The Effect of In Vitro Growth Conditions on the Resistance of Acanthamoeba Cysts

The Effect of In Vitro Growth Conditions on the Resistance of Acanthamoeba Cysts

CELINE COULON,a,b NATHALIE DECHAMPS,c THIERRY MEYLHEUC,d ANNE COLLIGNON,b GERALD MCDONNELLe

and VINCENT THOMASa

aSTERIS SA R&D, Fontenay-aux-Roses, 92260, France, andbDepartement de Microbiologie, Faculte de Pharmacie, Universite de Paris-Sud XI, Chatenay-Malabry, 92290, France, and

cCytometry and cell sorting facility, Institut de Radiobiologie Cellulaire et Moleculaire, Direction des Sciences du Vivant, Commissariat al’Energie Atomique, Fontenay, 92260, France, and

dPlateforme de Microscopie MIMA2, UMR1319 INRA/AGROPARISTECH, Micalis B2HM, Massy, 91300, France, andeSTERIS Limited, STERIS House, Jays Close Viables, Basingstoke, Hants, RG22 4AX, United Kingdom

ABSTRACT. Despite increasing concerns of direct pathogenicity and/or their role as hosts for other microorganisms there arecurrently no standard methods for the inactivation of amoebae that belong to the genus Acanthamoeba. Methods used to growamoebae and produce cysts for these tests may be important as they can dramatically modify cyst susceptibility. We comparedresistance of cysts produced from trophozoites grown in peptone-yeast extract-glucose broth or by feeding on HEp-2 cells andthen encysted in Neff’s medium. We observed that trophozoites grown using HEp-2 cells as a nutrient source produce cysts thatare significantly more resistant to SDS and to most biocides tested, including heat. Increased resistance is likely due to a higherproportion of mature cysts presenting thicker cell walls as demonstrated using transmission electron microscopy. This was con-firmed by calcofluor white staining demonstrating higher cellulose content in cysts produced from trophozoites grown using HEp-2 cells as a feeding source. These results demonstrate that not only methods used to produce cysts from trophozoites are critical,but that methods used to grow trophozoites before encystment should also be chosen carefully. This should be taken into accountfor the development of protocols to evaluate biocides and antimicrobials against amoebal cysts.

Key Words. Amoeba, biocide, encystment, growth condition, human cells, protozoa.

F REE-LIVING Amoebae belonging to the genus Acantha-moeba are ubiquitous in the environment. They are

mostly found in water but also in soil, air, and dust(Rodriguez-Zaragoza 1994). In a recent clinical study, Acan-thamoeba sp. have also been reported from the urine of criti-cally ill patients in intensive care units: 17 of 63 samplescollected from urinary catheters were positive for Acantha-moeba sp. (Santos et al. 2009). The organism’s life cycle iscomposed of two distinct stages: a metabolically active stagewith feeding and replicating trophozoites, and a dormant stagewith the formation of a cyst that resists environmental stressin case of unfavourable conditions (Neff et al. 1964). Thesecysts are double-walled and contain high proportions of cellu-lose (Dudley et al. 2009).

Several Acanthamoeba strains are pathogenic for humans,being mostly associated with amoebal keratitis, a sight-threat-ening corneal infection that can affect immunocompetent indi-viduals. Reported incidence of the disease has increaseddramatically over the last three decades presumably due toincreased use of contact lenses; it has been reported to be inthe range of 0.15–1.4 per million (Dart et al. 2009). Granu-lomatous amoebic encephalitis, disseminated diseases, and skinlesions have also been reported with Acanthamoeba sp., butare currently only rarely observed (Visvesvara et al. 2007).

In addition to their intrinsic pathogenicity, Acanthamoebacan potentially harbour intracellularly various bacterial, viral,and eukaryotic species pathogenic for human and animals(Greub and Raoult 2004). Among 539 bacterial species listedas being pathogenic for humans and/or animals, 102 to datehave been described in the literature as being able to resistand potentially proliferate within amoebae (Thomas et al.2010). This includes well known hospital-acquired pathogens,such as Acinetobacter sp., Enterobacter sp., Legionella sp., var-ious mycobacterial species, Pseudomonas sp., and Serratia sp.(Thomas et al. 2010). Survival of bacteria within amoebalcysts has also been demonstrated for several of these species,

including mycobacteria. Cysts are thus considered as potentialreservoirs of pathogenic microorganisms in water networksand it has been suggested that testing mycobacteria protectedwithin cysts may be necessary to properly evaluate decontami-nation procedures for water and endoscopes (Adekambi et al.2006).

We have recently published data regarding the intrinsicresistance of amoebal cysts to disinfection treatments used inhealthcare settings (Coulon et al. 2010). This study demon-strated that several chemical formulations containing peraceticacid, hydrogen peroxide or orthophtalaldehyde presentedgreater efficacy than glutaraldehyde, as did ethanol andsodium hypochlorite; however, some of these treatmentsrequired relatively long incubation times to achieve cyst inacti-vation. In this study we chose to use Neff medium to producecysts as it has been recommended due to its ability to inducesynchronous encystment and high yield of cysts (Neff et al.1964). However, it has also been reported that Neff methodsuffers limitations, and in some cases cysts produced usingother methods, such as encystment on nonnutrient agar(NNA) plates, have proved more resistant to disinfection(Hughes et al. 2003). On the basis of recent studies demon-strating that Acanthamoeba strains lose their abilities to encystsynchronously and show reduced temperature tolerance uponprolonged axenic culture (Koehsler et al. 2008; Pumidonminget al. 2010), and that growth on cell layers reactivates attenu-ated properties (Koehsler et al. 2009), we considered if cystsproduced from trophozoites grown on cell layers are moreresistant to disinfection treatments and consequently moreappropriate to test biocides used to reprocess critical medicaldevices.

