INTRODUCTION
Highly pathogenic avian influenza (HPAI) A/H5 viruses of the A/goose/Guangdong/1/1996 (GsGd) lineage first emerged in Hong Kong in 1997 (
1). Since 2003, these viruses have spread to Eurasia and Africa, and the GsGd lineage is the only lineage of HPAI viruses that are endemic in poultry. As a result of continuous circulation in poultry, the hemagglutinin (HA), the main antigenic determinant of influenza A viruses, of A/H5 GsGd viruses diversified in several genetic and antigenic clades (
2). Moreover, due to the segmented nature of the influenza A virus genome, viruses carrying GsGd HAs reassorted with cocirculating low-pathogenic avian influenza (LPAI) viruses from domestic and wild birds, increasing the genetic diversity of A/H5 GsGd viruses. In particular, the H5 HA of clade 2.3.4.4 was found to reassort frequently with neuraminidase (NA) genes of other subtypes than N1, yielding A/H5 viruses with various HA-NA combinations: A/H5N1, A/H5N2, A/H5N3, A/H5N5, A/H5N6, and A/H5N8 (
3).
In contrast to A/H5N8 viruses, which spread worldwide in a global fashion via wild migratory birds (
4), the distribution of A/H5N6 viruses has until recently been limited to China, Laos, and Vietnam. In China, they have replaced A/H5N1 viruses and became the dominant A/H5 GsGd circulating virus in poultry (
5). However, A/H5N6 viruses genetically related to the viruses from China were isolated during the 2016-2017 winter from wild migratory birds in Korea and Japan, where they also caused poultry outbreaks (
6–8), suggesting a potential for A/H5N6 viruses to spread via wild bird migration also.
A/H5N6 viruses are the only clade 2.3.4.4 viruses that have crossed the species barrier and infected humans. The first human cases of infection with the two circulating lineages of A/H5N6 viruses were both reported in China (
9,
10). As of 8 December 2017, a total of 17 patients infected with A/H5N6 have been reported, and 6 of these patients died (
11). Almost all patients infected with A/H5N6 viruses had documented exposure to infected poultry at live poultry markets, suggesting poultry exposure as a potential source of zoonotic transmission (
12). The first reassortant A/H5N6 viruses identified in April 2013 and the viruses isolated from the first three human cases in 2014 were carrying the internal genes of A/H5N1 viruses. Based on the origin of their HA and NA genes, these viruses belong to two genotypes (G1 and G2). After 2014, multiple other genotypes emerged upon reassortment with circulating A/H9N2 or A/H7N9 viruses (
5,
12,
13), resulting in the definition of 34 genotypes (
5).
The global spread via wild birds and complex reassortment history of clade 2.3.4.4 viruses required a detailed molecular and phenotypic characterization, especially of the newly emerged A/H5N6 viruses. Several research groups previously reported on the low pathogenicity and lack of airborne transmissibility in the ferret model of the Korean, European, and North American A/H5N8 viruses and North American A/H5N2 viruses, probably partially due to the presence of internal genes originating from wild bird LPAI viruses (
14–16). In contrast, A/H5N6 viruses carry internal genes coming from A/H5N1, A/H7N9, and A/H9N2 viruses, which are poultry-adapted viruses that have themselves caused zoonotic infections or acted as donors of internal genes to most other zoonotic influenza viruses (
5).
Here, we focused our investigation on A/H5N6 A/Guangzhou/39715/2014 (GZ/14) virus, which belongs to genotype G1 and caused one of the first human cases of A/H5N6 infection in 2014 (
17).
In vitro characterization of phenotypes that have been associated with airborne (respiratory droplets and aerosols) transmission of A/H5N1 viruses, such as receptor binding preference, thermostability, pH stability, and polymerase activity was performed. The potential of A/H5N6 GZ/14 to transmit via the airborne route between ferrets was assessed, as well as its pathogenicity in the ferret model. While our study suggests that the public health risk posed by this A/H5N6 virus is low, the A/H5N6 GZ/14 showed a high polymerase activity mediated by the E627K substitution in PB2, replicated to higher titers in the respiratory tracts of ferrets and was more pathogenic than a clade 2.3.4.4 A/H5N8 virus (
14).
