Maliya Alia MALEK

Plague in Maghreb

Thèse de Doctorat

2016

Pathologies Humaines & Maladies Infectieuses

Faculté de Médecine de Marseille

Unité de Recherche sur les Maladies Infectieuses et Tropicales Emergentes (URMITE)

AIX-MARSEILLE UNIVERSITE

FACULTE DE MEDECINE DE MARSEILLE

ECOLE DOCTORALE DES SCIENCES DE LA VIE ET DE LA SANTE

Thèse de Doctorat Pour obtenir le grade de Docteur d’Aix-Marseille Université

Discipline : Pathologie humaine Spécialité : Maladies infectieuses

Mademoiselle Maliya Alia MALEK

La Peste au Maghreb

Soutenance le 05/07/2016 devant le jury :

Rapporteur : Monsieur le Professeur Max MAURIN Rapporteur : Monsieur le Docteur Florent SEBBANE Examinateur : Monsieur le Professeur Directeur de thèse : Monsieur le Professeur Michel DRANCOURT Co-Directeur de thèse : Monsieur le Docteur Idir BITAM

Unité de Recherche sur les Maladies Infectieuses et Tropicales Emergentes URMITE, UMR CNRS 7273, IRD 198, INSERM 1095 Faculté de Médecine, Marseille 13005, France.

Directeur : Prof Didier RAOULT

Avant Propos

Le format de présentation de cette thèse correspond à une recommandation de la spécialité Maladies Infectieuses et

Microbiologie, 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 des règles qui lui sont imposées et qui comportent un format de these 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 exhaustive 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.

Prof. Didier Raoult

Sommaire

1. Résumé………………...……………………………………….…..….... 1

2. Abstract………………………………………………………………..... 3

3. Introduction et objectifs..…………………………..………………..….. 7

4. Chapitre I : La peste en Afrique du Nord……………….……………... 17

4.1. Review : Plague in Arab Maghreb, 1940-2015 a review………..… 21

5. Chapitre II : Foyers de peste enzootique, Algérie…..…....………….… 43

5.1 Article 1 : Enzootic plague foci, Algeria……………………..……… 45

6. Chapitre III : Halo-tolérance et révélation des foyers de peste...……. 55

6.1 Article 2 : Yersinia pestis halotolerance illuminates plague reservoirs 57

7. Conclusion et perspectives...... …………………...... 87

8. Références.………………..……….………………………………..…. 89

Remerciements

Pour commencer, je voudrais exprimer mes très vifs remerciements à un génie de la recherche, mon directeur de thèse, Michel DRANCOUR, pour m’avoir intégrée à son équipe et pour m’avoir donné les moyens de parvenir au bout de ce travail de recherche et au terme de cette thèse. Il fallait une bonne dose d’intuition pour m’accepter au départ, de patience pour m’encadrer quand j’avais besoin d’une direction et de confiance pour me permettre d’apporter les idées qui ont fait de mon travail une véritable production scientifique. Je suis heureuse d’avoir côtoyé une personne qui réunit toutes ces qualités et je lui souhaite beaucoup de réussite dans la poursuite de son titanesque travail qui l’attend à l’IHU Méditerranée Infection, et auquel j’espère avoir apporté une modeste pierre.

Une profonde gratitude à mon co-directeur de thèse Idir BITAM avec qui le chemin fut long et tumultueux. Sa confiance en moi m’a rendue maîtresse de mon travail et chef des lieux de son service. Nous avons parcouru bien des terres à la recherche de cette « peste » et nous avons fini par avoir la main dessus, espérons que notre travail sera à l’origine de bien des perspectives alléchantes.

Ma reconnaissance au Professeur Didier RAOULT pour son accueil dans son prestigieux laboratoire.

Je tiens à remercier Monsieur le Professeur Max MAURIN ainsi que le Docteur Florent SEBBANE, membres de mon jury, d’avoir accepté d’évaluer minutieusement mon travail.

A Olga, Sandrine, Abdel, Claire et Maya qui m’ont soutenue ces années, un grand merci.

À la P3 team : Nathalie, Jérôme, Jean-Marc mais surtout Muriel, à qui je dois mes premiers pas dans le NSB3 et qui a été aussi bien formatrice sur paillasse que dans la vie…

A Michèle pour sa patience et sa disponibilité, sans oublier Christine et Stéphanie à l’APHM.

Mes remerciements également à Annick et son équipe de choc : les techniciennes de l’URMITE qui avec leur patience et leur confiance ont su apporter une touche inoubliable à la réalisation de ce travail. Sans oublier bien sûr LE couple de l’Unité, Marielle et Sylvain à qui je souhaite une vie heureuse avec la petite Amandine.

Je n’oublierai jamais les encouragements prodigués avec générosité venant du cœur de la personne par qui tout se fait et sans qui rien ne se fait : Valérie, Francine la joie de l’Unité ainsi que Micheline pour son accueil et ses conseils pour notre intégration.

J’ai beaucoup profité des discussions abordées avec Claude, que ce soit pour m’apporter une précieuse aide face à un problème à résoudre ou bien que nous partagions notre intérêt pour les merveilles de la nature.

J’adresse également de chaleureux remerciements à ceux qui ont ajouté leurs conseils, leur expérience, leur bonne volonté et leur bonne humeur à mon chemin durant ces quatre années. Les doctorants devenus docteurs, en particulier Shady, Rita, mais aussi Joanne et Sophia qui ont été ma famille à Marseille ! À tous une excellente continuation ! Ainsi que tous ceux que j’oublierai de mentionner et qui ont garni mon chemin de ces moments forts, propres à égayer le temps de la recherche.

Un immense remerciement à madame Yasmina Benazzoug une femme que j’admire et sans qui tout cela n’aurai jamais été possible.

Je suis heureuse d’avoir pu explorer des territoires aussi éloignés dans mon pays d’origine qu’est l’Algérie et espère dans l’avenir avoir la chance de découvrir d’autres merveilleux horizons dans un autre pays encore plus grand et avec des paysages encore plus diverses, le Brésil, mon pays natal.

Et pour finir, un grand saut vers l’extérieur…

Ma profonde gratitude

A mes parents, les plus merveilleux que l’on puisse avoir… merci pour l’éducation pleine d’Amour que vous m’avez inculquée, grâce à vous j’ai hérité d’un grand cœur comme le vôtre qui me mènera très loin. Merci pour votre confiance qui m’a permis de me propulser dans la vie. Sans vous, je n’aurai pas pu concrétiser ces rêves, c’est de vous que me vient l’ambition de dépasser les limites.

A Hannah mon adorable sœur qui a toujours su trouver les mots adéquats pour révéler mes capacités et me soutenir dans les moments les plus durs.

A mes frères Raouf, Lies, Chakib et Arslan, les 4 mousquetaires… qui m’ont depuis mon jeune âge, chouchoutée, dorlotée mais surtout soutenue et encouragée.

A mes belles-sœurs qui sont de précieuses sœurs…

A ma joie, mon bonheur... mon neveu Fares et mes nièces Ranya, Amina, Chérifa et Baya qui par leur Amour, leur innocence, leur tendresse redonnent goût à la vie.

Et enfin à toi, qui est mon présent et sera mon avenir, toi qui me renforce et me complète

mon Mehrez … 1. Résumé

Yersinia pestis est l'agent causal de la peste, une maladie à transmission vectorielle enzootique infectant les rongeurs et leurs puces, l’Homme étant un hôte accidentel.

Y. pestis est classée par les Centers for Diseases Control comme un agent potentiel de bioterrorisme [http://www.bt.cd.gov/agent/plague]. Ce bacille persiste toujours dans la nature maintenu par un cycle enzootique dans des foyers de peste, ce qui conduit à la réémergence de la maladie en cas de proximité des populations humaines et ce jusqu’à plusieurs décennies plus tard. Il est donc important pour la communauté médicale de connaitre ces foyers afin d’établir le bon diagnostic en présence d’un cas de peste, prescrire le traitement efficace et surtout limiter la propagation au reste de la population. Dans notre travail nous nous sommes intéressés aux pays d’Afrique du Nord où une réémergence a eu lieu après des années de ‘silence’ et ce à trois reprises dans les années 2000. Nous avons, dans un premier temps, répertorié dans une revue les différents épisodes ayant eu lieu ainsi que le nombre de cas sur six pays du Maghreb:

Mauritanie, Maroc, Algérie, Tunisie, Libye et Egypte sur une période de 75 années

à compter de 1940 en mettant en évidence l’importation de la maladie dans ces pays puis un mode de contamination négligé dans la littérature, à savoir la transmission par voie orale de Y. pestis. Cette revue nous a conduits à travailler ensuite en Algérie pour mettre en évidence des foyers de peste encore inconnus et de confirmer l’existence du risque dans ceux déjà décrits. Nous y avons mené une étude rodontologique sur 237 micromammifères provenant de 11 régions du Nord

Algérien et avons ainsi confirmé deux foyers (Mascara, Laghouat) et mis en

Résumé |1 évidence trois nouveaux foyers (M’Sila, Biskra et Cap Djinet) porteurs d’un nouveau génotype (déterminé par Multi Spacer Typing) de Y. pestis de biotype Orientalis. Ce même travail nous a permis d’ajouter l’espèce Apodemus sylvaticus à la liste des rongeurs pestiférés.

La projection des foyers de peste ainsi actualisés sur une carte géographique et

écologique du Maghreb nous a conduits à observer que les foyers de peste étaient situés à proximité de points d’eau saumatre. Une étude statistique a confirmé une corrélation significative entre foyer de peste/eau salée en révélant une proximité minimale <3 km en comparaison à des zones d’eau douce. Sur cette base, une nouvelle campagne de prélèvements d’échantillons environnementaux salés en

Algérie a permis l’isolement d’une souche Y. pestis Algeria 3 d’un sol salé. Cette découverte a été confortée par l’observation expérimentale de la résistance de Y. pestis à un milieu hyper-salé à 150g/L NaCl se traduisant par un protéome spécifique en réponse à ce stress et mettant en évidence une forme d’adaptation de type forme L de la bactérie dans ce type d’environnement inanimé. Notre travail éclaire de façon originale un facteur méconnu de persistance tellurique de Y. pestis, conditionnant la réémergence de la peste dans des foyers séculaires au

Maghreb contrairement aux rivages Nord de la Méditerranée où la peste autochtone a disparu depuis un siècle.

Mots clés : Yersinia pestis, peste sylvatique, foyers, surveillance, persistence,

Afrique du Nord.

Résumé |2 2. Abstract:

Yersinia pestis is the causative agent of plague, an endemic vector-borne disease infecting rodents and their fleas, man is an accidental host. Y. pestis is classified by the Centers for Disease Control as a potential agent of bioterrorism

[http://www.bt.cd.gov/agent/plague]. This bacillus persists in nature held by an enzootic cycle of plague outbreaks, which led to the re-emergence of the disease in case of near human populations until several decades later. It is therefore important for the medical community to know these foci in order to establish the correct diagnosis in the presence of plague cases, prescribe effective treatment and especially limit the spread to the rest of the population.

In our work we focus on North African countries where re-emergence occurred after years of 'silence' and three times in the 2000s. We have initially, listed in a review different episodes that took place and the number of cases in six North African countries: Mauritania, Morocco, Algeria, Tunisia, Libya and Egypt over a period of

75 years from 1940 highlighting the importation of the disease in those countries and a way of contamination neglected in the literature, the oral transmission of Y. pestis.

This review has led us to work in Algeria to highlight unknown fever outbreaks and confirm the existence of risk in those already described. We have conducted a rodontologic study on 237 mammals from 11 regions of North Algeria and thus have

Abstract | 3 confirmed two outbreaks (Mascara, Laghouat) and demonstrated three new outbreaks (M'Sila, Biskra Cap Djinet) holders of a new genotype (Multi Spacer

Typing) of Y. pestis Orientalis biotype. The same work has enabled us to add

Apodemus sylvaticus species to the list of rodents plague.

The projection of updated plague foci on a geographical and ecological map of the

Maghreb led us to observe that plague foci were located near the brackish water points. A statistical study confirmed a significant correlation between plague focus / saltwater revealing minimal proximity <3 km in comparison to freshwater areas. On this basis, a new campaign of salty environmental sampling in

Algeria has allowed the isolation of a Y. pestis strain Algeria 3 from a saline soil.

This discovery was confirmed by experimental observation of the resistance of

Y. pestis in a hyper-saline environment till 150g / L NaCl resulting in a specific proteome in response to stress and highlighting the L-form as a type form of adaptation of the bacteria in such inanimate environment.

Our work in an original way illuminates an unknown factor telluric persistence of

Y. pestis, conditioning the re-emergence of plague in secular homes in the

Maghreb unlike Northern shores of the Mediterranean where the indigenous plague disappeared for a century.

Key-words: Yersinia pestis, sylvatic plague, foci, surveillance, persistence, North

Africa.

Abstract | 4

Introduction & objectifs

3. Introduction et objectifs.

La peste, ce fléau synonyme de terreur, est l’un des maux de l’humanité le plus connu. Etymologiquement, le mot peste dérive latin « pestis » qui désigne une maladie contagieuse associée à la destruction. Cette maladie mortelle persiste actuellement dans des foyers de peste dans de nombreux pays en Afrique, en Asie et en Amérique après que plusieurs épidémies ont été rapportées et ce depuis la Bible. La première épidémie documentée est la Peste Justinienne qui a débuté vers 532, se propageant depuis l’Egypte vers le Moyen Orient, traversant le Bassin Méditerranéen pour atteindre la Turquie (Biraben, 1975). La seconde épidémie, la mieux décrite et la plus dévastatrice, la Peste Médiévale est apparue en 1334 en Chine ensuite propagée vers l'ouest le long des grandes routes commerciales vers la mer Noire, puis vers les Indes et la Crimée pour ensuite atteindre l’Europe où elle décima plus d’un tiers de la population (Gottfried, 1983). La troisième pandémie débuta dans la province du Yunnan au cours de la seconde moitié du XIXe siècle. C’est au cours de cette épidémie qui atteint Hong Kong en 1894 que l’agent causal a été initialement isolé par le bactériologiste Franco-Suise Alexandre Yersin, élève de Louis Pasteur (Yersin, 1894). Yersin a isolé le micro-organisme responsable de la peste à partir d’un extrait de ganglion infecté, dénommé bubon. Suite à la mise en culture d’un extrait de bubon, Yersin a pu identifier des colonies blanches, qui se sont avérées être des coccobacilles à Gram négatif à coloration bipolaire strictement immobiles. Inoculés aux souris, ces bacilles tuaient les animaux en 1 à 3 jours. Yersin a pu par la suite élaborer un sérum antipesteux à partir du sang de chevaux (Yersin, 1897) puis un vaccin à partir d’une souche atténuée a été mis au point par Waldemar Haffkin (Haffkin, 1906). Le bacille fut initialement dénommé Pasteurella pestis

Introduction et objectifs |7 mais, en 1962, il fût renommé Yersinia pestis en hommage à Alexandre Yersin (van Loghem, 1944). La peste est décrite comme étant une zoonose essentiellement véhiculée par les micromammifères et plus précisément les rongeurs (chiens de prairie, marmottes, rats, écureuils…) (Pollitzer, 1954). En 1898, le biologiste médecin Français Paul-Louis SIMOND démontre que les puces sont vectrices de la bactérie (Simond, 1998). En 1904, Bacot démontre le mécanisme de transmission de la peste par les puces : suite à un repas sanguin sur un hôte infecté, les bactéries ingérées colonisent l’intestin moyen et produisent un biofilm dont la taille évite l’excrétion des bactéries avec les fèces et dont la croissance finit par obstruer la partie supérieure du tube digestif, le proventricule; la puce est qualifiée de bloquée. Ainsi, lors d’un repas sanguin d’une puce bloquée, le sang aspiré est stoppé au contact de l’amas bactérien, se contamine puis est régurgité au point de piqûre (Bacot, 1915 ; Bacot, 1914).

