This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Toropainen, M.
Right arrow Articles by Käyhty, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Toropainen, M.
Right arrow Articles by Käyhty, H.

 Previous Article  |  Next Article 

Infection and Immunity, October 2008, p. 4546-4553, Vol. 76, No. 10
0019-9567/08/$08.00+0     doi:10.1128/IAI.00418-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Pneumococcal Haemophilus influenzae Protein D Conjugate Vaccine Induces Antibodies That Inhibit Glycerophosphodiester Phosphodiesterase Activity of Protein D{triangledown}

Maija Toropainen,1* Anna Raitolehto,1 Isabelle Henckaerts,2 Dominique Wauters,2 Jan Poolman,2 Pascal Lestrate,2 and Helena Käyhty1

National Public Health Institute (KTL), Department of Vaccines, Helsinki, Finland,1 GlaxoSmithKline Biologicals, Rixensart, Belgium2

Received 4 April 2008/ Returned for modification 24 May 2008/ Accepted 11 July 2008


arrow
ABSTRACT
 
Haemophilus influenzae outer membrane protein D (PD) is a glycerophosphodiester phosphodiesterase (GlpQ) activity-possessing virulence factor and a promising vaccine antigen, providing 35.3% efficacy against acute otitis media caused by nontypeable H. influenzae (NTHI) when it was used as a carrier protein in a novel pneumococcal PD conjugate (Pnc-PD) vaccine. To study if PD-induced protection against NTHI could be due to antibodies that inhibit or neutralize its enzymatic activity, a GlpQ enzyme inhibition assay was developed, and serum samples collected from Finnish infants before and after Pnc-PD vaccination were analyzed for enzyme inhibition and anti-PD immunoglobulin G (IgG) antibody concentration. Before vaccination at age 2 months, the majority (84%) of infants (n = 69) had no detectable anti-PD IgG antibodies, and all were enzyme inhibition assay negative (inhibition index, <20). At age 13 to 16 months, all infants receiving three or four doses of Pnc-PD had detectable anti-PD IgG antibodies and 36% (8/22 infants) of the infants receiving three doses and 26% (6/23 infants) of the infants receiving four doses of Pnc-PD were inhibition assay positive (inhibition index, ≥20). No significant rise in anti-PD IgG antibodies or enzyme inhibition among control vaccinees (n = 24) receiving three doses of hepatitis B vaccine was detected. A modest correlation (rs, ~0.66) between anti-PD IgG concentration and enzyme inhibition was detected; however, their kinetics were clearly different. These data suggest that measurement of antibody responses that inhibit PD's enzymatic activity could be a useful tool for assessing Pnc-PD vaccine-induced protective immunity against NTHI.


arrow
INTRODUCTION
 
Unencapsulated, nontypeable Haemophilus influenzae (NTHI) is a frequent commensal of the human nasopharynx but is also the common cause of respiratory tract infections, such as otitis media (OM), sinusitis, bronchitis, and pneumonia (12, 22). Prevention of NTHI infections would provide considerable health and economic benefits. Thus, efforts have been directed toward identifying bacterial structures with potential as vaccine antigens. Of these, the outer membrane protein D (PD) is one of the most promising (25).

PD (also known as LPD) is a conserved 42-kDa outer membrane-associated lipoprotein (8). It belongs to the glycerophosphodiester phosphodiesterase (GlpQ) protein family and shows 78% amino acid similarity to the periplasmic nonlipidated GlpQ protein in Escherichia coli (21) and 90% amino acid similarity to the lipoprotein homologue in Pasteurella multocida (17). Similar to other members of this protein family, PD displays GlpQ activity, catalyzing the hydrolysis of glycerophosphodiesters to sn-glycerol 3-phosphate and the corresponding alcohol (21).

The hpd (or glpQ) gene, encoding lipo-PD, has been cloned and sequenced from several strains (4, 9, 29). The protein is genetically and antigenically conserved (4, 29) and is present in all typeable H. influenzae and NTHI strains tested thus far (3). Deviating from the nonlipidated GlpQ homologue in E. coli (15) and the lipidated GlpQ homologues in P. multocida (17) and Treponema pallidum (27), which are all located in the periplasm, in NTHI PD is proposed to be exposed to the cell surface (3).

The specific function(s) of PD is not known; however, previous in vivo and in vitro studies suggest that it is involved in NTHI pathogenesis. In an experimental rat OM model, a 100-fold higher concentration of PD-deficient mutant than PD-expressing wild-type bacteria was required to induce OM after direct injection of bacteria into the middle ear (10). Likewise, in a human nasopharyngeal tissue culture model using the same wild-type and mutated bacteria, the PD-deficient mutant caused much less damage to ciliated epithelial cells and loss of cilia than the wild-type, PD-expressing bacteria did (7). The mechanism(s) behind PD's virulence properties is not clear but may involve its GlpQ activity, either directly or indirectly (6).

Recently, a recombinant nonacylated form of PD (rPD) was used successfully as a novel carrier protein in a pneumococcal conjugate vaccine (Pnc-PD) (25). In a pediatric efficacy trial in the Czech Republic and in Slovakia, an efficacy of 35.3% (95% confidence interval [CI], 1.8% to 57.4%) against acute OM caused by NTHI was detected, associated with a 41.4% (95% CI, –4.9% to 67.3%) reduction in the nasopharyngeal NTHI carriage rate (25). The mechanism(s) for how PD induces protective immunity is currently unclear, but it seems to be antibody mediated, as passive immunization with a pediatric human serum pool generated against polysaccharide-PD conjugate vaccines conferred approximately 34% protection against the development of ascending NTHI-induced OM in a chinchilla viral-bacterial coinfection model (23).

