End of Oligodeoxynucleotides Abrogates Toll-Like Receptor 9

Dec 18, 2009 - Peptide Conjugation at the 5′-End of Oligodeoxynucleotides Abrogates ... Synthetic oligodeoxynucleotides containing unmethylated CpG...
0 downloads 0 Views 842KB Size
Download PDF
Bioconjugate Chem. 2010, 21, 39–45

39

Peptide Conjugation at the 5′-End of Oligodeoxynucleotides Abrogates Toll-Like Receptor 9-Mediated Immune Stimulatory Activity Mallikarjuna Reddy Putta, Fu-Gang Zhu, Daqing Wang, Lakshmi Bhagat, Meiru Dai, Ekambar R. Kandimalla, and Sudhir Agrawal* Idera Pharmaceuticals, Inc., 167 Sidney Street, Cambridge, Massachusetts 02139. Received June 4, 2009; Revised Manuscript Received November 30, 2009

Bacterial and synthetic DNA containing unmethylated CpG motifs act as ligands of Toll-like receptor 9 (TLR9). Our earlier studies showed that 5′-accessibility of synthetic oligodeoxynucleotides containing CpG motif (ODN) is required for TLR9-mediated immune stimulatory activity. Blocking the 5′-end of ODN through conjugation to a variety of moieties reduces immune stimulatory activity (Bioconjugate Chem. 2002, 13, 966-974). In the present study, we conjugated a model peptide, a 28-amino-acid-long β-amyloid peptide, to either the 5′- or the 3′-end of an ODN via C3 and C6 alkyl linkers. We compared the immune stimulatory activity of the resulting conjugates with that of a parent ODN without conjugation in TLR9-transfected cells, mouse spleen cell cultures, and in vivo in mice. ODN with the peptide conjugated at the 3′-end via C3 and C6 linkers had immune stimulatory activity similar to that of the parent ODN in both in vitro and in vivo in mice. On the contrary, conjugation of peptide at the 5′-end of the ODN significantly abrogated immune stimulatory activity. In conclusion, the results presented here demonstrate that peptide/protein conjugation to ODN is optimal at the 3′-end with either C3 or C6 linker and conjugation at the 5′-end leads to significant loss of TLR9-mediated immune stimulation.

INTRODUCTION Toll-like receptor 9 (TLR9) belongs to a family of conserved pathogen-associated molecular pattern recognition receptors and is predominantly expressed in the endosomal membranes of human B cells and plasmacytoid dendritic cells (pDCs). Synthetic oligodeoxynucleotides containing unmethylated CpG motifs (ODNs) act as ligands of TLR9 (1). Stimulation of TLR9 with ODN leads to the activation of MyD88-dependent NF-κB andMAPKpathways,resultinginTh1-typeimmuneresponses(2,3). Agonists of TLR9 are currently at various stages of clinical evaluation for the treatment of cancers, allergies, and infectious diseases and as adjuvants with vaccines (1–4). Agonists of TLR9 (ODNs) induce potent Th1-type immune responses and show encouraging results as adjuvants when coadministered with DNA/protein/peptide vaccines (5–12). In addition to coadministration of ODN with vaccines/antigens (proteins), studies have been carried out to deliver both protein and ODN to the same antigen-presenting cell (APC) either by coformulation of both agents in liposomes or by conjugating proteins to ODN to increase immune responses (13). Some of the peptides/proteins conjugated to ODNs include Amb a 1, OVA, β-galactosidase, and HIV gp120 (14–17). Proteins and peptides are commonly conjugated to the 5′-end of ODN, because the chemical modifications required for conjugation reactions at the 5′-end are easily accomplished during the solidphase synthesis (18). Recently, a conjugate of Amb a 1-ODN failed to meet the end points in a clinical study, prompting us to perform the present study to further elucidate the impact of 5′-conjugation of peptide to ODN on immune stimulatory activity. Our extensive structure-activity relationship studies of ODNs have identified critical structural features in the pentose sugar (19–22), phosphate backbone (23), nucleobases (24, 25), and nucleosides (26) required for TLR9 stimulation. Our * Tel.: 617-679-5501, Fax: 617-679-5572; e-mail: sagrawal@ iderapharma.com.

previous studies showed that the 5′-end of an ODN must be accessible for TLR9-mediated immune stimulatory activity. Modifications such as conjugation of di-, tetra-, or longer oligonucleotide, a fluorescein molecule or hairpin structures at the 5′-end block 5′-accessibility, resulting in significant abrogation of TLR9-mediated immune stimulatory activity (27, 28). In the present study, we synthesized 5′- and 3′-conjugates of ODN and a model 28-mer β-amyloid peptide using C3 and C6 linkers and compared their immune stimulatory activity with that of the parent ODN in TLR9 transfected HEK293 and J774 cell lines, mouse spleen cell cultures, and in vivo in mice.

EXPERIMENTAL PROCEDURES Reagents. Bromoacetyl-β-amyloid peptide (BrAc-DAEFRHDSGYEVHHQKLVFFAEDVGSNK-OH) was purchased from AnaSpec (San Jose, CA). Disulfide functionalized ODNs 5′-CTATCTGACGTTCTCTGT-3′-(CH2)6S-S(CH2)6OH (7), HO(CH2)6S-S(CH2)6-5′-CTATCTGACGTTCTCTGT-3′ (8), 5′CTATCTGACGTTCTCTGT-3′-(CH2)3S-S(CH2)3OH (9), and HO(CH2)3S-S(CH2)3-5′-CTATCTGACGTTCTCTGT-3′ (10) and unmodified ODNs 1 and 6 were synthesized as described below. DTT was obtained from Sigma-Aldrich (St. Louis, MO). Illustra NAP-25 columns were purchased from GE Healthcare (Piscataway, NJ). Thiol-modifier C6 S-S phosphoramidite and C3 S-S controlled-pore glass solid support were obtained from ChemGenes (Wilmington, MA) and Glen Research (Sterling, VA), respectively. Regular 5′-O-DMT-3′-phosphoramidites and reverse 3′-O-DMT-5′-phosphoramidites of dA, dC, dG, and T were obtained from Applied Biosystems (Foster City, CA) and ChemGenes (Wilmington, MA), respectively. Synthesis and Purification of Disulfide-Functionalized ODNs. Both 3′- and 5′-disulfide functionalized ODNs 7-10 (Scheme 1) as well as unmodified ODNs 1 and 6 (Table 1) with phosphorothioate (PS) backbone were synthesized on a MerMade 6 DNA synthesizer (Bioautomation, Inc., Plano, TX) using β-cyanoethylphosphoramidite chemistry. Synthesis was carried out on a 10-µmol scale as DMT-off in either 3′ f 5′ or

10.1021/bc900425s  2010 American Chemical Society Published on Web 12/18/2009

40 Bioconjugate Chem., Vol. 21, No. 1, 2010

Putta et al.

Scheme 1. Schematic Representation of the Synthesis of 3′- and 5′-Linker-Disulfide Functionalized ODNs 7-10a

a Reagents and conditions used on automated DNA synthesizer: (a) Detritylation: with 3%TCA in DCM; 2 min. (b) Coupling: with 0.05 M phosphoramidite in acetonitrile and 0.25 M ethylthiotetrazole (ETT), 6 min. (c) Capping: with Cap A (acetic anhydride/pyridine/THF) and Cap B (N-methylimidazole in THF), 3 min. (d) Oxidation: with 3% Beaucage reagent in acetonitrile, 6 min. (e) Coupling: with 0.05 M thiol-modifier C6 S-S phosphoramidite in acetonitrile and 0.25 M ETT in acetonitrile, 6 min. (f) Cleavage and deprotection off the synthesizer: with 28-30% NH4OH solution at 56 °C, 16 h. CPG stands for controlled-pore glass solid support.

Table 1. Oligodeoxynucleotide Sequences and Analytical Data of TLR9 Agonist and Its Peptide Conjugates accessibility

mol. wt.b

purity (% FLP)c

ODN number

sequencea

3′-end

5′-end

calcd

obsd

HPLC

CGE

1 2 3 4 5 6

5′-CTATCTGACGTTCTCTGT-3′ 5′-CTATCTGACGTTCTCTGT-3′-CH2(CH2)4CH2S-P P-SCH2(CH2)4CH2-5′-CTATCTGACGTTCTCTGT-3′ 5′-CTATCTGACGTTCTCTGT-3′-CH2CH2CH2S-P P-SCH2CH2CH2-5′-CTATCTGACGTTCTCTGT-3′ 5′-CTATCTCACCTTCTCTGT-3′

free conjugated free conjugated free free

free free conjugated free conjugated free

5705 9219 9219 9177 9177 5624

5706 9218 9217 9178 9174 5625

98 98 92 99 94 98

97 92 95 94 90 98

a All are oligodeoxynucleotides with phosphorothioate backbone; P ) β-amyloid (1-28) peptide. b Molecular weights as determined by MALDI-ToF mass spectrophotometer; calcd and obsd indicate calculated and experimental values, respectively. c FLP ) full length product; CGE ) capillary gel electrophoresis.

5′ f 3′ direction as needed. Beaucage sulfurizing agent was used as an oxidizing agent to obtain the PS backbone modification (29). ODN 7 with C6 linker disulfide modification at 3′end was achieved using 5′-dC solid support and 5′-phosphoramidites followed by final coupling with C6 S-S linker phosphoramidite (Sheme 1A). On the other hand, ODN 8 with C6 linker disulfide modification at the 5′-end was achieved using 3′-dT solid support and 3′-phosphoramidites followed by final coupling with C6 S-S linker phosphoramidite (Scheme 1B). Conversely, ODNs 9 and 10 with C3 linker disulfide modification at 3′- and 5′-ends were achieved using 3′- and 5′phosphoramidites, respectively, on C3 S-S thiol-modifier solid support (Scheme 1C,D). After the synthesis, ODNs were cleaved from the solid support and deprotected using 28-30% ammonium hydroxide solution at 56 °C for 16 h. Solid support was removed by filtration; excess ammonia was removed under vacuum. The remaining solution was diluted with 5 vol equiv of distilled water, and crude ODNs were purified on a Waters (Milford, MA) 600 HPLC system using a Source 15Q resin (GE Healthcare Biosciences, Piscataway, NJ) column (20 cm × 250 mm) at 85 °C. A 25 mM Tris-HCl, pH 7.5 buffer containing 20% acetonitrile was used as buffer A and 2.5 M NaCl in buffer A was used as buffer B. HPLC purification was performed at a flow rate of 10 mL/min running100% buffer A

