On October 18, 2011, a symposium was held at Albert Einstein College of Medicine to celebrate the last 50 years of research on cilia that, largely because of the contributions of the speakers presenting (Fig. 1), has led to the excitement captured in the new biology of cilia—a biology that has significant consequences in developmental dynamics. In this review based on my concluding remarks, I attempt to capture the spirit of the presentations for the reader by citing a key reference for each speaker. The excuse for this anniversary event was of my first article on cilia published 50 years ago (Satir, 1961) from the cradle of cell biology—the laboratory of Porter and Palade (Fig. 2)—at Rockefeller Institute for Medical Research (now Rockefeller University). Cell biology was in its infancy (it was also the year of the first meeting of the American Society for Cell Biology), but there was already in England a group under the leadership of Sir James Gray (1928, 1955) interested in how cilia move. Electron microscopy was the new tool with which to study cell organelles. Cilia attracted attention of the early electron microscopists, especially Irene Manton (1952), Keith Porter, Don Fawcett (1954), because of their consistent nano organization, the 9+2 axoneme in motile cilia, that Porter realized was related to the centriole and to 9+0 axoneme of nonmotile modified cilia that functioned as sense receptors in, for example, the crown cells of fishes or the photoreceptors of the human eye (Porter, 1957). The big question was how were the EM images related to the mechanism of ciliary movement? When I began my work, important contributions to this question had already been made by Björn Afzelius (1959) and Ian Gibbons (Gibbons and Grimstone, 1960; Gibbons, 1961). Like James Gray and Ian, I thought that a lot could be learned from studying the cilia of the clam gill and eventually this proved to yield the first proofs of the sliding microtubule mechanisms of the axoneme (Satir, 1965, 1968).
It is important to realize how primitive biochemistry and genetics also were 50 years ago. Gel electrophoresis was just beginning; virtually no protein sequencing was completed and sequencing was long and involved; techniques for nucleic acid sequencing had not yet been invented. Sir John Randall at Kings College, London had the idea that the problem of ciliary motility could be solved by means of a combined mutant and EM analysis with a ciliated cell suitable for genetic analysis—Chlamydomonas (c.f. Randall, 1969)—but before the discovery of the relationship of structure to sliding—including Gibbons' seminal discovery of the axonemal ATPase—dynein (Gibbons, 1963; Gibbons and Rowe, 1965) and confirmation of microtubule sliding in isolated axonemes (Summers and Gibbons, 1971)—this was premature. Clearly it is premature no longer and questions that seemed impossible to answer regarding the motile mechanism are now being attacked and answered by means of Chlamydomonas, in the new biology of cilia, as illustrated by the work and presentation of Win Sale (c.f., Gokhale et al., 2009). Gibbons has continued the exploration of dynein (c.f., Kon et al., 2009). Moreover, the conservation of cilia structure, motile mechanism and protein composition from Chlamydomonas to man has pointed the way to understanding human ciliary biology as illustrated by the work and presentations of George Witman (c.f. Pazour and Witman, 2008), Heymut Omran (c.f. Fliegauf et al., 2007), Daniela Nicastro (cf. Heuser et al., 2009), and Martina Brueckner (c.f. Fakhro et al., 2011).
In the 1960s, a major locus of ciliary biology in the United States was the Department of Zoology at the University of Chicago, where Frank Child had a faculty appointment and I joined him—after a bit Birgit joined me, bringing another ciliated cell—Tetrahymena—from Copenhagen. The group was enhanced by Joel Rosenbaum, then a postdoc; Sid Tamm, Gary Borisy, students; and then Fred Warner and George Witman. Rosenbaum and Child (1967) and colleagues (Rosenbaum et al., 1969) began to study how cilia assembled after amputation in a variety of flagellates and ciliates, with Joel at Yale eventually focusing on Chlamydomonas and a molecular and structural analysis of the intraflagellar transport machinery (c.f., Rosenbaum and Witman, 2002; Pigino et al., 2009) and its ever widening applications (e.g., Finetti et al., 2009).
The first descriptions of nonmotile 9+0 cilia on ordinary vertebrate cells in a variety of unexpected tissues came very shortly after the description of 9+0 modified cilia of sense organs. The term “primary cilium” was applied to these structures by Sorokin (1962) to distinguish them from the multiple motile cilia of respiratory epithelia. From the beginning, people actually studying primary cilia did not think they were vestigial (e.g., Barnes, 1961). The inverse relationship between the presence of primary cilia and cell division was noted early on (Fonte et al., 1971) and many people also suggested that primary cilia were sensory structures (c.f., Wheatley, 1982; Poole et al., 1985). However, rigorous proof of sensory function, even for olfactory cilia, was lacking.
