Crop Science Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Popi, J.
Right arrow Articles by Pauls, K.P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Popi, J.
Right arrow Articles by Pauls, K.P.
Agricola
Right arrow Articles by Popi, J.
Right arrow Articles by Pauls, K.P.
Crop Science 40:619-625 (2000)
© 2000 Crop Science Society of America

CROP BREEDING, GENETICS & CYTOLOGY

Random Amplified Polymorphic DNA–Based Evaluation of Diversity in the Hierarchical, Open-Ended Population Enrichment Maize Breeding System

J. Popia, J. Rajnprehtb, L.W. Kannenbergb and K.P. Paulsb

a Monsanto Canada Inc., 1288 Glanworth Drive, London, ON, N6N 1H1, Canada
b Dep. of Plant Agriculture, Univ. of Guelph, ON, N1G 2W1, Canada. Contribution of the Dep. of Plant Agriculture, Univ. of Guelph, Guelph, ON, N1G 2W1, Canada

jpopi{at}dekalb.com


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 Materials and methods
 Results and discussion
 REFERENCES
 
Commercial maize germplasm has a narrow genetic base. The hierarchical, open-ended population enrichment (HOPE) maize (Zea mays L.) breeding system was created with the dual purpose of developing useful inbred lines and of diversifying the maize breeding germplasm. The objective of this study was to estimate the genetic diversity in the HOPE populations at initiation and after 15 yr of operation. The HOPE system has two heterotic population sets (A and B). Each set has a hierarchical structure and consists of four open-ended populations: elite (E), high (H), intermediate (I), and low (L). New germplasm is added to the system, and high-performing entries from a lower level can be advanced to the next higher level. Increasingly stringent recurrent selection procedures are employed at each higher level of the hierarchy. Twelve random amplified polymorphic DNA (RAPD) primers were used to analyze 19 plants from the initial and last cycle of selection for each of the four levels of the two population sets. The greatest polymorphism (65.9%) was observed for the last selection cycle of the L level of the B set, while the last selection cycle of the E level of the A set was the least polymorphic (39.2%). Intrapopulation diversity was the greatest for the L level, and the smallest for the E level. Diversity appeared to be enhanced due to introgressions and advancements of material in the hierarchy. Principal component (PC) analysis indicated that the populations of the lower levels were more similar to one another than were the two E populations. All polymorphisms present in commercial check hybrids were also observed in the HOPE populations, while 32 of the observed 91 bands were polymorphic in the HOPE system and monomorphic in the hybrids. This study indicates that HOPE populations offer variability not present in commercial material and supports the HOPE system as a source of nontraditional germplasm for maize breeding programs.

Abbreviations: Dp, intrapopulation diversity • E, elite • H, high • HOPE, hierarchical open-ended population enrichment • I, intermediate • L, low • ME, modified ear-to-row selection • PC, principal component • PCR, polymerase chain reaction • PI, performance index • RAPD, random amplified polymorphic DNA • RRS, reciprocal recurrent selection


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
Materials and methods
 Results and discussion
 REFERENCES
 
THE HOPE BREEDING SYSTEM

for maize was created in 1977 with two purposes: (i) to serve as a base for developing useful inbred lines and (ii) to diversify the germplasm used in maize breeding programs. The system consists of two sets (A and B) of four populations arranged in a hierarchical manner, based on agronomic performance: low (L), intermediate (I), high (H), and elite (E) (Fig. 1) . By 1992, inclusive, more than 850 accessions had been introgressed into HOPE (Table 1) . All new germplasm enters the system at the L level and some also may be introgressed into the I and H levels. Also, more than 70 high-performing entries from a lower level have been advanced to the next higher level of the hierarchy (Table 1). Increasingly stringent recurrent selection procedures are employed at each higher level of the hierarchy with the intent of crafting the expected extensive diversity at the lower levels into the perhaps less diverse but presumably more useful breeding germplasm at the E level.



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 1 The hierarchical, open-ended population enrichment (HOPE) maize breeding system. RRS is reciprocal recurrent selection

 

View this table:
[in this window]
[in a new window]
  		Table 1 Total number of accessions introduced into the eight hierarchical, open-ended population enrichment (HOPE) maize populations until 1992, inclusive

 
The response to selection in the HOPE system was evaluated after 5 yr of selection (Cramer, 1986; Cramer and Kannenberg, 1992), but variability in the system was not assessed. While numerous studies have evaluated molecular diversity of maize hybrids (Smith et al., 1991, 1992; Smith and Smith, 1991) and inbreds (Melchinger et al., 1991; Livini et al., 1992; Messmer et al., 1992), little information is available on molecular diversity in closed maize populations. To our knowledge, the only published results are those of Labate et al. (1997), who evaluated molecular genetic variation in two maize populations that had undergone 12 cycles of reciprocal recurrent selection. The objective of our study was to estimate, using molecular markers, the diversity in the HOPE populations at the initiation of the system and following 15 yr of operation. In addition, the study included several commercial hybrids to allow the variability in the HOPE program to be compared with the variability available in conventional germplasm.


