| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Genetics, Vol. 173, 1007-1021, June 2006, Copyright © 2006
doi:10.1534/genetics.105.053165
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Division of Biological Sciences, University of Missouri, Columbia, Missouri 65211-7400
1 Corresponding author: University of Missouri, 117 Tucker Hall, Columbia, MO 65211-7400.
E-mail: birchlerj{at}missouri.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
-ZeinA) could be simultaneously visualized with the retroelement probes, emphasizing the value of the retroelement probes for cytogenetic studies of Zea and Tripsacum.
Except for one autotetraploid species, Zea species are diploids with a basal set of 10 chromosomes derived from an ancient polyploidization event. Tripsacum species represent a range of ploidies with total chromosome numbers in multiples of 18. The basal set of 18 chromosomes likely represents a recent polyploidization event (GAUT et al. 2000). These genera constitute an excellent model of the types of changes that occur during evolutionary divergence including chromosomal rearrangements, polyploidization events, and divergence of repetitive elements such as retrotransposons (GAUT et al. 2000). A comparison of the retroelement distribution in these related species would increase understanding of the mechanisms that shape plant evolution and the processes of retrotransposon expansion.
In addition to providing an evolutionary model of genome evolution, Zea and Tripsacum have agronomic value. Tripsacum dactyloides is a forage crop and can be hybridized with maize to produce partially fertile offspring (MANGELSDORF and REEVES 1931). Thus, Tripsacum might serve as a source of germplasm for maize crop improvement and vice versa. Understanding how these genera are related would facilitate such reciprocal utilization of tools and traits. Cytological approaches could play a role in determining their relationship by visualizing the changes in genomic elements and their physical arrangement among the Zea and Tripsacum species and allow visualization of gene flow resulting from breeding programs.
Improvements in chromosome spreading procedures, development of a collection of DNA elements that can be used as fluorescence in situ hybridization (FISH) probes to identify each chromosome (KATO et al. 2004), and detection of specific genetic loci on chromosomes (KOUMBARIS and BASS 2003; ANDERSON et al. 2004; KATO et al. 2006) have expanded the number of cytological tools available to maize researchers.
This report describes the application of karyotypic methods for the study of the evolution of abundant elements in the Zea and Tripsacum genomes, the extension of maize tools to Tripsacum, and the development of additional techniques to create a tool kit for use in subsequent studies of these species. First, specific sequences from retrotransposons, including previously characterized maize retroelements and Tripsacum retroelement sequences isolated in this study, were used as FISH probes onto chromosomes from a "trispecies" hybrid containing chromosomes from Zea mays, Z. diploperennis, and T. dactyloides to examine the distribution of retrotransposons among Zea and Tripsacum. The use of the trispecies hybrid allowed the relative intensities of hybridization of the retroelements to different genomes to be determined. Because many of the retroelement probes hybridized more intensely to one genome than to another, they were used as genome paints and their effectiveness was compared to genomic in situ hybridization (GISH). Examination of GISH images revealed variation of centromeric elements and heterochromatic knob elements among Zea and Tripsacum.
Second, the retroelement genome paints were demonstrated to be a useful addition to a cytological tool kit for Zea and Tripsacum. Several FISH probes known to label maize chromosomes were used in combination with the retroelement FISH probes to simultaneously distinguish different genomes and detect the chromosomal location of specific DNA elements in interspecific hybrids. Additionally, retroelement genome paints were used to demonstrate introgression of Tripsacum chromatin into maize lines. Applications of these techniques to basic questions of genome evolution and organization as well as practical uses in breeding programs are discussed.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Maize x Tripsacum hybrids were produced using tetraploid T. dactyloides pollen for silks of a maize popcorn variety (Super Gold). Maize x Z. diploperennis hybrids (maize variety NC300 x Z. diploperennis, PI 441932) were provided as kernels by Jim Holland (North Carolina State University). T. andersonii was provided by Denise Costich (Cornell University).
The Maize Genetics Cooperation Stock Center (MGCSC) provided four lines derived from addition lines produced by Walton Galinat. Two did not have any detectable alien chromatin (stock center field nos. 2000-9-1@ and 2000-8B-2@) and two did (described in this report, stock center field nos. 2000-7-5@ and 2000-W23-24/8A-2). The 2-T, T-2 translocation line produced by Marjorie Maguire was also obtained from the MGCSC (stock record 220B).
Genomic DNA for library construction and Southern blotting was collected from leaf tissue. Maize genomic DNA was from the inbred line, W22. Genomic DNA from T. dactyloides was collected from variety "Pete." Leaf tissue from T. laxum and Z. diploperennis for genomic DNA isolation was provided by Mary Eubanks. Z. luxurians was grown from kernels provided by Mary Eubanks (PI#306615). Sorghum bicolor seed (variety Sudan) was grown for leaf tissue.
Genomic library production and screening:
A Fosmid library made from T. dactyloides genomic DNA was constructed using a production kit from Epicentre according to the manufacturer's directions [catalog (cat.) no. CCFOS110]. Ninety-six colonies were hand picked and used to inoculate a 96-well square bottom plate. Plasmids were isolated using a 96-well plasmid isolation kit.
Small-insert libraries were made from nebulized genomic DNA from T. dactyloides and Z. diploperennis. The DNA was nebulized according to the manufacturer's directions (cat. no. 45-0072; Invitrogen, San Diego). Nebulized DNA was subjected to electrophoresis and DNA that was 1200–3000 bp in length was extracted. The extracted DNA was treated with Epicentre's (Madison, WI) End-It enzyme mixture (cat. no. ER0720) to produce blunt ends. The blunt end DNA was then incubated with Taq polymerase and dATP to add an "A" overhang and then cloned into Promega's (Madison, WI) pGemT TA-cloning vector (cat. no. A3600). Ninety-six colonies were grown and plasmids isolated using a 96-well plasmid isolation kit.
