On November 2nd, the Neurotechnology Industry Organization revealed its top 10 neuroscience trends for 2007. While most of these trends had to do with technology, which makes sense because the company has a vested interest in neuroscience technology, trend #9 was that “new research continues to link neurogenesis to treatment of depression.” 
While some researchers would question the tenability of that statement, the NIO is correct in its assessment that this is one of the “hottest” topics in neuroscience today. The reasons for this are two-fold. The first reason is that neurogenesis is still a relatively new idea with broad-ranging implications. If the brain can create new functional cells in some regions of the brain, then one might speculate that it will eventually be able to create cells for any region in the brain. An understanding of the adult stem cells in the brain might bring us one step closer to that elusive fountain of youth. The next reason that the potential link between neurogenesis and depression is considered so important is that major depressive disorder is a widespread pathological disorder, affecting 16.2% of US adults at some point in their lifetime (Kesller et al., 2003). Novel models of understanding the disease could lead to improved pharmaceutical drugs capable of treating depression more acutely. The neurogenesis hypothesis is one model that has been suggested for helping to explain how unipolar depression forms and how it may be treated. This stated hypothesis is accepted to have two components: one, that deficiency of neurogenesis may be culpable for the onset of depression, and two, that current methods of treating depression may work in large part by fixing an abnormality in neurogenesis to the hippocampal region (Sapolsky, 2004). To test this neurogenesis hypothesis, a number of possible correlations between what we know about depressive disorder and what we know about neurogenesis need to be addressed. Also, a number of questions isolating the mechanisms behind neurogenesis and depression need to be answered. This paper will attempt to review the testing of the neurogenesis hypothesis.
One of the fundamental assumptions of the neurogenesis hypothesis is that the hippocampal region in general is crucial to the development and treatment of depression. One attempt to find a correlation here is based on the idea that subjects who show depressive symptoms will show variance in the sizes of their hippocampal regions. Videbech and Ravnkilde (2004) conducted such a survey of studies which had measured hippocampal volume using MRI machines. They ensured that each study had identified which subjects qualified as having a major depressive disorder using a replicable method, and that each study had controlled for other variables such as age and drug abuse. They found that there was a significant effect size which correlated depression to a decrease in the volume of the hippocampus in both hemispheres of the brain. Since neurogenesis would be expected to increase the volume of the hippocampus, their results suggest that neurogenesis may have a role in depression. This paper will touch more on this point later. But while there is a correlation between the size of the hippocampus and depression, does the idea that the hippocampus causes or affects the symptoms of depression make intuitive sense? One of the commonly accepted symptoms of major affective disorder is cognitive impairment. Sweeny et al. (2000) tested groups of unipolar depressed subjects, bipolar subjects, and control subjects using computer programs designed to determine cognitive aptitude. Their study found that unipolar depressive patients predominantly showed deficiencies in episodic memory, but not much else. This is consistent with the known functions of the hippocampal regions in the brain (Ebmeimer et al, 2006). This link provides indirect evidence that the hippocampus plays at least somewhat of an important role in major depressive disorder. However, given how complicated the connections between known brain regions are known to be, it would be difficult to pinpoint exactly how much responsibility for depressive disorder can be placed on the hippocampal region of the brain.
However, just because the hippocampus plays a role in depression does not mean that neurogenesis in the hippocampus plays a functional role in the disease. Indeed, in past it has been somewhat unclear whether the newly generated neurons are even able to integrate themselves into the complicated neural networks of the hippocampus. Recently it has become more and more apparent that the new neurons are able to integrate themselves until existing neural networks in the hippocampus. Henriette van Praag et al. (2001) attempted to show this very fact—that stem cells can integrate and begin to act just as mature granule cells do in the dentate gyrus of the hippocampus. In order to do so, the researchers used a variety of staining techniques, including the use of green fluorescent protein (GFP) and 5-bromodeoxyuridine (BrdU). They then allowed their adult mice to live for a calculated period of time before they were killed: either a short amount of time to see how the cells are proliferating immediately after staining, or after a longer amount of time to see how the cells are integrating themselves into the neural network of the dentate gyrate. Based on the dendrite spine length of the stem cells, among other factors, they found that the cells were able to integrate themselves into the neural networks of the dentate gyrus. So it is possible for neurogenesis to lead to new functional neurons, which makes one wonder exactly what purpose, if any, these cells have in the brain.