The development of PD-based vaccines against NTHI would be facilitated if there was a functional assay correlating with protective efficacy. To study if PD-induced protection could be due to antibodies that inhibit, i.e., neutralize, its enzymatic activity, a GlpQ enzyme inhibition assay was developed, and pre- and postvaccination serum samples collected from infants given three or four doses of Pnc-PD vaccine during a previous immunogenicity and safety study in Finland (24) were analyzed for enzyme inhibition and anti-PD immunoglobulin G (IgG) antibody concentrations.


arrow
MATERIALS AND METHODS
 
Subjects, vaccines, vaccination, and sampling. The characteristics of the Finnish Pnc-PD conjugate vaccine safety and immunogenicity study participants, the vaccines, and the vaccination method have been described elsewhere (24). Of the total of 152 infants completing the whole study, 71 (47%) were included in the present study (Fig. 1). Group 1 vaccinees (Pnc-PD booster group) received the 11-valent Pnc-PD conjugate vaccine (GlaxoSmithKline Biologicals [GSK Bio], Rixensart, Belgium) at 2, 4, 6, and 12 to 15 months. Group 2 vaccinees (PncPS booster group) received Pnc-PD at 2, 4, and 6 months and a 23-valent pneumococcal polysaccharide vaccine (PncPS) (Pneumovax 23; Aventis Pasteur, Lyon, France) at 12 to 15 months. Group 3 vaccinees (control group) received hepatitis B vaccine (Engerix-B; GSK Bio) at 2, 4, and 6 months and Pnc-PD at 12 to 15 months. Diphtheria-tetanus-acellular pertussis-inactivated poliovirus-H. influenzae type b vaccine (Infanrix-Polio + Hib; GSK Bio) was given at 2, 4, and 6 months concomitantly with the Pnc-PD or the control vaccine, but at a separate injection site (opposite limb). Blood samples were obtained prior to dose 1 (2 months of age), 28 days after dose 3 (7 months of age), and immediately prior to (12 to 15 months of age) and 4 or 28 days after (12 to 15 or 13 to 16 months of age) the booster dose.


Figure 1
View larger version (14K):
[in this window]
[in a new window]

  		FIG. 1. Selection of serum samples for the present study. Pnc-PD, 11-valent pneumococcal Haemophilus influenzae PD conjugate vaccine; PncPS, 23-valent pneumococcal polysaccharide vaccine; HBV, hepatitis B vaccine.

Serum samples used in this study. For assessing immune responses to PD, serum samples taken prior to dose 1 (2 months of age) and immediately prior to (control group [n = 24]) or 28 days after (Pnc-PD group [n = 23] and PncPS group [n = 22]) the booster vaccination given at 12 to 15 months of age were used (Fig. 1). For assessing kinetics of antibody responses, serum samples taken from Pnc-PD (n = 6), PncPS (n = 6), and control (n = 6) vaccinees prior to dose 1 (at 2 months of age), 28 days after the primary series (at 7 months of age), immediately prior to the booster dose (at 12 to 15 months of age), and 28 days after the booster dose (at 13 to 16 months of age) were used (Fig. 1). The samples were stored at –20°C until they were analyzed.

For substrate titration studies, a serum pool was prepared from 10 inhibition assay-positive serum samples with sufficient volumes by mixing the sera at equal proportions.

Measurement of anti-PD IgG concentration by EIA. IgG antibodies to PD were measured by enzyme immunosorbent assay (EIA) as previously described, using nonacylated rPD as the coating antigen (25). The results were expressed as EIA units (U) per ml. The assay cutoff was 100 U/ml. For the calculation of geometric mean (GM) IgG antibody concentrations, samples with antibody concentrations of <100 U/ml were assigned a value of 30 U/ml.

Measurement of inhibition of GlpQ activity. (i) Chemicals and enzymes. β-NAD hydrate (NAD; Sigma-Aldrich), sn-glycero-3-phosphate dehydrogenase (GpDH [EC 1.1.1.8]; Fluka), L-{alpha}-glycerophosphorylcholine (GPC) (Sigma), and hydrazine monohydrate (Sigma-Aldrich) were all purchased from Sigma-Aldrich, and glycine (Merck) and calcium chloride (CaCl2; Merck) were purchased from VWR International. rPD was produced at GSK Bio as previously described (1).

(ii) GlpQ enzyme inhibition assay. To measure the inhibition of PD's GlpQ activity, twofold serial dilutions (1:2 to 1:32; 50 µl per well) of serum in 0.8 M hydrazine-0.2 M glycine-10 mM CaCl2 buffer, pH 9.0, were prepared in the wells of a 96-well microtiter plate (Corning Costar). rPD diluted in 0.8 M hydrazine-0.2 M glycine-10 mM CaCl2 buffer, pH 9.0, was added (0.5 µg per ml; 50 µl per well), and the plate was incubated for 1 h at 22°C.

After incubation, GlpQ activity was determined by a coupled photometric assay, essentially as described previously (14, 15), by adding 150 µl of a reaction mixture containing NAD, GpDH, and GPC in 0.8 M hydrazine-0.2 M glycine-10 mM CaCl2 buffer, pH 9.0, to each well. After 5 min of incubation at room temperature, the rate of NAD reduction was measured by recording the absorbance of each well at room temperature at 340 nm with a microplate reader (Multiskan MS; Thermo Fisher Scientific [formerly called Thermo Electron Corporation], Waltham, MA), using a kinetic processor (Ascent software; Thermo Fisher Scientific) with a reading interval of 1 min for a duration of 25 min. The final reaction concentrations in the 0.25-ml assay mixture were 0.5 mM NAD, 10 U/ml of GpDH, 0.8 mM GPC, and 0.1 µg/ml of rPD, unless otherwise stated. The assay and dilution buffer was 0.8 M hydrazine-0.2 M glycine-10 mM CaCl2, pH 9.0. Reaction buffer was made fresh every week (14), and the final reaction solutions were prepared fresh daily just before use from stock solutions stored at –70°C (NAD and GPC) or 4°C (GpDH and rPD). For the calculation of enzymatic activity (NADH production [pmol] per min), a molar absorbance coefficient of 6,300 M–1 cm–1 was used for NADH absorbance at 340 nm.