Dot blots were prepared using the Bio-Dot apparatus (cat. no. M1706545; Bio-Rad, Hercules, CA) according to the manufacturer's instructions. Maize Cot-100 DNA was prepared as described (ZWICK et al. 1997). Autoclaved genomic DNA (maize, T. dactyloides, or Z. diploperennis) and maize Cot-100 DNA were labeled with 32P-dCTP using a random decamer priming reaction (DECAprime II, cat. no. 1456; Ambion, Austin, TX) and hybridized overnight to the dot blots in hybridization solution (cat. no. H-7033; Sigma, St. Louis) at 65°. Following a quick rinse in 0.2x SSC, 0.1x SDS, the blots were washed two times at 65° for 30 min each. Blots were then labeled with a refillable ink pen spiked with 32P-dCTP and exposed to a phosphor-imaging plate for 
6 hr. Phosphor-imaging plates were analyzed using a FUJIFILM fluorescent image analyzer (FLA-2000), FUJI Image Reader V1.5E, and processed using FUJI Image Gauge V3.45.
Tripsacum genomic clones TC#5 and TC#25 were labeled with 32P-dCTP and probed onto a Southern blot prepared with genomic DNA from S. bicolor, T. dactyloides, T. laxum, Z. mays, Z. diploperennis, and Z. luxurians cleaved with HindIII. Hybridization was at 65° overnight followed by two washes 30 min each with 0.2x SSC, 0.1% SDS at 65° and visualization by exposure to a phosphor-imaging plate for 4–6 hr.
Subclones from fosmids that hybridized in a Tripsacum-specific pattern were produced by digestion of fosmid DNA with HindIII or SacI and ligated into a pBluescript vector that had been digested with the appropriate restriction enzyme. Plasmids were isolated from several colonies and used as templates for a PCR reaction using T3 and T7 primers. PCR products were used as templates for producing fluorescent probes, which were tested by FISH for specific hybridization to Tripsacum.
All plasmids from the small-insert library and fosmid subclones showing a Tripsacum-specific hybridization pattern were sequenced using M13F, M13R, T3, or T7 primers. Fosmids were end sequenced using the T7 primer. Sequences were compared against data in GenBank using the BlastN and/or BlastX algorithm (ALTSCHUL et al. 1990). The sequences of the elements that are enriched in Tripsacum chromatin have been deposited in GenBank (TF-B5-2, DQ223961; TF-B5-3, DQ223960; TF-B8-15, DQ223962; TC#5, DQ223963; TC#12, DQ223964; TC#24, DQ223965; and TC#25, DQ223966).
Chromosome spread preparation and FISH:
Chromosome preparations were prepared as described previously (KATO et al. 2004) with modifications described below. Root tips from maize x Tripsacum hybrids, maize x Z. diploperennis, and the trispecies hybrid were collected from adult plants by choosing young rapidly growing root tips. Root tips with many cells in mitosis were obtained from plants that had recently been propagated (1–3 weeks) or from older plants whose roots were trimmed, watered, and allowed to regrow for 3–6 days without watering. Root tips from other material were collected from seeds that had been germinated in moist vermiculite for 3 days at 30°.
The maize retroelement probes Grande, Huck, Prem2/Ji, Tekay, Cinful, and Opie were cloned as described (MROCZEK and DAWE 2003) except that a Promega pGemT TA-cloning kit was used. After sequencing to confirm the identity of the inserts, plasmids were used as templates for PCR reactions to produce DNA for fluorescent labeling. Prem1 was amplified from a clone provided by Jeffrey Bennetzen. DNA from the Tripsacum small-insert library was PCR amplified from the respective plasmid using M13F and M13R (or T3 and T7) primers as were the 180-bp knob repeat, 5S gene cluster, TR-1, NOR, Cent4, CentC, rp1, rp3, and the -zeinA cluster inserts from plasmids described elsewhere (KATO et al. 2004, 2006). Fosmid DNA was prepared using a QIAGEN (Valencia, CA) miniprep kit after induction to high copy number using the induction solution provided in the fosmid production kit from Epicentre (cat. no. CCFOS110). Fluorescent labeling was as described previously (KATO et al. 2006).
FISH was performed as described previously (KATO et al. 2004) except that the slides were not fixed with formaldehyde after UV-crosslinking. For GISH, the following amounts of probe were included in a total volume of 7 µl: FITC-labeled genomic DNA from Tripsacum, 50 ng; autoclaved unlabeled maize DNA, 2.5 µg (50x) or 5 µg (100x); Grande, 100 ng; Huck, 100 ng; Prem1, 100 ng; and 180-bp knob repeat, 40 ng. For other FISH experiments, 40–200 ng of each probe were used per slide in a total volume of 5 µl per slide.
Images were taken either by an Optronics MagnaFIRE CCD camera mounted on a Zeiss Universal microscope using a 100x plan apo oil immersion lens (for most images) or by a Sensys CCD camera (for the rp1, rp3, and -zeinA pictures) using a 60x plan apo oil immersion lens on an Olympus IX70 inverted microscope controlled by Metamorph software. Images were processed as described previously (KATO et al. 2004; LAMB et al. 2005) with the following addition. For some images where light intensity was not uniform, the image was severely blurred using the "Gaussian blur" function of Photoshop 7.0 and the blurred image was subtracted from the original to reduce the background.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
|
In the trispecies hybrid, several of the retroelement probes labeled nine chromosomes most intensely. To determine which of the Zea genomes is more strongly labeled and to confirm the pattern seen in the trispecies hybrid, these elements were also probed onto chromosomes from a Z. mays x Z. diploperennis F1 hybrid. Small heterochromatic blocks, called knobs, are present near the termini of chromosomes in Z. diploperennis whereas they tend to be larger and positioned interstitially in chromosomes of maize (MCCLINTOCK et al. 1981; KATO 1984). The locations of the knobs in the Z. mays x Z. diploperennis hybrid were determined by hybridization of the 180-bp knob repeat together with simultaneous hybridization of the Huck and Opie retroelements, each probe labeled with a different fluorescent molecule. The chromosomes with large internal knobs hybridized intensely with the Huck and Opie probes (Figure 1A). Therefore, the 10 weakly labeled chromosomes were determined to be from Z. diploperennis.
T. andersonii is thought to be an allopolyploid that combines the genomes of a Tripsacum species and Z. luxurians (TALBERT et al. 1990). Consistent with this hypothesis, chromosome spreads of T. andersonii contained 10 chromosomes that were strongly labeled by the maize retroelement probes except for Tekay, which hybridized to all the chromosomes. The remaining chromosomes were labeled by the Tripsacum-specific clones described below (supplemental Figure 2 at http://www.genetics.org/supplemental/).
