a Department of Plant Taxonomy, Wageningen Agricultural University, P.O. Box 8010,6700 ED Wageningen, The Netherlands b Centre for Plant Breeding and Reproduction Research (CPRO-DLO), P.O. Box 16, 6700 AA Wageningen, The Netherlands
MATERIALS AND METHODS
Key Words: Asteraceae Cichorium gene pool Lactuca Lactuceae phylogenetic relationships ribosomal DNA internal transcribed spacer (ITS) Taraxacum
MATERIALS AND METHODS
An economically important member of the genus Lactuca is the cultivated lettuce (Lactuca sativa), while the closely related species L. serriola, L. saligna, and L. virosa are important genitors for lettuce breeding. In recent years, the use of improved breeding techniques has extended the lettuce gene pool to L. tatarica (Chupeau et al., 1994; Maisonneuve et al., 1995). In order to make an accurate choice of the most promising wild species to further broaden the gene pool, more insight is needed into the taxonomic relationships of Lactuca. Therefore, a study was undertaken to evaluate the various Lactuca classifications and to elucidate the link between species relationships and the possibilities of sexual and somatic hybridization.
Three major generic divisions for Lactuca are those of Stebbins (1937), Tuisl (1968), and Feráková (1977). Stebbins (1937) applies a broad genus definition and includes Mulgedium Cass., Lactucopsis Schultz-Bip. ex Vis. et Panc., Phaenixopus Cass., Mycelis Cass., and part of Cicerbita Wallr. Tuisl (1968) takes the genus in a narrow sense, separating the genera Mulgedium, Scariola F.W. Schmidt (= Phaenixopus Cass.), Cicerbita, Cephalorrhynchus Boiss., and Steptorhamphus Bunge. Feráková (1977) takes an intermediate position including Mulgedium, Lactucopsis, and Phaenixopus/Scariola in Lactuca as sections Mulgedium (Cass.) C.B. Clarke, Lactucopsis (Schultz-Bip. ex Vis. et Panc.) Rouy., and Phaenixopus (Cass.) Benth., respectively. Feráková (1977) regards Mycelis, Cicerbita, Steptorhamphus, and Cephalorrhynchus as separate genera.
In order to evaluate the relationships of Lactuca s.l., a number of less closely related genera were included in our research, viz. Prenanthes L., Chondrilla L., Taraxacum Weber, Sonchus L., and Cichorium L. Stebbins (1953) considers Prenanthes to be related to Lactuca s.l., forming part of the so called Prenanthes-Lactuca line. Chondrilla, Taraxacum, and Sonchus are generally considered to be more distantly related to Lactuca, while the affinities of Cichorium are unclear (see, e.g., Bremer, 1994). In this paper the subgeneric classification of Feráková (1977) and the generic classification of Bremer (1994) will be used as a starting point (Table 1).
To facilitate the evaluation of Lactuca relationships and classifications, we focused on five research topics: (1) the disputed distinction of the species L. sativa, L. serriola, L. dregeana, and L. aculeata; (2) the assumed intermediate position of L. altaica between L. serriola and L. saligna (Feráková, 1977); (3) the position of L. sativa/L. serriola, L. saligna, and L. virosa relative to each other (see Koopman and De Jong, 1996); (4) the boundaries of the genus Lactuca and its subgeneric division as proposed by Feráková (1977) in relation to the recognition of the genera Mycelis, Cicerbita, and Steptorhamphus; and (5) the taxonomic position of Cichorium (Blackmore, 1981; Bremer, 1994) in relation to the monophyly of the Prenanthes-Lactuca line (Stebbins, 1953). These research topics are discussed in relation to various Lactuca classifications and morphological, crossability, cytological, isozyme, and molecular data from the literature. In conclusion, an adjusted genus concept for Lactuca is proposed based on the ITS-1 results and the subdivision of Feráková (1977). This genus concept is discussed in relation to the gene-pool concept of Harlan and De Wet (1971).
|MATERIALS AND METHODS|
MATERIALS AND METHODS
The ITS-1 was amplified in two steps. In the first step the entire ITS region, including ITS1 and 2 plus the interjacent 5.8S rRNA gene, was amplified using the primers "ITS5" and "ITS4" from White et al. (1990). In each PCR (polymerase chain reaction), 10 ng of nuclear DNA were used in a total volume of 25 µL, containing 10 mmol/L Tris-HCl pH 9.0, 50 mmol/L KCl, 1.5 mmol/L MgCl2, 0.01% (w/v) gelatin, 0.1% Triton X-100, 100 µmol/L of each dNTP, 50 ng of each primer, and 0.15 units polymerase (Super Trouper, HT Biotechnology Ltd., UK). Template DNA was denatured for 3 min at 94°C, followed by 30 cycles of 45 s at 94°C, 45 s at 55°C, and 1.45 min at 72°C on a Hybaid thermal cycler. Final extension consisted of 3 min at 72°C. In the second step, ITS-1 was amplified with the primers "ITS 5" and "ITS 2" from White et al. (1990) using 40 µL of a 1000 times dilution of the entire ITS-PCR product in a total volume of 100 µL. Again, the template was denatured for 3 min at 94°C, this time followed by 15 cycles of 45 s at 94°C, 45 s at 60°C, and 1.45 min at 72°C. The final extension consisted of 3 min at 72°C. The amplified product was purified for sequencing by cutting out the ITS-1 band after electrophoresis on a 2% agarose gel in TBE buffer. DNA was extracted from the agarose by electro-elution, followed by phenol/chloroform extraction and ethanol precipitation according to standard procedures (Sambrook, Fritsch, and Maniatis, 1989). Sequencing of ITS-1 sequences was performed with the Promega Silver Sequence(TM) DNA Sequencing System (Promega, Madison, Wisconsin, USA), according to the manufacturer's instructions, for part of the accessions. The ITS-1 was sequenced in both directions with the above-mentioned primers "ITS5" and "ITS2" using an annealing temperature of 55°C and 60 cycles on a Perkin-Elmer Thermal Cycler 480. Sequencing reactions were run on a 6% polyacrylamide gel. After silver staining, the gels were recorded on Kodak Duplicating RA 1 film. For another part of the accessions, ITS-1 sequencing was performed with the Applied Biosystems DyeDeoxy Terminator Cycle Sequencing Kit containing the Amplitaq FS DNA polymerase (Perkin-Elmer Applied Biosystems, Foster City, California, USA). Again, ITS-1 was sequenced in both directions with the primers "ITS 5" and "ITS 2" as above. Sequences were run on an Applied Biosystems 373. The few accessions sequenced with both methods gave identical results. All sequences obtained were deposited in the EMBL, GenBank, and DDBJ Nucleotide Sequence Databases under the accession numbers GBANAJ228605 to GBANAJ228661 (Table 3).
