13.2: Luria and Delbrück- Discovering the origin of mutations

Darwin and Wallace lacked a clear understanding of where genetic variation came from, how it might be stored or passed on from one generation to the next. An important question that arose early in the history of evolutionary studies was whether mutations were due to chance (stochastic) events or whether they were somehow purposefully generated in response to the "needs of the organism". Darwin assumed that evolution involved random variations that arise in individuals. A Lamarckian mechanism involves induced (and purposeful) responses by individuals.^[458] In the absence of a clear understanding of how genetic information is stored, replicated, and passed from generation to generation, there was really no way to distinguish between Darwinian and Lamarckian mechanisms, although Lamarckian mechanisms seemed more direct.^[459]

To understand how this question was resolved, consider the classic experiment, known as the Luria-Delbrück experiment after Salvador Luria (1912-1991) and Max Delbrück (1906-1981), the scientists who carried it out.^[460] Their study was published in 1943, before DNA was recognized as the genetic material and well before anyone understood how genetic information was stored.^[461] Luria and Delbrück examined the resistance of bacteria to viral infection. They used bacteria that could be infected and killed by a specific bacteriophage. Mutations arose spontaneously. When they appeared they rendered the bacteria, and their off-spring, immune to viral infection. The question Luria and Delbrück asked was, do phage resistance mutations appear randomly all of the time or does the presence of the virus 'induce" their appearance. Is immunity learned or lucky?^[462] If the generation of phage resistance mutations is adaptive, then we would expect that the frequency of resistance producing mutations will be more or less uniform from one population to the next – repeating experiments on different cultures should produce resistant bacteria at approximately the same rate in each (top panel next↓page). If, on the other hand, the mechanism occurs by chance, that is stochastically (middle panel ↓), we would expect that the number of mutational events will vary dramatically from one population (culture) to the next - the variation in the frequency of phage resistance (and the mutations that produce it) between independent populations will be large.

Luria and Delbrück started a number of bacterial cultures to which they then added enough virus (at the time of the horizontal red line in the top two panels) to kill every sensitive bacterium. They then plated out the cultures and counted the number of phage-resistant bacteria present, each of which grew up into a macroscopic clonal colony. The number of phage resistant cells in a culture reflects when, in the history of the culture, the resistance mutation appeared. If the resistance mutation appeared early in the history of the culture, as in the red-boxed culture (→) it would be common, whereas if it appeared late, it would be rare. The two models (induced/Lemarckian versus random/Darwinian) make dramatically different predictions. In the induced/Lemarckian model, the variation in the numbers of resistant bacteria between cultures is expected to be low, since resistance arises through a common “inductive”, physiological process, even though we do not know how that process works. In contrast, in the spontaneous/Darwinian model we expect large variations, with many cultures having no resistant bacteria and some having many. When the mutation occurs late, or not at all, as in lower panel, population 2, there will be few or no phage resistant cells. If the mutation occurs early there will be many resistant bacteria. Luria and Delbrück calculated what the two models predicted. The observed results (black bars) matched the prediction for the spontaneous/Darwinian mechanism, leading them to conclude that, at least in this system, mutations occurred independently of the presence of the virus.

To date there is no evidence that environmental factors can specifically induce the generation of beneficial or useful mutations. What can happen, however, is that the general (non-specific) mutation rate can increase in response to various stress conditions, arising from internal or environmental effects. Typically an increased mutation rate involves effects on the efficiency of DNA error repair systems, which leads to increased levels of genetic variation upon which selection can act.^[463]

The ability to control mutation rates occurs within the vertebrate immune system, through a process known as somatic hypermutation.^[464] This process is involved in the maturation of the immune response and the generation of increasingly specific antibodies, a topic well beyond our scope here. That said, the mechanism is known; these cells activate a gene that encodes an “activation-induced deaminase” or AID (OMIM:605257). The AID protein acts on cytosine residues in DNA to generate uracils that, when repaired, replace the original C:G base pair with an A:T base pair. The other genes in these cells appear to be at least partially protected by “selective targeting of AID and gene-specific, high-fidelity repair of AID-generated uracils”.^[465]

---