WO2022195451A1 - Redox-active compound for use in the method for treating diseases due to dysfunction of the mitochondrial respiratory chain complexes i, ii, iii - Google Patents

Abstract

Redox-active compound or a pharmaceutically acceptable salt thereof for use in the method for treating a disease selected from among the diseases due to dysfunction of the mitochondrial respiratory chain complexes I, II, III. This redox-active compound or a salt thereof have a standard reduction potential comprised between -0.15 and +0.25 volts and is selected from the group consisting of pyocyanin, indigo, indigo carmine, 2,6-dichloroindophenol, 2-chloroindophenol, indophenol and indophenol blue.

Description

REDOX-ACTIVE COMPOUND FOR USE IN THE METHOD FOR TREATING DISEASES DUE

TO DYSFUNCTION OF THE MITOCHONDRIAL RESPIRATORY CHAIN COMPLEXES I, II, III

DESCRIPTION

The present invention relates to a redox-active compound for use in the treatment of a disease selected from among the diseases due to dysfunction of the mitochondrial respiratory chain complexes I, II, III, including Parkinson’s disease, a neurodegenerative disorder whose pathogenesis has now been unequivocally proven to be contributed to by complex I deficiency.

The present invention further relates to a pharmaceutical composition comprising said redox-active compound and at least one excipient, and to the use of the aforesaid pharmaceutical composition in the method for treating diseases due to dysfunction of the mitochondrial respiratory chain complexes I, II, III.

BACKGROUND

Mitochondria synthesise the ATP needed for most cellular functions. A reduction in ATP synthesis may be due to a functional deficiency in the components of the respiratory chain (RC), namely complexes I, II, III, IV (OXPHOS). Diseases due to dysfunction of the mitochondrial respiratory chain complexes are caused by a reduction in electron flow leading to a reduction in ATP synthesis and give rise to further metabolic alterations, which characterize the overall pathophysiology. Patients with complex III (Cl 11) functional deficiency develop multiple and progressive neurological disorders which lead to premature death (Ghezzi, D. & Zeviani, doi:10.1042/ebc20170099). These defects have been associated with mutations in several genes (Ghezzi, D. & Zeviani, doi:10.1042/ebc20170099; Ghezzi, D. et al, doi:10.1038/ng.76; Bottani, E. et al, doi:10.1016/j.molcel.2017.06.001). For example, the nonsense and missense mutations reflected in three protein products, TTC19 (Ghezzi, D. et al, doi: 10.1038/ng.761 ), LYRM7 (Invernizzi, F. et al, doi: 10.1002/humu.22441 ) and BCS1L (De Lonlay, P. et al, doi:10.1038/ng706), cause defects in the assembly/stability and, consequently, in the functional activity of Clll, with decreased ubiquinol:cytochrome c oxidoreductase activity. Nonsense mutations in the gene encoding the TTC19 protein lead to the onset of a heterogeneous syndromic spectrum, consisting of several signs of neurological failure (Ghezzi, D. et al, doi:10.1038/ng.76), including progressive encephalomyopathy, severe psychiatric symptoms, Leigh syndrome and cerebellar ataxia (Ghezzi, D. & Zeviani, doi:10.1042/ebc20170099; Ghezzi, D. et al, doi:10.1038/ng.761). In particular, LYRM7 and BCS1L are chaperone proteins involved in Clll biogenesis.

Some animal models summarising Clll deficiency diseases have been used to study their pathophysiology. A Drosophila melanogaster model characterized by TTC19 knock-out (KO) shows a Clll activity deficiency associated in humans with multiple neurological symptoms (Ghezzi, D. et al, doi:10.1038/ng.761). A model of transcriptional repression of TTC19 (“knock down”, KD) in zebrafish ( Danio rerio) shows obvious alterations in development and thus in embryonic morphology (Costa, R. et al, doi: 10.1016/j.celrep.2019.07.050).

Recently, a mouse KO for TTC19 was also obtained, showing an approximately 50% reduction in complex III activity and a phenotype characterized by progressive neurodegeneration with the onset of multiple motor deficiencies (Bottani, E. et al, doi:10.1016/j.molcel.2017.06.001). Recent scientific work describes how the Alternative Oxidase (AOX) of Ciona intestinalis expressed in Clll and CIV-deficient cells (cytochrome c oxidase, COX) (Leveen, P. et al, doi: 10.1002/hep.24031 ; Fernandez-Ayala, D. J. et al, doi:10.1016/j.cmet.2009.03.004) replaces Clll and CIV, transferring electrons directly from the ubiquinone pool to molecular oxygen, without “pumping” protons from the matrix to the intermembrane space. It is believed that the unblocking of electron flow, and the subsequent restart of proton pumping by the Cl, operated by AOX expression, may increase ATP production in Clll or CIV-deficient cells (Fernandez-Ayala, D. J. et al, doi:10.1016/j.cmet.2009.03.004). However, the possible use of AOX as a therapy in Clll or CIV deficiencies is controversial (Dogan, S. A. et al, doi:10.1016/j.cmet.2018.07.012).

A number of mitochondriogenesis-stimulating compounds have also been tested by promoting the action of PGC1-a, the major transcriptional coactivator of the mitochondrial bioenergy program (Viscomi, C. et al, doi:10.1016/j.cmet.2011.04.011; Cerutti, R. et al, doi:10.1016/j.cmet.2014.04.001; Khan, N. A. et al, doi: 10.1002/emmm.201403943). Among these, AICAR, the AMPK agonist, acts by increasing the phosphorylation of PGC1-a (a modification that promotes its activation) and the levels of nicotinamide riboside (NR; a precursor of NAD⁺), the substrate (and activator) of Sirtuin 1 (a nuclear deacetylase that in turn activates PGC1-a). However, its administration did not induce a recovery of cellular respiration or prolong the life of Cl I l-deficient animals (Purhonen, J. et al, doi:10.1096/fj.201800090R), while the ketogenic diet, also proposed as an activator of mitochondrial biogenesis, produced only a modest benefit (Purhonen, J. et al, doi: 10.1038/s41598-017-01109-4).

With regard to complex I, it can be inactivated, for example, due to mutations in the NDUFS4 gene, i.e. in subunit 4 of NADH:ubiquinone oxidoreductase, which condition leads to the development of Leigh syndrome, a severe childhood-onset encephalopathy (e.g.: Van de Wal, M. et al, doi: 10.1093/brain/awab426).

A pathology associated with functional problems of complex I that affects, instead, elderly patients is Parkinson’s disease. Parkinson’s disease is the second most common disabling neurodegenerative disease (after Alzheimer’s disease) with an incidence of 1.8% in the population over 60 years of age. The characteristic degeneration of the dopaminergic nervous system has been closely related to functional deficiencies of the mitochondrial I complex by a long series of studies (e.g.: Schapira, A.H. et al, doi:10.1016/s0140- 6736(89)92366-0; Mizuno, Y. et al, doi:10.1016/0006-291x(89)91141-8; Schapira, A.H. et al, doi: 10.1111/j.1471-4159.1990.tb02325.x; Schapira, A.H. et al, doi: 10.1111/j.1471-4159.1990.tb05809.x; Janetzky, B. et al, doi: 10.1016/0304-3940(94)90372-7; Mann, V.M. et al, https://doi.org/10.1002/ana.410360612). The symptoms of Parkinson’s disease can be caused by exposure to complex I inhibitors such as MPTP, a drug (e.g.: Langston, J.W. et al, doi: 10.1126/science.6823561), and rotenone, a pesticide (e.g.: Betarbet, R. et al, doi: 10.1038/81834). According to this widely accepted aetiological model, complex I dysfunctions leads to a decrease in the transmembrane potential of the mitochondria, with the consequent permanence of the PINK1 kinase on the outer mitochondrial membrane and, even downstream of its activity, the progressive degradation of the mitochondrial network, bioenergetic cell deficiency, and neuroinflammation (for the latter, e.g.: Quinn, P.M.J. et al, https://doi.org/10.1186/s40478-020- 01062-w; Sugumar, M. et al, doi: 10.1007/s10072-021 -05551-1). The malfunction of the respiratory chain causes excessive production of reactive oxygen species (ROS), which promote autophagy and cell death. Finally, a recent study conducted in a murine model (Gonzalez-Rodhguez, P. et al, doi: 10.1038/s41586-021 -04059-0) further reinforces the discovery that complex I dysfunctions alone are sufficient to induce a progression to parkinsonism.

