The goal of this research was to determine the effects of chronic, orally-administered cannabidiol on a variety of phenotypes, including neuroinflammation, neurodegeneration, and spontaneous seizures. The Cln1^−/− mouse model reliably and naturally displays all of these pathological and clinical signs. It is important to note that we used highly purified cannabidiol, not a “whole-plant” extract. Whole-plant extracts typically contain hundreds, if not thousands, of chemicals native to the cannabis sativa plant. We sought to elucidate the individual contribution of the increasingly-popular and widely-available cannabinoid, CBD.

Our LC–MS/MS proteomic data confirmed that putative targets of CBD are present at near normal levels in the Cln1^−/− mouse brain, even near the end stage of disease. This is interesting in light of the fact that CLN1 disease is profoundly neurodegenerative and suggests that neurons expressing CBD receptors may be partially and preferentially spared. This also provided the impetus to proceed with the CBD study. Importantly, chronic oral administration of ~ 100 mg/kg of CBD was well tolerated and had no obvious deleterious effects. This regimen did not lead to behavioral or physical changes that we monitor daily that would be cause for veterinary attention (wasting, hunched posture, scruffy hair, sunken eyes, labored breathing, etc.), and it did not increase markers of pathology in WT mice treated with CBD compared to WT mice treated with vehicle (Supplemental Figure S2). Not unexpectedly, chronic administration of CBD had no effect on the progressive accumulation of SCMAS in the brains of treated Cln1^−/− mice. Cannabidiol is not known to mediate any mechanism that would increase PPT-1 enzyme activity or otherwise decrease SCMAS accumulation. Therefore, it is important to keep in mind that any effects of CBD would be observed in the context of continued disease progression.

Markers of a neuroimmune response, such as astrocytosis (GFAP) and microgliosis (CD68), are dramatically increased in the untreated Cln1^−/− mouse brain and can be observed as early as 1–3 months of age^22,23,25. Here we show a significant decrease in both GFAP and CD68 in areas of both the thalamus and the cortex following chronic CBD treatment. This reduction of cellular markers of disease is remarkable considering the continued disease progression and the fact that the animals were evaluated at a relatively late stage of disease. A growing body of literature reports anti-inflammatory effects of CBD^1,2. In the current study, whole brain homogenates from animals treated with CBD showed neither a decrease in pro-inflammatory nor an increase in anti-inflammatory cytokines/chemokines. It is possible that chronic treatment with CBD might have had a more pronounced effect earlier in the disease process. The apparent discordance between the cellular and humoral responses could be due to the different analytical approaches. Immunofluorescent imaging allows for the sensitive detection of regional differences. In contrast, regional differences in cytokine/chemokine levels could be masked when analyzing whole brain homogenates. We chose to analyze cytokine levels at the whole brain level because, 1) receptors targeted by CBD are spread throughout the brain, 2) neuropathology is widespread by late disease stage in CLN1, and 3) the etiology of seizures in the Cln1^−/− mouse is unknown, so probing only specific brain regions may have been limiting. Alternatively, CBD could have differential effects on the cellular and humoral responses through as yet unknown mechanisms. Future studies will be required to determine the effects of CBD treatment at various time points to determine the temporal nature of the reductions in astrogliosis and microglial activation. In addition, it will be critical to determine if the effects of CBD are enhanced when combined with treatments that target the underlying defect in CLN1 disease, PPT-1 deficiency. It is possible that CBD could synergize with another therapy to reduce both the cellular and humoral immune responses.

One of the hallmarks of CLN1 disease is profound neurodegeneration and cortical atrophy. In fact, at a terminal stage of disease, the mass of an affected child’s brain can be as little as 50% that of an age-matched unaffected child^14,15. Likewise, the Cln1^−/− mouse also has profound neurodegeneration, cortical atrophy and significantly decreased brain mass^27,30. Although there appeared to be a slight increase in cortical thickness with CBD treatment, this did not reach statistical significance. In addition, there was no significant change in neuron counts in either the thalamus or the cortex following chronic CBD treatment. Similar to the neuroimmune response, it will be important to determine the temporal effects on neurodegeneration and whether or not CBD can synergize with therapies designed to increase PPT-1 activity.

Here we report a more comprehensive characterization of the spontaneous epilepsy phenotype in the mouse model of INCL. From our data it is clear that Cln1^−/− mice experience seizures earlier than previously reported^22,29. Seizures clearly develop prior to 7 months of age, and are likely to appear, albeit at a low frequency, even before 6 months of age. This finding tells us what recent pathological data^22,23,31,32 have revealed in abundance: treatment is most likely to be successful when initiated early in the disease process. This supports our strategy of beginning CBD treatment at 1 month of age, which is prior to the development of any measurable pathological or clinical signs. Our findings also uncovered a significant sex difference in seizure frequency in untreated Cln1^−/− mice, with female mice exhibiting significantly more seizures than males during the optimal observation period. A higher proportion of female mice exhibit at least one seizure per bin at an earlier time point than male mice, a difference that persists through the end of life. This is consistent with previous findings in CLN3 disease patients (juvenile Batten disease) where females typically display a more severe trajectory of symptoms compared to males³³. Our findings also align with research in the CLN6 (variant late infantile NCL) mouse in that onset of some behavioral signs were earlier and more severe in females compared to males^34,35. Although it is tempting to extrapolate findings between various forms of CLN disease, this should be viewed cautiously since these diseases can have fundamentally different causes. For example, both CLN3 and CLN6 disease result from defects in transmembrane proteins, whereas CLN1 involves deficiency in a soluble lysosomal enzyme. Regardless, the current findings reveal the merit of further research into the nature of sex differences in CLN1 disease.