Au cours de la troisième pandémie, la peste se propagea en Afrique, en Australie, en Europe, au Moyen Orient, en Amérique du Nord et en Amérique du Sud (Butler, 1983). Actuellement, la peste n’est pas aussi dévastatrice qu’au cours des deux pandémies historiques, mais l’existence de foyers fait en sorte que la maladie persiste encore. Il est donc indispensable de tenter de prévenir sa ré-émergence en définissant les facteurs impliqués dans le cycle pour éviter d’autres épidémies. Les modèles actuels de l'écologie de la peste sont complexes, impliquant de nombreuses espèces de rongeurs et de plus de 80 espèces de puces à partir desquelles l’agent pesteux a été isolé dont Xenopsylla cheopis, la puce du rat, qui serait le principal vecteur de Y. pestis (Burroughs, 1947). Une basse température, un taux d’hygrométrie élevé ainsi qu’un surpeuplement du réservoir sont des facteurs qui conduisent à une propagation vectorielle

8| Introduction et objectifs accrue de Y. pestis (Pollitzer, 1954). Les travaux de Baltazard et Mollaret en Iran ont montré que Y. pestis peut survivre en restant virulente plusieurs années dans les cadavres et le terrier des animaux morts, jusqu’à la recolonisation des terriers par de nouveaux individus (Mollaret, 1963a; Mollaret, 1963b). Trois autres modes de transmission sont décrits pour la peste: contact direct, inhalation et ingestion. Ce dernier a été rarement étudié expérimentalement (Butler, 1982) et rapporté dans la littérature au cours de quelques épidémies au cours desquelles les camélidés et les chèvres, animaux sensibles à la peste, étaient la source d’une peste peut-être d’origine alimentaire pour les populations en contacts avec ces animaux (Christie, 1980). La peste présente une pathogénicité exceptionnelle pour les humains chez qui la mortalité est très élevée en l’absence de traitement. Selon le mode de transmission on distingue différentes formes cliniques parmi lesquelles la plus observée est la peste bubonique faisant suite à une inoculation de Y. pestis, essentiellement par piqûre de puce. Les bactéries migrent par les vaisseaux lymphatiques vers le premier relai ganglionnaire lymphatique. L'infection implique généralement les ganglions lymphatiques de l'aine, des aisselles et du cou. Les bactéries se répliquent dans le ganglion lymphatique qui devient cliniquement inflammatoire et douloureux (le bubon). La peste bubonique non traitée présente une mortalité de 30-75%, qui reste très élevée à 4-15% sous traitement antibiotique (Inglesby, 2000).

Au cours de la peste septicémique, l’infection se propage dans le sang et un choc septique en découle. Cela peut soit suivre une piqûre de puce par progression de Y. pestis dans la circulation sanguine (barrière lymphatique inefficace contre le pathogène) ou par le contact direct sur une peau lésée (Doll, 1994; Pollitzer, 1954). Les patients peuvent avoir des saignements de la peau et des muqueuses et des hémorragies dans les

Introduction et objectifs |9 organes dues à une coagulation intra-vasculaire disséminée, entrainant une ischémie vasculaire et des nécroses gangreneuses. Les symptômes peuvent apparaître le jour même et les patients peuvent mourir dans les 24 heures en l’absence de traitement. La peste pneumonique est la présentation la moins commune mais la seule contagieuse, présentant une mortalité très élevée (Inglesby, 2000). L’infection pulmonaire peut être primaire par l'inhalation de gouttelettes ou secondaire à une septicémie (Pollitzer, 1954). L'infection se présente comme une maladie broncho pneumonique avec une douleur thoracique, une toux, un essoufflement et une hémoptysie. Les complications comprennent coagulation intra-vasculaire disséminée ainsi qu’un syndrome de détresse respiratoire aiguë évoluant ainsi rapidement vers une septicémie. La peste pneumonique et la peste septicémique ne présentant pas d’adénite, peuvent être de diagnostic retardé voire rétrospectif lorsque les cas ne font pas partie d’une épidémie ou sont liés à un autre contexte épidémiologique spécifique (Raoult, 2013).

Les mesures prophylactiques et thérapeutiques ont permis de réduire considérablement la mortalité et de limiter la contagion Les sulfonamides ont été les premiers antibiotiques utilisés contre la peste à partir de 1938 suivi de la streptomycine en 1946 (Pollitzer, 1954).

• La streptomycine - réduit le taux de mortalité à 4-15% lorsqu'elle est administrée par voie parentérale (Butler, 1995).

• La gentamicine - est considérée comme une deuxième ligne de traitement et les taux de réussite parentérale peuvent être similaires à la streptomycine (Coppes, 1980).

10| Introduction et objectifs • Les fluoroquinolones - aussi efficaces que les aminoglycosides et les tétracyclines au cours d’expérimentations induites et peuvent être administrés par voie orale.

• Le chloramphénicol - il traverse la barrière hémato-encéphalique, mais cela n'a pas été confirmé et le médicament est associé à une toxicité.

En prophylaxie seront utilisées la tétracycline (sauf chez l'enfant) et la doxycycline (Smith, 1995). La doxycycline en absence de gentamicine peut être administrée chez les femmes enceintes dans le cadre d’une peste pneumonique (Czeizel, 1997).

Ces mesures prophylactiques doivent être utilisées dans les groupes suivants (Inglesby, 2000 ; Perry, 1997):

• Les personnes piquées par les puces pendant une épidémie.

• Les personnes exposées à des tissus ou liquides provenant d'animaux infectés par la peste.

• Les personnes vivant dans une maison où un patient a développé la peste bubonique.

• Les personnes en contact étroit avec les personnes soupçonnées de peste pneumonique.

Quelques souches de Y. pestis ont été décrites comme étant résistantes aux antibiotiques. En 1995, une souche Y. pestis 17/95 isolée d’un patient Malgache présentait une multi-résistance portée par un plasmide fréquemment rencontré chez les entérobactéries (Galimand, 1997). Une autre souche Y. pestis 16/95 isolée également à Madagascar à partir d’ un autre patient présentait une résistance à la streptomycine (Guiyoule,

Introduction et objectifs |11 2001). En Mongolie, une autre multi-résistance a été décrite sur une souche Y. pestis MNG3122 qui ne présentait une sensibilité qu’à la ciprofloxacine. Contrairement aux souches Malgaches, elle a été isolée à partir d’un rongeur (Kiefer, 2012).

Y. pestis est capable de croître sur une large gamme de température et de pH. Elle cultive dans un milieu incubé de 4 à 40°C et dont le pH peut varier entre 5,0 et 9,6 (Brubaker, 1972). Cependant, la bactérie croît de manière optimale à 28°C dans un milieu tamponné à pH 7,5. Y. pestis se caractérise par différents critères biochimiques et certains de ces critères (capacité à réduire les nitrates et à fermenter le glycérol) ont été utilisés pour regrouper différentes souches en trois biovars principaux: Antiqua, Médiévalis et Orientalis (Devignat, 1951). Les souches du biovar Antiqua sont capables de réduire les nitrates et de fermenter le glycérol. Les souches du biovar Médiévalis réduisent uniquement les nitrates alors que celles du biovar Orientalis ne sont capables de ne fermenter que le glycérol. Malgré ces divergences, les trois biovars sont aussi bien virulents chez les animaux que les humains (Brubaker, 1972).

Sur le plan évolutif, Y. pestis aurait émergé il y a environ 1,500- 20,000 ans de Y. pseudotuberculosis un entéropathogène responsable, le plus souvent, de maladie bénigne du tube digestif chez l’Homme (Achtman, 1999). Malgré les différences d’expression du pouvoir pathogène entre ces deux pathogènes, leurs chromosomes partagent 97% d’identité nucléotidique. Par ailleurs, tous deux possèdent un plasmide de virulence de 75 kilobases (Michiels, 1990). Ce plasmide est essentiel à leur virulence (Michiels, 1990). C’est pourquoi, il a été dénommé pYV pour plasmide Yersinia virulence. Il porte une trentaine de gènes codant un appareil de sécrétion de type III et des exotoxines Yops dénommées « Yersinia outer

12| Introduction et objectifs membrane proteins » (Michiels, 1990). Cet appareil permet à la bactérie, suite à son contact avec les phagocytes, d’injecter les toxines Yop dans le cytoplasme qui inhibent la phagocytose et la réponse inflammatoire et induisent la production de cytokines anti-inflammatoires et l’apoptose.

Le génome de Y. pestis se distingue cependant de celui Y. pseudotuberculosis par une perte importante de matériel génétique et l’inactivation des séquences codantes et l’accumulation de séquences d’insertion (Chain, 2004). La perte de matériel génétique fonctionnel (uréase, phospho-estérase) a joué un rôle essentiel dans la capacité de la bactérie à produire une infection transmissible dans la puce (Sun, 2014). Le génome de Y. pestis se distingue aussi de celui de l’ancêtre commun par la présence de deux plasmides pFra et le pPla qui jouent tous deux un rôle essentiel dans la peste (Sodeinde, 1992 ; Sebbane, 2006 ; Sebbane2001).

Avec une taille de 9.5 kb le plasmide pPla (ou pPCP1 : pesticine, coagulase, activateur du plasminogène) est spécifique à Y. pestis. La protéase Pla est essentielle pour la dissémination de Y. pestis depuis le point de piqûre cutanée vers les tissus profonds (sang, foie et rate), (Sodeinde, 1992 ; Sebbane, 2006). La protéase Pla est décrite comme étant impliquée dans la dégradation du ligand Fas-L de la protéine transmembranaire Fas, médiateur de l’apoptose et de la réponse inflammatoire chez la cellule hôte, afin de faciliter la colonisation des poumons par Y. pestis (Caulfield, 2014).

Le plasmide pFra (ou pMT1) porte le gène ymt (Yersinia murine toxine) qui code une toxine murine de type phospholipase D (PLD) protégeant la bactérie des facteurs antibactériens de la puce (Hinnebusch, 2002). Il porte également l’opéron caf1MA1 gouvernant la synthèse d’une pseudo-capsule qui confère une résistance à la phagocytose in vitro (Du,

Introduction et objectifs |13 2002). La sous-unité Caf1 interfère également avec la voie de signalisation de l’Interleukine 1 (Abramov, 2002). Le rôle de l’opéron caf dans la virulence bactérienne dépend du fond génétique de l’hôte (Sebbane, 2009) mais aussi du mode d’inoculation ; l’opéron est important chez les souris, son absence diminue l'incidence de la peste bubonique après une piqûre de puce mais pas suite à l’inoculation intradermique de culture bactérienne.

Au cours de son émergence, Y. pestis aurait également acquis un phage filamenteux YpfФ (Chouikha, 2010). Au cours de la microévolution du bacille de la peste, le phage serait resté dans les diverses lignées en tant qu'élément extra chromosomique stable jusqu’à intégrer le chromosome bactérien pour devenir un prophage stable dans le biovar Orientalis. Des mesures de virulence chez la souris suggèrent que le phage participerait à la virulence bactérienne, mais en l’absence d’expérience de complémentation, son rôle exact demeure méconnu (Chouikha, 2010).

La survie de Y. pestis dans le mésogastre des puces dépend de l'activité de la phospholipase D (Hinnebusch, 2002). Les mutants dépourvus de phospholipase D subissent une perte d'intégrité de la membrane externe, et disparaissent rapidement à l'intérieur de l'intestin moyen des puces un jour après l'infection. La toxine Ymt constituerait une protection de Y. pestis contre un agent bactériolytique généré dans le mésogastre des puces lors de la digestion du repas sanguin. Fait intéressant, la transformation de Y. pseudotuberculosis et E. coli avec le gène ymt de Y. pestis augmente considérablement leur capacité à infecter l'intestin moyen des puces (Hinnebusch, 2002).

Les modifications génétiques décrites jusqu’à présent permettent donc d’expliquer un nouveau type de virulence, de nouvelles voies de

14| Introduction et objectifs contamination et de portage du genre Yersinia à l’origine de l’existence de la peste par Y. pestis. Cependant les facteurs à l’origine de sa survie dans des environnements inanimés ainsi que sa ré-émergence périodique dans des foyers circonscrits ne sont pas encore clairement définis. Nous avons étudié la peste au Maghreb en commençant par revue de la littérature reprenant l’ensemble des épidémies décrites au Maghreb depuis 1940, date des derniers rapports complets effectués dans cette région de l’Afrique. Sur la base de cette revue, nous avons actualisé sur le terrain, les foyers de peste en Algérie où nous avons découvert de nouveaux foyers. Enfin, nous avons découvert que Y. pestis était une bactérie halotolérante dans des foyers de peste tellurique au Maghreb. Ces découvertes permettent de mieux rendre compte de l’établissement de foyers de peste dans l’hémisphère Nord

Introduction et objectifs |15

Chapitre I

4. Chapitre I : La peste en Afrique du Nord

4.1. Plague in Arab Maghreb, 1940-2015 : A Review

Historiquement en Afrique du Nord, la peste frappe la ville d’Oran en 1542 avec une férocité telle que d’après Haedo, les espagnols se trouvent obligés de quitter la ville. En 1552 elle atteint Alger, où elle est décrite jusqu’après

1584, avec des recrudescences en 1555, 1559, 1561, 1565, 1571 et 1584. Le taux de mortalité était si élevé que le tiers de la population fut anéanti en

1572. En 1582, la peste frappe Constantine avec une mortalité telles que les dates des faits historiques notables de cette période sont indiqués par

« l’année de la peste ». Elle réapparait à Alger en 1601, 1605, 1613, 1620 et en 1620 et accompagnée de famine en 1640 et en 1647. En 1650 la peste passe d’Alger pour atteindre Biskra ; elle ravage les oasis des Zibans et le l’O.Rhir, où 70.000 décès sont dénombrés dans cette région selon El Aïchi.

Des épidémies marquèrent les années 1654, 1659, 1660, 1663, 1675, 1676 et 1678 à Alger. Ensuite, la peste est signalée à Tunis, à Bougie, à Oran et à

Constantine où elle a fait 500 morts par jour. Elle y dura jusqu’en 1700. Et ce n’est qu’en 1731 que la terrible peste de Marseille de 1720 atteint Alger, elle se propage jusqu’à atteindre Tlemcen en 1738. En 1749 elle réapparait sur Alger. Constantine fut touchée en 1754 et Collo en 1758. Entre 1783 et

Revue de Littérature |17 1788, la peste est rapportée dans la province de Constantine puis s’étend à

Tunis en 1784, à Tabarka et à la Calle. En 1792 des corsaires arrivent de

Constantinople et infectent Alger où la peste persiste jusqu’en 1804, s’étendant à Dellys, Bône, Tunis en 1796, Constantine en 1797, Tlemcen

1798 tuant dans les Zibans jusqu’à 300 personnes par jour. Entre temps elle réapparait de nouveau à Oran, en 1794 et en 1799. En 1816 elle est réintroduite par des pèlerins à Bône puis s’étend à Constantine puis peu après est officiellement déclarée à Alger. Cette épidémie durera jusqu’en

1822 ; elle se propage sur tout le Nord de l’Afrique et ne fut pas la dernière.

Dès 1896 un ou deux cas apparaissent chaque année en Algérie jusqu’en

1904 où durant quatre ans le nombre de cas total atteignait 25, des cas qui restent cependant sporadiques qui s’étendent d’Oran à la Calle (Annaba).

En 1907 une épidémie se traduit par 57 cas puis jusqu’en 1923 des cas apparaissent tous les deux à trois ans (Raynaud1924). En 1926 elle réapparait à Oran, puis à Constantine en 1931. Les épisodes ayant suivi ont eu lieu courant la seconde guerre mondiale et la situation ne permettait pas de stopper les épidémies qui touchèrent également les forces alliées.

Dans notre revue, nous avons revu l’ensemble des cas répertoriés depuis

1940 dans chacun des six pays de l’Afrique du Nord et ce jusqu’en 2015.

Nous avons cherché quelles ont été les causes de ces épidémies et leur mode

18| Revue de Littérature de contamination. Ce travail de revue nous a permis de mettre en exergue un mode négligé de contamination par voie orale dans les populations nomades, en contact direct avec les cheptels contaminés et dont les migrations contribuaient à la propagation de la maladie.

Revue de Littérature |19

Chapitre I Plague in Arab Maghreb, 1940-2015: a review

Maliya Alia MALEK 1,2, Idir BITAM 2, Michel DRANCOURT 1 1. Faculté de Médecine, Université de la Méditerranée, France, 2. Faculté des Sciences Biologiques, Université des Sciences et de la Technologie Houari Boumediene, Algeria

Submitted to Journal : Frontiers in Medicine Specialty Section: Infectious Diseases Article type: Review Article Manuscript ID: 183223 Received: 22 December 2015; Accepted: 20 May 2016; Published: 03 June 2016 Available on line : http://journal.frontiersin.org/article/10.3389/fpubh.2016.00112 Frontiers website link: www.frontiersin.org Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest

Author contribution statement

MAM reviewed data and drafted the manuscript. IB drafted the manuscript. MD decided of the topic, reviewed data and drafted the manuscript.

