Shousun C Szu1, Feng-Ying C Lin1, Rachel Schneerson1,
John B Robbins1
Vi An Ho2, Tran Cong Thanh3, Phan Van Bay2,
Dang Duc Trach4, Ha Ba Khiem3
1National
Institutes of Health,
Betheda, MD. U.
S. A.
2Dong Thap Provincial Hospital – Cao Lanh, 3Pasteur
Institut in Ho Chi Minh City,
4National Institute of Hygiene and Epidemiology, Hanoi.
Typhoid fever remains a common and serious disease that is becoming more difficult to treat because of widespread resistance to antibiotics. More than 100 years of research have neither provided a vaccine for routine vaccination of children, nor a consensus as to what are the essential virulence factors, the protective antigen or a quantitative measurement of the host factor that confers immunity to Salmonella typhi. We think that our findings with the Vi capsular polysaccharide of S. typhi have help to solve these problems.
Surface polysaccharides of bacteria that cause systemic infection are both essential virulence factors and protective antigens. The capsular polysaccharides of Gram– and Gram+ pathogens and the O-specific polysaccharides of Gram– confer virulence by shielding the bacterium against complement. The polysaccharides have no pharmacologic activity except their ability to elicit antibodies.
Enteric fever is the consequence of bacteremia and its complications caused by ingestion of Salmonella: almost all cases are caused by Groups A, B, C, and D Salmonella. The most common and severe enteric fever is typhoid fever that is caused S. typhi, the only member of this Genus that possesses a capsular polysaccharide, Vi.
Salmonella are classified into groups on the basis of their O-specific polysaccharides. S. typhi is an inhabitant of and pathogen for humans only. This pathogen is described as a clone - with the exception of a mutant strain in Indonesia with a different flagellar antigen. All S. typhi isolated from the blood of patients have a Vi capsular polysaccharide.
S. typhi were classified as Group D according to Kauffmann and White. The serologic specificity of the group D LPS is conferred by the outer domain called the O-specific polysaccharide. Group D antisera, produced by multiple intravenous injections of heat-killed salmonellae, failed to agglutinate S. typhi from the blood ("inagglutinable"): agglutinability was restored only after the salmonellae were boiled.
In 1931, Felix and Pitt discovered the surface antigen (denoted as Vi for virulence) that accounted for the inability of group D antisera to agglutinate S. typhi. Electron microscopy shows that the Vi is evenly distributed over the surface of S. typhi. Inagglutinable S. typhi were significantly more virulent for mice than agglutinable strains. Vaccination with inagglutinable, but not with agglutinable, strains protected against S. typhi in mice and only antisera with Vi antibodies conferred passive protection to mice. They reported that blood isolates of S. typhi regularly have Vi and commented about the loss of their inagglutinability (Vi) following passage on media in the laboratory and by storage. Frits Kauffmann, the most eminent bacteriologist of the time, refuted the notion that Vi was an essential virulence factor and a controversy was ignited about the role of this antigen that lasts until today.
Vi is occasionally detected on other Salmonella such as Salmonella paratyphi C. But Vi on non-typhoidal Salmonella does not prevent agglutination of the organism by antibody to the O-antigen (agglutinable) and antibody to the Vi does not protect mice against challenge with these organisms. Elaine Daniels et al., showed that this difference between the Vi on non-typhoidal Salmonella and S. typhi was quantitative.
During the 1890's, typhoid fever in the Boer War prompted development of a vaccine prepared according to the teachings of Pasteur - isolate and inactivate the causative organism, then inject it. But the protective antigen of S. typhi was not known and after many laboratory passages, the strain used for vaccine production lost it potency (Vi). In addition, the presence of the lipopolysaccharide (endotoxin) in this vaccine caused a high rate of adverse reactions.
Because the effectiveness of cellular vaccines was questioned, WHO-sponsored trials in the 1950's of two types of cellular vaccines: Vi-rich or K type vs Vi-poor or L type made according to protocols published by Walter Reed Army Institute of Research. A representative trial in Pristina, Yugoslavia showed both types conferred protection against typhoid but that the Vi-rich had a statistically significant higher degree of efficacy.
In an elegant series of experiments, lasting over a decade. Maurice Landy showed that active immunization with Vi or Vi-containing organisms and passively administered human IgG antibodies to Vi protected mice challenged with S. typhi. He concluded that the cellular vaccine was, in fact, a Vi vaccine. On the basis of these extensive experiments, clinical evaluation of purified Vi was attempted. But Vi prepared by the methodology of that time was too pyrogenic for human use. A clinically acceptable vaccine was prepared after Vi was treated with 1.0 N hydrochloric acid at 100°C for 24 hours. Despite its failure to protect chimpanzees challenged with S. typhi, a clinical trial of this heat and acid-treat Vi was attempted. This Vi failed to protect healthy adults challenged with S. typhi. Landy's clinical trial was never published in a refereed journal but "succeeded" in founding a new field of immunology that assumed secretory immunity, and not serum antibodies, was required to confer protection against typhoid fever and other enter pathogens.
A modern analysis of polysaccharide vaccines provided an explanation for the failure of the acid and heat-treated Vi to prevent typhoid. Vi, a linear homopolymer of a D (1®4) galacturonic acid N-acetylated at C2 and O-acetylated at C3, is rigid, with the two acetyl moieties exposed on its surface and whose carboxyl group hidden. Its immunogenicity, similar to those of the capsular polysaccharide vaccines meningococci, pneumococci and Haemophilus influenzae type b, is related to its molecular weight and structural integrity. The treated Vi had a reduced molecular weight and its surface had been changed by removal of its O-acetyl and N-acetyl moieties.
