SclI: An Abundant Chromosome Scaffold Protein Is a Member of a Family of Putative ATPases with an Unusual Predicted Tertiary Structure

Here, we describe the cloning and characterization of ScII, the second most abundant protein after topoisomerase II, of the chromosome scaffold fraction to be identified. ScII is structurally related to a protein, Smclp, previously found to be required for accurate chromosome segregation in Saccharomyces cerevisiae. SclI and the other members of the emerging family of SMCl-like proteins are likely to be novel ATPases, with NTP-binding A and B sites separated by two lengthy regions predicted to form an a-helical coiled-coil. Analysis of the SclI B site predicted that SclI might use ATP by a mechanism similar to the bacterial recN DNA repair and recombination enzyme. SclI is a mitosis-specific scaffold protein that colocalizes with topoisomerase II in mitotic chromosomes. However, SclI appears not to be associated with the interphase nuclear matrix. ScII might thus play a role in mitotic processes such as chromosome condensation or sister chromatid disjunction, both of which have been previously shown to involve topoisomerase II. I N interphase human cells, chromosomal DNA molecules totaling `o2 m long are packaged into nuclei that are only ̀ ol0 #m in diameter. At mitosis, the chromosomes become further condensed by about fourfold. This overall 10,000-fold compaction of the DNA is accomplished by a hierarchy of DNA and chromatin packaging (Earnshaw, 1991). At the lowest level, the DNA fiber is compacted sixto sevenfold by winding around the histone core of the nucleosome, generating fibers of `ol0 nm diameter (Kornberg, 1974). At the second level, association of histone H1 with the 10-nm fiber causes the fiber to shorten and thicken to `o30 nm in diameter, bringing the overall compaction of the DNA to `o40-fold (Finch and Klug, 1976; Thoma et al., 1979; Horowitz et al., 1994). How the remaining 250-fold compaction of the 30-rim fiber is accomplished remains a matter of active investigation and considerable controversy (Earnshaw, 1991). At present, the most widely accepted model for higher order chromosome structure proposes that the 30-nm fiber is gathered into loops, each containing ,o50-100 kb of DNA, and tethered to nonhistone proteins of the nuclear scaffold or matrix (Laemmli et al., 1978). This model proposes that at the onset of mitosis, the scaffold proteins at the base of the loops associate with one another, thus pulling the chromosomal loop domains closer together. The aggregates of Address all correspondence to William C. Earnshaw, Department of Cell Biology and Anatomy, Johns Hopkins Medical School, 725 North Wolfe Street, Baltimore, MD 21205. Phone: (410) 955-2591. The present address for Edgar R. Wood is Division of Cell Biology, Burroughs Wellcome Co., 3030 Cornwallis Road, Research Triangle Park, NC 27709. chromosomal scaffolding with their associated loops are thought to form either rosettes that coil along a helical path (Comings and Okada, 1971; Rattner and Lin, 1985; Boy de la Tour and Laemmli, 1988) or stack above one another to form minibands (Pienta and Coffey, 1984). Recent microscopy analysis using DNA fluorochromes under conditions where they bind preferentially to AT-rich or GC-rich DNA has suggested that mitotic chromosome arms consist of a more or less tightly coiled axial region of AT-rich DNA with loops of GC-rich DNA protruding from it (Saitoh and Laemmli, 1994). The loop models of chromosome organization all suggest, however, that the chromatin fiber is packed into the final chromosome structure, and special molecules must exist that bind to the chromatin and define the base of each loop domain. At present, both the DNA sequences and polypeptide components that comprise this putative loop-fastener complex are unknown, although candidates for both have been suggested. The polypeptide components have been suggested to be components of the mitotic chromosome scaffold (or nuclear matrix) Adolph et al., 1977a, 1977b; Izaurralde et al., 1989; Zhao et al., 1993). The DNA sequences are known variously as MARs or SARs ~atr ix or scaffold attachment _regions) (Mirkovitch et al., 1984; Gasser et al., 1989). Chromosome scaffold proteins comprise the 5-10% of nonhistone chromosomal proteins that remain insoluble after treatment of isolated metaphase chromosomes with nucleases and subsequent extraction under a variety of conditions, including high salt (2 M NaC1), low ionic strength (dextran sulfate/heparin), or chaotropes (lithium diiodosalicylate) (Lewis and Laemmli, 1982). Although the chromo© The Rockefeller University Press, 0021-9525/94/10/303/16 $2.00 The Journal of Cell Biology, Volume 127, Number 2, October 1994 303-318 303 on A uust 1, 2017 jcb.rress.org D ow nladed fom some scaffold is in reality a biochemical fraction, the term has been widely interpreted as describing a structural network within mitotic chromosomes. This, in part, results from the observation that isolated chromosome scaffolds retain the overall chromosomal morphology, with paired sister chromatids and condensed centromeres (Adolph et al., 1977b; Earnshaw and Laemmli, 1983). However, the role, if any, played by chromosome scaffold proteins in chromosome structure and function remains an important unsolved question. The first chromosome scaffold protein to be conclusively identified was DNA topoisomerase II (Earnshaw et al., 1985; Berrios et al., 1985; Gasser et al., 1986) (initially termed ScI [Lewis and Laemmli, 1982]). This protein turns out to be the major component of the chromosome scaffold fraction (Gasser et al., 1986; Heck and Earnshaw, 1986). Several independent mapping techniques revealed that topoisomerase II is concentrated in the axial region of expanded mitotic chromosomes, and that it is largely absent from the expanded chromosomal loop domains (Earnshaw and Heck, 1985; Gasser et al., 1986). Functional studies support the notion that topoisomerase 1I plays an essential role in mitotic chromosome structure and function both early and late in mitosis. The protein is required for normal chromosome condensation in fission yeast (Uemura et al., 1987), and also for chromosome condensation in vitro when interphase nuclei or naked DNA are added to mitotic extracts prepared from Xenopus eggs (Adachi et al., 1991; Hirano and Mitchison, 1993). The role of topoisomerase II during chromosome condensation is not known. On the one hand, it is possible that the enzyme is simply required to sort out DNA entanglements that impede orderly chromosome condensation. For example, the mitotic condensation process may serve as a rectification mechanism, whereby neighboring chromosomes are untangled from one another so that they can assort independently during mitosis (Holm, 1994). On the other hand, topoisomerase II may actually make a structural contribution to the condensed chromosome (Earnshaw et al., 1985; Gasser et al., 1986; Adachi et al., 1991). This could occur through interactions of the protein with the SAR/MAR sequences that have been proposed to form the base of chromosomal loop domains (Adachi et ai., 1989). The notion that topoisomerase II plays a structural role in chromosomes is controversial, even when results obtained with the same experimental system are compared (Adachi et al., 1989; Hirano and Mitchison, 1993). Genetic analysis in the yeasts has revealed that topoisomerase II is required for disjunction of sister chromatids at anaphase (DiNardo et al., 1984; Holm et al., 1985; Uemura and Yanagida, 1986). This function is also conserved in vertebrates, as shown both by drug treatments of cultured cells (Downes et al., 1991; Clarke et al., 1993) and by analysis of sister chromatid disjunction in Xenopus cell cycle extracts (Shamu and Murray, 1992). It has been speculated that assembly of topoisomerase II into the chromosomal structure might be important for regulation of its action during disjunction of sister chromatids (Earnshaw et al., 1985). These studies of topoisomerase II provided the first concrete evidence that members of the chromosome scaffold fraction actually do play an important role in mitotic chromosome structure and function. However, with the exception of CENP-B (Eamshaw et al., 1984; Earnshaw and Rothfield, 1985), CENP-C (Earnshaw et ai., 1984; Earnshaw and Rothfield, 1985), CENP-E (Yen et al., 1991), and the INCENPs (Cooke et al., 1987), all of which are concentrated in and around the centromere, topoisomerase 1I has remained the only member of this fraction to be characterized. Other abundant members of the fraction, including ScII (135 kD) (Lewis and Laemmli, 1982) and SclII (140 kD), have remained unstudied. In this paper, we provide the first characterization of ScII. We prepared antibodies to chicken ScII and used them to obtain eDNA clones encoding the chicken polypeptide. Like topoisomerase II, ScII is concentrated in the axial region of swollen chromosomes throughout the entire length of the chromosome arms. Biochemical fractionation confirms that ScII is a prominent component of the mitotic chromosome scaffold fraction. However, the protein associates only very loosely with interphase nuclei, with t>95 % leaking out into the cytoplasm during Dounce homogenization. Thus, ScII is not a component of the nuclear matrix. DNA sequence analysis reveals that SclI is a member of an emerging family of proteins with two internal regions of coiled-coil and highly conserved NTP-binding motifs at the amino and carboxy termini. The best characterized member of this family, Smclp, is required for accurate chromosome segregation in the budding yeast Saccharomyces cerevisiae (Strunnikov et al., 1993). Analysis of the deduced polypeptide sequence, together with previous results, suggests that SclI may be a chromosomal enzyme that may function in a complex with topoisomerase II. Materials and Methods Isolation of Chicken Chromosome Scaffold Proteins and Production of Guinea Pig Antibody Mitotic chromosomes were isolated from chicken lymphohlastoid cell line MSB-1 as previously described (Earnshaw et al., 1985). Scaffolds were prepared by subjecting chromosomes to nuclease digestion and 2 M NaCI extraction (Adolph et al., 1977b). The pelleted scaffold fraction was applied onto preparative SDS-PAGE, the gels were stained with Coomassie blue, and a 135-kD hand was excised and used to immunize a guinea pig to produce antiserum 5132 (Earnshaw et al., 1985).