Here’s an interesting question: how long can living organisms survive in hibernation? We know that many animals such as bears, squirrels and bats hibernate seasonally, but I am talking about bacteria, which can enter a state of suspended animation and survive for very long periods. By “very long” I mean possibly millions of years. Hundreds of millions of years.

This question is interesting for several reasons. First, just what kind of chemistry does it take to do that? We know that DNA degrades over time. The random influences of temperature, radiation, or environmental chemistry ought to degrade any huge and complex molecule over such long time spans. How could bacterial DNA survive intact for millions of years?

Second, if it is true that bacteria can survive so long, what a remarkable opportunity this would be to investigate the process of evolution! Consider the organisms alive today. Although it is generally believed that archeobacteria found in certain inhospitable environments (such as mid-ocean ridges, or deep beneath the Earth) are the oldest forms of life, they can offer only limited insight into what early life might have been like. All forms of life continue to evolve, including archeobacteria. Although such bacteria look very primitive to us, they have also evolved and changed continuously for 3-4 billion years. Who knows what they looked like 3 billion years ago? Or even 100 million years ago? How could we know, unless we found a 100 million year old bacterial spore, and were able to sequence it? Since bacteria do not reproduce while in the spore state, they are not evolving. Their DNA is essentially unchanged. Such an organism could provide a 100 million year old snapshot of evolution, frozen in time. If we could then sequence its genome and compare it to similar life forms today, we would see how a hundred million years of evolution have brought about their changes.

It is also interesting from the point of view of panspermia – the hypothesis that life was seeded to Earth from space. Or even the reverse – life from Earth reaching other planets or other extraterrestrial habitable environments. We know that certain things (like large impact events) can propel material into space at speeds exceeding escape velocity. However, because of the immense distances involved, such material could take thousands of years to reach another solar planet, or millions of years to reach nearby stars. But if there are bacteria that can hibernate for millions of years, then such ideas become more practical.

I was reminded of these questions today when I read a story on New Scientist about bacteria found on the sea floor near Svalbard, an island off the coast of Norway. These are arctic bacteria, so you’d expect them to grow at cold temperatures. However, when the scientists incubated them, they had a surprise. In addition to the expected growth peak at about 20 °C, they observed a second growth peak at 55 °C.

Obviously, the sample contained different types of bacteria, one of which was thermophilic, with spore activation at 50 °C and peak growth at 55 °C. Why do these bacteria exist on the arctic sea floor, where the temperature never reaches 55 °C?

The DNA of these bacteria showed that they resemble bacteria known to exist deep under the Earth’s crust, where the temperatures are warmer. Typically, such bacteria are found associated with petroleum deposits. This might just be coincidence though, since most of our drilling underneath the sea floor is to look for oil. At any rate, the conditions in their normal habitat are hot and anoxic. So what were they doing on the arctic sea floor?

The temperature at the deep sea floor is near zero. It continues to rise in the Earth below the sea bed, at a rate of about 2-3 °C per 100 meters depth. So the normal habitat of these bacteria, with a peak growth rate at 55 °C, would be about 2 to 3 kilometers beneath the sea floor. Note that there are bacterial species that live at different depths below the sea floor – mesophiles, thermophiles (like these bacteria), and hyperthermophiles (in even deeper and more ancient sediments). It has been estimated that about 1/2 to 5/6 of the Earth’s entire bacterial biomass lives in the Earth beneath the sea floor[ref 1]. Some studies estimate that 1/10 to 1/3 of the Earth’s entire biomass is contained in these sub-sea-floor bacteria. So we are talking about very large bacterial populations, and an extremely diversified ecosystem.

There are a few different ways in which bacteria from deep sub-sea-floor regions could move to the ocean floors, and vice versa. The ocean floor constantly accumulates sediment, and therefore anything on the sea floor will be eventually buried deep beneath the sea floor. But this process takes millions of years, even hundreds of millions of years. So if these bacteria are in their normal life cycle, it would take up to a hundred million years for them to be buried deep enough to have a satisfactory environmental temperature in which to germinate. Can they live that long?

Another, somewhat faster process, involves the circulation of ocean water into the deep sediments below the sea floor. This takes about a million years, which is much faster the burying of sediments, but still pretty long. This could give the microbes a boost, take them at least part way through their journey down to 2500 meters. It could also explain how they happened to end up on the sea floor, through the deep sea bed circulation.

The article itself [PDF of the full article can be found here] concludes that the bacteria are engaged in their normal life cycle. Depending upon which cycle we are talking about (burial under accumulating sediments until they reach their target depth, or movement through oceanic sub-sea-floor circulation, or some combination of the two), this would mean that the normal life cycle for these bacteria ranges from a million to possibly a hundred million years.

