The role of cannabinoids in adult neurogenesis

Abstract

The processes underpinning post-developmental neurogenesis in the mammalian brain continue to be defined. Such processes involve the proliferation of neural stem cells and neural progenitor cells (NPCs), neuronal migration, differentiation and integration into a network of functional synapses within the brain. Both intrinsic (cell signalling cascades) and extrinsic (neurotrophins, neurotransmitters, cytokines, hormones) signalling molecules are intimately associated with adult neurogenesis and largely dictate the proliferative activity and differentiation capacity of neural cells. Cannabinoids are a unique class of chemical compounds incorporating plant-derived cannabinoids (the active components of Cannabis sativa), the endogenous cannabinoids and synthetic cannabinoid ligands, and these compounds are becoming increasingly recognized for their roles in neural developmental processes. Indeed, cannabinoids have clear modulatory roles in adult neurogenesis, probably through activation of both CB₁ and CB₂ receptors. In recent years, a large body of literature has deciphered the signalling networks involved in cannabinoid-mediated regulation of neurogenesis. This timely review summarizes the evidence that the cannabinoid system is intricately associated with neuronal differentiation and maturation of NPCs and highlights intrinsic/extrinsic signalling mechanisms that are cannabinoid targets. Overall, these findings identify the central role of the cannabinoid system in adult neurogenesis in the hippocampus and the lateral ventricles and hence provide insight into the processes underlying post-developmental neurogenesis in the mammalian brain.

Tables of Links

TARGETS
GPCRsa	Enzymese
CB1 receptor	Akt (PKB)
CB2 receptor	DAGLα
Ligand-gated ion channelsb	DAGLβ
NMDA receptor	ERK
Ion channelsc	FAAH
TRPV1	Monoacylglycerol lipase
Catalytic receptorsd	mTOR
EGFR	PI3K
IL-1 receptor, type I
IL-1 receptor, type II
TrkB (BDNF receptor)

LIGANDS
2-AG	Glutamate
5-HT	HU-210
Δ9-THC	HU-308
AA	IGF-1
ACEA	IL-1β
Adenosine	IL-6
AM251	IL-10
AM630	JWH-133
AM1241	NGF
Anandamide (AEA)	Noradrenaline
BDNF	SR141716A
Cannabidiol	SR144528
Cannabinol	TNF-α
Dopamine	URB597
FGF-2	VEGF
GABA	WIN55,212-2

These Tables list key protein targets and ligands in this article which are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Pawson et al., 2014) and are permanently archived in the Concise Guide to PHARMACOLOGY 2013/14 (^a,b,c,d,eAlexander et al., 2013a,b,c,d,e,,,,).

Introduction

For decades, the true plasticity of the mammalian CNS was underestimated and the adult brain was long considered to be a post-mitotic organ incapable of self-regeneration. However, pioneering work in the 1960s by Joseph Altman and colleagues challenged this long-standing dogma (Altman and Das, 1965). In this groundbreaking publication, Altman provided the first evidence that new neurons were generated in the adult rat hippocampus. Subsequent experiments demonstrated that adult neurogenesis was not specific to the hippocampus, with the adult olfactory bulb identified as another brain region where new neurons are added to existing circuitry throughout life (Altman, 1969). In spite of this work, the concept of post-developmental neurogenesis in the mammalian brain was subject to contemporary scepticism; currently, however, the phenomenon of adult neurogenesis is widely studied and research in the intervening years has confirmed adult neurogenesis in the murine hippocampus (Cameron et al., 1993; Kempermann et al., 1997), while the lateral ventricles (Lois and Alvarez-Buylla, 1993), regions adjacent to the ventricles (such as striatum and septum), as well as the thalamus and hypothalamus (Pencea et al., 2001) have been shown to be capable of generating new neurons during adulthood. In the human brain, evidence continues to mount to support the absence of neurogenesis in the adult human neocortex (Rakic, 2006). However, adult neurogenesis has been described in the hippocampus (Eriksson et al., 1998), the lateral ventricles (Sanai et al., 2004) and more recently in the striatum (Ernst et al., 2014).

Cannabinoids incorporate the active components of the hemp plant Cannabis sativa (the plant-derived cannabinoids), the endogenous cannabinoids (endocannabinoids) produced in humans and animals and the synthetic cannabinoid compounds. The cannabinoid system is linked with all aspects of human physiology and elicits diverse effects by activating the G protein-coupled cannabinoid receptors (CB) type 1 (CB₁) and type 2 (CB₂) subtypes, the expression of which has been localized on glia, immune cells and neurons throughout the CNS (Downer, 2011). A body of data indicates that cannabinoid ligands control cell genesis in the adult brain, regulating cell proliferation and overall neurogenesis in the mammalian brain (Kochman et al., 2006; Mackowiak et al., 2007). Furthermore, neural progenitor cells (NPCs) express a functional endocannabinoid system (Aguado et al., 2005; Compagnucci et al., 2013) and are producers of endogenous cannabinoids (Butti et al., 2012). Such findings, alongside a number of knockout studies targeting enzymes involved in the biosynthesis and degradation of endocannabinoids (Aguado et al., 2005; Gao et al., 2010), in addition to CB₁ (Jin et al., 2004) and CB₂ receptors (Palazuelos et al., 2006), place the cannabinoid system as a key player in the processes underlying adult neurogenesis.

