Origin of plantlets and callus obtained from chile pepper anther cultures

Summary Androgenesis occurred from chile pepper (Capsicum annuum L.) anthers incubated in a continuous warm environment (29° C) with continuous light. Forty plantes and embryoids were retrieved from anther cultures and anllyzed for isozyme markers. Of these, 35 exhibited a single allele for markers suggesting microspore origin, while 5 were heterozygous indicating somatic tissue origin. Chromosome numbers were confirmed for 21 plantlets, of which 16 were haploid and 5 were diploid. However, two plants exhibited a single allele for an isozyme marker but possessed the diploid chromosome number, suggesting spontaneous doubling. Anther cultures also produced callus. Nearly 92% of the slow-growing calli sampled were heterozygous for the isozyme marker, suggesting somatic tissue origin. More than 46% of the fast-growing calli exhibited only one allele for the marker, indicating microspore origin. Callus did not regenerate plantlets. The occurrence of both heterozygous and homozygous diploid plantlets from pepper anther cultures has important implications for applied breeding programs.     

Linkage analysis of sex determination in the honey bee (Apis mellifera)

A colony-level phenotype was used to map the major sex determination locus (designatedX) in the honey bee (Apis mellifera). Individual queen bees (reproductive females) were mated to single drones (fertile males) by instrumental insemination. Haploid drone progeny of an F1 queen were each backcrossed to daughter queens from one of the parental lines. Ninety-eight of the resulting colonies containing backcross progeny were evaluated for the trait ‘low brood-viability’ resulting from the production of diploid drones that were homozygous atX. DNA samples from the haploid drone fathers of these colonies were used individually in polymerase chain reactions (PCR) with 10-base primers. These reactions generated random amplified polymorphic DNA (RAPD) markers that were analyzed for cosegregation with the colony-level phenotype. One RAPD marker allele was shared by 22 of 25 drones that fathered low brood-viability colonies. The RAPD marker fragment was cloned and partially sequenced. Two primers were designed that define a sequence-tagged site (STS) for this locus. The primers amplified DNA marker fragments that cosegregated with the original RAPD marker. In order to more precisely estimate the linkage betweenX and the STS locus, another group of bees consisting of progeny from one of the low-brood viability colonies was used in segregation analysis. Four diploid drones and 181 of their diploid sisters (workers, nonfertile females) were tested for segregation of the RAPD and STS markers. The cosegregating RAPD and STS markers were codominant due to the occurrence of fragment-length alleles. The four diploid drones were homozygous for these markers but only three of the 181 workers were homozygotes (recombinants). Therefore the distance betweenX and the STS locus was estimated at 1.6 cM. An additional linked marker was found that was 6.6 cM from the STS locus.     

Detection of lethal yellowing phytoplasma in coconut plantlets obtained through in vitro germination of zygotic embryos from the seeds of infected palms

Lethal yellowing (LY) is a disease caused by 16SrIV phytoplasmas that has devastated coconut plantations in the Americas. An alternative means of phytoplasma spread is through seeds. Therefore, we used a novel approach based on plumules from the embryos of LY‐diseased coconut palms. We cultured the plumules in vitro to determine the presence of phytoplasma DNA in the plantlets. In the first assay, 185 embryos were obtained. The results showed positive detection in 20 samples (11%) with the nested PCR and in 59 samples (32%) with the TaqMan real‐time PCR. A second assay was designed to trace plumules to their respective embryos and haustorial tissues to determine whether they had derived from an embryo with positive LY detection; a total of 124 embryos were obtained. The results showed no positive detection with the nested PCR and positive detection in 42 of the haustorial tissue samples (32%) with the TaqMan real‐time PCR. The 124 plumules isolated from the embryos were cultivated under in vitro conditions and divided into two groups. Group A was followed for shoot formation and Group B was followed to the plantlet stage. After 3 months of cultivation, 33 cultures (50%) within Group A became necrotic; the rest were analysed to evaluate LY phytoplasma DNA with the TaqMan real‐time PCR assay and 14 (42%) tested positive. After 18 months of cultivation, 20 cultures (34%) within Group B became necrotic. The rest were analysed for the detection of the LY phytoplasma DNA, and 15 and 11 (39% and 29%) of the samples tested positive with the TaqMan real‐time PCR and nested PCR assays, respectively. Blast analysis of the sequenced products revealed that the sequences showed 99% homology with LY‐phytoplasma subgroup 16SrIV‐A. The results presented here demonstrate, for the first time, the occurrence of the transmission of LY phytoplasmas from coconut embryos to plantlets.     

