| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Genetics and Molecular Biology |
2Mount Auburn Cemetery, 580 Mount Auburn Street, Cambridge, Massachusetts 01890 USA; 3Royal Botanic Garden Edinburgh, 20A Inverleith Row, Edinburgh EH3 5LR, Scotland, UK
Received for publication May 14, 2002. Accepted for publication September 17, 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: chloroplast inheritance • cpDNA • hybrid origins • RAPDs • species separation • Taxaceae • Taxus x hunnewelliana • Taxus x media
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Long-standing questions of placing the order Taxales in the gymnosperms and the position of the family Taxaceae, as it relates with other families of conifers, have been clarified by recent molecular studies (Chaw et al., 1993
; Cheng et al., 2000
). Yet species delimitation remains a problem, and progress in phytochemical and pharmacological studies has suffered as a result (Appendino, 1995
). Since the classification of Pilger (1903)
, which recognized only a single species with seven geographical subspecies, many other authors have proposed alternative schemes. Their descriptions of the various taxa have often presented overlapping (and sometimes contradictory) criteria. One recent classification suggested recognizing 25 different species (Spjut, 2000
). Phenotypic plasticity throughout the species of Taxus (variability of a given taxon across a range of environmental conditions), a varying range of leaf and stem characteristics observable on a single plant, and a poor understanding of the differences between juvenile and adult morphologies, seem to account for much of the confusion. Although recent molecular phylogenetic analyses using internal transcribed spacer (ITS) sequences suggested some level of genetic diversity (Li et al., 2001
), the degree of genotypic distinctness that might exist among the various species, according to current species delimitation, has not yet been determined. In the early 1900s, a series of hybridization events was reported to have occurred in the United States between three commonly recognized species, Taxus baccata L. (the European yew), T. cuspidata Siebold & Zucc. (the Japanese yew), and T. canadensis Marsh. (the Canadian yew). These species are among the less controversial taxa and can be assigned distinct ranges with reasonable confidence based on collections and surveys conducted over the past century.
The reputed hybridizations between Taxus baccata and T. cuspidata, which produced T. x media Rehder, and between T. cuspidata and T. canadensis, which produced T. x hunnewelliana Rehder, were not deliberate. In the two initial crosses, at the Hunnewell Pinetum near Boston, Massachusetts, USA, and concurrently at the Hicks Nursery on Long Island, New York, USA, the maternal/paternal identities of the parents are only partially known. Pertinent information obtained from the few available sources (Hatfield, 1921
; Chadwick and Keen, 1976
; Cochran, 1992
), although limited, allow us to ascertain the maternal parentage as follows: the original Taxus x hunnewelliana seedlings were grown from seed collected from T. canadensis; the original T. x media seedlings from the Boston area were grown from seed of T. baccata ‘Fastigiata’ (the Irish yew); and the original seedlings of T. x media from the Long Island area were grown from seed of T. cuspidata.
Attempts to characterize the hybrids morphologically have been difficult. The remarks of the individuals who first grew the plants, and of Alfred Rehder, who published the first descriptions (Rehder, 1923
), all suggest a wide range of variability among the seedlings. The offspring apparently exhibited a range of characters that mixed alternately one or the other parent's characteristics in the many stem and leaf characters, growth habits, reproductive strategies, and cold hardiness. In some cases, a heterosis effect could be observed in these characters, and recent phytochemical research has noted that some hybrid cultivars possess higher taxane content than some species sampled (van Rozendaal et al., 1999
, 2000
), as well as higher cyanogenic activity (Appendino, 1995
). Two forms of Taxus x hunnewelliana are in the trade, one resembling T. canadensis more than T. cuspidata, the other being just the opposite (Chadwick and Keen, 1976
). As for the Taxus x media group, an array of different forms exists in the more than 150 cultivars that have been selected and which dominate the nursery trade in America. Because of the widespread planting of these, it is possible that they have given rise to the exotic yews reported as naturalizing in the northeastern United States (Hils, 1993
), as the seeds are easily dispersed by birds. Occasionally, one finds mention of cultivars that do not differ significantly from another, or even more interestingly, that a cultivar of one of the parent species, Taxus cuspidata ‘Thayerae’, resembles the T. x media group more than the species (den Ouden and Boom, 1965
).