Document US-A-10,123,985 describes methods, compositions and systems for treating mitochondrial disorders by administering aspartate, or an analogue or pro-drug thereof, or an agent that increases intracellular levels of aspartate. Document US-B-9,334,250 describes multifunctional radical quenchers for treating diseases associated with reduced mitochondrial function resulting in decreased ATP production and/or oxidative stress and/or lipid peroxidation. Document US-A-2014/148446 discloses compositions and uses thereof as antioxidants and/or neuroprotective agents for treating medical conditions associated with oxidative stress and/or neural damage, such as, for example, neurological diseases, disorders and trauma, and therefore in the treatment of diseases associated with the central nervous system, disorders and trauma. Document WO-A-2020/028222 describes compounds and methods for treating neurological or mitochondrial diseases, including epilepsy.

Document US-A-2013/345312 describes the treatment of mitochondrial diseases using naphthoquinones.

The scientific publication Robin, A.J. et al, doi: 10.1016/J.TI PS.2012.03.010, describes the biological properties that make mitochondria important in determining an individual’s health and disease and also describes pharmacological strategies developed to address mitochondrial dysfunction. The object of the present invention is therefore to provide a class of novel compounds adapted to induce increased ATP production in diseases due to deficiency of mitochondrial respiratory chain complex I - complex III.

A further object of the present invention is that such compounds have a good safety profile.

An object of the invention is also to provide a pharmaceutical composition for use in the aforesaid diseases due to CI-CIII deficiency.

These objects are achieved by a redox-active compound or a pharmaceutically acceptable salt thereof for use in the method for treating a disease selected from among the diseases due to dysfunction of the mitochondrial respiratory chain complexes I, II, III, according to independent claim 1. In particular, the objects of the invention are achieved by the use of a redox- active compound with a standard reduction potential comprised between -0.15 and +0.25 volts.

The objects are further achieved by a pharmaceutical composition comprising a therapeutically effective amount of a combination of at least one redox-active compound or a pharmaceutically acceptable salt thereof and at least one pharmaceutically acceptable excipient and, further, by the use of said pharmaceutical composition in the method for treating a disease selected from among the diseases due to dysfunction of the mitochondrial respiratory chain complex I, II, III, according to claims 18 and 19, respectively.

Further features of the invention are described in the dependent claims.

The features and advantages of the present invention are highlighted in the detailed description given below and in the embodiments provided by way of illustration and not limitation.

DETAILED DESCRIPTION OF THE INVENTION

As already mentioned, the present invention relates to a redox-active compound or a pharmaceutically acceptable salt thereof for use in the method for treating a disease selected from among the diseases due to dysfunction of the mitochondrial respiratory chain complexes I, II, III.

The redox-active compound or a salt thereof has a standard reduction potential comprised between -0.15 and +0.25 volts.

Here and in this description, the expression redox-active compound means a compound that is active in chemical reactions in which the oxidation number of the atoms changes, i.e. in which there is a passage of electrons from one chemical species to another. The expression redox-active compound is herein considered to be synonymous for redox-cycler compound. The term “redox cyclers” refers to compounds capable of carrying out reduction and oxidation cycles, such as in particular cycles characterized by the acquisition of an electron (reduction), to form radical species that can then react with an acceptor (oxidation), regenerating the original molecule.

As is well known in the art, the standard reduction potential measures the propensity of a chemical or biological species to gain or lose electrons by ionisation (Lu Y., Marshall N.M. https://doi.org/10.1007/978-3-642-16712- 6_19). An oxidative-reductive reaction can in fact be divided into two semi reactions: one in which a chemical species is oxidised and one in which a chemical species is reduced. The standard reduction potential of a compound is the potential measured under standard conditions by means of a hydrogen electrode. In biological systems, the standard reduction potential is defined at pH 7, using a hydrogen electrode at a partial pressure of hydrogen of 1 bar. Based on the results of the experiments conducted by them, the inventors believe that certain redox-active compounds having a standard electrochemical potential close to that of the ubiquinol/ubiquinone pair (Manago, A. et al, doi:10.1089/ars.2014.5979) are able to accept electrons from ubiquinol and act as “electron carriers” by transferring these electrons to cytochrome c and, in part, directly to oxygen. This increases respiration, resulting in an improvement in phenotypes linked to diseases due to deficiency of the mitochondrial respiratory chain complexes l-lll.

According to one aspect of the invention, it provides for administering to a subject affected by a disease due to dysfunction of the mitochondrial respiratory chain complexes I, II, III a therapeutically effective amount of the redox-active compound or of a pharmaceutically acceptable salt thereof. Preferably, this therapeutically acceptable amount is such that it results in sub micromolar nominal concentrations, which are low enough not to cause severe oxidative stress.

The inventors have indeed observed that high concentrations of these redox- active compounds can, in some cases, cause oxidative stress and even be toxic to the body (Manago, A. et al, doi:10.1089/ars.2014.5979), as in the case of the compound pyocyanin (Jayaseelan, S. et al, doi: 10.1007/s11274-013- 1552-5) (described later).

Surprisingly, instead, the use of concentrations in the sub-micromolar range, preferably in the concentration range <1pM, allows to increase respiration and improve clinical phenotypes in vivo, as clearly demonstrated by experiments conducted with the organisms of the model fruit fly and zebrafish affected by TTC19 deficiency.

Preferably, furthermore, the aforesaid redox-active compound is administered orally to the subject.

According to one aspect of the invention, the redox-active compound is conjugated to at least one chemical group imparting improved pharmacological properties to the resulting molecule, such as, by way of non-limiting example, different solubility, increased bioavailability, increased shelf-life, more favourable pharmacokinetic characteristics, specific tropism towards an organ (e.g., brain) and/or, at the cellular level, towards mitochondria.

This conjugation may be of reversible or hardly reversible type.

In the case of hardly reversible conjugation, it may involve linkers and be made with a carbon-carbon bond, with an ether or amine bridge, with a sulfide, sulfoxide, sulfone.

In the case of reversible conjugation (creation of pro-drugs), linkers may be used and the binding between the active ingredient, i.e. the redox-active compound, and such a chemical group may use one or more functions selected from carboxylic ester, carbamoyl ester, sulfonate, disulfide, acetal, ketal, carbonate, amide, imine, urea.

According to one aspect of the invention, this chemical group is oligoethylene glycol. In particular, such an oligoethylene glycol is (CH₂CH₂0)_n-R, where n can take a value comprised between 1 and 20 and R can have any structure. According to one aspect of the invention, such a chemical group is a sugar, preferably an oligosaccharide or a polysaccharide.

By way of example not to be considered as limiting, it can be glucose, mannose, arabinose, sucrose, cellulose, carboxymethylcellulose, chitin (polyacetylglucosamine).

According to one aspect of the invention, such a chemical group is a polyhydroxylated alkyl chain. This chain can be linear, branched, mono- or polycyclic, e.g. such a chain may contain one or more cycles formed by C-C, C-O-C, C-NR-C bonds.

According to one aspect of the invention, this chemical group is a sulfonate (RSO3-).

According to one aspect of the invention, such a chemical group is a natural or non-natural amino acid.

According to one aspect of the invention, such a chemical group is a natural or non-natural peptide.

By way of example not to be considered as limiting, such a peptide may be an organotropic or mitotropic peptide, including oligoarginine sequences. According to one aspect of the invention, the redox-active compound is conjugated to at least one mitochondriotropic group.

It is widely known in the literature that the term “mitochondriotropic group” refers to a chemical entity/group, usually a cation, that can be directed and concentrated in the mitochondria, preferably avoiding metabolic changes during transit through the cytoplasm.

Preferably, said mitochondriotropic group is triphenylphosphonium, methyldiphenylphosphonium or dequalinium.

More preferably, said mitochondriotropic group is triphenylphosphonium.

Figure 15 shows how conjugation of pyocyanin with the mitochondriotropic triphenylphosphonium group increases the percentage of ATP produced by cells with complex I dysfunction.

The mitochondriotropic group can be conjugated to the redox-active compound by means of an appropriate chain (linker), such as, by way of non-limiting example, aliphatic chains, aromatic chains, chains presenting heteroatoms (such as OEG/PEG chains), etc. The choice of the nature and length of the chain is likely to influence the degree of hydrophobicity of the conjugated compound.

A person skilled in the art is able to assess the appropriate nature and length of the chain in order to obtain a conjugated compound that can accumulate rapidly and abundantly in the mitochondria, driven by the mitochondrial membrane potential and, further, he is able to choose the most appropriate method among those known to carry out the conjugation between said mitochondriotropic group and the redox-active compound of the invention.

It is not excluded that, according to variants of the invention, such a mitochondriotropic group is different from triphenylphosphonium or that such a mitochondriotropic group is not present.