Linkage analysis of sex determination in the honey bee (Apis mellifera)

A colony-level phenotype was used to map the major sex determination locus (designatedX) in the honey bee (Apis mellifera). Individual queen bees (reproductive females) were mated to single drones (fertile males) by instrumental insemination. Haploid drone progeny of an F1 queen were each backcrossed to daughter queens from one of the parental lines. Ninety-eight of the resulting colonies containing backcross progeny were evaluated for the trait low brood-viability resulting from the production of diploid drones that were homozygous atX. DNA samples from the haploid drone fathers of these colonies were used individually in polymerase chain reactions (PCR) with 10-base primers. These reactions generated random amplified polymorphic DNA (RAPD) markers that were analyzed for cosegregation with the colony-level phenotype. One RAPD marker allele was shared by 22 of 25 drones that fathered low brood-viability colonies. The RAPD marker fragment was cloned and partially sequenced. Two primers were designed that define a sequence-tagged site (STS) for this locus. The primers amplified DNA marker fragments that cosegregated with the original RAPD marker. In order to more precisely estimate the linkage betweenX and the STS locus, another group of bees consisting of progeny from one of the low-brood viability colonies was used in segregation analysis. Four diploid drones and 181 of their diploid sisters (workers, nonfertile females) were tested for segregation of the RAPD and STS markers. The cosegregating RAPD and STS markers were codominant due to the occurrence of fragment-length alleles. The four diploid drones were homozygous for these markers but only three of the 181 workers were homozygotes (recombinants). Therefore the distance betweenX and the STS locus was estimated at 1.6 cM. An additional linked marker was found that was 6.6 cM from the STS locus.     

Identification of molecular markers associated with mite resistance in coconut (Cocos nucifera L.).

Coconut mite (Aceria guerreronis 'Keifer') has become a major threat to Indian coconut (Co?cos nucifera L.) cultivators and the processing industry. Chemical and biological control measures have proved to be costly, ineffective, and ecologically undesirable. Planting mite-resistant coconut cultivars is the most effective method of preventing yield loss and should form a major component of any integrated pest management stratagem. Coconut genotypes, and mite-resistant and -susceptible accessions were collected from different parts of South India. Thirty-two simple sequence repeat (SSR) and 7 RAPD primers were used for molecular analyses. In single-marker analysis, 9 SSR and 4 RAPD markers associated with mite resistance were identified. In stepwise multiple regression analysis of SSRs, a combination of 6 markers showed 100% association with mite infestation. Stepwise multiple regression analysis for RAPD data revealed that a combination of 3 markers accounted for 83.86% of mite resistance in the selected materials. Combined stepwise multiple regression analysis of RAPD and SSR data showed that a combination of 5 markers explained 100% of the association with mite resistance in coconut. Markers associated with mite resistance are important in coconut breeding programs and will facilitate the selection of mite-resistant plants at an early stage as well as mother plants for breeding programs.     

Tetraploidization events by chromosome doubling of nucellar cells are frequent in apomictic citrus and are dependent on genotype and environment

Background and Aims

Polyploidy is a major component of plant evolution. The citrus gene pool is essentially diploid but tetraploid plants are frequently encountered in seedlings of diploid apomictic genotypes. The main objectives of the present study were to establish the origin of these tetraploid plants and to ascertain the importance of genotypic and environmental factors on tetraploid formation.

