Three CpG oligodeoxynucleotide classes differentially enhance antigen-specific humoral and cellular immune responses in mice

Vaccine 29 (2011) 5778–5784

Contents lists available at ScienceDirect

Vaccine journal homepage: www.elsevier.com/locate/vaccine

Three CpG oligodeoxynucleotide classes differentially enhance antigen-specific humoral and cellular immune responses in mice Ying Liu a,b , Xing Luo a,b , Chunting Yang b , Sanke Yu a , Honglin Xu b,∗ a b

Department of Veterinary Medicine, Northwest A&F University, Shaanxi 712100, People’s Republic of China Department of Virology, National Vaccine and Serum Institute, No. 4 San Jian Fang Nan Li, Chao Yang District, Beijing 100024, People’s Republic of China

a r t i c l e

i n f o

Article history: Received 27 February 2011 Received in revised form 18 April 2011 Accepted 17 May 2011 Available online 12 June 2011 Keywords: CpG oligodeoxynucleotides CpG-ODN Vaccine Adjuvant

a b s t r a c t Synthetic oligodeoxynucleotides containing unmethylated CpG-dinucleotides (CpG-ODNs) are immunostimulatory in a broad spectrum of species. Extensive studies provide evidence that CpGODNs are effective as immunotherapeutics and vaccine adjuvants in various clinical settings. Three major classes of immunostimulatory CpG-ODNs are well characterized according to their in vitro activities and chemical compositions. However, it remains largely unclear whether and how these differences translate in vivo and in particular when used as vaccine adjuvants. In the present study, a panel of CpG-ODNs, including four representative sequences respectively from each class, was used to characterize their adjuvant activities in mice. The results demonstrated that three CpG-ODN classes can differentially affect antigen-specific humoral and cellular immune responses. Specifically, the Band C-class CpG-ODNs induce a potent Th1-biased immunity with comparable antibody levels as well as CD4+ and CD8+ T cell responses. In contrast, although the A-class CpG-ODNs can weakly enhance antibody titers and CD8+ T cell response regarding cytotoxic activity, they are not able to change the IgG1/IgG2a ratio or elicit antigen-specific, IFN-␥-secreting CD4+ and CD8+ T cells. Consistent with this, three CpG-ODN classes provide differential antigen-specific protection against Listeria monocytogenes, an intracellular bacterial infection. In conclusion, our study provides not only better knowledge about the adjuvant activities of three CpG-ODN classes but also implications for the rational design of CpG-ODN adjuvants. © 2011 Elsevier Ltd. All rights reserved.

1. Introduction Synthetic oligodeoxynucleotides containing unmethylated CpG-dinucleotides (CpG-ODN) are broadly immunostimulatory in various species including humans and mice [1–3]. CpG-ODNs directly activate dendritic cells (DCs) and B cells through Toll-like receptor 9 (TLR9) [4]. Other immune cells that lack TLR9, such as nature killer (NK) cells and T cells, are indirectly activated by CpG-ODN-induced cytokines and chemokines [1–3]. The response to CpG-ODNs skews the immune milleu in favor of Th1-type responses, an effect that underlies their use as immunoprotective agents and vaccine adjuvants in various clinical settings, particularly in the prevention and therapy of cancers, infectious diseases and allergies [1–3]. Extensive preclinical as well as clinical studies provide evidence that CpG-ODNs are effective for each of these applications [1–3].

∗ Corresponding author. Tel.: +86 10 65756772. E-mail address: [email protected] (H. Xu). 0264-410X/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.vaccine.2011.05.087

The immune properties of CpG-ODNs vary depending on CpG motifs, flanking sequence, length, backbone modification and formation of secondary and tertiary structures [1]. According to their in vitro activities and chemical compositions, three major classes of immunostimulatory CpG-ODNs are well described [5–8]. Aclass CpG-ODNs have a native phosphodiester palindromic central region flanked by nuclease resistant phosphorothiate backbones at both ends [5,6]. They form stable but complex higher-ordered structures known as G-tetrads through 3 poly G interaction [5,6]. A-class CpG-ODNs are characterized by strong plasmacytoid dendritic cell (pDCs) and NK cell activation, resulting in high levels of IFN-␣ and IFN-␥ production [5,6]. B-class CpG-ODNs are completely phosphorothioate-modified and do not form secondary structures [5]. These molecules are characterized by strong B cell activation [5]. C-class CpG-ODNs are also phosphorothioate sequences with a 3 palindrome permitting the formation of duplexes [5,7,8]. They have immune activities intermediate to the A- and B-classes, and can activate not only B cells but also pDCs and NK cells as well [5,7,8]. However, it remains largely unknown whether and how these differences of three CpG-ODN classes translate in vivo and in par-

