RESEARCH NEWS & VIEWS But how does the heat of reaction enable the enzyme molecule to move? Here, it must be noted that proteins exist in a world in which Brownian motion is governed by viscous forces, rather than by inertia6,7. Coasting is not an option — continuous force generation is required. And not all types of motion can cause translational diffusion6. To explain their observations, Riedel and colleagues develop a simple model whereby the heat generated from each catalytic cycle is transmitted through the enzyme as a pressure wave (Fig. 1). In this model, the active site must be asymmetrically placed — that is, not at the enzyme’s centre of mass. The pressure wave creates differential stress at the enzyme–solvent interface, which in turn propels the enzyme. The authors call this a ‘chemoacoustic’ effect. The model itself is barren of microscopic details and assumes that the enzyme, solvent water and their interface are individually homogeneous. This simple view nevertheless gives reasonable estimates of the force generated at the enzyme– water interface from a modest fluctuation in volume created by the thermal activation of a protein’s motional modes. Over the past few decades it has become abundantly clear that proteins (including enzymes) are dynamic entities rich in motion over a vast range of timescales. The picosecond-to-nanosecond time frame relevant to Riedel and co-workers’ findings is no exception. Sophisticated molecular-dynamics simulations8 suggest that transmission of energy through a protein can be remarkably fast — on the order of 5 ångströms per picosecond (1 picosecond is 10–12 seconds) — and non-uniformly distributed. The complexity of the internal motion of protein molecules has also been exemplified using nuclear magnetic resonance spectroscopy9. This complexity is particularly apparent in the context of applied pressure10, which is highly relevant to the present study. But perhaps most importantly, proteins have elements (often termed ‘foldons’) that cooperatively fold to adopt a particular substructure and dictate many of the kinetic and thermodynamic properties of protein molecules11. Given all this complexity, the precise mechanism by which pressure waves move through protein molecules remains uncertain. What happens when the transient expansion of an enzyme arrives at the enzyme–water interface is also more complicated than the situation in Riedel and colleagues’ simple model. Theory12,13, simulation14 and experiment15 suggest that the interaction of a protein surface with surrounding water molecules is context-dependent and variable. Furthermore, evidence14,16 seems to suggest that proteins can induce long-range ordering of water, beyond the traditional ‘hydration layer’ of water molecules that immediately surrounds a dissolved protein molecule. Clearly there is much to do to fully understand how heat generated by
enzyme catalysis is dissipated and how this can result in a locomotive protein molecule. Finally, the most intriguing question of all is whether the anomalous diffusion of protein molecules through the chemoacoustic effect is a product of evolution or simply an accidental unselected result of heat flow in proteins. It seems reasonable to think that enzymes might have evolved the capacity to seek ‘greener pastures’ of substrates. But this view is more complicated than it might at first look6,7, and any selective advantage may ultimately be quite obscure. ■ A. Joshua Wand is in the Department of Biochemistry and Biophysics, University of Pennsylvania, Philadelphia 19104-6059, USA. e-mail: [email protected] 1. Fischer, E. Ber. Dt. Chem. Ges. 27, 2985–2993 (1894). 2. Riedel, C. et al. Nature 517, 227–230 (2015). 3. Muddana, H. S., Sengupta, S., Mallouk, T. E., Sen, A.
& Butler, P. J. J. Am. Chem. Soc. 132, 2110–2111 (2010). 4. Sengupta, S. et al. J. Am. Chem. Soc. 135, 1406– 1414 (2013). 5. Paxton, W. F., Sundararajan, S., Mallouk, T. E. & Sen, A. Angew. Chem. Int. Edn 45, 5420–5429 (2006). 6. Purcell, E. M. Am. J. Phys. 45, 3–11 (1977). 7. Golestanian, R. Phys. Rev. Lett. 102, 188305 (2009). 8. Sharp, K. & Skinner, J. J. Proteins 65, 347–361 (2006). 9. Igumenova, T. I., Frederick, K. K. & Wand, A. J. Chem. Rev. 106, 1672–1699 (2006). 10. Fu, Y. et al. J. Am. Chem. Soc. 134, 8543–8550 (2012). 11. Englander, S. W. & Mayne, L. Proc. Natl Acad. Sci. USA 111, 15873–15880 (2014). 12. Sharp, K. A., Nicholls, A., Fine, R. F. & Honig, B. Science 252, 106–109 (1991). 13. Chandler, D. Nature 437, 640–647 (2005). 14. Heyden, M. & Tobias, D. J. Phys. Rev. Lett. 111, 218101 (2013). 15. Nucci, N. V., Pometun, M. S. & Wand, A. J. Nature Struct. Mol. Biol. 18, 245–249 (2011). 16. Grossman, M. et al. Nature Struct. Mol. Biol. 18, 1102–1108 (2011). This article was published online on 10 December 2014.
