Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Vaccines Against Diarrhea Caused by Noncholera Bacteria
Diarrheal infections are estimated to cause 1.7 billion disease episodes and 700,000 deaths globally each year, mainly in children in low- and middle-income countries. In addition to these direct consequences, infectious diarrhea is a major cause of malnutrition leading to stunting and also, even independently of malnutrition, to significant intellectual and other cognitive disturbances. Travelers to endemic areas are also at high risk for being infected by enteric pathogens. However, although the majority of infectious diarrheas and their negative consequences for human health would be preventable by effective vaccination, there are as yet only licensed vaccines against three enteric pathogens, Salmonella enterica serovar Typhi ( S . Typhi), Vibrio cholerae , and rotavirus. Available and future vaccines against these pathogens are described in other chapters of this book and will not be discussed here. However, intense efforts are under way to develop vaccines against additional important agents, in particular enterotoxigenic Escherichia coli (ETEC) and Shigella . This chapter focuses on vaccine development efforts against ETEC and Shigella , and briefly discusses the development of vaccines against Campylobacter , Shiga toxin–producing E. coli (STEC), and nontyphoidal Salmonella (NTS). While ETEC and Shigella have exclusively human reservoirs, Campylobacter , STEC, and NTS are also zoonotic pathogens with known animal reservoirs (chicken, cattle, etc.). Our discussion is limited to vaccines for human use, but it should be noted that there are ongoing efforts to also develop veterinary vaccines which could indirectly reduce the spread of pathogens to human populations.
MECHANISMS OF DISEASE AND IMMUNITY
Enteric pathogens differ in the way they cause infection and disease. This influences what type of immune responses they elicit and how vaccination may protect against infection and disease. Pathogenic pathways for bacterial enteropathogens include, for example: (a) colonization of the intestinal mucosa without invasion or morphological damage but with induction of diarrhea by means of powerful secreted enterotoxins (e.g., V. cholerae and ETEC); (b) adhesion to the mucosa and induction of enterocyte effacement by injection of bacterial proteins into the epithelial cells (e.g., enteropathogenic E. coli ); (c) induction of enterocyte killing by producing powerful exotoxins that block cellular protein synthesis (e.g., enterohemorrhagic E. coli [EHEC/STEC]); (d) local invasion and destruction of the mucosa (e.g., Shigella ); (e) local invasion of the mucosa and induction of inflammation (e.g., NTS); and (f) translocation across the mucosa, entry into the bloodstream and spread to distal organs (e.g., S. typhi and S. paratyphi A and B).
Secretory immunoglobulin A ( SIgA) produced by plasma cells in the intestinal lamina propria provides an important first line of defense against enteric infections. Dimeric immunoglobulin (Ig) A is actively transported through the epithelial barrier into the lumen, where it acquires secretory component (SC), and is released in the form of SIgA. SIgA is resistant to degradation and can neutralize bacterial enterotoxins (and viruses) and inhibit bacterial epithelial attachment, thereby preventing bacterial colonization and cell invasion. Low levels of IgG may also be present in gastrointestinal secretions. However, the small intestine (at least in individuals from industrialized country settings) is essentially nonpermeable to blood proteins, as long as it is uninflamed. This is in contrast to the lower respiratory tract and the female genital tract, which are both relatively permeable to transudated IgG antibodies. Consistent with this observation, the noninflammatory infections of the small intestine, such as cholera and ETEC, represent examples where vaccine protection is mediated by locally produced SIgA antibodies and probably requires at least primary vaccination via the oral route. Parenteral immunization may induce a transient SIgA response, but only in individuals who have been previously primed by natural exposure or oral immunization (see also Clemens et al., chapter on cholera vaccines in this volume). However, many young children in developing countries suffer from environmental enteropathy, a syndrome that includes a chronic state of inflammation of the proximal small intestine, small bowel bacterial overgrowth and excitation of the innate immune system associated with inflammation. In such children, circulating IgG antibodies may transudate to the mucosal surface of the gut and possibly contribute to protection. However, IgG is rapidly degraded by intestinal proteases which, together with the normally low levels and lack of function of complement in intestinal secretions, limits the protective significance of IgG in the gut. Long-lasting immunological memory, mainly through the generation of long-lived memory B cells in intestinal Peyer patches and lymphoid follicles in response to infection or oral immunization, can provide persistent protection long after the acute SIgA antibody response has dissipated. The memory B cells can rapidly evolve into IgA-secreting plasma cells upon re-exposure to antigen and elicit a rapid anamnestic SIgA antibody response which may control the infection before it causes disease. Mucosal memory B cells can be very long lived; for example, functional antigen-specific B-cell memory has been demonstrated for more than 10 years after primary oral cholera vaccination.
Cell-mediated protection against those pathogens that enter the host via mucosal surfaces can be mediated by intraepithelial and lamina propria CD4 + helper and CD8 + cytotoxic T lymphocytes. These cell-mediated responses are believed to contribute to the durable protection that has been observed in subjects immunized with Ty21a live oral typhoid vaccine. Numerous T-helper cells producing interferon-γ (Th1) or interleukin (IL)-17A (Th17) are often found in the lamina propria throughout the gastrointestinal tract. Although believed to be major effector T cells against Helicobacter pylori and intracellular organisms such as S. typhi and S . paratyphi , their protective mode of action in the gut mucosa is not fully understood. There are also data suggesting an important role for IL-17A in promoting intestinal IgA production and secretion. , T follicular helper (Tfh) cells support antibody production and affinity maturation to protein antigens in mucosal germinal centers, thereby having an important indirect role in forming protective antibody responses in the gut. Quantitation of circulating Tfh cells assessed within a week after primary vaccinations with an oral inactivated ETEC vaccine correlated with memory intestine-derived vaccine specific IgA responses 1–2 years later, suggesting that activated Tfh cells are mobilized into blood after oral vaccination and may indeed be used as biomarkers of vaccine specific mucosal memory in humans.
MUCOSAL VERSUS PARENTERAL VACCINATION
Traditional parenterally injected vaccines primarily induce systemic immune responses and are generally poor in inducing mucosal immunity. In contrast, mucosal vaccines, administered via oral, nasal, or sublingual routes, generally give rise to both mucosal and systemic immune responses. The mucosal response is anatomically compartmentalized reflecting the migratory properties of lymphocytes activated at different mucosal inductive sites, thus imposing distinct constraints on the choice of mucosal route for vaccine administration. In general, the strongest immune response is obtained at the site of vaccine application and next after that in anatomically adjacent or evolutionarily linked sites. An example of the latter is the intestine–mammary gland link in lactating women which ensures that the breastfeeding baby receives epidemiologically relevant breast milk SIgA antibodies from the mother resulting from and reflecting the current microbial exposure in the mother’s intestine.
While vaccination via the oral route has the potential to prevent clinical illness by all enteric pathogens, parenteral vaccination can be used as an alternative route for inducing protection against those enteric infections in which the pathogen is translocated across the intestinal epithelium before causing disease. An example is shigellosis (bacillary dysentery) in which translocated Shigella infect enterocytes from the basolateral side. At this point the organisms can readily be attacked by serum-derived antibodies in concert with complement and phagocytic neutrophils. Likewise, serum antibodies can effectively attack pathogens that cause disease by inducing inflammation in the submucosal lymphoid tissues at the site of infection (e.g., NTS serovars and Campylobacter jejuni ) or after entering the bloodstream (such as S. typhi or S. paratyphi ).
Previous exposure to the pathogen leading to mucosal immunological priming also increases the ability of parenteral vaccination to protect against mucosal infections. As mentioned above, a general rule seems to be that parenteral vaccines may provide some degree of boosting in already primed individuals, but they are unable to induce an effective mucosal response in individuals who have not been primed by previous mucosal exposure to the pathogen or by oral vaccination. As described below, ongoing efforts to develop vaccines against ETEC are mainly focused on oral administration, while both oral and parenteral vaccination is considered for Shigella, Campylobacter , STEC, and NTS.