Y. Liu et al. / Vaccine 29 (2011) 5778–5784 Table 1 CpG-ODNs panel used in the present study.

5779

2.3. B cell proliferation assay

Classes

Names

Sequences

References

A

D19 1585 2216 2204

GGTGCATCGATGCAGGGGGG GGGGTCAACGTTGAGGGGGG GGGGGACGATCGTCGGGGGG GGGGTCATCGATGAGGGGGG

[12] [6] [6] [6]

B

D2 1826m 1018 2006

TGTCGTCGTCGTTTGTCGTTTGTCGTT TCCATGTCGTTCCTGTCGTT TGACTGTGAACGTTCGAGATGA TCGTCGTTTTGTCGTTTTGTCGTT

[26] [26] [27] [28]

C

C10109 C274 C583 C594

TCGTCGTTTTACGGCGCCGTGCCG TCGTCGAACGTTCGAGATGAT TCGAACGTTCGAACGTTCGAAT TCGTCGATCGATCGAGATGAT

[9] [8] [8] [8]

ticular when used as vaccine adjuvants [3]. The prototypic B-class sequences are well studied and proven to be effective Th1 adjuvants in humans and various animals including mice [1–3]. Limited reports have shown that C-class sequences can also induce a comparable or even better humoral immune response than that of the B-class sequences in mice and guinea pigs [5,9]. Based on antibody levels elicited in rhesus monkeys using HBsAg or OVA as antigens, two studies concluded that A-class CpG-ODNs are as effective as B-class CpG-ODNs, while the third study suggested that the former could be superior to the latter [10–12]. These studies only compared a single sequence or at most a sequence mixture from either class [5,10–12]. Consequently, no clear conclusion could be drawn from these limited data. Moreover, although B-class CpGODNs can enhance both CD4+ and CD8+ T cell responses, little is known for the other two classes in this regard, and no study has yet compared their adjuvant effects on the cellular branch of immune responses. Therefore, it is necessary to comprehensively characterize the adjuvant activities of three CpG-ODN classes, which could provide better knowledge about their in vivo activities and lead to more potent vaccine adjuvants. In the present study, a panel of CpGODNs including four representative sequences respectively from each class was used to characterize their adjuvant activities. The results demonstrated that three CpG-ODN classes can differentially affect antigen-specific humoral and cellular immune responses in mice. 2. Materials and methods 2.1. Oligodeoxynucleotides Completely or partially phosphorothioate-modified CpG-ODNs were synthesized by TaKaRa Biotech (Dalian, China) and purified by ion-exchange HPLC. The final products had more than 95% of full-length sequences and contained undetectable levels of endotoxin. The CpG-ODN sequences used in the present study are listed in Table 1, in which the phosphodiester backbones in A-class sequences are underlined. 2.2. IFN- secretion assay Splenocytes were isolated from naïve BALB/c mice, erythrocytes were lysed by NH4 Cl treatment, and viability of the cells was more than 95% as verified by trypan blue exclusion assay. The cells were cultured in RPMI-1640 medium supplemented with 10% fetal calf serum (Hyclone) at 2 × 106 cells/well in 96-well plates. CpG-ODNs were added at a final concentration of 1 ␮M. The plates were incubated for 48 h in a 37 ◦ C/5% CO2 humidified incubator, and supernatants were assayed for IFN-␥ production with commercially available ELISA kit (BD Biosciences).