Adult neurogenesis

Adult neurogenesis can be loosely divided into four stages: proliferation of neural stem cells (NSCs) and NPCs, migration, neuronal differentiation and finally integration into functional synaptic networks. The two regions in which adult neurogenesis has been most extensively studied are the dentate gyrus of the hippocampus and the lateral ventricles. NSCs in the dentate gyrus reside predominantly in the subgranular zone (SGZ) where four types (type I, type IIa, type IIb and type III) have been characterized based upon proliferation rate, protein expression and morphology. In the murine forebrain, all newborn neurons are derived from type I NPCs, these cells possess a glial-like radial process, although are predominantly unipolar/bipolar in contrast to multipolar astrocytes, express glial fibrillary acidic protein (GFAP) and the intermediate filament protein nestin (Garcia et al., 2004). Type I NSCs are characterized by a low rate of proliferation (Ahn and Joyner, 2005). In contrast, type IIa cells are non-radial, do not express GFAP and exhibit a considerably higher proliferation rate compared with the relatively quiescent type I cells. Type IIa cells maintain nestin expression and both cell types are positive for the Sox gene family (Suh et al., 2007). Type IIb cells maintain important properties of stem cells as they uphold expression of nestin and Sox, but begin to express markers of neuronal committed progenitors, in particular the microtubule-associated protein doublecortin (DCX). If local conditions are favourable, type IIb cells can mature to the nestin negative/DCX positive early neuronal type III cell (Kronenberg et al., 2003).

In lateral ventricles, the subventricular zone (SVZ) contains the majority of ventricular NSCs and is one of the key regions of the brain where neurogenesis occurs throughout adulthood (Curtis et al., 2007). Three cell types have been discovered in the SVZ: type B cells much resemble type I cells in the SGZ; they are GFAP positive, possess a radial process and have a relatively low proliferation rate. Type C cells in the SVZ are reminiscent of type II cells in the SGZ as they are GFAP negative, non-radial and highly proliferative. Both cell types express nestin and Sox (Doetsch et al., 1997). Type A cells represent a population of neuroblasts which migrate at a rate of 30 000 per day along the rostral migratory stream (RMS) to the olfactory bulb (Alvarez-Buylla et al., 2001).

NSCs in both the dentate gyrus and the lateral ventricles have the capacity to produce cells that differentiate to neurons, astrocytes and oligodendrocytes (Gage, 2000). Neuroblasts originating in the SVZ primarily differentiate into olfactory bulb interneurons (Luskin, 1993). Under the right conditions, NSCs in the dentate gyrus can migrate to the granular cell layer and give rise to granular cells that integrate into the hippocampal circuitry forming glutamatergic synapses with granular neurons, interneurons and pyramidal cells in cornu ammonis region 3 (Toni et al., 2008). It has been suggested that these new born granular cells begin to resemble mature neurons, with regard to both their morphology and electrophysiological properties after approximately 4 weeks, although the maturation process continues for several months (Suh et al., 2009). In the young adult rat hippocampus, approximately 9000 new cells are generated each day with 50% of these cells expressing neuronal markers within 5–12 days. Although survival rate is low, it has been estimated that each month the number of new granular cells generated equates to about 6% of the total granular cell number (Cameron and McKay, 2001).

Extrinsic signals in adult neurogenesis

NSC/NPCs are highly sensitive to their microenvironment (i.e. their stem cell niche) and extrinsic signalling molecules largely dictate the proliferative activity and differentiation capacity of these cells. The functions of neurotrophic factors as extrinsic signalling molecules in adult neurogenesis continues to be unravelled, with strong evidence indicating that Trk receptors (and p75NTR co-receptor) are abundant on dividing progenitor cells in the adult primate SVZ/SGZ (Tonchev et al., 2007), with a body of literature indicating that brain-derived neurotrophic factor (BDNF) is a central player in adult neurogenesis. A common method of labelling proliferating cells in the dentate gyrus is to administer the thymidine analogue 5-bromo-2'deoxyuridine (BrdU), which incorporates into the DNA of cells during the S-phase of the cell cycle thus allowing the post-mortem identification of cells that have undergone proliferation. Intrahippocampal infusion of BDNF has been shown to increase the number of cells positive for BrdU and the neuron-specific protein neuronal nuclei in adult rats (Scharfman et al., 2005), while dentate gyrus-specific BDNF RNA interference reduces net neurogenesis in rats by impairing the survival of immature neurons (Taliaz et al., 2010). Similarly, NPC-specific deletion of the high-affinity BDNF receptor TrkB in mice compromises dendritic development and the survival capacity of immature neurons (Bergami et al., 2008), while BDNF-TrkB signalling has been shown to be imperative for hippocampal NSC proliferation in mice (Li et al., 2008). Of note, two other neurotrophic factors have been implicated in the regulation of adult neurogenesis; nerve-growth factor (NGF) has been shown to increase cell proliferation (Birch and Kelly, 2013) and immature neuron survival (Frielingsdorf et al., 2007) in the rat dentate gyrus, while VEGF has also been shown to induce cell proliferation (Jin et al., 2002) and promote immature neuron survival (Schanzer et al., 2004) in the SVZ and SGZ of the adult rat.

In addition to neurotrophic factors, data indicate that several growth factors, including insulin-like growth factor-1 (IGF-1) and FGF-2 are extrinsic factors involved in the regulation of adult neurogenesis. Indeed, s.c. or intraventricular infusion of IGF-1 enhances neurogenesis in the adult rat hippocampus (Aberg et al., 2000), while data from Zhao et al. (2007) demonstrate that conditional deletion of FGFR1 impairs the proliferation of NPCs in the dentate of adult mice (Zhao et al., 2007).

Neurotransmitters are also important regulators of neurogenesis in the adult brain. In particular, Bolteus and Bordey (2004) demonstrated that GABA has a direct effect on migrating neuroblasts in the adult mouse SVZ (Bolteus and Bordey, 2004), while many other studies have delineated the role of GABA in the regulation of NSC proliferative activity, fate decision and synaptic integration of immature neurons (Pallotto and Deprez, 2014). Similarly, glutamate can influence both proliferation and survival of NPCs; activation of the NMDA glutamate receptor has an inhibitory effect on cell proliferation and net neurogenesis in the rat (Cameron et al., 1995) and, in a somewhat paradoxical fashion, induction of LTP at the perforant path-dentate gyrus pathway in rats increases proliferation and survival of NPCs/immature neurons via a NMDA receptor-dependent mechanism (Bruel-Jungerman et al., 2006). Furthermore, the NMDA receptor has been shown to regulate survival of neuroblasts migrating from the mouse SVZ (Platel et al., 2010). Taken together, this suggests a complex role for glutamate in neurogenesis regulation. Additionally, monoamine neurotransmitters such as 5-HT, noradrenaline and dopamine have also been identified as neurogenic modulators, either via direct links in the case of dopamine (Van Kampen et al., 2004) or due to the fact that antidepressants and antipsychotics targeting these systems can affect neurogenesis (Dranovsky and Hen, 2006).

