cropsci_c07-09-0497 Hall.indd

Flax (Linum usitatissimum L.) is being evaluated as a crop platform for the production of bio-industrial and nutraceutical products. An important consideration for the release of any novel trait is the potential for gene fl ow to wild or weedy relatives and the impact it may have on their populations. The potential for gene introgression from transgenic fl ax to wild relatives, the occurrence, the phylogeny of fl ax wild relatives and reported interspecifi c hybridization was reviewed to initiate the evaluation of environmental risk of novel fl ax in Canada. The genus Linum contains approximately 230 species which are distributed in many parts of the world and may grow in sympatry with cultivated fl ax. Interspecifi c hybridization and cytogenetic studies between fl ax and congeneric species demonstrated that cultivated fl ax has the ability to hybridize and form viable F1 plants with at least nine species of Linum (L. africanum, L. angustifolium, L. corymbiferum, L. decumbens, L. fl occosum, L. hirsutum, L nervosum, L. pallescens, and L. tenue). Hybridization of fl ax with many other wild relatives has either not been studied or reported. However, based on the evidence of reported hybridization with wild or weedy relatives, gene fl ow from fl ax to wild or weedy relatives is possible in several species native to North America, depending on species distribution, sympatry, concurrent fl owering, ploidy level, and sexual compatibility. Amit J. Jhala and Linda M. Hall, Dep. of Agricultural, Food and Nutritional Science, Univ. of Alberta, Edmonton, AB, T6G 2P5 Canada. Linda M. Hall, Alberta Agriculture and Food, 410 Agriculture/Forestry Building, Univ. of Alberta, Edmonton, AB, T6G 2P5 Canada. Jocelyn C. Hall, Dep. of Biological Sciences, Biological Sciences Center, Univ. of Alberta, Edmonton, AB, T6G 2E9 Canada. Received 6 Sept. 2007. *Corresponding author ([email protected]). Abbreviations: AFLP, amplifi ed fragment length polymorphism; APG, Angiosperm Phylogeny Group; CFIA, Canadian Food Inspection Agency; EMBO, European Molecular Biology Organization; GM, genetically modifi ed; ITS, internal transcribed spacer region; NCRPIS, North Central Regional Plant Introduction Station; PGRC, Plant Gene Resources of Canada; RAPD, random amplifi cation of polymorphic DNA; sad2, stearoyl-ACP desaturase II. Published in Crop Sci. 48:825–840 (2008). doi: 10.2135/cropsci2007.09.0497 © Crop Science Society of America 677 S. Segoe Rd., Madison, WI 53711 USA All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher. R e p ro d u c e d fr o m C ro p S c ie n c e . P u b lis h e d b y C ro p S c ie n c e S o c ie ty o f A m e ri c a . A ll c o p y ri g h ts re s e rv e d . 826 WWW.CROPS.ORG CROP SCIENCE, VOL. 48, MAY–JUNE 2008 Fiber fl ax prospered in North America for many years as production followed settlers westward (Hammond and Miller, 1994). Fiber fl ax has been cultivated in the Netherlands and probably in Belgium and northern France since ancient times. Today, fi ber fl ax is grown primarily in China, Russia, Egypt, and near the northwestern European coast for the production of high quality linen (Vromans, 2006). In 2004, linseed was grown in 47 countries, and seed production was 1.903 million metric tonnes (Smith and Jimmerson, 2005). Canada, China, and the United States together are responsible for 64% of the total world fl ax seed output. Canada is currently the world’s leader in the production and export of fl ax seed, a position it has held since 1994. In 2006, Canada produced 1.041 million tonnes of fl ax seed (Statistics Canada, 2006) and exported 80 to 90% of the total production, mainly to Europe, the U.S., Japan, and South Korea (Flax Council of Canada, 2007a). Flax was among the fi rst crop species to be both genetically engineered with Agrobacterium mediated transformation and transformed with genes of