UNCONTROLLED CELL DIVISION

In the roughly 1 in 1,000 people affected by PKD, multiple fluid-filled cysts develop in their kidneys due to uncontrolled cell proliferation. The end result is often renal failure. Much research in mice and humans has been devoted to understanding this devastating disorder. Yet, the ground-breaking results came from seemingly unrelated work on the unicellular, flagellated green algae, Chlamydomonas.

In the early 1990's, researchers in Joel Rosenbaum's laboratory at Yale University* observed tiny particles moving up and down the length of Chlamydomonas' two flagellae. This was the first observation of IFT, the tightly controlled process by which cilia and flagella are built, maintained, and resorbed when needed. Cilia are formed by addition of large protein complexes to their distal ends, farthest away from the cell body. IFT particles act like tiny molecular trucks, shuttling protein complexes to the distal end and back again, traveling on a continuous track of motor proteins.

Rosenbaum's group soon cloned 18 or so genes involved in IFT. They searched Genbank for homologs. "One in particular stood out," he says. "When this gene was absent in mouse, it had all the characteristics of autosomal dominant PKD." These mice likely lacked the ability to assemble some essential cilium. But while primary cilia were known to exist on kidney cells, "they'd been relegated to the dustbin of being unimportant for half a century," he says. He and collaborators examined kidney tubules in the affected mice using scanning electron microscopy; they found that primary cilia were indeed either stunted or entirely absent.¹

Courtesy William J. Snell and Jummin Pan

A fluorescence micrograph of the biflagellated green alga, Chlamydomonas reinhardtii, shows its ~12 μm-long flagella. Chlamydomonas use these microtubule filled organelles not only for motility, but also for cilia-generated signaling during sensory transduction.

They were puzzled at first because most human cases of PKD are caused by defects in two proteins seemingly unrelated to cilia. Polycystin 1, a receptor-like protein, and polycystin 2, a calcium channel, form a complex in the cell membrane. Greg Pazour and his group at the University of Massachusetts Medical School put the pieces together when they showed that these two proteins normally reside on primary cilia of kidney cells.² If the cilia are defective, these proteins cannot properly localize and are prevented from carrying out their functions.

But these functions are still unknown. Jing Zhou and her group at Harvard Medical School have shown that bending the primary kidney cilium causes an influx of calcium through polycystin 2, suggesting that the cilium acts as a mechanoreceptor.³ This hypothesis is controversial, because a link has yet to be established between calcium influx in the cilium and control of cell division, says Pazour. "Polycystin 2 mutants show no calcium influx when the cilium is bent, but whether that is why you get PKD is still unknown," he says.

It has long been thought that primary cilia might be involved in control of cell division. Studies done in the 1970s showed that ciliary microtubules are formed using the same centrioles that organize the mitotic spindle, and primary cilia are well known to disappear and reappear during different cell cycle phases.

Donald Ingber, a coauthor on the Zhou study and also at Harvard, suggests a mechanism that might link ciliary bending to cell proliferation. Ingber studies cellular response to mechanical forces, particularly during development when cells are experiencing many changes in their external environment. "Distortion of the cytoskeleton activates signaling cascades [that] regulate apoptosis, growth, and differentiation," he says. And while it is a "bigger jump to say that forces on a cilium might affect these processes, the cilium is linked to the cytoskeleton," he adds. This provides a potential connection.

DEVELOPMENTAL DIFFICULTIES

© 2003 Current Opinion

A primary cilium stained green (antibody to acetylated tubulin) in a cultured human retinal pigmented epithelial cell. The actin cytoskeleton appears in red (phalloidin), and the nuclei in blue (DAPI). Scale bar = 5 μm. (From G.J. Pazour, G.B. Witman, Curr Opin Cell Biol, 15:105–10, 2003.)

Other genetic disorders show a connection between ciliary dysfunction and development. Primary ciliary dyskinesia, or PCD, is frequently caused by mutations in the dynein arms of microtubules in motile cilia. Patients with PCD have the defects one might expect given the known functions of motile cilia in human tissues, according to Brian Hackett of Washington University, St. Louis, who studies lung development. Symptoms of PCD include recurrent respiratory infections due to inadequate lung clearance, and fertility problems caused by defective sperm and poorly functioning oviducts.

But these mutations have less obvious effects as well. Patients with a form of PCD known as Kartagener syndrome have a condition called situs inversus, in which the usual left-right positioning of the organs is reversed. According to Hackett, situs inversus is thought to arise very early in development. Around the time of gastrulation, the embryo forms a node, which carries motile cilia. The cilia beat, producing a leftward flow of fluid, which is detected by immotile sensory cilia on the left side, "telling the left side of the embryo that that's the left side," says Hackett.

Other genetic disorders imply a role for primary cilia in development as well. Katsanis and his group study Bardet-Biedl syndrome (BBS), a complex genetic disorder recently found to be ciliary in origin. Patients with BBS show multiple, seemingly unrelated defects, including increased tendencies toward obesity and diabetes, polydactyly (extra fingers or toes), retinal degeneration, kidney malformation, and situs inversus. These patients also have an increase in mental disorders, such as schizophrenia, depression, autism, and learning disorders. BBS is associated with eight distinct genetic loci in humans.

