J. gen. ViroL (1981), 54, 103-114
103
Printed in Great Britain
Differential Distribution of Virus and Histological Damage in the Lower
Respiratory Tract of Ferrets Infected with Influenza Viruses of Differing
Virulence
By C. S W E E T , l* J. C. M A C A R T N E Y , 2 R. A. B I R D , 1
D. C A V A N A G H , I ~ " M. H. C O L L I E , 1 R. H. H U S S E I N I 1 AND H. S M I T H 1
Departments of 1Microbiology and 2Pathology, University of Birmingham, P.O. Box 363,
Birmingham B15 2TT, U.K.
(Accepted 22 December 1980)
SUMMARY
The distribution of four strains of influenza virus [A/PR/8/34 (HON 1) and clone
64d (attenuated for ferrets) and clones 64c and 7a (virulent for ferrets) of the
recombinant virus A/PR/8/34-A/England/939/69 (H3N2)] in the lower respiratory
tract (trachea, bronchi and the hilar, intermediate and outer alveolar zones of the
lung) of ferrets was monitored daily for 4 days after intranasal inoculation. On day 1,
some animals had high virus titres in all the tissues but in other animals virus was
undetectable, irrespective of the virus strain. Two days after inoculation increase of
virus contents of all tissues tended to be restricted. On days 3 and 4, the virulent
clones (64c and 7a), in contrast to the attenuated strains (A/PR/8/34 and clone
64d), consistently infected the lower respiratory tissues. However, for all infected
animals the virus contents of the hilar zones of the lungs were higher than those in
the intermediate zones, while the alveolar zones were relatively free from virus.
Quantitative estimations of the mild histological damage occurring in the lower
respiratory tract 3 to 6 days after inoculation also indicated that bronchial and
bronchiolar tissue were more susceptible to influenza virus than alveolar tissue and
that clones 64c and 7a produced more damage than the other two strains. In
agreement with the relative viral contents of clones 64c and 7a in the bronchi and in
the hilar and intermediate zones of the lung, clone 64c produced more damage than
clone 7a in the bronchi and less in the bronchioles of the lung parenchyma.
INTRODUCTION
Human influenza is primarily an infection of the upper respiratory tract but tracheobronchitis can occur (Stuart-Harris, 1965). Although influenza pneumonia is rare (Mulder &
Hers, 1972; Stuart-Harris, 1965) tests of lung function show that there is a general alteration
in many aspects of pulmonary function during influenza infection (Camner et al., 1973;
Douglas, 1975; Homer & Gray, 1973; Johanson et al., 1969; Kennedy et al., 1965; Little et
al., 1979; Picken et al., 1972) even when the lungs appear normal by clinical and X-ray
examination (Douglas, 1975). These studies suggest that subclinical influenza involvement of
the lower respiratory tract is prevalent and that the severity of disease may well be determined
by the extent of involvement of these areas, especially in the young and old.
In an attempt to establish the viral and host factors which determine the extent of infection
in the lower respiratory tract, ferret influenza has been used as a model for the human disease.
As in humans, influenza in the ferret is mainly an upper respiratory tract infection
t Present address: Houghton Poultry Research Station, Houghton, Huntingdon, Cambridgeshire PE17 2DA,
U.K.
Downloaded from www.microbiologyresearch.org by
88.99.165.207
On: Sun, 18 Jun 2017 09:58:20
0022-1317/81/0000-4482 $02.00 © 1981 SGMIP:
104
c.
SWEET
AND
OTHERS
(Stuart-Harris, 1965; Sweet & Smith, 1980) but tracheitis, bronchitis, bronchiolitis and a rare
pneumonitis have been reported (Liu, 1955; Mulder & Hers, 1972; Shope, 1934;
Stuart-Harris, 1965). Observations on members of two series of recombinants (Campbell et
al., 1979; Matsuyama et aL, 1980; Toms et al., 1976, 1977a) and the A/New Jersey/8/76
(HswlN1) influenza virus (Toms et al., 1977b) showed that the extent of lung infection as
judged by the recovery of infectious virus from macerates of whole lung varied with the strain
of virus. Thus, for the recombinant virus system A/PR/8/34-A/England/939/69 (H3N2),
the virulent clones 64c and 6 produced peak infections higher than those of the virulent clone
7a and the virulent parent A/England/939/69 (H3N2). An attenuated clone 64d and the
attenuated parent A/PR/8/34 (HON1) produced little lung infection even though with the
latter virus the nasal infection was high, if not higher, than that formed by the virulent clones
(Toms et al., 1976, 1977a; Matsuyama et al., 1980). However, because the whole lung
macerates used included portions of major bronchi and their hilar subdivisions, the capacities
of different strains to attack anatomical and histological areas within the lower respiratory
tract were not tested in a differential manner.
In this study, we report the amounts of infectious virus found in different subdivisions of
the ferret respiratory tract and a quantitative assessment of the associated damage revealed
by histology. Two virulent (clones 7a and 64c) and two attenuated (clone 64d and
A/PR/8/34) strains of influenza virus have been used for these comparisons. It should be
emphasized that the terms 'virulent' and 'attenuated' are used here with respect to the viruses'
capacity to produce disease in ferrets. However, for all strains, except clone 64c, the terms
also apply to their virulence for man (Matsuyama et al., 1980).
METHODS
Influenza viruses. Clones 7a, 64c and 64d of the recombinant influenza virus
A/PR/8/34-A/England/939/69 (H3N2) and the parent virus A/PR/8/34 (HON 1) (A/PR/8)
were described previously (Matsuyama et al., 1980; Sweet et al., 1974a) together with the
preparation of seed and working stocks.
Infectivity assays. Infectivity assays in eggs (50% egg infectious dose, EIDs0) were
performed as previously described (Sweet et al., 1974 b).
lntranasal inoculation of ferrets. This was performed as described by Toms et al. (1976).
Measurement of virus infectivity in macerates of nasal turbinates, trachea, bronchi and
subregions of the lung. Taking care to prevent contamination, the nasal turbinates were
removed from freshly killed ferrets and stored at --70 °C. The lung, trachea and heart were
removed en bloc and placed in a Petri dish.
The lung and trachea were orientated in the Petri dish with the heart uppermost and then
the heart and all associated connective and adipose tissue were removed. The trachea was cut
off about 1 cm above its bifurcation into the major bronchi and then the bronchi were
removed at their entry to the lung lobes (Fig. i). The two major lobes of both the right and
left lung were numbered 1 to 4 anti-clockwise starting from the right upper lobe and each lobe
was divided into three subregions (Fig. 1). The hilar zone contained the lobar bronchi, blood
vessels, lymph glands and connective tissue with a minimum of parenchyma; the intermediate
zone contained mainly bronchi and non-respiratory bronchioles between which lie packed
lobules of parenchyma; and the cortical or alveolar zone consisted predominantly of
parenchyma (Herrnheiser & Hinson, 1954). Each sample was stored at - 7 0 °C.
After thawing, all samples were macerated in a Sorvall omnimixer in 3 ml Hanks' balanced
salt solution (Wellcome Reagents) containing 1% (w/v) bovine serum albumin, 0.044 % (v/v)
sodium bicarbonate, 100 units/ml penicillin, 100 pg/ml streptomycin and 120 pg/ml
chloramphenicol. The cups of the omnimixer were washed with 1 ml Hanks' solution, the
fluids pooled and centrifuged at 1400 g for 10 min at 4 °C. The supernatants were titrated in
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:58:20
Influenza virus virulence
105
Trachea
J
!
Main bronchi
fi
Hilar zone
Intermediate zone
Alveolar zone
Fig. 1. Diagram of the regions of the middle and lower respiratory tract that were examined separately
for virus content; lung lobe 2 (lowerlobe of the right lung) is shown.
10- to 12-day-old hens' eggs. The results were recorded as the total amount of virus in the
tissue piece, i.e. the virus titre/ml of homogenate supernatant multiplied by the volume of the
latter.
Quantitative histopathology. The lungs, trachea and heart were dissected en bloc and
inflated under a pressure of 25 cm H20 with 10% formal saline fixative containing 3%
sucrose. The upper end of the trachea was clamped and the organs placed in sucrose--formal
saline for 48 h. After fixation, each lobe from the right and left lung was sectioned sagitally
into slices 3 m m in thickness, dehydrated in alcohol, cleaned and embedded flat in Paraplast
(Lancer, St. Louis, Mo., U.S.A.). Blocks from the right and left main extrapulmonary bronchi
were similarly sampled and processed. An average of five blocks was obtained from each
major lung lobe and a single block included the whole of each main bronchus. Sections to
include the whole cut surface of each block were cut at 5/~m and stained with haematoxylin
and eosin. In preliminary studies blocks were step-sectioned at intervals of 25 ~ n to
determine whether the distribution of histological lesions was adequately sampled by the
above method.
Because unpublished observations on animals infected with the A/New Jersey/8/76
(HswlN1) virus indicated that only microscopic lesions could be expected, a modification to
the point counting method for assessing area and volume proportions in lungs described by
Dunnill (1962) was adopted. Sections were projected at a constant magnification using a
Leitz projecting microscope on to a screen ruled out in a grid of 2 cm squares so as to
produce points 2 cm apart. The number of points overlying areas of acute inflammation
[defined as polymorphonuclear (PMN) leucocyte infiltration] were noted and expressed as a
percentage of the total number of points counted. The whole of each section was covered by
this method. A total of approx. 7000 points were counted in all the lobes of the lung for each
animal and approx. 500 points were counted in each bronchus from each animal. In sections
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:58:20
106
c. SWEET AND OTHERS
(a) 1
n
l'L II.--,. ~
(b) 1
8
2
~6
o
11_
I
IH
(c) 1
2
,
.m_i
_
NT BR L1 L2 L3 L4
TR
3
no
NT B R L I L2
TR
Tissue
i
n
L3 L4
NT BRL1
TR
L2
L3 L4
Fig. 2. Total titres of virus given as 50% egg infectious dose (EIDso) at day 1 post-infection for (a)
A/PR/8/34, (b) clone 64d, (c) clone 64c and (d) clone 7a in nasal turbinates (NT), trachea (TR),
bronchi (BR), and in hilar (i), intermediate (El) and alveolar (1~) zones of lung lobes L1, L2, L3 and L4.
Three animals (1, 2, 3) were used for each virus. Each histogram represents the titre of virus in an
individual tissue or subregion.
of lung, grid points overlying bronchi (defined as airways containing mural cartilage) were not
included in the counts.
RESULTS
Groups of ferrets were inoculated intranasally with 105 50 % ferret infectious doses (FIDso)
(Toms et al., 1976; Matsuyama et al., 1980) of each virus (A/PR/8, clones 64d, 64c and 7a)
and three animals were examined for virus in their respiratory tissues at daily intervals for 4
days. Additional groups of ferrets were inoculated as above; three animals were killed 2, 3, 4,
5, 6 and 8 days post-infection and the lower respiratory tract was examined for
histopathological changes.
Virus contents of respiratory tissues
One day after intranasal inoculation the nasal turbinates of all animals were heavily
infected with each of the four strains (Fig. 2). Some animals (one infected with A/PR/8, one
with clone 64c and two with clone 7a) showed virus, often at high level, in all three areas of
most lung lobes (Fig. 2). Other ferrets (one for each virus) showed no evidence of virus in any
lung lobe and isolations were made only occasionally from the lung lobes of the remaining
animals. Thus, at this early stage of infection, the main differences in virus contents of the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:58:20
107
I n f l u e n z a virus virulence
(a) 1
i --
--
U
~
m.,.--
ILlffl
(b) l
2
J
e~
"~ 8
(c) 1
2
(d) 1
2
]
4
3
d
NT B R L 1
TR
L2
L3
L4
oH,
NT BRL1
TR
L2
L3
L4
_
.
3
NT B R L 1
TR
L2
L3
L4
Tissue
Fig. 3. Total titres of virus given as EIDs0 at day 2 post-infection for (a) A/PR/8/34, (b) clone 64d, (c)
clone 64c and (d) clone 7a in nasal turbinates (NT), trachea (TR), bronchi (BR), and in hilar (11),
intermediate (D) and alveolar ([]) zones of lung lobes L1, L2, L3 and L4. Three animals (1, 2, 3) were
used for each virus. Each histogram represents the titre of virus in an individual tissue or subregion.
lung lobes seemed to be between individual ferrets rather than between the strains of virus
(Fig. 2). However, in previous unpublished comparisons of virus contents of the whole lung
(i.e. all three areas of all four lobes) five out of 15 ferrets infected with clone 7a contained
>/10 s EIDso of virus in their lungs compared with none of nine ferrets infected with clone
64d, suggesting that clone 7a may have an increased capacity for reaching the lung early
in infection.
Two days after inoculation, the virus titres in nasal turbinate macerates were similar to
those on day 1 for A/PR/8 and decreased, markedly for clone 64d and slightly for clones 64c
and 7a (Fig. 3). Little virus was found in the trachea and bronchi of animals inoculated with
A/PR/8 and clone 64d, but titres were moderately high in a few animals inoculated with
clones 64c and 7a (Fig. 3). The high levels of virus found in some lung lobes of some animals
on day 1 (Fig. 2; nine lobes had virus titres > 105 EIDs0 in at least one lobe of the three
zones) were not evident on day 2 (Fig. 3; only two lobes had virus titres > l0 s EIDso).
A/PR/8 was present in the subregions of many lung lobes (Fig. 3 a) but at barely detectable
levels. Clone 64d was rarely detected (Fig. 3 b). Clone 64c was sporadically isolated from the
hilar and intermediate zones of some lobes (Fig. 3 c). Only animals infected with clone 7a
consistently showed significant levels of virus in most lung lobes (Fig. 3 d).
Three days after inoculation, nasal turbinate titres remained relatively high for A/PR/8
and at the reduced level of day 2 for clone 64d, but those for clones 64c and 7a had declined
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:58:20
108
c.
SWEET
AND
OTHERS
(a) 1
2
3
(b)
2
3
(c)
2
3
~6
[]
4
~
8
IIHlhtJ
NT 8R L1
TR
L2
L3
L4
NT BR L1
TR
L2
L3
L4
NT BRL1
TR
L2
L3
L4
Tissue
Fig. 4. Total titres of virus given as EIDs0 at day 3 post-infection for (a) A/PR/8/34, (b) clone 64d, (e)
clone 64c and (d) clone 7a in nasal turbinates (NT), trachea (TR), bronchi (BR), and in hilar (HI),
intermediate ([3) and alveolar ([]) zones of lung lobes L1, L2, L3 and L4. Three animals (1, 2, 3) were
used for each virus. Each histogram represents the titre of virus in an individual tissue or subregion.
(Fig. 4). Only one o f three animals infected with A / P R / 8 or clone 64d showed appreciable
levels of virus in the trachea and bronchi correlating with significant titres in the lung (Fig.
4a, b). In contrast, much virus was found in the trachea and bronchi of all animals infected
with clones 64c and 7a (Fig. 4c, d). O f the eight animals having virus in the trachea and
bronchi (Fig. 4 a to d), seven had bronchial titres greater than or equal to those in the trachea.
Only single animals infected with A / P R / 8 or clone 64d had appreciable titres of virus in
their lung lobes (Fig. 4 a, b), whereas clones 64c and 7a were found in the lungs of all animals
(Fig. 4 e, d). However, for all infected animals the titres in the hilar zones were greater than
those in the intermediate zones, which in turn were greater than those in the alveolar zones;
thus, for 24 lung lobes containing greater than 103 EIDs0 of virus in at least one zone (Fig. 4),
the titres in the hilar zones were greater than those in the intermediate and alveolar zones in
17 and 22 respectively and, in 20 lobes, the titres in the intermediate zones were greater than
those in the alveolar zones. Isolations of clone 64c from the lung lobes were more sporadic
than for clone 7a and only four out of 12 hilar zones contained greater than 105 EIDs0 of
clone 64c compared with 10 out of 12 zones for clone 7a; corresponding figures for the
intermediate zones were 4 and 10 respectively for clones 64c and 7a. In contrast, all three
animals infected with clone 64c had bronchial titres greater than 106 EIDs0 compared with
one animal infected with clone 7a (Fig. 4).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:58:20
Influenza virus virulence
109
3
(a) 1
(b) 1
8
2
3
4
if2
~
8
3
4
2
8
3
6
NT BRL1
TR
L2 L3 L4
NT BR LI L2 L3 L4
TR
Tissue
NT BRL1
TR
L2 L3 L4
Fig. 5. Total titres of virus given as EIDs0 at day 4 post-infection for (a) A/PR/8/34, (b) clone 64d, (c)
clone 64c and (d) clone 7a in nasal turbinates (NT), trachea (TR), bronchi (BR), and in hilar (el),
intermediate (t-q)and alveolar ([]) zones of lung lobes L1, L2, L3 and L4. Three animals (1, 2, 3)were
used for each virus. Each histogram represents the titre of virus in an individual tissue or subregion.
F o u r d a y s after inoculation, the pattern was similar to that on d a y 3 but the levels o f virus
were generally lower (Fig. 5). The important findings were: (i) of 10 animals having virus in
the trachea and bronchi nine had bronchial titres greater than those in the trachea; (ii) the
amounts of clone 64c and 7a in the lower respiratory tract were larger than those o f the other
two strains; (iii) of the 26 lung lobes having greater than 103 EIDs0 of virus in at least one
zone, the titres for the hilar zones were greater than those for the intermediate and alveolar
zones in 20 and 23 lobes respectively and, in 19 lobes, the titres in the intermediate zones
were greater than those in the alveolar zones; (iv) five out of 12 hilar zones of animals infected
with clone 64c contained more than 105 EIDs0 o f virus c o m p a r e d with eight out o f 12 hilar
zones infected with clone 7a; and (v) in two animals infected with clone 64c, the titres in
the bronchi were greater than 10 6 EIDs0 c o m p a r e d with one animal infected with clone 7a
(Fig. 5).
Quantitative assessment of histological damage in the lower respiratory tract
The histopathological effects of intranasal inoculation were similar to classical descriptions
of ferret influenza using ferret-adapted influenza strains (Stuart-Harris, 1965): namely,
necrosis and degeneration o f bronchial and bronchiolar columnar epithelium with focal
ulceration and a P M N leucocyte infiltrate associated with oedema; alveolar inflammation,
when it occurred, was centred on bronchiolar lesions. In contrast to lesions induced with
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:58:20
110
c . SWEET AND OTHERS
Table 1. Quantitative assessment of bronchial inflammation in ferrets 3 to 6 days following
infection with influenza viruses
% Inflammation* in right (R) and left (L) bronchi with:
A/PR/8
Day after
inoculation Ferret no.
3
1
2
3
1
2
3
1
2
3
1
2
3
Clone 64d
R
L
R
L
ND?
ND
ND
ND
0
0
0
0
ND
ND
0
0
0
0
ND
0
0
0
0
0
0
0
0
0
2.4(0.7) 9.1(1.3)
9.9(1.2) 8.1(1.3)
0
0
3.6(0.9) 3.4(0.8)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Clone 7a
Clone 64c
R
A
R
L
0
0
0
0
5.1(1.6)
4.1(1.3)
0
0
0
0
4.1(1.3)
0
0
17.4(1-9)
0.8(0.4)
8.8(1-2)
9.8(1-8)
0
8.1(1.2)
2.1(0.8)
4.5(1.4)
3'8(0.8)
0.3(0,3)
2.9(0.8)
4.2(1.4)
7.4(1.9)
ND
4.0(1-0)
0.5(0.5)
0
3.2(0-9)
L
25(2.1)
0
10'5(1"9) 14.9(1.5)
0
6(1.1)
0
0
9.2(1.2)
22.2(1.8)
44-3(2.7)
15.2(1.9)
6.3(1-3)
10.4(1.6)
0
0
0
* Assessed as described in Methods; standard errors in parentheses.
~"so, Not done.
ferret-adapted virus strains, the lesions encountered in this study were mild in all animals.
Histopathological evidence of damage post-dated rises in the virus contents of the lower
respiratory tract by approx. 24 h. Thus, changes were first noted on day 3 after virus
inoculation and rose to a peak on day 5 and thereafter declined. Epithelial regeneration
started on day 6 and increased by day 8. Bronchial and parenchymal inflammation were
qualitatively most evident in animals infected with clones 7a and 64c and almost non-existent
in animals infected with clone 64d. Strain A / P R / 8 appeared intermediate in this respect.
Acute inflammation and necrosis of mural bronchial glands was a prominent feature in many
animals.
Quantitative comparison of the acute inflammatory damage induced by viruses was
hampered by two factors: (i) the minor degree of damage; and (ii) the extremely focal nature
of the changes. In many animals only one lobe or one main bronchus showed any evidence of
damage. Counts for individual lobes of each lung were therefore amalgamated in order to
minimize standard errors, and expressed as a percentage for the whole lung (right and left).
Counts on bronchi were not amalgamated. No attempt was made to express counts in terms
of volume proportions of affected lung (Dunnill, 1962) as the standard errors of the
percentages were relatively high in most cases (Tables 1 and 2) and measurements of lung
volumes would have introduced further errors.
Despite these difficulties, the quantitative results (Tables 1 and 2) confirmed the qualitative
observations noted above. Using non-parametric testing, because normal distributions cannot
be assumed, the differences between viruses observed on days 3 and 4 were not significant in
either the bronchi or parenchyma. However, on days 5 and 6 the differences were significant.
In the bronchi, inflammation was significantly greater with clone 64c than with clones 7a (P
< 0.05), 64d (P < 0.01) and A / P R / 8 (P < 0.01). There was no significant difference
between the inflammation induced by clone 7a and A / P R / 8 ; both produced greater bronchial
inflammation than clone 64d (P < 0.01 and P < 0.05 respectively). In the parenchyma the
relative damage induced by clones 64c and 7a was reversed. Thus; parenchymal
inflammation (predominantly bronchiolar) was significantly greater with clone 7a than clone
64c (P < 0.05), clone 64d (P < 0.01) and A / P R / 8 (P < 0.01). Also, parenchymal
inflammation with clone 64c was significantly greater than with clone 64d (P < 0.05) and
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:58:20
Influenza virus virulence
111
Table 2. Quantitative assessment of parenchymal inflammation in lungs of ferrets 3 to 6 days
following infection with influenza viruses
Day after
inoculation
•
Ferret no.
% Parenchymal inflammation* with:
z.
,
A/PR/8
Clone 64d
Clone 64c
Clone 7a
3
1
2
3
ND'~
ND
ND
0
0.3(0"08)
ND
0.27(0'06)
0.97(0.12)
0
0.06(0.03)
0.04(0.04)
0'03(0"03)
4
1
2
3
0
0
0.02(0.01)
0
0.06(0.03)
0
0
0-18(0.04)
0.46(0.08)
0.09(0.09)
0.08(0.08)
0.16(0.16)
5
!
