|Home | About | Journals | Submit | Contact Us | Français|
Current strategies to help tobacco smokers quit have limited success as a result of the addictive properties of the nicotine in cigarette smoke. We hypothesized that a single administration of an adeno-associated virus (AAV) gene transfer vector expressing high levels of an anti-nicotine antibody would persistently prevent nicotine from reaching its receptors in the brain. To test this hypothesis, we constructed an AAV.rh10 vector that expressed a full length, high affinity, anti-nicotine antibody derived from the Fab fragment of the anti-nicotine monoclonal antibody NIC9D9 (AAVantiNic). In mice treated with this vector, blood concentrations of the anti-nicotine antibody were dose-dependent, and the antibody showed high specificity and affinity for nicotine. The antibody shielded the brain from systemically administered nicotine, reducing brain nicotine concentrations to 15% of those in naive mice. The amount of nicotine sequestered in the serum of vector-treated mice was over 7 times greater than that in non-treated mice, with 83% of serum nicotine bound to IgG. Treatment with the AAVantiNic vector blocked nicotine-mediated alterations in arterial blood pressure, heart rate and locomotor activity. In summary, a single administration of a gene transfer vector expressing a high affinity anti-nicotine monoclonal antibody elicited persistent (18 weeks), high titers of an anti-nicotine antibody that obviated the physiologic effects of nicotine. If this degree of efficacy translates to humans, AAVantiNic could be an effective preventative therapy for nicotine addiction.
Cigarette smoking is a common addiction, with significant societal effects. Approximately 20% of the adults in the U.S. smoke cigarettes, and cigarette smoking accounts for 1 of every 5 deaths in the USA (1). Cigarette smoke causes chronic obstructive pulmonary disease (COPD) and lung cancer, and smoking is associated with an increased risk of cardiovascular disease, and a variety of non-lung neoplasms (2–5). Smoking-related health care and loss of productivity costs in excess of $193 billion annually in the United States (5).
Although each puff of cigarette smoke contains more than 4000 chemicals, the addictive properties of cigarette smoking derive from nicotine, a 162 Da alkaloid that represents 0.6–3.0% of the dry weight of tobacco (6–8). Most nicotine is pyrolized at the cigarette tip, but each cigarette typically delivers to the smoker 1.0 to 1.5 mg nicotine, which passes across the alveoli and into the blood stream, taking about 10 to 19 seconds to reach the brain (9–11). There, nicotine binds to the nicotinic acetylcholine receptor, triggering the conversion of L-tyrosine to dopamine, with resulting pleasure, reduced stress, alterations in blood pressure and heart rate, heightened alertness and increased ability to process information (12–14).
Despite the devastating health effects of nicotine addiction, current strategies of drug intervention and counseling to help smokers quit are mostly ineffective, with a 70 to 80% recidivism rate within 6 months (15). Current anti-smoking medications include nicotine replacement therapies, varenicline (a nicotinic receptor partial agonist) and bupropion (an anti-depressant) (16–18) but none have demonstrated high rates of efficacy and some have the potential for serious side effects; varenicline for example has recently been associated with adverse cardiovascular effects (15,17,19). One approach to treating nicotine addiction has been to develop an anti-nicotine vaccine, in which anti-nicotine antibodies bind to nicotine in the blood, preventing the drug from crossing the blood brain barrier and reaching its cognate receptors in the brain (20,21). Vaccines have had limited success, possibly as a result of failure to evoke a sufficiently high titer of a high-affinity antibody to nicotine. We therefore hypothesized that an adeno-associated viral gene transfer vector could be designed to express a known, high affinity anti-nicotine antibody at titers that would prevent nicotine from reaching the brain. Because AAV vectors can mediate persistent expression, we expected that this approach would require only a single vaccine administration. To evaluate this strategy, we generated AAVrh.10antiNic.Mab (referred to as AAVantiNic), a serotype rh.10 adeno-associated virus expressing NIC9D9, a high affinity anti-nicotine monoclonal antibody (22,23).
HEK 293 cells infected with the AAVantiNic vector (Fig. 1A) secreted IgG antibody, as demonstrated by coomassie blue stained SDS-PAGE and Western analysis (Fig. 1B,C). To assess the ability of AAVantiNic to express and maintain high titers of anti-nicotine antibody in serum, we injected C57Bl/6 mice intravenously with AAVantiNic at 3 doses: 109, 1010 or 1011 genome copies (gc). Using an anti-nicotine ELISA, we demonstrated the dose-dependence of the antibody, with the 1011gc group showing the highest serum concentrations of antibody at a mean titer of 1.1 ± 0.2 mg/ml at week 9 (Fig. 2A). This same dose generated a high antibody titer at 4 weeks (0.9 ± 0.1 mg/ml), which remained high until 18 weeks (1.3 ± 0.1 mg/ml), the longest time point evaluated (Fig. 2B). A competitive ELISA showed that the expressed anti-nicotine antibody had a higher affinity for nicotine than for the nicotine metabolites nornicotine and cotinine (Fig. 2C). The Kd for nicotine was 43 ± 20 nmol/l (Fig. 2D).
