news and views

Almost Planet X S. C. Tegler and W. Romanishin

Optical and infrared observations of a bright object in the outer Solar System reveal it to be surprisingly large — almost as big as Pluto’s moon. It could be the first of many such discoveries. hen Neptune, the eighth planet in the Solar System, was discovered in 1846, astronomers began looking for a ninth. Inspired by irregularities in Neptune’s orbit, Percival Lowell began searching in 1905 for what he called Planet X. But it wasn’t until 1930, 14 years after Lowell’s death, that Clyde Tombaugh of the Lowell Observatory in Flagstaff, Arizona, discovered Pluto. The search for Planet X was over. Or was it? Pluto turned out to be much smaller than expected, and some astronomers, including Tombaugh, continued to search for a tenth planet, beyond Pluto. No one has found a tenth planet, but during the past decade astronomers have discovered that Pluto is not alone. It seems that Pluto and its moon Charon are the largest known members of an ancient ring or belt of icy bodies, known as the Kuiper belt1,2. The primitive nature of the material in this belt might hold important clues about the formation and evolution of the Solar System. Unfortunately, apart from Pluto and Charon, these objects are so faint that studying their most basic properties requires a herculean effort. On page 446 of this issue, Jewitt, Aussel and Evans3 present the best measurements yet of the size and reflectivity of another body in the belt. Their work raises the possibility that Pluto is not the only Planet X, but perhaps one of several. The body studied by Jewitt and colleagues, now called Varuna4 (Fig. 1), was detected last November by the Spacewatch telescope in Arizona. It is the third brightest object in the belt, after Pluto and Charon. The optical and infrared measurements made by Jewitt et al. show that Varuna is slightly smaller than Charon, making it the third largest known object in the belt, but with a much darker surface than Charon. The results suggest that Pluto and Charon are not uniquely large objects, and that a continuum of sizes may exist. We can now imagine that bodies even larger and more distant than Pluto will be found. Such objects have so far escaped detection because of their extreme faintness, which is due in part to the feeble illumination from the Sun, and in part to their very dark surfaces. The idea of a belt of icy bodies surrounding the outer planets goes back to 1930, soon after the discovery of Pluto. Leonard5 was the first to publish the idea, which was later expanded upon by Kuiper6 and Edgeworth7.

W

NATURE | VOL 411 | 24 MAY 2001 | www.nature.com

Figure 1 The size and surface darkness of the ninth planet Pluto, its moon Charon, and the newly discovered object Varuna, the three largest known members of an ancient belt of icy objects that surround the eight planets closest to the Sun. Varuna was discovered by the Spacewatch team at the University of Arizona4, who regularly scan the skies for asteroids and comets. Infrared and optical measurements by Jewitt and colleagues3 suggest that Varuna is by far the largest body so far discovered in the belt, apart from Pluto and Charon. This work implies that Pluto and Charon are not unusually large, and that there may be other, as yet undiscovered objects in the outer reaches of the Solar System, perhaps including ‘planets’ even larger than Pluto.

For 60 years, Pluto and its moon Charon were the only known objects in the Kuiper belt. The third was found in 1992 (ref. 1), following advances in telescope and detector technology. There are now nearly 400 known objects in the belt, although there may be hundreds of thousands of 100-km-sized bodies and possibly billions of 10-km-sized objects that have yet to be discovered8. Although there are so many Kuiper belt objects (KBOs), the total mass of the belt is probably only about a tenth the mass of the Earth8. All the evidence to date suggested that Pluto was a uniquely large object within the Kuiper belt. Presently, the best way to estimate the diameter of a KBO is to measure the amount of sunlight it reflects, using an optical telescope. At a given distance, the brighter the object, the larger the surface area facing the Earth. But there is a caveat. The brightness of an object depends not only on the area of the reflecting surface, but also on the albedo, © 2001 Macmillan Magazines Ltd

