Enhancing immunogenicity of antigens through sustained intradermal delivery using chitosan microneedles with a patch-dissolvable design

Accepted Manuscript Enhancing Immunogenicity of Antigens through Sustained Intradermal Delivery using Chitosan Microneedles with a Patch-Dissolvable Design Mei-Chin Chen, Kuan-Ying Lai, Ming-Hung Ling, Chun-Wei Lin PII: DOI: Reference:

S1742-7061(17)30678-5 https://doi.org/10.1016/j.actbio.2017.11.004 ACTBIO 5157

To appear in:

Acta Biomaterialia

Received Date: Revised Date: Accepted Date:

21 August 2017 1 November 2017 2 November 2017

Please cite this article as: Chen, M-C., Lai, K-Y., Ling, M-H., Lin, C-W., Enhancing Immunogenicity of Antigens through Sustained Intradermal Delivery using Chitosan Microneedles with a Patch-Dissolvable Design, Acta Biomaterialia (2017), doi: https://doi.org/10.1016/j.actbio.2017.11.004

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Enhancing Immunogenicity of Antigens through Sustained Intradermal Delivery using Chitosan Microneedles with a Patch-Dissolvable Design

Mei-Chin Chen,* Kuan-Ying Lai, Ming-Hung Ling, and Chun-Wei Lin

Department of Chemical Engineering, National Cheng Kung University, Tainan, Taiwan (ROC)

*Correspondence to: Mei-Chin Chen, PhD Associate Professor Department of Chemical Engineering National Cheng Kung University Tainan, Taiwan 70101 Tel: +886-6-275-7575 # 62696 Fax: +886-6-234-4496 E-mail: [email protected]

1

Abstract Reducing the dosage required for vaccination is highly desirable, particularly in cases of epidemic emergencies. This study evaluated the potential of a chitosan microneedle (MN) system with a patch-dissolvable design for low-dose immunization. This system comprises antigen-loaded chitosan MNs and a hydrophilic polyvinyl alcohol/polyvinyl pyrrolidone supporting array patch, which provides extra strength to achieve complete MN insertion and then quickly dissolves in the skin to reduce patch-induced skin irritation. After insertion, MNs could be directly implanted in the dermal layer as an intradermal (ID) depot to allow a sustained release of the model antigen ovalbumin (OVA) for up to 28 days. We found that rats immunized with MNs containing low-dose OVA (approximately 200 µg) had persistently high antibody levels for 18 weeks, which were significantly higher than those observed after an intramuscular injection of full-dose OVA (approximately 500 µg), demonstrating at least 2.5-fold dose sparing. Moreover, OVA-encapsulated chitosan MNs had superior immunogenicity to OVA plus chitosan solution, indicating that MN-based delivery and prolonged skin exposure can further enhance chitosan’s adjuvanticity. Therefore, this patch-dissolvable MN system offers a needle-free, accurate, and reliable ID delivery of antigens and has potential as a sustained ID delivery device to improve vaccine efficacy and facilitate dose sparing with existing vaccines.

Keywords: adjuvanticity; chitosan; degradability; immunostimulation; sustained delivery; vaccination

2

1. Introduction Vaccines contribute markedly to protecting human health by releasing a small amount of pathogens to induce immune responses. However, some vaccines, such as DNA, toxoids, and protein vaccines, are not highly immunogenic and must be administered in large amounts, multiple doses, or combination with adjuvants to achieve immunization [1,2]. A high vaccine dose not only stimulates the immune system to produce more antibodies but also leads to an increase in side effects. For example, a high-dose flu vaccine, produced for elderly patients, can cause short-term aches and flu-like symptoms. Many traditional vaccines must be administered in several divided doses, which results in a costly and inconvenient regimen. Moreover, vaccination becomes ineffective if a patient misses one or more injections. A vaccine adjuvant is an agent that can enhance antigen-specific immune responses and direct the type of immune response induced. Formulating vaccines with appropriate adjuvants may reduce the required antigen dose and vaccination number, increase vaccine efficacy in infants and elderly patients, provide broad protection, and induce more rapid and long-lasting immune responses. A more comprehensive understanding of the mechanisms of adjuvants is important to developing next-generation vaccines. Chitosan, a natural polysaccharide produced by chitin deacetylation, has been extensively used as a main material for tissue engineering and drug delivery because of its biocompatibility, biodegradability, and ease of production. Recently, many studies have reported that chitosan is an effective adjuvant for injectable and mucosal vaccines [3−6]. Its possible mechanisms of action include protecting antigens from degradation, forming a depot at the inoculation site, facilitating antigen uptake and presentation, and activating innate and adaptive immune responses [4,7,8]. A recent study further demonstrated that chitosan can drive dendritic cell maturation by inducing type I interferons and can promote cellular immunity by engaging the DNA sensor cyclic GMP–AMP synthase–STING pathway [4]. Microneedles (MNs) can deliver vaccines directly into the skin that contains many 3

