news and views
Palaeoclimatology
A warm future in the past William R. Howard
*Carbonate Marine System During Oxygen Isotope Stage 11, American Geophysical Union Spring Meeting, Baltimore, Maryland, USA, 27–30 May 1997.
418
Figure 1 Climate and chemistry of the past 500,000 years. All the climatic variables b to f (refs 2, 8–11, respectively) are related to solar forcing, a (ref. 12), determined by changes in the Earth’s orbit (c, d and e relative to average over the Holocene — the past 10,000 years or so). The long, hot interglacial of stage 11 occurred under similar orbital conditions to those we have now — so is it a model for our own time?
Oxygen isotope stages 1
3
5
6
7
8
9
10
11
12
a
June 65 °N solar radiation
W m–2
550 530 510 490 470
b Ice volume
Marine d18O (standard deviations)
450
1 0 -1 1
c Sea surface temperature
0
°C
-1 -2 -3 -4
Subpolar Southern Ocean
d
5
°C
0 -5
Benthic d13C (‰)
-10
g cm–2 kyr–1
T
followed by early returns to glacial conditions (Fig. 1). Many other aspects of the climate were strange. In deep-sea records of MIS 11, subpolar waters in both hemispheres were at their warmest (L. Burckle, Lamont-Doherty Earth Obs.), following one of the coldest glacial stages of the past 500,000 years (Fig. 1c, d). Deep-ocean carbon isotope data show a maximum in the production of North Atlantic Deep Water (D. Hodell, Univ. Florida). Tracers of downward particle flux in the equatorial Pacific show that biological productivity was at a minimum in MIS 11 (R. Murray, Boston Univ.). Calcium car-
Relative to Holocene
o understand the impact of a possibly warmer future climate, geologists are searching the past for warmer-thanpresent interglacial intervals — examples of what we may expect in a world with more greenhouse gases than ours1. A common tactic is to look at the last interglacial (around 120,000 years ago), but a period 423,000 to 362,000 years ago may fit the bill better, because the Earth’s orbital geometry has been similar during the Holocene (the present interglacial period) to what it was then. This period is known to palaeoclimatologists as stage 11 or ‘MIS 11’, according to a numbering system for glacial advances and retreats marked in the marine oxygen isotope record (Fig. 1). At a recent symposium*, evidence emerged of extreme climatic variation and peculiar interplays between ocean temperature, thermohaline circulation, plankton ecology, sea level and reef growth during MIS 11, all of which may provide insight into the response of the natural carbon cycle to future climate change. Changes in the Earth’s orbital geometry explain the strong 41,000-year and 23,000year rhythms in climate over the past five million years (due to changes in axial tilt and orbital precession, respectively). During the past 600,000 years, however, the dominant rhythm has been a 100,000-year cycle. Although this climate cycle is at a similar frequency to orbital eccentricity variation, the radiative impact of that orbital cycle is too small — at least in linear climate models2 — to produce the large swings in the climate record. MIS 11 is the warm peak of one such ‘100k’ cycle, with two intriguing features. First, climate change at the 12–11 transition is nearly the most severe of the past half-million years (Fig. 1), yet it comes at a time of low orbital eccentricity, thus of low precessional amplitude. Second, interglacial conditions last longer. Subsequent interglacials (stages 9, 7, and 5) show short periods of warmth
bonate production peaked in subpolar and subtropical oceans (R. Schneider, Univ. Bremen), reflecting a temperature-related shift in plankton ecology from diatoms to calcareous plankton. In shallow-water environments, the evolution of several major reef systems (such as the Great Barrier Reef) accompanied increased reef carbonate production (A. Droxler, Rice Univ.; P. Davies, Sydney Univ.). The production of carbonate in continental shelves and in mid-latitude open-ocean environments may partly explain the extreme dissolution of carbonate sediments during MIS 11 throughout ocean basins such as the Indian and Pacific3. (In the oceans’ carbonate cycle, biological formation of hard body parts from carbonate outstrips river input of the component ions, and dissolution from the seabed makes up the difference. In other words, to produce more carbonate in one place you must dissolve more elsewhere.) Perhaps barrier-reef tracts were established during stage 11 because tropical continental shelves were flooded (A. Droxler; C. Kievman, Kean Coll.). How high did the seas rise? Beach deposits in Alaska, Bermuda and the Bahamas (P. Hearty, Bahamas) and uplifted reef terraces in Indonesia4 seem to imply that the ocean stood as much as twenty metres above its present level. But this is uncertain: estimating sea level for MIS 11 is complicated by the fact that dating beaches
Subpolar North Atlantic
e 0
-1 South Atlantic