In this paper, we address the questions of species distinctness in Taxus baccata, T. canadensis, and T. cuspidata and attempt to discover reliable means of identifying Taxus species, hybrids, and cultivars through molecular analyses. By also assessing the origins and parentage of the putative hybrids between these species, we have an opportunity to test whether assumptions about chloroplast inheritance in the genus (potentially useful as a taxonomic tool) can be verified.
Our investigations employed two data sets and several analytical approaches. First, a broad-spectrum genomic analysis using RAPDs examined a sampling of representative individuals from the parent species and hybrids. The use of RAPDs in this study was chosen because it is a sensitive and relatively fast method of detecting genetic variation based on the polymerase chain reaction (PCR) (Williams et al., 1990
). It has limitations, including the reproducibility of bands attributable to reaction conditions (Pérez, Albornoz, and Domínguez, 1998
). But if it is used carefully, its results are reliable. Many effective applications of the technique have been used to distinguish intraspecific genetic variation and the detection of hybrids or clones (for examples see Wolff and Peters-van Rijn, 1993
; Hollingsworth et al., 1998
; Herbert et al., 2002
).
In addition to RAPDs, chloroplast markers were sought to clarify the variation found between species, and to corroborate directions of hybridization. Based on ultrastructural analyses, the direction of chloroplast inheritance in Taxus was assumed to be paternal, as it is in other conifers (Pennell and Bell, 1988
; Mogensen, 1996
; Anderson and Owens, 1999
). Information from the molecular analyses of these organelles alone is not sufficient to prove hybridization, but when taxa are suspected of involvement with hybrids, the combined data from these different approaches can provide powerful insights (Isoda et al., 2000
).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
DNA extraction
Total genomic DNA was isolated from approximately 150 mg of leaf tissue that had been stored in liquid nitrogen. The protocol used was based on the Doyle and Doyle (1987)
cetyltrimethyl ammonium bromide (CTAB) micropreparation method, modified by adding an ammonium acetate wash (Weising et al., 1995
) for further purification. The DNA was dissolved in TE (10 mmol/L Tris-HCl, pH 8.0, 1 mmol/L EDTA [ethylenediaminetetraacetic acid]) to a final concentration of 2–4 ng/µL.
RAPDs PCR
All 25 µL PCRs were conducted using a Perkin Elmer 9600 machine (Perkin Elmer, Applied Biosystems Division, Foster City, California, USA), and the RAPD PCR followed the method of Hollingsworth et al. (1998)
exactly except for the addition of 0.5 µL formamide (100%) per reaction.
Out of 35 primers (Operon Technology, Alameda, California, USA) tested on representative parent species samples, 10 were selected and used in the analysis of the full set of samples (Table 1). Selected primers showed optimal banding patterns with clearly distinguishable bands, with polymorphisms evident among the different parent species. Electrophoresis was conducted for 2 h at 120 V, using 1.6% agarose gels (in 1x tris-borate-EDTA [TBE] buffer) with 5 µL of PCR product from each sample plus 3 µL loading dye, including a 1-kilobase (kb) pair ladder (Gibco BRL, Rockville, Maryland, USA) as a reference standard to facilitate calculations of fragment size. Staining was achieved with 1.5 µL ethidium bromide (10 ng/mL) added to the gel.
|
), and, as a precaution, the data matrix was subjected to an analysis excluding its 27 weakest bands to see whether the results were affected. Data were organized in a binary matrix, with 1 indicating band presence and 0 indicating band absence. This matrix was analyzed using the R-Package for Multivariate and Spatial Analysis suite of programs (Casgrain and Legendre, developers; http://www.fas.umontreal.ca/biol/casgrain/en/labo/R/index.html). Initially, a similarity matrix was generated (version R 4.0) using the Jaccard index [Sj = a/(a + b + c), where a represents bands common to both samples X and Y, b equals bands unique to sample X, and c equals bands unique to sample Y; Jaccard, 1912] and converted to a distance index (Dj = 1 – Sj). This was then represented in a scatter matrix using the Principal Coordinate module (version R 3.01). A further principal coordinates analysis (PCO) was run, after excluding the hybrids, to determine if the species' spatial relationships would be affected by their presence.