MATERIALS AND METHODS

Four representative Acanthamoeba strains were used: twoenvironmental isolates (T4 genotype) and two collectionstrains (Acanthamoeba castellanii CCAP 1501/10 and Acantha-moeba polyphaga CCAP 1501/18). They were routinely culti-vated at 28 °C in sterile peptone–yeast extract–glucose (PYG)medium containing for 5 L of distilled water: 100 g proteosepeptone (BD Difco, Temse, Belgium), 10 g yeast extract (BD

Corresponding Author: V. Thomas, STERIS SA R&D, 18 routedu Panorama, 92260 Fontenay-aux-Roses, France—Telephonenumber: + 33-146548557; FAX number: + 33-146549843; e-mail:[email protected]

J. Eukaryot. Microbiol., 0(0), 2012 pp. 1–8© 2012 The Author(s)Journal of Eukaryotic Microbiology © 2012 International Society of ProtistologistsDOI: 10.1111/j.1550-7408.2012.00612.x

1

Published bythe International Society of ProtistologistsEukaryotic Microbiology

The Journal of

Difco), 4.9 g MgSO4·7H2O, 5 g sodium citrate·2H2O, 0.1 g Fe(NH4)2(SO4)2·6H2O, 1.7 g KH2PO4, 1.97 g Na2HPO4·7H2O,45 g glucose, and 0.295 g CaCl2 (Greub and Raoult 2002).

Trophozoites to be used for cyst production were grownusing two different methods: growth in sterile PYG mediumor feeding on HEp-2 cells. The HEp-2 cells (American TypeCulture Collection, ATCC) CCL-23 were cultivated at 37 °Cin 150 cm² flasks with Dulbecco’s modified Eagle medium(DMEM) + 5% (v/v) foetal bovine serum. When cells reachedconfluence, DMEM was removed and replaced by 0.9% (w/v)NaCl containing ~ 106 amoeba trophozoites; flasks were thenincubated at 28 °C to allow amoebal growth and subculturedthree times using this method (Koehsler et al. 2009). Troph-ozoites cultivated using both methods were then encysted inNeff’s medium for 1 wk at 33 °C as previously described(Coulon et al. 2010; Neff et al. 1964).

To evaluate viability of cysts obtained from both condi-tions, they were stained with calcofluor white and propidiumiodide (PI) just after encystment in Neff’s medium. The samestaining was used to evaluate viability of cysts treated with0.5% (w/v) SDS for 10 min to inactivate nonmature cysts.Briefly, trophozoites were encysted in Neff medium containing0.005% (v/v) calcofluor white (Fluka 18909, Saint-Quentin-Fallavier, France) for 7 d at 33 °C to detect cellulose-contain-ing structures (Chavez-Munguia et al. 2005). They were thenincubated for 10 min with 2 lg/ml PI in Page’s amoeba saline(PAS) before or after SDS treatment, washed in PBS, andobserved with an IX 71 inverted microscope equipped with anX-cite® 120Q fluorescence illuminator (Olympus, Rungis,France). Proportions of PI-stained cysts were counted from atleast 400 cysts.

Transmission electron microscopy was used to evaluate dif-ferences in morphologies of cysts obtained from trophozoitescultivated in PYG or by feeding on HEp-2 cells. After encyst-ment for 7 d in Neff’s medium, cysts of strain Acanthamoebasp4 were collected, treated with 0.5% (w/v) SDS for 10 min,washed in PBS, and fixed in 2.5% (v/v) glutaraldehyde forone night, postfixed for 1 h in 1% (w/v) OsO4 in 0.1 M caco-dylate buffer overnight, stained with 0.25% (w/v) uranyl ace-tate overnight, dehydrated, and embedded in epoxy resin.Thin sections (i.e. 50–60 nm) were stained for 5 min with ura-nyl acetate, then for 5 min with lead citrate, and finally exam-ined with a PHILIPS EM208 electron microscope (PHILIPS,Eindhoven, The Netherlands).

For flow cytometry and fluorescence microscopy analysis,trophozoites cultivated on PYG medium or HEp-2 cells wereencysted in Neff medium containing 0.005% (v/v) calcofluorwhite (Fluka 18909) for 7 d at 33 °C. After 7 d incubationcysts were detached with a cell scrapper, harvested, concen-trated by centrifugation to reach 106 cysts/mL, treated with0.5% (w/v) SDS, fixed in 200 ll methanol for 10 min, andwashed two times in PBS. Stained cysts were then analysedand sorted with an Influx cell sorter (Becton Dickinson, LePont de Claix, France), equipped with Coherent Innova 308-UV and 307 argon ion lasers (Courtaboeuf, France), tuned toemit 100 mW of UV (all lines) and 200 mW of 488 nm. For-ward scatter (FSC) and log side-scatter (SSC) were detectedoff the 488-nm laser, whereas UV-excited calcofluor white flu-orescence intensity was detected through a 560/50 bandpassfilter (SemRock, Rochester, NY, USA) (Bayer et al. 2007). Aminimum of 50,000 events were stored in listmode for off-lineanalysis, and 3,000–15,000 events presenting clearly distin-guishable calcofluor white fluorescence compared with back-ground fluorescence observed in unstained samples wereanalysed, and clearly separated populations based on SSCchannel (log) and calcofluor white fluorescence were sorted in