antigen-presenting cells, and they thus have the potential for engendering higher immunogenicity than do intramuscular (IM) injections [9,10]. We previously developed chitosan MNs as a depot that enables targeted delivery of antigens to the skin for 2 weeks [11]. Compared with IM vaccination, intradermal (ID) vaccination by using the chitosan MNs resulted in significantly stronger immune responses, which lasted for 6 weeks. However, it is unknown whether prolonged exposure of the skin to chitosan further enhances chitosan adjuvanticity and whether chitosan MNs can yield a marked dose-sparing effect, which is beneficial in reducing vaccine costs and increasing vaccine supply. In the present study, we developed novel chitosan MNs with a patch-dissolvable design and investigated their potential as an implantable vaccine carrier to reduce the antigen dose required for vaccination; moreover, we elucidated the possible roles of these MNs in improving the immunogenicity of an antigen. The MN patch comprises antigen-loaded chitosan MNs and a dissolvable polyvinyl alcohol/polyvinyl pyrrolidone (PVA/PVP) supporting array patch (Fig. 1). This supporting array can not only offer strength to facilitate the complete insertion of the MNs into the skin, but also dissolve in the skin’s interstitial fluid spontaneously within a few minutes. After dissolution, the chitosan MNs can be retained at the injection site for sustained antigen release and immune activation. Because of the patch-dissolvable design of the present MN system, the patch need not be removed from the skin surface after MN insertion, thus further enhancing user convenience, compared with our previous MN system [11]. In this study, the composition of the chitosan MNs was adjusted to develop a more long-lasting antigen depot. We assume that retaining chitosan and antigens in the skin for longer periods may permit the prolonged stimulation of immune cells to generate more robust and persistent immune responses. The use of low-dose vaccines is cost-effective and important for the safety of vaccine recipients because it can alleviate concerns associated with some adverse events caused by immunization. To explore the dose-sparing potential of the chitosan MNs, we compared the immune responses elicited by MNs containing low-dose or full-dose ovalbumin (OVA) with those elicited by the standard IM vaccination of full-dose

4

OVA with or without chitosan solution. Moreover, the in vivo skin insertion and embedding potentials of the MNs, skin response after MN insertion, and MN degradability and antigen release were investigated.

2. Materials and Methods 2.1. Fabrication of OVA-Loaded Chitosan–PVA/PVP MN Patches 2.1.1. PDMS Molds for MNs, Supporting Arrays, and Pressing Tool Fig. S1 shows the photographs of the master structures of MNs (a and a1), supporting arrays, and pressing tools (b and b1). The insets in Fig. S1 show the detailed specifications of each metallic structure, consisting of 81 (9 × 9) microstructures. Polydimethylsiloxane (PDMS) MN molds were fabricated as exact inverse replicates of these metallic structures, according to a published method [11]. The prepared PDMS molds were subsequently used to prepare

the

chitosan

MNs,

PVA/PVP

supporting

array

patch,

and

poly(L-lactide-co-D,L-lactide) (PLA) pressing tool. 2.1.2. Preparation of OVA-Containing Chitosan Gel A chitosan gel (9 wt%) was prepared according to a literature procedure [11]. To maintain antigen activity during encapsulation, antigens (OVA or Alexa Fluor® 594–OVA conjugate; Alexa 594–OVA) were first dissolved in a trehalose aqueous solution, and the mixture was subsequently blended with the chitosan gel. The mixture was evaporated at 37 °C to form a viscous gel containing 0.07 wt% OVA, 1.3 wt% trehalose, and 13 wt% chitosan for casting. To visualize the MNs within the skin, chitosan was fluorescently labeled with fluorescein 5(6)-isothiocyanate (FITC–chitosan), according to a previously reported procedure [12], and then prepared as mentioned in the preceding descriptions to form a viscous gel for the casting process. 2.1.3. Preparation of the PVA/PVP Supporting Array Patch and PLA Pressing Tool To prepare the dissolvable supporting array, a 50 wt% PVA/PVP aqueous solution (0.5 g; 5

PVA:PVP weight ratio = 1:1) was placed on the PDMS mold and centrifuged in a swinging bucket rotor at 5100 rpm and 30 °C for 30 min. The filled mold was air dried at room temperature (RT) for 1 day and then placed in an oven at 37 °C for 1 day. A PLA pressing tool was fabricated by casting molten PLA pellets onto the PDMS mold in vacuum at −70 cm Hg for 4 h at 190 °C. The samples were then placed in a freezer at 4 °C for 10 min before being released from the mold. 2.1.4. Integration of Chitosan MNs and PVA/PVP Supporting Arrays To fabricate chitosan MNs, 180 mg of chitosan gel containing OVA and trehalose was placed on the PDMS mold and centrifuged in a swinging bucket rotor at 5100 rpm and 30 °C for 2 h (Fig. 2). To force the chitosan gel further into the mold cavities, approximately 0.1 g of the 50 wt% PVA/PVP solution was added on the mold surface and then air dried at RT for 1 day. In addition, to remove the excess polymer, the patch formed on the mold surface was cut with a razor blade. The fabricated PLA pressing tool was used to sufficiently compress the chitosan MNs into the mold cavities. To integrate the supporting array and MNs, the top of the PVA/PVP supporting array patch was slightly moistened with the 50 wt% PVA/PVP solution and subsequently aligned and inserted into the filled MN mold under a stereomicroscope (Fig. 2). This assembly was placed in an oven at 37 °C for 20 min to form the chitosan–PVA/PVP MN patch. 2.1.5. Measurement of OVA Loading Amount in MNs To determine the antigen content in MNs, the MNs were dissolved in deionized water with a mixing speed of 100 rpm at 4 °C for 5 days until complete dissolution. The amount of OVA extracted from the MNs was quantified from the supernatant using a BCA Protein Assay Reagent Kit (Pierce Chemicals, Rockford, IL). 2.2. Skin Insertion Test All animal procedures were approved by the Institutional Animal Care and Use Committee of National Cheng Kung University (NCKU), and the experiments were conducted in accordance with the guidelines of the Laboratory Animal Center of NCKU. To