Identification of Tripsacum clones (fosmids and small inserts) that paint Tripsacum chromosomes:
The distribution of maize retroelements in the trispecies hybrid indicates that many of the maize retroelement families have expanded after the divergence of the Zea and Tripsacum lineages. However, the Tripsacum genome size is comparable to that of maize (GAUT et al. 2000), suggesting that other retroelement families have expanded in the Tripsacum genome. This hypothesis would be supported by the discovery of retroelements that paint Tripsacum chromosomes but not Zea in the trispecies hybrid. Additionally, Tripsacum-specific retroelement probes would complement the maize retroelements as useful tools to distinguish the chromatin of different species in interspecific hybrids.
To obtain probes that would label Tripsacum chromatin, small libraries of Tripsacum genomic DNA were screened for clones, which could be used as FISH probes to distinguish Tripsacum chromosomes in a similar fashion as the maize retroelements. A fosmid library was generated and DNA from 96 clones was spotted in triplicate onto nylon membranes. The membranes were then probed with radiolabeled maize DNA, Cot-100 maize DNA, or genomic DNA from T. dactyloides. The dot blots were then examined for clones that were labeled by the Tripsacum genomic DNA but not by maize DNA. This approach allowed many clones to be screened at one time.
Three fosmids, TFF-B5, TF-B8, and TF-F7, that were labeled on the dot blot more intensely by Tripsacum than by maize were used as red FISH probes simultaneously with the Huck probe labeled in green on a trispecies hybrid chromosome spread. Two of these fosmids, TF-B5 and TF-B8, hybridized intensely to Tripsacum but not to Zea chromosomes. TF-F7 hybridized to the 5S ribosomal gene clusters (Figure 2 and supplemental Figure 3 at http://www.genetics.org/supplemental/). A number of clones, TF-A4, TF-C12, TF-G6, and TF-H7, that had little or no signal on the dot blot when hybridized with either the maize or the Tripsacum radiolabeled DNA were tested as FISH probes onto the trispecies hybrid. All of these probes gave weaker FISH signals than the fosmids that had strong dot blot signals and required longer exposure times to clearly visualize the fluorescent signals. Two fosmids, TF-C12 and TF-G6, hybridized more intensely to the Tripsacum chromosomes, giving a signal that allowed the Tripsacum chromosomes to be clearly distinguished from the Zea ones. Fosmid clone TF-H7 specifically labeled the Tripsacum chromosomes but it hybridized in a punctate pattern instead of the more uniform hybridization seen for retroelement probes. Fosmid TF-A4 labeled the Tripsacum chromosomes slightly more intensely than the Zea ones. Several of the dot blot signals resulting from hybridization with maize genomic DNA were stronger than those with Tripsacum DNA. When fosmids of this category, TF-B6 and TF-E12, were used as FISH probes, they strongly labeled the Zea chromosomes (Figure 2 and supplemental Figure 3).
|
89% identity). These 46 bp are contained in a 130-bp stretch that is 83% identical to a Setaria italice transposable element. In addition to the region shared between the subclones, TF-B8-15 contained a second region homologous to a maize retroelement (>86% identity for 1438 bp to gi:46200524) and a region with homology to multiple rice genomic regions.
Additional Tripsacum-specific clones were obtained from a small-insert library made from genomic DNA sheared to 1.5 kb. Clones were dotted to duplicate nylon membranes and probed with 32P-radiolabeled genomic DNA. Twenty-three clones that were strongly labeled on the dot blot (supplemental Figure 4 at http://www.genetics.org/supplemental/) were amplified via PCR to generate template DNA for fluorescent labeling. Many of the most intensely hybridizing spots produced multiple bands in a laddered pattern when subjected to gel electrophoresis, suggesting that they contained ribosomal DNA, knob sequences, or other tandemly arranged elements (data not shown). These products were not used and the remaining PCR products were labeled and screened by hybridization to trispecies hybrid chromosome spreads, allowing identification of four additional clones (TC#5, -12, -24, and -25) that hybridized strongly to Tripsacum chromosomes and weakly to moderately to Zea chromosomes. The remaining clones hybridized with slightly greater intensity to Tripsacum chromosomes than to Zea (TC#3, -13, and -17), to specific chromosomal regions such as the NOR (TC#11 and -19) or knobs (TC#4, -7, and -14), or did not hybridize at all (Figure 3; supplemental Figure 5 at http://www.genetics.org/supplemental/).
|
The clones that strongly labeled the Tripsacum chromosomes were sequenced and the results compared against the GenBank database using BlastN and/or BlastX. All of the clones were similar to many maize genomic sequences, including sequences annotated as retrotransposons. TC#24 and TC#25 shared several blocks of highly similar sequence (92% identity for 165 bp, 81% for 215 bp, 79% for 126 bp, 96% for 30bp, and 100% for 23 bp). The sequences identified in this study that are highly abundant in Tripsacum appear to represent four different retroelements [(1) TF-B5-3 and TF-B8-15, (2) TC#24 and TC#25, (3) TC#5, and (4) TC#12].
A small-insert library was also prepared from Z. diploperennis and dotted in the same manner as the Tripsacum library. The membranes were hybridized to maize genomic DNA, stripped, and hybridized to DNA from Z. diploperennis. Two clones were selected, one that hybridized more intensely to maize DNA (ZD#9) and one that hybridized equally (ZD#10). These clones were hybridized to the trispecies hybrid chromosome spreads. They labeled the chromosomes in a pattern consistent with the dot blot hybridization; the Tripsacum chromosomes were only weakly labeled (supplemental Figure 5 at http://www.genetics.org/supplemental/). The clones were sequenced and both were found to be homologous to many maize sequences. ZD#9 was homologous to several sequences annotated as Huck elements (e.g., 94% identity to gi:67043718). ZD#10 was homologous to intergenic sequences from a variety of clones including centromeric BAC 15C5 (gi:37514986) (NAGAKI et al. 2003). The regions of homology to BAC 15C5 included shadowspawn retroelement fragments (bp 43,649–43,409, 86% identity; 54,757–54,517, 86% identity; and 61,400–61,082, 89% identity) and a region putatively identified as degenerate retroelement sequence (bp 5587–5887, 83% identity and 7071–7404, 87% identity). The latter region has previously been subcloned and used as a FISH probe on maize chromosomes giving a dispersed hybridization pattern similar to that of ZD#10 (NAGAKI et al. 2003).