for 2 min, and then increasing buffer B from 0% to 30% in 10 min then 30% to 70% in 40 min. The major absorbance peak detected at 260 nm was collected, solvent was evaporated by rotoevaporation, and the residue was redissolved in RNAasefree water and desalted as following. Then ODNs were desalted on a C18 reverse-phase HPLC column and dialyzed against United States Pharmacopea-quality sterile water for irrigation (Braun). All ODNs synthesized were lyophilized and reconstituted in distilled water, and the concentrations were determined by measuring UV absorbance at 260 nm. The purity of all ODNs was determined by analytical anion-exchange HPLC (DNA Pack100 column) and capillary gel electrophoresis (P/ACE MDQ, Beckman Coulter, Fullerton, CA). The sequence integrity was characterized by matrixassisted laser desorption/ionization-time-of-flight (MALDI-ToF) mass spectrometry (Micro MX, Waters, USA). Preparation of Free Thiol-Functionalized ODNs 11-14 (Scheme 2). About 2.5 mg of each of the 3′- or 5′-disulfide functionalized ODNs 7-10 were dissolved separately in 2.25 mL of 100 mM sodium phosphate, pH 7.2 buffer and treated with 0.25 mL of 1 M DTT solution in 100 mM sodium phosphate, pH 7.2 buffer at room temperature. The mixture was vortexed, centrifuged, and the reaction progress monitored by MALDI/ToF mass spectral analysis. The reaction was completed in about 3.5 h, and the mixture was purified on Illustra NAP-

Bioconjugate Chem., Vol. 21, No. 1, 2010 41 Scheme 2. Schematic Showing the Synthesis of 3′- and 5′- ODN-β-Amyloid Peptide Conjugatesa

a Reagents and conditions: (a) 100 mM DTT in 100 mM sodium phosphate buffer pH 7.2, rt, 3.5 h; (b) DMSO/100-mM sodium phosphate buffer pH 7.2, rt, 24 h. Free 3′- and 5′-ends are highlighted.

25 column using water by following the supplier’s protocol. Thiol-functionalized ODNs 11-14 obtained from ODNs 7-10, respectively, were stored frozen immediately after elution and lyophilized to avoid dimerization. The purity and sequence integrity of ODNs were determined by analytical anionexchange HPLC (DNA Pack100 column), capillary gel electrophoresis, and MALDI-ToF mass spectrometry. Synthesis and Purification of ODN-Peptide Conjugates (2-5) (Scheme 2). About 2 mg of bromoacetyl-β-amyloid peptide 15 in 0.1 mL of DMSO was treated with 2.5 mg of 3′or 5′-thiol functionalized ODN (11-14) in 0.9 mL of 100 mM sodium phosphate, pH 7.2 buffer, vortexed and centrifuged, and the mixture was allowed to react at room temperature. The reaction progress was monitored by analytical anion-exchangeHPLC and MALDI/ToF mass spectral analysis. After 24 h, ODN-peptide conjugates (2-5) were purified on an anion exchange HPLC equipped with Source 15Q column at 85 °C using a gradient of solvent B (25 mM Tris-HCl, pH 7.5 buffer containing 20% acetonitrile and 2.5 M NaCl) over solvent A (25 mM Tris-HCl, pH 7.5 buffer containing 20% acetonitrile). Pure fractions were desalted on a C18 cartridge, dialyzed, and lyophilized to get conjugate as a white fluffy solid. Conjugates were reconstituted in distilled water, and concentrations were determined by measuring UV absorbance at 260 nm. The purity of ODNs 2-5 was determined by analytical anion-exchange HPLC (DNA Pack100 column), capillary gel electrophoresis, and denaturing PAGE. The sequence integrity of ODNs was characterized by MALDI-ToF mass spectrometry.

HEK293 Cell Assays. HEK293 cells stably expressing mouse TLR9 (Invivogen, San Diego, CA) were cultured and transiently transfected with secreted form of human embryonic alkaline phosphatase (SEAP) reporter plasmid (pNifty2-Seap) (Invivogen) as described previously (30). After transfection, medium was replaced with fresh culture medium, ODNs were added to the cultures, and the cultures were continued for 18 h. At the end of ODN treatment, culture supernatants were harvested and used for measuring SEAP activity by the QuantiBlue method following the manufacturer’s protocol (Invivogen). The data are shown as fold increase in NF-κB activity over PBS control. J774 Cell Cultures, Preparation of Nuclear Extracts, and EMSA. For NF-κB activation studies, murine J774 macrophage cells (ATCC) were plated at a density of 5 × 106 cells/ well in 6-well plates and cultured as described previously (31). Cells were treated with ODNs at 1 µg/mL concentration for 1 h. Nuclear extracts were prepared using the nuclear extraction kit (Sigma) according to the manufacturer’s instructions, and protein concentrations were determined by Bradford assay. Samples were either used immediately or stored frozen at -70 °C. NF-κB activation in the nuclear extracts was monitored by electrophoretic mobility shift assay (EMSA) as described previously (31). Mouse Spleen Cell Cultures. Spleen cells from C57BL/6 mice were prepared and cultured in RPMI complete medium consisting of RPMI 1640 with 10% fetal bovine serum, 100 U/mL penicillin, 100 µg/mL streptomycin, and 2 mM Lglutamine. Mouse spleen cells were plated in 96-well plates at

42 Bioconjugate Chem., Vol. 21, No. 1, 2010

Putta et al.

Figure 1. Denaturing PAGE analysis of purified ODNs 1-5 and 11-14 (lane numbers correspond to ODN numbers used in the text). Lane M is marker ODNs of defined length.

5 × 106 cells/mL and incubated with different concentrations of ODNs. Culture supernatants were collected after 24 h. The levels of cytokines and chemokines in the culture supernatants were measured using a mouse multiplex kit on an Applied Cytometry Systems Luminex 100 or 200 instrument, and the data were analyzed using StarStation software version 2.0. The required reagents were purchased from Invitrogen (Carlsbad, CA). In Vivo Studies in Mice. Four-to-six-week-old female C57BL/6 mice were obtained from Taconic Farms, Germantown, NY. The animal studies reported in the paper were carried out following Idera’s IACUC guidelines and approved protocols. Mice (n ) 3) were injected subcutaneously (s.c.) with 20 µg of ODN (1 mg/kg). Blood was collected by retro-orbital bleeding 2 h after ODN administration, and serum IL-12 levels were determined by sandwich ELISA.

RESULTS Synthesis of ODN-β-Amyloid Peptide Conjugates. We have synthesized ODN-peptide conjugates by covalently attaching NR-(bromoacetyl)peptide (amino acids 1-28 fragment of the β-amyloid protein) to 5′- or 3′-thiol modified 18-mer phosphorothioate ODNs with C6 or C3 linkers. In this approach, first we synthesized ODNs functionalized with C6- and C3-linker disulfide moieties at 3′- and 5′-ends as shown in Scheme 1 (ODNs 7-10). ODNs 7-10 were converted to activated-thiolfunctionalized ODNs 11-14 by cleaving the disulfide bridge with DTT (Scheme 2). Then, ODN-peptide conjugates 2-5 (with thioether linkages) with C6 and C3 linkers were obtained by adding the bromoacetyl peptide 15 to ODNs 11-14 with an activated thiol function at either the 3′- or 5′-end (Scheme 2). ODN 1 served as a parent immune stimulatory agent without peptide conjugation. ODN 6 was a control that had the same length and structure as that of ODN 1 but lacked an immune stimulatory motif. All ODNs were characterized by MALDI/ToF mass spectral analysis for their sequence integrity, and the purities were checked by HPLC and capillary gel electrophoresis (Table 1). All ODNs were also analyzed by denaturing slab gel polyacrylamide gel electrophoresis (PAGE). Figure 1 shows PAGE profiles of ODNs 1-5 and 11-14. ODNs 11-14 (with a thiol functional group at the 3′- or 5′-end) had mobility similar to that of ODN 1, which is an 18-mer. ODN-β-amyloid peptide conjugates 2-5 moved slower as that of a 46-mer ODN marker, suggesting the formation of peptide conjugates (Figure 1). Activity of ODNs in TLR9-Expressing HEK293 Cells. TLR9 agonists activate immune signaling pathways, including the NF-κB pathway (32). To determine the effects of 3′- and 5′-conjugation of peptide to ODN, as well as the linker length between ODN and peptide, HEK293 cells transfected with mouse TLR9 and SEAP reporter gene were stimulated with ODNs 1-6 and examined for NF-κB activation. The results of dose-dependent NF-κB activation over PBS control are shown in Figure 2. The control ODN 6 (lacking a stimulatory motif) exhibited no activity. ODNs 1-5 showed activation of NF-κB to different levels for 3′- and 5′-conjugates (Figure 2). ODNs 2

Figure 2. Dose-dependent activation of NF-κB in HEK293 cells expressing mouse TLR9 by ODNs 1-6. Data shown are representative of three independent experiments.

