Enhancing the therapeutic efficacy of CpG oligonucleotides using biodegradable microparticles

Enhancing the therapeutic efficacy of CpG oligonucleotides using biodegradable microparticles

Advanced Drug Delivery Reviews 61 (2009) 218–225 Contents lists available at ScienceDirect Advanced Drug Delivery Reviews j o u r n a l h o m e p a ...

575KB Sizes 0 Downloads 13 Views

Advanced Drug Delivery Reviews 61 (2009) 218–225

Contents lists available at ScienceDirect

Advanced Drug Delivery Reviews j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / a d d r

Enhancing the therapeutic efficacy of CpG oligonucleotides using biodegradable microparticles ☆ Padma Malyala ⁎, Derek T. O'Hagan, Manmohan Singh Novartis Vaccines & Diagnostics, 350 Massachussets Ave., Cambridge MA, 02139, USA

a r t i c l e

i n f o

Article history: Accepted 15 December 2008 Available online 11 January 2009 Keywords: Immunopotentiator Antigen Polylactide-co-glycolide CpG Charged particles

a b s t r a c t Oligonucleotides, with specific sequence surrounding CpG motifs, appear to be very effective for the induction of a potent Th1 responses. This molecule represents pathogen-associated molecular patterns (PAMPs) that allows the pathogen recognition receptors (PRRs) present on innate immune cells to recognize them and become activated. PAMPs and related compounds are often labelled as immunopotentiators, allowing a clear distinction between them and particulate delivery systems such as emulsions, liposomes, virus-like particles and microparticles. Microparticles prepared from biodegradable, biocompatible polyesters, and poly (lactide co-glycolide) (PLG). They have been proven to be a good particulate delivery system for the co-delivery of antigens and adjuvants. PLG has been used in humans for many years as a resorbable suture material and controlled-release drug delivery systems. It has been demonstrated that antigen presenting cells (APCs) efficiently uptake the PLG microparticles (∼ 1 μm) both in vivo and in vitro. After uptake, the PLG subsequently induces an antigen specific CTL response in rodents. Several groups, including our group, have evaluated CpG as an immunopotentiator in various formulations and delivery systems (i.e. emulsions and particulate systems). This review will discuss in detail the work conducted so far with CpG using PLG microparticles as a delivery system. We will also discuss the advantages and enhancement of immune properties of formulating CpG (soluble, adsorbed, and encapsulated forms) with PLG microparticles along with future directions for these microparticles with CpG. © 2009 Elsevier B.V. All rights reserved.

Contents 1.

Introduction . . . . . . . . . . . . . . . . . . . . . 1.1. CpG . . . . . . . . . . . . . . . . . . . . . . 1.2. PLG . . . . . . . . . . . . . . . . . . . . . . 1.3. Formulation . . . . . . . . . . . . . . . . . . 2. CpG with PLG microparticles . . . . . . . . . . . . . 2.1. CpG in soluble form with PLG microparticles. . . 2.2. CpG in adsorbed form with PLG microparticles . 2.3. CpG in encapsulated form in PLG microparticles . 2.4. Encapsulated CpG with GBS . . . . . . . . . . 3. CpG in soluble form . . . . . . . . . . . . . . . . . 4. CpG with other delivery systems . . . . . . . . . . . 4.1. PLG nanoparticles and Liposomes . . . . . . . . 4.2. Alum . . . . . . . . . . . . . . . . . . . . . 4.3. Other nanoparticulate delivery systems . . . . . 4.4. MF 59. . . . . . . . . . . . . . . . . . . . . 5. Future directions . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

219 219 219 219 219 219 220 221 221 222 223 223 223 223 224 224 224

☆ This review is part of the Advanced Drug Delivery Reviews theme issue on “CpG Oligonucleotides as Immunotherapeutic Adjuvants: Innovative Applications and Delivery Strategies.” ⁎ Corresponding author. Novartis Vaccines & Diagnostics, 350 Massachussets Ave., Cambridge MA, 02139, USA. Tel.: +1 510 923 8477; fax: +1 510 923 2586. E-mail address: [email protected] (P. Malyala). 0169-409X/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.addr.2008.12.009

P. Malyala et al. / Advanced Drug Delivery Reviews 61 (2009) 218–225

1. Introduction Modern vaccine formulations are steering away from live attenuated viruses and bacterial toxoids to more defined and purified recombinant subunit proteins. Although these antigens are highly relevant for sero protection, often they are poorly immunogenic due to lack of an innate immune stimulus and need an adjuvant to obtain an effective immune response. A delivery vehicle that can package and transport the antigen and adjuvant to the targeted site of action is essential for an ideal vaccine formulation Vaccine formulations are thus composed of three critical components: antigen, adjuvant, and delivery system. Antigen(s) are defined as a protein against which adaptive immune responses are elicited. An adjuvant is able to stimulate the innate immune system. A delivery system is defined as a vehicle to ensure optimal presentation to both innate and adaptive immune systems. 1.1. CpG An effective vaccine formulation should have the capability to produce both neutralizing antibodies and a potent T-cell response [1]. The chemical properties of the adjuvant determine coordinated priming, clonal expansion, T-cell effector function, and memory pool while the physical properties of antigen determine the uptake of antigen, which leads to MHC-restricted antigen presentation [1]. Production of co-stimulatory molecules and soluble mediators, such as CD40, B7 family, IL12, and IFN-⁎ are dependent on the adjuvant [1]. CpG stimulates the innate immune system by different mechanisms such as modifying physical properties of an antigen, resulting in enhanced antigen uptake [2,3] or assist in transport into draining lymph nodes. Adjuvants also act by the mechanism of targeting pattern recognition receptors (PRRs) such as toll like receptors (TLRs), NOD-like receptors (NLRs), cytosolic receptors, and membrane-associated lectins [4]. CpG appears to be very effective for the induction of potent Th1 responses, in contrast to Alum which is a Th2 inducer [21]. The molecule, bacterial DNA (CpG containing optimized oligo sequences), represents a pathogen-associated molecular pattern (PAMP) [5–7]. It targets toll like receptor 9(TLR9) receptors, which activates maturation of dendritic cells and facilitates cross presentation of antigens [8]. In humans, B cells and plasmacytoid dendritic cells are the only immune cells that are known to express TLR9 and to be activated by CpG. In mice, TLR9 is broadly expressed on both the major dendritic cell subtypes, plasmacytoid and myeloid dendritic cells, as well as in B cells, macrophages, and monocytes [9]. These differences in mouse and human models makes it challenging to extrapolate mouse data to humans, when evaluating CpG as an adjuvant [9].

