**3. Effects of diazoxide in multiple sclerosis**

#### **3.1. KATP channels and neuroinflammation**

Adenosine triphosphate (ATP)-dependent potassium (KATP) channels play important roles in many cellular functions by coupling cell metabolism to electrical activity. First detected in cardiac myocytes, they are also expressed in a wide number of cell types such as pancreatic β-cells, skeletal and smooth myocytes, neurons and microglia. In these cells, KATP channels regulate potassium fluxes across the cell membrane when glucose is available in sufficient conditions, which couple the electrical activity of the cell to energy metabolism [61, 62]. An increase in the cellular ATP concentration leads to channel closure and membrane depolari‐ zation. On the contrary, metabolic inhibition opens KATP channels and suppresses electrical activity.

Functional KATP channels in the cell membrane are assembled as a heterooctameric complex [63] from two structurally distinct subunits: the regulatory sulphonylurea receptor (SUR) and the pore-forming inwardly rectifying potassium channel (Kir) subunit 6.1 or 6.2. While ATP inhibits the KATP channel by directly binding to the cytoplasmic Kir6 domains, activators, such as potassium channel openers (KCOs), and inhibitors, such as sulphonylurea drugs [64], bind SUR to modulate the channel. Similar KATP channels have also been described in the mito‐ chondria, located on the inner membrane of these organelles where they play a crucial role in the maintenance of mitochondrial homeostasis and function [65].

In recent years, KATP channels have attracted increasing interest as targets for drug develop‐ ment. Their pivotal role in a plethora of physiological processes has been underscored by recent discoveries linking potassium channel mutations to various diseases. The second generation of KCOs with an improved in vitro or in vivo selectivity has broadened the chemical diversity of KATP channel ligands. On the other hand, considering the unique role that KATP channels play in the maintenance of cellular homeostasis, KCOs add their potential in promoting protection against metabolic stress to the already existing pharmacotherapy. Studies in animal models showed KCO effects in several pathological situations such as hypertension, cardiac ischaemia or asthma, which indicates a broad therapeutic potential for KCOs. For example, in the smooth muscle of blood vessels and pancreatic β-cells, diazoxide (7-chloro-3-methyl-4H-1,2,4 benzothiadiazine 1,1-dioxide) binds with similar affinities to SUR1 and SUR2B subunits of KATP channels, increases membrane permeability to potassium ions and induces hyperpolari‐ zation [64]. In these cells, diazoxide-induced hyperpolarization inhibits the opening of voltagegated calcium channels, and results in vasorelaxation and inhibition of insulin secretion [66]. As a consequence, diazoxide has been used since the 1970s for treating malignant hypertension and hypoglycaemia in the United States, Canada, and most European countries [67].

κB pathway and glial reactivity, this indirect neuroprotection is also of main interest for the

Several epidemiologic studies and clinical trials have been performed to test the preventive and therapeutic effects of EGCG in neurodegenerative diseases such as Alzheimer's, Parkin‐ son's or Huntington's diseases [59]. In general, these studies show no EGCG-related preven‐ tion of neuronal and glial cell death in patients [59]. In 2013, a multicentric Phase II national study to investigate the anti-inflammatory and neuroprotective effects of EGCG in 120 RRMS patients was started (SuniMS Study, ClinicalTrials.gov identifier: NCT00525668, sponsored by Charite University, Berlin, Germany). At the moment, no results are available for this study.

