Class and subclass anti-pneumococcal antibody responses in splenectomized patients

Antibody titres against pneumococcal capsular and cell wall antigens and the immune response to polyvalent pneumococcal vaccine were measured in 21 splenectomized patients and 12 healthy controls. Most individuals possessed anti-pneumococcal capsular polysaccharide antibodies of IgG, IgA and IgM classes. The anti-capsular IgG was predominantly of the IgG1 and IgG2 subclasses; only occasional individuals had any detectable titre in IgG3 or IgG4 subclass. Most individuals responded to immunization with Pneumovax. There was no clear difference between groups of control and splenectomized subjects, although three of the splenectomized patients had undetectable pre-immunization anti-capsular titres in one or more subclass which failed to rise following immunization. All subjects tested had anti-phosphocholine antibodies in IgG, IgA and IgM classes with the exception of a single splenectomized patient who lacked detectable anti-phosphocholine IgM. Pre-immunization titres where similar in healthy controls and splenectomized patients. There was no demonstrable rise in anti-phosphocholine titre following immunization with Pneumovax.     

A newly characterized group of anti-phosphocholine antibodies: group III antibodies

Four anti-phosphocholine (PC) monoclonal antibodies (MAbs) are described that can be clearly distinguished from group I and II anti-PC antibodies [Chang et al. (1982) J. Immun. 128, 702-706]. They bind optimally and suboptimally PC-histone coated ELISA plates equally well, in contrast to group I and II MAbs which bind much less well to suboptimally coated plates. On optimally coated plates their fine specificity for PC analogues is group II-like, whereas on suboptimally coated plates it is group I-like; group I and II MAbs each maintain their own characteristic fine specificity patterns on either type of plate. At the molecular level, all four MAbs utilize VK24 light chains and share a 5.2 kb rearranged band in Southern blot analysis (using a JH probe and EcoRI restricted DNA). The term "group III" is proposed to define and distinguish these anti-PC antibodies.     

Rheumatoid factors in cystic fibrosis: associations with disease manifestations and recurrent bacterial infections.

Sera from a total of 78 patients infected with Wuchereria bancrofti or appropriate controls were assayed for anti-phosphocholine antibodies in an enzyme-linked immunosorbent-assay (ELISA), using phosphocholine as an antigen. Anti-PC antibodies (both IgM and IgG) were observed in patients with all clinical forms of filariasis but, unexpectedly, were not significantly different from those of normal controls. Among the filariasis patients, however, individuals with patent microfilaraemia had significantly lower IgM-anti-PC titres (P less than 0.05) than did those of the other clinical groups. Competitive studies showed that 'naturally occurring' anti-PC antibodies interfere with PC antigen detection. Transient increases in the levels of circulating PC antigen were found by 3 days after treatment of microfilaraemic individuals with the filaricidal drug diethylcarbamazine. The subsequent decreases in antigen levels were generally associated with increased IgM- and IgG-anti-PC antibody levels. Such changes likely affect not only the ability to monitor the parasitological status of such individuals, but also the patient's subsequent immune interactions with filarial parasites.     

A model for the hapten-induced enhancement of monoclonal antibody binding in ELISA

The unusual binding properties of three anti-phosphocholine (PC) monoclonal antibodies (MAbs) in ELISA has been reported [Hall T. J. (1987) Molec. Immun. 24, 773-777]. These MAbs gave reproducible 2-4 fold increases in binding to PC-protein coated plates in the presence of PC (approximately 2 mM), nitrophenyl PC (approximately 10 micron) and other PC analogues. A model is proposed here in which the complementary and opposite binding of hapten and PC-protein in the combining sites of these MAbs explains the hapten enhanceable binding in ELISA.     

