Translation of therapeutic vaccines for addiction, cancer, or other chronic noncommunicable diseases has been slow because only a small subset of immunized subjects achieved effective Ab levels. We hypothesize that individual variability in the number of naive and early-activated hapten-specific B cells determines postvaccination serum Ab levels and vaccine efficacy. Using a model vaccine against the highly abused prescription opioid oxycodone, the polyclonal B cell population specific for an oxycodone-based hapten (6OXY) was analyzed by flow cytometry paired with Ag-based magnetic enrichment. A higher frequency of 6OXY-specific B cells in either spleen biopsies or blood, before and after immunization, correlated to subsequent greater oxycodone-specific serum Ab titers and their efficacy in blocking oxycodone distribution to the brain and oxycodone-induced behavior in mice. The magnitude of 6OXY-specific B cell activation and vaccine efficacy was tightly correlated to the size of the CD4+ T cell population. The frequency of enriched 6OXY-specific B cells was consistent across various mouse tissues. These data provide novel evidence that variations in the frequency of naive or early-activated vaccine-specific B and T cells can account for individual responses to vaccines and may predict the clinical efficacy of a therapeutic vaccine.

Vaccines have been the most effective medical intervention for infective diseases (1) and have shown promising proof of concept for the treatment of chronic diseases including drug addiction (2, 3), cancer (4), and Alzheimer’s (5). Yet, therapeutic vaccine efficacy is only met in the subset of immunized animal or human subjects with the highest serum Ab levels against drugs of abuse (6, 7), tumor-associated small carbohydrate and peptide Ags (8, 9), or β amyloid–derived peptides (5). The mechanism(s) underlying individual variability in vaccine efficacy is poorly understood, and it is unclear whether the generation of an effective response, or the lack thereof, is due to vaccine design, disease heterogeneity, host genetics, or interactions between the vaccine and the host immune system.

The development of vaccines against noncommunicable diseases has the potential to significantly affect public health (1). For instance, drug addiction is a worldwide concern (10). In the United States, tobacco use is the leading preventable cause of death and is responsible for ∼480,000 deaths annually (11). Similarly, ∼2.1 million people in the United States are addicted to prescription opioid analgesics, and ∼17,000 overdose deaths are attributable to opioid analgesics annually (11). Intravenous drug use is also a known vehicle for blood-borne pathogens, including HIV (11). Despite the high prevalence of opiate abuse, only a few medications exist (12, 13). Also, no treatments are approved for cocaine and methamphetamine dependence, supporting the need for new therapies (14). Vaccines, combined with current medications, may offer a treatment option for drug addiction. Vaccines against drugs of abuse consist of drug-derived haptens bound to foreign immunogenic carriers that stimulate T cell–dependent B cell activation to generate Abs that bind drugs of abuse in serum, decreasing the distribution of free drug (i.e., unbound) to the brain and addiction-related behaviors.

In germinal centers (GC), Ag-specific GC B cells interact with CD4+ T follicular helper cells (TfH) (15, 16) and mature into either long-lived memory or Ab-secreting B cells (17), which are critical to generate effective Abs. We hypothesize that the frequency of hapten-specific B cells and carrier-specific CD4+ T cells prior to immunization contributes to vaccine efficacy, as the number of naive peptide-specific CD4+ T cells can dictate the magnitude of a response to immunization (18). Moreover, CD4+ T cells specific for tetanus toxoid or keyhole limpet hemocyanin (KLH), proteins commonly used as carriers in conjugate vaccines, are present in different individual frequencies in human blood (19). Similarly, B cells specific for chicken OVA, PE, or other model proteins have different population sizes in the naive mouse repertoire (2022). These data suggest that the frequency of T and B cells specific for peptides or proteins may explain some of the variability observed in vaccinated individuals. In contrast, it is not known whether the initial size of the polyclonal hapten–specific B cell subsets shapes the postvaccination response. In this study, we examined the extent to which the number of polyclonal naive and early-activated hapten-specific B cells correlated to the efficacy of therapeutic vaccines for drug addiction and whether the frequency of hapten-specific B cells in blood provided predictive markers of vaccine clinical efficacy.

To address these central questions in vaccinology, we adopted a state-of-the-art Ag-based enrichment strategy paired with flow cytometry analysis of Ag-specific B cells in the entire B cell repertoire in naive and immunized mice (20) and applied it to study scarce B cells specific for haptens used in addiction vaccines (23). In this study, we tested the extent to which the hapten-specific B cell subsets relate to the efficacy of a conjugate immunogen consisting of an oxycodone-based hapten (6OXY) conjugated to KLH (6OXY-KLH), which showed promising preclinical efficacy in blocking oxycodone and hydrocodone distribution to the brain and their behavioral effects (24, 25), including blockage of oxycodone self-administration (26). Our results suggest that the frequency of the naive and early-activated 6OXY-specific B cells, before and after immunization, correlates with subsequent vaccine efficacy against oxycodone. These findings were supported by the reduction in 6OXY-specific B cells upon selective depletion of CD4+ T cells resulting in loss of vaccine efficacy and by the observation that increased numbers of carrier-specific CD4+ T cells prior to immunization correlate to greater vaccine efficacy. We discuss the implications of these data with respect to the possibility of identifying patients most likely to benefit from vaccination and how our findings could aid in the development of therapeutic vaccines against drug addiction or other noncommunicable chronic diseases.

A first study analyzed hapten-specific B cells by spleen biopsy before and after immunization in three mouse cohorts (n = 24 BALB/c mice). A second study analyzed B cells in blood before and after immunization in an independent cohort of mice (n = 24 BALB/c mice). We established a priori that within-group linear regression analysis required at least 12 mice for each group, based on previous studies (23, 25). A control study tested whether CD4+ T cells depletion prevented vaccine efficacy (n = 6 C57BL/6 mice), and then, whether the number of carrier-specific CD4+ Th cells in spleen biopsy prior to immunization correlated with vaccine efficacy (n = 9 C57BL/6 mice). We established a priori that t test and one-way ANOVA between groups required at least six mice per group. Mouse studies were performed in a treatment-blind fashion.

Mouse studies conformed to the U.S. Department of Health and Human Services’ Guide for the Care and Use of Laboratory Animals, the Minneapolis Medical Research Foundation, and University of Minnesota Animal Care and Use Committees.

Oxycodone was obtained through the National Institute on Drug Abuse Drug Supply Program and Sigma-Aldrich (St. Louis, MO). Drug doses and concentrations are expressed as the weight of the base.

An oxycodone-based hapten with a tetraglycine linker at the C6 position (6OXY) was synthesized from oxycodone. For ELISA, 6OXY was conjugated by carbodiimide chemistry to OVA (Sigma-Aldrich). For immunization studies, the 6OXY was conjugated by carbodiimide chemistry to KLH (Thermo Fisher, Rockford, IL) and to OVA previously conjugated to the 2W1S peptide (EAWGALANWAVDSA; Genscript) and designated as 6OXY-OVA2W.

