Abstract In this report we review α-satellite DNA (AS) sequence data to support the following proposed sce- nario of AS evolution. Centromeric regions of lower pri- mate chromosomes have solely “old” AS based on type A monomeric units. Type A AS is efficiently homoge- nized throughout the whole genome and is nearly identi- cal in all chromosomes. In the ancestors of great apes, a divergent variant of the type A monomer acquired the ability to bind CENP-B protein and expanded in the old arrays, mixing irregularly with type A. As a result, a new class of monomers, called type B, was formed. The “new” AS families were established by amplification of divergent segments of irregular A-B arrays and spread to many chromosomes before the human-chimpanzee-goril- la split. The new arrays contain regularly alternating monomers of types A and B. New AS is homogenized within an array with little or no homogenization between chromosomes. Most human chromosomes contain only one new array and one or a few old arrays. However, as a rule only new arrays are efficiently homogenized. Apparently, in evolution, after the establishment of the new arrays homogenization in the old arrays stopped. Notably, kinetochore structures marking functional cent- romeres are also usually formed on the new arrays. We propose that homogenization of AS may be limited to ar- rays participating in centromeric function. Introduction Satellite DNAs consist of long tandem arrays of short re- peated sequences that form the centromeric regions of all higher eukaryote chromosomes. They are sometimes also found in subtelomeric or other chromosomal loca- tions. Satellite DNAs are implicated in centromeric func- tions, such as segregation in mitosis and meiosis, recog- nition and pairing of homologous chromosomes, sister chromatid attachment, and formation of kinetochore structures (reviewed in Willard 1998). Notably, satellite DNAs form the most rapidly evolving compartment of the genome. Close species often have non-orthologous satellite DNAs at homologous chromosomal locations (Csink and Henikoff 1998). Alpha-satellite DNA of primates (AS) is by far the most studied satellite DNA species and represents a try- ing ground for all conceptual models (see Willard 1991, 1998 for review). Alpha-satellite DNA is sufficient to form a mitotic centromere in human cells; however, other DNA sequences can also perform this task. In any case, in all human chromosomes AS participates in centromeric function (reviewed in Warburton 1999). In humans almost every chromosome has its own unique family of AS. This has been highly instrumental in the development of “interphase cytogenetics”, making it possible to enumerate homologous chromosomes during interphase and to identify rearranged and small marker chromosomes. One should also expect AS studies to be highly useful for clarifying the events that formed the human karyotype: the composition of centromeric re- gions is one of the major differences between the chro- mosomal maps of man and the great apes. Here we present an overview of AS sequence studies and propose a scenario of AS evolution from lower pri- mates to man. It is based partly on the ideas and observa- tions published by Warburton et al. (1993), Kipling and Warburton (1997), Csink and Henikoff (1998) and Laurent et al. (1999). For this report we have re-typed all AS sequences mentioned in order to bring up to date their original classification into subfamilies by the authors. Edited by: T. Hassold I. Alexandrov ( ✉ ) · A. Kazakov · I. Tumeneva · Y. Yurov Mental Health Research Center, Russian Academy of Medical Sciences, Zagorodnoe sh.2, Moscow 113152, Russia e-mail: ivanalx@hotmail.com V. Shepelev Institute of Molecular Genetics, Russian Academy of Sciences, Moscow, Russia Chromosoma (2001) 110:253–266 DOI 10.1007/s004120100146 ORIGINAL ARTICLE Ivan Alexandrov · Alexei Kazakov · Irina Tumeneva Valery Shepelev · Yuri Yurov Alpha-satellite DNA of primates: old and new families Received: 9 March 2000 / In revised form: 2 February 2001 / Accepted: 13 March 2001 / Published online: 16 May 2001 © Springer-Verlag 2001