Finally, to analyse the metabolic effects of EGCG and to assess the importance of lipid oxidation in fuel muscle's energy metabolism and its relationship with muscle weakness and fatigue in RRMS patients, the same institution completed a clinical research trial in 2013 (ClinicalTri‐ als.gov identifier: NCT01417312). Results showed that EGCG given to MS patients over 12 weeks improves muscle metabolism during moderate exercise to a greater extent in men than in women, possibly because of sex-specific effects on autonomic and endocrine control. These results indicate that EGCG could be a promising treatment for MS, with a good- and wellknown safety profile and an interesting combinability with other treatments. However, as the EGCG pass through the blood-brain barrier in humans is still controversial, its bioavailability

Adenosine triphosphate (ATP)-dependent potassium (KATP) channels play important roles in many cellular functions by coupling cell metabolism to electrical activity. First detected in cardiac myocytes, they are also expressed in a wide number of cell types such as pancreatic β-cells, skeletal and smooth myocytes, neurons and microglia. In these cells, KATP channels regulate potassium fluxes across the cell membrane when glucose is available in sufficient conditions, which couple the electrical activity of the cell to energy metabolism [61, 62]. An increase in the cellular ATP concentration leads to channel closure and membrane depolari‐ zation. On the contrary, metabolic inhibition opens KATP channels and suppresses electrical

Functional KATP channels in the cell membrane are assembled as a heterooctameric complex [63] from two structurally distinct subunits: the regulatory sulphonylurea receptor (SUR) and the pore-forming inwardly rectifying potassium channel (Kir) subunit 6.1 or 6.2. While ATP inhibits the KATP channel by directly binding to the cytoplasmic Kir6 domains, activators, such as potassium channel openers (KCOs), and inhibitors, such as sulphonylurea drugs [64], bind SUR to modulate the channel. Similar KATP channels have also been described in the mito‐ chondria, located on the inner membrane of these organelles where they play a crucial role in

treatment of MS and other neurodegenerative diseases.

remains to be characterized [60].

316 Trending Topics in Multiple Sclerosis

activity.

**3. Effects of diazoxide in multiple sclerosis**

the maintenance of mitochondrial homeostasis and function [65].

**3.1. KATP channels and neuroinflammation**

Our laboratory has described the expression of KATP channels in microglia [68–70], through which they control the release of a diversity of inflammatory mediators, such as NO, IL-6 or TNF-α [71, 72]. We also evidenced that reactive microglia increase the expression of the KATPchannel components Kir6.1, Kir6.2, SUR1 and SUR2B [69, 73]. With this increased microglial expression, KCOs may modulate neuroinflammation. In this line, we and other authors have documented that pharmacological activation of KATP channels with diazoxide can exert CNS neuroprotective and anti-inflammatory effects against excitotoxicity, ischaemia, trauma and neurotoxicants [71, 72, 74, 75]. For example, any rat brain microinjection of glutamate ana‐ logues triggers a persistent process that leads to progressive atrophy with a widespread neuronal loss and a concomitant neuroinflammation [76]. This neurodegenerative process is reduced by diazoxide oral treatment, which ameliorated microglia-mediated inflammation and reduced neuronal loss [73].

Controlling the extent of microglial activation may offer prospective clinical therapeutic benefits for inflammation-related neurodegenerative disorders [73]. In this context, increased expression of KATP channels by activated microglia and the specific anti-inflammatory actions of diazoxide reveal the use of this drug as a therapeutic agent to treat MS inflammatory processes.

#### **3.2. KATP channel openers and experimental autoimmune encephalomyelitis**

We analysed the putative neuroprotective effects of diazoxide on an EAE murine model of MS by its oral administration in the classical EAE MOG35-55 mouse model for preclinical studies of MS [72]. The doses tested (0.8 and 0.05 mg/kg) were below those that induce blood glucose increase (>1 mg/kg) and both of them ameliorated clinical signs of EAE, being 0.8 mg/kg more effective [72]. Then, this dose was tested in two different experimental designs: (a) a preventive paradigm, in which diazoxide was administrated daily starting on the same day animals were immunized and (b) a palliative paradigm, in which mice were treated with diazoxide daily starting when they reached a clinical score of ≥1. In both cases, the treatment showed similar effectiveness and mice treated with diazoxide obtained a lower clinical score during the chronic phase of EAE. However, in the preventive paradigm, diazoxide could not delay EAE onset or reduce the number of animals developing EAE. This indicates that diazoxide effects could be mediated mainly by neuroprotection rather than by immunosuppression. The histopatholog‐ ical analysis of injured spinal cords confirmed that diazoxide elicited a significant reduction in myelin and axonal loss accompanied by a decrease in glial activation and neuronal damage, but it did not affect the number of infiltrating lymphocytes positive for CD3 and CD20.