In summary, many retroelement DNA sequences were assayed for their relative intensity of hybridization as FISH probes onto the trispecies hybrid. Sequences generally hybridized to the chromosomes in the following patterns: (1) strong on all three species; (2) strong and equal on Z. mays and Z. diploperennis but weak or not at all on Tripsacum; (3) strong on Z. mays, intermediate on Z. diploperennis, weak on Tripsacum; and (4) weak or intermediate on both Zea species and strong on Tripsacum. However, there were subtle differences in hybridization within each of these four general categories. This information is summarized in Table 1.
|
Because knobs from all species were labeled using GISH, using this technique to detect interspecific rearrangements could be complicated. Retroelement painting was therefore attempted in conjunction with GISH to increase the ability to discriminate different genomes. GISH was performed by allowing a mixture of Tripsacum genomic DNA labeled in green; an excess of the maize retroelements Grande, Huck, and Prem1 labeled in red; and the 180-bp knob repeat labeled in red to hybridize to chromosome spreads in the presence of unlabeled maize DNA (Figure 4). This procedure painted the Zea chromosomes in one color, red, and the Tripsacum chromosomes in another, green, while the knobs are labeled with both colors.
|
To determine which centromeric components were responsible for the Tripsacum GISH signal, FISH with known centromeric components was performed. When trispecies hybrid chromosomes were probed with the maize tandemly repeated centromere element CentC, all centromeres were labeled (Figure 4), although some of the maize centromeres had weaker signals than others, consistent with previous reports (ANANIEV et al. 1998; KATO et al. 2004). A full-length clone of a centromere-specific retrotransposon, CRM, hybridized most intensely to maize, second most to Z. diploperennis, and least to the Trispacum centromeres (Figure 4). To determine whether divergence of the CRM element in Tripsacum explained the centromere-specific GISH signal, primers that amplify several regions of the CRM element (including the LTR, gag/pol genes, and reverse transcriptase) were used to obtain DNA for generating FISH probes from Tripsacum genomic DNA template, using a low annealing temperature. The hybridization pattern using those probes was similar to that observed for the complete labeled maize element (data not shown). This control indicates that the difference in hybridization intensity between the three species is likely due to copy number expansion of CRM in the maize lineage.
Because specific CRM variants are known to be present in different copy numbers among maize centromeres, chromosomes from a hybrid of maize x Z. diploperennis were probed with the LTRs of two different CRM variants, CRM1a and CRM2a, to determine if both variants were more abundant in maize than in Z. diploperennis (supplemental Figure 6 at http://www.genetics.org/supplemental/). CRM2a hybridized more intensely to maize chromosomes than to those of Z. diploperennis. CRM1a also labeled the maize centromeres with greater intensity than those of Z. diploperennis, but the difference was less pronounced than that of CRM2a. CRM was also tested on a maize x T. dactyloides F1 hybrid and in T. andersonii. In both species, the intensity of Tripsacum centromere labeling was much less than that of the Zea centromeres (supplemental Figure 1 at http://www.genetics.org/supplemental/ and data not shown).
Use of retroelement paints and GISH to identify introgressed Tripsacum chromatin in maize lines:
Two of the Tripsacum-specific retroelements, subclone TF-B5-3 and TC#25 (red) and two maize retroelements, Prem2 and Huck (green), were applied to material obtained from the Maize Genetics Cooperation Stock Center that was thought to contain Tripsacum chromatin in a maize background. This collection consisted of four lines descended from maize–Tripsacum addition lines produced by Walton Galinat (GALINAT et al. 1963; GALINAT and MANGELSDORF 1966) and one line identified by Marjorie Maguire containing a translocation between maize chromosome 2 and a Tripsacum chromosome (MAGUIRE 1957; MAGUIRE 1960a,b). GISH was also performed on these lines using labeled Tripsacum genomic DNA (green), several maize retroelements (red), and the 180-bp knob repeat (red). This experiment allowed the utility of the newly cloned Tripsacum elements in identifying introgressed Tripsacum chromatin to be tested and compared to a GISH approach (Figure 5).
|
When plants containing Tripsacum chromatin were selfed, individuals that were homozygous for the translocation chromosomes were recovered and grown to maturity. The homozygous replacement line derived from the Maguire material was male sterile as previously reported (MAGUIRE 1957, 1960a,b), but the two homozygous replacement lines from the Galinat material were both male and female fertile.
Additional probes:
A collection of repetitive sequences has been applied to maize chromosome spreads as FISH probes, allowing the somatic karyotype to be determined (KATO et al. 2004). Probes from this collection and other probes that hybridize to single loci were applied to spreads containing both Zea and Tripsacum chromosomes to determine if maize probes can be universally applied to these closely related species and if they would be valuable in identifying specific Tripsacum chromosomes. The Zea chromosomes could be identified in these spreads because the probes were applied simultaneously with labeled retroelements to paint the maize and Tripsacum genomes differentially. The Zea chromosomes serve as a positive control because the probes have been applied previously to maize. Repetitive elements tested include a knob repeat (TR-1), a chromosome 4-specific probe (Cent4), the 5S ribosomal DNA, the 17S ribosomal RNA (the NOR), and a microsatellite repeat (TAGn) (Figure 6). Probes that label specific loci included the rp1, rp3, and -zeinA gene clusters (Figure 7). All probes tested gave signals on the Z. diploperennis and T. dactyloides chromosomes in the trispecies hybrid. No maize chromosomes were labeled by the 5S ribosomal RNA probed so the missing maize chromosome was identified as chromosome 2.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
11.9 million years ago (SWIGONOVA et al. 2004). Subsequently, the two diverged maize genomes were united through a polyploidization event 4.8 million years ago followed by extensive rearrangements, gene loss, and divergence and other modifications leading to the current diploidized maize genome (SWIGONOVA et al. 2004). Comparing the amount of DNA between maize and sorghum shows that the maize genome is more than twice as large as that of sorghum and the difference is thought to result from expansion of LTR-type retroelement families (SANMIGUEL and BENNETZEN 1997; GAUT et al. 2000). When the adh1-F genomic region of maize was examined, it was found to contain retroelements inserted into other retroelements in a nested fashion that accounted for >60% of the total sequence (SANMIGUEL et al. 1996). Because the process of transposition ensures that both LTRs are identical upon insertion into the genome, sequence divergence of the LTRs provides estimations of the timeframe of transposition. Using LTRs as a molecular clock, the approximate ages of several nested retroelements were determined in this region. Insertion dates for the various retroelements ranged from 5 MYA to near the present, including many elements, such as Ji and Huck, that show continual insertion over this period (SANMIGUEL et al. 1998).
