Overcoming the Aging Systemic Milieu to Restore Neural Stem Cell Function

Stem Cell Aging Has Been Linked to Altered Stem Cell Niches and the Systemic Milieu

A ging organisms gradually lose the ability to maintain homeostasis. Differentiated cells, especially those that are postmitotic, have a limited ability to maintain their state over time and need to be replaced. Multicellular organisms have evolved rejuvenative mechanisms based on populations of progenitor and adult stem cells that act as a reservoir for the production of newly differentiated cells to help maintain organismal integrity. Although these stem cell populations are insufficient to replace all aging differentiated cells completely in most animals, they likely help to extend longevity relative to organisms that have fewer or lack such reservoirs. However, recent work indicates that adult stem cell function declines with organismal age and that this loss is often associated with alteration of the stem cell niche and systemic milieu that helps maintain stem cell function. ¹ Perhaps this is not terribly surprising given that the stem cell niches and the systemic milieu are themselves composed of differentiated cells subject to aging. But why don't stem cells simply replace all aging differentiated cells to avoid deterioration of their milieu? The answer is that evolution only optimizes successful reproduction and not necessarily organismal integrity. Evolution has not selected for omnipotent stem cell-based repair systems for most animals, with the possible exception of certain asexual organisms such as hydra and some planarian species, which require such repair systems to reproduce. In the absence of such efficient repair systems, a gradual decline in the function of un-replaced or incompletely replaced aging cells eventually compromises the function of near and distant stem cells, as well as other differentiated cells, leading to loss of homeostasis.

Mechanisms of Stem Cell Aging

Loss of stem cell function in aging has been hypothesized to be due to several mechanisms: Depletion due to differentiation, loss of capacity for self-renewal or senescence, loss of lineage specificity (ability to make correct, fully functional progeny), or malignant transformation. These dysfunctional states can result from altered intrinsic metabolism or altered extrinsic influences (e.g., changes in trophic factors, inflammatory factors, etc). Among the best-studied examples of decreased stem cell function are the shift to an adipogenic program in adult myoblasts ² and in tendon-derived stem/progenitor cells ³ with age. A good example of stem cell depletion is loss of melanocyte stem cells by genotoxic stress resulting in gray hairs. ⁴ The best-understood mechanism of stem cell dysfunction in which the environment plays a clear role is the decline of skeletal muscle stem cell (satellite cell) function with age. Satellite cells' capability to repair damaged skeletal muscle is controlled by a balance between notch and wnt signaling ⁵ and is gradually lost with age. However, the loss is reversible. Connecting the circulatory system of a young mouse with an old mouse restores activity of old satellite cells. ⁶ Although the precise factors from plasma that cause this effect are unknown, it has been established that increased transforming growth factor-β (TGF-β) in the local environment of the satellite cells can block satellite cell function, ⁷ and it has been hypothesized that a systemic TGF-β antagonist neutralizes TGF-β in young but not old animals. ⁸ These results support the paradigm of age-associated stem cell dysfunction by defined changes in the systemic milieu.

Cytokine CCL11/Eotaxin-1 Increases with Age and Inhibits Neurogenesis

In an important study that extends our understanding of stem cell function during aging, Villeda et al. show that cytokine CCL11/eotaxin-1 serum levels increase with advancing age, impairing neurogenesis and cognitive function. ⁹ Adult neurogenesis, which occurs in blood vessel-rich neurogenic niches in the subventricular zone (SVZ) and the subgranular zone (SGZ) of the hippocampus, may play a role in learning and cognition in mammals, although there is some controversy in the literature as to exactly which cognitive processes are affected. ^{10 –12} First, Villeda et al. ⁹ showed that in their mouse cohort, as expected, old mice had increased neuroinflammation, decreased neurogenesis, and synaptic plasticity and deficits in cognitive behaviors (contextual fear conditioning and spatial learning in a radial arm water maze [RAWM]).

To test the hypothesis that age-related changes in the systemic milieu might cause decreased neurogenesis and cognitive loss, mice were subjected to heterochronic (young–old) parabiosis whereby the circulatory systems of old and young mice are surgically linked. In the young parabionts, there was a decrease of SOX-2 (also called SRY [sex determining region Y]-box 2)-expressing neural precursors, doublecortin (Dcx)-expressing newly born neurons, and overall cell proliferation as assessed by bromodeoxyuridine (BrdU) incorporation compared to young–young parabiotic pairs. Reciprocally, there was an increase in neurogenesis as measured by these parameters in the old parabionts compared to old–old parabiotic–paired controls. Taken together, these results support the hypothesis that the systemic milieu modulates neurogenesis during aging. The authors ruled out direct migration of cells into the brains of the parabionts by using green fluorescent protein (GFP)-expressing transgenic mice as one of the pair of connected animals. Consistent with the idea that blood from old mice makes soluble inhibitory factors, injection of plasma from old mice inhibited neurogenesis in young mice. Furthermore, young mice receiving old plasma showed alterations in two narrow tests of cognition: Decreased freezing in contextual, but not cued, memory tests, and impaired learning and memory for platform location in the RAWM test.

