| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Systematics |
2Department of Botany, University of Wisconsin, Madison, Wisconsin 53706 USA; 3Institut für Pharmazie und Molekulare Biotechnologie (IPMB), Universität Heidelberg, Abt. Biologie, Im Neuenheimer Feld 364, D-69120 Heidelberg, Germany
Received for publication October 28, 2003. Accepted for publication March 12, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Lamiaceae • Mentheae • phylogenetics • rbcL • Rosmarinus • Salvia • stamen • trnL-F
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Himmelbaur and Stibal, 1933–1935
; Claßen-Bockhoff et al., 2003
). The assumption among previous researchers has been that this peculiar pollination mechanism has only evolved once within the Mentheae, and thus, Salvia is monophyletic.
|
|
|
study of Salvia separated the genus into 12 sections and remains the most widely accepted treatment of Salvia. No comprehensive treatment of the genus has been completed since then, despite the recognition of over 500 new species of Salvia. Many researchers have modified Bentham's (1876)
ultimate subgeneric arrangement (Briquet, 1897
; Stibal, 1934
, 1935
; Pobedimova, 1954
; Hruby, 1962
; El-Gazzar and Watson, 1968
), but no easily employable, clearly defined subgeneric arrangement exists (Hedge, 1974
). Many recent researchers have avoided the troublesome issue of Bentham's subgeneric groupings by describing "species-groups" or small (often monotypic) sections (Epling, 1939
; Hedge, 1974
, 1982a
, b
). Thus Salvia raises a number of important systematic and evolutionary questions. Is Salvia monophyletic? How are other genera of Mentheae related to Salvia? How many major lineages are contained within Salvia? What are the large-scale relationships of New World Salvia lineages? Where did the major lineages originate? We address these questions with a molecular systematic investigation of all major groups within Salvia and allied genera of Mentheae.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
. Earlier work investigating rbcL in the Lamiaceae (Kaufmann and Wink, 1994
; Wink and Kaufmann, 1996
) resulted in rbcL sequences being obtained for over 80 species of Nepetoideae. This project expands on that sampling and presents 51 new trnL-F sequences and a total of 127 rbcL sequences. Accessions, vouchers, and GenBank numbers are available in the Appendix (see Supplemental Data accompanying the online version of this article). The data matrix for the rbcL analysis consists of 127 taxa, and the combined analysis of rbcL and trnL-F consists of 55 taxa. Attempts were made to use the same DNA for the rbcL and trnL-F sequencing; however, this was not possible for 19 of the taxa sampled. In four cases in the combined analysis, the rbcL sequence was obtained from a different species than the trnL-F sequence (Thymus serpyllum [trnL-F]/T. vulgaris [rbcL], Agastache urticifolia [trnL-F]/A. foeniculum [rbcL], Marrubium supinum [trnL-F]/M. incanum [rbcL], Mentha spicata [trnL-F]/M. longifolia [rbcL]). Although combining data from different members of a genus is not ideal, the placement of the four genera relative their sister groups did not change in single region analyses. This, in concert with the fact that our goal was not defining the placement of these genera but defining the lineages of Salvia, we feel justifies our decision to combine these sequences. The sampling of taxa is concentrated in the subfamily Nepetoideae (112 species sampled, representing 22 genera), particularly tribe Mentheae (105 species sampled, representing 18 genera) and the genus Salvia (70 species sampled). The extensive sampling within Salvia was designed to sample across the morphological diversity of the genus. Verbena officinalis of the Verbenaceae was chosen as an outgroup.
Extractions, amplification, and sequencing
Total genomic DNA was extracted using the DNeasy Plant Mini kit (Qiagen, Valencia, California, USA). Extractions were made from fresh leaves, leaves dried in silica gel, leaves from herbarium material, and frozen leaves. Polymerase chain reaction (PCR) amplification and cycle sequencing followed the methods described elsewhere (Conti et al., 1996
; Givnish et al., 2000
). The PCR product was purified either with a QIAquick PCR purification kit (Qiagen) or with an AmPure PCR purification kit (Agencourt, Beverly, Massachusetts, USA). Sequenced products were precipitated in ethanol and sodium acetate to remove excess dye terminators or cleaned with CleanSEQ Sequencing Reaction Clean-up system (Agencourt). Contiguous alignments were edited using Sequencher vs. 3.0 (Gene Codes, Ann Arbor, Michigan, USA).
Sequences were aligned visually in SeAl v. 2.0a7 (Rambaut, 2001
). Indels in the trnL-F data set were coded using the guidelines of Baum et al. (1994)
. Regions of ambiguous alignment were excluded from the analyses.