300 ll sterile water, concentrated to 50 ll by high-speed cen-trifugation (21,000 g for 5 min), deposited in wells of a Tef-lon-coated microscopy slide, dried at 37 °C, mounted withProlong Gold antifade reagent (Invitrogen, Villebon, France),and observed with a Olympus IX 71 inverted microscopeequipped with a X-cite® 120Q fluorescence illuminator (Olym-pus).

For disinfection tests, cysts were harvested in one-fourth ofstrength Ringer’s solution, treated with 0.5% (w/v) SDS for10 min to lyse nonmature cysts (Dudley et al. 2005) and disso-ciate aggregates (Coulon et al. 2010), then washed two timesin one-fourth of strength Ringer’s solution and stored at 4 °Cfor testing within 15 d. To test biocide efficacy, cyst suspen-sions were adjusted to 107–108 cysts/ml in one-fourth ofstrength Ringer’s solution and approx. 106–107 cysts werediluted 1/10 and incubated for 10–30 min in biocide suspen-sions (Table 1). Treated cysts were then re-suspended for5 min in Dey-Engley neutralizing broth (Buck and Rosenthal1996), then collected by centrifugation at 21,000 g for 5 min,suspended in saline buffer Ringer’s solution and seriallydiluted on Escherichia coli ATCC 25922 lawns on NNAplates. After 7 d incubation at 28 °C, NNA plates wereobserved for regrowth of trophozoites, and log reductionswere calculated using the Spearman–Karber method asdescribed for virucidal studies (Dougherty 1964). This methodis mainly used for the calculation of mycoplasma or virustitres and relies on the principle of quantal dose responserelationship. It has been proposed to calculate reduction ofamoebal titres after biocidal treatment (http://www.fda.gov/downloads/MedicalDevices/NewsEvents/WorkshopsConferences/UCM130739.ppt). To prepare NNA plates, 1.5 g agar (BDDifco) was diluted in 100 ml of Page’s modified Neff’s amoebasaline (PAS) containing for 1 L of distilled water: 120 mgNaCl, 4 mg MgSO4·7H2O, 4 mg CaCl2·2H2O, 142 mgNa2HPO4, and 136 mg KH2PO4.

Scanning electron microscopy was used to observe the effectof disinfection treatments on morphology of Acanthamoebasp4 cysts. Briefly, treated and nontreated cysts were immersedin 2.5% (v/v) glutaraldehyde in 0.1 M sodium cacodylate buf-fer, pH 7.4, overnight at 4 °C. The fixative was removed, andsamples were rinsed three times for 10 min in the 0.1 Msodium cacodylate solution, pH 7.4. The samples underwentprogressive dehydration by soaking in a graded series of etha-nol (i.e. 50–100%) before critical-point drying under CO2.

Table 1. List of biocides and contact times tested in this study.

Biocides

Finalconcentrations

of activecompounds

pH ofsolutions

(measured at21 °C)

Contacttimes(min)

Unformulated biocidesSodium hypochlorite 2,500 ppm 11.3 30Hydrogen peroxide 7.5% 3.4 30

Formulated biocidesGlutaraldehyde-based

product2%

glutaraldehyde5.6 30

Ortho-phtalaldehyde-basedproduct

0.55% OPA 7.5 30

PAA-based productSTERIS-20 (55 °C)

0.2% PAA 6.2 10

Heat65 °C N/A 7.2 10

All biocides were tested at recommended temperatures (room tem-perature, with the exception of STERIS-20).

2 J. EUKARYOT. MICROBIOL., 0, NO. 0, JANUARY 2012

Samples were mounted on aluminium stubs with double-sidedadhesive carbon discs (10 mm diam.) and sputter coated withgold-palladium (Polaron SC7640; Elexience, Verrieres-le-buisson, France) for 200 s at 10 mA. Samples were visualizedat 2 kV by field emission gun scanning electron microscopy assecondary electron images with a Hitachi S4500 instrument(Elexience, Verrieres-le-buisson).

All disinfection tests and PI-staining tests were performedat least four times. Fisher’s exact test was used to comparenumbers of PI-stained cysts and numbers of wells in whichamoebal regrowth was clearly observed after serial-dilutingtreated cysts onto E. coli-layered NNA plates. Statistical sig-nificance was set at an alpha level of 0.05.

RESULTS

Propidium iodide (PI) staining experiments demonstratedthat cysts obtained from trophozoites fed HEp-2 cells wereless frequently stained by PI than cysts obtained from PYG-grown trophozoites (Fig. 1). This difference in proportions ofPI-stained cysts was also generally observed after treatmentwith SDS, except for strain Acanthamoeba sp1, which showedhigh variability. When observed by transmission electronmicroscopy, PYG-derived cysts of Acanthamoeba sp4 consistedof two distinct populations: rounded immature cysts for whichonly an endocyst could be detected (Fig. 2, white arrows), andmore “mature” cysts for which exocyst structures were alsoobserved (Fig. 2, black arrows). The cysts derived fromgrowth on HEp-2 cells were more homogenous, with a major-ity of mature cysts being detected (Fig. 3). Furthermore,numerous structures resembling lipid droplets were observedin cysts obtained from trophozoites fed with HEp-2 cells(Fig. 4) whereas they were generally not found in cystsobtained from PYG-grown trophozoites (data not shown).