There are other putative examples of extremely long-living bacteria, but the claims tend to break down upon further examination. For example, Vreeland and others found a species of Bacillus enclosed in the brine inside a salt crystal in the Salado salt formation in New Mexico [full PDF here]. These crystals are from the Permian, 250 million years old. Their team was able to successfully revive the bacterial spores and grow them in culture, and named them Bacillus strain 2-9-3, subsequently called B. permians. However, it was not clear whether these are actually 250 million year old bacterial spores trapped inside the salt crystal, or if they are bacteria that only recently migrated into the salt crystal.

There are some studies that cast doubt on the age of the bacteria. The gene for the 16S ribosomal subunit was sequenced, and has been compared to the same gene from modern day halophilic bacteria. One study reports that “The B. permians sequence differs from that of S. marismortui by only one transition and one transversion out of the 1,555 aligned and unambiguously determined nucleotides” [ref 2].  S. marismortui is a moderately halophilic species from the Dead Sea. It seems very unlikely that a 250 million year old bacterium would differ from a modern species by only 2 mutations on that gene. Phylogenetic studies comparing B. permians to closely related organisms do not place B. permians at an ancestral position in the phylogenetic tree. So at best, there are serious doubts about the 250 million year claim.

Another example is the claim by Cano et al that they revived spores from many bacterial species found in the stomach of a bee that had been trapped in amber. The amber was mined in the Dominican Republic, and is 25-40 million years old. One of these strains was identified as Bacillus sphaericus, which is a common species even today, and is often found inside the bodies of insects. B. sphaericus is listed as a single species, but like many bacteria, it is really a complex of several different subgroups. There are at least 5 recognizable DNA subgroups in various strains of B. sphaericus. The type isolated from the ancient amber was named BCA16, and eventually 1482 base pairs from its 16S ribosomal subunit gene were sequenced. These have been compared to various types of modern B. sphaericus.

There is about 80% similarity to one modern strain – NRS 592 [ref 3]. This sort of leaves the question open – it’s not possible to definitely conclude that the spores found in the amber are millions of years old, but it’s not possible to rule it out either. Interestingly, the NRS 592 type that is most similar to the ancient strain is not primarily an insect pathogen. Other types of B. sphaericus which are known to be insect pathogens today share much smaller similarities to BCA16.

Another species isolated from ancient amber by Cano and his group was tentatively named Staphylococcus succinus [full PDF here]. The 16S ribosomal subunit gene for this bacterium differs from its closest modern day homologue (a urinary tract pathogen called Staphylococcus saprophyticus) by 19 substitutions out of 1525 aligned base pairs. This is certainly far larger than the reported homology for B. permians, which is supposed to be even older (250 million years). I don’t know if this is enough to support the ancient origin of S. succinus, but it certainly seems to deny the claim for B. permians. If a 25-35 million year old bacterial spore differs from modern species by 19 substitutions on that gene, how could a 250 million year old spore only differ from modern relatives by only 2 substitutions on the same gene?

There are a number of other similar claims, which I will not cover. Here are some of the papers if you are interested:

DNA sequences from a fossil termite in Oligo-Miocene amber and their phylogenetic implications. DeSalle R, Gatesy J, Wheeler W, Grimaldi D. Science 1992 Sep 25; 257 (5078): 1933-6.

This was one of the first papers about supposedly ancient DNA (25-30 million years old). Unlike the previous examples, this is not bacterial DNA, this is from a termite, which makes it harder to believe. On one hand, the fossil was preserved in amber, which is good. On the other hand, it’s not a spore, it’s a whole multicellular organism, which makes it hard to believe that the DNA could survive that long. The 18S ribosomal subunit gene was sequenced, and was found to be similar to modern termites and roaches.

Amplification and sequencing of DNA from a 120-135-million-year-old weevil. Cano RJ, Poinar HN, Pieniazek NJ, Acra A, Poinar GO Jr. Nature 1993 Jun 10; 363(6429): 536-8.

This one is from another insect, the weevil. This is even older, the weevil was found in Lebanese amber that is 120-135 million years old. Both 16S and 18S ribosomal unit genes were sequenced. Two short fragments from the 18S show that it’s possibly from some extinct weevil.

PCR jumping in clones of 30-million-year-old DNA fragments from amber preserved termites (Mastotermes electrodominicus). DeSalle R, Barcia M, Wray C. Experientia 1993 Oct 15; 49(10): 906-9.

More from the DeSalle group, also 30 million year old termites preserved in amber.