Comparison of a number of large genomic regions revealed that the regions between genes are very different for different inbred lines due largely to the presence or absence of various retroelements (BRUNNER et al. 2005). The retroelements found to be present at genomic locations from only one inbred tend to be more recently inserted than shared elements (nonhomologous elements, mean 0.91 MYA, median 0.61; shared retroelements, mean 1.34 MYA, median 1.16) while many older elements, common or not, were truncated in some fashion (BRUNNER et al. 2005). Transposition is a continual process that can be documented to 5 million years ago in maize with several different families actively expanding throughout that entire time. Unless selection pressures act to maintain an element's integrity, mutations and sequence elimination deteriorate elements after they have inserted into a genome. After enough changes have accumulated, ancient insertions become unrecognizable. In rice, the half-life for a retrotransposon has been estimated to be <6 million years (MA et al. 2004).
The four retroelement sequences identified in this study that are enriched in Tripsacum [(1) TF-B5-3 and TF-B8-15, (2) TC#24 and TC#25, (3) TC#5, and (4) TC#12] have highly similar homologs in maize. Hybridization to Southern blots of several related species demonstrated that TC#5 and TC#25 are abundant in two different Tripsacum species, are present in the Zea species, and are absent or highly diverged in Sorghum (Figure 3). The two sequences isolated from Z. diploperennis and several of the clones containing genomic fragments from Tripsacum hybridized with equal or greater intensity to maize than to the genome from which they were isolated. Thus, elements expanded in the maize lineage have closely related homologs in the Tripsacum genome and vice versa. The reason why some retroelements have expanded in one lineage and remained dormant in others is unknown. However, because they are abundant and have a high rate of genomic turnover, retroelements make ideal candidates for probes that will hybridize to multiple locations in a genome but not to genomes of more distantly related species. Because individual families of retroelements can rapidly expand from a few copies in a genome to thousands, species that are separated only by a few million years will share many elements but are likely to have some families that have differentially expanded.
For closely related species, measuring retroelement abundance may contribute to understanding their taxonomy. The elements Prem1, Opie, Prem2/Ji, but especially Huck and Grande hybridized with greater intensity to maize than to chromosomes from Z. diploperennis, while Cinful and Tekay hybridized with almost equal intensity. For many of the elements whose copy number has been estimated previously [Huck, Prem2/Ji, and Opie (MEYERS et al. 2001)], the hybridization pattern reflects the predicted copy number. In Z. diploperennis, many families of retroelements are less abundant than in maize (MEYERS et al. 2001), suggesting that the Z. diploperennis lineage experienced a cessation of retroelement expansion subsequent to its divergence from the other Zea species. Since some retroelement families, including Tekay and TC#13, which both predate the split between Zea and Tripsacum, are present in nearly equal numbers among the Zeas, the presence of fewer copies of many retroelement families is probably not due to widespread retroelement elimination. Also, a study on the retroelement Grande found that the ratio of the LTR copy number to the internal sequence copy number was 2, suggesting that most elements are intact instead of being found as partial remnants resulting from an elimination process. Additionally, all subfamilies of Grande were found in each of the Zea species, so expansion must have occurred before the split of the Zea species (GARCIA-MARTINEZ and MARTINEZ-IZQUIERDO 2003).
The element Huck exemplifies the differences in expansion of copy number that can occur among different lineages. This difference can be used to distinguish different genomes cytologically. Dot blot estimates of Huck in T. dactyloides (35,600–50,800 copies), Z. diploperennis (73,500–104,800), Z. mays (165,700–236,000), and Z. luxurians (246,900–351,800) show significantly different copy numbers among the species (MEYERS et al. 2001). Huck had expanded in the progenitor genome but after the split between Zea and Tripsacum, it ceased rapid expansion in the Tripsacum lineage. In the Zea lineage, Huck continued to increase in copy number but in the line that led to Z. diploperennis, expansion stopped or slowed while the lineages leading to Z. mays and Z. luxurians experienced continued retrotransposition. The difference between Z. luxurians and Z. mays must have appeared rapidly as only 0.63–0.7 million years separate the two species (HILTON and GAUT 1998). Thus, the Huck element hybridizes with different intensities to chromosomes from the three species present in the trispecies hybrid individual (Z. mays, Z. diploperennis, T. dactyloides), allowing them to be readily classified (Figures 2 and 3).
In contrast to Huck, the Tekay element hybridized strongly to all chromosomes in the trispecies hybrid (Figure 1). On the basis of polymorphisms in the glb1 and adh1 genes, maize and Tripsacum lineages are estimated to have diverged 4.5–4.8 MYA (HILTON and GAUT 1998). Tekay was the oldest element identified in the adh region with an insertion timeframe of 5 MYA (SANMIGUEL et al. 1998), which would predate the split between Zea and Tripsacum. Therefore, the abundance of the Tekay element in both Zea and Tripsacum may be explained by an amplification of the Tekay family in the progenitor line that gave rise to the two genera, although it is also possible that it has experienced transpositions in the separate lineages. Except for Tekay, the other elements tested (Huck, Prem1, Prem2/Ji, Opie, Grande, and Cinful) hybridized only weakly to Tripsacum chromosomes, supporting the idea that Tekay has the oldest date of expansion of the major extant maize retroelement families.
