Ionizing air affects influenza virus infectivity and prevents airborne-transmission

Abstract

By the use of a modified ionizer device we describe effective prevention of airborne transmitted influenza A (strain Panama 99) virus infection between animals and inactivation of virus (>97%). Active ionizer prevented 100% (4/4) of guinea pigs from infection. Moreover, the device effectively captured airborne transmitted calicivirus, rotavirus and influenza virus, with recovery rates up to 21% after 40 min in a 19 m(3) room. The ionizer generates negative ions, rendering airborne particles/aerosol droplets negatively charged and electrostatically attracts them to a positively charged collector plate. Trapped viruses are then identified by reverse transcription quantitative real-time PCR. The device enables unique possibilities for rapid and simple removal of virus from air and offers possibilities to simultaneously identify and prevent airborne transmission of viruses.

Full text

1
SCIENTIFIC RepoRts | 5:11431 | DOI: 10.1038/srep11431
www.nature.com/scientificreports
Ionizing air aects inuenza virus
infectivity and prevents airborne-
transmission
Marie Hagbom
1,*
, Johan Nordgren
1,*
, Rolf Nybom
2
, Kjell-Olof Hedlund
3
, Hans Wigzell
2
&
Lennart Svensson
1
By the use of a modied ionizer device we describe eective prevention of airborne transmitted
inuenza A (strain Panama 99) virus infection between animals and inactivation of virus (>97%).
Active ionizer prevented 100% (4/4) of guinea pigs from infection. Moreover, the device eectively
captured airborne transmitted calicivirus, rotavirus and inuenza virus, with recovery rates up to
21% after 40 min in a 19 m
3
room. The ionizer generates negative ions, rendering airborne particles/
aerosol droplets negatively charged and electrostatically attracts them to a positively charged
collector plate. Trapped viruses are then identied by reverse transcription quantitative real-time
PCR. The device enables unique possibilities for rapid and simple removal of virus from air and oers
possibilities to simultaneously identify and prevent airborne transmission of viruses.
ere is an urgent need for simple, portable and sensitive devices to collect, eliminate and identify viruses
from air, to rapidly detect and prevent outbreaks and spread of infectious diseases
1
. Each year, infectious
diseases cause millions of deaths around the world and many of the most common infectious pathogens
are spread by droplets or aerosols caused by cough, sneeze, vomiting etc.
2–5
. Knowledge of aerosol trans-
mission mechanisms are limited for most pathogens, although spread by air is an important transmission
route for many pathogens including viruses
6
.
Today no simple validated technology exists which can rapidly and easily collect viruses from air and
identify them. e problem is not the analyzing technique, since molecular biological methods such as
real-time PCR enable a sensitive detection system of most pathogens
7–9
. e diculty is to develop an
eective sampling method to rapidly collect small airborne particles including viruses from large vol-
umes of air. Furthermore, the sampling method should be robust with easy handling to enable a wide
distribution and application in many types of environment. At present, the most commonly used tech-
niques aimed to collect pathogens from air are airow and liquid models
10–15
. ese systems are complex,
and their eciency has not been thoroughly evaluated.
Spread of infectious diseases in hospitals can be most signicant
16–18
. In many situations there is a
need for a pathogen- and particle-free environment, e.g. in operation wards, environments for immuno-
suppressed patients as well as for patients with serious allergies. is makes it desirable to have a method
not only for collection and identication
19
, but also for eliminating virus and other pathogens from air
20
.
Ozone gas has been shown to inactivate norovirus and may be used in empty rooms to decontaminate
surfaces, however in rooms with patients ozone should not been used due to its toxicity
21
. Generation
of negative ions has previously been shown to reduce transmission of Newcastle disease virus
22,23
and
several kind of bacteria
24,25
in animal experimental set-ups.
e ionizing device used in this study operates at 12 V and generates negative ionizations in an elec-
tric eld, which collide with and charge the aerosol particles. ose are then captured by a positively
1
Division of Molecular Virology, Department of Clinical and Experimental Medicine, University of Linköping, 581
85 Linköping, Sweden.
2
Department of Microbiology, Karolinska Institute, Stockholm, Sweden.
3
Department
of Diagnostics and Vaccine, Swedish Institute for Communicable disease Control, Stockholm, Sweden.
*
These
authors contributed equally to this work. Correspondence and requests for materials should be addressed to L.S.
(email: lennart.t.svensson@liu.se)
Received: 27 November 2014
Accepted: 13 May 2015
Published: 23 June 2015
OPEN

www.nature.com/scientificreports/
2
SCIENTIFIC RepoRts | 5:11431 | DOI: 10.1038/srep11431
charged collector plate. For safety reasons, the collector plate has a very low current, less than 80μ A,
however the ionizer accelerates a voltage of more than 200,000 eV, which enables high production of
several billion electrons per second. Moreover, this device does not produce detectable levels of ozone
and can thus be safely used in all environments.
is technique is known to eectively collect and eliminate cat-allergens from air
26
. Aerosolized rota-
virus, calicivirus and inuenza virus particles exposed to the ionizing device were attracted to the col-
lector plate and subsequently identied by electron microscopy and reverse transcription quantitative
real-time PCR techniques. Most importantly, we demonstrate that this technology can be used to prevent
airborne-transmitted inuenza virus infections.
Results
Visualization and eciency of aerosol sampling as determined by electron microscopy. To
develop and validate the ionizing technique for collection and identication of viral pathogens, we used
several viruses of clinical importance; calicivirus, rotavirus and inuenza virus (H3N2, strain Salomon
Island) as well as latex particles. Canine calicivirus (CaCV, strain 48) was used as a surrogate
27
for human
norovirus, the aetiological agent behind the “winter vomiting disease”, causing outbreaks of great clinical
and economic importance
28
. Rhesus rotavirus was used as a surrogate marker for human rotavirus
29
.
e device (Fig.1a) consists of a small portable 12 volt operated ionizer, with a collector plate of pos-
itive charge attached to the ionizer, attracting negative particles from the air by electrostatic attraction.
