Am. J. Bot. Large Type Edition
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
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 (38)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Roa, A. C.
Right arrow Articles by Tohme, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Roa, A. C.
Right arrow Articles by Tohme, J.
Agricola
Right arrow Articles by Roa, A. C.
Right arrow Articles by Tohme, J.
(American Journal of Botany. 2000;87:1647-1655.)
© 2000 Botanical Society of America, Inc.

Cross-species amplification of cassava (Manihot esculenta) (Euphorbiaceae) microsatellites: allelic polymorphism and degree of relationship1

Ana C. Roa2,0, Paul Chavarriaga-Aguirre0, Myriam C. Duque0, María M. Maya3,0, Merideth W. Bonierbale4,0, Carlos Iglesias5,0 and Joe Tohme6,0

0 Cassava Program and Biotechnology Research Unit, Centro Internacional de Agricultura Tropical (CIAT), A.A. 6713, Cali, Colombia

Received for publication August 25, 2000. Accepted for publication February 10, 2000.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
Microsatellite amplification was performed on cassava (Manihot esculenta) and six other different species (all wild) of the Manihot genus. We used ten pairs of microsatellite primers previously developed from cassava, detecting 124 alleles in a sample of 121 accessions of the seven species. The number of alleles per locus ranged from four to 21 alleles, and allelic diversity was greater in the wild species than in cassava. Seventy-nine alleles, including unique ones, were detected in the wild species but were not found in the crop. The lower level of heterozygosity in some wild species probably resulted from a combination of fine-scale differentiation within the species and the presence of null alleles. Overall, microsatellite primers worked across the genus, but, with increasing genetic distance, success in amplifying loci tended to decrease. No accession of M. aesculifolia, M. carthaginensis, and M. brachyloba presented a banding pattern at locus Ga-140; neither did one appear for M. aesculifolia at locus Ga-13. Previous work with amplified fragment length polymorphism (AFLP) markers and this microsatellite analysis show that these three wild taxa are the most distant relatives of the crop, whereas the wild forms M. esculenta subsp. flabellifolia and M. esculenta subsp. peruviana appear to be the closest.

Key Words: cassava • cross-amplification • Euphorbiaceae genetic diversity • heterozygote deficiency • Manihot species • microsatellite loci


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
Cassava (Manihot esculenta Crantz) is a root crop produced by small farmers in tropical Africa, Asia, and Latin America. A staple food, it occupies fourth place after rice, sugar cane, and maize as a source of calories in the human diet (Cock, 1985 ; Best and Henry, 1994 ). It is also used by small-scale industries. Based on different kinds of evidence, the crop may have been first cultivated in one of two possible places: southern Mexico and Guatemala (Rogers, 1963 ; Renvoize, 1972 ) or the Amazon Basin in Brazil (Nassar, 1978 ; Allem, 1987, 1994 ).

Several wild species have been proposed as being the closest related taxa to the crop: M. aesculifolia, widely distributed in Central America (Rogers and Appan, 1973 ; Bertram, 1993 ); M. carthaginensis, which originated in the Caribbean Coastal regions of Colombia and Venezuela (Bertram, 1993 ); M. esculenta subsp. flabellifolia and M. esculenta subsp. peruviana (Allem, 1994 ); and M. tristis (Allem, 1987 ).

Molecular markers such as restriction fragment length polymorphisms (RFLPs) (Bertram, 1993 ; Fregene et al., 1994 ) and amplified fragment length polymorphisms (AFLPs) (Roa et al., 1997 ) have been used to search for cassava's closest relatives. Chloroplast and ribosomal DNA RFLPs analyses suggested two pairs of probable ancestors of cassava: M. aesculifolia and M. carthaginensis (Bertram, 1993 ), and M. tristis and M. esculenta subsp. flabellifolia (Fregene et al., 1994 ). Recent analysis carried out on seven taxa with AFLPs showed that M. esculenta subsp. flabellifolia and M. esculenta subsp. peruviana are the closest relatives to cassava (Roa et al., 1997 ).

Microsatellites or simple sequence repeats (SSRs) are tandem repetitive DNA sequences of two to five nucleotides long (Akagi et al., 1997 ). Accumulated evidence shows that they are widely dispersed in all eukaryotic genomes (Tautz, Trick, and Dover, 1986 ; Morgante and Olivieri, 1993 ; Dow, Ashley, and Howe, 1995 ; Byrne et al., 1996 ), for example, the dinucleotide repeats (AT/TA)n and (GA/CT)n are the most common found in higher plants (Morgante and Olivieri, 1993 ; Dow, Ashley, and Howe, 1995 ; Akagi et al., 1997 ; Steinkellner et al., 1997 ). To amplify microsatellites, specific primers are designed on the flanking regions of the SSRs. These markers are abundant, codominant, highly polymorphic, even within populations, spread throughout the genome, easily amplified by polymerase chain reaction (PCR), and the great majority are probably selectively neutral (Ashley and Dow, 1994 ). They offer great potential for studies on parentage (Dow and Ashley, 1996 ), gene flow within and between populations (Innan, Terauchi, and Miyashita, 1997 ), mapping and breeding (McCouch et al., 1997 ), and extent and maintenance of genetic diversity (Byrne et al., 1996 ; Witsenboer, Vogel, and Michelmore, 1997 ).