According to one aspect of the invention, the redox-active compound is selected from the group consisting of pyocyanin, indigo, indigo carmine, 2,6-dichloroindophenol, 2-chloroindophenol, indophenol and indophenol blue. According to one aspect of the invention, the redox-active compound is pyocyanin.

Pyocyanin (PYO), lUPAC name 5-methylphenazin-1-one, is a molecule secreted by the bacterium Pseudomonas aeruginosa, known to be toxic at high concentrations (50-100 mM) and to cause oxidative stress (Jayaseelan, S. et al, doi: 10.1007/si 1274-013-1552-5). The low molecular weight and the zwitterionic characteristics of PYO make this molecule particularly suitable for easily crossing cell membranes and the blood-brain barrier (Jayaseelan, S. et al, doi: 10.1007/si 1274-013-1552-5). The PYO has a standard reduction potential corresponding to approximately +0.045 V.

In particular, the inventors surprisingly found that low concentrations of PYO (<1mM) are not toxic in cells and in vivo and, on the contrary, increase respiration and improve the clinical and morphological phenotypes observed in fruit flies and zebrafish affected by TTC19 deficiency.

The inventors have indeed discovered that PYO can accept electrons from both NADH and ubiquinol and can in turn yield electrons, preferably to cytochrome c, but also to molecular oxygen. This partly explains the release of mitochondrial ROS and thus the toxicity of PYO when used at high concentrations.

However, experiments conducted by the applicants showed that the beneficial stimulatory effects on respiration produced by PYO at concentrations <1mM clearly outweighed the induction of ROS, leading to an increase in ATP of up to 300% in in vivo models of diseases due to dysfunction of the mitochondrial respiratory chain complexes, particularly Clll dysfunction.

PYO was also able to improve the bioenergetic profile in fibroblasts obtained from five patients with dysfunctions in three different factors involved in the assembly and/or stabilisation of complex III, namely TTC19, BSC1L and LYRM7.

In addition to these cell systems, beneficial effects have also been observed in vivo in model organisms such as Drosophila melanogaster and Danio rerio (fruit fly and zebrafish).

PYO has also shown to exert a positive effect on the morphology of mitochondria, which is closely linked to the bioenergetic efficiency of these organelles.

As can be deduced from Figure 3B, PYO induces an increase in Mfn2 expression, a protein crucial for mitochondrial fusion associated with the activation of mitochondrial metabolism.

The induction by PYO of both an instantaneous increase in respiration and a more lasting and delayed effect of this respiration (after three days of treatment) indicates that low non-cytotoxic concentrations of PYO may promote the maintenance of mitochondrial homeostasis, probably through mild ROS production and consequent activation of mitohormesis. Moderate oxidative stress can also induce an increase in mitochondrial fusion and biogenesis to re-establish appropriate rates of respiration and ATP production, thereby controlling metabolic adaptation.

The recovery of mitochondrial functions thanks to the use of PYO is documented by: (i) stabilisation of membrane potential, (ii) increase in Mfn2 levels and mitochondrial fusion, (iii) increase in PGC-1a expression.

An increase in mitochondrial biogenesis may also occur through the effect of PYO on the Wnt signalling pathway, as it results in increased PGC1a levels (Dott, W. et al, doi:10.1016/j.redox.2013.12.028).

PYO is therefore able to increase the levels of ATP produced by OXPHOS without causing excessive oxidative stress in cells of both humans and other organisms.

The hypothesised mechanism therefore provides that PYO restores mitochondrial function by transporting electrons from NADH and/or ubiquinol to cytochrome c by bypassing the defective complex (complex I or complex III, as the case may be), thus acting directly on ATP production by restoring the bioenergetic efficiency of the mitochondria.

Still advantageously, it was observed that long-term treatment of mice with low concentrations of PYO caused no toxic effects or significant increase in oxidative stress.

According to one aspect of the invention, the redox-active compound is indigo, whose lUPAC name is 2-(3-hydroxy-1 H-indol-2-yl)indol-3-one.

According to one aspect of the invention, the redox-active compound is indigo carmine, whose lUPAC name is disodium (2-(3-hydroxy-5-sulfonate-1 H-indol- 2-yl)-3-oxoindole-5-sulfonate). The standard reduction potential of indigo carmine is approximately -0.13 V.

According to one aspect of the invention, the redox-active compound is chosen from 2,6-dichloroindophenol and 2-chloroindophenol, preferably 2,6- dichloroindophenol. The standard reduction potential of 2,6-dichloroindophenol is approximately +0.217 V. The standard reduction potential of 2-chloroindophenol is approximately +0.23 V.

According to one aspect of the invention, the redox-active compound is indophenol, whose lUPAC name is 4-(4-hydroxyphenyl)iminocyclohexa-2,5- dien-1-one.

According to one aspect of the invention, the redox-active compound is indophenol blue, whose lUPAC name is 4-[4- (dimethylamino)phenyl]iminonaphthalen-1-one. The data set out so far and the examples given below indicate that the redox- active compound of the present invention is particularly suitable for use in the method for treating diseases due to dysfunction of the mitochondrial respiratory chain complexes I, II, III.

In particular, these diseases include Leigh syndrome (LS), various other mitochondrial encephalomyopathies, mitochondriopathy with lactic acidosis and stroke-like episodes (MELAS), Leber hereditary optic neuropathy (LHON), fatal infantile cardiomyopathy and lactic acidosis, macrocephaly with progressive leukodystrophy, unspecified encephalopathies, dystonia, sporadic or maternally inherited myopathies, SDH complex deficiency syndrome (SCD), hereditary paraganglioma/phaeochromocytoma syndrome, familial paraganglioma syndrome, Carney-Stratakis syndrome, progressive optic atrophy, ataxia, myopathy, childhood leukoencephalopathy with SDHAF1 and SDHB mutations, myopathy with or without myoglobinuria, tubulopathy, hepatopathy and encephalopathy, GRACILE syndrome, congenital metabolic acidosis, liver failure, encephalopathy/encephalomyopathy, slowly progressive encephalopathy/rapidly progressive neurological failure, cerebellar ataxia, septo-optic dysplasia, familial hepatopathy and ketoacidotic coma, rhabdomyolysis and myoglobinuria, epilepsy, anaemia, multisystem disorders, mitochondrial sensorineural hearing loss, hypertrophic cardiomyopathy, Alpers-Huttenlocher disease, neurological disorders due to complex III deficiency, cardioencephalomyopathy, leukodystrophy and Parkinson’s disease.

According to one aspect of the invention, the aforesaid disease is a disease due to dysfunction of the mitochondrial respiratory chain complex I.

According to a further aspect, the aforesaid disease is Parkinson’s disease. Parkinson’s disease, in particular, is a neurodegenerative disorder, the pathogenesis of which has now been unequivocally proven to be contributed to by complex I deficiency.

As can be seen from Figures 13A and 14, the redox-active compounds pyocyanin and indigo are capable of increasing ATP in cells exhibiting complex I deficiency, i.e. cells having a mitochondrial condition such as to develop Parkinson’s disease.

A pharmaceutical composition is also part of the present invention which comprises a therapeutically effective amount of a combination of at least one redox-active compound or a pharmaceutically acceptable salt thereof, as defined above, including variants, and at least one pharmaceutically acceptable excipient.

The term “pharmaceutically acceptable excipient” means a compound or a mixture thereof that is optimal for use in a composition formulated for the treatment and/or prevention of a disease, in particular diseases due to dysfunction of the mitochondrial respiratory chain complexes I, II, III.

Excipients suitable for such use are sweeteners, diluents, disaggregants, glidants, dyes, binders, lubricants, stabilisers, adsorbents, preservatives, surfactants, humectants, flavourings, softeners, film-forming substances, emulsifiers, wetting agents, release retardants, colloids, non-permeant compounds and mixtures thereof, and others per se known from the pharmaceutical industry.

The person skilled in the art is able to determine whether and to what extent particular pharmaceutical excipients may serve one or more functions in relation to their relative quantitative contribution in the final composition, what other excipients are present and the route of administration of the composition. Preferably, the pharmaceutical composition of the present invention is formulated in a form suitable for oral administration.

Non-limiting examples of a form suitable for oral administration are tablets, capsules, powders, syrups, elixirs, suspensions, solutions, emulsions, sachets and cachets or formulations suitable for inhalation such as aerosols, solutions or powders.

It is not excluded that the composition of the present invention is formulated in a form suitable for parenteral administration.

Non-limiting examples of a form suitable for parenteral administration are an aqueous buffer solution and an oily suspension.

It is hereby specified that the parenteral administration is meant to be the intramuscular, intravenous, intradermal, subcutaneous, intraperitoneal, intrasplenic administration.