2
3
0-02(0.02)
0-03(0.01)
0.04(0.02)
0-03(0.02)
0.2(0.06)
0.02(0.02)
0-7(0.09)
1.4(0.14)
0.46(0.08)
2.43(0.25)
0.3(0.09)
1.59(0.22)
6
1
2
3
0-01(0.01)
0
0
0
0.02(0.02)
0.06(0.03)
0.64(0.07)
0.08(0.02)
0.05(0.02)
2.66(0.06)
3.81(0.06)
1.96(0.25)
* Assessed as described in Methods. The table includes bronchioles where the main inflammatory changes
occurred and the focal alveolar changes around the bronchioles, but the bronchi were excluded. Standard errors
are in parentheses.
t ND, Not done.
A / P R / 8 (P < 0.01) but there was no significant difference between clone 64d and A/PR/8.
Results from day 2 and day 8 are not shown in the tables and were not analysed as there was
no acute inflammatory reaction in tissues on these days.
DISCUSSION
In contrast to previous studies where infection was assessed at only two sites (i.e. in nasal
washings and in homogenates of whole lung) and numerous ferrets were examined (Toms et
al., 1976, 1977a; Matsuyama et al., 1980), the present work concentrated on gaining more
precise information about virus distribution and histological damage in the respiratory tract
and necessarily (there were 15 virus estimations on tissues from each animal) only relatively
few animals were surveyed. Nevertheless, the results were sufficiently consistent for valid
conclusions to be drawn, particularly the fact that, in the lower respiratory tract, more virus
and histological damage was found in the bronchi and in those regions of the lung (hilar and
intermediate zones) where ciliated epithelium was present than in the alveolar region.
The presence of virus in homogenates of any part of the respiratory tract does not
necessarily mean that virus replication has occurred there because virus can be moved by
mucociliary action or drainage. However, the fact that histological evidence of acute
inflammation and necrosis was seen in the sites where most virus was found suggests that
replication was in fact occurring there.
The titres of all four strains in homogenates of nasal turbinates confirmed the patterns of
nasal infection that were indicated by virus contents of nasal washings. The reasons for these
patterns are known, namely, the restrictive influence 2 days after inoculation of inflammatory
phagocytes, non-specific inhibitors and interferon on all virus strains coupled with the
differential effect of fever on their replication (Husseini et al., 1981; Sweet et al., 1977, 1978;
Toms et al., 1976, 1977a).
With regard to the lower respiratory tract, high levels of virus were found 1 day after
inoculation, irrespective of strain, in all parts of the lower tracts of some animals, although it
was virtually absent from these sites in others. This suggested that the efficiency of
mucociliary clearance differed in individual ferrets and, in some cases, could not correct
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:58:20
112
c . SWEET AND OTHERS
'spillover' of virus from the high levels in the upper respiratory tract. However, present and
previous experiments indicated that, with clone 7a at least, replication may have occurred in
the lower respiratory tract during the first day of infection.
Two days after inoculation, when restrictive influences curbed nasal infection (see above),
the high levels of virus found after 1 day in some lung zones of a few animals were not seen
(Fig. 2 and 3) and only in the case of clone 7a were there signs of consistent infection and
then at a moderate level (Fig. 3). If Fig. 2 and 3 are fairly representative of the positions on
days 1 and 2, bearing in mind the small number of ferrets involved, it appears that similar
restrictive influences operate on the second day in the lower respiratory tract to hinder
replication or inactivate some virus. Of the probable restrictive factors in the lung, such as
alveolar macrophages, interferon and fever, only the latter has been studied and then only for
alveolar tissue (Matsuyama et al., 1980; Sweet et al., 1978). Infection of organ cultures of
alveolar tissue at normal and pyrexial temperatures indicated that fever, which occurs from
about 1-5 to 2.5 days post-inoculation for all strains, could restrict their replication in the
alveoli at that time. However, the effect of pyrexial temperatures on replication in bronchial
and bronchiolar tissue in organ culture has not yet been investigated.
