Exercise and voluntary wheel running have not only been shown to increase neurogenesis  but also to improve memory and learning tasks  and exert beneficial effects on diseases of the CNS, such as stroke, by facilitating motor recovery in rats . Frequent long-term sequelae after bacterial meningitis are cognitive deficits as a consequence of hippocampal damage in the course of this infectious disease. Conversely, hippocampal neurogenesis has been shown to increase after meningitis in both rodents and humans [13–15], but the effects of CNS inflammation on neurogenesis remain controversial [24, 25].
The increase in neurogenesis after brain injury possibly reflects an endogenous potential of cell replacement. Mice exposed to an enriched environment after recovery from bacterial meningitis, as a rehabilitative approach, did not show additive elevation of the meningitis-induced increase of hippocampal neurogenesis, suggesting a limited capacity of the brain to further increase the already elevated neurogenesis after bacterial meningitis. Accordingly, learning performance was not better in mice exposed to an enriched environment after meningitis compared to animals with meningitis who received no further stimuli . The present study aimed to investigate whether exercise prior to infection could alleviate the course of disease and increase neurogenesis, as well as improve spatial learning performance, thereby attenuating long-term sequelae after meningitis.
In this study, mortality due to Streptococcus pneumoniae meningitis was significantly decreased in mice that had been exercised by means of running wheels prior to infection, compared to animals that were kept under standard conditions. It can be hypothesized that improved cardiorespiratory capacity achieved by regular exercise facilitated survival and that an altered immune response towards a less pro-inflammatory direction was also responsible for decreased mortality. Indeed, the decrease of mortality was accompanied by a tendency towards lower bacterial burden as well as towards lower pro-inflammatory cytokine levels of IL-6 and TNF alpha and lower corticosterone levels in runners compared to sedentary mice. Overall, the extent of the inflammatory and stress response to bacterial meningitis was decreased by physical exercise. This observation is in accordance with other studies showing a decreased release of TNF alpha in response to lipopolysaccharide challenge in physically active animals , and adjuvant therapy with TNF alpha converting enzyme inhibitor-prolonged survival in rats with Streptococcus pneumoniae meningitis . Similar to this mouse model of meningitis, cortisol levels were significantly increased in the cerebrospinal fluid of patients with pneumococcal meningitis , and running appeared to have beneficial effects by reducing pre-infection corticosterone levels.
Since proliferative activity showed no difference between running and sedentary mice 4 days after infection, we propose that running prior to infection was unable to further increase the cell proliferation in the dentate gyrus stimulated also by the infection, but favored the differentiation of new-born cells into the neuronal lineage. Comparison of BrdU-labeled cells 7 weeks after meningitis revealed a lower amount of proliferating cells in running mice. One cannot differentiate whether this difference is due to fewer proliferating cells in the acute phase of meningitis (BrdU was applied on day 3 to 5 after induction of meningitis) or whether more initially labeled cells died or went into apoptosis by the end of the experiments 7 weeks after meningitis.
The number of TUC-4 and Dcx-expressing young neurons in the dentate gyrus was significantly increased in running mice 4 days after infection, suggesting a stimulating effect of physical exercise on neurogenesis in addition to the stimulating effects of meningitis alone. Seven weeks after meningitis, the number of Dcx-expressing cells no longer differed between the groups, but the percentage of initially proliferating BrdU-labeled cells that differentiated into adult neurons was significantly higher in running mice in comparison to sedentary controls. Running probably exerted its stimulating effects on neurogenesis by modulating the inflammatory response and reducing corticosterone levels, as the latter has already been shown in healthy mice . However, this modulation may not be the only reason for the beneficial effects of running. Other parameters such as the modulation of growth factors may also contribute to the increase of neurogenesis after meningitis and exercise. Promotion of growth factor pathways by running resulted in both stimulation of hippocampal neurogenesis and in functional improvement of memory tasks . In a rat model of traumatic brain injury, physical exercise 6 weeks before head trauma resulted in a better adaptation to oxidative challenge and prevented against ROS (reactive oxygen species)-mediated inhibition of the Na+, K + −ATPase .
The increase of hippocampal neurogenesis in this study was not accompanied by improved spatial learning: no difference in Morris water maze performance was observed between runners and non-runners 6 weeks after meningitis. If there were possible positive effects of running on cognitive performance, these effects were most probably erased by the damage caused by meningitis, and running prior to meningitis did not overcome these post-inflammatory sequelae. Whether learning and hippocampal neurogenesis positively correlate is still a matter of debate. There are studies postulating a correlation between neurogenesis and cognitive performance [33–37] while other studies imply partially contradictory results [13, 26, 38–41]. Furthermore, there are studies in which adult neurogenesis was virtually completely suppressed by radiation  or genetic manipulation  and in which learning and memory performance of the rodents remained unimpaired [44–47]. These divergent results may be in part a consequence of technical differences and varying time points of evaluation . The detection of alterations in spatial learning in the Morris water maze is a challenge because even healthy rodents without any manipulation perform differently in this task depending on species differences, genetic background, high intraindividual variation and motivation; all of which lowers the chance of detecting significant differences .
In conclusion, voluntary wheel running prior to the induction of the infection prolonged the survival time and decreased the mortality of bacterial meningitis, attenuated the inflammatory response and, furthermore, increased the differentiation of proliferating cells into adult neurons in the dentate gyrus suggesting a beneficial and protective effect of running. Physical exercise increased the ability of the host organism to survive Streptococcus pneumoniae meningitis with the aid of antibiotic therapy, but the neuroregenerative effect of this mechanism with respect to neuropsychological outcome appeared small under the conditions of the present study.