Enriched Environment and White Matter in Aging Brain

Authors

  • Shu Yang,

    1. Department of Histology and Embryology, Capital Medical University, Beijing 100069, People's Republic of China
    2. Department of Histology and Embryology, Chongqing Medical University, Chongqing 400016, People's Republic of China
    Search for more papers by this author
  • Wei Lu,

    1. Department of Histology and Embryology, Chongqing Medical University, Chongqing 400016, People's Republic of China
    Search for more papers by this author
  • De-Shan Zhou,

    1. Department of Histology and Embryology, Capital Medical University, Beijing 100069, People's Republic of China
    Search for more papers by this author
  • Yong Tang

    Corresponding author
    1. Department of Histology and Embryology, Chongqing Medical University, Chongqing 400016, People's Republic of China
    • Department of Histology and Embryology, Chongqing Medical University, Chongqing 400016, People's Republic of China
    Search for more papers by this author
    • Fax: +86-23-68485589


Abstract

Normal aging is commonly associated with decreased cognitive functions, which could be conspicuously alleviated by enriched environment (EE) with physical, social, and sensory stimuli, suggesting that aging brain still has intriguing plasticity. Multiple researches have been carried out to explore the structural and the molecular changes in aging brain, which would be considered for evidences that EE regulated brain plasticity. Because there is no significant neuron loss in aging cerebral cortex and the white matter is crucial for cognitive functions, this review focused on the age-related white matter changes and the effects of EE on aged white matter. Data from our stereology laboratory revealed that age-related spatial memory declines had more to do with white matter alterations, which were due to marked demyelination and loss of oligodendrocytes in the white matter. We also demonstrated that EE recovered spatial memory impairment and increased white matter volume by promoting marked remyelination in aged brain. This review approached the issue that EE might contribute to normal aging and be beneficial for those suffering from demyelinated diseases. Anat Rec, 2012. © 2012 Wiley Periodicals, Inc.

An Overview of Enriched Environment and Brain Functions

Even in the absence of neuro-degenerated diseases, normal aging is commonly associated with decreased cognitive functions (Buckner, 2004). Fore example, long-term memory and working memory declined throughout life and more so in advanced aging (Nyberg et al., 1996; Park et al., 1996). Old rats showed a reduced performance in the Morris maze spatial task (Rapp and Gallagher, 1996). An association between normal aging and decline in episodic memory has also been reported (Grady and Craik, 2000). For decades, people tried hard to find ways to delay or prevent the age-associated cognitive declines. In 1906, the term “plasticity” was introduced into neuroscience by Ernesto Lugano, an Italian psychiatrist (Berlucchi, 2002). From that time on, multiple researches on the neural plasticity, which was the capacity of cerebral structural and functional adaptation to the external interventions, were carried out. Sixty years ago, the important studies in this field were initiated by Donald Hebb, who occasionally observed that rats that were allowed to run around freely in his house were better problem-learners and had better memory than rats reared in the laboratory (Hebb, 1949). Later in 1964, Bennett et al. (1964) demonstrated that the chemistry and morphology of the brain could be experientially altered. Since then, the brain was considered to be mutable to “environmental enrichment.” In 1978, the enriched environment (EE) was first defined as “a combination of complex inanimate and social stimulation” by Rosenzweig et al. (1978). In an enriched environmental condition, animals are group housed (6–12 per cage) and provided with a variety of stimuli such as running wheels, ropes, platforms, tunnels, boxes, balls, figurines, building blocks, and so forth, which are replaced and rearranged frequently (Bennett et al., 1964, 1969). Besides, animals living in EE have free access to social interactions and opportunities for sensory and physical manipulations as well as various learning opportunities, compared with those living in impoverished environmental condition (IE) or social environmental condition (SE), in which the animals are housed individually or in small group (two to four) in regular lab cages without any stimulus objects (Rosenzweig and Bennett, 1996). In a battery of subsequent experiments, the effects of enriched environment on behavioral tests were developed. Rats raised under EE from weaning learned more quickly and accurately than IE rats assessed by the total errors, the number of correct choices to the first error, and the number of correct in the first 17 choices in 17-arm radial maze (Juraska et al., 1984). Enriched adult rats (3 months) exhibited high performance levels, by exploiting procedural competencies and working memory abilities in eight-arm radial maze (Leggio et al., 2005). Environmental enrichment improved spatial memory performance in Morris water maze of young rats and mice (Paylor et al., 1992; Williams et al., 2001). Moreover, the beneficial effects of enriched postoperative housing condition on locomotor activity in home cage, reactivity to novelty, spatial working and reference memory in the Morris water maze and eight-arm radial maze, and learning in the Hebb–Williams maze were particularly obvious in adult rats with lesions of hippocampus, subiculum, or entorhinal cortex (Galani et al., 1997, 1998).