According to one aspect of the invention, the pharmaceutical composition is used in the method for treating a disease selected from among the diseases due to dysfunction of the mitochondrial respiratory chain complex l-lll, preferably among the mitochondrial diseases due to dysfunction of the mitochondrial respiratory chain complex I, more preferably Parkinson’s disease.

BRIEF DESCRIPTION OF THE DRAWINGS

- In Figure 1 , sublethal concentrations of pyocyanin recover ATP production and cellular respiration in murine embryonic fibroblasts lacking the Clll assembly factor TTC19. In Figure 1A, the viability of MEF TTC19^+/ and ^_/ was assessed by MTS assay after 24 hours of treatment to determine the useful range of sub-lethal PYO concentrations. Staurosporine 4 mM was used as a positive control. The mean percentages ± SEM (standard error of the mean) of the absorbance of MTS measured at 490 nm were presented for the untreated control (n=9). Figure 1 B shows the rate of reduction of cytochrome c (expressed as AA550_nm) in isolated mitochondria in which complexes I, III and IV are inhibited by rotenone (2 pM), antimycin A (2 pg/ml) and NaN₃ (2.5 mM), respectively. PYO was added, at the concentrations indicated, after the addition of decylubiquinol (75 pM). The rate of reduction of cytochrome c in the presence of PYO was normalised with respect to the rate under control conditions, taken as a reference (1.0) (n=5 for each PYO concentration; mean values ± SEM are shown). Asterisks denote statistically significant differences. Figure 1C shows the Oxygen Consumption Rate (OCR) of MEF TTC19^+/ and ^_/ measured in the presence of 1.5 pM PYO (mean values ± SEM, n=3). The values were compared to the basal respiration recorded before the addition of PYO, which was considered to be 100%. In Figure 1D, bioenergetic parameters were calculated using OCR values measured after the sequential addition of modulators of mitochondrial function: oligomycin (2 pg/ml); FCCP (600 nM); antimycin A (1 pM)(n=3). Figure 1 E shows the quantification of respiratory complex III activity in mitochondria isolated from MEF TTC19^+/ and Clll activity was assessed under basal conditions and in the presence of PYO, and normalised to citrate synthetase activity (means ± SEM, n=3). Figure 1F shows the OCR measured on MEF TTC19^+/ and ^_/ to compare the basal respiratory capacities thereof. The values are means ± SEM (n=8). In Figure 1G, OCR was assayed in MEF TTC19^+/ to assess the ability of PYO to recover cellular respiration after inhibition of Clll with Antimycin A. The values were compared to basal respiration measured before the addition of Antimycin A, set as 100% (means ± SEM; n=4). Figures 1 H-J show the OCR levels measured in liver samples of mice TTC19^wt and The ability of PYO to increase OCR was assessed both under basal conditions (Figure 1 H) and after complete blockade of Clll with Antimycin A (Figure 1 J) (means ± SEM; n=3). In Figure 1K the ATP content of mitochondrial origin was measured in MEF TTC19^+/ and ^_/ after PYO treatment. Oligomycin-treated preparations were used as controls. The same number of attached cells was treated in fresh culture medium containing 5.5 mM 2-deoxyglucose (2DG) instead of glucose, to inhibit glycolysis. The values are presented as a percentage of the signal emitted by luciferase in cells MEF TTC19^+/ , to compare the amounts of ATP of mitochondrial origin in the two cell lines. The data are means ± SEM (n=4-5). The statistical significance of the differences was calculated for all panels (ANOVA or Student’s t-test; *=p<0.05; ^**=p<0.01 ; ^***=p<0.001 ).

- In Figure 2, the sublethal concentrations of pyocyanin are not toxic in vitro. In Figure 2A the mitochondrial production of ROS from cells MEF TTC19^+/ (left) and ^_/ (right) was determined by measuring the fluorescence intensity of MitoSox^®. Cells were treated with or without 1.5 mM PYO. Antimycin A and Rotenone were used as positive controls for ROS production. The values are expressed as a percentage of the basal fluorescence value measured before additions, and are shown as means ± SEM (n=4). Figures 2B-C show the quantification of lipid peroxidation observed in MEF TTC19^+/ and ^v , treated with or without 1.5 pM PYO for 72 h (Figure 2B) or 2 months (Figure 2C). Cumene hydroperoxide (Figure 2C) was used as a positive control. Means ± SEM (n=11-14 fields from 3 different biological replicates). Figures 2D-E show the quantification of protein oxidation detected by Oxyblot in MEF TTC19^+/ and treated with or without 1.5 pM PYO for 72 h (Figure 2D) or 2 months (Figure 2E) (means ± SEM, n=3). In Figure 2F the mitochondrial potential of MEF TTC19^/_ was determined by monitoring the fluorescence intensity of TMRM after addition of 1.5 pM PYO. The values are expressed as percentages of basal fluorescence measured before treatments and are means ± SEM (n=4).

- In Figure 3, PYO improves mitochondrial morphology in embryonic fibroblasts of mouse TTC19⁷. Figure 3A shows representative TEM images. MEF TTC19^+/ and ^v untreated or treated for 72 hours with 1.5 pM PYO are also shown. After the treatments, the cells were fixed, the images were acquired, and the mean cross-sectional area of 400 mitochondria was calculated for each of the cell lines and conditions using the Image J program. The quantification is presented in the right-hand side of the Figure. Figures 3B-C show the expression of Mitofusin 2 (Figure 3B) and PGC1 alpha (Figure 3C) proteins immunovisualized by Western blot. MEF TTC19^+/ and ^_/ were not treated or were treated for 24, 48 or 72 hours with 1.5 mM PYO, as shown in the Figure. The upper part reports the quantification by densitometry (means ± SEM) derived from 4-8 independent experiments, while the lower part shows representative WBs. Na⁺K⁺ ATPase, b-actin or Vinculin were used as loading controls. In Figure 3D, PYO significantly increases the Wnt signal in a zebrafish reporter strain exposed to 100 nM PYO, a non-lethal concentration, for 24 hours (n>30). The statistical significance of the differences was calculated for all panels (ANOVA or Student’s t-test; *=p<0.05; **=p<0.01; ***=p<0.001).

- In Figure 4, PYO increases respiration and ATP production, and leads to recovery of mitochondrial morphology in human fibroblasts from patients with mutations in the TTC19 gene. Figure 4A shows viability assays (MTS) that were performed on normal human fibroblasts and fibroblasts of a patient with a mutation in the TTC19 gene (human fibroblasts, pt#1). The cells were treated with or without different dosages of PYO for 24 hours to determine what was the highest concentration of PYO without negative effects on cell survival. Staurosporine 4 pM was used as a positive control. Data are reported as means ± SEM of percentages of MTS absorbance at 490 nm compared to the untreated sample, indicated in the graph as “ref” (n=3-4). In Figure 4B, the ATP content of mitochondrial origin was measured in normal human fibroblasts and in fibroblasts from patients after treatment with PYO. Oligomycin was used as a control. The cells were treated for 1 hour in a fresh medium in which glucose had been replaced with 5.5 mM 2-deoxyglucose (2DG) to inhibit glycolysis. The values are reported as percentages of luciferase signal compared to untreated normal human fibroblasts (in grey). The values are means ± SEM (n=4-5). In Figure 4C, the oxygen consumption rate (OCR) was measured in fibroblasts of patients. Data were compared to basal respiration measured before the addition of antimycin A, which was considered to be 100% reference, and are means ± SEM (n=3).

- In Figure 5, PYO recovers mitochondrial morphology in human fibroblasts from patients with mutations in the TTC19 gene, without causing oxidative toxicity. Figure 5A shows the quantification of TEM images of normal human fibroblasts and fibroblasts from patient #1 , untreated or treated for 72 hours with 0.8 mM PYO. Analysis performed like in Figure 3A.

