Karin Lemmer, 1 Martin Mielke, 2 Christine Kratzel, 1 Marion Joncic, 1 Muhsin Oezel, 3 Georg Pauli 4 and Michael Beekes 1 INTRODUCTION

Transcription

1 Journal of General Virology (2008), 89, DOI /vir Decontamination of surgical instruments from prions. II. In vivo findings with a model system for testing the removal of scrapie infectivity from steel surfaces Karin Lemmer, 1 Martin Mielke, 2 Christine Kratzel, 1 Marion Joncic, 1 Muhsin Oezel, 3 Georg Pauli 4 and Michael Beekes 1 Correspondence Karin Lemmer LemmerK@rki.de Michael Beekes BeekesM@rki.de 1 P24, Transmissible Spongiform Encephalopathies, Robert Koch-Institut, Nordufer 20, Berlin, Germany 2 FG 14, Applied Infection Control and Hospital Hygiene, Robert Koch-Institut, Nordufer 20, Berlin, Germany 3 ZBS4, Centre for Biological Safety Imaging Techniques for Rapid Morphology-Based Diagnostics of Infectious Organisms, Robert Koch-Institut, Nordufer 20, Berlin, Germany 4 ZBS1, Centre for Biological Safety Highly Pathogenic Viruses, Robert Koch-Institut, Nordufer 20, Berlin, Germany Received 24 August 2007 Accepted 29 September 2007 The unusual resistance of agents causing transmissible spongiform encephalopathies (TSEs) to chemical or thermal inactivation requires special decontamination procedures in order to prevent accidental transmission of these pathogens by surgical instruments. In the search for effective, instrument-compatible and routinely applicable decontamination procedures, a previous study [Lemmer, K., Mielke, M., Pauli, G. & Beekes, M. (2004). J Gen Virol 85, ] identified promising reagents in an in vitro carrier assay using steel wires contaminated with the diseaseassociated prion protein, PrP Sc. In the follow-up study presented here, these reagents were validated for their decontamination potential in vivo. Steel wires initially loaded with 3¾10 5 LD 50 of 263K scrapie infectivity were implanted into the brains of hamsters after treatment for decontamination and subsequently monitored for their potential to trigger clinical disease or subclinical cerebral PrP Sc deposition within an observation period of 500 days. It was found that routinely usable reagents such as a commercially available alkaline cleaner (ph 12.2) applied for 1 h at 23 6C or for 10 min at 55 6C and a mixture of 0.2 % SDS and 0.3 % NaOH (ph 12.8) applied for 5 or 10 min at 23 6C achieved removal of 263K scrapie infectivity below the threshold of detection (titre reduction of 5.5 log 10 units). The increasing use during the past few years of similar model systems by different research groups will facilitate comparison and integration of findings on the decontamination of steel surfaces from prions. Methods identified as highly effective in the 263K steel wire model need to be validated for human TSE agents on different types of instrument surfaces. INTRODUCTION Transmissible spongiform encephalopathies (TSEs) are fatal neurodegenerative diseases affecting the central nervous system (CNS) of animals and humans. This group of diseases includes scrapie in sheep, bovine spongiform encephalopathy (BSE) in cattle and chronic wasting disease in cervids, as well as Creutzfeldt Jakob disease (CJD) and variant CJD (vcjd) in humans. The pathognomonic hallmark of TSEs is the deposition in the CNS of a disease-associated form of the prion protein (PrP) with an aberrantly altered folding and/or aggregation structure. According to the prion hypothesis, the causative and transmissible agent of TSEs consists of proteinaceous infectious particles or prions that are composed essentially of misfolded PrP, referred to as PrP Sc (Prusiner, 1982, 1998). Whether TSE agents contain other factors in addition to the prion protein, e.g. polyanions (Deleault et al., 2007), remains to be established. Precautionary measures aiming at the prevention of iatrogenic CJD or vcjd transmissions in nosocomial settings have to take into account several different modes of transmission, particularly via blood and blood products, organ and tissue transplants, and surgical instruments and medical devices (Beekes et al., 2004). With respect to the G 2008 SGM Printed in Great Britain

2 Decontamination of surgical instruments from prions latter, reliable decontamination procedures that effectively counter the theoretical risk of CJD and vcjd transmission from asymptomatic pre- or subclinically infected humans without compromising the practical requirements for routine reprocessing of surgical instruments are required. However, the unusually high resistance of TSE agents to conventional methods of inactivation (reviewed by Taylor, 2000, 2004) and their high binding affinity to, as well as their tenacity on, steel surfaces (Zobeley et al., 1999; Flechsig et al., 2001) constitute a substantial challenge for the development of effective yet instrumentcompatible and routinely applicable decontamination procedures. In the search for effective and instrument/material-friendly decontamination procedures that can be applied to routine reprocessing of surgical instruments, we previously used an in vitro carrier assay in which infectious 263K scrapie brain homogenate from hamsters was fixed to the surface of steel wires (Lemmer et al., 2004). Batches of these wires were subjected to a screening approach using various chemical decontamination procedures, and the degrading, destabilizing and detaching activity of the tested reagents against PrP Sc was examined. In the in vitro assay, routinely usable reagents such as a mixture of 0.2 % SDS and 0.3 % (0.075 M) NaOH, a commercially available alkaline cleaner, 5 % SDS and a disinfectant containing 0.2 % peracetic acid (PAA) and NaOH in low concentrations ( M) were found to exert a pronounced decontamination