6

evaluate the skin insertion and embedding potentials, the prepared MNs were applied to a porcine cadaver or rat dorsal skin by using a homemade applicator with an application force of approximately 10 N per patch for 5 min until the supporting array dissolved in the skin (Fig. S2). Before insertion, male Sprague–Dawley rats (age: 4–5 weeks; weight: approximately 150 g) were anesthetized, and their back hair was shaved. The MN insertion site was stained with a blue tissue-marking dye for 1 min to identify the stratum corneum perforation sites. The skin was then excised for histological examination. 2.3. Transepidermal Water Loss Measurement To evaluate skin resealing after MN treatment, transepidermal water loss (TEWL) was determined using a Delfin VapoMeter (Delfin Technologies Ltd., Kuopio, Finland). TEWL readings were continuously monitored until they recovered to the baseline values, thus indicating the recovery of skin barrier function. Control rats received hair trimming only. 2.4. Skin Irritation Assessment To assess the skin irritation potential of the MN treatment, erythema was assessed through a spectrophotometer (ColorLite sph900, ColorLite GmbH, Germany) [13,14]. Measurements were made by placing the apparatus probe gently on the skin surface to record the color reflectance. The colorimeter was calibrated against a color standard before the measurement, according to the manufacturer’s instructions. Readings were recorded in triplicate at every site, and the mean redness value (a*) of the skin was calculated. The change in erythema was reported as a change from the baseline value, ∆ a*, calculated as ∆ a* = a*t (at time t after starting the study) − a*0 (at time 0, prior to treatment; PT). The treated areas were also visually examined for skin damage. 2.5. In Vivo MN Degradation and OVA Retention To assess the degradability of the chitosan MNs and retention of the antigens, Alexa 594–OVA-loaded FITC–chitosan MNs were implanted into

the dorsal skin of

Sprague–Dawley rats. The rats were killed at different time points, and the MN insertion sites were excised for histological section and confocal microscopy analyses. The specimens

7

were visualized using a confocal laser scanning microscopy (CLSM) instrument operated at excitation wavelengths of 488 nm for FITC–chitosan and 559 nm for Alexa 594–OVA. Two-dimensional (2D) x−z image was constructed from confocal Z-stack images. 2.6. In Vivo Fluorescence Imaging To visualize OVA retention at the insertion sites, the anesthetized rats were imaged at the indicated time points after the insertion of the Alexa 594–OVA-loaded MNs by using an in vivo imaging system (IVIS) equipped with Alexa Fluor® 594 filters (excitation filter = 580−620 nm; emission filter = 610−630 nm). This system was used to analyze collected fluorescence data, which are expressed as photon flux (photons/s/cm2/steradian) [15]. Moreover, the fluorescence intensity measured in the selected region at each time point was normalized to the initial fluorescence value obtained at 2 h after MN insertion (i.e., day 0) [16]. 2.7. Immunization in Sprague–Dawley Rats Sprague–Dawley rats were randomized into five groups: (1) the IM saline group, in which the rats were intramuscularly injected with 200 µL of phosphate-buffered saline (PBS); (2) the IM full-dose OVA group, in which the rats were intramuscularly injected with a single full dose (500 µg) of OVA in PBS (200 µL); (3) the IM full-dose OVA + CS group, in which the rats were intramuscularly injected with PBS (200 µL) plus a full dose (500 µg) of OVA and same amounts of chitosan (1 mg) and trehalose (0.1 mg) as in group (4); and (4) the MN full-dose and (5) low-dose groups, in which the rats received single full-dose (500 µg OVA/patch) and low-dose (200 µg OVA/patch) MN vaccination through MN insertion into the dorsal skin, respectively. Blood was collected from the jugular vein at specific time points and allowed to clot at 4 °C. Serum was obtained after centrifugation at 3000 rpm for 5 min at 4 °C. Aliquots of sera were stored at −20 °C until analysis. 2.8. Enzyme-Linked Immunosorbent Assay for Specific Total Immunoglobulin G Levels OVA-specific IgG levels in rat serum were quantified using the enzyme-linked immunosorbent assay, according to a previously published procedure [11]. Briefly, 96-well

8

plates were coated with OVA (200 µg/well) in a 0.05 M carbonate buffer (pH 9.5) at RT for 1 day and then blocked with 1% (w/v) of bovine serum albumin (BSA) diluted in PBS at RT for 2 h. The samples were diluted to 1:10 with PBS containing 0.66% BSA and incubated for 1 day at 4 °C. After being washed with 0.05% Tween-20 in PBS for three times, the plates were incubated with horseradish peroxidase-conjugated donkey anti-rat immunoglobulin G (IgG; 1:10,000) antibodies at RT for 2 h. Subsequently, the plates were rewashed, and enzyme activity was detected with 3,3′-5,5′-tetramethylbenzidine solution (100 µL/well). The color was allowed to develop until the optical density (OD) at 450 nm of the highest absorbing well reached approximately 1.8 [17]. The reaction was stopped by adding 0.5 M sulfuric acid (50 µL/well). 2.9. Statistical Analysis The differences between two groups were analyzed using the one-tailed Student’s t test by using SPSS (Chicago, Ill, USA). Data are presented as mean ± standard deviation (SD). A p value of < 0.05 was considered to indicate a statistically significant difference.

3. Results 3.1. Characterization of Chitosan MNs with a Patch-Dissolvable Design An implantable chitosan MN system with a patch-dissolvable design was developed by separately fabricating FITC-chitosan MNs and a PVA/PVP supporting array patch (Figs. 3a and 3a1) and then integrating them into one device (Figs. 3b and 3b1). The prepared MN patch was composed of 81 (9 × 9) microstructures with a 1000-µm center-to-center spacing between adjacent needles. The base width and height of the MNs and supporting arrays were 300 and 600 µm, respectively. The detailed dimensions of each structure are shown in the insets of Fig. 3. 3.2. Skin Insertion and Implantation Potential Previously, the drug delivery efficiency of polymeric MNs was often limited because of the incomplete insertion of MNs into the skin [18,19]. Owing to the skin’s inherent elasticity,

9

MN puncture causes notable skin deformation, resulting in partial MN insertion [20]. The height of the supporting array developed in the current study can offer an additional length for MNs to overcome skin indentation during MN applications, thus engendering complete MN insertion and efficient drug delivery. After applying the MN to porcine cadaver skin for 5 min, we observed that the PVA/PVP supporting array was completely dissolved in the skin (Fig. 4a). The treated skin was subsequently stained with a blue tissue-marking dye to determine the insertion ratio. As shown in Fig. 4b, a complete array of blue spots (9 × 9) was visible on the skin surface, demonstrating that all chitosan MNs were successfully inserted into the skin. Figs. 4c and 4d show histological sections of the rat skin after insertion of the FITC−chitosan MNs. We observed that the FITC–chitosan MNs (green, Fig. 4d) were completely inserted and then embedded in the rat skin at a depth of approximately 850 µm. This implantation depth indicates that the developed MNs were inserted through the epidermal layer and remained in the dermal layer. According to the H&E-stained skin section (Fig. S3), the epidermis and dermis thicknesses of the 5-week-old rats were 60 ± 7 and 975 ± 93 µm (n = 5), respectively. 3.3. Recovery of Skin Barrier Function TEWL is a noninvasive in vivo measure of water loss from the stratum corneum [21−23]. The stratum corneum corresponds to the outermost layer of the skin and is primarily responsible for skin barrier function. MN insertion significantly increased the TEWL value (p < 0.05, n = 5), demonstrating that the MNs disrupted the integrity of the stratum corneum barrier (Fig. 5). This value gradually returned to the baseline value within 6 h, indicating that microchannels or micropores created by the MNs were rapidly resealed and MN-induced skin damage was minor and recoverable. 3.4. Skin Irritation Skin redness or mild erythema is common in patients who have received MN treatment [13,14,24,25]. To assess MN insertion-induced skin irritation, erythema was quantified by skin color reflectance measurements by using a spectrophotometer [26]. MN insertion 10

significantly increased the erythema index (∆ a*, p < 0.05, n = 5; Fig. 6a) and resulted in visible skin redness (Fig. 6b) at 15 min. Such erythema was determined to last for a few hours, with the skin returning to its normal appearance after 1–3 days (Fig. 6). The presence of micropores on the skin surface provides preliminary evidence of skin resealing. We observed that the pore size became smaller at 15 min, and we found no micropores at 6 h after MN treatment (Fig. 6b), which corresponded to the TEWL results (Fig. 5). These results indicate that the proposed MNs induced mild skin irritation or damage, which dissipated within 24 h. 3.5. In Vivo MN Degradation and OVA Retention in Rat Skin To visualize the MN degradation and OVA release from the MNs, the Alexa 594–OVA-loaded FITC–chitosan MNs were applied to the rat skin, and the insertion sites were subjected to CLSM analysis and histological examination. Fig. 7 presents the confocal micrographs of the skin at varying depths and the corresponding 2D x−z images obtained on day 1. As shown in these 2D x−z images, the implanted MNs (green) were located at depths between 200 and 800 µm, which are equivalent to the dermal region. The red fluorescence of the Alexa 594–OVA surrounded the green one of the FITC–chitosan MNs (merged image), demonstrating that the encapsulated antigen was released from the implanted MNs to the skin. Fig. 8 shows the histological sections of the rat skin after MN insertion for 7, 14, 21, and 28 days. The implanted MNs (green) gradually became smaller, demonstrating that the chitosan MNs can be degraded within the skin. Even on day 28, the red fluorescence of the Alexa–OVA was observed in the small MN fragments, indicating that chitosan MNs can function as a depot of antigens at the insertion sites for 4 weeks. To understand the in vivo release behavior of OVA from the implanted MNs, we used the IVIS to detect the persistence of the OVA fluorescence signal at the insertion sites. Fig. 9a shows that the fluorescent signal of the Alexa 594–OVA gradually decreased but was still detectable 21 days after insertion, signifying that chitosan MNs enable prolonged antigen exposure in the skin for at least 3 weeks. The fraction of OVA remaining in the MNs was 11

determined by measuring the intensity of the fluorescence signal in the skin at specific time points compared with the maximal intensity obtained at 2 h after MN insertion (Fig. 9b). As shown in Fig. 9b, approximately 50% of OVA was released from the MNs within 2 days, and almost no fluorescence was detected after 3 weeks. However, during this period, the thickness of the rat skin, including the epidermis and dermis, increased from approximately 950 (on day 1) to 1200 µm (on day 21) as the rats grew (Fig. S4), which resulted in the migration of the MNs deeper into the skin (Fig. 8). The skin thickness was determined by microscopic examination of skin’s histological sections (Fig. S4). In in vivo imaging, light was detected at the surface of the animals per square centimeter per steradian. The in vivo fluorescent signal decreased with an increase in tissue depth. Such light attenuation may have resulted in the overestimation of OVA release. These results demonstrate the degradability of the chitosan MNs and their potential for the sustained intradermal delivery of OVA for 3−4 weeks. 3.6. Antigen Dose-Sparing Effect Induced by Chitosan MNs To understand whether applying chitosan in the form of an MN depot enhances its adjuvanticity and whether chitosan MNs induce a clear antigen dose-sparing effect, we immunized the rats with MNs containing low-dose (200 µg) or full-dose (500 µg) OVA and compared them with those intramuscularly injected with full-dose OVA (500 µg) or OVA (500 µg) plus chitosan solution. Fig. 10 shows the OVA-specific IgG levels in rat serum after immunization. Among all immunized groups, the group intramuscularly injected with full-dose antigen alone had the lowest antibody levels, except at week 4; however, the levels were significantly enhanced in the groups injected with chitosan solution or received the chitosan MNs. These results reveal that chitosan is an effective adjuvant that can promote antigen-specific immune responses, when delivered either in the form of a solution or an MN depot. Compared with the same dose of OVA and the adjuvant, the MN-based delivery of OVA induced significantly stronger immune responses than did the IM injection of OVA plus chitosan solution from week 8 until week 18 (p < 0.05, n = 5). This result indicates that the 12

immunogenicity of the OVA plus chitosan solution was further increased when OVA was encapsulated and delivered by the chitosan MNs. Such an increase suggests that MN-based delivery and prolonged antigen retention in the skin can substantially enhance specific antibody formation. Notably, rats immunized with MNs containing low-dose OVA had persistently high antibody levels; the levels were significantly higher than those obtained from rats intramuscularly injected with full-dose OVA (p < 0.01, n = 5, except at week 4) and were equivalent to those obtained from rats intramuscularly injected with full-dose OVA plus chitosan solution (p > 0.05, n = 5). We observed that the chitosan MNs enabled immunization with a reduced dose of antigen, from 500 to 200 µg, thus resulting in at least a 2.5-fold dose reduction. These results demonstrate that the administration of chitosan in MN form can further enhance its adjuvanticity and that chitosan MNs have great potential for low-dose immunization.