- Figure 6 shows the characterisation of dTTC19 KO flies treated with PYO. In Figure 6A, total ATP levels were measured in 5-day-old male flies {dTTC19 KO) and in controls {w¹¹¹⁸), after 12 or 24 hours of starving (3 groups of 10 individuals each). The difference between KO and controls was significant for 24-hour starving (**p<0.01). In Figure 6B, PYO toxicity was assessed in wild-type male flies {w¹¹¹⁸) after injection of PYO (1.0 at 200 pmol) into the haemolymph (50 individuals in groups of 10). Survival rate was assessed at 24, 48 and 72 hours after injection. In Figure 6C, total ATP levels were measured in 5-day-old control ( w ¹¹¹⁸ and dTTC19 KO) or PYO-treated ( w ¹¹¹⁸ PYO and dTTC19 KO PYO) flies after 24 hours of starving (3 groups of 10 individuals each). Treatment with PYO restored ATP production in dTTC19 KO flies to the level of the control w¹¹¹⁸ (Student’s t-test: **p<0,01). In Figure 6D, PYO induces a partial recovery of sensitivity to mechanical stress (“bang sensitivity”) in 12-day-old flies with impaired locomotor activity. The sensitivity to the “bang sensitivity” test was assessed in 12-day-old KO flies after injection of 1 pmol PYO ( dTTC19 KO PYO) and compared to that of the controls ( w ¹¹¹⁸\ control w¹¹¹⁸ flies for injection, and control dTTC19 KO flies for injection), with n>80 male individuals per genotype. The percentages of flies (and 95% confidence intervals, Cl) reaching the threshold dimensions (2.8, 5.6, 8.4, 11.2 cm) are shown for the mentioned genotypes, with and without PYO injection. In Figure 6E, the toxicity of repeated injections (1 pmol PYO or control solution for injections) was assessed in wild-type male w¹¹¹⁸ flies (60 individuals). The percentage of surviving flies was determined at 24, 48 and 72 hours after the second injection.

- Figure 7 shows the characterisation of KO zebrafish for TTC19 treated with PYO. In Figure 7A, oxygen consumption rate (OCR) was analysed in zebrafish CtrlMO (control morpholino), KD TTC19 and KD TTC19 treated with 100 nM PYO. The respiratory capacity of 72 hpf fish (hours post fertilisation) was analysed using Oxygraph technology. The values of A[pmol0₂]/fish x min for treated CtrlMO, Ttc19spMO and Ttc19spMO fish, respectively, are shown. The t-test for independent data was used for statistical analysis (*=p<0.05, **=p<0.01). The bars indicate the standard error of the mean (SEM). Figure 7B shows the survival of 72 hpf zebrafish as a function of PYO concentration. The fish were exposed to the drug, at the concentrations indicated, for 24 hours. Viability and the touch evoked response (ETR) were assessed. Figure 7C shows assays of touch evoked escape in 72 hpf zebrafish. The evaluation of embryo movements was carried out by measuring the distance (cm) between the initial (centre of the Petri dish) and final positions of the embryo considered as a linear distance travelled starting from the centre. Statistical analysis: t-test for independent data (*=p<0.05, ***=p<0.001). The graph reports means and medians. Figure 7D shows the traces of the movements of 40 fish/Petri dish (circle). The fish were placed in the centre of the plate and stimulated with a touch to the tail. Those treated with 100 nM PYO show a partial movement recovery. In Figure 7E, the reduction of TTC19 expression (knock-down) induces shortening of the body and tail malformation compared to untreated fraternal controls. 100 nM PYO administered in water at 12 hpf result in partial recovery of malformations and decrease mortality. In Figure 7F the peripheral nervous system was studied in Tg(mnx1 :mGFP), where GFP specifically stains motor neurons in vivo. Transcriptional repression of TTC19 reduces GFP expression and the number of motor neurons in the yolk extension region (N/Area). The addition of PYO does not complement these deficient parameters. Flowever, PYO increases axonal development compared to DMSO-treated embryos. The values are means ± SEM. Quantification of the number and length of motor neurons in the region of the yolk extension are shown in the Figure. The image shows the number of embryos manipulated and the number of motor neurons counted in each condition. Statistical significance was determined by means of a two-tailed Student’s f-test (*= p<0.05; **=p<0.01; ***=p<0.001).

- In Figure 8, PYO is not toxic in vivo to wild-type mice. Figure 8A shows the effect of PYO (10 nmol/gpc) on the body weight of the mice. Weekly measurements of mouse weight (grams) and mean ± standard deviation (SD) are also shown. Male (n=4) and female (n=4) C57BL6/J mice are shown. In Figure 8B, plasma levels of inflammatory cytokines (TNF-a, IL-1 b, IL-6) are not significantly altered by PYO treatment (10 nmol/gpc). Individual data are charted. The bars indicate the means ± SEM. Representative TEM images of liver sections from untreated or PYO-treated (10 nmol/gpc) mice are reported in Figure 8C. No visible alterations were found in nuclei, mitochondria, endoplasmic reticulum and other organelles. The scale reference is present in each photo.

- In Figure 9, PYO increases energy efficiency in fibroblasts of two additional patients with mutations in TTC19. In Figure 9A, ATP of mitochondrial origin was measured in fibroblasts from patients #2 (n=11 independent biological replicates) and #3 (n=7 independent biological technical replicates) after PYO treatment. In Figure 9B the mitochondrial production of ROS from fibroblasts of patients #2 and #3 was determined by measuring the intensity of Mitosox^® fluorescence. The data are means ± SEM (n=4 independent biological replicates).

- In Figure 10, PYO increases bioenergetic efficiency in fibroblasts of a patient with a homozygous pathogenic mutation of LYRM7. In Figures 10A- B, ATP of mitochondrial origin was measured in human control fibroblasts and in fibroblasts from a patient carrying a homozygous mutation in the gene encoding LYRM7 (patient #4), with or without PYO treatment. Oligomycin-treated cells were used as controls. The cells were treated for one hour in a fresh medium in which glucose was replaced by 5.5 mM 2- deoxyglucose (2DG). The values are reported as percentages of luciferase signal compared to untreated control human fibroblasts (in grey) (n=4 independent biological replicates). The values are means ± SEM. In Figure 10B, the OCR from fibroblasts of patient #4 was measured in the presence or absence of 0.8 mM PYO. The values were normalised against basal respiration recorded before addition of the compound, taken as 100% value, and are means ± SEM (n=3 independent biological replicates). The maximum respiration is reported to the right in the Figure. The values are means ± SEM with reference to the untreated sample (n=3 independent biological replicates).

- In Figure 11, PYO recovers mitochondrial function in fibroblasts of a patient with a homozygous pathogenic mutation of BCS1L. Figure 11A shows MTS viability assays that were performed on human fibroblasts from a patient carrying a homozygous mutation in the BCS1L gene (patient #5), like in Figure 4A. In Figure 11 B, mitochondrial ATP was measured in control human fibroblasts and in fibroblasts from patient #5, like in Figure 4B. The values are means ± SEM (n=5 independent biological replicates). Figure 11 C shows the stimulated respiration of fibroblasts of patient #5. The values are means ± SEM referring to an untreated sample (n=4 independent biological replicates). Figure 11 D shows the immunovisualization by Western blot and the relative quantification of the expression levels of the indicated proteins in human control fibroblasts and in fibroblasts of patient #5, which were treated and untreated with 0.8 mM PYO for 72 h (means ± SEM, for anti-PGC1a n=4 independent biological replicates, for anti-catalase n=6 independent biological replicates, for anti- SOD-1 n=5 independent biological replicates).

- Figure 12 shows in vivo experiments with PYO in mice with Ttc19 deletion (complex I dysfunction). A Treadmill Test performed on mice with deletion of the TTC19 gene, after injection of PYO or DMSO as a control, is shown. Values refer to the indicated time-points after administration and are means ± SEM with reference to the untreated sample (n=3 independent biological replicates).

- Figure 13 shows in vitro and in vivo experiments with pyocyanin in a murine model with complex I deficiency (NDUFS4^/). PYO increases bioenergetic efficiency in fibroblasts of mice with homozygous pathogenic deletion of NDUFS4. Figure 13A shows the relative percentages of ATP of mitochondrial origin in Ndufs4^+I~ mouse fibroblasts (controls) and in Ndufs4^~l~ mouse fibroblasts (with homozygous deletion of the gene encoding for NDUFS4), treated or not with PYO. The cells were treated for one hour in a medium in which glucose had been replaced by D-galactose. The values are reported as percentages of luciferase signal compared to control murine fibroblasts treated with DMSO (n=4 independent biological replicates). Oligomycin-treated cells were used as controls. The values reported are means ± SEM. Figure 13B shows the Rotarod test performed on mice with deletion of the Ndufs4 gene, after injection of PYO or DMSO (control). The values reported are means ± SEM with reference to the sample not treated with PYO (n=3 independent biological replicates).

- Figure 14 shows in vitro experiments with other redox-active compounds: indigo. Treatment with indigo increases bioenergetic efficiency in fibroblasts of mice with homozygous pathogenic deletion of the gene encoding for TTC19. ATP of mitochondrial origin was measured in control mouse fibroblasts and in mouse fibroblasts with homozygous deletion in the gene encoding TTC19, with or without Indigo treatment. Oligomycin-treated cells were used as controls. The cells were treated for one hour in a fresh medium in which glucose was replaced by 2-DG to block glycolysis. The values are reported as percentages of luciferase signal compared to untreated control murine fibroblasts (in grey) (n=4 independent biological replicates). The values reported are means ± SEM.

