Avant-Propos Le Format De Présentation De Cette Thèse Correspond À Une Recommandation À La Spécialité

Avant-propos Le format de présentation de cette thèse correspond à une recommandation à la spécialité

Maladies infectieuses, à l’intérieur du Master des Sciences de la Vie et de la Santé qui dépend de l’Ecole Doctorale des Sciences de la Vie de Marseille.

Le candidat est amené à respecter les règles qui lui sont imposées et qui comportent un format de thèse utilisé dans le Nord de l’Europe et qui permet un meilleur rangement que les thèses traditionnelles. Par ailleurs, la partie introduction et bibliographie est remplacée par une revue envoyée dans un journal afin de permettre une évaluation extérieure de la qualité de la revue et de permettre à l’étudiant de commencer le plus tôt possible une bibliographie sur le domaine de cette thèse. Par ailleurs, la thèse est présentée sur article publié, accepté, ou soumis associé d’un bref commentaire donnant le sens général du travail. Cette forme de présentation a paru plus en adéquation avec les exigences de la compétition internationale et permet de se concentrer sur des travaux qui bénéficieront d’une diffusion internationale.

Professeur Didier RAOULT

Remerciements

J’adresse mes remerciements aux personnes qui ont contribué à la réalisation de ce travail.

En premier lieu, au Professeur Didier RALOUT, qui m’a accueillie au sein de l’IHU Méditerranée Infection.

Au Professeur Serge MORAND, au Docteur Marie KEMPF et au Professeur Philippe

COLSON de m’avoir honorée en acceptant d’être rapporteurs et examinateurs de cette thèse.

Je souhaite particulièrement remercier :

Le Professeur Jean-Marc ROLAIN, de m’avoir accueillie dans son équipe et m’avoir orientée et soutenue tout au long de ces trois années de thèse.

Le Docteur Fadi BITTAR, mon co-directeur de thèse, qui m’a beaucoup appris et aidée à toujours trouver des solutions pour avancer.

Le Professeur Stéphane RANQUE, pour sa collaboration et m’avoir fourni un environnement de travail adéquat.

Je remercie également :

MOMO, le technicien qui m’a appris à si bien faire mes manips et qui m’a toujours fourni tout le nécessaire pour l’avancement de mes projets.

A toute la JMR-Team, si exceptionnelle, Linda, Edgarthe, Mouna, Adèle, Sophie, Lucie, David, Ousmane, Miharimamy, Ahmed, Reem, Ayline, Meryem, Youssouf sans oublier Muriel.

A mes deux coéquipières du combat scientifique Rym et Tania. Je ne vous remercierai jamais assez pour votre soutien et réconfort.

A ma meilleure amie, Nawal qui a toujours su m’épauler et me tirer vers le haut.

A mes amis et collègues, Ravah, Liliane et Fatima.

Je remercie particulièrement ma sœur Rima sans qui je ne serai jamais la personne que je suis devenue aujourd’hui. A Aissa qui m’a toujours orientée et soutenue dans mes choix. A Aksel et Lahna que j’aime énormément. A mes chers parents, pour tout leur amour et soutien durant ces longues années d’études.

A ma belle-famille, merci pour votre encouragement.

A Ghillas qui m’a toujours encouragée, soutenue et aimée et sans qui je n’aurai jamais réussi à aller au bout de mes ambitions. MERCI pour tout.

SOMMAIRE Résumé ...... 7

Abstract ...... 9

Introduction ...... 11

Partie I : Revue de la littérature sur les mécanismes de résistance aux antifongiques et identification des molécules médicamenteuses pour lesquelles une activité antifongique a été décrite ...... 16

Article 1: Drug-repurposing in medical mycology: the use of off-label compounds as antifungals to overcome the emergence of multidrug-resistant fungi ...... 19

Partie II : Identification de nouvelles molécules antimicrobiennes par criblage d’une chimiothèque de molécules « Prestwick chemical library » de biodisponibilité connue et approuvée par la FDA contre une collection de souches résistantes……………………. 50

Article 2: Identification of new antimycotic agents by screening of the current FDA-approved drugs against emerging invasive molds……………………………………………………… 54

Partie III : Repositionnement de molécules médicamenteuses pour la prise en charge des infections fongiques invasives, en monothérapie ou en association aux antifongiques communément utilisés en pratique clinique………………………………………………. 68

Article 3: In vitro polymyxin activity against clinical multidrug-resistant fungi…………… 72

Article 4: Repurposing of ribavirin as an adjunct therapy for invasive fungal disease………. 83

Partie IV : Identification du mécanisme d’action de la ribavirine sur Candida albicans, et

élaboration de ses éventuelles cibles spécifiques………………………………………110

Partie V: Annexes…………………………………………………………………………. 113

Article 5: Zidovudine: a salvage therapy for mcr-1 plasmid-mediated colistin-resistant bacterial infections?...... 116

Article 6: Colistin-and Carbapenem-Resistant Klebsiella pneumoniae Clinical_Isolates:

Algeria……………………………………………………………………………………… 127

Article 7: Genome sequence and description of Olsenella timonensis sp. nov. isolated from human fecal sample in France……………………………………………………………… 134

Conclusion et perspectives………………………………………………………………... 156

Activités scientifiques……………………………………………………………………... 159

Références…………………………………………………………………………………. 160

Résumé

Les infections fongiques invasives constituent un sérieux problème de santé publique dans le monde ; cette situation se complique par la disponibilité d’un faible nombre d’antifongiques utilisés en pratique clinique. Étant des organismes eucaryotes, les champignons sont phylogénétiquement plus proches de l'hôte humain que les bactéries, limitant ainsi le nombre de cibles spécifiques exploitables dans le développement de médicaments antifongiques. A ceci, vient s’ajouter l’émergence de nombreuses espèces qui présentent une résistance à au moins, une classe d’antifongiques usuels. De ce fait, la réutilisation de composés chimiques commercialisés auparavant et approuvés par la FDA, communément appelé « repositionnement des médicaments » constitue une solution pertinente et applicable à court terme, pour une meilleure prise en charge des mycoses invasives.

Par conséquent, une revue a été rédigée pour élucider les différents mécanismes de résistance aux quatre classes d’antifongiques couramment utilisés mais surtout pour déterminer les molécules repositionnées comme antifongiques par des études antérieures afin d’orienter notre travail en utilisant ce concept. Dans un second temps, 1280 molécules médicamenteuses, constituant la chimiothèque Prestwick (Prestwick, Illkirch graffenstaden, France), ont été testées sur des souches de champignons multirésistants (levures et champignons filamenteux) d’intérêt clinique, isolées à l’Hôpital la Timone de Marseille. Le criblage primitif à une concentration de 10 µM avait permis l’identification de plusieurs molécules capables d’inhiber la croissance fongique, à des pourcentages ≥ 90% et ≥ 70%, des levures et champignons filamenteux respectivement.

Par la suite, notre travail s’est focalisé sur deux molécules médicamenteuses : la colistine et la ribavirine. Les concentrations minimales inhibitrices de ces dernières ont été déterminées, 7

de même que leur activité fongicide ou fongistatique sur une large collection de souches. Des combinaisons synergiques avec les antifongiques habituels ont été mises au point notamment celles de la ribavirine en association avec l’amphotéricine B, l’itraconazole et le voriconazole qui sont actives, entre autres, sur les souches de Candida albicans multirésistantes.

Le but de notre troisième travail a été de comprendre le mécanisme d’action de la ribavirine, un antiviral, sur les Candida albicans et d’identifier sa potentielle cible. Pour se faire, les analogues des cibles de la ribavirine chez le virus de l’hépatite C, retrouvés chez les Candida albicans notamment les enzymes inosine-5’-monophosphate déshydrogénase (IMPDH) et l’ARN polymérase ont été étudiés. Des systèmes PCR et séquençage ont été développés pour détecter et analyser les gènes IMH3 et RPO21 qui codent pour les enzymes IMPDH et ARN polymérase respectivement chez les Candida.

Enfin, dans le cadre d’élargissement de mes champs d’activité au cours de cette thèse, des projets annexes ont été réalisés notamment celui de l’étude moléculaire des mécanismes de résistance à la colistine et la description d’une nouvelle espèce bactérienne, Olsenella

Timonensis.

Mots clés : Résistance aux antimicrobiens, antifongiques, agents mycotiques émergents, alternatives thérapeutiques, chimiothèque Prestwick, repositionnement des médicaments.

Abstract The increasing incidence of invasive infections caused by pathogenic fungi is a major worldwide concern; a serious situation to which the limited number of available effective antifungals to face it, is another problem. Being eukaryotic organisms, these fungal pathogens are phylogenetically closer to the human host than bacterial pathogens. Consequently, the development of a new specific drug targeting fungi (i.e. a new antifungal drug) is a difficult and long task. In addition to this limited therapeutic arsenal, there is a dramatic increase in the incidence of fungal strains resistant to antifungals. Hence, there is a constant need for other compounds that possess antifungal properties. An interesting thought in this field has been addressed via the drug repurposing, where FDA-approved drugs could be tested and used in another therapeutic class. Therefore, a literature review was prepared to report off-label drugs repurposed as antifungal in previous studies.

Secondly, by applying drug-repurposing approach, Prestwick Chemical Library

(Prestwick, Illkirch graffenstaden, France) containing 1,280 compounds previously approved by the Food and Drug Administration (FDA) was tested against multidrug-resistant fungi

(yeasts and filamentous fungi) recovered from La Timone Hospital in Marseille. Primary FDA approved drugs screening at fixed concentration of 10 µM, allowed us to identify several fungal growth inhibitors. These hits were associated with fungal growth inhibitions ≥ 90% and ≥ 70% against the tested yeasts and filamentous fungi respectively.

Among these non-standard antifungals, we focused our study on both colistin and ribavirin drugs. Minimum inhibitory concentrations of these compounds were determined against a large collection of strains, and time-kill curves were performed to establish their fungicidal or fungistatic activity. Moreover, synergistic combinations with the current 9

antifungal agents were examined; notably, association of ribavirin with either amphotericin B, itraconazole or voriconazole active against multidrug-resistant Candida albicans.

The aim of the third part of our work was to identify the mechanism of action of ribavirin, an antiviral compound, on Candida albicans and its potential target. So, we focused our work on the analogue of ribavirin target in hepatitis C virus, present in Candida albicans namely inosine-5'-monophosphate dehydrogenase (IMPDH) and RNA polymerase enzymes.

We designed PCRs and sequencing systems to detect and analyse IMH3 and RPO21 genes that encode IMPDH and RNA polymerase enzymes respectively.

Finally, to expand my research fields, additional projects were carried out including a molecular characterization of the resistance to colistin in bacterial strains collected from Algeria and the description of a new species Olsenella timonensis.

Key words: Antimicrobial-resistance, antifungals, emerging fungal strains, therapeutic alternatives, Prestwick chemical library, drug-repurposing.

Introduction Les infections fongiques invasives sont des affections relativement graves causées essentiellement par des champignons opportunistes et ubiquitaires notamment Candida et

Aspergillus, cependant, le nombre de cas induits par d’autres genres ou espèces ont augmenté de façon considérable au cours des dernières années. Cette diversité d’agents étiologiques qui induit une hausse de cas de mycoses recensées, pourrait être expliquée par de nombreux facteurs. (i) L’amélioration de la prise en charge thérapeutique d’un grand nombre de pathologies a induit une augmentation de l’espérance de vie des sujets âgés, qui représentent un terrain de prédilection aux infections fongiques invasives, en raison de leurs défenses immunitaires affaiblies. (ii) Un important taux d’immunodépression relatif à une augmentation de l’incidence de certaines affections telles que les cancers, infections à VIH mais également les maladies chroniques comme que le diabète. (iii) Une augmentation du nombre de pathogènes connus en raison de l’optimisation des moyens de diagnostic et l’amélioration des techniques d’identification et de caractérisation des agents microbiens [1]. De ce fait, il est difficile de contrôler l’évolution des infections fongiques d’autant plus que leur survenue chez des sujets sains, sans pathologie associée, est souvent décrite [2].

Le nombre d’agents mycotiques impliqués dans les infections chez l’homme pourrait dépasser les 400 espèces [3] et pour traiter ces mycoses, il n’existe que quatre classes d’antifongiques utilisées en pratique courante, à savoir les polyènes, les azolés, les echinocandines et la flucytosine. L’utilisation de ces antifongiques est limitée par le développement d’une néphrotoxicité suite une thérapie souvent prolongée (Amphotéricine B), les diverses interactions médicamenteuses qui induisent une contre-indication de certaines associations (Les azolés) et l’absence de cibles spécifiques induisant de nombreux effets secondaires [4]. A toutes ces contraintes vient se surajouter la fréquence d’échec clinique du 11

traitement antifongique, en raison du développement rapide de la résistance, à au moins une classe thérapeutique, rencontrée dans la prise en charge d’infections à germes invasifs et hautement pathogènes notamment Candida albicans, Aspergillus et les mucors [1]. En mycologie médicale, il existe deux types de résistance, la résistance intrinsèque, naturellement présente chez toutes les souches d’une même espèce et la résistance acquise, induite par une pression de sélection sur l’isolat en question [5].

Face à ces nombreuses problématiques, des alternatives thérapeutiques doivent être mises en place afin d’améliorer la prise en charge des infections fongiques. De l’étude préclinique à l’autorisation de mise sur le marché (AMM), le développement d’un nouveau médicament est très long et pourrait mettre jusqu’à vingt ans avant d’être disponible pour le patient, en tenant compte du fait que le rapport risque-bénéfice devrait être en faveur du patient

[6]. Cependant, il existe actuellement plusieurs milliers de médicaments pour traiter diverses maladies à travers le monde et la découverte de nouvelles drogues s’est longtemps limitée à l’observation empirique des effets produits par certaines substances naturelles sur le cours de la maladie, induisant ainsi une formulation aléatoire des différentes classes thérapeutiques utilisées de nos jours. Outre cela, le développement de molécules spécifiques à chaque maladie n’a été prôné que plusieurs années plus tard. C’est de là que le concept du repositionnement des médicaments a émané et qui a pour but, de tester une molécule pour une indication thérapeutique différente de celle pour laquelle elle a toujours été prescrite. Cette stratégie s’applique à des traitements déjà disponibles sur le marché [7].

Les avantages de cette approche pourraient se résumer dans deux notions fondamentales : rapidité et connaissance. Rapidité, car un médicament ayant une AMM est déjà passé par toutes les étapes d’évaluation de son effet toxique chez l’homme et son innocuité a déjà été démontrée.

Connaissance, car une fois sur le marché, un médicament reste pour toute la durée de son utilisation dans une phase de pharmacovigilance. Ce suivi dans le temps, en conditions réelles et sur une large population, permet d’améliorer et d’élargir la compréhension des effets d’un médicament avant son éventuel repositionnement [8]. Les premiers repositionnements étaient le fruit du hasard et d’observations cliniques fortuites. Par exemple, il a été observé que des personnes atteintes de la maladie de Parkinson présentant une progression particulièrement lente de cette pathologie, étaient toutes traitées pour un cancer et recevaient un traitement pour stimuler leur système immunitaire. Ce médicament est à présent étudié pour un potentiel effet neuroprotecteur dans la maladie de Parkinson.

Par conséquent, en utilisant le concept du repositionnement des médicaments, l’objectif de l’ensemble des travaux de cette thèse était d’identifier d’éventuels nouveaux agents antifongiques à partir des molécules déjà décrites et approuvées par la FDA afin d’assurer une bonne prise en charge des infections fongiques invasives. Les travaux ont été scindés en deux projets principaux :

-Identification de nouvelles molécules antimicrobiennes par criblage d’une chimiothèque de molécules « Prestwick Chemical Library » de biodisponibilité connue et approuvées par la FDA contre une collection de souches résistantes.

-Repositionnement de molécules médicamenteuses identifiées à partir du criblage primitif, pour la prise en charge des infections fongiques invasives, en monothérapie ou en association aux antifongiques communément utilisés en pratique clinique.

Dans un premier temps, une recherche bibliographique a été effectuée afin de déterminer les molécules médicamenteuses antérieurement repositionnées pour la prise en charge des infections fongiques. Les relations de synergie retrouvées entre les antifongiques non-standards

et les antifongiques usuels, ont été également rapportées. Par ailleurs, un aperçu des différents mécanismes de résistance aux agents antifongiques a été fait et l’intérêt de la réalisation des tests de sensibilité aux antifongiques pour le diagnostic et la prise en charge clinique des patients a été discuté. Toutes ces données ont été présentées sous forme d’une revue (Partie I).

La résistance aux antimicrobiens (antifongiques ou antibiotiques) représente un sérieux problème de santé publique et constitue une contrainte relativement difficile à contourner par les cliniciens. Ces derniers sont forcés de réduire le panel de molécules médicamenteuses à prescrire lors de la prise en charge de certaines infections à germes invasifs [9]. C’est pourquoi, le repositionnement des médicaments pourrait apporter une solution rapide et efficace à cette situation. Dans cette partie de notre travail, après la mise au point des différents protocoles utilisés, nous avons testé 1280 molécules médicamenteuses, appartenant à 291 classes thérapeutiques différentes, contre une collection de souches multirésistantes d’intérêt clinique

(Partie II).

Le repositionnement de médicaments antifongiques passe par une étape initiale qui consiste en la réalisation du criblage d’un nombre important de molécules médicamenteuses déjà existantes sur le marché, contre les agents mycotiques multirésistants. Dans un deuxième temps, vient la caractérisation de l’activité in vitro des molécules d’intérêt, sélectionnées sur des critères bien définis. Cette caractérisation consiste en l’identification de la concentration minimale inhibitrice (CMI), l’activité fongicide ou fongistatique de la molécule, les tests de synergie avec les antifongiques traditionnels, mais également l’exploration du mécanisme d’action du ̏ hit ̋ en question (Partie III).

La ribavirine est un analogue de la guanosine qui possède une activité à large spectre contre de nombreux virus à ARN et ADN. C’est le traitement de première intention de l’hépatite

C en association avec l’interféron- α [10]. Après la caractérisation de son activité antifongique, nous nous sommes intéressés au mécanisme d’action de la ribavirine sur les souches de Candida albicans et à l’hétérogénéité de son efficacité, car en effet, pour certaines souches de Candida albicans, les CMI restent relativement élevées. On a spéculé que la cible de cette molécule chez les C. albicans pourrait être identique à celle décrite chez le virus de l’hépatite C. Parmi les cibles communes à ces deux pathogènes, on s’est focalisé sur les enzymes Inosine-

5’monophosphate Déshydrogénase (IMPDH) et l’ARN polymérase. C’est pourquoi, on a développé des systèmes PCR et séquençage des gènes IMH3 et RPO21 qui codent pour les enzymes IMPDH et ARN polymérase respectivement. L’analyse des séquences de ces gènes nous permettrait d’identifier d’éventuelles mutations, pouvant expliquer le mécanisme d’action de cette molécule (Partie IV).

Les mécanismes de résistance aux antimicrobiens ont été décrits, aussi bien chez les champignons que chez les bactéries. De nombreuses souches bactériennes résistantes aux carbapénèmes, ont été recensées partout dans le monde, ce qui a induit la réintroduction de la colistine comme un traitement de dernier recours des infections causées par des bactéries Gram négatif. Cependant, l’émergence de souches bactériennes résistantes à la colistine, a suscité l’intérêt des scientifiques pour l’études des supports moléculaires pouvant médier cette résistance. Dans ce contexte, un autre aspect de notre travail nous a amené à étudier les mécanismes moléculaires de la résistance à la colistine et aux carbapénèmes chez des souches de Klebsiella pneumoniae, isolées de l’hôpital d’Annaba, en Algérie. Par ailleurs, la description d’une nouvelle espèce bactérienne Olsonella timonensis, a été réalisée. Cette bactérie a été isolée dans le cadre du projet culturomique de l’IHU Méditerranée Infection de Marseille et a fait l’objet d’une étude taxono-génomique.

Partie I Revue de la littérature sur les mécanismes de résistance aux antifongiques et identification des molécules médicamenteuses pour lesquelles une activité antifongique a été décrite.

Avant-propos

L’immunodépression, qu’elle soit secondaire aux infections à VIH ou au traitement associé aux greffes d’organes, a induit une augmentation de la vulnérabilité de l’hôte humain aux infections fongiques [11]. Ces conditions ont créé un environnement optimal pour la survenue d’infections opportunistes et ce n’est donc pas le fruit du hasard si des agents fongiques non pathogènes auparavant, deviennent de nouveaux agents mycotiques.

L’émergence de ces nouveaux pathogènes en clinique se fait en parallèle d’une augmentation de la résistance aux antifongiques, qui est un trait inhérent à la sélection naturelle. Dans cette revue de littérature, les mécanismes de résistance propres à chaque classe d’antifongiques sont

élucidés.

Par ailleurs, la prise en charge des infections fongiques devient plus difficile, étant donné le temps nécessaire au développement et la mise sur le marché d’un nouveau médicament. De plus, l'utilisation de médicaments conventionnels tels que les azolés et les polyènes se caractérise souvent par des échecs cliniques, en particulier chez les personnes immunodéprimées [12]. Ainsi, nous nous sommes intéressés aux différents rapports publiés antérieurement sur l'activité antifongique des médicaments antimicrobiens non traditionnels et sur les combinaisons synergiques qui améliorent l'activité des antifongiques usuels. Les mécanismes d’action des médicaments antifongiques repositionnés décrits auparavant ont été

également rapportés dans cette revue.

Aujourd'hui, en utilisant le principe du repositionnement, on constate que les agents antimicrobiens ‘non traditionnels’, qui étaient auparavant prescrits pour traiter des maladies non infectieuses, présentent des propriétés antimicrobiennes. Il est important d’aller plus loin dans d'exploration de ces composés. Dans ce contexte, nous avons présenté des rapports dans lesquels

des molécules appartenant à des classes thérapeutiques variables (allant des antiinflammatoires aux antipsychotiques) possèdent des propriétés antifongiques non négligeables. Toutefois, en raison des toxicités associées à certains de ces médicaments, leur application clinique peut être limitée. Il est également important de démontrer si ces " nouveaux " antifongiques avaient une activité chez l’hôte humain, et surtout pour le traitement d’infections fongiques invasives. À cette fin, l'utilisation de modèle animal pour la reproduction des tableaux cliniques associés aux maladies humaines, devrait être envisagée afin d'établir clairement les avantages thérapeutiques de ces composés chimiques.

Il est important de noter que certains de ces composés agissent, in vitro, en synergie avec les agents antimicrobiens traditionnels, comme les azoles. Cette caractéristique est encourageante afin que ces composés puissent être utilisés à de faibles concentrations sans engendrer d’effets secondaires sur le plan clinique. Néanmoins, l’utilisation de molécules repositionnées impose une certaine vigilance notamment lors de l’utilisation de composés qui ciblent les organismes eucaryotes et des médicaments qui peuvent induire une immunosuppression, afin d'obtenir le résultat thérapeutique souhaité sans effets indésirables.

Article 1: Review:

Drug-repurposing in medical mycology: the use of off-label compounds as

antifungals to overcome the emergence of multidrug-resistant fungi

Hanane Yousfi, Jean-Marc Rolain and Fadi Bittar.

In Progress « Draft manuscript »

1 Title: Drug-repurposing in medical mycology: the use of off-label compounds as antifungals

2 to overcome the emergence of multidrug-resistant fungi

3 Authors: Hanane Yousfi1, 2, Jean-Marc Rolain1, 2 and Fadi Bittar1, 2*

4 Affiliations :

1 5 Aix Marseille Univ, IRD, APHM, MEPHI, Marseille, France.

6 2IHU Méditerranée Infection, Marseille, France.

7 *Corresponding author:

8 Fadi Bittar, IHU Méditerranée Infection, 19-21 boulevard Jean Moulin, 13005 Marseille,

9 France. Email:[email protected]

10 Keys words: Drug-repurposing, antifungals, yeasts, emerging fungi, multidrug-resistance,

11 therapeutic alternatives, new targets.

20 Abstract

21 Immunodepression, whether due to HIV infections or organ transplant treatment, has

22 increased the vulnerability of the human host to fungal infections. These conditions have

23 created an optimal environment for the occurrence of opportunistic infections, and it is no

24 coincidence that previously non-pathogenic fungal agents are becoming new mycotic

25 pathogens. The emergence of these new pathogens in clinical setting is concomitant to the

26 antifungal resistance increase. In this literature review, the resistance mechanisms specific to

27 each class of antifungal agents are described.

28 In addition, difficulties encountered with fungal infections management associated to

29 the time required for new drug development need to consider other therapeutic alternatives.

30 Moreover, the use of conventional drugs as azoles and polyenes is often characterized by

31 clinical failure, particularly in immunocompromised individuals. Thus, the aim of this study is

32 to provide an overview of the antifungal activity of non-traditional antimicrobial drugs,

33 review their eventual mechanisms of action and the synergistic combinations that improve the

34 activity of conventional antifungal drugs.

41 42 43

44 I. Introduction

45 Most public health organizations but also the World Health Organization do not have a

46 fungal infection surveillance program 1, while 1.6 million deaths were registered each year

47 among a high number of people suffering from severe fungal diseases 2. Fungal infections

48 have been long time not documented and recognized in view of the critical need to treat

49 severe bacterial and viral infections but because the most common fungal causative agents are

50 opportunistic, there is persistence of disease symptoms during antibiotic therapy and this

51 make an attention on this group of highly pathogenic microbes 3. Particularly, invasive fungal

52 infections affect patients with a compromised immune system associated to medical

53 vulnerabilities including, immunosuppression encountered in several pathologic conditions

54 namely; hematologic malignancies 4, HIV infection and cancer chemotherapeutics 5. The

55 status of the host constitutes the most significant factor, nevertheless, there are other aspects

56 that may have a major effect 6. In addition, ageing population is now prone to invasive fungal

57 infection because of increasing longevity and improvement of diagnostic methods for

58 diagnosis of opportunistic fungal infections 7.

59 Fungal infections can also appear in healthy patients with no associated pathologies 8 so it

60 is very difficult to control the spread of fungal infections. More than, there is an important

61 diversity of fungal agents’ groups including yeast and yeast-like species such as Candida,

62 Cryptococcus and moulds such as Aspergillus. These pathogens are implicated in a variety of

63 infectious diseases, with varying prevalence and clinical outcomes, which complicate their

64 management 9.

65 To treat a large number of fungal infections, there are only 4 therapeutic classes used in

66 clinical practices, i.e polyenes, flucytosine, azoles and echinocandins (as prophylaxis or as

67 empirical, preventive or specific treatments, monotherapies or in combinations). Although

68 these drugs remain active, they have a significant number of limitations that complicate their

69 routine use as, off-target toxicity, drug-drug interaction, clinical failure and long-term

70 treatment 10.

71 In addition to the complex control of this growing public health problem, emerging

72 resistance to at least 1 class of existing antifungal agents and the poor clinical response of

73 many isolates to antifungal therapy push to raise the warning about the potential danger of

74 such a situation. Antifungal resistance remains universal although it may vary depending on

75 the species, geography and available therapeutic alternatives 9.

76 Faced to all these difficulties for the management of invasive fungal infections,

77 including; limited number of antifungal drugs, several side effects of current antifungals,

78 emerging of MDR-fungal strains and clinical failure, we have three main situations to

79 consider, in order to establish an immediate and appropriate measures to adopt. Firstly, it is

80 often claimed that reducing the prescription and consumption of antimicrobials could

81 overcome the development of resistance mechanisms, however, the use of antifungal agents to

82 treat 75% of women with vulvovaginal infection caused by Candida species 11, and nearly 3

83 million chronic lung infections induced by Aspergillus 1, is unavoidable. In addition, cases of

84 azoles-resistant aspergillosis have been detected on 4 continents where it was suggested that it

85 would be controlled by the use of azoles in agriculture 12 and recent study described evolution

86 of drug resistance in a naïve-antifungal patient with chronic Candida infection 13. Therefore,

87 reducing antifungals consumption will not improve the difficult management of invasive

88 fungal diseases.

89 Secondly, development of antifungal drugs represents a major challenge for the

90 pharmaceutical industry, since fungi are eukaryotic organisms, which implies the existence of

91 a close evolutionary relationship with their human hosts 14. Many studies have shown that the

92 drug development process is very expensive and many factors impact drug improvement, but

93 time and risk are the two most important determinants of this strategy 15. Consequently, the

94 development of new molecules to substitute current drugs will take more time before its use to

95 treat patients. Finally, we dispose of nearly 3732 drugs approved to treat various diseases. So,

96 identifying new use of investigational drugs that are outside of the primary medical

97 indication, commonly known as « drug-repurposing » can be faster, cheaper and more patient

98 secure with fewer side effects 16. Subsequently, the aim of this review, is to report the non-

99 traditional drugs already repositioned as antifungals, active against the most common

100 emerging multidrug-resistant fungi, in order to facilitate access to documentation of possible

101 therapeutic alternatives for invasive fungal infections treatment. We also discuss the necessity

102 of laboratory diagnosis for fungal infections and the several constraints encountered.

103 Does the antifungal susceptibility testing have a part in clinical fungal practice?

104 Microbiological resistance is described as the non-sensitivity of the isolate, to an antifungal

105 agent tested in vitro, using the Clinical and Laboratory Standards Institute (CLSI) or

106 European Committee on Antimicrobial Susceptibility Testing (EUCAST) protocols, where

107 the MIC obtained remain higher than breakpoint values descried for each species. It can be

108 intrinsic (primary) without prior exposure to the drug or acquired (secondary) due to a

109 pressure of selection applied by previous treatment with the antifungal agent tested.

110 In the other side, clinical resistance is described as the absence of clinical patient

111 improvement and in vivo treatment failure when antifungal agent which exhibited in vitro

112 efficacy against the fungal causative pathogen was used 6,17. Therefore, clinical and

113 microbiological resistance are two completely unrelated concepts because clinical response

114 cannot be predicted on the basis of in vitro antifungal susceptibility testing 17. The latter

115 finding may be supported by the “90-60 rule” described by EUCAST where sensitive

116 infections respond to adequate therapy about 90% of the time, while resistant infections only

117 respond 60% of the time. However, patient-specific factors and pharmacokinetic parameters

118 have a considerable influence on the clinical success determination 18. The antifungal

119 susceptibility test is a very active area of medical mycology, basing on the Minimum

120 Inhibitory Concentration (MIC) concept, performed using several routine daily practices as E-

121 test, commercial Sensititre YeastOne plates, Fungitest or Vitek and broth microdilution, as

122 outlined by CLSI and EUCAST protocols. The ability to generate MICs is of limited

123 importance without the support of the relevant interpretation and ability to evaluate its

124 significance in clinical terms. There is a significant variability in results obtained with the

125 various techniques (e.g E-test was more successful in detecting the amphotericin B-resistant

126 isolates compared to the reference microdilution method) 19, then clinicians are still finding

127 difficulties in the interpretation of antifungal susceptibility results because for some species

128 such as Candida and Aspergillus, breakpoints values are determined (Table 1) but most MICs

129 obtained for other uncommon and/or less frequent species, are arbitrarily interpreted 20. Al-

130 Hatmi et al have shown that, Fusarium strain isolated from a case of mycetoma was found to

131 have elevated MICs of itraconazole, but the itraconazole improved and completely cured the

132 patient 21 and in another study, it was shown that for patients suffering from Candidaemia, no

133 clear gradient in clinical failure frequency was observed through the increase in fluconazole

134 MIC values 22. Consequently, it is worth to emphasize that the predictability of clinical

135 success or failure, in response to the administration of a specific antifungal agent does not

136 depend only on MIC for the causative agent, so, performing antifungal susceptibility testing is

137 not required for every specific clinical fungal strain in standard use but only in some

138 conditions including; (i) epidemiological studies to establish the sensitivity profiles and drug

139 resistance rates of infection strains against commonly used antifungals in a particular centre

140 for efficient choice of antifungal empirical regimen, (ii) assessment of antifungal potency of

141 new developed compound 23, (iii) investigation of sensitivity profiles of Candida strains

142 isolated from candidemia as well as deep-seated Candida infections in order to select the most

143 efficient long-term therapy to use, (iv) assessment of susceptibility profiles of fungal strains

144 isolated from recurrent mucosal infections for efficient choice of alternative treatment

145 modalities 24 especially in immunocompromised patients where it was described that

146 treatment failure in patients suffering from HIV-infection with oropharyngeal and

147 oesophageal candidiasis, was often predicted by in vitro resistance test 25.

148 II. Current antifungal agents: mechanisms of resistance

149 Polyenes: the first antifungal agent developed in 1950s, for use in clinical practices

150 with a large spectrum of activity against yeast and filamentous fungi. The two most clinically

151 relevant members of this class are nystatin and amphotericin B (amB) but, pimaricin can also

152 be used 26. Amphotericin B is formulated as original desoxycholate and as 2 lipid-based

153 formulas: liposomal amphotericin B (L-AmB) and amphotericin B lipid complex (ABLC).

154 Polyenes are macrocyclic organic molecules, a natural products of an actinomycete of the soil,

155 Streptomyces nodosus 27.

156 Mechanisms of polyene-resistance: resistance to polyenes is mainly related to the changes in

157 the lipid structure of the membrane and subsequently a modification in its fluidity and

158 absorbency. The principal altered effects implicated in polyene resistance relate to enzymes

159 that are involved in the synthesis of ergosterol. Deficiencies of ERG2 and ERG3, which code

160 for the isomerase of C-8 sterol and 15,6-desaturase, respectively, induce quantitative and

161 qualitative modiﬁcations in membrane sterols, which affect, consequently, the quantity and

162 availability of ergosterol for polyene activity. Boosted activity of catalases, which decrease

163 oxidative injury represent another prospective polyene-resistance mechanism 28.

164 Amphotericin B-resistance is frequently described for many Aspergillus strains, and this is not

165 mediated by the alteration pathway of ergosterol content, as previously described. One of the

166 suggested ways to A. terreus to be resistant is to inhibit the Ras signal pathway by Hsp90 and

167 Hsp70, thus causing interruption in aqueous spores formation 29 (Table 2).

168 Azoles: are the best conventional antifungal agents in medical treatment today,

169 discovered in 1980s. They had a major effect on the treatment of invasive fungal infections,

170 over the last 35 years.

171 The use of the first-generation imidazoles including ketoconazole, miconazole, clotrimazole is

172 restricted primarily to the treatment of superficial fungal diseases. These were substituted with

173 the second-generation triazoles as fluconazole and itraconazole, to join a wide variety of

174 applications. Later, to surmount some constraints of efficacy and to prevent resistant

175 pathogens, third-generation azoles including voriconazole, posaconazole, efinaconazole and

176 isavuconazole have been established with extended range interest 30.

177 Mechanisms of Azoles-resistance: Four mechanisms of resistance to azoles have been

178 demonstrated, and there is no predominance between these mechanisms. (i) Reduction of

179 azoles plasmatic concentrations by surexpression of membrane-associated transporters

180 performing a multidrug efflux pumps function, encoded by the CDR gene families 17,31. (ii)

181 Target site changes induced by mutation in ERG11 gene encoding for the lanosterol C14a-

182 demethylase, resulting in blockage in azoles binding to appropriate enzymatic location. (iii)

183 Up-regulation of enzyme target, inducing an increase in the intracellular concentration of

184 ERG11p. It may be reached across gene amplification, and both enhanced transcription level

185 or reduced gene product degradation. (iv) Reduction of a toxic product 14a-methy-3,6-diol

186 amount obtained usually from 14a-methylfecosterol. This mechanism, named bypass pathway

187 development, is induced by mutation on the ERG3 gene. Thus, functional membrane was

188 conserved with inhibition of azoles action on biosynthesis of ergosterol pathway 17 (Table 2).

189 Echinocandins: are semisynthetic cyclic lipopeptides produced by the fermentation

190 broth of several fungi, discovered in early 1974. Echinocandins, including essentially

191 caspofungin, anidulafungin and micafungin, have been an essential adjunct to the growing

192 arsenal of products against invasive fungal diseases. They exert in vitro and in vivo fungicidal

193 action against most Candida species and fungistatic activity against Aspergillus 32.

194 Mechanisms of echinocandins-resistance: Candida resistance to echinocandins, have been

195 related to several mutation points in FKS gene encoding for component of β-GS enzyme,

196 especially in two points “hot-spot” zones, HS1 and HS2 of FKS1 gene. It was described that

197 the main and apparent catalytic subunit of β-GS is encoded by the FKS1 gene. This process of

198 resistance has similarly been shown in C. albicans and non-albicans Candida species.