SPECIFIC VACCINES
Enterotoxigenic Escherichia coli
Epidemiology
ETEC is among the most common bacterial causes of diarrhea-associated morbidity and mortality in children younger than age 5 years in developing countries. The peak incidence occurs in children 6–36 months of age, where ETEC together with rotavirus and Shigella are the most common pathogens. ETEC is also a major cause of the malnutrition and various cognitive disturbances noted as a sequalae of infectious diarrhea. ETEC is also the number-one cause of travelers’ diarrhea, reported to account for 20–60% of all travelers’ diarrhea globally with at least 20% of affected travelers being bedridden for part of their trip and 40% changing their itinerary because of diarrhea ( reviewed in reference ). Travelers’ diarrhea, primarily caused by ETEC, is also the most common medical problem for military personnel from industrialized countries deployed in less-developed areas of Asia or Africa, with an average incidence of 29% per month and in hyperendemic regions up to 60% per month.
Clinical Manifestations and Mechanisms of Disease and Immunity
Disease caused by ETEC is characterized by profuse watery diarrhea lasting for several days. Infection is usually self-limited but may lead to dehydration and malnutrition in young children. ETEC causes disease after ingestion of contaminated food or water by attachment to the small bowel mucosa by so-called colonization factor (CF) antigens, mostly fimbriae, followed by bacterial multiplication and production of heat-labile toxin (LT) and/or a heat-stabile toxin (ST). , Immune protection against ETEC infection and disease is believed to be mediated mainly, if not exclusively, by intestinal-mucosal SIgA antibody responses against CF antigens and LT inhibiting bacterial epithelial attachment and colonization and toxin binding, respectively. These main virulence factors and immune mechanisms are illustrated in Fig. 20.1 . ,
More than 25 types of CF antigens have been identified on human ETEC isolates, which has guided vaccine development efforts, with a few types occurring most frequently, including CFA/I and “coli surface” antigens CS1 through CS6. The CS6 antigen is an outer membrane antigen rather than a fimbrial protein, which may occur alone or together with either CS4 or CS5 fimbrial antigens on clinical isolates of ETEC. Importantly, immune responses against one CF type are in general not cross-protective against other types. Therefore, a broadly effective vaccine would require inclusion of multiple CF antigens. Notwithstanding this, a few vaccine-relevant immunological cross-reactions have been identified between certain CF antigens which can broaden the protective efficacy of vaccines containing representatives of such antigens. , Thus, the currently clinically most advanced ETEC vaccine ETVAX, described more fully below, has been shown by its CFA/I antigen to induce an antibody response against not only CFA/I positive ETEC, but also cross-reacts against ETEC bacteria expressing any of the “CFA/I family” antigens CS1, CS14 or CS17, and by its CS5 antigen inducing antibodies that cross-react with CS7-expressing ETEC.
Once attached to the intestinal epithelium, ETEC produces LT and/or ST enterotoxins, which induce the characteristic watery diarrhea. LT, which is highly homologous to cholera toxin (CT), is composed of a single toxic-active A subunit (LTA) and five identical B subunits, comprising the cell-binding domain (LTB). After binding via the LTB domain to receptors on intestinal cells, mainly, like cholera toxin, to GM1 ganglioside receptors but also to structurally related glycoproteins, the A subunit induces activation of adenylate cyclase, leading to increased production of cyclic adenosine monophosphate (cAMP), which affects ion channels and causes massive fluid secretion. Short-term protection against ETEC disease has been documented in persons immunized with a CTB subunit containing cholera vaccine because of immunologic cross-reactivity between B subunits of CT and LT. ST is a small peptide toxin (18–19 amino acids) that stimulates guanylate cyclase in target cells and elicits cyclic GMP mediated intestinal fluid secretion and diarrhea. Efforts to use ST as a vaccine antigen have been problematic because of the low immunogenicity of this small toxin and other considerations discussed later.