To test whether AAVantiNic could express sufficient anti-nicotine antibody to shield the brain from systemically administered nicotine, we challenged AAVantiNic-treated mice intravenously with [3H]nicotine (11,24) (Fig. 3). One minute after administration, total serum nicotine concentrations in AAVantiNic-treated mice (516 ng/ml) were 7.2 times higher than in naive control mice (71.7 ng/ml) (p<10−5). In the AAVantiNic-treated mice, 83% of the serum nicotine was bound to IgG. Conversely, nicotine concentrations in the brain of AAVantiNic-treated mice (12.2 ng/g brain) were 15% of those in non-treated naive control mice (79.2 ng/g brain), a 47-fold reduction in the ratio of brain to blood nicotine concentrations in the AAVantiNic-treated mice (p<10−5).
Systemic administration of nicotine robustly changed cardiovascular parameters in mice treated with the AAV control vector. Nicotine caused a 37% reduction in mean arterial pressure and a 46% reduction in heart rate in the control mice within 25 min (Fig. 4). In contrast, nicotine did not induce these blood pressure and heart rate responses in mice treated with AAVantiNic. As a control, we administered phosphate buffered saline (PBS) in place of nicotine to AAVantiNic-treated mice; no changes were seen in the mean arterial pressure or heart rate.
To determine whether AAVantiNic prevents nicotine-induced suppression of locomotion, AAVantiNic-treated mice were repeatedly challenged (4 to 7 weeks after vector administration) subcutaneously with 12.5 μg (0.5 mg/kg) of nicotine. Ambulatory activity was assessed for 15 min after nicotine administration for each of 10 challenges for 3 weeks (Fig. 5A). Naive control mice injected with nicotine showed a marked nicotine-dependent decrease in locomotor activity on all days tested, while AAVantiNic-treated mice showed the same ambulatory activity profile as the control mice given saline instead of nicotine (Fig. 5A). At day 18 of the nicotine challenge study (7 weeks after administration of AAVantiNic), we measured the cumulative distance traveled after nicotine administration. Naive control mice exhibited a nicotine-induced suppression of ambulatory activity (2.15 ± 0.30 m over 15 min, p<0.003) compared to saline-injected control mice, while AAVantiNic mice showed no nicotine-induced reduction in locomotor activity (6.38 ± 1.20 m), similar to control mice given saline (5.00 ± 0.75 m; p>0.6; Fig. 5B). The cumulative vertical activity profile demonstrated similar AAVantiNic-mediated protection from nicotine. Control mice given nicotine displayed 16.8 ± 3.0 sec (over 15 min) of vertical activity (p<10−5 compared to saline injected control mice), and AAVantiNic mice given nicotine exhibited 180.2 ± 26.7 sec of vertical activity, similar to the 200.6 ± 36.6 sec seen in the control mice given saline instead of nicotine (p>0.9; Fig. 5C).
The challenge for a successful vaccine for nicotine addiction is that it must evoke a high titer, high affinity, specific antibody in a broad spectrum of genetic backgrounds, a hurdle difficult to achieve through active vaccination. Our approach successfully avoids this obstacle by using the AAVrh.10 gene expression vector to effectively and systemically deliver the anti-nicotine monoclonal, NIC9D9. Our previous studies measuring the expression of a reporter transgene from intravenously delivered AAVrh.10 vectors in a mouse model show very high transduction efficiency in the liver with only minor transduction in other organs (25). Here, we have shown that an AAVantiNic vector effectively expresses a full-length, high affinity anti-nicotine monoclonal antibody and that a single administration of AAVantiNic to mice results in persistent, high concentrations of the antibody in serum at a 20-fold molar excess over the serum nicotine concentration of 70–110 ng/ml (greater than the nicotine serum concentrations of a continuous smoker) (11,26,27). In mice treated with AAVantiNic, parenterally administered nicotine became bound to antibody and was sequestered in the blood, preventing the drug from reaching the central nervous system, even at a challenge dose higher than a chronic smoker. Mice expressing the vector-derived antibody did not respond to nicotine with the usual alterations on the cardiovascular system or on ambulatory and vertical activity. If AAVantiNic shows efficacy in a rodent model of human nicotine self-administration this therapeutic approach would be a good candidate for human clinical studies.