activity on PrP Sc. Here, we present the results of a follow-up study in which these and other candidate reagents for decontamination were validated by bioassay. In this study, contaminated steel wires were implanted intracerebrally into hamsters after reprocessing and subsequently monitored for their potential to trigger clinical or subclinical (i.e. asymptomatic) infection within an observation period of 500 days. METHODS Reagents for decontamination. The following reagents were tested (Lemmer et al., 2004): 1.0 M NaOH, 2.5 % NaOCl ( p.p.m. available chlorine), 5 % SDS, 0.25 % PAA (titrated before use), an alkaline cleaner for medical devices (Baier et al., 2004) at concentrations of 0.5 and 1 %, a mixture of 0.2 % SDS and 0.3 % (0.075 M) NaOH, and a disinfectant containing 0.2 % PAA and NaOH in the range of % ( M). Distilled water served as a control. Contamination of wires. Preparation of the stainless steel wires (4 mm60.25 mm; Forestadent) and of the 263K scrapie brain homogenate from hamsters used for contamination were performed as described in detail previously (Lemmer et al., 2004). Carriers were coated in batches of 12 wires by incubation on a thermomixer in 150 ml of 10 % or further diluted scrapie brain homogenate in 2 ml tubes for 2 h at 37 uc under constant shaking at 700 r.p.m. Control batches of wires were either similarly incubated in 150 ml of a 10 % normal hamster brain homogenate or not incubated in brain homogenate. Subsequent to coating, wires were air-dried overnight in Petri dishes. Processing of wires for decontamination. Air-dried wires coated with 10 % scrapie brain homogenate were incubated for decontamination in solutions (1.5 ml) of the respective test reagents on a thermomixer (400 r.p.m.) at the temperatures and for the times specified in Tables 2 and 3. Subsequently, wires were rinsed in distilled water (once for 1 min, followed by four 10 min washes in a volume of 45 ml each) and again air-dried in Petri dishes. A proportion of control wires and of wires that underwent chemical decontamination was subjected to further treatment by steam sterilization. For this purpose, dried wires were sealed in sterilization paper and exposed to porous load steam sterilization at 121 uc for 15 min (control wires only) or at 134 uc for 5 min (control wires and chemically decontaminated wires). Scanning electron microscopy (SEM). SEM was performed for visual examination of the surfaces of contaminated wires after or without decontamination. Wires were mounted on sample stubs, sputter-coated with 5 nm gold/palladium (Polaron Sputter Coating unit E 5100; GaLa Instrumente) and examined using a LEO 1530 scanning electron microscope with field emission gun (Carl Zeiss SMT AG) between 3 and 5 kv. SEM magnifications were in the range of Studies in animals The studies in animals complied with German legal regulations and were approved by the responsible authorities. Male and female Syrian hamsters were infected by implantation of scrapie-contaminated wires into the brain using a stereotaxic instrument (Stoelting). Wires incubated in normal brain homogenate or native wires that were not coated with brain homogenates or subjected to reprocessing procedures were used as negative controls. Approximately 8-week-old hamsters were sedated with isoflurane for anaesthesia by intramuscular injection of a mixture of ketamine and xylazine. Anaesthetized hamsters were placed in the stereotaxic instrument using ear bars with short tips. One wire was implanted into each animal at the following coordinates: bregma, 22 mm/mediolateral 2 mm (Yan et al., 2004). For exact dorsoventral positioning of the wires at a depth of 4 8 mm (upper and lower ends, respectively) below the outer skull surface, we used a specially crafted needle with a mandrel as a pushing bar (Hero). All recipients of wires were marked with a transponder. The intracerebral implantation of wires was well tolerated by the animals. Hamsters were monitored at least twice a week for clinical signs of scrapie. When terminally affected with scrapie, at pre-defined time points during the incubation period or after asymptomatic survival for an observation period of 500 days post-implantation (p.im.), hamsters were sacrificed by CO 2 asphyxiation. End-point titration. For end-point titration, wires were incubated as described above in 263K scrapie hamster brain homogenates that had been serially diluted in logarithmic steps in normal brain homogenate over a range of dilutions from to After drying, wires were rinsed in distilled water (once for 1 min, followed by four 10 min washes in a volume of 45 ml each) to remove unfixed tissue debris (which might affect the titration by being unevenly attached to different carriers) and dried again. Survival times and attack rates for the development of terminal scrapie were monitored in hamsters for an observation period of 500 days p.im. The end-point dilution of infectivity was calculated from the observed rates of terminal scrapie per dilution according to the Spearman Kärber method as described by Bonin (1973). Bioassay of wire decontamination. The efficiency of test reagents and procedures for surface decontamination was assessed by monitoring the survival times and attack rates for the development of terminal scrapie following cerebral wire implantation in reporter animals over an observation period of 500 days p.im. Bioassays 349