Methods

Tetraploid seedlings from 30 diploid apomictic genotypes were selected by flow cytometry and genotyped with 24 single sequence repeat (SSR) markers to analyse their genetic origin. Embryo rescue was used to grow all embryos contained in polyembryonic seeds of ‘Tardivo di Ciaculli’ mandarin, followed by characterization of the plantlets obtained by flow cytometry and SSR markers to accurately establish the rate of tetraploidization events and their potential tissue location. Inter-annual variations in tetraploid seedling rates were analysed for seven genotypes. Variation in tetraploid plantlet rates was analysed between different seedlings of the same genotype (‘Carrizo’ citrange; Citrus sinensis × Poncirus trifoliata) from seeds collected in different tropical, subtropical and Mediterranean countries.

Key Results

Tetraploid plants were obtained for all the studied diploid genotypes, except for four mandarins. All tetraploid plants were identical to their diploid maternal line for SSR markers and were not cytochimeric. Significant genotypic and environmental effects were observed, as well as negative correlation between mean temperature during the flowering period and tetraploidy seedling rates. The higher frequencies (20 %) of tetraploids were observed for citranges cultivated in the Mediterranean area.

Conclusions

Tetraploidization by chromosome doubling of nucellar cells are frequent events in apomictic citrus, and are affected by both genotypic and environmental factors. Colder conditions in marginal climatic areas appear to favour the expression of tetraploidization. Tetraploid genotypes arising from chromosome doubling of apomictic citrus are extensively being used as parents in breeding programmes to develop seedless triploid cultivars and have potential direct use as new rootstocks.     

Production of Androgenetic Zebrafish (Danio Rerio)

To help investigate the evolutionary origin of the imprinting (parent-of-origin mono-allelic expression) of paternal genes observed in mammals, we constructed haploid and diploid androgenetic zebrafish (Danio rerio). Haploid androgenotes were produced by fertilizing eggs that had been X-ray irradiated to eliminate the maternal genome. Subsequent inhibition of the first mitotic division of haploid androgenotes by heat shock produced diploid androgenotes. The lack of inheritance of maternal-specific DNA markers (RAPD and SSR) by putative diploid and haploid androgenotes confirmed the androgenetic origin of their genomes. Marker analysis was performed on 18 putative androgenotes (five diploids and 13 haploids) from six families. None of 157 maternal-specific RAPD markers analyzed, some of which were apparently homozygous, were passed on to any of these putative androgenotes. A mean of 7.7 maternal-specific markers were assessed per family. The survival of androgenetic zebrafish suggests that if paternal imprinting occurs in zebrafish, it does not result in essential genes being inactivated when their expression is required for development. Production of haploid androgenotes can be used to determine the meiotic recombination rate in male zebrafish. Androgenesis may also provide useful information about the mechanism of sex determination in zebrafish.     

An integrated molecular linkage map of diploid wheat based on a <Emphasis Type="Italic">Triticum boeoticum</Emphasis> × <Emphasis Type="Italic">T. monococcum</Emphasis> RIL population

