Introduction

Adjuvants are compounds employed in vaccine formulation to improve the effect of co-administered antigens, inducing a stronger, faster, and longer-lasting immune response than generated by the antigen only [1]. Different adjuvants act on the immune system in distinct ways, directly related to their mechanism of action, which induces a diversity of responses in the immune system [2]. Understanding the mechanism of action and cell recruitment kinetics of new adjuvants is essential to the development and improvement of vaccines. Although adjuvants have been employed since the 1920s, only a few are licensed for human use. Aluminum hydroxide (Alum) was the first adjuvant licensed for human use, in 1930 [3]. MF59 (water-in-oil squalene-based emulsion), Imiquimod (R-837), and Resiquimod (R-848) [4], have also been licensed for human use, mostly in Europe. However, as many adjuvants have been discovered empirically, their mechanisms of action and cell recruitment kinetics remain unclear [5].

Glucopyranosyl Lipid A Stable Emulsion (GLA-SE), Montanide Gel, and R-848 are developed adjuvants that have shown great promise in vaccine tests. GLA-SE, used in a vaccine against the Influenza virus, stimulated an increase of humoral and cellular immune response with induction mainly of TNF-α and IFN-γ cytokines [6]. Montanide ISA 720 was evaluated associated with Leishmania donovani antigen, inducing a strong Delayed-Type Hypersensitivity (DTH) response, with the production of TNF-α and IFN-γ [7]. R-848 was also able to induce a type 1 immune response, with activation of myeloid and plasmacytoid dendritic cells, lymphocytes, and macrophages [8,9]. However, no data have been published on the time course of the recruitment of innate immune components by these adjuvants. Therefore, we aimed to evaluate the effect of the Montanide Pet Gel A (MPGA), Glucopyranosyl Lipid A Stable Emulsion (GLA-SE), and Resiquimod (R-848) adjuvants in the skin of mice inoculated intradermal route. The time course evaluated spanned from 1 to 336h, fully characterizing the immune response profile stimulated by the adjuvants and allowing a detailed comparison with the classic type II adjuvant, Aluminum hydroxide (Al(OH)₃).

Materials and Methods

This study was approved by the Ethical Committee on the Use of Animals (CEUA) of the Federal University of Ouro Preto (protocol 2011/26). Male outbred Swiss mice (4 to 8-week-old) were divided into five groups (n=5 mice/group/time): (1) Control group received sterile saline (NaCl 0.9%, pH 7.2-7.4); (2) Al(OH)₃ group received a 180 μg of Aluminum hydroxide (Sigma-Aldrich Co St. Louis, MO, USA); (3) MPGA group received 50 μL of Montanide Pet Gel A (Seppic, Paris, France); (4) GLA-SE group received a 50 μg of Glucopyranosyl Lipid A Stable Emulsion (Alexis Biochemicals, San Diego, USA); and (5) R-848 group received a 25 μg of Resiquimod (Sigma Chemical CO, St Louis, USA). Mice received a single 50 μL dose of each adjuvant by intradermal route at the dorsum previously shaved, and were evaluated at several time points (1, 12, 24, 48, 96, 168, and 336h). At each time point the animals were anesthetized intraperitoneally with sodium thiopental 2.5% (Tiopentax - Cristalia^®, São Paulo, Brazil, 30 mg/Kg). Mice received a lethal dose of sodium thiopental 2.5% and necropsied to collect the skin from the inoculation site. 

Skin from adjuvant inoculated mice were collected and divided into two parts. One of these was preserved in 10% buffered formalin for morphometric analysis. Posteriorly, it was dehydrated, embedded in paraffin, cut by microtome into 5 μm sections, mounted on slides and stained with Hematoxylin-Eosin (HE) for histological procedures [10]. The rest of the fragment was stored at -80^ºC until use.

The dermis/epidermis inflammatory cell infiltrates were measured analyzing 20 images of randomly selected fields (total area 1.5x10⁶ μm², 40x objective). Leica QWin software (Leica Microsystems, Wetzlar, Germany) was used for analysis of images by counting the cell nuclei present in tissue sections. The skin appendages such as hair follicles and sebaceous glands were excluded from the morphometric analysis as a standard procedure.

The slides with skin sections were used for immune cellular characterization in previously quantified inflammatory cell infiltrates. The cell profile was attained by differential leukocyte count of main inflammatory cells (neutrophils, macrophages, eosinophils, and lymphocytes) [10].

Briefly, skin samples from the inoculation site were weighed and received 700 μL of inhibitor protease (Sigma Chemical Co, St Louis, USA) and were homogenized in a Tissuelyser (Quiagen, USA). The homogenate samples were centrifuged at 10,000 x g for 10 minutes at 4 ºC and the supernatants were collected for cytokine measurement.  The cytokines IFN-γ, TNF-α, IL-6, IL-10, and IL-4 were measured by Cytometric Bead Array (BD Biosciences) [10]. The events were acquired on the FACSCalibur cytometer (BD Biosciences). The results were obtained using the FCAP Array^TM v1.0.2 software (BD Biosciences, San Jose, CA, USA) and expressed by picograms per milliliter (pg/mL) of total protein.

Statistical analyzes were performed using the GraphPad Prisma 8.0 software (Prism Software, Irvine, CA, USA). The data were plotted as absolute values ± Standard Deviation (SD) from duplicate sets of experiments. The Kolmogorov-Smirnoff test was used to confirm the normality of the data. Analysis of variance (one-way ANOVA) was performed, followed by the Turkey multiple comparison test to determine the significant differences between groups concerning inflammatory cell infiltrates and differential leukocyte count. The statistical significance was set at p-value < 0.05.

Results

The analysis of cell recruitment in the mice skin showed that all adjuvants’ groups increased recruitment of cells compared with the C group, however with distinct kinetics. In addition, the adjuvants (MPGA, GLA-SE, and R-848), demonstrated an increase in cells recruited at the site of inoculation when compared with classic type II adjuvant Al(OH)₃. MPGA and GLA-SE groups showed a similar response, albeit with an earlier and more durable cell recruitment in the former. Furthermore, GLA-SE was able to induce a significant increase in the number of cells at most of the time points evaluated (12, 24, 48, 96, and 168h), whereas MPGA showed a delay in the increase of cell recruitment (Figure 1a). Representative photomicrographs of histological sections of the inflammatory cell infiltrate in the mice skin at the inoculation site were shown in Figure 1b. Generally, the inflammatory cell infiltrates were observed in the muscle layer and deep dermis, causing enlargement of the entire dermis.    

The adjuvants MPGA and GLA-SE displayed similar cell recruitment profiles with an increase of neutrophils at 12 hours, 40% and 67% respectively. At 168 hours, these percentages decreased to 2% and 8%, respectively. We observed a neutrophils replacement by macrophages, with advancement of kinetics, at 168 hours in MPGA (23%) and GLA-SE (28%) groups. In contrast, the R-848 group had a similar cell recruitment profile with Al(OH)₃ group (Figure 1c).

Please maintain this figure in large size, because in the PDF file it is impossible to see the caption. Thus, we suggest using half a page for this figure.

MPGA and GLA-SE groups showed increased levels of IL-6 at 12 hours, and increased levels of TNF-α and IFN-γ at 24 hours. R-848 and Al(OH)₃ groups displayed a similar cytokine profile inducing higher levels of IL-6, TNF-α, and IL-10 at 1 hour, although Al(OH)₃ also induced IL-4 cytokine (Figure 2).

Discussion

The present study investigated the inflammatory cell infiltrates, cell recruitment and cytokine production, overtime after intradermal inoculum of GLA-SE, MPGA, and R-848 adjuvants. GLA-SE and MPGA adjuvants promoted a strong induction of pro-inflammatory cytokines production, resulting in intense and long-lasting cell recruitment. The IFN-γ and TNF-α production after stimulation by these compounds (GLA-SE and MPGA) were more intense and more prolonged than observed for the other cytokines. On the other hand, R-848 and Al(OH)₃ demonstrated a similar pattern of cell recruitment, however only Al(OH)₃ adjuvant displayed IL-4 production as expected. GLA-SE, a synthetic compound derived from LPS, is a Toll-like receptor 4 (TLR4) agonist, known to induce strong cellular activation with promotion of type 1 immune response [6,11]. Here, we show that this response starts as early as 1 hour after inoculation, as indicated by IL-6 production, with strong peaks of TNF-α detected at 12 hours and high production of IFN-γ at 24 hours. These results contrast with those of Lambert et al. (2012) [12], who reported a delayed presence of IL-6 and no difference in TNF-α production after subcutaneous injection of GLA-SE. Moreover, our time course experiment showed that cytokine production persists for 48 hours, decaying after that to normal levels. The early detection of IL-6 in the skin allows faster antigen recognition by APCs, and the persistence of TNF-α and IFN-γ might explain the intense and long-lasting cell recruitment in the dermis demonstrated. This prolonged immune response is an important property for vaccine development as it allows better antigen recognition and posterior specific T-cell presentation.

MPGA and GLA-SE displayed a similar pattern of cytokine production.  Furthermore, MPGA persisted for 336 hours at the inoculation site, probably due to its gel-based composition, consistent with the results by  Vialle et al. (2010) [13] obtained with intramuscular injection in guinea pigs.  This gel-based depot system might explain the persistence of IFN-γ levels for 96 hours. Regarding inflammatory process, GLA-SE and MPGA adjuvants promoted the earliest (12h and 48h) and most intense cell recruitment among all adjuvants tested, around 3-fold more than the control group. Moreover, cell persistence was very long, lasting until 168 hours for GLA-SE and 336 hours for MPGA at levels 2.7-fold and 2.2-fold higher than control and 2.15-fold and 1.4-fold higher than (Al(OH)₃), respectively. TLR4 agonists, such as GLA-SE, activate the immune system via TRIF and MyD88 pathways, which are related to the induction of chemokines/cytokines involved in cell recruitment and presentation pathways [12]. The persistence of MPGA in the inoculation site and long-lasting inflammation demonstrate that this gel-based composition presents a strong ability to induce the immune system to engage in intense cell recruitment. Vialle and colleagues [13] demonstrated that in association with OVA, MPGA induced sustained antigen release overtime, with recruitment of the innate immune system by a pro-inflammatory profile in the injected muscle and improved phagocytosis of the antigen increasing the activity of Antigen Presentation Cells (APCs). Al(OH)₃ adjuvant presents the same depot system mechanism, promoting around 1.7-fold more cell recruitment than the control group, however it is not able to induce strong inflammation as GLA-SE and MPGA.

Additionally, R-848 did not present remarkable cell recruitment to the dermis. This pattern was observed in Al(OH)₃ and R-848 groups may be explained by the fact that these adjuvants induced the production of IL-10, an important immunoregulatory cytokine [14]. The persistence and intensity of the responses induced by GLA-SE and MPGA make these adjuvants particularly promising in vaccine formulations, once may provide a more constant and longer antigen presentation than achieved with currently approved adjuvants.  The inflammatory process profile generated by the adjuvants varied over time. MPGA and GLA-SE adjuvants promoted high neutrophil recruitment at 12 hours that was persistent until 24 hours after inoculation. The presence of this cell at the inoculation site acts as the first line of defense against pathogens, releasing cytokines that induce the recruitment of other cell types such as macrophages [15]. At later time points, the cell profiles changed to a reduced number of neutrophils and an increased number of macrophages. Macrophages together with dendritic cells (DCs) are among the main cells responsible for phagocytosis and antigen presentation to T cells  [16]. The quick time recruitment of neutrophils and long-lasting macrophages creates a good microenvironment for antigen presentation, confirming the potential of these adjuvants for vaccine development. In contrast, R-848 did not show a significant cell profile variation during the time course. Differently, Kwissa et al. (2012) [17] demonstrated that intramuscular immunization of nonhuman primates with R-848 induced neutrophil increase on PBMC, although was not observed the cell migration from blood to the muscle.

Our results demonstrate the potential of adjuvants in stimulating the cellular immune response, which is directly associated with the dose inoculated and the route of administration [18]. Most existing vaccines use subcutaneous or intramuscular routes. The advantage of the intradermal route is an easier and faster access route, minimally invasive, and allows lower antigen dose, related to reducing the costs of a vaccine [19]. Here, we demonstrate that the intradermal route is as efficient as other routes in inducing immune system activation. Moreover, a complete-time course study evaluating a profile of innate immune responses completely elucidates the behavior of response triggered by different adjuvants, thus allowing the choice of a better adjuvant for each vaccine.

Conclusion

In this sense, based on our results we conclude that GLA-SE adjuvant is the most promising to compose vaccine candidates against intracellular pathogens due to its capacity to promote important pro-inflammatory cytokines production with high and long-lasting cell recruitment, which creates a perfect microenvironment for cell activation and posterior antigen presentation.