Keywords : Yersinia pestis, Plague, outbreak, Foci, Maghreb Arab

Revue de Littérature |21 REVISED VERSION

Abstract

We reviewed the epidemiology of 49 plague outbreaks which resulted in about 7,612 cases in 30 localities in the Arabic Maghreb (Mauritania, Morocco, Algeria, Tunisia, Libya and Egypt) over 75 years. Between 1940 and 1950, most cases recorded in Morocco (75%) and Egypt (20%), resulted from plague imported to Mediterranean harbours and transmitted by rat ectoparasites. In contrast, the re-emergence of plague in the southern part of Western Sahara in 1953 and in northeast Libya in 1976, was traced to direct contact between nomadic populations and infected goats and camels in natural foci, including the consumption of contaminated meat, illustrating this neglected oral route of contamination. Further familial outbreaks were traced to human ectoparasite transmission. Efforts to identify the factors contributing to natural foci may guide where to focus the surveillance of sentinel animals in order to eradicate human plague, if not Y. pestis from the Arab Maghreb. Ethics statement

(Authors are required to state the ethical considerations of their study in the manuscript including for cases where the study was exempt from ethical approval procedures.)

Did the study presented in the manuscript involve human or animal subjects: No

22| Revue de Littérature

Chapitre I

Introduction Plague, a deadly zoonosis caused by the Gram-negative bacterium Yersinia pestis, historically affected the Arab Maghreb (comprising Mauritania with the Western Sahara, Morocco, Algeria, Tunisia, Libya and Egypt) over a period of at least two millennia (Raoult et al., 2013)This is a review of plague cases from 1940 until the present day in the Arab Maghreb, with a view to restricting the evolution of the disease in this region and to identify features 6differentiating its epidemiology in the Arab Maghreb from other regions of the world (Kugeler et al., 2015)

Methods Information provided in this review has been compiled from the World Health Organization’s weekly report, information issued by the Algerian Ministry of Health, and published papers retrieved from the NCBI’s Pub Med. We also reviewed the Bulletin de la Société de Pathologie Exotique published in French and communications by professors M. Baltazard and H.H. Mollaret from the Pasteur Institutes. We also consulted grey literature to supplement officially reported data regarding the distribution of plague outbreaks, foci and the movements of the affected populations. Those data were collected from testimonies from local populations. The fact that plague records were not as systematic as they were over the same period in the US, necessarily limit available data (Kugeler et al., 2015).

Revue de Littérature |23 REVISED VERSION

Results Descriptive epidemiology. Between 1940 and 2015, a total of 49 plague outbreaks resulted in 7,612 cases in 30 localities in the Arab Maghreb (supplementary video). Reports made in 1940-1945 during the Second World War (WWII) did not detail the age, sex, and mode of contamination of patients. After 1945, plague cases were more accurately reported. We distinguished two periods between 1940 and 2015 with identifiable epidemiological patterns.

First period, 1940-1950. During WWII, plague affected all the Arab Maghreb countries except for Mauritania and accounted for 6,801/7,612 (89%) cases. Morocco alone recorded approximately 5,500 (71%) cases in the regions of Chaouia (20%), Agadir (14%), Marrakech (12%), Rabat, Doukkala and Port Lyautey (1%) (Pollitzer, 1954). In April 1940, preceded by a murine outbreak, plague affected Agadir and spread to Marrakech where it caused 498 cases five months later. Two outbreaks in 1941 in inland south Morocco, suggesting they resulted from endemic foci (Blanc and Baltazard, 1945a). In 1942, 25 cases notified in Casablanca probably resulted from the train transport of grains by rail from the southern foci where 583 cases were declared. In 1943 and 1944, plague spread to Rabat and caused 393 cases in Port Lyautey and 227 cases in Marrakech, respectivelY. Early in 1944, Casablanca suffered another episode of 79 bacteriologically confirmed cases ; followed by an 828-case devastating episode which was the last to be recorded in Morocco. In Algeria, eight cases were recorded in Algiers in the early 40s, 94 cases in 1944 and 11 cases in 1945. The same year in Oran, where the French Nobel-prize winning author Albert Camus located

24| Revue de Littérature

his novel ‘La Peste’ (The Plague), six pneumonic plague cases occurred (Roux and Mercier, 1946). Another two cases in 1946 and six cases in 1950 (Pollitzer, 1954) seemed to have been imported cases except for two local cases (Signoli et al., 2007). In Tunisia, twelve cases were recorded in 1940 and one in 1941. In August 1944, a fatal index case was confirmed by blood smear examination and isolation of Y. pestis in laboratory animal at Ferryville (currently known as Menzel Bourguiba). A final 37-case outbreak included 25 European patients. In Libya, 12 confirmed and two suspected cases were noted in 1939-1943 in a locality 12 km from Tripoli (Kaul, 1949). In Egypt, 452 cases were recorded in 1940 in the province of Assiut in Upper Egypt and some sporadic cases were also reported in Port Said. In November 1943, an outbreak was observed in the Suez Canal area; and in the Ismailia district and Port Said in 1944 accounting for a total of 862 cases. Plague reappeared in February 1945 causing 19 cases in ports along the canal to gradually dwindling with two cases in 1946. In January 1947, a 15-case outbreak took place in Alexandria after a 12 years gap and was the last reported outbreak in Egypt. It was reported that during this period the plague had essentially port profile in contrast to previous outbreaks that took place in rural areas (Tomich, 1947).

Second period, 1951-2015. In Mauritania, plague increased to a cumulate effective of 3%. In January 1953, a plague outbreak may have taken place in the Río de Oro, south-western Sahara but this episode was not microbiologically documented and was not declared to the WHO (Alonso, 1971). According to the Mauritanian Ministry of Health, 74 human cases were identified between Boulanouar and Bir Quendouz after similar human cases had been observed in 1951 in Bir Moghrein, a quasi desert region in

Revue de Littérature |25 REVISED VERSION north Mauritania (Alonso, 1971). Ten years later, four cases were diagnosed at Bir Igueni, an area inhabited by 3,000 nomads (Alonso, 1971). Investigation uncovered nine deaths in the previous four months. Nine fatal cases in the Nasri area also remained unreported to the WHO. Cases, without bacteriological confirmation and medical supervision, were collected by interviewing nomads who reported the first fatal case of plague in a woman from the Rio de Oro. Other cases include another woman, an eight-year-old girl and the grave digger, who had been in close contacts with the index case. On 16 October, at Port Etienne, five cases were reported of family members with a cervical tumefaction. Of them, four died. The doctor P. Bres reported a total of 11 deaths for this episode (Alonso, 1971). Four cases were declared at the Al Mounek camp and two others in mobile camps were identified one further north, at Bir Tenchi and Aguedat Iguenine, along with sick cattle. Investigation by M. Baltazard and J.M. Alonso discovered that plague had been moving from southern Morocco to Mauritania for years (Alonso, 1971). A 23-case outbreak reported the same year in northwest Mauritania was microbiologically documented (Alonso, 1971). At the same period at Fort Coppolani, an outbreak included a dozen cases and one deceased woman presenting buboes without bacteriological identification. In Libya, after a thirty years gap, an 18-case outbreak occurred in the Nofila area in l972. In 1976, an outbreak occurred in the northeast at the village of Al-Azzizat. Near Tobruk in the village of Krom el Kheil, five family cases occurred. All recovered after early antibiotic treatment but only four were plague positive. In Al- Azzizat, one human and four goats cases were reported. In January 1977, an 11-case outbreak took place in Jadu in southwest of Tripolitania (Christie, 1980). In September 1984, eight bubonic plague cases occurred at two

26| Revue de Littérature

locations 25 km and 60 km from Tobruk at the Egyptian border. Libya declared an episode in June 2009 in Betnane, a semi-nomadic area in the coastal town of Tobruk (Tarantola et al., 2009). Another possible outbreak of plague comprising of more than 20 cases occurred at Tobruk during the Libyan revolution in May 2011 but due to the political situation of the country no scientific evidence was provided. No natural plague focus had ever been described and confirmed in Algeria until the outbreak in Kehailia and neighbouring villages, 30 km southwest of Oran on 22 June 2003 (Bitam et al., 2006). Indeed, after a 53-year gap, on 4 June the University Hospital Center of Oran received an 11-year-old boy with signs of septicaemia. A dozen adults with similar symptoms were diagnosed with bubonic plague in the area of Oran with nine urban and rural cases, one case in Mascara and two cases in Ain Temouchent (Bertherat et al., 2007). In July 2008 a new episode took place for the first time in the Laghouat area in a nomad camp in Thait El Maa with four cases (Bitam et al., 2010). This was the last episode of plague in Algeria until the present day.

Clinical features and outcomes. While penicillin G was shown to be ineffective, sulfonamides and serotherapy allowed for effective prophylaxis (Mafart et al., 2004). Streptomycin was first used in 1947 during the Haifa outbreak (Haddad and Valero, 1948). Accordingly, we estimated mortality during the first period (1940-1950) to be of 63% significantly higher than the 28% mortality during the 1951-2015 periods (P < 0.05). Clinical descriptions indicated that 87% patients had inguinal or cervical buboes, 7% had a primary pulmonary plague and 6% had a septicaemic form of plague. Mauritanian cases described in 1953, 1963 and 1967 were all sporadic cases which occurred

Revue de Littérature |27 REVISED VERSION suddenly in a limited region and were described as cervical bubonic plague (Alonso, 1971). Inhabitants observed an increase in the commensally rodent population, but no unusual rodent mortality before the outbreak. Furthermore, the appearance of cases in a location which is 100-km distant does not mean expression of different foci at the same time but the same focus because the isolates all belonged to the same strain and are classified in clusters which are different than to those of America and Africa suggesting a non-imported plague (Bertherat et al., 2007). In that region, close contacts with Meriones could be explained by lack of food which leads the rodent to get into seeds’ reserve. In the last Oranian outbreak all twelve confirmed cases presented buboes. Plague was confirmed by a rapid diagnostic test (Chanteau et al., 2000) and Y. pestis was isolated from bubo or blood in six patients. In Laghouat, four people had swollen lymph nodes, one patient who developed pulmonary form died (Bitam et al., 2010). The other three patients were treated with doxycycline, rifampicin and intramuscular gentamicin after bacteriological confirmation of plague by culturing Y. pestis from bubo aspirate. Rats or Y. pestis-susceptible rodents were absent in the nomads’ environment. In Libya, in 2009, the first case was a child with pneumonic plague, who died; two family members of the child and two women presented a bubonic form (Tarantola et al., 2009). In the Maghreb episodes, a few Y. pestis isolates have been fully characterized but it was confirmed that strains responsible for the 2003 and 2008 episodes in Algeria were of the Orientalis biotype as were four strains isolated in 1940-1945 in Morocco and Algeria (Guiyole et al., 1994; Cabanel et al., 2013; Bitam et al., 2010); whereas strain responsible for the 2009 episode in Lybia was of the Medievalis biotype (Cabanel et al., 2013). Further pulsotype analyses indicated that the 2009 Lybia episode was most likely

28| Revue de Littérature

due to an unique strain contrary to the 2003 Algeria episode which was most likely due to several, closely related strains (Cabanel et al., 2013). Sources of human plague. Both wild rodents and domestic mammals have been sources of 8 zoonotic plague in the Arab Maghreb. In Mauritania all the cases were accompanied by sick cattle and significant rodent mortality. Psammomys obesus, Gerbillus gerbillus, Gerbillus pyramidum and Gerbillus nanus are major reservoirs (Klein, 1975a). Jaculus jaculus, carriers of Synosternus cleopatrae, are of particular interest due to the density that could be involved in the epidemic process; and the fact that their movements exceed that of other gerbils (Klein, 1975b). In Morocco, Rattus norvegicus is found in the north and Rattus rattus and Rattus rattus alexandrinus in the south. Within the country, many Meriones jirds were also found. The most common flea was Xenopsylla cheopis. In Algeria, as in Mauritania, plague-resistant Meriones shawii carrying the X. ramesis flea contribute to plague persistence. As for the 2004 outbreak, it was observed that the villagers had carried out a pest control campaign around the time of the fatal index paediatric case. Moreover, villagers suffered from poor housing with R. norvegicus infestation in wheat storage areas. From September 2004 to May 2005, our team found a 21% prevalence of Y. pestis in ninety X. cheopis fleas collected from R. rattus, R. norvegicus, Mus musculus and M. spretus in this area (Bitam et al., 2006). Moreover, Y. pestis was cultured from two M. shawii animals, a plague-resistant rodent species captured in Laghouat in January 2009, suggesting a new plague focus in Algeria (Bitam et al., 2010). This observation was confirmed by a further field investigation from 2009 to 2012, capturing 237 rodents revealing the persistence of plague in Oran and Laghouat and finding three new plague foci at Cap Djinet, Biskra and M’Sila (Malek et al., 2015). In

Revue de Littérature |29 REVISED VERSION particular, the Apodemus sylvaticus wood mouse positive for Y. pestis for the first time in addition to Rattus rattus, Meriones shawii, Psamommys obesus, Mus spretus and Crocidura russula (Malek et al., 2015). In Tunis, there was a predominance of Rattus norvegicus where 75 were infected, along with Rattus rattus alexandrinus black rats (29 infected), Rattus rattus (13 infected), Mus gentilis and Mus azoricus mice of which respectively nine and three were found to be pestiferous (Pollitzer, 1954). The predominant flea was Xenopsylla cheopis while Leptopsylla segnis and Nosopsyllus fasciatus appeared rarely. In Libya, Gerbillus gerbillus and Meriones shawi were the most common rodents and Gerbillus gerbillus were captured inside nomad tents (Pollitzer, 1954). Meriones libycus is a more widespread species and is comparatively resistant to plague. It has also been found to be seropositive for plague. In the north plague focus, Meriones libycus, Meriones caudatus, Psamommys obesus and Meriones shawi were present and carrying Xenopsylla ramesis, Xenopsylla cheopis, Xenopsylla taractes and Nosopsyllus henleyi fleas. In Egypt, Rattus norvegicus in ports, Rattus rattus at warehouses and Acomys cahirinus present at along the Suez Canal were in close contacts with populations (Pollitzer, 1954). X. cheopis is the most common flea followed by Leptopsylla segnis. Infestation of Rattus rattus by fleas was one of the principal factor for the high prevalence of plague in Upper Egypt were it was particularly to be found. In these situations, controlling wild rodents and their flea ectoparasites proved to be more difficult than controlling the commensally species, due to difficulties in locating burrows and runways and the wide dispersion of rodent populations rendering difficult to precisely decide on the limits of the area to be treated. Before the appearance of DDT and in some areas to over 80 years, flea and rodent

30| Revue de Littérature

control was carried-out by fumigating burrows with cyanide gas through insufflations of HCN dusts or granules. While the results of fumigation are often dramatic, this method had several shortcomings. In large burrow systems, the fumigant was often too light to reach all parts of the burrow system and rodents could escape its effects. Therefore, 10% DDT dust was one of the most common and effective compounds used in rodent flea control programs. However, widespread emergence of insecticide resistance in populations of several important vectors including X. cheopis, and the increased concern over environmental contamination, alternative compounds have been used in the more recent episodes. Most of these compounds are effective against both adult and larval fleas. These alternative insecticides include the organo-phosphorus, carbamate, pyrethroid and insect growth regulator compounds shown to be effective in field trials. The powders applied on the slopes or in the holes (commensally rodents) or inside burrows (wild rodents) proved to be effective against flea vectors. Rodents through spots powder on their slopes or out of their burrows pick insecticide powder in their fur and spread when they lick themselves, thus killing ectoparasites fleas. In modern-day Maghreb, WHO- recommended insecticides include the pyrethroids with residual effect such as Deltamethrin powder and liquid formulations of permethrin. Semi- domestic and domestic animals are another source of plague. Buboes are seen in domestic mammals such as cats, dogs and camels (Rust et al., 1971). In camels, the principal companion of nomad tribes, buboes have long been known as ‘gudda’ in the Arabian Peninsula and have been linked to human plague bubo by nomads. Camel plague has been described by Sotnikov in 1974 in Africa, Eurasia (USSR), Asia (Mongolia, China, India) and the Middle East (Iran, Iraq) (McGrane and Higgins, 1985). Camels may present

