Sequence Diversity in the Tetraploid Zea perennis and the Closely Related Diploid Z. diploperennis: Insights From Four Nuclear Loci

Corresponding author: Peter Tiffin, Department of Ecology and Evolutionary Biology, 321 Steinhaus Hall, University of California, Irvine, CA 92697-2525., ptiffin{at}uci.edu (E-mail)

Communicating editor: D. CHARLESWORTH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Polyploidy has been an extremely common phenomenon in the evolutionary history of angiosperms. Despite this there are few data available to evaluate the effects of polyploidy on genetic diversity and to compare the relative effects of drift and selection in polyploids and related diploids. We investigated DNA sequence diversity at four nuclear loci (adh1, glb1, c1, and waxy) from the tetraploid Zea perennis and the closely related diploid Z. diploperennis. Contrary to expectations, we detected no strong evidence for greater genetic diversity in the tetraploid, or for consistent differences in the effects of either drift or selection between the tetraploid and the diploid. Our failure to find greater genetic diversity in Z. perennis may result from its relatively recent origin or demographic factors associated with its origin. In addition to comparing genetic diversity in the two species, we constructed genealogies to infer the evolutionary origin of Z. perennis. Although these genealogies are equivocal regarding the mode of origin, several aspects of these genealogies support an autotetraploid origin. Consistent with previous molecular data the genealogies do not, however, support the division of Zea into two sections, the section Zea and the section Luxuriantes.

THIRTY to 70% of angiosperm species have polyploid events in their evolutionary history (STEBBINS 1950 ; MASTERSON 1994 ), and hence polyploid events are extremely common in the evolutionary history of flowering plants. Polyploidy is not, however, limited to plants. Yeast, insects, reptiles, mammals, and fish also have polyploid events in their evolutionary histories (LEWIS 1980 ; POSTLETHWAIT et al. 1998 ; SEOIGHE and WOLFE 1998 ). Yet, despite the evolutionary importance of polyploidy, there have been few studies of its effect on genetic diversity at the DNA level.

Polyploids are often classified into allopolyploids, which result from interspecific hybridization, or autopolyploids, which form through intraspecific chromosomal duplications (STEBBINS 1947 ). These modes of polyploid formation have different effects on the genome; allopolyploid results in twice the number of loci, whereas autopolyploid results in the same number of loci but twice the number of alleles segregating at each locus. However, both modes of formation are expected to result in polyploids having more genetic diversity than closely related diploids. Allopolyploids are expected to have greater genetic diversity because allopolyploids merge two independent evolutionary lineages (CLAUSEN et al. 1945 ; LEWIS 1980 ; GRANT 1981 ).

In contrast, when autopolyploid species are formed they will have equal or less genetic diversity than the progenitor diploid species. However, because tetrasomic inheritance doubles the number of alleles segregating at each locus, autopolyploids have larger effective population sizes (N_e) than their diploid progenitors (all other things, like population size, being equal). Genetic drift is slowed with a larger N_e, and thus autopolyploids should maintain greater levels of neutral genetic variation than their diploid ancestors (MOODY et al. 1993 ). The effect of selection on nonneutral variation is also expected to differ between autotetraploids and related diploids. For example, in a diploid population a recessive allele with a frequency of 0.1 is exposed to selection in 1% of individuals; whereas, a recessive allele in an autotetraploid population must reach a frequency of 0.3 before it will be exposed to selection in 1% of individuals. This difference in the efficacy of selection means that recessive alleles will be maintained in populations for much longer periods of time in autopolyploids than diploids (HALDANE 1930 ; WRIGHT 1938 ; HILL 1971 ; OTTO and WHITTON 2000 ). However, partially dominant alleles may respond to selection faster, and thus be maintained for shorter periods of time in autopolyploids than diploids (OTTO and WHITTON 2000 ).

Previous empirical investigations on the molecular diversity of tetraploids have relied primarily on isozyme surveys. The results from these studies are generally consistent with theoretical expectations of higher genetic diversity in tetraploids than their diploid progenitors (SOLTIS and SOLTIS 1993 ). However, isozymes capture only a fraction of allelic polymorphisms present in populations, and thus allozyme data are of limited use for assessing the relative effects of drift and selection. DNA sequence data provide both a more detailed measure of genetic diversity and a means to compare the relative effects of drift and selection between a polyploid and its diploid relatives. For example, interspecific differences in the frequency distributions of polymorphisms or in the ratio of nonsynonymous to synonymous variation may reflect differences in the relative strength of selection within the two species (SAWYER et al. 1987 ; MCDONALD and KREITMAN 1991 ; AKASHI 1999 ). Similarly, slowed drift in an autopolyploid may be reflected by higher diversity at synonymous sites. In short, comparisons of DNA sequence diversity between a polyploid and closely related diploids should provide insight into the relative strengths of the evolutionary forces acting on the different species.

Recent studies on gene expression and genomic divergence have begun to shed light on differences in the evolution of allopolyploids and diploids (SONG et al. 1995 ; CRONN et al. 1999 ; GALITSKI et al. 1999 ; SMALL and WENDEL 2000 ; WENDEL 2000 ). In contrast, DNA sequence diversity in autotetraploids and closely related diploids has not been investigated. In this article we present data on DNA sequence diversity at four nuclear loci–adh1, glb1, c1, and waxy–in the tetraploid plant Zea perennis and its closest diploid relative Z. diploperennis. Z. perennis and Z. diploperennis are morphologically similar, primarily outcrossing species that are endemic to the Cerro de San Miquel, Sierra de Manantlan, in southwestern Jalisco, Mexico (ILTIS et al. 1979 ; BENZ et al. 1990 ). Both species are congeners of the domesticate Z. mays ssp. mays and are the only perennial species in the genus Zea.

