Molecular Cytogenetic Comparisons Between Model Plant And Several Plant Genomes

In this study, comparisons among model plants (rice and Arabidopsis thaliana) and several dicotyledonous and monocotyledonous genomes were performed using fluorescence in situ hybridization (FISH) technique together with other cytogenetic methods. The main results are as follows:1. Comparative genomic analysis was conducted using comparative genomic in situ hybridization (cGISH) with biotin-labeled Arabidopsis and rice genomic DNA probes. The cGISH signals, unlike those of in situ hybridization with single- or low- copy DNA sequence probes, were detected without ocascade amplification in our experiments, so they actually represented the hybridization of the repetitive DNAs in the genomic probes to their homologous sequences distributed along the target chromosomes.The labeled Arabidopsis genomic DNA yielded signals on all chromosomes in sixtested species including dicots Brassica oleracea var. capitata, tomato and Vicia faba aswell as monocots rice, maize and barley. The signals were basically dispersedlydistributed along chromosomes and exhibited a tendency that the larger the target genomewas, the more the signals were shown, and the more dispersed the signals were along thelengths of the chromosomes. Obviously, the signals in the nucleolar organizing regions(NORs) were more prominent than those in other chromosomal regions in all testedspecies, indicating that labeled Arabidopsis genomic DNA could be used for NORmapping in other plants. The terminal signals on tomato chromosomes were proved to bethe hybridization of the Arabidopsis telomeric repeats. In all species tested, signalsoccurred mainly at interstitial sites as well as chromosomal ends, and a few signalsappeared in centromeric or pericentromeric regions. There were no signals or only a fewsignals in the centromere adjacent heterochromatic regions of V. faba and the chromosomeknobs of maize, suggesting that the repetitive DNAs within these prominentheterochromatic regions is not conservative among species or genera. As analyzed indetail, the cGISH signal patterns of barley, which were different from the C- andN-banding patterns of this species, could readily discriminate individual chromosome,suggesting that cGISH with Arabidopsis genomic DNA probe is also an effectivechromosome banding technique for plant chromosome analysis.The labeled rice genomic DNA produced a large number of dispersed signals on all chromosomes in tomato, V.faba, maize, barley, sorghum, rye and millet. The NOR signals in most target species were less strongly shown compared to those of Arabidopsis genomic DNA probe. Obviously, there were uneven signals in tomato, and more signals were shown in interstitial regions, while fewer signals in some of its centromeric, pericentromeric and subterminal regions. Differently, the dense signals in V. faba were basically evenly distributed along the chromosomes except the centromere adjacent heterochromatic regions in which few signals were shown. The chromosomes of five gramineous species showed very large numbers of signals. Comparatively, it was the most abounding and dense in rye, the sparsest in sorghum and millet, and the middle in maize and barley for their signal distribution. It indicated that the quantity of the conserved and homologous repetitive DNAs shared by these species are different considerably. The signals in maize, barley and rye were basically evenly distributed along the chromosomes, while those in sorghum and millet showed uneven distribution. Most centromeric or pericentromeric regions in the gramineous species showed signals. Few signals were shown in the maize knobs, the proximal heterochromatic areas of sorghum and the subterminal heterochromatic regions of rye, suggesting that the repetitive DNAs involved in these large heterochromatic regions are not conservative among species or genera. Hybridization at different stringency revealed that the repetitive sequences showing more than 65% homology are very rich between rice and maize or barley, while those of more than 90% homology are rare. The sorghum, millet and barley genomic DNA probes all resulted in large numbers of signals on maize chromosomes, further indicating a large amount of conserved repetitive sequences among species of different gramineous subfamilies. The signals on rice and millet chromosomes produced by labeled maize genomic DNA displayed obvious uneven distribution, and were markedly fewer than those on maize chromosomes produced by labeled rice or millet genomic DNA.The above results revealed the existence of many homologous repetitive DNAs besides rDNA and telomeric repeats among different plant genomes. These repetitive DNAs should be established before the divergence of dicotyledon and monocotyledon, different families or different subfamilies, and their higher homology has been maintained between distantly related species because of their low rate of evolution. Our investigation indicated that the ancient conserved repetitive sequences have undergone considerable amplification and homogenization in large plant genomes during evolution, suggesting that, like the production of new repetitive DNAs, the amplification of ancient conserved repetitive sequences should also play an important role in the expansion of genome size in plants.Our investigation demonstrated that, because of its ability for revealing the conserved repetitive DNAs among plant genomes and the distribution characteristics of these