Revue de Littérature |31 REVISED VERSION the three forms of plague and die within 20 days after an incubation period of one to six days. Not only camels but also gazelles, goats, sheep and hares die inexplicably and although not forming part of the cycle of plague were described as being responsible for contamination (Pollitzer, 1954; Fedorov, 1960). Four patients with adenitis were in close contacts with cattle including five sheep with superficial adenitis in the Al Mounek camp (Alonso, 1971). In 1976 in Libya four patients contracted plague four days after slaughtering and skinning a camel. The camel was eaten by some villagers after having been in contact with it including one who resold the meat. Seven adults exhibiting a serologically-confirmed bubonic plague were also reported and the reseller’s daughter presented with groin bubo. Villagers testify that the slaughter took place following an illness contracted by the camel which presented a swollen neck gland but no study states whether it was confirmed as plague positive or not. In Krom-el-Kheil the outbreak started by a father who killed and skinned a sick goat two days before his admission to hospital. The goat skin was kept and treated by the woman at home, where recent and older dead rats were found. It was therefore supposed that rat flea were responsible for the goat infection. In Al Azzizat, a 12-year-old boy, who had also skinned a sick goat, contracted the disease and recovered with anti-biotherapY. Four goats from the herd were tested and one of them was revealed as being seropositive to Y. pestis. In January 1977, an outbreak in Jadu in southwest of Tripolitania, involved 11 cases following the slaughter of a dying sheep, confirmed by Y. pestis isolation. In April-May 1967, an epizooty of rodents and gazelles and one child died of adenitis in Aguedat Iguenine was reported in a , a permanent nomad camp in Mauritania (Alonso, 1971). Four months before the first outbreak, fleas in tents and the absence of dead rodents were observed with

32| Revue de Littérature

deaths in camels, cattle, sheep, goats and even donkeys. Human cases suffering from bubonic plague may have resulted from contacts with sick camels which had been killed. Camel buboes cultured Y. pestis, Orientalis biotype but plague remained undeclared to the WHO (Alonso, 1971). In 32 patients, plague could be traced to direct animal contacts with goats, sheep and camels. In particular, eating camel meat recovered from a sick animal was documented in Mauritania and Libya for the second period. In addition, the epizooty in camels had resulted in human cases by affecting nomadic tribes. There is one confirmation of a published report of oral contamination. Christie et al. (1980) proposed an oral transmission route of in the 1976 Libyan outbreak. Goat and sheep are also considered to be sentinel animals indicating plague in a given focus (Pollitzer, 1954). The 1977 outbreak in Libya confirmed the high risk of direct contact with carcasses from livestock infected with plague. In addition to their migrant lifestyle leading to close contacts with rodents and thereby promoting the spread of the epidemic to other regions, some nomads’ tribes consumed dried rodent meat and traded in rodent furs (Audoin-Rouzeau, 2003). The majority of plague foci in the Maghreb are located in sub-desert nomadic regions. To the east there is the Libyan focus and to the west the north-east focus of Mauritania. In low-lying areas, where the habitat is particularly favorable both to the survival of the plague bacillus by the presence of sensitive rodent species and the availability of watering points in demand by migrants for stopovers and the presence of plants which camels fed upon (Klein, 1975b). Infected patients were a last source for secondary cases in four outbreaks. In seven patients, primary pneumonic plague could not be traced to direct air-borne transmission from an index case resulting from this route of contamination We observed a significant involvement of

Revue de Littérature |33 REVISED VERSION human ectoparasites such as Pulex irritans fleas and Pediculus humanus lice (Blanc and Baltazard, 1941a; Blanc and Baltazard,1941b). Experimental research undertaken in Morocco showed the possible transmission of plague by human ectoparasites, fleas and body lice stating that without human ectoparasites, bubonic plague epidemics are not possible (Blanc and Baltazard,1942). Moreover, careful investigation of cases means that family cases could be described and that transmission from person to person by human ectoparasites could be highlighted (Blanc and Baltazard, 1945b).

Discussion Despite limitations due to under-reporting of plague in Mauritania in 1963- 1967, and in Libya and Egypt in 1984 to avoid isolation and quarantine, and poor clinical descriptions of cases during the first period, we are confident that data reported here on plague over 75 years in the Arab Maghreb are reasonably sound enough to draw a picture of an evolving situation and to identify prospects for the next decade. The epidemiology of plague dramatically changed over 75 years in the Arab Maghreb with a sharp decrease in its overall prevalence from 6,801/22,946,800 inhabitants in the 40s to 43/92,586,424 inhabitants in the 2000s. Likewise, prognosis dramatically changed with mortality taking a sharp 0.02% decline to a residual mortality of 4.10-5 in the 2000s. Moreover, plague epidemiology changed from mixed rural and urban epidemics to rural epidemics involving nomads who currently remain the sole populations to be affected by deadly plague. Nomads mainly acquired plague through close contacts with domestic goats and camels. In particular, the consumption of poorly cooked meat from slaughtered sick animals is a source of plague for these

34| Revue de Littérature

populations; illustrating that deadly plague could be transmitted orally, as also described in Saudi Arabia in the form of severe pharyngitis in 1994 (Bin Saeed et al., 2005) and in Jordan in 1997 (Arbaji et al., 2005). Although this route of contamination has been neglected, animal models show that Y. pestis could cause deadly septicaemia after intragastric inoculation, without stool excretion (Butler et al., 1982). This situation is not surprising considering that Y. pestis was shown to have evolved from Yersinia pseudotuberculosis, a pathogen responsible for digestive tract infection, after chromosome reduction and acquisition of three plasmids (Achtman et al., 1999). Accordingly, Y. pestis retained the Yersiniagenus capability to enter the digestive tract, further acquiring the capability to cross the digestive tract barrier to provoke deadly septicaemia. These observations should be taken into consideration for further risk assessments of human plague. The fact that plague re-emerged in the very same location after decades of absence (Bitam et al., 2006; Bitam et al., 2010; Tarantola et al., 2009), and was genetically documented as being local one (Cabanel et al., 2013) along with continuous documentation of zoonotic plague illustrate the presence of plague foci in the Arab Maghreb. Evidence that Y. pestis persists in the soil under natural (Eisen et al., 2008) and experimental (Ayyadurai et al., 2008) conditions suggests that plague foci are telluric, where burrowing mammals could be infected by contacts with infected soil. Accordingly, nomads used to avoid the regions where plague cases had occurred, described as ‘cursed areas’. Respect for this rule even led people to believe that plague had been eradicated. This situation in the Arab Maghreb is similar to that reported in the United States (Kugeler et al., 2015) but contrasts sharply with the situation in neighboring Europe. Both Europe and the Arab Maghreb were exposed for two millennia to three

Revue de Littérature |35 REVISED VERSION historical plague pandemics, and recent genotyping data suggest that it may have established stable reservoirs in Europe during the 14th-17th century epidemics (Seifert et al., 2016). Nevertheless, plague did not established stable foci in Europe beyond historical period and no autochtonous case for 50 years (Schmid et al., 2015). Understanding the factors contributing to such contrasting situations may contribute to the overall understanding of the disease and its prevention.In conclusion, the unique epidemiological and clinical features of plague in the Arab Maghreb create a comprehensive view of plague as being a deadly infection residing in telluric foci which are sources for zoonotic plague transmitted to populations by direct and ectoparasite- borne contacts with infected animal and the consumption of infected meat. Primary digestive plague is a neglected yet deadly form of infection. In the Arab Maghreb countries, efforts should be made to understanding factors conditioning telluric plague foci and to strengthening surveillance of sentinel animals and ectoparasites. These efforts should be pursued to strengthen prevention in nomadic populations including hygiene in homes and cooking.

36| Revue de Littérature

Supplementary material legend Video: Kinetic cases of plague in the Arabic Maghreb, 1940-2015.

Revue de Littérature |37 REVISED VERSION

Financial support This study was financially supported by URMITE, IHU Méditerranée Infection, Marseille, France; and by the A*MIDEX project (n° ANR-11-IDEX-0001-02) funded by the «Investissements d’Avenir» French Government program, managed by the French National Research Agency (ANR). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Conflict of interest The authors declare no conflicts of interest

38| Revue de Littérature

References Achtman, M., Zurth, K., Morelli, G., Torrea, G., Guiyoule, A., Carniel, E. (1999). Yersinia pestis, the cause of plague, is a recently emerged clone of Yersinia pseudotuberculosis. Proc Natl Acad Sci U S A. 96, 14043–48. Alonso, J.M. (1971). Contribution à l’étude de la peste en Mauritanie. [in French] PhD Thesis. Université de Paris VI, Faculté de Médecine Pitié-Salpêtrière. Arbaji, A., Kharabsheh, S., Al-Azab, S., Al-Kayed, M., Amr, Z.S., Abu Baker, M., et al. (2005). A 12-case outbreak of pharyngeal plague following the consumption of camel meat, in north-eastern Jordan. Ann Trop Med Parasitol. 99, 789–93. Audoin-Rouzeau, F. (2003). Les chemins de la peste : Le rat, la puce et l’homme. [in French] Presses universitaires de Rennes. Ayyadurai, S., Houhamdi, L., Lepidi, H., Nappez, C., Raoult, D., Drancourt, M. (2008). Long-term persistence of virulent Yersinia pestis in soil. MicrobiologY. 154, 2865-71. doi:10.1099/mic.0.2007/016154-0. Bertherat, E., Bekhoucha, S., Chougrani, S., Razik, F., Duchemin, J.B., Houti, L., et al. (2007). Plague reappearance in Algeria after 50 years, 2003. Emerg Infect Dis. 13, 1459-62. doi:10.3201/eid1310.070284 Bin Saeed, A.A., Al-Hamdan, N.A., Fontaine, R.E. (2005). Plague from eating raw camel liver. Emerg Infect Dis. 11, 1456–7. Bitam, I., Ayyadurai, S., Kernif, T., Chetta, M., Boulaghman, N., Raoult, D., Drancourt, M. (2010). New rural focus of plague, Algeria. Emerg Infect Dis. 16, 1639-40. doi:10.3201/eid1610.091854 Bitam, I., Baziz, B., Rolain, J.M., Belkaid, M., Raoult, D. (2006). Zoonotic focus of plague, Algeria. Emerg Infect Dis. 12, 1975-7. doi:10.3201/eid1212.060522

Revue de Littérature |39 REVISED VERSION

Blanc, G., and Baltazard, M. (1941a). Recherches expérimentales sur la peste ; l'infection le la puce de l'homme : Pulex irritans L. [in French] Maroc Médical. 217. Blanc, G., and Baltazard, M. (1941b). Recherches expérimentales sur la peste. L'infection du pou de l'homme Pediculus corporis du Geer. [in French] C.R. Acad. Sci. 213, 851. Blanc, G., and Baltazard, M. (1945a). Documents sur la peste. [in French] Arch Inst Pasteur Maroc. 5, 349-54. Blanc G,. and Baltazard M. (1945b). Recherches sur le mode de transmission naturelle de la peste bubonique et septicémique. [in French] Arch de l’Institut Pasteur du Maroc. 3, 173-348. Blanc, G., and Baltazard, M. (1942). Rôle des ectoparasites humains dans la transmission de la peste. [in French] Bull Acad Med. 1126, 448. Butler, T., Fu, Y.S., Furman, L., Almeida, C., Almeida, A. (1982). Experimental Yersinia pestis infection in rodents after intragastric inoculation and ingestion of bacteria. Infect Immun. 36, 1160–7. Cabanel, N., Leclercq, A., Chenal-Francisque, V., Annajar, B., Rajerison, M., Bekkhoucha, S., Bertherat, E., Carniel, E. (2013). Plague Outbreak in Libya, 2009, Unrelated to Plague in Algeria. Emerg Infect Dis. 19, 230–6. doi:0.3201/eid1902.121031. Chanteau, S., Rahalison, L., Ratsitorahina, M., Mahafaly, M., Rasolomaharo, M., Boisier, P., et al. (2000). Early diagnosis of bubonic plague using F1 antigen capture ELISA assay and rapid immunogold dipstick. Int J Med Microbiol. 290, 279– 83. Christie, A.B., Chen, T.H., Elberg, S.S. (1980). Plague in Camels and Goats: Their Role in Human Epidemics. J Infect Dis. 141, 724-6. Eisen, R.J., Petersen, J.M., Higgins, C.L., Wong, D., Levy, C.E., Mead, P.S., et al. (2008). Persistence of Yersinia pestis in soil

40| Revue de Littérature

under natural conditions. Emerg Infect Dis. 14, 941- 3. doi: 10.3201/eid1406.080029. Fedorov, V.N. (1960). Plagues in camels and its prevention in the USSR. Bull World Health Organ. 23, 275–81. Guiyoule, A., Grimont, F., Iteman, I., Grimont, P.A., Lefèvre, M., Carniel, E. (1994). Plague pandemics investigated by ribotyping of Yersinia pestis strains. J Clin Microbiol. 32, 634-41. Haddad, C., and Valero, A. (1948). Streptomycin in bubonic plague. Br Med J. 1, 1026. Kaul, P.M. (1949Prevalence of plague in the world in recent years. World Health Organization. Epidemiol Vital Stat Report. 2, 143–62. Klein, J.M., Poulet, A.R., Simonkovich, E. (1975a). Observations écologiques dans une zone enzootique de peste en Mauritanie 1. Les rongeurs, et en particulier Gerbillus gerbillus, Olivier, 1801 (Rodentia, Gerbillinae). [in French] Cah. O.R.S.T.O.M., sér. Ent. Méd. et Parasitol. 13, 13-28. Klein, J.M., Alonso, J.M., Baranton, G., Poulet, A.R., Mollaret, H.H. (1975b). La peste en Mauritanie. [in French] Med Mal Infect. 5, 198-207. Kugeler, K.J., Staples, J.E., Hinckley, A.P., Gage, K.L., Mead, P.S. (2015). Epidemiology of Human Plague in the United States, 1900–2012. Emerg Infect Dis. 21, 16-22. Mafart, B., Brisou, P., Bertherat, E. (2004). [Plague outbreaks in the Mediterranean area during the 2nd World War, epidemiology and treatments]. [in French] Bull Soc Pathol Exot. 97, 306-10. Malek, M.A., Hammani, A., Beneldjouzi, A., Bitam, I. (2015). Enzootic plague foci, Algeria. New Microbes New Infect. 4, 13-6. doi:10.1016/j.nmni.2014.11.003 McGrane, J.J., and Higgins, A.J. (1985). Infectious diseases of the camel: viruses, bacteria and fungi. Br Vet J. 141, 529-47.

Revue de Littérature |41 REVISED VERSION

Pollitzer, R. (1954). Plague. Geneva: World Health Organization. http://apps.who.int/iris/bitstream/10665/41628/1/WHO_MON O_22.pdf?ua=1 Raoult, D., Mouffok, N., Bitam, I., Piarroux, R., Drancourt, M. (2013). Plague: history and contemporary analysis. J Infect. 66, 18-26. doi.10.1016/j.jinf.2012.09.010. Roux, A.H., and Mercier, C. (1946). Sur cinq cas de peste pulmonaire primitive dont trois suivis de guérison, observés à l’hôpital civil d’Oran. [in French] Bull Soc Exot. 39, 173-8. Rust, J.H., Dan, Jr, Cavanaugh, C., O'Shita, R., Marshall, J.D. (1971). The Role of Domestic Animals in the Epidemiology of Plague. J Infect Dis. 124, 522-6. Seifert, L., Wiechmann, I., Harbeck, M., Thomas, A., Grupe, G., Projahn, M., Scholz, H.C., Riehm, J.M. (2016). Genotyping Yersinia pestis in historical plague: evidence for long-term persistence of Y. pestis in Europe from the 14th to the 17th CenturY. PLoS One.11, e0145194. Schmid, B.V., Büntgen, U., Easterday, W.R., Ginzler, C., Walløe, L., Bramanti, B., Stenseth, N.C. (2015). Climate-driven introduction of the Black Death and successive plague reintroductions into Europe. Proc Natl Acad Sci U S A. 112, 3020-5. doi:10.1073/pnas.1412887112. Signoli, M., Chevé, D., Adalian, P., Boëtsch, G., Dutour, O. (2007). Plague: epidemics and societies. Firenze: Firenze university press. http://digital.casalini.it/9788884534897. Tarantola, A., Mollet, T., Gueguen, J., Barboza, P., Bertherat, E. (2009). Plague outbreak in the Libyan Arab Jamahiriya. Euro Surveill. 14. pii:19258. doi: 10.3201/eid1902.121031 Tomich, P.Q. (1947). Preliminary sylvatic plague studies in the Suez Canal zone. J Roy Egypt Med Ass. 30, 239-4

42| Revue de Littérature Chapitre II

5. Chapitre II : Foyers de peste enzootique, Algérie

5.1 Enzootic plague foci, Algeria

Nous avons vu que nous étions en présence de peste sylvatique en Afrique

du Nord permettant une ré-émergence des épidémies après de nombreuses

années, à plusieurs reprises au cours des années 2000. Afin de mieux

connaitre ces foyers de peste et dans la perspective de mettre en évidence de

nouveaux foyers de peste, nous avons mis en place un projet de surveillance

d’animaux sentinelle en Algérie. Ce travail de terrain a permis de mettre en

évidence trois nouveaux foyers de peste (M’Sila, Biskra et Cap Djinet)

porteurs d’un nouveau génotype (Multi Spacer Typing) de Y. pestis de

biotype Orientalis. Ce même travail nous a permis d’ajouter l’espèce

Apodemus sylvaticus à la liste des rongeurs pestiférés.

