Induction of a protective capsular polysaccharide antibody response to a multiepitope DNA vaccine encoding a peptide mimic of meningococcal serogroup C capsular polysaccharide

Abstract

Systemic infection by encapsulated organisms, such as Neisseria meningitidis, is a major cause of morbidity and mortality worldwide, especially in individuals less than 2 years of age. Antibodies directed at the capsular polysaccharide are shown to be protective against disease by inducing complement-dependent bactericidal activity. The current polysaccharide vaccine has been shown to be poorly immunogenic in high-risk groups and this is probably related to its T-independent properties. An alternative approach to eliciting a T-dependent serum immunoglobulin G (IgG) antibody response to encapsulated pathogens is DNA vaccination. We assessed the immunogenicity of a multiepitope DNA vaccine encoding a T-cell helper epitope and a peptide mimic of N. meningitidis serogroup C. The DNA construct induced a significant anti-polysaccharide antibody response that was bactericidal. Mice immunized with the DNA construct were subsequently protected against challenge with a lethal dose of N. meningitidis serogroup C.

Introduction

Neisseria meningitidis, a Gram-negative encapsulated bacterium, is a major cause of morbidity and mortality worldwide.¹N. meningitidis is a leading cause of bacterial meningitis and septicaemia in young children, especially those under 2 years of age.^2,3 Despite the availability of antibiotics and a polysaccharide vaccine, >350 000 cases of meningococcal disease occur annually throughout the world.^4,5 Under optimal conditions of identification and therapy, the mortality rate is at least 10% and the morbidity is as high as 30%.^6,7 Meningococci are classified into serogroups according to the structure of their capsular polysaccharides.^8–10 To date, 13 serogroups have been identified; however, four serogroups, namely A, B, C and Y, account for ≈90% of meningococcal meningitis cases.^5,7

In the United States, serogroup C alone is responsible for 45% of the cases of meningococcal meningitis each year.⁷ The group C meningococcal polysaccharide vaccine is based on the observation that antibodies directed at the polysaccharide capsule protect against disease by inducing complement-dependent bactericidal activity.^11–14 Although it is currently recommended for controlling meningococcal outbreaks and epidemics, the polysaccharide vaccine has several limitations. First, group C polysaccharide fails to elicit protective levels of antibodies in children < 2 years of age, the age-group with the highest incidence, morbidity and mortality. Second, reimmunization in this age-group causes suppression, rather than enhancement, of the immune response.^15–17 The poor efficacy and limited duration of the immune response in young children is attributed to the T-independent nature of the polysaccharide vaccine, as the response to these antigens develops late in ontogeny.^18–20 A polysaccharide–protein conjugate converts group C polysaccharide from a T-cell independent antigen into an immunogenic and effective vaccine at all ages.^21–26 Limitations of conjugates are the requirement of at least three, and possibly four, immunizations in infancy to achieve optimal efficacy. In addition, studies have shown that the antibody response to the polysaccharide component of the conjugate continues to possess characteristics of a T-independent immune response.^27–29

An alternative approach to the development of a T-dependent immune response to a carbohydrate is the use of anti-idiotypic (anti-Id) immunoglobulins that mimic the polysaccharide.^30–33 Anti-Id immunoglobulins can act as surrogate images of the carbohydrate antigen and can induce a functional, carbohydrate antibody response.^34,35 Previously, we identified an anti-Id immunoglobulin that mimics the polysaccharide of N. meningitidis serogroup C.³⁶ This anti-Id immunoglobulin elicited a T-dependent anti-polysaccharide antibody response that was protective in animal studies.^36,37 Further immunization studies demonstrated that a peptide, spanning the immunogenic CDR3 region of the anti-Id immunoglobulin, when complexed to proteosomes induced protection against a lethal challenge with meningococci.³⁸ These findings illustrate that a peptide mimic of a capsular polysaccharide can induce a T-dependent immune response.

