GENETIC VARIABILITY IN CORN LANDRACES FROM SOUTHERN BRAZIL

Maydica 53 (2008): 151-159 GENETIC VARIABILITY IN CORN LANDRACES FROM SOUTHERN BRAZIL P. Wietholter, M.J. Cruz de Melo Sereno, T. de Freitas Terra, S...
Author: Phillip Oliver
3 downloads 0 Views 195KB Size
Maydica 53 (2008): 151-159

GENETIC VARIABILITY IN CORN LANDRACES FROM SOUTHERN BRAZIL P. Wietholter, M.J. Cruz de Melo Sereno, T. de Freitas Terra, S. Delmar dos Anjos e Silva1, J.F. Barbosa Neto* Dep. Plantas de Lavoura, Faculdade de Agronomia, Universidade Federal do Rio Grande do Sul (UFRGS), Av. Bento Gonçalves, 7712, C.P. 776, CEP 91501-970 Porto Alegre, Brasil 1 Embrapa Clima Temperado, Embrapa, C. P. 403, 96.001-970 Pelotas, Brasil Received September 15, 2008

ABSTRACT - Corn (Zea mays subsp. mays) is one of the economically most important crop species in the world. It is the cultivated species that perhaps shows the greatest genetic variability and currently more than 250 landrace varieties have been reported. The objective of this study was to characterize the genetic variability in corn landrace varieties from Southern Brazil using phenotypic and molecular analysis. Eleven phenotypic traits were analyzed, of which eight showed significant differences, indicating distinction among the landrace varieties. Microsatellite (SSR) and AFLP markers were used in the molecular analysis. Both the phenotypic and molecular analyses indicated high genetic variability and also allowed the separation of the germplasm into groups of genetic similarity. The average similarity in the molecular markers analyses was 0.74. The results suggested that the varieties analyzed could be useful in corn genetic breeding programs. KEY WORDS: Zea mays; Genetic variability; SSR; AFLP.

INTRODUCTION Corn (Zea mays subsp. mays) is an important crop for human and animal feeding. This species is cross-pollinated, annual, monoecious with 2n = 20 chromosomes, and is currently considered as an ancient polyploid (GAUT et al., 2000). A striking characteristic is its wide genetic variability that resulted in more than 250 landrace varieties (PATERNIANI and GOODMAN, 1977). The origin of the corn genetic variability may be from polyploidization (GAUT et al., 2000) and from the intense presence of transposable elements (KIDWELL, 2002), while the development of the races is probably a consequence of hybridization and selection (PATERNIANI and GOODMAN, 1977). * For correspondence (fax: +55-51-3337-7519; e.mail: jfbn@ ufrgs.br).

Generally landrace varieties have lower yields than improved varieties and commercial hybrids. However, in addition to serving as an important food source, these varieties are fundamental for maize breeding programs, which frequently require germplasm to develop superior genotypes. According to NASS et al. (1993), only 14% of corn germplasm in Brazil has been used in breeding programs. One of the main reasons is the limited information regarding this germplasm. Therefore genetic variability analysis has been an important tool because it enables identification and characterization of genetic variability in this species (REIF et al., 2004). Estimates of genetic variability in corn varieties by phenotypic analysis have been conducted (CAMUSSI, 1979; ANDRADE et al., 2002; TEIXEIRA et al., 2002). However, although this type of analysis permits observation of a species in relation to plant architecture and performance under biotic or abiotic stresses, the environment effect is always a confounding factor. Molecular analysis has been an important tool in studies to identify genetic variability in corn. Molecular markers enable inferences regarding the genetic structure of the available population, which helps in planning strategies to germplasm conservation and use. At the beginning of the 1990s RFLP were used to identify genetic similarity in corn lines (SMITH et al., 1990). CARVALHO et al. (2004) analyzed 79 corn landraces and suggested that management practices used by small farmers in Brazil have contributed to maintain the genetic variability in this germplasm. Similarly, microsatellite markers (SSR) have been useful in evolutionary studies on the Zea genus. MATSUOKA et al. (2002) analyzed populations from all the species of the Zea genus and indicated that the SSR were powerful phylogenetic molecular markers both for intra- and interspecific studies of the genus. The use of SSR molecular markers also

