Detection of PUUV and identification of the virus origin

The colony of bank voles was tested for the presence of PUUV-reactive antibodies using an enzyme-linked immunosorbent assay (ELISA) in 2008, and the result was negative. However, in December 2012, serological analyses using two assays (an in-house IgG ELISA and a commercial immunochromatographic Rapid test (ReaScan Ab-Dect Puumala IgG test, Reagena, Toivala, Finland; [42]) performed on 17 old individuals (aged 203–265 days) from generation 14 of the selection experiment showed that all of them were positive for PUUV-reactive IgG antibodies (S1 Table). In contrast, five bank voles from a different colony housed in separate rooms within the same facility tested negative for the antibodies. Subsequent RT-PCR targeting the small (S) segment, using RNA derived from lung tissue, confirmed infection in all 17 seropositive individuals (S1 Table). The pairwise nucleotide sequence similarities ranged between 99.1% and 100% (S2 Table). Similar levels of pairwise nucleotide sequence similarity were observed for the corresponding medium (M) and large (L) segment sequences (S2 Table).

Between August 2009 and March 2011, wild-trapped bank voles from nine populations, along with their offspring, were housed in the same facility for a separate project (see Materials and Methods). They were therefore considered a potential source of PUUV in the main colony. To identify the origin of the PUUV incursion, a phylogenetic analysis of the colony-derived PUUV sequences together with reference sequences of PUUV strains from various clades was performed. The results showed that the colony-derived sequences clustered with those of viruses detected in bank voles captured in Teleśnica Oszwarowa in southern Poland, one of the nine populations from which the live-trapped voles originated [43,44]. These sequences were clearly separated from those from Mikołajki in northern Poland [43,44] and from other sequences of the Russian clade (S1A–C Fig). Pairwise comparisons of partial S, M, and L segment sequences showed high similarity to sequences from Teleśnica Oszwarowa (S segment 98.0–99.8%; M segment 98.5–99.8%; L segment 99.2–99.7%), but much lower similarity to sequences from Mikołajki (S segment 82.8–83.5%; M segment 80.4–80.5%; L segment 86.6–86.8%) at the nucleic acid level (S2 Table). These results indicated that wild voles captured in Teleśnica Oszwarowa and temporarily held in the same facility were the source of the virus (see Materials and Methods).

Serological tests performed on humans after the virus had been detected in voles showed that all individuals who worked directly with the infected bank voles, as well as technicians who had changed the air filters in the ventilation system, had anti-PUUV IgG antibodies and thus had developed immunity against PUUV. Importantly, however, all five tested bank voles representing another colony maintained in the same facility were not infected, even though they were housed in rooms not separated by a sanitary barrier from those of the infected colony (S1 Table). Similarly, people who worked only with voles from the other colony, or worked in the facility without direct contact with the voles, as well as people who worked in the same building but outside the facility, were not positive for PUUV-specific antibodies. These results implied that the temporary maintenance of the colony did not pose an immediate risk to people who had already worked with the infected colony and had developed immunity [45], or to people working in the same building without contact with infected voles or contaminated dust. This was the basis for the decision of the sanitary authorities not to exterminate the colony immediately, but granting time for the attempted eradication of the virus (initially 6 months, later extended to 10 months).

Outline of the eradication program

The plan for PUUV eradication (Fig 1) was founded on two previously mentioned reports: (a) outside the host organism, PUUV loses its infectivity at room temperature within a maximum of about 15 days [25], whereas (b) offspring of infected mothers are protected by MatAb, presumably at least until 60 days of age [28]. The voles in our colony are weaned at the age of 17 days. Thus, virus transferred from the maternal cage on the bodies or fur of juveniles is likely to remain infectious only until the voles are 32 days old (17 days up to weaning plus 15 days post-weaning). This gives, optimistically, a “safety time window” of 28 days (60 minus 32 days), or longer, before protection by MatAb becomes ineffective. Therefore, theoretically, if all voles from the parental generation (genP) are reproduced at the same time and euthanized immediately after weaning their offspring on day 17, the virus should not be effectively transmitted to the offspring generation (genO).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Timeline of Puumala virus eradication.