4. Discussion This study developed implantable chitosan MNs with a patch-dissolvable design for the sustained ID delivery of antigens (Fig. 1) and demonstrated their antigen dose-sparing potential (Fig. 10). The MNs were combined with a dissolvable PVA/PVP supporting array patch that provides an additional length and strength to ensure the complete insertion of the MNs (Figs. 3 and 4). We demonstrated the successful and complete implantation of the chitosan MNs in the dermal layer as a long-lasting antigen depot (Figs. 7 and 8). Inserting the whole vaccine-loaded MNs into the skin guarantees the delivery of the intended dose and avoids vaccine wastage. Targeted antigen delivery to the dermal layer that contains numerous antigen-presenting cells can induce stronger immunogenicity than can an IM injection [27−30]. We previously reported a chitosan MN system, mounted on a PLA supporting array patch, for transdermal delivery of antigens [11]. In that system, because the PLA patch cannot be dissolved in the skin, it should be detached or removed from the MNs after 13

insertion. However, the incorrect removal of the array patch may partially pull the MNs out of the skin, resulting in failed insertion. In this study, we highlighted the dissolvable design of the supporting array patch. Using this new MN system, users do not need to separate the array patch from the MNs because it can be dissolved in the skin within a few minutes of insertion, which is markedly superior to our previous system. This novel generation of chitosan MN patches can provide a more feasible, convenient, and user-friendly option for long-term delivery of vaccines to the skin. Slow antigen release and prolonged antigen presentation are beneficial for generating strong and durable antibody responses [31,32]. In this study, the implanted MNs could remain at the injection site for 4 weeks (Fig. 8), thus allowing the sustained release of the encapsulated antigens to provide immunostimulation. After being implanted in the skin, the MNs were gradually hydrated and swollen by absorbing the skin interstitial fluid, thus resulting in an initial release. The IVIS results indicate that approximately half of the encapsulated OVA was released from the MNs within the first 2 days (Fig. 9), which may act as a priming dose to elicit a rapid immune response. Subsequently, the OVA remaining in the MNs was slowly released for 28 days because of the degradation of the chitosan MNs (Fig. 8). The extended release of antigens may further boost the immune response to maintain high antibody levels. A prime-boost immunization regimen has been considered an efficient strategy for generating protective immune responses and inducing immunity [33−35]. Ovalbumin (OVA), a major protein extracted from chicken egg white, is a commonly used model antigen to study various immunological questions [36]. Generally, the OVA dose used for immunization in mice was in the range between 2−100 µg [37−40]. However, considering a 10-fold difference in body weight between the rat and the mouse (150−200 g vs. 15−20 g), we estimate the maximum OVA dose used for rats would be 1000 µg [11]. In this study, MNs containing 500 and 200 µg of OVA were used as a full dose and a low dose for immunization in rats. According to our review of the relevant literature, this is the first study to investigate the antigen dose-sparing potential of chitosan MNs. We observed that specific antibody

14

responses induced by a single IM injection of full-dose OVA (500 µg) markedly decreased at week 6, whereas the corresponding responses elicited by MN immunization with single low-dose OVA (200 µg) remained at a relatively high level for at least 18 weeks (Fig. 10). These results reveal that the ID delivery of antigens by using the chitosan MNs enabled at least a 2.5-fold dose reduction for antigens, which may result in cost savings and increased access to limited vaccine supplies. Immunization with a single dose of nonliving antigen vaccines is generally difficult, because a single injection typically fails to elicit robust and durable immunity. In this study, we demonstrated that single low-dose ID immunization with the implantable chitosan MNs elicited higher and more sustained specific antibody responses than did IM immunization with full-dose OVA, indicating that the proposed MNs has potential to reduce the dose of vaccines and number of inoculations required to elicit effective immunity. Chitosan MNs can not only serve as an efficient ID delivery system but also exert their promising adjuvant activity by forming a depot for the prolonged release of antigens and by activating dendritic cells [4] for promoting immune responses.

5. Conclusion This study revealed that strong and persistent antibody responses for at least 18 weeks can be generated by administering a single dose of antigen encapsulated in chitosan MNs. The proposed MNs act as devices for ID vaccination and as adjuvants for sustained antigen release and immunostimulation to improve the immunogenicity of the delivered antigens. The proposed chitosan MN-based delivery system resulted in at least a 2.5-fold antigen dose reduction. The patch-dissolvable design renders the MN system easier and more convenient for use in patients. These results suggest that implantable chitosan MNs can reduce the cost of vaccination and serve as a single-dose vaccine delivery device for improving vaccine coverage.

15

Acknowledgments We thank the support from the laboratory animal center of NCKU and the technical services provided by the Bio-image Core Facility of the National Core Facility Program for Biotechnology, Ministry of Science and Technology, Taiwan. The authors gratefully acknowledge the financial support of the Ministry of Science and Technology of Taiwan (MOST 105-2628-B-006-008-MY3 and MOST 106-2221-E-006-058-MY3) and Wallace Academic Editing for editing this manuscript.

Supporting Information Available: materials; bright-field micrographs and detailed specifications of the master structures of MNs, supporting arrays, and pressing tools; schematic illustration of how to use a homemade applicator for the MN insertion; H&E-stained histological sections of rat skin; histological sections of rat skin at Days 1 and 21 after insertion of MNs.

Disclosure The authors declare that they have no competing interests.

