lable at ScienceDirect

Microbial Pathogenesis xxx (2014) 1e5

Contents lists avai

Microbial Pathogenesis

journal homepage: www.elsevier .com/locate/micpath

Looking at protists as a source of pathogenic viruses

Bernard La Scola a, b, *

a Aix-Marseille University, Unit�e de Recherche sur les Maladies Infectieuses et Tropicales Emergentes (URMITE), UM63, CNRS 7278, IRD 198, INSERM U1095,Facult�es de M�edecine et de Pharmacie, Marseille, Franceb Institut Hospitalo-Universitaire (IHU) M�editerran�ee Infection, Pole des Maladies Infectieuses et Tropicales Clinique et Biologique, F�ed�eration deBact�eriologie-Hygi�ene-Virologie, Centre Hospitalo-Universitaire Timone, Assistance Publique, Hopitaux de Marseille, Marseille, France

a r t i c l e i n f o

Article history:Received 4 September 2014Accepted 5 September 2014Available online xxx

Keywords:ProtozoaGiant virusMimivirusMarseillevirusNCLDV

* Pole des Maladies Infectieuses, Assistance PubliquAix Marseille Universite, URMITE, UM63, CNRS 7278, Ide M�edecine, 27 Bd Jean Moulin, 13385 Marseille ced

E-mail addresses: bernard.lascola@univ-amu.fr, las

http://dx.doi.org/10.1016/j.micpath.2014.09.0050882-4010/© 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: La Scoladx.doi.org/10.1016/j.micpath.2014.09.005

a b s t r a c t

In the environment, protozoa are predators of bacteria and feed on them. The possibility that someprotozoa could be a source of human pathogens is consistent with the discovery that free-living amoebaewere the reservoir of Legionella pneumophila, the agent of Legionnaires' disease. Later, while searching forLegionella in the environment using amoeba co-culture, the first giant virus, Acanthamoeba polyphagamimivirus, was discovered. Since then, many other giant viruses have been isolated, including Marseil-leviridae, Pithovirus sibericum, Cafeteria roenbergensis virus and Pandoravirus spp. The methods used toisolate all of these viruses are herein reviewed. By analogy to Legionella, it was originally suspected thatthese viruses could be human pathogens. After showing by indirect evidence, such as sero-epidemiologicstudies, that it was possible for these viruses to be human pathogens, the recent isolation of some ofthese viruses (belonging to the Mimiviridae and Marseilleviridae families) in humans in the context ofpathologic conditions shows that they are opportunistic human pathogens in some instances.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

In the environment, protozoa are predators of bacteria and feedon them. The most studied of these protists as potential vectors ofpathogens are free-living amoebae. Free-living amoebae (FLA) areubiquitous unicellular eukaryotes that have been isolated world-wide from soil, water, air and even vertebrates. In the environment,they colonize natural biofilms, grazing on them to search for bac-teria, their major sources of nutrients. In fact, free-living amoebaeare natural phagocytes that feed on large particles (>0.5 mm) in theextracellular environment, independently of a recognition systemand based only on particle size [1]. They supposedly do not feed onsmaller particles that are able to pass through a 0.20 mm filter [2];however, as some may phagocytose small viruses (<0.2 mm), it ispossible that this phagocytosis may be mediated by specificrecognition in some instances.

Historically, the search for pathogens in protozoa followed thediscovery of Legionella pneumophila, the agent of Legionnaires'disease in humans. After that, TJ Rowbotham used FLA as a supportto isolate Legionella spp. from the environment, especially cooling

e-Hopitaux de Marseille andRD 198, INSERM 1095, Facult�eex 05, France.colab@yahoo.fr.