Acknowledgments

This work was supported by Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG), and the Instituto Nacional de Ciência e Tecnologia em Doenças Tropicais (INCT-DT). The authors acknowledge the Brazilian agencies CNPq (grant number 435224/2018-2), FAPEMIG (grant number APQ-03505-13 - Programa de Pesquisa para o SUS–PPSUS, APQ-01373-14 - PRONEX, APQ-01008-14, APQ-02638-17, INV-00037-14, APQ 02577-18, APQ 02286-21 and APQ-02556-18), CAPES (Finance Code 001), and UFOP for funding (grant number 082017080). A.B.R., B.M.R., and C.M.C. are grateful to CNPq for fellowships. 

Figures

References

1. Ramon G (1926) Procedes pour acroitre la production des antitoxins. Ann Inst Pasteur 40: 1-10.
2. Singh M, O'Hagan DT (2003) Recent advances in veterinary vaccine adjuvants. Int J Parasitol 33: 469-478.
3. Brewer JM (2006) (How) do aluminium adjuvants work? Immunol Lett 102: 10-15.
4. Apostolico JS, Lunardelli VA, Coirada FC, Boscardin SB, Rosa DS (2016) Adjuvants: Classification, Modus Operandi, and Licensing. J Immunol Res 2016.
5. O'Hagan DT, De Gregorio E (2009) The path to a successful vaccine adjuvant--'the long and winding road'. Drug Discov Today 14: 541-551.
6. Coler RN, Baldwin SL, Shaverdian N, Bertholet S, Reed SJ, et al. (2010) A synthetic adjuvant to enhance and expand immune responses to influenza vaccines. PLoS One 5: e13677.
7. Mutiso JM, Macharia JC, Taracha E, Wafula K, Rikoi H, et al. (2012) Safety and skin delayed-type hypersensitivity response in vervet monkeys immunized with Leishmania donovani sonicate antigen delivered with adjuvants. Rev Inst Med Trop Sao Paulo 54: 37-41.
8. Gibson SJ, Lindh JM, Riter TR, Gleason RM, Rogers LM, et al. (2002) Plasmacytoid dendritic cells produce cytokines and mature in response to the TLR7 agonists, imiquimod and resiquimod. Cell Immunol 218: 74-86.
9. Wang JP, Zhang L, Madera RF, Woda M, Libraty DH (2012) Plasmacytoid dendritic cell interferon-alpha production to R-848 stimulation is decreased in male infants. BMC Immunol 13: 35.
10. Vitoriano-Souza J, Moreira N, Teixeira-Carvalho A, Carneiro CM, Siqueira FA, et al. (2012) Cell recruitment and cytokines in skin mice sensitized with the vaccine adjuvants: saponin, incomplete Freund's adjuvant, and monophosphoryl lipid A. PLoS One 7: e40745.
11. Coler RN, Day TA, Ellis R, Piazza FM, Beckmann AM, et al. (2018) The TLR-4 agonist adjuvant, GLA-SE, improves magnitude and quality of immune responses elicited by the ID93 tuberculosis vaccine: first-in-human trial. NPJ Vaccines 3: 34.
12. Lambert SL, Yang CF, Liu Z, Sweetwood R, Zhao J, et al. (2012) Molecular and cellular response profiles induced by the TLR4 agonist-based adjuvant Glucopyranosyl Lipid A. PLoS One 7: e51618.
13. Vialle R, Dupuis L, Deville S, Bertrand F, Gaucheron J, et al. (2010) Microgel particulate adjuvant: characterisation and mechanisms of action. Procedia in Vaccinology 2: 12-16.
14. Gautam S, Kumar R, Maurya R, Nylen S, Ansari N, et al. (2011) IL-10 neutralization promotes parasite clearance in splenic aspirate cells from patients with visceral leishmaniasis. J Infect Dis 204: 1134-1137.
15. Su Y, Richmond A (2015) Chemokine Regulation of Neutrophil Infiltration of Skin Wounds. Adv Wound Care (New Rochelle) 4: 631-640.
16. Nyambura LW, Jarmalavicius S, Walden P (2018) Impact of Leishmania donovani infection on the HLA I self peptide repertoire of human macrophages. PLoS One 13: e0200297.
17. Kwissa M, Nakaya HI, Oluoch H, Pulendran B (2012) Distinct TLR adjuvants differentially stimulate systemic and local innate immune responses in nonhuman primates. Blood 119: 2044-2055.
18. Guy B (2007) The perfect mix: recent progress in adjuvant research. Nat Rev Microbiol 5: 505-517.
19. Kim YC, Jarrahian C, Zehrung D, Mitragotri S, Prausnitz MR (2012) Delivery systems for intradermal vaccination. Curr Top Microbiol Immunol 351: 77-112.