   Materials and methods
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and methods
Results and discussion
 REFERENCES
 
Synthesis of the HOPE System
The HOPE system has two population sets (A and B), with four levels each (Fig. 1). For the initial populations, 43 components (open-pollinated varieties, composites, synthetics, and double-cross hybrids) were assigned, six to eight components to a population, to the H, I, and L levels of each set. The assignment of components was based on available data regarding the heterotic pattern and agronomic performance of the entries. For each of the initial gene pools, the components were crossed in a diallel fashion, and subsequently all unique three-way crosses were made. Finally, the progenies of these crosses were intercrossed for one generation without selection. Recurrent selection was initiated in these six populations in 1977. Each of the two E populations (i.e., EA and EB) were formed in 1980 by intercrossing in a diallel pattern four and three S2 lines that traced back to three half-sib families selected from each of the original HA and HB populations, respectively.

Breeding Procedures in the HOPE System
Detailed information about the original version of the HOPE system is given by Cramer (1986) and Cramer and Kannenberg (1992), while the revised version is described by Kannenberg and Falk (1995). A brief description of both versions follows.

At the L level, stratified mass selection (Gardner, 1961) is employed, and with this procedure a cycle is completed every year. Each population is grown in 0.5-ha isolation plots and open-pollinated. The fields are divided at harvest into 200 grids, and the best five plants are harvested in each grid. Based on individual ear performance index (PI), which is the ratio of grain yield to grain moisture, the best ear from each grid is selected. Seed for the following cycle of selection is obtained by bulking equal quantities from each of the selected ears and the ears of new introductions. The best selections of the two L levels also are included in the corresponding I level evaluation trial.

The modified ear-to-row (ME) procedure (Lonnquist, 1964) was used initially at both the I and H levels. This procedure also required 1 yr to complete a cycle. Progeny trials included 144 entries, grown at two locations with two replications each. In the isolation block, the detasseled female rows were open-pollinated by a bulk of males representing all entries. Based on progeny trials, the best female entries were identified. At the I level, four to five ears were selected in each of the best 29 families, while at the H level, six to eight ears were selected from each of the best 18 families. The selected ears were planted ear-to-row in the next cycle of selection.

Based on the study of Cramer (1986), the ME procedure at both the I and H level was substituted after nine cycles of selection with other recurrent selection procedures. At the I level, the half-sib procedure, as suggested by Compton and Comstock (1976), was applied. This method requires 1 yr per cycle with the use of a winter nursery. For each cycle, 196 half-sib families are evaluated at two locations. Of these 196 families, 150 are from the previous selection cycle of the I level, and 46 are the best entries from the L level and any introductions being evaluated for introgression into the I level. The best 30 families are selected and residual seed of these is sown ear-to-row in the winter nursery for intercrossing, using bulked pollen from two to four selected males from each of 15 rows on the remaining 15 rows of females, and vice versa. Five ears are selected from each of the 30 entries to obtain 150 half-sib families for the following cycle of selection.

The procedure used in the revised version of the system at the H level is S2 progeny recurrent selection, which requires 3 yr per cycle of selection. S1 seed is produced in a winter nursery by selfing selected plants from the previous cycle of selection. The S1 progenies are planted in the summer nursery, and the best plants within visually selected progenies are selfed to produce S2 seed. Further visual selection is performed at harvest. In the following summer, 100 S2 families (including mostly material from the H level, but also S2s of the best entries from the I level and of any potential introgressions) are evaluated at two locations in two replication trials. Residual S2 seed of the best 20 entries is planted ear-to-row in a winter nursery for the first recombination, using the paired block (10 rows) approach described for the half-sib method; the cycle is completed with a second recombination in the following summer.