We then analysed the neuroprotective properties of diazoxide in vitro and ex vivo [77]. In this study, diazoxide effectively protected NSC-34 motoneurons against oxidative, excitotoxic and inflammatory insults. It also enhanced the expression and nuclear translocation of Nrf2 in these cells as well as in the spinal cord of EAE animals orally administered with diazoxide. This demonstrated that one of the mechanisms of actions implied in the neuroprotective role of diazoxide is mediated by the activation of Nrf2 expression and nuclear translocation. Finally, diazoxide decreased neuronal death in organotypic hippocampal slice cultures after excito‐ toxicity and preserved myelin sheath in organotypic cerebellar cultures exposed to proinflammatory demyelinating damage. Thus, diazoxide is a neuroprotective agent against oxidative stress-induced damage and cellular dysfunction.

We finally studied the putative actions of diazoxide on autoimmune key processes during EAE such as antigen presentation and lymphocyte activation and proliferation [78]. For this, we analysed KATP channel expression in CD4-positive T cells and the proliferative of lymphocyte response in the EAE model. When we used whole splenocytes to test whether diazoxide could modulate lymphocyte proliferation, results showed a significant inhibition of the lymphocyte proliferative rate both in vitro and in vivo. Also, the expression of dendritic cell activation markers such as CD83, CD80, CD86 or major histocompatibility complex class II was reduced in cultures treated with diazoxide. However, we observed no inhibition of cell proliferation when isolated CD4-positive T lymphocytes were used instead of whole splenocytes. Diazoxide also failed to inhibit the expression of lymphocyte activation markers. These results suggest that although diazoxide does not directly suppress lymphocyte activation and proliferation, it could modulate lymphocyte activity by regulating antigen presentation. These discrete effects indicate again that diazoxide treatment attenuates EAE pathology with no immuno‐ suppressive effects.

Taken together, at the doses studied, our results demonstrate novel actions of diazoxide as an anti-inflammatory drug that present beneficial effects on EAE through neuroprotection. At the functional level, diazoxide is a neuroprotective agent against oxidative stress-induced damage and cellular dysfunction (**Figure 2**). It attenuates EAE pathology not by causing lymphocyte suppression but by modulating immune communication, decreasing glial harmful activation and promoting myelin and neuronal protection. Thus, treatment with this widely used and well-tolerated drug may be a useful therapeutic intervention in ameliorating MS disease.

Neuroprotection: A New Therapeutic Approach of Relapsing Remitting Multiple Sclerosis http://dx.doi.org/10.5772/63730 319

**Figure 2.** Diazoxide modifies microglial reactivity in brain. Drawing of the effects of diazoxide in the microglial reac‐ tion during MS. (1) Diazoxide (Dzx) binds to mitochondrial KATP channels, induces depolarization of the mitochondrial internal membrane (MIM) and potentiates the H+ gradient generated by the electron transport chain (ETC). This en‐ hances both ATP synthesis and calcium concentration by activation of ATP synthase and the mitochondrial calcium uniporter (MCU), respectively. Calcium in the mitochondria also increases the tricarboxylic acid cycle (TCA) flux by activation of deshydrogenases and enhances ATP production [73]. Dzx also activates KATP channels from the plasma membrane, which modifies the cell response to activation signals, (2) activates Nrf2 and (3) through a mechanism that remains to be described. ARE, antioxidant response element; NE, nuclear envelope.