- Figure 15 shows in vitro experiments with pyocyanin (a mitochondriotropic derivative). The mitochondriotropic derivative of PYO increases bioenergetic efficiency in fibroblasts of mice with homozygous pathogenic deletion of the gene encoding for TTC19. ATP of mitochondrial origin was measured in control mouse fibroblasts and in mouse fibroblasts with homozygous deletion in the gene encoding TTC19, with or without treatment with PYO- TPP. Oligomycin-treated cells were used as controls. The cells were treated for one hour in a fresh medium in which glucose was replaced by 2-DG to block glycolysis. The values are reported as percentages of luciferase signal compared to untreated control murine fibroblasts (in black) (n=2 independent biological replicates). The values are means ± SEM. EXAMPLES

The following examples are given in order to provide the person skilled in the art with a complete description of how the compounds and compositions of the present invention have been studied. These examples are to be intended purely as not-limiting examples.

Example 1. Materials and methods Example 1.1 Reagents

Piocyanin, oligomycin, staurosporine, rotenone, antimycin A, NaN₃, FCCP were purchased from Sigma-Aldrich. All compounds were dissolved in DMSO. For the experiments on Drosophila melanogaster, PYO was dissolved in EtOH. For each batch, the pyocyanin concentration was checked by HPLC analysis. Example 1.2 Cell cultures

The different cell lines were cultured in DMEM supplemented with 10% foetal bovine serum, 10 mM HEPES, 100 U/ml penicillin and 100 U/ml streptomycin, 1X non-essential amino acids. The cells of patients were obtained from the BioBank of the Besta Institute, Milan. All cells were maintained at 37°C in an atmosphere with 5% CO2.

Example 1.3 Stocks of flies and their maintenance

Flies were reared on a standard cornmeal diet, at 23°C, 70% relative humidity, with a 12-hour light/dark cycle (LD 12:12). For starving experiments, the flies were transferred on a 1% agar medium for appropriate times. The dTTC19 KO knock-out line was generated by deleting the entire encoding sequence (CDS) of the CG15173 gene using the CRISPR/Cas9 system (WellGenetics, Taipei City, Taiwan). The w¹¹¹⁸ strain, used for genetic manipulations, was analysed as a control because of its genomic identity.

Example 1.4 Zebra fish lines and morpholino models

Wild zebrafish of the Tiibingen (Tϋ) strain and reared at the zebrafish service of the University of Padua were used. Zebrafish embryos were analysed using a Leica M165FC microscope and images were captured with a Nikon DS-F12 digital camera. Ttc19spMO were obtained as described in Costa, R. et al, doi: 10.1016/j.celrep.2019.07.050.

Example 1.5 Animal care and handling

Experiments with animals and their care followed the guidelines of the institutional authorities, and were approved by the Italian authorities in charge. C57BL6/J mice of both sexes were used for toxicity experiments: PYO was administered 5 days a week for 9 weeks.

Example 1.6 Isolation of mitochondria and cytochrome c reduction assay

The explanted mouse liver was immediately immersed in isolation medium (250mM sucrose, 5mM FIEPES, 2mM EGTA; pH 7.5) at melting ice temperature. Liver tissue was homogenised in the same solution and mitochondria were isolated by differential centrifugation, as described in Leanza, L. et al, doi:10.1016/j.ccell.2017.03.003. To measure Clll and PYO activity in vitro, the protocol described in Manago, A. et al, doi: 10.1089/ars.2014.5979 was followed.

Example 1.7 MTS assay and determination of ATP concentration To measure cell viability and determination of ATP concentration, experiments were performed as described in Costa, R. et al, doi: 10.1016/j.celrep.2019.07.050.

Example 1.8 Oxygen consumption measurement

Oxygen consumption rate (OCR; respiration) was measured using an XF24 Extracellular Flow Analyser (Seahorse, Bioscience) and data analysis was performed as reported in Dott, W. et al, doi:10.1016/j.redox.2013.12.028. Respiration was measured in small samples of livers of WT and TTC19^/_ mice at 37°C in 2 ml chambers of Oroboros Oxygraph two-channel respirometers, using DatLab software.

Example 1.9 ROS production and mitochondrial morphology The cells were incubated for 30 min at 37°C and in the dark with 1 mM MitoSox^®, and ROS production was measured as in Leanza, L. et al, doi: 10.1016/j.ccell.2017.03.003. Transmission electron microscopy was performed as described in Leanza, L. et al, doi: 10.1016/j.ccell.2017.03.003. Example 1.10 Western blotting

50 pg of proteins were loaded into each lane of 4%-12% NuPage Bis-Tris prefabricated gels (Thermo Fisher Scientific). The following primary antibodies, all diluted in 5% skimmed milk prepared in TBS, were used as in Costa, R. et al, doi:10.1016/j.celrep.2019.07.050: mitofusin-2 (1:1000 in 5% skimmed milk - Abnova); PGC1a (1:1000 in TBS + 5% BSA - Merck Millipore); ND2 (1:500 in TTBS - Thermo Fisher Scientific); CAT (1:1000 in TBS + 5% BSA -Abeam). Example 1.11 Injection procedure in flies

As described in Ghezzi, D. et al, doi:10.1038/ng.761 , 4-day-old male flies were anaesthetised with C0₂ (for max. 5 min), and inoculated using injection needles connected to a cell injector and to a micromanipulator. 1 to 200 pmol of PYO diluted in Ringer’s solution or Ringer with Brilliant Blue FCF dye were administered. The percentage of surviving flies was determined at 24, 48 and 72 hours after the second injection.

Example 1.12 Bang test on flies

1 hour before the experiment, male flies were transferred into plastic tubes (10 flies per tube), as described in Merkling, S. et al, doi: 10.1038/nprot.2015.071. The mechanical stimulus was obtained by placing the tube on a bench Vortex operating at maximum power for 10 seconds. The tube with the flies was then placed vertically in a graduated cylinder in front of a video camera connected to a computer. The number of flies able to climb up the inner wall of the tube up to a predefined height (2.8, 5.6, 8.4, 11.2 cm) in 30 seconds was counted. W¹¹¹⁸ and dTTC19 KO flies injected with Ringer’s solution and water were used as a control. Untreated w¹¹¹⁸ flies were used as an additional control. Example 1.13 Treatment of zebrafish embryos in vivo

20 wild-type zebrafish were distributed in a 6-well plate and treated with DMSO (vehicle) or PYO for 24 hours, starting at 24 hpf (end of the experiment at 48 hpf). The drugs were administered in 4 ml of fish water. Embryo analysis was performed using a Leica M165FC microscope. The experiment was repeated three times. For methods of experiments on Wnt signalling see Costa, R. et al, doi:10.1016/j.celrep.2019.07.050.

Example 1.14 Touch evoked escape reaction in zebrafish The assay of the touch evoked escape reaction can be used to assess muscle performance and stimulus perception (Granato, M. et al, https://doi.Org/10.1242/dev.123.1.399). A 72 hpf zebrafish typically escapes a tactile stimulus by following in most cases a linear centrifugal path until it reaches the edge of the Petri dish. The evoked escape reaction was induced by a single gentle stimulation on the tail of the larvae, using a thin polypropylene tip of a 200 pi pipette. With a single short startle, the embryo reaches its new final position. The video recordings were obtained with a standard digital video camera. The experiment was repeated three times, and each time the test involved more than 20 embryos. The data were processed with the Davinci Resolve program.

Example 1.15 Treatment of mice in vivo

WT C57BL/6J mice were injected with PYO 10 nmol/gpc. The injections were performed 5 times a week for a period of 2 months.

Then, the mice were killed and their organs explanted and immediately frozen in liquid nitrogen. Organ samples were used to verify the possible effects of PYO in vivo. Histology was performed as described in Leanza, L. et al, doi: 10.1016/j.ccell.2017.03.003.

Example 1.16 Enzyme-linked immunosorbent assay (ELISA)

The concentrations of TNF-a, IL-1 b and IL-6 in murine plasma samples were determined by ELISA technique (DuoSet ELISA R&D Systems) following the manufacturer’s instructions. The determination was based on the absorbance values of the recombinant murine standard.