Data sets and outgroup selection
Two data sets were analyzed. (1) One set contained all different sequences found in this study, supplemented with GenBank sequences GBANL13957 (L. sativa), GBANL48143 (L. perennis), GBANL48151 (P. purpurea), GBANL48301 (S. asper), and GBANL48337 (T. officinale). Sequences that were found to be identical among different accessions were entered only once in this data set. In the case of identical sequences among accessions, one accession was chosen arbitrarily to represent the sequence. The selected accessions and the corresponding accessions with identical sequences are listed in Table 2. The ingroup species were either members of Lactuceae subtribe Lactucinae or, in the case of Cichorium intybus, unassigned to a subtribe (Bremer, 1994). Three species from subtribes related to Lactucinae were used as outgroup species: T. officinale and C. juncea (Lactuceae subtribe Crepidinae) and S. asper (Lactuceae subtribe Sonchinae). Since this data set was too large to perform a branch and bound search, it was analyzed using a heuristic search. As a result, not all most parsimonious trees were found. (2) A smaller data set was compiled containing consensus sequences of the subtribe Lactucinae species only. The consensus sequences were obtained by merging all sequences of a species found in our study using the ambiguity codes of the NC-IUB for both ambiguous and variable positions. Additionally, all gaps present in any of the accessions of a species were introduced into its consensus sequence. In this consensus data set, P. purpurea was the subtribe Lactucinae species most distantly related to Lactuca (Stebbins, 1953) and therefore it was used as outgroup. The resulting data set was small enough to enable a branch and bound search yielding all possible most parsimonious trees.
Bootstrap values were calculated in 1000 replications of a heuristic search with ACCTRAN, multistate taxa interpreted as uncertainty (first data set) or polymorphisms (second data set), collapse of zero-length branches, simple taxon addition, and TBR without MULPARS. The amount of phylogenetic signal in the data sets was determined from the tree-length distribution of 100 000 random trees (multistate taxa interpreted as uncertainty) using the g1 statistic (Hillis and Huelsenbeck, 1992). Sequence divergence values between species were calculated in PHYLIP 3.572 (Felsenstein, 1993) with DNADIST and the Kimura two-parameter method. The transition/transversion ratios were calculated in MacClade 3.04 (Maddison and Maddison, 1993) as the average ratio across 100 most parsimonious trees for the first data set and as the average ratio across all 17 most parsimonious trees for the second data set. A neighbor-joining tree (based on the sequence divergence values) and a maximum-likelihood tree (using the empirical base frequencies) were calculated in PHYLIP 3.572 for both data sets. A distance matrix containing both absolute and mean pairwise distances between the accessions was generated with PAUP (Table 3).
MATERIALS AND METHODS
First data set
The total aligned length in the data set containing all sequences was 269 bp, including 166 variable sites, 142 of which were phylogenetically informative. The amount of phylogenetic signal was highly significant: the g1 of -0.75 is considerably lower than the critical value of -0.09 (P= 0.01; Hillis and Huelsenbeck, 1992). In the 1200 replications with random taxon addition, 878 shortest trees of 434 steps, a CI of 0.57 and a RI of 0.85 were found. The search with simple taxon addition and MULPARS yielded 900 trees (the extent of the tree buffer) of 434 steps, a CI of 0.57, a RI of 0.85, and an average transition/transversion ratio across the first 100 trees of 1.26.
In all trees generated, the species L. sativa, L. serriola, L. dregeana, and L. altaica form a clade with highly similar sequences. The L. aculeata sequence is distinguishable but only slightly different from that of the species in this clade. The sequences of all L. sativa accessions in our data set were identical to each other, to that of both L. dregeana accessions, to that of six of the eight L. serriola accessions and to one of the two L. altaica accessions. The L. serriola and L. altaica sequences that differed from the joint L. sativa/serriola/dregeana/altaica sequence deviated in only one or two positions, while the L. sativa GenBank accession GBANL13957 differed in four positions (Table 3).
The various accessions of the species outside the sativa/serriola/dregeana/altaica clade form clearly separated basal clades containing accessions of one species only. Apparently, the intraspecific variation within these species is small compared to the interspecific variation, which makes the ITS-1 a good character to identify them.