Z. perennis is generally thought to be an autotetraploid that originated from a Z. diploperennis-like ancestor (CLAUSEN et al. 1945 ). An autotetraploid origin is supported by morphological (ILTIS et al. 1979 ), ribosomal internal transcribed spacer (ITS) sequence (BUCKLER and HOLTSFORD 1996 ), chromosomal knob (KATO and LOPEZ 1990 ), and chloroplast restriction fragment length polymorphism (DOEBLEY et al. 1987 ) similarities. Chromosome segregation patterns in Z. perennis, however, are equivocal with regard to origin. RANDOLPH 1955 reported Z. perennis to form a prevalence of tetravalents at meiosis, as expected from an autotetraploid. However, SHAVER 1962 found approximately equal frequencies of tetravalents and bivalents at meiosis. These data suggest that Z. perennis is either an autotetraploid that has partially diploidized or Z. perennis is actually an allotetraploid. If Z. perennis is an allotetraploid, the high numbers of tetravalents formed at meiosis may indicate that closely related species hybridized to form the polyploid or that the genetic contribution of the two progenitor species was not equal, perhaps due to more extensive introgression from one of the two progenitor species.

An allotetraploid origin is also suggested by an investigation of restriction sites that identified a chloroplast haplotype from the Piedra Ancha population of Z. perennis (accession per6 in this study) that is considerably different from chloroplast haplotypes isolated from Z. diploperennis or other Z. perennis individuals (DOEBLEY 1989 ). This atypical chloroplast haplotype, although not found in any other Zea species, is more closely related to Z. mays than any other Zea species and thus seems to support the possibility of an allotetraploid origin of Z. perennis (DOEBLEY 1989 ).

The main objective of this study is to investigate the effects of polyploidy on genetic diversity by comparing DNA sequence diversity in a presumed autotetraploid to its most closely related diploid relative. In addition, because there is some uncertainty as to the origin of Z. perennis, we ask whether genealogical data support an intraspecific or interspecific origin of this species. Sequence data for the four genes we studied are also available from all other diploid species within the genus Zea, providing opportunities both to estimate the relatedness of Z. perennis alleles to those of other possible progenitors and to evaluate phylogenetic relationships among Zea species.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Sampling DNA sequences:
We PCR amplified the following sections of DNA: a 1400-bp portion of adh1 (alcohol dehydrogenase-1), a 1200-bp portion of glb1 (globulin-1, a nonessential seed storage protein), a 750-bp portion of c1 (a bHLH transcriptional regulator of enzymatic genes in the anthocyanin biosynthetic pathway), and a 1400-bp portion of waxy (granule bound starch synthase; MASON-GAMER et al. 1998 ). adh1 and glb1 are located within 12 cM of one another on chromosome 1, and c1 and waxy are located within 30 cM of one another on chromosome 9. waxy was amplified using PCR with 35 cycles of 1 min at 94°, 2 min at 62°, and 2 min at 72° (forward primer, 5' tgcgagctagacaacatcatgcgcc 3'; reverse primer, 5' agggcgcggccacgttctcc 3'); and c1 was amplified using PCR with 35 cycles of 1 min at 94°, 2 min at 60°, and 2 min at 72° (forward primer, 5' cactggggatccttagttactggcatg 3'; reverse primer, 5' cataggtaccagcgtgctgttccagtagt 3'). Details of the PCR amplification conditions and primers for adh1 and glb1 are reported in EYRE-WALKER et al. 1998 (adh1) and HILTON and GAUT 1998 (glb1). In most cases, sequences were amplified from a single individual from each of eight accessions of Z. diploperennis and six accessions of Z. perennis (Table 1). However, we were unable to obtain full-length sequences for adh1 from Z. diploperennis accession 2 or glb1 from Z. perennis accession 2. DNA was extracted from leaves using QIAGEN (Chatsworth, CA) DNeasy DNA extraction kits.

We sampled multiple alleles at each locus from all Z. perennis and several Z. diploperennis individuals. For each of the diploids we performed a single PCR reaction. PCR products were cloned into pGEM-T vectors (Stratagene, La Jolla, CA) and plasmid DNA was isolated from three clones using QIAGEN minipreps. Each of these clones was partially sequenced, using one of the amplification primers. Nonidentical clones from a single individual were then sequenced in their entirety using the PCR primers as well as internal primers (except for c1 for which only the two PCR primers were used), Perkin Elmer (Norwalk, CT) Big-Dye sequencing chemistry, and an ABI 377 automated sequencer.

Because we were unsure of the origin of Z. perennis and we wanted to ensure sufficient sample to detect evidence for a possible allotetraploid origin, we sampled Z. perennis much more extensively than Z. diploperennis. Our primary motivation for this extensive sampling was that if loci within Z. perennis were disomic (i.e., the loci were duplicated), we wanted to be certain that we sampled alleles from both loci. For each Z. perennis individual we performed two PCR reactions. The products from each of these reactions were cloned and a total of 12 clones were chosen for each Z. perennis individual, six colonies from each of the two original PCR reactions. Each of these 12 clones was partially sequenced using one end primer to identify alleles that differed within individuals. Nonidentical clones from a single individual were sequenced in their entirety as described above. This sampling scheme resulted in full sequences of multiple alleles from each individual.

More than half of the alleles identified after the initial data collection contained one or more unique single base pair changes (or "singletons") relative to the remainder of the sequences. Because singletons in individual cloned products can represent either true sequence variation or polymerase error, we reamplified all DNAs containing singletons to determine which singletons were true variants and which were the results of PCR error. We checked singletons in Z. diploperennis by reamplifying and resequencing the appropriate allele from each individual. The task was more burdensome in Z. perennis because we did not know the number of different alleles present within each tetraploid individual. It was therefore not always possible to determine if a clone obtained upon reamplification represented the allele without the singleton (thus indicating that the initial singleton was due to PCR error) or an allele that was not sampled previously. Because of this uncertainty, we employed an extensive sampling strategy to confirm the presence of singletons. For each tetraploid DNA from which at least one allele contained a singleton, we performed two additional PCR reactions, cloned the products from these reactions, and isolated and partially sequenced 6 isolates from each reaction (a total of 12 isolates from each tetraploid DNA). Assuming that each tetraploid plant contained four distinct alleles, and that there was no amplification bias in the PCR reactions, sampling 12 isolates results in a >95% probability of resampling the allele that initially contained the singleton. Approximately one-half of the singletons in the initial data set were confirmed by this strategy, while the other half were assumed to have resulted from polymerase error and excluded from the data set. Our estimated rate of polymerase error was 1 in 1200 bp, which is similar to previously reported polymerase error for Zea DNAs (EYRE-WALKER et al. 1998 ).

Corrected sequences were aligned manually along with previously published sequences from other Zea species. Sequences have been submitted to GenBank (Table 1).