Diploid A genome species of wheat harbour immense variability for biotic stresses and productivity traits, and these could be transferred efficiently to hexaploid wheat through marker assisted selection, provided the target genes are tagged at diploid level first. Here we report an integrated molecular linkage map of A genome diploid wheat based on 93 recombinant inbred lines (RILs) derived from Triticum boeoticum × Triticum monococcum inter sub-specific cross. The parental lines were analysed with 306 simple sequence repeat (SSR) and 194 RFLP markers, including 66 bin mapped ESTs. Out of 306 SSRs tested for polymorphism, 74 (24.2%) did not show amplification (null) in both the parents. Overall, 171 (73.7%) of the 232 remaining SSR and 98 (50.5%) of the 194 RFLP markers were polymorphic. Both A and D genome specific SSR markers showed similar transferability to A genome of diploid wheat species. The 176 polymorphic markers, that were assayed on a set of 93 RILs, yielded 188 polymorphic loci and 177 of these as well as two additional morphological traits mapped on seven linkage groups with a total map length of 1,262 cM, which is longer than most of the available A genome linkage maps in diploid and hexaploid wheat. About 58 loci showed distorted segregation with majority of these mapping on chromosome 2A^m. With a few exceptions, the position and order of the markers was similar to the ones in other maps of the wheat A genome. Chromosome 1A^m of T. monococcum and T. boeoticum showed a small paracentric inversion relative to the A genome of hexaploid wheat. The described linkage map could be useful for gene tagging, marker assisted gene introgression from diploid into hexaploid wheat as well as for map based cloning of genes from diploid A genome species and orthologous genes from hexaploid wheat.     

Phylogenetic analysis of Oryza species, based on simple sequence repeats and their flanking nucleotide sequences from the mitochondrial and chloroplast genomes

Simple sequence repeats (SSR) and their flanking regions in the mitochondrial and chloroplast genomes were sequenced in order to reveal DNA sequence variation. This information was used to gain new insights into phylogenetic relationships among species in the genus Oryza. Seven mitochondrial and five chloroplast SSR loci equal to or longer than ten mononucleotide repeats were chosen from known rice mitochondrial and chloroplast genome sequences. A total of 50 accessions of Oryza that represented six different diploid genomes and three different allopolyploid genomes of Oryza species were analyzed. Many base substitutions and deletions/insertions were identified in the SSR loci as well as their flanking regions. Of mononucleotide SSR, G (or C) repeats were more variable than A (or T) repeats. Results obtained by chloroplast and mitochondrial SSR analyses showed similar phylogenetic relationships among species, although chloroplast SSR were more informative because of their higher sequence diversity. The CC genome is suggested to be the maternal parent for the two BBCC genome species (O. punctata and O. minuta) and the CCDD species O. latifolia, based on the high level of sequence conservation between the diploid CC genome species and these allotetraploid species. This is the first report of phylogenetic analysis among plant species, based on mitochondrial and chloroplast SSR and their flanking sequences.     

Ploidy of somatic embryos and the ability to regenerate plantlets in melon (Cucumis melo L.)

Summary The number of chromosomes in cells of callus, somatic embryos and regenerated plantlets during somatic embryogenesis were examined in two cultivars of melon (Cucumis melo L.). Somatic embryos were diploid (50.0%/32.1%), tetraploid (38.5%/57.5%) and octoploid (11.5%/10.4%) whereas in callus cells diploidy (41.9%/43.3%), tetraploidy (27.9%/25.8%), octoploidy (11.6%/15.5%) and a low frequency of other types of ploidy and aneuploidy were observed. Mixoploid somatic embryos were not observed. These results suggest that the somatic embryos were selectively differentiated from diploid, tetraploid and octoploid cells, and that endopolyploidization of cultured cells occurred before the start of cell division leading to somatic embryogenesis. The ratio of diploid to tetraploid (1.30/0.55) in somatic embryos was less than that in callus cells (1.50/1.68) while ratios of diploid to octoploid (4.35/3.09) and tetraploid to octoploid (3.35/5.52) in somatic embryos were greater than those in callus cells (3.61/2.80 and 2.40/1.67). Therefore, it appears that the ability of callus cell to differentiate into somatic embryos increases in the following order: octoploid < diploid < tetraploid. Regenerated plantlets were diploid (65.5%/55.1%) and tetraploid (34.5%/44.9%). No octoploid plantlets were observed. The ratio of diploid to tetraploid in regenerated plantlets (1.72/1.23) was greater than that in somatic embryos. Therefore, it appears that the ability of somatic embryos to develop into plantlets increases in the following order: octoploid < tetraploid < diploid.     