16

References [1] G.J. Nabel, Challenges and opportunities for development of an AIDS vaccine, Nature 410 (2001) 1002–1007. [2] M. Kundi, New hepatitis B vaccine formulated with an improved adjuvant system, Expert Rev. Vaccines 6 (2) (2007) 133–140. [3] Y.M. Vasiliev, Chitosan-based vaccine adjuvants: incomplete characterization complicates preclinical and clinical evaluation, Expert Rev. Vaccines 14 (1) (2015) 37–53. [4] E.C. Carroll, L. Jin, A. Mori, N. Muñoz-Wolf, E. Oleszycka, H.B. Moran, S. Mansouri, C.P. McEntee, E. Lambe, E.M. Agger, P. Andersen, C. Cunningham, P. Hertzog, K.A. Fitzgerald, A. G. Bowie, E.C. Lavelle, The vaccine adjuvant chitosan promotes cellular immunity via DNA sensor cGAS-STING-dependent induction of type I interferons, Immunity 44 (3) (2016) 597–608. [5] K. Siddhapura, H. Harde, S. Jain, Immunostimulatory effect of tetanus toxoid loaded chitosan nanoparticles following microneedles assisted immunization, Nanomedicine 12 (1) (2016) 213–222. [6] H.S. Oberoi, Y.M. Yorgensen, A. Morasse, J.T. Evans, D.J. Burkhart, PEG modified liposomes containing CRX-601 adjuvant in combination with methylglycol chitosan enhance the murine sublingual immune response to influenza vaccination, J. Control. Release 223 (10) (2016) 64–74. [7] D. Fong, P. Grégoire-Gélinas, A.P. Cheng, T. Mezheritsky, M. Lavertu, S. Sato, C.D. Hoemann, Lysosomal rupture induced by structurally distinct chitosans either promotes a type 1 IFN response or activates the inflammasome in macrophages, Biomaterials 129 (2017) 127–138. [8] B.Y. Chua, T. Sekiya, M. Al Kobaisi, K.R. Short, D.E. Mainwaring, D.C. Jackson, A single dose biodegradable vaccine depot that induces persistently high levels of antibody over a year, Biomaterials 53 (2015) 50–57.

17

[9] J.M. Arya, L. Dewitt, M. Scott-Garrard, Y.W. Chiang, M.R. Prausnitz, Rabies vaccination in dogs using a dissolving microneedle patch, J. Control. Release 10 (2016) 19–26. [10] E.S. Esser, A. Romanyuk, E.V. Vassilieva, J. Jacob, M.R. Prausnitz, R.W. Compans, I. Skountzou, Tetanus vaccination with a dissolving microneedle patch confers protective immune responses in pregnancy, J. Control. Release 236 (2016) 47–56. [11] M.C. Chen, S.F. Huang, K.Y. Lai, M.H. Ling, Fully embeddable chitosan microneedles as a sustained release depot for intradermal vaccination, Biomaterials 34 (12) (2013) 3077–3086. [12] Y.H. Lin, F.L. Mi, C.T. Chen, W.C. Chang, S.F. Peng, H.F. Liang, H.W. Sung, Preparation and characterization of nanoparticles shelled with chitosan for oral insulin delivery, Biomacromolecules 8 (1) (2007) 146–152. [13] N.K. Brogden, M. Milewski, P. Ghosh, L. Hardi, L.J. Crofford, A.L. Stinchcomb, Diclofenac delays micropore closure following microneedle treatment in human subjects, J. Control. Release 163 (2) (2012) 220–229. [14] S.M. Bal, J. Caussin, S. Pavel, J.A. Bouwstra, In vivo assessment of safety of microneedle arrays in human skin, Eur. J. Pharm. Sci. 35 (3) (2008) 193–202. [15] K. Lee, J.D. Kim, C.Y. Lee, S. Her, H. Jung, A high-capacity, hybrid electro-microneedle for in-situ cutaneous gene transfer, Biomaterials 32 (30) (2011) 7705–7710. [16] M. Lee, C.G. Park, B.K. Huh, S.N. Kim, S.H. Lee, R. Khalmuratova, J.W. Park, H.W. Shin, Y.B. Choy, Sinonasal delivery of resveratrol via mucoadhesive nanostructured microparticles in a nasal polyp mouse model, Sci. Rep. 7 (2017) 40249. [17] X. Chen, T.W. Prow, M.L. Crichton, D.W. Jenkins, M.S. Roberts, I.H. Frazer, G.J. Fernando, M.A. Kendall, Dry-coated microprojection array patches for targeted delivery of immunotherapeutics to the skin, J. Control. Release 139 (3) (2009) 212–220.

18

[18] S.F. Lahiji, M. Dangol, H. Jung, A patchless dissolving microneedle delivery system enabling rapid and efficient transdermal drug delivery, Sci. Rep. 5 (2015) 7914. [19] L.Y. Chu, S.O. Choi, M.R. Prausnitz, Fabrication of dissolving polymer microneedles for controlled drug encapsulation and delivery: bubble and pedestal microneedle designs, J. Pharm. Sci. 99 (10) (2010) 4228–4238. [20] Q.L. Wang, D.D. Zhu, X.B. Liu, B.Z. Chen, X.D. Guo, Microneedles with controlled bubble sizes and drug distributions for efficient transdermal drug delivery. Sci. Rep. 6 (2016) 38755. [21] Y.A. Gomaa, D.I. Morrow, M.J. Garland, R.F. Donnelly, L.K. El-Khordagui, V.M. Meidan, Effects of microneedle length, density, insertion time and multiple applications on human skin barrier function: assessments by transepidermal water loss, Toxicol. In Vitro 24 (7) (2010) 1971–1978. [22] J. Gupta, H.S. Gill, S.N. Andrews, M.R. Prausnitz, Kinetics of skin resealing after insertion of microneedles in human subjects, J. Control. Release 154 (2) (2011) 148–155. [23] C. Curdy, A. Naik, Y.N. Kalia, I. Alberti, R.H. Guy, Non-invasive assessment of the effect of formulation excipients on stratum corneum barrier function in vivo, Int. J. Pharm. 271 (1-2) (2004) 251–256. [24] Y.C. Kim, J.H. Park, M.R. Prausnitz, Microneedles for drug and vaccine delivery, Adv. Drug Deliv. Rev. 64 (14) (2012) 1547–1568. [25] J. Arya, S. Henry, H. Kalluri, D.V. McAllister, W.P. Pewin, M.R. Prausnitz, Tolerability,

usability

and

acceptability

of

dissolving

microneedle

patch

administration in human subjects, Biomaterials 128 (2017) 1–7. [26] Z.R. Healy, A.T. Dinkova-Kostova, S.L. Wehage, R.E. Thompson, J.W. Fahey, P. Talalay, Precise determination of the erythema response of human skin to ultraviolet radiation and quantification of effects of protectors, Photodermatol. Photoimmunol. Photomed. 25 (1) (2009) 45–50.

