DNA Vaccine Delivery

October 2, 2011
Rocky Cranenburgh

BioPharm International, BioPharm International-10-02-2011, Volume 2011 Supplement, Issue 7

Development of the ideal DNA vaccine requires the optimization of delivery strategies and plasmid vectors.

A DNA vaccine is a plasmid produced in Escherichia coli that contains an antigen gene controlled by a promoter that functions only in animal cells, usually the cytomegalovirus promoter. The DNA vaccine plasmid must be delivered to antigen-presenting cells (APCs) of the host, where the antigenic protein is expressed, processed, and presented to the immune system. This approach works efficiently in mice and other animal models when DNA is delivered by intramuscular injection, and DNA vaccines have been licensed for veterinary applications, yet no DNA vaccine has been approved for use in humans. A key problem is achieving the delivery of sufficient plasmid to the APCs. This article discusses the approaches designed to achieve delivery, including a focus on the use of live bacterial vectors.


Since DNA vaccination was developed in the early 1990s, the most common method for immunization has been intramuscular injection of DNA. The DNA is usually dissolved in water or an isotonic saline solution, with the inclusion of an adjuvant if necessary. The aim is to get the plasmid into APCs, which in muscle tissue are primarily dendritic cells. There, the DNA is transported to the nucleus for expression, and the resulting polypeptide is processed and presented on the cell surface in a major histocompatibility complex (MHC). DNA vaccination can therefore produce a protective CD8+ cytotoxic T lymphocyte (CTL) response via MHC Class I (1). Also important are bystander effects, whereby antigens expressed and released from adjacent myocytes (i.e., muscle cells) are taken up by APCs, thus triggering antibody production through MHC Class II. This effect enables DNA vaccines to be developed to target a wide range of bacterial, viral, and parasitic diseases, in addition to the rapidly expanding field of cancer immunotherapy.

The first licensed DNA vaccine was West Nile-Innovator DNA, approved in 2005 for the immunization of horses against West Nile Virus. It was given in two intramuscular doses 2–4 weeks apart, then as a single dose annually (2). However, this vaccine has now been discontinued by Pfizer following their acquisition of the developers, Wyeth's Fort Dodge Animal Health. Also in 2005, Novartis Animal Health (Basel, Switzerland) gained approval for Apex-IHN, a vaccine against infectious hematopoietic necrosis virus in salmon, which encodes a viral glycoprotein and is administered by a single intramuscular dose of 10 µg DNA (3).

Despite promising results in animals, DNA vaccine efficacy has been disappointing in human clinical trials. This is generally because of a much lower specific immune response generated by injected plasmid DNA alone, and has led to the adoption of a prime-boost vaccination strategy whereby the plasmid DNA injection is followed by a boost with the same antigen gene in a viral vector. This heterologous boosting strategy has significantly increased the specific immune response. Commonly-used attenuated viral vectors include those based on modified vaccinia ankara (derived from the smallpox vaccine), lentivirus (mainly derived from HIV), and adenoviruses of primate origin. Viral vectors are expensive and complex to manufacture compared with DNA, thus eliminating the advantages of the DNA vaccine approach such as lower cost and simple, generic production processes. An additional disadvantage is the potential of a host immune response directed against the vector. Development of the ideal DNA vaccine therefore requires the optimization of delivery strategies and plasmid vectors.


A traditional approach for inserting DNA into microbial and animal cells in culture has been by the application of an electric pulse, which creates transient pores in the cell membrane and allows entry of the negatively charged DNA molecules. Electroporation has been adapted from its in vitro single-cell origins to enhance in vivo delivery to tissues following intramuscular injection, and enables a significant increase in transfection efficiency and immune response over injection alone. The disadvantages include pain at the injection site due to the electric pulse, and the requirement for specialized electroporation devices. The most common approaches use a disposable grid consisting of injection and electrode needles attached to the device that provides the electric pulse. The DNA solution is injected, needle electrodes are inserted through the skin and into muscle tissue, and an electric field applied to facilitate DNA entry into dendritic cells and myocytes.