Synthetic fragments of Vibrio cholerae O1 Inaba O-specific polysaccharide bound to a protein carrier are immunogenic in mice but do not induce protective antibodies

Development of Vibrio cholerae lipopolysaccharide (LPS) as a cholera vaccine immunogen is justified by the correlation of vibriocidal anti-LPS response with immunity. Two V. cholerae O1 LPS serotypes, Inaba and Ogawa, are associated with endemic and pandemic cholera. Both serotypes induce protective antibody following infection or vaccination. Structurally, the LPSs that define the serotypes are identical except for the terminal perosamine moiety, which has a methoxyl group at position 2 in Ogawa but a hydroxyl group in Inaba. The terminal sugar of the Ogawa LPS is a protective B-cell epitope. We chemically synthesized the terminal hexasaccharides of V. cholerae serotype Ogawa, which comprises in part the O-specific polysaccharide component of the native LPS, and coupled the oligosaccharide at different molar ratios to bovine serum albumin (BSA). Our initial studies with Ogawa immunogens showed that the conjugates induced protective antibody. We hypothesized that antibodies specific for the terminal sugar of Inaba LPS would also be protective. Neoglycoconjugates were prepared from synthetic Inaba oligosaccharides (disaccharide, tetrasaccharide, and hexasaccharide) and BSA at different levels of substitution. BALB/c mice responded to the Inaba carbohydrate (CHO)-BSA conjugates with levels of serum antibodies of comparable magnitude to those of mice immunized with Ogawa CHO-BSA conjugates, but the Inaba-specific antibodies (immunoglobulin M [IgM] and IgG1) were neither vibriocidal nor protective in the infant mouse cholera model. We hypothesize that the anti-Inaba antibodies induced by the Inaba CHO-BSA conjugates have enough affinity to be screened via enzyme-linked immunosorbent assay but not enough to be protective in vivo.     

The M603 idiotype is lost in the response to phosphocholine in terminal deoxynucleotidyl transferase-deficient mice

The majority of anti-phosphocholine (PC) antibodies induced by the PC epitope in Proteus morganii (PM) express the M603 idiotype (id), which is characterized by an invariant Asp to Asn substitution at the V(H):D(H) junction. To elucidate the molecular basis by which M603-like B cells acquire the mutations resulting in this invariant substitution, we analyzed the immune response to PC-PM in terminal deoxynucleotidyl transferase (TdT) gene knockout (KO) mice. In the absence of TdT, T15-id antibodies comprised 80-100% of the primary response to PC-PM. Less than 10% of the response in wild-type mice is T15-id(+). In TdT KO mice, the secondary response to PC-KLH was higher than in wild-type mice and was dominated by the germ-line T15-id. About 10% of this response, in both TdT KO and wild-type mice, comprised M167-id(+) antibodies. Additionally, none of the functionally rearranged V1/DFL16.1/J(H)1 cDNA isolated from PC-PM-immunized TdT KO mice showed the Asp/Asn substitution characteristic of PC-binding, PC-PM-induced M603-like antibodies. These data indicate that production of M603-id antibody is TdT dependent, while generation of M167-id antibody is TdT independent, and that in the absence of competition from M603-like B cells, T15-id B cells can respond to PC-PM.     

Patterns of protective associations differ for antibodies to P. falciparum‐infected erythrocytes and merozoites in immunity against malaria in children

Acquired antibodies play an important role in immunity to P. falciparum malaria and are typically directed towards surface antigens expressed by merozoites and infected erythrocytes (IEs). The importance of specific IE surface antigens as immune targets remains unclear. We evaluated antibodies and protective associations in two cohorts of children in Papua New Guinea. We used genetically‐modified P. falciparum to evaluate the importance of PfEMP1 and a P. falciparum isolate with a virulent phenotype. Our findings suggested that PfEMP1 was the dominant target of antibodies to the IE surface, including functional antibodies that promoted opsonic phagocytosis by monocytes. Antibodies were associated with increasing age and concurrent parasitemia, and were higher among children exposed to a higher force‐of‐infection as determined using molecular detection. Antibodies to IE surface antigens were consistently associated with reduced risk of malaria in both younger and older children. However, protective associations for antibodies to merozoite surface antigens were only observed in older children. This suggests that antibodies to IE surface antigens, particularly PfEMP1, play an earlier role in acquired immunity to malaria, whereas greater exposure is required for protective antibodies to merozoite antigens. These findings have implications for vaccine design and serosurveillance of malaria transmission and immunity.     

Antibodies against pneumococcal C-polysaccharide are not protective

The ability of antibodies against C-polysaccharide (C-Ps) to protect against experimental pneumococcal infection was examined in a mouse model. No protection against types 6A and 14 pneumococcal infection could be demonstrated neither with mouse monoclonal antibodies against C-Ps, specific for phosphorylcholine (PC) or for C-Ps backbone, nor for polyclonal rabbit immunsera against C-Ps. The monoclonal antibody with PC-specificity was protective against infection with type 27 pneumococcus, that has PC as part of its capsular polysaccharide. Type-specific mono- and polyclonal antibodies were highly protective against infection with the homologous type. We conclude that no species-specific protection can be achieved against intraperitoneal Streptococcus pneumoniae infection with optimally capsulated bacteria in outbred mice by passive immunization with antibodies to C-Ps.     

