Keywords

9.1 Introduction

Aging is a complex, multifactorial process influenced by many unknown genetic and environmental factors. It is associated with progressive decline in normal cell and organ functioning. An important hallmark of aging is ‘inflamm-aging’, a state of chronic, low-grade inflammation, caused by an elevated concentration of inflammatory markers in the circulation (Calder et al. 2017). The balance between pro- and anti-inflammatory cytokines in the healthy adult brain is shifted with aging towards a pro-inflammatory state (Franceschi 2007). This immunological fragile state makes the aged brain more susceptible to diseases, infection, and stress, which might even influence the onset of age-related neurodegenerative brain diseases (Franceschi and Campisi 2014; Gorle et al. 2016).

With life expectancy exponentially increasing, age-related diseases will become an emerging epidemic and a tremendous public health issue due to the high costs of dementia care. Numbers are predicted to increase to 152 million in 2050 and there are over 9.9 million new cases of dementia each year worldwide (Report 2018), in addition no treatment to reverse or halt disease progression exists. Increasing evidence indicates that degeneration of the choroid plexus can result in brain deficits and contribute to cognitive impairment. Therefore, extensive insights in the aging choroid plexus are essential in understanding age-associated neurodegenerative and neuroinflammatory disorders and pave new ways for therapy.

The role of the choroid plexus in health and disease is being increasingly recognized and it has been reported to play a central role during aging (Gorle et al. 2016; Vandenbroucke 2016; Marques et al. 2017). The choroid plexus is a highly vascularized brain structure, consisting of a monolayer of choroid plexus epithelial cells firmly interconnected by tight junctions (De Bock et al. 2014), that form one of the brain barriers, called the blood-cerebrospinal fluid (CSF) barrier. Together with other brain barriers, the blood-CSF barrier assures a balanced and well-controlled micro-environment in the central nervous system (CNS), providing protection against external insults such as toxins, infectious agents, and peripheral blood fluctuations (Gorle et al. 2016). The choroid plexus produces CSF and receives input from both circulatory, autonomic and immune system. It can respond as a key regulator to local changes of different physiological signals by changing its secretome, including proteins (Marques and Sousa 2015; Silva-Vargas et al. 2016) and extracellular vesicles (EVs) (Balusu et al. 2016b). The normal functioning of the choroid plexus is severely affected during aging and this dysfunction is even aggravated in age-related neurodegenerative brain diseases like Alzheimer’s disease (Balusu et al. 2016a). Understanding how the function and activity of the choroid plexus is altered in aging might lead to the identification of strategies to attenuate aging-associated cognitive decline and related diseases (Baruch et al. 2014; Vandenbroucke 2016; Gorle et al. 2016).

9.2 Morphological Changes of the Choroid Plexus Epithelium Upon Aging

Several reports have been published describing the morphological alterations of the choroid plexus epithelium upon aging (Fig. 9.1), which is comparable to other secretory epithelia. Across species epithelial atrophy and weight increase have been observed (Wen et al. 1999), however slightly different modifications have been described according to species.

Fig. 9.1
figure 1

Schematic representation of the changes at the choroid plexus during aging. Several morphological changes are observed at the choroid plexus: the choroid plexus epithelial cells are flattened with an irregular nucleus and shortened microvilli, more Biondi rings and lipofuscin are present, and the basement membrane is thickened and contains fragmented vessels, collagen fibers, hyaline bodies, calcifications, and psammomas. Functionally multiple alterations were shown: increased cerebrospinal fluid (CSF)/serum albumin ratio, reduced metabolic activity, decreased extracellular vesicles (EVs) in the CSF, increased levels of lactate and vasopressin, and altered immune cell recruitment (linked with changes in the interferon (IFN) balance). Key: CSF: cerebrospinal fluid, EV: extracellular vesicles, IFN: interferon, IL: interleukin, LDH: lactate dehydrogenase, SDH: succinate dehydrogenase

Full size image

In humans, the height of the epithelial cells decreases with approximately 11% during life and the cells become more flattened (Serot et al. 2000). The aged cell cytoplasm contains protein inclusions called Biondi ring tangles. In addition, the presence of lipofuscin deposits can be found. Since this age pigment is a product of lipid peroxidation by free oxygen radicals it will probably alter the cell functioning (ZS-Nagy et al. 1995). The nuclei become more irregular in elderly and have a flattened shape (Serot et al. 2000, 2001). Moreover, the epithelial basement membrane has been reported to become thicker with aging. Also the stroma of the aged choroid plexus is thicker and contains collagen fibers, hyaline bodies, calcifications, and psammomas (i.e. dystrophic calcifications) (Eriksson and Westermark 1986; Jovanovic et al. 2004; Sturrock 1988; Wen et al. 1999). An age-associated increase in size and volume density of the psammoma bodies has been described (Zivkovic et al. 2017). The arterial walls become thicker, especially the media and adventitia, while the blood vessel volume density decreases and elastic fibers are fragmented (Serot et al. 2000; Shuangshoti and Netsky 1970; Zivkovic et al. 2017), resulting in a reduced contact area between the blood and the choroid plexus epithelium.