A kiwifruit (Actinidia spp.) linkage map based on microsatellites and integrated with AFLP markers

A genetic map of kiwifruit (Actinidia spp.) was constructed using microsatellite and AFLP markers and the pseudo-testcross mapping strategy. (AC)_n and (AG)_n microsatellite repeats were first isolated from Actinidia chinensis (2n = 2x = 58) enriched genomic libraries and tested for segregation in the interspecific cross between the diploid distantly related species A. chinensis and A. callosa. Some 105 microsatellite loci of the 251 initially tested segregated in the progeny in a 1:1 ratio as in a classical backcross, or in a ratio which could mimic the backcross, and were mapped using 94 individuals. AFLP markers were then produced using MseI and EcoRI restriction enzymes and 15 primer combinations. Nearly 10% of loci showed a distorted segregation at α = 0.05, and only 4% at α = 0.01, irrespectively to the marker class. Two linkage maps were produced, one for each parent. The female map had 203 loci, of which 160 (71 SSR and 89 AFLP) constituted the framework map at a LOD score ≥ 2.0. The map was 1,758.5 cM(K) long, covering 46% of the estimated genome length. The male map had only 143 loci, of which 116 (28 SSR, 87 AFLP and the sex determinant) constituted the framework map. The map length was only 1,104.1 cM(K), covering 34% of the estimate genome length. Only 35 SSR loci were mapped in the male parent because 18% of SSR loci that were characterised did not amplify in A. callosa, and 48% were homozygous. The choice of parents in the pseudo-testcross is critically discussed. The sex determinant was mapped in A. callosa. Received: 27 July 2000 / Accepted: 31 October 2000     

Chromosome isolation by flow sorting in Aegilops umbellulata and Ae. comosa and their allotetraploid hybrids Ae. biuncialis and Ae. geniculata

This study evaluates the potential of flow cytometry for chromosome sorting in two wild diploid wheats Aegilops umbellulata and Ae. comosa and their natural allotetraploid hybrids Ae. biuncialis and Ae. geniculata. Flow karyotypes obtained after the analysis of DAPI-stained chromosomes were characterized and content of chromosome peaks was determined. Peaks of chromosome 1U could be discriminated in flow karyotypes of Ae. umbellulata and Ae. biuncialis and the chromosome could be sorted with purities exceeding 95%. The remaining chromosomes formed composite peaks and could be sorted in groups of two to four. Twenty four wheat SSR markers were tested for their position on chromosomes of Ae. umbellulata and Ae. comosa using PCR on DNA amplified from flow-sorted chromosomes and genomic DNA of wheat-Ae. geniculata addition lines, respectively. Six SSR markers were located on particular Aegilops chromosomes using sorted chromosomes, thus confirming the usefulness of this approach for physical mapping. The SSR markers are suitable for marker assisted selection of wheat-Aegilops introgression lines. The results obtained in this work provide new opportunities for dissecting genomes of wild relatives of wheat with the aim to assist in alien gene transfer and discovery of novel genes for wheat improvement.     

In Vitro Induction of Haploid, Diploid, and Triploid Androgenic Embryoids and Plantlets in Datura metal L.

Pollen plantlets of Datura metel L. have been successfully rearedin vitro through anther culture on a nutrient medium supplementedwith coconut milk (15 per cent v/v) The early segmentation patternof the proembryos and their subsequent differentiation intoheart, torpedo, and cotyledonary stages were traced. More than100 plantlets were grown to maturity on soil. They grew normallyand produced flowers. Root-tip and leaf-squash preparationsshowed variable numbers of chromosomes in their cells, and thisindicated the occurrence of ploidy plants among the population.Frequency determinations showed the predominant types to bediploid (70 per cent), triploid (24 per cent), and haploid (6per cent). Differences in the morphology of mature plants inregard to floral size, meiotic irregularities in PMCs, formationof giant pollen grains, differential fruitset, somatic mosaicismin respect of chromosome number, and aneuploidy of cells, wererecorded. The origin of diploid and triploid plants is discussed.     

Improvement in efficiency of microspore culture to produce doubled haploid canola (<Emphasis Type="Italic">Brassica napus</Emphasis> L.) by flow cytometry

The efficiency of production of doubled haploid plants in canola (Brassica napus L.) breeding programmes is reduced when large numbers of haploid and infertile plants survive until flowering. We used flow cytometry to determine ploidy status and predict subsequent fertility of microspore-derived plantlets from three canola genotypes, with or without colchicine treatment of microspore suspensions. Young leaf tissue was sampled from microspore-derived plantlets within 1 week of transfer to soil, and processed immediately by flow cytometry. The process was repeated on the same plants 3–5 weeks later. Of the 519 plants transferred to soil, 57.2% were consistently haploid at both sample times, 33.5% were consistently diploid at both sample times, and the remainder (9.2%) were uncertain or inconsistent in ploidy status across sampling times. Of the 518 plants that survived to flowering, 32.4% were diploid at both times of sampling and fertile (set seed) and 46.3% were haploid at both sampling times and infertile. Another 10.8% were haploid at both sampling times and fertile, but had low pollen viability and seed set, and some were triploid or of uncertain ploidy level. Colchicine treatment of microspore suspensions significantly increased the proportion of diploid plants from 9.7 to 69.7%, with significant variation among genotypes. Evidence from simple sequence repeat marker loci indicated that diploid and fertile plants from the control treatment (no colchicine) were derived from spontaneously doubled haploid gametes, rather than unreduced gametes or somatic tissue. Flow cytometry at the first sample time was very efficient in detecting diploid plants of which 94.2% were subsequently fertile.     

Advanced backcross QTL analysis in barley (Hordeum vulgare L.)

This paper reports on the first advanced backcross-QTL (quantitative trait locus) project which utilizes spring barley as a model. A BC(2)F(2) population was derived from the initial cross Apex ( Hordeum vulgare ssp. vulgare, hereafter abbreviated with Hv) x ISR101-23 ( H. v. ssp. spontaneum, hereafter abbreviated with Hsp). Altogether 136 BC(2)F(2) individuals were genotyped with 45 SSR (simple sequence repeat) markers. Subsequently, field data for 136 BC(2)F(2) families were collected for 13 quantitative traits measured in a maximum of six environments. QTLs were detected by means of a two-factorial ANOVA with a significance level of P < 0.01 for a marker main effect and a marker x environment (M x E) interaction, respectively. Among 585 marker x trait combinations tested, 86 putative QTLs were identified. At 64 putative QTLs, the marker main effect and at 27 putative QTLs, the M x E interaction were significant. In five cases, both effects were significant. Among the putative QTLs, 29 (34%) favorable effects were identified from the exotic parent. At these marker loci the homozygous Hsp genotype was associated with an improvement of the trait compared to the homozygous Hv genotype. In one case, the Hsp allele was associated with a yield increase of 7.7% averaged across the six environments tested. A yield QTL in the same chromosomal region was already reported in earlier barley QTL studies.     

Rapid attainment of a doubled haploid line from transgenic maize (<Emphasis Type="Italic">Zea mays</Emphasis> L.) plants by means of anther culture

Summary We present a strategy for establishing a transgenic doubled haploid maize line from heterozygous transgenic material by means of anther culture. Compared to conventional inbreeding, the in vitro androgenesis technique enables a faster generation of virtually fully homozygous lines. Since the androgenic response is highly genotype-dependent, we crossed transgenic, non-androgenic plants carrying a herbicide resistance marker gene (pat, encoding for phosphinothricin acetyl transferase) with a highly androgenic genotype. The transgenic progenies were used as donor plants for anther culture. One transgenic and three non-transgenic doubled haploid lines have been established within approximately 1 yr. The homozygosity of all four doubled haploid lines was tested by analysis of simple sequence repeat (SSR) markers at 19 different loci. Polymorphisms were found between the lines but not within the lines indicating the homozygous nature of the entire plant genome gained by anther culture. Southern blot analysis revealed that the transgenic donor plants and their doubled haploid progeny exhibited the same integration pattern of the pat gene. No segregation of the herbicide resistance trait has been observed among the progeny of the transgenic doubled haploid line.     

