Diarrhoeal Diseases (Updated February 2009)
Shigellosis is endemic throughout the world where it is held responsible for some 120 million cases of severe dysentery with blood and mucus in the stools, the overwhelming majority of which occur in developing countries and involve children less than five years of age   . About 1.1 million people were estimated to die from Shigella infection each year, with 60% of the deaths occurring in children under 5 years of age. More recent estimates fix the Shigella disease burden at 90 million episodes and 108 000 deaths per year [Lanata, personal communication]. In addition, about 500 000 cases of shigellosis are reported each year among military personnel and travellers from industrialized countries.
The disease is characterized by a short period of watery diarrhoea with intestinal cramps and general malaise, soon followed by permanent emission of bloody, mucoid, often mucopurulent stools. Outbreaks of dysentery due to S dysenteriae type 1 are frequent in poor populations living in crowded settings where hygiene is poor and sanitation non existent. Acute complications may occur that include peritonitis and septicaemia, especially in malnourished children, and the severe Haemolytic Uremic Syndrome (HUS) with renal failure. Since the late 1960s, pandemic waves of Shigella dysentery have hit sub-Saharan Africa, Central America and South and South-East Asia, often striking areas of political upheaval and natural disaster. During the 1994 genocide in Rwanda, approximately 20 000 Rwandan refugees who had fled into the North Kivu region of Zaire died in one month from dysentery caused by a strain of S dysenteriae that was resistant to all commonly used antibiotics.
Shigellosis is transmitted from humans-to-humans by the faecal-oral route via contaminated food and water or through person-to-person contact, as often observed in institutionalized populations. Transmission by house flies has also been documented  . Infection is common among travellers and military troops deployed in camps with less than optimal hygiene conditions .
In the absence of an existing effective vaccine, the ever increasing frequency of antimicrobial-resistant Shigella strains worldwide has become a major source of concern   During a survey of 600 000 persons of all ages residing in Bangladesh, China, Pakistan, Indonesia, Vietnam and Thailand, Shigellas were isolated in 5% of the 60 000 diarrhoea episodes detected between 2000 and 2004 and the majority of the bacterial isolates were resistant to amoxicillin and cotrimoxazole  . Similarly, during a 36-month surveillance study in a rural district in Thailand, where incidence of shigellosis was measured to be 4/1000/year in less than 5 years-old children, 95% of the S sonnei and S flexneri isolates were resistant to tetracycline and cotrimoxazole, and 90% of the S flexneri isolates were also resistant to ampicillin and chloramphenicol  . A similar finding was made in North Jakarta, Indonesia, where a surveillance study done between August 2001 and July 2003 found that children aged 1 to 2 years had a high incidence of shigellosis (32/1000/year) with 73% to 95% of the isolates being resistant to ampicillin, trimethoprim-sulfamethoxazole, chloramphenicol and tetracycline .
Three major species of Shigella are responsible for bacillary dysentery: S. sonnei, S. flexneri and S. dysenteriae. A fourth species, S boydii, is responsible for scattered disease foci. These species are further subdivided into subtypes on the basis of the antigen specificity of the O-polysaccharide portion of their LPS . S. sonnei , which has a single serotype, is the causative agent of most shigellosis in industrialized countries where it accounts for 77% of cases (compared to 15% in developing countries), but it also has become predominant in Thailand in recent years. S. flexneri, which has 14 serotypes and subtypes, is endemic in developing countries (60%) and is the most frequently isolated species worldwide. The predominant serotypes are S flexneri 2a, followed by 1b, 3a, 4a and 6. Untypable Shigellas have also recently emerged as a significant cause of diarrhoea. The serotype 1 of S. dysenteriae (Sd1) is of particular concern due to its expression of the Shiga toxin. It is the cause of epidemic dysentery and can cause vicious outbreaks in confined populations, especially refugee camps. A major obstacle to the control of Sd1 is its resistance to antimicrobial drugs. However, recent surveillance data from Bangladesh and India show that, for unknown reasons, Sd1 seems to have disappeared from these regions.