The timing and range of the “safety window” set by the end of genP-produced PUUV infectivity and the end of post-weaning protection by MatAb (green text) was estimated based on [25] and [28]. Abbreviations: genP: parental generation, genO: offspring generation, genG: grand-offspring generation, IVC: individually ventilated cages, IFA: indirect fluorescent antibody assay, MatAb: maternal antibodies, RT-qPCR: quantitative reverse transcription-polymerase chain reaction, PUUV: Puumala hantavirus.

https://doi.org/10.1371/journal.ppat.1013693.g001

This apparently simple plan, however, was based on assumptions that could have been overly optimistic. Moreover, it posed several technical problems, particularly challenging when one considers the scale and conditions under which the work was performed. First, in order to preserve genetic variation in the 16 lines, we aimed to eradicate the virus from the offspring of at least 240 families, i.e., more than 1000 individuals. Second, all the work had to be performed in a facility that did not guarantee full protection against virus transfer between cages. These issues will be addressed in the Discussion.

The major problem was that the plan had to assume that some offspring would not be fully protected by MatAb and would become infected. Therefore, the offspring had to be kept in individually ventilated cages (IVC), ideally of a type warranting full protection against virus transmission, which was far beyond the available financial resources. Instead, the weaned voles were placed in AERO Mouse Green Line IVC (Tecniplast, Bugugiatte, Italy), which, although equipped with HEPA filters, were not designed to warrant full protection against virus transmission. Next, the animals and their environment had to be frequently tested to detect and eliminate infected individuals before they could infect others. However, negative test results suggesting the absence of virus-specific antibodies or RNA in samples that could be collected intra vitam (faeces, blood) do not provide firm evidence of virus absence. Definitive evidence can be obtained only by post mortem analyses. Therefore, to the simple “one-step” plan outlined above, we added another safety step. After confirming that none of the adult genO voles were seropositive, they were mated to produce another generation (grand-offspring: genG). After weaning genG, all genO individuals were euthanized and tested for the presence of PUUV RNA in lung samples. Finally, tests for the presence of PUUV were also performed on the next two generations.

The eradication programme implementation

Procedures on parental generation (genP).

The eradication programme started on 1 February 2013 (day 1; Fig 1). Preliminary tests for the presence of PUUV-reactive IgG antibodies (ReaScan Ab-Dect test), performed on 45 females at the onset of the programme, indicated that none of the 12 individuals aged 94–117 days was seropositive, 8 of 12 individuals aged 121–132 days were seropositive, and all individuals older than 132 days were seropositive (Fig 2, S3 Table). The results indicated that MatAb can provide protection for at least as long as reported in previous studies. However, to ensure that the potential mothers can transmit MatAb to their offspring, mating should be better postponed until all potential mothers are seropositive. It is also worth noting that, above the age of 132 days, the level of antibodies was not significantly correlated with age (Pearson r = -0.08, p = 0.73, n = 21), and that among the oldest individuals both very high and low values were observed (just above the “positive” threshold of 15; Fig 2).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Relationship between anti-PUUV IgG antibody level measured as a score in the ReaScan Ab-Dect test and age of the voles.

The results were obtained for 45 females from the parental generation (genP) tested to assess the age at which most animals become seropositive. The threshold values are: negative < 5, ambiguous (not observed), positive > 15.

https://doi.org/10.1371/journal.ppat.1013693.g002

To ensure that all females were infected, used bedding was mixed between cages (4, 3, and 2 weeks before the planned mating). Another series of preliminary tests (indirect fluorescent antibody assay, IFA), performed on days 19–20, showed that 12 of 28 individuals aged 108–119 days were seropositive, and only 9 of 25 individuals aged 121–126 days were positive. Thus, the results of both preliminary tests indicated that, to ensure that most of the potential mothers had developed anti-PUUV IgG antibodies, the females should be mated at an age of at least 130 days. Ideally, mating should be postponed even further, but the health and safety authorities only permitted us to keep the infected animals for a strictly limited time. Furthermore, all females giving birth must be tested for the presence of anti-PUUV antibodies before their offspring are included in the next steps of the programme.

