SEARCH

SEARCH BY CITATION

Keywords:

  • BCG vaccination;
  • Chemokines;
  • Cytokines;
  • Heparin-binding haemagglutinin (HBHA);
  • tuberculosis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

Heparin-binding haemagglutinin (HBHA)-specific immune responses have been linked to protection against tuberculosis (TB). We investigated the hypothesis that BCG vaccination of human infants primes an HBHA-specific response, using multiplex to measure secreted cytokines and chemokines following HBHA and Mycobacterium tuberculosis purified protein derivative (PPD) stimulation of diluted whole blood samples from BCG-vaccinated or -unvaccinated infants. Of 42 analytes measured, 24 and 32 significant, BCG-associated increases were detected in response to HBHA and PPD, respectively. Both response profiles included Th-1, Th-2, Th-17 and inflammatory cytokines and chemokines (e.g. IFN-γ, TNF-α, IL-5, IL-10, IL-13, IL-17, MIP-1α and MIP-1β). We also found that six of the seven responses most closely correlated with IFN-γ were common to both HBHA and PPD. Notably, all HBHA-specific secretion of cytokines and chemokines from infant samples was dependant on previous BCG vaccination. Also, long-term persistence of HBHA-specific responses was found in adolescents with evidence of infant BCG vaccination. This study demonstrates for the first time BCG priming of an HBHA-specific immune response in infants that is characterised by a broad cytokine and chemokine signature. It also suggests a number of BCG vaccination associated, HBHA-induced responses that should be useful for future studies of biomarkers of protection against TB.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

The current global burden of tuberculosis (TB) disease is one of the greatest challenges to public health. That the problem is focussed in tropical countries is in part due to the lack of protective efficacy of the BCG vaccine against adult pulmonary disease at lower latitudes [1]. Efforts towards the development of a more efficacious vaccine for TB should ensure that the resultant immune response is an improvement upon that induced by BCG in the regions where it is least effective.

Infant BCG vaccination is a consistently efficacious and cost-effective preventative measure against childhood forms of TB [2, 3]. There is also evidence that BCG not only affords protection to infants in more equatorial regions but that this protection is against all forms of TB and persists for 15–20 years [4]. The efficacy of infant BCG vaccination must therefore be considered when new TB vaccines designed to improve protection against adult pulmonary TB are developed and tested.

Although homologous boosting of the BCG vaccine has proven ineffective [5], the immunity induced by prior BCG vaccination may be effectively boosted with selected antigens. Such a heterologous approach has seen the induction of strong, polyfunctional CD4+ T-cell responses in humans, specific for the mycobacterial antigen 85A, which was delivered via the modified virus Ankara vector following previous BCG priming [6, 7]. The antigen heparin-binding haemagglutinin (HBHA) is exposed on the surface of a variety of mycobacterial species and promotes binding to host epithelial cells [8]. This feature enhances the pathogenicity of Mycobacterium tuberculosis via the facilitation of extra-pulmonary dissemination of mycobacteria [9]. The detection of HBHA-specific immune responses in healthy patients who were latently infected but not in patients with active disease suggests a protective role for HBHA-specific responses [10]. Such a role is supported by the presence of HBHA-specific CD8+ T cells in latent human TB and the role of these cells in maintaining the latent state [11, 12]. In mouse models, immunisation with HBHA in the presence of a strong Th-1-driving adjuvant has been shown to provide protection similar to that induced by BCG vaccination. Furthermore, BCG administered to neonatal mice can prime HBHA-specific immune responses that, when later boosted with HBHA protein, provided significantly better protection against TB than BCG alone [13-16].

We show here that a broad cytokine and chemokine response was detected in HBHA-stimulated, diluted whole blood from BCG-vaccinated infants, indicating for the first time that BCG vaccination can prime an HBHA-specific immune response in humans.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

PPD and HBHA-specific cytokine/chemokine responses in recently BCG-vaccinated infants

In order to examine the cytokine/chemokine response to myco-bacterial antigens in infants recently vaccinated with BCG, diluted whole blood samples taken from infants approximately 3 months after vaccination or from age-matched controls were stimulated with M. tuberculosis purified protein derivative (PPD) or with HBHA for 7 days. PHA stimulation was included as a positive control. The median concentrations of each analyte measured by multiplex assay in culture supernatants for PPD and HBHA stimulation are displayed in Figure 1. Significant differences (p < 0.0004; Mann–Whitney U test with Bonferonni correction) in antigen-stimulated analyte responses between BCG-vaccinated and -unvaccinated groups are indicated. Thirty-two of 42 and 24 of 42 analytes were significantly increased in the BCG-vaccinated group after PPD and HBHA stimulation, respectively. Interestingly, whereas in the unvaccinated group analyte concentrations were essentially undetectable upon stimulation with HBHA, stimulation of samples from the unvaccinated group with PPD resulted in most cases in substantial secretion of analytes. PHA-induced secretion was detected in both vaccinated and unvaccinated infant groups for 39 of the 42 analytes. There were no significant differences in any PHA-stimulated response between vaccinated and unvaccinated groups (see Supporting Information Fig. 1). Responses to IL-3, IL-4 and IL-15 were not detected in either group to any stimulation.

image

Figure 1. PPD- or HBHA-specific cytokine/chemokine response profiles in BCG-vaccinated and unvaccinated infants. Cytokine and chemokine profiles were measured in tissue culture supernatants after 7 days stimulation of diluted whole blood with either (A) PPD (5 μg/mL) or (B) HBHA (10 μg/mL). Harvested supernatants from duplicate wells were pooled and stored at −80°C prior to multiplex bead array analysis. Black bars represent measurements in samples from infants 3 months after BCG vaccination (n = 20), and grey bars represent measurements in samples from age-matched unvaccinated infants (n = 18). Bars indicate the median cytokine concentration, and error bars show the interquartile range. Non-parametric Mann–Whitney tests were used to determine the statistical significance of the differences between the vaccinated and unvaccinated infant groups. *p-values of <0.0004 were considered significant following the application of the Bonferonni correction rule to allow for multiple comparisons. Cytokine and chemokine measurements that were significantly increased as a result of BCG vaccination are indicated by an asterisk (*).