DISCUSSION
Here, we report on a detailed
in vitro and
in vivo characterization of the clade 2.3.4.4 A/H5N6 GZ/14 virus in order to evaluate the risk for humans. Sequence analysis of A/H5N6 GZ/14 virus revealed a few amino acid substitutions that have previously been described as mammalian adaptation markers for other A/H5 viruses of the GsGd lineage: 94N (
25), 133A (
25), and 235P (
26) in HA, which have been associated with increased binding of A/H5N1 viruses to human-type receptors, and E627K in PB2, which has been associated with increased replication of influenza viruses
in vitro and
in vivo at temperatures equivalent to those of the mammalian upper respiratory tract (URT) (
21,
22,
27). Despite the fact that our knowledge of the genetic markers of mammalian adaptation is growing, performing risk assessment using genetic data alone might be misleading because of context dependency. Therefore, we sought to perform a phenotypic characterization of A/H5N6 GZ/14 by focusing on the phenotypes that have been associated with airborne transmission of A/H5N1 GsGd viruses (
28,
29): a change in receptor binding preference of HA from avian-type to human-type receptors, α2,3-SA and α2,6-SA, respectively, increased HA thermostability and acid stability and increased polymerase activity.
Using a resialylated TRBC assay, A/H5N6 GZ/14, as well as A/H5N8 ck/NL/14, was shown to display a typical pattern of avian influenza viruses by binding exclusively to α2,3-SA. These data are in accordance with previous studies, in which A/H5N6 GZ/14 and another human A/H5N6 isolate, A/Sichuan/26221/2014, were found to bind only to α2,3-SA (
30,
31). Interestingly, A/H5N6 viruses isolated from healthy ducks in China bound in a similar fashion to both α2,3-SA and α2,6-SA as well as to both upper (trachea) and lower (alveoli) respiratory epithelium (
32). A/H5N6 viruses isolated from wild bird carcasses in Hong Kong also displayed a dual binding phenotype (
31). The substitutions that mediated the binding of these avian A/H5N6 viruses to human-type receptors were not identified. Although most clade 2.3.4.4 viruses have retained α2,3-SA specificity, it was shown that the emergence of clade 2.3.4.4 viruses was accompanied with the capacity of these viruses to bind to fucosylated sialilosides, mediated by K218Q and S223R, substitutions that are conserved in clade 2.3.4.4 HAs (
33). Of note, such modifications cannot be detected using the resialylated TRBC, in which only the nature of the bondage between the terminal sialic acid and the antepenultimate sugar is evaluated.
HA acid stability has been described as an important host range factor, associated with adaptation of avian and swine influenza viruses to the human host, airborne transmission of avian viruses, and pandemic potential (
28,
29,
34–37). The HAs of A/H5N6 GZ/14 and of A/H5N8 ck/NL/14 were unstable compared to those of a human A/H3N2 virus and an avian-origin A/H5N1 transmissible virus assessed in syncytium formation and thermostability assays. A/H5N6 GZ/14 and A/H5N8 ck/NL/14 do not possess known substitutions that have been reported to increase the acid stability and/or thermostability of A/H5 viruses from other clades of the GsGd lineage, such as H103Y, T315I, or K58I (HA2 numbering) (
28,
29,
38,
39).
Using a minireplicon assay, we demonstrated that A/H5N6 GZ/14 possessed a high polymerase activity, which was mediated by E627K in PB2. Out of the 16 A/H5N6 human isolates, seven isolates carry the PB2-E627K, and one isolate, A/Sichuan/26221/2014, carries the PB2-D701N substitution, which has also been associated with increased replication and transmission in mammalian hosts (
20,
40,
41). Acquisition of PB2-E627K might be the result of adaptation to the human host, although this substitution has also been detected in A/H5 avian virus isolates from other clades, as well as in other avian-origin influenza viruses such as A/H7N9 (
42).
Of the three virus traits that conferred airborne transmissibility to A/H5N1 viruses between ferrets, A/H5N6 GZ/14 possesses only one of these traits, which is high polymerase activity. On the basis of the results of these
in vitro assays, we predicted that A/H5N6 GZ/14 would not be transmissible via the airborne route, which was confirmed using the ferret model. Ferrets are used as animal models to study airborne transmission of influenza viruses, as it was shown that human influenza viruses transmit via the airborne route between ferrets, while avian influenza viruses do not, corresponding to what is also observed in humans (
28). Our results are in accordance with a study on the transmissibility of clade 2.3.4.4 A/H5N6 viruses, isolated from healthy ducks, that were transmitted only via the direct contact route but not via the airborne route between ferrets (
32). Other viruses of clade 2.3.4.4, A/H5N8 or A/H5N2, were also found to not transmit via the airborne route between ferrets (
14–16).