Phylogenetic analysis
Phylogenetic relationships within Salvia and Mentheae were evaluated in a two-step approach. The first involved a 55-taxon data set (36 species of Salvia) using rbcL and trnL-F sequences. The combined data set was analyzed using maximum parsimony (MP), maximum likelihood (ML), and Bayesian analyses. The heuristic MP analysis (Fitch, 1971
) in PAUP* 4.0b10 (Swofford, 2002
) used 100 random addition sequences, with 10 trees held at each step during stepwise addition, and tree bisection and reconnection (TBR) branch swapping to explore the possibility of multiple islands of most parsimonious trees (Maddison, 1991
). To assess congruence between the rbcL and trnL-F data sets, 100 replicates of the partition homogeneity test (Farris et al., 1995
) were conducted using a full heuristic search, simple taxon addition, TBR branch swapping, and saving all most parsimonious trees. Although the partition homogeneity test has been criticized (Yoder et al., 2001
), the test has merit as a first assessment for congruence of data sets (Hipp et al., 2004
). Topological constraints based on previous systematic hypotheses were used in full heuristic searches with 100 random addition replicates to evaluate the number of extra steps required to force taxa together. Bootstrap (Felsenstein, 1985
) and Bremer support values (Bremer, 1988
; Donoghue et al., 1992
) were used to evaluate support for relationships within the resulting trees. Bootstrap values were obtained through an heuristic search on all characters, with 1000 replicates and 10 random addition sequences with TBR replicates with no more than 5000 trees saved per replicate. Bremer support values were obtained using reverse topological constraints, with 100 random addition replicates for each search, as suggested by Swofford (1993)
, and implemented by Baum et al. (1994)
. Maximum likelihood analyses were conducted on the individual and combined data sets as implemented in PAUP*. Optimality criteria were explored using Modeltest version 3.06 (Posada and Crandall, 1998
). Heuristic ML searches with TBR branch-swapping were conducted. Bayesian analyses in MrBayes version 3.0 (Huelsenbeck and Ronquist, 2001
) were done with 1 000 000 cycles, running four Markov chain Monte Carlo (mcmc) chains, a temperature of 0.2 with branch lengths saved, all substitution rates equal, and no among-site variation. Four separate Bayesian chains were conducted starting with a different random seed each time.
The second approach involved an expanded rbcL data set with 127 taxa (71 species of Salvia) using only MP. Because a full heuristic MP search was not feasible with the expanded rbcL data, an iterative approach was used to develop a strict consensus tree following methods of Catalán et al. (1997)
. An initial heuristic search was performed in the same manner as in the 55- taxa analysis, with each replicate limited to 1 min and the number of saved trees limited to 2000. A strict consensus tree was created from this analysis and loaded as a constraint in a second heuristic search and performed in the same way as the original (with no limit placed on number of trees saved), searching for equally or more parsimonious trees not compatible with the constraint. If trees of equal or lesser length than the previous maximum parsimony trees were found, a new strict consensus tree was created from those trees and the previous consensus tree and was then loaded as a constraint in the next iteration. This process was continued until no equally parsimonious trees incompatible with the previous strict consensus tree were found in 1000 random addition sequences, each limited to 2 min.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Analysis of trnL-trnF in the 55-taxon data set
The aligned length of the trnL-F data set was 1271 base pairs. With regions of ambiguous alignment or ambiguous sequences excluded, the total length of included characters was 914 base pairs. Of the 914 characters in the 55 taxa included in the combined analysis, 689 were constant, 109 variable characters were parsimony-uninformative, and 116 characters (12.7%) were potentially parsimony-informative. Fitch parsimony analysis of the trnL-F region of the 55 taxa included in the combined analysis found 1728 equally parsimonious trees of 319 steps (CI = 0.83, RI = 0.92, RC = 0.77). Twenty-two indel events were scored for the trnL-F data set, of which 15 were potentially parsimony-informative. In the combined analysis, indel events were not included, but are noted on Fig. 4. Analyses were also performed with the indel events included (Fig. 5). The analysis of trnL-F with indels included contained 936 characters of which 642 were constant, 147 variable characters were parsimony-uninformative, and 147 characters were potentially parsimony-informative. Fitch parsimony analysis of the trnL-F region of the 55 taxa found 180 equally parsimonious trees of 352 steps (CI = 0.83, RI = 0.92, RC = 0.76).
|
|
Combined rbcL and trnL-trnF analysis in the 55-taxon data set
The aligned length of the combined data set was 2771 base pairs. With regions of ambiguous alignment or ambiguous sequences excluded, the total length of included characters was 2285. Of those 2285 characters, 1868 were constant, 417 were variable, and 214 (9.4%) were potentially parsimony-informative. Fitch parsimony analysis of the combined trnL-F and rbcL data set of the 55 taxa included in the combined analysis found 180 equally parsimonious trees of 686 steps (CI = 0.69, RI = 0.84, RC = 0.58). Constraining Salvia to be monophyletic resulted in a shortest tree 19 steps longer than unconstrained trees. Constraining Salvia + Dorystaechas as monophyletic resulted in a shortest tree seven steps longer. Constraining Salvia + Dorystaechas + Rosmarinus + Perovskia to be monophyletic resulted in a tree two steps longer. Templeton tests rejected the trees produced from the search constraining all Salvia together, but did not reject trees produced from the other constraints (P > 0.05).
The topology of trees produced with indels included did not differ from trees produced from sequence data alone. The tree shown (Fig. 4) was produced without indels included. Indel events with a CI = 1.0 were mapped onto the combined phylogeny (Fig. 4).