3 K. Lemmer and others after chemical and thermal treatments or after chemical decontamination followed by steam sterilization were performed in groups of six hamsters each. Substances and formulations newly identified in our previous in vitro carrier assay as promising reagents for decontamination (Lemmer et al., 2004) and showing efficient removal of infectivity in the first set of bioassays (see Table 3, Bioassay group 1) were retested with independently prepared and processed wires (see Table 3, Bioassay group 2). However, this could not be done with the 1 % alkaline cleaner applied at 23 uc for 1 h due to restrictions on animal experiments (bioassay data on a similar regimen applied to scrapie brain homogenate have been published previously by the Robert Koch Institute; Baier et al., 2004). For the important reference treatments of 1 M NaOH and 2.5 % NaOCl, bioassays were repeated once without steam sterilization. Unusual or unclear results were validated by duplicate bioassays where necessary, but retesting was not performed when a regimen clearly failed to decrease the level of infectivity to or below the threshold of detection in the first set of bioassays. Time-course study of cerebral PrP Sc deposition. To examine the temporal spatial course of cerebral PrP Sc deposition, animals were implanted with unprocessed wires coated with 10 % 263K brain homogenate and sacrificed at 1, 5, 15, 30, 45 and 60 days p.im., as well as at the terminal stage of scrapie, for paraffin-embedded tissue (PET) blot examination. PET blotting. PET blot analysis of cerebral PrP Sc deposition was performed to detect subclinical scrapie infections in animals that did not show onset of disease during the period of clinical observation. Brains were cut coronally in four blocks and the brain tissue block containing the wire was fixed in 4 % paraformaldehyde whilst the remaining blocks were stored at 270 uc. To minimize tissue damage in the area of the wire channel, in most cases the upper part of the fixed tissue block was separated from the lower part by a horizontal cut near the upper wire end before the steel carrier was removed. Subsequently, PET blotting of blocks containing empty wire channels was carried out as described elsewhere (Schulz-Schaeffer et al., 2000; Thomzig et al., 2004). RESULTS SEM of steel wire surfaces For visual assessment of the efficacy of selected decontamination procedures, we examined the surface appearance of wires for several of the tested chemical treatments by SEM. Native steel wires washed in 2 % Triton X-100 for 15 min under constant ultrasonication (Lemmer et al., 2004) showed even surfaces with no prominent roughness (Fig. 1a). In contrast, wires coated with 263K brain homogenate, air-dried and rinsed in water were covered with prominent deposits of residual brain homogenate (Fig. 1b). After decontamination with 1.0 M NaOH for 60 min at 23 uc, the wire surface again resembled that of native steel wires (Fig. 1c). Similarly efficient cleaning was also observed for treatment with 2.5 % NaOCl (1 h at 23 uc; not shown), 1 % alkaline cleaner (10 min at 55 uc; not shown), 0.2 % SDS/0.3 % NaOH (10 min at 23 uc; Fig. 1d) and disinfectant containing PAA and NaOH used for 120 min (Fig. 1e). In contrast, incubation of wires in 0.25 % PAA (5 or 60 min at 23 uc) resulted in a strongly encrusted wire surface (Fig. 1f) indicating incomplete cleaning. End-point titration of the binding of 263K infectivity to steel wires In order to define the sensitivity of the steel wire bioassay and to establish a dose response relationship for the assessment of titre reductions achieved by the tested decontamination procedures, we performed an end-point titration experiment. Sets of wires were incubated in 150 ml 263K scrapie brain homogenate that had been diluted from to After coating and drying, the wires were rinsed in water and again air-dried. Following cerebral implantation of the wires, survival times and attack rates for the development of terminal scrapie were monitored in hamsters for an observation period of 500 days. The results are summarized in Table 1. The scrapie brain homogenate used for the end-point titration was produced from donor hamsters in the terminal stage of scrapie. As found repeatedly in our laboratory over the last two decades, such brain tissue contains an infectivity titre of approximately 10 9 intracerebral LD 50 (LD 50i.c. ) per gram of tissue (Beekes et al., 1996; Diringer et al., 1997, 1998; Weber et al., 2007). The endpoint titration experiment revealed that the dilution of scrapie brain homogenate leading to 1 LD 50 following intracerebral implantation (LD 50i.c.imp ) for wires prepared as described above was Therefore, carriers incubated with the dilution of scrapie brain homogenate were associated with approximately 0.3 LD 50i.c.imp per wire. Accordingly, wires coated with the dilution of scrapie brain homogenate could be expected to carry an initial infectious load of approximately ( )LD 50i.c.imp. Bioassays of chemical decontamination of steel wires The efficacy of the decontamination of steel wires from 263K scrapie agent was assessed by comparing the initial load of infectivity with that remaining attached to the wires after processing. The infectious titre of test wires before processing was LD 50i.c.imp (note that, other than for carriers used in the end-point titration experiment, wires were not rinsed in water before decontamination and thus may have been contaminated with unfixed tissue debris in addition to fixed infectivity). Residual wire infectivity was deduced from attack rates and survival times using the dose response relationship established in the end-point titration experiment (Table 1). Our in vivo assay allowed us to demonstrate titre reductions of 5.5 log 10 units (logs) when no residual infectivity could be detected after the processing of wires. In cases where infectivity was at the threshold of detection, i.e. in the order of approximately 1 LD 50i.c.imp or less per wire, this indicated a titre reduction in the range of 5.0 to 5.5 logs. The findings (attack rates and survival times) of our in vivo assays obtained after chemical and thermal processing of carriers, or subsequent to chemical decontamination 350 Journal of General Virology 89