To determine optimal time collection parameters, latex particles with sizes ranging from < 1 to > 10 μ m
were nebulized into a room of 19 m
3
. Testing revealed that 40–60 min was required to eliminate > 90%
of free latex particles in the air as determined by real-time particle counting (PortaCount Plus). e
particle counter can detect particles with size greater than 0.02 M. Visualization by scanning electron
microscopy (SEM) on grids from active- and inactive ionizer collector plates showed that accumulation
of latex particles was dramatically enhanced on active ionizer collector plates compared to the inactive
(Fig.1b,c). Next, high numbers of rotavirus and formalin-inactivated inuenza virus were aerosolized
Figure 1. Airpoint ionizer with collector plate (size 13 × 35 cm) (a). e ionizing device was developed
based of the Ion-Flow Ionizing Technology from LightAir AB, Solna, Sweden and was modied by installing
a plastic-cup with a conductive surface of 47 mm in diameter, with positive charge, as the collector plate;
Aerosolized and trapped latex particles (>1 to < 10 μ m) on active (b) and inactive (c) ionizer, (bar = 10 μ M);
Rotavirus (d); and inuenza virus (H1N1; strain Salomon Island) (e) trapped on active ionizer,
(Bar = 50 nm).

www.nature.com/scientificreports/
3
SCIENTIFIC RepoRts | 5:11431 | DOI: 10.1038/srep11431
under the same conditions. While, aer 40 min the inactive collector plates contained few (< 5) rotavirus
and inuenza virus, the active collector contained > 50 virus particles, as determined by transmission
electron microscopy (TEM), (Fig.1d,e).
Ionizing air and electrostatic attraction collects aerosol-distributed viruses as determined by
RT-qPCR. We next determined the capacity of RT-qPCR technology to quantitate the capacity of
the ionizer technique to collect and concentrate viruses. ree independent experiments with each of
the three viruses were carried out using the same virus concentrations in each experiment (Fig.2a–c).
Although several steps are involved from collection to detection the system was robust as to reproduci-
bility. e RT-qPCR data shows that the active collector is concentrating and collecting virus 1500–3000
times more ecient as compared to the inactive collector (Table1). When dierent dilutions of virus
was used for aerosol production the proportion of aerosolized virus collected on the active collector was
normally in the range of 0.1–0.6% for CaCV, rotavirus and inuenza virus. A reproducible nding with
Figure 2. Real-time PCR on trapped rotavirus (a), calicivirus (b) and inuenza virus (H1N1; strain
Salomon Island) (c). Note that no inuenza virus was detected on the inactive ionizer.
Table 1. Collection eciency of aerosolized CaCV, rhesus rotavirus (RRV) and Inuenza A virus in
various concentrations as determined by RT-q PCR. a) is experiment was performed only once.

www.nature.com/scientificreports/
4
SCIENTIFIC RepoRts | 5:11431 | DOI: 10.1038/srep11431
regard to CaCV was a signicant increase in relative recovery at the lowest concentrations increasing to
10–20% of the total amount of virus aerosolized (Table1).
Ionizing air reduces calicivirus and rotavirus infectivity. Next we determined if collected viruses
retained their infectivity aer being exposed to negative ions and/or aer being exposed to the positively
charged collector plate. Five mL of cell culture medium (Eagles Minimal Essential Media (Eagles MEM))
containing 1 × 10
6
peroxidase forming units of rotavirus respectively of CaCV were aerosolized and col-
lected during 40 min to an active collector plate, containing 1 mL of Eagles MEM. CaCV in cell culture
medium was also directly exposed to an active and inactive collector plate, without being aerosolized.
Viral infectivity was determined essentially as described
30
and the ratio between viral genome copy num-
bers versus infectivity was compared between aerosolized virus, virus exposed to active- and inactive
collector plates and the viral stocks. CaCV exposed to an active collector plate, without being aerosolized,
showed a slight reduction in infectivity (~40%) in comparison to virus that have been trapped on an
inactive collector plate (Table2). In contrast, the infectivity of aerosolized viruses was greatly reduced
by > 97%, indicating that ionization of the aerosol accounts for the vast majority of infectivity reduction,
and not the exposure to the charged collector plate.
Further support that ionizing was the mechanism by which viruses lost infectivity comes from exper-
iments were rotavirus was nebulized without ionizing and allowed to be trapped to an inactive collector
plate. Collectors were located at 30 cm from the nebulizer. e result concluded that the genome copy
versus infectivity ratio was unchanged from that of the viral stock, thus suggesting that inactivation of
virus is associated with ionized air.
Ionizing air and electrostatic attraction prevents airborne-transmitted inuenza A/Panama
virus infection between guinea pigs. Next we took advantage of an established inuenza guinea
pig model
31–33
to study if ionizing air and electrostatic attraction could prevent airborne aerosol and
droplet transmitted inuenza A/Panama (Pan/99) virus infection between guinea pigs. e airborne/
droplet transmission model was established essentially as described
31
using two separate cages with the
ionizer placed between the cages (Fig.3). Four guinea pigs were infected by intranasal route as described
with 5 × 10
3
pfu of Pan/99
31
and placed in cage “A” (Fig.3). At 30 hours post infection (h p.i.) 4 uninfected
guinea pigs were placed in cage “B” 15 cm from the cage with infected animals as illustrated in Fig.3,
with no physical contact. e ionizer was placed between cages “A” and “B”. Two identical experiments
were performed, one with active ionizer placed between the cages and one with an inactive ionizer.