In the absence of serum (wells containing rPD plus reaction buffer), the enzymatic activity of rPD dropped during the 1-hour incubation period to about one-half that without the incubation step. This was prevented by the addition of normal human serum. Hence, as the reference wells to which the enzymatic activities of the pre- and postvaccination serum samples were compared, a twofold serially (1:2 to 1:32) diluted serum pool prepared from unimmunized infant serum without detectable anti-rPD IgG antibody was used in duplicate. The reagent control contained plain assay buffer without added rPD and serum, and the serum control contained reference serum or test serum (both at a 1:2 dilution) without rPD.

For calculation of the inhibition index, the background readings for serum control wells without PD were subtracted from those for the corresponding wells containing rPD. The inhibition index was thereafter calculated for each well as follows: ([GlpQ activityReference pool – GlpQ activityTest serum]/GlpQ activityReference pool) x 100.

On the basis of the distribution of inhibition indexes for serum samples taken before vaccination (99th percentile, 16.1; n = 69), the threshold for the inhibition assay was set to 20. Thus, sera with inhibition indexes of <20 at a 1:2 dilution were considered enzyme inhibition assay negative and sera with inhibition indexes of ≥20 were considered enzyme inhibition assay positive.

Statistical methods. In the statistical analysis of inhibition assay data, nonparametric tests were used, and in the analysis of anti-PD IgG concentrations, parametric tests were used after log transformation of data. For groupwise comparisons, inhibition indexes were subjected to Kruskal-Wallis nonparametric analysis of variance (ANOVA), followed by the Mann-Whitney U test, if indicated. Differences in pre- and postvaccination inhibition indexes were analyzed with the Wilcoxon signed rank test. In groupwise comparisons of anti-PD IgG data, one-way ANOVA was used, followed by Tukey's honestly significant difference test, if indicated. Differences in pre- and postvaccination anti-PD IgG concentrations were analyzed with the paired t test, and differences in anti-PD IgG concentrations among inhibition assay-positive and inhibition assay-negative serum samples were analyzed with the unpaired t test. Spearman's correlation coefficient (rs) was used to assess the relationship between the inhibition index and anti-PD IgG concentration. For all comparisons, P values of <0.05 were considered significant. All analyses were done with SPSS software (SPSS 15.0; SPSS Inc., Chicago, IL).


arrow
RESULTS
 
Anti-PD IgG concentrations prior to and after Pnc-PD vaccination. To assess the immune response to vaccination, serum samples taken prior to vaccination (at 2 months of age) and 28 days after booster vaccination (at 13 to 16 months of age), when Pnc-PD booster group vaccinees had received four doses and PncPS booster group vaccinees had received three doses of Pnc-PD conjugate (Fig. 1), were first analyzed for anti-PD IgG concentration by EIA. As a control, serum samples from control vaccinees taken prior to vaccination (at 2 months of age) and immediately prior to the booster dose at 12 to 15 months of age, before any Pnc-PD vaccination (Fig. 1), were used.

Prior to vaccination at 2 months of age, the majority (84%) of the infants did not have detectable concentrations (≥100 U/ml) of IgG antibodies against PD (Fig. 2a to c). The GM antibody concentrations for the three study groups varied from 33 to 48 U/ml, with no statistically significant differences between the groups (P = 0.23; one-way ANOVA). Twenty-eight days after booster vaccination, all Pnc-PD (Fig. 2a) and PncPS (Fig. 2b) booster group vaccinees had detectable concentrations of antibodies to PD. The GM (95% CI) antibody concentrations increased 130-fold, from 33 U/ml (27 to 39 U/ml) to 4,302 U/ml (2,979 to 6,213 U/ml), and 61-fold, from 44 U/ml (28 to 68 U/ml) to 2,677 U/ml (1,712 to 4,185 U/ml), in the Pnc-PD and PncPS booster groups, respectively, compared to prevaccination concentrations at 2 months of age. For both groups, the increase in anti-PD IgG concentration was statistically significant (P < 0.01; paired t test), and the postbooster GM antibody concentrations were >66-fold higher (P < 0.01; Tukey's honestly significant difference test) than those in control group samples taken at 12 to 15 months of age. In the control group, no significant rise in anti-PD IgG concentration from age 2 months to 12 to 15 months was detected (GM [95% CI], 48 U/ml [32 to 71 U/ml] versus 40 U/ml [33 to 49 U/ml]; P = 0.15 [paired t test]) (Fig. 2c).


Figure 2
View larger version (28K):
[in this window]
[in a new window]

  		FIG. 2. Enzyme inhibition indexes and anti-PD IgG antibody concentrations before and after Pnc-PD conjugate vaccination. (Top) Anti-PD IgG antibody concentrations. (Bottom) Inhibition indexes at 1:2 serum dilution. Serum samples were taken prior to dose 1 (at 2 months of age) and immediately prior to (control group; at 12 to 15 months of age) or 28 days after (at 13 to 16 months of age) the booster dose, when the Pnc-PD group (n = 23) had received four doses, the PncPS booster group (n = 22) had received three doses, and the control group (n = 24) had received zero doses of Pnc-PD conjugate. GM anti-PD IgG concentrations at 2 and 12 to 15 or 13 to 16 months of age and median inhibition indexes at 2 and 12 to 15 or 13 to 16 months of age are shown. % ≥20, percentage of enzyme inhibition assay-positive sera at 2 and 12 to 15 or 13 to 16 months of age. The overlapping lines in panel c were separated (i.e., jittered) by ±3% to 15% to show them individually.

