Pediculus Humanus Corporis by Use of Melting Curve Analysis Genotyping

AIX-MARSEILLE UNIVERSITE ÉCOLE DOCTORALE DES SCIENCES DE LA VIE ET DE LA SANTÉ FACULTÉ DE MÉDECINE DE MARSEILLE Unité de Recherche sur les Maladies Infectieuses et Tropicales Emergentes (URMITE), UMR 63, CNRS 7278, IRD 198, Inserm 1095

Thèse présentée pour obtenir le grade universitaire de docteur Discipline : Pathologie Humaine Spécialité : Maladies Infectieuses

Rezak DRALI

Poux humains : Différenciation, distribution phylogéographique, Host-Switching et contrôle

Soutenue le 15 Décembre 2014 devant le jury :

Mr le Professeur Pierre MARTY Président du Jury/Rapporteur Mr le Docteur Arezki IZRI Rapporteur Mr le Professeur Philippe BROUQUI Directeur de thèse Mr le Professeur Didier RAOULT Co-directeur de thèse

À Tassadit, mon épouse. Merci pour ton soutien indéfectible À ma mère À mes enfants Taous, Mokrane et Ania.

Sommaire

Avant-propos 7 Résumé 8 Abstract 11 Introduction générale 15 Revue: Taxonomy, lifestyle and current genetic advances of primate’s lice. 23 Chapitre 1: Différenciation pou de tête - pou de corps 59 Article I: Distinguishing Body Lice from Head Lice by Multiplex 63 Real-Time PCR Analysis of the Phum_PHUM540560 Gene. Article II: Bartonella quintana in body lice from scalp hair 75 of homeless persons, France. Article III: Detection of Bartonella quintana in African 81 Body and Head Lice. Chapitre 2: Distribution phylogéographique des poux 91 humains contemporains et anciens. Article IV: A New Clade of African Body and Head Lice Infected 95 by Bartonella quintana and Yersinia pestis – Democratic Republic of the Congo. Article V: Studies of ancient lice reveal unsuspected past 107 migrations of vectors. Article VI: Evidence of sympatry of Clade A and Clade B head lice 119 in a pre-Columbian Chilean mummy from Camarones. Chapitre 3: Host-Switching 127 Article VII: Host switching of human lice to New World monkeys in Amazonia 131

5 Chapitre 4: Détection et monitoring de la résistance moléculaire 161 des poux de corps à la perméthrine Article VIII: Detection of a knockdown resistance mutation 165 associated with permethrin resistance in the body louse Pediculus humanus corporis by use of melting curve analysis genotyping. Article IX: Effect of permethrin-impregnated underwear on 177 body lice in sheltered homeless persons: a randomized controlled trial. Conclusions et perspectives 189 Références 191 Annexes 201 Article X: Matrix-Assisted Laser Desorption Ionization– 203 Time of Flight Mass Spectrometry for Rapid Identification of Tick Vectors. Mini review: Typhus in World War I. 213 Remerciements 219

6 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 thèse utilisé dans le Nord de l’Europe et qui permet un meilleur rangement que les thèses traditionnelles. Par ailleurs, la partie introduction et bibliographie est remplacée par une revue envoyée dans un journal afin de permettre une évaluation extérieure de la qualité de la revue et de permettre à l’étudiant de commencer le plus tôt possible une bibliographie 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.

Professeur Didier RAOULT

7 Résumé

Les poux hématophages de primates sont des ectoparasites spécifiques de leurs hôtes avec qui ils ont Co évolué depuis environ 25 millions d’années. Ceux infestant l’Homme sont sans doute les mieux étudiés, particulièrement le pou de tête (Pediculus humanus capitis) et le pou de corps (Pediculus humanus humanus). Ce sont deux écotypes indiscernables occupant chacun une niche écologique: les cheveux pour le pou de tête et les vêtements pour le pou de corps. La pédiculose due au pou de tête touche chaque année des centaines de millions d'enfants à travers le monde nonobstant la classe sociale à laquelle ils appartiennent alors que le pou de corps infeste spécialement les populations n’ayant pas un accès facile aux conditions standard d’hygiène, telles les sans-abri, les prisonniers et les réfugiés de guerre. Le pou de corps représente une menace réelle pour l’Homme en raison de son rôle de vecteur dans la transmission de trois maladies graves ayant tué des millions de personnes à travers l’histoire de l’humanité à savoir : le typhus épidémique, la fièvre des tranchées et la fièvre récurrente causées par Rickettsia prowazekii,

Bartonella quintana, et Borrelia recurrentis respectivement. Le pou

8 de corps est également soupçonné dans la transmission d'un quatrième agent pathogène fatidique, Yersinia pestis, l'agent de la peste.

L'analyse de ADN mitochondrial a permis de classer les poux humains en trois clades A, B et C où seul le clade A distribué mondialement comprend à la fois des poux de tête et des poux de corps.

Durant cette thèse, nous avons voulu apporter des réponses à un certain nombre de questions restées posées à travers le traitement de certaines thématiques qui nous semblaient importantes. Ainsi nous avons cherché à (i) trouver un moyen pour différencier entre le pou de tête et le pou de corps, (ii) en savoir d’avantage sur la distribution phylogéographique mondiale des poux humains, (iii) étudier les poux anciens pour nous aider à comprendre la circulation des vecteurs, la circulation des agents pathogènes transmis par ces vecteurs et les flux migratoires des hôtes de ces vecteurs, (iv) étudier le phénomène de

Host-Switching à travers P. mjobergi le poux de singe du Nouveau

Monde et (v) trouver un moyen efficace de contrôle des poux de corps infestant les sans-abri à Marseille à travers la détection et le monitoring de la résistance des poux aux insecticides.

Nous avons obtenu des résultats concrets dans chacune des thématiques abordées, résultats par ailleurs valorisés par des

9 publications scientifiques. En effet, nous avons (i) mis en place un outil moléculaire qui permet de différencier pour la première fois entre le pou de tête et le pou de corps qui a montré efficacité sur le terrain,

(ii) mis en évidence l’existence d’un nouveau clade mitochondrial

(Clade D) renfermant des poux de tête et des poux de corps susceptible de vectoriser B. quintana et Y. pestis, (iii) retracé les migrations humaines à travers l'analyse de poux anciens provenant de différentes périodes et localisations, (iv) démontré pour la première fois que P. mjobergi est génétiquement proche du pou humain et confirmé l’hypothèse qu’à l’origine P. mjobergi était un pou humain qui a été transféré aux singes du Nouveau Monde par les premiers

Hommes à avoir atteint le continent américain il y a des milliers d’années et (v) mis en place un outil de détection et de contrôle de la résistance moléculaire des poux à la perméthrine. Cet outil fut particulièrement utile dans l'étude clinique que nous avons menée pour déterminer si l'utilisation de sous-vêtements imprégnés d'insecticide offrait une protection efficace à long terme contre les poux de corps infestant les personnes sans-abri.

Mots clés : pou de tête, pou de corps, différenciation, distribution phylogéographique, poux anciens, Host-Switching, contrôle des poux.

10 Abstract

Primate sucking lice are obligate host-specific parasites that have Co evolved with their hosts for over 25 million years. Lice infesting humans are probably the best studied, particularly head louse

(Pediculus humanus capitis) and body louse (Pediculus humanus humanus). These are two indistinguishable ecotypes each occupying an ecological niche: hair for head louse and clothing for the body louse. Pediculosis due to head louse affects each year hundreds of millions of children worldwide despite the social class to which they belong while body louse infests especially populations that have not ready access to standard conditions of hygiene, such as the homeless, prisoners and war refugees. Body louse represents a real threat to humans because of its role as vector for the transmission of three deleterious diseases that have killed millions of people, namely epidemic typhus, trench fever and relapsing fever caused by Rickettsia prowazekii, Bartonella quintana, and Borrelia recurrentis respectively. The body louse is also suspected in the transmission of a fourth fateful pathogen, Yersinia pestis, the agent of plague.

Mitochondrial DNA allowed the classification of human lice in three

11 clades designed A, B and C where only clade A that is distributed worldwide comprises both head and body lice.

During my PhD, some thematic that seemed important have been addressed. Thus we aimed to (i) find a way to differentiate between human head and body lice, (ii) learn more about the worldwide phylogeographic distribution of lice (iii) study of ancient human lice which can help to understand the circulation of vectors, the flow of vector-borne pathogens and the migratory flu hosts of these vectors,

(iv) study the host-switching phenomenon through P. mjobergi the lice that parasite New World monkeys and (v) find an efficient way to control body lice infesting homeless people in Marseille.

In each of these issues, we obtained concrete results that have led to scientific publications. Indeed, we (i) implemented a molecular tool to differentiate for the first time between head and body louse, (ii) we highlighted the existence of a fourth mitochondrial clade (Clade D) comprising head and body lice that can vectorize B. quintana and Y. pestis, (iii) we traced human migration through the analysis of ancient lice from different periods and different area, (iv) we demonstrated for the first time that P. mjobergi is genetically close to human louse and confirmed the hypothesis that initially P. mjobergi was a human louse

12 has been transferred to New World monkeys by the first humans who have reached the American continent thousands of years ago and (v) we have implemented a tool for detecting and monitoring the molecular resistance to permethrin of body lice that parasite sheltered homeless persons in Marseille. This tool was particularly useful in the clinical study we conducted to determine whether the use of long- lasting insecticide–treated underwear provides effective long-term protection against body lice in homeless persons.

Keywords: head louse, body louse, differentiation, phylogeographic distribution, ancient lice, host-switching, lice control.

Introduction générale

Les poux (Insecta: Phthiraptera) sont des ectoparasites obligatoires des oiseaux et des mammifères. A ce jour, environ 4.900 espèces ont été répertoriées et réparties dans quatre sous-ordres: d’une part les poux mâcheurs (chewing / biting lice) comprenant les Rhynchophthirina, les Ischnocera et les Amblycera et d’autre part les anoploures

(Anoplura) suceurs de sang [1]. Les données morphologiques [2] combinées aux données moléculaires [3] ont montré qu'il y avait une relation para phylétique entre les poux mâcheurs et les anoploures. La récente découverte de deux fossiles de poux, le premier vieux de 44 millions d’années parasitait les oiseaux [4] et le second 100 millions d’années appartenaient à la famille Liposcelididae [5] a permis d’estimer l’âge des poux à plusieurs millions d’années. Les méthodes de datation moléculaire ont démontré que les poux ont survécu à la grande extinction des espèces du Crétacé-Paléogène (K-Pg) survenue il y’a 65 millions d’années [6].

Aptères et ne possédant pas d’hôte intermédiaire, les poux ont dû Co

évoluer et s’adapter au microenvironnement de leur hôte. De ce fait, la transmission des poux s’effectue lors des contacts directes entre hôtes conspécifiques [7].

15 Le processus de cospeciation hôtes-parasites est exemplifié à travers la relation des poux avec leurs hôtes [8], cependant ce processus n’est pas toujours parfait comme énoncé dans la règle de Fahrenholz qui

édicte que les phylogénies des parasites et de leurs hôtes doivent se refléter l’une et l’autre [9]. Parfois, cette association est marquée par une série d'événements historiques tels le changement d'hôte “host- switching” ou encore la duplication du parasite [10,11]. A juste titre, le phénomène de host-switching a été rapporté deux fois chez les primates. D’abord en 1938, lorsque Ewing a suggéré que Pediculus mjobergi le pou des singes du Nouveau Monde était à l'origine un pou humain (P. humanus) transmis par les premiers Hommes ayant traversé le détroit de Béring il y'a quelques milliers d'années [12].

Puis, en 2007, lorsque Reed et al. suggérèrent que l’Homme avait hérité de Pthirus pubis, le pou du pubis, à travers les contacts directs ayant eu lieu entre gorilles et hominidés archaïques il y’a environ 3 millions d’années [13].

Les poux du genre Pediculus infestant l’Homme sont les mieux

étudiés en raison notamment de l’intérêt médical qu’ils suscitent. Ce genre fut initialement établit par Linnaeus en 1758. Deux décennies plus tard en 1778, De Geer proposa les nomenclatures Pediculus

16 capitis de Geer 1778 et Pediculus corporis de Geer 1778 pour différencier le pou de tête du pou de corps [14]. Hopkins en 1952 a proposé une nomenclature qu’il souhaitait finale eu égard au foisonnement d’appellations données aux poux: Pediculus humanus humanus Linnaeus, 1758 pour le pou de corps et Pediculus humanus capitis de Geer 1778 pour le pou de tête [15].

Le pou de tête vit et se multiplie sur les cheveux alors que le pou de corps occupe une niche écologique différente, à savoir les plis et les coutures de vêtements [14]. La pédiculose due au pou de corps est très répandue parmi les populations précaires telles les pauvres, les sans- abri, les prisonniers et les réfugiés de guerre [16], tandis que les poux de tête affectent des centaines de millions d'écoliers à travers le monde indépendamment des conditions d'hygiène causant du prurit et une perte de sommeil dans certains cas [17]. Plusieurs études comparatives basées sur des critères morphologiques ont été réalisées sans pour autant avoir réussi à différencier entre le pou de tête et le pou de corps.

Dès 1919, Nuttall concluait déjà que les deux écotypes appartenaient à la même espèce puisque les deux écotypes ne présentaient aucune différence concernant les points essentiels [18].

17 Durant un laps de temps, la pigmentation des poux fut en débat. En

1917, Hindle à partir de ses expériences de croisement/sélection avait conclu que la pigmentation des poux était un caractère héréditaire [19].

Hypothèse réfutée par Nuttall en 1919 puis Ewing en 1926 qui considéraient que la pigmentation des poux adultes était due à la couleur du fond (background) utilisée durant les différents stade de développement larvaire [20,21]. Plus récemment, une étude comparant les données phénotypiques et génotypiques des poux d’origine

Africaine avait montré qu’il n’existait aucune concordance entre la couleur et le génotype des poux [22].

L’analyse moléculaire des gènes mitochondriaux a permis d’inférer une classification phylogéographique robuste de Pediculus humanus.

En effet, les poux humains du genre Pediculus ont été classés dans trois clades A, B et C où seul le Clade A distribué à travers le monde comprend aussi bien des poux de tête que des poux de corps [23]. Le

Clade B comprenant des poux de tête retrouvés jusqu’ici sur le continent américain, en Europe de l’ouest, en Australie et en Algérie.

Le Clade C a été retrouvé au Népal, en Ethiopie et au Sénégal [24].

L’analyse moléculaire de gènes mitochondriaux des poux récupérés sur des momies précolombiennes a montré que la présence des Clades

18 A et B sur le continent Américain était antérieure à l'arrivée des colons européens [25,26]. Ce qui renforce l’hypothèse selon laquelle le Clade

B a une origine Américaine qui par la suite a été redistribué dans le vieux monde par les colons ayant regagné l’Europe [24].

La taille prédictive du génome de P. humanus (108 Mb) faisait de lui un excellent candidat au séquençage du génome [27]. En 2010,

Kirkness et al. [28] publiaient une première version du génome du pou de corps et de son endosymbiont Candidatus Riesia pediculicola. Ce génome renferme 10.773 gènes codants parmi lesquels 163 étaient spécifiques à P. humanus [28].

La comparaison du profil transcriptionnel des poux de tête et de corps avait révélé que les deux écotypes avaient le même nombre de gènes excepté le gène phum_PHUM540560 absent chez le pou de tête [29] .

A ce jour, seul le pou de corps est reconnu comme vecteur de trois maladies délétères ayant tué des millions de personnes à travers l’histoire de l’humanité qui sont le typhus épidémique, la fièvre des tranchées et la fièvre récurrente causées par Rickettsia prowazekii,

Bartonella quintana et Borrelia recurrentis respectivement [30]. Le pou de corps est également soupçonné dans la transmission d'un quatrième agent pathogène mortel, Yersinia pestis, l'agent de la peste

19 [31,32]. Au cours des dernières années, l'ADN de B. quintana a été détecté chez les poux de tête de Clade A [16,32–34] et de Clade C

[35,36], l'ADN de B. recurrentis chez les poux de tête de Clade C [37].

Avant l'avènement des pédiculicides, le contrôle des poux était basé sur les moyens physiques tel le peignage, l’épouillage à la main, le rasage et le blanchiment de vêtements [38]. Au cours des dernières décennies, différentes molécules chimiques ont été développées pour lutter contre la pédiculose. Malheureusement les poux ont pu développer une résistance contre la plupart de ces molécules [39].

L’ivermectine semble donner de bons résultats dans le traitement des poux de tête [40] bien qu’une résistance potentielle à cette molécule eut été démontrée dans des conditions de laboratoire [41].

Durant cette thèse nous avons voulu apporter notre contribution dans le domaine de la recherche sur les poux humains. Ainsi, nous avons commencé par rédiger une revue de littérature intitulée « The lice of primates » pour donner un aperçu sur le positionnement phylogénétique des poux humains au sein des anoploures.

Nous avons organisé ce manuscrit de thèse autour de 4 chapitres couvrant chacun une thématique différente à savoir (1) la différenciation entre le pou de tête et le pou de corps, (2) la

20 distribution phylogéographique des poux humains contemporains et anciens, (3) étude du phénomène de Host-Switching et (4) la détection et le monitoring de la résistance moléculaire des poux de corps à la perméthrine.

Revue de littérature: Taxonomy, Lifestyle and Current Genetic Advances of Primate’s Lice

Revue proposée au journal Annual Review of Entomology

1 Taxonomy, Lifestyle and Current Genetic Advances of Primate’s Lice

3 Rezak Drali1 and Didier Raoult1

5 1Aix Marseille Université, URMITE, UM63, CNRS 7278, IRD 198, Inserm 1095,

6 27 Boulevard Jean Moulin 13005 Marseille, France

8 Corresponding author: Raoult, D. (mailto:[email protected])

24 Keywords: lice, primates, co evolution, host-switching 1

25 25 Abstract

26 The relationship between lice (Insecta: Phthiraptera) and their host, birds and mammals, is

27 very old, it dates back to tens of millions of years. Thus, their status as obligate parasite

28 makes lice remarkable markers of the evolution of their hosts. On the other hand, among the

29 4,900 species of lice identified, only the human body louse Pediculus humanus humanus, is

30 known to be a vector of three serious diseases that killed millions of people throughout the

31 history of mankind. Here we reviewed the various studies that targeted lice parasitizing

32 primates, with particular attention to human lice of the genus Pediculus. We also provided an

33 overview on the current taxonomy of primates.

49 2

26 50 Lice

51 Lice (Phthiraptera) are permanent obligate parasitic insects that infest birds and mammals.

52 They are about 4,900 known species distributed into four suborders: chewing or biting lice

53 including Rhynchophthirina, Ischnocera and Amblycera; sucking lice grouped under the name

54 of Anoplura (Johnson et al., 2004). The morphological data (Lyal, 1985a) supported by

55 molecular evidences (Barker et al., 2003) showed that there was a paraphyletic relation

56 between the chewing lice and Anoplura . The discovery of two rare fossils allowed estimating

57 the age of lice to about tens millions of years. Indeed, the first fossil that parasitized birds was

58 44 million years and the second was 100 million years and belongs to family Liposcelididae,

59 commonly called book-louse, a close relatives to parasitic lice (Grimaldi and Engel, 2006).

60 Molecular dating methods showed that the radiation of the louse suborders started before the

61 mass extinctions of species 65 million years ago (Ma), corresponding to Cretaceous-

62 Paleogene (K-Pg) boundary (Smith et al., 2011).

63 Because they are wingless and have no intermediate hosts, lice coevolved with their host to

64 adapt to its microenvironment and the transmission of lice occurs mostly through direct

65 contact between conspecific hosts (Barker, 1991). Lice are probably the most successful

66 pattern of host-parasite cospeciation (Hafner and Nadler, 1988), however, the cospeciation

67 process is not always perfect as stated in Farenholz's rule which enacts that host and parasite

68 phylogenies should mirror each other (Brooks, 1979). Occasionally this association is marked

69 by a range of historical events such as host switching, parasite duplication, sorting event and

70 failure of the parasite to speciate in response to host speciation (Johnson et al., 2003;Page and

71 Charleston, 1998).

72 Rightly in primates, the host-switching phenomenon has been described at least twice. Firstly

73 in 1938 when Ewing suggested that Pediculus mjobergi the louse parasitizing New World

74 monkeys was originally a human louse (P. humanus) transmitted to the monkey by the first 3

27 75 humans who crossed the Bering Strait it there's thousands of years (Ewing, 1938). Hypothesis

76 we have also confirmed thanks to the molecular analysis of howler monkey lice from

77 Argentina – South America (Drali, unpublished data). Then in 2007 when Reed et al. based

78 on molecular and phylogenetic analyzes suggested that the presence of the genus Pthirus in

79 humans would be the result of host-switching from archaic gorillas to archaic hominids

80 roughly 3 Mya throughout direct contact between them (Reed et al., 2007).

81 Characteristics

82 Lice are wingless, dorsoventrally flattened insects (Price and Graham, 1997).

83 Hemimetabolous, they pass through three nymphal stages before becoming adults (Lyal,

84 1985a). The mouthparts consist of mandibles for chewing lice allowing them to feed upon the

85 skin (feathers, fur, and squamae) and sometimes the blood of their hosts, and stylet (piercing

86 mouthparts) for sucking lice (Johnson and Clayton, 2003;Light et al., 2010). Initially, the

87 sucking lice ancestors lived free in burrows and nests of vertebrates and have chewing

88 mouthparts (Light et al., 2010;Lyal, 1985a). The development of mouthparts allowing some

89 of these lice to take blood meals has created a dependence on their hosts and thus have

90 become obligate parasites (Light et al., 2010). The tibiotarsal claws of sucking lice have

91 adapted for grasping host hairs (Durden, 2001).

92 Unlike the chewing lice, that have large heavily sclerotized heads that are as wide as the

93 prothorax, sometimes wider, the sucking lice have heads that are narrower than their

94 prothorax. Yet, Rhynchophthirina belonging to chewing lice group constitute an exception

95 with their mandibles that are worn at the end of a long snout which enables them to penetrate

96 the thick skin of their host to feed (Johnson and Clayton, 2003).

97 Classification

98 Phthiraptera would be derived from original Pscopteran-like ancestor is divided into two

99 groups the chewing lice and the sucking lice (Johnson et al., 2004). About 4400 species

28 100 compose the group of chewing lice which includes three major families: Amblycera (1182

101 species of bird lice and 162 species of mammal lice), Ischnocera (2683 species of bird lice

102 and 377 species of mammal lice) and Rhynchophthirina (3 species of mammal lice conﬁned

103 to elephants and pigs) (Cruickshank et al., 2001). The Anoplura group comprises about 540

104 species that parasitize mammals exclusively, two-third are parasites of rodents (Durden and

105 Musser, 1994b). Concerning the primates (Mammalia), of the 78 genera recognized, including

106 our own species (Finstermeier et al., 2013), 61.7% are parasitized by lice. Among these,

107 53.3% are parasitized by sucking lice, 13.3% by chewing lice and 5% by both sucking and

108 chewing lice (Light, 2005).

109 Behaviors

110 Lice usually do not leave their hosts without which they can only survive a short period (two

111 days) (Barker, 1994;Johnson and Clayton, 2003). The females lay eggs on the fur, hair or

112 feathers of their hosts at a suitable temperature until hatching for avoiding desiccation and

113 other physical factors (Price and Graham, 1997). Lice move to conspecific host when the

114 latter is in close contact, but exceptionally they can cling to the abdomen of haematophagous

115 flying insects to reach to another host (phoresis) (Durden, 1990;Harbison and Clayton, 2011).

116 Usually, chewing lice parasitize only a single species of hosts at a time (Johnson and Clayton,

117 2003).

118 Ecology

119 Of the 4400 species of chewing lice identified, 550 are hosted by mammals (Cruickshank et

120 al., 2001). The primates are parasitized by four genera of chewing lice among which three

121 belong to Ischnocera and the fourth belongs to Amblycera (Table 1).

122 Sucking lice are about 540 known species distributed in 50 genera and 15 families that spread

123 worldwide (Light et al., 2010).

29 124 Six genera of sucking lice were found in primates. Three belong to the family Polyplacidae

125 and the three other genera, each belonging to a different family, namely Pediculidae,

126 Pedicinidae and Pthiridae (Table 1) (Durden and Musser, 1994b;Price and Graham, 1997).

127 Among the 29 orders of mammals that are distributed in 153 families, 1,200 genera and 5,400

128 species (Wilson and Reeder, 2005), eight orders are not known to harbor sucking lice

129 including Monotremata, Marsupialia, Xenarthra, Pholidota, Chiroptera, Cetacea, Proboscidea,

130 and Sirenia (Kim, 2006).

131 Phylogeny of lice

132 The molecular phylogeny of lice was inferred after the analysis of 18S rRNA sequences in 33

133 species (Barker et al., 2003). This analysis showed that Amblycera is the sister-group to all

134 other lice but the Rhynchophthirina is sister to the Anoplura. Ischnocera is sister to the

135 Rhynchophthirina and Anoplura, which gives the following configuration (Amblycera

136 (Ischnocera (Anoplura, Rhynchophthirina))) (Barker et al., 2003) (Figure 1).

137 Primates (Mammalia)

138 Primates, term meaning "prime, first rank", represent an order of mammals, including

139 humans, apes, monkeys and prosimians (Wilson and Reeder, 2005). Arboricoles for the most

140 part, primates live in tropical forests. There are nevertheless partially terrestrial species such

141 as baboons or fully terrestrial species such as geladas and humans (Reed and Fleagle, 1995).

142 Until recently, the taxonomy of primates has undergone many changes. Today based on

143 mitogenomic phylogeny, primates including our own species comprise 480 species distributed

144 in 78 genera originated from a common ancestor during the Cretaceous/Paleocene boundary

145 approximately 80 to 90 Mya (Finstermeier et al., 2013;Perelman et al., 2011). Three suborders

146 compose primate’s order: Strepsirrhini, Tarsiiformes and Simiiformes (Perelman et al., 2011).

147

148

30 149 Strepsirrhini

150 Strepsirrhini include Lorisiformes (galagos, pottos, and lorises), Chiromyiformes (Malagasy

151 aye-aye) and Lemuriformes (Malagasy lemurs) (Roos et al., 2004). Insectivores, Strepsirrhini

152 are provided with a tail covered with fur, they have an elongated hairless and wet nose and

153 round eyes adapted to nocturnal activities. They have large mobile ears, sensitive tactile hairs

154 and a strong sense of smell (Mittermeier et al., 1999). Strepsirrhini characterizes mainly from

155 other primates by the presence of a toothcomb to the front of their teeth that consists of four

156 incisors and two canines, all elongated and facing forward. The dental comb is used to

157 retrieve the gum trees on which they feed, but also for grooming/delousing (Seiffert et al.,

158 2003). The habitat of Strepsirrhini overlaps two continents, Africa and Asia. Lorises are found

159 in equatorial Africa and Southeast Asia, the galagos in sub-Saharan African forests and the

160 Lemurs are endemic in Madagascar (Mittermeier et al., 1999). However, based on the results

161 obtained by the analysis of short interspersed elements (SINEs) combined with complete

162 mitochondrial cytochrome b sequences from all recognized strepsirrhine genera, Roos and

163 collaborators conclude that strepsirrhines originated from Africa and that Madagascar and

164 Asia were colonized by respective single immigration events (Roos et al., 2004).

165 Tarsiiformes (tarsiers)

166 Tarsiiformes are small nocturnal primates, including three genera (Tarsius, Cephalopachus

167 and Carlito) allopatrically distributed in the islands of Southeast Asia (Groves and Shekelle,

168 2010). Their weight varies between 80 and 150 g, they have round heads with very large eyes

169 that are directed forward and very mobile ears (Ankel-Simons, 2000).

170 For a long time, the position of Tarsiiformes relative to Strepsirrhini and Anthropoidea in the

171 phylogeny of primates was unclear (Yoder, 2003). Finally, the mitochondrial genome analysis

172 allowed to resolve the issue by showing the subdivision of primates into infra-orders

31 173 Strepsirrhini and Haplorhini, with tarsiers as sister group of Anthropoidea (Finstermeier et al.,

174 2013).

175 Simiiformes

176 Composed of Platyrrhini (New World monkeys) and Catarrhini that include Cercopithecoidea

177 (Old World monkeys) and Hominoidea (great apes and gibbons) (Perelman et al., 2011).

178 Platyrrhini

179 Platyrrhini comprises 5 families, 6 subfamilies, 19 genera, and 199 species that occupy large

180 areas in Central and South America (Rylands and Mittermeier, 2009). Callitrichidae, Cebidae

181 (Cebinae and Saimiriinae), Aotidae, Pitheciidae (Pitheciinae and Callicebinae) and Atelidae

182 (Alouattinae and Atelinae) compose the five families (Rylands and Mittermeier, 2009).

183 The shape of their nose earned them the name of Platyrrhini meaning "flatted nose" (Laitman,

184 2011). Their body size is extremely variable (about a factor of 100): 120 grams for the pygmy

185 marmoset (Cebuella pygmaea) and 10-12 kg for the muriqui (Brachyteles arachnoides) and

186 gray woolly monkey (Lagothrix cana) (Di Fione and Campbell, 2007). Platyrrhini are

187 provided with a long prehensile tail that can be controlled and used to hold onto branches

188 (Schmitt et al., 2005). According to performed studies, Platyrrhini separated from Catarrhini

189 during the Eocene about 45 million years ago (Finstermeier et al., 2013;Perelman et al.,

190 2011). The arrival of Platyrrhini the in the American continent remains ambiguous however,

191 the phylogeny resolves the relative divergence pattern among families members from a

192 common ancestor about 20 million years (Hodgson et al., 2009).

193 Catarrhini

194 Catarrhini includes Cercopithecoidea (Old World monkeys) and Hominoidea (great apes and

195 gibbons) (Perelman et al., 2011).

196

197

32 198 Cercopithecoidea

199 Named Old World monkeys in opposition to the New World monkeys, Cercopithecoidea are

200 distributed mainly in Africa and Asia (Rowe et al., 1996). As for Platyrrhini, they owe their

201 name to the form of their nose whose openings and directed downwardly (Groves, 1993).

202 Cercopithecidae includes two extant subfamilies, Colobinae (the leaf-eating monkeys) and

203 Cercopithecinae (the cheek-pouch monkeys) (Groves, 1993). Colobines are arboreal and

204 possess a long nonprehensile tails while cercopithecines have well-developed thumbs and tails

205 of varying lengths (Rowe et al., 1996). The body size of cercopithecines can varies from 1 kg

206 for talapoin (Miopithecus talapoin) to over 37 kg for the large olive baboon (Papio

207 hamadryas anubis) (Rowe et al., 1996).

208 Cercopithecidae comprises 21 genera. 10 belong to subfamily Colobinae in which 7 have an

209 Asian distribution (Nasalis, Simias, Presbytis, Semnopithecus, Trachypithecus, Pygathrix and

210 Rhinopithecus) and 3 have an African distribution (Colobus, Piliocolobus and Procolobus).

211 11 genera belong to subfamily Cercopithecinae that are endemic to Africa with exception of

212 the genus Macaca which is now mainly found in Asia (Allenopithecus, Cercocebus,

213 Cercopithecus, Chlorocebus, Erythrocebus, Lophocebus, Macaca, Mandrillus, Miopithecus,

214 Papio and Theropithecus) (Wilson and Reeder, 2005).

215 Old World monkeys separated from hominoids about 32 Ma and diverged into the subfamilies

216 Cercopithecinae and Colobinae in the Early Miocene (Finstermeier et al., 2013;Perelman et

217 al., 2011).

218 Hominoidea

219 Hominoidea includes two families: Hominidae (great apes) and Hylobatidae (gibbons, small

220 apes) (Perelman et al., 2011). Four extant genera compose the family Hominidae:

221 chimpanzees (Pan), gorillas (Gorilla), orangutans (Pongo), and humans (Homo) (Hirai et al.,

222 2012). Orangutans live in Asia, chimpanzees and gorillas in Africa and humans are distributed

33 223 worldwide (Rowe et al., 1996). Bipedalism is certainly the most characteristic feature of the

224 Hominidae, though not widely used by African apes which prefer to move on knuckles

225 (Richmond and Strait, 2000).

226 In 1990s, the analysis of whole mitochondrial sequences has allowed showing that humans

227 have a sister relationship with chimpanzees among Hominidae and therefrom, the phylogeny

228 of the Hominidae has been established (Horai et al., 1995).

229 The family Hylobatidae is composed from four genera (Hylobates, Hoolock, Nomascus, and

230 Symphalangus) (Groves, 2005). They live in Southeast Asia (India, China, Malay Peninsula,

231 Java, Borneo, and Sumatra) (Rowe et al., 1996). Given their lower size relative to the other

232 Hominoidea, the name lesser apes was given to the gibbons (Hirai et al., 2012). The four

233 genera of family Hylobatidae are phylogenetically divergent. Indeed, each genus of gibbons is

234 monophyletic (Roos and Geissmann, 2001).

235 Hominoids separated into gibbons (Hylobatidae) and great apes and humans (Hominidae) in

236 the Early Miocene (23 to 5 Mya). Within Hylobatidae, Nomascus separated first (7.8 Mya),

237 followed by the divergence of Symphalangus and Hylobates 6.2 Mya.

238 In Hominidae, orangutans (Pongo) diverged 15.2 MYA from the African great apes and

239 humans, while Gorilla separated from the Homo + Pan Clade 8.4 MYA. Finally, chimpanzees

240 and humans separated in the Latest Miocene, about 5.9 MYA (Finstermeier et al.,

241 2013;Perelman et al., 2011).

242 Lice of primates

243 Sucking lice

244 Genus Pediculus

245 Pediculus includes three known species that parasite primates: P. schaeffi the louse of

246 chimpanzees, P. mjobergi the louse of New World monkeys and P. humanus the louse of

247 humans (Durden and Musser, 1994b).

34 248 P. schaeffi was described for the first time in 1910 by Fahrenholz thanks to specimens found

249 in monkeys from a zoo in Hamburg, Germany (Durden and Musser, 1994b). P. schaeffi can

250 be found on any part of the body of their host but with a preference for head, groin and

251 armpits (Allen et al., 2013). By using a molecular clock analysis of mtDNA performed studies

252 showed that P. schaeffi diverged from its sister taxon P. humanus about 5.5 million years ago

253 (Kittler et al., 2003;Reed et al., 2004).

254 New World monkeys also harbored louse of the Pediculidae family (Ferris, 1916). Ferris gave

255 the name P. mjobergi to this species in reference to Mjoberg who was the first to describe it in

256 1910 (Ferris, 1951). In a study we have just completed, we reported the case of transfer of

257 human lice to the New World monkeys. Indeed, the morphological examinations and genetic

258 analyzes performed on the P. mjobergi lice collected from the howler monkeys Alouatta

259 caraya have supported these findings (Drali, unpublished Data).

260 Humans, chimpanzees and New World monkeys are not the only ones to harbor lice of the

261 genus Pediculus, in 1951, Ferris inventoried the siamang (Hylobates syndactylus) among the

262 hosts of Pediculus humanus (Ferris, 1951).

263 Pediculus humanus

264 Of lice infesting primates, those parasitizing humans are the best studied particularly head and

265 body lice, Pediculus humanus capitis and Pediculus humanus humanus respectively (Durden

266 and Musser, 1994b).

267 Taxonomy

268 Linnaeus in 1758 was the first who established the genus Pediculus. In 1778, De Geer

269 proposed the nomenclatures P. capitis de Geer 1778 and P. corporis de Geer 1778 to

270 discriminate between head and body louse (De Geer, 1778). Different terminologies have

271 been proposed to name the two variants: P. cervicalis Latreille 1803 and P. consobrinus

272 Piaget 1880 for head louse, P. vestimenti Nitzsch 1818 and P. tabescentium Alt 1824 for body 11

35 273 louse (Nuttall, 1917). In 1952 Hopkins suggested to give the final names: Pediculus humanus

274 humanus Linnaeus, 1758 to the body louse and Pediculus humanus capitis de Geer, 1778 to

275 the head louse (Hopkins, 1952).

276 Head and body lice ecotypes

277 The classification and nomenclature of human lice genus Pediculus derived from their

278 ecology. Indeed, the head louse lives, breeds and lays its eggs at the base of the hair of the

279 head while the body louse occupies a different ecological niche namely garments where it

280 lays its eggs in the seams and folds (De Geer, 1778).

281 Several comparative studies based on morphological criteria were conducted without reaching

282 differentiates between head lice and body lice. Early as 1919, Nuttall concluded that the two

283 ecotypes represent the extremes in the variation of the same species since they are

284 undistinguishable in all essential point of structure (Nuttall, 1919c).

285 For a while, pigmentation of lice was under debate. In 1917, Hindle starting from its breeding

286 experiments concluded that the pigmentation in lice is an inherited character (Hindle, 1917).

287 Assumption refuted by Nuttall in 1919 relayed later by Ewing in 1926 who considered that

288 the pigmentation of adult lice is due to the color of background used during nymphal

289 development (Ewing, 1938;Nuttall, 1919b). More recently, a study comparing the phenotypic

290 and genotypic data of human lice from Africa showed that no congruence between louse color

291 and genotype has been identified (Veracx et al., 2012).

292 Pediculosis due to the body louse is prevalent in precarious populations such as poor

293 populations, homeless, prisoners and war refugees (Sangare et al., 2014), while head lice

294 affects hundreds of millions of schoolchildren worldwide regardless hygienic conditions with

295 symptoms that include itching and in some cases loss of sleep (Chosidow et al., 1994).

296

297

36 298 Mitochondrial clades

299 Thanks to mitochondrial genes, a robust phylogeographic classification of Pediculus humanus

300 was inferred. Indeed, human lice are distributed into 3 clades designed Clade A, B and C

301 where only Clade A that is spread worldwide includes both head and body lice (Reed et al.,

302 2004). Clade B comprises head lice found in the Americas, Western Europe, Australia and

303 North Africa and Clade C including head lice was found in Nepal, Ethiopia and Senegal

304 (Boutellis et al., 2014). Analysis of mitochondrial gene of Pre-Columbian mummies lice

305 showed that Clades A and B were previously present on the Americas before the coming of

306 European settlers (Boutellis et al., 2013a;Raoult et al., 2008). This supports an American

307 origin for Clade B and a dispersal in the Old World with European colonists returning to

308 Europe (Boutellis et al., 2014). Recently, we described a fourth clade, Clade D comprising

309 head and body lice in Democratic Republic of the Congo (Figure 2).

310 Genome of Pediculus humanus

311 Until very recently (Kelley et al., 2014), Pediculus humanus had the smallest genome sizes

312 known in insects (ca. 108 Mb) at that time (Johnston et al., 2007) what made him an excellent

313 candidate for sequencing. In 2010, Kirkness et al. (Kirkness et al., 2010) published a first

314 version of the body louse genome, and its primary bacterial endosymbiont Candidatus Riesia

315 pediculicola. This genome contained 10,773 predicted protein-coding genes, 163 of which are

316 specific to it (Kirkness et al., 2010). It contains 58 microRNAs genes (Kirkness et al.,

317 2010;Olds et al., 2012). Transposable elements represent only 1% of the genome and GC

318 percentage is very low, about 28%, which would explain the small size of this genome

319 (Kirkness et al., 2010). Thirty seven P450s were found in louse, far of the number of P450

320 found in other insects, an advantage to explore the insecticide resistance in human lice (Lee et

321 al., 2010;Yoon et al., 2011). P. humanus is equipped with a complete genetic repertoire

322 enable him to carry out its essential biological functions, but compared to Drosophila 13

37 323 melanogaster, some genes associated with sensing and responding to the environment are in

324 reduced number (Kirkness et al., 2010).

325 Mitochondrial genome

326 Unlike bilateral animals from which the 37 mitochondrial (mt) genes are arranged on a single

327 circular chromosome, the head and body lice have their mt genes on 20 minichromosomes

328 respectively. Each minichromosome has a size of three to four kilobases and contains one to

329 three genes in addition to a large non coding region that includes a control region (Shao et al.,

330 2012). This organization would be due to the lack of the mitochondrial single-stranded

331 binding protein (mtSSB) involved in mitochondrial genome replication (Kirkness et al.,

332 2010).

333 Genome of Candidatus Riesia pediculicola endosymbiont

334 The genome of Candidatus Riesa pediculicola comprises less than 600 genes distributed

335 between a linear chromosome and plasmid that includes specifically genes required for the

336 synthesis of pantothenate, an indispensable vitamin deﬁcient in the louse diet (Kirkness et al.,

337 2010).

338 Recently, the sequencing of the genome of the P. schaeffi endosymbiont namely Candidatus

339 Riesa pediculischaeffi and its comparison to the genome of Candidatus Riesia pediculicola

340 showed that the two species diverged roughly 5.4 Mya ago (Boyd et al., 2014).

341 Head and body louse transcriptome

342 The comparison of the body and head louse transcriptome revealed that the two ecotypes had

343 the same numbers of genes except one gene missing in the head louse (PHUM540560) (Olds

344 et al., 2012) . The refined analysis of partial sequence of this gene in both body and head lice

345 showed that the gene was also present in the head louse but with a rearranged sequence.

346 Based on the difference between the two sequences, a multiplex RT-PCR was developed to

347 differentiate between the two ecotypes (Drali et al., 2013). The tool developed was

38 348 particularly effective in field since it able to determine that lice populations collected on the

349 head and clothing of the same homeless in Marseille had a body louse genotype (Drali et al.,

350 2014).

351 Vector competence of human lice

352 So far, body louse is the only known that can transmit at least three deleterious diseases that

353 have killed millions of people, namely epidemic typhus, trench fever and relapsing fever

354 caused by Rickettsia prowazekii, Bartonella quintana, and Borrelia recurrentis respectively

355 (Raoult and Roux, 1999). It is also suspected in the transmission of a fourth lethal pathogen,

356 Yersinia pestis, the agent of plague (Houhamdi et al., 2006;Piarroux et al., 2013). In recent

357 years, the DNA of B. quintana was found in head lice belonging to Clade A (Bonilla et al.,

358 2009;Boutellis et al., 2012;Piarroux et al., 2013;Sangare et al., 2014) and Clade C (Angelakis

359 et al., 2011;Sasaki et al., 2006). DNA of B. recurrentis found in head lice Clade C (Boutellis

360 et al., 2013b) and DNA of Y. pestis was detected in head lice of Clade A and D (Figure 2).

361 Treatment

362 Before the advent of pediculicides, the louse control was mainly by physical means such

363 combing, louse and nit removal, shaving and laundering of clothing (Mumcuoglu and Zias,

364 1988).

365 In recent decades, different chemical molecules with various mechanism of action have been

366 developed for the treatment of pediculosis. Unfortunately lice developed a resistance against

367 most of these molecules namely organochloride (DDT and Lindane), synthetic pyrethroids

368 (Permethrin), organophosphate (Malathion), and Carbamate (Carbaryl), (Clark et al., 2013).

369 Macrocyclic lactone (Ivermectin) seems to be working for the treatment of head lice (Pariser

370 et al., 2013) although a potential resistance to this molecule has been demonstrated under

371 laboratory conditions (Yoon et al., 2011).

39 372 Today, the availability of louse genome opens new perspectives in understanding its biology

373 vector competence which will help to consider new ways to fight against lice infestation.

374 Genus Pedicinus

375 Genus Pedicinus includes 16 species that parasitize 41 species of Old World monkeys

376 distributed among Colobinae and Cercopithecinae (Durden, 2001;Price and Graham, 1997).

377 More than one species of Pedicinus can parasitize different hosts but the opposite is rare

378 (Durden and Musser, 1994a). However, erratic cases where two different species of Pedicinus

379 that parasite the same host species have been reported in Colobinae; in Africa where two non-

380 interbreeding populations of lice were found on both groups of red colobus (Scholl et al.,

381 2012) and in Southeast-Asia on douc langurs (Pygathrix spp.) (Mey, 2010). In both cases, the

382 authors believe that this is due to host-switching that can occur during direct contact between

383 the respective hosts of both species of Pedicinus (Mey, 2010;Scholl et al., 2012).

384 Kuhn in 1967, cited by Price in 1997, reported cases of infestation of White-handed gibbon

385 (Hylobates lar) by Pedicinus eurygaster, the only case reported in literature (Kuhn and

386 Ludwig, 1967;Price and Graham, 1997).

387 Other sucking lice

388 The group of lemurs of Madagascar are hosting three genera of lice (Lemurpediculus,

389 Lemurphthirus and Phthirpediculus) all belonging to the family Polyplacidae (Durden and

390 Musser, 1994a). Comprising 190 species distributed in 20 genera, Polyplacidae is the largest

391 family of Anoplura (Durden and Musser, 1994b). Members of this family parasitize mainly

392 rodents and insectivores (Price and Graham, 1997).

393 Genus Pthirus

394 Pthirus pubis usually known as the crab or pubic louse is the third type of louse that

395 parasitizes humans (Durden and Musser, 1994a). Pt. pubis is prevalent worldwide and is

396 present in all categories of the population. It is sexually transmitted and has often been found

40 397 in combination with sexually transmitted infections (Anderson and Chaney, 2009). Crab louse

398 lives in pubic hair but he can be found also on the eyelashes and / or eyebrows of the children

399 (Chosidow, 2000). The second louse belonging to the family Pthiridae, Pthirus gorillae is the

400 parasite of gorillas (Durden and Musser, 1994a). Then, the legitimate question facing

401 scientists was: why humans and gorillas shared the same genus of lice? The molecular

402 analyzes of both species showed that P. pubis and P. gorillae were sister taxa that diverged 3

403 to 4 million years ago (Reed et al., 2007), more freshly than the gorilla/human-chimp split

404 which is estimated at about 9 million years ago (Wilkinson et al., 2011). The proposed

405 scenario assumes that the Pthirus lineage from archaic gorillas switched to the archaic human

406 who transmitted it to his descendants (Reed et al., 2007).

407 Chewing lice

408 Genus Trichophilopterus

409 This genus contains only one species namely Trichophilopterus babakotophilus for whom

410 Mjöberg (Nuttall, 1919a) established the monotypic family Trichophilopteridae that parasites

411 only the lemurs of Madagascar (Price and Graham, 1997). Trichophilopterus has been

412 classified for a time in the family Philopteridae (parasites of birds) by Ferris (1933), finally

413 taxonomists have found that it was quite different to constitute a family itself (Price and

414 Graham, 1997).

415 Genus Felicola

416 Genus Felicola is composed of 55 species of which 54 are parasites of carnivores. The fifty

417 fifth species is found in Lorisidae, the Asian-African family of primates (Lyal, 1985b;Perez

418 and Palma, 2001).