4 Decontamination of surgical instruments from prions Table 1. End-point titration of infectivity attached to steel wires coated with serially diluted 263K scrapie brain homogenate and subsequently implanted into the brain of Syrian hamsters For contamination, wires were incubated in 150 ml serially diluted homogenate. Attack rates for terminal scrapie are specified as the number of animals that developed terminal symptoms out of the total number of challenged animals. Survival times until the development of terminal scrapie are provided in days p.im. (mean±sd), and survival times in bold (.500 days p.im.) refer to hamsters that were sacrificed at the indicated time points without having developed clinical symptoms of scrapie. As determined by the Spearman Kärber method (Bonin, 1973), the end-point dilution of scrapie brain homogenate leading to 1 LD 50i.c.imp on wires was Accordingly, after coating with scrapie brain homogenate serially diluted to , wires were calculated to carry the indicated loads of infectivity. 263K scrapie brain homogenate Steel wire bioassay Dilution factor Titre (LD 50i.c. ml 1 ) Infectivity of coating preparation (LD 50i.c. ) Infectivity per wire (LD 50i.c.imp ) Attack rate for terminal scrapie Survival time (days p.im.) /6 94± /6 100± /6 113± /6 134± /6 161± /6 231±80, /6 198, / /6 503 followed by steam sterilization, and the resulting levels of residual infectivity as well as the titre reductions achieved by the tested decontamination procedures are summarized in Tables 2 and 3, respectively. As shown in the lower part of Table 2, negative-control wires did not trigger the onset of clinical disease. Distilled water. Irrigation of wires in distilled water (once for 1 min, followed by four 10 min washes in a volume of 45 ml each) at room temperature produced virtually identical survival times of 93 and 94 days p.im. in two independent experiments (Bioassay group 1, Table 2; and end-point titration, Table 1). Previously, we showed that simple processing in water resulted in considerable amounts of PrP Sc being detached from the wire surface and released into the aqueous phase (Lemmer et al., 2004). Together, these data indicate that our coated test wires were contaminated (i) with a fixed infectivity of LD 50i.c.imp, which was not removable by irrigation in water, and (ii) with additional unfixed material of unknown infectivity, resulting in a total initial load of LD 50i.c.imp. NaOH and NaOCl. Incubation in 1.0 M NaOH or 2.5 % NaOCl (containing p.p.m. available chlorine) for 1 h at 23 uc left no clinically detectable infectivity attached to the wires and produced titre reductions of 5.5 logs. Alkaline cleaner. Incubation of the wires in the commercially available alkaline cleaner at a concentration of 1 % (ph 12.2, as measured at room temperature) for 10 min at 55 uc or for 60 min at 23 uc resulted in complete removal of detectable infectivity and a titre reduction of 5.5 logs. However, a shortening of the incubation time to 5 min at 55 uc led in one of the two independent bioassay groups to clinical disease in one hamster at 386 days p.im. Here, the reduction in infectivity was.5 to 5.5 logs. Lowering the concentration of the cleaner to 0.5 % (ph 11.9) for 5 or 10 min at 55 uc caused high attack rates in the reporter animals and decreased the titre reductions down to 4 5 logs in one bioassay group. SDS. Exposure of wires to 5 % SDS (ph 7.1) for 60 min at 23 or 90 uc resulted in incomplete decontamination and did not prevent high attack rates. The titre reduction found after incubation at 90 uc and at 23 uc was 4 to,5 logs and 1 to,2 logs, respectively. SDS/NaOH. When contaminated wires were processed in a mixture of 0.2 % SDS and 0.3 % NaOH (ph 12.8) for 5 or 10 min at 23 uc, no residual infectivity attached to the carriers could be detected in the bioassay. Within our threshold of detection, the findings indicated complete decontamination with a reduction in infectivity of 5.5 logs. PAA. Following treatment of wires with 0.25 % PAA for 1 h at 23 uc or 10 min at 55 uc, large amounts of residual infectivity were found. PAA showed almost no decontaminating effect and produced a titre reduction of less than 1 log. PAA/NaOH. The disinfectant containing 0.2 % PAA, NaOH in the range M according to the specification of the manufacturer (ph 8.9) and tensidic additives left substantial amounts of agent attached to the wires after incubation times of 60, 90 and 120 min at 23 uc. The bioassay results indicated a reduction in infectivity of 2 to,3 logs

5 K. Lemmer and others Table 2. Bioassay results for reprocessed wires subjected to different decontamination procedures after coating with 263K scrapie brain homogenate For contamination, wires were incubated in 150 ml 10 % 263K scrapie brain homogenate, providing an initial infectivity load of approximately LD 50i.c.imp per wire. Although heat sterilization was carried out in most cases at 134 uc for 5 min, one group of reporter animals was implanted with wires that had been autoclaved at 121 uc for 15 min, as indicated. Attack rates and survival times are as described for Table 1. RT, Room temperature. Chemical Concn Time Temperature (min) (6C) Bioassay group 1 Bioassay group 2 + Sterilization 5 min 134 6C + Sterilization 5 min 134 6C Attack rate Survival time (days p.im.) Attack rate Survival time (days p.im.) Attack rate Survival time (days p.im.) Attack rate Survival time (days p.im.) None (rinsed in dh 2 O only) RT 