Flow cytometric analysis performed with cysts of the fourdifferent strains obtained from trophozoites cultivated withthe two different methods demonstrated the presence of oneor two separated populations depending on strains andmethod used to grow trophozoites (Fig. 5–9 and Table 2).When analysing mean SSC and mean fluorescence intensitiesof well-defined populations for every strain we found thatpopulations presenting the highest SSC and highest fluores-cence values in the HEp-2-fed population always had higher

Fig. 1. Resistance to SDS of cysts obtained from the four differ-ent Acanthamoeba strains grown in peptone-yeast extract-glucose(PYG)- or on HEp-2 cells. Proportions of cysts stained with propidiumiodide before (white bars) or after (grey bars) treatment with 0.5%(w/v) SDS for 10 min are represented. ** P < 0.001.

COULON ET AL.—RESISTANCE OF ACANTHAMOEBAE TO DISINFECTION 3

SSC and brighter fluorescence than PYG-grown populations(Table 2). Microscopic observation of sorted populations dem-onstrated that populations with higher SSC and bright calco-fluor white fluorescence presented typical cysts shapes withirregular, thick, and highly fluorescent cyst walls (Fig. 7, 9)whereas populations with lower SSC and dim calcofluor whitefluorescence were composed of rounded cysts presenting thin-ner and visually less fluorescent cell walls (cf. Fig. 6).

As previously described (Coulon et al. 2010), cysts producedfrom PYG-grown trophozoites encysted in Neff medium wereefficiently inactivated (> 4 log10 reduction) by most biocidesand for all strains tested (Fig. 10). The glutaraldehyde-basedproduct and a 7.5% (v/v) hydrogen peroxide solution werenot fully efficient on several strains even after 30-min expo-sure. Cysts produced from trophozoites grown onto HEp-2cells and encysted in Neff medium were generally more resis-tant to biocides, with the exception of the peracetic acid-basedproduct that was still very active (> 4 log10 reduction) on thefour tested strains after only 10 min exposure (Fig. 10). Forall strains, cysts produced from trophozoites grown ontoHEp-2 cells were significantly more resistant to 10-min expo-sure to heat and to a 30-min exposure to the glutaraldehyde-based product compared with cysts produced from trophozo-ites grown in PYG. For three of four amoebal strains, cystsproduced from trophozoites grown on HEp-2 cells were alsomore resistant to 30-min exposure to 0.25% (v/v) bleach or7.5% (v/v) hydrogen peroxide; higher resistance to a 30-min

Fig. 2–4. Ultrastructure of cysts of strain Acanthamoeba sp 4.grown in peptone-yeast extract-glucose (PYG) or on HEp-2 cells andencysted in Neff medium for 7 d at 33 °C. “Mature” cysts (blackarrows) were observed together with rounded cells resembling “non-mature” cysts (white arrows) in PYG-derived cysts (Fig. 2). “Mature”cysts were mainly observed in HEp-2-derived cysts (Fig. 3). Notethese “mature” cysts displayed “mature” endocyst walls and oftencontained numerous lipid droplets (Fig. 4). Bars: 10 lm (2, 3) and2 lm (4).

Table 2. Mean side-scatter channel (SSC) and calcofluor white fluorescence intensities of the different cysts populations analysed in thisstudy.

Sample

Dim (or full) population Bright population (if present)

% cysts SSC Fluorescence % cysts SSC Fluorescence

A.castellanii CCAP 1501/10 - PYG >95.0 92 405 - - -A.castellanii CCAP 1501/10 - Hep-2 79.0 101 186 19.9 402 530A. polyphaga CCAP 1501/18 - PYG >95.0 450 496 - - -A. polyphaga CCAP 1501/18 - Hep-2 >95.0 562 788 - - -Acanthamoeba sp1- PYG 62.4 198 348 35.9 665 1179Acanthamoeba sp1- Hep-2 64.9 296 488 32.2 973 1511Acanthamoeba sp4- PYG 53.8 122 171 45.1 300 558Acanthamoeba sp4- Hep-2 >95.0 517 773 - - -

Values are indicated for one or two populations depending if one or two differentiated populations were observed during analysis. When twopopulations were detected, calcofluor white-bright populations were demonstrated by microscopy to correspond to more mature cysts comparedwith calcofluor white-dim populations.

Fig. 5–7. Flow cytometric analysis of cysts of Acanthamoeba sp4 obtained from trophozoites grown in peptone-yeast extract-glucose(PYG). Cysts were stained with calcofluor white during encystment in Neff medium for 7 d at 33 °C. Different populations were observed andsorted by flow cytometry and then observed with a fluorescence microscope. “Non-mature” as well as “mature” cysts were obtained from troph-ozoites grown in PYG (Fig. 5), and corresponding pictures of the sort regions A (Fig. 6), and B (Fig. 7), respectively.


exposure to the Ortho-phtalaldehyde (OPA)-based productwas observed for two strains (Fig. 10).