Previous studies have reported on the low pathogenicity of reassortant viruses of clade 2.3.4.4, such as the Korean, European, and American A/H5N8 viruses and American A/H5N2 viruses in ferrets (
14–16). However, in contrast to these viruses that had acquired internal genes from LPAI viruses from wild birds, A/H5N6 GZ/14 possesses internal genes from the A/H5N1 GsGd lineage and also the adaptive substitution E627K in PB2 that might increase its pathogenicity in mammalian hosts. Ferrets are used as a model organism to study the pathogenesis of influenza viruses, as their susceptibility to infection with human and avian influenza viruses is similar to that of humans, and they develop respiratory disease similar to that observed in humans. Using the ferret model, we showed that the pathogenicity of A/H5N6 GZ/14 largely depends on the route of inoculation, as described previously for A/H5N1 viruses (
24). Upon intranasal inoculation, the pathogenicity of A/H5N6 GZ/14 was intermediate between the pathogenicities of A/H5N1 IN/05 and A/H5N8 ck/NL/14 (
14). The virus was detected in the olfactory bulb as early as 3 dpi, and it is possible that A/H5N6 GZ/14 would also be able to spread to the brain as seen with A/H5N1 IN/05, 7 days after intranasal inoculation (
24,
43). Ferrets lost on average 13% of their body weight over the time of the experiment, displayed more clinical signs than A/H5N8 ck/NL/14-inoculated ferrets, but did not progress to pneumonia as seen in A/H5N1 IN/05-inoculated animals. Differences between A/H5N6 and A/H5N1 might just be due to the route of inoculation—intranasal—which results in virus deposition in the lower respiratory tracts of some ferrets but not necessarily all (
23,
24). Therefore, to assess the ability of A/H5N6 GZ/14 to cause pneumonia, which is observed in most humans infected with A/H5 GsGd viruses, ferrets were inoculated intratracheally with 10
6 TCID
50 of virus. Upon intratracheal challenge, all ferrets developed a severe and fatal pneumonia characterized by severe DAD. High levels of replication of A/H5N6 GZ/14 in the lungs of infected ferrets upon deposition of the virus in the lungs correlates with the study by Hui et al., in which A/H5N6 GZ/14 was shown to replicate to very high titers in lung explants (
31). Sun et al. (
32) also showed that A/H5N6 viruses isolated from healthy ducks were not as pathogenic in ferrets as a 2.3.4 A/H5N1 virus, but these A/H5N6 viruses carried internal genes from different origin.
To conclude, the results from this study demonstrate that an A/H5N6 virus of the G1 genotype is not fully adapted to mammalian hosts and therefore presents a low risk to humans, although it caused more severe disease in ferrets than other clade 2.3.4.4 viruses. However, considering the diversity of genotypes among A/H5N6 viruses, more extensive characterization and risk assessment of A/H5N6 viruses of different genotypes is required in order to completely assess the risks. Moreover, owing to its pathogenicity in ferrets and its antigenic divergence from other 2.3.4.4 A/H5 Gs/Gd viruses, A/H5N6 GZ/14 virus could be used as a challenge candidate for vaccination/challenge studies to assess the ability of different vaccination strategies to elicit protective responses to emerging Gs/Gd HPAI viruses.
MATERIALS AND METHODS
Cells.
Madin-Darby canine kidney (MDCK) cells (ATCC) were cultured in Eagle’s minimal essential medium (EMEM) (Lonza) supplemented with 10% fetal bovine serum (FBS) (Greiner), 100 U/ml penicillin (PEN) (Lonza), 100 U/ml streptomycin (STR) (Lonza), 2 mM l-glutamine (l-Glu) (Lonza), 1.5 mg/ml sodium bicarbonate (NaHCO3) (Lonza), 10 mM HEPES (Lonza) and 1× nonessential amino acids (NEAA) (Lonza). 293T cells were cultured in Dulbecco modified Eagle’s medium (DMEM) (Lonza) supplemented with 10% FBS, 100 U/ml PEN, 100 U/ml STR, 2 mM l-Glu, 1 mM NaHCO3, and 1× NEAA. Vero cells were cultured in Iscove’s modified Dulbecco’s medium (IMDM) (Lonza) supplemented with 10% FBS, 100 U/ml PEN, 100 mg/ml STR, and 2 mM l-Glu.