Rodent models show similar epithelial disruptions of the choroid plexus epithelium compared to humans (Serot et al. 2001; Sturrock 1988). In elderly rats the epithelial cells lose height, approximately 15%, and become more flattened. The cells show an irregular, elongated nucleus and shortened microvilli, causing a decrease in the choroid plexus epithelium-CSF contact area. Lipid vacuoles are present in the cytoplasm of the choroidal epithelial cells. Irregular fibrosis has been described in the stroma of elderly rats together with thickening of the basement membranes (Serot et al. 2001; Sturrock 1988).

Age-associated reduction in contact area between blood-choroid plexus epithelium and choroid plexus epithelium-CSF due to morphological alterations, together with the changes in choroidal proteins involved in CSF production (Masseguin et al. 2005), negatively influence the CSF production (Vandenbroucke 2016). These morphological changes will result in functional alterations, which may consequently have an impact on brain homeostasis.

9.3 Functional Alterations of the Choroid Plexus in Aging

9.3.1 CSF Dynamics

9.3.1.1 CSF Production and Secretion

One of the major functions of the choroid plexus is CSF production and secretion. CSF flows from the choroid plexus through the ventricular system to the subarachnoid space and continues to the spinal column. The classic theory suggests that CSF flow is pulsatile and generated by cardiac pulsations and pulmonary respiration (Khasawneh et al. 2018; Sakka et al. 2011). CSF not only provides mechanical support to the brain (Segal 2000) but also helps to remove toxic catabolites of the brain metabolism (Brown et al. 2004). Furthermore, CSF can be considered as a route of communication within the brain as it carries hormones, growth factors, and neurotransmitters between different areas of the brain (Kaur et al. 2016; Marques et al. 2011; Silva-Vargas et al. 2016; Preston 2001; Brown et al. 2004; Strazielle and Ghersi-Egea 2000).

The adult brain contains a constant volume of 150 ml CSF, of which 25 ml in the brain ventricles and 125 ml in the subarachnoid compartments. The total CSF production is about 500 ml per day in healthy individuals at a rate of about 0.3–0.4 ml per minute and is completely replaced about four times a day (Brown et al. 2004; Khasawneh et al. 2018). CSF is for 99% composed of water with the remaining 1% accounted for by proteins, ions, neurotransmitters, and glucose. Ion concentrations of Na+, Cl−, and Mg2+ are higher in CSF than the levels in plasma, while K+ and Ca2+ concentrations are lower (Bulat and Klarica 2011; Sakka et al. 2011). The majority of the total CSF volume (60–90%) is being produced and secreted by the choroid plexus epithelium, the remaining CSF originates from the brain interstitial fluid, ependyma, and cerebral capillaries (Redzic and Segal 2004; Sakka et al. 2011). CSF secretion by the choroid plexus is dependent on active translocation of ions and water from the basolateral membrane to the cytoplasm and subsequently across the apical membrane into the brain ventricles (Brown et al. 2004). Transport of Na+, Cl−, K+ , and HCO3 − takes place via different transporters present on the choroid plexus epithelium. Na+ and Cl− are transported into the epithelial cells by Cl−/HCO3 − and Na+ linked Cl−/HCO3 − transporters present on the basolateral surface (Lindsey et al. 1990). Translocation of these ions creates an osmotic gradient which drives water transport facilitated by aquaporins (AQP) on the epithelial surface (Liddelow 2015). At the apical surface, the Na+, K+-ATPase plays an important role creating an osmotic gradient which facilitates transfer of various molecules in and out the choroid plexus epithelium (Pershing and Johanson 1982; Plotkin et al. 1997; Redzic and Segal 2004; Speake et al. 2001; Johanson et al. 2008). Besides the Na+, K+-ATPase transporter, also the electrogenic sodium-bicarbonate cotransporter (NBCe2) facilitates Na+ ion transport into the CSF. Interestingly, a knockout of this NBCe2 cotransporter resulted in significant remodeling of choroid plexus epithelium including abnormal mitochondrial distribution, cytoskeletal protein expression, CSF electrolyte imbalance, and neurological impairment (Kao et al. 2011), reflecting the importance of cotransporter in the normal physiology of the nervous system (Christensen et al. 2018).

In elderly, the CSF production is reduced as shown in multiple studies in human, rat, and sheep (Table 9.1). The reduced expression of choroidal proteins involved in CSF secretion such as carbonic anhydrase II and AQP1 have been described in aging rat models and sheep. In addition, Na+, K+-ATPase mRNA levels decrease with age (Chen et al. 2009; Kvitnitskaia-Ryzhova and Shkapenko 1992; Masseguin et al. 2005). Next to a decreased CSF production rate, the mean CSF pressure declines steadily after the age of 50. In comparison to a 20–49 year old group, the 50–54 age group showed a reduction of 2.5% and this even increased to 13% in individuals older than 70 years of age (Fleischman et al. 2012).

