Codon usage in Chlamydia trachomatis is the result of strand-specific mutational biases and a complex pattern of selective forces

The patterns of synonymous codon choices of the completely sequenced genome of the bacterium Chlamydia trachomatis were analysed. We found that the mo st important source of variation among the genes results from whether the s equence is located on the leading or lagging strand of replication, resulti ng in an over representation of G or C, respectively, This can be explained by different mutational biases associated to the different enzymes that re plicate each strand. Next we found that most highly expressed sequences are located on the leading strand of replication, From this result, replicatio nal-transcriptional selection can be invoked. Then, when the genes located on the leading strand are studied separately, the correspondence analysis d etects a principal trend which discriminates between lowly and highly expre ssed sequences, the latter displaying a different codon usage pattern than the former, suggesting selection for translation, which is reinforced by th e fact that Ks values between orthologous sequences from C,trachomatis and Chlamydia pneumoniae are much smaller in highly expressed genes. Finally, s ynonymous codon choices appear to be influenced by the hydropathy of each e ncoded protein and by the degree of amino acid conservation. Therefore, syn onymous codon usage in C. trachomatis seems to be the result of a very comp lex balance among different factors, which rises the problem of whether the forces driving codon usage patterns among microorganisms are rather more c omplex than generally accepted.