For B cell analysis, the 6OXY hapten was conjugated to PE (Prozyme, Hayward, CA) and designated as (6OXY)n-PE (23). PE was conjugated to the fluorescent dye Alexa Fluor 647 (AF647; Molecular Probes, Life Technologies, Grand Island, NY), and their conjugation was verified by UV spectrometry (23). To increase the B cell analysis sensitivity, the 6OXY, or the tetraglycine linker as control, was conjugated to the N-term of a EAWGALANAWKV peptide containing a biotin-(PEG)4 moiety (b) attached to a C-term lysine (K) amine side chain (below synthesis details). The resulting 6OXY(Gly)4-EAWGALANAW(K)bV and the control peptide NH2-(Gly)4EAWGALANAW(K)bV conjugates were attached, respectively, to a streptavidin (SA)-PE conjugate (SA-PE; Prozyme) or a SA-PE labeled with AF647 to provide [6OXY(Gly)4EAWGALANAW(K)bV]4-SA-PE (designated as (6OXY)4-SA-PE) and the decoy reagent [NH2-(Gly)4EAWGALANAW(K)bV]4-SA-PE-AF647 (designated as (control peptide)4-SA-PE-AF647). The (6OXY)4-SA-PE and the (control peptide)4-SA-PE-AF647 were purified by standard dialysis, and concentrations were measured by UV spectrometry. The 2W1S:I-Ab monomers were prepared, biotinylated, and attached to SA conjugated to either PE or allophycocyanin (27).

The 6OXY hapten or the linker alone were conjugated to the N-term of the EAWGALANAWKV peptide containing a biotin-(PEG)4 moiety (b) attached to the C-term lysine (K) side chain. Peptides were prepared by solid-phase synthesis using standard Fmoc protection as described previously (28). For N-term 6OXY hapten conjugation, the N-term Fmoc group was removed and the resulting free amine reacted with the hapten carboxylate using HBTU as the coupling agent followed by global acidic deprotection and resin cleavage (28). For C-terminal side chain biotinylation, a Fmoc-Lys(Dde) residue was incorporated during peptide synthesis. The Dde protecting group was removed using hydrazine, the resulting free amine coupled to a biotin-(PEG)4-N-succinimidyl ester followed by global deprotection, and conjugates purified by reversed-phase HPLC and characterized via nuclear magnetic resonance and liquid chromatography-mass spectroscopy (MS).

Peptide synthesis was carried out using an automated solid-phase peptide synthesizer (PS3; Protein Technologies, Memphis, TN) using standard Fmoc/HCTU-based chemistry using amino acid building blocks with typical acid-labile side-chain protecting groups. Synthesis began on preloaded Fmoc-Gly-Wang resin (0.10 mmol), and the peptide chain was elongated using HCTU/N-methylmorpholine–catalyzed, single coupling steps with 4 eq of both protected amino acids and HCTU for 30 min. To facilitate biotinylation at the end of the synthesis, the penultimate lysine residue was installed using Fmoc-Lys(Dde)-OH to provide an orthogonally protected side-chain amine. Following complete chain elongation, the peptide’s N terminus was deprotected with 5% piperidine in dimethylformamide (v/v), and the presence of the resulting free amine was confirmed by ninhydrin analysis. The resin containing the peptide was washed with CH2Cl2, dried in vacuo overnight, weighed, and divided into two portions for further synthesis on a reduced scale. Using 50 μmol peptide, the free N terminus was acylated with the oxycodone analog (20 mg, 50 μmol, 1 eq) in the presence of DIEA (8.6 μl, 5 μmol, 0.1 eq) and HCTU (21 mg, 50 μmol, 1 eq) in DMF (5 ml) for 10 h. After acylation was judged complete by ninhydrin analysis, the resin-bound peptide was washed thoroughly with CH2Cl2 and dried in vacuo for 4 h. The peptide was then reacted with 5% hydrazine in DMF (5 ml, v/v) to orthogonally remove the Dde protected side chain. After verifying that deprotection was complete by ninhydrin analysis, the peptide was washed with CH2Cl2, dried, and then reacted with biotin-(PEG)4-N-succinimidyl ester (44 mg, 75 μmol, 1.5 eq) in the presence of DIEA (8.6 μl, 5.0 μmol, 0.1 eq) in DMF (5 ml) 1 eq. After biotinylation was judged complete by ninhydrin testing, the peptide was cleaved from the resin along with simultaneous side-chain deprotection by treatment with Reagent K containing TFA (10 ml), crystalline phenol (0.5 g), 1,2-ethanedithiol (0.25 ml), thioanisole (0.5 ml), and H2O (0.5 ml) for 2 h at room temperature. The released peptide was collected and combined with TFA washes of the resin before precipitation of the peptide in chilled Et2O (100 ml). The crude solid peptide was collected by centrifugation, the supernatant was removed, and the resulting pellet was washed twice with cold Et2O (50 ml) repeating the centrifugation and supernatant removal steps each time. The crude peptide was purified using a semipreparative C18 reversed-phase HPLC column with detection at 214 nm and eluted with a gradient of solvent A (H2O/0.1% TFA, v/v) and solvent B (CH3CN/0.1% TFA, v/v). The crude peptide (115 mg) was dissolved in a DMF/H2O solution (1:5 v/v, 25 ml), applied to the column equilibrated in solvent A, and eluted using a linear gradient of (0–70% solvent B over 1.5 h at a flow-rate of 5 ml/min). Fractions were analyzed for purity with an analytical C18 reversed-phase HPLC column using a linear gradient (0–100% solvent B over 60 min at a flow rate of 1 ml/min) and detected at 214 nm. Fractions containing peptide product of at least 90% purity were pooled and concentrated by lyophilization to yield 23 mg (20% yield) of a white solid. A small amount (<1 mg) of the resulting purified peptide was dissolved in 10 μl 0.1% TFA/CH3CN and diluted 1:50 in a mixture of CH3CN/H2O (1:1 v/v) prior to MS analysis. MS was performed using a 50-μl injection and collecting 3000 scans. ESI-MS: calculated for C108H155N25O32S [M+2H]+2 1173.0497, found 1173.0399.

Sterile surgeries were performed under isofluorane anesthesia. Through a paramedian incision, the caudal portion of the spleen (∼50%) was resected, and wounds were sutured. Veterinary care consisted of 5 mg/kg s.c. Meloxicam (Boehringer Ingelheim, Fremont, CA) and 33 mg/kg s.c. Rocephin (Sandoz, Princeton, NJ).

Male BALB/c and C57BL/6 mice (Harlan Laboratories, Madison, WI) were housed with a 12-h light/12-h dark cycle, and fed ad libitum. In B cell studies, mice were immunized s.c. on days 0, 14, and 28 with 25 μg of either 6OXY-KLH or KLH absorbed to undiluted alum (Alhydrogel85; Brenntag Biosector, Frederikssund, Denmark) in a final volume of 0.2 ml (25). In the CD4 depletion study, mice were administered anti-CD4 mAb (0.1 mg in PBS, i.p.) for three consecutive days prior to the first immunization and then twice weekly for the duration of the study (29). The rat IgG2b mAb against mouse CD4 was produced from hybridoma GK1.5 cells (American Type Culture Collection, Manassas, VA). In the carrier-specific CD4+ T cell study, mice were immunized s.c. with 25 μg 6OXY-OVA2W absorbed to undiluted alum on days 0, 14, and 28.