A natural extension of the observation that peptides can mimic polysaccharides is the development of DNA constructs that induce polysaccharide antibodies. DNA immunization is based on the concept that endogenous expression of an encoded peptide by host cell machinery will stimulate an immune response similar to that induced by a pathogen.^39–41 Several studies have demonstrated the potential advantages of DNA immunization.^42,43 First, DNA vaccines are inexpensive to produce and easily constructed. Second, DNA vaccines are thermally stable at high and low temperatures, avoiding the need for cold-chain transport. This is essential for delivering the vaccine to areas of endemic disease. Third, DNA vaccines are malleable and can be altered to modulate the magnitude and orientation of the immune response.^44–46 This allows the insertion of multiple DNA-encoded epitope sequences, representing protein as well as carbohydrate antigens, into the DNA vector.^47–51 An advantage of this approach is the induction of a multispecific immune response directed at various immunologically relevant epitopes.^52,53 The ability to modulate the immune response to specific protein and carbohydrate epitopes makes DNA vaccines ideal for use against encapsulated organisms, where antibodies can be directed at both the polysaccharide and relevant protein epitopes. Including epitopes that encode peptide mimics of various serogroup polysaccharides, and those encoding conserved cell-wall components (i.e. proteins), allows the construction of a multiepitope DNA vaccine that has the potential of protecting against various meningococcal strains.

In this present study, we constructed a multiepitope DNA vaccine, encoding a peptide mimic of meningococcal serogroup C capsular polysaccharide (MCPS), a universal T-cell helper epitope, and a secretory leader sequence. Immunization with this DNA construct resulted in a functional anti-MCPS antibody response. Live challenge studies demonstrated that the immune response elicited by the DNA construct was protective against a lethal dose of meningococci.

Materials and methods

Capsular polysaccharides and bacterial strains

The MCPS was purchased from the National Institute for Biological Standards and Control (Potters Bar, Hertfordshire, UK). Meningococcal strains C-11 and C-35E were kind gifts from G. M. Carlone (Centers for Disease Control and Prevention, Atlanta, GA), and were used for bactericidal assays and lethal challenge.

DNA construct

The DNA expression vector, pcDNA3.1 (Invitrogen, Carlsbad, CA), was used for the construction of our multiepitope design. Sense and antisense oligonucleotides were synthesized (by Integrated DNA Techniques, Coralville, IA) to encode: MRYMILGLLALAAVCSAA,⁵⁴ a secretory leader sequence derived from the adenovirus E3; KQIINMQAVGKAMYA,⁵⁵ a T-cell helper epitope from human immunodeficiency virus (HIV) gp120; and KQIINMQAVGKAMYAARIYYRYDGFAY, a T-cell helper epitope from HIV gp120 in tandem with the B-cell epitope (MCPS peptide mimic). Purified oligonucleotides were resuspended in distilled water to a final concentration of 1 nmol/µl. For the P3A DNA construct, the leader-sequence oligonucleotide was cloned into the pcDNA3.1 vector at the HindIII and NotI restriction sites. Subsequently, the T- and B-cell epitope-encoded oligonucleotide was cloned into the vector between the NotI and XbaI restriction sites. For the P3C DNA construct, the oligonucleotides encoding the secretory leader sequence and T-cell epitope were cloned into the vector, as described above. Briefly, the pcDNA3.1 expression vector was double digested at 37° with the appropriate restriction enzymes. Annealed sense and antisense strands of the oligonucleotides were ligated into the digested vector for 16 hr at 16°. Transformation of the ligate was performed in Top 10F′Escherichia coli competent cells (Invitrogen), which were plated out onto agar plates containing 50 µg/ml ampicillin. Transformed bacterial clones were screened for an insert by the polymerase chain reaction (PCR) using appropriate primers. Positive clones were further screened for correct orientation of the insert by DNA sequencing using a Thermosequenase radiolabelled terminator sequencing kit (Amersham, Cleveland, OH). Plasmid DNA for immunizations was prepared using the endofree Qiagen-tip 500 (Maxi) kit (Qiagen Inc., Santa Clara, CA) according to the manufacturer's protocol. Endotoxin-free phosphate-buffered saline (PBS) was used to resuspend DNA.