Table 9.1 Changes in CSF secretion
Full size table

9.3.1.2 CSF Absorption

The site of CSF absorption is still a point of discussion in the research field. For decades it was believed that CSF returns to the venous blood in the brain sinuses through the arachnoid villi and granulations (Kida et al. 1988). These arachnoid granulations are projections of the arachnoid membrane into the dural venous sinuses. The driving force of the absorption of CSF into the venous bloodstream is a difference in fluid pressure between the subarachnoid space and the venous system. As such, fluid is driven out of the granulations into the circulation (Damkier et al. 2013). In addition, CSF absorption sites have been identified on meningeal recesses of spinal and cranial nerve roots, particularly the trigeminal and cochlear nerve (Sakka et al. 2011). Recently, dynamic imaging suggested that lymphatic outflow might be the major outflow route for CSF. Using non-invasive imaging techniques, the authors were able to demonstrate that tracers added to the CSF rapidly reach the lymph nodes using perineural routes through the foramina of the skull to finally reach the peripheral blood (Ma et al. 2017; Proulx et al. 2017). Interestingly, this lymphatic outflow system showed significant decline in aged mice (Ma et al. 2017).

9.3.1.3 CSF Turnover and Circulation

Moderate brain tissue atrophy that occurs during healthy aging, leading to an increase in total cranial CSF compartment and volume, affects the turnover or replacement time of CSF (Table 9.2) (Preston 2001). The decreased production and secretion of CSF together with increased CSF volume results in a longer CSF turnover with age. These observations have been confirmed by reduced clearance of radio-iodinated human serum albumin from the brain in individuals around 62 years of age (Henriksson and Voigt 1976). Similarly, reduced clearance of 3H-polyethylene glycol and 125I-Amyloid beta (Aβ) (1–40) was observed in older rats (Preston 2001). Cross-sectional studies in healthy humans show a doubling of CSF volume between the age of 30 and 70 years (Foundas et al. 1998; Matsumae et al. 1996a). In elderly humans, CSF turnover is reduced to two times daily in comparison to three to four times in young adults (Chiu et al. 2012; Johanson et al. 2008). In humans, additional factors have been described that contribute to the reduced CSF turnover, including increased resistance for CSF drainage by fibrosis present in the arachnoid membranes and an increase in central venous pressure, both noted to be increased in normal aging (Preston 2001; Bellur et al. 1980; Rubenstein 1998).

Table 9.2 CSF volume, turnover and clearance
Full size table

The diminished CSF production, secretion, and reduced CSF turnover rate might have serious complications. In senescence, alterations in CSF composition due to reduced turnover could bring about inadequate distribution of nutritive components and trophic factors, and the diminished CSF clearance leads to accumulation of toxic compounds and waste products from the brain (Marques et al. 2017; Preston 2001). Both will result in increased cellular stress and changes in the cerebral metabolism and blood flow, disrupting cognitive and motor functions (Rubenstein 1998), eventually influencing age-related cognitive decline and development of age-associated neurological diseases (Emerich et al. 2005). In addition, adult neural stem cells contact the CSF in the ventricular-subventricular stem cell niche of which the lateral choroid plexus is an important component. This implicates that secreted factors and toxic compounds accumulated due to diminished CSF clearance can impact the neural stem cells, which are especially sensitive to age-related changes (Silva-Vargas et al. 2016).

CSF production, clearance rate, and CSF flow are altered during aging, thereby affecting brain homeostasis. Interestingly, a highly organized pattern of ependymal cilia is responsible for the transport of CSF in the ventricles of the mouse brain. Coordinated cilia beating patterns collectively give rise to a network of fluid flows that allow for precise CSF directional flow, which may control substance distribution in the ventricle. A cilia-based switch was discovered that reliably and periodically alters the flow pattern and may control substance distribution in the ventricle (Faubel et al. 2016). However, it remains to be determined whether changes in beating patterns occur in aging and whether this affects the distribution of components throughout the brain. Peak CSF volume flow is altered in aging and higher aqueductal peak CSF flow velocities were described in elderly healthy volunteers (Gideon et al. 1994). In addition, the total cerebral blood flow decreases with aging and consequently CSF stroke volumes (i.e. mean volume of CSF passing through the aqueduct during both systole and diastole) and pulsations were significantly reduced in elderly (Stoquart-Elsankari et al. 2007).

9.3.1.4 Choroid Plexus Biochemistry

Choroid plexus functioning and metabolism are largely energy-dependent. All the homeostatic and secretory functions of the choroid plexus are linked to energy dependent mechanisms, explaining the huge number of mitochondria in the choroid plexus epithelial cells. Morphometric studies in different model organisms indicate that mitochondria constitute 10–14% of the choroid plexus cytoplasm (Cornford et al. 1997). In addition, proteome analysis in rats identified a total of 1400 proteins in the choroid plexus of which a high percentage (33.5%) are mapped to metabolism, e.g. several enzymes like hydrolases, oxidoreductases, and transferases. The presence of a substantial number of mitochondrial proteins in the proteome analyses suggests a high mitochondrial density (Sathyanesan et al. 2012). Aging however leads to a reduced metabolic activity of the choroid plexus epithelial cells, as demonstrated by in vitro choroid plexus cultures (Emerich et al. 2007). Additionally, the number of epithelial cells deficient in cytochrome C oxidase has been shown to be increased with age (Cottrell et al. 2001).