The partition homogeneity test of the two data sets suggests significant incongruity between the rbcL and trnL-F partitions compared to random partitions of the same size (P < 0.01). The most significant source of incongruence between the two data sets is the placement of a Mentha/Origanum/Thymus clade (hereafter refered to as the Mentha clade). The different positions of the Mentha clade are well-supported by each data set, respectively (trnL-F bootstrap = 100%; rbcL bootstrap = 75%). In the trnL-F analysis, this clade is placed in an unresolved polytomy including Lepechinia, Horminum, Rosmarinus, Perovskia, Salvia, Dorystaechas, Agastache, and Glechoma (Fig. 5). In the 55-taxa data set, the rbcL analysis places the Mentha clade sister to Salvia clade I. However, in the rbcL analysis Mentha itself has a relatively long branch length (10 steps under ACCTRAN), and long branch attraction may well be responsible for the peculiar and incongruent placement of the Mentha clade in the rbcL analysis. Despite data set incongruence, well-supported clades are generally supported by both regions (Fig. 4): Salvia clade I (bootstrap = 100%), Dorystaechas + Salvia clade II (bootstrap = 100%), Salvia subg. Calosphace (bootstrap = 100%), Salvia clade III (bootstrap = 90%), tribe Mentheae (bootstrap = 99%), and subfamily Nepetoideae (bootstrap = 100%). Each of the above clades are also maintained in the expanded rbcL analysis (Fig. 6).
|
Analysis of rbcL in 127-taxa data set
In the expanded rbcL analysis including 127 taxa, 1063 characters were constant, 308 were variable, and 185 (13.5%) were potentially parsimony-informative. Fitch parsimony analysis of the rbcL region for the 127 taxa included in the expanded rbcL analysis found trees of 705 steps (CI = 0.50, RI = 0.83, RC = 0.41). Although, as expected with rbcL, resolution within the Salvia clades is quite limited, the integrity of the major lineages (Salvia clades I, II, III) suggested by the combined analysis and by trnL-F alone are maintained in this expanded sampling (Fig. 6). Constraining Salvia to be monophyletic resulted in a shortest tree five steps longer than the most parsimonious unconstrained tree. Constraining Salvia + Dorystaechas as monophyletic resulted in a most parsimonious tree three steps longer. Constraining Salvia + Dorystaechas + Rosmarinus + Perovskia to be monophyletic resulted in a tree four steps longer. Templeton tests failed to significantly reject the constraint trees (P > 0.05).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Relationships within Mentheae
In view of the polyphyletic nature of Salvia as evidenced by these molecular results, understanding the evolution of Salvia requires a broader perspective within the family. Although considerable molecular phylogenetic work has examined the relationships within Lamiales and the placement of Lamiaceae (e.g., Olmstead et al., 1992
, 1993
, 2000
, 2001
; Wagstaff and Olmstead, 1997
) and relationships within some mint genera (Clerodendrum, Steane et al., 1997
, 1999
; Monarda, Prather et al., 2002
; Satureja, Cantino and Wagstaff, 1998
; Sideritis, Barber et al., 2002
; Teucrium, El Oualidi et al., 1999
), comparatively little has been done to examine relationships among genera in the mint family (Cantino and Sanders, 1986
; Wagstaff et al., 1995
, 1998
; Wink and Kaufmann, 1996
). Salvia and 122 other genera form the subfamily Nepetoideae, the best supported of the eight subfamilies of the Lamiaceae (Wagstaff et al., 1998
). The Nepetoideae is defined by hexacolpate pollen, an investing embryo, myxospermy, presence of rosmarinic acid, cpDNA restriction sites, and DNA sequence data (Cole, 1992
; Wagstaff, 1992
; Fig. 4).
The treatments of relationships within the subfamily Nepetoideae have varied considerably between researchers, ranging from 18 tribes (Wunderlich, 1967
) to four (Cantino et al., 1992
; Wagstaff et al., 1995
). Based on cpDNA restriction site analyses, Wagstaff et al. (1995)
recognized four monophyletic lineages within the Nepetoideae: tribe Lavanduloideae (consisting of Lavandula), tribe Ocimeae (defined by declinate stamens and consisting of 52 genera), tribe Elsholtzieae (defined by spreading stamens and consisting of six genera), and the tribe Mentheae, consisting of Salvia and 72 other genera. A monophyletic Mentheae, as suggested by Wagstaff et al. (1995)
, is supported by our analyses with trnL-F and rbcL.
In his treatment of Nepetoideae for Genera Plantarum, Bentham (1876)
placed the genus Salvia within his tribe Monardeae, defined by two ascending stamens. He recognized three ad hoc groups within Monardeae based on staminal structure: Meriandreae (Perovskia, Dorystaechas, and Meriandra), Salvieae (Salvia, Salviastrum [now included within Salvia], Audibertia [now included within Salvia], and Rosmarinus), and Monardeae (Monarda, Blephilia, and Ziziphora). The two thecae of each stamen of Meriandreae are parallel, those of Monardeae are divergent, and those of Salvieae are separated by an elongate connective. It should be noted that the unsampled Zhumeria has been suggested as a fourth genus in Meriandreae based on its staminal structure, pubescence, and corolla (Bokhari and Hedge, 1976
). Recent molecular investigations (Wagstaff et al., 1995
; Prather et al., 2002
) suggest that Bentham's subtribe Monardeae is not closely related to the subtribes Meriandreae and Salvieae. However, the relationships of the genera in the Meriandreae and Salvieae have been a source of debate in the literature. Although numerical analyses support an expanded Meriandreae to include Rosmarinus and Salvia (El Gazzar and Watson, 1970
), anatomical and morphological investigations (Bokhari and Hedge, 1971
) call into question the monophyly of an expanded Meriandreae (sensu Bentham, 1876
). Because of the conflict surrounding tribal relationships within the Nepetoideae and the non-monophyly of Salvia suggested by our molecular data (Figs. 4–6), any attempt to address the phylogenetics of and evolution within Salvia must be done at the next higher but clearly monophyletic level—the tribe Mentheae (sensu Cantino et al., 1992
). An important aspect will be a thorough reevaluation of subtribal relationships within the Mentheae (sensu Cantino et al., 1992
) in light of a well-supported molecular phylogeny.
The nature of rbcL makes it a suboptimal choice at the subgeneric and of limited use at the subfamilial level in the Lamiaceae. The slowly evolving nature of the region, in concert with the large sample size, imparts little chance of high support being obtained for clades. However, owing to the large number of sequences previously obtained for the rbcL in the Lamiaceae, we felt compelled to complete the sampling to represent all putative lineages of Salvia to investigate whether, even with a suboptimal and slowly evolving region, multiple lineages of Salvia were supported. The rbcL consensus tree (Fig. 6) supports the existence of the same three clades of Salvia identified by trnL-F (Fig. 5).
Staminal structure in Salvia
Central to any investigation into the phylogenetics, species radiation, or pollination biology of Salvia is the structure, ontogeny, and function of its peculiar stamens. Staminal morphology is the major defining character of Salvia, as well as being integral to the current subgeneric organization of the genus. Our preliminary research investigating the chloroplast regions rbcL and trnL-trnF demonstrates a polyphyletic Salvia and thus multiple origins of this staminal structure (Fig. 7).
|
Although the elongation of the connective into a lever mechanism may seem an exceedingly unlikely event to have occurred independently at least twice and probably three times in the Mentheae, initial morphological investigations suggest the lever mechanism found in Salvia clade I is in fact not homologous to the lever mechanism found in Salvia clade II (Fig. 7). In all of Salvia clade I, the aborted (or expressed) posterior theca is at the distal end of the connective. In Salvia clade II (particularly subg. Calosphace), it seems likely that the aborted posterior theca is expressed as a small projection next to the attachment with the filament. The actual lever in Salvia clade II would then consist of an extension of tissue on top of the connective, rather than an extension of the connective that is seen in Salvia clade I. Issues of pollination biology and biomechanics of the Salvia flower are currently being investigated in the laboratory of Regine Claßen-Bockhoff (Claßen-Bockhoff et al., 2003
), and future integration of this work with our phylogenetic results will be instrumental in understanding the ecological and functional aspects of staminal morphology in Salvia.
Phylogenetics of Salvia clade I
Salvia clade I appears to include all members of Bentham's subg. Sclarea and Leonia and those members of subg. Salvia not included in Salvia clade III. The eight species of New World Salvia belonging to Salvia clade I include four species of sect. Heterosphace (Walker and Elisens, 2001
), three species in subg. Salviastrum, and Salvia penstemonoides Kunth et Bouche. The three species of subg. Salviastrum, all native to Texas and northern Mexico, were originally placed by Scheele (1849)
in a separate genus. Although the three species of Salviastrum express the elongate connective diagnostic of Salvia, Scheele placed them in their own genus based on a dense annulus in the calyx. Torrey (1859)
soon thereafter noted the close affinities of Salviastrum and Salvia sect. Heterosphace and included Salviastrum as a section within Salvia. Bentham (1876)
, however, agreed with Scheele and maintained Salviastrum as its own genus. The most recent treatment of Salviastrum (Whitehouse, 1949
, p. 153) agrees with Torrey's treatment and suggests, "except for the dense ring of hairs in the calyx throat ... there are no common differences which will separate them from other species of Salvia. If the species included in Bentham's section Heterosphace are included in Salvia, then undoubtedly the Salviastrum section should be included, for they are closely linked by their similar calices which are alike in form and accrescence." The molecular data, identifying a monophyletic lineage consisting of New World Heterosphace + Salviastrum, support the suggestions of Whitehouse.
Salvia penstemonoides is a narrowly endemic species native to south-central Texas, thought to be extinct until its rediscovery in 1987 (Clebsch, 1997
). Originally described as belonging to Bentham's sect. Eusphace (Kunth and Bouche, 1848
), its rbcL sequence is exactly the same as S. texana (but differs at one site with trnL-trnF) and indicates its affinities with sect. Salviastrum. Its calyx (although lacking an annulus) and leaf morphology are more similar to Salviastrum than any other New World Salvia. Mitotic chromosome counts by the senior author (data not shown) totaled 2n = 28 for S. penstemonoides. The only other New World species with known counts of n = 14 or 2n = 28 in Salvia are the New World Heterosphace (Walker and Elisens, 2001
). If Salviastrum is maintained as its own section (albeit a paraphyletic taxon), S. penstemonoides should be included within it.
Salvia clade II
Salvia clade II (Figs. 4–6) is exclusively New World, is composed of the subg. Calosphace and the sect. Audibertia, and is likely sister to Dorystaechas. Dorystaechas is a monotypic genus restricted to southwest Anatolia placed by Bentham (1876)
in his "subtribe" Meriandreae along with Perovskia and Meriandra based on its two expressed stamens and parallel anther thecae. Dorystaechas is closely enough related to Salvia clade II that the molecular data cannot rule out its inclusion in clade II. Bentham suggested that Dorystaechas was most closely allied to the Indian and Himalayan genus Meriandra. Bokhari and Hedge (1976)
suggested the recently described Asian genus Zhumeria as a fourth genus assignable to the Meriandreae based on its staminal structure, pubescence, and corolla. The anthers of Dorystaechas and Meriandra are separated by slightly elongated connective (Bokhari and Hedge, 1971
). The connective in Perovskia is somewhat swollen, but not elongated. Following an analysis of the tribe Meriandreae based on anatomical and morphological characters, Bokhari and Hedge (1971)
concluded that the genera Perovskia, Meriandra, and Dorystaechas were in fact not likely to be closely related. Although we have yet to sample Meriandra, our molecular analysis supports this assertion as it relates to the genera Perovskia and Dorystaechas. Molecular evidence suggests Dorystaechas is likely sister to Salvia clade II, and Perovskia, together with Rosmarinus, sister to Salvia clade I (Fig. 4). The placement of Dorystaechas, Meriandra, and the members of Salvia clade III are currently being further investigated and will no doubt be central to fully understanding phylogenetics within the Mentheae, evolution of staminal structure within the Mentheae, and the origins of New World species of Salvia.
The New World section Audibertia and subg. Calosphace appear to be monophyletic sister groups in the combined analysis. Salvia sect. Audibertia is restricted to the California Floristic Province and adjacent deserts. The section contains 19 species, four of which (sect. Echinosphace Benth.) are distinct enough to have been separated into a section separate from sect. Audibertia by various authors (Bentham, 1832
; Neissess, 1983
). Section Echinosphace is characterized as having expressed and fertile posterior anther thecae, arachnoid pubescence, and a base chromosome number of 16. Section Audibertia is separated from sect. Echinosphace by phytochemical characters, by sterile or entirely aborted posterior anther thecae, and a base chromosome number of 15 (Neissess, 1983
). Our results suggest Neissess' sect. Echinosphace (in our data represented by S. greatai and S. californica) is sister to all other members of sect. Audibertia sensu lato. This relationship is consistent with previous work suggesting the isolated taxonomic nature of the latter two taxa (Epling, 1938
; Emboden, 1971
; Neisess, 1983
).
Subgenus Calosphace is a large assemblage of nearly 500 species separated from sections Audibertia and Echinosphace on the basis of sterile, fused posterior anther thecae forming a lever mechanism important in pollination (Fig. 3). Whereas Bentham (1876)
included Calosphace as part of Salvia, he treated Audibertia as a separate genus. Epling, whose revision of Calosphace remains the most complete treatment of the subgenus (Epling, 1939
), considered Audibertia and Calosphace as closely related (Epling, 1938
). Neissess (1983)
, who investigated relationships within sect. Audibertia based on morphological, chemical, and pollen characters, suggested sect. Audibertia was not allied with subg. Calosphace and more closely related to Rosmarinus and Old World groups of Salvia included in Bentham's subg. Leonia. Analyses of megagametophyte types (Carlson and Stuart, 1936
) and chromosome data (Epling et al., 1962
) in Salvia also suggested Old-World affinities of sect. Audibertia. Our results support a monophyletic subg. Calosphace sister to sect. Audibertia.
Salvia clade III
One of the more surprising results is the existence of a potential third clade of Salvia, clade III (Figs. 4–6). All species of this clade are East Asian (Hedge, 1998
) and correspond to Bentham's sect. Drymosphace. Section Drymosphace is defined by being large, herbaceous, glutinose plants with hastate leaves. The flowers are unique with the upper lip of the calyx short tridentate, a falcate and often compressed corolla, and connivent posterior anther thecae stretched forward. The rbcL analysis suggests this clade likely includes additional species outside sect. Drymosphace (e.g., S. barrelieri, S. fruticosa [both Mediterranean], S. digitaloides [Asian]) (Fig. 6). Staminal structure of S. glutinosa and S. hians is unique among the species of Salvia so far examined, with two connivent club-shaped posterior thecae producing no pollen and elongate anterior thecae producing pollen (Fig. 3). This unusual staminal structure is also found in S. koyamae, a Japanese species (J. B. Walker, personal observation). Future inclusion of more East Asian taxa in a combined analysis may elucidate the circumscription of this clade and help to define the morphological characters defining the clade as well as the broader relationships of Salvia clade III to the remainder of Salvia.
Biogeographic issues
A number of interesting biogeographical issues are raised with these molecular systematic results. Although the larger biogeographic history of Salvia and relatives in the Mentheae will need additional taxon sampling in the tribe for clarification, several finer scale biogeographic issues are clarified. Two of these issues involving New World taxa, the relationship of sect. Audibertia and subg. Calosphace (Salvia clade II) and the relationships within sect. Heterosphace (Salvia clade I), are briefly addressed here.
The putative sister relationship between sect. Audibertia and subg. Calosphace has interesting biogeographic implications. Although the California Floristic Province (CFP) is home to innumerable examples of distinct lineages, few groups respect the "boundaries" of the CFP as clearly as this group of Salvia. With the exception of adjacent desert regions, the 19 species of sect. Audibertia occur only in the CFP. More surprisingly, of the 500 species of subg. Calosphace (275 of which occur in Mexico [Ramamoorthy and Lorence, 1987
] and 18 in the Sonoran desert [Shreve and Wiggins, 1964
]) not one species occurs in California, Oregon, or Washington. Those that occur in Baja California (eight species) are all restricted to central and southern parts of the peninsula (Wiggins, 1980
). Why a group that has undergone explosive species radiations in the New World and is represented in nearly every region from southern South America to Canada should not have a single representative in the CFP is a question of considerable interest.
The second biogeographic issue revolves around sect Heterosphace in Salvia clade I. Bentham (1848)
was the first to recognize the close affinity of a group of southern African species to a southwestern group of North American species of Salvia that he placed in sect. Heterosphace. Our results further support the close relationship of the New World members of sect. Heterosphace (S. henryi, S. summa, S. roemeriana, S. lyrata) with Old World species. The fact that the New World members of sect. Heterosphace (together with sect. Salviastrum and S. penstemonoides) form a weak monophyletic group suggests a single dispersal from the Old World to the New, with subsequent diversification to give rise to the four species of sect. Heterosphace and three species of sect. Salviastrum in the New World. Defining a sister group to this lineage awaits further sampling of African species of Salvia.
Taxonomic implications for Salvia and relatives in Mentheae
A number of alternative taxonomic approaches are possible in dealing with the non-monophyly of Salvia. One approach would involve a nomenclatural change at the generic level of over 500 members of the genus Salvia. As the type species for the genus is Salvia officinalis L. (Britton and Brown, 1913
) belonging to Salvia clade I, all but eight of the 500 species of New World Salvia would transfer to new genera if Salvia were to be split. For example, Salvia subg. Calosphace would become Calosphace (Bentham) Rafinesque and Salvia sect. Audibertia would become Audibertia Bentham. A second approach would be the "sinking" of Dorystaechas, Perovskia, Rosmarinus, and possibly other genera into Salvia. A third approach would be to employ phylogenetic nomenclature, retaining current species binomials while naming clades to which they belong. This approach has been advocated for mints and their relatives (Cantino et al., 1997
, 1998
).
Our research continues to investigate relationships within this important group of plants, and we believe it is premature to present any taxonomic rearrangements before we have completed sampling across the diversity of Salvia and related genera. Regardless of which approach is embraced, one would be hard-pressed to find a group of plants containing more species of horticultural, ethnobotanical, and culinary import than the Nepetoideae which includes sage (Salvia), savory (Satureja), balm (Melissa), rosemary (Rosmarinus), thyme (Thymus), bee balm (Monarda), catnip (Nepeta), oregano (Origanum), basil (Ocimum), and mint (Mentha). This popularity leads one to hope that as many of these genera are demonstrated to be non-monophyletic (e.g., Salvia, Satureja [Cantino and Wagstaff, 1998
], Lepechinia [Hart, 1983
]), any modifications to nomenclature at the generic level will take into account the significant broader impact of any changes (Sytsma and Pires, 2001
).
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Barber J. C. J. Francisco-Ortega A. Santos-Guerra K. G. Turner R. K. Jansen 2002 Origin of Macaronesian Sideritis L. (Lamioideae: Lamiaceae) inferred from nuclear and chloroplast sequence datasets. Molecular Phylogenetics and Evolution 23: 293-306[CrossRef][ISI][Medline]
Baum D. A. K. J. Sytsma P. C. Hoch 1994 The phylogeny of Epilobium L. (Onagraceae) based on nuclear ribosomal DNA sequences. Systematic Botany 19: 363-388[CrossRef][ISI]
Bentham G. 1832 Labiatarum genera et species. Ridgeway, London
Bentham G. 1848 Labiatae. In A. De Candolle [ed.], Prodromus systematis naturalis regni vegetabilis, Vol. 12, 27–603. Treuttel et Wurtz, Paris, France
Bentham G. 1876 Labiatae. In G. Bentham and J. D. Hooker, Genera Plantarum, Vol. 2, 1160–1196. Reeve and Co., London, UK
Bokhari M. H. I. C. Hedge 1971 Observations on the tribe Meriandreae of the Labiatae. Notes, Royal Botanical Garden, Edinburgh 31: 53-67[Medline]
Bokhari M. H. I. C. Hedge 1976 Zhumeria (Labiatae): anatomy, taxonomy and affinities. Iran Journal of Botany 1: 1-10
Bremer K. 1988 The limits of amino acid sequence data in angiosperm phylogenetic reconstruction. Evolution 42: 795-803[CrossRef][ISI]
Briquet J. 1897 Labiatae. In A. Engler and K. Prantl, Die naturlichen Pflanzenfamilien nebst ihrer Gattungen und wichtigeren Arten Vol. IV, 183–287. Wilhelm Engelmann, Leipzig, Germany
Britton N. L. A. Brown 1913 An illustrated flora of the northern United States, Canada and the British possessions. Charles Scribner's Sons, New York, New York, USA
Cantino P. D. A. M. Harley S. J. Wagstaff 1992 Genera of Labiatae: status and classification. In R. M. Harley and T. Reynolds [eds.], Advances in labiate science, 511–522. Royal Botanic Gardens, Kew, UK
Cantino P. D. R. G. Olmstead S. J. Wagstaff 1997 A comparison of phylogenetic nomenclature with the current system: a botanical case study. Systematic Biology 46: 313-331[CrossRef][ISI]
Cantino P. D. W. Sanders 1986 Subfamilial classification of the Labiatae. Systematic Botany 11: 163-185
Cantino P. D. S. J. Wagstaff 1998 A reexamination of North American Satureja s.l. (Lamiaceae) in light of molecular evidence. Brittonia 50: 63-70
Cantino P. D. S. J. Wagstaff R. G. Olmstead 1998 Caryopteris (Lamiaceae) and the conflict between phylogenetic and pragmatic considerations in botanical nomenclature. Systematic Botany 23: 369-386[CrossRef][ISI]
Carlson E. B. Stuart 1936 Development of spores and gametophytes in certain New World species of Salvia. New Phytologist 35: 68-91[CrossRef]