Foyer de peste enzootique, Algérie |43

ORIGINAL ARTICLE Enzootic plague foci, Algeria

M. A. Malek1,2, A. Hammani3, A. Beneldjouzi4 and I. Bitam2 1) Aix Marseille Université, URMITE, UM 63, UMR_S 1095 UMR 7278, 13385 Marseille, France, 2) Laboratoire VALCORE, Faculté des Sciences, Université M’Hamed Bougara Boumerdès (UMBB), Boumerdès, 3) Faculté des Sciences Biologiques et Agronomiques, Université Mouloud Mammeri, Tizi Ouzou and 4) Institut Pasteur d’Alger, Dély Ibrahim, Algeria

Abstract

In Algeria, PCR sequencing of pla, glpD and rpoB genes found Yersinia pestis in 18/237 (8%) rodents of five species, including Apodemus sylvaticus, previously undescribed as pestiferous; and disclosed three new plague foci. Multiple spacer typing confirmed a new Orientalis variant. Rodent survey should be reinforced in this country hosting reemerging plague. New Microbes and New Infections © 2014 The Authors. Published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases.

Keywords: Algeria, Molecular typing, North Africa, Plague, Yersinia pestis Original Submission: 28 March 2014; Revised Submission: 1 November 2014; Accepted: 8 November 2014 Article published online: 4 December 2014 Corresponding author. I. Bitam, Laboratoire VALCORE, Faculté des Sciences, Université M’Hamed Bougara, 9 Boumerdès (UMBB), 35000 Boumerdès, Algeria E-mail: [email protected]

Foyer de peste enzootique, Algérie |45

Introduction

Plague, a deadly infection caused by the bacterium Yersinia pestis, is reemerging in some North African countries, including Libya and Algeria [1–3]. In Algeria, plague reappeared after 50 years of silence with two consecutive episodes in 2003 in Oran [1] and in 2008 in a small camp of nomads in the Thait El Maa area in Laghouat province [2]. In both outbreaks, patients originated from rural areas where they raised animals. Confirmation of the two Algerian outbreaks was made by using molecular investigations of the presence of Y. pestis in rodents and in rodents' fleas [1,2]. When the disease broke out in the Oran area in 2003, no plague focus had been described for decades in Algeria after rodent surveys were dropped. Therefore, in an effort to depict the current activity of plague foci in Algeria, we initiated a rodent study and molecular investigations of rodents captured in nine regions of Algeria.

Methods

Yearly field missions were conducted in 2009 to 2012, primarily in northern Algeria (Fig. 1). These missions aimed to better understand the diversity of small mammals, including rodents maintaining Y. pestis in zoonotic foci throughout the country. All catches were made on private farms from November 2009 to February 2012 by using BTS (Besançon Technique Service; INRA, Montpellier, France) and Sherman Trap (H. P. Sherman Traps, Tallahassee, FL, USA). After morphological identification, rodents were humanely killed; the spleen was extracted and stored individually in a

46| Foyer de peste enzootique, Algérie

sterile Eppendorf tube in ethanol (70%) before being tested in Marseille, France, in a biological security level 3 laboratory. Ethanol-preserved spleens were rinsed with sterile distilled water for 2 minutes, and total DNA was extracted by using the NucleoSpin DNA purification tissue kit according to the manufacturer's instructions (Macherey and Nagel, Düren, Germany). Real-time PCRs were performed by using a CFX 96 Real Time PCR System (Applied Biosystems, Coignières, France). Negative controls, consisting of noninfected Balb/c mice spleen total DNA, were introduced every five samples in all PCR experiments. In a first step, a 98 bp fragment of the plasminogen activator gene (pla) was amplified as previously described [4]. Confirmation was done by further partial PCR amplification and sequencing of the glpD gene encoding the glycerol-3-phosphate dehydrogenase [5] and on positive specimens by partial PCR amplification and sequencing of a 100 bp fragment of rpoB gene that encodes the β subunit of RNA polymerase [6]. Positive specimens were further genotyped by multiple spacer typing (MST) by sequencing PCR-amplified spacers YP1, YP3, YP4, YP5, YP7 and YP8, as previously described [7]. Gene sequences obtained with an ABI 3130Xl Genetic Analyzer (Applied Biosystems) were compared with those available in GenBank by using the nucleotide–nucleotide BLAST (blastn) program (available from http://www.ncbi.nlm.gov/BLAST/), and spacer sequences were compared with those previously reported [7].

Foyer de peste enzootique, Algérie |47

Results A total of 237 rodents were captured in the geographical area indicated in Fig. 1. While negative controls remained negative, pla fragments were amplified in 44/237 (18.5%) spleen specimens, with a cycle threshold value ranging from 27.64 to 34.35. Pla-positive specimens were collected from two Rattus norvegicus in Tlemcen; six Rattus rattus in Mascara; one Meriones shawii in Laghouat; two M. shawii in Biskra; one R. rattus in Batna; four M. shawii, three Psammomys obesus, one Mus spretus and two R. rattus in M'Sila; and 13 Mus spretus, six Apodemus sylvaticus and three Crocidura russula in Cap Djinet. Among 44 pla-positive specimens, 18 (41%) were further positive for both the glpD gene and for rpoB amplification in fourR. rattus from Mascara; one M. shawii from Laghouat; two M. shawii from Biskra; two M. shawii, twoP. obesus, two R. rattus and one M. spretus from M'Sila; and two C. russula, one M. spretus and oneA. sylvaticus from Cap Djinet. These specimens were regarded as definitely positive for Y. pestis. glpDsequences exhibited 100% identity to the reference sequence for biovar Orientalis (GenBank accession numbers AL590842 and YPO3937) characterized by a 93 bp deletion. Multispacer sequence typing yielded the same profiles in all the specimens, including spacer YP1 type 1; spacer YP3, type 5; spacer YP4, type 1; spacer YP5, type 1; and spacer YP8, type 2. YP7 spacer was sequenced in only nine specimens as a result of a limitation of the materials, and yielded a type 9. Altogether, MST data indicated a new MST type 20 in the Orientalis biovar.

48| Foyer de peste enzootique, Algérie

FIG. 1. Map of Algeria indicating number of Yersinia pestis–positive captured rodents in 12 areas. Tlemcen: five Rattus norvegicus. Aïn Témouchent: one Rattus rattus and one Meriones shawii. Mascara: 14 R. rattus. Laghouat: six M. shawii. Djelfa: eight R. rattus, four M. shawii and four Mus spretus. M’Sila: 22 M. shawii, 12 Psammomys obesus, 11 R. rattus, ten Mus spretus, five Jaculus jaculus and one Atelerix algirus. Biskra: 16 M. shawii. Batna: four P. obesus, two R. rattus and one Mus spretus. Algiers: two R. rattus. Boumerdès: three A. algirus and three R. rattus. Cap Djinet: 58 Mus musculus, 24 Crocidura russula, 13 Apodemus sylvaticus, six Lemniscomys barbarus, one R. rattus.

New Microbes and New Infections © 2014 The Authors. Published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases, NMNI, 4, 13–16 This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/)

Foyer de peste enzootique, Algérie |49

TABLE 1. Rodent species captured according to Algerian locality and PCR sequencing results

New Microbes and New Infections © 2014 The Authors. Published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases, NMNI, 4, 13–16 This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/)

50| Foyer de peste enzootique, Algérie

Discussion

Here, we achieved a renewed picture of plague enzooty in Algeria by using PCR sequencing of Y. pestis in field rodents (Table 1). Data were authenticated by the fact that no positive control was used and negative controls remained negative, eliminating the possibility of in-laboratory contamination. Here, tentative culture could not be achieved because of the lack of a level 3 biosafety laboratory in Algeria, and because of conservation of the specimens in 70% ethanol for safe transport and analysis [8]. PCR is routinely used to monitor ectoparasites and sentinel animals for bacteria causing emerging zoonotic diseases, such as Rickettsia spp. and Bartonella spp [9]. As for plague, PCR sequencing has been widely used to investigate the presence of Y. pestis in wild rodents and ectoparasites [10–12], as well as in humans [13]. Here, results obtained by pla partial amplification were confirmed by glpD gene amplification and MST. We observed that 59% of spleen specimens positive for pla were not confirmed by further molecular analyses. It was recently reported that the pla gene may not be specific for Y. pestis after its detection in tissues of uninfected R. rattus and R. norvegicus animals [16]. Therefore, any pla gene result must be confirmed by additional evidence for Y. pestis. In Algerian rodents, we observed only biovar Orientalis confirmed by glpD glycerol-negative sequencing and MST. This in agreement with recent observations thatY. pestis infecting patients in Algeria belonged to biovar Orientalis [1]. In neighbouring Libya, however, patients were found to be infected by the biovar Medievalis [17]. Unlike our study, Libyan and Algerian 2003 typing were analysed by pulsed-field gel electrophoresis,

Foyer de peste enzootique, Algérie |51 which requires large amounts of cultured microorganisms, but subculture alters ribotyping classification, and biovar identification was performed on the basis of their ability to ferment glycerol and reduce nitrate [17]. Among the 11 sites we investigated in Algeria, five (46%) yielded evidence of plague foci. Two of these foci were previously known in Mascara and Laghouat [1,2], whereas M'Sila, Biskra and Cap Djinet sites have not been registered as plague foci for 50 years. Underreporting of the plague as a result of the lack of laboratories for confirmation of the diagnosis contributes to making the epidemiologic situation and the actual impact of the disease difficult to assess. While 18 of the rodent species we studied that were positive forY. pestis were previously known as plague reservoirs [18], we found for the first time that the wood mouse (A. sylvaticus) was positive for Y. pestis. This species, never before described as pestiferous, is found in northwestern Africa along the entire coastal plain in Morocco, Algeria and Tunisia but is absent in Libya and Egypt because its southern range is limited by desert habitats [19]. This species also lives in many Mediterranean islands and has a large area of distribution in continental Europe [20]. It is abundant, and in some places, it is considered to be a pest species [21]. A. sylvaticus is probably resistant to plague, as we captured only living animals, and this one showed no particular signs of disease. However, plague foci are thought to result from a subtle balance between plague-susceptible and plague-resistant rodents [22]. Our work illustrates the necessity for regular observation of wild rodents in order to disclose active plague foci and to update the actual activity of plague in Algeria. In particular, discovery of new enzootic plague foci should alert doctors of the possibility of this diagnosis for patients exposed in areas endemic for reservoir animals.

52| Foyer de peste enzootique, Algérie

Conflict of Interest

None declared.

Acknowledgements.

This study was supported by Unité de Recherche sur les maladies Infectieuses et Tropicales, Institut Hospitalier Universitaire «Méditerranée Infection», Marseille, France.

References

1. Bitam I., Baziz B., Rolain J.M. Zoonotic focus of plague, Algeria. Emerg Infect Dis. 2006;12:1975–1977. 2. Bitam I., Ayyadurai S., Kernif T. New rural focus of plague, Algeria. Emerg Infect Dis. 2010;16:1639–1640. 3. Tarantola A., Mollet T., Gueguen J. Plague outbreak in the libyan Arab Jamahiriya. Euro Surveill.2009;14:19258. 4. Charrel R.N., La Scola B., Raoult D. Multi-pathogens sequence containing plasmids as positive controls for universal detection of potential agents of bioterrorism. BMC Microbiol. 2004;4:1–11. 5. Tran T.N.N., Signoli M., Fozzati L. High throughput, multiplexed pathogen detection authenticates plague waves in medieval Venice, Italy. PLoS One. 2011;6:1–5. 6. Drancourt M., Aboudharam G., Signoli M. Detection of 400-year- old Yersinia pestis DNA in human dental pulp: an approach to the diagnosis of ancient septicemia. Proc Natl Acad Sci U S A. 1998;95:12637–12640.

Foyer de peste enzootique, Algérie |53

7. Drancourt M., Roux V., Vu Dang L. Genotyping, Orientalis- like Yersinia pestis, and plague pandemics.Emerg Infect Dis. 2004;10:1585–1592. 8. Ayyadurai S., Flaudrops C., Raoult D. Rapid identification and typing of Yersinia pestis and other Yersiniaspecies by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry. BMC Microbiol. 2010;12:1–7. 9. Kumsa B., Parola P., Raoult D. Molecular detection of Rickettsia felis and Bartonella henselae in dog and cat fleas in Central Oromia, Ethiopia. Am J Trop Med Hyg. 2014;90:457–462. 10. Riehm J.M., Tserennorov D., Kiefer D. Yersinia pestis in small rodents, Mongolia. Emerg Infect Dis.2011;17:1320–1322. 11. Griffin K.A., Martin D.J., Rosen L.E. Detection of Yersinia pestis DNA in prairie dog–associated fleas by polymerase chain reaction assay of purified DNA. J Wildl Dis. 2010;46:636–643. 12. Adjemian J.Z., Adjemian M.K., Foley P. Evidence of multiple zoonotic agents in a wild rodent community in the eastern Sierra Nevada. J Wildl Dis. 2008;44:737–742. 13. Begier E.M., Asiki G., Anywaine Z. Pneumonic plague cluster, Uganda, 2004. Emerg Infect Dis.2006;12:460–467. 16. Pollitzer R. World Health Organization; Geneva: 1954. Plague; pp. 266–267. Monograph Series 22. 17. Khammes N., Aulagnier S. Diet of the wood mouse, Apodemus sylvaticus, in three biotopes of Kabylie of Djurdjura (Algeria) Folia Zool. 2007;56:243–252. 18. Wilson D.E., Reeder D.M. 3rd ed. Smithonian Institution Press; Washington, DC: 2005. Mammals species of the world. A taxonomic and geographic reference. 19. Schlitter D., Van der Straeten E., Amori G. IUCN Red List of Threatened species. International Union for Conservation of Nature (IUCN); Gland, Switzerland: 2012. Apodemus sylvaticus. 20. Perry R.D., Fetherston J.D. Yersinia pestis—etiologic agent of plague. Clin Microbiol Rev. 1997;10:35–66.

New Microbes and New Infections © 2014 The Authors. Published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases, NMNI, 4, 13–16 This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/)

54| Foyer de peste enzootique, Algérie Chapitre III

6. Chapitre III : Halo-tolérance et révélation des foyers de

peste

6.1 Article 2 : Yersinia pestis halotolerance illuminates

plague reservoirs

La localisation des foyers de peste en Algérie sur une carte géographique

nous a permis d’émettre l’hypothèse de leur co-localisation avec des

environnements inanimés salés. Cette hypothèse a été confortée par

l’analyse statistique de co-localisation des foyers de peste au Maghreb avec

les environnements salins et non-salins ; l’isolement de Y. pestis à partir

d’un prélèvement de sol salé, mais pas de prélèvements de sols non-salés ;

et la démonstration expérimentale de la résistance d’une souche de Y. pestis

biotype Orientalis, sous forme L dans des bouillons à salinité croissante.

Cette observation peut s’appliquer non seulement au Maghreb mais à

l’ensemble des foyers de peste de l’Hémisphère Nord

Halo-tolérance et révélation des foyers de peste |55

Yersinia pestis halotolerance illuminates plague reservoirs

Maliya Alia Malek*,†, Idir Bitam*,†, Jérôme Terras*, Jean Gaudart§, Said Azza*, Christophe Flaudrops*, Catherine Robert*, Didier Raoult *, Michel Drancourt *

* Aix Marseille Université, URMITE, UMR 63, CNRS 7278, IRD 198, Inserm 1095, Faculté de Médecine, 27 Bd Jean MOULIN, 13385 Marseille Cedex 5, France † Laboratoire Biodiversité et Environnement : Interactions Génomes, Faculté des Sciences Biologiques Université des Sciences et de la Technologie Houari Boumediene, El Alia, Bab Ezzouar 16111, Algérie. § Aix-Marseille Université, UMR912 SESSTIM (INSERM/IRD/AMU), Faculté de Médecine, 27 Bd Jean Moulin, 13385 Marseille Cedex 5, France

*Corresponding author: Michel Drancourt, Unité de Recherche sur les Maladies Infectieuses et Tropicales Emergentes, Faculté de Médecine, 27 Boulevard Jean Moulin, 13385 Marseille Cedex 5, France. Tel: 33 4 91 32 43 75. Fax: 33 4 91 38 77 72. Email: [email protected]

Manuscript submitted to Nature Communications: 3rd May 2016 Under revision: 15 June 2016

Abstract word count: 176 Text word count: 3389 Number of figures: 4 Number of Supplementary files: 4

Key words: Yersinia pestis, halotolerance, soil, Chott, NaCl.