Download figure to PowerPoint

In order to quantify the magnitude of the increases in analyte concentrations measured after antigen stimulation, the ratio of the median responses between the vaccinated and unvaccinated groups was calculated and used to rank the analytes by magnitude of response induced by either PPD or HBHA (Table 1). PPD was observed to stimulate, in order, IFN-γ, sCD40L and MIP-1α most strongly. The most prominently secreted analytes following HBHA stimulation were, in order, IP-10, MCP-1 and IL-8. To compare the overall profiles of secretion to each antigen, the magnitudes of response are represented as adjacent heat maps (Fig. 2A). A number of differences can be seen between the PPD- and HBHA-stimulated analyte profiles. As indicated above and in Table 1, the four analytes most strongly induced by each antigen do not overlap. These prominent responses are dominated by chemokines for HBHA, whilst for PPD they include Th-1-associated cytokines (IFN-γ, sCD40L), the growth factor GM-CSF, and the chemokine MIP-1α. Despite these differences, a significant correlation (Spearman's Rank correlation coefficient, r = 0.52; p = 0.0013) was detected between the magnitude of increase in biomarker concentrations (ratio of median responses in vaccinated and unvaccinated groups) after PPD or HBHA stimulation (Fig. 2B).

Table 1. Magnitude of PPD- and HBHA-specific cytokine and chemokine responses following infant BCG vaccination
PPD-specific responsesHBHA-specific responses
Cytokine/ chemokineaVaccinated: unvaccinatedp-valuebCytokine/ chemokineVaccinated: unvaccinatedp-value
  1. a

    Analytes were ranked in order of increasing magnitude as determined by the ratio of median response in the vaccinated group to median response in the unvaccinated group.

  2. b

    Non-parametric Mann–Whitney tests were used to determine the statistical p-values for each increase.

IFN-γ976.3<0.0001IP-1011220.5<0.0001
sCD40L270.9<0.0001MCP-15360.0<0.0001
MIP-1α157.1<0.0001IL-84571.6<0.0001
GM-CSF133.1<0.0001MDC532.1<0.0001
IP-10126.6<0.0001IL-1α121.3<0.0001
IL-1777.5<0.0001IFN-γ111.6<0.0001
sIL-2Rα68.0<0.0001MIP-1β105.3<0.0001
IL-1365.1<0.0001IL-696.60.0005
MDC27.4<0.0001GM-CSF94.6<0.0001
IL-12p4026.9<0.0001GRO68.20.0015
IL-1α26.8<0.0001TNF-α54.9<0.0001
MIP-1β25.7<0.0001sCD40L54.3<0.0001
IL-525.3<0.0001sIL-2Rα50.5<0.0001
TNF-α22.4<0.0001MIP-1α47.2<0.0001
IL-619.9<0.0001IL-1Rα45.6<0.0001
MCP-317.7<0.0001FGF-244.7<0.0001
VEGF15.6<0.0001MCP-339.50.0078
Flt-3L14.4<0.0001Fractalkine35.80.0013
IL-1012.8<0.0001VEGF32.1<0.0001
IL-811.1<0.0001IL-1317.3<0.0001
TNF-β9.7<0.0001IL-1715.10.0001
Eotaxin8.6<0.0001G-CSF14.10.0013
IL-1β7.7<0.0001IL-510.10.0001
IL-26.9<0.0001IL-79.10.0003
FGF-26.8<0.0001Eotaxin9.1<0.0001
GRO6.80.001IL-12p408.80.0003
G-CSF6.70.0001IL-1β8.50.0146
Fractalkine6.5<0.0001Flt-3L7.20.0007
IFN-α24.80.0021IL-106.5<0.0001
IL-12p704.8<0.0001TNF-β5.90.0001
IL-74.3<0.0001IL-24.1<0.0001
IL-1Rα4.0<0.0001TGF-α2.40.0032
TGF-α3.3<0.0001EGF2.10.1186
EGF3.20.0587IFN-α21.00.0819
IL-92.60.0001IL-31.00.806
MCP-11.70.0355IL-41.00.806
IL-31.00.0173IL-91.00.6129
IL-41.00.3031IL-12p701.00.1494
IL-151.01.000IL-151.01.000
PDGF-AA1.00.4091PDGF-AA1.00.6129
PDGF-AB/BB1.00.099PDGF-AB/BB1.00.8738
RANTES1.00.8966RANTES1.00.3769
image

Figure 2. BCG vaccination associated hierarchical increases in cytokines and chemokines following PPD or HBHA stimulation. (A) The heat maps represent the ratio of median responses in the vaccinated group (n = 20) to median responses in the unvaccinated group (n = 18) for each of the analytes measured. (B) The PPD- and HBHA-specific response ratios (vaccinated:unvaccinated) were used as co-ordinates to plot the position of each cytokine/chemokine on a scatter diagram. The association between the magnitudes of responses of the individual analytes to each antigen was determined by calculating Spearman's correlation coefficient. Only analytes for which a significant BCG vaccination associated increase was detected on stimulation with either PPD or HBHA are shown.

Download figure to PowerPoint

Thus, there was a broad array of cytokines/chemokines induced in vitro in response to either PPD or HBHA stimulation of diluted whole blood from BCG-vaccinated infants as compared with an age-matched, unvaccinated cohort. Strikingly, the HBHA-specific response, unlike the PPD-specific response, was entirely absent in unvaccinated infants. There was a significant overall association between the magnitude of individual cytokine/chemokine responses to PPD and HBHA, although there were also differences in the biomarkers most strongly induced by each antigen.