DNA immunization studies

Eight-week-old, female BALB/c mice (n = 6) were immunized intramuscularly (i.m.) in the anterior tibialis with purified DNA or intraperitoneally (i.p.) with MCPS. Mice were sedated before immunization using a mixture of Ketamine HCl and Xylazine, at 80 mg/kg and 16 mg/kg, respectively. Mice were immunized on days 0, 21, and 63 with 100 µg of DNA suspended in 100 µl of PBS and 40 µg of aluminium phosphate gel adjuvant (Superfos Biosector a/s, Vedbaek, Denmark). The test group consisted of mice immunized with the P3A construct containing the leader sequence and T- and B-cell epitopes. Controls were mice immunized with the P3C construct containing the leader sequence, T-cell epitope but lacking a B-cell epitope, and mice immunized i.p. with 5 µg of MCPS on day 0. All mice were tail-vein bled on day 0 (preimmunization) and on days 28, 42, 56, 70, and 84 postimmunization, and the sera were screened for anti-MCPS immunoglobulin.

MCPS enzyme-linked immunosorbent assay (ELISA)

MCPS antibodies were measured by ELISA. Briefly, 96-well microtitre plates were coated overnight at 37° with 50 µl/well MCPS and methylated human serum albumin (NIBSC) at a final concentration of 10 µg/ml. Wells were washed three times in PBS/0·1% Brij^® 35 Solution detergent, and blocked with 200 µl of blocking buffer (PBS containing 10% heat-inactivated calf serum and 0·1% Brij). Serial dilutions of postimmune sera, in blocking buffer, were added to the microtitre wells and incubated overnight at 4° with positive- and negative-control sera present on each plate. Bound antibody was detected after 1 hr of incubation with peroxidase-conjugated anti-mouse immunoglobulin (Ig)M, IgG, IgG1, IgG2a, IgG2b and IgG3 (Southern Biotech, Birmingham, AL) at a 1 : 3000 dilution in blocking buffer, and developed with substrate buffer containing 0·02%o-phenylenediamine. Absorbance levels were read at 490 nm with an enzyme immunoassay (EIA) reader (Bio-Tek instruments, Winooski, VT). Anti-MCPS immunoglobulin titres were calculated as the dilution of serum corresponding to 25% of the maximum optical density obtained with the positive reference serum.

Serogroup C bactericidal assays

The serum bactericidal assay was performed as described previously.⁵⁶N. meningitidis serogroup C strain C-11 was grown overnight at 37° in 5% CO₂ on brain heart infusion agar supplemented with 1% heat-inactivated horse serum (BHIA/S). A working bacterial seed lot was prepared by resuspending overnight colonies in Dulbecco's PBS containing 0·1% glucose, to yield a final concentration of 2 × 10⁴ colony-forming units (CFUs)/ml. Dilutions of heat-inactivated postimmune sera, ranging from 1 : 8 to 1 : 4096, were added to a 96-well tissue-culture plate. An equivalent cell/complement mixture (baby rabbit serum; Pelfreeze, Brown, Deer, WI) was added to all wells of the tissue-culture plate, excluding the control wells, and incubated at 37°. Controls included wells without postimmune sera, wells without a complement source, and wells with positive control sera. At 0- and 60-min time-points, aliquots from the wells were plated out onto BHIA/S plates using the agar-tilt method and the plates were incubated overnight at 37° with CO₂. The bactericidal titre was reported as the lowest serum dilution yielding > 50% killing compared with time zero.

Challenge experiments

Additional groups of mice (n = 6) were immunized with DNA or MCPS. Mice were challenged on day 56 with N. meningitidis serogroup C, strain 35E, using a murine model for meningococcal infection.^57,58 Seven days prior to challenge, mice received 1000 mg/kg iron dextran i.p. to enhance susceptibility to the bacteria. The lethal dose 50% (LD₅₀) value was determined in iron-loaded naïve mice challenged with 10³−10⁵ CFU of meningococci. Immunized iron-treated mice were challenged i.p. with 10 × LD₅₀ under sterile conditions with strain 35E. Ninety-six hours postchallenge, survivors were killed and their sera were collected for analysis.