Shigella invade the colonic epithelium by transcytosis through M cells and penetration into the epithelial cell layer by the basolateral membrane, then spread laterally from cell-to-cell. This invasive ability is due to several virulence factors encoded by a high molecular weight virulence plasmid, including IpaB, which shows strong affinity for the CD44 hyaluronic acid receptor and IpaC, which forms pores through the basolateral membrane of epithelial cells, allowing the bacterium to penetrate into the cytoplasm of the cell. The bacteria multiply in the cytoplasm and eventually kill the host cell while moving towards adjacent epithelial cells by a process of polymerisation/depolymerisation of the actin tubules mediated by the VirA virulence factor  . Their pathogenicity is also due to several enteric toxins, including Shigella enterotoxin 1 (ShET-1), which is encoded by chromosal gene set present in S flexneri strains 2a, enterotoxin 2 (ShET-2), the product of gene sen, which is carried by a large virulence plasmid common to most Shigella spp, and the Shiga toxin, which is encoded by the stx gene from a phage of S. dysenteriae. ShET-1 is a classical enterotoxin made of an active A subunit associated with five B subunits. The Shiga toxin inhibits protein synthesis in eukaryotic cells via inactivation of ribosomal RNA, leading to cell death. The toxin is cytotoxic, neurotoxic and enterotoxic. It targets glomerular epithelial cells and central nervous system microvascular endothelial cells, causing an haemolytic-uremic syndrome (HUS) and seizures  . Sd1 also causes a rapid increase in the cell membrane permeability of infected macrophages and destroy their mitochondrial function .
Epidemiology and volunteer studies have shown that protective immunity to Shigellas is directed to the O-somatic antigen and is narrowly type-specific, which has hampered the development of an effective Shigella vaccine. Strong mucosal sIgA anti-O-antigen antibody responses follow wild-type infection and experimental challenge  . Passive administration of cow's milk immunoglobulin that contained high-titer anti S flexneri 2a antibodies protected volunteers from experimental challenge with wild type S flexneri, whereas cow's milk with low titers of such antibodies did not  . In addition, cell-mediated immunity mechanisms, including IFN-?-secreting T cells, seem to play a role in recovery and immunity  . Candidate shigellosis vaccines currently in advanced development include both polysaccharide conjugate and live attenuated vaccines and mostly focus on the most frequently isolated S flexneri 2a and S sonnei, as well as on Sd1, due to the severity of cases    .
Polysaccharide conjugate vaccines
Parenteral conjugate vaccines were produced from purified Shigella lipopolysaccharide (LPS) from the relevant bacterial serotypes that were conjugated to tetanus toxoid, recombinant Pseudomonas aeruginosa exotoxin A (PsA) or CRM9 mutant diphtheria toxin. These vaccines, which were developed at the NIH, were shown to be 74% efficacious against disease when tested in field trials with Israeli military volunteers   and demonstrated safety and immunogenicity in 4-7 year-old children   . A Phase III trial of S flexneri 2a and S sonnei conjugate vaccines was recently completed in young children in Israel  . O-antigen-specific IgG antibody elicited by conjugate polysaccharide vaccines prevent in vitro invasion of Shigella into Caco-2 cells in culture and may be curative  . The use of synthetic oligosaccharides that mimic the O-antigen protective epitopes and are conjugated to appropriate protein carriers offers promise for an improved and cheaper future generation of conjugate Shigella vaccines .
Live attenuated vaccines
Definite progress has been made in the field of candidate live oral attenuated shigellosis vaccines, but the small margin that exists between under-attenuation responsible for excessive reactogenicity of the vaccine, especially in children, and over-attenuation leading to insufficient immunogenicity in human subjects, especially in developing countries, remains a major problem.
Serially passaged streptomycin-dependent (SmD) S. flexneri 1, 2a, and 3a and S. sonnei strains were shown in former Yugoslavia to be attenuated in adults, healthy and institutionalized children and to constitute candidate oral vaccines that were protective in 82-100% of cases   . However, side effects due to reversion of the mutation were observed and the development of the vaccine was discontinued.
