Ht

CDCa

b

c

d

a

ARRAA

KHTTNC

1

aBtsa

ddad

PT

h0

Vaccine 32 (2014) 6240–6250

Contents lists available at ScienceDirect

Vaccine

j o ur na l ho me page: www.elsev ier .com/ locate /vacc ine

BHA vaccination may require both Th1 and Th17 immune responseso protect mice against tuberculosis

laudie Verwaerdea,b,c,d,∗, Anne-Sophie Debriea,b,c,d, Christophe Dombud,amien Legrandc, Dominique Razea,b,c,d, Sophie Lechera,b,c,d, Didier Betbederd,amille Lochta,b,c,d

Inserm U1019, Lille, FranceCNRS UMR8204, Lille, FranceInstitut Pasteur de Lille, Center for Infection and Immunity of Lille, Lille, FranceUniv Lille Nord de France, Lille, France

r t i c l e i n f o

rticle history:eceived 14 November 2013eceived in revised form 12 August 2014ccepted 8 September 2014vailable online 22 September 2014

eywords:BHAuberculosish17 response

a b s t r a c t

Almost one century after the discovery of the BCG vaccine, tuberculosis remains a major cause of globalmortality and morbidity, emphasizing the urgent need to design more efficient vaccines. The heparin-binding haemagglutinin (HBHA) appears to be a promising vaccine candidate, as it was shown to affordprotection to mice against a challenge infection with Mycobacterium tuberculosis when combined withthe strong adjuvant DDA/MPL (dimethyldioctadecyl-ammonium bromide/monophosphoryl lipid A), aTLR4 ligand. In this study, we investigated the immunological response and protection of mice immu-nized with HBHA formulated in lipid-containing nanoparticles and adjuvanted with CpG, a TLR9 ligand.Subcutaneous immunization with this HBHA formulation led to a marked Th1 response, characterizedby high IFN-� levels, but no significant IL-17 production, both in spleen and lung, in contrast to DDA/MPL

anoparticlepG

MPL-formulated HBHA, which induced both IFN-� and IL-17. This cytokine profile was also observed inBCG-primed mice and persisted after M. tuberculosis infection. No significant protection was obtainedagainst challenge infection after vaccination with the nanoparticle-CpG formulation, and this was asso-ciated with a failure to mount a memory immune response. These results suggest the importance of bothTh1 and Th17 immune responses for vaccine-induced immunity.

© 2014 Elsevier Ltd. All rights reserved.

. Introduction

Tuberculosis (TB) remains a major cause of global mortalitynd morbidity despite large vaccination coverage [1] with theacille Calmette-Guérin (BCG), the only vaccine available againsthis disease. Although BCG protects reasonably well infants againstevere and deadly forms of TB [2], it provides insufficient protectiongainst the pulmonary form of the disease in adults [3].

In the two last decades, many efforts have been devoted to theevelopment of new vaccines against TB [4–6]. Some vaccine can-

idates aim at replacing BCG, such as recombinant forms of BCG orttenuated forms of Mycobacterium tuberculosis. Others are beingesigned to strengthen BCG-induced immunity. Only a handful of

∗ Corresponding author at: Center for Infection and Immunity of Lille, Institutasteur de Lille, 1, rue du Prof. Calmette, F-59019 Lille Cedex, France.el.: +33 3 20 87 11 54; fax: +33 3 20 87 11 58.

E-mail address: claudie.verwaerde@ibl.cnrs.fr (C. Verwaerde).

ttp://dx.doi.org/10.1016/j.vaccine.2014.09.024264-410X/© 2014 Elsevier Ltd. All rights reserved.

antigens have been tested so far as potential booster vaccines; oneof them is HBHA, a protein antigen involved in extrapulmonarydissemination of M. tuberculosis [7–9]. It is strongly recognized byT cells from latently infected individuals in contrast to patientswith active TB, and a decline in HBHA-specific T cell responses inlatently infected subjects was shown to be associated with risk ofprogression to TB disease [9–11].

In mouse models, HBHA has been shown to provide protec-tion against challenge with M. tuberculosis [12,13]. However, theprotective effect of HBHA in mice strongly depends on the use ofthe Th1 polarizing adjuvant dimethyldioctadecyl-ammonium bro-mide/monophosphoryl lipid A (DDA/MPL). As this adjuvant is notsuitable for human use, we explored alternative ways to simulateHBHA-mediated protective immune responses.

As an attractive antigen delivery system, we tested here

phospholipid-rich nanoparticles to vectorize HBHA. The use ofnanoparticles has a number of advantages, including the protec-tion of the antigen against degradation, potential cell targetingand cellular antigen delivery, and the possibility to combine

C. Verwaerde et al. / Vaccine 32 (2014) 6240–6250 6241

F elial c( ). HBHa

spiiatntsTcTm

2

2

Ccb

w

2

1mU

ig. 1. Epithelial cell delivery of HBHA using nanoparticles. Human bronchial epitha) or formulated with 30 �g of PGLA particle (b), NP particle (c) or NPL particle (dnd fixation (×63).

everal antigens and immunostimulartory molecules. Furthermore,hospholipid-rich nanoparticles have already undergone clinical

nvestigations in humans [14]. We found that HBHA formulatedn phospholipid-rich nanoparticles induced humoral and cellularnti-HBHA responses in mice, which could be strongly enhanced byhe addition of CpG ODN. Nevertheless, this formulation providedo protection against challenge with M. tuberculosis, in contrasto HBHA formulated with DDA/MPL. This lack of protection inpite of the strong Th1 response was accompanied by a lack ofh17 cytokine induction by the nanoparticle-CpG formulation. Inontrast, the formulation with DDA/MPL induced both Th1- andh17-type responses, suggesting that the induction of Th17 cellsay play a role in protection against tuberculosis by HBHA.

. Materials and methods

.1. Animals and cells

All experiments were performed with 6-week-old female57Bl/6 mice (Charles River Laboratory, L’Arbresle, France). Forhallenge infection with pathogens, mice were transferred to aiosafety level 3 laboratory and housed in isolators.

Human bronchial epithelial SV40-transformed cells (16HBE)ere maintained in DMEM/F12 medium with 10% FCS (Invitrogen).

.2. Bacteria and antigen preparation

M. tuberculosis H37Rv and the BCG Pasteur strain (isolate173P2, WHO, Stockholm, Sweden) were grown in liquid Sautonedium or on solid Middlebrook 7H11 medium (Difco, Detroit,SA). Stock solutions were titrated and stored at −80 ◦C until use.

ells were incubated for 30 min at 37 ◦C with 10 �g FITC-labelled HBHA either aloneA delivery into cells was observed using a fluorescence microscope after washing

HBHA was purified from BCG cultures by chromatography onheparin-Sepharose CL-6B, followed by HPLC, as described [7,10].For FITC labelling, HBHA was isolated from a recombinant Mycobac-terium smegmatis overproducing the BCG HBHA (unpublished data).Purified protein derivative (PPD) was purchased from StatensSerum Institute (Copenhagen, Denmark).

2.3. FITC labelling of HBHA

One milligram of HBHA was labelled with 100 �g FITC byovernight incubation in 0.1 M carbonate/bicarbonate buffer (pH9.5) at room temperature. FITC-labelled HBHA was separated fromfree FITC by two passages on a Sephadex G10 column.

2.4. Nanoparticles and antigen formulation

PGLA nanoparticles were prepared as previously described [15].Polysaccharide cationic nanoparticles (NP) were prepared frommaltodextrin; NPL are NP with anionic phospholipids incorporatedin their core [16,17]. HBHA (5 �g) was incorporated into nanopar-ticles by incubation for 15 min with nanoparticles (15 �g) in a finalvolume of 30 �l.

2.5. Immunization, prime-boost protocol and challenge infection

Animals were immunized three times at 2-week intervals with

5 �g HBHA encapsulated into 15 �g nanoparticles with or with-out 10 �g of CpG (ODN 1826, Invivogen). Alternatively, mice wereimmunized with 5 �g HBHA emulsified in 200 �l with DDA/MPL(150 �g DDA and 25 �g MPL/injection; Sigma).

6242 C. Verwaerde et al. / Vaccine 32 (2014) 6240–6250

Fig. 2. Antibody and cytokine immune responses after s.c. immunization of mice with HBHA formulated in NPL particles. (A) C57Bl/6 mice (n = 6–7/group) were immunized3 L (HBa ults st CFU

nmetEfwTo

2

[(ovuct

2

ouoe

times with 5 �g of HBHA formulated with NPL particles (nano-HBHA) or DDA/MPnti-HBHA responses were determined 1 week after the last immunization. The reshe means. (B) Mice were immunized as in A and challenged 4 weeks later with 100

In prime-boost experiments, mice were first primed subcuta-eously (s.c.) with 5 × 105 colony forming units (CFU) BCG and 8onths later boosted three times at a 2-week interval with differ-

nt preparations of HBHA. Animals were challenged at the indicatedime points by aerosol with ∼100 CFU of M. tuberculosis H37Rv.ight weeks later, their spleens were divided in two equal parts, oneor CFU counting and the other for splenocyte culture. The right lungas used for CFU counting and the left lung for cell preparations.

he bacterial loads were determined by plating serial dilutions ofrgan homogenates onto Middlebrook 7H11 medium agar.

.6. Lymphocyte preparations and cytokine production

Spleen and lung cells were prepared as previously described18,19]. Cells were added at 7.5 × 105 cells per well in 96-well platesBD Falcon) and stimulated with medium, 10 �g/ml HBHA or PPDr 1 �g/ml of Concanavalin A (Sigma). Cell supernatants were har-ested 48 h later, and cytokine levels were determined by ELISAsing the BD OptEIATM kit (for IFN-�, IL-5, IL-10) or a pair of spe-ific monoclonal antibodies (for IL-17; BD Pharmingen) accordinglyo manufacturer’ protocols.

.7. Antibody determination

Microtiter plates were coated overnight at 4 ◦C with 100 �l

f HBHA solution (1 �g/ml in carbonate buffer, pH 9.5) and sat-rated with 3% BSA in PBS for 2 h. Serum samples were addednto the wells and kept overnight at 4 ◦C. After washing, differ-nt goat anti-mouse Ig isotype-horseradish peroxidase conjugates

HA-DDA/MPL) and their antibody (sera diluted 1/400) and cellular (IFN-� and IL-5)hown are representative of 3 independent experiments. Horizontal lines representMtb by aerosol. The bacterial load were analysed 8 weeks later in spleen and lung.

(Southern Biotech, AL, USA), were added, and the plates wereincubated for 1 h. After washing, the presence of antibodies wasrevealed with the TMB substrate (BD).

2.8. Statistical analysis

The results were analysed using the Mann–Whitney U test(GraphPad Prism program). A value of P ≤ 0.05 was considered tobe significant.

3. Results

3.1. Selection of the optimal nanoparticle to deliver HBHA

Three types of nanoparticles were tested in order to deliverHBHA: an anionic nanoparticle composed of PGLA, and two pos-itively charged nanoparticles composed of maltodextrin, one ofwhich containing a lipid core composed of 70% phospholipids(termed, respectively, NP and NPL). The optimal formulation wasdetermined in an in vitro cell assay using human bronchial epithelialcells incubated with the different FITC-labelled HBHA-containingnanoparticles. As shown in Fig. 1, the best cell delivery wasobserved with HBHA incorporated into NPL. This formulation wasthus used in the following study.

