Ancient temperatures inferred from DNA
"Where was you hid to see all that?" he cried. "It seems to me that you knows a great deal more than you should."? - The Complete Sherlock Holmes
"Our DNA is a coded description of the worlds in which our ancestors survived. And isn't it an arresting thought? We are digital archives of the African Pliocene, even of Devonian seas; walking repositories of wisdom out of the old days. You could spend a lifetime reading in this ancient library and die unsated by the wonder of it."? -- Richard Dawkins, Unweaving the Rainbow
Like many of the characters baffled by Sherlock Holmes, I am repeatedly amazed by the detailed inferences my fellow scientists are able to draw about events in the distant past. This week's paper:
Palaeotemperature trend for Precambrian life inferred from resurrected proteins
is a good example. Eric Gaucher and colleagues at the University of Florida and DNA2.0 Inc. used protein sequences from a variety of modern bacteria species to infer the protein sequences of their distant and more recent ancestors...
They then synthesized copies of those ancestral proteins and measured their stability at different temperatures. Based on the reasonable assumption that the ancestral proteins were most stable at the temperatures to which the ancestral bacteria were usually exposed, they calculated the average temperature of the environment where ancestral bacteria lived at various times in the past. They inferred that, about 3.5 billion years ago, bacteria were exposed to temperatures similar to hot springs, about 70 degrees Celsius (160 Fahrenheit). Similarly, they inferred that more recent bacteria, only (!) one billion years ago, saw lower temperatures,: about 40 C (100 F).
Every method has possible sources of error. For example, there is some disagreement on the details of the family tree of the bacteria, a key input to this sort of analysis. Therefore, the authors compared two alternative family trees. This gave only small differences in the temperature optimum of the earliest bacteria (65 vs. 73 C). Comparing two methods that are unlikely to have the same type of error, because they are based on unrelated assumptions, is even better. So they also compared their protein-based temperature estimates with estimates from stable isotope ratios -- relative amounts of two nonradioactive isotopes whose values in sediments are thought to depend on the temperature at the time the sediments settled out of the ocean. The rough agreement between the two different methods is much more convincing than either method alone.
FIg. 3 from Gaucher EA, Govindarajan S, Ganesh OK. (2008) Palaeotemperature trend for precambrian life inferred from resurrected proteins. Nature 451: 704-708 (Copyright NPG; noncommercial reproduction under Fair Use rule.)