Sequence analyses:
For each of the four loci, genealogies were constructed using the neighbor-joining method (SAITOU and NEI 1987 ) on the basis of Kimura's two-parameter genetic distance, using a sequence from Tripsacum dactyloides as an outgroup. Data were resampled 1000 times for bootstrap analyses. When constructing genealogies, we included sequences from other members of the genus Zea. Genealogies were constructed with PAUP* version 4.0b (SWOFFORD 1990 ). The data included sequences from adh1 (GenBank nos. L08587, L08588, L08589, L08590, L08591, AF044289, AF044290, AF044291, AF044292, AF044293, AF044294, AF044295, AF044296, AF044297, AF044298, AF044299, AF044300, AF044301, AF044302, AF044303, AF044304, AF044305, AF044306, AF044307, AF045548, and AF293887, AF293888, AF293889, AF293890, AF293891, AF293892, AF293893, AF293894; EYRE-WALKER et al. 1998), c1 (GenBank nos. AF292540, AF292541, AF292542, AF292543, AF292544, AF292545, AF292546, AF292547, AF292548, AF292549, AF292550, AF292551, AF292552, AF292553; HANSON et al. 1996 ; L. ZHANG, A. PEEK, D. DUNAMS and B. S. GAUT, unpublished results), glb1 (GenBank nos. U28017, X59084, and AF064212, AF064213, AF064214, AF064215, AF064216, AF064217, AF064218, AF064219, AF064220, AF064221, AF064222, AF064223, AF064224, AF064225, AF064226, AF064227, AF064228, AF064229, AF064230, AF064231, AF064232, AF064233; HILTON and GAUT 1998), and waxy (GenBank nos. AF292500, AF292501, AF292502, AF292503, AF292504, AF292505, AF292506, AF292507, AF292508, AF292509, AF292510, AF292511, AF292512, AF292513, AF292514, AF292515, AF292516, AF292517, AF292518, AF292519, AF292520, AF292521, AF292522, AF292523, AF292524, AF292525, AF292526, AF292527, AF292528, AF292529, AF292530, AF292531, AF292532, AF292533, AF292534, AF292535, AF292536, AF292537, AF292538, AF292539; L. ZHANG, A. PEEK, D. DUNAMS and B. S. GAUT, unpublished results).

Estimates of genetic diversity, (TAJIMA 1983 ) and (WATTERSON 1975 ), were calculated separately on the basis of silent (synonymous and intron sites) and nonsynonymous sites. The recombination rate at each locus C (= 4Nc) was estimated using the method of HUDSON 1987 . Evidence for nonneutral evolution was investigated using the tests of HUDSON et al. (HKA; 1987), TAJIMA 1989 , MCDONALD and KREITMAN (MK; 1991), and FU and LI 1993 . HKA tests were performed with pairs of genes, resulting in six HKA tests for both Z. perennis and Z. diploperennis. For all HKA and MK tests, T. dactyloides sequences were used as the reference species. Ratios of the number of synonymous to nonsynonymous segregating sites and the frequency of singleton and nonsingleton segregating sites in Z. diploperennis and Z. perennis were compared using 2 x 2 contingency tests (SAWYER et al. 1987 ; AKASHI 1999 ). Sites that were polymorphic in one species were included in the analyses, regardless if that site was fixed or polymorphic in the second species. Polymorphism metrics and tests of selection were calculated with DnaSP version 3.0 (ROZAS and ROZAS 1999 ). All nucleotide sites (synonymous, nonsynonymous, and intron sites) were used for phylogenetic reconstruction and indels were excluded from all analyses.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Patterns of DNA sequence diversity within and among species:
Estimates of DNA sequence diversity, and , made separately on silent sites and nonsynonymous sites provided no clear evidence for more genetic diversity in tetraploid Z. perennis than diploid Z. diploperennis (Table 2). Similarly, estimates of C, the number of recombination events per gene (HUDSON et al. 1987 ), were also not consistently different across genes between species.

To determine whether the selective histories differed between the two species and among genes, we applied four tests of neutrality to the sequence data. These tests—Tajima's D (Table 2, all P > 0.2), Fu and Li's D (Table 2, all P > 0.10), Fu and Li's F (all P > 0.10, data not shown), and HKA (all P > 0.3, data not shown)—revealed no evidence of deviation from the neutral model. MK tests also were nonsignificant for c1, glb1, and waxy. In contrast, MK tests of adh1 data indicated significant departure from neutral evolution in Z. perennis (P < 0.05). However, this result is not significant after Bonferroni correction for multiple tests, and it seems unlikely that adh1 is under strong selection, given that HKA tests, intraspecific tests of selection, and a previous study of adh1 in Zea (EYRE-WALKER et al. 1998 ) have failed to reject neutral evolution for this gene.

A primary motivation for this study was to compare the distribution and pattern of polymorphism between a tetraploid and closely related diploid species. Interspecific differences in the frequency distributions of nonsynonymous to synonymous polymorphisms and rare to common polymorphisms can reflect differences in the relative strengths of selection and drift operating within each species (SAWYER et al. 1987 ; AKASHI 1999 ). We therefore compared the ratio of nonsynonymous to synonymous polymorphic sites with Z. perennis vs. Z. diploperennis. The ratio did not differ significantly between the two species for any of the loci (Fisher's exact test, all P > 0.20) or when data from all genes were analyzed together (Fisher's exact test, P > 0.5; Table 3).

We also compared the distribution of singleton vs. nonsingleton polymorphisms at synonymous, nonsynonymous, and all sites. For all genes and at all classes of sites the ratio of singleton to nonsingleton polymorphisms was lower in Z. perennis than Z. diploperennis. For three of the four genes these differences were not significant (G-test, all comparisons P > 0.10; Table 3), whereas at glb1 the distribution of singleton to nonsingleton polymorphisms at all sites, including introns, differed significantly between the two species (P < 0.01). Likewise, a contingency test comparing the distribution of singleton to nonsingleton polymorphic sites over all genes was significant (P < 0.01), although the significance of this comparison was largely due to the glb1 data. Removing the glb1 data resulted in a difference that was only marginally significant (P = 0.062). Overall, the tetraploid appears to have a lower relative frequency of singletons, although the significance of this result is due largely to one of the four genes, glb1.