C L I M ATE SC I E NCE
Unburnable fossil-fuel reserves How much more of Earth’s fossil fuels can we extract and burn in the short- to medium-term future and still avoid severe global warming? A model provides the answer, and shows where these ‘unburnable’ reserves are. See Letter p.187 M I C H A E L J A KO B & J É R Ô M E H I L A I R E
C
umulative carbon dioxide emissions must be less than 870 to 1,240 gigatonnes between 2011 and 2050 if we are to have a reasonable chance of limiting global warming to 2 °C above the average global temperature of pre-industrial times1. But the carbon contained in global resources of fossil fuels is estimated2 to be equivalent to about 11,000 Gt of CO2, which means that the implementation of ambitious climate policies would lead to large proportions of reserves remaining unexploited (Fig. 1). On page 187 of this issue, McGlade and Ekins2 comprehensively quantify the regional distribution of fossil-fuel reserves that should not be burned between 2010 and 2050, by modelling a broad range of scenarios based on least-cost climate policies. Several studies have previously analysed the global long-term implications of climatechange mitigation on fossil-fuel markets3–5. The novelty of the present study stems from the detailed regional representation of fossilfuel reserves used in the authors’ model, which are based on well-established data sources. In each of the 16 regions modelled, fossil fuels are divided into 21 categories that include various
1 5 0 | N AT U R E | VO L 5 1 7 | 8 JA N UA RY 2 0 1 5
© 2015 Macmillan Publishers Limited. All rights reserved
types of coal, oil and gas. Each category further accounts for key characteristics, such as recoverable resources, production and trade costs, as well as natural decline rates of production (the rates of fall that would occur in the absence of any further investment). This approach allows the authors to emphasize differences in unburnable fossil-fuel reserves. About 80%, 50% and 30% of coal, gas and oil reserves, respectively, would need to remain below Earth’s surface if the world is to limit an increase in global mean temperature to 2 °C. The uneven distribution of unburnable carbon has far-reaching consequences for fossil-fuel owners. For example, the Middle East, which holds the bulk of conventional oil reserves, would need to leave about 40% of those reserves underground. This corresponds to about 8 years of global production at current levels6 (87 million barrels per day). Similarly, countries with large coal endowments would face great challenges. China and India would have to discard 66% of their reserves, whereas Africa would have to leave 85% of them. In addition, the United States, Australia and countries of the former Soviet Union would need to leave more than 90% of their coal reserves underground, in stark contrast to the renaissance of coal
HARVARD PROJECT ON CLIMATE AGREEMENTS
NEWS & VIEWS RESEARCH
Atmosphere 870–1,240 Gt CO2
Fossil resources 11,000 Gt CO2
Figure 1 | Fossil-fuel resources exceed atmospheric disposal space for carbon emissions. McGlade and Ekins2 report that the carbon contained in fossil-fuel reserves (equivalent to 11,000 gigatonnes of carbon dioxide) is much more than the amount that can be emitted as CO2 to the atmosphere (870–1,240 Gt) if global warming is to be limited to 2 °C above the average global temperature of pre-industrial times. (Figure adapted from ref. 14.)