Ratio of infectious virus particles to virus genes per PCR-reaction as
quantied by RT-qPCR
Exposed
to charged
collector
Exposed to
uncharged
collector
Reduction
of
infectivity
Aerolized
virus
Aerolized
virus captured
Reduction
of
infectivity
CaCV 0.74 × 10
−4
1.24 × 10
−4
40.1% 2.96 × 10
−2
< 7.83 × 10
−4a
> 97.4%
a
RRV n.d. n.d. n.d. 4.86 × 10
−1
< 7.66 × 10
−3a
> 98.4%
a
Table 2. Reduction of infectivity of Canine Calicivirus (CaCV) and Rhesus Rotavirus (RRV).
a
Under
detection limit (10 peroxidase forming units/mL) on the infectivity assay.
Figure 3. Set-up design of inuenza virus (H3N2, Pan/99) aerosol-transmission experiments between
guinea pigs. Guinea pigs (n = 4) were intranasally infected with 5 × 10
3
pfu of Pan/99 virus in 100 uL
(50 uL in each nostril). All four infected animals were placed into an experimental cage “A”. At 30 h p.i. four
naïve uninfected guinea pigs were placed in cage “B” . Air-ow from le to right. Air exchanged 17x/day.
Filled rectangle = ionizer.

www.nature.com/scientificreports/
5
SCIENTIFIC RepoRts | 5:11431 | DOI: 10.1038/srep11431
Uninfected animals in cage “B” were exposed for 24 hours with airow from cage “A” hosting the 4
infected guinea pigs and then placed in individually ventilated cages for the next 21 days, to ensure that
the only time-point for being infected was the 24 hours when they were exposed to air from infected
animals in cage “A”. RT-qPCR of lung- and trachea biopsies examined at 54 h p.i. from the nasally exper-
imentally infected animals, revealed that 3 out of 4 guinea pigs in both experiments were positive for
inuenza.
We assessed transmission of infection from animals in cage “A” to exposed uninfected animals in cage
“B” by development of an immune response 21 days post exposure. e results shown in Fig.4 illus-
trate that when the ionizer was inactive, 3 of the 4 uninfected but exposed animals developed a serum
IgG inuenza-specic immune responses. In contrast, none of the 4 animals in cage “B” developed an
immune response to inuenza virus when the ionizer was active (Fig.4). Furthermore, inuenza virus
RNA could be detected by RT-qPCR, albeit at low concentration, on the collector plate from the active
ionizer but not with the inactive ionizer, showing that the ionizing device indeed collected virus excreted
from the infected animals in cage “A”.
Discussion
We describe a simple ionizing device operating at 12 volt that can prevent spread of airborne transmit-
ted viral infections between animals in a controlled setting, whilst simultaneously collecting virus from
air for rapid identication. Coupled with sensitive RT-qPCR assays, this sampling method enabled fast
detection and highly sensitive quantication of several human clinically important viruses such as inu-
enza virus, rotavirus and calicivirus. e device consists of a small portable ionizer, where a sampling
cup of positive charge is attached to the ionizer attracting negative particles from the air. Important
advantages with this novel ionizing device is the simple handling, high robustness as well as the wide
applicability to airborne pathogens.
e observation that signicantly higher numbers of rotavirus and CaCV particles were detected
on the active ionizer compared to the inactive ionizer (~1500–3000 times), led to the conclusion that
this technique can actively and eciently collect viral particles from air. Similarly, visualization of latex
particles by SEM revealed that latex particles of all sizes investigated were concentrated on the active
collector. It is interesting to note that a broad range of particles sizes, from 35 nm to 10 μ m was concen-
trated, suggesting a wide application range of the technology. However, too large particles may decrease
the recovery since these are proposed to remain for less time in the air
33,34
.
Interestingly, when we aerosolized low amounts of CaCV, (1.56 × 10
4
gene copies and 1.87 × 10
3
gene copies), we observed collection recoveries of 10.6 and 21%, respectively. is markedly increased
Figure 4. Active ionizer prevents aerosol transmitted inuenza virus (H3N2, Pan/99) infection between
guinea pigs. While the active ionizer prevented 4 of 4 exposed guinea pigs from developing an immune
response to inuenza virus, 3 of 4 animals were infected when the inactive ionizer was used. Graph shows
antibody titers by ELISA before infection (pre-serum 1, 2, 3 and 4) and at day 21 post-exposure to inuenza
virus (post-serum 1, 2, 3 and 4). Briey, inuenza virus H1N1; (SBL Inuenza Vaccine, Sano Pasteur,
Lyon, France) were coated on ELISA plates and incubated with two-fold dilutions of pre- and post- guinea
pig sera, followed by biotinylated rabbit-anti-guinea pig antibody, HRP conjugated streptavidin and TMB
substrate as described in Methods. Cut o (dashed line) value (0.284 OD) was the mean of the negative
controls + 2 SD.

www.nature.com/scientificreports/
6
SCIENTIFIC RepoRts | 5:11431 | DOI: 10.1038/srep11431
eciency, with smaller amounts of virus distribution in air, could be due to less aggregation of virus-virus
or virus-cell debris particles more long lasting airborne, and thus leads to stronger electrostatic attraction
by the collector. Furthermore, it is likely that much particles end up at the walls of the collector plate or
on areas adjacent to the collector plate on the ionizer; and are subsequently not quantied by real-time
PCR; thus underestimating the electrostatic eect. When aerolizing higher virus concentrations, this
eect can thus lead to lower estimates of recovery. Using CaCV, rotavirus and inuenza virus, we per-
formed three independent experiments for each concentration of aerosolized virus in order to assess
the robustness of the assay throughout all steps (collection with active ionizer, RNA extraction, cDNA
synthesis and real-time PCR). Although several steps are involved from collection to detection we found
the assay to be highly robust since the minimum and maximum quantity of virus from each independent
measurement was always within a range of 1 log (Fig.2).