f
2
1 Subpolar Southern Ocean 0
0
100
200
300
400
production
Matt McGue is in the Department of Psychology and the Institute of Human Genetics, 75 East River Road, University of Minnesota, Minneapolis, Minnesota 55455, USA. e-mail:
[email protected]
1. Galton, F. Hereditary Genius: An Inquiry into its Laws and Consequences (Macmillan, London, 1869). 2. Devlin, B., Daniels, M. & Roeder, K. Nature 388, 468–471 (1997). 3. Neisser, U. et al. Am. Psychol. 51, 77–101 (1996). 4. Herrnstein, R. J. & Murray, C. The Bell Curve: Intelligence and Class Structure in American Life (Free Press, New York, 1994). 5. Plomin, R. & Loehlin, J. C. Behav. Genet. 19, 331–342 (1989). 6. McClearn, G. E. et al. Science 276, 1560–1563 (1997). 7. McGue, M., Bouchard, T. J. Jr, Iacono, W. G. & Lykken, D. T. in Nature, Nurture and Psychology (eds Plomin, R. & McClearn, G. E.) 59–76 (Am. Psychol. Assoc., Washington DC, 1993). 8. Price, B. Am. J. Hum. Genet. 2, 293–352 (1950). 9. Broman, S. H., Nichols, P. L. & Kennedy, W. A. Preschool IQ: Prenatal and Early Developmental Correlates (Erlbaum, Hillsdale, NJ, 1975).
Carbonate (CaCO3) North Atlantic deep water accumulation rate
life, or in the womb, and directly affect the development of the brain. Research on the nature and nurture of IQ is converging on the view that human intellectual ability has a strong, but malleable, biological basis — a convergence that Galton would, no doubt, have found quite congenial.
500
Age (thousands of years ago)
Nature © Macmillan Publishers Ltd 1997
NATURE | VOL 388 | 31 JULY 1997
news and views and reefs of that age is difficult. Deep-sea d18O records (including some new data presented at the symposium) show isotopic depletions during MIS 11 that are consistent with a sea-level high, but in the isotope records the effects of warmer temperatures cannot be easily untangled from the effects of decreased ice volume. To produce such a high sea level would require the collapse of at least one major ice sheet (candidates include the West Antarctic and Greenland Ice Sheets). The stability of these ice sheets in the face of high-latitude warming is one of the main questions in climate-change research. These extremes in the ocean’s thermal and geochemical state all lower its ability to hold CO2, leaving more in the atmosphere. First, CO2 is less soluble in a warmer ocean. Second, as sea level and surface temperature rise, a greater proportion of biological production is in the form of calcium carbonate. This increased production and burial of calcium carbonate lowers the ocean’s alkalinity, again reducing its ability to hold on to CO2. Third, decreased biological production of soft tissue reduces the export of carbon from the surface (in contact with the atmosphere) to the deep ocean5. These changes in the ocean’s carbonate–carbon cycle during MIS 11 point to what might be an important set of interactions between temperature, ecology and CO2. Model simulations of oceanic CO2 uptake have recently begun to take such feedbacks into account6. Was the stage-11 warming driven by increased atmospheric CO2 concentrations? The Vostok ice core now provides a palaeoclimate record extending to about 425,000 years ago7. Preliminary measurements suggest that atmospheric CO2 levels during stage 11 are comparable to those of later interglacials, which would mean that the link between warming mechanisms and the carbon cycle may not be so simple. However, ice disturbance at the stage-11 level may have altered the CO2 record (D. Raynaud, LGGE, Grenoble). The symposium addressed only some of the evidence from and implications of the stage-11 warm interval. Was the climate stable, or punctuated by high-frequency ‘flickers’? What was the global distribution of sea-surface temperature? Stage 11 has not been studied in detail, in part because few climate records have reached it. But new cores, extracted by Ocean Drilling Program technology and long gravity-driven piston cores in high-accumulation-rate drift deposits, are beginning to provide more evidence. Stage 11 may provide some important perspectives on future climate change. Are we adding greenhouse gases to a climate that, left to its own devices, should be on the way into another glaciation, as in the short-lived stage 5; or one that would stay warm, or get NATURE | VOL 388 | 31 JULY 1997
warmer, for another 40,000 years, as in stage 11? Are we setting in motion a chain of biogeochemical feedbacks that will add an oceanic contribution to the anthropogenic greenhouse-gas enrichment? William R. Howard is at the Cooperative Research Centre for the Antarctic and Southern Ocean Environment, University of Tasmania, GPO Box 252-80, Hobart, Tasmania 7001, Australia. 1. Burckle, L. H. Quat. Sci. Rev. 12, 825–831 (1993). 2. Imbrie, J. et al. Paleoceanography 8, 699–735 (1993). 3. Farrell, J. & Prell, W. L. Paleoceanography 4, 447–466 (1989).