In aforementioned researches, only brain functions of young or adult animals were studied. It was then raised the question that whether the brain functions of aged animals might be affected as well. Following work indicated that the effects of enrichment stimulation were not limited to adult subjects. The quality of exploration expressed by corrected alternation scores or by the response to spatial change was improved by enrichment sometimes in 22-month-old rats (van Waas and Soffié, 1996). EE reduced the slowness of acquisition, reversed the short-term memory deficit and promoted the retention of the short signal assessed in 23-month-old rats in a Symbolic Delayed Matching to Sample Task (Soffié et al., 1999). In Hebb–Williams maze task, a type of intelligence test for examining problem-solving ability, the results of 12 problems showed that the enriched rearing condition improved the learning ability in all aged rats (25-month old) (Kobayashi et al., 2002). The spatial memory acquisition was improved and the development of a spatial bias in spatial probe trials was accelerated in enriched 28-month-old mice (Frick and Fernandez, 2003), which was consistent with Lores-Arnaiz et al. (2006). In addition, EE modified exploration activity, cognition, and motor functions of aged rats, valued in Bridges's test, passive avoidance test, open-field test and Marshall test (Fernandez et al., 2004). It then became important to discover what was responsible for these behavioral improvement observed in aged subjects.

White Matter Changes During Aging Process

To explore the targets of the EE, the first step is to figure out what's going on in the brain when aging. Some cellular and molecular changes in the aged brain have been documented since a few decades ago. Hippocampal neurogenesis showed an exponential decrease with age in both rodents and primates (Kuhn et al., 1996; Leuner et al., 2007). Aged rhesus monkeys had a lower frequency of multiple-synapse boutons (MSBs) and a lower number of synaptic contacts per MSB in the dentate gyrus outer molecular layer (Hara et al., 2011). Compared to young adults, aged monkeys exhibited a significant decrease in the NMDA receptor (Gazzaley et al., 1996) and brain dopamine (Wenk et al., 1989), both of which are the major neurotransmitters in cortex and hippocampus. Age-dependent reductions in brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF) expression were both detected in hippocampus of aged rats (Terry et al., 2011; Zeng et al., 2011). In 1955, Brody put forward that there was a significant and progressive loss of human cortical neurons during normal aging (as much as 49%) by examination on the subjects between 18 and 95 years of age, which received a great deal of publicity and was followed by several studies in which similar conclusions were made (Colon, 1972; Shefer, 1973; Devaney and Johnson, 1980; Henderson et al., 1980). Therefore, when referring to the age-related decline in brain functions, people considered it to neuron loss in the cortex in past few decades. However, there were still a very few disagreement. Haug et al. concluded that the previously reported numerous neuron losses with aging was likely due to the different shrinkages of the brain sections between young and old individuals during tissue processing, which was as much as 15% more in youth (Haug, 1984; Haug et al., 1984). They further concluded no age-associated change in the number of neurons of entire human cortex after making corrections for the difference in the amount of shrinkage (Haug, 1985). Terry et al. (1987) suggested that the earlier reports of the neuron loss in normal aging brains might be interfered by some materials from Alzheimer patients that were not excluded before examinations. It was not until the 1980s that a new unbiased stereological method for precise number estimation in three-dimension, the dissector, was developed by Denmark scientists (Sterio, 1984; Gundersen, 1986; Gundersen et al., 1988). Using this unbiased stereological method, Pakkenberg and Gundersen (1997) challenged the previous conclusions on the “density of neurons” and claimed that there was no significant change of the “total number of neurons” in the normal aged human brain cortex. The new viewpoint was soon confirmed by a set of following findings in world famous labs that the decline of cognition with aging could not be attributed to nonsignificant neuron loss in aged cerebral cortex (Morrison and Hof, 1997; Peters et al., 1998). What is the neural basis for the age-related deficits in cognition and memory if there is no significant neuron loss in aged cerebral cortex?

