Characteristic Distribution of GC Contents for Proteins with Charge Periodicity of 28 Residues

Nuclear genomes of vertebrates are composed of mosaic structures having different GC content (GC content is the molar ratio of guanine + cytosine in nucleotide sequence). These structures are called isochores, and each segment length is about 300kb [1]. Previous research has shown that different GC region of chromosome is correlated with several properties: the staining intensity of chromosomal bands [2] and gene density in GC-rich regions and GC-poor regions of isochores [3]. Then, most of the genes expressed ubiquitously in various tissues as a housekeeping gene are in the GC-rich regions [4]. It is considered that the isochores are related to the gene expression [3]. In addition, the GC content of gene coding region is related to the GC content of surrounding region of genome, for example GC-rich genes tend to present in GC-rich region of isochores and vice versa [5]. These facts indicate that GC content is one of the most important factors for considering the meaning of genes in terms of the expression in cells. In the accompany paper, we showed that some of proteins coded in the total genome have charge periodicity of 28 residues (PCP28). PCP28 were obtained by using the autocorrelation function of electric charges of amino acid sequences and had the significant peak at 28 residues periodicity in the plot of the function. More PCP28 were found in the eukaryotic genomes than in the prokaryotic genomes. Most of PCP28 are nuclear proteins which play important roles, such as gene regulation, genome replication, and so on. For more understanding of the role of PCP28, we investigated the GC content of PCP28. In this work, we analyzed the human genome focusing on PCP28, GC content, and gene expression. First, we investigated the population of PCP28 from the human genome as a function of the GC content. The result indicated that the GC content for PCP28 located in the nucleus was very low while that for PCP28 in other localization showed broad distribution. The low GC distribution is usually correlated with the low expression level of PCP28 [3]. Therefore, we investigated the expression pattern of PCP28 in several types of cells by using the published human genes expression data of microarray (http://symatlas.gnf.org).