Splenocytes from naïve BALB/c mice (isolated as described above) were labeled with 5 ␮M CFSE, and cultured in RPMI-1640 medium supplemented with 10% fetal calf serum (Hyclone) at 1 × 106 cells/well in 24-well plates. CpG-ODNs were added at a final concentration of 2 ␮g/ml. The plates were incubated for 72 h in a 37 ◦ C/5% CO2 humidified incubator, and the cells were stained with PE-conjugated anti-B220 antibody (BD Biosciences). The cells were collected on a FACSCalibur machine (BD Biosciences) and the data were analyzed with FlowJo software (Tree Star). 2.4. Mice immunization Six to eight weeks old female BALB/c or C57BL/6 mice (SPF grad) were immunized once or twice intramuscularly (i.m.) in 100 ␮l of total volume with 20 ␮g ovalbumin (OVA) adsorbed to 100 ␮g Al(OH)3 (alum) alone or in combination with 10 ␮g CpG-ODNs. Sera were collected 4 weeks post-priming for serological analysis, and analysis of cellular immune responses and listeria challenge were performed 10 days post-boosting. 2.5. ELISA 96-well MaxiSorp plates (Nunc) were coated overnight at 4 ◦ C with OVA at 200 ng/well. The plates were washed with PBS containing 0.05% Tween 20 (PBST) and blocked 1 h with PBS containing 5% calf serum (PBSC). Sera were diluted in PBSC by 2-fold serial dilution and incubated at room temperature for 1 h. After washing with PBST, the bound antibodies were detected with HRP-conjugated anti-mouse IgG (Sigma–Aldrich), IgG1 and IgG2a (Southern Biotech). Antibody titers were determined for each sample as end-point dilution with an arbitrary cutoff OD value of 0.100. Antibody titers of each group were transformed logarithmically and expressed as geometric means ± standard errors (GMT ± SE). 2.6. ELISPOT The number of IFN-␥-secreting splenocytes in response to OVA antigen was determined by ELISPOT using commercial available kit (BD Biosciences). Briefly, 96-well filtration plates were coated overnight at 4 ◦ C with anti-mouse IFN-␥ antibody at 500 ng/well. After blocking, the plates were overlaid with 5 × 105 splenocytes/well, and incubated for 20 h in a 37 ◦ C/5% CO2 humidified incubator in the presence of 5 ␮g/ml OVA. After extensive washing, the bound IFN-␥ was detected with biotinylated anti-mouse IFN-␥ antibody and HRP-conjugated streptavidin. Spots were counted on a CTL ImmunoSpot Analyzer machine (CTL). 2.7. Intracellular cytokine staining Splenocytes were isolated 10 days post-boosting as described above. The cells were cultured in RPMI-1640 medium supplemented with 10% fetal calf serum (Hyclone) at 4 × 106 cells/well in 96-well plates. OVA257–264 (SIINFEKL) peptide was used for re-stimulation at a concentration of 5 ␮g/ml. The plates were incubated for 6 h in a 37 ◦ C/5% CO2 incubator. For the last 3 h of re-stimulation, 10 ␮M monensin was added to block the secretion of cytokines. The cells were harvested and stained with FITC-conjugated anti-mouse CD8 antibody (BD Biosciences). After fixation with 4% paraformaldehyde and permeabilization with 0.15% saponin (Sigma–Aldrich), the cells were then stained with APC-conjugated anti-mouse IFN-␥ antibody (BD Biosciences). The cells were collected on a FACSCalibur machine (BD Biosciences) and the data were analyzed with FlowJo software (Tree Star).