The immune system can also heavily influence the fate of NSCs/NPCs with the antiproliferative and antineuronal differentiative effects of inflammatory cytokines such as IL-6, IL-1β and TNF-α (Kohman and Rhodes, 2013). Elsewhere, the pro-neurogenic effects of the anti-inflammatory cytokine IL-10 have been demonstrated in the amyloid precursor protein/presenilin protein 1 transgenic mouse (Kiyota et al., 2012). Importantly, a body of data indicates that cross-talk may exist between inflammatory mediators (particularly TNF-α) and NSCs/NPCs that may have important consequences for neural development and repair in disease states. Indeed, central administration of TNF-α to rats increases BrdU incorporation in SVZ cells (Wu et al., 2000), while inhibiting endogenous TNF-α signalling regulates the proliferative capacity of mouse neural precursor cells (Rubio-Araiz et al., 2008). In support of this, clear evidence indicates that this cytokine is up-regulated in the mouse brain during demyelination and remyelination, enhancing the proliferative capacity of oligodendrocyte progenitor cells (Arnett et al., 2001). Furthermore, Katakowski et al. (2007) have shown that TNF-α-converting enzyme proteolysis promotes stroke-induced SVZ progenitor cell neurogenesis in rats (Katakowski et al., 2007), indicating that TNF-α signalling may intricately impact neural development and brain repair, particularly in stroke pathogenesis.

Finally, several hormones including thyroid hormones (Remaud et al., 2014), glucocorticoids and, perhaps more speculatively, oxytocin (Schoenfeld and Gould, 2012) have been linked to neurogenesis regulation.

Intrinsic signals in adult neurogenesis

A large body of research has delineated the multiple mechanisms regulating events associated with adult neurogenesis, including cell proliferation, differentiation, maturation, migration and integration of neural cells into neuronal networks (Gage, 2000). Furthermore, through studies predominantly performed in rodents, the complexity of the cellular and molecular signalling processes regulating neurogenesis in the mammalian brain continues to be deciphered. It is now clear that key intrinsic signalling pathways involving Sonic Hedgehog (Shh), Wnt, bone morphogenetic protein (BMP), Notch and transcription factors are intimately associated with adult neurogenesis (Faigle and Song, 2013).

Shh is a signalling glycoprotein which acts through the patched 1 (Ptc1)–smoothened (Smo) receptor complex to activate intricate signal transduction pathways involved in the development of the CNS (Ruiz i Altaba et al., 2002). Indeed, Ptc and Smo are expressed in the adult hippocampus (Traiffort et al., 1998) and conditional deletion of Smo reduces the proliferation of progenitor cells in the postnatal hippocampus and SVZ (Machold et al., 2003). In support of this, pharmacological inhibition of Shh signalling has been shown to reduce granule cell proliferation in the adult rat dentate gyrus (Lai et al., 2003). More recent evidence also indicates that Shh signalling mediates cellular migration in the adult mouse mammalian brain (Balordi and Fishell, 2007), indicating the multifaceted role of Shh signalling in neurogenesis.

The Wnt signalling pathway is a long-standing player in the regulation of adult neurogenesis (McMahon and Bradley, 1990). Wnt ligands are a family of glycoproteins that play a role in the maturation of neurons, remodelling of axons and the maintenance of adult tissue homeostasis (Clevers and Nusse, 2012). Indeed, Wnt signalling, via β-catenin, mediates cellular differentiation in adult-derived mouse hippocampal progenitor cells (Lie et al., 2005) and data elsewhere indicates that Wnt-mediated neurogenesis requires NeuroD1 in adult mouse hippocampal NPCs (Gao et al., 2009). Overall, loss of function of Wnt signalling is strongly associated with determining the development of CNS disorders (De Ferrari and Inestrosa, 2000; Lovestone et al., 2007).

BMPs are members of the TGF-β superfamily and consist of at least 20 growth factors that act as key regulators of axonal growth in a number of neuronal populations (Hegarty et al., 2013). Indeed, clear evidence indicates that BMPs act as potent inhibitors of neuronal differentiation in the adult mouse SVZ (Lim et al., 2000), while Mira et al. (2010) have demonstrated that inhibition of BMP signalling in adult mouse SGZ neural precursor cells differentially regulates neurogenesis.

The components of the Notch signalling pathway are expressed in the SVZ and SGZ of the adult mammalian brain and data indicates that this pathway, through the inhibition of proneural genes, is a key regulator of neurogenesis in the CNS (Irvin et al., 2004). Indeed, Notch signalling is associated with reducing the adult mouse neural progenitor pool (Hitoshi et al., 2002) and promoting the self-renewal of nestin-expressing cells in the adult mouse SGZ (Ables et al., 2010). Interestingly, recent evidence indicates that cross-talk between Notch and EGFR signalling exist, with downstream consequences on NSCs/NPCs in the adult mouse SVZ (Aguirre et al., 2010). Furthermore, Notch 1 knockout mice demonstrate a reduction in dendritic trees associated with granule cells in the mouse dentate gyrus (Ables et al., 2010), highlighting the intrinsic role of Notch signalling in an array of neurodevelopmental cellular processes.