Statistical analysis

Analysis of pre- and postimmune sera antibody and bactericidal titres was performed by the non-parametric Mann–Whitney U-test (SPSS, Chicago, IL). Statistical analysis on postchallenge sera was performed using Fisher's exact test (SPSS). A P-value of < 0·05 was considered statistically significant.

Results

Construction of DNA vaccines

For the purpose of our study, a secretory leader sequence from adenovirus E3 was included in the DNA construct to increase exogenous expression of our B- and T-cell encoded peptides. The T-cell helper epitope from HIV gp120 is considered a ‘universal’ T-cell epitope owing to the lack of human leucocyte antigen (HLA) restriction and its ability to bind to both major histocompatibility complex (MHC) class I and II molecules.⁵⁵

Oligonucleotides encoding the leader sequence, the T-cell helper epitope and the peptide mimic of the MCPS B-cell epitope, were cloned into the appropriate restriction sites of the pcDNA3.1 expression vector. Bacterial colonies containing inserted oligonucleotides were identified using the PCR. To ensure that the oligonucleotides were inserted in the correct orientation and reading frame, clones identified as positive by the PCR were sequenced. A plasmid vector that demonstrated the correct orientation and insertion was selected for further analysis. Plasmid DNA for immunizations was prepared and resuspended in endotoxin-free PBS.

MCPS antibody response to DNA immunization

Groups of BALB/c mice (n = 6) were immunized on days 0, 21, and 63 with 100 µg of the indicated DNA construct emulsified in aluminium-phosphate gel adjuvant. The positive vaccine control consisted of mice immunized on day 0 with 5 µg of MCPS. Sera were obtained and the anti-MCPS immunoglobulin response was determined by ELISA throughout the course of the study. The results of these studies are shown in Fig. 1 and indicate that mice immunized with P3A DNA produced an anti-MCPS IgM antibody response that was significantly higher (P < 0·05) than that of mice immunized with P3C DNA (negative control). The increase of antibody levels in the negative control group, after subsequent boosts, are attributed to the administration of adjuvant, which enhances non-specific immune responses. The antibody response in the P3A DNA-immunized mice was higher than that determined for the MCPS-immunized group (positive control) after week 8, but this difference did not reach statistical significance. An anti-MCPS IgG antibody response was not detected in the P3A or P3C DNA groups, despite the presence of a T-cell epitope. A minimal anti-MCPS IgG3 antibody response (titres from 5 to 30) was detected in two MCPS-immunized mice on day 56, but the results were not statistically significant (data not shown). The immunization study was repeated three times, yielding similar antibody titre results on each occasion.

Figure 1.

Mice (n = 6) were immunized with 100 µg of DNA and 40 µg of aluminum phosphate gel adjuvant on days 0, 21, and 63. Positive controls were immunized with 5 µg of meningococcal serogroup C capsular polysaccharide (MCPS) on day 0. The anti-MCPS immunoglobulin response was determined by enzyme-linked immunosorbent assay (ELISA). The geometric mean immunoglobulin M (IgM) antibody titre of each group is represented by a column. *Statistical significance of groups compared with the negative control (P ≤ 0·05).

Bactericidal antibodies

To determine the functional activity of postimmune sera, bactericidal assays were performed. Immunization with the P3A DNA construct induced significantly greater bactericidal titres compared to mice immunized with P3C DNA (P = 0·002), on day 56 (Fig. 2). Bactericidal titres for mice immunized with P3A DNA were calculated and ranged from 8 to 256, whereas all mice immunized with P3C DNA had titres of < 4. Mice immunized with MCPS, as positive controls, also had significantly higher bactericidal titres (32–128) than the negative control (P = 0·005), but these titres were not significantly different from levels elicited by mice immunized with P3A DNA. These data indicate that immunization with P3A DNA induces antibodies that are functional and bactericidal against N. meningitidis serogroup C.

