BANANASTOCK
By the late 1960s, many investigators had observed unknown filaments in developing muscle cells that did not appear to be either actin or myosin filaments — the two major cytoskeletal elements in muscle. In 1968, on the basis of electron-microscopy studies of cultured differentiating muscle and fibroblast cells, Holtzer and colleagues reported a third type of filament. The key experiment was to treat dividing cells, which have few identifiable actin microfilaments, with mitotic inhibitors to break down spindle microtubules. The remaining predominant filament type was novel with a diameter (10 nm) in between that of actin (
6 nm) and myosin (15 nm) — hence, the name intermediate filaments (IFs).
The Holtzer team recognized that the 10-nm filaments might, in fact, represent a heterogeneous class, which turned out to be correct (see Further reading). Yet, how were these histologically diverse IFs unified at the molecular level? In the early 1980s, Weber and Geisler sequenced several IFs and discovered that they share a conserved structural domain that forms a double-stranded coiled coil of 
-helices. The variety in size was shown to arise from the N-terminal and C-terminal extensions flanking the coiled-coil structure.
Interestingly, the name "intermediate filaments" was given to these filaments by Holtzer because they were intermediate in diameter between actin filaments and myosin filaments, not microtubules, as it has come to be known.
Gregg Gundersen
Insight into the possible roles of IFs came from several studies, which showed that IFs are also present in the nucleus. Initially, Gerace, Blum and Blobel showed that the three predominant polypeptides present in a fraction from rat liver nuclei localize exclusively at the nuclear periphery and coincide with the nuclear lamina — a protein meshwork that underlies the inner nuclear membrane and is associated with nuclear pore complexes. They also noticed that, concomitant with the disassembly of the nuclear envelope in prophase, the major lamina polypeptides (or lamins) become dispersed, until telophase, when the nuclear envelope reassembles. In a follow-up study, Gerace and Blobel showed that the disassembly of the nuclear lamina results from the reversible depolymerization of the lamins, which is correlated with their reversible phosphorylation. In 1986, Aebi and colleagues reported the structural and assembly properties of lamins, and confirmed by sequence analysis that they are in fact a type of IF.
Together, these discoveries marked the identification of the third cytoskeletal filamentous system.
