Prospective Study of Systemic and Mucosal Immune Responses in ...

2 downloads 0 Views 2MB Size Report
Apr 15, 1991 - Dysenteric Patients to Specific Shigella Invasion Plasmid ... correlates of specific protection against shigellosis, we examined chronological ...
Vol. 59, No. 7

INFECTION AND IMMUNITY, JUlY 1991, p. 2341-2350 0019-9567/91/072341-10$02.00/0 Copyright (C 1991, American Society for Microbiology

Prospective Study of Systemic and Mucosal Immune Responses in Dysenteric Patients to Specific Shigella Invasion Plasmid Antigens and Lipopolysaccharides RICHARD A. OBERHELMAN, 12t* DENNIS J. KOPECKO,3 EDUARDO SALAZAR-LINDO,2 EDUARDO GOTUZZO,2 JERRY M. BUYSSE,3 MALABI M. VENKATESAN,3 AUGUSTO YI,2 CARMEN FERNANDEZ-PRADA,2 MIGUEL GUZMAN,2 RAUL LEON-BARUA,2 AND R. BRADLEY SACK' Division of Geographic Medicine, Department of International Health, The Johns Hopkins University School of Hygiene and Public Health, 615 North Wolfe Street, Baltimore, Maryland 21205'; Universidad Peruana Cayetano Heredia, A.P. 5045, Lima 100, PerM2; and Department of Bacterial Immunology,

Walter Reed Army Institute of Research, Washington, D.C.3 Received 26 November 1990/Accepted 15 April 1991

Shigellosis is a major cause of infant morbidity and mortality in developing countries. To find immunological correlates of specific protection against shigellosis, we examined chronological samples of sera, stool extracts, duodenal aspirates, and saliva samples from 39 adults and 22 children with shigellosis from Peru for the presence of specific antibody to invasion plasmid antigens (Ipa) common to all virulent Shigella strains, by using both a whole-organism enzyme-linked immunosorbent assay (ELISA) and a Western blot (immunoblot) assay. Antibody responses to lipopolysaccharide (LPS) from Shigella serotypes both homologous and heterologous to the infecting strain were also determined by ELISA. ELISAs showed that the highest serum immunoglobulin G (IgG) antibody titers to Shigella whole organisms both with and without surface Ipa were found in adults and malnourished children, the two groups with the shortest and longest durations of disease, respectively. Mucosal IgA antibody titers to Shigella strains decreased over time to a much greater extent than serum IgG titers, and IgA to Ipa in mucosal secretions was found in adults and well-nourished children but not in malnourished children. The presence of mucosal antibody to Ipa may limit the spread and severity of the infection, as indicated by the prolonged illness observed in malnourished children who have no significant mucosal antibody to Shigella Ipa. Serum antibody titers to the Ipa antigens were high relative to anti-Shigella LPS antibody titers, especially in pediatric patients. In contrast to the anti-Ipa responses observed, no differences in antibody responses to LPS in children compared by nutritional status were found. High levels of serum and mucosal cross-reacting antibody to heterologous serotype LPS were found between Shigellaflexneri serotypes la and 2a. Different patterns of immune response to Ipa proteins and LPS that may aid in the definition of Shigella antigens important in host protection were observed in adults, well-nourished children, and malnourished children.

Shigella infections continue to be a major cause of morbidity and mortality in developing countries, accounting for 1.5 to 14% of the cases of pediatric diarrheal disease in various studies (3, 10, 22, 23). This pathogen has not only been shown to cause acute gastroenteritis but has also been associated with chronic diarrhea in malnourished children (18). In our patient population, a higher mortality rate occurred with Shigella infections than with infections caused by other enteropathogens, as observed in a series of pediatric inpatients with diarrhea followed in Lima (6 deaths of 28 Shigella cases versus 0 deaths of 34 control cases). Aside from the promotion of improved personal hygiene, which is a measure designed to decrease the prevalence of all enteropathogens, the present control measures for shigellosis consist of antibiotic therapy alone. Unfortunately, multiple antibiotic resistance in shigellae is now commonplace. Candidate antidysentery vaccines have been pursued for many