sequences along chromosomes, the cGISH technique developed in our study could be anew approach for comparative genomic analysis in plants, whose role cannot be replaced by other chromosome banding techniques as well as molecular methods.2. Using an improved GISH procedure, fluorescence in situ hybridization withgenomic DNA to its own chromosomes (called self-genomic in situ hybridization,self-GISH) was conducted in six selected plant species different in genome size.Non-uniform distribution of the fluorescence-labeled probe DNA was observed in alltested species. The hybridization patterns differed among species and were correlated withthe genome size. The chromosomes in the small genome, A. thaliana, were labeled only inthe pericentromeric regions as well as in the NORs. The signals in rice, sorghum and B.oleracea var. capitata that are all relatively small in genome size were dispersed along thechromosome lengths with a predominant distribution in the pericentromeric, proximal andother heterochromatic regions. All the chromosomes in both maize and barley, whosegenomes are very large, were densely labeled, with strongly labeled and lightly labeled ornon-labeled regions being arranged alternatively throughout the lengths. Additionally,enhanced signal bands were shown in all centromeric and pericentromeric regions in B.oleracea var. capitata, and in most centromeric, all pericentromeric as well as someintercalary regions in barley. The enhanced signal band patterns in barley were proved tobe consistent with the N-banding pattern as well as the FISH banding pattern of (GAA)nmicrosatellite of this species, permitting identification of all the chromosomes. Thesimilarity between the self-GISH signal patterns and the Cot-1 DNA FISH signal patternsin rice , and the correspondence between the self-GISH patterns in Arabidopsis and riceand the distribution characteristics of repetitive sequences in the two genomes revealed bywhole genome sequencing indicates that the self-GISH signals actually resulted from thehybridization of genomic repetitive sequences to the chromosomes. Except some stronglylabeled repeated genes such as 45S rDNA, most strongly labeled regions should begene-poor regions, while the lightly labeled or non-labeled regions should be gene-richregions. Hence, the self-GISH technique should be useful for revealing the genomicorganization of genes and repetitive sequences at chromosome level as well thechromosome structural differentiation associated with repetitive sequences in plants.3. Mitotic chromosome spreads of 16 plant species belonging to six families were analyzed using an improved combined PI and DAPI (CPD) staining procedure. Fluorescence in situ hybridization (FISH) with 45S rDNA probe was conducted sequentially on the same spreads to evaluate the efficiency and sensitivity of the technique. Fluorochrome staining with chromomycin A3 (CMA)-DAPI also was conducted to clarify the properties of the sequences involved in the CPD banded regions. Our results revealed that all of the rDNA sites (NORs) in the species tested were efficiently shown as red bands by CPD staining, and the number and position of the bands corresponded precisely to those of the 45S rDNA FISH signals, indicating that the detection sensitivity of CPD staining is similar to that of FISH. In 10 of the species tested including Aegilops squarrosa,Allium sativum, Oryza sativum ssp. indica, Oryza officinalis, Pisum sativum, Secale cereale, Setaria italica, Sorghum vulgare, Vicia faba and Zea mays, CPD bands were exhibited exclusively in their NORs, while in other six species including Hordeum vulgare, Allium cepa, Psophocarpus tetragonolobus, Arabidopsis thaliana, Brassica oleracea var. capitata and Lycopersicon esculentum, CPD bands appeared in chromosomal regions other than their NORs. The CPD bands were in accordance with the CMA bands in all species tested. Likewise, human chromosomes also showed CPD bands being completely in agreement with the CMA bands. These results indicated there existed GC-rich sequences in the CPD bands and that the improved CPD staining procedure was specific for GC-rich regions in plant genomes. Our investigation not only clarified the banding mechanism of CPD staining, but also developed a new approach for detection of NORs and other GC-rich regions in paints.Sequential CPD staining and FISH with two different 45S rDNA clones on meiotic pachytene and mitotic metaphase chromosomes were performed in tomato. 10 red CPD bands were shown on eight pachytene bivalents, and 12 bands were shown on six pairs of mitotic metaphase chromosomes. The CPD bands exhibited on mitotic metaphase chromosomes corresponded to the prominent bands exhibited on the pachytene chromosomes. The distinctive CPD bands, which could be constantly and clearly detected using the CPD staining procedure we improved, provide new landmarks for chromosome identification in tomato. FISH with the tomato 45S rDNA clone revealed very strong signal(s) in the satellite(s) on the short arm of chromosome 2 as well as weak signals in five CPD banded regions at pachytene or four pairs of CPD banded regions at metaphase. However, FISH with the wheat 45S rDNA clone (pTa71 plasmid) only revealed signals in the satellites. Considering the difference in sequence between the two rDNA clones, we infer that only the satellite contain the coding regions of 45S rDNA unit in tomato, i.e. tomato has only one pair of 45S rDNA sites.On the basis of chromosome lengths, DAPI and CPD banding patterns, together