Vol.:(0123456789)1 3

Genetica (2017) 145:235–240 DOI 10.1007/s10709-017-9955-0

SHORT COMMUNICATION

Meiotic pairing as an indicator of genome composition in polyploid prairie cordgrass (Spartina pectinata Link)

Jeffrey W. Bishop1 · Sumin Kim1 · María B. Villamil1 · D. K. Lee1 · A. Lane Rayburn1 

Received: 11 July 2016 / Accepted: 31 January 2017 / Published online: 27 February 2017 © Springer International Publishing Switzerland 2017

Introduction

The neopolyploidization of prairie cordgrass (Spartina pectinata Link) has presented a unique opportunity to study the genetic evolution of Spartina, a genus comprised exclusively of polyploid species. Prairie cordgrass is a rhizomatous, perennial C4 grass native to North Amer-ica, with a distribution extending from southern Texas to northern coastal regions of Canada (USDA-NRCS 2002). Natural populations of prairie cordgrass exist as tetraploid (2n = 4x = 40), hexaploid (2n = 6x = 60), and octoploid (2n = 8x = 80) cytotypes, with the tetraploids and octop-loids well established in the United States (Church 1940; Reeder 1977; Kim et  al. 2012b). The hexaploid cytotype currently has only been observed in a single location in central Illinois in immediate proximity to tetraploid cyto-types, and because the polyploidization event generating the hexaploid cytotype occurred recently, they are consid-ered neohexaploids (Kim et al. 2012b).

Polyploidy is important in prairie cordgrass because it is known to affect plant phenotype. Kim et al. (2012a) reports increased numbers of chromosomes in hexaploid prai-rie cordgrass corresponding to larger stomata size, larger spikelets, increased tiller mass, delayed flowering time, and taller plants in comparison to tetraploid prairie cordgrass. Differences in phenotype may explain why hexaploid plants are observed to be competitive with the tetraploid plants. Given the proximity between the tetraploid and hexaploid cytotypes in central Illinois, it is hypothesized that the tetraploid produced the hexaploid through the union of a reduced tetraploid gamete and an unreduced tetraploid gamete (Kim et  al. 2012a). Study of chromosome pairing in tetraploid, hexaploid, and octoploid populations may provide valuable information in testing this hypothesis and assessing the genome compositions of all three cytotypes.

Abstract The existence of neopolyploidy in prairie cordgrass (Spartina pectinata Link) has been documented. The neohexaploid was discovered coexisting with tetra-ploids in central Illinois, and has been reported to exhibit competitiveness in the natural environment. It is hypoth-esized that the natural tetraploid cytotype produced the hexaploid cytotype via production of unreduced gametes. Meiosis I chromosome pairing was observed in tetraploid (2n = 4x = 40), hexaploid (2n = 6x = 60), and octoploid (2n = 8x = 80) accessions and the percentage of meiotic abnormality was determined. Significant differences in meiotic abnormality exist between tetraploid, hexaploid, and octoploid cytotypes. An elevated incidence of abnor-mal, predominantly trivalent pairing in the neohexaploid suggests that it may possess homologous chromosomes in sets of three, in contrast to the tetraploid and octoploid cytotypes, which likely possess homologous chromosomes in sets of two. Abnormal chromosome pairing in the hexa-ploid may result in unequal allocation of chromosomes to daughter cells during later stages of meiosis. Chromo-some pairing patterns in tetraploid, hexaploid, and octop-loid cytotypes indicate genome compositions of AABB, AAABBB, and AABBA′A′B′B′, respectively.

Keywords Chromosomes · Homology · Polyploidy · Meiosis

* A. Lane Rayburn arayburn@illinois.edu

1 Department of Crop Sciences, University of Illinois, 1102 S. Goodwin Ave, Urbana, IL 61801, USA

236 Genetica (2017) 145:235–240

1 3

Examination of chromosome pairing patterns is highly useful for determining genome composition as it provides accurate data regarding chromosome homology (Heyne 1987). This method has been used to analyze genome composition in species of Triticum (Heyne 1987) and Ley-mus (Wang and Jensen 1994). In this study, a cytogenetic approach to examine meiotic division during microsporo-genesis was utilized.

Materials and methods

Prairie cordgrass (Spartina pectinata Link) plants were dis-covered inhabiting a drainage ditch of farmland in Cham-paign County, IL (Graves et al. 2016). In this location, three areas were identified in which hexaploid and tetraploid plants grew in close proximity. Each of the hexaploid and tetraploid plants were sampled and transplanted to a field plot in 2012, arranged in a completely randomized design with three replications of each accession. An additional hexaploid accession (99C) derived from a tetraploid acces-sion (99A) was included in the field plot (Kim et al. 2010). The tetraploid accessions include 1034x, 1094x, 99A, and MBB4x and the hexaploid accessions include 1036x, 99C, MID6x, and MBB6x (Kim et al. 2013). Geographically and phylogenetically unrelated tetraploids (19–103 and 40–101) and octoploids (19–108 and PCG109) were also used in this study, and were established in separate field plots (Kim et al. 2013).

Prairie cordgrass spikelets were collected from each accession during five different stages of development. The early stage was characterized by a fully collared, upright flag leaf and a swollen flag leaf sheath. Boot stage was defined by emergence of the first spikelet. The heading stage was marked by visibility of the inflorescence pedun-cle, and stigma and anther stages were characterized by emergence of stigmas and anthers, respectively. During each stage, one inflorescence per plant was collected from at least three plants per accession and fixed in a Carnoy solution of ethanol, chloroform, and acetic acid in a ratio of 6:3:1 and stored at 4 °C. After 24  h the spikelets were removed from the Carnoy solution and stored in 70% etha-nol at 4 °C. Anthers were screened for meiosis I using 1% propionic carmine stain, and slides were viewed at a total magnification of 400× using phase contrast microscopy. Florets containing anthers undergoing meiotic diakinesis and metaphase I were immersed in alcoholic hydrochloric acid-carmine stain (Snow 1963) for a minimum of 7 days, after which chromosome squashes were prepared in 45% acetic acid. Between one and three florets were observed for each plant, and at least one plant was observed for each accession. Slides were viewed using an Olympus BX61 microscope at a total magnification of 1000x through a

phase contrast objective lens. Images were captured using an Olympus U-CMAD3 camera.

Chromosome pairing was considered normal if all chromosomes in a cell were clearly visible and paired as bivalents. The presence of multivalent chromosomes was the primary criterion for identifying abnormal chromo-some pairing. Other factors such as chromosome failure to align on the central plane during metaphase I or failure to migrate to the cell poles during anaphase I and telophase I also indicated abnormal pairing. After scoring meiotic cells as normal or abnormal, meiotic index (MI, %) was calcu-lated following Love (1951). However, the calculation of meiotic index was modified from Love (1951) so that indi-vidual cells were scored rather than quartets. The meiotic index was calculated as follows:

Meiotic Index (MI)% = (Number of normal cells × 100)

÷ (Total number of cells scored)

Statistical analysis

Mean of the MI for the accessions was analyzed using PROC GLIMMIX in SAS 9.4 (SAS Institute Inc., Cary, NC) and mean comparisons were obtained with the PDIFF option of the LSMEANS statement setting the probabil-ity of Type I error or alpha level (α) at 0.05. A total of 33 hexaploid meiocytes were analyzed intensively, in which the number of univalent, bivalent, trivalent, and higher multivalent associations were counted. The total number of each association type was calculated across all meiocytes, and a Chi-Squared test was employed to test for independ-ence between trivalent and higher multivalent frequency.

Results and discussion

Tetraploid and octoploid cytotypes showed normal biva-lent chromosome pairing during diakinesis and metaphase I (Figs.  1a, b, 2a, b), in agreement with Marchant (1968) and Reeder (1977). The current study expands upon previ-ous research, as chromosome segregation and cell division during later stages of meiosis I was evaluated in addition to chromosome pairing. Equal segregation of chromosomes during anaphase I and the generation of two identical nuclei during telophase I was observed (Figs. 1c, d, 2c, d). High values for the Meiotic Index of tetraploid and octoploid accessions reflect the observed meiotic normality (Table 1).

The study of chromosome pairing in the tetraploid cytotype provides information about chromosome homol-ogy (Jackson 1982; Heyne 1987), which may indicate whether the tetraploid is autopolyploid or allopolyploid. In autopolyploids, the presence of more than two homol-ogous chromosomes enables chromosomes to pair in multivalent associations (Ramsey and Schemske 2002).

237Genetica (2017) 145:235–240

1 3

Over time, autopolyploids may undergo the process of diploidization, after which homologous chromosomes no longer pair in multivalent associations and only pair in bivalents (Attia et al. 1987; Ramsey and Schemske 2002; Doyle et  al. 2008; Doyle and Egan 2010). It is possible that the tetraploid cytotype is autopolyploid and that dip-loidization has occurred, explaining bivalent chromo-some pairing. However, previous analysis of Waxy gene sequences has indicated that tetraploid S. pectinata con-tains two divergent copies (A and B) of the Waxy gene (Fortune et  al. 2007). One hypothesis presented by For-tune et  al. (2007) for this observation suggests that the tetraploid cytotype is an allopolyploid containing two distinct genomes (AABB), rather than an autopolyploid (AAAA). Chromosomes in the same genome would therefore pair in bivalents due to homology, and segre-gate normally during meiosis. These outcomes were observed in the current study. Furthermore, recent gen-otyping-by-sequencing (GBS) and Kompetitive allele-specific PCR (KASP) analyses indicated that disomic inheritance occurs in tetraploid S. pectinata (Graves et al.

2015; Crawford et  al. 2016), supporting the hypothesis that the tetraploid is allopolyploid (AABB).

The hexaploid cytotype in this study demonstrated a complete departure from normal chromosome pairing and segregation, with a Meiotic Index ranging from 0.0 to 0.6 (Table  1). The majority of microspores showed multiva-lent chromosome pairing to varying degrees as depicted in Fig.  3a, b, though the formation of trivalent chromo-somes was observed most often. Occasionally, groupings of four or more chromosomes were documented, as shown in Fig. 3b. There was also a high occurrence of univalent chromosomes, which often failed to align in the center of the cell during metaphase I (Fig.  3a, b) and subse-quently may not migrate to the cell poles during anaphase I (Fig. 3c). Finally, chromosome pairs failed to migrate to cell poles during anaphase I and telophase I (Fig.  3c, d). These abnormalities may result in either unequal segrega-tion or complete loss of chromosomes during cytokinesis.

The hexaploid cytotype of prairie cordgrass is hypoth-esized to have originated from the tetraploid cytotype through the union of an unreduced gamete with a reduced gamete (Kim et  al. 2012a). An allopolyploid genome

Fig. 1 Normal tetraploid cells: a diakinesis showing 20 normal bivalents. Example of normal bivalent in circle. b Metaphase I showing 20 normal bivalents. Example of normal bivalent in circle. c Anaphase I showing normal segregation. d Telophase I showing normal cytokinesis. Total magnification is 1000×, bar indicates 10 µm

238 Genetica (2017) 145:235–240

1 3

composition of AABB in the tetraploid, as suggested, would lead to an allopolyploid genome composition of AAABBB in the hexaploid. This means that it would have three homologous chromosomes, so chromosomes should mostly be present as univalents, bivalents, or trivalents, with occasional higher multivalents due to homoeologous pairing (Attia et  al. 1987; Ramsey and Schemske 2002;

Doyle et al. 2008; Doyle and Egan 2010). In this study, uni-valent, bivalent, and trivalent associations did occur most frequently, with higher multivalents occurring at a sig-nificantly lower rate than trivalents (χ2 = 139.2, p = 0.01) (Fig. 4). These data support the hypothesis that the hexa-ploid is AAABBB.

Fig. 2 Normal octoploid cells: a diakinesis showing 40 normal bivalents. Example of normal bivalent in circle. b Metaphase I showing 40 normal bivalents. Example of normal bivalent in circle. c Anaphase 1 showing normal segregation. d Telophase I showing normal cytokinesis. Total magnification is 1000×, bar indicates 10 µm

Table 1 Percent abnormality of accessions and ploidy levels, and meiotic index of accessions

† Significant differences between accessions (P < 0.05) were tested. Different letters indicate significant dif-ferences

Accession Ploidy No. of Plants No. of Cells Abnormality (%) Meiotic index†

1034x 4x 2 819 0.8 99.0A

1094x 4x 1 951 0.4 99.6A

19–103 4x 2 1400 0.1 99.9A

40–101 4x 1 844 0.0 100.0A

99A 4x 2 465 15.2 85.1B

MBB4x 4x 2 1281 0.9 99.0A

1036x 6x 1 559 100.0 0.0C

MID6x 6x 2 595 100.0 0.0C

99C 6x 1 850 99.2 0.6C

MBB6x 6x 1 636 100.0 0.0C

19–108 8x 2 1374 0.9 99.1A

PCG109 8x 2 975 0 100.0A

239Genetica (2017) 145:235–240

1 3

The tetraploid cytotype likely gave rise to the octoploid cytotype as well (Krahulcová and Krahulec 2000; Levin 2002). A tetraploid genome composition of AABB would lead to an octoploid genome composition of AAAABBBB. This means that there are four homologous chromosomes,

and therefore chromosomes could pair together as univa-lents, bivalents, trivalents, or quadrivalents (Attia et  al. 1987; Ramsey and Schemske 2002; Doyle et  al. 2008; Doyle and Egan 2010). Only bivalent pairing was docu-mented in this study. There may be a genetic mechanism causing homologous chromosomes to pair in a bivalent fashion, similar to mechanisms documented in other grass species (Rajhathy 1971; Evans and Macefield 1972). However, such a mechanism would likely be present in the hexaploid as well. There is no evidence that this type of mechanism is present in the hexaploid, considering the high number of univalent and multivalent chromosomes observed.

A more plausible explanation for bivalent chromosome pairing in the octoploid is an origin from two divergent tetraploid plants that hybridized. This hybridization event (AABB × A′A′B′A′), followed by chromosome-doubling, would create an octoploid with a genome composition of AABBA′A′B′B′. Bivalent chromosome pairing would then occur based on homology. This final postulation regarding the octoploid origin completes the hypothesis of allopoly-ploidy in all three cytotypes. Therefore, based upon chro-mosome pairing data and previous phylogenetic data, there is evidence that the tetraploid is AABB, the hexaploid is AAABBB, and the octoploid is AABBA′A′B′B′.

Fig. 3 Abnormal hexaploid cells: Metaphase I containing (a) univalents (U), bivalents (B), and trivalents (T). (b) Meta-phase I containing univalents (U), bivalents (B), and higher order multivalents (M—tri-valents and higher pairing structures). (c) Anaphase I containing 6 lagging chromo-somes indicated by arrows. (d) Anaphase I containing 2 lagging chromosomes indicated by arrows. Total magnification is 1000×, bar indicates 10 µm

Fig. 4 Total chromosome associations across hexaploid meiocytes

240 Genetica (2017) 145:235–240

1 3

Acknowledgements This work was supported by funding from the North Central Regional Sun Grant Center at South Dakota State Uni-versity through a grant provided by the United States Department of Agriculture (Award Number 2010-38502-21861) and by the National Institute of Food and Agriculture, U.S. Department of Agriculture, Hatch project under ILLU-802-952.

Compliance with ethical standards

Conflict of interest The authors declare that they have no conflict of interest.

References

Attia T, Busso C, Röbbelen G (1987) Digenomic triploids for an assessment of chromosome relationships in the cultivated diploid Brassica species. Genome 29:326–330

Church GL (1940) Cytotaxonomic studies in the Graminae Spartina, Andropogon, and Panicum. Am J Bot 27:263–271

Crawford J, Brown PJ, Voigt T, Lee DK (2016) Linkage map-ping in prairie cordgrass (Spartina pectinata Link) using gen-otyping-by-sequencing. Mol Breeding 36:62. doi:10.1007/s11032-016-0484-9

Doyle JJ, Egan AN (2010) Dating the origins of polyploidy events. New Phytol 186:73–85