Associations amongst cytokine/chemokine responses to PPD and HBHA

Spearman's Rank correlation coefficients were calculated between pairs of cytokines/chemokines to assess which responses were associated in terms of magnitude within individual BCG vaccinees. The degree of correlation between responses was much more apparent following HBHA stimulation than upon PPD stimulation. For HBHA, of the 276 different correlations between all 24 cytokines/chemokines for which there was a positive response, 105 (38%) had coefficients of r ≥ 0.75. For PPD, only 46 of the 496 (9%) different correlations between 32 positive responses had correlation coefficients of r ≥ 0.75.

As IFN-γ is considered an important component of the immune response to TB, we were interested to determine which cytokine/chemokine responses were most or least closely associated with IFN-γ responses in BCG-vaccinated infants. Following PPD stimulation, the magnitude of sCD40L secretion was most closely associated with the magnitude of IFN-γ secretion (r = 0.86), whilst IL-17 was least strongly associated with IFN-γ (r = −0.01). Following HBHA the equivalent strongest and weakest associations were GM-CSF (r = 0.96) and MIP-1α (r = 0.51) (Fig. 3). Figure 4 displays cytokines and chemokines in order of strength of correlation with the IFN-γ responses in both the PPD- and the HBHA-stimulated samples. For both antigens, GM-CSF, FGF-2, sCD40L, IL-12p40, IL-1α and sIL-2Rα comprised six of the seven cytokines/chemokines most closely associated with IFN-γ.

image

Figure 3. Association between the magnitudes of the IFN-γ response in BCG-vaccinated infants and of other cytokine responses. (A, B) Spearman's correlation coefficient was used to determine the strength of associations of different analytes with the IFN-γ response. Plots show the strongest and weakest associations with IFN-γ observed after (A) PPD and (B) HBHA stimulation of samples from BCG-vaccinated infants (n = 20).

Download figure to PowerPoint

image

Figure 4. Spectrum of correlations between the magnitude of the IFN-γ response in BCG-vaccinated infants and the magnitude of other cytokine responses. (A, B) Significant cytokine or chemokine responses after antigen stimulation were compared with IFN-γ responses following infant BCG vaccination (n = 20). Spearman's correlation coefficients are represented for (A) PPD-specific and (B) HBHA-specific responses.

Download figure to PowerPoint

In summary, amongst individual BCG-vaccinated infants there was a tendency for different cytokines/chemokines measured to respond in a related fashion following HBHA stimulation compared with that following PPD stimulation where the correlation between different analyte responses was less marked. There was agreement between the two antigen-induced responses as to which cytokines/chemokines were most strongly correlated with IFN-γ. A major difference between the two antigens, however, was the requirement for BCG vaccination for any HBHA-specific response and the absence of these in unvaccinated infants.

HBHA-specific antibody responses are not mounted following BCG vaccination of infants

In order to determine whether BCG vaccination of infants resulted in HBHA-specific antibody responses, sera from vaccinated and unvaccinated infants were investigated for the presence of IgG. HBHA protein was subjected to SDS-PAGE in multiple lanes that were then cut into strips on the resultant blot. Figure 5A shows the presence of equal quantities of HBHA protein (visible as a 28 kDa band) in multiple lanes on a representative blot following Ponceau S staining. Strips were probed with individual sera and specific antibody binding was detected using alkaline phosphatase conjugated, anti-human IgG secondary antibody. When compared with unvaccinated controls, there was no evidence of HBHA-specific antibody in sera from BCG-vaccinated infants (Fig. 5B). The 28 kDa ladder band and the HBHA band visible following probing with the mouse anti-HBHA monoclonal antibody E4 shows where the HBHA protein is located on the adjacent blots.

image

Figure 5. Western blot analysis of sera from BCG-vaccinated and -unvaccinated infants for the presence of HBHA-specific antibodies. HBHA protein was loaded onto multiple lanes of an SDS polyacrylamide gel and run adjacent to a pre-stained protein ladder. Following blotting, nitrocellulose membranes were stained with Ponceau S to confirm the presence of equal quantities of HBHA protein in each lane at the correct molecular weight (28 kDa). (A) A representative blot is shown. (B) Lanes were then cut into strips and each strip probed with individual serum samples from BCG-vaccinated (n = 18) or -unvaccinated (n = 17) infants. After washing, antibody binding was assessed using a secondary anti-human IgG AP conjugate and appropriate substrate. Representative strips are shown. As a positive control, one strip was probed with the anti-HBHA mouse monoclonal antibody E4 and a secondary anti-mouse IgG AP conjugated used for detection (right).

Download figure to PowerPoint

Persistence of HBHA-specific cytokine/chemokine responses following BCG vaccination

Prior to 2005, BCG vaccination in the United Kingdom was routinely given to adolescents between the ages of 13 and 14 years. In regions where TB rates were higher, BCG was also administered to infants. As a result, we were able to access archived supernatant samples collected at baseline in a previous study of BCG vaccination in adolescents. Supernatants were collected and stored following stimulation of diluted venous blood with HBHA as described above for infant samples. It was possible to identify a sample set from adolescents who had been vaccinated as infants as these individuals had a positive tuberculin skin test and a BCG scar (n = 16). Unvaccinated adolescents were identified following a negative tuberculin skin and having no evident BCG scar (n = 20). Multiplex testing of these supernatants from HBHA-stimulated samples revealed an array of cytokines and chemokines detectable in adolescents who were BCG vaccinated as infants (Fig. 6). Unlike samples from unvaccinated infants, those from adolescents with no evidence of previous BCG vaccination displayed some HBHA-induced responsiveness in terms of cytokine and chemokine secretion. Of the 42 analytes tested, five demonstrated significantly greater responses in BCG-vaccinated adolescents (IP-10, IL-1α, IFN-γ, GM-CSF and sIL-2Rα; p < 0.001, Mann–Whitney U test with Bonferonni correction). Using a less-stringent cutoff level for significance (p < 0.05 without Bonferonni correction), an additional ten analytes were increased in vaccinated adolescents (MCP-1, IL-8, MDC, TNF-α, sCD40L, MIP-1α, IL-1Rα, IL-17, TNF-β, IL-2).

