Ovalbumin and Poly(i:c) Encapsulated Dendritic Cell-Targeted Nanoparticles for Immune Activation in the Small Intestinal Lymphatic System

Abstract

Here, we describe antigen and adjuvant encapsulated dendritic cell-targeted nanoparticles for immune activation in the small intestinal lymphatic system to inhibit melanoma development. This strategy was demonstrated using chondroitin sulfate-coated nanoparticles (OPGMN) grafted with glycocholic acid and mannose for cationic liposomes encapsulated with ovalbumin as an antigen and polyinosine-polycytidylic acid as a cancer-specific adjuvant. OPGMN was absorbed in the gastrointestinal tract and delivered to the lymph nodes when orally administered. Oral delivery of OPGMN induced increased dendritic cell maturation compared to the intradermal route in the lymph node and induced T helper type 1 and type 2 responses, such as immunoglobulin G1 and G2c, interferon-gamma, and interleukin-2, in the blood. Repeated oral administration of OPGMN increased the population of CD3⁺CD8⁺ T cells, CD44^highCD62L^low memory T cells, and CD11b⁺CD27⁺ natural killer cells in the blood. OPGMN completely prevented melanoma development in the B16F10-bearing C57BL/6 mouse model by reducing the population of CD4⁺CD25⁺Foxp3⁺ regulatory T cells in the blood. This strategy is expected to prevent the recurrence of tumors after various cancer treatments.

Graphical Abstract

This study describes the non-specific immune response and consequent inhibition of tumor formation by oral administration of bile acid-containing ovalbumin and poly(i:c) encapsulated nanoparticles. This nanoformulation can be applied as a platform technology that provides a route through which nanoparticles containing various active ingredients can be absorbed from the small intestine and delivered to the lymph nodes.

1. Introduction

Melanoma is the most dangerous type of skin cancer and begins in cells known as melanocytes.^[1] Melanoma occurs primarily in people with low levels of the skin pigment melanin who are exposed to ultraviolet light, have poor immune function, and rare genetic conditions such as xeroderma pigmentosum.^[2] Melanoma can grow deep into the skin and spread through the lymph nodes (LNs) to other organs, reducing patient survival rates.^[3] Most recurrent melanoma spreads to many organs in the patient at a faster growth rate than prior to treatment, making tumor treatment more difficult.^[4] To improve the survival rate of melanoma patients, a method is needed that can prevent the development and recurrence of melanoma. However, it is difficult to predict and prevent melanoma metastasis and recurrence because it is affected by multiple variables, such as the subtype, cancer stage, histology, genetic factors, patient-related factors, stage at diagnosis, and treatment method.^[5–6] Immunotherapy has great potential as a method for preventing the development and recurrence of melanoma, since it does not need to consider variables as a method to overcome the immune suppression of tumors by triggering the patient’s immune system against cancer cells.^[7] Currently, most melanoma immunotherapies are aimed at treating tumors that have already metastasized or recurred rather than trying to prevent cancer from occurring or recurring.^[8–9] Therefore, cancer vaccines that can increase patient immunity during the early stages of melanoma (before the occurrence of melanoma metastasis) have great potential for preventing metastasis and recurrence of melanoma after treatment.

The lymphatic system is a network of lymphatic vessels and LNs where immune cells circulate and dendritic cells (DCs) induce antigen presentation and immune activation.^[10] Lymphatic vessels provide channels for antigens, antigen-presenting cells, and lymphocytes to travel from tissue to drain LNs. Delivery of antigens and adjuvants to the lymphatic system is the most important factor of successful cancer immunotherapy. In previous reports, bile acid-containing nanoparticles, when administered orally, were absorbed in the distal ileum via an apical sodium-dependent bile acid transporter (ASBT) and could be transported through the small intestinal lymphatic system.^[11] Therefore, oral delivery of cancer vaccines containing bile acids represents an attractive route to prevent the development and recurrence of melanoma since the largest population of DCs is in the intestine and LNs.^[12–14]

Here, we describe a strategy to prevent the development or recurrence of melanoma by inducing immune preactivation through oral delivery of bile acid-containing dendritic cell-targeted antigen and adjuvant encapsulated nanoparticles (OPGMN). In this study, we utilized a liposomal nanoparticle that is coloaded with ovalbumin (OVA) and polyinosine-polycytidylic acid (PIC) and coated with chondroitin sulfate-g-mannose (CSM) and chondroitin sulfate-g-glycocholic acid (CSG). OVA, a protein antigen, is widely used for eliciting cellular and humoral immune responses in cancer immunotherapy.^[15] OVA has been reported to facilitate efficient major histocompatibility complex class one presentation and generate CD8⁺ cytotoxic T lymphocytes in DCs.^[16–17] PIC is a synthetic double-stranded polyribonucleotide and an interferon (IFN) inducer that plays the role of an antiviral and anticancer agent as adjuvants for cancer vaccines that promote the maturation of DCs and enhance cancer-specific immune responses.^[18–19] Mannose was used as a DC-specific interaction moiety to improve antigen presentation by the mannose receptor on the surface of myeloid DCs.^[20–21] Glycocholic acid (GCA) is a type of bile acid and is used to increase the oral bioavailability of OPGMN in the gastrointestinal tract (GIT).^{[11, 22]} OPGMN was able to move directly into the LNs while maintaining intact nanoparticles when administered orally. Oral administration of OPGMN induced increased DCs maturation compared to intradermal (ID) injection in the LNs. OPGMN induces T helper type 1 (Th1) and T helper type 2 (Th2) immune responses to increase interleukin 2 (IL-2) and IFN-gamma (IFN-γ) in the blood (Scheme 1). The immune preactivation of OPGMN was sufficient to completely inhibit the development of B16F10 melanoma by increasing the number of central memory T cells (T_CM), cytotoxic T cells, and natural killer (NK) cells and decreasing the number of regulatory T cells (Tregs) in the blood. These results were confirmed in the B16F10 melanoma mouse model and CT26 colorectal carcinoma mouse model.