199 Mutations in FKS gene modify the glucan synthase enzyme kinetics inducing a significant

200 increase in the inhibitory concentration 33 (Table 2).

201 5-fluorocytosine (5-FC): a pyrimidine analogue, was developed in 1957 as an

202 antimetabolite. While it has not identified any potential as an antitumor therapy, it is currently

203 used for the treatment of specific fungal infections, by inhibiting RNA and DNA synthesis.

204 Mechanisms of 5FC-resistance: Flucytosine is infrequently used as monotherapy because of

205 the frequent and common development of resistance9. A specific point mutation in FUR1

206 gene encoding uracil phosphoribosyl transferase, inducing a high resistance to ﬂucytosine and

207 5-ﬂuorouracil in various fungi species. Additionally, other points mutation in the genes FCY1

208 and FCY2 which encode cytosine deaminase enzyme, resulting in Candida spp ﬂucytosine

209 resistance 34 (Table 2).

210 IV. Repurposed new off-label drugs as antifungal compounds against fungal

211 pathogens

212 Resistance of clinically relevant causative organisms to approved systemic antifungal

213 compounds has been reported, although the incidence of emergence differs across drug

214 classes 35. Resistance to at least one class of antifungal agents is a concern of most existing

215 fungal species; however, some major pathogens have a relatively high resistance rate and

216 constitute serious public health, especially, Candida albicans, Cryptococcus spp and

217 Aspergillus spp. In addition, other life-threatening and emerging pathogens, including not

218 previously well-identified/characterized species and opportunistic multidrug-resistant ones,

219 are increasingly reported. These include Candida auris, Scedosporium/Lomentospora spp.,

220 Fusarium spp. and other Mucorales. Drug repositioning or repurposing approaches allow new

221 indications for previously approved drugs that are marketed for other conditions. This

222 approach offers many benefits over de novo drug development. For the pharmaceutical

223 market, the pharmacokinetic and pharmacodynamic profiles previously established provide

224 acceleration with reduced costs in the development pipeline 36 .

225 IV.1. Antimicrobial repurposed drugs

226 Polymyxins antibiotics: are cyclic, positively charged peptides, obtained naturally from

227 Gram-positive bacteria, such as Paenibacillus polymyxa. Among polymyxin molecules

228 described, two have been used in clinical settings: polymyxin B (PMB) and polymyxin E

229 (colistin) 37. Almost at high concentrations, MICs ranging from 16 to 128 µg/ml, polymyxins

230 antibiotics showed antifungal activity against several MDR- yeasts and molds, except for

231 Aspergillus species (Table 3). The authors demonstrated a fungicidal activity of colistin

232 against Candida albicans, Cryptococcus neoformans and Rhodotorula mucilaginosa, with

233 minimum fungicidal concentrations ranging from 2 to 4 times MICs. Colistin showed to

234 induce membrane damages on MDR-C. albicans38.

235 In another hand, using checkerboard microdilution assay, synergistic activity was showed

236 with colistin-amphotericin B and colistin-itraconazole associations, against Candida albicans

237 and Lichtheimia corymbifera strains, respectively. In addition, colistin-fluconazole

238 combination was described to induce synergistic activity against Rhodotorula mucilaginosa

239 yeast38.

240 Ribavirin (RBV): is a purine nucleoside analogue that produces broad-spectrum activity

241 against many RNA and DNA viruses. Ribavirin is used to treat hepatitis C virus (HCV) in

242 combination with interferon-α 39. Tournu et al., recently identified, ribavirin as a potential C.

243 albicans vacuole disrupting agent 40. After that, another study demonstrated fungistatic

244 activity of RBV against Candida albicans including MDR strains and fungicidal activity

245 against C. parapsilosis. The authors showed RBV MIC ranging mostly from 0.37 to 3.02

246 µg/ml against C. albicans, C. parapsilosis and C. tropicalis. (Yousfi et al, 2019). Synergistic

247 activity was observed when ribavirin was combined with either amphotericin B, fluconazole

248 or itraconazole, against MDR-C. albicans.

249 Oxyclozanide: is an halogenated salicylanilide, widely used as antiparasitic veterinary drugs

250 against helminths and ectoparasites 41. Pic et al, showed that oxyclozanide exhibited 58% and

251 99% growth inhibition against C. albicans at 10 and 100 μM, respectively, as compared to the

252 control. The activity of oxyclozanide was assessed in sensitive, and azole- and echinocandin-

253 resistant clinical C. albicans isolates with diverse molecular resistance mechanisms, with

254 MICs values ranging from 16 to 32 µg/ml 42. The authors propose a mechanism of action by

255 uncoupling the mitochondrial electron transport from oxidative phosphorylation and

256 disturbing the mitochondrial membrane potential.

257 Chloroquine: is an important antimalarial drug, used for this indication since 1940. Webber

258 and co-workers documented that treatment of macrophages parasitized by cryptococcal cells,

259 with chloroquine leads to accumulation of this drug inside macrophages forming iron-

260 complexes that kill cryptococcal cells. It is also, shown to be effective against drug-resistant

261 C. albicans biofilms 43. In addition, adding chloroquine in combined therapy with drugs as

262 azoles, biofilms became susceptible to these conventional antifungals. Further, Shinde et al

263 showed that chloroquine induce inhibition of normal ergosterol synthesis inducing inhibition

264 of morphogenesis of Candida albicans strains. Furthermore, Huang et al, reported using

265 functional genomics approach that chloroquine inhibits the thiamine transporters Thi7, Nrt1,

266 and Thi72 in yeasts. Thiamine plays crucial role in glucose metabolism, where cells can

267 derive energy to support cellular processes 44.

268 Quinacrine: a water-soluble acridone derivative, was widely used for the prevention and

269 treatment of malaria during World War II and remains available as a highly active therapeutic

270 agent against giardiasis. Quinacrine in monotherapy, has been shown in vitro to be eﬀective

271 for the prevention and treatment of Candida albicans bioﬁlms with MICs range of 64- 256

272 µg/ml. In addition, combination of quinacrine with fluconazole against mature biofilms had

273 no interaction (FICI=1.25). However, both amphotericin B and caspofungin had synergy with

274 quinacrine against mature biofilm, with FICI equal to 0.37 and 0.31 respectively.

275 Benzimidazoles (mebendazole, albendazole, flubendazole, and triclabendazole): are a broad-

276 spectrum nematicide and taenicide anthelmintic drugs. Joffe et al., demonstrated

277 benzimidazoles efficacy in inhibition of the growth of C. neoformans. They showed that the

278 most efficient compounds were mebendazole and flubendazole with same MICs values of

279 0.3125 µM. They focused their study on mebendazole, because it efficiently penetrates the

280 brain in animal models 45. Mebendazole has a fungicidal activity against C. neoformans

281 associated to its action against intracellular C. neoformans 46.

282 IV.2. Anti-inflammatory repurposed drugs

283 Aspirin and Ibuprofen: are commonly used to ameliorate fever and other symptoms of

284 different illness. Recently, Ogundeji et al repurposed aspirin and ibuprofen as alternative

285 drugs to control the growth of cryptococcal cells. They showed that C. neoformans strains

286 were more sensitive than C. gattii strains and ibuprofen had a greater inhibitory effect than

287 aspirin on all strains at each drug concentration tested 47. Other study demonstrated that the

288 effects of ibuprofen are dose-dependent, where at high concentration (10 mg/ml) Candida

289 cells are killed, whereas at a low concentration, 5 mg/ml the drug was fungistatic. The

290 combined effect of Ibuprofen and fluconazole did not yield total cryptococcal growth

291 inhibition. Nonetheless, synergistic outcomes were observed. These two drugs were able to

292 effect synergism at concentrations that were lower than their individually deﬁned MICs 47.

293 Ibuprofen has also been shown to act in synergy with fluconazole when tested in combined

294 therapy against Candida albicans with FIC < 0.5 48.

295 In addition, the combined effect of Ibuprofen and AMB did not yield total cryptococcal

296 growth inhibition. Nonetheless, synergistic outcomes were observed. These two drugs were

297 able to effect synergism at concentrations that were lower than their individually deﬁned

298 MICs 47.

299 Auranofin: Inhibits several inﬂammatory pathways and has been in clinical use since 1985 as

300 a disease-modifying antirheumatic drug used to slow down or stop the progression of this

301 rheumatic disorder. Wiederhold et al showed that auranoﬁn displays activity against all

302 Candida albicans isolates tested with MIC values <1 µg/ml 49.

303 IV.3. Antipsychotic drugs

304 Haloperidol/Trifluperidol: are an FDA-approved oral antipsychotic that were recently

305 discovered to possess antifungal activity towards a drug-sensitive C. albicans strains with

306 MICs values < 4 µg/ml. The authors demonstrated that the two antipsychotic have a very

307 similar effect as fluconazole 50. Combination of Bromperidol-derivative with Posaconazole

308 (POS) had a lowest FICI value of 0.13, which showed decrease of MIC alone of azole and

309 antipsychotic compound from >32 and >128 µg/mL to 2 and 8 µg/mL (16-fold reduction in

310 MIC values for both drugs), respectively. The authors demonstrated that the combinations of

311 either posaconazole or voriconazole and bromperidol derivatives could synergistically inhibit

312 the growth of an azole-resistant C. albicans that otherwise would not have responded to high

313 concentrations of either drugs given in the assay. In addition, strong synergy was observed

314 between bromperidol derivative with POS against C. glabrata and Aspergillus terreus 51.

315 Sertraline: are an anti-psychotic agent. Sertraline has been reported to be fungicidal against

316 C. neoformans 52,53. Zhai et al showed sertraline antifungal potential against several clinical

317 isolates of C. neoformans (MIC 2-6 µg/mL). These authors also showed in vivo antifungal

318 activity in a murine model of cryptococcosis since sertraline reduces the fungal burden. The

319 patients treated with sertraline had faster cryptococcal cerebrospinal fluid clearance and lower

320 relapse rates than those reported in the past 54. The fungicidal combination of sertraline-

321 fluconazole was effective against all species tested, including drug-resistant clinical isolates

322 of Candida, and in an in vivo model of C. neoformans infection in Galleria mellonella. In

323 another hand, Nayak et al tested the sertraline-fluconazole combination against 53 C.

324 neoformans strains belonging to several serotypes. The authors showed that a large number of

325 isolates (31/53) responded with synergy (FICI ≤ 0.5) to the sertraline/fluconazole combination

326 55. In addition, Rossato et al showed the synergistic activity of sertraline-AMB association

327 against a large panel of C. neoformans strains 56.

328 Chlorpromazine and Triﬂuoperazine: are dopamine antagonist and work by acting on a

329 number of receptors in the brain 57. Schizophrenia is the most common medical indication of

330 these drugs. Vitale et al, reported the antifungal properties of these phenothiazine compounds

331 against Aspergillus species (A. fumigates, A. ustus, A. terreus) zygomycetes (Absidia

332 corymbifera, Rhizopus oryzae, R. Microspores), and Scedosporium species (S. apiospermum,

333 S. prolificans). Both drugs inhibited the growth of all fungi tested at concentrations of 16 to

334 64 µg/ml 58. The combination of AMB with chlorpromazine against C. neoformans

335 demonstrated a predominant synergism profile. Thirty clinical isolates of C. neoformans var.

336 grubii isolated from cerebrospinal fluid (CSF) of HIV-positive patients were tested in this

337 study in presence or absence of induced polysaccharide capsule56.

338 IV.4. Miscellaneous drugs repurposed

339 Atorvastatin/Simvastatin: are a statins, major lowering-cholesterol drugs. Macreadie et al,

340 demonstrated a strong inhibition of growth of C. albicans tested, with MIC=100 µM 59.

341 Tamoxifen and Toremifene: are estrogen receptor antagonists triphenylethylenes, usually

342 used in breast cancer treatment. Butts et al. demonstrated fungicidal activity of these two

343 compounds against C. neoformans. An important facet of C. neoformans pathogenesis is its

344 ability to replicate within the phagolysosomes of macrophages, thus, the authors showed

345 tamoxifen and toremifene play an important antifungal activity against C. neoformans within

346 macrophages 60. Tamoxifen was synergistic with AMB (FICI, 0.5; 2 g/ml tamoxifen–0.25

347 g/ml AMB). In another hand, the combination of toremifene (2 µg/ml) and FLU (2 µg/ml)

348 inhibited all cryptococcal growth, giving a fractional inhibitory concentration index of 0.50.

349 In vivo, 200 mg/kg of body weight per day of tamoxifen reduced kidney fungal burden (1.5

350 log10 CFU per g tissue; P 0.008) in a murine model of disseminated candidiasis. In addition,

351 tamoxifen is a known inhibitor of mammalian calmodulin, and tamoxifen-treated yeast show

352 phenotypes consistent with decreased calmodulin function, including lysis, decreased new

353 bud formation, disrupted actin polarization, and decreased germ tube formation 61.

354 Disulfiram: is an alcohol antagonist drug used in clinical practices for many years. Khan et

355 al, reported its antifungal potential. The MIC 50 and MIC 90 of disulfiram for yeast isolates is

356 4 and 8 μg/ml, respectively, and the MIC range is 1-16 μg/ml for both fluconazole sensitive

357 and resistant strains. Interestingly, disulfiram also showed fungicidal activity on Aspergillus

358 spp. with MIC 50 and MIC 90 of 2 and 8 μg/ml, respectively 62. Shukla et al demonstrated

359 that disulfiram reverses Cdr1p-mediated drug resistance by interaction with both ATP and

360 substrate-binding sites of MDR-Candida transporter and may be useful for antifungal therapy

361 63.

362 V. Conclusion

363 The rise in the emergence of new mycotic agents and drug resistance drives the need to find

364 new and/or alternative drugs. Today, a picture is emerging where non-traditional

365 antimicrobial agents, which were previously prescribed to treat non-infectious conditions,

366 show antimicrobial properties. It is worth our time to explore these compounds further. In this

367 context, we presented reports where compounds ranging from anti-inflammatory drugs to

368 anti-psychotic drugs are documented to control fungal growth. However, due to toxicities

369 attributed to some of these drugs, their clinical application may be limited. It is also important

370 to demonstrate if these ‘new’ antifungals will work in host cells, and probably in some cases,

371 in situations of advanced disease. Towards this end, the usage of animals in modelling human

372 disease should be considered in order to clearly establish the therapeutic benefits of these

373 chemical compounds.

374 An important note is that some of these compounds in in vitro assays act in synergy with

375 traditional antimicrobial agents, such as azoles. This is one attribute that is encouraging in that

376 it implies these compounds can be added at low concentrations where clinically they could

377 not exert any negative physiological outcomes. Additionally, the added effect of these

378 chemical compounds in combination therapy enhances the efficiency of traditional

379 antimicrobial drugs, which under normal conditions may be fungistatic. We should point out

380 that careful consideration must be taken when attempting repurposing approach, especially in

381 the case of compounds that generally target eukaryotic organisms and drugs that can induce

382 immunosuppression, in order to realize the desired therapeutic outcome to the exclusion of

383 adverse effects.

384 Acknowledgments

385 We thank CookieTrad for English reviewing. 386 387 Funding information

388 This work was supported by the French Government under the «Investissements d’avenir »

389 (Investments for the Future) program managed by the Agence Nationale de la Recherche

390 (ANR, fr: National Agency for Research), (reference: Méditerranée Infection 10-IAHU-03)

391 This work was supported by Région Provence-Alpes-Côte d’Azur and European funding FEDER PRIM

392 Transparency declarations

393 No conflict of interest or financial disclosure to declare for all authors.

394 References

395 1. Gordon D. Brown, DavidW. Denning, Neil A.R. Gow, StuartM. Levitz, Mihai G.

396 Netea, Theodore C. White. Hidden killers: human fungal infections. Sci Transl Med.

397 2012;4(165):1-9. doi:10.1126/scitranslmed.3004404

398 2. Denning DW. Calling upon all public health mycologists. Eur J Clin Microbiol Infect

399 Dis. 2017;36(6):923-924. doi:10.1007/s10096-017-2909-8

400 3. Srinivasan A, Lopez-ribot JL, Ramasubramanian AK. Drug resistance Overcoming

401 antifungal resistance. Drug Discov Today Technol. 2014;11:65-71.

402 doi:10.1016/j.ddtec.2014.02.005

403 4. Prinsloo G, Dobson R, Brent A, Mammoli A. The epidemiology of fungal infections in

404 patients with hematologic malignancies: the SEIFEM-2004 study Livio. AIP Conf

405 Proc. 2016;1734:1068-1075. doi:10.1063/1.4949226

406 5. Armstrong-James D, Meintjes G, Brown GD. A neglected epidemic: Fungal infections

407 in HIV/AIDS. Trends Microbiol. 2014;22(3):120-127. doi:10.1016/j.tim.2014.01.001

408 6. Alcazar-fuoli L, Mellado E. Current status of antifungal resistance and its impact on

409 clinical practice. Br J Haematol. 2014;166(April):471-484. doi:10.1111/bjh.12896

410 7. Meurman JH, Hämäläinen P, Janket SJ. Oral infections in older adults. Aging health.

411 2006;2(6):1013-1023. doi:10.2217/1745509X.2.6.1013

412 8. Gupta R, Goel N, Gupta A, Gupta KB, Chaudhary U, Sood S. A rare fungal infiltration

413 of lungs in a healthy young girl. Case Rep Pulmonol. 2011;2011.

414 doi:10.1155/2011/917089

415 9. Geddes-Mcalister J, Shapiro RS. New pathogens, new tricks: Emerging, drug-resistant

416 fungal pathogens and future prospects for antifungal therapeutics. Ann N Y Acad Sci.

417 2018;1435(1):57-78. doi:10.1111/nyas.13739

418 10. Perfect JR. The antifungal pipeline: a reality check HHS Public Access.

419 2018;16(9):603-616. doi:10.1038/nrd.2017.46

420 11. Denning DW, Kneale M, Sobel JD, Rautemaa-Richardson R. Global burden of

421 recurrent vulvovaginal candidiasis: a systematic review. Lancet Infect Dis.

422 2018;18(11):e339-e347. doi:10.1016/S1473-3099(18)30103-8

423 12. Kontoyiannis DP. Antifungal resistance: An emerging reality and a global challenge. J

424 Infect Dis. 2017;216(Suppl 3):S431-S435. doi:10.1093/infdis/jix179

425 13. Demers EG, Biermann AR, Masonjones S, et al. Evolution of drug resistance in an

426 antifungal-naive chronic Candida lusitaniae infection. Proc Natl Acad Sci.

427 2018;115(47):12040-12045. doi:10.1073/pnas.1807698115

428 14. Hall RA, Noverr MC. Fungal interactions with the human host: exploring the spectrum

429 of symbiosis. Curr Opin Microbiol. 2017;40:58-64. doi:10.1016/j.mib.2017.10.020

430 15. Dimasi JA, Feldman L, Seckler A, Wilson A. Trends in risks associated with new drug

431 development: Success rates for investigational drugs. Clin Pharmacol Ther.

432 2010;87(3):272-277. doi:10.1038/clpt.2009.295

433 16. Pushpakom S, Iorio F, Eyers PA, et al. Drug repurposing: progress, challenges and

434 recommendations. Nat Rev Drug Discov. 2018. doi:10.1038/nrd.2018.168

435 17. Kanafani ZA, Perfect JR. Resistance to antifungal agents: Mechanisms and clinical

436 impact. Chinese J Infect Chemother. 2010;10(4):320. doi:10.1086/524071

437 18. Rex JH, Pfaller MA. Has Antifungal Susceptibility Testing Come of Age ? DEGREE

438 OF CORRELATION BETWEEN IN VITRO AND IN. Clin Infect Dis. 2002;35:982-

439 989.

440 19. Peyron F, Favel A, Michel-Nguyen A, Gilly M, Regli P, Bolmström A. Improved

441 detection of amphotericin B-resistant isolates of Candida lusitaniae by Etest. J Clin

442 Microbiol. 2001;39(1):339-342. doi:10.1128/JCM.39.1.339-342.2001

443 20. Rex JH, Pfaller MA, Walsh TJ, et al. Antifungal susceptibility testing: Practical aspects

444 and current challenges. Clin Microbiol Rev. 2001;14(4):643-658.

445 doi:10.1128/CMR.14.4.643-658.2001

446 21. Al-Hatmi A, Curfs-Breuker I, de Hoog G, Meis J, Verweij P. Antifungal Susceptibility

447 Testing of Fusarium: A Practical Approach. J Fungi. 2017;3(2):19.

448 doi:10.3390/jof3020019

449 22. Fernández-Ruiz M, Guinea J, Lora-Pablos D, et al. Impact of fluconazole susceptibility

450 on the outcome of patients with candidaemia: data from a population-based

451 surveillance. Clin Microbiol Infect. 2017;23(9):672.e1-672.e11.

452 doi:10.1016/j.cmi.2017.01.014

453 23. Arikan S. Current status of antifungal susceptibility testing methods. Med Mycol.

454 2007;45(7):569-587. doi:10.1080/13693780701436794

455 24. Hospenthal DR, Murray CK, Rinaldi MG. The role of antifungal susceptibility testing

456 in the therapy of candidiasis. Diagn Microbiol Infect Dis. 2004;48(3):153-160.

457 doi:10.1016/j.diagmicrobio.2003.10.003

458 25. Walsh TJ, Gonzalez CE, Piscitelli S, et al. Correlation between in vitro and in vivo

459 antifungal activities in experimental fluconazole-resistant oropharyngeal and

460 esophageal candidiasis. J Clin Microbiol. 2000;38(6):2369-2373.

461 26. Rogers TR. Antifungal drug resistance: Does it matter? Int J Infect Dis.

462 2002;6(SUPPL. 1). doi:10.1016/S1201-9712(02)90154-2

463 27. Nett JE, Andes DR. Antifungal Agents: Spectrum of Activity, Pharmacology, and

464 Clinical Indications. Infect Dis Clin North Am. 2016;30(1):51-83.

465 doi:10.1016/j.idc.2015.10.012

466 28. Scorzoni L, de Paula e Silva ACA, Marcos CM, et al. Antifungal therapy: New

467 advances in the understanding and treatment of mycosis. Front Microbiol. 2017;8:DOI:

468 10.3389/fmicb.2017.00036. doi:10.3389/fmicb.2017.00036

469 29. Blatzer M, Blum G, Jukic E, et al. Blocking Hsp70 enhances the efficiency of

470 amphotericin B treatment against resistant Aspergillus terreus strains. Antimicrob

471 Agents Chemother. 2015;59(7):3778-3788. doi:10.1128/AAC.05164-14

472 30. Allen D, Wilson D, Drew R, Perfect J. Azole antifungals: 35 years of invasive fungal

473 infection management. Expert Rev Anti Infect Ther. 2015;13(6):787-798.

474 doi:10.1586/14787210.2015.1032939

475 31. Prasad Dr R, Rawal MK. Efflux pump proteins in antifungal resistance. Front

476 Pharmacol. 2014;5:1-13. doi:10.3389/fphar.2014.00202

477 32. Aguilar-Zapata D, Petraitiene R, Petraitis V. Echinocandins: The Expanding

478 Antifungal Armamentarium. Clin Infect Dis. 2015;61(6):604-611.

479 doi:10.1093/cid/civ814

480 33. Pfaller MA. Antifungal Drug Resistance : Mechanisms , Epidemiology , and

481 Consequences for Treatment. Am J Med. 2012;125(1):S3-S13.

482 doi:10.1016/j.amjmed.2011.11.001

483 34. Edlind TD& SKK. Mutational analysis of ﬂucytosine resistance in Candida glabrata.

484 Antimicrob Agents Chemother. 2010;54:4733–4738.

485 35. Robbins N, Caplan T, Cowen LE. Molecular Evolution of Antifungal Drug Resistance.

486 Annu Rev Microbiol. 2017;71(1):753-775. doi:10.1146/annurev-micro-030117-020345

487 36. Chigaev A, Hudson L, Thompson T, et al. Drug repurposing from an academic

488 perspective. Drug Discov Today Ther Strateg. 2011;8(3-4):61-69.

489 doi:10.1016/j.ddstr.2011.10.002

490 37. Biswas S, Brunel JM, Dubus JC, Reynaud-Gaubert M, Rolain JM. Colistin: An update

491 on the antibiotic of the 21st century. Expert Rev Anti Infect Ther. 2012;10(8):917-934.

492 doi:10.1586/eri.12.78

493 38. Yousfi H, Ranque S, Rolain J, Bittar F. In vitro polymyxin activity against clinical

494 multidrug-resistant fungi. 2019;7:1-10.

495 39. Te HS, Randall G, Jensen DM. Mechanism of action of ribavirin in the treatment of

496 chronic hepatitis C. Gastroenterol Hepatol (N Y). 2007;3(3):218-225.

497 doi:10.1059/0003-4819-155-7-201110040-00017

498 40. Tournu H, Carroll J, Latimer B, et al. Identification of small molecules that disrupt

499 vacuolar function in the pathogen Candida albicans. PLoS One. 2017;12(2):1-21.

500 doi:10.1371/journal.pone.0171145

501 41. Swan GE. The pharmacology of halogenated salicylanilides and their anthelmintic use

502 in animals : review article. J S Afr Vet Assoc. 2012;70(2):61-70.

503 doi:10.4102/jsava.v70i2.756

504 42. Pic E, Sellam A. OUP accepted manuscript. Zool J Linn Soc. 2019:1-4.

505 doi:10.1093/zoolinnean/zly093

506 43. Shinde RB, Rajput SB, Raut JS, Karuppayil SM. An in vitro repositioning study

507 reveals antifungal potential of chloroquine to inhibit growth and morphogenesis in

508 Candida albicans. J Gen Appl Microbiol. 2013;59(2):167-170.

509 doi:10.2323/jgam.59.167

510 44. Huang Z, Srinivasan S, Zhang J, et al. Discovering Thiamine Transporters as Targets of

511 Chloroquine Using a Novel Functional Genomics Strategy. PLoS Genet. 2012;8(11):4-

512 13. doi:10.1371/journal.pgen.1003083

513 45. Bai RY, Staedtke V, Wanjiku T, et al. Brain penetration and efficacy of different

514 mebendazole polymorphs in a mouse brain tumor model. Clin Cancer Res.

515 2015;21(15):3462-3470. doi:10.1158/1078-0432.CCR-14-2681

516 46. Joffe LS, Schneider R, Lopes W, et al. The anti-helminthic compound mebendazole

517 has multiple antifungal effects against Cryptococcus neoformans. Front Microbiol.

518 2017;8(MAR):1-14. doi:10.3389/fmicb.2017.00535

519 47. Ogundeji AO, Pohl CH, Sebolai OM. Repurposing of Aspirin and Ibuprofen as

520 Candidate Anti- Cryptococcus Drugs. Antimicrob Agents Chemother. 2016;60(8):4799-

521 4808. doi:10.1128/AAC.02810-15.Address

522 48. Scott EM, Tariq VN, McCrory RM. Demonstration of synergy with fluconazole and

523 either ibuprofen, sodium salicylate, or propylparaben against Candida albicans in vitro.

524 Antimicrob Agents Chemother. 1995;39(12):2610-2614. doi:10.1128/AAC.39.12.2610

525 49. Wiederhold NP, Patterson TF, Srinivasan A, et al. Repurposing auranofin as an

526 antifungal: In vitro activity against a variety of medically important fungi. Virulence.

527 2017;8(2):138-142. doi:10.1080/21505594.2016.1196301

528 50. Stylianou M, Kulesskiy E, Lopes JP, Granlund M, Wennerberg K, Urban CF.

529 Antifungal application of nonantifungal drugs. Antimicrob Agents Chemother.

530 2014;58(2):1055-1062. doi:10.1128/AAC.01087-13

531 51. Holbrook SYL, Garzan A, Dennis EK, Shrestha SK, Garneau-Tsodikova S.

532 Repurposing antipsychotic drugs into antifungal agents: Synergistic combinations of

533 azoles and bromperidol derivatives in the treatment of various fungal infections. Eur J

534 Med Chem. 2017;139:12-21. doi:10.1016/j.ejmech.2017.07.030

535 52. Rhein J, Nielsen K, Boulware DR, Meya DB. Sertraline for HIV-associated

536 cryptococcal meningitis – Authors’ reply. Lancet Infect Dis. 2016;16(10):1111-1112.

537 doi:10.1016/S1473-3099(16)30340-1

538 53. Zhai B, Wu C, Wang L, Sachs MS, Lin X. The antidepressant sertraline provides a

539 promising therapeutic option for neurotropic cryptococcal infections. Antimicrob

540 Agents Chemother. 2012;56(7):3758-3766. doi:10.1128/AAC.00212-12

541 54. Rhein J, Morawski BM, Hullsiek KH, et al. Efficacy of adjunctive sertraline for the

542 treatment of HIV-associated cryptococcal meningitis: An open-label dose-ranging

543 study. Lancet Infect Dis. 2016;16(7):809-818. doi:10.1016/S1473-3099(16)00074-8

544 55. Nayak R, Xu J. Effects of sertraline hydrochloride and fluconazole combinations on

545 Cryptococcus neoformans and Cryptococcus gattii. Mycology. 2010;1(2):99-105.

546 doi:10.1080/21501203.2010.487054

547 56. Santurio JM, Zanette RA, Chassot F, Alves SH, Rossato L, Loreto ÉS. In vitro

548 synergistic effects of chlorpromazine and sertraline in combination with amphotericin

549 B against Cryptococcus neoformans var. grubii. Folia Microbiol (Praha).

550 2016;61(5):399-403. doi:10.1007/s12223-016-0449-8

551 57. Beveridge A. The Creation of Psychopharmacology. J R Soc Med. 2017;95(11):568-

552 569. doi:10.1177/014107680209501118

553 58. Vitale RG, Afeltra J, Meis JFGM, Verweij PE. Activity and post antifungal effect of

554 chlorpromazine and trifluopherazine against Aspergillus , Scedosporium and

555 zygomycetes. Mycoses. 2007;50:270-276. doi:10.1111/j.1439-0507.2007.01371.x

556 59. Macreadie IG, Johnson G, Schlosser T, Macreadie PI. Growth inhibition of Candida

557 species and Aspergillus fumigatus by statins. FEMS Microbiol Lett. 2006;262(1):9-13.

558 doi:10.1111/j.1574-6968.2006.00370.x

559 60. Butts A, Koselny K, Chabrier-Roselló Y, et al. Estrogen receptor antagonists are anti-

560 cryptococcal agents that directly bind EF hand proteins and synergize with fluconazole

561 in vivo. MBio. 2014;5(1):1-11. doi:10.1128/mBio.00765-13

562 61. Dolan K, Montgomery S, Buchheit B, DiDone L, Wellington M, Krysan DJ.

563 Antifungal activity of tamoxifen: In vitro and in vivo activities and mechanistic

564 characterization. Antimicrob Agents Chemother. 2009;53(8):3337-3346.

565 doi:10.1128/AAC.01564-08

566 62. Khan S, Singhal S, Mathur T, Upadhyay DJ, Rattan A, Range MIC. Antifungal

567 Potential of Disulfiram. Med Mycol. 2007;48:109-113.

568 63. Shukla S, Sauna ZE, Prasad R, Ambudkar S V. Disulfiram is a potent modulator of

569 multidrug transporter Cdr1p of Candida albicans. Biochem Biophys Res Commun.

570 2004;322(2):520-525. doi:10.1016/j.bbrc.2004.07.151

571

MIC breakpoint (µg/ml) Strains Antifungals CLSI EUCAST SENSITITRE

Candida spp Amphotericin B 1 1 NI Anidulafungin 0.5 0.032 2 Caspofungin 0.5 / 2 Micafungin 0.5 0.016 2 Fluconazole 64 4 64 Itraconazole 1 0.064 1 Posaconazole NI 0.064 NI Voriconazole 0.5 0.25 4 Isavuconazole NI NI NI Flucytosine 32 NI 32 Aspergillus spp Amphotericin B NI 2 NI Itraconazole NI 2 NI Posaconazole 8 0.25 NI Voriconazole NI 2 NI Isavuconazole NI 1 NI

Table 1: Resistance interpretive breakpoints for in vitro susceptibility testing. Antifungal Mechanisms of action Clinical indications Side effects Mechanisms of resistance classes Polyenes: -Membrane -Endemic fungal infection -Phlebitis -Deficiencies in ERG2 and ERG3 genes. permeabilization by ion -Ergosterol synthesis alteration. Amphotericin B channels formation -Cryptococcal meningitis -Electrolyte defects - Quantitative and qualitative modiﬁcations Nystatin in membrane sterols. -Cell content leakage -Topical Candidal infections -Renal toxicity Pimaricin -Inhibition of RAS signal and interruption of spores’ formation (Aspergillus spp). Azoles: -Inhibition of cytochrome -All invasive candidiasis -Generic symptoms: -Surexpression of efflux pumps function. Fluconazole P450 nausea, abdominal pain, -Mutations in ERG11 gene inducing Itraconazole -Cryptococcal meningitis emesis, diarrhea blockage in azoles binding. -Alteration of fungal Voriconazole -Up-regulation of enzyme target. membrane fluidity and -Aspergillus fumigatus -Hepatotoxicity Posaconazole agility. infections -By-pass pathway development by ERG3 Efinaconazole gene mutation. Isavuconazole Echinocandins: -Inhibition of β-1,3-glucan synthase (β-GS) -Invasive candidiasis. -Hypersensitivity -Mutations on FKS gene (encoding β-GS). Micafungin -Interruption of fungal cell -Anaphylactic choc Caspofungin wall biosynthesis -Invasive aspergillosis -Modification of glucan synthase kinetics. Anidulafungin -Disruption of amino- -Cryptococcosis -Gastrointestinal affect. -Mutations on FUR1 gene (encoding uracil acetylation of tRNA -Hepatotoxicity phosphoribosyl transferase) 5-Flucytosine -Invasive candidiasis if -Bone marrow depression -inhibition of thymidylate treatment failure. -Mutations on FCY1 gene (encoding cytosine synthetase enzyme deaminase enzyme)

Table 2: Current antifungal agents: mechanisms of action, clinical indications, side effects and mechanisms of antifungal drugs resistance. Antifungal Drug Clinical activity MIC ranges Antifungal mechanism of action References activity

C. albicans C. neoformans R. mucilaginosa Polymyxins (Colistin and Gram-negative S. apiospermum 16-128 µg/ml Membrane damages on Candida albicans. Yousfi et al (2019) polymyxin B) bacterial infections L. prolificans F. oxysporum F. solani R. oryzae

C. albicans Disruption of vacuoles function and/or Yousfi et al (2019) Ribavirin Hepatitis C 0.37- 3.02 µg/ml C. tropicalis integrity of Candida albicans strains. Tournu et al (2017) C. parapsilosis Uncoupling the mitochondrial electron transport from oxidative phosphorylation Oxyclozanide Parasitosis C. albicans 16-32 µg/ml Pic et al (2018) and disturbing the mitochondrial membrane potential.

Shinde et al (2013) C. neoformans NI -Inhibition of normal ergosterol biosynthesis inducing morphogenesis disruption.

Chloroquine Malaria C. albicans 31.2-250 µg/ml -Fungal growth inhibition via blocking thiamine transportation. Huang et al (2012) Quinacrine Helminthiasis C. albicans 64-256 µg/ml Inhibition of ﬁlamentation Kulkarny et al (2014)

Morphological alterations by reducing Mebendazole Helminthiasis C. neoformans 0.03-0.31 µg/ml Joffe et al (2017) capsular dimension. C. gatii Candida spp 1-16 µg/ml Inhibition the oligomycin-sensitive ATPase Khan et al (2007) Disulfiram Alcoholism C. neoformans 1-2 µg/ml activity of Cdr1pv, multidrug transporter of Shukla et al (2004) Aspergillus spp 1-16 µg/ml Candida. C. albicans 4- >16 µg/ml S. apiospermum 2- >16 µg/ml Auranofin Rheumatoid arthritis Not reported Wiederhold et al (2017) L. prolificans 8-16 µg/ml C. neoformans >16 Reduction of the ergosterol content in the cell membrane and alteration of the Simvastatin C. albicans properties of the polysaccharide capsule; hypercholesterolemia 100 µM increase in the production of ROS by Ribeiro et al (2017) Atorvastatin C. gatii macrophages; and reduction of yeast phagocytosis and the intracellular proliferation rate.