Scanning electron microscopy (SEM) allowed observationof several distinct cyst morphologies in cysts populationsobtained from Acanthamoeba sp4 trophozoites grown in PYGor fed HEp-2 cells (Fig. 11–15). Wrinkled cysts (Fig. 11) weremostly detected in cyst populations obtained from HEp-2derived trophozoites, whereas a mix of wrinkled and roundedcysts (Fig. 12) was observed in cyst populations obtained fromPYG-grown trophozoites. Surfaces of wrinkled cysts had agranular aspect whereas surfaces of rounded cysts had a fibril-lar aspect. Rounded cysts presenting fibrillar circumvolutionson their surface (Fig. 13) were only observed in cyst popula-tions obtained from HEp-2-grown trophozoites and were inminority in this population. Whatever condition was used togrow trophozoites before encystment, we did not observe anyobvious morphological change of cysts treated with hydrogenperoxide, the OPA-based product, and the PAA-based prod-uct (data not shown). PYG- and HEp-2 derived cysts treatedwith sodium hypochlorite had a swollen appearance withridges (Fig. 14). Fibrils were more clearly separated on thesurface of PYG-derived rounded cysts heated at 65 °C for10 min (Fig. 15), whereas the surface of wrinkled cysts wasleft unchanged (data not shown).

DISCUSSION

As noted in the introduction, there is a critical need to vali-date treatment efficacy against Acanthamoebae, particularlyagainst cysts, and methods to be used to test the efficacy ofbiocides against cysts of Acanthamoeba sp. are under discus-sion (Anger and Lally 2008). As for other micro-organisms,test methodology has a potentially high impact on conclusionsthat can be drawn from such tests. It has been reported thatdifferent acanthamoebal strains present different susceptibili-ties to biocides and that methods used to produce cysts fromtrophozoites also have a dramatic importance on final results

(Anger and Lally 2008; Beattie et al. 2003; Buck et al. 2000;Hughes et al. 2003). Importantly, cysts derived from labora-tory strains that have been maintained in prolonged axenicculture are more susceptible to disinfection than those derivedfrom the original isolate (Hughes et al. 2003). Furthermore, ithas been reported that trophozoites grown in sterile PYG andincubated in Neff’s encystment medium produce roundedtrophozoites and immature cysts, but not double-walledmature cysts even after more than 4 wk incubation (Beattieet al. 2003). Despite this, encystment in Neff medium is stillcurrently used to produce cysts as it is easy to set-up and toproduce large amounts of cysts.

As recent studies demonstrated that unnecessary metabolicfeatures, such as encystment and protease activities that arelost or at least down-regulated after prolonged axenic culture,can be recovered after growth on human cells (Koehsler et al.2008, 2009; Pumidonming et al. 2010), we tested if trophozo-ites cultivated on HEp-2 cells produce cysts that are moreresistant than cysts obtained from PYG-grown trophozoites.We demonstrated it is indeed the case: HEp-2 derived cystswere more resistant to SDS treatment and contained higherproportions of lipid droplets associated with mature cysts(Bowers and Korn 1969). Calcofluor white staining experi-ments confirmed higher proportions of more mature cystsobtained from trophozoites fed with HEp-2 cells, resulting inhigher cellulose content in the cyst walls and higher resistanceto disinfection (Connell et al. 2001; Lloyd et al. 2001). Cystspresenting high fluorescence values generally had higher SSCvalues, likely due to increased refraction by the cyst walls.Thus, even if not prevailing in the whole cysts population,these more mature cysts are likely responsible for the observedincreased resistance to biocides. Whether additional specificfactors over- or under-expressed in the presence of mamma-lian cells also play a significant role in this increased resistanceremains to be determined.

Comparison between strains is difficult as differencesbetween fluorescence values are not directly linked to observed

Fig. 8–9. Flow cytometric analysis of cysts of Acanthamoeba sp4 obtained from trophozoites grown on HEp-2 cells. More than 95%“mature” cysts were obtained from trophozoites grown on HEp-2 cells (Fig. 8), and corresponding picture of sort region C (Fig. 9). Region D(Fig. 8) indicates the same area as region A (Fig. 5), but did not contain an identifiable subpopulation. Calcofluor white-low, side-scatter chan-nel (SSC)-low populations essentially contained cellular debris, and no intact cells were retrieved.


differences between resistances to biocides: as an example,cysts obtained from HEp-2-fed trophozoites are less resistant(P < 0.05) to hydrogen peroxide for Acanthamoeba sp1 thanfor Acanthamoeba sp4, despite having higher mean fluores-cence value. Previous studies reported no significant differ-ences in FSC for cysts obtained from PYG-growntrophozoites using different methods and presenting differentsusceptibilities to biocides (Hughes et al. 2003). Accordingly,we did not observe such differences in our experiments andcould clearly distinguish populations only due to differences incalcofluor white staining.

Nonmature forms mainly observed in cyst preparationsobtained from PYG-grown trophozoites resembled cysts in theearly stages of exocyst synthesis that have been previouslydescribed by others (Bowers and Korn 1969; Pasternak et al.

