Skin appendages such as hair follicles, sebaceous glands, and sweat glands have also been shown to play an important role in increasing skin permeability, particularly for large molecules. Appendages create a channel in the stratum corneum and facilitate the dermal absorption of topically applied compounds. Another method that has been researched extensively for enhancing skin permeability is the microneedle technology. Microneedles can disrupt the stratum corneum in the micron scale and thus can be used to enhance drug transportation into or across the skin. In this chapter, cutaneous DNA immunization by modifying the hair follicle cycle and by microneedle application is discussed in details.

Hair follicular targets mainly include four sites: infundibulum, sebaceous gland, bulge region, and hair bulb (Knorr et al. 2009; Wosicka and Cal 2010). The infundibulum region creates the major entry point for topically applied substances, especially in the lower part where the epidermis is not fully differentiated. APCs, mast cells, and other immune cells are also located inside and around the infundibular epithelium, which makes HFs a potential target for cutaneous immunization. Sebaceous gland is another important therapeutic target site. It is associated with the etiology of androgenetic alopecia, acne, and other sebaceous gland dysfunctions; therefore, it can be used for the therapy of these diseases. The bulge region in HFs contains epithelial stem cells, which has a high proliferative capacity and multipotency. Thus, the bulge region is of particular interest for gene therapy targeting congenital hair disorders or genetic skin disorders. Finally, the hair bulb with the dermal papilla and the hair matrix connected to the blood vessels is important in the regulations of hair growth and pigmentation (Knorr et al. 2009; Wosicka and Cal 2010).

Along with HF morphology, transfollicular delivery also depends upon the functional status of HFs. Each HF goes through a highly regulated continuous cycle comprised of three major stages: anagen, catagen, and telogen (Paus and Cotsarelis 1999). In anagen (growth) phase, cells proliferate rapidly and continuously to form the inner root sheath and migrate upward to form the hair shaft. Catagen (regressing) phase is the end of mitosis where cell death of the lower follicle segment occurs. In telogen (resting) phase, the hair eventually shed. Under physiological conditions, 85–90 % of human scalp follicles are within the anagen phase (lasting for 2–6 years), 1–2 % of scalp follicles are in the catagen phase (lasting for 2–4 weeks), and about 10 % are in the telogen phase (lasting for 2–4 months) (Paus and Cotsarelis 1999).

Slominski et al. reported that the characteristics of the anagen phase include thickened dermal and epidermal layers of the skin, increased size of HFs, extension of follicles deep into the dermal adipose tissue, and initiation of melanin synthesis (Slominski et al. 1991). Otberg et al. found that the penetration of topically applied substances into the infundibulum is dependent on the active and inactive states of HFs. Active HFs exhibiting hair growth (anagen) and/or sebum production are open for the penetration process, but telogen follicles do not support any follicular penetration (Otberg et al. 2004).

Domashenko et al. showed that topical application of a lacZ gene-encoding plasmid complexed with liposomes (lipoplexes) led to efficient in vivo transfection of hair follicular cells, and the transfection was dependent upon the stage of the hair follicular cycle (Domashenko et al. 2000). Transfection occurred only at the onset of the growing stage of the hair follicular cycle created by hair depilation. Expression of β-galactoside gene was detected only in the HFs of mice transfected with lipoplexes that contained lacZ plasmid on the first, second, or third day after hair depilation. Transfection 3 days after the depilation did not result in any β-galactosidase activity. During the first 3 days after depilation, the HFs were in anagen-onset stage. Thus, the authors concluded that the timing of plasmid application when the HFs are just entering the anagen stage is critical for efficient in vivo transfection of HFs (Domashenko et al. 2000). Gupta et al. also supported this idea by reporting that gene transfection of HFs occurs only during the anagen-onset stage (Gupta et al. 2001). Thus, the anagen stage of HFs is important for transfollicular penetration.

Fan et al. demonstrated that topical application of plasmids encoding lacZ or hepatitis B surface antigen (HBsAg) to untreated mouse skin can induce specific immune responses (Fan et al. 1999). The authors also reported that topical gene transfer is dependent on the presence of normal HFs. Plasmid DNA was detected in and around the HFs in normal hair-bearing skin. The skin that lacks normal HFs failed to retain plasmid DNA. The authors further showed that normal HFs are important for the induction of immune responses. Application of plasmid DNA onto a skin area that lacks normal HFs in immunocompetent mice failed to generate an immune response against HBsAg (Fan et al. 1999).

Anagen phase can be artificially induced by experimental methods including hair plucking, vigorous shaving, or chemical exposure such as caustic materials and depilatory agents.

Ishimatsu-Tsuji et al. defined mouse HF cycle by different gene expression patterns following hair plucking using wax (Ishimatsu-Tsuji et al. 2005). The period of up to 2 days after hair plucking (early anagen) is defined as the defense response phase, where mainly inflammation-related genes are upregulated. The period of 3–6 days following hair plucking (mid-anagen) is defined by keratinocyte proliferation processes, whereas late anagen and early catagen phase are characterized by multiple processes including keratin intermediate filament formation (Ishimatsu-Tsuji et al. 2005).

We hypothesized that topical application of plasmid DNA onto a skin area wherein the HFs are in the anagen-onset stage will induce a stronger immune response than when the HFs are in the telogen stage. In order to test this hypothesis, the HFs on the dorsal side of 6-week-old mice were induced (from telogen stage) into anagen-onset stage by hair plucking with a Veet® wax strip from Reckitt Benckiser, Parsippany, NJ, USA (i.e., “cold” waxing). Veet® wax strips come in the form of pre-waxed strips. Strip was placed over the hair-trimmed skin area and rapidly removed to detach the hair from the skin surface. Two days later, plasmid that encodes anthrax protective antigen (PA63) protein (pGPA) was applied onto the plucked skin area. Cholera toxin was used as an adjuvant. Our data showed that cutaneous immunization by applying the pGPA plasmid onto a skin area wherein the HFs had been induced into anagen-onset stage induced a stronger anti-PA antibody response, increased the antigen gene expression, and enhanced the retention of the plasmid DNA, as compared to topical application of the same pGPA plasmid onto a skin area wherein the HFs were left in the telogen stage (Shaker et al. 2007).

In the “cold” waxing-based studies, the hair plucking was performed by “cold” waxing, which in some cases can result in significant breakages of the hair at or below the surface of the skin. In order to remove the hair completely and expand the HF significantly, we used a “warm” waxing method to pluck hair, expecting that it would further enhance the penetration of plasmid DNA into the HFs and thus induce a stronger immune response than “cold” waxing-based hair plucking. In the “warm” waxing method, warm wax in viscous liquid form is directly applied onto a hairy skin area. The warmth of the wax is expected to allow the HF pores to expand and thus complete removal of the hair from the follicles becomes possible (Sloat et al. 2012a).

The immune responses induced by cutaneous immunization with the pGPA plasmid onto a skin area wherein the hair had been plucked 2 days earlier using “cold” versus “warm” waxing techniques were evaluated and compared. Plucking the hair by “warm” waxing resulted in an anti-PA IgG titer significantly higher than by “cold” waxing (Sloat et al. 2012a). Relatively more PA63 gene fragments were recovered from the skin area where the hair was plucked by “warm” waxing than by “cold” waxing, indicating that “warm” waxing led to more DNA uptake and/or retention in the skin than “cold” waxing (Sloat et al. 2012a).

Based on the studies performed in our lab, we conclude that the immune response induced by cutaneous DNA immunization via HFs can be enhanced by modifying the HFs or HFs cycle.