Figure 3. Activation of NF-κB in J774 murine macrophages upon treatment with ODNs. Autoradiogram showing NF-κB activation at 1 µg/mL concentration of indicated ODNs. Lane numbers correspond to ODN numbers used in the text. Experiments were carried out as described in the Experimental Section. Data shown are of one representative experiment of three independent experiments.

and 4 (with peptide conjugated at the 3′-end) induced levels of NF-κB activation similar to that of parent ODN 1, suggesting that conjugation of peptide at the 3′-end has minimal effect on TLR9-mediated immune stimulatory activity of ODN (Figure 2). Additionally, spacer length between ODN and peptide (2 with C6-linker vs 4 with C3 linker) has insignificant effect on immune stimulation. In contrast, ODNs 3 and 5 (with peptide conjugated at the 5′-end) induced lower levels of NF-κB than ODNs 1, 2, and 4 (Figure 2). These results support our previous observations that an accessible 5′-end is required for TLR9 recognition and stimulation and that blocking of the 5′-end of an ODN by conjugation of a bulkier ligand abrogates immune stimulation (27, 28). Activity of ODNs in J774 Cells. To further study the effects of 3′- and 5′-conjugation of peptide to ODN and the role of linker length on NF-κB activation, J774 mouse macrophage cells, which express TLR9 (32), were stimulated with 1 µg/mL ODNs 1-6 and the nuclear extracts were examined for NF-κB activation by EMSA. The results are presented in Figure 3. The presence of bands corresponding to NF-κB with ODNs 1, 2, and 4 suggest that the transcription factor NF-κB was activated (Figure 3). In contrast, the band intensity with ODNs 3 and 5 was minimal, suggesting that conjugation of peptide at the 5′end of ODN results in the loss of TLR9-mediated immune activation (Figure 3). Activity of ODNs in Mouse Spleen Cell Cultures. To study the impact of peptide conjugation with different linker lengths at the 3′- or 5′-end of ODN in primary cells, we evaluated ODNinduced production of cytokines in C57BL/6 mouse spleen cell cultures. The induction of IL-12, IL-6, MIP-1R, and TNF-R at 10 µg/mL concentration of ODN in spleen cell cultures is shown in Figure 4. The response was dependent on the dose of ODN (Supporting Information Figure 1). Control ODN 6 (lacking a stimulatory motif) induced no or insignificant levels of IL-12, IL-6, MIP-1R, and TNF-R in mouse spleen cell cultures. ODNs

Bioconjugate Chem., Vol. 21, No. 1, 2010 43

Figure 4. IL-12, IL-6, MIP-1R, and TNF-R induction profiles of ODNs 1-6 (10 µg/mL concentration) in C57BL/6 mouse spleen cell cultures. Spleen cells were cultured in the presence of ODNs for 24 h, and the levels of secreted cytokines in culture supernatants were measured by multiplex luminex assay as described in the Experimental Section. Each value is an average of three replicate wells, and the results are of one representative experiment of three independent experiments.

levels of serum IL-12 than did ODNs 1, 2, and 4 (Figure 5). These in vivo data are in agreement with the results observed in cell-based assays.

DISCUSSION

Figure 5. In vivo IL-12 induction by ODNs 1-6 administered s.c. to C57BL/6 mice at 1 mg/kg dose. Blood was collected 2 h post-ODN administration, and serum IL-12 levels were determined by ELISA as described in the Experimental Section. Data shown are representative of three independent experiments. * p < 0.001.

2 and 4 (peptide conjugated with C6 and C3 linkers at the 3′end) induced similar levels of IL-12, IL-6, MIP-1R, and TNF-R to that of parent ODN 1 (Figure 4). In contrast, ODNs 3 and 5 (peptide conjugated with C6 and C3 linkers at the 5′-end) induced lower levels of IL-12, IL-6, MIP-1R, and TNF-R than did ODNs 1, 2, and 4 (Figure 4). These results are consistent with HEK293 and J774 cell culture results and further confirm that blocking the 5′-end of ODN results in decreased TLR9mediated immune stimulatory activity. Activity of ODNs In Vivo in Mice. We then examined the in vivo IL-12 induction profiles of ODNs 1-6 in mice to further understand the effects of site of peptide conjugation. ODNs were administered subcutaneously to C57BL/6 mice at a dose of 1 mg/kg, and serum IL-12 levels were determined by sandwich ELISA 2 h post-ODN administration. As shown in Figure 5, control ODN 6 induced no IL-12, whereas ODN 1 induced higher levels of IL-12 compared with ODN 6 suggesting that the stimulatory motif is required for immune responses. ODNs 2 and 4 (peptide conjugated with C6 and C3 linkers at the 3′end) induced slightly higher levels of IL-12 than did ODN 1 (Figure 5). In contrast, ODNs 3 and 5 (peptide conjugated with C6 and C3 linkers at the 5′-end) induced significantly lower

