Safety and Colonization of Two Novel virG(icsA)-based Live Shigella sonnei Vaccine Strains in Rhesus Macaques (Macaca mulatta)

Shigella are gram-negative bacterium that cause bacillary dysentery (shigellosis). Symptoms include diarrhea and discharge of bloody mucoid stools, accompanied by severe abdominal pain, nausea, vomiting, malaise, and fever. Persons traveling to regions with poor sanitation and crowded conditions become particularly susceptible to shigellosis. Currently a vaccine for Shigella has not been licensed in the United States, and the organism quickly becomes resistant to medications. During the past 10 y, several live attenuated oral Shigella vaccines, including the strain WRSS1, have been tested in humans with considerable success. These Phase I vaccines lack the gene for the protein VirG also known as IcsA, which enables the organism to disseminate in the host target tissue. However, 5% to 20% of the vaccinated volunteers developed mild fever and brief diarrhea, and the removal of additional virulence-associated genes from the vaccine strain may reduce or eliminate these side effects. We administered 2 Shigella sonnei vaccines, WRSs2 and WRSs3, along with WRSS1 to compare their rates of colonization and clinical safety in groups of 5 rhesus macaques. The primate model provides the most physiologically relevant animal system to test the validity and efficacy of vaccine candidates. In this pilot study using a gastrointestinal model of infection, the vaccine candidates WRSs2 and WRSs3, which have additional deletions in the enterotoxin and LPS modification genes, provided better safety and comparable immunogenicity to those of WRSS1.Abbreviations: IcsA, intercellular spread protein A (VirG protein); Ipa, invasion plasmid antigen; LPS, lipopolysaccharideVarious species of Shigella cause bacillary dysentery, also known as shigellosis, in humans and other primates.^{1,18,31,37,49} Symptoms include diarrhea, with various degrees of mucus and hematochezia, accompanied by severe abdominal pain, nausea, vomiting, malaise, and fever. Shigella is a gram-negative bacterium whose genomic sequence is very similar to that of Escherichia coli and is phylogenetically considered a pathotype of E. coli.²³ Shigella infections are spread by the fecal–oral route through the consumption of contaminated food and water or by mechanical vectors such as insects. Persons traveling in areas with poor sanitation and crowded conditions become particularly susceptible to such diseases.^2,39,48,56 In colonies of research macaques, Shigella infections are spread by the fecal–oral route, originating from addition of animals that are asymptomatic carriers, either from the wild or from other colonies.^1,55 Currently no Shigella vaccine has been licensed in the United States, although several are under development.^{25,32,35,38,54} Antibiotics are used as treatment therapy, but many Shigella isolates are multidrug-resistant, increasing the need for a preventive vaccine for travelers, military personnel, and children in endemic areas.^{16,34,41,44,57} Rhesus macaques represent an excellent model for studying Shigella because primates are the only known animal model that simulate natural human infection, including dysentery after oral challenge.^{11-13,17,28,40,47,46} Other animal models, such as the mouse pulmonary model and the guineapig ocular and intranasal models, mimic specific, isolated steps in Shigella pathogenesis.^14,19,26There are 4 major serogroups of Shigella and one or more serotypes within each serogroup. This classification is based on antigenic differences in the O-antigen polysaccharide component of the outer membrane-associated lipopolysaccharide (LPS). Estimates from the Centers for Disease Control (Atlanta, GA) indicate that approximately 400,000 cases of shigellosis occur in the US annually, with an estimated 165 million cases worldwide each year.²² These occur predominantly in developing countries, where the population most affected is children younger than 5 y.²²Various in vitro cell culture models of infection, as well as studies in animal models including gastrointestinal infection in nonhuman primates, have contributed to the current understanding of Shigella pathogenesis.