8/8 93±4 3/6 347±64; 502 3/6 318±21; Sterilization 15 min 121 6C 5/5* 112±6 NaOH 1.0 M / / /6 502 NaOCl 2.5 % /5* 502 0/5* 502 0/6 501 Alkaline cleaner 1.0 % / / /5* 386; 503 0/ / /5* 502 0/ / / % /6 378±18; 0/ /6 324±47 0/5* /6 276±116; 0/5* 503 4/6 328±83; 0/ SDS 5 % /6 240± /5D 109±3 SDS/NaOH 0.2 %/ / /5* 502 0/ / % / / / /6 502 PAA 0.25 % /6 97± /6 96±1 Disinfectant with PAA/NaOH Negative controls Wires incubated in normal brain homogenate Native wires (uncoated and not reprocessed) 0.2 %/ % /5* 115±4 2/6 294±66; /6 124±5 0/ /6 122±6 3/6 339±38; 504 0/ /5* 534 6/6 132±4 1/6 477; 502 *One out of six animals challenged with an implanted wire died for reasons unrelated to scrapie; accordingly, the number of animals in the respective group was set to n55. DOne animal of this group, which developed no clinical symptoms until 512 days p.im., had no steel wire implanted into the brain as revealed by post-mortem examination. 352 Journal of General Virology 89

6 353 Table 3. Residual infectivity and titre reductions determined after decontamination of steel wires coated with 263K scrapie brain homogenate Residual wire infectivity was deduced from attack rates and survival times (Table 2) using the dose response relationship established in the end-point titration experiment (Table 1), and titre reductions were calculated by comparing the residual infectivity with the contamination load prior to decontamination. Red., reduction in titre; Res. inf., residual infectivity; RT, room temperature; UD, undetectable; ND, not done. Chemical Concentration Time (min) Temperature (6C) Res. inf./wire (LD 50i.c.imp ) Bioassay group 1 Bioassay group 2 Red. (log 10 ) + Sterilization 5 min 134 6C Res. inf./wire (LD 50i.c.imp ) Red. (log 10 ) Res. inf./wire (LD 50i.c.imp ) Red. (log 10 ) + Sterilization 5 min 134 6C Res. inf./wire (LD 50i.c.imp ) None (rinsed in RT *.0 to,3.5 to,5.5 ND ND.0 to,3.5 to,5.5 dh 2 O only) NaOH 1.0 M UD 5.5 UD 5.5 UD 5.5 ND ND NaOCl 2.5 % UD 5.5 UD 5.5 UD 5.5 ND ND Alkaline cleaner 1.0 % 5 55 UD 5.5 UD to,3.5 to 5.5 UD UD 5.5 UD 5.5 UD 5.5 UD UD 5.5 ND ND ND ND ND ND 0.5 % to 3 5 to 5.5 UD to 4to,5 UD to 3 5 to 5.5 UD to,3.5 to 5.5 UD 5.5 SDS 5 % to to,5 ND ND ND ND ND ND to 1to,2 ND ND ND ND ND ND SDS/NaOH 0.2 %/0.3 % 5 23 UD 5.5 UD 5.5 UD 5.5 UD UD 5.5 UD 5.5 UD 5.5 UD 5.5 PAA 0.25 % to to,1 ND ND ND ND ND ND Disinfectant with PAA/NaOH 0.2 %/0.075 % % Red. (log 10 ) to 0to,1 ND ND ND ND ND ND to 2to,3.0 to,3.5 to 5.5 ND ND ND ND to 2to,3 UD 5.5 ND ND ND ND to 2to,3.0 to,3.5 to to 2to,3.0 to,3.5 to *The titre reduction due to irrigation in dh 2 O could not be quantified in our experimental setup. Decontamination of surgical instruments from prions

7 K. Lemmer and others LD 50i.c.imp. Accordingly, the titre reduction was in the range of.5 to,5.5 logs, whereas steam sterilization at 121 uc for 15 min decreased the agent titre by approximately 2 logs. Bioassays of chemical decontamination followed by steam sterilization After steam sterilization of wires for 5 min at 134 uc subsequent to chemical treatments that had already achieved complete decontamination alone, titre reductions consistently remained 5.5 logs. If wires were sterilized at 134 uc for 5 min after treatment with 0.5 % alkaline cleaner (5 or 10 min at 55 uc), an additive effect with respect to inactivation of infectivity was observed: the attack rates went down to 0/5 or 0/6 and titre reduction increased to 5.5 logs (Tables 2 and 3). With the disinfectant containing PAA and NaOH as the main components, steam sterilization at 134 uc for 5 min yielded an improvement in titre reduction from 2to,3 logs (disinfectant only) to.5 to 5.5 logs (disinfectant plus steam sterilization) (Tables 2 and 3). Fig. 1. Appearance of steel wire surfaces after different treatments as revealed by SEM. (a) Native wires washed in 2 % Triton X-100 but not coated or reprocessed. (b) Wires coated with a 10 % 263K scrapie brain homogenate, air-dried and rinsed in water. (c) Test wires decontaminated with 1.0 M NaOH (60 min at 23 6C). (d) Test wires decontaminated with 0.2 % SDS/0.3 % NaOH (10 min at 23 6C). (e) Test wire treated with disinfectant containing PAA and NaOH (120 min at 23 6C). (f) Test wires incubated in 0.25 % PAA (60 min at 23 6C). Bars: (a d, f) left, 50 mm, right, 5 mm; (e) 5 mm. Bioassays of thermal decontamination of steel wires Sterilization. Steam sterilization at 134 uc for 5 min without chemical decontamination reduced the load of infectivity on steel wires from LD 50i.c.imp present after irrigation in water to a level in the range of.0 to,3 PET blot analyses Examination of brains from hamsters that remained free of scrapie symptoms during clinical observation for 500 days. Subclinical TSE infection could not be detected by PET blotting in any of the animals that survived without scrapie symptoms until termination of the end-point titration experiment. PET blot analyses of brain samples from the decontamination bioassays yielded a positive result in only one animal that did not show clinical symptoms during the observation period of more than 500 days p.im. (Fig. 2). In this case, a steel wire processed by steam sterilization at 134 uc for 5 min had been implanted into the reporter animal. Apart from this case, subclinical PrP Sc deposition could not be detected in any of the asymptomatic animals around the cerebral wire channel or in other brain regions examined by PET blotting. Time course of cerebral PrP Sc deposition after implantation of 263K-coated steel wires without previous decontamination. In order to complement the PET blot findings shown in Fig. 2 for one asymptomatic hamster infected with a minimal dose of agent after thermal wire sterilization, we examined the temporal spatial course of cerebral PrP Sc deposition in a worst-case scenario. Here, steel wires coated with the maximum infectivity load were implanted without previous decontamination. Subsequently, the deposition of PrP Sc was monitored by PET blotting at 1, 5, 15, 30, 45 and 60 days p.im., as well as at the terminal stage of scrapie, in coronal sections of tissue blocks that contained the wire channels. 354 Journal of General Virology 89