152

P. WIETHOLTER, M.J. CRUZ DE MELO SERENO, T. DE FREITAS TERRA, S. DELMAR DOS ANJOS E SILVA, J.F. BARBOSA NETO

allowed the determination of the genetic diversity of the Mexico corn accessions (REIF et al., 2006). Some studies carried out recently based on molecular markers have showed significant reduction in the genetic variability of maize (TENAILLON et al., 2001; WARBURTON et al., 2008). All these results indicated that the germplasm identification can be used as source of genetic variability for breeding programs. According to WARBURTON et al. (2008), there are many single alleles present in landrace varieties that are not being exploited that could contribute significantly to the development of new varieties and hybrids. In the last 10 years a large number of landraces germplasm maintained by farmers for decades has been collected by our program. The objective of this study was to evaluate the genetic variability present in these landraces using phenotypic and molecular analyses aiming to contribute with useful germplasm to maize breeding programs. MATERIAL AND METHODS Field experiments were conducted at the Estação Experimental of Universidade Federal do Rio Grande do Sul (EEA/UFRGS), located in Eldorado do Sul (Brazil), and at Embrapa Clima Temperado, located in Pelotas (Brazil). Thirty-seven corn landraces from different Southern Brazil regions were collected by Embrapa Clima Temperado (Brazil) and FEPAGRO (Brazil). Plant material consisted of these landraces and four control varieties (Table 1). The control genotypes were two improved open pollinated varieties (BR-451 and Pampa) and two commercial hybrids (NB 3311 and Tork). A completely randomized design with three replications was used. Each plot consisted of two 5 m rows, separated by 0.7 m. Thinning to one plant each 20 cm was done about 20 days after sowing keeping 25 plants per plot. At the V3 stage (RITCHIE et al., 1992) 100 kg/ha nitrogen was applied. The analyzed traits were selected based on their importance for corn breeding programs and were closely linked to aspects of adaptation and productivity in southern Brazil. The following phenotypic traits were assessed: female and male flowering (degrees day), plant height (m), ear height (m), percentage of root and stalk lodging, percentage of rotten ears, kernel weight (mg), ear length (cm), ear diameter (cm), and number of rows per ear. In the experiments conducted at Embrapa Clima Temperado only plant height, ear height, and percentage of root and stalk lodging were assessed. The results were submitted to analysis of variance and the means were compared by Tukey test at 5% of probability. The genetic variability was estimated based on Euclidian test which is recommended for the phenotipics quantitative traits (CROSSA and FRANCO, 2004). The plant material for molecular analysis was extracted from the experiments conducted at EEA/UFRGS in Eldorado do Sul. Leaves were collected from 20 plants per genotype and kept in plastic bags containing silica to dehydrate the material. The DNA

was extracted from a bulk of 20 plants per genotype according to the MURRAY and THOMPSON (1980) protocol and quantified by comparison with DNA lambda in 0.9% agarose gel. The DNA working solutions were standardized at the concentration of 10 ηg/μl. The molecular markers assessed were the microsatellite (SSR) and amplified fragment length polymorphism (AFLP) types. SSR primers were selected accordingly their distribution on the maize chromosomes and, whenever possible, SSR were related to enzymes involved in important metabolic processes, as flooding tolerance. The SSR reactions were prepared for a 20 μl volume. Each reaction contained 10.2 μl sterilized water, 2 μl Buffer 10 X (Invitrogen), 0.6 μl MgCl2 (Invitrogen), 0.4 μl dNTP mix (Invitrogen), 0.3 μl F primer (forward), 0.3 μl R primer (reverse), 0.2 μl Taq-DNA Polymerase enzyme (Invitrogen, 5u/μl), and 6 μl de DNA. The DNA was amplified using a touchdown type program in a thermocycler. The amplified DNA fragments were separated in 3% agarose gel, stained with ethydium bromide and visualized under ultraviolet light. The 100 pb DNA Ladder marker (Invitrogen) was used as molecular weight standard. PCR reactions and gel visualization were carried out for all individuals together for each primer. Twenty-one primers distributed on all the corn chromosomes were tested. The AFLP reactions were conducted according to the protocol by VOS et al. (1995) using a combination of the Mse1 and Pst1 restriction enzymes. The selective amplification was performed using three different primer combinations (P1+M1, P1+M2, and P1+M3, where: P1 was 5’-GACTGCGTAGGTGCAGAAA-3’, M1 was 5’-GATGAGTCCTGAGTAACAC-3’, M2 was 5’GATGAGTCCTGAGTAACAG-3’, and M3 was 5’-GATGAGTCCTGAGTAACCG-3’). The amplified fragments were separated in acrylamide gel with silver nitrate staining. The number of loci generated was quantified by primer combinations. The genetic variability was estimated based on the SSR and AFLP data using the NEI and LI (1979) genetic distance. A table was constructed to join the data of the two markers where each SSR primer was considered as a locus. For the ALFP marker, each band observed in the gel was considered as a locus and only the presence or absence of this band was assessed in each population.

