Assessment of the removal and inactivation of influenza viruses H5N1 and H1N1 by drinking water treatment

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 6

Avai lab le a t www.sc iencedi rec t .com

journa l homepage : www.e lsev ie r . com/ loca te /wat res

Assessment of the removal and inactivation of influenzaviruses H5N1 and H1N1 by drinking water treatment

Dorothee Lenes a,*, Nathalie Deboosere b, Florence Menard-Szczebara a, Jerome Jossent a,Virginie Alexandre b, Claire Machinal a, Michele Vialette b

a Veolia Environment, Research and Innovation, Chemin de la Digue, BP76, 78603 Maisons-Laffitte Cedex, Franceb Institut Pasteur de Lille, Microbiological Safety Unit, 1 rue du Professeur Calmette, 59019 Lille Cedex, France

a r t i c l e i n f o

Article history:

Received 24 September 2009

Received in revised form

14 January 2010

Accepted 16 January 2010

Available online 25 January 2010

Keywords:

Avian influenza viruses

H5N1

H1N1

Drinking water

Physical treatment

Disinfection

* Corresponding author. Tel.: þ33 1 34 93 31E-mail address: [email protected]

0043-1354/$ – see front matter ª 2010 Elsevidoi:10.1016/j.watres.2010.01.013

a b s t r a c t

Since 2003, there has been significant concern about the possibility of an outbreak of avian

influenza virus subtype H5N1. Moreover, in the last few months, a pandemic of a novel swine-

origin influenza A virus, namely A(H1N1), has already caused hundreds of thousands of human

cases of illness and thousands of deaths. As those viruses could possibly contaminate water

resources through wild birds excreta or through sewage, the aim of our work was to find out

whether the treatment processes in use in the drinking water industry are suitable for eradi-

cating them. The effectiveness of physical treatments (coagulation–flocculation–settling,

membrane ultrafiltration and ultraviolet) was assessed on H5N1, and that of disinfectants

(monochloramine, chlorine dioxide, chlorine, and ozone) was established for both the H5N1

and H1N1 viruses.

Natural water samples were spiked with human H5N1/H1N1 viruses. For the coagulation–

settling experiments, raw surface water was treated in jar-test pilots with 3 different coagu-

lating agents (aluminum sulfate, ferric chloride, aluminum polychorosulfate). Membrane

performance was quantified using a hollow-fiber ultrafiltration system. Ultraviolet irradiation

experiments were conducted with a collimated beam that made it possible to assess the

effectiveness of various UV doses (25–60 mJ/cm2). In the case of ozone, 0.5 mg/L and 1 mg/L

residualconcentrationsweretestedwitha contact timeof 10min.Finally, forchlorine,chlorine

dioxide and monochloramine treatments, several residual oxidant target levels were tested

(from 0.3 to 3 mg/L) with contact times of 5–120 min. The infectivity of the H5N1 and H1N1

viruses in water samples was quantified in cell culture using a microtiter endpoint titration.

The impact of coagulation–settling on the H5N1 subtype was quite low and variable. In

contrast, ultrafiltration achieved more than a 3-log reduction (and more than a 4-log

removal in most cases), and UV treatment was readily effective on its inactivation (more

than a 5-log inactivation with a UV dose of 25 mJ/cm2). Of the chemical disinfection

treatments, ozone, chlorine and chlorine dioxide were all very effective in inactivating

H5N1 and H1N1, whereas monochloramine treatment required higher doses and longer

contact times to achieve significant reductions.

Our findings suggest that the water treatment strategies that are currently used for

surface water treatment are entirely suitable for removing and/or inactivating influenza A

viruses. Appropriate preventive actions can be defined for single disinfection treatment

plants.

ª 2010 Elsevier Ltd. All rights reserved.

31; fax: þ33 1 34 93 31 10.m (D. Lenes).er Ltd. All rights reserved.

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 62474

1. Introduction

Influenza viruses are enveloped RNA viruses belonging to the

Orthomyxoviridae family. Wild birds constitute the natural

reservoir of all avian influenza viruses (AIVs) which cause

asymptomatic or low pathogenic infection (Webster et al.,

1992). These viruses replicate in the respiratory, but also in

the gastro-intestinal tract in ducks (Webster et al., 1978). Shed

in large amount in the faeces, they can spread between birds

through the faecal-oral route. Domestic poultry are also

susceptible to influenza.

Influenza viruses can be divided into two groups based on

their clinical manifestations in chicken. Low pathogenic (LP)

AIVs involve milder respiratory diseases, while highly-path-

ogenic (HP) strains, usually H5 or H7, cause severe illness and

high mortality. Due to structural differences at the cleavage

site of the precursor of the viral external glycoproteins hae-

magglutinin (HA), these HPAIV are characterized by a larger

tissue tropism and can replicate in vascular and lymphatic

systems, damaging vital organs and tissues (Alexander, 2000).

Surveillance studies in wild birds in North America and

Europe have revealed a high prevalence of AIV of low viru-

lence for poultry, with higher isolation rate in wild ducks

(Hinshaw et al., 1980; Wallensten et al., 2007). LPAIV H5 and H7

introduced into poultry from wild birds may mutate into HP

viruses after a short or long time of circulation in the poultry

population (Van Reeth, 2007). Apparent insertion or substitu-

tion at their HA cleavage site and the role of the polymerase

and non structural internal genes have been shown for H5

viruses in relation to their virulence (Alexander, 2000; Lycett

et al., 2009; Neumann et al., 2009). Moreover, AIVs are able to

reassort or adapt in an intermediate permissive host, such as

pigs and poultry species to be transmitted to humans (Webby

et al., 2007). Human influenza viruses, replicating in the

respiratory tract, are often associated with both upper and

lower respiratory tract symptoms and can be highly patho-

genic (Van Reeth, 2007).

The HPAIV H5N1 was first reported as the cause of 18

human infections in 1997 in Hong-Kong (Xu et al., 1999). From

2003, it began to spread from Asia to Europe and subsequently

to both North and West Africa in 2005 and 2006 (Neumann

et al., 2009; Chan et al., 2009). The spread of HPAIV H5N1

devastated domestic poultry populations and resulted in the

largest and most lethal H5N1 virus outbreak in humans to

date. H5N1 mainly affects the lower respiratory tract in

human beings: it causes respiratory infection sometimes

leading to acute respiratory distress symptom and also multi-

organ failure, with rapid disease progression and high

lethality rate (Malik Peiris et al., 2009; Neumann et al., 2009). It

has been detected in faecal samples, suggesting replication in

the intestine too (Buchy et al., 2007; Ng and To, 2007).

There is currently considerable worldwide concern over

the possibility of a pandemic outbreak of avian influenza, as

isolated human cases of the HPAIV H5N1 continue to occur in

several Asian countries, and this strain could mutate into

a form that could readily spread between humans (WHO,

2007). Most human cases had a history of very close contact

with infected poultry. Inhalation of infectious droplets is

probably the most common route of infection (Brankston

bodies where waterfowl gather, and, moreover, these viruses

can persist for quite long periods of time in water (on the order

of one or several months), even if it may vary among the AIV

isolates and according to water characteristics such as pH,

salinity and temperature (Stallknecht et al., 1990a,b; Brown

et al., 2007, 2009; Zhang et al., 2007). Potential concentrations

of AIV in surface water were estimated by Schijven et al.

(2005), using a quantitative microbial risk assessment

method: the mean values could vary between 10�5 and 10�2/L

considering different scenarios (volume of water resources).

Nevertheless, the oral ingestion or aspiration of contaminated

water could be a possible mode of human contamination,

even if there is no evidence reported. Surface water resources

(reservoirs, lakes and rivers), and groundwater aquifers under

the influence of surface water could be potential routes by

which the H5N1 virus could enter the drinking water supply.

Moreover, in April 2009, a global outbreak of a novel swine-

origin influenza A H1N1 (S-OIV) virus began in Mexico (Malik

Peiris et al., 2009). This so-called ‘‘swine flu’’ or ‘‘A (H1N1)’’

virus is generated by multiple reassortment events in swine

and humans (Smith et al., 2009) and is transmitted between

human beings. In early June 2009, the World Health Organi-

zation declared the outbreak to be a pandemic, although it

also noted that the virus was globally of ‘‘moderate severity’’.