A new penetration-enhancing technique “cyanoacrylate skin surface stripping” (CSSS) was introduced by Schaefer and Lademann in 2001 and has been shown to be a promising method for transfollicular penetration by opening up the follicular infundibula (Schaefer and Lademann 2001). CSSS facilitates follicular penetration of vaccines by removing cellular debris and sebum from the hair follicular openings (Schaefer and Lademann 2001). Toll et al. used CSSS technique to deliver fluorescent microspheres of size 0.75–6.0 μm on human skin and demonstrated that without pretreatment with CSSS, the transfollicular microsphere penetration was below 25 % of the total number of HFs with a maximum penetration depth of 1500 μm in the hair follicles. However, CSSS pretreatment of the skin increased the transfollicular microsphere penetration to 35–87 % of the HFs with a penetration depth of more than 2300 μm (Toll et al. 2004).

Multiple efforts have also been made to explore nanoparticles as transfollicular carrier for vaccine delivery inside the skin. Vogt et al. investigated the penetration of 40, 750, and 1500 nm nanoparticles across human skin and their possible targeting to APCs inside the skin (Vogt et al. 2006). In their study, CSSS was performed prior to applying nanoparticles cutaneously. Flow cytometric data showed that after cutaneous application, only the 40 nm particles entered epidermal LCs, clearly indicating that only smaller nanoparticles can deeply penetrate into hair openings and through the follicular epithelium into perifollicular dermis. Thus, it was concluded that 40 nm nanoparticles, but not 750 or 1500 nm nanoparticles, are suitable for cutaneous immunization via the HFs (Vogt et al. 2006).

Mahe et al. performed a study to evaluate the potential of using the HF pathway for vaccine delivery. The authors studied the penetration of 40 or 200 nm fluorescent polystyrene nanoparticles as well as virus particles and demonstrated that both polystyrene nanoparticles and the modified vaccinia Ankara (MVA) viral particles penetrated deeply into HFs and were internalized by epidermal and dermal APCs. Topical application of ovalbumin-encoding plasmid DNA or MVA induced both humoral and cellular immune response. IgG response induced by ovalbumin-encoding plasmid DNA alone after cutaneous administration was found similar to that induced by ovalbumin protein delivered cutaneously in combination with cholera toxin adjuvant. However, in this study, authors preferred to perform tape stripping at the place of CSSS before applying vaccines onto the skin surface (Mahe et al. 2009).

In a clinical study performed by Vogt et al., it was demonstrated that cutaneous administration of influenza protein-based vaccine using CSSS technique promoted both CD4⁺ T and CD8⁺ T cell immune responses in human subjects, whereas intramuscular injection of the same vaccine generated only CD4⁺ T cell response (Vogt et al. 2008).

Another clinical study based on CSSS procedure was performed by Combadiere et al. (Combadiere et al. 2010). Immune response generated by cutaneous application of a combined tetanus and influenza vaccine suspension after CSSS procedure in the upper left arm area was compared with the intramuscular immunization with same vaccine. Results showed that cutaneous application of the vaccine was safe and that the inactivated influenza vaccine induced a CD8⁺ T cell response. Tetanus specific cellular response level was not detected (Combadiere et al. 2010).

Microneedles are one of the new technologies studied in transdermal drug delivery. This is a simple, pain-free, and minimally invasive technology that helps the delivery of large molecular weight and hydrophilic compounds across skin. Microneedles consist of a plurality of microprojections of different shapes, sizes, and heights, which are attached to a base support. Application of microneedles onto skin surface creates micron size transport pathways, which allow the permeation of macromolecules inside the skin. According to Kaushik et al., microneedles perforate the stratum corneum, avoiding any contact with nerve fibers and blood vessels present in the dermis layer (Kaushik et al. 2001). Therefore, the major benefit of using microneedles is the pain-free delivery of both small and large molecular weight compounds (Kaushik et al. 2001; Prausnitz 2004).