3.2. Induction of immune responses with HBHA-NPL

HBHA encapsulated in NPL particles (nano-HBHA) was injecteds.c. three times in mice, and both humoral and cellular immune

C. Verwaerde et al. / Vaccine 32 (2014) 6240–6250 6243

Fig. 3. Effect of addition of CpG to HBHA preparation. (A) C57Bl/6 mice (n = 5/group) were immunized 3 times with 5 �g of HBHA formulated with NPL particles (nano-HBHA),with NPL particles and co-incorporated CpG (nano-HBHA-CpG), with NPL particles and CpG given at a different site (nano-HBHA/CpG) or with DDA/MPL (HBHA-DDA/MPL),and their anti-HBHA IgG and IgG1, IgG2a, (sera diluted 1/200 for IgG, 1/500 for IgG1, 1/100 for IgG2a,) and cellular (IFN-�, IL-5 and IL-17) anti-HBHA response were determined1 week after the last immunization. Negative controls included mice immunized with NPL particles without HBHA and CpG (nano) and mice immunized with NPL withoutHBHA but with incorporated CpG (nano-CpG). The results shown are representative of two independent experiments. Horizontal lines represent the means. (B) C57Bl/6 mice(n = 5/group) were immunized three times with 5 �g of HBHA formulated with DDA or NPL particles plus 25 �g MPL (DDA-HBHA/MPL or nano-HBHA/MPL) or plus 15 �gCpG (DDA-HBHA/CpG or nano-HBHA/CpG). IFN-� and IL-17 production were determined 1 week after the last immunization after in vitro stimulation of splenocytes withH dividiw btain

rHTSa

BHA (10 �g/ml).Results are expressed in the form of ratios which are obtained by

ith the corresponding MPL-containing formulation. They are the mean of values o

esponses were analysed one week later. As a comparator,

BHA formulated with DDA/MPL (HBHA-DDA/MPL) was used.he negative control consisted of nanoparticles without HBHA..c. immunization with nano-HBHA led to the production ofnti-HBHA antibodies (Fig. 2A), although at a level inferior to

ng the cytokine levels found in CpG-containing formulation by the values obtaineded with 2 and 3 pooled spleen cell preparations for each group.

that obtained with HBHA-DDA/MPL. Upon in vitro stimulation

with HBHA, splenocytes from mice immunized with nano-HBHAproduced lower IFN-�, but higher IL-5 levels than splenocytesfrom mice immunized with HBHA-DDA/MPL (Fig. 2A). Whenmice were challenged by aerosol with M. tuberculosis, significant

6244 C. Verwaerde et al. / Vaccine 32 (2014) 6240–6250

Fig. 4. Cytokine profile induced by nano-HBHA-CpG in BCG-primed mice (A) and after mycobacterial challenge (B). C57Bl/6 mice (n = 5/group) were s.c injected with BCGand 10 weeks later immunized twice with 5 �g of HBHA formulated with NPL particles (nano-HBHA), with NPL particles and co-incorporated CpG (nano-HBHA-CpG) or withDDA/MPL (HBHA-DDA/MPL). (A) The cellular anti-HBHA response (IFN-�, IL-5 and IL-17) was determined in the spleen 1 week after the last immunization. NPL particleswithout HBHA and CpG (nano) served as a negative control. (B) Four weeks after the last immunization, animals were i.n. challenged with 2.5 × 105 BCG bacteria. Four weeksl and thN

p(

3

TcaisGCitaf

ater, both splenocytes and lung cells were in vitro stimulated with HBHA or PPD,

PL particles without HBHA and CpG (nano) served as a negative control.

rotection was only seen in HBHA-DDA/MPL vaccinated miceFig. 2B).

.3. Addition of CpG into NPL

In order to enhance nano-HBHA immunogenicity, we added theLR9 ligand CpG, an already approved adjuvant for clinical appli-ation, to nano-HBHA. CpG ODN1826 was selected because it is

very potent Th1 adjuvant in murine models both in vitro andn vivo, [20], which was previously used with success in vaccinetrategies against intracellular pathogens [21,22]. Ten microgrampG was either co-incorporated with HBHA into NPL (nano-HBHA-pG) or administered at approximately 1 cm from the nano-HBHA

njection site (nano-HBHA/CpG). As shown in Fig. 3A, the addi-ion of CpG induced an increase in total anti-HBHA IgG production,pproaching the level obtained with HBHA-DDA/MPL. In particularor the Th1-related IgG2a, a marked increase was observed upon the

e IFN-� and IL-17 production determined in the supernatants collected 48 h later.

addition of CpG. This effect was only observed for nano-HBHA-CpG, but not for nano-HBHA/CpG. Upon in vitro stimulation ofsplenocytes with HBHA, the IFN-� response of nano-HBHA-CpG-immunized mice was comparable to that of HBHA-DDA/MPL-immunized mice, whereas only a weak IFN-� response was seen inthe nano-HBHA and nano-HBHA/CpG groups (Fig. 3A). The additionof CpG, either within the nanoparticle or given at a different site,completely abolished the IL-5 production seen in the nano-HBHAgroup. Finally, in contrast to HBHA-DDA/MPL-immunized mice, nosignificant amount of IL-17 was produced in any other group. Thus,a highly polarized and Th1-type restricted immune response wasinduced in mice immunized with nano-HBHA-CpG.

In order to determine whether the low IL-17 production

observed in the nano-HBHA-CpG group was due to the immunos-timulant CpG, or to the carrier (i.e. nanoparticles), we performeda back to back comparison of the two carriers combined withthe two immunostim ulants and examined the relative IL-17 and

C. Verwaerde et al. / Vaccine 32 (2014) 6240–6250 6245

Fig. 5. Lack of protection in s.c. nano-HBHA-CpG immunized, BCG-primed mice against aerosol challenge with M. tuberculosis. C57Bl/6 mice (n = 5–7/group) were s.c. primedwith PBS or 5 × 105 CFU BCG and 8 months later immunized 3 times with 5 �g HBHA formulated with NPL particles and co-incorporated CpG (nano-HBHA-CpG) or withD d CpGa CFU Mw

Irvie(ttI

3t

aapCbtwmwigc

tblmHwrssEaI

DA/MPL (HBHA-DDA/MPL). Mice immunized with NPL particles without HBHA ans a negative control. The mice were challenged 8 weeks later by aerosol with ∼100eeks later. Horizontal lines represent the means.