The widespread use of microsatellites has been limited by the fact that PCR primers require a high degree of homology to work, implying that novel species-specific markers would have to be isolated when starting the analysis of a new species (Steinkellner et al., 1997 ). Success in the cross-species amplification of any DNA sequence is inversely related to the evolutionary distance between the two species (Steinkellner et al., 1997 ). Hence, research on species relationships has increasingly focused on assessing the ability of SSR primers to amplify the same loci across different species and genera (Byrne et al., 1996 ; Katzir et al., 1996 ; Isagi and Suhandono, 1997 ; Smulders et al., 1997 ; Steinkellner et al., 1997 ; Witsenboer et al., 1997 ).

For cassava, 14 different primer sequences were designed to amplify SSRs containing mostly perfect or imperfect GA repeats. The primers were tested on 522 accessions of the cultivated cassava core collection conserved at CIAT and showed heterozygosity values between 0.00 and 0.88, with as many as 15 different alleles at one locus (Chavarriaga-Aguirre et al., 1998 ).

To evaluate the ability of these primers to amplify SSR loci in congeneric species, a set of six wild Manihot species and a diverse sample of cassava were assessed with ten pairs of primers. The sequence conservation of the flanking primer regions is discussed, together with the utility of using a codominant marker, such as microsatellites, to assess the genetic diversity and degree of relationship between cassava and its wild relatives.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 LITERATURE CITED
 
Plant material
A sample of 121 accessions was selected from seven species. For cassava (referred as ESC in tables and figures), 38 accessions were selected from the cassava core collection held in trust at CIAT. The core collection was previously set according to geographic, morphological, and biochemical parameters as described by Hershey et al. (1994) . The remaining 83 accessions were chosen, based on previous analysis with AFLPs (Roa et al., 1997 ), to represent the wild taxa M. aesculifolia (AES), M. brachyloba (BLO), M. carthaginensis (CTH), M. esculenta subsp. flabellifolia (ESC-FLA), M. esculenta subsp. peruviana (ESC-PER), and M. tristis (TST) (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. Germplasm of Manihot species selected for microsatellite analysis

 
DNA extraction
DNA samples were prepared by first drying young leaves at 48°C for 20 h, then grinding them, using a mortar and pestle, until a fine powder was obtained. The DNA was then extracted in eppendorf tubes, according to the following protocol (C. Colombo, 1995, personal communication, Montpellier, France): 0.1 g of ground tissue was resuspended in 1.5 mL of hexadecyltrimethyl-ammonium bromide (CTAB) buffer (0.1 mol/L Tris-HCl pH 8.0; 1.25 mol/L NaCl; 0.02 mol/L ethylenediaminetetra acetic acid (EDTA); 4% CTAB, and 0.1% 2-mercaptoethanol) and incubated at 65°C for 1 h 30 min. After two chloroform-isoamylic extractions, the aqueous phase was collected and the nucleic acids precipitated with cold isopropanol. The pellet collected by centrifugation was washed with 75% ethanol and dried at room temperature. The nucleic acids were dissolved in T10E1 (10 mmol/L Tris-HCl (pH 8.0), 1 mmol/L EDTA) and treated with RNAse A (10 µg/mL). Electrophoresis and fluorometry determined DNA quality and concentration.

Primers and PCR conditions
DNA from each sample was amplified, using phosphoramidite-labeled primers (Perkin Elmer/Applied Biosystems, Foster City, California, USA, or Research Genetics, Huntsville, Alabama, USA). Multiplex PCR reactions were set according to the method developed by Chavarriaga-Aguirre et al. (1998) . Primer pairs were combined as follows: one quadruple with primers for the loci Ga-21, Ga-126, Ga-134, and Ga-136; and three duplex for the loci Ga-16/Ga-140; Ga-13/Gagg-5, and Ga-12/Ga-131. The optimal annealing temperature was the same for all primer sets. PCR reactions and amplification profiles were carried out according to Chavarriaga-Aguirre et al. (1998) , except that no polymerase activation cycle at 94°C for 10 min was carried out before amplification. In this study we used AmpliTaq polymerase (Perkin Elmer/Applied Biosystems), while Chavarriaga et al. (1998) used AmpliTaq-Gold, which required an activation cycle as mentioned above.

Electrophoresis conditions and allele sizing
A volume of 1.5 µL of the PCR product, combined with 0.5 µL of the internal size standard (GeneScan TAMRA 500, Perkin Elmer/Applied Biosystems) and 2 µL of deionized formamide, mixed with loading buffer (5:1), was denatured at 98°C for 2 min. This mixture was loaded on 4% polyacrylamide denaturing gels, containing 6 mol/L urea. Electrophoresis was carried out with 1x Tris-borate-EDTA (TBA) buffer on an ABI Prism 377 automatic DNA sequencer (Perkin Elmer/Applied Biosystems) using the Gene Scan Analysis Software version 2.0.2. The GeneScan module used to run the gels was GS 36C–2400 [36 cm well-to-read (WTR), virtual filter C, 3000 V/2400 scans/h in 2.5 h].