Inactivation of viruses by electrostatic attraction has only been briey investigated
35
. In the present
study, rotavirus and CaCV lost signicant (> 97%) infectivity (ratio; CaCV from 3.0 × 10
−2
to < 7.8 × 10
−4
and rotavirus from 4.9 × 10
−1
to < 7.6 × 10
−3
) in ionized air as determined by a ratio of infectivity ver-
sus gene copies. e mechanism of inactivation was not explicitly investigated in this study, but inac-
tivation mechanisms may include reactive species and/or increased protein charge levels, which could
inactivate virus as previously described
36,37
. Reduced infectivity has been proposed to be due to reactive
oxygen species and ozone, through lipid- and protein peroxidation reactions that may cause damage
and destruction to the viral lipid envelope and protein capsid
36
. In particular, protein peroxidation may
play a key role in the inactivation of non-enveloped viruses, such as adenovirus, poliovirus and other
enteroviruses such as rota- and caliciviruses. Enveloped viruses are suggested to lose infectivity due to
lipid peroxidation. However, the cytotoxity of ozone creates a major obstacle for the clinical application
of ozone. It has been shown hat increasing the ion concentration of the air eciently protect chickens
from air-born transmission of lethal Newcastle disease virus infection
23
. e exact mechanism of nega-
tive ion inactivation of viruses has not been shown and needs to be further investigated. However, in a
study using generation of negative and positive ions, inuenza virus was inactivated although ozone level
was negligible (0.005 ppm or less)
37
.
Our device released a steady-state ozone concentration below the detection limit (0.002 ppm) as
tested by VTT (Technical Research Center of Finland, Tampere, Finland) and by Air Resources Board
in the US, thus ozone cannot in this case be a contributor of viral inactivation. However, reactive radi-
cals such as •O
2
-
may be generated, which may contribute to inactivation through damage to either the
protein or the nucleic acid structure of the viruses
37
. As infectivity was not lost when virus was nebulized
into the air of the room without ionization and only slightly reduced when applied directly on the pos-
itively charged collector plate, it is suggested that most reduction of infectivity may be due to increased
negative charged levels, presumably resulting in changes in isoelectric point and thus structural changes
of the capsid. As the two viruses investigated are non-enveloped, lipid modication can be ruled out.
Most interesting, and of great clinical signicance of this study was the novel nding that the ionizing
device could detect and prevent inuenza virus infection in a controlled setting, mimicking “authentic”
conditions. Our intranasal infection protocol was essentially as previously described
31,33
using Harley
guinea pigs and 5 × 10
3
plaque forming units (pfu) of Pan/99 inuenza virus. As guinea pigs of the
Hartley strain are highly susceptible to human inuenza A virus strain Pan/99 (H3N2), with an infec-
tious dose (ID
50
) of 5 pfu
31
, this makes the viral strain most appropriate for these studies. Moreover,
Lowen and co-workers have shown 100% transmission of Pan/99 by aerosol to guinea pigs
38,39
. Previous
studies have also shown that the used infectious dose results in a viral growth peak around day 3 p.i. in
both lungs and nasal passages in this animal model
31
, a time point when naïve animals in our study was
exposed to air from the infected animals.
We found, by assessing development of the immune response, that 3 of 4 uninfected guinea pigs
became infected aer exposure to animals inoculated with 5 × 10
3
pfu of Pan/99. ese susceptibility
gures are similar to those of Mubareka and co-workers
33
who found that 2 of 3 guinea pigs became
infected following short-range aerosol transmission with a dose of 10
3
pfu whereas 3 of 3 animals became
infected with a infectious dose of 10
6
pfu. We examined the immune response at 21 days p.i., a time point
when Lowen and co-workers
40
previously have found that naturally Pan/99 infected guinea pigs had
developed a signicant immune response.
e mode of inuenza virus transmission includes direct contact with individuals, exposure to
virus-contaminated objects (fomites) and inhalation of infectious aerosols. Previous studies using the
guinea pig animal model have indicated that aerosol and not fomites is the principal route of Pan/99
transmission between guinea pigs
33
. Aerosol released virus from inoculated animals could be detected
on the active collector plate by RT-qPCR, albeit at very low gene copy numbers. Using the guinea pig as
a host model, Lowen et al.
38
showed that aerosol spread of inuenza virus between animals is dependent
upon both the relative humidity and temperature. ey found that low relative humidity of 20–30% and
temperature of 5 °C was most favourable, with no transmission detected at 30 °C. In our set-up system,
the temperature was between 20–21 °C and relative humidity between 35–36.2%.
e described ionizing device coupled with RT-qPCR assays has a clear diagnostic potential. e
easy handling, low cost, free of ozone production, robustness, high eciency and low-voltage (12 volt)
operation enables large-scale use. Locations critical for infectious spread, such as airplanes, hospitals,
day-care centres, school environments and other public places could thus be monitored and controlled

www.nature.com/scientificreports/
7
SCIENTIFIC RepoRts | 5:11431 | DOI: 10.1038/srep11431
by the collection and analysis of airborne viruses and other pathogens on the collector plate. e device
also show potential for transmission prevention, although the potency needs to be further investigated in
real-life settings. We conclude that this innovative technology hold great potential to collect and identify
viruses in environmental air.
Methods
Study design. e experimental room has grounded metal walls, with a volume of 19 m
3
(B250*L330*H235cm). One active and one inactive ionizer device, designed for collection and analysis
of particles in the air, were placed in the room at equal distance (215 cm) from the nebulizer (Aiolos
Albatross, Aiolos, Sweden), with a distance of 64 cm between the ionizers. A particle counter (PortaCount
Plus, TSI Incorporated, USA) was used before and during the experiment. Before the start of aerosol
experiments, the room was emptied on particles by the active ionizer and the collector plate was dis-
carded before the experiments begun and replaced with a new collector plate. e experiments were
continued until the particle counts were back to basal level, usually reached within 40 min. Humidity
and temperature conditions were measured initial to each aerosol experiments.