FN-� production depending on the immunostimulant added. Theesults, expressed as fold changes of cytokine levels measured afteraccination with CpG- compared to MPL-containing formulations,ndicated that, when compared to MPL, the addition of CpG withither the DDA or the NPL carriers led to a drop in IL-17 synthesisFig. 3B). In contrast, the addition of CpG in either carrier formula-ion led to an increase in IFN-� production. These results indicatehat the immunostimulant CpG was the major driver of the lowL-17 and high IFN-� production, rather than the carrier NPL.

.4. Nano-HBHA-CpG-induced cytokine profile is maintained inhe context of BCG priming or after challenge

We have previously shown that HBHA administrations givenfter BCG priming strongly increases BCG-mediated protectiongainst M. tuberculosis challenge [12]. Here, we tested whether BCGriming influences the cytokine profile induced by nano-HBHA-pG immunization. Mice received s.c. BCG, followed 10 weeks latery two s.c. nano-HBHA, nano-HBHA-CpG or HBHA-DDA/MPL injec-ions, given at a 2-week interval. One week later, the splenocytesere restimulated with HBHA, and the secreted cytokines wereeasured. After BCG priming, we observed significant IFN-� buteak IL-5 secretion in the nano-HBHA-CpG and HBHA-DDA/MPL

mmunized groups, and only barely IL-17 in the nano-HBHA-CpGroup (Fig. 4A), indicating that BCG priming did not modify theytokine profile induced by nano-HBHA-CpG immunization.

To investigate whether this polarized immune response is main-ained following pulmonary infection, mice were BCG-primed,oosted twice with nano-HBHA-CpG and then intra-nasally chal-

enged with BCG. Four weeks later the cytokine responses wereeasured in spleen and lung cells upon stimulation with PPD orBHA. Whereas no marked difference in the IFN-� productionas seen between groups, immunization with nano-HBHA-CpG

esulted in a general reduction of IL-17 synthesis, both in lung andpleen (Fig. 4B). This reduction was seen after both HBHA and PPD

timulation, indicating a general inhibition of the IL-17 production.ven the basal production of IL-17 in unstimulated lung cells wasbolished in animals receiving nano-HBHA-CpG. This absence ofL-17 production was also noticed in the nano-HBHA group, but

(nano) or with NPL without HBHA but with incorporated CpG (nano-CpG) served. tuberculosis, and the bacterial burden was determined in the spleens and lungs 8

affected essentially the specific anti-HBHA response, and not theanti-PPD response.

3.5. Lack of protection after nano-HBHA-CpG immunization

In order to investigate the potential protective effect ofnano-HBHA-CpG against M. tuberculosis challenge in BCG-primedmice, mice were immunized three times with different formu-lations of HBHA 8 months after BCG priming and challenged by aerosol with M. tuberculosis 8 weeks later, according to theprotocol defined previously [12]. As shown in Fig. 5, only animalsimmunized with HBHA-DDA/MPL were protected against aerosolchallenge in the lungs (0.7 log reduction). A trend (0.3 log reduction)was observed after immunization with nano-HBHA-CpG, althoughthis did not reach statistical significance. This could to be due in partto the presence of CpG, since a slight reduction was also observedwith nano-CpG in the absence of HBHA (BCG/nano vs BCG/nano-CpG; 0.2 log reduction). However, this CpG effect was only foundin BCG-primed animals, but not in control mice (i.e. PBS/nano-CpGvs PBS/nano), suggesting a booster effect of CpG on BCG priming aspreviously reported [23]. No significant protection was observed inthe spleen with or without BCG vaccination, although there was atrend in the reduction of the bacterial burden was seen 8 monthsafter BCG administration, (respectively, 0.43 and 0.87 log reductionbetween PBS-nano vs BCG-nano and PBS/nano-CpG vs BCG/nano-CpG primed animals). These results are somewhat in contrastto earlier work [12]. However, in the previous study a differentmouse strain was used (Balb/c), as well as another route of infec-tion (intraveneous), which may explain the differences in theseresults.

3.6. Lack of memory response in BCG-primed, nano-HBHA-CpGimmunized mice after challenge

In an attempt to understand why nano-HBHA-CpG immuniza-

tion conferred only weak protection in spite of the strong Th1response, we analysed the immune response developed in miceafter infection. As indicated in Fig. 6A whereas the anti-HBHA anti-body response was maintained after challenge of HBHA-DDA/MPL

6246 C. Verwaerde et al. / Vaccine 32 (2014) 6240–6250

Fig. 6. Lack of anti-HBHA memory induced by nano-HBHA-CpG immunization. C57Bl/6 mice (n = 5–7/group) were s.c. primed with PBS or 5 × 105 CFU BCG and 8 monthslater immunized 3 times with 5 �g HBHA formulated with NPL particles and co-incorporated CpG (nano-HBHA-CpG) or with DDA/MPL (HBHA-DDA/MPL). Mice immunizedwith NPL particles without HBHA and CpG (nano) or with NPL without HBHA but with incorporated CpG (nano-CpG) served as a negative control. The mice were challenged 8weeks later by aerosol with ∼100 CFU M. tuberculosis and sacrificed 8 weeks after challenge. (A) The anti-HBHA IgG, IgG1 and IgG2a levels were measured in the sera either 1week after the last immunization or at the time of sacrifice (sera diluted 1/400 for IgG, 1/200 for IgG1 and IgG2a). Individual mice were analysed, and each symbol representsa different mouse. Horizontal lines represent the means. (B) IFN-�, IL-5 and IL-17 levels were analysed on spleen and lung cell culture supernatants after in vitro stimulationwith HBHA.