Divergence of knobs and centromeric elements revealed by GISH:
Previous work to characterize the knob sequences has revealed that there are many sequence differences between the knobs of maize and Tripsacum but that a strong sequence conservation exists for at least a portion of the 180-bp knob element (DENNIS and PEACOCK 1984). Because of the sensitivity of detection, even a small amount of probe can be visualized and because the knobs consist of thousands of tandemly arrayed elements, a small amount of unblocked probe will be concentrated at discrete spots on the chromosomes and visualized as Tripsacum labeling on maize chromosomes. Because Tripsacum chromatin introgression was suspected for some lines analyzed (discussed below), maize knob sequences were labeled in a different color than the Tripsacum genomic DNA and hybridized simultaneously. The rationale is that if hybridization of the labeled Tripsacum DNA is seen on a maize chromosome, the colocalization of that signal with the maize knob signal will allow the possibility of introgression of a Tripsacum segment to be ruled out.
This approach resulted in strong labeling of maize knobs by the labeled knob sequence (in red) and weak labeling by Tripsacum genomic DNA (in green). On Tripsacum chromosomes, the opposite was observed: strong labeling by Tripsacum genomic DNA and weak labeling by maize knob sequences. The difference in intensity of hybridization reflects the divergence between the knob sequences of the two genomes.
In addition to the maize knobs, discrete signals from the labeled Tripsacum genomic DNA were visualized at the centromeres of the Tripsacum and some Zea chromosomes of the trispecies hybrid. In an analogous fashion to the knobs, the intensity of labeling of the Zea centromeres was markedly less than those of Tripsacum. Two types of elements are known to be abundant at the centromeres of grasses, a tandemly arrayed centromeric satellite repeat and a family of centromere-specific retrotransposons (ANANIEV et al. 1998; CHENG et al. 2002; ZHONG et al. 2002). When CRM was used as a probe, the maize chromosomes were labeled most intensely, even when CRM sequences amplified from Tripsacum genomic DNA were used as FISH probes. Therefore, the GISH signal at Tripsacum centromeres is not due to an increased abundance of CRM elements or to their divergence from maize homologs. When a maize version of CentC was used as a FISH probe, the centromeres of Zea and Tripsacum were labeled with equal intensity (although some maize centromeres had very little CentC signal). Therefore, copy number of CentC does not account for the GISH signal at Tripsacum centromeres. The presence of a Tripsacum-specific subset of CentC elements or other Tripsacum-specific centromere sequences is a potential explanation for Tripsacum centromere labeling in GISH preparations (Figure 4).
Sequencing of large stretches of DNA from centromeric regions of maize and rice has revealed the abundance of a family of centromere-specific retroelements at centromeres (ZHONG et al. 2002; NAGAKI et al. 2004a,b). Because many of the elements are intact and have nearly identical LTRs, it is assumed that these elements resulted from recent insertion events. Solo LTRs from these elements have also been identified and the pattern of distribution of these elements suggests that their frequent insertion and elimination (perhaps through unequal crossovers between the LTRs) plays a role in the evolution of sequences at centromeres. The insertion and elimination of CRM retrotransposons and flanking sequences could contribute to the widely varying amounts of centromeric satellite among centromeres of a given karyotype and among homologs of different varieties (CHENG et al. 2002; KATO et al. 2004; LAMB et al. 2004). CRM1a signal intensity was fairly uniform among the chromosomes of maize and Z. diploperennis while CRM2a intensity was quite different (supplemental Figure 6 at http://www.genetics.org/supplemental/). The relatively weak signal intensity in centromeres from Tripsacum and Z. diploperennis using maize CRM elements could reflect the active expansion of different CRM versions in the different lineages of the three species. However, given the relative scarcity of CRM elements in Z. diploperennis and Tripsacum and the uniformity of the centromeric satellite copy number among the chromosomes (CentC is also fairly uniform on Z. luxurians, data not shown), it is possible that CRM retrotransposon activity is especially high at centromeres in maize.
Development of cytological tools for examination of Zea and Tripsacum:
Use of abundant retroelements as FISH probes to identify genomes:
Although GISH has been traditionally used to distinguish genomes cytologically, there are some drawbacks to that approach. Although many sequences, especially retroelements, may have diverged between species, there may be tandem arrays of conserved elements found in both, such as the megabases of knob sequences present in tandem arrays in Zea and Tripsacum. When GISH was performed using labeled DNA from T. dactyloides on trispecies hybrid chromosome spreads, the Tripsacum chromatin was differentially hybridized but the knobs from both species were also intensely labeled. As a result, while entire chromosomes are readily distinguished, this approach could be limited when applied to material thought to contain introgressed alien chromatin. Large knobs are present in different locations in various maize lines and small knobs can be identified on virtually every chromosome arm using FISH and long exposure times (ADAWY et al. 2004; J. LAMB, A. KATO, J. MEYERS and J. BIRCHLER, unpublished data). Thus, there is a possibility of confusing small knobs in maize with introgressed material. In general, the presence of shared repetitive elements arranged in tandem repeats may also create "false" GISH signals when the procedure is used with any related species. The addition of abundant maize retroelements and the 180-bp knob repeat labeled in a different color than the Tripsacum genomic DNA increased the power of the GISH technique by providing a second confirmation of the identity of the respective chromatin (Figures 4 and 7).
In contrast to GISH, using abundant retroelements to distinguish genomes allows for long exposure times and intense labeling of Tripsacum euchromatin. Use of previously characterized maize retroelements as FISH probes allowed the distinction of maize and Tripsacum genomes in the allopolyploid species T. andersonii (supplemental Figure 2 at http://www.genetics.org/supplemental/), in a maize x Tripsacum intergeneric hybrid (supplemental Figure 1 at http://www.genetics.org/supplemental/), and in the trispecies hybrid. Tripsacum-specific retroelements were not previously reported so it was necessary to isolate them from genomic libraries prior to testing them as FISH chromosome paints.