^4,42 Shigella organisms target the distal region of the colon and rectum, where the bacteria are captured by specialized M-cells located within the follicle-associated epithelium. The M-cells deliver bacterial antigens, which include bacterial LPS and invasion plasmid antigen (Ipa) proteins, to the underlying antigen-presenting macrophages and dendritic cells.³⁶ Virulent Shigella strains escape macrophages by a cytotoxic effect, causing release of the bacteria and inflammatory cytokines such as IL1β, TNFα, and IL18. The bacteria then invade adjacent intestinal epithelial cells at the basolateral side, through interaction between bacterial proteins and multiple signaling molecules within the host cell.^36,42Shigella are capable of orchestrating this uptake into nonphagocytic epithelial cells, a process termed invasion, through the secretion of proteins that are highly conserved among all virulent strains. These proteins are encoded by a large, 213-kb plasmid, referred to as the invasion plasmid or the virulence plasmid.^50-52 Each bacterium ultimately is engulfed within an endocytic vacuole. Subsequent lysis of the vacuole sets the bacterium free within the epithelial cell cytoplasm, where it replicates and moves to adjacent epithelial cells, with the help of the intercellular spread (Ics) A protein, also known as VirG.^3,24,45 The VirG(IcsA)-assisted intra- and intercellular spread of the bacteria within the epithelial tissue contributes significantly to the loss of epithelial cell integrity and accompanying tissue injury. Shigella strains with loss of the virG(icsA) gene are significantly attenuated in all animal models of virulence.⁵ Several Shigella vaccine candidates that have undergone Phase 1 clinical trials are principally attenuated due to lack of the VirG(IcsA) protein, these include S. flexneri 2a strain SC602, S. sonnei strain WRSS1, and S. dysenteriae 1 vaccine strain WRSd1.^{5,10,15,21,53}The S. sonnei first-generation vaccine candidate WRSS1 has been tested in several Phase I inpatient and outpatient trials.^15,21,33 WRSS1, like the previous S. flexneri 2a vaccine candidate, SC602, was shown to be safe in Phase I trials when orally administered at a dose of 10³ to 10⁴ CFU.^15,21,33 Both SC602 and WRSS1 colonized well as evidenced by fecal excretion of the vaccine strain for 5 to 7 d. The vigorous immune response seen with both these vaccine candidates was correlated with the robust colonization. However, 5% to 20% of the SC602- and WRSS1-immunized volunteers showed mild and transient symptoms of diarrhea and fever. In light of new knowledge about Shigella pathogenesis as well as data from other clinical trials, 2 new and presumably safer S. sonnei derivatives have been constructed (WRSs2 and WRSs3). In addition to loss of virG(icsA), WRSs2 has deletions in 2 genes, senA and senB, that are present on the virulence plasmid.^29,52 senA (also known as shet2-1 and ospD3) encodes the enterotoxin ShET2-1, which has been shown to cause fluid accumulation in rabbit ileal loops.^7,8 senB (also known as shet2-2 and ospD2) encodes a similarly sized protein as SenA and shows 40% identity at the amino acid level. The enterotoxic activity of ShET2-2 remains to be demonstrated.^20,29 Like WRSs2, WRSs3 lacks virG(icsA) and the 2 enterotoxin genes with the additional deletion of the virulence plasmid-based msbB2 gene. Lack of the msbB2 gene product has previously been shown to produce a less-toxic LPS molecule, which might help to reduce the symptoms of fever seen in WRSS1-immunized volunteers.^20,29 How these additional mutations, especially that of msbB2, will affect colonization of these strains is unclear, given that effective colonization is critical in the elicitation of immune response and protection. The primary goal of this pilot study was to compare the safety and colonization of 2 novel S. sonnei vaccine candidate strains (WRSs2 and WRSs3), with those of the clinically tested vaccine strain WRSS1, by using a gastrointestinal model of infection. The hypothesis tested was that the 2 novel Shigella vaccine strains would colonize the vaccinated animal and induce an immune response similar to that of WRSS1 but with less frequent and less severe clinical side effects.