Scheme 1.

Schematic representation of the strategy of oral vaccine delivery and tumor-forming prevention.

2. Results and Discussions

2.1. Preparation and Characterization of OPGN

To shield the cationic liposome (CL) surface and ASBT-mediated absorption, CSG was synthesized following a previous report.^[23] Briefly, primary amine groups synthetically introduced to the carboxylate group on GCA were used to couple to carbodiimide-activated carboxyl groups on chondroitin sulfate (CS) in aqueous conditions. The successful synthesis of CSG was confirmed by ¹H-NMR.^[22]

To impart the DC targeting effect to the nanoparticles, CSM was synthesized according to Figure S1a. The carboxyl group of CS was activated by 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS) and synthesized by binding with the primary amine of 4-aminophenyl α-D-mannopyranoside (mannose). The successful synthesis of CSM was confirmed by a -CH peak of mannose appearing at 7.0 ppm using ¹H-NMR (Figure S1b). The degree of mannose conjugation on CS was estimated by comparing the intensities of the -CH₃ peak at 1.8 ppm from CS. The degree of conjugation was determined to be 5.3 mannose per 10 repeat units of CS.

CL, composed of 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP) and 1,2-dioleoyl-sn-glycerol-3-phosphocholine (DOPC), was fabricated using a thin-film hydration method to attach OVA and PIC on the surface of the liposomes by electrostatic interactions. Briefly, an anionic charge of OVA and PIC solutions were added to CL and shaken at 50 rpm for 2 hours. The anionic charge of CS, CSG, or a mixture of CSG and CSM (50:50 w/w) solution was added to the OVA and PIC attached CL solution. The unencapsulated OVA, PIC, CS, CSG, and CSM were removed by dialysis. To clarify the identity of OPGMN (OVA and PIC encapsulated CSG and CSM coated nanoparticle), various compositions of nanoparticles, such as CL (empty nanoparticle), GMN (CSG and CSM coated empty nanoparticle), PGMN (PIC encapsulated CSG and CSM coated nanoparticle), OGMN (OVA encapsulated CSG and CSM coated nanoparticle), OPCN (OVA and PIC encapsulated CS coated nanoparticle, and OPGN (OVA and PIC encapsulated CSG coated nanoparticle), were prepared, and their properties such as average size, PDI, zeta-potential, and loading efficiency were analyzed as shown in Table 1. The resulting CL exhibited a positive zeta potential (+61 mV), which changed to negative (−34 to −41 mV) after coating with CSG and a mixture of CSG and CSM. PGMN and OGMN exhibited loading efficiencies of 82% PIC and 84% OVA, respectively. OPGN and OPGMN exhibited reduced loading efficiencies of 53–55% OVA and 71–73% PIC due to the limited drug loading space of liposomes. However, the nanoparticle preparation method of encapsulating a drug by electrostatic interaction on the liposome surface exhibited improved drug encapsulation efficiency compared with the method of encapsulating the drug inside the liposome core.^[24]

Table 1.

Composition and characterization of CSG- and CSM-coated OVA- and PIC-loaded liposomes.

Formulation	DOPC (mg)	DOTAP (mg)	OVA (mg)	PIC (mg)	CS (mg)	CSG (mg)	CSM (mg)	Size (d. nm)	PDI	Zeta-potential (mV)	OVA Loading efficiency (%)	PIC Loading efficiency (%)
CL	38	39	-	-	-	-	-	128 ± 1	0.183 ± 0.006	61 ± 1	-	-
GMN			-	-	-	40	40	133 ± 2	0.193 ± 0.009	−34 ± 4	-	-
PGMN			-	1	-	40	40	158 ± 3	0.119 ± 0.046	−40 ± 1	-	82 ± 4
OGMN			2.5	-	-	40	40	150 ± 2	0.101 ± 0.025	−38 ± 2	84 ± 2	-
OPCN			2.5	1	80	-	-	154 ± 3	0.143 ± 0.008	−44 ± 3	55 ± 3	72 ± 2
OPGN			2.5	1	-	80	-	171 ± 4	0.155 ± 0.003	−38 ± 1	53 ± 4	73 ± 2
OPGMN			2.5	1	.	40	40	157 ± 3	0.184 ± 0.025	−41 ± 1	55 ± 2	71 ± 3

CL: Cationic liposome, GMN: CSG and CSM coated nanoparticle, PGMN: PIC encapsulated CSG and CSM coated nanoparticle, OGMN: OVA encapsulated CSG and CSM coated nanoparticle, OPCN: OVA and PIC encapsulated CS coated nanoparticle, OPGN: OVA and PIC encapsulated CSG coated nanoparticle, OPGMN: OVA- and PIC-encapsulated CSG- and CSM-coated nanoparticles

