. Retrospective birth dating of cells in humans. Cell. 2005 Jul 15;122(1):133-43. PubMed.


Please login to recommend the paper.


Make a Comment

To make a comment you must login or register.

Comments on Primary Papers and News

  1. About 40 years ago, Joseph Altman and his colleagues used the 3H-thymidine autoradiographic method to birthdate cells in the brains of adult rats and cats, and reported evidence for neurogenesis in the olfactory bulb, hippocampus, and neocortex (1-5). These findings were corroborated and extended by Michael Kaplan and his colleagues by combining 3H-thymidine autoradiography with electron microscopy (16-19). Despite this rather large body of work, the findings were not well-received by the neuroscience community, and adult neurogenesis in the mammalian brain remained a matter of debate (17,27, 28). In the late 1990s, the use of a newer technique, bromodeoxyuridine (BrdU) labeling, in combination with confocal microscopy, led to new evidence in support of this phenomenon and the acceptance of adult neurogenesis in the olfactory bulb and the hippocampus (11,13,14,24). These two areas have been named “neurogenic” because they are now widely believed to support adult neurogenesis. However, similar evidence has been reported for “non-neurogenic” brain regions, including the neocortex, striatum, amygdala, hypothalamus, and substantia nigra (6,7,9,12,30,32). One of the most disputed claims is that of adult neurogenesis in the neocortex. We and others have reported positive evidence for adult-born neurons in the neocortex of rat and monkey (7,9,13,15). However, others have not corroborated these findings (20,22,23), or have found evidence for adult neurogenesis only under conditions of damage (21,25,31). Methodological flaws have been proposed as explanations for putative false positive and false negative data on this subject (13,26). The development and application of new methods is clearly needed to resolve the debate about adult neurogenesis in neocortex and other “non-neurogenic” brain regions.

    The recent publication in Cell by Spalding and colleagues (29) reported the use of a novel and highly innovative method to search for adult neurogenesis in human postmortem brain tissue. This method was designed with the following facts in mind: 1) Levels of atmospheric C14 were very high during the time of intensive nuclear weapons testing (in the mid 1950s to early 1960s); and 2) carbon atoms in the DNA of cells do not turn over. Thus, the relative content of C14 in a population of neurons, particularly those generated during or immediately after the weapons testing, can be used to birthdate the cells. This method can be used for examining cell populations, but not individual cells, because C14 incorporation into DNA is a rare event, even at the time when C14 atmospheric levels were very high. Nonetheless, for large homogeneous populations of cells, an average age can be estimated. In tissues comprising multiple cell types, cells can be sorted and C14 levels can be assayed in subpopulations. Using this approach, the average age of cells in two regions of the adult human brain autopsy tissue—the cerebellum and occipital cortex—were estimated. The authors found that cells in the occipital cortex were younger than cells in the cerebellum, but when neurons and non-neurons were separated, the occipital cortex neurons were the oldest, almost as old as the individual. Thus, the findings support the view that cortical neurons are generated during development. The authors leave open the possibility that adult neurogenesis occurs in brain regions that were not examined in the study.

    While it might appear that these findings rule out the possibility of adult neurogenesis, at least in the occipital cortex of humans, there are some important points to consider. The first is whether or not this method is sufficiently sensitive to detect adult neurogenesis in any brain region. This can be tested by examining an area where this phenomenon is well-established and occurs at a relatively high rate. For the human brain, there is only one such area—the hippocampus. In 1998, Eriksson and colleagues reported evidence for adult neurogenesis in the hippocampus of adult human cancer victims, using BrdU labeling (11). Spalding and colleagues (29) did not examine the hippocampus in the present study, so this issue will require further experimentation.

    The second question is whether a substantially lower rate of adult neurogenesis could be detected with this method. Studies reporting new neurons in putatively “non-neurogenic” brain regions indicate that the rates of neuronal addition are very low relative to those in the hippocampus (8,9,15,32). Since carbon dating does not allow for the birthdating of individual cells, a relatively large proportion of adult-born neurons would be necessary to detect an average age difference. The authors suggest that detectability limits are about 1 percent for their method (29), so one out of 100 neurons must be produced in adulthood to find adult neurogenesis in a given region. If adult-generated neurons survived for long periods of time, accumulating over decades, then the 1 percent detectability limit may be sufficient. However, studies in animals report that adult-generated neurons have a relatively short lifespan and many new neurons appear to die shortly after their production (10,15). This would preclude the ability to detect an accumulation of adult neurogenesis over a very long period of time with a method that cannot examine individual cells.

    In summary, the paper by Spalding and colleagues (29) presents a creative and novel approach to investigating the age of neurons in the brains of humans born around the time of nuclear bomb testing. Further characterization and refinement of this method will undoubtedly provide useful information about brain development and function. In the meantime, we can conclude that most, if not all, neurons in the occipital cortex of humans are generated during development. The possibility for adult neurogenesis in this area remains only if its rate is exceedingly low and the new neurons do not have a lengthy lifespan.

    View all comments by Elizabeth Gould

This paper appears in the following:


  1. Dating the Birth of Human Cells—Carbon 14 Runs Rings around Competition