Fig. 10. Cysticidal effect of treatments including biocides andmoist heat against cysts of Acanthamoeba castellanii CCAP 1501/10,Acanthamoeba polyphaga CCAP 1501/18, and two genotype T4 strainspreviously isolated from river water (Acanthamoeba sp1) and hospitalwater (Acanthamoeba sp4). Log10 reductions are reported. ■: troph-ozoites were grown in peptone-yeast extract-glucose and encysted inNeff medium; : trophozoites were grown on HEp-2 cells and ency-sted in Neff medium. Statistically significant differences observed forthe efficacy of treatments against cysts obtained according to bothmethods are indicated with stars: * P < 0.05; ** P < 0.001. “>”: Max-imum log10 reduction was observed.

Fig. 11–15. Fine structural aspects of the surface of cysts ofAcanthamoeba sp4 grown in either peptone-yeast extract-glucose(PYG) or by feeding on HEp-2 cells. “Mature” cysts (Fig. 11A, 11B)were observed together with “non-mature” cysts (Fig. 12A, 12B) inPYG-derived cysts whereas “mature” cysts were largely predominantin HEp-2 -derived cysts. Rounded cysts presenting fibrillar circumvo-lutions on their surface were only observed in HEp-2 -derived cystsand could represent “non-mature” cysts in this population (Fig. 13A,13B). “Mature” cysts treated with bleach presented a swollen aspect(Fig. 14A, 14B). “Non-mature” cysts heated at 65 °C presented fibrilswith a more resolved aspect (Fig. 15A, 15B) compared with nonheat-ed cysts (see Fig. 12A, 12B). Bars: 4.29 lm (left column), 750 nm(right column).


1970). It has also been suggested that cyst wall synthesis startswith the formation of the exocyst, is then followed by the for-mation of an endocyst consisting of fibrous material (cellu-lose), and the intercyst space is filled with filaments connectingthe two structures (Lemgruber et al. 2010). Our observationstend to confirm that the exocyst also has a fibrillar aspect inits early developmental stage (Chavez-Munguia et al. 2005).Well-separated fibrils observed at the surface of cysts heatedat 65 °C suggest that lipid-containing material has beenremoved by heat. In previous reports authors suggested thatswelling of the cyst wall observed after treatment with highpolyhexamethylenebiguanide concentrations could be due toeffect of the biocide on the inner cyst wall (Khunkitti et al.1998). Based on observations reported in our study this couldalso be the case after treatment with bleach.

As already reported by others (Shoff et al. 2008; Srikanthand Berk 1993), we observed that degrees of resistance to bio-cides also depended on the strains studied and biocides testedin addition to growth in PYG or on cells. This could be dueto subtle differences in cyst wall composition that could notbe detected with analysis performed in this study. Some of thetreatments tested in our study (i.e. aldehyde-based treatments)are currently widely used for high level disinfection. Of note,resistance of trophozoites and cysts to glutaraldehyde hasalready been reported by others (Coulon et al. 2010; Greuband Raoult 2003). Published data concerning heat-inactivationof Acanthamoeba sp. are variable: Turner et al. (2000)observed that trophozoites of A. castellanii were inactivatedfollowing a 30-min exposure at 46 °C, whereas a temperatureof 56 °C was necessary to inactivate the same number of cystsproduced from PYG-grown trophozoites encysted in Neffmedium. Aksozek et al. (2002) reported that a temperature of65 °C for more than 5 min was fully efficient to inactivatetrophozoites of A. castellanii cultivated in PYG and encystedon NNA plates, whereas Ludwig et al. (1986) reported thatcysts of A. castellanii and A. polyphaga directly encysted inPYG were inactivated only after exposure to moist heat for10 min at 80 °C. The same exposure conditions did not com-pletely inactivate cysts of thermo-tolerant Acanthamoeba sp.infected with L. pneumophila (Storey et al. 2004). Increasedheat-resistance is a major concern as heat-shock treatments at65 °C are commonly used to control proliferation of L. pneu-mophila in water distribution networks (Mouchtouri et al.2007). Acanthamoebae sp. are frequently isolated from hospitalwater networks (Thomas et al. 2006) and incomplete efficacycould thus result in increased risk of patient contamination withamoebal cysts harbouring pathogenic bacterial species.

Our observations with SEM suggest that the methoddescribed by Koehsler et al. (2009) to produce mature cystsfrom trophozoites grown on HEp-2 cells is of major interestfor the evaluation of disinfection efficacy. The method is rela-tively simple to set-up and does not require special media orequipment. Cysts produced from trophozoites grown on HEp-2 cells are potentially more similar to amoebal cysts freshlyisolated from patients, without requiring the use of in vivomodels that are more difficult to set-up and can only be usedin labs equipped with animal care facilities. However, asimportant variability in sensitivity to biocides is observed fromone amoebal strain to another, it is difficult to fully demon-strate capacity of the method to restore biocide resistances ofPYG-derived cysts at their initial level. To specifically achievethis goal, new Acanthamoeba isolates should be collected fromclinical and environmental samples and disinfection tests per-formed against these fresh isolates. The new isolates shouldthen be cultivated in PYG for several dozen generations, andnew disinfection tests performed at regular intervals before or

after reactivation by passage on cell monolayers. In addition,the effect of long-term growth of amoebae on HEp-2 cellsrequires further investigation. Another important point toconsider is the potential effect of the presence of intracellularbacteria on encystment and the resistance of cysts to disinfec-tion: a recent report suggests that A. castellanii Strain Ma(ATCC 50370), a strain currently used for testing biocide effi-cacy (Gatti et al. 1998; Kilvington et al. 2008), harbours apreviously unrecognized mycobacterial endosymbiont relatedto the Mycobacterium avium complex (Glaser et al. 2011).These results should be taken into account for the develop-ment of normalized protocols for the evaluation of biocidesagainst protozoa, including amoebal cysts.