C. Verwaerde et al. / Vaccine 32 (2014) 6240–6250 6247

Fig. 7. Lack of protection after i.n. immunization with nano-HBHA-CpG.C57Bl/6 mice (n = 7/group) were either i.n. or s.c. immunized three times with 5 �g of HBHA formulated with NPL particles and co-incorporated CpG (nano-HBHA-CpG) orwith DDA/MPL (HBHA-DDA/MPL). Mice immunized with NPL particles without HBHA but with incorporated CpG (nano-CpG) served as a negative control. Animals werea eks lat( ernatad

iLmbH

erosol challenged 4 weeks later with ∼100 CFU M. tuberculosis and sacrificed 8 weA) IFN-�, IL-5 and IL-17 levels were analysed on spleen and lung cell culture supetermined in the spleen and lungs. Horizontal lines represent the means.

mmunized mice, it decreased in the nano-HBHA-CpG group.

ikewise, while spleen cells from HBHA-DDA/MPL immunized ani-als secreted diverse cytokines upon HBHA stimulation, only

ackground cytokine secretion levels were observed in nano-BHA-CpG group (Fig. 6B). Weaker responses to HBHA were also

er.nts after in vitro stimulation with HBHA for 48 h and (B) the bacterial loads were

found in the lungs of the nano-HBHA-CpG group, compared to the

HBHA-DDA/MPL group.

Thus, despite the strong immunogenicity of HBHA achievedafter s.c. delivery with nanoparticles plus CpG, this antigen formula-tion appeared unable to maintain a robust T cell immunity. Several

6 ccine

s1tor

3s

bnwlniaiIHpac

4

dTNatcr

ietrHs[tar[

nEinNuPVtioHpeHwp[

248 C. Verwaerde et al. / Va

tudies have suggested that the generation of vaccine-induced IL-7 may be prerequisite for long-lasting memory immunity againstuberculosis [24–26]. The absence of production of this cytokine inur vaccination protocol may thus account for the lack of memoryesponse observed.

.7. Intranasal immunization elicits IL-17 synthesis but is notufficient for protection

Since the intranasal (i.n.) vaccination has been reported toe the preferred route for promoting Th17 responses [27–29],ano-HBHA-CpG was administered three times i.n. to mice, whichere then challenged by aerosol with M. tuberculosis 4 weeks

ater. Control groups consisted of animals s.c. immunized withano-HBHA-CpG or HBHA-DDA/MPL. In the i.n. nano-HBHA-CpG-

mmunized mice, we observed measurable, but low IL-17 synthesisfter in vitro HBHA stimulation both in the spleens and the lungs,n contrast to s.c. nano-HBHA-CpG-immunized mice (Fig. 7A). ThisL-17 production remained, however, far below that obtained byBHA-DDA/MPL immunization, which was also the case for theroduction of IFN-�. Intranasal vaccination with nano-HBHA-CpGlso did not result in significant protection against M. tuberculosishallenge (Fig. 7B).

. Discussion

In this study we evaluated the capacity of nanoparticles toeliver HBHA, a potential subunit vaccine candidate against TB.he best delivery of HBHA into cells was achieved with a cationicPL, which is consistent with the high content of HBHA in basicmino acids [30]. These NPL are of small size (60 nm), facilitatingheir migration to the draining lymph node and consequently theontact with antigen-presenting cells and the initiation of immuneesponses [31–33].

S.c. injection of HBHA encapsulated in NPL induced anti-HBHAmmune responses. The addition of a low dose of CpG stronglynhanced this response, to a level comparable to that obtained withhe strong adjuvant formulation HBHA-DDA/MPL. This enhancedesponse was only observed when CpG was co-entrapped withBHA into nanoparticles, allowing for concomitant delivery to the

ame antigen-presenting cells, and confirming previous reports34–37]. CpG directed the anti-HBHA immune response mainlyowards a Th1 type profile, characterized by IFN-� productionnd antibodies of the IgG2a isotype. It suppressed the Th2-elated IL-5 cytokine production, consistent with previous reports38–41].

No or only very low levels of IL-17 were detected after s.c. immu-ization with nano-HBHA-CpG, in contrast to HBHA-DDA/MPL.ven in BCG-primed animals boosted with nano-HBHA-CpG,mmunization did not induce IL-17, and no IL-17 was induced inano-HBHA-CpG-immunized mice after M. tuberculosis challenge.either in the spleen nor in the lungs was IL-17 detected after stim-lation with HBHA or with other mycobacterial antigens, such asPD, indicating a generalized suppression of the Th17 response.arious cytokines, such as IL-4, IL-10 or IL-27, have been described

o inhibit IL-17 synthesis [42]. However, we did not detect signif-cant differences in the production of these cytokines at the timef sacrifice between animals immunized with nano-HBHA-CpG vsBHA-DDA/MPL (data not shown). The link between CpG and IL-17roduction is complex, and conflicting reports are found in the lit-rature, showing either a positive or a negative association [43–49].

owever, these studies used different protocols of administration,hich may have impacted on the nature of the different antigen-resenting cell populations present at the site of immunization50].

32 (2014) 6240–6250

It has now increasingly become clear that the production ofIFN-� is important but not sufficient for protection against TB[51–53], and the results reported here support this idea. Althoughnano-HBHA-CpG administration induced a strong IFN-� response,this did not result in significant protection against M. tuberculosischallenge infection, whereas HBHA-DDA/MPL provided significantprotection, without superior IFN-� production. The main differ-ence between these two formulations resided in the differentialproduction of IL-17. This cytokine was recently shown to play acrucial role in the formation of granulomas during M. tuberculo-sis infection [54] and in the establishment of a strong memory Tcell response [24]. Studies performed with IL-17KO or IL-17RA-/-mice indicated that IL-17 is not essential to control mycobacterialgrowth, but its lack resulted in reduced survival late in the courseof infection [24,25,55].