The connections between neurons in the cortex or subcortex ensure the conduction of nerve pulse and the integrity of brain functions. The white matter is the major concentration of the connections in the central nervous system, which is mainly composed of nerve fibers, glias, and intercellular substances. Therefore, more and more neuroscientists paid attention to the researches on the white matter. Merier-Ruge et al. (1992) demonstrated the age-related shrinkages of the cerebral gyri were resulted from the shrinkages of the subcortical white matter according to the measurements on the profile and the length density of capillaries and the distance between capillaries in the precentral gyrus, gyrus rectus, and corpus callosum of 33 human brains aged 5–93 years old. Common age-related loss of white matter volume and a much smaller decline of gray matter volume were directly observed on the in vivo MR images (Albert, 1993; Christiansen et al., 1994; Guttmann et al., 1998; Jernigan et al., 2001; Lemaitre et al., 2005). More advanced diffusion tensor imaging (DTI) technique detected white matter tract disruption during normal aging (O'sullivan et al., 2001; Shenkin et al., 2005). However, the white matter volume obtained with MRI techniques may be affected by a noise of neuroimaging parameters and the fibers in the white matter could not be precisely visualized on MRI and DTI. Thus it is presently not known whether DTI measures of white matter changes reflected demylination, axonal loss, or other form of white matter disruption. Using Cavalieri's principal on the autopsy of 94 normal human brains (62 males aged 19–87 years and 32 females aged 18–93 years), Pakkenberg and Gundersen (1997) found that the white matter volume in aged human brains was decreased by 28%. Tang et al. (1997) and Marner et al. (2003) afterward reported that the total length of the myelinated fibers in the white matter was significantly decreased by 45% from 20 to 80 years old by the use of stereological methods. The limitation of aforementioned studies was that they could not completely exclude the brains from people with early stages of Alzheimer's disease because approximately half of those over 80, living at home and apparently “normal,” actually show evidence of mild dementia when given a formal neuropsychological evaluation. The inclusion of the possible early Alzheimer brains could cause chaos in the assessment of brain changes with “normal aging.” Therefore, to further verify whether the white matter changes during normal aging, our lab recently used 6-month (young), 18-month (middle-aged), and 28-month (old-aged) male and female rats, which are considered not to develop Alzheimer's disease with aging (Erickson and Barnes, 2003), to do researches on the age-related white matter changes. After perfusion-fixation, the tissue blocks were sampled from the entire white matter in an isotropic, uniform, and random way and the microscopic pictures of nerve fibers were captured randomly. The white matter volume was calculated according to the Cavalieri's principle and the myelinated fiber volume, length and diameter were estimated with the unbiased stereological techniques (Table 1). Our results showed that the white matter volume of middle-aged male rats significantly decreased by 38.8% when compared with young males, while it did not decrease in middle-aged female rats comparing with young females, but significantly decreased by 42.7% in old-aged females (Yang et al., 2008, 2009). We indicated that the age-associated white matter shrinkage was resulted from a distinguished loss of the myelinated fibers in the white mater. The distributions of the myelinated fibers with classified diameters and myelin sheath thicknesses of young, middle-aged, and old-aged rats suggested that the age-related myelinated fiber loss was mainly due to a marked loss of the small myelinated fibers with thin sheathes (Yang et al., 2009). The myelinated fiber loss might be owing to: (1) universal axon degeneration with myelinolysis; (2) myelinolysis only; (3) both above. To figure out the reason for the myelinated fiber loss, it was necessary to investigate the age-related changes of the unmyelinated fibers in the white matter. The unmyelinated fibers with large and small diameters significantly degenerated together during aging process and the marked loss of small myelinated fibers to some extent replenish the loss of the small unmyelinated fibers in the white matter (Li et al., 2009). They concluded that it was the general demyelination of small myelinated fibers that induced the decline of the myelinated fiber length in aging brain white matter (Li et al., 2009). Given that the fibers in different white matter regions connect to different function domains in cortex, the alterations in aged white matters were probably heterogeneous. Sullivan et al. (2006) observed on DTI selective age-related degradation of anterior callosal fiber bundles, the principal white matter structure enabling interhemispheric information transfer. Our team confirmed that there was significant age-related atrophy of the corpus callosum (CC) (Xu et al., 2009a). We also further found that the frontal region in CC, which was the most associated with cognition and memory (Budson and Price, 2005), was more vulnerable to aging process than other regions of the CC (Xu et al., 2009a). In the frontal CC, the total length of the myelinated fibers with smaller diameter was decreased by 28.1% in old-aged rats (Xu et al., 2009a).