Enzyme inhibition indexes prior to and after Pnc-PD vaccination. Next, the serum samples were analyzed in the enzyme inhibition assay. Prior to vaccination at 2 months of age, all infant sera were enzyme inhibition assay negative (inhibition index of <20 at a serum dilution of 1:2) (Fig. 2d to f). The median inhibition index varied from 2.6 to 3.8, with no statistically significant differences between the study groups (P = 0.55; nonparametric ANOVA). Twenty-eight days after booster vaccination, 26% (6/23 vaccinees) of the Pnc-PD booster group vaccinees (receiving four doses of Pnc-PD conjugate) (Fig. 2d) and 36% (8/22 vaccinees) of the PncPS booster group vaccinees (receiving three doses of Pnc-PD conjugate) (Fig. 2e) were enzyme inhibition assay positive. The median (25th, 75th percentiles) inhibition index increased 5.0-fold, from 2.6 (1.0, 4.1) to 12.9 (8.5, 21.9), and 3.2-fold, from 3.6 (0, 9.2) to 11.5 (4.8, 36.2), in the Pnc-PD and PncPS booster groups, respectively, compared to prevaccination levels at 2 months of age. For both groups, the increase in inhibition index was statistically significant (P < 0.05; Wilcoxon signed rank test) and the median postbooster inhibition index was >3.4-fold higher (P < 0.05; Mann-Whitney U test) than that for control group samples taken immediately prior to the first Pnc-PD vaccination at 12 to 15 months of age. In the control group, no rise in inhibition index from 2 months to 12 to 15 months of age was detected (median inhibition index [25th, 75th percentiles], 3.8 [1.6, 6.2] versus 3.3 [0.9, 9.1]; P = 0.46 [Wilcoxon signed rank test]) (Fig. 2f), and all vaccinees remained enzyme inhibition assay negative. Thus, similar to the case for anti-PD IgG concentration, there was a significant increase in enzyme inhibition index after three or four doses of Pnc-PD conjugate.

Correlation between inhibition index and anti-PD IgG concentration. To study the association between enzyme inhibition and antibody concentration, postbooster inhibition indexes were plotted against anti-PD IgG concentrations, and the GM anti-PD IgG concentrations among inhibition assay-positive and inhibition assay-negative sera were calculated. There was a moderate, statistically significant positive correlation between inhibition indexes and anti-PD IgG concentrations in the Pnc-PD (rs = 0.67; P < 0.001) and PncPS (rs = 0.66; P = 0.001) booster groups (Fig. 3). In both vaccine groups, the GM IgG concentration was >2.8-fold higher among inhibition assay-positive sera than among inhibition assay-negative sera (P < 0.001; unpaired t test), though especially in the Pnc-PD booster group, some sera remained enzyme inhibition assay negative despite high anti-PD IgG concentrations.


Figure 3
View larger version (8K):
[in this window]
[in a new window]

  		FIG. 3. Correlation between inhibition index and anti-PD IgG antibody concentration. Serum samples were taken 28 days after the booster dose at 13 to 16 months of age, when Pnc-PD booster group vaccinees (n = 23) had received four doses and PncPS booster group vaccinees (n = 22) had received three doses of Pnc-PD conjugate. GM anti-PD IgG concentrations among enzyme inhibition assay-positive (inhibition index, ≥20) or enzyme inhibition assay-negative (inhibition index, <20) serum samples are shown. rs, Spearman's correlation coefficient.

Kinetics of inhibition index and anti-PD IgG concentration. To study the association between inhibition index and anti-PD IgG concentration in more detail, serum samples from six Pnc-PD and six Pnc-PS booster vaccinees taken prior to vaccination (at 2 months of age), 28 days after the primary series (at 7 months of age), and immediately prior to and 28 days after (from vaccinees 2 to 12 only) booster vaccination (given at 12 to 15 months of age) were analyzed for enzyme inhibition and anti-PD IgG concentration. As a control, serum samples from six randomly selected control vaccinees were included in the analysis.

Prior to vaccination at 2 months of age, none of the samples from Pnc-PD booster group vaccinees were enzyme inhibition assay positive, and none had detectable IgG antibodies to PD (Fig. 4a, vaccinees 1 to 6). Twenty-eight days after the primary series (at 7 months of age), the inhibition index rose clearly in all but one (vaccinee 3) of the six Pnc-PD vaccinees selected for kinetic studies, associated with a concomitant increase in anti-PD IgG concentration. Prior to booster doses at 12 to 15 months of age, the inhibition indexes increased further in all but two vaccinees (vaccinees 3 and 6). At the same time, the anti-PD IgG concentrations remained essentially the same or slightly decreased in all but one vaccinee (vaccinee 5), whose antibody concentration increased 1.4-fold (10,500 to 14,900 U/ml) compared to the concentration after the primary series (at 7 months of age). Surprisingly, no increase in inhibition index was detected in the four postbooster samples (vaccinees 3 to 6) taken 28 days after the booster vaccination (at 13 to 16 months of age). At the same time, the anti-PD IgG concentrations increased in these four Pnc-PD-boosted vaccinees, by 1.4- to 3.3-fold. In the Pnc-PS booster group, similar clear increases in inhibition index with constant or decreasing antibody concentrations for post-primary series (at 7 months of age) to prebooster (at 12 to 15 months of age) samples were detected in four (vaccinees 7, 8, 9, and 11) of the six vaccinees included in the kinetic analysis (Fig. 4b, vaccinees 7 to 12); for vaccinee 7, the increases in antibody concentration and enzyme inhibition from prebooster (at 12 to 15 months of age) to postbooster (at 13 to 16 months of age) vaccination were probably due to a natural encounter with H. influenzae. In the control group, no major changes in anti-PD IgG concentrations (Fig. 4c) or inhibition indexes (Fig. 4d) were detected during the entire 11- to 14-month follow-up period. Thus, despite a moderate positive correlation between inhibition index and anti-PD IgG concentration postbooster, their kinetics were clearly different.