The 6OXY-specific serum IgG Ab titers were measured by ELISA (25). The 6OXY-specific serum IgG Abs selectively bind oxycodone, hydrocodone, and oxymorphone (24, 25, 30).

A week after the last immunization, mice were injected s.c. with 2.25 mg/kg oxycodone, and antinociception was tested on a hot plate set at 54°C to determine the following behavioral endpoints: hind paw lifts and flicks or reaching a maximal cutoff of 60 s to avoid tissue damage (25). The percent maximal possible effect (MPE%) was calculated as (postdrug latency − baseline latency)/(maximal cutoff − baseline latency) × 100 (25). After the hot plate test, mice were euthanized by CO2, and serum and brain oxycodone concentrations were measured by gas chromatography coupled to MS (25). Concentrations represent the total of protein-, Ab-bound, and free drug in each sample.

Partial spleen samples were collected from each mouse and disaggregated as described previously (23). Whole blood was harvested by facial vein puncture, collected in tubes containing heparin 50 USP units (Hospira, Lake Forest, IL), and processed using the ammonium-chloride-potassium lysing buffer (Lonza, NJ). Single-cell suspensions from each mouse were centrifuged at 1600 rpm for 5 min at 4°C and resuspended to a final volume of 200 μl sorter buffer (DPBS + 2% FBS, 0.1% sodium azide) including Fc block (2.4G2, 2% rat serum, 0.1% sodium azide; BioXCell, West Lebanon, NH). Either PE-AF647 or (control peptide)4-SA-PE-AF647 was added at a final concentration of 5 nM and incubated at room temperature for 5 min. The (6OXY)n-PE or the (6OXY)4-SA-PE conjugate was then added at a concentration of 5 nM and incubated for 25 min at 4°C. Samples were washed in ice-cold sorter buffer, resuspended to a final volume of 200 μl sorter buffer, mixed with 25 μl anti-PE magnetic beads (Miltenyi Biotec, Auburn, CA), and incubated for 15 min at +4°C. The cells were resuspended in sorter buffer and passed through a magnetized LS column (Miltenyi Biotec). For each sample, bound fractions were collected, whereas flow-through fractions were used for calibration and compensation. Fractions were centrifuged at 1600 rpm for 5 min at 4°C and resuspended in 100 μl sorter buffer as described previously (20, 23). From each individual mouse partial splenectomy or blood sample, fluorescent counting beads (Accucheck; Invitrogen, Frederick, MD) were used to calculate total numbers of live lymphocytes, B cells, and hapten-specific B cells in the bound fraction (23).

Cell suspensions were incubated with fluorochrome-labeled anti-mouse Abs for the following B cell surface markers: FITC anti-GL7, PE-Cy7 anti-B220, allophycocyanin anti-IgM, AF700 anti-CD38, and eF450 anti-IgD; and for the following allophycocyanin–eF780 labeled anti-mouse Abs for non-B cell surface markers: anti-CD90.2, anti-CD11c, anti–Ly-6G, and anti-F4/80. All Abs were from eBioscience (San Diego, CA) with the exception of FITC anti-GL7 (BD Pharmingen). Cells were fixed in formaldehyde (Cytofix/Cytoperm; BD Biosciences, San Diego, CA), washed with permeabilization buffer (BD Biosciences), and incubated with the Pacific Orange–labeled anti-mouse Ab anti-Ig H and L chains, both intracellular and extracellular (Invitrogen), as described previously (20).

The gating strategy used to analyze 6OXY-specific B cells was previously described in detail (23). Total B cells were identified as Ig-positive cells that did not express the non-B cell markers CD90.2 (T cells), Gr-1 (neutrophils), CD11.c (dendritic cells), and F4/80 (macrophages). B cells binding the 6OXY hapten are distinguished from B cells bound to the PE-based decoys by comparing fluorescence between (6OXY)n-PE and PE-AF647 or (6OXY)4-SA-PE and (control peptide)4-SA-PE-AF647. 6OXY-specific B cells were identified as either Ighigh Ab-secreting cells (ASC) or B220high non-ASC. In contrast, the B220negIgMID cells were excluded from B cell analysis (23), because this population consisted of non-B cells that bound circulating Ab via FcRs (31). 6OXY-specific B220high B cells were classified as CD38lowGL7high germinal center (GC) or CD38high naive and memory B cells. CD38high GL7neg B cells were further identified as IgMhigh B cells or as IgMneg and IgDneg switched Ig (swIg) B cells as described previously (20, 23). When specified, 6OXY-specific GC B cells were quantified as CD38+GL7high or CD38negGL7high B cells as previously described (22) and shown in Supplemental Fig. 2. We have previously shown that enriched 6OXY-specific B cells are specific for free oxycodone and hapten (23).

Tissue samples were disaggregated to obtain a single-cell suspension. For CD4+ and CD8+ analysis, samples were directly labeled for T and non-T surface markers as described below. For detection of 2W1S-specific CD4+ T cells, samples were first coincubated with 10 nM 2W1S:I-Ab-PE, 2W1S:I-Ab–allophycocyanin, and BV421 anti-CXCR5 for 60 min at room temperature (27), washed, mixed with anti-PE and anti-allophycocyanin magnetic beads, and isolated through magnetic enrichment (27). Enriched single-cell suspensions were labeled with PE-Cy7 anti-CD45R, PE-Cy7 anti-CD11b, PE-Cy7 anti-CD11c, PE-Cy7 anti-F4/80, PE anti-CD4, BV510 anti-CD8a, allophycocyanin–eFluor780 anti-CD4, PerCP-eFluor710 anti-CD90.2, FITC anti-CD279, and AF700 anti-CD44. Using a flow cytometer, after identification of singlets through standard forward light scatter-W and forward light scatter-A gating, T cells were first identified as CD90.2high and then as CD8high T cells or CD4high T cells (27). Within the CD4high T cells, 2W1S-specific CD4+ T cells were identified as PEhigh and allophycocyaninhigh (27). Within the 2W1S-specific T cell gate, we identified CXCR5high T cells (27). However, CD4+ CXCR5high TfH are not usually found in the spleen of naive mice but in this study served as a site-specific population-specific internal negative control. T cell numbers were quantified using fluorescent counting beads (Accucheck; Invitrogen), whereas compensation was performed using Ultracorp eBeads (eBioscience, San Diego, CA).

B and T cell analyses were performed on a four-laser (405, 488, 561, and 640 nm) LSR II Fortessa using FACSDiva 7.0 (BD Biosciences), and processed with FlowJo (Tree Star, Ashland, OR).