use currently under way in many places7. Gas-fired power plants emit less CO2 per unit of energy produced than coal-fired plants, and so ‘unconventional’ sources of natural gas, such as shale gas, have been touted as a bridge to the projected global transition to carbonfree, renewable energy technologies (although this bridging role has recently been challenged8). Encouraged by the recent shale-gas production boom in the United States, several world regions, including China, India, Africa and the Middle East, are seeking to unlock their large endowments or increase existing production. However, McGlade and Ekins’ analysis shows that Africa and the Middle East would have to leave their entire unconventional gas resources underground, and that about 10% of the combined endowment of China and India (which includes substantial amounts of coal-bed methane) could be produced. McGlade and Ekins’ figures, computed for the period 2010–50, show that the amounts of unburnable fossil fuels are modestly sensitive to the availability of carbon capture and sequestration technology. When this technology is not available, even less coal, oil and gas can be extracted, and natural gas must be used in preference to coal because of the gas’s lower ratio of emissions to energy produced. The future use of CO2-removal technologies might allow further extraction of all fossil fuels after 2050, but there are many uncertainties associated with predicting the availability of these young technologies. The authors’ insights echo calls9 in the past
few years for society to divest itself of fossil fuels. Such calls have been made by organizations in an attempt to influence institutional investors, such as pension funds, to shift their portfolios towards clean-energy investments. These organizations also draw attention to a potential bursting of the ‘carbon bubble’ that would result from the adoption of ambitious climate policies, leading to severe devaluations of fossil-fuel reserves, which are currently worth about US$27 trillion9. Fossil-fuel companies must therefore ask themselves whether they should continue to invest in exploration for, and processing of, oil, gas and coal, or risk losing billions of dollars of stranded assets. Given the political influence of the fossil-fuel industry, policy-makers must design solutions that ensure stakeholders’ acceptance. Importantly, McGlade and Ekins’ results clearly highlight the distributional challenge of climate policy: imposing a limit on the use of fossil fuels transfers economic benefits (known as rents) from resource owners to those who obtain the right to use the remaining burnable reserves. Hence, successful climate policy will crucially hinge on the question of whether this ‘climate rent’ can be shared in an equitable way that also ensures resource owners are compensated for their losses4. This could be achieved by an appropriate allocation of emissions permits in an international carbon market, or by payments through the Green Climate Fund (which was set up by the United Nations to assist developing countries in adopting practices that counter climate change). Other proposals include alleviating national debt in exchange for emissions reductions10, or using some part of the climate rent to finance access to basic infrastructure services, such as water, sanitation and electricity11. But given the crucial role of energy in economic development, how can countries be convinced to forgo the use of fossil fuels if this is perceived to imperil primary policy objectives such as poverty reduction? During the US–Africa Leaders’ Summit last August, for example, Tanzania’s energy minister, Sospeter Muhongo, said12: “We in Africa, we should not be in the discussion of whether we should use coal or not. In my country of Tanzania, we are going to use our natural resources because we have reserves which go beyond 5 billion tons.” Only a global climate agreement that compensates losers and is perceived as equitable by all participants can impose strict limits on the use of fossil fuels in the long term. By identifying potential winners and losers of climatechange mitigation, analyses such as the one by McGlade and Ekins provide valuable support for the design of such an agreement, and inform short-term measures that can pave the way to an accord13. ■ Michael Jakob and Jérôme Hilaire are at the Potsdam Institute for Climate Impact
50 Years Ago The Schizophrenia Research Fund has been established to support research into problems connected with mental illness in general and schizophrenia in particular … Initial impetus has been given to the fund by a gift of £50,000 from the Rothschild family, and the establishment of a Schizophrenia Research Fellowship, to which Dr. D. Straughan has been appointed … Dr. Straughan’s contract is for seven years. His work … has been concerned with pharmacological aspects of mammalian brain physiology, and he will concentrate on the biochemical basis of schizophrenia. It is hoped that this initial effort will attract interest in, and support for, work in the immense field of research bearing on the problems of mental health. From Nature 9 January1965