Haloperidol/Trifluperidol Psychosis C. albicans < 4 µg/ml Not identified Stylianou et al (2014)

Finasteride Prostatic hyperplasia C. albicans NI Inhibition of ﬁlamentation Routh et al (2013)

Sertraline Psychosis C. neoformans 2-6 µg/ml Not identified Zhai et al (2012)

Tamoxifen /Toremifene Breast cancer C. albicans 8 to 64 µg/ml Binding to the two essential EF-hand Butts et al (2014) proteins calmodulin 1 (Cam1) and C. neoformans calmodulin-like protein (Cml) and Dolan et al (2009) prevention of Cam1 from binding to its well-characterized substrate calcineurin

C. neoformans Stress induction via activation of the high- osmolarity glycerol (HOG) pathway, and Ogundeji et al Aspirin/Ibuprofen Inflammation C. gatii 5-10 mg/ml activation of reactive oxygen species (ROS)- Pina Vaz et al (2000) Candida spp mediated membrane damage C. neoformans Aspergillus spp A. corymbifera Chlorpromazine Schizophrenia R. oryzae 1-16 µg/ml NI Vitale et al (2007) R. microspores S. apiospermum S. prolificans Chavez-Dozal et al Doxorubicin Cancers C. albicans NI Inhibition of filamentation (2014)

Table 3: Off-label drugs repurposed as antifungal agents.

Partie II

Identification de nouvelles molécules antimicrobiennes par criblage d’une chimiothèque de molécules « Prestwick Chemical

Library » de biodisponibilité connue et approuvée par la FDA

contre une collection de souches multi-résistantes.

Avant-propos

Le concept du repositionnement des médicaments permet d'identifier de nouvelles indications thérapeutiques pour les médicaments déjà existants sur le marché [13]. Ce concept constitue la principale approche appliquée dans notre étude qui permettra un gain de temps et une limitation plus rapide et efficace du développement de la résistance aux antimicrobiens traditionnels, afin d’assurer une meilleure prise en charge des patients.

La première étape du processus de repositionnement consiste à choisir parmi les différentes chimiothèques de molécules commercialisées, celle qui soit la plus adaptée pour la réalisation de nos tests in vitro. Dans un premier temps, une étude bibliographique a été réalisée sur un ensemble de chimiothèques, en se basant sur quelques critères de sélection entre autres ; l’innocuité des molécules. On a retenu la chimiothèque « Prestwick Chemical Library » qui contient 1280 molécules, toutes approuvées par la FDA « Food and Drug Administration » pour leur innocuité chez l’humain. La chimiothèque nous a été fournie sous forme de 16 plaques de

96-puits, chacune contenait 80 molécules déposées dans un ordre bien défini et à une concentration de 10 mM, diluées dans 100 µl de DMSO.

Dans un deuxième temps, nous avons mis au point le protocole à utiliser pour tester les

1280 molécules, selon les recommandations CLSI « Clinical and Laboratory Standard Institute

». Pour ce faire, on a évalué la sensibilité de 4 souches de contrôle qualité (Candida krusei

ATCC 6258, Candida parapsilosis ATCC 22019, Aspergillus flavus ATCC 204304, Aspergillus fumigatus ATCC 204305) au trois antifongiques conventionnels (itraconazole, amphotéricine

B, fluconazole), en utilisant trois méthodes différentes. Les résultats ont été obtenus sous forme de CMI, par trois tests de sensibilité aux antimicrobiens, notamment CMI en milieu liquide selon les recommandations CLSI, E-test et les plaques commercialisées Sensititre YeastOne.

Après analyse et comparaison des valeurs de CMI obtenues par les trois méthodes, une concordance a été constatée. De même, les CMI obtenues appartenaient aux limites de CMI recommandées pour les souches de références. Par conséquent, le protocole de test d’activité en milieu liquide selon les recommandations CLSI sera utilisé dans la suite de nos travaux.

Par la suite, un criblage primitif des 1280 molécules médicamenteuses à une concentration de 10 µM a été réalisé sur une collection de souches avec des profils de résistance différents. Les résultats ont été obtenus sous forme de densités optiques (DO) par un spectrophotomètre et les pourcentages d’inhibition de la croissance de l’agent mycosique par chacune des molécules testées ont été calculés. Le calcul des pourcentages d’inhibition a été réalisé en appliquant l’équation suivante : % de l’inhibition de la croissance fongique = (DO du contrôle positif - DO du puit test ) * 100 / (DO du contrôle positif – DO du contrôle négatif), en tenant compte du fait que le puit du contrôle négatif ne contient que le milieu de culture seul, le puit du contrôle positif contient la souche microbienne testée dans son milieu de culture approprié, et enfin, le puit test contient la souche fongique avec la molécule testée.

Le criblage primitif a été effectué sur une collection de souches de champignons (levures et filamenteux), provenant de l’hôpital La Timone de Marseille. Les souches ont été isolées de différents prélèvements cliniques, notamment le sang, le liquide d’aspiration bronchique et le liquide céphalo-rachidien. La prise en charge clinique des infections induites par ces souches

était difficile en raison de leurs profils de résistance associés à un échec thérapeutique.

Le test de sensibilité, in vitro, des 1280 molécules a été réalisé sur des champignons filamenteux émergeants, incluant les espèces suivantes : Aspergillus, Fusarium, Scedosporium,

Rhizopus et Lichtheimia. Toutes les souches présentaient une résistance à au moins une classe

52 d’antifongiques, notamment la résistance aux azolés et amphotéricine B. Seules les molécules qui ont présenté un pourcentage d’inhibition de la croissance fongique ≥ 70% ont été retenues.

Une large activité antifongique de quatre molécules a été discutée, notamment celle du clioquinole, alexidine dihydrochloride, hexachlorophene et thonzonium bromide. Ces drogues sont des médicaments antimicrobiens qui permettent avec toutes les autres molécules rapportées dans cette étude, d’enréchir le panel d’antifongiques potentiellement utilisables pour le traitement d’infections fongiques à germe invasifs (Article 2).

Le criblage primitif de la chimiothèque a été également réalisé sur des souches de levures multirésistantes notamment Candida albicans, Cryptococcus neoformans et Rhodotorula mucilaginosa et Candida auris et les résultats seront exposés dans les parties suivantes.

Article 2:

Identification of new antimycotic agents by screening of the current FDA-approved drugs against emerging invasive molds.

Hanane Yousfi, Stéphane Ranque, Jean marc Rolain and Fadi Bittar

Submitted to « Medical Mycology »

1 Title: Identification of new antimycotic agents by screening of the current FDA-approved

2 drugs against emerging invasive molds.

3 Authors list : Hanane Yousfi1,3, Stéphane Ranque2,3, Jean marc Rolain1,3 and Fadi Bittar1,3*

4 Affiliations :

1 5 Aix Marseille Univ, IRD, APHM, MEPHI, Marseille, France.

6 2 Aix Marseille Univ, IRD, APHM, SSA, VITROME, Marseille, France.

7 3IHU Méditerranée Infection, Marseille, France.

8 *Corresponding author:

9 Fadi Bittar, IHU Méditerranée Infection, 19-21 boulevard Jean Moulin, 13005 Marseille,

10 France. Email:[email protected]

11 Keys words: Drugs repurposing, Prestwick chemical library, emerging fungi, multidrug-

12 resistant molds.

13 Abstract: 100 words

14 Text: 1200 words

15 References: No more than 20 references

16 Table: 1

17 Figure: 1

18 Brief report: Medical Mycology

21 22 Abstract

23 The incidence of uncommon mold infections increased in the last decade. Particularly,

24 zygomycetes are the most encountered of these molds, which are resistant to several

25 antifungals. Consequently, it is urgent to find new compounds active against such invasive

26 species. Here, 1280 FDA-approved drugs were tested against six multidrug-resistant molds

27 including, Aspergillus, Fusarium, Scedosporium, Rhizopus and Lichtheimia species. We

28 identified several hits inducing fungal growth inhibition ≥ 70%. Clioquinol, alexidine

29 dihydrochloride, hexachlorophene and thonzonium bromide have a broad antifungal activity

30 against all strains tested. This study enriches the panel of antifungals that could be used in the

31 treatment of filamentous mycosis.

44 45 Until recently, more than 400 fungal species have been involved as opportunistic

46 pathogens in human infection [1]. Unfortunately, to treat these various fungal infections

47 induced by a such high number of fungi, only 4 therapeutic classes, namely polyenes, azoles,

48 echinocandins and flucytosine, are available and are used as prophylaxis or as specific

49 treatment either in monotherapy or in combination in clinical practices [2]. Although these

50 antifungal drugs remain active, they have some limitations that may complicate their routine

51 uses such as an off-target toxicity, a drug-drug interaction and a poor clinical response

52 especially to face emerging resistant isolates [3]. Antifungal resistance is a major concern that

53 has been described in most fungal pathogens; on the one hand the main human fungal

54 pathogens, including Candida albicans, Cryptococcus spp and Aspergillus spp present a

55 relatively high resistance rate and therefore they constitute serious public health concerns. On

56 the other hand other life-threatening and emerging pathogens, including opportunistic

57 multidrug-resistant (MDR) ones are increasingly reported [4]. These include

58 Scedosporium/Lomentospora spp., Fusarium spp. and other Mucorales [5]. Thus, the aim of

59 this study is to determine available off-label drugs with potential antifungal actions against

60 emerging multidrug-resistant molds. This promising strategy, which is commonly known as «

61 repurposing approach », is useful to prevent a possible treatment failure and the development

62 of drug-resistance [6].

63 For this purpose, we screened the Prestwick Chemical Library (Prestwick, Illkirch

64 graffenstaden, France), a molecules-library containing 1,280 compounds previously approved

65 by the Food and Drug Administration (FDA) and belonging to 291 different therapeutic

66 classes, to identify inhibitors of growth of the tested multidrug-resistant fungi. All molecules

67 were tested at a fixed concentration of 10 µM. Fungal inoculum was prepared in RPMI-1640

68 medium (Sigma Aldrich, St Louis, France) according to the Clinical and Laboratory

69 Standards Institute (CLSI) protocol. Proportion of fungal growth inhibition was calculated 70 using Optical Density (OD) values measured using plate reader spectrophotometer (Multiskan

71 spectrum, Thermo Scientific, France) at a wavelength = 405 nm compared to the same

72 untreated strain. Six fungal strains isolated from different clinical samples, were tested

73 including; Aspergillus calidoustus (bronchial aspiration), Fusarium oxysporum (nails),

74 Fusarium solani (eye), Rhizopus oryzae (sinus biopsy), Lomentospora prolificans (blood) and

75 Lichtheimia corymbifera (eye). All strains were recovered from La Timone University

76 Hospital in Marseille and are therapeutically unmanageable with the current antifungal agents.

77 Aspergillus flavus ATCC 204304 and Aspergillus niger ATCC 200930 were also tested and

78 used as a quality control.

79 Antifungals susceptibily of all strains was determined using E-test method. Posaconazole

80 (POS), itraconazole (ITR), voriconazole (VRC), isavuconazole (ISA) and amphotericin B

81 (AMB) were tested.

82 High AMB Minimum Inhibitory Concentrations (MICs) were recorded for all tested strains,

83 especially for Rhizopus oryzae, Lomentospora prolificans and Fusarium solani (MICs > 32

84 µg/ml), while MICs of azoles were variable but remained higher than 32 µg/ml for Fusarium,

85 Lomentospora and Rhizopus species. Based on the primary FDA-approved drugs screening,

86 we identified 3 compounds active against R. oryzae, 11 compounds active against F.

87 oxysporum, 12 active against F. solani, 15 molecules active against L. prolificans and A. niger

88 ATCC, 13 molecules active against A. calidoustus and finally 14 drugs active against both A.

89 flavus ATCC and L. corymbifera (Table 1). All retained compounds induced fungal growth

90 inhibition ≥ 70 % (Table 1). Identified hits belong to six different classes including

91 antifungals, antibacterials, antiseptics, anthelmintics, antineoplastics and other miscellenaous

92 drugs (Table 1). Not surprisingly, most of hits are antifungals (50%). Among the non-

93 antifungal hits identified as effective compounds against the tested clinical molds, we noted

94 four particular drugs which showed a broad antifungal activity against at least five of the six 95 clinical tested strains, namely clioquinol, alexidine dihydrochloride, hexachlorophene and

96 thonzonium bromide (Figure 1).

97 Clioquinol (5-chloro-7-iodo-8-quinolinol) is initially produced as topical antiseptic

98 and also used as oral treatment of intestinal amoebiasis [7]. However, in the early 1970s,

99 clioquinol was discarded as oral therapy due to its association to the development of subacute

100 myelo-optic neuropathy [7]. Currently, clioquinol is topically used to treat skin infections and

101 considered as effective zinc chelator in the treatment of Alzheimer's disease [7]. Recently,

102 You et al reported the important activity of 3% clioquinol cream against a large number of

103 fungal species [8]. Using modified agar diffusion assay, the authors showed that the inhibition

104 zone was biggest for Candida tropicalis, Candida guilliermondii, Aspergillus terreus and F.

105 solani [8]. In our study, clioquinol activity testing was performed using microdilution broth

106 method and showed antifungal growth inhibition of all tested mold strains except R. oryzae.

107 Alexidine dihydrochloride (AXD) is bis-guanide molecule, initially identified for its

108 antibacterial properties but it is also found to have anti-inflammatory and anti-cancer effects

109 [9]. In this study, AXD exert a broad antifungal activity against A. calidoutus (81%), F.

110 solani (82%), F. oxysporum (81%), L. prolificans (77%) and L. corymbifera (77%). Recently,

111 Mamouel et al, reported an antifungal activity of AXD against a diverse range of MDR-fungi

112 including C. albicans, Candida auris, Cryptococcus neoformans and the filamentous ones;

113 Scedosporium apiospermum, Aspergillus fumigatus, L. corymbifera and also R. oryzae.

114 Moreover, they showed a synergistic activity between AXD and fluconazole against

115 fluconazole-resistant C. albicans [10] . Intriguingly, no effect of AXD was observed in our

116 test against R. oryzae, at concentration equal to 10 µM (i.e. 5.8 µg/ml) (Table 1), whereas the

117 aforementioned study showed that AXD MIC80 = 3 µg/ml against R. oryzae strain. This

118 discordance is may be due to the resistance profile of R. oryzae strain used in our study; our

119 strain is resistant to all tested azoles including posaconazole (MIC >32 µg/ml), in contrast 120 with R. oryzae strain reported in Mamouel et al’s study which showed posaconazole MIC =

121 0.25 µg/ml. To our knowledge, no previous study has reported the activity of AXD against

122 Fusarium species and L. prolificans.

123 Hexachlorophene (HCP) is a chlorinated bisphenol with important bacteriostatic

124 properties against several Gram-positive bacteria (including Staphylococcus) used in

125 dermatological preparations (antiseptic) [11]. HCP showed important fungal growth inhibition

126 against all tested strains except A. calidoustus and A. niger (Table 1). It is the first report of

127 antifungal properties of HCP against invasive fungal strains. However, studies conducted on

128 mice and human stem cells have shown a clear chronic negative effect of HCP on human cells

129 and that oral or vaginal administration of important doses of HCP was embryotoxic and

130 teratogenic in rats [11].

131 Thonzonium bromide (TB) is a monocationic detergent that promotes tissue contact

132 and is used in corticosporin drops to help the penetration of active ingredients (antibacterial

133 and anti-inflammatory drugs) through ear wax. Its antifungal activity was shown by previous

134 screening of the Prestwick chemical library against planktonic C. albicans strain [10], and it

135 was identified by Chan et al study as a specific ATPase inhibitor in C. albicans [12].

136 Consequently, the high antifungal activity of TB against all mold species tested in our study

137 has not been yet described and constitutes very useful results for future investigation of its

138 antifungal activity.

139 Despite the current advances in diagnosis and prevention of invasive fungal infections,

140 the incidence of mycosis, treatment failure and mortality remain excessively high in

141 immunocompromised patients. Thus, the results of this study raise the panel of antifungal

142 agents that can be considered in the treatment of such complex infections. Further

143 investigation of antifungal properties of the off-label drugs identified in this study, each alone 144 or in combination with another antifungal drug, can facilitate the management of MDR mold

145 infections.

146 Acknowledgments

147 We thank CookieTrad for English reviewing.

148 Funding information

149 This work was supported by the French Government under the «Investissements d’avenir »

150 (Investments for the Future) program managed by the Agence Nationale de la Recherche

151 (ANR, fr: National Agency for Research), (reference: Méditerranée Infection 10-IAHU-03)

152 This work was supported by Région Provence-Alpes-Côte d’Azur and European funding

153 FEDER PRIM

154 Transparency declarations

155 No conflict of interest or financial disclosure to declare for all authors.

156

157

158

159

160

161

162

163

164

165 166 References

167 [1] B.E. de Pauw. What are fungal infections?, Mediterr. J. Hematol. Infect. Dis. 3 (2011)

168 DOI 10.4084/MJHID.2011.001.

169 [2] L. Scorzoni, A.C.A. de Paula e Silva, C.M. Marcos, P.A. Assato, W.C.M.A. de Melo,

170 H.C. de Oliveira, C.B. Costa-Orlandi, M.J.S. Mendes-Giannini, A.M. Fusco-Almeida.

171 Antifungal therapy: New advances in the understanding and treatment of mycosis,

172 Front. Microbiol. 8 (2017) DOI: 10.3389/fmicb.2017.00036.

173 [3] A.M. Nicola, P. Albuquerque, H.C. Paes, L. Fernandes, F.F. Costa, E.S. Kioshima,

174 A.K.R. Abadio, A.L. Bocca, M.S. Felipe. Antifungal drugs: New insights in research &

175 development, Pharmacol.doi:10.1016/j.pharmthera.2018.10.008.

176 [4] D.A. Enoch, H.A. Ludlam, N.M. Brown. Invasive fungal infections: A review of

177 epidemiology and management options, J. Med. Microbiol. 55 (2006) 809–818.

178 [5] J. Geddes-Mcalister, R.S. Shapiro. New pathogens, new tricks: Emerging, drug-

179 resistant fungal pathogens and future prospects for antifungal therapeutics, Ann. N. Y.

180 Acad. Sci. 1435 (2018) 57–78.

181 [6] S.M. Corsello, J.A. Bittker, Z. Liu, J. Gould, P. Mccarren, J.E. Hirschman, S.E.

182 Johnston, A. Vrcic, B. Wong, M. Khan, R. Narayan, C.C. Mader, A. Subramanian, R.

183 Todd, C. Chase, HHS Public Access, Nat. Med. 23 (2017) 405–408.

184 [7] S.R. Bareggi, U. Cornelli. Clioquinol: Review of its mechanisms of action and clinical

185 uses in neurodegenerative Disorders, CNS Neurosci. Ther. 18 (2012) 41–46.

186 [8] Z. You, X. Ran, Y. Dai, Y. Ran. Clioquinol, an alternative antimicrobial agent against

187 common pathogenic microbe, J. Mycol. Med. 28 (2018) 492–501.

188 [9] X. Zhu, J. Gao, P.Y. Ng, A. Qin, J.H. Steer, N.J. Pavlos, M.H. Zheng, Y. Dong, T.S.

189 Cheng. Alexidine Dihydrochloride Attenuates Osteoclast Formation and Bone 190 Resorption and Protects Against LPS-Induced Osteolysis, J. Bone Miner. Res. 31

191 (2016) 560–572.

192 [10] Z. Mamouei, A. Alqarihi, S. Singh, S. Xu, M.K. Mansour, A.S. Ibrahim, P. Uppuluri.

193 Alexidine Dihydrochloride Has Broad-Spectrum Activities against Diverse Fungal

194 Pathogens, MSphere. 3 (2018) 1–11.

195 [11] M. Leśniak, R. Zdanowski, M. Suska, A. Brewczyńska, W. Stankiewicz, M. Kloc, J.Z.

196 Kubiak, S. Lewicki. Effects of Hexachlorophene, a Chemical Accumulating in Adipose

197 Tissue, on Mouse and Human Mesenchymal Stem Cells, Tissue Eng. Regen. Med. 15

198 (2018) 211–222.

199 [12] C.Y. Chan, C. Prudom, S.M. Raines, S. Charkhzarrin, S.D. Melman, L.P. De Haro, C.

200 Allen, S.A. Lee, L.A. Sklar, K.J. Parra, Inhibitors of V-ATPase proton transport reveal

201 uncoupling functions of tether linking cytosolic and membrane domains of V 0 subunit

202 a (Vph1p), J. Biol. Chem. 287 (2012) 10236–10250.

203

204 Figure legend

205 Figure 1: Venn diagram of non-antifungal compounds active against emerging fungal molds,

206 with growth inhibition > 70%. Compound name Aspergillus Aspergillus Aspergillus Fusarium Fusarium Lomentospora Rhizopus Lichtheimia flavus niger calidoustus solani oxysporum prolificans oryzae corymbifera Antifungals Tiabendazole - + (70%) - + (81%) - - - - Voriconazole - - + (79%) + (80%) + (78%) - - - Butenafine Hydrochloride - - - + (72 %) + (76%) - - - Econazole nitrate + (79%) - + (83%) + (72%) - - - - Amphotericin B - + (74%) - + (88%) + (70%) - - - Enilconazole + (81%) + (70%) + (79%) - + (82%) - - - Terbinafine - - - - + (80%) - - - Haloprogin + (71%) - - - - + (80%) - + (80%) Liranaftate + (80%) - + (82%) - - + (72%) - + (72%) Amorolfine hydrochloride - - - - - + (80%) - + (80%) Itraconazole + (81%) + (74%) + (80%) - - - - - Oxiconazole Nitrate - - + (79%) - - - - - Butoconazole nitrate - + (73%) ------Sulconazole nitrate - + (74%) ------Posaconazole - + (71%) ------Miconazole + (81%) ------Naftifine hydrochloride + (75%) ------

Antibacterials Chloroxine - - + (77%) + (87%) + (80%) - - - Alexidine dihydrochloride + (70%) + (76%) + (81%) + (82%) + (81%) + (77%) - + (77%) Dequalinium dichloride - - - - - + (81%) - + (81%) Methyl benzethonium chloride - + (73%) - - - + (81%) - + (81%) Chlorhexidine - - + (70%) - - + (71%) - + (70%) Clotrimazole + (74%) - + (78%) - - + (72%) - + (71%) Sertaconazole nitrate - - + (77%) - - - - - Anthelmintics Tiabendazole - + (70%) - + (81%) - - - - Fenbendazole - + (71%) ------Albendazole - + (74%) ------Parbendazole + (73%) + (70%) ------

Antiseptics Thonzonium bromide + (72%) + (78%) + (73%) + (88%) + (83%) + (83%) + (72%) + (83%) Hexachlorophene + (80%) - - + (79%) + (81%) + (81%) + (77%) + (81%) Clioquinol + (71%) + (75%) + (73%) + (85%) + (77%) + (82%) - + (82%) Antineoplastics Azaguanine-8 - - - - - + (73%) - + (74%) Floxuridine - - - + (92%) + (79%) + (73%) - + (72%) Camptothecine + (71%) ------Miscellaneous drugs Pentetic acid (Chelator) - - - - - + (73%) + (73%) + (73%) Disulfiram (Alcohol addiction) - - - - - + (74%) - + (70%) 207 208 Table 1: List of compounds showing activity (+) against the tested mold strains after screening of the Prestwick Chemical Library. (%):

209 proportion of fungal growth inhibition. (-): no activity or < 70%.

Partie III

Repositionnement de molécules médicamenteuses pour la prise en charge des infections fongiques invasives, en monothérapie ou en association aux antifongiques communément utilisés en pratique

clinique.

Avant-propos

Les infections causées par des agents mycosiques, particulièrement les infections sévères ou profondes, ont considérablement augmenté et leurs agents étiologiques se sont diversifiés ces dernières années. Parallèlement, les possibilités thérapeutiques se sont limitées

à uniquement quatre classes d’antifongiques, compliquant ainsi la prise en charge des patients

[14]. Les mycoses profondes et systémiques sont dues essentiellement à deux genres de champignons opportunistes, Candida et Aspergillus, mais le nombre de cas causés par d'autres genres ou espèces sont en augmentation. Par conséquent, il est urgent de mettre en place des alternatives thérapeutiques au traitement antifongique traditionnel, particulièrement depuis l’émergence de plusieurs souches hautement résistantes à la thérapie antifongique de base.

Parmi les molécules antérieurement repositionnées comme potentiels antifongiques, on trouve la colistine, qui est un antibiotique utilisé en dernier recours pour traiter les infections induites par les bactéries Gram-négatif multi-résistantes [15]. L’activité antifongique de la colistine sur des levures telles que les Candida tropicalis a été démontrée [16], de même que son activité sur des espèces de champignons filamenteux telles que

Fusarium et Rhizopus [17]. C’est pourquoi, dans cette partie du travail, on a testé l’action de cette molécule sur une collection de souches fongiques d’intérêt clinique résistantes à au moins une classe d’antifongiques. On a montré que la colistine exerce une activité antifongique sur l’ensemble des souches testées sauf sur celles appartenant à l’espèce

Aspergillus, avec des CMI allant de 16 à 128 µg/ml. On a également démontré l’activité fongicide de la colistine sur les souches multi-résistantes de C. albicans, Cryptococcus neoformans et Rhodotorula mucilaginosa. En utilisant la microscopie à fluorescence, une action de perméabilisation de la membrane cellulaire a été constatée sur une souche de C.

albicans multi-résistante traitée avec la colistine, permettant potentiellement d’expliquer son caractère fongicide.

Néanmoins, les CMI de la colistine obtenues dans cette étude restent relativement

élevées et difficiles à atteindre en concentration plasmatique chez le patient, en raison de l’importante néphrotoxicité de cette molécule. C’est pourquoi, des tests de synergie ont été réalisés avec les antifongiques usuels et on a montré la présence d’une importante activité synergique entre la colistine et l’amphotéricine B et l’itraconazole contre Candida albicans et

Rhodotorula mucilaginosa respectivement. De même, l’association de la colistine au fluconazole a montré une activité synergique sur la souche R. mucilaginosa.

En conclusion, les résultats de cette étude ont permis d’apporter une alternative thérapeutique au traitement d’infections fongiques à germes invasifs. L’utilisation de la colistine en association avec d’autres antifongiques reste une solution envisageable et efficace pour la prise en charge de certains cas de mycoses qui restent très difficile à traiter en pratique clinique (Article 3).

Les infections à Candida ou candidoses sont habituellement des infections opportunistes qui ne se manifestent qu’en présence de facteurs prédisposants. Leur spectre clinique s’étend des formes localisées (cutanées et/ou muqueuses) d’une grande fréquence en clinique, aux atteintes invasives rencontrées chez des terrains d’immunodépression [1]. Le genre Candida compte un peu moins de 200 espèces mais C. albicans représente la principale levure impliquée en pathologies humaines. Elle représente plus de 70 % des isolats et induit plus de 50 % des candidémies [18]. Ces infections sont habituellement traitées par des azolés

et l’amphotéricine B mais l’efficacité de ces molécules est compromise par le développement de la résistance aux antifongiques. Par conséquent, il est important d’identifier d’autres médicaments ayant une activité antifongique sur ce type de levures.

En utilisant le concept du repositionnement, un criblage initial des 1280 molécules a

été réalisé sur deux souches de Candida albicans (Candida albicans ATCC 90028 et Candida albicans multi-résistant). Vingt-et-une molécules médicamenteuses présentant une activité antifongique non-traditionnelle ont été identifiées, dont la ribavirine qui est un antiviral, habituellement utilisé dans la prise en charge de l’hépatite C en association à l’interféron-a.

Ultérieurement, la sensibilité à la ribavirine de 100 souches de Candida appartenant différentes espèces, a été testée. On a montré que la ribavirine présente, in vitro, une activité antifongique importante sur l’ensemble des souches avec des CMI comprises majoritairement entre 0,37 et 3,02 µg/ml. En revanche, des CMI élevées ≥ 24.16 µg/ml ont été constatées pour les souches de C. krusei, C. glabrata, C. lusitaniae et certaines souches de C. albicans. On a aussi démontré que la résistance de ces souches à la ribavirine n’était pas le résultat d’une augmentation de l’activité des pompes à efflux membranaires.

Afin d’étudier la relation de synergie qui pourrait exister entre la ribavirine et les antifongiques, nous avons réalisé des tests d’association entre la ribavirine et amphotéricine

B, fluconazole et voriconazole. Une importante activité synergique a été démontrée avec les trois associations sur la souche de Candida albicans multi-résistante. Ces résultats sont très prometteurs quant à l’utilisation de la ribavirine en monothérapie ou en association, pour la prise en charge des mycoses invasives. Cependant, l’activité in vivo, de la ribavirine sur des modèles murins devrait être démontrée afin de conforter l’intérêt de l’utilisation de cet antiviral en mycologie médicale (Article 4).

Article 3:

In vitro polymyxin activity against clinical multidrug-resistant fungi

Hanane Yousfi, Stéphane Ranque, Jean-Marc Rolain and Fadi Bittar

Published in «Antimicrobial Resistance and Infection Control»

Yousfi et al. Antimicrobial Resistance and Infection Control (2019) 8:66 https://doi.org/10.1186/s13756-019-0521-7

RESEARCH Open Access In vitro polymyxin activity against clinical multidrug-resistant fungi Hanane Yousfi1, Stéphane Ranque2, Jean-Marc Rolain1 and Fadi Bittar1*

Abstract Background: Although antifungals are available and usually used against fungal infections, multidrug-resistant (MDR) fungal pathogens are a growing problem for public health. Moreover, fungal infections have become more prevalent nowadays due to the increasing number of people living with immunodeficiency. Thus, previously rarely- isolated and/or unidentified fungal species including MDR yeast and moulds have emerged around the world. Recent works indicate that polymyxin antibiotics (polymyxin B and colistin) have potential antifungal proprieties. Therefore, investigating the in vitro activity of these molecules against clinical multidrug-resistant yeast and moulds could be very useful. Methods: In this study, a total of 11 MDR yeast and filamentous fungal strains commonly reported in clinical settings were tested against polymyxin antibiotics. These include strains belonging to the Candida, Cryptococcus and Rhodotorula yeast genera, along with others belonging to the Aspergillus, Fusarium, Scedosporium, Lichtheimia and Rhizopus mould genera. The fungicidal or fungistatic action of colistin against clinical yeast strains was determined by the time-kill study. Further, a checkerboard assay for its combination with antifungal agents, usually used in clinical practices (amphotericin B, itraconazole, voriconazole), was carried out against multi-drug resistant fungal strains. Results: Polymyxin B and colistin exhibited an antifungal activity against all MDR fungal strains tested with MICs ranging from 16 to 128 μg/ml, except for the Aspergillus species. In addition, colistin has a fungicidal action against yeast species, with minimum fungicidal concentrations ranging from 2 to 4 times MICs. It induces damage to the MDR Candida albicans membrane. A synergistic activity of colistin-amphotericin B and colistin-itraconazole associations against Candida albicans and Lichtheimia corymbifera strains, respectively, and colistin-fluconazole association against Rhodotorula mucilaginosa, was demonstrated using a checkerboard microdilution assay. Conclusion: colistin could be proposed, in clinical practice, in association with other antifungals, to treat life- threatening fungal infections caused by MDR yeasts or moulds. Keywords: Polymyxin antibiotics, MDR-fungi, Repurposing-drug, Candida albicans, Molds

Background most common drugs used in different clinical situations; Invasive fungal diseases treatment is challenged by the azole drugs display a large spectrum of activity against restricted number of available antifungal drugs; with only both yeast and filamentous fungi. (iii) Pyrimidine four different classes of antifungals being available to treat analogues, including 5-flucytosine, are used in combin- a large number of fungal-associated diseases [1]. (i) Poly- ation with other antifungals to treat yeast infections but enes are the first antifungals available in clinical practice have little action against most moulds. (iv) Finally, echino- with two drugs mainly used: amphotericin B and nystatin. candins, the newest class of antifungals include caspofun- (ii) Azole antifungals, such as fluconazole, itraconazole, gin, micafungin and anidulafungin that display fungicidal voriconazole, posaconazole and isavuconazole, are the activity against ascomycetes yeast species [2]. In addition to this limited therapeutic arsenal, there * Correspondence: [email protected] has been a dramatic and worldwide increase in the 1Aix Marseille Univ, IRD, APHM, MEPHI, IHU Méditerranée Infection, 19-21 boulevard Jean Moulin, 13005 Marseille, France incidence of fungal infections [3]. In fact, along with the Full list of author information is available at the end of the article

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Yousfi et al. Antimicrobial Resistance and Infection Control (2019) 8:66 Page 2 of 10

main mycosis agents, such as Candida albicans, Asper- concentrations (MICs) ranging from 30 to 75 μg/ml [14]. gillus fumigatus and Cryptococcus neoformans [4], other More recently, polymyxin sensitivity of life-threatening life-threatening and emerging pathogens, including not moulds, such as Fusarium and Rhizopus species, has previously well-identified/characterized species and been described [15, 16]. This study aimed to test the in opportunistic multidrug-resistant (MDR) ones, are vitro activity of polymyxin against the most common increasingly reported. These include Candida auris, clinical MDR yeasts and moulds and to assess colistin Scedosporium/Lomentospora spp., Fusarium spp. and activity, the mechanism of action and the synergy of Mucorales [2]. Indeed, several factors can explain this colistin-antifungals associations that could be used in increasing incidence of fungal infections; the increasing the treatment of invasive fungal infections. number of patients with immunodeficiency (ex. HIV, cancer and transplant patients), the ageing of the popu- Methods lation [5] and improved detection and diagnostic Fungal isolates methods [2]. However, the severity of such fungal infec- In this study, eleven clinical fungi recovered at the Uni- tions varies depending on the site of infection (superfi- versity hospital of Marseille were used. Four yeasts cial or deep-seated) and the immune status of the belonging to Candida, Cryptococcus and Rhodotorula concerned patients. One major characteristic of these species and seven moulds belonging to Fusarium, emerging fungal pathogens is their highly-resistant pro- Scedosporium, Lichtheimia, Rhizopus and Aspergillus file to antifungal drugs. Therefore, these disseminated species were tested (Table 1). Isolates were identified infections caused by MDR yeasts and moulds are diffi- using Matrix-Assisted Laser Desorption/Ionisation mass cult to treat [6], leading thus to a significant increase of spectrometry (MALDI-TOF MS) [17] and microscopic morbidity and mortality, in immunocompromised methods for the filamentous fungi species. These strains patients but also in healthy individuals [7]. Antifungal were isolated from different clinical samples including resistance can be intrinsic, called primary resistance, or blood culture, cerebrospinal fluid, nails, bronchial aspir- acquired, also called secondary resistance. Many resist- ation and ocular samples (Table 1). Candida krusei ance mechanisms have been described, such as biofilm ATCC 6258, Candida parapsilosis ATCC 22019, Asper- formation (especially in Candida albicans), failure of gillus fumigatus ATCC 204205, Aspergillus flavus ATCC intracellular drug accumulation or drug target alter- 204304, Escherichia coli LH1 [18], Escherichia coli 1R4 ations [8]. During the past decade, genome plasticity of [19] and Klebsiella pneumoniae 853 [20] were used as human fungal pathogens has been strongly associated susceptibility testing quality controls. with their ability to acquire resistance to antifungals [9]. That is why many studies suggest to use other pharma- Phenotypic profiles determination cological classes and re-purposing old drugs either as a Antifungal susceptibility testing was performed using single antifungal agent or in combination with known two different methods: E-test (BioMérieux, Marcy antifungal drugs [10]. l’Etoile, France) and commercial broth microdilution In this regard, antimicrobial peptides (AMPs) have plates; Sensititre®YeastOne® (Thermo Fisher Scientific, received attention as prospective compounds for a Schwerte, Germany). The MICs obtained for each anti- further discovery of new antimycotics. More than 2700 fungal tested against yeasts and moulds (Table 1) was antimicrobial peptides have been identified and the compared to the breakpoints provided by the manufac- number is growing [11]. Among the AMPs commonly turers or to the epidemiological cutoff as previously used in therapeutic practices, there are polymyxins described [21] in order to assess the susceptibility of which are cyclic, positively charged peptides, obtained each strain to the different antifungal agents tested. naturally from Gram-positive bacteria, such as Paeniba- cillus polymyxa. Among polymyxin molecules described, Polymyxin susceptibility testing two have been used in clinical settings: polymyxin B Colistin and PMB MICs were performed using the broth (PMB) and polymyxin E (colistin) [12]. This class of microdilution method as outlined by the Clinical and antibiotics has been discarded in the early 1980s because Laboratory Standards Institute (CLSI) (M38-A, Vol. 22, of their neuro- and nephro-toxicity. However, poly- NO. 16 for filamentous fungi and M27-A2, Vol. 22, NO. myxins were recently reintroduced in the antimicrobial 15 for yeasts). Serial colistin and PMB (Sigma Aldrich, therapy as a last option to treat infections caused by St Louis, France) dilutions ranging from 0.5 to 256 μg/ multidrug-resistant Gram-negative bacteria [13]. In ml were prepared in RPMI-1640 (Sigma Aldrich, St addition to its antibacterial action, polymyxins were Louis, France) with glutamine and without bicarbonate shown in the early 1970s to have antifungal activity medium buffered to pH 7.0 with MOPS (Sigma Aldrich, against various Candida species, including Candida tro- St Louis, France) buffer. Fungal inoculums were picalis with polymyxin E Minimum inhibitory prepared in the test medium and adjusted to 0.5 Yousfi et al. Antimicrobial Resistance and Infection Control (2019) 8:66 Page 3 of 10