Optimization of culture conditions for in vitro rooting of argan (Argania spinosa L.)]

The root system produced of in vitro organ plantlets is of poor quality and not efficient for the transfer to out-door conditions. To overcome such problems, experimentation was undertaken where the effects of growth regulators, nitrogen, sugar, activated charcoal and coconut fiber were tested on root induction and elongation. Modified Murashige and Skoog with half strength salt was used as a basal medium. Root induction (85%) with a mean of 16 roots per explant was obtained when shoots were grown, under dark conditions for 14 days, with a combination of two auxins (IBA and NNA), added at equal concentrations (5 mg.L-1). Secondary roots, 10 cm long, were initiated in 12% of the cultures in presence of 5 g.L-1 activated charcoal. Further improvements in the growth of the primary and secondary roots were obtained when semi-solid medium was substituted with a substrate composed of coconut fibers (80 g) mixed with semi-solid medium (35 mL) and agar (2.5 g.L-1).     

In vitro production of haploid and doubled haploid plants from pollinated ovaries of maize (Zea mays)

Haploid induction has potential application for maize breeding. This paper reports that maize haploid plants have been induced by in vitro culture of pollinated ovaries. From a total of 26,400 cultured ovaries, 24 haploid plants were obtained and two of them were doubled after colchicine treatment. The maximum frequency of gynogenesis was 0.17% at 19.5 h post-pollination (HPP). The results showed that HPP was an important factor affecting plant induction from ovaries. Regenerated diploid R₀ plants were then subjected to genetic analysis using SSR molecular markers. One R₀ plant, whose progeny revealed a high level of homogeneity for several agro-morphological traits, was homozygous at 20 loci tested, with 11 showing paternal and 9 maternal banding pattern. This demonstrates that it is feasible to induce maize haploid plants by in vitro culture of pollinated ovaries.     

Sequence Divergence of Microsatellites and Phylogeny Analysis in Tetraploid Cotton Species and Their Putative Diploid Ancestors

To determine the level of microsatellite sequence differences and to use the information to construct a phylogenetic relationship for cultivated tetraploid cotton （Gossypium spp.） species and their putative diploid ancestors, 10 genome-derived microsatellite primer pairs were used to amplify eight species, including two tetraploid and six diploid species, in Gossypium. A total of 92 unique amplicons were resolved using polyacrylamide gel electrophoresis. Each amplicon was cloned, sequenced, and analyzed using standard phylogenetic software. Allelic diversities were caused mostly by changes in the number of simple sequence repeat （SSR） motif repeats and only a small proportion resulted from interruption of the SSR motif within the locus for the same genome. The frequency of base substitutions was 0.5%-1.0% in different genomes, with only few indels found. Based on the combined 10 SSR flanking sequence data, the homology of A-genome diploid species averaged 98.9%, even though most of the amplicons were of the same size, and the sequence homology between G. gossypioides （Ulbr.） Standl. and three other D-genome species （G. raimondii Ulbr., G. davidsonii Kell., and G. thurberi Tod.） was 98.5%, 98.6%, and 98.5%, respectively. Phylogenetic trees of the two allotetraploid species and their putative diploid progenitors showed that homoelogous sequences from the A- and D-subgenome were still present in the polyploid subgenomes and they evolved independently. Meanwhile, homoelogous sequence interaction that duplicated loci in the polyploid subgenomes became phylogenetic sisters was also found in the evolutionary history of tetraploid cotton species. The results of the present study suggest that evaluation of SSR variation at the sequence level can be effective in exploring the evolutionary relationships among Gossypuim species.