5780

Y. Liu et al. / Vaccine 29 (2011) 5778–5784

2.8. In vivo cytotoxic T lymphocyte (CTL) assay The in vivo CTL assay was performed as described by Barber et al. [13]. In brief, splenocytes from naïve C57BL/6 were labeled with 5 ␮M or 0.5 ␮M CFSE. The CFSEhigh cells were pulsed with 5 ␮g/ml OVA257–264 peptide for 4 h at 37 ◦ C/5% CO2 . The cells were washed and pooled at a 1:1 ratio and transferred retro-orbitally into the immunized recipient mice (2 million cells per mouse in 100 ␮l volume). After overnight (about 15 h) killing, splenocytes were isolated from the recipients and analyzed using flow cytometry for target cell clearance. Gated on CFSE+ cells, the percentage of antigen-specific killing was calculated using the following formula: 100 − [(% CFSEhigh in immunized mice/% CFSElow in immunized mice)/(%CFSEhigh in naïve mice/% CFSElow in naïve mice)] × 100. 2.9. Listeria challenge The recombinant OVA-expressing Listeria monocytogenes strain LM-OVA was provided by Lefrancois [14]. LM-OVA was grown in brain–heart infusion (BHI) broth supplemented with 5 ␮g/ml erythromycin. Ten days post-boosting, the mice were challenged retro-orbitally with 1 × 106 CFU of the bacteria in 100 ␮l of saline. The bacterial dose was verified by plating on BHI agar plates. Three days after infection, the spleens were homogenized and lysed in sterile water with 0.2% NP-40, serial dilutions of homogenates were plated on BHI agar plates, and colonies were counted after incubation at 37 ◦ C for 24 h. 2.10. Statistical analysis Statistical significance of differences was determined by performing an analysis of variance (ANOVA) test, followed by Dunnett’s t-test, where p < 0.05 was considered to be statistically significant. 3. Results 3.1. Verification of three CpG-ODN classes with distinct immunostimulatory activities In order to comparatively study the vaccine adjuvant activities of different classes of CpG-ODN, we first collected from literatures a panel of published representative sequences, which include four A-class CpG-ODNs (D19, 1585, 2216 and 2204), four B-class CpG-ODNs (D2, 1826m, 1018 and 2006), and four C-class CpG-ODNs (C10109, C274, C583 and C594). Murine splenocytes were incubated with the sequences to verify their in vitro immunostimulatory activities. As expected, all the A-class CpGODNs induced high levels of IFN-␥ but no proliferation of B cells (Fig. 1A and B). In contrast, the B-class CpG-ODNs induced B cell proliferation but no IFN-␥ secretion (Fig. 1A and B). The C-class CpG-ODNs had properties intermediate to the A- and Bclass sequences, inducing weak but detectable levels of IFN-␥ and strong B cell proliferation as that of the B-class sequences (Fig. 1A and B). 3.2. Three CpG-ODN classes differentially enhance the humoral immune responses To compare the adjuvant effects of three CpG-ODN classes on humoral immune responses, groups of six BALB/c mice were immunized i.m. once with OVA adsorbed to alum alone (the control group) or in combination with various classes of CpG-ODNs. Sera were collected 4 weeks later to measure the antigen-specific antibody levels by ELISA. Compared with the control group, all

Fig. 1. Verification of three CpG-ODN classes with distinct immunostimulatory activities. Splenocytes from naïve BALB/c mice were incubated with various classes of CpG-ODNs. Supernatants were harvested at 48 h to measure IFN-␥ production by ELISA (A), and B cell proliferation was analyzed at 72 h as the dilution of CFSE by flow cytometry (B). Histograms are gated on B220+ cells and overlaid with the control cells without CpG-ODNs stimulation (gray histogram). The results are reproducible in two experiments and the representative data are shown here.

the B- and C-class CpG-ODNs induced significant and comparable increase of the antigen-specific IgG titers by 8–25 folds (Fig. 2A). However among the A-class CpG-ODNs, only 2216 and 2204 weakly increased the antigen-specific IgG levels by 3–4 folds (Fig. 2A). Analysis of the antigen-specific IgG isotypes revealed the similar trends of IgG1 and IgG2a levels as that of IgG for all classes of CpG-ODNs (Fig. 2B and C). Specifically, the A-class CpG-ODNs induced no increase or weak increase of both isotypes by 3–5 folds, while all the B- and C-class CpG-ODNs induced 4–11fold increase of the IgG1 titers and dramatic increase of the IgG2a titers by about 200–300 folds (Fig. 2B and C). According to the antigen-specific IgG1 to IgG2a ratio, all the B- and Cclass CpG-ODNs induced a Th1-biased humoral immune response as expected (Fig. 2D). Unexpectedly but interestingly, although 2216 and 2204 consistently enhanced the antigen-specific IgG as well as the isotype levels, they induced a decent Th2-type of humoral immune response as that of the control group, which was immunized with the classical Th2-biased adjuvant, alum (Fig. 2D).

