Prokaryotic and eukaryotic integral membrane proteins have similar architecture Rajneesh Kumar Gaur Æ Girija Arun Natekar Received: 7 November 2008 / Accepted: 2 March 2009 / Published online: 8 March 2009 Ó Springer Science+Business Media B.V. 2009 Abstract Integral membrane proteins constitute a major constituent of lipid bilayer both in prokaryotes and eukaryotes. The statistical analysis was carried out to determine the bias in amino acid distribution between prokaryotic and eukaryotic integral membrane proteins (pIntMPs and eIntMPs). Our results indicate that both pIntMPs and eIntMPs demonstrate the striking similarity in amino acid distribution in their transmembrane and extra- membranous region. pIntMPs have relatively greater functional importance for Gly and Asn in comparison to eIntMPs. Keywords Integral membrane protein  Prokaryote  Eukaryote  Transmembrane domain  Extramembrane region  Amino acid composition Abbreviations IntMPs Integral membrane proteins pIntMPs Prokaryotic integral membrane proteins eIntMPs Eukaryotic integral membrane proteins Introduction Integral membrane proteins (IntMPs) execute a number of cellular functions necessary for maintaining the homeostasis of the cell. Membrane proteins constitute about 30% of the entire protein content of the cell and are often very complex [1]. Due to their involvement in a variety of cell functions such as cell signaling, translocation of nutrients and ions, more than half of all membrane proteins are predicted to be pharmaceutical targets [2]. Membrane proteins are classified as peripheral and transmembrane types depending whether they traverse the membrane or not. All membrane protein structures solved to date show that transmembrane domains fold as either single alpha helices, bundles of alpha helices or beta strands assembled in beta barrels [3]. Membrane protein structures available in PDB are very small in number as a result of difficulties in crystallizing them [4]. The number of membrane proteins whose X-ray crystal structures are known is still very small and repre- sents only *0.2% of all solved protein structures [5]. Sequential analysis of databases of experimentally con- firmed membrane proteins has been used for predicting the transmembrane regions of alpha-helices [6] and to some extent the beta strand topology [7]. The well-defined membrane spanning structural elements and their adjoining regions are sufficiently exploited to determine the amino acid bias. As a result of statistical analysis, the positive-inside rule was first established for bacterial integral membrane protein [8]. Positive inside rule formulated on the basis of preferential distribution of charged residues, mainly Lys and Arg, flanking the hydro- phobic transmembrane domain on the cytoplasmic side. Recent comparative analysis of 107 genomes shows that the positive-inside rule holds true in most but not all organisms [8]. Much information is available on the secondary R. K. Gaur (&) Redox Biology Center, N118 Beadle Center, University of Nebraska, 1901 Vine Street, Lincoln, NE 68588-0662, USA e-mail: meetgaur@gmail.com G. A. Natekar Biosystems Laboratory, Chemical Engineering Department, Indian Institute of Technology, Bombay, Powai, Mumbai 400076, India 123 Mol Biol Rep (2010) 37:1247–1251 DOI 10.1007/s11033-009-9497-3