Here are some papers that talk more generically about extracting ancient DNA:

1. Very old DNA. DeSalle R, Grimaldi D. Curr Opin Genet Dev 1994 Dec; 4(6): 810-5.
2. Ancient DNA: using molecular biology to explore the past. Brown TA, Brown KA. Bioessays 1994 Oct; 16(10): 719-26.
3. Implications of ancient DNA for phylogenetic studies. DeSalle R Experientia 1994 Jun 15; 50(6): 543-50.
4. Rapid isolation of DNA from fossil and museum specimens suitable for PCR. Cano RJ, Poinar HN. Biotechniques 1993 Sep; 15(3): 432-4, 436.
5. The range of life in amber: significance and implications in DNA studies. Poinar GO Jr. Experientia 1994 Jun 15; 50(6): 536-42.

So how plausible are these accounts of finding ancient DNA and being able to sequence it? Many modern microbiologists consider it very implausible. The main reason is that DNA is a very large and complex molecule, which degrades rapidly in the environment. The problem is radioactivity in the soil and rocks. Even small levels of radioactivity which are naturally found everywhere, should degrade DNA within a few hundreds of years. Bacterial spores contain no active DNA repair enzymes. How could the DNA survive that long?

A bacterial spore is a remarkably hardy entity. The thick walls offer considerable protection to the contents. The core, which contains the nuclear material and other cellular machinery is very low in water content, which also helps in preserving the material.

We know that live bacteria can take a lot of abuse. For example, Deinococcus radiodurans can stand cold, vacuum, acid, dehydration, and massive amounts of radiation. It has been shown to tolerate 10,000 grays of radiation (5 grays are lethal to humans). It’s not obvious why such an organism should evolve – after all, there are no habitats on Earth with such extreme radioactivity. The highest natural radiation is found in some areas of Iran, and it’s only about 260 milligrays per year. A year is far in excess of a bacterium’s lifetime, and even so the total cumulative radiation in a year would be far below the capacity of this organism to tolerate.

Some people have suggested that resistance to radiation and resistance to drying use the same mechanism, so the bacteria were really selected for resistance to drying, and resistance to radiation was just a side effect. This is certainly possible, but the fact remains that many of the mechanisms for the resistance depend upon the organism being alive and active. D. radiodurans, for example, has multiple copies of its genome for redundancy, and very active DNA repair enzymes. It has special mechanisms for annealing split strands. However, these are all active mechanisms, and don’t occur in bacterial spores.

What about passive mechanisms? We know that in D. radiodurans, the DNA is very tightly coiled into toroids. Does this confer some extra resistance? Some studies have found high levels of manganese associated with the DNA. Manganese complexes can act as antioxidants. Some researchers have also suggested that the chief damage from ionizing radiation is not to the DNA, but rather to the associated proteins. Manganese complexes could be protecting the proteins. These passive mechanisms would work in bacterial spores, but are they enough to ensure survival for millions of years? Who knows.

The amber fossils may be similarly protected. We know that amber slows down the degradation of biological material. This has been shown in studies on proteins. Proteins in all life forms on Earth are made of amino acids which are “left handed”, that is, they are levorotatory stereoisomers. However, simply over time, all amino acids racemize – spontaneously convert from one stereoisomer to the other. In living organisms, this is not a problem, because proteins are constantly being destroyed and replaced. Since all new proteins are formed with levorotatory amino acids, this is the predominant stereoisomer seen in living organisms.

However, after the cell dies, no new proteins are being synthesized. In time, all the remaining amino acids racemize, and after an extended period of time, the amino acids will be about half and half – roughly 50% in the levo form, and the other 50% in the dextro form. But it has been shown that this process of racemization is significantly slowed down if the organic material is preserved in amber. So if amber can protect proteins, perhaps it can protect DNA as well. Again, the question is for how long, and under what conditions. Could DNA protected by amber, then buried in inert rock or inside a salt crystal, with a low ambient radioactivity level survive for a long time? For millions of years? How about a whole bacterial spore, presumably with manganese complexes, or something functionally equivalent that preserves it?

I don’t think this question has been answered yet, but it seems like people are slowly chipping away at the notion that DNA (or bacterial spores) couldn’t survive beyond a few hundred years.

References:

1. Leg 201 synthesis: Controls on microbial communities in deeply buried sediments. Jørgensen, B.B., D’Hondt, S.L., and Miller, D.J. (Eds.). Proceedings of the Ocean Drilling Program, Scientific Results Volume 201. [PDF]

2. The Permian Bacterium that Isn’t. Graur, D and Pupko, T. Mol. Bio and Evolution. (2001) 18(6): 1143-1146. [PDF]

3. DNA similarity analysis of a putative ancient bacterial isolate obtained from amber. Yousten, AA and Rippere, KE. FEMS Microbiol Letters. 152 (1997): 345-347. [PDF]