See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/271399985

Molecular Diversity and Genetic Structure of Guineagrass (Panicum maximum

Jacq.), a Tropical Pasture Grass

Article  in  Tropical Plant Biology · December 2011

DOI: 10.1007/s12042-011-9081-6

CITATIONS READS

10 287

6 authors, including:

Adna Sousa Liana Jank

Universidade Federal da Paraíba Brazilian Agricultural Research Corporation (EMBRAPA)
47 PUBLICATIONS   378 CITATIONS    118 PUBLICATIONS   861 CITATIONS   

SEE PROFILE SEE PROFILE

Tatiana Campos
University of Campinas
50 PUBLICATIONS   488 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

GENETIC DIVERSITY AS A STRATEGY FOR THE MANAGEMENT AND USE OF GERMPLASM OF Vitis sp. OF THE AGRONOMIC INSTITUTE - IAC View project

Genomic selection for cross-pollination crops View project

All content following this page was uploaded by Anete Pereira de Souza on 16 July 2015.

The user has requested enhancement of the downloaded file.

Tropical Plant Biol.
DOI 10.1007/s12042-011-9081-6

Molecular Diversity and Genetic Structure of Guineagrass

(Panicum maximum Jacq.), a Tropical Pasture Grass
Adna Cristina Barbosa de Sousa & Liana Jank &
Tatiana de Campos & Danilo Augusto Sforça &
Maria Imaculada Zucchi & Anete Pereira de Souza

Received: 30 April 2011 / Accepted: 1 September 2011

# Springer Science+Business Media, LLC 2011

Abstract Guineagrass (Panicum maximum Jacq.) is a from microsatellite-enriched libraries. Flanking primers
forage grass found in tropical and subtropical regions. It were designed for 116 microsatellite loci and screened
is an apomictic and tetraploid species from Africa. The using a sample of 25 guineagrass accessions. The thirty
objective of this study was to evaluate the genetic diversity selected polymorphic microsatellites employed in this study
of guineagrass accessions sampled from its regions of produced a total of 192 bands when evaluated in the 396 P.
origin, which is in Tanzania and Kenya. In this study, a total maximum accessions, with an average of 6.4 bands per
of 396 accessions were analyzed, and a collection of microsatellite. Four genetic clusters were identified in the
reproducible and informative microsatellites was devel- collection using STRUCTURE analysis, and these results
oped. Thirty microsatellites were employed to characterize were confirmed using AMOVA. The largest genetic
these accessions. A total of 576 clones were sequenced variation was found within clusters (65.38%). This study
revealed that the collection of accessions from the P.
maximum region of origin was a rich source of genetic
Communicated by Yin-Long Qiu
variability. The geographical distances and genetic similar-
Electronic supplementary material The online version of this article ities among accessions did not indicate a significant
(doi:10.1007/s12042-011-9081-6) contains supplementary material,
which is available to authorized users. association between genetic and geographical variation,
supporting the natural interspecific crossing between P.
A. C. B. de Sousa : T. de Campos : D. A. Sforça : A. P. de Souza
maximum, P. infestum and P. trichocladum as the origin of
Genetic Engineering and Molecular Biology Center (CBMEG),
University of Campinas (UNICAMP), the high genetic variability and the existence of an agamic
CP 6010, Campinas, SP CEP 13083-970, Brazil complex formed by these three species.

L. Jank
Keywords Genetic diversity . Genetic resources .
Embrapa Beef Cattle, Forage Breeding Department, Brazilian
Agricultural Research Corporation, Megathyrsus maximus . Microsatellite markers . Tropical
CP 154, Campo Grande, MS CEP 79002-970, Brazil forage

M. I. Zucchi
Abbreviations
Agronomic Institute of Campinas,
Pólo Apta Centro Sul – Rod. SP 127 km 30, AMOVA analysis of molecular variance
Piracicaba, SPCP 28, CEP 13400-970, Brazil CTAB cetyltrimethylammonium bromide
D discriminating power
A. P. de Souza
DNA deoxyribonucleic acid
Biology Institute, Plant Biology Department (DBV),
University of Campinas (UNICAMP), EMBRAPA Brazilian Agricultural Research Corporation
CP6109, Campinas, SP CEP 13083-970, Brazil IPTG isopropyl β-D-1-thiogalactopyranoside
MCMC Markov Chain Monte Carlo
A. P. de Souza (*)
NJ neighbor joining
Universidade Estadual de Campinas (UNICAMP), CBMEG,
CP 6010, 13083-970, Campinas, SP, Brazil ORSTOM Institut Français de Recherche Scientifique
e-mail: [email protected] pour le Développement en Coopération
 Tropical Plant Biol.

PCA principle component analysis Développement—IRD). The germplasm includes tetraploid

PCR polymerase chain reaction apomictic and tetraploid artificially induced sexual plants
PIC polymorphism information content and may be considered representative of the existing natural
SSRIT simple sequence repeat identification tool variability of the species due to the eco-geographical scope
of the collecting expeditions (Savidan et al. 1989).
The P. maximum germplasm comprises a high level of
phenotypic variability (Jank et al. 1994; Jank et al. 1997).
Introduction However, the genetic diversity of the germplasm has not
been evaluated at the molecular level. Elucidating the genetic
Guineagrass (Panicum maximum Jacq., Megathyrsus max- variability of the germplasm is important for the develop-
imus Jacq. Simon BK and Jacobs SWL) is an important ment of breeding strategies, the selection of accessions and
tropical forage grass native to Africa, where high genetic the conservation of these genetic resources. With the
diversity is found (Burton et al. 1973). Because of its high development of molecular marker technologies, scientific
yield and nutritional content and wide adaptability to tools have become available to efficiently describe the
diverse ecological niches, guineagrass has been widely structure of the genetic diversity present in the germplasm
introduced and exploited in most tropical and subtropical or within cultivars and the diversity among populations
countries, including Brazil, Japan, the USA and Australia without the need for phenotyping (Bolaric et al. 2005).
(Nakajima 1978; Smith 1979; Savidan 1982; Duke 1983). Among the available molecular markers, microsatellites are
Guineagrass belongs to the family Poaceae, the subfamily the most promising marker for genomic applications and are
Panicoideae and the tribe Paniceae. The species is considered highly informative (Gupta and Varshney 2000). Micro-
apomictic (of the gametophytic aposporous type) (Savidan satellite markers consist of short (1–6 bp) tandem repeat
2000; Jain et al. 2006), and its apomictic accessions are DNA sequences randomly dispersed throughout the genome.
autotetraploid (2n=4x=32) (Combes and Pernès 1970; These sequences are locus-specific, polymorphic and exhibit
Bogdan 1977); however, sexual plants in nature have been a co-dominant segregation pattern (Gaitán-Solís et al. 2003).
observed and identified as diploid (2n=2x=16) (Nakajima et The allelic diversity at microsatellite loci, caused by
al. 1979). P. maximum forms an agamic complex with the variations in the number of repeats in the core sequence, is
botanical species P. infestum Anders and P. trichocladum K. likely caused by polymerase slippage and deficient DNA
Schum (Muir and Jank 2004). Because these three species repair during DNA replication (Field and Wills 1996).
possess the same chromosome number (2n=4x=32) and Microsatellites have been proven effective for estimating
intercross freely, intermediate or hybrid accessions may be genetic diversity and genetic relationships and for predicting
found in the natural grass populations in East Africa both the genetic value of selected accessions derived from
(Savidan and Pernès 1982). In the natural population, 7% intraspecific crosses and the performance of their hybrid
of accessions are diploid and reproduce sexually (Pernès progenies (Varshney et al. 2005; Ebina et al. 2007; Chandra
1975). It is thought that these diploid sexual forms intercross and Tiwari 2010). This paper presents the genetic diversity
with P. maximum, P. infestum and P. trichocladum through of 396 P. maximum accessions from the Embrapa Beef Cattle
spontaneous haploidization and recurrent tetraploidization germplasm bank using 30 microsatellite loci.
(Savidan and Pernès 1982). Natural grass populations can be
divided into three types: monomorphic, polymorphic discon-
tinuous and polymorphic continuous. The polymorphic Results and Discussion
discontinuous populations include well-differentiated bio-
types, while the polymorphic continuous populations include Sequence Analysis and Microsatellite Loci Polymorphism
mixed populations representing diploid sexual and tetraploid
apomictic biotypes. Prior studies have suggested the possi- The library enriched for dinucleotide repeat motifs (CT8
bility that the polymorphic continuous population originated and GT8) exhibited high levels of microsatellite enrichment.
from the crossing of a population with the diploid sexual A total of 576 clones were isolated and sequenced. Of these
pool (Pernès 1975). clones, 323 sequences (61%) contained microsatellites, and
The P. maximum germplasm, available at the Brazilian 236 (48.7%) were suitable for designing primers. Redundant
Agricultural Research Corporation—Embrapa Beef Cattle sequences accounted for 23% of the microsatellite-containing
(Mato Grosso do Sul, Campo Grande, Brazil), was clones. The size of the inserts varied between 360 and 900 bp,
introduced in 1982 by a cooperative agreement with the with an average size of 500 bp. Screening of the library
former Institut Français de Recherche Scientifique pour le indicated that 80% of the microsatellites consisted of simple
Développement en Coopération (ORSTOM) in France dinucleotide motifs (perfect and imperfect), while 20% were
(known today as the Institut de Recherche pour le composed of compound motifs. The maximum number of
Tropical Plant Biol.