Y. Liu et al. / Vaccine 29 (2011) 5778–5784

5781

Fig. 2. Three CpG-ODN classes differentially enhance the humoral immune responses against OVA. Groups of BALB/c mice (n = 6) were immunized i.m. with OVA (20 ␮g) adsorbed to alum (100 ␮g) alone (control) or in combination with various CpG-ODNs (10 ␮g). Sera were collected 4 weeks later to measure the antigen-specific IgG(A), IgG1(B), IgG2a(C) and IgG1/IgG2a ratio(D). Antibody titers for each group of mice were transformed logarithmically and expressed as GMT ± SE. Compared with the control group, NS-Not Significant, *p < 0.05, **p < 0.01 and ***p < 0.001 was determined by ANOVA, followed by Dunnett’s t-test.

3.3. Three CpG-ODN classes differentially enhance the cellular immune responses To compare the adjuvant effects of three CpG-ODN classes on cellular immune responses, groups of C57BL/6 mice were immunized i.m. twice with OVA plus alum and various classes of CpG-ODNs. Ten days post-boosting immunization, antigen-specific T cell responses were assessed in vitro or in vivo by various assays. As revealed by in vitro re-stimulation of splenocytes with OVA protein, all the B- and C-class CpG-ODNs induced comparable increase of the antigen-specific, IFN-␥-secreting splenocytes, while no response could be seen for the A-class CpG-ODNs (Fig. 3A). These OVAspecific, IFN-␥-secreting splenocytes were confirmed to be CD4+ T cells by an intracellular IFN-␥ staining assay, in which anti-CD4, anti-CD8 and anti-IFN-␥ triple-color staining was performed after overnight re-stimulation of the splenocytes with OVA protein (data not shown). Intracellular cytokine staining assay and in vivo CTL assay were performed to evaluate OVA257–264 peptide-specific CD8+ T cell response. Consistent with the CD4+ T cell response, all the B- and C-class CpG-ODNs induced comparable antigen-specific CD8+ T cell response as revealed by both assays (Fig. 3B–E). In contrast, more complicated but interesting results were seen for all the A-class CpG-ODNs. Despite they can not enhance the antigen-specific CD8+ T cell response in terms of IFN-␥-secreting (Fig. 3B and C), those mice immunized with the A-class CpG-ODNs indeed elicited an in vivo CTL response, as demonstrated by the clearance of about 40% of the OVA257–264 peptide-pulsed target cells (Fig. 3D and E). An intracellular bacterial infection model was used to further examine the antigen-specific T cell-mediated protection induced with three CpG-ODN classes. Ten days post-boosting immunization, the mice were challenged retro-orbitally with a recombinant

strain of L. monocytogenes expressing OVA as a model antigen. Compared with the naïve mice, all the groups immunized with the B- and C-class CpG-ODNs showed significant protection, as revealed by about 3–4 log 10s lower of bacterial burdens in the spleens (Fig. 3F). In contrast among the A-class CpG-ODNs, 2204 did not show protection while the other sequences only decreased the bacterial loads by 1–2 log 10s (Fig. 3F). 4. Discussion Three major classes of immunostimulatory CpG-ODNs are well characterized based on their in vitro activities on immune cells and chemical modification of backbones [5–8]. However, little is known whether and how these differences translate in vivo when used as vaccine adjuvants [3]. Here a panel of representative CpG-ODNs from each class was compared for their adjuvant effects in mice using OVA as a model antigen. The results showed clearly that antigen-specific humoral and cellular immunity can be affected differentially by three CpG-ODN classes. More specifically, the B- and C-class CpG-ODNs are both potent Th1-type adjuvants, inducing comparable antibody levels as well as CD4+ and CD8+ T cell responses. In contrast, despite the A-class CpG-ODNs can weakly enhance antibody titers and CD8+ T cell response regarding cytotoxic activity, they are not able to change the IgG1/IgG2a ratio or elicit antigen-specific, IFN-␥-secreting T cell response. Consequently, three CpG-ODN classes provide different levels of antigen-specific protection against L. monocytogenes infection. Jegerlehner et al. recently demonstrated that TLR9 expression in B cells but not other antigen presenting cells (APCs) is critical for a B-class CpG-ODN to enhance Th1-type humoral immune response [15]. Given the facts that C-class CpG-ODNs can activate both B cells and pDCs [5,7,8], and the latter can induce plasma cell differ-