Clustering analyses were performed with PAUP*, version 4.0b10 (Swofford, 2002
), using the unweighted pair-group method algorithm (UPGMA). Distance settings used the standard distance (total character distance); the Nei-Li distance in PAUP is inappropriate for RAPD data because PAUP uses the formula 
= –(3/2)ln[(41/2r – 1)/3] from Nei and Li (1979)
, which gives a distance in number of substitutions per site in a restriction motif, assuming six-base pair cutters (r = 6) (J. Wilgenbusch, Florida State University, personal communication). Estimates of statistical support for the resulting clusters were obtained in a UPGMA bootstrap analysis over 10 000 replicates.
Pairwise genetic distances were obtained based on the Jaccard indices (see above) and genetic indices, such as pair-wise differences between all pairs both within the species and hybrid groups and between the species, were then calculated.
Chloroplast trnL-F region
The PCR reactions were set up for each of the parent species (T. baccata no. 3, T. canadensis no. 17, and T. cuspidata no. 10) and a negative control, for amplification of the trnL-F chloroplast intron-spacer region using the trn-c and trn-f primers (Taberlet et al., 1991
). The PCR products were obtained with the following 50-µL reaction mix: 33.25 µL H2O, 5 µL buffer (as for RAPDs), 5 µL dNTPs (2 mmol/L), 2.5 µL MgCl2 (50 mmol/L), 1.5 µL primer each (10 µmol/L), 0.25 µL Taq polymerase (5 units/µL), and 1.0 µL of template DNA (2–4 ng/µL). The PCR conditions were as follows: (1) one cycle of 94°C for 1 min; (2) 35 cycles of 94°C for 45 s, 55°C for 45 s, 72°C for 3 min; (3) one cycle of 72°C for 10 min.
The PCR products were checked with electrophoresis on a 1% agarose gel, using a 1-kb ladder as size standard, and purified using Qiagen QIAquick Purification columns (Westburg BV, Leusden, Netherlands) according to the manufacturer's protocol. Sequencing was achieved using the Thermosequenase II dye terminator sequencing premix kit (Amersham Biosciences, Little Chalfont, Buckinghamshire, UK) according to the manufacturer's recommendations, on a Peltier Thermal Cycler PTC-200 (MJ Research, Watertown, Massachusetts, USA) using the trn-c, trn-d, and trn-f primers (Taberlet et al., 1991
). Sequencing products were analyzed on an ABI 377 sequencer (Perkin Elmer). The resulting alignment of sequences, including primer sites, yielded a fragment of 899 base pairs (bp). The sequences have been submitted to GenBank (T. baccata AY083113/AY083069; T. canadensis AF506837; T. cuspidata AF506838).
The three trnL-F sequences of the three parent species were analyzed with Web-Cutter (http://www.firstmarket.com/cutter/cut2.html) to identify appropriate cutting enzymes that could recognize species-specific restriction sites; there was a single cut site for XbaI (T/CTAGA, at positions 150–155 of the sequences) in T. baccata and T. cuspidata (T. canadensis had the sequence motif TATAGA at these positions), and a single cut site for VspI (AT/TAAT, at positions 558–563 of the sequences) in T. baccata only (T. canadensis and T. cuspidata sequences at these positions were ATTACT). Thus, Taxus canadensis could remain uncleaved. Taxus baccata could be cleaved at two sites, leaving fragment lengths of 150, 409, and 340 bp, while T. cuspidata could be cleaved once, leaving fragment lengths of 150 and 749 bp.
For restriction for each sample, the following digestion mix was prepared: 6.7 µL of H2O, 0.5 µL of VspI (10 units/µL), 0.5 µL of XbaI (10 units/µL) (both enzymes from Promega Corporation, Southampton, UK), 0.3 µL of acetylated bovine serum albumin (BSA). Ten microliters of the digestion mix was added to 20 µL of PCR product (at 2–4 ng/µL concentration) from each sample. The solutions were then incubated for 2 h at 37°C in a PE 9600 PCR machine. Ten microliters of the digest was analyzed on a 3% agarose gel containing 4 µL ethidium bromide (10 ng/mL), at 120 V for 3 h. Size calibration was facilitated by a 1-kb size standard.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The analysis showed the lowest variation within species among the samples of Taxus canadensis with 20.6% of their bands occurring as polymorphic. The highest number of polymorphic bands was seen in the Taxus baccata group (86.5% polymorphism), with T. cuspidata in between (76.8%). An even higher level of band diversity was detected in the hybrid group T. x media (94.9%). By contrast, the other hybrid group, T. x hunnewelliana, showed much less variation (49%).