419 Genus Cebidicola

420 Belonging to the family of Trichodectidae, genus Cebidicola comprises three species found in

421 New World monkeys living in Brazil: C. amiatus found in Ring-tailed monkey (Alouatta 17

41 422 ursina) and Woolly spider monkey (Brachyteles arachnoïdes); C. extrarius found in Red

423 howler monkey; C. semiarmatus found in Rufous-handed howler monkey (Alouatta belzebul),

424 Black howler (Boero and Boerhinger, 1963), Brown howler (Alouatta guariba) and Ring-

425 tailed monkey (Price and Graham, 1997).

426 Genus Aotiella

427 Two species of Aotiella a parasite of South American night monkeys (Aotidae) have been

428 described: A. aotophilus hosted by (Aotus azarai) and A. hershkovitzi found in gray-necked

429 night monkey (Aotus trivirgatus) (Price and Timm, 1995).

430 Co evolution and Host-Switching

431 The partnership hosts - lice parasites old of 65 million years is marked by a succession of

432 events like cospeciation, coevolution, host-switching and adaptation (Barker, 1991;Barker,

433 1994;Hafner and Page, 1995;Smith et al., 2011). For lice, Host-switching is not accidental but

434 rather a vital necessity. This is also the case for blood sucking insects that can feed in

435 fortuitous host when its favorite host becomes scarce or is not accessible (Takken and

436 Verhulst, 2013). Physiological factors like famine and/or abundance of accessible new hosts

437 may favor to host-switching (Takken and Verhulst, 2013).

438 Host-switching phenomenon seems marked by strict rules. Indeed, according to Clayton, lice

439 cannot establish viable populations on novel hosts that differ in size from the native host

440 (Clayton et al., 2003). In real life, this rule does not always respected. Indeed, in 1940s, a

441 colony of human body lice (Orlando strain) was adapted to take blood meals on rabbits

442 (Culpepper, 1944). Cases of infestation with Pediculus and Pedicinus were reported in

443 gibbons we thought no parasitized by lice just like orangutans (Kuhn and Ludwig, 1967;Price

444 and Graham, 1997). New World monkeys have inherited human lice (Ewing, 1938). Indeed,

445 P. mjobergi has well adapted to its new host. It is even likely that recombination events can

446 take place between the two populations of lice on the occasion of their meeting (Drali,

42 447 unpublished data). Humans and gorillas share the louse of genus Pthirus as a result of host-

448 switching from archaic gorillas to archaic hominids roughly 3 Mya throughout direct contact

449 between them (Reed et al., 2007).

450 Concluding remarks

451 The study of lice is essential for understanding complex processes such as coevolution,

452 cospeciation, adaptation and host-switching. Through carrying out this literature review, we

453 realized that little is known about lice parasitizing primates, apart from the human louse of the

454 genus Pediculus. Thus, some primates can harbor lice belonging to different sub-orders. This

455 is the case of Lemuridae, Indridae and Cebidae (Table 1). Whether for humans an hypothesis

456 was proposed to explain his hosting of two genera of sucking lice belonging to two different

457 families (Reed et al., 2004), we do not yet know how certain primates can harbor lice

458 belonging to different sub-orders. Is this the result of a rehosting followed by an adaptation as

459 in P. mjobergi? Further investigations and larger collection of specimens in the field will be

460 necessary to try to answer this question. The sequencing of the genomes could be a real

461 breakthrough in understanding biology, epidemiology and phylogenetic relationship of the

462 different species of lice.

463

464

465

466

467

468

469

470

471

43 472 References Cited

473

474 1. Allen, J. M., C. O. Worman, J. E. Lihgt, and D. L. Reed, 2013, Parasitic Lice Help to 475 Fill in the Gaps of Early Hominid History., in JF Brinkworth and K Pechenkina eds., 476 Primates, Pathogens, and Evolution: New York, Springer, p. 161-186.

477 2. Anderson, A. L., and E. Chaney, 2009, Pubic lice (Pthirus pubis): history, biology and 478 treatment vs. knowledge and beliefs of US college students: Int.J.Environ.Res.Public 479 Health, v. 6, no. 2, p. 592-600.

480 3. Angelakis, E., G. Diatta, A. Abdissa, J. F. Trape, O. Mediannikov, H. Richet, and D. 481 Raoult, 2011, Altitude-dependent Bartonella quintana genotype C in head lice, 482 Ethiopia: Emerg.Infect.Dis., v. 17, no. 12, p. 2357-2359.

483 4. Ankel-Simons, F., 2000, Primate anatomy: an introduction. Academic Press.

484 5. Barker, S. C., 1991, Evolution of host-parasite associations among species of lice and 485 rock-wallabies: coevolution? (J. F. A. Sprent Prize lecture, August 1990): 486 Int.J.Parasitol., v. 21, no. 5, p. 497-501.

487 6. Barker, S. C., 1994, Phylogeny and classification, origins, and evolution of host 488 associations of lice: Int.J.Parasitol., v. 24, no. 8, p. 1285-1291.

489 7. Barker, S. C., M. Whiting, K. P. Johnson, and A. Murrell, 2003, Phylogeny of the lice 490 (Insecta, Phthiraptera) inferred from small subunit rRNA.: Zoologica Scripta, v. 32, 491 no. 5, p. 407-414.

492 8. Boero, J. J., and I. K. Boerhinger, 1963, Reflexiones sobre un nuevo caso de Pediculus 493 mjöbergi en el mono aullador Alouatta caraya.: Rev.Fac.Agr.Vet., v. 15, no. 3, p. 87- 494 98.

44 495 9. Bonilla, D. L., H. Kabeya, J. Henn, V. L. Kramer, and M. Y. Kosoy, 2009, Bartonella 496 quintana in body lice and head lice from homeless persons, San Francisco, California, 497 USA: Emerg.Infect.Dis., v. 15, no. 6, p. 912-915.

498 10. Boutellis, A., L. Abi-Rached, and D. Raoult, 2014, The origin and distribution of 499 human lice in the world: Infect.Genet.Evol., v. 23, p. 209-217.

500 11. Boutellis, A., R. Drali, M. A. Rivera, K. Y. Mumcuoglu, and D. Raoult, 2013a, 501 Evidence of sympatry of clade a and clade B head lice in a pre-Columbian Chilean 502 mummy from Camarones: PLoS.One., v. 8, no. 10, p. e76818.

503 12. Boutellis, A., O. Mediannikov, K. D. Bilcha, J. Ali, D. Campelo, S. C. Barker, and D. 504 Raoult, 2013b, Borrelia recurrentis in head lice, Ethiopia: Emerg.Infect.Dis., v. 19, no. 505 5, p. 796-798.

506 13. Boutellis, A., A. Veracx, E. Angelakis, G. Diatta, O. Mediannikov, J. F. Trape, and D. 507 Raoult, 2012, Bartonella quintana in head lice from Senegal: 508 Vector.Borne.Zoonotic.Dis., v. 12, no. 7, p. 564-567.

509 14. Boyd, B. M., J. M. Allen, V. de Crecy-Lagard, and D. L. Reed, 2014, Genome 510 Sequence of Candidatus Riesia pediculischaeffi, Endosymbiont of Chimpanzee Lice, 511 and Genomic Comparison of Recently Acquired Endosymbionts from Human and 512 Chimpanzee Lice: G3.(Bethesda.).

513 15. Brooks, D. R., 1979, Testing the context and extent of host-parasite coevolution.: 514 SysT.Zool., v. 28, p. 229-307.

515 16. Chosidow, O., 2000, Scabies and pediculosis: Lancet, v. 355, no. 9206, p. 819-826.

516 17. Chosidow, O., C. Chastang, C. Brue, E. Bouvet, M. Izri, N. Monteny, S. Bastuji- 517 Garin, J. J. Rousset, and J. Revuz, 1994, Controlled study of malathion and d- 518 phenothrin lotions for Pediculus humanus var capitis-infested schoolchildren: Lancet, 519 v. 344, no. 8939-8940, p. 1724-1727.

45 520 18. Clark, J. M., K. S. Yoon, S. H. Lee, and B. R. Pittendrigh, 2013, Human lice: Past, 521 present and future control.: Pesticide Biochemistry and Physiology, v. 106, p. 162- 522 171.

523 19. Clayton, D. H., S. E. Bush, B. M. Goates, and K. P. Johnson, 2003, Host defense 524 reinforces host-parasite cospeciation: Proc.Natl.Acad.Sci.U S A, v. 100, no. 26, p. 525 15694-15699.

526 20. Cruickshank, R. H., K. P. Johnson, V. S. Smith, R. J. Adams, D. H. Clayton, and R. D. 527 Page, 2001, Phylogenetic analysis of partial sequences of elongation factor 1alpha 528 identifies major groups of lice (Insecta: Phthiraptera): Mol.Phylogenet.Evol., v. 19, no. 529 2, p. 202-215.

530 21. Culpepper, G. H., 1944, The rearing and maintenance of a laboratory colony of the 531 body louse.: The American journal of tropical medicine and hygiene, v. 1, no. 5, p. 532 327-329.

533 22. De Geer,C. Mémoires pour servir à l'histoire des Insectes. [7], 62-68. 1778. Stokholm, 534 Hesselberg. 535 Ref Type: Serial (Book,Monograph)

536 23. Di Fione, A., and C. J. Campbell, 2007, The Atelines: Variation in ecology, behavior, 537 and social organization., in CJ Campbell, A Fuentes, KC MacKinnon, M Panger, and 538 SK Bearder eds., Primates in Perspective: New York, Oxford University Press, p. 155- 539 185.

540 24. Drali, R., A. Boutellis, D. Raoult, J. M. Rolain, and P. Brouqui, 2013, Distinguishing 541 body lice from head lice by multiplex real-time PCR analysis of the 542 Phum_PHUM540560 gene: PLoS.One., v. 8, no. 2, p. e58088.

543 25. Drali, R., A. K. Sangare, A. Boutellis, E. Angelakis, A. Veracx, C. Socolovschi, P. 544 Brouqui, and D. Raoult, 2014, Bartonella quintana in body lice from scalp hair of 545 homeless persons, France: Emerg.Infect.Dis., v. 20, no. 5, p. 907-908.

46 546 26. Durden, L. A., 1990, Phoretic relationships between sucking lice (Anoplura) and flies 547 (Diptera) associated with humans and livestock.: Entomologist, v. 109, no. 3, p. 191- 548 192.

549 27. Durden, L. A., 2001, Lice (Phthiraptera)., in WM Samuel, MJ Pybus, and AA Kocan 550 eds., Parasitic diseases of wild mammals.: Ames: Iowa State University Press, p. 3-17.

551 28. Durden, L. A., and G. G. Musser, 1994a, The Mammalian Hosts of The Sucking Lice 552 (Anoploura) of the World: A Host-Parasite List.: Bull.Soc.Vector Ecol., v. 19, p. 130- 553 168.

554 29. Durden, L. A., and G. G. Musser, 1994b, The sucking lice (Insecta, Anoplura) of the 555 world: a taxonomic checklist with records of mammalian hosts and geographical 556 distributions.: Bulletin of the American Museum of Natural History, v. 218, p. 1-90.

557 30. Ewing, H. E., 1938, The Sucking Lice of American Monkeys.: The Journal of 558 Parasitology, v. 24, no. 1, p. 13-33.

559 31. Ferris, G. F., 1916, A catalogue and lost list of the Anoplura. San Francisco, The 560 Academy.

561 32. Ferris, G. F., 1951, The sucking lice: Mem.Pacific Coast Entomol.Soc., v. 1, p. 1-320.

562 33. Finstermeier, K., D. Zinner, M. Brameier, M. Meyer, E. Kreuz, M. Hofreiter, and C. 563 Roos, 2013, A mitogenomic phylogeny of living primates: PLoS.One., v. 8, no. 7, p. 564 e69504.

565 34. Grimaldi, D., and M. S. Engel, 2006, Fossil Liposcelididae and the lice ages (Insecta: 566 Psocodea): Proc.Biol Sci., v. 273, no. 1586, p. 625-633.

567 35. Groves, C., 1993, Order Primates., in DE Wilson and DM Reeder eds., Mammal 568 Species of the World: A Taxonomic and Geographic Reference: Washington, D.C., 569 Smithsonian Institution, p. 243-278.

47 570 36. Groves, C., 2005, Order Primates., in DE Wilson and DM Reeder eds., Mammal 571 Species of the World: A Taxonomic and Geographic Reference: Washington, D.C., 572 Smithsonian Institution, p. 243-278.

573 37. Groves, C., and M. Shekelle, 2010, The genera and species of Tarsiidae.: International 574 Journal of Primatology, v. 31, no. 6, p. 1071-1082.

575 38. Hafner, M. S., and S. A. Nadler, 1988, Phylogenetic trees support the coevolution of 576 parasites and their hosts: Nature, v. 332, no. 6161, p. 258-259.

577 39. Hafner, M. S., and R. D. Page, 1995, Molecular phylogenies and host-parasite 578 cospeciation: gophers and lice as a model system: Philos.Trans.R Soc Lond B Biol 579 Sci., v. 349, no. 1327, p. 77-83.

580 40. Harbison, C. W., and D. H. Clayton, 2011, Community interactions govern host- 581 switching with implications for host-parasite coevolutionary history: 582 Proc.Natl.Acad.Sci.U S A, v. 108, no. 23, p. 9525-9529.

583 41. Hindle, E., 1917, Notes on the biology of Pediculus humanus L.: Parasitology, v. 9, p. 584 259-265.

585 42. Hirai, H., H. Imai, and Y. Go, 2012, Post-Genome Biology of Primates. Tokyo, 586 Springer.

587 43. Hodgson, J. A., K. N. Sterner, L. J. Matthews, A. S. Burrell, R. A. Jani, R. L. Raaum, 588 C. B. Stewart, and T. R. Disotell, 2009, Successive radiations, not stasis, in the South 589 American primate fauna: Proc.Natl.Acad.Sci.U S A, v. 106, no. 14, p. 5534-5539.

590 44. Hopkins,GHE. The correct names of the body and head lice of man. [85], 91-92. 1952. 591 The Entomologist. 592 Ref Type: Serial (Book,Monograph)

593 45. Horai, S., K. Hayasaka, R. Kondo, K. Tsugane, and N. Takahata, 1995, The recent 594 African origin of modern humans revealed by complete sequences of hominoid

48 595 mitochondrial DNAs.: Proceedings of the National Academy of Sciences., v. 92, no. 2, 596 p. 532-536.

597 46. Houhamdi, L., H. Lepidi, M. Drancourt, and D. Raoult, 2006, Experimental model to 598 evaluate the human body louse as a vector of plague: J.Infect.Dis., v. 194, no. 11, p. 599 1589-1596.

600 47. Johnson, K. P., R. J. Adams, R. D. Page, and D. H. Clayton, 2003, When do parasites 601 fail to speciate in response to host speciation?: Syst.Biol, v. 52, no. 1, p. 37-47.

602 48. Johnson, K. P., and D. H. Clayton, 2003, The biology, ecology, and evolution of 603 chewing lice., in RD Price, RA Hellenthal, RL Palma, KP Johnson, and DH Clayton 604 eds., The chewing lice: world checklist and biological overview.: Illinois Natural 605 History Survey Special Publication, p. 449-476.

606 49. Johnson, K. P., K. Yoshizawa, and V. S. Smith, 2004, Multiple origins of parasitism in 607 lice: Proc.Biol Sci., v. 271, no. 1550, p. 1771-1776.

608 50. Johnston, J. S., K. S. Yoon, J. P. Strycharz, B. R. Pittendrigh, and J. M. Clark, 2007, 609 Body lice and head lice (Anoplura: Pediculidae) have the smallest genomes of any 610 hemimetabolous insect reported to date: J.Med.Entomol., v. 44, no. 6, p. 1009-1012.

611 51. Kelley, J. L., J. T. Peyton, A. S. Fiston-Lavier, N. M. Teets, M. C. Yee, J. S. Johnston, 612 C. D. Bustamante, R. E. Lee, and D. L. Denlinger, 2014, Compact genome of the 613 Antarctic midge is likely an adaptation to an extreme environment: Nat.Commun., v. 614 5, p. 4611.

615 52. Kim, K. C., 2006, Blood-sucking lice (Anoplura) of small mammals: True parasites., 616 in S Morand, BR Krasnov, and R Poulin eds., Micromammals and Macroparasites.: 617 Springer Japan, p. 141-160.

618 53. Kirkness, E. F. et al., 2010, Genome sequences of the human body louse and its 619 primary endosymbiont provide insights into the permanent parasitic lifestyle: 620 Proc.Natl.Acad.Sci.U.S.A, v. 107, no. 27, p. 12168-12173.

49 621 54. Kittler, R., M. Kayser, and M. Stoneking, 2003, Molecular evolution of Pediculus 622 humanus and the origin of clothing: Curr.Biol., v. 13, no. 16, p. 1414-1417.

623 55. Kuhn, H. J., and H. W. Ludwig, 1967, Die Affenlause der Gattung Pedicinus.: 624 Zeitschrift fur zoologische Systematik und Evolutionsforschung, v. 5, p. 144-297.

625 56. Laitman, J. T., 2011, A (New World monkey) tree grows in Brooklyn: 626 Anat.Rec.(Hoboken.), v. 294, no. 12, p. 1953-1954.

627 57. Lee, S. H. et al., 2010, Decreased detoxification genes and genome size make the 628 human body louse an efficient model to study xenobiotic metabolism: Insect Mol.Biol, 629 v. 19, no. 5, p. 599-615.

630 58. Light, J. E., 2005, Host-parasite cophylogeny and rates of evolution in two rodent- 631 louse assemblages (Doctoral dissertation, Faculty of the Louisiana State University 632 and Agricultural and Mechanical College in partial fulfillment of the requirements for 633 the degree of Doctor of Philosophy in The Department of Biological Sciences by 634 Jessica E. Light BS, University of Michigan).,

635 59. Light, J. E., V. S. Smith, J. M. Allen, L. A. Durden, and D. L. Reed, 2010, 636 Evolutionary history of mammalian sucking lice (Phthiraptera: Anoplura): 637 BMC.Evol.Biol, v. 10, p. 292.

638 60. Lyal, C. H., 1985a, Phylogeny and classification of the Psocodea, with particular 639 reference to the lice (Psocodea: Phthiraptera).: Systematic Entomology, v. 10, no. 2, p. 640 145-165.

641 61. Lyal, C. H. C., 1985b, A cladistic analysis and classification of trichodectid mammal 642 lice (Phthiraptera: Ischnocera): Bull.Brit.Mus.(Nat.Hist.) Entomol., v. 51, p. 187-346.

643 62. Mey, E., 2010, The Pedicinus species (Insecta, Phthiraptera, Anoplura, Pedicinidae) 644 on douc langurs (Pygathrix spp.).: Vietnamese Journal of Primatology, v. 4, p. 57-68.

50 645 63. Mittermeier, R. A., A. B. Rylands, and W. R. Konstant, 1999, Primates of the world: 646 An introduction, in RM Nowak ed., Walker's Mammals of the World (6th ed.): Johns 647 Hopkins University Press, p. 1-52.

648 64. Mumcuoglu, K. Y., and J. Zias, 1988, Head lice, Pediculus humanus capitis 649 (Anoplura, Pediculidae) from hair combs excavated in Israel and dated from 650 the first century B.C. to the eighth century A.D.: J.Med.Entomol., v. 25, p. 545-547.

651 65. Nuttall, G., 1919a, The biology of Pediculus humanus.: Parasitology, v. 10, p. 201- 652 220.

653 66. Nuttall, G. H., 1917, Studies on Pediculus. I. The copulatory apparatus and the process 654 of copulation in Pediculus humanus.: Parasitology, v. 9, no. 2, p. 293-324.

655 67. Nuttall, G. H., 1919b, The biology of Pediculus humanus L. (Supplementary notes).: 656 Parasitology, v. 11, p. 201-220.

657 68. Nuttall, G. H., 1919c, The systematic position, synonymy and iconography of 658 Pediculus humanus and Phthirus pubis.: Parasitology, v. 11, no. 3-4, p. 329-346.

659 69. Olds, B. P. et al., 2012, Comparison of the transcriptional profiles of head and body 660 lice: Insect Mol.Biol, v. 21, no. 2, p. 257-268.

661 70. Page, R. D., and M. A. Charleston, 1998, Trees within trees: phylogeny and historical 662 associations: Trends Ecol.Evol., v. 13, no. 9, p. 356-359.

663 71. Pariser, D. M., T. L. Meinking, and W. G. Ryan, 2013, Topical ivermectin lotion for 664 head lice: N.Engl.J.Med., v. 368, no. 10, p. 967.

665 72. Perelman, P. et al., 2011, A molecular phylogeny of living primates: PLoS.Genet., v. 666 7, no. 3, p. e1001342.

51 667 73. Perez, J. M., and R. L. Palma, 2001, A new species of Felicola (Phthiraptera: 668 Trichodectidae) from the endangered Iberian lynx: another reason to ensure its 669 survival.: Biodiversity and Conservation, v. 10, p. 929-937.

670 74. Piarroux, R., A. A. Abedi, J. C. Shako, B. Kebela, S. Karhemere, G. Diatta, B. 671 Davoust, D. Raoult, and M. Drancourt, 2013, Plague epidemics and lice, Democratic 672 Republic of the Congo: Emerg.Infect.Dis., v. 19, no. 3, p. 505-506.

673 75. Price, R. D., and O. H. Graham, 1997, Chewing and Sucking Lice as Parasites of 674 Mammals and Birds.: Technical Bulletin, no. 1849.

675 76. Price, R. D., and R. M. Timm, 1995, The chewing louse genus Aotiella (Phthiraptera: 676 Gyropidae) from the South American night monkeys, Aotus (Primates: Cebidae).: 677 Proceedings of the Entomological Society of Washington, v. 97, no. 3, p. 659-665.

678 77. Raoult, D., D. L. Reed, K. Dittmar, J. J. Kirchman, J. M. Rolain, S. Guillen, and J. E. 679 Light, 2008, Molecular identification of lice from pre-Columbian mummies: 680 J.Infect.Dis., v. 197, no. 4, p. 535-543.

681 78. Raoult, D., and V. Roux, 1999, The body louse as a vector of reemerging human 682 diseases: Clin.Infect.Dis., v. 29, no. 4, p. 888-911.

683 79. Reed, D. L., J. E. Light, J. M. Allen, and J. J. Kirchman, 2007, Pair of lice lost or 684 parasites regained: the evolutionary history of anthropoid primate lice: BMC.Biol., v. 685 5, p. 7.

686 80. Reed, D. L., V. S. Smith, S. L. Hammond, A. R. Rogers, and D. H. Clayton, 2004, 687 Genetic analysis of lice supports direct contact between modern and archaic humans: 688 PLoS.Biol, v. 2, no. 11, p. e340.

689 81. Reed, K. E., and J. G. Fleagle, 1995, Geographic and climatic control of primate 690 diversity: Proc.Natl.Acad.Sci.U S A, v. 92, no. 17, p. 7874-7876.

52 691 82. Richmond, B. G., and D. S. Strait, 2000, Evidence that humans evolved from a 692 knuckle-walking ancestor: Nature, v. 404, no. 6776, p. 382-385.

693 83. Roos, C., and T. Geissmann, 2001, Molecular phylogeny of the major hylobatid 694 divisions: Mol.Phylogenet.Evol., v. 19, no. 3, p. 486-494.

695 84. Roos, C., J. Schmitz, and H. Zischler, 2004, Primate jumping genes elucidate 696 strepsirrhine phylogeny: Proc.Natl.Acad.Sci.U S A, v. 101, no. 29, p. 10650-10654.

697 85. Rowe, N., J. Goodall, and R. A. Mittermeier, 1996, The Pictorial Guide to the Living 698 Primates. East Hampton, New York, Pogonias Press.

699 86. Rylands, A. B., and R. A. Mittermeier, 2009, The Diversity of the New World 700 Primates (Platyrrhini): An Annotated Taxonomy, in PA Garber, A Estrada, JC Bicca- 701 Marques, EW Heymann, and KB Strier eds., South American Primates. Comparative 702 Perspectives in the Study of Behavior, Ecology, and Conservation.: Springer, p. 23-54.

703 87. Sangare, A. K. et al., 2014, Detection of Bartonella quintana in African Body and 704 Head Lice: Am.J.Trop.Med.Hyg., v. 91, no. 2, p. 294-301.

705 88. Sasaki, T., S. K. S. Poudel, H. Isawa, T. Hayashi, S. Seki, T. Tomita, K. Sawabe, and 706 M. Kobayashi, 2006, First Molecular Evidence of Bartonella quintana in Pediculus 707 humanus capitis (Phthiraptera: Pediculidae), Collected from Nepalese Children: 708 J.Med.Entomol., v. 43, no. 1, p. 110-112.

709 89. Schmitt, D., M. D. Rose, J. E. Turnquist, and P. Lemelin, 2005, Role of the prehensile 710 tail during ateline locomotion: experimental and osteological evidence: 711 Am.J.Phys.Anthropol., v. 126, no. 4, p. 435-446.

712 90. Scholl, K., J. M. Allen, F. H. Leendertz, C. A. Chapman, and D. L. Reed, 2012, 713 Variable microsatellite loci for population genetic analysis of Old World monkey lice 714 (Pedicinus sp.): J.Parasitol., v. 98, no. 5, p. 930-937.

53 715 91. Seiffert, E. R., E. L. Simons, and Y. Attia, 2003, Fossil evidence for an ancient 716 divergence of lorises and galagos: Nature, v. 422, no. 6930, p. 421-424.

717 92. Shao, R., X. Q. Zhu, S. C. Barker, and K. Herd, 2012, Evolution of extensively 718 fragmented mitochondrial genomes in the lice of humans: Genome Biol Evol., v. 4, 719 no. 11, p. 1088-1101.

720 93. Smith, V. S., T. Ford, K. P. Johnson, P. C. Johnson, K. Yoshizawa, and J. E. Light, 721 2011, Multiple lineages of lice pass through the K-Pg boundary: Biol Lett., v. 7, no. 5, 722 p. 782-785.

723 94. Takken, W., and N. O. Verhulst, 2013, Host preferences of blood-feeding mosquitoes: 724 Annu.Rev.Entomol., v. 58, p. 433-453.

725 95. Veracx, A., A. Boutellis, V. Merhej, G. Diatta, and D. Raoult, 2012, Evidence for an 726 African cluster of human head and body lice with variable colors and interbreeding of 727 lice between continents: PLoS.One., v. 7, no. 5, p. e37804.

728 96. Wilkinson, R. D., M. E. Steiper, C. Soligo, R. D. Martin, Z. Yang, and S. Tavare, 729 2011, Dating primate divergences through an integrated analysis of palaeontological 730 and molecular data: Syst.Biol, v. 60, no. 1, p. 16-31.

731 97. Wilson, D. E., and D. M. Reeder, 2005, Mammal species of the world: a taxonomic 732 and geographic reference. JHU Press.

733 98. Yoder, A. D., 2003, The phylogenetic position of genus Tarsius: whose side are 734 you on?, in PC Wright, EL Simons, and S Gursky eds., Tarsiers: past, present, and 735 future: New Jersey, Rutgers University Press, p. 161-175.

736 99. Yoon, K. S., J. P. Strycharz, J. H. Baek, W. Sun, J. H. Kim, J. S. Kang, B. R. 737 Pittendrigh, S. H. Lee, and J. M. Clark, 2011, Brief exposures of human body lice to 738 sublethal amounts of ivermectin over-transcribes detoxification genes involved in 739 tolerance: Insect Mol.Biol., v. 20, no. 6, p. 687-699. 740

54 Table 1. The lice of primates

Suborder Family Genus Primate Family Primate Genus Ischnocera Philopteridae Trichophilopterus Lemuridae Lemur and Eulemur Indridae Indri and Propithecus Trichodectidae Felicola Lorisidae Nycticebus Cebidicola Cebidae Alouatta and Brachyteles Amblycera Gyropidae Aotiella Aotus Anoplura Polyplacidae Lemurpediculus Cheirogaleidae Cheirogaleus and Microcebus Phthirpediculus Lemuridae Eulemur

Megaladapidae Lepilemur 55 Indridae Avahi and Propithecus Lemurphthirus Galagonidae Galago, Galagoides and Otolemur Pediculidae Pediculus Hominidae Pan and Homo Cebidae Alouatta, Cebus, Cacajao, Pithecia and Ateles Hylobatidae Hylobates Pedicinidae Pedicinus Cercopithecidae Cercocebus, Cercopithecus, Erythrocebus, Macaca, Papio, Lophocebus, Chlorocebus, Miopithecus, Colobus, Nasalis, Presbytis, Procolobus, Trachypithecus and Semnopithecus Pthiridae Pthirus Hominidae Gorilla and Homo Figure 1. Maximum-likelihood (ML) phylogram of cox1 mt gene in primates. The symbols represent the different genera of lice that parasitize different primate families.

56 Figure 2. Maximum-likelihood (ML) phylogram of cytb mt gene. ML bootstrap that support values greater than 75 are located above the nodes. Mitochondrial clade memberships are indicated to the right of each tree. GenBank accession numbers, manuscript lead author, locality are indicated for each louse specimen. Localities are abbreviated as follows: California (CA), Florida (FL), Georgia (GA), Maryland (MD), Massachusetts (MA), Papua New Guinea (PNG), Republic Democratic of the Congo (RDC), United Kingdom (UK), and Utah (UT). 57

Chapitre 1 : Différenciation pou de tête - pou de corps

Préambule

Le pou de tête et le pou de corps appartiennent-ils à la même espèce?

Une question restée en débat pendant plus de deux siècles. En effet, une fois détachés de leurs niches écologiques respectives, ils ne sont plus différenciables [42]. La comparaison de différents critères morphologiques chez les deux types de poux tels la forme, la taille ou encore la pigmentation n’a pas réussi à les séparer [43]. Certains auteurs soutiennent que les différences morphologiques observées entre les deux écotypes seraient dues à une adaptation à leur environnement. À juste titre, des poux de tête élevés sous les conditions habituelles des poux de corps deviennent progressivement indiscernables de ces derniers [44,45].

C’est finalement le séquençage du génome du pou de corps [28] qui a permis de faire une percée dans le traitement de la question. En effet, dans une étude visant à comparer le profil transcriptionnel du pou de tête et du pou de corps, Olds et al. avaient soutenu que seul le gène phum_PHUM540560 séparait les deux types de poux. Pour les auteurs, ce gène était présent et transcrit chez le pou de corps, mais absent chez le pou de tête [29].

61 Dans la foulée, nous avons ciblé et réussi à amplifier une séquence partielle de 187 pb de ce gène aussi bien chez des poux de tête que des poux de corps appartenant au Clade A provenant de 13 pays, 5 continents. L’alignement des séquences obtenues a révélé l’existence de plusieurs polymorphismes entre les séquences des deux types de poux. Nous nous sommes appuyés sur ces différences pour mettre en place un outil moléculaire basé sur de la RT PCR multiplexe capable de différencier rapidement entre le pou de tête et le pou de corps [42].

Cet outil s’est révélé particulièrement utile sur le terrain pour connaître le statut génotypique des poux infectés par B. quintana

[16,46].

Article I: Distinguishing Body Lice from Head Lice by Multiplex Real-Time PCR Analysis of the Phum_PHUM540560 Gene

PLoS One 8: e58088

Distinguishing Body Lice from Head Lice by Multiplex Real-Time PCR Analysis of the Phum_PHUM540560 Gene

Rezak Drali1,2, Amina Boutellis1, Didier Raoult1, Jean Marc Rolain1, Philippe Brouqui1* 1 Aix Marseille Universite´, URMITE, UM63, CNRS 7278, IRD 198, Inserm 1095, Marseille, France,2 Service des Ente´robacte´ries et Hygie`ne de l’Environnement, Institut Pasteur d’Alge´rie, Algiers, Algeria

Abstract

Background: Body louse or head louse? Once removed from their environment, body and head lice are indistinguishable. Neither the morphological criteria used since the mid-18th century nor the various genetic studies conducted since the advent of molecular biology tools have allowed body lice and head lice to be differentiated. In this work, using a portion of the Phum_PHUM540560 gene from the body louse, we aimed to develop a multiplex real-time polymerase chain reaction (PCR) assay to differentiate between body and head lice in a single reaction.

Materials and Methods: A total of 142 human lice were collected from mono-infested hosts from 13 countries on five continents. We first identified the louse clade using a cytochrome b (CYTB) PCR sequence alignment. We then aligned a fragment of the Phum_PHUM540560 gene amplified from head and body lice to design-specific TaqMan� FAM- and VIC- labeled probes.

Results: All the analyzed lice were Clade A lice. A total of 22 polymorphisms between the body and head lice were characterized. The multiplex real-time PCR analysis enabled the body and head lice to be distinguished in two hours. This method is simple, with 100% specificity and sensitivity.

Conclusions: We confirmed that the Phum_PHUM540560 gene is a useful genetic marker for the study of lice.

Citation: Drali R, Boutellis A, Raoult D, Rolain JM, Brouqui P (2013) Distinguishing Body Lice from Head Lice by Multiplex Real-Time PCR Analysis of the Phum_PHUM540560 Gene. PLoS ONE 8(2): e58088. doi:10.1371/journal.pone.0058088 Editor: Ramy K. Aziz, Cairo University, Egypt Received August 29, 2012; Accepted January 30, 2013; Published February 28, 2013 Copyright:� 2013 Drali et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: The authors have no support or funding to report. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected]

Introduction 18S ribosomal RNA has enabled the sub-Saharan African phylogenetic group of lice to be distinguished from a second Body and head lice are hematophagous ectoparasites that are group that encompasses the remainder of the lice worldwide [8,9]. specific to humans [1] and have different ecologies. The body An analyses of the mitochondrial cytochrome b (CYTB) and louse, Pediculus humanus corporis, lives and multiplies in clothing, cytochrome oxidase I (COI) genes, have allowed the differentia- whereas the head louse, Pediculus humanus capitis, lives and lays its tion of three clades of lice. Clade A contains both body and head eggs on hair [2,3]. The body louse is known as a vector of three lice that are distributed worldwide. Clade B contains head lice life-threatening infectious diseases: epidemic typhus, caused by encountered in America, Europe and Australia, whereas Clade C Rickettsia prowazekii; relapsing fever, caused by Borrelia recurrentis; and contains head lice found in Ethiopia, Nepal and Senegal [10–14]. trench fever, caused by Bartonella quintana [4,5]. Recently, a method targeting intergenic spacers that utilizes four Distinguishing body from head lice has always been a challenge. highly polymorphic markers has revealed associations between the Once a louse leaves its biotope (head or clothes), it becomes sources and genotypic distributions of lice [15,16]. Nevertheless, indistinguishable from other lice, which has presented a critical none of the above genetic studies were able to differentiate problem in historical and paleobiological studies of lice. between body and head lice. In 2010, the sequencing of the entire Since the mid-18th century, morphological criteria such as size, genome of P. humanus corporis provided new perspectives for shape and color gradation have been used to differentiate body understanding the relationship between the biology and genetics of and head lice into two distinct species [6]. In 1978, the use of microscopes to observe body and head lice collected from the louse [17]. More recently, a study comparing the transcrip- Ethiopians with double infestations allowed a researcher to tional profiles of body and head lice reported that the two types of conclude that the lice represented two distinct species, Pediculus lice had a single, 752-base pair (bp) difference in the Phum_- humanus Linnaeus and Pediculus capitis De Geer. He based his assertion PHUM540560 gene, which encodes a hypothetical, 69-amino on the length of the tibia of the louse’s middle leg [7]. acids protein of unknown function [18]. Based on the alignment of The advent of molecular biology and gene sequencing has led to a portion of the two Phum_PHUM540560 gene sequences, we the development of genetic studies to address issues concerning have designed a novel multiplex real-time PCR assay to efficiently louse phylogeny. The investigation of the gene that encodes the differentiate, for the first time, between body and head lice

PLOS ONE | www.plosone.org 651 February 2013 | Volume 8 | Issue 2 | e58088 Differentiating Body from Head Louse collected from a mono-infested host. This assay has been tested by Research, Waltham, MA, USA). The final reaction volume was analyzing a large collection of worldwide specimens belonging to 20 ml, with 0.4 U of Phusion polymerase (Finnzymes, Thermo Clade A, the only clade known to contain both body and head lice. Scientific, Vantaa, Finland), 4 ml of 5x Phusion buffer, 0.5 mM of each primer, 0.16 mM dNTP mix and 30–50 ng of genomic Materials and Methods DNA. The following cycling conditions were used for the amplifications: an initial 30-s denaturation at 98uC; 35 cycles of Ethics statement denaturation for 5 s at 98uC and annealing for 30 s at 56uC (for Lice from foreign countries were obtained from the private CYTB gene) or 59uC (for the Phum_PHUM540560 gene); and a frozen collection of our laboratory (The URMITE/WHO final 15 min extension at 72uC. The amplification was completed Collaborative Research Center). The lice in that collection were by a 5-min extension at 72uC. Subsequently, the PCR products required for various epidemiological and entomological studies or were subjected to electrophoresis on 1.5% agarose gels with to perform diagnoses abroad and were sent to our laboratory as a ethidium bromide staining and were then purified using Nucleo- WHO reference facility. The specimens were collected according Fast 96 PCR Plates (Macherey-Nagel EURL, Hoerdt, France) to the ethics laws of each country; however, because lice are not according to the manufacturer’s instructions. part of the human body, lice removed from individuals are not Bidirectional DNA sequencing of the targeted PCR products considered to be human samples in most countries. The body lice was performed using the 3130XL genetic analyzer (Applied were collected from clothing, and the head lice were removed from Biosystems, Courtaboeuf, France) with the BigDye Terminator hair, with the verbal consent of the infested individuals. Written v1.1 cycle (Applied Biosystems). The electropherograms obtained consent was not obtainable in the majority of cases because most for each sequence were analyzed using Chromas Pro software of the subjects were illiterate. However, in most instances, the (Technelysium PTY, Australia). investigator, local authorities and/or village chief approved and were present when it was performed. Phylogenetic analysis The lice collected in France were obtained from homeless The DNA sequences were aligned using the multi-sequence individuals during a registered epidemiological study (French alignment software CLUSTAL X, version 2.0.11. The partial Bioethics laws nu 2011–814). Informed consent was obtained from CYTB gene sequences were aligned with sequences available from these subjects, and the study was approved by the ‘‘Comite´ de GenBank. The percentages of similarity were determined using the Protection des Personnes Sud Mediterrane´e I’’ on January, 12, MEGA 5 software package (Molecular Evolution Genetic Anal- 2011 (ID RCB: 2010-A01406-33). ysis, The Biodesign Institute, AZ, USA) [20]. The PhyML The anonymity of the individuals who provided the lice used in phylogeny software was used to create an unrooted phylogenetic the present genetic analysis was preserved. tree based on the DNA sequences using maximum likelihood (ML) 100 bootstrap replicates [21]. Sampling A total of 142 lice, including 88 body lice and 54 head lice, were Real-time PCR and PCR products sequencing collected from mono-infested human hosts. The head lice were � TaqMan FAM- and VIC-labeled probes (Table 2) specific to collected exclusively from the hair, and the body lice were body and head lice, respectively, were designed for the sequences collected exclusively from clothing. No lice were collected from the obtained in this study. Both probes contained a TAMRA neck or the beard; the purpose of this precaution was the quencher dye at the 39 end. The probes were synthesized by avoidance hybrid lice, as previously reported [7]. The strain Applied Biosystems (Courtaboeuf, France). information, geographic origin and anatomical sources (body or Monoplex and multiplex real-time PCRs were performed in the head) of the analyzed lice are provided in Table 1. CFX96 thermal cycler (Bio-Rad Laboratories, Foster City, CA, USA). The final reaction volume of 20 ml contained 5–20 ng of the DNA preparation DNA template, 10 ml of 2x QuantiTect Probe PCR Master Mix Prior to DNA isolation, each louse was immersed in 70% (Qiagen), 0.5 mM of each primer and 0.2 mM of the FAM- or ethanol for 15 min and was then rinsed twice in sterile water. VIC-labeled probes. A monoplex protocol designed to optimize Total genomic DNA was extracted using the QIAamp Tissue Kit the conditions for the multiplex real-time PCR was used: a (QIAGEN, Hilden, Germany) according to the manufacturer’s denaturation step at 95uC for 15 min; and 40 cycles of 95uC for instructions. The extracted DNA was assessed for quantity and 15 s and 60uC for 45 s. The multiplex real-time PCR was quality using a NanoDrop instrument (Thermo Scientific, performed using the optimized conditions that were determined in 2 Wilmington, United Kingdom) before being stored at 20uC [19]. the monoplex real-time PCR assay. Each reaction contained 10 ml of 2x QuantiTect Probe PCR Master Mix (Qiagen), 0.5 mM of Conventional PCR and sequencing each primer, 0.2 mM of each fluorogenic probe, 5–20 ng of the Two conventional PCR experiments were performed in this DNA template, adjusted to a final volume of 20 ml with the study. The first was performed to identify the Clades of the addition of nuclease-free dH2O. The cycling parameters consisted collected lice by amplifying and sequencing a 347-bp fragment of of 95uC for 15 min and 40 cycles of 95uC for 15 s and 60uC for the mitochondrial cytochrome b (CYTB) gene [7]. The second 1 min. To evaluate the specificity (the ability of the test to identify PCR targeted a 187-bp fragment of the Phum_ PHUM540560 negative results) and sensitivity (the ability of the test to identify gene using a pair of primers designed in this study and based on positive results) of the developed method, all the products of the the Phum_PHUM540560 gene sequence available from GenBank multiplex real-time PCR amplifications were sequenced, and these (Pediculus humanus corporis strain USDA 1103172108290, GenBank sequences were used as the gold standard reference. accession no. NW_002987859.1 GI: 242022583). The obtained PCR products from three body lice and three head lice were Results sequenced to enable comparison of the body and head lice DNA sequences. All the PCRs were performed using the primers The concentration of the genomic DNA extracted from the 142 outlined in Table 2 and a PTC-200 automated thermal cycler (MJ lice analyzed in this study ranged from 5 to 20 ng/ml.

PLOS ONE | www.plosone.org 662 February 2013 | Volume 8 | Issue 2 | e58088 Differentiating Body from Head Louse

Table 1. The Clade A lice examined in this study and the results of the real-time PCR assay.

Country Town/province Analysis channel results Number FAM-positive VIC-positive

Body lice France Marseille 15 15 0 Hungary Budapest 10 10 0 Nepal Pokava 9 9 0 China Inner Mongolia Province 5 5 0 Tiligi 7 7 0 Japan Tokyo 10 10 0 Madagascar Borenty village 9 9 0 Kenya Nairobi 10 10 0 USA Orlando 13 13 0 Head lice USA Washington 6 0 6 Brazil Sao Cristovao 6 0 6 Amazonia 8 0 8 Madagascar Bedaro village 12 0 12 Senegal Dakar 3 0 3 Australia Brisbane 5 0 5 Papua New Guinea Highlands 5 0 5 New Zeland Auckland 9 0 9

doi:10.1371/journal.pone.0058088.t001

Genotypic distribution of the lice based on the acid and the I19N (ATT.AAT) mutation that would replace mitochondrial CYTB gene isoleucine with asparagine. A 347-bp DNA fragment was successfully amplified from the The remainder of the polymorphisms were spread throughout CYTB gene in all 142 lice. Direct sequencing and multiple the first intron and included the insertion of nucleotides at two alignments of the obtained sequences revealed that all the lice different locations: approximately nucleotide (nt) 96+11 ins.G and belonged to Clade A (data not shown). nt 96+80, (ins.CT). This triplex insertion resulted in the amplification of a 190-bp fragment from the head lice and a Characterization of the partial Phum_PHUM540560 gene 187-bp fragment from the body lice (Figure 1). in body and head lice Real-time PCR and PCR product sequencing The multiple alignments of the partial Phum_PHUM54056 The monoplex real-time PCR results demonstrated that the gene sequences obtained from the six analyzed lice revealed 22 FAM-labeled probe was specific to the body lice and that the VIC- polymorphisms between body and head lice (Figure 1). The first labeled probe was specific to the head lice. This assay was exon contained two point mutations: a silent (CCA.CCC) optimized by testing louse specimens from known anatomical transversion affecting codon 18 that would not change the amino locations.

Table 2. The oligonucleotide primers and probes used in this study.

Name Purpose Sequence 59R39

Cytb_F Forward sequencing primer partial GAGCGACTGTAATTACTAATC cytochrome b gene Cytb_R Reverse sequencing primer GGACCCGGATAATTTTGTTG partial cytochrome b gene Phum540560_F Forward sequencing primer GTCACGTTCGACAAATGTT partial Phum_PHUM540560 gene Phum540560_R Reverse sequencing primer TTTCTATAACCACGACACGATAAAT partial Phum_PHUM540560 gene BL probe Specific to the body lice FAM-CGATCACTCGAGTGAATTGCCA-TAMRA HL probe Specific to the head lice VIC-CTCTTGAATCGACGACCATTCGCT-TAMRA

doi:10.1371/journal.pone.0058088.t002

PLOS ONE | www.plosone.org 673 February 2013 | Volume 8 | Issue 2 | e58088 Differentiating Body from Head Louse

Figure 1. Primer and probe alignments with partial Phum_PHUM540560 gene sequences from body and head lice[35]. A portion of the Phum_PHUM540560 gene sequences from body and head lice were aligned with the primers and probes designed for the multiplex RT-PCR assay. Part of the first exon spanings nucleotides 1 to 64 was analyzed. The forward and reverse primer sequences are boxed in black. The FAM- and VIC-labeled probe sequences are boxed in purple and green, respectively. The nucleotides in blue represent single-nucleotide polymorphisms that are specific to head lice. The nucleotides in black represent single-nucleotide polymorphisms that are specific to body lice. BL: body louse; HL: head louse; NW_002987859.1: Pediculus humanus corporis strain USDA 1103172108290 Phum_PHUM540560 (gene sequence available in GenBank). doi:10.1371/journal.pone.0058088.g001

The multiplex real-time PCR assay clearly identified and for the VIC-labeled probe contained sequences specific to head simultaneously differentiated among the 142 lice included in this lice (100% sensitivity and 100% specificity; data not shown). work. Specifically, the signal emitted by the FAM-labeled probe was detected only in the body louse samples, whereas the signal Discussion emitted by the VIC-labeled probe was detected only in the head louse samples (Figure 2). No signals were detected in the non- Presently, comparisons of the body and head lice genomes are template controls (NTCs). The Ct values obtained in this assay are not possible because the head louse genome is not yet available. outlined in Table S1. The sequencing of the 142 PCR products Recently published comparative transcriptional profiles of both has confirmed our results. In addition, 100% of the samples that body and head lice demonstrated that among the nine genes with were positive for the FAM-labeled probe contained sequences differential expression, only one gene was absent in the head louse specific to body lice, and 100% of the samples that were positive but present in the body louse [18]. We considered this difference to be a possible opportunity for distinguishing body lice from head

Figure 2. Amplification curves from multiplex real-time PCR assays. Figure 1A. Real-time PCR amplification curves for body lice using a partial Phum_PHUM540560 gene in the FAM channel (495–520). Figure 1B. Amplification curves for head licee louse using a partial Phum_PHUM540560 gene in the VIC channel (522–544). doi:10.1371/journal.pone.0058088.g002

PLOS ONE | www.plosone.org 684 February 2013 | Volume 8 | Issue 2 | e58088 Differentiating Body from Head Louse lice. Unexpectedly, our first PCR amplification of the 187- bp clothing. In such a situation, finding lice on the head challenges fragment of the Phum_PHUM540560 gene produced a PCR the ‘‘head louse definition’’, making an identification tool useful. product from both head and body louse samples, suggesting that at Recent studies have suggested that head and body lice can be least a portion of the gene was present in both types of lice. The mixed in people infested with both types of lice [28]. Although sequencing of the PCR products revealed significant differences head and body lice do not interbreed in the wild [7], fertile hybrids between the sequences from the head and body lice, which may with an intermediate morphology [30] have been reported under explain why the head louse sequence failed to amplify in Old’s laboratory conditions [31,32]. Moreover, several observational experiment [18]. studies have also suggested that head lice may become body lice In this study, we exploited this sequence variation in the partial when raised under the appropriate conditions [33,34]. Our Phum_PHUM540560 gene to discriminate between body and technique can be used to identify heterozygous specimens, which head lice from a global collection of lice collected from mono- may prove valuable for studies on the population dynamics of lice. infested hosts originating from five different continents. Lice from This work confirmed that the Phum_PHUM540560 gene may Clade A were used because Clade A is the only currently be a useful genetic marker for the study of lice. However, the recognized clade that includes both head and body lice [10,13,14], genetic differences between head and body lice do not put back the two types of lice that our assay was developed to distinguish. into question whether head and body lice are conspecific [11]. The Finally, our choice of specimens was based on the commonly ability to distinguish between head and body lice may facilitate recognized definitions of body and head lice. Under these future research into the behavior of Clade A body and head lice. conditions, we developed a multiplex real-time PCR assay that is rapid (two hours) and simple and has 100% specificity and Supporting Information sensitivity. The purpose of this study was to distinguish between body and Table S1 Ct values obtained in multiplex real-time PCR for head lice, a long-standing challenge. Resolving this challenge has differentiating between body and head louse. become even more important because both head and body lice (DOCX) have been reported to harbor Bartonella quintana, the trench fever agent, raising the question of whether head lice, similar to body Acknowledgments lice, can transmit the agent [22,23]. Currently, B. quintana DNA has been detected only in head lice collected from impoverished The text was edited by American Journal Experts under Certificate people in situations where co-infestations with body lice are Verification Key: 6B5B-1E56-39B6-59E4-BE7F. We thank Idir BITAM, possible [24–27]. In fact, co-infestations have been recently Institut Pasteur d’Alge´rie, Algiers, Algeria; and Christophe ROGIER, reported in the same homeless population [28]. One study of Institut Pasteur de Madagascar, Antananarivo. head lice collected from schoolchildren in France failed to detect B. quintana [29]. The ability to distinguish body lice from head lice Author Contributions will help advance our understanding of the role of head louse in Conceived and designed the experiments: PB DR JMR. Performed the the transmission of B. quintana. Moreover, 22% of the homeless experiments: RD AB. Analyzed the data: RD AB DR JMR PB. Wrote the people who frequent shelters in Marseille, France are infested with paper: RD PB. lice, and some people can harbor more than 10,000 lice in their

References 1. Durden LA (2002) Lice (Phthiraptera). In: Mullen G, Durden LA, editors. 14. Reed DL, Smith VS, Hammond SL, Rogers AR, Clayton DH (2004) Genetic Medical and veterinary entomology. 45–65. analysis of lice supports direct contact between modern and archaic humans. 2. Burgess IF (2004) Human lice and their control. Annu Rev Entomol 49: 457– PLoS Biol 2: e340. 10.1371/journal.pbio.0020340 [doi]. 481. 10.1146/annurev.ento.49.061802.123253 [doi]. 15. Li W, Ortiz G, Fournier PE, Gimenez G, Reed DL, et al. (2010) Genotyping of 3. Light JE, Toups MA, Reed DL (2008) What’s in a name: the taxonomic status of human lice suggests multiple emergencies of body lice from local head louse human head and body lice. Mol Phylogenet Evol 47: 1203–1216. populations. PLoS Negl Trop Dis 4: e641. 10.1371/journal.pntd.0000641 [doi]. 4. Brouqui P, Stein A, Dupont HT, Gallian P, Badiaga S, et al. (2005) 16. Veracx A, Boutellis A, Merhej V, Diatta G, Raoult D (2012) Evidence for an Ectoparasitism and vector-borne diseases in 930 homeless people from African Cluster of Human Head and Body Lice with Variable Colors and Marseilles. Medicine (Baltimore) 84: 61–68. Interbreeding of Lice between Continents. PLoS One 7: e37804. 10.1371/ 5. Raoult D, Roux V (1999) The body louse as a vector of reemerging human journal.pone.0037804 [doi]; PONE-D-12-05586 [pii]. diseases. Clin Infect Dis 29: 888–911. 17. Kirkness EF, Haas BJ, Sun W, Braig HR, Perotti MA, et al. (2010) Genome 6. Veracx A, Raoult D (2012) Biology and genetics of human head and body lice. sequences of the human body louse and its primary endosymbiont provide Trends Parasitol. S1471-4922(12)00163-8 [pii];10.1016/j.pt.2012.09.003 [doi]. insights into the permanent parasitic lifestyle. Proc Natl Acad Sci U S A 107: 7. Busvine JR (1978) Evidence from double infestations for the specific status of 12168–12173. 1003379107 [pii];10.1073/pnas.1003379107 [doi]. human head lice and body lice (Anoplura). Systematic Entomology 3: 1–8. 18. Olds BP, Coates BS, Steele LD, Sun W, Agunbiade TA, et al. (2012) 8. Leo NP, Barker SC (2005) Unravelling the evolution of the head lice and body Comparison of the transcriptional profiles of head and body lice. Insect Mol Biol lice of humans. Parasitol Res 98: 44–47. 10.1007/s00436-005-0013-y [doi]. 21: 257–268. 10.1111/j.1365-2583.2012.01132.x [doi]. 9. Yong Z, Fournier PE, Rydkina E, Raoult D (2003) The geographical segregation 19. Drali R, Benkouiten S, Badiaga S, Bitam I, Rolain JM, et al. (2012) Detection of of human lice preceded that of Pediculus humanus capitis and Pediculus a knockdown resistance (kdr) mutation associated with permethrin resistance in humamus humanus. C R Biol 326. the body louse Pediculus humanus corporis using melting curve analysis 10. Kittler R, Kayser M, Stoneking M (2003) Molecular evolution of Pediculus genotyping. J Clin Microbiol. JCM.00808-12 [pii];10.1128/JCM.00808-12 humanus and the origin of clothing. Curr Biol 13: 1414–1417. [doi]. S0960982203005074 [pii]. 20. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, et al. (2011) MEGA5: 11. Leo NP, Campbell NJ, Yang X, Mumcuoglu K, Barker SC (2002) Evidence molecular evolutionary genetics analysis using maximum likelihood, evolution- from mitochondrial DNA that head lice and body lice of humans (Phthiraptera: ary distance, and maximum parsimony methods. Mol Biol Evol 28: 2731–2739. Pediculidae) are conspecific. J Med Entomol 39: 662–666. msr121 [pii];10.1093/molbev/msr121 [doi]. 12. Light JE, Allen JM, Long LM, Carter TE, Barrow L, et al. (2008) Geographic 21. Guindon S, Gascuel O (2003) A simple, fast, and accurate algorithm to estimate distributions and origins of human head lice (Pediculus humanus capitis) based large phylogenies by maximum likelihood. Syst Biol 52: 696–704. on mitochondrial data. J Parasitol 94: 1275–1281. GE-1618 [pii];10.1645/GE- 54QHX07WB5K5XCX4 [pii]. 1618.1 [doi]. 22. Angelakis E, Rolain JM, Raoult D, Brouqui P (2011) Bartonella quintana in 13. Raoult D, Reed DL, Dittmar K, Kirchman JJ, Rolain JM, et al. (2008) head louse nits. FEMS Immunol Med Microbiol 62: 244–246. 10.1111/j.1574- Molecular identification of lice from pre-Columbian mummies. J Infect Dis 197: 695X.2011.00804.x [doi]. 535–543.

PLOS ONE | www.plosone.org 695 February 2013 | Volume 8 | Issue 2 | e58088 Differentiating Body from Head Louse

23. Brouqui P, Raoult D (2006) Arthropod-borne diseases in homeless. Ann N Y Acad 29. Bouvresse S, Socolovshi C, Berdjane Z, Durand R, Izri A, et al. (2011) No Sci 1078: 223–235. evidence of Bartonella quintana but detection of Acinetobacter baumannii in 24. Bonilla DL, Kabeya H, Henn J, Kramer VL, Kosoy MY (2009) Bartonella head lice from elementary schoolchildren in Paris. Comp Immunol Microbiol quintana in body lice and head lice from homeless persons, San Francisco, Infect Dis 34: 475–477. S0147-9571(11)00076-2 [pii];10.1016/j.ci- California, USA. Emerg Infect Dis 15: 912–915. mid.2011.08.007 [doi]. 25. Boutellis A, Veracx A, Angelakis E, Diatta G, Mediannikov O, et al. (2012) 30. Busvine JR (1948) The head and body races of Pediculus humanus L. Bartonella quintana in head lice from Senegal. Vector Borne Zoonotic Dis 12: Parasitology 39: 1–16. 564–567. 10.1089/vbz.2011.0845 [doi]. 31. Bacot AW (1917) A contribution to the bionomics of Pediculus humanus (vestimenti) 26. Sasaki T, Poudel SK, Isawa H, Hayashi T, Seki N, et al. (2006) First molecular and Pediculus capitis. Parasitology 9: 228–258. evidence of Bartonella quintana in Pediculus humanus capitis (Phthiraptera: 32. Mullen G, Durden LA (2009) Medical and Veterinary Entomology. Academic Pediculidae), collected from Nepalese children. J Med Entomol 43: 110–112. Press, San Francisco. 27. Angelakis E, Diatta G, Abdissa A, Trape JF, Mediannikov O, et al. (2011) 33. Alpatov V, Nastukova OA (1955) Transformation of the head form of Pediculus Altitude-dependent Bartonella quintana genotype C in head lice, Ethiopia. humanus into the body form under changed conditions of existence. Bull Moscow Emerg Infect Dis 17: 2357–2359. 10.3201/eid1712.110453 [doi]. Nat Hist Res Soc 60: 92. 34. Nuttall G (1919) The biology of Pediculus humanus. Supplementary notes. 28. Veracx A, Rivet R, McCoy KD, Brouqui P, Raoult D (2012) Evidence that head Parasitology 11: 201–221. and body lice on homeless persons have the same genotype. PLoS One 7: 35. Corpet F (1988) Multiple sequence alignment with hierarchical clustering. e45903. 10.1371/journal.pone.0045903 [doi];PONE-D-12-20556 [pii]. Nucleic Acids Res 16: 10881–10890.

PLOS ONE | www.plosone.org 706 February 2013 | Volume 8 | Issue 2 | e58088 Table S1. Ct values obtained in multiplex real-time PCR for differentiating between body and head louse

Ct values Louse Country Town/Province FAM VIC Body louse_1003/4 21,56 NA Body louse_1160/1 20,84 NA Body louse_1200/2 20,02 NA Body louse_1229/6 18,33 NA Body louse_1237/9 19,04 NA Body louse_1237/10 20,19 NA Body louse_1237/11 20,11 NA Body louse_1237/16 20,24 NA France Marseille Body louse_1237/17 20,09 NA Body louse_1237/19 20,51 NA Body louse_1240/1 20,87 NA Body louse_1245/4 20,72 NA Body louse_1246/1 18,74 NA Body louse_2002/2 19,04 NA Body louse_2215 20,10 NA Body louse_104 22,02 NA Body louse_105 25,10 NA Body louse_106 26,51 NA Body louse_107 23,90 NA Body louse_108 24,17 NA Hungary Budapest Body louse_109 22,10 NA Body louse_110 27,59 NA Body louse_111 22,74 NA Body louse_112 23,19 NA Body louse_113 25,00 NA Body louse_80 23,08 NA Body louse_82 21,58 NA Inner Mongolia Body louse_83 23,50 NA Province Body louse_83 30,84 NA Body louse_86 31,98 NA Body louse_87 20,68 NA China Body louse_88 22,17 NA Body louse_89 26,41 NA Body louse_90 20,49 NA Tiligi Body louse_91 22,43 NA Body louse_92 21,94 NA Body louse_93 22,84 NA Body louse_94 21,74 NA Body louse_95 21,28 NA Body louse_96 20,08 NA Japan Tokyo Body louse_97 20,89 NA Body louse_98 21,43 NA

71 Body louse_99 23,07 NA Body louse_100 21,86 NA Body louse_101 22,04 NA Body louse_102 22,07 NA Body louse_103 22,42 NA Body louse_70 21,66 NA Body louse_71 21,71 NA Body louse_72 23,48 NA Body louse_73 23,08 NA Body louse_74 22,29 NA Nepal Pokava Body louse_75 23,22 NA Body louse_76 26,08 NA Body louse_78 24,11 NA Body louse_79 23,88 NA Body louse_223 30,34 NA Body louse_224 35,20 NA Body louse_226 33,64 NA Body louse_227 30,26 NA Body louse_228 36,03 NA Kenya Nairobi Body louse_229 30,68 NA Body louse_230 31,23 NA Body louse_231 30,21 NA Body louse_232 32,72 NA Body louse_233 35,21 NA Body louse_175 20,46 NA Body louse_176 21,04 NA Body louse_177 20,10 NA Body louse_178 20,12 NA Body louse_179 22,11 NA Madagascar Borenty village Body louse_180 21,87 NA Body louse_181 23,76 NA Body louse_182 23,45 NA Body louse_183 23,02 NA Body louse_PDL1 19,28 NA Body louse_PDL3 18,88 NA Body louse_PDL4 18,66 NA Body louse_PDL6 18,62 NA Body louse_PDL11 18,90 NA Body louse_PDL13 18,57 NA Body louse_PDL14 19,13 NA USA Laboratory colony Body louse_PDL7 18,92 NA Body louse_PDL2 18,30 NA Body louse_PDL9 17,53 NA Body louse_PDL10 19,01 NA Body louse_PDL4 21,18 NA Body louse_PDL5 21,03 NA Head louse_13 NA 31,52 USA Washington

72 Head louse_14 NA 32,36 Head louse_15 NA 32,04 Head louse_16 NA 35,44 Head louse_17 NA 37,34 Head louse_19 NA 36,01 Head louse_9 NA 34,43 Head louse_10 NA 35,87 Head louse_11 NA 33,92 Head louse_12 NA 39,17 Amazonia Head louse_14 NA 35,69 Head louse_172 NA 37,51 Head louse_173 NA 36,56 Brazil Head louse_175 NA 37,32 Head louse_20 NA 35,47 Head louse_21 NA 37,84 Head louse_23 NA 35,22 Sao Cristovao Head louse_24 NA 39,33 Head louse_25 NA 38,18 Head louse_29 NA 23,20 Head louse_30 NA 36,56 Head louse_31 NA 35,45 Head louse_34 NA 38,16 Australia Brisbane Head louse_36 NA 37,01 Head louse_37 NA 37,88 Head louse_50 NA 36,43 Head louse_51 NA 36,33 Head louse_54 NA 36,26 Papua New Guinea Highlands Head louse_55 NA 36,21 Head louse_56 NA 37,64 Head louse_60 NA 35,71 Head louse_61 NA 35,94 Head louse_62 NA 37,74 Head louse_63 NA 36,43 Head louse_64 NA 38,07 New Zeeland Auckland Head louse_66 NA 37,20 Head louse_67 NA 36,89 Head louse_68 NA 37,45 Head louse_69 NA 37,16 Head louse_160 NA 24,45 Head louse_161 NA 24,79 Head louse_162 NA 24,54 Head louse_163 NA 36,13 Head louse_165 NA 24,09 Madagascar Bedaro village Head louse_166 NA 28,02 Head louse_168 NA 26,78 Head louse_170 NA 26,23 Head louse_171 NA 36,16

73 Head louse_172 NA 28,20 Head louse_173 NA 29,55 Head louse_174 NA 27,44 Head louse_48 NA 27,54 Head louse_95 NA 28,17 Senegal Dakar Head louse_96 NA 24,03 NTC NA NA

NA: signal not detected; NTC: non template control

Article II: Bartonella quintana in Body Lice from Scalp Hair of Homeless Persons, France

Emerging Infectious Diseases 20: 907-908.

LETTERS

Address for correspondence: Pietro E. (5). While surveying for trench fever controls. Negative controls were in- Varaldo, Unità di Microbiologia, Dipartimento among homeless persons in shelters cluded in each assay. di Scienze Biomediche e Sanità Pubblica, in Marseille, France during October Of the hatched larvae, 5 (6%) of Università Politecnica delle Marche, Via Tronto 2012–March 2013, we investigated the 83 recovered from clothing and 10/A, 60126 Ancona, Italy; email: pe.varaldo@ the presence of B. quintana DNA in 7 (11%) of 66 from the hair (Table) univpm.it nits, larvae, and adult lice collected of 4 of the 7 dually infested persons from mono-infested and dually infest- were positive for B. quintana DNA ed persons and determined the geno- (online Technical Appendix Table 1 types of the specimens. wwwnc.cdc.gov/EID/article/20/5/13- The persons included in this study 1242-Techapp1.pdf). Of the 840 adult received long-lasting insecticide-treat- body lice, 174 (21%) collected from ed underwear; lice were collected by 42 (53%) of 80 of the mono-infested removing them from clothing, includ- persons contained B. quintana DNA ing underwear, pants, and shirts. Be- (Table, online Technical Appendix 2). Bartonella cause body lice reside in the clothing of The multiplex real-time PCR that tar- quintana in Body infested persons except when feeding, geted the PHUM540560 gene clearly they are sometimes called clothing lice. identified all nits, larvae, and adult lice Lice from Scalp A total of 989 specimens were as belonging to the body lice lineage. Hair of Homeless tested, including 149 (83 from cloth- Negative controls remained negative Persons, France ing and 66 from hair) first–instar lar- in all PCR-based experiments. vae hatched in the laboratory from For 2 decades, B. quintana DNA To the Editor: Bartonella quin- eggs collected from 7 dually infested has been regularly detected in lice col- tana is a body louse–borne human persons, and 840 adult body lice collected from the heads of persons living pathogen that can cause trench fever, lected from the clothing of 80 mono- in poverty, but it had not been detected bacillary angiomatosis, endocardi- infested patients. We included DNA in head lice that infest schoolchildren tis, chronic bacteremia, and chronic isolated from 3 nits collected from the (7,8). All of the lice collected during lymphadenopathy (1). Recently, B. hair of a mono-infested person who this study that tested positive for B. quintana DNA was detected in lice had previously been confirmed as pos- quintana from homeless persons were collected from the heads of poor and itive for B. quintana 6) (Table). body lice, including some that were homeless persons from the United Total DNA was extracted by us- recovered from hair. This observation States, Nepal, Senegal, Ethiopia, and ing an EZ1 automated extractor (QIA- supports our assertion that body lice the Democratic Republic of the Congo GEN, Courtaboeuf, France) and sub- are not confined to the body. The 3 eggs and in nits in France (2,3). The head jected twice to real-time PCR specific that were removed from the hair of a louse, Pediculus humanus capitis, and for B. quintana. The first PCR targeted mono-infested homeless person whose the body louse, Pediculus humanus the 16S-23S intergenic spacer region. samples tested positive for B. quintana humanus, are obligatory ectoparasites Positive samples were confirmed by were also body lice. During the clinical that feed exclusively on human blood using a second real-time PCR target- examination, no adult head lice or adult (4). Outside of their habitats, the 2 ing the yopP gene (6). Samples that body lice were found on that person, con- ecotypes are morphologically indis- tested positive for B. quintana DNA firming that the patient had been heav- tinguishable (1). Sequence variation were analyzed by multiplex real-time ily infested with body lice in the past, in the PHUM540560 gene discrimi- PCR that targeted the PHUM540560 not head lice. The nits were most likely nates between head and body lice by gene (5). We used head and body lice laid by body lice that migrated toward determining the genotype of the lice that had known genotypes positive the patient’s head. When a member of

Table. Distribution of Bartonella quintana DNA in nits, larvae, and adult body lice collected from hair and clothing of homeless persons in shelters, Marseille, France, October 2012–March 2013* No. persons No. (%) lice positive for Location Dually infested, n = 7 Monoinfested, n = 80 B. quintana DNA Reference Hair Nits 0 3 3 (100) (6) Hatched larvae 66 0 7 (10.60) This study Clothing Hatched larvae 83 0 5 (6.00) This study Adults 0 840 174 (20.70) This study *All lice were identified as body lice. Study participants were provided with long-lasting insecticide-treated underwear, and killed body lice were collected from the clothing of infested persons.

Emerging Infectious Diseases • www.cdc.gov/eid77 • Vol. 20, No. 5, May 2014 907 LETTERS this research team (DR) collected the References Myasthenia Gravis eggs from the hair shaft, they were 1. Brouqui P. Arthropod-borne diseases Associated with found ≈3 3.5 cm from the hair fol- associated with political and social disor- licle. Because hair grows ≈1.25 cm per der. Annu Rev Entomol. 2011;56:357–74. Acute Hepatitis E month, the louse infestation occurred http://dx.doi.org/10.1146/annurev-ento- Infection in ≈3 months before egg collection (6). 120709-144739 2. Boutellis A, Mediannikov O, Bilcha Immunocompetent Homeless persons that we have KD, Ali J, Campelo D, Barker SC, et al. monitored for many years are often Borrelia recurrentis in head lice, Ethiopia. Woman heavily infested by body lice but are Emerg Infect Dis. 2013;19:796–8. also occasionally infested with head 3. Piarroux R, Abedi AA, Shako JC, To the Editor: Hepatitis E vi- Kebela B, Karhemere S, Diatta G, et al. lice. Before genetic tools that differen- Plague epidemics and lice, Democratic rus (HEV) is a common cause of tiate the head and body louse lineages Republic of the Congo. Emerg Infect Dis. acute hepatitis in developing coun- were available (5), it was speculated 2013;19:505–6. http://dx.doi.org/10.3201/ tries. The course of acute hepatitis E that body lice may have originated eid1903.121542 is usually benign, except in pregnant 4. Gratz NG. Emerging and resurging from head lice (9). From our study, it is vector-borne diseases. Annu Rev Entomol. women and in immunocompromised clear that under conditions of massive 1999;44:51–75. http://dx.doi.org/10.1146/ patients, who are prone to a lethal or infestation, body lice can migrate and annurev.ento.44.1.51 chronic outcome of the disease. Since colonize hair; the opposite may also be 5. Drali R, Boutellis A, Raoult D, 2001, hepatitis E has been emerg- Rolain JM, Brouqui P. Distinguishing body true. However, there is no evidence that lice from head lice by multiplex real-time ing in industrialized countries, and body lice are capable of causing an out- PCR analysis of the Phum_PHUM540560 neurologic manifestations such as break of lice living on the head, as hap- gene. PLoS ONE. 2013;8:e58088. http:// Guillain-Barré syndrome, brachial pens among schoolchildren that have dx.doi.org/10.1371/journal.pone.0058088 neuritis, transverse myelitis, and cra- 6. Angelakis E, Rolain JM, Raoult D, been found to be infested only by head Brouqui P. Bartonella quintana in nial nerve palsies have been reported lice. This suggests that body lice can- head louse nits. FEMS Immunol Med in patients with acute or chronic forms not thrive in the environment of head Microbiol. 2011;62:244–6. http://dx.doi. of the disease (1–6). Most cases with lice, which infest millions of children org/10.1111/j.1574-695X.2011.00804.x neurologic manifestations have been 7. Fournier PE, Ndihokubwayo JB, Guidran J, worldwide (10), further suggesting Kelly PJ, Raoult D. Human pathogens characterized by infection with geno- that outbreaks of trench fever are most in body and head lice. Emerg Infect Dis. type 3 HEV. Data are not available to likely not linked to head lice in indus- 2002;8:1515–8. http://dx.doi.org/10.3201/ indicate whether this association be- trialized countries. In conclusion, by eid0812.020111 tween HEV infection and neurologic 8. Bouvresse S, Socolovshi C, Berdjane Z, analyzing lice harvested from the heads Durand R, Izri A, Raoult D, et al. No manifestations is related to a specific and clothing of homeless persons, we evidence of Bartonella quintana but antigenic stimulus provided by HEV have shown that the 2 ecotypes belong detection of Acinetobacter baumannii or is linked to the more comprehen- to the same body lice population. in head lice from elementary schoolchil- sive assessment for such neurologic dren in Paris. Comp Immunol Microbiol conditions that is available in industri- The text has been edited by American Infect Dis. 2011;34:475–7. http://dx.doi. org/10.1016/j.cimid.2011.08.007 alized countries or to a reporting bias. Journal Experts under certificate verifica- 9. Li W, Ortiz G, Fournier PE, Gimenez G, We report a case of HEV infection in tion key 51F8-8A17-1F51-90DD-705C. Reed DL, Pittendrigh B, et al. Geno- an immunocompetent woman who typing of human lice suggests multiple emergencies of body lice from local head had muscle-specific kinase (MuSK) Rezak Drali, louse populations. PLoS Negl Trop Dis. antibody–positive myasthenia gravis Abdoul Karim Sangaré, 2010;4:e641. http://dx.doi.org/10.1371/ associated with HEV replication. Amina Boutellis, journal.pntd.0000641 A 33-year-old woman was hos- Emmanouil Angelakis, 10. Chosidow O, Chastang C, Brue C, Bouvet E, Izri M, Monteny N, et al. pitalized in France for subacute as- Aurélie Veracx, Controlled study of malathion and thenia and intermittent symptoms Cristina Socolovschi, d-phenothrin lotions for Pediculus huma- including dysarthria, dysphagia, Philippe Brouqui, nus var capitis-infested schoolchildren. muscle weakness, and diplopia. She and Didier Raoult Lancet. 1994;344:1724–7. http://dx.doi. org/10.1016/S0140-6736(94)92884-3 had no family history of autoimmune disease and no notable personal Author affiliations: Aix Marseille Université, Address for correspondence: Didier Raoult, medical history; she had not received Marseille, France; and Institut Hospitalo- Institut Hospitalo-Universitaire Méditerranée any recent vaccinations and had not Universitaire Méditerranée Infection, Infection, 27 boulevard Jean Moulin, 13385 traveled outside France during the 13005, Marseille Marseille CEDEX 5, France; email: didier. previous year. Physical examination DOI: http://dx.doi.org/10.3201/eid2005.131242 [email protected] showed no pyramidal, vestibular, or

908 Emerging Infectious Diseases • www.cdc.gov/eid78 • Vol. 20, No. 5, May 2014 Article DOI: http://dx.doi.org/10.3201/eid2005.131242 Bartonella quintana in Body Lice from Scalp Hair of Homeless Persons, France

Technical Appendix

Detailed distribution of B. quintana DNA among lice from mono- infested and dually infested homeless persons, France

Technical Appendix Table 1: Distribution of B. quintana DNA among lice from dually infested homeless persons, France, October 2012–March 2013 Lice collected from body Lice collected from head ID code homeless persons Lice tested B. quintana no. (%) Lice tested B. quintana no. (%) 32 5 0 4 0 33 5 2 (40.00) 2 1 (50.00) 40 5 1 (20.00) 4 0 89 29 0 6 0 B 17 0 30 5 (16.60) D 11 2 (18.20) 10 1 (10.00) Nov 11 0 10 0 Total 83 5 (6.00) 66 7 (10.60)

Technical Appendix Table 2: Detail of distribution of B. quintana DNA among lice from mono-infested homeless persons, France, October 2012–March 2013 ID no. homeless persons Body lice B. quintana 1 3 0 2 3 1 3 3 3 4 3 0 5 3 0 6 3 0 7 3 3 8 3 3 1001 69 13 1002 17 7 1003 18 6 1005 3 0 1022 10 0 1023 3 3 1029 15 9 1034 3 1 1037 8 2 1038 101 3 1040 5 1 1051 2 0 1052 74 15 1053 19 2 1054 3 1 1059 19 0 1060 3 3 1065 4 2 1066 2 1 1070 7 3 1087 23 4 1101 2 2 1105 1 0

79 ID no. homeless persons Body lice B. quintana 1151 10 5 1152 17 2 1153 1 0 1155 6 2 1158 4 0 1159 5 4 1160 3 0 1161 4 0 1163 1 1 1200 9 4 1201 24 0 1202 1 0 1203 1 0 1204 2 0 1205 1 0 1206 5 0 1208 21 2 1209 2 2 1210 23 13 1211 16 10 1212 5 1 1213 4 0 1222 46 1 1223 28 5 1224 2 0 1225 2 0 1227 2 0 1228 2 0 1229 18 3 1230 2 0 1231 6 1 1232 2 0 1234 10 1 1236 4 0 1237 44 19 1238 2 0 1239 2 0 1240 11 5 1241 1 1 1242 5 4 1244 2 0 1245 3 0 1246 7 0 2002 1 0 2016 34 0 2040 1 0 2101 1 0 2103 2 0 Total (%) 840 174 (20.7)

Article III: Detection of Bartonella quintana in African Body and Head Lice

American Journal of Tropical Medicine and Hygiene 91: 294-301.

Detection of Bartonella quintana in African Body and Head Lice

Abdoul Karim Sangare´, Amina Boutellis, Rezak Drali, Cristina Socolovschi, Stephen C. Barker, Georges Diatta, Christophe Rogier, Marie-Marie Olive, Ogobara K. Doumbo, and Didier Raoult* Unite´ de Recherche sur les Maladies Infectieuses et Tropicales Emergentes (URMITE), Unite´ Mixte de Recherche (UMR)63, 7278 Centre National de la Recherche Scientifique (CNRS), Institut de Recherche pour le De´veloppement (IRD) 198, Institut National de la Sante´ et de la Recherche Me´dicale (INSERM) 1095, University of Aix, Marseille, France; IRD, Campus Commun Universite´ Cheikh Anta Diop (UCAD)-IRD of Hann, Dakar, Senegal; Parasitology Section, School of Chemistry and Molecular Biosciences (SCMB), University of Queensland, Brisbane, Queensland, Australia; Pasteur Institute of Madagascar, Ambohitrakely, Madagascar; University of Bamako, Malaria Research and Training Center (MRTC)/De´partement d’Epidemiologie des Affections Parasitaires (DEAP)/Faculte´ de Me´decine de Pharmacie et d’Odontostomatologie (FMPOS)-Unite´ Mixte Internationale (UMI)3189, Bamako, Mali

Abstract. Currently, the body louse is the only recognized vector of Bartonella quintana, an organism that causes trench fever. In this work, we investigated the prevalence of this bacterium in human lice in different African countries. We tested 616 head lice and 424 body lice from nine African countries using real-time polymerase chain reaction targeting intergenic spacer region 2 and specific B. quintana genes. Overall, B. quintana DNA was found in 54% and 2% of body and head lice, respectively. Our results also show that there are more body lice positive for B. quintana in poor countries, which was determined by the gross domestic product, than in wealthy areas (228/403 versus 0/21, P< 0.001). A similar finding was obtained for head lice (8/226 versus 2/390,P= 0.007). Our findings suggest that head lice in Africa may be infected by B. quintana when patients live in poor economic conditions and are also exposed to body lice.

INTRODUCTION mitted by body lice, but recently, DNA of B. quintana has also been found in head lice collected from homeless individuals in Sucking lice (Anoplura) are hematophagous wingless 9 16 17 11 1 Nepal, the United States, France, Senegal, insects that can infest birds and mammals. Among these and Ethiopia.10,18 hosts, humans constitute the preferred host for only two spe- The detection of B. quintana in African lice remains limited cies: Pediculus humanus and Pthirus pubis (pubic lice). to only a small number of countries. Therefore, the objectives P. humanus includes two morphotypes: P. humanus morpho- of this study were to investigate the presence of B. quintana in type capitis (head lice) and P. humanus morphotype humanus 2 head and body lice in different areas of African countries (body lice). Each louse has a specific ecotype; head lice live suffering from poverty, social instability, or war and identify and lay eggs in the hair and are prevalent in all countries and the relationship between louse phenotypes and genotypes. all levels of society, whereas body lice live in clothing and multiply when cold, promiscuity, and lack of hygiene are pres- 3 ent. Louse coloration was described at the beginning of 20th MATERIALS AND METHODS century. The variability in louse color on a single host may be affected by not only the color of the skin but also, the color of Ethics statement. Lice from African countries were obtained the hair and clothing.4,5 from the private frozen collection of our laboratory (The Unite´ Many genetic studies have been conducted on human lice— de Rcherche sur les Maladies Infectieuses et Tropicales first based on 18S ribosomal RNA6,7 and then, the mitochon- Emergentes [URMITE]/World Health Organization [WHO] drial genes (cytochrome oxidase subunit 1 [Cox1] and cyto- Collaborative Research Center). The lice in that collection chrome b [Cytb]). These studies allowed scientists to classify were required for various epidemiological and entomological lice into three different clades: Clade A, the most common studies or to perform diagnoses abroad, and they were sent to clade found worldwide and comprised of both head and body our laboratory as a WHO reference facility. Body lice were lice; Clade B, composed of only head lice and found in Cen- collected from clothing and head lice were removed from hair tral and North America, Europe, and Australia8; and Clade C, with the verbal consent of the infested individuals. Written including black head lice found in Nepal,9 Ethiopia,10 consent was not obtainable in the majority of cases, because and Senegal.11 most of the subjects were illiterate. However, in most instances, The body louse is linked to poverty. Its transmission occurs the investigator, local authorities/Institutional Review Boards in crowded environments, such as homeless shelters, refugee (IRBs), and/or village/family chief approved and were present camps, and jails, especially when hygienic standards are lack- when collection was performed (with individual consent). ing.12 The body louse is the main vector of three pathogenic Sampling/country. Human lice samples were collected bacteria: Bartonella quintana, the agent of trench fever; from patients in various regions of Africa (Figure 1). Nine Borrelia recurrentis, the agent of louse-borne relapsing fever; countries were investigated. In total, 1,040 lice were col- and Rickettsia prowazekii, the agent of epidemic typhus.13,14 lected: 616 head lice and 424 body lice. Exactly 381 head lice B. quintana is a Gram-negative bacterium that causes trench were collected in Senegal (2011), 75 head lice and 22 body fever, bacillary angiomatosis, endocarditis, chronic bacteremia, lice were collected in Madagascar (2009, 2011, and 2012), and chronic lymphadenopathy.15 It has typically been trans- 14 head and body lice were collected in Ethiopia (2011), 92 head lice were collected in Mali (2010), 37 body lice were collected in Kenya (1999), 166 body lice were collected in *Address correspondence to Didier Raoult, URMITE, UM63, 7278 Rwanda (2011), 10 head and body lice were collected in CNRS, IRD 198, Inserm 1095, University of Aix, 27 Bd Jean Moulin, Burundi (2008), 35 head lice and 154 body lice were collected 13005 Marseille, France. E-mail: [email protected] in Congo (2010), and 9 head lice and 21 body lice were

29483 BARTONELLA QUINTANA IN AFRICAN LICE 295

Figure 1. Map of Africa showing the prevalence of B. quintana in head and body lice collected in the various study areas during 1999–2012. collected in Algeria (2000). In our laboratory, each sample in half on the longitudinal plane, and one-half was stored at was photographed on dorsal and ventral sides with a camera −20°C for subsequent analysis. (Olympus DP71, Rungis, France). Lice were preserved in Distribution of regions. We classified these countries into 70% alcohol, rinsed two times with sterile distilled water for two regions: poor (with the gross domestic product [GDP]< 2 minutes, and dried with filter paper. Samples were then cut $1,844 per capita: Madagascar, Ethiopia, Mali, Kenya,

Figure 2. Phenotypes of head and body lice collected from African countries during 1999–2012 (Head lice: A, B and C; Body lice: D, E, F and G).

84 296 SANGARE´ AND OTHERS

Rwanda, Burundi, and Congo) and wealthy (with GDP³ that encodes a hypothetical intracellular effector.17 Each real- $1,844 per capita: Algeria and Senegal) areas. GDP being an time PCR assay was performed using a CFX96 TM REAL- economic indicator of the wealth produced annually in a partic- Time System C1000 Thermal Cycler (Bio-Rad Laboratories, ular country for our classification, we estimated the average per Foster City, CA). For each PCR assay, two negative and capita GDP of all regions.19 The average GDP corresponding positive controls were used. The identification of B. quintana to the total of GDP per capita from nine countries divided by was confirmed by amplification of both ITS2 and yopP genes. nine ($16,600/9= $1,844) was then compared with the infection Genotypic status of lice. All samples that tested positive for rate of B. quintana in lice. B. quintana DNA were analyzed by multiplex real-time PCR DNA preparation and detection of B. quintana in lice. that targeted a portion of the Phum PHUM540560 gene. This After incubation at 56°C in a dry bath overnight, the lice were assay allowed for the discrimination of body lice from head extracted on an automaton EZ1 device using the QIAamp lice, which has described previously.20 As a positive control, Tissue Kit (Qiagen, Hilden, Germany) according the manu- we used a head louse and body louse with genotypic statuses facturer’s recommendations. The DNA was used as a tem- that were known (VIC-positive for the head louse and FAM- plate in a real-time polymerase chain reaction (PCR) assay positive for the body louse). Monoplex and multiplex real- targeting a portion of the Bartonella 16S–23S intergenic time PCRs were performed in a CFX96 Thermal Cycler spacer (ITS) region and a specific B. quintana gene (yopP) (Bio-Rad Laboratories, Foster City, CA).

Figure 3. Phylogenetic tree of lice based on the Cytb (using the ML method with 100 bootstrap replicates).

85 BARTONELLA QUINTANA IN AFRICAN LICE 297

Standard PCR and sequencing. Lice genotype was identi- with lice. Statistical analysis was performed with Epi Info 6 fied by the analysis of partial (347-bp) mitochondrial Cytb (www.cdc.gov/epiinfo/Epi6/EI6dnjp.htm), and aP value of DNA. The primers used for the Cytb gene were CytbF1 (5¢- < 0.05 was considered significant. GAGCGACTGTAATTACTAATC-3¢) and CytbR1 (5¢-GGA Phylogenetic analysis. Each DNA sequence was aligned CCCGGATAATTTTGTTG-3¢). Overall, 10 lice belonging to using multisequence alignment software (CLUSTALX, ver- each region were tested with a standard PCR assay targeting sion 2.0.11). The ChromasPro program (Technelysium PTY, the Cytb gene as described previously.8 Each amplification Australia) was used to analyze, assemble, and correct sequences. reaction was performed on a PTC-200 thermocycler machine The sequence similarities were determined using MEGA 5, (MJ Research, Waltham, MA). The PCR reaction contained and phylogenetic trees were obtained using the maximum 21 9.9mL water, 4mL buffer 5 + Fusion HF Buffer, 0.4mL 10 mM likelihood (ML) method with 100 bootstrap replicates. deoxynucleoside triphosphates (dNTPs), 1.25mL each primer, We compared our sequences with other sequences that are 0.2mL Phusion polymerase (Finnzymes, Thermo Scientific, present in GenBank. Vantaa, Finland), and 3mL DNA template to obtain a total volume of 20mL. The cycling conditions were 98°C for 30 sec- RESULTS onds, 35 cycles of 98°C for 5 seconds, 56°C for 30 seconds, and 72°C for 15 seconds, and a final extension time of Morphological analysis. Phenotypical examination of the 5 minutes at 72°C. The success of the PCR amplification was lice showed that all head lice from Senegal, Madagascar, and then detected by the migration of the PCR product on 2% Ethiopia have a black color (Figure 2A–C). The color of all agarose gel (Electrophoresis grade; Invitrogen, Carlsbad, CA) body lice in Madagascar and Rwanda is black; in Kenya, all prepared with 0.5% Tris Borate ethylenediaminetetraacetic body lice are brown, whereas in Ethiopia, all body lice are gray (EDTA) (TBE; Euromodex, Lake Placid, NY) and charged (Figure 2D–G). During our study, the phenotypes of lice from with a solution of 0.5% ethidium bromide (Invitrogen, Congo, Burundi, Mali, and Algeria were not investigated, Carlsbad, CA). Purification of PCR-amplified products was because they had been destroyed previously for other studies. performed using distilled RNase-DNase free water on Phylogenetic analysis. On the basis of phylogenetic study NucleoFast 96 PCR plates (Macherey-Nagel EURL, Hoerdt, of lice Clade A based on ITSs (ultispacer typing method France). Sequencing of positive samples was performed [MST])22 and through the analysis of Cytb in this work using the ABI Prism Big Dye Terminator Cycle Sequencing (Figure 3), our results show that head lice from Algeria, Kit, version 1.1 (Applied Biosystems, Foster City, CA) accord- Madagascar, Burundi, and Senegal belong to Clade A2. ing to the manufacturer’s instructions. Clade C was found in head lice from Ethiopia, Mali, and Data analysis. Microsoft Excel was used for data manage- Senegal. Clade B was found in head lice from Algeria. Based ment. Descriptive statistics, such as percentages and means, on Cytb analysis, all head and body lice that are positive for were computed to summarize the proportions of infestations B. quintana DNA belong to Clade A2 (Table 1).

Table1 Distribution of B. quintana DNA in head lice (N= 616) and body lice (N= 424) collected from African countries during 1999–2012 Lice analyzed Positive B. quintana DNA (%) Cytb

Country/area Head lice Body lice Head lice Body lice Head lice Body lice Senegal Yeumbeul 101 0 0− Clade A2− Malika 69 0 0− Clade A2− Dielmo village 71 0 2 (2.8)− Clade A2− Ndiop village 77 0 0 (0)− Clade C− Keur Massar 63 0 0− Clade A2− Total 381 0 2 (0.52) Madagascar Borenty village 0 9− 1 (11.1)− Clade A2 Tsiroanomandidy 21 13 0 0 Clade A2 Clade A2 Anjozorobe 39 0 2 (5.1)− Clade A2− Bedaro village 15 0 0− Clade A2− Total 75 22 2 (2.66) 1 (4.54) Ethiopia Gondar 14 14 0 0 Clade C Clade A2 Mali Diankabou 92 0 0− Clade C− Kenya Nairobi 0 37− 27 (72.9)− Clade A2 Rwanda Kigali 0 166− 149 (89.7)− Clade A2 Burundi Bujumbura 10 10 0 1 (10) Clade A2 Clade A Congo RDC 35 154 6 (17.1) 50 (32.5) Clade A2 Clade A2 Algeria Batna 9 21 0 0 Clades A and B Clade A2 Total 616 424 10 (1.6) 228 (53.7)

86 298 SANGARE´ AND OTHERS

Molecular detection of B. quintana in lice. The DNA of are from Rwanda, 27 of 37 (72.9%) body lice are from Kenya, B. quintana was detected in 10 of 616 (1.6%) head lice (the 50 of 154 (32.5%) body lice are from the Democratic Repub- mean cycle thresholds [ct] value ± SD: 30.64 ± 6.34). Specifi- lic of Congo,23 1 (11.1%) body louse is from Madagascar, and cally, B. quintana DNA was found in 2 of 71 (2.8%) head lice 1 (10%) body louse is from Burundi. No B. quintana DNA from Senegal (Dielmo village), 2 of 39 (5.1%) head lice was detected in the Madagascar area of Tsiroanomandidy, from Madagascar (Anjozorobe), and 6 of 35 (17.1%) head Ethiopia, or Algeria. There are significantly more body lice lice from Congo.23 No B. quintana DNA was detected in with B. quintana DNA than head lice (54% versus 1.6%,P< Senegal (Yeumbeul, Malika, Ndiop village, and Keur Massar), 0.001) (Table 1). Madagascar (Tsiroanomandidy and Bedaro village), Ethiopia, Genotypic status of positive lice to B. quintana DNA. Mali, Burundi, or Algeria (Table 1). Among the African Clade A lice positive for B. quintana The DNA of B. quintana was detected in 228 of 424 (54%) DNA, all head lice had the head louse genotype, and all body body lice (the mean ct value ± SD: 33.19 ± 3.79). Among 228 lice had the body louse genotype. No signal was detected in B. Quintana-positive body lice, 149 of 166 (89.7%) body lice the negative controls.

Table2 Prevalence of infections of B. quintana in body lice (this study and the literature) Country/source Year of collection Percent (no. B. quintana/no. tested) Source Algeria Schoolchildren 2000 0 (0/21) This study Homeless in Batna 2001 0 (0/33) 30 Madagascar Local population 2009/2011/2012 4.5 (1/22) This study Ethiopia Bahir Dar 2011 0 (0/14) This study Poor regions (Jimma) 2010 3 (1/33) 29 Poor regions (Jimma) 2010 18 (76/424) 10 Kenya Local population 1999 72.9 (27/37) This study Rwanda Jail 2011 89.7 (149/166) This study Jail 2001 2.3 (6/262) 30 Burundi Refugee camp 2008 10 (1/10) This study During typhus outbreak 1997 0 (0/10) 32 Refugee camp 1997 9.5 (6/63) 32 After outbreak in camp 1998 14.3 (13/91) 32 During typhus outbreak 1997 0 (0/10) 30 Refugee camp 1997 9.5 (6/63) 30 After typhus outbreak 1998 14.3 (13/91) 30 Refugee camp 1998 21 (8/38) 30 Refugee camp 2000 90 (100/111) 30 Refugee camp 2001 93.9 (31/33) 30 Congo Local population 2010 32.5 (50/154) This study Refugee camp 1998 0 (0/7) 32 Refugee camp 1998 0 (0/7) 30 Local population 2010 32.5 (50/154) 23 France SDF (homeless) 1997 20 (3/15) 34 SDF (homeless) 1998 4 (3/75) 32 SDF (homeless) 2000 26 (42/161) 33 SDF (homeless) 1998–2001 9.8 (32/324) 30 Tunisia Homeless in Sousse 2000 0 (0/3) 30 Zimbabwe Homeless in Harare 1998 16.7 (2/12) 30–32 The Netherlands Homeless in Utrecht 2001 36 (9/25) 30 United States SDF (homeless) 2007–2008 33.3 (11/33) 16 Nepal Street children and children in slums 2002 20 (4/20) 9 Russia Homeless in Moscow 1998 12.3 (33/268) 32 Australia Homeless 2001 0 (0/2) 30 Peru Andean rural population NA 1.4 (1/73) 32 Andean rural population NA 0 (0/10) 30 NA= not available.

87 BARTONELLA QUINTANA IN AFRICAN LICE 299

Relationship between B. quintana in body or head lice and lected in this country (Figure 2). In addition, it seems that the socioeconomic level. We compared our results with the life color of lice is more complex than indicated in the first find- conditions of the regions tested and found that the body lice ings provided in 1926 by Ewing,25 and as suggested by a recent that were B. quintana-positive were more likely to be found in study,10 lice color may be independent of the host’s skin color. poor countries than wealthy countries (228/403 versus 0/21, On the basis of phylogenetic study of lice Clade A, based P< 0.001). The same finding was made for head lice (8/226 on ITSs (MST)22 and through the analysis of Cytb in this work versus 2/390,P= 0.007). (Figure 3), we confirmed that head lice in Senegal are Clades We correlated the GDP to the level of B. quintana infec- A2 and C11 and that head lice in Ethiopia are Clade C.10 In tion. Our results indicate that the higher the GDP in a region, addition, we have found that head lice in Madagascar, the less prevalent that B. quintana DNA was and vice versa Burundi, Congo, and Algeria belong to Clade A2 and that (correlationr=−0.178). head lice in Mali are Clade C (Figure 3). The body lice of Algeria, Ethiopia, Madagascar, Kenya, Burundi, and Rwanda DISCUSSION belong to Clade A2; it is the main clade in sub-Saharan Africa.26 Therefore, the geographical distribution of lice This study of 1,040 human lice of the genus Pediculus col- seems to be complex and independent of the phenotype. lected from nine African countries has enabled us to better B. quintana is a re-emerging pathogen that is responsible specify some characteristics of human lice. The phenotypic for a range of clinical manifestations in humans.27 It has long study allowed us to confirm several previously described been established that trench fever can be transmitted by the observations, including the black phenotype of head lice in body louse.28 However, the role played by the head louse as a Senegal and Ethiopia.10,11,24 For the first time, we have reservoir or vector of B. quintana remains unclear.29 In total, established the phenotype of lice collected in Madagascar B. quintana was found in 54% and 2% of body and head lice and Kenya. Body lice from Madagascar and Rwanda are collected in Africa, respectively. Here, we find that there are black; body lice collected in Kenya are brown. The head lice more B. quintana in body lice than head lice. This finding from Madagascar had the same black color as body lice col- could probably be explained through the role played by

Table3 Prevalence of infections of B. quintana in head lice (this study and the literature) Country/source Year of collection Percent (no. B. quintana/no. tested) Source Algeria Schoolchildren 2000 0 (0/9) This study Schoolchildren NA 0 (0/18) 30 Burundi Schoolchildren 2008 0 (0/10) This study Schoolchildren NA 0 (0/20) 30 Senegal Rural community 2011 0.52 (2/381) This study Rural community NA 6.9 (19/274) 11 Ethiopia Poor region 2011 0 (0/14) This study Poor regions (Jimma) 2010 7 (19/271) 10 Poor regions (Jimma) 2010 9.2 (6/65) 29 Congo Local population 2010 17.1 (6/35) This study Local population 2010 17.1 (6/35) 23 Madagascar Local population 2010–2011 2.6 (2/75) This study Mali Schoolchildren 2010 0 (0/92) This study Portugal Schoolchildren NA 0 (0/20) 30 China Schoolchildren NA 0 (0/23) 30 Thailand Schoolchildren NA 0 (0/29) 30 Australia Schoolchildren NA 0 (0/3) 30 United States SDF (homeless) 2007–2008 25 (3/12) 16 Nepal Street children and children in slums 2002 9.5 (2/21) 9 Russia Schoolchildren NA 0 (0/10) 30 France Schoolchildren NA 0 (0/20) 30 Schoolchildren 2008–2009 0 (0/288) 31 SDF (homeless) 2008 100 (3/3) 17 NA= not available.

88 300 SANGARE´ AND OTHERS the body louse in transmission of diseases. With the new B. quintana-positive were more likely to be found in poor method of multiplex real-time PCR assay with the Phum_ countries (GDP< $1,844 per capita) than wealthy countries PHUM540560 gene, all head lice were genotyped as head lice (GDP> $1,844 per capita; 8/226 versus 2/390,P< 0.007). by the signal emitted by the VIC-labeled probe specific to Head lice are prevalent around the world and in all levels head lice, and all body lice were genotyped as body lice by of society. Thus, head lice may be infected with B. quintana the signal emitted by the FAM-labeled probe specific to body when the host is coinfected with body lice in a precarious lice. In our study, the infection rate was higher in Rwanda (149/ environment in which proper hygiene is lacking. 166 versus 6/262,P< 0.001) (Table 2) and lower in Burundi (1/ In conclusion, it has been estimated that trench fever 10 versus 158/346,P= 0.025) than the rate reported in 2002 by affected several million people, especially in Russia and on Fournier and others.30 These differences could be explained by the Eastern, Central, and Western European fronts during the living conditions in jail in Rwanda and probably, the sam- World War II.12,15 Our study on the prevalence of B. quintana ple size in this study in Burundi. We have confirmed the pres- in lice from various geographical and socioecological situa- ence of B. quintana in body lice in Rwanda and Burundi.30 tions associated with clinical information will help to evaluate B. quintana has also been detected in body lice from homeless the role of head lice in the transmission of this disease. people in Zimbabwe (16.7%, 2/12).30 This bacterium has not been found in body lice in Tunisia.30 We detected B. quintana Received December 5, 2013. Accepted for publication February 14, for the first time in body lice collected in Madagascar (4.54%, 2014. 1/22) and Kenya (72.9%, 27/37). The high rate of B. quintana in Published online June 16, 2014. body lice in some African countries (Kenya, Rwanda, and Acknowledgments: The authors thank Mr. Jean-Michel Berenger Congo) may reflect the low socioeconomic level of the study (entomologist) and the molecular biology team of URMITE Marseille population. We tested this hypothesis by comparing the aver- for their technical support. age GDPs of nine different countries with the level of Authors’ addresses: Abdoul Karim Sangare´, Amina Boutellis, Rezak B. quintana infection. We found that the higher the GDP Drali, Cristina Socolovschi, Georges Diatta, and Didier Raoult, increases, the lower the level of B. quintana decreases and vice URMITE, UM63, 7278 CNRS, IRD 198, Inserm 1095, University of versa (correlationr=−0.178). For example, for countries with Aix, Marseille, France, and IRD, Campus Commun UCAD-IRD of Hann, Dakar, Senegal, E-mails: [email protected], amina. high GDP, this hypothesis could be justified through the previ- [email protected], [email protected], cr_socolovschi@yahoo. 30 ous work by Fournier and others in 2002, which reported 0% com, [email protected], and [email protected]. Stephen C. B. quintana in head lice in Portugal, China, Thailand, Australia, Barker, Parasitology Section, School of Chemistry and Molecular Bio- Algeria, and France.30 This percentage of infection with sciences (SCMB), University of Queensland, Brisbane, Queensland, B. quintana remains stable (0%) in body lice in Australia, Australia, E-mail: [email protected]. Christophe Rogier and Marie- 31 Marie Olive, Pasteur Institute of Madagascar, Ambohitrakely, Algeria, and Peru. In France, Bouvresse and others also Madagascar, E-mails: [email protected] and [email protected]. found the same result (0%) in head lice. Roux and Raoult32 Ogobara K. Doumbo, University of Bamako, MRTC/DEAP/FMPOS- reported a rate of 1.4% in body lice in Peru. For countries with UMI3189, Bamako, Mali, E-mail: [email protected]. low GDP, this hypothesis could be confirmed by the work of Piarroux and others23 in 2010, which reported 32.5% B. quintna in body lice in Congo, and the work of Fournier and others,30 REFERENCES which found a higher prevalence of 93.9% B. quintana in body 1. Barker SC, 1994. Phylogeny and classification, origins, and evolu- lice in Burundi (a refugee camp in 2001). Elsewhere, we com- tion of host associations of lice. Int J Parasitol 24: 1285–1291. pared the infection rate in B. quintana body lice in people in 2. Light JE, Toups MA, Reed DL, 2008. What’s in a name: the Africa with the infection rate in B. quintana body lice in home- taxonomic status of human head and body lice. Mol Phylogenet less people in Marseille, France,30,32,33 and the result was statis- Evol 47: 1203–1216. 3. Badiaga S, Brouqui P, 2012. Human louse-transmitted infectious tically significant (228/424 versus 77/560,P< 0.001) (Table 2). diseases. Clin Microbiol Infect 18: 332–337. In this study, we show a relationship between B. quintana pres- 4. Nuttall GHF, 1917. The biology of Pediculus humanus. Parasitol- ence in body lice and socioeconomic level, and we found that ogy 10: 80–185. the body lice that were B. quintana-positive were more likely 5. Busvine JR, 1946. On the pigmentation of the body louse to be found in poor countries (GDP< $1,844 per capita) than pediculus humanus L. In Proceedings of the Royal Entomo- logical Society of London. Series A, General Entomology wealthy countries (GDP> $1,844 per capita; 228/403 versus (Vol. 21, No. 10–12, pp. 98–103). Blackwell Publishing Ltd. 0/21,P< 0.001). 6. Yong Z, Fournier PE, Rydkina E, Raoult D, 2003. The geograph- Head lice can also be infected with B. quintana. In our ical segregation of human lice preceded that of Pediculus study, we detected the DNA of B. quintana in 1.6% (10/616) humanus capitis and Pediculus humanus humanus. C R Biol 326: 565–574. of head lice collected: in Senegal, 0.52% (2/381); in Madagascar 7. Leo NP, Barker SC, 2005. Unravelling the evolution of the head for the first time, 2.66% (2/75); in the Congo, 17.1% (6/35). lice and body lice of humans. Parasitol Res 98: 44–47. B. quintana was detected in 7% (19/274) of head lice collected 8. Raoult D, Reed DL, Dittmar K, Kirchman JJ, Rolain JM, in Dakar, Senegal11 and 9.2% (6/65) of head lice pools and Guillen S, Light JE, 2008. Molecular identification of lice from 7% (19/271) of head lice on persons living in the poorest pre-Columbian mummies. J Infect Dis 197: 535–543. 10,29 9. Sasaki T, Poudel SK, Isawa H, Hayashi T, Seki N, Tomita T, areas of Jimma, Ethiopia. No B. quintana DNA was Sawabe K, Kobayashi M, 2006. First molecular evidence detected in head lice from schoolchildren in Marseille, of Bartonella quintana in Pediculus humanus capitis France,30,31 contrary to the work by Angelakis and others,10 (Phthiraptera: Pediculidae), collected from Nepalese children. which have reported this result in the homeless population. J Med Entomol 43: 110–112. 10. Angelakis E, Diatta G, Abdissa A, Trape JF, Mediannikov O, In Russia, Portugal, Algeria, Burundi, China, Thailand, and 30 Richet H, Raoult D, 2011. Altitude-dependent Bartonella Australia, no B. quintana was found in schoolchildren quintana genotype C in head lice, Ethiopia. Emerg Infect Dis (Table 3). In this study, we also found the head lice that were 17: 2357–2359.

89 BARTONELLA QUINTANA IN AFRICAN LICE 301

11. Boutellis A, Veracx A, Angelakis E, Diatta G, Mediannikov O, epidemics and lice, Democratic Republic of the Congo. Emerg Trape JF, Raoult D, 2012. Bartonella quintana in head lice Infect Dis 19: 505–506. from Senegal. Vector Borne Zoonotic Dis 12: 564–567. 24. Veracx A, Boutellis A, Merhej V, Diatta G, Raoult D, 2012. 12. Brouqui P, 2011. Arthropod-borne diseases associated with polit- Evidence for an African cluster of human head and body lice ical and social disorder. Annu Rev Entomol 56: 357–374. with variable colors and interbreeding of lice between conti- 13. Cutler SJ, 2010. Relapsing fever–a forgotten disease revealed. nents. PLoS ONE 7: e37804. J Appl Microbiol 108: 1115–1122. 25. Ewing HE, 1926. A revision of the American lice of the genus 14. Raoult D, Ndihokubwayo JB, Tissot-Dupont H, Roux V, Pediculus, together with aconsideration of the significance of Faugere B, Abegbinni R, Birtles RJ, 1998. Outbreak of epi- their geographical and host distribution. Proc US Natl Museum demic typhus associated with trench fever in Burundi. Lancet 68:1–30. 352: 353–358. 26. Boutellis A, Drali R, Rivera MA, Mumcuoglu KY, Raoult D, 15. Raoult D, Roux V, 1999. The body louse as a vector of 2013. Evidence of sympatry of clade A and clade B head lice reemerging human diseases. Clin Infect Dis 29: 888–911. in a pre-Columbian Chilean mummy from Camarones. PLoS 16. Bonilla DL, Kabeya H, Henn J, Kramer VL, Kosoy MY, 2009. ONE 8: e76818. Bartonella quintana in body lice and head lice from homeless 27. Foucault C, Brouqui P, Raoult D, 2006. Bartonella quintana persons, San Francisco, California, USA. Emerg Infect Dis 15: characteristics and clinical management. Emerg Infect Dis 12: 912–915. 217–223. 17. Angelakis E, Rolain JM, Raoult D, Brouqui P, 2011. Bartonella 28. Bacot A, 1921. On the probable identity of Rickettsia pediculi quintana in head louse nits. FEMS Immunol Med Microbiol 62: with Rickettsia quintana. BMJ 1: 156–157. 244–246. 29. Cutler S, Abdissa A, Adamu H, Tolosa T, Gashaw A, 2012. 18. Boutellis A, Mediannikov O, Bilcha KD, Ali J, Campelo D, Bartonella quintana in Ethiopian lice. Comp Immunol Microbiol Barker SC, Raoult D, 2013. Borrelia recurrentis in head lice, Infect Dis 35: 17–21. Ethiopia. Emerg Infect Dis 19: 796–798. 30. Fournier PE, Ndihokubwayo JB, Guidran J, Kelly PJ, Raoult D, 19. CIA Factbook, 2012. The World Factbook. Central Intelligence 2002. Human pathogens in body and head lice. Emerg Infect Agency. Available at: https://www.cia.gov/library/publications/ Dis 8: 1515–1518. the-world-factbook/index.html. 31. Bouvresse S, Socolovshi C, Berdjane Z, Durand R, Izri A, Raoult 20. Drali R, Boutellis A, Raoult D, Rolain JM, Brouqui P, 2013. D, Chosidow O, Brouqui P, 2011. No evidence of Bartonella Distinguishing body lice from head lice by multiplex real- quintana but detection of Acinetobacter baumannii in head lice time PCR analysis of the Phum_PHUM540560 gene. PLoS from elementary schoolchildren in Paris. Comp Immunol ONE 8: e58088. Microbiol Infect Dis 34: 475–477. 21. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S, 32. Roux V, Raoult D, 1999. Body lice as tools for diagnosis 2011. MEGA5: molecular evolutionary genetics analysis using and surveillance of reemerging diseases. J Clin Microbiol 37: maximum likelihood, evolutionary distance, and maximum par- 596–599. simony methods. Mol Biol Evol 28: 2731–2739. 33. La SB, Fournier PE, Brouqui P, Raoult D, 2001. Detection and 22. Li W, Ortiz G, Fournier PE, Gimenez G, Reed DL, Pittendrigh culture of Bartonella quintana, Serratia marcescens, and B, Raoult D, 2010. Genotyping of human lice suggests multiple Acinetobacter spp. from decontaminated human body lice.J emergencies of body lice from local head louse populations. Clin Microbiol 39: 1707–1709. PLoS Negl Trop Dis 4: e641. 34. Brouqui P, La Scola B, Roux V, Raoult D, 1999. Chronic 23. Piarroux R, Abedi AA, Shako JC, Kebela B, Karhemere S, Bartonella quintana bacteremia in homeless patients. N Engl Diatta G, Davoust B, Raoult D, Drancourt M, 2013. Plague J Med 340: 184–189.

Chapitre 2 : Distribution phylogéographique des poux humains contemporains et anciens

Préambule

Les poux comptent parmi les plus anciens parasites obligatoires de l'Homme, ce qui fait d’eux un excellent marqueur de l'évolution et de la migration des espèces du genre Homo au fil du temps [23]. Les nombreuses études basées sur les analyses de phylogénie des gènes mitochondriaux ont montré que les poux humains étaient (i) distribués

à l’intérieur de trois clades, les Clades A, B et C et (ii) une relation phylogéographique existait entre ces trois clades [47,48]. Ainsi, si le

Clade A composé de poux de tête et de poux de corps avait une distribution géographique mondiale, les deux autres clades avaient une distribution géographique plus restreinte. le Clade B d’origine américaine [26] est retrouvé maintenant en Europe, en Australie et en

Algérie alors que le Clade C est distribué en Afrique et au Népal [49].

Dans ce chapitre nous présentons les résultats que nous avons obtenus grâce à l’analyse de poux humains rares et précieux, ce qui nous a valu quelques surprises.

Dans le premier papier, nous rapportons pour la première fois l’existence d’un quatrième clade mitochondrial (Clade D) en

République Démocratique du Congo. Comme le Clade A, le Clade D

93 renferme des poux de tête et des poux de corps chez qui on a détecté de l’ADN de B. quintana et de Y. pestis.

Dans le deuxième papier, l'analyse d’anciennes lentes de poux de tête provenant d'Israël datant de 2 périodes différentes, chalcolithique

(4.000 avant JC) et période islamique (650-810 AD), a permis d'affirmer qu'elles appartenaient probablement à des personnes originaires de l'Afrique de l'Ouest en raison de leur appartenance à un sous clade mitochondrial spécifique à cette région.

Dans le troisième papier, nous avons confirmé l’origine américaine du

Clade B à travers l’analyse des lentes de poux de momies précolombienne datant de 4.000 ans [25].

Article IV: A New Clade of African Body and Head Lice Infected by Bartonella quintana and Yersinia pestis – Democratic Republic of the Congo

Accepté avec révisions mineures dans American Journal of Tropical Medicine and Hygiene

American Journal of Tropical Medicine & Hygiene

� � �

� � � �� For Peer Review ��

��

�

97 Page 1 of 9 American Journal of Tropical Medicine & Hygiene

1 2 3 1 �� 4 5 2 �� 6 7 3 �� 8 9 4 �� 10 11 12 5 �� 13 14 6 �� 15 16 7 �� 17 18 8 �� 19 For Peer Review 20 21 9 �� 22 23 10 �� 24 25 11 �� 26 27 12 �� 28 29 30 13 �� 31 32 14 �� 33 34 15 �� 35 36 16 �� 37 38 17 � 39 40 41 18 �� 42 43 19 �� 44 45 20 �� 46 47 21 � 48 49 50 22 �� 51 52 23 � 53 54 24 � 55 56 57 58 59 60 1

98 American Journal of Tropical Medicine & Hygiene Page 2 of 9

1 2 3 25 �� 4 5 26 �� 6 7 27 �� 8 9 28 �� 10 11 12 29 �� 13 14 30 �� 15 16 31 �� 17 18 � 32 �� 19 For Peer Review 20 21 33 �� 22 �� 23 34 �� 24 25 35 �� 26 27 36 �� 28 29 � 30 37 �� 31 32 38 �� 33 34 39 �� 35 36 40 �� 37 38 �� 41 �� 39 40 �� 41 42 �� 42 43 43 �� 44 45 44 �� 46 47 45 �� 48 49 50 46 �� 51 52 47 �� 53 54 48 �� 55 56 57 58 59 60 2

99 Page 3 of 9 American Journal of Tropical Medicine & Hygiene

1 2 3 49 �� 4 5 50 �� 6 7 51 �� 8 9 52 �� 10 11 12 53 �� 13 14 54 �� 15 16 55 �� 17 18 56 �� 19 For Peer Review 20 21 57 �� 22 � 23 58 �� 24 25 59 �� 26 27 60 �� 28 29 �� 30 61 �� 31 �� 32 62 �� 33 34 63 �� 35 36 64 �� 37 38 65 �� 39 40 � 41 66 �� 42 43 67 �� 44 45 68 �� 46 47 69 �� 48 49 50 70 �� 51 52 71 �� 53 54 72 �� 55 56 73 �� 57 58 59 60 3

100 American Journal of Tropical Medicine & Hygiene Page 4 of 9

1 2 3 74 �� 4 5 75 �� 6 7 76 �� 8 9 77 �� 10 11 12 78 �� 13 14 79 �� 15 16 80 �� 17 18 81 �� 19 For Peer Review 20 21 82 �� 22 23 83 �� 24 25 84 �� 26 27 85 �� 28 29 30 86 �� 31 32 87 �� 33 34 88 �� 35 36 89 �� 37 38 90 �� 39 40 41 91 �� 42 43 92 �� 44 45 93 �� 46 47 94 �� 48 49 50 95 �� 51 52 96 �� 53 54 97 �� 55 56 98 �� 57 58 59 60 4

101 Page 5 of 9 American Journal of Tropical Medicine & Hygiene

1 2 3 99 �� 4 5 100 �� 6 7 101 �� 8 9 102 �� 10 11 12 103 �� 13 14 104 �� 15 16 105 �� 17 18 106 �� 19 For Peer Review 20 21 107 �� 22 23 108 �� 24 25 109 �� 26 27 110 �� 28 29 30 111 �� 31 32 112 �� 33 34 113 �� 35 36 114 �� 37 38 115 �� 39 40 41 116 �� 42 43 117 �� 44 45 118 �� 46 47 119 �� 48 49 50 120 �� 51 52 121 � 53 54 122 � 55 56 123 � 57 58 59 60 5

102 American Journal of Tropical Medicine & Hygiene Page 6 of 9

1 2 3 124 �� 4 5 125 �� 6 7 126 �� 8 9 127 �� 10 11 12 128 �� 13 14 129 �� 15 16 130 �� 17 18 131 �� 19 For Peer Review 20 21 132 �� 22 23 133 �� 24 25 134 �� 26 27 135 �� 28 29 30 136 � 31 32 137 � 33 34 138 � 35 36 139 � 37 38 140 � 39 40 41 141 � 42 43 142 � 44 45 143 � 46 47 144 � 48 49 50 145 � 51 52 146 � 53 54 147 � 55 56 148 � 57 58 59 60 6

103 Page 7 of 9 American Journal of Tropical Medicine & Hygiene

1 2 3 149 �� 4 5 150 �� 6 7 151 �� 8 9 152 �� 10 11 12 153 �� 13 14 154 �� 15 16 155 �� 17 18 156 �� 19 For Peer Review 20 21 157 �� 22 23 158 �� 24 25 159 �� 26 27 160 �� 28 29 30 161 �� 31 32 162 �� 33 34 163 �� 35 36 164 �� 37 38 165 �� 39 40 41 166 �� 42 43 167 �� 44 45 168 �� 46 47 169 �� 48 49 50 170 �� 51 52 171 �� 53 54 172 �� 55 56 173 �� 57 58 59 60 7

104 American Journal of Tropical Medicine & Hygiene Page 8 of 9

1 2 3 174 �� 4 5 175 �� 6 7 176 �� 8 9 177 �� 10 11 12 178 �� 13 14 179 �� 15 16 180 �� 17 18 181 �� 19 For Peer Review 20 21 182 �� 22 23 183 �� 24 25 184 �� 26 27 185 �� 28 29 30 186 �� 31 32 187 �� 33 34 188 �� 35 36 189 � 37 38 190 � 39 40 41 191 � 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 8

105 (B) Borrelia recurrentis Rickettsia prowazekii pestisYersinia Bartonella quintana Body louse Body louse louse Head louse Head louse Head louse Head 106 American Journal of Tropical Medicine & Hygiene

For Peer Review (A) Figure: Maximum-likelihood (ML) phylogram of the mitochondrial cytochrome ) Maximum-likelihood ( cytb Figure: b the mitochondrial (ML) phylogram of cytochrome gene. Page 9 of 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Article V: Studies of Ancient Lice Reveal Unsuspected Past Migrations of Vectors

Accepté avec révisions mineures dans American Journal of Tropical Medicine and Hygiene

107

American Journal of Tropical Medicine & Hygiene

��

��For� �� Peer Review � ��

��

109 Page 1 of 9 American Journal of Tropical Medicine & Hygiene

1 2 3 1 LRH: DRALI AND OTHERS 4 5 2 RRH: PAST MIGRATIONS OF VECTORS 6 7 3 Studies of ancient lice reveal unsuspected past migrations of vectors 8 9 4 10 11 12 5 Rezak Drali, Kosta Y. Mumcuoglu, Gonca Yesilyurt, Didier Raoult 13 14 6 �� 15 16 7 �� 17 18 8 �� 19 For Peer Review 20 21 9 �� 22 23 10 �� 24 25 11 �� 26 27 12 �� 28 29 30 13 Abstract��The analysis of ancient head louse eggs recovered from Israel dating from 31 32 14 Chalcolithic period (four millennium B.C.) and early Islamic period (650 – 810 A.D.) allowed 33 34 15 to affirm that they probably belonged to people originating from West Africa because of their 35 36 16 belonging to the mitochondrial sub clade specific to that region. 37 38 17 39 40 41 18 42 43 19 44 45 20 Address correspondence to: Didier Raoult, Unité de Recherche sur les Maladies Infectieuses 46 47 21 et Tropicales Emergentes (URMITE), Faculté de Médecine, 27 Blvd Jean Moulin, 13385 48 49 50 22 Marseille cedex 5, France. Email: [email protected] 51 52 23 53 54 24 � 55 56 25 ��, ancient lice, paleo-entomology, past migration of vectors� 57 58 59 60 1

110 American Journal of Tropical Medicine & Hygiene Page 2 of 9

1 2 3 26 Head lice (��) and body lice (��) 4 5 27 are strict bloodsucking ectoparasites of humans, and each species lives in a specific ecological 6 7 28 niche: hair for head lice, clothing for body lice.1 Hundreds millions of children worldwide are 8 9 29 continually infested by head lice unrelated to hygienic conditions, resulting in insomnia and 10 11 2 12 30 itching. Body lice exclusively infest populations exposed to stressful life conditions, such as 13 14 31 the homeless, prisoners and war refugees, and can serve as a vector for three serious humans 15 16 32 diseases, namely, epidemic typhus, trench fever and relapsing fever caused by �� 17 18 1 33 ��, �� and ��, respectively. Body lice are also 19 For Peer Review 20 3 21 34 suspected to be able to host and transmit ��, the agent of plague. Genetically, 22 23 35 human lice are distributed into three mitochondrial clades (A, B and C), whereby only Clade 24 25 36 A comprises head and body lice.4 Clade A is found on all continents, whereas Clade B, with 26 27 37 an assignment of an American origin, is present in Europe and Australia and was recently 28 29 5 30 38 characterized in North Africa. Lastly, Clade C is to date limited to Africa and Asia. 31 32 39 Lice are among the oldest parasites of humans, thus representing an excellent marker 33 34 40 of the evolution and migration of the Homo species over time.4 For centuries, archaeologists 35 36 41 excavating soil in different locations around the world have found nits, lice and/or combs on 37 38 6 42 mummies or human remains with ages varying between 300 and 10,000 years (see for a 39 40 41 43 review). The most ancient specimens (nits from hair, 10,000 years old) were found in Brazil, 42 7 43 44 South America. However, few molecular data on ancient lice are available. In 2008, Raoult 44 45 45 et al. showed that the most prevalent and well-distributed clade of lice (A) had a pre- 46 47 46 Columbian presence on the American continent.8 In 2013, Boutellis et al� confirmed this 48 49 50 47 result and demonstrated that Clade B also had a presence of at least 4,000 years in America 51 9 52 48 and that Clades A and B could live in sympatry. 53 54 55 56 57 58 59 60 2

111 Page 3 of 9 American Journal of Tropical Medicine & Hygiene

1 2 3 49 In the present work, we obtained and analyzed ancient head louse eggs recovered from 4 5 50 Israel dating from two different periods, namely, the Chalcolithic (four millennium B.C.) and 6 7 51 early Islamic periods (650 – 810 A.D.). 8 9 52 In 2014, two lots of ancient head louse nits recovered from human remains in Israel 10 11 12 53 were sent to our laboratory for molecular analysis. A total of seven samples were examined. 13 14 54 Five operculated nits were found on hair remains in the Cave of the Treasure located at Nahal 15 16 55 Mishmar, in the Judean Desert, which date to the Chalcolithic period.10 The two other nit 17 18 56 samples were found on human hair remains belonging to a population who lived in Nahal 19 For Peer Review 20 21 57 Omer in the Arava Valley (between Dead Sea and Red Sea) during the Early Islamic Period 22 23 58 (650 – 810 A.D.). The site appears to have been a way-station on the north-south Arava route 24 25 59 and on the Spice Route between Petra and Gaza.11,12 The operculum on the eggs was most 26 27 60 likely detached during the centuries, but the embryos were still present inside the eggs. 28 29 30 61 The nits were photographed using an Axio Zoom V16 (Carl Zeiss AG, Germany). 31 9 32 62 DNA extraction was performed as previously described. To determine the clade of the 33 34 63 specimens, an 89-bp fragment of the mitochondrial cytochrome b gene (��) was targeted. A 35 36 64 sensitive tool based on the real-time PCR with a hydrolysis probe that was useful to detect 37 38 65 and amplify ancient DNA was developed. A pair of primers (the reverse primer was 39 40 41 66 degenerate) that can amplify the different known clades of lice was used: cytbF (5’- 42 43 67 AGTGTGAGGAGGGTTTTCAG-3’) and cytbR (5’-CAAACCCCAAYAAVAYAAACGG- 44 45 68 3’). A TaqMan© FAM-labeled probe that contained non-fluorescent quencher conjugated to a 46 47 69 minor groove binder (MGB) at the 3’ was designed: FAM- 48 49 50 70 CTACTTTAGAGCGGTTGTTTACTC-MGB (Applied Biosystems, Courtaboeuf, France). 51 52 71 Real-time PCR was performed using the CFX96 thermal cycler (Bio-Rad 53 54 72 Laboratories, Foster City, CA, USA). The final reaction volume of 20 µL contained 3 µL of 55 56 73 the DNA template, 10 µL of 2x QuantiTect™ Probe PCR Master Mix (Qiagen), 0.5 µM of 57 58 59 60 3

112 American Journal of Tropical Medicine & Hygiene Page 4 of 9

1 2 3 74 each primer and 0.2 µM of the FAM-labeled probe. The thermocycling parameters consisted 4 5 75 of 95°C for 15 min and 40 cycles of 95°C for 10 s and 60°C for 30 s. No-template controls 6 7 76 (NTCs) were included in RT-PCR assay. The products of the real-time PCR amplifications 8 9 77 were sequenced as previously described.9 The nucleotide sequences obtained were aligned, 10 11 9 12 78 and phylogenetic analyses were performed as previously described. 13 14 79 The appearance of nits under the microscope shows that they were well preserved 15 16 80 despite the many centuries spent underground. As shown in Figure 1, we can distinguish the 17 18 81 eyes, legs and claws of embryos still inside the operculated nits. 19 For Peer Review 20 21 82 RT-PCR was positive for the seven samples of ancient DNA tested (32 < Ct < 34). The Ct 22 23 83 value was 20 in the positive control; no fluorescence was detected in the negative control. 24 25 84 A Maximum-Likelihood phylogenetic analysis performed for the cytb gene showed that the 26 27 85 ancient hair nits belonging to the Chalcolithic and early Islamic periods are part of Clade C. 28 29 30 86 Within Clade C, these nits form a sub-clade with lice found in Senegal (Figure 2). 31 32 87 Nonetheless, the ancient DNA sequences are quite unique and are different compared to 33 34 88 known sequences of contemporary lice. 35 36 89 In this work, the first molecular data on lice infesting the ancient inhabitants of the 37 38 90 Near East were compiled. The tool implemented here based on the real-time PCR was 39 40 41 91 effective in preventing the difficulties associated with the analysis of ancient DNA, which is 42 13 43 92 often damaged or in a very low concentration. 44 45 93 Interestingly, this study reveals the presence of a mitochondrial genotype, Clade C, in 46 47 94 a Near East region that has thus far only been found in Africa (Ethiopia and Senegal) and Asia 48 49 5 50 95 (Nepal). Specifically, these old nits are included in a subgroup in Clade C comprising 51 52 96 contemporary lice that are characterized only in West Africa. The presence of this genotype in 53 54 97 the Near East is necessarily linked to migration flows of humans through the ages. The slave 55 56 98 trade practiced by Arab tribes to supply the Near and Middle East in manpower existed for a 57 58 59 60 4

113 Page 5 of 9 American Journal of Tropical Medicine & Hygiene

1 2 14 3 99 long time. The genetic information available today suggests that significant gene flow most 4 5 100 likely occurred within the past ∼2,500 years between sub-Saharan African and Near East 6 7 101 populations.15 8 9 10 102 Thus, the study of ancient lice provides a wealth of interesting information for 11 12 103 understanding and elucidating the migration of our species throughout history and also 13 14 104 information about the circulation of pathogens transmitted by lice. Indeed, lice have proven to 15 16 105 be an excellent recorder, allowing explanations of what occurred when discovering certain 17 18 106 mass graves. In 2006, Raoult et al. concluded that louse-borne infectious diseases that 19 For Peer Review 20 21 107 affected nearly one-third of Napoleon’s soldiers buried in Vilnius might have been a major 22 13 23 108 factor of mortality in the French retreat from Russia. 24 25 109 In conclusion, we assert that the study of ancient lice is very useful to understand the 26 27 110 circulation of vectors, the flow of vector-borne pathogens and the migratory flu hosts of these 28 29 30 111 vectors. 31 32 112 �� 33 34 113 We thank Jean-Michel BERENGER (Entomologist) and Abdul Karim SANGARÉ for their 35 36 114 technical support. 37 38 115 �� 39 40 41 116 The authors declare they have no conflict of interest. 42 43 117 ��: Rezak Drali and Didier Raoult, Unité de Recherche sur les Maladies 44 45 118 Infectieuses et Tropicales Emergentes (URMITE), Centre National de la Recherche 46 47 119 Scientifique No. 7278 (CNRS7278), Institut de Recherche pour le Développement No. 198 48 49 50 120 (IRD198), Institut National de la Santé et de la Recherche Médicale Unité No. 1095 51 52 121 (InsermU1095), Institut Hospitalo-Universitaire Méditerranée-Infection, Aix-Marseille 53 54 122 Université, Marseille, France, Emails: [email protected], [email protected]. 55 56 123 Kosta Y. Mumcuoglu and Gonca Yesilyurt, Department of Microbiology and Molecular 57 58 59 60 5

114 American Journal of Tropical Medicine & Hygiene Page 6 of 9

1 2 3 124 Genetics, The Kuvin Center for the Study of Infectious and Tropical Diseases, Hadassah 4 5 125 Medical School, The Hebrew University, Jerusalem, Israel, Emails: [email protected], 6 7 126 [email protected]. 8 9 127 10 11 12 128 References 13 14 129 15 16 1. Raoult D, Roux V, 1999. The body louse as a vector of reemerging human diseases. �� 17 130 18 131 �� : 888-911. 19 For Peer Review 20 21 132 2. Chosidow O, Chastang C, Brue C, Bouvet E, Izri M, Monteny N, Bastuji-Garin S, 22 133 Rousset JJ, Revuz J, 1994. Controlled study of malathion and d-phenothrin lotions 23 24 134 for �� var ��-infested schoolchildren. ��: 1724-1727. 25 26 135 3. Houhamdi L, Lepidi H, Drancourt M, Raoult D, 2006. Experimental model to evaluate 27 28 136 the human body louse as a vector of plague. �� 1589-1596. 29 30 31 137 4. Reed DL, Smith VS, Hammond SL, Rogers AR, Clayton DH, 2004. Genetic analysis of 32 138 lice supports direct contact between modern and archaic humans. �� e340. 33 34 35 139 5. Boutellis A, Abi-Rached L, Raoult D, 2014. The origin and distribution of human lice in 36 140 the world. �� 209-217. 37 38 39 141 6. Mumcuoglu KY, 2008. Human lice: Pediculus and Pthirus. In: Raoult, Drancourt, 40 41 142 editors. Paleomicrobiology: Past Human Infections. Berlin: Springer-Verlag. pp. 42 143 215-222. 43 44 45 144 7. Araujo A, Ferreira LF, Guidon N, Maues Da Serra FN, Reinhard KJ, Dittmar K, 2000. 46 145 Ten thousand years of head lice infection. �� 269. 47 48 49 146 8. Raoult D, Reed DL, Dittmar K, Kirchman JJ, Rolain JM, Guillen S, Light JE, 2008. 50 147 Molecular identification of lice from pre-Columbian mummies. �� 51 52 148 535-543. 53 54 55 149 9. Boutellis A, Drali R, Rivera MA, Mumcuoglu KY, Raoult D, 2013. Evidence of 56 150 sympatry of clade A and clade B head lice in a pre-Columbian Chilean mummy from 57 58 151 Camarones. �� e76818. 59 60 6

115 Page 7 of 9 American Journal of Tropical Medicine & Hygiene

1 2 3 152 10. Bar-Adon P, 1980. The cave of the treasure: The finds from the caves in Nahal 4 Mishmar. ��. 5 153 6 7 154 11. Baginski A, Shamir O, 1995. Early Islamic textiles, basketry and cordage from Nahal 8 9 155 Omer, Israel. �� 21-42. 10 11 156 12. Negev A, 1966. The Date of the Petra-Gaza Road. �� 12 13 157 89-98. 14 15 158 13. Raoult D, Dutour O, Houhamdi L, Jankauskas R, Fournier PE, Ardagna Y, Drancourt 16 17 159 M, Signoli M, La VD, Macia Y, Aboudharam G, 2006. Evidence for louse- 18 19 160 transmittedFor diseases inPeer soldiers of Napoleon's Review Grand Army in Vilnius. �� 20 161 �� 112-120. 21 22 23 162 14. Lewis B, 1992. Race and slavry in the Middle East: an historical enquiry. Oxford 24 163 University Press. 25 26 27 164 15. Richards M, Rengo C, Cruciani F, Gratrix F, Wilson JF, Scozzari R, Macaulay V, 28 29 165 Torroni A, 2003. Extensive female-mediated gene flow from sub-Saharan Africa 30 166 into near eastern Arab populations. �� 1058-1064. 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 7

116 Page 8 of 9 117 American Journal of Tropical Medicine & Hygiene

For Peer Review Figure Operculated 1: Figure head louse recovered egg from Israel dating from Chalcolithic period (four millennium B.C.) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 Clade C Chalcolithic-30 Chalcolithic-5 Clade-C39_Senegal P.schaeffi Clade A Clade-C42_Ethiopia Chalcolithic-36 Chalcolithic-29 Islamic-11 Clade B Clade-C44_Nepal Chalcolithic-27 Islamic-9

50 Clade-A12

92 Clade-A5

Clade-B36 55 118 Clade-A17

Clade-B33 56

62 American Journal of Tropical Medicine & Hygiene 0.05 For Peer Review Maximum-likelihood (ML) phylogram of cytochrome b gene. b gene. cytochrome of phylogram (ML) 2. Maximum-likelihood Figure to indicated are memberships clade Mitochondrial nodes. the above located are 50 than greater values support that bootstrap ML tree.each of right the Page 9 of 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Article VI: Evidence of Sympatry of Clade A and Clade B Head Lice in A Pre-Columbian Chilean Mummy from Camarones

PLoS One 8: e76818.

119

Evidence of Sympatry of Clade A and Clade B Head Lice in a Pre-Columbian Chilean Mummy from Camarones

Amina Boutellis1, Rezak Drali1, Mario A. Rivera2, Kosta Y. Mumcuoglu3, Didier Raoult1* 1 Unite´ de Recherche sur les Maladies Infectieuses et Tropicales Emergentes: URMITE, Aix Marseille Universite´, UMR CNRS 7278, IRD 198, INSERM 1095. Faculte´ de Me´decine, 27 Bd Jean Moulin, Marseille, France,2 Programa Identidad del Fin del Mundo. Universidad de Magallanes-Mineduc, Punta Arenas, Chile,3 Department of Microbiology and Molecular Genetics, The Kuvin Center for the Study of Infectious and Tropical Diseases, Hadassah Medical School, The Hebrew University, Jerusalem, Israel

Abstract Three different lineages of head lice are known to parasitize humans. Clade A, which is currently worldwide in distribution, was previously demonstrated to be present in the Americas before the time of Columbus. The two other types of head lice are geographically restricted to America and Australia for clade B and to Africa and Asia for clade C. In this study, we tested two operculated nits from a 4,000-year-old Chilean mummy of Camarones for the presence of the partial Cytb mitochondrial gene (270 bp). Our finding shows that clade B head lice were present in America before the arrival of the European colonists.

Citation: Boutellis A, Drali R, Rivera MA, Mumcuoglu KY, Raoult D (2013) Evidence of Sympatry of Clade A and Clade B Head Lice in a Pre-Columbian Chilean Mummy from Camarones. PLoS ONE 8(10): e76818. doi:10.1371/journal.pone.0076818 Editor: Michael Knapp, Bangor University, United Kingdom Received April 22, 2013; Accepted August 29, 2013; Published October 30, 2013 Copyright:� 2013 Boutellis et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: These authors have no support or funding to report. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected]

Introduction have contributed to populating the Americas. In the current study, two operculated nits from a mummy found in Camarones, Chile, Pediculus humanus capitis is an ancient human parasite most likely were tested to identify the mitochondrial phylotypes of the lice. associated with humans since our pre-hominid ancestor and dispersed throughout the world by early human migrants [1]. Materials and Methods Louse infestation in ancient human populations has been recorded in different geographic regions of the world [2] and even affected Hair samples from seven mummies from Camarones 15-D, wealthy social classes, such as the 15th-century King of Naples, Chile, carbon-dated to ca. 4,000 B.C., were examined for the Ferdinand II of Aragon [3]. The oldest head louse nit was found presence of head lice (Figure 1), and in six hair samples, the nits of on a hair from an archaeological site in northeastern Brazil and head lice were found. No body louse was found on these was dated to 8,000 B.C. [4]. The oldest such finding in Asia is mummies. The material (nits fixed on hair) of one mummy was 9,000 years old, obtained from a hair sample from an individual stored in 70% ethyl alcohol for the last 10 years and was sent to who lived in the Nahal Hemar cave in Israel [5]. Head lice have our laboratory in Marseilles in September 2012. also been found at archaeological sites in the southwestern USA, A first screening of the quality of the nits was performed using the Aleutian Islands, Peru, Greenland and Mexico and on our ZEISS zoom microscope with a fixed camera (ZEISS AXIO mummies that were Incan sacrifices [4]. Recently, another ZOOM.V16. France), and then two operculated nits were selected discovery of lice was reported for a Maitas Chiribaya mummy for our study (Figure 2). The two nits were rinsed twice in sterile from Arica, in northern Chile, dating to 670–990 A.D. (calibrated) water then crushed with a scalpel. The total genomic DNA was [6]. The evidence for the presence of ectoparasites on ancient extracted and eluted in a 50 ml volume with a QIAamp Tissue Kit Americans indicates that head lice most likely arrived with the first (Qiagen, Courtaboeuf, France) with an EZ1 apparatus, as human colonists who entered the Americas [7]. However, by described by the manufacturer. The extracted genomic DNA amplifying mitochondrial DNA (mtDNA) of part of two genes concentrations were 1.8 ng/ml and 1.3 ng/ml for the two nits, and (Cytb and Cox1) belonging to 10,000-year-old lice collected from the DNA samples were stored at220uC under sterile conditions to Peruvian mummies, Raoult and colleagues demonstrated that the avoid cross-contamination until further processing. The DNA of worldwide clade A louse was in the Americas before the time of the two nits was amplified using a suicide nested polymerase chain Columbus [8]. Two other clades of head lice have been reported reaction (PCR) protocol (re-amplification without positive control) and have a specific geographical distribution: clade C is specifically with a partial Cytb gene (270 bp) primers, as previously described restricted to head lice in Ethiopia, Nepal and Senegal [9], whereas [11]. The same primers were used for both amplifications. PCR the clade B head lice are found in North and Central America, reactions were prepared on ice and contained 3ml of the DNA Australia and certain European countries [7]. However, the origin template, 4ml of 5X HF Phusion Buffer, 250mM of each of clade B head lice remains unknown because no lice with this nucleotide, 0.5mM of each primer, 0.2ml of Phusion DNA phylotype have been reported in Asia or in any region suspected to Polymerase (Thermo Scientific, Lithuania) and water (DNase and

PLOS ONE | www.plosone.org 1211 October 2013 | Volume 8 | Issue 10 | e76818 Head Lice Nits in Pre-Columbian Mummy

RNase-Free) to a final reaction mixture volume of 20ml. The PCRs were performed in a PTC-200 automated thermal cycler (MJ research, Waltham, MA, USA). The cycling conditions were 98uC for 30 sec; 40 cycles of 5 sec at 98uC, 30 sec at 56uC, 15 sec at 72uC; and a final extension time of 5 min at 72uC. All of the experiments were performed in a location free of louse DNA, under a hood with air capture and with sterilized instruments that were used only once. The negative controls remained negative. The success of the PCR amplification was then verified by migration of the PCR product on a 2% agarose gel. The NucleoFast 96 PCR Plates (Macherey-Nagel EURL, France) and BigDye Terminator version 1.1 cycle sequencing-ready reaction mix (Applied Biosystems, Foster City, CA) were then used to purify the PCR products to be sequenced directly in both directions with the same primers used in the PCR amplification. The ABI 3100 automated sequencer (Applied Biosystems) resolved the sequenced products. The program Chromas Pro software (Technelysium PTY, Australia) was used to analyze, assemble and correct the sequences. In addition to our newly obtained data, 18 samples belonging to the worldwide clade A [8], 19 samples belonging to clade B from USA, UK [11] and Honduras [7] and 28 samples belonging to clade C from Senegal [9] and Ethiopia [12] were used. The Phylogeny Reconstruction was performed from the DNA sequences using the Maximum Likelihood (ML) with 100 Bootstrap Replications within the MEGA 5 software with complete deletion, Tamura-Nei model (nucleotide) of substitution model was used automatically [13].

Figure 1. Picture of the Mummy. Mummy 23 from Camarones 15-D, Ethics statement Northern Chile� Mario A Rivera. Mummies from Camarones were excavated in 1990 by a team doi:10.1371/journal.pone.0076818.g001 of investigators under the direction of Mario A. Rivera, they have permit from Consejos Monumentos Nacionales, authorization

Figure 2. Picture of the Pre Columbian nit. A pre-Columbian nit isolated from a Chilean mummy (No. 1) (h: hair; o: operculum; c: cementum) (picture taken with a ZEISS AXIO ZOOM.V16). doi:10.1371/journal.pone.0076818.g002

PLOS ONE | www.plosone.org 1222 October 2013 | Volume 8 | Issue 10 | e76818 Head Lice Nits in Pre-Columbian Mummy

Figure 3. Cytb phylogenic analysis. The phylogenic tree based on ML method of the two pre-Columbian Chilean nits based on the partial Cytb gene (270 bp). HL: head louse, BL: body louse. The numbers on the branches are bootstrap values. doi:10.1371/journal.pone.0076818.g003 number 355, November 12th, 1987. Samples were donated by phylogenetic tree, it was found that one nit belonged to the Museo Arqueologia San Miguel Azapa, Universidad de Tarapaca, worldwide clade A (Genbank accession nuKF498963), whereas the Arica, Chile and permission was obtained from the said Museum second nit belonged to the clade B (Genbank accession to access the collections [10]. nuKF498962) (figure 3). One base distinguished chilean mummy clade A head lice from other sequences found in GenBank: Results position 7 (G in mummy nit lice versus a gap in others) (Figure 4). There is no difference between the Chilean mummy clade B nit The DNA of the two nits (848 mm and 912 mm long, and other sequences present in GenBank (Figure 4). respectively) was amplified and sequenced for the Cytb gene of 270 bp. After assembling the sequences and analyzing the

PLOS ONE | www.plosone.org 1233 October 2013 | Volume 8 | Issue 10 | e76818 Head Lice Nits in Pre-Columbian Mummy

Figure 4. Sequences alignment. Alignment of the clade A and clade B sequences (Chilean mummy’s nit and other sequences present in GenBank) P: Philippines, G: Germany, I: Iran, U: United State of America, T: Taiwan, C: Canada, S: Senegal, Pa: Papua New Guinea, H: Honduras. doi:10.1371/journal.pone.0076818.g004

Discussion been reported only in Africa; and A3, which is specific to American lice [14]. Clade B may have developed in North and We report the first identification of both the clade A and the Central America before Columbus and is now spreading clade B genotypes existing in sympatry in two nits isolated from throughout the world. This clade’s origin predates modern Homo human remains from pre-Columbian Chile. Lice with clade A and sapiens by an order of magnitude (ca. 1.18 million years) [1]. In clade B mtDNA are not uncommon and were reported to be contrast, Clade C is mostly confined to Africa and Asia. present on a single human head in both the USA and Honduras The present work confirms that the origin of clade B was on several occasions [8]. Clade B was first identified in America America before the arrival of Columbus, but it will be interesting and the United Kingdom and was later found in other European to test other mummies from Asia, which is reputed to have peopled countries and Australia [7]. Until now, the origin of clade B was the Americas. unknown, but according to the clade’s current distribution, it was speculated that this louse phylotype was imported into Europe by Europeans returning from America [1]. Currently, the most likely Author Contributions theory is that the clade A louse issued from Africa and was Conceived and designed the experiments: DR KM. Performed the distributed worldwide, given the clade’s three different chromo- experiments: AB RD. Analyzed the data: AB RD DR. Wrote the paper: somal signatures: A1, which is found worldwide; A2, which has AB RD KM MR DR. Collected samples: MR KM.

PLOS ONE | www.plosone.org 1244 October 2013 | Volume 8 | Issue 10 | e76818 Head Lice Nits in Pre-Columbian Mummy

References 1. Reed DL, Smith VS, Hammond SL, Rogers AR, Clayton DH (2004) Genetic 8. Raoult D, Reed DL, Dittmar K, Kirchman JJ, Rolain JM, et al. (2008) analysis of lice supports direct contact between modern and archaic humans. Molecular identification of lice from pre-Columbian mummies. J Infect Dis 197: PLoS Biol 2: e340. 10.1371/journal.pbio.0020340 [doi]. 535–543. 10.1086/526520 [doi]. 2. Rick FM, Rocha GC, Dittmar K, Coimbra CE, Reinhard K, et al. (2002) Crab 9. Boutellis A, Veracx A, Angelakis E, Diatta G, Mediannikov O, et al. (2012) louse infestation in pre-Columbian America. J Parasitol 88: 1266–1267. Bartonella quintana in head lice from Senegal. Vector Borne Zoonotic Dis 12: 10.1645/0022–3395 (2002) 088 [1266: CLIIPC]2.0.CO;2 [doi]. 564–567. 10.1089/vbz.2011.0845 [doi]. 3. Fornaciari G, Giuffra V, Marinozzi S, Picchi MS, Masetti M (2009) ‘Royal’ 10. Rivera MA, Mumcuoglu K, Mathney RT, Matheny DG (2008) Head lice eggs, pediculosis in Renaissance Italy: lice in the mummy of the King of Naples Anthropophthirus capitis, from mummies of the Chinchorro tradition, Camarones Ferdinand II of Aragon (1467–1496). Mem Inst Oswaldo Cruz 104: 671–672. 15-D, Northern Chile. Chungara, Revista de Antropologia Chilena 40: 31–39. S0074–02762009000400026 [pii]. 11. Li W, Ortiz G, Fournier PE, Gimenez G, Reed DL, et al. (2010) Genotyping of 4. Araujo A, Ferreira LF, Guidon N, Maues Da Serra FN, Reinhard KJ, et al. human lice suggests multiple emergencies of body lice from local head louse (2000) Ten thousand years of head lice infection. Parasitol Today 16: 269. populations. PLoS Negl Trop Dis 4: e641. 10.1371/journal.pntd.0000641 [doi]. S0169–4758 (00) 01694-X [pii]. 12. Angelakis E, Diatta G, Abdissa A, Trape JF, Mediannikov O, et al. (2011). 5. Mumcuoglu K (2008) Pediculus and Pthirus. In Paleomicrobiology – Past Altitude-dependent Bartonella quintana genotype C in head lice, Ethiopia. Emerg Human Infections. Raoult, D. & M. Drancourt (eds). Springer, Berlin, 215–222. Infect Dis; 17: 2357–2359. 10.3201/eid1712.110453 [doi]. 6. Arriaza B, Orellana NC, Barbosa HS, Menna-Barreto RF, Araujo A, et al. 13. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, et al. (2011) MEGA5. (2012) Severe head lice infestation in an Andean mummy of Arica, Chile. Molecular evolutionary genetics analysis using maximum likelihood, evolution- J Parasitol 98: 433–436. 10.1645/GE-2903.1 [doi]. ary distance, and maximum parsimony methods. Mol Biol Evol; 28: 2731–39. 7. Light JE, Allen JM, Long LM, Carter TE, Barrow L, et al. (2008) Geographic msr121 [pii];10.1093/molbev/msr121 [doi]. distributions and origins of human head lice (Pediculus humanus capitis) based 14. Boutellis A, Veracx A, Jonatas A, Raoult D (2013) Amazonian head lice-specific on mitochondrial data. J Parasitol 94: 1275–1281. GE-1618 [pii];10.1645/GE- genotypes are putatively pre-Columbian. Am J Trop Med Hyg 88(6):1180–4. 1618.1 [doi]. 10.4269/ajtmh.12–0766 [doi].

PLOS ONE | www.plosone.org 1255 October 2013 | Volume 8 | Issue 10 | e76818

Chapitre 3: Host-Switching

127

Préambule

L’association étroite hôte-parasite peut conduire à une cospeciation.

Ce phénomène est bien illustré par la relation poux-hôtes. En effet, les poux sont des ectoparasites obligatoires des oiseaux et des mammifères avec qui le partenariat a commencé il y’a plus de 65 millions d’années [6,7,50].

Parfois, cette association est marquée par une série d'événements historiques tels le changement d'hôte “host-switching” ou encore la duplication de parasite [10,11].

Depuis un siècle, la relation pou – hôte entre Pediculus mjobergi et les primates du Nouveau Monde est restée une énigme du fait de la ressemblance morphologique frappante de ce dernier avec les poux humains.

Pour investiguer ces observations centenaires, nous avons recueilli des poux de singes hurleurs d’Argentine et nous les avons comparés à des poux humains que nous avons recueillis dans un village d’Amazonie isolé qui a échappé à la mondialisation.

Cette étude a permis de lever l'ambiguïté existante grâce aux premières analyses moléculaires et génétiques effectuées sur P. mjobergi.

129 En effet, les résultats montrent que P. mjobergi a été transmis aux singes par les premiers humains arrivés sur le continent américain il y’a des milliers d'années. Ce résultat remet en cause le paradigme de coévolution stricte entre le pou et son hôte.

130

Article VII: Host switching of Human Lice to New World Monkeys in Amazonia

Soumis comme Research Reports dans Proceedings of the National Academy of Sciences

131

1 Host switching of human lice to new world monkeys in Amazonia

3 Rezak Dralia, Laurent Abi-Rachedb, Amina Boutellisa, Félix Djossouc, Stephen C. Barkerd and

4 Didier Raoulta

6 aAix Marseille Université, URMITE, UM63, CNRS 7278, IRD 198, Inserm 1095, 13005

7 Marseille, France.

8 bCentre National de la Recherche Scientifique, Institut de Mathématiques de Marseille - Unité

9 Mixte de Recherche 7373, Equipe ATIP, Aix-Marseille Université, 13331 Marseille, France.

10 cService de maladies infectieuses et tropicales, Centre hospitalier de Cayenne, 97306 Cayenne

11 Cedex. Guyane française.

12 dDepartment of Parasitology, School of Chemistry and Molecular Biosciences, The University

13 of Queensland, Brisbane, Queensland, Australia.

15 Corresponding author: Didier Raoult

16 mailto:[email protected]

18 PNAS Research Reports

24 Key words: Pediculus mjobergi, Clade B, Amazonian head louse, host-switching,

25 cospeciation

133 26 Abstract (204/250)

27 The coevolution between a host and its obligate parasite is exemplified in the sucking lice that

28 infest primates (1-4). In the context of close lice-host partnerships and cospeciation, Pediculus

29 mjobergi, the louse of New World primates, has long been puzzling because its morphology

30 resembles that of human lice (5,6). To investigate the possibility that P. mjobergi was

31 transmitted to monkeys from the first humans who set foot on the American continent

32 thousands of years ago, we obtained and compared P. mjobergi lice collected from howler

33 monkeys from Argentina to human lice gathered from a remote and isolated village in

34 Amazonia that has escaped globalization (7). Morphological examinations were first

35 conducted and verified the similarity between the monkey and human lice. The molecular

36 characterization of several nuclear and mitochondrial genetic markers in the two types of lice

37 revealed that one of the P. mjobergi specimens had a unique haplotype that clustered with the

38 haplotypes of Amazonian head lice that are prevalent in tropical regions in the Americas, a

39 natural habitat of New World monkeys. Because this phylogenetic group forms a separate

40 branch within a human clade of sequences that is of American origin, this finding indicates

41 that human lice have transferred to New World monkeys.

134 51 Significance Statement (116/120)

52 Lice are permanent obligate parasitic insects that infest birds and mammals. Because they are

53 wingless and have no intermediate hosts, lice coevolved with their hosts and acclimated to

54 their microenvironments. The transmission of lice mostly occurs through direct contact

55 between conspecific hosts. In exceptional cases, lice may migrate to another host and adapt to

56 their new ecological niches. These events are known as the host-switching phenomenon. In

57 this study, we report a case of human lice transfer to New World monkeys. This event

58 happened thousands of years ago when the first men reached the American continent. The

59 morphological examinations and genetic analyses performed on Pediculus mjobergi lice

60 collected from howler monkeys, Alouatta caraya, support these findings.

135 76 Introduction

77 A close host-parasite association can lead to cospeciation. This phenomenon is exemplified in

78 lice (Insecta: Phthiraptera), the obligate ectoparasites of birds and mammals, as their

79 partnership with their hosts has taken place for over 65 million years (1-4).

80 Humans can be infested by two types of sucking lice (Anoplura), Pediculus humanus and

81 Phthirus pubis, the crab louse (8). P. h. capitis (head louse) and P. h. humanus (body louse)

82 are ecotypes that have spread worldwide as modern humans have moved out of Africa over

83 the past 100,000 years (9). The molecular analysis of the mitochondrial (mt) genes

84 cytochrome oxidase subunit 1 (cox1) and cytochrome b (cytb) has allowed for their

85 classification into three haplogroups, which are designated A, B and C (10). Of these groups,

86 only haplogroup A is distributed worldwide and comprises both head and body lice (11,12).

87 Haplogroup B consists of head lice found in the Americas, Western Europe, Australia and

88 North Africa (13). Haplogroup C consists of head lice found in Nepal, Ethiopia and Senegal

89 (12,14,15). In addition to the inter-haplogroup diversity, human lice also demonstrate intra-

90 haplogroup diversity, which is illustrated by the many distinct A and B haplotypes (16-18).

91 The molecular analysis of lice collected from Pre-Columbian mummies showed that

92 haplogroups A and B were already present in the New World before the arrival of European

93 settlers (19,20). This finding supports an American origin for haplogroup B, followed by a

94 dispersal into the Old World by European colonists returning to Europe (13).

95 Interestingly, humans are not the only species in the Americas that harbor lice of the genus

96 Pediculus. In 1916, Ferris reported that New World monkeys harbored a sucking louse of the

97 Pediculidae family, suborder Anoplura (21). The first description of this louse was reported in

98 1910 by Mjoberg, who named it Pediculus affinis (22). Ferris suggested replacing the name

99 affinis with mjobergi (23). P. mjobergi has been found in three of the five families of

136 100 monkeys in the New World, the Cebidae (capuchin monkeys), the Atelidae (howler and spider

101 monkeys) and the Pitheciidae (titi monkeys) (22,24,25).

102 To investigate the relationship between P. humanus and P. mjobergi lice, we obtained and

103 compared P. mjobergi lice with lice recovered from a remote and isolated village in

104 Amazonia that has escaped globalization.

105 Results

106 P. mjobergi specimens highly resemble human lice

107 A comparison between the six adult P. mjobergi specimens from two wild howler monkeys

108 and 19 Amazonian human head lice showed that the P. mjobergi specimens are

109 morphologically similar to human lice (Figure 2). In particular, P. mjobergi were the same

110 gray color as the contemporary head lice recovered in the USA and were not the darker color

111 of the Amazonian head lice (Figure 2).

112 P. mjobergi mtDNA sequences belong to the mtDNA clades of human lice

113 To determine if the morphological similarities between the P. mjobergi specimens and human

114 lice are the result of convergent evolution or a recent common ancestor, we evaluated four

115 mtDNA markers (cytb, cox1, 16S rRNA, and nad2). Although each marker was targeted for

116 amplification from each of the six P. mjobergi specimens, only one marker was amplified

117 from all six individuals (cytb), and the remaining three markers (cox1, 16S rRNA, and nad2)

118 were only amplified from P. mjobergi individual #4 (Figure 3). Because the positive and

119 negative PCR amplification controls worked well in all of the amplification experiments, this

120 finding suggests that for P. mjobergi individuals #1-3 and #5-6, either the DNA extracted was

121 not of sufficient quality or the targeted genes had a sequence at the primer sites that was too

122 divergent.

123 The targets that could be amplified by PCR were sequenced, and the sequences generated

124 were aligned with the publicly available sequences. The P. mjobergi sequences obtained were

137 125 highly similar to the sequences obtained from human lice. To investigate their precise

126 relationship, we performed phylogenetic analyses. Maximum-likelihood (ML) and Neighbor-

127 Joining (NJ) phylogenetic analyses were performed for each of the four mtDNA genes and

128 showed that, in all four analyses, the human and P. mjobergi sequences split into three well-

129 supported haplogroups that corresponded to haplogroups A, B and C (Figure 4). In all four

130 analyses, the sequences of the monkey louse P. mjobergi #4 clustered with the Amazonian

131 head lice in haplogroup B. Similarly, the analysis of the cytb gene revealed that the remaining

132 monkey lice (P. mjobergi #1-3 and #5-6) have the widespread haplotype A5 of haplogroup A

133 (Figure 4A).

134 To investigate the geographical distribution of the haplotypes more precisely, we assembled a

135 dataset of 707 cytb sequences of body and head lice (424 from GenBank and 283 from our

136 laboratory). These 707 sequences span 36 worldwide geographic locations on five continents

137 and represent 32 distinct haplotypes, 21 from haplogroup A (66%), five from haplogroup B

138 (16%) and six from haplogroup C (SI Appendix). While P. mjobergi #1-3 and #5-6 have

139 haplotype A5, which is common worldwide (80% of locations and 49% of the 707 analyzed

140 human lice), the P. mjobergi #4 haplotype is closely related to haplotype B54 (Figure 5A),

141 which is unique to Amazonian head lice; these two haplotypes only differ in five nucleotides

142 that result in only a single amino acid difference (Lys129Ile) (SI Appendix).

143 We performed a similar analysis for the cox1 gene. Even though the sequences included in

144 this analysis were primarily collected from across the America (16), we were also able to

145 include 79 sequences from head and body lice from other parts of the world as well as the 33

146 lice sequences obtained in this study (SI Appendix). The 562 sequences collected represent 37

147 haplotypes, fifteen from haplogroup A, fifteen from haplogroup B and two from haplogroup

148 C. As for cytb, P. mjobergi #4 has a unique haplotype that is part of a subgroup of haplogroup

138 149 B that contains haplotype B18 from Argentina and Mexico and haplotypes B29-B30 from

150 Amazonia (Figure 5B).

151 This analysis result indicates that the P. mjobergi mtDNA sequences belong to the mtDNA

152 clades of human lice. In particular, specimen P. mjobergi #4 had, for all four markers,

153 sequences belonging to the B haplogroup, a haplogroup that most likely originated in the

154 Americas. Therefore, in addition to the morphological similarities, P. mjobergi lice and

155 human lice also share highly related mtDNA genomes, which suggests that their

156 morphological similarities are not the result of convergent evolution but are the result of their

157 close genetic relationship.

158 P. mjobergi nuclear sequences are related to the sequences of human lice

159 To further investigate whether P. mjobergi and human lice are closely related, we extended

160 our analysis to include nuclear markers, three genes and four intergenic spacers (Figure 3).

161 All three nuclear genes were amplified from P. mjobergi #4 as well as one gene from P.

162 mjobergi individual #3 (18S rRNA), but no PCR product was obtained for specimens #1-2

163 and #5-6 (Figure 3). A comparison of the sequences obtained in these amplifications with

164 publicly available human lice sequences showed that the P. mjobergi sequences are highly

165 related to the human lice sequences but are unique sequences in all cases. In particular, the

166 18S rRNA sequences from P. mjobergi individual #3 and P. mjobergi #4 are both unique (SI

167 Appendix).

168 The intergenic spacers PM1 and PM2 could be amplified from P. mjobergi #4, and PM1 was

169 also amplified from P. mjobergi #3 (Figure 3). Both spacers were also amplified from the 19

170 Amazonian head lice included in this study (see Methods). For PM1, 15 of the 19 Amazonian

171 head lice shared genotype 13 (G13) with human lice from Africa and Europe belonging to a

172 group that also includes P. mjobergi #3 (Figure 6A). The P. mjobergi #4 PM1 sequence,

173 however, belongs to genotype 21 (G21), which was previously characterized in Mexico. The

139 174 remaining Amazonian head lice belong to either genotype 36 (G36), which was already

175 characterized in Amazonia, or a new genotype that is highly related to G36 (Figure 6A). For

176 the PM2 marker, P. mjobergi #4 and Amazonian head lice share genotype 47 (G47) with

177 human lice from Africa, Oceania and America (Figure 6B).

178 This analysis of nuclear markers supports the results from the mtDNA analysis and shows that

179 the P. mjobergi specimens are closely related to human lice.

180 Discussion

181 In this study, we report for the first time molecular data for Pediculus mjobergi, which is

182 known as the louse of New World monkeys. The six individual lice analyzed were

183 morphologically similar to Pediculus humanus, the human lice (Figure 2). This observation

184 supports two previous descriptions reported in 1938 and 1983 (5,6).

185 The mtDNA analysis revealed that the New World monkey lice analyzed in this study have

186 cytb haplogroup A and B haplotypes, whereas the Amazonian head lice only have haplogroup

187 B haplotypes. Interestingly, while five P. mjobergi had a haplogroup A haplotype (A5) that is

188 prevalent (49% of analyzed human lice) and well distributed (80% of locations), P. mjobergi

189 #4 had a novel haplogroup B haplotype. This novel haplotype belongs in a group with

190 Amazonian head lice (Figure 5) that are prevalent in tropical areas in the Americas (Mexico,

191 Amazonia and northeast Argentina), a natural habitat of New World monkeys.

192 Because of the genetic and geographic proximity of the lice evaluated in this study, this

193 finding suggests that an exchange of genetic material has taken place between human lice and

194 New World monkey lice during their casual encounters. Consistent with this model, contact

195 between monkeys and humans has never been interrupted in South America. For example,

196 when monkeys are not hunted for their meat, they can become pets (26).

197 In blood sucking insects, including mosquitoes, host preference demonstrates a high degree of

198 flexibility when the favorite host species becomes scarce or is not accessible (27). Under these

140 199 conditions, physiological factors (hunger) and the physical abundance of available new hosts

200 may contribute to host switching (27). In this study, we show that this type of host switching

201 has occurred with human lice transmitted to monkeys. Moreover, the evidence suggesting that

202 two different mitochondrial genotypes (Clade A and Clade B) are prevalent, leads us to

203 conclude that the P. humanus-P. mjobergi shift is not a unique event because it appears to

204 have occurred at least twice. Among closely related mammals, louse-host cospeciation

205 primarily results from allopatry (28,29). Conversely, the sympatric life of related hosts can

206 result in an exchange, and potentially a recombination, of their respective lice, which would

207 overrule the strict louse-host coevolution paradigm.

208 Materials and Methods

209 Lice samples. For P. mjobergi, six adult specimens were collected from two wild howler

210 monkeys, Alouatta caraya, located in northeast Argentina. The first four lice were collected

211 from monkey #B2188 from the National Park Iguaza, Province of Misiones. The other two

212 lice were found on monkey B1395 from the Province of Corientes (Figure 1).

213 For the human lice in the Americas, 19 Amazonian human head lice were included in this

214 study. We recovered these specimens in 2013 from members of the Wayapi community (7)

215 living in “Trois-Sauts”, a remote and isolated village on the Oyapock River along the border

216 between French Guyana and Brazil (Figure 1).

217 Human head and body lice collected in France were included as positive controls.

218 Lice were photographed on their dorsal and ventral sides using a fixed camera (Olympus

219 DP71, Rungis, France).

220 DNA samples. Genomic DNA was extracted using the QIAamp DNA tissue extraction kit

221 (Qiagen, Hilden, Germany) in an EZ1 apparatus following the manufacturer's instructions.

222 DNA from each louse was eluted in 100 µl of TE buffer and stored at -20 °C.

141 223 PCR amplifications. Seven genes, three nuclear genes [18S rRNA, glycerol-3-phosphate

224 dehydrogenase (GPD) and RNA polymerase II largest subunit (RPII)] and four mtDNA

225 [(cytb), cytochrome oxidase subunit 1 (cox1), 16S ribosomal RNA, and NADH

226 dehydrogenase subunit 2 (nad2)] were investigated. We also targeted intergenic regions using

227 two highly polymorphic microsatellites (PM1 and PM2) as previously described (30).

228 All primers used for these experiments are described in the SI Appendix.

229 PCR amplifications were conducted in a Peltier PTC-200 model thermal cycler (MJ Research

230 Inc., Watertown, Mass. USA). PCR reactions were prepared on ice and contained 3 μl of

231 DNA template, 4 μl of Phusion HF Buffer, 250 μM of each nucleotide, 0.5 μM of each

232 primer, 0.2 μl of high fidelity Phusion DNA Polymerase (Finnzymes, Thermo Scientific,

233 Vantaa, Finland) and water to a final reaction mixture volume of 20 μl. The cycling

234 conditions were 98°C for 30 sec; 35 cycles of 5 sec at 98°C, 30 sec at Tm (SI Appendix), 15

235 sec at 72°C; and a final extension time of 5 min at 72°C. PCR positive and negative controls

236 were included in each assay. The success of PCR amplification was then verified by

237 electrophoresis of the PCR product on a 1.5% agarose gels.

238 Sequencing. NucleoFast 96 PCR Plates (Macherey-Nagel EURL, France) were used to purify

239 the PCR products before sequencing. The purified PCR products were then sequenced in both

240 directions (with the PCR primers) using a BigDye Terminator version 1.1 cycle sequencing-

241 ready reaction mix (Applied Biosystems, Foster City, CA) and an ABI 3100 automated

242 sequencer (Applied Biosystems). The program Chromas Pro software (Technelysium PTY,

243 Australia) was used to analyze, assemble and correct the sequences.

244 The sequences obtained in this study and those representing the haplotypes determined from

245 the set of used sequences have been deposited in GenBank (KM579408 - KM579584).

246 Sequence analysis. For each of the target genes, the nucleotide sequences obtained in this

247 study were aligned with the sequences available on public databases (GenBank) using

142 248 CLUSTALX 2.0.11 (31). MEGA 6 was used for phylogenetic analyses and tree

249 reconstruction (32). NJ phylogenetic analysis was performed using the Maximum Composite

250 Likelihood method with 200 replicates. ML analyses were performed using the Jukes-Cantor

251 model with 200 replicates.

252 To determine the phylogeographic relationships among the lice, median-joining networks

253 were assembled for the lice haplotypes of two mitochondrial markers, cytb and cox 1, using

254 the method of Bandelt with the program Network version 4.6.1.1 (33). The sequence between

255 nucleotide positions 370-642 of cytb and 748-1026 of cox1 was determined for all the

256 specimens. Partial gene sequences were aligned with the sequences available in GenBank.

257 The percent similarities were determined using MEGA6 (32).

258 Ancestral sequences were reconstructed for each node of the cytb phylogenetic tree using the

259 marginal reconstruction approach with BASEML of the PAML4 software package (34).

260

261

262

263

264

265

266

267

268

269

270

271

272

143 273 Reference List

274

275 1. Barker,S.C. (1991) Evolution of host-parasite associations among species of lice and 276 rock-wallabies: coevolution? (J. F. A. Sprent Prize lecture, August 1990). Int. J. 277 Parasitol., 21, 497-501.

278 2. Barker,S.C. (1994) Phylogeny and classification, origins, and evolution of host 279 associations of lice. Int. J. Parasitol., 24, 1285-1291.

280 3. Hafner,M.S. and Page,R.D. (1995) Molecular phylogenies and host-parasite 281 cospeciation: gophers and lice as a model system. Philos. Trans. R Soc Lond B Biol 282 Sci., 349, 77-83.

283 4. Smith,V.S., Ford,T., Johnson,K.P., Johnson,P.C., Yoshizawa,K. and Light,J.E. (2011) 284 Multiple lineages of lice pass through the K-Pg boundary. Biol Lett., 7, 782-785.

285 5. Ewing,H.E. (1938) The Sucking Lice of American Monkeys. The Journal of 286 Parasitology., 24, 13-33.

287 6. Maunder,J.W. (1983) The Appreciation of lice. Proc. R. Inst. Great Britain., 55, 1-31.

288 7. Woerther,P.L., Angebault,C., Lescat,M., Ruppe,E., Skurnik,D., Mniai,A.E., 289 Clermont,O., Jacquier,H., Costa,A.D., Renard,M. et al. (2010) Emergence and 290 dissemination of extended-spectrum beta-lactamase-producing Escherichia coli in the 291 community: lessons from the study of a remote and controlled population. J. Infect. 292 Dis., 202, 515-523.

293 8. Durden,L.A. (2001) Lice (Phthiraptera). In Samuel,W.M., Pybus,M.J. and Kocan,A.A. 294 (eds.), Parasitic diseases of wild mammals. Ames: Iowa State University Press, pp. 3- 295 17.

296 9. Ascunce,M.S., Toups,M.A., Kassu,G., Fane,J., Scholl,K. and Reed,D.L. (2013) 297 Nuclear genetic diversity in human lice (Pediculus humanus) reveals continental 298 differences and high inbreeding among worldwide populations. PLoS. One., 8, 299 e57619.

144 300 10. Reed,D.L., Smith,V.S., Hammond,S.L., Rogers,A.R. and Clayton,D.H. (2004) Genetic 301 analysis of lice supports direct contact between modern and archaic humans. PLoS. 302 Biol, 2, e340.

303 11. Reed,D.L., Light,J.E., Allen,J.M. and Kirchman,J.J. (2007) Pair of lice lost or 304 parasites regained: the evolutionary history of anthropoid primate lice. BMC. Biol., 5, 305 7.

306 12. Xiong, H., Campelo, D., Pollack, R. J., Raoult, D., Shao, R., Alem, M., Ali, J., Bilcha, 307 K., Barker, S. C. (2014). Second�generation sequencing of entire mitochondrial 308 coding�regions (� 15.4 kb) holds promise for study of the phylogeny and taxonomy of 309 human body lice and head lice. Medical and veterinary entomology, 28(S1), 40-50.

310 13. Boutellis,A., Abi-Rached,L. and Raoult,D. (2014) The origin and distribution of 311 human lice in the world. Infect. Genet. Evol., 23, 209-217.

312 14. Light,J.E., Toups,M.A. and Reed,D.L. (2008) What's in a name: the taxonomic status 313 of human head and body lice. Mol. Phylogenet. Evol., 47, 1203-1216.

314 15. Veracx,A., Boutellis,A. and Raoult,D. (2013) Genetic recombination events between 315 sympatric Clade A and Clade C lice in Africa. J. Med. Entomol., 50, 1165-1168.

316 16. Ascunce,M.S., Fane,J., Kassu,G., Toloza,A.C., Picollo,M.I., Gonzalez-Oliver,A. and 317 Reed,D.L. (2013) Mitochondrial diversity in human head louse populations across the 318 Americas. Am. J. Phys. Anthropol., 152, 118-129.

319 17. Leo,N.P., Campbell,N.J., Yang,X., Mumcuoglu,K. and Barker,S.C. (2002) Evidence 320 from mitochondrial DNA that head lice and body lice of humans (Phthiraptera: 321 Pediculidae) are conspecific. J. Med. Entomol., 39, 662-666.

322 18. Light,J.E., Allen,J.M., Long,L.M., Carter,T.E., Barrow,L., Suren,G., Raoult,D. and 323 Reed,D.L. (2008) Geographic distributions and origins of human head lice (Pediculus 324 humanus capitis) based on mitochondrial data. J. Parasitol., 94, 1275-1281.

325 19. Boutellis,A., Drali,R., Rivera,M.A., Mumcuoglu,K.Y. and Raoult,D. (2013) Evidence 326 of sympatry of clade a and clade B head lice in a pre-Columbian Chilean mummy 327 from Camarones. PLoS. One., 8, e76818.

145 328 20. Raoult,D., Reed,D.L., Dittmar,K., Kirchman,J.J., Rolain,J.M., Guillen,S. and 329 Light,J.E. (2008) Molecular identification of lice from pre-Columbian mummies. J. 330 Infect. Dis., 197, 535-543.

331 21. Ferris,G.F. (1916) A catalogue and lost list of the Anoplura. The Academy, San 332 Francisco.

333 22. Durden,L.A. and Musser,G.G. (1994) The Mammalian Hosts of The Sucking Lice 334 (Anoploura) of the World: A Host-Parasite List. Bull. Soc. Vector Ecol., 19, 130-168.

335 23. Ferris,G.F. (1951) The sucking lice. Mem. Pacif. Coast Entomol. Soc..

336 24. Finstermeier,K., Zinner,D., Brameier,M., Meyer,M., Kreuz,E., Hofreiter,M. and 337 Roos,C. (2013) A mitogenomic phylogeny of living primates. PLoS. One., 8, e69504.

338 25. Price,M.A. and Graham,O.H. (1997) Chewing and sucking lice as parasites of 339 mammals and birds. Technical Bulletin., 1849, 1-309.

340 26. Horwich,R.H. (1998) Effective solutions for howler conservation. International 341 Journal of Primatology, 19, 579-598.

342 27. Takken,W. and Verhulst,N.O. (2013) Host preferences of blood-feeding mosquitoes. 343 Annu. Rev. Entomol., 58, 433-453.

344 28. Mayr,E. (1944) Systematics, and the Origin of Species. Columbia University Press., 345 New York, NY.

346 29. Mayr,E. (1963) Animal Species, and Evolution. Harvard university Press., Cambridge, 347 MA.

348 30. Li,W., Ortiz,G., Fournier,P.E., Gimenez,G., Reed,D.L., Pittendrigh,B. and Raoult,D. 349 (2010) Genotyping of human lice suggests multiple emergencies of body lice from 350 local head louse populations. PLoS. Negl. Trop. Dis., 4, e641.

351 31. Larkin,M.A., Blackshields,G., Brown,N.P., Chenna,R., McGettigan,P.A., 352 McWilliam,H., Valentin,F., Wallace,I.M., Wilm,A., Lopez,R. et al. (2007) Clustal W 353 and Clustal X version 2.0. Bioinformatics., 23, 2947-2948.

146 354 32. Tamura,K., Stecher,G., Peterson,D., Filipski,A. and Kumar,S. (2013) MEGA6: 355 Molecular Evolutionary Genetics Analysis version 6.0. Mol. Biol. Evol., 30, 2725- 356 2729.

357 33. Bandelt,H.J., Forster,P. and Rohl,A. (1999) Median-joining networks for inferring 358 intraspecific phylogenies. Mol. Biol. Evol., 16, 37-48.

359 34. Yang,Z. (2007) PAML 4: phylogenetic analysis by maximum likelihood. Molecular 360 biology and evolution., 24, 1586-1591. 361

362

363

364

365

366

367

368

369

370

147 Trois-Sauts village, French Guyana

National Park Iguaza, 148 Province of Misiones, Argentina

Province of Corientes, Argentina

Figure 1: Geographical localization of lice sampling New World monkey lice: Argentina Human head lice: french Guyana 149

Figure 2: Morphological comparisons show that P. mjobergi lice highly resemble human lice (1, 2) human head lice from USA; (3, 4) P. mjobergi from Argentina; (5, 6) human head lice from Amazonia. Male; female Amplified Not amplified 150

Figure 3: PCR results of genetic markers in monkey lice P. mjobergi. Lice # (1-4) were collected from monkey [B2188]. Lice # (5 and 6) were collected from monkey [B1395]. (A) (B) Body louse Head louse Haplogroup A Head & body lice Haplogroup A includes sequences of P.mjobergi #1-3 and #5-6.

Haplogroup B Haplogroup B

Haplogroup C

151 (C) (D)

Haplogroup A Haplogroup A

Haplogroup B Haplogroup B

Haplogroup C Haplogroup C

Figure 4. P. mjobergi sequences highly related to sequences obtained from human lice. Maximum-likelihood (ML) and neighbor-joining phylograms resulting from analysis of mt genes: (A) cytb, (B) cox1, (C) 16S rRNA and (D) Nad2. Bootstrap support values (greater than 80) are located above and below the nodes, respectively. Mt clade memberships are indicated to the right of each tree. [A] Africa [B] Asia Europe America Oceania P. mjobergi 152 [C]

Figure 5A. Statistical parsimony network for the cytb gene haplotypes found in haplogroup A, haplogroup B and haplogroup C. Each connecting branch represents a single mutational step. Sizes are scaled and represent relative frequencies. [A] [B] Africa Asia Europe America Oceania P. mjobergi 153

[C]

Figure 5B. Statistical parsimony network for the cox1 gene haplotypes found in haplogroup A, haplogroup B and haplogroup C. Each connecting branch represents a single mutational step. Sizes are scaled and represent relative frequencies. KM579411 G13 Amazonia EU928831 G28 Russia EU928841 G38 France (A) KM579409 G13 P. mjobergi #3 (B) EU928843 G40 Portugal Africa KM579412 G13 Amazonia EU928830 G27 Russia KM579414 G13 Amazonia EU928847 G44 Mexico Asia EU928862 G59 France KM579415 G13 Amazonia KM579435 G39 Orlando strain Europe KM579416 G13 Amazonia EU928805 G2 France KM579417 G13 Amazonia EU928806 G3 Russia America EU928809 G6 KM579418 G13 Russia Amazonia EU928810 G7 France Oceania KM579419 G13 Amazonia KM579434 G6 France KM579421 G13 Amazonia EU928851 G48 Russia EU928856 G53 P. mjobergi KM579422 G13 Amazonia Russia EU928857 G54 Russia KM579423 G13 Amazonia EU928833 G30 Russia KM579425 G13 Amazonia EU928834 G31 Russia EU928835 G32 Russia KM579427 G13 Amazonia KM579428 G33 Nepal KM579426 G13 Amazonia EU928837 G34 France EU928793 G13 France, Portugal Senegal EU928817 G14 Burundi KM579443 EU928785 G5 Burundi, Ethiopia P. mjobergi #4 KM579444 - KM579454 G47 Amazonia EU928790 G10 Burundi KM579442 G47 Amazonia EU928786 G6 Burundi KM579441 G47 Australia KM579436 - KM579439 G47 Algeria EU928783 G3 Burundi, Rwanda KM579433 G47 Australia EU928801 G21 Mexico EU928861 G58 Mexico EU928794 G14 Russia EU928853 G50 France 154 KM579430 G61 Nepal KM579408 G21 P. mjobergi #4 EU928846 G43 Russia EU928803 G23 Russia EU928840 G37 Mexico EU928798 G18 France Russia Mexico Ethiopia, Rwanda EU928839 G36 Russia Mexico EU928795 G15 Russia EU928855 G52 EU928854 G51 France EU928796 G16 Russia EU928852 G49 France KM579420 G36 Amazonia KM579429 G60 Nepal KM579413 Amazonia EU928821 G18 Rwanda EU928829 G26 Rwanda KM579410 G36 Amazonia EU928828 G25 Burundi JX178753 G36 Amazonia EU928811 G8 Rwanda KM579431 G62 EU928792 G12 Burundi Ethiopia KM579440 G64 Ethiopia EU928791 G11 Burundi EU928826 G23 Rwanda EU928782 G2 Burundi EU928823 G20 Rwanda Russia EU928787 G7 Burundi EU928848 G45 EU928825 G22 Burundi EU928781 G1 Burundi EU928824 G21 Rwanda JQ652455 G35 Burundi EU928822 G19 Burundi EU928789 G9 Burundi KM579432 G63 Ethiopia EU928814 G11 Rwanda EU928802 G22 Burundi EU928813 G10 Burundi EU928799 G19 Russia EU928820 G17 Burundi EU928819 G16 Burundi EU928860 G57 France 0.002 EU928859 G56 Rwanda

0.005

Figure 6. Phylogenetic organization of human lice and P. mjobergi based on two nuclear intergenic spacers, PM1, PM2, using the maximum likelihood method. (A) (B)

KM579466 P. mjobergi #3 GTA -- AAA ------GTTC - GCCGGCC KM579492 P. mjobergi #4 TGTCTGTGGCG TT KM579467 P. mjoberg i #4 GTA -- AAA ------GTTC - GCCAGTA KM579490 Amazonia HL TGTCTATGGCG TT KM579482 France BL AF139484 Russia HL GCA -- AAA ------GTCA - ACAAGTC TGTCTATGGCG TT KM579491 Ethiopia HL AY236410 France HL GCT -- AAA ------GTTCGACCAGTC TGTCTATGACG TT KM579481 USA BL FJ267394 USA HL GTA -- AAA ------GTTC - GCCAGTC TGTCCATAGTG CT AY316876 UK HL FJ267396 USA HL GTA -- AAA ------ATTC - GCCAGTC AATCTATGGCG TT FJ267397 HL UT USA GTA -- AAA ------ATTC - GCCAGTC AY316877 Laos HL AATCTATGGCG TT AY316878 Ecuador HL AATCTATGGCG TT AY236418 Thailand HL GTA -- AAA ------GTTCAGCCAGTC AY316881 Germany HL AATCTATGGCG TT KM579475 Algeria HL GTA -- AAA ------GTTC - GCCAGTC AY316882 Germany HL AATCTATGGCG TT KM579476 Algeria HL GTA -- AAA ------GTTC - GCCAGTC AY316886 UK HL KM579477 Algeria HL GTA -- AAA ------GTTC - GCCAGTC AATCTATGGCG TT AY316887 Germany HL AATCTATGGCG TT KM579468 Australia HL GTT -- AAA ------GTTC - GCCAGTC AY316888 Germany HL KM579474 Amazonia HL GTT -- AAA ------GTTC - GCCAGTC AATCTATGGCG TT AY316890 Nepal HL AY077775 Australia HL GTT -- AAA ------GTTC - GCCAGTC AATCTATGGCG TT AY236414 HL Portugal GTT -- AAA ------GTTC - GCCAGTC AY316891 Ethiopia HL AATCTATGGCG TT AY236417 HL China GTT -- AAA ------GTTC - GCCAGTC AY316900 Nepal HL AATCTATGGCG TT EF570919 Indonesia HL GTA -- GTAACGA --- GTTC - GCCAGTC AY316901 Papua New Guinea AATCTATGGCG TT AY316905 Ecuador HL KM579469 Nepal BL GCA -- GTAACGACGAGT CA - ACAAGTC AATCTATGGCG TT AY316906 Ecuador HL AF139479 Russia BL GCA -- GTAACGACGAGT CA - ACAAGTC AATCTATGGCG TT AF139480 USA BL GCA -- GTAACGACGAGT CA - ACAAGTC AY316907 Germany HL AATCTATGGCG TT AY236416 Netherlands BL GCA -- GTAACGACGAGT CA - ACAAGTC AY316909 Germany HL AATCTATGGCG TT KM579470 Nepal BL GCA -- GTAACGACGAGT CA - ACAAGTC 155 AY316911 UK HL AATCTATGGCG TT AY589938 Nepal BL GCA -- GTAACGACGAGT CA - ACAAGTC FJ267455 USA HL AATCTATGGCG TT AY589941 Iran HL GCA -- GTAACGACGAGT CA - ACAAGTC AY316903 Papua New Guinea HL FJ267398 Canada BL GCA -- GTAACGACGAGT CA - ACAAGTC AATCTATGGCG TT AY589940 Iran HL GCA -- GTAACGACGAGT CA - ACAAGTC FJ267456 USA HL AATCTATGGCG TT AY589942 Nepal BL GCA -- GTAACGACGAGT CA - ACAAGTC FJ267457 USA HL AATCTATGGCG TT AF139488 Tunisia BL GCA -- GTAACGACGAGT CA - ACAAGTC FJ267459 Canada BL AATCTATGGCG TT AF139478 France BL GCA -- GTAACGACGAGT CA - ACAAGTC AY316879 Panama HL KM579471 France BL GCA -- GTAACGACGAGT CA - ACAAGTC AATCTGTGGCG TT AY316880 Panama HL AATCTGTGGCG TT AY236411 Algeria BL ACA -- GTAACGACGAGT CA - ACAAGTC AY316883 Ethiopia HL AY589939 Nepal HL GCA -- GTAACGACGAGT CA - ACCAGTC AACCTATGGTG TC AY316885 Ethiopia HL AF139481 Peru BL GCA -- GTAACGACGAGT CA - ACCAGTC AACCTATGGTG TC AY316898 Ethiopia HL FJ267395 USA HL GTA -- ATATTCGTACGT TC - GCCAGTC AACCTATGGTG TC AY316884 Ethiopia HL KM579479 Nepal HL GCA -- GTTTCGACGAGT CA - ACAAGTC AATCCATGGCG TT KM579480 USA BL GCA -- GTTTCGACGAGT CA - ACAAGTC AY316892 Ethiopia HL AATCCATGGCG TT FJ267458 USA HL FJ267399 Burundi BL GCATAGTAATGA - CAGACA - ATAAGTC AATCCATGGCG TT AY236412 Rwanda BL GCATAGTAATGA - CAGACA - ATAAGTC AY 316889 Nepal HL AATCTATGGTG TC AF139486 Burundi BL GCATAGTAATGA - CAGACA - ATAAGTC AY316894 Ethiopia HL AATCTATGGTG TC KM579472 Ethiopia BL GCATAGTAATGA - CAGACA - ATAAGTC AY316897 Ethiopia HL AATCTATGGTG TC KM579473 Ethiopia BL GCATAGTAATGA - CAGACA - ATAAGTC AY316893 Ethiopia HL AF139482 Zimbabwe BL GCACGGTAATGA - CAGACA - ATAAGTC AACCCATGGTG TT AY316895 Ethiopia HL AY236413 Burundi HL GCATAGTAATGA - TAGACA - ATAATTC AATTCATGGTG TC AY236415 Rwanda HL GCATAGTAATGA - TAGACA - ATAATTC AY316896 Ethiopia HL AATCCATAGTC CT

AY316899 Nepal HL AATCTATGGTG TT Supplementary file 1: Alignment of variable nucleotide positions AY316902 Papua New Guinea HL AATCTACGGCG TT AY316904 Papua New Guinea HL AATCTACGGCG TT

(A) 18S rRNA gene; (B) RPII gene; (C) GPD gene AY316908 Germany HL AATCTATGGTG CT AY316910 Germany HL BL: body louse; HL: head louse AATCTATGGTG CT (C)

KM579584 P. mjobergi 4# ATTTTATTTTTTAAA AAAACATGAAACTGA ATTATAA - TCG

KM579582 Nepal BL ATTTTAT - TTTCAAAATACTAGA CTGGGAAATTGC - G - CGA

KM579583 Nepal BL ATTTTAT - TTTTAAAATACTAGA CTGGTAAATTGC - G - CGA

KM579575 Algeria HL A --- TATTTTTTAAAAAAC CATGAAACTGAATTA T - ATCGA KM579576 Algeria HL A --- TATTTTTTAAAAAAC CATGAAACTGAATTA T - ATCGA

KM579580 Nepal HL G --- AAT - TTTTAAAAAACCATA AAACTAAATTAC - ATCGA KM579578 Ethiopia HL G --- AAT - TTTTAAAAAACCATA AAACTAAATTAC - ATCGA

KM579577 Amazonia HL GTATGTGAAAATAAT AAGACATAAAACTAA ACTATTAACGA

156 KM579579 Ethiopia BL GGATGTGCAAATGAA ATACCATACAACTAG TCTATTAATTG

Supplementary file 1: Alignment of variable nucleotide positions (A) 18S rRNA gene; (B) RPII gene; (C) GPD gene BL: body louse; HL: head louse A B 157

Body louse Head louse Head & body lice

Haplogroup C Haplogroups A & B Haplotype 54 Supplementary file 2. Haplogroup B Haplogroups A & C Haplotype 39 A . Chronogram for the lice haplotypes resulting from analysis cytb gene. Haplogroup A Haplogroups B & C B. Nucleotide positions shared between the different haplotypes * Non synonymous nucleotide position Acc. No. Acc. KM579559 7 KM579563 1 KM579564 8 KM579568 1 KM579551 2 KM579554 1 KM579555 1 KM579566 5 KM579560 1 KM579561 1 KM579538 1 KM579539 3 KM579543 3 KM579545 1 KM579546 17 KM579562 51 KM579565 11 KM579550 20 KM579567 19 KM579569 18 KM579552 Total 20 KM579547 10 KM579548 16 KM579549 346 KM579542 2 2 KM579558 8 37 KM579553 USA 4 4 KM579541

Peru 1 1 KM579540 1 1 KM579544

Panama 1 11 84 707 Mexico

Honduras 14 1 16 21 1 KM579557 KM579556 Ecuador

Chille 1 20 76 Canada Brazil 8 2

Amazonia 20 19

Papua-New-Guinea 1 7 8

New Zeeland 1017110 1 1 3 1 1 934 9

Australia 10 6 United Kingdom Portugal Hungary

Germany 3 28 France

Yemen 1 44 9 10 4 32 57 16 18 50 10 7 2 1 8 10

Taiwan 1 1 1 4 1 Russia 1 2

Philippines 9 1 17 Nepal 20 Mongolia

Laos 1 1 45 5 27 1 44 4 10 4 38

Japan 10 Iran

China 6 5 5 Senegal 6 Rwanda Madagascar Kenya Ethiopia Burundi 26 18 85 7 18 13 11 6 2 10 1 6 83 14 31 1 cytb gene haplotype frequency of human lice per location worldwide worldwide lice location human of per frequency 3:file cytb gene haplotype Supplementary Algeria

Haplotype 1 C43 C44 51 7 C42 B32 B33 B34 A14 A53 B36 24 C39 A16 A19 A45 A52 B54 1 C40 C41 A17 2 A18 A11 A12 A13 A2 A3 A6 3 A7 A9 1 A10 A1 1 A4 A5 18A8 21 7 17 13 2

158 Europe Oceania Africa Asia America Acc. No. 1 KM579520 1 KM579522 2 KM579523 2 KM579524 2 KM579502 2 KM579503 1 KM579504 2 KM579506 3 KM579507 7 KM579494 1 KM579496 1 KM579498 1 KM579499 18 KM579493 Total 15 KM579521

Canada 2 12 KM579510 21 2 KM579518 1 KM579519 Mexico 14 7 561 3411 3 KM579511 1 4 KM579512 1 1 KM579513 1 KM579515 1 KM579516 1 KM579517 22 2 KM579505 1 2 KM579495 1 KM579497 10 7 239 KM579500 92 4 113 KM579501 USA 42 5 75 KM579508

Panama 2 2 KM579514

Honduras 25 17 42 KM579509 1 Amazonia 15 Ecuador Peru Colombia 6 1 2 2 1 18 Argentina 10

Papua-New-Guinea 9 195 2 2 1 4 Cook Islands Australia 3 1 9 235 2 1 2 17 29 2 177 1 3 United Kingdom 24 1 Norway

France 1 Yemen 5 1 2 2 2 3 1 Philippines 7 Nepal Mongolia Ethiopia Burundi Algeria 3 2 4 1 414 1 1 230

Haplotype A1 A2 A3 12 B19 B20 B21 B22 B23 B24 B25 B26 B27 B28 B29 2 C31 2 C32 B30 A4 A5 A6 A7 A8 2 A9 A10 A11 A12 A13 A14 1 A15 1 3 B16 B17 B18 cox1 gene haplotype frequency of human lice per location worldwide humanlicelocation of per frequency haplotype 4:filegene cox1 Supplementary

159

Chapitre 4 : Détection et monitoring de la résistance moléculaire des poux de corps à la perméthrine

161

Préambule

La prévalence des poux de corps chez les personnes fréquentant les foyers pour sans-abri à Marseille atteint 22% [51]. Il était donc indispensable de mettre en œuvre des stratégies de lutte contre le pou de corps vecteur de trois maladies ayant tué des millions de personnes par le passé [30]. Comme on l’avait constaté, les poux de corps sont extrêmement contagieux et peuvent se propager par contact direct entre les personnes ou par le biais de la literie dans les foyers. Les mesures thérapeutiques classiques telles le changement fréquent de vêtements ou le lavage des couvertures à 50 °C restent difficiles à appliquer auprès de ces personnes qui vivent dans des conditions précaires. L'ivermectine par voie orale réduit le prurit et la prévalence des infestations par les poux de corps chez ces patients mais l'effet reste transitoire [52,53]. Ces résultats montrent que l'éradication complète de ces ectoparasites chez les personnes sans-abri demeure un défi permanent.

Un essai clinique ayant pour objectif l’éradication des poux de corps par le port de sous-vêtements imprégnés de perméthrine dans les populations défavorisées de Marseille avait été mis en place. Lors de cette étude, il était important de vérifier la sélection possible de la

163 résistance moléculaire à la perméthrine par évaluation de la prévalence des mutations conférant cette résistance avant, pendant et après l’étude. Pour cela, il fallait développer une méthode rapide, fiable et permettant la réalisation de plusieurs centaines de tests avant le début de l’essai clinique.

Nous avions opté pour une méthode de génotypage par PCR en temps réel avec sondes d’hybridation qui s’était révélée sensible et spécifique [54].

Les résultats de l’essai clinique avaient suggéré d’éviter l’usage de la perméthrine en raison de l’augmentation de la résistance kdr chez les populations de poux de corps ciblées [55].

164

Article VIII: Detection of a Knockdown Resistance Mutation Associated with Permethrin Resistance in the Body Louse Pediculus humanus corporis by Use of Melting Curve Analysis Genotyping

Journal of Clinical Microbiology 50:2229-2233.

165

Detection of a Knockdown Resistance Mutation Associated with Permethrin Resistance in the Body Louse Pediculus humanus corporis by Use of Melting Curve Analysis Genotyping

Rezak Drali, Samir Benkouiten, Sékéné Badiaga, Idir Bitam, Jean Marc Rolain and Philippe Brouqui

J. Clin. Microbiol. 2012, 50(7):2229. DOI: Downloaded from 10.1128/JCM.00808-12. Published Ahead of Print 9 May 2012.

Updated information and services can be found at:

http://jcm.asm.org/content/50/7/2229 http://jcm.asm.org/

These include: SUPPLEMENTAL MATERIAL http://jcm.asm.org/content/suppl/2012/06/12/50.7.2229.DC1.ht ml

on July 5, 2012 by INIST-CNRS BiblioVie REFERENCES This article cites 30 articles, 7 of which can be accessed free at: http://jcm.asm.org/content/50/7/2229#ref-list-1

CONTENT ALERTS Receive: RSS Feeds, eTOCs, free email alerts (when new articles cite this article), more»

Information about commercial reprint orders: http://journals.asm.org/site/misc/reprints.xhtml To subscribe to to another ASM Journal go to: http://journals.asm.org/site/subscriptions/

167 Detection of a Knockdown Resistance Mutation Associated with Permethrin Resistance in the Body Louse Pediculus humanus corporis by Use of Melting Curve Analysis Genotyping

Rezak Drali,a,b Samir Benkouiten,a Sékéné Badiaga,a Idir Bitam,c Jean Marc Rolain,a and Philippe Brouquia Aix-Marseille Université, URMITE CNRS-IRD, UMR 6236/198 IHU Méditerranée Infection, Faculté de Médecine et de Pharmacie, Marseille, Francea; Service des Entérobactéries et Hygiène de l’Environnement, Institut Pasteur d’Algérie, Algiers, Algeriab; and Service d’Ecologie des Systèmes Vectoriels, IRD 198, Institut Pasteur d’Algérie, Algiers, Algeriac Downloaded from Louse-borne diseases are prevalent in the homeless, and body louse eradication has thus far been unsuccessful in this population. We aim to develop a rapid and robust genotyping method usable in largefield-based clinical studies to monitor permethrin resistance in the human body louse Pediculus humanus corporis. We assessed a melting curve analysis genotyping method based on real-time PCR using hybridization probes to detect the M815I-T917I-L920F knockdown resistance (kdr) mutation in the paraorthologous voltage-sensitive sodium channel (VSSC)� subunit gene, which is associated with permethrin resistance. The 908-bp DNA fragment of the VSSC gene, encoding the� subunit of the sodium channel and encompassing the three mutation sites, was PCR sequenced from 65 lice collected from a homeless population. We noted a high prevalence of the 3 indicated mutations in the body lice collected from homeless people (100% for the M815I and L920F mutations and 56.73% for the T917I muta- http://jcm.asm.org/ tion). These results were confirmed by melting curve analysis genotyping, which had a calculated sensitivity of 100% for the M815I and T917I mutations and of 98% for the L920F mutation. The specificity was 100% for M815I and L920F and 96% for T917I. Melting curve analysis genotyping is a fast, sensitive, and specific tool that is fully compatible with the analysis of a large number of samples in epidemiological surveys, allowing the simultaneous genotyping of 96 samples in just over an hour (75 min). Thus, it is perfectly suited for the epidemiological monitoring of permethrin resistance in human body lice in large-scale clinical studies. on July 5, 2012 by INIST-CNRS BiblioVie

he human body louse Pediculus humanus corporis(P. h. corpo- (11). This resistance is generated by three point mutations, result- Tris) is a hematophagous ectoparasite that lives and multiplies ing in the amino acid substitutions M815I, T917I, and L920F in in clothing. Body lice are vectors of 3 major infectious diseases: the paraorthologous voltage-sensitive sodium channel (VSSC)� epidemic typhus caused by Rickettsia prowazekii, relapsing fever subunit (17). According to various authors, these mutations are caused by Borrelia recurrentis, and trench fever caused by Barto- often found en bloc as a resistant haplotype (7). In Marseille, we nella quintana( 24). The poor living conditions and crowded sit- wished to conduct a pilot prospective study to eradicate body uations in homeless, war refugee, or natural disaster victim pop- louse infestations in the homeless individuals frequenting the ulations provide ideal conditions for the spread of lice (23). Body shelters by providing them with underwear impregnated with a louse infestation has been observed in 22% of the sheltered home- solution of 0.4% permethrin. For this purpose, it was critical to be less population in Marseille, France, causing trench fever, epi- able to evaluate permethrin resistance before and during the study demic typhus, and relapsing fever (4), and coinfestation with head using a rapid and specific molecular method. lice is often noted. Consequently, all measures that can be used to There are currently several different methods available for de- decrease the burden of these ectoparasites in homeless people are tecting the mutations responsible for kdr in head lice, including warranted to avoid the spread and/or outbreak of these diseases. PCR-restriction fragment length polymorphism (PCR-RFLP) Clothing change as well as the use of ivermectin to eradicate the (15) and quantitative sequencing (QS), real-time PCR amplifica- body lice in a cohort of homeless individuals in Marseille was tion of a specific allele (rtPASA), and serial invasive signal ampli- unsuccessful (10), suggesting that the complete eradication of this fication reaction (SISAR) (6). Nevertheless, these methods are ectoparasite is a true challenge (1, 5). It was therefore essential for time-consuming and not applicable to the investigation of large us to investigate other strategies to reach this goal. quantities of arthropods. We report here a rapid and reliable sin- During the 1980s, a study conducted by the U.S. Army demonstrated that the use of uniforms impregnated with 0.125 mg/ 2 cm permethrin effectively eradicated body lice (28). Received 26 March 2012 Returned for modification 17 April 2012 The resistance to permethrin, known as knockdown resistance Accepted 27 April 2012 (kdr), wasfirst identified in the housefly Musca domestica( 8) and Published ahead of print 9 May 2012 is due to the presence of mutations in the gene encoding the� Address correspondence to Jean Marc Rolain, [email protected], or subunit of the sodium channel that is responsible for the depolar- Philippe Brouqui, [email protected]. ization of nerve cells (9). To date, permethrin resistance has never Supplemental material for this article may be found at http://jcm.asm.org/. been studied in P. h. corporis. However, in the head louse Pediculus Copyright © 2012, American Society for Microbiology. All Rights Reserved. humanus capitis(P. h. capitis), resistance to permethrin wasfirst doi:10.1128/JCM.00808-12 reported in France in 1994 and throughout the world thereafter

July 2012 Volume 50 Number 7 Journal of Clinical168 Microbiology p. 2229–2233 jcm.asm.org 2229 Drali et al.

gle-step method to allow the monitoring of kdr in human body mutations, T917I (ACA�ATA) and L920F (CTT�TTT). In addition to lice. the two primer pairs, two hybridization probes, an anchor probe and a reporter probe, were designed by Tib Molbiol (DNA Synthesis Service, Berlin, Germany) to detect each of the three targeted mutations (see Table MATERIALS AND METHODS S1 in the supplemental material). These probes are alternatively labeled Body louse populations. Body lice were collected from volunteers at 2 withfluorescein and LC Red 640fluorochromes. The probes hybridize homeless shelters in Marseille before the beginning of a permethrin clin- head to tail to the region containing the mutation site to genotype, allow- ical trial. The research complied with all relevant federal guidelines and ing the transfer of thefluorescence energy offluorescein to that of Light institutional policies (ID RCB: 2010-A01406-33). Alternatively, homeless Cycler Red 640 (Fig. 1). The emittedfluorescence is measured continu- individuals were given new clothes, and their louse-infested clothing was ously during the melting phase, during which the temperature is gradually removed and brought to the laboratory for louse removal. A colony of increased, allowing the determination of the melting temperature (Tm) of in-house human body lice maintained on rabbits and never exposed to the reporter probe. The reactions were performed using a LightCycler 480 permethrin has been used as a wild-type control. Two head lice collected (Roche Diagnostics Corp.) in 96-well plates (LightCycler Multiwell Plate in France and Algeria and stored in our laboratory were also used as 96). In afinal volume of 20�l, 10�l 2� QuantiTect Probe PCR Master Downloaded from possibly resistant specimens. Mix (Qiagen), 0.5 mM (each) primer, 0.2 mM anchor probe, 0.2 mM Genomic DNA extraction. Prior to DNA extraction, the collected lice probe reporter, and between 5 and 20 ng DNA were mixed. Two different were immersed in 70% ethanol for 15 min and then rinsed with distilled programs were used for the detection of three mutations in this study (see water. Each louse was cut longitudinally into two parts. DNA extraction Table S2 in the supplemental material). was performed using a Qiagen tissue kit (Hilden, United Kingdom), ac- Data analysis. The sensitivity and specificity of the melting curve anal- cording to the manufacturer’s instructions. The extracted DNA was as- ysis genotyping method were tested using a contingency table represent- sessed for quantity and quality using a NanoDrop instrument (Thermo ing thefindings of the method test compared to the gold standard of Scientific, Wilmington, United Kingdom) before being stored at�20°C. sequencing. The statistical analyses were performed using SPSS for Win- PCR amplification and sequencing of the 908-bp fragment of the dows, version 17.2. http://jcm.asm.org/ VSSC gene. By analogy with head lice, Lee et al. (17) hypothesized that acquired resistance to permethrin in body lice was associated with the RESULTS same mutations. Consequently, we chose to use the already-reported spe- PCR amplification and sequencing of the 908-bp fragment of cific primers 5 HL-QS (5 -ATTTTGCGTTTGGGACTGCTGTT-3 ) and the VSSC gene. A total of 65 lice, including 52 body lice and 1 head 3 HL-QS (5 -CCATCTGGGAAGTTCTTTATCCA-3= = ) to amplify= the 908-bp= DNA= fragment of the VSSC gene (16). The Phusion= high-fidelity louse from homeless persons, 11 body lice from the laboratory DNA polymerase (Finnzymes, Thermo Scientific, Vantaa, Finland) was colony, and 1 head louse from a schoolchild in Algeria, were used used. The reactions were performed in afinal volume of 50�l with 1 U for this assay. The concentration of genomic DNA extracted from Phusion polymerase, 10�l 5� Phusion buffer, 0.5 mM (each) primer, the lice varied from 10 to 50 ng/�l. on July 5, 2012 by INIST-CNRS BiblioVie 0.16 mM deoxynucleoside triphosphates (dNTPs), and 30 to 50 ng of The 908-bp DNA fragment of the VSSC gene encoding the� DNA. The amplification consisted of 35 cycles (98°C for 5 s, 56°C for 30 s, subunit of the sodium channel was successfully amplified in all 65 and 72°C for 15 s), preceded by an initial phase at 98°C for 30 s and lice tested. Direct sequencing and multiple alignments of these followed by a termination phase at 72°C for 5 min. The PCRs were per- fragments allowed us to deduce the complete reference sequence. formed using the Mastercycler Thermocycler (Eppendorf, Hamburg, Four introns of various sizes (87, 86, 88, and 85 bp) and three Germany). After amplification, all of the PCR products were analyzed by exons (141, 174, and 163 bp) were found (see Fig. S1 in the sup- electrophoresis on 1.5% agarose gels using ethidium bromide staining. Bidirectional DNA sequencing of the targeted 908-bp PCR products plemental material). The genotyping results of the 3 mutation was performed using the 3130XL genetic analyzer (Applied Biosystems, sites in the 65 lice tested are summarized in Table 1, with several Courtaboeuf, France) with the BigDye Terminator v1.1 cycle (Applied different profiles identified. Of the specimens genotyped using the Biosystems). The electropherograms obtained for each sequence were an- Sanger sequencing method and designated RG (reference geno- alyzed using ChromasPro software. A multiple alignment was performed type), 31% exhibited a resistant haplotype (RRR), 29% showed an by the ClustalW method using the Basic Local Alignment Search Tool RHR haplotype, 22% showed an RSR haplotype, and 18% showed (BLASTn for nucleotide comparisons) available at http://blast.ncbi.nlm a wild-type haplotype (SSS). An alignment of the 65 obtained .nih.gov/. sequences indicated that all of the lice tested, which contained at Cloning of the 908-bp DNA fragments. As a high-quality source of least one of the 3 target mutations, also had a single nucleotide target DNA to optimize the test, we amplified DNA fragments from two polymorphism in the second intron, at approximately nucleotide body lice that had previously been sequenced and shown to be homozygous for the three mutations and cloned them into the pGEM-T Easy (nt) 2484� 4, A� T (data not shown). vector systems (Promega, Madison, WI) according to the manufacturer’s Melting curve analysis genotyping. The real-time PCR tech- recommendations. Thefirst louse exhibited a wild-type susceptible geno- nique wasfirst improved by testing the known genotype clones. type (SSS), and the second displayed a resistant genotype (RRR). To ob- Each of the three mutations was targeted individually. The results tain a heterozygous genotype (HHH), we mixed the two cloned DNA obtained with the melting curve analysis and sequencing were fragments in equal proportions (1:1). R, S, and H represented homozy- superimposable for the three mutation sites. As expected, the gous resistant, homozygous wild-type, and heterozygous genotypes, re- wild-type allele showed a peak with a higherT m than that of spectively. One clone per louse was chosen to conduct our experiments. the mutated allele because the detection probe complementary to The plasmids containing the cloned fragments were extracted using the the mutated allele was detached earlier in the presence of a mis- alkaline lysis method (3). match during the melting step. In addition, theT of the duplex Melting curve analysis genotyping. The melting curve analysis geno- m typing method is based on real-time PCR with hybridization probes, using probe/mutated codon is lower than that of the duplex probe/wild- fluorescent resonance energy transfer (FRET) technology. Two sets of type codon. A double peak is characteristic of heterozygotes (see primers were designed to amplify specific regions: (i) a 144-bp genomic Fig. S2 in the supplemental material). fragment of thefirst exon to characterize the M815I (ATG�ATT) muta- Thus, the G�T transversion responsible for the replacement of tion and (ii) a 203-bp fragment of the third exon to detect the 2 other a methionine with an isoleucine at position 815 and the C�T

2230 jcm.asm.org 169 Journal of Clinical Microbiology Knockdown Resistance in Body Louse Downloaded from http://jcm.asm.org/

FIG 1 Schematic diagram showing the hybridization of the FRET probes. For the detection of the kdr mutations, M815I (ATG�ATT) in theﬁrst exon and T917I (ACA�ATA) and L920F (CTT�TTT) in the third exon, 2 amplicons (a to c) spanning the region containing the polymorphic sites are ampliﬁed with 2 sets of

primers (primers Seq1 F [forward] and Seq1 A [reverse] [a] and primers Seq2 S [forward] and Seq2 A [reverse] [b and c]). on July 5, 2012 by INIST-CNRS BiblioVie

TABLE 1 Validation of the method of genotyping by comparing the transitions causing the replacements of a threonine with an iso- results of melting curve analysis (predicted genotype) and Sanger leucine at position 917 and a leucine with an phenylalanine at sequencing (reference genotype)a position 920 each shift theT m by 7 to 8°C. No. of lice in exon with mutation, To ensure that the results reflected the targeted fragments, we determined by method have verified the PCR products by sequencing after genotyping Third exon (data not shown). First exon, We then focused on the 65 lice for which the sequence of in- M815I, T917I, L920F, terest was previously determined using the Sanger sequencing ATG�ATT ACA�ATA CTT�TTT Louse type (no. method. The genotypes for each louse obtained by the melting of lice tested) Phenotype PG RG PG RG PG RG curve analysis genotyping, designated PG (predicted genotype), Body lice from S 0 0 13 13 1c 0 are reported in Table 1. homeless (52) H 0 0 18 19 0 0 Sensitivity and specificity of the real-time PCR with hybrid- R 52 52 21b 20 51 52 ization probe technique. Of the 195 tests performed, only two results were discordant with those obtained by the reference Head louse from S 0 0 0 0 0 0 method: a false positive for the T917I mutation and a false nega- homeless (1) H 0 0 1 1 0 0 R 1 1 0 0 1 1 tive for the L920F mutation (Table 1). The calculated sensitivity was 100% for the M815I and T917I mutations and 98% for the Human body S 11 11 11 11 11 11 L920F mutation; the specificity was 100% for M815I and L920F lice bred on H 0 0 0 0 0 0 and 96% for T917I. rabbits (11) R 0 0 0 0 0 0

Head louse from S 1 1 1 1 1 1 DISCUSSION Algeria (1) H 0 0 0 0 0 0 In contrast to head lice, for which the kdr frequency was some- R 0 0 0 0 0 0 times as high as 0.90, the permethrin resistance of the body lice a Abbreviations: PG, predicted genotype obtained by melting curve analysis genotyping was still unknown. Consequently, to promote the eradication of method; RG, reference genotype obtained by the Sanger sequencing method; R, body lice in the homeless population, we have planned a clinical resistant homozygote in which both alleles are mutated; S, susceptible homozygote in study using permethrin-impregnated underwear. Monitoring the which neither allele is mutated; H, heterozygote in which only one of the two alleles is mutated. resistance of body lice to permethrin in the homeless population b False-positive result. before and during the clinical trial was determined to be manda- c False-negative result. tory. To achieve this goal, we opted to use a robust and simple

July 2012 Volume 50 Number 7 170 jcm.asm.org 2231 Drali et al. single-step melting curve analysis genotyping method. Since its tive, and specific tool that is fully compatible with the analysis of a establishment in 1997 (19), this method has been particularly use- large number of samples in epidemiological surveys, allowing the ful in detecting known mutations causing human diseases such as simultaneous genotyping of 96 samples in just over an hour (75 min). cancer (27) and other disorders (30). The melting curve analysis Therefore, this method is perfectly suited to the epidemiological genotyping technique is also used in otherfields, such as molecu- monitoring of permethrin resistance in large-scale clinical studies. lar haplotyping (21), bacteriology (22), parasitology (26), and vi- ACKNOWLEDGMENTS rology (25). Using this method allowed us to determine the genotype of human lice at each mutation site. Furthermore, its ability This study was funded in part by PHRC 2010 from the French Ministry of to detect heterozygotes makes it particularly useful for following Health. the dynamics of the kdr mutation in a targeted population of lice. The text has been edited by American Journal Experts under certificate verification key 7DCD-BB1D-3851-ACC7-FB0D. Despite the two inaccurate results, a false positive and a false neg- We gratefully thank Didier Raoult from URMITE Marseille for sug- ative for the T917I and L920F mutations, respectively, in our total

gestions and help with this study and Amina Boutellis, Jean Christophe Downloaded from of 195 tests, the melting curve analysis genotyping method proved Lagier, Elisabeth Botelho-Nevers, Mathieu Million, Djamel Thiberville, to have great sensitivity (98% to 100%) and excellent specificity Nadim Cassir, and Aurélie Veracx for their assistance. (96% to 100%) in the detection and characterization of allelic REFERENCES resistance to permethrin in human lice. The false-positive result could be due to differences in the strength of probe binding (2) or 1. Badiaga S, et al. 2008. The effect of a single dose of oral ivermectin on pruritus in the homeless. J. Antimicrob. Chemother. 62:404–409. polymerization errors during the insertion of nucleotides by the 2. Bass C, et al. 2007. Detection of knockdown resistance (kdr) mutations in Taq polymerase during real-time PCR. The resequencing of this Anopheles gambiae: a comparison of two new high-throughput assays sample confirmed its initial genotypic status, which was heterozy- with existing methods. Malar. J.6:111. gous. Conversely, the false-negative result could have been due to 3. Birnboim HC, Doly J. 1979. A rapid alkaline extraction procedure for http://jcm.asm.org/ screening recombinant plasmid DNA. Nucleic Acids Res.7:1513–1523. a manipulation error during the distribution of the DNA samples 4. Brouqui P, et al. 2005. Ectoparasitism and vector-borne diseases in 930 or the reagent mixtures into the 96-well plates. Several tools de- homeless people from Marseilles. Medicine (Baltimore) 84:61–68. scribed in the literature are available to monitor both phenotypic 5. Chosidow O. 2000. Scabies and pediculosis. Lancet 355:819–826. and genotypic resistance traits in head lice. In addition to being 6. Clark JM. 2009. Determination, mechanism and monitoring of knock- faster than conventional phenotypic bioassays, which are tedious down resistance in permethrin-resistant human head lice, Pediculus humanus capitis. J. Asia Pac. Entomol. 12:1–7. and not adapted to such large-scale purposes (18), these tech- 7. Clark JM. 2010. Permethrin resistance due to knockdown gene mutations niques can simultaneously detect the allele frequencies and point is prevalent in human head louse populations. Open Dermatol. J.4:63–68. mutations associated with resistance to permethrin without gene 8. Davies TG, Field LM, Usherwood PN, Williamson MS. 2007. DDT, pyre- on July 5, 2012 by INIST-CNRS BiblioVie sequencing (14). thrins, pyrethroids and insect sodium channels. IUBMB Life 59:151–162. 9. Dong K. 2007. Insect sodium channels and insecticide resistance. Invert. The eradication of body lice in the homeless is a major chal- Neurosci.7:17–30. lenge. We implemented several different clinical trials during the 10. Foucault C, et al. 2006. Oral ivermectin in the treatment of body lice. J. past 10 years tofight louse infestations without success ( 4). Al- Infect. Dis. 193:474–476. though resistance to DDT was reported at the end of the 1940s and 11. Hemingway J, Ranson H. 2000. Insecticide resistance in insect vectors of in the early 1950s in the United States, southeast Asia, Egypt, and human disease. Annu. Rev. Entomol. 45:371–391. 12. Hodgdon HE, et al. 2010. Determination of knockdown resistance allele Iran (13, 20), to the best of our knowledge, no data are available frequencies in global human head louse populations using the serial inva- for permethrin resistance in P. h. corporis. In this work, we noted a sive signal amplification reaction. Pest Manag. Sci. 66:1031–1040. high prevalence of the 3 indicated mutations in the body lice col- 13. Hurlbut HS, Peffly RL, Salah AA. 1954. DDT resistance in Egyptian body lected from homeless people (100% for the M815I and L920F lice. Am. J. Trop. Med. Hyg.3:922–929. 14. Kim HJ, Symington SB, Lee SH, Clark JM. 2004. Serial invasive signal mutations and 56.73% for the T917I mutation). As suggested for amplification reaction for genotyping permethrin-resistant (kdr-like) hu- head lice, under the selective pressure of permethrin, the muta- man head lice, Pediculus capitis. Pestic. Biochem. Physiol. 80:173–182. tions M815I and L920F are thefirst to take place. This event would 15. Kristensen M. 2005. Identification of sodium channel mutations in hu- be followed by the onset of the T917I mutation (12), which is man head louse (Anoplura: Pediculidae) from Denmark. J. Med. Ento- considered the main cause of this resistance, while the M815I and mol. 42:826–829. 16. Kwon DH, Yoon KS, Strycharz JP, Clark JM, Lee SH. 2008. Determi- L920F mutations would reduce permethrin sensitivity (29). These nation of permethrin resistance allele frequency of human head louse findings suggest that the population of body lice infesting home- populations by quantitative sequencing. J. Med. Entomol. 45:912–920. less people in Marseille has been subjected to selective pressure. 17. Lee SH, et al. 2003. Sodium channel mutations associated with knock- Overuse of synthetic pyrethroids to eradicate the various insect down resistance in the human head louse, Pediculus capitis (De Geer). Pestic. Biochem. Physiol. 75:79–91. pests, notably the increasing infestation with bedbugs in the last 4 18. Lee SH, et al. 2000. Molecular analysis of kdr-like resistance in perme- decades, has been well documented and probably accounts for the thrin-resistant strains of head lice, Pediculus capitis. Pestic. Biochem. level of resistance reported here. Physiol. 66:130–143. The presence of a single nucleotide polymorphism (SNP), at 19. Lyon E, Wittwer CT. 2009. LightCycler technology in molecular diag- approximately nt 2484� 4, A�T, at the beginning of the second nostics. J. Mol. Diagn. 11:93–101. 20. McLintock J, Zeini A, Djanbakhsh B. 1958. Development of insecticide intron of the 908-bp target sequence of the VSSC gene, was an resistance in body lice in villages of North-Eastern Iran. Bull. World unexpectedfinding. This polymorphism was found exclusively in Health Organ. 18:678–680. those lice that displayed at least one of the 3 tested mutations and 21. Motovska Z, et al. 2010. Platelet gene polymorphisms and risk of bleeding could represent an informative SNP for the presence or absence of in patients undergoing elective coronary angiography: a genetic substudy of the PRAGUE-8 trial. Atherosclerosis 212:548–552. these mutations; however, further investigations are required to 22. Randegger CC, Hachler H. 2001. Real-time PCR and melting curve anal- address this possibility. ysis for reliable and rapid detection of SHV extended-spectrum beta- In conclusion, melting curve analysis genotyping is a fast, sensi- lactamases. Antimicrob. Agents Chemother. 45:1730–1736.

2232 jcm.asm.org 171 Journal of Clinical Microbiology Knockdown Resistance in Body Louse

23. Raoult D, Foucault C, Brouqui P. 2001. Infections in the homeless. 28. Sholdt LL, Rogers EJ, Jr, Gerberg EJ, Schreck CE. 1989. Effectiveness of Lancet Infect. Dis.1:77–84. permethrin-treated military uniform fabric against human body lice. Mil. 24. Raoult D, Roux V. 1999. The body louse as a vector of reemerging human Med. 154:90–93. diseases. Clin. Infect. Dis. 29:888–911. 29. SupYoon KS, Symington SB, Lee SH, Soderlund DM, Clark JM. 2008. 25. Ratcliff RM, Chang G, Kok T, Sloots TP. 2007. Molecular diagnosis of Three mutations identified in the voltage-sensitive sodium channel alpha medical viruses. Curr. Issues Mol. Biol.9:87–102. subunit gene of permethrin-resistant human head lice reduce the perme- 26. Safeukui I, et al. 2008. Evaluation of FRET real-time PCR assay for rapid thrin sensitivity of houdefly Vssc1 sodium channels expressed in Xenopus detection and differentiation of Plasmodium species in returning travel- oocytes. Insect. Biochem. Mol. Biol. 38:296–308. lers and migrants. Malar. J.7:70. 30. Wee L, Vefring H, Jonsson G, Jugessur A, Lie RT. 2010. Rapid geno- 27. Schnittger S, et al. 2006. KIT-D816 mutations in AML1-ETO-positive typing of the human renin (REN) gene by the LightCycler instrument: AML are associated with impaired event-free and overall survival. Blood identification of unexpected nucleotide substitutions within the selected 107:1791–1799. hybridization probe area. Dis. Markers 29:243–249. Downloaded from http://jcm.asm.org/ on July 5, 2012 by INIST-CNRS BiblioVie

July 2012 Volume 50 Number 7 172 jcm.asm.org 2233

5’ HL-QS

1st exon 2nd exon 3rd exon Intron 1 Intron 2 Intron 3 Intron 4 (a) 141 bp 174 bp 163 bp 87 bp 86 bp 88 bp 85 bp 3’ HL-QS

815 917 920 H D M D K L T F V L C CAC GAC ATG GAT AAA TTA ACA TTC GTC CTT TGC Wild type + + I + + + I + + F + *** *** **T *** *** *** *T* *** *** T** *** Mutated

(b)

Supplementary figure 1. (a) A diagram of the 908 bp fragment of the VSSC gene encoding the α-subunit of a sodium channel in P. humanus corporis, containing the three mutations, M815I (ATG / ATT), T917I (ACA / ATA) and L920F (CTT / TTT). (b) Electropherograms of the wild- type sequence in the region of interest obtained from a body louse that was raised on a rabbit in the laboratory.

173

(a) (c)

Supplementary figure 2. Genotyping of the mutation site (b) (T917I) using melting curve analysis. (a) Melting curves: in Wild Type green, the homozygous mutant (TT); in blue, wild-type (CC); Homozygote CC TT in red, the compound heterozygote (CT); and in azure the Heterozygote negative control. (b) Derivative melting peaks obtained from CT the melting curves. (c) Electropherograms of the (T917I) mutation site in 3 samples. Top, the wild-type allele bearing a

C., Middle, heterozygote with two alleles, one wild-type and Negative control one mutated (CT). Bottom, the mutated allele showed a T at the same position.

174 Supplementary Table 1. DNA sequences of the primers and probes used to detect kdr mutations in body lice.

Exon Primer or Sequence and a.a. probe positiona Seq1 F (forward) ACCCATTCGTCGAATTATTCATAACT exon 1 Seq1 A (reverse) CCCCGCATTAAAATTAAATTTTTAC

Seq1 mut TCTGTCCATGTCTTTATCCATGTC-FLb M815I

Anchor Seq1 LC-640-TGATGATCCAAAGCCATAAATAGTGTGTT-P Seq2 S (forward) TTTTTTTCTTTTTATGACGAAAC Seq2 A (reverse) CCCGTGTAATTTTTTCCA exon 3

Anchor 1 AATTTCAATTATGGGTCGAACTGTTGG-FL T917I Sensor 1 C LC-640-GCTTTGGGTAATTTAACATTCGTC-P and L920F Sensor 2 C ATTCGTCCTTTGCATTATCATATTCAT-FL Anchor 2 LC-640-TTTGCCGTTATGGGAATGCAACTT-P

LC-Red 640 and fluorescein are fluorophores. The 3’ end of the one of the two probes in each pair was phosphorylated to prevent probe elongation by Taq polymerase during the PCR.

Bold, codons; underlined, mutated bases. a The amino acids are numbered is according to GenBank accession no. AY191155 b FL, fluorescein; LC-Red 640, LightCycler-Red 640; P, phosphorylated.

175 Supplementary Table 2. Programs for real-time PCR to detect kdr mutations using the Light

Cycler apparatus 480.

LightCycler© 480 Program Mutations Primers Size Denaturation Amplification Melting 95 °C; 01 sec 95 °C; 1 sec M815I Seq1F/Seq1A 144 bp 60 °C; 10 sec 40 °C; 30 sec 72 °C; 06 sec 80 °C; 95 °C; 15 min T917I 95 °C; 01 sec continuous and or Seq2S/Seq2A 203 bp 51 °C; 15 sec transition rate: L920F 72 °C; 10 sec 0.1 °C/sec

176

Article IX: Effect of Permethrin-Impregnated Underwear on Body Lice in Sheltered Homeless Persons: a Randomized Controlled Trial

JAMA Dermatology 150:273-279.

177

Research

Original Investigation Effect of Permethrin–Impregnated Underwear on Body Lice in Sheltered Homeless Persons A Randomized Controlled Trial

Samir Benkouiten, MPH; Rezak Drali, MSc; Sékéné Badiaga, MD, PhD; Aurélie Veracx, PhD; Roch Giorgi, MD, PhD; Didier Raoult, MD, PhD; Philippe Brouqui, MD, PhD

Supplemental content at IMPORTANCE The control of body lice in homeless persons remains a challenge. jamadermatology.com

OBJECTIVE To determine whether the use of long-lasting insecticide–treated underwear provides effective long-term protection against body lice in homeless persons.

DESIGN, SETTING, AND PARTICIPANTS A randomized, double-blind, placebo-controlled trial was conducted in February and December 2011 in 2 homeless shelters (Madrague Ville and Forbin) in Marseille, France. Of the 125 homeless persons screened for eligibility, 73 body lice–infested homeless persons, 18 years or older, were enrolled.

INTERVENTIONS Body lice–infested homeless persons were randomly assigned to receive 0.4% permethrin–impregnated underwear or an identical-appearing placebo for 45 days, in a 1:1 ratio, with a permuted block size of 10. Visits were scheduled at days 14 and 45. Data regarding the presence or absence of live body lice were collected.

MAIN OUTCOMES AND MEASURES The primary and secondary end points were the proportions of homeless persons free of body lice on days 14 and 45, respectively. Mutations associated with permethrin resistance in the body lice were also identified.

RESULTS Significantly more homeless persons receiving permethrin-impregnated underwear than homeless persons receiving the placebo were free of body lice on day 14 in the intent-to-treat population (28% vs 9%;P = .04), with a between-group difference of 18.4 percentage points (95% CI, 1.4-35.4), and in the per-protocol population (34% vs 11%; P = .03), with a between-group difference of 23.7 percentage points (95% CI, 3.6-43.7). This difference was not sustained on day 45. At baseline, the prevalence of the permethrin- resistant haplotype was 51% in the permethrin group and 44% in the placebo group. On day 45, the permethrin-resistant haplotype was significantly more frequent in the permethrin group than in the placebo group (73% vs 45%,P < .001).

CONCLUSION AND RELEVANCE Permethrin–impregnated underwear is more efficient than Author Affiliations: Unité de placebo at eliminating body louse infestations by day 14; however, this difference was not Recherche sur les Maladies sustained on day 45. The use of permethrin may have increased the resistance to permethrin Infectieuses et Tropicales in body lice and thus must be avoided. Émergentes, Aix-Marseille Université, Marseille, France (Benkouiten, Drali, Badiaga, Veracx, Raoult, Brouqui); TRIAL REGISTRATION clinicaltrials.gov Identifier: NCT01287663 Institut Hospitalo-Universitaire Méditerranée Infection, Marseille, France (Benkouiten, Drali, Veracx, Giorgi, Raoult, Brouqui); Service des Entérobactéries et Hygiène de l'Environnement, Institut Pasteur d'Algérie, Alger, Algérie (Drali); Institut National de la Santé et de la Recherche Médicale, Aix-Marseille Université, Marseille, France (Giorgi). Corresponding Author: Philippe Brouqui, MD, PhD, Aix Marseille Université, URMITE, 27 bd Jean JAMA Dermatol. doi:10.1001/jamadermatol.2013.6398 Moulin, 13005 Marseille, France Published online December 4, 2013. ([email protected]).

Downloaded From: http://archderm.jamanetwork.com/ by American Medical Association, June Robinson on 12/30/2013 Research Original Investigation Permethrin–Impregnated Underwear

omelessness is a major social and public health prob- S.A., Paris, France), was prepared as a 1:20 emulsion in lem worldwide. The prevalence of body lice in shel- water as recommended by the manufacturer. The impregna- H tered homeless persons varies from 7% to 22%.1 Body tion was performed by an independent person in the Public lice are known vectors of Bartonella quintana, Rickettsia prowa- Hospitals of Marseille laundry. Sets of underwear (T-shirt, zekii, and Borrelia recurrentis, which cause trench fever, epi- underpants, and socks) were placed into the emulsion for 15 demic typhus, and relapsing fever, respectively.2 Conse- minutes, completely saturated, removed, and allowed to quently, all measures that can be used to decrease the burden dry. Once dry, the underwear was odorless. According to the of body lice infestation in homeless persons, and more gen- manufacturer’s instructions, the permethrin–impregnated erally in persons living in crowded and unhygienic environ- underwear is effective up to 6 months and even after 6 ments, are warranted to avoid the spread and/or outbreak of washes. Other sets of underwear were treated identically these diseases. but without the permethrin formulation. The permethrin- Pediculus humanus humanus, the human body louse, is a impregnated and placebo underwear were identical in host-specific hematophagous ectoparasite that lives in the appearance but were labeled discreetly and then stored in 2 clothes. Body lice are extremely contagious and can be spread separate boxes until use. through body contact, shared clothing, or shared bedding in overcrowded conditions.3 The classic therapeutic measures for Participants body lice infestations are the frequent changing or washing of To recruit a sufficient number of participants, 2 independent the infected person’s clothes and blankets at 50°C and the fre- study cohorts of homeless persons were performed. In study quent treatment of bedding with insecticides.4 However, in our A, homeless persons were recruited in February 2011 from 2 experience with the sheltered homeless persons in Marseille, shelters (Madrague Ville and Forbin) in the city of Marseille. France, these measures have had little success.5 Oral ivermec- In study B, different homeless persons were recruited in De- tin reduces the prevalence of body lice infestations and pru- cember 2011 from the same 2 shelters. Each facility provides ritus in homeless persons, but the effect is transient.6,7 These nighttime shelter for a mean of 300 homeless persons who stay findings suggest that the complete eradication of this ecto- in the shelter overnight and leave it in the morning. Home- parasite in homeless persons remains a challenge.8 less persons with a self-reported diagnosis of pruritus and/or The pyrethroids are the major commercially available pe- with body lice were screened. Homeless persons were eli- diculicides. All World Health Organization–recommended in- gible for inclusion in the study if they were 18 years or older, secticide-treated mosquito nets are pyrethroid based.9 The im- were able to provide consent, declared that they slept at least pregnation of clothing with a pyrethroid emulsion has been 3 nights per week in 1 of the 2 shelters, and had at least 1 live reported to eradicate body lice after a single application to mili- body louse recovered on examination. The exclusion criteria tary uniforms, even after 20 washes,10 and may provide long- were the presence of cutaneous superinfection or intrave- lasting protection. The clinical safety and effectiveness of topi- nous drug use. cal permethrin in humans have been reported previously.11,12 We conducted our randomized, double-blind, placebo- Randomization and Interventions controlled study to determine whether the use of long- Homeless persons were randomly assigned to the interven- lasting permethrin–treated underwear provides long-term pro- tion group with sealed, opaque envelopes in a 1:1 ratio with a tection against louse proliferation in sheltered homeless permuted block size of 10. Participants and investigators were persons. Our secondary aim was to assess the mutations as- unaware of the treatment assignments throughout the study. sociated with permethrin resistance in the body lice. Visits were scheduled on days 14 and 45. Data on the presence or absence of live body lice, whether the clothes had been changed between visits, and the occurrence of adverse events Methods were collected. All participants received their protocol underwear on day Study Design 1 (baseline) and at each follow-up visit (on days 14 and 45) un- Our study was a double-center, double-blind, randomized, pla- der the supervision of the investigators. The underwear could cebo-controlled intervention trial. Homeless persons were also be changed between the follow-up visits in the shelters given underwear treated with permethrin or an identical- at the request of the individual (with respect to the assigned appearing placebo for 45 days. The protocol was approved by group). The used underwear was collected by the same ento- our institutional review board (January 24, 2011; reference mologist for detailed visual inspection. This evaluator was 2010-A01406-33), and the study was performed in accor- trained in the technique for detecting and counting live body dance with the good clinical practices recommended by the lice from the infested underwear. Dead and living lice were dif- Declaration of Helsinki and its amendments. All participants ferentiated; lice were considered to be dead if they were not provided written informed consent. This study is registered moving. Homeless persons were excluded from further study with clinicaltrials.gov (identifier NCT01287663). if they had any manifestations suggesting adverse effects, and specific treatment was given as needed. The final visit on day Underwear Preparation 45 was regarded as the end of the study for every participant. An 8% (8-g/L) permethrin formulation for impregnation, which If persistent live body lice were found at this visit, homeless is commercially available under the label Barrage Insect (S.P.C.I. persons were offered a single dose of oral ivermectin (12 mg).6

E2 JAMA Dermatology Published online December 4, 2013 jamadermatology.com

Downloaded From: http://archderm.jamanetwork.com/ by American Medical Association, June Robinson on 12/30/2013 Permethrin–Impregnated Underwear Original Investigation Research

Figure. Study Flow of Participants

125 Homeless persons assessed for eligibility

52 Excluded 52 Not infested with body lice

73 Randomized

40 Randomized to permethrin 33 Randomized to placebo

8 Lost to follow-up at day 14 5 Lost to follow-up at day 14

Included in analysis of primary end point Included in analysis of primary end point 40 Included in intent-to-treat population 33 Included in intent-to-treat population 27 Included in per-protocol population 28 Included in per-protocol population

13 Lost to follow-up at day 45 9 Lost to follow-up at day 45

Screening and inclusion process for Included in analysis of secondary end point Included in analysis of secondary end point participants of the randomized 40 Included in intent-to-treat population 33 Included in intent-to-treat population controlled trial and flow of 27 Included in per-protocol population 24 Included in per-protocol population participants through each stage of the study.

Outcome Measures and Safety End Points points between the permethrin and placebo groups when cal- The primary efficacy end point was the proportion of home- culating the proportion of homeless persons free of body lice less persons free of body lice (defined as absence of living body on day 14 with a 2-sidedα = .05, assuming an anticipated ef- lice in the underwear) 14 days after treatment. The secondary fect between 20 and 40 percentage points in the placebo group. efficacy end point was the same assessment 45 days after treat- In our experience, approximately 70% of pruritus symptoms ment. End points were assessed on the basis of the exhaus- are due to body lice infestations, and presuming that the rate tive examination of live body lice in all collected underwear. of individuals lost to follow-up could be up to 30%, we pre- The pruritus that normally accompanies body lice infes- dicted that we would need to screen 122 homeless persons. tation may be exacerbated temporarily after dermal expo- Analyses were conducted in accordance with the intent-to- sure to permethrin.13 Physical examinations were performed treat and per-protocol principles. In the intent-to-treat analy- at scheduled visits, and adverse events were recorded during sis, only the homeless persons who were present at the sched- the 45-day study period. The prevalence and severity of pru- uled follow-up visits were included. For the intent-to-treat ritus, graded from 0 to 3 (0, none; 1, mild; 2, moderate; and 3, analysis, loss to follow-up was considered a treatment fail- severe), were assessed at each visit. ure. The Pearsonχ 2 test and Fisher exact test, as appropriate, Another objective was to investigate the evolution of per- were applied to analyze the primary and secondary end points methrin resistance in the body lice. The permethrin resis- of efficacy, and 95% CIs for the difference between the suc- tance of body lice was determined in a representative ran- cess rates in the study groups were calculated. Thet test for dom sample of body lice collected from all body lice–infested independent groups and Mann-Whitney test, as appropriate, homeless persons and stratified by the level of infestation of were used to investigate the safety end point of mean pruri- the homeless persons (see eTable 1 in the Supplement). A melt- tus score and the continuous variables.P≤ .05 (2-tailed test) ing curve analysis genotyping method,14 based on a previ- was established as the level of significance for all tests. Sta- ously reported real-time polymerase chain reaction using hy- tistical analyses were performed using SPSS statistical soft- bridization probes, was used to detect the 3 mutations (M815I, ware, version 17.2 (SPSS Inc). T917I, and L920F), identified in the voltage-sensitive sodium channelα-subunit gene, responsible for knockdown resistance (kdr). According to the literature, these 3 mutations de- Results fine the RRR haplotype, which confers permethrin resistance in head lice.15,16 Participants The trial profile is summarized in the Figure. Of the 125 Statistical Analysis homeless persons screened for eligibility in February and We estimated that approximately 60 body lice–infested home- December 2011, 73 (58%) were eligible on the basis of the less persons (30 in each group) would need to be enrolled to presence of live body lice (40 in the permethrin group and provide 90% power to detect a difference of 40 percentage 33 in the placebo group) and were consequently randomized

jamadermatology.com JAMA Dermatology Published online December 4, 2013 E3

Downloaded From: http://archderm.jamanetwork.com/ by American Medical Association, June Robinson on 12/30/2013 Research Original Investigation Permethrin–Impregnated Underwear

Table 1. Demographic and Baseline Characteristics of the Study Groups

No. (%) of Study Participantsa Permethrin Placebo Characteristic (n = 40) (n = 33) P Value Age, mean (SD), y 56.4 (14) 57.62 (12) .69 Men 37 (92) 33 (100) .24 Madrague Ville Shelter 36 (90) 31 (94) .68 Marginal homelessb 20 (50) 20 (61) .36 Homeless with >50 lice 18 (45) 19 (58) .28 a Data are presented as number (percentage) of study participants Duration of homelessness≤24 mo 15 (38) 19 (58) .08 unless otherwise indicated. Pruritus 40 (100) 33 (100) b Classified by shelter staff.

Table 2. Effect of Treatment on Days 14 and 45 in Study A and Study B Combined

No./Total (%) Outcome Measure Permethrin Placebo Difference, % (95% CI)P Value Intent-to-treat population Body lice–free homeless after 14 d 11/40 (28) 3/33 (9) 18.4 (1.4 to 35.4) .04 Body lice–free homeless after 45 d 11/40 (28) 9/33 (27) 0.2 (−20.3 to 20.8) .98 Per protocol Body lice–free homeless after 14 d 11/32 (34) 3/28 (11) 23.7 (3.6 to 43.7) .03 Body lice–free homeless after 45 d 11/27 (41) 9/24 (38) 3.2 (−23.6 to 30.0) .81

into the control and treatment groups (Figure). They were 45 (mean [SD], 148.4 [427.6] in the permethrin group vs 147.6 predominantly male (96%), were mostly from the Madrague [272.3] in the placebo group;P = .68) (data not shown). Ville shelter (92%), and had a mean (SD) age of 56.9 (13.3) years (age range, 20-79 years). Approximately 45% reported Adverse Events being homeless for less than or equal to 24 months. Baseline No adverse events were reported in any treated homeless per- characteristics were similar between the treatment groups sons. The prevalence of pruritus was reduced in both groups, (Table 1). with no significant differences in the proportion of homeless persons free of pruritus between the permethrin group and the Primary and Secondary Outcomes placebo group on day 14 (8 of 32 [25%] vs 6 of 27 [22%];P = .80), In the intent-to-treat population, 11 of 40 homeless persons with an odds ratio of 1.16 (95% CI, 0.34–3.91), or on day 45 (8 (28%) were free of live body lice on day 14 (primary end point) of 27 [30%] vs 8 of 24 [33%];P = .77), with an odds ratio of 0.84 in the permethrin group compared with 3 of 33 (9%) in the pla- (95% CI, 0.25-2.75) in the per-protocol population. The mean cebo group (P = .04), with a between-group difference of 18.4 (SD) pruritus score at baseline was 2.53 (0.69) in the perme- percentage points (95% CI, 1.4-35.4) (Table 2). This propor- thrin group and 2.24 (0.90) in the placebo group. No signifi- tion was also significantly greater in the permethrin group than cant differences were found in the mean reduction in pruri- in the placebo group in the per-protocol population (34% vs tus score from baseline between the permethrin group and the 11%;P = .03), with a between-group difference of 23.7 percent- placebo group on day 14 (–0.68 vs –0.28; 95% CI, –0.97 to 0.17; age points (95% CI, 3.6-43.7). P = .17) and on day 45 (–0.92 vs –0.45; 95% CI, –1.15 to 0.21; With respect to the secondary efficacy end point, in the P = .17). intent-to-treat population, 11 of 40 homeless persons (28%) were free of live body lice on day 45 in the permethrin group Permethrin Resistance of Body Lice compared with 9 of 33 (27%) in the placebo group (28% vs 27%; Of the 34 035 live body lice that were collected, 371 were used P = .98), with a between-group difference of 0.2 percentage to assess permethrin resistance because at least 1 louse per in- points (95% CI, –20.3 to 20.8). In addition, no significant dif- fested homeless persons was selected (see eTable 1 in the ference was found between the 2 proportions in the per- Supplement): 187 were collected in study A (91 on day 1 [44 protocol population (41% vs 38%;P = .81), with a between- from 18 homeless persons from the permethrin group and 47 group difference of 3.2 percentage points (95% CI, –23.6 to 30.0) from 17 homeless persons from the placebo group] and 96 on (Table 2). day 45 [43 from 8 homeless persons from the permethrin group Significant reductions from the baseline in the mean num- and 53 from 7 homeless persons from the placebo group]) and ber of body lice were observed on day 14 and day 45 in the per- 184 in study B (67 on day 1 [32 from 16 homeless persons from methrin and placebo groups (see eTable 2 in the Supple- the permethrin group and 35 from 14 homeless persons from ment). However, no significant difference was found between the placebo group] and 117 on day 45 [56 from 8 homeless per- the 2 groups on day 14 (mean [SD], 176.1 in the permethrin sons from the permethrin group and 61 from 8 homeless per- group vs 104 [202.1] in the placebo group;P = .18) or on day sons from the placebo group]).

E4 JAMA Dermatology Published online December 4, 2013 jamadermatology.com

Downloaded From: http://archderm.jamanetwork.com/ by American Medical Association, June Robinson on 12/30/2013 Permethrin–Impregnated Underwear Original Investigation Research

Table 3. Evolution of the Permethrin-Resistant Haplotype in Body Lice During the Survey Period

Permethrin Placebo Haplotype Day 1 Day 45P Value Day 1 Day 45P Value Study A RRR, No./total (%)a 17/44 (39) 31/43 (72) 18/47 (38) 17/53 (32) .002 .51 RRR, % (95% CI)a,b 41.9 (40.4-43.4) 67.4 (66.4-68.4) 21.1 (20.8-21.4) 26.2 (25.3-27.0) Study B RRR, No./total (%)a 22/32 (69) 41/56 (73) 18/35 (51) 34/61 (56) .65 .68 RRR, % (95% CI)a,b 77.8 (76.1-79.6) 75.0 (74.1-76.0) 56.9 (55.1-58.6) 59.9 (58.8-61.0) Studies A and B RRR, No./total (%)a 39/76 (51) 72/99 (73) 36/82 (44) 51/114 (45) .004 .90 RRR, % (95% CI)a,b 62.8 (61.9-63.7) 71.8 (71.3-72.3) 34.3 (33.8-34.9) 44.8 (44.3-45.4)

a Considered permethrin resistant. b Inference based on the subrandom sample of body lice stratified based on the initial infestation level of the individual.

Table 4. Evolution of the Allele Frequency of the T917I and L920F Mutations in Body Lice During the Survey Period

No./Total (%) Permethrin Placebo Mutation Day 1 Day 45P Value Day 1 Day 45P Value Study A T917I 18/44 (41) 31/44 (72) .003 21/47 (45) 17/53 (32) .19 L920F 43/44 (98) 43/43 (100) .99 37/47 (79) 53/53 (100) <.001 Study B T917I 22/32 (69) 45/56 (80) .21 18/35 (51) 34/61 (56) .68 L920F 32/32 (100) 56/56 (100) >.99 34/35 (97) 61/61 (100) .77 Studies A and B T917I 40/76 (53) 76/99 (77) .008 39/82 (48) 51/114 (45) .69 L920F 75/76 (99) 99/99 (100) .89 71/82 (87) 114/114 (100) <.001

At baseline, the prevalence of the permethrin-resistant day 45 and is accompanied by increasing permethrin resis- haplotype (Table 3) among the lice collected was 51% in the per- tance in body lice collected from homeless persons. methrin group and 44% in the placebo group. On day 45, the Resistancetopermethrinhasbeenreportedintheheadlouse permethrin-resistant haplotype was significantly more fre- Pediculus humanus capitis in many parts of the world.17-20 quent in the permethrin group than in the placebo group (73% This is the first trial, to our knowledge, that tests permethrin– vs 45%;P < .001). impregnated underwear on body lice in sheltered homeless At baseline, the prevalence of permethrin-resistant mu- persons, combined with molecular detection of mutations asso- tations (Table 4) was established as 99% for the L920F muta- ciated with permethrin resistance in body lice. The kdr allele fre- tion and 53% for the T917I mutation in the permethrin group quencyinthebodylicepopulationatbaselinewasunexpectedly and 87% for the L920F mutation and 48% for the T917I muta- high in study A (averaging 38%). The relative ease with which tion in the placebo group. The prevalence of the T917I muta- this group of mutations was identified within the modern tion increased significantly from baseline to day 45 in the per- body louse population may be related to prior exposure to methrin group (from 53% to 77%;P = .008) but remained stable dichlorodiphenyltrichloroethane, which likely involved in the placebo group (from 48% to 45%;P = .69). The preva- kdr-like mechanisms in some cases, in a manner similar to the lence of the L920F mutation increased significantly in the pla- rapid increase of permethrin resistance in the head louse cebo group from baseline to day 45 (from 87% to 100%; populations shortly after the introduction of synthetic pyre- P < .001). The prevalence of the M815I mutation was estab- throids in Europe and worldwide.17-20 Resistance of the body lished as 100% for all samples. louse to dichlorodiphenyltrichloroethane was reported at the end of the 1940s and early 1950s.21-23 This finding could suggest that, although no resistance to permethrin was reported Discussion in the body lice in Western Europe before this study, some lice are currently resistant to pyrethroids through dormant cross- In this randomized controlled trial, long-lasting permethrin– resistance attributed to the kdr mechanism. treated underwear is more efficient in the elimination of body In our study, the increase of the prevalence of the T917I louse infestations than placebo in the short term (ie, on day mutation interestingly coincided with the loss of perme- 14), but the difference with the placebo was not sustained by thrin efficacy. This result is coherent with that obtained in a

jamadermatology.com JAMA Dermatology Published online December 4, 2013 E5

Downloaded From: http://archderm.jamanetwork.com/ by American Medical Association, June Robinson on 12/30/2013 Research Original Investigation Permethrin–Impregnated Underwear

previous study16 that concluded that the T917I mutation Second, recent studies25 suggest that head and body lice alone was responsible for most of the target site insensitiv- can be mixed in persons infested with both head and body lice. ity reported in the resistant RRR haplotype. In addition, a Whether these lice are conspecific remains controversial,25-30 recent study24 reported a prevalence of 5% of the T917I and although head and body lice do not interbreed in the mutation, against 70% for both the M815I and L920F muta- wild,31 fertile hybrids that have intermediate morphologic tions, in an Egyptian head louse population for which selec- characteristics32 have been reported under laboratory tion with pyrethroid-based pediculicides was expected to conditions.33,34 Moreover, several observational studies35-38 be low. These findings suggested that head lice may have have also suggested that head lice could become body lice when acquired the M815I and L920F mutations first and then, raised under the laboratory conditions. once these 2 mutations are present, rapidly acquired the Third, it appears that only a change of underwear could re- T917I mutation, high levels of nerve insensitivity, and resis- duce body lice infestation (equivalent efficiency of placebo at tance after they were again placed under pyrethroid selec- day 45 compared with permethrin). This occurrence would not tive pressure, leading to control failure. be expected outside a controlled study. In fact, at the start of Our trial had several limitations that merit consider- this study, the mean number of body lice per homeless per- ation. First, we believe that head and body lice are trans- sons was high (approximately 450 body lice per subject), and ferred among people when they come in close personal con- despite the availability of the protocol underwear in shelters, tact in the shelter, and thus the resistance to permethrin is mosthomelesspersonschangedtheirclothesonlyduringsched- also shared. Indeed, of the 11 homeless persons who were uled follow-up visits and had not washed them between visits. free of live body lice on day 14 in the permethrin group, 2 In conclusion, this trial clearly demonstrates that the use were reinfected on day 45. Moreover, the frequency of the of permethrin–impregnated underwear had the consequence L920F mutation increased significantly from baseline to day of increasing the percentage of permethrin-resistant body lice 45 in the placebo group in study A, suggesting that L920F- in sheltered homeless persons. These findings lead us to rec- mutated lice had been transferred between the permethrin ommend avoiding the use of permethrin to treat body lice in- and placebo groups. festations, although implementing new strategies is crucial.

ARTICLE INFORMATION Émergentes; and the infectious diseases specialists 9. WHO recommended long-lasting insecticidal Accepted for Publication: June 27, 2013. for helpful discussions and active participation in mosquito nets. http://www.who.int/whopes/Long the study. We also thank the directors and the staff _lasting_insecticidal_nets_Jul_2012.pdf. Accessed Published Online: December 4, 2013. of the 2 shelters for their assistance. November 19, 2012. doi:10.1001/jamadermatol.2013.6398. Correction: This article was corrected on 10. Sholdt LL, Rogers EJ Jr, Gerberg EJ, Schreck CE. Author Contributions: Dr Brouqui had full access December 5, 2013, for an error in the first sentence Effectiveness of permethrin-treated military to all of the data in the study and takes of the Results section of the Abstract. uniform fabric against human body lice. Mil Med. responsibility for the integrity of the data and the 1989;154(2):90-93. accuracy of the data analysis. REFERENCES Study concept and design: Benkouiten, Badiaga, 11. Tomalik-Scharte D, Lazar A, Meins J, et al. Raoult, Brouqui. 1. Brouqui P, Stein A, Dupont HT, et al. Dermal absorption of permethrin following topical Acquisition of data: Benkouiten, Drali, Badiaga, Ectoparasitism and vector-borne diseases in 930 administration. Eur J Clin Pharmacol. Veracx. homeless people from Marseilles. Medicine 2005;61(5-6):399-404. Analysis and interpretation of data: Benkouiten, (Baltimore). 2005;84(1):61-68. 12. Young GD, Evans S. Safety and efficacy of DEET Drali, Giorgi. 2. Raoult D, Roux V. The body louse as a vector of and permethrin in the prevention of arthropod Drafting of the manuscript: Benkouiten, Drali, reemerging human diseases. Clin Infect Dis. attack. Mil Med. 1998;163(5):324-330. Badiaga, Giorgi, Raoult. 1999;29(4):888-911. 13. Hollister LE. AMA Drug Evaluations Annual Critical revision of the manuscript for important 3. Badiaga S, Brouqui P. Human louse-transmitted 1991. JAMA. 1991;266(3):424. intellectual content: Veracx, Raoult, Brouqui. infectious diseases. Clin Microbiol Infect. 14. Drali R, Benkouiten S, Badiaga S, Bitam I, Rolain Statistical analysis: Benkouiten, Giorgi. 2012;18(4):332-337. Obtained funding: Brouqui. JM, Brouqui P. Detection of a knockdown resistance Administrative, technical, or material support: 4. Izri A, Chosidow O. Efficacy of machine mutation associated with permethrin resistance in Benkouiten, Badiaga, Raoult. laundering to eradicate head lice: the body louse Pediculus humanus corporis by use Study supervision: Benkouiten, Drali, Badiaga, recommendations to decontaminate washable of melting curve analysis genotyping. J Clin Brouqui. clothes, linens, and fomites. Clin Infect Dis. Microbiol. 2012;50(7):2229-2233. 2006;42(2):e9-e10. Conflict of Interest Disclosures: None reported. 15. Lee SH, Gao JR, Yoon KS, et al. Sodium channel 5. Badiaga S, Raoult D, Brouqui P. Preventing and mutations associated with knockdown resistance in Funding/Support: This study was supported by controlling emerging and reemerging transmissible the human head louse, Pediculus capitis (De Geer). national grant PHRC 2010 from the French Health diseases in the homeless. Emerg Infect Dis. Pestic Biochem Physiol. 2003;75:79-91. Ministry to Dr Brouqui. 2008;14(9):1353-1359. 16. SupYoon K, Symington SB, Hyeock Lee S, Role of the Sponsor: The French Health Ministry 6. Foucault C, Ranque S, Badiaga S, Rovery C, Soderlund DM, Marshall Clark J. Three mutations had no role in the design and conduct of the study; Raoult D, Brouqui P. Oral ivermectin in the identified in the voltage-sensitive sodium channel collection, management, analysis, and treatment of body lice. J Infect Dis. alpha-subunit gene of permethrin-resistant human interpretation of the data; and preparation, review, 2006;193(3):474-476. head lice reduce the permethrin sensitivity of house or approval of the manuscript; decision to submit fly Vssc1 sodium channels expressed in Xenopus the manuscript for publication. 7. Badiaga S, Foucault C, Rogier C, et al. The effect of a single dose of oral ivermectin on pruritus in the oocytes. Insect Biochem Mol Biol. 2008;38(3): Additional Contributions: We thank the homeless homeless. J Antimicrob Chemother. 296-306. individuals who were involved in this study; the 2008;62(2):404-409. 17. Clark JM. Permethrin resistance due to medical and pharmacy students, interns and knockdown gene mutations is prevalent in human fellows, and researchers of the Unité de Recherche 8. Chosidow O. Scabies and pediculosis. Lancet. sur les Maladies Infectieuses et Tropicales 2000;355(9206):819-826.

E6 JAMA Dermatology Published online December 4, 2013 jamadermatology.com

Downloaded From: http://archderm.jamanetwork.com/ by American Medical Association, June Robinson on 12/30/2013 Permethrin–Impregnated Underwear Original Investigation Research

head louse populations. Open Dermatol J. 24. Hodgdon HE, Yoon KS, Previte DJ, et al. man [in Russian]. Med Parazitol (Mosk). 2010;4:63-68. Determination of knockdown resistance allele 1995;(1):23-25. 18. Durand R, Bouvresse S, Andriantsoanirina V, frequencies in global human head louse 31. Busvine JR. Evidence from double infestations Berdjane Z, Chosidow O, Izri A. High frequency of populations using the serial invasive signal for the specific status of human head lice and body mutations associated with head lice pyrethroid amplification reaction. Pest Manag Sci. lice (Anoplura). Syst Entomol. 1978;3:1-8. 2010;66(9):1031-1040. resistance in schoolchildren from Bobigny, France. 32. Busvine JR. The head and body races of J Med Entomol. 2011;48(1):73-75. 25. Veracx A, Rivet R, McCoy KD, Brouqui P, Raoult Pediculus humanusL. Parasitology. 19. Gao JR, Yoon KS, Lee SH, et al. Increased D. Evidence that head and body lice on homeless 1948;39(1-2):1-16. frequency of the T929I and L932F mutations persons have the same genotype. PLoS One. 2012;7(9):e45903. 33. Bacot AW. A contribution to the bionomics of associated with knockdown resistance in Pediculus humanus (vestimenti) and Pediculus permethrin-resistant populations of the human 26. Olds BP, Coates BS, Steele LD, et al. capitis. Parasitology. 1917;9:228-258. head louse, Pediculus capitis, from California, Comparison of the transcriptional profiles of head Florida, and Texas. Pestic Biochem Physiol. and body lice. Insect Mol Biol. 2012;21(2):257-268. 34. Mullen G, Durden LA. Medical and Veterinary Entomology. San Diego, CA: Academic Press; 2009. 2003;77(3):115-124. 27. Li W, Ortiz G, Fournier PE, et al. Genotyping of 20. Yoon KS, Gao JR, Lee SH, Clark JM, Brown L, human lice suggests multiple emergencies of body 35. Howlett FM. Notes on head- and body-lice and Taplin D. Permethrin-resistant human head lice, lice from local head louse populations. PLoS Negl upon temperature reactions of lice and mosquitoes. Pediculus capitis, and their treatment. Arch Trop Dis. 2010;4(3):e641. Parasitology. 1917;10:186-188. Dermatol. 2003;139(8):994-1000. 28. Leo NP, Hughes JM, Yang X, Poudel SK, 36. Nuttall GHF. The biology of Pediculus humanus: 21. Hurlbut HS, Altman RM, Nibley C Jr. DDT Brogdon WG, Barker SC. The head and body lice of supplementary notes. Parasitology. 1919;11:201-221. resistance in Korean body lice. Science. humans are genetically distinct (Insecta: 37. Alpatov VV, Nastukova OA. Transformation of 1952;115(2975):11-12. Phthiraptera, Pediculidae): evidence from double the head form of Pediculus humanus into the body 22. Eddy GW, Cole MM, Couch MD, Selhime A. infestations. Heredity (Edinb). 2005;95(1):34-40. form under changed conditions of existence. Bull Resistance of human body lice to insecticides. 29. Leo NP, Campbell NJ, Yang X, Mumcuoglu K, Soc Nat Moscow. 1955;60:79-92. Public Health Rep. 1955;70(10):1035-1038. Barker SC. Evidence from mitochondrial DNA that 38. Levene H, Dobzhansky T. Possible genetic 23. McLINTOCK J, Zeini A, Djanbakhsh B. head lice and body lice of humans (Phthiraptera: difference between the head louse and the body Development of insecticide resistance in body lice Pediculidae) are conspecific. J Med Entomol. louse. Am Nat. 1959;93:347-353. in villages of North-Eastern Iran. Bull World Health 2002;39(4):662-666. Organ. 1958;18(4):678-680. 30. Khudobin VV. The adaptive potentials of human head and clothes lice when parasitizing on

jamadermatology.com JAMA Dermatology Published online December 4, 2013 E7

Downloaded From: http://archderm.jamanetwork.com/ by American Medical Association, June Robinson on 12/30/2013 Supplementary Online Content Benkouiten S, Drali R, Badiaga S, et al. Effect of Permethrin-Impregnated Underwear on Body Lice in Sheltered Homeless Persons: A Randomized Controlled Trial eTable 1. Stratified random sampling of the levels of body lice infestations eTable 2. Evolution of the mean number of body lice between days 1 and 14 and days 1 and 45 in each subgroup This supplementary material has been provided by the authors to give readers additional information about their work.

186 eTable 1. Stratified random sampling of the levels of body lice infestations Day 1 Day 45

Licea, Min–Max Homeless, No. Liceb, % Lice testedc, No. /total Homeless, No. Liceb, % Lice testedc, No./total

1–9 16 33.33 12/41 5 100 17/17

10–49 19 5 21/404 11 15 47/316 50–99 12 2.50 21/832 4 7.50 20/272 100–499 13 1.25 44/3554 6 3.75 56/1485 500–999 8 0.63 37/5711 3 1.88 44/2360 1000–1999 5 0.31 23/7571 2 0.94 29/3099 Total 73 158/18113 31 213/7549 a Total number of body lice per homeless person b Percentage of body lice drawn per homeless person c Number of body lice tested

187

eTable 2. Evolution of the mean number of body lice between days 1 and 14 and days 1 and 45 in each subgroup Day 1 Day 45 P Value Day 1 Day 45 P Value

Permethrin Lice, No. (%) 8907 (100) 5460 (61.3) 8339 (100) 4008 (48.1) Lice, Mean (SD) 278.3 (472.7) 176.1 (500.9) 0.001 308.8 (502.8) 148.4 (427.6) 0.001 Placebo Lice, No. (%) 7545 (100) 2912 (38.6) 6078 (100) 3542 (58.3) Lice, Mean (SD) 269.5 (431.0) 104 (202.1) 0.001 253.3 (440.8) 147.6 (272.3) 0.04 a Total number of body lice per homeless person

Conclusions et perspectives

Cette thèse s’inscrit dans la continuité des nombreux travaux sur le pou humain réalisés à l’URMITE au sein des équipes dirigées par le

Professeur Didier Raoult, un pourvoyeur essentiel des connaissances scientifiques dans ce domaine durant les vingt dernières années.

A travers cette thèse, nous avons pu répondre à certaines questions concernant les poux humains restées en débat depuis longtemps.

Ainsi, nous avons mis en place un outil moléculaire qui permet de différencier pour la première fois entre le pou de tête et le pou de corps du clade mitochondrial le plus distribué à travers le monde

(Clade A). Un outil qui a démontré son utilité sur le terrain notamment pour déterminer le génotype des poux infectés par des pathogènes.

Nous avons mis en évidence l’existence d’un nouveau clade mitochondrial, le Clade D renfermant des poux de tête et des poux de corps susceptibles de vectoriser B. quintana et Y. pestis. Nous avons mis en place un système pour pouvoir retracer les migrations des humains et des pathogènes à travers l'analyse de poux anciens provenant de différentes périodes et de différentes localisations. Nous avons démontré que P. mjobergi était à l’origine un pou humain qui a

été transféré aux primates du Nouveau Monde par les premiers

189 Hommes à avoir atteint le continent américain il y a des milliers d’années. Cela fût possible grâce aux premières analyses génétiques effectuées sur P. mjobergi. Nous avons mis en place un outil de détection et de contrôle de la résistance moléculaire des poux à la perméthrine. Outil qui fût particulièrement utile dans l'étude clinique que nous avons menée pour déterminer si l'utilisation de sous- vêtements imprégnés d'insecticide offrait une protection efficace à long terme contre les poux de corps infestant les personnes sans-abri.

Durant cette thèse, j’ai eu la chance d’accéder à un matériel biologique d’une rareté exceptionnelle, et comme on l’a constaté, dans chacune des thématiques où ce matériel fut impliqué, les résultats obtenus étaient surprenants. A l’ère du séquençage à haut débit, il serait très intéressant de séquencer le maximum de poux provenant de différents endroits à travers le monde pour effectuer une analyse globale des génomes. Ce qui permettra d’en savoir un peu plus sur l’odyssée de l’espèce humaine et des pathogènes vectorisés par les poux. Car comme on le soupçonne, ces informations doivent être enregistrées et bien gardées par le plus intimes de nos parasites.

190 Références

1. Johnson KP, Yoshizawa K, Smith VS (2004) Multiple origins of parasitism in lice. Proc Biol Sci 271: 1771-1776. 10.1098/rspb.2004.2798 [doi];HM75VQVC7RH4PK25 [pii].

2. Lyal CH (1985) Phylogeny and classification of the Psocodea, with particular reference to the lice (Psocodea: Phthiraptera). Systematic Entomology 10: 145-165.

3. Barker SC, Whiting M, Johnson KP, Murrell A (2003) Phylogeny of the lice (Insecta, Phthiraptera) inferred from small subunit rRNA. Zoologica Scripta 32: 407-414.

4. Wappler T, Smith VS, Dalgleish RC (2004) Scratching an ancient itch: an Eocene bird louse fossil. Proc Biol Sci 271 Suppl 5: S255- S258. 10.1098/rsbl.2003.0158 [doi].

5. Grimaldi D, Engel MS (2006) Fossil Liposcelididae and the lice ages (Insecta: Psocodea). Proc Biol Sci 273: 625-633.

6. Smith VS, Ford T, Johnson KP, Johnson PC, Yoshizawa K, Light JE (2011) Multiple lineages of lice pass through the K-Pg boundary. Biol Lett 7: 782-785. rsbl.2011.0105 [pii];10.1098/rsbl.2011.0105 [doi].

7. Barker SC (1991) Evolution of host-parasite associations among species of lice and rock-wallabies: coevolution? (J. F. A. Sprent Prize lecture, August 1990). Int J Parasitol 21: 497-501. 0020- 7519(91)90053-A [pii].

191 8. Hafner MS, Nadler SA (1988) Phylogenetic trees support the coevolution of parasites and their hosts. Nature 332: 258-259. 10.1038/332258a0 [doi].

9. Brooks DR (1979) Testing the context and extent of host-parasite coevolution. SysT Zool 28: 229-307.

10. Johnson KP, Adams RJ, Page RD, Clayton DH (2003) When do parasites fail to speciate in response to host speciation? Syst Biol 52: 37-47. JF8959128JU409RB [pii].

11. Page RD, Charleston MA (1998) Trees within trees: phylogeny and historical associations. Trends Ecol Evol 13: 356-359. S0169- 5347(98)01438-4 [pii].

12. Ewing HE (1938) The Sucking Lice of American Monkeys. The Journal of Parasitology 24: 13-33.

13. Reed DL, Light JE, Allen JM, Kirchman JJ (2007) Pair of lice lost or parasites regained: the evolutionary history of anthropoid primate lice. BMC Biol 5: 7. 1741-7007-5-7 [pii];10.1186/1741- 7007-5-7 [doi].

14. De Geer C (1778) Mémoires pour servir à l'histoire des Insectes. 62-68.

15. Hopkins GHE (1952) The correct names of the body and head lice of man. 91-92.

16. Sangare AK, Boutellis A, Drali R, Socolovschi C, Barker SC, Diatta G, Rogier C, Olive MM, Doumbo OK, Raoult D (2014)

192 Detection of Bartonella quintana in African Body and Head Lice. Am J Trop Med Hyg 91: 294-301.

17. Chosidow O, Chastang C, Brue C, Bouvet E, Izri M, Monteny N, Bastuji-Garin S, Rousset JJ, Revuz J (1994) Controlled study of malathion and d-phenothrin lotions for Pediculus humanus var capitis-infested schoolchildren. Lancet 344: 1724-1727.

18. Nuttall GH (1919) The systematic position, synonymy and iconography of Pediculus humanus and Phthirus pubis. Parasitology 11: 329-346.

19. Hindle E (1917) Notes on the biology of Pediculus humanus L. Parasitology 9: 259-265.

20. Ewing HE (1926) A revision of the American lice of the genus Pediculus, together with a consideration of the significance of their geographical and host distribution. Proc US Nat Mus 68: 1-30.

21. Nuttall GH (1919) The biology of Pediculus humanus L. (Supplementary notes). Parasitology 11: 201-220.

22. Veracx A, Boutellis A, Merhej V, Diatta G, Raoult D (2012) Evidence for an African cluster of human head and body lice with variable colors and interbreeding of lice between continents. PLoS One 7: e37804. 10.1371/journal.pone.0037804 [doi];PONE-D-12- 05586 [pii].

23. Reed DL, Smith VS, Hammond SL, Rogers AR, Clayton DH (2004) Genetic analysis of lice supports direct contact between

193 modern and archaic humans. PLoS Biol 2: e340. 10.1371/journal.pbio.0020340 [doi].

24. Boutellis A, Abi-Rached L, Raoult D (2014) The origin and distribution of human lice in the world. Infect Genet Evol 23: 209- 217.

25. Boutellis A, Drali R, Rivera MA, Mumcuoglu KY, Raoult D (2013) Evidence of sympatry of clade a and clade B head lice in a pre-Columbian Chilean mummy from Camarones. PLoS One 8: e76818.

26. Raoult D, Reed DL, Dittmar K, Kirchman JJ, Rolain JM, Guillen S, Light JE (2008) Molecular identification of lice from pre- Columbian mummies. J Infect Dis 197: 535-543. 10.1086/526520 [doi].

27. Johnston JS, Yoon KS, Strycharz JP, Pittendrigh BR, Clark JM (2007) Body lice and head lice (Anoplura: Pediculidae) have the smallest genomes of any hemimetabolous insect reported to date. J Med Entomol 44: 1009-1012.

28. Kirkness EF, Haas BJ, Sun W, Braig HR, Perotti MA, Clark JM, Lee SH, Robertson HM, Kennedy RC, Elhaik E, Gerlach D, Kriventseva EV, Elsik CG, Graur D, Hill CA, Veenstra JA, Walenz B, Tubio JM, Ribeiro JM, Rozas J, Johnston JS, Reese JT, Popadic A, Tojo M, Raoult D, Reed DL, Tomoyasu Y, Kraus E, Mittapalli O, Margam VM, Li HM, Meyer JM, Johnson RM, Romero-Severson J, Vanzee JP, Alvarez-Ponce D, Vieira FG, Aguade M, Guirao-Rico S, Anzola JM, Yoon KS, Strycharz JP,

194 Unger MF, Christley S, Lobo NF, Seufferheld MJ, Wang N, Dasch GA, Struchiner CJ, Madey G, Hannick LI, Bidwell S, Joardar V, Caler E, Shao R, Barker SC, Cameron S, Bruggner RV, Regier A, Johnson J, Viswanathan L, Utterback TR, Sutton GG, Lawson D, Waterhouse RM, Venter JC, Strausberg RL, Berenbaum MR, Collins FH, Zdobnov EM, Pittendrigh BR (2010) Genome sequences of the human body louse and its primary endosymbiont provide insights into the permanent parasitic lifestyle. Proc Natl Acad Sci U S A 107: 12168-12173. 1003379107 [pii];10.1073/pnas.1003379107 [doi].

29. Olds BP, Coates BS, Steele LD, Sun W, Agunbiade TA, Yoon KS, Strycharz JP, Lee SH, Paige KN, Clark JM, Pittendrigh BR (2012) Comparison of the transcriptional profiles of head and body lice. Insect Mol Biol 21: 257-268. 10.1111/j.1365-2583.2012.01132.x [doi].

30. Raoult D, Roux V (1999) The body louse as a vector of reemerging human diseases. Clin Infect Dis 29: 888-911. 10.1086/520454 [doi].

31. Houhamdi L, Lepidi H, Drancourt M, Raoult D (2006) Experimental model to evaluate the human body louse as a vector of plague. J Infect Dis 194: 1589-1596. JID36487 [pii];10.1086/508995 [doi].

32. Piarroux R, Abedi AA, Shako JC, Kebela B, Karhemere S, Diatta G, Davoust B, Raoult D, Drancourt M (2013) Plague epidemics

195 and lice, Democratic Republic of the Congo. Emerg Infect Dis 19: 505-506.

33. Bonilla DL, Kabeya H, Henn J, Kramer VL, Kosoy MY (2009) Bartonella quintana in body lice and head lice from homeless persons, San Francisco, California, USA. Emerg Infect Dis 15: 912-915. 10.3201/eid1506.090054 [doi].

34. Boutellis A, Veracx A, Angelakis E, Diatta G, Mediannikov O, Trape JF, Raoult D (2012) Bartonella quintana in head lice from Senegal. Vector Borne Zoonotic Dis 12: 564-567. 10.1089/vbz.2011.0845 [doi].

35. Angelakis E, Diatta G, Abdissa A, Trape JF, Mediannikov O, Richet H, Raoult D (2011) Altitude-dependent Bartonella quintana genotype C in head lice, Ethiopia. Emerg Infect Dis 17: 2357- 2359. 10.3201/eid1712.110453 [doi].

36. Sasaki T, Poudel SKS, Isawa H, Hayashi T, Seki S, Tomita T, Sawabe K, Kobayashi M (2006) First Molecular Evidence of Bartonella quintana in Pediculus humanus capitis (Phthiraptera: Pediculidae), Collected from Nepalese Children. J Med Entomol 43: 110-112.

37. Boutellis A, Mediannikov O, Bilcha KD, Ali J, Campelo D, Barker SC, Raoult D (2013) Borrelia recurrentis in head lice, Ethiopia. Emerg Infect Dis 19: 796-798. 10.3201/eid1905.121480 [doi].

38. Mumcuoglu KY, Zias J (1988) Head lice, Pediculus humanus capitis (Anoplura, Pediculidae) from hair combs excavated in

196 Israel and dated from the first century B.C. to the eighth century A.D. J Med Entomol 25: 545-547. 39. Clark JM, Yoon KS, Lee SH, Pittendrigh BR (2013) Human lice: Past, present and future control. Pesticide Biochemistry and Physiology 106: 162-171.

40. Pariser DM, Meinking TL, Ryan WG (2013) Topical ivermectin lotion for head lice. N Engl J Med 368: 967. 10.1056/NEJMc1215548 [doi].

41. Yoon KS, Strycharz JP, Baek JH, Sun W, Kim JH, Kang JS, Pittendrigh BR, Lee SH, Clark JM (2011) Brief exposures of human body lice to sublethal amounts of ivermectin over- transcribes detoxification genes involved in tolerance. Insect Mol Biol 20: 687-699. 10.1111/j.1365-2583.2011.01097.x [doi].

42. Drali R, Boutellis A, Raoult D, Rolain JM, Brouqui P (2013) Distinguishing body lice from head lice by multiplex real-time PCR analysis of the Phum_PHUM540560 gene. PLoS One 8: e58088.

43. Veracx A, Raoult D (2012) Biology and genetics of human head and body lice. Trends Parasitol 28: 563-571. S1471- 4922(12)00163-8 [pii];10.1016/j.pt.2012.09.003 [doi].

44. Alpatov WW, Nastjukova OK (1955) Transformation of the head form of Pediculus humanus L. into the body form under the influence of changed living conditions. Bull Soc Nat Moscow 60: 79-92.

197 45. Bacot A (1917) A contribution to the bionomics of Pediculus humanus (vestimenti) and Pediculus capitis. Parasitology 9: 228- 258.

46. Drali R, Sangare AK, Boutellis A, Angelakis E, Veracx A, Socolovschi C, Brouqui P, Raoult D (2014) Bartonella quintana in body lice from scalp hair of homeless persons, France. Emerg Infect Dis 20: 907-908. 10.3201/eid2005.131242 [doi].

47. Kittler R, Kayser M, Stoneking M (2003) Molecular evolution of Pediculus humanus and the origin of clothing. Curr Biol 13: 1414- 1417.

48. Light JE, Toups MA, Reed DL (2008) What's in a name: the taxonomic status of human head and body lice. Mol Phylogenet Evol 47: 1203-1216.

49. Boutellis A, Abi-Rached L, Raoult D (2014) The origin and distribution of human lice in the world. Infect Genet Evol 23: 209- 217. S1567-1348(14)00020-3 [pii];10.1016/j.meegid.2014.01.017 [doi].

50. Hafner MS, Sudman PD, Villablanca FX, Spradling TA, Demastes JW, Nadler SA (1994) Disparate rates of molecular evolution in cospeciating hosts and parasites. Science 265: 1087-1090.

51. Brouqui P, Stein A, Dupont HT, Gallian P, Badiaga S, Rolain JM, Mege JL, La SB, Berbis P, Raoult D (2005) Ectoparasitism and vector-borne diseases in 930 homeless people from Marseilles.

198 Medicine (Baltimore) 84: 61-68. 00005792-200501000-00006 [pii].

52. Badiaga S, Foucault C, Rogier C, Doudier B, Rovery C, Dupont HT, Castro P, Raoult D, Brouqui P (2008) The effect of a single dose of oral ivermectin on pruritus in the homeless. J Antimicrob Chemother 62: 404-409. dkn161 [pii];10.1093/jac/dkn161 [doi].

53. Foucault C, Ranque S, Badiaga S, Rovery C, Raoult D, Brouqui P (2006) Oral ivermectin in the treatment of body lice. J Infect Dis 193: 474-476. JID35187 [pii];10.1086/499279 [doi].

54. Drali R, Benkouiten S, Badiaga S, Bitam I, Rolain JM, Brouqui P (2012) Detection of a knockdown resistance mutation associated with permethrin resistance in the body louse Pediculus humanus corporis by use of melting curve analysis genotyping. J Clin Microbiol 50: 2229-2233. JCM.00808-12 [pii];10.1128/JCM.00808-12 [doi].

55. Benkouiten S, Drali R, Badiaga S, Veracx A, Giorgi R, Raoult D, Brouqui P (2014) Effect of permethrin-impregnated underwear on body lice in sheltered homeless persons: a randomized controlled trial. JAMA Dermatol 150: 273-279. 1782130 [pii];10.1001/jamadermatol.2013.6398 [doi].

199

Annexes

201

Article X: Matrix-Assisted Laser Desorption Ionization–Time of Flight Mass Spectrometry for Rapid Identification of Tick Vectors

Journal of Clinical Microbiology, 51(2), 522-528.

203

Matrix-Assisted Laser Desorption Ionization −Time of Flight Mass Spectrometry for Rapid Identification of Tick Vectors

Amina Yssouf, Christophe Flaudrops, Rezak Drali, Tahar Kernif, Cristina Socolovschi, Jean-Michel Berenger, Didier Raoult and Philippe Parola J. Clin. Microbiol. 2013, 51(2):522. DOI: 10.1128/JCM.02665-12. Published Ahead of Print 5 December 2012. onoddfo on October 24, 2014 by guest Downloaded from

Updated information and services can be found at: http://jcm.asm.org/content/51/2/522

These include:

REFERENCES This article cites 32 articles, 4 of which can be accessed free at: http://jcm.asm.org/ http://jcm.asm.org/content/51/2/522#ref-list-1

CONTENT ALERTS Receive: RSS Feeds, eTOCs, free email alerts (when new articles cite this article), more» on October 24, 2014 by guest

Information about commercial reprint orders: http://journals.asm.org/site/misc/reprints.xhtml To subscribe to to another ASM Journal go to: http://journals.asm.org/site/subscriptions/

205 Matrix-Assisted Laser Desorption Ionization–Time of Flight Mass Spectrometry for Rapid Identiﬁcation of Tick Vectors

Amina Yssouf,a Christophe Flaudrops,b Rezak Drali,a Tahar Kernif,a Cristina Socolovschi,a Jean-Michel Berenger,a Didier Raoult,a Philippe Parolaa Aix Marseille Université, Unité de Recherche en Maladies Infectieuses et Tropicales Emergentes, UM63, CNRS 7278, IRD 198, Inserm 1095, WHO Collaborative Center for Rickettsioses and Other Arthropod Borne Bacterial Diseases, Faculté de Médecine, Marseille, Francea; Assistance Publique des Hôpitaux de Marseille, CHU Timone, Pôle Infectieux, Marseille, Franceb

A method for rapid species identification of ticks may help clinicians predict the disease outcomes of patients with tick bites and may inform the decision as to whether to administer postexposure prophylactic antibiotic treatment. We aimed to establish a matrix-assisted laser desorption ionization–time offlight mass spectrometry (MALDI-TOF MS) spectrum database based on the Downloaded from analysis of the legs of six tick vectors: Amblyomma variegatum, Rhipicephalus sanguineus, Hyalomma marginatum rufipes, Ix- odes ricinus, Dermacentor marginatus, and Dermacentor reticulatus. A blind test was performed on a trial set of ticks to identify specimens of each species. Subsequently, we used MALDI-TOF MS to identify ticks obtained from the wild or removed from patients. The latter tick samples were also identified by 12S ribosomal DNA (rDNA) sequencing and were tested for bacterial infections. Ticks obtained from the wild or removed from patients (R. sanguineus, I. ricinus, and D. marginatus) were accurately identified using MALDI-TOF MS, with the exception of those ticks for which no spectra were available in the database. Further- more, one damaged specimen was correctly identified as I. ricinus, a vector of Lyme disease, using MALDI-TOF MS only. Six of the 14 ticks removed from patients were found to be infected by pathogens that included Rickettsia, Anaplasma, and Borrelia http://jcm.asm.org/ spp. MALDI-TOF MS appears to be an effective tool for the rapid identification of tick vectors that requires no previous expertise in tick identification. The benefits for clinicians include the more targeted surveillance of patients for symptoms of potentially transmitted diseases and the ability to make more informed decisions as to whether to administer postexposure prophylactic treatment.

icks are obligate hematophagous parasites of the order Acari. clear internal transcribed spacer 2 (ITS2), have been developed to TThese arthropods can feed on every known class of vertebrate identify arthropods, including ticks (9). However, there is cur- on October 24, 2014 by guest and can bite people (1). Ticks are currently the second leading rently no PCR assay that can distinguish tick species, and ideal vector of human infectious diseases and can carry bacterial (1), PCR primer pairs that can amplify the relevant gene fragments are viral (1a), and protozoan pathogens (2). However, only in 1982, not available. with the identification of Borrelia burgdorferi as the etiological In addition to the technical and logistical drawbacks of PCR agent of Lyme disease, was the major effect of ticks on public assays, this approach is further limited by the availability of gene health recognized, leading to an increased awareness of tick-borne sequences in GenBank. Protein profiling by matrix-assisted laser diseases (3). Since then, more than 15 tick-borne rickettsioses desorption ionization–time offlight mass spectrometry (MALDI- have emerged throughout the world (4). TOF MS) is now increasingly common for the routine identifica- The removal of a tick from the human body is a common tion of microorganisms in clinical microbiology (10). This revo- situation, and patients may visit a physician with an attached or lutionary, reliable, and cost-effective technique is simpler and removed tick. Certain tick species are well-known vectors of hu- faster than conventional phenotypic and molecular methods for man diseases, such that identifying the species, which will alert the the identification of human pathogens (10). physician to the diseases that may have been transmitted, is clini- The MALDI-TOF MS approach wasfirst applied to arthropods cally helpful if such information is obtained quickly (1). Indeed, for the differentiation of Drosophila species (11). It was found that recent studies confirm that the use of doxycycline prophylaxis protein extracts obtained from whole specimens generated repro- following an Ixodes tick bite is useful for the prevention of Lyme ducible spectra (11). Species-specific protein profiles have also disease (4). Similar postexposure regimens could also prevent been be used to differentiate three species of aphids (insects that tick-borne relapsing fever in areas of endemicity (5). However, for feed on plants) (12). In 2011, a blind test in which 111 wild spec- prophylactic treatment to be effective, it must be delivered shortly imens were compared to the database profiles of Culicoides species after potentially infectious ticks are removed from patients (6). Ticks species can be morphologically identified using taxonomic keys for endemic species in several geographic regions (1). Received 4 October 2012 Returned for modification 25 October 2012 However, morphological identification can be difficult because it Accepted 18 November 2012 requires some entomological expertise, and it is difficult to iden- Published ahead of print 5 December 2012 tify a specimen that is damaged or at an immature stage of its life Address correspondence to Philippe Parola, [email protected]. cycle (1). Molecular methods, such as the sequencing of the mito- Copyright © 2013, American Society for Microbiology. All Rights Reserved. chondrial 12S (7), 18S (8), and 16S ribosomal DNAs (rDNAs) (7), doi:10.1128/JCM.02665-12 mitochondrial cytochrome oxidase subunit 1 (COX1), and nu-

522 jcm.asm.org Journal of Clinical Microbiology206 p. 522–528 February 2013 Volume 51 Number 2 MALDI-TOF MS Identiﬁcation of Ticks

TABLE 1 Arthropods used to establish the reference database of MALDI-TOF spectra and arthropods used in the blind test No. of specimens used to No. of specimens used for the blind test create the database procedure Geographical Arthropod for database origin Sourcea Total Sex and/or stageb Total (score range) Sex and/or stageb Ticks Amblyomma variegatum Senegal LC 18 6 M, 5 F, 7 N 2 (2.248–2.357) 1 M, 1F Rhipicephalus sanguineus France LC 10 5 M, 5 F 2 (1.995- 2.135) 1 M, 1 F France Vegetation 3 (1.715–1.745) M France Vegetation 2 (1.701–1.884) F Dermacentor marginatus France LC 10 5 M, 5 F 2 (2.232–2.519) 1 M, 1 F Dermacentor reticulatus Croatia LC 10 5 M, 6 F 3 (2.119- 2.281) 1 M, 2 F Hyalomma marginatum ruﬁpes Senegal LC 12 6 M, 6 F 2 (1.992–2.352) 1 M, 1 F Ixodes ricinus France LC 7 N 2 (1.706–1.831) 1 M, 1 F France Vegetation 7 (1.304–1.878) 3 M, 2 F, 2 N Rhipicephalus sulcatus Senegal Animal 2 (1.009–1.166) 1 M, 1 F Downloaded from Haemaphysalis concinna France Vegetation 2 (0.838–1.062) 1 M, 1 F

Other arthropods Ctenocephalides felis England LC 17 11 F, 6 M 2 (2.103–2.123) 1 M, 1 F Pediculus humanus corporis USA LC 12 7 M, 6 F 2 (1.788–1.797) 1 M, 1 F Triatoma infestans Bolivia LC 12 6 M, 6 F 2 (1.733–1.779) 1 M, 1 F Cimex lectularius France LC 12 6 M, 6 F 2 (2.145–2.429) 1 M, 1 F Culex pipiens France LC 7 F 2 (1.655–1.749) F Apis mellifera France Field 2 (0.732–0.778) 1 M, 1 F http://jcm.asm.org/ Pyrrhocoris apterus France Field 2 (0.759–0.998) 1 M, 1 F Blaptica dubia India Field 2 (0.707–1.066) 1 M, 1 F Tenibrio molitor France Field 2 (0.755–0.958) 1 M, 1 F a LC, laboratory colonies. b M, male; F, female; N, nymph stage. showed that MALDI-TOF MS can differentiate species of Culi- Queensland, Australia), and were compared with sequences from Gen- coides biting midges collected in thefield ( 13). More recently, a Bank. The unused tick body parts were stored at�80°C for subsequent on October 24, 2014 by guest MALDI-TOF MS study of seven ticks reported that whole ticks or analyses. body parts, excluding the legs, generate spectra that are sufficient MALDI-TOF procedure. (i) Preparation of samples. Each specimen for species identification (14). was placed in a 1.5-ml microcentrifuge tube and immobilized or anesthe- The objective of the present study was to investigate the use of tized at�20°C for 30 min ( 11). Subsequently, the specimens were rinsed once with distilled water. After the specimens were dried with paper, four MALDI-TOF MS for the rapid differentiation of tick species using legs were removed with scalpels. One of two different homogenization only their legs. Our goals were to establish a reference database, to solutions was used depending on the specimen’s size (11). The tick legs evaluate the MALDI-TOF MS-based identification system in a were manually homogenized in 60�l of 70% formic acid and 60�l of 50% blind test, and to evaluate this new identification tool using ticks acetonitrile in 1.5-ml microcentrifuge tubes using pellet pestles (Fischer removed from patients. Scientific). The legs of mosquitoes,fleas, and lice were also homogenized in 20�l of formic acid and 20�l of acetonitrile. Triatomine legs were MATERIALS AND METHODS homogenized in 100�l of 70% formic acid and 100�l of 50% acetonitrile. Arthropods. To establish a reference database, we used laboratory-reared All homogenates were centrifuged at 10,000 rpm for 20 s, and 1�l of each hard ticks,fleas, lice, triatomines, mosquitoes, and bedbugs ( Table 1). All supernatant was spotted onto a steel target plate (Bruker Daltonics) in specimens were fresh and nonengorged and maintained at room temper- quadruplicate (13). One microliter of a CHCA matrix suspension com- ature for less than 1 week postmortem. The tick specimens included Am- posed of saturated�-cyano-4-hydroxycinnamic acid (Sigma), 50% ace- blyomma variegatum, Hyalomma marginatum rufipes, Rhipicephalus san- tonitrile, 10% trifluoroacetic acid, and high-performance liquid chroma- guineus, Ixodes ricinus, Dermacentor marginatus, and Dermacentor tography (HPLC)-grade water was directly spotted onto each sample on reticulatus. For the blind test, other specimens of the same species and the target plate to allow cocrystallization (15). The target plate was dried specimens of other arthropod families were used as controls. Starved ticks for several minutes at room temperature before insertion into the collected in thefield or removed from animals were characterized by MALDI-TOF MS instrument. MALDI-TOF MS (Table 1). A total of 14 ticks removed from 14 patients (ii) MALDI-TOF MS parameters. Protein mass profiles were acquired in France from June to August 2012 were morphologically characterized using a Microflex MALDI-TOF mass spectrometer (Bruker Daltonics) using standard taxonomic keys as D. marginatus, R. sanguineus, I. ricinus, with Flex Control software (Bruker Daltonics). We performed measure- Rhipicephalus sp., and Ixodes sp. (Table 2). One extensively damaged hard ments in the linear positive-ion mode (15) between 2 and 20 kDa, and tick removed from a patient was also tested. After the morphological each spectrum corresponded to ions obtained from 240 laser shots per- identification procedure, four legs were removed for MALDI-TOF MS formed in six regions of the same spot. The obtained spectra were pro- (see below) and sequencing assays. DNA was extracted from tick body cessed using Flex Analysis, version 3.3, and MALDI Biotyper, version 3.0, halves, and a 360-bp fragment of the mitochondrial 12S rDNA sequence software. was amplified by PCR and sequenced (14a). The sequences were analyzed (iii) Spectral analysis and reference database creation. To study the using ChromasPro, version 1.34 (Technelysium Pty, Ltd., Tewantin, reproducibility of the species spectra, the average spectral profiles ob-

February 2013 Volume 51 Number 2 207 jcm.asm.org 523 Yssouf et al.

TABLE 2 Identification of ticks removed from patients using morphological, MALDI-TOF MS, and molecular methods and the results of the detection of Rickettsia, Borrelia, Anaplasma, and Bartonella DNA Molecular identification (% similarity with the indicated GenBank sequence) of:a Morphological MALDI-TOF MS identification identification (score) Ticke Bacteria D. marginatus D. marginatus (1.83) D. marginatus (99.3% AM410570.1) D. marginatus D. marginatus (1.798) D. marginatus (99.4% AM410570.1) D. marginatus D. marginatus (1.799) D. marginatus (99.4% AM410570.1) R. slovacad,R. raoultii d R. sanguineus R. sanguineus (1.957) R. sanguineus (99.1% JX206975.1) R. massiliaed I. ricinus I. ricinus (1,737) I. ricinus (99.5% JN248424.1) Damaged I. ricinus (1,376) Not obtainedc I. ricinus I. ricinus (1.321) I. ricinus (99.3% JN248424.1) Rhipicephalus sp. Not identifiedb R. bursa (100% AM410572.1) Ixodes sp. I. ricinus (1.308) I. ricinus (99.2% JM248424.1) B. garinii (99.7% JN828689.1) Rhipicephalus sp. R. sanguineus (2.117) R. sanguineus (99.2% AY559843.1) R. conoriid I. ricinus I. ricinus (1.728) I. ricinus (99.7% JN248424.1)A. phagocytophilum d Downloaded from I. ricinus I. ricinus (1.823) I. ricinus (99.2% JN248424.1) I. ricinus I. ricinus (1.641) I. ricinus (99.06% AY945481.1) Incompletely described bacterium (97.6% AY776167.1), B. miyamotoi (100% FJ874925.1) I. ricinus I. ricinus (1.668) I. ricinus (100% JN248424.1) D. marginatus D. marginatus (1.83) D. marginatus (99.3% AM410570.1) D. marginatus D. marginatus (1.798) D. marginatus (99.4% AM410570.1) R. slovacad, R. raoultiid D. marginatus D. marginatus (1.799) D. marginatus (99.4% AM410570.1) R. massiliaed a DNA extracted from laboratory colonies of uninfected ticks was used as a negative control. DNA extracted from Rickettsia montanensis, Bartonella elizabethae, Borrelia crocidurae,

and Anaplasma phagocytophilum was used as a positive control. http://jcm.asm.org/ b No R. bursa spectrum was available in our MALDI-TOF MS database. c After two attempts. d qPCR. e By 12S RNA sequencing. tained from the four spots for each specimen of the same species were sanguineus DNA samples (19). Rickettsia slovaca- and Rickettsia raoultii- analyzed and compared using the ClinProTools, version 2.2, program specific qPCRs were performed on positive Dermacentor DNA samples (Bruker Daltonics). Using MALDI Biotyper, version 3.0, a database con- (17). Borrelia-positive samples were confirmed by Borrelia-specific qPCR taining the studied arthropod groups was assembled from at leastfive amplification of the internal transcribed spacer (ITS) (19), and identifi- on October 24, 2014 by guest specimens of each sex for all of the studied arthropod species, excluding cation was performed by sequencing a 750-bp gene fragment of theflaB mosquitoes, for which the majority of tested specimens were females. The gene (20). reproducible spectra were used in the database. (iv) Study validation using a blind test and identification of ticks RESULTS from thefield and patients. An intercomparison of the reference spectra of each species was performed to interrogate the database (14, 16). The MALDI-TOF MS spectrum analysis and database assembly.A “start identification” function in the MALDI Biotyper allows the identifi- total of 120 arthropod specimens, including 67 ticks representing cation of each specimen to be tested, and the results were presented as log six species, were subjected to MALDI-TOF MS analysis. The pro- score values. A blind test was then performed with specimens from our tein spectral profiles of all tested arthropod species were very sim- laboratory colonies for which there were corresponding reference spectra ilar between specimens of the same species within the mass range in our database. For each species, two new specimens were used. We also of 2 to 20 kDa, and the signal intensities were very strong. The tested two specimens of arthropods not included in our database, includ- alignment of the spectra of several specimens of the same species ing bees,firebugs, roaches, and beetles. Identification scores were ob- showed that the major identified protein peaks were present in tained for each spectrum of the tested sample. All ticks collected in the each specimen of the same species (Fig. 1). The analysis in Clin- field and removed from patients were tested against the database. When each tick was tested, particular attention was paid to determine if the ProTools of the tick protein profiles yielded spectra comprising spectral profile matched profiles for the reference samples of the same sex between 60 and 136 peaks in the mass range of 2 to 20 kDa, with an as the specimen. The Mantel Haenszel test (Epi Info, version 7, program) average of 106 peaks per spectrum. The spectra corresponding to was used to evaluate the quality of the sex identification. the legs of other arthropods (Culex pipiens, Triatoma infestans, (v) Cluster analysis. We performed hierarchical clustering of the mass Ctenocephalides felis, Pediculus humanus, and Cimex lectularius) spectra of all tested species that were present in or absent from the data- comprised 47 to 64, 94 to 120, 75 to 93, 46 to 76, and 87 to 132 base using the mean spectrum projection (MSP) dendrogram function of peaks, respectively, in the mass range of 2 to 20 kDa. MALDI Biotyper, version 3.0. The objective was to determine whether the The spectra for each species of ticks were shown to be repro- ticks could be clustered using this approach. ducible (Fig. 2) for all tested groups after spectral analysis and Molecular detection of bacteria in ticks removed from patients. All alignment. This reproducibility was observed for both the male DNA samples of ticks removed from patients were screened for Rickettsia spp. by quantitative real-time PCR (qPCR) analysis of a fragment of the and female specimens. Subsequently, a database was created and gltA gene (17). To screen for Bartonella spp., an internally transcribed loaded with all of the spectra for these species. spacer was targeted, whereas a fragment of the 16S rRNA gene was tar- Study validation using a blind test. Querying spectra in the geted for Borrelia spp. (18). Rickettsia massiliae-specific qPCR and Rick- database yielded satisfactory results, with scores between 2.224 ettsia conorii-specific qPCR were performed on positive Rhipicephalus and 2.764. The blind test of the reference database, in which two

524 jcm.asm.org 208 Journal of Clinical Microbiology MALDI-TOF MS Identiﬁcation of Ticks Downloaded from

FIG 1 Matrix-assisted laser desorption ionization–time offlight (MALDI-TOF) mass spectral profiles for the legs of different tick species. (A) Hyalomma(Hy) marginatum rufipes male and female. (B) Rhipicephalus(Rh) sanguineus male and female in the range of 2 to 20 kDa. Intens, signal intensity; au, arbitrary units. http://jcm.asm.org/ specimens of each of the six tick species maintained in the labora- with scores between 1.749 and 2.519. The spectra of bees,firebugs, tory as well as the remaining arthropod groups were assayed, suc- roaches, and beetles, for which references were not included in the cessfully identified all species groups. The majority (69%) of spec- database, scored lower, between 0.778 and 1.066. Regarding the imens were identified to the species level with scores of�2. All sex of the specimen, it was not possible to definitively discriminate specimens tested had spectra that matched the reference spectra male and female ticks based on their MS spectra. However, the on October 24, 2014 by guest

FIG 2 Comparison of the A. variegatum spectral profiles to assess reproducibility and to identify an A. variegatum specimen. (A) Alignment of the spectra of several specimens of A. variegatum (gel view) showing that the major identified protein peaks were present in each specimen of this species. Arb u, arbitrary units. (B) Average peak list for the A. variegatum spectral profiles. (C) Results obtained for the blind testing of an A. variegatum specimen against the database using MALDI Biotyper, version 3.0, software (correct identification with a score of 2.357).

February 2013 Volume 51 Number 2 209 jcm.asm.org 525 Yssouf et al. Downloaded from http://jcm.asm.org/

FIG 3 Dendrogram obtained by cluster analysis of spectra obtained from laboratory-reared ticks,field-collected ticks, and ticks removed from patients. The ticks were analyzed using MALDI Biotyper software. M, male; F, female; N, nymph; L, ticks from laboratory colonies; F, ticks collected in thefield; P, ticks collected from patients. ticks’ spectral profiles more frequently matched those of the cor- not identified by MALDI-TOF MS (no reference spectra in our rect sex than the incorrect sex, with 16/22 (72%) matches for database). After two attempts, we were unable to obtain a high- on October 24, 2014 by guest males and 19/29 (65%) matches for females (P� 0.007). quality 12S RNA gene sequence for the damaged tick that was MALDI-TOF MS using ticks from thefield and patients. The removed from a patient. This specimen was identified as I. ricinus field-collected ticks for which the species were represented in the by MALDI-TOF MS. Finally, the tick that was morphologically database (I. ricinus and R. sanguineus) were identified correctly by identified as Rhipicephalus sp. and was not identified by MALDI- MALDI-TOF MS (scores of�1.8).R. sulcatus and Haemaphysalis TOF MS was identified by sequencing as Rhipicephalus bursa concinna ticks from thefield (absent from our MALDI-TOF MS (100% agreement with AM410572 in GenBank), for which no database) did not match any known spectra. Eleven morphologi- spectrum was available in the MALDI-TOF database. Six of the 14 cally identified adult ticks (I. ricinus, R. sanguineus, and D. mar- ticks removed from patients were found to be infected by bacteria, ginatus) removed from patients were correctly identified by including Rickettsia massiliae, R. conorii, R. slovaca, R. raoultii, MALDI-TOF MS (identification scores between 1.321 and 2.117). Borrelia garinii, Borrelia miyamotoi, and Anaplasma phagocytophi- An Ixodes sp. nymph removed from a patient was identified asI. lum (Table 1). DNA sequences closely related to those of an in- ricinus using MALDI-TOF MS. Additionally, the tick that was completely described bacterium (21) were also detected in one extensively damaged was identified as I. ricinus based on its four specimen of I. ricinus. spectra, with a score of 1.376. However, one specimen (Rhipiceph- alus sp.) removed from a patient could not be identified by MALDI-TOF MS, as the spectrum did not match that of any spec- DISCUSSION imens in the database. Although other factors should be considered, the identification of For the phylogenetic analysis, we performed hierarchical clus- ticks that have bitten or been removed from patients is thefirst tering of the mass spectra of different species of tested ticks that step in assessing the risk of infection (1). In this study, all ticks were present in or absent from our reference database. In the removed from patients were potential vectors of pathogens that generated dendrogram, specimens of the same tick species, in- depend on the involved tick species for propagation. In Europe,D. cluding R. sanguineus, I. ricinus, and D. marginatus from the lab- marginatus is one of the primary vectors of R. slovaca and R. raoul- oratory,field, and patients, clustered together with the exception tii, two spotted fever group rickettsiae responsible for tick-borne of one specimen (Fig. 3). lymphadenopathy (TIBOLA), also called Dermacentor-borne Molecular identification of ticks and the detection of bacte- necrosis erythema and lymphadenopathy (DEBONEL) or SENLAT, ria. 12S gene sequencing was performed on 13 of 14 specimens which is defined as scalp eschar and neck lymphadenopathy after removed from patients. The results corroborated those obtained a tick bite (21a). I. ricinus is a major vector of several bacterial by morphological analysis and MALDI-TOF MS for most samples agents, including the causative agent of Lyme borreliosis, Borrelia (Table 1). R. sulcatus and H. concinna collected from thefield were garinii( 3), B. miyamotoi, which causes relapsing fever (21b),A.

526 jcm.asm.org 210 Journal of Clinical Microbiology MALDI-TOF MS Identification of Ticks phagocytophilum, the causative agent of human granulocytic ana- results of the MALDI-OF MS analysis of arthropod specimens. To plasmosis (22), and Rickettsia helvetica, an emerging pathogen (4). expand our database, we will test a large number of tick species. Rhipicephalus sanguineus, the brown dog tick, is a primary vec- Some species are available in collections. However, it seems that tor of Rickettsia conorii, the causative agent of life-threatening long-term storage in 70% ethanol reduces the reproducibility of Mediterranean spotted fever (23), Rickettsia rickettsii is the vector the MALDI-TOF MS spectra (13, 25). It will also be useful to of Rocky Mountain spotted fever in southern regions of the evaluate the use of MALDI-TOF MS for the identification of fro- United States (23a), andR. massiliae is an emerging pathogen zen specimens. (23b). Rhipicephalus bursa is a known vector of several cattle par- In conclusion, we showed in this study that MALDI-TOF MS is asites as well as a putative vector of Crimean-Congo hemorrhagic an efficient approach for the rapid identification of tick vectors. fever in certain regions of the world, such as Turkey (24). The This method was used for thefirst time to identify ticks removed potential for these ticks to transmit disease agents was illustrated from patients. The results were obtained rapidly relative to the in this study by the detection of several pathogens in the ticks that time required for molecular methods, and the completion of this had bitten patients. Thisfinding highlights the clinical need for the assay does not require any specific entomological expertise. One species identification of ticks. damaged tick removed from a patient was successfully identified

Our results suggest that the MALDI-TOF MS spectra of pro- using MALDI-TOF MS as I. ricinus, a vector of Lyme disease. The Downloaded from tein extracts from tick legs are a suitable tool for identifying ticks. knowledge of the tick species can be used to inform the clinician’s The results obtained using this method corroborated those from decision as to whether to prescribe prophylactic doxycycline treat- morphological and molecular identification methods. ment. At the very least, the identity of the tick species can tell Overall, 63% of the laboratory tick specimens were identified physicians which specific clinical signs they should look for in by MALDI-TOF MS with scores of�2, which are considered to be their patients. The rapid identification of ticks, and most likely of reliable scores for the identification of bacterial species (15). other arthropod vectors, is now possible in any laboratory with a Ticks collected in thefield or removed from patients were re- MALDI-TOF MS system. In our unit, results are now available for

liably identified with lower score values, but each specimen’s spec- clinicians in less than 1 h, with no requirement for entomological http://jcm.asm.org/ trum matched a reference spectrum. Only one tick was not iden- expertise. Our database contains reference spectra for ticks re- tified by MALDI-TOF MS, R. bursa (identification confirmed by moved from humans in our area. This database can be shared and 12S sequencing). This tick was not identified because the corre- used directly by any clinical microbiology laboratory equipped sponding spectrum was not present in our database. The species with a MALDI Biotyper system. Because MALDI-TOF studies not represented in our database, which were tested as controls with bacteria and yeast have demonstrated significant variation in (fleas, mosquitoes, bugs, bees, and beetles), had low scores, less the protein profile based on geographical region (27), it will be than 1.1, and none of their spectra matched the reference spectra interesting to test ticks of the same species from a variety of geo- in the database. graphical regions to determine if this technique can be used as a on October 24, 2014 by guest The establishment of a cutoff score for accurate identification regional or global tool. We will continue to add new reference would be ideal. When ticks removed from patients were tested, spectra to our database to test the ability of this MS method to most (76%) were identified with scores of�1.7, and all were iden- discriminate closely related species and to define a definitive cutoff tified with scores of�1.3. However, it is not known if a score of 1.3 score for species identification. Finally, it will be informative to can be used as a definitive cutoff in other regions. The ticks used to determine whether MALDI-TOF MS can be used to identify not construct the database are representative of the tick species in only tick vectors but also the microorganisms with which the ticks France and correspond to the species that have been removed are infected. from French patients, including returned travelers, and sent to our reference centers in the past 15 years (P. Parola, unpublished ob- ACKNOWLEDGMENTS servations). In addition, the vectors included in this study are not We thank Arnaud Canet for providing ticks collected in thefield, Nicolas closely related, and it must be confirmed that this MALDI-TOF Armstrong for technical assistance, and Guillaume Lacour for providing MS method can differentiate closely related species, e.g., Ixodes mosquito specimens. species. However, when D. marginatus specimens removed from patients were tested against the database, which contained refer- REFERENCES ence spectra for this species and for the closely related speciesD. 1. Parola P, Raoult D. 2001. Ticks and tickborne bacterial diseases in hu- reticulatus, the identification of the specimens was unequivocal. mans: an emerging infectious threat. Clin. Infect. Dis. 32:897–928. Using the MSP dendrogram function of MALDI Biotyper, ver- 1a.Hubalek Z, Rudolf I. 2012. Tick-borne viruses in Europe. Parasitol. Res. 111:9–36. sion 3.0, all but one tick (R. bursa) clustered according to their 2. Gray J, Zintl A, Hildebrandt A, Hunfeld KP, Weiss L. 2010. Zoonotic genera, species, and strain. A similar discrepancy was observed in babesiosis: overview of the disease and novel aspects of pathogen identity. a preliminary study using the entire body of the tick (14). There- Ticks Tick-Borne Dis.1:3–10. fore, although MALDI-TOF MS has opened new doors for the 3. Stanek G, Wormser GP, Gray J, Strle F. 2012. Lyme borreliosis. Lancet phylogenetic study of arthropods and bacteria, additional data are 379:461–473. 4. Parola P, Paddock CD, Raoult D. 2005. Tick-borne rickettsioses around still needed, and detailed studies should be conducted. the world: emerging diseases challenging old concepts. Clin. Microbiol. Contrary to a recent report (14), we found that the use of tick Rev. 18:719–756. legs appears to be an effective means to identify tick vectors if a 5. Balicer RD, Mimouni D, Bar-Zeev Y, Levine H, Davidovitch N, Ankol reference database is available. The remainder of the body can be OH, Zarka SS 2010. Post exposure prophylaxis of tick-borne relapsing fever. Eur. J. Clin. Microbiol. Infect. Dis. 29:253–258. reserved for other purposes, such as the detection of pathogens, as 6. Piesman J, Hojgaard A. 2012. Protective value of prophylactic antibiotic performed in this study. However, various storage conditions (10, treatment of tick bite for Lyme disease prevention: an animal model. Ticks 25) and chemical extraction methods (26) appear to influence the Tick-Borne Dis.3:193–196.

February 2013 Volume 51 Number 2 211 jcm.asm.org 527 Yssouf et al.

7. Norris DE, Klompen JS, Keirans JE, Black WC. 1996. Population ge- 19. Socolovschi C, Reynaud P, Kernif T, Raoult D, Parola P. 2012. Rick- netics of Ixodes scapularis (Acari: Ixodidae) based on mitochondrial 16S ettsiae of spotted fever group, Borrelia valaisiana, and Coxiella burnetii in and 12S genes. J. Med. Entomol. 33:78–89. ticks on passerine birds and mammals from the Camargue in the south of 8. Mangold AJ, Bargues MD, Mas-Coma S. 1998. 18S rRNA gene sequences France. Ticks Tick-Borne Dis.3:355–360. and phylogenetic relationships of European hard-tick species (Acari: Ixo- 20. Assous MV, Wilamowski A, Bercovier H, Marva E. 2006. Molecular didae). Parasitol. Res. 84:31–37. characterization of tickborne relapsing fever Borrelia, Israel. Emerg. In- 9. Song S, Shao R, Atwell R, Barker S, Vankan D. 2011. Phylogenetic and fect. Dis. 12:1740–1743. phylogeographic relationships in Ixodes holocyclus and Ixodes cornuatus 21. Skarphedinsson S, Jensen PM, Kristiansen K. 2005. Survey of tickborne (Acari: Ixodidae) inferred from COX1 and ITS2 sequences. Int. J. Parasi- infections in Denmark. Emerg. Infect. Dis. 11:1055–1061. tol. 41:871–880. 21a.Parola P, Rovery C, Rolain JM, Brouqui P, Davoust B, Raoult D. 2009. 10. Seng P, Rolain JM, Fournier PE, La Scola B, Drancourt M, Raoult D. Rickettsia slovaca and R. raoultii in tick-borne rickettsioses. Emerg. Infect. 2010. MALDI-TOF-mass spectrometry applications in clinical microbiol- Dis. 15(7):1105–1108. ogy. Future Microbiol.5:1733–1754. 21b.Platonov AE, Karan LS, Kolyasnikova NM, Makhneva NA, Toporkova 11. Feltens R, Gorner R, Kalkhof S, Groger-Arndt H, and von Bergen M. MG, Maleev VV, Fish D, Krause PJ. 2011. Humans infected with relaps- 2010. Discrimination of different species from the genus Drosophila by ing fever spirochete Borrelia miyamotoi, Russia. Emerg. Infect. Dis. 17: intact protein proﬁling using matrix-assisted laser desorption ionization 1816–1823. mass spectrometry. BMC Evol. Biol. 10:95. doi:10.1186/1471-2148-10-95. 22. Dumler JS. 2012. The biological basis of severe outcomes in Ana-

12. Perera MR, Vanstone VA, Jones MG. 2005. A novel approach to identify plasma phagocytophilum infection. FEMS Immunol. Med. Microbiol. Downloaded from plant parasitic nematodes using matrix-assisted laser desorption/ 64:13–20. ionization time-of-flight mass spectrometry. Rapid Commun. Mass Spec- 23. Rovery C, Raoult D. 2008. Mediterranean spotted fever. Infect. Dis. Clin. trom. 19:1454–1460. North Am. 22:515–530. 13. Kaufmann C, Schaffner F, Ziegler D, Pfluger V, Mathis A. 2012. 23a.Demma LJ, Traeger MS, Nicholson WL, Paddock CD, Blau DM, Ere- Identification offield-caught Culicoides biting midges using matrix- meeva ME, Dasch GA, Levin ML, Singleton J, Jr, Zaki SR, Cheek JE, assisted laser desorption/ionization time offlight mass spectrometry. Par- Swerdlow DL, McQuiston JH. 2005. Rocky Mountain spotted fever from asitology 139:248–258. an unexpected tick vector in Arizona. N. Engl. J. Med. 353:587–594. 14. Karger A, Kampen H, Bettin B, Dautel H, Ziller M, Hoffmann B, Suss J, Klaus C. 2012. Species determination and characterization of develop- 23b.Parola P, Socolovschi C, Jeanjean L, Bitam I, Fournier PE, Sotto A, Labauge P, Raoult D. 2008. Warmer weather linked to tick attack and

mental stages of ticks by whole-animal matrix-assisted laser desorption/ http://jcm.asm.org/ ionization mass spectrometry. Ticks Tick-Borne Dis.3:78–89. emergence of severe rickettsioses. PLoS Negl. Trop. Dis.2:e338. doi: 10 14a.Beati L, Keirans JE. 2001. Analysis of the systematic relationships among .1371/journal.pntd.0000338. ticks of the genera Rhipicephalus and Boophilus (Acari: Ixodidae) based on 24. Estrada-Pena A, Bouattour A, Camicas A, Walker AR. 2004. Ticks of mitochondrial 12S ribosomal DNA gene sequences and morphological domestic animals in the Mediterranean region: a guide to identification of characters. J. Parasitol. 87:32–48. species. University of Zaragoza, Zaragoza, Spain. 15. Fournier PE, Couderc C, Buffet S, Flaudrops C, Raoult D. 2009. Rapid 25. Kaufmann C, Ziegler D, Schaffner F, Carpenter S, Pfluger V, Mathis A. and cost-effective identification of Bartonella species using mass spec- 2011. Evaluation of matrix-assisted laser desorption/ionization time of trometry. J. Med. Microbiol. 58:1154–1159. flight mass spectrometry for characterization of Culicoides nubeculosus 16. Sauer S, Freiwald A, Maier T, Kube M, Reinhardt R, Kostrzewa M, biting midges. Med. Vet. Entomol. 25:32–38. 26. Carbonnelle E, Mesquita C, Bille E, Day N, Dauphin B, Beretti JL,

Geider K. 2008. Classification and identification of bacteria by mass spec- on October 24, 2014 by guest trometry and computational analysis. PLoS One3:e2843. doi: 10.1371 Ferroni A, Gutmann L, Nassif X. 2011. MALDI-TOF mass spectrometry /journal.pone.0002843. tools for bacterial identification in clinical microbiology laboratory. Clin. 17. Bechah Y, Socolovschi C, Raoult D. 2011. Identification of rickettsial Biochem. 44:104–109. infections by using cutaneous swab specimens and PCR. Emerg. Infect. 27. Cassagne C, Ranque S, Normand AC, Fourquet P, Thiebault S, Planard Dis. 17:83–86. C, Hendrickx M, Piarroux R. 2011. Mould routine identification in the 18. Parola P, Diatta G, Socolovschi C, Mediannikov O, Tall A, Bassene H, clinical laboratory by matrix-assisted laser desorption ionization time-of- Trape JF, Raoult D. 2011. Tick-borne relapsing fever borreliosis, rural flight mass spectrometry. PLoS One6(12):e28425. doi: 10.1371/journal Senegal. Emerg. Infect. Dis. 17:883–885. .pone.0028425.

528 jcm.asm.org 212 Journal of Clinical Microbiology

Mini review: Typhus in World War I

Microbiology today 41 (2), 68 – 71.

213

Ty p h u s in World War I

Fig. 1. Coloured electron micrograph of a bacterium of the genus Rickettsia. CNRI/Science Photo Library Rezak Drali, Philippe Brouqui & Didier Raoult

Epidemic typhus has always accompanied disasters striking he original description of typhus humanity. Famine, cold and wars are its best allies. Typhus, also is thought to have been made in T1546 by Fracastoro, a Florentine known as historical typhus, classic typhus, sylvatic typhus, red physician, in his treatise of infectious diseases: De contagione et contagiosis louse disease, louse-borne typhus, and jail fever has caused morbis. His observations during the mortality and morbidity through the centuries, and on the Eastern Italian outbreaks in 1505 and 1528 allowed him to separate typhus from Front during World War I it led to the death of thousands. the other pestilential diseases. It also recognised the transmission of typhus

68 Microbiology Today May 14 | www.sgm.ac.uk 215 (a)

(b)

from human-to-human. The term Fig. 2. R. prowazekii-infected (a) or uninfected (b) dead P. humanus humanus. The louse infected with ‘exanthematic typhus’ was introduced in R. prowazekii became red and developed rectal bleeding before dying. Reproduced from Houhamdi, L. & others 1760 by the French physician, Boissier (2002), J Infect Dis 186, 1639–1646; license no. 3337041496430; Oxford University Press de Sauvages. Thanks to PCR testing of dental pulp from ancient remnants Prize in 1928 for his fndings. Nicolle from typhus after confrming Ricketts’ of bodies from graves, we now have was able to transmit the typhus from observations. In 1916, Rocha Lima evidence that typhus and trench fever humans to chimpanzees and then to described the bacterium and named were involved in the decimation of the macaques through blood transmission it Rickettsia prowazekii in honour of besiegers of Douai, 1710–1712, during and, fnally, from macaque to macaque Ricketts and Prowazek. the War of the Spanish Succession, via a body louse. Body lice infected by R. prowazekii and afflicted the soldiers of Napoleon’s Between 1903 and 1908, Ricketts become red and die shortly thereafter Grand Army in Vilnius in 1812 after their identifed Rickettsia rickettsii (Fig. 1), the (Fig. 2). Humans are the principal retirement from Russia (Table 1). agent of spotted fever that is closely reservoir of typhus during outbreaks. In 1909, epidemic typhus was related to the agent of typhus. In 1910, However, a zoonotic reservoir of R. found to be transmitted by Pediculus he contracted typhus and died in Mexico prowazekii exists. In addition to the humanus humanus, the body louse, by while conducting his experiments. detection of antibodies against R. Charles Nicolle, and he received a Nobel In 1914, von Prowazek in turn died prowazekii in a wide range of domestic and wild animals, R. prowazekii was Table 1. Potential typhus outbreaks through the history of mankind isolated from the blood of Egyptian Period Outbreak Probability Other possible disease donkeys and from the spleens, ﬂeas and lice of the ﬂying squirrel in Florida, USA. 15th century Conquest of Granada Likely R. prowazekii was also isolated from 16th century Mexico Likely Smallpox Hyalomma spp. ticks recovered from 16th century Hungarian disease Likely livestock in Ethiopia and Amblyomma 1710–1712 War of Spanish Succession Proven Trench fever spp. ticks in Mexico. (France, Europe) A typhus outbreak requires the 1812 Napoleonic Wars Proven Trench fever occurrence of both body louse outbreak (Vilnius, Eastern Europe) and a case of bacteraemic typhus 1914–1918 World War I (Russia, Europe) Proven Trench fever (Brill–Zinsser disease or epidemic 1917–1925 Bolshevik Revolution (Russia) Proven Other louse-borne typhus) (Fig. 3). These two conditions are often combined in wartime, where diseases stress, lack of hygiene and non-changing 1940–1945 World War II (Europe, North Africa) Proven Trench fever of clothes during the winter months are 1997 Burundi Civil War (Central Africa) Proven Trench fever common.

Microbiology Today May 14 | www.sgm.ac.uk 69 216 “ “ lesions on the skin. R. prowazekii enters to the extremities (Fig. 4). The rash, Epidemic typhus is an the skin through these lesions or by which may be hard to see in darker- the contamination of conjunctivae or skinned individuals, except in the axilla, unpredictable disease that can mucous membranes with louse faeces is classically described as sparing the containing rickettsiae. Infection through palms and soles. Gangrene and necrosis suddenly re-emerge when social the aerosols of faeces-infected dust has of toes and fngers that necessitates also been reported and is the major risk amputation has been observed. organisation is disrupted among physicians. Neurologic symptoms include confusion and drowsiness. Coma, seizures and Clinical manifestations of epidemic focal neurologic signs may develop in a Body louse outbreak typhus minority of patients. The mortality rate The body louse is a blood-sucking Epidemic typhus is a life-threatening, varies from 0.7 to 60% for untreated ectoparasite, specifc to humans, that acute exanthematic feverish disease cases, depending on the age of the lives and multiplies in clothing. During that is primarily characterised by the patient, with a case fatality ratio lower its life cycle of approximately 35 days, abrupt onset of fever with painful than 5% in patients less than 13 years the female louse lays an average of 200 myalgia, a severe headache, malaise old. In self-resolving cases, R. prowazekii eggs, which can increase the number and a rash. Non-specifc symptoms can persist for life in humans, and under “of lice from a few to thousands on sometimes include a cough, abdominal stressful conditions recrudescence may the same individual. The body louse pain, nausea and diarrhoea. The rash occur as a milder form of Brill–Zinsser ingests an average of fve meals a day, that is characteristic of epidemic typhus disease. R. prowazekii bacteraemia generating extremely dry dejections. It classically begins a few days after the occurs in Brill–Zinsser disease so it can injects various substances when biting onset of symptoms, appearing as a red initiate an outbreak of epidemic typhus that cause itching, compelling the host macular or maculopapular eruption on when body lice are present on the to scratch vigorously thereby generating the trunk that later spreads centrifugally infected individuals.

(1) (2) (3) Body lice infection R. prowazekii-infected body lice infestation

Wars Crowding, lack of hygiene Brill–Zinsser and cold and no clothing change Body lice outbreak Disease Typhus outbreak

Fig. 3. Outbreak of epidemic typhus. When social organisation is disrupted, (1) body louse outbreak occurs among defenceless populations and (2) the presence of a case of Brill–Zinsser can initiate (3) an outbreak of Fig. 4. Diffuse petechial rash of epidemic typhus. epidemic typhus. Credit Credit

70 Microbiology Today May 14 | www.sgm.ac.uk 217 relapsing fever and caused by Borrelia recurrentis.

Conclusion Epidemic typhus is an unpredictable disease that can suddenly re-emerge when social organisation is disrupted, as Fig. 5. Soldiers’ kit bags being placed into gas chambers to be deloused during World War I. Otis Historical was observed in 1997 among Burundi’s Archives, National Museum of Health and Medicine/Science Photo Library Civil War refugees in central Africa. Wars are optimal conditions for body Typhus in the First World War standards of hygiene among the troops louse proliferation and their associated The declaration of war by Austria to prevent body lice infestations (Fig. 5). diseases. Thus, the control of lice with against Serbia in 1914 following the On the Western Front, although the combination of oral Ivermectin, assassination of Archduke Ferdinand body lice were also endemic among clean clothes and insecticides will help quickly expanded into an uncontrollable the troops, there was no outbreak of to avoid disasters caused by typhus, global conﬂict in World War I. typhus. The situation lacked the R. trench fever and relapsing fever during On the Eastern Front, intense prowazekii bacteraemia to trigger a humanitarian catastrophes. shelling of Serbian cities destroyed typhus epidemic, as had happened on the existing infrastructure and drove the Eastern Front. Another disease, Rezak Drali, Philippe Brouqui & the population to the streets, and at described for the frst time and also Didier Raoult least 20,000 Austrians were taken vectored by the body louse, was raging Aix Marseille Université, URMITE, UM63, prisoner by the Serbs. There was a in the trenches among the troops. It CNRS 7278, IRD 198, Inserm 1095, 13005 lack of physicians and other medical is caused by the bacterium Bartonella Marseille, France and Institut Hospitalo- professionals because they had been quintana and was named trench fever. Universitaire Méditerranée Infection, seconded to the army, which led to the On the Russian front, throughout the 13005, Marseille, France; Corresponding rapid collapse of the health status of last two years of the conﬂict and during author Tel. +33 491 32 43 75 defenceless populations. Malnutrition, the Bolshevik revolution, approximately [email protected] overcrowding and a lack of hygiene 2.5 million deaths were recorded. paved the way for typhus. In November Typhus was latent in Russia long before Further reading 1914, typhus made its frst appearance the beginning of World War I. The Bavaro, M. F. & others (2005). History of US among refugees and prisoners, and it mortality rate rose from 0.13 per 1,000 military contributions to the study of Rickettsial then spread rapidly among the troops. in peacetime to 2.33 per 1,000 in 1915. diseases. Mil Med 170 (Suppl. 1), 49–60. One year after the outbreak of hostilities, Soldiers and refugees imported typhus Bechah, Y. & others (2008). Epidemic typhus. typhus killed 150,000 people, of whom and propagated it across the country. Lancet Infect Dis 8, 417–426. 50,000 were prisoners in Serbia. A third It was during the hard winter of 1917– Raoult, D., Woodward, T. & Dumler, J. S. of the country’s doctors suffered the 1918 that the biggest outbreak of typhus (2004). The history of epidemic typhus. Infect same fate. The mortality rate reached in modern history began in a Russia that Dis Clin N Am 18, 127–140. an epidemic peak of approximately 60 to was already devastated by famine and Tschanz, D. W. (2008). Typhus Fever on the 70%. This dramatic situation dissuaded war. The great epidemic started in the Eastern Front in World War I. Insects, Disease the Germano-Austrian commandment big cities and eventually reached the and History Website, Entomology Group of from invading Serbia in an attempt to distant lands of the Urals, Siberia and Montana State University, microb.io/1kLupfa prevent the spread of typhus within their Central Asia. (accessed 18 January 2008). borders. Drastic measures were taken, After World War I, between 1919 Zinsser, H. (1935). Rats, lice and history: A such as the quarantine of people with and 1923, there were fve million deaths chronicle of pestilence and plagues. New York: the frst clinical signs of the disease, in Russia and Eastern Europe because Black Dog and Leventhal Publishers. but attempts were also made to apply of a third disease vectored by body lice,

Microbiology Today May 14 | www.sgm.ac.uk 71 218 Remerciements

Je commencerai par remercier le Professeur Didier RAOULT pour avoir accepté de m’accueillir dans son laboratoire. Je suis honoré et extrêmement fier d’avoir travaillé sous votre direction. Permettez-moi de vous exprimer mon profond respect.

Je remercie le Professeur Philippe BROUQUI, mon directeur de thèse et avant cela mon responsable de Master. Je vous exprime ma gratitude pour m’avoir dirigé, conseillé et aidé à mener à terme cette thèse.

Merci au Professeur Jean-Marc ROLAIN. J’ai beaucoup apprécié d’avoir travaillé avec vous. Vous avez toujours répondu présent à mes sollicitations.

Je remercie tous mes co-auteurs que je ne pourrais citer individuellement car et ils sont nombreux.

Merci au Docteur Idir BITAM. Vos encouragements et vos conseils m’ont toujours poussé à aller de l’avant. J’espère que l’avenir nous réserve une longue et fructueuse collaboration.

Je remercie mes rapporteurs d’avoir accepté d’évaluer ce travail.

Je remercie toute l’équipe de l’URMITE et particulièrement tous les membres avec qui j’ai été amené à travailler.

Enfin, je remercie vivement l’Institut Pasteur d’Algérie, l’institution à laquelle j’appartiens qui m’a permis d’entreprendre et mener à terme cette thèse.

219