Genealogical analysis:
Several aspects of the genealogies (Fig 1 Fig 2 Fig 3 Fig 4) support an autotetraploid origin of Z. perennis from Z. diploperennis. First, with one exception we found no well-supported clades (i.e., bootstrap support >50%) that contained alleles from either Z. perennis or Z. diploperennis and any other Zea species. Second, several well-supported clades contained alleles isolated from both Z. perennis and Z. diploperennis. Third, the majority of clades that contained Z. perennis alleles also contained Z. diploperennis alleles or were sister to clades containing Z. diploperennis alleles. The only strong evidence for an autotetraploid origin, however, comes from the c1 locus; all c1 alleles isolated from these two species formed a single monophyletic clade.

View larger version (18K): In this window In a new window Download PPT slide   Figure 1. Neighbor-joining reconstructions of the genealogical relationships among adh1 alleles. Bootstrap values for nodes supported in >50% of 1000 bootstrap replicates are given above the branches leading to those nodes. Zea perennis and Z. diploperennis alleles are named as indicated in Table 1. Z. mays ssp. mays (M), Z. mays ssp. parviglumis (P), Z. mays ssp. mexicana (MX), and Z. luxurians (L) sequences were published previously (see MATERIALS AND METHODS for references).

View larger version (15K): In this window In a new window Download PPT slide   Figure 2. Neighbor-joining reconstructions of the genealogical relationships among glb1 alleles. Allele designations are described in Fig 1.

View larger version (16K): In this window In a new window Download PPT slide   Figure 3. Neighbor-joining reconstructions of the genealogical relationships among c1 alleles. Allele designations are described in Fig 1.

View larger version (19K): In this window In a new window Download PPT slide   Figure 4. Neighbor-joining reconstructions of the genealogical relationships among waxy alleles. Allele designations are described in Fig 1.

The genealogies are not, however, entirely consistent with an autotetraploid origin. The strongest evidence against an autotetraploid origin comes from the adh1 locus where one clade of Z. perennis alleles (per7a, per4b, per1a, and per2b) is sister to a clade of Z. mays ssp. mays alleles (Fig 1). However, the node joining these two clades and other internal nodes within the adh1 genealogy are poorly supported, and thus any inferences regarding the genealogical placement of these alleles should be viewed with caution. It is also possible that these adh1 alleles are the result of introgression of Z. mays ssp. alleles into Z. perennis. Nevertheless, these alleles could be interpreted as evidence of an allotetraploid origin for Z. perennis.

In addition to the genealogical data, the ratio of shared polymorphisms and fixed differences between Z. perennis and three of the diploid species of the genus Zea (Z. diploperennis, Z. luxurians, and Z. mays ssp. parviglumis; Table 4) is not entirely consistent with an autotetraploid origin. If Z. perennis were an autotetraploid we expect the ratio of shared polymorphisms to fixed differences to be much higher between Z. perennis and Z. diploperennis than between Z. perennis and any of the other species. The ratio of shared polymorphisms to fixed differences between Z. perennis and Z. diploperennis is, in fact, higher than between other species pairs but is not significantly higher than the ratio between Z. perennis and Z. mays ssp. parviglumis.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Genetic diversity:
The primary purpose of this study was to compare genetic diversity between a tetraploid, Z. perennis, and a closely related diploid Z. diploperennis. In theory, allotetraploids should have higher genetic diversity than diploids because they are formed through hybridization of two divergent genomes. Likewise, autopolypoidy can slow the effects of both drift and selection on genetic diversity, potentially resulting in both greater diversity and different patterns of molecular diversity in autotetraploids than diploids (HALDANE 1930 ; WRIGHT 1938 ; HILL 1971 ; MOODY et al. 1993 ). Contrary to these expectations, we detected little evidence for differential patterns of drift and selection in tetraploid Z. perennis relative to its closest diploid relative Z. diploperennis.

There are at least two possible explanations for the absence of substantial differences in the pattern and level of genetic diversity between the tetraploid and the diploid. One possibility is that Z. perennis is an autotetraploid that originated too recently to allow for the accumulation of mutations that differentiate the two species. We estimated the time of Z. diploperennis-Z. perennis divergence using a method of NEI 1987 (p. 277), which is based on the average number of nucleotide substitutions within and between populations, and estimated substitution rates of 5 x 10^-9–30 x 10^-9 substitutions per site per year (WOLFE et al. 1987 ; GAUT and CLEGG 1991 ). The estimated divergence times based on the high and low substitution rates, respectively, were 63,500–381,000 years (adh1), 194,700–1,168,000 years (glb1), 41,800–251,000 years (c1), and 24,000–142,000 years (waxy). Thus the preponderance of evidence supports divergence within the past few hundred thousand years.

A recent divergence between these species is also supported by the large number of polymorphisms that are shared between Z. perennis and Z. diploperennis and several strongly supported clades that are composed predominantly of Z. perennis or Z. diploperennis alleles but also contain alleles from the other species. Interestingly, the glb1 gene, which shows no evidence of lineage sorting, is also the only gene with a ratio of common to rare polymorphisms that is significantly higher in Z. perennis than Z. diploperennis. This, together with higher ratios of common to rare polymorphisms in all four genes, as would be expected if drift were slowed in the tetraploid (MOODY et al. 1993 ), could be indicative of increasing diversity in Z. perennis.

A second possible explanation for finding approximately equal amounts of genetic diversity in the tetraploid and diploid is that demographic forces associated with polyploid formation offset the forces that lead to greater diversity. Theoretical arguments for greater genetic diversity in autotetraploids are based on the assumption that increasing the number of chromosomes per individual increases effective population size. This implicitly assumes that the tetraploid and diploid species have had approximately equal population sizes. However, autotetraploid formation may involve a diversity-reducing population bottleneck (STEBBINS 1980 ). Nevertheless, our data provide no evidence for a severe bottleneck in the formation of Z. perennis. If Z. perennis has experienced a severe bottleneck, then the Tajima's D's should be both negative and lower than Tajima's D's calculated from Z. diploperennis; but this was not the case (Table 2). It is possible that Z. perennis has had multiple origins and this is the reason for finding little evidence for a bottleneck. Distinguishing multiple origins from gene flow and segregation of ancient polymorphisms is not possible from our data; however, the wide distribution of Z. perennis alleles within the genealogies suggests that multiple origin events are a possibility. The primary point of this discussion is that the demographic forces associated with tetraploid formation may offset the evolutionary forces contributing to greater genetic diversity in tetraploids.