RESULTS AND DISCUSSION Landrace varieties are populations used by farmers over several generations resulting also in selection for innumerous traits (HARLAN, 1992). This selection was probably most effective for qualitative traits, due to their high heritability. In the landraces analyzed in this study, grain colors were white, yellow, orange, purple, and striped, while the grain type ranged from dent, semi-dent, flint, semi-flint, and floury (Table 1). The color and grain type variables can be considered qualitative traits, and are governed by fewer genes with little environmental effect. This fact allowed more efficient selection by the farmers, contributing to stabilizing the traits in each population. Analysis of variance indicated significant differ-

GENETIC VARIABILITY IN CORN LANDRACES

153

TABLE 1 - Abbreviation, region of collection (Santa Catarina State - SC, Serra do Nordeste/Rio Grande do Sul State - SN, Planalto Médio/Rio Grande do Sul State - PM, Vale do Uruguai/Rio Grande do Sul State - VU, Grandes Lagoas/ Rio Grande do Sul State - GL, and Litoral/Rio Grande do Sul State - LI), grain type (dent - D, flint - F, semi-dent - SD, semi-flint - SF, and floury - FLO), and grain color (white - W, yellow - Y, orange - OR, striped - ST, and purple - PU) for the 41 populations under study. ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Population Abbreviation Region of Collection Grain Type Grain Color ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 8 Carreiras Amarelo 8CA SN D Y 8 Carreiras Branco

8CB

SC

D

W

Açoriano Branco

AcoBr

LI

FLO

W

Amarelão

Amar

PM

D

Y

Amarelão Comum

AmarC

GL

D

Y

Argentino Flint

ArgFl

GL

F

OR

Assis Brasil (1)

AB31

GL

F

OR

Assis Brasil (2)

AB16

GL

F

OR

Bico de Ouro

BOuro

SN

D

OR

BR – 451

BR451

GL

SD

W

Branco Argentino

BrArg

GL

F

W

Branco Duro

BrDur

GL

F

W

Brasino

Bras

SN

D

ST

Cabo Roxo

CabRx

SC

SD

PU

Cabo Roxo Misto

CbRxM

SN

D

PU

Caiano Amarelo

CaiAm

VU

D

Y

Caiano Branco

CaiBr

VU

D

W

Caiano Rajado

CaiRj

GL

D

ST

Catete Amarelo

CatAm

GL

D

Y

Cinquentinha

Cinqu

SN

D

W

Col1

GL

SF

PU,Y

Colonial (1) Colonial (2)

Col2

GL

F,SD

PU,ST

Colonial Vermelho

ColV

VU

F,D

PU

ComAm

PM

D

Y

Comum Amarelo Cultivar Brancão

CB

SN

D

W

Cunha

Cun1

SN

D

Y

Cunha

Cun2

LI

D

Y

Cunha Sabugo Duplo

CunSD

PM

D

Y

Cunha Sabugo Fino

CunSF

PM

D

Y

Dente de Cão

DCao

VU

D

Y

Dente de Ouro

DOuro

GL

D

OR

Ferro

SN

F

OR

Pampa

GL

D

ST

POP - 5 (1)

P544

GL

F

W

POP - 5 (2)

P521

GL

F

W

Pururuca Branco

PurBr

SN

F

W

Ferro Pampa

Roxo Índio

RxIn

GL

FL,SD,D

PU

Sabuguinho

Sabug

PM

D

Y

Sabuguinho Amarelo

SabAm

SN

D

Y

NB 3311

N3311

-

SD

Y

TORK Tork SD Y –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

154

P. WIETHOLTER, M.J. CRUZ DE MELO SERENO, T. DE FREITAS TERRA, S. DELMAR DOS ANJOS E SILVA, J.F. BARBOSA NETO

TABLE 2 – Means for plant height (m), ear height (m), male flowering (degrees day), female flowering (degrees day), kernel weight (mg), ear length (cm), and number of rows per ear in 41 corn populations. ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Plant Ear Male Female Kernel Ear Number Populations Height Height Flowering Flowering Weight Lenght of Rows ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 8CB 2.20 1.26 816 891 367 19 8 CaiBr