6OXY-specific B cell numbers, oxycodone concentrations, and MPE% were compared by two-sided paired or unpaired t test within—or between—groups. 6OXY-specific B cells across groups were compared by one-way ANOVA followed by Tukey’s post hoc test. The relationship between the 6OXY-specific B cells, CD4+ T cells, carrier-specific CD4+ T cells, and 6OXY-specific titers, oxycodone concentrations and oxycodone MPE% within each group was analyzed by either Pearson or Spearman correlation after performing the D’Agostino and Pearson omnibus normality test. Analyses were performed using Graph Pad 6.0 (La Jolla, CA).

To test the hypothesis that the initial size of the polyclonal hapten–specific B cell population determines vaccine efficacy, we analyzed B cells in naive and immunized mice using fluorescent Ag-based magnetic enrichment paired with flow cytometry (23). In two independent studies, the 6OXY-specific B cell repertoire was analyzed in either spleen biopsy or blood samples collected before and after immunization with 6OXY-KLH or unconjugated KLH in alum (Fig. 1A). Immunized mice were tested for serum Abs, vaccine effect on oxycodone distribution to serum and to the brain, and oxycodone-induced behavior. In the cohort of mice that was analyzed by partial splenectomy at day 35, we determined 6OXY-specific serum IgG titers (Fig. 1B) and their functional effects following injection with oxycodone (2.25 mg/kg, s.c.). This dose of oxycodone elicits maximal responses in the hot plate analgesia test providing an effective strategy to screen vaccine efficacy against opioids (25, 28). As expected, immunization with 6OXY-KLH increased serum oxycodone concentrations (Fig. 1C), reduced distribution of oxycodone to the brain, and oxycodone antinociception (Fig. 1D, 1E).

Having determined that partial splenectomy did not interfere with the 6OXY-KLH efficacy and knowing that partial splenectomy did not interfere with the IgMhigh memory B cell population in mice (32), we took advantage of this approach to test the extent to which vaccine efficacy against oxycodone correlated to the naive and early-activated 6OXY-specific B cell populations. In spleen biopsy samples from naive mice, we detected a substantial number of 6OXY-specific B220highGL7CD38highIgMhigh B cells (2950 ± 455, mean ± SEM), but very low numbers of 6OXY-specific B220highGL7CD38highIgMnegIgDneg swIg B cells, B220highCD38lowGL7high GC B cells, or 6OXY-specific B220lowIghigh ASC B cells (Fig. 2). 6OXY-specific ASC B cells were detected 14 d after the first immunization, but this response waned by day 35 (Fig. 2A, 2B). Multiple immunizations with 6OXY-KLH increased 6OXY-specific GC and swIg B cells (Fig. 2A, 2C, 2D). In contrast, no differences were detected in the number of 6OXY-specific IgMhigh B cells before and after immunization (Fig. 2A, 2E). Longitudinal B cell analysis by spleen biopsy showed that 6OXY-KLH elicits 6OXY-specific GC and swIg B cells, which are critical for generation of long-term plasma and memory B cell responses.

Vaccine efficacy depends on adequate levels of serum Abs that bind oxycodone in serum, thus preventing oxycodone distribution to the brain and oxycodone-induced behaviors (24, 25). Thus, we tested the extent to which the number of hapten-specific B cells prior to immunization correlated with vaccine efficacy. After spleen biopsies, the first cohort of naive BALB/c mice (n = 24) was randomly assigned to either the 6OXY-KLH or KLH immunization groups. The number of 6OXY-specific B cells was not different, in spleen biopsy samples, between the two groups prior to immunization (Supplemental Fig. 1A–C). As expected, immunization with 6OXY-KLH absorbed on alum significantly increased serum oxycodone concentrations after oxycodone administration (Supplemental Fig. 1D). Retention of oxycodone in serum correlated with the appearance of 6OXY-specific serum IgG Abs after immunization (Fig. 3A), as previously described in spleen-intact mice and rats (24, 25). Interestingly, a higher number of 6OXY-specific IgMhigh and swIg B cells prior to immunization correlated to greater 6OXY-specific serum IgG titers 7 d after the last immunization (Fig. 3B, 3C). In contrast, no relationship was found between the 6OXY-specific GC or ASC B cells prior to immunization and Abs (data not shown).

There was a trending association between increased levels of 6OXY-specific serum IgG Abs and decreased distribution of oxycodone to the brain (Fig. 3D). Strikingly, a higher number of 6OXY-specific IgMhigh B cells prior to immunization correlated to lower brain oxycodone concentrations in the 6OXY-KLH group (Fig. 3E). In contrast, no significant relationship was found between 6OXY-specific swIg (r2 = 0.28, p = 0.08; Fig. 3F), GC, or ASC B cells prior to immunization and brain oxycodone in vaccinated mice (data not shown). These data showed that the presence of a higher number of hapten-specific IgMhigh B cells, and to a lesser extent swIg B cells, prior to immunization correlated to greater Ab titers and vaccine efficacy, supporting our hypothesis that the initial number of hapten-specific B cells underlies individual responses to vaccines. As a cautionary note, three or four individual data points may have contributed to determine the significant relationship between hapten-specific IgMhigh and swIg B cells and vaccine effects (Fig. 3). However, all data points were included in the statistical analysis because our study focused on individual variability in response to vaccines and the relationship between B cells and multiple independent parameters of vaccine efficacy. The lack of association between Ab titers or vaccine efficacy and numbers of 6OXY-specific GC or ASC B cells prior to immunization is consistent with the notion that these B cell phenotypes are generated only after vaccination.

Immunization with 6OXY-KLH increased the number of 6OXY-specific GC B cells in the spleen compared with KLH and naive groups (Fig. 2C). Similarly, compared with naive mice (Fig. 4A), the ratio between 6OXY-specific GC B cells and other B cell phenotypes changed 14 d after vaccination (Fig. 4B). Individual distributions of B cell subsets (e.g., the ratio between GC:IgMhigh:swIg:ASC) were qualitatively different suggesting variable GC responses, which may underlie individual differences in vaccine efficacy.

To better understand the polyclonal B cell responses to 6OXY-KLH and to test whether vaccination elicits characteristic GC activation, we further characterized 6OXY-specific GC B cells as GC-independent CD38+GL7high or GC-independent CD38GL7high B cells (22). Prior to immunization, 6OXY-specific CD38+GL7high B cells generally outnumbered CD38GL7high B cells (451 ± 73 versus 107 ± 23, mean ± SEM, p < 0.001; Supplemental Fig. 2). In contrast, immunization with 6OXY-KLH specifically increased the number of 6OXY-specific CD38GL7high B cells compared with KLH and naive groups (Supplemental Fig. 2), thus increasing the ratio of CD38GL7high over CD38+GL7high B cells (Fig. 4C), which reflects GC activation.