Halo-tolérance et révélation des foyers de peste |57 Abstract

Two millennia after swiping over Europe and North Africa, plague established permanent foci in North Africa but not in neighboring Europe.

Mapping human plague foci reported in North Africa for 70 years indicated they are all significantly located at <3 kilometers from the Mediterranean seashore or the edge of salted lakes named chotts. In Algeria, culturing 352 environmental specimens naturally containing 0.5 to 70 g/L NaCl yielded only one Yersinia pestis Orientalis biotype isolate in one 40 g/L NaCl chott soil specimen. Culturing Y. pestis in broth steadily enriched in NaCl indicated survival up to 150 g/L NaCl as L-form variants exhibiting a distinctive and unique MALDI-TOF mass spectrometry peptide profile.

Further transcriptomic analyses found the upregulation of several outer- membrane proteins including TolC efflux pump and OmpF porin implied in osmotic pressure regulation. Salt tolerance of Y. pestis L-form may play a role in the maintenance of natural plague foci in North Africa and beyond, as these geographical correlations could be extended to 31 plague foci in the northern hemisphere (from 15°N to 50°N).

58|Halo-tolérance et révélation des foyers de peste

Significance

Halotolerance of the plague agent Yersinia pestis is shown to drive its persistence in soil, a discovery which may help focusing plague survey in some restricted salt areas.

Plague is a deadly infectious disease caused by the bacterium Yersinia pestis (1,2). Post-genomic analyses confirmed that Y. pestis was derived from the environmental bacteria Yersinia pseudotuberculosis 1,500 to

20,000 years ago in Central Asia and gradually spread from east to west along the historical tracks of human migration such as the Silk Road (3-5).

Y. pestis reached Europe and North Africa where it caused a first pandemic called the Justinian pandemic between 541-767 AD; then a medieval pandemic between 1346 and the end of the eighteenth century (6). These two historical pandemics, characterized by an explosive mortality, killing up to half of the urban populations in a few months, have been microbiologically confirmed by the detection of specific nucleotidic sequences (7-9) and the reconstitution of the entire genome of several strains (5,10,11). This epidemic regimen, which has never been observed thereafter, was likely fueled by indirect inter-human transmission by human ectoparasites such as lice (12-17) and the Pulex irritans fleas (18). A third pandemic began in the Hong Kong area in the late nineteenth century (2). It

Halo-tolérance et révélation des foyers de peste |59 is currently responsible for hundreds of deaths every year. It exhibits a different epidemiological regimen characterized by epidemics circumscribed in time and in more limited geographical areas that are called plague foci (19). These epidemics are linked to transmission of Y. pestis by small wild mammal ectoparasites without inter-human transmission (1).

Moreover, small outbreaks are linked with contact with infected cats (20) or consumption of infected meat (21) and sporadic cases described at the southern border of the United States are due to direct contact with infected wild animal carcasses (22).

Strikingly, two millennia after swiping over Europe and North

Africa, plague established foci in later regions but not in neighboring

Europe (4). In North Africa, plague foci are still active as illustrated by the resurgence of the plague after 53 years of silence in Oran, Algeria, for which genetic analyses confirmed a local and not imported strain of Y. pestis (23). In these foci, plague is a zoonosis since the Y. pestis bacterium is introduced into populations from infected animals (1). These constitute a link in an epidemiological chain involving a balanced transmission between plague-susceptible and plague-resistant species (24).

It is likely that, ultimately, animals become contaminated from infected soil (25,26). Indeed, Y. pestis was isolated from rodent burrow soil

60|Halo-tolérance et révélation des foyers de peste

several years after any animal had actually lived there (27). Also in natural conditions, a strain of Y. pestis was isolated from the ground at the point of death of a mountain lion with plague, three weeks after the death of the animal (22). Experimental data have confirmed several times the persistence of living Y. pestis up to 28 months after artificial inoculation of soil (28,29).

The reasons for the persistence of plague foci in North Africa and not in neighboring Europe are not understood. In this context, we observed that in North Africa plague foci were significantly located at the periphery of chotts, which are salty areas with a salt content from 10 g/L to saturation

(300-400g/L), higher than that of the seas and oceans (30). We isolated a new strain of Y. pestis in Algeria in just one chott soil sample containing

40 g/L of salt. Finally, we showed that the persistence of Y. pestis in soil samples artificially inoculated with this strain was the same in the presence of salt, but as L-form like variants that had been poorly described for this bacterial species.

Halo-tolérance et révélation des foyers de peste |61

Figure 1.

A) Location of human plague foci in six countries in North Africa, 1940- 2015. Plague foci are significantly located <3 km of salt source (Mediterranean sea and chotts).

● Salt water; ● Fresh water; ● 1-10 cases; ● 10-100 cases ;●> 100 cases.

B) Boxplot of minimum distances to plague foci (y axis, distance in km).

C) Boxplot of mean distances to plague foci (y axis, distance in km).

62|Halo-tolérance et révélation des foyers de peste

RESULTS Co-localization of plague foci and chotts, North Africa. In North Africa, plague reemerged in Oran, Algeria, in 2003 (31), 53 years after a previous episode in the same city and in Tobruk, Libya, in 2009 (32). Both outbreaks are commonly located on the edge of the Mediterranean Sea with a 0-35 g/L salinity in addition to the edge of the Sebkha with up to 400 g/L salinity for the Algerian cases. We mapped the human plague foci reported in North Africa for 75 years (33) and observed that these foci were all located at a distance <3 kilometers from the sea or from the edge of a chott, which designates an inland salty area (Figure 1). Salt water ponds were significantly closer to plague foci than non-salt water ponds, according to the minimum distance (median [IQR] 2.67 km [3.76] vs 4.12 km [4.68], p<0.001) (Figure 1).

Figure 2. Yersinia pestis forms filamentous colonies in 150g/L NaCl-broth (A) and small colonies (B): right panel, control; left panel, Y. pestis exposed to 150g/L NaCl. Salt water ponds were also significantly closer in mean to plague foci (median [IQR] 56.30 km [1.87] vs 64.83 km [44.22], p<0.001) (Figure 3).

Halo-tolérance et révélation des foyers de peste |63

localization plagueof withlakes salt foci in North hemisphere.

-

Co SupplementaryFigure 1.

64|Halo-tolérance et révélation des foyers de peste

Furthermore, we measured a significant spatial association between human plague for 31 foci and saline water in comparison to freshwater on the northern hemisphere (from 15°N to 50°N) (Supplementary Figure 1).

Significant statistical analysis showed that the median distance to a plague focus is greater for freshwater sources than for salt springs. Three quarters of the salt springs are located less than 3 km away while three quarters of freshwater sources are further than 3 km away. Indeed, the median distance of a plague focus to a salt water point is 0.89 (IQR=2.66) with a minimum of 0.01 km and a maximum of 7.74 km, while the median distance to a freshwater source is 4.63 (IQR=2.7) km with a minimum of 0.09 km and a maximum of 9.95 km.

Isolation of Y. pestis in chott. To investigate the presence of Y. pestis in chotts in Algeria, we collected 208 soil samples comprising: 120 soil samples collected in chotts (salinity, 25 to 70 g/L) and 88 soil samples collected outside chotts (salinity, 0.5 to 5g/L); and 144 water samples of which 98 contained salt water (50 g/L NaCl) and 46 contained freshwater

(0.05 to 5g/L NaCl) (Table 1). Culturing these 352 samples in a safety level

3 laboratory yielded one Y. pestis isolate (named Y. pestis Algeria3) in one chott soil specimen containing 4% NaCl against zero isolates from low

Halo-tolérance et révélation des foyers de peste |65 salinity control soil samples. The Y. pestis Algeria3 isolate was biotyped

Orientalis Y. pestis by whole genome sequencing (GenBank accession number FAUR00000000). In order to confirm this observation, we compared the survival of Y. pestis Algeria1, another Orientalis isolate from

Algeria (34), after its experimental inoculation in three different natural soils collected in Algeria, sterilized by autoclaving and supplemented with

0.5g/L NaCl or 40g/L NaCl. There was no difference in the 5-week survival of Y. pestis Algeria1 in these two seeded soil conditions (Table 2) (p=015).

Survival in salted soil was not restricted to just the Algeria3 isolate.

66|Halo-tolérance et révélation des foyers de peste

Supplementary Table 1. List of samples collected in North Algeria for the detection of Y. pestis.

Halo-tolérance et révélation des foyers de peste |67 Supplementary Table 2. Survival of Y. pestis Algeria 1 in a 40g/L salt after 5-week inoculation : number of colonies after 48-hour incubation on blood agar at 28°C, 5% CO2.

7 days 14 days 21 days 28 days 35 days Mean

Soil 1 1a 214 209 210 205 198 207.2

1b 228 224 219 221 206 219.6

1c 203 201 204 198 189 199

Control 1 236 216 224 216 202 218.8

Soil 2 2a 232 228 219 224 213 223.2

2b 221 216 214 217 211 215.2

2c 229 209 213 213 207 214.2

Control 2 219 212 215 210 197 210.6

Soil 3 3a 237 223 229 231 219 227.8

3b 216 213 211 205 196 208.2

3c 227 221 217 218 203 217.2

Control 3 222 209 213 205 197 209.2

Salt-induced Y. pestis L-forms. To further characterize Y. pestis cells exposed to salt, we cultured Y. pestis Algeria1 in trypticase-soy broth supplemented with an increasing concentration of salt in the presence of a control cultured in parallel in standard broth. After a seven-week culture in broth containing 150 g/L NaCl, Y. pestis formed large filamentous colonies

68|Halo-tolérance et révélation des foyers de peste

that troubled the broth (Figure 2A). Subculture of these colonies on 5% sheep-blood agar yielded opaque white microcolonies with a smooth surface, a regular round shape and an entire edge like the control strain but smaller (0.9-1 mm) (Figure 2A). Such colonies were less sticky and easier to take off completely losing their natural viscosity than the control Y. pestis cell colonies, which formed sticky strands when touched by an inoculation loop. The colony became whiter to opaque when we poured trypticase-soy broth on the Petri dish. Gram staining disclosed morphological changes of

Y. pestis for salt concentrations starting at 50g/L NaCl; at lower NaCl concentrations, Y. pestis showed a classic morphology featuring individual coccobacilli whereas for concentrations above 50 g/L NaCl, Y. pestis featured aggregate cocci of a small size. Negative staining and electron microscopy showed the control culture to contain bacilli measuring 1,340 ±

72 µm by 810 ± 21 µm (Figure 2D) whereas Y. pestis exposed to 150 g/L

NaCl exhibited a round shape and measured 973 ± 32 µm by 901 ± 25 µm

(P < 0.05). Inclusion electron microscopy showed that, compared to control

Y. pestis cells, cells of Y. pestis exposed to 150 g/L NaCl exhibited a thinner cell membrane and a thinner inner membrane, which eventually vanished as reported in spheroplasts (L-forms) (35) (Figure 2E).

Halo-tolérance et révélation des foyers de peste |69 Salinity induces a specific transcriptional program in L-form Y. pestis.

The analysis of protein spectra obtained by MALDI-TOF-MS spectrometry confirmed the identification of Y. pestis Algeria1 with an identification score of 2.299-2.5; and revealed a reproducible pattern with the same distribution but with slight differences in absolute intensity between those obtained with the reference strain (Figure 3).

Figure 3. Electron microscopy of Y. pestis control (A) and Y. pestis exposed for seven weeks to 150g/L NaCl (B).

Furthermore, of the 102 spots obtained by 2D differential gel electrophoresis separation (Figure 4A), 68 were identified by MS analysis which yielded 42 proteins belonging to 17 different COGs (Figure 4B).

70|Halo-tolérance et révélation des foyers de peste

Figure 4. Proteomics of Y. pestis exposed to 150g/L NaCl: (A): MALDI- TOF mass spectrometry (a: control; b: exposed) (B). Representative 2D differential gel electrophoresis (DIGE) analysis of Y. pestis proteins (C): string network of DIGE analysis of Y. pestis proteins. Each individual sample from Y. pestis Orientalis wild type and 150g/L NaCl exposed Y. pestis and a pooled reference sample were labeled with Cy5, Cy3, and Cy2, respectively, and were then separated on the same gel using the 2D- DIGE system. Three images were obtained from each gel and an overlay of dye scan images was also obtained. Selected protein spots exhibiting an ANOVA score lower or close to 0.05 and a change of at least 1.5-fold intensity are indicated by circles and spot numbers as indicated in Table (D): Analysis, according the COG family, of upregulated and downregulated proteins identified by Mass spectrometry after 2D- differential gel electrophoresis separation.

Halo-tolérance et révélation des foyers de peste |71 COG M, C and E contained the majority of the differentially regulated proteins. More particularly, upregulation was observed for outer-membrane proteins OmpF, OmpA and TolC (hasF), an outer-membrane efflux pump; and proteins involved in energy production such as L-lactate dehydrogenase. Up-regulation was also observed for proteins involved in cellular processes and signaling such as YaeT; as well as proteins involved in metabolism such as lldD, nuoD, atpD, atpA, aspA, proV, ureC, ppSA, lamB and fadL; and proteins involved in information storage and processing such as ssb and rarA and finally other proteins poorly characterized. On the other hand, the D-lactate dehydrogenase, dihydrolipoamide dehydrogenase, beta succinyl-CoA synthetase subunit and catalase katE and katY were downregulated.

72|Halo-tolérance et révélation des foyers de peste

DISCUSSION

We present several lines of evidence that the plague agent Y. pestis is surviving in salt soil environments such as the ones encountered in so- called chotts in North Africa. Indeed, after we observed significant co- localization of plague foci reported for 70 years in animals and humans in

North Africa, with salt chotts, we cultured one Y. pestis Algeria3 isolate from one chott soil specimen containing 40 g/L NaCl, whereas no isolate was produced from control soil specimens containing 0.5-1 g/L salt.

Algeria3 is the first soil isolate of Y. pestis in Africa, as the three previous isolates made over 120 years were in Asia (2, 27) and America (22).

Experimental studies further confirmed that Y. pestis survived in soil containing 40 g/L salt and in hypertonic broth containing up to 150 g/L salt.

Y. pestis exposed to salt formed round cells looking like L-forms.

More than one century ago, ―involution forms‖ were reported after exposure of Y. pestis to 35 g/L NaCl (36-37). ―Involution forms‖ are convincingly identical to L-forms photographed after culture of Y. pestis in 20 g/L salt broth (38). Interestingly, several Russian reports made in the 70-80s described Y. pestis L-forms in rodents and their ectoparasites collected in plague foci (39).

Halo-tolérance et révélation des foyers de peste |73 Y. pestis did not sustain direct exposure to high NaCl concentrations, but rather a progressive, stepwise exposure to high salinity.

This observation suggests either the selection of halotolerant variants within the population or, more likely, a progressive adaptation to increasing NaCl concentrations. L-forms may reflect an adaptation of Y. pestis to osmotic pressure by decreasing the surface in contact with the environment. In line with these morphological observations, we noticed upregulation of several outer-membrane proteins including efflux pumps such as TolC (linked to the inner membrane AcrB), exporting toxic compounds to the bacteria (40), and OmpF porin regulating the osmotic pressure to maintain cell permeability (41). Whereas Escherichia coli OmpF is upregulated by low osmolarity and repressed by high osmolarity, Y. pestis OmpF expression is not repressed by high osmolarity but is incredibly upregulated (41).

Alternatively, halotolerance in Y. pestis may rely on the Na+/H+ antiporter

(nhaA and nhaB), recently described as necessary for the virulence of Y. pestis (42). It has been shown that the Na+/H+ antiporters in the model of

Vibrio cholerae, the causative agent of cholera, confer Na+ resistance involving the NADH-quinone oxidoreductase pump (NQR) (43).

Extending our geographic investigations in North Africa to the

North hemisphere, we observed that plague foci are indeed significantly located close to salt sources (Supplementary Figure 1). This observation is

74|Halo-tolérance et révélation des foyers de peste

no longer true for the south hemisphere, where the majority of human cases are reported today.