The mammalian brain depends on glucose as main energy source and a continuous supply is essential to sustain neural activity (Siesjo 1978; Simpson et al. 2007). Glucose provides energy for physiological brain functioning (biosynthesis of neurotransmitters, maintenance of action potentials, information processing) by oxidative metabolism and tight regulation of the glucose metabolism is necessary (Mergenthaler et al. 2013). Glucose transporter proteins transfer glucose from the blood circulation to the brain. The blood-CSF barrier expresses GLUT1 (Redzic 2011; Serot et al. 2003; Simpson et al. 2007). Disruption of the glucose metabolism and the pathways involved in glucose delivery can have pathophysiological consequences and lead to brain diseases. There is compelling evidence that the aging tissue is unable to maintain appropriate energy output. During aging, the expression of enzymes necessary for anaerobic respiration and oxidative phosphorylation, such as lactate dehydrogenase (LDH) and succinate-dehydrogenase (SDH), are diminished and consequently energy production in choroid plexus epithelial cells decreases (Fig. 9.1) (Emerich et al. 2005; Ferrante and Amenta 1987; Gorle et al. 2016). Both LDH and SDH play a key role in glucose metabolism and show a major reduction with age, respectively 9 and 26% (Ferrante and Amenta 1987; Preston 2001). Impairment of glucose dependent energy transduction mechanisms may influence the functional activity of the choroid plexus epithelial cells (Ferrante and Amenta 1987). In addition, in humans CSF levels of lactate increase with age (Fig. 9.1). Since CSF lactate and brain lactate concentration correlate closely, this might suggest a decline in the efficiency of glucose metabolism in brain tissue (Yesavage et al. 1982).

Different imaging methods have been developed to allow non-invasive brain measurements. Functional magnetic resonance spectroscopy (fMRS) is used for the measurement of metabolite concentrations in the human brain (Jahng et al. 2016). In addition, alterations of the glucose metabolism in the choroid plexus can be visualized and measured in vivo with dynamic fluorodeoxyglucose positron emission tomography (dynamic 18F-FDG-PET). By using this technique, the dynamic uptake of FDG in the choroid plexus and CSF can be measured over time. A recent study showed the presence of decreased glucose metabolism in Alzheimer’s disease patients. Conversely, dynamic uptake was higher in CSF for Alzheimer’s disease patients. The activity of the choroid plexus gradually decreases in patients with cognitive decline. This results in the disturbance of the glucose exchange at the blood-CSF barrier and alters the CSF-choroid plexus glucose equilibrium (Daouk et al. 2016).

9.3.1.5 Iron Metabolism

Iron is an essential element for different metabolic processes, tissue homeostasis, and brain functioning. However, in excessive amounts, iron becomes toxic for cells. Therefore, the iron delivery in the brain is strictly regulated through receptor mediated endocytosis of iron-bound transferrin by the blood-CSF barrier and the blood-brain barrier (Morris et al. 1992; Deane et al. 2004; Rouault et al. 2009; Hubert et al. 2019). During aging, decreased metabolic activity, increased oxidative stress, impaired barrier functioning, impaired protein secretion, and diminution of the CSF flow might all affect the iron metabolism and iron-mediated toxicity (Marques et al. 2009, 2007; Chen et al. 2012b). Moreover, pro-inflammatory cytokines like IL-6, which are increased in the blood with age, can influence the secretion of hepcidin by choroid plexus epithelial cells. Hepcidin is a central regulator of iron homeostasis and secretion is influenced through the Stat3 signal transduction pathway ( Chongbin et al. 2014; Chen et al. 2008; Villeda et al. 2011; Rouault et al. 2009; Hubert et al. 2019; Leitner and Connor 2012; Lu et al. 1995).

9.3.2 Growth Factors and Hormones Secreted by the Choroid Plexus

The choroid plexus is uniquely located at the interface between blood and CSF. It expresses many receptors for growth factors and hormones, such as growth hormone (GH), prolactin, corticotrophin-releasing hormone, vasopressin, and leptin, in order to respond to local and peripheral signals (Kaur et al. 2016; Marques et al. 2011; Silva-Vargas et al. 2016). In this way, the choroid plexus is a key component in neuroendocrine regulation, having an impact on hormonal signaling and in addition, also the choroid plexus functioning is regulated by a variety of hormones (Preston 2001). Several studies revealed that different hormones and neuropeptides might be actively processed by the choroid plexus, namely GH, nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), vascular endothelial growth factor (VEGF), insulin-like growth factor (IGF1 and 2), and insulin-like growth factor binding protein 2 (IGFBP2) (Emerich et al. 2005; Holm et al. 1994; Nilsson et al. 1996; Vega et al. 1992). IGF2 plays an important role in cell growth, in development, and maintenance of the nervous system and regulates the functional plasticity of the adult brain (Lenoir and Honegger 1983; Mill et al. 1985; Mozell and Mcmorris 1991). The production of IGF partially depends on the presence of GH (Cohen et al. 1992). In elderly humans, reduced binding of GH to the choroid plexus has been reported, resulting in reduced activity of IGF2, which might have an impact on epithelial cell growth and repair (Nilsson et al. 1992; Preston 2001). Next to changes in IGF, in vitro studies showed that VEGF secretion in aged choroid plexus epithelial cells was reduced compared to young epithelial cells (Emerich et al. 2007).

