| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
2 Department of Botany, University of Hawaii at Manoa, Honolulu, Hawaii 96822 USA; and 3 Department of Biology, Eastern Washington University, Cheney, Washington 99004 USA
Received for publication September 29, 1999. Accepted for publication January 4, 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Asteraceae • Calycadenia • chromosome races • Compositae • cytogenetics • hybridization • Madiinae • tarweeds
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), at least four of which are highly variable. Here we deal primarily with one of the polymorphic species, C. pauciflora A. Gray.
Keck's circumscription of Calycadenia pauciflora included a series of relatively well-known populations in the North Coast Range, reportedly extending from Sonoma to Glenn counties; to these he added some poorly known populations from across the Central Valley in the Sierran foothills of Calaveras and Stanislaus counties, California (Munz and Keck, 1959
; Keck, 1960
). The unifying features of the populations on opposite sides of the Central Valley were primarily the small capitula with white flowers and the generally small stature of the plants. More critical examination and biosystematic studies revealed the very distinctive nature of the Sierran foothill plants and resulted in their recognition as a separate species, C. hooveri G. Carr (Carr, 1975a
).
Even with the exclusion of the Sierran component, the residual North Coast Range populations of Calycadenia pauciflora comprise a very heterogeneous assemblage. Biosystematic examination of these populations (Carr, 1975b
) revealed two distinct elements, one with n = 5 (race Pauciflora) and a second one with n = 6 that is further divisible into four chromosome races (Elegans, Healdsburg, Ramulosa, and Tehama) differentiated primarily by reciprocal translocations. Morphologically, races Elegans and Healdsburg have larger capitula and coarser stems with a more dominant central axis compared to Pauciflora, Ramulosa, and Tehama (Figs. 1–12). Although these chromosome races were all treated as belonging to Calycadenia pauciflora in an earlier biosystematic study (Carr, 1975b
), more recent studies (Carr and Carr, 1983
, and unpublished data) have led to amalgamation of C. ciliosa Greene, C. fremontii A. Gray, and the two larger-headed chromosome races (Elegans and Healdsburg) formerly included in C. pauciflora (Carr and Carr, 1993
). This enlarged species receives the earliest published name, C. fremontii A. Gray.
|
had reported the first collection of Calycadenia fremontii s.s. since 1937 and only the second collection following its initial discovery in 1846 by Fremont. In the report of their rediscovery Hayes, Schlising, and Wurlitzer (1979)
referred to two small populations of C. fremontii separated by only a short distance.
Our interest in the Calycadenia populations previously known as C. ciliosa (Carr, 1977
; Carr and Carr, 1983
) led us to examine the newly reported populations of C. fremontii. While one of these populations proved to be C. fremontii s.s., the smaller heads, diffuse habit, and wiry stems of plants in the other population suggested closer affinities with populations remaining in Calycadenia pauciflora s.s. (Figs. 3–6, 9–12). The most obvious distinction of this newly discovered population appeared to be its location on the floor of the Central Valley, somewhat separated from other populations of C. pauciflora in the North Coast Range.
Neither molecular nor cytogenetic data have yet satisfactorily explained the complex pattern of relationships among the populations of Calycadenia pauciflora and C. fremontii (Carr, 1975b
; Carr, 1977
; Carr and Carr, 1983
; Baldwin, 1993
; Baldwin and Markos, 1998
). In an attempt to help resolve this situation, we undertook a crossing program designed to establish the position of the newly discovered population in the framework of chromosome evolution available for Calycadenia pauciflora (Carr, 1975b
). This paper reports the results of this program of hybridization in terms of fertility and chromosomal relationships, and establishes yet another chromosome race (Wurlitzer) among the small-headed, white-flowered populations of Calycadenia from northern California.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
) were established in greenhouse culture from cypselae that had been collected from the field and held in cold storage (Table 1). Procedures related to hybridization, assessment of pollen fertility based on stainability of grains, and preparation of floral bud material for analysis of chromosome pairing behavior during meiosis followed those detailed in earlier studies of Calycadenia (e.g., Carr, 1975b
). Production of hybrids was facilitated by the self-incompatible nature of all populations used in this study. Although 300–500 or more meiocytes of each hybrid combination were available for study, the frequencies of the various chromosome configurations were not quantified.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
). The expression of NO function of a given NO chromosome varies among populations, and this variation is especially apparent in hybrids between populations, where a hierarchy of NO function is commonly seen (see also McClintock, 1934
; Navashin, 1934
). Only rarely is there evidence of function of all four NO chromosomes in a single cell. However, by observing several cells of different progenies of a given hybrid combination, all four NO chromosomes can usually be detected. The limited number of cells that could be illustrated here did not allow documentation of this phenomenon. Nevertheless, NO chromosomes have considerable utility as markers that facilitate recognition of these hybrids based on their meiotic chromosome configurations. Where available this information is included in the analyses.