Allele sizing was performed, using the third-order least-square algorithm, which relies on regression analysis to build a best-fit size-calling curve. It compensates for any fragments that may run anomalously (Perkin-Elmer, 1995 ). Allele sizes were determined for all 121 individuals and compared, using Genotyper Software version 1.1 (Perkin-Elmer, 1995 ).

Data analysis
Presence and absence of alleles were registered for all 121 individuals. Only the strongest bands were considered alleles because lighter bands may have been stutter bands that resulted from slippage of the Taq polymerase during PCR (Lagercrantz, Ellengran, and Anderson, 1993 ; Wu and Tanksley, 1993 ). The number of alleles and percentage of polymorphism were calculated for each locus and species.

Levels of differentiation among taxa and degrees of nonrandom association of alleles within species, allele frequencies, heterozygosity, and Fis values were estimated for each locus–species combination, using the PopGene software (version 1.2; Yeh, Rongcai, and Boyle, 1997 ). Fis, called the fixation index or inbreeding coefficient, is the correlation between the number of homologous alleles within individuals, with reference to the local population. It is also a measure of heterozygote deficiency or excess (Avise, 1994 ). All individuals were assumed to be diploid. Therefore, when only one allele was observed at a locus, the individual was considered to be a homozygote.

At microsatellite loci, species heterozygote deficiencies were observed and may be explained by the presence of null alleles (Brookfield, 1996 ). To estimate how the frequency of null alleles (rb) affects the observed heterozygosity (Hobs), rb was calculated according to Brookfield (1996) ,

where Hexp is the expected heterozygosity.

A similarity matrix, calculated by using the Nei–Li index (Nei and Li, 1979 ), was used to construct a dendrogram to show the degree of genetic similarity within and among taxa. The dendrogram was constructed by employing the option TREE and the unweighted pair group method (UPGMA) with the NTSYS program (version 1.8; Rohlf, 1994 ). The dendrograms and similarity indexes obtained with AFLPs (Roa et al., 1997 ) were compared with those generated by the SSRs. Cophenetic values for each dendrogram and the assembly of a cophenetic matrix for each marker type were calculated, using the MXCOMP option of the NTSYS program. The Mantel matrix test was used to compare cophenetic matrices (Sokal and Rohlf, 1995 ).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 LITERATURE CITED
 
Cross-species amplification
Crucial to the use of SSRs for assessing the genetic variation within and among species is the ability of primers from one species to amplify homologous loci in related species (Kijas, Fowler, and Thomas, 1995 ). At eight loci, primers amplified the target sequence across all species tested (Table 2), but in the wild species M. aesculifolia, M. brachyloba, and M. carthaginensis, no products were amplified at locus Ga-140. Similarly, with the locus Ga-13, no PCR products were observed for the first species (Table 2). At four of the eight loci, amplification products were absent in some accessions.


View this table:
[in this window]
[in a new window]
  		Table 2. Cross-species amplification of ten microsatellite loci in seven Manihot taxa. "+" denotes successful amplification, "-" denotes no amplification

 
Number of alleles and polymorphism in Manihot
A total of 124 alleles were found in 121 individuals, using ten microsatellite loci. The number of alleles detected in all species varied greatly among loci. The highest number, 21 alleles, was found at locus Ga-131, and was followed by locus Ga-126 with 20 alleles. Five SSR loci showed between ten and 20 alleles, and fewer than ten alleles were detected in three loci, four alleles being the lowest number found (for locus Ga-13) (Fig. 1).



View larger version (36K):
[in this window]
[in a new window]
  		Fig. 1. Number and distribution of alleles at microsatellite loci (Gagg-5 to Ga-140) in Manihot species. Black bars = alleles unique to cassava; white bars = alleles shared between cassava and wild taxa; hatched bars = alleles unique to wild taxa (gray: present in the two subspecies of M. esculenta; white: absent in the two subspecies of M. esculenta)

 
At all loci, a higher number of alleles were detected in wild species than in the cultivated cassava (Fig. 1). Considering the entire set of germplasm, 79 alleles were present in the wild species but absent in cassava. Forty-two alleles were shared between cassava and the wild conspecific taxa and just three SSR alleles were found only in the crop, suggesting a larger pool of SSR alleles for the wild species. Analyzing only cassava and the wild forms of M. esculenta, 47 additional alleles were present in the subspecific taxa. Thus, 32 unique alleles were found in species outside the M. esculenta complex (Fig. 1).

The percentage of polymorphic loci in the seven Manihot species ranged from 60 to 100% (Table 2), corresponding to a mean polymorphism of 83%. The taxa with the highest level of polymorphic loci (100%) were the cultivated species M. esculenta and the wild form M. esculenta subsp. peruviana, followed by M. esculenta subsp. flabellifolia with 90% polymorphism. Manihot carthaginensis and M aesculifolia exhibited 90 and 80% polymorphism, respectively, and 60% of the SSR loci were polymorphic in M. brachyloba and M. tristis (Table 2).