"A year and a half ago, I knew nothing about cilia," says Katsanis. Then his group cloned the BBS8 gene, which they found was homologous to a prokaryotic protein domain known as pilF. PilF is involved in the formation of pili, finger-like appendages of bacteria. Based on this clue, they localized the BBS8 gene product to the basal body, the structure connecting the cilium to the cell body. They and others have now established that the other seven BBS genes also encode proteins involved in ciliary function or biogenesis. "All the pieces fell into place," he says, once they recognized the connection with ciliary dysfunction.

But the implications go beyond this very rare disease. Katsanis emphasizes that the study of BBS has identified a specific cellular defect whose role can now be investigated in more common conditions. "We are studying BBS because it is multisystemic, hitting essentially every system in the human body," says Katsanis. "The same principles are going to be applicable to some experimentally intractable systems, such as diabetes, obesity, and schizophrenia."

The connection between primary cilia and disorders such as obesity is only a matter of speculation for now. Katsanis suggests that a predisposition to gain weight might be caused by ciliary defects in some specific organ. He points out that this organ could be the brain. Obesity might be a "neurodevelopment issue, caused by changes in the wiring of the hypothalamus," he says. Such changes might prevent proper operation of the feedback loop leading to satiation, for example.

The idea that primary cilia might affect brain development is supported by the observation that nearly all brain neurons possess a primary cilium on their cell bodies. He and others have detected changes in brain development in mice with ciliary dysfunction. The prevalence of mental disorders in patients with BBS also lends support. And recent studies show that primary cilia on some neurons carry serotonin and somatostatin receptors, potential pathways for developmental signals.

A CELLULAR ANTENNA

Clearly, ciliary dysfunction can have profound effects on development in complex organisms. Researchers are now looking at cilia with known functions to get clues as to how primary cilia might be involved in development. They are focusing on the fact that many cilia are involved in sensory functions.

In vertebrates, "Cilia are critical for sensory perception; we see, smell, and hear with cilia," says William Snell, who studies the sensory role of the flagellum in Chlamydomonas. Retinal cells contain modified primary cilia, which explains why retinal degeneration develops in some patients with BBS and PCD.

Beds of sensory cilia in the nose play a key role in olfaction. Katsanis and his colleagues recently showed that patients with BBS have impaired ability to smell, a previously unknown symptom. He and Randall Reed, who studies olfaction at Johns Hopkins, plan to exploit the olfactory system as a model for how defects in BBS-linked genes prevent proper ciliary function. They also plan to study the roles of these genes in olfaction.

Both primary and motile cilia are well known to play a sensory role in simpler organisms such as Chlamydomonas and Caenorhabditis elegans. Maureen Barr of the University of Madison-Wisconsin studies sensory aspects of feeding and mating behaviors in male C. elegans. Early on, she had identified a gene called lov-1 (for "location of vulva"). Defects in this gene prevent males from properly sensing the presence of a mating partner, and from completing mating behavior if a partner is found. The lov-1 gene product, LOV-1, was localized to the cell bodies and primary cilia of male-specific sensory neurons. Lov-1 is a homolog of pkd-1, the gene encoding polycystin 1. "At the time it was really surprising and confusing," she says, but now it is understood that both are ciliary proteins.

Polycystin 1 is a very large membrane protein containing more than 4,000 amino acid residues. Besides eleven transmembrane spans, much of the protein (50%–75%) forms a large extracellular domain. Barr hypothesizes that polycystin 1 and LOV-1 are receptors that receive signals from the outside environment. But, "what activates them may be different," she says. "The mating signal could be chemical, or mechanical, or both." And clearly the consequences of signal reception are different in the two systems as well. She plans to take advantage of the well-established knowledge of C. elegans genetics to identify new components in the signaling pathway.

There is no shortage of candidate genes. Cilia are extremely complex organelles, composed of hundreds of proteins. Recently, several groups have compared the genomes of ciliated organisms, such as humans and C. elegans, with those of nonciliated organisms, such as Arabidopsis and Saccharomyces cerevisiae. These groups identified a total of nearly 700 genes associated with the presence of cilia in an organism.⁴⁵ The genes encode not only the proteins that make up the cilium itself, but also IFT components and ciliary signaling pathways.

From the proteomics side, Greg Pazour's lab has identified at least 300 proteins in the Chlamydomonas flagellum using mass spectrometry in an as yet unpublished study. Understanding the functions of these proteins will go a long way towards explaining how IFT works, and elucidating the roles of cilia in cell signaling and development. As Larry Goldstein, who studies motor proteins involved in intracellular transport at University of California, San Diego, puts it, "Cilia are amazing organelles that still have a lot to teach us."

Megan M. Stephan mstephan@the-scientist.com

* The author works in a Yale University laboratory that is unaffiliated with Dr. Rosenbaum's laboratory.