As might be expected from the low number of polymorphic bands, Taxus canadensis had the lowest pair-wise difference (0.114). Taxus baccata showed the highest species value (0.510), followed by T. cuspidata (0.451). Taxus x hunnewelliana had a value intermediate between its putative parents (0.255), while the diversity within the T. x media group was similar to T. baccata (0.517). When calculating the diversity per band position, similar relationships between the groups were found (Table 2).
|
Species-specific bands
While many bands showed a high frequency of occurrence within a group, 21 appeared solely in a given species and in every sample within that species. These often (but not always) appeared in the hybrids as well. An indication of the frequency with which these markers were found in the hybrid groups can be seen in Table 3. Four of the five Taxus baccata species-specific bands occurred within the T. x media group, in a range of samples (from 19 to 96%). Of the six Taxus cuspidata species-specific bands (excluding T. cuspidata ‘Thayerae’), five were found in the T. x media group (from 4 to 58%). All six appeared in the T. x hunnewelliana group (mostly in 100% of the samples). Similarly, six of the 10 markers for Taxus canadensis appeared in the T. x hunnewelliana group, with a high level of group uniformity (100% in four of the six bands). While none of the Taxus baccata bands, as expected, were present in T. x hunnewelliana, one species-specific band from T. canadensis (OPP-20-1309) appeared in six samples of the T. x media group.
|
Principal coordinates analysis
Relationships between species and hybrids, based on the principal coordinates analysis, were visualized best as a two-dimensional graph (Fig. 1). A decision to use only the first two factors in the PCO, which contained the clear majority of variance (36.9%), would satisfy the Kaiser criterion (Kaiser, 1960
). In this analysis the organization of the species and hybrid groups into distinct clusters and the relative positions of hybrid groups was clearly seen. This clear separation was most noticeable in the triangular arrangement of the Taxus baccata, T. cuspidata, and T. x media groups. The same formation seemed repeated on a smaller scale in the Taxus cuspidate, T. canadensis, and T. x hunnewelliana groups. When the same analysis was run with the hybrid groups excluded, and when a reanalysis was run with the weakest bands removed, the species maintained both the separation of clusters and their triangular orientation (data not shown).
|
The UPGMA cluster analysis
As in the PCO, the relationships seen in the UPGMA dendrogram showed the clear separation of the three parent species and two groups of hybrids (Fig. 2). Within Taxus baccata, bootstrap values were low, with the species cluster supported by a bootstrap value (BS) of 71%. The Taxus baccata group clustered strongly (93% BS) with the T. x media hybrid group (itself supported as a cluster with a moderate 71% BS). Similarly, a close connection could be seen between the species Taxus canadensis (a highly supported cluster with 100% BS) and its hybrid group T. x hunnewelliana (albeit not supported by a bootstrap value). Sister to this same cluster was Taxus cuspidata, the other suspected parent in the T. x hunnewelliana hybrids.
|
Within the Taxus x media group, 10 of the samples paired together in well-supported clusters (Fig. 2). Yet the duplicate samples of the same cultivar designation (Taxus x media ‘Hicksii’ [nos. 33 and 47 ], T. x media ‘Hillii’ [nos. 34 and 48 ], and T. x media ‘Flushing’ [nos. 31 and 44]) did not pair together. The reanalysis with the weak bands omitted gave identical topologies (data not shown).