It has been shown previously that CBD protects against seizures in multiple mouse models of induced epilepsy. These include models of PTZ-, audiogenic-, and thermal-induced seizures^7–11. We are unaware of the use of CBD to reduce naturally occurring seizures in a murine model. In addition, the overwhelming majority of studies administer CBD over a relatively short period of time (days to weeks). The current study differs significantly from previous work in that the Cln1^−/− mouse naturally develops seizures as a result of increasing disease pathology (i.e., there is no need to artificially induce seizures). Given that CLN1 disease is progressive, and the spontaneous seizures appear later in the disease, we also administered CBD for at least six months. It is clear that chronic, oral administration of CBD alone does not significantly reduce seizure frequency. In addition, treatment with CBD did not delay the onset of seizures in the Cln1^−/− mouse. Given the reduction in astrocytosis and microgliosis, the complete lack of any effects on seizures is perhaps a bit surprising. We previously showed that decreasing these cellular markers in the Cln1^−/− mouse via a small molecule, brain-penetrant anti-inflammatory drug (Minozac) is associated with a small but significant decrease in seizure frequency²⁶. One explanation for this apparent discrepancy is that Minozac was specifically designed as an anti-inflammatory agent and is likely more potent than CBD.

Anecdotally, caregivers of patients with infantile Batten disease report anti-seizure effects of cannabis products. This could be due to what is commonly referred to as the “entourage effect”, which hypothesizes that phytocannabinoids work best in the context of terpenes and flavonoids that occur naturally in the cannabis sativa plant. It is possible that a “whole-plant” cannabis product would provide greater benefit to pathological markers of disease, or seizure frequency, compared to CBD alone. Unfortunately, due to the relatively crude nature of such products it is virtually impossible to determine the specific effects of CBD, even if it is one of the major components of these preparations. Further, as is the case with many human conditions that involve epilepsy, patients are often taking a regimen of 3–5 anti-seizure drugs prior to the addition of a cannabis product^3–5. This makes it even more difficult to determine any positive effects attributable solely to CBD or any other phytocannabinoid. It is certainly possible that CBD could enhance the effects of other anti-seizure drugs or vice versa. It has been reported that CBD can increase the concentration of active metabolites of other anti-seizure medications. The most robust evidence comes from CBD-clobazam interactions³⁶.

Where previous research has shown various protective effects of CBD at lower doses than used here^1,37–40, the lack of an effect on seizure occurrence reported in the murine model of CLN1 disease may be explained by some key differences. First, route of administration affects brain and plasma concentrations of CBD greatly; i.p. delivery leads to a faster and greater concentration of drug in the brain compared to oral delivery¹⁷. As noted above, repeated i.p. delivery negatively affects Cln1^−/− mice, as would repeated oral gavage, so the voluntary oral dosing was the most practical and clinically relevant route of administration. Still, it is clear that we delivered measurable drug to the brain, which peaks at 6 h following oral administration¹⁷. Another important consideration is the use of induced states of epilepsy/inflammation/pain in the literature compared to the spontaneous seizures that develop in Cln1^−/− mice. This murine model displays profound neurodegeneration that progresses rapidly, leading to naturally-occurring seizures that may simply be more difficult to treat compared to seizure models that must be induced via external intervention. Finally, we tested only one dose of CBD in the current study; evaluation of various doses used in previous work^1,38,39 may yield different results.

Taken together, we provide evidence for positive effects of chronic CBD on the neuroimmune response in the mouse model of CLN1 disease. Importantly, this was observed using CBD in the presence of ongoing disease and in the absence of any adjunct anti-inflammatory, neuroprotective, or anti-epileptic treatments. Although 100 mg/kg of CBD alone is not sufficient to prevent or reduce the naturally occurring seizures in the Cln1^−/− mouse, it might enhance the effects of other anti-seizure drugs or treatment strategies designed to target the fundamental genetic defect. Finally, it is possible that a higher dose of CBD may be necessary to achieve positive anti-seizure benefits. The dose (~ 100 mg/kg) of CBD administered in the current study is within the range reported in the literature(1-600 mg/kg)^{10,11,40–42}, but it is possible that an even higher dose is required to affect seizures in the context of such profound disease. Alternatively, a lower dose may produce more desirable results, especially should it be used in combination with other anti-seizure drugs as in the case of clinical trials for Dravet Syndrome and Lennox-Gastaut Syndrome. Given the thorough characterization of the Cln1^−/−mouse seizure phenotype and the ability to chronically and non-invasively administer CBD, these questions can now be addressed directly.