Recently, several transcription factors have been highlighted for their role in adult neurogenesis. In addition to the long-standing role of cAMP response element-binding protein (CREB) in regulating cell development (Finkbeiner et al., 1997), more recent data indicate that CREB phosphorylation robustly enhances progenitor cell proliferation and controls the survival of new neurons in the adult mouse hippocampus in vivo (Jagasia et al., 2009). Interestingly, overexpression of Ascl1 transcription factor regulates the fate of oligodendrocytes in the mouse SGZ in vivo (Jessberger et al., 2008) and both the orphan nuclear receptor Tlx (Zhang et al., 2008) and Sox2 gene family (Ferri et al., 2004) are central in regulating NSC proliferation in the mouse hippocampus. In support of this data indicating that transcription factors are strongly linked to neural differentiation in the rodent brain in vivo, further evidence has identified that Tbr2 (Hodge et al., 2012) and distal-less (Brill et al., 2008) are also associated with neural differentiation in the mouse dentate gyrus and olfactory bulb respectively.

Cannabinoids

The Cannabis plant has been utilized by humans in several capacities for thousands of years and Western medicine has recognized its therapeutic potential since the late 1800s (Reynolds, 1890). Today, this potential is still recognized (Robson, 2014) and the properties of the endocannabinoid system continue to be deciphered.

The CB₁ receptor was first described and cloned in the early 1990s (Matsuda et al., 1990; Gerard et al., 1991); it was found to be abundantly expressed throughout the CNS, and, in particular, in areas associated with learning and memory including the hippocampus (Herkenham et al., 1990). A second cannabinoid receptor, the CB₂ receptor, was also cloned in the 1990s (Munro et al., 1993) where it was initially thought to be localized to the periphery; however, its expression in the CNS has been demonstrated (Gong et al., 2006). Shortly after the identification of these receptors [receptor nomenclature follows (Alexander et al., 2013a)], their endogenous ligands, known as endocannabinoids, were discovered. The two endocannabinoids that have been studied in most detail are N-arachidonoylethanolamide (also known as anandamide; AEA) (Devane et al., 1992) and 2-arachidonoylglycerol (2-AG) (Mechoulam et al., 1995). AEA is a phospholipid-derived molecule that is an agonist at the CB₁ and CB₂ receptor; it is detectable peripherally in the plasma and throughout the mammalian brain; in particular. it is found at high concentrations in the hippocampus, cerebellum and cortex (Felder and Glass, 1998). AEA is rapidly synthesized in neurons following depolarization and subsequent Ca²⁺ influx (Dimarzo et al., 1994). 2-AG, similar to AEA, is synthesized in an activity-dependent manner, is ubiquitously found in the CNS and is both a CB₁ and CB₂ receptor agonist; however, the concentration of 2-AG is up to 1000 times that of AEA (Sugiura et al., 1995). In neuronal signaling, endocannabinoids function as retrograde neurotransmitters; they are synthesized and released by a postsynaptic neuron and activate receptors on presynaptic neurons (Wilson and Nicoll, 2001). Deactivation of endocannabinoids occurs through specific enzymatic reactions. Fatty acid amide hydrolase (FAAH) is an intracellular membrane-bound enzyme that degrades fatty acid amides and it is responsible for inactivating AEA by catalyzing its breakdown to arachidonic acid (AA) and ethanolamine (Cravatt et al., 1996). Deactivation of 2-AG is primarily achieved by the enzyme monoacylglycerol lipase again producing AA (Dinh et al., 2002).

In addition to endogenous cannabinoid receptor ligands, other classes of cannabinoids have been identified. The identification of Cannabis plant-derived cannabinoids, or phytocannabinoids, including cannabinol, cannabidiol (CBD) and the main psychoactive component of the plant Δ⁹-tetrahydrocannabinol (THC), preceded the discovery of endocannabinoids by several decades (Mechoulam et al., 2014). To date, it has been suggested that there is over 100 phytocannabinoids and novel cannabinoids continue to be isolated from the C. sativa plant (Radwan et al., 2009). Moreover, many synthetic agonists, inverse agonists and antagonists of the cannabinoid receptors have been produced. The synthetic cannabinoids HU-210 and R-(+)-WIN55212 show a high affinity for both the CB₁ and CB₂ receptor (Rinaldi-Carmona et al., 1994), while selective agonists have also been identified including the CB₁ selective agonist arachidonyl-2′-chloroethylamide (ACEA) (Hillard et al., 1999) and the CB₂ selective agonist JWH-133 (Huffman et al., 1999). Other synthetic ligands that bind cannbinoid receptors but evoke inhibitory effects include SR141716A and SR144528 which exert CB₁ and CB₂ selectivity respectively (Rinaldi-Carmona et al., 1994; 1998,), as well as the high-affinity CB₁ ligand AM251 (Gatley et al., 1996) and the high-affinity CB₂ ligand AM630 (Ross et al., 1999). Several lines of evidence suggest that theses ligands not only result in receptor antagonism but also inverse agonism (Pertwee, 2005).