Chloroplast trnL-F region
Sequencing of the chloroplast trnL-F region in the three parent species indicated 16 polymorphic sites (1.9%) differentiating the species. The maximum sequence divergence, found between Taxus baccata and T. canadensis, was 1.44% while between both T. baccata and T. cuspidata and T. canadensis and T. cuspidata, it was 1.10%. After the double digestion, the observed banding patterns showed consistent species-specific chloroplast markers present in the three species' sample groups, with one exception: the sample of Taxus cuspidata ‘Thayerae’ (no. 15) possessed a T. baccata group chloroplast type. In the Taxus x hunnewelliana hybrid group, one sample (no. 20) possessed a chloroplast type of the T. cuspidata species group, while the rest had the T. canadensis type. Among the samples of Taxus x media, all samples showed either one or the other chloroplast type from the T. baccata or T. cuspidata species groups (Table 3).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), subspecies level (dos Santos et al., 1994
), population level (Herbert et al., 2002
), and cultivar level (Wolff and Peters-van Rijn, 1993
). Sampling size limitations in the species populations have to be considered when interpreting our results, although the species samples used covered most of their respective geographical ranges. Furthermore, RAPD primer selection allows a limitation in hypervariable band positions that are likely to be saturated on the gel, reducing problems of homoplasies and scoring errors due to similar fragment length of non-homologous bands. These two factors have resulted in DNA fingerprints well representative of the individual Taxus species. This is particularly reflected in the clarity of definition and distinction among the groups sampled. The sensitivity of RAPDs in detecting variation at different levels has allowed a number of interesting observations regarding species, hybrids, and hybrid cultivars. The ability of the RAPD data to differentiate the same species and hybrid groups by the PCO and UPGMA analyses, and to show the same relative distance relationships between the species when the hybrids were excluded, supports the integrity of the data set. Of the 21 species-specific marker bands observed, eight (including at least two from each species) can be obtained from a single primer, OPA-7. The minimum number of primers needed to distinguish among the cultivars of Taxus x media included here is two (OPA-20, OPC-6). However, the reliability of these primers as a cultivar identification tool should be determined by testing additional cultivars.
Species recognition
The delimitation of the three species studied is well supported by the RAPD data. In addition to the species-specific banding observed, the differences between species were suggested by the data set through different analyses; both PCO and UPGMA portrayed groups with distinct genomes. Pollen fertility studies also support their independent species status; these indicate an impaired meiosis in the reputed hybrids, shown by a reduction in the formation of functional pollen and a lower pollen viability rate (Collins, 2001
; E. Anderson, University of Victoria, British Columbia, personal communication).
Chloroplast genome typing additionally supports the species' distinctness. The identification of three different chloroplast types suggests a long period of isolation and the development of separate species genotypes. To elucidate species relationships and migration patterns throughout the genus, more comprehensive species sampling is required.
The rather low diversity observed in Taxus canadensis, also detected in a population study by Senneville et al. (2001)
, may be due partly to the limited sampling from its distribution range and the fact that habitat availability is more restricted compared with the other species. It may be more affected, however, by the tendency of T. canadensis toward widespread clonal reproduction and monoecy, a reproductive strategy that is unique within Taxus (Wilson, Buonopane, and Allison, 1996
). Inbreeding potential could be considered significant, as no apparent self-incompatibility exists (Allison, 1993
). Taxus baccata and T. cuspidata showed considerably higher diversity than T. canadensis, in line with their dioecy and wider distributions.
Hybrid origins and status
The positions of the two reputed hybrid groups, as shown in the PCO, implied nearly equal distances from their respective parent species. The distinctness of each group was further borne out by their clustering in the UPGMA analysis. Thus, these hybrids represent crosses between distinct parental genotypes. The presence of the various species-specific marker bands in both Taxus x media and T. x hunnewelliana is evidence for the genetic contributions from the respective parental genomes. The chloroplast markers provide further evidence. Both hybrid groups contain samples that have either one or the other parental chloroplast type, suggesting that reciprocal crosses have occurred in both hybrid groups. An alternative explanation could be organelle leakage (Mogensen, 1996
), but, given the nearly equal frequencies of both parental cpDNA types in the T. x media group, reciprocal crosses are the more parsimonious explanation. The two types of Taxus x hunnewelliana in the trade, each resembling one parent species more than the other (Chadwick and Keen, 1976
), could be explained by the occurrence of reciprocal crosses. Of particular interest is the inheritance observed in Taxus x media ‘Hatfieldii’ (no. 46) and the two T. x media ‘Hicksii’ samples (nos. 33 and 47). According to the available records, which describe these very first Taxus x media hybrids, the seed collected by Theophilus Hatfield from T. baccata ‘Fastigiata’ and the seed collected by the Hicks Nursery from T. cuspidata specify the direction of the cross in each instance. By finding the T. cuspidata chloroplast type in T. x media ‘Hatfieldii’ and the T. baccata chloroplast type in both samples of T. x media ‘Hicksii’, we can verify that the chloroplast is paternally inherited, as predicted by Pennell and Bell (1988)
and Anderson and Owens (1999)
.