Allelic distribution and frequencies
Figure 2 shows the different allele sizes scored among the Manihot species. At most of the loci, allele sizes varied from 100 to 175, except for loci Ga-134 and Ga-126 where sizes were from >200 base pairs (bp) to 405 bp. The smallest difference between the highest and lowest values of allele size length was 6 bp at locus Ga-13, and the largest difference (225 bp) was detected at locus Ga-126. The allele sizes scored at the other remaining loci presented differences between 14 and 53 bp (Fig. 2). In most cases, the number of unique alleles (i.e., amplified products in just one individual or one species) was positively correlated with the total number of alleles per locus and their size differences. For instance, 13 was the maximum number of unique alleles detected at locus Ga-126. Manihot aesculifolia presented seven of the 13 unique alleles detected at this locus (Fig. 2). At locus Ga-131, with 21 alleles and 53 bp of difference between the shortest and largest alleles, a total of seven unique alleles were scored (Fig. 2).



View larger version (56K):
[in this window]
[in a new window]
  		Fig. 2. Allelic distribution and frequencies in each of the Manihot taxa at different microsatellite loci. Bar colors: gray = frequency 0 to <=0.33; cross-hatched = frequency >0.33 to <=0.66; black = frequency >0.66 to 1.00. AES = M. aesculifolia; BLO = M. brachyloba; CTH = M. carthaginensis; FLA = M. esculenta subsp. flabellifolia; PER = M. esculenta subsp. peruviana; TST = M. tristis; ESC = M. esculenta (i.e., cassava)

 
At most of the SSR loci, alleles were detected at frequencies <0.33, both within and among the seven Manihot taxa (Fig. 2). Nevertheless, at the loci Gagg-5 and Ga-13, some alleles showed frequencies >0.66 in more than two species (Fig. 2). Unique alleles were also observed, with few exceptions, at low frequencies. Only for M. brachyloba and M. carthaginensis were alleles detected at frequencies of 0.7–1 at four SSR loci and at locus Gagg-5, respectively (Fig. 2). For the locus Ga-140 where no amplification was obtained for M. aesculifolia, M. Brachyloba, and M. carthaginensis, the alleles were evenly distributed among the remaining species and the frequencies varied mainly from 0 to 0.33. At the intraspecific level, in species such as M. esculenta subsp. peruviana, a larger percentage (60%) of the alleles was present at very low frequency (<0.1) (Fig. 2).

Level of heterozygosity in Manihot
Outcrossing plants with dioecious floral morphology are expected to have high genetic heterozygosity within populations (Avise, 1994 ). The Manihot species evaluated fit into this group of plants, but, in contrast to the theory, in 42 of 70 (60%) locus-species comparisons, the observed number of heterozygotes (Hobs) was less than expected (Hexp) (Table 3). Pooling across all ten loci, a heterozygote deficit was found in five of the seven taxa, all being wild species. The only species with the same overall observed (mean Hobs) and expected heterozygosity (mean Hexp) figures was M. esculenta (Table 3).


View this table:
[in this window]
[in a new window]
  		Table 3. Heterozygosity values (Hexp and Hobs), frequency of null alleles ([cf2]r[cf1][cf15]b[cf1]), and fixation index (Fis) at microsatellite loci of different Manihot species

 
Forty-six of 70 (65.7%) Fis values were greater than zero (Table 3), indicating that nonrandom association of alleles can occur in the species tested (Gibbs et al., 1997 ). According to Gibbs et al., (1997) , Fis values cannot be calculated (referred as NA in Table 3) when only one allele is present in the species or a second allele is present in only one individual.

The rb estimator, which estimates the frequency of null alleles in each species at each locus, varied from -0.14 in M. esculenta (locus Gagg-5) up to 0.33 in M. esculenta subsp. peruviana (locus Ga-134). When overall rb was calculated for all loci in each species, the lowest (0.0) and highest (0.20) values were also obtained at M. esculenta and M. esculenta subsp. peruviana, respectively (Table 3).

Cluster analysis
Figure 3 depicts the clustering of Manihot accessions into groups that corresponded, in most cases, with their taxonomic classification. With these ten SSR loci, the differentiation of every genotype was not possible in 12 cases, corresponding to accessions of the same populations in M. brachyloba, M. carthaginensis, M. esculenta subsp. peruviana, and M. tristis. Notwithstanding, the species M. aesculifolia, M. brachyloba, and M. carthaginensis formed discrete groups (clusters 7, 3, and 5, respectively) with <30% of similarity to the cultivated species (cluster 1). Most accessions from M. esculenta subsp. flabellifolia, M. esculenta subsp. peruviana, and M. tristis formed a mixed cluster (2), which was closer to cassava than the remaining species. Within this mixed cluster, the accessions of M. esculenta subsp. flabellifolia and some of M. esculenta subsp. peruviana were closest to cassava. As was shown by Roa et al. (1997) , using AFLP markers on these species, the most similar group to cassava is the mixed cluster formed by the taxa proposed by Allem (1994) as the wild forms of the crop, followed by M. tristis.