One of the most parsimonious simple taxon addition trees is shown in Fig. 1. The topology of this tree is identical to that of the strict consensus of all 900 shortest simple taxon addition trees generated, except for its intraspecific branches that often form polytomies in the strict consensus tree (not shown). The topologies of the neighbor-joining tree and the maximum-likelihood tree were comparable with that of the simple taxon addition tree in the main aspects. All trees show a "L. sativa" clade containing L. sativa, L. serriola, L. dregeana, and L. altaica, with L. aculeata as a sister group of this "L. sativa" clade. These five species are part of a larger clade that also includes L. saligna and L. virosa and thus contains all subsection Lactuca species examined. This "subsection Lactuca" clade is part of a larger "Lactuca" clade containing all Lactuca species except for the subsection Cyanicae species L. perennis and L. tenerrima and the Asiatic species L. indica. Within this "Lactuca" clade, all trees contain the "subsect. Lactuca" clade, L. quercina, and a clade containing L. tatarica, L. sibirica, and L. viminea. Apart from the "Lactuca" clade, all trees contain a clade including L. perennis, C. plumieri, L. tenerrima, and S. tuberosus, with L. perennis as a sister group of C. plumieri, and L. tenerrima as a sister group of S. tuberosus. A separate clade that is present in all trees consists of M. muralis and C. alpina. The species C. intybus, P. purpurea, T. officinale, C. juncea, and S. asper branch off in the same order in all the trees. However, in the neighbor-joining and the maximum-likelihood trees P. purpurea and T. officinale branch off as sister groups and not sequentially. The minor differences between the various trees mostly originate in differences in sister-group relationships. (1) In the neighbor-joining tree L. saligna and L. virosa are sister groups, while the L. saligna/L. virosa clade is a sister group of the clade containing L. sativa. In both the simple taxon-addition consensus tree and the maximum-likelihood tree, L. virosa and the L. sativa clade are sister groups (see Fig. 1). (2) L. sibirica and L. tatarica are sister groups in both the neighbor-joining tree and the maximum-likelihood tree, while L. sibirica and L. viminea are sister groups in the simple taxon-addition consensus tree. (3) The clade containing L. perennis, C. plumieri, L. tenerrima and S. tuberosus is a sister group of the clade containing M. muralis and C. alpina in both the neighbor-joining and the maximum-likelihood trees, while the larger clade containing these six species is a sister group of a clade containing all other Lactuca species. These sister-group relationships are not present in the simple taxonaddition consensus tree. (4) Prenanthes and Taraxacum are sister groups in both the neighbor-joining and the maximum-likelihood trees, but not in the simple taxon-addition consensus tree. (5) The position of L. indica is different in the various trees: in the neighbor joining tree it is a sister group of L. perennis/C. plumieri, in the maximum-likelihood tree of M. muralis/C. alpina, and in the simple taxon-addition tree of the large Lactuca clade.
Second data set
The total aligned length in the data set with subtribe Lactucinae species was 258 bp, with 118 variable sites, 57 of which were phylogenetically informative. The g1 statistic of -0.52 was highly significant (<-0.20, P = 0.01), indicating the presence of sufficient phylogenetic signal. With multistate taxa interpreted as uncertainty, 17 trees of 203 steps were obtained with a CI of 0.70 and a RI of 0.64. A set of trees with identical topology, but a length of 255 steps, a CI of 0.76, a RI of 0.64, and an average transition/transversion ratio of 1.43 was obtained with multistate taxa interpreted as polytomies. The 50% majority rule consensus of the 17 trees is shown in Fig. 2.
MATERIALS AND METHODS
The work of Kesseli and Michelmore (1986) was excluded from the discussion since the uncertain identities of L. saligna, L. dregeana, and two of the three L. virosa accessions in their data set hamper an unambiguous interpretation of the results in terms of species relationships.
Distinction among L. sativa, L. serriola, L. dregeana, and L. aculeata
The extremely close relationship between L. sativa and L. serriola is apparent in the good interspecific crossability (e.g., Thompson, Whitaker, and Kosar, 1941; Lindqvist, 1960; De Vries, 1990), the identical karyotype (Lindqvist, 1960; Chatterjee and Sharma, 1969; Globerson, Netzer, and Sacks, 1980; Haque and Godward, 1985; Koopman and De Jong, 1996), chromosome banding pattern (Koopman, De Jong, and De Vries, 1993), and DNA content (Koopman and De Jong, 1996). Furthermore, in phenetic analyses of nuclear RFLP data (Kesseli, Ochoa, and Michelmore, 1991) and nuclear AFLP data (Hill et al., 1996) L. sativa and L. serriola clustered together. RFLP analysis of mtDNA showed a high proportion of shared fragments (Vermeulen et al., 1994), isozyme analysis of foliar esterases shows patterns common to both species (Roux, Chengjiu, and Roux, 1985), and a phenetic analysis of SDS-electrophoresis patterns of seed proteins (De Vries, 1996) showed L. sativa and L. serriola as completely interlaced groups. Numerical morphological analyses of plant morphological data showed L. sativa and L. serriola as separate but partly overlapping groups (De Vries and Van Raamsdonk, 1994; Frietema de Vries, Van der Meijden, and Brandenburg, 1994; Frietema de Vries, 1996). While De Vries and Van Raamsdonk (1994) maintain L. sativa and L. serriola as separate species, Frietema de Vries (1996) considers the overlap enough to regard them as conspecific. Using the ITS-1 sequences, L. sativa and L. serriola cannot be distinguished since six out of eight L. serriola sequences were identical to the single sequence that was characteristic to all L. sativa accessions in our study. Taking into account this large overlap in ITS-1 sequence and that of characters as shown in the literature cited, it would seem most appropriate to regard L. sativa as conspecific with L. serriola.
The close relationship of L. dregeana with L. serriola is indicated by its morphology (Zohary, 1991) and the fact that L. serriola x L. dregeana crosses (Koopman, unpublished data) yielded large numbers of plump seeds. The ITS-1 sequence data from our study corroborate this relationship since the sequences of L. dregeana, L. sativa, and most of the L. serriola accessions were identical. Therefore, L. dregeana should be considered conspecific with L. serriola as well.
The position of L. aculeata close to L. serriola is clear from the plant morphology (Zohary, 1991), crossability (Globerson, Netzer, and Sacks, 1980; Koopman, unpublished data), chromosome morphology (Globerson, Netzer, and Sacks, 1980), and isozyme analyses of foliar esterases (Roux, Chengjiu, and Roux, 1985). The ITS-1 sequence of L. aculeata differed only slightly from those of L. sativa and L. serriola, confirming the close relationship between L. aculeata and L. sativa/serriola. Further information will be needed to determine the relationship more accurately.
The close relationship of L. sativa, L. serriola, L. dregeana, and L. aculeata (and L. altaica) as a group is stressed by their pairwise distances. The maximum distance within this group is four positions excluding the L. sativa GenBank accession and six positions including it. This is comparable with the largest intraspecific distances within the closely related species L. saligna and L. virosa, which are four and six, respectively (Table 3).