The surprising level of genetic diversity in the Taxus x media hybrids, as evidenced by the number of polymorphic RAPD bands and the mean pair-wise difference (both considerably higher than the values noted for T. canadensis), can only be explained by the many sexually produced nursery introductions mentioned in the literature (e.g., Chadwick and Keen, 1976
) and the fact that several different hybridization events have occurred over time.
The inability to find identical banding profiles in the duplicate cultivar samples (Taxus x media ‘Flushing’, ‘Hicksii’, and ‘Hillii’) can perhaps be attributed to the practice in some nurseries of introducing cultivars as seedling lots. Asexual production of cultivars was sometimes employed. In those cases where it was, we would expect to find identical genotypes in an RAPD analysis. Although that was not the case here, we only had three duplicate samples available. No two cultivar samples had identical PCR profiles, which could indicate the origin of cultivars asexually from existing ones. This may be explained by the fact that we sampled only about 10% of the Taxus x media cultivars in the trade. A larger sampling in both cases stated above would probably uncover such cases of genetic identity.
For those samples in the Taxus x media group possessing bands not seen in either parent species, including the T. canadensis marker band OPP-20-1309, the limited parent species sampling might account for the absence. If a larger genotypic representation of Taxus baccata or T. cuspidata were studied, the origin of those bands might be better explained. One explanation could be that the band found in the T. x media group represents a length, but not sequence, homoplasy with that found in Taxus canadensis.
Yet another possibility that could explain the diversity in the Taxus x media hybrids is the likelihood of having F2 generation or backcross hybrids, which would not be surprising given the casual production practices followed by some of the nurseries. In these cases, RAPD analysis would be unable to distinguish with certainty between F1 and F2 hybrids. However, the possibility of F2 hybrids might explain the unusual position of Taxus x media ‘Green Mountain’ (no. 45) in the PCO, where it was drawn slightly in the direction of T. baccata.
Finally, given the widespread planting of Taxus x media cultivars in the northeastern United States, it is possible that a hybrid would backcross with one of the parent species (which are also grown in the area). Again, it is unlikely that RAPDs could distinguish these cases, except by a somewhat removed spatial distance on the PCO analysis. Instead, it would be useful to examine the available organelle markers that could indicate parental inheritance. In Taxus x media ‘Green Mountain’ (no. 45), which might represent such a backcross, its T. baccata chloroplast type prevents us from doing this, since that chloroplast type is common among the T. x media group.
The equally removed position in the PCO of Taxus cuspidata ‘Thayerae’ (no. 15) from the other T. cuspidata samples and its separate clustering in the UPGMA analysis make this sample another candidate for a backcross, since it possesses a Taxus baccata chloroplast type, yet is not among the T. x media hybrids. Wilson (1926)
and Rehder (1931)
described this introduction as arising from seedlings of a Taxus cuspidata plant on the Thayer Estate near Boston. The pollen parent is again unknown, but this sample's position near the T. cuspidata group suggests a backcross with a T. x media plant possessing the T. baccata chloroplast type. Our analysis of the sample named T. x media ‘Thayerae’ in this study (no. 40), which behaved in every way like the T. x media group and which likely originated from a T. baccata pollen parent, sheds light on a perplexing taxonomic question. The names Taxus x media ‘Thayerae’ and T. cuspidata ‘Thayerae’ both appear in many published treatments of Taxus species and hybrids. Despite the fact that applying the same cultivar epithet ‘Thayerae’ to two different taxa in the same genus violates current nomenclature rules (ICNCP Art. 26.1: Trehane et al., 1995
), the situation may have been caused by the dual origins of a genetically mixed (two paternal parents) seedling lot.
Conclusions
The degree of species segregation suggested by our data implies a level of genetic diversity greater than might be anticipated for a genus once regarded as monotypic (with geographic subspecies). Phenotypic plasticity influenced by habitat as well as seasonal effect on phenetic characters have no doubt played a role in past taxonomic difficulties. Yet despite the fact that Taxus presents challenges of identification in the field and herbarium, the distinct genotypes found here cannot be ignored. The clear differences noted between the three species studied indicate that the current species delimitation is well founded, especially in the light of the characteristics in the hybrids derived from them.