Several lines of evidence support the feasibility of an efficacious ETEC vaccine. In animal models, antibodies to LT and CF antigens have been shown to cooperate synergistically to provide protection. , Intestinal antibodies protected against challenge with ETEC strain H10407, that expresses CFA/I fimbriae, LT and ST, in 90% of volunteers who had received passive oral milk-derived immunoglobulin from cows hyperimmunized with several fimbriated ETEC strains and LT. Studies in a “Controlled Human Infection Model” (CHIM) have also demonstrated effective protection against ETEC reinfection with the same strain, and epidemiologic studies support that an initial ETEC infection protects against subsequent reinfection with ETEC expressing the same or cross-reactive CF antigens.
Enterotoxigenic Escherichia coli Vaccine Candidates
Although the diversity of ETEC, with more than 100 O serogroups and more than 25 recognized CF antigens, represents a major challenge for vaccine development, several ETEC vaccine candidates are under development. Most vaccines are intended for oral administration and based on toxin-derived antigens alone or combined with the most prevalent CF antigens, the latter being either purified or expressed on the bacterial surface. These vaccines are in different stages of preclinical and clinical development. Currently, there are five different vaccine candidates in the clinic or poised to begin clinical development (see Table 20.1 ). , One vaccine, the oral cholera vaccine Dukoral is licensed in several countries for prevention of ETEC travelers’ diarrhea based on its demonstrated 50–70% short-term cross-protective efficacy through its B subunit (CTB) component against LT-mediated ETEC diarrhea and a field trial in Bangladesh provided an impressive 86% efficacy against life-threatening diarrhea due to LT-positive ETEC. Among the ETEC specific vaccine candidates, the ETVAX/dmLT oral vaccine is the most advanced being in Phase IIb clinical testing. There are two other orally delivered live-attenuated combined Shigella-ETEC candidates: the ShigETEC is in Phase I trials and the CVD 1208S-122 prototype vaccine is poised to enter Phase I studies. The ETEC fimbrial tip adhesin (FTA) from the Naval Medical Research Center and PATH and the Multi-epitope fusion technology (MEFA) vaccine candidate from the University of Illinois and John Hopkins University are both parenteral candidates in early clinical and preclinical development, respectively.
TABLE 20.1
Enterotoxigenic Escherichia coli Candidate Vaccines Tested in Humans
| Vaccines | Development Stage | References |
|---|---|---|
| Killed Whole-Cell Oral Vaccines | ||
| ETVAX | ||
| Multivalent ETEC vaccine; Inactivated recombinant E. coli strains overexpressing CFA/I, CS3, CS5, and CS6 + LCTBA B subunit toxoid ± dmLT adjuvant | Phase I/II trial in Bangladesh Phase IIb trial in Finland/Benin |
|
| rCTB-CF vaccine (now replaced by ETVAX) | ||
| Inactivated ETEC strains expressing CFA/I, CS2, CS3, CS4, and CS5 + rCTB | Phase III trials in North American travelers to Mexico and Guatemala; Phase III trial in children in Egypt | |
| Live-Attenuated Oral Whole-Cell Vaccines | ||
| ACE527 | ||
| ETEC strains with deletions in aroC , ompC, and ompF expressing CFA/I, CS1, CS2, CS3, CS5, and CS6, and LTB ± dmLT | Phase IIb trial in the United States | |
| Subunit Oral Vaccines | ||
| Dukoral | ||
| Oral killed whole-cell + rCTB cholera vaccine (with demonstrated efficacy of the rCTB component against diarrheal disease caused by LT-positive ETEC) | Licensed Phase III trials in Bangladesh, in Finnish travelers to Morocco, and North American travelers to Mexico |
|
| dmLT | ||
| Double mutant LT (R192G/L211A) | Phase I trial in the United States | |
| CS6 | ||
| Free protein or in microspheres | Phase I trial in the United States | |
| Subunit Transcutaneous/Intradermal Vaccines | ||
| LT patch | Phase III trial in the United States | |
| CF Tip Proteins | ||
| CfaE + mLT (LTR192G) | Phase I trial in the United States |
CF, colonization factor; CS, coli surface antigen; dmLT, double mutant labile toxin; ETEC, enterotoxigenic Escherichia coli ; LCTBA, heat-labile toxin B subunit/cholera toxin B subunit hybrid protein; LT, labile toxin; mLT, single mutant LT; rCTB, recombinant cholera toxin subunit B.