Vi prepared by modern methods was evaluated in 5 to 44 year-olds in Nepal and in school-age children in the Republic of South Africa. Vi elicited a ³4 fold rise in Vi antibodies in about 70% of recipients and conferred about 70% protection against typhoid fever. These results were confirmed recently in Southwest China.
WHO Requirements have been issued for Vi which is now licensed by national authorities more than 70 countries. But Vi is not suitable for routine immunization because of its age-related and T-cell independent properties.
There is difficulty in making diagnosis of typhoid fever. Gilman et al showed that, assuming a 100% recovery of S. typhi from the bone marrow, only 50% of patients had a positive blood culture and considerably less had the organism recovered from other sites including the stool and urine. The previous conception that the incidence of typhoid fever is very low in young children is probably related to difficulty and reluctance to take blood samples from this age group. Further, media composed of simple meat infusions are not effective in isolating S. typhi from the blood.
Studies in Indonesia with the orally-administered attenuated Ty21a strain and more recently in New Delhi showed that at least half of typhoid fever cases were in young children. In a population-based study in collaboration with workers in Dong Thap Province and the Pasteur Institut in Ho Chi Minh City, Dr. Kimi Lin showed that the annual attack rate of typhoid fever was highest among children less than 15 years of age: it was 413/100,000 in this age group and 358/100,000 in 2 to 4 year-olds. The probability of positive blood culture is related to the duration of fever and to the number of blood samples: the yield of S. typhi from blood cultures from children after 3 days of fever was 4.6 percent compared to 17 percent from those with ³7 days of fever. We have extrapolated these findings to show that probably all people in South Vietnam get typhoid fever by the age of 40 years.
Several synthetic schemes to form conjugates were used to improve the immunologic properties of Vi as was used successfully for H. influenzae type b. Technically, the unusual structure of Vi interfered with the usual approaches to this problem. Two linkers, N-succinimidyl-3-(2-pyridyl dithio) propionate (SPDP) and adipic acid dihydrazide (ADH), were used to form conjugates between Vi and several carrier proteins including the non-toxic fully antigenic recombinant protein of Pseudomonas aeruginosa (rEPA). Vi conjugates, using SPDP as the linker were more immunogenic than Vi in primates and, most importantly, elicited a booster response in that species. Conjugates using ADH as the linker proved to be even more immunogenic in animals and were then evaluated in Dong Thap province. Clinical trials showed that Vi-rEPA elicited a booster response in 2 to 4 year-olds (T-cell dependent) and induced about 3 times higher antibody levels than Vi injected into 5–14 year-olds (enhanced immunogenicity).
The safety and efficacy of Vi-rEPA were evaluated in a double-blinded, placebo-controlled, randomized trial of 2–5 year-olds of 16 communes in Dong Thap Province. Participants (11,091) received 2 injections 6 weeks apart of Vi-rEPA or saline. Active surveillance for typhoid was conducted for 27 months. No serious side reactions were observed. Typhoid was diagnosed by isolation of S. typhi from blood cultures after ³3 days of fever (³37.5°C). S. typhi was isolated from 4 fully vaccinated recipients of Vi-rEPA compared to 48 of placebo (efficacy 91.5%, C.I. 77.1 to 96.6). Among 773 children who received only 1 injection, there was 1 case in the Vi-rEPA group and 8 in the placebo. Cases were distributed evenly by age and throughout the study. Passive surveillance in the 18 months since the study ended showed 16 cases among vaccinees and 2 in placebos (87.5% efficacy).
Vi-rEPA-induced IgG anti-Vi was assayed by ELISA from 4 randomly selected recipients from each commune each month. 100 percent of the recipients of Vi-rEPA had ³10-fold rise of their IgG anti-Vi 10 weeks after 2nd injection. At 6 months IgG anti-Vi levels in the recipients of Vi-rEPA was 22.5 ELISA units (EU) or 35-fold higher than those of the placebos (p<0.001). Vi-rEPA-induced IgG anti-Vi levels decreased to 10.7 EU at 2 years; still 18.8-fold higher than of the controls (p<0.001). The decline of IgG anti-Vi levels at 2 years was 1.4 fold in the 4 to 5-year-olds compared to 2.4 fold in the 2 to 3-year-olds (NS).
We emphasize two findings. First, the efficacy of Vi-rEPA (91.5%) is the highest reported for any typhoid vaccine. It is likely that Vi-rEPA will be considerably more effective than the current vaccines in individuals >5 years. Second, is the efficacy of Vi-rEPA in young children. Should Vi-rEPA, administered to infants with DTP elicit levels of anti-Vi comparable to those in the 2 to 5 year-olds, Vi-rEPA could be included into the Expanded Program on Immunization of the WHO.
Lastly, we think that the principle derived from this study, that serum IgG antibodies confer immunity to a systemic disease caused by an enteric pathogen, can be extended to non-typhoidal salmonellae of Groups A, B, C, and D, Shigella, Vibrio cholerae O1 and O139, and to Escherichia coli O157.
Copyright
2003 by Pasteur Institute In HCM, All rights reserved. Contact us : pasteur@pasteur-hcm.org.vn
167 Pasteur Street, District 3, HoChiMinh City, Vietnam. Tel : (84-8) 8200739 -
Fax : (84-8) 8231419.