Position of L. altaica
Based on morphological characters, Feráková (1977) regarded L. altaica as a species intermediate between L. serriola and L. saligna. The close relationship with L. serriola is indicated by the results of crossing experiments of Thompson, Whitaker, and Kosar (1941) and Lindqvist (1960). The latter did not even distinguish L. altaica as a separate species but considered it a primitive form of L. sativa. Furthermore, in (unpublished) crossing experiments by the present first author, L. altaica behaved like L. serriola. Crosses between L. serriola and L. altaica yielded many normal seeds of good quality, while crosses between L. saligna and L. altaica yielded only small seeds of poor quality, comparable to seeds from saligna x serriola crosses. The ITS-1 results corroborate the close relationships between L. altaica and L. serriola, since the ITS-1 sequence of one L. altaica accession in our study was identical to that of the most common L. serriola sequence and that of the other deviated in only two autapomorphic positions. Except for some morphological characters, no evidence was found to support a close relationship to L. saligna. Therefore, L. altaica should be considered conspecific with L. serriola rather than intermediate between L. serriola and L. saligna.
Position of L. sativa/serriola, L. saligna, and L. virosa relative to each other
Four different possibilities regarding the position of L. saligna and L. virosa relative to each other and to L. sativa/serriola appear from literature. (1) Data on plant morphology (De Vries and Van Raamsdonk, 1994), SDS-electrophoresis patterns of seed proteins (De Vries, 1996), and isozyme analysis of foliar esterases (Roux, Chengjiu, and Roux, 1985) indicate a close relationship between L. sativa/serriola and L. virosa, while L. saligna is the more distinct species. (2) Data on crossability (Thompson, Whitaker, and Kosar, 1941; Lindqvist, 1960; De Vries, 1990), karyotype (Lindqvist, 1960; Koopman and De Jong, 1996), chromosome banding pattern (Koopman, De Jong, and De Vries, 1993), and DNA content (Koopman and De Jong, 1996) indicate that L. sativa/L. serriola occupy an intermediate position between L. saligna and L. virosa, but closer to L. saligna than to L. virosa. In this case, L. virosa is the more distinct species. The position of L. sativa/serriola closer to L. saligna than to L. virosa is corroborated by data on nuclear AFLP(TM)s (Hill et al., 1996). (3) Hill et al. (1996) stated that their AFLP results were similar to previously published nuclear RFLP (restriction fragment length polymorphism) results of Kesseli, Ochoa, and Michelmore (1991). However, in contrast with the phenetic tree based on AFLPs, the RFLP tree shows that L. saligna and L. virosa are more related to each other (in terms of similarity) than to L. sativa/serriola. (4) A completely different indication of the species relationships within subsect. Lactuca was given by Vermeulen et al. (1994). In their analysis of mtDNA RFLPs, the proportion of bands shared by L. serriola and L. virosa was higher than that shared by L. serriola and L. sativa. This large proportion of identical bands would indicate that L. virosa is at least as related to L. serriola as is L. sativa, which is in conflict with all previously mentioned results.
In most of our ITS-1 based phylogenies, L. virosa is more closely related to L. sativa/serriola than is L. saligna, which corroborates the results of the isozyme and morphological analyses. However, the order in which L. virosa and L. saligna branch off in the parsimony tree is determined by only one synapomorphy (Fig. 1). In the neighbor-joining tree L. virosa and L. saligna are sister groups, which corroborates the RFLP results. Given these ambiguities, the ITS-1 results must be considered inconclusive as to the position of L. sativa/serriola, L. saligna, and L. virosa relative to each other.
The apparently conflicting indications of species relationships reported in literature can be brought in line by postulating (as far as we know for the first time) that L. virosa is a hybrid taxon. In that case, the mtDNA RFLPs, which are inherited maternally, can be interpreted as indicating a serriola-like species as a female parent to the hybrid. The contribution of a yet unknown male parent is expressed in two L. virosa characters that are unique within subsect. Lactuca, namely the winged black achene and the presence of one pair of satellite chromosomes instead of two.
In the new hybrid species, different classes of DNA evolved differently, as discussed in Koopman, De Jong, and De Vries (1993). Extensive chromosome rearrangements and dynamic changes of large blocks of repetitive DNA could give rise to the deviating L. virosa karyotype, chromosome banding pattern, DNA content, and crossability. The unique DNA sequences reflected in the RFLP and isozyme patterns, as well as in the plant morphology, evolved differently. The ITS-1 results can be explained by assuming that after the hybridization event the ITS-1 sequences derived from both parents were subjected to a homogenization process in the new hybrid (see Elder and Turner  for a discussion on homogenization). The homogenization gave rise to a new ITS-1 sequence characteristic for the hybrid species, while the original parental sequences were lost.
Position of Cichorium
The taxonomic position of Cichorium within the Lactuceae is unclear because important characters such as pappus type and pollen morphology are not phylogenetically interpretable. On the one hand, Cichorium has a pollen type too widespread within the Lactuceae to be useful for clarifying genus relationships (Blackmore, 1981). On the other hand, Cichorium has a pappus type that is unique within the Lactuceae and thus useless for phylogenetic purposes as well (Bremer, 1994). According to Stebbins (1953) and Jeffrey (1966), Cichorium and Lactuca are not closely related. Stebbins (1953) recognizes eight subtribes within the Cichorieae and assigns Cichorium and Lactuca sensu Stebbins (1937) to different subtribes. Cichorium is placed in the subtribe Cichorinae, while Lactuca is part of the so called Prenanthes-Lactuca line of subtribe Crepidinae. Jeffrey (1966) recognizes five groups within the Cichorieae and although both Cichorium and Lactuca are placed in the same group, Cichorium is separated as the monogeneric Cichorium subgroup while Lactuca is placed in the broadly defined Crepis subgroup. Recently, Vermeulen et al. (1994) concluded from mitochondrial RFLP data that Cichorium is more closely related to Lactuca and Cicerbita than to Chondrilla, Taraxacum, and Sonchus. On the other hand, Whitton, Wallace, and Jansen (1995) found that using chloroplast DNA restriction site variation, Cichorium appears more closely related to Prenanthes, Chondrilla, and Taraxacum than to Lactuca and Cicerbita and least related to Sonchus. Our ITS-1 data on C. intybus support the view of Vermeulen et al. (1994) that Cichorium is closely related to Lactuca. According to the ITS-1 data, Cichorium is more related to Lactuca than is Prenanthes, which places Cichorium within Stebbins' Prenanthes-Lactuca line or Jeffrey's Crepis subgroup.