Figure 2.

Functional activity in postimmune sera was determined by serum bactericidal assays on day 56. Geometric mean bactericidal titres for each group are indicated by columns; individual mouse titres are denoted by circles. Some circles may be representative of more than one mouse in a group. *Statistical significance of groups compared with the negative control (P ≤ 0·05).

Protective efficacy of DNA-immunized mice

A challenge experiment was performed to further assess the functional activity of the immune response induced in mice by immunization with DNA. Groups of mice were immunized as described above. Mice immunized with MCPS were boosted with 5 µg of MCPS 7 days prior to challenge, since mice immunized with polysaccharide alone on day 0 died from meningitis. Mice received 1000 mg/kg iron dextran 1 week prior to challenge to enhance the susceptibility to meningococcal infections, and were subsequently challenged with a 10 × LD₅₀ dose of N. meningitidis serogroup C strain 35E. The results of these studies demonstrate that mice immunized with P3A DNA and MCPS were protected against meningococcal infection, with 100% survival. Five of the six mice immunized with P3C DNA (negative control) died within 24 hr postchallenge. This challenge was repeated and yielded similar results.

Antibody isotype profiles postchallenge

To evaluate the anti-MCPS immunoglobulin levels postchallenge among survivors in both P3A and P3C DNA groups, sera were collected 96 hr after challenge. ELISAs were performed to measure the antibody isotypes present in the sera collected. The results of these assays demonstrate no significant difference in anti-MCPS IgM levels of the survivors immunized with P3A DNA, MCPS and the one survivor immunized with P3C DNA (data not shown). Prior to challenge, IgG was not detectable in P3A DNA-immunized mice, whereas postchallenge anti-MCPS IgG were detected in all sera. ELISAs evaluating the different IgG isotypes in sera, demonstrated high levels of IgG1 and IgG3, and moderate levels of IgG2a and IgG2b, in mice immunized with P3A DNA (Fig. 3a). The survivor immunized with P3C DNA had no detectable IgG1, IgG2a or IgG2b, but a low level of IgG3 (Fig. 3b). Two mice immunized with MCPS demonstrated low levels of IgG3 prechallenge, whereas IgG1, IgG2b and IgG3 were only detected postchallenge in the majority of the sera (Fig. 3c). No anti-MCPS IgG2a was detected in survivors immunized with MCPS. Overall, the anti-MCPS IgG1, IgG2a, IgG2b and IgG3 titres were significantly higher in mice immunized with P3A DNA compared to the survivor immunized with P3C DNA (P = 0·003, 0·007, 0·007, and 0·014, respectively). In addition, postchallenge anti-MCPS IgG titres, of all isotypes tested, were significantly higher in mice immunized with P3A DNA compared to those immunized with MCPS (P = 0·026, 0·030, 0·0001, 0·003, respectively).

Figure 3.

Mice immunized with DNA and meningococcal serogroup C capsular polysaccharide (MCPS) were challenged with a lethal dose of Neisseria meningitidis serogroup C strain 35E. Ninety-six hours postchallenge, sera were collected from survivors and anti-MCPS IgG1, IgG2a, IgG2b, and IgG3 were determined by enzyme-linked immunosorbent assay (ELISA). Geometric mean pre- and postchallenge antibody titres for each isotype are represented by columns; individual mouse titres are denoted by circles. Some circles may be representative of more than one mouse. *Statistical significance of the test group compared with controls (P ≤ 0·05).

