Download Persistent influenza C virus possesses distinct functional properties

Journal of General Virology (1994), 75, 2189-2196. Printed in Great Britain
2189
Persistent influenza C virus possesses distinct functional properties due to
a modified HEF glycoprotein
Manfred Marschall,'* Georg Herrler, 2 Christoph B6swald, 1 Gisela Foerst'
and Herbert Meier-Ewert ~
1 Abteilungfiir Virologie, Institutfiir Medizinische Mikrobiologie und Hygiene, Technische Universitiit Miinchen,
Biedersteiner Strafle 29, D W-80802 Miinchen and 2 Institut fiir Virologie, Philipps- Universitiit Marburg, D W-35037
Marburg, Germany
A model of long term viral persistence has been
established by selecting a spontaneous mutant strain of
influenza C/Ann Arbor/1/50 virus in a permanent cartier
culture of MDCK cells. Infectivity and cell tropism are
mainly determined by the multifunctional viral membrane glycoprotein (HEF). HEF analysis was aimed at
identifying a putative correlation between sequence and
function, i.e. receptor binding, enzymatic activity,
antigenicity and rate of infection. The current experimental picture is summarized by the following
findings: (i) C/Ann Arbor/I/50 persistent virus carries a
modified receptor-binding sequence, (ii) receptor-binding activity is altered, as indicated by a higher efficiency
in recognizing low amounts of the receptor determinant
N-acetyl-9-O-acetylneuraminic acid, (iii) direct attachment to cell surfaces differs from that of wild-type virus,
as measured by slower kinetics of viral elution, (iv)
receptor-destroying enzymatic activity is diminished, (v)
characteristic features of virion surface morphology are
altered or unstable, (vi) persistent-type HEF epitopes
are distinguishable by monoclonal antibodies from wildtype and (vii) viral infectivity is intensified for cells
bearing a low number of receptors. The sum of these
changes highlights a structurally and functionally
modified HEF glycoprotein that allows long term viral
persistence. In order to clarify which of the described
points are required for the persistent viral phenotype, a
working concept is presented.
Introduction
binding, receptor-destroying activity and membrane
fusion (reviewed by Herrler & Klenk, 1991). Point
mutations in the predicted receptor-binding pocket have
been demonstrated to change the binding preference for
modified receptors (Szepanski et al., 1992) and, consequently, to influence the reproduction rate in cell culture
(Umetsu et al., 1992). The receptor-destroying function
of HEF was proposed to be required for entry into target
cells at a step prior to fusion (Strobl & Vlasak, 1993).
The fusion efficiency between viral and cellular membranes was attributed to pH-dependent, conformational
changes o f H E F (Formanowski et al., 1990). In the as yet
unexplained phenomenon of persistent infection, HEF
seems to undergo structural and functional variation.
This was proposed after initial studies on virus antigenicity, haemagglutination inhibition and plaque morphology (Camilleri & Maassab, 1988).
Given this background we sought to correlate HEF
variability with the persistence of infection. HEF
sequence studies were carried out in parallel with
experiments on its function and were supplemented by
data on the structure and infectivity of virions. Here we
present evidence that influenza C virus persistence is
Persistent infections with RNA viruses are supposed to
be achieved by different viral strategies, e.g. restricted
viral replication, interference with cell regulation and
escape from immune surveillance (reviewed by Oldstone,
1991). The support of survival and longevity of infected
cells is a critical prerequisite for viral persistence in
culture models (Celma & Fernandez-Mufioz, 1992;
Urabe et al., 1992), as well as in the natural host
(Kandolf et al., 1991). In order to maintain the infectious
process over long periods the selection of lowcytotoxicity variants and reduced virus shedding appear
to be essential (Clavo et al., 1993; Urabe et al., 1993).
Events stabilizing virus-cell interactions, such as the
inhibition of differentiation and of programmed cell
death (Montgomery et al., 1991 ; Levine et al., 1993), as
well as alteration of the tissue tropism (Shioda et al.,
1992; Subbarao et al., 1993), have been repeatedly
observed.
In influenza C virus infection, vital functions for
infectivity are served by the single viral surface glycoprotein (HEF). The functions of HEF include receptor
0001-2359 © 1994SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 05:31:44
2190
M. Marschall and others
associated with specific functional changes displayed on
the virion surface.
Methods
Virus and cell culture. Influenza C virus strains Ann Arbor/I/50
wild-type (C/AA-wt) or a persisting variant (C/AA-pi) were grown in
MDCK cells or embryonated eggs as described previously
(Formanowski & Meier-Ewert, 1988). MDCK cells and persistently
infected MDCK cells were grown in Dulbecco's minimum essential
medium containing 10% (v/v) fetal calf serum. The persistently
infected culture was maintained at 33 °C with biweekly feedings by
replacing 50 % of the growth medium.
Amplification o f virus sequences by PCR. Virion RNA was extracted
from infectious allantoic fluids according to the guanidinium thiocyanate protocol (Chomczynski & Sacchi, 1987). One lag of RNA was
reverse-transcribed in 10 lal of 100 mM-Tri~HC1 pH 8.3, 10 mM-MgC12,
140 mM-KC1, 10 mM-DTT, 400 laM-dNTPs and 3 pmol of influenza
virus universal primer Unil (5' AGCAAAAGCAGG 3') (Robertson,
1979) using avian myeloblastosis virus reverse transcriptase (12.5 U;
Boehringer Mannheim). The mixture was incubated at 42 °C for 60 rain
and the reaction terminated by boiling for 1 min. Subsequent PCR was
performed in a 30 cycle programme as described by Manuguerra et al.
(1992) with temperature levels of 50 °C for annealing (1 min), 72 °C for
polymerization (3 min) and 94°C for denaturation (1 min). The
verification of amplification products was achieved by agarose gel
electrophoresis. Oligonucleotide primers for the HEF-encoding segment 4 were chosen from nucleotide positions 22 to 52 (5' ATGTTTTTCTCATTACTCTTGATGTTGGGCC 3') and 1024 to 1048
(3' CCGTCTCTTAGACTGGTACGTCACC 5') (position numbering
according to Buonagurio et al., 1985).
Nucleotide sequencing. PCR products were purified using NICK
columns (Sephadex G-50), as described by the manufacturer
(Pharmacia) and used directly as sequencing templates. The nucleotide
sequences of both strands were determined by the dideoxynucleotide
chain termination method (Sanger et al., 1977). Primers were between
15 and 17 nucleotides in length. They were synthesized using a
Bioscience Award synthesizer and were desalted with NAP-5 columns
(Sephadex G-25; Pharmacia). Sequencing reaction products were
separated on 6 to 8 % urea/polyacrylamide gels in a Bio-Rad SequiGen sequencing cell.
Resialylation o f erythrocytes. Erythrocytes from 1-day-old chicks
were treated with neuraminidase and resialylated using c~2,3- and c~2,6sialyltransferases (Boehringer Mannheim) as described elsewhere
(Schultze et al., 1990; Szepanski et al., 1992). The enzyme substrate,
CMP-activated
N-acetyl-9- O-acetylneuraminic
acid
(CMPNeu5,9Ac2), was kindly provided by Professor Dr Dr R. Brossmer.
Using different amounts of substrate, batches of erythrocytes were
obtained that differed in the level of surface-bound Neu5,9Ac 2.
Following incubation for 3 h at 37 °C, the erythrocytes were washed,
resuspended in PBS to a final concentration of 1% and used for
haemagglutination assays.
Attachment assay. Two-million trypsinized MDCK cells per reaction
were rinsed with PBS and incubated with 50 haemagglutinating units of
influenza C virus in a 50 lal volume. Virus attachment to cell surfaces
was achieved by incubation for 30 min at room temperature. The cells
were then rinsed again and incubated in 50 tal of PBS at 37 °C for the
times indicated in order to allow the membrane-attached virus to elute
or penetrate. The process was stopped by non-detergent formaldehyde
fixation onto glass slides (3 % in PBS, 15 min at room temperature).
Surface fluorescence study of bound particles was performed by
staining with influenza C virus-specific rabbit antisera in a standard
protocol for indirect immunofluorescence (Marschall et al., 1989).
Acetylesterase assay. Virus was pelleted from allantoic fluids, rinsed
and resuspended in PBS. The protein concentration was determined
with a conventional test kit (Bio-Rad). One lag of the prepared virus
was used to measure viral acetylesterase activity by incubation with the
substrate analogue p-nitrophenyl acetate (1 mM in PBS) in a final
volume of 0-5 ml. The turnover of the colour substrate at 37 °C was
monitored for the times indicated by measurement of the A400 of the
reaction (Marschall et al., 1993).
Determination o f infectivity. Chick embryo fibroblasts (CEF) were
prepared from 10-day-old fertilized eggs. After mechanical grinding,
embryo tissues were incubated in 0.25 % (w/v) trypsin for 1 h at 37 °C
and filtered through a sterile gauze mesh. The solubilized cells were
pelleted and seeded in 24-well microtitre plates (Nunc) in minimal
essential medium (MEM) containing 10 % fetal calf serum. When the
cells became confluent, adhering fibroblasts were washed to remove
remaining cell debris and then utilized for infectivity assays. Two
sublines of Madin-Darby canine kidney cells, MDCK I cells (high
levels of virus receptor) and MDCK II (low levels of virus receptor),
were seeded as described above, but grown in Dulbecco's MEM.
Infections were performed with serial dilutions of influenza C virus for
1 h at 33 °C. The virus inoculum was removed, followed by two washes
with PBS. Incubation post-infection (p.i.) was performed at 33 °C for
3 days with reduced fetal calf serum (2 % v/v). Virus yield was
determined from the supernatant by standard haemagglutination
microtitration using 1% chicken erythrocytes.
Electron microscopy. Cell culture-grown virus was pelleted from the
supernatant (90000 g for 2 h) after removal of cell debris (3000 g for
15 min). Pellets were rinsed in PBS pH 7.2 and resuspended by mild
souication on ice. Using carbon-coated grids, the samples were
negatively stained with 2 % (w/v) potassium phosphotungstate pH 6.0
for 15 s and analysed immediately.
Results
Persistent virus carries a modified receptor-binding
sequence that is associated with an enhanced binding
efficiency
In order to characterize critical steps in the persistent
virus cycle, we investigated its receptor binding properties in resialylation experiments (Fig. l a and b).
Neuraminidase-treated erythrocytes from 1-day-old
chicks were incubated with CMP-activated 9-0acetylated sialic acid and one of the two different
sialyltransferases. In this way erythrocytes were obtained
bearing Neu5,9A% bound to surface glycoconjugates
either in 0~2,3 linkage to Galfll,3GalNAc or in a2,6
linkage to Galfll,4GlcNAc. Further variation was
obtained by varying the amount of CMP-Neu5,9Ac z. As
shown in Fig. 1, persistent virus was more efficient in
recognizing 9-O-acetylated sialic acid in both the c~2,3 (a)
and ~2,6 linkage (b) than the parental wild-type virus.
The only case where haemagglutination was effected by
both viruses to the same titre is shown in Fig. 1 (b); in this
case erythrocytes were incubated with a2,6-sialyltransferase and 50 pmol of CMP-activated sialic acid.
When the amount of CMP-Neu5,9A% was lowered
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 05:31:44
Influenza C virus persistence
T a b l e 1. HEF sequence variation with respect to
256
receptor binding and cell tropism
(a)
Influenza C virus isolates
64
r
C / A A / 1 / 5 0 (wild-type)t
C / A A / 1 / 5 0 (persisting
variant)
C / J H B / 1 / 6 6 (receptorbinding variant):~
C/YA/4/88 (cell-adapted
variant):~
C/YA/4/88
(neutralization-resistant)
variant):~
¢.-
16
m
m.
m
0-1
¸
0-25
0.5
2
CMP-9-O-Ac-Neu5Ac (nmol)
256
(b)
<
e~
.=
2191
rZ
64
r-"
Amino acid sequence*
275-SPYT GNSGDTPT MQCD- 290
L I
I
N
N
* The bold-print sequence represents the receptor-binding pocket, as
determined by analogy with influenza A virus (Weis et al., 1988) and
with the variants aligned below.
t The region shown for C / A A / 1 / 5 0 (wild-type) is totally conserved
in different strains, e.g. C/Johannesburg/1/66, C/Yamagata/10/81 or
C/California/78 (Buonagurio et al., 1985).
:~ Sequence comparison between C / A A / 1 / 5 0 (persisting variant)
and the depicted isolates was based on publications by Szepanski et al.
(1992), Umetsu et al. (1992) and Matsuzaki et al. (1992), respectively.
~100
16
90
= 80
-~
.~ 70
.~ 60
50
4
30
J
2.5
5
10
25
CMP-9-O-Ac-Neu5Ac (pmol)
50
128
20
10
0
3
2.5
32
2
g
,* 1.5
1
Adult
1-day-old
Asialylated
Chicken erythrocytes
I
0.5
/AA-wt
/AA-pi
0
0
Fig. 1. Virus-receptor binding studies. The ability of influenza C / A n n
A r b o r ~ l ~ 5 0 virus wild-type (C/AA-wt) and the persistent variant
(C/AA-pi) to use 9-O-acetylated sialic acid in two different linkage
types as a receptor determinant was tested. Neuraminidase-treated
erythrocytes were resialylated by incubation with either
Galfll,3GalNAc~2,3-sialyltransferase (a) or Galfll,4GlcNAc72,6sialyltransferase (b) in the presence of different amounts of CMPNeu5,9Ac 2. Following sialylation, the erythrocytes were used to
determine the haemagglutinating titre of C/AA-wt (dark bars) and
C/AA-pi (light bars). Agglutination of erythrocytes from 1-day-old
chickens (c) was determined for C/AA-pi (light bars), C/AA-wt
(medium bars) and as a standard virus control C / J H B / t / 6 6 (dark
bars). Neuraminidase-treated erythrocytes (Asialylated) were included
as a negative control.
10
30
60 min
Time (min)
Fig. 2. Viral RDE activities were determined by haemagglutination
elution (a) and acetylesterase assay (b). (a) Conventional haemagglutination reactions, using 1% chicken erythrocytes and equivalent
dilutions of egg-grown influenza C / A n n Arbor/1/50 virus were assayed
for 30 rain at room temperature. Thereafter the microtitre plates were
shifted to 37 °C in a water bath and observed for RDE-mediated
elution. The continued reversion of haemagglutination to elution is
expressed as percentage relative to the initial haemagglutination titre.
(b) Acetylesterase activity of particles was quantified in vitro by the
colour reaction with a substrate analogue. Reactions were performed in
triplicate for control and quickly measured at fixed time points. The
profiles of the test shown relate to 1 gg of total protein each.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 05:31:44
2192
M. Marschall and others
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
Mock
Adsorption
120 rain 37 °C
120 min on ice
C/AA-wt
C/AA-pi
Fig. 3. Kinetics of virus attachment to cell surfaces. InfluenzaC/Ann Arbor/l/50 virus (a to d) and its persistent variant (C/AA-pi;
e to h) were processedin an MDCK cell attachment assay and visualizedby indirect immunofluorescence.Mock controls (a and e) are
a cell control without virus. Adsorption of virus was stopped after 30 min at room temperature (b and f) or was followedby washing
and 120 min incubation for elution/penetration at 37 °C (c and g) or on ice (d and h), respectively.
fivefold, the receptors generated by the transferase
reaction were insufficient for haemagglutination by
C/AA-wt, whereas the persistent virus still showed
activity, even when the amount of substrate was reduced
20-fold. Moreover, persistent virus also had a significant
advantage over wild-type when the haemagglutination
test was performed with erythrocytes containing 9-0acetylated sialic acid in an ~2,3 linkage (Fig. 1 a). These
results indicate that C/AA-pi has an increased affinity
for Neu5,9Ac2 compared to that of the parental virus.
The high efficiency of receptor-binding by C/AA-pi is
also evident from its unexpected ability to agglutinate
erythrocytes from 1-day old chicks (Fig. l c). It is
important to note that these erythrocytes contain 9-0acetylated sialic acid in such a low amount that they are
resistant to the haemagglutinating activity of all influenza
C virus strains analysed so far (Herrler & Klenk, 1987a).
The results shown in Fig. 1 (c) indicate that the persistent
virus is unique among influenza C viruses in its ability to
utilize these low amounts of Neu5,9Ac 2 for the agglutination of erythrocytes.
Nucleotide sequence analysis (Table 1) revealed a
change of two amino acids located close to one another
within the predicted receptor-binding pocket of H E F at
positions 284 (Thr--* Leu) and 286 (Thr ~ Ile). This
alteration was demonstrated repeatedly in several
samples of persistent genomes isolated and amplified
independently (i.e. different cell culture and egg passages).
Wild-type virus, conversely, was found to be genetically
homogeneous with respect to the conserved receptorbinding sequence as shown by a strain-specific PCR
technique. Since residue 284 (Thr ~ Leu) is altered by a
double point mutation in two neighbouring nucleotides,
we positioned the extreme 3' end of the respective P C R
primer exactly at this genome location and hence
successful amplification was strictly reserved for the
persistent type (Marschall & Meier-Ewert, 1993).
Viral receptor-destroying enzyme (RDE) activity and the
rates of cell surface elution
The R D E of influenza C virus is a neuraminate-Oacetylesterase (Herrler et al., 1985). R D E activity
counteracts the effect of haemagglutination. We
attempted to find strain-specific differences in the R D E
activity by the reversal of haemagglutination (Fig. 2a).
The kinetics of elution from erythrocytes, comparing
wild-type and persistent virus, revealed a constant
retardation of about 70 min in the case of persistent-type
virions, regarding complete RDE-mediated haemagglutination clearance.
Quantification of the enzymatic H E F activity was
accomplished by an in vitro acetylesterase assay (Fig.
2b). Virus preparations, equivalent in amounts o f total
protein, were tested and the results corroborated the
previous findings. Persistent virus was reduced in
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 05:31:44
Influenza C virus persistence
2193
substrate reactivity as shown by a lowered signal at the
turnover saturation point (60 min).
These inherent features o f persistent virus were
confirmed at the level o f cell surface reactivity. Both
types o f virions were allowed to attach to M D C K cells
and were subsequently tested for the continuous presence
o f m e m b r a n e - b o u n d viral antigen (Fig. 3). The efficiency
o f binding and elution was quantified by microscopic cell
counting with respect to the rate o f specific fluorescence.
Virus clearance occurred with an efficiency o f approximately 90 % after 120 min at 37 °C, in the case o f wildtype virus. Persistent virus, however, was clearly retained
on the cell surface after the same incubation period
(reduction o f only 10 %). As a control, the reaction was
carried out on ice and no difference was observed
between the two strains, i.e. b o u n d viral particles o f both
types remained constantly detectable. This p r o m p t e d us
to test the effectiveness o f a potent R D E inhibitor, 3,4dichloroisocoumarin (Vlasak et al., 1989), in our assay.
The presence o f 500 I.tM o f inhibitor during the 120 min
incubation at 37 °C reduced the wild-type specific surface
clearance by approximately 50 %, indicating that acetylesterase activity is involved in this process but is not the
sole cause o f the effects observed. The constant binding
o f persistent virus was unaffected under these conditions
(data not shown).
Structural divergence of persistent virions
The surface o f influenza C virus particles is typically
organized in a highly ordered H E F glycoprotein structure, i.e. hexagons (Flewett & Apostolov, 1967). Electron
microscopic analysis o f C / A A - w t demonstrated a hexagonal pattern o f spikes on the virion surface (Fig. 4).
Table 2. Virus variant-specific antibody reactivity
Haemagglutination
inhibition titrest
Influenza C
virus-specific
antibodies*
C/AA-wt
C/AA-pi
Ratio S
H1/66§
Nil
FC1.16.3.3¶
FCGB4.6.3¶
FCDD4.1¶
DA2.6.4¶
DC6.6.6-7-2¶
400
100
128000
6400
5
40
5
400
100
64000
12800
80
40
40
1/1
1/1
2/1
1/2
1/16
1/1
1/8
* All antibodies tested were raised against virus strain C/JHB/1/66.
t Absolute values represent the maximum dilution factor of the
respective antibody achieving inhibition in the standard haemagglutination inhibition tests.
$ Relative values represent the ratio between the antibody's
reactivity towards wild-type and persistent virus, respectively.
§ Polyspecific rabbit antiserum used as a positive control.
I] Non-immune rabbit serum used as a negative control.
¶ HEF glycoprotein-specific MAbs.
Fig. 4. Virion morphology. Influenza C/Ann Arbor/I/50 virus wildtype (a and b) and its persistent variant (c and d) were freshly prepared
from infected cell supernatants by ultracentrifugation. Negatively
stained particles were photographed. Note the differences in conservation of the hexagonal surface pattern. Bar marker in (d) represents
400 nm in (b) and (d) and 250 nm in (a) and (c).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 05:31:44
2194
M. Marschall and others
Contrastingly, the persistent variant was shown to be
largely devoid of the wild-type morphology, showing a
strikingly smooth appearance. The paucity of hexagons
was noticed in independent virion specimens. The
influences of storage, preparation or staining conditions
on this structural aberration have still to be determined.
In order to gain insight into the consequences of
structural variation, we tested influenza C virus strains
for antibody binding measured by haemagglutination
inhibition (Table 2). Monoclonal antibodies (MAbs)
directed against H E F (strain C / J H B / 1 / 6 6 ; kindly
provided by Dr R . W . Compans) displayed variable
reactivities on a scale from very high (1:128000) to
almost ineffective (1:5) with both viruses tested. There
was no total loss of any of the epitopes tested for in the
persistent-type virus. However, an intriguing finding is
that there is at least one MAb (FCDD4.1) which reacted
to a significantly higher (16:1) ratio in favour of
persistent virus. This effect indicates a decisive modification in H E F epitopes and antigenicity.
The infectious yieM of persistent virus increases at low
receptor levels
The consequences of H E F alteration to the biological
behaviour of the virus were addressed by analysing
infectivity and virus multiplication (Fig. 5). CEFs were
utilized to correlate the titres of infecting and progeny
virus (Fig. 5b). Progeny rates were mostly similar to
those of wild-type virus (Fig. 5a). For both viruses,
optimal inoculum dilutions for maximal progeny rates
were evident, with a slightly higher infectivity for C / A A pi, but this varied in separate experiments. M D C K II
cells were also assayed with the same virus preparations.
This subline of M D C K cells was shown to be resistant to
infection by influenza C / J H B / 1 / 6 6 virus due to the low
receptor number phenotype of its cell surface (Herrler &
Klenk, 1987b). As shown in Fig. 5, M D C K II cells also
behave in a non-susceptible way towards influenza
C / A A / 1 / 5 0 wild-type virus. Samples of medium taken 3
days and 7 days p.i. were mostly devoid of measurable
progeny virus. Persistent virus, however, produced
marked yields in the period from days 3 to 7 p.i. This
important difference was reproduced several times using
persistent virus from independent sources and passages.
Interestingly, the optimal m.o.i, was constantly determined at unexpectedly high dilutions (e.g. 1 : 16 in Fig.
5, corresponding to only 3 haemagglutinating units/ml
in the inoculum). As controls, M D C K I cells which carry
high receptor concentrations and are commonly used for
influenza C virus studies, were infected in parallel. Here
the progeny rates for both viruses were equivalent.
Infectivity profiles and virus quantities in these cells were
mostly identical, which is in direct contrast to the
100
(a)
//•.,
10
/
)
zJ
x~
\
\
e...
_=
\
0.1
.
2
3
i
A
.
4
5
100
6
7
(b)
10
\
0.1
x
0
1
?
2
3
4
Inoculum dilution (log2)
Fig. 5. Infectivityof wild-typeand persistent virus for CEF, MDCK I
(high receptor number) and MDCK II cells (low receptor number).
Confluent cell layers, grown in microtitre plates, were tested for
influenzaC/Ann Arbor/1/50 virusproduction (a, C/AA-wt; b, C/AApi) under optimized conditions. Titres of inoculum (shaded bars) and
progeny virus (lines) were determined by haemagglutination assay.
Representative values (one experiment out of eight in parallel) are
shown, from experimentsperformed with virus derived from different
egg-passage levels. Symbols: &, CEF cells 3 days p.i.; x, MDCK I 3
days p.i. ; O, MDCK II cells 3 days p.i. ; II, MDCK II cells 7 days p.i.
situation in M D C K II cells. This cell-specific discrepancy
argues strongly against a general, low permissiveness of
M D C K cells or a defective wild-type virus preparation.
Discussion
In this study of a persistent influenza C virus strain, we
describe mutations in the putative receptor-binding
pocket of H E F in association with unusual receptorbinding, R D E and structural properties. Furthermore an
increase in infectivity under conditions of low cell
receptor concentrations was noted and the possibility of
host preference, concerning the cell tropism, is evident.
Specific binding to cells carrying a limited number of
receptor sites will consequently lead to the specialization
for a subpopulation of target cells. Analogous receptorbinding variants of influenza C virus with mutations in
the same position or in the vicinity of the ones depicted
here have been published (Table 1) and the importance
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 05:31:44
Influenza C virus persistence
of receptor determinants for the cell tropism is well
characterized (Herrler & Klenk, 1987b). A reinforcing
effect, illustrating the idea of virus benefit at low receptor
concentrations, is the impairment of R D E activity.
Persistent virus might be handicapped in dealing with
high cellular receptor levels, in terms of lack of elution
(i.e. inefficient cleavage from invalid receptors on nonpermissive cells or binding to lysed membranes by
default). Cells of a low receptor number type appear to
become profitable targets for this virus variety.
Importantly, R D E activity is considered to be a ratelimiting factor in influenza C virus reproduction. Recently this function was shown to be required for entry
into target cells (Strobl & Vlasak, 1993), therefore an
impairment might contribute to the self-limitation of the
viral cycle under persistence. Serine 71 of H E F was
described as being located in the R D E active site,
between amino acids 68 and 74, and is highly conserved
among influenza C virus strains (Herrler et al., 1988). We
also found this site to be conserved in persistent virus by
nucleotide sequencing (data not shown). Since this
sequence was claimed to be essential for enzymatic
activity, dramatic changes would be predicted to lead to
a complete loss of function. The measured reduction in
R D E activity, however, is more likely to be the result of
an altered conformation in adjacent tertiary structures as
proposed by Table 2 and Fig. 4: changes in epitope
accessibility were shown to be concomitant with alterations in virion morphology (at least under stressed
conditions, e.g. microscopy staining). The significance of
these structural effects has still to be clarified in detail,
but information from a persistence model with lymphocytic choriomeningitis virus (LCMV) illustrates this
point: a single amino acid change in the LCMV
glycoprotein was shown to be associated v¢ith an unusual
epitope reactivity and with persistence in vivo (Salvato et
al., 1991). The effects of alterations in the virus surface
have also been discussed in a report about two mutated
capsid genes of poliovirus which were sufficient to confer
a persistent phenotype (Calvez et al., 1993).
Nucleotide sequencing and functional analysis of
persistent-type H E F were performed with virus derived
from different hosts, i.e. cell culture and embryonated
hen eggs. Persistently infected M D C K cells have been
cultured for more than 5 years and virus isolation from
the supernatant was carried out several times. Thus the
propagation of cell-grown C/AA-pi in fertilized eggs was
achieved by serial allantoic inoculation and all virus used
in these studies originated from passage levels four and
five. The conservation of the persistent viral phenotype
was proven by the re-infection and re-establishment of
long term persistence in fresh M D C K cells (unpublished
observation). Furthermore, in the biochemical tests
described, persistent virus appeared to be invariable.
2195
These findings suggest that the determinant for persistence is genetically stable and that no host-specific
selection of revertants is evident.
These results imply a discrete effect of the described
changes in H E F on the establishment of influenza C
virus persistence. Given this, our hypothesis to explain
the trigger for viral persistence includes several further
observations. We propose the possible involvement of
more than one critical determinant. (i) Divergence of the
virion glycoprotein seems to confer a distinct functional
prerequisite for a non-cytocidal, persistent viral life
cycle. The selection of moderately or low productive
strains, evolving by long term persistence, has been
reported for other virus systems, e.g. measles virus
(Celma & Fernandez-Mufioz, 1992; Hirano et al., 1993)
and influenza B virus (Clavo et al., 1993). These
characteristics in virus replication accompany an adaptation to suitable cellular subpopulations. (ii) The
connection between cell tropism and an extreme variation of viral glycoproteins is impressively illustrated by
persistence of human immunodeficiency virus (Shioda et
al., 1992). Yet, apart from this, (iii) we have experimental
evidence for the elusive stability of viral R N A during
non-productive phases (Marschall et al., 1993) and for
the ability to maintain R N A persistence only in certain
cell types (M. Marschall, G. Foerst & H. Meier-Ewert,
unpublished results). In line with our observations are
studies showing the extraordinarily long half-life of
inactivated influenza A viral R N A segments (Cane &
Dimmock, 1990) and on the characterization of a unique
virus regulatory protein from persistent infection (Lucas
et al., 1988). Considering these aspects we favour the
notion that besides H E F variation, additional regulatory
functions of other virus genes might be involved. For
C/AA-pi virus, investigations of the non-structural
coding region and gene products are under way.
We express our thanks to Dr H. F. Maassab, Department of
Epidemiology,Universityof Michigan, Ann Arbor, Mich., U.S.A. for
providing us with the persistently infected cell culture, to Dr J. S.
Robertson, National Institute for Biological Standards and Control,
Potters Bar, Hertfordshire, U.K. for help with nucleotide sequencing,
to Dr R. W. Compans, Department of Microbiologyand Immunology,
Emory University School of Medicine, Atlanta, Ga, U.S.A., and to
Mrs G. Terfloth, Institute for Anatomy, Technical University of
Munich, Germany for cooperation in electronmicroscopyas well as to
Dr I. Chaloupka for reading the manuscript. M.M. is indebted to the
research group of Dr H. Wolf, Institute for Medical Microbiologyand
Hygiene, Universityof Regensburg, Germany for stimulating scientific
discussions. This work was supported by Deutsche Forschungsgemeinschaft, Me422/3-1.
References
BUONAGURIO, D., NAKADA, S., DESSELBERGER,U., KRYSTAL, M. &
PALESE, P. (1985). Noncumulative sequence changes in the hemag-
glutinin genes of influenza C virus isolates. Virology 146, 221-232.
CALVEZ, V., PELLETIER, I., BORZAKIAN,S. 8z COLBERE-GARAPIN,F.
(1993). Identificationof a region of the poliovirus genome involved
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 05:31:44
2196
M . Marschall and others
in persistent infection of HEp-2 cells. Journal of Virology 67,
44324435.
CAMILLEm, J. & MAASSAB, H. (1988). Characteristics of a persistent
infection in Madin-Darby canine kidney cells with influenza C virus.
Intervirology 29, 178-184.
CANE, C. & DIMMOCK, N. "(1990). Intracellular stability of the gene
encoding influenza virus hemagglutinin. Virology 175, 385 390.
CELMA, M. L. & FERNANDEZ-Mtrg~OZ, R. (1992). Measles virus gene
expression in lytic and persistent infections of a human lymphoblastoid cell line. Journal of General Virology 73, 2203 2209.
CHOMCZWqSKI, P. & SACCHI, N. (1987). Single-step method of RNA
isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Analytical Biochemistry 162, 156-159.
CLAVO, A., MAASSAB,H. & SHAW, M. (1993). A persistent infection in
MDCK cells by an influenza type B virus. Virus Research 29, 21-31.
FLEWETT, T. H. & APOSTOLOV, K. (1967). A reticular structure in the
wall of influenza C virus. Journal of General Virology 1, 297-304.
FORMANOWSKI, F. & MEIER-EWERT, H. (1988). Isolation of the
influenza C virus glycoprotein in a soluble form by bromelain
digestion. Virus Research 10, 177-192.
FORMANOWSKI, F., WHARTON, S. A., CALDER, L. J., HOFBAUER, C. &
MEIER-EWERT, H. (1990). Fusion characteristics of influenza C
viruses. Journal of General Virology 71, 1181-1188.
HERRLER, G, & KLENK, H.-D. (1987a). Restoration of receptors for
influenza C virus on chicken erythrocytes by incubation of bovine
brain gangliosides. In The Biology of Negative Strand Viruses, pp.
63-67. Edited by B. W. J. Mahy & D. Kolakofsky. Amsterdam:
Elsevier.
HERRLER, G. & KLENK, H.-D. (1987b). The surface receptor is a major
determinant of the cell tropism of influenza C virus. Virology 159,
102-108.
HERRLER, G. & KLENK, H.-D. (1991). Structure and function of the
HEF glycoprotein of influenza C virus. Advances in Virus Research
40, 213 234.
HERRLER,G., ROTT, R, KLENK,H.-D., MOLLER,H.-P., SHUKLA,A. &
SCHALrER, R. (1985). The receptor-destroying enzyme of influenza C
virus is neuraminate-O-acetylesterase. EMBO Journal 4, 1503 1505.
HERREER, G., Mt~LTHAUP,G., BEYREUTH~R,K. & KEENK, H.-D. (1988).
Serine 71 of the glycoprotein HEF is located at the active site of the
acetylesterase of influenza C virus. Archives of Virology 102, 269-274.
HIRANO, A., AYATA, M., WANe, A. & WON~, T. (1993). Functional
analysis of matrix proteins expressed from cloned genes of measles
virus variants that cause subacute panencephalitis reveals a common
defect in nucleocapsid binding. Journal of Virology 67, 1848-1853.
KANDOLF, R., KLINGEL, K., MERTSCHING, H., CANU, A., HOHENADL,
C., ZELL, R., REIMANN, B., HEIM, A., MCMANtJS, B., FOtJLIS, A.,
SCHUETHEISS, H.-P., ERDMANN, E. & RIECKER, G. (1991). Molecular
studies on enteroviral heart disease: pattern of acute and persistent
infections. European Heart Journal 12, 49-55.
LEVINE, B., HUANG, Q., ISAACS,J., REED, J., GRIFFIN, D. & HARDWICK,
M. (1993). Conversion of lytic to persistent alphavirus infection by
the bcl-1 cellular oncogene. Nature, London 361,739-742.
LUCAS, W., WHITAKER-DOWLING, P., KAIFER, C. & YOUNGNER, J.
(1988). Characterization of a unique protein produced by influenza
A virus recovered from a long-term persistent infection. Virology
166, 620-623.
MANUGUERRA, J.-C., HANNOUN, C., NICOLSON, C. & ROBERTSON, J.
(1993). Genie amplification of the entire coding region of the HEF
RNA segment of influenza C virus. Journal of Virological Methods
41, 59-76.
MARSCHALL, M. & MEIER-EWERT, H. (1993). Studies on the expression
of influenza C viral RNA and proteins in persistently infected
MDCK cells during nonproductive intervals. IXth International
Congress of Virology, Glasgow, Abstract P50-12.
MARSCHALL, M., MOTZ, M., LESER, U., SCHWARZMANN,F., OKER, B.
& WOLF, H. (1989). Hepatitis B virus surface antigen as a reporter of
promoter activity. Gene 81, 109 117.
MARSCHALL, M., BOSWALD,C., SCHULER, A., YOUZBASHI,E. & MEIEREWERT, H. (1993). Productive and non-productive phases during
long-term persistence of influenza C virus. Journal of General
Virology 74, 2019-2023.
MATSUZAKI, M., SUGAWARA,K., ADACHI, K., HONGO, S., NISHIMURA,
H., KITAME, F. & NAKAMURA, K. (1992). Location of neutralizing
epitopes on the hemagglutinin-esterase protein of influenza C virus.
Virology 189, 79-87.
MONTGOMERY, L.B., KAO, C.-Y. Y., VERDIN, E., CAHILL, C. &
MARATOS-FLIER, E. (1991). Infection of a polarized epithelial cell line
with wild-type reovirus leads to virus persistence and altered cellular
function. Journal of General Virology 72, 2939-2946.
OLDSTOYE, M. (1991). Molecular anatomy of viral persistence. Journal
of Virology 65, 6381-6386.
ROBERTSON, J. S. (1979). 5' and 3' terminal nucleotide sequences of the
RNA genome segments of influenza virus. Nucleic Acids Research 6,
3745-3757.
SALVATO,M., BORROW,P., SHIMOMAYE,E. & OLDSTONE,M. (1991).
Molecular basis of viral persistence: a single amino acid change in
the glycoprotein of lymphocytic choriomeningitis virus is associated
with suppression of the antiviral cytotoxic T-lymphocyte response
and establishment of persistence. Journal of Virology 65, 1863 1869.
SANGER, F., NICKLEN, S. & COUESON, A. R. (1977). DNA sequencing
with chain-terminating inhibitors. Proceedings of the National
Academy of Sciences, U.S.A. 74, 5463 5467.
SCHULTZE, B., GROSS, H.-J., BROSSMER,R., KLENK, H.-D. & HERRLER,
G. (1990). Hemagglutinating encephalomyelitis virus attaches to Nacetyl-9-O-acetylneuraminic acid-containing receptors on erythrocytes: comparison with coronavirus and influenza C virus. Virus
Research 16, 185 194.
SHIODA, T., LEVY, J. & CHENG-MAYER, C. (1992). Small amino acid
changes in the V3 hypervariable region of gp120 can affect the T-cellline and macrophage tropism of human immunodeficiency virus type
1. Proceedings of the National Academy of Sciences, U.S.A. 89,
9434-9438.
STROBE, B. • VEASAK, R. (1993). The receptor-destroying enzyme of
influenza C virus is required for entry into target cells. Virology 192,
679 682.
SUBBARAO,K., LONDON, W. & MURPHY, B. (1993). A single amino acid
in the PB2 gene of influenza A virus is a determinant of host range.
Journal of' Virology 67, 1761-1764.
SZEPANSKI,S., GROSS,H., BROSSMER,R., KEENK, H.-D. & HERRLER, G.
(1992). A single point mutation of the influenza C virus glycoprotein
(HEF) changes the viral receptor-binding activity. Virology 188,
85 92.
UMETSU,Y., SUGAWARA,K., NISHIMURA,H , HONGO, S., MATSUZAKI,
M., KITAM~, F. & NAKAMURA,K. (1992). Selection of antigenically
distinct variants of influenza C viruses by the host cell Virology 189,
740-744.
URABE, M., TANAKA, T. & TOBITA, K. (1992). MDBK cells which
survived infection with a mutant of influenza virus A/WSN and
subsequently received many passages contained viral M and NS
genes in full length in the absence of virus production. Archives of
Virology 130, 547~,62.
URABE, M., TANAKA,T., ODAGIRI,T., TASHIRO, M. & TOBITA, K.
(1993). Persistence of viral genes in a variant of MDBK cells after
productive replication of a mutant of influenza virus A/WSN.
Archives of Virology 128, 97 110.
VLASAK, R., MUSTER, T., LAURO, A., POWERS, J. & PALESE, P. (1989).
Influenza C virus esterase: analysis of catalytic site, inhibition and
possible function. Journal of Virology 63, 2056-2062.
WEIS, W., BROWN, J., CUSACK, S., PAULSON,J., SKEHEL,J. & WILEY, D.
(1988). Structure of influenza virus haemagglutinin complexed with
its receptor, sialic acid. Nature, London 333, 42~431.
(Received 10 January 1994; Accepted 7 February 1994)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 14 May 2017 05:31:44