the percentage of sunlight the surface reflects. Material as black as charcoal has an albedo of about 4%, whereas material as reflective as frost has an albedo of about 40%. So optical brightness measurements alone cannot distinguish between a large, dark object and a small, highly reflective one. Determining independent values for the diameter and albedo requires measuring both the sunlight reflected and the thermal infrared light emitted by the body. This technique has been successfully used to determine the diameter and albedo of asteroids in the inner Solar System. But KBOs are much harder to detect in the infrared because they emit radiation at wavelengths that can be easily absorbed by the Earth’s atmosphere. Some objects, known as Centaurs, are thought to have escaped from the Kuiper belt and can be found throughout the outer Solar System, sometimes passing even closer to the Sun, not far from Earth8. As the Sun warms them, some of the icy 423

news and views material evaporates in a cloud of dust and gas, which is occasionally visible as the tail of a comet. Until now, astronomers have assumed that KBOs have an albedo similar to the closer Centaur objects and comets8, whose albedo is typically 4%. Such an assumption is fraught with danger because the Sun’s warmth may trigger chemical and physical changes in Centaurs that alter their surface reflectivity. Varuna is so bright that Jewitt and colleagues have been able to measure simultaneously its optical reflectivity and infrared emission; they found that it has a diameter of 900 km and an albedo of 7%. Its surface is darker than that of Pluto and Charon, but its albedo is slightly higher than the value previously assumed for KBOs. Only Pluto and its satellite Charon have larger diameters than Varuna (2,400 km and 1,200 km, respectively). So Varuna closes the gap between the largest previously known KBO, which was around 600 km in diameter (assuming an albedo equal to that of Varuna), and the smallest planet in the Solar System, Pluto. Varuna and the handful of Centaurs whose albedos have been measured all have dark surfaces, which tells us there is little surface frost or ice. We already know that Pluto and Charon have very bright surfaces9. Because of its size, Pluto is a very different beast to the other KBOs — it is large enough to retain a tenuous atmosphere, resulting in a global surface frost. Charon, on the other hand, has a similar diameter to Varuna, but its albedo (40%) is about six times that of Varuna. Why do Varuna and Charon have

such different albedos? It could be because the surface of Charon is covered in frozen water10, perhaps a result of whatever process caused it to become a moon of Pluto. More measurements of thermal emission are needed to determine whether dark surfaces are ubiquitous in the Kuiper belt, as well as to obtain reliable diameters. Thermal infrared observations of these faint objects are extraordinarily difficult from the ground, but much easier in space. The Space Infrared Telescope Facility (SIRTF) satellite is expected to measure the diameters and albedos of dozens of KBOs after its launch in 2002. Such observations are our best chance of finding more planets beyond Pluto. ■ S. C. Tegler is in the Department of Physics and Astronomy, Northern Arizona University, Flagstaff, Arizona 86011, USA. e-mail: [email protected] W. Romanishin is in the Department of Physics and Astronomy, University of Oklahoma, Norman, Oklahoma 73019, USA. e-mail: [email protected] 1. 2. 3. 4.

Jewitt, D. & Luu, J. Nature 362, 730–732 (1993). http://cfa-www.harvard.edu/cfa/ps/lists/TNOs.html Jewitt, D., Aussel, H. & Evans, A. Nature 411, 446–447 (2001). McMillan, R. & Larsen, J. Minor Planet Electronic Circular 2000–X02 (Minor Planet Center, Cambridge, MA, 2000). 5. Leonard, F. C. Leaflet Astron. Soc. Pacific 1, 121–124 (1930). 6. Kuiper, G. P. in Astrophysics: A Topical Symposium (ed. Hynek, J.) 357–427 (McGraw-Hill, New York, 1951). 7. Edgeworth, K. E. J. Br. Astron. Assoc. 53, 181–188 (1943). 8. Jewitt, D. C. & Luu, J. X. in Protostars and Planets IV (eds Mannings, V., Boss, A. P. & Russell, S. S.) 1201–1229 (Arizona Univ. Press, Tucson, 2000). 9. Tholen, D. J. & Buie, M. W. in Pluto and Charon (eds Stern, S. A. & Tholen, D. J.) 193–219 (Arizona Univ. Press, Tucson, 1997). 10. Marcialis, R. L., Rieke, G. H. & Lebofsky, L. A. Science 237, 1349–1351 (1987).