Figure 4
View larger version (15K):
[in this window]
[in a new window]

  		FIG. 4. Kinetics of vaccine responses. Serum samples were taken prior to dose 1 (at 2 months of age), 28 days (at 7 months of age) after dose 3, and immediately prior to (at 12 to 15 months of age) and 28 days after (at 13 to 16 months of age) the booster dose. (a) Anti-PD IgG antibody concentrations (•) and inhibition indexes ({circ}) at a 1:2 serum dilution for six Pnc-PD group vaccinees. (b) Anti-PD IgG antibody concentrations (•) and inhibition indexes ({circ}) at a 1:2 serum dilution for six PncPS group vaccinees. Anti-PD IgG antibody concentrations (c) and inhibition indexes (d) at a 1:2 serum dilution are also given for six control vaccinees.

Inhibition index as a function of serum dilution. Figure 5 shows the inhibition indexes of all inhibition assay-positive sera (n = 14) as a function of serum dilution. There was a linear relationship between inhibition index and serum dilution. For four samples, inhibition was detected up to a 1:16 dilution, and for two samples, it was detected up to the highest dilution (1:32) tested.


Figure 5
View larger version (17K):
[in this window]
[in a new window]

  		FIG. 5. Postvaccination inhibition indexes in inhibition assay-positive sera as a function of serum dilution. Serum samples (n = 14) were taken 28 days after the Pnc-PD or PncPS booster at 13 to 16 months of age, when the vaccinees had received three or four doses of Pnc-PD conjugate.

Mechanism of postvaccination serum-induced enzyme inhibition. To study the nature of enzyme inhibition induced by anti-PD antibodies, a serum pool was prepared from inhibition assay-positive samples, and its enzyme inhibitory capacity was tested at a 1:4 dilution at different substrate concentrations (0.06 to 40 times the PD Km). After reaching a maximum value at a substrate concentration of about 1.6 mM, no significant increase in enzymatic activity up to a concentration of 6.4 mM was detected (Fig. 6), suggesting that the neutralizing antibodies did not inhibit PD's enzymatic activity by binding to the substrate binding site directly.


Figure 6
View larger version (10K):
[in this window]
[in a new window]

  		FIG. 6. Enzymatic activity of PD in the presence and absence of a postvaccination serum pool at different substrate concentrations. The postvaccination serum pool was prepared from inhibition assay-positive (n = 10) postbooster serum samples with sufficient volumes by mixing the sera at equal proportions, and its inhibitory activity was tested at a 1:4 dilution.

Enzyme inhibition assay reproducibility. After sensitizing the inhibition assay by reducing the amount of PD to one-half (final concentration, 0.05 µg/ml) of the original concentration to decrease serum consumption, a set of six inhibition assay-positive postbooster samples with different inhibition indexes was tested at a 1:4 dilution on three separate days to assess interassay reproducibility. The reproducibility of the assay was good (Fig. 7), with a mean coefficient of variation of 19.7% for interassay comparisons.


Figure 7
View larger version (26K):
[in this window]
[in a new window]

  		FIG. 7. Reproducibility of enzyme inhibition assay. Six (a to f) inhibition assay-positive postbooster samples from the Pnc-PD and PncPS booster groups were tested on three separate days at a 1:4 dilution.


arrow
DISCUSSION
 
Studies addressing the immunogenicity of novel PD conjugate vaccines would be greatly facilitated if there was a relevant and robust test to assess the functionality of anti-PD antibodies alongside conventional quantitative antibody assays. In the present study, a novel enzyme inhibition assay was developed to study if antibody to PD could inhibit or neutralize its GlpQ activity, and pre- and postvaccination serum samples collected from infants vaccinated with the Pnc-PD conjugate or a control hepatitis B vaccine were evaluated. The results presented show that this application was entirely feasible, suggesting that the enzyme inhibition assay could be a useful tool in assessing the immunogenicity of Pnc-PD conjugates.

Despite the amount of work done on vaccines to prevent OM caused by NTHI, the mechanism for how systematic immunization prevents infection at mucosal surfaces still remains obscure and may vary depending on the target antigen. In previous studies, the lack of circulating bactericidal antibodies to NTHI has been associated with susceptibility to OM caused by this organism (5, 28). While moderate bactericidal antibody responses in experimental animals have been detected following PD immunization, their existence and role in protection in humans remain unclear (1, 2). Our efforts to detect opsonophagocidal activity in sera from infants immunized with Pnc-PD have also been unsuccessful (our unpublished data). Thus, in the present study, a completely different approach was chosen to assess the possible biological activity of anti-PD antibodies. We detected GlpQ neutralizing antibody responses in 26% of the Pnc-PD booster group vaccinees and in 36% of the PncPS booster group vaccinees compared to none of the control vaccinees. These results coincide well with the results of a recent Pnc-PD efficacy trial in the Czech Republic and Slovakia, where an efficacy of 35.3% against acute OM caused by NTHI was detected. These results are also in line with previous passive protection studies with chinchillas, where a pediatric human serum pool generated from Pnc-PD vaccinees conferred ~34% protection against OM after intranasal challenge with NTHI (23).