ACKNOWLEDGMENTS

This study was partially funded by the Direction Generaledes Entreprises (convention no. 072906389) under the Frenchcompetitiveness cluster Medicen Region Ile-de-France.

LITERATURE CITED

Adekambi, T., Ben Salah, S., Khlif, M., Raoult, D. & Drancourt, M.2006. Survival of environmental mycobacteria in Acanthamoebapolyphaga. Appl. Environ. Microbiol., 72:5974–5981.

Aksozek, A., McClellan, K., Howard, K., Niederkorn, J. Y. & Ali-zadeh, H. 2002. Resistance of Acanthamoeba castellanii cysts tophysical, chemical, and radiological conditions. J. Parasitol.,88:621–623.

Anger, C. & Lally, J. M. 2008. Acanthamoeba: a review of its poten-tial to cause keratitis, current lens care solution disinfection stan-dards and methodologies, and strategies to reduce patient risk. EyeContact Lens, 34:247–253.

Bayer, J., Grunwald, D., Lambert, C., Mayol, J. F. & Maynadie, M.2007. Thematic workshop on fluorescence compensation settings inmulticolor flow cytometry. Cytometry B Clin Cytom., 72:8–13.

Beattie, T. K., Seal, D. V., Tomlinson, A., McFadyen, A. K. &Grimason, A. M. 2003. Determination of amoebicidal activities ofmultipurpose contact lens solutions by using a most probablenumber enumeration technique. J. Clin. Microbiol., 41:2992–3000.

Bowers, B. & Korn, E. D. 1969. The fine structure of Acanthamoebacastellanii (Neff strain) II. Encystment. J. Cell Biol., 41:786–805.

Buck, S. L. & Rosenthal, R. A. 1996. A quantitative method to evalu-ate neutralizer toxicity against Acanthamoeba castellanii. Appl. Envi-ron. Microbiol., 62:3521–3526.

Buck, S. L., Rosenthal, R. A. & Schlech, B. A. 2000. Methods usedto evaluate the effectiveness of contact lens care solutions and othercompounds against Acanthamoeba: a review of the literature.CLAO J., 26:72–84.

Chavez-Munguia, B., Omana-Molina, M., Gonzalez-Lazaro, M.,Gonzalez-Robles, A., Bonilla, P. & Martinez-Palomo, A. 2005.Ultrastructural study of encystation and excystation in Acantha-moeba castellanii. J. Eukaryot. Microbiol., 52:153–158.

Connell, C., Rutter, A., Hill, B., Suller, M. & Lloyd, D. 2001. Ency-station of Acanthamoeba castellanii: dye uptake for assessment byflow cytometry and confocal laser scanning microscopy. J. Appl.Microbiol., 90:706–712.

Coulon, C., Collignon, A., McDonnell, G. & Thomas, V. 2010. Resis-tance of Acanthamoeba cysts to disinfection treatments used inhealth care settings. J. Clin. Microbiol., 48:2689–2697.

Dart, J. K., Saw, V. P. & Kilvington, S. 2009. Acanthamoeba keratitis:diagnosis and treatment update 2009. Am. J. Ophthalmol., 148:487–499.

Dougherty, R. M. 1964. Animal virus titration techniques. In: Harris,R. C. J. (ed.), Techniques in Experimental Virology. AcademicPress, New York. p.169–223.

Dudley, R., Jarroll, E. L. & Khan, N. A. 2009. Carbohydrate analysisof Acanthamoeba castellanii. Exp. Parasitol., 122:338–343.

Dudley, R., Matin, A., Alsam, S., Sissons, J., Maghsood, A. H. &Khan, N. A. 2005. Acanthamoeba isolates belonging to T1, T2, T3,


T4 but not T7 encyst in response to increased osmolarity and cystsdo not bind to human corneal epithelial cells. Acta Trop., 95:100–108.

Gatti, S., Cevini, C., Bruno, A., Penso, G., Rama, P. & Scaglia, M.1998. In vitro effectiveness of povidone-iodine on Acanthamoebaisolates from human cornea. Antimicrob. Agents Chemother.,42:2232–2234.

Glaser, K. C., Hetrick, N. D. & Molestina, R. E. 2011. Evidence fora previously unrecognized mycobacterial endosymbiont in Acantha-moeba castellanii strain Ma (ATCC® 50370TM). J. Eukaryot. Micro-biol., 58:75–76.

Greub, G. & Raoult, D. 2002. Crescent bodies of Parachlamydiaacanthamoeba and its life cycle within Acanthamoeba polyphaga: anelectron micrograph study. Appl. Environ. Microbiol., 68:3076–3084.

Greub, G. & Raoult, D. 2003. Biocides currently used for broncho-scope decontamination are poorly effective against free-living amoe-bae. Infect. Control Hosp. Epidemiol., 24:784–786.

Greub, G. & Raoult, D. 2004. Microorganisms resistant to free-livingamoebae. Clin. Microbiol. Rev., 17:413–433.

Hughes, R., Heaselgrave, W. & Kilvington, S. 2003. Acanthamoebapolyphaga strain age and method of cyst production influence theobserved efficacy of therapeutic agents and contact lens disinfec-tants. Antimicrob. Agents Chemother., 47:3080–3084.