2.2. Lymph Node Delivery by Oral Administration of OPGN

Here, we elucidated the GIT absorption and delivery pathway using mannose-free OPGN to demonstrate that OPGN can be absorbed from the GIT and transported to lymph nodes by bile acid transporter rather than antigen-presentation by DCs. In previous studies, we have reported that bile acids containing nanoparticles are transitioned for at least 4 hours in the GIT, where they can be absorbed at the ileum and delivered to the lymph nodes.^{[11, 25]} The size of OPGN did not change significantly in distilled water (DW), pH 1.8 fasted state simulated intestinal fluid (FaSSIF), and pH 5.8 fasted state simulated gastric fluid (FaSSGF) and maintained the 150–170 nm size (Figure 1a). Due to the stable electrostatic interactions between strong acidic and basic components, the size of OPGN was not noticeably changed by gastric and intestinal pH. This means that when OPGN is administered orally, OVA and PIC encapsulated in OPGN are well protected while passing through the GIT. To evaluate the ASBT-mediated cellular uptake and tissue penetration efficacy of OPGN, we prepared Alexa 647-labeled OVA-encapsulated OPGN and OPCN. OPGN (with GCA) exhibited a 2-fold increased cellular uptake efficacy in ASBT-expressing SK-BR-3 cells compared with OPCN (without GCA) (Figure 1b). OPGN showed a significantly increased absorption in the ileum, where ASBT was most expressed, compared to OPCN through an everted gut sac model (Figure 1c).^[11] These results suggest that GCA in OPGN enhances cellular uptake and tissue penetration by interaction with ASBT on the epithelial cells of in vitro and ex vivo models. To clarify whether nanoparticles could be delivered to the LNs, we prepared Alexa 647-labeled OVA and rhodamine-labeled PIC-encapsulated OPGN and OPCN. Oral gavage of OPGN (OG, OPGN_OG) exhibited a highest LNs delivery efficiency compared to oral administration of OPCN (OPCN_OG) and intradermal injection of OPGN (OPGN_ID) (Figure 1d). The presented fluorescence intensities were normalized based on lymph node fluorescence intensities from untreated mice. The LNs delivery efficiency of OPGN_OG has significantly decreased in the cycloheximide pre-treated mice (OPGN_OG + cycloheximide). Cycloheximide can block the chylomicron pathway without affecting other absorption pathways.^[26] In the confocal microscopy images, OPGN_OG and OPGN_ID were localized in the LNs after 3 hours post-administration into C57BL/6 mice (Figure 1e). The fluorescence signals of OPGN_OG and OPGN_ID were maintained at least 24 hours post-administration (Figure S2). These results suggest that OPGN can be mainly transported to LNs via the chylomicron pathway. Based on these differences in cellular uptake, intestinal permeability, and LNs reach between OPGN and OPCN, we assumed that GCA in the OPGN plays a crucial role in OPGN intestinal absorption and LNs migration. However, since there are various variables in the GIT environment, further studies are needed to clearly elucidate the mechanisms of OPGN uptake and LNs delivery.

Figure 1.

a) Size distribution of OPGN in DW, pH 1.8 FaSSGF, and pH 5.8 FaSSIF. b) Flow cytometer analysis of OVA, OPCN, and OPGN (Alexa-647 OVA concentration = 5 μg/mL). c) OVA transport of three different EGS tissues (duodenum, jejunum, and ileum) in Krebs-Ringer solution containing OVA, OPCN, and OPGN (n=4). d) Normalized fluorescence intensity of Alexa 647-labeled OVA in the axillary, brachial, and mesenteric LNs (n=4). e) Confocal images of the axillary LNs. After 3 hours administration, axillary LNs were harvested and frozen section with 10 μm thickness. White scale bar is 250 μm. *P value < 0.05. ***P value < 0.001.

2.3. In Vivo DC Maturation

To determine whether OPGMN at the LNs affected DCs maturation, various doses of OPGN were administered OG. DCs isolated after 7 days of administration from axillary LNs displayed a shift from the immature to the mature phenotype through upregulation of the costimulatory molecules CD40, CD80, and CD86. Despite additional antigen and adjuvant doses, OPGN_100 μg (100 μg of OVA and 52 μg of PIC per mouse) and OPGN_200 μg (200 μg of OVA and 104 μg of PIC per mouse) exhibited similar or reduced DC maturation compared to OPGN_50 μg (50 μg of OVA and 26 μg of PIC per mouse), respectively (Figure S3). GMN_OG displayed DC maturation similar to that of the nontreated group, clarifying the effect of OPGN on DC maturation. Therefore, we used OPGN_50 μg for all further experiments. Comparing the DC maturation of OPGMN and OPGN to evaluate DC targeting properties by mannose, OPGMN_OG exhibited higher DC maturation than OPGN_OG and OPGMN_ID at the same dose (Figure 2). This result indicates that mannose in OPGMN increases DC maturation efficacy and that OPGMN delivery to LNs via OG induced increased DC maturation efficiency compared to the conventional ID route.^[27] Additionally, these results demonstrate that OPGN and OPGMN properly maintain their function in GIT environments, including extreme pH variations and various digestive enzymes after oral administration.

Figure 2.

Flow cytometer analysis of matured DCs in the axillary LN. a) OPGN and OPGMN were administered by ID or OG in C57BL/6 mice (n=4). Statistical analysis of b) CD11c⁺MHC class II⁺CD40⁺, c) CD11c⁺MHC class II⁺ CD80⁺, and d) CD11c⁺MHC class II⁺ CD86⁺ matured DCs. After 7 days each administration, axillary LNs harvested and a fixed number of cells (1 × 10⁵) per sample were analyzed by a flow cytometer. **P value < 0.01. ***P value < 0.001.

2.4. In Vivo Immune Responses to Repeated OPGMN Administration

To evaluate the antibody-triggering ability of OPGMN, a total of four nanoparticle administrations were performed of OG or ID weekly, and the amount of total immunoglobulin G (IgG) was measured one-week postadministration from each dose. The concentration of total IgG increased in OPGMN_OG compared with OPGMN_ID and OPGN_OG, and the highest IgG increase was observed after the fourth repeated administration (Figure 3a). The amounts of Th1 and Th2 markers IgG2c and IgG1 were measured to determine whether the increase in IgG caused immune stimulation in Th1 and Th2 responses (Figure 3b, c). The Th1-related isotypes of IgG2c were twice as high as those of IgG1 (Figure 3d). OPGMN_OG induced significantly increased total IgG, IgG1, and IgG2c production capacity compared to GMN_OG (0 μg of OVA and 0 μg of PIC per mouse), OGMN_OG (50 μg of OVA and 0 μg of PIC per mouse), PGMN_OG (0 μg of OVA and 26 μg of PIC per mouse), and OPGMN_ID (Figure 3e–g). OPGNM showed a 2.5-fold increase in total IgG, a 3.3-fold increase in IgG1, and a 2.6-fold increase in IgG2c compared to PGMN. This enhanced antibody production capacity suggested the adjuvant capacity of OVA. These results indicate that OPGMN_OG increased IgG production by repeated administration and induced both of Th1 and Th2 signaling pathways. To clarify whether OPGMN_OG could induce the Th1 signaling pathway, we evaluated the IFN-γ and IL-2 concentrations as Th1 response markers after administering each nanoparticle type four times. Untreated mice, GMN, OGMN, and PGMN were used as control groups to clearly identify the increase in IFN-γ and IL-2 concentrations. OPGMN_OG produced a 13-fold increase in IFN-γ (Figure 3h) and a 1.4-fold increase in IL-2 (Figure 3i) compared to the untreated group. IFN-γ plays a key role in activating cellular immunity and thus stimulates antitumor immune responses. IFN-γ is considered potentially useful for adjuvant immunotherapy against various types of cancer. IFN-γ can overcome tumor progression by inhibiting angiogenesis in tumor tissues, inducing Treg apoptosis.^[28] IL-2 is one of the major cytokines that has multiple effects on the immune system.^[29] IL-2 is produced primarily by antigen-stimulated CD4⁺ T cells; however, it can also be produced by CD8⁺ T cells, NK cells, and activated DCs. IL-2 plays an important role in differentiating CD4⁺ T cells into various subsets. It can promote CD8⁺ T cells and NK cytotoxic activity and can promote naive CD4⁺ T cell differentiation by regulating the T cell differentiation program in response to antigen.^[30–31] Thus, the increase in IFN-γ and IL-2 by OPGMN may play an important role in the prevention of tumor formation by promoting immune effects against tumors. To clearly demonstrate OVA-specific effects, we collected the spleens of mice administered four doses of each sample. The spleens were collected and cultured in OVA-containing culture medium to measure the production of IFN-γ and IL-2. OPGMN_OG exhibited a 21-fold increase in IFN-γ (Figure 3j) and a 5-fold increase in IL-2 (Figure 3k) concentration compared to the untreated group. These results suggest that the direct lymph node delivery method by oral administration of OPGMN induces both Th1/Th2 responses and maximizes the immune activation effect.

Figure 3.

a) Total IgG, b) IgG1, and c) IgG2c concentration (conc.) in serum (n=4). GMN, OPGN, and OPGMN were weekly administered OG in C57BL/6 mice. OPGMN was weekly ID injection in C57BL/6 mice. d) IgG2c versus IgG1 ratio of OPGMN_OG. e) Total IgG, f) IgG1, and g) IgG2c concentration (conc.) in serum after 4-times sample administration (n=4). h) IFN-γ and i) IL-2 conc. in serum after 4-times sample administration in C57BL/6 mice. OVA-specific j) IFN-γ and k) IL-2 conc. in spleen culture medium (n=4). GMN, OGMN, PGMN, and OPGMN were weekly administered OG or ID in C57BL/6 mice. After 4-times weekly administration of each sample, the spleen was harvested and incubated for 2 days with OVA (0.1 mg/mL). *P value < 0.05. **P value < 0.01. ***P value < 0.001.

2.5. In Vivo Cell Population of Cytotoxic T cells and T_CM

We evaluated how the increased IFN-γ and IL-2 levels induced by OVA-specific responses affected the number of cytotoxic T cells and T_CM. GMN, OGMN, PGMN, and OPGMN were administered OG weekly a total of four times. One week after each sample administration, peripheral blood mononuclear cells (PBMCs) were obtained to identify the number of CD3⁺CD8⁺ T cells and CD44^highCD62L^low T_CM. Oral administration of OPGMN resulted in a significantly increased CD3⁺CD8⁺ cytotoxic T cell population compared to the untreated group (Figure 4a–c and Figure S4). The number of CD3⁺CD8⁺ T cells increased 1.6- and 2.2-fold compared to the untreated group when OPGMN was repeatedly administered 3 times (Figure 4b) and 4 times (Figure 4c). The CD44^highCD62L^low T_CM was observed to maintain a 2.4- to 2.8-fold increase in the OPGMN group compared with the nontreated group after 4 repeated administrations (Figure 4d–f and Figure S5). OGMN and PGMN induced a slight increase in the population of CD3⁺CD8⁺ T cells and CD44^highCD62L^low T_CM. However, the increased population of CD3⁺CD8⁺ T cells and CD44^highCD62L^low T_CM was significantly lower than that of the OPGMN group. The numbers of CD3⁺CD8⁺ T cells and CD44^highCD62L^low T_CM populations in the GMN were similar to those in the untreated group, which meant that the nanoparticles without OVA and PIC did not introduce any immunity. These results demonstrated that repeated oral administration of OPGMN clearly increased the population of CD3⁺CD8⁺ T cells and CD44^highCD62L^low T_CM. This increase in the number of CD3⁺CD8⁺ T cells might have been caused by a variety of concomitant processes, such as DC maturation and cytokine production, and an increase in CD44^highCD62L^low T_CM.

Figure 4.

Flow cytometry analysis of CD3⁺CD8⁺ T cells and CD44^highCD62L^low T_CM in C57BL/6 mice. GMN, OGMN, PGMN, and OPGMN were administered OG weekly to C57BL/6 mice (n=4). The statistical significance of the CD3⁺CD8⁺ T cell population at a) 2 weeks, b) 3 weeks, and c) 4 weeks postadministration. The statistical significance of the CD44^highCD62L^low T_CM population at d) 2 weeks, e) 3 weeks, and f) 4 weeks postadministration. Each week postvaccination, PBMCs were collected and stained with anti-CD3, anti-CD8, anti-CD44, and anti-CD62L antibodies. CD16/32 antibody was used to block NK cells, B cells, monocytes, granulocytes and platelets. A fixed number of cells (1 × 10⁵) per sample was measured using a flow cytometer. *P value < 0.05. **P value < 0.01. ***P value < 0.001.

2.6. Inhibition Efficacy of Immune Preactivation Timing-Dependent Tumor Development

In this study, we evaluated whether an immune response preactivated by OPGMN could prevent tumor formation because the immune-triggering effects differed according to the frequency of OPGMN administration. In addition, the tumor prevention effect in CT26 colorectal cancer was also evaluated whether the immune response pre-activated by OPGMN could be applied to the prevention of various tumorigenesis as well as B16F10 melanoma. Administration of OPGMN was started at different times according to the schedule shown in Figure 5a, and a total of four doses were administered at weekly intervals. On Day 0, 5 × 10⁴ counts of B16F10 cells and CT26 cells were inoculated into each male C57BL/6 mouse and Balb/c mouse, respectively. For the control group, PBS was administered weekly OG from 2 weeks before cancer inoculation. OPGMN exhibited significant tumor growth inhibitory effects in both B16F10 melanoma and CT26 colorectal adenocarcinoma when starting dosing before cancer inoculation (Before 1 week and Before 2 weeks) than when starting dosing simultaneously (Day 0) or after (After 1 week) cancer inoculation (Figure 5b–d). When OPGMN was administered 2 weeks before, the formation of B16F10 melanoma was prevented by 100% (Figure 5e) and colorectal carcinoma was prevented by 60% (Figure 5f). Notably, in the Before 2 weeks group, despite the administration of OPGMN ending after 7 days of cancer inoculation, no melanoma was found until after 33 days of cancer inoculation. Individual tumor volume measurements revealed tumor occurrence in all PBS-treated mice and 0–60% prevention of tumor formation in the After 1 week, Day 0, and Before 1 week groups (Figure S6). In this study, the prevention efficacy of tumor formation by OPGMN was evaluated in B16F10-bearing C57BL/6 mice and CT 26-bearing Balb/c mice that did not express OVA. Of note, the immune-triggering effect of OPGMN is a nonspecific cancer type effect. The reason OPGMN is applicable for the treatment of tumors that do not express OVA can be explained by the role of PIC. PIC is a Toll-like receptor-3 agonist that was developed to mimic pathogen infection and boost immune system activation to promote anticancer therapy.^[32–33] PICs can be used alone for tumor treatment and to control the immune response to antigens as adjuvants.^[34–37] These different tumor-forming prevention effects in melanoma and colorectal adenocarcinoma may in part be relevant to the immune environment of tumors because B16F10 melanoma in C57BL/6 mice is characterized by lower tumor-mediated immune tolerance than CT26 colorectal adenocarcinoma in Balb/c mice, which has been shown to increase immunity when the growth environment of the tumor is altered by the drug.^[38] Repeated oral administration of OPGMN did not induce mouse weight loss or liver dysfunction regardless of administration timing (Figure 5g,h and Figure S7). To clearly demonstrate that OPGMN exhibited a complete melanoma prevention effect in the Before 2 weeks dosing group, the OVA, PIC, and OVA/PIC bulk materials started dosing 2 weeks before cancer inoculation, and the tumor occurrence was monitored. Tumors were observed in all mice with the same dosing schedule as the Before 2 weeks group (Figure S8). These results suggest that prior immune activation by OPGMN prevented subsequent tumor formation and improved tumor prevention efficacy compared to conventional injectable cancer vaccines.^[39–41] Thus, we believe that oral administration of OPGMN exhibits sufficient prevention of tumor development by non-specific immune preactivation, and we expect it to be applicable to various types of cancer recurrence prevention.

Figure 5.

Vaccination timing-dependent tumor-forming prevention. a) Timeline for the cancer inoculation and OPGMN vaccination. OPGMN was administered depending on the vaccination timeline. Tumor volume measurement of b) B16F10 tumor-bearing C57BL/6 mice and c) CT26 tumor-bearing Balb/c mice (n=5). d) Images of B16F10 tumor-bearing mice after 10- and 18-days cancer inoculation. Red circles indicate tumor region. Individual tumor volume measurement of Before 2 weeks of OPGMN administered in e) B16F10 tumor-bearing C57BL/6 mice and f) CT26 tumor-bearing Balb/c mice. Body weight measurement of g) B16F10 tumor-bearing mice and h) CT26 tumor-bearing Balb/c mice (n=5). *P value < 0.05. **P value < 0.01.

2.7. In Vivo Cell Population of Cytotoxic T cells, T_CM, and Tregs

To clarify the prevention effect of tumor formation on OPGMN administration timing, we evaluated the population of cytotoxic T cells, T_CM, Tregs, and NK cells in PBMCs 21 days after cancer inoculation following the dosing schedule, as shown in Figure 5a. The number of CD3⁺CD8⁺ T cells increased 1.7-fold in the Before 2 weeks group compared to the PBS group (Figure 6a). The number of CD44^highCD62L^low T_CM increased 1.6-fold in the Before 2 weeks group compared to the PBS group (Figure 6b). The number of CD4⁺CD25⁺Foxp3⁺ Tregs decreased 9.2-fold in the Before 2 weeks group compared to the PBS group (Figure 6c). The number of CD11b⁺CD27⁺ NK cells increased 6.9-fold in the Before 2 weeks group compared to the PBS group (Figure 6d). When dosing begins with OVA, PIC, and OVA/PIC bulk materials before 2 weeks of cancer inoculation, there was no difference among the numbers of CD3⁺CD8⁺ T cells, CD44^highCD62L^low T_CM, CD4⁺CD25⁺Foxp3⁺ Tregs, and CD11b⁺CD27⁺ NK cells (Figure 6e–h). These results suggest that the number of Tregs decreased the most and the number of NK cells, cytotoxic T cells, and T_CM increased the most when OPGMN was administered 2 weeks before tumor inoculation. Further studies are needed to elucidate which immune signaling pathway is responsible for these phenomena.

Figure 6.

Flow cytometry analysis of a) CD3⁺CD8⁺ T cells, b) CD44^highCD62L^low T_CM, c) CD4⁺CD25⁺Foxp3⁺ Tregs, and d) CD11b⁺CD27⁺ NK cells in OPGMN-administered B16F10 tumor-bearing C57BL/6 mice. OPGMN was administered OG at different timings from 2 weeks before to 1 week after transplantation. Flow cytometry analysis of e) CD3⁺CD8⁺ T cells, f) CD44^highCD62L^low T_CM, g) CD4⁺CD25⁺Foxp3⁺ Tregs, and h) CD11b⁺CD27⁺ NK cells in the PBMCs of OVA, PIC, and OVA/PIC mice administered OG weekly from 2 weeks before cancer inoculation (n=5). CD45⁺CD3^-CD56⁺ cells were gated as NK cells. A fixed number of cells (1 × 10⁵) per sample was measured using a flow cytometer for CD3⁺CD8⁺ T cells, T_CM, CD4⁺CD25⁺Foxp3⁺ Tregs, and CD11b⁺CD27⁺ NK cells. *P value < 0.05. **P value < 0.01. ***P value < 0.001.

3. Conclusion

Here, we described an oral cancer vaccine that exhibits desirable immune-triggering effects to prevent melanoma development/recurrence. Repeated oral administration of OPGMN induced the maturation of DCs and promoted both Th1 and Th2 immune responses. OPGMN increased the number of T_CM, cytotoxic T cells, and NK cells while decreasing that of Tregs according to the timing of administration. Repeated administration of OPGMN before tumor development inhibited 100% of B16F10 melanoma formation and 60% of CT26 colorectal adenocarcinoma formation, which was maintained even after the end of OPGMN administration. OVA and PIC loaded OPGMN presented effective immune response through the small intestinal lymphatic system, which can be expanded to various and multiple antigen-specific immunities using an oral delivery system for cancer treatment. These results indicate that OPGMN may be applicable for the prevention of various tumor metastasis and recurrence in patients after various cancer treatments, such as chemotherapy, radiotherapy, and surgery.

4. Experimental Section

Synthesis of CSG:

Prior to conjugation of GCA directly to CS (10–40 kDa, injectable grade, Yantai Dongcheng Biochemicals Co., Ltd. Shandong, China), the carboxyl group of GCA was modified with ethylenediamine (EDA, Sigma, St. Louis, MO, USA). GCA (3 g, 6.44 mmol) was dissolved in 30 mL of dimethylformamide. Excess N,N’-dicyclohexylcarbodiimide (DCC, 1.73 g, 8.38 mmol, Sigma, St. Louis, MO, USA), NHS (960 mg, 8.38 mmol, Sigma, St. Louis, MO, USA), and EDA (5.10 g, 322 mmol) were added to the GCA solution. The mixture was stirred at 50°C for 24 hours. After the reaction, dicyclohexylurea and unreacted EDA were removed by filtration through a 0.22 μm polyethersulfone syringe filter and rotary evaporation at 80°C. The remaining solution was precipitated in ethyl acetate, and the pellet was collected. The pellet was dissolved in dimethylformamide and precipitated in acetonitrile. The precipitate was dried in vacuo for 24 hours. The collected pellet of aminated GCA (GCA-EDA) was dissolved in 10 mL of DW, filtered through a 0.45 μm polytetrafluoroethylene syringe filter and lyophilized. CS (500 mg, 0.056 mmol) was dissolved in 5 mL of 0.1 M MES buffer at pH 6.0. Then, EDC (642 mg, 3.3 mmol, ProteoChem, Denver, CO, USA) and NHS (384 mg, 3.3 mmol) were added to the CS solution and stirred for 30 minutes. GCA-EDA (1.35 g, 2.6 mmol) was dissolved in 15 mL of 0.1 M MES buffer at pH 6.0, added to the CS solution and stirred for 24 hours. The reacted solution was precipitated three times in 60% cold ethanol. The precipitate was dried in vacuo for 24 hours. The synthesized CSG was characterized by the degree of substitution with Varian Mercury 400 ¹H NMR spectrophotometers (Varian Inc., Palo Alto, CA, USA).

Synthesis of Mannose-g-Chondroitin Sulfate (CSM):

CS (1 g, 0.11 mmol) was dissolved in 10 mL of 0.1 M MES buffer at pH 6.0. Then, EDC (500 mg, 2.6 mmol) and NHS (300 mg, 2.6 mmol) were added to the CS solution and stirred for 30 minutes. 4-Aminophenyl α-D-mannopyranoside (mannose, 710 mg, 2.6 mmol, Sigma, St. Louis, MO, USA) was dissolved in 7 mL of 0.1 M MES buffer at pH 6.0, added to the CS solution and stirred for 24 hours. The reacted solution was precipitated in 60% cold ethanol. The precipitate was dried in vacuo for 24 hours. The synthesized CSM was characterized by the degree of substitution using a Varian Mercury 400 ¹H NMR spectrophotometer.

Preparation of CSG- and CSM-Coated OVA- and PIC-Encapsulated Nanoparticles:

Thirty-eight milligrams of 1,2-dioleoyl-sn-glycerol-3-phosphocholine (DOPC, Echelon Bioscience, Salt Lake City, UT, USA) and 39 mg of 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP, CordenPharma, Plankstadt, Germany) were dissolved in 10 mL of chloroform. To make a thin film on the 250 mL round bottom flask, the solvent was removed by rotary evaporation at 40°C. The film was rehydrated in 10 mL of DW using a VWR Scientific Aquasonic 75 T Ultrasonic cleaner for 5 min at RT. The liposome solution was passed through 100 nm and 200 nm filters (Whatman Nuclepore, Maidstone, UK) at 60°C using a membrane extruder LiposomeFast^® LF-50 (Avestin, Inc., ON, Canada). Chicken egg white ovalbumin (OVA, Sigma, St. Louis, MO, USA) and PIC (low molecular weight, InvivoGen, San Diego, CA, USA) were dissolved in DW and added to the liposome solution and then shaken at 50 rpm for 2 hours. GMN, OGMN, PGMN, OPCN, OPGN, and OPGMN were prepared by mixing cationic charged nanoparticles with anionic charged CS, CSG, or a mixture of CSG and CSM by electrostatic interactions. In brief, 10 mL of OVA- or/and PIC-incorporated CL was mixed in 2 mL of CS or CSG (80 mg/mL) or a 1:1 mixture of CSG and CSM (80 mg/mL) and then stirred for 2 min. The unloaded OVA and PIC were removed by dialysis against DW (cutoff Mw 50 kDa membrane). The loading efficiency of OVA and PIC was evaluated using a BCA protein assay kit (Fisher Scientific Test Kit Pierce™, Pittsburgh, PA, USA) and propidium iodide (Sigma, St. Louis, MO, USA), respectively ^[42]. The stability of OPGN in the GIT simulated fluids was evaluated in DW, pH 1.8 FaSSGF, and pH 6.5 FaSSIF (n=3). The size and zeta potential of each liposome solution were measured using dynamic light scattering (Zetasizer 3000, Malvern Instruments, USA).

ASBT-Mediated Cellular Uptake:

The ASBT-mediated cellular uptake efficacy of OPGN was estimated in ASBT-expressing SK-BR-3 (human breast carcinoma, ATCC^® HTB-30™) cells. The cells were maintained in RPMI-1640 medium containing 10% fetal bovine serum and 1% streptomycin-ampicillin under 5% CO₂ at 37°C. The cells were seeded into 6-well plates at a density of 1 × 10⁵ cells per well. Alexa 647-labeled OVA (Thermo Fisher Scientific, Waltham, MA, USA) was encapsulated on the CL and coated with CS and CSG. Prior to sample treatment of each cell, all samples were diluted in serum-free RPMI-1640 medium. The cells were treated with OVA, OPCN, and OPGN and incubated for 30 minutes. After 30 minutes of incubation, the cells were washed with PBS and analyzed using a flow cytometer.

Ex vivo Transport Efficacy in the Small Intestine:

All animal experiments were approved by the University of Utah’s Animal Care and Use Committee. To perform the everted gut sac (EGS) assay, the small intestine was harvested from 200–225 g male Sprague–Dawley (SD) rats (Charles River Laboratories International, Inc.). The duodenum, jejunum, and ileum were separated into 5 cm lengths, and each segment was placed with the luminal side out. After one end of the segment was tied, the inner space was filled with oxygenated culture medium, and another end was tied to make an EGS with a length of 4 cm. EGS was placed in a Krebs-Ringer solution containing OVA, OPCN, and OPGN (Alexa 647-labeled OVA concentration = 50 μg/mL) at 37°C for 90 minutes. The amount of OVA transported per unit tissue area was analyzed by fluorescence intensity (excitation/emission; 651/667 nm) using Alexa-647 calibration curves in Krebs-Ringer solution.