Table 1 Phenotypic profiles and Colistin, PMB MICs of fungal strains tested in this study Clinical samples MICs (μg/mL) Anid Mica Caspo Flu 5-Fc Posa Itra Vorico AB Isavu Ct PMB Candida krusei ATCC 6258 / 0.06 0.12 0.25 32 8 0.25 0.12 0.25 1 ND 64 32 Candida parapsilosis ATCC 22019 / 0.5 0.5 0.12 1 0.12 0.03 0.008 0.03 0.5 ND 64 16 Rhodotorula mucilaginosa Endobucal > 8 > 8 > 8 128 0.06 1 0.5 2 1 ND 32 16 Cryptococcus neoformans Blood > 8 > 8 > 8 2 1 0.03 0.12 0.03 0.5 ND 32 16 Candida albicans H5 CSF 0.015 0.03 0.05 16 1 0.5 0.5 0.25 0.25 ND 128 128 Candida albicans H6 Nails 0.06 0.03 0.12 > 256 0.12 > 8 32 256 1 ND 128 64 Aspergillus fumigatus ATCC 204205 / ND ND ND ND ND 0.008 0.015 0.12 2 ND > 256 > 256 Aspergillus flavus ATCC 204304 / ND ND ND ND ND 0.06 0.06 1 4 ND > 256 > 256 Aspergillus calidoustus Bronchial aspiration ND ND ND ND ND 1 6 4 4 0.25 256 128 Fusarium oxysporumY5 Nails ND ND ND ND ND > 32 > 32 2 4 > 32 64 16 Fusarium solani Y6 Ocular sample ND ND ND ND ND > 32 > 32 > 32 > 32 > 32 64 16 Rhizopus oryzae Y9 Sinus biopsy ND ND ND ND ND > 32 > 32 > 32 > 32 > 32 128 64 Lomentospora prolificans Y8 Blood ND ND ND ND ND > 32 64 > 32 > 32 > 32 32 16 Scedosporium apiospermum F2 Bronchial aspiration ND ND ND ND ND 2 0.75 > 32 4 ND 16 16 Lichtheimia corymbifera ST87 Eyes ND ND ND ND ND 2 2 > 32 2 2 32 32 Anid - Anidulafungin; Mica - Micafungin; Casp - Caspofungin; Flu - Fluconazole; 5-Fc - 5-Flurocytosin; Pos - Posaconazol, Itra - Itraconazole; Vori - Voriconazole; AB - Amphotericin B; Isavu – Isavuconazole; Ct – Colistin; PMB- Polymyxin B; MIC - Minimum Inhibitory Concentration; ND: Not Done

MacFarland. A 1:100 dilution followed by a 1:20 dilution to 100% when the concentration of an antibiotic reaches were performed on yeast strains to obtain a final inocu- to the MIC as the OD of the tested well is quasi-equal lum of 0.5 to 2.5 × 103 CFU/mL, whereas only a 1:50 di- to the OD of the blank well. lution was done for moulds with a final inoculum of approximately 0.4 to 5 × 104 CFU/mL. It is important to Colistin time-kill experiment and minimum fungicidal mention that fresh conidia of filamentous fungi were ob- concentration (MFC) determination tained after approximately 7 days of incubation at 35 °C Colistin time-kill study was performed, as previously on potato dextrose agar. Then, 100 μl of the fungal/bac- described [22]. 9 ml of the fungal suspension was ad- terial inoculum was added into each colistin or justed to a 0.5 McFarland turbidity. One ml of the ad- PMB-containing wells. Plates were incubated at 37 °C for justed fungal suspension was added to 9 ml of either 24 h (Candida spp., bacterial strains) or 48 h (Cryptococ- RPMI-1640 medium, as a control, or to a solution of cus) and at 35 °C for 48 h (filamentous fungi and Rho- growth medium supplemented with an appropriate con- dothorula mucilaginosa). The susceptibility to polymyxin centration of antibiotic solution. The colistin concentra- antibiotics was assessed on the basis of visual observations in the resulting solutions were 0.5, 1, 2, 4, 8, 16, tion of growth or inhibition of the isolate in the culture and 32 times the MICs for the tested isolates. Then, the media. The Resazurin (Sigma Aldrich, St Louis, France) tubes were incubated at 37 °C on an orbital shaker. At 0, was used to indicate the growth of any microorganism 6, 12, 24, 36, and 48 h following the introduction of the by a culture medium colour shift from blue to pink. tested isolate into the solutions tubes, 100 μl aliquots Then, the inhibition rate was calculated, after measuring were taken from each test solution. Different serial dilu- the optical density (OD) value by using a plate reader tions were performed on these aliquots, and a 10 μl spectrophotometer (Multiskan spectrum, Thermo Scien- aliquot from each dilution was streaked on Sabouraud tific, France), as follows: % of fungal growth inhibition agar plates (Biomérieux, France) and incubated for = (OD of untreated well - OD of tested well)*100 / (OD approximately 24 h for colony count determination. of untreated well – OD of blank well); where a blank Then, the Minimum fungicidal concentration (MFC) well contains only the RPMI medium (i.e. without any was determined as the lowest antibiotic concentration fungal strain and without any antibiotic agent), an un- leading to no significant growth or less than three treated well contains a given strain in the RPMI medium colonies on Sabouraud agar plates in comparison to the without any antibiotic agents and a tested well contains growth control. The experimental data was analyzed both a given strain in the RPMI medium and a given using the GraphPad Prism 5.3 software (GraphPad Inc., antibiotic agent. Thus, the growth inhibition rate is close San Diego, CA, USA) to obtain time-kill curves. Yousfi et al. Antimicrobial Resistance and Infection Control (2019) 8:66 Page 4 of 10

Study of the mechanism of action of colistin on Candida quality control strains corresponded to the recom- albicans species mended 24 h and 48 h - MICs limits of microbroth dilu- We used propidium iodide (PI), a membrane-impermeable tion method outlined in the CLSI protocol. This DNA stain, to demonstrate the eventual fungicidal activity confirms the adequateness of the method used to deter- of colistin by its ability to induce cell membrane damages mine susceptibility profiles for all isolates. C. albicans [23]. Colistin-treated and untreated cells were suspended H6 was resistant to all azole antifungal agents tested in PBS and stained with 5 μg/ml of PI for 20 min in the while C. albicans H5 was only resistant to fluconazole dark at room temperature. PI fluorescence was examined (Table 1). Both strains remained sensitive to echinocan- under a fluorescence microscope at λex = 535 nm and λem dins and amphotericin B antifungals. Rhodotorula muci- = 617 nm. laginosa was resistant in vitro to fluconazole with MIC = 128 μg/ml and to all echinocandin agents. Cryptococ- Colistin / antifungals association checkerboard testing cus neoformans was also resistant to echinocandin class Checkerboard broth microdilution method was used to (Table 1). test the synergy of colistin with three antifungal agents All filamentous fungi tested were resistant to ampho- commonly used in clinical settings; amphotericin B, flu- tericin B with a high MIC for Rhizopus oryzae and Sce- conazole and itraconazole (Sigma Aldrich, St Louis, dosporium species (> 32 μg/ml). Fusarium, Scedosporium France). Firstly, MICs of individual agents were deter- and Rhizopus species were also resistant to azoles (vori- mined because the range of concentration of drugs to conazole, itraconazole and posaconazole) usually used in test the associations was established according to these clinical settings (Table 1). The pyrimidine analogue MICs. Eight doubling dilutions of the two agents being 5-fluorocytosine and echinocandin agents were not tested (i.e. colistin and antifungals) were prepared in the tested against moulds because of their relatively poor susceptibility testing medium RPMI-1640 (1MIC, ½ activity against filamentous fungi. MIC, ¼ MIC, 1/8 MIC, 1/16 MIC, 1/32 MIC, 1/64 MIC, 1/128 MIC). 50 μl of each agent was added in wells of a Polymyxins exhibited in vitro, antifungal activity against microtiter plate to provide a total of 64 drug combina- MDR yeasts and filamentous fungi tions. Additional rows were used to determine the MIC Susceptibility testing showed clear endpoints with a of each antimicrobial agent alone by adding 100 ml of 100% growth inhibition of the MDR strains tested. The each agent. The fungal inoculum was prepared accord- MICs of Escherichia coli LH1 (MIC = 8 μg/ml) (16), ing to the CLSI standard protocol and 100 μl was added Escherichia coli 1R4 (MIC = 8 μg/ml) (17) and Klebsiella to each well. The plates were incubated under optimal pneumoniae 853 (MIC = 64 μg/ml) [20] were within the growth conditions for yeasts and filamentous moulds. quality control ranges. The results were analysed and MIC100 were determined The polymyxins MICs against the fungi strains ranged visually and by optical density measurements on a mi- from 16 to 128 μg/ml (Table 1), except for Aspergillus croplates reader (Multiskan Spectrum, Thermo Scien- fumigatus and A. flavus strains, which appeared to be tific), based on a reduction in absorbance compared to not sensitive to both colistin and PMB, with MICs the free drug control wells. Then, we calculated the ≥256 μg/ml. Among the 11 clinical strains tested, 4 iso- Fractional Inhibitory Concentrations (FICs), where FIC1 lates including Lichtheimia corymbifera, Lomentospora (Colistin) = MIC of colistin in the combination/MIC of prolificans and Scedosporium apiospermum exhibited the colistin alone and FIC2 (antifungal drug) = MIC of anti- highest susceptibility to polymyxin molecules, with fungal drug in the combination/MIC of antifungal drug MICs ranging from 16 μg/ml to 32 μg/ml; in contrast alone. The fractional inhibitory index (FIX) is the sum of Aspergillus calidoustus exhibited the lowest susceptibility FIC1 and FIC2 and was interpreted as follows: if the FIX to colistin and PMB, with MICs of 256 μg/ml and is ≤0.5, then there is synergy between the tested antimi- 128 μg/ml, respectively. crobials; if it is > 0.5 but ≤1 then there is additivity between the tested antimicrobials; if it is > 1 but ≤4, Colistin presents fungicidal activity against clinical MDR there is indifference between the tested antimicrobials, yeasts and if the FIX is > 4, that means that there is an antag- The colistin MICs ranged from 16 to 128 μg/ml for the onism between the tested antimicrobials. yeast isolates. As shown on Fig. 1, no inhibitory effect was observed at colistin concentrations equal to 0.5X Results MICs and the curves were nearly identical to those Resistance phenotypic profiles of the emerging fungal found in the controls for all species tested. At colistin pathogens used in this study concentrations equal to 1X MICs, fungistatic effect was The MICs obtained with the different antifungal classes observed (Fig. 1). In contrast, colistin fungicidal activities tested against the 4 Candida spp. and Aspergillus spp. against C. albicans, C. neoformans and R. mucilaginosa Yousfi et al. Antimicrobial Resistance and Infection Control (2019) 8:66 Page 5 of 10

Fig. 1 Time-kill kinetics of colistin against four fungal strains (C. albicans, C. krusei, C. neoformans and R. mucilaginosa). The colistin concentrations used are as following: (▲) control (no colistin added), (♦) 0.5X MIC, (■) 1X MIC,(●) 2X MIC, (■) 4X MIC were noted, no later than 12th hour of incubation, with Synergistic activity of colistin with itraconazole, MFC ≥ to two-fold of its MIC values (Fig. 1). For C. albi- amphotericin B and fluconazole cans strains, the MFC was 256 μg/ml, but regrowth was Based on checkerboard association testing (Fig. 3), observed at the 36th hour of incubation. The latter colistin-itraconazole and colistin-amphotericin B exhib- phenomenon was not observed in Cryptococcus neofor- ited a synergistic activity against MDR C. albicans and mans or in Rhodotorula mucilaginosa species. the mucoralean Lichtheimia corymbifera. We noted that Moreover, a red fluorescence was observed by fluores- itraconazole MIC decreased from 32 μg/ml to 2 μg/ml cence microscopy (Fig. 2) indicating the presence of (for C. albicans H6), with a FIX = 0.5, and from 2 μg/ml interaction between PI and nuclear DNA of C. albicans to 0.5 μg/ml (for L. corymbifera), with a FIX = 0.2 when H6 treated with a colistin concentration equal to MFC. combined with colistin (64 μg/ml and 0.5 μg/ml, respect- This observation allowed us to conclude that colistin ively) (Table 2). The MIC of amphotericin B decreased can induce cell membrane damage which provide fur- from 2 μg/ml to 0.5 μg/ml (for L. corymbifera) and from ther evidence that it can lead to cell death, confirming 1 μg/ml to 0.5 μg/ml (for C. albicans H6) when com- its fungicidal activity. bined with colistin (0.5 μg/ml and 1 μg/ml, respectively)

Fig. 2 Fluorescence microscopy of Candida albicans H6 after treatment with 5 μg/ml of propidium iodide. a: fluorescence image of cells treated with 256 μg/ml (2X MIC) of colistin for 24 h. b: Brightfield image of cells treated with colistin for 24 h. Scale bar: 2 μm Yousfi et al. Antimicrobial Resistance and Infection Control (2019) 8:66 Page 6 of 10

Fig. 3 Plots of the checkerboard assays for the combinations of colistin with 3 antifungals (fluconazole, itraconazole and amphotericin b). Each dot presents the MICs of colistin (x-axis) and the antifungal agent (y-axis) used in the combination against R. mucilaginosa (a), C. albicans H5 (b), L. corymbifera (c and e) and C. albicans H6 (d and f)

Table 2 MIC and FIX values of colistin in combination with antifungals from the checkerboard assay Strains tested Agents in combination MIC alone (μg/ml) MIC in the combination (μg/ml) FIC FIX Outcome Rhodotorula mucilaginosa Colistin 16 1 0.06 0.5 Synergy Fluconazole 128 64 0.5 Candida albicans H5 Colistin 128 1 0.007 1 Additivity Fluconazole 16 16 1 Candida albicans H6 Colistin 128 64 0.5 0.5 Synergy Itraconazole 32 2 0.06 Colistin 128 1 0.007 0.5 Synergy Amphotericin B 1 0.5 0.5 Lichtheimia corymbifera ST87 Colistin 32 0.5 0.01 0.2 Synergy Itraconazole 2 0.5 0.25 Colistin 32 0.5 0.01 0.2 Synergy Amphotericin B 2 0.5 0.25 FIC; Fractional Inhibitory concentration = MIC of the agent in the combination/MIC of the agent alone. FIX; Fractional Inhibitory Index is the sum of the FICs of the agents in the combination Yousfi et al. Antimicrobial Resistance and Infection Control (2019) 8:66 Page 7 of 10

(Table 2). Interestingly, colistin acted in synergy with flu- [15, 24]. Indeed, in Ben-Ami et al study, the colistin conazole against R. mucilaginosa but not against MICs against the fourteen clinical Rhizopus spp. tested Candida albicans H5 (Table 2). Although the colistin were variables and ranged between 16 and 32 μg/ml, but MICs decreased from 128 μg/ml to 1 μg/ml when com- MICs of antifungal agents were not mentioned [15]. So bined with fluconazole (16 μg/ml), no synergy against C. the discordance of MIC results between the Ben-Ami et albicans H5 was observed with a FIX = 1 (Table 2). al study and our work could be explained by the eventual high resistance level of our strain to azole Discussion agents which can be due to the over expression of Our study demonstrated that polymyxin antibiotics have, efflux-pumps and/or other mechanisms [27]. It is aside from their antibactericidal activity, an antifungal important to mention that PMB MIC against R. oryzae activity, especially against multidrug-resistant Candida, was equal to 64 μg/ml in this study compared to 32 μg/ Rhodotorula and Cryptococcus yeast isolates, but also ml obtained by Zhai et al study [24]. against resistant filamentous fungi, such as Scedospor- Finally, among the pertinent emerging fungal pathogens ium, Rhizopus and Lichtheimia species. The MICs shown by several studies, Scedosporium and Lomentospora obtained against Candida spp. ranged between 64 and spp. are often notified [28]. They can induce a broad range 128 μg/ml for colistin and mainly between 16 and 64 μg/ of diseases; from colonisation in cystic fibrosis patients ml for PMB. The latter is in concordance with PMB (for Scedosporium spp.) to disseminated severe infections MICs already reported by Zeidler et al and Zhai et al in immuno-compromised hosts (for Lomentospora prolifi- studies [7, 24] confirming the validity of our results and cans). Although, the colistin MIC against S. apiospermum the reproducibility of the used technique. Although we strain tested here is within the colistin MICs range previ- obtained a MIC of PMB equal to 16 μg/ml against C. ously described by Schemuth et al, the colistin MIC ob- neoformans, a lower MIC (MIC = 8 μg/ml) against this tained for L. prolificans was higher than that described in species has been previously described [24]. This could this previous study (32 μg/ml versus 12 μg/ml) [29]. possibly be explained by the resistance phenotype of the Nevertheless, it is worthy to note that MICs90 were used strain tested in our study. Moreover, colistin has not by Schemuth et al [29] whereas MICs100 were used in our been shown, in the early 1970s, to be an effective study. Finally, to the best of our knowledge, colistin and molecule against three strains belonging to the genus PMB activities against Lichtheimia corymbifera have not Rhodotorula [14]. However, to the best of our know- been previously reported. ledge, neither colistin MICs against C. neoformans nor In human studies, a single dose of 75 to 150 mg of PMB MICs against R. mucilaginosa have been reported colistin produced bioactive serum colistin concentra- elsewhere. tions ranging from 6 to 18 μg/ml; higher serum colistin On the other hand, the filamentous fungi isolates concentrations (13 to 32 μg/ml) were measured during tested in this study constitute the most common the prolonged therapy of patients with cystic fibrosis emerging cause of human mould infections with an [15]. Therefore, the obtained MICs of colistin and PMB increase being reported from various geographical sites are difficult to be achieved with IV administration, [2], including particularly Aspergillus and Fusarium spp. mainly due to their renal and neurological toxicities and Here, colistin and PMB MICs ranging from 16 to 64 μg/ the risk of frequent selection of bacterial resistant ml have been observed against the F. oxysporum and F. strains. Solani strains which are resistant to almost all azole an- However, the efficacy of polymyxin molecules on a tifungals and amphotericin B. The same range of MICs large number of MDR fungi can be considered advanta- has been reported in previous study where 12 Fusarium geous to treat bacterial and fungal co-infections that spp. were tested against PMB but not against colistin occur frequently in immunocompromised patients [30] [16]. The absence of colistin and PMB efficacy against and cystic fibrosis (CF) patients. Chronic bacterial and several Aspergillus spp. has been reported in various fungal colonization of the respiratory tract secretions is studies with MICs > 256 μg/ml [24, 25] which are similar the main cause of morbidity and mortality in CF pa- to our results. Despite that Aspergillus spp. are tients. Therefore, it would be helpful to use a treatment remaining the first cause of mould infections, mucormy- that is active on both bacteria and fungi in this context. cosis is increasingly reported in immune-compromised It is worthy to note that, in clinical practice, colistin is ad- patients and is associated with an elevated rate of mor- ministered by inhalation in CF patients as prophylaxis and tality (40–70%) even under an appropriate therapy [26]. also as a treatment against Pseudomonas aeruginosa infec- Among mucoralean pathogens, Rhizopus is the main fre- tion [31]. In addition, aerosolised colistin treatment, is quently identified genus in human infections. In our used in ventilator-associated pneumonia (VAP) cases study, colistin and PMB MICs against R. oryzae are one caused by MDR bacteria in intensive care unit setting to two folds higher than those reported in earlier studies [32]. Interestingly, in a recent in vivo study, Landersdorfer Yousfi et al. Antimicrobial Resistance and Infection Control (2019) 8:66 Page 8 of 10

et al [33] observed high epithelial lining fluid and low permeability leading to a leakage of intracellular compo- plasma colistin concentrations following the administra- nents [41]. Therefore, and as supported by PI staining re- tion of only a pulmonary dose through jet nebulization, sults (Fig. 2), it is likely that antifungal azoles ease the confirming a benefit of the local administration of colistin polymyxins’ action and add a potential damage to the fun- in comparison to its IV treatment [34]. Moreover, a pro- gal membrane which results in a synergistic potency of the spective study conducted on 18 patients with chronic lung combined drugs. Moreover, colistin MIC values significantly disease showed that nebulized colistin is effective and decreased from 128 μg/ml to 1 μg/ml and from 32 μg/ml to improves the quality of life, without presenting side effects 0.5 μg/ml against C. albicans and L. corymbifera respect- and without selecting colistin-resistant isolates in treated ively when it was associated with amphotericin B. The asso- patients [35]. So high-dose nebulized colistin could be ciation of the fungal membrane permeabilization induced proposed against pulmonary life-threatening MDR fungi, by amphotericin B via ion channel formation [42] with the without increasing colistin plasma concentration, and thus probable membrane damage occurred by colistin could ex- avoiding colistin’s toxicity. plain the decrease of MICs and the synergistic effect be- Similar to CF cases, the use of polymyxin antibiotics tween colistin and amphotericin B. can improve the poor prognosis of fungal keratitis, due Thus, despite the elevated MICs of colistin found in our to the emergence of MDR fungal pathogens, particularly work against multidrug-resistant yeast and moulds, the Fusarium spp. [36], and to the limited ocular penetra- use of colistin, in combination with other antifungal tion of antifungals [37]. Notably, PMB can be formulated agents, remains an excellent way to avoid the development for ophthalmic use [16], which is described as a highly of fungal resistance and to decrease the antifungal effect- effective drug on bacterial corneal ulcerations [38]. ive concentration usually used in clinical settings [16, 22]. Moreover, the use of such antimicrobial agent consti- Colistin is one of many AMPs already used in clinical tutes a potential alternative treatment that may improve settings [11]. So, in addition to the colistin-antifungal the outcome in some critical infections caused by MDR combination evaluated in this study, other AMPs could fungi, such as the recent MDR Fusarium keratitis-case further be tested to potentiate the antifungal activity of report in a 46-year-old man who was still declining even existing antifungal compounds. For example, Wakabaya- the maximal therapeutic support and therapeutic kerato- shi et al, previously described the synergistic effect of plasty [36]. lactoferin, a human antimicrobial peptide, with clotrima- Several approaches could be used to overcome the zole against C. albicans [43]. Moreover, lactoferin development of antifungal resistance in the treatment of induced an important decrease of all azoles’ MICs tested fungal diseases. Aside from the discovery of new effect- against azole-resistant Candida spp. [43]. Consequently, ive agents, one realistic alternative option would be to natural or synthetic AMPs, have been identified as an enhance the activity of existing agents. Combination original therapeutic alternative that could be investigated therapies exploit the chances for better efficacy, by medical researchers and pharmaceutical companies. decreased toxicity and reduced development of drug Using the same approach which was used herein, resistance [39]. A previous study demonstrated an in another AMP, less toxic than polymyxins such as bacitra- vitro synergy between colistin and echinocandins in sev- cin or gramicidin analogues, could be tested as monother- eral pathogenic yeasts, namely C. albicans, C. glabrata, apy or in association with antifungals against MDR fungi. C. tropicalis, C. parapsilosis and C. krusei, as well as in fluconazole-resistant C. albicans strains [7]. Conclusion To the best of our knowledge, no previous studies Our findings demonstrate that polymyxins display a have tested the activity of colistin in combination with broad-spectrum activity against common MDR fungi other antifungal agents against fungi of the genera Rho- especially those which are difficult to manage in clinical set- dotorula and Lichtheimia. tings. Unfortunately, polymyxins’ MICs against these MDR A high decrease of colistin’s MICs was observed when strains are higher than those that could clinically be used in it was combined with azoles (with fluconazole against R. human therapy, thus the use of such high toxicity-associ- mucilaginosa and with itraconazole against either C. ated concentration of polymyxins presents the major limita- albicans or L. corymbifera, Table 2 and Fig. 3). It is well tion of their application in clinical mycology practice. known that the main mechanism of action of azoles is However, colistin seems to induce C. albicans membrane the inhibition of enzymes that transform lanosterol into damages and to act in synergy with either itraconazole or ergosterol, a major lipid of the fungal membrane. This amphotericin B (each also acting on the fungal membrane). inhibition alters both the permeability and fluidity of We therefore suggest that colistin (at a ‘safe’ reduced dose) fungal membrane [40]. On the other hand, polymyxins can be used in combination with currently available anti- are well known for weakening the outer membrane in fungal drugs, as a last resort option, against life-threatening Gram-negative bacteria and the disruption of its MDR fungi. Yousfi et al. Antimicrobial Resistance and Infection Control (2019) 8:66 Page 9 of 10

Abbreviations 9. Gulshan K, Moye-Rowley WS. Multidrug resistance in fungi. Eukaryot Cell. AMPs: Antimicrobial peptides; CF: Cystic fibrosis; CLSI: Clinical and Laboratory 2007;6:1933–42. Standards Institute; FIC: Fractional inhibitory concentration; FIX: Fractional 10. Peyclit L, Baron SA, Yousfi H, Rolain JM. Zidovudine: a salvage therapy for inhibitory index; MDR: Multidrug resistant; MFC: Minimal Fungicidal mcr-1 plasmid-mediated colistin-resistant bacterial infections? Int J Concentration; MIC: Minimum inhibitory concentration; PBS: Phosphate- Antimicrob Agents. 2018;(1):11–3. buffered saline; PI: Propidium iodide; PMB: Polymyxin B 11. Bondaryk M, Staniszewska M, Zielińska P, Urbańczyk-Lipkowska Z. Natural antimicrobial peptides as inspiration for Design of a new Generation Acknowledgements Antifungal Compounds. J Fungi. 2017;3. https://doi.org/10.3390/jof3030046. The author thanks Magdalen Lardière and CookieTrad for English reviewing. 12. Landman D, Georgescu C, Martin DA, Quale J. Polymyxins revisited. Clin Microbiol Rev. 2008;21:449–65. Funding 13. Olaitan A, Morand S, Rolain JM. Mechanisms of polymyxin resistance: This work was supported by the French Government under the « acquired and intrinsic resistance in bacteria. Front Microbiol. 2014;5:1–18. Investissements d’avenir » (Investments for the Future) program managed by 14. Nicholls MWN. Polymyxin sensitivity of Candida tropicalis. J Med Microbiol. the Agence Nationale de la Recherche (ANR, fr: National Agency for 1970;3:529–38. Research), (reference: Méditerranée Infection 10-IAHU-03). 15. Ben-Ami R, Lewis RE, Tarrand J, Leventakos K, Kontoyiannis DP. Antifungal This work was supported by the Région Provence-Alpes-Côte-d’Azur and activity of colistin against Mucorales species in vitro and in a murine model European funding FEDER PRIMI. of Rhizopus oryzae pulmonary infection. Antimicrob Agents Chemother. 2010;54:484–90. Availability of data and materials 16. Hsu LH, Wang HF, Sun PL, Hu FR, Chen YL. The antibiotic polymyxin B All data are available in Table 1. exhibits novel antifungal activity against Fusarium species. Int J Antimicrob Agents. 2017;49:740–8. Authors’ contributions 17. Cassagne C, Normand AC, L’Ollivier C, Ranque S, Piarroux R. Performance of JMR and FB were involved in the conception and design of the study. HY MALDI-TOF MS platforms for fungal identification. Mycoses. 2016;59:678–90. performed the experiments. HY, SR, JMR and FB analysed and interpreted 18. Olaitan A, Chabou S, Okdah L, Morand S, Rolain JM. Dissemination of the the data. HY and FB co-wrote the manuscript. All authors have read and mcr-1 colistin resistance gene. Lancet Infect Dis. 2016;16:147. approved the final manuscript. 19. Leangapichart T, Gautret P, Brouqui P, Memish ZA, Raoult D, Rolain JM. Acquisition of mcr-1 plasmid-mediated colistin resistance in Escherichia coli Ethics approval and consent to participate and Klebsiella pneumoniae during hajj 2013 and 2014. Antimicrob Agents – Not applicable. Chemother. 2016;60:6998 9. 20. Lalaoui R, Bakour S, Livnat K, Assous MV, Diene SM, Rolain JM. Spread of Consent for publication Carbapenem and Colistin-resistant Klebsiella pneumoniae ST512 clinical Not applicable. isolates in Israel: a cause for vigilance. Microb Drug Resist. 2018. https://doi. org/10.1089/mdr.2018.0014. 21. Espinel-Ingroff A, Arthington-Skaggs B, Iqbal N, Ellis D, Pfaller MA, Messer S, Competing interests et al. Multicenter evaluation of a new disk agar diffusion method for The authors declare that they have no competing interests. susceptibility testing of filamentous fungi with voriconazole, posaconazole, itraconazole, amphotericin B, and caspofungin. J Clin Microbiol. 2007;45: Publisher’s Note 1811–20. Springer Nature remains neutral with regard to jurisdictional claims in 22. Pankey G, Ashcraft D, Kahn H, Ismail A. Time-kill assay and etest evaluation published maps and institutional affiliations. for synergy with polymyxin B and fluconazole against Candida glabrata. Antimicrob Agents Chemother. 2014;58:5795–800. Author details 23. Kwolek-Mirek M, Zadrag-Tecza R. Comparison of methods used for assessing 1Aix Marseille Univ, IRD, APHM, MEPHI, IHU Méditerranée Infection, 19-21 the viability and vitality of yeast cells. FEMS Yeast Res. 2014;14:1068–79. boulevard Jean Moulin, 13005 Marseille, France. 2Aix Marseille Univ, IRD, 24. Zhai B, Zhou H, Yang L, Zhang J, Jung K, Giam CZ, et al. Polymyxin B, in APHM, SSA, VITROME, IHU-Méditerranée Infection, 19-21 boulevard Jean combination with fluconazole, exerts a potent fungicidal effect. J Moulin, 13005 Marseille, France. Antimicrob Chemother. 2010;65:931–8. 25. Short G, Rennison C, Gould K, Fisher A. Activity of Colistin against Received: 14 December 2018 Accepted: 10 April 2019 filamentous Fungi isolated from lung transplant recipients. J Hear Lung Transplant. 2010;29:S185–6. 26. Buchanan WL, Schaufele RL, Knudsen TA, Sein T, Zaoutis TE, Chiou CC, et al. References Epidemiology and outcome of Zygomycosis: a review of 929 reported 1. Morio F, Jensen RH, Le PP, Arendrup MC. Molecular basis of antifungal drug cases. Clin Infect Dis. 2005;41:634–53. resistance in yeasts. Int J Antimicrob Agents. 2017;50:599–606. 27. Pfaller MA. Antifungal drug resistance : mechanisms , epidemiology , and 2. Geddes-Mcalister J, Shapiro RS. New pathogens, new tricks: emerging, drug- consequences for treatment. Am J Med. 2012;125:S3–13. resistant fungal pathogens and future prospects for antifungal therapeutics. 28. Pellon A, Ramirez-Garcia A, Buldain I, Antoran A, Martin-Souto L, Rementeria Ann N Y Acad Sci. 2018. https://doi.org/10.1111/nyas.13739. A, et al. Pathobiology of Lomentospora prolificans: could this species serve as 3. Dannaoui E. Antifungal resistance in Mucorales. Int J Antimicrob Agents. a model of primary antifungal resistance? Int J Antimicrob Agents. 2017;50:617–21. 2018;51:10–5. 4. Alcazar-fuoli L, Mellado E. Current status of antifungal resistance and its 29. Schemuth H, Dittmer S, Lackner M, Sedlacek L, Hamprecht A, Steinmann E, impact on clinical practice. Br J Haematol. 2014;166:471–84. et al. In vitro activity of colistin as single agent and in combination with 5. Kauffman CA. Fungal infections in older adults. Clin Infect Dis. antifungals against filamentous fungi occurring in patients with cystic 2001;33:550–5. fibrosis. Mycoses. 2013;56:297–303. 6. Kanafani ZA, Perfect JR. Resistance to antifungal agents: mechanisms and 30. Udeani TKC, Moses J, Uzoechina A, Okwori AEJ, Okwosa CN. Microbial clinical impact. Clin Infect Dis. 2008;46:120–8. aetiologic agents associated with pneumonia in immunocompromised 7. Zeidler U, Bougnoux ME, Lupan A, Helynck O, Doyen A, Garcia Z, et al. hosts. African J Infect Dis. 2010;4:1–6. Synergy of the antibiotic colistin with echinocandin antifungals in candida 31. Elborn JS, Hodson M, Bertram C. Implementation of European standards of species. J Antimicrob Chemother. 2013;68:1285–96. care for cystic fibrosis - control and treatment of infection. J Cyst Fibros. 8. Vale Silva LA. Molecular mechanisms of resistance of Candida spp. to 2009;8:211–7. membrane-targeting antifungals. In: Coste AT and Vandeputte P, editors. 32. Abdellatif S, Trifi A, Daly F, Mahjoub K, Nasri R, Ben Lakhal S. Efficacy and Antifungals: from genomics to resistance and the development of novel toxicity of aerosolised colistin in ventilator-associated pneumonia: a agents: Caister Academic Press, Poole, UK; 2015. p. 1–26. prospective, randomised trial. Ann Intensive Care Springer Paris. 2016;6:1-11. Yousfi et al. Antimicrobial Resistance and Infection Control (2019) 8:66 Page 10 of 10

33. Landersdorfer Cornelia B, Nguyen T-H, Linh Thuy Lieu GN, Bischof RJ, Meeusen EN, Li J, Nation RL, et al. Substantial targeting advantage achieved by pulmonary Administration of Colistin Methanesulfonate in a large. Antimicrob Agents Chemother. 2017;61:1–15. 34. Carillo C, Pecoraro Y, Anile M, Poggi C, Oliva A, Amore D, et al. Colistin- based treatment of multidrug-resistant gram-negative bacterial pulmonary infections after lung transplantation. Transplant Proc. 2019;51:202–5. 35. Steinfort DP, Steinfort C. Effect of long-term nebulized colistin on lung function and quality of life in patients with chronic bronchial sepsis. Intern Med J. 2007;37:495–8. 36. Sara S, Sharpe K, Morris S. Multidrug-resistant Fusarium keratitis: diagnosis and treatment considerations. BMJ Case Rep. 2016;3. https://doi.org/10.1136/bcr-2016-215401. 37. Maharana P, Sharma N, Nagpal R, Jhanji V, Das S, Vajpayee R. Recent advances in diagnosis and management of mycotic keratitis. Indian J Ophthalmol. 2016;64:346. 38. Bosscha MI, Van Dissel JT, Kuijper EJ, Swart W, Jager MJ. The efficacy and safety of topical polymyxin B, neomycin and gramicidin for treatment of presumed bacterial corneal ulceration. Br J Ophthalmol. 2004;88:25–8. 39. Foucquier J, Guedj M. Analysis of drug combinations: current methodological landscape. Pharmacol Res Perspect. 2015;3:1–11. 40. François IEJA, Cammue BPA, Borgers M, Ausma J, Dispersyn GD, Thevissen K. Azoles: mode of antifungal action and resistance development. Effect of miconazole on endogenous reactive oxygen species production in Candida albicans. Antiinfect Agents Med Chem. 2006;5:3–13. 41. Trimble MJ, Mlynarcik P, Kolar M, Hancock REW. Polymyxin: alternative mechanisms of action. Cold Spring Harb Perspect Med. 2016;6:1–22. 42. Gray KC, Palacios DS, Dailey I, Endo MM, Uno BE, Wilcock BC, et al. Amphotericin primarily kills yeast by simply binding ergosterol. Proc Natl Acad Sci. 2012;109:2234–9. 43. Wakabayashi H, Abe S, Teraguchi S, Hayasawa H, Yamaguchi H. Inhibition of hyphal growth of azole-resistant strains of Candida albicans by triazole antifungal agents in the presence of lactoferrin- related compounds. Antimicrob Agents Chemother. 1998;42:1587–91.