Vaccines Containing Killed Whole-Cell Bacteria Plus B Subunit
Since it has proved to be difficult to develop effective immunity against the low-molecular-weight STs (see section on toxoid vaccine candidates below), the predominant strategy to protect against both LT-mediated and ST-mediated ETEC diarrhea is to develop an oral vaccine that by inducing mucosal immunity will inhibit intestinal attachment and colonization of ETEC, especially ST-producing ETEC, and will hinder bacterial production of enterotoxins and prevent their release directly onto the epithelial cells. For additional protection against LT-induced diarrhea, the vaccine should preferably also contain or express an LT “toxoid” component such as the LTB or CTB subunits; as mentioned above. The oral cholera vaccine Dukoral through its CTB component has demonstrated significant short-term protective efficacy against LT-mediated ETEC diarrhea. Based on this strategy, a first-generation oral ETEC vaccine consisting of a combination of formalin-inactivated ETEC bacteria expressing various prevalent CF antigens and recombinant CTB subunit (rCTB) was developed by researchers at the University of Gothenburg and extensively tested in Phase I and Phase II trials. , This vaccine, given orally with a buffer in a two- or three-dose regimen, was safe and induced intestinal mucosal immune responses in 70–90% of Swedish, Bangladeshi, and Egyptian vaccinees of different ages. However, when tested in two Phase III trials in North American travelers to Mexico and Guatemala and a Phase III study in 6- to 18-month-old children in Egypt, the vaccine showed only a modest protective efficacy against overall (usually mild) ETEC diarrhea. , (also L. Bourgeois, personal communication, 2012). Nevertheless, in both of the traveler studies in adults the vaccine reduced more severe ETEC diarrheal disease by almost 80%. Thus, vaccine protection against ETEC appears to be in line with observations for already licensed enteric vaccines against cholera and rotavirus disease, which have greater impact on moderate and severe illness than on milder disease. It is also in line with the finding that a CTB-containing oral cholera vaccine (the prototype of Dukoral) when tested in a large field trial in Bangladesh gave an overall 67% protection against diarrheal disease (mainly of moderate severity) caused by LT-producing ETEC and no less than 86% protection against life-threatening disease caused by such bacteria.
A second-generation, further improved oral killed ETEC vaccine (ETVAX) based on the same approach has then been developed. The vaccine composition has been enhanced by: (a) the inactivated vaccine strains contain 4- to 15-fold increased amounts of CFA/I, CS3, and CS5 antigens, and also contain the CS6 antigen in an immunogenic form (which was missing in the first-generation vaccine); (b) the rCTB is replaced by a hybrid rCTB/LTB subunit protein (LCTB A ) eliciting a stronger anti-LT immune response; and (c) the double-mutant LT protein (dmLT), serving both as mucosal adjuvant and extra LT antigen, is included in the vaccine. The dmLT molecule (R192G/L211A), which has two mutations in the enzymatic A subunit compared to the native LT toxin, is nontoxic, and has shown potent adjuvant activity when tested in combination with various vaccines in mice and in in vitro experiments with human cells. When the improved ETEC vaccine was given to mice, strong intestinal-mucosal IgA antibody responses were induced (as well as serum IgG and IgA responses) to each of the vaccine CFs as well as to LTB, and responses to the vaccine were further enhanced by the dmLT adjuvant. Based on these data, a Phase I study was undertaken in Swedish adult volunteers (30–35 subjects per group) given either the vaccine alone, the vaccine together with 10 or 25 µg of dmLT adjuvant, or placebo. Adverse reactions were few and mild with no differences between subjects receiving vaccine with or without dmLT or placebo. The majority of the vaccinated subjects (74% of all vaccine recipients and 83% of those receiving vaccine plus low-dose dmLT) showed significant mucosal IgA responses to all five of the primary vaccine antigens. The mucosal IgA antibodies were measured from cultured gut-derived blood lymphocytes (the antibody in lymphocyte supernatants method; ALS) and/or as specific SIgA antibodies in stool samples. Addition of low-dose dmLT to the vaccine significantly enhanced ALS responses to the CS6 antigen, which is the CF present in lowest amounts in the vaccine.