potential agronomic value (McHughen, 2002). Several novel traits have been expressed in fl ax including chlorsulfuron and metsulfuron methyl resistance (McSheff rey et al., 1992), glufosinate-ammonium resistance (McHughen and Holm, 1995), and glyphosate resistance ( Jordan and McHughen, 1988). Only one transgenic fl ax cultivar, CDC Triffi d (McHughen et al., 1997), was released in Canada in 1998 for unconfi ned use in fi elds with persistent herbicide residues (CFIA, 2004b), but it was deregistered almost immediately at the request of the fl ax industry (Flax Council of Canada, 2007b). Although transgenic fl ax may have been a solution to signifi cant agronomic issues such as weed control (McHughen, 1989) or disease resistance (Polyakov et al., 1998), a concern over the market’s reaction to the import of genetically modifi ed material halted all transgenic development of the crop, even for primary use in paint and fl ooring industries and animal feed as a co-product. Since then, the EU is moving toward being more open to bioproducts and transgenic crops (Hricova, 2002; Breithaupt, 2004; see also Millam et al., 2005). GM crops are now grown worldwide, and the number of species and the area under production continues to increase ( James, 2003; Nap et al., 2003). One of the critical concerns that must be addressed before the release of a novel crop is the potential movement of transgenes from GM crops to wild populations (Raybould and Gray, 1993; CFIA, 2004a). A better understanding of crop-towild gene fl ow is essential for ecological risk assessment of the potential for transgene spread (Dale, 1993; Conner et al., 2003). In addition, the potential impact on biodiversity (Wilkinson et al., 2003) and genetic resources must be evaluated (Ellstrand, 1988; Andow and Alstad, 1998). Risk assessment of transgenic fl ax including transgene movement from transgenic fl ax to its weedy relatives is in progress (Hall et al., 2006). We hypothesize that L. usitatissimum is more likely to hybridize with closely related species having a similar ploidy level, genome, and chromosome pairing. Our objective is to establish the potential risk of gene fl ow from transgenic fl ax before experimental testing, based on (i) biology, distribution, and fl owering phenology of closely related species to fl ax, (ii) relatedness and crossability, and (iii) probability of interspecifi c hybridization and introgression between transgenic fl ax and its wild or weedy species. TAXONOMY AND PHYLOGENY Flax is a member of the family Linaceae which is composed of 22 genera (Vromans, 2006) and approximately 300 species (Hickey, 1988; Heywood, 1993). Linaceae is placed in the order Linales by some taxonomists (Cronquist, 1981), but most recently the family has been placed in the order Malpighiales (APG II, 2003). Important genera in the family includes: Linum (230 species), Hugonia (40 species), Reinvardtia (two–four species), Anisadenia (two species), Roucheria (eight species) and Radiola (Heywood, 1993). The genus Linum is traditionally divided into fi ve sections, Linum, Linastrum, Cathartolinum, Dasylinum, and Syllinum (Winkler, 1931) with an additional section, Cliococca, added by Ockendon and Walters (1968). Cultivated fl ax, Linum usitatissimum, is placed in the section Linum. The taxonomy and classifi cation of Linum has changed with increased knowledge. Many researchers classifi ed Linum species either on the basis of morphological characters or center of origin (Linnaeus, 1857; De Candolle, 1904; Tammes, 1925; Vavilov, 1926; Winkler, 1931; Dillman, 1933; Dillman, 1953; Richharia, 1962). Alternatively, other researchers grouped Linum species based on chromosome number (Kikuchi, 1929; Nagao, 1941; Ray, 1944; Osborne and Lewis, 1962; Gill, 1966; Ockendon, 1971; Chennaveeraiah and Joshi, 1983; Gill, 1987). However, there is no single prevailing classifi cation scheme for