Table 1. Age-related changes in the white matter parameters
AnimalV (wm) (mm3)L (mf) (km)V (mf) (mm3)V (ms) (mm3)D (mf) (μm)
  1. V (wm): mean volume of white matter; L (mf): total length of myelinated fibers; V (mf): total volume of myelinated fibers; V (ms): total volume of myelin sheaths; D (mf): arithmetic mean diameter of myelinated fibers. Mean value of each variable (Mean), standard deviation (SD), observed coefficient of variation (OCV), and coefficient of error (CE) are given.

Young     
Mean119.6129.358.226.80.56
SD7.528.89.43.60.03
OCV (%)6.322.316.113.55.4
CE (%)1.316.511.810.52.1
CE2/OCV2 (%)4.354.753.760.514.1
Middle-aged     
Mean110.0119.375.434.90.68
SD7.151.36.85.20.10
OCV (%)6.543.09.115.115.7
CE (%)1.615.97.37.82.1
CE2/OCV2 (%)6.113.764.426.71.5
Old-aged     
Mean68.561.443.018.60.77
SD11.429.211.06.90.17
OCV (%)16.747.625.537.223.4
CE (%)1.420.410.412.32.4
CE2/OCV2 (%)0.711.311.06.90.7

According to our previous studies, we considered that the break down of myelin sheaths induced the marked myelinated fiber loss in the aged white matter. That the degenerations of the myelin sheaths surrounding axons decreased the conduction velocity was possibly responsible for the impaired cognition and memory of the elderly (O'sullivan et al., 2001; Bartzokis et al., 2003). By the use of stereological methods and electric microscope technique, we detected age-related morphological alterations of myelin sheaths, including the local splitting of the major dense line to accommodate dense cytoplasm derived from the oligodendrocytes, the opening-up of the intraperiod line to surround a fluid-filled space, the formation of large sheathes and double sheaths (Xu et al., 2009b), which were consistent with previous literatures (Peters et al., 2000, 2001; Sandell and Peters, 2001, 2003; Peters and Sethares, 2002). The oligodendrocytes in the central nerve system are responsible for the myelination, thus the quantity and the quality of the oligodendrocytes could influence the myelin sheaths formation during aging process. Sandell and Peters (2002) found an increase of oligodendrocyte number in the optic nerve of aged monkey. However, the number of oligodendrocytes counted in their research was actually the number of oligodendrocyte profiles in the two-dimensional sections, which could lead to bias. Also, the result from the optic nerve could not represent the age-related changes of the oligodendrocytes in the entire white matter. In the recent research in our lab, Chen et al. (2011) used the optical fractionator to count the total number of the oligodendrocytes immunohistochemically stained by CNPase in the whole white matter of young- and old-aged rats. Results showed that the total number of oligodendrocytes in the white matter was significantly decreased when aging (Chen et al., 2011).

The age-related white matter alterations or hyperintensities, as well as the loss of the myelinated fibers and the oligodendrocytes therein, may have important implications in the age-related cognitive and mnemonic impairments (De Groot et al., 2000; Krammer et al., 2007). The hyperintensities on MRI could explain some of the intellectual impairment in the elderly, especially slowing of distinct motor and attentional functions, as well as mental processing (Ylikoski et al., 1993). Further researches on brain DTI revealed the relationship between the white matter tract disruption and the age-related cognitive decline (O'sullivan et al., 2001; Shenkin et al., 2005). Madureira et al. (2006), based on the leukoaraiosis and disability in the elderly study (LADIS) population baseline data, pointed that age-related white matter changes influenced the performance on memory, executive functions, and speed/motor control. By the use of Morris water maze, we found that the spatial learning ability of rats was declined during aging since aged rats performed increased swim time (latency) to reach the platform, which was in line with the loss of white matter and myelinated fibers therein (Yang et al., 2009). Though the correlation was not much statistically significant, the consistency of the declines in spatial learning ability and white matter measures was instructive (Yang et al., 2009). Unlike dead neurons, the alterations in myelinated fibers and oligodendrocytes in the white matter could be alleviated or restored, thus the cognitive and mnemonic impairments associated with normal aging would be retarded. The plasticity on aged white matter is helpful for successful aging.