Importantly, IL-17 was shown to be involved in vaccine-inducedimmunity [24–26], and the results presented here also suggest therole of IL-17 in maintaining vaccine-induced memory, even if thedefinitive proof of its role in HBHA-conferred protection will implyits depletion by specific anti-IL-17 antibodies. Challenge infectionfailed to induce a recall of both humoral and cellular responsesin mice s.c. immunized with nano-HBHA-CpG, in contrast to ani-mals immunized with HBHA-DDA/MPL, which induced a markedIL-17 synthesis and showed persistent immune responses afterchallenge. One hypothesis is that this general immunosuppressionmay be due to the generation of regulatory T and B cells follow-ing CpG administration, and possibly amplified by M. tuberculosischallenge. There is indeed a fine balance between the generationof Th17 and Treg lineages. Although originating from a same pre-cursor, these two CD4+ T cell subsets diverge early, depending ontheir cellular environment, into distinct phenotypes with oppositeactivities [56]. Illustrating this complexicity is the opposite effectof CpG on the generation or suppression of Treg activity describedin the literature [57–59]. Whether in our model, CpG indeed gen-erates Treg cells awaits further investigation, especially throughkinetic studies on the cytokine profile and the phenotype of thecells that expand in the lungs after immunization and challengeinfection.

In addition to the choice of the adequate combination of adju-vants, the site of antigen administration and the nature of thedifferent immune cell types present at this site, are also impor-tant to optimize immune responses [60]. The mucosal tissuerepresents an attractive site to induce protective immunity, par-ticularly against M. tuberculosis, for which it is the natural portalof entry. It has been shown that pulmonary immunization withAg85B associated with CpG in nanoparticles elicits both IFN-�and IL-17 synthesis, whereas intradermal injection with the samepreparation failed to induce IL-17 [51]. In that study, only immu-nization via the pulmonary route resulted in some protectionagainst a challenge infection. By contrast, in another study, pul-monary administration of Ag85A with CpG was shown to inducehigh levels of IFN-� but no IL-17 and failed to protect mice againstchallenge infection [52]. In this study, we found that i.n. immuniza-tion with nano-HBHA-CpG induced measurable but low levels ofIL-17 in addition to IFN-�, confirming the importance of the immu-nization site to orient the immune response. However, both theseIL-17 and IFN-� levels were low and not sufficient to protect ani-mals against M. tuberculosis challenge. This weak immunogenicityis rather surprising since we have previously shown that NPL is ableto induce a marked immune response to the encapsulated antigenby the i.n. route [36]. HBHA has the capacity to interact with heparinsulphate present at the surface of epithelial cells, which may there-

fore potentially have impacted on the transit of the nano-particlesthrough the epithelium. We are now investigating new formula-tions of encapsulated HBHA, hoping to obtain improved mucosalimmunity and protection.

ccine

Tnds

A

MgFac

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

terium tuberculosis infection. Immunology 2007;121(4):508–17.

C. Verwaerde et al. / Va

In conclusion, this work suggests that in addition to the effectorh1 immune response, the protection conferred by HBHA immu-ization against Mycobacterium tuberculsosis infection may alsoepend on the development of a Th17 type response, previouslyhown to be involved in vaccine-induced immune memory.

cknowledgements

We are grateful to Carine Rouanet for helpful discussion and toarc Loyens for HBHA preparation. This work was supported by a

rant from the European Commission (contract number HEALTH-3-2009-241745), Framework 7 program, NEWTB-VAC (Discoverynd preclinical development of new generation tuberculosis vac-ines.

eferences

[1] World Health Organization. Weekly epidemiological record. World HealthOrganization; 2011. p. 509–20.

[2] Colditz GA, Berkey CS, Mosteller F, Brewer TF, Wilson ME, Burdick E, et al.The efficacy of bacillus Calmette-Guerin vaccination of newborns and infantsin the prevention of tuberculosis: meta-analyses of the published literature.Pediatrics 1995;96:29–35.

[3] Colditz GA, Brewer TF, Berkey CS, Wilson ME, Burdick E, Fineberg HV, et al.Efficacy of BCG vaccine in the prevention of tuberculosis—meta-analysis of thepublished literature. JAMA 1994;271:698–702.

[4] Ottenhoff TH, Kaufmann SH. Vaccines against tuberculosis: where are we andwhere do we need to go. PLoS Pathog 2012;8(5):e1002607.

[5] Kaufmann SH. Tuberculosis vaccine development: strength lies in tenacity.Trends Immunol 2012;33(7):373–9.

[6] Brennan MJ, Stone MR, Evans T. A rational vaccine pipeline for tuberculosis. IntJ Tuberc Lung Dis 2012;16(12):1566–73.

[7] Menozzi FD, Rouse JH, Alavi M, Laude-Sharp M, Muller J, Bischoff R, et al. Iden-tification of a heparin-binding hemagglutinin present in mycobacteria. J ExpMed 1996;184:993–1001.

[8] Pethe K, Bifani P, Drobecq H, Sergheraert C, Debrie A-S, Locht C, et al.Mycobacterial heparin-binding hemagglutinin and laminin-binding proteinshare antigenic methyllysines that confer resistance to proteolysis. Proc NatlAcad Sci USA 2002;99:10759–64.

[9] Pethe K, Alonso S, Biet F, Delogu G, Brennan MJ, Locht C, et al. Theheparin-binding haemagglutinin of Mycobacterium tuberculosis is required forextrapulmonary dissemination. Nature 2001;412:190–4.

10] Masungi C, Temmerman S, Van Vooren JP, Drowart A, Pethe Menozzi FD, et al.Differential T and B cell responses against Mycobacterium tuberculosis heparin-binding hemagglutinin adhesin in infected healthy individuals and patientswith tuberculosis. J Infect Dis 2002;185:513–20.

11] Corbière V, Pottier G, Bonkain F, Schepers K, Verscheure V, Lecher S, et al. Riskstratification of latent tuberculosis defined by combined interferon gammarelease assays. PLoS One 2012;7:e43285.

12] Rouanet C, Debrie AS, Lecher S, Locht C. Subcutaneous boosting with heparinbinding haemagglutinin increases BCG-induced protection against tuberculo-sis. Microbes Infect 2009;11:995–1001.

13] Parra M, Pickett T, Delogu G, Dheenadhayalan V, Debrie A-S, Locht C,et al. The mycobacterial heparin-binding hemagglutinin is a protective anti-gen in the mouse aerosol challenge model of tuberculosis. Infect Immun2004;72:6799–805.

14] Stephenson I, Zambon MC, Rudin A, Colegate A, Podda A, Bugarini R, et al. PhaseI evaluation of intranasal trivalent inactivated influenza vaccine with nontox-igenic Escherichia coli enterotoxin and novel biovector as mucosal adjuvants,using adult volunteers. J Virol 2006;80(10):4962–70.

15] Chang J, Paillard A, Passirani C, Morille M, Benoit JP, Betbeder D, et al. Trans-ferrin adsorption onto PLGA nanoparticles governs their interaction withbiological systems from blood circulation to brain cancer cells. Pharm Res2012;29(6):1495–505.

16] Paillard A, Passirani C, Saulnier P, Kroubi M, Garcion E, Benoit JP, et al. Positivelycharged, porous, polysaccharide nanoparticles loaded with anionic moleculesbehave as ‘stealth’ cationic nanocarriers. Pharm Res 2007;27:126e33.

17] Dombu C, Carpentier R, Betbeder D. Influence of surface charge and innercomposition of nanoparticles on intracellular delivery of proteins in airwayepithelial cells. Biomaterials 2012;33(Dec (35)):9117–26.

18] Verwaerde C, Delanoye A, Macia L, Tailleux A, Wolowczuk I. Influence of high-fat feeding on both naive and antigen-experienced T-cell immune response inDO10.11 mice. Scand J Immunol 2006;64(5):457–66.

19] Paget C, Ivanov S, Fontaine J, Blanc F, Pichavant M, Renneson J, et al. Potentialrole of invariant NKT cells in the control of pulmonary inflammation and CD8+T cell response during acute influenza A virus H3N2 pneumonia. J Immunol

2011;186(10):5590–602.

20] Hartmann G, Weeratna RD, Ballas ZK, Payette P, Blackwell S, et al. Delineation ofa CpG phosphorothioate oligodeoxynucleotide for activating primate immuneresponses in vitro and in vivo. J Immunol 2000;164:1617–24.

[

32 (2014) 6240–6250 6249

21] Al-Mariri A, Tibor A. Protection of BALB/c mice against Brucella abortus 544 chal-lenge by vaccination with bacterioferritin or P39 recombinant proteins withCpG oligodeoxynucleotides as adjuvant. Infect Immun 2001;69(8):4816–22.

22] Rhee EG, Mendez S, Shah JA, Wu CY, Kirman JR, et al. Vaccination withheat-killed leishmania antigen or recombinant leishmanial protein and CpGoligodeoxynucleotides induces long-term memory CD4+ and CD8+ T cellresponses and protection against leishmania major infection. J Exp Med2002;195(12):1565–73.

23] Taniguchi K, Takii T, Yamamoto S, Maeyama J, Iho S, Maruyama M, et al.Reactivation of immune responses against Mycobacterium tuberculosis byboosting with the CpG oligomer in aged mice primarily vaccinated withMycobacterium bovis BCG. Immun Ageing 2013;10(1):25.

24] Khader SA, Bell GK, Pearl JE, Fountain JJ, Rangel-Moreno J, Cilley GE, et al.IL-23 and IL-17 in the establishment of protective pulmonary CD4+ T cellresponses after vaccination and during Mycobacterium tuberculosis challenge.Nat Immunol 2007;8(4):369–77.

25] Wozniak TM, Saunders BM, Ryan AA, Britton WJ. Mycobacterium bovisBCG-specific Th17 cells confer partial protection against Mycobacteriumtuberculosis infection in the absence of gamma interferon. Infect Immun2010;78(10):4187–94.

26] Lin Y, Slight SR, Khader SA. Th17 cytokines and vaccine induced immunity.Semin Immunopathol 2010;32(1):79–90.

27] Dubin PJ, Kolls JK. Th17 cytokines and mucosal immunity. Immunol Rev2008;226:160–71.

28] Zygmunt BM, Rharbaoui F, Groebe L, Guzman CA. Intranasal immunizationpromotes Th17 immune responses. J Immunol 2009;183(11):6933–8.

29] Kolls JK, Khader SA. The role of Th17 cytokines in primary mucosal immunity.Cytokine Growth Factor Rev 2010;6:443–8.

30] Menozzi FD, Bischoff R, Fort E, Brennan MJ, Locht C. Molecular characterizationof the mycobacterial heparin-binding hemagglutinin, a mycobacterial adhesin.Proc Natl Acad Sci USA 1998;95:12625–30.

31] Reddy ST, van der Vlies AJ, Simeoni E, Angeli V, Randolph GJ, O’Neil CP, et al.Exploiting lymphatic transport and complement activation in nanoparticle vac-cines. Nat Biotechnol 2007;25(10):1159–64.

32] Manolova V, Flace A, Bauer M, Schwarz K, Saudan P, Bachmann MF. Nanopar-ticles target distinct dendritic cell populations according to their size. Eur JImmunol 2008;38(5):1404–13.

33] Debin A, Kravtzoff R, Vaz Santiago J, Cazales L, Sperandio S, Melber K, et al.Intranasal immunization with recombinant antigens associated with newcationic particles induces strong mucosal as well as systemic antibody and CTLresponses. Vaccine 2002;20:2752–63.

34] Malyala P, Chesko J, Ugozzoli M, Goodsell A, Zhou F, Vajdy M, et al. The potencyof the adjuvant, CpG oligos, is enhanced by encapsulation in PLG microparticles.J Pharm Sci 2008;97:1155–64.

35] Fischer S, Schlosser E, Mueller M, Csaba N, Merkle HP, Groettrup M, et al.Concomitant delivery of a CTL-restricted peptide antigen and CpG ODNby PLGA microparticles induces cellular immune response. J Drug Target2009;17(8):652–61.

36] Krishnamachari Y, Salem AK. Innovative strategies for co-delivering antigensand CpG oligonucleotides. Adv Drug Deliv Rev 2009;61:205–17.