While coadministration of ODN and vaccine candidates has yielded promising results (11, 12), attempts are being made to deliver both antigen (protein/peptide) and ODN to the same APC to enhance cross-presentation and cross-priming (17). Liposomal formulations and conjugation of protein/peptide antigen to ODN are commonly employed to deliver both of the components to the same APC (17). Our previous studies showed that an accessible 5′-end of an ODN is required for TLR9-mediated immune responses and blocking the 5′-end results in loss of immune stimulatory activity (27, 28). In the present study, we have conjugated a 28-mer β-amyloid peptide that is not immunogenic to distinguish ODN-mediated TLR9 activation from peptide/ protein-induced immune responses and study the role of 5′accessibility of ODN for TLR9-mediated immune responses. In vitro studies carried out in TLR9 transfected cell line, mouse macrophage J774 cell line, and mouse primary splenocytes showed that conjugation of peptide at the 5′end of ODN significantly reduces TLR9-mediated immune responses. Conjugation of peptide at the 3′-end of ODN resulted in similar activity to that of parent ODN. Moreover, we have also examined whether the proximity of peptide to the 3′- or 5′-end of the ODN would have an effect on TLR9-mediated immune stimulation of 3′- and 5′conjugates by synthesizing and studying peptide-ODN conjugates using C3- and C6-linkers. ODN conjugated with peptide at the 3′-end through C3- and C6-linkers showed similar TLR9-mediated immune responses to that of the parent ODN, suggesting that the effects observed in these studies are not dependent on the proximity of peptide to the 3′- or 5′-end of ODN, but because an accessible 5′-end of ODN is required for TLR9-mediated immune stimulation. These results were further confirmed in acute administration studies in vivo in mice, in which the parent compound without peptide conjugation and the two conjugates that have peptide attached at the 3′-end of ODN either via C6- or C3-

44 Bioconjugate Chem., Vol. 21, No. 1, 2010

linker (2 and 4) are significantly more immune stimulatory than those with peptide conjugated at the 5′-end (3 and 5). Our previous studies have demonstrated that differences in immune stimulatory activity of 3′- and 5′-conjugates is due to the requirement of 5′-end accessibility for TLR9-mediated immune responses, but not as a result of their differences in cellular uptake, as both types of conjugates are taken up by cells to the same extent (28). Further, these studies have shown that conjugation of a mono, di, tetra, or oligonucleotide through 5′-5′ linkage or a fluorescein moiety, but not a phosphorothioate group, at the 5′-end significantly reduces TLR9-mediated immune responses (28). These results together with the present studies with a peptide conjugated at the 5′- and 3′-end of ODN suggest that an accessible 5′-end of ODN is required for TLR9mediated immune responses and conjugation of bulkier moieties like peptides/proteins at the 5′-end of ODN results in loss of TLR9-mediated immune responses. Several peptide/protein-ODN conjugates are currently at various stages of preclinical and clinical evaluation for the treatment of various disease indications (13–17, 33–37). These studies have utilized conjugation of peptides at the 5′-end of ODN due to ease of incorporation of chemical modifications required for conjugation. The present studies described above suggest that conjugation of peptides or proteins at the 5′-end of ODN does not benefit the TLR9 agonist-mediated immune responses to the full extent. A ragweed allergen, Amb a 1, conjugated at the 5′-end of an ODN has been the most extensively studied and evaluated in human clinical trials (15, 33, 34). However, the clinical trial results have failed to achieve the end points. We have also conjugated other peptides of 23 and 28 amino acids to ODN and observed a loss of TLR9-mediated activity with 5′-, but not 3′-, conjugates (data not shown). Taken together, the present results clearly demonstrate that conjugation of large molecules such as peptides/proteins at the 5′-end of an ODN could lead to significant loss of TLR9-mediated immune stimulatory activity, thereby compromising the advantage provided by TLR9-mediated adjuvant activity. Therefore, conjugation of peptides/proteins at the 3′-end of ODN would be beneficial. Supporting Information Available: A figure showing concentration-dependent cytokine induction in mouse spleen cell cultures is available. This material is available free of charge via the Internet at http://pubs.acs.org.

LITERATURE CITED (1) Hemmi, H., Takeuchi, O., Kawai, T., Kaisho, T., Sato, S., Sanjo, H., Matsumoto, M., Hoshino, K., Wagner, H., Takeda, K., and Akira, S. (2000) A Toll-like receptor recognizes bacterial DNA. Nature 408, 740–745. (2) Yamamoto, S., Yamamoto, T., Nojima, Y., Umemori, K., Phalen, S., McMurray, D. N., Kuramoto, E., Iho, S., Takauji, R., Sato, Y., Yamada, T., Ohara, N., Matsumoto, S., Goto, Y., Matsuo, K., and Tokunaga, T. (2002) Discovery of immunostimulatory CpG-DNA and its applications to tuberculosis vaccine development. Jpn. J. Infect. Dis. 55, 37–44. (3) Agrawal, S., and Kandimalla, E. R. (2005) Role of Toll-like receptors in antisense and siRNA. Nat. Biotechnol. 22, 1533– 1537. (4) O’Neill, L. A., Bryant, C. E., and Doyle, S. L. (2009) Therapeutic targeting of Toll-like receptors for infectious and inflammatory diseases and cancer. Pharmacol. ReV. 61, 177– 197. (5) Klinman, D. M., Currie, D., Gursel, I., and Verthelyi, D. (2004) Use of CpG oligodeoxynucleotides as immune adjuvants. Immunol. ReV. 199, 201–216.