In the cohort of mice analyzed by spleen biopsy 14 d after the first immunization, 6OXY-KLH effectively increased serum oxycodone concentrations (Fig. 4D). There was also greater GC activation, as shown by the increased ratio of 6OXY-specific CD38 to CD38+GL7high GC B cells in individual mice, correlating with greater efficacy on oxycodone serum distribution (Fig. 4E). In fact, increased number of 6OXY-specific CD38GL7high B cells (r = 0.73, p < 0.0001; data not shown) and 6OXY-specific Ighigh B cells (Fig. 4F) 14 d after immunization correlated with greater subsequent efficacy in increasing serum oxycodone concentrations at 35 d. Collectively, these data support our hypothesis that the size of the early-activated hapten-specific B cell population contributes to the conjugate vaccine efficacy.

To test translational validity, we studied the extent to which the size of the 6OXY-specific B cell population in blood correlated to vaccine efficacy against oxycodone. To identify rare high-affinity 6OXY-specific B cells in small blood volumes, we have developed an enrichment reagent consisting of the 6OXY hapten conjugated to a short biotinylated peptide bound to SA-PE, which provided the (6OXY)4-SA-PE fluorescent conjugate of known haptenization ratio. Using this tool, we analyzed the frequency of polyclonal 6OXY-specific B cells in the blood of a cohort of BALB/c mice (n = 24) before and after immunization with either 6OXY-KLH or KLH in alum. Immunization was effective in decreasing distribution of oxycodone to the brain (Fig. 5A), and the increased reduction of brain oxycodone concentration was dependent on early generation of serum Ab titers against oxycodone (Fig. 5B). Instead, an increased frequency of 6OXY-specific IgMhigh B cells in enriched blood samples collected prior to immunization correlated with greater vaccine efficacy in blocking oxycodone brain distribution in mice (Fig. 5C). Furthermore, immunization with 6OXY-KLH increased the blood frequency of 6OXY-specific IgMhigh, swIg, and GC B cells compared with the KLH control group (Fig. 5D–F). In contrast, 6OXY-specific ASC B cells were barely detected in blood before or 14 d after immunization. Analysis of 6OXY-specific B cells in blood was performed 14 d after immunization to be consistent with analysis performed by spleen biopsy (Figs. 14) and our previous report (23).

An increased frequency of 6OXY-specific IgMhigh B cells in enriched blood samples collected 14 d after immunization correlated with subsequent greater vaccine effect on oxycodone distribution to serum (Fig. 5G), brain (Fig. 5H), and prevention of oxycodone behavioral effects (Fig. 5I). A similar significant relationship was found between 6OXY-specific GC and swIg B cells and vaccine efficacy (Supplemental Fig. 3). The association between vaccine efficacy and 6OXY-specific B cells either before or shortly after immunization supports the hypothesis that naive and early-activated hapten-specific B cells are derived from the same B cell compartment and that the initial size of the 6OXY-specific B cell population correlates to vaccine success.

To further support the finding that early B cell clonal expansion 14 d after immunization dictated subsequent therapeutic effects at day 35, we tested 6OXY-KLH efficacy following CD4+ T cell depletion in C57BL/6 mice (n = 6 each group). Anti-CD4 mAb–mediated selective depletion of CD4+ T cells (Fig. 6A) led to decreased frequency of 6OXY-specific IgMhigh and swIg B cells in blood (Fig. 6B). In this study, 6OXY-specific ASC and GC B cells were not detected in blood after immunization (data not shown), which reflected the observed overall lower frequencies of 6OXY-specific B cells in enriched blood from C57BL/6 mice compared with BALB/c. As expected, CD4+ T cell depletion blunted vaccine efficacy against oxycodone (Fig. 6C).

Then, we tested the extent to which the population size of carrier-specific T cells related to vaccine efficacy. To analyze carrier-specific naive CD4+ T cells, we used a peptide:MHC class II tetramer–based enrichment method (33). This strategy has been previously used to track Ag-specific T cells during the course of a bacterial infection in mice (34). Prior to immunization, by means of spleen biopsy, we determined the number of CD4+ T cells specific for the 2W1S variant of peptide52–68 epitope from the I-Eα-chain (33). We then immunized C57BL/6 mice with the 6OXY-OVA labeled with the 2W1S peptide (6OXY-OVA2W, n = 9). As expected, higher numbers of 2W:I-Ab–specific CD4+ T cells prior to immunization were tightly correlated to vaccine efficacy (Fig. 6D). These data suggest that 6OXY-specific B cells and carrier-specific CD4+ T cells were highly interrelated and both contributed to vaccine efficacy.

Before vaccination, the frequency of 6OXY-specific B cell subsets among total B cells in the enriched bound fraction (Fig. 7) was consistent across spleen biopsy, blood, and the previously analyzed pooled peripheral lymph nodes and spleen samples (23). The 6OXY-specific B cell repertoire consisted mainly of IgMhigh B cells (Fig. 7). From the total B cells detected in enriched spleen biopsy samples (246,600 ± 32,000, mean ± SEM, n = 23 mice), 1.5 ± 0.2% were 6OXY-specific IgMhigh B cells (2,950 ± 455, mean ± SEM), whereas 6OXY-specific IgMhigh B cells constituted 0.8 ± 0.12% of total B cells in lymph nodes and spleen samples (Fig. 7A). From the total B cells detected in enriched blood samples (5440 ± 610, mean ± SEM, n = 24 mice), 2.2 ± 0.3% were 6OXY-specific IgMhigh B cells (125 ± 20, mean ± SEM, n = 24). These data showed that analysis of hapten-specific B cells can be performed in various tissues, including peripheral blood, prior to immunization.

The translation of therapeutic vaccines against drug addiction has been slow because clinical efficacy is achieved only in the small subset of immunized subjects that showed the highest levels of drug-specific serum Abs. Previously, we found that a greater number of hapten-specific B cells correlated to greater efficacy of oxycodone vaccines consisting of conjugate immunogens containing different structurally related haptens (23) or administered in different adjuvant formulations (28). In this study, we hypothesized that clinically significant responses to addiction vaccines may be predicted from the number of hapten-specific B cells in unimmunized subjects and by the number of activated hapten–specific B cells present soon after immunization. Using an Ag-based enrichment strategy paired with flow cytometry analysis, we found that a higher frequency of naive and early-activated hapten-specific B cells in spleen biopsies and blood correlated with subsequent greater vaccine efficacy against oxycodone. In addition, CD4+ T cell–dependent B cell activation was required for vaccine efficacy, and the number of carrier-specific CD4+ T cells prior to immunization correlated with vaccine efficacy. Our data suggest that vaccine clinical efficacy may be increased by analyzing blood from patients for the presence of hapten-specific B cells to determine which subjects are most likely to benefit from vaccination. This strategy may help physicians to personalize vaccine-based therapy or to monitor clinical trials at early stages. Although more studies are needed before extending these results beyond addiction vaccines, it is tempting to speculate that the efficacy of other therapeutic vaccines containing haptens derived from tumor-associated small carbohydrate and peptide Ags, or Alzheimer’s β amyloid–derived peptides may be similarly influenced and predicted by the number of Ag-specific B cells at the time of vaccination.