100 Years Ago There is a widespread but erroneous belief in official circles, and among wealthy philanthropists, to the effect that you can hire a scientific discoverer and then say to him, “Discover me this” or “Discover me that” (naming to him a possible and greatly desired piece of new knowledge), and that he will thereupon proceed right away to make the discovery which you want … But valuable and important scientific discovery cannot be produced directly in response to orders given and money expended. You cannot manufacture scientific discovery like soap. The great difficulty, in the first place, is to catch that rare and evasive creature — a scientific discoverer — and when you have found him you have to humour him and let him do as he fancies. Then he will discover things, but probably not the things which either you or he wanted or expected. From Nature 7 January 1915
8 JA N UA RY 2 0 1 5 | VO L 5 1 7 | N AT U R E | 1 5 1
© 2015 Macmillan Publishers Limited. All rights reserved
RESEARCH NEWS & VIEWS Research, Head of Research Domain III, 14412 Potsdam, Germany. M.J. is also at the Mercator Research Institute on Global Commons and Climate Change, 10829 Berlin, Germany. e-mails: [email protected]; [email protected] 1. Clarke, L. et al. in Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (eds Edenhofer, O. et al.) Ch. 6 (Cambridge Univ. Press, 2014). 2. McGlade, C. E. & Ekins, P. Nature 517,
187–190 (2015). 3. McCollum, D., Bauer, N., Calvin, K., Kitous, A. & Riahi, K. Clim. Change 123, 413–426 (2014). 4. Bauer, N. et al. Clim. Change http://dx.doi. org/10.1007/s10584-013-0901-6 (2013). 5. Jewell, J. et al. Clim. Change Econ. 04, 1340011 (2013). 6. BP Statistical Review of World Energy 2014; www. bp.com/en/global/corporate/about-bp/energyeconomics/statistical-review-of-world-energy.html (2014). 7. Steckel, J. C., Jakob, M., Marschinski, R. & Luderer, G. Energy Policy 39, 3443–3455 (2011). 8. McJeon, H. et al. Nature 514, 482–485 (2014). 9. Carbon Tracker & Grantham Institute Unburnable Carbon 2013: Wasted Capital and Stranded Assets; www.carbontracker.org/report/
CANCE R
Resistance through repopulation Bladder-cancer cells have been found to release prostaglandin E2 when they are killed by chemotherapy. Paradoxically, this molecule stimulates the proliferation of surviving cancer stem cells, leading to tumour repopulation. See Letter p.209 I A N F. TA N N O C K
I
nitial or acquired resistance to anticancer drugs limits their effectiveness. Many papers have described mechanisms underlying such drug resistance, but most have evaluated molecular changes in single tumour cells, with the emphasis on stable genetic changes that may be induced by therapy or selected for. However, causes of drug resistance are multiple and complex, and in
Relative number of tumour cells
a
this issue, Kurtova et al.1 (page 209) explore an alternative mechanism — the repopulation of tumours by cancer stem cells that survive treatment. The authors show that, in certain bladder tumours, repopulation is stimulated by the hormone-like lipid molecule prostaglandin E2 (PGE2), which is released from tumour cells that are killed by the initial chemotherapy. Furthermore, the authors show that repopulation can be inhibited by the drug celecoxib, which inhibits PGE2 synthesis.
Chemotherapy with accelerating repopulation
b Chemotherapy with inhibition of repopulation
1.0
1.0
10–1
10–1
10–2
10–2 0
1
2
3
2 3 4 5 6 0 1 Chemotherapy cycles (at three-week intervals)
4
5
6
Figure 1 | Inhibiting repopulation enhances chemotherapy effects. a, Repopulation of a tumour with cells that survive chemotherapy can lead to patterns of tumour shrinkage and regrowth during and after each cycle of chemotherapy. This repopulation often accelerates, even if there is no change in the effectiveness of the chemotherapy in killing tumour cells during successive courses. b, Selective inhibition of repopulation between treatment cycles can lead to continued loss of tumour cells. Kurtova et al.1 show that this approach works in mouse models of bladder cancer, by using the drug celecoxib to inhibit the production of the molecule prostaglandin E2, which they found stimulates repopulation from cancer stem cells in this tumour. 1 5 2 | N AT U R E | VO L 5 1 7 | 8 JA N UA RY 2 0 1 5
© 2015 Macmillan Publishers Limited. All rights reserved
wasted-capital-and-stranded-assets (2013). 10. Fenton, A., Wright, H., Afionis, S., Paavola, J. & Huq, S. Nature Clim. Change 4, 650–653 (2014). 11. Jakob, M. & Edenhofer, O. Oxford Rev. Econ. Policy (in the press). 12. www.scientificamerican.com/article/africa-needsfossil-fuels-to-end-energy-apartheid 13. Jakob, M. et al. Nature Clim. Change 4, 961–968 (2014). 14. Edenhofer, O., Flachsland, C., Jakob, M. & Lessmann, K. The Atmosphere as a Global Commons — Challenges for International Cooperation and Governance Discuss. Pap. 13-58; http:// belfercenter.ksg.harvard.edu/files/hpcadp58_ edenhofer-flachsland-jakob-lessmann.pdf (2013).