with FISH signals of 45S and 5S rDNA, detailed karyotype of the winged bean was constructed. FISH of the 45S rDNA probe on mitotic chromosomes revealed three pairs of 45S rDNA sites including one major and two minors. The major pair of sites was located in the unique secondary constrictions of the satellite chromosome pair, while the minors were located in the positions near the centromeres on the long or short arms of two pairs of non-satellite chromosomes. For 5S rDNA, only one pair of sites was detected using FISH. FISH with pAtT4 probe proved that Arabidopsis-type telomeric repeats were located at the ends of all chromosomes. Distinct DAPI and CPD banding patterns were simultaneously shown after CPD staining. CPD bands were shown at all 45S rDNA sites and centromeric regions. Each chromosome pair could be identified with DAPI and CPD banding patterns in combination with chromosome measurements.The localization of rDNA sites by CPD staining together with FISH of 45S rDNA probe in the 16 plant species revealed that the 45S rDNA sites occurred mainly in the short armsinstead of the long arms, and interstitial and terminal rDNA sites appeared at similar frequency. Secondary constrictions appeared in most interstitial rDNA sites, but the orientation of the rDNA clusters relative to the secondary constrictions differed among interstitial sites. The chromosomal distribution of 45S rDNA sites in plants could be classified into 12 types based on the differences in the location of rDNA sites and the orientation of rDNA clusters.4. Comparative analysis of the patterns of rDNA chromatin organization in interphase nuclei of 11 plant species belonging to 6 families was conducted using Interphase-FISH technique. Our results revealed that two types of rDNA hybridization signals, knob and spot, were shown by 45S rDNA probe in the interphase nuclei of all tested species, indicating the existence of similar patterns of rDNA chromatin organization at interphase in higher plants. Some differences in interphase rDNA organization were also found among species. The knobs strongly fluorescing were usually located at the periphery of the nucleoli or inside the nucleoli and represented condensed inactive rDNA chromatin. The spots less strongly fluorescing were similar in size and dispersed in the nucleoli. The spots associated with or emanating from the knobs were observed in a portion of the nuclei in each tested species. It was obvious that the more the spots were shown, the smaller the knobs became, and the number of spots was correlated with the activity of the cell. It was concluded that spots resulted from decondensation of knobs, and the decondensation of rDNA sites was the cytogenetic manifestation of active rRNA gene transcription. The expected thin threads of chromatin between spots were not observed in all tested species. Our observations were in favor of the inference recently proposed by other studies that spot is the structural unit in the organization of active ribosomal genes and the site that the rRNA synthesis takes place.In each tested species, the number of spots varied considerably among interphase nuclei at different stage, suggesting the number of activated rRNA genes differed in the nuclei at different stage. Obviously, the knobs could not be decondensed thoroughly even if the maximum number of spots were shown in the nuclei. Sequential CPD and silver staining in maize revealed that the vast majority of the rDNA sites did not participate in the formation of the nucleolus. In maize, sorghum and Aegilops tauschii, the size and intensity of CPD bands or FISH signals of homologous rDNA sites in metaphase cells were almost identical, indicating the similar number of rRNA genes between the two homologous sites. However, the level of expression usually differed between their homologous sites as revealed by rDNA FISH signals in interphase nuclei. This was confirmed by the difference in the length of secondary constriction of homologous chromosomes and the differences in size for Ag-NORs and nucleoli in maize.In P. tetragonolobus the major sites could display as knobs together with spots at a higher frequency, while the minor sites usually showed perinucleolar knobs or knobs distant from the nucleoli, suggesting the different level of expression between the major and minor rDNA sites. Silver staining revealed variable number of Ag-NORs on mitoticchromosomes. One pair of Ag-NORs of the major rDNA sites was observed in every mitotic cell, but the Ag-NORs of the minor sites only occurred in a proportion of the recorded cells. The occurrence frequency of Ag-NORs of the minor sites on chromosome 8 was much higher than that of the minor sites on chromosome 1. Moreover, only the Ag-NORs of the minor sites on chromosome 8 appeared in the cells showing four Ag-NORs. Such differential activation and level of expression among non-homologous rDNA sites were not observed in Sagittaria sagittifolia, which also has three pairs of rDNA sites, but the copy number of rRNA genes are comparatively similar among different pairs of sites.The 45S rDNA probe produced a large number of spots in the nucleolar regions of A. thaliana prophase cells, indicating that rDNA transcription was very active during prophase. In each species tested, the rDNA sites in prometaphase cells displayed decondensed FISH signals as well as many spots. Silver staining showed prominent nucleoli in prometaphase cells of maize. These facts indicated active rDNA transcription during prometaphase.