8 Decontamination of surgical instruments from prions Fig. 2. PET blot detection of PrP Sc deposition in the brain of a subclinically infected hamster. The animal remained free of scrapie symptoms until termination of the bioassay experiment at 502 days after implantation of a 263K-coated steel wire that had been subjected to steam sterilization for 5 min at 134 6C. Coronal sections encompassing the wire channel (W) were cut from rostral to caudal (a c). PrP Sc was not found directly adjacent to the wire channel, but foci of deposition (filled arrows) were detectable in different brain areas. The cluster of hole-like structures located proximal to the wire channel possibly originated from blood vessels (BV). Fig. 3 shows the findings on the course of PrP Sc deposition. Sections were selected from different coronal areas for optimal display of the conspicuous features found by PET blotting. Following wire implantation, initial tissue lesions due to the invasive procedure were visible at 1 day p.im. At this time point and at 5 days p.im., no PrP Sc, which typically presents in the PET blot as violet-stained granular deposits, could be detected in the examined brain sections. However, at 15 days p.im., tiny PrP Sc deposits became visible in the area of the wire channel in sections from two out of three examined hamsters. The locations of PrP Sc foci detected in association with the wire channel were not identical in these two animals or in hamsters examined at 30 days p.im., indicating that cerebral infection was established from different sites of contaminated wires. At 45 days p.im., the process of PrP Sc deposition already exhibited considerable spread (not shown), and at 60 days p.im., strong immunostaining for PrP Sc was seen through- Fig. 3. PET blot detection of cerebral PrP Sc accumulation after implantation of 263Kcoated steel wires without previous decontamination. Coronal sections encompassing the wire channel (W) from animals sacrificed at 1, 5, 15, 30 and 60 days p.im., as well as at the terminal stage of scrapie, are shown. PrP Sc could be detected as violet-stained granular immunoreactive material from 15 days p.im. (filled arrow) onwards. Detected PrP Sc deposition started at the wire channel and proceeded subsequently to almost all areas in the examined brain tissue specimens. W, Wire channel

9 K. Lemmer and others out many areas, particularly of the thalamus. At the terminal stage of disease, most areas in the examined brain region were positive for PrP Sc. DISCUSSION We used the 263K scrapie hamster model, a paradigm that has been applied in many TSE inactivation studies (reviewed by Taylor, 2000, 2004), in conjunction with a steel wire assay (Zobeley et al., 1999; Flechsig et al., 2001; Fichet et al., 2004) to investigate the decontamination of surgical instruments and medical devices. The study presented in this report aimed to validate our previous in vitro carrier experiments on the detachment, destabilization and degradation of PrP Sc bound to steel surfaces, in which we identified routinely usable candidate reagents for efficient surface decontamination (Lemmer et al., 2004). Similar model systems have been increasingly employed during the past few years by different research groups (Yan et al., 2004; Baxter et al., 2005; Fichet et al., 2007), a fact that facilitates direct comparison as well as integration of independently achieved findings. The high degree of interlaboratory comparability of the used in vivo carrier assay is highlighted by strikingly consistent end-point titration results obtained by us and others (Fichet et al., 2004, 2007). Reagents achieving complete removal of detectable infectivity In accordance with a comprehensive body of published data on NaOH and NaOCl (Kimberlin et al., 1983; Brown et al., 1986; Taylor et al., 1994; Taylor, 2000; Flechsig et al., 2001; Rutala & Weber, 2001), we confirmed that treatment of contaminated steel wires with 1.0 M NaOH or a NaOCl solution containing at least p.p.m. available chlorine for 1 h at room temperature resulted in no detectable infectivity on the wires and achieved titre reductions of 5.5 logs. Almost identical reductions of at least 5.6 logs were observed for these reagents in the steel wire assay by Fichet et al. (2004). When testing further reagents in the steel wire assay, we and others found that different commercially available alkaline cleaners that can be applied for routine reprocessing of surgical instruments also achieved, under appropriate conditions, complete removal of detectable infectivity and titre reductions of 5.5 logs (this study) or 5.6 logs (Fichet et al., 2004). Most interestingly, a mixture of 0.2 % SDS/0.3 % NaOH was found to be highly effective for decontamination. With this formulation, complete elimination of detectable infectivity, indicating a titre reduction of 5.5 logs, was achieved after incubation for only 5 or 10 min at 23 uc. Of note, subclinical infections after implantation of wires that had been decontaminated by the methods described above were not found in any of the asymptomatic reporter animals at the termination of the experiment at 500 days p.im. by PET blotting for cerebral PrP Sc deposition. In contrast, when steel wires coated with the maximum infectivity load