One quarter to half of patients infected by S-OIV had diarrhea

or vomiting (Dawood et al., 2009; Perez-Padilla et al., 2009;

WHO, 2009). Moreover, studies reported viral shedding in

stool of patients infected with seasonal H1N1 influenza

viruses (Chan et al., 2009). Therefore, one should also consider

the contamination risk of water used for human water supply,

by influenza viruses, via potential contamination of urban

wastewater.

H1N1 and H5N1 subtypes present common structural

properties of enveloped viruses with negative sense, single-

stranded, segmented RNA. Available data, concerning resis-

tance of AIV to different chemical agents and physical

conditions, revealed no differences in relation with strains

subtypes (reviewed by De Benedictis et al., 2007). However

both subtypes have been experimentally studied to compare

mainly cell tropism and genetic sequences possibly involved

in virulence. H5N1 had the most diverse cell tropism and the

highest viral replication efficiency in both cell lines and

allantoic fluid, in comparison with S-OIV (Li et al., 2009).

Moreover distinct differences between HA proteins were

shown, with intrinsic disorder distributions in amino acid

sequences of these proteins observed for virulent strains

(such as H5N1) by comparison with less virulent ones (such as

H1N1 (A/PR/8/34)) (Goh et al., 2009).

The aim of the work reported here was to assess the

performance of different drinking water treatment processes

(coagulation–flocculation–settling, membrane ultrafiltration,

ultraviolet treatment, chemical disinfection) for removing and

inactivating the H5N1 virus, and also to conduct a few inac-

tivation experiments (chemical disinfection) on a seasonal

influenza H1N1, in order to determine the effectiveness of

typical groundwater or surface water treatment plants, and

identify preventive actions that could be used in a context of

pandemic outbreaks.

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 6 2475

2. Material and methods

2.1. Preparation of virus

As the experiments were performed before the recent A

(H1N1) outbreak, a seasonal strain was used. Strain A/PR/8/34

of influenza A virus subtype H1N1 (VR-1469) was obtained

from the American Type Culture Collection (LGC Promochem,

Strasbourg, France). This strain was isolated from a patient

living in Puerto Rico in 1934 and was derived by adaptation of

egg-passage to Madin Darby Canine Kidney cells (MDCK).

HPAIV H5N1 (A/Cambodia/408008/2005 clade 1) and MDCK

cells were obtained from the Pasteur Institutes in Cambodia

(Buchy et al., 2007) and in Paris, respectively. Virus propagation

and titration methods were provided by Pasteur Institute,

Paris. The influenza viruses were propagated in MDCK cells

and titers expressed as median 50% tissue-culture infectious

dose (TCID50/mL). Briefly, maintenance medium (Minimum

Essential Medium [MEM] supplemented with 5% fetal bovine

serum, 8 mM tricine buffered solution [Sigma–Aldrich, St

Louis, USA], 2 mM L-glutamine, 100 U/mL penicillin and 100 mg/

mL streptomycin) was removed from 75-cm2 tissue-culture

flasks containing complete cell monolayers (about 1.5 � 107

cells). The monolayers were rinsed with Dulbecco’s PBS (Invi-

trogen, Carlsbad, CA), and inoculated with 2 mL of serum-free

maintenance medium containing 1.5� 104 TCID50 of influenza

viruses (Multiplicity Of Infection – MOI – of about 10�3). Viruses

were allowed to adsorb to host cells for 1 h at room tempera-

ture before adding 8 mL of viral propagation medium per flask

(the serum-free maintenance medium described above, sup-

plemented with 1 mg/mL of TosylPhenylChloromethylKeton

(TPCK)-treated trypsin [Sigma–Aldrich]). Flasks were incu-

bated at 35 �C under 5% CO2 for 2 or 3 days to allow viral

replication to occur. The infective supernatants were har-

vested and centrifuged at about 2000 � g for 10 min at 4 �C to

remove cell debris. A first stock of H5N1 virus with a final titer

of approximately 108 TCID50/mL was obtained and stored at

�80 �C until use. Further H5N1 and H1N1 supernatants were

concentrated by centrifuging at 10,000� g for 4 h 30 min at 4 �C.

The pellets were suspended in appropriate volumes of Dul-

becco’s Phosphate Buffered Saline, to obtain final titers of

approximately 109 TCID50/mL. Concentrated virus solutions

were stored at �80 �C until use.

ATCC 15597-B1 MS2 bacteriophages were obtained from

the National Collection of the Pasteur Institute of Paris. The

phages were replicated using Escherichia coli (W1485) as

a bacterial host strain. Virus stocks were stored at �80 �C. A

viral suspension of 3 � 1011 PFU/mL in phosphate buffer was

used to spike the environmental matrix.

2.2. Infectivity assays

The infectivity of the influenza viruses was quantified by

using a microtiter endpoint titration.

Samples of viruses-inoculated water were diluted 1:5 for

coagulation–flocculation treatment, and 1:2 for the other

treatments by adding MEM to avoid toxic effects on cells. Serial

ten-fold dilutions (10�1–10�8) were then made in the dilution

medium, which was the previously-described serum-free

maintenance medium. MDCK cells were prepared the day

before titration. Each well of a 96-well microtiter plate received

100 mL of 3� 105 cells/mL suspended in maintenance medium.

One-day-old confluent cells monolayers were rinsed with

dilution medium, and infected with 100 mL of the appropriate

virus dilution. Plates were incubated at room temperature for

1 h. Each well was then supplemented with 100 mL dilution

medium containing 2 mg/mL of TPCK-treated trypsin. Plates

were covered and incubated at 35 �C under 5% CO2 for 3–5 days.

Light microscopy was used to examine the plates for cyto-

pathic effects (CPE). Endpoints were reported as 100% mono-

layer destruction with TCID50 values, calculated by the

Spearman and Karber method (Hamilton et al., 1977) as

described in the European standard NF EN 14476 (2005).

The detection limit of these assays ranged from 6.6 to

1.7 � 101 TCID50/mL, depending on the dilution and volume of

the water tested.

MS2 F-specific RNA phages were quantified by double agar

layer plaque assay method (ISO 10705-1, 1995). The viral

concentration was expressed in plaque forming units per

milliliter (PFU/mL).

2.3. Reagents and analytical procedures

Three coagulants were tested in the clarification tests:

aluminum sulfate (solid Al2(SO4)3,18H2O – VWR), ferric chlo-

ride (41% FeCl3 – Brenntag), WAC-HB (aluminum poly-

chorosulfate, 8.3% Al2O3 – Arkema).

For chlorine disinfection, a commercial solution of bleach

(9.6% – 115 Cl2 g/L) diluted in demineralized water (100 mg Cl2/

L) was used. The residual chlorine was determined using N,N-

diethyl-p-phenylenediamine (DPD) colorimetric method

(Hach Lange kit 8021, USEPA accepted method. The procedure

is adapted from the Standard Methods for the Examination of

Water and Wastewater, 2005, 4500-Cl G for drinking water).

A 100-mg ClO2/L chlorine dioxide solution was produced

using a specific generator (Prominent, Bellozon). Residual levels

were determined using DPD method after adding glycine (Hach

Lange kit 10126, USEPA accepted methods. The procedure is

adapted from the Standard Methods for the Examination of

Water and Wastewater, 2005, 4500-ClO2P for drinking water).

Monochloramine was produced by mixing equal volumes

of a solution of bleach (2.76 g Cl2/L) and a solution of NH4Cl

(2.107 g/L) in demineralized water, thus leading to a Cl2:N ratio

of 5:1; residual levels were measured using the indophenol

method (Hach Lange 10171, U.S. Patent 6,315,950).

Finally, gaseous ozone was produced by a generator (BMT

802X) within O2 extracted from ambient air using an oxygen

generator (Platinum S, Invacare). For security reasons, ozone

production was carried out outside the L3 laboratory. The

ozone was then analyzed and directly placed in a hermetic gas

pocket (Tedlar, Arelko). This gas pocket was then rapidly

transferred into the L3 laboratory. The residual ozone was

determined using the indigo method (Hach Lange 8311. The

procedure is adapted from the Standard Method for the

Examination of Water and Wastewater, 2005, 4500-O3).

Sodium thiosulfate (0.1 N – Hach) was used to quench

ozone, chlorine, chlorine dioxide and monochloramine

(Takizawa et al., 1973; White, 1999; Wang et al., 2006) after

scheduled contact times.

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 62476

2.4. Experimental natural water

All disinfection experiments were performed using water

from a surface water treatment plant operated by Veolia

Water in France. Membrane ultrafiltration (UF), ultraviolet

treatment (UV), and chemical disinfection experiments were

carried out on ozonated water (after clarification and ozona-

tion treatment), whereas raw water was used for coagulation

tests.

Characteristics of the different types of water are summa-

rized in Table 1.

2.5. Experimental procedures

Experiments with HPAIV H5N1 and with subtype H1N1 were

conducted at Pasteur Institute, Lille, in biosafety level (BSL) 3

and BSL2 facilities respectively. Experiments with MS2 bacte-

riophages were carried out at Veolia Environment laboratory.

All experiments were implemented at ambient tempera-

ture (22 �C � 2 �C). Influenza virus suspensions were directly

diluted in the water being studied to obtain an initial concen-

tration of 105–106 TCID50/mL that would make it possible to

detect 2–4-log inactivation in water. The MS2 suspension was

pre-diluted in sterile phosphate buffer (PB, 0.1 mM, pH 7.2).

Virus counts were performed just after the test treatments.

2.5.1. Bench scale system for the coagulation–flocculation–settling testCoagulation–flocculation and settling (also called clarification)

are the first vital chemical treatment stages in most of the

surface water treatment plants. Their aim is to remove dis-

solved solids. Common coagulants are iron or aluminum salts;

the applied doses depend on water characteristics (essentially

turbidity and/or organic carbon content) and may vary

between a few to dozens of ppm.

The coagulation–flocculation and settling experiments

were carried out in a jar-test apparatus (Minimix laboratory

mixer, ECEngineering) with four 500-mL jars.

For each experiment, 1.7 L of raw water were spiked with

H5N1 virus in order to obtain an initial concentration of about

105 TCID50/mL. 30 ppm of commercial coagulant (ferric chlo-

ride, aluminum sulfate, WAC-HB) was added to 400 mL of

Table 1 – Characteristics of the experimental natural water.

Parameters Raw water 1 characteristics(1st and 2nd jar-tests)a

Raw(3

Conductivity (mS/cm at 25 �C) 623

pH 7.8

Total alkalinity (F�) 27.5

Total hardness (F�) 31.4

Turbidity (FNU) 6

Total organic

carbon (mg/L)

1.7

Manganese (mg/L) 0.029

Iron (mg/L) 0.16

Ammonia (mg/L) 0.28

a Raw water was used for the clarification experiments.

b Clarified ozonated water was used for the ultrafiltration and disinfecti

spiked water, and then stirred during 3 min at 160 rpm. The

water was then flocculated for 15 min at 45 rpm. After floccu-

lation, settling occurred (15 min). Finally, samples were taken

from the supernatant for pH and turbidity measurement, virus

counting and assessment of the log removal. A control jar

containing raw water spiked with virus, but without coagulant,

was also tested under the same conditions.

2.5.2. Bench scale pilot for membrane ultrafiltrationMembrane ultrafiltration is a high-performance treatment

process that may be used either in groundwater or surface

water treatment plants, after clarification processes or (for

high-quality water) instead of clarification processes.

Membrane ultrafiltration (UF) performance was measured

using 6 bench scale experimental modules prepared from

hollow-fiber membranes: 3 modules for the MS2 prior exper-

iments and 3 modules for the H5N1 experiments. A protocol

was first established to prepare the membranes in order to

obtain a stable permeability value. Each bench scale module

(performed in triplicate for each virus) was tested prior to the

experiments in order to check its integrity (Machinal et al.,

2006). The membranes consisted of internal filtration

hollow-fibers of polyethersulfone with a molecular weight

cut-off of 150 kDa. Each module was composed of 6 fibers,

each with a filtration surface of 50 cm2.

In previous studies, optimal operating conditions had been

determined (Langlet et al., 2008, 2009) that would limit viral

aggregation and adhesion to the membrane, and quantify MS2

virus removal by membrane filtration processes without any

risk of overestimation. For preliminary MS2 experiments,

a volume of viral suspension (up to 106 PFU/mL) was prepared

in a 0.1 mM sterile PB solution at neutral pH; the aim was to

characterize the membranes. For the H5N1 experiments,

natural water was spiked with virus in order to obtain an initial

concentration of about 105 TCID50/mL in order to assess the

total removal of H5N1 virus under more realistic conditions.

A total volume of 2.7 L of water was spiked with each virus.

In each experiment, 3 modules were successively placed in the

filtration apparatus, and filtration was carried out under

a pressure of 0.3–0.4 bar in frontal mode and for a variable time,

depending on the permeability of the membranes. Constant

volumes of suspension were filtered per membrane, with a flow

water 2 characteristicsrd and 4th jar-tests)a

Clarified/ozonated watercharacteristics (min–max values)b

611 576–660

7.85 7.45–7.9

23.6 22.7–26.1

29.4 27.6–30.8

15 <0.2–0.48

2.5 0.6–1.7

0.015 <0.005–<0.02

<0.02 <0.005–0.02

0.20 <0.05–<0.1

on (UV, chlorine, chlorine dioxide, chloramine) experiments.

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 6 2477

rate of 14 mL/min of permeate. 20 mL samples of water were

taken at the inlet and at the outlet (permeate) of the ultrafil-

tration apparatus, at different filtration times (beginning,

middle and end of filtration cycle, corresponding to 140, 260 and

380 mL of filtered water respectively), and were analyzed.

2.5.3. Ozone treatmentOzone is a powerful oxidant that is commonly used in surface

water treatment plants, either for oxidation or disinfection

purposes.

Due to the high oxidant consumption of the virus

production medium, concentrated H5N1 and H1N1 solutions

were used for the ozone experiments so that a higher dilution

rate could be used. 70 mL of spiked water (containing initial

virus concentrations of about 105 and 106 TCID50/mL for H5N1

and H1N1, respectively) were placed in a 100 mL batch reactor

consisting of a glass funnel with a septum. Ozone gas

(60–140 g/Nm3) was injected with a syringe directly into the

water phase. The target residual concentrations of ozone were

0.5 mg/L and 1 mg/L (which are representative of typical

applied residuals) after 10 min of contact time (in real ozone

reactors, contact times may vary between a few to 10 min).

The doses required were determined prior to the experiment

on the spiked water.

The system was then shaken for 4 min on a shaker (Ika, VX8),

in order to ensure the transfer of O3 into the water phase. After

another 6 min without shaking (total contact time ¼ 10 min),

samples were taken for virus counting (after adding sodium

thiosulfate) and measurement of the residual ozone.

One control reactor was included in each experiment,

consisting of spiked water to which no ozone had been added.

2.5.4. Ultraviolet radiationUV treatment can be considered as a high-performance disin-

fection process that may be applied either on groundwater or

surface water before the final chlorine-based disinfection.

The effect of ultraviolet (UV) treatment was investigated at

bench scale using a 254 nm collimated beam apparatus. This

consisted of one 8-W mercury low vapour pressure lamp

(Calgon Carbon) emitting light mainly at 253.7 nm. The inci-

dent light intensity was measured with a radiometer IL1400A

(International Light, Newburyport, US).

25 mL of spiked water (containing an initial concentration

of H5N1 of about 106 TCID50/mL) were poured into a Petridish

(90 mm), and placed on a magnetic stirrer under the colli-

mated beam. The depth of the sample was 0.44 cm. UV

transmission through the spiked water was measured (97,5%),

and the radiation times required to obtain 25, 40 and 60 mJ/

cm2 (typical encountered doses) were calculated using

a program developed by Bolton (519-741-6283, 2002) which

takes into account all the parameters impacting the irradia-

tion of water samples. A non-irradiated control sample was

always included.

2.5.5. Chlorine, chlorine dioxide and monochloraminetreatmentChlorine-based disinfectants are applied in most plants (either

groundwater or surface water) for disinfection purposes and

also to maintain a residual disinfectant in the distribution

system.

The initial disinfectant doses were chosen to achieve

oxidant residuals representative of typical drinking water

treatments. The target residual levels were 0.3, 1 and 1.5 mg/L

for chlorine and chlorine dioxide, and 0.3, 1, 2 and 3 mg Cl2/L

for monochloramine, after a contact time of 5 min.

Preliminary tests were conducted on water to which the

virus production medium was added. Residual oxidant levels

were then monitored over 2 h. Moreover, the residual oxidant

was monitored over 5 min in 10 mL of spiked water before

each experiment.

120 mL of spiked water (with an initial virus concentra-

tion of about 105 TCID50/mL for H5N1 and H1N1) were

disposed in hermetic plastic ware covered with aluminum.

Oxidant was added while stirring gently (magnetic stirrer –

IKA, RO5). Samples were taken after predetermined contact

times (5 min, 15 min, 30 min, 60 min and 120 min) for

measurement of the residual oxidant and virus counting

(after adding sodium thiosulfate). Ct values (the concentra-

tion of disinfectant, C [mg/L], multiplied by the contact time,

t [min]) were used to express the effectiveness of the various

disinfectants.

One control reactor was included in each experiment con-

sisting of spiked water to which no oxidant had been added.

Due to the high oxidant consumption rate, complementary

experiments were conducted with concentrated H5N1 and

H1N1 for the chlorine experiments.

3. Results and discussion

3.1. Impact of coagulation–flocculation–settling onH5N1

Two jar-test experiments, comparing the 3 different coagu-

lants (aluminum sulfate Al2(SO4)3, ferric chloride FeCl3,

aluminum polychorosulfate WAC-HB), were carried out in

duplicate on 2 types of water, taken from different rivers (their

characteristics are summarized in Table 1).

Fig. 1 shows the results of the first jar-test experiment.

Viral titers were determined in water samples before (control)

and after clarification with the 3 reagents.

The H5N1 concentrations obtained in raw water before

clarification were 104 and 103 TCID50/mL in the 1st and 2nd jar-

tests, respectively, when the target virus concentration in

spiked raw water was 105 TCID50/mL. An initial loss of about

1–2-log of the H5N1 virus was therefore observed just after

spiking the raw water. 3 hypotheses can be drawn: viral

adsorption on storage recipient, as described by Butot et al.

(2007), aggregation, or inactivation. Since no decrease of

viral RNA quantities was observed in the aqueous phase by

real-time RT-PCR (data not shown), the phenomenon of viral

adsorption was therefore excluded. Either loss of infectivity or

viral particles aggregation could result in the reduction of the

number of infectious particles. Some studies concerning

aggregation and surface properties of bacteriophages, which

are non-enveloped viruses, showed that an increase in the

size of the particles occurred when pH becomes lower than

the virus isoelectric point (pI) (Langlet et al., 2007, 2008). A few

studies have been conducted to determine the pI of influenza

A viruses on H1N1 strains and have reported values of about 5

1.58E+03

3.75E+02

5.00E+01

8.90E+02

2.81E+02

2.81E+01

2.81E+02

6.65E+02

1.19E+04

1.19E+03

1,00E+01

1,00E+02

1,00E+03

1,00E+04

1,00E+05

)tnemirepxefodne(lortnoC3)4OS(2lA3lCeFBHCAW

H5

N1

(T

CID

50

/m

L)

Jar-test 1Jar-test 2Control 1Control 2

Fig. 1 – Impact of coagulation–flocculation–settling on H5N1 spiked in raw water – Results of the first jar-test experiment (1st

and 2nd replicates).

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 62478

(Miller et al., 1944; Oster, 1951; Zhirnov, 1991). Even if avian

influenza viruses are enveloped viruses, these data suggest

that viral particles were mainly suspended (and not aggre-

gated) in the experimental raw water which was character-

ized by a pH value higher than 7 (Table 1). Thus the initial

decrease of infectious particles number would be due to viral

inactivation, probably resulting from a virucidal effect of the

raw water on H5N1.

Furthermore, at the end of the experiments, an additional

1-log infectious titer decrease was observed in the control

water sample, while no change in the viral RNA quantities was

observed (data not shown), suggesting once again that the raw

water itself may have had a virucidal effect on H5N1.

5.00E+01

< 1.7E+011.83E+01

3.49E+01

1.58E+03

3.75E+03

1,00E+01

1,00E+02

1,00E+03

1,00E+04

1,00E+05

3lCeFBHCAW

H5

N1

(T

CID

50

/m

L)

Fig. 2 – Impact of coagulation–flocculation–settling on H5N1 spik

(3rd and 4th replicates).

Compared to the control sampled at the end of the exper-

iment, adding a coagulation agent to the water did not display

any great removal, except for WAC-HB (where a decrease of

1–1.5-log was observed).

Fig. 2 shows the results of the second jar-test experiment. As

in the first jar-test experiment, an initial inactivation of about

2-log was observed in the raw water, with a further 0.7–1-log

inactivation at the end of the experiments, without adding any

coagulating agent. However, adding the coagulant resulted in

the removal of an additional 1-log (0.9–1.3-log) of viruses.

According to WHO (2006), the removal of viruses by coagu-

lation–flocculation–settling process is quite low and varies

between 30% and 70%, depending on the coagulant, the pH,

1.70E+01

3.75E+02

1.96E+01

3.75E+02

)tnemirepxefodne(lortnoC3)4OS(2lA

Jar-test 3Jar-test 4Control 3Control 4

ed in raw water – Results of the second jar-test experiment

0,00

0,50

1,00

1,50

2,00

2,50

3,00

3,50

4,00

4,50

5,00

Beginning Middle End

Filtration time

Lo

g rem

oval o

f H

5N

1 (lo

g10 T

CID

50/m

L)

UF module 1UF module 2UF module 3

Arrows mean "> 4.47 log removal"

Fig. 3 – Log removal of H5N1 spiked in treated water by ultrafiltration.

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 6 2479

temperature, alkalinity and turbidity. Several authors

(LeChevallier and Au, 2004; Hijnen and Medema, 2007) have

highlighted the great variability of virus removal by clarifica-

tion. Studies report log removals from 0.15-log to more than 4-

log. In our experiments, if the inactivation of H5N1 due to the

raw water itself is considered separately, the effect of coagu-

lation–flocculation and settling was low: it varied between

0 and 1.5-log, and was more significant in the 2nd experiment,

when raw water was characterized with a higher turbidity

(about 15 FNU comparing to 6 FNU in the 1st experiment). This

is quite consistent with the literature (Rao et al., 1988). It must

be noted that the experimental initial pH was about 7.8, which

is relatively high and do not facilitate removal of viruses by

coagulation (Harrington et al., 2001; Hijnen and Medema, 2007).

3.2. Removal of H5N1 by ultrafiltration

The results of H5N1 removal by the 3 UF modules are repre-

sented in Fig. 3.

The turbidity of water was 0.25 FNU for these experiments.

Most of titer results in the UF permeate were below the

detection limit of 6.80 TCID50/mL.

More than a 4.4-log removal of influenza viruses was

observed in all the experiments and for all the different

filtration times, except for the last module for which only

a 3.3-log removal was found at the 30-min filtration time

Table 2 – Comparison of the effectiveness of UF membranes in

Mean log removal (3 modules) [min. value;max. value]

Beginning (w

MS2 phages NA

H5N1 >4.46 [4.44

NA ¼ not available.

sampling. Nevertheless, the results led to the conclusion that

UF filtration is a very effective process against H5N1.

Experiments were also conducted on the MS2 phage in

order to compare the H5N1 results with those for the removal

of a model virus. MS2 is commonly used as an indicator of

virus removal by membrane filtration because it is similar in

size (25 nm), shape, and nucleic acid make up to poliovirus

and hepatitis A, which are small pathogenic enteric viruses

(Valegard et al., 1990). With its low size and low pI, it repre-

sents a worst-case for removal in membrane filtration

processes (Langlet et al., 2009). Avian influenza viruses are

characterized by a larger particle size (80–120 nm). Table 2

shows that the removal of the H5N1 virus by UF modules

was at least as high as that of the MS2 phage, confirming that

MS2 can be a conservative surrogate for virus removal by

ultrafiltration, not only for enteric viruses such as poliovirus,

hepatitis A or calicivirus (Jacangelo et al., 2006), but also for

this particular highly-pathogenic virus.

3.3. Inactivation of H5N1 and H1N1 by ozone

Enveloped viruses are recognised as being highly sensitive to

disinfectants because they require an intact lipid envelope to

attach to and infect host cells, and this envelope can be

damaged by chemical agents (Rice et al., 2007). Therefore water

disinfection processes such as ozone or chlorine-based

removing H5N1 and MS2 spiked in treated water.

Filtration time

10 min) Middle (w20 min) End (w30 min)

3.85 [3.44; 4.39] 4.11 [3.98; 4.21]

; >4.47] >4.44 [4.41; >4.47] >4.1 [3.37; >4.47]

Table 3 – Impact of low pressure UV on the H5N1 virusand the MS2 bacteriophage spiked in treated water.

Fluence (mJ cm�2) 25 40 60

Log inactivation

H5N1 (C0 ¼ 6.42 log/mL) >5.50 >5.50 >5.50

MS2 (C0 ¼ 5.01 log/mL) 1.87 2.88 3.65

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 62480

disinfection, which are designed to inactivate more resistant

enteric viruses, are expected to be highly effective on influenza

viruses.

Ozone experiments were carried out using concentrated

viruses, because of a high ozone consumption obtained with

an initial non-concentrated viral inoculum. The results are

shown in Fig. 4.

The residual levels of ozone obtained were 0.4 and 1.2 mg/L

for H5N1, 0.5 and 1.1 mg/L for H1N1 respectively, after a 10-min

contact time. The figure shows that even with residual ozone

levels of 0.4–0.5 mg/L, a contact time of 10 min is completely

adequate to achieve at least a 4-log inactivation of H5N1 and

H1N1 (concentrations below the detection limit of 6.6 TCID50/

mL). Unfortunately, lower contact times were not tested.

Those experiments highlighted the effectiveness of this

disinfectant against influenza viruses, as against many other

viruses. Indeed, USEPA (1999) defined the Ct required to obtain

2, 3 and 4-log reduction of viruses, based on poliovirus (which

is one of the most resistant viruses). At 20 �C, a 4-log reduction

is achieved with a Ct of less than 1 mg/L.min.

The experiments conducted by Murray et al. (2008) on

various type of viruses, including H1N1 (A/PR/8/34), confirmed

that enveloped viruses show extreme sensitivity to ozone.

They suggested that ozone exposure reduce viral infectivity by

lipid peroxidation and subsequent lipid envelope and protein

capsid damage.

3.4. Inactivation of H5N1 by ultraviolet

On the contrary to ozone that inactivates microorganisms

primarily by damaging cellular structures, UV radiation inac-

tivates microorganisms by damaging their nucleic acid,

thereby preventing them from replicating. It induces damages

mainly in the pyrimidines nucleotides (Jagger, 1967).

The 3 different UV fluences (25, 40 and 60 mJ/cm2) were

tested in duplicate in treated water characterized by a UV

6.31E+04

8.51E+05

6.6<

1,00E+00

1,00E+01

1,00E+02

1,00E+03

1,00E+04

1,00E+05

1,00E+06

Control t0 0.5 mg O3/L - 10 min

H5

N1

a

nd

H

1N

1 (T

CID

50

/m

L)

H5N1H1N1O3 residual (H5N1 experiO3 residual (H1N1 experi

Arrows mean "

Fig. 4 – Inactivation of H5N1 and H1N1s

transmission of 97.5%. A comparison with MS2 is shown in

Table 3. Concentrations of H5N1 at the end of the UV experi-

ments were always below the threshold of detection

(<10 TCID50/mL), indicating that UV radiation had very effec-

tively inactivated H5N1. More than a 5.5-log inactivation was

achieved with the lowest UV fluence tested (25 mJ cm2), which

corresponds to the minimum dose that is currently used in

drinking water treatment plants equipped with this disinfec-

tion process. H5N1 seems to be very sensitive to UV radiation,

in contrast to MS2, for which the log inactivations were found

to be 1.9, 2.9 and 3.6 respectively for the 3 UV fluences tested.

MS2 inactivation results are completely consistent with liter-

ature data (Oppenheimer et al., 1993; Wiedenmann et al., 1993;

Meng and Gerba, 1996; Sommer et al., 1998) indicating that

MS2 is quite resistant to UV. Lucio-Forster et al. (2006) con-

ducted experimental UV radiation assays, but not directly on

H5N1. They chose H5N2, a low pathogenicity virus, as

a surrogate. They also concluded that the sensitivity of this

avian influenza virus to low pressure UV radiation was high

(a 2.7-log inactivation was observed at 10 mJ/cm2 in PB) sug-

gesting that H5N1 would likely be inactivated in water at

typical UV fluences. This was confirmed by our experiments.

3.5. Inactivation of H5N1 and H1N1 bymonochloramine

Four doses of monochloramine were tested for target residuals

after 5 min of contact times. The results for monochloramine

1.51E+05

4.79E+05

6.6<

1 mg O3/L - 10 min Control (end of experiment)0

0,2

0,4

0,6

0,8

1

1,2

1,4

ments)ment)

< detection limit"

piked in treated water with ozone.

0,0

0,5

1,0

1,5

2,0

2,5

3,0

3,5

4,0

0 20 40 60 80 100 120 140

Contact time (min)

H2

NC

l re

sid

ua

ls

(m

g/L

C

l2

)

Target monochloramine residual 0.3 mg Cl2/LTarget monochloramine residual 1 mg Cl2/LTarget monochloramine residual 2 mg Cl2/LTarget monochloramine residual 3 mg Cl2/L

Fig. 5 – Monochloramine residuals in treated water during H5N1 experiments.

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 6 2481

residuals are shown in Fig. 5. The water oxidant demand was

quite low. Fig. 6 shows the inactivation of H5N1 viruses vs.

monochloramine dose for different contact times. 0.3 mg/L of

residual monochloramine is not sufficient to achieve a good

inactivation of H5N1: even after 120 min of contact, the inac-

tivation of H5N1 was only about 1 log. Nevertheless, with doses

that are currently used in water treatment plants – 1, 2 and

3 mg/L – and contact times of 60 min, 30 min and 15 min

respectively, more than a 4-log inactivation of the H5N1 virus

can be achieved in drinking water. A Ct of about 60 mg/L.min is

therefore suitable to obtain more than a 4-log inactivation,

which allows to say that H5N1 is readily inactivated by mon-

ochloramine. In contrast, according to USEPA (1999), a Ct

higher than 700 mg/L.min is required at 20 �C in order to obtain

1,0E+0

1,0E+1

1,0E+2

1,0E+3

1,0E+4

1,0E+5

1,0E+6

0 20 40 60

Contac

H5

N1

(T

CID

50

/m

L)

Fig. 6 – H5N1 inactivation w

a 4-log reduction of Hepatitis A virus which can be considered

as relatively resistant to chlorine-based disinfectants

(Peterson et al., 1983; Sobsey et al., 1988; Li et al., 2004).

A few experiments were also conducted with H1N1 plus 1

and 2 mg/L of monochloramine. The results that are shown in

Fig. 7 highlight the good effectiveness of this disinfectant

versus the H1N1 virus. Indeed, at the lower residual tested

(1 mg/L), a contact time of 15 min was sufficient to achieve

more than a 4-log inactivation of the H1N1 virus. This implies

that H1N1 is not a conservative surrogate of the inactivation of

H5N1 with monochloramine.

The viral inactivation mechanism by chloramine is still not

very clear and may be dependent on virus type and disinfec-

tion dose (USEPA, 1999), since some studies demonstrated an

80 100 120 140

t time (min)

Target monochloramine residual 0.3 mg Cl2/LTarget monochloramine residual 1 mg Cl2/LTarget monochloramine residual 2 mg Cl2/LTarget monochloramine residual 3 mg Cl2/L

Detection limit

ith monochloramine.

1,0E+00

1,0E+01

1,0E+02

1,0E+03

1,0E+04

1,0E+05

1,0E+06

0 20 40 60 80 100 120 140

Contact time (min)

H1

N1

(T

CID

50

/m

L)

0

0,5

1

1,5

2

2,5

H2N

Cl resid

ual (m

g/L

C

l2)

H1N1 - Target monchloramine residual 1 mg/L (run A)H1N1 - Target monochloramine residual 1 mg/L (run B)H1N1 - Target monochloramine residual 2 mg/LH2NCl residual - Target residual 1 mg/L (run A)H2NCl residual - Target residual 1 mg/L (run B)H2NCl residual - Target residual 2 mg/L

Detection limit

Fig. 7 – H1N1 inactivation with monochloramine.

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 62482

action on RNA and other studies suggest mechanisms

involving protein coat.

3.6. Inactivation of H5N1 and H1N1 by chlorine dioxide

Chlorine dioxide experiments were performed with residual

levels of chlorine dioxide of 0.3, 1 and 1.5 mg/L (after a contact

time of 5 min) and initial concentrations of 104–105 TCID50/mL

of H5N1. In all the experiments, ClO2 residuals remained quite

stable, and the inactivation enabled to obtain undetectable

concentrations of H5N1 virus, even for the shortest contact

time. Therefore, more than a 4-log inactivation of the H5N1

virus was obtained with concentrations of chlorine dioxide

0,0

0,5

1,0

1,5

2,0

2,5

3,0

3,5

4,0

4,5

0 20 40 60

Contact t

Ch

lo

rin

e re

sid

ua

l (m

g/L

C

l2

)

Fig. 8 – Chlorine residuals in treated water during H5N1 ex

from 0.3 mg/L during 5 min, meaning that the inactivation of

avian influenza virus with chlorine dioxide is very successful.

Similar results were obtained with the H1N1 virus: a single

experiment was conducted with the same contact times and

a chlorine dioxide residual of 0.5 mg/L (data not shown). The

log inactivation achieved was greater than 4.5 after only 5 min.

No H1N1 virus was detected in the disinfected samples.

H5N1 and H1N1 influenza viruses are effectively inacti-

vated by chlorine dioxide, and show less resistance than other

types of viruses. The USEPA (1999) defined Ct inactivation

curves on the basis of hepatitis A virus inactivation: for a 4-log

inactivation, it was assumed that a Ct of about 10–15 mg/L.min

was required at 20 �C, which is much more important than the

80 100 120 140

ime (min)

Target chlorine residual 0.3 mg/LTarget chlorine residual 1 mg/LTarget chlorine residual 1.5 mg/LTarget chlorine residual 0.3 mg/L - Concentrated H5N1Target chlorine residual 0.5 mg/L - Concentrated H5N1Target chlorine residual 1 mg/L - Concentrated H5N1Target chlorine residual 1.5 mg/L - Concentrated H5N1

periments (with or without a viral concentration step).

1,0E+0

1,0E+1

1,0E+2

1,0E+3

1,0E+4

1,0E+5

1,0E+6

035202510150

Contact time (min)

H5

N1

(T

CID

50

/m

L)

Target chlorine residual 0.3 mg/L

Target chlorine residual 1 mg/L

Target chlorine residual 1.5 mg/L

Target chlorine residual 0.3 mg/L - Concentrated H5N1

Target chlorine residual 0.5 mg/L - Concentrated H5N1

Target chlorine residual 1 mg/L - Concentrated H5N1

Target chlorine residual 1.5 mg/L - Concentrated H5N1

Detection limit

Fig. 9 – H5N1 inactivation with chlorine (with or without a viral concentration step).

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 6 2483

Ct of 2.5 mg/L min that we found for a 4-log inactivation of

H5N1 or H1N1.

Concerning the inactivation mechanisms of chlorine

dioxide, the primary mode of action on viruses is still not well

understood: chlorine dioxide could react either with periph-

eral structures or nucleic acids (Li et al., 2004; USEPA, 1999)

depending particularly on virus type.

3.7. Inactivation of H5N1 and H1N1 by chlorine

As in the ozone experiments, the viral inoculum displayed

a high initial chlorine demand, which led to rapid decrease of

chlorine residuals as shown in Fig. 8.

In order to obtain a chlorine residual of about 0.3 mg/L after

a contact time of 5 min, an initial dose of 2.3 mg/L was

1,0E+0

1,0E+1

1,0E+2

1,0E+3

1,0E+4

1,0E+5

1,0E+6

0 20 40 60

Contact tim

H1

N1

(T

CID

50

/m

L)

H1

Fig. 10 – H1N1 inactivation with chlorin

required. Due to the high oxidant consumption, complemen-

tary experiments were conducted using concentrated H5N1

and H1N1.

H5N1 inactivation occurred very rapidly with the residual

levels tested. Fig. 9 shows that even after a contact time of

5 min, the inactivation of an initial population of about

105 TCID50/mL exceeded 4-log, at a pH of 7.45.

Nevertheless, the concentration protocol of H5N1 was then

applied for a second set of experiments in order to allow

a higher dilution of the virus stock. This dilution made it

possible to limit the oxidant demand, and to obtain stable

residual levels (see Fig. 8). The pH of the experiment was 8.

Unfortunately, an inactivation of only 2.8-log could be

demonstrated in these experiments due to a low concentra-

tion of H5N1 in the stock (the initial titer of the spiked water

80 100 120 140

e (min)

0,00

0,50

1,00

1,50

2,00

2,50

H2N

Cl resid

ual (m

g/L

C

l2)

N1 Chlorine residual

Detection limit

e (after a viral concentration step).

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 62484

did not exceed 103.8 TCID50/mL). However, the results showed

that H5N1 is readily inactivated by chlorine (2.8-log inactiva-

tion with 0.3 mg/L residual chlorine after a contact time of

5 min).

In the case of H1N1, the experiment conducted with

a chlorine residual of about 1.7 mg/L, and a pH of 7.8

demonstrated a considerable inactivation of the virus: more

than a 4-log inactivation was achieved after a contact time of

5 min (Fig. 10).

These experiments highlight the effectiveness of chlorine

against the H5N1 and H1N1 viruses. This is consistent with the

results reported by Rice et al. (2007), showing that more than

a 3-log reduction of the H5N1 virus can be obtained with

a chlorine dose of 2 mg/L (leading to a residual of 0.5–1 mg/L,

due to the high consumption of the oxidant by the inoculum)

and a contact time of 1 min at pH 7 or pH 8. Lucio-Forster et al.

(2006) also demonstrated the effectiveness of chlorine on

another avian influenza virus (H5N2): a Ct of 8 mg/L.min

enabled to completely inactivate the virus (more than a 3.7-log

inactivation).

As for chlorine dioxide and monochloramine, research has

shown that chlorine is capable of damaging either the

external structures of viruses (proteins or lipids) or the nucleic

acid (USEPA, 1999; Li et al., 2002). It may depend on virus type

and chlorine concentrations.

4. Conclusions

This study revealed the impact of water treatment processes

on the removal and inactivation of the H5N1 virus. Assuming

that the impact of a multi-barrier treatment plant can be

approximated by summing the log reduction of the different

treatment processes, our data show that treatment plants

designed to supply drinking water from surface water

resources, which are generally composed of a coagulation–

flocculation–settling process, a 1st stage filtration (sand

filtration for example), an ozonation process, a 2nd stage

filtration (granular activated carbon filtration for example),

and a final disinfection, would readily be effective in removing

or inactivating H5N1. Although the clarification process

(coagulation–flocculation–settling) does not significantly

reduce H5N1 concentrations in water, a combination of

ozonation and final disinfection processes (chlorine, chlorine

dioxide or monochloramine) could allow at least a 7-log

inactivation, provided that the required oxidant residuals and

contact times are applied. Effective treatment conditions can

therefore be defined, and are quite similar to those currently

used on site. In high-performance treatment plants that use

membrane ultrafiltration and/or UV irradiation, the inactiva-

tion will be considerably enhanced as these processes enable

to achieve a further 4-log reduction of H5N1.

Nevertheless, preventive actions can be recommended for

water treatment sites supplied with groundwater that only

use chlorine or monochloramine disinfection, if a pandemic

situation arises. If the resource is potentially exposed to the

influence of surface water, an increase in disinfectant dose

levels can be recommended. Mobile UV-treatment stations

could also be proposed to ensure secure drinking water

distribution systems.

As for the H1N1 virus, current data does not enable to

conclude if the experimental strain is completely represen-

tative of the novel pandemic one. However it would seem

logical that these strains would not significantly differ in their

susceptibility to disinfectants. The few disinfection experi-

ments conducted are very optimistic and showed that ozone,

chlorine dioxide, chlorine and monochloramine are

completely suitable for inactivating this virus.

Further research is currently in progress, using virus

models, in order to elucidate the impact of virus characteris-

tics on their adsorption properties and on the effectiveness of

physical processes (coagulation–flocculation–settling, sand

filtration, microfiltration and ultrafiltration).

Acknowledgements

We would like to thank Dr. Philippe Buchy (Pasteur Institute,

Cambodia) for providing the virus H5N1 A/Cambodia/408008/

2005.

r e f e r e n c e s

Alexander, D., 2000. A review of avian influenza in different birdspecies. Veterinary Microbiology 74 (1–2), 3–13.

Brankston, G., Gitterman, L., Hirji, Z., Lemieux, C., Gardam, M.,2007. Transmission of influenza A in human beings. LancetInfectious Diseases 7, 257–265.

Brown, J.D., Swayne, D.E., Cooper, R.J., Burns, R.E., Stallknecht, D.E.,2007. Persistence of H5 and H7 avian influenza viruses in water.Avian Diseases 51, 285–289.

Brown, J.D., Goekjian, G., Poulson, R., Valeika, S., Stallknecht, D.E.,2009. Avian influenza virus in water: infectivity is dependent onpH, salinityand temperature.VeterinaryMicrobiology136,20–26.

Buchy, P., Mardy, S., Vong, S., Toyoda, T., Aubin, J., Miller, M.,Touch, S., Sovann, L., Dufourcq, J., Richner, B., Tu, P., Tien, N.,Lim, W., Peiris, J., Van der Werf, S., 2007. Influenza A/H5N1virus infection in humans in Cambodia. Journal of ClinicalVirology 39 (3), 164–168.

Butot, S., Putallaz, T., Croquet, C., Lamothe, G., Meyer, R.,Joosten, H., Sanchez, G., 2007. Attachment of enteric viruses tobottles. Applied Environmental Microbiology 73 (16),5104–5110.

Chan, M., Lee, N., Chan, P., Leung, T., Sung, J., 2009. Fecaldetection of influenza A virus in patients with concurrentrespiratory and gastrointestinal symptoms. Journal of ClinicalVirology 45 (3), 208–211.

Dawood, F.S., Jain, S., Finelli, L., Shaw, M.W., Lindstrom, S.,Garten, R.J., Gubareva, L.V., Xu, X., Bridges, C.B., Uyeki, T.M.,2009. Emergence of a novel swine-origin influenza A (H1N1)virus in humans. The New England Journal of Medicine 361.

De Benedictis, P., Beato, M.S., Capua, I., 2007. Inactivation of avianinfluenza viruses by chemical agents and physical conditions:a review. Zoonoses Public Health 54 (2), 51–68.

Goh, G., Dunker, A., Uversky, V., 2009. Protein intrinsic disorderand influenza virulence: the 1918 H1N1 and H5N1 viruses.Virology Journal 3 (6), 69.

Hamilton, R.R.C., etal., 1977. Trimmed Spearman-Karbermethod forestimating median lethal concentrations in toxicity bioassays.Environmental Science and Technology 11 (7), 714–719.

Harrington, G.W., Chen, H.W., Harris, A.J., Xagoraraki, I., et al.,2001. Removal of Emerging Waterborne Pathogens. AWWAResearch Foundation and American Water Works Association,USA, 188 pp.

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 6 2485

Hijnen, W.A.M., Medema, D.G., 2007. Elimination ofMicroorganisms by Drinking Water Treatment Processes,third ed. IWA, Kieuwegein, The Netherlands.

Hinshaw, V., Webster, R., Turner, B., 1980. The perpetuation oforthomyxoviruses and paramyxoviruses in Canadianwaterfowl. Canadian Journal of Microbiology 26 (5), 622–629.

ISO 10705-1, 1995. Water Quality-detection and Enumeration ofBacteriophages, Part1. Enumeration of F-specific RNABacteriophages. International Organization forStandardization, Geneva, Switzerland.

Jacangelo, J.G., Patania Brown, N.L., Madec, A., Schwabb, K., et al.,2006. Micro- and Ultrafiltration Performance SpecificationBased on Microbial Removal. AWWA Research Foundationand American Water Works Association, USA, p. 212.

Jagger, J., 1967. Introduction to Research in UltravioletPhotobiology. Englewood Cliffs, NJ: Prentice-Hall, Inc.

Langlet, J., Gaboriaud, F., Gantzer, C., 2007. Effects of pH on plaqueforming unit counts and aggregation of MS2 bacteriophage.Journal of Applied Microbiology 103 (5), 1632–1638.

Langlet, J., Gaboriaud, F., Duval, J.F.L., Gantzer, C., 2008.Aggregation and surface properties of F-specific RNA phages:implication for membrane filtration processes. WaterResearch 42 (10–11), 2769–2777.

Langlet, J., Ogorzaly, L., Schrotter, J.C., Machinal, C., Gaboriaud, F.,Duval, J.F.L., Gantzer, C., 2009. Efficiency of MS2 phage and Qb

phage removal by membrane filtration in water treatment:applicability of a real-time RT-PCR method. Journal ofMembrane Science 326, 111–116.

LeChevallier, M.W., Au, K.K., 2004. Water Treatment andPathogen Control: Process Efficiency in Achieving SafeDrinking Water. World Health Organisation, p. 136.

Li, I.,Chan, K.,To, K.,Wong,S.,Ho, P.,L,S.K.,Woo,P., Tsoi, H., Chan, J.,Cheng, V., Zheng, B., Chen, H., Yuen, K., 2009. Differentialsusceptibility of different cell lines to swine-origin influenza AH1N1, seasonal human influenza A H1N1, and avian influenza AH5N1 viruses. Journal of Clinical Virology 46, 325–330.

Li, J.W., Xin, Z.T., Wang, X.W., Zheng, J.L., Chao, F.H., 2002.Mechanisms of inactivation of hepatitis A virus by chlorine.Applied and Environmental Microbiology 68 (10), 4951–4955.

Li, J.W., Xin, Z.T., Wang, X.W., Zheng, J.L., Chao, F.H., 2004.Mechanisms of inactivation of hepatitis A virus in water bychlorine dioxide. Water Research 38, 1514–1519.

Lucio-Forster, A., Bowman, D.D., Lucio-Martinez, B., Labare, M.P.,Butkus, M.A., 2006. Inactivation of the avian influenza virus(H5N2) in typical domestic wastewater and drinking watertreatment systems. Environmental Engineering Science 23 (6),897–903.

Lycett, S., Ward, M., Lewis, F., Poon, A., Kosakovsky Pond, S.,Brown, A., 2009. Detection of mammalian virulencedeterminants in highly pathogenic avian influenza H5N1viruses: multivariate analysis of published data. Journal ofVirology 83 (19), 9901–9910.

Machinal, C., Torres, S., Schrotter, J.C., 2006. Expertise centre ofexcellence ARAMIS dedicated to membrane plant design andoperation. Desalination 200 (1–3), 757–759 (11/12).

Malik Peiris, J.S., Poon, L.L.M., Guan, Y., 2009. Emergence ofa novel swine-origin influenza A virus (S-OIV) H1N1 virus inhumans. Journal of Clinical Virology 45 (3), 169–173.

Meng, Q.S., Gerba, C.P., 1996. Comparative inactivation of entericadenovirus, poliovirus and coliphages by ultravioletirradiation. Water Research 30 (11), 2665–2668.

Miller, G.L., Lauffer, M.A., Stanley, W.M., 1944. Electrophoreticstudies on PR8 influenza virus. Journal of ExperimentalMedicine 80 (6), 549–559.

Murray, B.K., Ohmine, S., Tomer, D.P., Jensen, K.J., Johnson, F.B.,Kirsi, J.J., Robison, R.A., O’Neill, K.L., 2008. Virion disruption byozone-mediated reactive oxygen species. Journal ofVirological Methods 153, 74–77.

NF EN 14476, 2005. Antiseptiques et Desinfectants Chimiques.Essai Virucide Quantitatif de Suspension Pour lesAntiseptiques et Desinfectants Chimiques Utilises enMedecine Humaine. National Organization forStandardization, France.

Neumann, G., Chen, H., Gao, G.F., Shu, Y., Kawaoka, Y., 2009.H5N1 influenza viruses: outbreaks and biological properties.Cell Research 2009, 1–11.

Ng, W.F., To, K.F., 2007. Pathology of human H5N1 infection: newfindings. The Lancet 370, 1106–1108.

Oppenheimer, J.A., Hoagland, J.E., et al., 1993. Microbialinactivation and charcaterization of toxicity and by-productsoccurring in reclaimed wastewater disinfected with UVradiation. In: Planning, Design & Operation of EffluentDisinfection Systems, Whippany, NJ.

Oster, G., 1951. The isoelectric points of some strains oftobacco mosaic virus. Journal of Biological Chemistry 190(1), 55–59.

Perez-Padilla, R., de la Rosa-Zamboni, D., Ponce de Leon, S.,Hernandez, M., Quinones-Falcon, I.F., Bautista, E., Ramirez-Venegas, A., Rojas-Serrano, J., Ormsby, C.E., Corrales, A.,Higuera, A., Mondragon, E., Cordova-Villalobos, J., 2009.Pneumonia and respiratory failure from swine-origininfluenza A (H1N1) in Mexico. The New England Journal ofMedicine 361 (7), 680–689.

Peterson, D., Hurley, T.R., Hoff, J.C., Wolfe, L.G., 1983. Effect ofchlorine treatment on infectivity of hepatitis A virus. AppliedEnvironmental Microbiology 45 (1), 223–227.

Rao, V.C., Symons, J.M., Ling, G.A., Wang, G.P., Metcalf, T.G.,Hoff, F.J.C., Melnick, J.L., 1988. Removal of hepatitis A virusand rotavirus by drinking water treatment. Journal of AWWA80 (2), 59–67.

Rice, E.W., Adcock, N.J., Sivaganesan, M., Brown, J.D.,Stallknecht, D.E., Swayne, D.E., 2007. Chlorine inactivation ofhighly pathogenic avian influenza virus (H5N1). EmergingInfectious Diseases 13 (10), 1568–1570.

Standard Methods for the Examination of Water and Wastewater,21st ed., 2005 American Public Health Association (APHA),American Water Works Association (AWWA) & WaterEnvironment Federation (WEF), 1368 pp.

Schijven, J.F., Teunis, P.F.M., de Roda Husman, A.M., 2005.Quantitative Risk Assessment of Avian Influenza VirusInfection via Water. RIVM report 703719012/2005, p. 33.

Smith, G., Vijaykrishna, D., Bahl, J., Lycett, S., Worobey, M.,Pybus, O., Ma, S., Cheung, C., Raghwani, J., Bhatt, S., Peiris, J.,Guan, Y., Rambaut, A., 2009. Origins and evolutionarygenomics of the 2009 swine-origin H1N1 influenza A epidemic.Nature 459 (7250), 1122–1125.

Sobsey, M.D., Fuji, T., Shields, P.A., 1988. Inactivation ofhepatitis A virus and model viruses in water by free chlorineand monochloramine. Water Science and Technology 20,385–391.

Sommer, R., Haider, T., et al., 1998. Time dose reciprocity in UVdisinfection of water. Water Science and Technology 38 (12),145–150.

Stallknecht, D.E., Shane, S.M., Kearney, M.T., Zwank, P.J., 1990a.Persistence of avian influenza virus in water. Avian Diseases34, 406–411.

Stallknecht, D.E., Kearney, M.T., Shane, S., Zwank, P.J., 1990b.Effects of pH, temperature, and salinity on persistence of avianinfluenza viruses in water. Avian Diseases 34, 412–418.

Takizawa, M., Okuwari, A., Okabe, T., 1973. The chemicalbehaviour of low valence sulphur compounds. VIII. Theoxidation of sodium thiosulfate with ozone. Bulletin of theChemical Society of Japan 46, 3785–3789.

USEPA, 1999. Alternative Disinfectants and Oxidants GuidanceManual. US Environmental Protection Agency, Office of Water,United States Environmental Protection Agency, p. 346.

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 2 4 7 3 – 2 4 8 62486

Valegard, K., Liljas, L., Fridborg, K., Unge, T., 1990. The three-dimensional structure of the bacterial virus MS2. Nature 346, 36.

Van Reeth, K., 2007. Avian and swine influenza viruses: ourcurrent understanding of the zoonotic risk. VeterinaryResearch 38 (2), 243–260.

Wallensten, A., Munster, V., Latorre-Margalef, N., Brytting, M.,Elmberg, J., Fouchier, R., Fransson, T., Haemig, P., Karlsson, M.,Lundkvist, A., Osterhaus, A., Stervander, M., Waldenstrom, J.,Bjorn, O., 2007. Surveillance of influenza A virus in migratorywaterfowl in northern Europe. Emerging Infectious Diseases13 (3), 404–411.

Wang, L.L., Hung, Y., Shammas, K.N., 2006. AdvancedPhysicochemical Treatment Processes. In: Handbook ofEnvironmental Engineering, vol. 4. Humana Press Inc, 690 pp.

Webby, R., Webster, R., Richt, J., 2007. Influenza viruses in animalwildlife populations. Current Topics in Microbiology andImmunology 315, 67–83.

Webster, R., Yakhno, M., Hinshaw, V., Bean, W., Murti, K., 1978.Intestinal influenza: replication and characterization ofinfluenza viruses in ducks. Virology 84 (2), 268–278.

Webster, R., Bean, W., Gorman, O., Chambers, T., Kawaoka, Y.,1992. Evolution and ecology of influenza A viruses.Microbiological Revue 56 (1), 152–179.

White, G.C., 1999. Handbook for Chlorination and AlternativeDisinfectants. John Wiley and Sons, New York.

WHO, 2006. Guidelines for Drinking-water Quality. IncorporatingFirst and Second Addenda, third ed. World Health Organization.

WHO, 2007. Review of Latest Available Evidence on PotentialTransmission of Avian Influenza (H5N1) Through Water andSewage and Ways to Reduce the Risks to Human Health.World Health Organization. http://www.who.int/water_sanitation_health/emerging/h5n1background.pdf, p. 37.

WHO, 2009. Human infection with pandemic A (H1N1) 2009influenza virus: clinical observations in hospitalized patients,Americas, July 2009-update. Weekly Epidemiological Records84, 305–308.

Wiedenmann, A., Fisher, B., et al., 1993. Disinfection of hepatitis Avirus and MS-2 coliphage in water by ultraviolet irradiation:comparison of UV-susceptibility. Water Science andTechnology 27 (3–4), 335–338.

Xu, X., Subbarao, K., Cox, N., Guo, Y., 1999. Geneticcharacterization of the pathogenic influenza A/Goose/Guangdong/1/96 (H5N1) virus: similarity of its haemagglutiningene to those of H5N1 viruses from the 1997 outbreaks in HongKong. Virology 261 (1), 15–19.

Zhang, G., Shoham, D., Gilichinsky, D., Davydov, S., Castello, J.,Rogers, S., 2007. Evidence of influenza a virus RNA in Siberianlake ice. Journal of Virology 81 (5), 2538.

Zhirnov, O., 1991. The anomalous isoelectric properties ofinfluenza virus matrix protein M1. Vopr Virusol 36 (3), 191–194.