New entries can be found at #11 (Y. Kamei, et al., Cell, 85[3]:403-14, 3 May 1996; with 51 citations during May-June 1997) and #14 (V.V. Ogryzko, et al., Cell, 87[5]:953-9, 29 November 1996; with 49 cites this period). These two papers address a fundamental question of molecular biology: how do transcription factors, which switch genes on and off, gain access to the specific portions of DNA whose activity they are supposed to be controlling? The E. coli sequence, at #13, helps put the problem in perspective.

   In the firehose of outpourings about DNA it is easy to lose sight of just what an astounding molecule it is, and how tightly it has to be packed into the cell. Way back in antediluvian 1970, textbooks pointed out that "the average DNA of a bacterium has proportions equivalent to a violin string 1 millimeter across and 400 meters long." The complete sequence of E. coli--the average bacterium referred to--allows us to update that. E. coli DNA is 4,639,221 base pairs long, for a total length of almost 1.6 millimeters. Expand its diameter to the 1 millimeter violin string, and its length reaches 788 meters. That jump from 400 to almost 800 is probably the result of E. coli having far more genes, of far greater complexity, than were imagined in the 1970s, but the cell itself has not grown to accomodate our new estimate of the size of its DNA.

   The packing problem is even more acute in eukaryotes, where there is considerably more non-coding DNA to bundle into the nucleus. One of the packaging units of the chromosome is the nucleosome, a length of DNA wrapped around a core that consists of two copies of each of four different histone molecules. Biochemical studies of active genes suggest that the histone proteins are acetylated, which probably disrupts the structure of the nucleosome and allows transcriptional factors access to the DNA.

   Some transcription regulators turn out to be histone acetyltransferases (HATs), which lends weight to the idea that acetylation is a key step in activation. But they lack crucial characteristics; some attack histones only when free, not when bound up in nucleosomes, while others do not act on all four core histones. The papers at #11 and #14 show that another transcriptional activator is a novel HAT.

   CBP (for CREB binding protein) interacts strongly with phosphorylated cAMP element-binding protein (CREB), while p300 is a closely related molecule with very similar functions. Michael Rosenfeld at the Howard Hughes Medical Institute at the University of California, San Diego, assembled a team to investigate CBP/p300 and discovered that CBP interacts directly with several different nuclear receptors (paper #11). It also binds to a cofactor--an SRC-1 protein--with a suggestion that this cofactor enhances the specificty of the activation. With many related SRC-1 coactivators, each tuned, as it were, to a different gene, the cell would have a mechanism for integrating the demands of a variety of activation signals.

   In paper #14, Yoshihiro Nakatani's group at the National Institute of Child Health and Human Development, Bethesda, Maryland, demonstrated that CBP acetylates all four core histones in the nucleosome, and does so in an entirely novel way. The SRC-1 coactivator is also a HAT. Nakatani's model envisages the two HATs forming a complex on specific promotor elements and then acetylating the histone tails in a manner determined by the promotor. "With two activators, maybe it opens up the DNA better," Nakatani tells Science Watch.