Three and 4 days after inoculation, the distribution of virus in the zones of lung-infected
animals suggested that, irrespective of strain, the epithelium of the bronchi and bronchioles
was more easily and heavily infected than the cells of the alveolar zone. This is supported by
previous studies using fluorescent antibody in which replication in the ciliated epithelial cells
of the bronchi and bronchioles was indicated by cytoplasmic and nuclear fluorescence and
desquamation of some fluorescing bronchiolar ceils (Liu, 1955). In the alveolar region
fluorescence was noted only in alveolar macrophages (Liu, 1955) although, in some studies
with ferret-adapted strains, fluorescence occurred in alveolar cells (Mulder & Hers, 1972;
Stuart-Harris, 1965) indicating that, as in organ cultures (Cavanagh et al., 1979; Kingsman
et al., 1977), these cells will support some replication. The infection pattern was further
supported by the fact that histological damage in alveoli appeared to be secondary to damage
in adjacent ciliated epithelium in bronchioles and bronchi and involved a greater relative
percentage area in bronchi than parenchyma. The reasons for the differential infection of the
lower respiratory tissues are unknown. Alveolar cells can support virus replication both in
organ cultures and in vivo (see above) so the relative lack of alveolar involvement might be
due to destruction of virus or virus-infected cells by alveolar macrophages or other host
defence mechanisms. The strong adherence of influenza virus to cilia (Dourmashkin &
Tyrrell, 1970; Gould et aL, 1972) may promote infection of bronchial and bronchiolar
tissue in vivo. Also, the ciliated epithelial cells of the bronchi and bronchioles may, like those
of the nasal mucosa (Cavanagh et al., 1979; Kingsman et al., 1977), be better able to support
virus replication and to release virus than alveolar cells. Thus, infection might develop and
spread more easily in ciliated epithelium. Bronchial replication probably supplied some of the
virus found in the trachea as a result of mucociliary action. Taking this into account, the
lower virus titres in the trachea compared with those of the bronchi (Fig. 4 and 5) and the fact
that histological damage in the trachea was minimal and focal (J. C. Macartney et al., unpublished observations) support conclusions from scanning electron microscopy that virus
replication in the trachea is restricted to small foci (Chevance et aL, 1978). If true, this
restriction is interesting since it occurs in ciliated epithelium.
The virus distributions on days 3 and 4 not only clearly indicated that clones 64c and 7a
infected bronchial tissue better than clone 64d and A/PR/8 but also suggested that clone 64c
infected the bronchi better than clone 7a and the lower regions (hilar and intermediate zones)
less well. These differences in distribution of the strains were substantially in agreement with
the relative amount of histological damage induced by them in either the major bronchi or
lung parenchyma. Thus, clones 64c and 7a induced greater overall damage than either clone
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:58:20
Influenza virus virulence
113
64d or A/PR/8, and clone 64c damaged the bronchi more than clone 7a and the bronchioles
of the lung parenchyma less. The differences were statistically significant despite the low
levels of damage and the high standard errors on measurements. Overall, the relative
percentage bronchial inflammation appeared more significant than parenchymal damage
(Tables 1 and 2). Clearly, the reasons for the differences between strains in their ability to
infect the lower respiratory tract must be sought in their relative capacities to infect bronchial
and bronchiolar tissue and not alveolar tissue, which seems little affected by even the virulent
strains.
This w o r k w a s s u p p o r t e d by the Medical R e s e a r c h Council. W e gratefully a c k n o w l e d g e the
technical assistance o f M r s L. H a r p e r .
REFERENCES
CAMNER, P., JARSTRAND, C. & PHILIPSON, K. (1973). Tracheobronchial clearance in patients with influenza.
American Review of Respiratory Disease 108, 131-135.
CAMPBELL, D., SWEET, C. & SMITH, H. (1979). Comparisons of virulence of influenza virus recombinants in ferrets
in relation to their behaviour in man and their genetic constitution. Journal of General Virology 44, 37-44.
CAVANAGH, D., MITKIS, F., SWEET, C., COLLIE, M. n. & SMITH, U. (1979). The localization of influenza virus in the
respiratory tract of ferrets: susceptible nasal mucosa cells produce and release more virus than susceptible
lung cells. Journal of General Virology 44, 505-514.
CHEVANCE, L. G., SIMONE-LAVOINE,N., VALANCOGNE-GROSJEAN,S., LESOURD, M. & AYMARD-HENRY, M. (1978).
Scanning and transmission electron microscopy study of ferret respiratory mucosa infected with influenza A
virus. Annales de Microbiologie 129A, 177-206.
DOUGLAS, R. G. (1975). Influenza in man. In The Influenza Viruses and Influenza, pp. 395-481. Edited by E. D.
Kilbourne. London: Academic Press.
DOURMASHKIN,R. R. & TYRRELL, D. A. J. (1970). Attachment of two myxoviruses to ciliated epithelial cells. Journal
of General Virology 9, 77-88.
DUNNILL, M. S. (1962). Quantitative methods in the study of pulmonary pathology. Thorax 17, 320-328.