2.45

1.46

1004

1054

433

18

Cinqu

2.09

1.11

816

891

300

14

14 14

PurBr

2.09

1.26

930

1031

333

18

14

Ferro

2.16

1.31

939

1040

333

18

14

Amar

2.60

1.50

987

1040

400

18

14

8CA

2.17

1.17

930

966

333

18

8

DCao

2.20

1.24

987

1040

300

15

24

BOuro

2.36

1.48

987

1040

366

19

14

Cun1

2.30

1.24

987

1040

266

16

18

Sabug

2.63

1.64

987

1054

333

17

12

Bras

2.46

1.50

1005

1040

333

18

14

CabRx

2.49

1.50

987

1054

333

17

14

BR451

1.83

1.05

987

1040

333

17

16

AcoBr

1.75

1.04

882

1040

266

16

12

BrDur

2.08

1.33

987

1040

333

19

12

P544

2.04

1.12

930

1040

300

19

14

P521

2.00

1.11

882

1040

300

19

16

AB31

1.74

0.99

1005

1040

333

19

18

AmarC

2.13

1.23

1013

1011

366

17

12

AB16

1.87

1.09

987

966

300

20

12

Col1

1.90

1.09

873

966

333

19

14

Col2

2.05

1.18

996

1040

366

21

14

ColV

2.14

1.27

1005

1040

366

19

16

CaiBr

2.12

1.31

1005

1040

366

15

14

BrArg

2.29

1.37

930

966

300

17

12

ArgFl

2.21

1.36

930

1040

400

17

12

DOuro

2.03

1.16

873

966

300

19

14

CatAm

2.28

1.31

930

966

366

17

12

RxIn

2.19

1.31

987

1040

333

19

12

CaiRj

2.29

1.39

987

1040

400

18

15

Pampa

2.23

1.26

987

1040

300

21

14

CbRxM

1.96

1.14

996

1040

266

14

14

ComAm

1.75

1.11

1018

1054

366

18

12

Cun2

1.80

1.10

996

1040

233

13

12

CunSD

2.16

1.14

873

1040

300

16

18

CunSF

1.93

1.15

1005

1040

300

19

12

CaiAm

1.90

1.19

996

1040

300

14

12

SabAm

2.89

1.80

828

1246

204

15

14

N3311

1.89

0.96

816

873

300

18

16

Tork

1.68

1.01

891

966

300

20

18

Mean

2.13

1.25

950

1021

326

18

14

LSD 5% 0.49 0.43 200.7 259.8 19 5.7 0.8 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

GENETIC VARIABILITY IN CORN LANDRACES

ences for plant height, ear height, male and female flowering, kernel weight, ear length, number of rows per ear, and root lodging. These results suggested the presence of genetic variability among the assessed genotypes (MELO et al., 2001). The mean plant height was 2.13 m, with values ranging from 1.68 m (Tork) to 2.89 m (SabAm). The ear height presented a mean value of 1.25 m ranging from 0.96 m to 1.80 m. Female flowering ranged from 873 degrees day (N3311) to 1246 degrees day (SabAm), while male flowering ranged from 816 degrees day (8CB, Cinqu and N3311) to 1018 degrees day (ComAm). The grain weight varied from 266 mg (CbRxM) to 433 mg (CaiBr). The ear length values varied from 13 cm (Cun2) to 21 cm (Col2 and Pampa). The amplitude observed in the number of rows per ear trait was eight (8CB) to 24 (DCao) (Table 2). Variation for percentage of rotten ears, stalk lodging, and ear diameter was not significant among the genotypes. Root lodging was significant, but not important because it was observed only in the BOuro genotype (15%), which also had more stalk lodging (14%). Ear diameter was not variable probably due to the measuring method that considered the cob and grains together, and did not take into consideration variation in the cob and grain depths. AZAR et al. (1997) observed significant variation in 35 corn races (white grain and floury endosperm) for flowering (male and female), weight of 100 kernels, ear length and diameter, plant and ear height, and other traits not assessed in these experiments. Generally, our results confirmed the existence of variability in the analyzed landraces and were in agreement with other studies (ANDRADE et al., 2002; TEIXEIRA et al., 2002). The joint analysis of the experiments in Eldorado do Sul and Pelotas did not show the existence of genotype x location interaction. FEDERIZZI et al. (1993) also did not detect genotype x location interaction in oat (Avena sativa) in Southern Brazil, but the authors observed strong genotype x year interaction. Otherwise, MELO et al. (2001) evaluated several agronomic traits in ten corn hybrids in Lavras and Ijaci, both in Minas Gerais (Brazil), and detected the presence of cultivar x location interactions. In our study, the genotype x year interaction could not be assessed because the experiment was carried out in only one year. However, the objective of the present study was initially to assess the germplasm variability of the Southern Brazil and thus the genotype x year interaction was not considered relevant,

155

FIGURE 1 - Clustering obtained by the genetic similarity matrix from the data generated based on the phenotypic assessments based on Euclidian Distance.

because the group of superior populations could be maintained in subsequent years. The refinement of the analysis with some selected populations would require more detailed assessment of the genotype x year interaction. Furthermore, the traits assessed, especially plant height, flowering time and ear height, have medium to high heredity (LOCATELLI et al., 2002; GAMA et al., 2003) and are not particularly affected by the genotype x year interaction (LOCATELLI et al., 2002). Based on the phenotypic evaluation the germplasm was grouped according to its divergence. This type of analysis is important because it prevents including many crosses, leading to reduced costs in the breeding program in addition to improving the assessment precision. The estimation of the Euclidian distance generated a dendrogram where six groups were observed (Fig. 1). The first group consisted of only the SabAm genotype that was very different and inferior to the others regard-

156

P. WIETHOLTER, M.J. CRUZ DE MELO SERENO, T. DE FREITAS TERRA, S. DELMAR DOS ANJOS E SILVA, J.F. BARBOSA NETO

TABLE 3 - Microsatellite primers used in the analysis, locus, chromosome location, number of alleles detected, and size of the bands. ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Primer Loci Location (bin) Nº alleles Size (pb) ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– umc 1547 Sterol methyl transferase 2 1.01 2 60 - 84 umc 1026 umc1026 2.04 2 130 - 150 umc 1683 umc1683 3.04 4 90 - 130 phi 073 Glutathione-s-transferase 3.05 1 93 umc 1594 umc1594 3.10 2 130 - 150 umc 1294 umc1294 4.02 2 270 - 300 nc 004 Alcohol desydrogenase2 4.03 4 150 - 200 phi 074 Zein protein 22.1 4.04 3 90 - 210 umc 1173 Histone deacetylase homolog 4.09 3 150 - 170 umc 1197 Catalase 3 4.11 2 80 - 105 umc 1097 umc1097 5.00 2 100 - 130 umc 1056 Peroxidase 13 5.03 6 110 - 160 umc 1341 Replication origin activator 2 6.05 - 6.06 3 105 - 140 umc 1545 Heat shock protein 3 7.00 3 50 - 90 umc 1433 umc1433 7.02 2 50 - 90 umc 1627 Oxygen evolving complex 23 8.03 2 150 - 170 umc 1172 Pyruvate decarboxylase 1 8.04 2 150 - 170 umc 1636 umc1636 9.02 3 100 - 150 phi 065 Phosphoenolpyruvate carboxylase 1 9.03 3 140 - 160 phi 016 Sucrose synthase 1 9.04 4 140 - 180 phi 032 Sucrose synthase 1 9.04 2 190 - 250 Mean – – 2.7 – –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

ing the traits analyzed. The second group also consisted of a single genotype (CunSF). The third group was formed by the flint genotypes Cun2 and CbRxM, which were characterized by high plant height and ear insertion and late female and male flowering. The fourth group was formed by genotypes that in general were characterized as medium female flowering and average plant and ear height. Two subgroups were observed in this group, one presenting only white kernel genotypes and the other containing white and yellow kernel genotypes. The fifth group was formed by two hybrids (N3311 and Tork) and two landrace varieties (CunSD and ColV) where the hybrids formed one subgroup and the landrace varieties another. The subgroup formed by the hybrids was characterized by low plant height, low ear height, early female and male flowering and high row number per ear. The two landrace varieties in the second subgroup were characterized by presenting medium to high plant height and high row number per ear. The row number trait was probably responsible for the grouping of these four genotypes. Apparently, none of the other traits contributed to the formation of

this group, showing the difference of the hybrid testers compared to the landrace varieties. The last and largest group contained the two open pollination controls (Pampa and BR451) and the remaining 22 landrace varieties. This group was similar in performance regarding the traits male and female flowering and plant height. Three subgroups were observed within this last group. It could be observed that approximately 80% of the varieties that formed the first subgroup (DCao, BOuro, ArgFl, Cun1, CaiRj, CatAm, Bras, Douro, RxIn, Pampa and Col1) presented a flint-type grain and the varieties that made up the second subgroup (Cinqu, CabRx, AB16 and Amar) performed similarly regarding the kernel row number per ear. No specific trait could be found among those analyzed that could have contributed to the formation of the third subgroup. Apparently the traits that most contributed to the formation of the groups were plant height, ear insertion, female flowering, male flowering and kernel row number per ear. The other traits (root and stalk lodging, ear length, ear diameter, and kernel weight) showed little contribution to the formation of the groups and subgroups.