Ionizer technology and device. e ionizing device used in this study was developed on the basis of
the ion-ow ionizing technology from LightAir AB, Solna, Sweden (www.lightair.com) and was modied
for this work by the Department of Microbiology, Karolinska Institute, Stockholm, Sweden. e device
(size of 13 × 35 cm) was modied by installing a plastic-cup with a conductive surface of 47 mm in diam-
eter (GP plastindustri, Gislaved, Sweden) as the collector plate (Fig.1). e collector plate has for safety
reasons a very low current, less than 80 μ A, however the ionizer accelerates an extremely high voltage of
more than 200,000 eV. e ionizer creates electrons, which will render surface molecules of particles in
air negatively charged thus attracting them to the positively charged collector plate. is device generates
approximately 35 000 billion electrons per second (www.lightair.com) with a steady-state ozone concen-
tration below the detection limit (0,002 ppm) as tested by VTT Technical Research Center of Finland,
Tampere, Finland. It has also been ozone tested and certied by ARB (Air Resources Board) in the US.
Aer the end of the sampling period, the ionizer was turned o, and the collector plate was covered
with a lid and stored at − 20 °C until analysis. Viruses captured on the collector plates were analyzed by
a RT-qPCR for rotavirus, CaCV and inuenza virus, and the results from the active- and inactive ionizers
were compared. Scanning- and transmission electron microscopy were used for visualization of collected
viruses and latex-particles.
Aerosol experiments of virus and latex particles. Dierent amounts of rhesus rotavirus (geno-
type G3P[3]), inuenza virus (strain H1N1, Salomon Island, inactivated) and CaCV strain 48 (genus
Vesivir us) were diluted in water to a nal volume of 5 mL. In aerosol experiments for scanning electron
microscopy and infectivity analysis, virus was diluted in Eagles MEM. Virus suspensions in dierent
concentrations were distributed as aerosols in the room by the use of a nebulizer. Each experiment was
performed in triplicates and collection of aerosolized virus and latex particles were performed during
40 min.
Transmission electron microscopy (TEM). Carbon/Formvar-coated 400 mesh copper grids were
placed on the collector during aerosol experiment with inuenza- and rotavirus. Grids were then rehy-
drated in Eagles MEM containing 1% bovine serum albumin (BSA) before being negatively stained with
2% phosphotungstic acid and analyzed by TEM. Ten grid squares were analyzed per specimen and the
number of virus particles per unit area was calculated.
Scanning electron microscopy (SEM). Collected samples were added on the surface of a poly-
carbonate 0.6 μ m lter (Nucleopore, Inc) which was tted to an airtight gadget (GP Plastindustri AB,
Gislaved, Sweden). e lter was dried in room temperature, coated with 40 Å thick layer of ionized gold
and analyzed by SEM (Philips High Resolution SEM 515). e method has previously been used and
reported in studies of cytomegalovirus as well as cerebrospinal uid
41–43
.
Extraction of viral RNA from collector plates. e attached viral particles on the collector plates
were lyzed with 1 mL of viral lysis buer (buer AVL, QIAamp viral RNA mini kit) added directly into
the collector plate and immediately proceeded for extraction of viral RNA using QIAamp Viral RNA
Mini Kit (Cat.no: 52906 Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Each
sample was eluted with 60 μ L of RNase-free water containing 0.04% sodium azide (AVE buer; Qiagen,
Hilden, Germany).
Reversed transcription of extracted viral RNA. 28 μ L of the extracted viral RNA was mixed with
50 pmol of Pd(N)
6
random hexamer primer (GE-Healthcare, Uppsala, Sweden) and quickly chilled on
ice for 2 min, followed by the addition of one Illustra Ready-To-Go reverse transcriptase PCR (RT-PCR)
bead (GE-Healthcare, Uppsala, Sweden) and RNase-free water to a nal volume of 50 μ L. For rotavirus,
an initially denaturation step at 97 °C for 5 min was performed to denature the dsRNA. e reverse

www.nature.com/scientificreports/
8
SCIENTIFIC RepoRts | 5:11431 | DOI: 10.1038/srep11431
transcription (RT) reaction was carried out for 40 min at 42 °C to produce the cDNA later used for
real-time PCR.
Quantitative real-time PCR for rotavirus. Rhesus rotavirus was detected and quantied using a
LUX real-time PCR assay as described previously
7
. is real-time PCR uses labeled primers with dier-
ent uorophores for each VP6 subgroup and external plasmid standards for semi-quantication
44
.
Quantitative real-time PCR for CaCV. CaCV was detected and quantied using a SYBR green assay
on a ABI prism 7500 (Applied Biosystems, Foster City, CA) with primers; (nal concentration 200 nM)
CaCV-3 (5-ACCAACGGAGGATTGCCATC-3´ (nucleotides 393 to 410 according to GenBank acces-
sion no. AF053720) and CaCV-4 (5´-TAGCCGATCCCACAAGAAGACA-3´ (nucleotides 452 to 474),
specic for CaCV strain 48. e reaction was performed with 2 μ L cDNA in 10 μ L 2X SYBR Green PCR
Master Mix (Applied Biosystems) and water to a nal volume of 20 μ L. e following cycling program
was used: 95 °C for 10 min followed by 45 cycles of 95 °C for 15 seconds and 60 °C for 1 min. Melting
curve analysis was performed immediately aer PCR completion, by heating at 95 °C for 15 seconds,
followed by cooling to 60 °C for 1 min and subsequent heating to 95 °C at 0.8 °C min
−1
with continu-
ous uorescence recording. Melting temperatures were determined on all samples using the Sequence
Detection Soware version 1.3.1 (Applied Biosystems) and visualized by plotting the negative derivatives
against temperature.