Ichor Medical Systems (San Diego, CA) has developed the TriGrid Delivery System (TriGrid) a hand-held electroporation device that has been demonstrated to increase DNA vaccine delivery efficiency by as much as 1000-fold over injection alone. Four electrodes are arranged in a diamond shape around the central needle, which is a conventional single-use hypodermic syringe that is inserted into the device, and an electric pulse is administered upon injection (see Figure 1a). Ichor has demonstrated greatly enhanced potency of DNA vaccines delivered with the TriGrid in several animal models using DNA vaccines that encode antigens such as anthrax, hepatitis B virus, and tumor-associated antigens. The device is being evaluated in several clinical trials, and results demonstrating enhanced responses to a prophylactic HIV DNA vaccine delivered with TriGrid compared with conventional needle injection have been published recently (4). In addition, Ichor has shown in animal models that delivery of DNA encoding the protein therapeutics erythropoietin and interferon-β may be a viable alternative to injection of the recombinant proteins.

Figure 1: Devices for plasmid DNA delivery indluding (a) TriGrid and (b) CELLECTRA with applicator (for intramuscular electroporation), (c) DermaVax (for transdermal electroporation) and (d) ZetaJet (for subcutaneous or intramuscular high-pressure injection).

A therapeutic plasmid DNA application (although not a vaccine) that was licensed for pigs in 2008 is LifeTide SW, developed by VGX Animal Health (The Woodlands, TX) (5). The plasmid produces growth hormone-releasing hormone and is administered to sows of breeding age to increase the number of piglets weaned, and requires only a single treatment. Plasmid delivery is by direct injection into skeletal muscle, followed by electroporation using a portable electrokinetic device. The device, called CELLECTRA, was developed by Inovio (see Figure 1b) and uses a sterile, disposable electroporation needle array. Inovio has also demonstrated 1000-fold greater DNA delivery compared with injection alone. The device is equally suited to DNA vaccine applications, and users of CELLECTRA have targeted a range of pathogens including HIV, Clostridium difficile, hepatitis C virus, and malaria parasites. Inovio has an ongoing Phase II clinical trial for cervical-cancer therapy (human papilloma virus therapeutic vaccine) using the device. Additional ongoing human clinical trials include Phase I trials for prophylactic vaccines for influenza (i.e., avian and seasonal) and HIV, and a Phase I clinical trial for HIV therapy.

While ample evidence indicates electroporation enhances the delivery and immunogenicity of injected DNA vaccines, it has the drawback of increased discomfort at the administration site due to the electrical pulses. However, initial human clinical-trial data with intramuscular delivery indicates that the electroporation procedure is tolerable for therapeutic and even prophylactic applications. Furthermore, continued improvements to device design are likely to result in increased acceptability of the procedure. A novel approach to delivering electrical pulses within target tissue that may increase tolerability is the generation of electrical fields by piezoelectricity, as used in a contactless, noninvasive device developed by Inovio.

Another interesting alternative to penetrative electrodes has been developed by MagneGene (Lake Forest, CA), which uses contactless magnetic electroporation. This technique relies on a strong, rapidly changing magnetic field generated by a paddle-shaped device to induce rotating electric fields in the host tissue. This device has been evaluated in guinea pigs following injection with reporter plasmid, where enhanced in vivo gene expression was demonstrated. Although the magnetic field causes muscle spasms, these subside immediately after use of the device, and there is no pain following the procedure (6).