Viruses.
All eight gene segments of a human virus isolate, A/H5N6 A/Guangzhou/39715/2014 (GZ/14), were amplified from copy DNA by PCR using specific primers and cloned in a modified version of the bidirectional reverse genetics plasmid pHW2000 (
44). Recombinant viruses were rescued by reverse genetics upon transfection of 293T cells as previously described (
44). For
in vivo experiments, a full recombinant A/H5N6 GZ/14 virus was used. For
in vitro experiments (thermostability, resialylated turkey red blood cell [TRBC], and hemagglutination inhibition assays), recombinant viruses containing six gene segments of A/Puerto Rico/8/1934 (PR/34), the HA without the multibasic cleavage site (MBCS) and the NA of A/H5N6 GZ/14 or A/H5N8 A/chicken/Netherlands/EMC-3/2014 (ck/NL/14) were used. Control viruses for thermostability and resialylated red blood cell assays were recombinant viruses containing seven gene segments of PR/34 and the corresponding HA without the MBCS (if applicable). Prototypic viruses used in hemagglutination inhibition assay were recombinant viruses containing six gene segments of PR/34, the HA without the MBCS, and the NA of the corresponding virus with the exception of the A/goose/Eastern China/1112/2011 and A/gyrfalcon/Washington/41088-6/2014, which was a recombinant virus carrying seven genes of PR/34 and the corresponding HA. Virus stocks were propagated up to three times in MDCK cells and titrated in MDCK cells as described below. The sequences of the virus stocks used for every experiment were checked using Sanger sequencing. Virus names were abbreviated as follows: A/Indonesia/5/2005 (IN/05), A/Netherlands/213/2003 (NL/03), A/Mallard/Sweden/49/2002 (ml/SW/02), A/Hong Kong/156/1997 (HK/97), A/Vietnam/1194/2004 (VN/04), A/turkey/Turkey/1/2005 (tk/TK/05), A/Anhui/1/2005 (AN/05), A/gyrfalcon/Washington/41088-6/2014 (gy/WAS/14), and A/goose/Eastern China/1112/2011 (gs/EC/11).
Titrations.
MDCK cells were inoculated with 10-fold serial dilutions of virus stocks, nose swabs, throat swabs, or homogenized tissue samples. The cells were washed with phosphate-buffered saline (PBS) 1 h after inoculation and cultured in infection medium, consisting of EMEM supplemented with 100 U/ml PEN, 100 U/ml STR, 2 mM
l-Glu, 1.5 mg/ml NaHCO
3, 10 mM HEPES, 1× NEAA, and 20 μg/ml trypsin (
N-tosyl-
l-phenylalanine chloromethyl ketone [TPCK]-treated trypsin; Sigma). Three days after inoculation, supernatants of cell cultures were tested for agglutinating activity using TRBCs as an indicator of virus replication. Infectious virus titers were calculated from 4 replicates each of the homogenized tissue samples, nose swabs, and throat swabs and from 10 replicates of the virus stocks by the method of Reed and Muench (
45).
Ferret experiments.
Ferrets were housed and experiments were conducted in strict compliance with European guidelines (European Union [EU] directive on animal testing 86/609/EEC) and Dutch legislation (Experiments on Animals Act, 1997). All animal experiments were approved by the independent animal experimentation ethical review committee “stichting dier experimenten commissie (DEC) consult” (Erasmus MC permit number EMC4028). Research projects involving laboratory animals can be executed only if they are approved by the animal experiments committee (DEC). The DEC considers the application and pays careful attention to the effects of the intervention on the animal and its discomfort and weighs this against the social and scientific benefit to humans or animals. The researcher is required to keep the effects of the intervention to a minimum, based on the three R’s (refinement, replacement, reduction).
(i) Ferret transmission experiments.
Four ferrets were inoculated intranasally with a total dose of 106 TCID50 of virus distributed into the two nostrils (250 µl per nostril). Each donor ferret was then placed in a transmission cage. One day after inoculation, one naive recipient ferret was placed opposite of each donor ferret. Each transmission pair was housed in a separate transmission cage designed to prevent direct contact but to allow airflow from the donor ferret to the recipient ferret. Nose and throat swabs were collected at 1, 3, 5, and 7 days postinoculation (dpi) from donor ferrets and at 1, 3, 5, 7, and 9 days postexposure (dpe) from the recipient ferrets and subjected to endpoint titration in MDCK cells. Donor ferrets were euthanized at 7 dpi and recipient ferrets were euthanized at 14 dpe to allow assessment of seroconversion.