Interestingly, the growth factors secreted by the choroid plexus into the CSF may be involved in the proliferation, differentiation, and survival of neural progenitor cells in the subventricular zone (Falcao et al. 2012; Lun et al. 2015). More recently, it was shown that the lateral ventricle choroid plexus affects the behavior of neural stem cells of the ventricular-subventricular zone by the secretion of several factors promoting for colony formation and proliferation (Silva-Vargas et al. 2016). The functional effect of the lateral ventricle secretome changes throughout life, with activated neural stem cells being especially sensitive to age-related changes. The lateral ventricle choroid plexus is an important compartment that contributes to the age-related changes of the ventricular-subventricular zone stem cells. Transcriptome analysis revealed two proteins, BMP5 and IGF1, that might play an important role in these age-dependent effects of the choroid plexus (Silva-Vargas et al. 2016). The expression of both proteins decreases with aging, BMP5 levels are lower in aged human CSF and systemic IGF levels decrease with aging (Baird et al. 2012; Bartke et al. 2013). Furthermore, a study revealed that implants of young choroid plexus in rats were potently neuroprotective, whereas the choroid plexus implants from aged rats were only modestly effective and less potent. This study links aging with a diminished neuroprotective capacity of the choroid plexus epithelial cells (Emerich et al. 2007).

Vasopressin is a neurohormone produced by the hypothalamus, involved in the regulation of blood pressure. A high density of Vasopressin receptors (V1) is present at the choroid plexus. The activation of the V1 receptors regulates CSF production by decreased choroidal blood flow or by the effect on the choroidal epithelial cells’ ion channels (Chodobski and Szmydynger-Chodobska 2001; Faraci et al. 1988). Vasopressin is able to reduce the efflux of Cl− ions by regulating the Na+, K+, 2Cl−cotransporter, and maintains the volume of the choroidal epithelial cells. Vasopressin levels in blood and CSF can vary substantially, and elevated levels of vasopressin have been found in the CSF of old rats and in the plasma of elderly humans (Fig. 9.1) (Frolkis et al. 1999). In addition to vasopressin, also angiotensin II and endothelin-1, secreted by the choroid plexus, can affect the choroidal blood flow and CSF secretion (Kaur et al. 2016). A reduced production and secretion of CSF could influence the delivery of many components to the brain and may interfere with the normal physiological pathways (Kaur et al. 2016; Preston 2001).

The Klotho protein is a transmembrane protein that was identified as aging-suppressor (Kuro et al. 1997). A defect in Klotho gene expression in mice accelerates aging-like phenotypes and results in a syndrome that resembles human aging including a short lifespan, impaired cognition (Uchida et al. 2001; Shiozaki et al. 2008), abnormal brain pathology, infertility, arteriosclerosis (Arking et al. 2003), skin atrophy, osteoporosis (Ogata et al. 2002), and emphysema (Kuro et al. 1997). Vice versa, the overexpression of Klotho in mice extends life span and improves memory (Kurosu et al. 2005; Li et al. 2019). Moreover, gene expression analysis of brain white matter in rhesus monkeys also indicated the implication of Klotho in the regulation of brain aging (Duce et al. 2008). In humans, a functional variant of the KLOTHO (KL) gene showed to be associated with high-density cholesterol, blood pressure, stroke, and longevity (Arking et al. 2002, 2005). The Klotho protein functions as a circulating hormone that represses intracellular signals of insulin and IGF1, an evolutionarily conserved mechanism for extending life span. In Klotho-deficient mice the disruption of insulin and IGF1 signaling lead to the improvement of aging-like phenotypes, suggesting that Klotho-mediated inhibition of insulin and IGF1 signaling contributes to its anti-aging properties (Kurosu et al. 2005). Klotho is predominantly secreted by the choroid plexus, the distal tubule cells of the kidney, and parathyroid glands; high levels of Klotho are expressed in the choroid plexus of juvenile and adult mice, humans, and mammals (Kuro 2010). Soluble Klotho has been demonstrated to be present in human CSF and serum (Imura et al. 2004; Semba et al. 2014). Aging is associated in mice with decreased klotho expression in the choroid plexus (Zhu et al. 2018). Moreover, CSF klotho concentrations are lower in older versus younger cognitive healthy individuals and in addition, CSF klotho concentrations are significantly lower in Alzheimer’s disease patients compared to adults without cognitive problems (Semba et al. 2014). Selective depletion of klotho in the choroid plexus triggered the expression of multiple proinflammatory factors and macrophage infiltration into the choroid plexus. Furthermore, experimental reduction of klotho in the choroid plexus demonstrated enhanced microglial activation in the hippocampus following peripheral stimulation with lipopolysaccharide (Zhu et al. 2018). These results suggest that klotho depletion from the choroid plexus could contribute to the age-dependent priming of microglia for activation by peripheral infections (Henry et al. 2009; Zhu et al. 2018). In primary macrophage cultures, Klotho suppressed the activation of the NLRP3 inflammasome by enhancing fibroblast growth factor (FGF)23 (Zhu et al. 2018). This suggests that Klotho controls the brain-immune systems interface in the choroid plexus. Moreover, Klotho depletion in aging or disease may weaken this barrier and promote immune-mediated neuropathogenesis (Zhu et al. 2018). In addition, Klotho secreted by the choroid plexus might enhance oligodendrocyte maturation and myelination of the CNS (Chen et al. 2013). In this way it may play a role in the prevention of myelin degeneration in the aging brain (Chen et al. 2013; Semba et al. 2014).