The next piece of evidence surrounding the neurogenesis hypothesis is how it fits into the way that antidepressant treatments work. While the rationale that they work on serotonin reuptake and monoamine oxidase inhibitors is widely accepted, the fact that these drugs generally take two to three weeks to start working is not an obvious result. Some complex mechanisms of serotonin reuptake and monoamine oxidase inhibitors have been suggested to explain this phenomenon (Celada et al, 2004). However, the neurogenesis hypothesis could simplify our understanding of how the antidepressant drugs work. Indeed, it makes a certain amount of intuitive sense that the new neurons generated by the antidepressants would require a delay before they could begin to work themselves into the circuitry of the hippocampus. Some studies have found that new neurons may begin to be important for trace memory in the hippocampus after just 1-2 weeks (Shors et al, 2001). Zhao et al (2006) studied the development of newly formed granule cells in the dentrate gyrus of adult rats and mice. In order to do so, they injected GFP viruses into adult animals to indentify which cells had been newly formed at the time of injection. They then killed the animals at varying points of their development, isolated the cells that had been marked by the virus, and analyzed the stages of development that each of these cells had gone through. Based on how developed (ie, how long) the dendritic spines of the neurons were at the time of death, the researchers in the study attempt to draw conclusions about how they functioned in the animal’s brain at the time of death. They found that the dendritic spine had already started to grow by around 16 days in both adult rat and mice brains. These results, combined with the results of the previous study that new neurons help form trace memories after just 1-2 weeks, suggest that animals may be able to use the neurons to form new memories very soon after they appear in the dendrite gyrus. While abstracting results from animals to humans is tricky, the animal results suggest that the stem cells could begin to integrate themselves into the hippocampus around 2-3 weeks after they are formed, near the same time that the antidepressant drugs begin to work. Therefore, the neurogenesis hypothesis could help to simplify one of the not-so-well understood mechanisms of how andippressant drugs work, providing some indirect evidence for the neurogenesis hypothesis, since scientists have been taught to revere Occam’s razor.
The other piece of correlative evidence that supports the neurogenesis hypothesis is the relationship between stress and depression. There is a lot of data behind the theory that external stressors can lead to depression, and behind the idea that depressed subjects are less able to raise their cortisol levels when faced with a challenge (Miller et al, 2005). It has also been shown in double-blind studies that antiglucocorticoids can act as antidepressants if the patient is diagnosed as hypercortisolemic, which many depressed patients are (Wolkowitz et al, 1999). Additionally, excess of adrenal steroid and stress appear to damage neurons in the hippocampus, negatively affecting long-term potentation of neurons, cognition, and memory (Pavlides et al, 2002). Based on this relationship, scientists became curious as to whether or not stress inhibited neurogenesis of the hippocampus. Gould et al (1998) tested whether or not the growth of granule cells in the dendrite gyrus of adult primates would be affected by stressful experiences. In order to create a stressful experience, the researchers placed one male marmoset monkey in the cage of another male of the same species for an hour. The monkeys reacted to this by behaving subordinately: they remained still in one part of the cage in order to avoid a fight with the male that they had been placed with. This was considered a reasonable induction of stress in the “intruder” monkey. After the hour, the monkeys were removed from the cage and injected with BrdU, which acts as a marker of proliferating cells and their offspring. Two hours later, once the stem cells had a chance to form, the monkeys were killed, and sections of their brains were analyzed. The animals that were placed under the stress were then compared to control monkeys that had not been put in the stressful situation, and the monkeys that had been not inducted to the stress were found to have more stem cell proliferation. This result helps to illuminate the fact that stress is a potent inhibitor of neurogenesis, at least in non-human animals. However, since it would probably be unethical to force a human to undergo a very stressful experience in an experimental design, there is less evidence for this phenomenon in human subjects. If stress did not cause an inhibition of neurogenesis, that would be solid evidence against the neurogenesis hypothesis. As it is, the fact that stress might inhibit neurogenesis provides some more correlative evidence in support of it.