image

Figure 6. HBHA-specific cytokine/chemokine response profiles in BCG-unvaccinated adolescents or adolescents who were BCG vaccinated as infants. Cytokine and chemokine profiles were measured in tissue culture supernatants after 7 days stimulation of diluted whole blood with HBHA (10 μg/mL). Harvested supernatants from duplicate wells were pooled and stored at −80°C prior to multiplex bead array analysis. Black bars represent measurements in samples from adolescents with evidence of previous BCG vaccination (positive tuberculin skin test and a BCG scar; n = 16). Grey bars represent measurements in samples from unvaccinated adolescents (negative tuberculin skin test and no BCG scar evident; n = 20). Bars indicate the median cytokine concentration, and error bars show the interquartile range. Non-parametric Mann–Whitney tests were used to determine the statistical significance of the differences between the vaccinated and unvaccinated adolescents. *p-values of <0.001 were considered significant following the application of the Bonferonni correction rule to allow for multiple comparisons. Cytokines and chemokine measurements that were significantly increased as a result of BCG vaccination are indicated by an asterisk (*).

Download figure to PowerPoint

In summary, these results support the hypothesis that HBHA-specific cytokine and chemokine responses that are detectable following BCG vaccination of infants are maintained for at least 13–14 years. The data also suggest that, in the context of the assay conditions used and analytes measured in this study, some sensitivity to HBHA stimulation develops between the ages of approximately 25 weeks and 13 years that is not associated with BCG vaccination.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

Although BCG vaccination of infants can protect against childhood forms of TB [2] and, according to one study carried out in Brazil, against pulmonary TB for up to 20 years [4], its limited efficacy in certain settings has initiated a search for more effective vaccination strategies. The boosting of specific T-cell responses that have been primed by BCG with defined mycobacterial antigens is one such strategy. As BCG is a complex organism, rich in antigenic structures, it is important to determine that selected antigens do prime T cells and that these T cells are potent effectors and not prone to subjugation by stronger or more downregulatory responses following vaccination. One way of demonstrating potency is for the response to a particular antigen to exhibit multifunctionality. Such a demonstration is increasingly consistent with the detection of a number of expressed effector molecules, which may be either secreted or cell-surface associated. The former may be detected in tissue culture supernatants following antigen stimulation by multiple ELISAs or, as in this study, by multiplex bead arrays.

We detected an extensive and functionally broad cytokine/chemokine response following stimulation of diluted blood from recently BCG-vaccinated infants with mycobacterial antigens. Measurements revealed significant increases in secretion of 32 of the 42 or 24 of the 42 cytokines and chemokines when vaccinated infants were compared with an unvaccinated control group for responses to PPD and HBHA, respectively. We have previously shown using a smaller, 21-plex bead array that the response to PPD in BCG-vaccinated infants was also broad in terms of cytokines and chemokines detected [17]. That finding is confirmed here and extended to include an even broader array of cytokines and chemokines. Of the 15 responses detected in the previous study, only IL-4 was not found to be significantly increased after BCG in the current study. Conversely, the results presented here demonstrate significant increases in Eotaxin, IL-1β, IL-7 and IL-12p70, all of which displayed no statistical difference between vaccinated and unvaccinated infants previously [17]. One possible explanation for the differences is the increased mean age at which infants received BCG vaccination in the current study (11.5 weeks) as compared to the previous study (7 weeks). Differences in cytokine secretion have been reported when responses were compared between infant cohorts receiving BCG at different ages [18, 19], although in those studies the age difference was more pronounced (birth versus 10 weeks or birth versus 4.5 months), and both were in African settings. Another possibility is that we have previously observed differences in analyte detection sensitivity between the 21-plex kit used previously and the 42-plex kit used in the current study, which could also account for the difference seen.

The results presented here demonstrate for the first time the priming of an HBHA-specific immune response in humans following BCG vaccination. This is revealed in the 24 cytokines and chemokines that were detected following HBHA stimulation of diluted whole blood samples from BCG-vaccinated infants, and that were largely undetectable in equivalent assays carried out on samples from unvaccinated infants using a sensitive multiplex bead array assay. There was no evidence of a BCG vaccination induced antibody response to HBHA present in infant sera. Such antibodies have been detected in samples from TB patients [8]. This could represent differences in exposure of the humoral immune system to HBHA from M. tuberculosis as opposed to Mycobacterium bovis BCG or differences in the distribution of antigen during infection rather than vaccination. However, the results presented strongly support the hypothesis that BCG vaccination of UK infants is priming an HBHA-specific T-cell response. We also present data that indicate the persistence of infant BCG-primed, HBHA-specific T-cell responses into adolescence. Although the BCG-associated response profile was not so distinct at this age (i.e. the differences between those vaccinated and unvaccinated were less pronounced), this was in part due to the increased responsiveness to HBHA in unvaccinated individuals. It was previously shown that there were no differences in IFN-γ responses in BCG-vaccinated adults compared with those who were unvaccinated [20]. Together with our data, this suggests that responses to HBHA do wane with time and are eventually indistinguishable from background.