ETVAX safety was further documented in an age-descending trial of 475 subjects in Bangladesh. , Immunizations started with a full adult dose of vaccine plus dmLT in adults followed by administration of different amounts of vaccine and dmLT combinations in children 2–5 years, children 1–2 years, and finally infants aged 6–11 months. ETVAX was safe in all age groups, most solicited events were mild, with vomiting being the most common event in vaccine recipients. Vomiting was reduced by using fractional adult doses without loss of immunogenicity. The addition of dmLT did not change the safety profile. Mucosal IgA antibody responses in intestine-derived lymphocyte secretions (ALS) were detected against all primary vaccine antigens (CFA/I, CS3, CS5, CS6, and the LCTB A toxoid) in most adult participants and among the two older age groups of children. Responses were less frequent and of lower magnitude in infants aged 6–11 months than in older children. Fecal secretory IgA immune responses were recorded against at least three of the vaccine antigens in 78 (56%) of 139 infants aged 6–11 months who were vaccinated. Addition of the adjuvant dmLT enhanced the magnitude, breadth, and kinetics (based on number of responders after the first dose of vaccine) of immune responses in infants.
The safety, immunogenicity, and efficacy of ETVAX was recently assessed in a double-blind, randomized placebo-controlled, efficacy trial in Finnish travelers to Benin, West Africa. The trial demonstrated good immunogenicity with strong serum IgA and IgG responses to LTB and excellent safety. Preliminary results indicate significant efficacy against moderate-to-severe ETEC-attributable disease ( P = 0.006), as well as against severe diarrhea due to any infectious cause ( P = 0.03). Antibiotic or antisecretory drug treatment was given to significantly fewer vaccine responders than to placebo recipients ( P = 0.03), suggesting that ETVAX reduced disease severity. Phase II/III trials of ETVAX in adults and children are underway in Zambia and the Gambia.
Live-Attenuated Vaccines
Several observations established the feasibility of immunoprophylaxis against ETEC using live oral vaccines. In a challenge model of ETEC diarrhea in adult volunteers, an initial clinical diarrheal infection caused by a fimbriated LT/ST ETEC strain conferred approximately 80% protection against clinical illness upon subsequent homologous rechallenge. Immunization with a single 5 × 10 10 colony-forming unit (CFU) dose of an ETEC strain (serotype O6:H16 expressing CS1 and CS3) that had lost the genes encoding LT and ST stimulated significant intestinal SIgA anti-CF antibody responses and conferred 75% protection against challenge with a heterologous ETEC serotype (O139:H28) expressing CS1 and CS3 and LT and ST. More recently, the ACE527 live vaccine was developed containing three attenuated ETEC strains with deletions in aroC , ompC , and ompF , which collectively express CFA/I, CS1, CS2, CS3, CS5, and CS6, as well as LTB. In a Phase I trial, two doses of ACE527 (10 10 and 10 11 CFUs) were well tolerated, and the higher vaccine dosage induced a high frequency of gut-derived IgA ALS responses against LTB as well as CFA/I, CS6, and to a lesser extent CS3. Irrespective of dose, immune responses to other antigens were low and infrequent. These observations were extended in a Phase IIb study in subjects who received two 10 11 CFU doses and were then challenged with ETEC strain H10407 (O78, CFA/I, LT/ST). In this study, 19% of vaccinees, but none of the placebo recipients, vomited upon immunization, and other gastrointestinal symptoms also tended to be more common in the vaccine group. A nonsignificant 27% reduction was seen in moderate to severe diarrhea, the primary end point, although the vaccine decreased the intestinal colonization by the challenge strain. In a follow-up study, however, the ACE527 vaccine coadministered with dmLT adjuvant provided significant (66%) protection against ETEC challenge, while the nonadjuvanted vaccine was not protective. However, this candidate is not currently under active development.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here