Natural anti-TNP antibodies from rainbow trout interfere with viral infection in vitro

Normal and viral-infected rainbow trout (RT) were tested for serum antibody activity against self and nonself antigens. Particularly high titres of anti-trinitrophenyl (TNP) antibodies were noted, as in other fish species. To analyse this, the anti-TNP antibodies were isolated by affinity chromatography and their capacity to interfere with viral infection in vitro was studied. We selected RT fibroblasts as target cells, and two common pathogenic viruses in trout, a rhabdovirus, viral haemorrhagic septicaemia virus (VHS) and a birnavirus, the infectious pancreatic necrosis virus (IPN). Anti-TNP antibodies were examined for their capacity to neutralize VHS and IPN viruses. Data obtained show that the anti-TNP antibodies, even at high concentrations, only partially neutralized virus. In contrast, when anti-TNP antibodies were assayed for their protective activity using RT fibroblast cells infected with VHS or IPN viruses, results showed high protective activity, regardless of serum origin or of the virus used, when the antibodies were added to the cell culture after viral infection. Therefore, our experiments indicate that the protective activity does not seem to be due to a direct interaction of the antibodies with the viruses. It is suggested that virus-modified cell surface self structures exhibit new epitopes which interact with the anti-TNP antibodies. Such an interaction would allow anti-TNP antibodies to participate in a non-specific defence mechanism against viral infection.     

Antibodies to lipooligosaccharide of a Brazilian purpuric fever isolate of Haemophilus influenzae biogroup aegyptius lack bactericidal and protective activity.

The immunological basis for protection against Brazilian purpuric fever (BPF), a fulminant infection of young children associated with bacteremia with Haemophilus influenzae biogroup aegyptius, is unknown. Candidate antigens to which protective antibodies may be directed include cell surface proteins and lipooligosaccharide (LOS). We studied the activity of antisera to LOS purified from a BPF H. influenzae biogroup aegyptius isolate. Anti-LOS antisera contained anti-LOS antibody by enzyme immunoassay and immunoblot and no detectable anti-outer membrane protein antibodies by immunoblot. Anti-LOS antisera had minimal bactericidal activity and were not protective against the homologous strain in an infant rat model of bacteremia. Antiserum to whole bacterial cells had a titer of anti-LOS antibody similar to that of anti-LOS antisera and was bactericidal and protective. Removal of anti-LOS antibodies from anti-whole cell antiserum by affinity chromatography did not result in a loss of bactericidal activity. Serum from a normal adult contained anti-LOS antibodies and had bactericidal activity. However, anti-LOS antibodies purified from this serum did not have detectable bactericidal activity. These studies suggest that anti-LOS antibodies produced in rats are not bactericidal and do not contribute to protection against experimental bacteremia with BPF strains of H. influenzae biogroup aegyptius.     

Avidity as a determinant of the protective efficacy of human antibodies to pneumococcal capsular polysaccharides

Antibodies reactive with capsular polysaccharides are considered the principal mediators of immunity against invasive diseases caused by Streptococcus pneumoniae. In this study, we tested the hypothesis that anti-pneumococcal capsular polysaccharide (PPS) antibody avidity can influence protective efficacy. We measured the avidities of individual adult postvaccination immunoglobulin G2 (IgG2) antibodies to PPS serotypes 6B and 23F and examined the relationship between avidity and opsonophagocytic and mouse-protective activities. The avidities of PPS 6B- and PPS 23F-specific IgG2 antibodies ranged from 6 to 31 nM-1 and from 3 to 20 nM-1, respectively. We observed an inverse correlation between the magnitude of avidity and the amount of antibody required to protect mice against lethal bacteremia caused by serotype 6B pneumococci. Similarly, higher-avidity antibodies were more effective than lower-avidity antibodies in vitro in mediating complement-dependent opsonophagocytosis of both 6B and 23F pneumococci. These data suggest that in adults, PPS antibodies are sufficiently polymorphic to possess biologically significant variations in avidity. We conclude that avidity functions as an important determinant of anticapsular antibody protective efficacy against pneumococci.     