Example 1.17 Quantification and statistical analysis

The statistical significance of the differences was assessed for each treatment using one-way (Dunnett’s test as post-hoc) analysis of variance (ANOVA) or t-test. All statistical analyses were performed using GraphPad Prism software. Example 2. Results

Example 2.1 Effects of PYO in murine embryonic fibroblasts carrying TTC19 deletion, a complex III stabilising factor

Murine embryonic fibroblasts (MEF) from TTC19⁷ mice were initially used (Bottani, E. et al, doi: 10.1016/j.molcel.2017.06.001). MEF TTC19^+/ was used as a control. At high concentrations (50-100 mM), similar to those found in Pseudomonas infections, PYO is toxic due to its redox activity. The non-toxic concentration range was therefore preliminarily identified for TTC19^+/ and ^_/ cell lines: exposure to [PYO] < 3mM has no effect on the survival of these cells, at least within 24 hours (Fig. 1 A).

The next step was to check, using isolated mitochondria, whether PYO, at these non-toxic concentrations, was capable of taking over CIN’s role, i.e. accepting electrons from ubiquinol by transferring them to cytochrome c, thereby reconstituting the electron flow along the respiratory chain. Mouse liver mitochondria permeabilised with alamethicin were used for this purpose. Functionality at the Clll level was assessed by measuring the reduction of cytochrome c in the presence of inhibitors of respiratory chain complexes I, III and IV (rotenone, antimycin A and NaN₃, respectively). Following the addition of decylubiquinol and 0.8 or 1.5 mM PYO, there was a greater reduction in cytochrome c (as indicated by an increase in absorbance at 550 nm, attributed to the chemical reduction of this mobile electron carrier), even in the presence of all inhibitors (Fig. 1 B). These results indicate that PYO can accept electrons from decylubiquinol and use them to reduce cytochrome c, thus taking over the oxidation-reduction functions of Clll. Consistently, 1.5 mM PYO causes a clear increase in both basal and stimulated respiration (Figs. 1C-D), measured as mitochondrial oxygen consumption rate. In accordance with the decrease in protein expression of a Clll subunit and in the activity of the complex (Fig. 1 E), respiration is lower in MEF TTC19^/_ cells than in MEF TTC19^+/ heterozygous cells used as a control (Fig. 1 F).

To further demonstrate that PYO is able to increase respiration even in the absence of functioning Clll, complex III was blocked in MEF TTC19^+/ with antimycin A, and then PYO was added. It is important to note here that the PYO-induced respiration decreases in the presence of complex IV inhibitors (Fig. 1G). To test whether PYO can have a positive effect not only on cells in culture, but also in tissues particularly affected by the disease, liver from wild- type and TTC19⁷ mice was isolated, and respiration was measured using a high-resolution oxygraph. Oxygen consumption, reduced in TTC19^/_ mitochondria compared to TTC19^+/ controls (Fig. 1H) (Costa, R. et al, doi:10.1016/j.celrep.2019.07.050), was increased by PYO, both under basal conditions (Fig. 11) and in the presence of Antimycin A (Fig. 1J). This increase was found to be sensitive to CIV inhibition, confirming that PYO transfers electrons largely to cytochrome c rather than directly to oxygen.

Higher respiration should correspond to greater ATP synthesis. To measure mitochondrial ATP levels, cells were cultured in the presence of 5.5 mM 2-deoxyglucose (2-DG) to eliminate the contribution of glycolysis to cellular ATP synthesis. Relating the data to ATP levels in MEF TTC19^+/ , a reduction is observed in TTC19^/_ cells (Fig. 1K), as expected given the decreased Clll activity observed in various tissues (Granato, M. et al, https://doi.Org/10.1242/dev.123.1.399). Incubation of TTC19^/_ cells with 0.8 or 1.5 mM PYO approximately doubles the level of ATP of mitochondrial origin. PYO can transfer electrons not only to cytochrome c, but also to molecular oxygen. This is why, at high concentrations, it induces ROS production in mitochondria, ATP depletion and ultimately cell death. On the other hand, at low concentrations PYO only increases mitochondrial ROS formation by about 10% in both cell lines, a moderate increase that does not compromise cell survival (Fig. 2A). In addition, these ROS levels do not induce lipid peroxidation, neither after 72 hours of incubation (Fig. 2B), nor after treatment of the cells for 2 months (Fig. 2C). Similarly, protein oxidation was not increased (Fig. 2D for data after 72 hours, and Fig. 2E for the 2-month treatment). PYO also stabilises the transmembrane potential of mitochondria and has a hyperpolarising effect (Fig. 2H).

The manifestations of cell and tissue pathology linked to the deficiency of oxidative phosphorylation are often associated with ultrastructural alterations of the mitochondrial cristae. It was then verified whether the mitochondrial morphology was modified in TTC19^/_ vs. TTC19^+/ cells and whether PYO at low concentrations had any effect on organelle structure. Comparison of the size and shape of mitochondria in TEM (transmission electron microscopy) images of TTC19^/_ cells (vs. TTC19^+/) demonstrated that in untreated homozygous cells the mitochondria are more fragmented, and with a significantly smaller overall area in the images (Fig. 3A). Treatment with PYO for 72 hours leads to an almost complete normalisation of the size of the organelles. Consistently, Mitofusin 2 expression (Mfn2), a protein crucial for mitochondrial fusion, increases in PYO-treated cells (Fig. 3B). PGC-1a levels, the main mediator of the process, were measured to clarify whether mitochondrial biogenesis was increased following PYO treatment. The relative expression of PGC-1a (Fig. 3C) was greater in PYO-treated cells, suggesting that PYO induces both mitochondrial fusion and biogenesis.

It is known that cells with complex III defects are characterized by a reduced signal in the Wnt pathway due to the deficiency in mitochondrial ATP synthesis. It was then verified whether PYO, which as mentioned increases ATP levels, was also able to recover the signal in the Wnt pathway. For this purpose, the zebrafish line ( Danio rerio) Tg(7xTCFX.Iasiam:EGFP)ia4, a canonical Wnt signalling reporter line, was used (Costa, R. et al, doi:10.1016/j.celrep.2019.07.050). PYO, at low concentrations, has been shown to induce this recovery in zebrafish (Fig. 3D).

Example 2.2 PYO increases respiration and mitochondrial ATP levels in human fibroblasts from patients with mutations in TTC19, a Clll quality control factor

The effects of PYO on human fibroblasts obtained from three different patients carrying pathogenic mutations in TTC19 were observed. Similarly to what was observed in MEFs, sub-lethal concentrations of PYO lead to an increase in ATP content in human fibroblasts, both in healthy, control cells and in cells featuring a truncated form of TTC19 (Fig. 4A). The ATP content of the patients’ cells is lower than that of healthy cells (Fig. 4B), but incubation with 0.4 or 0.8 mM PYO restores mitochondrial ATP to the same levels as control fibroblasts. Exposure to sub-lethal concentrations of PYO is associated with a slight increase in basal respiration, as in the case of MEF TTC^^Fig. 4C). Mitochondria from mutant human fibroblasts show fragmentation of their reticulum compared to control fibroblasts, but PYO can reconstitute the elongated phenotype (Fig. 5A).

Example 2.3 PYO increases ATP level in vivo and compensates for motor failure in the Drosophila KO model forTTC19

The effects of PYO on locomotor activity in vivo were also assayed. Firstly, a Drosophila melanogaster line not expressing TTC19 was used. These flies have reduced Clll activity and show a strong sensitivity to the “bang test”, i.e. reduced neuromotor activity (Cerutti, R. et al, doi:10.1016/j.cmet.2014.04.001). Adult dTTC19 KO flies, obtained with CRISP/Cas9 technology, also demonstrate strong sensitivity to the bang test and markedly reduced mitochondrial ATP production after 24 h of starving (Fig. 6A). Prior to its use, the toxicity of PYO was evaluated in wild-type flies ( w ¹¹¹⁸) to define the range of non-toxic dosages in Drosophila. Injection of <50 pmol of PYO/fly into the haemolymph has been shown to have no toxic effects (Fig. 6B). It was also observed that treatment of 5-day-old dTTC19 KO flies with 1.0 pmol PYO/fly results in a strong increase in ATP content (Fig. 6C) and reduces sensitivity to the bang test (i.e. increases locomotor activity) and restores the activity of 12-day-old, almost completely paralysed flies (Fig. 6D). When comparing the percentages of flies demonstrating recovery from the bang test in the experimental groups consisting of the wild-type flies that had been injected with saline solution as a control and of the KO flies treated similarly, again as a control population, the differences were significant for all rise rates, for both young and old flies. In conclusion, PYO results in an increase in the ATP level of dTTC19 KO flies and allows recovery of biochemical and behavioural phenotypes similar to those of wild-type flies, without causing toxicity in vivo (Fig. 6E).