To determine the identity of the soluble factors, Villeda et al. used enzyme-linked immunosorbent assay (ELISA) analysis to test the levels of 66 cytokines, chemokines, and other factors in heterochronic parabionts and in unpaired old mice. They determined that CCL2, CCL11, CCL12, CCL19, haptoglobin, and β-2-microglobulin were elevated in both groups of animals. The authors then focused on CCL11, which was also observed to show increased levels with age in analysis of random human plasma samples. CCL11 was an auspicious choice because it was previously shown to inhibit neurogenesis in human neural precursors in culture, ¹³ although Villeda et al. ⁹ were apparently unaware of this work. Consistent with this prior work, CCL11 inhibited neurogenesis of mouse primary neural stem/progenitor cells (NPCs) as measured by neurosphere formation and neural differentiation of the human nTERA cell line in culture. Injections of CCL11 into young mice inhibited neurogenesis, but could be rescued by co-administration of CCL11 neutralizing antibody. CCL11-injected young mice showed similar cognitive defects to young mice treated with plasma.

The evidence taken together suggests that systemic CCL11 is upregulated during aging and plays a role in inhibiting neurogenesis, resulting in some decline in cognitive function. The authors claim that this is the first report of a specific systemic factor associated with aging that affects adult stem cell function. However, it is important to note that the authors did not report that injection of anti-CCL11 neutralizing antibodies could restore neurogenesis in old animals. It is unclear whether this experiment had been attempted, but it would significantly strengthen the case for CCL11 being the critical serum factor that mediates systemic inhibition of neurogenesis during aging. Because the authors suggest that blood be more extensively characterized for other factors that may contribute to both inhibition and promotion of neurogenesis, it is likely that multiple factors play a role in inhibition.

Effects of Integration Between the Immune System and Central Nervous System

Neurogenesis has been previously shown to be sensitive to inhibition by inflammation. Increased inflammation is a hallmark of aging in mammals, thus decreased neurogenesis has been hypothesized to be caused by aging-associated inflammation. The work of Villeda et al. is not inconsistent with this hypothesis, but the involvement of CCL11 suggests that the connection between altered immune system function and neurogenesis may be more subtle. Several recent reports support the idea that the immune system and the brain may be more tightly integrated than previously appreciated. For example, meningeal T cells support learning, perhaps through secretion of interleukin-4 (IL-4), which maintains a state of “muted inflammation.” ^14,15 CCL11 can inhibit IL-4 activity, ¹⁶ potentially increasing inflammation in the critical brain regions, ¹⁵ although as mentioned above CCL11 can directly affect NPCs, so this effect may be at most a secondary one. Another instructive example is that the ratio of CD4 to CD8 T cells in heterogeneous mouse strains directly correlates with the levels of neurogenesis seen in any particular strain, suggesting a deep connection between T cell function and neurogenesis. ¹² A potentially intimate connection between the immune system and the central nervous system would add to the number of potential ways by which aging-associated decline of function in one cell population could critically impact another cell population in mammals.

Medical Implications

Although neurogenesis plays a role in memory and cognitive function in mice, it is only part of a more complex set of mechanisms to ensure plasticity. ¹¹ There are other aspects of brain function that decline during aging ¹⁷ that may be addressed by potential intervention. ^18,19 To the extent that neurogenesis is a potential target to improve age-related decline in cognition, there are several reported means to achieve increased neurogenesis: Supplementation with curcumin ²⁰ and apigenin, ²¹ exercise, ^22,23 and dietary restriction. ^{24 –26} The evidence that these may be helpful in humans comes from animal models and must be considered preliminary.

In situations in which inflammation blocks neurogenesis, such as inflammation associated with cranial radiation therapy or lipopolysaccharide injection, indomethacin has been reported to restore neurogenesis. ²⁷ Interestingly, indomethacin has been reported to act similarly to CCL11 on eosinophils at low doses, but decreases expression of CCR3, the CCL11 receptor, at high concentrations. ²⁸

Targeting CCL11 function must be considered completely speculative at this time, because it may be necessary to inhibit the function of several plasma-borne factors to effectively induce neurogenesis in older people. Substances that have been reported to inhibit CCL11 function include nobiletin ²⁹ and 7,4′-dihydroxy flavone from Glycyrrhiza uralensis. ^30,31 However, there is no available evidence that these molecules will induce neurogenesis, although it is possible that they would be able to achieve some beneficial effect, especially in conjunction with treatments that appear to stimulate neurogenesis.

Conclusion

Subtle specific changes in the stem cell milieu, among them altered serum levels of specific factors such as CCL11, may play a more significant functional role in normal human aging than more dysfunctional cell states, such as cellular and replicative senescence, that have been hypothesized to play critical roles in aging through promotion of increased inflammation. ³² The existence of potentially important specific age-associated molecular changes in stem cells opens the door to the discovery and development of interventions to slow aging.