Article 5:

Repurposing of ribavirin as an adjunct therapy for invasive fungal disease

Hanane Yousfi, Carole Cassagne, Stéphane Ranque, Jean-marc Rolain and Fadi Bittar

Revised in « Antimicrobial Agents and Chemotherapy »

1 Title: Repurposing of ribavirin as an adjunct therapy for invasive fungal disease

2 Running title: In vitro ribavirin activity against Candida species.

3 Authors list: Hanane Yousfi1,3, Carole Cassagne2,3, Stéphane Ranque2,3, Jean-marc Rolain1,3

4 and Fadi Bittar1,3*

5 Affiliations :

1 6 Aix Marseille Univ, IRD, APHM, MEPHI, Marseille, France.

7 2 Aix Marseille Univ, IRD, APHM, SSA, VITROME, Marseille, France.

8 3 IHU Méditerranée Infection, Marseille, France

9 *Corresponding author:

10 Fadi Bittar, IHU Méditerranée Infection, 19-21 boulevard Jean Moulin, 13005 Marseille,

11 France. Email: [email protected]

12 Abstract words count: 247

13 Importance words count: 94

14 Number of words: 2526

15 Number of tables: 1

16 Number of figures: 3

17 Supplementary data: Table S1

18 Number of references: 35 19 Abstract

20 The use of antifungal agents in clinical settings is limited by the appearance of drug resistance

21 and adverse side effects. There is, therefore, an urgent need to develop new drugs to

22 strengthen the treatment of invasive fungal diseases. The aim of this study is to describe the

23 potential repurposing of ribavirin as an adjunct therapy against Candida spp.

24 Primary screening of Prestwick chemical library against Candida albicans ATCC 90028 and

25 fluconazole-resistant Candida albicans was performed. Subsequently, we evaluated the

26 response of 100 Candida spp strains to ribavirin, an antiviral agent, using the broth

27 microdilution method. We checked the involvement of efflux pump activity in the

28 development of ribavirin-resistance. We studied time-kill curves and performed a

29 checkerboard assay for ribavirin-antifungals combinations study.

30 Twenty-one nonstandard antifungal compounds were identified, including ribavirin. Ribavirin

31 had, in vitro, an antifungal activity against 63 Candida strains including C. albicans, C.

32 parapsilosis and C. tropicalis, with a minimum inhibitory concentrations (MICs) ranging

33 from 0.37 to 3.02 µg mL-1, while MICs for C. krusei, C. glabrata, C. lusitaniae and some C.

34 albicans remain high (≥ 24.16 µg mL-1). No relation was observed between efflux pump

35 activity and ribavirin-resistance. Ribavirin exhibited a fungistatic activity against multidrug-

36 resistant (MDR) C. albicans and a fungicidal activity against C. parapsilosis strain. In

37 addition, ribavirin acted synergistically with either ﬂuconazole, voriconazole or amphotericin

38 B against MDR-C. albicans strain.

39 The presented study highlights the potential clinical application of ribavirin, alone or in

40 association with other antifungal agents, as an adjunct anti-Candida drug. 41 Importance

42 The rise of new fungal agents emergence and drug-resistant strains requires more researches

43 for the development of new and/or alternative drugs. Consequently, our in vitro study

44 addresses some of these current therapeutic issues by repurposing ribavirin, alone or in

45 association with either amphotericin B, or voriconazole or fluconazole, as a candidate for the

46 treatment of infection by Candida spp, especially Candida albicans, C. parapsilosis, and C.

47 tropicalis. Although the exact mechanism of action is unknown, further in vivo studies are

48 needed to explore the clinical interest of ribavirin in the treatment of invasive fungal diseases.

61 62 Introduction

63 Fungi are considered to be one of the main causes of human infections, particularly among

64 immunocompromised individuals and hospitalized patients with serious immunosuppressive

65 conditions such as HIV and organ transplantation (1). Infections due to Candida spp are one

66 of the most common invasive fungal diseases (2). In a recent study, Candida spp were

67 identified as the most frequent cause of bloodstream infection in hospitalized patients (3).

68 Only three classes of antifungal agents are currently used to treat invasive Candida species

69 infections, namely azoles (ex. fluconazole), polyenes (ex. amphotericin B) and echinocandins

70 (ex. caspofungin) (4). Although these classes of antifungal agents are usually active against

71 many infections, their applicability and efﬁcacy could be compromised due to their high

72 toxicity and the development of drug resistance (5). Hence, there is a constant need for other

73 compounds that possess antifungal properties. An interesting thought in this field has been

74 addressed via the drug repurposing, where FDA-approved drugs could be tested and used in

75 another therapeutic class (6). Applying this approach by screening 1,920 compounds

76 belonging to three FDA-libraries drugs against Candida albicans strains, Tournu et al.,

77 recently identified, ribavirin, a purine nucleoside analog, as a potential C. albicans vacuole

78 disrupting agent (7). Ribavirin is a guanosine analog that produces broad-spectrum activity

79 against many RNA and DNA viruses. It was discovered in 1972 by Witkowski and coworkers

80 (8). Following this discovery, many investigations have demonstrated its efficacy in the

81 treatment of pediatric respiratory syncytial virus infection in 1986 (9). Over a decade later,

82 ribavirin was used to treat hepatitis C virus (HCV) in combination with interferon-α (10) and

83 since then ribavirin is being used essentially for this indication. Here, we investigated the in

84 vitro activity of ribavirin against different clinical Candida species, including emerging

85 fluconazole-resistant Candida albicans. We also evaluated the eventual synergy between

86 ribavirin and common antifungal agents. 87 Results and discussion

88 By testing the FDA-approved library of 1,280 drugs against C. albicans ATCC 90028

89 (quality control) and C. albicans (Q181103513) that have high resistance to fluconazole and

90 all echinocandin antifungal agents, 21 non-standard antifungal hits showed fungal growth

91 inhibition greater than 90%. These primary compounds include 7 antibacterial drugs, 3

92 antihelmintic agents and 11 compounds belonging to different therapeutic classes (Table 1).

93 The principal aim of our study is to identify new molecules whose antifungal action is not

94 known and/or well investigated and which can be used in clinical practices without any

95 constraints for patients. Among these 21 hits obtained, 14 of them have already been well

96 identified and studied in previous works (Table 1), confirming the validity and the

97 reproducibility of our results and our screening technique. However, to our knowledge, the

98 fungal growth inhibition induced by the remaining 7 compounds including pinaverium

99 bromide, avermectin B1, triclabendazole, tetraethylenepentamine pentahydrochloride,

100 thioguanine, and anthralin, has never been previously reported neither well studied as well as

101 the ribavirin compound (Table 1).

102 Among these unknown antifungal hits, we focused our study on ribavirin drug, which presents

103 a percentage of fungal growth reduction equal to 96%, for several purposes; (i) Ribavirin has

104 been shown to disrupt vacuolar function in the pathogen C. albicans (7), however, no report

105 describes directly its antifungal efficacy. (ii) Most of these 7 hits have been designed for a

106 specific non-infectious indication (antispastic, antilipemic, antineoplastic, antipsoriatic)

107 (Table 1) whereas ribavirin was approved to treat liver infectious disease (11), so its

108 repurposing as an antifungal agent could be admissible. (iii) Reversible hemolytic anemia is

109 the unique adverse effect observed with ribavirin treatment, unlike the cytotoxic effect that of

110 other molecules, such as thioguanine and avermectin B1, may have on human cells due to

111 their non-selective activities (12, 13). (iv) Ribavirin can be administered orally (capsule or 112 tablet), intravenously and through inhalation (11), while topical application is only possible

113 for some molecules such as anthralin (14).

114 That is why, in this study, ribavirin susceptibility testing against a large panel of 100 Candida

115 spp strains was performed, and ribavirin MICs were determined. Consequently, we report here

116 the efficacy of this compound against a collection of 60 Candida spp strains (total No = 100

117 strains), including 6 C. parapsilosis, 5 C. tropicalis, and 49 C. albicans, with MIC ranging

118 from 1.56 to 12.5 µM (i.e. 0.37 to 3.02 µg mL-1). Among these 60 clinical strains tested, C.

119 parapsilosis and C. tropicalis showed the greatest susceptibility to ribavirin with MICs

120 mainly ≤ 3.02 µM (i.e. ≤ 0.75 µg mL-1). Interestingly, ribavirin was effective against the

121 clinical strain that was resistant to both fluconazole and echinocandin agents (C. albicans

122 Q181103513), with a low MIC equal to 6.25 µM (i.e. 1.51 µg mL-1). On the other hand, the

123 ribavirin MICs obtained against C. krusei, C. lusitaniae, C. auris, and C. glabrata were

124 mainly ≥ 100 µM (i.e. ≥ 24.16 µg ml-1) (Fig 1).

125 In order to study fungicidal and fungistatic activities of ribavirin compound, time-kill curves

126 were performed with C. albicans and C. parapsilosis. Ribavirin exhibited fungistatic activity

127 against MDR C. albicans and C. albicans ATCC 90028 strains tested. At ribavirin

128 concentrations equal to 0.5 X MIC, no inhibitory effect was observed and the curve was

129 nearly identical to those for the control. At concentrations ≥ 1 X MIC, a concentration-

130 independent fungistatic effect was observed. In contrast, ribavirin fungicidal activity against

131 C. parapsilosis was noted with Minimal Fungicidal Concentration (MFC) ≥ to two-fold its

132 MIC value ( MFC = 3.12 µM; i.e. 1.5 µg/mL) (Fig 2).

133 Ribavirin is part of the WHO Model List of Essential Medicines, as an effective and safe drug

134 used in a health system. Ribavirin is administered orally at a dose of 1,200 mg (600 mg twice

135 a day) during 4 weeks to achieve a stable plasmatic concentration of 2.2 µg mL-1 (11). A

136 retrospective study performed by J.F. Jen et al, on patients infected with the HCV, showed 137 that 49% of the sustained virological response was achieved at week 4 via ribavirin plasmatic

138 concentrations that ranged from 3.5 to 4 µg mL-1. In addition, the authors preconized the

139 achieving of higher ribavirin plasmatic concentrations in the treatment of a patient infected

140 with HCV genotype 1 (15). Moreover, a prospective open-label study in patients infected with

141 HCV genotype 1, demonstrated that the increase in the ribavirin treatment dose (maximum

142 daily dose of 3,600 mg) resulted in ribavirin plasmatic concentrations of 5.37 µg mL-1. In this

143 case, undetectable HCV RNA on 9/10 patients during the fourth week was obtained but

144 associated with more frequent side effects, such as anemia (16). Thus, plasma concentrations

145 equal to the MICs obtained from most of the Candida spp strains tested in this study can be

146 easily achieved. The main side effect of ribavirin treatment is reversible hemolytic anemia

147 that may be observed in patients with chronic hepatitis C (17). The use of this drug for a short

148 time therapy, along with the use of hematopoietic growth factors, can manage the risk of

149 severe anemia in maintaining an effective concentration.

150 Nonetheless, some Candida strains were resistant to ribavirin, including some C. albicans

151 strains and all the tested C. glabrata, C. krusei and C. lusiatnia strains. Among the many drug

152 resistance mechanisms that have been described in yeast, one specific or non-specific

153 mechanism is the overexpression of membrane-associated carriers acting as drug efﬂux

154 pumps (18). Thus, we supposed that reduced permeability of the membrane to ribavirin drug

155 and the potential role of efflux pumps in the extrusion of ribavirin molecule can be the

156 mechanism which explains the ribavirin-resistance. That is why we checked the efflux-pumps

157 activity using CCCP (carbonyl cyanide 3-chloro-phenylhydrazone) nonspecific efflux-pumps

158 inhibitor and the specific calcium channel-blocker; verapamil against ribavirin-resistant

159 strains. The activity of ribavirin did not change with the addition of CCCP or verapamil. No

160 changes were reported in the ribavirin MICs of all strains tested after EPIs treatment, which

161 could indicate the potential normal transport of ribavirin in the isolates tested, although we did 162 not quantify the intracellular level of ribavirin. This result leads us to the possibility of the

163 existence of a specific pathway that could induce the development of ribavirin resistance in

164 yeasts.

165 Mutation frequencies of C. albicans ATCC on RPMI medium containing 1.52, 3.04, 6.08 µg

166 mL-1 of ribavirin were 1.5x10-4, 1.1x10-4, 9.2x10-5 respectively. Mutation frequencies of C.

167 albicans (Q181103513) on RPMI-agar plates with 3.04, 6.08 and 12.16 µg mL-1 of ribavirin

168 were 1.6x10-4, 1.13x10-4, 9.8x10-4 respectively. The mutation frequency was approximatively

169 about 10-4 which is relatively higher than that reported for classical antifungals (19). This

170 result indicates that it is preferable to use ribavirin for short time or in association with other

171 antimicrobial drugs. Therefore, in addition to the discovery of new agents effective against

172 fungi, a pragmatic alternative would be to improve the activity of old antifungal agents by

173 using a combination of drugs. Thus, testing the ribavirin-antifungals combinations could be

174 very useful to increase the antimicrobial spectrum of the usual antifungals used and to

175 minimize the development of drug resistance and side effects by reducing the drug

176 concentration. Ribavirin with amphotericin B, fluconazole and voriconazole combinations

177 exhibited a synergistic activity against MDR C. albicans. We found that the MIC of

178 amphotericin B is decreased from 1 µg/mL to 0.005 µg/ml when associated to 0.37 µg/ml of

179 ribavirin compound, inducing a total fungal growth inhibition (Figure 3) with FICI= 0.25. The

180 MIC of voriconazole decreased from 128 µg/ml to 0.5 µg/ml with ribavirin concentration

181 equal to 0.37 µg/ml. Interestingly, ribavirin had synergistic activity with fluconazole against

182 fluconazole-resistant C. albicans, in which ribavirin concentration was lowered from 1.5

183 µg/ml to 0.18 µg/ml when combined with a concentration of fluconazole equal to 1 µg/ml

184 (Fig 3). However, there is no interaction between ribavirin and the antifungal agents when

185 tested against ribavirin-resistant species, as C. glabrata, C. krusei, and C. lusitaniae. 186 Mechanisms of action of ribavirin on the HCV have been correctly elucidated, where four

187 pathways were described; (i) inhibition of HCV replication (RNA polymerase), (ii) inhibition

188 of IMPDH enzyme, (iii) induction of RNA mutagenesis and (iv) adaptation of immune

189 response to the HCV(11). In contrast, the mechanism of ribavirin action on Candida species

190 remains unclear. Thus, the most important perspective of this study will be the

191 characterization of the mode of action of ribavirin against Candida species.

192 Materials and methods

193 Screening of 1280 FDA-approved drugs against C. albicans

194 Using the same concept of drugs-repurposing, an initial FDA-approved library of 1,280 drugs

195 (Prestwick, Illkirch graffenstaden, France) (20) screen was performed against Candida

196 albicans ATCC 90028 (quality control) and clinical Candida albicans strain, which is

197 resistant to fluconazole and all echinocandins agents at a fixed concentration of 10 µM. The

198 fungal inoculum was prepared using RPMI-1640 medium (Sigma Aldrich, St Louis, France)

199 according to the Clinical and Laboratory Standards Institute (CLSI) protocol. Sixteen of 96-

200 well plates were used. Each plate contains 80 compounds, the first column served for the

201 positive control with untreated fungi and the last one contains the medium as negative control.

202 The plates were incubated 24 hours at 37°C. Then, the optical density (OD) was obtained

203 using a spectrophotometer and the percentages of fungal growth inhibition were calculated in

204 relation to the untreated wells.

205 Fungal strains

206 A collection of 96 Candida spp strains was tested, including 72 C. albicans, 5 Candida

207 parapsilosis, 8 Candida glabrata, 5 Candida tropicalis, 3 Candida krusei and 3 Candida

208 lusitaniae strains with different phenotypic profiles (Table S1). The species of all isolates

209 were identified using Matrix-assisted laser desorption/ionization mass spectrometry (MALDI- 210 TOF MS) (21). All strains were recovered from La Timone University Hospital in Marseille

211 and were isolated from different clinical samples, mainly from blood culture, urine and

212 vaginal swabs (Supplementary Table 1). C. albicans ATCC 90028, C. parapsilosis ATCC

213 22019, C. auris DSMZ 21092 and C. krusei ATCC 6258 were used as quality control (Table

214 S1).

215 Antifungals susceptibility testing

216 Antifungal susceptibility testing was performed using commercial broth microdilution plates;

217 Sensititre®YeastOne® (Thermo Fisher Scientific, Schwerte, Germany) containing 9

218 antifungals belonging to the 4 therapeutic classes, namely polyene (amphotericin B), azoles

219 (fluconazole, posaconazole, voriconazole and itraconazole), echinocandin (anidulafungin,

220 caspofungin and micafungin) and 5-flucytosine. The minimum inhibitory concentrations

221 (MICs) obtained for Candida spp strains tested (Table S1) were compared to the breakpoints

222 provided by the manufacturers or, to the Clinical and Laboratory Standards Institute (CLSI)

223 epidemiological cutoffs, to assess the susceptibility of each strain to the different antifungal

224 agents tested.

225 Ribavirin susceptibility testing

226 Using the broth microdilution method as outlined by the CLSI, ribavirin MICs were

227 performed. Serial ribavirin (Sigma Aldrich, St Louis, France) dilutions ranging from 0.2 to

228 100 µM (i.e. 0.047 to 24.16 µg mL-1) were tested in RPMI-1640 medium. The MIC of

229 ribavirin was defined as the concentration that resulted in complete growth inhibition

230 compared to untreated control wells. This later was determined visually after 24 hours of

231 incubations at 37°C using OD by spectrophotometric measurement.

232 Time-kill assay 233 In order to assess the fungicidal or fungistatic activity of ribavirin compound against three

234 fungal strains (2 C. albicans and 1 C. parapsilosis), ribavirin time-kill analysis was performed

235 as previously described (22); (C. albicans ATCC 90028 was used as quality control). The

236 ribavirin action was investigated at 4 different concentrations (0.5xMIC, 1xMIC, 2xMIC,

237 4xMIC) and the fungal growth was followed-up by the CFU mL-1 number after 2, 6, 12, 24,

238 36, 48 hours of incubation at 37°C. The results were analyzed using the GraphPad Prism 5.3

239 software (GraphPad Inc., San Diego, CA, USA).

240 Testing of efflux-pumps activity

241 The implication of efflux-pumps in ribavirin resistance mechanism in some ribavirin-resistant

242 Candida spp strains (2 C. glabrata, 1 C. krusei, 1 C. lusiataniae, 1 C. albicans) was verified

243 using CCCP (carbonyl cyanide 3-chloro-phenylhydrazone) and verapamil inhibitors, which

244 decrease the MICs when resistance was correlated to this mechanism (23). The CCCP and

245 verapamil inhibitors were tested at two final concentrations (0.5 µg mL-1, 10 µg mL-1). To

246 ensure that these concentrations do not affect yeast growth, control growth wells containing

247 efflux-pumps inhibitors (EPIs) alone in RPMI medium were added. We determined ribavirin

248 MICs for all strains before and after adding EPIs to the RPMI medium.

249 Mutant frequency

250 The frequency of spontaneous mutations was determined on C. albicans ATCC 90028 and

251 multidrug-resistant (MDR)- C. albicans (Q181103513) by plating 100 µl of growing yeast

252 with appropriate dilution (104 CFU mL-1) on ribavirin-free plates and without dilution (106

253 CFU mL-1) on ribavirin-containing plates (19). RPMI-1640 agar plates with different ribavirin

254 concentrations (1xMIC, 2xMIC, 4xMIC) were prepared as previously described (24).

255 Colonies grown after 24 hours of incubation at 37°C were counted. The mutation frequency is 256 the ratio of the mutants obtained (CFU mL-1) divided by the total inoculum plated (106 CFU

257 mL-1) on ribavirin-agar plates.

258 Synergy evaluation of ribavirin-antifungals combinations

259 In order to exploit the potential for improved efficacy, reduced toxicity and reduced risk of

260 drug resistance development (25), ribavirin-antifungal agent combinations were tested against

261 6 Candida spp strains, including the MDR C. albicans strain (Q181103513), C. albicans

262 ATCC 90028, C. albicans (Q181208813), C. glabrata (Q181198565), C. glabrata

263 (Q181203672) and C. krusei (Q181208438). Based on the checkerboard association assay,

264 combinations of ribavirin with each of the following antifungals: amphotericin B, fluconazole

265 and voriconazole (Sigma Aldrich, St Louis, France) were tested. The effects of the various

266 combinations of the two compounds were analyzed using the Fractional Inhibitory

267 Concentration Index (FICI) calculated as previously described (25). 268 Acknowledgments

269 We thank CookieTrad for English reviewing.

270

271 Funding information

272 This work was supported by the French Government under the «Investissements d’avenir »

273 (Investments for the Future) program managed by the Agence Nationale de la Recherche

274 (ANR, fr: National Agency for Research), (reference: Méditerranée Infection 10-IAHU-03)

275 This work was supported by Région Provence-Alpes-Côte d’Azur and European funding

276 FEDER PRIM

277 Transparency declarations

278 No conflict of interest or financial disclosure to declare for all authors. 279 References

280 1. Geddes-Mcalister J, Shapiro RS. 2018. New pathogens, new tricks: Emerging, drug-

281 resistant fungal pathogens and future prospects for antifungal therapeutics. Ann N Y

282 Acad Sci 1435:57–78.

283 2. Singh R, Parija SC. 2012. Candida parapsilosis: An emerging fungal pathogen. Indian

284 J Med Res 136:671–673.

285 3. Spivak ES, Hanson KE. 2018. Candida auris: an Emerging Fungal Pathogen. J Clin

286 Microbiol 56:e01588-17.

287 4. Bassetti M, Peghin M, Timsit JF. 2016. The current treatment landscape: Candidiasis. J

288 Antimicrob Chemother 71:ii13-ii22.

289 5. Pfaller MA. 2012. Antifungal Drug Resistance : Mechanisms , Epidemiology , and

290 Consequences for Treatment. American Journal Medicine 125:S3–S13.

291 6. Ogundeji AO, Pohl CH, Sebolai OM. 2016. Repurposing of Aspirin and Ibuprofen as

292 Candidate Anti- Cryptococcus Drugs. Antimicrob Agents Chemother 60:4799–4808.

293 7. Tournu H, Carroll J, Latimer B, Dragoi AM, Dykes S, Cardelli J, Peters TL, Eberle

294 KE, Palmer GE. 2017. Identification of small molecules that disrupt vacuolar function

295 in the pathogen Candida albicans. PLoS One 12:1–21.

296 8. Witkowski JT, Robins RK SR. 1972. Design, synthesis, and broad spectrum antiviral

297 activity of 1-D-ribofuranosyl-1, 2, 4-triazole-3-carboxamide and related nucleosides.

298 Journal Medical Chemother 15: 1150–1154.

299 9. Krilov LR. 2001. Respiratory Syncytial Virus: Update on Infection, Treatment, and

300 Prevention. Curr Infect Dis Rep 3:242–246. 301 10. Te HS, Randall G, Jensen DM. 2007. Mechanism of action of ribavirin in the treatment

302 of chronic hepatitis C. Gastroenterol Hepatol (N Y) 3:218–25.

303 11. Thomas E, Ghany MG, Liang TJ. 2012. The Application and Mechanism of Action of

304 Ribavirin in Therapy of Hepatitis C. Antivir Chem Chemother 23:1–12.

305 12. Truong M, Monahan LG, Carter DA, Charles IG. 2018. Repurposing drugs to fast-track

306 therapeutic agents for the treatment of cryptococcosis. PeerJ 1–18.

307 13. Soyuncu S, Oktay C, Berk Y EC. 2007. Abamectin intoxication with coma and

308 hypotension. Clin Toxicol 3:290–300.

309 14. Wu SZ, Wang S, Ratnaparkhi R BW. 2018. Treatment of pediatric alopecia areata with

310 anthralin: A retrospective study of 37 patients. Pediatr Dermatol 6:817–820.

311 15. Jen JF, Glue P, Gupta S, Zambas D HG. 2000. population pharmacokinetic and

312 pharmacodynamic analysis of ribavirin in patients with chronic hepatitic C. Ther drug

313 monito 22:555–565.

314 16. Lindahl K, Stahle L, Bruchfeld A SR. 2005. High-dose ribavirin in combination with

315 standard dose peginterferon for treatment of patients with chronic hepatitis C.

316 Hepatology 41:275–279.

317 17. Bodenheimer HC, Lindsay KL, Davis GL, Lewis JH, Thung SN, Seeff LB. 1997.

318 Tolerance and efficacy of oral ribavirin treatment of chronic hepatitis C: a multicenter

319 trial. Hepatology 26:473–7.

320 18. Prasad Dr R, Rawal MK. 2014. Efflux pump proteins in antifungal resistance. Front

321 Pharmacol 5:1–13.

322 19. Locke JB, Almaguer AL, Zuill DE, Bartizal K. 2016. Characterization of in vitro

323 resistance development to the novel echinocandin CD101 in Candida species. 324 Antimicrob Agents Chemother 60:6100–6107.

325 20. Peyclit L, Baron SA, Yousfi H, Rolain JM. 2018. Zidovudine: A salvage therapy for

326 mcr-1 plasmid-mediated colistin-resistant bacterial infections?. Int J Antimicrob

327 Agents 52:11–13.

328 21. Cassagne C, Normand AC, L’Ollivier C, Ranque S, Piarroux R. 2016. Performance of

329 MALDI-TOF MS platforms for fungal identification. Mycoses 59:678–690.

330 22. Klepser ME, Wolfe EJ, Jones RN, Nightingale CH, Pfaller MA. 1997. Antifungal

331 Pharmacodynamic Characteristics of Fluconazole and Amphotericin B Tested against

332 Candida albicans. Antimicrob Agents Chemother 41:1392–1395.

333 23. Cuevas O, Pela T, Sa M. 2006. Fluconazole resistance mechanisms in Candida krusei :

334 The contribution of efflux-pumps. Med Mycol 575–578.

335 24. Jorgensen JH, Pfaller MA, Carroll KC, Funke G, Landry ML, Richter SS, Warnock D.

336 2015. Manual of Clinical Microbiology, 11th ed.

337 25. Foucquier J, Guedj M. 2015. Analysis of drug combinations: current methodological

338 landscape. Pharmacol Res Perspect 3:1–11.

339 26. Wiederhold NP, Patterson TF, Srinivasan A, Chaturvedi AK, Fothergill AW, Wormley

340 FL, Ramasubramanian AK, Lopez-Ribot JL. 2017. Repurposing auranofin as an

341 antifungal: In vitro activity against a variety of medically important fungi. Virulence

342 8:138–142.

343 27. Khan S, Singhal S, Mathur T, Upadhyay DJ, Rattan A, Range MIC. 2007. Antifungal

344 potential of Disulfiram. Med Mycol 48:109–113.

345 28. De Cremer K, Lanckacker E, Cools TL, Bax M, De Brucker K, Cos P, Cammue BPA, 346 Thevissen K. 2015. Artemisinins, new miconazole potentiators resulting in increased

347 activity against Candida albicans biofilms. Antimicrob Agents Chemother 59:421–426.

348 29. Miletti KE, Leibowitz MJ. 2000. Pentamidine inhibition of group I intron splicing in

349 Candida albicans correlates with growth inhibition. Antimicrob Agents Chemother

350 44:958–966.

351 30. Olsen I. 2014. Attenuation of Candida albicans virulence with focus on disruption of its

352 vacuole functions. J Oral Microbiol 6: 1–6.

353 31. Kasper L, Miramn P, Jablonowski N, Wisgott S, Wilson D, Brunke S, Hube B. 2015.

354 Antifungal activity of clotrimazole against Candida albicans depends on carbon

355 sources, growth phase and morphology. J Med Microbiol 64:714–723.

356 32. Calamari SE, Bojanich MA, Barembaum SR, Berdicevski N, Azcurra AI. 2011.

357 Antifungal and post-antifungal effects of chlorhexidine, fluconazole, chitosan and its

358 Combinations on Candida albicans. Med Oral Patol Oral Cir Bucal 16:23–28.

359 33. Mendling W, Weissenbacher ER, Gerber S, Prasauskas V, Grob P. 2016. Use of locally

360 delivered dequalinium chloride in the treatment of vaginal infections: a review. Arch

361 Gynecol Obstet 293:469–484.

362 34. Kim K, Zilbermintz L, Martchenko M. 2015. Repurposing FDA approved drugs against

363 the human fungal pathogen, Candida albicans. Ann Clin Microbiol Antimicrob 14:1–

364 11.

365 35. Niewerth M, Kunze D, Seibold M, Schaller M, Korting HC, Hube B. 2003. Ciclopirox 366 olamine treatment affects the expression pattern of Candida albicans genes encoding

367 virulence factors, iron metabolism proteins, and drug resistance factors. Antimicrob

368 Agents Chemother 47:1805–1817.

369 370 Figures legends

371 Fig 1. Distribution of ribavirin MICs for all clinical Candida spp strains tested in this study.

372 C.parapsilosis (brown), C. glabrata (green), C. auris (yellow), C. albicans (red), C.

373 krusei (blue), C. lusitaniae (black), C. tropicalis (grey), MICs – Minimum Inhibitory

374 Concentrations.

-1 375 Fig 2. Values for Log10 of number of CFU mL versus time for Candida strains tested

376 against ribavirin; C. albicans Q181103513 (A), C. albicans ATCC 90028 (B) and C.

377 parapsilosis Q181208447 (C) at the following concentrations: control (■ filled square),

378 0.5XMIC (▲ triangle), 1XMIC (● circle), 2XMIC (♦ lozenge), 4XMIC (─ line).

379 Fig 3. Ribavirin-antifungal agent associations testing, against MDR-Candida albicans

380 Q181103513.

381 A; Chequerboard assays using resazurin coloration indicate fungal growth (Pink) and 100%

382 of fungal growth inhibition (Blue).

383 B; Results from FICI calculation of various ribavirin-antifungal combinations. A synergistic

384 effect was considered when FICI is ≤ 0.5 (Green) and no interaction was considered if FICI is

385 > 0.5 (Red).

386 PC – Positive control (Untreated Candida albicans strain), NC – Negative control (RPMI

387 medium). 388

Drug Therapeutic class Fungal growth inhibition (%) References Auranofin Analgesic 99 26 Disulfiram Antabuse effect 98 27 Pinaverium bromide Antispastic 98 Unknown Avermectin B1 Antihelmintic 98 Unknown Pyrvinium pamoate Anthelmintic 97 28 Triclabendazole Anthelmintic 96 Unknown Pentamidine isethionate Antiprotozoal 96 29 Tetra ethylenepentamine pentahydrochloride Antilipemic 94 Unknown Ribavirin Antiviral 96 7 Pentetic acid Chelating/Radioprotectant 98 29 Thioguanosine Antineoplastic 96 Unknown Anthralin Antipsoriatic 96 Unknown Thonzonium bromide Antiseptic 98 30 Clioquinol Antiamebic/Antiseptic 98 7 Clotrimazole Antibacterial 98 31 Chloroxine Antibacterial 98 32 Chlorhexidine Antibacterial/Antiseptic 98 32 Dequalinium dichloride Antibacterial/Antiseptic 98 33 Methyl benzethonium chloride Antibacterial 98 34 Benzethonium chloride Antibacterial/Antiseptic 98 34 Ciclopirox ethanolamine Antibacterial 97 35 389 Table 1. Hits obtained by the primary screening of 1,280 FDA-approved drugs of Prestwick chemical library at 10 µM against both C. albicans

390 ATCC 90028 and fluconazole-resistant C. albicans (strain Q181103513). 1 2 Fig 1. Distribution of ribavirin MICs for all clinical Candida spp strains tested in this study.

3 C. parapsilosis (brown), C. glabrata (green), C. auris (yellow), C. albicans (red), C. krusei

4 (blue), C. lusitaniae (black), C. tropicalis (gre), MICs – Minimum Inhibitory Concentrations. 5 A.

6 7 B.

8 9 C.

10 -1 11 Fig 2. Values for Log10 of number of CFU mL versus time for Candida strains tested

12 against ribavirin; C. albicans Q181103513 (A), C. albicans ATCC 90028 (B) and C.

13 parapsilosis Q181208447 (C) at the following concentrations: control (■ square), 0.5XMIC

14 (▲ triangle), 1XMIC (● circle), 2XMIC (♦ lozenge), 4XMIC (─ line). 15 (A) (B)

18 19 Fig 3. Ribavirin-antifungal agent associations testing, against MDR-Candida albicans

20 Q181103513.

21 A; Chequerboard assays using resazurin coloration indicate fungal growth (Pink) and 100%

22 of fungal growth inhibition (Blue).

23 B; Results from FICI calculation of various ribavirin-antifungal combinations. A synergistic

24 effect was considered when FICI is ≤ 0.5 (Green) and no interaction was considered if FICI is

25 > 0.5 (Red). 26 PC – Positive control (Untreated Candida albicans strain), NC – Negative control (RPMI

27 medium).