An explanation of why it was so difficult to demonstrate sensory and signaling functions of primary cilia is that when primary cilia were first discovered in EM, the nature of the cell and ciliary membrane was not clear (c.f. Robertson, 1981). From my perspective, only after Branton (1966) used freeze fracture to demonstrate that the membrane had a lipid bilayer substratum into which proteins were inserted did the nature of the membrane come into focus. It was in this context that Bernie Gilula (then my graduate student) and I used freeze fracture (and instruction from Branton's lab) to discover and describe the ciliary necklace (Gilula and Satir, 1972) and Birgit used the technique to describe the dynamic fusion rosette in ciliates (Satir et al., 1973). From this type of information and much more, Singer and Nicolson (1972) constructed the fluid mosaic model of the membrane. Suddenly everyone in cell biology realized that local protein composition of membranes could vary and that a specific part of the cell membrane, such as the ciliary membrane, could have specific transmembrane receptors traversing it.
The elaboration of the studies of 50 years ago has been very rich especially in the areas of ciliary growth, assembly, and signaling. Greg Pazour and colleagues' extrapolation from Chlamydomonas mutants to kidney primary cilia (Pazour et al., 2000) has blossomed into the expanding study of human ciliopathies, ushering in the new biology of cilia, which Pazour continues to exploit (e.g., Keady et al., 2011). The Oak Ridge polycystic kidney transgenic mouse developed by Brad Yoder and others has proven to be an especially valuable tool (Sharma et al., 2008). The study of ciliary signaling pathways (c.f. Goetz and Anderson, 2010; Satir et al., 2010), particularly in fibroblasts (Schneider et al., 2010), cardiogenesis (Clement et al., 2009), and nervous system development and function (Liem et al., 2012) has proceeded quickly and continuously as discussed by Soeren Christensen and Kathryn Anderson.
It is perhaps worthwhile to ask what key questions remain to be answered. Certainly the list is long—for example, we await a more complete understanding of the relationship between cilia and cellular energy metabolism including mTOR (c.f., Boehlke et al., 2010)—but let me highlight two issues of basic cell and ciliary biology that I consider crucial.
First, why is it necessary for specific signaling molecules to be localized to cilia? Clearly, ciliary localization is critical for proper functioning of those molecules—think polycystin, PDGFR alpha, patched, and smoothened in Hh signaling—defective sequestration leads to polycystic disease, defective neurogenesis, and cancer. But why is sequestration of the signal important? Different molecules including different specific receptors and certain transcription factors are sequestered in the cilia of different tissues. Why so? Must strength of the signal be compared with other signals coming from nonciliary receptors? If so, why are different signals chosen for comparison or is there a common final path? Where and how is the comparison done? What are the consequences? Or is there some other reason for ciliary localization that I or we are missing?
Second, what is the relationship of the cilium to cellular coordinates—i.e., anterior/posterior, left/right gradients within cells and tissues? To illustrate this, I return to the ciliated L cells of the mussel gill (Fig. 3) whence this journey began. There is almost a crystalline pattern to the ciliated surface of the four rows of cells, with each cilium essentially surrounded by six microvilli and six other cilia (Fig. 4). The cilia are set in a grid of microtubules above an interweaving f-actin web (Reed et al., 1984), both of which (in Xenopus cilia) are essential to the normal patterning in different ways (Werner et al., 2011). To remind you that because dynein arms always run clockwise around the axoneme when viewed base to tip, the motile ciliary axoneme clearly has a left and right side, separated by the axis passing between the central pair. The left side of the cilium can arbitrarily be defined as the normal direction of active doublets in the effective stroke and hence activity of dynein arms on doublets 9-4. Can this analysis be extended into the basal body? (Note that every basal body like every axoneme in Figures 3 and 4 is oriented in the identical way) or to nonmotile primary cilia where bending is passive? What is the biochemical/structural basis for the determination of left and right in basal bodies/primary cilia? How does the left side of the cilium align with the anterior/posterior and left/right directions of the cell—for example in the kidney? Or in the migrating fibroblast, where the primary cilium points in the direction of migration? Obviously this question relates to the Wnt/PCP signaling system (c.f., Borovina et al., 2010), but the functional pathway remains obscure. I call your attention to the importance of such orientation for the ciliated protista (Bell et al., 2008) but also for the subsequent alignment of the mitotic apparatus in kidney cells (Menezes and Germino, 2009; Delaval et al., 2011) as well as for the activation of signaling at the left side of the embryonic node (Antic et al., 2010; Hashimoto et al., 2010).
We have come a long way in the past 50 years, but the excitement is far from over. The new biology of cilia continues to present fascinating and deep questions in cell and developmental biology. Let us hope to be participants in the search for answers for some time to come.