GENETIC VARIABILITY IN CORN LANDRACES

Regarding the molecular analysis, in the SSR analysis 60 alleles were detected in the 23 loci analyzed, with an average of 2.7 alleles per locus (Table 3). REIF et al. (2006) reported a mean of 7.8 alleles per locus in 25 accessions of Mexican races. MATSUOKA et al. (2002) reported an average of 12.6 in 16 teosinto and corn race populations and 6.9 in 101 hybrids. One of the factors that could have lead to a larger number of alleles per locus was the number of genotypes analyzed, that is, the more genotypes analyzed, larger the possibility of finding different alleles. However, only the hybrids in the study by MATSUOKA et al. (2002) had a superior number of genotypes (101) compared with the number of landrace varieties assessed in the present study (37), indicating that the low average number of alleles per locus found may have other causes. A factor that might be related was the relatively recent origin of the varieties studied. The close origin and the short period for differentiation might have led to lower levels of allele variation. Another factor was the elimination of bands with size different than expected based on the SSR primer description, resulting in elimination of real bands. In addition, the analytical method adopted in this study using 3% agarose gel may have contributed to the lower number of alleles detected. Polyacrylamide gels have a larger allele detection capacity with few base pairs of difference. However, although we know that SSR analysis in agarose gels may result in underestimation of the variability, there are several studies that indicate its use, especially because it is a cheap and a quick technique (SENIOR et al., 1998). LEGESSE et al. (2007) assessed 56 corn genotypes from Ethiopia and Zimbabwe with 27 SSR loci (3% agarose gel) and reported a mean of 3.85 band/locus. In this study, the grouping pattern was consistent with the information available from the pedigree and it was concluded that the variability detected using the SSR could contribute to the effective use of the lines to exploit the heterosis and the formation of genetically diverse populations in the corn breeding programs in Ethiopia. MENKIR et al. (2004) also assessed genetically diversity in mid altitude tropical corn lines with SSR markers and using 2% agarose gel to separate into heterotic groups. They observed clustering as expected, and concluded that the grouping according to the molecular data could serve as a base to establish heterotic groups with larger genetic similarity within the groups. LABORDA et al. (2005) assessed the genetic diversity in 85 tropical corn genotypes using the AFLP and

157

FIGURE 2 - Clustering obtained by the genetic similarity matrix from data generated by the SSR and AFLP markers based on the Nei and Li coefficient.

SSR markers and agarose gels at 4%. Similarly to other analysis they observed that the molecular markers were useful to organize the genetic variability in order to use in plant breeding programs. Furthermore, the SSR primers used might have presented little polymorphism in the group of varieties analyzed. Variation of two to six alleles per locus was observed. The allele sizes ranged from 60 to 300 base pairs (Table 3). Regarding the AFLP, the primer combination analyzed determined the occurrence of 136 polymorphic bands, with an average of 45 bands per combination. The loci ranged from 250 to 1600 base pairs. The NEI and LI (1979) coefficient of similarity was estimated from a data set joining SSR and AFLP analysis. The values ranged from 0.35 (SabAm and Tork) to 0.88 (Ferro and Amar), where the average value was 0.74. Cluster analysis produced seven groups, considering the average value of 0.74 as the cutoff point (Fig. 2). Five of the seven groups consisted of a single genotype (PurBr, CbRxM, N3311, Tork, and SabAm). Another group consisted of the

158

P. WIETHOLTER, M.J. CRUZ DE MELO SERENO, T. DE FREITAS TERRA, S. DELMAR DOS ANJOS E SILVA, J.F. BARBOSA NETO

landraces Cun2, CunSD, CunSF, and CaiAm. The last group consisted of the other genotypes. No clustering by location of collection, grain type or color was observed, nor regarding the other traits assessed in the field. This fact occurred because the molecular markers used were chosen randomly and not in function of the agronomic traits assessed. Although it was not the objective of this study, the landraces could not be grouped based on the race groups reported by PATERNIANI and GOODMAN (1977) and SANCHÉZ et al. (2007). The experiment did not include tester populations from these races and this analysis was impossible. The most similar genotypes in relation to the agronomic traits (DCao/BOuro) also presented high similarity in the molecular assessment (0.83). Similarly, the least similar genotypes for the agronomic traits (SabAm e Tork) also presented low similarity in the molecular assessment (0.35), indicating that although the primers used were chosen randomly and the genotype disposition in the dendogram was different, the results obtained in both the analyses were consistent. Both the phenotypic and the molecular assessment showed a relatively high average similarity value (0.66 and 0.74, respectively) and were approximately related with the values reported in other genetic similarity analysis studies on landrace varieties of the species (MELO et al., 2001; CARVALHO et al., 2002, 2004). In general, corn landrace varieties presenting around 70% genetic similarity have been reported as of considerable genetic variability. Thus the landrace varieties analyzed in this study could be very useful for breeding new superior varieties. Genotypes that could have potential as sources of genetic variability in corn breeding programs, based on phenotypic data, included 8CB for earliness, and plant and ear height; CaiBr for kernel weight; Col2 for ear length, and DCao for the number of rows per ear. One of the main limiting factors to the use of landrace varieties in breeding programs is the fear that the genetic variability, instead of contributing to specific traits, may destroy a set of important traits already fixed in the commercial varieties, such as earliness and plant height. This fact may lead to a reduction in the population mean for these traits and even break linkage blocks of important and fundamental traits for adaptation. However, this study indicted that, regarding the phenotypic traits, the majority of the landrace varieties were not statistically different from the tester hybrids, suggesting that the landrace germplasm would not damage the previously established traits. Taking into consideration the richness of