Effects of Enriched Environment on the White Matter

Because the neocortex and the hippocampus are considered to be close to cognition and memory, the researches on the enriched environmental effects have been mainly focused on the two of the above brain regions. Rats or monkeys living in EE exhibited increases in neuronal size, dendritic spines, and branching and synaptic protein levels in the hippocampus and prefrontal cortex (Diamond, 1967; Globus et al., 1973; Greenough and Volkmar, 1973; Kozorovitskiy et al., 2005), fast excitatory postsynaptic potential (fEPSP) in dentate gyrus and CA1 of hippocampus (Green and Greenough, 1986; Foster and Dumas, 2001), extracellular glutamate and GABA in the CA3 area of the hippocampus (Segovia et al., 2006), levels of NGF and BDNF in hippocampus (Neeper et al., 1995; Pham et al., 1999a,b), even impressive neurogenesis in the dentate gyrus (Kempermann et al., 1997; Nilsson et al., 1999). With the realization of an important role that the white matter plays in brain functions, people tried to find out whether EE have positive effects on the white matter as well. When Rhesus monkeys were raised in EE from 2 to 12 months, the size of corpus callosum viewed on MRI was significantly increased compared to control monkeys (Sanchez et al., 1998). They thought that this finding might reflect an increase in the number, myelination, and/or size or packing density of the myelinated axons in the corpus callosum (Sanchez et al., 1998). On DTI, some studies on adult human brains showed that trained subjects (Baduk and juggling) had larger white matter (Scholz et al., 2009; Lee et al., 2010); and the changes in fractional anisotropy (FA) within developmental and adult white matter between trained subjects and nontrained subjects were statistically significant (Bengtsson et al., 2005; Scholz et al., 2009; Takeuchi et al., 2010), which might indicated a train-induced increase in the myelination of white matter (Bengtsson et al., 2005; Scholz et al., 2009; Takeuchi et al., 2010). Moreover, it was recently documented that experience dependent myelination in white matter extended into very old age (65–80years) (Lövdén et al., 2010). These aforementioned results revealed that white matter could be modified by external factors and spotlighted as an intervention target. However, “You can't say based on the functional imaging and fractional anisotropy what is going on at a cellular level.” said Adeline Vanderver, a neurologist and white matter expert at the National Children's Medical Center in Washington, DC (Vance, 2009). Therefore, it is necessary to study the effects of EE on the white matter at a cellular level, for example, the myelinated fibers and the myelin-forming oligodendrocytes.

Previous research showed that premature eyelid opening accelerated the onset of myelination (Tauber et al., 1980). Although the result came from optic nerves, it was helpful to understand the myelination process in central nervous system. The number of myelinated axon profiles in the splenial corpus callosum of rats reared under EE from weaning to 55 days of age was increased according to an ultrastructural analysis (Juraska and Kopcik, 1988). Szeligo and Leblond (1977) examined the influence of rearing environment on fiber tracts in CC during suckling and found an increase of oligodendrocytes in the visual cortex of EC rats. The greater volume fraction of oligodendrocyte nuclei in the visual cortex of rats reared for 30 days postweaning in EE has also been demonstrated (Sirevaag and Greenough, 1987). Beyond developmental period, preliminary result indicated that the number of myelinated axons in splenial corpus callosum of adult rats exposed to EE was increased compared with their control littermates (Briones et al., 1999). Although very few, these studies suggested that oligodendrocytes and myelination process were sensitive to experience and not limited in developmental timeframe, but extended to adulthood. However, there were still some questions remained to be explored: (1) whether myelinated fibers and oligodendrocytes are sensitive to enriched environment even in aging process; (2) whether the general myelinated fibers and oligodendrocytes in the whole white matter, instead of some local cerebral regions, are sensitive to enriched environment; (3) previous morphological studies had some technical limits on estimating myelinated fibers and oligodendrocytes, in which it was the cell or axon profiles that were observed.