As stated earlier in this paper, one of the more widely-accepted pathological symptoms of depressive disorder is that it decreases the volume of the hippocampus. However, the mechanisms that lead to this decrease in volume are not well understood. If they were found to be due to an inhibition of neurogenesis, perhaps from stress, this result would support the neurogenesis hypothesis. If the reasons for the decrease in volume were found to be from some other source, perhaps the death of existing hippocampal cells because of a lack of necessary nutrients, then it would provide support against the neurogenesis hypothesis (Sapolsky, 2004). Unfortunately, there is little data on the human central nervous system, in large part because it would be highly unethical to perform the stain and then brain slice technique that is used in animals, and in part because data on post-mortem tissue has thus far yielded inconclusive results (Feldmann et al, 2007). Despite the dearth of information, some scientists have attempted to make a few estimates about how widespread the neurogenesis is. Cameron and McKay (2001) posited that the amount of BrdU marker that had been used in previous studies was not ample, and only a portion of the stem cells were being found. Using a higher concentration of BrdU along with a thymide marker on adult rats, they found that over 9000 new stem cells were produced in the dentate gyrus of the hippocampus, for a total of over 250,000 a month. As Gould and Gross (2002) point out, this is out of a total of 1 to 2 million total neurons in the dentate gyrus of the adult rat, suggesting that the newly formed stem cells play a large role in the hippocampus. However, once again, it is difficult to extrapolate this data to humans, who could have entirely different amounts of cells in their dentate gyrus, and different amount of new cells formed each day. Even if hippocampal volume were shown to be lowered because of impairment in neurogenesis, this data would not prove a causal both parts of the neurogenesis hypothesis. Instead it must be shown that neurogenesis has an impact on the symptoms of depression (Feldmann et al, 2007), otherwise the data will remain simply correlative.
One study that attempted to get to the core of the neurogenesis hypothesis looked at the development of learned helplessness behavior in rats. Vollmayr et al. (2003) looked at whether or not the proliferation rate of new stem cells in the dentate gyrus would vary in animals that showed symptoms of learned helplessness versus those that did not. They subjected groups of rats to inescapable foot shocks, and decided based on their behavior whether they were displaying learned helplessness or not. The newly forming cells in the dentate gyrus were marked with BrdU, and the amounts of these cells were compared between those animals that had undergone learned helplessness and those that had not. In order to support the causal portion of the neurogenesis hypothesis, these results should have found that those animals that had experienced learned helplessness would have less cells, because the depression-like symptoms should have impaired their neurogenesis. Instead, the researchers found no significant difference between the two groups, and given that the sample size (200) of the study was so high, these results hurt the case for the neurogenesis hypothesis substantially. Additionally, there is some evidence that a decrease in neurogenesis will not necessarily lead to depressive symptoms, as it should based on the causal portion of the neurogenesis hypothesis (Malberg and Duman, 2003). These researchers also made their finding using a learned helplessness model in rats through foot shocks. This evidence seems to discount the possibility that there is a direct causal link between neurogenesis and depression. However, just as it was important to express caution over optimistic results for the neurogenesis hypothesis that relied on animal models earlier, it is important not to be too pessimistic about the hypothesis based solely on this animal data. Nevertheless, the results from the Vollmayr et al. and Malberg and Duman studies indicate that, as of now, we have no reason to suspect that there is a causal link between neurogenesis and depression.
The neurogenesis hypothesis has two parts. While the part about the causal link may have been refuted for now, the part about how current depression treatments work remains relevant. Researchers have found that many of the major antidepressant pharmaceutical agents stimulated neuron cell proliferation in the dentate gyrus of adult rats (Kodama et al., 2004). Additionally, the drugs in this study showed no effect on the number of cells in the subventrical zone or primary motor cortex, suggesting that they were isolated to areas where neurogenesis would take place. Electroconvulsive therapy is another method for treating depression that is widely held to be useful. Madsen et al. (2000) gave rats a series of either 1 or 10 electroconvulsive series, and stained their cells with BrdU marker at different points after the treatment. He found that the rats which had been given electroconvulsive therapy had higher levels of new stem cells in the dentate gyrus, and that these results showed themselves in a graded manner, so those rats that had undergone more electroconvulsive therapy showed a greater increase in neurogenesis. The only anti-depressive treatment that does not appear to yield an increase of new cells in the dentate gyrus is transcranial magnetic stimulation (Czeh et al., 2001). Although it is tempting to dismiss this results as non-significant because TMS is a new and relatively unproven method of treating depression, the careful scientist must review all the information at his or her disposal. Nevertheless, there appears to some sort of relationship between neurogenesis and the action of many of the methods for treating depression.