In vivo effect of cannabinoids on adult neurogenesis

In addition to the various neurogenesis regulators discussed earlier, there is considerable evidence to suggest that both exogenous and endogenous cannabinoids can control cell genesis in the adult brain, although the effects can vary considerably according to the cannabinoid, dose and duration of administration (see Table 2014). What appears to be a common characteristic of both synthetic (Mackowiak et al., 2007) and plant-derived (Kochman et al., 2006) cannabinoids is that acute administration has no effect on cell proliferation or overall neurogenesis in the hippocampus; however, chronic administration of exogenous cannabinoids has been shown to affect the process. For example, chronic treatment with the potent synthetic cannabinoid HU-210, a drug that has a high affinity for both CB₁ and CB₂ receptors, enhances both proliferation and survival of cells in the rat dentate gyrus (Jiang et al., 2005). Similarly, chronic administration of the CB₂ selective agonist HU-308 also exhibits proliferative-enhancing affects (Palazuelos et al., 2012), raising the possibility that these effects may be mediated, at least in part, by CB₂ receptor signalling. This is supported by evidence that a number of BrdU⁺ cells in the dentate gyrus are reduced in CB₂-deficient mice (Palazuelos et al., 2006). In contrast to this, chronic administration of another synthetic CB₁/CB₂ agonist WIN55,212-2 to rats during adulthood was found to have no effect on the number of immature neurons in the dentate gyrus, however, interestingly, administration during adolescence decreased the number of immature neurons, an affect that is attributed to selective suppression of dorsal but not ventral hippocampal neurogenesis (Abboussi et al., 2014). Further contrasting effects are observed in the aged brain where WIN55,212-2 administration partially restored age-related deficits in hippocampal neurogenesis in rats (Marchalant et al., 2009), suggesting a unique temporal role for cannabinoid receptors in the regulation of neurogenesis throughout the lifespan. The effects of the phytocannabinoid Δ⁹-THC appear to be dose- and/or time-dependent; 3 weeks of oral administration of a weekly escalating dose of Δ⁹-THC was found to have no effect on cell proliferation in the mouse dentate gyrus (Kochman et al., 2006), whereas, 6 weeks of oral administration of a static dose of Δ⁹-THC has been shown to decrease cell proliferation without having an effect on overall neurogenesis in mice (Wolf et al., 2010). Interestingly, the study by Wolf et al. (2010) found that chronic administration of another phytocannabinoid CBD also decreased proliferation but, strikingly, and perhaps appearing somewhat counterintuitive, is that CBD induced a substantial increase in net neurogenesis by a CB₁ receptor-dependent mechanism (Wolf et al., 2010). These data are supported by evidence that repeated administration of CBD to wild-type mice increases hippocampal NPC proliferation via CB₁ receptors, which may underlie the anxiolytic effect of CBD in chronically stressed animals (Campos et al., 2013).

Table 1.

Literature assessing the in vivo effects of cannabinoids in neurogenesis

Treatment	Measurement	Observation	Reference
HU-210	Cell proliferation in the dentate gyrus in adult rats	Enhanced	Jiang et al. (2005)
HU-308	Hippocampal progenitor proliferation in adult mice	Enhanced	Palazuelos et al. (2012)
WIN55,212-2	Dorsal hippocampal neurogenesis during adolescence	Reduced	Abboussi et al. (2014)
WIN55,212-2	Age-related deficits in hippocampal neurogenesis	Partial restoration	Marchalant et al. (2009)
Δ9-THC/CBD	Precursor cell proliferation in the dentate gyrus	Reduced	Wolf et al. (2010)
CBD	Cell survival in the dentate gyrus	Enhanced	Wolf et al. (2010)
CBD	Number of BrdU+ cells colocalized with NeuN+ cells in hippocampus	Enhanced	Campos et al. (2013)
DAGL inhibitor	Cell proliferation in the adult SVZ	Reduced	Goncalves et al. (2008)
URB597/AEA/WIN55,212-2	Adult hippocampal NPC proliferation	Enhanced	Aguado et al. (2005)
WIN55,212-2/JWH-133/URB597	Progenitor cell proliferation in the SVZ	Enhanced	Goncalves et al. (2008)
AM251	Cell proliferation in the SGZ	Enhanced	Hill et al. (2006)
AM251	Cell proliferation in the SGZ	Enhanced at 24 h/ reduced at 48 h	Wolf et al. (2010)
FAAH−/−	Cell proliferation in the dentate gyrus of adult mice	Enhanced	Aguado et al. (2005)
DAGLα−/−	Cell proliferation and number of DCX+ neurons in the hippocampus	Reduced	Gao et al. (2010)
DAGLβ−/−	Cell proliferation in the hippocampus	Reduced	Gao et al. (2010)
CB1−/−	Cell proliferation in the dentate gyrus and SVZ	Reduced	Jin et al. (2004)
			Kim et al. (2006)
CB1−/−	Number of BrdU+ cells colocalized with S100β+ cells in the SGZ and granule cell layer of the dentate gyrus	Reduced	Aguado et al. (2006)
CB1−/−	Number of BrdU+ cells colocalized with NeuN+ cells in the SGZ and granule cell layer of the dentate gyrus	Enhanced	Aguado et al. (2006)
CB1−/−	Kainic acid-induced hippocampal NPC proliferation	Reduced	Aguado et al. (2007)
CB1−/−	Cortical thickness	Reduced at P2	Diaz-Alonso et al. (2012)
SR141716A	Cell proliferation in the SVZ	Enhanced	Jin et al. (2004)
JTE-907/AM630	Cell proliferation in the SVZ	Reduced	Goncalves et al. (2008)
CB2−/−	Number of BrdU+ cells in dentate gyrus	Reduced	Palazuelos et al. (2006)

JTE-907 and AM630 are CB₂ receptor antagonists. NeuN, neuronal nuclei.

The CB₁ receptor inverse agonist AM251 is often used to oppose the effects of endocannabinoids at the receptor and acute administration of this drug increases cell proliferation in the SGZ 24 h post-treatment (Hill et al., 2006; Wolf et al., 2010). However, this increase reverts to a decrease from 48 h onwards (Wolf et al., 2010), again suggesting a complex temporal role for cannabinoid signalling in NSC fate. Chronically, the same inverse agonist was found to have no effect (Rivera et al., 2011); however, it has been shown to block the proliferative-enhancing effects of aerobic exercise (Hill et al., 2010). This raises the possibility that endocannabinoid signalling via the CB₁ receptor may not be important for basal regulation of NPCs, but rather is essential for mediating the effects of exercise, which is a well-established, potent neurogenesis stimulator (van Praag, 2009). Another drug used to inhibit endocannabinoid activity, the CB₁ and transient receptor potential cation channel subfamily V member 1 (TRPV1) antagonist SR141716A, has been shown to increase cell proliferation in the dentate gyrus and the lateral ventricles of mice (Jin et al., 2004). This effect was observed in both wild-type and CB₁, but not TRPV1, knockout mice. Furthermore, Aguado et al. (2006) have observed reduced astrogliogenesis and increased neurogenesis in CB₁-deficient mice (Aguado et al., 2006). These findings illustrate that multiple receptors are responsible for the effects of cannabinoids on neurogenesis, which may account for the complexity of the results observed.