View larger version (35K):
[in this window]
[in a new window]
  		Fig. 3. Dendrogram of genetic similarity in Manihot species, drawn from a UPGMA analysis of Nei–Li estimates based on a microsatellite analysis. The nine clusters shown are discussed in the text. 1 = M. esculenta, that is, cassava; 2 = M. esculenta subspp. flabellifolia (EF) and peruviana (P), and M. tristis (TST); 3 = M. brachyloba (BLO); 4 = outlying population of M. esculenta subsp. flabellifolia; 5 = M. carthaginensis (CTH); 6 = outlying accessions of M. carthaginensis; 7 = M. aesculifolia (AES)

 
With respect to the intraspecific variability, microsatellites showed a high heterogeneity within all taxa studied, even in cassava. The range varied from <10% within some populations of species such M. brachyloba, 50% within the sample of cassava and M. aesculifolia, and even further in the case of M. carthaginensis (70% of variability). One population of M. esculenta subsp. flabellifolia (cluster 4) and two accessions classified as M. carthaginensis (cluster 6) appeared to be very distant from their conspecifics (Fig. 3).

Comparison between AFLPs and microsatellites
To compare the results obtained with AFLP and microsatellite markers in a common group of germplasm, the Mantel matrix correspondence test was used. Similarity matrices and UPGMA dendrograms with each marker class were generated with 96 selected Manihot materials (data not shown). All the cophenetic correlation coefficients between the similarity matrices and cophenetic matrices obtained with both marker types were high, 0.81 and 0.87, respectively, and statistically significant (probability of random Z < obs. Z: P = 1.00).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
LITERATURE CITED
 
The present work intended to determine the extent to which pairs of primers designed for the amplification of SSR loci in cassava can be used for assessing its wild relatives. The utility of the cassava primers to produce PCR -amplified products across the genus was demonstrated. Nevertheless, as it has been pointed out in other works on SSR loci conservation in plant species, a decline of amplification success was observed with increase of genetic distance (Isagi and Suhandono, 1997 ; Steinkellner et al., 1997 ; White and Powell, 1997 ; Witsenboer et al., 1997 ). In M. aesculifolia, M. carthaginensis, and M. brachyloba, the species most distant from cassava, no amplification products were detected. In species with low degrees of relationship, therefore, the same SSRs loci cannot be found. The result obtained with this codominant marker agrees with the findings reported for AFLP markers used with the same species (Roa et al., 1997 ): that M. aesculifolia and M. carthaginensis are the wild species most distantly related to cassava.

According to Smulders et al. (1997) , the lack of amplification of an allele in certain accessions can be the result of divergence in the sequences flanking the microsatellite, creating a null allele. The production of an undetectable amount of PCR product is another explanation given by Smulders et al. (1997) and Lavi et al. (1994) . Nevertheless, these studies were carried out, using a detection method that is less sensitive than fluorescence. In our survey, null alleles were confirmed after several repetitions of the assay, using the same conditions and also different annealing temperatures, to ensure that no reaction failure existed. Nonamplifying or null alleles have not been highlighted in literature on plant molecular ecology, but, from human studies, two facts emerge. First, these alleles are common ; and, second, where flanking sequences were obtained for nonamplifying alleles, a mutation was found to have occurred in one of the priming sites (Pemberton et al., 1995 ). The exclusion of even a few nonamplifying homozygotes can have dramatic effects on the interpretation of genotype frequency distributions and could lead to mistaken interpretations about the level of inbreeding in a population (Pemberton et al., 1995 ).

Microsatellite variation was clearly detected in the set of germplasm studied. The number of alleles varied greatly among the SSR loci evaluated, but, in all, more alleles were found in the wild species than in cassava. Almost 64% of the scored alleles were present in the wild taxa while absent from cassava. Considering only cassava and the wild subspecies, 47 additional alleles were found in the second group, suggesting a larger pool of alleles for the M. esculenta subspecies than for cassava. This result shows that the primary gene pool of cassava, composed of the wild forms and a few other species (Allem, 1994 ), contains a valuable source of diversity that could be useful for improving cassava.

The high level of polymorphism (83%) obtained with this type of marker makes it a powerful tool for assessing genetic diversity in crop plants and its relatives. Ashley and Dow (1994) proposed that the high polymorphism found at microsatellite loci is related to the mechanism of mutation and the high rate at which this occurs. Nevertheless, the polymorphism rate reported here was lower than the AFLP polymorphism (98.6%) obtained by Roa et al. (1997) in the same species. This could be explained by the lack of amplification products in some Manihot species at certain SSR loci.

Currently, different models try to explain the mutational dynamics of the microsatellite loci. According to Slatkin's model (1995) , they mutate in an unconstrained fashion, meaning that no limit exists on the number of possible allele sizes. However, evidence suggests that biases may exist in mutation direction and therefore a limit does exist to maximum allele size (Lehmann, Hawley, and Collins, 1996 ; Goodman, 1997 ). Considering the length differences that were found for most of the SSR loci evaluated, the amplification patterns obtained may correspond to one locus with several alleles. Nevertheless, the length difference between the shortest and longest allele for locus Ga-126 (>100 steps) may indicate the presence of other loci for these dinucleotide repeats in M. aesculifolia.

The substantial number of unique alleles (and null alleles) present in species such as M. aesculifolia, M. carthaginensis, and M. brachyloba suggests that their separation from cassava on a evolutionary scale took place long ago, allowing mutation to generate new alleles. Similar results have been reported in animal studies (Gibbs et al., 1997 ). The wild forms of M. esculenta (subspp. flabellifolia and peruviana) also presented a larger number of unique alleles than did the cultivated species, indicating a larger pool of SSR alleles in the wild relatives. However, substantial conclusion on Manihot evolution or cassava wild relatives awaits a full generic phylogenetic revision including both molecular and morphological data. This study, evaluating the ability of selected primers to amplify SSR loci at the generic level, is preliminary to such a definitive study.