9.3.3 Barrier Permeability and Transport by the Choroid Plexus

The blood-CSF barrier ensures a stable, balanced, and well-controlled micro-environment of the brain, which is necessary for proper functioning of the CNS. Transport across the barrier is restricted by tight junctions between the choroid plexus epithelial cells and require transporter and receptor systems in a directional way (Redzic 2011; Saunders et al. 2013). The choroid plexus produces CSF by passive filtration of fluid across the fenestrated capillaries and regulated secretion of molecules across the choroid plexus epithelial cells (Brinker et al. 2014), together with active production of molecules by the choroid plexus epithelial cells (Thouvenot et al. 2006). Dysregulation of choroid plexus transporters and tight junction complexes subsequently reflects into CSF compositional changes. Several studies have reported a compromised blood-CSF barrier in response to inflammatory signals (Brkic et al. 2015; Marques and Sousa 2015; Vandenbroucke et al. 2012). As described previously, aging is associated with morphological changes of the choroid plexus epithelial cells and is in addition associated with a state of low grade, chronic inflammation or inflamm-aging which might lead to the loss of the barrier function at the choroid plexus. Loss of barrier integrity might result in leakage of components from the blood circulation into the CSF, thus changing CSF composition. In agreement with this, a study performed by Chen et al., showed increased blood-CSF permeability for proteins upon aging in sheep. However, no complete disruption of the barrier is present since the passage of larger molecules was still prevented (>109.51–120 kDa) (Chen et al. 2009, 2012a). Studies conducted in healthy elderly individuals report only small changes in CSF composition: the concentration of molecules including Transthyretin (TTR) (Serot et al. 2003; Kleine et al. 1993b), alpha2-macroglobulin (Garton et al. 1991; Kleine et al. 1993b), and IgG (Blennow et al. 1993a; Garton et al. 1991; Kleine et al. 1993b; Chen et al. 2018) increases slightly with age. The CSF versus serum albumin ratio is used to evaluate blood-CSF barrier functioning and an increased variability in the CSF/serum albumin ratio has been observed from the age of 45 years, indicating that the blood-CSF barrier is compromised in elderly humans (Fig. 9.1) (Blennow et al. 1993a, b, c). However, it is difficult to determine whether this is the result of increased blood-CSF barrier permeability or altered clearance of the proteins (Preston 2001; Serot et al. 2003, 1997). Often these elevated levels are interpreted as blood-CSF barrier integrity loss.

Choroid plexus epithelial cells are able to secrete EVs, including exosomes, into the CSF as a mechanism of blood-brain communication (Balusu et al. 2016b). EVs are membrane-derived vesicles that can enclose specific repertoires of proteins, lipids, and RNA molecules (Van Niel et al. 2018; Mathieu et al. 2019) and are able to transport these molecules both to adjacent and distant cells (Baixauli et al. 2014; Mittelbrunn and Sanchez-Madrid 2012; Paolicelli et al. 2018). In the CNS, EVs have shown to mediate intercellular communication over long range distances and are believed to be important for the cross-talk between neurons and glial cells in the brain (Paolicelli et al. 2018). Inflammation, which is also present in the aging brain, was shown to induce an increase in EV production, together with an altered EV content (Balusu et al. 2016b). The number of EVs present in the CSF declines in elderly humans and their miRNA content changes throughout life (Fig. 9.1) (Tietje et al. 2014). However, the size of the EVs and their size distribution did not change during aging (Tietje et al. 2014; Yang et al. 2015). EVs in the CSF can be produced by different cell types and no data is currently available on EV production by the choroid plexus epithelial cells during aging. However, if affected, this might have consequences for the nutrient delivery to the brain. As an example, exosomes, a specific type of EVs, which are secreted via the fusion of multivesicular bodies with the plasma membrane, are important for the delivery of folate, an important vitamin for the brain, across the choroid plexus epithelial cells into the CSF (Grapp et al. 2013).