If Z. perennis is in fact an autotetraploid, then our finding of approximately equal DNA sequence diversity in Z. perennis and its closest diploid relative appears inconsistent with the general finding that autotetraploids have a higher number of allozyme alleles than their diploid progenitors (reviewed in SOLTIS and SOLTIS 1993 ). However, allozyme studies that find greater genetic diversity in autotetraploids than diploids may be biased by unequal sampling. SOLTIS and SOLTIS 1993 reviewed data from six studies comparing autotetraploids to their diploid progenitors and found a greater number of allelic types in the tetraploid than the diploid species in five of six of these comparisons. However, at least five of these studies sampled considerably more alleles from the tetraploid (1.5 to nearly 4 times more alleles were sampled from the tetraploids; sampling data from the sixth study were not available). Because greater sampling effort is likely to result in the detection of a greater number of allelic types, the results from these allozyme studies must be interpreted with caution. In contrast, estimates of and correct for sample size and thus these measures of diversity should be robust to deviations in sampling intensity.

Because genetic diversity has been measured for other diploid species of Zea, we can compare sequence diversity among several other members of the genus (Table 5). At the four loci we studied, both Z. perennis and Z. diploperennis harbor lower levels of genetic diversity than Z. mays ssp. parviglumis and in general have lower diversity than the domesticate Z. mays ssp. mays. In contrast, Z. perennis has higher diversity than Z. luxurians, whereas Z. diploperennis and Z. luxurians have approximately equal amounts of diversity. The exception to these general patterns is the c1 locus. At c1, Z. mays ssp. mays has lowest diversity and Z. perennis has lower diversity than either Z. luxurians or Z. diploperennis. c1 has previously been hypothesized to evolve in response to selection during domestication, which could explain the low level of diversity in Z. mays ssp. mays. However, it is not clear why Z. perennis should also have relatively low diversity at this locus, especially since we found no evidence of selection having acted on this locus in this taxon. Altogether, these sequence data are in broad agreement with earlier isozyme studies of Zea species. Both sequence and isozyme data show that among all Zea species, Z. mays ssp. parviglumis has the greatest amount of genetic diversity and Z. luxurians has the least amount. However, sequence and isozyme data do not agree with regard to the relative levels of diversity within Z. perennis and Z. diploperennis. The sequence data revealed generally higher diversity in Z. perennis than Z. diploperennis or Z. luxurians, whereas isozyme data revealed higher diversity in Z. diploperennis than in Z. perennis or Z. luxurians (DOEBLEY et al. 1984 ).

View this table: In this window In a new window   Table 5. values and their standard errors (in parentheses) of adh1, glb1, c1, and waxy from five taxa of Zea

Origin of Z. perennis and phylogenetic relationships among Zea species:
Genealogical data may be useful for inferring the evolutionary origin of tetraploid species (OKANE et al. 1996 ; SMALL et al. 1998 ; BARRIER et al. 1999 ; SMALL and WENDEL 2000 ; WENDEL 2000 ). Alleles sampled from an allopolyploid are expected to form two clades, one of which is either interspersed with or sister to clades formed by alleles from each of the parental taxa. In contrast, alleles sampled from an autopolyploid are expected to group phylogenetically with alleles from the species from which the polyploid formed. If the autopolyploid evolved recently, the alleles from the polyploid should be interspersed with alleles from the diploid progenitor; whereas, alleles from older autopolyploids should form clades that are monophyletic sister clades to the diploid progenitors.

Although the genealogical relationships among the sequences we sampled are not well enough resolved to provide conclusive evidence, our data largely support an autopolyploid origin of Z. perennis from a Z. diploperennis-like ancestor. This support comes from four observations. First, with one exception there were no well-supported clades that contained alleles from Z. perennis and any other Zea species. Second, several well-supported clades contained alleles isolated from both Z. perennis and Z. diploperennis. Third, the majority of clades that contained Z. perennis alleles also contained Z. diploperennis alleles or were sister to clades containing Z. diploperennis alleles. The strongest evidence for an autotetraploid origin, however, comes from the c1 locus; all c1 alleles isolated from these two species formed a single monophyletic clade. These genealogical inferences are consistent with the preponderance of evidence from morphological, isoenzyme, and meiotic pairing data that also support an intraspecific origin of Z. perennis (CLAUSEN et al. 1945 ; SHAVER 1962 ; KATO and LOPEZ 1990 ).

Two aspects of the data, however, appear to conflict with an autotetraploid origin but are consistent with Z. perennis forming through interspecific hybridization between ancestors of Z. diploperennis and Z mays ssp. parviglumis or a now-extinct Zea species (DOEBLEY 1989 ). First, one clade of Z. perennis adh1 sequences clusters with Z. mays ssp. mays sequences, although the branches connecting these clades have weak phylogenetic support. The second apparent conflict with an autotetraploid origin is that the ratio of fixed to shared polymorphisms between Z. perennis to Z. mays ssp. parviglumis is approximately equal to the ratio between Z. perennis and Z. diploperennis (Table 4). However, this too is mitigated by the high ratio of fixed to shared polymorphisms between Z. diploperennis and Z. mays ssp. parviglumis, suggesting that the high ratio of shared polymorphisms may be due to lineage sorting and not reflective of an interspecific origin. Although evidence for an interspecific origin is certainly no stronger than evidence for an autopolyploid origin, we cannot exclude the possibility that Z. perennis was formed through interspecific hybridization. If Z. perennis was formed through interspecific hybridization it may be most similar to Z. diploperennis today because of a high rate of Z. diploperennis introgression, or because Z. diploperennis genes were preferentially maintained as the Z. perennis genome evolved.