The levels of heterozygosity found in Manihot were, in most cases, lower than expected. Species heterozygote deficiency can be explained as a result of different factors: (a) unrecognized genetic structure within populations, (b) inbreeding due to consanguineous mating, or (c) presence of null alleles, such that many apparent homozygotes are, in reality, heterozygotes between a visible and a null allele (Pemberton et al., 1995 ; Brookfield, 1996 ). Direct evidence for null alleles and genetic differentiation between populations was collected (rb and Fis values). Assuming that the heterozygote deficit is totally due to null alleles, the rb values suggest that such alleles are present, on the average, at relatively high frequencies (~11%) in Manihot species.

The other component of heterozygote deficiency implies a genetic differentiation, that is, in allele sizes and frequencies, within and between populations of a species. Similar results were obtained in the AFLP study (Roa et al., 1997 ). This fine-scale differentiation within the species could be reflected in a substantial difference between the observed and expected heterozygosity. In the crop sample, 60% of the observed alleles in all loci were at frequencies between 0.1 and 0.99 (Fig. 2). This feature indicates a low differentiation among accessions and could explain why the difference between observed and expected heterozygosity was not significant. However, in species such as M. esculenta subsp. peruviana, a larger percentage (60%) of alleles was present at low frequencies (<0.1) (Fig. 2), indicating a high differentiation among the accessions at the SSR loci. A combination of the presence of null alleles and fine-scale genetic differentiation may have been the cause of the heterozygote deficiency in Manihot species, but we could not quantify, using our data, the importance of each factor as a cause of the heterozygote deficit.

The cluster analysis of the seven Manihot species with SSR markers showed that the so-called wild forms of M. esculenta are the most closely related taxa to cassava among those studied here. The high correlation indexes obtained between the SSR and AFLP markers mean that a similar genetic structure of Manihot is depicted by both marker types. Dominant (AFLPs) (Roa et al. 1997 ) and codominant (SSR) markers demonstrated that cassava probably arose from the wild Brazilian forms M. esculenta subsp. flabellifolia and M. esculenta subsp. peruviana, as described by Allem (1994) .

The utility of a codominant marker such as microsatellites for accurately assessing the occurrence of heterozygote genotypes in species with an outbreeding mode of reproduction such as cassava is clear. High levels of intraspecific variation were detected in most of the Manihot species surveyed, making SSR markers suitable for studies of population differentiation. This factor is essential for establishing the criteria for conservation of populations of a species under in situ and ex situ conditions.


   FOOTNOTES
 
1 The authors thank Dr. Morry Levy, Purdue University, for his statistical advice on microsatellite analysis, Dr. Gary Kochert, University of Georgia, for generously providing the cassava microsatellite primers, and Elizabeth de Paéz for editorial assistance. This research was supported in part by a grant from the Instituto Colombiano para el Desarrollo de la Ciencia y la Tecnología "Francisco José de Caldas" (COLCIENCIAS). Back

2 Current address: Center for the Application of Molecular Biology to International Agriculture-CAMBIA GPO Box 3200, Canberra, ACT 2601, Australia. Back

3 Current address: Purdue University, Department of Biological Sciences, Lilly Hall, Room G-420A, West Lafayette, Indiana 47907-1392 USA. Back

4 Current address: Centro Internacional de la Papa-CIP, Apartado 1558, Lima 12, Peru. Back

5 Current address: Weaver Popcorn NC, 1000 North 325 W, PO Box 207, New Richmond, Indiana 47967 USA. Back

6 Auther for reprint requests (Tel.: ++57-2-4450000 ext. 3265; fax: ++57-2-4450073; e-mail: j.tohme{at}cgiar.org ). Back


   LITERATURE CITED
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
Akagi, H., Y. Yokozeki, A. Inagaki, and T. Fujimura. 1997 Highly polymorphic microsatellites of rice consist of AT repeats, and a classification of closely related cultivars with these microsatellite loci. Theoretical and Applied Genetics 94: 61–67.[CrossRef][ISI]

Allem, A. C. 1987 Manihot esculenta is a native of the neotropics. Plant Genetic Resources Newsletter (International Board of Plant Genetic Resources) 71: 22–24.

———. 1994 The origin of Manihot esculenta Crantz (Euphorbiaceae). Genetic Resources and Crop Evolution 41: 133–150.[CrossRef]

Ashley, M. V., and B. D. Dow. 1994 The use of microsatellite analysis in population biology: background, methods and potential applications. In B. Schierwater, B. Streit, G. P. Wagner, and R. Desalle [eds.], Molecular Ecology and Evolution: Approaches and Applications, 185–201. Birkhäuser Verlag, Basel, Switzerland.

Avise, J. C. 1994 Molecular markers, natural history and evolution. Chapman and Hall, New York, New York, USA.

