When William Earnshaw struck out from his postdoc with Ulrich Laemmli for his new lab at Johns Hopkins University (Baltimore, MD) in 1981, he took with him Laemmli's rather controversial idea that a chromosome scaffold of nonhistone proteins might be responsible for the radial loop structure of chromatin. Laemmli's studies of metaphase chromosomes, which had been depleted of histones, supported such a model (Paulson and Laemmli, 1977; Marsden and Laemmli, 1979).
At Hopkins, Earnshaw soon learned, “people used antibodies for everything,” so he followed suit. After eight months spent isolating human chromosomes, digesting the DNA, then extracting away the more soluble proteins, he had three presumptive scaffold protein bands with which to immunize guinea pigs. Paranoid, he spray painted dots on the guinea pigs' backs so as not to lose track of them (it paid off when one ended up in someone else's cage).
He found a protein that reproducibly turned up on mitotic chromosomes and in subcellular fractions thought to hold the scaffold components, but he had no idea what the protein was. A fortunate lunch with Leroy Liu, who worked on topoisomerase 2, an enzyme known to untangle DNA strands by cutting and religation, led to some Western blot swapping that revealed Earnshaw's protein as topo 2 (Earnshaw et al., 1985). It was the first localization of a nonhistone protein to mitotic chromosomes.
Further investigation with his collaborator Margarete Heck localized topo 2 to the base supports of the radial loops of chromatin by immunofluorescence. The antibody did not cause global condensation of chromosomes in vitro as did other DNA-binding antibodies. Most bivalent DNA antibodies would bind along all lengths of chromatin and cause condensation through cross-linking. With anti-topo 2, only very localized condensation was seen along the axial regions of chromosomes, and Earnshaw argued that the protein must be stably localized there (Earnshaw and Heck, 1985).
“We now know that it probably doesn't have a structural role,” says Earnshaw. Instead, topo 2 is required for untangling chromosomes and appears to be involved in compaction, perhaps in concert with the condensin complex. But the work opened the door for the scaffold hypothesis to flourish. At the time, critics said that the scaffold proteins might simply be a precipitation artifact. Now, the scaffold is envisioned as a protein core or network that regulates the higher-order structure of chromosomes—possibly by binding stretches of DNA called scaffold attachment regions that can be up to 100 kb apart. This creates a loop of histone-wound DNA (for review see Swedlow and Hirano, 2003). Recent work from the Earnshaw lab has shown that knocking out a key member of the condensin complexes can abolish the entire chromosome scaffold (Hudson et al., 2003).
An image from the 1985 study became, in January 1986, the first picture ever used on the cover of the JCB. Earnshaw says the paper's most lasting legacy may be the first use of antibodies to study the structure of chromosomes.