The genus Zea has been separated into two sections, section Luxuriantes, composed of Z. luxurians, Z. diploperennis, and Z. perennis; and section Zea, composed of Z. mays ssp. mays, Z. mays ssp. mexicana, Z. mays ssp. Parviglumis, and Z. mays ssp. huehuetenagensis (DOEBLEY 1990 ). This division is supported by chloroplast restriction site polymorphism data (DOEBLEY et al. 1987 ) and weakly supported by morphological (DOEBLEY 1983 ) and isoenzyme data (DOEBLEY et al. 1984 ). In contrast, none of the four loci we studied formed a monophyletic group that included Z. perennis, Z. diploperennis, and Z. luxurians. Moreover, the distribution of shared polymorphisms and fixed differences between Z. perennis, Z. diploperennis, and Z. luxurians suggests that both Z. diploperennis and Z. perennis are more distantly related to Z. luxurians than to Z. mays ssp. parviglumis (Table 4). As such, our data do not support the section Luxuriantes as a natural grouping, but this conclusion is subject to the caveat that Z. mays ssp. parviglumis is known to segregate ancient alleles, thus complicating phylogenetic conclusions (GAUT and CLEGG 1993 ). However, previous molecular sequence data, including ITS data (BUCKLER and HOLTSFORD 1996 ), limited restriction site data from the IGS (ribosomal intergeneric spacer) and 5SDNA (ZIMMER et al. 1988 ), and Magellan retrotransposon sequence (PURUGGANAN and WESSLER 1994 ) also do not support section Luxuriantes as a natural grouping.

Our data also provide some evidence that the phylogenetic relationships among the genus Zea may be complicated by introgression between taxa. In particular, a glb1 allele from Z. diploperennis (dip1a, Fig 3) was very similar to and formed a well-supported clade with two Z. mays ssp. mays alleles. In contrast, all other Z. perennis and Z. diploperennis glb1 alleles formed well-supported clades that contained no alleles from other Zea species. Previous investigations of nuclear genes in the genus Zea also provide evidence of possible introgression between Z. mays ssp. mays and Z. diploperennis. Isozyme surveys of Zea found a Z. diploperennis individual with two alleles that were otherwise unknown in Z. diploperennis but common in Z. mays ssp. mays (DOEBLEY et al. 1984 ). In addition, several Z. mays ssp. mays-like ITS sequences were isolated from Z. diploperennis and Z. perennis individuals (BUCKLER and HOLTSFORD 1996 ). Introgression between Z. mays ssp. mays and Z. diploperennis may not be surprising given that hybrids between these species are reported to be viable and fertile (POSPUPULETI and GALINAT 1982 ; BENZ et al. 1990 ). Moreover, farmers in the Sierra de Manantlan region of Mexico, where Z. perennis and Z. diploperennis grow, reportedly hybridize Z. diploperennis and Z. mays ssp. mays in an attempt to improve productivity and impart pest resistance to traditional maize races (BENZ et al. 1990 ). These hybrids are often left to grow in the maize field for 3–5 years. Because populations of Z. diploperennis often surround maize fields in this area, the hybrids may provide a bridge for introgression of Z. mays ssp. mays alleles into the Z. diploperennis gene pool.

	ACKNOWLEDGMENTS

This work was improved by discussions with J. D. Bever. We also thank E. S. Buckler, R. L. Small, L. Zhang, and two anonymous reviewers for comments on the manuscript. This work was supported by National Science Foundation grant 98-15855 to B.S.G. P.T. was supported in part by National Research Initiative Competitive Grants Program/U.S. Department of Agriculture Award 99-35301-8076.

Manuscript received October 23, 2000; Accepted for publication February 5, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AKASHI, H., 1999  Inferring the fitness effects of DNA mutations from polymorphism and divergence data: statistical power to detect directional selection under stationarity and free recombination. Genetics 151:221-238[Abstract/Free Full Text].

BARRIER, M., B. G. BALDWIN, R. H. ROBICHAUS, and M. D. PURUGGANAN, 1999  Interspecific hybrid ancestry of a plant adaptive radiation: allopolyploidy of the Hawaiian silversword alliance (Asteraceae) inferred from floral homeotic gene duplications. Mol. Biol. Evol. 6:1105-1113.

BENZ, B. F., L. R. SANCHEZ-VELASQUEZ, and F. J. SANTANA MICHEL, 1990  Ecology and ethnobotany of Zea diploperennis: preliminary investigations. Maydica 35:85-98.

BUCKLER, E. S. and T. P. HOLTSFORD, 1996  Zea systematics: ribosomal ITS evidence. Mol. Biol. Evol. 13:612-622[Abstract].

CLAUSEN, J., D. D. KECK, and W. H. HIESEY, 1945  Experimental studies on the nature of species. II. Plant evolution through amphidiploidy and autopolyploidy with examples from the Madiinae. Carnegie Inst. Wash. 564:1.

CRONN, R. C., R. L. SMALL, and J. F. WENDEL, 1999  Duplicated genes evolve independently after polyploid formation in cotton. Proc. Natl. Acad. Sci. USA 96:14406-14411[Abstract/Free Full Text].

DOEBLEY, J., 1983  The maize and teosinte male inflorescence—a numerical taxonomic study. Ann. Mo. Bot. Gard. 70:32-70.

DOEBLEY, J., 1989  Molecular evidence for a missing wild relative of maize and the introgression of its chloroplast genome into Zea perennis. Evolution 43:1555-1559.

DOEBLEY, J., 1990  Molecular systematics of Zea (Gramineae). Maydica 35:143-150.

DOEBLEY, J., M. M. GOODMAN, and C. W. STUBER, 1984  Isoenzymatic variation in Zea (Gramineae). Syst. Bot. 9:203-218.

DOEBLEY, J. F., W. RENFROE, and A. BLANTON, 1987  Restriction site variation in the Zea chloroplast genome. Genetics 117:139-147[Abstract/Free Full Text].

EYRE-WALKER, A., T. L. GAUT, H. HILTON, D. L. FELDMAN, and B. S. GAUT, 1998  Investigation of the bottleneck leading to the domestication of maize. Proc. Natl. Acad. Sci. USA 95:4441-4446[Abstract/Free Full Text].

FU, Y. and W. LI, 1993  Statistical tests of neutrality of mutations. Genetics 133:693-709[Abstract].

GALITSKI, T., A. G. SALDANHA, C. A. STYPES, E. S. LANDER, and G. R. FINK, 1999  Ploidy regulation of gene expression. Science 285:251-253[Abstract/Free Full Text].

GAUT, B. S. and M. T. CLEGG, 1991  Molecular evolution of alcohol dehydrogenase 1 in members of the grass family. Proc. Natl. Acad. Sci. USA 88:2060-2064[Abstract/Free Full Text].