219

Polylactide-co-glycolide (PLG), a biodegradable polymer, is a prime choice to synthesize microparticles. It is a biocompatible polyester and has extensive safety data in humans, as it has been used as resorbable suture material and as a controlled-release drug delivery system. Moreover, it is feasible to co-deliver both the antigen and adjuvant (antigen absorbed or entrapped) with PLG microparticles [12]. 1.3. Formulation To ensure optimum presentation to both innate and adaptive immune systems, the three key components of a vaccine (antigen(s), adjuvant, and delivery vehicle) need to be effectively formulated. Furthermore, safety is likely to improve by formulation, which is especially important for novel adjuvants. PLG microparticles are typically synthesized by a double emulsion method to prepare a water in oil in water emulsion followed by solvent evaporation [13]. CpG has been formulated with PLG microparticles in different ways. It has been added in soluble form, adsorbed to charged PLG microparticles, and encapsulated within the microparticles. These different types of formulations, along with their methods of preparation and immunogenicity effects, will be discussed individually in the following sections. 2. CpG with PLG microparticles 2.1. CpG in soluble form with PLG microparticles Our group has modified PLG microparticles, by adding anionic surfactants, to develop charged microparticles [12]. The charge on the surface of the microparticles aids in the surface adsorption of antigens. Typically, antigen is adsorbed onto charged microparticles by overnight adsorption at 4 °C in buffered PLG microparticles and freeze dried with lyo protectants [12]. Soluble CpG is added in the soluble form to reconstituted lyophilized antigen adsorbed PLG microparticles at the time of immunization [10]. CpG added in soluble form is little or not associated with PLG. An in vitro release at time zero indicates a release of 85–90% of CpG with the rest being released in 24 h (Fig. 1) [10]. CpG in soluble form has been evaluated in several studies with a range of antigens and has provided a varied immune response, depending on the antigen and the delivery vehicle. Studies evaluating mucosal and humoral responses obtained from mice and rat studies with intranasal and parenteral routes and influenza, meningitis, HIV and dinitrophenylated bovine serum albumin antigens are discussed in this section.

1.2. PLG As stated earlier, a suitable vaccine delivery system will enable the packaging and transportion of the antigen and adjuvant to the targeted site of action, as well as optimize its presentation to the innate and adaptive immune systems. Particulate systems serve the function of a delivery system due to similarities in size to pathogens that they combat. Particulate carriers are typically less than 5 μm in size and are thus taken up by the antigen presenting cells (APC), macrophages, and dendritic cells [10]. Microparticles enable self assembly of particles and also offer the advantage of presenting multiple copies of antigens on their surface, resulting in B cell activation [11]. In addition to presenting multiple copies, the particulate delivery system traps and retains the antigen in local lymph nodes and protects it from degradation, ensuring longer duration. Microparticles co-deliver adjuvants to the targeted site of action and prevent systemic distribution at the site of injection. There are different particulate systems, such as, those from biodegradable polymers, liposomes, virus-like particles, and emulsions.

Fig. 1. In vitro release of CpG from four types of CpG formulations: soluble, adsorbed, 50% encapsulated, and 100% encapsulated. Release of CpG from microparticles was determined by reconstituting lyophilized PLG/CpG formulations with water and incubating at room temperature for 4 weeks. Samples were collected at time zero, 7 days,14 days, 21 days, and 28 days. The PLG particles were separated by centrifugation and the amount of CpG present in the supernatant was determined by UV spectroscopy.

220

P. Malyala et al. / Advanced Drug Delivery Reviews 61 (2009) 218–225

In a mouse study evaluating two antigens, HIV-1 gp 120 protein and protein from Neisseria meningitides serotype B (Men B) with PLG microparticles, Kazzaz et al. compared the adjuvant effect of encapsulated MPL and RC 529 PLG formulations to their soluble counterparts and soluble CpG [14]. The encapsulated formulations were made by encapsulating MPL or RC 529 in PLG microparticles and the encapsulated adjuvants mixed with antigen adsorbed PLG microparticles. The study found that the functional titers for Men B antigen with encapsulated MPL and RC529 were comparable to soluble CpG added to PLG microparticles (Table 1). With gp120 antigen, IgG responses from encapsulated MPL and RC529 were comparable or higher to soluble CpG formulation. However, when soluble CpG group was compared to soluble MPL or soluble RC529, the soluble CpG group had the highest HIV-specific IgG and IgG2a titers. Also, soluble MPL and RC529 had poorer antibody responses when compared to their formulated counterparts (data not shown). Wack et al. evaluated CpG in soluble form with influenza proteins as the antigen and delivery vehicles were PLG, MF59, Alum, and Calcium Phosphate [14]. Although PLG with CpG did produce higher functional titers in mice when compared to PLG/Flu alone, MF59 with CpG yielded highest functional titers across all three influenza strains [14]. MF59 will be discussed in detail in Section 4 along with other delivery systems. CpG has been shown to be an effective mucosal immune modulator in the rat animal model [18]. In rodents, tear IgA antibodies are important mediators for ocular immune defense. An important inductive site for tear IgA responses in rats is nasal-associated lymphoid tissue (NALT). Gill and Montgomery determined the effect of soluble CpG on induction of rat tear IgA responses with soluble antigen and antigen encapsulated in PLG particles [18]. The antigen used was dinitrophenylated bovine serum albumin (DNP-BSA) administered intranasally (IN). Data from the study showed significant elevated levels of IgA antibodies in tears and saliva with CpG. Higher levels were elicited with antigen encapsulated in PLG microparticles and soluble CpG when compared to soluble antigen and soluble CpG [18]. A compilation of immunogenicity data for PLG/antigen with and without soluble CpG can be seen in Table 2. From the studies discussed above, it can be noted that there is an adjuvant effect observed with soluble CpG (Table 1). However, the immunogenicity effect with soluble CpG is not superior to other adjuvants when they (RC529, MPL, and QS21) are added in encapsulated form (Table 1) [14]. Moreover, the encapsulated adjuvants had higher potency when compared to their soluble counterparts, thus exemplifying the importance of adjuvant delivery. As mentioned earlier, CpG has been formulated in both adsorbed and encapsulated forms. The section below reviews the enhancement of immunogenicity with CpG adsorbed to PLG.