Naturally acquired passive protective activity against Neisseria meningitidis Group C in the absence of serum bactericidal activity

The hallmark of immunity to meningococcal disease is a bactericidal titer in serum of > or =1:4 measured with human complement, but this threshold titer may underestimate the extent of protection. We used the infant rat model of meningococcal bacteremia to measure group C passive protective activity in serum samples from 91 unimmunized adults living in California. A total of 35 sera (38.5%) had passive protective activity. Sera with complement-mediated bactericidal titers of > or =1:4 were 3.4-fold more likely to confer protection (89%) than nonbactericidal sera (26%; P < 0.0001). Thus, bactericidal titers of > or =1:4 are a marker of protection, but this threshold lacks sensitivity for predicting protective activity. We investigated the 73 sera with bactericidal titers of <1:4 to determine the basis of protective activity. The 19 sera with protective activity had a higher geometric mean group C anticapsular antibody concentration (0.72 microg/ml) than the 54 sera that lacked protective activity (0.16 microg/ml; P < 0.001). Thus, protective activity in the absence of bactericidal activity was associated with higher concentrations of anticapsular antibodies, but not all sera with anticapsular antibodies conferred protection. Of 18 nonbactericidal sera with anticapsular antibody concentrations between 0.31 and 0.99 microg/ml, the 11 sera that conferred protection had a higher mean antibody avidity constant (21.9 nM(-1)) than the 7 nonprotective sera (14.6 nM(-1); P < 0.03). Thus, in sera with titers of <1:4, protective activity is associated with higher-avidity group C anticapsular antibodies, which are present in concentrations insufficient to elicit complement-mediated bacteriolysis in vitro but sufficient to confer protection in an in vivo bacteremia model.     

Evaluation of transferrin-binding protein 2 within the transferrin-binding protein complex as a potential antigen for future meningococcal vaccines.

Because the meningococcal transferrin receptor was shown to elicit bactericidal and protective antibodies in laboratory animals, we undertook a study of the protective role of each of the polypeptides within the Tbp1-Tbp2 complex. We developed a procedure to purify from Neisseria meningitidis B16B6 the two proteins in milligram amounts and raised specific antisera in rabbits and mice. Only antisera specific for Tbp2 displayed bactericidal activity against the parent strain. Mice immunized with purified Tbp2 survived a lethal challenge to a similar degree as animals immunized with the Tbp1-Tbp2 complex, demonstrating that Tbp2 played an important role in the protective activity observed with the complex. Both Tbp1- and Tbp2-specific antisera inhibited transferrin binding to the purified receptor in a solid-phase binding assay, suggesting that the antibodies were able to interact with the Tbp1 molecule only when it was removed from its membrane environment. Finally, Tbp2-specific immunoglobulins were able to lower the growth rate of the meningococci when human transferrin was their sole iron source. Therefore, in all four different systems tested, Tbp2 or antibodies specific for Tbp2 displayed biological characteristics close to those of the Tbp1-Tbp2 complex. This suggests that Tbp2 plays an important role in the protective activity of the complex, eliciting antibodies that are not only bactericidal but also inhibitory for meningococcal growth.     

Modified Opsonization,Phagocytosis, and Killing Assays To Measure Potentially Protective Antibodies against Pneumococcal Surface Protein A

The standard opsonophagocytosis killing assay (OPKA) for antibodies to pneumococcal capsular polysaccharide was modified to permit an evaluation of the protection-mediating antibodies to pneumococcal surface protein A (PspA). We found that by increasing the incubation time with the complement and phagocytes from 45 min to 75 min, the protective activity was readily detected. In another modification, we used a capsule type 2 target strain that expressed PspA but not pneumococcal surface protein C (PspC). With these modifications separately or in combination, rabbit antisera to the recombinant α-helical or proline-rich domains of PspA mediated >50% killing of the target strain. The ability of normal human sera to mediate the killing of pneumococci in this modified OPKA correlated with their levels of antibodies to PspA and their ability to protect mice against fatal infection with a type 3 strain. Passive protection of mice against pneumococci and killing in the modified OPKA were lost when normal human sera were adsorbed with recombinant PspA (rPspA) on Sepharose, thus supporting the potential utility of the modified OPKA to detect protective antibodies to PspA. In the standard OPKA, monoclonal antibodies to PspA were strongly protective in the presence of subprotective amounts of anti-capsule. Thus, the currently established high-throughput OPKA for antibodies to capsule could be modified in one of two ways to permit an evaluation of the opsonic efficacy of antibodies to PspA.     