Studies utilizing gene knockdown technology to limit the activity of the endocannabinoid system have provided compelling evidence linking cannabinoids and neurogenesis in the adult brain. Knockdown of the enzyme responsible for AEA hydrolysis, FAAH, increases cell proliferation in the dentate gyrus of adult mice (Aguado et al., 2005), while Goncalves et al. (2008) have demonstrated that chronic inhibition of the enzyme responsible for the production of 2-AG almost completely abolished cell proliferation in the mouse SVZ, while inhibiting FAAH also increased neurogenesis (Goncalves et al., 2008). These findings illustrate the importance of basal endocannabinoid tone in maintaining neurogenesis. Elsewhere, complete knockdown of the α subtype of the DAG lipase α (DAGLα) enzyme reduces brain 2-AG and AEA levels by approximately 80% and 40%, respectively, and furthermore leads to a decrease in cell proliferation rate and a 50% reduction in immature DCX positive neurons in the mouse hippocampus (Gao et al., 2010). The same study shows that a reduction in central 2-AG alone can also interfere with neurogenesis; knockdown of the DAGLβ subtype reduces 2-AG levels in the brain without significantly affecting AEA and results in a decrease in cell proliferation in the hippocampus. Further evidence supporting a role for endocannabinoid signalling in adult hippocampal neurogenesis can be found in studies involving cannabinoid receptor knockout animals; a CB₁^−/− genotype is accompanied by a 50% decrease in proliferating cells in the dentate gyrus (Jin et al., 2004; Kim et al., 2006). Furthermore, Aguado and colleagues (2007) have demonstrated that kainic acid-induced hippocampal NPC proliferation is attenuated in CB₁^−/− mice, indicating the role of CB₁ in neurogenesis induced by excitotoxicity. Intricate data from the same group indicates that CB₁^−/− mice have reduced cortical thickness at postnatal day 2, indicating the integral role of CB₁ receptors in controlling the specification of upper- and deep-layer cortical neurons (Diaz-Alonso et al., 2012). Finally, CB₂^−/− animals also exhibit a decreased proliferation rate illustrating the importance of both the CB₁ and CB₂ receptors (Palazuelos et al., 2006). Taken together, these studies suggest that the endocannabinoid system, acting via multiple complex mechanisms, is a key player in the regulation of adult neurogenesis in vivo.

In vitro effect of cannabinoids on adult neurogenesis

It is known that NPCs (Aguado et al., 2005) express a functional endocannabinoid system and are targeted by cannabinoids to promote neurosphere generation and NPC proliferation (see Table 2000). In addition, endocannabinoids are central in regulating neural differentiation and migration. Indeed, in embryonic murine precursors derived from the cortex, AEA enhances cell differentiation towards a neuronal lineage via a CB₁-dependent mechanism (Compagnucci et al., 2013). Furthermore, using freshly dissected RMS tissue from the postnatal brain, Oudin et al. (2011) have shown that endocannabinoid tone is central in controlling neuroblast migration from RMS explants (Oudin et al., 2011). Elsewhere, Butti et al. (2012) demonstrate that SVZ adult mouse NPCs are producers of AEA and that AEA regulates spontaneous EPSCs in medium spiny neurons (Butti et al., 2012). Furthermore, the synthetic cannabinoid WIN-55,212-2, in addition to the selective FAAH inhibitor, URB597, have been shown to promote neurosphere generation, while WIN-55,212-2, URB597 and endocannabinoids (both AEA and 2-AG) increase the number of BrdU⁺ NPCs from dissociated neurospheres (Aguado et al., 2005). In further experiments from this group using postnatal rat cortical neural progenitors, WIN-55,212-2, URB597, AEA and 2-AG increased the number of GFAP⁺ cells with a concomitant decrease in β-tubulin III⁺ cells after differentiation for 2 days, indicating the progliogenic action of synthetic and endogenous cannabinoids during the differentiation process (Aguado et al., 2006). Elsewhere, the CB₂ specific agonist AM1241 has been shown to promote the proliferation/differentiation of human NSCs in the presence of the HIV-1 glycoprotein Gp120, and furthermore, AM1241 prevents DNA fragmentation induced by administration of Gp120, which suggests a neuroprotective role of CB₂ receptors against impaired neurogenesis, with relevance to the cognitive deficits seen in HIV-1 patients (Avraham et al., 2014). Indeed, CB₂ knockout reduces the self-renewal (as determined by neurosphere generation in vitro) of murine embryonic cortical NPCs (Palazuelos et al., 2006), while both HU-308 and JWH-133 increase both primary neurosphere generation and neural progenitor self-renewal in vitro (Palazuelos et al., 2006). Rubio-Araiz et al. demonstrated that both CB₁ (ACEA) and CB₂ (JWH-056) agonists stimulate the proliferation of primary murine cortical neurospheres (Rubio-Araiz et al., 2008) and recently it has also been demonstrated that hemopressin (a CB₁ inverse agonist) promotes oligodendroglial differentiation within SVZ NSC/NPC cultures derived from neonatal mice (Xapelli et al., 2014). In support of this, the CB₁ receptor agonist ACEA promotes murine neural precursor differentiation via CB₁, with the CB₂ receptor agonist JWH-133 being ineffective (Compagnucci et al., 2013).

Table 2.

