Mucosal Vaccines: Insights from Different Fields

Researchers are joining forces to understand mucosal immunity and develop mucosal vaccines

By Andreas von Bubnoff

Even though HIV transmission most often occurs at mucosal surfaces, there is still much that is unknown about the role of mucosal immunity in blocking the virus (see The great barrierIAVI Report, March-April 2008). But a recent meeting on Modern Mucosal Vaccines, Adjuvants & Microbicides, held in Porto, Portugal from October 22-24, showed that research is now underway that may help fill this knowledge gap. “I was hoping that we were able to rejuvenate this whole area,” said Pearay Ogra, a professor of pediatrics at the State University of New York, who was one of the organizers of the meeting, which attracted about 100 researchers. He said it succeeded in bringing together people from varied backgrounds including microbicide research, immunology, and vaccinology. Indeed, the meeting was quite unique in that sense, said Nicholas Mantis, a research scientist at the New York State Department of Health.

Accordingly, the research presented covered diverse topics, including the application of antiretrovirals and antibodies as microbicides, live vaginal bacteria microbicides, using plants to manufacture proteins for vaccines, and utilizing microparticles to enhance the induction of immune responses. The researchers also discussed broader topics like which type of immune response is the most relevant to measure in mucosal tissues and novel routes of administering mucosal vaccines.

Joining forces

Several speakers emphasized the importance of microbicide and vaccine researchers working together to develop approaches to prevent mucosal acquisition of HIV. The potential for synergy became evident in a presentation by Martin Cranage, a professor and chair of molecular vaccinology at St. George’s, University of London, who described research where an intra-rectally applied gel containing 1% of the antiretroviral tenofovir protected six out of nine Indian rhesus macaques against intra-rectal simian immunodeficiency virus (SIV) challenge (1). He also mentioned an ongoing project with Robin Shattock, also at St. George’s, in which the researchers are testing a gel-based version of gp140 protein that is applied vaginally in rabbits, nonhuman primates (NHPs), and humans.

In the tenofovir study, the macaques that remained uninfected and showed no systemic antibody response to SIV after being protected by the tenofovir gel nevertheless showed an SIV-specific T-cell response both locally in the gut, as well as systemically. This suggests that rectal exposure to the virus in the presence of tenofovir might have a similar effect to vaccination, Cranage said. “This is the potential bridge between microbicides and vaccines,” he said.

Laurel Lagenaur, a senior scientist at California-based Osel Inc. described the development of a live vaginal protein-based microbicide by introducing genes into Lactobacilli, bacteria that normally live in the vaginal mucosa. Deficiency in these bacteria is associated with increased acquisition of sexually transmitted infections, including HIV. She showed that it is possible to colonize NHP vaginas with a transgenic Lactobacillus strain expressing the protein cyanovirin, which binds HIV surface glycoproteins. In addition, the protein was actually expressed by the bacteria in the macaques. Next, the company plans to conduct an in vivo challenge study in NHPs.

Larry Zeitlin, president of San Diego-based Mapp Biopharmaceutical, also emphasized a synergy between vaccinology and microbicides. He described the use of a combination of monoclonal antibodies to herpes simplex virus (HSV) and CCR5, a cellular chemokine receptor used by HIV to enter its target cells, for a vaginal microbicide called mapp66. He said that mapp66 can neutralize both HSV and HIV in vitro. The company is planning Phase I trials with this microbicide later this year or in early 2009. “As we are developing a microbicide like this and learn what the protective dose is,” Zeitlin said, “that would provide some guidance for vaccinologists in terms of what neutralizing titers we need to target in the vagina.”

A serving of transgenic potatoes

Zeitlin’s talk also illustrated another theme at the conference: the potential of using plants in vaccine production. To produce the monoclonal antibodies for the mapp66 microbicide, his company uses tobacco plants. Zeitlin said the antibodies can be made in accordance with good manufacturing practice (GMP) standards, much faster and cheaper than when using mammalian cell culture. GMP is a set of standards required by regulatory agencies like the US Food and Drug Administration (FDA) for products that are tested in humans (see Cooking up candidates,IAVI Report, Jan.-Feb. 2008).

