Journal of Hospital Infection (2009) 72, 342e350

Available online at www.sciencedirect.com

www.elsevierhealth.com/journals/jhin

New hospital disinfection processes for bothconventional and prion infectious agentscompatible with thermosensitive medicalequipment

S. Lehmann a,b,*, M. Pastore a,*, C. Rogez-Kreuz c, M. Richard d,M. Belondrade a, G. Rauwel e, F. Durand e, R. Yousfi c, J. Criquelion e,P. Clayette c, A. Perret-Liaudet d

a CNRS, Institut de Genetique Humaine UPR 1142, Universite Montpellier 1, Franceb CHU Montpellier, Laboratoire de Biochimie, Hopital Saint Eloi, Montpellier, Francec Neurovirology Laboratory, SPI-BIO, CEA, Fontenay aux Roses, Franced Hospices Civils de Lyon, CBPE, Service de Neurobiologie, Lyon, Francee Laboratoires ANIOS Lille Hellemmes, France

Received 7 March 2009; accepted 27 March 2009Available online 21 June 2009

KEYWORDSCopper;CreutzfeldteJakobdisease;Decontamination;Prion protein

* Joint corresponding authors. Addfax:þ33 4 99 61 99 31.

E-mail address: Sylvain.Lehmann@

0195-6701/$ - see front matter ª 200doi:10.1016/j.jhin.2009.03.024

Summary With the detection of prions in specific tissues in variant andsporadic CreutzfeldteJakob diseases, efficient decontamination for humantransmissible spongiform encephalopathy (TSE) agents, that is compatiblewith medical equipment, has become a major issue. We previously de-scribed the cleavage of prions on exposure to copper (Cu) and hydrogenperoxide (H2O2) and have used this property to develop efficient priondecontamination processes. To validate this approach, in-vitro assays ongenuine human and animal prions using both brain homogenates and steelwires to mimic contamination of medical equipment were conducted.In-vivo experiments using steel wire in the hamster 263 K model were thenused to evaluate the effect on prion infectivity. Assays on classical patho-gens following international norms completed these prion experiments.In-vitro data confirmed the full decontamination efficacy of H2O2/Cuon different TSE strains. Combination of Cu with peracetic acid, used forendoscope disinfection, also revealed improved prion decontamination.Animal assay demonstrated efficacy on TSE infectivity of H2O2/Cu alone

ress: CNRS, Institut de Genetique Humaine UPR 1142, Montpellier, F-34000, France. Tel.:/

igh.cnrs.fr

9 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.

Prion decontamination 343

or in combination with detergents (reduction factor �5.25 log10). Assays onclassical pathogens confirmed the disinfection properties of the differentprocesses. Taken together, these new disinfection processes are efficientfor both conventional and prion infectious agents and are, compatible withthermosensitive medical equipment. They can be adapted to hospitals’and practitioners’ routine use, and they present reduced risks for theenvironment and for healthcare professionals.ª 2009 The Hospital Infection Society. Published by Elsevier Ltd. All rightsreserved.

Introduction

Transmissible spongiform encephalopathies (TSEs),or prion diseases, are fatal neurodegenerative dis-orders that include scrapie and bovine spongiformencephalopathy (BSE) in animals, and Creutz-feldteJakob disease (CJD) in humans.1 HumanTSEs are unique in that they occur in sporadic(sCJD), genetic or infectious forms. These infec-tious forms are generally connected to a specifichistory of exposure to prions, in the past throughcannibalism and more recently through medicalprocedures.1,2 In 1996 in the UK a novel formnamed variant CJD (vCJD) was identified, linkedto consumption of BSE-tainted products.3 Onlya limited number of patients with vCJD have beendescribed (see website of the UK CreutzfeldteJakob Disease Surveillance Unit), but this newform, characterised by blood transmission and lym-phoid prion tissue distribution, raised awareness onnew TSE iatrogenic transmission risks.4,5 TSE infec-tivity is often associated with the presence of anabnormal protease-resistant form, PrPres, of thenormal host prion protein (PrP), and recent worksdemonstrated the presence of PrPres in peripheralorgans of patients with vCJD and sCJD, includingtonsil, appendix, spleen or muscles.5,6 In addition,the presence of asymptomatic carriers in the popu-lation was demonstrated in the UK.4 While knownCJD iatrogenic risks were previously linked to con-tact with the central nervous system, these findingsraised the spectre of transmission through commonsurgical and medical procedures, including endos-copy.7 Consequently, specific preventive recom-mendations have been proposed in medicalpractice.8 They include the use of disposablematerial, the identification of patients at risk(with previous neurosurgery, blood transfusion,extractive hormone treatment) and the decontam-ination using validated methods.