GAUT, B. S. and M. T. CLEGG, 1993  Molecular evolution of the Adh1 locus in the genus Zea. Proc. Natl. Acad. Sci. USA 90:5095-5099[Abstract/Free Full Text].

GRANT, V., 1981 Plant Speciation, Ed. 2. Columbia University Press, New York.

HALDANE, J. B. S., 1930  Theoretical genetics of autopolyploids. J. Genet. 22:359-372.

HANSON, M. A., B. S. GAUT, A. O. STEC, S. I. FUERSTENBERG, and M. M. GOODMAN et al., 1996  Evolution of anthocyanin biosynthesis in maize kernels: the role of regulatory and enzymatic loci. Genetics 143:1395-1407[Abstract].

HARTL, D. L., and A. G. CLARK, 1997 Principles of Population Genetics. Sinauer Associates, Sunderland, MA.

HILL, R. R., 1971  Selection in autotetraploids. Theor. Appl. Genet. 41:181-186.

HILTON, H. and B. S. GAUT, 1998  Speciation and domestication in maize and its wild relative: evidence from the Globultin-1 gene. Genetics 150:863-872[Abstract/Free Full Text].

HUDSON, R. R., 1987  Estimating the recombination parameter of a finite population model without selection. Genet. Res. 50:245-250[Medline].

HUDSON, R. R., M. KREITMAN, and M. AGUADE, 1987  A test of neutral molecular evolution based on nucleotide data. Genetics 116:153-159[Abstract/Free Full Text].

ILTIS, H. H., J. F. DOEBLEY, R. GUZMAN, and B. PAZY, 1979  Zea diploperennis (Gramineae): a new teosinte from Mexico. Science 203:186-188[Abstract/Free Full Text].

KATO, Y. and R. LOPEZ, 1990  Chromosome knobs of the perennial teosintes. Maydica 35:125-141.

LEWIS, W. H., 1980 Polyploidy in species populations, pp. 103–144 in Polyploidy, edited by W. H. LEWIS. Plenum Press, New York.

MASON-GAMER, R., C. WEIL, and E. KELLOGG, 1998  Granule-bound starch synthase: structure, function, and phylogenetic utility. Mol. Biol. Evol. 15:1658-1673[Abstract].

MASTERSON, J., 1994  Stomatal size in fossil plants: evidence for polyploidy in majority of angiosperms. Science 264:421-423[Abstract/Free Full Text].

MCDONALD, J. H. and M. KREITMAN, 1991  Adaptive protein evolution at the Adh1 locus in Drosophila. Nature 351:652-654[Medline].

MOODY, M. E., L. D. MUELLER, and D. E. SOLTIS, 1993  Genetic variation and random drift in autotetraploid populations. Genetics 134:649-657[Abstract].

NEI, M., 1987 Molecular Evolutionary Genetics. Columbia University Press, New York.

OKANE, S. L., B. A. SCHAAL, and I. A. ALSHEHBAZ, 1996  The origins of Arabidopsis suecica (Brassicaceae) as indicated by nuclear rDNA sequences. Syst. Bot. 21:559-566.

OTTO, S. P. and J. WHITTON, 2000  Polyploid incidence and evolution. Annu. Rev. Genet. 34:401-437[Medline].

POSPUPULETI, C. V. and W. C. GALINAT, 1982  Zea diploperennis. I. Its chromosomes and comparative cytology. J. Heredity 73:168-170[Abstract/Free Full Text].

POSTLETHWAIT, J. H., Y. L. YAN, M. A. GATES, S. HORNE, and A. AMORES et al., 1998  Vertebrate genome evolution and the zebrafish gene map. Nat. Genet. 18:345-359[Medline].

PURUGGANAN, M. D. and S. R. WESSLER, 1994  Molecular evolution of magellan, a maize Ty3/gypsy like retrotransposon. Proc. Natl. Acad. Sci. USA 91:11674-11678[Abstract/Free Full Text].

RANDOLPH, L. F., 1955 History and origin of corn. II. Cytogenetic aspects of the origin and history of corn: a monograph, pp. 26–61 in Corn and Corn Improvement, edited by G. F. Sprague. Academic Press, New York.

ROZAS, J. and R. ROZAS, 1999  DnaSP version 3: an integrated program for molecular population genetics and molecular evolution analysis. Bioinformatics 15:174-175[Abstract/Free Full Text].

SAITOU, N. and M. NEI, 1987  The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425[Abstract].

SAWYER, S. A., D. E. DYKHUIZEN, and D. L. HARTL, 1987  Confidence interval for the number of selectively neutral amino acid polymorphisms. Proc. Natl. Acad. Sci. USA 84:6225-6228[Abstract/Free Full Text].

SEOIGHE, C. and K. H. WOLFE, 1998  Extent of genomic rearrangement after genome duplication in yeast. Proc. Natl. Acad. Sci. USA 95:4447-4452[Abstract/Free Full Text].

SHAVER, D. L., 1962  A study of meiosis in perennial teosinte, in tetraploid maize and in their tetraploid hybrid. Caryologia 15:43-57.

SMALL, R. L. and J. F. WENDEL, 2000  Copy number lability and evolutionary dynamics of the Adh gene family in diploid and tetraploid cotton (Gossypium). Genetics 155:1913-1926[Abstract/Free Full Text].

SMALL, R. L., J. A. RYBURN, R. C. CRONN, T. SEELANAN, and J. F. WENDEL, 1998  The tortoise and the hare: choosing between noncoding plastome and nuclear ADH sequences for phylogeny reconstruction in a recently diverged plant group. Am. J. Bot. 85:1301-1315[Abstract/Free Full Text].

SOLTIS, D. E. and P. S. SOLTIS, 1993  Molecular data and the dynamic nature of polyploidy. Crit. Rev. Plant Sci. 12:243-273.

SONG, K., P. LU, K. TANG, and T. C. OSBORN, 1995  Rapid genome change in synthetic polyploids of Brassica and its implication for polyploid evolution. Proc. Natl. Acad. Sci. USA 92:7719-7723[Abstract/Free Full Text].

STEBBINS, G. L., 1947  Types of polyploids: their classification and significance. Adv. Genet. 1:403.

STEBBINS, G. L., 1950 Variation and Evolution in Plants. Columbia University Press, New York.

