Download Differing Infection Patterns of Dengue and Yellow Fever Viruses in a

1270
Differing Infection Patterns of Dengue and Yellow Fever Viruses in a Human
Hepatoma Cell Line
Philippe Marianneau, Anne-Marie Steffan, Cathy Royer,
Marie-Thérèse Drouet, André Kirn, and Vincent Deubel
Unité des Arbovirus et Virus des Fièvres Hémorragiques, Institut
Pasteur, Paris, and Laboratoire de Virologie de La Faculté de
Médecine, INSERM U74, Strasbourg, France
Dengue (DEN) and yellow fever (YF) viruses are responsible for human diseases with
symptoms ranging from mild fever to hepatitis and/or hemorrhages. Whereas DEN virus
typically induces only limited foci of necrosis in the liver, YF virus infection is characterized
by devastating lesions. In a human hepatoma cell line (HepG2), the kinetics of DEN and YF
virus replication and release from the cells and the nature of host cell response to viral infection
were compared. DEN virus infection was characterized by the early appearance of intracellular
viral antigens, major ultrastructural cytopathic changes as early as 32 h after infection, extensive apoptotic cell death, and a low production of infectious particles. In contrast, YF
virus grew exponentially to high titers and induced cytopathic changes only 72 h after infection.
Differences between the infection processes of the two viruses observed in the hepatoma cell
line may explain the different liver pathologies.
Yellow fever (YF) and dengue (DEN) viruses are members
of the flavivirus genus in the family Flaviviridae. They are positive-strand RNA viruses associated with a capsid protein and
have an envelope assembled by budding through intracellular
membranous structures [1]. These agents cause the most significant arthropodborne viral diseases in tropical countries and
are responsible for infections with a large diversity of manifestations, from mild fever to hepatitis and/or hemorrhage. The
pattern of symptoms depends on both viral and host factors
[2]. In the first stages of infection, viruses may replicate in a
variety of cells in different organs and are subsequently released
into the bloodstream, from which they are taken up by hematophagous vectors. Although the majority of flaviviruses
have the capacity to cross the blood-brain barrier in natural
infections, the tropism of YF and DEN viruses may have
changed during evolution, from nervous tissue to reticuloendothelial cells [3], with the liver as the ultimate target organ [4].
The hepatic pathology of these two viruses is different. In the
course of YF infection, although the earliest cytopathologic
changes and viral antigens appear in Kupffer cells, the hepatic
parenchyma is the main target. The hepatocellular damage is
marked by cell degeneration in the mesolobular and midzonal
regions of the hepatic lobules, corresponding to coagulative
necrosis of hepatocytes with microvacuolar fatty infiltrations
Received 15 April 1998; revised 6 July 1998.
Financial support: Direction des Recherches Etudes et Techniques (grant
95.1378). P.M. is supported by a grant from the Fondation Roux, Institut
Pasteur, Paris.
Reprints or correspondence: Dr. Vincent Deubel, Unité des Arbovirus et
Virus des Fièvres Hémorragiques, Institut Pasteur, 25, rue du Dr Roux,
75724 Paris Cedex 15, France ([email protected]).
The Journal of Infectious Diseases 1998; 178:1270–8
q 1998 by the Infectious Diseases Society of America. All rights reserved.
0022-1899/98/7805-0006$02.00
[5]. In DEN virus disease, the hepatic lesions resemble those
of the early stage of YF virus infection but are less severe, with
central and paracentral focal necrosis, hypertrophy of Kupffer
cells, mild fatty changes, and patchy portal mononuclear cell
infiltration [6]. Sinusoidal acidophilic Councilman bodies,
which may correspond to apoptotic hepatocytes [7], are observed in both infections [2]. It is not known whether the differences in the extent of pathologic changes induced by these
two viruses in the liver are due to different cytopathic events
in the hepatocytes. In YF virus infection, damage to hepatocytes is presumably a direct result of viral infection, as these
cells are the major site of viral replication [8, 9]. The target cells
for DEN virus have not been clearly identified [8, 10], although
the presence of virus-specific antigens and RNA inside hepatocytes suggests that these cells may also be susceptible to infection by DEN virus [11] (Couvelard A, Marianneau P, unpublished data). We previously demonstrated that a human
hepatoma cell line, HepG2, which conserved numerous characteristics of differentiated hepatocytes [12], is permissive for
DEN virus infection and that infected cells died by an apoptotic
mechanism as a direct consequence of the viral infection and
not of a bystander effect [13].
To determine whether the differences in the extent of hepatic
lesions associated with DEN and YF virus infections can be
attributed to differences in the replication strategy of the viruses
in the hepatocytes or to immunopathologic mechanisms [14],
we compared infections of HepG2 cells with low-passaged clinical isolates of the two viruses. We investigated differences in
the kinetics of appearance of viral antigen–containing cells, in
the production of infectious virus particles and the propagation
of the infection in cell culture, and in the strength and rapidity
of both the cytopathic effect and the induction of apoptosis.
Differences in virus morphogenesis and in subsequent cyto-
JID 1998;178 (November)
HepG2 Cell Infection by DEN and YF Viruses
pathic effects were also sought in an ultrastructural study of
HepG2 cells infected with the two viruses.
Materials and Methods
Cells and Viruses
Human hepatoma cell line HepG2 (American Type Culture Collection, Rockville, MD) was cultured at 377C in Dulbecco’s modified Eagle medium (DMEM; Gibco BRL, Gaithersburg, MD) supplemented with 10% heat-inactivated fetal calf serum (FCS) and 4
mM glutamine. Mosquito Aedes pseudoscutellaris AP61 cells were
propagated in monolayers at 287C in Leibovitz L-15 medium
(Gibco BRL) supplemented with 0.55% tryptose phosphate and
10% FCS.
YF virus strain DakHD 78359 was recovered on AP61 cells from
the serum of a child, collected 1 day before his death in North
Cameroon in 1990 [15]. The virus isolate was used at its fourth
passage on AP61 cells. DEN serotype 1 virus OSTER was recovered on Aedes albopictus C6/36 cells from a serum sample collected
from a patient in French Guiana in 1989 and was used at its third
passage [16, 17]. Virus stocks were prepared and titrated on AP61
cells as described previously [16, 17]. The titers of YF and DEN
viruses were 4 3 10 6 and 5 3 10 7 focus-forming units (ffu)/mL.
DEN virus was purified from supernatants by polyethylene glycol
precipitation and sucrose gradient centrifugation for electron microscopy experiments requiring a high MOI [16, 17].
Virus Production and Indirect Immunofluorescence Assay (IFA)
HepG2 cells were grown in an 8-microchamber Lab-Tek slide
(Nunc, Roskilde, Denmark) for 24 h. Cells were infected with YF
and DEN viruses at various MOIs from 1 to 50 ffu/cell. At various
times after infection, supernatants were collected for virus titration,
and cells were fixed with 3% paraformaldehyde in PBS. Viral antigens in infected cells were detected by IFA as previously described,
using DEN- or YF-specific hyperimmune mouse ascitic fluids (1:
100) and fluorescein-conjugated anti-mouse IgG (1:100) [16, 17].
Transmission Electron Microscopy
HepG2 cell monolayers were grown in Lab-Tek microchambers
and infected with YF virus at an MOI of 20 ffu/cell or with DEN
virus at MOIs of 20 and 100 ffu/cell. Cells were fixed with 2.5%
glutaraldehyde in 75 mM sodium cacodylate (pH 7.3) containing
4% sucrose, 1 mM MgCl2, and 1 mM CaCl2, and postfixed in 1%
OsO4. The cells were then dehydrated and embedded in LX112
(Ladd Research Industries, Burlington, VT) and observed under a
microscope (EM410; Philips Electronic Instruments, Eindhoven,
The Netherlands).
Detection of Apoptosis
Propidium iodide staining.
HepG2 cells were grown in 8microchamber Lab-Tek slides for 24 h and infected with various
MOIs of DEN or YF virus in medium containing 2% FCS. Cell
monolayers were incubated for 1 h at 377C, washed six times with
1271
DMEM, and cultured in medium containing 2% FCS. At various
times after infection, the cells were fixed in paraformaldehyde for
20 min, washed three times with PBS, treated with 50 mM NH4Cl
in PBS at room temperature for 10 min, washed three times with
PBS, and finally permeabilized with 0.5% Triton X-100 in PBS at
room temperature for 4 min. Viral antigens in infected cells were
detected by IFA as described above. Cells were then treated with
200 mg/mL RNase A in phosphate buffer (0.2 M, pH 7.0) at 377C
for 1 h and washed three times with PBS, and nuclei were stained
with 20 mg/mL propidium iodide in 0.1% sodium citrate, pH 7.0,
at room temperature for 10 min. The slides were washed in PBS,
mounted in 40% glycerol in PBS, and observed under a fluorescence
microscope (Leitz, Rockleigh, NJ).
Terminal deoxynucleotidyl transferase–mediated dUTP nick-end
labeling (TUNEL) analysis. TUNEL analysis (Boehringer Mannheim, Indianapolis) with fluorescein-dUTP was used to label DNA
strand breaks induced during apoptosis. Cells were treated as described for propidium iodide staining. Cells were permeabilized,
incubated for 30 min at 377C with the TUNEL label, washed with
PBS, and mounted with 40% glycerol in PBS. Fluorescent cells
were counted under a fluorescence microscope.
Detection of DNA fragmentation. Fragmentation of cellular
DNA into characteristic apoptotic ladders was assessed as previously described [18]. HepG2 cells were infected with DEN or YF
virus at an MOI of 20 ffu per cell in DMEM containing 2% FCS,
incubated at 377C for 1 h, washed three times with DMEM, and
cultured in medium containing 2% FCS. At 48 and 72 h after
infection, 107 cells were washed in PBS and lysed in 0.5 mL of lysis
buffer (10 mM Tris-HCl, pH 7.4; 1 mM EDTA, pH 8.0; 0.2% Triton
X-100) containing 100 mg/mL proteinase K, at room temperature
for 1 h. The lysed samples were centrifuged at 15,000 g for 15 min
to pellet the nuclei, and the cytoplasmic DNA was precipitated
from the supernatant with 5 M NaCl in 1 vol of isopropanol at
2207C for 1 h. The DNA precipitates were recovered by centrifugation at 13,000 rpm at 47C for 15 min, air-dried at room temperature for 15 min, and resuspended in sterile water. DNA was
then loaded onto a 1.8% agarose gel in Tris-acetate (40 mM)–
EDTA (0.001 M) buffer.
Results
Kinetics of Viral Multiplication
HepG2 cells were infected with YF and DEN virus at MOIs
of 1, 20, and 50 ffu/cell. Viral antigen–containing cells, detected
by immunofluorescence, were counted at 24, 48, and 72 h after
infection. Twenty-four hours after infection, ∼17%–45% of the
DEN virus–infected cells (figure 1A) displayed speckled fluorescent labeling throughout the cytoplasm, in some cases very
intense in the perinuclear area (figure 2A). At later time points,
the infected cell count never increased above 1.4 times that after
24 h, whatever the MOI used. The maximum infected cell count
was reached 48 h after infection but did not exceed 65% of the
cells present, even with the highest infectious dose of DEN
virus. In the case of YF virus, the kinetics of appearance of
viral antigen–containing cells was different. The number of YF
virus antigen–containing cells was very low 24 h after infection
1272
Marianneau et al.
JID 1998;178 (November)
Figure 1. Analysis of sensitivity of HepG2 cells to infection by DEN and YF virus. HepG2 cells were infected with DEN (solid line) or YF
virus (dashed line) at MOIs of 1 (v), 10 (m), and 20 (m). At various times after infection, % of viral antigen-positive cells was calculated by
immunofluorescence assay (A), and virus particles produced in supernatants were titered on AP61 cells (B). Each point represents mean for 2
determinations in duplicate. Error bars are omitted for clarity.
(∼10 times lower than DEN virus–infected cells), but the proportion of cells infected increased exponentially with time of
infection to reach 100% 72 h after infection with the highest
infectious dose (figure 1A). Fluorescently labeled YF virus proteins appeared homogeneously distributed in the cytoplasm
(figure 2B), and this pattern was also different from that observed in DEN virus–infected cells.
Kinetics of the Production of Infectious Virus Particles
The release of infectious virus particles in the culture medium
was followed by titration on AP61 cells (figure 1B). YF virus
multiplication seemed to be very efficient: The number of in-
fectious particles detected increased with time to a maximum
titer of 9 3 10 5 ffu/mL 48 h after infection with an MOI of 20.
In contrast, the number of infectious DEN virus particles produced was ∼100–1000 times lower.
Detection of Apoptotic Cells
DEN virus–infected HepG2 cells die ∼40 h after infection
by an apoptotic mechanism, directly due to viral replication
and not to a bystander effect [13]. We examined whether DEN
and YF viruses differed in their capacity to induce apoptosis
in hepatocytes. HepG2 cells were infected at various MOIs and
tested by propidium iodide staining or TUNEL analysis to
Figure 2. Immunofluorescent staining of viral antigens. HepG2 cells were infected with DEN virus (A) or YF virus (B) and fixed
24 h after infection. Viral antigens were detected by use of DEN- or YF-specific hyperimmune mouse ascitic fluids.
JID 1998;178 (November)
HepG2 Cell Infection by DEN and YF Viruses
Figure 3. Kinetics of apoptotic cell death of infected HepG2 cells.
HepG2 cells were infected with DEN (solid line) or YF virus (dashed
line) at MOIs of 1 (v), 10 (m), and 20 (m). At various times after
infection, number of apoptotic nuclei was determined by immunofluorescent terminal deoxynucleotidyl transferase–mediated dUTP nickend labeling. Each point represents mean for 2 determinations in duplicate. Error bars are omitted for clarity.
apoptotic nuclei. These two different techniques gave identical
results. For both viruses, no apoptotic cells were detected 24
h after infection (figure 3). Forty-eight hours after infection
with the highest MOI, the number of apoptotic cells in DEN
virus–infected cultures was higher than that in YF virus–infected cultures. The percentage of apoptotic cells increased with the time of infection with both viruses but remained higher whatever the MOI used with DEN virus.
Isolation of DNA from infected cultures at 72 h after infection
confirmed the occurrence of an apoptotic process in cells infected with both viruses and the greater extent of this phenomenon in DEN virus–infected than YF virus–infected HepG2
cells (figure 4).
1273
tures of DEN virus particles sequestered by the invaginating
plasma membrane (figure 5G).
HepG2 infection with DEN virus. One of the most prominent features in DEN virus–infected HepG2 cells was a very
rapid (within 32 h) cytopathic effect. The effect consisted first
in the proliferation and dilatation of the endoplasmic reticulum
(figure 6A, 6B) and second in the formation of huge contiguous
vacuoles invading the whole cytoplasm and forming a network
around the nucleus (figure 6A, 6E). The ultrastructural evidence
for the infection of these cells was the presence of intracellular
virus particles. The frequent continuity between these vacuoles
and the perinuclear cisternae suggested that they were of endoplasmic reticulum origin (figure 6A). A small number of virus
particles were found inside intracytoplasmic vesicles (figure 6C).
Extracellular DEN virus particles were rarely observed. The
ultrastructure of the virus particles associated with cellular debris (figure 6D) seems to have been altered (cf. inset in figure
6C). Apoptotic cells with chromatin condensation and nuclear
fragmentation were visible as early as 48 h after infection and
more numerous 72 h after infection (figure 6F).
HepG2 infection with YF virus. The ultrastructural features
of HepG2 cells infected with YF virus were different from those
observed with DEN virus. At a low magnification, YF virus–infected cells or virus-producing cells were easily recognizable: Numerous electron-dense inclusions, not observed in
DEN virus–infected cells, were scattered throughout the cytoplasm, especially in the perinuclear region (figure 7A). At a
higher magnification, other features typical of flavivirus infection were observed, with proliferation of intracellular membranous structures (figure 7B) and dilated rough endoplasmic
reticulum (figure 7C) into which virus particles were budding.
Numerous nascent YF virus particles were found inside the
rough endoplasmic reticulum (figure 7C, 7E), and sometimes
Ultrastructural Study of Infected HepG2 Cells
Penetration of DEN virus particles. HepG2 cells were infected with highly purified and concentrated virus at an MOI
of 100 for 1 h at 377C and were then fixed for electron microscopy. Numerous DEN virus particles were found attached
to the plasma membrane (figure 5A). They entered the cells via
several different mechanisms. Many single particles seemed to
be internalized via clathrin-coated pits (figure 5B, 5C) or small
uncoated vesicles (figure 5D). Several particles forming aggregates at the cell surface could be engulfed simultaneously by a
mechanism of phagocytosis or macropinocytosis via typical cytoplasmic processes (figure 5E). These particles were found enclosed inside phagosomes (figure 5F). We obtained several pic-
Figure 4. Effect of DEN and YF virus infection on DNA fragmentation. Soluble DNA was extracted from HepG2 cell lysates 72 h after
mock infection (M.I.) or infection at MOI of 20 with indicated virus.
Sizes of DNA markers are indicated.
1274
Marianneau et al.
JID 1998;178 (November)
Figure 5. Ultrastructural study of penetration of DEN virus into HepG2 cells 1 h after infection at MOI of 100. A, Attachment of numerous
DEN virus particles to the membrane (original magnification, 3130,000). B and C, Uptake of single particles via clathrin-coated pits (3130,000).
D, Uptake of DEN virus particle by uncoated vesicle (3170,000). E, Engulfment of several particles by cytoplasmic process (342,500). F, Numerous
DEN virus particles enclosed in intracytoplasmic vacuole (386,000). G, Capture of virus particles by invaginating plasma membrane (arrows)
(368,000).
they formed small crystalline arrays in the cytoplasm (figure
7F). No significant alterations in intensively virus-producing
cells were observed 48 h after infection, although after 72 h,
some infected cells displayed an intense vacuolization and vesiculization of the cytoplasm (figure 7D). Only a few cells in the
apoptotic state were observed (figure 7G).
Discussion
Postmortem analysis reveals that the degree of hepatic lesions
associated with DEN and YF diseases is different. The whole
liver seems to be damaged by an invasive process during YF
virus infection, whereas in DEN infection the lesions are mostly
well-limited foci of necrosis. To evaluate the possibility that the
type and degree of hepatocyte cell damage caused by YF and
DEN viruses could be different, we comparatively analyzed the
viral replication and the response to viral infection in a human
hepatoma cell line, HepG2. Two low-passaged isolates repre-
sentative of YF and DEN serotype 1 virus were used in these
experiments.
Cellular receptors specific for flavivirus envelope glycoprotein
have not been identified on mammalian cells [17, 19]. Although
electron microscopy studies have revealed that receptor-mediated endocytosis of YF virus in Vero cells occurred through
coated vesicles [20], the mechanism by which DEN virus penetrates target cells remains unclear. We showed in this study
that numerous DEN virus particles attached to the plasma
membrane before apparently entering cells by diverse mechanisms. The first mechanism appeared to be receptor-mediated
endocytosis in clathrin-coated vesicles, consistent with the mode
of entry of other flavivirus, including YF, Kunjin, and West
Nile viruses [20–22]. The second mechanism involved macropinocytosis or phagocytosis, with the engulfment of the viruses
by cytoplasmic processes, as already described for DEN virus
in blood monocytes [23]. This clathrin-independent route may
explain the broad spectrum of cells of different origins sensitive
JID 1998;178 (November)
HepG2 Cell Infection by DEN and YF Viruses
1275
Figure 6. Ultrastructural features of HepG2 cells infected with DEN virus at MOI of 20. A, Cytopathic effect observed 48 h after infection,
with extensive proliferation of endoplasmic reticulum (arrows), and enlargement of vacuoles to form network, which are sometimes in continuity
with (*) perinuclear cisternae (original magnification, 360,000). B, Dilatation of endoplasmic reticulum. C, Presence of DEN virus particles inside
intracytoplasmic vesicles 48 h after infection (329,000). Inset: 2 virus particles inside vesicle (arrow) (3135,000). D, Extracellular clumps of DEN
virus particles with altered ultrastructure associated with cellular debris, 48 h after infection (380,000). E, Typical cytopathic effect with huge
vacuoles forming network around nucleus 48 h after infection (34300). F, Characteristic aspect of apoptotic cell 48 h after infection, with
densification and fragmentation of nucleus (39400).
1276
Marianneau et al.
JID 1998;178 (November)
Figure 7. Ultrastructural features of YF virus–infected HepG2 cells 72 h after infection at MOI of 20. A, Typical electron-dense areas corresponding to sites of accumulation of nucleocapsids (original magnification, 313,000). B, Proliferation of membranous structures inside cytoplasm
(arrows) (368,000). C, Enlargement of rough endoplasmic reticulum containing several YF virus particles (357,600). D, Cytopathic effect in
HepG2 cell with intense proliferation of intracytoplasmic vesicles (318,200). E, Higher magnification of YF viruses in cytoplasmic vesicle shown
in figure 4C (3108,000). F, Intracytoplasmic virus particles forming crystalloid area (3108,000). G, Apoptotic cell displaying nuclear fragmentation
and shrinkage of cytoplasm (34300).
JID 1998;178 (November)
HepG2 Cell Infection by DEN and YF Viruses
to DEN and other flavivirus infections in vitro. Direct penetration of the virus particle by fusion with the plasma
membrane, as shown for DEN virus in mosquito cells, was not
observed [23]. The different modes of penetration of DEN virus
into HepG2 cells are thus very similar to those described for
other flaviviruses. The pathway of YF virus penetration into
HepG2 cells was not studied because the titer of purified and
concentrated YF virus was not sufficiently high.
The number of infected cells (viral antigen–containing cells)
was much higher 24 h after DEN than YF infection, suggesting
that DEN virus uses a more efficient pathway of infection,
leading to earlier viral protein synthesis. The patterns of immunofluorescent labeling in infected HepG2 cells were different
and consisted of large perinuclear inclusions and small cytoplasmic foci of DEN virus antigens and of homogeneous YF
virus antigens distributed throughout the cytoplasm. These differences in the cellular localization of the viral antigens have
already been described for DEN virus–infected Vero cells [22]
and YF virus–infected hamster kidney cells [24]. Presumably,
therefore, the different cytoplasmic distributions of the viral
proteins are a consequence of the virus rather than the host
cells.
DEN virus infection of HepG2 cells led to only a small
amount of infectious particles released and a slight increase in
the number of viral antigen–containing cells with the time of
infection. In contrast, YF virus replicated to high titers and
infected all exposed cells. These results are in accordance with
the ultrastructural study, which evidenced that YF virus particles were more frequently observed than DEN virus particles
in the cytoplasm, forming crystalloid areas, and in vesicles. In
contrast, DEN virus–infected Vero cells produced 2–4 log more
virus [17] and showed numerous virus particles inside intracytoplasmic vacuoles or clustered in the cytoplasm in crystalloid
arrays [25]. Furthermore, neither the huge electron-dense intracytoplasmic areas corresponding to the accumulation of nucleocapsids [20] nor the extensive proliferation of membranous
structures observed in YF virus–infected HepG2 cells were obvious with DEN virus. Whereas numerous rough endoplasmic
reticulum–derived vacuoles typical of flavivirus infection in
mammalian and mosquito cells [25] were present in HepG2
cells infected with both viruses, very early striking cytopathologic changes corresponding to an exaggerated dilatation of
endoplasmic reticulum cisternae with a rarification of the cytoplasm were found only in DEN virus–infected HepG2 cells.
Although this virus-induced extended vacuolization does not
correspond to cytologic changes generally occurring during
apoptosis, it is probable that these cells subsequently die by
apoptosis, as they decrease in number between 48 and 72 h
after infection concomitantly with an increase in the number
of apoptotic cells.
We observed that DEN virus–infected cells died rapidly by
apoptosis. The highly productive YF virus–infected cells, in
contrast, did not show early cytopathic changes. It was unclear
1277
whether the defect in the production of mature infectious DEN
virus particles is the cause or the consequence of the early
cytopathic effect and the apoptotic process. However, we have
previously reported that the apoptotic process in DEN virus–infected HepG2 cells could be generated by the accumulation of viral proteins in the endoplasmic reticulum [13]. Any
defect in DEN morphogenesis in HepG2 cells may thus have
two consequences: first, a very low production of infectious
virus particles, and second, the induction of an early apoptotic
process. We are now investigating the cause of the very low
production of DEN virus particles in HepG2 cells. Preliminary
results showing the presence of large amounts of intracellular
viral RNA and proteins reinforce the idea that a defect in viral
morphogenesis is a significant factor.
The results of our studies, showing the release of only small
amounts of infectious DEN virus from infected HepG2 cells
contrasting with the large amounts of YF virus particles produced, allow us to hypothesize on the events in hepatocytes
during the course of natural infection. Our observations may
explain the differences between the pathologic changes caused
in the liver. YF infection is characterized by devastating liver
lesions following extensive growth of the virus in liver cells,
whereas in DEN infection, the low production of viral progeny
by infected hepatocytes could lead to limited foci of infection.
Moreover, the spread of DEN virus infection may be contained
by an early death of infected hepatocytes, the apoptotic bodies
being rapidly cleared from the “necrotic foci” by surrounding
phagocytic cells (Couvelard A, Marianneau P, unpublished
data). In contrast, the delayed appearance of apoptotic cell
death in the liver of YF virus–infected patients may be too late
when most of the target cells have already been infected, and
the result may be severe liver destruction.
Further investigations using the in vitro model of human
hepatoma cells and other virus isolates may allow elucidation
of virus and cellular factors involved in the pathogenesis of
hemorrhagic fever and hepatitis.
References
1. Rice CM. Flaviviridae: the viruses and their replication. In: Fields BN, Knipe
DM, Howley PM, eds. Fields virology. Philadelphia: Lippincott-Raven,
1996:931–59.
2. Monath TP, Heinz FX. Flaviviruses. In: Fields BN, Knipe DM, Howley PM,
eds. Fields virology. Philadelphia: Lippincott-Raven, 1996:961–1034.
3. Monath TP. Yellow fever and dengue—the interactions of virus, vector and
host in the re-emergence of epidemic disease. Semin Virol 1994; 5:133–45.
4. Innis B. Dengue and dengue hemorrhagic fever. In: Porterfield JS, ed. Exotic
viral infections. London: Chapman & Hall Medical, 1995:103–46.
5. Digoutte JP, Cornet M, Deubel V, Downs WG. Yellow fever. In: Porterfield
JS, ed. Exotic viral infections. London: Chapman & Hall Medical, 1995:
67–102.
6. Monath TP. Pathobiology of the flaviviruses. In: Schlesinger S, Schlesinger
MJ, eds. The Togaviridae and Flaviviridae. New York: Plenum Press,
1986:375–440.
1278
Marianneau et al.
7. Kerr JFR, Searle J, Halliday WJ, et al. The nature of piecemeal necrosis in
chronic active hepatitis. Lancet 1979; 2:827–8.
8. Hall WC, Crowell TP, Watts DM, et al. Demonstration of yellow fever and
dengue antigens in formalin-fixed paraffin-embedded human liver by immunohistological analysis. Am J Trop Med Hyg 1991; 45:408–17.
9. Deubel V, Huerre M, Cathomas G, et al. Molecular detection and characterization of yellow fever virus in blood and liver specimens of a nonvaccinated fatal human case. J Med Virol 1997; 53:212–7.
10. Monath TP, Ballinger ME, Miller BR, Salaun JJ. Detection of yellow fever
viral RNA by nucleic acid hybridization and viral antigen by immunocytochemistry in fixed human liver. Am J Trop Med Hyg 1989; 40:663–8.
11. Lucia HL, Kangwanpong D. Identification of dengue virus–infected cells in
paraffin-embedded tissue using in situ polymerase chain reaction and
DNA hybridization. J Virol Methods 1994; 48:1–8.
12. Knowles BB, Searls DB, Aden DP. Human hepatoma-derived cell lines. In:
Chisari FV, ed. Advances in hepatitis research. New York: Masson, 1984:
196–202.
13. Marianneau P, Cardona A, Edelman L, Deubel V, Desprès P. Dengue virus
replication in human hepatoma cells activates NF-kappa B which in turn
induces apoptotic cell death. J Virol 1997; 71:3244–9.
14. Igarashi A. Impact of dengue virus infection and its control. FEMS Immunol
Med Microbiol 1997; 18:291–300.
15. Vicens R, Robert V, Pignon D, Zeller H, Ghipponi PM, Digoutte JP. Epidemic of yellow fever in north Cameroun in 1990: first isolation of yellow
fever virus in Cameroun. Bull World Health Organ 1990; 71:173–6.
16. Desprès P, Frenkiel MP, Deubel V. Differences between cell membrane fusion
17.
18.
19.
20.
21.
22.
23.
24.
25.
JID 1998;178 (November)
activities of two dengue type-1 isolates reflect modifications of viral structure. Virology 1993; 196:209–19.
Marianneau P, Megret F, Olivier R, Morens DM, Deubel V. Dengue 1 virus
binding to human hepatoma HepG2 and simian VERO cell surfaces differs. J Gen Virol 1996; 77:2547–54.
Malorni W, Rivabene R, Santini MT, Donelli G. N-acetylcysteine inhibits
apoptosis and decreases viral particles in HIV-chronically infected U937
cells. FEBS Lett 1993; 327:75–8.
Chen Y, Maguire T, Marks RM. Demonstration of binding of dengue virus
envelope protein to target cells. J Virol 1996; 70:8765–72.
Ishak R, Tovey DG, Hovard T. Morphogenesis of yellow fever 17D in infected
cell cultures. J Gen Virol 1988; 69:325–35.
Gollins SW, Porterfield JS. Flavivirus infection enhancement in macrophages:
an electron-microscopic study of viral cellular entry. J Gen Virol 1985; 66:
469–73.
Ng ML, Lau LC. Possible involvement of receptors in the entry of Kunjin
virus into Vero cells. Arch Virol 1988; 100:199–211.
Hase T, Summers PL, Eckels KH. Flavivirus entry into cultured mosquito
cells and human peripheral blood monocytes. Arch Virol 1989; 104:
129–43.
McGavran MH, White JD. Electron microscopic and immunofluorescent
observations on monkey liver and tissue culture cells infected with the
Asibi strain of yellow fever virus. Am J Pathol 1964; 45:501–17.
Matsumura T, Stoller V, Schlesinger RW. Studies on the nature of dengue
viruses. Structure and development of dengue virus in Vero cells. Virology
1971; 46:344–55.