Lactoferrin (LF) is an iron-binding glycoprotein of 692 amino acids in length, and is present in breast milk and mucous secretions [1
]. LF is a multifunctional protein that is involved in iron transportation in the intestines and blood, the inhibition of viral and bacterial infections, the modulation of immunity, and nonspecific immune responses [1
]. The excellent antiviral activity of LF has been shown to prevent several viral infections, particularly ones caused by intestinally and mucosally-transmitted viruses, by binding to viral particles or viral receptors on the host cell membrane [3
]. Positively-charged LF can interact with negatively-charged compounds such as glycosaminoglycan (GAG), thus inhibiting virus-receptor interaction [5
]. In addition to interacting with GAGs, LF prevents viral infections by binding to dendritic cell-specific intercellular adhesion molecule 3-grabbing non-integrin (DC-SIGN) and low-density lipoprotein receptors (LDLR) [6
Dengue is an important mosquito-borne viral disease in tropical and subtropical regions. The responsible pathogen, dengue virus (DENV), consists of four distinctive serotypes: DENV-1, -2, -3, and -4. The infection can be mild or can result in serious clinical presentations, causing the mild dengue fever (DF), the more serious dengue hemorrhagic fever (DHF), or dengue shock syndrome (DSS) [8
]. DENV is an encapsidated and enveloped virus consisting of genome of 11 kb positive-sense, single-stranded RNA that encodes 10 viral proteins, including a capsid, a premembrane/membrane, and an envelope (E) protein; and seven nonstructural proteins, designated NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5 [9
The initial step of viral infection is the most critical: the interaction and subsequent binding between virions and viral receptor(s) on the host cellular membrane [10
]. The DENV E protein has been shown to interact with GAG, DC-SIGN, and other possible receptors on the host cellular membrane [11
]. Highly sulfated GAG, heparan sulfate (HS), and multi-sulfur chemicals (e.g., suramin), interact with the amino acid residues 284 to 310 and 386 to 411 of the DENV E protein, and inhibit DENV infection [11
]. Thus, it has been suggested that the interaction of the dengue virus E protein with highly sulfated GAG on host cells is an important determinant for virus infection [11
]. The C-type transmembrane receptor DC-SIGN has also been shown to be a prominent DENV receptor [13
]. The DENV uses DC-SIGN as a receptor to infect human dendritic cells [12
]. Several carbohydrate-binding compounds, such as mannose-specific plant lectins from Hippeastrum hybrids (HHA), have been shown to block DENV infection by binding either to the virus or to DC-SIGN [14
Mosquito-borne viruses including Sindbis virus and Semliki Forest virus were inhibited by a LF–HS interaction [16
]. Our previous study has shown that bovine lactoferrin (bLF) blocks Japanese encephalitis virus (JEV) infection by binding to HS and LDLR, but not by binding to virus particles [6
]. Effective antiviral drugs are currently unavailable for treating DENV infection. In this study, we investigate the antiviral activity of bLF against DENV in vitro and in vivo. The results show that bLF blocks DENV-2 entry into cells by binding to HS, DC-SIGN, and LDLR, decreasing viral replication, and resulting in a reduction of morbidity in a mouse model.
Bovine LF exhibits antiviral activities by binding to and interacting with viral particles, such as hepatitis C virus, rotavirus, poliovirus, echovirus, enterovirus [3
], or with viral receptors on the cellular membrane, as in the cases of poliovirus, hepatitis B virus, and enterovirus [3
]. In addition, we and other research groups demonstrated that bLF reduces infection of arboviruses such as Sindbis virus, Semliki Forest virus, Toscana virus, and JEV by binding membrane-bound viral receptor candidate(s) [6
]. In the present study, we demonstrate that bLF prevents DENV infection by interfering with the virus entry machinery.
Multiple steps are involved in DENV entry into cells, including receptor binding, endocytosis, penetration, fusion, and uncoating 1 [10
]. Many reports have shown that the bLF blocks virus infection primarily at early steps of virus entry [1
]. Here, our results show that the presence of bLF prior to or during virus infection reduces virus yield and exhibits significant anti-DENV-2 activity. These results are consistent with our previous JEV study showing that bLF inhibits dengue virus infection mainly at the steps of virus attachment [6
]. Also, bLF added after viral attachment inhibited DENV-2, and that might associate with the multiple enzymatic activities of internalized lactoferrin, especially RNase, [25
] or with innate immunity triggered by bLF through a Toll-like receptor [26
In order to understand the mechanism of the inhibition, the IC50
of bLF added at different steps of DENV-2 infection was calculated and compared with those of other arboviruses, namely JEV, ZIKV, and CHIKV [6
] (Table 2
). Overall, bLF added during virus infection was efficient to inhibit arboviruses infection; when added before virus addition, bLF significantly inhibited DENV-2, JEV, and CHIKV, but not ZIKV; and when added after virus addition, bLF significantly inhibited DENV-2, JEV, and ZIKV, but not CHIKV. It seemed that DENV-2 was more sensitive to the inhibitory effect of bLF that inhibited binding/entry and post-entry steps in the DENV-2 life cycle. The inhibitory effect of bLF at the DENV-2 post-entry step requires further investigation.
N it appears at the first toimeHS is the first molecule to have been identified as a possible receptor for DENV-2 [11
]. The internalization of DENV-2 is also mediated by HS [27
]. A previous study demonstrated that bLF binds electrostatically to HS via a direct interaction of negatively-charged HS and positively-charged bLF [16
]. In this study, bLF strongly prevents DENV-2 from binding to the cellular membrane, and also inhibits DENV-2 at the penetration step. Using HS-expressing and deficient CHO cells, our results support that bLF blocks DENV-2 infection by interacting with HS on the cellular membrane.
Bovine LF blocks the interaction between gp120 and DC-SIGN expressed on dendritic cells, and inhibits the process of transmission of human immunodeficiency virus (HIV) from dendritic cells to T cells [7
]. The ion-binding region of bLF was considered important for interacting with DC-SIGN and related to its antiviral effect [7
]. The mosquito-borne DENV initially infects mature human dendritic cells via engagement of receptor DC-SIGN, and viral replication occurs in DC-SIGN-expressing mature dendritic cells [12
]. This study indicated that the interaction between DC-SIGN and bLF plays a role in the inhibitory effect of bLF against DENV-2 infection.
LDLR is also a possible receptor for several viruses of the Flaviviridae
family, such as hepatitis C virus (HCV) and JEV [6
]. The amount of LDLR on the surface of Huh-7 cells was increased at the early stage of DENV infection [30
], and LDL had showed an inhibition against DENV infection [31
]. In the present study, DENV-2 infection was inhibited by chicken anti-LDLR antibody and rLDLR. These results implied that LDLR might be involved in DENV-2 entry. Our previous report indicates that bLF binds to LDLR and inhibits JEV entry into cells [6
]. Here, the anti-DENV-2 effect of bLF was attenuated by the addition of rLDLR. The LDLR are responsible for regulating the extracellular cholesterol concentration; cholesterol-lowering drugs such as statin increase LDLR expression, thereby reducing the serum cholesterol level [32
]. Thus, a comprehensive study on the role of LDLR in DENV-2 infection and a risk assessment study to determine the effect of serum cholesterol level and dengue infection may be required to investigate the impact of cholesterol–lowering drugs on DENV infection.
Here, we demonstrate that bLF blocks DENV infection primarily by binding to several possible receptors for DENV-2. However, the antiviral activities of bLF were varied against DENV-1-4. This observation might reflect that some dengue serotypes may infect cells via serotype-specific mechanisms. Serotype-specific components on cellular membranes regulating the entry of the dengue virus into cells have been identified [33
], namely, the 37/67-kDa high-affinity laminin receptor that has been identified as a serotype-specific receptor for DENV-1 [34
Most anti-DENV agents in development bind to a specific target, such as a viral protein [35
] or a molecule on or among cells [15
]. Here, we demonstrate that bLF acts as an anti-DENV agent by binding to multiple specific targets that might prevent the development of a resistant virus against bLF, but this needs to be proved in a future study. Although bLF might act as a promising anti-DENV agent and orally administered bLF absorbs by intestinal receptor and appears in the blood [36
], several aspects of the anti-DENV activity of bLF need to be characterized comprehensively in the future, including which lobe (N or C lobe) of the bLF molecule exerts anti-DENV activity [22
], the optimal dosage and time (before or after infection) to administer bLF in vivo, and the anti-DENV activity of bLF administered orally. In addition, an additive effect has been indicated between lactoferrin and ribavirin, which is another potential anti-DENV drug [37
4. Materials and Methods
4.1. Viruses and Cell Lines
Dengue viruses were propagated in and harvested from the supernatant of infected-C6/36 cell lines, and stored at −70 °C. Virus strains included DENV-1 Hawaii, DENV-2 16681, DENV-3 H-87, and DENV-4 H-241. Mosquito C6/36 and monkey Vero cells were cultured in minimum essential medium (MEM) (Life Technologies, Carlsbad, CA, USA) with 10% fetal bovine serum (FBS) (Biological industries, Kibbutz, Israel) at 28 or 37 °C, respectively, with 5% CO2. The HS-expressing CHO-K1 and HS-deficient CHO-pgsA745 cells were grown in Ham’s F12 Nutrient Mixture (GIBCO, Gaithersburg, NY, USA) with 10% FBS at 37 °C with 5% CO2. Human monocyte THP-1 cells were cultured in Roswell Park Memorial Institute (RPMI) medium (Life Technologies, Carlsbad, CA, USA) with 10% FBS in 37 °C with 5% CO2.
4.2. Plaque Assay and Plaque Reduction Assay
Plaque assays were used for viral titration, and plaque reduction assays to study the antiviral effect of bLF. The procedures were similar to those described in our previous report, with some modification [6
]. Vero cells were seeded in six-well plates (105
cells/well) and incubated at 37 °C overnight. The cultured Vero cells were infected with 200 µL of serially diluted virus for 1.5 h at 37 °C; the cell plates were rocked every 20 min during infection. Subsequently, each well was covered with 4 mL of 1.1% methyl cellulose in DMEM with 1% FBS and incubated for seven days at 37 °C. Cells were fixed with 10% formaldehyde for 40 min and stained with 0.5% crystal violet overnight. Plaque number was counted and expressed as the plaque forming unit per mL (PFU/mL).
Iron-unsaturated (apo) bLF (Sigma, St. Louis, MO, USA) (purity > 85%) was dissolved in deionized water and diluted with DMEM. In plaque reduction assays similar to the plaque assay, Vero cells were treated with various concentrations of bLF (0, 0.1, 25, 50, 100, 200 μg/mL) at different stages of virus infection, including prior to viral attachment (pre-treatment), during viral infection (co-treatment), after viral attachment (post-attachment), and at all three stages. Plaque number was counted and compared with an untreated virus control.
4.3. Viral binding Assay
Suspended Vero cells were harvested using 5.0 mM ethylenediaminetetraacetic acid (EDTA) in PBS. Cells (106) were incubated with or without 200 μg/mL bLF at 4 °C for one hour. After washing with ice-cold PBS, bLF-treated cells were incubated with DENV-2 at a multiplicity of infection (MOI) of 50 for 1 h at 4 °C and washed three times with ice-cold PBS. The cells were fixed with 0.5% formaldehyde, and stained using the monoclonal antibody 4G2 and fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse at 4 °C for one hour for each procedure. After washing, the membrane-bound DENV-2 was detected and analyzed via flow cytometer (Beckman, Porterville, CA, USA).
4.4. Infectious Center Assay
An infectious center assay was used to study the role of HS in the anti-DENV-2 activity of bLF described in the previous report [6
]. The HS-expressing CHO-K1 and HS-deficient CHO-pgsA745 cells were harvested from cultured flasks via treatment with 5 mM EDTA in PBS. Cells (106
) were mixed with 200 μL/mL bLF for 1 h at 37 °C. After centrifugation and a PBS wash, LF-treated cells were infected with DENV-2 at MOI of 5 for 1 h at 37 °C, and washed three times with PBS. The infected cells were serially diluted ten-fold and seeded on a monolayer of Vero cells in six-well plates at 37 °C for one hour. Then, each well was covered by 4 mL 1.1% methyl cellulose in DMEM with 1% FBS, and incubated at 37 °C for seven days. Cells were then fixed using 10% formaldehyde for 40 min, and stained with 0.5% crystal violet overnight. Plaques were counted and expressed as infectious rate (infection rate = plaques per well/cells per well).
4.5. Blocking Experiments to Investigate the Role of LDLR
Blocking experiments were used to study further the role of LDLR on DENV-2 infection described in a previous report [6
]. Vero cell monolayers in six-well plates were incubated with 5 μg/well of chicken anti-LDLR antibody (Millipore, Billerica, MA, USA) at 37 °C for one hour. Following a wash, the treated cells were infected with approximately 200 PFU of DENV-2 for one hour, and a plaque assay was carried out as described above.
We also treated Vero cell monolayers in six-well plates with recombinant human LDLR at 200 ng/mL (R&D systems, Inc., Minneapolis, MN, USA) and bLF at 200 µg/mL along or together for one hour at 37 °C. After a PBS wash, the bLF and rLDLR-treated cells were infected with 200 PFU of DENV-2 in order to study the attenuation of rLDLR on anti-DENV-2 activity of bLF. The antiviral effects of the treatments were estimated in a protocol similar to the plaque reduction assay.
4.6. Determing the Role of DC-SIGN
The expression of DC-SIGN is inducible in human monocyte-derived THP-1 cells. The THP-1 cells were cultured with recombinant IL-4 (3.0 µg/mL), GM-CSF (1.5 μg/mL), TNF-α (0.3 μg/mL) and ionomycin (3 μg/mL) (all recombinant proteins from PEPROTECH, Rehovot, Israel), in RPMI medium with FBS, according to a previous report [19
]. The culture medium was changed on the third and fifth days. After seven days, the cells were fixed and stained with mouse anti-human DC-SIGN (AbD Serotec, Oxford, UK) and FITC-conjugated goat anti-mouse IgG (Invitrogen, Grand Island, NY USA). The expression of DC-SIGN on THP-1 cells was detected and analyzed via flow cytometer (Beckmen, Los Olivos, CA, USA). The DC-SIGN-expressing and non-expressing THP-1 cells were used to study the role of DC-SIGN in the anti-DENV-2 activity of bLF in an infectious center assay.
4.7. Indirect Immunofluorescence Assay
Vero cells were seeded in the wells of a chamber slide (Millipore, Billerica, MA, USA), cultured at 37 °C overnight, and incubated with a mixture of DENV-2 and bLF at 37 °C for one hour. After one day, the cells were fixed with 4% paraformaldehyde-PBS at room temperature for 20 min and then washed with PBS. After blocking with 3% BSA-PBS at 37 °C for one hour, the cells were stained with anti-DENV-2 mouse hyperimmune ascitic fluid (MHIAF), followed by FITC-conjugated goat anti-mouse IgG (KPL, Gaithersburg, MD, USA) in 1% Evans blue. The images were viewed using an OLYMPUS CKX41. The infection rate (infection rate = total IFA-positive cells/total counted cells) was estimated by counting positive cells under three independent fields (>100 cells/field).
4.8. Mouse Challenge Experiments
The mouse experiment protocol was approved by the Committee on the Ethics of Animal Experiments of National Chung Hsing University (Approval No: 97-76, Date: 24 December 2008), following guidelines from the care and use manual of the National Laboratory Animal Center, Taiwan. Efforts were made to minimize suffering, and the mice were euthanized with 50% CO2 and cervical dislocation. Pregnant BALB/c mice were purchased from the National Laboratory Animal Center, National Science Council, Taiwan. One-day-old suckling mice were inoculated with 20 µL of 25 LD50 (1 × 104 PFU) of DENV-2 (16681) or pre-mixed DENV-2 (25 LD50) with 200 mg bLF by intracranial (i.c.) injection. Signs of illness, including dehydration, hind leg paralysis, and a lack of interest in suckling were observed daily and recorded. These mice were sacrificed at 13 days post-inoculation and the brains were collected for virus titer determination by plaque assay.
4.9. Statistical Analysis
Student’s two-tailed t-test and one-way ANOVA were respectively used to analyze the differences between two groups, and multiple groups by GraphPad Prism v5.01. p < 0.05 was indicated as a statistically difference.