At the E level, reciprocal recurrent selection (Comstock et al., 1949) is employed. This procedure is similar to the S2 progeny method used at the H level, with the exception that the S2 lines from one set (e.g., EA) are topcrossed to a tester consisting of a composite of S4 lines from the previous cycle of the complementary population (EB). In the second summer, 100 topcrosses (mostly E level material, but also some topcrosses of the best H level entries) are evaluated at two locations in two replication trials. In the initial cycle of selection at the E level, S1 lines instead of S2 lines were used to make topcrosses, so that cycle was completed in 2 yr, while all subsequent selection cycles have required 3 yr.

The system is open to new introductions, which are first screened for per se performance to determine the appropriate level in the hierarchy, and for performance of the topcrosses to the two E populations to determine the appropriate set. All accessions are introduced into an L population in the form of a cross with the E level of the same set. This is to prevent "swamping" of the L populations due to the large number of accessions that may be added in any given year and to enforce the A x B heterotic pattern throughout the system. Introductions with high per se performance are also included in the evaluation trials of the I and H levels. No new accessions are introgressed directly into the E-level populations. Until 1992, inclusive, 856 accessions entered the HOPE system, while 72 entries were advanced through the hierarchy (Table 1).

Selection up to 1988 was based on PI and on resistance to lodging. Performance index was used as a selection criterion because it incorporates grain yield and grain moisture into one selection parameter. Selection for grain yield without consideration of moisture would have resulted in an undesired shift toward later maturity. Since 1989, at the E, H, and I levels, a rank selection system based on whole-plot grain yield and grain yield of standing plants has been used. Grain moisture at harvest is monitored and entries with high moisture are culled to avoid shifting of the populations to later maturities. At the L level, the selection procedure was not changed.

Plant Material
Random amplified polymorphic DNA (RAPD) analyses were performed on individual plants from randomly intermating populations of the initial (C0) and last cycle (Cn) of each level of the two population sets. Preliminary research was done to determine the number of plants to be sampled from each population. Based on this information (data not shown), nineteen plants were analyzed from each of the 16 HOPE populations. A group of 11 single cross hybrids (Table 2) was also included in the analyses, with one plant being analyzed for each of the hybrids.


View this table:
[in this window]
[in a new window]
  		Table 2 Hybrids used in random amplified polymorphic DNA (RAPD) analyses

 
DNA Extraction
The DNA extraction method was a modification of the procedure described by Edwards et al. (1991). Extraction buffer was prepared by mixing 10 µL of Tris-HCl pH 7.4, 2.5 mL of 5 M NaCl, 2.5 µL of 0.5 M EDTA pH 8.0, 2.5 µL of 10% sodium dodecyl sulphate (SDS), and 32.5 µL of double-distilled H2O. Shoots of 4-d-old seedlings were removed and kept at -25°C prior to DNA extraction. Frozen tissue of individual plants was homogenized with an electric grinder ({approx}500 rpm) in a sterilized Eppendorf tube in 300 µL of extraction buffer. The homogenate was vortexed and then centrifuged for 5 min in a microfuge. After adding 300 µL of 1:1 chloroform–phenol to the supernatant, the mixture was vortexed and centrifuged for 5 min. Chloroform (300 µL) was added to the aqueous phase and the mixture was vortexed and centrifuged for 5 min. Twenty percent polyethylene glycol containing 6 M NaCl (300 mL) was mixed with the aqueous phase, and the sample was placed at -20°C for 1 hr. The sample was centrifuged and the DNA pellet was air dried and dissolved in 100 µL of extraction buffer at 4°C overnight. The sample was centrifuged for 1 min and the supernatant was used for RAPD analysis.

DNA Amplification and Separation
The DNA content in each sample was determined by measuring the fluorescence of a sample stained with Hoechst using a Shimatzu spectrofluorometer RF-5000 (Shimatzu Scientific, Columbia, MD). The samples were diluted with extraction buffer to a concentration of 8 ng µL-1 DNA. The polymerase chain reaction (PCR) mixture contained 7.5 µL of sterile double-distilled H2O, 5.0 µL of 0.5 mM dNTP (Promega, Madison, WI), 2.5 µL 10x reaction buffer (GIBCO BRL, Grand Island, NY), 1.5 µL 25 mM MgCl2 (GIBCO BRL), 4.0 µL 10 mg mL-1 primer, 1.5 µL 1 unit mL-1 Taq polymerase (GIBCO BRL), and a 3-µL sample of DNA (24 ng). The PCR reactions were performed in a MJ Research Programmable Thermal Controller (PTC-100, MJ Research, Waltham, MA), with 45 cycles of 1 min at 94°C (denaturation), 1 min at 36°C (annealing), and 2 min at 72°C (elongation). DNA fragments were separated by agarose-gel electrophoresis (1.4% gel, w/v), visualized with ethidium bromide and photographed with Polopan 667 film (Polaroid, Cambridge, MA).