We have previously shown that a single immunization with the 6OXY-KLH immunogen selectively increased the polyclonal hapten–specific B cell population in pooled peripheral lymph nodes and spleen (23). In the current study, one immunization or multiple boosts with 6OXY-KLH in alum adjuvant increased the polyclonal hapten–specific B cell population, which includes the ASC, GC, IgMhigh, and swIg B cell subsets, in either the spleen or blood compartments, which is consistent with previous analysis (23), and reports of NP-specific B cells in mice immunized with NP-KLH (35) or PE-specific B cells in mice immunized with PE (22). These results suggest that immunization with 6OXY-KLH in alum adjuvant induces GC formation and plasma and memory B cells, supporting previous evidence of preclinical efficacy.

In the spleen, more hapten-specific IgMhigh naive and memory B cells, and at least in part switched memory B cells, before immunization correlated to greater postvaccination Ab responses and vaccine efficacy in blocking distribution of oxycodone to the brain. Also in spleen, the numbers of 6OXY-specific GC and ASC B cells shortly after immunization correlated to subsequent vaccine efficacy. In blood, the presence of increased numbers of 6OXY-specific IgMhigh B cells before immunization, or 6OXY-specific IgMhigh, GC, and swIg B cells at 2 wk after vaccination, correlated to increased vaccine ability to prevent brain distribution and block oxycodone antinociception. These results, summarized in Supplemental Table I, suggest that the frequencies of 6OXY-specific IgMhigh and swIg B cell subsets are the best predictors of vaccine efficacy. Moreover, these data suggest that the initial size of the polyclonal hapten–specific naive B cell population correlates to the size of the early-activated B cell population shortly after vaccination, which, during GC formation, determines the magnitude of the Ab response against oxycodone and vaccine efficacy. Our study is supported by findings that various preimmunization serologic, transcriptomic, and cellular parameters in human blood, including the frequency of B cell subsets, were stable over time and correlated with Ab responses in subjects immunized with seasonal and pandemic influenza vaccines (36). The hypothesis that vaccine-specific B cells may provide predictive biomarkers of vaccine efficacy is further supported by a correlation found between switched memory B cells in human PBMC and the efficacy of an H1N1 influenza vaccine (37).

Immunization with 6OXY-KLH increased GC B cells, and a more detailed analysis of GC B cells in spleen biopsies revealed that 6OXY-KLH stimulated expansion of genuine CD38negGL7high GC B cells, which is a hallmark of longer-lived swIg memory B cell through GC formation (22). In contrast, generation of CD38+GL7high B cells is a critical feature of GC-independent memory formation, which leads to differentiation in IgMhigh and shorter-lived swIg memory B cells (22). The presence of CD38negGL7high GC B cells after immunization supports the 6OXY-KLH’s ability to generate swIg memory B cells rather than IgMhigh B cells. In fact, increased GC activation and the frequency of hapten-specific GC B cells tightly correlated to vaccine efficacy against oxycodone. These data suggest that the lack of addiction vaccine efficacy in patients displaying low Ab titers is not due to GC-independent hapten-specific B cell responses. Rather, these data indicate that after immunization, variability in the timing, and quality and magnitude of GC activation might underlie individual differences in therapeutic vaccine efficacy against drugs of abuse or other small molecules.

Selective depletion of CD4+ Th cells in secondary lymphoid organs prevented activation of hapten-specific B cells and blunted vaccine efficacy. These results suggest that a large number of naive and early-activated hapten-specific B cells, in the presence of a sizeable carrier-specific CD4+ T cell population, may be required for optimal T cell–dependent B cell activation and vaccine efficacy against oxycodone. Indeed, using carrier proteins labeled with known T cell epitopes and peptide:I-Ab tetramers, to study naive CD4+ T cells, we found that the initial population of carrier-specific CD4+ T cells prior to immunization correlated to vaccine efficacy against oxycodone. Overall, our results are consistent with findings that increased numbers of GC B cells are correlated to, and necessary for, increased Ag-specific TfH cell responses (38, 39). However, the novelty of our study is that we have established a relationship between the polyclonal hapten-specific B cell and carrier-specific CD4+ T cell populations and vaccine efficacy in wild-type BALB/c and C57BL/6 mice.

In this study, we reported a correlation between the size of the polyclonal naive and early-activated hapten-specific B cell populations and subsequent measures of vaccine efficacy against oxycodone. In this study, we have focused on vaccine’s effects shortly (7 d) after the last immunization to be consistent with our previous vaccine development work (2426, 28). Clinical evaluation of nicotine and cocaine vaccines involved immunization regimens of three to five monthly doses followed by booster injections every 2 or 6 mo to maintain high serum Ab levels, showing half lives of ∼2–3 mo (6, 7, 40). In future investigations, it will be important to address how the frequency of naive hapten–specific B cells, or alternatively carrier-specific CD4+ T cells, influences long-term adaptive immune responses to vaccines against drugs of abuse.

The current findings, and the reported methods for analysis of Ag-specific B and T cells, may aid the development of novel mechanism-based strategies to increase the overall efficacy of vaccines against drugs of abuse. Analysis of hapten-specific B cells and carrier-specific CD4+ T cells may speed vaccine discovery by screening the lead haptens that best bind B cells, novel vaccine carriers, and delivery platforms that enhance stimulation of T cell help or by comparing candidate conjugate immunogens to predict their likelihood of clinical efficacy.

Using our enrichment strategy, hapten-specific B cells were easily detected in mouse blood before and after immunization. Application of this approach to human PBMC may inform development of therapeutic vaccines suitable for human testing and companion diagnostic tests to identify patients that will likely respond to vaccine therapy. Indeed, B cell–based predictive biomarkers may aid patient stratification and design of adaptive clinical trials. The clinical benefit of such strategies is supported by evidence that pre-existing serum Abs for a blood group determinant and a viral glycan correlated with overall survival in patients immunized with a candidate prostate cancer vaccine (41, 42). Analysis of the vaccine-specific B and T cell repertoire may be combined with high-throughput immune repertoire sequencing (43) to accelerate development of vaccines against chronic noncommunicable diseases.

We thank Minneapolis Medical Research Foundation for purchasing a BD Biosciences LSRIIFortessa, Marc Jenkins’ group for providing the 2W1S:I-Ab tetramers and critical support, and Paul Pentel for helpful discussion. We also thank Maria Luisa Gelmi for establishing an institutional and educational agreement between the Facoltà di Scienze del Farmaco, Università degli Studi di Milano and the University of Minnesota Medical School.

This work was supported by National Institutes of Health Grant DA034487 (to M.P.) and the Minneapolis Medical Research Foundation Translational Research Award (to M.P.).

The online version of this article contains supplemental material.

Abbreviations used in this article:

ASC

Ab-secreting cell

GC

germinal center

KLH

keyhole limpet hemocyanin

MS

mass spectroscopy; non-ASC, non–Ab-secreting cell

6OXY

oxycodone-based hapten

SA

streptavidin

swIg

switched Ig

TfH

T follicular helper cell.