Doyle JJ, Flagel LE, Paterson AH, Rapp RA, Soltis DE, Soltis PS, Wendel JF (2008) Evolutionary genetics of genome merger and doubling in plants. Annu Rev Genet 42:443–461

Evans GM, Macefield AJ (1972) The suppression of homoeologous pairing by B-chromosomes in a Lolium species hybrid. Nature 236:110–111

Fortune PM, Schierenbeck KA, Ainouche AK, Jacquemin J, Wendel JF, Ainouche ML (2007) Evolutionary dynamics of Waxy and the origin of hexaploid Spartina species (Poaceae). Mol Phylo-genet Evol 43:1040–1055

Graves H, Rayburn AL, Gonzales-Hernandez JL, Nah G, Kim D, Lee DK (2015) Validating DNA polymorphisms using KASP assay in prairie cordgrass (Spartina pectinata Link) populations in the US. Front Plant Sci 6:1271. doi:10.3389/fpls.2015.01271

Graves H, Rayburn AL, Kim S, Lee DK (2016) Chloroplast DNA var-iation within prairie cordgrass (Spartina pectinata Link) popula-tions in the US. J Syst Evol 54(2):104–112

Heyne EG (1987) Wheat and wheat improvement. American Society of Agronomy, Inc.; Crop Science Society of America, Inc.; Soil Science Society of America, Inc; Madison, WI, USA

Jackson RC (1982) Polyploidy and diploidy: new perspectives on chromosome pairing and its evolutionary implications. Am J Bot 69(9):1512–1523

Kim S, Rayburn AL, Lee DK (2010) Genome size and chromosome analyses in prairie cordgrass. Crop Sci 50:2277–2282

Kim S, Rayburn AL, Boe A, Lee DK (2012a) Neopolyploidy in Spar-tina pectinata Link: 1. Morphology analysis of tetraploid and hexaploid plants in a mixed natural population. Mol Phylogenet Evol 298:1073–1083

Kim S, Rayburn AL, Parrish A, Lee DK (2012b) Cytogeographic dis-tribution and genome size variation in prairie cordgrass (Spar-tina pectinata Bosc ex Link). Plant Mol Biol Rep 30:1073–1079

Kim S, Rayburn AL, Voigt TB, Ainouche ML, Ainouche AK, Lee DK (2013) Chloroplast DNA intraspecific phylogeography of prairie cordgrass (Spartina pectinata Bosc ex Link). Plant Mol Biol Rep 31:1376–1383

Krahulcová A, Krahulec F (2000) Offspring diversity in Hieracium subgen. pilosella (Asteraceae): new cytotypes from hybridiza-tion experiments and from open pollination. Fragm Flor Geobot 45:239–255

Levin DA (2002) The role of chromosomal change in plant evolution. Oxford University Press, Oxford

Love RM (1951) Varietal differences in meiotic chromosome behav-ior of Brazilian wheats. Agron J 43:72–76

Marchant CJ (1968) Evolution in Spartina (Graminae) III. Species chromosome numbers and their taxonomic significance. J Linn Soc Bot 60(383):411–417

Rajhathy T (1971) The allopolyploid model in Avena. In: G. Kimber, G.P. Redei (eds). Stadler Symposia 3:71–87

Ramsey J, Schemske DW (2002) Neopolyploidy in flowering plants. Annu Rev Ecol Evol Syst 33:589–639

Reeder J (1977) Chromosome numbers in western grasses. Am J Bot 64(1):102–110

Snow R (1963) Alcoholic hydrochloric acid-carmine as a stain for chromosomes in squash preparations. Stain Technol 38:9–13

United States Department of Agriculture, Natural Resources Con-servation Service (2002) Prairie cordgrass plant guide. Natural Resources Conservation Service, Washington, DC. http://plants.usda.gov/plantguide/pdf/pg_sppe.pdf. Accessed 13 Feb 2014

Wang RRC, Jensen KB (1994) Absence of the J genome in Leymus species (Poaceae: Triticeae): evidence from DNA hybridization and meiotic pairing. Genome 37:231–235