Infectious TSE agents have a long record ofextreme resistance to most decontaminationprocesses.9 Hence, decontamination reference

methods include the use of high concentration ofsodium hydroxide or sodium hypochlorite.9,10 How-ever, these chemicals are not compatible withvarious pieces of medical equipment, and theypresent a serious handling hazard for healthcareemployees. While being less efficient, prion inacti-vation by autoclaving at 134 �C (porous and steam)has also been proposed and is now often imple-mented in hospitals. However, this procedure isnot compatible with thermosensitive equipment,and alternative approaches using alkaline, acidic,enzymatic, or phenolic compounds, as well as gas-eous hydrogen peroxide, have then been proposedby several groups.11e15 In practice, the use ofthese decontamination methods is limited. Oneimportant issue is represented by their suitabilityfor routine disinfection (in terms of practicability,safety risk, device, cost, etc.). For endoscopes,which are particularly sensitive to extreme pH,chloride oxidants or high temperature, peraceticacid (PAA, which yields to an incomplete prion de-contamination) is nevertheless the recommendedmethod in France as a replacement for glutaralde-hyde, which is known to stabilise prions and to behazardous for healthcare employees.11,16

Following the observation that cellular PrPC

binds Cu, we and others have attempted to de-cipher the role of this metal ion in prion biol-ogy.17,18 We have demonstrated that both normalPrPC and the pathological PrPres isoform undergoa specific cleavage on exposure to copper (Cu)and hydrogen peroxide (H2O2).

19,20 We thereforeproposed to use this property to develop efficientprion decontamination processes. In the presentwork, we extended the initial work and confirmedthe decontamination effect of H2O2/Cu on genuinehuman prions, and we validated this new approachin-vitro and in-vivo using steel wires to mimiccontamination of medical equipment.21 We thendesigned full disinfection processes validated forboth conventional and TSE infectious agents, com-patible with thermosensitive medical equipment,and adapted for hospitals’ and practitioners’ use.

344 S. Lehmann et al.

Methods

Reagents for decontamination

The following reagents were tested: H2O2/Cu [(hy-drogen peroxide 7.5%/CuSO4 500 mmol/L (ANIOS)],AF [alkaline detergent, surfactants, chelatant(ANIOS)], PAA (5% PAA, 23% hydrogen peroxide,acetic acid and water), PAA/Cu [PAA supplementedwith Cu: PAA (600 ppm)þ 500 mmol/L of CuSO4], Np[detergent for manual process: tri-enzymaticcleaner solution (ANIOS)], Dp [detergent anddisinfectant for manual process: tri-enzymaticdetergent and disinfectant solution (ANIOS)], Nmp[detergent for washer (machine) process (used at40 �C): tri-enzymatic cleaner solution (ANIOS)].

Infectious material

Brain homogenate (BH) samples were obtainedfrom brains of either healthy (non-infected withTSE) or experimental 263 K-infected hamsters,RML- or 6PB1-infected mice at the terminal stageof the disease. For human brain samples, in allcases, informed consent for research was obtainedand the material had appropriate ethical approvalfor use in this project. BSE brain was generouslyprovided by Thierry Baron (AFSSA, Lyon France).For each sample, 10% or 20% BH (weight/volume)in phosphate-buffered saline (PBS) or 5% glucosesolution were prepared. The homogenates werethen filtered through a 20-gauge needle beforestorage at �80 �C.

