The Mass Question from Nature 27 February 1902

Home , Edward Witten

news and views in cloned mammals, this is a valid concern. in treating human diseases; the idea is to use Finally, the number of live cloned offspring some of a patient’s nuclei to produce geneti- produced per number of nuclear transfers cally identical early embryos, from which ES was pitifully small — approaching one in a cells would be generated and used to grow thousand. So we are still left wondering healthy replacement tissues in vitro. But whether the nuclei of differentiated cells can unless there is a real breakthrough in finding only be reprogrammed under exceptional a source of adult nuclei that can be efficiently circumstances. reprogrammed, all the talk about this ‘thera- 100 YEARS AGO Why do nuclei from different adult tis- peutic cloning’ will come to nothing. In the In reviewing my child’s book, “Beautiful sues vary in the efficiency with which they largest study in mice to date12, only 35 ES-cell Birds”, F. E. B., writing in your columns, can be reprogrammed? Could it be due to lines were generated from over a thousand says, “Why should he select the ‘beautiful whether or not the tissues contain stem nuclear transfers (an efficiency of just 3.4%). birds’ only, and, by implication, condone the cells? That remains to be seen. The idea that This will not be acceptable in humans, where massacre of birds that have not that stem-cell nuclei may be more easily repro- eggs will be hard to come by. Reprogram- advantage?” The question is a misstatement grammed came from using ES cells — which ming adult cells directly, without an oocyte of fact, which I hope you will allow me to are even more developmentally versatile intermediate, would seem a more viable show, though I can only do so by quoting than adult stem cells — as a source of nuclei alternative. Surely the time has come for the myself. On the last page — which I daresay for nuclear transfer7,8. These studies showed cloners to turn their attention to the molec- F. E. B. did not get to — there is this: that more cloned blastocysts developed to ular mechanisms of nuclear reprogramming “ ‘Mother, promise not to wear any feathers the point of live birth when the source of in the egg, and to use the information to except the beautiful ostrich feathers that you nuclei was ES cells rather than adult cells. enhance the potential of adult cells for use in look so lovely in?’ As soon as she promised, But fewer embryos produced from ES-cell cell-based therapies. ■ then all the beautiful birds in the world (and nuclei reached the blastocyst stage, meaning Janet Rossant is at the Samuel Lunenfeld Research that means all the birds, for all birds are that the overall percentages of live offspring Institute and the Department of Molecular and beautiful) will be saved,” &c. (The italics are per nuclear transfer were not very different. Medical Genetics, University of Toronto, mine). This is the final promise and the goal One problem with ES-cell-derived clones is 600 University Avenue, Toronto, Ontario to which I have been leading. May I ask that their expression of ‘imprinted’ genes — M5G 1X5, Canada. F. E. B. whether, if he wished to arouse a genes that are specifically expressed from e-mail: [email protected] child’s interest and sympathies in any subject, either the maternally derived or the paternally 1. Hochedlinger, K. & Jaenisch, R. Nature 415, 1035–1038 (2002); he would choose the more or less salient derived chromosome, but not from both — online 10 February 2002 (10.1038/nature718). material to do it with? Edmund Selous 9 2. Gurdon, J. B. J. Embryol. Exp. Morphol. 10, 622–640 (1962). is often abnormal. By contrast, a study of 3. Gurdon, J. B. & Laskey, R. A. J. Embryol. Exp. Morphol. cloned mice derived from adult cell nuclei 24, 227–248 (1970). I would commend to Mr. Selous Dr. Samuel showed that several imprinted genes were 4. Wilmut, I., Schnieke, A. E., McWhir, J., Kind, A. J. & Johnson’s sound remark concerning a quite Campbell, K. H. Nature 385, 810–813 (1997). expressed normally. So stem cells from fetal 5. Wakayama, T., Perry, A. C., Zuccotti, M., Johnson, K. R. & analogous statement. An orchard, observed or adult tissues (see, for example, refs 10, 11) Yanagimachi, R. Nature 394, 369–374 (1998). the Doctor, would be properly described as might be a better choice than ES cells for 6. Nagy, A., Rossant, J., Nagy, R., Abramow-Newerly, W. & barren of fruit, even if subsequent research Roder, J. C. Proc. Natl Acad. Sci. USA 90, 8424–8428 (1993). testing the reprogramming potential of 7. Humpherys, D. et al. Science 293, 95–97 (2001). discovered a dozen apples and pears upon stem-cell nuclei. 