genetic variability that exists within the species and that few divergences were observed between the landrace and breeding material for the basic adaptation traits, there are certainly many other traits that were not assessed in this study that could be incorporated into the germplasm in the breeding programs, without causing the loss of those desirable traits already available in the programs. The distribution of the genotypes in the dendrogram generated from the phenotypic and molecular data was different. However, it was observed that many landrace varieties were not significantly different from the hybrids, considering the main adaptive traits (time of flowering, plant and ear height). Otherwise, the landrace populations were very different from the commercial material based on the molecular data, indicating that there may be a considerable portion of variability that has not yet been exploited. Thus the landrace varieties that were least different from the hybrids regarding the adaptive traits assessed, but that showed genetic divergence by molecular markers, could be used to analyze specific traits. These landraces would not decrease the population mean for the adaptive traits and could, because of their molecular divergence, contribute with others of importance including disease resistance, flooding tolerance, drought tolerance, and nutritional quality. This is supported by the fact that at least 50% of the SSR markers assessed are related to enzymes involved in important metabolic pathways. Thus this germplasm should be investigated more thoroughly to detect traits in its genetic diversity that has potential for use in breeding programs. Based on this information it would be possible to introduce these useful genes from the landrace varieties into commercial genotypes using breeding techniques, as for example the AB-QTL technique (TANKSLEY and NELSON, 1996). ACKNOWLEDGMENTS - The authors thank the CNPq and CAPES for funding.

REFERENCES ANDRADE R.V., M.X. SANTOS, A.S. FERREIRA, A.C. OLIVEIRA, 2002 Avaliação de acessos de milho crioulo coletados na região central do Brasil. Rev. Bras. Milho Sorgo 1: 67-74. AZAR C., D.E. MATHER, R.I. HAMILTON, 1997 Maize landraces of the St. Lawrence-Great Lakes region of North America. Euphytica 98: 141-148. CAMUSSI A., 1979 Numerical taxonomy of italian populations of maize based on quantitative traits. Maydica 24: 161-174.

GENETIC VARIABILITY IN CORN LANDRACES

CARVALHO V.P., P.M. RUAS, C.F. RUAS, J.M. FERREIRA, R.M.P. MOREIRA, 2002 Assessment of genetic diversity in maize (Zea mays L.) landraces using inter simple sequence repeat (ISSR) markers. Crop Breed. Appl. Biotech. 2: 553-564. CARVALHO V.P., C.F. RUAS, J.M. FERREIRA, R.M.P. MOREIRA, P.M. RUAS, 2004 Genetic diversity among maize (Zea mays L.) landraces assessed by RAPD markers. Genet. Mol. Biol. 27: 228-236. CROSSA J., J. FRANCO, 2004 Statistical methods for classifying genotypes. Euphytica 137: 19-37. FEDERIZZI L.C., J.F. BARBOSA NETO, F.I.F. CARVALHO, L.V.M. VIAU, J.L. SEVERO, E.L. FLOSS, A. ALVES, J. ALMEIDA, A.C. SILVA, 1993 Estabilidade do rendimento de grãos em aveia: efeito do uso de fungicidas. Pesq. Agropec. Bras. 28: 465-472. GAMA E.E.G., M.X. SANTOS, R.G. FERRÃO, W.F. MEIRELES, C.A.P. PACHECO, S.N. PARENTONI, P.E.O. GUIMARÃES, 2003 Potencial genético de um sintético de milho de grãos duros para formação de híbridos. Ciência Rural 33: 615-619. GAUT B.S., L.T. MAUD, A. PEEK, M.C. SAWKINS, 2000 Maize as a model for the evolution of plant nuclear genomes. Proc. Nat. Acad. Sci. USA 97: 7008-7015. GOWER J.C., 1971 A general coefficient of similarity and some of its properties. Biometrics 27: 857-874. HARLAN J.R., 1992 Crops and man. Madison: American Society of Agronomy. 284 p. KIDWELL M.G., 2002 Transposable elements and the evolution of genome size in eukaryotes. Genetica 115: 49-63. LABORDA P.R., K.M. OLIVEIRA, A.A.F. GARCIA, M.E.A.G.Z. PATERNIANI, A.P. SOUZA, 200 Tropical maize germplasm: what can we say about its genetic diversity in the light of molecular markers? Theor. Appl. Genet. 111: 1288-1299. LEGESSE B.W., A.A. MYBURG, K.V. PIXLEY, A.M. BOTHA, 2007 Genetic diversity of African maize inbred lines revealed by SSR markers. Hereditas 144: 10-17. LOCATELLI A.B., L.C. FEDERIZZI, V. NASPOLINI FILHO, 2002 Capacidade combinatória de nove linhagens endogâmicas de milho (Zea mays L.) em dois ambientes. Ciência Rural 32: 365-370. MATSUOKA Y., S.E. MITCHELL, S. KRESOVICH, M. GOODMAN, J. DOEBLEY, 2002 Microssatelites in Zea - variability, patterns of mutations, and use for evolutionary studies. Theor Appl Genet. 104: 436-450. MELO W.M.C., R.G.V. PINHO, J.B. SANTOS, D.F. FERREIRA, 2001 Utilização de caracteres morfoagronômicos e marcadores moleculares para a avaliação da divergência genética entre híbridos de milho. Rev. Ceres 48: 195-207. MENKIR A., A. MELAKE-BERHAN, C. THE, I. INGELBRECHT, A. ADEPOJU, 2004 Grouping of tropical mid-altitude maize inbred lines on the basis of yield data and molecular markers. Theor. Appl. Genet. 108: 1582-1590. MURRAY M., W.F. THOMPSON, 1980 Rapid isolation of high-molecular-weight plant DNA. Nuc. Acid Res. 8: 4321-4325.