Brain homogenate assays

Diluted BH (2%) were mixed for 30 min with a PBSsolution containing various concentrations of de-contamination reagents. This solution was thenmixed with an equal volume of 2� TritoneDOClysis buffer (150 mmol/L NaCl, 0.5% Triton X-100,0.5% sodium deoxycholate, and 50 mmol/L TriseHCl) and incubated for 15 min on ice. One-tenthof the solution was added to 5� Laemmli bufferto be used as proteinase K-negative (PK�) samples.The remaining samples were digested with PK(20 mg/mL) for 30 min at 37 �C. Digestion wasterminated by the addition of Pefabloc (4 mmol/Lfinal) and PrPres was recovered by centrifugationat 20 000 g at 4 �C (Benchtop, Beckmann CoulterInc., Fullerton, CA, USA) or by methanol precipi-tation. Pellets were dissolved with an equalvolume of 2� Laemmli buffer. All samples wereboiled for 5 min before western blot analysisusing a mixture of SAF60, SAF69, and SAF70antibodies as described.20

Steel wire in-vitro assays

Steel wires used for in-vitro assays were artificiallycontaminated with BH as described with slightmodifications as follows:14 for experimental TSE,10% BH were directly incubated with steel wires,whereas for human TSE, in order to increase sensi-tivity without affecting the global response of theassay (not shown), CJD-PrPres-enriched pellet ob-tained by sodium phosphotungstic acid precipita-tion was added to 10% BH of CJD patients beforecontamination of 40 steel wires, during 2 h at37 �C under gentle agitation.5 These stainless steelmaterials were dried overnight under a class II mi-crobiological safety cabinet at room temperature.Treatments by different solutions were performedunder gentle agitation (w50 rpm). A wash step inwater was systematically performed after treat-ment. For PK-treated samples, PK was incubated1 h at 37 �C directly on steel wires before desorp-tion of remaining PrPres by incubation in Laemmlibuffer at 100 �C during 10 min in a sonicator bath.Eluted samples were analysed by western blottingusing the TeSeE� Sheep/Goat western blot assayfrom Bio-Rad (Gif sur Yvette, France).

Animal bioassay

Syrian golden hamsters aged 6e8 weeks were usedin this study. Eight animals were included in each‘wire’ group and four animals were included ineach ‘BH’ group. Each animal was identified bymeans of an electronic tag and housed in level 3care facilities officially registered for prion exper-imental studies on rodents (agreement number A92-032-02 for animal care facilities, agreementnumber 92-189 for animal experimentation).

Stainless steel wire segments of 5 mm lengthwere artificially contaminated with the 263 Kstrain by immersion in 20% 263 K-infected BH for2 h at room temperature and dried overnight undera class II microbiological safety cabinet at roomtemperature.

Wires were treated under gentle agitation(w50 rpm) with different reagents. A water rinsewas performed between and at the end of eachtreatment.

After treatments, wires were individually im-planted into the prefrontal subcortical region ofanesthetised hamsters. Experimental animals wereexamined physically once to twice a week. Clinicalsigns of TSE and deaths were recorded. Animals atthe terminal stage of the disease were euthanisedand diagnosis of TSE was confirmed by detection ofPrPres in brains by western blot using the

Prion decontamination 345

monoclonal peroxidase-conjugated anti-rodentPrP SAF 83 antibody.22 All surviving healthy animalswere euthanised 12 months after inoculation andwere confirmed to be PrPres negative by westernblot. Reduction factors were calculated using thefour-parameter standard curve (Table I).

Results

In-vitro impact on PrPres from brainhomogenates (BH)

From the perspective of generating realistichospital disinfection processes, we selecteda new formulation of Cu at 500 mmol/L witha higher concentration of H2O2 at 7.5%, recognisedto be efficient against conventional infectious

Table I Animal assay (titration and wire experiments)

Sample

263 K BH, dilution:10�3

10�4

10�5

10�6

10�7

10�8

10�9

10�10

10�11

10�12

Wires:Control (no treatment)Water (10 min)H2O2/Cu (30 min)AF (10 min)Autoclaving (steam, 134 �C, 18 min)NpeNpeH2O2/Cu (10 mine5 mine15 min)DpeDpeH2O2/Cu (10 mine5 mine15 min)NpeDpeH2O2/Cu (10 mine5 mine15 min)NmpeNmpePAA/Cu (10 mine5 mine10 min, 40 �C)