(ii) Ferret pathogenesis experiments.
Six ferrets were inoculated intranasally with a total dose of 10
6 TCID
50 of A/H5N6 GZ/14 by applying 250 μl of virus suspension dropwise to each nostril. Clinical scores in all groups were assessed every day. Activity status was scored as follows: 0, alert and playful; 1, alert and playful only when stimulated; 2, alert but not playful when stimulated; 3, neither alert nor playful when stimulated. Dyspnea was characterized by open-mouth breathing with exaggerated abdominal movement (
46). Body weight was recorded daily. Throat and nose swabs were collected every day and were stored at −80°C in transport medium (Hanks’ balanced salt solution containing 0.5% lactalbumin [Sigma], 10% glycerol [Sigma], 200 U/ml PEN, 200 mg/ml STR, 100 U/ml polymyxin B sulfate [Sigma], and 250 mg/ml gentamicin [Gibco]) for endpoint titration in MDCK cells. At 3 and 6 dpi, three animals from each group were euthanized by exsanguination under anesthesia and were necropsied according to a standard protocol (
46). Tissues harvested for virological examination (right nasal turbinates, trachea, right bronchus, right lungs, tracheobronchial lymph nodes, right tonsils, right olfactory bulb, right cerebrum, right cerebellum, heart, liver, spleen, kidney, adrenal gland, pancreas, and jejunum) were homogenized in transport medium using the FastPrep system (MP Biomedicals) with two one-quarter-inch ceramic sphere balls, centrifuged 1,500 ×
g for 10 min, aliquoted, and stored at −80°C for endpoint titration in MDCK cells. Tissues harvested for histological examination (nasal turbinates, larynx, trachea, left bronchus, left lung, tracheobronchial lymph nodes, heart, liver, spleen, kidney, adrenal gland, esophagus, duodenum, jejunum, colon, pancreas, mesenteric lymph node, left olfactory bulb, left cerebrum, left cerebellum, and eye lid) were fixed in 10% neutral buffered formalin, embedded in paraffin, sectioned at 4 μm, and stained with hematoxylin and eosin (HE) for examination by light microscopy.
In a second experiment, three ferrets were inoculated intratracheally with 3 ml of virus dilution containing 106 TCID50. The experimental design and procedures were similar to those performed for the intranasal inoculation.
Immunohistochemistry (IHC).
For detection of influenza A virus antigen, sequential slides of all tissues were stained with a primary antibody against the influenza A virus nucleoprotein (NP) as described previously (
46). In each staining procedure, an isotype control was included as a negative control, and a lung section from a cat infected experimentally with A/H5N1 was used as a positive control.
Resialylated turkey red blood cell assay.
All sialic acids (SA) were removed from the surfaces of TRBCs by incubating 1% TRBCs in PBS with 50 mU of Vibrio cholerae NA (VCNA) (Roche) in 8 mM calcium chloride (CaCl2) at 37°C for 1 h. After two washing steps with PBS, VCNA-treated TRBCs were resuspended in PBS containing 1% bovine serum albumin (BSA) to a final concentration of 0.5% TRBCs. Removal of SA was confirmed by observation of a complete loss of hemagglutination of the TRBCs by control influenza A viruses. Subsequently, resialylation was performed using 0.5 mU of α2,3-N-sialyltransferase (Sigma) or 25 mU of α2,6-N-sialyltransferase (Sigma) and 1.5 mM CMP-sialic acid (Merck) at 37°C for 2 h in a final volume of 75 μl to produce α2,3-TRBCs and α2,6-TRBCs, respectively. After two washing steps, the TRBCs were resuspended in PBS containing 1% BSA to a final concentration of 0.5% TRBCs. Resialylation was confirmed by hemagglutination of viruses with known receptor specificity: A/H5N1 IN/05 and A/H3N2 NL/13. The receptor specificity of A/H5N8 ck/NL/14 and A/H5N6 GZ/14 was tested by performing a standard hemagglutination assay with the modified TRBCs.
Syncytium formation assay.