The NO region is located in the short arm of each NO chromosome (see Carr, 1975b
, fig. 15), and these arms are also comparatively heterochromatic. The net effect is a suppression of chiasma formation in the short arms of NO chromosomes. However, translocations involving NO chromosomes are a recurrent theme in the differentiation of the chromosome races of the C. pauciflora–C. fremontii complex, including those described herein. The suppression of chiasma formation in the short arms of NO chromosomes has considerable effect on the frequencies of various chromosome configurations observed in a given hybrid, and in some cases may prevent detection of the theoretical maximum meiotic chromosome association (MMCA).
Chromosomes associated mostly as pairs and multivalents prior to the reductional division of meiosis in all hybrid combinations. Univalents were only infrequently encountered and very rarely did a single cell contain more than one. The maximum chromosome associations ranged from a ring of six and three pairs in Tehama x Wurlitzer to a chain of ten and one pair in Elegans x Wurlitzer. A more detailed characterization of each hybrid combination is presented below.
Elegans (n = 6) x Wurlitzer (n = 6)
A very frequent chromosome configuration seen in this hybrid combination was one linear chain of four, one linear and one branched chain of three, and one ring pair of chromosomes (Fig. 14). In other cells articulation of the linear chain of three and the linear chain of four into a chain of seven chromosomes was commonly observed (Fig. 15). The apparent maximum meiotic chromosome association (MMCA) in this interracial hybrid is a chain of ten and one ring pair (Fig. 16) and was seen in only a few cells. The infrequency of this configuration is expected because of the position of a NO chromosome at each end of the chain of seven and the requirement for rare chiasma formation between NO arms in order to concatenate the branched chain of three and chain of seven into a chain of ten (cf. Figs. 15 and 16). The two chromosomes not involved in chain formation consistently form a ring pair, suggesting their complete homology. Thus, the observed MMCA of a chain of ten appears to identify all of the chromosomes involved in gross structural alterations. The equivalent of a minimum of four reciprocal chromosome translocations would be required to produce these results. The mean pollen stainability of seven individuals of this hybrid combination was 13% (Table 2, Fig. 13).
|
|
Pauciflora (n = 5) x Wurlitzer (n = 6)
The MMCA observed in this hybrid combination was frequently encountered, consisting of a multiple of seven and two rod pairs The multiple of seven exhibited a branch near one end and a heteromorphic ring at the other end (Fig. 21). Several other permutations involving smaller associations were also commonly observed, e.g., those depicted in Figs. 19 and 20 in which the heteromorphic ring is seen as a separate element. An identical heteromorphic ring appears in all interracial hybrids involving Race Pauciflora (cf. Carr, 1975b
). Although it sometimes resembles a ring of three isochromosomes, the heteromorphic ring consists of one large metacentric chromosome (Fig. 22) from Race Pauciflora and a much smaller chromosome from Race Wurlitzer (or potentially any of the other six-paired races described here). Remarkably, this large chromosome expresses homology distally with both arms of the small (Wurlitzer) chromosome in the heteromorphic ring and proximally (near the centromere) with the distal portion of a second chromosome from Race Wurlitzer. Exactly the same behavior is seen in all of the hybrids between Race Pauciflora and six-paired races reported earlier, although pairing details of the remaining chromosomes differ from one combination to the next (Carr, 1975b
).