5782

Y. Liu et al. / Vaccine 29 (2011) 5778–5784

Fig. 3. Three CpG-ODN classes differentially enhance the cellular immune responses against OVA. Groups of C57BL/6 mice were primed on day 0 and boosted on day 28 i.m. with OVA (20 ␮g) plus alum (100 ␮g) and various CpG-ODNs (10 ␮g). Analysis of the cellular immune responses (A)–(E) and listeria challenge (F) were performed 10 days post-boosting. (A) The number of IFN-␥-secreting splenocytes in response to OVA protein re-stimulation was determined by ELISPOT (n = 3). (B) and (C) Splenocytes were stained for intracellular IFN-␥ after OVA257–264 peptide re-stimulation (+ peptide, gray column) or no stimulation as control (− peptide, blank column). The representative plots (B) and the total number (C) of IFN-␥-secreting CD8+ T cells in the spleens are shown (n = 6). (D) and (E) CFSE-labeled, peptide-loaded or un-loaded C57BL/6 splenocytes were transferred into the immunized recipient mice. In vivo killing of the target cells was assessed 15 h later by flow cytometry. Histograms are gated on CFSE+ cells (D) and the Mean ± SE of antigen-specific killing from indicated group of mice are shown in (E) (n = 6). (F) The mice were challenged retro-orbitally with LM-OVA at a dose of 1 × 106 CFU/mouse, and the bacterial burdens in the spleens were determined 3 days later (n = 12). The dash line shows the detecting limit of 50 CFU/spleen. Compared with the control group, NS-Not Significant, *p < 0.05, **p < 0.01 and ***p < 0.001 was determined by ANOVA, followed by Dunnett’s t-test.

Y. Liu et al. / Vaccine 29 (2011) 5778–5784

entiation through type I interferon (IFN-I) and interleukin-6 [16], therefore it is reasonable to speculate that C-class CpG-ODNs could be more potent adjuvants compared with B-class CpG-ODNs. However, as demonstrated in the present study, all the B- and C-class CpG-ODNs induce comparable Th1-type humoral immunity. This result is somewhat unexpected but consistent with Jegerlehner’s finding [15], suggesting that the immunostimulatory activities on pDCs and NK cells could be redundant in the presence of strong B cell activation when C-class CpG-ODNs are used as vaccine adjuvants. The prototypic B-class CpG-ODNs are well-characterized Th1type adjuvants, eliciting both CD4+ and CD8+ T cell responses when combined with protein antigens in both humans and mice [1–3]. Although they can activate various APCs in mice, at least one study clearly showed that B-class CpG-ODN-induced Th1 CD4+ T cell response is dependent on direct activation of B cells [17]. Eliciting effective CD8+ T cell response against exogenous protein antigens requires cross-presentation and cross-priming by professional APCs, such as DCs and B cells [18]. Interestingly, B cells are proven to be necessary and sufficient for the cross-priming of antigen-specific CD8+ T cells by B-class CpG-ODNs [17,19,20]. Our study showed that C-class CpG-ODNs can enhance both CD4+ and CD8+ T cell responses similarly as B-class sequences, although they have additional immunostimulatory activities on pDCs and NK cells [5,7,8]. Therefore, in supporting the previous findings [17,19,20], our observation also suggest that direct activation of B cells could be the primary immunopotentiating mechanism whatever B- or Cclass CpG-ODNs are used as vaccine adjuvants. However, we still need to use B cell-deficient ␮MT mice or the mice with conditional deficiency of TLR9 in B cells to further characterize the potential mechanism. Previous studies established that IFN-I can potently enhance humoral immunity and promote IgG isotype switching, and adjuvant effects mediated by CpG-ODNs are diminished in IFN-I receptor or IFN-␥ deficient mice [21–24]. Therefore, A-class CpGODNs could be potent vaccine adjuvants considering they can induce large amounts of IFN-I and IFN-␥ through strong activation of both pDCs and NK cells [5,6]. However, although four A-class CpG-ODNs used in the present study are comparable regarding their in vitro immunostimulatory activities, when compared with the B- and C-class sequences, they showed weaker adjuvant activities, which is also inconsistent and variable among different sequences and different immune parameters examined. Furthermore, A-class CpG-ODNs are not able to change the IgG1/IgG2a ratio or induce IFN-␥-secreting CD4+ and CD8+ T cells, but indeed elicit a detectable CD8+ CTL response. This phenomenon has never been revealed before and cannot be simply explained by any existing immunological theory because it is well recognized that induction of CD8+ CTL response requires Th1 immune milleu. In contrast to our findings, several previous studies performed in rhesus monkeys demonstrated that A-class CpG-ODNs are as effective as or even superior to B-class sequences for protein antigens [10–12]. This discrepancy could be possibly contributed by different animal models used in the studies and need to be further clarified in the future. Collectively, through comprehensive characterization of the adjuvant activities of three CpG-ODN classes, our results provide implications for the rational design of CpG-ODN adjuvant. Compared with A-class CpG-ODNs, B- and C-class sequences are more efficient in inducing both humoral and cellular immunity. The safety and efficacy of B-class CpG-ODNs as vaccine adjuvants has been confirmed in human clinical trials [1–3]. However, for both A- and C- class CpG-ODNs, the safety needs to be further evaluated considering that IFN-I production by pathological activation of pDCs has been demonstrated to be involved in psoriasis [25]. Therefore, choice of B- instead of A- and C-class CpG-ODNs as vac-