While we found a moderate positive correlation between inhibition index and anti-PD IgG concentration at 28 days postbooster, some sera, especially those from the Pnc-PD booster group, remained enzyme inhibition negative despite high anti-PD IgG levels. The antibody kinetics measured by the two assays also differed. These differences were most evident during the period from 7 months (post-primary series) to 12 to 15 months (prebooster) of age, when there was a clear rise in inhibition index without a concomitant rise in anti-PD IgG antibody concentration in two-thirds of the Pnc-PD booster group and the Pnc-PS booster group sera analyzed, and from 12 to 15 months (prebooster) to 13 to 16 months (postbooster) of age, when there was no increase in inhibition index among the four Pnc-PD booster group sera analyzed, despite a moderate rise in anti-PD IgG antibody concentration. The reason for the late increase in enzyme inhibition compared to anti-PD IgG levels after the primary vaccination series is unclear but could be due to affinity maturation of the antibody response, resulting in antibodies with increased functional, i.e., inhibitory, activity. Alternatively, the specificities of anti-PD antibodies in serum samples taken at the later time point and those taken 1 month after the third immunization may be different. In mice infected with the protozoan parasite Trypanosoma cruzi, nonneutralizing antibodies to the immunodominant domain of the trans-sialidase were detected as early as 8 to 10 days after infection, while neutralizing antibodies directed to the enzymatic domain of the protein were not detectable until 30 days after infection (16).

It was interesting that no significant differences in enzyme inhibition at 28 days postbooster between the two vaccine groups receiving three (PncPS booster group) or four (Pnc-PD booster group) doses of Pnc-PD vaccine were detected. This could have been due to a lack of a booster antibody response in the Pnc-PD group vaccinees in general. However, the approximately 1.6-fold higher GM anti-PD antibody concentration in the Pnc-PD booster group than in the PncPS booster group postbooster and the clear rise in anti-PD IgG concentration from prebooster to postbooster among Pnc-PD vaccinees (our unpublished observations) question this possibility. Alternatively, taking into consideration the late maturation of enzyme inhibition compared to anti-PD IgG concentration after the primary vaccination series, the time point used for the collection of postbooster samples (28 days after the booster dose) might not have been optimal for enzyme inhibition studies. In future, additional later sampling time points after booster vaccination might be useful to assess the significance of the booster dose for enzyme inhibition responses in general.

In an experiment assessing the effect of increasing substrate concentration on enzyme activity in the presence of inhibitor (a serum pool prepared from postbooster samples collected after three or four doses of Pnc-PD), no significant rise in PD's enzymatic activity up to a concentration of 6.4 mM (~40 times the PD Km) was detected. This suggests that the inhibitory antibodies did not directly bind and block the substrate binding site but instead bound to some different epitope and either changed the conformation of PD or disturbed its function by steric hindrance. Additional studies are needed to study the nature of anti-PD antibody-mediated enzyme inhibition in more detail.

We did not address the exact mechanism for how neutralizing antibody to PD might provide protection in vivo against OM. This could be due to prevention of GlpQ's deleterious effects on host glycophospholipids or due to interference with bacterial glycerophospholipid metabolism. Alternatively, or additionally, neutralizing antibodies to PD might interrupt NTHI colonization indirectly by interfering with PD-mediated acquisition and incorporation of phosphorylcholine (ChoP) into bacterial lipopolysaccharide (LPS) (6).

Similar to other bacterial species that inhabit the human respiratory tract, including Streptococcus pneumoniae (11, 20) and Neisseria spp. (13, 26), for NTHI the expression of ChoP on the bacterial surface appears to be important for the adherence and persistence of bacteria on the mucosal membranes, probably through mediating bacterial adherence to the receptor for platelet activating factor and by increasing NTHI resistance to the host antimicrobial peptides expressed in the upper respiratory tract (19, 30, 31, 33). In H. influenzae, the expression of ChoP on LPS is phase variable and under the control of four genes, namely, licA, -B, -C, and -D, of the lic-1 locus (32). Similar to S. pneumoniae, H. influenzae lacks a pathway for de novo synthesis of choline and acquires it exclusively from the environment. Of the four proteins encoded by the lic-1 locus, LicB, showing similarity to eukaryotic choline permeases, and LicA, a phase-variable choline kinase, have been postulated to be responsible for the transport of choline from the environment and its subsequent phosphorylation, respectively (34). A pyrophosphorylase, LicC, then catalyzes the formation of nucleoside diphosphocholine from choline phosphate and a nucleoside triphosphate, while LicD, a putative diphosphonucleoside choline transferase (18), has been postulated to be responsible for the transfer of ChoP from choline diphosphonucleoside to its final location on LPS (34). In the absence of free choline, NTHI is able to use GPC, an abundant degradation product of host phospholipids also used as the substrate in the present enzyme inhibition studies, as an alternative source of choline (6). This requires the presence of PD but also an in-frame translation of the licA gene (6), suggesting that PD's role might be the provision of free choline through its GlpQ activity, catalyzing the hydrolysis of GPC to glycerol-3-phosphate and choline. By inhibiting the utilization of GPC as the source of choline, and hence the decoration of LPS with ChoP, antibody to PD may thus interfere with the colonization of NTHI on respiratory mucosa and prevent the spread of bacteria through the Eustachian tube into the middle ear. In the future, studies addressing GlpQ's substrate specificity could be useful to assess the significance and biological function of PD for NTHI in general and to estimate how antibody-induced inhibition of PD's GlpQ activity relates to protection. The ability of purified rPD- or PD-expressing lic-1 deletion mutant bacteria to cause damage to ciliated epithelial cells could also be evaluated in the absence and presence of specific antibodies to gain a more comprehensive view of the direct and indirect roles of PD in NTHI virulence and the possible protective mechanism of neutralizing antibodies.

To conclude, a simple and reproducible enzyme immunoinhibition assay was developed, allowing the detection of functional antibody responses to PD after Pnc-PD conjugate vaccination. Potentially, this application could be extended to other bacterial vaccine candidates possessing enzymatic activity as well. Additional studies are needed to verify the value of a GlpQ enzyme inhibition assay alongside other potential functional antibody assays, such as serum bactericidal activity assay, as a correlate for Pnc-PD vaccine-induced protection against NTHI.


arrow
ACKNOWLEDGMENTS
 
Matti Sarvas is thanked for helpful discussions on enzyme assays, and Mika Lahdenkari is thanked for help with statistical analysis.