Transdermal DNA immunization involves the use of arrays of microneedles, each a few hundred microns long, to pierce the barrier of the stratum corneum (i.e., the skin's outer layer, typically 10–20 µm thick) and deliver the vaccine (7). The skin has a high concentration of APCs called Langerhans cells, which makes it an attractive target. The advantages of microneedles are that they are easy to administer and cause less pain at the injection site than does a conventional needle. There are various strategies to achieve vaccination using microneedles. The simplest involves arrays of microneedles that are used to pierce or scarify the stratum corneum, thus increasing permeability to a topically applied DNA solution. This approach is suitable for preclinical studies, but is not scalable. Alternatively, the microneedles can be coated with the dried vaccine which dissolves following administration. Hollow microneedles can be used to inject solutions containing the vaccine. These can be submillimeter arrays or scaled-down needles in the range of 1–2 mm. An interesting alternative to fixed, disposable microneedles is a solid, soluble microneedle array that is either formulated or coated with the vaccine. This array is inserted into the skin, where the microneedles dissolve or degrade, leaving only the backing to be disposed of and thus eliminating contaminated sharps.

Smallpox DNA vaccination of mice was demonstrated using the Easy Vax device supplied by Cellectis therapeutics (Paris), and was the first example of microneedle-mediated electroporation (8). The gun-shaped Easy Vax combines the electroporation and microneedle approaches, with cutaneous Langerhans cells as the targeted APCs. The plasmid DNA was dried onto the tips of the eighty-microneedle array and inserted into the skin to enable the DNA to dissolve, followed by the delivery of six electric pulses.

While the microneedle approach is gaining interest for delivery of therapeutic and antigenic recombinant proteins, only a few studies in mice have shown immune protection against microneedle-administered DNA vaccines. An additional problem when translating animal results to humans relates to the amount of DNA that can be loaded onto a microneedle array. A typical human clinical trial injected dose of 2–4 mg DNA is already relatively small compared with the murine equivalent from which it is extrapolated, yet this would be prohibitively high to expect to dissolve intradermally from a dried formulation on a microneedle array, which leaves liquid DNA solution injection as the only viable strategy.

Cellectis therapeutics also supplies the DermaVax electroporator (see Figure 1c), which requires a DNA solution to be injected sub-dermally using a hypodermic syringe, followed by electroporation through a series of pulses from 2 mm needle electrodes attached to the bench-top DermaVax apparatus. The DermaVax system induces a 100- to 1000-fold greater increase in gene expression over injection alone and is being used for the delivery of DNA vaccines against HIV-1, as well as vaccines against prostate and colorectal cancers in human clinical trials (9).

Inovio Pharmaceuticals has developed a transdermal electroporation device called CELLECTRA-3P. In pilot studies in nonhuman primates, Inovio demonstrated that intradermal administration of a smallpox vaccine or an H5N1 vaccine successfully protected animals from a lethal challenge of monkeypox or an H5N1 infection, respectively (10, 11). The device is now in Phase I clinical trials for a universal influenza vaccine. Inovio has also developed a prototype minimally invasive electroporation device that has been demonstrated for influenza DNA vaccine delivery.


Gene guns have been used since the advent of DNA vaccination for high-pressure transdermal delivery of microbeads coated with DNA (biolistics). PowderMed (Oxford, UK, acquired by Pfizer) developed a handheld particle-mediated epidermal delivery (PMED) device that uses a high-pressure helium minicylinder to fire DNA-coated gold microbeads into the epidermis. Because the gold beads enter the cell cytoplasm and target more APCs than does needle injection, greater CTL responses can be generated for the same amount of DNA (11). In one example, PMED was used to deliver 2 µg of two plasmids to rhesus macaques: one with an H1N1 hemagglutinin gene, the other expressing the adjuvant GM–CSF. This method generated good antibody and CTL responses (12). However, limitation to the microbead approach is the amount of DNA that can be coated onto beads, which could restrict its ability to be scaled up for human applications.

Merial (Duluth, GA) developed the pioneering canine melanoma DNA vaccine Oncept, which was fully approved in 2010. It contains the human tyrosinase gene, delivered (following surgical tumor removal) in four doses at two-week intervals, followed by a single dose every six months. Oncept is administered using a needle-free Canine Transdermal Device developed by Bioject Medical Technologies (Portland, OR) (13). Bioject specializes in the development of needle-free injection systems that use high-pressure injection to fire a stream of the DNA vaccine solution through the epidermis. Their latest device, ZetaJet, is small and spring-activated for ease of use in the field, and fires up to 0.5 mL into skin or muscle tissue.


The oral route for vaccination avoids needles and reduces the logistical issues and costs of implementing vaccination programs requiring trained healthcare professionals, which would be particularly beneficial in developing countries. Oral vaccination also stimulates a mucosal immune response, which is important because mucosal surfaces are a more common route of entry for many pathogens than is the skin. Much of the research in oral DNA vaccine delivery has focused live bacterial vectors (LBVs) derived from attenuated bacterial pathogens Salmonella and Shigella, enteric species that are able to replicate the high copy-number DNA vaccine plasmids normally produced in E. coli. These attenuated pathogens carry mutations of biosynthetic or invasive genes that eliminate their pathogenicity and ability to persist in host tissues or the environment.

Figure 2: Oral recombinant vaccine delivery using live bacterial vectors. (a) The live bacterial vectors (LBVs) are ingested and travel into the small intestine, where they invade the lining of the ileum through M cells and enter Peyer’s patches. (b) The LBVs invade antigen presenting cells and are phagocytosed. Shigella can escape the phagosome and enter the cytoplasm, while Salmonella modifies the phagosome and persists until it is digested by fusion with a lysosome. MHC is major histocompatibility complex.

LBVs that have been ingested travel from the stomach into the small intestine. Here, Salmonella and Shigella invade microfold (M) cells and enter lymphatic nodules called Peyer's patches. (see Figure 2). Once inside the Peyer's patches, the bacteria become targets for (or actively enter) macrophages, where they are internalized in membrane-bound phagosomes. At this point, one of two strategies is employed: Shigella escapes from the phagosome into the cytoplasm, while Salmonella alters the composition of the phagosome to survive and replicate (14). However, the precise mechanism of DNA vaccine delivery using bacterial vectors is not understood. For Shigella, the cells are thought to eventually lyse in the cytoplasm, releasing the DNA vaccine plasmids which are then transported to the nucleus by the host cell. Attenuating mutations that increase the rate of cell lysis have been shown to improve DNA vaccine delivery using Shigellaflexneri. For phagosome-bound Salmonella, the potential mechanism is less obvious, but could be due to a bystander effect in which DNA released from host cells that have undergone apoptosis due to Salmonella invasion is carried to APCs in membrane blebs (14). As with other DNA delivery routes, once the plasmid enters the nucleus, the antigen gene is expressed, and the protein processed for MHC presentation. Another advantage of bacterial vectors is potent immunostimulatory properties, due mainly to the lipid A (endotoxin) in the cell membranes, which eliminates the need for an adjuvant. The direct targeting of the mucosal immune system means that vector-directed immune responses are not likely to inhibit the repeated use of live bacterial vectors.

Drug development suffers from the high attrition rate in translating promising preclinical results in animal models into success in human clinical trials. One variable in DNA vaccine development is the relative dose between preclinical and clinical subjects. For example, a DNA vaccination by conventional needle injection of a 20 g mouse typically requires 100 µg of DNA to generate a protective immune response. To scale this to a 70-kg human would involve injecting 350 mg, which is prohibitive in terms of both the volume required and the associated cost (15). A typical vaccine dose used in human clinical trials is 2–4 mg DNA, which represents only 0.5–1% of the scaled dose. The extent to which the lack of scalability has affected human DNA vaccine development is debatable, as the first vaccine approved was effective in an even larger animal: the horse. But in contrast, Salmonella-based approaches use the same dose in terms of bacteria and plasmid DNA in humans as in mice. This similarity allows the murine dose to be representative, because S. enterica serovar Typhimurium in mice mimics the invasion and pathogenicity of S. enterica serovar Typhi in humans.