8. Rideout, W. M. III et al. Nature Genet. 24, 109–110 (2000). two or three trees. Now Mr. Selous’ book is These questions are of fundamental 9. Inoue, K. et al. Science 295, 297 (2002). called “Beautiful Birds.” It is not called 10.Goodell, M. A. et al. Nature Med. 3, 1337–1345 (1997). interest, but of course have practical implica- 11.Toma, J. G. et al. Nature Cell Biol. 3, 778–784 (2001). “Birds.” It is clear, too, what Mr. Selous means tions. One of the potential uses of cloning is 12.Wakayama, T. et al. Science 292, 740–743 (2001). by “beautiful.” His plates and the greater part of his descriptions deal with the Paradiseidæ, Humming Birds, and other birds which High-energy physics everyone calls beautiful. I do not find chapter after chapter relating to partridges, quails, sparrows, and other “plain” birds. F. E. B. The mass question From Nature 27 February 1902. Edward Witten 50 YEARS AGO Do the elementary particles known as neutrinos have mass? Yes, In Nature of January 19, p. 92, a translation according to recent experiments. But how much? A surprising — and was published of resolutions passed at a controversial — result suggests that the answer is not what we thought. conference held in Moscow last June on the theory of chemical structure in organic eutrinos were long believed to be, like a new type of nuclear decay process. If this chemistry. It was stated there that “The photons, massless particles that always somewhat controversial finding holds up, Conference has clearly demonstrated the Ntravel at the speed of light. In the past it implies that the three types of neutrino soundness of the theory of the structure of few years, by studying neutrinos emitted by have almost the same mass, and gives us a organic compounds due to the great Russian the Sun or created by cosmic rays in the window on physics that goes far beyond our scientist, A. M. Butlerov; this theory lies at Earth’s atmosphere, physicists have learned present knowledge. the basis of the whole of modern organic that neutrinos actually have tiny but non-zero To put the mass of the neutrino in con- chemistry”. The theory of resonance or masses, roughly ten million times smaller text, consider the mass of other elementary mesomerism was said to be “directly than the mass of an electron. These masses particles. The electron, for example, is opposed to the basic thesis of Butlerov’s are believed to result from physical processes about 1,800 times lighter than the proton or theory”, and it was condemned as physically occurring at energies well beyond those neutron, and about 200,000 times lighter untenable and sterile. Such sweeping claims of known particle interactions. In Modern than the heaviest known elementary parti- require examination. Physics Letters A, Klapdor-Kleingrothaus cles, which are the W and Z bosons and the From Nature 1 March 1952. and colleagues1 now claim to have observed top quark (Fig. 1). Why these masses vary

NATURE | VOL 415 | 28 FEBRUARY 2002 | www.nature.com 969

before physicists started trying to go beyond Matter Force it. They wanted to build a unified theory that would motivate the existence of elementary Quarks Leptons Bosons particles and forces, rather than just describ- Electron 2,3 Electron Up Down Photon (Electromagnetic force) ing them, as the standard model does . In this 0.0005 neutrino 0.003 0.006 0 < 2 x 10–9 more ambitious framework — optimisti- cally dubbed ‘grand unification’ — lepton- Muon Muon Charm Strange W, Z (Weak force) 0.106 neutrino 1.3 0.1 80.4/91.2 number conservation is not automatic. Thus, < 2 x 10–9 a new perspective emerged4–6: lepton number Tau Tau Top Bottom Gluon should be very nearly conserved in nature (Strong force) 1.777 neutrino 175 4.3 0 because it is exactly conserved in the well- –9 < 2 x 10 tested standard model; but it should be very slightly violated by the effects of grand unification. Figure 1 The standard model of high-energy physics: fundamental particles and their masses (in If lepton number is not conserved, it no GeV c1 2, where c is the speed of light). Leptons and quarks interact through exchange of the particles longer provides a way of distinguishing a associated with three forces (weak, strong and electromagnetic) to form the matter we see around us. neutrino from an antineutrino. They could, The fourth fundamental force, gravity, cannot yet be described within the framework of the standard in fact, be two forms of the same particle. model. Although we do not yet understand why, the matter particles form three ‘families’ in order of This particle