy rice genotype, RR28, collected in California. Sequence analysis of the p-VATPase region of this genotype indicated that it was a unique, private haplotype genetically most similar to O. rufipogon—specifically O. rufipogon A100945-1 from southeast Asia used in crossing trials in California during the late 1970s. These M-101 crosses with O. rufipogon accessions produced weedy rice with red pericarp and shattering seeds to be used as a stem rot- resistant parent in breeding programs. California weedy rice RR28 appears to be a unique variant not found in our collection of thorough California weedy rice sampling. Moreover, included only one California cultivar in their study. This recently released cultivar is likely not the exclusive source of domestication alleles, so our study was designed to capture a comprehensive set of crop alleles from a wider range of California cultivars. Londo and Schaal reported that RR28 shares alleles with NSGC 5936 and California cultivar accession M205. The California ‘red’ rice specimen in our study, from the same county as the one used by [22], are four gourmet varieties with red bran. A single genotype cannot explain the population-level evolutionary history of weedy rice in California. We employed an extensive, broad sampling of wild, weedy, and cultivated genotypes to test the hypothesis of a wild/domestic hybrid in California. This sampling approach enabled us to examine how global and gourmet specialty rice sources have contributed to the localized evolution of weedy rice. The intermediate morphology of California weedy rice alone—strawhull awned and pigmented pericarp—does not necessarily indicate a crop-wild hybrid origin. Indeed, we found no evidence to suggest an escaped O. rufipogon source of weedy rice . From an evolutionary standpoint, vertical grow rack system the amount of morphological variation exhibited by this emergent lineage of weedy rice demonstrates its rapid adaptive potential and ability to persist in agroecosystems.
Zhang et al. also used UPGMA cluster analysis and principal component analysis to show that a spontaneously emergent weedy rice lineage is more closely related to rice cultivar grown in the sample field than with other neighboring cultivars or weedy varieties,supporting a de-domestication hypothesis that weedy rice can be derived from cultivated rice as we show in this study . The possibility that a widely cultivated species has a propensity to feralize under selection pressure variation has implications for crop management and necessitates further investigation on both the agroecological and molecular evolutionary levels .Morphological traits used to classify major ecotypes of US weedy rice , including awnedness, plant height and culm abundance, seed shattering, spikelet fertility , flowering, grain weight, and leaf size, are readily identifiable in the field . However, determining the evolutionary pathways to weedinessis imperative but challenging because de-domestication can follow different trajectories and proceed cryptically. Some weedy rice are visually indistinguishable from cultivars except for the shattering phenotype because some weedy rice have the same overall appearance and grain size as the cultivar, and have white pericarp, but this is not the case of California or other US weedy rice. Also, seed morphology differences between weedy and cultivated rice in the field may only become apparent after hulling when the red caryopsis is visible,effectively concealing the weed during invasion. Emergence, establishment, and spread of weedy rice in places that have no wild Oryza provide clues to how feral forms cryptically evolve. For example, re-seedingwith previously grown cultivated rice in Malaysia has selected for weedy ecotypes which shatter their seed. In instances where farmers cease cultivating a landrace to take advantage of a new cultivar with better yield, the agroecological conditions are altered to prime the environment for de-domestication to proceed. Abandoned land race rice can establish, and because farmers are familiar with the similar appearance of volunteer land race varieties under cultivation, they may not be identified as de-domesticated lineages. This suggests that management strategies must include monitoring at a smaller scale to ensure that escaping individuals with slightly higher variance in weedy traits are immediately identified rather than considered environmental variants of the cultivar. The historical eradication of weedy rice from California was largely due to intensive management including control of water and use of certified seed. However, unintentional dispersal of weedy individuals via a connectivity corridor such as a river or irrigation system could be responsible for either resurgence or maintenance of weedy rice populations.Evolution of weedy rice by de-domestication is not simply “domestication in reverse,” and involves the interplay of a greater number of mechanistic drivers than rice domestication. Temporal variation in US weedy rice flowering strategies and shared haplotypes with crop ancestors suggest hybridization and evolution on a short timescale. The gene or processes conferring a wild or weedy trait during de-domestication may not be the same ones responsible during domestication. However, because variation is limited to ancestral standing variation and novel mutations, weeds evolving directly from crops consist of fewer de-domesticated haplotypes, making these cases ideal for testing hypotheses about adaptive evolution during dedomestication. Understanding the origin of weedy rice in Californiawill be useful in managing this weed and controlling the evolution of future feral ecotypes. Pinpointing the molecular evolution and genomic factors involved in the weediness associated with this endoferal origin of weedy rice in California is currently underway . In the absence of directed husbandry, domesticated lines will maintain or acquire weedy or wild traits to ensure success in resource capture, survival, and fecundity. This dedomestication, or evolutionary reversion to wild-type morphology , can happen through processes involving hybridization with endemic or introduced relatives and involve either natural selection acting on standing genetic variation in the domesticated populations or gene flow from wild relatives. Modes of weed evolution include a domesticate origin with genetic contributions from weedy or wild relatives , or solely domesticate-derived forms . Seed dispersal is pivotal to the establishment and maintenance of grasses, where the level of shattering is directly related to a weed’s fitness. Because non-shattering is selected for in domesticated grasses, weeds evolving from crops would have to revert to shattering either through gene flow or by re-evolving the trait, assuming there is no standing variation for this trait in domesticated populations. Extended seed dormancy could also increase fitness of a weed, but would be selected against in a crop because cultivation would be difficult.Therefore, the reversion to seed dormancy would be selected for in weedy species.In addition, it is possible that pigmented pericarp may be selected for in weedy rice evolution due to its association with the ability of seeds to remain dormant for long periods of time.