By screening either a fosmid or a small-insert library, clones were identified that paint the Tripsacum genome. Many crop species have a similar (or lower) gene density than maize including barley, rye, wheat, and oat (BENNETZEN and KELLOGG 1997). Because of the ease in creating small-insert or fosmid libraries, the simplicity of using FISH to screen the clones, and the relatively high number of desirable clones, it would be reasonable to use this approach to create genome-distinguishing "paints" for other species. In some cases, such paints may allow genomes to be distinguished when GISH cannot do so. Using retroelement paints instead of GISH also simplifies the simultaneous detection of other FISH probes.
Maize–Tripsacum addition and replacement lines:
Maize x Tripsacum F1 hybrids can be backcrossed with maize pollen, yielding viable seeds that result from unreduced gametes (MANGELSDORF and REEVES 1932). In just a few backcross generations, most of the Tripsacum chromosomes are lost with a few exceptions (MAGUIRE 1957, 1963). Having a collection of maize lines with each of the different Tripsacum chromosomes would be a valuable tool in anchoring and extending the Tripsacum genetic map. To create a number of maize–Tripsacum addition lines, maize lines with various recessive markers were hybridized to Tripsacum and used as the paternal parent in subsequent backcrosses. By selecting for individuals in which the recessive allele was complemented each generation, the presence of the corresponding Tripsacum chromosome would also be selected (GALINAT et al. 1963; GALINAT and MANGELSDORF 1966).
Five lines produced in this way were provided for analysis by the MGCSC. Three of the lines still retained detectable Tripsacum chromatin but as translocation chromosomes involving a maize chromosome rather than as a maize–Tripsacum addition line with an intact chromosome. One of the lines was described previously to contain a translocation between a maize chromosome and the added Tripsacum chromosome (MAGUIRE 1957) but the other two represent novel translocations that have arisen during maintenance at the stock center. Because the translocation chromosomes can be found as homozygotes with a total chromosome count of 20, it appears that the Tripsacum chromatin has replaced the maize chromatin and is transmittable through both the male and the female gametophytes. Although crossovers between maize and Tripsacum chromosomes are probably rare events, they are the simplest explanation for the type of translocation observed. The recovery of three independent translocations that replace the maize chromatin with a Tripsacum functional equivalent raises the possibility of using Tripsacum as a source of germplasm for maize modification. The retroelement paints developed in this study will facilitate identification and tracking of introgressed Tripsacum chromatin in maize lines.
Extension of maize FISH probes to T. dactyloides and Z. diploperennis:
In concert with the fluorescently labeled retroelements, additional probes can be used in a normal fashion because blocking genomic DNA is not added. Although the GISH procedure does not preclude the use of additional probes, repetitive DNA probes need to be added in larger amounts to obtain similar intensities (data not shown) and some signals may not be easy to visualize. Numerous probes (180-bp knob, TR-1, NOR, 5S, Cent4, and "TAG" microsatellite), which label maize chromosomes, allowing each homolog to be distinguished in somatic preparations, and three probes (rp1, rp3, and -zeinA), which label gene clusters and hybridize to single locations in maize, were applied to chromosome preparations from the trispecies hybrid. Each of these probes hybridized to the Tripsacum and Z. diploperennis chromosomes, demonstrating the general applicability of probes developed in maize to other Zea species and to Tripsacum.
Advances in FISH techniques have allowed specific maize loci to be detected, anchoring the genetic map to a physical location (KOUMBARIS and BASS 2003; ANDERSON et al. 2004; KATO et al. 2006; WANG et al. 2006). In the related Sorghum species, the use of BACs has allowed detailed cytogenetic maps to be developed (ZWICK et al. 1998; ISLAM-FARIDI et al. 2002; KIM et al. 2002), opening up the possibility of cytogenetic examination of chromosomal rearrangements among species in this genus. Many of these BACs can be used as FISH probes in maize or rice (ZWICK et al. 1998; KOUMBARIS and BASS 2003). As an alternative to using BACs as FISH probes to detect specific genetic loci, small and unique DNA elements, such as individual genes, can be used as probes. The current limit of consistent detection in maize is 3 kb (WANG et al. 2006) and as sequencing efforts proceed on the maize genome, numerous probes against specific loci will be developed. The trispecies hybrid provides an excellent means of testing the functionality of additional maize FISH probes in Tripsacum and Z. diploperennis because it contains an internal positive control in the form of the maize chromosomes. As more probes become available, they could be used to improve the genetic map of Tripsacum and will facilitate cytogenetic comparisons of these species.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
ADAWY, S. S., R. M. STUPAR and J. JIANG, 2004 Fluorescence in situ hybridization analysis reveals multiple loci of knob-associated DNA elements in one-knob and knobless maize lines. J. Histochem. Cytochem. 52: 1113–1116.
ALTSCHUL, S. F., W. GISH, W. MILLER, E. W. MYERS and D. J. LIPMAN, 1990 Basic local alignment search tool. J. Mol. Biol. 215: 403–410.[CrossRef][Medline]
ANANIEV, E. V., R. L. PHILLIPS and H. W. RINES, 1998 Chromosome-specific molecular organization of maize (Zea mays L.) centromeric regions. Proc. Natl. Acad. Sci. USA 95: 13073–13078.
ANDERSON, L. K., N. SALAMEH, H. W. BASS, L. C. HARPER, W. Z. CANDE et al., 2004 Integrating genetic linkage maps with pachytene chromosome structure in maize. Genetics 166: 1923–1933.
BENNETZEN, J. L., and E. A. KELLOGG, 1997 Do plants have a one-way ticket to genomic obesity? Plant Cell 9: 1509–1514.[CrossRef][Medline]
BRUNNER, S., K. FENGLER, M. MORGANTE, S. TINGEY and A. RAFALSKI, 2005 Evolution of DNA sequence nonhomologies among maize inbreds. Plant Cell 17: 343–360.
CHENG, Z., F. DONG, T. LANGDON, S. OUYANG, C. R. BUELL et al., 2002 Functional rice centromeres are marked by a satellite repeat and a centromere-specific retrotransposon. Plant Cell 14: 1691–1704.
DENNIS, E. S., and W. J. PEACOCK, 1984 Knob heterochromatin homology in maize and its relatives. J. Mol. Evol. 20: 341–350.[CrossRef][Medline]
GALINAT, W. C., 1988 The origin of corn, pp. 1–31 in Corn and Corn Improvement, edited by G. F. SPRAGUE. American Society of Agronomy, Madison, WI.
