IMS 4112 and VLP of HBV as Th1 Adjuvants for a Recombinant Protein of HIV-1

Background: Current thinking suggests that vaccination approaches against the HIV-1 should be directed to elicit a Th1 cell-mediated immunity, neutralizing antibodies and/or ADCC mediating antibody responses. Also, experimental evidences suggest that both the systemic and mucosal compartments of the immune system must be stimulated to achieve protection. In this regard, effective adjuvants are necessary. Thus, we developed the multiantigenic vaccine candidate TERAVAC that comprise the recombinant protein CR3 of HIV-1 and the surface (S) and core (C) virus like particles (VLP) of Hepatitis B virus (HBV) as adjuvant to promote a Th1 immune response. On the other side, the new Th1 adjuvant Montanide^® IMS 4112 (IMS 4112), an oil-in-water micro-emulsion was developed by SEPPIC.

Purpose of study: The aim of this work was to compare the adjuvant effect of IMS 4112 versus HBV VLP on the CR3(HIV-1)-specific immune response in Balb/c mice using schedules that combine nasal and/or subcutaneous inoculations.

Results: We found a better Th1 response with HBV VLP but a higher IgG response in vagina with IMS 4112 after i.n. inoculation. In s.c. immunization similar immune responses were detected using both adjuvants. When combining i.n. and s.c. coadministration, the HBV VLP seemed to require less immunization doses to achieve detectable proliferation of CD4⁺ and CD8⁺ lymphocytes in the systemic compartment.

Conclusion: Overall, HBV VLP seem to be a better adjuvant for the recombinant CR3 protein of HIV-1 than the IMS 4112 at the concentration it was used in this study.

Abbreviations

VLP: Virus Like Particles; HBV: Hepatitis B virus; HBsAg (S): Surface Antigen of the HBV; HBcAg (C): Core Antigen of the HBV; i.n. Intranasal; s.c.: Subcutaneous; AlOOH: Aluminum Hydroxide Adjuvant.

Introduction

Numerous evidences suggest that long-lasting cell immunity would be necessary to eliminate cells infected with the human immunodeficiency virus type 1 (HIV-1). For instance, the protective role of CD8⁺ CTL responses in the elimination of such cells during primary infection and in controlling virus replication during the chronic phase is well documented [1]. Depletion of CD8⁺ lymphocytes in simian immunodeficiency virus (SIV)-infected rhesus monkeys during chronic infection resulted in a rapid and marked increase in viremia. In this animal model, viral control was achieved after reappearance of SIV-specific CD8⁺ T cells, supporting the importance of these cells in vaccination approaches against HIV-1 [2,3].

Considering that HIV-1 is mainly a mucosal transmitted infectious pathogen, vaccination efforts are re-focused on mucosal routes [4]. Belyakov and colleges reported the induction of HIV-1 specific interferon gamma (IFN-γ) secreting CD8⁺ T cells in mice intestinal mucosa following mucosal, parenteral or heterologous mucosal-parenteral coadministration [5,6]. Additionally, they showed that CD8⁺ T cells mediated preservation of mucosal CD4⁺ T cells from depletion in macaques after mucosal challenge with pathogenic Simian/Human Immunodeficiency Virus (SHIV) [6].

Other evidences suggest the significant role of the humoral response (i.e.; antibody-dependent cell-mediated cytotoxicity (ADCC) or serum neutralizing antibodies (Nabs)) in controlling HIV infection [7-9]. Moreover, some authors have suggested that secretion of HIV-specific IgA and IgG in the vaginal mucosa might also play an important role in the prevention of HIV-1 transmission through sexual intercourse [10-14]. Therefore, immunogens to induce HIV-1-specific IgA and IgG antibodies in the genital tract might be considered for prophylactic vaccination against HIV-1. All together, these elements support the development of novel vaccine candidates for parenteral and mucosal inoculation in order to elicit anti-HIV-1 cellular and humoral responses in the systemic and mucosal compartment [15]. In this regard, it is known that simultaneous administration through a mucosal and parenteral route of HIV native proteins [16] or a recombinant chimeric antigen [17] enhanced the overall immune response in comparison with single route immunizations.

Another important piece in the vaccination puzzle is the choice of the appropriate adjuvant. Although during the last years a new generation of compounds has emerged, aluminum salts have been extensively used for more than 60 years for veterinary and human vaccines due to their safety, low production cost and verified adjuvant effect with many antigens [18,19]. In the past decade, SEPPIC, one of the leader enterprises for adjuvant engineering worldwide, developed a new generation of products, registered under the trade name Montanide^® IMS, to obtain vaccine formulations with a good balance between safety and efficiency. These substances ready to dilute oil-in-water micro-emulsions based on the blend of undisclosed GRAS (Generally Recognized as Safe) immuno-stimulatory compounds and excipients. Some of these adjuvants have been administered to animals in veterinary vaccines with promising results [20-22] but their experimental use for human vaccines is almost unexplored. In this regard, IMS 4112 was developed for human vaccines production to elicit a Th1 biased immune response against intracellular pathogens (i.e.; viruses).

A vaccine candidate against HIV-1 must satisfy three requirements: first, HIV-related antigen(s) (recombinant or not) to stimulate a specific B and/or T cell immunity; second, a safe adjuvant formulation to modulate and enhance the immune response to a particular Th profile and third, stimulation of the mucosal compartment. According to these requirements, we produced the recombinant protein CR3. It comprises B-cell, Th and CTL epitopic rich regions from several subtypes B HIV-1 isolates. This recombinant antigen was coadministered with the Hepatitis B Virus (HBV) surface (HBsAg) and nucleocapsid (HBcAg) antigens through the nasal or subcutaneous route to Balb/c mice and we observed the induction of anti-CR3 (HIV) cellular and humoral immune responses. For subcutaneous immunization the antigens were adjuvated in aluminum hydroxide (AlOOH) and nasal administration were carried out with physiological buffers [23]. Furthermore, simultaneous administration of the above mentioned multiantigenic mixture (named TERAVAC) through the nasal and subcutaneous routes to Balb/c mice promoted a Th1-biased CR3 (HIV)-specific cellular and humoral immune responses [17,24]. In this regard, the VLP of HBV behaves as Th1 adjuvants enhancing the specific anti-CR3 (HIV) immune response and as delivery antigens for effective nasal inoculation [23,25].

