As noted above, CB1 receptors are primarily expressed in neurons, while CB2 receptors are primarily expressed in immune cells. A large number of small molecules have been developed to modulate the function of CB1/CB2 receptors, and many of these cross the BBB. In addition, several of these molecules have been evaluated in clinical trials and safety data has been established. One of the major gaps in current treatments for the LSDs is the lack of therapies for the CNS, and therefore repurposing these ECS molecules is a logical approach. Such ECS effector molecules have been shown to reduce inflammation, slow cell death processes, improve mitochondrial, lysosomal and other cell functions, and enhance other metabolic pathways that are commonly defective in the LSDs (Table 1).

As an example, an early paper showed that activation of CB1 receptors with the endocannabinoid AEA or with THC, the main psychoactive compound in cannabis, led to sphingomyelin degradation in cultured astrocytes through activation of a neutral sphingomyelinase [30]. In the LSD acid sphingomyelinase deficient Niemann-Pick disease (i.e., types A and B Niemann-Pick disease, ASMD), sphingomyelin storage is the primary metabolic abnormality [31], suggesting that CB1 activation may be a reasonable approach for the treatment of this disorder. Importantly, although the sphingomyelin storage in ASMD begins in lysosomes, it rapidly extends to the plasma membrane, mitochondria and other cell compartments, making it potentially accessible to the neutral sphingomyelinase activity [32]. Although an ERT is in the final stages of clinical development for this disorder (Olipudase alfa®), it does not impact the CNS disease that occurs in about half of all ASMD patients [33].

We therefore decided to evaluate this approach in a mouse model of type A and B NPD (acid sphingomyelinase deficient mice, ASMKO) [34], and found the unexpected downregulation of CB1 on the surface of neurons from these mice [35]. This was due to entrapment of the receptor within lysosomes, indicating that CB1 signaling was likely abnormal in this disorder and further supporting the activation of CB1 as a therapeutic strategy. Since direct activation of CB1 with synthetic or natural agonists leads to psychotropic effects, we chose an indirect strategy using small molecules that inhibit the enzyme fatty acid amide hydrolase (FAAH), leading to an elevation of endogenous endocannabinoids, including AEA, and several other bioactive lipids (e.g., palmitoylethanolamide [PEA] and oleoylthanolamide [OEA]). AEA primarily acts on CB1 receptors and to a lesser extent CB2, while PEA and OEA act on the PPAR-α receptor. Many FAAH inhibitors have been developed that cross the BBB, and several have entered clinical trials [23].

We first evaluated the ability of three such FAAH inhibitors to reduce sphingomyelin levels in neurons from ASMKO mice. Significant sphingomyelin reduction was observed, which could be prevented using an inhibitor of neutral sphingomyelinase [35]. We then treated ASMKO mice with one FAAH inhibitor, and found improvements in many pathologic markers, including reduction of sphingomyelin in the brain and other tissues. Importantly, several neurological clinical endpoints also were improved, and lifespan was significantly extended as well. A schematic depiction illustrating the putative mechanism of action of FAAH inhibition in ASMD is shown in Fig. 1. We also found that treatment of cells from patients with type C Niemann-Pick disease, due to mutations in the NPC1 gene, led to reduction of sphingomyelin and, importantly, reduction in cholesterol storage. This is consistent with previous work showing that reduction of sphingomyelin in type C Niemann-Pick disease cells using recombinant acid sphingomyelinase led to correction of cholesterol trafficking [36]. While these examples may be specific to diseases with significant sphingomyelin storage, activation of CB1 receptors has many other beneficial effects that could impact all LSDs, including prevention of neurodegeneration, inhibition of pain, remyelination and others.

In contrast to CB1, CB2 receptors are primarily expressed on immune cells both in the periphery and CNS. Therefore, unlike CB1, direct agonists of CB2 receptors do not have major psychotropic effects, and have been safely used in animal models of various diseases. The main effect of these direct CB2 agonists is to reduce inflammatory responses in autoimmune and other inflammatory diseases, including neuroinflammatory diseases, although additional positive effects on prevention of cell death, repair of the BBB, and other cell abnormalities have been reported [20, 21].

To initially evaluate the role of CB2 in the LSDs, we chose mouse models for two diseases: acid ceramidase deficiency (Farber disease) and mucopolysaccharidosis type IIIA (Sanfilippo A; MPS IIIA) [37, 38]. Although the initiating storage material in these two diseases is distinct (the lipid ceramide and the glycosaminoglycan heparan sulfate, respectively), both of these disorders are characterized by early activation of inflammatory pathways in the periphery and CNS. Consistent with this inflammatory disease, CB2 expression was found to be markedly elevated in several tissues from these animals (Fig. 2). Treatment studies using a CB2 agonist are underway, but preliminary findings have already revealed a slowing of disease progression and extension of lifespan in the Farber disease animals, consistent with the reduction of inflammation and other positive effects attributed to these molecules. Thus, CB2 receptors are not only a potential new target for treatment, but also may serve as a biomarker to indicate macrophage infiltration and inflammation in different LSDs.