B, Looking at protists as a s

towers, the most common environmental reservoir of Legionella atthe origin of epidemics. Using the procedure developed for thispurpose, TJ Rowbotham and others could isolate Legionella frominfected human samples, including sputum and stool [3,4]. Inaddition to Legionella, several bacterial species were also isolatedand named amoeba-resisting microorganisms (ARMs). These ARMsconsist mainly of bacteria that belong to various phylogeneticclades dispersed throughout the prokaryotic tree [5], and amongthese facultative and obligate intracellular species, some are humanpathogens. However, while searching for ARMs, a giant viruses wasdiscovered, resembling bacteria in its size. This first one wasAcanthamoeba polyphaga mimivirus. After an inaugural work on thesurvival of Coxiella burnetii in Acanthamoeba castellanii [6], our in-terest in amoeba-associated organisms continued when Richard J.Birtles took up a post-doctoral fellowship in the laboratory,bringing along a collection of obligate intra-amoebal bacterialparasites. The parasites were recovered by Dr Tim Rowbotham overa period of nearly twenty years from environmental water samplescollected as part of the Legionnaires' disease outbreak in-vestigations by his employer, the Public Health Laboratory Serviceof England and Wales. Most of these bacteria were referred to as“Legionella-like amoebal pathogens” (LLAPs) [7]. In addition tocultures of LLAPs were two cultures of apparently Gram-positivecoccoid bacteria, referred to as “the Bradford coccus” and “Hall'scoccus”, the latter of which had been sent to Dr Rowbotham by a

ource of pathogenic viruses, Microbial Pathogenesis (2014), http://

B. La Scola / Microbial Pathogenesis xxx (2014) 1e52

colleague in the USA and later identified as a new genus, Para-chlamydia acanthamoeba [8].

However, all of the attempts to amplify 16S rDNA from theBradford coccus failed despite the use of PCR assays that incorpo-rated different sets of “universal” pan-bacteria primers. To searchfor a particular cell wall structure that could explain the inefficiencyof our DNA extraction protocols, we decided to examine the ul-trastructure of the Bradford coccus by electron microscopy [9]. Toour great surprise, we observed icosahedral particles resemblinggiant Iridoviruses within infected amoebae. The suspicion of theviral origin of the Bradford coccus was confirmed by further pre-liminary work, in which we showed that it contained a largedouble-stranded DNA chromosome coding for typical viral genesand underwent an eclipse-phase replication typical of viruses.Furthermore, we now know that the assembly of virus particlestakes place in specific intracellular locations, which have beentermed “virus factories” when previously observed for viruses,including Iridoviruses [10]. A particle diameter of 600 nm and agenome size of 1.2 Mb made this virus the largest known ever andthe first member of the giant viruses, later called giruses. Phylogenyindicated that it groups with other nucleocytoplasmic large DNAviruses (NCLDVs) including the Iridoviridae, Baculoviridae, Phy-codnaviridae and Poxviridae. After this seminal work and theincidental isolation of Mimivirus, the search for viruses of protistsbegan, a strategy developed by our laboratory that was then fol-lowed by several other laboratories across the world.

In all of those cases, protist-associated viruses have been iso-lated according to two different strategies. The most commonlyused strategy is to reproduce the isolation of Mimivirus by inocu-lating samples on axenic FLA using amoebae as a support for cul-ture, exactly as is done with the culture of intracellular bacteriawith cell lines. In that case, samples (human or environmental) areinoculated on an amoebal monolayer cultivated in buffer with an-tibiotics that prevent the multiplication of bacteria, and the cul-tures are examined to detect amoeba lysis, indicative of an amoebapathogen that can potentially be a giant virus. In such cases,amoebae from the genus Acanthamoeba, A. polyphaga andA. castellanii have been primarily used, but recently, other protozoasuch as Harmanella vermiformis have been tentatively used (un-published data). The second strategy consists of isolating protistsand then searching for the presence of giant viruses developing inthem. This strategy allows the isolation of unique species of giantviruses although it is less commonly used. These different strate-gies are analyzed here with a specific view on isolates fromhumans.

2. Isolation of giant viruses using amoeba co-cultures

The first isolate was Acanthmoeba polyphaga mimivirus, and itwas isolated by the original method described by TJ Rowbotham toisolate Legionella sp [4]. In this method, isolationwas performed onA. polyphaga strain AP L1501/3A. This strain was chosen because itcould be produced axenically in PYG medium and because it is aslow-growing amoeba. The speed of growth is important becausethe relative number of Legionella and amoebae is critical to observeLegionella growth. If there are too many amoebae, they can encystbefore their lysis due to Legionella growth is detected, and if thereare too few amoebae, the growth of Legionella may be missed. Thestrain A. polyphaga (Linc-AP1) we used for propagating Mimivirusand for isolating giant viruses was provided by TJ Rowbotham to ustwenty years ago, and it is thought to be comparable to strain APL1501/3A. We replaced our co-culture on shell vials with a 6- to 12-well microplate system, allowing us to test more samples at thesame time [11e13]. Due to the massive multiplication of Mim-iviruses and Marseilleviruses, we do not believe that this relative

Please cite this article in press as: La Scola B, Looking at protists as a sdx.doi.org/10.1016/j.micpath.2014.09.005

amount of viruses to amoeba is critical to isolate Mimiviruses.However, we showed that the multiplication of APMV is dependenton the virus/amoeba ratio [14]. Our study suggested that a lowvirus/amoeba ratio was more efficient for the production of infec-tious particles; at a virus/amoeba ratio (virus in TCID50) of a factorof 0.01, Mimivirus was more productive than at any other ratio.However, we believe that in natural conditions, this ratio is smaller.Therefore, in our recent procedures, we performed two blind sub-cultures on the amoebae before searching for amoeba lysis [13]. Inour first studies searching for amoeba pathogens, after inoculatinghundreds of environmental samples without a mix of antibioticscontaining vancomycin and colistin, we isolated almost exclusivelybacteria [15,16]. We isolated few giant viruses, including the secondMimivirus isolate shown to be infected by a virophage, which wenamed Mamavirus, and the first Marseillevirus isolate [17,18]. Ourcolleague G. Greub, who was searching mostly for Parachlamydia,had the same experience and isolated a unique giant virus, whichrepresents of a new lineage of Marseillevirus he named Lausannevirus [19].

In order to increase our chances of isolating viruses, we modi-fied our culture strategy by isolating and testing the antibioticsusceptibility of the contaminating bacteria [20]. This allowed us toisolate 19 giant viruses from 105 environmental samples, includingMarseilleviruses and Mimiviruses, for an isolates/samples rationever obtained at such a level since then. This strategy allowed us,for example, to identify the presence of different lineages withinMimiviruses, A, B and C, with one of these lineages correspondingto a Mimivirus later named Megavirus [21]. This strategy alsoallowed us to isolate a second virophage, which was later demon-strated to be integrated in the Mimivirus genome and which wouldbe named later as a provirophage [22]. The major difficulty inobtaining this high isolates/samples ratio was clearly the fine detailof the procedure. To avoid the meticulous observation of micro-plates searching for amoeba lysis, we added a blind enrichment andsearched for lysis on agar plates [23]. After 3 days of culture inamoeba without antibiotics, the supernatant was subcultured to amicroplate with antibiotics. Then, after 2 days, the supernatant wasinoculated as a drop on an agar plate seeded with a monolayer ofA. polyphaga. The virus was detected by observing a clear areacorresponding to amoeba lysis. This procedure allowed us to isolate11 new Marseilleviridae strains and four new Mimiviridae strains.This procedure was thus shown to be highly efficient, as it allowedthe isolation of the first Mimiviruses of human origin [24,25], but ithas some limitations, the major being that it cannot be used withhighly motile protozoa. Other research groups used this procedureto discover additional isolates of Marseilleviruses and Mimiviruses,the last being isolated in Brazil [26]. With now more than 7000inoculated samples, we have isolated 43 strains of Mimiviridae and17 strains of Marseilleviridae [13], whereas at the same time,another team isolated and described two new virus families, Pan-doravirus and Pithovirus, using A. castellanii as a support forisolating amoeba-associated viruses [27,28]. Although these virusesseem to be mostly environmental, a recent work suggests thathumans are in contact with them through contact lens-associatedkeratitis [29]. The team of JM Claverie first isolated Megavirus chi-lensis, which corresponds to the Mimiviridae we identified aslineage C. It was isolated from sea water in Chile [21]. In their work,they supplemented 1 L of sea water with 4% rice. The mixture wasthen incubated at room temperature in the dark to allow the het-erotrophic bacteria to grow and to be fed on by protozoa, allowingthe protozoa to expand and thus increase the viral population. Afterfiltration, the membrane was mixed with antimicrobial agents(penicillin, gentamicin, streptomycin, and fungizone) for 3 daysthen inoculated on several Acanthamoeba species in microplates. Itis not clear if the membranewas inoculated on A. castellanii, but the

ource of pathogenic viruses, Microbial Pathogenesis (2014), http://

B. La Scola / Microbial Pathogenesis xxx (2014) 1e5 3

experiments was later performed in this species. The same teamisolated Pandoraviruses on A. castellanii from sediments [28]. In thiswork, they replaced gentamicin with amoxicillin, concentratedliquid samples through PEG precipitation, and adapted theirA. castellanii strain to increasing concentrations of amphotericin B(up to 5 mg/ml) and to a new medium called PPYG/NaCl. Therationale for changing the medium was not indicated by the au-thors. From these works with the isolation of two new virus fam-ilies, A. castellanii could be a more permissive amoeba thanA. polyphaga, which is efficient for isolating Mimiviruses and Mar-seilleviruses but could fail to isolate Pandora or Pithovirus families.In some instances, weworked with A. castellanii for viruses isolatedfrom A. polyphaga but that grew better on A. castellanii, such asMamavirus [18,30]. Following the concept of isolating new giantviruses associated with protozoa, we inoculated 5 never-testedgenera of protozoa to isolate new giant virus families; flow cy-tometer detection was used to detect protozoa lysis using a tech-nique derived from one used to detect marine viruses [31]. UsingHartmannella vermiformis,wewere able to isolate 7 viruses of a newvirus family that are currently under characterization in our labo-ratory (unpublished data). This isolation shows that this strategycould be efficient to isolate and describe new virus families in thefuture.

3. Isolation of protozoa infected by viruses

Cafeteria roenbergensis virus (CroV) is the only protozoa virus todate to have been discovered by isolation in association with itsnatural host [32]. The procedure consisted of using an enrichedseawater medium and adding yeast extract to promote bacterialgrowth. Cultures were then kept at room temperature in the dark,allowing protozoa to grow. One culture of CroV even showed aprogressive loss of virulence that was identified later as the seconddescribed virophage, named Mavirus [33]. This strategy is mostlikely the more promising regarding the perspective of isolatingnew viral families associated with protozoa but is currently highlydifficult to perform. However, the procedure seems to be poorlyamenable to high-throughput isolation due to the need for rela-tively large volumes of cultures.

4. Human pathogens from protozoa

The first arguments about the human pathogenicity of protozoaviruses were based on four sero-epidemiological studies. The firstone tested 887 samples from Canadian patients, including 376 withcommunity-acquired pneumonia and 511 healthy controls. Anti-bodies against APMV were present at higher levels in patients withpneumonia compared with the controls (9.66%e2.3%, respectively)[34]. Patients with pneumonia and antibodies to APMV morefrequently experienced a second hospitalization. The second studyconducted in an ICU also showed a higher prevalence of antibodiesto APMV in patients with pneumonia (19.2%) when compared withthe controls, which were all negative [34]. The third study analyzedthe sera from 157 patients admitted to ICUs for 210 cases ofpneumonia over 18 months [35]. This work showed that 19% ofpneumonia patients were exposed to several amoeba-associatedmicroorganisms simultaneously, including five exposures toAPMV. In 2009, a prospective study followed a cohort of 300 pa-tients with mechanical ventilation and observed the appearance ofantibodies to APMV in 59 patients [36]. A unique study not per-formed by our laboratory was performed in a reference center forrespiratory insufficiency in Holland. It showed that among 118patients, 3 had antibodies to APMV, including one with associatedexacerbation of respiratory symptoms and elevation of antibodytiters between two serial samples [37]. We have nevertheless

Please cite this article in press as: La Scola B, Looking at protists as a sdx.doi.org/10.1016/j.micpath.2014.09.005

observed a decrease in seroprevalence in our latest studies. Weformed the hypothesis that our APMV antigen was modified withsuccessive sub-cultures made for antigen production. This hy-pothesis was confirmedwhenwe showed that iterative subcultureswere associated with a genome reduction in our strain and asso-ciated with the loss of surface proteins involved in the formation offibrils [38].

In addition to these seroepidemiologic studies, molecularstudies attempting to detect viruses of protozoa were performed.DNA from APMV was detected in the bronchoalveolar lavage (BLA)of a patient in our first study [34], and then in our more recentworks in BLA and stool samples [24,25]. However, several studiesdid not detect APMV DNA in respiratory samples [39e42]. We havesome doubt regarding a possible contamination in our amplifica-tions, but a recent metagenomic study showed the presence ofAPMV DNA in naso-pharyngeal aspirates from patients with severepneumonia [43]. There are several explanations for the discrepancybetween studies. In the study of Larcher C. et al, 491 respiratorysamples were tested, mostly nasal, naso-pharyngeal and oro-pharungeal, along with 5 BLAs and endotracheal aspirates. In thestudy by Dare R.K. et al, only naso-pharyngeal aspirates werestudied, as Vanspauwen M.J. et al. tested sputum samples in theirfirst work. If samples from upper respiratory airways are bestadapted to “conventional” viruses that colonize the upper respi-ratory epithelium before lung invasion, they are likely not adaptedto viruses of protozoa such as APMV because these viruses infectprofessional phagocytes through phagocytosis, similar to bacteria[44,45]. Therefore, in most studies, BLAs were rarely tested, andfinally, APMV detectionwas positive only in two studies performedon BLAs. Moreover, the primers used to detect APMV were latershown, after isolation and analysis of a larger collection of APMV, tobe far from “universal” APMV primers due to high nucleotidicpolymorphisms [9,46].

Evidence of APMV infections was in fact obtained from alaboratory-acquired infection and from recent direct isolation ofthe virus. After the first isolation of APMV, APMVwas considered tobe an environmental non-pathogen. In 2004, a technician in ourlaboratory manipulating large amounts of APMV developed sub-acute pneumonia with cough, fever and thoracic pain [47]. He didnot improve with antibiotic therapy but recovered slowly after 2weeks of evaluation. All of the tests for an etiologic agent werenegative, but we observed that he exhibited seroconversion toAPMV. Two-dimensional Western blots confirmed seroconversionwith reactivity against 23 proteins of APMV. Finally, wewere able toisolate APMV from patients with pneumonia. The first humanisolate of APMVwas LBA111 from the BLA of a Tunisian patient withpneumonia [24]. Another APMV, Shanvirus, was isolated from thestool of a Tunisian patient with pneumonia [25]. This isolation islogical for this highly resistant virus [48,49] and is analogous to theisolation of two other phagocyte pathogens responsible for pneu-monia, L. pneumophila andMycobacterium tuberculosis [3,50]. Thesestrains belong to lineage C of Mimiviridae as Megavirus chiliensis[51]. These isolations were the results of the high-throughputtechniques detailed above.

Mimiviridae are not the only protozoa viruses isolated fromhuman. The first isolate ofMarseilleviridaewas Senegalvirus, whichwas isolated from the stool of a healthy Senegalese man usingamoeba co-culture [52]. The second isolate close to Marseilleviruswas isolated from human blood and named GBM for “giant bloodMarseillevirus” [53]. First, the genome of a Marseilleviridae viruswas detected from a healthy blood donor by using pyrosequencingon DNA extracted from the donor's blood. Among the 20,238 ob-tained reads (NCBI access number: PRJNA183996), 67.7% were fromviruses. Most had homology with Anelloviridae, especially Torqueteno virus (TTV) and TTV-like virus, SEN virus, and TTV midi virus

ource of pathogenic viruses, Microbial Pathogenesis (2014), http://

B. La Scola / Microbial Pathogenesis xxx (2014) 1e54

and TTV-like mini virus. However, 2.5% of the reads were related toMarseillevirus. After assembly, the 2 obtained contigs were map-ped on two distinct regions of Marseillevirus corresponding topositions 111,593 and 125,241 (13,649 bp) and positions 210,367and 220,539 (10,173 bp). PCRs targeting Marseillevirus were posi-tive on this blood. The virus-laden fraction of this patient's serumwas concentrated using 3 Marseillevirus-specific monoclonal an-tibodies in order to perform additional pyrosequencing (NCBI ac-cess number: PRJNA185405). Reads were assembled and, takentogether, allowed reconstructing a consensus sequence of357 433 bp with high homology to Marseillevirus (436/617 codingsequences with homology to Marseillevirus and 3/6173 with ho-mology to Lausannevirus). Transmission electronic microscopyallowed observing particles with viral morphology and a sizecompatible with that of Marseilleviridae (216.7 ± 4.7 nm). In situhybridization using a specific probe detected a particle with anidentical size. As GBM was detected in the blood donor, a furtherstudy compared the frequency of Marseillevirus DNA in blood do-nors compared with that of a patient with thalassemia andrepeated blood transfusions [54]. The DNA of Marseillevirus wasdetected in 4% of blood donors compared with 9.1% of patients withthalassemia, suggesting transmission through transfusions and itsoccurrence in healthy humans. The same result was observed byserology, as 22.7% of patients with thalassemia had antibodies toMarseillevirus compared with 12.6% of blood donors. Finally, thesame virus was detected by PCR, fluorescence in situ hybridization,and immunohistochemistry in an 11-month-old child with adenitis[55].

5. Conclusion

A decade after the description of Mimivirus, the first giant virus,the pathogenicity of protozoa-associated viruses is only beginningto be understood. Future experiments will have to be performed todefine their overall pathogenicity for humans, including searchingfor new genera of these viruses in the human environment usingnot only Acanthamoeba sp. but also other human-associated pro-tozoa as a support for co-culture.

References

[1] Aderem A, Underhill DM. Mechanisms of phagocytosis in macrophages. Annu.Rev. Immunol. 1999;17:593e623.

[2] Audic S, Robert C, Campagna B, et al. Genome analysis of Minibacteriummassiliensis highlights the convergent evolution of water-living bacteria. PLoSGenet. 2007 Aug 24;3(8):e138.

[3] Rowbotham TJ. Isolation of Legionella pneumophila serogroup 1 from humanfeces with use of amebic cocultures. Clin. Infect. Dis. 1998;26:502e3.

[4] Rowbotham TJ. Isolation of Legionella pneumophila from clinical specimens viaamoebae, and the interaction of those and other isolates with amoebae. J. Clin.Pathol. 1983;36:978e86.

[5] Moliner C, Fournier PE, Raoult D. Genome analysis of microorganisms living inamoebae reveals a melting pot of evolution. FEMS Microbiol. Rev. 2010May;34(3):281e94.

[6] La Scola B, Raoult D. Survival of Coxiella burnetii within free-living amoebaeAcanthamoeba castellanii. Clin. Microbiol. Infect. 2001;7(2):75e9.

[7] Birtles RJ, Rowbotham TJ, Raoult D, Harrison TG. Phylogenetic diversity ofintra-amoebal legionellae as revealed by 16S rRNA gene sequence compari-son. Microbiology 1996;142:3525e30.

[8] Birtles RJ, Rowbotham TJ, Storey C, Marrie TJ, Raoult D. Chlamydia-like obligateparasite of free-living amoebae. Lancet 1997;349(9056):925e6.

[9] La Scola B, Audic S, Robert C, et al. A giant virus in amoebae. Science 2003 Mar28;299(5615):2033.

[10] Raoult D, Audic S, Robert C, et al. The 1.2-megabase genome sequence ofMimivirus. Science 2004 Nov 19;306(5700):1344e50.

[11] La Scola B, Mezi L, Weiller PJ, Raoult D. Isolation of Legionella anisa using anamoebal coculture procedure. J. Clin. Microbiol. 2001;39(1):365e6.

[12] La Scola B, Barrassi L, Raoult D. Isolation of new fastidious a Proteobacteriaand Afipia felis from hospital water supplies by direct plating and amoebal co-culture procedures. FEMS Microbiol. Ecol. 2000;34:129e37.

[13] Pagnier I, Reteno DG, Saadi H, et al. A decade of improvements in Mimiviridaeand Marseilleviridae isolation from amoeba. Intervirology 2013;56(6):354e63.

Please cite this article in press as: La Scola B, Looking at protists as a sdx.doi.org/10.1016/j.micpath.2014.09.005

[14] Abrahao JS, Boratto P, Dornas FP, et al. Growing a giant: evaluation of thevirological parameters for mimivirus production. J. Virol. Methods 2014Oct;207C:6e11.

[15] Pagnier I, Raoult D, La Scola B. Isolation and identification of amoeba-resistingbacteria from water in human environment by using an Acanthamoeba poly-phaga co-culture procedure. Environ. Microbiol. 2008 May;10(5):1135e44.

[16] Evstigneeva A, Raoult D, Karpachevskiy L, La Scola B. Amoeba co-culture of soilspecimens recovered 33 different bacteria, including four new species andStreptococcus pneumoniae. Microbiology 2009 Feb;155(Pt 2):657e64.

[17] Boyer M, Yutin N, Pagnier I, et al. Giant Marseillevirus highlights the role ofamoebae as a melting pot in emergence of chimeric microorganisms. Proc.Natl. Acad. Sci. U. S. A. 2009 Dec 22;106(51):21848e53.

[18] La Scola B, Desnues C, Pagnier I, et al. The virophage as a unique parasite of thegiant mimivirus. Nature 2008 Sep 4;455(7209):100e4.

[19] Thomas V, Bertelli C, Collyn F, et al. Lausannevirus, a giant amoebal virusencoding histone doublets. Environ. Microbiol. 2011 Jun;13(6):1454e66.

[20] La Scola B, Campocasso A, N'Dong R, et al. A burden of new giant viruses fromthe environment and their tentative characterization by MALDI-TOF mass-spectrometry. Intervirology 2010;53:344e53. Karger ed.

[21] Arslan D, LeGendre M, Seltzer V, Abergel C, Claverie JM. Distant Mimivirusrelative with a larger genome highlights the fundamental features of Mega-viridae. Proc. Natl. Acad. Sci. U. S. A. 2011 Oct 18;108(42):17486e91.

[22] Desnues C, La Scola B, Yutin N, et al. Provirophages and transpovirons as thediverse mobilome of giant viruses. Proc. Natl. Acad. Sci. U. S. A. 2012 Oct30;109(44):18078e83.

[23] Boughalmi M, Saadi H, Pagnier I, et al. High-throughput isolation of giantviruses of the Mimiviridae and Marseilleviridae families in the Tunisianenvironment. Environ. Microbiol. 2013 Jul;15(7):2000e7.

[24] Saadi H, Pagnier I, Colson P, et al. First isolation of Mimivirus in a patient withpneumonia. Clin. Infect. Dis. 2013 Aug;57(4):e127e34.

[25] Saadi H, Reteno DG, Colson P, et al. Shan virus: a new mimivirus isolated fromthe stool of a Tunisian patient with pneumonia. Intervirology 2013;56(6):424e9.

[26] Campos RK, Boratto PV, Assis FL, et al. Samba virus: a novel mimivirus from agiant rain forest, the Brazilian Amazon. Virol. J. 2014;11:95.

[27] LeGendre M, Bartoli J, Shmakova L, et al. Thirty-thousand-year-old distantrelative of giant icosahedral DNA viruses with a pandoravirus morphology.Proc. Natl. Acad. Sci. U. S. A. 2014 Mar 18;111(11):4274e9.

[28] Philippe N, LeGendre M, Doutre G, et al. Pandoraviruses: amoeba viruses withgenomes up to 2.5 Mb reaching that of parasitic eukaryotes. Science 2013 Jul19;341(6143):281e6.

[29] Scheid P, Balczun C, Schaub GA. Some secrets are revealed: parasitic keratitisamoebae as vectors of the scarcely described pandoraviruses to humans.Parasitol. Res. 2014 Jul 18 (Epub ahead of print).

[30] Yoosuf N, Yutin N, Colson P, et al. Related giant viruses in distant locations anddifferent habitats: Acanthamoeba polyphaga moumouvirus represents a thirdlineage of the Mimiviridae that is close to the megavirus lineage. Genome Biol.Evol. 2012;4(12):1324e30.

[31] Martinez MJ, Swan BK, Wilson WH. Marine viruses, a genetic reservoirrevealed by targeted viromics. ISME J. 2014 May;8(5):1079e88.

[32] Fischer MG, Allen MJ, Wilson WH, Suttle CA. Giant virus with a remarkablecomplement of genes infects marine zooplankton. Proc. Natl. Acad. Sci. U. S. A.2010 Nov 9;107(45):19508e13.

[33] Fischer MG, Suttle CA. A virophage at the origin of large DNA transposons.Science 2011 Apr 8;332(6026):231e4.

[34] La Scola B, Marrie TJ, Auffray JP, Raoult D. Mimivirus in pneumonia patients.Emerg. Infect. Dis. 2005;11(3):449e52.

[35] Berger P, Papazian L, Drancourt M, La Scola B, Auffray JP, Raoult D. Ameba-associated microorganisms and diagnosis of nosocomial pneumonia. Emerg.Infect. Dis. 2006;12(2):248e55.

[36] Vincent A, La SB, Forel JM, Pauly V, Raoult D, Papazian L. Clinical significance ofa positive serology for mimivirus in patients presenting a suspicion ofventilator-associated pneumonia. Crit. Care Med. 2009 Jan;37(1):111e8.

[37] Vanspauwen MJ, Franssen FM, Raoult D, Wouters EF, Bruggeman CA,Linssen CF. Infections with mimivirus in patients with chronic obstructivepulmonary disease. Respir. Med. 2012 Dec;106(12):1690e4.

[38] Boyer M, Azza S, Barrassi L, et al. Mimivirus shows dramatic genome reduc-tion after intraamoebal culture. Proc. Natl. Acad. Sci. U. S. A. 2011 Jun21;108(25):10296e301.

[39] Larcher C, Jeller V, Fischer H, Huemer HP. Prevalence of respiratory viruses,including newly identified viruses, in hospitalised children in Austria. Eur. J.Clin. Microbiol. Infect. Dis. 2006 Nov;25(11):681e6.

[40] Dare RK, Chittaganpitch M, Erdman DD. Screening pneumonia patients formimivirus. Emerg. Infect. Dis. 2008 Mar;14(3):465e7.

[41] Costa C, Bergallo M, Astegiano S, et al. Detection of Mimivirus in bron-choalveolar lavage of ventilated and nonventilated patients. Intervirology2012;55(4):303e5.

[42] Vanspauwen MJ, Schnabel RM, Bruggeman CA, et al. Mimivirus is not afrequent cause of ventilator-associated pneumonia in critically ill patients.J. Med. Virol. 2013 Oct;85(10):1836e41.

[43] Lysholm F, Wetterbom A, Lindau C, et al. Characterization of the viralmicrobiome in patients with severe lower respiratory tract infections, usingmetagenomic sequencing. PLoS One 2012;7(2):e30875.

[44] Ghigo E, Kartenbeck J, Lien P, et al. Ameobal pathogen mimivirus infectsmacrophages through phagocytosis. PLoS Pathog. 2008 Jun;4(6):e1000087.

ource of pathogenic viruses, Microbial Pathogenesis (2014), http://

B. La Scola / Microbial Pathogenesis xxx (2014) 1e5 5

[45] Silva LC, Almeida GM, Oliveira DB, et al. A resourceful giant: APMV is able tointerfere with the human type I interferon system. Microbes Infect. 2014Mar;16(3):187e95.

[46] Vincent A, La Scola B, Papazian L. Advances in Mimivirus pathogenicity.Intervirology 2010;53:304e9. Karger ed.

[47] Raoult D, Renesto P, Brouqui P. Laboratory infection of a technician bymimivirus. Ann. Intern Med. 2006 May 2;144(9):702e3.

[48] Slimani M, Pagnier I, Boughalmi M, Raoult D, La SB. Alcohol disinfectionprocedure for isolating giant viruses from contaminated samples. Intervirol-ogy 2013;56(6):434e40.

[49] Boratto PV, Dornas FP, Andrade KR, et al. Amoebas as mimivirus bunkers:increased resistance to UV light, heat and chemical biocides when viruses arecarried by amoeba hosts. Arch. Virol. 2014 May;159(5):1039e43.

[50] Bonnave PE, Raoult D, Drancourt M. Gastric aspiration is not necessary for thediagnosis of pulmonary tuberculosis. Eur. J. Clin. Microbiol. Infect. Dis. 2013Apr;32(4):569e71.

Please cite this article in press as: La Scola B, Looking at protists as a sdx.doi.org/10.1016/j.micpath.2014.09.005

[51] Colson P, de L X, Yutin N, et al. “Megavirales”, a proposed new order foreukaryotic nucleocytoplasmic large DNA viruses. Arch. Virol. 2013Dec;158(12):2517e21.

[52] Lagier JC, Armougom F, Million M, et al. Microbial culturomics: paradigm shiftin the human gut microbiome study. Clin. Microbiol. Infect. 2012 Dec;18(12):1185e93.

[53] Popgeorgiev N, Boyer M, Fancello L, et al. Marseillevirus-like virus recoveredfrom blood donated by asymptomatic humans. J. Infect. Dis. 2013 Oct1;208(7):1042e50.

[54] Popgeorgiev N, Colson P, Thuret I, et al. Marseillevirus prevalence in multi-transfused patients suggests blood transmission. J. Clin. Virol. 2013 Dec;58(4):722e5.

[55] Popgeorgiev N, Michel G, Lepidi H, Raoult D, Desnues C. Marseillevirusadenitis in an 11-month-old child. J. Clin. Microbiol. 2013 Dec;51(12):4102e5.

ource of pathogenic viruses, Microbial Pathogenesis (2014), http://