1
Rappuoli
R.
,
Pizza
M.
,
Del Giudice
G.
,
De Gregorio
E.
.
2014
.
Vaccines, new opportunities for a new society.
Proc. Natl. Acad. Sci. USA
111
:
12288
12293
.
2
Pentel
P. R.
,
LeSage
M. G.
.
2014
.
New directions in nicotine vaccine design and use.
Adv. Pharmacol.
69
:
553
580
.
3
Shen
X. Y.
,
Orson
F. M.
,
Kosten
T. R.
.
2012
.
Vaccines against drug abuse.
Clin. Pharmacol. Ther.
91
:
60
70
.
4
Astronomo
R. D.
,
Burton
D. R.
.
2010
.
Carbohydrate vaccines: developing sweet solutions to sticky situations?
Nat. Rev. Drug Discov.
9
:
308
324
.
5
Liu
B.
,
Frost
J. L.
,
Sun
J.
,
Fu
H.
,
Grimes
S.
,
Blackburn
P.
,
Lemere
C. A.
.
2013
.
MER5101, a novel Aβ1-15:DT conjugate vaccine, generates a robust anti-Aβ antibody response and attenuates Aβ pathology and cognitive deficits in APPswe/PS1ΔE9 transgenic mice.
J. Neurosci.
33
:
7027
7037
.
6
Hatsukami
D. K.
,
Jorenby
D. E.
,
Gonzales
D.
,
Rigotti
N. A.
,
Glover
E. D.
,
Oncken
C. A.
,
Tashkin
D. P.
,
Reus
V. I.
,
Akhavain
R. C.
,
Fahim
R. E.
, et al
.
2011
.
Immunogenicity and smoking-cessation outcomes for a novel nicotine immunotherapeutic.
Clin. Pharmacol. Ther.
89
:
392
399
.
7
Martell
B. A.
,
Orson
F. M.
,
Poling
J.
,
Mitchell
E.
,
Rossen
R. D.
,
Gardner
T.
,
Kosten
T. R.
.
2009
.
Cocaine vaccine for the treatment of cocaine dependence in methadone-maintained patients: a randomized, double-blind, placebo-controlled efficacy trial.
Arch. Gen. Psychiatry
66
:
1116
1123
.
8
Gilewski
T. A.
,
Ragupathi
G.
,
Dickler
M.
,
Powell
S.
,
Bhuta
S.
,
Panageas
K.
,
Koganty
R. R.
,
Chin-Eng
J.
,
Hudis
C.
,
Norton
L.
, et al
.
2007
.
Immunization of high-risk breast cancer patients with clustered sTn-KLH conjugate plus the immunologic adjuvant QS-21.
Clin. Cancer Res.
13
:
2977
2985
.
9
Miles
D.
,
Roché
H.
,
Martin
M.
,
Perren
T. J.
,
Cameron
D. A.
,
Glaspy
J.
,
Dodwell
D.
,
Parker
J.
,
Mayordomo
J.
,
Tres
A.
, et al
.
2011
.
Phase III multicenter clinical trial of the sialyl-TN (STn)-keyhole limpet hemocyanin (KLH) vaccine for metastatic breast cancer.
Oncologist
16
:
1092
1100
.
10
UNODC. 2013. World Drug Report 2013
.
11
CDC
.
2014
.
Wide-ranging Online Data for Epidemiologic Research.
CDC
,
Atlanta, GA
.
12
Dodrill
C. L.
,
Helmer
D. A.
,
Kosten
T. R.
.
2011
.
Prescription pain medication dependence.
Am. J. Psychiatry
168
:
466
471
.
13
Stotts
A. L.
,
Dodrill
C. L.
,
Kosten
T. R.
.
2009
.
Opioid dependence treatment: options in pharmacotherapy.
Expert Opin. Pharmacother.
10
:
1727
1740
.
14
Skolnick
P.
,
Volkow
N. D.
.
2012
.
Addiction therapeutics: obstacles and opportunities.
Biol. Psychiatry
72
:
890
891
.
15
McHeyzer-Williams
M.
,
Okitsu
S.
,
Wang
N.
,
McHeyzer-Williams
L.
.
2012
.
Molecular programming of B cell memory.
Nat. Rev. Immunol.
12
:
24
34
.
16
Crotty
S.
2011
.
Follicular helper CD4 T cells (TFH).
Annu. Rev. Immunol.
29
:
621
663
.
17
Taylor
J. J.
,
Jenkins
M. K.
,
Pape
K. A.
.
2012
.
Heterogeneity in the differentiation and function of memory B cells.
Trends Immunol.
33
:
590
597
.
18
Jenkins
M. K.
,
Moon
J. J.
.
2012
.
The role of naive T cell precursor frequency and recruitment in dictating immune response magnitude.
J. Immunol.
188
:
4135
4140
.
19
Geiger
R.
,
Duhen
T.
,
Lanzavecchia
A.
,
Sallusto
F.
.
2009
.
Human naive and memory CD4+ T cell repertoires specific for naturally processed antigens analyzed using libraries of amplified T cells.
J. Exp. Med.
206
:
1525
1534
.
20
Pape
K. A.
,
Taylor
J. J.
,
Maul
R. W.
,
Gearhart
P. J.
,
Jenkins
M. K.
.
2011
.
Different B cell populations mediate early and late memory during an endogenous immune response.
Science
331
:
1203
1207
.
21
Taylor
J. J.
,
Martinez
R. J.
,
Titcombe
P. J.
,
Barsness
L. O.
,
Thomas
S. R.
,
Zhang
N.
,
Katzman
S. D.
,
Jenkins
M. K.
,
Mueller
D. L.
.
2012
.
Deletion and anergy of polyclonal B cells specific for ubiquitous membrane-bound self-antigen.
J. Exp. Med.
209
:
2065
2077
.
22
Taylor
J. J.
,
Pape
K. A.
,
Jenkins
M. K.
.
2012
.
A germinal center-independent pathway generates unswitched memory B cells early in the primary response.
J. Exp. Med.
209
:
597
606
.
23
Taylor
J. J.
,
Laudenbach
M.
,
Tucker
A. M.
,
Jenkins
M. K.
,
Pravetoni
M.
.
2014
.
Hapten-specific naïve B cells are biomarkers of vaccine efficacy against drugs of abuse.
J. Immunol. Methods
405
:
74
86
.
24
Pravetoni
M.
,
Le Naour
M.
,
Harmon
T. M.
,
Tucker
A. M.
,
Portoghese
P. S.
,
Pentel
P. R.
.
2012
.
An oxycodone conjugate vaccine elicits drug-specific antibodies that reduce oxycodone distribution to brain and hot-plate analgesia.
J. Pharmacol. Exp. Ther.
341
:
225
232
.
25
Pravetoni
M.
,
Le Naour
M.
,
Tucker
A. M.
,
Harmon
T. M.
,
Hawley
T. M.
,
Portoghese
P. S.
,
Pentel
P. R.
.
2013
.