years now. However, no vaccine candidates have thus far proven highly useful. The logical development of vaccines is hampered by the lack of information on what constitutes specific immunity to shigellae. Limited epidemiological surveys (5) and studies of experimentally infected humans and animals (8) have indicated that Shigella serotype-specific immunity occurs following bacillary dysentery. The average length of effective immunity to bacillary dysentery has not been systematically analyzed. Studies of early prototype attenuated live Shigella oral vaccines (8) verified the development of serotypespecific protection following immunization. Although these and other early studies showed that Shigella infection in the gut generally causes an increase in anti-Shigella serum antibodies (8), induced immunity to Shigella has not yet been demonstrated to correlate strictly with the presence of specific anti-Shigella serum antibodies (7). In fact, parenterally administered, killed whole-cell Shigella vaccines stimulated in recipients a good anti-Shigella serum antibody response but failed to protect animals or humans against shigellosis (8). The important questions of what constitutes the basis of serotype-specific immunity, what the average length of serotype-specific protection following natural Shigella infection is, and whether any lipopolysaccharide

* Corresponding author. t Present address: Department of Tropical Medicine, Tulane University School of Public Health and Tropical Medicine, 1501 Canal Street, 5th Floor, New Orleans, LA 70112.

2341

2342

OBERHELMAN ET AL.

(LPS)-based, cross-serotype protective immunity exists remain unanswered. Recent work designed to facilitate vaccine development has concentrated on finding virulence or antigenic factors that are common to all pathogenic Shigella and enteroinvasive Escherichia coli strains. The discovery of large 180- to 210-kb plasmids in virulent Shigella and enteroinvasive E. coli strains (15) that code for enteroinvasion-related proteins which are antigenically similar in different species (12) has led to speculation about stimulating cross-protective immunity directed against these proteins. Transfer of this large plasmid from an invasive Shigella strain to a nonpathogenic E. coli strain has resulted in the transfer of the enteroinvasive capacity in tissue culture, demonstrating the role of some protein products of the plasmid in conferring invasivity (26). Enteroinvasion-related proteins encoded by this plasmid have been characterized and designated invasion plasmid antigens (Ipa), including IpaA (78 kDa), IpaB (62 kDa), IpaC (42 kDa), and IpaD (39 kDa) (1, 12, 28). These Ipa proteins have been shown to be potent immunogens in the sera of experimentally infected monkeys and Thai children with natural infection (20). Studies of infected rhesus monkeys have recently shown greater mucosal immunoglobulin G (IgG) and IgA responses to whole organisms and water extracts of virulent, plasmid-containing Shigella strains than to avirulent, plasmid-negative Shigella strains (4). However, studies of specific mucosal responses to Shigella infections in humans have been lacking. The purpose of the present analysis was to better define the natural history of both systemic and mucosal IgG and IgA responses to Shigella infections in humans. Thus, a prospective study in a group of Peruvian adults and children with shigellosis was conducted to determine Shigellaspecific antibody titers and the presence of antibody to specific Ipa proteins and specific Shigella LPS. This systemic and mucosal antibody information was compared with various clinical characteristics of the patient group, such as age, nutritional status, and duration of disease, to obtain correlative evidence of antibody-mediated protection. MATERIALS AND METHODS Patient enrollment and recorded clinical data. Infants, children, and adults with clinical dysentery (defined as three or more unformed stools in 24 h with gross fecal blood and/or greater than 25 fecal leukocytes per high-power field) were identified in a group of inpatients and outpatients at the Hospital General Base Cayetano Heredia, Lima, Peru. Stool cultures were used to identify patients with Shigella infection, and those with positive Shigella cultures were enrolled, after informed consent, in a follow-up program to obtain samples for immunologic investigation. Clinical data, recorded for all patients, included characteristics of the stools, state of hydration, nutritional status, and duration of illness. The nutritional status was determined as percent weight for age by the classification of Gomez et al. (11) as follows: well nourished, percent weight for age, >90%; first-degree (10) malnutrition, 75 to 90%; second-degree (20) malnutrition, 60 to 75%; and third-degree (30) malnutrition,