As with PPD, the HBHA response profile in infants was mixed in that it included Th-1-associated and pro-inflammatory cytokines and chemokines (e.g. IFN-γ, TNF-α, GM-CSF, IL-1α, IL-8, IL-17) as well as Th-2-associated and regulatory biomarkers (IL-5, IL-10, IL-13). As we have previously discussed [17, 21], the appearance of Th-2 and regulatory cytokines in data sets such as this may represent a mixed immune response to BCG that incorporates biomarker signatures other than those associated with Th-1 responses or may represent the initiation of regulatory negative feedback loops in response to the concentrated cytokine milieu that develops in longer term assays such as those described here. It has also been suggested that BCG vaccination of infants may induce low-frequency populations of induced regulatory T cells that could account for the detection of cytokines such as IL-10 [22]. Interestingly, none of the Th-2 or regulatory markers remained significantly increased in the vaccinees in the adolescent group.

Fewer analytes were found to be specifically increased in samples from vaccinated infants following HBHA stimulation as compared with that following PPD stimulation (24 versus 32). The undefined and complex nature of PPD as an antigen is likely to stimulate many types of cells in whole blood cultures, including CD4+ T cells of different specificities as well as other types of immune cells. We have previously observed, for example, natural killer cell responses following PPD stimulation of samples from BCG-vaccinated adolescents [21]. Further evidence of the complex nature of the immune response to PPD, albeit in vivo, is seen following transcriptional profiling of human immune responses in human tuberculin skin test biopsies [23]. Those investigators revealed a similarly multifaceted immune response that includes Th-1, CTL, innate and IFN-γ-stimulated immune signatures. As a purified protein antigen, HBHA would be expected to stimulate fewer antigen-specific T-cell responses and therefore a less-exaggerated cytokine milieu in comparison with those for PPD. The quantitative differences in key cytokines such as IFN-γ and TNF-α and reduced feedback loop signals to other cells might then lead to the qualitative differences in the number of different cytokines or chemokines detected following HBHA and PPD stimulation.

It is interesting to note the prominence of the four chemokine responses in BCG-vaccinated infants following HBHA stimulation (IP-10, MCP-1, IL-8 and MDC) when considered in terms of fold increase over the unvaccinated group. In fact, the large fold increase in these four chemokines following HBHA stimulation is more a function of the absence of any response in the unvaccinated group. This is not the case for PPD stimulation, which induces relatively prominent IP-10, MCP-1, IL-8 and MDC responses in unvaccinated infants. As a PPD-specific Th-1 response appears to be absent in unvaccinated UK infants at this age (as indicated here by the lack of an IFN-γ response and also seen previously [24]), it appears that PPD possesses the capacity to stimulate certain chemokine or inflammatory-like responses even in a BCG-unprimed immune system.

Of the cytokines previously measured in studies designed to demonstrate the multifunctionality of mycobacteria-specific T cells [7, 22], IFN-γ, MIP-1β and TNF-α were prominent in the response to HBHA (respective increases of 112-, 105- and 55-fold in BCG-vaccinated infants). Although increased fourfold in the vaccinated group, the presence of IL-2 in the biosignature of the HBHA response was much less marked, a finding that may relate to the kinetics of IL-2 release and degradation in the 7-day whole blood assay. Other Th-1 cytokines that might be predicted to participate in an anti-TB immune response and which this and the previous study have implicated in the PPD-specific and now the HBHA-specific response in BCG vaccinees are IL-1α and IL-6.

The importance of IFN-γ as a component of anti-TB immune responses has been acknowledged and is supported by studies in patients and mice with disabling mutations in the IFN-γ receptor [25-28]. For this reason, it has for some years been the cytokine of choice for investigators wishing to demonstrate an immune response with the potential for effective protection against TB [6, 29]. However, the magnitude of the IFN-γ response does not provide a simple read-out of protective immunity [30-33]. It is now thought that a biosignature representing protective immunity will require components in addition to IFN-γ, although the inclusion of IFN-γ appears to be important. This premise was the rationale behind our attempt in the current study to ascertain which cytokine and chemokines were most strongly associated with IFN-γ responses in individual infants vaccinated with BCG. After HBHA or PPD stimulation, GM-CSF, FGF-2, sCD40L, IL-12p40, IL-1α and sIL-2Rα comprised six of the seven cytokines/chemokines most closely associated with IFN-γ. Furthermore, IFN-γ, GM-CSF, IL-1α and sIL-2Rα were also four of the five analytes that remained significantly increased in BCG-vaccinated adolescents.

Given the extensive HBHA-specific cytokine and chemokine responses identified in this study, the authors recognise the importance of identifying the cellular sources of these molecules in order to further dissect the nature of any potential protective immune mechanisms these responses might bestow upon BCG-vaccinated individuals. Although this analysis was beyond the scope of the current study, we believe that such an investigation would form a logical follow up to the current work.

A potential limitation of this study is that the BCG-vaccinated and -unvaccinated infant cohorts were recruited from neighbouring UK National Health Service regions that had different BCG vaccination policies. As policy is determined by rates of TB, the vaccinated cohort was recruited from a region with a higher rate of TB than the unvaccinated cohort. As we did not determine the TB exposure status of the parents and families, we cannot exclude the possibility of differences in intrauterine or post-partum exposure of infants to TB-related antigens or maternally derived immune mediators between the two cohorts. However, we think that the risk of exposure of infants to TB in this way is small. The majority (86%) of the BCG-vaccinated infant cohort were British Caucasians yet most of TB cases in the region from which they were recruited (86%) were in non-UK-born patients [34]. Across London as a whole, of all the non-UK-born cases of TB, 83% originate from either South Asia or sub-Saharan Africa [35]. Together, these figures suggest that the infants in the BCG-vaccinated cohort are unlikely to have come from families with TB cases.

In summary, we present here data that indicate a broad and extensive cytokine and chemokine profile in BCG-vaccinated infants that is specific for the mycobacterial antigen HBHA and that remains detectable into adolescence. These data indicate the potential of HBHA for the boosting of BCG-primed responses and points the way to studies that will more closely define this immune response and characterise its cellular components. The use of a multiplex assay to measure 42 distinct cytokines or chemokines has revealed a small panel of prominent, BCG vaccination dependant responses that are specific for HBHA antigen and warrant further investigation in the search for protective biomarkers against TB.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

Recruitment and blood sampling

UK born, BCG-vaccinated infants (n = 21) were recruited after obtaining parental, informed consent. The mean vaccination age was 11.5 weeks (standard deviation: 6.8 weeks). Venous blood samples (heparinised for whole blood assays and without anticoagulant for serum samples) were obtained approximately 3 months post BCG vaccination (mean sample time: 13.0 weeks post vaccination, standard deviation: 3.4 weeks). Samples were also obtained from an age-matched cohort of unvaccinated infants (n = 18). The mean blood sample age of the vaccinated and unvaccinated cohorts was 25.0 weeks (standard deviation: 5.4 weeks) and 24.9 weeks (standard deviation: 5.1 weeks), respectively. The vaccinated and unvaccinated cohorts were recruited from neighbouring National Health Service regions with different BCG vaccination policies based on the rates of TB (>40 per 10,000 population and <40 per 100,000 population, respectively). The ethnic composition of the two cohorts were 18 of the 21 Caucasian (British), 2 of the 21 Caucasian (other) and 1 of the 21 mixed race (Caucasian and Asian) for the vaccinated cohort and 14 of the 18 Caucasian (British), 3 of the 18 Caucasian (other) and 1 of the 18 mixed race (Caucasian and Asian) for the unvaccinated cohort. Informed consent was obtained from a parent or guardian of all infant participants. Adolescent samples were obtained from year 8 pupils (13–14 years old) who were invited to participate in the BCG study. Previously BCG-vaccinated adolescents (n = 16) were identified following a positive tuberculin skin test (Heaf grade 1 or above) and confirmation of a healed BCG scar. Unvaccinated adolescents were identified following a negative tuberculin skin test (Heaf grade 0) and with no apparent BCG scar. Informed written consent was obtained from parents and verbal consent given by participants. Approval for the infant and adolescent studies was given by the Redbridge and Waltham Forest Health Authority Local Research Ethics Committee, and the Ethics Committee of the London School of Hygiene & Tropical Medicine.

Diluted whole blood assays

One in ten diluted blood was cultured in 96-well U-bottomed plates (Corning, Tewksbury, MA, USA) in a final volume of 200 μL of medium (RPMI 1640; 2 mM L-glutamine; Invitrogen, Paisley, UK). Antigens used were PPD (Statens Serum Institute, Denmark) at 5 μg/mL and native HBHA at 10 μg/mL. HBHA was purified by heparin-Sepharose chromatography as described [8], followed by reverse-phase high-pressure liquid chromatography as described elsewhere [10]. Phytohaemagglutinin (PHA; Sigma-Aldrich, Dorset, UK) was used as a positive control stimulant at 5 μg/mL. Cultures were incubated for 7 days in humidified incubators at 37°C, 5% CO2 after which culture supernatants were collected and stored at −80°C.

Multiplex bead array

The human 42-plex cytokine/chemokine MilliplexTM MAP premixed kit was used to assess cytokine and chemokine content of culture supernatants according to the manufacturer's instructions (cat no. MPXHCYTO60KPMX42, Millipore Corp, St. Charles, MO, USA): IL-1α, IL-1β, IL-1RA, IL-2, IL-2RA, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-12p40, IL-12p70, IL-13, IL-15, IL-17, sCD40L, EGF, Eotaxin, Flt3L, FGF-2, Fractalkine, G-CSF, GM-CSF, Gro, IFN-α2, IFN-γ, IP-10, MCP-1, MCP-3, MDC, MIP-1α, MIP-1β, PDGF-AA, PDGF-AB/BB, RANTES, TGF-α, TNF-α, TNF-β and VEGF. Data were acquired using the Biorad Luminex reader using Bioplex manager 4.1 software. For each cytokine, the standard curve consisted of six fivefold dilutions between 10,000 and 3.2 pg/mL.

SDS-PAGE and western blots

HBHA protein prepared as described above was subjected to SDS-PAGE (0.8 μg per lane) on a 15% polyacrylamide gel. A protein ladder (PageRuler Plus; Fermentas, St. Leon-Rot, Germany) was run simultaneously. After blotting onto nitrocellulose membranes, protein was visualised by emersion in Ponceau S (Sigma) for 5 min. Membranes were cut into strips and each strip blocked in 1% BSA then probed with a 1/100 dilution of infant serum in TBS-T. Antibody binding was detected using alkaline phosphatase visualisation (ProtoBlot II AP system, Promega, Madison, WI, USA). The HBHA-specific mouse E4 monclonal antibody [8] was used to confirm detection of HBHA on separate western blot strips.

Data management and statistical analysis

All data points for which analyte concentrations measured were below 3.2 pg/mL (i.e. the value of the lowest standard dilution) or for which the concentration was below the lower limit of quantitation were assigned a value of 1.6 pg/mL. When measured concentrations were above the upper limit of quantitation, data points were assigned the highest measurable concentration for that analyte. Data from one vaccinated infant were excluded from the analysis due to abnormally high background analyte secretion. Data were analysed using Excel 2007 (Microsoft), STATA/IC version 10.1 (StataCorp LP) and SPSS version 16.0 (SPSS Inc). The Mann–Whitney U test was used to test for significant differences between vaccinated and unvaccinated infants for each analyte. The cutoff p-values for significance were determined after applying the Bonferonni correction for multiple comparisons. Spearman's Rank correlation coefficient was used to determine the association between the fold increase in biomarker responses due to vaccination following HBHA or PPD stimulation as well as to determine the association between IFN-γ and other biomarker responses in the BCG-vaccinated group.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