With a mean pollen stainability of 26% for nine individuals, this hybrid combination is apparently the most fertile, even though other combinations may have simpler meiotic configurations. Higher than expected pollen stainability has also been observed in most other hybrids involving Race Pauciflora (Carr, 1975b
). Fertility in these hybrids is probably enhanced by the compensating effect of the large metacentric chromosome contributed by the Race Pauciflora genome. This chromosome combines the genetic material equivalent to two chromosomes from the n = 6 parent and would genetically buffer some of the products of meiosis that would otherwise be inviable.
It was not possible to locate all of the NO chromosomes in this combination with certainty but at least one of the chromosomes in each of the two bivalents that appear as rods in most cells function as nucleolar organizers (Figs. 19–21). The level of pollen stainability suggests that these rod pairs may be completely homologous and that all of the chromosomal differences of the parents may be reflected in the observed MMCA. If this is the case, then both chromosomes of each of the rod pairs must be NO chromosomes and the failure to observe them all functioning is very likely due to the hierarchy of nucleolar organizer activity in hybrids mentioned earlier. The MMCA observed in hybrids between these races indicates that the genomes of Pauciflora and Wurlitzer are differentiated by the equivalent of a minimum of three reciprocal chromosome translocations.
Ramulosa (n = 6) x Wurlitzer (n = 6)
In this hybrid combination a rare MMCA of one linear chain of eight and two ring pairs was observed (Fig. 24). In other cells the chain of eight was rarely replaced by two chains of four or more commonly a chain of six and an extra pair (cf. Fig. 23). Other permutations were observed and some cells even had six pairs of chromosomes, although some of the pairs were heteromorphic. The chain of six frequently observed appears to have a NO chromosome at each end. The rod pair that adds to the chain of six to make the rare chain of eight is also composed of two NO chromosomes. Placement of the NO chromosomes probably accounts for the rarity of the chain of eight and also for the lack of observation of a ring of eight that possibly occurs at some extremely low frequency. The observed MMCA of hybrids between Ramulosa and Wurlitzer indicates that the genomes of these races are differentiated by the equivalent of at least three reciprocal chromosome translocations. The mean pollen stainability of six individuals of this hybrid combination was 16%.
Tehama (n = 6) x Wurlitzer (n = 6)
The observed MMCA of a ring of six and three pairs (Fig. 27) was seen in several cells but was not common. In many cells a linear or sometimes branched chain of six replaced the ring of six. In other cells the chain of six was replaced by a multiple of four and an additional pair (Figs. 25 and 26), or not uncommonly a chain of three, a pair, and a univalent. The very unusual multiple of four ("frying pan" configuration) in Figs. 25 and 26 is identical to that seen in many other cells of this combination but nothing like it has been observed in other Calycadenia hybrids. It is peculiar in having three chromosomes with their centromeres in a ring and a fourth chromosome with its centromere and one arm outside of the ring. This unusual configuration could be explained in a variety of ways, but it appears probable that a history of pericentric inversion plays a role (see Discussion).
Of the remaining three bivalents not involved in multiple associations, one or two commonly formed ring bivalents and rarely all three formed ring bivalents. Both chromosomes of the pair that usually formed a rod were frequently involved in nucleolar organization. Thus, it appears likely that the observed MMCA reflects all of the gross chromosomal restructuring that differentiates the parental races. The multiple of six in this configuration indicates that the parental genomes differ by the equivalence of a minimum of two reciprocal chromosome translocations and probably a pericentric inversion. The mean pollen stainability of nine individuals of this hybrid combination was 19%.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). In contrast, Carr and Carr (1983)
were able to relate a series of five chromosome races of what is now considered C. fremontii in single-step fashion. Similarly, Kyhos (1965)
documented the stepwise origin of chromosomally differentiated species of Chaenactis.
There are at least three alternatives that independently or collectively may account for the observed paucity of single-step relationships among the races investigated in this study. First, it is possible that one or more of the races could have originated precipitously by saltational reorganization of chromosomes. There is experimental evidence that Race Pauciflora could have originated from Race Tehama in this fashion (Carr, 1980
). In this case the mean pollen stainability of the structural heterozygotes was 44%, far above any of the Race Wurlitzer hybrids reported here but comparable to the Pauciflora hybrids with two other races examined in the earlier study of C. pauciflora (Carr, 1975b
).
