| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Dep. of Crop Sciences, Univ. of Illinois at Urbana-Champaign, 320 ERML, 1201 W. Gregory, Urbana, IL 61801
* Corresponding author (arayburn{at}uiuc.edu)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
0.95 pg in A. palmeri to 1.4 pg in A. tuberculatus. Overlap of genome sizes among the species does exist; however, due to the reproductive biology of the species, this overlap does not preclude the detection of interspecific hybrids. Any dioecious weedy Amaranthus plant in Illinois that has a genome size between 1.3 pg and 1.1 pg is probably a hybrid. Thus, the potential for genome size determination to reveal interspecific hybrids among weedy Amaranthus species has been demonstrated.
Abbreviations: FI, fluorescence inhibitor • RBG Kew, Royal Botanical Gardens, Kew, UK
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
One way in which genetic material could be exchanged is by introgressive hybridization. Reports have indicated that, in plants, interspecific hybrids occur in nature (Stebbins, 1950). Although this evolutionary importance has been debated for decades, given the right circumstances, interspecific hybrids could play a crucial role in the evolution of many plant species, especially those labeled noxious weeds. The greatest potential for introgressive hybridization to have an impact in the evolution of a species is in areas of "much-disturbed habitats" (Stebbins, 1950). One of the most continually disrupted areas is the agricultural land of the Midwest. Given the intense farming practices used and the nearly constant changes in Best Management Practices, agrichemicals used, etc., the landscape is in a continual state of flux. If interspecific hybridization does occur in Amaranthus, introgressive hybridization could be very important in the evolution and adaptation of these weedy species.
Murray (1940) reported that hybrids could be made among various monoecious and dioecious pigweed species. Wetzel et al. (1999) and Franssen et al. (2001) reported the production of interspecific hybrids between two dioecious pigweed species. Tranel et al. (2002) reported interspecific hybridization between a monoecious and a dioecious pigweed species. In all the above reports, the hybrids were made under controlled, isolated conditions. Trucco et al. (2005) demonstrated interspecific hybrids between A. hybridus and A. tuberculatus under field conditions. Thus, it is possible that hybrids between pigweeds could occur naturally in agronomic fields.
Jeschke et al. (2003) and Wetzel et al. (1999) indicated that a major problem with assessing natural hybridization in pigweeds is the ability to identify such hybrids. Jeschke et al. (2003) suggested that genome size could be used to identify hybrids between A. hybridus and A. tuberculatus. Using flow cytometry to detect hybrid plants has become common practice (Barre et al., 1998; Williams et al., 2002; Sisko et al., 2003; Nimura et al., 2003; Horita et al., 2003). Critical to the usefulness of genome size in identifying hybrids is the DNA content of the parental types. To identify hybrids, the parents need to differ significantly with respect to nuclear genome size. Such is the case with respect to A. hybridus and A. tuberculatus, the two most common pigweeds in Illinois. The difference in genome size is enough that the genome size of the hybrid is significantly different from both parents (Jeschke et al., 2003). Tranel et al. (2002) determined that flow cytometry could be used to definitively identify hybrids between A. hybridus and A. tuberculatus made under laboratory conditions, and Trucco et al. (2005) demonstrated that flow cytometry could be used to detect such hybrids under controlled field conditions. However, under agronomic field conditions, not only are A. hybridus and A. tuberculatus present but other pigweeds are also sympatric. Thus, the potential of using genome size in assessing natural hybrids is dependent on the genome size of all the pigweeds in a given locale.
The purpose of this study was to determine the genome size of pigweed species found in Illinois. Flow cytometry was used to assess the genome size of several populations of each species to document what, if any, genome size variation exists among these species and, if it does exist, to determine if it is useful in identifying interspecific pigweed hybrids.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
In both experiments, the factor species and accessions were considered fixed. Accessions were nested within species, and two and three replications were used for experiment 1 and experiment 2, respectively. The linear model used in both experiments was
 | [1] |
i is the effect of ith species, ßj(i) is the effect of the jth accession within the ith species, and 
ijk is the error term. The statistical analysis was done with the MIXED procedure of SAS (SAS, 2003). Least significant differences were calculated based on appropriate degrees of freedom and standard errors for the differences of means. To compare data among studies, a third experiment was run. The two accessions of A. hybridus were analyzed. Two accessions of another species, A. retroflexus (MH 84 and an accession from RBG KEW) were also analyzed. Four plants of each accession were grown, and nuclei were isolated and examined as described above. W22 was used as an internal standard.