Evaluation of major generic concepts in Lactuca
According to Feráková (1977) two main generic delimitations of Lactuca are possible: taking the genus in a broad sense according to Stebbins (1937) or treating the genus in a narrow sense according to Tuisl (1968). Feráková (1977) herself takes an intermediate position. More recently, Shih (1988) published an extremely narrow genus concept.
The genus Lactuca according to Stebbins (1937) is characterized by a corolla tube generally more than half as long as the ligule, a pappus containing at least some bristles that are no more than four-celled in cross section at the base, and a flattened achene with an expanded pappus disc and two lateral ribs or wings more pronounced than the others. The inflorescences are many-headed panicles or racemes. Based on this description, Lactuca sensu Stebbins includes the genera Mulgedium, Lactucopsis, Phaenixopus, Mycelis, and Cicerbita, but Cicerbita alpina is excluded because of its coarse pappus and nearly columnar, only slightly compressed achene. Stebbins did not mention Steptorhamphus, but it probably should be included as well since it fits Stebbins' genus description of Lactuca in general terms, notwithstanding the relatively coarse pappus. Tuisl (1968) takes the genus Lactuca in a narrow sense and recognizes Mulgedium, Scariola (= Phaenixopus as recognized by Stebbins), Cicerbita, and Steptorhamphus as separate genera. Lactuca is characterized by two equal rows of pappus hairs, a distinct beak, and many-flowered heads. The genera Lactuca, Mulgedium, and Scariola are separated from Steptorhamphus and Cicerbita by the presence of two rows of equal pappus hairs. The genus Mycelis, which was not mentioned by Tuisl, can also be separated from Lactuca by this character. Lactuca is separated from Mulgedium and Scariola because of its distinct beak, while Scariola is separated from Mulgedium by its few-flowered heads. According to Tuisl's genus definition, Lactuca sibirica should be included in Mulgedium as well, although the species was not treated in Tuisl (1968). Tuisl's genus concept also necessitates the separation of Lactucopsis from Lactuca (Feráková, 1977). Feráková (1977) takes an intermediate position between Stebbins (1937) and Tuisl (1968). She limits the genus Lactuca to species with two equal rows of pappus hairs, thus excluding Steptorhamphus, Mycelis, and Cicerbita, but includes Tuisl's genera Mulgedium and Scariola as sections within Lactuca. As a result, Feráková (1977)'s genus Lactuca comprises the sections Phaenixopus (= Scariola as recognized by Tuisl), Mulgedium, Lactucopsis, and Lactuca, the latter of which is divided into subsections Lactuca and Cyanicae (see also Table 1). More recently, Shih (1988) narrowed down the genus Lactuca including only species with numerous yellow florets and pale brown achenes with a clearly distinct, filiform beak. This would limit the genus to species from Feráková's subsection Lactuca, but strictly speaking it would exclude L. virosa as well as L. sativa cultivars with white or blackish achenes.
Our ITS-1 phylogeny is in line with the view of Stebbins (1937) that Lactuca in a broad sense should include Mulgedium, Lactucopsis, Phaenixopus, Mycelis, and Cicerbita and that Lactuca can be separated from the closely related genus Prenanthes. However, the exclusion of C. alpina from Lactuca s.l. is not supported since C. alpina fell within the group of species from Lactuca sensu Stebbins. The ITS-1 data conflict with the subgeneric division of Tuisl (1968) on several points. The separation of Mulgedium (represented by L. tatarica and L. sibirica) from Scariola (= Phaenixopus, represented by L. viminea) is not fully supported since all trees generated showed a sibirica/tatarica/viminea clade (Figs. 1, 2) and part of the analyses show a smaller sibirica/viminea clade nested within it (see results section and Fig. 1). The distinction between Lactuca and Cicerbita/Steptorhamphus or between Cicerbita and Mycelis is not supported either, since species from the different genera form clades together (Figs. 1, 2). The ITS-1 data partly corroborate the genus concept of Feráková. On the one hand, a clade containing all subsect. Lactuca species is present in all ITS-1 analyses, as is a larger clade containing all Lactuca species except L. perennis and L. tenerrima. On the other hand, the distinction between sect. Mulgedium and sect. Phaenixopus is not confirmed, nor is the position of subsect. Cyanicae. The sect. Mulgedium species L. tatarica and L. sibirica form a clade with the sect. Phaenixopus species L. viminea, while the subsect. Cyanicae species L. perennis and L. tenerrima do not form a clade with Lactuca species but with C. plumieri and S. tuberosus, respectively. The ITS-1 data fit the narrow generic concept of Shih (1988) to a large extent, since this concept would lead to the recognition of separate genera for most of the species used in our study and thus avoids the classification problems. As such, our results would also fit a separation of L. tatarica and L. sibirica (Shih, 1988) and of C. alpina and C. plumieri (Stebbins, 1937; Shih, 1991). However, because the genus description of Shih would exclude L. virosa (and even part of L. sativa) from Lactuca it conflicts with the ITS-1 data showing a L. sativa/ serriola/ saligna/ virosa clade.
Delimitation of Lactuca
Regarding the ITS-1 results, there are several options for delimiting the genus Lactuca.
1) A large and variable genus could be recognized, approximately according to Stebbins (1937), but with the inclusion of Cicerbita alpina. This would "group together species similar in habit, and those whose individual affinities are clearly with each other rather than with species excluded from the genus" however, "occasional, transitional species are found" (Stebbins, 1937). These transitional species show morphological characters grouping them with Lactuca as well as morphological characters grouping them with other genera. In our data set, C. alpina is such a transitional species, since its general habit and ITS-1 sequence place it well within Lactuca sensu Stebbins, but its pappus and achene do not fit his genus description. Including C. alpina in Lactuca sensu Stebbins would necessitate an expanded genus description including species with a coarse pappus and nearly columnar achenes. This new genus description would probably obscure the boundaries between Lactuca s.l. and Prenanthes or even less related genera and the genus would become unacceptably variable.
2) A solution to this problem would be to recognize a narrow genus Lactuca, identical to Feráková's subsect. Lactuca. Depending on the species concept used, the number of species contributing to this genus can be reduced by lumping L. sativa and the L. serriola-like species L. serriola, L. aculeata, L. scarioloides Boiss., L. azerbaijanica Rech., L. georgica Grossh., L. dregeana, and L. altaica, described in Zohary (1991). This solution would also fit an expanded genus description of Shih allowing white or blackish achenes, but necessitates the recognition of many additional genera (see, e.g., Shih, 1988).