Oral Lymphatic Delivery of OPGMN:

OPGN can be delivered to LNs through OG. Alexa 647-labeled OVA and rhodamine-labeled PIC (InvivoGen, San Diego, CA, USA) were encapsulated on the CL and coated with CS and CSG. Prior to OPCN and OPGN oral administration, mice were fasted for 6 hours before and 1 hour after administration with free access to water. Fluorescence-labeled OPCN and OPGN were administered ID or OG to 5-week-old C57BL/6 mice. After 3 hours and 24 hours of administration, LNs were harvested and cryo-dissected at a thickness of 10 μm. The sections were observed using confocal microscopy. To statistically compare the lymph node targeting efficacy of OPGN, 3 hours post-administration of OPCN and OPGN, the axillary, brachial, mesenteric, and LNs were harvested and meshed with a cell strainer (n=4). To evaluate the absorption pathway of OPGN_OG, C57BL/6 mice were pretreated intraperitoneal with 3 mg/kg of cycloheximide (OPGN_OG + cycloheximide) dissolved in saline (0.5 mg/mL). Harvested samples were centrifugated at 1,500 rpm for 10 minutes, separate supernatant and cell pellet. The cell pellet was lysis with lysate buffer. The Alexa-647 OVA fluorescence intensity of the mixture of supernatant and lysate solution was measured at excitation 650 nm and emission 665 nm. To minimize the fluorescence interference of tissues, the fluorescence intensity of non-treated mice was normalized to all sample values (n=4).

In Vivo DCs Maturation:

To evaluate DCs maturation, GMN, OPGN, and OPGMN were administered ID or OG to 5-week-old male C57BL/6 mice (n=4). Seven days postadministration, mice were sacrificed, and axillary LNs were harvested and passed through a 70-μm BD cell strainer. LN cells (1 × 10⁶) were stained with anti-CD11c and anti-MHC class II antibodies (Biolegend, San Diego, CA, USA). DC maturation was measured by staining with anti-CD40, anti-CD80, and anti-CD86 antibodies (Biolegend). A total of 1 × 10⁵ cells (without live-dead marker) per sample were measured by flow cytometry and analyzed with FlowJo software.

In Vivo Antibody and Cytokine Production:

To evaluate the antibody efficacy against repeated nanoparticle administration, OPGN and OPGMN were administered weekly to 5-week-old C57BL/6 mice (n=4). GMN was administered weekly to 5-week-old C57BL/6 mice as a control group. To administer OG, mice were fasted for 6 hours before and 1 hour after administration with free access to water. After one week postadministration, mouse serum was collected and analyzed using IgG, IgG1, and IgG2c enzyme-linked immunosorbent assay (ELISA, Invitrogen, Carlsbad, CA, USA). After four administrations of each sample, mouse serum and spleen were collected and analyzed with IFN-γ and IL-2 ELISA kits (Invitrogen, Carlsbad, CA, USA). To evaluate the efficacy of OVA-specific IFN-γ and IL-2 production, the harvested spleen was cultured with OVA (0.1 mg/mL) containing RPMI 1640 medium for 2 days. After 2 days of incubation, the cell culture medium was collected and analyzed using IFN-γ and IL-2 ELISA kits (n=4).

In Vivo Cytotoxic T cell and T_CM Populations:

GMN, OGMN, PGMN, and OPGMN were administered OG weekly to 5-week-old C57BL/6 mice (n=4). To administer OG, mice were fasted for 6 hours before and 1 hour after administration with free access to water. Each week postadministration, PBMCs were collected from facial vein and lysis the red blood cells. The PBMCs were stained with anti-CD8, anti-CD16/32, anti-CD44, and anti-CD62L antibodies (Biolegend). Anti-CD16/32 antibody (Biolegend) was used to block NK cells, B cells, monocytes, granulocytes, and platelets. A fixed number of cells (1 × 10⁵, without live-dead marker) per sample was measured using a flow cytometer and analyzed with FlowJo software.

Administration Timing-Dependent Tumor Development Inhibition:

To establish a tumor-bearing mouse model, B16F10 cells (murine melanoma, ATCC^® CRL-6475™) and CT26 (murine colon carcinoma, ATCC^® CRL-2638™) were implanted into 6-week-old male C57BL/6 and Balb/c mice, respectively. Briefly, 5 × 10⁴ cells in 100 μL of serum-free culture medium were subcutaneously injected into each mouse. Solutions of PBS and OPGMN were administered OG before 2 weeks, before 1 week, 0 days, after 1 week, and after 2 weeks of cancer inoculation (n=5). To administer OG, mice were fasted for 6 hours before and 1 hour after administration with free access to water. Tumor volumes were calculated using the following equation: volume = 0.5 × L × W², where “W” and “L” are the width and length of the tumor, respectively. For in vivo flow cytometry analysis, after 21 days of cancer inoculation, PBMCs were collected and stained with anti-CD3, anti-CD4, anti-CD8, anti-CD11b, anti-CD27, anti-CD44, anti-CD45, anti-CD56, anti-CD62L, anti-CD25, and anti-Foxp3 antibodies (Biolegend). A fixed number of cells (1 × 10⁵, without live-dead marker) per sample was measured using a flow cytometer for CD3⁺CD8⁺ cells, CD44^highCD62L^low T_CM, CD4⁺CD25⁺Foxp3⁺ Tregs, and CD11b⁺CD27⁺ NK cells.

In vivo Toxicity Study:

After four rounds of OPGMN weekly medication in 5-week-old C57BL/6 mice, serum and liver were harvested to assess liver function and toxicity. The collected serum was analyzed for a liver function index, including albumin, total protein, total bilirubin, aspartate aminotransferase (AST), alanine aminotransferase (ALT), and alkaline phosphatase (ALP), using a Hitachi 7180 automatic biochemical analyzer. Liver toxicity was analyzed using hematoxylin and eosin (H&E) staining.

Statistical Analysis:

The results from the in vitro studies are expressed as the mean ± standard deviation (SD). The results from the ex vivo and in vivo studies are expressed as the mean ± standard error of the mean (SEM). Differences between the values were assessed using one-way ANOVA and Tukey multiple comparison test.