9.4 Immune Cell Trafficking at the Choroid Plexus

Migration of immune cells into brain tissue and (limited) inflammatory reactions are fundamental mechanisms to sustain normal physiology, immune surveillance, host defense, and learning processes (Engelhardt and Coisne 2011; Garner et al. 2006; Ransohoff and Engelhardt 2012; Galea et al. 2007). However, these mechanisms are tightly controlled by the presence of different brain barriers. The blood-CSF barrier is perfectly located at the interface between blood and CSF to provide active immune surveillance and serves as active and selective gate for immune cell trafficking (Demeestere et al. 2015). The tightly connected choroidal epithelium limits paracellular transport of not only molecules, but also immune cells. Additionally, the choroid plexus contains fenestrated capillaries, allowing free communication between the stroma and peripheral blood (Demeestere et al. 2015). The choroid plexus stroma contains a large population of macrophages, dendritic cells, CD3+ and CD4+ T cells, and CX3CR1hi Ly6Clow monocytes (Shechter et al. 2013). Macrophages at the apical side of the choroid plexus are called epiplexus or Kolmer cells (Maslieieva and Thompson 2014) and are thought to contribute to the immune component of the blood-CSF barrier. In healthy conditions, the CSF contains CD4+ T cells, natural killer cells, and B cells (Ransohoff and Engelhardt 2012). Although leukocytes enter the CSF, in steady state conditions they do not invade the brain parenchyma (Shechter et al. 2013). Leukocyte infiltration however is modulated in response to disease or trauma like meningitis, multiple sclerosis, or peripheral inflammation. The cells can transmigrate from the blood across the fenestrated endothelium to enter the stroma matrix. After travelling through the stroma of the choroidal cells, the immune cells can, in response to specific triggers, cross the choroid plexus epithelium and enter the CSF where they are able to skew toward specific effector responses, including regulatory T cells, T helper 2 cells, and alternatively activated macrophages (Shechter et al. 2013). The CD4+ T cells present in the CSF are distinct from the T cell populations in the blood circulation and brain parenchyma, indicating that the influx of T cells via the choroid plexus into the CSF is highly regulated (Engelhardt and Ransohoff 2012). After entering the CSF, leukocytes might be able to cross the ependymal cell layer and migrate further into the brain parenchyma under inflammatory conditions or might travel to the arachnoidea via the CSF flow.

The expression of adhesion molecules, chemokines, and chemokine receptors control leukocyte trafficking across the blood-CSF barrier. Inflammation causes the upregulation of adhesion molecules in the choroid plexus epithelial cells, such as intercellular adhesion molecule 1 (ICAM-1), vascular cellular adhesion molecule 1 (VCAM-1), and mucosal vascular addressin cell adhesion molecule 1 (MADCAM-1) (Endo et al. 1998). The expression of cell adhesion molecules at the brain barriers could possibly be increased during aging because of the pro-inflammatory state of the brain. The choroid plexus also produces cytokines (e.g. interleukin-1β (Il-1β) and tumor necrosis factor α (TNFα)) and chemokines (e.g. C-X-C motif chemokine ligand (CXCL10) and monocyte chemoattractant protein 1 (MCP1)). These cyto- and chemokines are necessary for the activation and/or recruitment of immune cells during systemic inflammation (Demeestere et al. 2015). Interestingly, aging leads to an increased expression of Il-1β in the choroid plexus (Silva-Vargas et al. 2016).

In the healthy adult brain, a balance is present between pro- and anti-inflammatory cytokines, but Baruch and colleagues observed in the choroid plexus a shift towards a Th2-like pro-inflammatory state with increasing age (Baruch et al. 2013; Sparkman and Johnson 2008). This pro-inflammatory state is reflected by the reduced production of interferon (IFN)-γ and increased production of IL-4, which negatively affect brain functioning (Baruch et al. 2013). Mice lacking the IFNγ receptor show a decreased number of leukocytes in the CSF and premature cognitive decline (Baruch et al. 2013). A type I IFN signature was described in the aged choroid plexus (Baruch et al. 2014). In both mouse models and human samples, the choroid plexus showed an increased type I and decreased type II IFN dependent gene expression profile (Fig. 9.1). This type I IFN signature negatively influences type II IFN signaling, leading to a reduced expression of homing and trafficking molecules (Cd34, Madcam1, Ccl2, Cx3cr1, Cxcl13, Il2) that are required for leukocyte entry in the CSF during aging, eventually leading to increased brain inflammation and cognitive decline (Baruch et al. 2014; Kunis et al. 2013). Interestingly, blocking IFN type I signaling restored cognitive functioning and hippocampal neurogenesis and in addition was able to diminish astrogliosis and microgliosis, and increase the anti-inflammatory cytokine IL-10 in the hippocampus (Baruch et al. 2014). It was suggested that this IFN I signaling is a mechanism to attenuate neuroinflammation which eventually becomes detrimental to brain plasticity resulting in age-associated cognitive decline (Baruch et al. 2014). However, therapeutically targeting the IFN pathway could influence the immune surveillance at the choroid plexus, since a tight balance between type I IFNs and IFNγ is central in leukocyte entry and cognition (Deczkowska et al. 2016). The importance of this balance is reflected in the IFNβ treatment, which is used in the clinic to reduce clinical relapses in multiple sclerosis (Wingerchuk and Carter 2014). Patients treated with IFN experience several adverse effects including increased risk for developing depression, cognitive decline, and they develop Parkinson like symptoms (Manouchehrinia and Constantinescu 2012). Similarly, type I IFNs aggravate disease in multiple mouse models of Parkinson’s disease (Main et al. 2017).