this genus. The grouping of 41 Linum species proposed by Gill (1987), based on morphological, cytological, and interspecifi c compatibility evidence, will be followed in this paper. Phylogenetic studies based on molecular markers are limited. An amplifi ed fragment length polymorphism (AFLP) based phylogeny of 17 species of Linum is not compatible with traditional sections of the species (e.g., Winkler, 1931; Ockendon and Walters, 1968; Diederichsen and Richards, 2003), although there is evidence of fi ve species clusters (Vromans, 2006). McDill and Simpson (2005) conducted a more comprehensive phylogenetic study of Linum based on DNA sequence variation from multiple chloroplast markers and the nuclear encoded internal transcribed spacer region (ITS). Their analysis of approximately 70 species indicates that blue-fl owered R e p ro d u c e d fr o m C ro p S c ie n c e . P u b lis h e d b y C ro p S c ie n c e S o c ie ty o f A m e ri c a . A ll c o p y ri g h ts re s e rv e d . CROP SCIENCE, VOL. 48, MAY–JUNE 2008 WWW.CROPS.ORG 827 Linum has a large number of diploid species that exhibit a remarkable diversity in chromosome number including n = 8, 9, 10, 12, 14, 15, 16, 18, 30, and > 30 (Darlington and Wylie, 1955; Gill, 1987). Diversity in chromosome numbers may be due to polyploidy and aneuploidy (Chennaveeraiah and Joshi, 1983). Initial studies of the chromosome number of cultivated fl ax estimated the chromosome number to be 2n = 32 (Martzenitzin, 1927; Lutkov, 1939). However, later cytogenetic and interspecifi c hybridization studies confi rm the chromosome number to be 2n = 30 (Kikuchi, 1929; Dillman, 1938; Nagao, 1941; Ray, 1944; Richharia, 1962; Gill, 1966; Chennaveeraiah and Joshi, 1983). The reasons for the confl icting results were the small size of the chromosomes in Linum, the tendency of the observed fragments to retain some stain (Ray, 1944; Gill, 1966), and an accidental segmentation in the somatic mitosis (Martzenitzin, 1927). There were some disagreements among various researchers regarding the chromosome numbers of other Linum species (Table 1). For example, Kikuchi (1929) classifi ed L. alpinum as a member of group III with chromosome number n = 18, whereas, Ray (1944) and Nagao (1941) have grouped this species as n = 9 (Table 1). Gill (1966) indicated uncertainty in the chromosome number of this species (Table 1). The Linum alpinum specimen from which Kikuchi (1929) counted chromosomes may be a Japanese tetraploid (Simonet and Chopinet, 1939), which could account for the variability in the results. Linum narbonense was grouped as n = 14 (Ray, 1944), but Kikuchi (1929) and Nagao (1941) observed n = 9, and 2n = 18 and/ or 36 (Gill, 1966). Linum monogynum has been reported, with qualifi cation, as n = 43 and 2n = 86 (Kikuchi, 1929). Linum hirsutum has a variable reported chromosome count of n = 8 (Ray, 1944), n = 9 (Nagao, 1941), n = 15 (Seetaram, 1972), and n = 16 or 18 (Gill, 1966; Table 1). CENTER OF ORIGIN AND EVOLUTION OF L. USITATISSIMUM The center of origin of cultivated fl ax is uncertain (Lay and Dybing, 1989) with many existing theories. Among the eight independent centers of origin of the world’s most important cultivated plants (Vavilov, 1926), Linum species were reported to have originated in four, the Central Asiatic, the Near Eastern, the Mediterranean, and the Abyssinian Center. Gill (1987) and Richharia (1962) have also discussed these four probable centers of fl ax origin. Alternatively, other researchers believe that Egypt could be a center of dissemination (De Candolle, 1904). Finally, an area east of the Mediterranean toward India has been suggested as another center of origin because a diverse form of fl ax is found in the area (De Candolle, 1904; Zeven, 1982). The progenitor of cultivated fl ax is also uncertain (Gill, 1987). Many authors reported cultivated fl ax is derived from two or more ancestral