We conclude that salt is one of the factors contributing to the maintenance of plague foci in North Africa and Eurasia. L-form of Y. pestis may persist in these salt areas. Based on maps of salt in the north hemisphere, it may be possible to focus the surveillance of enzootic plague around salt sources in order to optimize the prevention of human plague in these countries.

Materials and methods

Analyzing co-localization of plague foci and chotts. In a first step, chotts were localized in the five North African countries Mauritania, Morocco,

Algeria, Tunisia and Lybia by using the Map Atlas on line

(http://www.unesco.org/languages-atlas/fr/atlasmap.html). Then plague outbreaks reported in these five countries over 75 years [Malek, unpublished data] were plotted. Coordinates corresponding to the middle of every focus were used for statistical analysis. Distance between each water pond and plague foci was estimated according to geographical coordinates.

For each water pond (either salt or non-salt water), the shortest distance

(minimum distance to plague foci) and the mean distance were estimated.

Halo-tolérance et révélation des foyers de peste |75 Salt and non-salt water categories were compared using a non-parametric

Wilcoxon test. Statistical analyses were performed using R3.1.3 software

(Copyright 2015 The R Foundation for Statistical Computing, Vienna,

Austria). In a second step, this analysis was extended to the northern hemisphere in areas between the fifteenth and forty-fifth parallel (from

15°N to 50°N). For the study we considered plague foci described in the last

50 years by the World Health Organization (WHO) as well as human cases reported and sylvatic plague, because in some countries no human cases do not mean no plague.

Sampling on an inanimate environment. Based on the assumption of co- localization of plague foci and a salty environment, we conducted a field mission for plague foci sampling in Algeria. We investigated Algerian plague foci described during the 2000s. In Oran (44) from the Sebkha (Aïn

Beida side, Daët el Bagrat and El Kerma side), Oued Chelif, Daiat Morsly,

Gharabas lake, Telamine lake and from Saline of Arzew. In Laghouat (34) from M’Zi oued, whose courses end up in Chott El Melghir. In M’Sila (45) from Chott el Hodna (Maarif side, Souamaa side and Ain el Khadra). In

Biskra (45) from Oumache, Ourelal, Bouchegroune and Tolga. Salinity was tested using a refractory technique (automatic salinity refractometer, Fisher

Scientific, Strasbourg, France).

76|Halo-tolérance et révélation des foyers de peste

Yersinia pestis strains. Y. pestis Algeria1 strain CSUR (Collection de

Souches de l’Unité des Rickettsies) P100, an Orientalis biotype isolate made from a rodent in Algeria (34), was used in this study. In addition, a

Y. pestis Algeria3 isolate was made directly from one natural soil specimen collected at Chott el Hodna, Algeria (Supplementary Figure 3).

Manipulations of Y. pestis were performed in a biosafety level 3 laboratory.

Yersinia cells were streaked from a frozen culture stock onto 5% sheep- blood agar (bioMérieux, La Balme-les-Grottes, France) for 48 hours at

28 °C. As for the experiment, Yersinia cells were grown in two tubes, each containing 10 mL trypcase soy broth (0.5% of sodium chloride in this medium; Becton-Dickinson, Grenoble, France) to the early stationary growth phase at 108 colony-forming units (CFU)/mL (OD600 = 2; Biolog model 21907 S/N 0744984)). One control tube and the second tube supplemented with 10g/L enzyme and endotoxin-free NaCl (Carlo Erba,

Mundolsheim, France) were incubated at 28 °C for 5 days. Then, Yersinia cells were subcultured on 5% sheep-blood agar (without NaCl addition)

(ref 43041, bioMérieux, Marcy l’Etoile, France) to check for cell viability and enumerate the colonies. Colony enumeration was done after the inoculum had been serially diluted 1:10 to 10E-6 dilution and digital pictures of colonies were recorded. An average was calculated for dilutions up to about 10 to 300 colonies. The counting CFU/mL is expressed in

Halo-tolérance et révélation des foyers de peste |77 log10. Also, acridine orange (Becton-Dickinson, Grenoble, France) staining was performed. Slides were washed, mounted with Fluoprep (bioMérieux) and examined using an optical microscope (x1,000) (Leica) and under the

Leica DM2500 Upright Fluorescence Microscope at x1,000 magnification.

The identification of all colonies was confirmed using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) and a Microflex system (Brücker Daltonics, Wissembourg, France) as previously described (14). In parallel, 5 mL of the inoculated salt-culture medium were mixed with 5 mL of a salt-trypcase soy broth solution to increase the concentration at 2% w/v NaCl. The remaining volume was kept to verify the viability of cells under these conditions by subcultures every

48 hours. This operation was repeated up to a final NaCl concentration of

15% w/v NaCl (150 g/L). Also, Y. pestis cells were directly subcultured into trypticase-soy broth containing either 5% w/v, 10% w/v or 15% w/v NaCl final concentration, incubated for 7 days and subcultured on sheep-blood agar every 48 hours. DNA was extracted from the strain using the QIAamp

DNA Mini Kit (Qiagen) according to the manufacturer’s protocol. PCR sequencing of partial yopT, caf1, ymT and pla Y. pestis genes (Table 3) was carried out in an Applied Biosystems 2720 thermal cycler (MJ Research) to demonstrate that the strain had kept its three plasmids: pFRa, pPla and pYV.

Experiments were done in triplicate.

78|Halo-tolérance et révélation des foyers de peste

Supplementary Table 3. Primers and probes used for PCR.

Electron microscopy. Control Y. pestis cells subcultured without NaCl and Y. pestis cells exposed to 150 g/L NaCl for 7 weeks were examined by electron microscopy using both negative staining and inclusion. Observations were made using a Morgagni 268D microscope (FEI, Philips, France) operating at X35-X280000 magnification. Negative stains were made by contrasting samples with a solution of 1% ammonium molbydate. Inclusions were performed as previously described (46) except that sample fixation was done in a 2% glutaraldehyde. For each grid, a minimum of 150 Y. pestis cells were observed, and a minimum of five grids were observed for each condition.

Halo-tolérance et révélation des foyers de peste |79 2D-DIGE analysis. We performed the Two-Dimensional (2D) Difference Gel Electrophoresis technique for separating complex mixtures of proteins for Y. pestis control and Y. pestis at the maximum level of salinity to visualize the effect of salt on proteome. The four replicates of each strain condition were prepared as in the standard 2D-PAGE using the2-D Clean- Up Kit. Sample preparation for 2D-DIGE: purified bacteria were resuspended in solubilization buffer (30 mM Tris, 7 M urea, 2 M thiourea and 4 % (w/v) 3- [(3 cholamidopropyl)-dimethylammonio]-1-propanesulfonate (CHAPS) and disrupted by sonication (three times for 60 s at power = 20 W without pulsing). Cell debris was removed by centrifugation (12,000 x g, 4 °C, 10 min), and soluble proteins were precipitated using the Plus One 2- DClean-Up Kit (GE Healthcare, Chalfont St. Giles, UK). The final pellet was resuspended in solubilization buffer. The protein concentration was determined using the Bio-Rad protein assay kit (Bio-Rad). After this step, the pH of the sample was adjusted if needed to 8.5 (the appropriate pH for sample labeling). For 2D-DIGE analysis and protein identification, four replicates from each medium were labeled with cyanine dyes (Cy3 or Cy5) in a ratio of 400 pmol CyDye to 50 µg of protein, according to the manufacturer’s instructions (GE Healthcare, Chalfont St Gilles, UK). An internal standard was created by combining equal amounts of protein from every sample and then labeling with Cy2 using the same ratio. Each sample was labeled for 30 min on ice in the dark and the reaction was quenched by the addition of 1 µl of 10 mM lysine. CyDye-labeled samples were combined during 2-D gel electrophoresis so that each gel contained a Cy2, a Cy3 and a Cy5 labeled sample. Two dimensional gel electrophoresis was carried out as previously described (47). After electrophoresis, gels were

80|Halo-tolérance et révélation des foyers de peste scanned at appropriate wavelengths to cyanines using the Typhoon FLA 9000 Imager according to the manufacturer’s protocol (GE Healthcare). Scans were acquired at 100 µm resolution. Images were cropped with Image Quant TM software (GE Healthcare) and further analyzed using the Progenesis SameSpots software version 4.0.3779 from Nonlinear Dynamics (Newcastle upon Tyne, UK) as described by the manufacturers. To determine significant differences in 2D spot abundance, an ANNOVA score (p-value) lower or close to 0.05 and a change of at least 1.5-fold between the wild type Y. pestis and the salt stress Y. pestis protein spots were required for spots to be selected for digestion and identification by MS analysis as previously described (47).

MS protein identification. After protein fractionation and relative quantity measurement by 2D-DIGE, all proteins from gel spots were reduced then alkylated and finally in-gel digested. The MALDI-TOF-MS spectra peaks lists were compared to Y. pestis protein sequences with our internal Mascot search engine (IHU).

Halo-tolérance et révélation des foyers de peste |81 References

1. Raoult D, Mouffok N, Bitam I, Piarroux R, Drancourt M (2013) Plague: history and contemporary analysis. J Infect 66:18-26.

2. Yersin A (1894) La peste bubonique à Hong Kong. Ann Inst Pasteur 2:428-430.

3. Morelli G, et al. (2010) Phylogenetic diversity and historical patterns of pandemic spread of Yersinia pestis. Nat Genet 42:1140–1143.

4. Schmid BV, et al. (2015) Climate-driven introduction of the Black Death and successive plague reintroductions into Europe. Proc Natl Acad Sci U S A 112:3020-3025.

5. Rasmussen S, et al. (2015) Early Divergent Strains of Yersinia pestis in Eurasia 5,000 years ago. Cell. 163:571-582.

6. Achtman M, et al. (2004) Microevolution and history of the plague bacillus, Yersinia pestis. Proc Natl Acad Sci U S A 101:17837–17842.

7. Drancourt M, et al. (2007) Yersinia pestis Orientalis in remains of ancient plague patients. Emerg Infect Dis 13:332-333.

8. Wiechmann I, Grupe G (2005) Detection of Yersinia pestis DNA in two early medieval skeletal finds from Aschheim (Upper Bavaria, 6th century A.D.). Am J Phys Anthropol 126:48-55.

9. Drancourt M, et al. (2004) Genotyping, Orientalis-like Yersinia pestis, and plague pandemics. Emerg Infect Dis 10:1585-1592.

10. Wagner DM, et al. (2014) Yersinia pestis and the Plague of Justinian 541—543 AD: a genomic analysis. Lancet Infect Dis 14:319-326.

11. Bos KI, et al. (2011) A draft genome of Yersinia pestis from victims of the Black Death. Nature 478:506-510.

12. Blanc G, Baltazard M (1941) Recherches expérimentales sur la peste. L'infection du pou de l'homme : Pediculus corporis de Geer. Maroc Médical N° 216, p. 39.

82|Halo-tolérance et révélation des foyers de peste

13. Houhamdi L, Lepidi H, Drancourt M, Raoult D (2006) Experimental model to evaluatethe human body louse as a vector of plague. J Infect Dis 194:1589-1596.

14. Ayyadurai S, Sebbane F, Raoult D, Drancourt M (2010) Body lice, Yersinia pestis Orientalis, and Black Death. Emerg Infect Dis 16:892- 893.

15. Drancourt M, Raoult D (2014) Yersinia pestis and the three plague pandemics. Lancet Infect Dis 14:918-919.

16. Drali R, et al. (2014) Bartonella quintana in body lice from scalp hair of homeless persons, France. Emerg Infect Dis 20:907-908.

17. Drali R, Mumcuoglu KY, Raoult D. ―Human Lice in paleoentomology and paleomicrobiology‖ in Didier Raoult & Michel Drancourt (eds) "Paleomicrobiology of humans" by ASM Press. Washington DC, 2016. 18. Blanc G, Baltazard M (1941) Recherches expérimentales sur la peste. L'infection le la puce de l'homme : Pulex irritans L. Maroc Médical N° 217, p. 81.

19. World Health Organization. Comité OMS expert de la peste, quatrième rapport. Genève, 1970.

20. Weniger BG, et al. (1984) Human bubonic plague transmitted by a domestic cat scratch. JAMA 251:927-928.

21. Bin Saeed AA, Al-Hamdan NA, Fontaine RE (2005) Plague from eating raw camel liver. Emerg Infect Dis 11:1456-1457.

22. Eisen R J, et al. (2008) Persistence of Yersinia pestis in soil under natural conditions. Emerg Infect Dis 14:941-943.

23. Cabanel N, et al. (2013) Plague Outbreak in Libya, 2009, Unrelated to Plague in Algeria. Emerg Infect Dis 19(2):230-236.

24. Baltazard M (1964) The conservation of plague in inveterate foci. J Hyg Epidemiol Microbiol Immunol VIII: 409-421.

Halo-tolérance et révélation des foyers de peste |83 25. Drancourt M, Houhamdi L, Raoult D (2006) Yersinia pestis as a telluric, human ectoparasite-borne organism. Lancet Infect Dis 6(4):234-241.

26. Bazanova LP, Maevskiĭ MP, Khabarov AV (1997) [An experimental study of the possibility for the preservation of the causative agent of plague in the nest substrate of the long-tailed suslik]. Med Parazitol (Mosk) 4:37-39. Russian. PubMed PMID: 8587510.

27. Karimi Y (1963) Natural preservation of plague in soil. Bull Soc Pathol Exot Filiales 56:1183–1186.

28. Ayyadurai S, et al. (2008) Long-term persistence of virulent Yersinia pestis in soil. Microbiology 154(Pt 9):2865-2871.

29. Mollaret HH (1963) Experimental preservation of plague in soil [in French]. Bull Soc Pathol Exot Filiales 56:1168-1182.

30. Pouget M (1980) Les relations sol-végétation dans les steppes Sud- algéroises. Thèse Doc Es sciences, Université Aix-Marseille, p.555.

31. Bertherat E, et al. (2007) Plague reappearance in Algeria after 50 years, 2003. Emerg Infect Dis 13:1459-1462.

32. Tarantola A, Mollet T, Gueguen J, Barboza P, Bertherat E (2009) Plague outbreak in the Libyan Arab Jamahiriya. Euro Surveill 14(26).

33. Malek MA, Bitam I, Drancourt M (2015) Plague in Arab Maghreb, 1940-2015: a review (submitted data).

34. Bitam I, et al. (2010) New rural focus of plague, Algeria. Emerg Infect Dis 16(10):1639-1640.

35. Klieneberger-Nobel E (1960) L-forms of bacteria. Academic Press, New York, NY (1):361-386.

36. Pollitzer R (1952) Plague studies. 5. Methods for laboratory diagnosis. Bull World Hlth Org 6:317-349.

84|Halo-tolérance et révélation des foyers de peste

37. Bibel DJ, Chen TH (1976) Diagnosis of plague: an analysis of the Yersin-Kitasato controversy. Bacteriol Rev 40:633-651.

38. Rowland S (1914) The morphology of the plague bacillus. J Hyg (Lond) 13(Suppl):418-422.

39. Zykin LF (1994) [The results of a 15-year study of the persistence of Yersinia pestis L-forms]. Zh Mikrobiol Epidemiol Immunobiol. Suppl 1:68-71. Russian.

40. Lister IM, Raftery C, Mecsas J, Levy SB (2012) Yersinia pestis acrAB-tolC in Antibiotic Resistance and Virulence. Antimicrob Agents Chemother 56:1120-1123.

41. Gao H, et al. (2011) Phenotypic and transcriptional analysis of the osmotic regulator OmpR in Yersinia pestis. BMC Microbiology 11:39.

42. Minato Y, et al. (2013) Na+/H+ antiport is essential for Yersinia pestis virulence. Infect Immun 81:3163-3172.

43. Herz K, Vimont S, Padan E, Berche P (2002). Roles of NhaA, NhaB, and NhaD Na+/H+ Antiporters in Survival of Vibrio cholerae in a Saline Environment. J Bacteriol 185:1236-1244.

44. Bitam I, Baziz B, Rolain JM, Belkaid M, Raoult D (2006) Zoonotic focus of plague, Algeria. Emerg Infect Dis 12:1975-1977.

45. Malek MA, Hammani A, Beneldjouzi A, Bitam I (2014) Enzootic plague foci, Algeria. New Microbes New Infect 4:13-16.

46. Adékambi T, et al. (2004) Amoebal coculture of "Mycobacterium massiliense" sp. nov. from the sputum of a patient with hemoptoic pneumonia. J Clin Microbiol 42:5493-5501.

47. Boyer M, et al. (2009) Giant Marseillevirus highlights the role of amoebae as a melting pot in emergence of chimeric microorganisms. Proc Natl Acad Sci U S A 106:21848-21853.