Our confidence in the abilities of the molecular techniques used here to identify Taxus species, hybrids, and cultivars (of sexual or asexual origin) was increased by using more than one technique. The sensitivity of RAPDs, as seen in both the PCO and UPGMA analyses, was complemented by data from the cpDNA analysis. Together, these techniques were capable of discerning much information about the status and origins of the samples that were studied.
It is hoped that further studies will benefit from the results presented here and that a comprehensive analysis of the entire genus, which could unravel evolutionary species relationships and migration patterns, will explore the genetic diversity that appears to exist between species of Taxus.
|
|
FOOTNOTES |
|---|
4 Author for reprint requests (telephone: 617-547-7105 x261; d.collins{at}mtauburn.com
)
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Anderson E. D. J. N. Owens 1999 Megagametophyte development, fertilization, and cytoplasmic inheritance in Taxus brevifolia. International Journal of Plant Sciences 160: 459-469[CrossRef]
Appendino G. 1995 The phytochemistry of the yew tree. Natural Products Reports 12: 349-360
Chadwick L. C. R. A. Keen 1976 A study of the genus Taxus. Research Bulletin 1086. Ohio Agricultural Research and Development Center, Wooster, Ohio, USA
Chaw S.-M. H. Long B.-S. Wang A. Sharkikh W.-H. Li 1993 The phylogenetic position of Taxaceae based on 18S rRNA sequences. Journal of Molecular Evolution 37: 624-630[Web of Science][Medline]
Cheng Y. R. G. Nicolson K. Tripp S.-M. Chaw 2000 Phylogeny of Taxaceae and Cephalotaxaceae genera inferred from chloroplast matK gene and nuclear rDNA ITS region. Molecular Phylogenetics and Evolution 14: 353-365[CrossRef][Web of Science][Medline]
Cochran K. D. 1992 A history of yews in the United States. In R. C. Hansen [ed.], Taxus and Taxol—a compilation of research findings. Special Circular 150-99. Ohio State University Extension Research, Ohio Agricultural Research and Development Center, Columbus, Ohio, USA
Collins D. J. 2001 An examination of the species Taxus baccata, T. cuspidata and T. canadensis, and their reputed hybrids. Master's thesis, University of Edinburgh and Royal Botanic Garden Edinburgh, Edinburgh, UK
Dempsey D. I. Hook 2000 Yew (Taxus) species—chemical and morphological variations. Pharmaceutical Biology 38: 274-280
den Ouden P. B. K. Boom 1965 Manual of cultivated conifers. Martinus Nijhoff, The Hague, Netherlands
dos Santos J. B. J. Nienhuis P. Skroch J. Tivang M. K. Slocum 1994 Comparison of RAPD and RFLP genetic markers in determining genetic similarity among Brassica oleracea L. genotypes. Theoretical and Applied Genetics 87: 909-915[Web of Science]
Doyle J. J. J. L. Doyle 1987 A rapid DNA isolation procedure for small quantities from fresh leaf tissue. Phytochemistry Bulletin 19: 11-15
Hatfield T. D. 1921 Raising yews from seed at Wellesley: study of relative hardiness of European and Asiatic types. Garden Magazine 33: 23-26
Herbert J. P. M. Hollingsworth M. F. Gardner R. R. Mill P. I. Thomas T. Jaffré 2002 Conservation genetics and phylogenetics of New Caledonian Retrophyllum (Podocarpaceae) species. New Zealand Journal of Botany 40: 175-188
Hils M. H. 1993 Taxaceae Gray: yew family. In Flora of North America Editorial Committee [eds.], Flora of North America north of Mexico, vol. 2, Pteridophytes and gymnosperms, 423–427. Oxford University Press, New York, New York, USA
Hollingsworth M. L. P. M. Hollingsworth G. I. Jenkins J. P. Bailey C. Ferris 1998 The use of molecular markers to study patterns of genotypic diversity in some invasive alien Fallopias (Polygonaceae). Molecular Ecology 7: 1681-1691[CrossRef]