As Kurtova and colleagues point out, cellular repopulation of normal tissues is crucial in wound healing and in the recovery of white blood cells of the immune system following chemotherapy. The rapidly proliferating, drug-sensitive precursor cells in bone marrow are preferentially killed by anticancer drugs, leading to a decrease in functional white cells in the blood about 10–14 days after treatment and leaving patients at risk of infection. The fall in white-blood-cell count stimulates the production of haematological growth factors, which signal slowly proliferating bone-marrow stem cells to divide, and the blood is usually repopulated with mature white cells in about three weeks, the usual interval between courses of clinical chemotherapy. But tumour repopulation also occurs, through cancer cells that survive repeated treatments, and this can accelerate with time2. It has long been recognized that repopulation occurs during courses of radiotherapy, and that the relative rates of repopulation in tumour and surrounding normal tissue can determine patient outcomes3. Kurtova and colleagues’ findings add to evidence that repopulation is also a cause of treatment failure during chemotherapy. Mathematical models show2 that accelerating repopulation during successive courses of treatment can lead to tumour remission then regrowth, as is observed in patients, without any change in the intrinsic sensitivity of the tumour cells to treatment drugs (Fig. 1). Inhibiting repopulation has the potential to improve the outcome of chemotherapy, provided that the inhibitors used are specific for tumour cells and do not prevent the essential repopulation in normal tissues. Although the mechanisms leading to tumour repopulation are not known, it has been assumed that it might occur as tumours shrink and the nutrient status of surviving cells improves as other cells die2. The surviving population will probably over-represent cells that were nutrient deprived and slowly proliferating, and thus resistant to drugs that target cell-cycle pathways. The surviving cells are also probably situated far from tumour blood vessels, because this location will have been subject to lower drug concentrations owing
enzyme catalysis is dissipated and how this can result in a locomotive protein molecule. Finally, the most intriguing question of all is whether the anomalous diffusion of protein molecules through the chemoacoustic effect is a product of evolution or simply an accidental unselected result of heat flow in proteins. It seems reasonable to think that enzymes might have evolved the capacity to seek ‘greener pastures’ of substrates. But this view is more complicated than it might at first look6,7, and any selective advantage may ultimately be quite obscure. ■ A. Joshua Wand is in the Department of Biochemistry and Biophysics, University of Pennsylvania, Philadelphia 19104-6059, USA. e-mail: [email protected] 1. Fischer, E. Ber. Dt. Chem. Ges. 27, 2985–2993 (1894). 2. Riedel, C. et al. Nature 517, 227–230 (2015). 3. Muddana, H. S., Sengupta, S., Mallouk, T. E., Sen, A.
& Butler, P. J. J. Am. Chem. Soc. 132, 2110–2111 (2010). 4. Sengupta, S. et al. J. Am. Chem. Soc. 135, 1406– 1414 (2013). 5. Paxton, W. F., Sundararajan, S., Mallouk, T. E. & Sen, A. Angew. Chem. Int. Edn 45, 5420–5429 (2006). 6. Purcell, E. M. Am. J. Phys. 45, 3–11 (1977). 7. Golestanian, R. Phys. Rev. Lett. 102, 188305 (2009). 8. Sharp, K. & Skinner, J. J. Proteins 65, 347–361 (2006). 9. Igumenova, T. I., Frederick, K. K. & Wand, A. J. Chem. Rev. 106, 1672–1699 (2006). 10. Fu, Y. et al. J. Am. Chem. Soc. 134, 8543–8550 (2012). 11. Englander, S. W. & Mayne, L. Proc. Natl Acad. Sci. USA 111, 15873–15880 (2014). 12. Sharp, K. A., Nicholls, A., Fine, R. F. & Honig, B. Science 252, 106–109 (1991). 13. Chandler, D. Nature 437, 640–647 (2005). 14. Heyden, M. & Tobias, D. J. Phys. Rev. Lett. 111, 218101 (2013). 15. Nucci, N. V., Pometun, M. S. & Wand, A. J. Nature Struct. Mol. Biol. 18, 245–249 (2011). 16. Grossman, M. et al. Nature Struct. Mol. Biol. 18, 1102–1108 (2011). This article was published online on 10 December 2014.