BH, brain homogenate; AF, alkaline detergentesurfactantsechelattein; Np, detergent for manual process; Dp, detergent and disinfe

a Calculated using the date of death of animals confirmed to beb Calculated using the four-parameter standard curve as follows:

a four-parameter standard curve. For each group of treated wires,dilution of inoculum presenting an equivalent infectivity. The log10

follows: log10 (corresponding dilution of the treated group) e log1

each group of animals, the mean duration was calculated includinlopathy-infected animals were confirmed as PrPres positive by westudy were confirmed as PrPres negative by western blot. Thistime and the variation in transmission rate. Indeed, the date of euin the mean duration. When compared with a standard curve pemethod is simpler and more precise. Indeed, it excludes a causethat may be dependent on the BH dilution even when dilutions ar

agents. This new formulation was tested usingour standard BH paradigm.20 It demonstrated a use-ful effect towards PrPres from human (sCJD, vCDJ)and animal (genuine BSE, 263 K) prions (Figure 1AeD).In fact for all these strains, PrPres signal com-pletely disappeared as early as 15 min after treat-ment. This was true for both PK� and PKþ

samples, thus avoiding the risk that modificationof PK efficacy could account for the results.23 Inthe same experiment, a strong alkaline solution(AF) was used as a control. As reported previ-ously, alkaline treatment acts by increasing PKsensitivity of PrPres, explaining the differentialsignal between PK� and PKþ samples.15 Overallboth treatments, H2O2/Cu and AF, resulted ina comparable disappearance of PrPres signal.

We also investigated whether the limited priondecontamination efficacy of PAA could be improved

Transmissionrate (%)

Time of death,days (SD)a

Log10

reductionb

100 90 (4) e100 110 (6) e100 139 (23) e100 155 (27) e100 223 (1) e75 284 (3) e0 >370 e0 >370 e0 >370 e0 >370 e

100 90 (4) e100 103 (16) 0.98

0 �370 �5.250 �370 �5.25

57 140 (27) 4.1143 133 (7) 4.5520 159 (0) �5.250 �370 �5.25

67 102 (12) 3.43

ant; PrPres, protease-resistant form of normal host prion pro-ctant for manual process; PAA, peracetic acid.PrPres positive by western blot.the mean duration on the dilutions of BH was plotted to definethe mean duration was reported on the curve to determine thereduction factor related to each treatment was calculated as

0 (corresponding dilution of the untreated control group). Forg all animals of the group: transmissible spongiform encepha-stern blot, and healthy animals euthanised at the end of thecalculation takes into account both the delay in incubationthanasia of healthy animals at the end of the study is includedrformed using wires contaminated with limit-diluted BH, this

of variability linked to the yield of PrPres coating on wirese performed in healthy homogenate.

PK + PK −

1 2 3 4 5 1 2 3 4 5

1 2 3 4

1 2 3 4

5 6 7

263K

vCJD

vCJD

PK + PK −

1 2 3 4 5 1 2 3 4 5

sCJD

A PK + PK −

1 2 3 4 5 1 2 3 4 5

BSE

B

DPK + PK −

1 2 3 4 5 1 2 3 4 5

vCJD

C

1 2 3 4 5 6

RML

E

1 2 3 4 5 6 7

sCJD

G H

F

Figure 1 In-vitro assays on human and animal PrPres (protease-resistant form of normal host prion protein). (AeD)Each panel represents experiments performed with the following infected brain homogenate (BH): sporadic Creutz-feldteJakob disease (sCJD) type 1 homozygous M/M at codon 129, new variant CJD (vCJD), genuine bovine spongiformencephalopathy (BSE) and hamster 263 K scrapie (263 K). Proteinase K-negative (PK�) and PK-positive (PKþ) laneswere obtained with the equivalent of 50 mg and 550 mg of BH per lane, respectively. Lane 1: untreated BH; lanes 2and 3: BH treated with H2O2/Cu (7.5%/500 mmol/L) for 15 min and 30 min, respectively; lanes 4 and 5: BH treatedfor 10 min with alkaline detergent (AF) at final concentrations of 0.5% and 1%, respectively. (E) 50 mg protein equiv-alent of mouse RML BH were analysed. Before western blot detection of PrP, BH were incubated for 30 min in the pres-ence of 80 mmol/L of H2O2 [equivalent to the concentration of H2O2 present in peracetic acid (PAA) solution]supplemented with 0, 125 and 500 mmol/L of Cu (lanes 1, 2 and 3, respectively). This was compared with the effectof PAA solution (600 ppm) supplemented with 0, 125 and 500 mmol/L of Cu (lanes 4, 5 and 6, respectively). (F) 550 mg

346 S. Lehmann et al.

Prion decontamination 347

by adding Cu to the working solution.11,16 Therationale for this was based on the fact that PAA isoften generated by an equilibrium between aceticacid and H2O2.