5783

cine adjuvants could ensure maximal immunopotentiating effects while avoiding unwanted activation of pDC.

Acknowledgements This work was funded by grants 2008ZX10002-002 (to Xu HL), 2010AA022907 (to Xu HL) and 2009ZX10004-404 (to Yang CT) from People’s Republic of China.

References [1] Krieg AM. Therapeutic potential of Toll-like receptor 9 activation. Nat Rev Drug Discov 2006;5(6):471–84. [2] Klinman DM. Immunotherapeutic use of CpG oligodeoxynucleotides. Nat Rev Immunol 2004;4(4):1–10. [3] Davis HL. TLR9 agonists for immune enhancement of vaccines. In: New Generation Vaccines. fourth ed. New York: Informa Healthcare USA Inc; 2010, 263–272. [4] Hemmi H, Takeuchi O, Kawai T, Kaisho T, Sato S, Sanjo H, et al. A Toll-like receptor recognizes bacterial DNA. Nature 2000;408(6813):740–5. [5] Vollmer J, Weeratna R, Payette P, Jurk M, Schetter C, Laucht M, et al. Characterization of three CpG oligodeoxynucleotide classes with distinct immunostimulatory activities. Eur J Immunol 2004;34(1):251–62. [6] Krug A, Rothenfusser S, Hornung V, Jahrsdorfer B, Blackwell S, Ballas ZK, et al. Identification of CpG oligonucleotide sequences with high induction of IFN-␣/␤ in plasmacytoid dendritic cells. Eur J Immunol 2001;31(7):2154–63. [7] Hartmann G, Battiany J, Poeck H, Wagner M, Kerkmann M, Lubenow N, et al. Rational design of new CpG oligonucleotides that combine B cell activation with high IFN-␣ induction in plasmacytoid dendritic cells. Eur J Immunol 2003;33:1633–41. [8] Marshall JD, Fearon K, Abbate C, Subramanian S, Yee P, Gregorio J, et al. Identification of a novel CpG DNA class and motif that optimally stimulate B cell and plasmacytoid dendritic cell functions. J Leukoc Biol 2003;73(6):781–92. [9] Gu M, Hine PM, Jackson WJ, Giri L, Nabors GS. Increased potency of BioThrax® anthrax vaccine with the addition of the C-class CpG oligonucleotide adjuvant CpG 10109. Vaccine 2007;25:526–34. [10] Verthelyi D, Wang VW, Lifson JD, Klinman DM. CpG oligodeoxynucleoties improve the response to hepatitis B immunization in healthy and SIV-infected rhesus macaques. AIDS 2004;18(7):1003–8. [11] Hartmann G, Marschner A, Viveros PR, Hennig CS, Eisenblatter M, Suh YS, et al. CpG oligonucleotides induce strong humoral but only weak CD4+ T cell responses to protein antigens in rhesus macaques in vivo. Vaccine 2005;23:3310–7. [12] Verthelyi D, Kenney RT, Seder RA, Gam AA, Friedag B, Klinman DM. CpG oligodeoxynucleotides as vaccine adjuvant in primates. J Immunol 2002;168:1659–63. [13] Barber DL, Wherry EJ, Ahmed R. Cutting edge: rapid in vivo killing by memory CD8T cells. J Immunol 2003;171(1):27–31. [14] Pope C, Kim SK, Marzo A, Masopust D, Williams K, Jiang J, et al. Organspecific regulation of the CD8T cellresponse to Listeria monocytogenes