The concept of microneedle first came in the 1970s; however, technological limitations at that time prevented the product concept from being executed (Gerstel and Place 1976). In the 1990s with the beginning of high-precision microelectronics industrial tools, microneedle manufacturing also became a reality. The first microneedle array reported in the literature was made from silicon wafer and was developed for gene delivery (Hashmi et al. 1995). In that study, the needles were inserted into nematodes to increase molecular uptake and gene transfection. After transfection, the gene of interest was expressed in the progeny of the injected worms (Hashmi et al. 1995). The first paper to report microneedles for transdermal drug delivery was published by Henry et al. in 1998, who demonstrated that microneedles can be used to increase skin permeability (Henry et al. 1998). An array of solid microneedles of 150 μm length was used to create micropores in human epidermis. These micropores increase the skin permeability by three orders of magnitude for a small model compound calcein. The transportation of calcein took place through the leakage pathways created between the needles and the skin. When the needles were removed from the skin, the skin permeability was increased by another order of magnitude (Henry et al. 1998). In a follow-up study, McAllister et al. studied the permeability of various compounds across cadaver skin using Franz diffusion cell and noted that insulin, bovine serum albumin, and latex nanoparticles (up to 100 nm size) could cross the skin pretreated with microneedles (McAllister et al. 2003). These early studies demonstrated that skin permeability can be increased using microneedles.

Microneedles are somewhat like traditional needles but are fabricated on the micron scale. The most commonly used fabrication materials are metals, silicon, silicon dioxide, polymers, and glass. The first few microneedle devices were fabricated from silicon (Henry et al. 1998; Hashmi et al. 1995). Later, other materials such as stainless steel (Verbaan et al. 2007), dextrin (Ito et al. 2006a, b), glass (Wang et al. 2000), ceramic (Ovsianikov et al. 2007), maltose (Kolli and Banga 2008), galactose (Donnelly et al. 2009), and various polymers (McAllister et al. 2003; Park et al. 2005) have also been used to fabricate microneedles. The most commonly used methods for the manufacturing of microneedles are chemical isotropic etching, injection molding, reactive ion etching, surface/bulk micromachining, polysilicon micromolding, lithography-electroforming-replication, and laser drilling (Park et al. 2005; Trichur et al. 2002; Yang and Zahn 2004; Davis et al. 2005; Stoeber and Liepmann 2005). The two basic designs of microneedles are in plane and out plane designs.

The use of microneedles in delivering vaccines into the epidermis and dermis compartments of skin is a very attractive approach for cutaneous vaccination. Four different microneedle designs have been developed for the delivery of vaccines inside the skin.

As discussed earlier, it is a well-known fact that skin offers a great immunologic environment compared to muscle and subcutaneous tissues. However, so far there is no simple, reliable, and safe method available to deliver vaccines into the skin efficiently in a cost-effective way to a large population of people. The microneedle device has the potential to address these problems to some extent. The painless microneedles not only can improve patient compliance but also enable the targeting of antigens to the immune cell-rich skin layers. In the past decades, many different vaccines and/or antigens, including live attenuated microorganisms, proteins, and plasmid DNA, have been delivered into the skin using microneedles. For example, influenza vaccination using whole inactivated influenza viruses coated on microneedles has been studied extensively. Studies have shown a complete protection against infection after microneedle-mediated cutaneous immunization with influenza virus A/Puerto Rico/8/34 (H1N1), influenza virus A/California/04/09 (H1N1), and influenza virus A/Aichi/2/68 (H3N2) (Zhu et al. 2009; Koutsonanos et al. 2009, 2011). Along with whole inactivated influenza virus, virus like particles (VLP) coated on microneedles were also studied. It was shown that cutaneous immunization with avian H5 influenza VLPs coated on stainless steel microneedles induced a stronger immune response in a mouse model, relative to intramuscular injection (Song et al. 2010a, b). Microneedles have also been used for the delivery of recombinant subunit vaccines, such as trimeric influenza hemagglutinin protein. Upon administration, protein antigen coated on microneedles induced an enhanced immunity, relative to the subcutaneous injection of the same protein antigen in mice (Weldon et al. 2011). In a study in our laboratory, we have demonstrated that ovalbumin chemically conjugated onto solid lipid nanoparticles and then administered cutaneously onto a skin area pretreated with microneedles induced a stronger immune response as compared to cutaneous immunization using ovalbumin alone (Kumar et al. 2011). The dose of the antigen determined whether the microneedle-mediated immunization can induce a stronger immune response than subcutaneous injection of the same antigen conjugated on the solid lipid nanoparticles (Kumar et al. 2011). Finally, plasmid DNA has also been explored extensively for microneedle-mediated cutaneous immunization, which is discussed below in details.