Sampling for infectivity studies with rotavirus and CaCV. To determine whether the ionizing
technology has any inuence on virus infectivity, rhesus rotavirus and CaCV was aerosolized and col-
lected on active ionizer collector plates, covered with 1 mL of Eagles MEM. Rotavirus (1 × 10
6
peroxidase
forming units) and CaCV (1 × 10
6
peroxidase forming units) was aerosolized, each in a total volume
of 5 mL and collected for 40 min followed by determination of viral infectivity and number of genome
copies.
To determine if ionized air per se had an eect on viral infectivity, rhesus rotavirus was aerosolized
and captured on a collector plate containing 1 mL of Eagles MEM, without ionization, placed at a dis-
tance of 30 cm from the nebulizer.
To determine if electrostatic attraction of the collector plate aected viral infectivity, rotavirus (2 × 10
5
peroxidase forming units) and CaCV (2 × 10
5
peroxidase forming units) in 1 ml of Eagles MEM, were
added to inactive and active collector plates for 40 min. e plates were subsequently stored at − 20 °C
until determination of viral infectivity and number of genome copies.
Determination of rotavirus and CaCV infectivity. Rotavirus stock and samples were diluted 1:10
in Eagles MEM and subsequently diluted in two-fold dilutions. Determination of viral infectivity was
performed as previously described on conuent Green monkey kidney cells (MA104) in 96-well plates
30
.
CaCV infectivity was determined essentially as for rotavirus with the modication that samples were
added to conuent Madin-Darby Canine Kidney (MDCK) cells in 48-well plates and infectivity deter-
mined as previously described
45
and conrmed by RT-qPCR. To determine the reduction of infectivity,
the ratio of viral genome copy numbers versus infectivity was compared between aerosolized virus, virus
exposed to active- and inactive collector plates and the viral stock.
Animals. Guinea pigs, strain Hartley, female, 300–350 g, were housed at Astrid Fagraeus Laboratory,
Karolinska Institute, according to approved guidelines from the Board of Agriculture and the Council of
Europes Convention on vertebrate animals used for scientic purpose. e experimental protocol was
approved by the Animal Ethics Committee in Stockholm (Permit Number: N177/11).
Airborne transmission of inuenza virus. We use a guinea pig animal model to investigate whether
the ionizing technique could prevent transmission of inuenza virus infection, since this model have
successfully been used as a model of aerosol transmission studies of inuenza virus
31,33
. Human inuenza
A virus, strain Pan/99 (kindly provided by Peter Palese, New York, USA) was used since this strain has
been shown to eectively replicates in the upper respiratory airways and eectively transmit by aerosols
but not by fomites in guinea pigs. Female guinea pigs, 300–350 g, strain Hartley, were housed at Astrid
Fagreaus Laboratorium, Solna, Stockholm (Ethical permission N177/11). Four animals were anesthetized
by an intra peritoneal injection of ketamin (Ketalar el Ketaminol) 50 mg/kg and xylazin (Rompun) 5 mg/
kg and infected intranasally with 5 × 10
3
pfu of Pan/99 virus in 100 uL (50 uL in each nostril). All four
infected animals were placed into the experimental cage (Fig.3, cage “A”). At 30 h p.i., four naïve unin-
fected guinea pigs were placed next to the transmission cage (Fig.3, cage “B”) at a distance of 15 cm. Air
owed freely between cages, but direct contact between inoculated and exposed animals was prohibited.
e four naïve guinea pigs were exposed for 24 hours and then put into separately individually venti-
lated cages, to ensure that no aerosol transmission occurred between the animals. Two identical exper-
iments were performed, with an active and inactive ionizer. e nasally infected animals were removed
aer the exposure time and lung and trachea biopsies were collected (54 h p.i.) and investigated for
inuenza virus by RT-qPCR. At 21 days post exposure, serum was collected from the uninfected exposed
guinea pigs and the prevalence of antibodies against inuenza A virus was determined by ELISA. Sera

www.nature.com/scientificreports/
9
SCIENTIFIC RepoRts | 5:11431 | DOI: 10.1038/srep11431
taken before exposure to the infected guinea pigs (pre-sera), and 21 days days aer exposure (post-sera)
were analyzed from each animal.
ELISA detection of inuenza A antibodies. Briey, 96-well plates (Nunc, 96 F MAXISORP,
Roskilde, Denmark) where coated with formalin-inactivated Inuenza A virus H1N1 (SBL Inuenza
Vaccine, Sanoil Pasteur, Lyon, France) diluted in coating buer (0.05 M sodium carbonate buer, pH
9.5–9.7) at 5 g/mL and incubated at + 4 °C over night. Wells were washed x3 (0.9% NaCl and 0.05%
Tween-20) and blocked with 3% BSA in PBS buer for 1 hour at 37 °C. Serum samples were diluted 1:100
and further in two-fold dilutions in dilution buer (PBS containing 0.5% BSA and 0.05% Tween-20), and
incubated for 90 min at 37 °C. Plates were then washed x5 and incubated for 60 min at 37 °C with sec-
ondary biotinylated goat-anti guinea pig antibody (Vector, BA-7000) and horseradish-peroxidase (HRP)
conjugated Streptavidin (DAKO, Denmark, P0397), both at a dilution of 1:3000. Plates were then washed
x5 and 100 L of tetramethyl benzidine (TMB) substrate (Sigma Aldrich, T-0440-16) was added to each
well, the reaction developed for 10 min and stopped by addition of 100 L of 2 M H
2
SO
4
. Absorbance was
measured at 450 nm in an ELISA reader (VersaMax, Molecular Devices). Cut o values were calculated
as the average value of negative controls OD and 2 times the SD.