GOULD, E. A., RATCLIFFE, S. A., BASARAB, O. & SMITH, n. (1972). Studies on the basis of localization of influenza
virus in ferret organ cultures. British Journal of Experimental Pathology 53, 31-36.
HERRNHEISER, G. & HINSON, K. F. W. (1954). An anatomical explanation of the formation of butterfly shadows.
Thorax 9, 198-210.
HO~ER, G. J. & GRAY, F. D. (1973). Effect of uncomplicated, presumptive influenza on the diffusing capacity of the
lung. American Review of Respiratory Disease 108, 866-869.
HUSSEINI, R. H., SWEEI~, C., COLLIE, M. H. & SMITH, H. (198 I). The relation of interferon and non-specific inhibitors
to virus levels in nasal washes of ferrets infected with influenza viruses of differing virulence. British Journal
of Experimental Pathology 62, 87-93.
JOHANSON, W. G., PIERCE, A. K. & SANFORD, J. P. (1969). Pulmonary function in uncomplicated influenza.
American Review of Respiratory Disease 100, 141-146.
KENNEDY, M. C. S., MILLER, D. L. & PEARSON, A. J. (1965). Respiratory function of recruits to the Royal Air Force
in health and during acute respiratory diseases. British Journal of Diseases of the Chest 59, 10-14.
KINGSMAN, S. M., TOMS, G. L. & SMITH, n. (1977). The localization of influenza virus in the respiratory tract of
ferrets: the different susceptibilities of fresh and maintained organ cultures. Journal of General Virology 37,
249-258.
LITTLE, J. W., DOUGLAS, R. G., HALL, W. J. & ROTH, F. K. (1979). Attenuated influenza produced by experimental
intranasal inoculation. Journal of Medical Virology 3, 177-188.
LIU, C. (1955). Studies on influenza infection in ferrets by means of fluorescein-labelled antibody. I. The
pathogenesis and diagnosis of the disease. Journal of Experimental Medicine 101,665-676.
MATSUYAMA, T., SWEET, C., COLLIE, M. H. & SMITH, H. (1980). Different aspects of virulence in ferrets exhibited by
influenza virus recombinants of known genetic constitution. Journal of Infectious Diseases 141, 351-361.
MULDER, J. & HERS, J. F. Ph. (1972). Influenza. Groningen: Wolters-Noordhoff.
PICKEN, J. J., NIEWOEHNER,D. E. & CHESTER, E. H. (1972). Prolonged effects of viral infections of upper respiratory
tract upon small airways. American Journal of Medicine 52, 738-746.
SHOPE, R. E. (1934). The infection of ferrets with swine influenza virus. Journal of Experimental Medicine 60,
49-61.
STUART-HARRIS, C. H. (1965). Influenza and Other Virus Infections of the Respiratory Tract. London: Edward
Arnold.
SWEET, C. & SMITH, n. (1980). Pathogenicity of influenza virus. Microbiological Reviews 44, 303-330.
SWEET, C., STEPHEN, J. & SMITH, H. (1974 a). The behaviour of antigenically related influenza viruses of differing
virulence on disulphide-linked immunosorbents. Immunochemistry 11,823-826.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:58:20
114
c.
SWEET
AND
OTHERS
SWEET, C., STEPHEN, J. ~ SMITH, H. (1974b). Purification of influenza viruses using disulphide-linked
immunosorbents derived from rabbit antibody. Immunochemistry 11,295-304.
SWEET, C., aIRD, R. A., TOMS, G. L., WOODWARD, C. G. & SMITH, H. (1977). Thermal stability and interaction with
ferret inflammatory exudates of two clones of influenza virus of differing virulence for both ferrets and man.
British Journal of Experimental Pathology 58, 635-643.
SWEEX, C., CAVANAGH, D., COLLIE, M. H. & SMITH, H. (1978). Sensitivity to pyrexial temperatures: a factor
contributing to virulence differences between two clones of influenza virus. British Journal of Experimental
Pathology 59, 373-380.
TOMS, G. L., BIRD, R. A., KINGSMAN, S. M., SWEET, C. & SMITH, H. (1976). The behaviour in ferrets of two closely
related clones of influenza virus of differing virulence for man. British Journal of Experimental Pathology 57,
37-48.
TOMS, G. L., DAVIES, J. A., WOODWARD, C. G., SWEET, C. & SMITH, H. (1977a). The relaxation of pyrexia and nasal
inflammatory response to virus levels in nasal washings of ferrets infected with influenza viruses of differing
virulence. British Journal of Experimental Pathology 58, 444-458.
TOMS, G. L., SWEET, C. ~ SMITH, H. (1977b). Behaviour in ferrets of swine influenza virus isolated from man.
Lancet i, 68-71.
(Received 3 November 1980)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 09:58:20