Considering the high HBV-HIV co-infection rate in some geographic areas, the multiantigenic formulation TERAVAC might elicit protective immunity against both pathogens. However, the production cost of the three antigens associated to such anti-HIV vaccine will increase its price in the market. Because of that reason, we have considered the possibility to replace the HBV antigens with another adjuvant formulation to avoid the extra cost associated with their production. Thus, the aim of the present work was to compare the adjuvant effect of HBV VLP and IMS 4112 on the CR3 (HIV)-specific immune response in schedules through the nasal, subcutaneous and simultaneous nasal-subcutaneous administration.

Materials and Methods

Antigens

The recombinant HIV-1 antigen CR3 is a multiepitopic protein which consist of B cell, Th and CTL epitopic regions comprising T1 and T2 from gp120, an epitope from gp41, another from Vpr, a fragment of p66/p55 (reverse transcriptase [RT]) protein (positions 2663-3109, HIV-1 SF2 provirus), a part of Nef (positions 8516-8818, HIV-1 LAI isolate), and a part of Gag (positions 1451-1696, HIV-1 SF2) [23]. It was purified from E.coli as a pyrogen free product with more than 95% purity (Technological Development Unit, CIGB, Havana, Cuba) [26]. The entire recombinant HBcAg of 183 aminoacids (hereafter referred as C) was expressed in E.coli and purified as described [27]. Recombinant HBsAg subtype Adw2 (hereafter referred as S) was taken from the production process of the Cuban hepatitis B vaccine Heberbiovac HB (CIGB, Havana, Cuba). This protein was expressed in the yeast Pichia pastoris and its purification procedure has been previously reported [28].

Immunizations

Female Balb/c mice, 6-8 weeks old were purchased from CENPALAB (Habana, Cuba). Three different immunization schedules through the intranasal (i.n.), subcutaneous (s.c) or simultaneous (s.c./i.n.) inoculations were carried out as described in Table 1. In all schedules, the animals received 5 µg of each antigen (CR3, C and S) per route diluted in phosphate buffer saline (PBS). For the intranasal inoculation, they were anesthetized by intraperitoneal administration of 10 µL ketamine (50 mg/mL), placed in a supine position and the immunogen dispensed slowly in 50 L (25 µL/nostril) using a pipette tip. Montanide^® IMS 4112 VG PR (SEPPIC, France) can be used in a range between 5 to 50% v/v in aquous phase. It was decided to use at 10% v/v ratio based on the mixed routes of inoculation (s.c./i.n.) targeted in this study. For subcutaneous immunizations, antigens were adjuvated in 1 mg/mL AlOOH (Superfos Biosector A/S, Vedbaek, Denmark) or 10% v/v IMS 4112.

Negative control groups were formulated without CR3 (Table 1; Schedule #1 and #2, groups 1 and 2; Schedule #3, groups 1-3). In all cases, immunogens were prepared the day before inoculations and stored at 4° C. All experiments were conducted according to institutional guidelines.

Biological fluids

Animals were bled by retro-orbital puncture according to Table 1. Sera were collected by centrifugation at 7 800g for 10 minutes (centrifuge 5415C, Eppendorff, Germany)

Vaginal washes were obtained 10 days after the final dose by reflushing 100 µL of sterile PBS with a micropipette. Supernatants were collected by centrifugation as described above and properly stored at -20°C until antibody detection.

Quantification of IFN-γ secreting cells

Ten days after the last doses, the frequency of CR3 (HIV)-specific IFN-γ secreting cells was detected in spleen of mice using an enzyme-linked immunospot (ELISPOT) assay as described [24]. Briefly, nitrocellulose-backed 96-well plates (MultiScreen-IP filter plates MAIPN4550; Millipore, USA) were coated with 100 μL of murine IFN-γ specific mAb AN18 (5 μg/mL; Mabtech Inc., USA) overnight at 4°C. Two dilutions (2 x 10⁵ and 1 x 10⁵) of freshly isolated splenocytes were incubated for 48 h at 37°C in 5% CO₂ in complete medium (RPMI 1640 [Invitrogen GIBCO, UK] supplemented with 10% fetal bovine serum [FBS], 2 mM glutamine, 2 mM sodium pyruvate, and antibiotics) with sterile and non-pyrogenic CR3, at 2.5μg/mL. Wells that contained cells in complete medium or 5 μg/mL Concanavalin A (Con A; Sigma, St. Louis, MO, USA) were used as negative and positive controls respectively. Secondary biotin-conjugated antibody R4-6A2 (0.5 μg/mL; Mabtech) was then added and incubated for 2 h at room temperature. Detection of CR3 (HIV)-specific IFN-γ secreting cells was achieved under a dissection microscope. Results were expressed as the number of spot-forming cells (SFC) per 10⁶ splenocytes after subtracting the spots of negative wells. Values above the mean number of spots in the placebo or negative group plus 3 standard deviations were considered positive.

In some experiments 20x10⁶ of splenocytes at 2 x10⁶ cells/mL where re-stimulated ex vivo with 2.5μg/mL CR3 protein at 37°C in 5% CO₂. After five days, half of the medium was removed and replaced with fresh medium and human recombinant IL-2 (BD Biosciences, USA) at 10 U/mL final concentration. At day seven, the ELISPOT assay was conducted as described above.

CR3 (HIV)-specific T Cells Proliferation

The assay to assess CR3 (HIV)-specific CD4⁺ and CR8 + T cell proliferation was described previously [17]. Briefly, ten days after the fifth co-administration, 20x10⁶ splenocytes were labelled with 5 μM 5,6-carboxyfluorescein diacetate succinimidyl ester (CFSE; Molecular Probes, Eugene, OR, USA) and cultured in the presence of 2.5 μg/mL CR3 and 5 μg/mL Con A for 5 days at 37 °C in 5% CO₂. After incubation, one million cells were stained with 0.2 μg of the CD8a-APC antibody (clone Ly-2) or the CD4-APC antibody (clone L3T4) (eBiosciences, San Diego, CA, USA). The events were acquired using a PAS III flow cytometer (Partec GmbH, Münster, Germany) and the analysis was performed with Flomax v2.4f software (Partec GmbH). One-hundred thousand viable lymphocytes including blasts were gated (gate R1) for the analyses. For each experimental condition we first calculate the percent of CD8⁺ or CD4⁺ T cells that divided after ex vivo stimulation (Q1) relative to the bulk of CD8⁺ or CD4⁺ T cells gated (Q2+Q1). For each group, results are presented as percent of CR3 (HIV)-specific T cell population that divided after ex vivo stimulation with CR3 (CFSE_low) after substracting the percent corresponding to unstimulated cells. We considered a positive response in experimental groups when the percent values were at least twofold higher compared to the corresponding negative control group.

Cytokine secretion in culture supernatants

Cytokine secretion was assessed by a sandwich ELISA in culture supernatants of splenocytes collected after five days re-stimulation with 2.5μg/mL CR3 protein. High binding capacity 96-well plates (Corning Life Sciences, USA) were coated with 50 μL anti-mouse IFN-γ clone AN18 (5 μg/mL; Mabtech Inc., USA) or anti-mouse IL-4 clone 11B11 (5 μg/mL; BD Pharmingen USA) overnight at 4°C in coating buffer (11 mM Na₂CO₃ and 35 mM NaHCO₃, pH 9.6). Then, plates were blocked with 100 μL of working solution (10% FBS in PBS) during 1 h at 37°C. Undiluted samples and the standard curve of recombinant mouse IFN-γ (0.39 ng/mL-50 ng/mL) or recombinant mouse IL-4 (0.95 pg/mL-30 pg/mL) were incubated 2 h at 37°C. Secondary biotin-conjugated anti-mouse IFN-γ, clone R4-6A2 (0.5 μg/mL; Mabtech) or biotin-conjugated anti-mouse IL-4, clone BVD6-24G2 (0.5 μg/mL, BD Pharmingen) was then added and incubated for additional 2 h at 37°C. Peroxidase-labeled streptavidin (Sigma, St. Louis, MO) was added at a 1:1000 dilution for 1 h at 37°C. Finally, colour reaction was developed with the substrate solution (52 mM Na₂HPO₄, 25 mM sodium citrate, o-phenylenediamine [OPD, 1 mg/mL] [Sigma-Aldrich, USA], and 0.1% H₂O₂) for 10 minutes at room temperature. The reaction was stopped with 25 μL of 3 M H₂SO₄. Then, the plates were read at 492 nm in a microplate reader (Sensident Scan 352; Labsystems, Helsinki, Finland). Washes (at least five) with 0.05% Tween 20 in PBS were carried out between each step. Standard curve and all reagents were diluted with working solution. To determine the cytokine concentration in the samples, we calculated the mean absorbance for each set of duplicate standards, controls and samples and subtracted the mean zero standard absorbance from each. Then, the logarithm of the absorbance values were interpolated into a linear log-log regression analysis plotting concentration of cytokine standards versus A₄₉₂. Finally, cytokine concentration was calculated as the antilog of the resulting value. The limit of detection (LoD) was calculated from the standard curve as described by Armbruster and Pry [29] using A₄₉₂ and the formula A₄₉₂ LoD = meanblank + 1.645 × (SDblank + SDlow concentration sample). Then, to obtain the LoD the same procedure as for the samples was followed.

Antibody response

The CR3 (HIV)-specific antibody responses in serum and vaginal fluid were measured by enzyme-linked immunosorbent assay (ELISA) [17]. Plates were coated overnight with 5 μg/mL of CR3 at 4°C in sodium acetate buffer, pH 5.2. After blockage with 2% skim milk in PBS for 1 h at 37°C, vaginal washes or serum samples diluted in 1% skim milk, and 1% Tween 20 in PBS were added for 2 h at 37°C. Rabbit anti-mouse total IgG, IgG₁, IgG_2a and IgA peroxidase conjugates (MP Biomedicals, USA) were incubated for 1 h at 37°C at appropriate dilutions. The reactions were then developed with the OPD substrate solution for 10 min at room temperature. The reaction was stopped with 50 μL of 3 M H₂SO₄. Finally, the plates were read at 492 nm in a microplate reader (Sensident Scan 352; Labsystems, Helsinki, Finland). Washes (at least five) with 0.05% Tween 20 in distilled water were carried out between each step. IgG titer was referred as standard units (SU) calculated by the interpolation of absorbance values at a fixed serum dilution into a linear regression analysis, plotting dilution versus ABS₄₉₂ of the standard curve from CR3-specific monoclonal antibody (CBSSCR.2, CIGB Santi Spiritus, Cuba). The titer was defined as the highest dilution that gave twice the absorbance of the negative control sera diluted 1:100. Titers are given as geometric mean with 95% CI from individual sera.

IgA and IgG responses from vaginal washes corresponded to 1:10 diluted samples. For IgG subclasses, results were expressed as mean ABS₄₉₂ at a fixed serum dilution. Threshold values were calculated as the average plus three standard deviations of the absorbance in the negative control group.

Statistical procedures

GraphPad Prism version 5.0 software (Graph- Pad Software, San Diego, CA, USA) was used to carry out statistical analyses. All titers were transformed to log10 for a normal distribution. For the non-seroconverting sera, an arbitrary titer of 10 was assigned for statistical processing. A p <0.05 was considered statistically significant.

Results

VLP of HBV induced a better CR3 (HIV)-specific Th1 immunomodulation than IMS 4112 adjuvant after nasal inoculation

As previously reported, immunization with the CR3 antigen alone did not induce a detectable Th1 response after i.n. administration [23]. In contrast, immunization of CR3 in a multiantigenic formulation with the VLP of HBV (C and S) through the nasal (in PBS) or subcutaneous route induced a CR3-specific Th1 pattern [23]. In an attempt to evaluate the new adjuvant IMS 4112, we examined the differential effects on the CR3 (HIV)-specific immune response of this compound versus the hepatitis B antigens after i.n. inoculation. After the fifth dose (Table 1; Schedule #1) the spleen of five Balb/c mice per group was aseptically removed and the secretion of IFN-γ and IL-4 and the frequency of IFN-γ secreting cells anti-CR3 (HIV) were determined after several days of in vitro re-stimulation with the protein. We found a higher median of IFN-γ secretion in culture supernatants in the group immunized with CR3+C+S (TERAVAC) (median, 37.6 ng/mL; [range,17-53]) compared to mice immunized with CR3_IMS 4112 (6.7 ng/mL; [2-13]) (p=0.0286; two-tailed Mann-Whitney test). CR3 (HIV)-specific IL-4 secretion was below the cut off value in all experimental groups (Figure 1A).

Figure 1: CR3 (HIV)-specific IFN- and IL-4 response induced by intranasal antigen delivery. Ten days after the fifth immunization the spleens of five mice per group were aseptically removed and cultured individually during 5 days with 2.5 μg/mL of CR3. IFN- and IL-4 secretion in culture supernatant was determined by ELISA (A) and the frequency of IFN- secreting cells was evaluated after re-stimulation by ELISPOT (B). Data represents mean + SD. The dotted (…) and dashed (---) lines represent the cut off value for IL-4 and IFN-, respectively. Different letters indicate statistical significance (p=0.0286; two-tailed Mann-Whitney test).

To further characterize the Th1 response, the frequency of IFN-γ secreting cells was also assessed in splenocytes of individual mice (Figure 1B). Although there was a trend in the group CR3_IMS 4112 to elicit a lower frequency of CR3-specific IFN-γ secreting cells compared to the group CR3+C+S (3 832 vs 6 905 SFC/10⁶ splenocytes, respectively), this difference did not achieve statistical significance (p=0.1143; two-tailed Mann-Whitney test). In any case, this result was in line with previous results of IFN-γ secretion.

Only IMS 4112 adjuvant induce CR3 (HIV)-specific IgG response in vagina after nasal inoculation

It is documented that antigen delivery by the intranasal route induces antigen-specific response at distal mucosal sites. Our group has previously reported that intranasal inoculation with CR3+C+S (TERAVAC) promote anti-C and anti-S, but not a detectable anti-CR3 IgA response in vaginal washes [23]. Here, we analyzed the effect of IMS 4112 on the anti-CR3 (HIV) IgA and IgG response in this organ ten days after the last immunization. Figure 2A shows that approximately 44% of mice immunized with CR3_IMS 4112 were positive for IgG secretion. This response was no detectable in the remaining groups. Furthermore, neither IMS 4112, nor the VLP of HBV could induce a detectable anti-CR3 (HIV) specific IgA response in vagina (Figure 2A). In the systemic compartment, after i.n immunization, none of the experimental groups achieved 100% IgG seroconvertion rate in sera and no statistical differences between CR3+C+S and CR3_IMS 4112 were observed (p=0.4267; two-tailed Mann-Whitney test) (Figure 2B).

Figure 2: Anti-CR3 (HIV) antibody responses after intranasal antigen delivery. Ten days after the fifth dose, vaginal washes and sera were collected. Mean ABS 492nm + SD of tested animals represent the CR3 (HIV)-specific IgA and IgG antibody levels in vagina (A). Serum anti-CR3 (HIV) IgG titers were also determined. Data represent geometric mean + 95 % CI (B). The dotted (…) and dashed (---) lines represent the cut off value for IgA and IgG, respectively. The seroconvertion frequency is indicated above the bars.

Similar Th1 adjuvant effect of IMS 4112 and VLP of HBV after subcutaneous administration

To further evaluate the adjuvant effect after parenteral immunization, we immunized similar experimental groups through the subcutaneous route (Table 1; Schedule #2). In these experiments the multiantigenic formulation CR3+C+S (TERAVAC) was also adjuvated in AlOOH as previously reported [25]. At the end of the immunizations, we measured the CR3 (HIV)-specific cellular immune response.

After three doses, the level of IFN-γ secretion in culture supernatants of CR3-stimulated splenocytes as well as the frequency of positive response was similar among experimental groups (p=0.2403; two-tailed Mann-Whitney test) (Figure 3A). Regarding the frequency of IFN-γ secreting cells measured by ELISPOT after re-stimulation, we performed a meta-analysis of two independent experiments. Figure 3B shows similar frequency of CR3 (HIV)-specific IFN-γ-secreting cells among groups immunized with CR3+C+S_AlOOH or CR3_IMS 4112 (1 130 vs 754 SFC/10⁶ splenocytes, respectively; p=0.7173; two-tailed Mann-Whitney test) (Figure 3B). To additionally characterize the immune response, we examined the CR3 (HIV)-specific IgG₁ and IgG_2a antibody production in sera of mice as a useful parameter to discriminate Th1-Th2 profiles [27,30]. A predominant IgG₁ response was found in all animal groups regardless of the immunogen. A lower IgG_2a secretion was also detected. No statistical differences were found between the CR3-IMS 4112 and the CR3+C+S_AlOOH group in the antibody response (IgG₁ or IgG_2a) (Figure 3C).

Figure 3: Th1 response after subcutaneous delivery of different antigen formulations. Ten days after the 3rd immunization mice spleens were aseptically removed and cultured individually during 5 days with 2.5 μg/mL of CR3. IFN- secretion in culture supernatant was determined by ELISA (A) and the frequency of IFN- secreting cells was evaluated after re-stimulation by ELISPOT (B). Both experiments are represented as mean + SD. The number of positive response is indicated above the bar. Anti-CR3 (HIV) IgG subclass pattern was analyzed in 7 mice per group at fixed serum dilution of 1:1 000 for IgG₁ and 1:100 for IgG_2a (C). The dashed line represents the cut off values.

Additionally, we determined the anti-CR3 (HIV) whole IgG titer in mice sera. A 100% seroconvertion and higher IgG geometric mean titers (8 801 SU) were achieved in the CR3+C+S_AlOOH group. The 90% of CR3_IMS 4112 immunized mice achieved a geometric mean titer of 423 SU, suggesting a better immunoenhancing effect of VLP of HBV on the overall CR3 (HIV)-specific IgG response after subcutaneous administration (p=0.0003; two-tailed unpaired t test with Welch’s correction) (Figure 4).

Figure 4: Whole anti-CR3 (HIV) IgG titer in mice sera 10 days after the 3rd dose of subcutaneous immunization. Data represents geometric mean with 95 % CI from 10 mice per group. The frequency of positive results is indicated above the bars. Letters are associated with statistical differences (p=0.0003; two-tailed unpaired t test with Welch’s correction).

Subcutaneous/intranasal co-administration of the CR3 protein with IMS 4112 and HBV antigens in AlOOH

Mucosal and parenteral co-administration usually enhances the response against a given antigen in comparison with single route schedules [16,31]. According to those reports, we decided to compare the adjuvant effect of hepatitis B VLP and IMS 4112 on the anti-CR3 (HIV) immune response using such schedules as well as the combined effect of both adjuvants (Table 1; Schedule #3). Thus, a group of mice was inoculated with HBV VLP subcutaneously and IMS 4112 by the nasal route. In this case adjuvants were selected considering previous results where HBV VLP provided a similar Th1 adjuvant effect to IMS 4112 but higher IgG levels in sera via s.c. and for the intranasal inoculation the IMS 4112 adjuvant because of the induction of IgG antibodies in vagina.

Figure 5: Frequency of CR3 (HIV)-specific IFN-g secreting cells induced by the s.c./ i.n coadministration regime. Ten days after 5th and 6th immunizations, splenocytes from 3 mice/group were stimulated ex vixo with 2.5 µg/mL of CR3 during 48 hours. Bars represent the frequency of IFN-g secreting cells quantified in pooled samples.

After five immunizations, only the CR3+C+S_AlOOH /CR3+C+S group showed detectable levels of anti-CR3 (HIV) IFN-γ secreting cells in mice spleen measured by ELISPOT assay. An additional sixth dose was required to induce a similar response when CR3 was adjuvated with IMS 4112 or when the combination of adjuvants was tested in the CR3+C+S_AlOOH/CR3_IMS 4112 group (Figure 5). In accordance with the above mentioned results, a detectable proliferative response of 16.5% CD4⁺ and 19.6% CD8⁺ T lymphocytes was only achieved in CR3+C+S_AlOOH/CR3+C+S immunized animals after five doses (Figure 6).

Figure 6: CR3 (HIV)-specific T cell proliferation induced after CR3+C+S_AlOOH/CR3+C+S s.c./i.n. coadministration. Ten days after the fifth dose, the splenocytes from three mice drawn randomly were isolated, stained with CFSE and re-stimulated ex vivo with CR3, concanavalin A (Con A) or complete medium (RPMI) during 5 days. Then, CD4⁺ and CD8⁺ lymphocyte labeling was performed in order to quantify proliferation by flow cytometry. One-hundred thousand viable lymphocytes including blasts (gate R1) were gated for the analyses and represented as dot-plot of CD4⁺APC or CD8⁺APC vs CFSE. For each experimental condition we first calculated the percent of CD8⁺ or CD4⁺ T cells that divided after ex vivo stimulation (Q1) relative to the bulk of CD8⁺ or CD4⁺ T cells gated (Q2+Q1). Then, the percent in RPMI medium was subtracted from the CR3 stimulation to obtain the CR3 (HIV)-specific proliferation (data at the right). In the negative control group C+S_AlOOH/C+S s.c./i.n. background values of 0% and 6.28% were observed for CD8⁺CFSE_low and CD4⁺CFSE_low, respectively.

To further characterize the adjuvant effect on the humoral response, we analyzed the whole anti-CR3 (HIV) IgG titers and subclasses in animal’s sera ten days after the fifth and the sixth dose (Figure 7). Overall, a predominant IgG₁ over an IgG_2a response was detected and no statistical differences were found among groups in relation with IgG subclasses or even doses (Figure 7A,B). Additionally, no statistical differences were observed in whole IgG titers among groups after the 5th (p=0.1070, one-way ANOVA) or the 6th dose (p=0.9189, one-way ANOVA) (Figure 7C).

Humoral CR3 (HIV)-specific IgA and IgG response were also assessed in vagina after the fifth and the sixth dose. Levels of IgA were undetectable in all the experimental groups (data not shown). A low IgG response and similar frequency of responder mice per group were observed without any significant statistical differences after the 5th (p=0.9081, one-way ANOVA) or the 6th dose (p=0.6030, one-way ANOVA) (Figure 7D).

Figure 7: Anti-CR3 (HIV) antibody response in sera and vagina after the 5th and the 6th dose of s.c/i.n. administration. Twelve and nine mice were screened ten days after the 5th and the 6th dose, respectiely. Individual IgG₁ (•) and IgG_2a (○) mean ABS₄₉₂ nm were assessed after the 5th (A) and the 6th dose (B) at fixed serum dilution (1:1 000 for IgG₁ and 1:100 for IgG_2a). Total IgG titers are represented as geometric mean with 95% CI (C). The level of CR3 (HIV)-specific IgG response was assessed in vaginal washes. Frequency of responders is indicated above the bar (D). The dashed line represents the cut off value. No statistical differences were observed neither between doses nor between groups.

Discussion

The purpose of this work was to compare the adjuvant effect of IMS 4112 versus the VLP of HBV after mucosal, parenteral and mucosal/parenteral immunizations of the recombinant multiantigenic protein CR3 of the HIV-1.

Previous studies demonstrate the Th1 adjuvant effect of the mixture of HBcAg and HBsAg on the CR3 (HIV)-specific cellular and humoral immune response after i.n. [23], s.c. [25] and even s.c./i.n. coadministration [24]. In this regard, the experimental evidence pointed out to the nucleic acid (double-stranded RNA, dsRNA) content inside the HBcAg as the primary cause [32]. Additionally, the nanoparticle size of both VLP of HBV, their spontaneous aggregation with CR3 [33] and the lipids content of HBsAg from the P. pastoris host [34], may all together further contribute to the adjuvant effect and effective nasal immunization.

During the last decade, SEPPIC developed a new generation of adjuvants registered under the name of Montanide^® IMS, which comprise a range of ready-to-dilute oil-in-water micro-emulsions. These adjuvants are dispersions of liquid particles, varying in size between 50 to 500 nm, in an aqueous phase containing an immunostimulating compound listed as GRAS substances. One of them, IMS 3012, has proved its advantage for a veterinary vaccine against Rhodococcus equi, inducing IgG2b subclass prevalence and the IFN-γ, IL-2 and IL-10 mRNA expression in vaccinated mares [35]. In a different set of experiments, IMS 3012 was the adjuvant of choice for hyperimmunization of equines for the production of polyvalent snake antivenom sera. In that experiment, 100% of horses in IMS 3012 immunized group responded against different snake toxins and showed minimum local reactions at injection site, demonstrating the safety of this adjuvant in veterinary vaccines [22]. Another group of researchers have studied the immunogenicity of a multi-allele and multi-antigenic vaccine against malaria using IMS 4112 as adjuvant. Their results showed the lowest serum IgG titers in rabbits immunized in the presence of IMS 4112 in comparison with other three experimental adjuvants. This was accompanied by the lowest inhibition of P. falciparum growth by the antibodies elicited in this group [36].

The results obtained in the present work showed a better Th1 adjuvant effect of the VLP of HBV than IMS 4112 after intranasal inoculation. It was evidenced by a significant higher IFN-γ secretion of splenocytes after ex vivo restimulation and a trend to a higher frequency of IFN-γ-secreting cells. However, IMS 4112 seems to be a better inducer of IgG response in vagina. IgA antibodies were not detectable in this organ. There are several explanations for that result. First of all, it is known that IgG antibody production is usually higher than IgA in this organ [37]. Additionally, it is also important to consider that samples of vaginal washes were collected in 0.1 mL and further diluted 1:10 for evaluation and such dilution factor might decrease the IgA concentration below the threshold value of detection of the ELISA assay. It is noteworthy that immunization of TERAVAC stimulates the production of gp120-specific IgG [23,38]. While the physiological significance of such IgG antibodies awaits further analysis, one can speculate that such Ig in the vaginal mucosa might contribute at some level of protection from infection as previously suggested by some authors [13,14]. A potential role of such IgG might be to trap HIV viral particles in the mucus as reported for herpes simplex virus serotype 1 (HSV-1) [39] and HSV-2 [39,40]. On the other hand, after intranasal immunization both adjuvants, IMS 4112 and the VLP of HBV, induced a similar level of IgG production and seroconvertion in the systemic compartment.

In contrast, after parenteral immunization (via s.c) we did not observed any statistical significant differences in relation to the secretion of IFN-γ in culture supernatant, the frequency of IFN-γ secreting cell or the production of IgG_2a antibodies in serum. All these parameters suggest a similar Th1 adjuvant effect in the systemic compartment. But, it is worth of remark that HBV-VLP per se are not able to induce a detectable frequency of IFN-γ secreting cell in the spleen of mice after several doses and AlOOH must be added to achieved a good response (E. Iglesias, unpublished results). Considering that fact, it is possible to speculate that IMS 4112 would have a better Th1 adjuvant effect after parenteral immunization than the mixture of HBV-VLP per se without AlOOH. The combined adjuvant effect of AlOOH and the HBV-VLP explain the significant higher whole IgG titers observed in sera of immunized animals compared to IMS 4112. Previous studies have shown that AlOOH do not stimulate any known Toll-like receptor (TLR) [41,42]. It is well-known that the adjuvant effect of AlOOH is mediated by slow release of the antigens (the so call “depot effect”), better uptake by antigen presenting cells (APC) and Nod-like receptor pyrin containing 3 (NLRP3) inflammasome-dependent and independent signaling pathways [43]. In this regard, AlOOH does not induce a strong commitment of the immune system to either a Th1 or Th2 pathway but it enhances the Th1 effect provided by the HBV VLP in TERAVAC [25].

The results obtained with the subcutaneous immunization seem contradictory with those obtained after intranasal inoculation when a better Th1 effect was reported for the HBV-VLP. Nevertheless, it might be influenced by qualitative/quantitative expression pattern of TLRs between the nasal mucosa and the systemic compartment and at the cellular level depending also on the differentiation stage of APCs [44] among other parameters still under investigation. For instance, a recent report evidenced different outcomes of the immune response elicited by dendritic cells (DC) when a specific TLR stimulation is produced or not at the same time with another TLR on the same APC [45]. In that sense, the immunological outcome promoted by a particular adjuvant formulation (that may stimulate an array of TLRs) will be the result of a complex mesh of interactions influenced by the anatomical localization and cellular state of differentiation. Then, all these features of the innate immunity could have played a role.

Although there are few reports on this subject, it is clear that TLR3 expression in nasal epithelial cells is abundant in humans and mice [46-48]. Additionally, as Asahi-Ozaki and coworkers have shown, intranasal administration in mice of the synthetic dsRNA, poly (I:C) stimulates the up-regulation of TLR3 and TLR7 in the nasal-associated lymphoid tissues (NALT) which in turn promotes activation of the NF-kB pathway and production of type I IFNs [49] which eventually leads to a strong immune response. They did not find a similar stimulation effect in the spleen. This differential up-regulation of TLR3 and 7 between the mucosal and systemic compartment after i.n. administration might explain why the HBcAg (i.e. the mixture of HBV VLP) induced a better Th1 immunodeviation than IMS 4112 after intranasal administration with a similar level of IgG production in sera. Unfortunately, the chemical composition of IMS 4112 has not been disclosed and because of that our analysis are very limited. Additionally, concentration of use of IMS 4112 was arbitrarily fixed at 10% v/v and we consider that a dose escalation of the adjuvant could have also brought further elements for analysis.

When considering the results obtained in the schedule for combined parenteral/mucosal coadministration, it is not possible to conclude superiority of some group over any other. A possible explanation for the fact that only in the CR3+C+S_AlOOH/CR3+C+S was possible to detect anti-CR3 (HIV) IFN-γ-secreting cells and proliferation of CD4⁺ and CD8⁺ lymphocytes when whole IgG titers, IgG₁, IgG_2a in sera or vaginal washes were not different among the groups, might be the low sensitivity of the assays. To reconcile such discrepancy, it is possible to speculate that frequency of CR3 (HIV) IFN-γ-secreting cells and proliferation of CD4⁺ and CD8⁺ lymphocytes in the group of mice immunized with IMS 4112 s.c./i.n. and HBV VLP in alum via s.c. and with IMS 4112 i.n. were near but below the threshold of detection of the assays and the values reported in the group of HBV-VLP (CR3+C+S_AlOOH/CR3+C+S) were above but near the same threshold. In any case, these results suggest a stronger adjuvant effect of the coadministration of VLP than any other combination tested. It is worth of remark that VLP for subcutaneous immunization were formulated with AlOOH.

Conclusion

In summary, our results suggest that HBV-VLP induce a better Th1 adjuvant effect than IMS 4112 at the concentration of 10% v/v after nasal immunization although the last adjuvant seems to be a better inducer of IgG in vagina. In the parenteral inoculation the adjuvant effect of IMS4112 and HBV VLP formulated with AlOOH behave in a very similar way. When using schedules of nasal and subcutaneous coadministration it seems that less inoculation doses are needed to achieve a detectable Th1 response with HBV-VLP. Taken together, these results suggest that HBV VLP are a better adjuvant for the recombinant CR3 protein of HIV-1 than the IMS 4112.

The authors thank Ismariley Revee and Sara Clark for technical assistance in the culture laboratory. We also thank Yalena Amador-Cañizares; M.Sc their critical review of English grammar. This work was financed by the Center for Genetic Engineering and Biotechnology, Havana, Cuba.

1. Ahlers JD, Belyakov IM (2010) New paradigms for generating effective CD8⁺ T cell responses against HIV-1/AIDS. Discov Med 9: 528-537. Link: https://goo.gl/Xx5dz1
2. Schmitz JE, Kuroda MJ, Santra S, Sasseville VG, Simon MA, et al. (1999) Control of viremia in simian immunodeficiency virus infection by CD8⁺ lymphocytes. Science 283: 857-860. Link: https://goo.gl/LSNw3O
3. Sircar P, Furr KL, Dorosh LA, Letvin NL (2010) Clonal repertoires of virus-specific CD8⁺ T lymphocytes are shared in mucosal and systemic compartments during chronic simian immunodeficiency virus infection in rhesus monkeys. J Immunol 185: 2191-2199. Link: https://goo.gl/dBN8qr
4. Demberg T, Robert-Guroff M (2009) Mucosal immunity and protection against HIV/SIV infection: strategies and challenges for vaccine design. Int Rev Immunol 28: 20-48. Link: https://goo.gl/j5dhjC
5. Belyakov IM, Ahlers JD, Nabel GJ, Moss B, Berzofsky JA (2008) Generation of functionally active HIV-1 specific CD8⁺ CTL in intestinal mucosa following mucosal, systemic or mixed prime-boost immunization. Virology 381: 10⁶-115. Link: https://goo.gl/izWZel
6. Belyakov IM, Isakov D, Zhu Q, Dzutsev A, Berzofsky JA (2007) A novel functional CTL avidity/activity compartmentalization to the site of mucosal immunization contributes to protection of macaques against simian/human immunodeficiency viral depletion of mucosal CD4⁺ T cells. J Immunol 178: 7211-7221. Link: https://goo.gl/81wUfr
7. Chung A, Rollman E, Johansson S, Kent SJ, Stratov I (2008) The utility of ADCC responses in HIV infection. Curr HIV Res 6: 515-519. Link: https://goo.gl/xaIATH
8. Bonsignori M, Pollara J, Moody MA, Alpert MD, Chen X, et al. (2012) Antibody-dependent cellular cytotoxicity-mediating antibodies from an HIV-1 vaccine efficacy trial target multiple epitopes and preferentially use the VH1 gene family. J Virol 86: 11521-11532. Link: https://goo.gl/TWQ1cS
9. Girard MP, Osmanov S, Assossou OM, Kieny MP (2011) Human immunodeficiency virus (HIV) immunopathogenesis and vaccine development: a review. Vaccine 29: 6191-6218. Link: https://goo.gl/PqYqko
10. Akagi T, Kawamura M, Ueno M, Hiraishi K, Adachi M, et al. (2003) Mucosal immunization with inactivated HIV-1-capturing nanospheres induces a significant HIV-1-specific vaginal antibody response in mice. J Med Virol 69: 163-172. Link: https://goo.gl/wc1RL4
11. Alfsen A, Iniguez P, Bouguyon E, Bomsel M (2001) Secretory IgA specific for a conserved epitope on gp41 envelope glycoprotein inhibits epithelial transcytosis of HIV-1. J Immunol 166: 6257-6265. Link: https://goo.gl/iYvnDh
12. Kaul R, Trabattoni D, Bwayo JJ, Arienti D, Zagliani A, et al. (1999) HIV-1-specific mucosal IgA in a cohort of HIV-1-resistant Kenyan sex workers. Aids 13: 23-29. Link: https://goo.gl/bng7Nj
13. Bomsel M, Tudor D, Drillet AS, Alfsen A, Ganor Y, et al. (2011) Immunization with HIV-1 gp41 subunit virosomes induces mucosal antibodies protecting nonhuman primates against vaginal SHIV challenges. Immunity 34: 269-280. Link: https://goo.gl/sDPiIn
14. Ghosh M, Fahey JV, Shen Z, Lahey T, Cu-Uvin S, et al. (2010) Anti-HIV activity in cervical-vaginal secretions from HIV-positive and -negative women correlate with innate antimicrobial levels and IgG antibodies. PLoS One 5: e11366. Link: https://goo.gl/GIz9Wy
15. Vajdy M (2006) Current efforts on generation of optimal immune responses against HIV through mucosal immunisations. Drugs R D 7: 267-288. Link: https://goo.gl/1UPGlu
16. Caputo A, Castaldello A, Brocca-Cofano E, Voltan R, Bortolazzi F, et al. (2009) Induction of humoral and enhanced cellular immune responses by novel core-shell nanosphere- and microsphere-based vaccine formulations following systemic and mucosal administration. Vaccine 27: 3605-3615. Link: https://goo.gl/1hx9sG
17. Iglesias E, Garcia D, Carrazana Y, Aguilar JC, Sanchez A, et al. (2008) Anti-HIV-1 and anti-HBV immune responses in mice after parenteral and nasal co-administration of a multiantigenic formulation. Curr HIV Res 6: 452-460. Link: https://goo.gl/L8zc1Z
18. Lindblad EB. (2004) Aluminium adjuvants--in retrospect and prospect. Vaccine 22: 3658-3668. Link: https://goo.gl/VKjMFb
19. Mahboubi A, Fazeli MR, Dinarvand R, Samadi N, Sharifzadeh M, et al. (2008) Comparison of the adjuvanticity of aluminum salts and their combination in hepatitis B recombinant protein vaccine in assessed mice. Iran J Immunol 5: 163-170. Link: https://goo.gl/FuLf9c
20. Deville S, Pooter A, Aucouturier J, Laine-Prade V, Cote M, et al. (2005) Influence of adjuvant formulation on the induced protection of mice immunized with total soluble antigen of Trichinella spiralis. Vet Parasitol 132: 75-80. Link: https://goo.gl/do5pcI
21. Cauchard J, Sevin C, Ballet JJ, Taouji S (2004) Foal IgG and opsonizing anti-Rhodococcus equi antibodies after immunization of pregnant mares with a protective VapA candidate vaccine. Vet Microbiol 104: 73-81. Link: https://goo.gl/7KFFqQ
22. Waghmare A, Deopurkar RL, Salvi N, Khadilkar M, Kalolikar M, et al. (2009) Comparison of Montanide adjuvants, IMS 3012 (Nanoparticle), ISA 206 and ISA 35 (Emulsion based) along with incomplete Freund's adjuvant for hyperimmunization of equines used for production of polyvalent snake antivenom. Vaccine 27: 10⁶7-1072. Link: https://goo.gl/Rnw8fc
23. Iglesias E, Thompson R, Carrazana Y, Lobaina Y, Garcia D, et al. (2006) Coinoculation with hepatitis B surface and core antigen promotes a Th1 immune response to a multiepitopic protein of HIV-1. Immunol Cell Biol 84: 174-183. Link: https://goo.gl/Tw3PnK