Table 2 Effect of soluble CpG on PLG/antigen formulations Formulations

Titer type

PLG/ antigen

PLG/antigen with CpG

With HIV antigen, gp 120 With flu antigen With DNP-BSA

Anti-gp120 IgG2a/IgG titer ratio Functional HI titers (anti H3N2) ng Ag-specific IgA/μg total IgA (8 days post two IN imm.) Functional BCA titers

1.24 700–1000 20–40

103.15 1500–2000 80–120

64

512

With Men B

A comparison of immunogenicity indicating titers for PLG/antigen with and without soluble CpG from different studies is presented [15–17].

2.2. CpG in adsorbed form with PLG microparticles Several groups sought to formulate CpG with PLG delivery system to improve its potency by promoting the uptake and delivery of the adjuvant into APCs. Singh et al. utilized charged microparticles to adsorb CpG on the surface of PLG microparticles and co deliver CpG adsorbed on cationic PLG along with antigen adsorbed to anionic PLG. CpG was surface adsorbed onto PLG microparticles by overnight rocking at 4 °C and lyophilized with lyo protectants [19]. The reconstituted lyophilized CpG adsorbed cationic PLG microparticles are admixed with antigen adsorbed anionic PLG microparticles at the time of immunization. An in vitro release assay at time zero of CpG adsorbed formulation indicates 92% adsorption of CpG to PLG microparticles. This is in contrast to CpG added in soluble form which is little or not associated with PLG. However, in vitro release data of CpG adsorbed formulation at 24 h indicates a 78% release of CpG, leaving about 22% adsorbed to PLG microparticles. CpG adsorbed has been evaluated with bacterial and viral protein antigens and the studies are discussed in this section. In Singh et al.'s study, they examined the ability of CpG adsorbed to elicit antibody and cytotoxic T lymphocyte (CTL) responses to HIV antigens, p55 gag and gp120, following intramuscular immunization in mice. The groups in the study for both antigens were soluble antigen mixed with soluble CpG, antigen adsorbed to PLG with no adjuvant, CpG adsorbed to PLG with soluble antigen, and PLG with antigen adsorbed mixed with PLG/CpG adsorbed. A significant adjuvant effect of antibody (Fig. 2) and CTL responses (data not shown) were observed with the group, PLG with antigen adsorbed mixed with PLG/CpG adsorbed, when compared to the other groups. Another antigen evaluated for CpG adsorbed formulations was in a collaboration study with Center for Biologics Evaluation Research (CBER) and U.S. Army Medical Research Institute of Infections and

Table 1 Immunogenicity comparison of soluble CpG with encapsulated adjuvant formulations Formulations

Men B SBA (functional titers)

gp120 IgG2a/ IgG titer ratio

gp120 serum cytokine responses pg/ml

Hep B % seroprotection post 1st dose

Soluble CpG Encapsulated RC529 Encapsulated MPL Encapsulated QS21

4096 8192

103.15 172.28

0–25 –

40% –

8192

84.62

200–250

81.60%





150–225

85.70%

Antigens used in the studies were Men B, gp120, and Hep B. Immunogenic functional titers for Men B, antibody titers ratio of anti-gp120 IgG2a/IgG serum IgG antibody titers, cytokine responses for gp120, and percent seroprotected against Hep B post immunization are tabulated [15–17].

Fig. 2. Effect of adsorbed CpG on immunogenicity of antigen adsorbed on anionic PLG microparticles. Graph depicts serum IgG titer ratio of PLG/antigen with and without CpG [10].

P. Malyala et al. / Advanced Drug Delivery Reviews 61 (2009) 218–225

Novartis was anthrax vaccine (AVA) [20]. Anthrax vaccine was adsorbed on alum and co delivered with CpG adsorbed onto PLG. The formulation, along with relevant controls, was administered into mice as a part of immunization schedule. The mice were later challenged with lethal doses of anthrax and survival monitored for 3 weeks. The immunized mice with CpG adsorbed onto PLG with AVA were protected within 1 week of immunization and also had correlating higher serum IgG titers (data not shown) [20]. The antigen from Neisseria meningitidis serotype B (Men B), a recombinant protein, was used by Malyala et al. to investigate the immunopotentiator effect of CpG with antigen adsorbed to Poly (lactide-co-glycolide) microparticles (PLG particles) [10]. The formulations evaluated include, CpG added in Soluble form, CpG adsorbed, and CpG encapsulated in these studies. CpG was adsorbed onto cationic PLG particles as previously described. The antibody titers indicated that adding CpG in adsorbed form, using Chitosan or CTAB for forming cationic PLG microparticles, did not have any significant effect over adding CpG in Soluble form, with the Men B antigen(Fig. 2) [10]. Data from the mouse studies with adsorbed CpG on PLG microparticles suggests that adjuvant effect is antigen specific as formulation by adsorption showed higher immunogenicity with gp120, p 55 gag, and AVA but not with Men B antigen. 2.3. CpG in encapsulated form in PLG microparticles Although the adsorbed CpG formulation did not have an enhanced immunogenic effect when compared to soluble CpG in the above study, CpG encapsulated in PLG microparticles did show a four fold increase in functional titers [10]. CpG exhibits significant enhancement of immunogenicity in soluble form when compared to other adjuvants and is often used as gold standard in adjuvant evaluation. This study was undertaken to further improve the potency of CpG by encapsulating CpG in PLG microparticles and compared to soluble and adsorbed CpG formulations. CpG, being a water soluble compound, offered great challenges to encapsulate during PLG microparticles synthesis due to its natural affinity for the external water phase in the water in oil in water (w/o/w) emulsification process. Two methods, complexation and viscosity enhancement, were employed to retain CpG in the internal phase. Complexation of CpG was accomplished using cationic chitosan, and this complex was added in the water phase during the synthesis of the primary water in oil emulsion. Based on the ratio of CpG and Chitosan, encapsulation efficiencies varied. Two fixed ratios,1000:1 and 1.4:1 of CpG and Chitosan, yielded encapsulation efficiencies of 50% and 100% when a 0.5% load of CpG was targeted w/w PLG (Fig. 4). 100% encapsulation was achieved when CpG was completely complexed with Chitosan in the ratio of 1.4 and 1 respectively. An analysis of the supernatant from the centrifuged CpG–