C L I M ATE SC I E NCE
Unburnable fossil-fuel reserves How much more of Earth’s fossil fuels can we extract and burn in the short- to medium-term future and still avoid severe global warming? A model provides the answer, and shows where these ‘unburnable’ reserves are. See Letter p.187 M I C H A E L J A KO B & J É R Ô M E H I L A I R E
C
umulative carbon dioxide emissions must be less than 870 to 1,240 gigatonnes between 2011 and 2050 if we are to have a reasonable chance of limiting global warming to 2 °C above the average global temperature of pre-industrial times1. But the carbon contained in global resources of fossil fuels is estimated2 to be equivalent to about 11,000 Gt of CO2, which means that the implementation of ambitious climate policies would lead to large proportions of reserves remaining unexploited (Fig. 1). On page 187 of this issue, McGlade and Ekins2 comprehensively quantify the regional distribution of fossil-fuel reserves that should not be burned between 2010 and 2050, by modelling a broad range of scenarios based on least-cost climate policies. Several studies have previously analysed the global long-term implications of climatechange mitigation on fossil-fuel markets3–5. The novelty of the present study stems from the detailed regional representation of fossilfuel reserves used in the authors’ model, which are based on well-established data sources. In each of the 16 regions modelled, fossil fuels are divided into 21 categories that include various
1 5 0 | N AT U R E | VO L 5 1 7 | 8 JA N UA RY 2 0 1 5
© 2015 Macmillan Publishers Limited. All rights reserved
types of coal, oil and gas. Each category further accounts for key characteristics, such as recoverable resources, production and trade costs, as well as natural decline rates of production (the rates of fall that would occur in the absence of any further investment). This approach allows the authors to emphasize differences in unburnable fossil-fuel reserves. About 80%, 50% and 30% of coal, gas and oil reserves, respectively, would need to remain below Earth’s surface if the world is to limit an increase in global mean temperature to 2 °C. The uneven distribution of unburnable carbon has far-reaching consequences for fossil-fuel owners. For example, the Middle East, which holds the bulk of conventional oil reserves, would need to leave about 40% of those reserves underground. This corresponds to about 8 years of global production at current levels6 (87 million barrels per day). Similarly, countries with large coal endowments would face great challenges. China and India would have to discard 66% of their reserves, whereas Africa would have to leave 85% of them. In addition, the United States, Australia and countries of the former Soviet Union would need to leave more than 90% of their coal reserves underground, in stark contrast to the renaissance of coal
HARVARD PROJECT ON CLIMATE AGREEMENTS
NEWS & VIEWS RESEARCH
Atmosphere 870–1,240 Gt CO2
Fossil resources 11,000 Gt CO2
Figure 1 | Fossil-fuel resources exceed atmospheric disposal space for carbon emissions. McGlade and Ekins2 report that the carbon contained in fossil-fuel reserves (equivalent to 11,000 gigatonnes of carbon dioxide) is much more than the amount that can be emitted as CO2 to the atmosphere (870–1,240 Gt) if global warming is to be limited to 2 °C above the average global temperature of pre-industrial times. (Figure adapted from ref. 14.)
use currently under way in many places7. Gas-fired power plants emit less CO2 per unit of energy produced than coal-fired plants, and so ‘unconventional’ sources of natural gas, such as shale gas, have been touted as a bridge to the projected global transition to carbonfree, renewable energy technologies (although this bridging role has recently been challenged8). Encouraged by the recent shale-gas production boom in the United States, several world regions, including China, India, Africa and the Middle East, are seeking to unlock their large endowments or increase existing production. However, McGlade and Ekins’ analysis shows that Africa and the Middle East would have to leave their entire unconventional gas resources underground, and that about 10% of the combined endowment of China and India (which includes substantial amounts of coal-bed methane) could be produced. McGlade and Ekins’ figures, computed for the period 2010–50, show that the amounts of unburnable fossil fuels are modestly sensitive to the availability of carbon capture and sequestration technology. When this technology is not available, even less coal, oil and gas can be extracted, and natural gas must be used in preference to coal because of the gas’s lower ratio of emissions to energy produced. The future use of CO2-removal technologies might allow further extraction of all fossil fuels after 2050, but there are many uncertainties associated with predicting the availability of these young technologies. The authors’ insights echo calls9 in the past
few years for society to divest itself of fossil fuels. Such calls have been made by organizations in an attempt to influence institutional investors, such as pension funds, to shift their portfolios towards clean-energy investments. These organizations also draw attention to a potential bursting of the ‘carbon bubble’ that would result from the adoption of ambitious climate policies, leading to severe devaluations of fossil-fuel reserves, which are currently worth about US$27 trillion9. Fossil-fuel companies must therefore ask themselves whether they should continue to invest in exploration for, and processing of, oil, gas and coal, or risk losing billions of dollars of stranded assets. Given the political influence of the fossil-fuel industry, policy-makers must design solutions that ensure stakeholders’ acceptance. Importantly, McGlade and Ekins’ results clearly highlight the distributional challenge of climate policy: imposing a limit on the use of fossil fuels