The role of antibody in enhancing dengue virus infection
Introduction
Dengue viruses (DENV) are mosquitos-borne single-strand RNA flaviviruses [1] that cause mild dengue fever (DF), severe dengue hemorrhagic fever (DHF), and dengue shock syndrome (DSS) [2]. There are 50–100 million DF cases each year [3], with more than two billion people being at risk in the tropical regions of Africa, Asia, and South America [2].
Dengue virus has four distinct serotypes (DENV 1–4) whose cross-reactive immune responses contribute to increased disease severity following heterologous infections. Generally, primary infections result in either asymptomatic or mild DF disease and, following virus clearance, in life-long immunity against that serotype [4]. Secondary infections with a different serotype are either cleared [4] or induce severe disease with DHF/DSS [5]. It is unclear if infections with two serotypes result in life-long immunity [6] or if tertiary and quaternary infections may occur [7].
The mechanisms responsible for the severity of secondary dengue infections are not completely understood. One hypothesis postulates that cross-reactive antibodies are responsible for the enhancement of the infection, in a mechanism called “antibody-dependent enhancement” (ADE) [8], [9], [10]. ADE works as follows. When a patient is first infected with one dengue strain, the host produces neutralizing antibodies specific to that strain [8], [11]. After the primary infection is eliminated, long-lived antibody producing plasma cells specificfor the first virus strain persist in the body. When infection with a second dengue serotype occurs, antibodies from the primary infection bind the second virus but do not neutralize it. Instead, phagocytes recruited to clear the virus-antibody immune complexes internalize non-neutralized virus and become infected in the process [11]. The antibody-dependent enhancement makes vaccination difficult, as failure to immunize against all strains will expose the population to the risk of more severe infections. A vaccine trial attempting to vaccinate against all four serotypes and use cross-reactivity to the patient’s advantage is ongoing [12].
An alternative hypothesis postulates that CD4 and CD8 T cells specific to the first dengue serotype dominate the cellular immune responses to heterologous virus, leading to a phenomenon called “original antigenic sin” (OAS) [13], [14], [15], [16]. OAS works as follows. During secondary heterologous infection pre-existing lower avidity cross-reactive CD4 and CD8 T cells rather than higher avidity strain-specific CD4 and CD8 T cells dominate the cellular immune responses. This results in inefficient elimination of cells infected with the heterologous virus. Recent studies have questioned the negative effect of OAS suggesting that genetic rather than immunological factors contribute to disease severity [17].
In this study we investigate the role of ADE in enhancing the severity of the disease by developing a mathematical model of host–virus interaction and predicting possible mechanisms that can explain the observed virus expansion and loss during primary and secondary dengue infections. We use the model to determine the role of cross-reactive antibodies during DF and DHF-inducing secondary infections. It has been hypothesized that the effect of ADE results in increased infectivity of target cells [4]. Our model, however, can better explain patient data when the cross-reactive antibodies interfere with the non-neutralizing antibody effects, by reducing the phagocyte-mediated removal of antibody-virus immune complexes.
The paper is structured as follows. In Section 2, we develop a model of primary infection and analyze its long term dynamics. In Section 3, we develop a model of secondary infection that accounts for the role of strain-specific and cross-reactive antibodies in disease enhancement. We provide the asymptotic analysis of the model. In Section 4, we numerically investigate the models. We use the primary infection model to infer unknown virus–host parameters by comparing it to known primary infection facts. We use the secondary infection model to determine the role of antibodies in inducing severe disease by fitting the model to dengue virus data from patients experiencing DF and DHF. We end with a discussion.
Section snippets
Model of antibody responses to primary dengue virus infection
We model the host–virus dynamics during primary infection with a dengue virus. The model describes the interaction between uninfected monocytes, T, infected monocytes, I, dengue virus, V, resting and activated B lymphocytes, B and Ba, plasma cells, P, and antibody, A, as follows. Uninfected monocytes are produced at constant rate s, die at per capita rate dT, and become infected at constant rate β. Infected monocytes die at per capita rate δ > dT due to both virus-induced and
Model of antibody responses to secondary dengue infections
Following primary dengue infection, virus is eliminated and long-lived plasma cells and antibodies specific to the strain persist in the body. This is described in model (1) by the stability of S4, with plasma cells and the antibodies at their carrying capacities, KP and KA. If the patient gets reinfected with the same virus serotype, virus elimination is faster due to the presence of pre-existing antibodies. Without loss of generality, we assume that the first infection has been eliminated
Parameter values
There are on average T0 = 4 × 105 monocytes per ml of blood in humans [22]. Healthy monocytes die at per capita rate dT = 0.01 per day [23]. We assume that monocytes at steady-state before virus detection T0 = s/dT, therefore, s = 4 × 103 per ml per day. We assume there is a small number of infected monocytes are the time of virus detection I0 = 3 × 10−4 per ml. They die at rate δ = 3.5 per day [24]. V0 = 357 RNA per ml, corresponding to the limit of detection [25]. Virus is produced at rate p
Discussion
We developed a mathematical model of antibody responses to dengue primary infection and used it to determine unknown parameters that describe observed host–virus quantitative characteristics, such as high level viremia followed by virus clearance and delayed antibody responses which become detectable after virus resolution. We assumed that antibodies are present at levels below detection at the time of infection and modeled the neutralizing and non-neutralizing effects of antibody on virus
Acknowledgments
R.N.B. and S.M.C. acknowledge support from Virginia Tech startup fund.
References (35)
Dengue
Lancet
(2007)- et al.
Intrinsic antibody-dependent enhancement of microbial infection in macrophages: Disease regulation by immune complexes
Lancet Infect. Dis.
(2010) - et al.
Human antibodies against dengue enhance dengue viral infectivity without suppressing type I interferon secretion in primary human monocytes
Virology
(2011) - et al.
Mathematical model of antiviral immune response I. data analysis, generalized picture construction and parameters evaluation for hepatitis B
J. Theor. Biol.
(1991) - et al.
Sensitivity analysis of a nonlinear lumped parameter model of HIV infection dynamics
Bull. Math. Biol.
(2004) - et al.
Antibodies from patients with dengue viral infection mediate cellular cytotoxicity
J. Clin. Virol.
(2006) - et al.
Flaviviridae Intervirol.
(1985) - et al.
Secondary infection as a risk factor for dengue hemorrhagic fever/dengue shock syndrome: an historical perspective and role of antibody-dependent enhancement of infection
Arch. Virol.
(2013) Assessing the potential of a candidate dengue vaccine with mathematical modeling
PLoS Negl. Trop. Dis.
(2012)- et al.
Neutralizing antibodies after infection with dengue 1 virus
Emerg. Infect. Diseases
(2007)