24. Iglesias E, Garcia D, Marquez G, Prieto YC, Sanchez J, et al. (2012) Two mucosal-parenteral schedules to coadminister a multiantigenic formulation against HIV-1 in Balb/c mice. Int Immunopharmacol 12: 487-493. Link: https://goo.gl/a9YzGT
25. Iglesias E, Franch O, Carrazana Y, Lobaina Y, Garcia D, et al. (2006) Influence of aluminum-based adjuvant on the immune response to multiantigenic formulation. Viral Immunol 19: 712-721. Link: https://goo.gl/04oxE6
26. Iglesias E, Ruiz M, Carrazana Y, Cruz LJ, Aguilar A, et al. (2001) Chimeric proteins containing HIV-1 T cell epitopes: Expression in E.coli, purification and induction of antibodies in mice. J Biochem Mol Biol Biophys 5: 109-112. Link: https://goo.gl/tGTseV
27. Lobaina Y, Garcia D, Abreu N, Muzio V, Aguilar JC (2003) Mucosal immunogenicity of the hepatitis B core antigen. Biochem Biophys Res Commun 300: 745-750. Link: https://goo.gl/QmEChE
28. Hardy E, Martinez E, Diago D, Diaz R, Gonzalez D, et al. (2000) Large-scale production of recombinant hepatitis B surface antigen from Pichia pastoris. J Biotechnol 77: 157-167. Link: https://goo.gl/mjKD7G
29. Armbruster DA, Pry T (2008) Limit of blank, limit of detection and limit of quantitation. Clin Biochem Rev 29 Suppl 1: S49-52. Link: https://goo.gl/KFeHgR
30. Debin A, Kravtzoff R, Santiago JV, Cazales L, Sperandio S, et al. (2002) Intranasal immunization with recombinant antigens associated with new cationic particles induces strong mucosal as well as systemic antibody and CTL responses. Vaccine 20: 2752-2763. Link: https://goo.gl/lHQP9p
31. Buonaguro L, Visciano ML, Tornesello ML, Tagliamonte M, Biryahwaho B, et al. (2005) Induction of systemic and mucosal cross-clade neutralizing antibodies in BALB/c mice immunized with human immunodeficiency virus type 1 clade A virus-like particles administered by different routes of inoculation. J Virol 79: 7059-7067. Link: https://goo.gl/7u6g6j
32. Riedl P, Stober D, Oehninger C, Melber K, Reimann J, et al. (2002) Priming Th1 immunity to viral core particles is facilitated by trace amounts of RNA bound to its arginine-rich domain. J Immunol 168: 4951-4959. Link: https://goo.gl/jh0Jwb
33. Guillen G, Aguilar JC, Dueñas S, Hermida L, Iglesias E, et al. (2013) Virus-like particles as nanovaccine candidates. Adv Nat Sci Nanosci Nanotechnol 4: 015005. Link: https://goo.gl/SblDT4
34. Lobaina Y (2015) Co-administración intranasal-parenteral de un candidato vacunal, que combina a los antígenos de superficie y nucleocápsida del virus de hepatitis B, como estrategia terapéutica para el tratamiento de la hepatitis B crónica. PhD thesis, Havana University Faculty of Biology. Link: https://goo.gl/KZq9BD
35. Taouji S, Nomura I, Giguere S, Tomomitsu S, Kakuda T, et al. (2004) Immunogenecity of synthetic peptides representing linear B-cell epitopes of VapA of Rhodococcus equi. Vaccine 22: 1114-1123. Link: https://goo.gl/rMxhz9
36. Kusi KA, Faber BW, Riasat V, Thomas AW, Kocken CH, et al. (2010) Generation of humoral immune responses to multi-allele PfAMA1 vaccines; effect of adjuvant and number of component alleles on the breadth of response. PLoS One 5: e15391. Link: https://goo.gl/xoWRDc
37. Mestecky J, Moldoveanu Z, Russell MW (2005) Immunologic uniqueness of the genital tract: challenge for vaccine development. Am J Reprod Immunol 53: 208-214. Link: https://goo.gl/Xfm3Mj
38. Iglesias E, Cruz O (2016) The Selection of the Aluminum-based Adjuvant Reconfigures the Exposure of the Target Antigen in the Multiantigenic Formulation TERAVAC. Drug Delivery Letters 6 (E-pub ahead of print):
39. Wang YY, Kannan A, Nunn KL, Murphy MA, Subramani DB, et al. (2014) IgG in cervicovaginal mucus traps HSV and prevents vaginal herpes infections. Mucosal Immunol 7: 1036-1044. Link: https://goo.gl/dVuwsx
40. Parr EL, Parr MB (1999) Immune responses and protection against vaginal infection after nasal or vaginal immunization with attenuated herpes simplex virus type-2. Immunology 98: 639-645. Link: https://goo.gl/WzgjSg
41. Gavin AL, Hoebe K, Duong B, Ota T, Martin C, et al. (2006) Adjuvant-enhanced antibody responses in the absence of toll-like receptor signaling. Science 314: 1936-1938. Link: https://goo.gl/Yi7EtO
42. Schnare M, Barton GM, Holt AC, Takeda K, Akira S, et al. (2001) Toll-like receptors control activation of adaptive immune responses. Nat Immunol 2: 947-950. Link: https://goo.gl/pibM3W
43. Pelka K, Latz E (2011) Getting closer to the dirty little secret. Immunity 34: 455-458. Link: https://goo.gl/zeSJ2D
44. Iwasaki A, Medzhitov R (2004) Toll-like receptor control of the adaptive immune responses. Nat Immunol 5: 987-995. Link: https://goo.gl/xvuo9f
45. Desch AN, Gibbings SL, Clambey ET, Janssen WJ, Slansky JE, et al. (2014) Dendritic cell subsets require cis-activation for cytotoxic CD8 T-cell induction. Nat Commun 5: 4674. Link: https://goo.gl/jqo8GE
46. McClure R, Massari P (2014) TLR-Dependent Human Mucosal Epithelial Cell Responses to Microbial Pathogens. Front Immunol 5: 386. Link: https://goo.gl/YfOMjv
47. Lin CF, Tsai CH, Cheng CH, Chen YS, Tournier F, et al. (2007) Expression of Toll-like receptors in cultured nasal epithelial cells. Acta Otolaryngol 127: 395-402. Link: https://goo.gl/GlXoZa
48. Heinz S, Haehnel V, Karaghiosoff M, Schwarzfischer L, Muller M, et al. (2003) Species-specific regulation of Toll-like receptor 3 genes in men and mice. J Biol Chem 278: 21502-21509. Link: https://goo.gl/puZQlC
49. Asahi-Ozaki Y, Itamura S, Ichinohe T, Strong P, Tamura S, et al. (2006) Intranasal administration of adjuvant-combined recombinant influenza virus HA vaccine protects mice from the lethal H5N1 virus infection. Microbes Infect 8: 2706-2714. Link: https://goo.gl/8rBjul