16 When Cu was added to PAA(Figure 1E and F) a clear diminution of PrP signalwas observed. However, the remaining signal (evenwhen concentrations of Cu >500 mmol/L were usede not shown) was higher than that obtained withoutPAA, suggesting that this product, or the chemicalenvironment, was interfering. This result wasencouraging in respect of improving washerdisinfection of endoscopes that are reliant on PAA.

In-vitro impact on PrPres adsorbed on steelwires

Steel wires contaminated with BH and dried tomimic medical equipment contamination weregenerated using sCJD and vCJD prions.14,21 Hospi-tal disinfectant processes that included a doublecleaning phase based on enzymatic detergent(Np) or enzymatic detergent disinfectant (Dp)scheme were combined with H2O2/Cu treatment.Combination of NpeNpeH2O2/Cu, DpeDpeH2O2/Cu or NpeDpeH2O2/Cu had an efficacy on sCJDor vCJD similar to that of H2O2/Cu alone(Figure 1GeH). Based on the semiquantitativeevaluation of the signal (lanes 1e3), we estimatedthat the reduction was �3 log10.

In-vivo prion steel wire assays

We used the standard steel wires assay in thehamster 263 K model and the SpearmanneKarbercalculation method for the titration curve (Table I).24

The infectious titre adsorbed on one wire beforeany treatment was 5.25 log10 median infectivedose (ID50). Reduction factors were calculated us-ing the four-parameter standard curve taking intoaccount both the modification of the incubation-period assay and the transmission rate. After onewater rinse, the infectivity was reduced by0.98 log10 (Table I). By contrast, the followingwater rinses did not further remove prion infectiv-ity from the wires (data not shown). The H2O2/Cusolution, as well as the alkaline agent AF, achieveda full decontamination effect with a reduction ininfectivity of >5.25 log10 and a transmission rate

protein equivalent of vCJD BH was analysed by western bloLane 1: untreated; lane 2: H2O2/Cu (80 mmol/L/500 mmol/500 mmol/L). (GeH) Western blot detection of PrPres eluted f1, 2, 3: untreated control wires contaminated with undilute4e5: wires contaminated with undiluted BH and treated a15 min) (lane 4), with NpeNpeH2O2/Cu (lane 5), DpeDpeH2

the different phases: 10 mine5 mine15 min).

of 0%. Steam autoclaving at 134 �C did not achievea complete prion decontamination (transmissionrate of 57%) which was in agreement with previousobservations.11

As expected, when the complete hospital pro-cedures were tested, a prion decontaminationeffect of >5.25 log10 was obtained when at leastone step included Dp (NpeDpeH2O2/Cu and DpeDpeH2O2/Cu). The titre reduction was 4.55 log10

when only Np was used (NpeNpeH2O2/Cu) suggest-ing a lesser efficiency of Np in comparison with Dp.Finally, the detergent steps for washer (machine)applications (NmpeNmp) followed by PAA comple-mented with Cu achieved a reduction in infectivityof 3.44 log10, which was 1 log10 more efficient thanthat described for PAA alone in similar experimen-tal conditions.11,16 The reduced H2O2 concentra-tion of the PAA solution and/or the effect of thePAA itself may explain the lower efficient priondecontamination observed.

Disinfection activities for classicalpathogens

From the perspective of hospital use we evaluatedthe disinfection activities for classical pathogensof the above-mentioned formulations using currentEuropean norms for bactericidal, mycobacterici-dal, fungicidal, virucidal and sporicidal activities(NF EN 13727, NF EN 14561, NF EN 14348, Pr EN14563, NF EN 13624, NF EN 14562, NF EN14476þ A1 and AFNOR NF T 72-230). These normsare being used by regulatory authorities to vali-date disinfection of medical equipment; they de-fine very precisely the agents to use, and theexperimental conditions to follow (in suspension,on carrier, etc.). H2O2/Cu treatment achieved inall cases the required efficacy (Table II).