This study was supported by funding from GlaxoSmithKline Biologicals, Rixensart, Belgium.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: National Public Health Institute, Department of Vaccines, Vaccine Immunology Laboratory, Mannerheimintie 166, FIN-00300 Helsinki, Finland. Phone: 358-9-47448873. Fax: 358-9-47448599. E-mail: maija.toropainen{at}ktl.fi Back

{triangledown} Published ahead of print on 21 July 2008. Back

Editor: A. Camilli


arrow
REFERENCES
 
    1
  1. Akkoyunlu, M., H. Janson, M. Ruan, and A. Forsgren. 1996. Biological activity of serum antibodies to a nonacylated form of lipoprotein D of Haemophilus influenzae. Infect. Immun. 64:4586-4592.[Abstract]
    2. 2
  2. Akkoyunlu, M., A. Melhus, C. Capiau, O. van Opstal, and A. Forsgren. 1997. The acylated form of protein D of Haemophilus influenzae is more immunogenic than the nonacylated form and elicits an adjuvant effect when it is used as a carrier conjugated to polyribosyl ribitol phosphate. Infect. Immun. 65:5010-5016.[Abstract]
    3. 3
  3. Akkoyunlu, M., M. Ruan, and A. Forsgren. 1991. Distribution of protein D, an immunoglobulin D-binding protein, in Haemophilus strains. Infect. Immun. 59:1231-1238.[Abstract/Free Full Text]
    4. 4
  4. Duim, B., P. Ruiter, L. D. Bowler, J. Dankert, and L. van Alphen. 1997. Sequence variation in the hpd gene of nonencapsulated Haemophilus influenzae isolated from patients with chronic bronchitis. Gene 191:57-60.[CrossRef][Medline]
    5. 5
  5. Faden, H., J. Bernstein, L. Brodsky, J. Stanievich, D. Krystofik, C. Shuff, J. J. Hong, and P. L. Ogra. 1989. Otitis media in children. I. The systemic immune response to nontypable Haemophilus influenzae. J. Infect. Dis. 160:999-1004.[Medline]
    6. 6
  6. Fan, X., H. Goldfine, E. Lysenko, and J. N. Weiser. 2001. The transfer of choline from the host to the bacterial cell surface requires glpQ in Haemophilus influenzae. Mol. Microbiol. 41:1029-1036.[CrossRef][Medline]
    7. 7
  7. Janson, H., B. Carlén, A. Cervin, A. Forsgren, A. B. Magnusdottir, S. Lindberg, and T. Runer. 1999. Effects on the ciliated epithelium of protein D-producing and -nonproducing nontypeable Haemophilus influenzae in nasopharyngeal tissue cultures. J. Infect. Dis. 180:737-746.[CrossRef][Medline]
    8. 8
  8. Janson, H., L. O. Heden, and A. Forsgren. 1992. Protein D, the immunoglobulin D-binding protein of Haemophilus influenzae, is a lipoprotein. Infect. Immun. 60:1336-1342.[Abstract/Free Full Text]
    9. 9
  9. Janson, H., L. O. Heden, A. Grubb, M. R. Ruan, and A. Forsgren. 1991. Protein D, an immunoglobulin D-binding protein of Haemophilus influenzae: cloning, nucleotide sequence, and expression in Escherichia coli. Infect. Immun. 59:119-125.[Abstract/Free Full Text]
    10. 10
  10. Janson, H., A. Melhus, A. Hermansson, and A. Forsgren. 1994. Protein D, the glycerophosphodiester phosphodiesterase from Haemophilus influenzae with affinity for human immunoglobulin D, influences virulence in a rat otitis model. Infect. Immun. 62:4848-4854.[Abstract/Free Full Text]
    11. 11
  11. Kharat, A. S., and A. Tomasz. 2006. Drastic reduction in the virulence of Streptococcus pneumoniae expressing type 2 capsular polysaccharide but lacking choline residues in the cell wall. Mol. Microbiol. 60:93-107.[CrossRef][Medline]
    12. 12
  12. Klein, J. O. 1997. Role of nontypeable Haemophilus influenzae in pediatric respiratory tract infections. Pediatr. Infect. Dis. J. 16:S5-S8.[CrossRef][Medline]
    13. 13
  13. Kolberg, J., E. A. Hoiby, and E. Jantzen. 1997. Detection of the phosphorylcholine epitope in streptococci, Haemophilus and pathogenic Neisseriae by immunoblotting. Microb. Pathog. 22:321-329.[CrossRef][Medline]
    14. 14
  14. Lang, G. 1984. l-(–)Glycerol 3-phosphate, p. 525-531. In H. U. Bergmeyer, J. Bergmeyer, and M. Grassl (ed.), Methods of enzymatic analysis, 3rd ed., vol. VI. Verlag Chemie, Weinheim, Germany.
    15. 15
  15. Larson, T. J., M. Ehrmann, and W. Boos. 1983. Periplasmic glycerophosphodiester phosphodiesterase of Escherichia coli, a new enzyme of the glp regulon. J. Biol. Chem. 258:5428-5432.[Abstract/Free Full Text]
    16. 16
  16. Leguizamon, M. S., O. E. Campetella, S. M. Gonzalez Cappa, and A. C. Frasch. 1994. Mice infected with Trypanosoma cruzi produce antibodies against the enzymatic domain of trans-sialidase that inhibit its activity. Infect. Immun. 62:3441-3446.[Abstract/Free Full Text]
    17. 17
  17. Lo, M., J. D. Boyce, I. W. Wilkie, and B. Adler. 2004. Characterization of two lipoproteins in Pasteurella multocida. Microbes Infect. 6:58-67.[CrossRef][Medline]
    18. 18
  18. Lysenko, E., J. C. Richards, A. D. Cox, A. Stewart, A. Martin, M. Kapoor, and J. N. Weiser. 2000. The position of phosphorylcholine on the lipopolysaccharide of Haemophilus influenzae affects binding and sensitivity to C-reactive protein-mediated killing. Mol. Microbiol. 35:234-245.[CrossRef][Medline]