Cutaneous DNA immunization has been performed using gene gun and electroporation techniques and was found to improve the immune responses by manyfolds in human and nonhuman primates (Donnelly et al. 2003). However, gene gun and electroporation techniques are tedious and require a vaccination protocol and equipment (Nicolas and Guy 2008). Microneedles as a new technology may deliver DNA vaccine painlessly into the skin and is potentially suitable for inexpensive mass production. The first study showing the delivery of DNA using microneedles was performed by Hashmi et al. in 1995. In their study, DNA-coated microprobes were injected in the cuticle of nematode Heterorhabditis bacteriophora to express a foreign gene, β-galactosidase (Hashmi et al. 1995). After this study, microneedle devices were shown to increase the skin permeability of plasmid DNA in vitro (McAllister et al. 2000). However, there was not any systematic in vivo study performed until 2002, when Mikszta et al. reported the first in vivo study using microneedles to deliver DNA vaccine topically into the skin. In that study, microenhancer arrays (MEAs, i.e., microneedles) were fabricated using potassium hydroxide etch technique. The array was fabricated on 1 cm² microchips and had the projection height ranging from 50 to 200 μm. Naked plasmid DNA encoding firefly luciferase was administered to mice topically using the MEAs. The arrays were dipped into plasmid DNA solution and scraped multiple times across the skin of mice to create microabrasions. The results demonstrated that the mean luciferase activity in groups treated with six or more passes of MEA were 1000- to 2800-fold higher in comparison to topical controls. Untreated skin was used as a control for possible delivery through HFs. The gene expression level in the MEA-treated groups was also found similar or greater than after intramuscular and intradermal injection of plasmids. The results showed that even a single pass of the MEA across the skin enabled gene transfer. The authors also used a plasmid DNA that encodes hepatitis B surface antigen and reported that MEA-based cutaneous immunization induced a stronger immune response with less variability after the second and third immunizations, as compared to immunization by hypodermic needle injection. Both hypodermic needle injection and MEA-based delivery induced similar antigen-specific cytotoxic T lymphocyte (CTL) responses (Mikszta et al. 2002). The significance of this study is that it demonstrates the feasibility of cutaneous DNA vaccination using microneedles.

In separate studies by Birchall et al. and Coulman et al., it has been demonstrated that microneedles can facilitate the delivery of plasmid DNA into the epidermis layer of the human skin, and the genes encoded by the plasmid can be expressed in the skin. Birchall et al. delivered pCMV-β plasmid into the skin with the help of silicon microneedles and demonstrated a detectable gene expression in the microchannels. In control groups, where DNA was delivered without microneedle treatment, expression was not observed. Coulman et al. also used silicon microneedle arrays to create micropores through the stratum corneum of human skin samples to test the ability of the conduits to facilitate the delivery of pCMV-β plasmid. Their results demonstrated that these micropores facilitated the delivery of plasmid DNA, and gene expression study confirmed that naked plasmid DNA was able to be expressed in excised human skin. However, the presence of limited conduits that were positive for gene expression indicated that there is a need to optimize the microneedle device morphology, the application method, and the DNA formulation in order to obtain a higher gene expression (Birchall et al. 2005; Coulman et al. 2006).

Alarcon et al. performed an animal study to use microneedles to deliver three different types of influenza vaccines: whole inactivated influenza virus, trivalent split-virion human vaccine, and plasmid DNA encoding the influenza virus hemagglutinin. In their study, rats were chosen as an animal model since rat skin is thicker than the total length of microneedles (i.e., 1 mm) used in the study. Antigens were administered intradermally using hollow microneedles and intramuscularly using hypodermic needles. Results showed that intradermal delivery using hollow microneedles generated an antibody response similar to intramuscular injection; however, dose-sparing effect was achieved when microneedles were used (Alarcon et al. 2007).