Influenza virus HA-induced cell fusion was tested in Vero-118 cells transfected with 5 μg of HA cloned in pCAGGs expression plasmid using Xtremegene transfection reagent according to the manufacturer’s instructions (Roche). One day after transfection, the cells were harvested using trypsin-EDTA and plated in 24-well plates. The next morning, the cells were washed, and medium was replaced with IMDM (Lonza) containing 10 μg/ml of trypsin. After 1 h, the cells were washed with PBS and exposed to PBS at pH 4.8, 4.9, 5.0, 5.1, 5.2, 5.3, 5.4, 5.5, 5.6, 5.7, 5.8, or 5.9 for 10 min at 37°C. Subsequently, the PBS was replaced by IMDM supplemented with 10% FBS, 100 U/ml PEN, 100 mg/ml STR, and 2 mM l-Glu. The next day, the cells were fixed using 80% ice-cold acetone, washed, and stained using a 20% Giemsa solution (Merck Millipore, Darmstadt, Germany). Visual inspection of the cell cultures for the presence of syncytia (multinucleated cells) was used to determine the pH threshold required for fusion.
Thermostability assay.
In short, viruses were diluted to 128 HA units (HAU)/25 μl using PBS. The samples were incubated in a thermal cycler for 30 min at temperatures of 50°C, 51.7°C, 54.3°C, 56°C, 58.5°C, and 60°C. Subsequently, the HA titer was determined by performing a hemagglutination assay using TRBCs.
Polymerase assay.
293T cell monolayers were transfected with 125 ng of luciferase reporter plasmid (firefly open reading frame flanked by the noncoding regions of the M segment under control of a PolI promoter) and 12.5 ng of an internal control plasmid (Renilla gene under the cytomegalovirus [CMV] promoter) together with the mix of PB2, PB1, PA, and NP plasmids, cloned in a bidirectional reverse genetics plasmid, in quantities of 125, 125, 125, and 250 ng, respectively. The transfected cells were incubated at 33°C or 37°C. After 24-h incubation, the supernatants were discarded, and the cell extracts were prepared in 100 μl of lysis buffer. The luciferase levels were assayed with luciferase assay system (Promega) and detected by a luminometer. Polymerase complexes of control viruses were included in the assay: a human H1N1 virus A/Hong Kong/54/1998 (HK/98), two human A/H5N1 viruses A/Vietnam/1203/2004 (VN/1203/04) and A/Indonesia/5/05 (IN/05), and an avian virus A/H5N2 A/mallard/Netherlands/3/1999 (ml/NL/99).
Hemagglutination inhibition assay.
Ferret antisera were prepared by intranasal inoculation with 500 µl recombinant virus containing six gene segments of PR/34 (
47), the corresponding HA without the MBCS, and the corresponding NA. Fourteen days postinoculation, ferrets were boosted by injecting subcutaneously 500 µl of a 1:1 mix of concentrated virus (>2,048 HAU) and Titermax Gold adjuvant (Sigma-Aldrich, St. Louis, MO, USA). Antisera were collected 2 weeks later. Ferret antisera were then pretreated overnight with receptor-destroying enzyme (VCNA) at 37°C and incubated at 56°C for 1 h the next day. Antisera were subsequently treated with 10% TRBCs at 4°C for an hour. Twofold serial dilutions of the antisera, starting at a 1:20 dilution, were mixed with 25 μl of a virus stock containing 4 hemagglutinating units and were incubated at 37°C for 30 min. Subsequently, 25 μl of 1% TRBCs was added, and the mixture was incubated at 4°C for 1 h. Hemagglutination inhibition (HI) was read and was expressed as the reciprocal value of the highest dilution of the serum that completely inhibited agglutination of virus and erythrocytes. The detection limit corresponds to half of the first antisera dilution (<10).
Biosafety.
All experiments were conducted within the enhanced animal biosafety level 3 (ABSL3+) facility of Erasmus MC. The ABSL3+ facility consists of a negative-pressure laboratory (−30 Pa) in which all in vivo and in vitro experimental work is carried out in class 3 isolators or class 3 biosafety cabinets, which also have negative pressure (less than −200 Pa). Although all experiments are conducted in closed class 3 cabinets and isolators, special personal protective equipment, including laboratory suits, gloves, and FFP3 facemasks, is used. Air released from the class 3 units is filtered by high-efficiency particulate air (HEPA) filters and then leaves via the facility ventilation system, again via HEPA filters. Only authorized personnel that have received the appropriate training can access the ABSL3+ facility. For animal handling in the facilities, personnel always work in pairs. The facility is secured by procedures recognized as appropriate by the institutional biosafety officers and facility management at Erasmus MC and Dutch and U.S. government inspectors.