159

NASS L.L., I.J. PELLICANO, A.C.C. VALOIS, 1993 Utilization of genetic resources for maize and soybean breeding in Brazil. Rev. Bras. Genet. 16: 983-988. NEI M., W.H. LI, 1979 Mathematical model for studying genetic variation in terms of restriction endonucleases. Proc. Nat. Acad. Sci. USA 70: 3321-3323. PATERNIANI E., M.M. GOODMAN, 1977 Races of maize in Brazil and adjacent áreas. Mexico: Centro Internacional de Mejoramiento de Maiz y Trigo. 95 p. REIF J.C., X.C. XIA, A.E. MELCHINGER, M.L. WARBURTON, D.A. HOISINGTON, D. BECK, M. BOHN, M. FRISCH, 2004 Genetic diversity determined within and among CIMMYT maize populations of tropical, subtropical and temperate germoplasm by SSR markers. Crop Sci. 44: 326-334. REIF J.C., M.L. WARBURTON, X.C. XIA, D.A. HOISINGTON, J. CROSSA, S. TABA, J. MUMINOVIÉ, M. BOHN, M. FRISCH, A.E. MELCHINGER, 2006 Grouping of accessions of Mexican races of maize revisited with SSR markers. Theor. Appl. Genet. 113: 177-185. RITCHIE S.W., G.O. BENSON, S. LUPKES, 1992 How Corn Plant Develops. Special Report nº48, Iowa State University Cooperative Extension Service, Ames, USA. SÁNCHEZ G.J.J., M.M. GOODMAN, C.W. STUBER, 2007 Racial diversity of maize in Brazil and adjacent areas. Maydica 52: 13-30. SENIOR M.L., J.P. MURPHY, M.M. GOODMAN, C.W. STUBER, 1998 Utility of SSRs for determining genetic similarities and relationships in maize using an agarose gel system. Crop Sci. 38: 1088-1099. SMITH O.S., J.S.C. SMITH, L.S. BOWEN, R.A. TENBORG, S.J. WALL, 1990 Similarities among a group of elite maize inbreds as measured by pedigree, F1 grain yield, grain yield, heterosis, and RFLPs. Theor. App. Genet. 80: 833-840. TANSKLEY S.D., J.C. NELSON, 1996 Advanced backcrosses QTL analysis: a method for the simultaneous Discovery and transfer of valuable QTLs from unadapted germplasm into the elite breeding lines, 1996. Theor. App. Genet. 92: 191-203. TEIXEIRA F.F., R.V. ANDRADE, A.C. OLIVEIRA, A.S. FERREIRA, M.X. SANTOS, 2002 Diversidade no germoplasma de milho coletado na região nordeste do Brasil. Rev. Bras. Milho Sorgo 1: 59-67. TENAILLON M.I., M.C. SAWKINS, A.D. LONG, R.L. GAUT, J.F. DOEBLEY, B.S. GAUT, 2001 Patterns of DNA sequence polymorphism along chromosome 1 of maize (Zea mays ssp. mays L.). Proc. Nat. Acad. Sci. USA 98: 9161-9166. VOS P., R. HOGERS, M. BLEEKER, M. REIJANS, T. LEE, M. HORNES, A. FRIJTERS, J. POT., J. PELEMAN, M. KUIPER, M. ZABEAU, 1995 AFLP: a new technique for DNA fingerprinting. Nuc. Acid Res. 23: 4407-4414. WARBURTON M.L., J.C. REIF, M. FRISCH, M. BOHN, C. BEDOYA, X.C. XIA, J. CROSSA, J. FRANCO, D. HOISINGTON, K. PIXLEY, S. TABA, A.E. MELCHINGER., 2008 Genetic diversity in CIMMYT nontemperate maize germoplasma: landraces, open pollinated varieties, and inbred lines. Crop Sci. 48: 617-624.

Suggest Documents