GALINAT, W. C., and P. C. MANGELSDORF, 1966 Genetic correspondence of Tripsacum chromosomes to their homeologs from corn. Maize Genet. Coop. Newsl. 40: 99–100.
GALINAT, W. C., P. C. MANGELSDORF and R. S. CHAGANTI, 1963 Segregation of genetically marked chromosomes in maize-Tripsacum hybrids. Maize Genet. Coop. Newsl. 37: 32–33.
GARCIA-MARTINEZ, J., and J. A. MARTINEZ-IZQUIERDO, 2003 Study on the evolution of the Grande retrotransposon in the Zea genus. Mol. Biol. Evol. 20: 831–841.
GAUT, B. S., M. LE THIERRY D'ENNEQUIN, A. S. PEEK and M. C. SAWKINS, 2000 Maize as a model for the evolution of plant nuclear genomes. Proc. Natl. Acad. Sci. USA 97: 7008–7015.
HILTON, H., and B. S. GAUT, 1998 Speciation and domestication in maize and its wild relatives: evidence from the globulin-1 gene. Genetics 150: 863–872.
ISLAM-FARIDI, M. N., K. L. CHILDS, P. E. KLEIN, G. HODNETT, M. A. MENZ et al., 2002 A molecular cytogenetic map of sorghum chromosome 1. Fluorescence in situ hybridization analysis with mapped bacterial artificial chromosomes. Genetics 161: 345–353.
KATO, A., J. C. LAMB and J. A. BIRCHLER, 2004 Chromosome painting using repetitive DNA sequences as probes for somatic chromosome identification in maize. Proc. Natl. Acad. Sci. USA 101: 13554–13559.
KATO, A., P. S. ALBERT, J. M. VEGA and J. A. BIRCHLER, 2006 Sensitive FISH signal detection in maize using directly labeled probes produced by high concentration DNA polymerase nick translation. Biotech. Histochem. (in press).
KATO, Y. T. A., 1984 Chromosome morphology and the origin of maize and its races. Evol. Biol. 17: 219–253.
KIM, J. S., K. L. CHILDS, M. N. ISLAM-FARIDI, M. A. MENZ, R. R. KLEIN et al., 2002 Integrated karyotyping of sorghum by in situ hybridization of landed BACs. Genome 45: 402–412.[Medline]
KOUMBARIS, G. L., and H. W. BASS, 2003 A new single-locus cytogenetic mapping system for maize (Zea mays L.): overcoming FISH detection limits with marker-selected sorghum (S. propinquum L.) BAC clones. Plant J. 35: 647–659.[CrossRef][Medline]
LAMB, J. C., J. THEURI and J. A. BIRCHLER, 2004 What's in a centromere? Genome Biol. 5: 239.[CrossRef][Medline]
LAMB, J. C., A. KATO and J. A. BIRCHLER, 2005 Sequences associated with A chromosome centromeres are present throughout the maize B chromosome. Chromosoma 113: 337–349.[CrossRef][Medline]
MA, J., K. M. DEVOS and J. L. BENNETZEN, 2004 Analyses of LTR-retrotransposon structures reveal recent and rapid genomic DNA loss in rice. Genome Res. 14: 860–869.
MAGUIRE, M. P., 1957 Cytogenetic studies of Zea hyperploid for a chromosome derived from Tripsacum. Genetics 42: 473–486.
MAGUIRE, M. P., 1960a A study of homology between a terminal portion of Zea chromosome 2 and a segment derived from Tripsacum. Genetics 45: 195–209.
MAGUIRE, M. P., 1960b A study of pachytene chromosome pairing in a corn–Tripsacum hybrid derivative. Genetics 45: 651–664.
MAGUIRE, M. P., 1963 High transmission frequency of a Tripsacum chromosome in corn. Genetics 48: 1185–1194.
MANGELSDORF, P. C., 1947 The origin and evolution of maize. Adv. Genet. 1: 161–207.
MANGELSDORF, P. C., and R. G. REEVES, 1931 Hybridization of maize, Tripsacum and Euchlaena. J. Hered. 22: 329–343.
MANGELSDORF, P. C., and R. G. REEVES, 1932 Genetic and cytological studies in hybrids of Zea and Tripsacum. Proc. 6th Int. Congr. Genet. 2: 134–135.
MCCLINTOCK, B., Y. T. A. KATO and A. BLUMENSHEIN, 1981 Chromosome Constitution of Races of Maize. Colegio do Postgraduados, Chapingo, Mexico.
MEYERS, B. C., S. V. TINGEY and M. MORGANTE, 2001 Abundance, distribution, and transcriptional activity of repetitive elements in the maize genome. Genome Res. 11: 1660–1676.
MROCZEK, R. J., and R. K. DAWE, 2003 Distribution of retroelements in centromeres and neocentromeres of maize. Genetics 165: 809–819.
NAGAKI, K., J. SONG, R. M. STUPAR, A. S. PAROKONNY, Q. YUAN et al., 2003 Molecular and cytological analyses of large tracks of centromeric DNA reveal the structure and evolutionary dynamics of maize centromeres. Genetics 163: 759–770.
NAGAKI, K., Z. CHENG, S. OUYANG, P. B. TALBERT, M. KIM et al., 2004a Sequencing of a rice centromere uncovers active genes. Nat. Genet. 36: 138–145.[CrossRef][Medline]
NAGAKI, K., P. NEUMANN, D. ZHANG, S. OUYANG, C. R. BUELL et al., 2004b Structure, divergence, and distribution of the CRR centromeric retrotransposon family in rice. Mol. Biol. Evol. 22: 845–855.
SANMIGUEL, P., and J. L. BENNETZEN, 1997 Evidence that a recent increase in maize genome size was caused by the massive amplification of intergene retrotransposons. Ann. Bot. 82: 37–44.[CrossRef][Medline]
SANMIGUEL, P., A. TIKHONOV, Y. K. JIN, N. MOTCHOULSKAIA, D. ZAKHAROV et al., 1996 Nested retrotransposons in the intergenic regions of the maize genome. Science 274: 765–768.