The primers for the PCR reaction were purchased from the University of British Columbia Biotechnology Laboratory. A preliminary investigation was done to determine the number of primers to be employed. Based on the data from the preliminary experiments (not shown), the PCR analyses were performed using 12 primers (Table 3) .


View this table:
[in this window]
[in a new window]
  		Table 3 Primers used in the polymerase chain reaction (PCR), and their sequences

 
Diversity Computations
Ninety-one RAPD bands were scored. The presence or absence of a band was denoted with 1 or 0, respectively. The MEGA software (Kumar et al., 1993) was used to compute the p-distance between all pairs of plants in a population. This parameter is the complement to the simple matching coefficient described by Gower (1985), and it is represented by the formula:

The mean of all pairwise distances in a population (Dp) was considered to be an estimate of the intrapopulation diversity.

To estimate the effect of introductions on intrapopulation diversity, all bands that were polymorphic only in the last cycle of selection were identified and reassigned the value from C0, that is, all values of 0 or 1. Dp was computed with the modified data to estimate the diversity that would have been observed in populations if there had been no introductions.

Diversity among the populations was determined by PC analysis, performed by SAS (SAS Institute, 1982). For every band, populations were scored with 1 if the band exhibited polymorphism, or 0 if the band was monomorphic across the 19 analyzed plants. The PC analysis, with the use of the correlation matrix, estimated interpopulation diversity by comparing the scores given to the populations for each of the 91 evaluated bands.


   Results and discussion
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
REFERENCES
 
Intrapopulation Diversity
The highest percentages of polymorphic bands in the A set were recorded for the L and I levels (Fig. 2) , and these remained stable at >61% to the last cycle of selection. For the upper two levels, the percentage of polymorphism dropped during selection, especially for EA, which dropped under 40% by the last cycle of selection. A comparison of the Dp values for the different levels in the A set indicated greater diversity in the lower levels than at the E level (Fig. 3) . At the lower three levels, especially in LA, Dp increased during selection, while a decrease was observed for EA.



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 2 Band polymorphism in the populations of the A and B sets of the hierarchical, open-ended population enrichment (HOPE) maize breeding system

 


View larger version (48K):
[in this window]
[in a new window]
  		Fig. 3 Diversity within the populations of the A and B sets of the hierarchical, open-ended population enrichment (HOPE) maize breeding system. Distance between every two plants in a population was determined by the complement to the simple matching coefficient (the number of bands for which the plants did not get the same score divided by the total number of bands); population diversity is the mean of all pairwise distances. Standard error is given for each diversity estimate

 
For the B set, polymorphisms increased slightly at all levels during selection; the largest increase was observed at the LB level (62.6–65.9%) (Fig. 2). Also, each successively higher level in the hierarchy had less polymorphism. The largest Dp occurred in LB, and the lowest in EB (Fig. 3). The diversity in the populations of the B set, with the exception of HB, remained stable during selection.

The PCR analysis confirmed the hierarchical structure of the HOPE system. The highest values for band polymorphism and Dp were detected at the L level in all but one case (band polymorphism in C0 of the A set), while the E level always had the lowest values (Fig. 2 and 3). With the exception of EA, HA, and IA, more polymorphism was detected in Cn than in C0 (Fig. 2). In contrast, in two closed populations subjected to reciprocal recurrent selection, Labate et al. (1997) detected a decrease of polymorphism from 99 to 75% and a 50% loss in gene diversity after 12 cycles of selection. However, one has to keep in mind that Labate et al. (1997) observed individual loci, while we looked at RAPD bands, which may represent more than one locus, thus having a greater chance of being polymorphic.

At all levels, polymorphisms of some bands disappeared during selection, while for other bands, polymorphisms were gained (Table 4) . In the A set, each level gained polymorphisms for four to five bands. The number of bands that became monomorphic during selection increased for each consecutively higher level of the hierarchy, ranging from four for LA to 16 for EA. For the B populations, the number of bands that became polymorphic varied from four for HB to eight for LB. In HB, three polymorphic bands became monomorphic during selection, while for each of the other levels, five bands became monomorphic.