Bertram, R. B. 1993 Application of molecular techniques to genetic resources of cassava (Manihot esculenta Crantz, Euphorbiaceae): interspecific evolutionary relationships and intraspecific characterisation. Ph.D. dissertation, University of Maryland, College Park, Maryland, USA.

Best, R., and G. Henry. 1994 Cassava: towards the year 2000. In International network for cassava genetic resources, 3–11. Report of the First Meeting of the International Network for Cassava Genetics Resources, CIAT, Cali, Colombia, 18–23 August. International Crop Network Series number 10. International Plant Genetic Resources Institute (IPGRI), Rome, Italy.

Brookfield, J. F. Y. 1996 A simple new method for estimating null frequency from heterozygote deficiency. Molecular Ecology 5: 453–455.[CrossRef][Medline]

Byrne, M., M. I. Marquez-García, T. Uren, D. S. Smith, and G. F. Moran. 1996 Conservation and genetic diversity of microsatellite loci in the genus Eucalyptus. Australian Journal of Botany 44: 331–341.[CrossRef]

Chavarriaga-Aguirre, P., M. M. Maya, M. W. Bonierbale, S. Kresovich, M. Fregene, J. Tohme, and G. Kochert. 1998 Microsatellites in cassava (Manihot esculenta Crantz): discovery, inheritance, and variability. Theoretical and Applied Genetics 97: 493–501[CrossRef][ISI]

Cock, J. H. 1985 Cassava: new potential for a neglected crop. Westview Press, Boulder, Colorado, USA.

Dow, B. D., and M. V. Ashley. 1996 Microsatellite analysis of seed dispersal and parentage of saplings in bur oak, Quercus macrocarpa. Molecular Ecology 5: 615–627.

———, ———, and H. F. Howe. 1995 Characterization of highly variable (GA/CT)-n microsatellites in the bur oak, Quercus macrocarpa. Theoretical and Applied Genetics 91: 137–141.[ISI]

Fregene, M. A., J. Vargas, F. Angel, J. Tohme, R. A. Asiedu, M. O. Akorada, and W. M. Roca. 1994 Chloroplast DNA and nuclear ribosomal DNA variability in cassava (Manihot esculenta Crantz) and its wild relatives. Theoretical and Applied Genetics 89: 719–727.[ISI]

Gibbs, H. L., K. A. Prior, P. J. Weatherhead, and G. Johnson. 1997 Genetic structure of the threatened eastern massasauga rattlesnake, Sistrurus c. catenatus: evidence from microsatellite DNA markers. Molecular Ecology 6: 1123–1132.[CrossRef][Medline]

Goodman, S. J. 1997 RST Calc: a collection of computer programs for calculating estimates of genetic differentiation from microsatellite data and determining their significance. Molecular Ecology 6: 881–885.[CrossRef]

Hershey, C., C. Iglesias, M. Iwanaga, and J. Tohme. 1994 Definition of a core collection for cassava. In International network for cassava genetic resources, 145–156. Report of the First Meeting of the International Network for Cassava Genetic Resources, CIAT, Cali, Colombia, 18–23 August. International Crop Network Series number 10. International Plant Genetic Resources Institute (IPGRI), Rome, Italy.

Innan, H., R. Terauchi, and N. T. Miyashita. 1997 Microsatellite polymorphism in natural populations of the wild plant Arabidopsis thaliana. Genetics 146: 1441–1452.[Abstract]

Isagi, Y., and S. Suhandono. 1997 PCR primers amplifying microsatellite loci of Quercus myrsinifolia Blume and their conservation between oak species. Molecular Ecology 6: 897–899.[CrossRef][Medline]

Katzir, N., Y. Danin-Poleg, G. Tzuri, Z., Karchi, U., Lavi, AND P. B. Cregan. 1996 Length polymorphism and homologies of microsatellites in several Cucurbitaceae species. Theoretical and Applied Genetics 93: 1282–1290.[CrossRef][ISI]

Kijas, J. M. H., J. C. S. Fowler, and M. R. Thomas. 1995 An evaluation of sequenced tagged microsatellite site markers for genetic analysis within Citrus and related species. Genome 38: 349–355.[Medline]

Lagercrantz, U., H. Ellengran, and L. Anderson. 1993 The abundance of various polymorphic microsatellite motifs differs between plants and vertebrates. Nucleic Acids Research. 21: 1111–1115.[Abstract/Free Full Text]

Lavi, U., M. Akkaya, A. Bhagwat, E. Lahav, and P. B. Cregan. 1994 Methodology of generation and characteristics of simple sequence repeat DNA markers in avocado (Persea americana M.). Euphytica 80: 171–177.[CrossRef][ISI]

Lehmann, T., W. A. Hawley, and F. H. Collins. 1996 An evaluation of evolutionary constraints on microsatellite loci using null alleles. Genetics 144: 1155–1163.[Abstract]

McCouch, S. R., X. Chen, O. Panaud, S. Temnykh, Y. Xu, Y. G. Cho, N. Huang, T. Ishii, AND M. Blair. 1997 Microsatellite marker development, mapping and applications in rice genetics and breeding. Plant Molecular Biology 35: 89–99.[CrossRef][ISI][Medline]