A fourth experiment comparing the genome size of A. tuberculatus and A. palmeri was performed. Seeds were planted and grown as previously described. After approximately 6 weeks, three leaves were taken from four plants per species. The leaves from each plant were analyzed on the flow cytometer, and genome size was determined as described previously. In addition, A. tuberculatus and A. palmeri leaves were chopped and ground together. The samples containing the mixed nuclei were analyzed as described previously. The differences between the two species' DNA contents were compared using ANOVA and LSD tests using the SAS statistical program.
The chromosome number was then determined for A. tuberculatus and A. palmeri. Root tips were rinsed with deionized water, placed in 8-hydroxyquinoline for 2 h, and fixed in Carnoy's solution. Root tips were obtained and stained with aceto-orcein, and root tip squashes were performed. A total of 12 spreads per species were analyzed for chromosome number. The chromosomes were counted using an Olympus BX61 microscope.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The genome size of the pigweed species ranged from 0.96 to 1.21 pg per 2C nucleus in experiment 1 and 0.95 to 1.23 pg per 2C nucleus in experiment 2 (Table 2). In both experiments, species were significantly different with respect to genome size (p < 0.001). Accessions within species were not significant at = 0.05. The order of genome size was similar between the experiments with the only exception being that A. blitoides had a slightly larger genome size than A. albus in experiment 1 and a slightly smaller genome size in experiment 2.
|
|
In the fourth experiment, cuttings from A. tuberculatus and A. palmeri were analyzed. There was no significant difference among plants within each species (Table 2). A statistically significant difference was observed between groups, with the overall average genome size of A. tuberculatus being 1.43 ± 0.01 per 2C nucleus, whereas the overall mean genome size of A. palmeri was 0.91 ± 0.01 per 2C nucleus (p < 0.0001). When co-chopped, the variation in genome size between the two species was 52% (Fig. 1)
. Both species were observed to have 32 chromosomes, with the chromosomes of A. palmeri appearing slightly smaller (Fig. 2)
.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
27% variation in genome size. Substantiating the reproducibility of the genome size data is that in both experiments, only one minor change in the order of the species was observed and that the actual picogram amounts were very similar. In addition, the coefficient of variation (CV) of the measured G1 peaks averaged 5.0%. This mean CV is higher than recommended by Rayburn (1993). Rayburn (1993) used maize and sorghum in describing the flow cytometric technique of DNA analysis. In the Amaranthus species, it is much more difficult to obtain CVs of 3%. However, Dole
el and Barto
(2005) recommended that in "difficult" plant species, CVs around 5.0% are acceptable. Although such genome size variation has been previously reported in some of the species examined in this study (Bennett et al., 2000), two major discrepancies occur between the previous study and the results presented in this study. As described by Rayburn (1993) and Greilhuber et al. (2005), inconsistency exists with respect to how genome sizes are reported. To facilitate comparisons among studies, genome sizes in this article refer to the amount of DNA found in G1 diploid cells and are expressed as the amount of pg per 2C nucleus.
The first discrepancy between the two studies is the total pg amounts reported for the species. Bennett et al. (2000) reported that A. retroflexus and A. spinosus had 1.7 and 1.9 pg per 2C nucleus, respectively. In the present study, A. retroflexus and A. spinosus were observed to have 1.22 and 1.07 pg per 2C nucleus, respectively. In addition, Bennett et al. (1998) reported that A. hybridus had 1.4 pg per 2C nucleus, whereas Tranel et al. (2002) and Jeschke et al. (2003) have reported A. hybridus to have 1.12 and 1.04 pg per 2C nucleus. One possible explanation of these differences is that the accessions observed by Bennett et al. (2000) were different from the accessions of the same species used in the U.S. studies and as such have more DNA per genome. To test this hypothesis, accessions of A. hybridus and A. retroflexus used by Bennett et al. (2000) were obtained from RGB KEW and analyzed in conjunction with accessions of the same two species used in this study. When the present study was repeated, the A. retroflexus used in the present study was found to have a genome size of 1.21 pg per 2C nucleus, consistent with the previous results of this study, whereas the accession obtained from RGB KEW was observed to have 1.16 pg per 2C nucleus, not the previously reported 1.7 pg (Table 2). The two species of A. hybridus had nearly identical genome sizes of 1.07 pg per 2C nucleus, again not consistent with the previously reported 1.4 pg per 2C nucleus. Thus, in both cases, the pigweed species genome sizes seem to be overestimated in Bennett et al. (2000).
One possible explanation of these observations is that in Tranel et al. (2002), in Jeschke et al. (2003), and in the present study, the maize line W22 was used as the internal standard. W22 is an inbred line of maize that has been reported to have 5.35 pg of DNA per 2C nucleus (Biradar and Rayburn, 1993). Bennett et al. (2000) used multiple standards, and the exact standard used in the pigweed determinations is not clear. Critical to absolute genomes size determinations is the choice of the reference standard (Johnston, 1999). The reference standard should be well defined. Because all the studies that reported the lower genome size for the pigweed species used W22 as a standard, chicken red blood cell nuclei were used to ensure that the 5.35 pg of W22 was correct. Upon comparing the chicken nuclei with Xenopus laevis nuclei defined as having 6.3 pg per 2C nucleus (Freeman and Rayburn, 2004), the chicken nuclei were determined to have a 2C genome size of 2.5 pg (data not shown), which is within the reported range of chicken nuclei (Johnston, 1999). The recalculated genome size of W22 using chicken red blood cells was 5.33 pg, which is nearly identical to the 5.35 pg previously reported. To obtain the numbers observed by Bennett et al. (2000), the chicken nuclei would have to have had a genome size of 4 pg, which would be well outside the reported range of chicken genome size and the newly recalibrated chicken nuclei used in this study. Therefore, the lower values for the pigweed species reported here seem to be correct. However, absolute genome size is not the only incongruity between the two studies.
According to Bennett et al. (2000), A. spinosus was reported to have the largest genome size of the pigweed species examined (1.9 pg per 2C nucleus). In this study, A. spinosus was observed to have one of the smallest genome sizes (1.07 pg per 2C nucleus). This is not a result of an incorrect standard. A miscalculated standard may result in an error of absolute DNA content, but the overall relative ranking of the species would be maintained. Therefore, another problem must exist. Another difference between the Bennett et al. (2000) study and this study is the method used to calculate genome size. Bennett et al. (2000) used microdensitometry on nuclei from fixed root tips, whereas flow cytometry of leaf tissue was used in the present study. Although both methodologies are extensively used, they require extreme care due to technical errors that can occur during analysis (Greilhuber, 1997; Price et al., 2000).
The major concern when using flow cytometry is the presence of fluorescence inhibitors (FIs) (Price et al., 2000). Although an internal standard was used during the flow cytometric analysis to compensate for FIs, the lower values observed in this study may cause concern that this compensation is not complete. However, Tranel et al. (2002), using flow cytometric analysis, observed that the F1 hybrid between two pigweed species had an intermediate genome size between the two parental genome sizes. The probability that in the F1 hybrid the exact amount of FIs would be present that would interact with the intermediate DNA present in the hybrid to give a fluorescence almost exactly half way between the two parent fluorescence intensities is low. Also, given the fact that an internal standard was used, such an exact interaction would be relatively rare. One might suspect that the maize internal standard used in this study and in Tranel et al. (2002) could react differently than pigweed nuclei to FIs and thus produce abnormal results. The co-chopping of the A. tuberculatus and A. palmeri leaves gave the same relative difference between the two species. Thus, the difference seems real and is not a result of using a maize internal control.
The lower estimate of genome size in A. spinosus in this study is also supported by recent phylogenic data. One might expect that the two dioecious species would be more closely related and therefore the genome sizes of these two species to be similar. However, recent reports have suggested that A. palmeri (a dioecious species) and A. spinosus (a monoecious species) may be closely related. Franssen et al. (2001) noted that the pollen morphology of A. palmeri was more similar to A. spinosus than any other pigweed species. Wasson and Tranel (in press) using AFLP analysis observed that phylogenetic analysis and principal coordinate analysis resulted in a clustering of A. palmeri, with A. spinosus indicating close genetic similarities. Given the apparent close genetic relationship and almost identical pollen morphology, the observation that these species would have similar genome sizes is not totally unexpected. Thus, the low genome size observed in A. spinosus is real and is not an artifact of the methodology.