infection. J Immunol 2001;166:3402–9. [15] Jegerlehner A, Maurer P, Bessa J, Hinton HJ, Kopf M, Bachmann MF. TLR9 signaling in B cells determines class switch recombination to IgG2a. J Immunol 2007;178:2415–20. [16] Jego G, Palucka AK, Blanck JP, Chalouni C, Pascual V, Banchereau J. Plasmacytoid dendritic cells induce plasma cell differentiation through type I interferon and interleukin 6. Immunity 2003;19:225–34. [17] Hoft DF, Eickhoff CS, Giddings OK, Vasconcelos JC, Rodrigues MM. Trans-sialidase recombinant protein mixed with CpG motif-containing oligodeoxynucleotide induces protective mucosal and systemic Trypanosoma cruzi immunity involving CD8+ CTL and B cell-mediated cross-priming. J Immunol 2007;179:6889–90. [18] Kurts C, Robinson BW, Knolle PA. Cross-priming in health and disease. Nat Rev Immunol 2010;10(6):403–14. [19] Jiang W, Lederman MM, Harding CV, Sieg SF. Presentation of soluble antigens to CD8+ T cells by CpG oligodeoxynucleotide-primed human naïve B cells. J Immunol 2011;186:2080–6. [20] Heit A, Huster KM, Schmitz F, Schiemann M, Busch DH, Wagner H. CpG-DNA aided cross-priming by cross-presenting B cells. J Immunol 2004;172:1501–7. [21] Bon AL, Schiavoni G, D’Agostino G, Gresser I, Belardelli F, Tough DF. Type I interferons potently enhance humoral immunity and can promote isotype switching by stimulating dendritic cells in vivo. Immunity 2001;14:461–70. [22] Bon AL, Thompson C, Kamphuis E, Durand V, Rossmann C, Kalinke U, et al. Enhancement of antibody responses through direct stimulation of B and T cells by type I IFN. J Immunol 2006;176:2074–8. [23] Van Uden JH, Tran CH, Carson DA, Raz E. Type I interferon is required to mount an adaptive response to immunostimulatory DNA. Eur J Immunol 2001;31:3281–90. [24] Rosa DS, Bastos KR, Bargieri DY, Tzelepis F, Nomizo A, Russo M, et al. Role of interferon-␥ during CpG oligodeoxynucleotide-adjuvanted immunization with recombinant proteins. Vaccine 2007;25:6007–17.

5784

Y. Liu et al. / Vaccine 29 (2011) 5778–5784

[25] Nestle FO, Conrad C, Tun-Kyi A, Homey B, Gombert M, Boyman O. Plasmacytoid predendritic cells initiate psoriasis through interferon-alpha production. J Exp Med 2005;202(1):135–43. [26] Xu HL, Wang SQ, Wang SF, Guo F, Lu RJ, Ruan L. Delineation of CpG motifs for activating human immune cells. Chin J Microbiol Immunol 2001;21(5): 471–5.

[27] Halperin SA, Nest G, Smith B, Abtahi S, Whiley H, Eiden JJ. A phase I study of the safety and immunogenicity of recombinant hepatitis B surface antigen coadministered with an immunostimulatory phosphorothioate oligonucleotide adjuvant. Vaccine 2003;21:2461–7. [28] Hartmann G, Krieg AM. Mechanism and function of a newly identified CpG DNA motif in human primary B cells. J Immunol 2000;164:944–53.