37] Wagner H. The immunogenicity of CpG–antigen conjugates. Adv Drug DelivRev 2009;61:243–7.

38] Yamamoto S, Yamamoto T, Kataoka T, Kuramoto E, Yano O, Tokunaga T.Unique palindromic sequences in synthetic oligonucleotides are requiredto induce IFN and augment IFN-mediated natural-killer activity. J Immunol1992;148(12):4072–6.

39] Lipford GB, Bauer M, Blank C, Reiter R, Wagner H, Heeg K. CpG-containing syn-thetic oligonucleotides promote B and cytotoxic T cell responses to proteinantigen: a new class of vaccine adjuvants. Eur J Immunol 1997;27:2340–4.

40] Klinman DM, Currie D, Gursel I, Verthelyi D. Use of CpG oligodeoxynucleotidesas immune adjuvants. Immunol Rev 2004;199(1):201–16.

41] Klinman DM, Klaschik S, Sato T, Tross D. CpG oligonucleotides as adjuvants forvaccines targeting infectious diseases. Adv Drug Deliv Rev 2009;61:248–55.

42] McGeachy MJ, Cua DJ. Th17 cell differentiation: the long and winding road.Immunity 2008;28(4):445–53.

43] Metwali A, Winckler S, Nedim Ince M, Elliott DE. CpG oligonucleotide stim-ulation of TLR9 inhibits IL-17 production from mesenteric lymph node cells.Gastroenterology 2008;134(4 (Suppl 1)):A–A651.

44] Chen L, Ahmed E, Wang T, Wang Y, Ochando J, Chong AS, et al.TLR signals promote IL-6/IL-17-dependent transplant rejection. J Immunol2009;182(10):6217–25.

45] Tigno-Aranjuez JT, Jaime R, Tuohy VK, Lehmann PV, Tary-Lehmann M.Encephalitogenicity of complete Freund’s adjuvant relative to CpG is linkedto induction of Th17 cells. J Immunol 2009;183(9):5654–61.

46] da Fonseca DM, Silva CL, Wowk PF, Paula MO, Ramos SG, Horn C, et al. Mycobac-terium tuberculosis culture filtrate proteins plus CpG Oligodeoxynucleotidesconfer protection to Mycobacterium bovis BCG-primed mice by inhibitinginterleukin-4 secretion. Infect Immun 2009;77(12):5311–21.

47] da Fonseca DM, Silva CL, Paula MO, Soares EG, Marchal G, Horn C, et al. Increasedlevels of interferon-gamma primed by culture filtrate proteins antigen andCpG-ODN immunization do not confer significant protection against Mycobac-

48] Ballester M, Nembrini C, Dhar N, de Titta A, de Piano C, Pasquier M, et al.Nanoparticle conjugation and pulmonary delivery enhance the protectiveefficacy of Ag85B and CpG against tuberculosis. Vaccine 2011;29(40):6959–66.

6 ccine

[

[

[

[

[

[

[

[

[

[

[motifs induces the de novo generation of Th17 cells in C57BL/6 mice. Eur J

250 C. Verwaerde et al. / Va

49] Todoroff J, Lemaire MM, Fillee C, Jurion F, Renauld JC, Huygen K, et al. Mucosaland systemic immune responses to Mycobacterium tuberculosis antigen 85Afollowing its co-delivery with CpG, MPLA or LTB to the lungs in mice. PLoS One2013;8(5):e63344.

50] Xu L, Wang C, Zhou Y, Ren T, Wen Z. CpG oligonucleotides induce the differentia-tion of CD4(+)Th17 cells by triggering plasmacytoid dendritic cells in adoptivelycell transfer immunotherapy. Immunol Lett 2012;142(1–2):55–63.

51] Hoft DF, Worku S, Kampmann B, et al. Investigation of the relationship betweenimmune-mediated inhibition of mycobacterial growth and other potential sur-rogate markers of protective Mycobacterium tuberculosis immunity. J Infect Dis2002;186:1448–57.

52] Cowley SC, Elkins KL. CD4+ T cells mediate IFN-�-independent control ofMycobacterium tuberculosis infection both in vitro and in vivo. J Immunol2003;171:4689–99.

53] Abebe F. Is interferon-gamma the right marker for bacille Calmette-Guérin-

induced immune protection? The missing link in our understanding oftuberculosis immunology. Clin Exp Immunol 2012;169(3):213–9.

54] Okamoto Yoshida Y, Umemura M, Yahagi A, O’Brien R, Ikuta K, Kishihara K, et al.Essential role of IL-17A in the formation of a mycobacterial Infection inducedgranuloma in the lung. J Immunol 2010;184:4414–22.

[

32 (2014) 6240–6250

55] Freches D, Korf H, Denis O, Havaux X, Huygen K, Romano M. Mice geneticallyinactivated in interleukin-17A receptor are defective in long-term con-trol of Mycobacterium tuberculosis infection. Immunology 2013;140(2):220–31.

56] Murphy KM1, Stockinger B. Effector T cell plasticity: flexibility in the face ofchanging circumstances. Nat Immunol 2010;11(8):674–80.

57] Moseman EA1, Liang X, Dawson AJ, Panoskaltsis-Mortari A, Krieg AM,et al. Human plasmacytoid dendritic cells activated by CpG oligodeoxynu-cleotides induce the generation of CD4 + CD25+ regulatory T cells. J Immunol2004;173(7):4433–42.

58] LaRosa DF, Gelman AE, Rahman AH, Zhang J, Turka LA, et al. CpG DNA inhibitsCD4 + CD25+ Treg suppression through direct MyD88-dependent costimula-tion of effector CD4+ T cells. Immunol Lett 2007;108(2):183–8.

59] Wu W1, Huang L, Mendez S. A live Leishmania major vaccine containing CpG

Immunol 2010;40(9):2517–27.60] Denis F, Alain S, Ploy MC. New routes of administration: epidermal,

transcutaneous mucosal ways of vaccination. Med Sci (Paris) 2007;23(4):379–85.