Suplementary Table 1: MIC values of ribavirin and the antifungals tested against 100 Candida spp strains used in this study. Values are in µg mL-1 , except the last column (µm)

les

Ribavirinµm

Ribavirin

Voriconazole

Posaconazole

Itraconazole

Clinicalsamp

Fluconazole

5-Flucytosine

Strainsspecies Micafungin

Caspofungin

References

Anidulafungin

Strain number AmphotericinB 1 Q181201658 C. parapsilosis Urine 0.25 1 1 1 0.25 0.25 0.3 0.015 <0.008 0.38 1.56 2 Q181208447 C. parapsilosis Blood culture 0.5 2 0.03 0.015 0.5 0.25 0.06 0.03 <0.008 0.76 3.12 3 8070833697 C. parapsilosis Blood culture 1 1 0.5 1 0.12 2 0.06 0.06 0.03 0.76 3.12 4 ATCC 22019 C. parapsilosis / 0.5 0.5 0.12 0.5 0.12 1 0.008 0.03 0.03 0.76 3.12 5 Q181261149 C. parapsilosis Blood culture 0.5 1 0.5 1 0.25 0.12 0.03 0.015 <0.008 0.38 1.56 6 Q181217615 C. parapsilosis Blood culture 0.5 1 1 1 <0.06 2 0.06 0.03 0.008 0.38 1.56 7 Q181201614 C. glabrata Kehr's drain 1 0.06 0.03 0.5 <0.06 8 8 1 0.25 >24.4 >100 8 Q181198565 C. glabrata Blood culture 0.5 0.03 0.06 0.03 <0.06 64 1 8 2 >24.4 >100 9 Q181203672 C. glabrata Blood culture 0.5 0.12 0.25 0.03 <0.06 128 1 >8 2 >24.4 >100 10 Q181202903 C. glabrata Sputum 1 0.03 0.03 0.5 0.06 16 4 8 1 >24.4 >100 11 Q181255715 C. glabrata Urine 0.5 <0.015 0.03 0.015 <0.06 64 16 4 8 >24.4 >100 12 Q181280604 C. glabrata Peritoneal fluid 1 0.015 0.06 0.015 <0.06 4 0.25 0.5 0.25 24.4 100 13 8070855959 C. glabrata Blood culture 1 <0.015 0.25 0.015 <0.06 128 16 8 8 >24.4 >100 14 Q181208658 C. glabrata Urine 1 <0,015 0.25 0.015 <0.06 128 16 8 8 24.4 100 15 DSM 21092 C. auris / 0.5 0.12 0.03 0.12 <0.06 4 0.06 0.015 0.03 24.4 100 16 ATCC 9028 C. albicans / 0.25 0.06 0.06 <0.008 0.12 0.25 0.06 0.03 <0.008 3.05 12.5 17 Q181103513 C. albicans Nails 1 8 >8 8 <0.06 256 0.25 0.25 0.12 1.52 6.25 18 Q181202764 C. albicans Prosthetic materiel 1 0.12 0.06 <0.008 0.12 <0.12 0.06 0.03 <0.008 3.05 12.5 19 Q181201638 C. albicans Drain fluid 1 0.03 0.03 <0.008 <0.06 0.12 0.015 0.015 <0.008 12.2 50 20 Q181201672 C. albicans Urine 0.25 1 1 1 1 0.25 0.03 0.015 <0.008 0.76 3.12 21 Q181202033 C. albicans Peritoneal fluid 1 0.06 0.015 <0.008 2 0.25 0.03 0.03 0.008 3.05 12.5 22 Q181199096 C. albicans Gastric biopsy 1 0.03 0.06 0.03 <0.06 0.5 0.06 0.03 0.008 3.05 12.5 23 Q181201157 C. albicans Vaginal swab <0.015 0.06 1 0.03 0.5 0.06 0.015 <0.008 0.03 3.05 12.5 24 Q181201404 C. albicans Lung biopsy 0.5 0.3 0.03 <0.008 0.25 0.5 0.06 0.03 <0.008 3.05 12.5 25 Q181207523 C. albicans Vaginal swab 0.5 0.015 0.03 0.008 <0.06 <0.12 0.015 0.008 <0.008 6.1 25 26 Q181207575 C. albicans Vaginal swab 1 0.015 0.015 0.015 0.5 0.5 0.12 0.06 <0.008 3.05 12.5 27 Q181208108 C. albicans Peritoneal fluid 1 0.06 0.06 0.015 <0.06 0.5 0.03 0.03 0.015 >24.4 >100 28 Q181207653 C. albicans Urine 1 0.015 0.03 0.015 <0.06 0.5 0.03 0.03 0.008 6.1 25 29 Q181207186 C. albicans Urine 0.5 0.03 0.03 0.008 <0.06 <0.12 0.015 0.008 <0.008 12.2 50 30 Q181206360 C. albicans Urine 0.5 0.015 0.015 0.008 0.25 0.12 0.015 0.008 <0.008 12.2 50 31 Q181208606 C. albicans Urine 1 0.015 0.03 0.015 0.5 0.25 0.06 0.03 <0.008 3.05 12.5 32 Q181208813 C. albicans Urine 1 0.06 0.06 <0.008 <0.06 0.5 0.06 0.03 <0.008 3.05 12.5 33 Q181208729 C. albicans Vaginal swab 0.5 0.015 0.03 0.25 0.06 0.015 0.008 0.015 0.008 6.1 25 34 Q181213218 C. albicans Urine 0.25 0.015 0.015 0.008 <0.06 0.12 0.015 0.008 <0.008 1.52 6.25 35 Q181202764 C. albicans Prosthetic materiel 1 0.12 0.06 <0.008 0.12 <0.12 0.06 0.03 <0.008 1.52 6.25 36 Q181198103 C. albicans Lung biopsy 0.5 0.03 0.12 0.015 <0.06 4 0.25 1 0.12 3.05 12.5 37 Q181203206 C. albicans Biopsy 0.5 0.06 0.06 0.015 <0.06 0.25 0.015 0.015 <0.008 0.76 3.12 38 Q181201380 C. albicans Sputum 1 <0.015 0.015 0.015 >64 0.5 0.06 0.03 <0.008 1.52 6.25 39 Q181199096 C. albicans Gastric biopsy 1 0,03 0.06 0.03 <0.06 0.5 0.06 0.03 0.008 3.05 12.5 40 Q181201157 C. albicans Vaginal swab <0.015 0.06 1 0.03 0.5 0.06 0.015 <0.008 0.008 3.05 12.5 41 Q181202348 C. albicans Vaginal swab <0.015 0.03 <0.06 0.015 0.25 0.015 <0.008 <0.008 0.008 1.52 6.25 42 Q181202493 C. albicans Sputum 0.25 0.03 0.03 0.008 0.06 0.12 0.03 0.008 <0.008 3.05 12.5 43 Q181202887 C. albicans Urine 0.5 0.03 0.03 0.008 <0.06 0.12 0.03 0.015 <0.008 3.05 12.5 44 Q181204443 C. albicans Kehr's drain 0.5 0.03 0.06 0.015 <0.06 0.5 0.06 0.03 <0.008 3.05 12.5 45 Q181205297 C. albicans Peritoneal fluid 1 0.06 0.06 0.015 0.5 0.25 0.03 0.015 <0.008 1.52 6.25 46 Q181204777 C. albicans Bile fluid 1 0.03 0.015 0.015 <0.06 0.25 0.03 0.015 <0.008 6.1 25 47 Q181203339 C. albicans Urine 1 0.06 0.06 0.008 <0.06 1 0.06 0.12 0.015 >24.4 >100 48 Q181207523 C. albicans Vaginal swab 0.5 0.015 0.03 0.008 <0.06 <0.12 0.015 0.008 <0.008 6.1 25 49 Q181207853 C. albicans Drain fluid 1 0.06 0.06 0.015 0.12 2 0.12 0.06 0.015 6.1 25 50 Q181208657 C. albicans Vaginal swab 1 0.06 0.06 0.015 1 0.25 0.03 0.015 <0.008 12.2 50 51 Q181207853 C. albicans Drain fluid 1 0.06 0.06 0.015 0.12 2 0.12 0.06 0.015 >24.4 >100 52 Q181208026 C. albicans Peritoneal fluid 0.5 0.03 0.03 0.015 <0.06 0.5 0.06 0.03 0.015 1.52 6.25 53 Q181205788 C. albicans Puncture fluid 0.5 <0.015 0.015 0.015 0.06 0.25 0.03 0.015 0.015 3.05 12.5 54 490890568 C. albicans Blood culture 1 0.5 0.12 0.008 0.5 1 0.5 0.25 0.03 3.05 12.5 55 Q181253001 C. albicans Blood culture 0.5 1 1 1 0.5 0.25 0.03 0.015 <0.008 1.52 6.25 56 Q181258251 C. albicans Blood culture 0.5 0.03 0.15 <0.008 0.25 0.5 0.12 0.03 0.015 3.05 12.5 57 Q181270397 C. albicans Vaginal swab 0.5 0.015 0.015 <0.015 0.12 0.25 0.03 0.015 <0.008 12.2 50 58 Q181266055 C. albicans Vaginal swab 0.5 0.008 0.008 0.008 0.06 0.12 0.06 0.03 0.008 1.52 6.25 59 Q181278775 C. albicans Puncture fluid 1 0.003 0.006 <0.008 0.006 0.5 0.12 0.06 0.015 3.05 12.5 60 Q181281649 C. albicans Intrauterine device 0.5 0.03 0.03 0.015 0.006 1 0.12 0.06 0.03 3.05 12.5 61 Q181267399 C. albicans Vaginal swab 0.5 0.015 0.015 0.008 0.06 0.12 0.06 0.03 0.008 1.52 6.25 62 Q181281412 C. albicans Peritoneal fluid 1 0.015 0.03 <0.008 0.06 1 0.12 0.06 <0.008 3.05 12.5 63 Q181281621 C. albicans Puncture fluid 0.5 0.06 0.06 <0.008 <0.006 0.5 0.06 0.03 <0.008 0.76 3.12 64 Q181263869 C. albicans Urine 0.5 <0.015 0.015 0.015 <0.06 0.12 0.015 0.015 <0.008 3.05 12.5 65 Q181260431 C. albicans Abscess 0.25 <0.015 0.008 0.008 0.06 0.12 0.06 0.03 <0.008 1.52 6.25 66 Q181261387 C. albicans Urine 1 0.03 0.015 <0.008 <0.06 0.5 0.06 0.03 0.015 1.52 6.25 67 Q181256109 C. albicans Prosthetic materiel 0.5 0.03 0.03 <0.008 <0.06 0.12 0.015 0.015 <0.008 12.2 50 68 Q181256542 C. albicans liver Biopsy 0.5 0.015 0.015 <0.008 <0.06 <0.12 0.015 0.008 <0.008 3.05 12.5 69 Q181264539 C. albicans Blood culture 0,5 <0.015 0.015 <0.008 >64 0.25 0.12 0.06 0.015 0.76 3.12 70 Q181252599 C. albicans Urine 1 0.25 0.06 0.03 0.12 1 0.12 0.12 0.06 6.1 25 71 Q181201405 C. albicans Lung biopsy 1 0.03 0.06 0.015 0.5 0.5 0.06 0.03 0.015 1.52 6.25 72 Q181235301 C. albicans Vaginal swab 1 <0.015 0.03 0.015 1 0.5 0.06 0.03 0.015 3.05 12.5 73 Q181218400 C. albicans Vaginal swab 1 0.015 0.03 <0.008 0.06 1 0.12 0.06 <0.008 12.5 50 74 Q181231655 C. albicans Urine 0.5 0.3 0.03 <0.008 0.25 0.5 0.06 0.03 <0.008 >24.4 >100 75 Q181201405 C. albicans Lung biopsy 0.5 0.015 0.015 <0.008 <0.06 <0.12 0.015 0.008 <0.008 >24.4 >100 76 Q181253570 C. albicans Urine 0.5 <0.015 0.15 <0.008 0.12 <0.012 0.03 0.15 <0.008 >24.4 >100 77 Q181212571 C. albicans Vaginal swab 0.25 0.03 0.03 0.008 <0.06 <0.12 0.015 0.008 <0.008 3.05 12.5 78 Q181211559 C. albicans Urine 0.5 0.12 0.015 0.12 1 2 1 0.5 0.06 12.2 50 79 Q181226987 C. albicans Urine 1 0.06 0.06 <0.008 64 0.25 0.06 0.03 0.015 6.1 25 80 Q181231263 C. albicans Urine 0.5 0.015 0.06 <0.008 <0.06 0.25 0.06 0.03 <0.008 3.05 12.5 81 Q181231467 C. albicans Urine 0.5 0.015 3 0.008 <0.06 <0.012 0.03 0.015 <0.008 3.05 12.5 82 Q181253408 C. albicans Urine 0.5 <0.015 <0.008 <0.008 0.25 0.25 0.06 0.015 <0.008 12.2 50 83 Q181260978 C. albicans Puncture fluid 0.25 0.06 0.12 0.12 0.5 2 1 0.5 0.25 12.2 50 84 Q181108884 C. albicans Cerebro-spinal fluid 0.25 0.015 0.05 0.03 1 16 0.5 0.5 0.25 1.52 6.25 85 Q181207854 C. albicans Drain fluid 1 0.06 0.06 0.015 0.12 2 0.12 0.06 0.015 1.52 6.25 86 Q181201405 C. albicans Lung biopsy 1 0.03 0.06 0.015 0.5 0.5 0.06 0.03 0.06 1.52 6.25 87 Q181207853 C. albicans Drain fluid 1 0.06 0.06 0.015 0.12 2 0.12 0.06 0.015 3.05 12.5 88 Q181208657 C. albicans Urine 1 0.06 0.06 0.015 0.25 0.03 <0.008 0.06 0.015 3.05 12.5 89 ATCC 6258 C. krusei / 1 0.06 0.25 0.12 8 32 0.12 0.25 0.25 >24.4 >100 90 Q181262141 C. krusei Prosthetic materiel 0.5 0.03 0.12 0.06 4 32 0.12 0.12 0.25 24.4 100 91 Q181263662 C. krusei Urine 1 0.03 0.25 0.12 2 16 0.12 0.12 0.06 12.2 50 92 Q181208438 C. krusei Bronchial aspirate 1 0.12 0.5 0.12 8 32 0.25 0.25 0.25 24.4 100 93 Q181263124 C. lusitaniae Blood culture 0.25 0.12 0.12 0.06 <0.06 0.12 0.03 0.015 <0.008 >24.4 >100 94 Q181266535 C. lusitaniae Vaginal swab 0.25 0.12 0.12 0.06 0.06 1 0.12 0.03 0.008 >24.4 >100 95 Q181206420 C. lusitaniae Urine 0.5 0.25 0.25 0.25 32 0.5 0.12 0.03 <0.008 >24.4 >100 96 Q181250041 C. tropicalis Puncture fluid 0.5 0.12 0.12 0.015 64 1 0.12 0.06 0.06 1.52 6.25 97 Q181264363 C. tropicalis Urine 1 0.03 0.03 0.03 <0.06 0.12 0.12 0.03 0.25 0.75 3.12 98 Q181257439 C. tropicalis Bronchial aspirate 0.5 0.03 0.06 0.03 64 1 0.12 0.12 0.06 0.75 3.12 99 8070845333 C. tropicalis Bronchial aspirate 1 0.12 0.12 0.03 >64 2 0.12 0.12 0.12 1.52 6.25 100 Q181203338 C. tropicalis Urine 0.5 0.06 0.12 0.015 64 1 0.12 0.06 0.06 0.75 3.12

Partie IV :

Identification du mécanisme d’action de la ribavirine sur

Candida albicans, et élaboration de ses éventuelles cibles

spécifiques

110

Avant-propos

La caractérisation du mode d'action des molécules antifongiques repositionnées fait partie intégrante du processus de découverte d’alternatives médicamenteuses nouvelles.

Depuis longtemps, des méthodes classiques ont été utilisées pour explorer le mode d'action des antifongiques, s'appuyant spécifiquement sur la microscopie classique et les approches biochimiques de base. Avec l'ère de la génomique fonctionnelle et des technologies d'expérimentation à haut débit, la mise en œuvre de stratégies de découverte des cibles médicamenteuses a radicalement été remodelée.

Dans cette partie du travail, après avoir caractérisé l’action antifongique de la ribavirine, sur les Candida albicans, on s’est intéressé à la compréhension de son mécanisme d’action ainsi que les cibles potentiellement spécifiques qu’elle pourrait avoir. Il existe quatre voies principales d’action de la ribavirine sur le virus de l’hépatite C notamment ; (i) inhibition de l’action de l’ARN polymérase virale, (ii) inhibition de l’inosine monophosphate déshydrogénase (IMPDH), (iii) induction de mutations au niveau de l’ARN et enfin (iv) modulation de la réponse immune. Notre travail s’est porté sur deux principales voies d’action de la ribavirine, à savoir l’inhibition de l’ARN polymérase et de l’IMPDH, qui constituent deux cibles existantes chez les Candida. Chez les Candida, il existe trois types d’ARN polymérase, et c’est l’ARN polymérase II qui est responsable de la synthèse d’ARN à partir d’une matrice d’ADN. Elle est constituée de plusieurs sous-unités dont la plus importante est la Rpb1 formée de 1728 acides aminés et codée par le gène RPO21. Quant à l’IMPDH, c’est une enzyme indispensable pour la conversion de l’inosine-5-monophosphate en xanthine-

5monophosphate, étape limitante de la synthèse de la guanosine qui rentre dans la structure du

111

matériel génétique. Cette enzyme est formée de 521 acides aminés et codée par le gène IMH3 situé sur le chromosome 2 du génome du Candida albicans.

De ce fait, on a désigné plusieurs jeux d’amorces spécifiques à ces deux gènes. On a réussi à amplifier les gènes en question, sur plusieurs souches de Candida albicans, sensibles et résistantes à la ribavirine. Par la suite, le séquençage des gènes nous a permis d’obtenir la structure nucléotidique des gènes afin d’effectuer une analyse comparative entre les séquences des souches sensibles à la ribavirine (CMI basses) et celles qui sont résistantes (CMI > 24

µg/ml). Ce travail est en cours et ses résultats seront communiqués sous forme d’article prochainement.

112

Partie V :

Annexes

113

Avant-propos

Une expansion importante de bactéries Gram-négatif résistantes au traitement par les carbapénèmes a été signalée par plusieurs rapports épidémiologiques. Cette situation a incité la réintroduction de la colistine, comme antibiotique de dernier recours, dans la prise en charge des infections induites par des bactéries multi-résistantes. En effet, la colistine a été longtemps abandonnée en raison de sa neurotoxicité et sa néphrotoxicité chez l’homme [19]. La colistine est un antibiotique polypeptidique et bactéricide qui a pour cible, le lipopolysaccharide (LPS) membranaire chargé négativement [19]. Jusqu’en 2015, la résistance à la colistine était principalement médiée par des mutations chromosomiques qui activent le système à deux composantes PhoP/PhoQ et PmrA/PmrB. Plus tard, Liu et al ont rapporté la première détection de la résistance plasmidique à la colistine médiée par le gène mcr-1 [20].

Ainsi, un screening des 1280 molécules de la chimiothèque, sur une souche d’Escherichia coli colistine-résistante a permis la première identification de l’activité antibactérienne de la zidovudine, un antirétroviral utilisé dans le traitement du Sida, sur une souche bactérienne résistante à la colistine avec un mécanisme de résistance plasmidique

(mcr1). Ensuite, l’activité de la zidovudine a été élucidée sur une collection de bactéries résistantes à la colistine et aux carbapénèmes incluant d’autres souches d’E. coli et des

Klebsiella pneumoniae. L’ensemble des souches testées étaient sensible à la zidovudine avec des CMI comprises entre 0.2 et 6.25 µM. Les CMI de la zidovudine obtenues dans cette étude pourraient être facilement atteintes en concentrations plasmatiques chez l’homme, ce qui faciliterait son utilisation en pratique clinique (Article 5).

114

Nous nous sommes également intéressés à l’étude épidémiologique des supports moléculaires de la résistance à la colistine, à travers le monde. C’est pourquoi, on a étudié trois isolats de K. pneumoniae résistants à la colistine, prélevés à l'hôpital universitaire d'Annaba, en

Algérie, chez trois patients différents qui ont en commun un antécédent de chirurgie urologique.

Cette étude a rapporté la première description de la résistance à la colistine associée à la présence d’une carbapénèmase OXA-48 chez des Klebsiella pneumoniae ST101, en Algérie,

(Article 6). Ceci m’avait permis d’apprendre toutes les techniques de biologie moléculaire à savoir la PCR, le séquençage et l’analyse MLST.

Une partie de ce travail a consisté en la description complète d’une nouvelle espèce, isolée dans le cadre du projet « culturomique » de l’IHU Méditerranée Infection. Il s’agit de la description de la souche bactérienne Olsenella timonensis, une bactérie Gram-négatif, anaérobique stricte provenant de selles d’un sujet âgé de 75 ans présentant un syndrome inflammatoire, en France. Elle possède un génome de 217, 673, 37 paires de bases avec un

GC% de 68, 08%. O. timonensis est sensible au céfoxitine, vancomycine, téicoplanine, linézolide, ciprofloxacine, sulfaméthoxazole / triméthoprime, doxycycline, érythromycine, clindamycine et rifampicine (Article 7).

La description de la nouvelle espèce m’avait permis d’apprendre les différentes techniques de base de la bactériologie.

115

Article 5:

Zidovudine: a salvage therapy for mcr-1 plasmid-mediated

colistin-resistant bacterial infections?

Lucie Peyclit, Sophie Baron, Hanane Yousfi, Jean-Marc Rolain

Published in « International Journal of Antimicrobial Resistance »

116

Accepted Manuscript

Zidovudine: a salvage therapy for mcr-1 plasmid-mediated colistin-resistant bacterial infections?

Lucie Peyclit , Sophie Baron , Hanane Yousﬁ , Jean-Marc Rolain

PII: S0924-8579(18)30089-X DOI: 10.1016/j.ijantimicag.2018.03.012 Reference: ANTAGE 5403

To appear in: International Journal of Antimicrobial Agents

Received date: 2 March 2018 Revised date: 12 March 2018 Accepted date: 15 March 2018

Please cite this article as: Lucie Peyclit , Sophie Baron , Hanane Yousﬁ , Jean-Marc Rolain , Zidovu- dine: a salvage therapy for mcr-1 plasmid-mediated colistin-resistant bacterial infections?, International Journal of Antimicrobial Agents (2018), doi: 10.1016/j.ijantimicag.2018.03.012

This is a PDF ﬁle of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its ﬁnal form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. ACCEPTED MANUSCRIPT

Zidovudine: a salvage therapy for mcr-1 plasmid-mediated colistin-resistant bacterial infections?

Lucie Peyclit1#, Sophie Baron1#, Hanane Yousfi1 and Jean-Marc Rolain*1.

# contributed equally to the work

Affiliations: 1 Aix Marseille Univ, IRD, APHM, MEPHI, IHU Méditerranée Infection,

Faculté de Médecine et de Pharmacie, 19-21 boulevard Jean Moulin, 13385 Marseille

CEDEX 05, France.

* Corresponding author: Jean-Marc Rolain, Aix Marseille Univ, IRD, APHM, MEPHI, IHU

Méditerranée Infection, Faculté de Médecine et de Pharmacie, 19-21 boulevard Jean Moulin,

13385 Marseille CEDEX 05, France. Phone: (33) 4 91 32 43 75. Email: jean- [email protected]

Keys words: multi-drug resistant Gram negative bacteria; mcr-1; zidovudine; last line antibiotic; azidothymidine

Over the last decade, antibiotic resistance has become a major new threat to human health and a worldwide problem prompting the WHO to prioritize the fight against multidrug-resistant bacteria. The WHO positioned last year Enterobacteriaceae as an antibiotic-resistant priority pathogen [1]. Among all threats, the newly described mcr-1 plasmid-mediated colistin- resistant mechanism [2], which has been described all over the world in Gram-negative bacteria [3], became the most important. We have recently reported that this threat is probably overestimatedACCEPTED since many old antibiotics are MANUSCRIPTstill effective and could be reintroduced into the pharmaceutical market to treat such infections [4–6]. The main problem is that those old but useful antibiotics are no longer available in many countries [7]. What could solve both problems of antibiotic-resistance and no-longer available drugs in the market is "drug repurposing" i.e the use of a FDA-approved drug as new therapeutic agents. Using this ACCEPTED MANUSCRIPT approach by screening a total of 1,163 FDA-approved drugs, Ng et al have recently found that zidovudine, an antiviral drug active against HIV, was effective in vitro against 13 carbapenem-resistant and colistin-susceptible Enterobacteriaceae (Escherichia coli and

Klebsiella pneumoniae) with MICs compatible with human use [8]. Zidovudine, a nucleoside reverse transcriptase inhibitor, was the first antiretroviral commercialized for the treatment of

HIV infection in 1987 [9]. This nucleotide and synthetic analog was initially used as an anti- cancer drug [9], but, from 1986, its antibacterial effect was revealed in bacteria [10,11] that possess a thymidine kinase homolog able to activate the drug [12]. Using a similar approach with a different FDA-approved library of 1,280 drugs (Prestwick, Illkirch graffenstaden,

France), we also identified zidovudine as an effective drug against Enterobacteriaceae. To our knowledge, no study already reported zidovudine efficacy on colistin-resistant Gram- negative bacteria. In our laboratory, we collected a high number of multi-drug resistant

Enterobacteriaceae including carbapenem and colistin-resistant strains isolated from different geographical areas worldwide for which molecular mechanisms of resistance have been previously characterized. We report here the efficacy of this compound against a serie of 40 strains including 16 Escherichia coli (12 colistin-resistant isolates, 3 carbapenem-resistant strains and 1 susceptible isolate) and 22 Klebsiella pneumoniae (11 colistin-resistant strains, 9 carbapenem-resistant isolates, one colistin- and carbapenem-resistant strain and one susceptible isolate). Molecular mechanisms of resistance to colistin include mcr-1, mgrB inactivation and PmrB mutation whereas resistance to carbapenems include blaNDM, blaKPC and blaACCEPTEDOXA-48 carbapenemase; details are given MANUSCRIPT in Table 1. Two strains of Staphylococcus aureus which are intrinsically resistant to zidovudine were also used as controls [13] (Table

1).

Zidovudine MICs were determined using broth microdilution dilution method in Mueller

Hinton II broth with a zidovudine (Sigma-Aldrich, Missouri, United States) concentration ACCEPTED MANUSCRIPT ranging from 0.2 to 100 µM (i.e. 0.053 µg/mL to 26.7 µg/mL). All strains were susceptible to zidovudine with MICs ranging from 0.2 to 6.25 µM i.e from 0.05 µg/mL to 1.67 µg/mL

(Table 1). Interestingly, zidovudine was also highly effective against a clinical isolate that was both carbapenem and colistin-resistant (K. pneumoniae strain 853; Table 1). The two S. aureus strains were resistant with MICs >100 µM (>26.7 µg/mL).

From a clinical point of view these findings are extremely important for the medical community because of the pharmacokinetic properties of zidovudine. In a study with healthy volunteers, a Cmax of 4 µM (1.07µg/ml) of zidovudine was obtained 1h after the administration of 200 mg of zidovudine in a single oral administration [14]. Similarly,

Wattanagoon et al have reported in healthy volunteers in Thailand a Cmax of 17.98 µM (4.8

µg/ml) after a single dose of 300 mg of zidovudine [15]. Zidovudine is used at the dose of 600 mg/day (300 mg twice a day) in HIV treatment suggesting that a plasmatic concentration above the MIC could be easily achieved. This dosage could be increased for example to 2400 mg/day (600 mg four times/day) since over dosage of 20g has been reported to be free of side effects in the past in humans [16]. Moreover, zidovudine is currently available either in tablet form or in intravenous form that could be used as a slow intravenous infusion (half-life of zidovudine is 1.1 hour) in case of severe sepsis or meningitis since zidovudine can achieve therapeutic concentrations in the cerebrospinal fluid [17]. Zidovudine could also be used during pregnancy and in children [18]. Finally, side effects, as hematotoxicity, are well known and are associated with long-term administration of zidovudine so they could easily be managedACCEPTED as salvage therapy if used for a relatively MANUSCRIPT short period of time. One question with the use of zidovudine as salvage therapy for multidrug-resistant bacterial infections is the possibility of resistance development. This has been already reported in E. coli in HIV patients treated with zidovudine [19] due to the loss of thymidine kinase activity

[13,20] preventing zidovudine phosphorylation and activation. Thus it seems reasonable to ACCEPTED MANUSCRIPT combine zidovudine with other drugs to circumvent the selection of zidovudine-resistant strains, for example fosfomycin or minocycline that are known to be effective against such strains [5,6] or tigecycline as recently reported by Ng et al. [8].

To conclude, our in vitro study demonstrates that drug repurposing is an effective way to

(re)discover existing drugs that may be able to solve some of the current problems of worldwide antibiotic-resistance. Indeed, the antibacterial activity of zidovudine was previously described in mice [11] and suspected in vivo in humans [21]. The revival of old antibiotics can be useful to fight bacteria resistant to multiple current antibacterial drug, both screening of large libraries of approved drug as well as of the literature for old data on antibacterial activities. Here zidovudine is identified as a leader in this context of worldwide emergence of colistin resistance. Case reports and clinical trials with zidovudine in combination with other drugs to treat patients infected with carbapenem-resistant and/or colistin-resistant bacteria definitely prompt us to add this drug to our therapeutic weapons against multi-drug resistant bacteria. As patents on zidovudine expired in 2005, several generics exist and this drug is less used as an antiretroviral therapeutic, its use as an antibacterial drug could be an affordable alternative.

Declarations

Funding

This work was supported by the French Government under the « Investissements d’avenir »

(Investments for the Future) program managed by the Agence Nationale de la Recherche (ANR,ACCEPTED fr: National Agency for Research), (reference: MANUSCRIPT Méditerranée Infection 10-IAHU-03) Competing Interests

None.

Ethical Approval

Not required ACCEPTED MANUSCRIPT

Acknowledgment

We want to thank Miss Emilie Lambourg for English correction.

References

[1] Tacconelli E, Carrara E, Savoldi A, Harbarth S, Mendelson M, Monnet DL, et al. Discovery, research, and development of new antibiotics: the WHO priority list of antibiotic- resistant bacteria and tuberculosis. Lancet Infect Dis 2017. doi:10.1016/S1473- 3099(17)30753-3. [2] Liu Y-Y, Wang Y, Walsh TR, Yi L-X, Zhang R, Spencer J, et al. Emergence of plasmid-mediated colistin resistance mechanism MCR-1 in animals and human beings in China: a microbiological and molecular biological study. Lancet Infect Dis 2016;16:161–8. doi:10.1016/S1473-3099(15)00424-7. [3] Baron S, Hadjadj L, Rolain J-M, Olaitan AO. Molecular mechanisms of polymyxin resistance: knowns and unknowns. Int J Antimicrob Agents 2016;48:583–91. doi:10.1016/j.ijantimicag.2016.06.023. [4] Rolain J-M, Abat C, Jimeno M-T, Fournier P-E, Raoult D. Do we need new antibiotics? Clin Microbiol Infect Off Publ Eur Soc Clin Microbiol Infect Dis 2016;22:408– 15. doi:10.1016/j.cmi.2016.03.012. [5] Dubourg G, Okdah L, Le Page S, Rolain J-M, Raoult D. In vitro activity of “old antibiotics” against highly resistant Gram-negative bacteria. Int J Antimicrob Agents 2015;46:718–20. doi:10.1016/j.ijantimicag.2015.09.008. [6] Cassir N, Rolain J-M, Brouqui P. A new strategy to fight antimicrobial resistance: the revival of old antibiotics. Front Microbiol 2014;5:551. doi:10.3389/fmicb.2014.00551. [7] Pulcini C, Mohrs S, Beovic B, Gyssens I, Theuretzbacher U, Cars O, et al. Forgotten antibiotics: a follow-up inventory study in Europe, the USA, Canada and Australia. Int J Antimicrob Agents 2017;49:98–101. doi:10.1016/j.ijantimicag.2016.09.029. [8] Ng SMS, Sioson JSP, Yap JM, Ng FM, Ching HSV, Teo JWP, et al. Repurposing Zidovudine in combination with Tigecycline for treating carbapenem-resistant Enterobacteriaceae infections. Eur J Clin Microbiol Infect Dis Off Publ Eur Soc Clin Microbiol 2018;37:141–8. doi:10.1007/s10096-017-3114-5. [9] Chow WA, Jiang C, Guan M. Anti-HIV drugs for cancer therapeutics: back to the future? Lancet Oncol 2009;10:61–71. doi:10.1016/S1470-2045(08)70334-6. [10] Elwell LP, Ferone R, Freeman GA, Fyfe JA, Hill JA, Ray PH, et al. Antibacterial activity and mechanism of action of 3’-azido-3’-deoxythymidine (BW A509U). Antimicrob Agents Chemother 1987;31:274–80. [11] Keith BR, White G, Wilson HR. In vivo efficacy of zidovudine (3’-azido-3’- deoxythymidine) in experimental gram-negative-bacterial infections. Antimicrob Agents Chemother 1989;33:479–83. [12] ACCEPTED Doléans-Jordheim A, Bergeron E, Bereyziat MANUSCRIPT F, Ben-Larbi S, Dumitrescu O, Mazoyer M-A, et al. Zidovudine (AZT) has a bactericidal effect on enterobacteria and induces genetic modifications in resistant strains. Eur J Clin Microbiol Infect Dis Off Publ Eur Soc Clin Microbiol 2011;30:1249–56. doi:10.1007/s10096-011-1220-3. [13] Lewin CS, Allen RA, Amyes SG. Mechanisms of zidovudine resistance in bacteria. J Med Microbiol 1990;33:235–8. doi:10.1099/00222615-33-4-235. [14] Taburet A-M, Naveau S, Zorza G, Colin J-N, Delfraissy J-F, Chaput J-C, et al. Pharmacokinetics of zidovudine in patients with liver cirrhosis. Clin Pharmacol Ther 1990;47:731–9. doi:10.1038/clpt.1990.101. ACCEPTED MANUSCRIPT

[15] Wattanagoon Y, Na Bangchang K, Hoggard PG, Khoo SH, Gibbons SE, Phiboonbhanakit D, et al. Pharmacokinetics of zidovudine phosphorylation in human immunodeficiency virus-positive thai patients and healthy volunteers. Antimicrob Agents Chemother 2000;44:1986–9. [16] Pickus OB. Overdose of zidovudine. N Engl J Med 1988;318:1206. doi:10.1056/NEJM198805053181817. [17] Balis FM, Pizzo PA, Murphy RF, Eddy J, Jarosinski PF, Falloon J, et al. The pharmacokinetics of zidovudine administered by continuous infusion in children. Ann Intern Med 1989;110:279–85. [18] Sperling R. Zidovudine. Infect Dis Obstet Gynecol 1998;6:197–203. doi:10.1155/S1064744998000404. [19] Lewin CS, Watt B, Paton R, Amyes SG. Isolation of zidovudine resistant Escherichia coli from AIDS patients. FEMS Microbiol Lett 1990;58:141–3. [20] Lewin CS, Allen R, Amyes SG. Zidovudine-resistance in Salmonella typhimurium and Escherichia coli. J Antimicrob Chemother 1990;25:706–8. [21] Casado JL, Valdezate S, Calderon C, Navas E, Frutos B, Guerrero A, et al. Zidovudine therapy protects against Salmonella bacteremia recurrence in human immunodeficiency virus- infected patients. J Infect Dis 1999;179:1553–6. doi:10.1086/314749.

ACCEPTED MANUSCRIPT ACCEPTED MANUSCRIPT

Table 1. MIC of colistin, imipenem and zidovudine against various colistin or carbapenemase-resistant strains.

MIC MIC MIC MIC Resistance Resistance Species Strain Origin Country colistin imipenem AZT AZT References phenotype mechanism (µg/mL) (µg/mL) (µg/mL) (µmol/L) Escherichia coli ATCC 25922 human United States none susceptible 1 0.19 0.835 3.125 ATCC

E. coli LH1 human Laos colistin mcr-1 6 0.125 0.104 0.4 [20]

E. coli LH30 human Laos colistin mcr-1 6 0.19 0.208 0.8 [20]

E. coli LH57 human Laos colistin mcr-1 6 0.125 0.417 1.56 [20]

E. coli 1R4 human Saudi Arabia colistin mcr-1 4 0.19 0.208 0.8 [21]

E. coli NCTC 13846 human England colistin mcr-1 4 0.19 0.208 0.8 NCTC

E. coli 6R human Saudi Arabia colistin mcr-1 4 0.125 0.417 1.56 [21]

E. coli SE65 human Algeria colistin mcr-1 8 0.19 1.67 6.25 [22]

E. coli LH257 human Laos colistin mcr-1 12 0.19 0.104 0.4 [20]

E. coli 134R human Saudi Arabia colistin mcr-1 3 0.19 0.208 0.8 [21]

E. coli 44A human Saudi Arabia colistin mcr-1 4 0.19 0.104 0.4 [21]

E. coli TH99 human Thailand colistin mcr-1 4 0.125 0.104 0.4 [20]

E. coli 235 chicken Algeria colistin mcr-1 4 0.125 0.208 0.8 [20]

E. coli CMUL64 human Lebanon carbapenem blaoxa-48 0.38 0.023 0.052 0.2 [23] Unpublished E. coli CSURP5142 human Marseille carbapenem bla 1 32 0.208 0.8 oxa-48 data

E. coli CSURP1954 human Algeria carbapenem blaNDM-5 0.25 > 32 0.052 0.2 [24] Klebsiella pneumoniae ATCC 13883 human unknown none susceptible 1 0.38 0.052 0.2 ATCC

K. pneumoniae LH70 human Laos colistin unknown 12 0.19 0.052 0.2 [25]

ACCEPTED MANUSCRIPT ACCEPTED MANUSCRIPT

K. pneumoniae LH12 human Laos colistin mgrB 32 0.19 0.104 0.4 [25]

K. pneumoniae FHM169 human France colistin mgrB 8 0.19 0.208 0.8 [25]

K. pneumoniae TH20 human Thailand colistin mgrB 32 0.19 0.417 1.56 [25]

K. pneumoniae TH28 human Thailand colistin mgrB 8 0.19 0.208 0.8 [25]

K. pneumoniae TH176 human Thailand colistin mgrB 12 0.19 < 0.052 < 0.2 [25]

K. pneumoniae LH17 human Laos colistin mcr-1 + PmrB T157P 12 0.125 0.835 3.125 [25]

K. pneumoniae LH92 human Laos colistin mcr-1 12 0.19 0.104 0.4 [25]

K. pneumoniae FHA60 human France colistin mcr-1 8 0.19 0.104 0.4 [25]

K. pneumoniae FHM128 human France colistin mcr-1 4 0.19 0.417 1.56 [25]

K. pneumoniae TH68 human Thailand colistin mcr-1 8 0.19 0.104 0.4 [25] Unpublished K. pneumoniae CSURP5123 human France carbapenem bla 1 6 0.104 0.4 oxa-48 data Unpublished K. pneumoniae CSURP5233 human France carbapenem bla 0.25 2 0.835 3.125 oxa-48 data Unpublished K. pneumoniae NCTC 13443 human unknown carbapenem bla 1 > 32 0.208 0.8 NDM-1 data Unpublished K. pneumoniae CSURP5141 human France carbapenem bla 0.25 4 0.835 3.125 NDM-1 data Unpublished K. pneumoniae CSUR5135 human France carbapenem bla 1 0.38 0.417 1.56 NDM-1 data

K. pneumoniae CSURP1572 human Algeria carbapenem blaKPC-3 0.125 8 0.104 0.4 [26] carbapenem bla Unpublished K. pneumoniae 853 human Israel KPC 48 > 32 0.104 0.4 colistin mgrB data Unpublished K. pneumoniae 1348 human Israel carbapenem bla < 2 8 0.835 3.125 KPC data Unpublished K. pneumoniae 695 human Israel carbapenem bla < 2 8 0.417 1.56 KPC data Unpublished K. pneumoniae 473 human Israel carbapenem bla < 2 6 0.417 1.56 KPC data Staphylococcus aureus CSURP1943 human France MRSA - - 26.27 < 100 [27]

ACCEPTED MANUSCRIPT ACCEPTED MANUSCRIPT

S. aureus ATCC 25923 unknown unknown none susceptible - - 26.27 < 100 ATCC MRSA: Methicillin-resistant Staphylococcus aureus.