As with Shigella, Listeria monocytogenes has the ability to invade M cells and escape from the phagosome, and has been used to demonstrate DNA vaccine delivery, although it cannot replicate the high copy-number plasmids that are universally used as DNA vaccines. E. coli strains have also been used as DNA vaccine vectors, having been genetically modified to express the invasion and phagosomal escape proteins from its pathogenic relatives.

Public acceptance of Salmonella and other pathogen-based vectors is important for this to be a viable strategy, so it is encouraging that a current licensed typhoid vaccine is a live attenuated S. enterica serovar Typhi marketed as Vivotif (Crucell, The Netherlands), which has been safely administered to more than 200 million patients over 25 years (16). Other attenuated enteric bacteria have also shown good safety profiles in recent clinical trials, and the increasing use of probiotics has spread the concept of ingesting beneficial bacteria. However, several issues must be overcome before live bacterial vectors can be commercialized. The most convenient oral delivery for adults would use a capsule containing lyophilized bacteria. An enteric coating protects against the stomach acid, but lyophilized bacteria are susceptible to bile when released into the small intestine. To mitigate this susceptibility, Cambridge University in collaboration with Cobra Biologics developed a bile-adsorbing resin formulation that allows water to penetrate, but adsorbs the bile acids (17). This formulation enables the bacteria to rehydrate for the few crucial minutes that are required to increase their resistance to the effects of bile, before their release as the capsule dissolves (see Figure 3).

Figure 3: Bile-adsorbing resin for oral capsular delivery of lyophilized enteric bacteria.

Another key issue to address is plasmid instability in bacterial cells. The antibiotic-resistance genes normally used to select and maintain plasmid-containing cells are strongly discouraged by regulatory authorities for reasons of biosafety when used in LBVs. FDA requires a valid justification and proof that these genes cannot be transferred to commensal bacteria (18). As there is arguably no longer a credible justification for their use, antibiotic resistance genes are not used by companies developing LBV strategies. Even if they were permitted, they contribute to a significant metabolic burden on the bacterial cell due to their constitutive expression and the high plasmid copy-number—the major factor in plasmid loss in the first place—and it is not acceptable practice to dose a vaccinee with antibiotics. Therefore, antibiotic-free plasmid selection and maintenance systems are essential for DNA vaccine delivery using LBVs, but most systems retain the metabolic burden of an alternative expressed selectable marker gene.

To address this problem, Cobra Biologics applied its operator–repressor titration (ORT) technology to Salmonella as ORT–VAC. ORT requires an essential bacterial chromosomal gene to be controlled by a promoter such as lac or tet. The cell is unable to grow in the absence of the chemical inducer because the repressor protein binds to the chromosomal operator and prevents gene expression. However, when the cell is transformed with a plasmid that also possesses the operator (a short, nonexpressed sequence), multiple copies of the plasmid titrate the repressor and enable expression of the essential gene and cell growth (see Figure 4).

Figure 4: Operator–repressor titration for stable maintenance of selectable marker gene-free plasmids in live bacterial vectors. (a) The natural promoter of an essential gene is replaced with an inducible promoter and repressor gene, such that a repressor protein binds to an operator sequence and prevents essential gene expression and cell growth. (b) When the cell is transformed with a multicopy plasmid that also contains the operator sequence, the repressor protein is titrated by binding to the operator, thus enabling essential gene expression and cell growth.

ORT–VAC has recently been applied for successful immunization using a tuberculosis DNA vaccine expressing the antigen mpt64 from Mycobacterium tuberculosis, generating greater T-cell responses than the injected DNA vaccine in mice (19). This technique reduced the pulmonary count of infecting M. tuberculosis following a challenge. The positive control was the standard murine dose of 100 µg of intramuscularly-injected DNA, but ORT–VAC achieved better results despite delivering only 0.01–0.1% of this plasmid dose in 107 –108 orally administered bacteria.


