© 2006 Nature Publishing Group

NEWS & VIEWS NATURE|Vol 439|2 February 2006

542

planetoids that populate this region of theSolar System — known as the scattered Kuiperdisk — in that its orbit is highly elliptical, witha distance of closest approach to the Sun (peri-helion) near the orbit of Neptune. This type oforbit is unlike that of Pluto or other objects inthe nearer, classical Kuiper belt, but is com-mon to many trans-neptunian objects. Suchbodies keep to less-elliptical orbits just beyondNeptune, lying 35 to 50 times farther from theSun than Earth is. Like Pluto, 2003 UB313 has asatellite. Physical observations3 of the surfaceof 2003 UB313 have also shown it to be remark-ably similar both to Pluto and to Neptune’ssatellite Triton (thought to be a capturedKuiper-belt object), both of which have highalbedos and methane ice on their surface.

These observations lead inevitably to thequestion of whether 2003 UB313 itself qualifiesas a planet. The minimum size required forthis has never been strictly defined. Until the nineteenth century, the term ‘planet’ wasuncontroversial: a planet was any object thathad a fairly circular orbit around the Sun, andthat did not show any cometary activity suchas a coma or a tail. By 1807, the word encom-passed Ceres, Pallas, Juno and Vesta, which arein the belt of objects between Mars and Jupiter.In the mid-nineteenth century, the growingnumber of such objects being discoveredcaused the term ‘asteroid’ or ‘minor planet’ to be coined. A similar situation is now loom-ing with Pluto, 2003 UB313 and the rest of the trans-neptunian objects. Although thelargest of these are much bigger than thelargest main-belt asteroids, they also form anensemble of bodies of probably very similarformation and evolutionary history.

Whichever way you care to count them, withthe discovery and measurement of the size of 2003 UB313 there are no longer nine majorplanets in the Solar System. A committee hasbeen formed by the International Astronomi-cal Union to mull over several competing defi-nitions of the term. One fairly simple solutionis that anything orbiting the Sun and largerthan a certain minimum size counts as a planet.The size is likely to be arbitrary: a round-num-ber radius of 1,000 km, or perhaps the size ofPluto itself. These bounds stem from a per-ceived need to maintain Pluto’s ‘traditional’ status as a planet, and would leave us with ten current planets, including 2003 UB313.

A second possible definition is that anythingwith a high enough mass to be spherical is aplanet. Although this is a physical criterion,determining whether or not an object meets it is complicated. It would also increase thenumber of known planets by several factors. A third alternative is that any object that has a unique orbit (which means that it is sufficiently far removed from other orbitingbodies), and that dominates its local environ-ment gravitationally, should be known as aplanet. This definition has the most solid sci-entific basis, as objects with similar formationand evolutionary histories would be grouped

together. It would leave us with eight planets:Pluto would not fulfil the criteria.

A final proposal for what constitutes a planettakes into account the fact that the currentlyknown major planets are themselves remark-ably different, and can be split into sub-categories. Mercury, Venus, Earth and Mars, forexample, are terrestrial planets whose composi-tions are dominated by rock. Jupiter and Saturnare gas giant planets dominated by their hydro-gen and helium envelopes. Uranus and Nep-tune are ice giant planets, dominated by gasesother than hydrogen and helium. The trans-neptunian objects, or ice dwarf planets, areprobably composed of large amounts of volatilesolids such as methane ice and water ice.

So ‘planet’ is a term, seemingly, that can bebent to any number of uses. But whatever theplanet-definition committee decides will be

secondary to the naked fact of our rapidlyexpanding scientific knowledge and under-standing of the Solar System. With furtherimprovements in instrumentation and soft-ware, there may be further surprises in store.■

Scott S. Sheppard is in the Department ofTerrestrial Magnetism, Carnegie Institution ofWashington, 5241 Broad Branch Road NW,Washington, DC 20015, USA.e-mail: sheppard@dtm.ciw.edu

1. Bertoldi, F., Altenhoff, W., Weiss, A., Menten, K. M. &Thum, C. Nature 439, 563–564 (2006).

2. Jewitt, D., Aussel, H. & Evans, A. Nature 411, 446–447 (2001).3. Brown, M., Trujillo, C. & Rabinowitz, D. Astrophys. J. 635,

L97–L100 (2005).4. Altenhoff, W., Bertoldi, F. & Menten, K. Astron. Astrophys.

415, 771–775 (2004).5. Grundy, W., Noll, K. & Stephens, D. Icarus 176, 184–191

(2005).6. Cruikshank, D. P. et al. Astrophys. J. 624, L53–L56 (2005).

MOLECULAR BIOLOGY

Prime-time progressStephen D. Bell

DNA is duplicated within a complex macromolecular machine. Insights intohow replication begins and how this is coordinated with progression of DNAsynthesis come from a diverse range of sources.