3) A third and in our opinion more favorable solution would be to recognize a genus Lactuca according to Feráková, but with the exclusion of subsect. Cyanicae. This would limit the genus to species with more than three ribs on the achenes. The adjusted genus concept has the benefit of corresponding to the gene pool of cultivated lettuce according to the gene-pool concept of Harlan and De Wet (1971), which facilitates its practical use and acceptance.
Lactuca and the lettuce gene pool
Within Feráková's Lactuca subsect. Lactuca a group of species can be identified that is closely related to and readily crossable with L. serriola and L. sativa, containing among others L. dregeana, L. aculeata, and L. altaica (Zohary, 1991). Since all these species have an ITS-1 sequence that is (nearly) identical to that of cultivated lettuce (L. sativa), this sequence can be considered characteristic for species contributing to the primary gene pool of lettuce (Harlan and De Wet, 1971). It is expected that the other members of the group, namely L. scarioloides, L. azerbaijanica, and L. georgica (Zohary, 1991) will show similar ITS-1 sequences and crossing behavior. The remaining subsect. Lactuca species L. saligna and L. virosa are characterized by their own distinct ITS-1 sequences and in the analyses they form a clade with the species from the primary genepool. In contrast to what is stated in Zohary (1991), L. saligna as well as L. virosa contribute to the secondary gene pool of cultivated lettuce, since both are partly interfertile with L. sativa (Lindqvist, 1960; Maisonneuve et al., 1995).
A group of species less related to cultivated lettuce (Lactuca sativa) branch off closely to the primary and secondary gene-pool species in ITS-1 based phylogenies. These are L. viminea from Feráková's section Phaenixopus, L. tatarica and L. sibirica from section Mulgedium, and L. quercina from section Lactucopsis. Hybridization data on these species are limited to L. viminea and L. tatarica. L. viminea is crossable with L. virosa (Groenwold, 1983) yielding a partly fertile hybrid, and L. tatarica can be somatically hybridized with L. sativa to produce a fertile hybrid (Chupeau et al., 1994; Maisonneuve et al., 1995). Since the genetic diversity of L. viminea and L. tatarica is not directly accessible for lettuce breeding but requires special techniques such as bridging species or somatic hybridization, they belong to the tertiary gene pool of L. sativa. Based on the species relationships proposed by Feráková (1977) and the ITS-1 results it can be expected that L. quercina and L. sibirica and the species from Lactuca sections Phaenixopus, Mulgedium, and Lactucopsis not included in our study, contribute to L. sativa's tertiary gene pool as well.
The species L. perennis and L. tenerrima from Lactuca subsect. Cyanicae (Feráková, 1977) do not form a clade with Lactuca species in the ITS-1 phylogenies, but with species from related genera outside the lettuce gene pool. This position outside the lettuce gene pool is corroborated by the fact that L. perennis is not crossable with subsect. Lactuca species (Thompson, Whitaker, and Kosar, 1941) and that the somatic hybrids between L. perennis and L. sativa reported on in Chupeau et al. (1994) and Maisonneuve et al. (1995), were completely sterile (B. Maisonneuve, INRA, France, personal communication). No literature data were available on the remaining subsect. Cyanicae species, but based on the relationships proposed by Feráková (1977), the ITS-1 results on L. tenerrima and L. perennis, and the literature data on L. perennis, it can be expected that the entire subsection falls outside the lettuce gene pool.
The limited data available on the remaining species used in our study do not indicate additional species that could contribute to the gene pool of cultivated lettuce. The tested species L. indica (Thompson, Whitaker, and Kosar, 1941) and S. tuberosus (= L. cretica Desf.; Thompson, 1943) were not crossable with any subsect. Lactuca species. Somatic hybridizations of L. sativa with L. indica (Mizutani et al., 1989) yielded colonies of hybrid callus, but no viable plants could be regenerated from these colonies. Somatic hybridizations between L. sativa and C. plumieri, C. juncea, C. intybus, and T. officinale completely failed (Chupeau et al., 1994; Maisonneuve et al., 1995). Therefore, the gene pool of cultivated lettuce seems limited to the Lactuca species sensu Feráková excluding subsect. Cyanicae, as indicated by the dotted line in Figs. 1 and 2. The position of L. indica relative to the species in the lettuce gene pool is still unclear since its position varied in the different phylogenetic analyses (see Results section and Figs. 1 and 2). However, because L. indica can be somatically hybridized with L. sativa to produce a viable callus, it is probably closely related to the species in the lettuce gene pool. Additional research, preferably involving other Asiatic species as well, should elucidate the position of L. indica.
Based on our present information, we propose an adjustment of the genus concept of Feráková (1977) implying exclusion of sect. Lactuca subsect. Cyanicae from Lactuca. This limits the genus to species with more than three ribs on the achenes. The adjusted genus coincides with the lettuce gene pool. Section Lactuca subsect. Lactuca comprises the primary and secondary gene pool, while the sections Phaenixopus, Mulgedium, and Lactucopsis comprise the tertiary gene pool. Section Lactuca subsect. Cyanicae is not included in Lactuca, nor does it belong to the lettuce gene pool. The position of L. indica needs further consideration.
The practical value of our study to plant breeders is that it points out the species contributing to cultivated lettuce's tertiary gene pool. These tertiary gene-pool species will become increasingly important in breeding programs since improved breeding techniques (Chupeau et al., 1994; Maisonneuve et al., 1995) will make them more easily accessible as a gene source.