has one state that spins one increasing mass. The observation of a rare nuclear decay by Klapdor-Kleingrothaus et al.1 suggests way and another state that spins the other that neutrino masses may not follow this trend, but are in fact similar in value. way (Fig. 2), just like a particle with mass, such as the electron. So if lepton number is so much is a mystery, even in the modern distinguish them by their electric charge. But not conserved, neutrinos could have mass. standard model of elementary particles. By there is another apparently conserved charge But this mass can only be very small, because contrast, until recently, neutrinos seemed to in interactions between elementary parti- it arises from effects that are absent in the be massless, and in the 1950s physicists cles: the lepton number. The electron and standard model. Direct measurement of thought they had worked out why. the neutrino are leptons, and the positron such a small mass is difficult, but studies The key is chirality. In biochemistry, and the antineutrino are antileptons. The of the decay of the tritium nucleus have chirality describes the ‘handedness’ of a number of leptons minus the number of demonstrated7 that one type of neutrino is molecule, which may look different from its antileptons in an interaction is called the lighter than about 2 electron volts. mirror image. A simple molecule such as lepton number. Leptons and antileptons can A more subtle way of looking for neutrino

H2O looks the same as its mirror image, but be created by many processes, such as the mass depends on the fact that there are three a more complex molecule such as dextrose decay of a neutron to a proton, an electron kinds of neutrino: the electron neutrino, the may not. That certain chiral molecules are and an antineutrino. In this example, there muon neutrino and the tau neutrino (which important in biology, and their mirror-image are no leptons at the outset (the neutron is are typically produced alongside electrons, molecules are not, is believed to reflect acci- a ‘baryon’), then one lepton (the electron) muons and tau leptons, respectively). This dents in the evolution of life, rather than any and one antilepton (the antineutrino) are leads to the possibility of an interesting inherent difference between the molecules. created, so the lepton number does not quantum-mechanical effect: while travelling Neutrinos have a similar kind of chirality. change. Indeed, it is conserved in all the usual through a vacuum, one type of neutrino can Elementary particles have an intrinsic elementary particle processes. convert spontaneously into another. This is quantum-mechanical ‘spin’. Most particles The concept of lepton-number conser- known as neutrino oscillation, and can only can spin in a right-handed or left-handed vation was derived from experiment, and happen if neutrinos have mass. sense around their direction of motion, but originally had no theoretical explanation There is now extensive evidence for neutrinos always spin in a left-handed sense behind it. In the 1970s, the newly developed neutrino oscillations, both from neutrinos (Fig. 2). Like chirality in biology, this property standard model of high-energy physics produced by cosmic rays in the Earth’s may conceivably have its origins in a chance offered some insight: given the particles atmosphere8,9 and from neutrinos produced event, in this case an accident of the Big Bang. assumed to exist in the standard model and by the Sun10. (The interpretation in terms Such an intrinsic chirality is impossible for the rules by which it is constructed, it is of neutrino oscillations has resolved a long- particles with mass (because the direction of actually impossible to violate lepton-number standing discrepancy11 between the number spin of a massive particle can be changed conservation. of neutrinos expected from the Sun and the by rotating the particle in its rest frame), so The standard model was barely in place number we actually detect.) In this fast- physicists concluded that neutrinos must moving area, experiment is well ahead of have zero mass. theory, and many important measurements Spin But there is a problem with this argu- a Neutrino are expected in the next few years. The results ment, and it has to do with antimatter. Every so far support the rough range of possible particle of elementary matter has a corre- neutrino masses that arises from grand- sponding antiparticle, with the same mass unification theory. The experiments have but opposite electric charge. For example, b Antineutrino Direction of motion also turned up a surprise: the measured