A second explanation for the failure to identify single-step cytogenetic relationships among the races of Calycadenia pauciflora is that although the races representing intermediate steps may have existed in the past, they may now be extinct or perhaps still extant but yet undiscovered. Indeed, efforts in the field have located many new populations of Calycadenia, some of them apparently representing new chromosome races such as Wurlitzer, described herein, and others in the C. fremontii alliance that remain to be characterized cytogenetically.
Finally, we may be suffering from taxonomic bias, i.e., we may be looking for the closest relationships in the wrong species. We have generally thought of resolution of chromosome races of C. fremontii and C. pauciflora as two separate problems, when in fact they may be one and the same. Two of the races previously treated as C. pauciflora (Elegans and Healdsburg) have already been shifted to C. fremontii, along with all of the known races of C. ciliosa. Races Elegans and Healdsburg are cytogenetically relatively close to Race Ciliosa but could be even closer to other races in C. fremontii, including some that have not even been adequately characterized as of yet (Carr and Carr, unpublished data). Unfortunately, the hybrids that would provide this information have not yet been synthesized.
Perhaps even the small-headed races (Pauciflora, Ramulosa, Tehama, and Wurlitzer) that seem to form a practical taxonomic unit now treated as C. pauciflora have counterparts among the races of C. fremontii that are cytogenetically more similar to them than they are to each other. Both cytogenetic and molecular data suggest this possibility. The cytogenetic clue that suggests a closer than anticipated link between small-headed races of C. pauciflora and certain races of C. fremontii is the appearance of two anomalous chromosome pairing configurations, one in the Wurlitzer–Tehama hybrids reported herein ("frying pan" configuration, Figs. 25 and 26), and another one (a peculiar loop-like configuration between two paired chromosomes) in several interracial hybrid combinations now treated as C. fremontii (Carr and Carr, 1983
). Both of these anomalies can be explained on the basis of heterozygosity for what may be the same pericentric inversion. A detailed pairing analysis and the potential mode of origin of the pericentric inversion that would account for the configurations seen in certain interracial hybrids of C. fremontii were presented earlier (Carr and Carr, 1983
). What appear to be identical pairing anomalies were also reported in the study of C. pauciflora (Carr, 1975b
).
The anomalous pairing configuration seen in the Tehama x Wurlitzer hybrids (Figs. 25 and 26) is not the same as that reported earlier but can be explained on the same basis. This multiple of four chromosomes has three chromosomes in a ring and the centromere and arm of the fourth chromosome outside of the ring. Moreover, in other cells the multiple extends to six chromosomes that may include the ring of three at one end, or alternatively may form a ring of six (Fig. 27). These observations could probably be explained in a variety of ways, perhaps most readily by invoking the presence of a duplicated chromosome segment. However, the explanation that seems most consistent with observed patterns of chromosomal differentiation in Calycadenia does not rely on the incorporation of a duplicated segment to account for the anomalous pairing. This scenario requires only reciprocal translocation of portions of both arms of one chromosome with terminal portions of each of two other chromosomes. The further requirement is that a sizable proximal portion including the centromere of the recurrent translocation chromosome be inverted. The necessary pericentric inversion could have occurred concomitant with the translocations, but this would have required a minimum of four simultaneous breaks and would also have resulted in an initial structural heterozygote with substantially reduced fertility, thus lowering the probability of fixation of the derived structural homozygote. Alternatively, the pericentric inversion and the translocations could have occurred stepwise in any order with resulting intermediate structural heterozygotes of comparatively high fertility.
Regardless of the mode of origin, a pericentric inversion such as this could account for all of the chromosome configurations observed in the Tehama x Wurlitzer hybrids, as well as the anomalous configurations seen in the earlier studies of C. pauciflora and C. fremontii. If these interpretations are correct and if the inversion has a single common origin, then it appears that C. pauciflora, even narrowly defined, is not strictly monophyletic and indeed, some of its chromosome races may be more closely related to races of C. fremontii than any of them are to each other.
