Evaluation of bioaerosol samplers for the detection and quantification of influenza virus from artificial aerosols and influenza virus-infected ferrets

Abstract

Background: Bioaerosol sampling devices are necessary for the characterization of infectious bioaerosols emitted by naturally-infected hosts with acute respiratory virus infections. Assessment of these devices under multiple experimental conditions will provide insight for device use.

Objectives: The primary objective of this study was to assess and compare bioaerosol sampling devices using a) an in vitro, environmentally-controlled artificial bioaerosol system at a range of different RH conditions and b) an in vivo bioaerosol system of influenza virus-infected ferrets under controlled environmental conditions. Secondarily, we also sought to examine the impact of NSAIDs on bioaerosol emission in influenza virus-infected ferrets to address its potential as a determinant of bioaerosol emission.

Methods: We examined the performance of low and moderate volume bioaerosol samplers for the collection of viral RNA and infectious influenza virus in vitroand in vivo using artificial bioaerosols and the ferret model of influenza virus infection. The following samplers were tested: the polytetrafluoroethylene filter (PTFE filter), the 2-stage National Institute of Occupational Safety and Health cyclone sampler (NIOSH cyclone sampler) and the 6-stage viable Andersen impactor (Andersen impactor).

Results: The PTFE filter and NIOSH cyclone sampler collected similar amounts of viral RNA and infectious virus from artificially-generated aerosols under a range of relative humidities (RH). Using the ferret model, the PTFE filter, NIOSH cyclone sampler and the Andersen impactor collected up to 3.66 log₁₀ copies of RNA/L air, 3.84 log₁₀ copies of RNA/L air and 6.09 log₁₀ copies of RNA/L air respectively at peak recovery. Infectious virus was recovered from the PTFE filter and NIOSH cyclone samplers on the peak day of viral RNA recovery.

Conclusion: The PTFE filter and NIOSH cyclone sampler are useful for influenza virus RNA and infectious virus collection and may be considered for clinical and environmental settings.

Keywords: bioaerosol samplers; bioaerosols; ferret model; influenza virus.

Figure 1

Artificial bioaerosol and environmental chambers. A, Custom‐built artificial aerosolization chamber with Collison nebulizer and B, Caron environmental test chamber for ferret experiments. PTFE filter (black), NIOSH cyclone sampler (red), and Andersen impactor (green) were used to sample influenza A virus–laden bioaerosols. All images were created or modified with SketchUp Pro software. Modified from Verreault, D

Figure 2

Size fractionation of influenza viral RNA and infectious virus collected by the NIOSH cyclone sampler. Predominately <1.0 μm particles formed at low and medium RH conditions. Viral RNA and infectious virus collected as A, percent of total H1N1 influenza virus RNA recovered, B, percent of total H1N1 influenza virus PFU recovered, C, percent of total H3N2 influenza virus RNA recovered, and D, percent of total H3N2 influenza virus PFU recovered by the NIOSH cyclone sampler according to particle size. Particle size ranges are >4.0 μm (black), 1.0‐4.0 μm (white), and < 1.0 μm (gray). Data are presented as means ± SEM, n = 3 for each relative humidity condition, n.d., not detected

Figure 3

No difference was determined for weight change or viral load between untreated and treated ferrets. Ferrets treated with meloxicam had significantly lower rectal temperatures on days 1 and 7 p.i. A, Weight (percent change from baseline) and B, temperature (°C) were recorded on day 1 and alternating days p.i. C, Viral load (log₁₀ PFU/mL) was determined from nasal washes on day 1 and alternating days p.i. Data are presented as means ± SEM, n = 4 ferrets per group. * P < .05

Figure 4

Polytetrafluoroethylene (PTFE) filter and NIOSH cyclone samplers retained viral RNA from the air within the environmental chamber used to house influenza virus–infected ferrets. A, Viral RNA (log₁₀ copies/L air) collected from untreated ferrets by the PTFE filter (black) and NIOSH cyclone sampler (>4.0 μm = dark gray, 1.0‐4.0 μm = white, and <1.0 μm = light gray). B, Viral RNA (log₁₀ copies/L air) collected from treated ferrets by the PTFE filter (black) and NIOSH cyclone sampler (>4.0 μm = dark gray, 1.0‐4.0 μm = white, and <1.0 μm = light gray). C, Infectious virus (total PFU) collected by the PTFE filter (black) and NIOSH cyclone sampler stage 3 (<1.0 μm, white) on day 3 p.i. One sample collected per day for each sampler (4 ferrets per sample), n.d., not detected

Figure 5

Viral RNA collected from the air emitted by IAV‐infected ferrets using the Andersen impactor. A, Viral RNA (log₁₀ copies/L air) collected from untreated ferrets by the Andersen impactor (>4.7 μm = gray, 2.1‐4.7 μm = white, and 0.65‐2.1 μm = black). B, Viral RNA (log₁₀ copies/L air) collected from treated ferrets by the Andersen impactor (>4.7 μm = gray, 2.1‐4.7 μm = white, and 0.65‐2.1 μm = black). One sample collected per day for each sampler (4 ferrets per sample)