Only two DNA vaccines (i.e., Apex–IHN and Oncept) and one DNA therapy (i.e., LifeTide) are currently on the market, all for veterinary applications. But despite a slow start, the DNA vaccine market is growing steadily, and the approval of a human DNA vaccine in the next few years would be a significant shot in the arm for the sector. There is no fundamental reason why DNA vaccines will not work in humans, and the success of the Oncept canine melanoma vaccine is significant, because it is one of only two licensed immunotherapeutic cancer vaccines. Overall sales in the DNA vaccine market totaled $141 million in 2008, and with a compound annual growth rate (CAGR) of 69.5%, are forecast to increase to $2.7 billion by 2014 (20). The value of human DNA vaccines was only $12 million in 2008, but it doubled a year later and is predicted to reach $2.3 billion by 2014 (149.6% CAGR) (20). In 2008 there were 95 open clinical trials involving plasmid DNA; the majority of these were for anticancer applications (see Figure 5) (13).

Figure 5: The number of open Phase I–III clinical trials involving plasmid DNA in 2008 (13).

For all recombinant vaccines, the selection of an antigen that is able to stimulate a protective, long-lasting immune response is the key to success. In addition to the antigen choice, effective delivery technologies are essential for realizing the promise of DNA vaccines, because there is a limit to what can be achieved through engineering of plasmids and adjuvant formulation. The fact that two out of the three plasmid-based medicines on the market use relatively novel approaches (i.e., electroporation and high-pressure transdermal delivery) demonstrates the capacity for innovation in this field. Effective oral delivery would represent a significant advance. The cost of DNA vaccines would be greatly reduced by a simple manufacturing process and no requirement for a viral boost or specialized delivery device.

ROCKY CRANENBURGH, PHD, is head of molecular biology at Cobra Biologics, Stafforshire UK, rocky.cranenburgh@cobrabio.com.


1. M.A. Lui, Immunol. Rev. 239, 62–84 (2010).

2. "West Nile-Innovator DNA", Fort Dodge Animal Health. E0257D 15M AP (Nov. 2008).

3. N.C. Simard, "The Road to Licensure of a DNA Vaccine for Atlantic Salmon", www.bion.no/filarkiv/2010/07/vaccine_simard.pdf, accessed Aug. 2011.

4. S. Vasan et al., PLoS One 6 (5), e19252 (2011).

5. K.E. Broderick et al., Hum. Vaccines 7, 22–28 (2011).

6. T. Kardos. "Contactless Magnetic Electroporation" presented at the DNA Vaccines Conference (San Diego, CA, 2011).

7. M.R. Prausnitz et al., Microbiol. Immunol. 333, 369–393 (2009).

8. J.W. Hooper et al., Vaccine 25, 1814–1823 (2007).

9. A.K. Roos et al., Mol. Ther. 13 (2), 320–327 (2006).

10. L.A. Hirao et al., J. Infect. Dis. 203 (1), 95–102 (2011).

11. D.J. Laddy et al., J. Virol. 83 (9), 4624–4630 (2009).

12. P.T. Loudon et al., PLoS One 5 (6), PLoS ONE doi:10.1371/journal.pone.0011021, Jun. 8, 2010.

13. M.A. Kutzler and D.B. Weiner, Nature Rev. Genetics 9, 776–788 (2008).

14. M. Shata et al., Mol. Med. Today 6, 66–71 (2000).

15. P.M. Smooker et al., Biotechnol Annual Rev. 10, 189–236 (2004).

16. G. Dietrich, A. Collioud, and S.A. Rothen. Biopharm Intl. 21 (10) s6–s14 (2008).

17. A. Edwards and N.K.H. Slater, Vaccine 27, 3897–3903 (2009).

18. FDA, Early Clinical Trials with Live Biotherapeutic Products: Chemistry, Manufacturing, and Control Information. (Rockville, MD, Sept. 2010).

19. J.M. Huang et al., Vaccine 28, 7523–7528 (2010).

20. J. Bergin, "DNA Vaccines: Technologies and Global Markets," BCC Research (2009)