Molecular motors

Switching on kinesin Manfred Schliwa and Günther Woehlke A crystal structure helps us to understand how an enzyme works. Better yet are crystal structures of the enzyme in different states of activity, which have now revealed the intricate workings of a molecular motor. wist-off caps are an effective way of sealing bottles. A small rotation of the cap in one direction tightens it, while a small turn in the opposite direction removes it from the bottle. A twist-off cap seems an unlikely model for a molecular motor protein, but, on page 439 of this issue1, Kikkawa and colleagues suggest that a roughly 20° rotation of the motor region of kinesin helps it to tighten its grip on the surface of the filamentous tracks it moves along. This finding was made possible by the crystallization of a motor region in two functionally significant states — one bound to the nucleotide adenosine diphosphate (ADP), and the other in complex with an analogue of adenosine triphosphate (ATP). The ‘ATP-like’ conformation had, until now, resisted all attempts

T

424

to crystallize it. The authors’ success ushers in a new era in the study of kinesin’s mechanics and chemistry. Kinesin motors move in one direction along linear tracks called microtubules by virtue of cyclical changes in the conformation (shape) of kinesin, driven by the hydrolysis of one molecule of ATP (thereby producing ADP) per step. They share this feature with a different type of motor, myosin, which uses actin filaments as a track. These two classes of motor have no apparent similarities in their amino-acid sequences. But their atomic structures, obtained by X-ray crystallography, reveal a remarkable conformational similarity between the central portion of myosin, which includes the nucleotide-binding site, and the ‘head’ © 2001 Macmillan Magazines Ltd

(the nucleotide-binding motor domain) of kinesin2,3. So it had been suspected that these two families of motor proteins operate according to similar principles4. This idea is reinforced by the findings of Kikkawa et al.1. In both myosin and kinesin, the nucleotide-binding site is flanked by two structural elements, referred to as switch I and switch II, by analogy to equivalent loops found in G proteins (signalling proteins that bind to GTP). The crystal structures of myosin and G proteins in different nucleotide-binding states (that is, bound to GTP or GDP for the G proteins, and to ATP or ADP for myosin) have been available for some time. These structures show that significant changes in the switch motifs help to communicate the hydrolysis of ATP or GTP to the rest of the molecule, switching off G proteins and driving the movement of myosin. The structures obtained by Kikkawa et al.1 reveal that equivalent structural permutations are linked to ATP hydrolysis in the switch I and switch II regions of kinesin (Fig. 1). Although not entirely unexpected, it is reassuring to see this expectation fulfilled in actual atomic structures. What are the consequences of the ATPdependent structural changes in the switch regions of kinesin? The most dramatic change occurs in a subdomain of the kinesin head. This subdomain, which the authors refer to as the ‘switch II cluster’, is linked to switch II, and consists of two structural features called Ȋ-helices, and three loops. One of the Ȋ-helices, Ȋ4 (sometimes referred to as the switch II helix), is a prominent component of the microtubule-binding face of the kinesin head5. This helix undergoes nucleotide-dependent changes in length — when the kinesin head is bound to ADP rather than the ATP analogue, the helix is longer by about two turns, at the expense of one of the loops, which shortens. The helix also shows nucleotide-dependent rotation of about 20° relative to the rest of the kinesin head (Fig. 1), causing other elements of the switch II cluster to follow this movement. This rotation, in turn, is communicated to the kinesin neck domains, which previous studies have shown to have a decisive role in the generation of force and movement by kinesin6. These crystal structures were of kinesin alone, not attached to microtubules. With just these structures available, Kikkawa et al. might have proposed that the generation of force by kinesin involves this slight rotation of the Ȋ4 helix relative to the rest of the motor domain and the microtubule. But, because they also obtained cryo-electronmicroscopic images of microtubule-bound kinesin heads in both activity states, a different picture emerges. Such images show lowresolution, three-dimensional contours of the kinesin head in contact with the microtubule surface. The atomic structures can be NATURE | VOL 411 | 24 MAY 2001 | www.nature.com