DNA is replicated by unzipping the doublehelix to expose the bases that act as a templatefor copying the genetic material. Both strandsof DNA serve as templates, and thus one dou-ble helix becomes two. Conceptually, this is asimple reaction, but the devil — as so often —is in the detail: the process is mediated by amultitude of proteins and turns out to bemechanically complex. A trio of papers in thisissue1–3 have made considerable headway inunderstanding the intricacies of replication.

One level of complexity in the replicationreaction comes from thefact that DNA polymer-ase, the enzyme that syn-thesizes the new DNA,cannot begin a stranditself. Rather, it extends ashort RNA ‘primer’ that isalready bound to the tem-plate. This means that anadditional enzyme, a pri-mase, is required to gener-ate the primer to start thepolymerase reaction4,5.

A second complexitylies in the fact that the twostrands of the DNA doublehelix are arranged anti-parallel to one another; theopposite directions aretermed 5� to 3� and 3� to 5�(from the positions of carbon atoms in the sugars

that make up the DNA backbone). However,DNA polymerase can synthesize DNA in onlyone direction: 5� to 3� (Fig. 1). So the 3� to 5�template strand — the ‘leading’ strand — canreadily be replicated, but how is the other, ‘lag-ging’ strand copied? This dilemma wasresolved by the discovery that the laggingstrand is replicated discontinuously. It is syn-thesized in short pieces, called Okazaki frag-ments, that are then joined together —essentially, the polymerase takes two ‘steps’ forward and then synthesizes one back. The

Figure 1 | DNA replication. The helicase unzips the double-strandedDNA for replication, making a forked structure. The primase generatesshort strands of RNA that bind to the single-stranded DNA to initiateDNA synthesis by the DNA polymerase. This enzyme can work only inthe 5� to 3� direction, so it replicates the leading strand continuously.Lagging-strand replication is discontinuous, with short Okazakifragments being formed that are later linked together.

5'

3'

3'

5'

5'

Leading strand

Primer

Helicase

DNA polymerase

Primase

Okazaki fragment

Lagging strand

2.2 News & Views 541 NEW MH 30/1/06 10:29 AM Page 542

Nature Publishing Group ©2006

© 2006 Nature Publishing Group

NATURE|Vol 439|2 February 2006 NEWS & VIEWS

543

difference in the synthesis of the two strandsmeans that, in principle, the leading strandrequires only a single priming event, whereasthe lagging strand needs a new primer for eachOkazaki fragment. And all these events mustsomehow be coordinated to produce twodaughter strands at roughly the same rate.

One of the outstanding conundrums con-cerning replication is how the process dealswith broken or damaged DNA, for examplethat generated by ultraviolet radiation, par-ticularly at the leading strand. Replicating thedamage could obviously be harmful for thedaughter cells. So does the replication machin-ery just stall and wait for the damage to befixed, or is there some way to sidestep thedamage and continue without copying thedamaged DNA?

Heller and Marians (page 557)1 addressedthis issue by biochemically mimicking a block-age of the leading-strand template. Their find-ings indicate that a new priming event canoccur downstream of the lesion, so that theportion of the leading strand encompassingthe lesion is not copied and the resulting DNAwill be single-stranded. This would seem tocontradict the prevalent theory that leading-strand synthesis is continuous. However, precisely this phenomenon is observed in bac-terial cells, where ultraviolet irradiation canlead to single-stranded gaps on both leadingand lagging strands. The gap is filled in after-wards using the remaining undamaged strandas a template. So presumably the existence of amechanism for re-priming the leading strandby generating a single-strand gap buys the celladditional time to deal with the damage with-out interrupting the essential process of DNAreplication.

How is the primase delivered to the sitewhere it needs to act? In bacteria, the primaseinteracts physically with the enzyme responsi-ble for unzipping the double helix, the replica-tive helicase. Some bacterial viruses that havetheir own replication machinery take this astep further by having the primase covalentlyattached to the helicase. This union of theunzipping and priming activities provides a mechanism to couple progression of thereplication ‘fork’ with depositing the primase

on the lagging strand. Heller and Marians1

reveal that the helicase is also crucial in re-priming the leading strand when replication isrestarted after DNA damage.

The coupling of helicase and primase pre-sents a potential dilemma. When the primaseis deposited on the lagging strand, it synthe-sizes RNA in the opposite direction to the pro-gression of the fork (Fig. 1). What happens tothe motion of the fork when this happens? Dothe enzymes uncouple, does the helicase runahead, or does the fork stall transiently?