Catalán P. E. A. Kellogg R. G. Olmstead 1997 Phylogeny of Poaceae subfamily Pooideae based on chloroplast ndhF gene sequences. Molecular Phylogenetics and Evolution 8: 150-166[CrossRef][ISI][Medline]
Claßen-Bockhoff R. P. Wester E. Tweraser 2003 The staminal lever arm mechanism in Salvia—a review. Plant Biology 5: 33-41[CrossRef][ISI]
Clebsch B. 1997 A book of salvias. Timber Press, Portland, Oregon, USA
Cole M. D. 1992 The significance of the terpenoids in the Labiatae. In R. M. Harley and T. Reynolds [eds.], Advances in labiate science, 315–324. Royal Botanic Gardens, Kew, UK
Conti E. A. Litt K. J. Sytsma 1996 Circumscription of Myrtales and their relationships to other rosids: evidence from rbcL sequence data. American Journal of Botany 83: 221-233[CrossRef][ISI]
Donoghue M. R. G. Olmstead J. F. Smith J. D. Palmer 1992 Phylogenetic relationships of Dipsacales based on rbcL sequences. Annals of the Missouri Botanical Garden 79: 333-345
El-Gazzar A. L. Watson 1968 The taxonomy of Salvia: a test of two radically different numerical methods. Journal of the Linnaean Society (Botany) 60: 237-250[Medline]
El-Gazzar A. L. Watson 1970 A taxonomic study of Labiatae and related genera. New Phytologist 69: 451-486[CrossRef][ISI]
El Oualidi J. O. Verneau S. Puech J. Y. Dubuisson 1999 Utility of rDNA, ITS sequences in the systematics of Teucrium section Polium (Lamiaceae). Plant Systematics and Evolution 215: 49-70[CrossRef][ISI][Medline]
Emboden W. A. 1971 The role of introgressive hybridization in the development of Salvia section Audibertia. Contributions in Science 208: 1-15[Medline]
Epling C. 1938 The Californian salvias. Annals of the Missouri Botanical Garden 25: 95-188[CrossRef][Medline]
Epling C. 1939 A revision of Salvia, subgenus Calosphace. Beihefte Feddes Repertorium specierum novarum regni vegetabilis 110: 1-383[Medline]
Epling C. H. Lewis P. Raven 1962 Chromosomes of Salvia: section Audibertia. Aliso 5: 217-221
Farris J. S. M. Kallersjo A. G. Kluge C. Bult 1995 Testing significance and incongruence. Cladistics 10: 315-319[CrossRef][ISI]
Felsenstein J. 1985 Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783-791[CrossRef][ISI]
Fitch W. M. 1971 Toward defining the course of evolution: minimum change for a specific tree topology. Systematic Zoology 20: 406-416[CrossRef][ISI]
Givnish T. J. T. M. Evans M. L. Zjhra P. E. Berry K. J. Sytsma 2000 Molecular evolution, adaptive radiation and geographic diversification in the amphiatlantic family Rapateaceae: evidence from ndhF sequence data. Evolution 54: 1915-1937[CrossRef][ISI][Medline]
Hart J. A. 1983 Systematics and evolution in the genus Lepechinia. Ph.D. dissertation, Harvard Uniersity, Cambridge, Massachusetts, USA
Hedge I. C. 1974 A revision of Salvia in Africa and the Canary Islands. Notes, Royal Botanical Garden, Edinburgh 33: 1-121[Medline]
Hedge I. C. 1982a Labiatae. In P. H. Davis [ed.], Flora of Turkey and the eastern Aegaean Islands, vol. 7, Labiatae, 36–463. Edinburgh University Press, Edinburgh, Scotland
Hedge I. C. 1982b Salvia. In K. H. Rechinger [ed.], Flora Iranica, vol. 150, Labiatae, 401–476. Akademische Druckund, Graz, Austria
Hedge I. C. 1998 Salvia. In Wu Zheng-yi and P. Raven [eds.], Flora of China, vol. 17, Verbenaceae through Solanaceae. Missouri Botanical Garden, St. Louis, Missouri, USA
Himmelbaur W. E. Stibal 1933–1935 Entwicklungsrichtungen in der blutenregion der gattung Salvia L. I–III. Biologia generalis 8: 449-474 9:129–150; 10:17–48 [Medline]
Hipp A. L. J. C. Hall K. J. Sytsma 2004 Phylogenetic accuracy, congruence between data partitions, and performance of the ILD. Systematic Biology 53: 81-89[CrossRef][ISI][Medline]
Hruby K. 1962 Key to the supraspecific taxa of the genus Salvia. Preslia 34: 368-373[Medline]
Huelsenbeck J. P. F. R. Ronquist 2001 MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754-755
Kauffmann M. M. Wink 1994 Molecular systematics of the Nepetoideae (family Labiatae): phylogenetic implications from rbcL gene sequences. Zeitschrift fur Naturforschung 49c: 635-645[Medline]
Kunth C. S. C. Bouche 1848 Index seminum in horto botanico berolininensi. Keonglicher Botanischer Garten, Berlin, Germany
Maddison D. R. 1991 The discovery and importance of multiple islands of most-parsimonious trees. Systematic Zoology 40: 315-328[CrossRef]
Neissess K. R. 1983 Evolutions, systematics and terpene relationships of Salvia section Audibertia. Ph.D. dissertation, University of California, Riverside, California, USA
Olmstead R. G. B. Bremer K. M. Scott J. D. Palmer 1993 A parsimony analysis of the Asteridae s.l. based on rbcL sequences. Annals of the Missouri Botanical Garden 80: 700-722[CrossRef][ISI]