Morgante, M., and A. M. Olivieri. 1993 PCR-amplified microsatellites as markers in plant genetics. Plant Journal 3: 175–182.[CrossRef][ISI][Medline]

Nassar, N. M. A. 1978 Conservation of the genetic resources of cassava (Manihot esculenta Crantz). Determination of wild species localities with emphasis on probable origin. Economic Botany 32: 311–320.[ISI]

Nei, M., and W.-H. Li. 1979 Mathematical model for studying genetic variation in terms of restriction endonucleases. Proceedings of the National Academy of Sciences, USA 76: 5269–5273.[Abstract/Free Full Text]

Pemberton, J. M., J. Slate, D. R. Bancroft, and J. A. Barrett. 1995 Nonamplyfying alleles at microsatellite loci: a caution for parentage and population studies. Molecular Ecology 4: 249–252.[Medline]

Perkin-Elmer. 1995 ABI Prism GenScan Analysis Version 2.0.2. Applied Biosystems Division. Foster City, California, USA.

———. 1995 Genotyper Software Version 1.1. Applied Biosystems Division. Foster City, California, USA.

Renvoize, B. S. 1972 The area of origin of Manihot esculenta as a crop plant-a review of the evidence. Economic Botany 26: 352–360.[ISI]

Roa, A. C., M. M. Maya, M. C. Duque, J. Tohme, A. C., Allem, and M. W. Bonierbale. 1997 AFLP analysis of relationships among cassava and other Manihot species. Theoretical and Applied Genetics 95: 741–750.[CrossRef][ISI]

Rogers, D. J. 1963 Studies of Manihot esculenta and related species. Bulletin of the Torrey Botanical Club 90: 43–54.[CrossRef]

———, and S. G. Appan. 1973 Manihot, Manihotoides (Euphorbiaceae). Flora Neotropica Monograph number 13. Hafner Press, New York, New York, USA.

Rohlf, F. J. 1994 NTSYS-pc: numerical taxonomy and multivariate system, version 1.80. Exeter Software, Setauket, New York, USA.

Slatkin, M. 1995 A measure of population subdivision based on microsatellite allele frequencies. Genetics 139: 457–462.[ISI][Medline]

Smulders, M. J. M., G. Bredemeijer, W. Rus-Kortekass, P. Arens, and B. Vosman. 1997 Use of short microsatellites from database sequences to generate polymorphisms among Lycopersicon esculentum cultivars and accessions of other Lycopersicon species. Theoretical and Applied Genetics 97: 264–272.

Sokal, R. R., and J. Rohlf. 1995 Biometry: the principles and practice of statistics in biological research, 3rd. ed. W.H. Freeman, New York, New York, USA.

Steinkellner, H., C. Lexer, E. Turetschek, and J. Glössl. 1997 Conservation of (GA)n microsatellite between Quercus species. Molecular Ecology 6: 1189–1194.[CrossRef]

Tautz, D., M. Trick, and G. A. Dover. 1986 Cryptic simplicity in DNA is a major source of genetic variation. Nature (London) 322: 652–656.[CrossRef][Medline]

White, G., and W. Powell. 1997 Cross-species amplification of SSR loci in the Meliaceae family. Molecular Ecology 6: 1195–1197.[CrossRef]

Witsenboer, H., J. Vogel, and R. W. Michelmore. 1997 Identification, genetic localisation, and allelic diversity of selectively amplified microsatellite polymorphic loci in lettuce and wild relatives (Lactuca spp.). Genome 40: 923–936.

Wu, K. S., and S. D. Tanksley. 1993 Abundance, polymorphism and genetic mapping of microsatellites in rice. Molecular and General Genetics 241: 225–235.

Yeh, F. C., Y. Rongcai, and T. Boyle. 1997 PopGene Version 1.2, Microsoft Window- based software for population genetics analysis: a quick user's guide. University Alberta/Center for International Forestry Research, Alberta, Canada.




This article has been cited by other articles:


Home page
 jashsHome page
Y. Wang, M. Kang, and H. Huang
Microsatellite Loci Transferability in Chestnut
J. Amer. Soc. Hort. Sci., September 1, 2008; 133(5): 692 - 700.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
M. E. Halsey, K. M. Olsen, N. J. Taylor, and P. Chavarriaga-Aguirre
Reproductive Biology of Cassava (Manihot esculenta Crantz) and Isolation of Experimental Field Trials
Crop Sci., January 16, 2008; 48(1): 49 - 58.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
E. Gaitan-Solis, M. C. Duque, K. J. Edwards, and J. Tohme
Microsatellite Repeats in Common Bean (Phaseolus vulgaris): Isolation, Characterization, and Cross-Species Amplification in Phaseolus ssp.
Crop Sci., November 1, 2002; 42(6): 2128 - 2136.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
C. M. Raker and D. M. Spooner
Chilean Tetraploid Cultivated Potato, Solanum tuberosum, is Distinct from the Andean Populations: Microsatellite Data
Crop Sci., September 1, 2002; 42(5): 1451 - 1458.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
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 (38)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Roa, A. C.
Right arrow Articles by Tohme, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Roa, A. C.
Right arrow Articles by Tohme, J.
Agricola
Right arrow Articles by Roa, A. C.
Right arrow Articles by Tohme, J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS