PLASMID21, 113-119(1989) In Viva Assembly of Chromatin on pBR322 Sequences Cloned into Yeast Plasmids FRANCISCO ESTRUCH, Jo& E. P&XZ-ORTfN, EMILIA MATALLANA, AND Lurs FRANCO Department of Biochemistry and Molecular Biology, University of Valencia, E-46100 Burjassot, Valencia, Spain Received October 3, 1988; revised January 16, 1989 In order to study the in vivo assembly of chromatin on prokaryotic DNA templates, we have transformed yeast cells with plasmids pAJ50 and pRB58, which contain pBR322 sequences. In both cases nucleosomes are assembled in vivo on pBR322 DNA, although the nucleosomes arc not homogeneous in sire. To explore whether there is any preference for nucleosome assembly along pBR322 sequences, we have used an indirect end labeling method. The results indicate that most nucleosomes are placed at random on pBR322, although the probability for histone octamers to interact with some short regions is somewhat reduced. These regions coincide with sequences in which the frequency distribution of nucleosomes reconstituted in vitro (E. Caffarelli et al. (1988) Eur. J. Biochem. 171, 497-501) is low. Sequence determinants that direct chromatin assembly in vitro seem thereby to act to some extent in vivo. 0 1989 Academic Press, Inc It has been known for some years that nu- cleosomes can be assembled in vitro by recon- stituting histones with prokaryotic DNA (see, for instance, Ruiz-Car&lo et al., 1979; Stein et al., 1979) and, in some instances, this pro- cess occurs in a nonrandom manner (Chao et al., 1979; Ramsay el al., 1984; McNamara et al., 1986). More recently, Caffarelli et al. (1988) have conducted an elegant study in which nucleosomes were observed by electron microscopy after psoralen crosslinking of minichromosomes reconstituted in vitro on pBR322. The position of nucleosomes was determined relative to the PstI and PvuII sites; as the polarity of the DNA molecules is not apparent under electron microscopy, the true distribution function of nucleosomes on DNA was calculated by Fourier transform of the data. Although a strict nucleosome positioning was not evident, several maxima and minima appear in the distribution function, revealing that interaction between the histone octamer and the DNA sequences is not entirely at random. chromatin structure of yeast centromeres, suggested that nucleosomes, if they occur, may be asynchronously aligned on the pBR322 se- quences of the plasmid pYe(CEN)4 1, in con- trast with those positioned on CEN sequences. To establish whether nucleosomes are as- sembled in vivo on pBR322 and whether their distribution resembles that found in vitro, we have studied the chromatin structure of pBR322 DNA in two yeast plasmids; pAJ50 (Jimenez and Davies, 1980) and pRB58 (Carlson and Botstein, 1982), the former car- rying the yeast LEU2 gene, and the latter con- taining the yeast SUC2 and URA3 genes. The chromatin structure of the SUC2 and LEU2 genes has been studied in our laboratory (Perez-Ortin et al., 1986a, 1987; Martinez- Garcia et al., 1989), and the structure of URA3 chromatin was studied by Thoma (1986). MATERIALS AND METHODS Plasmids On the other hand, the in viva assembly of The plasmids usedin this study were pAJ50 nucleosomes on prokaryotic DNA has re- and pRB58. A map of the relevant regions of ceived little attention. Bloom and Carbon these plasmids is given in Fig. 1. Plasmid (1982), in the course of their studies on the pAJ50 was originally obtained by JimCnez and 113 0147-619X/89 $3.00 Copyright 0 1989 by Academic p, Inc. All rights of reproduction in any form swerved.