Calcium hypothesis of neurodegeneration – an update

A couple of years ago it was my great privilege to co-edit a special issue of BBRC together with Maria Catia Sorgato and Ernesto Carafoli (Volume 483, Issue 4, pages 979–1194, published 19^th of February, 2017). The focus on this issue was on neurodegeneration, an area of huge interest and importance. Neurodegenerative disease field is also very wide, with multiple lines of research and directions taken by many excellent laboratories around the world. It is obviously impossible to cover all these ideas at sufficient depth in one issue. So, we decided to focus on 3 major themes, which are most close to our own research and interests (and probably because of that also seem to us to be most important). The themes we selected were calcium/synaptic dysregulation (Ilya), protein misfolding/prion-like mechanisms (Maria), and mitochondrial dysfunction (Ernesto). Of course, these is a lot of crosstalk within these themes, as mitochondrial dysfunction and Ca²⁺ dysregulation go hand-by-hand in causing protein misfolding, and protein misfolding impacts both Ca²⁺ signaling and mitochondrial function. When we solicited the reviews, we attempted to cover multiple neurodegenerative diseases so we can compare how these 3 different themes play out in the context of brain aging, Alzheimer’s disease (AD), Parkinson’s disease (PD), amyotrophic lateral sclerosis (ALS), Huntington’s disease (HD), prion disease and ataxias. An idea was to highlight common deficits that are shared by these disorders and also emphasize their unique features.

We were lucky to engage leaders in the field of neurodegeneration, and all reviews from this issue are very well cited since publication. One review that received particular attention and was very highly cited is a review by James Surmeier and his colleagues entitled “Calcium and Parkinson’s disease” [1]. In this review article the authors make very strong and logical case that neuronal calcium (Ca²⁺) plays a critical role in pathogenesis of PD, and focus on impact of one particular Ca²⁺ entry mechanism mediated by specific L-type voltage gated Ca²⁺ channels. First of all, they argued that selective vulnerability of substantia nigra pars compacta (SNc) neurons in PD may not be related to dopamine toxicity (as generally assumed), but rather related to the massive Ca²⁺ influx into these neurons during pacemaking activity that is attributable to their robust expression of Ca_v1.3 L-type voltage gated Ca²⁺ channels. Ca_v1.3 channels are a distinctive subset of Cav1 channels in that they open at relatively hyperpolarized membrane potentials (below spike threshold). Because SNc dopaminergic neurons are pacemaking, they spend most of their time in a membrane potential range where Cav1.3 channels can open, leading to persistently high intracellular Ca²⁺ concentrations. In this review (and in subsequent papers), Jim Surmeier and his colleagues argue that this influx of Ca²⁺ serves to drive mitochondrial respiration to meet the bioenergetic needs of SNc dopaminergic neurons. The downside of this design is that it results in sustained mitochondrial oxidant stress, which over decades leads to mitochondrial damage and the death of SNc dopaminergic neurons. Interestingly, this hypothesis also helps to explain why loss of function mutations that compromise mitochondrial oxidant defenses or turnover are associated with familial forms of PD.

In support of these ideas they discussed studies that demonstrated that Cav1 (L-type) channel inhibitors had beneficial effects in some animal models of PD. It was demonstrated in these studies that isradipine and nimodipine (both are inhibitors of Ca_v1 channels) attenuated the death of SNc neurons induced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) or 6-hydroxydopamine in rodent models. An additional, and perhaps the strongest, argument was based on epidemiological studies. They cited several retrospective analysis reports that consistently demonstrated that the patients taking dyhidropiridines (inhibitors of Ca_v1 channels) to control hypertension have up to 30% reduction in occurrence of PD. This effect was observed only for centrally penetrant dyhidropiridines and was not observed when hypertension was treated with other medications such as beta-adrenergic blockers.

“Ca²⁺ hypothesis of neurodegeneration” was initially proposed by Zaven Khachaturian for AD 30 years ago [2], and since then, pathological role of several distinct Ca signaling pathways in neurodegeneration had been proposed and applied by several different groups to aging, AD, PD and other disorders. Some of these mechanisms were discussed in review articles in the same special issue of BBRC [3–10], but until recently, none of them have been formally tested in Phase 3 trial in humans. The only previous attempt was the phase 2 AD trial (183 participants) of L-type Ca²⁺ channel inhibitor MEM-1003 (derivative of nimodipine) conducted by Memory Pharmaceuticals (). This study was completed in October 2007 with negative results and was never published.