were implanted without previous decontamination, PET blotting revealed PrP Sc deposition in the brain as early as 15 days p.im. (i.e. long before the development of clinical symptoms and terminal disease). Taken together, the incubation of steel wires in 1.0 M NaOH for 1 h at 23 uc, 2.5 % NaOCl for 1 h at 23 uc, an alkaline cleaner (used at a concentration of 1 %) for 10 min at 55 uc or 0.2 % SDS/0.3 % NaOH for 5 or 10 min at 23 uc led to complete decontamination in terms of detectable infectivity, as well as to efficient removal of residual brain material from the carriers. Reagents achieving incomplete removal of detectable infectivity Of the chemicals found to result in incomplete decontamination in our in vivo assay, 5 % SDS applied for 60 min at 90 uc achieved the highest titre reduction ( 4to,5logs), and a disinfectant containing PAA and NaOH resulted in a titre reduction of 2 to,3 logs. Although 5 % SDS and the disinfectant containing PAA and NaOH failed to result in complete removal of infectivity, the levels of titre reduction they achieved explain their apparently promising performance previously observed in the in vitro screening assay (Lemmer et al., 2004). PAA alone (used at a concentration of 0.25 % for 1 h at 23 uc or 10 min at 55 uc) did not show any relevant titre reduction (,1 log) under two different processing conditions. In accordance with our in vivo results, 0.25 % PAA was previously found to exert no significant detaching, destabilizing or degrading effects on PrP Sc in vitro (Lemmer et al., 2004). Similar in vivo results were reported by Yan et al. (2004) using 0.35 % PAA applied for 5 min at 23 uc, whilst Fichet et al. (2004) observed a titre reduction of 3.5 logs with 0.25 % PAA applied for 12 min at 55 uc. The exact reasons for these variations in the observed decontamination activity of PAA remain to be established but may result at least in part from slight differences in the applied procedures or reagents used. Effect of steam sterilization Steam sterilization of contaminated wires at 134 uc for 5 min (a reprocessing step for surgical instruments often used in nosocomial settings in Germany) without previous chemical decontamination yielded attack rates of 50 % in two independent bioassays and a titre reduction on the wires of.5 to,5.5 logs. Yan et al. (2004) observed an attack rate of 10 % after steam sterilization of wires at 134 uc for 18 min, whilst Fichet et al. (2004), again after steam sterilization at 134 uc for 18 min, found a 60 % attack rate with respect to the development of terminal scrapie and a titre reduction of logs. 356 Journal of General Virology 89

10 Decontamination of surgical instruments from prions Steam sterilization at 134 uc for 5 min subsequent to chemical treatments did not negatively affect decontamination efficiencies. Rather, steam sterilization consistently increased the reduction in infectivity below the limit of detection after incubation of wires in 1 % alkaline cleaner for 5 min at 55 uc or 0.5 % alkaline cleaner for 5 or 10 min at 55 uc. A similar but more pronounced cumulative effect was also obvious after steam sterilization of wires incubated in the disinfectant containing PAA and NaOH. However, it should be noted that the total titre reductions observed in this experiment were generally lower than would have been expected from the addition of individual titre reductions obtained by incubation in the disinfectant ( 2 to,3 logs) and by sterilization for 5 min at 134 uc (.5to,5.5 logs). This highlights the problem that titre reductions achieved by different decontamination procedures cannot simply be summed. Concluding remarks In the search for effective but instrument-compatible and routinely applicable procedures that allow reliable decontamination of surgical instruments from TSE agents, we carried out a steel wire assay in vivo to follow up on our previous in vitro experiments (Lemmer et al., 2004). We found that, under appropriate conditions, relatively mild reagents such as a commercially available alkaline cleaner (ph 12.2) or a mixture of 0.2 % SDS and 0.3 % NaOH (ph 12.8) could achieve complete removal of 263K scrapie infectivity and a titre reduction of 5.5 logs with or without subsequent steam sterilization for 5 min at 134 uc. With 0.2 % SDS/0.3 % NaOH, this effect occurred after treatment for only 5 or 10 min at 23 uc. The hamster-adapted scrapie strain 263K has been used as a model agent in many studies on the decontamination of TSE agents and, as outlined above, the 263K steel wire model has produced reproducible results in vitro and in vivo when applied to the testing of decontamination procedures for TSE agents. Recently, it was shown that inactivation results obtained with the steel wire assay for mouse-adapted BSE agent 6PB1 did not differ significantly from those obtained with 263K scrapie agent (Fichet et al., 2007). However, it has to be noted that wires used as a model for assessing the decontamination of surgical instruments from prions may be easier to clean than larger flat metal surfaces (Lipscomb et al., 2006) or complex medical devices. Furthermore, Peretz et al. (2006) reported that hamster-adapted Sc237 scrapie agent (which is thought to be identical to strain 263K) may be times less resistant to decontamination by acidic SDS than human strains of TSE agents. The latter finding refers to one specific and rather uncommon inactivation procedure, but together with the report by Lipscomb et al. (2006) suggests that decontamination procedures found in the 263K steel wire model should be validated for human TSE agents on different types of instrument surfaces. ACKNOWLEDGEMENTS We are grateful to Kristin Kampf, Angelika Mas Marques, Ulrich