    19. 19
  19. Lysenko, E. S., J. Gould, R. Bals, J. M. Wilson, and J. N. Weiser. 2000. Bacterial phosphorylcholine decreases susceptibility to the antimicrobial peptide LL-37/hCAP18 expressed in the upper respiratory tract. Infect. Immun. 68:1664-1671.[Abstract/Free Full Text]
    20. 20
  20. Mosser, J. L., and A. Tomasz. 1970. Choline-containing teichoic acid as a structural component of pneumococcal cell wall and its role in sensitivity to lysis by an autolytic enzyme. J. Biol. Chem. 245:287-298.[Abstract/Free Full Text]
    21. 21
  21. Munson, R. S., Jr., and K. Sasaki. 1993. Protein D, a putative immunoglobulin D-binding protein produced by Haemophilus influenzae, is glycerophosphodiester phosphodiesterase. J. Bacteriol. 175:4569-4571.[Abstract/Free Full Text]
    22. 22
  22. Murphy, T. F. 2003. Respiratory infections caused by non-typeable Haemophilus influenzae. Curr. Opin. Infect. Dis. 16:129-134.[Medline]
    23. 23
  23. Novotny, L. A., J. A. Jurcisek, F. Godfroid, J. T. Poolman, P. A. Denoel, and L. O. Bakaletz. 2006. Passive immunization with human anti-protein D antibodies induced by polysaccharide protein D conjugates protects chinchillas against otitis media after intranasal challenge with Haemophilus influenzae. Vaccine 24:4804-4811.[CrossRef][Medline]
    24. 24
  24. Nurkka, A., J. Joensuu, I. Henckaerts, P. Peeters, J. Poolman, T. Kilpi, and H. Käyhty. 2004. Immunogenicity and safety of the eleven valent pneumococcal polysaccharide-protein D conjugate vaccine in infants. Pediatr. Infect. Dis. J. 23:1008-1014.[CrossRef][Medline]
    25. 25
  25. Prymula, R., P. Peeters, V. Chrobok, P. Kriz, E. Novakova, E. Kaliskova, I. Kohl, P. Lommel, J. Poolman, J. P. Prieels, and L. Schuerman. 2006. Pneumococcal capsular polysaccharides conjugated to protein D for prevention of acute otitis media caused by both Streptococcus pneumoniae and non-typable Haemophilus influenzae: a randomised double-blind efficacy study. Lancet 367:740-748.[CrossRef][Medline]
    26. 26
  26. Serino, L., and M. Virji. 2000. Phosphorylcholine decoration of lipopolysaccharide differentiates commensal Neisseriae from pathogenic strains: identification of licA-type genes in commensal Neisseriae. Mol. Microbiol. 35:1550-1559.[CrossRef][Medline]
    27. 27
  27. Shevchenko, D. V., T. J. Sellati, D. L. Cox, O. V. Shevchenko, E. J. Robinson, and J. D. Radolf. 1999. Membrane topology and cellular location of the Treponema pallidum glycerophosphodiester phosphodiesterase (GlpQ) ortholog. Infect. Immun. 67:2266-2276.[Abstract/Free Full Text]
    28. 28
  28. Shurin, P. A., S. I. Pelton, I. B. Tager, and D. L. Kasper. 1980. Bactericidal antibody and susceptibility to otitis media caused by nontypable strains of Haemophilus influenzae. J. Pediatr. 97:364-369.[CrossRef][Medline]
    29. 29
  29. Song, X. M., A. Forsgren, and H. Janson. 1995. The gene encoding protein D (hpd) is highly conserved among Haemophilus influenzae type b and nontypeable strains. Infect. Immun. 63:696-699.[Abstract]
    30. 30
  30. Swords, W. E., B. A. Buscher, K. Ver Steeg II, A. Preston, W. A. Nichols, J. N. Weiser, B. W. Gibson, and M. A. Apicella. 2000. Non-typeable Haemophilus influenzae adhere to and invade human bronchial epithelial cells via an interaction of lipooligosaccharide with the PAF receptor. Mol. Microbiol. 37:13-27.[CrossRef][Medline]
    31. 31
  31. Tong, H. H., L. E. Blue, M. A. James, Y. P. Chen, and T. F. DeMaria. 2000. Evaluation of phase variation of nontypeable Haemophilus influenzae lipooligosaccharide during nasopharyngeal colonization and development of otitis media in the chinchilla model. Infect. Immun. 68:4593-4597.[Abstract/Free Full Text]
    32. 32
  32. Weiser, J. N., J. M. Love, and E. R. Moxon. 1989. The molecular mechanism of phase variation of H. influenzae lipopolysaccharide. Cell 59:657-665.[CrossRef][Medline]
    33. 33
  33. Weiser, J. N., N. Pan, K. L. McGowan, D. Musher, A. Martin, and J. Richards. 1998. Phosphorylcholine on the lipopolysaccharide of Haemophilus influenzae contributes to persistence in the respiratory tract and sensitivity to serum killing mediated by C-reactive protein. J. Exp. Med. 187:631-640.[Abstract/Free Full Text]
    34. 34
  34. Weiser, J. N., M. Shchepetov, and S. T. Chong. 1997. Decoration of lipopolysaccharide with phosphorylcholine: a phase-variable characteristic of Haemophilus influenzae. Infect. Immun. 65:943-950.[Abstract]

Infection and Immunity, October 2008, p. 4546-4553, Vol. 76, No. 10
0019-9567/08/$08.00+0     doi:10.1128/IAI.00418-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Toropainen, M.
Right arrow Articles by Käyhty, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Toropainen, M.
Right arrow Articles by Käyhty, H.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»