Olmstead R. G. C. W. Depamphilis A. D. Wolfe N. D. Young W. J. Elisens P. A. Reeves 2001 Disintegration of the Scrophulariaceae. American Journal of Botany 88: 348-361
Olmstead R. G. K. J. Kim R. K. Jansen S. J. Wagstaff 2000 The phylogeny of the Asteridae s.l. based on chloroplast ndhF gene sequences. Molecular Phylogenetics and Evolution 16: 96-112[CrossRef][ISI][Medline]
Olmstead R. G. H. J. Michaels K. M. Scott J. D. Palmer 1992 Monophyly of the Asteridae and identification of their major lineages inferred from DNA sequences of rbcL. Annals of the Missouri Botanical Garden 79: 249-265
Pobedimova E. G. 1954 Labiatae. In B. K. Shishkin [ed.], Flora of the U.S.S.R., vol. 21, Labiatae, 178–260. Izdatel'stvo Akademii Nauk SSR, Moscow, USSR
Posada D. K. A. Crandall 1998 Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817-818
Prather L. A. A. K. Monfils A. L. Posto R. A. Williams 2002 Monophyly and phylogeny of Monarda (Lamiaceae): evidence from the internal transcribed spacer (ITS) region of nuclear ribosomal DNA. Systematic Botany 27: 127-137[ISI][Medline]
Ramamoorthy T. P. D. H. Lorence 1987 Species vicariance in the Mexican flora and description of a new species of Salvia. Bulletin du Museum National d'histoire Naturelle 4: 167-175[Medline]
Rambaut A. 2001 Sequence alignment editor, version 2.0a6. Available online at http://evolve.zoo.ox.ac.uk/
Scheele A. 1849 Beitrage zur Flora von Texas. Linnaea 22: 584[Medline]
Shreve F. I. Wiggins 1964 Vegetation and flora of the Sonoran Desert, vol. 2. Stanford University Press, Stanford, California, USA
Sprengel C. K. 1793 Das entdeckte Geheimnis der Natur im Bau und in der Befruchtung der Pflanzen. Friedrich Vieweg dem aeltern, Berlin, Germany
Steane D. A. R. W. Scotland D. J. Mabberley R. G. Olmstead 1999 Molecular systematics of Clerodendrum (Lamiaceae): ITS sequences and total evidence. American Journal of Botany 86: 98-107
Steane D. A. R. W. Scotland D. J. Mabberley S. J. Wagstaff P. A. Reeves R. G. Olmstead 1997 Phylogenetic relationships of Clerodendrum s.l. (Lamiaceae) inferred from chloroplast DNA. Systematic Botany 22: 229-243
Stibal E. P. 1934 A revision of Chinese and East Burmese species of Salvia. Acta Horti Gotoburgensis 9: 101-165[Medline]
Stibal E. P. 1935 Revision der indischen und tibetanischen arten der gattung Salvia. Repertorium specierum novarum regni vegetabilis 39: 173-186[Medline]
Swofford D. L. 1993 PAUP*. Phylogenetic analysis using parsimony version 3.1.1. Illinois Natural History Survey, Champaign, Illinois, USA
Swofford D. L. 2002 PAUP*. Phylogenetic analysis using parsimony (*and other methods), version 4.0b10. Sinauer Associates, Sunderland, Massachusetts, USA
Sytsma K. J. J. C. Pires 2001 Plant systematics in the next 50 years—re-mapping the new frontier. Taxon 50: 713-732[CrossRef][ISI]
Torrey J. 1859 Botany of the boundary. Report of the U.S.–Mexican Boundary Survey 2: 132[Medline]
Wagstaff J. J. 1992 Phylogeny and character evolution in subfamily Nepetoideae (Labiatae). Ph.D. dissertation, Ohio University, Athens, Ohio, USA
Wagstaff S. J. L. Hickerson R. Spangler P. A. Reeves R. G. Olmstead 1998 Phylogeny in Labiatae s.l. inferred from cpDNA sequences. Plant Systematics and Evolution 209: 265-274[CrossRef][ISI]
Wagstaff S. J. R. G. Olmstead 1997 Phylogeny of Labiatae and Verbenaceae inferred from rbcL sequences. Systematic Botany 22: 165-179
Wagstaff S. J. R. G. Olmstead P. D. Cantino 1995 Parsimony analysis of cpDNA restriction site variation in subfamily Nepetoideae (Labiatae). American Journal of Botany 82: 886-892[CrossRef][ISI]
Walker J. B. W. J. Elisens 2001 A revision of Salvia section Heterosphace in western North America. Sida 19: 571-589[Medline]
Whitehouse E. 1949 Revision of Salvia section Salviastrum. Field and Laboratory 17: 151-165[Medline]
Wiggins I. 1980 Flora of Baja California. Stanford University Press, Stanford, California, USA
Wink M. M. Kauffman 1996 Phylogenetic relationships between some members of the subfamily Lamioideae inferred from nucleotide sequences of the rbcL gene. Botanica Acta 109: 139-148[ISI]
Wunderlich R. 1967 Ein Vorschlag zu einer naturlichen Gleiderung der Labiaten auf Grund der Pollenkorner, der Samenentwicklung und des reifen Samens. Oesterriechisce Botanische Zeitschrift 114: 383-483[CrossRef][Medline]
Yoder A. D. J. A. Irwin B. A. Payseur 2001 Failure of the ILD to determine data combinability for slow loris phylogeny. Systematic Biology 50: 408-424[ISI][Medline]
This article has been cited by other articles:
|
|
 |
|
  R. Classen-Bockhoff Floral Construction and Pollination Biology in the Lamiaceae Ann. Bot., August 1, 2007; 100(2): |