This work was supported by funds awarded to Professor Hazel M. Dockrell by the European Commission within the 6th framework program (FP6) TBVAC (contract number LSHP-CT-2003-503367) and the 7th framework programme (FP7) NEWTBVAC (contract number HEALTH-F3-2009-241745). We thank all the infants and mothers and school children who kindly agreed to participate in this study, the staff and school nurses at NHS Redbridge and NHS West Essex particularly Dr. Makki Hameed and Dr. Gladys Xavier and the project phlebotomist Mrs Shakuntala Patel.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information
  • 1
    Wilson, M. E., Fineberg, H. V. and Colditz, G. A., Geographic latitude and the efficacy of bacillus Calmette-Guerin vaccine. Clin. Infect. Dis. 1995. 20: 982991.
  • 2
    Colditz, G. A., Berkey, C. S., Mosteller, F., Brewer, T. F., Wilson, M. E., Burdick, E. and Fineberg, H. V., The efficacy of bacillus Calmette-Guerin vaccination of newborns and infants in the prevention of tuberculosis: meta-analyses of the published literature. Pediatrics 1995. 96: 2935.
  • 3
    Trunz, B. B., Fine, P. and Dye, C., Effect of BCG vaccination on childhood tuberculous meningitis and miliary tuberculosis worldwide: a meta-analysis and assessment of cost-effectiveness. Lancet 2006. 367: 11731180.
  • 4
    Barreto, M. L., Cunha, S. S., Pereira, S. M., Genser, B., Hijjar, M. A., Yury Ichihara, M., de Brito, S. C. et al., Neonatal BCG protection against tuberculosis lasts for 20 years in Brazil. Int. J. Tuberc. Lung Dis. 2005. 9: 11711173.
  • 5
    Dantas, O. M., Ximenes, R. A., de Albuquerque Mde, F., da Silva, N. L., Montarroyos, U. R., de Souza, W. V., Pereira, T. C. et al., A case-control study of protection against tuberculosis by BCG revaccination in Recife, Brazil. Int. J. Tuberc. Lung Dis. 2006. 10: 536541.
  • 6
    McShane, H., Pathan, A. A., Sander, C. R., Keating, S. M., Gilbert, S. C., Huygen, K., Fletcher, H. A. et al., Recombinant modified vaccinia virus Ankara expressing antigen 85A boosts BCG-primed and naturally acquired antimycobacterial immunity in humans. Nat. Med. 2004. 10: 12401244.
  • 7
    Beveridge, N. E., Price, D. A., Casazza, J. P., Pathan, A. A., Sander, C. R., Asher, T. E., Ambrozak, D. R. et al., Immunisation with BCG and recombinant MVA85A induces long-lasting, polyfunctional Mycobacterium tuberculosis-specific CD4+ memory T lymphocyte populations. Eur. J. Immunol. 2007. 37: 30893100.
  • 8
    Menozzi, F. D., Rouse, J. H., Alavi, M., Laude-Sharp, M., Muller, J., Bischoff, R., Brennan, M. J. et al., Identification of a heparin-binding hemagglutinin present in mycobacteria. J. Exp. Med. 1996. 184: 9931001.
  • 9
    Pethe, K., Alonso, S., Biet, F., Delogu, G., Brennan, M. J., Locht, C. and Menozzi, F. D., The heparin-binding haemagglutinin of M. tuberculosis is required for extrapulmonary dissemination. Nature 2001. 412: 190194.
  • 10
    Masungi, C., Temmerman, S., Van Vooren, J. P., Drowart, A., Pethe, K., Menozzi, F. D., Locht, C. et al., Differential T and B cell responses against Mycobacterium tuberculosis heparin-binding hemagglutinin adhesin in infected healthy individuals and patients with tuberculosis. J. Infect. Dis. 2002. 185: 513520.
  • 11
    Temmerman, S. T., Place, S., Debrie, A. S., Locht, C. and Mascart, F., Effector functions of heparin-binding hemagglutinin-specific CD8+ T lymphocytes in latent human tuberculosis. J. Infect. Dis. 2005. 192: 226232.
  • 12
    van Pinxteren, L. A., Cassidy, J. P., Smedegaard, B. H., Agger, E. M. and Andersen, P., Control of latent Mycobacterium tuberculosis infection is dependent on CD8 T cells. Eur. J. Immunol. 2000. 30: 36893698.
  • 13
    Temmerman, S., Pethe, K., Parra, M., Alonso, S., Rouanet, C., Pickett, T., Drowart, A. et al., Methylation-dependent Tcell immunity to Mycobacterium tuberculosis heparin-binding hemagglutinin. Nat. Med. 2004. 10: 935941.
  • 14
    Rahman, M. J. and Fernandez, C., Neonatal vaccination with Mycobacterium bovis BCG: potential effects as a priming agent shown in a heterologous prime-boost immunization protocol. Vaccine 2009. 27: 40384046.
  • 15
    Rouanet, C., Debrie, A. S., Lecher, S. and Locht, C., Subcutaneous boosting with heparin binding haemagglutinin increases BCG-induced protection against tuberculosis. Microbes. Infect. 2009. 11: 9951001.
  • 16
    Guerrero, G. G. and Locht, C., Recombinant HBHA boosting effect on BCG-induced immunity against Mycobacterium tuberculosis infection. Clin. Dev. Immunol. 2011. 2011: 730702.
  • 17
    Lalor, M. K., Smith, S. G., Floyd, S., Gorak-Stolinska, P., Weir, R. E., Blitz, R., Branson, K. et al., Complex cytokine profiles induced by BCG vaccination in UK infants. Vaccine 2010. 28: 16351641.
  • 18
    Burl, S., Adetifa, U. J., Cox, M., Touray, E., Ota, M. O., Marchant, A., Whittle, H. et al., Delaying bacillus Calmette-Guerin vaccination from birth to 4 1/2 months of age reduces postvaccination Th1 and IL-17 responses but leads to comparable mycobacterial responses at 9 months of age. J. Immunol. 2010. 185: 26202628.