Isoda K. S. Shirasi S. Watanabe K. Kitamura 2000 Molecular evidence of natural hybridization between Abies veitchii and A. homolepis (Pinaceae) revealed by chloroplast, mitochondrial and nuclear DNA markers. Molecular Ecology 9: 1965-1974[CrossRef][Medline]
Jaccard P. 1912 The distribution of flora in the alpine zone. New Phytologist 11: 37-50[CrossRef]
Kaiser H. F. 1960 The application of electronic computers to factor analysis. Educational and Psychological Measurement 20: 141-151[CrossRef][Web of Science]
Li J. C. C. Davis P. Del Tredici M. J. Donoghue 2001 Phylogeny and biogeography of Taxus (Taxaceae) inferred from sequences of the internal transcribed spacer region of nuclear ribosomal DNA. Harvard Papers in Botany 6: 267-274
Mogensen H. L. 1996 The hows and whys of cytoplasmic inheritance. American Journal of Botany 83: 383-404[CrossRef][Web of Science]
Nei M. W.-H. Li 1979 Mathematical model for studying genetic variation in terms of restriction endonucleases. Proceedings of the National Academy of Sciences, USA 76: 5269-5273
Pennell R. I. P. R. Bell 1988 Insemination of the archegonium and fertilization in Taxus baccata L. Journal of Cell Science 89: 551-559
Pérez T. J. Albornoz A. Domínguez 1998 An evaluation of RAPD fragment reproducibility and nature. Molecular Ecology 7: 1347-1357[CrossRef][Medline]
Pilger R. 1903 Taxaceae. In A. Engler and K. Prantl [eds.], Das Pflanzenreich, 18 (IV.5): 1–124. W. Engelmann, Leipzig and Berlin, Germany
Rehder A. 1923 Taxus media hybrid. Journal of the Arnold Arboretum 4: 106-109
Rehder A. 1931 Taxus cuspidata Sieb. & Zucc. f. Thayerae Wilson. Journal of the Arnold Arboretum 12: 59.
Senneville S. J. Beaulieu G. Daoust M. Deslauriers J. Bousquet 2001 Evidence for low genetic diversity and metapopulation structure in Canada yew (Taxus canadensis): considerations for conservation. Canadian Journal of Forest Research 31: 110-116[CrossRef]
Spjut R. W. 2000 Key to the species of Taxus in North America. World Botanical Associates website, URL: http://www.worldbotanical.com/TAXNA.HTM
Swofford D. L. 2002 PAUP*: phylogenetic analysis using parsimony (*and other methods), version 4. Sinauer Associates, Sunderland, Massachusetts, USA
Taberlet P. L. Gielly G. Pautou J. Bouvet 1991 Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17: 1105-1109[CrossRef][Web of Science][Medline]
Trehane P. C. D. Brickell B. R. Baum W. L. A. Hetterscheid A. C. Leslie J. Mcneill S. A. Spongberg F. Vrugtman 1995 International code of nomenclature for cultivated plants. Quarterjack Publishing, Wimborne, Dorset, UK
van Rozendaal E. L. M. S. J. L. Kurstjens T. A. van Beek R. G. van den Berg 1999 Chemotaxonomy of Taxus. Phytochemistry (Oxford) 52: 427-433[CrossRef][Web of Science]
van Rozendaal E. L. M. G. P. Lelyveld T. A. van Beek 2000 Screening of the needles of different yew species and cultivars for paclitaxel and related taxoids. Phytochemistry (Oxford) 53: 383-389[CrossRef][Web of Science][Medline]
Weising K. H. Nybom K. Wolff W. Meyer 1995 DNA fingerprinting in plants and fungi. CRC Press, Boca Raton, Florida, USA
Williams J. G. K. A. R. Kubelic K. J. Livak J. A. Rafalski S. V. Tingey 1990 DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Research 18: 6531-6535
Wilson E. H. 1926 Some yews and low-growing conifers. House and Garden 49: 82-83,150,154.
Wilson P. M. Buonopane T. D. Allison 1996 Reproductive biology of the monoecious clonal shrub Taxus canadensis. Bulletin of the Torrey Botanical Club 123: 7-15[CrossRef][Web of Science]
Wolff K. J. Peters-van Rijn 1993 Rapid detection of genetic variability in chrysanthemum (Dendranthema grandiflora Tzvelev) using random primers. Heredity 71: 335-341
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