4. Pirrazoli, P. A. Mar. Geol. 109, 221–236 (1993). 5. Broecker, W. S. & Peng, T.-H. in The Global Carbon Cycle (ed. Heimann, M.) 95–115 (Springer, New York, 1993). 6. Sarmiento, J. & Le Quéré, C. Science 274, 1346–1350 (1996). 7. Petit, J. R. et al. Nature 387, 359 (1997). 8. Howard, W. R. & Prell, W. L. Paleoceanography 7, 79–118 (1992). 9. Ruddiman, W. F., Raymo, M. E., Martinson, D. G., Clement, B. M. & Backman, J. Paleoceanography 4, 353–412 (1989). 10. Oppo, D. W., Fairbanks, R. G., Gordon, A. L. & Shackleton, N. J. Paleoceanography 5, 43–54 (1990). 11. Howard, W. R. & Prell, W. L. Paleoceanography 9, 453–482 (1994). 12. Berger, A. L. Quat. Res. 9, 139–167 (1978).
Protein engineering
Reading, writing and redesigning Michael Groß and Kevin W. Plaxco
ature doesn’t have a protein-folding problem — but we do. Although we know that the line-up of the aminoacid building-blocks in a protein encodes the architecture of its correctly folded (native) form, we still cannot properly read or write this code. That is, we find it difficult to predict structures from sequences, or to design sequences that fold into a desired structure. In 1994, this dilemma prompted George Rose and Trevor Creamer1 to offer a cash prize for the solution to a specific proteindesign problem. And in this month’s issue of Nature Structural Biology, Lynne Regan and co-workers report2 that they have redesigned a protein into an entirely different fold by exchanging only 50% of its amino acids. This remarkable achievement has earned them the prize, less than four years after it was announced. Engineering prizes have, arguably, driven technological advances in the past — from the first human-powered flight across the English Channel (Kremer prize), to the first steps in nanotechnology (Feynman prize). The protein-design equivalent of the first cross-Channel flight is the ‘Paracelsus chal-
N
a
lenge’, named after the sixteenth-century Swiss physician, philosopher and alchemist. In an editorial in Proteins, Rose and Creamer proposed the challenge, pledging $1,000 for the first person to show a pair of proteins which have distinctly different topologies, but which retain no less than 50% sequence identity. Such pairs are not found in nature — databases show that pairs of proteins sharing 25% of their amino-acid residues usually have very similar structures3. An identity score of 50% between two natural protein sequences would convince any biochemist that the proteins must be nearly identical in structure. On the other hand, because the structural identity of a protein is determined by roughly a quarter of its building blocks, the task — which is a bit like turning this article into a report about a tennis match by exchanging only half of the words — is a realistic one. In the language of proteins, most of the words seem to be space fillers that do not convey a specific meaning, so the challenge is to find the important words and rewrite them. Regan and coworkers have met the challenge with style. Rather than simply converting a given pro-
b
B1 domain
c
Janus
Rop
Figure 1 The two faces of protein folding. Dalal et al.2 have risen to meet the ‘Paracelsus challenge’, by designing a pair of proteins that have completely different folds yet have no less than 50% sequence identity. a, The starting material was the B1 domain of the streptococcal IgG-binding protein G, which is mainly b-sheet. b, By modifying just 28 of the 56 amino acids in this region, they created a new protein which they called Janus. Janus has 50% sequence identity with the B1 domain, yet its mainly helical structure resembles that of the RNA-binding protein Rop. c, Rop, a homodimeric fourhelix bundle protein. Blue boxes, residues from B1; red boxes, residues from Rop; black boxes, residues that are identical in B1 and Rop; white boxes, residues from neither B1 nor Rop. Nature © Macmillan Publishers Ltd 1997
419