The protective capacity of immune sera in experimental mouse salmonellosis is mainly due to IgM antibodies

Passive immunization was used to protect mice against a general infection caused by Salmonella typhimurium and our purpose was to compare the protective capacity of different immunoglobulin isotypes (classes and subclasses). Three antisera were studied, one pool of mouse serum against the envelope of rough bacteria, and two rabbit sera against smooth bacteria. Three different methods were used to separate isotypes. The consistent finding was that only IgM antibodies protected efficiently. A unit of IgG antibodies had an effect that was 1/50th of the IgM effect or less. This effect could have been due to a contamination by IgM. IgA appears to be non-protective like IgG. In two of the antisera a considerable proportion of protective antibodies were against a defined antigenic determinant (anti-0–4,5 or anti-0–9). IgG antibodies of these sera measured by the solid phase assay were also predominantly anti-0–4,5 or anti-0–9, respectively. This argues that the failure of IgG antibodies to protect cannot be explained by assuming that unlike IgM antibodies they are directed against “non-protective” determinants. We conclude that the observed difference between the protective capacities of IgM and IgG antibodies is due to C-region differences between the μ- and γ-chains.     

Biologic activities of antibodies to the neutral-polysaccharide component of the Pseudomonas aeruginosa lipopolysaccharide are blocked by O side chains and mucoid exopolysaccharide (alginate).

Virulent strains of Pseudomonas aeruginosa are either of a nonmucoid, lipopolysaccharide (LPS)-smooth or mucoid, LPS-rough phenotype, and immunity to these different variants is efficiently mediated by antibodies specific to O antigens or mucoid exopolysaccharide (also called alginate), respectively. In addition to O side chains and core polysaccharide components, the LPS of P. aeruginosa also contains neutral-polysaccharide components that express antigenic determinants common to many clinical isolates. We evaluated antibodies specific to neutral polysaccharides for the ability to mediate opsonic killing and protective immunity. Antibodies to these antigens mediated opsonic killing of poorly virulent nonmucoid LPS-rough isolates but not of isogenic strains with either a LPS-smooth or a mucoid phenotype. Antibodies to neutral-polysaccharide antigens also failed to protect neutropenic mice from challenge with modest doses of LPS-smooth P. aeruginosa strains (< 10(3) CFU per mouse), whereas O-antigen-specific antibodies were highly protective. Antibodies to neutral polysaccharides deposited significantly (P = 0.002) more C3 onto LPS-rough strains than did antibodies to O side chains, but this situation was reversed when isogenic LPS-smooth strains were tested. Given that protective immunity against P. aeruginosa must be directed against either nonmucoid LPS-smooth strains or mucoid LPS-rough strains, it appears that antibodies specific to neutral-polysaccharide antigens do not protect against P. aeruginosa infection. Lack of protection is likely due to the ability of both O side chains and mucoid exopolysaccharide (alginate) to interfere with the opsonic killing activity of neutral-polysaccharide-specific antibodies.     

Evidence for the presence of lipopolysaccharide in a ribonuclease-sensitive ribosomal vaccine of Pseudomonas aeruginosa.

To obtain information about the nature of the immunogens in the ribosomal vaccine (fraction II) of Pseudomonas aeruginosa, we studied the specificity of rabbit antibodies to fraction II. Crossed immunoelectrophoresis demonstrated the presence of antibodies which precipitated with ribosomal antigens, but not with lipopolysaccharide (LPS). By means of an enzyme-linked immunosorbent assay, antibodies to LPS were detected in antibodies to fraction II. Antibodies to fraction II could protect mice against a lethal challenge with P. aeruginosa. Absorption experiments demonstrated that the protective ability of antibodies to fraction II was due to antibodies to cell envelope antigens, whereas antibodies to ribosomal antigens did not contribute to the protection. Antibodies to LPS could be detected in mice 1 week after a single vaccination with fraction II. It was concluded that the protective activity of fraction II was due, at least in part, to the presence of LPS in the ribosomal vaccine. Treatment of fraction II with ribonuclease decreased the protective activity of the ribosomal vaccine. Addition of synthetic polyadenylic acid-polyuridylic acid restored the protective activity of ribonuclease-treated fraction II, indicating that RNA in the ribosomal vaccine might act as an adjuvant or a carrier in the presentation of the of the contaminating cell envelope antigens. The protective activity and the toxicity of fraction II were compared with the protective activity and the toxicity of fraction I, which contained cell envelope components, including LPS, and of purified LPS. The results indicated that protection by the ribosomal vaccine was associated with a slightly higher toxicity than was protection by fraction I, whereas purified LPS was the most toxic vaccine.     