Molecular evidence appears to further weaken the historical concept of these species; phylograms based on rDNA data failed to distinguish chromosome races of C. pauciflora as a distinct clade and often group Race Pauciflora very close to or with races of C. fremontii (Baldwin, 1993
; Baldwin and Markos, 1998
). Unfortunately, Race Pauciflora is the only small-headed one included in the molecular analyses, and only three cytogenetically characterized elements of C. fremontii (Corning, Elegans, and Lewiston) are included. Curiously, Race Elegans was displaced in the analyses such that strict monophyly for the group of races originally treated as C. pauciflora would not be achieved short of inclusion of all of the races of C. fremontii and C. pauciflora, together with C. oppositifolia Greene, a narrow endemic of Butte County with n = 7 (Baldwin, 1993
; Baldwin and Markos, 1998
).
Given the tremendous range of nearly continuous morphological and cytogenetic variation encompassed by such a group, it may represent the epitome of impracticality of adherence to cladistic methods in the recognition of formal taxonomic categories. On the other hand, additional molecular and cytogenetic data may help to make some sense out of what seems on the surface to be an intractable problem. Calycadenia certainly offers many opportunities for both avenues of investigation.
While the data presented here clearly indicate that Elegans and Wurlitzer are the most cytogenetically distant races, the nearest relative of Race Wurlitzer is not identified with any certainty. The lack of single-step transitions together with the realization that C. fremontii may harbor cytotypes relevant to any detailed understanding of evolutionary relationships of these races makes it unreasonable to speculate further on the matter at this time. We look forward to the possibility of further resolution of chromosome evolution of C. fremontii and the placement of these pieces of the puzzle into proper perspective. If this can be accomplished, the C. fremontii–C. pauciflora alliance will likely provide one of the most complex and instructive examples of chromosome evolution among plants.
|
FOOTNOTES |
|---|
4 Author for reprint request (e-mail:gerry{at}hawaii.edu
)
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
———, and S. Markos. 1998 Phylogenetic utility of the external transcribed spacer (ETS) of 18S–26S rDNA: congruence of ETS and ITS trees of Calycadenia (Compositae). Molecular Phylogenetics and Evolution 10: 449–463[CrossRef][ISI][Medline]
Carr, G. D. 1975a Calycadenia hooveri (Asteraceae), a new tarweed from California. Brittonia 27: 136–141[CrossRef][ISI]
———. 1975b Chromosome evolution and aneuploid reduction in Calycadenia pauciflora (Asteraceae). Evolution 29: 681–699[CrossRef][ISI]
———. 1977 A cytological conspectus of the genus Calycadenia (Asteraceae): an example of contrasting modes of evolution. American Journal of Botany 64: 694–703[CrossRef][ISI]
———. 1980 Experimental evidence for saltational chromosome evolution in Calycadenia pauciflora Gray (Asteraceae). Heredity 45: 107–112[ISI]
Carr, R. L., and G. D. Carr. 1983 Chromosome races and structural heterozygosity in Calycadenia ciliosa Greene (Asteraceae). American Journal of Botany 70: 744–755[CrossRef][ISI]
———, and ———. 1993 Calycadenia. In J. C. Hickman [ed.], The Jepson manual: higher plants of California, 218–219. University of California Press, Berkeley, California, USA
Hayes, M., R. Schlising, and H. Wurlitzer. 1979 Calycadenia fremontii rediscovered. Fremontia 7: 14–15
Keck, D. D. 1960 Subtribe Madiinae. In R. S. Ferris [ed.], Illustrated Flora of the Pacific States, vol. IV, 154–192. Stanford University Press, Stanford, California, USA
Kyhos, D. 1965 The independent aneuploid origin of two species of Chaenactis (Compositae) from a common ancestor. Evolution 19: 26–43[CrossRef][ISI]
McClintock, B. 1934 The relation of a particular chromosomal element to the development of the nucleoli in Zea mays. Zeitschrift fuer Zellforschung und Mikroskopische Anatomie 21: 294–328[CrossRef]
Munz, P. A., and D. D. Keck. 1959 A California flora. University of California Press, Berkeley, California, USA
Navashin, M. 1934 Chromosome alterations caused by hybridization and their bearing upon certain general genetic problems. Cytologia 5: 169–203
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