#### **3.3. Effects of diazoxide in MS patients**

starting when they reached a clinical score of ≥1. In both cases, the treatment showed similar effectiveness and mice treated with diazoxide obtained a lower clinical score during the chronic phase of EAE. However, in the preventive paradigm, diazoxide could not delay EAE onset or reduce the number of animals developing EAE. This indicates that diazoxide effects could be mediated mainly by neuroprotection rather than by immunosuppression. The histopatholog‐ ical analysis of injured spinal cords confirmed that diazoxide elicited a significant reduction in myelin and axonal loss accompanied by a decrease in glial activation and neuronal damage, but it did not affect the number of infiltrating lymphocytes positive for CD3 and CD20.

We then analysed the neuroprotective properties of diazoxide in vitro and ex vivo [77]. In this study, diazoxide effectively protected NSC-34 motoneurons against oxidative, excitotoxic and inflammatory insults. It also enhanced the expression and nuclear translocation of Nrf2 in these cells as well as in the spinal cord of EAE animals orally administered with diazoxide. This demonstrated that one of the mechanisms of actions implied in the neuroprotective role of diazoxide is mediated by the activation of Nrf2 expression and nuclear translocation. Finally, diazoxide decreased neuronal death in organotypic hippocampal slice cultures after excito‐ toxicity and preserved myelin sheath in organotypic cerebellar cultures exposed to proinflammatory demyelinating damage. Thus, diazoxide is a neuroprotective agent against

We finally studied the putative actions of diazoxide on autoimmune key processes during EAE such as antigen presentation and lymphocyte activation and proliferation [78]. For this, we analysed KATP channel expression in CD4-positive T cells and the proliferative of lymphocyte response in the EAE model. When we used whole splenocytes to test whether diazoxide could modulate lymphocyte proliferation, results showed a significant inhibition of the lymphocyte proliferative rate both in vitro and in vivo. Also, the expression of dendritic cell activation markers such as CD83, CD80, CD86 or major histocompatibility complex class II was reduced in cultures treated with diazoxide. However, we observed no inhibition of cell proliferation when isolated CD4-positive T lymphocytes were used instead of whole splenocytes. Diazoxide also failed to inhibit the expression of lymphocyte activation markers. These results suggest that although diazoxide does not directly suppress lymphocyte activation and proliferation, it could modulate lymphocyte activity by regulating antigen presentation. These discrete effects indicate again that diazoxide treatment attenuates EAE pathology with no immuno‐

Taken together, at the doses studied, our results demonstrate novel actions of diazoxide as an anti-inflammatory drug that present beneficial effects on EAE through neuroprotection. At the functional level, diazoxide is a neuroprotective agent against oxidative stress-induced damage and cellular dysfunction (**Figure 2**). It attenuates EAE pathology not by causing lymphocyte suppression but by modulating immune communication, decreasing glial harmful activation and promoting myelin and neuronal protection. Thus, treatment with this widely used and well-tolerated drug may be a useful therapeutic intervention in ameliorating MS disease.

oxidative stress-induced damage and cellular dysfunction.

suppressive effects.

318 Trending Topics in Multiple Sclerosis

Based on the positive preclinical studies and considering that diazoxide has been on the market for decades with an excellent safety profile, two Spanish biotechnological companies, Neurotec Pharma and Advancell, performed a clinical development programme to assess diazoxide efficacy and safety in MS patients. NEUROADVAN trial (ClinicalTrials.gov identifier: NCT01428726) was initiated in 2011 and ended in 2014. NEUROADVAN was a Phase IIa, multinational, double-blind, placebo-controlled clinical trial to evaluate the efficacy and safety of diazoxide. The drug was orally administered at two different doses and compared in 103 patients with RRMS versus placebo (1:1:1). The total duration of the treatment was 6 months. Additionally, patients were allowed to continue during an optional extension period, in the same arm of the study, in a blinded way, until study finalization.