A live hybrid attenuated Shigella strain expessing both S. flexneri 2a and S. sonnei O-antigens was developed as an oral vaccine (FS) at the Lanzhou Institute of Biological Products, using the T32 attenuated S flexneri strain initially developed by Istrati in Romania. Large field studies in China have demonstrated 61-65% protection against S. flexneri 2a and 57-72% protection against S. sonnei. A protective efficacy against heterologous Shigella species was also claimed  . However, the use of a 3-dose vaccination regimen with high doses of a live vaccine strain (>2x10E10 cfu) remains problematic at this time. Further field studies of the FS vaccine in toddlers and infants may help define the public health application of this vaccine in China.
A live, attenuated S. flexneri 2a strain (SC602), and an attenuated S. dysenteriae type 1 strain (SC599) carrying mutations in their icsA, iuc, iut and stxA genes, were developed at the Pasteur Institute, Paris. IcsA is an outer membrane protein that nucleates cellular actin, thereby allowing the cell-to-cell spread of the bacteria. Iuc iut are involved in the scavenging of Fe3+ ions via the production of siderophores (aerobactin). StxA encodes the catalytic subunit of the Shiga toxin. SC602 was tested in adult volunteers in the USA and in adults and children in Bangladesh in collaboration with the Walter Reed Army Research Institute (WRAIR) and IVI. A remarkable efficacy against challenge was observed in USA volunteers   , but results of immunogenicity were disappointingly low in young infants in the field, due to lack of colonization of the gut, perhaps as a consequence of the presence of maternal antibodies from breastfeeding, or due to over-attenuation of the vaccine candidate for this target population.
Numerous approaches at attenuating S flexneri 2a by targeted deletion of virulence genes including the set and sen genes have similarly been carried out at the Center for Vaccine Development (CVD) in Maryland, USA, resulting in an array of candidate vaccine strains such as CVD 1203, CVD 1204, CVD 1207, CVD 1208 and culminating in strain CVD 1208S, which recently underwent Phase I clinical trials   (For a review, see  ).
Investigators at WRAIR also constructed a series of attenuated S sonnei strains, resulting in strain WRSs1 which showed immunogenicity and clinical acceptability  . They also attenuated Sd1 by deletion of its stxAB and virG/icsA genes, resulting in strain WRSd1  , which, unfortunately, was only modestly immunogenic in human volunteers, probably because of insufficient colonization of the gastrointestinal tract. The defect has been corrected in a more recent generation Sd1 vaccine strains by reinserting the fnr gene (fumarate/nitrate reductase), a global transcriptional regulator. At the same time, a new S flexneri vaccine candidate strain, WRSf2G11, was constructed by deleting the icsA, set and sen genes .
Altogether, this well illustrates the difficulty met in trying to achieve the right balance between robust immunogenicity, especially in young children, optimal colonization and shedding patterns, and clinical tolerance of the attenuated strains, as well as the need to develop a multivalent vaccine to cover a spectrum of Shigella spps.
As an alternative strategy, several groups have attempted to express Shigella O-antigens in well-tolerated live vectors such as E coli, generating strain EcSf2a-2, or attenuated S typhi strainTy21a, but with only limited efficacy so far (reviewed in  ).
Other candidate Shigella vaccines
A formalin-inactivated S sonnei vaccine (SsWC) was developed as an oral, killed, whole-cell vaccine at the Johns Hopkins University in Baltimore, MD, USA, and recently tested in a Phase I trial on a small number of volunteers  . Similarly, Antex (USA) is developing a Shigella inactivated whole-cell vaccine as well as an oral travellers' diarrhoea vaccine (ActivaxTM) containing antigens from Campylobacter, Shigella and ETEC. These candidate vaccines will shortly undergo clinical testing.
Several subunit Shigella vaccines, including a parenteral nuclear protein/ribosomal vaccine developed by the International Vaccine Institute (IVI) and the Walter Reed Army Institute of Research (WRAIR), and a nasally administered proteosome vaccine consisting of Shigella LPS linked to micelles of the outer membrane protein of group B Neisseria meningitidis, still are at a preclinical stage. A novel formulation using a bacterial extract invasion complex named invaplex, which contains IpaB, IpaC, and LPS from S flexneri and S sonnei, was found to elicit protection against challenge in the guinea pig model  .