The arguments for the pathological role of Ca_v1.3-channel-mediated Ca²⁺ entry in PD that were summarized by Jim Surmeier and his colleagues in their review article [1] appear to be strong enough to convince National Institute of Neurological Disease and Stroke (NINDS) to fund a clinical trial in order to determine if treatment with isradipine could slow disease progression in early-stage PD patients (). It was a multi-center trial and the lead investigators of the trial were Tanya Simuni and Kevin Biglan. The trial was called STEADY-PD III and it was a 36-month, Phase 3, placebo-controlled study assessing the effectiveness of an immediate release format of isradipine (two daily 5 mg doses) in 336 participants with early PD who were not receiving dopaminergic therapy. The results of the STEADY-PD III were eagerly anticipated, as there was hope that approaching PD from a new direction may yield better outcome and provide positive guidance for the neurodegenerative disease field in general.

STEADY-PD III was completed in November 20, 2018 and results of the trial have not yet been published. However, a principle investigator of this trial Tanya Simuni presented initial analysis results at American Academy of Neurology Annual Meeting on May 7, 2019 in Philadelphia. I was not in attendance at this conference, but Tanya Simuni kindly provided me with the abstract of her presentation. Unified Parkinson Disease Rating Scale (UPDRS) was used as a primary outcome for this trial. The changes in UPDRS score over 36 months were 2.99 points (isradipine) and 3.26 points (placebo) with treatment effect of 0.27 points (p = 0.85). Therefore, the statistical significance was not reached on UPDRS measure. None of the secondary outcome measures reached statistical significance either. So, the results of this trial targeting CaV1.3 channels were negative.

What may be the reasons for these negative results? One possibility is that doses of isradipine used in the trial (immediate release formulation, 5 mg, twice a day) was not sufficient to substantially inhibit Ca_v1.3 channels in the brain. The dose was selected based on the maximal tolerated dose in the previously completed Phase II study. There was no obvious biomarker for “target engagement” in this trial, and it is possible that the dose of isradipine was too low. Testing another, more potent inhibitor of Ca_v1.3 channels may alleviate this problem in the future. In addition, the reliance upon an immediate release formulation of isradipine, rather than the controlled release format, may have led to long ‘troughs’ in the brain concentration of isradipine where there was essentially no inhibition of Cav1 channels. Another possibility is that the trial was conducted too late, and while inhibition of Ca_v1 channels may help to prevent the disease, it may exert much weaker effect on its progression once it is started. This interpretation is consistent with retrospective clinical data that demonstrated that patients taking dihydropiridines for hypertension have substantially reduced risk of PD. These patients typically are started on medication decades before the average age of diagnosis with PD (~ 60 years old). Thus, it is possible that Cav1.3 channel-mediated Ca influx may play an important role in initiating the disease, but once pathogenic processes start it is no longer a primary driver of the pathology.

Does the negative results obtained in the first trail mean that Ca_v1.3 channels are not a viable target for PD therapy? My opinion is that it is premature to draw a final conclusions based on a single trial. And negative results of the trail targeting Ca_v1.3 channels also should not undermine a more broad “Ca²⁺ hypothesis of neurodegeneration”, which involves additional Ca²⁺ signaling pathways. After all, multiple phase 3 trials were conducted in AD based on “amyloid hypothesis” [11–13] and some more trials being conducted currently despite the fact that all of them were negative so far. Potential reasons used to explain failure of these trials are very similar to the arguments regarding STEADY-PD III trial of isradipine. In my opinion supporters of “Ca²⁺ hypothesis of neurodegeneration” should take the same “learning by failing” approach as proponents of “amyloid hypothesis” do [12] and proceed with future trials using more potent and selective Ca_v1.3 channel inhibitors [14], as well as targeting other relevant Ca²⁺ targets such as ryanodine receptors [15], store-operated Ca²⁺ influx channels [16, 17] and calcineurin [18, 19]. Also, selecting patients at earliest possible stage of the disease (for example by using biomarkers) may increase chance of success in future trials of drugs against Ca²⁺-related targets. It is also possible that single target approach to prevent neurodegeneration may not be sufficiently effective, and a combination of agents targeting different biological pathways (for example distinct Ca²⁺ and dopamine signaling for PD or Ca²⁺ signaling and amyloid pathway for AD) will be eventually necessary to achieve meaningful clinical benefit in AD, PD and other neurodegenerative disorders.