The usefulness of ITS-1 sequences for phylogenetic analysis and evaluation of existing classifications in Lactuca and related genera has been demonstrated in this paper. However, regarding details as, for example, the position of L. sativa/serriola, L. saligna and L. virosa relative to each other, ITS-1 sequences were inconclusive and supplemental markers are needed to elucidate the relationships. Recently, AFLPs have become available as a tool for systematic studies of closely related species (Sharma, Knox, and Ellis, 1996; Huys et al., 1996; Keim et al., 1997; Kardolus, Van Eck, and Van den Berg, 1998). Although the relative contributions of restriction site variation and insertions/deletions to the variation sampled with AFLPs is not exactly known, AFLPs proved to be useful molecular markers for phylogenetic purposes. Since Hill et al. (1996) demonstrated that AFLPs can be applied in Lactuca, they seem suitable to add information to the ITS-1 phylogeny presented in this paper. Therefore, a phylogenetic analysis using AFLP data of all Lactuca accessions from the ITS-1 study will be carried out in the near future.
MATERIALS AND METHODS
Baldwin, B. G. 1992 Phylogenetic utility of the internal transcribed spacers of nuclear ribosomal DNA in plants: An example from the Compositae. Molecular Phylogenetics and Evolution 1: 316. [Medline]
. 1993 Molecular phylogenetics of Calycadenia (Compositae) based on ITS sequences of nuclear ribosomal DNA: chromosomal and morphological evolution reexamined. American Journal of Botany 80: 222238.
Bremer, K. 1994 Asteraceae: cladistics and classification. Timber Press, Portland, OR.
Blackmore, S. 1981 Palynology and intergeneric relationships in subtribe Hyoseridinae (Compositae: Lactuceae). Botanical Journal of the Linnean Society 82: 113.
Chatterjee, T., and A. K. Sharma. 1969 Cytotaxonomy of Cichorieae. Genetica 40: 577590.
Chupeau, M-C., B. Maisonneuve, Y. Bellec, and Y. Chupeau. 1994 A Lactuca universal hybridizer, and its use in creation of fertile interspecific somatic hybrids. Molecular and General Genetics 245: 139145.
Cronquist, A. 1985 History of generic concepts in the Compositae. Taxon 34(1): 610.
De Vries, I. M. 1990 Crossing experiments of lettuce cultivars and species (Lactuca sect. Lactuca, Compositae). Plant Systematics and Evolution 171: 233248.
. 1996 Characterization and identification of Lactuca sativa cultivars and wild relatives with SDS-electrophoresis (Lactuca sect. Lactuca, Compositae). Genetic Resources and Crop Evolution 43: 193202.
, and L. W. D. Van Raamsdonk. 1994 Numerical morphological analysis of lettuce cultivars and species (Lactuca sect. Lactuca, Asteraceae). Plant Systematics and Evolution 193:125141.
Elder, J. F., and B. J. Turner. 1995 Concerted evolution of repetitive DNA sequences in eukaryotes. Quarterly Review of Biology 70: 297320. [Medline]
Felsenstein, J. 1993 PHYLIP (Phylogeny Inference Package) version 3.572. Computer program distributed by the author, Department of Genetics, University of Washington, Seattle, WA.
Feráková, V. 1977 The genus Lactuca L. in Europe. Bratislava, Univerzita Komenského.
Frietema De Vries, F. T. 1996 Cultivated plants and the wild flora. Ph.D. dissertation, Rijksherbarium/Hortus Botanicus, Leiden, The Netherlands.
, R. Van der Meijden and W. A. Brandenburg. 1994 Botanical files on lettuce (Lactuca sativa). Gorteria supplement 2.
Globerson, D., D. Netzer, and J. Sacks. 1980 Wild lettuce as a source for improving cultivated lettuce. In J. W. Maxon Smith and F. A. Langton [eds.], Proceedings Eucarpia meeting on leafy vegetables, 8696. Glasshouse Crops Research Institute, Littlehampton.
Groenwold, R. 1983 Onderzoek van de relatie tussen Lactuca en Bremia Lactucae. Verslag van een voorlichtingsbijeenkomst voor slaveredelaars (vervolg). Zaadbelangen 37: 132.
Harlan, J. R., and J. M. J. De Wet. 1971 Towards a rational classification of cultivated plants. Taxon 20: 509517.
Haque M. Z., and M. B. E. Godward. 1985 Comparison between two genera, species and cultivars in Lactuceae. I. Karyotype analysis. Cytologia 50: 725738.
Hill, M., H. Witsenboer, M. Zabeau, P. Vos, R. Kesseli, and R. Michelmore. 1996 PCR-based fingerprinting using AFLPs as a tool for studying genetic relationships in Lactuca spp. Theoretical and Applied Genetics 93: 12021210.
Hillis, D. M., and J. P. Huelsenbeck. 1992 Signal, noise, and reliability in molecular phylogenetic analyses. Journal of Heredity 83: 189195.
Huys, G., R. Coopman, P. Janssen, and K. Kersters. 1996 High-resolution genotypic analysis of the genus Aeromonas by AFLP fingerprinting. International Journal of Systematic Bacteriology 46: 572580. [Abstract]
Iwatsuki, K., T. Yamazaki, D. E. Boufford, and H. Ohba. 1995 Flora of Japan, vol. IIIb. Kodansha Ltd., Tokyo.
Jeffrey, C. 1966 Notes on Compositae. I. The Cichorieae in east tropical Africa. Kew Bulletin 18: 427486.
Kardolus, J. P., H. J. Van Eck, and R. G. Van den Berg. 1998 The potential of AFLPs in biosystematics: a first application in Solanum taxonomy. Plant Systematics and Evolution 210: 87103.
Keim, P., A. Kalif, J. Schupp, K. Hill, S.E. Travis, K. Richmond, D.M. Adair, M. Hugh-Jones, C.R. Kuske, and P. Jackson. 1997 Molecular evolution and diversity in Bacillus anthracis as detected by amplified fragment length polymorphism markers. Journal of Bacteriology 179: 818824.
Kesseli, R. V., and R. W. Michelmore. 1986 Genetic variation and phylogenies detected from isozyme markers in species of Lactuca. Journal of Heredity 77: 324331.
, O. Ochoa, and R. W. Michelmore. 1991 Variation at RFLP loci in Lactuca spp. and origin of cultivated lettuce (L. sativa). Genome 34: 430436.
Kim, K.-J., and R. K. Jansen. 1994 Comparisons of phylogenetic hypotheses among different data sets in dwarf dandelions (Krigia, Asteraceae): additional information from internal transcribed spacer sequences of nuclear ribosomal DNA. Plant Systematics and Evolution 190: 157185.