221

Chitosan (1.4:1) complex by UV spectroscopy indicated absence of CpG in the supernatant (data not shown). In the other method of encapsulation, poly vinyl alcohol (PVA) was used as the viscosity enhancer, and CpG and PVA were mixed in the water phase prior to the primary emulsion. An encapsulation efficiency of 50% was achieved when a 0.5% load of CpG was targeted w/w PLG. An in vitro release of CpG from the formulations was done in water at room temperature under rocking to monitor its release. CpG had different release profiles based on the method of formulating CpG (Fig. 1). While the soluble and adsorbed CpG formulations had faster release profiles as mentioned before, the CpG formulation, which had about 100% encapsulation, released less than 20% CpG even at the end of 6 weeks yielding a flat release profile. CpG formulations with an initial 50% encapsulation released gradually with 100% release achieved over a period of 4 weeks. Men B protein was either surface adsorbed on CpG encapsulated PLG microparticles or on separate PLG microparticles. Presence of CpG in the microparticles did not affect the total adsorption and integrity of protein. Minimal protein was released at the end of 4 weeks in vitro release study indicating the association of protein with the microparticles. Men B adsorbed on separate PLG microparticles was reconstituted with water and mixed with lyophilized CpG encapsulated at time of immunization. The mouse animal study was conducted in four antigen doses in two separate studies. The doses evaluated were 1, 2, 10 and 20 μg and immunogenicity differences between formulations could be observed at 2 μg antigen dose, with a plateau effect observed at 10 μg and above. CpG–Chitosan complex with PLG/Men B was also evaluated in this elaborate study and the formulation showed lesser or comparable responses to soluble CpG, thus demonstrating the importance of encapsulating CpG–Chitosan complex in PLG microparticles, to ensure delivery at the site of action. However, the results obtained with CpG– Chitosan complex needs to be confirmed at the lower antigen dose of 2 μg. Soluble and adsorbed CpG formulations showed similar functional titers and were four times higher, when compared to PLG/Men B without CpG. Compared to soluble CpG, CpG encapsulated in PLG microparticles with 50% encapsulation efficiency exhibited a four fold increase in functional titers at 2 μg dose, and similar trends were observed with IgG and cytokine responses(Fig. 3) [10]. Immunogenicity of formulations with antigen, Men B on same or separate PLG microparticles was similar (data not shown). 2.4. Encapsulated CpG with GBS CpG was co-encapsulated with inactivated Group B streptococcus (GBS) antigen by University of Iowa and mucosal and humoral responses evaluated in mice [21]. Routes of administration were oral, vaginal and nasal, along with IM and IP. The objective of the study was

Fig. 3. Comparison of functional bactericidal titers for antigen, Men B in PLG/Men B with and without CpG. CpG, when added, was in soluble, adsorbed, and encapsulated forms. Dose of Men B antigen was 2 μg. The formulations in the figure are PLG/Men B is PLG/Men B without CpG, PLG/Men B + CpG soluble is PLG/Men with soluble CpG, PLG/Men B + PLG/CTAB/ CpG Adsorbed is PLG/Men B with CpG adsorbed on separate particles and PLG/Chitosan–CpG Encap/Men B is PLG/Men B with CpG encapsulated in the same particles. A four fold difference in BCA titers observed between soluble CpG and encapsulated CpG formulations while adsorbed CpG is similar to soluble CpG [10].

222

P. Malyala et al. / Advanced Drug Delivery Reviews 61 (2009) 218–225

Fig. 4. CpG load and encapsulated w/w PLG in CpG-OVA [8], CpG in PLG (low and high Chitosan) [8] and CpG/GBS in PLG [8]. Target load of CpG in PLG/CpG/GBS is less when compared to other studies and indicated in the smaller figure [8].

to evaluate both local mucosal response, at the site of colonization of the antigen, and humoral response to provide transplacental protection for the poorly immunogenic GBS. GBS-CpG encapsulated PLG formulation was prepared by the w/o/w method described above by adding GBS and CpG in the aqueous phase and then adding it to PLG oil phase in the primary emulsion. The primary emulsion was added to PVA aqueous solution prior to the secondary emulsion step. Excess PVA was subsequently washed off. Encapsulated efficiencies of both CpG and GBS were 10% of the loading levels added (Fig. 4). Controls in this study were GBS + CpG and GBS encapsulated in PLG with no CpG. GBS-CpG encapsulated PLG formulation elicited significantly higher response compared to the control groups and documented secretory IgA antibodies in vaginal washes and blood samples. This indicates possible protection against maternal GBS colonization and passive transplacental immunization for the fetus and the neonate [21]. In the fatal neurodegenerative prion diseases, normal alpha helical prion protein converts to abnormal beta sheet rich infectious isoform (PrPSc), which the immune system fails to recognize due to self tolerance [22]. Kaiser-Schulz et al. investigated the co-encapsulation of CpG with the antigen, recombinant tandem prion protein (PrP) in wild type mice to break self tolerance to the protein. Encapsulation of CpG and PrP was achieved by employing solvent evaporation technique. The antigen and adjuvant were emulsified with methylene chloride and the primary emulsion added to an aqueous phase containing 3% polyvinyl alcohol. Enhanced induction of specific T cell and humoral responses were demonstrated with PLG coencapsulating PrP and CpG. Depending on the mode of antigen preparation, not only was there an enhanced immune response, but CpG also influenced the refolding conditions and interaction [22]. Another group that investigated encapsulated CpG was Gomez et al. with Protamine and bee venom allergen phospholipase A2(PLA2) [23]. The antigen, PLA2 was encapsulated in PLG with and with out CpG (PLG-PLA and PLG-PLA-CpG) as well as with CpG and Protamine, a cationic protein (PLG-PLA-CpG-Protamine). Control groups of PLA adsorbed on Alum and soluble CpG added to PLG-PLA were included in the mouse study. Unlike the previous methods of micropartices synthesis, Gomez et al. employed a microextrusion based w/o/w sovent extraction technique. A static multilamination micromixer was used and primary emulsification was achieved by ultrasonication. The suspension was extruded from the micromixer and then added to an

aqueous solution of 0.5% PVA. The solvent evaporated suspension was filtered through a 0.8 um membrane filter and particles collected on the membrane and dried. The D50 for particle size ranged from 4 to 8 m and antigen loading efficiencies were in the range of 22–33%. Integrity of the antigen was intact on synthesis and storage. In vitro release of the antigen was higher and complete for the protamine containing formulation while MP-PLA2 released only 63% of antigen over 7 weeks period. Antibody and T cell responses were strongest with PLG coencapsulating CpG, PLA2, and protamine [23]. Yet another study that evaluated CpG coencapsulated with antigen was from the University of Iowa. Zhang et al. studied the effect of CpG coencapsulated with Ovalbumin(OVA) in PLG microparticles and compared the formulation to CpG–OVA fusion molecules and soluble CpG with OVA [8]. The responses were also compared to Alum formulation. For the encapsulated formulation, loading levels of antigen and CpG were 0.95 and 0.8% respectively w/w PLG with encapsulation efficiencies of about 23% and 34% respectively (Fig. 4). The particles were synthesized by double emulsion solvent evaporation method. Responses, both humoral and T cell, were strongest for CpG coencapsulateed with OVA in microparticles with CpG–OVA fusion molecules also generating a strong response when compared to Alum and soluble CpG formulation groups [8]. Data from different groups with CpG encapsulated formulations indicate enhancement of immunogenicity when compared to CpG added in soluble form which underlines the necessity of formulating the adjuvant to co-deliver the antigen and adjuvant to the targeted site of action. The encapsulated GBS study concluded that microencapsulation of both antigen and CpG was required to elicit an immune response, while the control groups, Soluble GBS & soluble CpG group and GBS encapsulated in PLG group, failed to elicit an immune response [21]. Kaiser-Schulz's group came to a similar conclusion in the Prion study. They explained that PLG with tPrP alone or Soluble antigen and Soluble CpG group failed to induce a detectable response as there was no co-internalization of antigen and CpG into the same APCs [22]. Citing other studies, Omez et al. also concluded that encapsulation of antigen, CpG and protamine ensures synchronized delivery of both into the same APCs and thus a better immune response when compared to antigen encapsulated or antigen encapsulated with protamine. Zhang et al. also stressed the need for colocalization of antigen and adjuvant for uptake by the APCs [8]. Encapsulation helps not only in optimized release of antigen and adjuvant but also improves the safety profile of the adjuvant by mitigating the toxicity associated with activity at non-targeted tissues [22]. 3. CpG in soluble form The most simple and common method of evaluating immunopotentiator effect of an adjuvant is to add the adjuvant in soluble form with the antigen. Many groups have evaluated CpG in soluble form to evaluate the effect of CpG on the antigen's potency and is also often used as a control, when evaluating other adjuvants. Soluble CpG has been studied in combination with other adjuvants, like polyphosphazenes, to increase immune response to antigens in mice [24]. Immunogenicity of two antigens, Hepatitis B virus surface antigen and formalin-inactivated bovine respiratory syncytial virus vaccine, was increased when CpG and phosphazene were added in the

Table 3 Effect of CpG on soluble antigens Formulations

Titer type

Antigen

Antigen with CpG

HepB DNP-BSA

Log10IgG2a titers (24 wks post imm) ng Ag-specific IgA/μg total IgA (8 days post two IN imm.)

2 6–10

3.5–4.0 50–85

A comparison of immunogenicity indicating titers for Antigen with and with out soluble CpG from different studies presented [8].

P. Malyala et al. / Advanced Drug Delivery Reviews 61 (2009) 218–225

soluble form to the soluble antigens (Table 3) [24,25]. Soluble CpG has also been shown to induce antitumor immunity [26]. In the development of a nasal vaccine for human papillomavirus type 16 virus-like particles, CpG and heat labile enterotoxin (HLT) were compared as mucosal adjuvants [27]. HLT proved to be the more potent adjuvant with higher humoral and cellular immune responses to airway immunization(data not shown) [27]. In the studies described throughout the review, where CpG formulations were being investigated for enhancement in immune responses, most studies had a control group of soluble CpG added to the soluble antigen [18,21]. In most of those studies, formulated CpG exhibited higher titers when compared to the soluble CpG + soluble antigen group (Table 3). In another study, CpG was evaluated in soluble form and its adjuvancy compared to formulated monophosphoryl lipid A and QS21 against hepatitis B surface antigen in healthy adult volunteers [28]. The formulated adjuvants, monophosphoryl lipid A and QS 21, performed superior to soluble CpG, which again stresses the need to formulate adjuvants [28]. Table 1, which shows immunogenicity indicating data from this study and from Kazzaz et al. study, discussed in Section 2.1, indicates that encapsulated adjuvants like MPL, RC529, and QS 21, are comparable or superior in immunogenicity to Soluble CpG formulations [13,26]. As can be seen from the studies in this section, results with soluble CpG vary depending on the antigen and route of administration. 4. CpG with other delivery systems Extensive work has been conducted with CpG and delivery systems other than biodegradable PLG microparticles, where CpG has either been formulated with the delivery system or added in soluble form to the antigen formulated to delivery vehicle. Formulation of CpG was done by adsorption, complexation, and encapsulation. Nanoparticlate delivery systems were made from PLG, Chitosan, and gelatin. 4.1. PLG nanoparticles and Liposomes PLG nanoparticles were used to codeliver tetanus toxoid and CpG and dose sparing of CpG oligodeoxynucleotide achieved, minimizing possible side effects of the adjuvant [29]. Cationic PLG nanoparticles and liposomes were used to package CpG in a pig paratyphoid vaccine study with mice [30]. Both the delivery systems exhibited strong immunostimulatory effect when compared to the control in the study. Liposomes were also used as a delivery system to coencapsulate CpG oligodeoxynucleotides with recombinant Leishmania major stress-inducible protein 1 [31]. The coencapsulated formulation enhanced immune response and protection against leishmaniasis in