ACCEPTED MANUSCRIPT

Article 6:

Colistin- and Carbapenem-Resistant Klebsiella pneumoniae Clinical_Isolates : Algeria

Hanane Yousfi, Linda Hadjadj, Iman Dandachi, Rym Lalaoui, Adil Merah, Kamel Amoura, Ahlem Dahi, Mazouz Dekhil, Naima Messalhi, Seydina M. Diene, Sophie Baron and Jean-Marc Rolain

Published in « Microbial Drug Resistance »

127

MICROBIAL DRUG RESISTANCE Volume 00, Number 00, 2018 ª Mary Ann Liebert, Inc. DOI: 10.1089/mdr.2018.0147

Colistin- and Carbapenem-Resistant Klebsiella pneumoniae Clinical_Isolates: Algeria

Hanane Yousﬁ,1 Linda Hadjadj,1 Iman Dandachi,1 Rym Lalaoui,1 Adil Merah,2 Kamel Amoura,2,3 Ahlem Dahi,2,3 Mazouz Dekhil,2,3 Naima Messalhi,2,3 Seydina M. Diene,1 Sophie Baron,1 and Jean-Marc Rolain1

This study investigates the molecular mechanisms of colistin and carbapenem resistance in Klebsiella pneumoniae ST101 strains. The three K. pneumoniae carried blaCTX-M-15, blaTEM-183, and blaSHV-106 genes and two coharbored blaOXA-48. As for colistin resistance, the isolates had amino acid substitutions in PmrA/B and a truncated mgrB gene in one isolate.

Keywords: Klebsiella pneumoniae, colistin resistance, carbapenem resistance, mgrB insertion

he prevalence of extended spectrum b-lactamase Identification of the isolates was done using matrix- T(ESBL) and carbapenemase-producing Klebsiella pne- assisted laser desorption and ionization time-of-flight mass umoniae isolates is constantly rising in clinical settings.1 spectrometry (Microflex; Bruker Daltonics).16 Antibiotic Consequently, colistin, a previously abandoned antimicro- susceptibility testing was performed by disk diffusion meth- bial agent due to its nephrotoxicity and neurotoxicity in od. Interpretation of results was done according to the Euro- humans, was reintroduced in clinical settings for the treat- pean Committee following the antimicrobial susceptibility ment of infections caused by multidrug-resistant (MDR) testing guidelines. The three isolates were resistant to colistin, organisms.2 Recently, there is an increasing number of re- aztreonam, ceftazidime, cefotaxime, ceftriaxone, cefoxitin, ported cases of combined resistance to antibiotics of the last gentamicin, ciprofloxacin, nalidixic acid, nitrofurantoin, and resort, carbapenems and colistin, in clinical settings.3,4 fosfomycin; however, they remained sensitive to amikacin, In Algeria, the first report of a colistin-resistant isolate was trimethoprim/sulfamethoxazole, tetracycline, and imipenem published in 2015 and described as an Acinetobacter bau- (Table 1). In addition, two of three K. pneumoniae strains mannii ST2 isolated from patients in the university hospital were resistant to chloramphenicol and ertapenem (Table 1). center of Beńi-Messous in Algiers. This isolate presented The minimum inhibitory concentration (MIC) of colistin, with a deleterious insertion of an amino acid named ‘‘ala- imipenem, and ertapenem for isolates was determined by nine’’ in the pmrB gene at position 163.5 Thereafter, the mcr-1 broth microdilution, which revealed that all isolates were plasmid-mediated colistin resistance gene was described in resistant to colistin (MIC ‡16 mg/mL) with only two of them Escherichia coli after its isolation from animals as well as being also resistant to ertapenem (MIC = 8 mg/mL); all iso- 6,7 Downloaded by Gothenburg University Library from www.liebertpub.com at 09/26/18. For personal use only. in clinical settings. In this study we report the first de- lates were imipenem susceptible (Table 1). It is to be men- tection of a colistin-resistant K. pneumoniae coharboring tioned that sensitivity to imipenem was further tested using OXA-48 carbapenemase, which was isolated from a hos- E-test. The latter revealed the presence of imipenem MICs of pital in Algeria. 0.25, 1.5, and 1 mg/mL in M5, M6, and M7 strains, respec- In 2016, three colistin-resistant K. pneumoniae isolates tively. The carbapenemase activity of the two carbapenem- were recovered in Annaba University hospital, in Algeria, resistant isolates (M6 and M7) was thereafter confirmed by a from three different patients who have in common a uro- positive modified Carba-NP test performed as previously logical surgery antecedent (Table 1). The patients were described.5 admitted to the infectious diseases unit for recurrent urinary Multilocus sequence typing (MLST) analysis, according to tract infection, where urine cytobacteriology and antibiotic the Pasteur schemes available at the Institute Pasteur’s MLST susceptibility testing were performed. Of note, two of the Website (www.pasteur.fr/mlst/), revealed that all of them aforementioned patients had previously received colistin for belonged to the same sequence type ‘‘ST101.’’ Real-time PCR treatment of their recurrent urinary tract infection. amplification of carbapenemase-encoding genes (blaOXA-48,

1Aix Marseille University, IRD, APHM, MEPHI, IHU-Mediterranean Infection, Marseille, France. 2Microbiology Department, Annaba University Hospital, Annaba, Algeria. 3Department of Infectious Diseases, Annaba University Hospital, Annaba, Algeria.

1 2 YOUSFI ET AL.

blaNDM, blaVIM, and blaKPC) and ESBL-encoding genes (blaCTX-M, blaTEM, and blaSHV) (Table 2) was performed. Later, standard PCR amplification and DNA sequencing of phoP/Q, mutations the positive ones showed that all isolates were blaCTX-M-15, blaTEM-183, and blaSHV-106 positive, with two of them, which are ertapenem-resistant, coharboring blaOXA-48. None of the isolates expressed blaNDM, blaVIM, or blaKPC. The molecular mechanism of colistin resistance was investigated by PCR pmrB, amplification and sequencing of the pmrA, pmrB, phoP, mutations phoQ, mgrB, crrAB, mcr-1, and mcr-2 genes (Table 2). The plasmid-mediated colistin resistance genes mcr-1 and mcr-2 were absent in the three K. pneumoniae strains. Since CrrAB was not amplified in any single isolate by the PCR method,

pmrA, we concluded that this two-component system is absent in A217V V212G, T256A WT mutations three analyzed isolates. As previously reported, the crrAB system was absent in some K. pneumoniae, associated with the probable insertion sequence-mediated deletion or substi- 17 903B WT A217V T246AWT WT WT T246Atution WT in the locus. Sequence analysis revealed no mutations mgrB, IS

mutations in the phoP and phoQ genes, but showed an inactivating , , , , insertion in the mgrB gene in one isolate (M5) at nucleotide , , , 94 with 95% identity at the nucleotide level with IS903B insertion sequence (IS5 family of insertion sequences). The SHV-106 TEM-183 SHV-106 TEM-183 OXA-48 SHV-106 TEM-183 OXA-48 A217V pmrA substitution was observed in two strains (M5 CTX-M-15 CTX-M-15 CTX-M15 bla bla bla bla bla bla bla bla and M6) with mutations in the pmrB gene for the three bla bla bla Isolated from Urine Samples of Patients from Algeria isolates (V212G, T256A, and T246A) (Table 1). Colistin is the last-line antibiotic for treatment of infections by Gram-negative bacteria such as K. pneumoniae and the ongoing emergence of colistin and carbapenem resistance represents a serious problem for the management of infections caused by these bacteria.18 This study is in accordance with recent studies that highlighted the emergence of colistin resistance in MDR K. pneumoniae arising from loss of function by inactivation of the mgrB gene and activation of the PmrA/B system inducing modiﬁcation of the lipopoly- 17–19 Klebsiella pneumoniae saccharide. The A217V pmrA mutation shown in this CTX, CRO, FOX, GEN, CIP, CHL, NA, NIT, FOS CRO, FOX, GEN, CIP, NA, NIT, FOS CRO, FOX, GEN, CIP, CHL, NA, NIT, FOS study was reported previously in colistin-resistant K. pneumoniae from Serbia, also belonging to ST101 and harboring

oxacin; CRO, ceftriaxone; CT, colistin; CTX, cefotaxime; ERT, ertapenem; FOS, fosfomycin; FOX, cefoxitin; GEN, 4

ﬂ blaOXA-48. In this study, the authors concluded that this

1 CT, ATM, CAZ, CTX, mutation in the pmrA gene could play a role in the develop- < g/mL) Resistance phenotype bla genes

l ment of colistin resistance. These data would strengthen the ( ERT MIC presumption that this mutation was responsible for colistin resistance. The pmrB mutation T246A detected in this study was also shown in polymyxin B-resistant K. pneumoniae 20 g/mL) isolated from rectal swabs in Brazil. In this study, the au- l ( IMP MIC thors suggest that the speciﬁc pmrB mutation (T246A) found Downloaded by Gothenburg University Library from www.liebertpub.com at 09/26/18. For personal use only. was not capable of producing polymyxin resistance alone, since this mutation was also found in polymyxin-susceptible

g/mL) isolates and was considered not deleterious by PROVEAN l ( CT MIC software. To our knowledge, the other pmrB mutations detected in this study (V212G and T256A) have not been described previously.2,21 There are only three reports of genomic investigation on

Colistin OXA-48-producing K. pneumoniae ST101 that are also

prescription colistin resistant (red in Fig. 1). The strains from two of those reports (Serbia and Turkey) have amino acid changes in the pmrB gene.4,21 The third study, from Tunisia, de- Description of Colistin- and Carbapenem-Resistant scribed that the colistin resistance of K. pneumoniae was date 1. due to an insertion sequence in the mgrB coding gene be- Isolation tween the nucleotides 123 and 124. The inserted sequence 44

Table does not match with any identiﬁed IS sequences. Thus, this is the ﬁrst description of colistin- and carbapenem- ATM, aztreonam; CAZ, ceftazidime; CHL, chloramphenicol; CIP, cipro gentamicin; IMP, imipenem; MIC, minimum inhibitory concentration; NA, nalidixic acid; NIT, nitrofurantoin; WT, wild type. M6 20/10/2016 Yes 16 2 8 CT, ERT, ATM, CAZ, M7 30/11/2016 Yes 64 1 8 CT, ERT, ATM, CAZ, CTX, M5 23/05/2016 No 64 0.25 Strains resistant K. pneumoniae ST101 in Algeria. COLISTIN–CARBAPENEM-RESISTANT K. PNEUMONIAE ST101 3

Table 2. Primer Sequences of Klebsiella pneumoniae Analyzed in This Study Target gene Type of PCR Primer name Sequence (5¢-3¢) Reference crrAB Standard PCR TupA_F AAGTCCCAAAAGAGGCAAAC Cheng et al.8 H236-2575_R GTGAGGCCATCAAATTCTCG mgrB Standard PCR MgrB_F ATTCTGCCGCTTTTGCTG Olaitan et al.9 MgrB_R CGTTTTGAAACAAGTCGATGA mcr-1/2 Standard PCR Mcr-1/2_F GCAGCATACTTCTGTGTGGTAC This study Mcr-1/2_R TATGCACGCGAAAGAAACTGGC phoP Standard PCR phoP_F CGATGGTTGATGAGCTGAAA This study phoP_R TGCTGAGCCGGTAATGCTGGA phoQ Standard PCR phoQ_F GACGTCCCATCAGTACATCAATGG This study phoQ_R GCAGCATACTTCTGTGTGGTAC pmrA Standard PCR pmrA_F CATAGCCAGAAGACGCTGAT This study pmrA_R CAATCTGACGCTGAACATGG pmrB Standard PCR pmrB_F GTGCCACTAACGGTGGTACA This study pmrB_R GTCTGGAGAGATCGGGTCAA CTX-M Real-time PCR CTX-M_RT_F CGGGCRATGGCGCARAC This study CTX-M_RT_R TGCRCCGGTSGTATTGCC CTX-M_RT_Probe CCARCGGGCGCAGYTGGTGAC Standard PCR CTX-M_ALL_STD_F TTTGCGATGTGCAGTACCAGTAA Edelstein et al.10 CTX-M_ALL_STD_R CGATATCGTTGGTGGTGCCATA SHV Real-time PCR SHV_RT_F TCCCATGATGAGCACCTTTAAA Roschanski et al.11 SHV_RT_R TCCTGCTGGCGATAGTGGAT SHV_RT_Probe TGCCGGTGACGAACAGCTGGAG Standard PCR SHV_STD_F ATTTGTCGCTTCTTTACTCGC Yagi et al.12 SHV_STD_R TTTATGGCGTTACCTTTGACC TEM Real-time PCR ALLTEM_RT_F TTCTGCTATGTGGTGCGGTA This study ALLTEM_RT_R GTCCTCCGATCGTTGTCAGA ALLTEM_RT_Probe AACTCGGTCGCCGCATACA CTATTCTCAGA Standard PCR ALLTEM_STD_F ATGAGTATTCAACATTTCCGTG Kruger et al.13 ALLTEM_STD_R TTACCAATGCTTAATCAGTGAG OXA-48 Real-time PCR OXA48_RT_F TCTTAAACGGGCGAACCAAG This study OXA48_RT_R GCGTCTGTCCATCCCACTTA OXA48_RT_Probe 6-FAM-AGCTTGATCGCCCTCGA TTTGG-TAMRA Standard PCR OXA48_STD_F TTGGTGGCATCGATTATCGG Poirel et al.14 OXA48_STD_R GAGCACTTCTTTTGTGATGGC KPC Real-time PCR KPC_RT_ F GATACCACGTTCCGTCTGGA This study KPC_RT_R GGTCGTGTTTCCCTTTAGCC KPC_RT_Probe 6-FAM-CGCGCGCCGTGACGGA AAGC-TAMRA VIM Real-time PCR ALLVIM_RT_F CACAGYGGCMCTTCTCGCGGAGA This study ALLVIM_RT_R GCGTACGTYGCCACYCCAGCC ALLVIM_RT_Probe 6-FAM-AGTCTCCACGCACTTTCATGA CGACCGCGTCGGCG-TAMR 15 Downloaded by Gothenburg University Library from www.liebertpub.com at 09/26/18. For personal use only. NDM Real-time PCR NDM_RT_F GCGCAACACAGCCTGACTTT Diene et al. NDM_RT_R CAGCCACCAAAAGCGATGTC NDM_RT_Probe 6-FAM-CAACCGCGCCCAACTTTG GC-TAMRA Real-time PCR

The analysis results of M5 colistin-resistant strain with be prevented. It is urgent to establish a powerful monitoring the mgrB truncation collected from a patient not treated with system in each hospital with perfect coordination between colistin showed that the clinical use of colistin may not be all Algerian hospitals to detect as soon as possible an epi- the only reason for the emergence of colistin resistance. demic infection and prevent the spread of such MDR bac- Another possibility is the horizontal transmission between teria inducing infections that are difﬁcult to treat.45 patients, who have in common a stay in the urological unit of the same hospital. However, this is probably not the case Acknowledgments here since only one isolate with the mgrB truncation was isolated. Thus, a possible spread of nosocomial infections to The author thanks TradOnline for English correction. a larger number of patients and healthy individuals should This work was supported by the French Government under 4 YOUSFI ET AL.

FIG. 1. Geographical distribution of the carbapenemase-producing Klebsiella pneumoniae ST101 with different colistin susceptibility phenotype. Other carbapenemases include KPC, NDM, and OXA-181. ColiR, colistin resistant.4,9,21–43

the ‘‘Investissements d’avenir’’ program managed by the 6. Yanat, B., J. Machuca, R.D. Yahia, A. Touati, and A´ . Agence Nationale de la Recherche (reference: Me´diterraneé Pascual. 2016. First report of the plasmid-mediated colistin Infection 10-IAHU-03). resistance gene mcr-1 in a clinical Escherichia coli isolate in Algeria. Int. J. Antimicrob. Agents 48:760–761. 7. Olaitan, A.O., S. Chabou, L. Okdah, S. Morand, and J.M. Disclosure Statement Rolain. 2016. Dissemination of the mcr-1 colistin resistance No conflict of interest or financial disclosure exists to gene. Lancet Infect. Dis. 16:147. declare for all authors. 8. Cheng, Y.H., T.L. Lin, Y.T. Lin, and J.T. Wang. 2016. Amino acid substitutions of CrrB responsible for resistance to colistin through CrrC in Klebsiella pneumoniae. Anti- References microb. Agents Chemother. 60:3709–3716. 1. Biswas, S., J.M. Brunel, J.C. Dubus, M. Reynaud-Gaubert, 9. Olaitan, A.O., S.M. Diene, M. Kempf, M. Berrazeg, S. J.M. Rolain. 2012. Colistin: an update on the antibiotic of Bakour, S.K. Gupta, B. Thongmalayvong, K. Akkhavong, the 21st century. Expert. Rev. Anti. Infect. Ther. 10:917– S. Somphavong, P. Paboriboune, K. Chaisiri, C. Komala- Downloaded by Gothenburg University Library from www.liebertpub.com at 09/26/18. For personal use only. 934. misra, O.O. Adelowo, O.E. Fagade, O.A. Banjo, A.J. Oke, 2. Olaitan, A.O., S. Morand, and J.M. Rolain. 2014. Me- A. Adler, M.V. Assous, S. Morand, D. Raoult, and J.M. chanisms of polymyxin resistance: acquired and intrinsic Rolain. 2014. Emergence of colistin resistance in Klebsiella resistance in bacteria. Front. Microbiol. 5:1–18. pneumoniae from healthy humans and patients in Lao PDR, 3. Potron, A., L. Poirel, E. Rondinaud, and P. Nordmann. Thailand, Israel, Nigeria and France owing to inactivation 2013. Intercontinental spread of OXA-48 beta-lactamase- of the PhoP/PhoQ regulator mgrB: an epidemiological and producing Enterobacteriaceae over a 11-year period, 2001 molecular study. Int. J. Antimicrob. Agents 44:500–507. to 2011. Eurosurveillance 18:20549. 10. Edelstein, M., M. Pimkin, I. Palagin, I. Edelstein, and L. 4. Novovic´, K., A. Trudic´, S. Brkic´, Z. Vasiljevic´, M. Kojic´, Stratchounski. 2013. Prevalence and molecular epidemi- D. Medic´, I. C´ irkovic´, and B. Jovcˇic´. 2017. Molecular ep- ology of CTX-M extended-spectrum beta-lactamase- idemiology of colistin-resistant, carbapenemase-producing producing Escherichia coli and Klebsiella pneumoniae in Klebsiella pneumoniae, Serbia, 2013–2016. Antimicrob. Russian hospitals. Antimicrob. Agents Chemother. 47: Agents Chemother. 24:e02550-16. 3724–3732. 5. Bakour, S., A.O. Olaitan, H. Ammari, A. Touati, S. Saoudi, 11. Roschanski, N., J. Fischer, B. Guerra, and U. Roesler. 2014. K. Saoudi, and J.M. Rolain. 2015. Emergence of colistin- Development of a multiplex real-time PCR for the rapid and carbapenem-resistant Acinetobacter baumannii ST2 detection of the predominant beta-lactamase genes CTX-M, clinical isolate in Algeria: first case report. Microb. Drug SHV, TEM and CIT-type AmpCs in Enterobacteriaceae. Resist. 21:279–285. PLoS One 9:e100956. COLISTIN–CARBAPENEM-RESISTANT K. PNEUMONIAE ST101 5

12. Yagi, T., H. Kurokawa, N. Shibata, K. Shibayama, and Y. Samuelsen, M.D. Asensi. 2011. Molecular epidemiology of Arakawa. 2000. A preliminary survey of extended-spectrum KPC-2-producing Klebsiella pneumoniae isolates in Brazil: B-lactamases (ESBLs) in clinical isolates of Klebsiella the predominance of sequence type 437. Diagn. Microbiol. pneumoniae and Escherichia coli in Japan. FEMS Microbiol. Infect. Dis. 70:274–277. Lett 184:53–56. 26. Markovska, R., T. Stoeva, I. Schneider, L. Boyanova, V. 13. Kruger, T., D. Szabo, K.H. Keddy, K. Deeley, J.W. Marsh, Popova, D. Dacheva, R. Kaneva, A. Bauernfeind, V. Mitev, A.M. Hujer, R.A. Bonomo, and D.L. Paterson. 2004. In- and I. Mitov. 2015. Clonal dissemination of multilocus fections with nontyphoidal Salmonella species producing sequence type ST15 KPC-2-producing Klebsiella pneu- TEM-63 or a novel TEM enzyme, TEM-131, in South moniae in Bulgaria. APMIS 123:887–894. Africa. Antimicrob. Agents Chemother. 48:4263–4270. 27. Yu, J., Y. Wang, Z. Chen, X. Zhu, L. Tian, L. Li, and Z. 14. Poirel, L., T. Naas, and P. Nordmann. 2010. Diversity, Sun. 2017. Outbreak of nosocomial NDM-1-producing epidemiology, and genetics of class D b-lactamases. Anti- Klebsiella pneumoniae ST1419 in a neonatal unit. J. Glob. microb. Agents Chemother. 54:24–38. Antimicrob. Resist. 8:135–139. 15. Diene, S.M., N. Bruder, D. Raoult, and J.M. Rolain. 2011. 28. Oteo, J., D. Saez, V. Bautista, S. Fernandez-Romero, J.M. Real-time PCR assay allows detection of the New Delhi Hernandez-Molina, M. Perez-Vazquez, B. Aracil, and J. metallo-b- lactamase (NDM-1)-encoding gene in France. Campos. 2013. Carbapenemase-producing Enterobact- Int. J. Antimicrob. Agents 37:544–546. eriaceae in Spain in 2012. Antimicrob. Agents Chemother. 16. Seng, P., M. Drancourt, F. Gouriet, B. La Scola, P.E. 57:6344–6347. Fournier, J.M. Rolain, and D. Raoult. 2009. Ongoing rev- 29. Kirveskari, J., A.J. Hakanen, and O. Monica. 2012. olution in bacteriology: routine identification of bacteria by Carbapenemase-producing Enterobacteriaceae in Finland: matrix-assisted laser desorption ionization time-of-flight the first years (2008–2011). J. Antimicrob. Chemother. 67: mass spectrometry. Clin. Infect. Dis. 49:543–551. 2860–2864. 17. Baron, S., L. Hadjadj, J.M. Rolain, and A.O. Olaitan. 2016. 30. Liapi,s E., A. Pantel, J. Robert, M.H. Nicolas-Chanoine, L. Molecular mechanisms of polymyxin resistance: knowns Cavalie´, N. Van der Mee-Marquet, C. De Champs, N. and unknowns. Int. J. Antimicrob. Agents 48:583–591. Aissa, C. Eloy, V. Blanc, C. Guyeux, D. Hocquet, J.P. 18. Olaitan, A.O., and J.M. Rolain. 2015. Interruption of mgrB Lavigne, and X. Bertrand. 2014. Molecular epidemiology in the mediation of colistin resistance in Klebsiella oxytoca. of OXA-48-producing Klebsiella pneumoniae in France. Int. J. Antimicrob. Agents 46:354–356. Clin. Microbiol. Infect. 20:1121–1123. 19. Poirel, L., A. Jayol, and P. Nordmann. 2017. Polymyxins: 31. Mshana, S.E., M. Fritzenwanker, L. Falgenhauer, E. Do- antibacterial activity, susceptibility testing, and resistance mann, and T. Hain. 2015. Molecular epidemiology and mechanisms encoded by plasmids or chromosomes. Clin. characterization of an outbreak causing Klebsiella pneumo- Microbiol. Rev. 30:557–596. niae clone carrying chromosomally located bla CTX-M-15 20. Aires, C.A.M., P.S. Pereira, M.D. Asensi, and A.P.D. at a German University-Hospital. BMC Microbiol. 15:122. Carvalho-Assef. 2016. MgrB mutations mediating poly- 32. Melegh, S., G. Schneider, M. Horva´th, F. Jakab, L. Emody,} myxin B resistance in Klebsiella pneumoniae isolates from and Z. Tigyi. 2015. Identification and characterization of rectal surveillance swabs in Brazil. Antimicrob. Agents CTX-M-15 producing Klebsiella pneumoniae clone ST101 Chemother. 60:6969–6972. in a Hungarian university teaching hospital. Acta. Micro- 21. Jayol, A., L. Poirel, A. Brink, M.V. Villegas, M. Yilmaz, biol. Immunol. Hung. 62:233–245. and P. Nordmann. 2014. Resistance to colistin associated 33. Mammina, C., C. Bonura, A. Aleo, T. Fasciana, T. Brunelli, with a single amino acid change in protein PmrB among G. Pesavento, R. Degl’Innocenti R, and A. Nastasi. 2012. Klebsiella pneumoniae isolates of worldwide origin. Anti- Sequence type 101 (ST101) as the predominant carbapenem- microb. Agents Chemother. 58:4762–4766. non-susceptible Klebsiella pneumoniae clone in an acute 22. Hrabak, J., J. Empel, T. Bergerova, K. Fajfrlik, P. Urbas- general hospital in Italy. Int. J. Antimicrob. Agents 39:543– kova, I. Kern-Zdanowicz, W. Hryniewicz, and M. Gniad- 545. kowski. 2009. International clones of Klebsiella 34. Frasson, I., E. Lavezzo, E. Franchin, S. Toppo, L. Barzon, pneumoniae and Escherichia coli with extended-spectrum A. Cavallaro, S.N. Richter, and G. Palu. 2012. Anti- b-lactamases in a Czech Hospital. J. Clin. Microbiol. 47: microbial treatment and containment measures for an ex- Downloaded by Gothenburg University Library from www.liebertpub.com at 09/26/18. For personal use only. 3353–3357. tremely drug-resistant Klebsiella pneumoniae ST101 isolate 23. Kaase, M., S. Schimanski, R. Schiller, B. Beyreiß, A. carrying pKPN101-IT, a novel fully sequenced blaKPC-2 Thu¨rmer, J. Steinmann, V.A. Kempf, C. Hess, I. Sobottka, plasmid. J. Clin. Microbiol. 50:3768–3772. I. Fenner, S. Ziesing, I. Burckhardt, L. von Mu¨ller, A. 35. Hashimoto, A., M. Nagamatsu, N. Ohmagari, K. Hayaka- Hamprecht, I. Tammer, N. Wantia, K. Becker, T. Holz- wa, Y. Kato, and T. Kirikae. 2014. Isolation of OXA-48 mann, M. Furitsch, G. Volmer, and S.G. Gatermann. 2016. carbapenemase-producing Klebsiella pneumonia ST101 Multicentre investigation of carbapenemase-producing Es- from an overseas traveler returning to Japan. Jpn. J. Infect. cherichia coli and Klebsiella pneumoniae in German Dis. 67:120–121. Hospitals. Int. J. Med. Microbiol. 306:415–420. 36. Lee, S.-Y., Y.-J. Park, J.K. Yu, S. Jung, Y. Kim, S.H. Jeong, 24. De Laveleye, M., T.D. Huang, P. Bogaerts, C. Berhin, C. and Y. Arakawa. 2012. Prevalence of acquired fosfomycin Bauraing, P. Sacre´, A. Noel, and Y. Glupczynski. 2017. resistance among extended-spectrum-lactamase-producing Increasing incidence of carbapenemase-producing Escher- Escherichia coli and Klebsiella pneumoniae clinical isolates ichia coli and Klebsiella pneumoniae in Belgian Hospitals. in Korea and IS26-composite transposon surrounding fosA3. Eur. J. Clin. Microbiol. Infect. Dis. 36:139–146. J. Antimicrob. Chemother. 67:2843–2847. 25. Seki, L.M., P.S. Pereira, M.da.P. de Souza M da, M.de.S. 37. Pena, I., J.J. Picazo, C. Rodr´ıguez-avial, and I. Rodr´ıguez- Conceiçaõ, E.A. Marques, C.O. Porto, E.M.L. Colnago, avial. 2014. Carbapenemase-producing Enterobacteriaceae C.de.F.M. Alves, D. Gomes, A.P.D.A.C. Assef, and O. in a tertiary hospital in Madrid, Spain: high percentage of 6 YOUSFI ET AL.

colistin resistance among VIM-1-producing Klebsiella pneu- 43. Maamar, E., S. Ferjani, A. Jendoubi, S. Hammami, Z. moniae ST11 isolates. Int. J. Antimicrob. Agents 43:460–464. Hamzaoui, L. Mayonnove-Coulange, M. Saidani, A. Kam- 38. Marcade, G., S. Brisse, S. Bialek, E. Marcon, V. Leflon- moun, A. Rehaiem, S. Ghedira, M. Houissa, I. Boutiba-Ben Guibout, V. Passet, R. Moreau, and M.-H. Nicolas-Chanoine. Boubaker, A. Slim, and V. Dubois. 2016. High prevalence of 2013. The emergence of multidrug-resistant Klebsiella gut microbiota colonization with broad-spectrum cephalo- pneumoniae of international clones ST13, ST16, ST35, ST48 sporin resistant Enterobacteriaceae in a Tunisian intensive and ST101 in a teaching hospital in the Paris region. Epide- care unit. Front. Microbiol. 7:1–10. miol. Infect. 141:1705–1712. 44. Mansour, W., M. Haenni, E. Saras, R. Grami, Y. Mani, A. 39. Ska´lova´, A., K. Chudeˇjova´, V. Rotova´, M. Medvecky, V. Ben Haj Khalifa, S. El Atrouss, M. Kheder, M. Fekih Sˇtudentova´, E. Chudaćˇkova´, P. Lavicka, T. Bergerova, V. Hassen, N. Boujaâfar, O. Bouallegue, and J.Y. Madec. Jakubu, H. Zemlickova, C.C. Papagiannitsis, and J. Hraba´k. 2017. Outbreak of colistin-resistant carbapenemase- 2016. Molecular characterization of OXA-48-like- producing Klebsiella pneumoniae in Tunisia. J. Glob. An- producing Enterobacteriaceae in the Czech Republic: evi- timicrob. Resist. 10:88–94. dence for horizontal transfer of pOXA-48-like plasmids. 45. Okdah, L., T. Leangapichart, L. Hadjadj, A.O. Olaitan, C. Antimicrob. Agents Chemother. 188:9–16. Al-Bayssari, R. Rizk, M. Hammoud, S.M. Diene, and J.M. 40. Venditti, C., L. Villa, A. Capone, D. Fortini, S. D’Arezzo, C. Rolain. 2017. First report of colistin-resistant Klebsiella Nisii, E. Bordi, V. Puro, M. Antonini, A. Carattoli, M.A. pneumoniae clinical isolates in Lebanon. J. Glob. Anti- Cataldo, N. Petrosillo, and A. Di Caro. 2016. Isolation of KPC microb. Resist. 9:15–16. 3-producing Enterobacter aerogenes in a patient colonized by MDR Klebsiella pneumoniae. New Microbiol. 39:310–313. 41. Pitart, C., M. Sole, I. Roca, A. Fabrega, J. Vila, and F. Address correspondence to: Marco. 2011. First outbreak of a plasmid-mediated Jean-Marc Rolain, PharmD, PhD carbapenem-hydrolyzing OXA-48-lactamase in Klebsiella Aix Marseille University, IRD, APHM, MEPHI, pneumoniae in Spain. Antimicrob. Agents Chemother. 55: IHU-Mediterranean Infection 4398–4401. Faculty of Medicine and Pharmacy 42. Izdebski, R., K. Bojarska, A. Baraniak, E. Literacka, M. 19-21 Boulevard Jean Moulin Herda, D. Zabicka,_ A. Guzek, M. Po´łgrabia, W. Hryniewicz, 13385 Marseille, CEDEX 05 and M. Gniadkowski. 2015. NDM-1 or OXA-48-producing France Enterobacteriaceae colonising Polish tourists following a terrorist attack in Tunis. Eurosurveillance 20:1–4. E-mail: [email protected] Downloaded by Gothenburg University Library from www.liebertpub.com at 09/26/18. For personal use only.

Article 7:

Genome sequence and description of Olsenella Timonensis sp. nov.

isolated from human fecal sample in France.

Hanane Yousfi, Liliane Okdah, Sokhna Ndongo, Nicholas Armstrong, Magali Richez, Bruno

Senghor, Didier Raoult, Jean Marc Rolain and Fadi Bittar

In progress « Draft manuscript »

134

1 Title: Genome sequence and description of Olsenella Timonensis sp. Nov. Isolated from

2 human fecal sample in France.

3 Authors Hanane Yousfi1, Liliane Okdah1, Sokhna Ndongo1, Nicholas Armstrong1, Magali

4 Richez1, Bruno Senghor1, Didier Raoult1, Jean Marc Rolain1 and Fadi Bittar1*

5 1 Aix Marseille Univ, IRD, APHM, MEPHI, IHU-Méditerranée Infection, Marseille, France.

6 * Corresponding author:

7 Fadi Bittar, IHU Méditerranée Infection, 19-21 boulevard Jean Moulin, 13005 Marseille,

8 France. Email: [email protected]

9 Keywords: Olsenella Timonensis sp. nov., genomic description, taxono-genomics, human

10 culturomics

11 Word count

12 Abstract: 136 words

13 Text: 2381 words

14 Figures: 3

15 Tables: 3

16 References: 29

20 21 Abstract

22 Olsenella Timonenesis strain P2300 is a gram-strain-positive, strictly anaerobic, non-

23 motile, non-sporulating coccobacillus, isolated from fecal swab of a men with inflammatory

24 syndrome, colonic polyp and diverticulosis, in France. Phylogenetic analyses, based on 16S

25 rRNA gene sequences, revealed that the isolate represented a new lineage within the genus

26 Olsenella of the family Coriobacteriaceae. The character of this bacterium, with the complete

27 genome sequence and annotation, will be investigated here. The major fatty acids were

28 Hexadecanoic acid (43 %), 9-Octadecenoic acid (35 %) and Octadecanoic acid (8 %). The

29 genome size is 217, 673, 37 bp and has a G+C content of 68, 08 mol%. It contains 1,944 protein-

30 coding and 58 RNA genes, including 52 tRNA genes and 6 rRNA genes. The Olsenella

31 timonensis strain P2300 strain exhibited 70% 16S rRNA similarity with Olsenella profusa, the

32 phylogenetically closest species in the family Coriobacteriaceae.

33 .

42 43 Introduction

44 Olsenella is one of the thirteen genera in the family Coriobacteriaceae. In 1991, the genus is

45 named in honor of Ingar Olsen, a contemporary Norwegian microbiologist, who was the first

46 to describe this strain as ‘Lactobacillus uli’ which transferred to the new genus Olsenella as

47 O.uli because of the divergence on 16S rRNA gene sequence and the presence of same

48 phenotypic characters (1). Another novel species O. profusa was also described. According to

49 the Dewhirst et al (2), the human oral cavity is the main habitat and the bovine rumen a likely

50 habitat of the Olsenella. Isolates of O. uli and O. profusa are regularly recovered from disease

51 sites in the human mouth and sometimes from blood of human with local oral or

52 gastrointestinal infections(3). Olsenella are found in the healthy and acidotic bovine rumen.

53 Molecular genetic studies have also reported the detection of Olsenella-related clones in the

54 gastrointestinal tracts of humans, pigs (4), wallabies (5), and chicken (6) and in divers’

55 anaerobic environmental sites (7). This indicate that the habitats of Olsenella generally

56 comprise the oral cavity and gastrointestinal tract of homeothermic vertebrates and that non-

57 animal habitats also exist.

58 In the present study, we present a set of features and a summary classification for a novel

59 species belonging to the genus Olsenella which was isolated from the stool of 42-year-old

60 man, living in France. This isolation was part of a ''culturomics'' study aiming at cultivating

61 the maximum number of bacterial species from human feces.

62 Materials and Methods

63 Sample collection and growth conditions

64 The strain P2300 was isolated in October 2015 from a fecal sample, collected from 73-year-

65 old French male, with an inflammatory syndrome and digestives antecedents of colonic

66 polyps and diverticulosis. The fecal specimen was pre-incubated during 3 days in anaerobic 67 hemoculture bottle supplemented with 4ml of blood and 4ml of rumen. Obtained inoculum

68 was incubated 72 hours on 5% sheep blood-enriched Columbia agar (Biomérieux, Marcy

69 l’Etoile, France) at 37°C. Growth was tested under aerobic and anaerobic conditions using

70 AnaeroGenTM Compact (BioMerieux Marcy l’Etoile, France). Many temperatures (22, 37, 42,

71 55°C) were tested to determine the optimal growth temperature of the strain. Optimal salt

72 concentration required for growth was determined by growing the strain at 5, 20, 50, 70 and

73 100% of NaCl.