19

[27] C.M. Fehres, J.J. Garcia-Vallejo, W.W. Unger, Y. van Kooyk, Skin-resident antigen-presenting cells: instruction manual for vaccine development, Front. Immunol. 4 (2013) 157. [28] M. Zaric, O. Lyubomska, O. Touzelet, C. Poux, S. Al-Zahrani, F. Fay, L. Wallace, D. Terhorst, B. Malissen, S. Henri, U.F. Power, C.J. Scott, R.F. Donnelly, A. Kissenpfennig, Skin dendritic cell targeting via microneedle arrays laden with antigen-encapsulated poly-D,L-lactide-co-glycolide nanoparticles induces efficient antitumor and antiviral immune responses, ACS Nano 7 (3) (2013) 2042–2055. [29] M. Zaric, O. Lyubomska, C. Poux, M.L. Hanna, M.T. McCrudden, B. Malissen, R.J. Ingram, U.F. Power, C.J. Scott, R.F. Donnelly, A. Kissenpfennig, Dissolving microneedle

delivery

of

nanoparticle-encapsulated

antigen

elicits

efficient

cross-priming and Th1 immune responses by murine Langerhans cells, J. Invest. Dermatol. 135 (2) (2015) 425–434. [30] P. Schipper, K. van der Maaden, S. Romeijn, C. Oomens, G. Kersten, W. Jiskoot, J. Bouwstra, Repeated fractional intradermal dosing of an inactivated polio vaccine by a single hollow microneedle leads to superior immune responses, J. Control. Release 242 (2016) 141–147. [31] L. Han, J. Xue, L. Wang, K. Peng, Z. Zhang, T. Gong, X. Sun, An injectable, low-toxicity phospholipid-based phase separation gel that induces strong and persistent immune responses in mice, Biomaterials 105 (2016) 185–194. [32] J.R. Adams, S.L. Haughney, S.K. Mallapragada, Effective polymer adjuvants for sustained delivery of protein subunit vaccines, Acta Biomater. 14 (2015) 104–114. [33] J.R. Vasconcelos, M.R. Dominguez, A.F. Araújo, J. Ersching, C.A. Tararam, O. Bruna-Romero, M.M. Rodrigues, Relevance of long-lived CD8+ T effector memory cells for protective immunity elicited by heterologous prime-boost vaccination, Front. Immunol. 3 (2012) 358. [34] A. Ciabattini, G. Prota, D. Christensen, P. Andersen, G. Pozzi, D. Medaglini, Characterization of the antigen-specific CD4(+) T cell response induced by

20

prime-boost strategies with CAF01 and CpG adjuvants administered by the intranasal and subcutaneous routes, Front. Immunol. 6 (2015) 430. [35] C.D. Lemke, S.M. Geary, V.B. Joshi, A.K. Salem, Antigen coated poly α-hydroxy acid based microparticles for heterologous prime-boost adenovirus based vaccinations, Biomaterials 34 (2013) 2524–2529. [36] A. Ikejiri, Y. Ito, S. Naito, K. Takada, Two- and three-layered dissolving microneedles for transcutaneous delivery of model vaccine antigen in rats, J. Biomater. Nanobiotechnol. 3 (2012) 325−334. [37] D. Mahony, A.S. Cavallaro, F. Stahr, T.J. Mahony, S.Z. Qiao, N. Mitter N, Mesoporous silica nanoparticles act as a self-adjuvant for ovalbumin model antigen in mice, Small 9 (2013) 3138−3146. [38] B.Y. Chua, T. Sekiya, M. Al Kobaisi, K.R. Short, D.E. Mainwaring, D.C. Jackson, A single dose biodegradable vaccine depot that induces persistently high levels of antibody over a year, Biomaterials 53 (2015) 50−57. [39] E.I. Wafa, S.M. Geary, J.T. Goodman, B. Narasimhan, A. K. Salem, The effect of polyanhydride chemistry in particle-based cancer vaccines on the magnitude of the anti-tumor immune response, Acta Biomater. 50 (2017) 417−427. [40] L. Han, J. Xue, L. Wang, K. Peng, Z. Zhang, T. Gong, X. Sun, An injectable, low-toxicity phospholipid-based phase separation gel that induces strong and persistent immune responses in mice, Biomaterials 105 (2016) 185−194.