On day 28 (Fig 1), 620 pairs were mated (35–40 per each of the 16 lines), and all remaining voles were euthanized. To increase the chance that only anti-PUUV IgG antibody-positive females were mated, and also to increase reproductive success, older individuals were chosen wherever possible (130–181 days old), but to ensure adequate representation in each of the lines, 25 females aged 117–128 days were also included. Wherever possible, the regular selection criteria used in the selection experiment were also considered [5]. Males were euthanized 12–13 days after mating (days 40–41). From day 17 after mating, females were checked daily for pregnancy and birth, and non-pregnant females were euthanized. The first births occurred 18 days post mating, and within 11 days 473 females gave birth (mostly in the first 6 days), while the remaining females were euthanized. On days 9–12 after parturition (days 54–63, Fig 1), cages were changed and IFA tests were performed on all 436 females that had reproduced and still had live offspring (S3 Table). Of the 18 females that had been mated at 117–127 days of age, only one tested negative for anti-PUUV IgG antibodies. Among the remaining 418 females that had been mated at 130 days or older, only three tested negative (mated at 131, 138, and 150 days of age). Further, we retained only mothers that (a) were seropositive, (b) gave birth up to 26 days post mating (to narrow the age span in the cohort of offspring to 9 days), and (c) had litters of 3 or more young. From those that met these criteria, we chose 280 (17–18 per line; 9 were mated below 130 days of age), whose offspring were weaned to IVC at the age of 17 days (days 63–71, Fig 1). Immediately after weaning, all genP voles and excess offspring were euthanized.

With the progressive reduction of the colony, all animal rooms and common work areas in the facility were carefully disinfected, including the ventilation system. From day 71 of the eradication programme, all potentially infectious material was confined to the IVC system, placed in a previously disinfected room equipped with an entirely independent ventilation system. All further work with the animals was also performed in this room. Tests for the presence of PUUV RNA in dust samples taken from various locations in the facility were performed repeatedly at a later date and, with a few exceptions, gave negative results. Results of these tests are detailed in the last section of the Results.

Identification of the PUUV origin in the bank vole colony

Multiple molecular investigations have indicated the evolution of PUUV within local bank vole reservoir populations, which is reflected in the close phylogenetic relationship of PUUV sequences from bank voles from the same site [46,47]. The current distribution of PUUV lineages is mainly shaped by the postglacial immigration of bank vole phylogroups [48]. Thus, central and western Europe were invaded by bank voles of the western phylogroup, distributing the central European (CE) PUUV lineage [49]. In contrast, in eastern and northern Europe the situation is much more complex. Different PUUV lineages can be harboured by different phylogroups of the bank vole. So far, in northern Poland exclusively the Latvian (LAT) PUUV lineage has been detected in bank voles of the Carpathian and Eastern phylogroups, whereas in southern Poland PUUV of the Russian (RUS) lineage has been demonstrated to be harboured by bank voles of the Carpathian and Eastern phylogroups [43,44].

The PUUV sequences of the 17 bank voles from generation 14 of the selection experiment (i.e., the grandparent generation in terms of the eradication programme) were almost identical, which suggests a single incursion event. To identify the potential origin of the PUUV incursion, these colony-derived sequences were compared to reference sequences of PUUV strains from various clades, indicating a high nucleotide sequence similarity and phylogenetic clustering with sequences from Teleśnica Oszwarowa, southern Poland. Therefore, the infection probably originated from the group of voles that were trapped in Teleśnica Oszwarowa and temporarily maintained (2009–2011) in the same facility as the selection experiment colony. Based on the eradication experiment described here, future analyses will allow determination of the mutation rate of PUUV in bank voles and comparison with the situation in cell culture [46].

The eradication scheme and its pitfalls

The simple plan for PUUV eradication outlined in the second section of the Results was based on two assumptions supported by fairly convincing empirical data: (a) PUUV loses its infectivity outside the host at room temperature within about 15 days [25], and (b) the offspring of infected mothers are protected by MatAb until the age of 60 days or longer [28], and on a simple calculation leading to the conclusion that if young voles are weaned and separated from any infected mother at the age of 17 days, the virus transmitted on their bodies will become non-infectious approximately 4 weeks before protection by MatAb ceases. Nevertheless, the plan was vulnerable to many potential problems, which should be considered by other researchers who may need to adopt this scheme in their work:

1. 1) The estimates of both the MatAb protection period and the PUUV infectivity duration in the environment were based on small or only moderately sized samples. Thus, neither the minimum of the former nor the maximum of the latter could be determined with high reliability. If the actual extremes deviated from the reported values by just 10 days, the optimistic 4-week safety window would become very narrow.
2. 2) The safety window considered for the entire colony is in practice much shorter because, even if all parents are mated at the same time, parturitions are spread over several days, which again narrows the safety window (in our case to 20 days).
3. 3) The plan requires that all potential mothers are infected well before mating and have developed a sufficiently high level of PUUV-specific antibodies to provide their young with effective protection by MatAb. To help achieve this, we used a paradoxical approach of intentionally infecting all animals before implementing the proper eradication procedure. However, the available quantitative antibody-detection tests require a large amount of blood and practically cannot be performed without euthanizing the animals. On the other hand, the available rapid immunoassays require only a drop of blood but provide only qualitative or semi-quantitative results. To our knowledge, no studies have been performed to determine whether and to what extent the efficiency of MatAb-mediated protection in newborns and juveniles depends on the level of anti-PUUV antibodies in the mothers. Thus, a mother “positive” according to such a qualitative or semi-quantitative test might not provide effective protection to the offspring.
4. 4) Moreover, we had to choose between three conflicting requirements: (a) maintaining a balanced representation of all 16 vole lines; (b) mating females at an age at which they were highly likely to have already developed antibodies; and (c) the understandable requirement of the health inspection to begin and complete the eradication process as quickly as possible. Therefore, in the case of lines with a small number of older females, we also mated younger females (aged 117–127 days) and retained the offspring of nine of these mothers. Although the IFA test showed the presence of anti-PUUV IgG antibodies in all of these mothers around 10 days after parturition, two of the younger mothers were the only ones that had inadequately protected their offspring against PUUV infection.
5. 5) Importantly, even a single incompletely protected juvenile would cause re-occurrence of the infection in the next generation unless the individual is isolated and eliminated before transmitting the virus. However, the virus starts spreading before PUUV-specific antibodies in blood can be detected [24]. Moreover, serological assays do not allow distinction between a young individual still carrying MatAb and one that has started to produce its own antibodies as a result of infection. Tests for the presence of viral RNA in saliva or excreta may provide positive evidence of infection but do not provide negative evidence (that the animal is not infected), because virus shedding is discontinuous [24]. Samples from the cage environment could be a more effective measure, but if applied soon after weaning they would result in false positives, because some amount of virus or viral RNA may be carried from maternal cages on the juveniles’ bodies. To summarize, no intra vitam method allows reliable detection of infection before the virus starts spreading, and in practice not even at the time when spreading already occurs.
6. 6) Thus, the eradication plan must assume that some young individuals may be infected. Therefore, all young animals must be kept in isolation, which requires the use of IVC and an appropriate cage-changing station to prevent virus transmission during animal maintenance procedures and blood testing. Such equipment, at the level of protection designed for preventing viral transmission (grade III), and with sufficient capacity to maintain several hundred individuals, was far beyond our financial reach, and the same may be true for most other researchers. We could afford to purchase only a grade II + IVC system (“Aero Green line”, Tecniplast, Italy), which, although equipped with HEPA filters, is not guaranteed to prevent viral transmission. In our case, the cage system and work scheme that we implemented proved to be effective: the few infected animals found in two cages did not spread the virus to other voles. However, this should not be regarded as strong evidence that such a system provides adequate protection in general.
7. 7) We initially assumed that the absence of PUUV-specific antibodies after a 55-day quarantine (a maximum of 15 days of PUUV infectivity outside the host plus a maximum of 40 days for developing own antibodies after infection) could be treated as evidence of successful eradication. However, the case of a single vole that did not show antibodies in the test conducted 82 days after its last contact with an infected vole or infectious material, yet was detected as seropositive on day 92 (at 131 days of age), changed this perspective. Regardless of the mechanism of delayed seroconversion, which remains unknown, or a false negative result in the initial testing, we concluded that no period could be considered long enough to provide definitive negative evidence. Therefore, the one-step eradication plan has no predefined moment at which success or failure can be declared with certainty.
8. 8) Negative evidence can be provided only by post mortem tests for the presence of viral RNA in lung tissue (or perhaps other organs), and by tests for the presence of viral RNA in the environment, which have proved to be extremely sensitive and allowed detection in dust samples for at least 3 months after infectious virus material was removed. Therefore, the eradication programme must a priori include the second step: producing, within the IVC system, a next generation of offspring and performing RNA analyses in the lung samples of all parents (or in another organ in which a particular virus resides).
9. 9) This raises the question of whether it would be better to schedule the second reproduction as early as possible, i.e., just after the animals reach sexual maturity. In the case of bank voles, this would be at around 60 days of age, which would considerably shorten the whole procedure. However, we would not recommend this because of the increased risk that some individuals are infected even though not yet detected as anti-PUUV-positive by serological testing, and hence an increased risk of transmitting the virus to the grand-offspring generation.
10. 10) This was not a planned experiment, but rather a “rescue operation” carried out on a tight budget and with a “fast-track design”, i.e., involving instant decisions about subsequent steps based on current information. Consequently, not all procedures were well optimised and some decisions were imprudent. Therefore, if we could, we would introduce a few changes to the procedure:

First, we would not have mated parental generation females that had not tested positive for the presence of anti-PUUV IgG antibodies before mating. Including in the programme the few younger females that did not have antibodies 10 days before mating was perhaps the only major fault. If we had not done so, we probably would not have had to cope with infected individuals in the offspring generation. Thus, we would have argued to the health and sanitary authorities for a further delay in starting the eradication programme. If permission had not been granted, we would have sacrificed the balanced representation of all vole lines. Also, if we could afford that, it would have been better to test all females before mating and during lactation with the more reliable semi-quantitative test (ReaScan Ab-Dect), rather than the much less costly qualitative IFA test. The variation in antibody levels was presumably very high (Fig 2), and such tests would have allowed selection, for further steps, only of offspring from females with high antibody levels. Similarly, all weaned young could have been individually tested with the semi-quantitative test for the level of MatAb, and only those with high levels retained. Possibly, in this way we would have avoided having incompletely protected young in the offspring generation, and the entire procedure would have been shortened.

Second, we would not attempt to detect viral RNA in the saliva of young animals 12 days after weaning. This is because contamination of samples with virus transmitted at weaning, which is still present in the environment, leads to “false alarms” and unnecessary euthanizing of animals. In fact, it is not worth investing in the tedious attempts to sample saliva, which inevitably also increases the risk of transmitting the virus between cages. On the other hand, sampling dust from cages is easy, provides integrated information about virus shedding over several days, and such tests have proven to be highly sensitive. Again, however, such tests should only be performed after a longer period since weaning and after at least two rounds of cage changes.

Third, after detecting the first few infected individuals about 50 days after weaning, we eliminated only their cagemates, but not their sibling or the cagemates of those siblings, which were not seropositive. We did this because, at that time, we did not know how many of the young animals were infected. After losing many animals due to a “false alarm”, we were reluctant to eliminate more without evidence of infection. Instead, the suspicious animals were tested more frequently. However, from the perspective of later experience—detecting the infected sibling more than 40 days later—we see that this decision was wrong, and that all those suspicious animals should have been eliminated as well.

Ethical statement

Animal welfare was monitored daily throughout the experiment. The procedures on animals performed as a part of the maintenance of the experimental colony were approved by the 1st Local Institutional Animal Care and Use Committee in Kraków (decision 68/2012), in accordance with the EU directive 2010/63/EU. According the regulations effective in 2013, the additional procedure of blood sampling performed exclusively for the purpose of the PUUV detection, rather than as a part of an experiment performed for research purposes, did not require a separate ethical permit.

Establishing the colony and the selection experiment

The colony of bank voles (Myodes glareolus syn. Clethrionomys glareolus) was established from wild voles captured in Niepołomice Forest near Kraków (southern Poland) in 2000 and 2001 [50,51]. In 2004, a multidirectional artificial selection experiment was initiated, with 4 lines selected for high swim-induced aerobic metabolism, 4 for herbivorous capability, 4 for predatory propensity, and 4 unselected control lines [5,6]. To preserve genetic polymorphism, in each of the 16 lines the offspring were reared from 16–20 families in each generation (two per year). The selection was effective in each of the three directions, and the unique experimental evolution model system has become a valuable resource for research in several areas, such as behaviour [52,53], metabolism and thermoregulation [6,54,55], molecular genetics [56,57], exercise physiology [58,59], neurophysiology [60,61], ageing [62–64], ecotoxicology [65], stress response [66,67], holobiont evolution [41,68] or evolution of dietary niche [69].

