Mannan-mediated gene delivery for cancer immunotherapy

Abstract

Recent years have seen a resurgence in interest in the development of efficient non-viral delivery systems for DNA vaccines and gene therapy. We have previously used oxidized and reduced mannan as carriers for protein delivery to antigen-presenting cells by targeting the receptors that bind mannose, resulting in efficient induction of cellular responses. In the present study, oxidized mannan and reduced mannan were used as receptor-mediated gene transfer ligands for cancer immunotherapy. In vivo studies in C57BL/6 mice showed that injection of DNA encoding ovalbumin (OVA) complexed to oxidized or reduced mannan-poly-L-lysine induced CD8 and CD4 T-cell responses as well as antibody responses leading to protection of mice from OVA+ tumours. Both oxidized and reduced mannan delivery was superior to DNA alone or DNA-poly-L-lysine. These studies demonstrate the potential of oxidized and reduced mannan for efficient receptor-mediated gene delivery in vivo, particularly as DNA vaccines for cancer immunotherapy.

Figure 1

Sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE) gel. (a) Coomassie blue stain and (b) periodic acid Schiff's (PAS) stain. Lane A, oxidized mannan (OxMan); lane B, poly l-lysine (PLL) and lane C, OxMan-PLL.

Figure 2

(a) Degree of complexation between oxidized mannan–poly l-lysine (OxMan-PLL) and ovalbumin (OVA) DNA at different PLL:DNA ratios (R). OxMan-PLL-DNA complexes of R values 0.1, 0.25, 0.4, 0.5, 0.75, 1, 2, 3, 4, 5 and 10 were used to evaluate the degree of complexation by assessing the extent of gel retardation in 0.6% agarose gel. Varying levels of complexation between the two components were observed at different R values, as reflected by the extent of retardation of bands. The degree of complexation decreased as the R value increased; 10 times excesses of PLL and DNA showed complete and no binding, respectively. The amount of DNA loaded into wells for each condition was the same (200 ng). (b) Gel retardation analysis for DNA and OxMan. DNA mixed with OxMan did not show signs of retardation (migration to the anode) and showed the same bands as DNA alone at all R values. (c) Gel retardation analysis for DNA and PLL. For PLL mixed with DNA, all groups showed complete retardation.

Figure 3

Cell toxicity assay. Poly l-lysine (PLL), oxidized mannan–poly l-lysine (OxMan-PLL), reduced mannan–poly l-lysine (RedMan-PLL), OxMan, RedMan or no carrier was added to J774 cells. [³H]thymidine was added after 16 hr and thymidine uptake by cells was measured. Error bars depict standard error of the mean for triplicate wells. A representative example of five different experiments is shown.

Figure 4

Proliferative responses in C57BL/6 mice immunized with different DNA vaccine formulations. C57BL/6 mice were immunized intradermally (i.d.) on days 0 and 14 with 10 μg (R = 0.4) or 50 μg (R = 2.0) oxidized mannan–poly l-lysine–ovalbumin DNA (OxMan-PLL-OVA DNA) or reduced mannan–poly l-lysine–ovalbumin DNA (RedMan-PLL-OVA DNA). Subsequently, 10–14 days after boost immunization, splenocytes were assessed for proliferation response to OVA (diamonds), OVA CD4 (squares) or CD8 T-cell epitopes (triangles) on days 1–5. Medium alone was used as a negative control (‘NEG’; crosses). (a) DNA 10 μg; (b) DNA 50 μg; (c) DNA-PLL 10 μg; (d) DNA-PLL 50 μg; (e) RedMan-PLL-DNA 10 μg; (f) RedMan-PLL-DNA 50 μg; (g) OxMan-PLL-DNA 10 μg; (h) OxMan-PLL-DNA 50 μg. Concanavalin A (Con A) was used as an internal positive control. Data are shown as mean counts per minute (c.p.m.) for triplicate wells ± standard error of the mean. One representative of two experiments is shown (n = 3–4 mice/group).

Figure 5

Interferon (IFN)-γ responses in C57BL/6 mice immunized with different DNA vaccine formulations. Mice were immunized intradermally (i.d.) on days 0 and 14 with 10 μg (R = 0.4) or 50 μg (R = 2.0) oxidized mannan–poly l-lysine–ovalbumin DNA (OxMan-PLL-OVA DNA) or reduced mannan–poly l-lysine–ovalbumin DNA (RedMan-PLL-OVA DNA). Subsequently, 10–14 days after boost immunization, IFN-γ responses to OVA (OVA), SIINFEKL (CD8) and OVA323-339 (CD4) were assessed by enzyme-linked immunosorbent spot-forming cell assay (ELISPOT). Mean spot-forming units for triplicate samples ± standard error are shown. This is a representative example of four different experiments (n = 3 mice/group).

Figure 6

Antibody responses in C57BL/6 mice immunized with different DNA vaccine formulations. C57BL/6 mice were immunized intradermally (i.d.) on days 0 and 14 with 10 μg (R = 0.4) or 50 μg (R = 2.0) oxidized mannan–poly l-lysine–ovalbumin DNA (OxMan-PLL-OVA DNA) or reduced mannan–poly l-lysine–ovalbumin DNA (RedMan-PLL-OVA DNA) and, 14 days after boost injection, serum was collected and immunoglobulin G (IgG) levels were assessed by enzyme-linked immunosorbent assay (ELISA). Individual mouse titres are shown with means denoted by a horizontal line. A representative example of three different experiments (n = 5–6 mice/group) is shown.

Figure 7

EG7 tumour challenge of C57BL/6 mice. C57BL/6 mice were immunized with 10 μg poly l-lysine (PLL)-DNA, 50 μg PLL-DNA, 10 μg reduced mannan (RedMan)-PLL-DNA, 50 μg RedMan-PLL-DNA, 10 μg oxidized mannan (OxMan)-PLL-DNA or 50 μg OxMan-PLL-DNA, and challenged with 1 × 10⁷ EG7 cells. The tumour sizes for individual mice on day 9 after challenge are shown. In addition, two groups of mice immunized with phosphate-buffered saline (PBS) or 50 μg OxMan-PLL-DNA were challenged with EL4 cells as controls to demonstrate ovalbumin (OVA)-specific antitumour responses in immunized mice. The tumour sizes for individual mice on day 9 after challenge are shown. The horizontal bar denotes mean tumour size (in mm²). Representative example of three different experiments (n = 6–7 mice/group) is shown.