Discussion

Capsular polysaccharides induce a T-independent immune response characterized primarily by the presence of anti-polysaccharide IgM, lack of isotype switching and a memory response, and are poorly immunogenic in the very young.^59,60 Alternative approaches to eliciting a T-dependent response to carbohydrate antigens consist of conjugate vaccines, anti-Id vaccines and, potentially, DNA vectors that encode peptide mimics of the polysaccharide capsule. A peptide-based DNA immunization targeted to carbohydrate antigens has distinct advantages. First, peptide-encoded DNA vaccines have the ability to induce carbohydrate cross-reactive humoral and cellular immune responses.⁶¹ Second, DNA vaccines can redirect the immune response to a T helper 1 (Th1) response, which is characterized by the presence of IgG2a in mice.⁶² IgG2a is a complement fixing and opsonizing antibody that is crucial for preventing infection by a bacterial pathogen. It has previously been noted that the redirection of a T helper 2 (Th2) to a Th1 immune response, induced by a peptide-based DNA vaccine, may help to overcome the immune tolerance to polysaccharide antigens seen in neonates.⁶¹ Finally, several peptide mimics from different capsular polysaccharides may be incorporated into a single DNA vector in order to elicit anti-polysaccharide immunoglobulin targeted to other meningococcal serogroups.

In a multiepitope DNA construct, several factors must be taken into consideration with regard to design. First, B-cell epitopes must assume and be presented in the correct conformation in order to adopt the shape of the nominal antigen. Otherwise, peptide epitopes may induce antibodies that do not recognize the bacterial antigen. Second, short peptide sequences are unstable and poorly immunogenic in vivo. T-cell help is required to induce an optimal immune response. The polymorphism of HLA molecules, to which peptides bind, affects the magnitude and direction of the T-cell help provided. Therefore, a ‘universal’ Th epitope should be included in the multiepitope construct, eliminating HLA restriction. Lastly, linking multiple epitopes in a single construct may form deleterious flanking sequences from neighbouring epitopes, potentially suppressing the presentation of the target epitopes.^63–66 This includes determinants created at the junction between two neighbouring epitopes, when linking peptides in tandem.⁶⁷ Several studies present conflicting results concerning the role of neighbouring sequences on the efficiency of expression.^68,69

We have developed a multiple epitope DNA vaccine encoding a ‘universal’ T-cell helper epitope and a peptide that mimics the polysaccharide capsule of N. meningitidis serogroup C. This construct was designed to increase exogenous expression of our peptide mimic and to simultaneously induce T-cell help, thus stimulating both a humoral and a cellular immune response. The leader sequence assists in the targeting and transport of the B-cell epitope into the endoplasmic reticulum (ER), thereby enhancing translation and endogenous expression of the peptide.⁷⁰ Likewise, a T-cell helper epitope provides sufficient T-cell help to antibody-producing B cells. By inserting a ‘universal’ T-cell epitope, HLA restriction is avoided and MHC class I and II molecules are targeted with the same affinity.^71,72

In this study we have shown that immunization with a DNA construct encoding a carbohydrate peptide mimic and a ‘universal’ T-cell helper epitope induces protective levels of anti-MCPS immunoglobulin. The feasibility of inducing polysaccharide antibody responses through DNA immunization has been reported previously in several studies,^73,74 but the functional activity of the antibody response was not characterized. Primary and booster immunizations with the P3A DNA construct elicited anti-MCPS immunoglobulins with bactericidal activity. Mice immunized with P3C DNA produced significantly lower antibody titres that lacked detectable functional activity. Studies by Maslanka et al.⁷⁵ and Borrow et al.⁷⁶ have demonstrated that standard serogroup C ELISAs measure both low-avidity, non-functional antibodies as well as high-avidity antibodies that correlate with protection. In addition, a relationship has been shown, by Goldschneider et al.,¹³ between the presence of bactericidal antibodies and levels of protection against infection. It is probable that the P3C DNA vector induced low-avidity, non-specific and consequently non-bactericidal antibodies that cross-react with MCPS. Therefore, even though moderate levels of anti-MCPS immunoglobulin were elicited by P3C DNA mice, the bactericidal data that measured high-avidity, functional anti-MCPS immunoglobulins, was a more accurate measurement of future protection.

