There is growing evidence that CSF plays an important role in physiological as well as pathophysiological processes of the brain including adult neurogenesis [28–30]. In a very recent study of our group, we found adult human leptomeningeal CSF being a promotor of survival, differentiation and astrogliogenesis of fetal NSCs from rat . The influence of CSF on adult NSCs remains however still enigmatic. In this study we therefore used human adult NSCs as an in vitro model to study the effects of adult leptomeningeal CSF on NSC behaviour including survival, self renewal and differentiation. The central finding of our study is that in vitro, adult CSF promotes survival and differentiation of ahNSCs, but drives the differentiation process towards astrogliogenesis. In accordance with these findings, the loss of stem cell potential is accelerated when cultured in adult CSF. These findings suggest that adult CSF contains key factors involved in the control of cellular proliferation and differentiation processes. This assumption is supported by in vivo findings, demonstrating that adult NSCs of the subventricular zone have transitory contact with the ventricular brain cavity and many of them still posses one microcilia which extends into the CSF [32–35]. In addition, it is known from previous studies [21–23] that embryonic CSF has a trophic influence on survival and differentiation of NSCs. However, the loss of stem cell potential is decelerated, which is contradictory to our findings in adult CSF and may explain the abundance of post-mitotic cells in adult CNS.
Based on the knowledge of embryonic CSF studies, it has already been postulated that diffusible factors in embryonic CSF regulate the three basic cellular behavioural parameters of neuroephitelial stem cells and that embryonic CSF may play a key role in brain development in vivo .
However, how CSF influences neuroectodermal cells during development remains enigmatic, but the components contained in CSF as well as CSF pressure and flow seem to play an important role [26, 36]. Regarding the components of CSF influencing neuroectodermal cell behaviour, recent investigations concentrated mainly on proteins, "membranous particles" and amino acids but also on growth factors such as FGF2 [21, 24, 25, 36, 37]. In avian and human CSF, many proteins with a known influence on cell survival, neural and glial differentiation, proliferation and signal transduction were found (such as transthyretin, serin, retinol binding protein, heparan sulfate, several apolipoproteins, and FGF2) [22, 24, 25]. Although it has also been demonstrated that the protein composition of embryonic CSF is more complex than that of adult CSF , our results indicate that adult CSF has the capacity to influence the behaviour of adult NSCs in the adult brain, too.
CSF as a beneficial environment for cell survival and growth has also been postulated by in vivo studies, investigating for example the behaviour of fetal NSCs after injection in the fourth ventricle of spinal cord lesioned rats with a good survival of grafted cells within the CSF [38, 39], as well as in a recent in vitro study of our own group with fetal rat NSCs . In the present study, CSF had a general stimulating effect demonstrated by the faster loss of self-renewing capacity and stronger cell extension outgrowth. Adult hNSCs differentiated predominantly into astrocytes (38% in CSF, 25% in standard media) when treated with CSF and to a lower extend into oligodendrocytes and neurons. After differentiation for 7 days, we found 9% of the cells to be β-tubulin+ in CSF. In standard culture media, 31% of the cells were β-tubulin III+. Therefore, our data strongly suggest inhibitory effects of CSF on neurogenesis of ahNSCs, but promoting effects on astroliogenesis. Possible factors in CSF influencing differentiation behaviour of NSCs are bone morphogenetic proteins (BMPs): Monoclonal antibodies against BMP7 were for example shown to inhibit CSF induced dendritic outgrowth of neurons. BMP4 was shown to induce neuronal differentiation of NSCs by activating the ERK and inhibiting the GSK3β pathway. Both described effects could be blocked by Noggin, a BMP inhibitor [40, 41]. These described effects of BMPs, do have an influence on neuronal differentiation only in fetal NSCs from rat. Until recently, it remained elusive whether these BMPs may also play a role in the observed effects of adult human CSF on astroglial differentiation, but we could show that parts of the CSF mediated effects on fetal rat NSCs could be blocked by Noggin. BMPs thus seem to be at least a part of the soluble factors in the CSF influencing stem cell behaviour in fetal rat NSCs. However, in the present work, we could not find any inhibiting effects of Noggin on CSF survival effects on adult human NSCs. This raises the question whether BMPs influence NSC behaviour only during ontogenesis and therefore have no influence on NSCs derived from the adult brain or whether BMPs might differential effects in human and rat cell systems.
It is well accepted that lumbar CSF is different from ventricular CSF because of both passive diffusion processes across the blood-CSF-barrier and suggested active secretion processes during the cranio-caudal circulation . Consistently, all herein described CSF effects can only be attributed to leptomeningeal CSF. Whether ventricular CSF has similar same effects on NSC behaviour remains elusive. The use of CSF out of ventricular drainages is however problematic, as ventricular drainages are used in patients with obstructed CSF circulation (for example cerebral aqueduct stenosis) or other reasons of elevated brain pressure with defect blood-brain- and blood-CSF-barrier (for example after major stroke, after intracerebral bleedings or inoperable brain tumours). Furthermore, CSF out of ventricular drainages is often contaminated by blood and altered by inflammatory processes, which is why it cannot be used for the examination of healthy CSF effects.