223

immunized BALB/c mice when compared with both CpG added in soluble form with antigen or antigen without the adjuvant. The results indicated the superiority of CpG ODN in its liposomal form over its soluble form and plays an important role in vaccine development strategies against leishmaniasis [31]. 4.2. Alum Alum is an age old delivery vehicle that is predominantly used in vaccine field. In a CpG bound to Alum study, data indicated that higher antibody response to malarial recombinant Plasmodium falciparum protein AMA1-C1 was achieved when the highest concentration of CpG was bound to Alum [32]. The study indicated physical association of the adjuvant to delivery vehicle is required rather than adding CpG in a free soluble form [32]. In another study, soluble CpG was found to increase the number of mouse responders with recombinant Pseudomonas aeruginosa ExoProtein A conjugates of AMA1 and Pfs25 on Alum [33]. 4.3. Other nanoparticulate delivery systems Other nanoparticulate delivery systems evaluated were chitosan nanoparticles, cationic gelatin nanoparticles and protamine nanoparticles. Chitosan nanoparticles were used as a delivery system to encapsulate pig IL-6 gene and CpG motifs for enhancement of immunity in mice [34]. Delivery by cationic gelatin nanoparticles strongly increased the immunostimulatory effects of CpG oligonucleotides [35]. Immunostimulatory properties of CpG-oligonucleotides were enhanced when encapsulated in protamine nanoparticles [36]. To ensure that protamine nanoparticles do not have an immunostimulatory effect, control ODN was encapsulated and this group had no immunostimulatory properties, thus supporting the use of particulate delivery systems like biodegradable protamine nanoparticles for the development of CpG-ODN-based therapeutics [36]. In a mouse study with hepatitis B surface antigen, a weak antigen, researchers found soluble CpG gave better results than adsorbed CpG to alginate coated nanoparticles [37]. They hypothesized that the free amount of the CpG ODN, released from alginate coated nanoparticles, was not high enough to stimulate the immune system to respond to the HBV antigen. The researchers discussed that the results were still not fully conclusive with regard to the advantages of the association of the CpG ODN to nanoparticles [37]. Once inside the target tissue, the ideal delivery system should release the CpG ODN, where most cell types have the capacity to take up CpG ODN via endocytosis or the CpG ODN internalization in cells would be facilitated by the use of CpG associated to suitable nanoparticles [37]. This does bring attention to the fact that adjuvant evaluation needs to be done in soluble and formulated forms, to best judge its adjuvancy.

Fig. 5. Evaluation of delivery vehicles with soluble CpG [38].

224

P. Malyala et al. / Advanced Drug Delivery Reviews 61 (2009) 218–225

4.4. MF 59 MF 59 is an oil in water nano emulsion and licensed to be used in more than 20 countries [38]. It is currently used in Fluad, which is a Novartis adjuvanted influenza vaccine. Addition of MF59 has improved cross reactivity against heterovariant strains and also increased immunogenicity in the geriatric population prone to influenza infections [38]. Wack et al. evaluated CpG in soluble form with influenza proteins as the antigen and PLG, MF59, Alum, and Calcium Phosphate as the delivery vehicles [38]. MF 59 ranked as the best delivery vehicle against the three trivalent flu strains without an adjuvant followed by PLG and Alum. A CpG effect was observed with all the four delivery vehicles evaluated as can be seen from Fig. 5 which compares the ratio of functional HI titers of formulations with and without CpG. Although PLG with CpG did produce higher functional titers in mice when compared to PLG/Flu alone, MF59 with CpG yielded highest functional titers across all three influenza strains when compared to the other delivery vehicles (Fig. 5) [38]. 5. Future directions The necessity of adding an adjuvant to modern vaccines is proved with data generated from research conducted with vaccine adjuvants. CpG appears to be a strong adjuvant in the field of vaccines due to its ability to generate strong humoral and cytokine responses [5–7]. Results from the studies investigating CpG indicate that immunogenicity of the adjuvant is greatly affected depending on the mode of delivery of the immunopotentiator. Unformulated CpG is a strong adjuvant and has often been used as a gold standard for evaluation of adjuvants. While soluble CpG showed an adjuvant effect when added to soluble antigen, studies which compared this group to soluble CpG added to antigen associated with a delivery vehicle proved that the latter group was more immunogenic [5–7]. Studies with CpG associated with a delivery vehicle were generally more immunogenic compared to CpG added in soluble form to an antigen or antigen associated with a delivery vehicle [5–7]. Encapsulated adjuvants like MPL, RC529, and QS 21, were comparable or superior in immunogenicity to soluble CpG formulations [13,26]. These encapsulated non CpG adjuvants also elicited stronger immune responses when compared to their soluble counterparts. As discussed in Section 2.3, which describes the strong immune responses of CpG encapsulated in PLG microparticles, several groups underlined the need to employ a delivery vehicle to ensure co-delivery of antigen and adjuvant to the same APCs based on the data their studies generated [13,26]. Thus, the first test of adjuvancy could be adding CpG in soluble form but formulating CpG could further increase its immunogenic effect and increase the number of respondents [33]. Adsorbed and encapsulated formulations have shown significant enhancement in immunogenicity when compared to the soluble counterpart, with encapsulated formulations exhibiting higher potency [33]. Exceptions to this general trend were generated in studies where unformulated CpG, added in soluble form, was similar or more potent when compared to formulated CpG [10,37]. This could be due to a strong association of CpG to the delivery vehicle that would prevent CpG from being taken up by the macrophages and dendritic cells and elicit its effect [37]. Targeted delivery of immunopotentiator to the site of action is an important factor for its effect. Its extent and duration are dependent on the optimized release of immunopotentiator. Soluble adjuvants, unassociated with the antigen or the delivery system, are characteristically delivered locally and distributed to distant tissues that might cause toxicity to non targeted cells. Formulating an immunopotentiator would overcome these deficiencies and protect degradation, target it to macrophages and dendritic cells, and enable co-delivery of

both antigen and immunopotentiator to the same cell type [33]. In addition to increasing potency, formulating CpG would help in dose sparing of antigen or the adjuvant and also decrease the toxicity of the adjuvant due to the low dose [37]. A great amount of work has been conducted with CpG and a wide breadth of antigens. However, most studies were in smaller animal models and more data needs to be generated in larger models. As stated in Section 1.1, expression of TLR 9 differs based on animal species. For instance, in humans, B cells and plasmacytoid dendritic cells are the only immune cells that are known to express TLR9 and be activated by CpG, whereas in mice, TLR9 is broadly expressed on both the major subtypes, plasmacytoid and myeloid dendritic cells, as well as in B cells, macrophages and monocytes [9]. Hence, one should be especially cautious when extrapolating mouse data to humans, during evaluation of CpG as an adjuvant [9]. CpG is currently being evaluated in the clinic with various antigens like Hep B viral antigen, influenza flu, and montanide, which is a cancer vaccine. It is being evaluated in unformulated form and also with Alum, and MF 59 delivery vehicles. However, there is no clinical evaluation with CpG formulated in PLG microparticles. To conclude, data obtained thus far stresses the need for evaluating CpG or any other adjuvant in both soluble and formulated forms with antigen and delivery system. Also, CpG needs to overcome major safety hurdles, as all new adjuvants will. References [1] A. Heit, D.H. Busch, H. Wagner, F. Schmitz, Vaccine protocols for enhanced immunogenicity of exogenous antigens, Int. J. Med. Microbiol. 298 (1–2) (2008) 27–32. [2] R. Schirmbeck, W. Bohm, K. Ando, F.V. Chisari, J. Reimann, Nucleic-acid vaccination primes hepatitis-b virus surface antigen-specific cytotoxic t-lymphocytes in nonresponder mice, J. Virol. 69 (10) (1995) 5929–5934. [3] J.M. Brewer, L. Tetley, J. Richmond, F.Y. Liew, J. Alexander, Lipid vesicle size determines the Th1 or Th2 response to entrapped antigen, J. Immunol.161 (8) (1998) 4000–4007. [4] S. Akira, K. Takeda, Toll-like receptor signalling, Nat. Rev. Immunol. 4 (7) (2004) 499–511. [5] R. Medzhitov, C.A. Janeway Jr., An ancient system of host defense, Curr. Opin. Immunol. 10 (1) (1998) 12–15. [6] A. Bendelac, R. Medzhitov, Adjuvants of immunity: harnessing innate immunity to promote adaptive immunity, J. Exp. Med. 195 (5) (2002) F19–23. [7] R. Medzhitov, Toll-like receptors and innate immunity, Nat. Rev. Immunol. 1 (2) (2001) 135–145. [8] X.D. Zhang, C.E. Weiner, G.J. Salem, AK., A comparative study of the antigen-specific immune response induced by co-delivery of CpG-PDN and antigen using fusion molecules or biodegradable microparticles, J. Pharm. Sci. 96 (12) (2007) 3283–3292. [9] A.E. Chutitorn Ketloya, Utaiwan Srichairatanakulb, Amporn Limsalakpetchb, Kosol Yongvanitchitb, Sathit Pichyangkulb and Kiat Ruxrungthama, Expression and function of Toll-like receptors on dendritic cells and other antigen presenting cells from non-human primates Vet, Immunol. Immunopathol. 125 (1–2) (2008) 18–30. [10] P Malyala, J. Chesko, M. Ugozzoli, M. Vajdy, F Zhou, A. Goodsell, T. O'Hagan D, M. Singh, The potency of the adjuvant, CpG oligos, is enhanced by encapsulation in PLG microparticles. J. Pharm. Sci. 97 (3) (2007) 1155–1164. [11] D.T. O'Hagan, M. Singh, J.B. Ulmer, Microparticle-based technologies for vaccines, Methods 40 (1) (2006) 10–19. [12] M. Singh, J. Kazzaz, M. Ugozzoli, P. Malyala, J. Chesko, D.T. O'Hagan, Polylactide-coglycolide microparticles with surface adsorbed antigens as vaccine delivery systems, Curr. Drug. Deliv. 3 (1) (2006) 115–120. [13] M. Singh, J. Kazzaz, M. Ugozzoli, J. Chesko, D.T. O'Hagan, Charged polylactide coglycolide microparticles as antigen delivery systems, Expert Opin. Biol. Ther. 4 (4) (2004) 483–491. [14] J. Kazzaz, M. Singh, M. Ugozzoli, J. Chesko, E. Soenawan, T. O'Hagan D, Encapsulation of the immune potentiators MPL and RC529 in PLG microparticles enhances their potency, J. Control. Release 110 (3) (2006) 566–573. [15] S. Somroop, P. Tongtawe, U. Chaisri, P. Tapchaisri, M. Chongsa-nguan, P. Srimanote, W. Chaicumpa, Traffic of antibody-secreting cells after immunization with a liposomeassociated, CpG-ODN-adjuvanted oral cholera vaccine, Asian Pac. J. Allergy Immunol. 24 (4) (2006) 229–238. [16] T. Naito, Y. Kaneko, D. Kozbor, Oral vaccination with modified vaccinia virus Ankara attached covalently to TMPEG-modified cationic liposomes overcomes pre-existing poxvirus immunity from recombinant vaccinia immunization, J. Gen. Virol. 88 (Pt 1) (2007) 61–70. [17] M. Tafaghodi, M.R. Jaafari, S.A. Sajadi Tabassi, Nasal immunization studies using liposomes loaded with tetanus toxoid and CpG-ODN, Eur. J. Pharm. Biopharm. 64 (2) (2006) 138–145. [18] R. Gill, P.C. Montgomery, Curr. Eye Res. 24 (3) (2002) 228–233. [19] M. Singh, G. Ott, J. Kazzaz, M. Ugozzoli, M. Briones, J. Donnelly, D.T. O'Hagan, Cationic microparticles are an effective delivery system for immune stimulatory CpG DNA, Pharm. Res. 18 (10) (2001) 1476–1479.