STEBBINS, G. L., 1980 Polyploidy in plants: unsolved problems and prospects, pp. 495–520 in Polyploidy, edited by W. H. LEWIS. Plenum Press, New York.

SWOFFORD, D. L., 1990 PAUP: Phylogenetic Analysis Using Parsimony, Version 3.0. Computer program distributed by the Illinois Natural History Survey, Champaign, IL.

TAJIMA, F., 1983  Evolutionary relationship of DNA sequences in finite populations. Genetics 105:437-460[Abstract/Free Full Text].

TAJIMA, F., 1989  Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics 123:585-595[Abstract/Free Full Text].

WATTERSON, G. A., 1975  On the number of segregating sites in genetical models without recombination. Theor. Popul. Biol. 7:188-193.

WENDEL, J. F., 2000  Genome evolution in polyploids. Plant Mol. Biol. 42:225-249[Medline].

WOLFE, K. H., W.-H. LI, and P. M. SHARP, 1987  Rates of nucleotide substitution vary greatly among plant mitochondrial, chloroplast and nuclear DNAs. Proc. Natl. Acad. Sci. USA 84:9054-9058[Abstract/Free Full Text].

WRIGHT, S., 1938  The distribution of gene frequencies in populations of polyploids. Proc. Natl. Acad. Sci. USA 24:372-377[Free Full Text].

ZIMMER, E. A., E. R. JUPE, and V. WALBOT, 1988  Ribosomal gene structure, variation and inheritance in maize and its ancestors. Genetics 120:1125-1136[Abstract/Free Full Text].

This article has been cited by other articles:

				J. Ross-Ibarra, M. Tenaillon, and B. S. Gaut Historical Divergence and Gene Flow in the Genus Zea Genetics, April 1, 2009; 181(4): 1399 - 1413. [Abstract] [Full Text] [PDF]

				H. Chen, P. L. Morrell, M. de la Cruz, and M. T. Clegg Nucleotide Diversity and Linkage Disequilibrium in Wild Avocado (Persea americana Mill.) J. Hered., March 14, 2008; (2008) esn016v1. [Abstract] [Full Text] [PDF]

				U. Arunyawat, W. Stephan, and T. Stadler Using Multilocus Sequence Data to Assess Population Structure, Natural Selection, and Linkage Disequilibrium in Wild Tomatoes Mol. Biol. Evol., October 1, 2007; 24(10): 2310 - 2322. [Abstract] [Full Text] [PDF]

				L.-B. Zhang and S. Ge Multilocus Analysis of Nucleotide Variation and Speciation in Oryza officinalis and Its Close Relatives Mol. Biol. Evol., March 1, 2007; 24(3): 769 - 783. [Abstract] [Full Text] [PDF]

				Q. Zhu, X. Zheng, J. Luo, B. S. Gaut, and S. Ge Multilocus Analysis of Nucleotide Variation of Oryza sativa and Its Wild Relatives: Severe Bottleneck during Domestication of Rice Mol. Biol. Evol., March 1, 2007; 24(3): 875 - 888. [Abstract] [Full Text] [PDF]

				M. Heuertz, E. De Paoli, T. Kallman, H. Larsson, I. Jurman, M. Morgante, M. Lascoux, and N. Gyllenstrand Multilocus Patterns of Nucleotide Diversity, Linkage Disequilibrium and Demographic History of Norway Spruce [Picea abies (L.) Karst] Genetics, December 1, 2006; 174(4): 2095 - 2105. [Abstract] [Full Text] [PDF]

				A. Liu and J. M. Burke Patterns of Nucleotide Diversity in Wild and Cultivated Sunflower Genetics, May 1, 2006; 173(1): 321 - 330. [Abstract] [Full Text] [PDF]

				D. A. Moeller and P. Tiffin Genetic Diversity and the Evolutionary History of Plant Immunity Genes in Two Species of Zea Mol. Biol. Evol., December 1, 2005; 22(12): 2480 - 2490. [Abstract] [Full Text] [PDF]

				K. Bomblies and J. F. Doebley Molecular Evolution of FLORICAULA/LEAFY Orthologs in the Andropogoneae (Poaceae) Mol. Biol. Evol., April 1, 2005; 22(4): 1082 - 1094. [Abstract] [Full Text] [PDF]

				P. K. Ingvarsson Nucleotide Polymorphism and Linkage Disequilibrium Within and Among Natural Populations of European Aspen (Populus tremula L., Salicaceae) Genetics, February 1, 2005; 169(2): 945 - 953. [Abstract] [Full Text] [PDF]

				P. Tiffin, R. Hacker, and B. S. Gaut Population Genetic Evidence for Rapid Changes in Intraspecific Diversity and Allelic Cycling of a Specialist Defense Gene in Zea Genetics, September 1, 2004; 168(1): 425 - 434. [Abstract] [Full Text] [PDF]

				P. Tiffin Comparative Evolutionary Histories of Chitinase Genes in the Genus Zea and Family Poaceae Genetics, July 1, 2004; 167(3): 1331 - 1340. [Abstract] [Full Text] [PDF]

				K. S. Caldwell, J. Dvorak, E. S. Lagudah, E. Akhunov, M.-C. Luo, P. Wolters, and W. Powell Sequence Polymorphism in Polyploid Wheat and Their D-Genome Diploid Ancestor Genetics, June 1, 2004; 167(2): 941 - 947. [Abstract] [Full Text] [PDF]

				R. M. Clark, E. Linton, J. Messing, and J. F. Doebley Inaugural Article: Pattern of diversity in the genomic region near the maize domestication gene tb1 PNAS, January 20, 2004; 101(3): 700 - 707. [Abstract] [Full Text] [PDF]

				S. E. Ramos-Onsins, B. E. Stranger, T. Mitchell-Olds, and M. Aguade Multilocus Analysis of Variation and Speciation in the Closely Related Species Arabidopsis halleri and A. lyrata Genetics, January 1, 2004; 166(1): 373 - 388. [Abstract] [Full Text] [PDF]

				P. Tiffin and B. S. Gaut Molecular Evolution of the Wound-Induced Serine Protease Inhibitor wip1 in Zea and Related Genera Mol. Biol. Evol., November 1, 2001; 18(11): 2092 - 2101. [Abstract] [Full Text] [PDF]