repeats among the microsatellites was 23 (perfect GA). different numbers of patterns (7 and 6, respectively),
Trinucleotides were less frequent (15%) and exhibited less exhibited similar discriminatory powers. Based on the
repeat units (three to five units). Tetranucleotides (3%) and estimates of PIC and D, 30 microsatellite loci were selected
hexanucleotides (2%) were also observed as simple repeats. to help characterize the P. maximum germplasm bank,
The most common repeat motif was the (TG)n/(CA)n group, which is composed of 396 accessions. The 30 selected
which represented over 79.5% of all microsatellites. microsatellite markers are depicted in Table 1.
Ultimately, a total of 116 primer pairs were designed and
tested using PCR amplification. Previously, 20 microsatellite Molecular Analysis of the Panicum maximum Germplasm
markers had been characterized among 25 P. maximum
accessions selected from the germplasm bank to investigate From the 30 microsatellite loci selected, a total of 192
polymorphism (Sousa et al. 2011). A total 96 newly bands were produced, with an average of 6.4 bands per
developed microsatellites were tested, of which 66 produced locus. Based on Jaccard’s similarity coefficient, the genetic
a product of the expected size. In total, 55 microsatellites variation among the accessions was estimated. The simi-
were polymorphic, and 11 were monomorphic (Table 1). Of larity values among the 396 accessions of P. maximum
the 66 microsatellites evaluated, 46 consisted of dinucleotide ranged from 0.16 to 0.86, with an average similarity of 0.32
repeats, 1 was a trinucleotide repeat and 9 were composed of (Supplementary Table 1). Among the 15 sexual plants (S7,
compound repeats. No correlation was observed between the S8, S9, S10, S11, S12, S13, S14, S15, S16, S17, S18, S19,
types or length of repeats and monomorphic marker S20 and S21), the mean similarity was 0.63.
behavior. A typical STRUCTURE analysis assumes a model in
A total of 318 bands were obtained from the 55 which there are K populations, each of which is characterized
polymorphic microsatellite markers. The number of bands by a set of band frequencies at each locus. Accessions are
from each microsatellite locus ranged from 3 to 16, with an assigned (probabilistically) to a particular population or
average of 5.8 bands per locus. This finding confirms the jointly to two or more populations if their genotypes indicate
high polymorphism of the markers. The number of bands that they are admixed. STRUCTURE analysis, combined
detected in each accession ranged from one to four (Fig. 1). with the computation of Evanno ΔK statistics, suggested a
These results confirm the autotetraploid nature of P. primary partition of the P. maximum accessions into four
maximum (Combes 1975). In autotetraploids, there are four clusters (K=4), with a number of accessions exhibiting
copies of each homologous chromosome. The resulting admixture (Fig. 2a and b). Therefore, the subpopulations
meiotic combination events may include quadrivalents, from the STRUCTURE analysis were grouped into four
trivalents, bivalents and univalents. In nature, P. maximum clusters (I, II, III and IV) (Fig. 3 and Supplementary Table 2),
exhibits diploid-tetraploid-haploid cycles (Savidan and with bootstrap values ranging from 49% to 99%. Cluster I
Pernès 1982) in crosses between apomictic accessions and (red) contained 50 apomictic accessions and cluster II (green)
sexual plants. At each locus, the number of bands and the contained 61 accessions, while the other two clusters, III
number of patterns were used to calculate the degree of (blue) and IV (yellow), contained, 107 and 178 apomictic
polymorphism. The polymorphism information content and sexual accessions, respectively. This approach success-
(PIC) values were calculated to assess marker informative- fully discriminated all of the accessions tested, with the
ness (Mateescu et al. 2005), and the discriminating power exception of duplicate accessions. Confirmed replicates
(D) of each locus was estimated to compare the efficiency (Table 2) in the germplasm bank were determined based on
of markers in varietal identification (Tessier et al. 1999). genetic and morphological analyses (Jank et al. 1997), and
The PIC values ranged from 0.19 to 0.89, with an average the replicates were grouped together in the same cluster. The
of 0.56. The D values ranged from 0.34 to 0.99, with an sexual plants were grouped with apomictic tetraploids in
average of 0.68. The highest PIC and D values were clusters II (S16), C3 (S18, S17, S12, S21 and S8) and C4
observed at the locus 2PMc217, which contained 16 bands. (S11, S13, S14, S19, S20, S15, S9, S7 and S10). These
Of the investigated loci, 70% exhibited more than a 50% sexual plants were selected from crosses between tetraploid-
probability of discriminating between two accessions. The induced sexual plants and apomictic accessions. The mode
analysis of the D values indicated that the efficiency of a of sexual reproduction had been previously identified for
given marker did not depend solely on the number of each accession using embryonic sac analysis with methyl
patterns it generated, as was reported by Tessier et al. salicylate clearing and Nomarski differential interference
(1999). For example, the loci 2PMc428, 2PMc40.1, contrast microscopy (Young et al. 1979). Accessions
2PMc194, 1PMc39.b and 1PMc55 each produced the same classified as sexual plants exhibited exclusively sexual
numbers of patterns and bands (6 and 5, respectively), but embryonic sacs (consisting of part of the stigma, an egg,
they demonstrated different discriminatory powers. In polar nuclei and antipodal cells) (Savidan 1982; Nakagawa
contrast, the loci 2PMc52 and 2PMc103, each generating 1990).
Table 1 Characteristics of the 66 microsatellite loci in 25 Panicum maximum Jacq. accessions: locus name, GenBank accession number, primer sequences (F: forward primer, R: reverse primer),
repeat motif from sequenced clone, product length in base pairs, melting temperature (Tm), number of bands (NA), number of banding pattern, polymorphic information content (PIC) and
discriminating power (D)

Locus/ GenBank accession no. Primer sequences (5′-3′) Repeat motif Product length (bp) Tm (°C) NA Banding pattern PIC D

2PMc8ab F: GCGTTGCTGCATGCGATACCT (TG)8 266 60° 6 7 0.66 0.76

FJ039711 R:GGGGACAAATGCGTTGAAATTAAAAATA
2PMc255 F: GCCGTGAAGACAAAGAGACC (CA)5 229 60° 4 6 0.51 0.66
FJ039712 R: GGAGAGCGAAGGGAGACATT
1PMs35.1 F: TACACTACGCCATTTTG (TG)6 198 51.4° 4 5 0.43 0.59
FJ039713 R: CTAATAGCTTCCTCAGTAATAG
1PMs43b F: ATGAAGCGGGGCGTGTAGTATT (TC)5 200 60° 6 6 0.60 0.72
FJ039714 R: TGGTGGGCGGTAAAGAGTAAAG
2PMc428b F: CTCTCAGTCCCACAGCACAC (CA)11 206 60° 6 5 0.57 0.70
FJ039715 R:TATTTGGGGATTGGGAGTAGTT
2PMc35 F: AGCACTGTGCACTAACCAAATG (GT)7 211 58.8° 4 4 0.49 0.63
FJ039716 R: CGTCTCCGTCCACCGATAG
2PMc39a F:AATGAGCTACCTTCTTG (TA)5 180 55° 1 – – –
HM235410 R:CATTTTAATTTTCCTGTC
2PMc376b F: CACCCATAACTGTAAAAGAA (GT)4GC(GT)5AT(GT)6 258 51.7° 12 12 0.87 0.98
FJ039717 R: CTGGAGTAGCAAGAGTGTT
1PMs96 F: ACAAAGATGGGGCGTGAAGAC (CA)5A(CA)2 252 60° 4 5 0.46 0.60
FJ039718 R: CTAGGTAGGCCGACAACAATGA
2PMc28 F: AACCCGCGCATTTACTACA (AC)6 241 55° 4 5 0.44 0.52
FJ039719 R: ATGGTTGCAGAGAAGAGATGAC
2PMc52b F: AGAATGGCACCTGGAGATAG (TG)7 235 55° 6 7 0.67 0.82
FJ039720 R: GGATAGGCCGAAAGAACAT
2PMc216a F:GGTTCCATATCCCACAC (GT)8 196 50° 1 – – –
HM235411 R:ATCTCCACATTTAGTATCAA
2PMc168b F: CCTCGCATTTTTCTGGATTTA (TG)5 213 60° 12 10 0.79 0.86
FJ039721 R: CATAGACGCACGCACACTCAC
2PMc40b F: ATATTTCCTCGAGATTTGTGTT (TG) 5CA(TG)4 254 55° 6 6 0.62 0.70
FJ039722 R: AAGCTTTGGGGATTAGTAGAA
2PMc7.12 F: TAAACTAGAGGACCCGTGTG (GT)7 269 60° 4 4 0.40 0.55
GU252057 R:TGTAGGCTCAAGAAAGGATT
2PMc9.9b F: GTGCGCGGCCAAGAAAAAGT (GT)6 202 58° 7 8 0.66 0.78
GU252058 R: CTCGAGGGGTGGATAGGACAGG
2PMc9.17b F: ATCAACGCTTTAATCCCTGTCC (CA)5 230 60° 7 6 0.64 0.75
GU252059 R: CATCGTCGTCCTCATCGTAGTC
2PMc282a F:CAGGAACATTATGAAAGTAT (CT)18 163 60° 1 – – –
HM235417 R:AAAAAGTTGCTCTAAAAAT
2PMc14b F: CAGCTCCGTCCCGTATCTCTAA (GT)7 190 60° 4 5 0.55 0.69
Tropical Plant Biol.
Table 1 (continued)
Locus/ GenBank accession no. Primer sequences (5′-3′) Repeat motif Product length (bp) Tm (°C) NA Banding pattern PIC D

GU252060 R: CCGCAGGGAAGCACTATGGT
2PMc19 F: ATGGTTAAAGATGTTGTGAGTG (AC)9 248 55° 3 3 0.22 0.37
Tropical Plant Biol.

GU252061 R: GAGGCTGAGTTCTTGGATAG
2PMc27 F: AAAAGTAGAAGCATTATCCAT (CA)9 217 60° 4 3 0.33 0.49
GU252062 R: TTGCAAAGTGAAAACATTAG
2PMc34 F: AGCACTGTGCACTAACCAAATG (TG)7 211 58.8° 4 4 0.47 0.61
GU252063 R: CGTCTCCGTCCACCGATAG
2PMc40.1b F: ATATTTCCTCGAGATTTGTGTT (GT)4CA(TG)5 254 52° 6 5 0.62 0.72
GU252064 R: AAGGTTTGGGGATTAGTAGAA
2PMc48.2 F: TTCTTTCTTTCCTGTC (CA)13 220 44° 4 3 0.30 0.44
GU252065 R: TTAGATGCTTGAGTTT
2PMc51 F: TCAGCAAGAAACATCCTCA (GA)23 244 60° 4 5 0.44 0.61
GU252066 R: TTCCATAACCCAAATCCTG
2PMc256a F:TGTTCCATTATTGTGTT (GA)9 215 60° 1 – – –
HM235412 R:ACTTTGTTATTGTGAGAA
2PMc55b F: GGTAGCGCTCTGTCCTCTTG (AC)10 220 60° 6 7 0.67 0.80
GU252067 R: GACGGCCTTTCGCTTATTTC
2PMc48b F: CCTGTCAAAAACTATGC (CA)13 231 55° 8 9 0.77 0.89
GU252068 R: GGGGAGACCTAACCA
2PMc60b F: ACAGTTAGCTTAGTGGTTG (CA)8 237 50° 4 6 0.55 0.71
GU252069 R: TATGAAGGAGTAAAAAGACA
2PMc62b F: TGCTGTTTCATACTCTCATT (AG)10 228 51.2° 5 6 0.59 0.74
GU252070 R: ACTGTCTGTTGCTTCACTG
2PMc73 F: TAGTTATGTCATTATTTAGCA (CA)5 233 40° 4 4 0.31 0.44
GU252071 R: AAGTCTTATTTAGTCATTTTG
2PMc285a F:ACTTGCATGTTTTTAT (GT)12 175 45° 1 – – –
HM235420 R:TTGTTCCATCGTCTAT
2PMc84 F: GATCTATAAAAGGAGGGAGCAG (CA)10 153 50° 4 4 0.42 0.57
GU252072 R: GGGGGTTACAAGCAGGTC
2PMc87b F: CCGCTACCTTTTTCTGTCTCCA (CT)5 248 60° 9 10 0.75 0.86
GU252073 R: CTCGGCGCAAGTTGAAGTTTT
2PMc90 F: AACGGTAGCTGGTGAAGA (CA)8 178 53.7° 4 5 0.46 0.55
GU252074 R: ATGTCGATGTGGCAAGTG
2PMc103b F: GCTACATTGGTCTTG (CT)16 282 60° 8 6 0.67 0.82
GU252075 R: GGCACTTCTTAGGATA
2PMc143b F: TTGATAGATACAGAGGAACTTG (CT)10 171 60° 10 11 0.79 0.92
GU252076 R: GGTGCCCATTAGATTGAA
2PMc247a F:GCTCCTTGCTTCACTTTTAT (CA)17 228 45° 1 – – –
HM235413 R:ATCCCGTCATTATTCCATT
Table 1 (continued)
Locus/ GenBank accession no. Primer sequences (5′-3′) Repeat motif Product length (bp) Tm (°C) NA Banding pattern PIC D

2PMc152b F: GGCCCGTCATGTAAGAAC (CA)6 275 60° 5 6 0.59 0.73

GU252077 R: GAG GCTGAGACCGAGTGG
2PMc158b F: GGAATAGCCCCAGATA (TC)6(CA)7 227 50° 8 7 0.70 0.88
GU252078 R: GGCTACCTTCATTGTTC
2PMc172 F: AAGCTAGCAGTTTGAT (CT)20 194 45° 4 3 0.22 0.36
GU252101 R: CGTAGGTATTGGAGTG
2PMc326.1Aa F:GAAAGCATGGGCACAC (CT)15AT(CT)3 244 55° 1 – – –
HM235419 R:TCGTCTCAAGGCATCC
2PMc173 F: AAGGGTATTAGGTTCTGCT (GA)12 258 55° 4 2 0.19 0.34
GU252080 R: CATGACTGACTGGATTAGG
2PMc175 F: TTCACGGTCAGATTCA (CA)6 243 45° 4 4 0.44 0.62
GU252081 R: TGCAGCTCATTTGTTT
2PMc178b F: ACCTGCTTGTTTTGCTTGTTTG (CA)6 226 60° 12 13 0.85 0.94
GU252082 R: AGGGCTGGCTCTGATTGG
2PMc194b F: CCACACGTCGCACTGATAAAAA (CA)9 245 60° 6 5 0.64 0.78
GU252083 R: CCCGAAGGCAGTAGGAGTAGAT
2PMc198 F: CAGAAAGAAGGAAGGAAAGGAA (CT)7 255 56.5° 3 5 0.29 0.37
GU252084 R: TCTAGCTGCATGCATAAACACT
2PMc433a F:GCATGTAGAGCACCAC (CA)5 180 48.9° 1 – – –
HM235414 R:TGTTGAAGTCAGCCTTAT
2PMc221 F: GCACGATGGGCTAAGG (GAA)5 198 53.3° 4 4 0.39 0.52
GU252085 R: GCGGCGGAACGATAA
2PMc217b F: TAACACGGGAGCTGAGGAACAT (GA)11 249 60° 5 6 0.68 0.77
GU252087 R: TGAACATAGCCAGGGAAAGGTC
2PMc247b F: GCTCCTTGCTTCACTTTTAT (CA)17 228 53.2° 16 14 0.89 0.99
GU252088 R: ATCCCGTCATTATTCCATT
2PMc302 F: GGCCTTACCCAATCCA (GT)5 210 51° 4 5 0.44 0.57
GU252089 R: TTCCCTTAACCAAATCACTT
2PMc326b F: CAATTCGTCCCTCGTCTA (CT)15 254 51° 12 12 0.79 0.90
GU252090 R: GGTTCCATGCACAAATAA
2PMc340a F:GGAGAATAAGAGAATG (GT)8 291 53.3° 1 – – –
HM235418 R:TAAGTAGGAGGTATGG
2PMc382b F: ACCCATGATCAGGCAGACAAGA (CA)10 236 60° 6 7 0.76 0.82
GU252091 R: GCAGGCAGGAAAGCAGTAACAC
2PMc389b F: CAGGTAACATCACAAGTA (CA)9 177 50° 7 6 0.72 0.84
GU252092 R: CTATAGGTAAAGCCAGTA
1PMc1.1b F: GGGGGGCGAGAGGGGAGAC (GT)2CT (GT)5 233 60° 6 6 0.59 0.72
GU252093 R: CGGGCGCAGTTTATGGTTGGT
Tropical Plant Biol.
Table 1 (continued)
Locus/ GenBank accession no. Primer sequences (5′-3′) Repeat motif Product length (bp) Tm (°C) NA Banding pattern PIC D

2PMc96a F:TCCTCCCCTTCTTTGTA (CA)7 237 50° 1 – – –

HM235415 R:TCTCTTCAGGTCTCCAC
Tropical Plant Biol.

1PMc13a F: TCGTCCGCCTGAGCAT (GT)9 209 57.7° 4 3 0.39 0.47

GU252094 R: ACGGCGCACCACTGAC
1PMc32 F: AACAGTTTGCAGATGGTAG (CA)2CG(CA)7 256 60° 4 5 0.41 0.57
GU252095 R: TTGAGGATTAATGAGAAGTC
1PMc35.2 F: AATTTTGTTATCCTGCTCCAC (GT)5 208 60° 3 4 0.33 0.47
GU252096 R: ACCCAAAGATAATTAGAACCTG
1PMc39.bb F: CCATCACTCGGGTCAG (CA)8 242 60° 6 5 0.62 0.77
GU252097 R: TTTCGGCAAAACATACA
1PMc53 F: AAAGGGGGTTACAAGCAGGTC (GT)2GA(GT)5 146 60° 4 4 0.45 0.59
GU252098 R: GATCTATAAAAGGAGGGAGCAGA
2PMc239.1Aa F:TAACAAGAGAAATAAACAA (GA)8 216 50° 1 – – –
HM235416 R:GGAGTAAAAGGACCAC
1PMc55b F: TCCCTCTAGAACCAAGCACA (GT)13 160 60° 6 5 0.60 0.74
GU252099 R: ATCAAGACACATCAAGAACACAT
1PMc72 F: GAAATCCGCCTCCACCAA (CA)6 195 60° 4 5 0.57 0.67
GU252100 R: TCCGGCGCCACTTCAT

a
Monomorphic loci
b
Microsatellite loci selected to characterize the germplasm of Panicum maximum Jacq
 Tropical Plant Biol.

Fig. 1 Allelic variation among

25 Panicum maximum acces-
2PMc87 M 250 bp
sions detected using silver-
stained 6% polyacrylamide gels.
M, molecular size marker (10 bp
DNA Ladder). ♦ one band; 230 bp
▪ two bands; • three bands;
* four bands ♦ •
1PMs43 M
220 bp

190 bp
• ♦
2PMc168
230 bp

180 bp
* * •

The tetraploid sexual plants were obtained through the The inheritance of apomixis in P. maximum is
use of colchicine to artificially double the chromosomes of determined by a single dominant gene or a group of genes
diploid sexual plants collected in East Africa for use in located close together, which results in sexual and
breeding programs (Savidan 1982; Nakagawa and Hanna apomictic progenies in a 1:1 ratio (Savidan 1983).
1992). For practical purposes, crosses between accessions Therefore, crosses between sexual and apomictic accessions
with different chromosomal numbers or with distinct result in 50% apomictic hybrids and 50% sexual hybrids.
meiotic behavior usually result in infertile progeny. In This process produces fixed superior accessions that can be
guineagrass, chromosome duplication using colchicine multiplied and entered into the selection process, presenting
allowed for the crossing of sexual plants with tetraploid new possibilities for breeding in this species. Thus, the
apomictic accessions to obtain fertile hybrids (Combes and selection of superior accessions may be conducted by
Pernès 1970; Pernès et al. 1975). selecting the best apomictic accessions from the germplasm
The original diploid sexual plant K189 was collected in bank or by crossing selected sexual and apomictic
Korogwe, Tanzania, and its doubling is the basis of most accessions that exhibit promising characteristics (Savidan
tetraploid sexual plants studied. Other plants collected near 1975, 1982, 1983).
the collection site of K189, accessions K187B, K190A, Cluster IV included the Brazilian cultivars Tanzania-1
K190B, K191, K192 and K193, were all grouped in cluster (ORSTOM T58), Mombaça (ORSTOM K190A) and Atlas
IV. This genetic background of the sexual progenitor may as well as the Cuban cultivar Likone (K5829) (ORSTOM
be the reason why most sexual plants were also grouped in G26). The Australian cultivar Green Panic (ORSTOM G15)
this cluster, despite some of their apomictic progenitors and the Brazilian cultivars Aries and Massai (ORSTOM
being grouped in other clusters. One of the apomictic T21) were found in cluster III. Despite being unable to
progenitors, G3, was grouped in cluster II, and the others, separate the accessions in the different clusters according to
280 and K211, were grouped in cluster III. The other their morphology, the commercial cultivars included in
progenitors for these sexual plants are C1 and G23, which cluster IV were all tall, wider-leafed plants, while the
were grouped in cluster IV. cultivars included in cluster III were short with narrower
The sexual plants that were grouped in cluster III, were leaves. The Massai cultivar is a natural hybrid derived from
derived from male progenitors C1 and G23, accessions that crosses between P. maximum and P. infestum that was
were grouped in cluster IV, as mentioned above. The collected in Dar-Bagamoyo, Tanzania. As a hybrid between
exception was S21, which is a tri-cross derived from male the two species cited, the cultivar’s inflorescences are
progenitors from cluster II and another from cluster III. It is intermediate between a panicle typical of P. maximum and a
noteworthy that all of the sexual plants in this cluster had raceme typical of P. infestum. The inflorescences exhibit
admixtures from cluster IV. primary branches and no secondary branches (Jank 1995;
Tropical Plant Biol.

Fig. 2 Determination of K,
the most probable number of
clusters, using STRUCTURE
software for 396 Panicum
maximum accessions. a. Log
probability of the data, L(K), as
a function of K averaged over
20 replicates, and (b). Ad-hoc
ΔK statistics as a function of K
calculated over 20 replicates

Euclides et al. 2000). In Brazil, cultivars Tanzania-1, little origin correlation, particularly in the south of Kenya
Mombaça and Massai are the most widely cultivated. They and north of Tanzania. This result indicates a close
were selected primarily for the following four characteristics: relationship between the accessions of P. maximum. One
leaf yield, leaf percentage, the ability to regrow seven days advantage of the PCA was that it allowed for the evaluation
after harvesting and pure seed yield (Jank et al. 1993). of the relationships between sets of two accessions, which
The STRUCTURE analysis indicated that clusters I, II, helped to visualize possible introgression between clusters.
III and IV possessed mixed-ancestry origins (Kenya and Cluster I was the most structured cluster, exhibiting
Tanzania). The P. maximum accessions that are preceded by exclusively apomictic accessions, with the exception of
T were collected in Tanzania (1969), KK was collected in the S16 sexual accession. Clusters II and III were more
Kenya (1969), K was collected in Kenya or Tanzania (1967) dispersed and shared gene pools.
and G or a number not preceded by a letter were provided by The genetic relationships among the P. maximum
African research institutions as seeds or cuttings, respectively. accessions did not indicate an association with their
Cluster I comprised 94% of the accessions from Kenya or geographical distribution. All of the groups exhibited wide
Tanzania collected in 1967. Clusters II, III and IV were more geographical distributions in Kenya and Tanzania. Some
diverse because they included accessions collected from the accessions collected in 1967 and others from African
two expeditions in East Africa (1967 and 1969) and African research institutions were closely related genetically (Cluster
research institutions. The level of genetic diversity within IV II, green), but they exhibited a broad geographical distribu-
(0.69) was higher than that of I (0.39), II (0.41) and III (0.49). tion, which included regions in the south of Kenya and the
Cluster IV exhibited the greatest total number of bands per north of Tanzania. In the first collecting expedition (1967),
locus and the highest numbers of bands. This results is one diploid sexual plant was discovered in the region of
justified because cluster IV had the highest number of Korogwe in Tanzania (Combes and Pernès, 1970; Savidan,
apomictic accessions (168 accessions) and sexual plants (10 1982), which suggests that the center of origin could have
plants). The results of the STRUCTURE analysis for variation been located in Tanzania. The wide distribution of the
within groups were confirmed using AMOVA (Table 3). The guineagrass accessions strongly supports intercrossing with
largest percentage of variation was determined within groups diploid sexual plants in natural habitats, which may have
(65.38%), and a smaller level of variation was observed preceded dispersal to other regions. Our data support the
among groups (34.62%). This result is consistent with the occurrence of natural interspecific crossings between P
apomictic mode of reproduction of P. maximum, in which the maximum, P infestum and P. trichocladum at the geograph-
intraspecific variability in apomictic species is large. ical origin of the high genetic variability, and these findings
Principal components analysis (PCA) was employed to additionally suggest the existence of an agamic complex
visualize individual accessions in a multivariate space formed by these three species. This possibility is supported
based on values of genetic similarity derived from the by various apomictic accessions of guineagrass being
proportion of the accessions in the data. This type of clustered according to their genetic similarity. The results
graphical representation enables the evaluation of the of the STRUCTURE analysis could represent the sexual
population structure and geometric distances among all of crossing events according to the conditions in natural
the accessions in the study. The first two principal habitats. According to Pernès (1975), natural populations
coordinates in the PCA accounted for 57.04% and can be divided into three main types: monomorphic,
14.52% of the total variation, respectively. The PCA polymorphic discontinuous and polymorphic continuous.
produced three distinct distributions of accessions (Fig. 4), Prior studies have suggested that the polymorphic continuous
which are colored in the figure according to the STRUC- population originated through the crossing of populations
TURE results. A scattergram of these two axes indicated from the diploid sexual pool. Our results indicate that the
 Tropical Plant Biol.

polymorphic discontinuous population corresponds with Fig. 3 Population structure analysis. Each accession is represented by„
cluster I (red), which contains only apomictic accessions. a thin vertical segment, which can be partitioned into K colored
segments that represent the individual estimated membership of the K
The polymorphic continuous populations corresponded to cluster. The colors of the bar correspond to one of the four clusters
the accessions included in clusters II (green), III (blue) and identified using the STRUCTURE program (I = red, II = green,
IV (yellow), which contain both apomictic accessions and III = blue and IV = yellow)
sexual plants.
The accessions that were introduced in Brazil and other and Pernès 1970). The accessions were introduced into
regions, specifically, Sri Lanka, Venezuela, Suriname, Brazil through a cooperative agreement with Embrapa
Australia, Guadalupe, S. Morocco and Vietnam, over the (Savidan et al. 1989).
past 400 years were grouped by admixture into four groups. Genomic DNA was extracted from freeze-dried leaf
All domesticated accessions exhibited wide geographical samples through using cetyltrimethylammonium bromide
distribution, with close genetic relationships within groups (CTAB) method (Doyle and Doyle 1990). DNA samples
of the accessions collected in Kenya and Tanzania, suggesting were quantified through comparison with known quantities
that these accessions were collected in the region of origin of λ-phage DNA on a 1% agarose gel.
(East Africa). Nevertheless, these accessions are apomictic,
and a relative scarcity of polymorphisms in the domesticated Constructing the Library, Screening for Microsatellite
accessions prevented the detection of more subtle genetic Repeat Sequences, DNA Sequencing and Designing
differences among closely related accessions. Primers
Microsatellites have become one of the most widely
employed molecular markers for genetic analysis. The The microsatellite markers used in this study were
enrichment of DNA fragments through the binding of developed from an enriched genomic library employing
microsatellite probes is a simple and efficient method for the protocol described by Billotte et al. (1999). The
the isolation of microsatellites and has been successfully extracted DNA (P. maximum cv. Tanzania ORSTOM-T58)
applied to a number of plant genomes. Molecular marker was digested using the RsaI restriction enzyme (Invitrogen,
analysis demonstrated the pattern of the distribution of the CA, USA) and ligated to the adapter sequences 5′-
genetic diversity and the population structure of the P. CTCTTGCTTACGCGTGGACTA-3′ and 5′-TAGTC
maximum germplasm. This study found that this germplasm CACGCGTAAGCAAGAGCACA-3′. The library was
collection was a rich source of genetic variability, providing enriched for dinucleotide sequences using (CT)8- and
the necessary raw material for breeding programs. In support (GT)8-biotinylated microsatellite primers with labeled
of the long-term conservation of germplasms, microsatellite probes. The selected DNA fragments were recovered using
markers may be employed to demonstrate that accessions or Streptavidin MagneSphere Paramagnetic Particles (Promega,
cultivars are true to type for the following reasons; to help WI, USA) with a biotinylated probe. After the DNA
ensure their proper maintenance; to determine the degree of fragments were recovered, magnetic selection was performed
relatedness among accessions or groups of accessions; to according to the manufacturer’s specifications. Selected
clarify the genetic structure, partitioning or variation among fragments were PCR-amplified using primer sequences
accessions, populations and species; and to help determine the complementary to the adapters and then ligated into the
presence of a specific gene or gene complex in particular pGEM-T vector (Promega, WI, USA). Escherichia coli
accessions (Koh et al. 1996). Therefore, the development of XL-1 Blue cells (Stratagene, CA, USA) were transformed
these microsatellite markers is an important first step toward with the recombinant plasmids and cultivated on Luria-
the development of a genetic linkage map and a better Bertani (LB) agar plates containing 100 μg mL−1 ampicillin
understanding of the genomic organization of P. maximum. (Sigma, Germany), 50 μg mL−1 X-galactosidase and
isopropyl β-D-1-thiogalactopyranoside (IPTG) (MBI Fer-
mentas, MD, USA). Single white colonies were transferred
Material and Methods to microplates for long-term storage at −80°C. A total of 576
recombinant colonies were selected and sequenced in both
Plant Material and DNA Extraction directions using T7 and SP6 promoter primers using an ABI
PRISM 377 DNA Sequencer (Applied Biosystems, CA,
A total of 396 P. maximum accessions were analyzed in this USA) with the BigDye terminator version 3.1 Cycle
study (Table 2). The accessions were obtained from the Sequencing Kit (Applied Biosystems, CA, USA). The
germplasm bank of the Brazilian Agricultural Research simple sequence repeat identification tool (SSRIT) (http://
Corporation—Embrapa Beef Cattle at Campo Grande, www.gramene.org/db/markers/ssrtool) was employed to
Mato Grosso do Sul, Brazil. They were collected in Kenya identify microsatellites present in non-redundant sequences
and Tanzania by the former ORSTOM, France (Combes (Temnykh et al. 2001). The sequences containing micro-
Tropical Plant Biol.
 Tropical Plant Biol.

Table 2 Information for the 396 Panicum maximum Jacq. accessions

Sample code Accession ID Origin RM Sample code Accession ID Origin RM

1 K190Ac Tanzania, Korogwe APO 199 K214 Kenya, L. Lunga-Mombasa APO

2 T58c Tanzania, Korogwe APO 200 K217 Kenya, L. Lunga-Mombasa APO
3 3Coloradoa Unknown APO 201 K218 Kenya, Mombasa APO
4 Natsuyutaka East Africa APO 202 K219 Kenya, Mombasa APO
5 C1 Africa APO 203 K220 Kenya, Mombasa-Voi APO
6 60 Congo, Brazzaville APO 204 K223 Kenya, Mombasa-Voi APO
7 77 Kenya APO 205 K224 Kenya, Mombasa-Voi APO
8 82 Cameroon, Nkol Bisson APO 206 K225 Kenya, Mombasa-Voi APO
9 87 Cameroon, Nlohelouem APO 207 K227 Kenya, Mombasa-Voi APO
10 93 Introduced from Brazil APO 208 K228 Kenya, Mombasa-Voi APO
11 174 Ivory Coast, Daloa APO 209 K237 Kenya, Voi APO
12 304 Zaire, Gandajika APO 210 K241 Kenya, Voi-Machakos APO
13 B2 Introduced in Brazil APO 211 K241D Unknown APO
14 B6 Introduced in Brazil APO 212 K244 Kenya, Voi-Machakos APO
15 B7 Introduced in Brazil APO 213 K249 Kenya, Nairobi APO
16 B10 Introduced in Brazil APO 214 KK7 Kenya, Meru-Embu APO
17 B12 Introduced in Brazil APO 215 KK8 Kenya, Meru-Embu APO
18 B19 Introduced in Brazil APO 216 KK10 Kenya, Meru-Embu APO
19 B22 Introduced in Brazil APO 217 KK12 Kenya, Meru-Embu APO
20 B26 Introduced in Brazil APO 218 KK14 Kenya, Meru-Embu APO
21 78 Introduced in Angola APO 219 KK14E Unknown APO
22 S7 IRD cross SEX 220 KK15 Kenya, Meru-Embu APO
23 S8 IRD cross SEX 221 KK16 Kenya, Meru-Embu APO
24 S9 IRD cross SEX 222 KK17 Kenya, Meru-Embu APO
25 S10 IRD cross SEX 223 KK18 Kenya, Meru-Embu APO
26 S11 IRD cross SEX 224 KK21 Kenya, Meru-Embu APO
27 S12 IRD cross SEX 225 KK23 Kenya, Meru-Embu APO
28 S13 IRD cross SEX 226 KK23E Unknown APO
29 S14 IRD cross SEX 227 KK25 Kenya, Meru-Embu APO
30 S15 IRD cross SEX 228 KK26 Kenya, Meru-Embu APO
31 S16 IRD cross SEX 229 KK33 Kenya, Meru-Embu APO
32 S17 IRD cross SEX 230 KK34 Kenya, Meru-Embu APO
33 S18 IRD cross SEX 231 GI Gabon, Irat APO
34 S19 IRD cross SEX 232 G2 Uganda, Serere APO
35 S20 IRD cross SEX 233 G3 Nigeria, Zaria APO
36 S21 IRD cross SEX 234 G4 Botswana, Tuli APO
37 S22 IRD cross APO 235 G5 Botswana, Nata APO
38 S23 IRD cross APO 236 G6 Zimbabwe, Melsetter APO
39 T4 Tanzania, Dar-Bagamoyo APO 237 G7 Zimbabwe APO
40 T7 Tanzania, Dar-Bagamoyo APO 238 G8 Zimbabwe, Marandellas APO
41 T11 Tanzania, Dar-Bagamoyo APO 239 G9 Zimbabwe, Victoria Falls APO
42 T11D Unknown APO 240 G10 Zimbabwe, Marandellas APO
43 T18 Tanzania, Dar-Bagamoyo APO 241 G11 Malawi APO
44 T19 Tanzania, Dar-Bagamoyo APO 242 G11E Unknown APO
45 T21c Tanzania, Dar-Bagamoyo APO 243 G12 Introduced in Angola APO
46 T23 Tanzania, Dar-Bagamoyo APO 244 G13 Zimbabwe, Marandellas APO
47 T24 Tanzania, Dar-Bagamoyo APO 245 G14 Zimbabwe, Marandellas APO
48 T45 Tanzania, Korogwe APO 246 G15 Zimbabwe, Marandellas APO
49 T46 Tanzania, Korogwe APO 247 G16 Introduced in Brazil APO
50 T60 Tanzania, Korogwe-Kilosa APO 248 G17 Zimbabwe APO
51 T62 Tanzania, Korogwe-Mikume APO 249 G18 Kenya APO
52 T65 Tanzania, Korogwe-Kilosa APO 250 G19 Kenya, Machakos APO
53 T68 Tanzania, Korogwe-Kilosa APO 251 G20 Kenya APO
Tropical Plant Biol.

Table 2 (continued)
Sample code Accession ID Origin RM Sample code Accession ID Origin RM

54 T72 Tanzania, Korogwe-Kilosa APO 252 G20E Unknown APO

55 T77 Tanzania, Kilosa-Mikume APO 253 G21 Kenya, Mac Kinnon Road APO
56 T81 Tanzania, Kilosa-Morogoro APO 254 G21E Unknown APO
57 T84 Tanzania, Kilosa-Morogoro APO 255 G22 Tanzania, Tengeru APO
58 T86 Tanzania, Mts. Uruguru APO 256 G23 Malawi, Nchizi APO
59 T91 Tanzania, Morogoro APO 257 G26 Kenya APO
60 T92 Tanzania, Morogoro APO 258 G27 Introduced in Angola APO
61 T95 Tanzania, Morogoro APO 259 G27E Unknown APO
62 T96 Tanzania, Morogoro-Dar APO 260 G28 South Africa APO
63 T97 Tanzania, Morogoro-Dar APO 261 G28A Unknown APO
64 T98 Tanzania, Morogoro-Dar APO 262 G30 Angola APO
65 T103 Tanzania, Morogoro-Dar APO 263 G31 Angola APO
66 T104 Tanzania, Morogoro-Dar APO 264 G5E Botswana, Nata APO
67 T108 Tanzania, Morogoro-Dar APO 265 G32 Angola APO
68 T109 Tanzania, Morogoro-Dar APO 266 G33 Angola APO
69 T110 Tanzania, Morogoro-Dar APO 267 G34 South Africa APO
70 T111 Tanzania, Morogoro-Dar APO 268 G35 Gabon, Irat APO
71 T113 Tanzania, Dar APO 269 G36 Zaire, Kinshasa-Nioki APO
72 T114 Tanzania, Dar APO 270 G38 Introduced in Sri-Lanka APO
73 T116 Tanzania, Dar APO 271 G39 Introduced in Sri-Lanka APO
74 T117 Tanzania, Dar APO 272 G40 Madagascar APO
75 T200 Tanzania APO 273 G41 Botswana, Mahalapye APO
76 T201 Tanzania APO 274 G42 Botswana, Ngamiland APO
77 15 Ivory Coast, Binao APO 275 G43 Botswana, Mahalapye APO
78 57 Ivory Coast, Daome, Niaouli APO 276 G45 South Africa APO
79 58 Central African Republic, Boukoko APO 277 G46 South Africa APO
80 64D Unknown APO 278 G47 South Africa APO
81 65 Congo APO 279 G48 South Africa APO
82 69 Zimbabwe, Melsetter APO 280 G50 South Africa APO
83 73 Introduced in Costa Rica APO 281 G51 South Africa APO
84 74 Angola APO 282 G52 Introduced in S. Vietnam APO
85 80 Cameroon, Yaounde APO 283 G54 Introduced in Marocco APO
86 81 Cameroon, Nkwonvone APO 284 G54E Unknown APO
87 88 Introduced in Brazil APO 285 G56 Tanzania, Kilosa APO
88 88A Unknown APO 286 G56E Unknown APO
89 88b Introduced in Brazil APO 287 G58 Nigeria, Nsukka APO
90 89 Introduced in Brazil APO 288 G58D Unknown APO
91 89B Unknown APO 289 G59 Nigeria, Nsukka APO
92 90 Introduced in Brazil APO 290 G59b Nigeria, Nsukka APO
93 92 Introduced in Brazil APO 291 G61 Introduced in Australia APO
94 92D Unknown APO 292 G62 Zimbabwe APO
95 96 South Africa APO 293 G64 Kenya, Rumuruti APO
96 96E Unknown APO 294 G68 Introduced in Australia APO
97 97 South Africa APO 295 G68A Unknown APO
98 97E Unknown APO 296 G69 South Africa APO
99 102 Introduced in Guadalupe APO 297 G70 Zimbabwe APO
100 103 Introduced in Guadalupe APO 298 G71 Introduced in Australia APO
101 105 Introduced in Guadalupe APO 299 G71E Unknown APO
102 105Db Unknown APO 300 G73 Introduced in Australia APO
103 106 Introduced in Guadalupe APO 301 G74 Introduced in Australia, Brisbane APO
104 112 South Africa APO 302 G75 Introduced in Brazil APO
105 112D Unknown APO 303 G76 Introduced in Australia, Brisbane APO
106 114 South Africa APO 304 G76D Unknown APO
 Tropical Plant Biol.

Table 2 (continued)
Sample code Accession ID Origin RM Sample code Accession ID Origin RM

107 114D Unknown APO 305 G77 Introduced in Australia, Brisbane APO
108 116 Malawi, Lilongwe APO 306 G77E Unknown APO
109 117 Unknown APO 307 G78 Introduced in Australia, Brisbane APO
110 118 Unknown APO 308 G78E Unknown APO
111 139 Tanzania, Tengeru-Moshi APO 309 G85 Introduced in Australia, Canberra APO
112 172 Nigeria, Lagos APO 310 G86 Introduced in Australia, Canberra APO
113 280 Kenya, Nanyuki APO 311 G88 Central African Republic, Bouar APO
114 309 Zaire, Gandajika APO 312 G89 Central African Republic, Bangui APO
115 353 Togo, Ganave APO 313 G89E Unknown APO
116 354 Togo, Ganave APO 314 G90 Introduced in Jamaica, Mona APO
117 S22Rb IRD cross APO 315 G90D Unknown APO
118 K4 Kenya, Nairobi APO 316 G91 Introduced in Jamaica, Mona APO
119 K8 Kenya, Nairobi APO 317 G93 Senegal, Dakar APO
120 K2 Kenya, Nairobi APO 318 G94 Ivory Coast, Tiantiebe APO
121 K4 Kenya, Nairobi APO 319 G95 Introduced in Venezuela APO
122 K5 Kenya, Nairobi APO 320 G96 Introduced in Venezuela APO
123 K15 Kenya, Rumuruti APO 321 G97 Introduced in Surinam APO
124 K23 Kenya, Rumuruti-Maralal APO 322 G98 Introduced in Brazil APO
125 K28 Kenya, Maralal APO 323 G99 Burundi, Bujumbura APO
126 K31 Kenya, Maralal-Rumuruti APO 324 G100 Introduced in Haiti APO
127 K32 Kenya, Maralal-Rumuruti APO 325 G109 Tanzania APO
128 K35 Kenya, Rumuruti-Nanyuki APO 326 87 Cameroon, Nlohelouem APO
129 K36 Kenya, Rumuruti-Nanyuki APO 327 Ariesc Introduced in Brazil APO
130 K38 Kenya, Rumuruti-Nanyuki APO 328 Atlasc Introduced in Brazil APO
131 K39 Kenya, Nanyuki-Nyeri APO 329 Japa Unknown APO
132 K39D Unknown APO 330 3697=74b Angola APO
133 K42 Kenya, Nanyuki-Nyeri APO 331 3808=89b Introduced in Brazil APO
134 K42D Unknown APO 332 3816=90b Introduced in Brazil APO
135 K47 Kenya, Meru APO 333 3859=94b Introduced in Brazil APO
136 K47D Kenya, Meru APO 334 3891=101b Introduced in Guadalupe APO
137 K48 Kenya, Meru APO 335 3905=102b Introduced in Guadalupe APO
138 K59 Kenya, Nkubu APO 336 3930=105b Introduced in Guadalupe APO
139 K59E Unknown APO 337 3981=114b Introduced in Guadalupe APO
140 K62 Kenya, Nkubu APO 338 4120=354b Togo, Ganave APO
141 K63 Kenya, Nkubu APO 339 4316=G19b Kenya, Machakos APO
142 K64 Kenya, Nkubu APO 340 4316B=G19b Gabon, Irat APO
143 K65 Kenya, Nkubu APO 341 4375=G26b Kenya APO
144 K68 Kenya, Meru-Embu APO 342 4391=G28b South Africa APO
145 K71 Kenya, Meru-Embu APO 343 4405A=G30b Nigeria, Zaria APO
146 K83 Kenya, Meru-Embu APO 344 4405B=G30b Nigeria, Zaria APO
147 K88 Kenya, Meru-Embu APO 345 4499=G40b Madagascar APO
148 K89 Kenya, Meru-Embu APO 346 4464=G36b Zaire, Kinshasa-Nioki APO
149 K93 Kenya, Embu APO 347 4502A=G41b Botswana, Tuli APO
150 K98 Kenya, Forthall APO 348 4502B=G41b Botswana, Mahalapye APO
151 K98D Unknown APO 349 4618=G58b Nigeria, Nsukka APO
152 K102 Kenya, Forthall-Nairobi APO 350 4634=G61b Introduced in Australia APO
b
153 K102R Kenya, Forthall-Nairobi APO 351 4651=G64b Kenya, Rumuruti APO
154 K103 Kenya, Forthall-Nairobi APO 352 4654 Unknown APO
155 K104 Kenya, Forthall-Embu APO 353 4669=G68b Introduced in Australia APO
156 K105 Kenya, Forthall-Nairobi APO 354 4707=G73b Introduced in Australia APO
157 K105A Unknown APO 355 4723=G75b Introduced in Brazil APO
158 K106 Kenya, Forthall-Nairobi APO 356 4731=G76b Introduced in Australia APO
159 K112 Kenya, Nairobi-Arusha APO 357 4804=G90b Introduced in Jamaica, Mona APO
Tropical Plant Biol.

Table 2 (continued)
Sample code Accession ID Origin RM Sample code Accession ID Origin RM

160 K115 Kenya, Nairobi-Arusha APO 358 4812=G91b Introduced in Jamaica, Mona APO
161 K116 Kenya, Nairobi-Arusha APO 359 4847=G96b Introduced in Venezuela APO
162 K117 Kenya, Nairobi-Arusha APO 360 4863=G98b Introduced in Brazil APO
163 K124 Tanzania, Arusha APO 361 4928=T93b Tanzania, Morogoro APO
164 K124D Unknown APO 362 K209 Kenya, L. Lunga-Mombasa APO
165 K124Rb Tanzania, Arusha APO 363 K16 Kenya, Th. Falls-Rumuruti APO
166 K125 Tanzania, Arusha APO 364 5274=K39b Kenya, Nanyuki-Nyeri APO
167 K126 Tanzania, Arusha APO 365 5282=K42b Kenya, Nanyuki-Nyeri APO
168 K130 Tanzania, Tengeru APO 366 5321=K47b Kenya, Meru APO
169 K138 Tanzania, Tengeru-Moshi APO 367 5932=K201b Tanzania, Tanga APO
170 K139 Tanzania, Tengeru-Moshi APO 368 5461=K130b Tanzania, Tengeru APO
171 K142 Tanzania, Tengeru-Moshi APO 369 5568=K145b Tanzania, Tengeru-Moshi APO
172 K145 Tanzania, Tengeru-Moshi APO 370 K159 Tanzania, Mts Pare Nord APO
173 K146 Tanzania, Tengeru-Moshi APO 371 K160 Tanzania, Mts Pare Nord APO
174 K146E Unknown APO 372 K162 Tanzania, Mts Pare Nord APO
175 K156 Tanzania, Moshi-Mombo APO 373 K172 Tanzania, Mts Pare Nord APO
176 K163 Tanzania, Mts Pare Nord APO 374 5819A=K175b Tanzania, Mts Pare Nord APO
177 K164 Tanzania, Mts Pare Nord APO 375 5819B=K175b Tanzania, Mts Pare Nord APO
178 K165 Tanzania, Mts Pare Nord APO 376 K176 Tanzania, Mts Pare Nord APO
179 K171 Tanzania, Mts Pare Nord APO 377 K177 Tanzania, Baron`s Falls APO
180 K173 Tanzania, Mts Pare Nord APO 378 K17 Kenya, Th. Falls-Rumuruti APO
181 K174 Tanzania, Mts Pare Nord APO 379 K25 Kenya, Rumuruti-Maralal APO
182 K174Rb Tanzania, Mts Pare Nord APO 380 K27 Kenya, Maralal APO
183 K175 Tanzania, Mts Pare Nord APO 381 6149=K28b Kenya, Maralal APO
184 K175D Unknown APO 382 K72 Kenya, Meru-Embu APO
185 K187B Tanzania, Mombo-Korogwe APO 383 K74 Kenya, Meru-Embu APO
186 K190B Tanzania, Korogwe-Tanga APO 384 K79 Kenya, Meru-Embu APO
187 K191 Tanzania, Korogwe-Tanga APO 385 K93 Kenya, Embu APO
188 K192 Tanzania, Korogwe-Tanga APO 386 K95 Kenya, Embu-Forthall APO
189 K193 Tanzania, Korogwe-Tanga APO 387 K99 Kenya, Forthall APO
190 K194 Tanzania, Korogwe-Tanga APO 388 K230 Kenya, Mombasa-Voi APO
191 K197 Tanzania, Korogwe-Tanga APO 390 K238 Kenya, Voi-Machakos APO
192 K201 Tanzania, Tanga APO 391 K240 Kenya, Voi-Machakos APO
193 K204 Tanzania, Mpirani-L. Lunga APO 392 K243 Kenya, Voi-Machakos APO
194 K205 Tanzania, Mpirani-L. Lunga APO 393 T106 Tanzania, Morogoro-Dar APO
195 K205E Unknown APO 394 T115 Tanzania, Dar APO
196 K206 Tanzania, Mpirani-L. Lunga APO 395 7676=KK18b Kenya, Meru-Embu APO
197 K211 Kenya, L. Lunga-Mombasa APO 396 KK20 Kenya, Meru-Embu APO
198 K212 Kenya, L. Lunga-Mombasa APO

Accession ID ORSTOM: Institut Français de Recherche Scientifique pour le Développement en Coopération; RM: reproductive mode; APO:
apomictic (2n=4x=32); SEX: sexual (2n=4x=32); a Panicum coloratum; b Replicates; c Brazilian cultivar

satellites were edited and clustered using LaserGene v. 5.03

software (DNAStar Inc.). MICROSAT software was utilized
Table 3 Molecular analysis of variance (AMOVA) based on to remove the restriction sites in the sequences. PrimerSelect
STRUCTURE results software (DNAStar Inc.) was used to design complementary
Source of Degrees of Sum of Variance Percentage of P-value primer pairs under the following conditions: an expected
variation freedom squares component components amplified product size between 150 and 300 bp, GC content
between 40% and 60%, melting temperature (Tm) between
Among Clusters 3 177.494 3.949Va 34.62 0.0000
45° and 60°C, primer length between 18 and 22 bp and with
Within Clusters 21 537.143 9.354Vb 65.38 –
no predicted hairpin or dimer formation.
 Tropical Plant Biol.

Fig. 4 Association among 396 Panicum maximum accessions revealed using principal components analysis (PCA) based on Jaccard’s similarity
coefficient calculated from 30 microsatellite loci. Samples are color-coded based on the STRUCTURE results

Polymerase Chain Reaction (PCR) Amplifications Data Analysis

and Genotyping
The PIC values were calculated for estimates of marker
PCR was performed in a total reaction volume of 25 μL informativeness according to the equation of Mateescu et
containing 0.5 ng of DNA template, 0.8 μM of each of the al. (2005);
forward and reverse primers, 100 μM dNTPs (MBI
X
n n X
X n
Fermentas, MD, USA), 1.5 mM MgCl2, 10 mM Tris-HCl, PIC ¼ 1  p2i  2p2i p2j
50 mM KCl and 0.5 U of Taq DNA polymerase i¼1 i¼1 j¼iþ1
(Invitrogen, CA, USA). All PCR amplifications were
where p is the frequency of the ith band, pj is the frequency
i
performed in a PTC-200 thermal cycler (MJ Research,
of the jth band and the summation extends over n bands. To
Waltham, MA,USA) under the following conditions: 94°C
compare marker efficiencies in varietal identifications, the
for 1 min, 30 cycles of 94°C for 1 min, a specific
D was estimated for each primer based on the following
temperature for 1 min and 72°C for 1 min and a final
formula;
extension at 72°C for 5 min. Amplification products were
genotyped using electrophoresis on 6% denaturing poly-
acrylamide gels in 1X TBE buffer. A 10 bp ladder X
l
Npj  1
Dk ¼ 1  pj ;
(Invitrogen, CA, USA) was used to determine the size of j¼1
N 1
the standards. The DNA fragments were visualized using
silver staining according to the method of Creste et al. where N is the number of individuals and pj is the
(2001). frequency of the jth pattern (Tessier et al. 1999).
Tropical Plant Biol.

Molecular data were scored for each accession based on Chandra A, Tiwari KK (2010) Isolation and characterization of
microsatellite markers from guineagrass (Panicum maximum)
the presence or absence of the band. These data were
for genetic diversity estimate and cross-species amplification.
employed to generate a binary matrix for all pairwise Plant Breed 129:120–124
combinations and to calculate Jaccard’s similarity coeffi- Coelho ASG (2002) BOOD version 3.0. Avaliação de dendrogramas
cient (Jaccard 1908) using NTSYS-pc version 2.1 software baseados em estimativas de distâncias/similaridades genéticas
através do procedimento de bootstrap. Departamento de Biologia
(Rohlf 2000). This information was utilized in a principle
Geral, Instituto de Ciências Biológicas, Universidade Federal de
components analysis (PCA) using DARwin software v. Goiás, Goiânia, Goiás, Brasil.
5.0.157 (Perrier and Jacquemound-Collet 2006). The Combes D (1975) Polymorphisme et modes de reproduction dans la
reliability of the generated cluster was also tested using section des Maximae du genre Panicum (Gramineae) en Afrique.
Paris: Mémoires ORSTOM 77:1–99.
bootstrap analysis with the BooD program with 1,000 Combes D, Pernès J (1970) Variations dans le nombres chromosomi-
iterations (Coelho 2002). STRUCTURE software version ques du Panicum maximum Jacq. en relation avec le mode de
2.2 (Pritchard et al. 2000) was used to generate a Bayesian reproduction. Compt Rendus Acad Sci 270:782–785
inference of the population structure. This method identifies Creste S, Tulmann-Neto A, Figueira A (2001) Detection of single
sequence repeat polymorphisms in denaturing polyacrylamide
clusters of genetically similar individuals from multilocus
sequencing gels by silver staining. Plant Mol Biol Rep 19:299–
genotypes without prior knowledge of their population 306
affinities. The model assumes K genetic clusters, with each Doyle JJ, Doyle JL (1990) Isolation of plant DNA from fresh tissue.
cluster having a characteristic set of band frequencies at Focus 12:13–15
Duke JA (1983) Panicum maximum Jacq. (Poaceae: Guineagrass,
each locus; a no-admixture model with correlated band
Hamilgrass). Center for new crops and plant products. Purdue
frequencies was assumed. As a preliminary step, the University, West Lafayette, IN. Available at: http://www.hort.
analysis was performed for a number of genetic clusters purdue.edu/newcrop/duke_energy/Panicum_maximum.html
(K) ranging from 2 to 20. Consistent results across runs (accessed on March 20, 2009).
Ebina M, Kouki K, Tsuruta S, Akashi R, Yamamoto T, Takahara M,
were obtained using a burn-in period of 100,000 repeats, Inafuku M, Okumura K, Nakagawa H, Nakajima K (2007)
followed by 200,000 Markov Chain Monte Carlo (MCMC) Genetic relationship estimation in guineagrass (Panicum maxi-
repeats. The most probable number for K was calculated mum Jacq.) assessed on the basis of simple sequence repeat
based on the method of Evanno et al. (2005) using an ad markers. Grassland Science 53:155–164
Evanno G, Regnaut S, Goudet J (2005) Detecting the number of
hoc statistic, ΔK, which represents the rate of change in the
clusters of individuals using the software STRUCTURE: a
log probability of the data between successive K values simulation study. Mol Ecol 18:2611–2620
rather than the log probability of the data. An analysis of Excoffier L, Laval G, Schneider S (2005) Arlequin ver. 3.0: An
molecular variance (AMOVA) was conducted using ARLE- integrated software package for population genetics data analysis.
Evol Bioinformatics Online 1:47–50
QUIN 1.1 software (Excoffier et al. 2005). The level of
Euclides VPB, Macedo MCM, Valério JR, Bono JAM (2000) Massai
significance for variance component estimates was deter- cultivar (Panicum maximum) a new forage option: adaptation and
mined using non-parametric permutation procedures using productivity characteristics. In: Reunião da Sociedade Brasileira
1,000 permutations. de Zootecnia, 37. Viçosa. Anais… pp. 45−51.
Field D, Wills C (1996) Long, polymorphic microsatellites in simple
organisms. Proc Roy Soc 263:209–215
Acknowledgments The authors are grateful to the Fundação de
Gaitán-Solís E, Duque MC, Edwards KJ, Tohme J (2003) Micro-
Amparo à Pesquisa do Estado de São Paulo (FAPESP) for financial
satellite in common bean (Phaseolus vulgaris): isolation, char-
support (Project 05/51010-0) and for a graduate fellowship to A.C.B.
acterization, and cross-species amplification in Phaseolus ssp.
Sousa (06/52953-8). The authors thank the Conselho Nacional de
Crop Sci 42:2128–2136
Desenvolvimento Científico e Tecnológico (CNPq) for research
Gupta PK, Varshney RK (2000) The development and use of
fellowships awarded to A.P. Souza and L. Jank.
microsatellite markers for genetics analysis and plant breeding
with special emphasis on bread wheat. Euphytica 113:163–185
Jaccard P (1908) Nouvelles recherches sur la distribution florale.
References Bulletin de la Société Vandoise des Science Naturelles 44:223–
270
Jain A, Roy AK, Kaushal P, Malaviya DR, Zadoo SN (2006)
Billotte N, Lagoda PJL, Risterucci AM, Baurens FC (1999) Isoenzyme banding pattern and estimation of genetic diversity
Microsatellite-enriched libraries: applied methodology for the among guineagrass germplasm. Genet Resour Crop Evol 53:339–
development of SSR markers in tropical crops. Fruits 54:277–288 347
Bogdan AV (1977) Panicum maximum. In: Bogdan AV (ed) Tropical Jank L (1995) Melhoramento e seleção de variedades de Panicum
pasture and fodder plants. Longman, London, pp 181–191 maximum. In: Simpósio sobre manejo da pastagem, 12. Piraci-
Bolaric S, Barth S, Melchinger AE, Posselt UK (2005) Molecular caba. Anais… FEALQ. pp. 21–58.
genetic diversity within and among German ecotypes in Jank L, Costa JCG, Savidan YH, Valle CB (1993) New Panicum
comparison to European perennial ryegrass cultivars. Plant Breed maximum cultivars for diverse ecosystems in Brazil. In: Interna-
124:257–262 tional Grassland Congress. Palmerston North. Proceedings…
Burton GW, Millot JC, Monson WG (1973) Breeding procedures for New Zealand Grassland Association. 17:509–511.
Panicum maximum Jacq. suggested by plant variability and mode Jank L, Savidan YH, Souza MT, Costa JCG (1994) Avaliação do
of reproduction. Crop Sci 13:717–720 germoplasma de Panicum maximum introduzido da África: 1.
 Tropical Plant Biol.

Produção forrageira. Revista da Sociedade Brasileira de Zootecnia Rohlf FJ (2000) NTSYS-pc numerical taxonomy and multivariate
Viçosa, MG 23:433–440 analysis system, version 2.1.
Jank L, Calixto S, Costa JCG, Savidan YH, Curvo JBE (1997) Savidan Y (1975) Herédité de l’apomixie contribution a l’étude de
Catalog of the characterization and evaluation of the Panicum l’hérédité de l’apomixie sur Panicum maximum Jacq. (analysis
maximum germplasm: morphological description and agronom- dês sacs embryonnaires). Biology 10:91–95
ical performance. Campo Grande, MS: Embrapa Gado de Corte, Savidan YH (1982) Nature et hérédité de l´apomixie chez Panicum
53p. (Embrapa Gado de Corte. Documentos, 68). maximum Jacq. ORSTOM Travaux et Documentos, 153. ORS-
Koh HJ, Heu MH, McCouch SR (1996) Molecular mapping of the ges TOM, Paris.
gene controlling the super-giant embryo character in rice (Oryza Savidan YH (1983) Genetics and utilization of apomixis for the
sativa L.). Theor Appl Genet 93:257–261 improvement of guineagrass (Panicum maximum Jacq.). In: Smith
Mateescu RG, Zhang Z, Tsai K, Phavaphutanon J, Burton-Wurster JA and Hays VW (ed.) Proc. Int. Grassl. Gongr., 14th. Lexington,
NI, Lust G, Quaas R, Murphy K, Acland GM, Todhunter RJ KY. 15–24. Westview Press, Boulder, CO. pp. 182–184.
(2005) Analysis of allele fidelity, polymorphic information Savidan YH (2000) Apomixis: genetics and breeding. Plant Breed
content, and density of microsatellites in a genome-wide Rev 18:10–86
screening for hip dysplasia in a crossbreed pedigree. J Hered Savidan Y, Pernès J (1982) Diploid-tetraploid-haploid cycles and the
96:847–853 evolution of Panicum maximum Jacq. Evolution 36:596–600
Muir JP, Jank L (2004) Guineagrass. In: Moser LE, Burson BL, Savidan YH, Jank L, Costa JCG, Valle CB (1989) Breeding Panicum
Sollenberger LE (eds) Warm-Season (C4) Grasses. Agronomy maximum in Brazil: 1. Genetic resources, modes of reproduction
Monograpy, 45:589–621. American Society of Agronomy, Crop and breeding procedures. Euphytica 41:107–112
Science Society of America, Soil Science Society of America, Smith RL (1979) Seed dormancy in Panicum maximum Jacq. Trop
Madison, WI. Agr 56:233–239
Nakagawa H (1990) Embryo sac analysis and crossing procedure for Sousa ACB, Jungmann L, Campos T, Sforça DA, Boaventura LR,
breeding apomictic guineagrass (Panicum maximum Jacq.). Jpn Silva GMB, Zucchi MI, Jank L, Souza AP (2011) Development
Agr Res Q 24:163–168 of microsatellite markers in Guineagrass (Panicum maximum
Nakagawa H, Hanna W (1992) Induced sexual tetraploids for breeding Jacq.) and their transferability to other tropical forage grass
guineagrass (Panicum maximum Jacq.). Science Council of Japan species. Plant Breed 130:104–108
and Japanese Society of Grassland Science 38:152–158 Temnykh S, DeClerck G, Lukashova A, Lipovich L, Catinhour S,
Nakajima K (1978) Comparison of major agronomic characters in McCouch S (2001) Computational and experimental analysis of
guineagrass and colored guineagrass. Jpn Agr Res Q 12:145– microsatellites in rice (Oryza sativa L.): frequency, length
151 variation, transposon associations, and genetic marker potential.
Nakajima K, Komatsu N, Mochizuki N, Suzuki S (1979) Isolation of Genome 11:1441–1452
diploid and tetraploid sexual plants in guineagrass (Panicum Tessier C, David J, Boursiquot P, Charrieir AJM (1999) Optimizations
maximum Jacq.). Jpn J Breed 29:228–238 of the choice of molecular markers for varietal identification in
Pernès J (1975) Schéme d’amelioration génétiques des complexes Vitis vinifera L. Theor Appl Genet 98:171–177
agamiques du type Panicum. Cahiers ORSTOM Série Biology Varshney RK, Graner A, Sorrells ME (2005) Genic microsatellite
10:67–75 markers in plants: features and applications. Trends Biotechnol
Perrier X, Jacquemound-Collet JP (2006) DARwin software. Available 23:48–55
from http://www.darwin.cirad.fr/darwin. Young BA, Sherwood RT, Bashaw EC (1979) Cleared-pistil and thick-
Pritchard JK, Stephens M, Donnelly P (2000) Inference of population sectioning techniques for detecting aposporous apomixes in
structure using multilocus genotypes data. Genetics 155:945–959 grasses. Can J Bot 57:1668–1672

View publication stats