the antiparticle of the electron, e1, is the ‘mixing angles’ (which determine the proba- positron, denoted e& . Similarly, the neutrino bility that neutrinos oscillate from one type has an antiparticle: the antineutrino. The to another) are much larger than theorists antineutrino has the opposite chirality to the Spin generally expected. neutrino — it always spins in a right-handed It seems logical to suspect that neutrino sense around its direction of motion (Fig. 2). Figure 2 Chirality is the spice of life. a, The mass results from the non-conservation of Apart from their chirality, how can you neutrino spins in a left-handed sense around its lepton number. But the neutrino-oscillation tell a neutrino from an antineutrino? They direction of motion; b, the antineutrino spins in measurements alone do not show that lepton are both electrically neutral, so we cannot a right-handed sense. number is not conserved. So can we do this

970 NATURE | VOL 415 | 28 FEBRUARY 2002 | www.nature.com

© 2002 Macmillan Magazines Ltd news and views in some other way? This is what Klapdor- prior to this claim, that a sensitivity 103 or 104 mechanical advantage6 and allows them to Kleingrothaus et al.1 claim to have done, by times greater than that of this experiment maintain similar capacities of force genera- observing the nuclear decay 76Ge➝76Se& 2e1. may be needed to conclusively observe the tion and bone loading. But this only applies This reaction is called neutrinoless double- violation of lepton-number conservation. to animals as large as 300 kg or so. Above this b-decay, as the final state contains two Such sensitivity suggests how difficult, as weight, further changes in muscle mech- electrons (historically known as b-particles) well as how potentially rewarding, future anical advantage are probably limited7, and and no antineutrinos — so the reaction experiments are likely to be. ■ sustaining force capacity for movement at violates the conservation of lepton number Edward Witten is at the School of Natural Sciences, greater speeds becomes a problem. by two units. Taken together with the oscil- Institute for Advanced Study, Olden Lane, So how fast might a 6,000-kg dinosaur lation measurements, and assuming that the Princeton, New Jersey 08540, USA. have moved? Previous estimates of the speed only relevant particles are the three known e-mail: [email protected] and locomotive capacity of dinosaurs and types of neutrino, the new result implies 1. Klapdor-Kleingrothaus, H. V., Dietz, A., Harney, H. L. & other extinct animals have been purely that the three neutrinos have approximately Krivosheina, I. V. Mod. Phys. Lett. A 16, 2409–2420 (2001). qualitative. Some models are based on the 2. Pati, J. & Salam, A. Phys. Rev. D 10, 275–289 (1974). equal masses, probably a few tenths of an 3. Georgi, H. & Glashow, S. Phys. Rev. Lett. 32, 438–441 (1974). limb motion deduced from the step length electron volt. This is a surprising result 4. Yanagida, T. in Proc. Workshop on Unified Theory and Baryon and stride frequency derived from fossilized because other particle families, such as Number in the Universe (eds Sawada, O. & Sugamoto, A.) 95–98 tracks1,2,4,8. However, such estimates depend (KEK, Tsukuba, 1979). quarks and the charged leptons, do not have 5. Gell-Mann, M. et al. in Supergravity (eds van Nieuwenhuysen, P. on assumptions about body mass distri- approximately equal masses (Fig. 1), and it & Freedman, D. Z.) 315–321 (North-Holland, Amsterdam, 1979). bution, limb posture and limb length, and will put a severe constraint on theories of the 6. Weinberg, S. in First Workshop on Grand Unification (eds about kinematic similarities between species. Frampton, P., Glashow, S. L. & Yildiz, A.) 347–362 (Math. Sci. 8 origin of neutrino masses. Press, Brookline, MA, 1980). The data from fossilized tracks uncovered Some caution is called for, however, 7. Bonn, J. et al. (MAINZ collaboration) Nucl. Phys. Proc. Suppl. so far suggest that large bipedal dinosaurs because of the exceptionally difficult nature 91, 273–279 (2001). moved at speeds of less than 5 m s11. But it of the experiment. Criticisms of the assump- 8. Fukuda, S. et al. (SuperKamiokande collaboration) Phys. Rev. may be that tracks left by faster-moving Lett. 85, 3999–4003 (2000). tions made by the authors in analysing the 9. Giacomelli, G. & Giorgini, M. (MACRO collaboration) dinosaurs just haven’t been discovered yet. background and extracting an extremely Preprint hep-ex/0110021 (2001); http://xxx.lanl.gov In their analysis of Tyrannosaurus, small signal have already been offered12,13. At 10.http://www.sns.ias.edu/~jnb/SNexperiments/experiments.html