Discussion

Our investigations into the functional relationshipbetween PrP and metal ions have demonstratedthat physiological PrPC undergoes a specific cleav-age on exposure to copper (Cu) and hydrogenperoxide (H2O2).

18,19 This cleavage phenomenon,linked to the Fenton reaction, was also observed

t after PK digestion and a 30 min incubation as follows.L); lane 3: PAA (600 ppm); lane 4: PAA/Cu (600 ppm/rom steel wires contaminated with sCJD and vCJD. Lanesd BH, or BH diluted 102 or 103 times, respectively; lanest room temperature: with H2O2/Cu (7.5%/500 mmol/L,O2/Cu (lane 6) and NpeDpeH2O2/Cu (lane 7) (times for

Table II Disinfection efficacy of H2O2/Cu (7.5%/500 mmol/L) for classical pathogens

Norm Organism Description Reduction factor (RF)

NF EN 13727 Enterococcus hirae ATCC10541Pseudomonas aeruginosaATCC 15442Staphylococcus aureus ATCC6538

Chemical disinfection.Quantitative bactericidal assayin suspension. Albumin 0.3 g/L

RF of 105 achieved in15 min with H2O2/Cu

NF EN 14561 Enterococcus hirae ATCC10541Pseudomonas aeruginosaATCC 15442Staphylococcus aureus ATCC6538

Quantitative bactericidalassay on carrier. Albumin 0.3 g/L

RF of 105 achieved in15 min with H2O2/Cu

NF EN 14348 Mycobacterium terrae ATCC5755 & avium ATCC 5769

Quantitative mycobactericidalassay in suspension. Albumin 0.3 g/L

RF of 104 achieved in15 min with H2O2/Cu

pr EN 14563 Mycobacterium terrae ATCC5755 & avium ATCC 5769

Quantitative mycobactericidalassay on carrier. Albumin 0.3 g/L

RF of 104 achieved in15 min with H2O2/Cu

NF EN 13624 Candida albicans ATCC 10231Aspergillus niger ATCC 16404

Chemical disinfection.Quantitative fungicidal assayin suspension. Albumin 0.3 g/L

RF of 104 achieved in15 min with H2O2/Cu

NF EN 14562 Candida albicans ATCC 10231Aspergillus niger ATCC 16404

Chemical disinfection.Quantitative fungicidal assayon carrier. Albumin 0.3 g/L

RF of 104 achieved in15 min with H2O2/Cu

NF EN 14476þ A1 Adenovirus type 1(75 ATCC VR-5)Poliovirus type 1 (LSc-2ab)

Chemical disinfection.Quantitative virucidal assayin suspension. Albumin 0.3 g/L

RF of 104 achieved in15 min with H2O2/Cu

AFNOR NF T72-230

Bacillus cereus CIP 7 803Bacillus subtilis var. nigerCIP 7 718Clostridium sporogenes51 CIP 7 939

Chemical disinfection. Quantitativesporicidal assay in suspension.Dilution/neutralisation method,at 20 �C

RF of 105 achieved in15 min with H2O2/Cu

pr EN 14563 Mycobacterium terraeATCC 5755

Chemical disinfection. Quantitativemycobactericidal assay on carrier.Albumin 3 g/L, erythrocytes 3 mL/L

RF of 102.2 for Dp-DpRF �104 for H2O2/CuRF �104 for Dp-Dp-H2O2/Cu

Mycobacterium aviumATCC 5769

Chemical disinfection. Quantitativemycobactericidal assay on carrier.Albumin 3 g/L, erythrocytes 3 mL/L

RF of 101.4 for Dp-DpRF �104 for H2O2/CuRF �104 for Dp-Dp-H2O2/Cu

348 S. Lehmann et al.

for PrPres suggesting that this property could beused for prion decontamination.20 The molecularmechanism responsible for this PrPres cleavageand degradation might be linked to an exchangeof the metal ion content of PrPres molecules, asso-ciated with a change in conformation and an in-crease in sensitivity to proteases and to theFenton reaction.25

Importantly, H2O2 had been already recognisedas an efficient disinfectant for conventional infec-tious agents, but at concentrations higher thanthose used in our initial study. In order to developa formulation that will be efficient on both prionsand conventional agents, we therefore focusedon formulations of Cu at 500 mmol/L with

a concentration of H2O2 as high as 7.5%. Thedisappearance of PrPres (Figure 1AeD) fromdifferent human and animal brain homogenatesconfirmed the efficacy of the formulation, inde-pendent of prion strain. Similar results wereobtained with a strong alkaline solution (AF) usedas a control.

