Antigen Uptake After Intradermal Microinjection Depends on Antigen Nature and Formulation, but Not on Injection Depth

Abstract

The skin is an attractive alternative administration route for allergy vaccination, as the skin is rich in dendritic cells (DCs) and is easily accessible. In the skin multiple subsets of DCs with distinct roles reside at different depths. In this study antigen (=allergen for allergy) formulations were injected in ex vivo human skin in a depth-controlled manner by using a hollow microneedle injection system. Biopsies were harvested at the injection site, which were then cultured for 72 h. Subsequently, the crawled-out cells were collected from the medium and analyzed with flow cytometry. Intradermal administration of ovalbumin (OVA, model antigen) solution at various depths in the skin did not affect the migration and maturation of DCs. OVA was taken up efficiently by the DCs, and this was not affected by the injection depth. In contrast, Bet v 1, the major allergen in birch pollen allergy, was barely taken up by dermal DCs (dDCs). Antigens were more efficiently taken up by CD14⁺ dDCs than CD1a⁺ dDCs, which in turn were more efficient at taken up antigen than Langerhans cells. Subsequently, both OVA and Bet v 1 were formulated in cationic and anionic liposomes, which altered antigen uptake drastically following intradermal microinjection. While OVA uptake was reduced by formulation in liposomes, Bet v 1 uptake in dDCs was increased by encapsulation in both cationic and anionic liposomes. This highlights the potential use of liposomes as adjuvant in intradermal allergy vaccine delivery. In conclusion, we observed that antigen uptake after intradermal injection was not affected by injection depth, but varied between different antigens and formulation.

Keywords: Bet v 1; allergen specific immunotherapy; antigen uptake; injection depth; intradermal injection; liposomes; ovalbumin.

Figure 1

Detection of DCs after intradermal injection of 10 μL with a hypodermic needle and syringe. The migrated DCs were collected from human ex vivo skin (circles = abdominal, squares = breast) explants after 72 h of culturing. DCs were defined by expression of both HLA-DR and CD11c. A representative plot of the DCs shows the gating (A). In the DC population 3 subsets were identified: CD1a⁺ dDCs (B), CD14⁺ dDCs (C), and LCs (D) (n = 10 independent experiments) showing each experiment and mean.

Figure 2

Dose dependency of OVA uptake is DC subset specific. OVA uptake by LCs, CD1a⁺, and CD14⁺ dDCs was investigated by flow cytometry as function of the OVA dose after intradermal administration in abdominal skin by using a conventional hypodermic needle-and-syringe. OVA uptake was measured and displayed as percentage of cells that had taken up OVA within migrated CD14⁺ dDCs (A), CD1a⁺ dDCs (B), and LCs (C). The data represents mean ± SEM (n ≥ 3 independent experiments) uptake was compared by one-way ANOVA with Tukey's multiple comparison test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 3

Uptake of OVA by skin-resident DCs is not dependent on injection depth. The uptake of OVA in CD14⁺ dDCs, CD1a⁺ dDCs, or LCs as function of the depth in abdominal skin at which 0.1 μg OVA was administered by using a single hollow microneedle or conventional hypodermic needle-and-syringe (ID). The data represents mean ± SEM (n ≥ 4 independent experiments). No statistical differences were found between the different depths (one-way ANOVA).

Figure 4

Not all antigens are taken up to the same extent by APCs. The uptake of fluorescent OVA or Bet v 1 in CD14⁺ dDCs, CD1a⁺ dDCs, and LCs after microinjection at 500 μm depth with 0.1 μg of antigen in abdominal skin (mean ± SEM; n ≥ 6 independent experiments). The percentage of cells that had taken up antigen was compared in a mixed-effects analysis with a Tukey's multiple comparison test. *p < 0.05, **p < 0.01.

Figure 5

Effect of fluorescently labeled liposomes on antigen uptake. The uptake of fluorescent OVA (A–C) and Bet v 1 (D–F) in different dDCs or LCs was measured (mean ± SEM; n ≥ 6 independent experiments). Antigens (0.1 μg) were injected at 500 μm depth in abdominal (n ≥ 4) or breast (n = 2) skin explants. The percentage of dDCs that had taken up antigen was compared in a mixed-effects analysis and Dunnett post-test to compare uptake to free antigen. *p < 0.05, **p < 0.01, ****p < 0.0001.

Figure 6

Normalized effect of fluorescently labeled liposomes on antigen uptake. The percentage of cells that had taken up free antigen, as displayed in Figure 5, was set at 100% (dotted line in all graphs). Percentage of cells that had taken up antigen after formulation into liposomes was normalized to the uptake of free antigen. The normalized uptake of both OVA (left) and Bet v 1 (right) incorporated in cationic liposomes (white) and anionic liposomes (black) after administration at 500 μm depth in abdominal (n ≥ 4) or breast (n = 2) skin explants is shown (mean ± SEM; n ≥ 6 independent experiments). Significant differences between dendritic cell subset and formulation were not found in a 2-way ANOVA for either antigen.

Figure 7

Liposome uptake by skin resident APCs. The uptake of fluorescently labeled cationic liposomes (left) and anionic liposomes (right) in different skin APCs was measured (mean ± SEM; n ≥ six independent experiments) after injection at 500 μm depth in abdominal (n ≥ 4) or breast (n = 2) skin explants. The percentage of dDCs or LCs that had taken up liposomes was compared in a 2-way ANOVA and showed no significant differences between OVA or Bet v 1 in each subset.