Putta et al. (6) McCluskie, M. J., and Davis, H. L. (1998) CpG DNA is a potent enhancer of systemic and mucosal immune responses against hepatitis B surface antigen with intranasal administration to mice. J. Immunol. 161, 4463–4466. (7) Oxenius, A., Martinic, M. M. A., Hengartner, H., and Klenerman, P. (1999) CpG-containing oligonucleotides are efficient adjuvants for induction of protective antiviral immune responses with T-cell peptide vaccines. J. Virol. 73, 4120–4126. (8) Schneeberger, A., Wagner, C., Zemann, A., Luhrs, P., Kutil, R., Goos, M., Stingl, G., and Wagner, S. N. (2004) CpG motifs are efficient adjuvants for DNA cancer vaccines. J. InVestig. Dermatol. 123, 371–379. (9) Zhu, F. G., Kandimalla, E. R., Yu, D., and Agrawal, S. (2007) Oral administration of a synthetic agonist of Toll-like receptor 9 potently modulates peanut-induced allergy in mice. J. Allergy Clin. Immunol. 120, 631–637. (10) Aurisicchio, L., Peruzzi, D., Conforti, A., Dharmapuri, S., Biondo, A., Giampaoli, S., Fridman, A., Bagchi, A., Winkelmann, C. T., Gibson, R., Kandimalla, E. R., Agrawal, S., Ciliberto, G., and La Monica, N. (2009) Treatment of mammary carcinomas in HER-2 transgenic mice through combination of genetic vaccine and an agonist of Toll-like receptor 9. Clin. Cancer Res. 15, 1575–1584. (11) Cooper, C. L., Davis, H. L., Morris, M. L., Efler, S. M., Krieg, A. M., Li, Y., Laframboise, C., Al Adhami, M. J., Khaliq, Y., Seguin, I., and Cameron, D. W. (2004) Safety and immunogenicity of CPG 7909 injection as an adjuvant to Fluarix infleuenza vaccine. Vaccine 22, 3136–3143. (12) Halperin, S. A., Dobson, S., McNeil, S., Langley, J. M., Smith, B., McCall-Sani, R., Levitt, D., Van Nest, G., Gennevois, D., and Eiden, J. J. (2006) Comparison of the safety and immunogenicity of hepatitis B virus surgace antigen co-administered with an immunostimulatory phosphorothioate oligonculetide and a licensed hepatitis B vaccine in healthy young adults. Vaccine 24, 20–26. (13) Krishnamachari, Y., and Salem, A. K. (2009) Innovative strategies for co-delivering antigens and CpG oligonucleotides. AdV. Drug DeliVery ReV. 61, 205–217. (14) Tighe, H., Takabayashi, K., Schwartz, D., Marsden, R., Beck, L., Corbeil, J., Richman, D. D., Eiden, J. J., Spiegelberg, H. L., and Raz, E. (2000) Conjugation of protein to immunostimulatory DNA results in a rapid, long-lasting and potent induction of cellmediated and humoral immunity. Eur. J. Immunol. 30, 1939– 1947. (15) Tighe, H., Takabayashi, K., Schwartz, D., Van Nest, G., Tuck, S., Eiden, J. J., Kagey-Sobotka, A., Creticos, P. S., Lichtenstein, L. M., Spiegelberg, H. L., and Raz, E. (2000) Conjugation of immunostimulatory DNA to the short ragweed allergen Amb a 1 enhances its immunogenicity and reduces its allergenicity. J. Allergy Clin. Immunol. 106, 124–134. (16) Heit, A., Maurer, T., Hochrein, H., Bauer, S., Huster, K. M., Busch, D. H., and Wagner, H. (2003) Toll-like receptor 9 expression is not required for CpG DNA-aided cross-presentation of DNA-conjugated antigens but essential for cross-priming of CD8 T cells. J. Immunol. 170, 2802–2805. (17) Daftarian, P., Sharan, R., Haqa, W., Ali, S., Longmate, J., Termini, J., and Diamonda, D. J. (2005) Novel conjugates of epitope fusion peptides with CpG-ODN display enhanced immunogenicity and HIV recognition. Vaccine 23, 3453–3468. (18) Agrawal, S., Christodoulou, C., and Gait, M. J. (1986) Efficient methods for attaching non-radioactive labels to the 5′ ends of synthetic oligodeoxyribonucleotides. Nucleic Acids Res. 14, 6227–6245. (19) Zhao, Q., Yu, D., and Agrawal, S. (1999) Site of chemical modifications in CpG containing phosphorothioate oligodeoxynucleotide modulates its immunostimulatory activity. Bioorg. Med. Chem. Lett. 9, 3453–3458. (20) Zhao, Q., Yu, D., and Agrawal, S. (2000) Immunostimulatory activity of CpG containing phosphorothioate oligodeoxynucleotide is modulated by modification of a single deoxynucleoside. Bioorg. Med. Chem. Lett. 10, 1051–1054.