Hutchinson and Garcia5 introduce an 11.Bahcall, J. N. & Davis, R. Jr in Essays in Nuclear Astrophysics any rate, planned future experiments using (eds Barnes, C. A., Clayton, D. D. & Schramm, D. N.) 243–285 approach based on estimates of the mini- 76 much larger quantities of Ge (or similar (Cambridge Univ. Press, 1982). mum muscle mass needed for fast running. nuclei) will achieve much greater sensi- 12.Ferruglio, F., Strumia, A. & Vissani, F. Preprint hep-ph/0201291 First they applied their analysis to alligators (2002); http://xxx.lanl.gov tivity. By extrapolating from the oscillation 13.Aalseth, C. E. et al. Preprint hep-ex/0202018 (2002); and chickens — two living relatives of measurements, many physicists have guessed, http://xxx.lanl.gov bipedal dinosaurs. The results show that alligators have less than half the muscle mass that they would need to run fast (if, like Biomechanics bipedal dinosaurs, they used only their hind limbs), whereas chickens have nearly twice the necessary hind-limb muscle mass. Walking with tyrannosaurs This agrees with the observed fact that chick- Andrew A. Biewener ens and many other avian bipeds are good runners, but alligators must support them- Tyrannosaurus terrorized the Earth — at least in the Hollywood version of selves on four limbs and move at relatively history. But an estimate of the muscle volume in its hind legs suggests that modest speeds. the mighty giant could only walk, not run. Hutchinson and Garcia then extended their analysis to estimate the limb muscle ver the course of history, vertebrates enough leg muscles to produce the forces mass of extinct animals and quantify their have evolved an enormous range of required for an animal of such size to run. locomotive performance. From fossil speci- Osizes, spanning well over six orders of The skeletal muscles in all animals are mens of Tyrannosaurus, the authors esti- magnitude in body mass. The largest and made of the same contractile proteins, so mated body and segment mass, worked out most captivating terrestrial giants were the their intrinsic capacity for generating force areas of muscle attachment, and deduced dinosaurs, and Tyrannosaurus — although is very nearly the same. The force that can the forces and moments that the creature’s not the largest at around 6,000 kg — is be produced depends on the cross-sectional leg muscles could have generated. Their perhaps the most famous and terrifying area of a muscle’s fibres. But as body size analysis rests on assumptions about the limb representative of this group. Some workers1,2 increases, the geometrical effects of scale posture and the magnitude of reaction forces have argued that bipedal tyrannosaurs and mean that muscle capability does not exerted by the ground on the limbs of other huge dinosaurs could not move fast increase proportionately. The force that a Tyrannosaurus, and about the kinematic because their size would have imposed severe muscle can generate increases less rapidly similarity between dinosaurs and living constraints on physiological and mechanical than body weight, so, despite their greater birds and mammals9. But their results show functions. But others claim that these volume, the muscles of larger animals gener- that, even if the creature used all its hind- creatures were much more athletic3,4. ate less force per unit weight. limb muscle mass, it could not have generated An obvious difficulty in resolving this In addition, the ability of an animal’s the forces necessary for running. They show argument is that dinosaurs have been extinct skeleton to support mechanical loads that for a chicken scaled up to 6,000 kg to run, for a long time, so reconstructing how they decreases with size because bone area does not it would need muscles in each leg equivalent moved is a challenge. But on page 1018 of this increase nearly as fast as an animal’s weight. to 99% of its body mass — which is obviously issue5, Hutchinson and Garcia introduce a Living terrestrial mammals can accommo- impossible. The results for smaller bipeds, new biomechanical approach to the prob- date these problems of scale by altering their however, show they probably could run lem, applying an analysis of living animals to limb posture when they run: larger animals quickly, in agreement with estimates of their their ancient dinosaur relative. They show run on more erect limbs than much smaller speeds from fossil tracks8. that Tyrannosaurus simply did not have large animals, which gives their muscles greater A pleasing aspect of Hutchinson and

NATURE | VOL 415 | 28 FEBRUARY 2002 | www.nature.com 971