Literature assessing the in vitro effects of cannabinoids in neurogenesis

Treatment	Measurement	Observation	Reference
HU-210/AEA	Proliferation of embryonic hippocampal NPCs/NSCs	Enhanced	Jiang et al. (2005)
HU-308	Proliferation of HiB5 NPCs	Enhanced	Palazuelos et al. (2012)
HU-308	Proliferation of cortical progenitors in organotypic cultures	Enhanced	Palazuelos et al. (2012)
AEA/ACEA	Differentiation of embryonic murine neural precursors derived from the cortex towards neural lineage	Enhanced	Compagnucci et al. (2013)
ACEA/JWH-133	Migration of Cor-1 NSC line	Enhanced	Oudin et al. (2011)
AM251/JTE-907/DAGL inhibitors	RMS neuroblast migration	Reduced	Oudin et al.(2011)
ACEA/JWH-133	RMS neuroblast migration	Enhanced	Oudin et al. (2011)
ACEA/JWH-056	Proliferation of neurospheres	Enhanced	Rubio-Araiz et al. (2008)
WIN-55,212-2/URB597	Neurosphere generation	Enhanced	Aguado et al. (2005)
WIN-55,212-2/URB597/ AEA/2-AG	Number of BrdU+ NPCs from dissociated neurospheres	Enhanced	Aguado et al. (2005)
WIN-55,212-2/URB597/ AEA/2-AG	Number of GFAP+ cells after differentiation of postnatal NPCs for 2 days	Enhanced	Aguado et al. (2006)
WIN-55,212-2/URB597/ AEA/2-AG	Number of β-tubulin III+ cells after differentiation of postnatal NPCs for 2 days	Decreased	Aguado et al. (2006)
AM1241	Proliferation/differentiation of human NSCs in presence of Gp120	Enhanced	Avraham et al. (2014)
CB2−/−	Neurosphere generation of murine embryonic cortical NPCs	Reduced	Palazuelos et al. (2006)
HU-308/JWH-133	Primary neurosphere generation and NPC self-renewal	Increased	Palazuelos et al. (2006)
Hemopressin	Oligodendroglial differentiation within SVZ NPC/NSC cultures	Increased	Xapelli et al. (2014)

Hemopressin is a CB₁ inverse agonist.

Mechanisms of cannabinoid-induced regulation of intrinsic/extrinsic signalling in adult neurogenesis

The cellular signalling events orchestrated by cannabinoids in NPCs continue to be elucidated, with particular roles for ERK, PI3K and Akt pathways suggested (see Figure 1). In particular, CB₂ couples to the ERK and PI3K/Akt cascades (Palazuelos et al., 2006; 2012; Molina-Holgado et al., 2007) and the CB₂ agonist HU-308 promotes the proliferation of NPCs via ERK and PI3K/Akt signalling (Palazuelos et al., 2006). In support of this, HU-308 is a robust activator of the PI3K/Akt pathway in the HiB5 hippocampal progenitor cell line (Palazuelos et al., 2012). Interestingly, mammalian target of rapamycin complex 1 (mTORC1) signalling is a target of the PI3K/Akt pathway and hence is central in neural cell survival/death decision; mTORC signalling also contributes to CB₂-regulated NPC proliferation. Indeed, HU-308 induces cell proliferation in both embryonic organotypic cortical slices and in adult hippocampal NPCs via an mTORC1-dependent mechanism (Palazuelos et al., 2012). Elsewhere, both CB₁ (ACEA) and CB₂ (JWH-056) agonists have been shown to stimulate the proliferation of mouse neural precursor cells via PI3K/Akt pathways (Molina-Holgado et al., 2007) and TNF-α signalling mechanisms (Rubio-Araiz et al., 2008). Both the synthetic cannabinoid HU-210 and AEA promote the proliferation of cultured embryonic hippocampal NPCs in a concentration-dependent manner involving G_i/o proteins and the ERK signalling pathways (Jiang et al., 2005). Further in vitro evidence indicates that ACEA enhances murine neural precursor differentiation to neurons by targeting ERK signalling (Compagnucci et al., 2013). In addition, ACEA reduces ERK phosphorylation in neural precursor cells and this reduction promotes neuronal differentiation. Using neurogenesis and PCR arrays, Compagnucci et al. (2013) recently demonstrated that CB₁ activation promotes the expression of genes involved in neuronal maturation and commitment to a neuronal lineage (Compagnucci et al., 2013). In contrast, the endogenous cannabinoid AEA has been shown to inhibit cortical neuron progenitor differentiation to mature neuronal phenotype, decrease the proliferation of primary postnatal murine NPCs (Soltys et al., 2010) and inhibit the differentiation of the human NSC line, HNSC.100 (Rueda et al., 2002). These events are CB₁ receptor-dependent and as AEA inhibits NGF-induced ERK activation in PC12 cells via CB₁ receptors, this suggests that AEA inhibits NPC differentiation through attenuation of the ERK pathway (Rueda et al., 2002).

Figure 1.

Endocannabinoid signalling regulates NPCs in the adult brain. Endocannabinoids acting in an autocrine and paracrine fashion may activate CB₁ and/or CB₂ receptors. CB₁ and CB₂ activity can induce both PI3K/Akt/mTORC and MEK/MAPK/CREB signalling pathways that influence cell proliferation, differentiation and survival, while also promoting integration of immature neurons into existing circuitry. In addition, CREB can induce transcription of BDNF that can directly influence cell fate and may also increase CB₁ expression and endocannbinoid production, possibly leading to positive feedback within the signalling system.

Further data elsewhere indicate that signalling involving CREB transcription factor may govern cannabinoid-induced regulation of NPCs. Indeed, exposure of murine NPCs to AEA promotes glial and neuronal differentiation, with a possible role for CREB (Soltys et al., 2010). Much data indicate that CREB is a cannabinoid target, with recent evidence indicating that CB₂ agonists target CREB signalling in the rat cortex after subarachnoid haemorrhage (Fujii et al., 2014) and cerebral ischaemia (Choi et al., 2013). In support of this, THC (Casu et al., 2005) and AEA (Isokawa, 2009) administration has been shown to regulate the expression of phosphorylated CREB in the rat cerebellum and hippocampus, respectively, while the CB₂ receptor agonist, trans-caryophyllene, promotes the phosphorylation of neural CREB (Choi et al., 2013).