Example 2.4 PYO increases respiration and recovers motility in knock-down zebrafish for TTC19

In addition to the Drosophila model system, the beneficial effect of PYO has been demonstrated in the corresponding Clll disease model in Danio rerio. Reducing TTC19 gene expression with a Morpholino oligomer (Merkling, S. et al, doi:10.1038/nprot.2015.071 (knockdown; KD), a reduction in respiration was observed (Fig. 7A). Mitochondrial oxygen consumption increased significantly following the addition of PYO to the fish water (Fig. 7A). PYO, added 6 hours after fertilisation (hours post-fertilisation, hpf), was replaced twice daily up to 72 hpf.

The concentration used was 100 nM, which is not toxic to zebrafish (Fig. 7B). Like Drosophila, Danio rerio was also able to recover, albeit only partially, its ability to move. In particular, the touch evoked response (ETR) (i.e. the ability of embryos to swim in response to a mechanical stimulation) is almost completely absent in KD fish for TTC19. Some individuals completely recovered the response after treatment with 100 nM PYO, and an improvement in ETR was observed in most of the remaining treated individuals (Figs. 7C and 7D). In addition, in PYO-treated fish there was an improvement in morphological parameters altered in animals KD for TTC19: shortening of the sagittal axis, craniofacial and tail deformities, reduced pigmentation (Fig. 7E) and an increase in the number of axonal junctions (Fig. 7F).

Example 2.5 Absence of toxicity in the mouse of the long-term treatment with PYO in vivo

It was also verified whether the administration of PYO for longer periods of time could cause toxic effects related to oxidative stress. Intraperitoneal injections of PYO (10 nmol/gpc) were therefore administered to adult wild-type mice (n=8) once daily for two months. This long-term treatment did not cause a growth stop in the animals (Fig. 8A) or signs of malaise, and also did not cause an inflammatory phenotype (Fig. 8B). In addition, the histological examination (haematoxylin-eosin staining) of various tissues showed no marked alterations and the ultrastructure of the nuclei and mitochondria in the liver of these mice was not altered (Fig. 8C).

Example 2.6 PYO causes the same beneficial effects in fibroblasts of patients with different mutations leading to Clll deficiency

To confirm that PYO can be used effectively in the treatment of Clll deficiency diseases due to mutations of other important factors besides TTC19 (Figs. 9A, 9B: data with other patients with TTC19 mutation), similar studies to those just described were also carried out on fibroblasts obtained from a patient with a homozygous mutation in LYRM7 (Figs. 10 A, B) and from a mutant in BCS1L (Figs. 11 A-D). The results were comparable to those obtained in patients with TTC19 deficiency, indicating that PYO at sub-lethal concentrations acts in the same way, leading to a significant improvement in mitochondrial function.

The following examples 2.7, 2.8, 2.9 and 2.10 and the corresponding Figures 12, 13, 14 and 15 relate to the effect of PYO, indigo and triphenylphosphonium-conjugated PYO in vivo in mice with Ttc19 deletion (complex I dysfunction), thus correlated to the efficacy of the compounds in Parkinson’s disease.

Example 2.7 In vivo experiments with PYO in mice with Ttc19 deletion (complex I dysfunction)

Fig. 12 represents a graph of the Treadmill Test performed on mice with Ttc19 gene deletion, after injection of PYO or DMSO as a control.

Example 2.8 In vitro and in vivo experiments with pyocyanin in murine model with complex I deficiency ( NDUFS4 ) PYO increases bioenergetic efficiency in fibroblasts of mice with homozygous pathogenic deletion of NDUFS4 (Fig. 13).

Example 2.9 In vitro experiments with other redox-active compounds: indigo Treatment with indigo increases bioenergetic efficiency in fibroblasts of mice with homozygous pathogenic deletion of the gene encoding for TTC19 (Fig. 14).

Example 2.10 In vitro experiments with pyocyanin (mitochondriotropic derivative)

The mitochondriotropic derivative of PYO increases bioenergetic efficiency in fibroblasts of mice with homozygous pathogenic deletion of the gene encoding for TTC19 (Fig. 15).

Claims

1) Redox-active compound or a pharmaceutically acceptable salt thereof for use in the method for treating a disease selected from among the diseases due to dysfunction of the mitochondrial respiratory chain complexes I, II, III, said redox-active compound or a salt thereof having a standard reduction potential comprised between -0.15 and +0.25 volts and being selected from the group consisting of pyocyanin, indigo, indigo carmine, 2,6- dichloroindophenol, 2-chloroindophenol, indophenol and indophenol blue.

2) Redox-active compound for use according to the preceding claim, characterized in that said treatment method provides for administering to a subject affected by said disease a therapeutically effective amount of said redox-active compound or of a pharmaceutically acceptable salt thereof.

3) Redox-active compound for use according to claim 1 or 2, characterized in that said disease is selected from the group consisting of Leigh syndrome (LS), mitochondrial encephalomyopathy, mitochondriopathy with lactic acidosis and stroke-like episodes (MELAS), Leber hereditary optic neuropathy (LHON), fatal infantile cardiomyopathy and lactic acidosis, macrocephaly with progressive leukodystrophy, unspecified encephalopathies, dystonia, sporadic or maternally inherited myopathies, SDH complex deficiency syndrome (SCD), hereditary paraganglioma/pheochromocytoma syndrome, familial paraganglioma syndrome, Carney-Stratakis syndrome, progressive optic atrophy, ataxia, myopathy, childhood leukoencephalopathy with SDHAF1 and SDHB mutations, myopathy with or without myoglobinuria, tubulopathy, hepatopathy and encephalopathy, GRACILE syndrome, congenital metabolic acidosis, liver failure, encephalopathy/ encephalomyopathy, slowly progressive encephalopathy/rapidly progressive neurological failure, cerebellar ataxia, septo-optic dysplasia, familial hepatopathy and ketoacidotic coma, rhabdomyolysis and myoglobinuria, epilepsy, anaemia, multisystem disorders, mitochondrial sensorineural hearing loss, hypertrophic cardiomyopathy, Alpers-Huttenlocher disease, neurological disorders due to complex III deficiency, cardioencephalomyopathy, leukodystrophy and Parkinson’s disease.

4) Redox-active compound for use according to claim 1, 2 or 3, characterized in that said disease is a disease due to dysfunction of the mitochondrial respiratory chain complex I. 5) Redox-active compound for use according to any one of the preceding claims, characterized in that said disease is Parkinson’s disease.

6) Redox-active compound for use according to any one of the preceding claims, characterized in that said redox-active compound is orally administered to said subject.

7) Redox-active compound for use according to any one of the preceding claims, characterized in that said redox-active compound is conjugated to at least one mitochondriotropic group, said mitochondriotropic group being a chemical group adapted to be directed and concentrated in the mitochondria.

8) Redox-active compound for use according to the preceding claim, characterized in that said mitochondriotropic group is triphenylphosphonium, methyldiphenylphosphonium or dequalinium.

9) Redox-active compound for use according to any one of the preceding claims, characterized in that said redox-active compound is pyocyanin.

10) Redox-active compound for use according to any one of claims 1 to 8, characterized in that said redox-active compound is indigo.

11) Redox-active compound for use according to any one of claims 1 to 8, characterized in that said redox-active compound is indigo carmine.

12) Redox-active compound for use according to any one of claims 1 to 8, characterized in that said redox-active compound is 2,6- dichloroindophenol.

13) Redox-active compound for use according to any one of claims 1 to 8, characterized in that said redox-active compound is 2-chloroindophenol.

14) Redox-active compound for use according to any one of claims 1 to 8, characterized in that said redox-active compound is indophenol.

15) Redox-active compound for use according to any one of claims 1 to 8, characterized in that said redox-active compound is indophenol blue.

16) Redox-active compound for use according to any one of the preceding claims, characterized in that said redox-active compound is conjugated to at least one chemical group adapted to improve the pharmacological properties of said redox-active compound.

17) Redox-active compound for use according to the preceding claim, characterized in that said chemical group is selected from oligoethylene glycol, a sugar, a linear, branched, mono or polycyclic polyhydroxylated alkyl chain, a sulfonate, a natural or non-natural amino acid, a natural or non-natural peptide.

18) Pharmaceutical composition comprising a therapeutically effective amount of a combination of at least one redox-active compound or a pharmaceutically acceptable salt thereof, as defined in any one of claims 1 to 17, and at least one pharmaceutically acceptable excipient.

19) Pharmaceutical composition as defined in claim 18 for use in the method for treating a disease selected from among the diseases due to dysfunction of the mitochondrial respiratory chain complex I, II, III, preferably among the diseases due to dysfunction of the mitochondrial respiratory chain complex I, more preferably Parkinson’s disease.