Transactions of the Royal Society of Tropical Medicine and Hygiene (2006) 100, 608614

available at www.sciencedirect.com

journal homepage: www.elsevierhealth.com/journals/trst

REVIEW

Pathogenesis of liver involvement during dengue viral infections

S.L. Seneviratne a,, G.N. Malavige b, H.J. de Silva c
a

Department of Clinical Immunology, Level 7, John Radcliffe Hospital, Oxford OX3 9DU, UK Department of Microbiology, Faculty of Medicine, University of Jayawardenapura, Sri Lanka c Department of Medicine, Faculty of Medicine, University of Kelaniya, Sri Lanka
b

Received 9 August 2005 ; received in revised form 21 October 2005; accepted 21 October 2005 Available online 17 February 2006

KEYWORDS
Dengue; Liver; Transaminases; Immune mechanisms; Clinical observations

Summary The dengue virus can infect many cell types and cause diverse clinical and pathological effects. We describe clinical and experimental observations that suggest that liver involvement occurs during dengue infections, and we outline the possible role played by host immune responses in this process. 2005 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved.

1. Introduction
Dengue is epidemic or endemic in virtually every tropical country. It is the most important viral haemorrhagic fever in the world. Infection may be clinically asymptomatic or give rise to undifferentiated fever, dengue fever (DF), dengue haemorrhagic fever (DHF) or dengue shock syndrome (DSS). The dengue virus, an RNA virus, belongs to the Flaviviridae family, and consists of four serotypes (DEN14). The virus can infect many cell types and cause diverse clinical and pathological effects. Its main effects are on the vascular, muscular and haematological systems. However, both clinical and experimental observations suggest that there is liver involvement during dengue infection. This liver dysfunction could be a direct viral effect on liver cells or be an adverse

consequence of dysregulated host immune responses against the virus. In this mini-review we outline the clinical and experimental observations of liver involvement during dengue infections, and discuss the possible role played by host immune responses in this process.

2. Clinical observations
Clinical evidence of liver involvement in dengue infections includes the presence of hepatomegaly and increased serum liver enzymes. Hepatomegaly is frequent and is commoner in patients with DHF than in those with DF. Several studies document raised serum transaminase levels in dengue infection. Transaminase levels are also higher in DHF/DSS than in DF and tend to return to normal 1421 d after infection. Kuo et al. (1992) evaluated 270 dengue patients and found abnormal aspartate transaminase (AST) and alanine aminotransaminase (ALT) levels in 93.3 and 82.2%, respectively. Most had mild to moderate increases, while levels

Corresponding author. Tel.: +44 1865 225993; fax: +44 1865 225990. E-mail address: suran200@yahoo.co.uk (S.L. Seneviratne).

0035-9203/$ see front matter 2005 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.trstmh.2005.10.007

Liver involvement in dengue infections more than 10 times the normal upper limit were seen in 11.2 and 7.4% of patients. Nimmannitya (1987), investigating 145 dengue patients, found ALT levels to be normal, slightly elevated or signicantly elevated in 74, 18 and 8% of patients, respectively. No mention of AST levels were found in this report. Wahid et al. (2000) studied 50 serologically conrmed cases of dengue (25 cases each of DF and DHF), and found serum AST and ALT levels to be significantly higher in patients with DHF. In addition, Mohan et al. (2000), who evaluated children with dengue (37 cases of DF, 16 with DHF and 8 with DSS), found abnormal transaminases in 96%, with higher levels in DHF/DSS. Of 1585 dengue patients (65% with primary dengue, 91% with DF) studied by Souza et al. (2004), during a dengue epidemic in Rio de Janeiro, Brazil, alterations in AST and ALT were seen in 63.4 and 45% of patients, with 3.8% having transaminase levels >10 times the upper limit of normal. Fulminant hepatic failure complicating severe DHF/DSS has also been documented, and is associated with a poor prognosis (Alvarez and Ramirez-Ronda, 1985; Lawn et al., 2003; Lum et al., 1993; Munasinghe and Rajasuriya, 1967; Nguyen et al., 1997; Subramanian et al., 2005). Nimmannitya et al. (1987) have reported 18 DHF cases with jaundice and encephalopathy, of whom 10 died. Varying abnormalities in liver enzymes appear to be present in most patients with symptomatic dengue infections, but they tend to recover soon (Pancharoen et al., 2002). There does not appear to be chronic liver damage as with the hepatitis B and C viruses. In a subgroup of predominantly DHF/DSS patients, severe liver dysfunction occurs and is a marker of poor prognosis. During some dengue epidemics, greater degrees of liver damage are seen (Ehrenkranz et al., 1971). Although this may be a consequence of different dengue serotypes having varying tissue trophism, this has not been widely studied. Some investigators suggest that liver damage may be potentiated by the intake of drugs (such as acetaminophen and anti-emetics) during the early phase of the illness, but others do not see this (Suvatte et al., 1990). The course appears not to be inuenced by concomitant hepatitis virus infection (Chung et al., 1992). Although hepatitis B virus (HBV) is hyperendemic in parts of South America and in the Far East, no evidence exists that HBV infection acts as a co-factor for hepatic damage in dengue infections. In dengue infections, elevations in serum AST appear to be greater than ALT levels. This differs from the pattern in viral hepatitis, in which ALT levels are usually higher than or equal to AST levels (Gholson et al., 1990), but it is similar to that seen with alcoholic hepatitis. The exact signicance of this pattern seen in dengue is uncertain. It has been suggested that it may be due to excess release of AST from damaged myocytes during dengue infections (Chung et al., 1992), but this has not been formally tested. Simultaneous measurement of muscle isoforms of lactate dehydrogenase and creatinine kinase may help further clarify this observation. The elevated AST levels tend to return to normal more rapidly than ALT levels. This is possibly because AST (12.522 h) has a shorter half-life than ALT (3243 h) (Hawker, 1991). Histological changes reported in the liver in dengue include: microvesicular steatosis, hepatocellular necrosis, Kupffer cell hyperplasia and destruction, Councilman bodies

609 and cellular inltrates at the portal tract (Bhamarapravati, 1989; Burke, 1968). Most reports are based on small numbers of samples obtained from fatal cases. The presence of thrombocytopenia and coagulative dysfunction makes it difcult to obtain samples from others. As such, one is unsure of the degree of changes present in those with milder disease. Steatosis occurs frequently in hepatitis of viral origin and no special signicance can be attributed to this process in dengue infections. Hepatocellular necrosis in dengue generally affects the midzonal area and sometimes the centrolobular area. Reasons for this pattern may be that hepatocytes in this zone are more sensitive to anoxia or the products of an immune response (e.g. cytokines and chemokines) or that the dengue virus preferentially infects cells in this zone. In fact, dengue viral RNA and protein have been detected in midzonal hepatocytes, mostly around necrotic foci. Councilman (acidophilic) bodies correspond to hepatocytes showing the characteristic morphology of apoptosis. Inammatory mononuclear cell inltrates (of varying intensity) are seen in most specimens studied so far. The dengue virus has been isolated from the liver of fatal cases (Burke, 1968; Rosen et al., 1989; Sumarmo et al., 1983). Some nd it the main organ from which virus could be isolated (Bhamarapravati, 1997; Huerre et al., 2001; Rosen et al., 1989). For instance, using mosquito inoculation techniques, DEN-2 or DEN-3 viruses were recovered from the livers of 5 of 17 fatal cases, but rarely from other tissues (Rosen et al., 1989). Using dengue-specic RT-PCR on liver samples from fatal cases, dengue RNA was detected in 11 of 15 (Rosen et al., 1999). Dengue RT-PCR has been done on parafn-embedded samples from autopsies of 10 children with a clinical diagnosis of DHF/DSS 17 years after their death. In 44, 80 and 43% of cases, DEN-2 RNA was detected in the liver, spleen and lymph nodes (Sariol et al., 1999). Dengue viral RNA has also been detected in midzonal hepatocytes of archived parafn-embedded autopsy tissues using an in-situ PCR method (Kangwanpong et al., 1995). Immunohistochemistry and in-situ hybridization have been used to localize dengue antigens in naturally infected human tissues (Jessie et al., 2004), with immunoperoxidase methods suggested to be reliable and specic in diagnosing dengue or yellow fever infection from human archive samples (Hall et al., 1991). Clinical manifestations in severe dengue disease and yellow fever are similar. Viruses causing them are closely related (both are aviruses, transmitted by the same group of vectors). Overall liver pathology in dengue appears to be similar to that observed during the early stages of yellow fever (Bhamarapravati, 1997). However, in yellow fever, liver cell necrosis tends to be more severe and extensive. In addition, immunouorescence patterns in infected HepG2 cells differ. While dengue viral antigens are found as large perinuclear inclusions and small cytoplasmic foci, yellow fever antigens are homogeneously distributed throughout the cytoplasm (Marianneau et al., 1998). DEN-2 infection of HepG2 cells leads to the release of only a small amount of infectious particles and a slight increase in the number of viral antigen-containing cells over time. By contrast, yellow fever viruses replicated to high titres and infected all exposed cells (Marianneau et al., 1998). Furthermore, while dengue virus-infected cells died rapidly by apoptosis, the

610 highly productive yellow fever virus-infected cells did not show early cytopathic changes (Marianneau et al., 1998). It is possible that early apoptosis of infected hepatocytes followed by their rapid clearance by surrounding phagocytic cells may help to limit the spread of the dengue virus. By contrast, the delayed appearance of apoptotic liver cell death in yellow fever may occur too late. At this stage most cells may have already been infected, causing severe liver destruction. The dengue virus is able to replicate in both hepatocytes and Kupffer cells (Huerre et al., 2001). Although dengue viruses can enter human Kupffer cells efciently, replicative infection of these cells appears not to be very efcient (Marianneau et al., 1999). This may be because most viral particles enter such cells by phagocytosis, a process known to lead to viral degradation. In the smaller number of Kupffer cells in which virus replication can occur, the virus probably enters by receptor-mediated endocytosis and fusion. Two phases of Kupffer cell activation have been described. The rst phase occurs shortly after infection with nitric oxide and IFN- production, with a second a few hours later, involving IL-6 and TNF- synthesis (Marianneau et al., 1999). Infection of hepatic cell lines HepG2 and THLE-3 with ChemiVaxTM -DEN14 and their parent viruses, wild-type DEN14 and YF17D, showed signicant differences in growth kinetics (Brandler et al., 2005). The YF17D virus produced higher titres and caused extensive cytopathic effects earlier than ChemiVaxTM -DEN14 or wild-type DEN14 viruses. The lack of growth of chimeric viruses in human hepatic cells suggests that these viruses may be less hepatotrophic than YF17D virus vaccine in humans.

S.L. Seneviratne et al. et al., 1997; Germi et al., 2002). HS is reported to be involved in entry of all four dengue serotypes into HepG2 cells (Thepparit et al., 2004). However, the degree of internalization varied between serotypes. At present, exact mechanisms of interaction (including the nature of molecules that facilitate entry) between the dengue virus and liver cells are poorly dened. Glucose regulated protein 78 (GRP78) was reported to be used by DEN-2 to gain entry into HepG2 cells (a human hepatoblastoma cell line) (Jindadamrongwech et al., 2004). Thepparit and Smith (2004) suggested that the DEN-1 virus (but not the other dengue serotypes) used the 37/67 high-afnity laminin receptor to enter liver cells. This is a non-integrin cell surface molecule expressed on normal human liver cells, with upregulated levels on liver carcinoma cells. Laminin receptor binding appeared to occur both directly or via HSdependent interactions. Interestingly, a similar mechanism has been previously proposed for entry of prion proteins (Gauczynski et al., 2001) and Sindbis viruses (Wang et al., 1992) into cells. Hilgard and Stockert (2000), studying DEN1 binding protein expression using the Huh7 liver cell line, found the virus able to bind two membrane proteins (approximately 33 and 37 kDa in size). It is possible that the 37-kDa protein described in this report is similar to the molecule described by Thepparit and Smith (2004). The permissiveness of a cell to infection by a virus is considered to be due to two factors: the ability of the virus to enter the cell, and the factors within the cell that enable the virus to replicate successfully. Cellular permissiveness to dengue virus infection appears to be modulated by viral serotype, strain and cell type. HepG2 cells show a marked inuence of cell physiology, with cells in the G2 phase of the cell cycle having a higher susceptibility to infection and a higher rate of virus production per cell (Phoolcharoen and Smith, 2004). The link between the cell cycle and the permissiveness of liver cells to infection with the dengue virus may point the way to understanding the greater susceptibility of children towards the more severe forms of dengue infection (Phoolcharoen and Smith, 2004). Under experimental conditions, HepG2 cell binding of DEN-1 and DEN-2 was found to be non-saturable (Suksanpaisan and Smith, 2003), in keeping with results using Huh7 cells (Hilgard and Stockert, 2000). It is suggested there is a degree of cooperation in dengue virus binding onto liver cells. The binding of one virus to a liver cell serves to facilitate binding of further virus particles, thus making it easier for successive particles to bind. It does this is by inducing conformational changes in cell surface binding molecules. Dengue viruses have varying effects on liver cell lines in vitro. DEN-2 were able to attach equally well to ve liver cell lines, but rates of replication and levels of virion production were higher in the differentiated cell lines (Hul7, PLC, M3B and Chang) than in a de-differentiated cell line (HA22T) (Lin et al., 2000b). It is possible that specic cell differentiation factors may play important roles in viral replication within hepatocytes. Studies aimed at identifying such factors may allow us to understand this process better. Heparin is able to inhibit DEN-2 virus invasion of the different liver cell lines (Lin et al., 2002). Dengue viruses have differential susceptibility to heparin inhibition. This property may be used for screening and selecting mutant viruses for use in different animal models.

3. Experimental observations
To infect cells, dengue viruses need rst to attach to host cell surfaces. This attachment process is considered a major determinant of the viral host range and tissue tropism. The dengue viral envelope protein has been implicated as the viral attachment protein (Chen et al., 1996). Following penetration, internalization occurs by either endocytosis or direct fusion. Evidence exists for both receptor-mediated and non-receptor-mediated entry pathways. There is considerable interest in determining the nature of cellular proteins used by different viruses to enter cells. Identication of such proteins may allow us to rationally design small peptide-based molecules that are able to prevent or modify such interactions. In the case of the dengue viruses, some of these molecules include: DC-specic ICAM3 grabbing non-integrin (DC-SIGN) used by the virus to gain entry into monocyte-derived dendritic cells (NavarroSanchez et al., 2003); the Fc receptor used in cases of secondary infection to gain entry into monocytes (Daughaday et al., 1981); and monocytemacrophage complement receptors (CR3) mediating IgM-dependent enhancement of avivirus replication (Cardosa et al., 1983). Several viruses are able to bind heparan sulphate (HS), a ubiquitous glycosaminoglycan found on the surface of cells. HS may act directly as a virus receptor or serve as a low-afnity virus-binding site, allowing the virus to accumulate before transfer to its high-afnity receptor (Chen

Liver involvement in dengue infections RANTES (regulated upon activation, normal T cell expressed and secreted), a CC chemokine has chemotactic activity for T cells, monocytes, natural killer cells and eosinophils. Levels are higher in patients infected with the dengue virus. Correlation between DEN-2 infection and upregulated RANTES gene expression in liver cells has been shown in vitro (Lin et al., 2000a). It has been hypothesized that dengue infection of liver cells upregulates RANTES mRNA expression by oxidant-dependent and -independent pathways. Following dengue infection, cellular apoptosis in the liver has been seen both in vivo and in vitro (Couvelard et al., 1999; Marianneau et al., 1996, 1997). The Councilman bodies are believed to be the remains of cells undergoing apoptosis (Huerre et al., 2001). Dengue virus infection of primary cultures of human Kupffer cells and a hepatoma cell line induces apoptosis, as evidenced by DNA laddering (Marianneau et al., 1997, 1999). Activation of the transcription factor NF-B has been implicated in the induction of apoptosis (Marianneau et al., 1997). NF-B decoys were able to inhibit HepG2 apoptosis, further pointing to the important role played by the NF-B pathway in this process. Modulating this activity may have future therapeutic potential in patients with severe dengue-induced liver dysfunction. Several mechanisms may be involved in dengue-induced liver cell apoptosis. These include direct cytopathic effects of the virus, mitochondrial dysfunction due to low ow hypoxia and the inuence of cellular and humoral immune factors in the liver (as is observed in other types of viral hepatitis). The apoptotic process appears to be independent of p53 (Thongtan et al., 2004). Increased levels of endoplasmic reticulum stress have been speculated to induce apoptosis in dengue, as has been previously proposed with Japanese encephalitis, another member of the Flavivirus genus (Su et al., 2002). Recently, dengue virus-induced TRAIL (tumour necrosis factor-related apoptosis-inducing ligand) expression has been suggested to be partly responsible for causing apoptosis (Matsuda et al., 2005). DEN-2 infection was shown to cause TRAIL promotor activation, whereas a proteosome inhibitor and TRAIL antibody were able to inhibit DEN-2induced apoptosis. An animal model for dengue was established by transplanting a human HepG2 cell line into severe combined immunodecient (SCID) mice, followed by intraperitoneal infection with DEN-2 virus 8 weeks later (An et al., 1999). During the early post-infection stages, high viral titres were found in the liver and serum but not the brain. Paes et al. (2005) have used BALB/c mice to study liver injury following inoculation with a DEN-2 virus isolated from a human patient. Although the mice did not present with clinical signs and survived the infection, hepatic injury was seen in all individuals. Dengue viral antigens were found in the liver. Elevated transaminase levels correlating with a peak of viraemia were noted.

611 attempt to outline its relationship to liver involvement in dengue. Both innate and adaptive host immune responses play important roles in determining the natural history of viral infections. Innate immune responses are induced rapidly and act as rst-line defences until specic adaptive immune responses come into play. Differences in antibody, T-cell and cytokine responses are seen among patients with uncomplicated DF or DHF/DSS. The immune enhancement and viral virulence hypotheses have been put forward to explain the causation of severe dengue disease. Antibody-dependent enhancement (ADE) attempts to explain the observation that individuals experiencing a secondary infection with a heterologous dengue viral serotype have a signicantly higher risk of developing DHF/DSS. During secondary dengue infections, preexisting non-neutralizing antibodies may form complexes with the virus and enhance its uptake and replication in macrophages (Halstead and ORourke, 1977). The viral virulence hypothesis is based on the observation that some dengue viruses have greater epidemic potential than others. These virulent strains replicate faster and to higher concentrations and hence cause higher levels of viraemia. Effective CD4+ and CD8+ T-cell responses have been suggested to play an important role in clearance of acute dengue viraemia. Following primary dengue, both serotypespecic and serotype-cross-reactive memory T cells are formed. On secondary exposure to a different viral serotype, most serotype-cross-reactive CD4+ and CD8+ T cells are able to augment infection by producing various cytokines (Kurane et al., 1990). During dengue infection, monocytes, B cells, T cells and mast cells produce large amounts of cytokines. Serum concentrations of TNF-, IL-2, IL-6 and IFN- are highest in the rst three days of illness while IL-10, IL-5 and IL-4 tend to appear later (Chaturvedi et al., 1999). It has been suggested that predominant Th2 responses (IL-4, IL-5) occur in DHF/DSS, whereas Th1 (IFN-) responses seem to protect against severe infections. Using MHC class I tetrameric complexes loaded with a peptide from the NS3 protein, expansion of CD8+ T cells with relatively low afnity for the currently infecting virus and higher afnity for serotypes presumed to have been encountered in the past were found (Mongkolsapaya et al., 2003). This was thought to result from the detrimental effects of original antigenic sin. It was argued that these inappropriate T cells contribute to immunopathology while doing little to clear the virus. Another explanation suggested for these ndings was that high antigenic load associated with a second dengue infection (due to immune enhancement), may preferentially drive high-afnity T cells into apoptosis, which would in turn increase the frequency of lower-afnity cells (Mongkolsapaya et al., 2003). T-cell responses in dengue infections need to be well regulated, to avoid specic downsides. For instance, in the process of viral elimination, damage to vital target organs may occur. As in the case of the other hepatitis viruses, dengue virus-specic CD4+ and CD8+ T cells may cause liver cell damage by direct cytolytic and/or cytokine-mediated effects. Cytokines produced during the immune response may also have ill effects, such as promoting the excessive deposition of extracellular matrix.

4. Immunopathogenic mechanisms
An excellent review on current knowledge of dengue pathogenesis has been recently published (Stephenson, 2005). Therefore, in this article, we will only describe salient features of the immune response during dengue infections and

612 Bhamarapravati et al. (1967) examined the liver pathology of 100 fatal paediatric DHF cases (aged 5 months to 14 years) and reported that cellular inltration was noted in 64 cases. Lymphocytoid cells, megakaryocytes and rarely neutrophils were observed in the sinusoids. Cellular inltration around the portal tract comprised lymphocytes, plasmacytoid cells and some histiocytes. In a recent study by Huerre et al. (2001), in ve fatal paediatric cases (aged 10 months to 6 years), little or no inammatory inltrate was found in four cases and a moderate periportal inltrate in one. Chen et al. (2004) infected immunocompetent C57BL/6 mice with high titres of DEN-2 strain 16681 and studied lymphocyte activation and hepatic cellular inltration. T cells in the infected mice were found to be activated and functionally active, as evidenced by the production of IFN-. Most of the activated cells were CD8+ T cells. Liver enzyme elevation and hepatic T-cell inltration were found to coincide with the kinetics of T-cell activation. Hepatic cellular inltrates consisted predominantly of T cells. While most CD4+ T cells clustered around the portal vein, most CD8+ T cells were scattered in the hepatic acinus. Flow cytometric analysis of liver inltrating T cells showed that nearly two-thirds were CD8+ T cells. Gagnon et al. (1999) have postulated that CD4+ cytotoxic T cells (CTLs) may mediate liver damage in dengue through a mechanism involving bystander lysis. CD4+ T-cell-mediated cytotoxicity is thought to occur via two main pathways: release of perforin and granzymes from the activated CTL or the interaction of Fas ligand on T cells with Fas on the target cell. The Fas/Fas-L pathway could contribute to the destruction of cells presenting viral antigens as well as nonantigen-presenting bystander cells that express Fas. Dengue viral capsid-protein-specic CD4+ T-cell clones were able to mediate bystander lysis of HepG2 and Jurkat cells. It has been suggested that dengue virus-specic CTL may be activated by dengue virus-infected Kupffer cells, and in turn lyse hepatocytes via a bystander mechanism. However, such bystander lysis has still not been demonstrated in vivo. At present, the part played by the host immune response in liver damage is unclear. Differences in the intensity of mononuclear cell inltrates in the liver have been reported in human and animal studies. Although it has been suggested that cellular inltrates are lower following arboviral infections than following hepatitis B or C infection, reports in human and animal models of dengue suggest signicant inltrates may occur. It is known that activated viral-specic T cells are able to bind specic receptors on virus-infected cells and induce their apoptosis. It is possible that some T cells that enter the liver may cause damage and either undergo apoptosis or leave the liver. They may also produce a range of cytokines/chemokines, and cause damage to target organs (such as the liver) at a distance from the site of immune activation.

S.L. Seneviratne et al. antigens have been found within hepatocytes, and the virus appears to be able to replicate in both hepatocytes and Kupffer cells, and dysregulated host immune responses may play an important causative role in liver damage. Modulating these immune responses may have a therapeutic potential. There are limitations in the investigation of liver involvement in dengue infection. Immunopathological lesions in the liver are difcult to study in patients with thrombocytopenia and coagulative dysfunction; present knowledge is based mainly on post-mortem specimens and animal models. Conicts of interest statement The authors have no conicts of interest concerning the work reported in this paper.

References
Alvarez, M.E., Ramirez-Ronda, C.H., 1985. Dengue and hepatic failure. Am. J. Med. 79, 670674. An, J., Kimura-Kuroda, J., Hirabaashi, Y., Yasui, K., 1999. Development of a novel mouse model for dengue virus infection. Virology 263, 7077. Bhamarapravati, N., 1989. Hemostatic defects in dengue hemorrhagic fever. Rev. Infect. Dis. 11 (Suppl. 4), S826S829. Bhamarapravati, N., 1997. Pathology of dengue infections, in: Gubler, D.J., Kuno, G. (Eds), Dengue and Dengue Haemorhagic fever. Cambridge University Press, Cambridge, pp. 115132. Bhamarapravati, N., Tuchinda, P., Boonypaknavik, V., 1967. Pathology of Thailand haemorrhagic fever: a study of 100 autopsy cases. Ann. Trop. Med. Parasitol. 61, 500510. Brandler, S., Brown, N., Ermak, T.H., Mitchell, F., Parsons, M., Zhang, Z., Lang, J., Monath, T.P., Gurakhoo, F., 2005. Replication of chimeric yellow fever virus-dengue serotype 14 virus vaccine strains in dendritic and hepatic cells. Am. J. Trop. Med. Hyg. 72, 7481. Burke, T., 1968. Dengue haemorrhagic fever: a pathological study. Trans. R. Soc. Trop. Med. Hyg. 62, 682692. Cardosa, M.J., Portereld, J.S., Gordon, S., 1983. Complement receptor mediates enhanced avivirus replication in macrophages. J. Exp. Med. 158, 258263. Chaturvedi, U.C., Elbishbishi, E.A., Agarwal, R., Raghupathy, R., Nagar, R., Tandon, R., Pacsa, A.S., Younis, O.I., Azizeh, F., 1999. Sequential production of cytokines by dengue virus-infected human peripheral blood leukocyte cultures. J. Med. Virol. 59, 335340. Chen, H.S., Lai, S.Y., Sun, J.M., Lee, S.H., Lin, Y.C., Wang, W.K., Chen, Y.C., Kao, C.L., King, C.C., Wu-Hsieh, B.A., 2004. Lymphocyte activation and hepatic cellular inltration in immunocompetent mice infected by dengue virus. J. Med. Virol. 73, 419431. Chen, Y., Maguire, T., Marks, R.M., 1996. Demonstration of binding of dengue virus envelope protein to target cells. J. Virol. 70, 87658772. Chen, Y., Maguire, T., Hileman, R.E., Fromm, J.R., Esko, J.D., Linhardt, R.J., Marks, R.M., 1997. Dengue virus infectivity depends on envelope protein binding to target cell heparan sulfate. Nat. Med. 3, 866871. Chung, H.K., Dar, I.T., Chi, S.C., Chi, K.L., Shue, S.C., Yun, F.L., 1992. Liver biochemical tests and dengue fever. Am. J. Trop. Med. Hyg. 47, 265270. Couvelard, A., Marianneau, P., Bedel, C., Drouet, M.T., Vachon, F., Henin, D., Deuble, V., 1999. Report of a fatal case of dengue infection with hepatitis: demonstration of dengue antigens in hepatocytes and liver apoptosis. Hum. Pathol. 30, 11061110. Daughaday, C.C., Brandt, W.E., McCown, J.M., Russell, P.K., 1981. Evidence for two mechanisms of dengue virus infection of

5. Conclusion
In summary, clinical and experimental observations suggest that liver involvement occurs during dengue infections. Clinical evidence includes hepatomegaly and increased serum liver enzymes, with liver involvement being more pronounced in the more severe forms of infection. Dengue viral

Liver involvement in dengue infections

adherent human monocytes: trypsin sensitive virus receptors and trypsin-resistant immune complex receptors. Infect. Immun. 32, 469473. Ehrenkranz, N.J., Ventura, A.K., Cuadrado, R.R., Pond, W.L., Porter, J.E., 1971. Pandemic dengue in Caribbean countries and the southern United States past, present and potential problems. N. Engl. J. Med. 285, 14601469. Gagnon, S.J., Ennis, F.A., Rothman, A.L., 1999. Bystander target cell lysis and cytokine production by dengue virus-specic human CD4+ cytotoxic T-lymphocyte clones. J. Virol. 73, 36233629. Gauczynski, S., Peyrin, J.M., Haik, S., Leucht, C., Hundt, C., Rieger, R., Krasemann, S., Deslys, J.P., Dormont, D., Lasmezas, C.I., Weiss, S., 2001. The 37-kDa/67-kDa laminin recepor acts as the cell-surface receptor for cellular prion protein. EMBO J. 20, 58635875. Germi, R., Crance, J.M., Garin, D., Guimet, J., Lortat-Jacob, H., Ruigrok, R.W., Zarski, J.P., Drouet, E., 2002. Heparan Sulfate mediated binding of infectious Dengue virus type 2 and yellow fever virus. Virology 292, 162168. Gholson, C.F., Provenza, J.M., Bacon, B.R., 1990. Hepatologic considerations in patients with parenchymal liver disease undergoing surgery. Am. J. Gastroenterol. 85, 487496. Hall, W.C., Crowell, T.P., Watts, D.M., Barros, V.L., Kruger, H., Pinheiro, F., Peters, C.J., 1991. Demonstration of yellow fever and dengue antigens in formalin-xed parafn-embedded human liver by immunohistochemical analysis. Am. J. Trop. Med. Hyg. 45, 408417. Halstead, S.B., ORourke, E.J., 1977. Antibody-enhanced dengue virus infection in primate leukocytes. Nature 265, 739741. Hawker, F., 1991. Liver dysfunction in critical illness. Anaesth. Intensive Care 19, 165181. Hilgard, P., Stockert, R., 2000. Heparan sulfate proteoglycans initiate dengue virus infection of hepatocytes. Hepatology 32, 10691077. Huerre, M.R., Lan, N.T., Marianneau, P., Hue, N.B., Khun, H., Hung, N.T., Khen, N.T., Drouet, M.T., Huong, V.T., Ha, D.Q., Buisson, Y., Deubel, V., 2001. Liver histopathology and biological correlates in ve cases of fatal dengue fever in Vietnamese children. Virchows Arch. 438, 107115. Jessie, K., Fong, M.Y., Devi, S., Lam, S.K., Wong, K., 2004. Localization of dengue virus in naturally infected human issues, by immunohistochemistry and in situ hybridization. J. Infect. Dis. 189, 14111418. Jindadamrongwech, S., Thepparit, C., Smith, D.R., 2004. Identication of GRP 78 (BiP) as a liver cell expressed receptor element for dengue virus serotype 2. Arch. Virol. 149, 915927. Kangwanpong, D., Bhamarapravati, N., Lucia, H.L., 1995. Diagnosing dengue virus infection in archived autopsy tissues by means of the in-situ PCR method: a case report. Clin. Diagn. Virol. 3, 165172. Kuo, C.H., Tai, D.I., Chang-Chien, C.S., Lan, C.K., Chiou, S.S., Liaw, Y.F., 1992. Liver biochemical tests and dengue fever. Am. J. Trop. Med. Hyg. 47, 265270. Kurane, I., Innis, B.L., Nimmannitya, S., Nisalak, A., Rothman, A.L., Livingston, P.G., Janus, J., Ennis, F.A., 1990. Human immune responses to dengue viruses. Southeast Asian J. Trop. Med. Public Health 21, 658662. Lawn, S.D., Tilley, R., Lloyd, G., Finlayson, C., Tolley, H., Newman, P., Rice, P., Harrison, T.S., 2003. Dengue haemorrhagic fever with fulminant hepatic failure in an immigrant returning to Bangladesh. Clin. Inf. Dis. 37, 14. Lin, Y.L., Liu, C.C., Chuang, J.I., Lei, H.Y., Yeh, T.M., Lin, Y.S., Huang, Y.H., Liu, H.S., 2000a. Involvement of oxidative stress, NF-IL-6, and RANTES expression in Dengue-2-virus-infected human liver cells. Virology 276, 114126. Lin, Y.L., Liu, C.C., Lei, H.Y., Yeh, T.M., Lin, Y.S., Chen, R.M., Liu, H.S., 2000b. Infection of ve human liver cell lines by dengue-2 virus. J. Med. Virol. 60, 425431.

613
Lin, Y.L., Lei, H.Y., Lin, Y.S., Yeh, T.M., Chen, S.H., Liu, H.S., 2002. Heparin inhibits dengue-2 virus infection of ve human liver cell lines. Antiviral Res. 56, 9396. Lum, L.C., Lam, S.K., George, R., Devi, S., 1993. Fulminant hepatitis in dengue infection. Southeast Asian J. Trop. Med. Public Health 24, 467471. Marianneau, P., Megret, F., Olivier, R., Morens, D.M., Deubel, V., 1996. Dengue 1 virus binding to human hepatoma HepG2 and simian Vero cell surfaces differs. J. Gen. Virol. 77, 25472554. Marianneau, P., Cardona, A., Edelman, L., Deubel, V., Despres, P., 1997. Dengue virus replication in human hepatoma cells activates NF-kappaB which in turn induces apoptotic cell death. J. Virol. 71, 32443249. Marianneau, P., Steffan, A.M., Royer, C., Drouet, M.T., Kirn, A., Deubel, V., 1998. Differing infection patterns of Dengue and Yellow fever viruses in a human Hepatoma cell line. J. Infect. Dis. 178, 12701278. Marianneau, P., Steffan, A.M., Royer, C., Drouet, M.T., Jaeck, D., Kirn, A., Deubel, V., 1999. Infection of primary cultures of human kupffer cells by dengue virus: no viral progeny synthesis, but cytokine production is evident. J. Virol. 73, 52015206. Matsuda, T., Almasan, A., Tomita, M., Tamaki, K., Saito, M., Tadano, M., Yagita, H., Ohta, T., Mori, N., 2005. Dengue virus-induced apoptosis in hepatic cells is partly mediated by Apo2 ligand/tumour necrosis factor-related apoptosis-inducing ligand. J. Gen. Virol. 86, 10551065. Mohan, B., Patwari, A.K., Anand, V.K., 2000. Hepatic dysfunction in childhood dengue infection. J. Trop. Pediatr. 46, 4043. Mongkolsapaya, J., Dejnirattisai, W., Xu, X.N., Vasanawathana, S., Tangthawornchaikul, N., Chairunsri, A., Sawasdivorn, S., Duangchinda, T., Dong, T., Rowland-Jones, S., Yenchitsomanus, P.T., McMichael, A., Malasit, P., Screaton, G., 2003. Original antigenic sin and apoptosis in the pathogenesis of dengue hemorrhagic fever. Nat. Med. 9, 921927. Munasinghe, D.R., Rajasuriya, K., 1967. Hepatitis in dengue-fever. Ceylon Med. J. 12, 222223. Navarro-Sanchez, E., Altmeyer, R., Amara, A., Schwartz, O., Fieschi, F., Virelizier, J.L., Arenzana-Seisdedos, F., Despres, P., 2003. Dendritic-cell-specic ICAM3-grabbing non-integrin is essential for the productive infection of human dendritic cells by mosquito-cell-derived dengue viruses. EMBO Rep. 4, 723728. Nguyen, T.L., Nguyen, T.H., Tieu, N.T., 1997. The impact of dengue haemorrhagic fever on liver function. Res. Virol. 148, 273277. Nimmannitya, S., 1987. Clinical spectrum and management of dengue haemorrhagic fever. Southeast Asian J. Trop. Med. Public Health 18, 392397. Nimmannitya, S., Thisyakorn, U., Hemsrichart, V., 1987. Dengue haemorrhagic fever with unusual manifestations. Southeast Asian J. Trop. Med. Public Health 18, 398406. Paes, M.V., Pinhao, A.T., Bareto, D.F., Costa, S.M., Oliveira, M.P., Nogueira, A.C., Takiya, C.M., Farias-Filho, J.C., Schatzmayr, H.G., Alves, A.M.B., Barth, O.M., 2005. Liver injury and viraemia in mice infected with dengue-2 virus. Virology 338, 236246. Pancharoen, C., Rungsarannont, A., Tisyakorn, U., 2002. Hepatic dysfunction in dengue patients with various severity. J. Med. Assoc. Thai. 85 (Suppl.), 298301. Phoolcharoen, W., Smith, D.R., 2004. Internalisation of the dengue virus is cell cycle modulated in HepG2, but not Vero cells. J. Med. Virol. 74, 434441. Rosen, L., Khin, M.M., Tin, U., 1989. Recovery of virus from the liver of children with fatal dengue: reections on the pathogenesis of the disease and its possible analogy with that of yellow fever. Res. Virol. 140, 351360. Rosen, L., Drouet, M.T., Deubel, V., 1999. Detection of dengue viral RNA by reverse tanscription-polymerase chain reaction in the liver and lymphoid organs but not in the brain in fatal human infection. Am. J. Trop. Med. Hyg. 61, 720724.

614
Sariol, C.C., Pelegrino, J.L., Marinez, A., Arteaga, E., Kouri, G., Guzman, M.G., 1999. Detection and genetic relationship of dengue virus sequences in seventeen-year-old parafnembedded samples from Cuba. Am. J. Trop. Med. Hyg. 61, 9941000. Souza, L.J., Alves, J.G., Nogueira, R.M.R., Neto, C.G., Bastos, D.A., da Siva Siqueira, E.W., Souto Filho, J.T.D., Cezario, T.A., Soares, C.E., Carneiro, R.C., 2004. Aminotransferase changes and acute hepatitis in patients with dengue fever: analysis of 1585 cases. Braz. J. Infect. Dis. 8, 156163. Stephenson, J.R., 2005. Understanding dengue pathogenesis: implications for vaccine design. Bull. World Health Organ. 83, 308314. Su, H.L., Liao, C.L., Lin, Y.L., 2002. Japanese encephalitis virus infection initiates endoplasmic reticulum stress and an unfolded protein response. J. Virol. 76, 41624171. Subramanian, V., Shenoy, S., Joseph, A.J., 2005. Dengue haemorrhagic fever and fulminant hepatic failure. Dig. Dis. Sci. 50, 11461147. Suksanpaisan, L., Smith, D.R., 2003. Analysis of saturation binding and saturation infection for dengue serotypes 1 and 2 in liver cells. Intervirology 46, 5055. Sumarmo, W.H., Jahja, E., Gubler, D., Suharyono, W., Sorensen, K., 1983. Clinical observations on virologically conrmed fatal

S.L. Seneviratne et al.

dengue infections in Jakarta, Indonesia. Bull. World Health Organ. 61, 693701. Suvatte, V., Vajaradul, C., Laohapand, T., 1990. Liver failure and hepatic encephalopathy in DHF/DSS: a correlation study with acetaminophen usage. Southeast Asian J. Trop. Med. Public Health 21, 694695. Thepparit, C., Smith, D.R., 2004. Serotype-specic entry of dengue virus into liver cells: identication of the 37-kilodalton/67kilodalton high-afnity laminin receptor as a dengue virus serotype 1 receptor. J. Virol. 78, 1264712656. Thepparit, C., Phoolcharoen, W., Suksanpaisan, L., Smith, D.R., 2004. Internalisation and propagation of the dengue virus in human hepatoma (HepG2) cells. Intervirology 47, 78 86. Thongtan, T., Panyim, S., Smith, D.R., 2004. Apoposis in dengue virus infected live cell lines HepG2 and Hep3B. J. Med. Virol. 72, 436444. Wahid, S.F., Sanusi, S., Zawawi, M.M., Ali, R.A., 2000. A comparison of the pattern of liver involvement in dengue hemorrhagic fever with classic dengue fever. Southeast Asian J. Trop. Med. Public Health 31, 259263. Wang, K.S., Kuhn, R.J., Strauss, E.G., Ou, S., Strauss, J.H., 1992. High-afnity laminin receptor is a receptor for Sindbis virus in mammalian cells. J. Virol. 66, 49925001.