Borchers, Patrizia Reckwald and Gudrun Holland for excellent technical support. This work was supported by the German Federal Ministry for Education and Research (BMBF, Grant PTJ-BIO ) and by the European Network of Excellence NeuroPrion. REFERENCES Baier, M., Schwarz, A. & Mielke, M. (2004). Activity of an alkaline cleaner in the inactivation of the scrapie agent. J Hosp Infect 57, Baxter, H. C., Campbell, G. A., Whittaker, A. G., Jones, A. C., Aitken, A., Simpson, A. H., Casey, M., Bountiff, L., Gibbard, L. & Baxter, R. L. (2005). Elimination of transmissible spongiform encephalopathy infectivity and decontamination of surgical instruments by using radio-frequency gas-plasma treatment. J Gen Virol 86, Beekes, M., Baldauf, E. & Diringer, H. (1996). Sequential appearance and accumulation of pathognomonic markers in the central nervous system of hamsters orally infected with scrapie. J Gen Virol 77, Beekes, M., Mielke, M., Pauli, G. & Baier, M. (2004). Aspects of risk assessment and risk management of nosocomial transmission of classical and variant Creutzfeldt Jakob disease with special attention to German regulations. In Prions. A Challenge for Science, Medicine and the Public Health System. Contributions to Microbiology, vol. 11, pp Edited by H. F. Rabenau, J. Ciantl & H. W. Doerr. Basel: Karger. Bonin, O. (1973). In Quantitativ-virologische Methoden, pp Stuttgart: Georg Thieme Verlag. Brown, P., Rohwer, R. G. & Gajdusek, D. C. (1986). Newer data on the inactivation of scrapie virus or Creutzfeldt Jakob disease virus in brain tissue. J Infect Dis 153, Deleault, N. R., Harris, B. T., Rees, J. R. & Supattapone, S. (2007). Formation of native prions from minimal components in vitro. Proc Natl Acad Sci U S A 104, Diringer, H., Beekes, M., Özel, M., Simon, D., Queck, I., Cardone, F., Pocchiari, M. & Ironside, J. W. (1997). Highly infectious purified preparations of disease-specific amyloid of transmissible spongiform encephalopathies are not devoid of nucleic acids of viral size. Intervirology 40, Diringer, H., Roehmel, J. & Beekes, M. (1998). Effect of repeated oral infection of hamsters with scrapie. J Gen Virol 79, Fichet, G., Comoy, E., Duval, C., Antloga, K., Dehen, C., Charbonnier, A., McDonnell, G., Brown, P., Lasmezas, C. I. & Deslys, J. P. (2004). Novel methods for disinfection of prion-contaminated medical devices. Lancet 364, Fichet, G., Comoy, E., Dehen, C., Challier, L., Antloga, K., Deslys, J. P. & McDonnell, G. (2007). Investigations of a prion infectivity assay to evaluate methods of decontamination. J Microbiol Methods 70, Flechsig, E., Hegyi, I., Enari, M., Schwarz, P., Collinge, J. & Weissmann, C. (2001). Transmission of scrapie by steel-surfacebound prions. Mol Med 7, Kimberlin, R. H., Walker, C. A., Millson, G. C., Taylor, D. M., Robertson, P. A., Tomlinson, A. H. & Dickinson, A. G. (1983). Disinfection studies with two strains of mouse-passaged scrapie agent. Guidelines for Creutzfeldt Jakob and related agent. J Neurol Sci 59, Lemmer, K., Mielke, M., Pauli, G. & Beekes, M. (2004). Decontamination of surgical instruments from prion proteins: in 357

11 K. Lemmer and others vitro studies on the detachment, destabilization and degradation of PrP Sc bound to steel surfaces. J Gen Virol 85, Lipscomb, I. P., Pinchin, H. E., Collin, R., Harris, K. & Keevil, C. W. (2006). Are surgical stainless steel wires used for intracranial implantation of PrP Sc a good model of iatrogenic transmission from contaminated surgical stainless steel instruments after cleaning? J Hosp Infect 64, Peretz, D., Supattapone, S., Giles, K., Vergara, J., Freyman, Y., Lessard, P., Safar, J. G., Glidden, D. V., McCulloch, C. & other authors (2006). Inactivation of prions by acidic sodium dodecyl sulfate. J Virol 80, Prusiner, S. B. (1982). Novel proteinaceous infectious particles cause scrapie. Science 216, Prusiner, S. B. (1998). Prions. Proc Natl Acad Sci U S A 95, Rutala, W. A. & Weber, D. J. (2001). Creutzfeldt Jakob disease: recommendations for disinfection and sterilization. Clin Infect Dis 32, Schulz-Schaeffer, W. J., Tschoke, S., Kranefuss, N., Drose, W., Hause-Reitner, D., Giese, A., Groschup, M. H. & Kretzschmar, H. A. (2000). The paraffin-embedded tissue blot detects PrP Sc early in the incubation time in prion diseases. Am J Pathol 156, Taylor, D. M. (2000). Inactivation of transmissible degenerative encephalopathy agents: a review. Vet J 159, Taylor, D. M. (2004). Resistance of transmissible spongiform encephalopathy agents to decontamination. In Prions. A Challenge for Science, Medicine and the Public Health System. Contributions to Microbiology, vol. 11, pp Edited by H. F. Rabenau, J. Ciantl & H. W. Doerr. Basel: Karger. Taylor, D. M., Fraser, H., McConnell, I., Brown, D. A., Brown, K. L., Lamza, K. A. & Smith, G. R. (1994). Decontamination studies with the agents of bovine spongiform encephalopathy and scrapie. Arch Virol 139, Thomzig, A., Spassov, S., Friedrich, M., Naumann, D. & Beekes, M. (2004). Discriminating scrapie and bovine spongiform encephalopathy isolates by infrared spectroscopy of pathological prion protein. J Biol Chem 279, Weber, P., Giese, A., Piening, N., Mitteregger, G., Thomzig, A., Beekes, M. & Kretzschmar, H. A. (2007). Generation of genuine prion infectivity by serial PMCA. Vet Microbiol 123, Yan, Z. X., Stitz, L., Heeg, P., Pfaff, E. & Roth, K. (2004). Infectivity of prion protein bound to stainless steel wires: a model for testing decontamination procedures for transmissible spongiform encephalopathies. Infect Control Hosp Epidemiol 25, Zobeley, E., Flechsig, E., Cozzio, A., Enari, M. & Weissmann, C. (1999). Infectivity of scrapie prions bound to a stainless steel surface. Mol Med 5, Journal of General Virology 89