To resolve these questions, 14-month (middle-aged) and 24-month (old-aged) SD rats were randomly assigned to either an EC or SC for 4 months prior to Morris water maze tests in our laboratory. Middle-aged female rats exposed to enriched environment performed much better in spatial tests than those under SE did. By the use of unbiased stereological methods we developed before, we found that the white matter volume, the myelinated fiber volume, the myelinated fiber length and the myelin sheath volume were increased induced by the way of EE both in middle-aged and old-aged rats (Fig. 1); but the mean diameters of the myelinated fibers were not significantly different between EE rats and SE rats. According to the distribution changes of myelinated fiber length with different diameters, we suggested the increases of white matter and myelinated fibers were likely due to the marked increase of smaller myelinated fibers (0.3–0.7 μm in diameter) but not the thickening of the myelin sheaths during aging. Our previous studies indicated that the total length of the myelinated fibers with small diameters (0.3–0.7 μm) in white matter was markedly lost during aging process, which was mainly due to the demyelination of the myelinated fibers while the axons were preserved (Marner et al., 2003; Li et al., 2009; Yang et al., 2009). Peters' research group also reported that there were no significant age-related changes in the axons even though the myelin sheaths were severely degenerated with aging (Peter et al., 2000; Peters and Sethares, 2002). The preserved axons made it possible for remyelination. Therefore, the enrichment-induced remyelination was considered to be one of the potential reasons for the marked increase of the myelinated fibers. Oligodendrocytes, the myelin forming cells, were involved in the remyelination process. Enriched housing conditions promoted the oligodendrocyte fate of subventricular zone (SVZ)-recruited cells in the experimental autoimmune encephalomyelitis (EAE) demyelination lesions of adult mouse (Magalon et al., 2007). In focal cortical ischemia caused by distal ligation of the middle cerebral artery, EE enhanced the generation of oligodendrocyte progenitor cells (NG2 positive polydendrocytes) (Komitova et al., 2006). A new arrival article revealed that the EE increased bromodeoxyuridine (BrdU)-positive cell numbers in the amygdala, almost all of which expressed the oligodendrocyte progenitor marker Olig2 (Okuda et al., 2009). We recently found that the total number of CNPase positive oligodendrocytes in corpus callosum was significantly increased in the enriched middle-aged (18 months) female rats (Zhao et al., 2011). CNPase was involved in the migration or expansion of oligodendrocyte membranes during myelination (Gravel et al., 1996, Yin et al., 1997) and was considered to play distinct roles in subcellular compartments of myelin that serve axon–glial communication (Rasband et al., 2005). Therefore, the increased oligodendrocytes could form myelin sheaths to coat the axons, thus the nerve conduction function could be restored in aging brains, which might be beneficial for the spatial memory enhancement observed by our group. Although statistical analysis showed no significant correlation between spatial memory enhancement and myelinated fiber parameter changes in white matter, we could not conclude that the myelinated fibers in the white matter are far away from the spatial abilities. It is likely that a combination of increased neurons, synapses and extracellular substances in cortex and hippocampus of aged rats (Saito et al., 1994; Lores-Arnaiz et al., 2006; Segovia et al., 2006), myelinated fibers in white matter observed in our research, as well as other factors still unknown, contribute to the enhanced spatial performance induced by exposure to an environmental enrichment.

Figure 1.

A: The effects of enriched environment on the white matter volume. B: The effects of enriched environment on the total volume of the myelinated fibers in the white matter. C: The effects of enriched environment on the total length of the myelinated fibers in the white matter. D: The effects of enriched environment on the total volume of the myelin sheaths and the total volume of the axons in the white matter. * P < 0.05, ** P < 0.01, *** P < 0.001. SE indicates standard environment and EE enriched environment.

CONCLUSION

Cognition declines during normal aging were associated with white matter alterations. The general demyelination of small myelinated fibers with thin myelin sheaths in the white matter was the primary cause of the myelinated fiber loss in aged white matter. Although the remyelination in central nervous system has been confirmed (Sim et al., 2002; Duce et al., 2006), the remyelination efficiency became lower in aged brain (Shields et al., 1999; Sim et al., 2002; Peters and Sethares, 2003). Increasing the natural processes to obtain therapeutic effects was therefore a major challenge. Compared with drugs, EE strategy is safer, less expensive, simpler to implement and produces fewer side effects. EE decreased the impairment of the brain function and protected the aged cerebral white matter structure. Thus, EE would be a viable alternative to benefit the normal aged human and those suffering from demyelinated diseases.

Acknowledgements

Chen Li, Wei Zhang, Lin Chen, Qiang Xu, Yuan-yu Zhao, and Xiao-yan Shi are highly appreciated.

Ancillary