Men and women aged 18–55 years with RRMS (McDonald criteria 2010 [79]) and an EDSS score of 0–5.0 were eligible for the study. The inclusion criteria required at least one relapse in the previous 2 years or the presence of at least one Gad1 lesion in the previous year. During 24 weeks, patients received one daily oral tablet with 0.3 or 4 mg diazoxide. MRI scans were performed at baseline and every 4 weeks until the end of the study. Patients enrolled in the follow-up study were subjected to an additional scan at week 48.

The primary efficacy end point was that the number of new Gad1 lesions appeared on T1 weighted sequences from weeks 4 to 24. This end point has been validated in many previous 6-month MS trials and was based on the diazoxide effects on microglia activation and bloodbrain barrier closing. Secondary MRI end points included cumulative number of lesions on T2-weighted sequences for all MRIs, cumulative number of lesions on T2-weighted sequences in the 6 months of the study (compared with the baseline MRI); cumulative number of combined unique active lesions (CUALs), addition of new or enlarged lesions on T2-weighted sequences that do not enhance with gadolinium and new Gad1 lesions for all MRIs; cumulative number of CUALs, addition of new or enlarged lesions on T2-weighted sequences that do not enhance with gadolinium and new Gad1 lesions in the 6 months after starting therapy (compared with the baseline MRI); number of patients without Gad1 lesions in T1-weighted sequences in the 6 months after starting therapy; and percentage brain volume change (PBVC). Secondary clinical end points also included the measurement of relapse-free status, relapse rate, number of relapses requiring corticosteroid treatment, time to first relapse during the trial, change in EDSS scale and quality of life. Secondary safety end points were monitorized during the 6 months of therapy up until 15 days after the last dose of diazoxide and included incidence, nature and severity of adverse events (AEs). Control of glucose levels, glycated haemoglobin and blood pressure were also monitorized.

The results of the clinical trial showed no differences in the diazoxide groups in the primary end point or in the other MRI variables associated with the presence of new lesions [80]. However, an interesting decrease in the PBVC in the patients who received diazoxide com‐ pared with placebo was found. The number of new T2/Proton Density lesions converting to black holes was not different between arms. Finally, in accordance with the small sample size of the trial, no differences were detected in clinical variables of relapse-free status, relapse rate, number of relapses requiring corticosteroid treatment, change in EDSS or quality of life.

There were six serious adverse events during the study but only a case with autoimmune hypothyroidism was considered to be related to the therapy that was not discontinued. Regarding the described clinical effects of diazoxide on glycaemia, the detected glucose blood levels were always within the limits of normality. This confirms that the diazoxide doses used in this trial were lower than those that induce hyperglycaemic and hypotensive effects.

As commented above, at the doses tested, diazoxide does not seem to have a significant effect that impedes the appearance of new Gad1 lesions. However, although patients were randomly distributed within the groups, at the beginning of the study those patients receiving diazoxide presented a higher number of Gad1 lesions than patients included in the placebo group. This initial higher disease activity in the group treated with diazoxide was maintained throughout the trial. These findings confirm the need to perform an extensive and accurate selection of patients, in terms of the activity of the disease, which must be carried out prior to the clinical trial.

Regarding the effects of diazoxide on brain atrophy, PBVC progressed more slowly in treated patients in a dose-dependent manner. This is consistent with a recent study that validates the measurement of brain atrophy as an outcome for Phase II trials in RRMS [81]. However, taking into account that treated patients had a more active disease, and that diazoxide presents vasodilator actions and peripheral resistance, the reduced atrophy found in patients treated with diazoxide may also be secondary to fluid shifts and not a true protection against brain injury. Nevertheless, the effects in slowing the progression of brain atrophy would require further validation.

To sum up, NEUROADVAN study indicates that diazoxide is a safe drug, well tolerated in patients with MS, but it was not possible to find evidence of efficacy in preventing the formation of new inflammatory lesions.