Disinfection of endoscopes relies in part on theuse of PAA, which has a limited prion inactivationcapacity.11,16 As PAA solutions are generated by anequilibrium between acetic acid and H2O2 we alsotested and confirmed that the addition of Cu couldimprove PAA efficacy (Figure 1EeF).16 Endoscopesare disinfected through a procedure that commonlyincludes a double detergent/disinfectant phase. To

Prion decontamination 349

mimic medical equipment contamination and asa preliminary study to animal assays, we thereforecombined this procedure with Cu/H2O2 and testedit on PrPres adsorbed on steel wires. H2O2/Cu treat-ment alone or in combination with enzymaticdetergent (Np) or enzymatic detergent disinfectant(Dp) achieved a complete disappearance of PrPresadsorbed on wires.

In parallel with these biochemical analyses, itwas important to complete the evaluation using thereference in-vivo experiments with steel wires inthe hamster 263 K model. As expected, water alonedid not result in significant reduction in prion infec-tious titre; PAA alone had a poor efficacy but thiswas improved by the addition of Cu. As reported be-fore, autoclaving did not achieve a complete priondecontamination.11,26 Detergent in combinationwith H2O2/Cu resulted, as did H2O2/Cu alone, incomplete prion decontamination similar to thatobtained with an alkaline treatment (�5.25 log10).

Using experimental protocols defined in therelevant French and European norms, we alsoconfirmed that the improved H2O2/Cu formulationcharacterised by an increased concentration ofperoxide reached the requested bactericidal, my-cobactericidal, fungicidal, virucidal and sporicidalactivities (Table II).

Using the most accepted animal model (263 Kstrain adapted to hamster), our results have justifiedfurther examination of prion decontamination pro-cedures based on the synergistic combinationH2O2/Cu alone or combined with detergents/dis-infectants. Experiments on a representative modelof human TSE are ongoing. Formulation and pro-cedures describedhere are compatible with thermo-sensitive equipment such as endoscopes and offerreduced risks for theenvironmentand for healthcareprofessionals. They are also adapted for use with ex-isting and routine hospital disinfection procedures,i.e. for conventional agents, meaning that theirimplementation will require minimal adaptation.

Acknowledgements

We thank J. Grassi and colleagues (CEA Saclay,France), J-M. Bilheude, C. Mazure and G. Nespou-lous (Bio-Rad, Gif sur Yvette, France) for providinganti-prion antibodies and technical advice; C.Creminon (CEA Saclay, France) for his insight intoH2O2/Cu anti-prion mechanism, and S. Lansardierefor technical help.

Conflict of interest statementThe CNRS has patented the H2O2/Cu anti-priondecontamination methods (patent published

25 November 2005, # 2870458). The companyANIOS has a research collaboration with theCNRS. G. Rauwel, F. Durand, and J. Criquelionare employees of the manufacturer ANIOS.ANIOS had no role in the collection and analysisof prion data and it contributed mainly in pro-viding working solutions, in testing for classicalpathogen disinfection and in the design of pro-cesses compatible with hospital practice andrisk management.

Funding sourcesThis work has been partly financially supportedby ANIOS.

References

1. Prusiner SB, Scott MR, Dearmond SJ, Cohen FE. Prion pro-tein biology. Cell 1998;93:337e348.

2. Wadsworth JD, Collinge J. Update on human prion disease.Biochim Biophys Acta 2007;1772:598e609.

3. Will RG, Ironside JW, Zeidler M, et al. A new variant ofCreutzfeldteJakob disease in the UK. Lancet 1996;347:921e925.

4. Hilton DA. Pathogenesis and prevalence of variant Creutz-feldteJakob disease. J Pathol 2006;208:134e141.

5. Wadsworth JD, Joiner S, Hill AF, et al. Tissue distribution ofprotease resistant prion protein in variant CreutzfeldteJakob disease using a highly sensitive immunoblottingassay. Lancet 2001;358:171e180.