P. Malyala et al. / Advanced Drug Delivery Reviews 61 (2009) 218–225 [20] H. Xie, I. Gursel, B.E. Ivins, M. Singh, D.T. O'Hagan, J.B. Ulmer, D.M. Klinman, P. Baldrick, D. Richardson, A.W. Wheeler, H. Aggerbeck, I. Heron, K. Miyamura, S. Nishio, A. Ito, R. Murata, R. Kono, CpG oligodeoxynucleotides adsorbed onto polylactide-co-glycolide microparticles improve the immunogenicity and protective activity of the licensed anthrax vaccine, Infect. Immun. 73 (2) (2005) 828–833. [21] S.K. Hunter, M.E. Andracki, A.M. Krieg, Biodegradable microspheres containing group B Streptococcus vaccine: immune response in mice, Am. J. Obstet. Gynecol. 185 (5) (2001) 1174–1179. [22] G.H. Kaiser-Schultz, A. Heit, L. Quintanilla-Martinez, F. Hammerschmnidt, S. Hess, L. Jennen, H. Rezaei, H. Wagner, H.M. Schatzl, Polylactide-coglycolide microspheres co-encapsulating recombinant tandem prion protein with CpG-oligonucleotide break self tolerance to prion protein in wild type mice and induce CD4 and CD8 T cell responses, J. Immunol. 179 (5) (2007) 2297–2807. [23] J.M. Martinez Gomez, S. Fischer, N. Csaba, T.M. Kundig, H.P. Merkle, B. Gander, P. Johansen, A protective allergy vaccine based on CpG and Protamine containing PLGA microparticles, Pharm. Res. 24 (10) (2007) 1927–1935. [24] B.P. Mutwiri G, H. Soita, L.A. Babiuk, Co-administration of polyphosphazenes with CpG oligodeoxynucleotides strongly enhances immune responses in mice immunized with Hepatitis B virus surface antigen, Vaccine 26 (22) (2008) 2680–2688. [25] O.M. Mapletoft JW, Kovacs-Nolan J, Latimer L, Mutwiri G, Babiuk LA, van Drunen Littel-van den Hurk S, Intranasal immunization of mice with a formalin-inactivated bovine respiratory syncytial virus vaccine co-formulated with CpG oligodeoxynucleotides and polyphosphazenes results in enhanced protection. J Gen Virol. 2008 Jan;89(Pt 1):250–60 89(1) (2008) 250–260. [26] K.C. Woo SJ, M.Y. Park, H.S. Kim, H.J. Sohn, J.S. Park, H.J. Kim, S.T. Oh, T.G. Kim, Coadministration of carcinoembryonic antigen and HIV TAT fusion protein with CpGoligodeoxynucleotide induces potent antitumor immunity, Cancer Sci. 99 (5) (2008) 1034–1039. [27] Z.R. Revaz V, C. Moser, J.T. Schiller, F. Ponci, M. Bobst, D. Nardelli-Haefliger, Humoral and cellular immune responses to airway immunization of mice with human papillomavirus type 16 virus-like particles and mucosal adjuvants, Antivir. Res. 76 (1) (2007) 75–85. [28] H.Y. Vandepapelière P, P. Moris, M. Van Mechelen, M. Janssens, M. Koutsoukos, P. Van Belle, F. Clement, E. Hanon, M. Wettendorff, N. Garçon, G. Leroux-Roels, Vaccine adjuvant systems containing monophosphoryl lipid A and QS21 induce strong and persistent humoral and T cell responses against hepatitis B surface antigen in healthy adult volunteers, Vaccine 26 (10) (2008) 1375–1386.

225

[29] M. Diwan, P. Elamanchili, M. Cao, J. Samuel, Dose sparing of CpG oligodeoxynucleotide vaccine adjuvants by nanoparticle delivery, Curr. Drug Deliv. 1 (2004) 405–412. [30] S.L. Wu M, S. Liu, J. Li, K. Wu, L. Wang, Y. Shen, K. Liu, Y. Zheng, X. Zhang, R. Gao, The effect of entrapment of CpG sequence with cationic PLG nanoparticles on the immune responses of mice to pig paratyphoid vaccine, J. Biomed. Eng. 22 (5) (2005) 975–979. [31] J.M. Badiee A, A. Samiei, D. Soroush, A. Khamesipour, Coencapsulation of CpG oligodeoxynucleotides with recombinant Leishmania major stress-inducible protein 1 in liposome enhances immune response and protection against leishmaniasis in immunized BALB/c mice, Clin. Vaccine Immunol. 15 (4) (2008) 668–674. [32] A.J. Mullen GE, G. Dobrescu, K. Rausch, L. Lambert, C.A. Long, A.P. Miles, A. Saul, Enhanced antibody production in mice to the malaria antigen AMA1 by CPG 7909 requires physical association of CpG and antigen, Vaccine 25 (29) (2007) 5343–5347. [33] R.K. Qian F, O. Muratova, H. Zhou, G. Song, A. Diouf, L. Lambert, D.L. Narum, Y. Wu, A. Saul, L.H. Miller, C.A. Long, G.E. Mullen, Addition of CpG ODN to recombinant Pseudomonas aeruginosa ExoProtein A conjugates of AMA1 and Pfs25 greatly increases the number of responders, Vaccine 26 (20) (2008) 2521–2527. [34] Z.H. Zhao ZZ, Q. Chen, D. Su, Z. Xie, Y.Y. Wang, Y. Yang, Z.Z. Wang, J.L. Li, K.Y. Wu, H. N. Wang, M.J. Meng, R. Gao, Promotion of immunity of mice to Pasteurella multocida and hog cholera vaccine by pig interleukin-6 gene and CpG motifs, Comp. Immunol. Microbiol. Infect. Dis. (2008) [Feb 13] [ahead of print]. [35] B.C. Zwiorek K, J. Battiany, G. Winter, S. Endres, G. Hartmann, C. Coester, Delivery by cationic gelatin nanoparticles strongly increases the immunostimulatory effects of CpG oligonucleotides, Pharm. Res. 25 (3) (2008) 551–562. [36] L.D. Kerkmann M, J. Weyermann, A. Marschner, H. Poeck, M. Wagner, J. Battiany, A. Zimmer, S. Endres, G. Hartmann, Immunostimulatory properties of CpGoligonucleotides are enhanced by the use of protamine nanoparticles, Oligonucleotides 16 (4) (2006) 313–322. [37] C.-d.-S.A. Borges O, J. Tavares, N. Santarém, A. de Sousa, G. Borchard, H.E. Junginger, Immune response by nasal delivery of hepatitis B surface antigen and codelivery of a CpG ODN in alginate coated chitosan nanoparticles, Eur. J. Pharm. Biopharm. 69 (2) (2008) 405–416. [38] A. Wack, B.C. Baudner, A.K. Hilbert, I. Manini, S. Nuti, S. Tavarini, H. Scheffczik, M. Ugozzoli, M. Singh, J. Kazzaz, E. Montomoli, G. Del Giudice, R. Rappuoli, D.T. O'Hagan, Combination adjuvants for the induction of potent, long-lasting antibody and T-cell responses to influenza vaccine in mice, Vaccine 26 (4) (2008) 552–561.