Lee et al. (page 621)2 examine the primingprocess using a model DNA replication system derived from a bacterial virus calledbacteriophage T7. This well-known systemcontains a covalently fused primase–helicasecalled gp4. The authors analyse the primingprocess using a single-molecule approach witha model DNA substrate and three purifiedproteins: gp4, the T7 DNA polymerase and anaccessory factor, thioredoxin.

They find that the rate of leading-strandsynthesis in the absence of lagging-strand synthesis is constant and uninterrupted. How-ever, when ribonucleotides are added to the reaction, permitting the primase to act on the lagging-strand template, the progression ofthe leading strand pauses briefly at definedpositions. These positions correlate well withthe preferred priming sites of the gp4 primase,and each pause lasts about five seconds, ingood agreement with the RNA synthesis rateof gp4 primase. However, in five seconds theleading-strand DNA polymerase can synthe-size about 1,000 bases of DNA. So pausing themachinery prevents an uncoupling of the syn-thesis of the two strands. The stalling mecha-nism remains unknown, although Lee et al.speculate that the primase domain may regulate the DNA-unzipping activity of gp4.Although T7 gp4 has coupled primase andhelicase domains in one protein, the physicalinteraction of primase and helicase seen inbacteria makes it likely that this pausing mightoccur in a broad range of organisms, even if itrequires two proteins.

Does all priming have to be performed by aspecialized primase? Some clues may comefrom the unusual replication process carried

out by ‘selfish DNA’ elements such as bacterio-phage. Some of these elements have their ownDNA replication machinery (such as phage T7,discussed above); but others, such as bacterio-phage M13, co-opt the host cell’s proteins toensure their own replication. In fact, M13 sub-verts the host gene-expression machinery toinitiate replication of its genome. The bacterio-phage uses the bacterial RNA polymerase —usually employed in copying DNA into RNA destined to make protein — to producethe RNA primer for its replication instead.Normally, RNA polymerase peels the RNA offDNA, acting like a wire stripper, resulting infree single-stranded RNA and re-forming dou-ble-stranded DNA. But, to serve as a primer forM13 replication, the RNA produced by thepolymerase must remain paired to DNA.

Severinov and colleagues (page 617)3 haveuncovered the basis of this altered behaviour.The key lies in the fact that the M13 template is a single-stranded DNA. RNA polymerase usually acts on double-stranded DNA, whichenters and leaves the enzyme molecule by specific channels, with only a short regionunwound in the centre of the enzyme (Fig. 2).As the two DNA strands bind back together, a lid-like structure peels the RNA off anddirects it out through a separate exit channel.But, because M13 is single stranded, there is no return to double-stranded DNA and theRNA–DNA hybrid becomes hyper-extended.This seems to force the hybrid molecule toreposition in the downstream channel of theRNA polymerase, with RNA extending beyondthe body of the enzyme ready to act as a primerfor DNA synthesis (Fig. 2). The authors pro-pose that this may be a general property of thereplication of a number of selfish elements in bacteria. Notably, several plant and animalviruses are proposed to extrude single-stranded regions at their DNA replication startsites, so perhaps RNA polymerase primesreplication in some of these situations too.

Thus, bacteria and their viruses employ arange of strategies to effect priming, to couplepriming events to the coordinated progressionof the replication fork, and to ensure that DNA damage need not be an absolute barrierto fork progression. The power of these recon-stituted bacterial and bacteriophage systems isapparent in the exquisite mechanistic detailthat can be gleaned from their analysis. Simi-larly detailed studies of the more complex primases found in higher organisms5 areeagerly anticipated. ■

Stephen D. Bell is in the MRC Cancer Cell Unit,Hutchison MRC Research Centre, Hills Road,Cambridge CB2 2XZ, UK.e-mail: sb419@hutchison-mrc.cam.ac.uk

1. Heller, R. C. & Marians, K. J. Nature 439, 557–562 (2006).2. Lee, J.-B. et al. Nature 439, 621–624 (2006).3. Zenkin, N., Naryshkina, T., Kuznedelov, K. & Severinov, K.

Nature 439, 617–620 (2006).4. Kornberg, A. & Baker, T. A. DNA Replication 2nd edn

(Freeman, New York, 1992).5. Frick, D. N. & Richardson, C. C. Annu. Rev. Biochem. 70,

39–80 (2001).

Figure 2 | Co-opting other enzymes for priming. The bacterial virus bacteriophage M13 replicates by using its host’s RNA polymerase (which usually makes the RNA encoded in genes) to make RNAprimers. Severinov et al.3 find that the single-stranded DNA that makes up the viral genome subvertsthe usual mechanism of the polymerase, which normally deals with double-stranded DNA.

RNA–DNA hybrid

RNA polymerase

Normal transcription

DNADNA

Lid

RNA

M13 priming by RNA polymerase

3'

2.2 News & Views 541 NEW MH 30/1/06 10:29 AM Page 543

Nature Publishing Group ©2006