Khunkitti, W., Hann, A. C., Lloyd, D., Furr, J. R. & Russell, A. D.1998. Biguanide-induced changes in Acanthamoeba castellanii: anelectron microscopic study. J. Appl. Microbiol., 84:53–62.

Kilvington, S., Heaselgrave, W., Lally, J. M., Ambrus, K. & Powell,H. 2008. Encystment of Acanthamoeba during incubation in multi-purpose contact lens disinfectant solutions and experimental formu-lations. Eye Contact Lens, 34:133–139.

Koehsler, M., Leitsch, D., Duchene, M., Nagl, M. & Walochnik, J.2009. Acanthamoeba castellanii: growth on human cell layers reacti-vates attenuated properties after prolonged axenic culture. FEMSMicrobiol. Lett., 299:121–127.

Koehsler, M., Leitsch, D., Furnkranz, U., Duchene, M., Aspock, H.& Walochnik, J. 2008. Acanthamoeba strains lose their abilities toencyst synchronously upon prolonged axenic culture. Parasitol.Res., 102:1069–1072.

Lemgruber, L., Lupetti, P., De Souza, W., Vommaro, R. C. & daRocha-Azevedo, B. 2010. The fine structure of the Acanthamoebapolyphaga cyst wall. FEMS Microbiol. Lett., 305:170–176.

Lloyd, D., Turner, N. A., Khunkitti, W., Hann, A. C., Furr, J. R. &Russell, A. D. 2001. Encystation in Acanthamoeba castellanii:development of biocide resistance. J. Eukaryot. Microbiol., 48:11–16.

Ludwig, I. H., Meisler, D. M., Rutherford, I., Bican, F. E., Langston,R. H. & Visvesvara, G. S. 1986. Susceptibility of Acanthamoeba to

soft contact lens disinfection systems. Invest. Ophthalmol. Vis. Sci.,27:626–628.

Mouchtouri, V., Velonakis, E. & Hadjichristodoulou, C. 2007. Ther-mal disinfection of hotels, hospitals, and athletic venues hot waterdistribution systems contaminated by Legionella species. Am. J.Infect. Control, 35:623–627.

Neff, R. J., Ray, S. A., Benton, W. F. & Wilborn, M. 1964. Inductionof synchronous encystation (differentiation) in Acanthamoeba spp.In: Prescott, D. M. (ed.), Methods in Cell Physiology, Vol. 1 Aca-demic Press, New York. p. 55–83.

Pasternak, J. J., Thompson, J. E., Schultz, T. M. & Zachariah, K.1970. A scanning electron microscopic study of the encystment ofAcanthamoeba castellanii. Exp. Cell Res., 60:290–298.

Pumidonming, W., Koehsler, M. & Walochnik, J. 2010. Acantha-moeba strains show reduced temperature tolerance after long-termaxenic culture. Parasitol. Res., 106:553–559.

Rodriguez-Zaragoza, S. 1994. Ecology of free-living amoebae. Crit.Rev. Microbiol., 20:225–241.

Santos, L. C., Oliveira, M. S., Lobo, R. D., Higashino, H. R., Costa,S. F., van der Heijden, I. M., Giudice, M. C., Silva, A. R. & Levin,A. S. 2009. Acanthamoeba spp. in urine of critically ill patients.Emerg. Infect. Dis., 15:1144–1146.

Shoff, M. E., Joslin, C. E., Tu, E. Y., Kubatko, L. & Fuerst, P. A.2008. Efficacy of contact lens systems against recent clinical andtap water Acanthamoeba isolates. Cornea, 27:713–719.

Srikanth, S. & Berk, S. G. 1993. Stimulatory effect of cooling towerbiocides on amoebae. Appl. Environ. Microbiol., 59:3245–3249.

Storey, M. V., Winiecka-Krusnell, J., Ashbolt, N. J. & Stenstrom, T.A. 2004. The efficacy of heat and chlorine treatment against ther-motolerant Acanthamoebae and Legionellae. Scand. J. Infect. Dis.,36:656–662.

Thomas, V., Herrera-Rimann, K., Blanc, D. S. & Greub, G. 2006.Biodiversity of amoebae and amoebae-resisting bacteria in a hospi-tal water network. Appl. Environ. Microbiol., 72:2428–2438.

Thomas, V., McDonnell, G., Denyer, S. P. & Maillard, J.-Y. 2010.Free-living amoebae and their intracellular pathogenic microorgan-isms: risks for water quality. FEMS Microbiol. Rev., 34:231–259.

Turner, N. A., Russell, A. D., Furr, J. R. & Lloyd, D. 2000. Emer-gence of resistance to biocides during differentiation of Acantha-moeba castellanii. J. Antimicrob. Chemother., 46:27–34.

Visvesvara, G. S., Moura, H. & Schuster, F. L. 2007. Pathogenic andopportunistic free-living amoebae: Acanthamoeba spp., Balamuthiamandrillaris, Naegleria fowleri, and Sappinia diploidea. FEMSImmunol. Med. Microbiol., 50:1–26.

Received: 02/15/11, 09/29/11, 11/07/11, 12/09/11; accepted: 12/10/11


Documents

The Effect of In Vitro Growth Conditions on the Resistance of Acanthamoeba Cysts