transfers economic benefits (known as rents) from resource owners to those who obtain the right to use the remaining burnable reserves. Hence, successful climate policy will crucially hinge on the question of whether this ‘climate rent’ can be shared in an equitable way that also ensures resource owners are compensated for their losses4. This could be achieved by an appropriate allocation of emissions permits in an international carbon market, or by payments through the Green Climate Fund (which was set up by the United Nations to assist developing countries in adopting practices that counter climate change). Other proposals include alleviating national debt in exchange for emissions reductions10, or using some part of the climate rent to finance access to basic infrastructure services, such as water, sanitation and electricity11. But given the crucial role of energy in economic development, how can countries be convinced to forgo the use of fossil fuels if this is perceived to imperil primary policy objectives such as poverty reduction? During the US–Africa Leaders’ Summit last August, for example, Tanzania’s energy minister, Sospeter Muhongo, said12: “We in Africa, we should not be in the discussion of whether we should use coal or not. In my country of Tanzania, we are going to use our natural resources because we have reserves which go beyond 5 billion tons.” Only a global climate agreement that compensates losers and is perceived as equitable by all participants can impose strict limits on the use of fossil fuels in the long term. By identifying potential winners and losers of climatechange mitigation, analyses such as the one by McGlade and Ekins provide valuable support for the design of such an agreement, and inform short-term measures that can pave the way to an accord13. ■ Michael Jakob and Jérôme Hilaire are at the Potsdam Institute for Climate Impact
50 Years Ago The Schizophrenia Research Fund has been established to support research into problems connected with mental illness in general and schizophrenia in particular … Initial impetus has been given to the fund by a gift of £50,000 from the Rothschild family, and the establishment of a Schizophrenia Research Fellowship, to which Dr. D. Straughan has been appointed … Dr. Straughan’s contract is for seven years. His work … has been concerned with pharmacological aspects of mammalian brain physiology, and he will concentrate on the biochemical basis of schizophrenia. It is hoped that this initial effort will attract interest in, and support for, work in the immense field of research bearing on the problems of mental health. From Nature 9 January1965
100 Years Ago There is a widespread but erroneous belief in official circles, and among wealthy philanthropists, to the effect that you can hire a scientific discoverer and then say to him, “Discover me this” or “Discover me that” (naming to him a possible and greatly desired piece of new knowledge), and that he will thereupon proceed right away to make the discovery which you want … But valuable and important scientific discovery cannot be produced directly in response to orders given and money expended. You cannot manufacture scientific discovery like soap. The great difficulty, in the first place, is to catch that rare and evasive creature — a scientific discoverer — and when you have found him you have to humour him and let him do as he fancies. Then he will discover things, but probably not the things which either you or he wanted or expected. From Nature 7 January 1915
8 JA N UA RY 2 0 1 5 | VO L 5 1 7 | N AT U R E | 1 5 1
© 2015 Macmillan Publishers Limited. All rights reserved
RESEARCH NEWS & VIEWS Research, Head of Research Domain III, 14412 Potsdam, Germany. M.J. is also at the Mercator Research Institute on Global Commons and Climate Change, 10829 Berlin, Germany. e-mails: [email protected]; [email protected] 1. Clarke, L. et al. in Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (eds Edenhofer, O. et al.) Ch. 6 (Cambridge Univ. Press, 2014). 2. McGlade, C. E. & Ekins, P. Nature 517,
187–190 (2015). 3. McCollum, D., Bauer, N., Calvin, K., Kitous, A. & Riahi, K. Clim. Change 123, 413–426 (2014). 4. Bauer, N. et al. Clim. Change http://dx.doi. org/10.1007/s10584-013-0901-6 (2013). 5. Jewell, J. et al. Clim. Change Econ. 04, 1340011 (2013). 6. BP Statistical Review of World Energy 2014; www. bp.com/en/global/corporate/about-bp/energyeconomics/statistical-review-of-world-energy.html (2014). 7. Steckel, J. C., Jakob, M., Marschinski, R. & Luderer, G. Energy Policy 39, 3443–3455 (2011). 8. McJeon, H. et al. Nature 514, 482–485 (2014). 9. Carbon Tracker & Grantham Institute Unburnable Carbon 2013: Wasted Capital and Stranded Assets; www.carbontracker.org/report/
CANCE R
Resistance through repopulation Bladder-cancer cells have been found to release prostaglandin E2 when they are killed by chemotherapy. Paradoxically, this molecule stimulates the proliferation of surviving cancer stem cells, leading to tumour repopulation. See Letter p.209 I A N F. TA N N O C K
I
nitial or acquired resistance to anticancer drugs limits their effectiveness. Many papers have described mechanisms underlying such drug resistance, but most have evaluated molecular changes in single tumour cells, with the emphasis on stable genetic changes that may be induced by therapy or selected for. However, causes of drug resistance are multiple and complex, and in
Relative number of tumour cells
a
this issue, Kurtova et al.1 (page 209) explore an alternative mechanism — the repopulation of tumours by cancer stem cells that survive treatment. The authors show that, in certain bladder tumours, repopulation is stimulated by the hormone-like lipid molecule prostaglandin E2 (PGE2), which is released from tumour cells that are killed by the initial chemotherapy. Furthermore, the authors show that repopulation can be inhibited by the drug celecoxib, which inhibits PGE2 synthesis.