Extraction of inuenza RNA from guniea pig tissue. RNA was extracted from trachea and lung
tissue of infected guinea pig. Briey, 100–250 mg of tissue were homogenized with a tissue homogenizer
and total RNA extracted with RNAeasy Midi Kit (Qiagen) according to the manufacturer’s instructions.
Quantitative real-time PCR for inuenza virus. To detect and quantify inuenza A virus on the
collector plates as well as in guinea pig tissue samples, we used a One-Step Taq Man real-time RT-PCR
assay
46
with primers F1-mxA (150 nM) (5´-AAGACCAATYCTGTCACCTCTGA-3´), F3-mxA (150 nM)
(5´-CAAGACCAATCTTGTCACCTCT GAC-3´) and R1-mxA (900 nM) (5´-TCCTCGCTCACTGGGCA
- 3´) and probes P1-Mx (110 nM) (5´-FAM-TTGTGTTCACGCTCACC–MGB–3´) and P2-Mx (110 nM)
(5´-FAM-TTTGTATTCACGCTCACCG–MGB -3´), with the Rotor-Gene Probe RT-PCR Kit (Qiagen).
e real-time PCR reaction was performed in a Corbett Rotor-Gene 6000 (Qiagen) with the following
cycling protocol: 50 °C for 10 min, followed by 45 cycles of 95 °C for 5 seconds and 57 °C for 15 seconds.
References
1. Lipsitch, M. et al. Transmission dynamics and control of severe acute respiratory syndrome. Science 300, 1966–1970, (2003).
2. Tellier, . Aerosol transmission of inuenza A virus: a review of new studies. J  Soc Interface 6 Suppl 6, S783–S790, (2009).
3. Stilianais, N. I. & Drossinos, Y. Dynamics of infectious disease transmission by inhalable respiratory droplets. J  Soc Interface
7, 1355–1366, (2010).
4. Teunis, P. F., Brienen, N. & retzschmar, M. E. High infectivity and pathogenicity of inuenza A virus via aerosol and droplet
transmission. Epidemics 2, 215–222, (2010).
5. Fowler, . A., Scales, D. C. & Ilan, . Evidence of airborne transmission of SAS. N Engl J Med 351, 609–611; author reply 609-
611, (2004).
6. Fiegel, J., Clare, . & Edwards, D. A. Airborne infectious disease and the suppression of pulmonary bioaerosols. Drug Discov
Today 11, 51–57, (2006).
7. Nordgren, J., Bucardo, F., Svensson, L. & Lindgren, P. E. Novel light-upon-extension real-time PC assay for simultaneous
detection, quantication, and genogrouping of group A rotavirus. J Clin Microbiol 48, 1859–1865, (2010).
8. Nordgren, J., Bucardo, F., Dienus, O., Svensson, L. & Lindgren, P. E. Novel light-upon-extension real-time PC assays for
detection and quantication of genogroup I and II noroviruses in clinical specimens. J Clin Microbiol 46, 164–170, (2008).
9. Macay, I. M., Arden, . E. & Nitsche, A. eal-time PC in virology. Nucleic Acids es 30, 1292–1305, (2002).
10. Booth, T. F. et al. Detection of airborne severe acute respiratory syndrome (SAS) coronavirus and environmental contamination
in SAS outbrea units. J Infect Dis 191, 1472–1477, (2005).
11. Hermann, J. . et al. Optimization of a sampling system for recovery and detection of airborne porcine reproductive and
respiratory syndrome virus and swine inuenza virus. Appl Environ Microbiol 72, 4811–4818, (2006).
12. Pyanov, O. V. et al. Using a bioaerosol personal sampler in combination with real-time PC analysis for rapid detection of
airborne viruses. Environ Microbiol 9, 992–1000, (2007).
13. Verreault, D., Moineau, S. & Duchaine, C. Methods for sampling of airborne viruses. Microbiol Mol Biol ev 72, 413–444, (2008).
14. Turgeon, N., Toulouse, M. J., Martel, B., Moineau, S. & Duchaine, C. Comparison of ve bacteriophages as models for viral
aerosol studies. Appl Environ Microbiol 80, 4242–4250, (2014).
15. Verani, M., Bigazzi, . & Carducci, A. Viral contamination of aerosol and surfaces through toilet use in health care and other
settings. American journal of infection control 42, 758–762, (2014).
16. Branston, G., Gitterman, L., Hirji, Z., Lemieux, C. & Gardam, M. Transmission of inuenza A in human beings. Lancet Infect
Dis 7, 257–265, (2007).
17. Sato, T. [Infection control in hospitals]. Masui 59, 56–63, (2010).
18. Hans, C. Infection control practices for avian inuenza in healthcare facilities. J Indian Med Assoc 104, 388, 390, 392 passim,
(2006).
19. Hatagishi, E. et al. Establishment and clinical applications of a portable system for capturing inuenza viruses released through
coughing. PloS one 9, e103560, (2014).
20. Bergeron, V. et al. Supplemental treatment of air in airborne infection isolation rooms using high-throughput in-room air
decontamination units. American journal of infection control 39, 314–320, (2011).
21. Hudson, J. B., Sharma, M. & Petric, M. Inactivation of Norovirus by ozone gas in conditions relevant to healthcare. e Journal
of hospital infection 66, 40–45, (2007).
22. Mitchell, B. W. & ing, D. J. Eect of negative air ionization on airborne transmission of Newcastle disease virus. Avian diseases
38, 725–732, (1994).
23. Estola, T., Maela, P. & Hovi, T. e eect of air ionization on the air-borne transmission of experimental Newcastle disease virus
infections in chicens. e Journal of hygiene 83, 59–67, (1979).

www.nature.com/scientificreports/
10
SCIENTIFIC RepoRts | 5:11431 | DOI: 10.1038/srep11431
24. Escombe, A. . et al. Upper-room ultraviolet light and negative air ionization to prevent tuberculosis transmission. PLoS medicine
6, e43, (2009).
25. Gast, . ., Mitchell, B. W. & Holt, P. S. Application of negative air ionization for reducing experimental airborne transmission
of Salmonella enteritidis to chics. Poultry science 78, 57–61, (1999).
26. Parvaneh, S. et al. A new method for collecting airborne allergens. Allergy 55, 1148–1154, (2000).
27. Mochizui, M. et al. A calicivirus isolated from a dog with fatal diarrhoea. Vet ec 132, 221–222, (1993).
28. Lee, B. Y. et al. Economic value of norovirus outbrea control measures in healthcare settings. Clinical microbiology and infection:
the ocial publication of the European Society of Clinical Microbiology and Infectious Diseases 17, 640–646, (2011).
29. Vesiari, T., autanen, T., Varis, T., Beards, G. M. & apiian, A. Z. hesus otavirus candidate vaccine. Clinical trial in children
vaccinated between 2 and 5 months of age. Am J Dis Child 144, 285–289, (1990).
30. Istrate, C. et al. Bone marrow dendritic cells internalize live F-81 bovine rotavirus and rotavirus-lie particles (F 2/6-GFP-VLP
and F 8*2/6/7-VLP) but are only activated by live bovine rotavirus. Scand J Immunol 65, 494–502, (2007).
31. Lowen, A. C., Mubarea, S., Tumpey, T. M., Garcia-Sastre, A. & Palese, P. e guinea pig as a transmission model for human
inuenza viruses. Proc Natl Acad Sci U S A 103, 9988–9992, (2006).
32. Pica, N., Chou, Y. Y., Bouvier, N. M. & Palese, P. Transmission of inuenza B viruses in the guinea pig. J Virol 86, 4279–4287,
(2012).
33. Mubarea, S. et al. Transmission of inuenza virus via aerosols and fomites in the guinea pig model. J Infect Dis 199, 858–865,
(2009).
34. Gralton, J., Tovey, E., McLaws, M. L. & awlinson, W. D. e role of particle size in aerosolised pathogen transmission: a review.
J Infect 62, 1–13, (2011).
35. ettleson, E. M. et al. Airborne virus capture and inactivation by an electrostatic particle collector. Environ Sci Technol 43,
5940–5946, (2009).
36. Murray, B. . et al. Virion disruption by ozone-mediated reactive oxygen species. Journal of virological methods 153, 74–77,
(2008).
37. Nishiawa, . & Nojima, H. Airborn virus inactivation technology using cluster ions generated by discharge plasma. Sharp Tech
J 86, 10–15, (2003).
38. Lowen, A. C., Mubarea, S., Steel, J. & Palese, P. Inuenza virus transmission is dependent on relative humidity and temperature.
PLoS Pathog 3, 1470–1476, (2007).
39. Lowen, A. C., Steel, J., Mubarea, S. & Palese, P. High temperature (30 degrees C) blocs aerosol but not contact transmission
of inuenza virus. J Virol 82, 5650–5652, (2008).
40. Lowen, A. C. et al. Blocing interhost transmission of inuenza virus by vaccination in the guinea pig model. J Virol 83,
2803–2818, (2009).
41. Andersson, J. et al. apid detection of cytomegalovirus using immune scanning electron microscopy. Journal of virological
methods 16, 253–262, (1987).
42. Wetterberg, L. et al. Micrometer-sized particles in cerebrospinal uid (CSF) in patients with schizophrenia. Neurosci Lett 329,
91–95, (2002).
43. Johansson, V. et al. Microscopic particles in two fractions of fresh cerebrospinal uid in twins with schizophrenia or bipolar
disorder and in healthy controls. PloS one 7, e45994, (2012).
44. Nordgren, J., indberg, E., Lindgren, P. E., Matusse, A. & Svensson, L. Norovirus gastroenteritis outbrea with a secretor-
independent susceptibility pattern, Sweden. Emerging infectious diseases 16, 81–87, (2010).
45. San Gabriel, M. C. et al. Identication of canine calicivirus capsid protein and its immunoreactivity in western blotting. e
Journal of veterinary medical science / the Japanese Society of Veterinary Science 59, 97–101, (1997).
46. Nilsson, A. C. et al. Longitudinal clearance of seasonal inuenza A viral NA measured by real-time polymerase chain reaction
in patients identied at a hospital emergency department. Scandinavian journal of infectious diseases 42, 679–686, (2010).
Acknowledgement
We acknowledge Dr Peter Palese for providing inuenza virus strain Pan/99 and antisera to this study.
is work was supported by the Swedish Research Council (LS) 320301.
Author Contributions
L.S., H.W., R.N., M.H. and J.N. designed the experiments, R.N. developed the ionizer device, prepared
the set-up of the experimental room and performed the scanning electron microscopy, L.S., M.H., J.N.
and R.N. performed the experiments, K.O.H. performed the transmission electron microscopy, M.H.
and J.N. performed the laboratory analysis, L.S. and H.W. interpreted the data, L.S., H.W., M.H. and J.N.
wrote the manuscript.
Additional Information
Competing nancial interests: e authors declare no competing nancial interests.
How to cite this article: Hagbom, M. et al. Ionizing air aects inuenza virus infectivity and prevents
airborne-transmission. Sci. Rep. 5, 11431; doi: 10.1038/srep11431 (2015).
is work is licensed under a Creative Commons Attribution 4.0 International License. e
images or other third party material in this article are included in the article’s Creative Com-
mons license, unless indicated otherwise in the credit line; if the material is not included under the
Creative Commons license, users will need to obtain permission from the license holder to reproduce
the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/