Subsequently following challenge with a 10 × LD₅₀ dose of live meningococci, mice immunized with the P3A DNA construct and MCPS survived. In contrast, only one out of six P3C-immunized mice (negative control) survived lethal challenge. A possible explanation for survival of this mouse is that meningococci tend to aggregate in solution. Despite our attempts to standardize the challenge procedure and dose, it is probable that the surviving mouse in the negative control group did not receive the 10 × LD₅₀ dose of meningococci. This suspicion is supported by the finding that 6 hr postchallenge, the mouse was not bacteraemic, whereas bacteria were detected in the blood of P3A- and MCPS-immunized mice.

Prechallenge antibody studies showed a predominant anti-MCPS IgM response, indicating the stimulation of a T-independent response. However, at 96 hr postchallenge, the anti-MCPS IgG1, IgG2a, IgG2b and IgG3 antibody titres were significantly greater in P3A DNA-immunized mice than in the survivors of both control groups. The absence of anti-MCPS IgG in P3A DNA-immunized mice, and low levels of IgG3 in MCPS mice prechallenge, suggests that IgM were protective against challenge, and that anti-MCPS IgG were not required for survival. The unexpected presence of various IgG isotypes in postchallenge sera from mice immunized with DNA indicates the induction of a mixed Th1 and Th2 immune response. Anti-MCPS IgG2a were detected only in mice immunized with the multiepitope DNA construct. The data also indicate that immunization with DNA resulted in a greater memory response in P3A DNA mice than in mice immunized with MCPS, owing to higher anti-MCPS IgG levels postchallenge. These results support previous observations, by other investigators, that DNA immunization can redirect the immune response from a Th2 response, primarily seen with polysaccharide vaccines, to a more desirable Th1 immune response.

In a T-independent immune response, carbohydrate antigens stimulate B cells by crosslinking surface IgM, leading to the production of antigen-directed antibodies. These antibodies are of relatively low affinity to antigen, with limited evidence for somatic hypermutation. Th cells support the differentiation and proliferation of B cells into plasma or memory B cells in a T-dependent immune response. This is characterized by somatic hypermutation and isotype switching from IgM to IgG or IgA antibodies. The immune response induced by our DNA construct was uniquely different from a ‘true’ T-independent or T-dependent response. Prior to challenge, characteristics of a T-independent response were detected, whereas postchallenge a T-dependent-like response was observed, with high levels of IgG and evidence of a memory response. Previous studies, using carbohydrate antigens, have described T-cell responses in a predominantly characterized T-independent response to the polysaccharide.^79–81 Studies from other investigators indicate that immunization with DNA encoding carbohydrate mimotopes could result in a T-dependent response with isotype switching.⁶² In addition, cellular response studies of carbohydrate-mimicking peptides indicate that mimotopes can activate T-cell subsets of Th1 and Th2 phenotypes.^61,74,82,83 Our results imply that B-cell memory and B-cell activation can occur without standard effector responses normally seen in a T-dependent immune response. This was also recently demonstrated by Laylor et al.,⁸⁴ where DNA vaccination resulted in a significant B-cell memory response in the absence of an overt effector response. Similarly, Klinman⁸⁵ reported that, in limiting-dilution fragment cultures, memory B-cell clones arise without detectable primary antibody production.

Previous immunization studies in our laboratory, using the MCPS peptide mimic complexed to proteosomes, resulted in a T-dependent immune response. However, the response to the same peptide, when encoded in a DNA vector, was markedly different. The results from our present study indicate that T-cell help is not adequately being provided prior to challenge. This may be attributed to the design and efficacy of the multiepitope DNA construct. First, the T-cell epitope included in the vector may not be effectively targeting T-cell help. A Th epitope from N. meningitidis may be more effective at stimulating Th cells and improving the immune response. Second, specific flanking sequences may be required for optimal expression of the T-cell epitope. These considerations will be addressed in the near future in order to optimize our DNA vector design.

We have shown that DNA immunization with a peptide mimic of MCPS results in a strong B-cell memory response in the absence of a strong effector response. We are currently expanding these studies and performing in-depth analysis of the nature of the immune response to DNA vaccines expressing carbohydrate peptide mimics. Our future results may influence the decisions regarding appropriate methods used for evaluating the efficacy of DNA vaccines.