View this table:
[in this window]
[in a new window]
  		Table 4 Numbers of bands with changes in polymorphisms during selection in the HOPE maize populations.{dagger}{ddagger}

 
Bands that were polymorphic only in the last cycle of selection were considered to be the result of direct introductions (for L, I, and H) or advancements in the hierarchy (for I, H, and E). Changes in polymorphism might also occur because of sampling errors; a band could be present, but missed in C0. However, for each gene pool, the number of bands that were polymorphic only in Cn varied from four to eight, and the probability that all these bands were present, but not detected in C0, is very small. For example, if the chance to miss a band is 0.1, the probability of missing four bands with the same frequency is 10-4. Thus, the polymorphic changes probably reflect the open-ended aspect of the HOPE system.

The impact of introductions on the diversity of the populations was determined by comparing the observed diversity values in Cn to those computed after reassigning the value from C0 to all bands that were polymorphic only in Cn. The increase in Dp due to new polymorphisms on the A side of the system varied from 0.01 for IA and HA to 0.07 for LA (Fig. 3). On the B side, the increase in Dp was least for HB and EB (0.01) and greatest for LB (0.03) (Fig. 3).

Interpopulation Diversity
A multivariate analysis revealed that the first principal component (PC1) accounted for 21.3% of the variability existing among the 16 HOPE populations, while PC2 and PC3 accounted for 12.2 and 10.6%, respectively. A three-dimensional graph of the first three PCs indicated a cluster containing the lower two levels of both population sets and the H level of the A set (Fig. 4) . The remaining six populations, belonging to the E level of both sets and to the H level of the B set, were distributed across a larger area. The fact that the L populations, especially in C0, were more similar to one another than the populations of the I, H, and E levels (Table 5) agrees with the data of Kermali (1994), who evaluated morphometric diversity within and among synthetics; he observed a decrease of interpopulation diversity for synthetics with a high intrapopulation diversity.



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 4 Principal component analysis of the hierarchical, open-ended population enrichment (HOPE) maize breeding system populations representing the initial (C0) and last (Cn) cycle of selection for the elite (E), high (H), intermediate (I), and low (L) levels of the A and B population sets

 

View this table:
[in this window]
[in a new window]
  		Table 5 Distance between the corresponding levels of the A and B sets in the initial (C0) and last (Cn) cycle of selection of the HOPE maize populations.{dagger} Distances were computed based on the first three principal components

 
The distances between the corresponding populations of the two sets, based on the first three PC (Table 5), suggest that divergence occurred between EA and EB during selection, as well as between LA and LB. At the E level this occurred possibly because of the use of reciprocal recurrent selection. The increase in distance at the L level could, similarly, be attributed to the fact that the large number of accessions that were added to each of the two L pools were previously crossed to the E population of the same population set. The populations of the H and I levels did not diverge during selection perhaps because the populations belonging to the two sets were improved independently from each other.

HOPE vs. Commercial Material
The comparison of the HOPE material with hybrids indicated that the populations possess variability not present in commercial material. All of the bands that were polymorphic in the hybrids were also polymorphic in at least one of the Cn HOPE populations (Table 6) . On the other hand, 32 of the 91 bands that were polymorphic in the HOPE system were not polymorphic in the hybrids. The comparison of EA and EB, the least diverse populations of HOPE, and the hybrids showed that 17 bands were polymorphic in Cn of the E level and monomorphic in the hybrids; five bands were polymorphic in the hybrids but not in the E level. Therefore, the HOPE germplasm has the potential to contribute to the diversity of commercial germplasm


View this table:
[in this window]
[in a new window]
  		Table 6 Comparison of random amplified polymorphic DNA (RAPD) band polymorphisms{dagger} between the last cycle of selection of the hierarchical, open-ended population enrichment (HOPE) maize breeding system and 11 hybrids. A total of 91 bands was evaluated

 

   ACKNOWLEDGMENTS
 
We wish to thank the National Science and Engineering Research Council of Canada and the Ontario Ministry of Education for the funding provided for this study.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
 REFERENCES
 
Part of a thesis by the senior author in partial fulfillment of the requirements for the Ph.D. degree.

Received for publication January 22, 1999.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
 REFERENCES
 




This article has been cited by other articles:


Home page
 Crop Sci.Home page
E.A. Lee, R. Chakravarty, B. Good, M.J. Ash, and L.W. Kannenberg
Registration of 38 Maize (Zea mays L.) Breeding Populations Adapted to Short-Season Environments
Crop Sci., November 21, 2006; 46(6): 2728 - 2733.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Popi, J.
Right arrow Articles by Pauls, K.P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Popi, J.
Right arrow Articles by Pauls, K.P.
Agricola
Right arrow Articles by Popi, J.
Right arrow Articles by Pauls, K.P.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome