Download Quantities of infectious virus and viral RNA recovered from sheep

Journal
of General Virology (2002), 83, 1915–1923. Printed in Great Britain
...................................................................................................................................................................................................................................................................................
Quantities of infectious virus and viral RNA recovered from
sheep and cattle experimentally infected with foot-and-mouth
disease virus O UK 2001
S. Alexandersen, Z. Zhang, S. M. Reid, G. H. Hutchings and A. I. Donaldson
Institute for Animal Health, Pirbright Laboratory, Ash Road, Pirbright, Woking, Surrey GU24 0NF, UK
The profiles of virus production and excretion have been established for sheep experimentally
infected with the UK 2001 strain of foot-and-mouth disease (FMD) virus by inoculation and by
direct and intensive contact. Virus replicated rapidly in the inoculated sheep, from which a peak
infectivity of airborne virus of 104n3 TCID50 per sheep per 24 h was recovered. Around 24 h later,
contact-infected sheep excreted airborne virus maximally. Similar amounts of airborne virus were
recovered from cattle. The excretion of virus by the sheep under these conditions fell into three
phases. First, a highly infectious period of around 7–8 days. Second, a period of 1–3 days soon
afterwards when trace amounts of viral RNA were recovered in nasal and rectal swabs. Third, at
4 weeks after exposure, the demonstration, by tests on oesophageal–pharyngeal samples, that
50 % of the sheep were carriers. The implications of the results and the variable role that sheep
may play in the epidemiology of FMD are discussed.
Introduction
Foot-and-mouth disease (FMD) is a virus disease of
domesticated and wild ruminants and pigs. FMD virus (FMDV)
is a member of the genus Aphthovirus within the family
Picornaviridae (Belsham, 1993) and has a significant degree of
variability – seven serotypes and more than 60 subtypes exist
(Haydon et al., 2001 ; Knowles et al., 2001). The isolate used in
the present studies belongs to serotype O and has specifically
been characterized as a ‘ PanAsian strain ’ (Knowles et al., 2001).
FMD is feared by farmers and veterinary authorities because of
its highly contagious nature and the drastic measures required
to eradicate the infection. As a consequence, FMD is the major
disease constraint to international trade in livestock and animal
products.
The contagious nature of FMD is a reflection of a number
of factors, including the wide host-range of the virus, the
amount of infectivity excreted by affected animals, the low
doses required to initiate infection and the many routes of
infection. FMD is most often spread when infected and
susceptible animals are brought into contact. Various mechanisms of transmission are possible in such circumstances –
between ruminant animals, the most common route is by the
inhalation of infectious droplets and droplet nuclei originating
mainly in the breath of the infected animals (Sellers et al.,
Author for correspondence : Soren Alexandersen.
Fax j44 1483 232448. e-mail soren.alexandersen!BBSRC.ac.uk
0001-8157 # 2002 SGM
1971 ; Sellers, 1971). The next most frequent mechanism of
spread is by the feeding of contaminated animal products, e.g.
milk and meat. Infection may also be spread by mechanical
means, for example when animals come into contact with the
virus on the surfaces of transport vehicles, milking machines or
on the hands of animal attendants. An additional mechanism is
the spread of FMDV by the wind. This occurs infrequently, as
it requires particular climatic and epidemiological conditions,
but is essentially uncontrollable and can be dramatic (Anonymous, 1969 ; Donaldson et al., 1982).
Sheep played a major role in the 2001 UK FMD epidemic
and it was important to obtain virological and aerobiological
data for that species infected with the UK strain of virus.
Infection of pigs has been the subject of a separate study
(Alexandersen & Donaldson, 2002). The current studies were
undertaken with the objectives of (i) generating virus and viral
RNA response curves for sheep infected with the O UK 2001
strain of FMDV and (ii) determining the amount of airborne
virus excreted by infected sheep and cattle for incorporation
into predictive models that simulate airborne spread.
Methods
Animals. Ten female, Dorset cross-bred sheep, weighing around
30 kg, were used. The sheep were shorn of their fleeces and placed in a
single room in a biosecure animal building. Six ‘ inoculated ’ sheep, i.e.
animals selected at random from the group, were infected by injection as
described below. Four ‘ contact ’ sheep were kept in the same room
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 03:23:19
BJBF
S. Alexandersen and others
throughout the experiment. The humidity was kept above 60 % and the
ventilation in the animal room was reduced to three air changes per hour.
Thus, the experimental design created optimal conditions for virus
transmission, simulating intensive, indoor conditions. In a subsequent
experiment, two heifers were used (Friesian Holstein breed, around 150
kg body weight).
The inoculated sheep were injected intradermally\subdermally in the
coronary band of a left fore foot (Burrows, 1968) with 0n5 ml of a stock
virus (for details, see below). The inoculum was diluted 1 : 10 in
MEM–HEPES (Eagle’s minimal essential medium with 20 mM HEPES
buffer and antibiotics). Each animal received around 10(n& TCID . The
&!
two heifers were inoculated with the same virus by subdermo-lingual
injection (Henderson, 1949, 1952).
Each day until 10 days post-inoculation (p.i.), the animals were
examined clinically and rectal temperatures were recorded. Blood samples
and nasal and rectal swabs were taken daily for the first 2 weeks after
inoculation (only on days 0–3 for the cattle). The samples were transferred
immediately to the laboratory ; swabs were processed immediately and
stored as mentioned below while blood samples were kept at 4 mC for
16–24 h and the serum separated. An aliquot of each serum was diluted
immediately 1 : 1 with Total nucleic acid lysis buffer (Roche) and stored at
room temperature for automated nucleic acid extraction (Roche) and
subsequent analysis by real-time 5h-nuclease RT–PCR. The rest of the
serum was frozen immediately and stored at k80 mC. Swabs were taken
in duplicate ; one swab was placed in 2 ml maintenance medium and
stored at k80 mC and the other was placed in 1 ml TRIzol (Life
Technologies) and stored at k80 mC. Oesophageal–pharyngeal (‘ probang ’) samples were taken from the sheep at 28 days after exposure.
These samples were shaken with an equal volume of buffer (pH 7n2–7n4)
and stored frozen at k80 mC until analysed. The sheep were killed at 28
days p.i. and the cattle at 3 days p.i.
Virus. The virus suspension was prepared from foot epithelium
obtained from a pig during the 2001 UK FMD epidemic. The virus isolate
is denoted FMDV O UKG 34\2001. A 10 % (w\v) suspension was made
in MEM–HEPES and stored as 0n5 ml aliquots at k80 mC. The titres of
this stock virus were 10)n) and 10(n' TCID \ml in primary bovine
&!
thyroid (BTY) cells (Snowdon, 1966) and in IB-RS-2 cells (De Castro,
1964 ; De Castro & Pisani, 1964), respectively.
Measurement of aerosol excretion of FMDV from sheep
and cattle infected with UKG 34/2001. Samples of the air in the
rooms containing infected sheep or cattle were collected on days 1 and
2 p.i. (sheep) and days 1 and 3 p.i. (cattle). These days were selected on
the basis of earlier findings (Donaldson et al., 1970 ; Sellers & Parker,
1969). In addition, air samples were collected from a series of three pairs
of inoculated sheep that were selected on the same days and placed for
about 15 min in a 610 l cabinet (Donaldson & Ferris, 1980). Multiple air
samples were collected. The relative humidity was kept high (above 60 %)
and therefore suitable for the survival of airborne FMDV (Donaldson,
1972).
Air sampling methods. Air samples from the animal rooms were
collected with an all-glass cyclone sampler operated for 20 min at a flow
rate of around 170 l\min (Gibson & Donaldson, 1986).
The air in the cabinet containing each pair of sheep was sampled with
a three-stage liquid impinger (May, 1966) containing a total of 30 ml
collecting fluid. The sampler was operated at 55 l\min for 5 min. The
collecting fluid employed in the impingers was MEM–HEPES with
antibiotics and 0n1 % (w\v) BSA (Donaldson et al., 1987 ; Gibson &
Donaldson, 1986).
The concentration of virus per litre of air was determined by endpoint titration of the collecting fluid of the particular air-sampler,
BJBG
multiplied by the volume of the collecting fluid and the flow of the
sampler. The amount of infectivity recovered was expressed as the total
amount of airborne FMDV in TCID per sheep or heifer per 24 h.
&!
Assay for virus. The infectivity of the collection fluid in air samplers
and in selected blood samples and swabs was determined by assay in
monolayer cultures of BTY cells in roller tubes (Donaldson et al.,
1987 ; Gibson & Donaldson, 1986). Tenfold dilution series of collecting
fluid samples were made and each dilution was inoculated onto five tubes.
Titres were calculated using the Karber equation according to Lennette
(1964). The specificity of any CPE was confirmed by antigen ELISA
(Ferris et al., 1988 ; Ferris & Dawson, 1988 ; Hamblin et al., 1984 ; Roeder
& Le Blanc Smith, 1987).
Quantitative RT–PCR. A quantitative RT–PCR method was used
to determine the amount of FMDV RNA in extracts of total nucleic acid
from blood and swab samples. Conversion to TCID equivalents\ml
&!
was made according to Alexandersen et al. (2001) and Oleksiewicz
et al. (2001). The assay method was similar to that used previously
(Alexandersen et al., 2001 ; Oleksiewicz et al., 2001) except that the
primers and the probe (patent pending) were altered so that the assay was
able to detect all isolates of FMDV (Reid et al., 2001). The specific
conditions used will be published in detail elsewhere (Reid et al., 2002).
All samples were stored in lysis buffer or in TRIzol until subjected to
automated total nucleic acid extraction in a MagnaPure LC robot (Roche).
All extractions involved 0n1 ml sample and the nucleic acid was finally
eluted in a volume of 0n1 ml. Thus, the nucleic acid was more dilute than
in earlier investigations (Alexandersen et al., 2001 ; Oleksiewicz et al.,
2001) ; however, the extraction had several advantages over the manual
method. Firstly, the extraction was very consistent and gave highly
reproducible results. Furthermore, because the samples were more dilute
and of a greater purity than in previous work, only a single dilution was
assayed (i.e. an amount corresponding to 0n003 ml initial sample in a
single RT–PCR). Fifty serum samples from the sheep on days 0–4 were
tested in duplicate. Eighty-five per cent of these duplicates fell within a
single Ct (Ct is ‘ cycle threshold ’, the first cycle where a sample can be
detected as positive). Fifteen per cent of the samples showed more
variability. The higher variability in this small percentage of samples was
probably caused by faulty or inaccurate manual pipetting and, subsequently, the system was changed to include a robotic arrangement of
the RT reaction and PCR. With this procedure of automated nucleic acid
extraction and robotic RT and PCR, the variability was consistently
within a single Ct for duplicate samples.
All estimations included standard reactions using samples with a
known content of FMDV (as determined by virus titration in cell culture)
and all quantifications were based on a comparison with standard curves
based on tenfold dilution series of cell culture supernatant at 40 h after
infection with FMDV (infectious titre determined by virus titration) as
described in detail previously (Alexandersen et al., 2001 ; Oleksiewicz et
al., 2001). The method is influenced minimally by sample type and has a
sensitivity (still in the quantitative range) of approximately 0n1
TCID \ml of either serum, cell culture or swab fluid. In most figures, the
&!
levels of FMDV RNA are given by the expression of 50kCt, in order for
samples with a higher content to be shown easily on the graphs. For
figures showing the relationship between real-time RT–PCR signal and
virus infectivity in cell culture, the RT–PCR values are given as
(50kCt)i2n3 in order to correlate the two dimensions, of which one had
a log scale (RT–PCR) and the other a log scale (cell-culture titration).
#
"!
Assay for antibodies. Serum samples were tested for the presence
of antibodies to FMDV by an ELISA using homologous virus passaged
in cell culture as described previously (Ferris, 1987 ; Ferris et al.,
1990 ; Hamblin et al., 1986a, b, 1987).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 03:23:19
FMD virus loads and excretion in sheep and cattle
Results
Airborne virus recovery and estimated aerosol
excretion and exposure doses
The amounts of airborne virus recovered from the animal
room containing inoculated and contact sheep, the cabinet
containing inoculated sheep and the animal room containing
inoculated cattle are shown in Tables 1 and 2. In brief, the data
show that sheep and cattle excreted up to 10%n$ TCID per
&!
animal per 24 h.
In order to make the titres comparable between the room,
which was ventilated (reduced to three air changes per hour),
and the cabinet, which was not ventilated, an amount of 10!n&
TCID was added to the quantity of virus recovered in room
&!
samples. Taken together, the samples collected from the
cabinet containing inoculated sheep and the animal room
containing all 10 sheep (both inoculated and contacts) indicated
that peak virus excretion by inoculated sheep occurred at 1 day
p.i., while the source of virus on the subsequent day most likely
originated from the contact sheep.
Clinical signs, virus load in serum and swabs and
seroconversion
Inoculated sheep developed local vesicular lesions (unruptured) at the injection sites as well as increased temperature of
one or more additional feet (generalization) by 1–2 days p.i.
and all six sheep showed mild clinical signs of lameness on days
2–5. Fever, defined as a temperature above 40 mC, began after
Table 1. Amounts of airborne FMDV recovered from the cabinet containing a pair of inoculated sheep or from the
animal room containing both inoculated and contact-infected sheep
Samples were collected with a three-stage (May) liquid impinger operating at 55 l\min for 5 min (cabinet samples) or a Cyclone sampler operating
at 170 l\min for 20 min (room samples). The mean amount of infectivity recovered was 10%n$ TCID per 24 h per inoculated sheep at 1 day p.i.
&!
Sample
Time of
sampling
Sheep
Cabinet samples (55 l/min for 5 min)
1
1 day p.i.
UI 97, UJ 03
2
1 day p.i.
UJ 01, UJ 00
3
2 days p.i.
UI 97, UJ 03
4
2 days p.i.
UI 96, UI 99
Room samples (170 l/min for 20 min)
1
1 day p.i.
All 10 sheep
2
2 days p.i.* All 10 sheep
Virus recovered
in sample (log10
TCID50/ml)
Concentration of
airborne virus
(TCID50/l)
0n8
0n4
Negligible
Negligible
0n7
0n27
Negligible
1n8†
0n2
Total airborne
virus (TCID50)
Estimated release of airborne
virus in 24 h (log10 TCID50)
190
4n74 for two sheep, 4n44 per sheep
75
4n34 for two sheep, 4n03 per sheep
No airborne virus detected
No airborne virus detected
No airborne virus detected (below detection limit)
631
4n7
* Collected at day 2 p.i. for inoculated sheep, which is probably day 1 post-infection for the contacts.
† An amount of 10%n( TCID was recovered from the room containing six inoculated and four contact-infected sheep. The data from the cabinet
&!
samples suggest that the main source of virus at this time (day 2 p.i.) was the four contact-infected sheep. This equates to 10%n" TCID per
&!
(contact) sheep, equivalent to about 10%n' TCID per 24 h per contact sheep at day 1 post-exposure (multiplied by three, i.e. an addition of 10!n&
&!
TCID , to account for reduction by ventilation ; see text).
&!
Table 2. Amounts of airborne FMDV recovered from infected cattle
Two heifers were sampled. Samples were collected with a Cyclone sampler operating at 170 l\min for 20 min. The mean amount recovered was
10%n$ TCID per 24 h per heifer.
&!
Sample
Time of
sampling
Concentration of virus
(log10 TCID50/ml)
Total airborne virus
(TCID50)
Estimated total release
of airborne virus in
24 h (log10 TCID50)
1
2
3
4
1 day p.i.
1 day p.i.
3 days p.i.
3 days p.i.
1n4
1n0
1n4
0n5
252
100
252
30
4n3
3n9
4n3
3n3
Mean release per heifer
(log10 TCID50 per 24 h
per animal)*
4n35
4n24
* An addition of 10!n& TCID was made to each value to account for ventilation ; see text.
&!
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 03:23:19
BJBH
S. Alexandersen and others
Fig. 1. FMDV genomes in serum samples
from sheep. The mean and standard
deviation (SD) in each group are shown in
relation to the number of days after the
start of the experiment. Signals are
expressed as 50kCt and, for certain
levels, the corresponding values for
TCID50 equivalents/ml are shown
[conversion to TCID50 equivalents/ml
made according to Alexandersen et al.
(2001) and Oleksiewicz et al., (2001)].
The start of fever (body temperature
above 40 mC) in the inoculated (4, n l
6) and contact-infected (#, n l 4)
groups is also indicated.
Fig. 2. Correlation of virus content in 40 nasal swab samples taken on
days 3 (), 4 (=), 5 (#) and 8 (5) titrated in BTY monolayer cell
cultures and calculated from real-time RT–PCR data. Titres are given on a
log10 scale. To achieve a linear correlation, the x-axis is expressed as
(50kCt)i2n3, because, in principle, the Ct value is on a log2 scale. Mean
values for days 3 (
), 4 (>) and 5 ($) are shown with surrounding
ellipses and day 8 (4) with a bar indicating the . The day 8 samples
were all negative by virus isolation.
Fig. 3. Correlation of virus content in 30 rectal swab samples taken on
days 3 (), 4 (=) and 8 (5) titrated in BTY monolayer cell cultures and
calculated from real-time RT–PCR data. Titres are given on a log10 scale.
To achieve a linear correlation, the x-axis is expressed as (50kCt)i2n3,
because, in principle, the Ct value is on a log2 scale. Mean values for days
3 (
), 4 (>) and 8 (4) are shown with surrounding circles indicating
the SD. The day 8 samples were all negative by virus isolation.
1–2 days and lasted up to 5 days. The mean time for the
development of fever was at day 2 in the inoculated group.
Among the contact group, only a single sheep showed signs
(increased temperature of a foot) at day 1 after inoculation of
the donor group. At days 2–6, more sheep from the contact
group showed signs of lameness so that, on day 6, all four
sheep had shown clinical signs of disease. Fever lasted from 1
to 3 days in this group. The mean time when fever was first
detected was at day 4 in this contact group (after inoculation
of the injected sheep). The time after exposure that lesions,
increased temperature or clinical signs of disease (lameness)
were first observed was on average 1n4 days for the inoculated
group and around 3 days for the contact group.
The cattle showed only mild signs of disease and had a
minor temperature increase. At 1 day p.i., both heifers had an
excess nasal secretion. At 2 days p.i., both heifers were lame on
one foot, although lesions were not obvious. At 3 days p.i., the
two heifers were killed. Examination revealed severe lesions on
the tongue (ruptured vesicles), but only mild generalized
lesions, in the form of small, ruptured vesicles on the dental
pad. In addition, small unruptured vesicles were found on one
or two feet. No rumenal lesions were observed.
The results obtained for the quantities of virus in the
inoculated sheep showed that viraemia (defined as detection of
FMDV genomic material in serum samples by real-time
RT–PCR) was detectable from 1 day p.i. and reached peak
values of around 10%n& TCID equivalents\ml at 2–3 days p.i.
&!
All of the inoculated sheep ceased to exhibit viraemia by 5–8
days p.i. All of the contact-infected sheep ceased to have
viraemia by day 8 (Fig. 1). Selected serum samples were also
tested by virus titration in cell culture (data not shown) and, as
before, strong correlation with the RT–PCR assay was found
BJBI
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 03:23:19
FMD virus loads and excretion in sheep and cattle
Fig. 4. FMDV genomes in nasal swab
samples from sheep. The mean as well as
the  in each group (inoculated sheep,
n l 6 ; contact sheep, n l 4) are shown
in relation to the number of days after
the start of the experiment. Signals are
expressed as 50kCt and the
corresponding values for TCID50
equivalents/ml are indicated for certain
levels. The corresponding levels of
airborne virus excreted from the sheep
are also indicated as the amount of
airborne virus excreted per sheep per
hour, as derived from the data in Table 1.
Fig. 5. FMDV genomes in rectal swab
samples. The mean and  in each group
(inoculated sheep, n l 6 ; contact sheep,
n l 4) are shown in relation to the
number of days after the start of the
experiment. See legend to Fig. 4 for
further details.
(Alexandersen et al., 2001) ; however, none of the serum
samples contained detectable levels of infectious FMDV after
day 7. However, low levels of viraemia (genomic material)
were found by RT–PCR in some day 9–11 samples. None of
the 10 sheep had detectable viraemia on days 12, 13 or 28
either by virus isolation or by RT–PCR (Fig. 1). However, the
signals for the day 9–11 serum samples corresponded to very
low FMDV genome levels, most likely below the detection
limit of cell culture. The results were repeatable, so it was
concluded that low-level viraemia (defined as presence of
FMDV RNA) can be seen for a few days in infected animals
after the initial peak viraemia is cleared by the developing
antibody response. In fact, similar patterns were seen with
nasal and rectal swab samples (see below).
Peak viraemia in both groups coincided with body
temperatures above 40 mC, which, as mentioned earlier, peaked
at days 2 and 4, respectively, for the inoculated and contact
sheep.
The correlation between the infectivity of swab samples
and TaqMan RT–PCR was compared by testing 40 nasal
swabs (taken at days 3, 4, 5 and 8 p.i.) and 30 rectal swabs
(taken at days 3, 4 and 8 p.i.) by both methods. A strong
correlation was found with samples containing more than 10#n!
TCID virus\ml ; however, samples with little or no detectable
&!
live virus were often positive in the RT–PCR and so this assay
appears to deliver the most accurate quantification of virus
load, even though the virus detected may not necessarily be
infectious (Figs 2 and 3). Nevertheless, infectivity and RT–PCR
reactivity were strongly correlated for the day 3, 4 and 5
samples. On day 8, all of the samples were negative for
infectivity (below detection limit), but several had a low
reaction by RT–PCR. This indicated that there was a strong
correlation between the assays and that samples that were
weakly positive by RT–PCR but negative in cell culture were
just below the detection limit of that particular assay. For the
rectal swabs, a moderately strong linear correlation between
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 03:23:19
BJBJ
S. Alexandersen and others
Fig. 6. Detection of antibodies in sera from inoculated sheep (solid line)
and contact-infected sheep (dotted line) by liquid-phase-blocking ELISA.
All negative samples have a titre of less than 1 : 16, while a titre of more
than 1 : 40 is considered positive (Donaldson et al., 1996). Inoculated
sheep, n l 6 ; contact sheep, n l 4.
samples and all the nasal swabs were negative, but one of 10
sheep had a positive rectal swab by RT–PCR. This particular
animal was a carrier both by virus isolation and by RT–PCR.
Interestingly, when the results of the RT–PCR analysis of the
10 sheep were divided into two groups of five sheep (five
carriers and five non-carriers), it was seen that the mean
number of days detected positive was higher in the carrier
group for both nasal and rectal swabs and, furthermore, that
the mean peak levels in these swabs were also higher. The
differences detected in the serum samples were less pronounced ; however, the mean peak viraemia was higher by
approximately 2 Ct (indicating that the peak viraemia of carrier
animals may be around fourfold higher than that of non-carrier
animals).
Discussion
infectivity and RT–PCR signal was found ; however, the
infectivity of these samples was clearly much lower than for
the nasal swabs, even for samples having a comparable signal
in RT–PCR. This suggested that the virus infectivity was
reduced by alimentary passage or, alternatively, that the rectal
swabs contained some material that reduced the sensitivity of
the cell-culture system. Day 8 samples were negative by virus
isolation, but a few samples were weakly positive in the
RT–PCR, while day 3 and 4 samples were positive for
infectivity at a low level (up to about 10#n! TCID ) but had a
&!
stronger signal by RT–PCR.
The virus loads found in the nasal and rectal swabs are
shown in Figs 4 and 5. The levels are generally higher in the
nasal swabs, with a mean of about 10$ TCID equivalents or
&!
about 10% in single samples, while rectal swabs contained 10#
TCID equivalents or less. FMDV RNA peaked in the nasal
&!
swabs around days 3–4 and in the rectal swabs at day 5.
The temporal development of antibodies is shown in Fig. 6.
Antibodies were detected in inoculated sheep from 4 or 5 days
p.i. and reached high levels by 7 days p.i. Contact sheep were
antibody-positive from day 6 or 7 and reached high levels by
day 8 or 9. The development of measurable levels of antibodies
correlated well with a sharp decrease in viraemia.
Samples from the two cattle were also examined (data not
shown in detail). Briefly, nasal swabs were positive on days
1–3 p.i. at a level corresponding to 10%n!–10&n! TCID
&!
virus\ml. This value is similar to that found for the day 3 and
4 sheep samples and in probang or in serum samples from
heifers at 1–3 days p.i. (data not shown).
Carriers
At 4 weeks after exposure, three of the six inoculated sheep
were carriers (two virus-positive and an additional sheep
positive by RT–PCR) and, of the four contact sheep, two were
carriers (one virus-positive and this one and another sheep
positive by RT–PCR). However, at 4 weeks p.i., all the serum
BJCA
A critical determinant of an FMD epidemic is the ability of
the virus infection to spread within and between premises
under field conditions. The rate and extent of spread will be
determined by the inherent properties of the virus, the amount
of virus excreted, the degree of direct and indirect contact, the
susceptibility of the species affected and the specific agricultural
and environmental conditions. An analysis of the first 10 days
of the 1967–68 UK epidemic, when spread was attributed to
the transmission of virus by the wind, resulted in widespread
transmission (Anonymous, 1969 ; Donaldson, 1979, 1986 ;
Donaldson et al., 1982 ; Gloster et al., 1982). Such widespread transmission is most likely to occur when there has
been airborne spread of virus from pigs to ruminants or,
alternatively, when there has been extensive contact between
infected and susceptible animals under conditions of dense
stocking. Recent analyses indicate that airborne spread over a
distance of more than 6–20 km is unlikely for most contemporary FMDV isolates (Donaldson et al., 2001 ; S.
Alexandersen, R. P. Kitching, J. H. Sorensen, T. Mikkelsen, J.
Gloster and A. I. Donaldson, unpublished data). However,
certain FMDV isolates may be excreted at very high levels
and may have caused airborne spread beyond 20 km (S.
Alexandersen, Z. Zhang and A. I. Donaldson, unpublished
data). The species infected at source is important, since the
quantity of airborne virus emitted by sheep and cattle is much
lower than that from pigs. Consequently, sheep and cattle are
much less likely to be the source of airborne spread than are
pigs (Donaldson et al., 1970, 2001 ; Donaldson, 1979 ; Sellers &
Parker, 1969 ; Sorensen et al., 2000). Another factor that may
result in widespread transmission is when infected animals are
moved through markets as part of extensive trading networks,
as was seen with sheep during the 2001 UK epidemic. Under
these conditions, spread may be rapid, distant and extensive.
Furthermore, the effect may be further amplified if the animals
are housed, since this will raise the aerial concentration of
FMDV and increase the probability, and potentially the load,
of infection. In contrast, transmission by direct or indirect
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 03:23:19
FMD virus loads and excretion in sheep and cattle
contact with infectious excretions, for example nasal fluid, will
probably be a less-efficient mechanism of transmission and thus
result in spread to a smaller proportion of animals.
The results described show that the UK 2001 strain of
FMDV was transmitted very rapidly from experimentally
inoculated to in-contact sheep. Within 48 h, the sheep infected
by contact were excreting airborne virus maximally under the
experimental conditions selected. The high dose of virus to
which both the inoculated and in-contact sheep were exposed
was probably the main factor that determined the speed of
transmission. Both categories of sheep were continuously and
closely confined, so transmission to the in-contact sheep could
have occurred by several different routes, e.g. by inhalation, by
ingestion or by intra-ocular or percutaneous infection. Considering that the inoculated sheep were excreting large
amounts of airborne virus within 24 h of inoculation and that
sheep are very susceptible to infection by the respiratory route
(Gibson & Donaldson, 1986), it is most likely that the contact
sheep were infected by the inhalation of highly infectious
droplets and droplet nuclei.
The incubation period in FMD depends on the species, the
strain of virus, the route of infection and, especially, the dose
of infection. In the present experiment, conducted under
intensive indoor conditions, the incubation period for the
inoculated sheep was 24–48 h. However, in a different
experiment with the same strain of virus, in which sheep were
exposed for 2 h to airborne virus from pigs, an incubation
period of 6–8 days was observed (Aggarwal et al., 2002). In
that experiment, the challenge dose would have been lower,
and this was probably the major factor that determined the rate
of spread through the groups of sheep. In the field, the dose of
exposure would be influenced by several factors. Management
activities resulting in the congregation of sheep, such as
lambing, shearing, deworming, clinical inspection etc., would
increase the challenge dose by the aerogenic route, especially
if the animals were indoors, but would also increase the
probability of transmission by additional routes. For example,
personnel handling the sheep could easily become contaminated by virus in nasal fluid, saliva, vesicular fluid, milk or
faeces and transfer it mechanically. The result would be shorter
incubation periods in individual sheep and an increased rate of
transmission through the flock. By contrast, incubation periods
would be longer and the rate of transmission slower among
sheep held outdoors under extensive conditions. It could be
hypothesized that in some circumstances the level of infection
might decline with time and be self-limiting. There is evidence
to support this hypothesis both during an epidemic of FMD
(Mackay et al., 1996) and from experimental studies using
conditions resembling less-intensive, outdoor conditions with
a much lower effective contact rate (Hughes et al., 2002).
The results obtained from the recovery of virus in samples
of air and from nasal and rectal swabs showed that the
excretion patterns of individual sheep could be subdivided into
three phases : first, an initial, highly infectious period lasting for
around 7–8 days ; second, shortly afterwards, a period lasting
for 1–3 days when trace amounts of virus RNA were detected
in swabs from the respiratory and alimentary tracts as well as
in serum samples ; and then third, at 4 weeks, the demonstration
that 50 % of the sheep were persistently infected, i.e. ‘ carriers ’.
The role of sheep in the epidemiology of FMD results
mainly from the first of these phases, since this is when sheep
are most infectious. It is of particular epidemiological importance that the emission of airborne virus by sheep occurs
from 1 or 2 days before clinical disease is likely to be apparent.
Even then, the proportion of sheep with obvious clinical signs
may be small (Donaldson & Sellers, 2000). The significance of
the second phase of excretion is not clear, but is almost
certainly of much less importance in terms of transmission.
However, this second phase may possibly be involved in the
development of carrier animals or possibly in the generation of
variant virus. The third phase, when a proportion of sheep may
become carriers, is variable. Around 45 % of experimentally
infected sheep have been shown to be carriers at 8 weeks and
25 % after 12 weeks (Burrows, 1968). The only evidence that
carrier sheep may have played a role in the epidemiology of
the disease is circumstantial, and this occurred 9 months after
infection (E. Stougaard, personal communication, 1983).
The peak amounts of airborne virus recovered from the
sheep and cattle in this study were similar to those obtained
with other strains of FMDV (Donaldson et al., 1970 ; Sellers et
al., 1970 ; Sellers & Parker, 1969). Airborne virus was detected
on only one of two days of sampling from sheep but on both
days 1 and 3 p.i. sampled from cattle. By contrast, other strains
of virus were recovered from sheep for up to 48 h longer,
although, in those studies, the pattern of very early excretion
from sheep was also evident (Donaldson et al., 1970 ; Sellers et
al., 1970 ; Sellers & Parker, 1969). The recovery, in earlier
studies, of amounts of infectivity over a longer period may
have been because the sampling devices collected larger
volumes of air and thus had greater sensitivity than our
system. The peak amounts of virus recovered from sheep and
cattle infected with the UK 2001 strain were only around onesixtieth of that recovered from infected pigs (Alexandersen &
Donaldson, 2002). This difference between ruminants and pigs
is considerably less than has been found for some other strains
of FMDV (Donaldson et al., 1970 ; Sellers, 1971 ; Sellers &
Parker, 1969 ; S. Alexandersen, Z. Zhang and A. I. Donaldson,
unpublished data).
The concentration of airborne virus in the room containing
the sheep was equivalent to 10−!n( TCID \l. Assuming that an
&!
adult sheep inspires about 15 l\min (Donaldson et al., 2001),
the dose received by the contact sheep was approximately
10$n' TCID per sheep over a 24 h period. With this
&!
concentration of virus in the air, the sheep are likely to have
inhaled an infectious dose within approximately 5 min
(Donaldson et al., 2001 ; Gibson & Donaldson, 1986).
The total amount of airborne FMDV emitted from an
infected pig farm is the product of the number of affected
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 03:23:19
BJCB
S. Alexandersen and others
animals, determined by careful clinical inspection, multiplied
by the excretion values for the particular virus. The risk of
airborne spread from the farm can then be assessed using an
atmospheric dispersion model and inputting the emission data
together with data about the climatic conditions and local
topography to simulate plumes of airborne virus (Donaldson et
al., 2001 ; Sorensen et al., 2001). The same can be done with
premises containing cattle, although the quantity of airborne
virus excreted will be much lower. For sheep premises, it would
be more difficult to make accurate estimates of virus emission
due to the cryptic nature of FMD in that species. Also, for
sheep kept under extensive conditions of husbandry, the
amounts of virus excreted at a given time are likely to be less
than intensively stocked animals due to the slower progression
of the disease in the flock.
Peak viraemia in both the inoculated and contact groups
correlated well with the pyrexic responses, which peaked at
days 2 and 4, respectively, for the inoculated and contact
sheep. The virus load found in the swab samples indicated that
the nasal tract, especially, contained significant amounts of
virus, while rectal swab samples, although often positive by
RT–PCR, were usually negative for or very low in tests for
infectivity. Interestingly, as reported previously (Alexandersen
et al., 2001), our quantitative RT–PCR assay was proportional
to infectivity on samples taken soon after infection, i.e. up to
about day 5 after exposure, and therefore before antibody
complexes would have been present. The correlation was
changed at later stages, presumably because the virus was
bound to antibody and not detected in infectivity assays but
still gave a positive reaction by RT–PCR. However, it is
obvious that the correlation of the two methods on rectal
swabs was different from the correlation on nasal swabs and
indicated a low infectivity of the rectal swab samples.
The clinical signs in the sheep infected with the UK 2001
strain were typical of FMD in that species ; however, the signs
in infected cattle were mild and accompanied by a minor
pyrexic response. In another experiment with the UKG
34\2001 FMDV strain in cattle (Aggarwal et al., 2002), the
lesions seen in infected cattle were also mild. However, these
results are at variance with observations in the field during the
UK epidemic, where cattle had severe lesions, especially on
the dental pad and gingiva (S. Alexandersen, R. P. Kitching,
J. H. Sorensen, T. Mikkelsen, J. Gloster and A. I. Donaldson,
unpublished data).
The development of antibodies around days 4–5 and 6–7 in
inoculated and contact sheep, respectively, coincided with a
sharp decline in viraemia. However, the decline caused by
circulating antibodies was more pronounced than that in the
nasal cavity as reflected in the nasal swabs. A reduced decline
in virus load despite the development of antibodies has
been suggested previously for epithelial lesions in pigs
(Alexandersen et al., 2001).
The number of samples from swabs and, to a lesser degree,
serum detected as positive was significantly higher in animals
BJCC
that became carriers and, furthermore, the mean peak levels
were higher. Thus, these studies may indicate a correlation
between peak virus load and duration of the virus load in
serum and swabs on the subsequent development of carrier
sheep.
We thank Tellervo Rendle, Linda Turner and Geoff Pero for excellent
technical assistance. Luke Fitzpatrick, Nigel Tallon, Darren Nunney and
Malcolm Turner are thanked for their assistance with the handling and
management of experimental animals. The research was supported by the
Department for Environment, Food and Rural Affairs (DEFRA), UK.
References
Aggarwal, N., Zhang, Z., Cox, S., Statham, R., Alexandersen, S.,
Kitching, R. P. & Barnett, P. V. (2002). Experimental studies with foot-
and-mouth disease virus, strain O, responsible for the 2001 epidemic in
the United Kingdom. Vaccine 20, 2508–2515.
Alexandersen, S. & Donaldson, A. I. (2002). Further studies to quantify
the dose of natural aerosols of foot-and-mouth disease virus for pigs.
Epidemiology and Infection 128, 313–323.
Alexandersen, S., Oleksiewicz, M. B. & Donaldson, A. I. (2001). The
early pathogenesis of foot-and-mouth disease in pigs infected by contact :
a quantitative time-course study using TaqMan RT–PCR. Journal of
General Virology 82, 747–755.
Anonymous (1969). Report of the Committee of Inquiry on Foot-and-Mouth
Disease (1968), part 1. Ministry of Agriculture, Fisheries & Food. London :
Her Majesty’s Stationery Office.
Belsham, G. J. (1993). Distinctive features of foot-and-mouth disease
virus, a member of the picornavirus family ; aspects of virus protein
synthesis, protein processing and structure. Progress in Biophysics &
Molecular Biology 60, 241–260.
Burrows, R. (1968). The persistence of foot-and mouth disease virus in
sheep. Journal of Hygiene 66, 633–640.
De Castro, M. P. (1964). Behaviour of the foot and mouth disease virus
in cell cultures : susceptibility of the IB-RS-2 cell line. Arquivos do Instituto
Biologico Sao Paulo 31, 63–78.
De Castro, M. P. & Pisani, R. C. B. (1964). The chromosome complement of the IB-RS-2 swine cell line susceptible to the foot and mouth
disease virus. Arquivos do Instituto Biologico Sao Paulo 31, 155–166.
Donaldson, A. I. (1972). The influence of relative humidity on the
aerosol stability of different strains of foot-and-mouth disease virus
suspended in saliva. Journal of General Virology 15, 25–33.
Donaldson, A. I. (1979). Airborne foot-and-mouth disease. Veterinary
Bulletin 49, 653–659.
Donaldson, A. I. (1986). Aerobiology of foot-and-mouth disease
(FMD) : an outline and recent advances. Revue Scientifique et Technique
Office International des Epizooties 5, 315–321.
Donaldson, A. I. & Ferris, N. P. (1980). Sites of release of airborne footand-mouth disease virus from infected pigs. Research in Veterinary Science
29, 315–319.
Donaldson, A. I. & Sellers, R. F. (2000). Foot-and-mouth disease. In
Diseases of Sheep, 3rd edn, pp. 254–258. Edited by W. B. Martin & I. D.
Aitken. Oxford : Blackwell Science.
Donaldson, A. I., Herniman, K. A., Parker, J. & Sellers, R. F. (1970).
Further investigations on the airborne excretion of foot-and-mouth
disease virus. Journal of Hygiene 68, 557–564.
Donaldson, A. I., Gloster, J., Harvey, L. D. & Deans, D. H. (1982). Use
of prediction models to forecast and analyse airborne spread during the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 03:23:19
FMD virus loads and excretion in sheep and cattle
foot-and-mouth disease outbreaks in Brittany, Jersey and the Isle of
Wight in 1981. Veterinary Record 110, 53–57.
Hughes, G. J., Mioulet, V., Haydon, D. T., Kitching, R. P., Donaldson,
A. I. & Woolhouse, M. E. J. (2002). Serial passage of foot-and-mouth
Donaldson, A. I., Gibson, C. F., Oliver, R., Hamblin, C. & Kitching, R. P.
(1987). Infection of cattle by airborne foot-and-mouth disease virus :
disease virus in sheep reveals declining levels of viraemia through time.
Journal of General Virology 83, 1907–1914.
minimal doses with O1 and SAT 2 strains. Research in Veterinary Science
43, 339–346.
Donaldson, A. I., Kitching, P. & Barnett, I. T. (1996). Foot-and-mouth
disease. In Manual of Standards for Diagnostic Tests and Vaccines, 1996 edn,
pp. 47–56. Paris : OIE.
Knowles, N. J., Samuel, A. R., Davies, P. R., Kitching, R. P. &
Donaldson, A. I. (2001). Outbreak of foot-and-mouth disease virus
Donaldson, A. I., Alexandersen, S., Sorensen, J. H. & Mikkelsen, T.
(2001). Relative risks of the uncontrollable (airborne) spread of FMD by
different species. Veterinary Record 148, 602–604.
Ferris, N. P. (1987). Development and use of Elisa in the control of footand-mouth disease. IAEA Proceedings 348, 65–77.
Ferris, N. P. & Dawson, M. (1988). Routine application of enzymelinked immunosorbent assay in comparison with complement fixation for
the diagnosis of foot-and-mouth and swine vesicular diseases. Veterinary
Microbiology 16, 201–209.
Ferris, N. P., Powell, H. & Donaldson, A. I. (1988). Use of pre-coated
immunoplates and freeze-dried reagents for the diagnosis of foot-andmouth disease and swine vesicular disease by enzyme-linked immunosorbent assay (ELISA). Journal of Virological Methods 19, 197–206.
Ferris, N. P., Kitching, R. P., Oxtoby, J. M., Philpot, R. M. & Rendle, R.
(1990). Use of inactivated foot-and-mouth disease virus antigen in
liquid-phase blocking ELISA. Journal of Virological Methods 29, 33–41.
Gibson, C. F. & Donaldson, A. I. (1986). Exposure of sheep to natural
aerosols of foot-and-mouth disease virus. Research in Veterinary Science
41, 45–49.
Gloster, J., Sellers, R. F. & Donaldson, A. I. (1982). Long distance
transport of foot-and-mouth disease virus over the sea. Veterinary Record
110, 47–52.
Hamblin, C., Armstrong, R. M. & Hedger, R. S. (1984). A rapid enzymelinked immunosorbent assay for the detection of foot-and-mouth disease
virus in epithelial tissues. Veterinary Microbiology 9, 435–443.
Hamblin, C., Barnett, I. T. & Hedger, R. S. (1986a). A new enzymelinked immunosorbent assay (ELISA) for the detection of antibodies
against foot-and-mouth disease virus. I. Development and method of
ELISA. Journal of Immunological Methods 93, 115–121.
Hamblin, C., Barnett, I. T. & Crowther, J. R. (1986b). A new enzymelinked immunosorbent assay (ELISA) for the detection of antibodies
against foot-and-mouth disease virus. II. Application. Journal of Immunological Methods 93, 123–129.
Hamblin, C., Kitching, R. P., Donaldson, A. I., Crowther, J. R. & Barnett,
I. T. (1987). Enzyme-linked immunosorbent assay (ELISA) for the
detection of antibodies against foot-and-mouth disease virus. III.
Evaluation of antibodies after infection and vaccination. Epidemiology and
Infection 99, 733–744.
Haydon, D. T., Bastos, A. D., Knowles, N. J. & Samuel, A. R. (2001).
Evidence for positive selection in foot-and-mouth disease virus capsid
genes from field isolates. Genetics 157, 7–15.
Henderson, W. M. (1949). The Quantitative Study of Foot-and-mouth
Disease Virus. ARC Report Series no. 8, pp. 1–50. London : Her Majesty’s
Stationery Office.
Henderson, W. M. (1952). A comparison of different routes of
inoculation of cattle for detection of the virus of foot-and-mouth disease.
Journal of Hygiene 50, 182–194.
serotype O in the UK caused by a pandemic strain. Veterinary Record 148,
258–259.
Lennette, E. H. (1964). General principles for laboratory diagnosis of
viral and rickettsial infections. In Diagnostic Procedures for Viral and
Rickettsial Diseases, pp. 48–51. New York : American Public Health
Association.
Mackay, D. K., Newman, B. & Sachpatzidis, A. (1996). 1994 Greek
FMD epidemic : analysis of the serological survey. Foot and Mouth Disease
Newsletter 1, 5.
May, K. R. (1966). Multistage liquid impinger. Bacteriological Reviews 30,
559–570.
Oleksiewicz, M. B., Donaldson, A. I. & Alexandersen, S. (2001).
Development of a novel real-time RT-PCR assay for quantitation of footand-mouth disease virus in diverse porcine tissues. Journal of Virological
Methods 92, 23–35.
Reid, S. M., Ferris, N. P., Hutchings, G. H., Zhang, Z., Belsham, G. J. &
Alexandersen, S. (2001). Diagnosis of foot-and-mouth disease by real-
time fluorogenic PCR assay. Veterinary Record 149, 621–623.
Reid, S., Ferris, N., Hutchings, G., Zhang, Z. & Alexandersen, S.
(2002). Detection of all seven serotypes of foot-and-mouth disease virus
by real-time, fluorogenic RT–PCR assay. Journal of Virological Methods (in
press).
Roeder, P. L. & Le Blanc Smith, P. M. (1987). Detection and typing of
foot-and-mouth disease virus by enzyme-linked immunosorbent assay : a
sensitive, rapid and reliable technique for primary diagnosis. Research in
Veterinary Science 43, 225–232.
Sellers, R. F. (1971). Quantitative aspects of the spread of foot and
mouth disease. Veterinary Bulletin 41, 431–439.
Sellers, R. F. & Parker, J. (1969). Airborne excretion of foot-and-mouth
disease virus. Journal of Hygiene 67, 671–677.
Sellers, R. F., Donaldson, A. I. & Herniman, K. A. (1970). Inhalation,
persistence and dispersal of foot-and-mouth disease virus by man. Journal
of Hygiene 68, 565–573.
Sellers, R. F., Herniman, K. A. & Donaldson, A. I. (1971). The effects of
killing or removal of animals affected with foot-and-mouth disease on the
amounts of airborne virus present in looseboxes. British Veterinary Journal
127, 358–365.
Snowdon, W. A. (1966). Growth of foot-and-mouth disease virus in
monolayer cultures of calf thyroid cells. Nature 210, 1079–1080.
Sorensen, J. H., Mackay, D. K., Jensen, C. O. & Donaldson, A. I.
(2000). An integrated model to predict the atmospheric spread of foot-
and-mouth disease virus. Epidemiology and Infection 124, 577–590.
Sorensen, J. H., Jensen, C. O., Mikkelsen, T., Mackay, D. K. &
Donaldson, A. I. (2001). Modelling the atmospheric dispersion of foot-
and-mouth disease virus for emergency preparedness. Physics and
Chemistry of the Earth 26, 93–97.
Received 19 October 2001 ; Accepted 26 March 2002
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 03:23:19
BJCD