Bioconjugate Chem., Vol. 21, No. 1, 2010 45 (21) Yu, D., Kandimalla, E. R., Zhao, Q., Cong, Y., and Agrawal, S. (2002) Immunostimulatory properties of phosphorothioate CpG DNA containing both 3′-5′- and 2′-5′-internucleotide linkages. Nucleic Acids Res. 30, 1613–1619. (22) Agrawal, S., and Kandimalla, E. R. (2001) Antisense and/or immunostimulatory oligonucleotide therapeutics. Curr. Cancer Drug Targets 1, 197–209. (23) Yu, D., Kandimalla, E. R., Zhao, Q., Cong, Y., and Agrawal, S. (2001) Immunostimulatory activity of CpG oligonucleotides containing nonionic methylphosphonate linkages. Bioorg. Med. Chem. 9, 2803–2808. (24) Yu, D., Kandimalla, E. R., Zhao, Q., Cong, Y., and Agrawal, S. (2001) Modulation of immunostimulatory activity of CpG oligonucleotides by site-specific deletion of nucleobases. Bioorg. Med. Chem. Lett. 11, 2263–2267. (25) Yu, D., Kandimalla, E. R., Zhao, Q., Bhagat, L., Cong, Y., and Agrawal, S. (2003) Requirement of nucleotide bases proximal to a CpG-motif for immunostimulatory activity of synthetic oligonucleotides. Bioorg. Med. Chem. 11, 459–464. (26) Yu, D., Kandimalla, E. R., Cong, Y., Tang, J., Tang, J. Y., Zhao, Q., and Agrawal, S. (2002) Design, synthesis, and immunostimulatory properties of CpG DNAs containing alkyllinker substitutions: role of nucleosides in the flanking sequences. J. Med. Chem. 45, 4540–4548. (27) Yu, D., Zhao, Q., Kandimalla, E. R., and Agrawal, S. (2000) Accessible 5′-end of CpG-containing phosphorothioate oligodeoxynucleotides is essential for immunostimulatory activity. Bioorg. Med. Chem. Lett. 10, 2585–2588. (28) Kandimalla, E. R., Bhagat, L., Yu, D., Cong, Y., Tang, J., and Agrawal, S. (2002) Conjugation of ligands at the 5′-end of CpG DNA affects immunostimulatory activity. Bioconjugate Chem. 13, 966–974. (29) Iyer, R. P., Egan, W., Regan, J. B., and Beaucage, S. L. (1990) 3H-1,2-Benzodithiole-3-one 1,1-dioxide as an improved sulfurizing reagent in the solid-phase synthesis of oligodeoxyribonucleoside phosphorothioates. J. Am. Chem. Soc. 112, 1253– 1254.

(30) Putta, M. R., Zhu, F. G., Li, Y., Bhagat, L., Cong, Y., Kandimalla, E. R., and Agrawal, S. (2006) Novel oligodexoynucleotide agonists of TLR9 containing N3-Me-dC or N1-MedG modifications. Nucleic Acids Res. 34, 3231–3238. (31) Kandimalla, E. R., Bhagat, L., Wang, D., Yu, D., Zhu, F. G., Tang, J., Wang, H., Huang, P., Zhang, R., and Agrawal, S. (2003) Divergent synthetic nucleotide motif recognition pattern: design and development of potent immunomodulatory oligodeoxyribonucleotide agents with distinct cytokine induction profiles. Nucleic Acids Res. 31, 2393–2400. (32) Stacey, K. J., Sweet, M. J., and Hume, D. A. (1996) Macrophages ingest and are activated by bacterial DNA. J. Immunol. 157, 2116–2122. (33) Creticos, P. S., Schroeder, J. T., Hamilton, R. G., BalcerWhaley, S. L., Khattignavong, A. P., Lindblad, R., Li, H., Coffman, R., Seyfert, V., Eiden, J. J., and Broide, D. (2006) Immunotherapy with a ragweed-toll-like receptor 9 agonist vaccine for allergic rhinitis. N. Engl. J. Med. 355, 14451455. (34) Broide, D. H. (2005) Immunostimulatory sequences of DNA and conjugates in the treatment of allergic rhinitis. Curr. Allergy Asthma Rep. 5, 182–185. (35) Heit, A., Huster, K. M., Schmitz, F., Schiemann, M., Busch, D. H., and Wagner, H. (2004) CpG-DNA aided cross-priming by cross-presenting B cells. J. Immunol. 172, 1501–1507. (36) Heit, A., Schmitz, F., O’Keeffe, M., Staib, C., Busch, D. H., Wagner, H., and Huster, K. M. (2005) Protective CD8 T cell immunity triggered by CpG-protein conjugates competes with the efficacy of live vaccines. J. Immunol. 174, 4373–4380. (37) Suzuki, M., Ohta, N., Min, W. P., Matsumoto, T., Min, R., Zhang, X., Toida, K., and Murakami, S. (2007) Immunotherapy with CpG DNA conjugated with T-cell epitope peptide of an allergenic Cry j 2 protein is useful for control of allergic conditions in mice. Int. Immunopharmacol. 7, 46–54. BC900425S