Kim, S.-C., D. J. Crawford, J. Francisco-Ortega, and A. Santos-Guerra. 1996 A common origin for woody Sonchus and five related genera in the Macaronesian islands: Molecular evidence for extensive radiation. Proceedings of the National Academy of Sciences, USA 93: 77437748.
Koopman,W. J. M., and J. H. De Jong. 1996 A numerical analysis of karyotypes and DNA amounts in lettuce cultivars and species (Lactuca subsect. Lactuca, Compositae). Acta Botanica Neerlandica 45: 211222.
, , and I. M. De Vries. 1993 Chromosome banding patterns in lettuce species (Lactuca sect. Lactuca, Compositae). Plant Systematics and Evolution 185: 249257.
Lindqvist, K., 1960 Cytogenetic studies in the serriola group of Lactuca. Hereditas 46: 75151.
Maddison, W. P., and D. R. Maddison. 1993 MacClade version 3.04. Computer program distributed by Sinauer Associates, Sunderland, MA.
Maisonneuve, B., M.-C. Chupeau, Y. Bellec, and Y. Chupeau. 1995 Sexual and somatic hybridization in the genus Lactuca. Euphytica 85: 281285.
Mizutani, T., X. J. Liu, Y. Tashiro, S. Miyazaki, and K. Shimasaki. 1989 Plant regeneration and cell fusion of protoplasts from lettuce cultivars and related wild species in Japan. Bulletin of the Faculty of Agriculture, Saga University 67: 109118.
Roux, L., Z. Chengjiu, and Y. Roux. 1985 Caractérisation de Lactuca sativa L. et des espèces apparentées par électrofocalisation des estérases. Agronomie 5: 915921.
Sambrook , J., E. F. Fritsch, and T. Maniatis. 1989 Molecular cloning, a laboratory manual, 2d ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Sang, T., D. J. Crawford, S.-C. Kim, and T. F. Stuessy. 1994 Radiation of the endemic genus Dendroseris (Asteraceae) on the Juan Fernandez Islands: evidence from sequences of the ITS regions of nuclear ribosomal DNA. American Journal of Botany 81: 14941501. [ISI]
, , and T.F. Stuessy. 1995 ITS sequences and the phylogeny of the genus Robinsonia (Asteraceae). Systematic Botany 20: 5564.
Sharma, S. K., M. R. Knox, and T. H. N. Ellis. 1996 AFLP analysis of the diversity and phylogeny of Lens and its comparison with RAPD analysis. Theoretical and Applied Genetics 93:751758.
Shih, C. 1988 Revision of Lactuca L. and two new genera of tribe Lactuceae (Compositae) on the mainland of Asia. Acta Phytotaxonomica Sinica 26: 382393.
. 1991 On circumscription of the genus Cicerbita Wall., and two new genera of Compositae from Sino-Himalayan region. Acta Phytotaxonomica Sinica 29: 394417.
Stebbins, G. L. 1937 Critical notes on Lactuca and related genera. Journal of Botany: 1218.
. 1953 A new classification of the tribe Cichorieae, family Compositae. Madroño 12: 6581.
Susanna, A., N. Garcia Jacas, D. E. Soltis, and P. S. Soltis. 1995 Phylogenetic relationships in tribe Cardueae (Asteraceae) based on ITS sequences. American Journal of Botany 82: 10561068.
Swofford, D. L. 1993 PAUP: Phylogenetic Analysis Using Parsimony, Version 3.1.1. Computer program distributed by the Illinois Natural History Survey, Champaign, IL.
Thompson, R. C. 1943 Further studies on interspecific genetic relationships in Lactuca. Journal of Agricultural Research 66: 4148.
, T. W. Whitaker, and W. F. Kosar. 1941 Interspecific genetic relationships in Lactuca. Journal of Agricultural Research 63: 91107.
Tuisl, G. 1968 Der Verwandtschaftskreis der Gattung Lactuca L. im Iranischen Hochland und seinen Randgebieten. Annalen des Naturhistorischen Museums in Wien 72: 587638.
Tutin, T. G., V. H. Heywood, N. A. Burges, D. M. Moore, D. H. Valentine, S. M. Walters, and D. A. Webb [eds.]. 1976 Flora Europaea, vol. 4. Cambridge University Press, Cambridge.
Vermeulen, A., B. Desprez, D. Lancelin, and H. Bannerot. 1994 Relationships among Cichorium species and related genera as determined by analysis of mitochondrial RFLPs. Theoretical and Applied Genetics 88: 159166.
Vosman, B., P. Arens, W. Rus-Kortekaas, and M. J. M. Smulders. 1992 Identification of highly polymorphic DNA regions in tomato. Theoretical and Applied Genetics 85: 239244.
White, T. J., T. Bruns, S. Lee, and J. Taylor. 1990 Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In M. A. Innis, D. H. Gelfand, J. J. Sninsky, and T. J. White [eds.], PCR protocols, a guide to methods and applications, 315322. Academic Press, San Diego, CA.
Whitton, J., R. S. Wallace, and R. K. Jansen. 1995 Phylogenetic relationships and patterns of character change in the tribe Lactuceae (Asteraceae) based on chloroplast DNA restriction site variation. Canadian Journal of Botany 73: 10581073.
Zohary, D. 1991 The wild genetic resources of cultivated lettuce (Lactuca sativa L.) Euphytica 53: 3135.
This article has been cited by other articles:
W. J. M. Koopman, M. J. Zevenbergen, and R. G. Van den Berg
Species relationships in Lactuca s.l. (Lactuceae, Asteraceae) inferred from AFLP fingerprints
Am. J. Botany, October 1, 2001; 88(10): 1881 - 1887.
[Abstract] [Full Text] [PDF]
G. J. Schmidt and E. E. Schilling
Phylogeny and biogeography of Eupatorium (Asteraceae: Eupatorieae) based on nuclear ITS sequence data
Am. J. Botany, May 1, 2000; 87(5): 716 - 726.
[Abstract] [Full Text]
|HOME||HELP||FEEDBACK||SUBSCRIPTIONS||ARCHIVE||SEARCH||TABLE OF CONTENTS|