Microsporogenesis, viability and morphology of pollen grain in accessions of Cynodon L. C. Rich. (Poaceae)

Microsporogenesis, viability and morphology of pollen grain in accessions of Cynodon L. C. Rich. (Poaceae)

South African Journal of Botany 118 (2018) 260–267 Contents lists available at ScienceDirect South African Journal of Botany journal homepage: www.e...

2MB Sizes 0 Downloads 5 Views

South African Journal of Botany 118 (2018) 260–267

Contents lists available at ScienceDirect

South African Journal of Botany journal homepage: www.elsevier.com/locate/sajb

Microsporogenesis, viability and morphology of pollen grain in accessions of Cynodon L. C. Rich. (Poaceae) D.M. Silva a, Y.D. Santos a, F.R.G. Benites b, V.H. Techio a,⁎ a b

Federal University of Lavras – UFLA, Lavras, Minas Gerais, Brazil Embrapa Gado de Leite, Juiz de Fora, Minas Gerais, Brazil

a r t i c l e

i n f o

Article history: Received 7 March 2018 Received in revised form 11 July 2018 Accepted 16 July 2018 Available online xxxx Edited by Johannes Van Staden Keywords: Forage grasses Pollen grain Meiosis Bermuda grass Star grass

a b s t r a c t Plants of the genus Cynodon were considered as weed for a long period. However, this scenario has changed when it was observed that many species could be used in forage farming and for the recovery of degraded areas. Due to the high expansive potential, Cynodon spp. became an option for producers and aroused interest for breeding programs, besides requiring further studies on genetic and cytogenetic variability. Research on meiosis in Cynodon are scarce since the inflorescences are small and the standardization of the stages of spikelet collections is difficult. The aim of the present study was to evaluate the microsporogenesis, viability and morphology of the pollen grains of two accessions of Cynodon dactylon var. dactylon (L.) Pers. (2n = 4x = 36, bermudagrass) and two accessions of Cynodon nlemfuensis var. nlemfuensis Vanderyst (2n = 2x = 18, stargrass). The slides were prepared by squash technique and stained with 2% carmine propionic. For the viability analyses of pollen grains, 2% propionic carmine and Alexander's stain were used. The measurements and the morphology of the pollen grains were defined from analyses with the acetolysis technique and scanning electron microscopy. The meiosis of the four accessions/species was regular. The staining tests showed variations in the viability rate of the pollen grain between the diploid and tetraploid species, being higher among the tetraploids. The pollen grains measured 21.1 and 28.8 μm on average for both species and were classified as oblate spheroidal, monocolpates and showed no ornamentation, such as spicules and other attachments with non-roughened, micro reticulated exine. © 2018 SAAB. Published by Elsevier B.V. All rights reserved.

1. Introduction The genus Cynodon L. C. Rich. belongs to the Poaceae family and their species show significant economic value due to their use as forage for animal feeding, erosion protection and sports field coverage (Neiva et al., 1999; Taliaferro et al., 2004; Wu and Taliaferro, 2009). The plants from this genus are grouped in Bermuda grasses containing stolons and rhizomes, and in Star grasses, showing only stolons, leaves and stems larger than those of Bermuda grasses (Harlan, 1970; Andrade et al., 2009). They are perennial grasses from tropical and subtropical climates, with wide distribution in Africa, but also present in South America (Caro and Sánchez, 1969; Harlan et al., 1970; Filgueiras and Valls, 2015). Cynodon dactylon (L.) Pers. belongs to the Bermuda grass group and is the most important species of the genus due to its prevalence, wide geographic distribution and its morphological variants, such as var. afghanicus; var. aridus; var. dactylon; var. coursii; var. elegans and var. polevansii. In the dactylon variety, three races are found: tropical, ⁎ Corresponding author at: Federal University of Lavras, Department of Biology, 37200000 Lavras, Minas Gerais, Brazil. E-mail address: [email protected]fla.br (V.H. Techio).

https://doi.org/10.1016/j.sajb.2018.07.026 0254-6299/© 2018 SAAB. Published by Elsevier B.V. All rights reserved.

temperate and seleucidus (Wu and Taliaferro, 2009). Cynodon nlemfuensis Vanderyst is the most promising of the star grass group since it has favorable characteristics as forage and is genetically related to C. dactylon (Wu and Taliaferro, 2009). Until the mid-1940s, Cynodon plants were considered as weeds due to their high adaptability, growth, and establishment in the environment. Still in 1950, farmers tried to control it by several methods, such as weeding, animal control and chemical control (Horowitz, 1996). However, according to Wu and Taliaferro (2009), important characteristics observed in this genus have aroused the interest in developing cultivars for use in forage farming, mainly due to the resistance to cold, adaptation to the temperate climate and grazing efficiency. When considering the high expansive potential, Cynodon spp. has become an option for agricultural producers (Neiva et al., 1999), since grazing is a low-cost source of feed (Primavesi et al., 2004). In Brazil, plants develop satisfactorily even in soils with low fertility (Sousa et al., 1998). Despite the importance, the numbers of species to the genus is still controversial, since there are studies reporting the occurrence of 20 species (Caro and Sánchez, 1969); 9 species and 10 varieties (Harlan et al., 1970); 10 species (Clayton et al., 2009) and, more recently, 13 species (“The Plant List, 2016”). In addition, cytogenetic studies for this

D.M. Silva et al. / South African Journal of Botany 118 (2018) 260–267

group are still scarce, particularly summarizing to chromosome counts and determining ploidy levels, ranging from diploid to hexaploid and 16 to 54 chromosomes (Avdulow, 1931; Taliaferro et al., 2004; Dhaliwal and Gupta, 2011), and basic chromosome number x = 9 (Avdulow, 1931; Forbes and Burton, 1963) and x = 8 (Dhaliwal and Gupta, 2011). Studies of meiosis for the genus are also scarce and some very old (Forbes and Burton, 1963; Gupta and Srivastava, 1970; Hanna and Burton, 1977; Brilman et al., 1982; Dhaliwal and Gupta, 2011). This type of study becomes an important ally to understand and follow the whole process for the formation of gametes and confirm the basic number of chromosomes. In addition, the monitoring of post-meiotic events, which result in the formation of pollen grain, is also fundamental, since they allow advancing in knowledge applied to taxonomy, palynology and breeding programs. Based on the above, the aim of the present study was to evaluate the microsporogenesis, viability, and morphology of the pollen grains of four accessions of Cynodon.

2. Material and methods 2.1. Plant material The evaluations were performed in four accessions of Cynodon (Table 1), provided by Embrapa Gado de Leite, Juiz de Fora, Minas Gerais State, Brazil, which are originated from the active germplasm bank of the United States Department of Agriculture-USDA, Tifton, GA, USA.

2.2. Evaluation of meiosis and pollen grain viability The spikelets were collected between 9 a.m. and 4 p.m., with temperatures varying between 25 °C and 35 °C and fixed in solution of ethyl alcohol solution: acetic acid: propionic acid (6:3:2) stored in microtubes at − 4 °C. For meiotic analysis, the spikelets were excised under a stereoscopic microscope to extract the anthers, which were cut and macerated on the slide with a drop of 45% acetic acid. Subsequently, the slides were prepared by squash technique (Guerra and Souza, 2002) and stained with 2% propionic carmine. Prior to the fixation step, the size of the spikelets was measured using a caliper, and subsequently the size of each spikelet/slides was related to the phases of meiosis observed in each slides assessed. For pollen viability, the procedure was similar to the meiosis, but two dyes were used: 2% propionic carmine and 2% Alexander's stain. A total of 200 grains of pollen/slide and five slides per dye were evaluated. For staining with propionic carmine, the pollen grains that showed highly stained red cytoplasm were considered as viable and the unstained as unviable. For Alexander's stain, purple-stained grains were considered as viable and weakly green-stained as unviable. The slides were evaluated on a light microscope (Carl Zeiss, AxioLabA1), equipped with a microcamera (AxioCam ICc1), for Table 1 Identification of the species, code/accession and chromosome number of the evaluated plants. Species

Code

Accession

Chromosome numbera

Cynodon dactylon var. dactylon Cynodon dactylon var. dactylon

EGL 9 EGL 10 ERX 7

PI 224141-29 PI 29117102 PL 18

2n = 4x = 36 2n = 4x = 36

b

2n = 2x = 18

b

2n = 2x = 18

Cynodon nlemfuensis var. nlemfuensis Cynodon nlemfuensis var. nlemfuensis a b

ERX 21

Chiavegatto et al., 2016. Plants originated from breeding program, developed by Embrapa Dairy Cattle.

261

capturing the images. Scott-Knott statistical test (p b 0.05) was applied using the SISVAR software (Ferreira, 2003) in pollen viability analyzes. 2.3. Evaluation of the pollen grain through acetolysis technique The employed technique followed the recommendations of Erdtman (1960) and modified according to Melhem et al. (2003). The collected spikelets were macerated using a needle in a microtube containing 45% acetic acid and centrifuged for 10 min at 2500 rpm. Subsequently, the supernatant was removed and replaced with distilled water, centrifuged again for 10 min at the same speed and discarded at the end. Then, 1 mL of acetolysis solution (1 sulfuric acid: 9 pure acetic anhydride) was placed in a water bath at 87 °C for 2 min. The material was then centrifuged twice, removing the supernatant after each centrifugation. After the last centrifugation, the supernatant was replaced with 50% glycerinated water. The material was stored in a microtube at 10 °C for 24 h until the preparation time of the slide. The glycerinated water was discarded and the material was placed in slides with portions of glycerinated gelatin (100 mL of distilled water, 100 mL of glycerin, 17 g of colorless gelatin and 1 g of phenol) subjected to heating in a heater plate and covered with coverslips. The slides were evaluated on a light microscope (Carl Zeiss, AxioLabA1), equipped with a microcamera (AxioCam ICc1), for capturing the images. For each accession, five slides were evaluated and the measurements of the polar axis (P), equatorial axis (E) and exine thickness were performed in 10 pollen grains per slide using ImageJ software version 1.44 (Research Services Branch, U.S. National Institutes of Health, Bethesda, MD, USA.). Pollen grains were classified according to Erdtman (1943). Subsequently, the data were submitted to the Scott-Knott test (p b 0.05) using the SISVAR software (Ferreira, 2003). 2.4. Analysis of pollen grain ultrastructure through scanning electron microscopy To observe the surface of the pollen grain, the samples were fixed in Karnovsky solution, until the moment of analyses. Samples were washed in 0.05 M cacodylate buffer (three times - 10 min each) and post fixed in 1% osmium tetroxide for 4 h at room temperature. Then, they were progressive dehydrated in acetone (25%, 50%, 75%, 90% and 100% - 3 times for 10 min), subjected to the critical desiccation point of CO2 in BAL-TEC equipment, CPD-030, fixed in metallic bracket with silver glue and covered with metallic gold (10 nm) in BAL-TEC apparatus, SCD-050. The anthers and pollen grains were observed and electromicrographed on a scanning electron microscope LEO-EVO 40, XVP. Pollen grains were classified according to the number of openings (Erdtman, 1943). 3. Results The meiosis analysis initially requires the definition of collection times of the floral bud and the definition of size and morphology that are ideal for the observation of meiocytes in different stages. Regarding the choice of the spikelet (Fig. 1), a gradient related to size, morphology, and staining was observed. In the smaller ones, about 1 mm and showing lighter staining, the initial stages of prophase I were found (Fig. 1A); in the spikelets of intermediate size, between 1.5 and 2 mm, the stages between metaphase I and metaphase II were found (Fig. 1B), and the larger ones, from 2.5 mm to 3 mm and darker staining, showed meiocytes between the stages of anaphase II to tetrad (Fig. 1C) and pollen grains, respectively (Fig. 1D). Tetraploidy (2n = 4× = 36) was confirmed for both accessions of C. dactylon var. dactylon (EGL 9 and EGL 10), with diplotene and diakinesis showing univalent, bivalent and tetravalent configurations (Fig. 2C-D), being one bivalent connected to the nucleolus and rarely two or three (Fig. 2D-E). In these accessions were observed

262

D.M. Silva et al. / South African Journal of Botany 118 (2018) 260–267

Fig. 1. Relationship between the spikelet size the and the meiosis phases of Cynodon sp. In the spikelets with (A) 1 mm were stages of prophase I; (B) 1.5 to 2 mm, from metaphase I to metaphase II; (C) 2.5 to 3 mm: anaphase II to tetrad; (D) N 3 mm: pollen grain. (The bar corresponds to 500 μm).

predominantly 2I + 2IV + 13II. For C. nlemfuensis var. nlemfuensis were invariably identified 9 bivalents in diakinesis and metaphases I (Fig. 3C-D), confirming diploidy (2n = 2× = 18), with a bivalent connected to the nucleolus (Fig. 3C) and, less frequently, two and three nucleolar bivalents. In the EGL 10, accession of C. dactylon var. dactylon was observed two terminal knobs and 12 chromomers on pachytenic chromosomes (Fig. 2B). Knobs were also observed in the other accessions (EGL9, ERX7 and ERX 21). However, the ERX 21 accession of C. nlemfuensis var. nlemfuensis showed a variation in the number of chromomers between 10 and 16, being more frequent 12 chromomers (Fig. 3B). In general, the meiosis of the four evaluated accessions/species was regular. The accession with the highest rate of abnormalities (2.04%) was EGL9 (Cynodon dactylon var. dactylon) and the lowest (0.26%) was observed in EGL 21 (Cynodon nlemfuensis var. nlemfuensis) (Table 2). Despite the occurrence of univalent and tetravalent configuration in C. dactylon var. dactylon, the other subphases of prophase I did not show an irregularity. Metaphases II and anaphases II with asynchronous nuclei were observed at a frequency of 7.7 and 1%, respectively (Fig. 2J), as well as telophases II with delayed chromosomes/chromatids (6.2%) were quantified (Table 2). However, in the tetrads, the nuclei were completely normal in both accessions. As observed in Brachiaria (Paula et al., 2016) and common bean (Lima et al., 2016), there are several explanations for this observations. One is that non-oriented chromosomes/chromatids, or those with irregular segregation, may have been included in the nuclei in telophase II or tetrads stages, where they formed normal meiotic products. Another possibility is that meiocytes with a higher rate of abnormalities undergo apoptosis, which could be a strategy to eliminate non-functional pollen grains (Paula et al., 2016). The accessions of Cynodon nlemfuensis var. nlemfuensis (ERX 7 and ERX 21), showed normal prophase I (Fig. 3A–C). The highest index of irregularities was observed from metaphase I, where non-oriented

chromosomes were found (9.4% for ERX 7 and 0.8% for ERX 21, Table 2). In addition, late chromosomes/chromatids were observed in telophase I and II at a frequency of 6 and 4.6% for the ERX 7 accession, respectively (Table 2). The ERX 21 accession showed this abnormality only in telophase I, at a very low frequency (lower than 1%). All the evaluated tetrads were normal in both accessions. The colorimetric tests showed that the accessions of C. dactylon var. dactylon presented pollen viability rates higher than 75% (Table 3). In C. nlemfuensis var. nlemfuensis, the highest percentage (56.02%) was obtained for ERX7 accession with 2% propionic carmine dye (Table 3, Fig. 4A–B). The estimated rate with both dyes for C. dactylon var. dactylon showed that there were no significant differences (p N 0.05) for the pollen grain viability between the accessions EGL 9 and EGL 10. This same observation was confirmed when analyzing each accession in the comparison among dyes (Table 3). For C. nlemfuensis var. nlemfuensis, the estimated viability rate with propionic carmine dye showed significant differences (p b 0.05) between accessions ERX 7 and ERX 21. The estimates did not differ statistically (p N 0.05) for the two accessions with Alexander's stain (Table 3, Fig. 4C-D). The analysis within each accession in the comparison among dyes showed that only for ERX 7 there were significant differences (p b 0.05) for the pollen viability determined by both dyes (Table 3). In addition, pollen grains of different sizes, viable, were observed at low frequencies in the four accessions, and the largest ones showed 28.8 μm on average and the lowest, 21.1 μm (Fig. 4B and D). Observations of the pollen grains obtained from the acetolysis technique showed that there is no statistical difference (p N 0.05) for the ratio between polar and equatorial axis among the accessions (Table 4). However, the polar axes of the pollen grains of the ERX 21 accession were statistically different from ERX 7, both C. nlemfuensis var. nlemfuensis and the two accessions of C. dactylon var. dactylon (Table 4). Regarding the thickness of exine, ectoexine and endoexine, the values observed for EGL 10 (C. dactylon var. dactylon) and ERX 7 (C. nlemfuensis var. nlemfuensis) were statistically the same (p N 0.05), as well as between the accessions EGL 9 (C. dactylon var. dactylon) and ERX 21 (C. nlemfuensis var. nlemfuensis) (Fig. 4E and Table 4). The pollen grains of the evaluated Cynodon accessions/species were morphologically identical and showed only one opening and were therefore classified as monocolpates and showed no ornamentation as spicules and other appendages with non-roughened, microreticulated exine (Fig. 4I). 4. Discussion The data obtained in relation to the size and color of the spikelets aided to select them for the meiotic and pollen viability analyzes and could subsidize further studies with species of the genus. This information is important because, according to Wu and Taliaferro (2009), research on the pairing and meiotic behavior of Cynodon are limited, especially tetraploids and hybrids, and probably one of the reasons is the reduced size of inflorescences and the difficulty to define the collection stages. Some authors have already described irregularities in the chromosome pairing in different genotypes of C. dactylon. Forbes and Burton (1963) reported one accession of C. dactylon had slightly irregular chromosome pairing (Is) averaging 1.6 or less. Malik and Tripathi (1968) indicated that chromosome paired as 18 IIs, however one to two quadrivalents were observed in some cells of tetraploid C. dactylon from India. Another study, Hanna and Burton (1977) reported cytogenetics and fertility in four cultivar of C. dactylon. The cultivar Coastcross-1 showed very irregular meiosis with means association of 5.07 I + 13.94 II + 0.41 III + 0.51 IV per cell. According to Obeso et al. (2014), the formation of bivalent chromosomes is the ideal for the occurrence of normal chromosomal

D.M. Silva et al. / South African Journal of Botany 118 (2018) 260–267

263

Fig. 2. Meiosis in Cynodon dactylon var. dactylon. (A) Leptotene; (B) Pachytene, arrows indicate terminal knobs; (C) Diplotene, arrows indicate tetravalent configurations and arrowheads indicate univalent configurations; (D and E) Diakinesis, arrows indicate bivalents connected to the nucleolus; (F) Metaphase I; (G) Anaphase I; (H) Telophase I; (I) Prophase II; (J) Metaphase II with asynchronous nuclei; (K) Anaphase II; (L) Telophase II with asynchronous nuclei; (M) Tetrad. (Bar corresponds to 10 μm).

segregation, because they contribute to the occurrence of bipolar attachment of the homologous centromere and the subsequent positioning of the chromosomes in the equatorial cell plate, events that contribute to ensure the regularity of segregation.

The classic behavior of the univalents is to show orientation and irregular segregation observed in the form of chromosomes with early or late rise during metaphases I and anaphases I, as observed in the results of the present study. These univalents

264

D.M. Silva et al. / South African Journal of Botany 118 (2018) 260–267

Fig. 3. Meiosis in Cynodon nlemfuensis var. nlemfuensis. (A) Leptotene; (B) Pachytene, arrows indicate terminal knobs; (C) Diakinesis, arrow indicates bivalent connected to the nucleolus; (D) Metaphase I with non-oriented chromosomes; (E) Anaphase I; (F) Telophase I; (G) Prophase II; (H) Metaphase II; (I) Anaphase II; (J) Telophase II with micronucleus; (K) Tetrad. (Bar corresponds to 10 μm).

can be maintained in the later phases and can generate micronuclei (Pagliarini, 2000). In species with irregular segregation, the correlation between this characteristic and pollen fertility and the seed production is observed (Pagliarini, 2000). Chromosomes in tetravalent configurations are also characterized by irregular segregation and result in gametes with unbalanced chromosome number. The tetravalents can segregate as II + II or as IV. Segregation as II + II is desirable, because it may result in balanced gametes (Sybenga, 1992). Failures in pairing may occur due to numerical or structural chromosomal alterations or due to hybrid combinations. In the latter case, the pairing is dependent on the degree of affinity among the chromosomes/genomes of the parental species involved in the crosses (Appels et al., 1998; Singh, 2003). When the bivalents are partially paired (homoeologous pairing), it is said that they have

lower chromosomal affinity, and the higher the number of univalents, the lower the affinity (Techio and Davide, 2007). Cynodon dactylon is recognized as allopolyploid (Zhi-Yun et al., 2013; Chiavegatto et al., 2016; Guo et al., 2017) The two accessions of C. dactylon evaluated in this study also showed the formation of multivalents, but predominance of bivalents, indicating that there is a partial homology among the chromosomes, characteristic of a segmental allopolyploid. Harlan (1970) and Harlan and De Wet (1969) suggested that C. dactylon var. dactylon (diploid) probably originated from the crossbreed of diploid parental C. dactylon var. aridus and C. dactylon var. afghanicus. More recently, Wu (2011) has suggested that the tetraploid tropical race of the dactylon variety comes from the fusion of nonreduced gametes of the diploid C. dactylon var. aridus. The allo or autotetraploid origin of C. dactylon may be elucidated from investigations

D.M. Silva et al. / South African Journal of Botany 118 (2018) 260–267 Table 2 Number of analyzed meiocytes and percentage of abnormal meiocytes (in parentheses) in accessions of Cynodon sp. Species (accession) Phase meiosis

C. dactylon var. dactylon (EGL 9)

C. dactylon var. dactylon (EGL 10)

C. nlemfuensis var. nlemfuensis (ERX 7)

C. nlemfuensis var. nlemfuensis (ERX 21)

PI MI AI TI PII MII AII TII TET Total

537 (2.23%) 91 (0%) 44 (0%) 49 (0%) 45 (0%) 39 (7.7%) 10 (1%) 32 (6.2%) 32 (0%) 879 (2.04%)

647 (1.5%) 290 (0%) 47 (0%) 68 (0%) 105 (0%) 67 (0%) 3 (100%) 21 (0%) 99 (0%) 1347 (0.96%)

483 (0%) 96 (9.4%) 15 (0%) 50 (6%) 105 (0%) 82 (0%) 35 (0%) 43 (4.6%) 176 (0%) 1085 (1.38%)

308 (0%) 254 (0.8%) 38 (0%) 165 (0.6%) 161 (0%) 62 (0%) 40 (0%) 55 (0%) 69 (0%) 1152 (0.26%)

Caption: PI (Leptotene/Zygotene, Pachytene, Diplotene/Diakinesis); MI (Metaphase I); AI (Anaphase I); TI (Telophase I); PII (Prophase II); MII (Metaphase II); AII (Anaphase II); TII (Telophase II); TET (Tetrad).

on meiosis and pairing configurations of the supposed parent species by genomic in situ hybridization (GISH). In allopolyploid species, such as C. dactylon, which have a hybrid origin, the cytological diploidization is thought to result from the divergence between homoeologous chromosomes, which may already exist and/or be accentuated at the onset

265

of polyploid formation (Le Comber et al., 2010) and involve the rearrangement of large chromosome fragments, or from the activity of Pairing homeologous (Ph) genes (Jenczewski and Alix, 2004). Another aspect to be highlighted in this study is the report of the variation from one to three nucleolar bivalents observed in C. dactylon var. dactylon and that had not been described in previous studies. In the analysis of the morphology of the pachytenic chromosomes of diploid plants of C. dactylon, Brilman et al. (1982) mentioned the presence of the nucleolus organizer region (NOR) in the chromosomal pair 4. In the diakinesis of the diploid cytotype of C. dactylon presented by Dhaliwal and Gupta (2011), a nucleolar bivalent is also observed. Typically, tetraploids, as evaluated in this study, are expected to exhibit more than one NOR. In mitotic chromosomes, by fluorescence in situ hybridization (FISH), Zhi-Yun et al. (2013) reported three and four 45S rDNA signals on two tetraploid Cynodon lineages, and in observations made by Chiavegatto (unpublished data) in accessions of C. dactylon var. dactylon, four 45S rDNA sites were invariably observed. The variable number of nucleolar bivalents may indicate extra locus of rDNA 45S in the accessions evaluated in this study. Intraspecific and intravarietal variations for this marker have been already described in other grasses, such as Setaria sphacelata (Nani et al., 2015), Lolium multiflorum (Bustamante et al., 2014), Lolium perenne (Książczyk et al., 2010), and in Brachiaria brizantha (Nani et al., 2016).

Table 3 Percentage of viable pollen grains in Cynodon accessions, using colorimetric tests with Alexander's stain and 2% propionic carmine. Viable (%) Species (accession)

C. dactylon var. dactylon (EGL 9)

C. dactylon var. dactylon (EGL 10)

C. nlemfuensis var. nlemfuensis (ERX 7)

C. nlemfuensis var. nlemfuensis (ERX 21)

87.38 Aa 75.81 Aa

94.94 Aa 88.07 Aa

56.02 Ab 18.41 Bb

18.52 Ac 24.08 Ab

Dye 2% propionic carmine Alexander

Equal lowercase letters in rows and equal capitals in columns do not differ statistically (p N 0.05) by Scott Knott test.

Fig. 4. Pollen grains of Cynodon sp. (A) Pollen grain stained with 2% propionic carmine dye, arrowhead indicates unviable pollen grain, arrow indicates pollen grain of different size; (B) pollen grain stained with Alexander's stain, arrowhead indicates unviable pollen grain, arrow indicates pollen grain of different size; (D) morphology of the pollen grains obtained by the acetolysis technique (I) morphology of the pollen grain obtained in scanning electron microscopy (The bar corresponds to 10 μm).

266

D.M. Silva et al. / South African Journal of Botany 118 (2018) 260–267

Table 4 Average values (μm) of the pollen grain characteristics of Cynodon sp. accessions. Species (Accession)

Endoexine

Ectoexine

Exine

Equatorial Axis

Polar axis

Ratio P/E

C. dactylon var. dactylon (EGL 9) C. dactylon var. dactylon (EGL 10) C. nlemfuensis var. nlemfuensis (ERX 7) C. nlemfuensis var. nlemfuensis (ERX 21)

0.411 b

0.436 b

0.847 b

19.081 b

19.149 b

1.006 a

0.489 a

0.483 a

0.972 a

19.322 b

19.543 b

1.012 a

0.512 a

0.520 a

1.032 a

22.932 a

22.676 a

0.990 a

0.425 b

0.422 b

0.847 b

16.259 c

16.264 c

1.001 a

Averages followed by the same letter in column do not differ statistically (p N 0.05) by Scott-Knott test (P ≤ 5%).

The abnormalities verified in meiosis analyses in the accessions of C. dactylon var. dactylon and C. nlemfuensis var. nlemfuensis were lower than 2%. In the accessions of C. dactylon var. dactylon (EGL 9 and EGL 10), the higher frequency of abnormalities occurred in meiosis II, especially in metaphase II of the EGL9 accession and anaphase II of the EGL 10. Dhaliwal and Gupta (2011), in the evaluations with tetraploid cytotypes of C. dactylon, observed a higher index of abnormalities in meiosis I, presenting delayed chromosomes in anaphase I (9.8 to 20.3%) and bridges in telophase I (4.8 to 9.5%). The asynchrony observed in metaphases II and anaphases II in C. dactylon did not reflect the formation of abnormal microspores, suggesting that they only represent an ephemeral delay of these chromosomes. The diploid accessions (C. nlemfuensis var. nlemfuensis - ERX 7 and ERX 21) had a higher rate of abnormality in meiosis I, especially in metaphase I and telophase I. The abnormalities observed only in telophase II of ERX 7 did not persist in tetrads, indicating that chromosomes/chromatids with irregular segregation were incorporated into the nuclei. The presence of terminal chromomeres and knobs in the pachytenic chromosomes had already been described by Brilman et al. (1982) in diploid clones of C. dactylon. The observations performed in the accessions evaluated in this study showed only two conspicuous terminal knobs and about 12 chromomeres. Knobs and chromomeres are regions of constitutive heterochromatin, highly condensed and observed on pachytenic chromosomes (Schulz-Schaeffer, 1980; Sumner, 2003). The fusion of one or more chromomeres can lead to the formation of knobs. According to Sumner (2003), there is a close correlation between the location of chromomeres and knobs with the G-band patterns, which are rich in AT bases. The pollen viability between the diploid (C. nlemfuensis var. nlemfuensis) and tetraploid (C. dactylon var. dactylon) accessions was statistically different (p N 0.05). The pollen viability estimates observed in this study for C. dactylon var. dactylon were superior to the results obtained by Dhaliwal and Gupta (2011), with five tetraploid genotypes of C. dactylon, whose average was 46.14%. The ERX 7 accession of C. nlemfuensis var. nlemfuensis showed higher pollen viability among the accessions of C. nlemfuensis var. nlemfuensis, despite showing a higher frequency of abnormalities. Pollen viability studies with the species C. nlemfuensis were not found. Future studies aiming at integrating this information into in vitro or in vivo germination tests may help to determine more precisely the fertility potential of the pollen grains of these species. In the viability analyses of the pollen grain, it should be consider that there can be significant variations among individuals of the same species and among samples coming from the same individual. Other variables may influence, such as abiotic factors and the difference among genotypes, which may contribute to the divergence of data obtained in the studies (Techio et al., 2006). In Poaceae, according to Silva et al. (2012), the decrease in pollen viability in plants with normal meiosis can often be related to inability to withstand desiccation. This may be one of the explanations for the results observed in the accessions of C. nlemfluensis evaluated in this study. The pollen grains were classified as oblate spheroidal, according to the classification proposed by Erdtman (1943), where the ratio between

polar and equatorial axis is between 0.88 and 1 μm. In a study by Liu et al. (2004), pollen grains from C. dactylon were classified as suboblates, with ratio between polar and equatorial axis from 0.79 to 1.11 μm, also different from the reported by Morgado et al. (2015) in which the pollen grains of C. dactylon were classified as prolate spheroid with measurement of the polar axis between 10.0 and 12.0 μm and equatorial axis between 9.0 and 12.0 μm. In the same study, Liu et al. (2004), observed, from transmission electron microscopy (TEM), that the pollen grains are insular, with compact exine, where the granule fusion (denomination of insular) appears in density of 1–2/μm2, reporting that these characteristics are common within the subfamily Chloridoideae, in which the genus Cynodon is inserted. The variations in the morphological types observed in our study in comparison with other authors can be explained by effects on pollen grain size and structure caused by the acetolysis technique, which was also used by Liu et al. (2004) and Morgado et al. (2015) to evaluate the pollen grain of C. dactylon. According to O'Keefe and Wymer (2015), this technique is the most widespread in studies of pollen morphology, but has the disadvantage of causing changes in size and, in some cases, in the pollen grain structure. These authors proposed an alternative technique based on enzymatic treatment in fresh pollen grains. Although increasing processing time, it is a non-toxic technique, easy to apply and comparable in costs to acetolysis. Furthermore, it does not cause damage even in fragile pollen grains. Besides the size of the pollen grain, Gupta and Srivastava (1970) and Dhaliwal and Gupta (2011) reported averages of 23.8 and 27.3 μm for diploid and triploid accessions and for diploids and tetraploids of Cynodon dactylon, respectively, similar to that observed in this study, with averages between 21.1 and 28.8 μm for both species. Based on the scanning electron microscopy (SEM) images, the pollen grains of the evaluated accessions are mono-apertured, i.e., they have only one opening through which the pollen tube emerges, with microreticulated exine, according to the classification of Punt et al. (2007) and did not present ornamentation that facilitate dispersion by means of entomophily. 5. Conclusion The accessions of C. dactylon var. dactylon and C. nlemfuensis var. nlemfuensis showed regular meiosis with low rates of abnormalities. The tetraploid accessions of C. dactylon var. dactylon showed higher rates of pollen viability compared to the diploid accessions of C. nlemfuensis var. nlemfuensis. The accessions show pollen grains oblate spheroidal, monocolpates and did not show ornamentation as spicules and other attachments with non-roughened, microreticulated exine. Acknowledgements The authors thank the National Council for Scientific and Technological Development - CNPq(301456/2015-1); the Foundation for Supporting Research of the State of Minas Gerais - FAPEMIG(0025917) for financial support to the research and the Coordination for the

D.M. Silva et al. / South African Journal of Botany 118 (2018) 260–267

Improvement of Higher Education Personnel - CAPES for the scholarship to the first author. References Andrade, C.M.S., de Assis, G.M.L., Fazolin, M., Goncalves, R.C., Sales, M.F.L., Valentin, J.F., Estrela, J.L.V., 2009. Grama-estrela-roxa: gramínea forrageira para diversificação de pastagens no Acre. Embrapa Acre, Rio Branco. 83 pp, Available at. https://www. embrapa.br/busca-de-publicacoes/-/publicacao/663655/grama-estrela-roxa– graminea-forrageira-para-diversificacao-de-pastagens-no-acre, Accessed date: 1 November 2015. Appels, R., Morris, R., Gill, B.S., May, C.E., 1998. Chromosome Biology. Kluwer Academic Publishers, Boston, Dordrecht, London. Avdulow, N.P., 1931. Karyosystematische unteruching der familie gramineen. Bulletin of Applied Botany, Genetics Supplement, London (428 pp.). Brilman, L.A., Kneebone, W.R., Endrizzi, J.E., 1982. Pachytene chromosome morphology of diploid Cynodon dactylon (L.) Pers. Cytologia 47, 171–181. Bustamante, F.O., Rocha, L.C., Torres, G.A., Davide, L.C., Mittelmann, A., Techio, V.H., 2014. Distribution of rDNA in diploid and polyploid Lolium multiflorum Lam. and fragile sites in 45S rDNA regions. Crop Science 54, 1–9. Caro, J.A., Sánchez, E.L., 1969. Las especies de Cynodon (Gramineae) de la República Argentina. Kurtziana 5, 191–252. Chiavegatto, R.B., Paula, C.M.P.de, Souza, S.F., Benites, F.R.G., Techio, V.H., 2016. Karyotype asymmetry in Cynodon Rich. (Poaceae) accessions. Genetics and Molecular Research 15, 1–12. Clayton, W.D., Harman, K.T., Williamson, H., 2009. GrassBase – the online world grass flora, Richmond. Available at. http://www.kew.org/data/grasses-db.html, Accessed date: July 2015. Dhaliwal, A.R.S., Gupta, R.C., 2011. Cytological study on three cytotypes of bermudagrass (Cynodon dactylon (L.) Pers.) from Haryana and Shiwalik Hills. Journal of Basic and Applied Biology 5, 7–12. Erdtman, G., 1943. An Introduction to Pollen Analysis. Chronica Botanica Company, Waltham. Erdtman, G., 1960. The acetolysis method - a revised description. Svensk Botanisk Tidskrift 54, 561–564. Ferreira, D.F., 2003. Sisvar® - Versão 4.6 (Build 61). Universidade Federal de Lavras, Brasil. Filgueiras, T.S., Valls, J.F.M., 2015. Cynodon in Lista de Espécies da Flora do Brasil. Available at. Jardim Botânico do Rio de Janeirohttp://floradobrasil.jbrj.gov.br/jabot/ floradobrasil/FB13144, Accessed date: 4 June 2015. Forbes, I., Burton, G.W., 1963. Chromosome numbers and meiosis in some Cynodon species and hybrids. Crop Science 3, 75–79. Guerra, M., Souza, M.J., 2002. Como observar cromossomos: um guia de técnica em citogenética vegetal, animal e humana. Funpec, São Paulo. Guo, Y., Wu, Y., Anderson, J.A., Moss, J.Q., Zhu, L., Fu, J., 2017. SSR marker development, linkage mapping, and QTL analysis for establishment rate in common Bermudagrass. The Plant Genome 10, 1–11. Gupta, P.K., Srivastava, A.K., 1970. Natural Triploidy in Cynodon dactylon (L.) Pers. Caryologia 23, 29–35. Hanna, W.W., Burton, G.W., 1977. Cytological and fertility characteristics of some hybrid bermudagrass cultivars. Crop Science 17, 243–245. Harlan, J.R., 1970. Cynodon species and their value for grazing and hay. Herbage Abstract 40, 233–238. Harlan, J.R., De Wet, J.M.J., 1969. Sources of variation in Cynodon dactylon (L.) Pers. Crop Science 9, 774–778. Harlan, J.R., De Wet, J.M.J., Rawal, K.M., Felder, M.R., Richardson, W.L., 1970. Cytogenetic studies in Cynodon L. C. Rich (Gramineae). Crop Science 10, 288–291. Horowitz, M., 1996. Bermudagrass (Cynodon dactylon): a history of the weed and its control in Israel. Phytoparasitica 24, 305–320. Jenczewski, E., Alix, K., 2004. From diploids to allopolyploids: the emergence of pairing control genes. Critical Reviews in Plant Sciences 23, 21–45. Książczyk, T., Taciak, M., Zwierzykowski, Z., 2010. Variability of ribosomal DNA sites in Festuca pratensis, Lolium perenne, and their intergeneric hybrids, revealed by FISH and GISH. Journal of Applied Genetics 51, 449–460. Le Comber, S.C., Ainouche, M.L., Kovarik, A., Leitch, A.R., 2010. Making a functional diploid: from polysomic to disomic inheritance. New Phytologist 186, 113–122. Lima, D.C., Braz, G.T., Reis, G. B.dos, Techio, V.H., Davide, L.C., Abreu, A.de F.B., 2016. Implications of mitotic and meiotic irregularities in common beans (Phaseolus vulgaris L.). Genetics and Molecular Research 15, 1–10.

267

Liu, Q., Zhao, N.X., Hao, G., 2004. Pollen morphology of the Chloridoideae (Gramineae). Grana 43, 238–248. Malik, C.P., Tripathi, R.C., 1968. Cytological evolution within the Cynodon dactylon complex. Biologisches Zentralblatt 87, 625–627. Melhem, T.S., Cruz-Barros, M.A.V.da, Corrêa, A.M.da S., Makino-Watanabe, H., SilvestreCapelato, M.S.F.S., Esteves, V.L.G., 2003. Variabilidade polínica em plantas de Campos do Jordão (São Paulo, Brasil). Boletim do Instituto de Botânica de São Paulo 16, 1–104. Morgado, L.N., Gonçalves-Esteves, V., Resendes, R., Ventura, M.A.M., 2015. Pollen morphology of Poaceae (Poales) in the Azores, Portugal. Grana 54, 282–293. Nani, T.F., Cenzi, G., Pereira, D.L., Davide, L.C., Techio, V.H., 2015. Ribosomal DNA in diploid and polyploid Setaria (Poaceae) species: number and distribution. Comparative Cytogenetics 9, 645–660. Nani, T.F., Pereira, D.L., Souza Sobrinho, F., Techio, V.H., 2016. Physical map of repetitive DNA sites in Brachiaria spp.: intravarietal and interspecific polymorphisms. Crop Science 56, 1769–1783. Neiva, J.N.M., Ferreira, A.C.H., Vasconcelos, V.R. de, Melo, F.I.O., Pimentel, J.C.M., Teixeira, M., 1999. Avaliação do crescimento e valor nutritivo do capim de burro (Cynodon dactylon (L) Pers.) no litoral cearense. Reunião Anual da Sociedade Brasileira de Zootecnia, Porto Alegre. Obeso, D., Pezza, R.J., Dawson, D., 2014. Couples, pairs, and clusters: mechanisms and implications of centromere associations in meiosis. Chromosoma 123, 43–55. O'Keefe, J.M.K., Wymer, C.L., 2015. An alternative to acetolysis: application of an enzymebased method for the palynological preparation of fresh pollen, honey samples and bee capsules. Palynology 41, 117–120. Pagliarini, M.S., 2000. Meiotic behavior of economically important plant species: the relationship between fertility and male sterility. Genetics and Molecular Biology 23, 997–1002. Paula, C.M. P.de, Figueiredo, K.G., Souza Sobrinho, F., Benites, F.R.G., Techio, V.H., 2016. Microsporogenesis analysis validates the use of artificially tetraploidized Brachiaria ruziziensis in breeding programs. Genetics and Molecular Research 15, 1–10. Primavesi, A.C., Primavesi, O., Correa, L., de, A., Silva, A.G., Cantarella, H., 2004. Adubação nitrogenada em capim coastcross: efeitos na extração de nutrientes e recuperação aparente do nitrogênio. Revista Brasileira de Zootecnia 33, 68–78. Punt, W., Hoen, P.P., Blackmore, S., Nilsson, S., Thomas, A.Le., 2007. Glossary of pollen and spore terminology. Review of Palaeobotany and Palynology 143, 1–81. Schulz-Schaeffer, J., 1980. Cytogenetics: Plants, Animals, Humans. Springer Science & Business Media, Berlin. Silva, L.A.C., Pagliarini, M.S., Santos, S.A., Silva, N., Souza, V.F., 2012. Chromosome number, microsporogenesis, microgametogenesis, and pollen viability in the Brazilian native grass Mesosetum chaseae (Poaceae). Genetics and Molecular Research 11, 4100–4109. Singh, R.J., 2003. Plant Cytogenetics. CRC Press, Florida. Sousa, F.B., Carvalho, F.C., Araújo Filho, J.A., 1998. Capim Gramão: Uma Opção Para o Nordeste brasileiro. EMBRAPA-CNPC. Circular Técnica n. 14 (Sobral). Sumner, A.T., 2003. Chromosomes: Organization and Function. John Wiley & Sons, Nova Jersey, L. Sybenga, J., 1992. Cytogenetics in Plant Breeding. Springer Science & Business Media, Berlin. Taliaferro, C.M., Rouquette, J.R.F.M., Mislevy, P., 2004. Bermudagrass and stargrass. In: Moser, L.E., Burson, B.L., Sollenberger, L.E. (Eds.), Warm-Season (C4) Grasses. Monograph vol. 45. American Society of Agronomy, Madison, pp. 417–438. Techio, V.H., Davide, L.C., 2007. Análise genômica em cromossomos de plantas com base no pareamento meiótico. Biotemas 20, 7–18. Techio, V.H., Davide, L.C., Pedrozo, C.Â., Pereira, A.V., 2006. Viabilidade dos grãos de pólen de acessos de capim-elefante, milheto e híbridos interespecíficos (capim-elefante x milheto). Acta Scientiarum Biological Science 1, 7–12. The plant list, 2016. A working list off all plant species. Available at. http://www. theplantlist.org/tpl1.1/search?q=cynodon, Accessed date: 2 May 2016. Wu, Y., 2011. Cynodon. In: Kole, C. (Ed.), Wild Crop Relatives: Genomic and Breeding Resources Millets and Grasses. Institute of Nutraceutical Research, USA, pp. 53–71. Wu, Y.Q., Taliaferro, C.M., 2009. Bermudagrass. In: Singh, R.J. (Ed.), Genetic Resources, Chromosome Engineering, and Crop Improvement: Forage Crops. CRC Press, New York, pp. 233–277. Zhi-Yun, G., Chao, X., Ming-Liang, Z., Miao, W., 2013. Distribution of rDNA loci and genome differentiation in tetraploid Cynodon. Indian Journal Genetics and Plant Breeding 73, 459–461.