Details of the breeding scheme and maintenance conditions are described in our earlier work [6,52]. Briefly, the colony has been maintained in a “conventional” facility, corresponding to Biosafety Level 1, appropriate for the intended research objectives, which assumed that the animals should not be maintained under germ-free conditions. Depending on the phase of breeding, the colony size ranges from about 1500–4000 individuals. The animals are kept in standard plastic mouse cages (1290D or 1264C, Tecniplast, Bugugiatte, Italy) with sawdust bedding, located in six climatic chambers providing fixed light conditions (16L:8D), a constant temperature (20°C), and appropriate ventilation. Food (Labofed H, Morawski, Kcynia, Poland) and water are available ad libitum.

Other voles in the facility

In the same facility, another permanent but small bank vole colony is maintained, established in the 1970s, which forms the basis of a separate research programme by our colleagues [70]. The animals are maintained in a partly separate space, which includes animal rooms and laboratory space. However, this space is not separated from the area where the voles from the selection experiment are kept by a sanitary barrier: it is connected to the common corridor by a standard door and uses the same washing and food storage rooms. Voles from this colony have not been infected by PUUV.

Between August and November 2009, a separate temporary colony of voles was established based on animals captured from nine wild populations (between 33 and 74 adults and subadults per population). This colony was used to study variation in physiological performance traits [71,72]. The animals were kept in a separate room in the same facility as the main colony. The maintenance conditions in this temporary colony were the same as in the main colony. However, the working regime was separate (people did not work in both colonies on the same day, and separate sets of cages, bottles, and other equipment were used). Again, however, the room was not separated by a strict sanitary barrier. Then, two generations of offspring of the wild animals were produced, and finally the entire colony was closed in March 2011. The wild-derived animals were present in the facility from the time when animals of generation 8 of the selection experiment were bred until the time when animals of generation 11 were bred.

Anaesthesia, euthanasia, and sample collection

For the purpose of the initial investigation, 22 voles (17 from the focal colony and 5 from the other colony maintained in the same facility) were anaesthetized with isoflurane (Aerane, Baxter, USA), and blood was collected by cardiac puncture using a heparinized syringe. Voles were then euthanized by administering increasing doses of isoflurane until cardiac arrest, dissected, and the heart and other internal organs were removed (which ensured death). The blood was centrifuged at 4000 rcf for 10 minutes and the plasma was separated. Plasma and samples of internal organs were stored at −80°C and carcasses at −20°C, and then transferred for further analyses performed later at the Friedrich-Loeffler-Institut.

For the purpose of the massive and repeated testing for the presence of PUUV-specific antibodies during the eradication procedure (see next section), 5–10 µl blood samples were taken from the femoral artery of live animals. Animals from the offspring generation (genO) that were detected or suspected to be infected with PUUV based on either the presence of anti-PUUV IgG antibodies or PUUV RNA in saliva, as well as their cagemates and siblings, were immediately euthanized by cervical dislocation. The euthanasia of large numbers of animals planned at the termination of subsequent generations was performed either by cervical dislocation or CO₂ inhalation.

Testing for PUUV presence during the eradication programme

Identification of the putative PUUV origin

Sanitary precautions

All work associated with standard animal maintenance, as well as collection of samples for PUUV testing (blood and tissue samples from voles, and dust samples from cages and the facility environment), was carried out by personnel who had been confirmed to possess anti-PUUV IgG antibodies and were therefore protected against reinfection with PUUV. Nevertheless, personnel wore anti-dust or medical masks, disposable gloves (usually latex gloves over veterinary gloves), disposable non-woven aprons with elasticated cuffs worn over medical aprons, and disposable non-woven shoe covers and caps. Personnel cleaning cages also wore waterproof aprons. After work, all personnel changed their clothing and disinfected their hands with Manusan (an antiviral, antibacterial, and antifungal liquid; Laboratoria Polfa Łódź, Warsaw). All work was performed under controlled conditions, minimising virus transmission. Manipulations requiring opening cages (i.e., changing cages, providing food, taking samples, removing used bedding) were performed using a cage-changing station and bedding disposal station with laminar airflow and HEPA filters (models CS5 and DS36, Tecniplast, Bugugiatte, Italy). All tools, equipment, working surfaces, chambers, corridors, and facility rooms (storage, cleaning, etc.) were continually disinfected with an antiviral agent (Virkon; LANXESS, Sudbury). The entire ventilation system in emptied rooms was disinfected soon after moving juveniles into the IVC system.