Halo-tolérance et révélation des foyers de peste |85

Conclusion & perspectives

7. Conclusion et perspectives

La peste, zoonose contractée essentiellement au contact des rongeurs, est une maladie qui sévit toujours de nos jours en Afrique, Asie et Amérique et fait partie des maladies actuellement ré-émergentes dans le monde. Au cours du XXème siècle, l’utilisation de traitements antibiotiques et le renforcement des mesures de santé publique ont réduit très fortement la morbidité et la mortalité dues à cette maladie, mais n’ont pas permis de la faire disparaître. En effet, sa principale capacité à "s’éteindre" pendant plusieurs années avant de réapparaître brutalement sous forme épidémique reste un phénomène mal compris. Nous avons précédemment confirmé la persistance tellurique Y. pestis sous forme virulente pendant au moins

9 mois dans un sol stérilisé humidifié et dépourvu de tout supplément nutritionnel (Ayyadurai2008). Ces résultats ont été confirmés par une

équipe américaine par l'observation de la persistance de Y. pestis dans un sol naturellement infecté (Eisen2008). Notre travail de thèse a permis de mettre en évidence trois foyers de peste qui étaient alors inconnus en Algérie. La surveillance des animaux sentinels nous a permis de confirmer l’existence du danger épidémique présent dans l’Oranie et à Laghouat (dernières régions à déclarer une peste humaine) (Bitam2010) mais également de décrire à nouveau les foyers « négligés » de Biskra, M’Sila et de Cap Djinet

Conclusions et perspectives |87 au littoral (Malek2015). Avec une description du rongeur Apodemus sylvaticus comme nouvellement porteur de peste sans signes cliniques apparants. Cette géolocalisation conduit donc à une hypothèse confirmée relatant la proximité des foyers de peste aux environnements salins (Malek, unpublished data), appelés chotts dans notre zone d’étude. Après l’isolement sur terrain d’une souche de Y. pestis Algeria 3 dans un tel environnement, l’étude expérimentale a confirmé la résistance de la bactérie aux milieux à forte concentration de NaCl avec adaptation par changement de conformation et d’expression protéique. Il s’agit de résultats préliminaires quant aux mécanismes de survie tellurique de Y. pestis dans son environnement naturel. Notre thèse est donc que la persistance de

Y. pestis dans le sol salé est un des déterminants des foyers de peste au

Maghreb.

Notre travail de thèse ouvre des perspectives de recherche quant au mode de transmission à partir du sol salé, en recherchant son association aux végétaux vivant dans ces environnements salins et consommés par les camélidés. Dans cette perspective, il serait intéressant de reprendre le mécanisme de contamination par voie orale qui, comme nous l’avons décrit, n’est pas négligeable. Les résultats de tels travaux permettraient probablement de compléter le cycle de la peste sylvatique au Maghreb et de guider la surveillance.

88|Conclusions et perspectives Références

8. Références

Abramov VM, Vasiliev AM, Khlebnikov VS, Vasilenko RN, Kulikova NL, Kosarev IV, Ishchenko AT, Gillespie JR, Millett IS, Fink AL & Uversky VN. 2002. Structural and functional properties of Yersinia pestis Caf1 capsular antigen and their possible role in fulminant development of primary pneumonic plague. J. Proteome Res. 1(4):1307-15. Sebbane F, Clayton OJ, Gardner D, Long D & Hinnebusch BJ. 2009. The Yersinia pestis caf1M1A1 Fimbrial Capsule Operon Promotes Transmission by Flea Bite in a Mouse Model of Bubonic Plague. Infect Immun. 77(3):1222–9.

Achtman M, Zurth K, Morelli G, Torrea G, Guiyoule A & Carniel E. 1999. Yersinia pestis, the cause of plague, is a recently emerged clone of Yersinia pseudotuberculosis. Proc Natl Acad Sci U S A. 96(24):14043-8. Erratum in: Proc Natl Acad Sci U S A 2000;97(14):8192.

Ayyadurai S, Houhamdi L, Lepidi H, Nappez C, Raoult D & Drancourt M. 2008. Long-term persistence of virulent Yersinia pestis in soil. Microbiology. 154:2865-71.

Bacot AW & Martin CJ. 1914. Observations on the mechanism of the transmisison of plague by fleas. Journal of Hygiene. 13:423–9.

Bacot AW. 1915. LXXX. Observations on the length of time that fleas (Ceratophyllus fasciatus) carrying Bacillus pestis in their alimentary canals are able to survive in the absence of a host and retain the power to re-infect with plague. J Hyg (Lond). 14(Suppl):770-3.

Références |89 Bacot AW. 1915. LXXXI. Further notes on the mechanism of the

transmission of plague by fleas. J Hyg (Lond). 14(Suppl):774-6.

Bercovier H, Mollaret HH, Alonso J-M, Brault J, Fanning GR, Steiger AG & Brenner DJ. 1980a. Intra- and Interspecies Relatedness of Yersinia pestis by DNA Hybridization and Its Relationship to Yersinia pseudotuberculosis. Current Microbiology 4: 207–12.

Biraben, J-N, & Le Goff J. 1975. The plague in the early middle ages. In R. Forster and O. Ranum (ed.), Biology of man in history. The Johns Hopkins University Press, Baltimore, Md. 48–80.

Bitam I, Ayyadurai S, Kernif T. 2010. New rural focus of plague, Algeria. Emerg Infect Dis. 16:1639–40.

Brubaker R R. 1972. The genus Yersinia: biochemistry and genetics of virulence. Curr. Top. Microbiol. Immunol. 57:111–158.

Burroughs AL. 1947. Sylvatic plague studies: The vector efficiency of nine species of fleas compared with Xenopsylla cheopis. J Hyg (Lond). 45(3):371-96.

Butler T. 1995. Yersinia species. Principles and Practice of Infectious Diseases. New York, NY: Churchill Livingstone. 2070-78.

Butler, T. 1983. Plague and other Yersinia infections. Plenum Press, New York, N.Y.

Carniel, E, Mazigh, D & Mollaret, H. H. 1987. Expression of iron- regulated proteins in Yersinia species and their relation to virulence. Infect. Immun. 55, 277–80.

90 | Références Caulfield AJ, Walker ME, Gielda LM & Lathem WW. 2014. The Pla protease of Yersinia pestis degrades fas ligand to manipulate host cell death and inflammation. Cell Host Microbe. 15(4):424-34.

Chain PSG, Carniel E, Larimer FW, Lamerdin J, Stoutland PO, Regala WM, Georgescu AM, Vergez LM, Land LM, Motin VL, Brubaker RR, Fowler J, Hinnebusch J, Marceau M, Medigue C, Simonet M, Chenal-Francisque V, Souza B, Dacheux D, Elliott JM, Derbise A, Hauser LJ & Garcia E. 2004. Insights into the evolution of Yersinia pestis through whole-genome comparison with Yersinia pseudotuberculosis. Proc Natl Acad Sci U S A. 101(38):13826-31.

Chanteau S, Rahalison L, Duplantier JM, Rasoamanana B, Ratsitorahina M, Dromigny JA, Laventure S, Duchemin JB, Boisier P, Rabeson D, Roux J. 1998. [Update on plague in Madagascar] [Article in French]. Med Trop . 58(2 Suppl):25-31.

Chouikha I, Charrier L, Filali S, Derbise A & Carniel E. 2010. Insights into the infective properties of YpfΦ, the Yersinia pestis filamentous phage. Virology. 407(1):43-52.

Chouikha I, Hinnebusch BJ. 2012. Yersinia--flea interactions and the evolution of the arthropod-borne transmission route of plague. Curr Opin Microbiol. 15(3):239-46.

Christie AB. 1980. Chen TH, Elberg SS. Plague in camels and goats: their role in human epidemics. J Infect Dis. 141:724–6.

Coppes JB. 1980. Plague in pregnancy. J Reprod Med. 25(2):91-5.

Références |91 Czeizel AE , Rockenbauer M. 1997. Teratogenic study of doxycycline. Obstet Gynecol. 89(4):524-8.

Derbise A, Chenal-Francisque V, Pouillot F, Fayolle C, Prévost MC, Médigue C, Hinnebusch BJ & Carniel E. 2007. A horizontally acquired filamentous phage contributes to the pathogenicity of the plague bacillus. Mol Microbiol. 63(4):1145-57.

Devignat R. 1951. Variétés de l'espèce Pasteurella pestis; nouvelle hypothèse. Bull World Health Organ.4(2):247–63.

Doll JM, Zeitz PS, Ettestad P, Bucholtz AL, Davis T & Gage K. 1994. Cat- transmitted fatal pneumonic plague in a person who traveled from Colorado to Arizona. American Journal of Tropical Medicine and Hygiene 51: 109–14.

Du Y, Rosqvist R & Forsberg A. 2002. Role of fraction 1 antigen of Yersinia pestis in inhibition of phagocytosis. Infect. Immun. 70(3):1453-60.

Eisen RJ, Petersen JM, Higgins CL, Wong D, Levy CE, Mead PS, Schriefer ME, Griffith KS, Gage KL, Beard CB. 2008. Persistence of Yersinia pestis in soil under natural conditions. Emerg Infect Dis. 6:941-3.

Galimand M, Guiyoule A, Gerbaud G, Rasoamanana B, Chanteau S, Carniel E, & Courvalin P. 1997. Multidrug resistance in Yersinia pestis mediated by a transferable plasmid. N. Engl. J. Med. 337:677- 80.

92 | Références Gottfried, R. S. 1983. The black death. Natural and human disaster in medieval europe. The Free Press, New York, N.Y.

Guerrant R, Walker D, Weller P. 2011. eds. Tropical Infectious Diseases. 3rd ed. New York: Saunders Elsevier. 276-83.

Guiyoule A, Gerbaud G, Buchrieser C, Galimand M, Rahalison L, Chanteau S, Courvalin P & Carniel E. 2001. Transferable plasmid- mediated resistance to streptomycin in a clinical isolate of Yersinia pestis. Emerg. Infect. Dis. 7:43-48.

Haffkine WM. 1906. Les inoculations antipesteuses. Bulletin de l’Institut Pasteur 4: 1–16

Hinnebusch BJ, Erickson DL. 2008. Yersinia pestis biofilm in the flea vector and its role in the transmission of plague. Curr Top Microbiol Immunol. 322:229-48.

Inglesby TV, Henderson DA, O'Toole T & Dennis DT. 2000. Safety precautions to limit exposure from plague-infected patients. JAMA. 284(13):1648-9.

Malek MA, Hammani A, Beneldjouzi A & Bitam I. 2015. Enzootic plague foci, Algeria. New Microbes New Infect. 4:13–6.

Michiels T, Wattiau P, Brasseur R, Ruysschaert J M & Cornelis G. (1990) Secretion of Yop proteins by Yersiniae. Infect Immun. 58(9): 2840– 2849.

Références |93 Mollaret HH (1963a) Conservation expérimentale de la peste dans le sol. [Experimental conservation of plague in the soil]. Bulletin de la Société de pathologie exotique 56: 1168–82.

Mollaret HH, Karimi Y, Eftekhari M & Baltazard M. 1963b. La peste de fouissement. Burrowing plague. Bulletin de la Société de pathologie exotique 50: 1186–93

Perry RD, Fetherston JD. 1997. Yersinia pestis--etiologic agent of plague. Clin Microbiol Rev. 10(1):35-66.

Raoult D, Mouffok N, Bitam I, Piarroux R & Drancourt M. 2013. Plague: history and contemporary analysis. J Infect. 66(1):18-26.

Raynaud L. «La peste en Algérie: l’épidémie de peste dans la région d’Alger. Cas de peste survenus dans la colonie de 1899 à 1924». 1924. Archives de l’Institut Pasteur d’Algérie. 303–61.

Sebbane F, Devalckenaere A, Foulon J, Carniel E & Simonet M. 2001. Silencing and reactivation of urease in Yersinia pestis is determined by one G residue at a specific position in the ureD gene. Infect Immun. 69(1):170–6.

Sebbane F, Jarrett CO, Gardner D, Long D & Hinnebusch BJ. 2006. Role of the Yersinia pestis plasminogen activator in the incidence of distinct septicemic and bubonic forms of flea-borne plague. Proc Natl Acad Sci USA. 103(14):5526–30.

94 | Références Simond M, Godley ML, Mouriquand PD. 1998. Paul-Louis Simond and his discovery of plague transmission by rat fleas: a centenary. J R Soc Med. 91(2):101-4.

Smith MD, Vinh SX, Hoa NT, Wain J, Thung D, White NJ. 1995. In vitro antimicrobial susceptibilities of strains of Yersinia pestis. Antimicrob Agents Chemother. 39:2153-54.

Sodeinde OA, Subrahmanyam Y, Stark K, Quan T, Bao YD & Goguen JD. 1992. A Surface Protease and the Invasive Character of Plague. Science 258: 1004–7.

Sun YC, Jarrett CO, Bosio CF & Hinnebusch BJ. 2014. Retracing the evolutionary path that led to flea-borne transmission of Yersinia pestis. Cell Host Microbe. 15(5):578-86. van Loghem JJ. 1944. The classification of the plague- bacillus. Antonie van Leeuwenhoek Journal of Microbiology and Serology 10: 15–16.

Yersin A. 1894. La peste bubonique à Hong-Kong. Ann Inst Pasteur. 8:662-7.

Yersin A. 1897. Sur la peste bubonique (séro-thérapie). Annales de l’Institut Pasteur 11: 81–93.

Références |95

Résumé Yersinia pestis est l'agent causal de la peste, une maladie à transmission vectorielle enzootique infectant les rongeurs et leurs puces, l’Homme étant un hôte accidentel. Y. pestis est classée par les Centers for Diseases Control comme un agent potentiel de bioterrorisme [http://www.bt.cd.gov/agent/plague]. Ce bacille persiste toujours dans la nature maintenu par un cycle enzootique dans des foyers de peste, ce qui conduit à la réémergence de la maladie en cas de proximité des populations humaines et ce jusqu’à plusieurs décennies plus tard. Il est donc important pour la communauté médicale de connaitre ces foyers afin d’établir le bon diagnostic en présence d’un cas de peste, prescrire le traitement efficace et surtout limiter la propagation au reste de la population.

Dans notre travail nous nous sommes intéressés aux pays d’Afrique du Nord où une réémergence a eu lieu après des années de ‘silence’ et ce à trois reprises dans les années 2000. Nous avons, dans un premier temps, répertorié dans une revue les différents épisodes ayant eu lieu ainsi que le nombre de cas sur six pays du Maghreb: Mauritanie, Maroc, Algérie, Tunisie, Libye et Egypte sur une période de 75 années à compter de 1940 en mettant en évidence l’importation de la maladie dans ces pays puis un mode de contamination négligé dans la littérature, à savoir la transmission par voie orale de Y. pestis. Cette revue nous a conduits à travailler ensuite en Algérie pour mettre en évidence des foyers de peste encore inconnus et de confirmer l’existence du risque dans ceux déjà décrits. Nous y avons mené une étude rodontologique sur 237 micromammifères provenant de 11 régions du Nord Algérien et avons ainsi confirmé deux foyers (Mascara, Laghouat) et mis en évidence trois nouveaux foyers (M’Sila, Biskra et Cap Djinet) porteurs d’un nouveau génotype (déterminé par Multi Spacer Typing) de Y. pestis de biotype Orientalis. Ce même travail nous a permis d’ajouter l’espèce Apodemus sylvaticus à la liste des rongeurs pestiférés.

La projection des foyers de peste ainsi actualisés sur une carte géographique et écologique du Maghreb nous a conduits à observer que les foyers de peste étaient situés à proximité de points d’eau saumâtre. Une étude statistique a confirmé une corrélation significative entre foyer de peste/eau salée en révélant une proximité minimale <3 km en comparaison à des zones d’eau douce. Sur cette base, une nouvelle campagne de prélèvements d’échantillons environnementaux salés en Algérie a permis l’isolement d’une souche Y. pestis Algeria 3 d’un sol salé. Cette découverte a été confortée par l’observation expérimentale de la résistance de Y. pestis à un milieu hyper-salé à 150g/L NaCl se traduisant par un protéome spécifique en réponse à ce stress et mettant en évidence une forme d’adaptation de type forme L de la bactérie dans ce type d’environnement inanimé. Notre travail éclaire de façon originale un facteur méconnu de persistance tellurique de Y. pestis, conditionnant la réémergence de la peste dans des foyers séculaires au Maghreb contrairement aux rivages Nord de la Méditerranée où la peste autochtone a disparu depuis un siècle.