Yasmin Thanavala, a professor of immunology and oncology at the Roswell Park Cancer Institute in New York, presented research in which people who had previously been vaccinated with hepatitis B vaccine were given transgenic raw potatoes expressing hepatitis B surface antigen. At least half showed increased hepatitis B antibody titer after eating two or three doses of the transgenic potatoes. This suggests that oral delivery of antigens in minimally processed plant materials can provoke immune responses, Thanavala said. It also is an advantageous way to deliver vaccine in developing countries, since it eliminates injection and the cold chain (2). “Apparently it works, [although] there are some limitations because these are raw potatoes,” said Jiri Mestecky, a professor of microbiology and medicine at the University of Alabama at Birmingham, adding that other researchers are working with other plants like tomatoes, for example, which might be more appealing for consumption.

Considering tolerance

One concern with delivering antigens in food is that the immune system may develop tolerance to them. Oral vaccines therefore need to contain additional “danger signals” that can alert the immune system. These warning signals often come from adjuvants, according to Jan Holmgren, director of the Vaccine Research Institute at the University of Gothenborg in Sweden. In developing countries, oral vaccines—especially live attenuated viral and bacterial vaccines—are often less effective than in developed countries, Holmgren said, adding that the reasons are not well understood. Possible explanations include nutritional deficiencies, competing microflora for live vaccines, and perhaps also that people there tend to be exposed to so many antigens that oral vaccination can be like “spitting in the sea,” said Holmgren.

Mestecky pointed out that for a novel antigen, the induction of tolerance depends on whether the first vaccination is mucosal or systemic. His group found that oral or nasal immunizations of humans with a novel antigen called keyhole limpet hemocyanin resulted in diminished T-cell responses following subsequent systemic immunizations with that same antigen. This only occurred when the first immunization was delivered mucosally, not when systemic preceded oral immunization. He said this could be a concern for HIV vaccines, in that vaccinating mucosally first with a novel antigen might induce tolerance that could dampen the cellular immune response to HIV.

The route can make a difference

There are many different routes to deliver mucosal vaccines. Typically vaccines are delivered directly to mucosal surfaces to induce a mucosal immune response. More traditional routes include oral or nasal administration, but novel routes like transdermal or sublingual administration also show some promise. Typically, the strongest response is at the vaccinated mucosa, with the next best at adjacent mucosae, although the nasal and perhaps sublingual routes can also stimulate a genital mucosal immune response.

Increasingly, researchers are finding that the choice of route can make a difference in the immune response. For example, Charani Ranasinghe, a research fellow at the Australian National University, and her colleagues showed that in mice, nasal priming followed by intramuscular or nasal boosting elicits a higher avidity CD8+ cytotoxic T Lymphocyte (CTL) response than a systemic (intramuscular) prime-boost with pox vectors expressing HIV genes (3). In addition, the researchers found that the high avidity of CTLs generated by the mucosal immunizations correlates with lower expression of interleukin-4 (IL-4) and IL-13 cytokines by CD8+ CTL, with IL-13 being especially important, according to knockout studies in mice. This is the first study which demonstrates the importance of IL-13 for CTL avidity in vitro or in vivo, Ranasinghe said.

Susan Barnett, a senior director for viral vaccine research at Novartis Vaccines and Diagnostics, presented data from a Phase I trial that showed that intranasal priming using gp140 Env protein combined with systemic boosting can elicit HIV-specific Immunoglobulin A (IgA) and IgG in cervicovaginal secretions. In the trial, led by David Lewis of St. George’s, University of London, women were primed several times intranasally with the Env protein with or without the adjuvant LTK63, a nontoxic mutant of Escherichia coli enterotoxin, and then boosted twice intramuscularly with the Env protein and a different adjuvant called MF59. “I don’t think anyone has gone into women and elicited a vaginal IgA response with an envelope based vaccine,” said Barnett, who was also part of a recent study that showed that intramuscular immunization alone or combined with intranasal immunization can protect rhesus macaques against a vaginal SIV/HIV hybrid virus, known as SHIV, challenge (4). “We can elicit vaginal mucosal responses with intranasal priming, that’s the take home [message].”

Open up and say aaaah...

Mucosal immune responses can also be induced by using a relatively novel route: sublingual immunization, in which liquid drops are applied under the tongue, according to Cecil Czerkinsky, deputy director of the International Vaccine Institute in Seoul. He showed that in mice sublingual immunization with ovalbumin plus cholera toxin adjuvant induced ovalbumin-specific mucosal antibody and CTL responses in the lung (5) and in the female reproductive tract. Sublingual immunization with live or inactivated influenza vaccine induced systemic and mucosal IgA and IgG antibodies and CTL responses and protected against lethal influenza challenge in mice (6). He said the sublingual area is good for immunizations because it is not keratinized, which makes it more permeable. It also contains dendritic cells that are similar to the skin’s Langerhans cells, and in mice, sublingually administered antigen does not go into the olfactory bulb epithelium, suggesting it is not neurotoxic. He also showed preliminary results of a human study that suggests that sublingual administration of recombinant cholera toxin B subunit is safe.

From potatoes to particles

Not only the route of administration, but also the formulation of a mucosal vaccine can make a difference in the immune response it induces. One novel formulation currently under investigation uses particles coated with antigenic proteins. Maarten van Roosmalen, a senior scientist at the Dutch company Mucosis, described a particle called GEM that is made by hot acid treatment of Lactococcus lactis, a type of bacteria used to produce some types of cheese. This removes the proteins, lipids, and DNA, and the particles are then used to carry antigens. The antigen-covered particles are much more efficiently taken up by antigen-presenting cells than antigen alone, and peptidoglycans in the particle stimulate the innate immune system, resulting in a more robust and stronger immune response, according to van Roosmalen. In mice, intranasal vaccination with GEM-based streptococcus and influenza vaccines induced IgA secretion in the nasopharynx, lung, and vagina, and protected from lethal challenge with these pathogens.

In general, particle-based vaccines may be more immunogenic than soluble antigens, said Ed Lavelle, a lecturer in the school of biochemistry and immunology at Trinity College. He used a mimetic of a lectin called UEA-1 that binds to intestinal microfold (M) cells in mice. M cells overlie Peyer’s patches in the intestine, sample proteins and antigens from the luminal side of the gastrointestinal tract, and present them to underlying immune cells. Lavelle found that in mice, association of ovalbumin and UEA-1 to polystyrene microparticles enhanced cellular immune responses to intranasal and oral delivery, compared with delivering these two substances in soluble form. “Just the fact that you put an antigen on a bead is enough to stimulate a decent immune response,” Mantis said.  

IgA: Not the Whole Story 

One important question is which mucosal antibody responses are the most critical. Antibody measurements in mucosal tissues typically focus on secretory Immunoglobulin A (IgA), but several speakers at the meeting cautioned that a potential role for other antibody types in protection should not be neglected. Lou Bourgeois, a scientific officer at the Program for Appropriate Technology in Health (PATH), said that secretory IgA is not always essential for mucosal immunity—for example, most people with IgA deficiency do not have an increased susceptibility to infections. He said researchers should also look at the protective effect of IgG at mucosal surfaces. Jan Holmgren of the Vaccine Research Institute in Sweden said that in IgA-deficient people, IgG and especially IgM responses might compensate and that at some mucosal surfaces, such as the lungs and the vagina, systemic antibodies may actually leak through—or get transported—to the mucosal surface.

Another issue is how to correctly measure mucosal immune responses. Some researchers use absorbent sponges called Weck-Cel to obtain vaginal and rectal secretions to measure antibody concentrations. But Jiri Mestecky of the University of Alabama at Birmingham pointed out that such measurements might be misleading because the sponges could damage the mucosal epithelium and therefore take up systemic antibodies that leak into mucosal surfaces. In mucosal tissues like the female genital tract it could then be hard to distinguish circulating IgG from IgG that has been produced locally, he added.

1.PLoS Med. 5, e157, 2008
2. Proc. Natl. Acad. Sci. 102, 3378, 2005
3. J. Immunol. 178, 2370, 2007
4. AIDS 22, 339, 2008
5. Vaccine 25, 8598, 2007
6. Proc. Natl. Acad. Sci. 105, 1644, 2008