forms (De Candolle, Linum species were sister to a predominantly yellow-fl owered lineage. These lineages initially diversifi ed in Eurasia and members of both the blue and yellow-fl owered lineages appear to have independently colonized North America. The subsequent diversifi cation of the yellow fl owered Linum species in North America includes members previously classifi ed as separate genera: Hesperolinon, Sclerolinon digynum, and Cliococca selaginoides. Karyotype number is not refl ective of phylogenetic relationships among Linum species. For example, an analysis based on RAPD data indicate that L. decumbens (2n = 30) is clustered with L. grandifl orum (2n = 16), not with other species that share the same chromosome number (e.g., L. angustifolium and L. usitatissimum; Fu et al., 2002). Linum perenne group can be easily distinguishable from other Linum species morphologically (Ockendon, 1968), but the molecular study of Vromans (2006) indicate that classifi cation among the L. perenne group is still complicated. Neither L. perenne nor L. austriacum form a specifi c group, even though L. austriacum is considered a member of L. perenne group (Diederichsen, 2007) and they have the same haploid karyotype number of nine (Nagao, 1941; Gill, 1987). Additional molecular studies have focused on withinspecies variation of L. usitatissimum. Mansby et al. (2000) used isozyme markers to study the genetic diversity in fl ax and defi ned fi ve groups, but with low variation within the groups. An unexpectedly high genetic diversity within accessions led to the conclusion that the large heterozygosity found in L. usitatissimum may be the result of more outbreeding than earlier believed (Mansby et al., 2000). This fi nding was unexpected as fl ax is reported to be an obligate inbreeding species (Durrant, 1986). In a study on geographic patterns of fl ax variability, Fu (2005) pointed out that accessions from the East Asian and European regions were most diverse, whereas accessions from the Indian subcontinent and Africa were the most distinct. Overall, comparatively more variation existed in landraces than cultivars. Considerable diff erence within and among the four groups of cultivated fl ax cultivars was observed in quantitative traits; however, RAPD and two qualitative characters did not show marked diff erences (Diederichsen and Fu, 2006). A molecular study comparing fi ber and oil fl ax indicated that fi ber cultivars have a narrower and more homogenous genetic base than oil cultivars (Fu et al., 2002). Vromans’ (2006) AFLP study supports this fi nding and he further speculated that linseed cultivars and a wild relative L. bienne could be important sources for the introduction of favorable traits to fi ber fl ax. VARIABILITY IN CHROMOSOME NUMBERS Karyotypic analysis of Linum species began more than a half century ago, which has allowed several species to be recognized and diff erentiated (Tutin et al., 1968). The genus R e p ro d u c e d fr o m C ro p S c ie n c e . P u b lis h e d b y C ro p S c ie n c e S o c ie ty o f A m e ri c a . A ll c o p y ri g h ts re s e rv e d . 828 WWW.CROPS.ORG CROP SCIENCE, VOL. 48, MAY–JUNE 2008 Table 1. Comparison of various groupings of Linum species based on chromosome numbers, including only those species for which cytological information is available. Kikuchi (1929) Ray (1944) Nagao (1941) Gill (1987) Group I (n = 9) L. altaicum Fisch. Group I (n = 8) L. grandifl orum Desf. Group I (n = 8) L. grandifl orum Desf. Group I (2n = 18) L. alpinum Jacq. (2n = 18,36) L. austriacum L. L. hirsutum L. L. altaicum Ldb. L. extraaxillare Kit. L. anglicum Mill. L. hologynum Reichb. L. austriacum L. L. lewisii Pursh L. grandifl orum Desf. (2n = 16,18) L. muilleri Moris. L. tenuifolium L. L. narbonense L. L. hologynum Reichb. L. perenne L. L. julicum Hayek. L. sibiricum DC L. lewisii Pursh. L. narbonense L. (2n = 18,28)