One way to isolate the second part of the hypothesis is to attempt to discover whether the anti-depressive treatments depend upon neurogenesis in order to work. In a controversial study, Santarelli et al. (2003) attempted to do just this. First, they formed a rodent model of depression whose symptoms were alleviated by the anti-depressant. They then attempted to disrupt neurogenesis in the hippocampus by delivering low-dose x-radiation to the hippocampus while attempting to spare the brain and the most of the rest of the body. This x-radiation yielded an 85% decrease of BrdU-positive stained cells in the subgranular zone of the dentate gyrus. They found that when the x-rays regionally restricted the neurogenesis of the hippocampus, the anti-depressant effects on their rodent model by the anti-depressant drug (fluoxetine) were blocked. Although their study was well controlled for internally, it has still received some criticism. The main critique of the article has been that the research design used to model depression, whether the animal was willing to feed in a novel environment, does not actually test for depression but instead for anxiety (Sapolsky, 2004). While this criticism does not refute their result entirely, it does bring up some doubt. So while this study is one of the strongest pieces of evidence supporting the second part of the neurogenesis hypothesis, that neurogenesis is the reason that anti-depressive treatments work, it is not perfect, and further replication or a study of animals closer genetically to humans could yield more significant results.
This paper has reviewed our current understanding of the relationship between neurogenesis and depression. As stated earlier, the neurogenesis hypothesis has two components. The first, that there is a causal relationship between neurogenesis and depression, relies on a large amount of correlative and indirect evidence, such as the relationship between stress and neurogenesis, and the relationship between stress and depression. However, if there is to be a causal relationship, then depression must lead to a decrease in neurogenesis, which is not the case (Vollmayr et al., 2003), at least in rodents. However, the second component of the neurogenesis hypothesis, which is that anti-depressive treatments work via the stimulus of neurogenesis, remains a viable idea. Santarelli et al. (2003) in particular seemed to show that stimulating neurogenesis is the mechanism through which antidepressant treatments work, although the fact that TMS does not appear to work via neurogenesis is one knock against this theory. Further research, especially research that somehow could replicate these studies on humans, while remaining ethically sound, would be extremely advantageous to the study of the relationship between neurogenesis and depression. This topic is likely to remain “hot” given that neurogenesis is such a new idea. Indeed, the fact that we may be beginning to understand one of its potential functions is fascinating. Despite the obvious reasons to be excited about this research, we must strive to remain fundamentally sound in our scientific study of the neurogenesis hypothesis and its implications for the treatment of depression.
H. A. Cameron and R.D. McKay, Adult neurogenesis produces a large pool of new granule cells in the dentate gyru, Journal of Computational Neurology 435 (2001), pp. 406-417.
P. Celada, M. Puig, M. Amargos-Bosch, A. Adell and F. Artigas, The therapeutic role of 5-HT1A and 5-HT2A receptors in depression, Journal of Psychiatry and Neuroscience 29 (2004), pp. 252–265.
B. Czeh, T. Michaelis, T. Watanabe, J. Frahm, G. de Biurrun, M. van Kampen, et al., Stress-induced changes in cerebral metabolites, hippocampal volume, and cell proliferation are prevented by antidepressant treatment with tianeptine, Proceedings of the National Academy of Sciences of the USA 98 (2001), pp. 12796–12801.
K.P. Ebmeier, C. Donaghey and J.D. Steele, Recent developments and current controversies in depression, Lancet 367 (2006), pp. 153–167.
E. Gould and C. Gross, Neurogenesis in adult mammals: Some progress and problems, Journal of Neuroscience 22 (2002), pp. 619–623.