Chemotherapy with accelerating repopulation
b Chemotherapy with inhibition of repopulation
1.0
1.0
10–1
10–1
10–2
10–2 0
1
2
3
2 3 4 5 6 0 1 Chemotherapy cycles (at three-week intervals)
4
5
6
Figure 1 | Inhibiting repopulation enhances chemotherapy effects. a, Repopulation of a tumour with cells that survive chemotherapy can lead to patterns of tumour shrinkage and regrowth during and after each cycle of chemotherapy. This repopulation often accelerates, even if there is no change in the effectiveness of the chemotherapy in killing tumour cells during successive courses. b, Selective inhibition of repopulation between treatment cycles can lead to continued loss of tumour cells. Kurtova et al.1 show that this approach works in mouse models of bladder cancer, by using the drug celecoxib to inhibit the production of the molecule prostaglandin E2, which they found stimulates repopulation from cancer stem cells in this tumour. 1 5 2 | N AT U R E | VO L 5 1 7 | 8 JA N UA RY 2 0 1 5
© 2015 Macmillan Publishers Limited. All rights reserved
wasted-capital-and-stranded-assets (2013). 10. Fenton, A., Wright, H., Afionis, S., Paavola, J. & Huq, S. Nature Clim. Change 4, 650–653 (2014). 11. Jakob, M. & Edenhofer, O. Oxford Rev. Econ. Policy (in the press). 12. www.scientificamerican.com/article/africa-needsfossil-fuels-to-end-energy-apartheid 13. Jakob, M. et al. Nature Clim. Change 4, 961–968 (2014). 14. Edenhofer, O., Flachsland, C., Jakob, M. & Lessmann, K. The Atmosphere as a Global Commons — Challenges for International Cooperation and Governance Discuss. Pap. 13-58; http:// belfercenter.ksg.harvard.edu/files/hpcadp58_ edenhofer-flachsland-jakob-lessmann.pdf (2013).
As Kurtova and colleagues point out, cellular repopulation of normal tissues is crucial in wound healing and in the recovery of white blood cells of the immune system following chemotherapy. The rapidly proliferating, drug-sensitive precursor cells in bone marrow are preferentially killed by anticancer drugs, leading to a decrease in functional white cells in the blood about 10–14 days after treatment and leaving patients at risk of infection. The fall in white-blood-cell count stimulates the production of haematological growth factors, which signal slowly proliferating bone-marrow stem cells to divide, and the blood is usually repopulated with mature white cells in about three weeks, the usual interval between courses of clinical chemotherapy. But tumour repopulation also occurs, through cancer cells that survive repeated treatments, and this can accelerate with time2. It has long been recognized that repopulation occurs during courses of radiotherapy, and that the relative rates of repopulation in tumour and surrounding normal tissue can determine patient outcomes3. Kurtova and colleagues’ findings add to evidence that repopulation is also a cause of treatment failure during chemotherapy. Mathematical models show2 that accelerating repopulation during successive courses of treatment can lead to tumour remission then regrowth, as is observed in patients, without any change in the intrinsic sensitivity of the tumour cells to treatment drugs (Fig. 1). Inhibiting repopulation has the potential to improve the outcome of chemotherapy, provided that the inhibitors used are specific for tumour cells and do not prevent the essential repopulation in normal tissues. Although the mechanisms leading to tumour repopulation are not known, it has been assumed that it might occur as tumours shrink and the nutrient status of surviving cells improves as other cells die2. The surviving population will probably over-represent cells that were nutrient deprived and slowly proliferating, and thus resistant to drugs that target cell-cycle pathways. The surviving cells are also probably situated far from tumour blood vessels, because this location will have been subject to lower drug concentrations owing