21

Figure Captions Figure 1. Schematic illustrations of chitosan microneedles (MNs) with a patch-dissolvable design, consisting of antigen-loaded chitosan MNs and a dissolvable polyvinyl alcohol/polyvinyl pyrrolidone supporting array patch. After insertion, the supporting array can be quickly dissolved in the skin, thus implanting the MNs in the dermal layer as an intradermal (ID) depot to allow sustained release of the antigen and activating antigen-presenting cells. Figure 2. Schematic illustrations of the fabrication process for the chitosan-PVA/PVP microneedle (MN) patch. PDMS: polydimethylsiloxane; OVA: ovalbumin; PVA/PVP: polyvinyl alcohol/polyvinyl pyrrolidone; PLA: poly(L-lactide-co-D,L-lactide). Figure 3. Bright-field micrographs of the PVA/PVP supporting array patch (a and a1) and FITC-chitosan MNs combined with the patch (b and b1). The insets show their detailed dimensions. Figure 4. Skin insertion ability. Bright-field micrographs of MN patch after being applied to a porcine cadaver skin for 5 min (a) and the MN-treated skin after being stained with blue tissue marking dye (b). Histological sections of the rat skin (c and d) after insertion of the FITC-chitosan MNs: bright-field (c) and fluorescence (d) micrographs. The green fluorescence in (d) indicated the implanted FITC-chitosan MNs. Figure 5. Transepidermal water loss (TEWL) from the skin after MN application (n = 5 rats for each group). Baseline indicated the rat skin without MN treatment. Data are presented as mean ± SD. Figure 6. Skin irritation assessment. The change in redness (∆ a*) of the rat skin after MN insertion in comparison to the redness before insertion (a) (n = 5 rats for each time point). Bright-field micrographs of rat skins at different time points after MN insertion (b). PT in (a) indicates prior to treatment with MNs. The scale bar in (b) represents 1 mm. Data in (a) are presented as mean ± SD. Figure 7. In vivo MN implantation and OVA release. Confocal micrographs of the skin at varying depths on Day 1 after insertion of Alexa 594-OVA-loaded FTIC-chitosan MNs and their 2D x−z images. The red fluorescence indicates the Alexa 594-OVA; the green fluorescence indicates the FITC-chitosan MNs. Figure 8. In vivo MN degradation and OVA retention in rat’s skin. Histological sections of the skin at Days 7, 14, 21, and 28 after insertion of Alexa 594-OVA-loaded FTIC-chitosan MNs. The red fluorescence indicates the Alexa 594-OVA; the green fluorescence indicates the FITC-chitosan MNs. The arrows in the bright-field images point to the implanted MNs. The scale bar represents 300 µm. Figure 9. In vivo skin retention profile of OVA. Fluorescence images of the rats were obtained using an in vivo imaging system at defined times after administration of Alexa 594-OVA-loaded MNs to the dorsal skin (a). The fraction of OVA remaining in the dorsal skin was determined by measuring the intensity of the

22

fluorescence signal in the inserted sites in comparison to the maximal intensity obtained at 2 h after administration (b) (n = 5 rats for each time point). Data in (b) are presented as mean ± SD. Figure 10. OVA-specific IgG levels of rats after administration of a single dose of OVA on day 0: rats were intramuscularly injected with saline only (IM saline), or saline containing 500 µg OVA (IM full-dose OVA), or 500 µg OVA + 1 mg chitosan + 0.1 mg trehalose (IM full-dose OVA + CS); rats were received with MNs containing 200 µg (MN low-dose OVA) or 500 µg OVA (MN full-dose OVA) (n = 4 rats for each group). The actual OVA amounts in low-dose and full-dose MNs were 216 ± 11 and 533 ± 37 µg per patch (n = 5 patches for each group), respectively. Data are presented as mean ± SD. *: p < 0.05; **: p < 0.01.

23

Figure 1 antigen

CS MN dissolvable supporting array

Epidermis

5 min Dermis

patch dissolution

skin insertion

MN degradation

MN depot for sustained intradermal delivery and activating antigen-presenting cells

Langerhans cell Dendritic cell

MN implantation

Figure 2 OVA/Chitosan gel

PVA/PVP solution

Cutting off the patch

Centrifugation

Drying

PDMS mold Chitosan-PVA/PVP MN patch

razor blade

Make chitosan MNs compact

PVA/PVP supporting array

Press MNs down

PVA/PVP sol. Drying

PLA pressing tool

Aligning & integrating

Figure 3

a1 600 μm

a

b1

MN

Supporting array

600 μm

b

600 μm

< 10 μm

Figure 4

a

b

2 mm

c

d

Figure 5 40

Baseline Microneedles (n = 5)

35

TEWL(g/m2h)

30 25 20 15 10 5 0 -1

0

1

2

3

Time (hr)

4

5

6

Figure 6

Change in Erythema, Δa* (AU)

a

b 5

(n = 5)

4

0 min

15 min

6h

1 day

3 2

1 0 -1

PT 15 min 2 h

6h

Time

1d

3d

5d

2D xz image

Figure 7

FITC-Chitosan FITC-Chitosan

Alexa 594-OVA

merged

0 µm

0 µm

0 µm

900 µm

900 µm

900 µm

900 µm

900 µm

900 µm

Day 28

Day 21

Day 14

Day 7

Figure 8 bright-field

FITC-Chitosan

Alexa 594-OVA

merge

Figure 9 b 1.2

a

Normalized Radiance

Epi-fluorescence

6.0 108 4.0

2.0

Alexa 594-OVA (n = 5)

1 0.8 0.6 0.4 0.2 0 0

0 day

2 day

7 day

7

14

21

Time (day)

14 day

21 day

Figure 10

OVA-specific IgG level absorbance at 450 nm

2.0

1.5

(n = 4) IM saline IM full-dose OVA IM full-dose OVA + CS MN low-dose OVA MN full-dose OVA ** **

** *

**

**

**

*

** **

**

*

1.0

0.5

0.0 0

2

4

6 8 10 week (post-vaccination)

12

14

18

Statement of Significance: This study developed implantable chitosan microneedles (MNs) with a patch-dissolvable design for the sustained intradermal (ID) delivery of antigens and demonstrated their antigen dose-sparing potential. We found that rats immunized with chitosan MNs containing low-dose OVA had persistently high antibody levels for 18 weeks, which were significantly higher than those observed after an intramuscular injection of full-dose OVA, demonstrating at least 2.5-fold dose sparing. Our results indicate that chitosan MNs can not only serve as an efficient vaccine delivery system but also exert their promising adjuvant activity by forming an ID depot for prolonged antigen exposure and activating dendritic cells for promoting immune responses.

Graphical Abstract