E. Gould, P. Tanapat, B.S. McEwen, G. Flugge and E. Fuchs, Proliferation of granule cell precursors in the dentate gyrus of adult monkeys is diminished by stress, Proceedings of the National Academy of Sciences of the United States of America 95 (1998), pp. 3168–3171.
R. E. Feldmann, A. Sawa and G. H. Seidler, Causality of stem cell based neurogenesis and depression—To be or not to be, is that the question?, Journal of Psychiatric Research 41 (2007), pp. 713-723.
R.C. Kessler, P. Berglund, O. Demler, R. Jin, D. Koretz, K. R. Merikangas, A. J. Rush, E. E. Walters, P. S. Wang, The epidemiology of major depressive disorder, Journal of the American Medical Association 289 (2003), pp. 3095-3105.
M. Kodama, T. Fujioka and R.S. Duman, Chronic olanzapine or fluoxetine administration increases cell proliferation in hippocampus and prefrontal cortex of adult rat, Biological Psychiatry 56 (2004), pp. 570–580.
Y.-B. Liu, P. A. Lio, J. F. Pasternak, and B. L. Trommer, Developmental changes in membrane properties and postsynaptic currents of granule cells in rat dentate gyrus, Journal of Neurophysiology 76 (1996), pp. 1074−1088.
J.E. Malberg and R.S. Duman, Cell proliferation in adult hippocampus is decreased by inescapable stress: reversal by fluoxetine treatment, Neuropsychopharmacology 28 (2003), pp. 1562–1571.
G.E. Miller, N. Rohleder, C. Stetler and C. Kirschbaum, Clinical depression and regulation of the inflammatory response during acute stress, Psychosomatic Medicine 67 (2005), pp. 679–687.
C. Pavlides, L.G. Nivon and B.S. McEwen, Effects of chronic stress on hippocampal long-term potentiation, Hippocampus 12 (2002), pp. 245–257.
K. Pham, B.S. McEwen, J.E. Ledoux and K. Nader, Fear learning transiently impairs hippocampal cell proliferation, Neuroscience 130 (2005), pp. 17–24.
H. van Praag, A. F. Schinder, B. R. Christle, N. Toni, T. D. Palmer, and F. H. Gage, Functional neurogenesis in the adult hippocampus, Nature 415 (2001), pp. 1030-1034.
L. Santarelli, M. Saxe, C. Gross, A. Surget, F. Battaglia and S. Dulawa et al., Requirement of hippocampal neurogenesis for the behavioral effects of antidepressants, Science 301 (2003), pp. 805–809.
R.M. Sapolsky, Is impaired neurogenesis relevant to the affective symptoms of depression?, Biological Psychiatry 56 (2004), pp. 137–139.
T. J. Shors, G. Miesegaes, A. Beylin, M. Zhao, T. Rydel and E. Gould, Neurogenesis in the adult is involved in the formation of trace memories, Nature 410 (2001), pp. 372–376.
J. A. Sweeney, J. A. Mkiec, and D. J. Kupfer, Neuropsychologic impairments in bipolar and unipolar mood disorders on the CANTAB neurocongitive battery, Biological Psychiatry 48 (2000), pp. 674-684.
P. Videbech, B. Ravnkilde, Hippocampal volume and depression: a meta-analysis of MRI studies, The American Journal of Psychiatry. 161 (2004), pp. 1957-1967.
B. Vollmayr, C. Simonis, S. Weber, P. Gass and F. Henn, Reduced cell proliferation in the dentate gyrus is not correlated with the development of learned helplessness, Biological Psychiatry 54 (2003), pp. 1035–1040.
O.M. Wolkowitz, V.I. Reus, T. Chan, F. Manfredi, W. Raum and R. Johnson et al., Antiglucocorticoid treatment of depression: double-blind ketoconazole, Biological Psychiatry 45 (1999), pp. 1070–1074.
C. Zhao, E.M. Teng, R.G. Summers Jr., G.L. Ming and F.H. Gage, Distinct morphological stages of dentate granule neuron maturation in the adult mouse hippocampus, Journal of Neuroscience 26 (2006), pp. 3–11.