Abstract
The nucleotide sequence of the DNA of maize streak virus (MSV) has been determined. The data were accommodated into one DNA circle of 2687 nucleotides, in contrast to previously characterised geminiviruses which have been shown to possess two circles of DNA. Comparison of the nucleotide sequences of the DNA of MSV with those of cassava latent virus (CLV) and tomato golden mosaic virus (TGMV) showed no detectable homology. Analysis of open reading frames revealed seven potential coding regions for proteins of mol. wt. greater than or equal to 10 000, three in the viral (+) sense and four in the complementary (-) sense. The position of likely transcription signals on the MSV DNA sequence would suggest a bidirectional strategy of transcription as proposed for CLV and TGMV. Nine inverted repeat sequences which have a potential of forming hairpin structures of delta G greater than or equal to -14 kcal/mol have been detected. Three of these hairpin structures are in non-coding regions and could be involved in the regulation of transcription and/or replication.