74 Identification by MALDI-TOF MS and 16S rRNA sequencing

75 Obtained colonies were isolated on 5% sheep blood-enriched Columbia agar, and identified

76 using MALDI-TOF MS and 16S rRNA sequencing. MALDI-TOF identification,

77 measurement and analysis were performed as previously described (8,9) by a Microflex

78 spectrometer with a MTP 96 MALDI-TOF target plate (Bruker Daltonics, Leipzig, Germany).

79 The obtained spectra are imported into the MALDI BioTyper software (version 2.0, Bruker)

80 and analyzed by standard pattern matching against our databases (Bruker database and

81 URMITE database, incremented with our data). When the identification score was ≥ 1.9 the

82 species level was determined; whereas a score between 1.9 and 1.7 allowed identification only

83 at the genus level and a score < 1.7 gave no identification. In this last case, the 16S rRNA

84 gene was sequenced as previously described (10,11) in order to get a molecular identification.

85 The threshold of 98.7% and 95% similarity level were used to define a new species or a new

86 genus, respectively, without performing DNA-DNA hybridization (12).

87 Phylogenetic analysis

88 The phylogenetic tree highlighting the phylogenetic position of this bacterium relative to

89 other species was identified. Sequences were recovered using a nucleotide blast against the

90 16S RNA Database of Silva’s “The All-Species Living Tree” project (LTPs119). A filter to 91 eliminate sequences with a size under 1,450 was applied. Pass filter sequences were aligned

92 using Muscle and phylogenetic inferences were obtained using the approximately maximum-

93 likelihood method within the Fast Tree software. The support local values were computed

94 using the Shimodaira-Hasegawa test. A filter using PhyloPattern (13) was applied to the tree

95 to remove duplicate species in the tree or poor taxonomic reference species.

96 Biochemical characterization and antibiotics susceptibility test

97 Phenotypic tests were performed using an API ZYM strip, an API 50CH strip and an API

98 20A strip (BioMerieux, Marcy l’Etoile, France) according to the manufacturer’s instructions.

99 In vitro susceptibility to antibiotics was determined using the disc diffusion method (i2a,

100 Montpellier, France) on Muller-Hinton agar. Oxidase (Becton, Dickinson and Company, Le

101 Pont de Claix, and France) and catalase assays (BioMerieux, Marcy l’Etoile, France) were

102 performed separately.

103 Morphological characterization by microscopy

104 To determine the Gram stain, we used the color Gram kit 2 (BioMérieux, Marcy l’Etoile,

105 France) and observed it under a DM1000 photonic microscope (Leica Microsystems).

106 A sporulation test was done by a thermal shock at 100°C for half an hour. A fresh colony was

107 observed between blades and slats using a Leica DM 1000 microscope (Leica Microsystems,

108 Nanterre, France) to assess its motility. In order to observe cell morphology, cells were fixed

109 with 2.5 % glutaraldehyde in 0.1M cacodylate buffer for at least 1h at 4°C. A drop of cell

110 suspension was deposited for approximately 5 minutes on glow-discharged formvar carbon

111 film on 400 mesh nickel grids (FCF400-Ni, EMS). The grids were dried on blotting paper and

112 cells were negatively stained for 10 s with 1% ammonium molybdate solution in filtered

113 water at RT. Electron micrographs were acquired with a Morgagni 268D (Philips)

114 transmission electron microscope operated at 80 keV. 115 Fatty acid methyl ester (FAME) analysis by GC/MS

116 Cellular fatty acid methyl ester (FAME) analysis was performed by GC/MS. Two samples were

117 prepared with approximately 8 mg of bacterial biomass per tube harvested from several culture

118 plates. Fatty acid methyl esters were prepared as described by Sasser (2006). GC/MS analyses

119 were carried out as described before (Dione, 2016). Briefly, fatty acid methyl esters were

120 separated using an Elite 5-MS column and monitored by mass spectrometry (Clarus 500 - SQ

121 8 S, Perkin Elmer, Courtaboeuf, France). Spectral database search was performed using MS

122 Search 2.0 operated with the Standard Reference Database 1A (NIST, Gaithersburg, USA) and

123 the FAMEs mass spectral database (Wiley, Chichester, UK).

124 DNA extraction and genome sequencing

125 Genomic DNA (gDNA) of Olsenella Timonensis P2300 strain was extracted in two steps. First,

126 a mechanical treatment was first performed by glass beads acid washed (G4649-500g Sigma)

127 using a FastPrep BIO 101 instrument (Qbiogene, Strasbourg, France) at maximum speed (6.5)

128 for 3x 30s. Then, after a 2 hours lysozyme incubation at 37°C, DNA was extracted on the EZ1

129 biorobot (Qiagen, Courtaboeuf, France) with EZ1 DNA tissues kit. The elution volume is 50µL.

130 gDNA was quantified by a Qubit assay with the high sensitivity kit (Life technologies,

131 Carlsbad, CA, USA) to 117.7 ng/µl gDNA was sequenced on the MiSeq Technology (Illumina

132 Inc, San Diego, CA, USA) with the mate pair strategy. The gDNA was barcoded in order to be

133 mixed with 11 other projects with the Nextera Mate Pair sample prep kit (Illumina). The mate

134 pair library was prepared with 1.5 µg of DNA using the Nextera mate pair Illumina guide. The

135 gDNA sample was simultaneously fragmented and tagged with a mate pair junction adapter.

136 The pattern of the fragmentation was validated on an Agilent 2100 BioAnalyzer (Agilent

137 Technologies Inc, Santa Clara, CA, USA) with a DNA 7500 labchip. The DNA fragments

138 ranged in size from 1.5 kb up to 11kb with an optimal size at 6.88 kb. No size selection was

139 performed and 533 ng of tagmented fragments were circularized. The circularized DNA was 140 mechanically sheared to small fragments with an optimal at 639 bp on the Covaris device S2 in

141 T6 tubes (Covaris, Woburn, MA, USA). The library profile was visualized on a High Sensitivity

142 Bioanalyzer LabChip (Agilent Technologies Inc, Santa Clara, CA, USA) and the final

143 concentration library was measured at 135.83 nmol/l. The libraries were normalized at 2nM

144 and pooled. After a denaturation step and dilution at 15 pM, the pool of libraries was loaded

145 onto the reagent cartridge and then onto the instrument along with the flow cell. Automated

146 cluster generation and sequencing run were performed in a single 39-hours run in a 2x251-bp.

147 Total information of 5.9 Gb was obtained from a 624 K/mm2 cluster density with a cluster

148 passing quality control filters of 96.33 % (11,600,000 passing filter paired reads). Within this

149 run, the index representation for strain GM2T was determined to 6.40 %. The 742,400 paired

150 reads were trimmed and then assembled.

151 Genome assembly, annotation and comparison

152 The genome’s assembly was performed with a pipeline that enabled to create an assembly

153 with different softwares (Velvet (14), Spades (15) and Soap Denovo (16)), on trimmed

154 (MiSeq and Trimmomatic (17) softwares) or untrimmed data (only MiSeq software). For each

155 of the six assemblies performed, GapCloser (16) was used to reduce gaps. Then

156 contamination with Phage Phix was identified (blastn against Phage Phix174 DNA sequence)

157 and eliminated. Finally, scaffolds which size was under 800 bp were removed and scaffolds

158 with a depth value lower than 25% of the mean depth were removed (identified as possible

159 contaminants). The best assembly was selected by using different criteria (number of

160 scaffolds, N50, number of N). For the studied strain, Velvet gave the best assembly, with a

161 depth coverage of 63x.

162 Open Reading Frames (ORFs) were predicted using Prodigal (18) with default parameters.

163 The predicted ORFs were excluded if they were spanning a sequencing gap region (contain

164 N). The predicted bacterial protein sequences were searched against the Clusters of 165 Orthologous Groups (COG) using BLASTP (E-value of 1e-03, coverage 0,7 and identity

166 percent 30%). If no hit was found, it search against the NR database using BLASTP (E-value

-03 167 of 1e , coverage 0,7 and identity percent of 30%). If the sequence’s length was smaller than

-05 168 80 amino acids, we used an E-value of 1e . The tRNAScanSE tool (19) was used to find

169 tRNA genes, while rRNA genes were identified using RNAmmer (20). Lipoprotein signal

170 peptides and the number of transmembrane helices were predicted using Phobius (21).

171 ORFans were identified if all the BLASTP performed didn't give positive results (E-value

-03 -05 172 smaller than 1e for ORFs with sequence size superior to 80 aa or E-value smaller than 1e

173 for ORFs with sequence length smaller than 80 aa). Such parameter thresholds have already

174 been used in previous works to define ORFans. The resistome was analysed using the

175 ARGANNOT (Antibiotic Resistance Gene Annotation) database and BLASTp in GenBank

176 (22). The exhaustive search for bacteriocin was performed using the database which is

177 available in our laboratory (Bacteriocins of the URMITE database BUR;

178 http://drissifatima.wix.com/bacteriocins). Protein sequences from this database allowed

179 putative bacteriocins from human gut microbiota to be identified using BLASTp methodology

180 (23). Analysis of the presence of polyketide synthase and non-ribosomal peptide synthase

181 (PKS/NRPS) was performed by discriminating the large size gene using the database in our

182 laboratory (24). Predicted proteins were compared against the non-redundant (nr) GenBank

183 database using blastp and were finally examined using antiSMASH (25). All annotation

184 process was performed in the Multi-Agent Software System DAGOBAH (26), including

185 Figenix (27) Libraries which provide pipeline analysis.

186 Species which must be compared were automatically retrieved from the 16s RNA tree using

187 PhyloPattern (13). For each selected species, the complete genome sequence, proteome

188 sequence and ORFeome sequence were retrieved from the FTP of NCBI. If one specific strain

189 didn’t have a complete and available genome, a complete genome of the same species was 190 used. If ORFeome and proteome were not predicted, Prodigal was used with default

191 parameters to predict them. The proteomes were analyzed with proteinOrtho (28). Then for

192 each couple of genomes, a similarity score (mean value of nucleotide similarity between all

193 couples of orthologous genes between the two genomes studied) was computed by AGIOS

194 tool (Average Genomic Identity Of gene Sequences) (29). Genome-to-Genome Distance

195 Calculator (GGDC) analysis was performed using the GGDC web server as previously

196 reported. An annotation of all proteomes was also realized to define the distribution of

197 functional classes of predicted genes according to the clusters of orthologous groups of

198 proteins (with same method than for genome annotation). The comparison process was

199 performed in the Multi-Agent Software System DAGOBAH (26), that includes Figenix (27)

200 Librairies which provided pipeline analysis and by using Phylopattern (13) for tree

201 manipulation.

202 Results and discussion

203 Strain identification and phylogenetic analyses

204 The isolated strain P2300 was obtained by direct culture of the stool sample on Columbia agar

205 supplemented with sheep blood (COS) after dilution. By MALDI-TOF MS analysis, we

206 obtained a score under 1.7. Therefore, the isolate was not a member of a known species

207 present in our database.

208 Phenotypic characterization

209 The strain Olsenella had grown at temperature 37°C after 72h incubation on COS. An

210 optimal growth was observed at 0 or 0.5% of NaCl, so the isolate did not require NaCl for

211 growth. The strain growth was obtained in strict anaerobic conditions. The motility test was

212 negative, and the strain was not able to form spores. Colonies which have grown in COS were

213 observed to be smooth, flat, circular, and 1-2mm in diameter as observed using an electron

214 microscope after negative staining (Figure 2). 215 Strain P2300 exhibited catalase and oxidase negative activity. Using API 50 CH strip,

216 positive reactions were observed for glycerol, D-arabinose, L-arabinose, D-ribose, D-xylose,

217 D-mannose, D-mannitol, N-acetylglucosamine, amygdalin, arbutin, esculin ferric citrate,

218 salicin, D-cellobiose, D-maltose, D-lactose, D-melibiose, D-saccharose, D-trehalose, Inulin,

219 D-melezitose, D-raffinose, Amidon, Glycogen, Gentiobiose, D-turanose, L-fucose, potassium

220 gluconate, and potassium 5-ketogluconate. Using API ZYM strip, positive reactions were

221 observed for α-glucosidase (maltase) and weakly positive reactions were observed for N-

222 acetyl-β-glucosaminidase and α-mannosidase. The antibiotic susceptibility test of strain

223 demonstrating that it was susceptible to cefoxitin, vancomycin, teicoplanin, linezolid,

224 ciprofloxacin, sulfamethoxazole/trimethoprim, doxycycline, erythromycin, clindamycin,

225 rifampicin. The major fatty acids were Hexadecanoic acid (43 %), 9-Octadecenoic acid (35

226 %) and Octadecanoic acid (8 %). Minor amounts of unsaturated, branched and other saturated

227 fatty acids were also detected

228 Genome properties

229 The genome is 2 176 737 bp long with 68, 08 % GC content. It is composed of 1 scaffold

230 (composed of 1contigs). Of the 2 002 predicted genes, 1 944 were protein-coding genes and

231 58 were RNAs (2 genes are 5S rRNA, 2 genes are 16S rRNA, 2 genes are 23S rRNA, 52

232 genes are TRNA genes). A total of 1 490 genes (76, 65%) were assigned as putative function

233 (by cogs or by NR blast). 102 genes were identified as ORFans (5, 25%).

234 The remaining genes were annotated as hypothetical proteins (288 genes => 14, 81%).

235 Conclusion

236 Considering the phenotypic, phylogenetic and genomic analyses described above, we propose

237 that strain P2300 represents the type strain of a novel species of the genus of Olsenella and we

238 propose the name Olsenella Timonensis. 239 Conflict of interest

240 The authors report no potential conflicts of interest or financial disclosure issues.

241 ACKNOWLEDGMENTS

242 Funding

243 This work was supported by the French Government under the « Investissements d’avenir »

244 (Investments for the Future) program managed by the Agence Nationale de la Recherche

245 (ANR, fr: National Agency for Research), (reference: Méditerranée Infection 10-IAHU-03)

246 References

247 1. Göker M, Held B, Lucas S, Nolan M, Yasawong M, del Rio TG, et al. Complete

248 genome sequence of Olsenella uli type strain (VPI D76D-27C T). Stand Genomic Sci.

249 2010;3(1):76–84.

250 2. Dewhirst FE, Spratt DA, Downes J, Wade WG, Paster BJ, Tzellas N, et al.

251 Characterization of novel human oral isolates and cloned 16S rDNA sequences that fall

252 in the family Coriobacteriaceae: description of Olsenella gen. nov., reclassification of

253 Lactobacillus uli as Olsenella uli comb. nov. and description of Olsenella profusa sp.

254 nov. Int J Syst Evol Microbiol. 2011;51(5):1797–804.

255 3. Kraatz M, Wallace RJ, Svensson L. Olsenella umbonata sp. nov., a microaerotolerant

256 anaerobic lactic acid bacterium from the sheep rumen and pig jejunum, and emended

257 descriptions of Olsenella, Olsenella uli and Olsenella profusa. Int J Syst Evol

258 Microbiol. 2011;61(4):795–803.

259 4. Li X, Jensen RL, Højberg O, Canibe N, Jensen BB. Olsenella scatoligenes sp. nov., a

260 3-methylindole- (skatole) and 4-methylphenol- (p-cresol) producing bacterium isolated

261 from pig faeces. Int J Syst Evol Microbiol. 2015;65(2015):1227.

262 5. Chhour KL, Hinds LA, Deane EM, Jacques NA. The microbiome of the cloacal 263 openings of the urogenital and anal tracts of the tammar wallaby, Macropus eugenii.

264 Microbiology. 2008;154(5):1535–43.

265 6. Maurer JJ, Lee M, Lu J, Idris U, Harmon B, Hofacre C, et al. Diversity and Succession

266 of the Intestinal Bacterial Community of the Maturing Broiler Chicken Maturing

267 Broiler Chicken. 2013;69(December):6816–24.

268 7. Bowman KS, Moe WM, Rash BA, Bae HS, Rainey FA. Bacterial diversity of an acidic

269 Louisiana groundwater contaminated by dense nonaqueous-phase liquid containing

270 chloroethanes and other solvents. FEMS Microbiol Ecol. 2006;58(1):120–33.

271 8. Seng P, Drancourt M, Gouriet F, La Scola B, Fournier P-E, Rolain JM, et al. Ongoing

272 revolution in bacteriology: routine identification of bacteria by matrix-assisted laser

273 desorption ionization time-of-flight mass spectrometry. Clin Infect Dis

274 9. Seng P, Rolain J-M, Fournier PE, La Scola B, Drancourt M, Raoult D. MALDI-TOF-

275 mass spectrometry applications in clinical microbiology. Future Microbiol. 2010

276 Nov;5(11):1733–54.

277 10. Khamis A, Raoult D, La Scola B. Comparison between rpoB and 16S rRNA gene

278 sequencing for molecular identification of 168 clinical isolates of Corynebacterium. J

279 Clin Microbiol. 2005 Apr;43(4):1934–6.

280 11. Drancourt M, Berger P, Raoult D. Systematic 16S rRNA gene sequencing of atypical

281 clinical isolates identified 27 new bacterial species associated with humans. J Clin

282 Microbiol. 2004 May;42(5):2197–202.

283 12. Kim M, Oh HS, Park SC, Chun J. Towards a taxonomic coherence between average

284 nucleotide identity and 16S rRNA gene sequence similarity for species demarcation of

285 prokaryotes. Int J Syst Evol Microbiol. 2014;64(PART 2):346–51.

286 13. Gouret P, Thompson JD, Pontarotti P. PhyloPattern: regular expressions to identify

287 complex patterns in phylogenetic trees. BMC Bioinformatics. 2009 Sep;10. 288 14. Zerbino DR, Birney E. Velvet: algorithms for de novo short read assembly using de

289 Bruijn graphs. Genome Res. 2008 May;18(5):821–9.

290 15. Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, Kulikov AS, et al.

291 SPAdes: a new genome assembly algorithm and its applications to single-cell

292 sequencing. J Comput Biol. 2012 May;19(5):455–77.

293 16. Luo R, Liu B, Xie Y, Li Z, Huang W, Yuan J, et al. SOAPdenovo2: an empirically

294 improved memory-efficient short-read de novo assembler. Gigascience. 2012

295 Dec;1(1):18.

296 17. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina

297 sequence data. Bioinformatics. 2014 Aug;30(15):2114–20.

298 18. Link. Prodigal (http://prodigal.ornl.gov/).

299 19. Lowe TM, Eddy SR. tRNAscan-SE: a program for improved detection of transfer RNA

300 genes in genomic sequence. Nucleic Acids Res. 1997 Mar;25(5):955–64.

301 20. Lagesen K, Hallin P, Rødland EA, Staerfeldt H-H, Rognes T, Ussery DW. RNAmmer:

302 consistent and rapid annotation of ribosomal RNA genes. Nucleic Acids Res.

303 2007;35(9):3100–8.

304 21. Käll L, Krogh A, Sonnhammer ELL. A combined transmembrane topology and signal

305 peptide prediction method. J Mol Biol. 2004 May;338(5):1027–36.

306 22. Gupta SK, Padmanabhan BR, Diene SM, Lopez-Rojas R, Kempf M, Landraud L, et al.

307 ARG-ANNOT, a new bioinformatic tool to discover antibiotic resistance genes in

308 bacterial genomes. Antimicrob Agents Chemother. 2014;58(1):212–20.

309 23. Drissi F, Buffet S, Raoult D, Merhej V. Common occurrence of antibacterial agents in

310 human intestinal microbiota. Front Microbiol. 2015;6:441.

311 24. Conway KR, Boddy CN. ClusterMine360: a database of microbial PKS/NRPS

312 biosynthesis. Nucleic Acids Res. 2013 Jan;41(Database issue):D402-7. 313 25. Weber T, Blin K, Duddela S, Krug D, Kim HU, Bruccoleri R, et al. antiSMASH 3.0-a

314 comprehensive resource for the genome mining of biosynthetic gene clusters. Nucleic

315 Acids Res. 2015 Jul;43(W1):W237-43.

316 26. Gouret P., Paganini J, Dainat J, Louati D., Darbo E., Ponatarotti P. LA. Integration of

317 Evolutionary Biology Concepts for Functional Annotation and Automation of Complex

318 Research in Evolution: The Multi-Agent Software System DAGOBAH, Evolutionary

319 Biology Concepts, Biodiversity, Macroevolution and Genome Evolution. 2011;71–87.

320 27. Gouret P, Vitiello V, Balandraud N, Gilles A, Pontarotti P, Danchin EGJ. FIGENIX:

321 intelligent automation of genomic annotation: expertise integration in a new software

322 platform. BMC Bioinformatics. 2005 Aug;6:198.

323 28. Lechner M, Findeiss S, Steiner L, Marz M, Stadler PF, Prohaska SJ. Proteinortho:

324 detection of (co-)orthologs in large-scale analysis. BMC Bioinformatics. 2011

325 Apr;12:124.

326 29. Ramasamy D, Mishra AK, Lagier J-C, Padhmanabhan R, Rossi M, Sentausa E, et al. A

327 polyphasic strategy incorporating genomic data for the taxonomic description of novel

328 bacterial species. Int J Syst Evol Microbiol. 2014 Feb;64(Pt 2):384–91.

329

330

331

332

333

334

335 Figure legends. 336 Figure 1. A consensus Phylogenetic tree based on 16S rRNA gene sequence comparisons,

337 showing the relationships between strain Olsenella Timonensis, and other type strains from

338 the genera Olsenella. Sequences were aligned using CLUSTALW, and phylogenetic

339 inferences were obtained using the neighbor-joining method in the MEGA 7 software

340 package. Numbers above the node are percentages of bootstrap values. The scale bar

341 represents 0.02 substitutions per nucleotide position.

342 Figure 2. Transmission electron microscopy of the Olsenella Timonensis, using Morgani 268

343 Dalton (Philips) at an operating voltage of 60 Kv. The scale bar represents 500nm.

344 Figure 3. Graphical circular map of the chromosome. From outside to the center: genes on the

345 forward strand colored by COG categories (only genes assigned to COG), genes on the

346 reverse strand colored by COG categories (only gene assigned to COG), RNA genes (tRNAs

347 green, rRNAs red), GC content and GC skew.

348

349

350

351

352

353

354

355

356

357

358 Table 1. Nucleotide content and gene count levels of the genome

Value % of total Description

[J] 168 9 Translation

[A] 0 0 Rna processing and modification

[K] 109 6 Transcription

[L] 59 3 Replication, recombination and repair

[B] 0 0 Chromatin structure and dynamics

[D] 21 1 Cell cycle control, mitosis and meiosis

[Y] 0 0 Nuclear structure

[V] 47 2 Defense mechanisms

[T] 65 3 Signal transduction mechanisms

[M] 66 3 Cell wall/membrane biogenesis

[N] 11 1 Cell motility

[Z] 0 0 Cytoskeleton

[W] 7 0 Extracellular structures

[U] 22 1 Intracellular trafficking and secretion

[O] 52 3 Posttanslational modification, protein turnover,chaperones [X] 17 1 Mobilome: prophages, transposons

[C] 62 3 Energy production and conversion

[G] 196 10 Carbohydrate transport and metabolism

[E] 146 7 Amino acid transport and metabolism

[F] 60 3 Nucleotide transport and metabolism

[H] 68 3 Coenzyme transport and metabolism

[I] 53 3 Lipid transport and metabolism

[P] 54 3 Inorganic ion transport and metabolism

[Q] 18 1 Secondary metabolites biosynthesis, transport and catabolism [R] 140 7 General function prediction only

[S] 64 3 Function unknown

_ 608 31 Not in COGs

359

360 361 Table 2. Number of genes associated with the 25 general COG functional category

Attribute Value % of totala Genome size (bp) 2176737 100

Coding region (bp) 1974535 91

G+C content (bp) 1481945 68

Total genes 2002 100

RNA genes 58 3 Protein-coding genes 1944 100

Protein associated to function prediction 1490 77

Protein associated to COGs 1336 69

Protein with peptide signals 171 9 Protein with transmembrane helices 453 23 362

363

364

365

366

367

368

369

370

371

372

373 374 Table 3. Cellular fatty acid composition (%)

Mean relative % (a) Fatty acid IUPAC name

16 :0 Hexadecanoic acid 42.6 ± 0.4

18 :ln9 9-Octadecenoic acid 34.5 ± 1.8

18 :0 Octadecanoic acid 7.9 ± 0.8

14 :0 Tetradecanoic acid 5.6 ± 0.4

18 :2n6 9,12-Octadecadienoic acid 3.1 ± 1.1

16 :ln7 9-Hexadecenoic acid 2.4 ± 0.6

15 :0 Pentadecanoic acid 1.7 ± 0.2

17 :0 anteiso 14-methyl-Hexadecanoic acid <1%

17 :ln8 9-Heptadecenoic acid <1%

Heptadecanoic acid 17 :0 <1%

375

Olsenella uli ATCC 49627 (NR036820)

100 Olsenella uli ATCC 49627 (NR115110) Olsenella uli DSM 7084 (NR116937) 37 Olsenella uli DSM 7084(NR074414) 77 Olsenella umbonata lac31(NR116936)

70 Olsenella profusa DSM 13989 (NR116938) 100 Olsenella profusa D315A-29 (NR036821)

55 Olsenella Timonensis P2300 (LT161892) Olegusella massiliensis KHD7 (NR146815) Atopobium minutum DSM 20586 (NR116940) 78 100 Atopobium parvulum DSM 20469(NR102936) 98 Atopobium parvulum JCM 10300(NR113159) 100 Atopobium rimae JCM 10299 (NR113038) 100 Atopobium rimae ATCC 49626 (NR036819) Olsenella scatoligenes SK9K4 (NR134781) Cryptobacterium curtum (AB019260)

0.01

Conclusions et perspectives

Ce travail a permis de montrer que l’utilisation de l’approche du repositionnement des médicaments, pourrait constituer une solution rapide et efficace pour contourner et prévenir l’émergence de la résistance aux antimicrobiens. Plusieurs molécules médicamenteuses possèdent des activités pharmacologiques, autres que celle pour laquelle elles ont été initialement mises sur le marché, ce qui favorise l’élargissement de leur champ de prescription.

En revanche, cette théorie peut se heurter à la notion de la prescription « off-label » qui reste légale mais engage la responsabilité du médecin prescripteur et l’oblige à justifier son choix de prescription par des études scientifiques d’efficacité et de sécurité. C’est pourquoi, outre le criblage des 1280 molécules médicamenteuses effectué sur les agents fongiques communément responsables d’infections invasives, on a focalisé notre travail sur un certain nombre de « hits » afin d’étudier de manière plus approfondie, leur efficacité in vitro.

Par ce travail, nous avons réussi à enrichir le panel d’antifongiques qui pourraient être utilisables pour la prise en charge des mycoses, grâce aux différents composés antifongiques

«non-standard» identifiés par le criblage primitif parmi les molécules approuvées par la FDA.

Nous avons aussi, démontré l’efficacité, in vitro, d’un antiviral ; la ribavirine en monothérapie ou en association aux antifongiques usuels, à savoir l’amphotéricine B, le fluconazole et le voriconazole, sur plusieurs espèces de Candida mais surtout sur les souches de Candida albicans multi-résistantes qui constituent un vrai challenge thérapeutique.

La colistine, un antibiotique utilisé pour le traitement des infections bactériennes invasives, pourrait trouver son efficacité dans la prise en charge des infections induites par des champignons hautement pathogènes, telles que la fusariose et les mucormycoses pulmonaires.

156

L’association d’infections bactériennes et fongiques est l’apanage des patients immunodéprimés, en raison des multiples facteurs prédisposants rencontrés chez cette population. En plus de son activité antibactérienne, la colistine présente un large spectre d’activité antifongique sur les agents mycosiques émergents et pourrait constituer une voie alternative efficace pour la prise en charge des patients présentant une co-infection bactérienne et fongique.

En laboratoire de mycologie, il est recommandé de réaliser un test de sensibilité aux antifongiques devant une infection fongique invasive, avant d’établir la conduite thérapeutique

à tenir. Les agents étiologiques sont souvent résistants à la plupart des antifongiques testés, in vitro, c’est le cas du Fusarium et du Rhizopus. Ceci induit souvent un retard de prise en charge et oblige la mise en route d’un traitement souvent probabiliste. C’est pourquoi, il serait intéressant d’automatiser la réalisation d’un criblage de chimiothèque médicamenteuse sur les champignons invasifs isolés en milieu hospitalier surtout, face à des cas d’infections systémiques difficiles à traiter.

En raison du lien étroit entre les champignons et l’hôte humain, il existe peu de cibles spécifiques aux antifongiques actuels, ainsi l’une des principales perspectives de notre étude serait d’identifier de nouveaux mécanismes d’action des antifongiques repositionnés. L’objectif de cette démarche serait de mettre au point de nouvelles cibles thérapeutiques susceptibles d’être utilisées par l’industrie pharmaceutique d’une part, et d’une autre part d’apporter une meilleure compréhension du mode d’action des composés identifiés afin d’éviter la survenue d’interactions médicamenteuses et d’effets indésirables. Dans ce contexte, l’action de la ribavirine sur deux enzymes angulaires de la viabilité fongique (IMPDH, ARN polymérase) est

157 en cours d’étude. On a réussi à désigner un système PCR et séquençage de ces enzymes et l’analyse comparative des séquences nucléotidiques pourrait nous aider à avoir une réponse à la question initiale qu’on s’était posé : la ribavirine, a-t-elle les mêmes cibles chez le Candida albicans que celles décrites chez le virus de l’hépatite C ?

Il est vrai que la réutilisation de médicaments approuvés par la FDA permet de contourner les différentes étapes de développement préclinique, car les composés utilisés ont déjà été qualifiés sur le plan de la pharmacologie, la pharmacocinétique et la toxicologie. Néanmoins, les processus réglementaires pour ajouter une indication thérapeutique à un médicament déjà sur le marché sont certes plus simples que pour une nouvelle thérapie mais au-delà des tests d’efficacité in vitro, il y a tout de même un certain nombre d’étapes à suivre notamment la réalisation de tests in vivo des différentes drogues repositionnées, sur un modèle animal afin d’approuver son activité non-traditionnelle nouvelle.

À la suite de cette étude, les scientifiques et les personnels de santé pourraient être rassurés, quant à leur inquiétude face à l’émergence de la résistance aux agents antimicrobiens, car on a montré qu’en dehors des thérapies conventionnelles couramment utilisées, des milliers d’autres molécules pourraient être exploitées. Ceci est valable en mycologie médicale mais aussi en bactériologie, parasitologie et dans la prise en charge des maladies rares et invasives.

158

Activités scientifiques

➢ Hanane Yousfi, Stéphane Ranque, Jean-Marc Rolain and Fadi Bittar. In vitro polymyxin activity against clinical multidrug-resistant fungi.

Poster présenté : - A la journée Infectiopôle Sud. IHU Méditerranée Infection. Marseille.

Juillet 2017

-A la 26 ème assemblée annuelle de l'école doctorale. Faculté de Médecine.

Marseille. Mai 2018

- Au Congrès Européen de la Microbiologie Clinique et des Maladies

Infectieuses (ECCMID). Amsterdam. Avril 2019

➢ Hanane Yousfi, Stéphane Ranque, Jean-Marc Rolain and Fadi Bittar. Innovative therapeutic alternatives to overcome antifungal drug resistance.

Poster présenté : -A la journée de la recherche laboratoire de microbiologie. Faculté de

Pharmacie. Marseille. Octobre 2018.

➢ Hanane Yousfi, Carole Cassagne, Stéphane Ranque, Jean-Marc Rolain and Fadi Bittar.

Repurposing of ribavirin as an adjunct therapy for invasive fungal disease.

Poster présenté : - A la journée Infectiopôle Sud. IHU Méditerranée Infection. Marseille.

Juillet 2018

-Au Congrès Européen de la Microbiologie Clinique et des Maladies

Infectieuses (ECCMI). Amsterdam. Avril 2019.

159

Références

1. J. Geddes-Mcalister, R.S. Shapiro, New pathogens, new tricks: Emerging,

drugresistant fungal pathogens and future prospects for antifungal therapeutics, Ann.

N. Y.

Acad. Sci. 1435 (2018) 57–78. doi:10.1111/nyas.13739.

2. R. Gupta, N. Goel, A. Gupta, K.B. Gupta, U. Chaudhary, S. Sood, A rare fungal

infiltration of lungs in a healthy young girl., Case Rep. Pulmonol. 2011.

doi:10.1155/2011/917089.

3. B.E. de Pauw, What are fungal infections?, Mediterr. J. Hematol. Infect. Dis. 3 (2011)

DOI 10.4084/MJHID.2011.001.

4. M.A. Pfaller, Antifungal Drug Resistance : Mechanisms , Epidemiology , and

Consequences for Treatment, Am. J. Med. 125 (2012) S3–S13.

doi:10.1016/j.amjmed.2011.11.001.

5. S.J.G. Matthew C. Fisher, Nichola J. Hawkins, Dominique Sanglard, Worldwide

emergence of resistance to antifungal drugs challenges human health and food

security, Science . 2018. 360:739–742.

6. Chigaev A, Hudson L, Thompson T, Young SM, Strouse J, Hjelle B, et al. Drug

repurposing from an academic perspective. Drug Discov Today Ther Strateg.

2011;8:61–9.

7. Pushpakom S, Iorio F, Eyers PA, Escott KJ, Hopper S, Wells A, et al. Drug

repurposing: progress, challenges and recommendations. Nat Rev Drug Discov.

Nature Publishing Group; 2018;

8. Butts A, Krysan DJ. Antifungal Drug Discovery: Something Old and Something New.

PLoS Pathog. 2012;8(9):9–11.

160

9. Kontoyiannis DP. Antifungal resistance: An emerging reality and a global challenge. J

Infect Dis. 2017;216(Suppl 3):S431–5.

10. Thomas E, Ghany MG, Liang TJ. The Application and Mechanism of Action of

Ribavirin in Therapy of Hepatitis C. Antivir Chem Chemother. 2012;23(1):1–12.

11. Jeannette Guarner. Human immunodeficiency virus and fungal infections. Semin

Diagn Pathol. 2017;34(4):325–31.

12. François IEJA, Cammue BPA, Borgers M, Ausma J, Dispersyn GD, Thevissen K.

Azoles: Mode of antifungal action and resistance development. Effect of miconazole

on endogenous reactive oxygen species production in Candida albicans. Antiinfect

Agents Med Chem. 2006;5(1):3–13.

13. Capoci IRG, Faria DR, Sakita KM, Rodrigues-Vendramini FAV, Bonfim-Mendonça P

de S, Becker TCA, et al. Repurposing approach identifies new treatment options for

invasive fungal disease. Bioorg Chem; 2019;84:87–97.

14. Roemer T, Krysan DJ. Unmet Clinical Needs , and New Approaches. Cold Spring

Harb Perspect Med. 2014;1–14.

15. Landman D, Georgescu C, Martin DA, Quale J. Polymyxins revisited. Clin Microbiol

Rev. 2008;21(3):449–65.

16. Nicholls MWN. Polymyxin sensitivity of Candida tropicalis. J Med Microbiol.

1970;3:529–38.

17. Ben-Ami R, Lewis RE, Tarrand J, Leventakos K, Kontoyiannis DP. Antifungal

activity of colistin against Mucorales species in vitro and in a murine model of

Rhizopus oryzae pulmonary infection. Antimicrob Agents Chemother. 2010;54:484–

90.

18. Bassetti M, Peghin M, Timsit JF. 2016. The current treatment landscape: Candidiasis.

J Antimicrob Chemother 71:ii13-ii22. 161

19. Baron S, Hadjadj L, Rolain J-M, Olaitan AO. Molecular mechanisms of polymyxin

resistance: knowns and unknowns. Int J Antimicrob Agents [Internet]. 2016

Dec;48(6):583–91

20. Liu Y-Y, Wang Y, Walsh TR, Yi L-X, Zhang R, Spencer J, et al. Emergence of

plasmid-mediated colistin resistance mechanism MCR-1 in animals and human beings

in China: a microbiological and molecular biological study. Lancet Infect Dis. 2016

Feb;16(2):161–8

162