Formation of new neurons (called neurogenesis) is considerably reduced in adults compared to infants. Considering the fact that new neurons would have to compete with the old ones for the space and connections (i.e. synaptic connections that are present between neurons), it makes sense that the old connections might sometimes be replaced by the new ones during neurogenesis. Consequently, if the old connections were storing memories then their replacement would fragment or completely delete those memories. If we take a step further this would imply that, compared to adult mice, the young mice would not be very good at making long-term memories because in their brains active neurogenesis is taking place, old neurons are replaced, and new connections are made- a process which should delete all the things that were stored in the old neuronal network.
Consistent with these ideas is the observation of ‘infant amnesia’ in many animals. Namely, animal infants, including human, forget things more readily than adults (can you remember your first, second or third birthdays?). With these premises scientists decided to investigate if neurogenesis could be the cause of infant amnesia in mice and other animals.
Firstly, to prove that the adult mice have reduced neurogenesis in dentate gyrus (part of brain within hippocampus that is linked to memory formation and storage) they injected a retrovirus into dentate gyrus (the retrovirus is a type of virus that will stably integrate into cell’s DNA). The retrovirus was carrying a green fluorescent protein gene (GFP), which after integration will be expressed by the cell. If neurons are dividing then the number of neurons with GFP signal will increase because it will be passed on to new cells from their progenitors but if cells don’t divide then the signal remains stable. The scientists show that in infant mice GFP signal increases and in adults it does not.
Next the study looked at formation of dentate gyrus-dependent memory in young and old mice. Old mice and young mice were trained by placing them in a specific room and giving a series of foot shocks. 28 days later the two age groups were placed in the same room (but no shock was given) and their reaction was observed. The old mice all froze in place in that room indicating the recollection of the shock-memory, however, the younglings showed no response suggesting that they have forgotten the link between the room and the foot shock training.
Okay, so now we know that old mice have little neurogenesis and that they remember the past events better than young mice. But no link between the two events has yet been shown. So, the next step is to increase the neurogenesis in old mice and see if they start forgetting things. As it turns out, neurogenesis in old mice can be increased by as much as 50% simply by making them exercise, namely running in a scroll-wheel. So after several days of exercise , the amount of neurogenesis was measured in the old mice by the same GFP method and an increased neurogenesis was found. Now again the same foot-shock experiment was done: mice are given foot-shocks, put into space with scroll-wheel (mice like to run so they are sure to exercise on it) and placed back into the same room where the foot-shocks were given. As predicted adult mice showed reduced freezing behaviour because of increased neurogenesis, moreover, if the mice exercised before foot-shock training the old mice were still able to recall the training well, indicating that the neurogenesis after the training prevented stable memory formation. The same results were also confirmed by administering drug that induces neurogenesis (and therefore led to forgetting in old mice) and using transgenic mice in which neurogenesis can be inhibited with a drug (the inhibition led to higher rates of remembering in mice that were allowed to exercise as well as the infant mice).
Different animals are born at different stages of brain maturity depending on the length of gestation period. For example, in guinea pigs gestation takes 65 days compared to 21 days in mice, so therefore, guinea pigs should have lower neurogenesis rates as their development after birth is already quite advanced in time. And indeed, when scientists used the same foot-shock experiment to test patterns of memory recall in young and old guinea pigs they found no difference between them. Not surprisingly, injection of drug that increases neurogenesis in guinea pigs led to decreased memory formation.
In conclusion, the study has determined that neurogenesis contributes to the loss of dentate gyrus-linked memory formation. Other studies, however, have previously shown the reverse effects of neurogenesis, i.e. suggesting that it improves memory. Overall, the picture now seems to be that neurons at different neurogenesis stages might play different roles in memory formation; for example, sufficiently mature neurons could provide more space for new memories. It sort of make sense, I mean if we extrapolate it to humans (and I know that from animal models this might not always be the best thing to do), for infants long-term memories are not as important because parents will provide the missing context for them. As we mature neurogenesis decreases and memories become more stable, while a low amount of new neurons formed contribute to an increased requirements for memory storage space. And for those who might now be scared to exercise because they might start forgetting stuff, well again the neurogenesis in adults is very low and a bit of increase in it thus far seems to be only a good thing. This study, for example, shows that exercise caused increase in hippocampus volume reverses the effects of ageing-linked memory loss and even improves it!
Akers KG, Martinez-Canabal A, Restivo L, Yiu AP, De Cristofaro A, Hsiang HL, Wheeler AL, Guskjolen A, Niibori Y, Shoji H, Ohira K, Richards BA, Miyakawa T, Josselyn SA, & Frankland PW (2014). Hippocampal neurogenesis regulates forgetting during adulthood and infancy. Science (New York, N.Y.), 344 (6184), 598-602 PMID: 24812394