It remains to be determined whether loss of barrier integrity is responsible for immune cell trafficking across the blood-CSF barrier. Independent of barrier impairment, other mechanisms might determine the leukocyte migration across the blood-CSF barrier. Nitric oxide, a negative regulator of leukocyte trafficking has been found to be upregulated at the choroid plexus during aging (Baruch et al. 2015). Additionally, transcellular migration events of leukocytes occurring in close proximity of the tight junctions have been undervalued and might have been mistaken for paracellular migration (Phillipson et al. 2008; Wewer et al. 2011; Wolburg et al. 2005). Steinmann et al. were able to demonstrate transcellular migration of leukocytes across the blood-CSF barrier after bacterial infection as well as T-cell transmigration after viral stimulation (Steinmann et al. 2013; Wewer et al. 2011). Moreover, polymorphonuclear (PMN) and monocytes differentially migrate in a human blood-CSF barrier model (Steinmann et al. 2013).

9.5 Choroid Plexus and Neurodegenerative Diseases

Interestingly, all the age-related morphological changes in choroid plexus structure described above, such as the flattening of epithelium, thickening of basement membrane, and lipofuscin deposits, are significantly more prominent in neurodegenerative diseases.

The most prevalent neurodegenerative disease, Alzheimer’s disease, is characterized by the decline of memory and other cognitive functions. It is a progressive deteriorating disease, eventually leading to loss of autonomy and ultimately patients require full-time medical care (Jost and Grossberg 1995). Pathologically, Alzheimer’s disease is defined by severe neuronal loss, resulting in the loss of brain volume, which is most pronounced around the medial temporal lobe areas, and particularly in the hippocampus. Furthermore, the aggregation of beta-amyloid (Aβ) in extracellular senile plaques and formation of intraneuronal neurofibrillary tangles consisting of hyperphosphorylated tau protein have been identified to play a major role in Alzheimer’s disease pathogenesis (Braak and Braak 1991).

Ultrastructural changes are similar to those described in aging namely epithelial cell atrophy. In humans, the cells decrease in height with approximately 22% compared to healthy controls (Serot et al. 2000, 2003). The cytoplasm of the epithelial cells contains multiple lipofuscin and Biondi tangles (Miklossy et al. 1998). The basement membrane of the epithelium is thickened and irregular. Apical microvilli become irregular and fibrotic (Serot et al. 2000; Jellinger 1976). The stroma contains calcifications and psammomas, hyaline bodies, and thickened vessel walls (Serot et al. 2003).

TTR, a highly expressed protein at the choroid plexus, is a carrier for the thyroid hormones, but also has the ability to bind to Aβ and prevents the aggregation and deposition of Aβ plaques in the brain (Marques et al. 2013; Schwarzman et al. 1994). Studies reporting the production and secretion of TTR by the choroid plexus during aging show conflicting data: both an increase and decrease of TTR in the CSF has been described (Kleine et al. 1993a; Redzic et al. 2005; Serot et al. 1997). The blood-CSF barrier expresses several transporter systems including low-density lipoprotein receptor-related protein 1 (LRP1), receptor for advanced glycation end products (RAGE), receptor glycoprotein 330/megalin (LRP2), and Pgp. The LRP and Pgp receptors are responsible for the receptor-mediated efflux of Aβ from the brain, while RAGE mediates the influx of Aβ into the brain (Marques et al. 2013; Storck et al. 2016). Expression of LRP2 in the choroid plexus is decreased during aging, but an increase of LRP1 and Pgp is observed as well as no difference in RAGE expression (Gorle et al. 2016; Pascale et al. 2011).

Several studies have reported the beneficial effect of the choroid plexus on the rejuvenation of damaged brain regions because of the production of neurotrophic factors (Thanos et al. 2010; Borlongan et al. 2004a, b; Bolos et al. 2014). Choroid plexus epithelial cells treated in vitro with Aβ peptide lead to increased proliferation and differentiation of neuronal progenitor cells (Bolos et al. 2014). Moreover, transplantation of healthy choroid plexus epithelial cells into the brain of an Alzheimer’s disease mouse model induced a significant reduction in brain Aβ and tau levels and improved memory of the animals (Bolos et al. 2014). Also in other neurodegenerative diseases, such as Huntington’s disease choroid plexus cell transplantation studies showed successful results (Emerich and Borlongan 2009).

9.6 Conclusions

The choroid plexus, that contains the blood-CSF barrier, accomplishes important functions in the CNS and actively contributes to brain homeostasis. The choroid plexus is able to respond to changes both in the periphery and the brain parenchyma. However, during aging, the morphology and normal functioning of the choroid plexus is severely compromised. Alterations in brain barrier transport mechanisms, CSF production and clearance, receptor-mediated signaling, enzymatic and metabolic activity, loss of barrier integrity, and insufficient distribution of nutrients have an effect on brain functioning and might influence cognitive performance. Understanding how the blood-CSF barrier is altered in aging and how it can contribute to these age-associated diseases, might lead to novel strategies to attenuate aging-associated cognitive decline and related diseases.