The Sox2 gene family regulate NSC proliferation in the hippocampus and recent evidence indicates that CB₁ receptor activation enhances the number of Sox2⁺ cells via Notch signalling in cultured mouse SVZ cells, suggesting that CB₁ receptor activation promotes the self-renewal of SVZ cultures (Xapelli et al., 2013). Cannabinoids also regulate the expression of the T-box transcription factor, Tbr, which may be central in mediating the neurogenic effects of cannabinoids. Indeed, Saez et al. (2014) has recently demonstrated that prenatal exposure of rats to WIN-55,212-2 differentially regulates the number of glutamatergic intermediate progenitors (Tbr2⁺) and post-mitotic neurons (Tbr1⁺) during embryonic development in the cortex (Saez et al., 2014). Interestingly, this indicates that prenatal exposure to WIN-55,212-2 impacts the differentiation of glutamatergic neurons in the developing cerebral cortex. In support of this, data from CB₁-deficient murine embryos indicate that there is a decrease in Tbr2⁺ cells in the SVZ (Diaz-Alonso et al., 2014) while Tbr1⁺ post-mitotic cells accumulate abnormally during embryogenesis in deep bins of the cortical plate of CB₁-deficient mice when compared with wild-type littermates (Diaz-Alonso et al., 2012).

Neurotrophic factors are strongly linked to adult neurogenesis and recent evidence suggests that there is functional interplay between BDNF and CB₁ receptors in the brain (De Chiara et al., 2010). In support of this, Maison et al. (2009) demonstrated that BDNF increases the expression of CB₁ receptors in rat cultured cerebellar granule neurons (Maison et al., 2009), while BDNF can also promote the production of cortical endocannabinoids (Lemtiri-Chlieh and Levine, 2010). In human studies, D'Souza et al. (2009) demonstrated that i.v. administration of THC enhances the expression of peripheral BDNF in serum (D'Souza et al., 2009) and this is supported by evidence that CB₂ receptor stimulation promotes BDNF expression in rat neurons (Choi et al., 2013). Recent evidence also suggests that CB₁ receptors can cross-talk with NGF signalling in adult mouse dorsal root ganglion neurons (Wang et al., 2014). In addition, intricate new data from Keimpema et al. (2013) indicate that NGF affects endocannabinoid signalling to promote cholinergic differentiation in mice (Keimpema et al., 2013).

A body of literature indicates that signalling involving adenosine, PKC, growth factors and IL-1 receptors may govern cannabinoid-induced regulation of NPCs. Indeed, using adult neural precursor cells prepared from the whole brains of 8-week-old mice, Shinjoy and Di Marzo (2013) recently demonstrated that the major non-THC phytocannabinoid, cannabichromene (CBC), promotes cell survival during differentiation while blunting cell differentiation into astroglia. The authors suggest the involvement of ERK, ATP and adenosine signalling cascades in mediating the effects of CBC on neural cells (Shinjyo and Di Marzo, 2013). Recent evidence also indicates that cannabinoids can target the actin-bundling protein fascin, which plays a role in the migration of neuroblasts and neural development (Sonego et al., 2013). Indeed, the CB₁ agonist ACEA controls the interaction between fascin and PKC, which indicates that CB₁-dependent signalling may regulate actin-bundling activity, with a subsequent effect on neuroblast migration (Sonego et al., 2013). EGFR signalling is key in controlling NSC survival, and using the Cor-1 NSC line, data from Sutterlin et al. (2013) demonstrate that CB₁ and CB₂ receptors cooperate with EGFR in the regulation of NSC expansion (Sutterlin et al., 2013). Similarly, the CB₁ receptor has been shown to couple activated FGF receptors to axonal growth in rat cerebellar granule neurons (Williams et al., 2003). Finally, Garcia-Ovejero et al. (2013) have demonstrated that both CB₁ and CB₂ receptors are co-expressed with IL-1 receptor, type I and IL-1 receptor, type II in mouse brain neurospheres and both ACEA and JWH-133 affect IL-1 signalling in primary cultures of mouse brain-derived neurospheres, increasing IL-1β, while decreasing IL-1Rα production by neurospheres. This is significant given that IL-1β negatively regulates neurosphere proliferation (Garcia-Ovejero et al., 2013).

Concluding remarks

While much progress has been made in recent decades in understanding the process of adult neurogenesis, the underlying mechanisms have yet to be fully elucidated. As highlighted in this review, the microenvironment clearly determines the rate of proliferation of NSCs and NPCs, their survival and their differentiation into mature neurons that are integrated into functional networks. Endocannabinoids may play pivotal roles in at least some of these phases of neurogenesis. Of particular interest are the varying temporal effects of synthetic, endogenous and plant-derived cannabinoids on the proliferation and survival phases of neurogenesis, indicating complex physiological regulation of this process that may be modulated by drugs that target the endocannabinoid system. The functional importance of neurogenesis has yet to be clarified; however, the weight of evidence indicates that impaired neurogenesis is associated with depression and cognitive impairment. Pharmacological targeting of the cannabinoid system as a regulator of neurogenesis may prove a fruitful strategy in the prevention or treatment of mood or memory disorders.

Glossary

2-AG

2-arachidonoylglycerol

AA

arachidonic acid

ACEA

arachidonyl-2′-chloroethylamide

AEA

anandamide

BDNF

brain-derived neurotrophic factor

BMP

bone morphogenetic protein

BrdU

5-bromo-2'deoxyuridine

CB

cannabinoid receptor

CBC

cannabichromene

CBD

cannabidiol

CREB

cAMP response element-binding protein

DAGL

DAG lipase

DCX

double cortin

FAAH

fatty acid amide hydrolase

GFAP

glial fibrillary acidic protein

IGF-1

insulin-like growth factor-1

mTORC1

mammalian target of rapamycin complex 1

NGF

nerve-growth factor

NPC

neural progenitor cell

NSC

neural stem cell

Ptc1

patched 1

RMS

rostral migratory stream

SGZ

subgranular zone

Shh

Sonic Hedgehog

Smo

smoothened

SVZ

subventricular zone

THC

Δ⁹-tetrahydrocannabinol

TRPV1

transient receptor potential cation channel subfamily V member 1

Conflict of interest

The authors declare that they have no conflict of interest.