Developmental biochemistry of cotton seed embryogenesis and germination: VII. Characterization of the cotton genome

Abstract

The DNA of cotton, Gossypium hirsutum, has been characterized as to spectral characteristics, buoyant density in CsCl, base composition, and genetic complexity. The haploid genome size is found to bo 0.795 pg DNA/cell. However, the amount of DNA per cell in the cotyledons increases during embryogenesis to an average ploidy level of 12 N in the mature seed cotyledons. Reassociation kinetics indicate that this increase is due to endoreduplication of the entire genome. Non-repetitive deoxynucleotide sequences account for approximately 60.5% of the cotton genome ( C 0t 1 2 pure ¶ ¶ Abbreviations used: C 0t , product of DNA concentration (mol nucleotido l −1) × time(s); C 0t 1 2 , that C 0t value at which one-half of a kinetic component of DNA will have reassociated; C 0t 1 2 pure, that C 0t value at which one-half of a kinetic component of DNA would have reassociated if only that component were present, calculated as the product of C 0t 1 2 × the fraction of the genome in that kinetic component; 5 MeC, 5-methylcytosine; rDNA, ribosomal RNA cistrons. = 437); highly repetitive sequences (> 10,000 repetition frequency) constitute about 7.7% of the genome. ( C 0t 1 2 pure = 4.6 × 10 −4 ) and intermediately repetitive sequences constitute the remaining 27% of the genome ( C 0t 1 2 pure = 1.46 ). Hybridization of 125I-labeled cytoplasmic ribosomal RNA to whole-cell DNA on filters and in solution indicate approximately 300 to 350 copies of the rRNA cistrons per haploid genome. The interspersion of repetitive sequences that reassociate between C 0t values of 0.1 and 50 with non-repetitive sequences of the cotton genome has been examined by determining the reassociation kinetics of DNA of varying fragment lengths and by the electron microscopy of reassociated molecules. About 60% of the genome consists of non-repetitive regions that average 1800 base-pairs interpersed with repetitive sequences that average 1250 base-pairs. Approximately 20% of the genome may be involved in a longer period interspersion pattern containing non-repetitive sequences of approximately 4000 base-pairs between repetitive sequences. Most of the individual sequences of the interspersed repetitive component are much smaller than the mass average size, containing between 200 and 800 base-pairs. Sequence divergence is evident among the members of this component. Highly repetitive sequence elements that are reassociated by a C 0t value of 0.1 average 2500 base-pairs in length, appear to have highly divergent regions and do not appear to be highly clustered. A portion of this highly repetitive component reassociates by C 0t = 10 −4 , zero-time binding DNA, and accounts for less than 3% of the genome. At least a third of these sequences appear by electron microscopy to be intramolecular duplexes (palindromes) of 150 to 200 base-pairs and to occur in clusters.