Reduced antinociception of opioids in rats and mice by vaccination with immunogens containing oxycodone and hydrocodone haptens.
J. Med. Chem.
56
:
915
923
.
26
Pravetoni
M.
,
Pentel
P. R.
,
Potter
D. N.
,
Chartoff
E. H.
,
Tally
L.
,
LeSage
M. G.
.
2014
.
Effects of an oxycodone conjugate vaccine on oxycodone self-administration and oxycodone-induced brain gene expression in rats.
PLoS ONE
9
:
e101807
.
27
Tubo
N. J.
,
Pagán
A. J.
,
Taylor
J. J.
,
Nelson
R. W.
,
Linehan
J. L.
,
Ertelt
J. M.
,
Huseby
E. S.
,
Way
S. S.
,
Jenkins
M. K.
.
2013
.
Single naive CD4+ T cells from a diverse repertoire produce different effector cell types during infection.
Cell
153
:
785
796
.
28
Pravetoni
M.
,
Vervacke
J. S.
,
Distefano
M. D.
,
Tucker
A. M.
,
Laudenbach
M.
,
Pentel
P. R.
.
2014
.
Effect of currently approved carriers and adjuvants on the pre-clinical efficacy of a conjugate vaccine against oxycodone in mice and rats.
PLoS One
9
:
e96547
.
29
Norian
L. A.
,
Kresowik
T. P.
,
Rosevear
H. M.
,
James
B. R.
,
Rosean
T. R.
,
Lightfoot
A. J.
,
Kucaba
T. A.
,
Schwarz
C.
,
Weydert
C. J.
,
Henry
M. D.
,
Griffith
T. S.
.
2012
.
Eradication of metastatic renal cell carcinoma after adenovirus-encoded TNF-related apoptosis-inducing ligand (TRAIL)/CpG immunotherapy.
PLoS One
7
:
e31085
.
30
Pravetoni
M.
,
Raleigh
M. D.
,
Le Naour
M.
,
Tucker
A. M.
,
Harmon
T. M.
,
Jones
J. M.
,
Birnbaum
A. K.
,
Portoghese
P. S.
,
Pentel
P. R.
.
2012
.
Co-administration of morphine and oxycodone vaccines reduces the distribution of 6-monoacetylmorphine and oxycodone to brain in rats.
Vaccine
30
:
4617
4624
.
31
Bell
J.
,
Gray
D.
.
2003
.
Antigen-capturing cells can masquerade as memory B cells.
J. Exp. Med.
197
:
1233
1244
.
32
Tracy
E. T.
,
Haas
K. M.
,
Gentry
T.
,
Danko
M.
,
Roberts
J. L.
,
Kurtzberg
J.
,
Rice
H. E.
.
2011
.
Partial splenectomy but not total splenectomy preserves immunoglobulin M memory B cells in mice.
J. Pediatr. Surg.
46
:
1706
1710
.
33
Moon
J. J.
,
Chu
H. H.
,
Hataye
J.
,
Pagán
A. J.
,
Pepper
M.
,
McLachlan
J. B.
,
Zell
T.
,
Jenkins
M. K.
.
2009
.
Tracking epitope-specific T cells.
Nat. Protoc.
4
:
565
581
.
34
Dileepan
T.
,
Linehan
J. L.
,
Moon
J. J.
,
Pepper
M.
,
Jenkins
M. K.
,
Cleary
P. P.
.
2011
.
Robust antigen specific th17 T cell response to group A Streptococcus is dependent on IL-6 and intranasal route of infection.
PLoS Pathog.
7
:
e1002252
.
35
King
I. L.
,
Fortier
A.
,
Tighe
M.
,
Dibble
J.
,
Watts
G. F.
,
Veerapen
N.
,
Haberman
A. M.
,
Besra
G. S.
,
Mohrs
M.
,
Brenner
M. B.
,
Leadbetter
E. A.
.
2012
.
Invariant natural killer T cells direct B cell responses to cognate lipid antigen in an IL-21‑dependent manner.
Nat. Immunol.
13
:
44
50
.
36
Tsang
J. S.
,
Schwartzberg
P. L.
,
Kotliarov
Y.
,
Biancotto
A.
,
Xie
Z.
,
Germain
R. N.
,
Wang
E.
,
Olnes
M. J.
,
Narayanan
M.
,
Golding
H.
, et al
.
2014
.
Global analyses of human immune variation reveal baseline predictors of postvaccination responses.
Cell
157
:
499
513
.
37
Frasca
D.
,
Diaz
A.
,
Romero
M.
,
Phillips
M.
,
Mendez
N. V.
,
Landin
A. M.
,
Blomberg
B. B.
.
2012
.
Unique biomarkers for B-cell function predict the serum response to pandemic H1N1 influenza vaccine.
Int. Immunol.
24
:
175
182
.
38
Baumjohann
D.
,
Preite
S.
,
Reboldi
A.
,
Ronchi
F.
,
Ansel
K. M.
,
Lanzavecchia
A.
,
Sallusto
F.
.
2013
.
Persistent antigen and germinal center B cells sustain T follicular helper cell responses and phenotype.
Immunity
38
:
596
605
.
39
Kitano
M.
,
Moriyama
S.
,
Ando
Y.
,
Hikida
M.
,
Mori
Y.
,
Kurosaki
T.
,
Okada
T.
.
2011
.
Bcl6 protein expression shapes pre-germinal center B cell dynamics and follicular helper T cell heterogeneity.
Immunity
34
:
961
972
.
40
Cornuz
J.
,
Zwahlen
S.
,
Jungi
W. F.
,
Osterwalder
J.
,
Klingler
K.
,
van Melle
G.
,
Bangala
Y.
,
Guessous
I.
,
Müller
P.
,
Willers
J.
, et al
.
2008
.
A vaccine against nicotine for smoking cessation: a randomized controlled trial.
PLoS One
3
:
e2547
.
41
Campbell
C. T.
,
Gulley
J. L.
,
Oyelaran
O.
,
Hodge
J. W.
,
Schlom
J.
,
Gildersleeve
J. C.
.
2013
.
Serum antibodies to blood group A predict survival on PROSTVAC-VF.
Clin. Cancer Res.
19
:
1290
1299
.
42
Campbell
C. T.
,
Gulley
J. L.
,
Oyelaran
O.
,
Hodge
J. W.
,
Schlom
J.
,
Gildersleeve
J. C.
.
2014
.
Humoral response to a viral glycan correlates with survival on PROSTVAC-VF.
Proc. Natl. Acad. Sci. USA
111
:
E1749
E1758
.
43
DeKosky
B. J.
,
Kojima
T.
,
Rodin
A.
,
Charab
W.
,
Ippolito
G. C.
,
Ellington
A. D.
,
Georgiou
G.
.
2015
.
In-depth determination and analysis of the human paired heavy- and light-chain antibody repertoire.
Nat. Med.
21
:
86
91
.

M.P. is the author of a patent on the oxycodone vaccine (Publication No. US20140093525-A1). The authors have no financial conflicts of interest.

Supplementary data