Development of a competitive enzyme linked immunosorbent assay to identify epitope specific antibodies in recipients of the U.S. licensed anthrax vaccine

Vaccination with anthrax vaccine adsorbed (AVA) results in the production of protective antigen (PA) specific antibodies, which play an important protective role against anthrax toxins. Analyzing the specificity of serum antibodies generated in response to AVA vaccination can provide insight into the mechanisms of protective immunity against this important pathogen. The goal of this study was to develop a competitive enzyme linked immunosorbent assay (cELISA) to test human immune serum for antibodies specific for a known lethal toxin neutralizing epitope in PA. PA-specific antibodies in sera from individuals who received the six-dose AVA vaccine series competed for binding to immobilized PA with monoclonal antibody F20G75, which binds to a linear epitope in domain 2 of PA and neutralizes lethal toxin activity in vitro. These results suggest that antibodies in human AVA vaccinee serum recognize the same epitope as F20G75, or one in close proximity to it, and may serve a protective role against anthrax lethal toxin. This assay may be used for serological confirmation of successful immunization against anthrax and for the identification of antibodies in human vaccinee serum that recognize protective epitopes on PA.     

The MHC class I-related FcRn ameliorates murine Lyme arthritis

The identification of the neonatal FcR (FcRn) as an IgG homeostasis regulator has led to research aimed at delineating a role for FcRn in humorally mediated disease. FcRn is a class I-related molecule that prolongs the half-life of serum IgG by preferentially binding IgG at low pH and inhibiting its degradation. Its role in protective immunity to infectious organisms is unknown. We investigated the function of FcRn in the murine model of Lyme arthritis, caused by infection with Borrelia burgdorferi. We infected FcRn(-/-) and wild-type mice with B. burgdorferi and monitored the development of arthritis. Infected FcRn(-/-) mice demonstrated decreased serum levels of anti-B. burgdorferi antibodies and borreliacidal activity. Moreover, these mutant mice developed increased ankle swelling and joint histopathology following infection. Our data suggest that FcRn ameliorates murine Lyme arthritis by preventing the degradation of protective borreliacidal antibodies.     

The C-terminal domain is essential for protective activity of the Bordetella pertussis adenylate cyclase-hemolysin.

The adenylate cyclase-hemolysin of Bordetella pertussis consists of a cell-invasive N-terminal adenylate cyclase domain linked to a C-terminal RTX hemolysin containing extensive glycine-rich repeats. The toxin is an essential virulence factor required in the initial stages of infection. Adenylate cyclase-hemolysin was also shown to be a potent vaccinating antigen inducing protection against B. pertussis colonization of the mouse respiratory tract. This protective activity depends on a posttranslational fatty-acylation modification. We used a set of deletion derivatives of the recombinant adenylate cyclase-hemolysin to localize the protective epitopes on the 1,706-residue toxin. We show that specific anti-adenylate cyclase-hemolysin antibodies present in the sera of B. pertussis-infected mice and humans are directed predominantly against the modification-and-repeat portion of the toxin, contained in the last 800 residues of the adenylate cyclase-hemolysin. These antibodies appear to recognize conformational epitopes present only in a structure formed by the intact C-terminal half of the toxin. There was no correlation between the capacity of the truncated adenylate cyclase-hemolysin derivatives to induce both toxin-neutralizing antibodies upon immunization of mice and protective immunity. However, only the truncated proteins which were recognized by the sera of infected mice and humans and which had their last 800 residues intact had the capacity to induce protection of mice against colonization by B. pertussis. This indicates that the structure of the modification-and-repeat region of adenylate cyclase-hemolysin is critical for its protective activity.