Given the wide range of genome sizes found in the Illinois pigweed species (50%), the initial appearance is that genome size would not be effective in identifying hybrids between pigweed species. For instance, if A. retroflexus would hybridize with A. spinosus, the expected genome size of the hybrid would be 1.13 pg per 2C nucleus, similar to the average genome size of A. powelli, which is 1.13 pg per 2C nucleus. Therefore, genome size would be of little use for the random screening of F1 pigweed hybrids in the field. However, an interesting dichotomy exists with respect to reproduction among the pigweed species. A. tuberculatus and A. palmeri are dioecious species, whereas the remaining pigweeds in Illinois are monoecious. The monoecious plants have a propensity to self-fertilize due to the proximity of the male flowers to the female flowers (Franssen et al., 2001). In addition, the pollen of the monoecious species does not seem to be as aerodynamic as the pollen from dioecious species. Therefore, interspecific hybridization among monoecious pigweeds would be relatively limited.
With interspecific hybridization being skewed toward dioecious by dioecious or dioecious by monoecious species, the use of genome size to identify interspecific hybrids becomes apparent. For example, if the two dioecious species sympatric in Illinois, A. tuberculatus and A. palmeri, were to hybridize, the F1 hybrid would theoretically have an intermediate DNA content between the two species. Interestingly, A. palmeri was observed in this study to have the lowest genome size of the weedy Amaranthus, whereas Tranel et al. (2002) and Jeschke et al. (2003) observed A. tuberculatus to have a genome size of 1.4 per 2C nucleus. Thus, this hybrid represents the hybridization of the species with the widest range of genome size. To ensure that the 1.4 pg per 2C nucleus estimate for A. tuberculatus was correct and that the genome sizes between the two dioecious species were that distinct, the genome size of A. tuberculatus and A. palmeri were reassessed. The genome size of A. palmeri was as expected from the previous experiments, whereas A. tuberculatus had 1.4 pg per 2C nucleus. Thus, any potential hybrid between these two species should have 1.15 pg per 2C nucleus. This genome size is similar to the genome size of several pigweed species; however, because this cross is between to two dioecious parents, the hybrid would also be dioecious. Because the remaining species are monoecious, the hybrid would be morphologically distinct and easily distinguished from the surrounding species.
In interspecific hybridization between dioecious and monoecious species, an interesting phenomenon occurs. Murray (1940) reported that, in dioecious by monoecious pigweed crosses, "dioeciousness is epistatic to monoeciousness." Simply put, F1 hybrids between monoecious and dioecious pigweeds are dioecious. Therefore, any dioecious pigweed plant that has a genome size of less than 1.3 pg of DNA per 2C nucleus and more than 1.1 pg of DNA per 2C nucleus is a potential hybrid. Although the exact parentage may be unknown, the amount of pigweed-interspecific hybridization in a given field population may be estimated by observing the number of dioecious pigweeds with intermediate genome sizes.
In conclusion, genome sizes in pigweed species found in Illinois range from 0.95 to 1.4 pg of DNA per 2C nucleus. The two species at the extreme range of the spectrum are the two dioecious species. Because both species have 32 chromosomes, this variation is due to intrachromosomal variation. Given the reproductive biology of the Amaranthus species, the dominance of dioeciousness, and the wide range of genome size variation between the dioecious species, genome size determinations have the potential to be useful in addressing questions concerning the amount of interspecific hybridization within the genus Amaranthus under field conditions.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 23, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
el, J., and J. Barto. 2005. Plant DNA flow cytometry and estimation of nuclear genome size. Ann. Bot. 95:99–110.el, M.A. Lysák, and M.D. Bennett. 2005. The origin, evolution, and proposed stabilizaton of the terms ‘genome size’ and ‘C-value’ to describe nuclear DNA contents. Ann. Bot. 95:255–260.This article has been cited by other articles:
|
|
 |
|
  T. A. Gaines, W. Zhang, D. Wang, B. Bukun, S. T. Chisholm, D. L. Shaner, S. J. Nissen, W. L. Patzoldt, P. J. Tranel, A. S. Culpepper, et al. Gene amplification confers glyphosate resistance in Amaranthus palmeri PNAS, January 19, 2010; 107(3): 1029 - 1034. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