6. Peden AH, Ritchie DL, Head MW, Ironside JW. Detection andlocalization of PrPSc in the skeletal muscle of patients withvariant, iatrogenic, and sporadic forms of CreutzfeldteJakob disease. Am J Pathol 2006;168:927e935.

7. Brown P, Gibbs Jr CJ, Rodgers-Johnson P, et al. Humanspongiform encephalopathy: the National Institutes ofHealth series of 300 cases of experimentally transmitteddisease. Ann Neurol 1994;35:513e529.

8. WHO. Infection control guidelines for transmissible spongi-form encephalopathies. Report 23e26 March 1999,WHO/CDS/CSR/APH/2000.3. Geneva: WHO; 1999.

9. Taylor DM. Inactivation of prions by physical and chemicalmeans. J Hosp Infect 1999;43(Suppl.):S69eS76.

10. Brown P, Rohwer RG, Gajdusek DC. Sodium hydroxidedecontamination of CreutzfeldteJakob disease virus. N EnglJ Med 1984;310:727.

11. Fichet G, Comoy E, Duval C, et al. Novel methods for disin-fection of prion-contaminated medical devices. Lancet2004;364:521e526.

12. Jackson GS, McKintosh E, Flechsig E, et al. An enzyme-detergent method for effective prion decontamination ofsurgical steel. J Gen Virol 2005;86:869e878.

13. Peretz D, Supattapone S, Giles K, et al. Inactivation ofprions by acidic sodium dodecyl sulfate. J Virol 2006;80:322e331.

14. Lemmer K, Mielke M, Pauli G, Beekes M. Decontaminationof surgical instruments from prion proteins: in vitro studieson the detachment, destabilization and degradationof PrPSc bound to steel surfaces. J Gen Virol 2004;85:3805e3816.

15. Baier M, Schwarz A, Mielke M. Activity of an alkaline‘cleaner’ in the inactivation of the scrapie agent. J HospInfect 2004;57:80e84.

350 S. Lehmann et al.

16. Taylor DM. Resistance of the ME7 scrapie agent to peraceticacid. Vet Microbiol 1991;27:19e24.

17. Brown DR, Qin K, Herms JW, et al. The cellular prion proteinbinds copper in vivo. Nature 1997;390:684e687.

18. Lehmann S. Metal ions and prion diseases. Curr Opin ChemBiol 2002;6:187e192.

19. McMahon HE, Mange A, Nishida N, Creminon C, Casanova D,Lehmann S. Cleavage of the amino-terminus of the prionprotein by reactive oxygen species. J Biol Chem 2001;276:2286e2291.

20. Solassol J, Pastore M, Crozet C, Perrier V, Lehmann S. Anovel copper-hydrogen peroxide formulation for priondecontamination. J Infect Dis 2006;194:865e869.

21. Zobeley E, Flechsig E, Cozzio A, Enari M, Weissmann C. In-fectivity of scrapie prions bound to a stainless steel surface.Mol Med 1999;5:240e243.

22. Bach S, Talarek N, Andrieu T, et al. Isolation of drugs activeagainst mammalian prions using a yeast-based screeningassay. Nat Biotechnol 2003;21:1075e1081.

23. Solassol J, Arlotto M, Lehmann S. Detection of prion afterdecontamination procedures: comparative study of stan-dard western blot, filter retention and scrapie-cell assay.J Hosp Infect 2004;57:156e161.

24. Karber G. Beitrag zur kollektiven Behandlung pharmakolo-gischer Reihenversuche. Arch Exp Pathol Pharmakol 1931;162:956e959.

25. Wadsworth JD, Hill AF, Joiner S, Jackson GS, Clarke AR,Collinge J. Strain-specific prion-protein conformation de-termined by metal ions. Nat Cell Biol 1999;1:55e59.

26. Taylor DM. Inactivation of transmissible degenerativeencephalopathy agents: a review. Vet J 2000;159:10e17.