  • 19
    Kagina, B. M., Abel, B., Bowmaker, M., Scriba, T. J., Gelderbloem, S., Smit, E., Erasmus, M. et al., Delaying BCG vaccination from birth to 10 weeks of age may result in an enhanced memory CD4 T cell response. Vaccine 2009. 27: 54885495.
  • 20
    Hougardy, J. M., Schepers, K., Place, S., Drowart, A., Lechevin, V., Verscheure, V., Debrie, A. S. et al., Heparin-binding-hemagglutinin-induced IFN-gamma release as a diagnostic tool for latent tuberculosis. PLoS One 2007. 2: e926.
  • 21
    Smith, S. G., Lalor, M. K., Gorak-Stolinska, P., Blitz, R., Beveridge, N. E., Worth, A., McShane, H. et al., Mycobacterium tuberculosis PPD-induced immune biomarkers measurable in vitro following BCG vaccination of UK adolescents by multiplex bead array and intracellular cytokine staining. BMC Immunol. 2010. 11: 35.
  • 22
    Soares, A. P., Scriba, T. J., Joseph, S., Harbacheuski, R., Murray, R. A., Gelderbloem, S. J., Hawkridge, A. et al., Bacillus calmette-guerin vaccination of human newborns induces T cells with complex cytokine and phenotypic profiles. J. Immunol. 2008. 180: 35693577.
  • 23
    Tomlinson, G. S., Cashmore, T. J., Elkington, P. T., Yates, J., Lehloenya, R. J., Tsang, J., Brown, M. et al., Transcriptional profiling of innate and adaptive human immune responses to mycobacteria in the tuberculin skin test. Eur. J. Immunol. 2011. 41: 32533260.
  • 24
    Lalor, M. K., Ben-Smith, A., Gorak-Stolinska, P., Weir, R. E., Floyd, S., Blitz, R., Mvula, H. et al., Population differences in immune responses to bacille Calmette-Guerin vaccination in infancy. J. Infect. Dis. 2009. 199: 795800.
  • 25
    Ottenhoff, T. H., De Boer, T., van Dissel, J. T. and Verreck, F. A., Human deficiencies in type-1 cytokine receptors reveal the essential role of type-1 cytokines in immunity to intracellular bacteria. Adv. Exp. Med. Biol. 2003. 531: 279294.
  • 26
    Pearl, J. E., Saunders, B., Ehlers, S., Orme, I. M. and Cooper, A. M., Inflammation and lymphocyte activation during mycobacterial infection in the interferon-gamma-deficient mouse. Cell Immunol. 2001. 211: 4350.
  • 27
    Flynn, J. L., Chan, J., Triebold, K. J., Dalton, D. K., Stewart, T. A. and Bloom, B. R., An essential role for interferon gamma in resistance to Mycobacterium tuberculosis infection. J. Exp. Med. 1993. 178: 22492254.
  • 28
    Cooper, A. M., Dalton, D. K., Stewart, T. A., Griffin, J. P., Russell, D. G. and Orme, I. M., Disseminated tuberculosis in interferon gamma gene-disrupted mice. J. Exp. Med. 1993. 178: 22432247.
  • 29
    Black, G. F., Weir, R. E., Floyd, S., Bliss, L., Warndorff, D. K., Crampin, A. C., Ngwira, B. et al., BCG-induced increase in interferon-gamma response to mycobacterial antigens and efficacy of BCG vaccination in Malawi and the UK: two randomised controlled studies. Lancet 2002. 359: 13931401.
  • 30
    Wedlock, D. N., Denis, M., Vordermeier, H. M., Hewinson, R. G. and Buddle, B. M., Vaccination of cattle with Danish and Pasteur strains of Mycobacterium bovis BCG induce different levels of IFNgamma post-vaccination, but induce similar levels of protection against bovine tuberculosis. Vet. Immunol. Immunopathol. 2007. 118: 5058.
  • 31
    Elias, D., Akuffo, H. and Britton, S., PPD induced in vitro interferon gamma production is not a reliable correlate of protection against Mycobacterium tuberculosis. Trans. R. Soc. Trop. Med. Hyg. 2005. 99: 363368.
  • 32
    Majlessi, L., Simsova, M., Jarvis, Z., Brodin, P., Rojas, M. J., Bauche, C., Nouze, C. et al., An increase in antimycobacterial Th1-cell responses by prime-boost protocols of immunization does not enhance protection against tuberculosis. Infect. Immun. 2006. 74: 21282137.
  • 33
    Mittrucker, H. W., Steinhoff, U., Kohler, A., Krause, M., Lazar, D., Mex, P., Miekley, D. et al., Poor correlation between BCG vaccination-induced T cell responses and protection against tuberculosis. Proc. Natl. Acad. Sci. USA 2007. 104: 1243412439.
  • 34
    Anderson, L., Moore, J., Kruijshaar, M., Pedrazzoli, D., Bradshaw, L., Crofts, J., Stagg, H. et al., Tuberculosis in the UK: annual report on tuberculosis surveillance in the UK, 2010. Health Protection Agency Centre for Infections, London, October 2010.
  • 35
    Anderson, C., Maguire, H., Anderson, S., Carless, J. and Kanfoudi, L., Tuberculosis in London 2009: annual report on tuberculosis surveillance in London. Health Protection Agency London Regional Epidemiology Unit, London, January 2011.
Abbreviations
HBHA

heparin-binding haemagglutinin

PPD

Mycobacterium tuberculosis purified protein derivative

TB

tuberculosis

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

Disclaimer: Supplementary materials have been peer-reviewed but not copyedited.

FilenameFormatSizeDescription
eji2341-sup-0001-figures1.pdf189KFigure S1. PHA-induced cytokine and chemokineresponses measured by multiplex assay following the 7 day stimulation of diluted whole blood from BCG-vaccinated (black bars) or BCG unvaccinated (grey bars) UK infants.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.