Previous studies with active anti-nicotine vaccines have attempted to generate a potent humoral immune response against nicotine. The challenge of this approach is that nicotine is a small non-immunogenic molecule as demonstrated by the absence of immunity in smokers. To impart immunopotency, nicotine (or a nicotine analog) must be coupled to a larger molecule to induce a potent anti-nicotine immune response (20,28,29). For example, AM1, a trans-3′-(hydroxymethyl) nicotine-derived nicotine hapten with a linker containing an ether moiety and a free carboxyl group for conjugation, has been attached to carriers such as tetanus toxin to create an anti-nicotine vaccine. In a rodent self-administration model, this vaccine increased motivation for nicotine self-administration (30) in effect, surmounting vaccine mediated blockade of nicotine. Three active immunotherapy vaccines have been in clinical trials, including TA-NIC [a nicotine analog linked to cholera toxin B (Celtic Pharma)], NicVAX [a nicotine analog linked to Pseudomonas aeruginosa exoprotein A (Nabi Pharmaceuticals)], and NicQb [a nicotine analog linked to particles of the bacteriophage Qβ (Cytos Biotechnology)] (17,31). The carriers for NicVAX and NicQB are conjugated to 3′-aminomethyl nicotine, while the TA-NIC carrier is conjugated to an N'-butyric acid adduct of (s) nicotine (17,30,32). The data that are publically available indicate that these vaccines are well tolerated, and the individuals with the highest circulating concentrations of antibodies are more likely to abstain from smoking (17). However, in all cases there has been a large variation among the trial participants in the amount of antibody generated, and only a relatively small percentage of the participants - those with the highest serum anti-nicotine antibody titers - 24.6% (vs 12.0% in the placebo group) have abstained from smoking in the phase 2b clinical trial (17,31–34). Our vector, AAVantiNic expresses an anti-nicotine antibody with high affinity (similar to that of the investigational vaccines NicVAX (35) and NicQb (29)), but with a much higher concentration of serum antibody effectively preventing nicotine from entering the brain at only one minute after administration, indicating that AAVantiNic may be more effective, with a therapeutic threshold at higher doses of administered nicotine (Table 1).
In addition to the addictive effects of nicotine, cigarette smoking is strongly linked to the development of a number of pathological conditions, including diseases of the cardiovascular system (5). Central neural pathways, in part, mediate nicotine's acute and chronic effects on cardiovascular control (36), although the mechanism(s) remain largely undefined. In line with previous findings (37,38), here we demonstrate a significant bradycardia and blood pressure reduction to parenteral nicotine administration at a resulting serum concentration higher than that of a typical smoker. This response is completely abolished in AAVantiNic-treated mice, which suggests that the vaccine also abrogates nicotine-mediated physiological changes, such as the adverse cardiovascular effects associated with cigarette smoking that most likely play a role in drug reinforcement beyond the neurological chemical addiction. The ability AAVantiNic to persistently sequester nicotine en route to its brain receptors, and abrogate the physiologic effects and addictive properties (12,13,39) as well as suppress nicotine-induced locomotor behaviors through repeated challenges of nicotine at serum concentrations equivalent to or greater than that attained by chronic cigarette smoking suggests that clinical translation is appropriate. We have completed safety studies for two AAVrh.10 vectors (carrying different transgenes), one of which is being used in an ongoing clinical trial for Batten disease in children (clinicaltrials.gov, NCT01161576). Immunotherapy using the AAV vector-directed expression of monoclonal antibodies as a treatmentfor drug addiction offers a unique opportunity to address a great societal problem and unlike current therapies directed toward smoking cessation, which require daily doses for the nicotine addict and lead to non-compliance this would be circumvented with the single administration of an AAVantiNic-based therapy.
AAVantiNic vector is based on the non-human primate-derived rh.10 capsid pseudotyped with AAV2 inverted terminal repeats surrounding the anti-nicotine antibody expression cassette (Fig. 1A). The expression cassette consists of cytomegalovirus (CMV) enhancer-chicken–β-actin promoter, the anti-nicotine monoclonal NIC9D9 heavy chain coding sequence, a 4-amino-acid furin cleavage site and the 24-amino-acid self-cleaving 2A peptide, the anti-nicotine monoclonal NIC9D9 light chain coding sequence, and the rabbit β-globin polyadenylation signal (22,23,40–42). The cDNA sequence of the NIC9D9 antibody heavy chain (IgG1) and light chain (κ chain) were cloned from the mouse hybridoma NIC9D9 (42), by using RNA ligase-mediated rapid amplification of cDNA ends (GeneRacer kit; Invitrogen) with mouse immunoglobulin gene specific primers. The negative control vector AAVrh.10antiPA.Mab (referred to as AAVcontrol) encoded an irrelevant antibody directed against anthrax protective antigen.
AAVantiNic was produced by Polyfect-mediated (Qiagen) cotransfection into human embryonic kidney 293 cells (HEK 293; American Type Culture Collection) of three plasmids, pAAVαNIC9D9 (600 μg), pAAV442 (600 μg), and pAdΔF6 (1.2 mg): (1) pAAVNIC9D9 is an expression plasmid containing (5′ to 3′) the AAV2 5′-inverted terminal repeat including packaging signal (ψ), the anti-mouse NIC9D9 antibody expression cassette and the AAV2 3′-inverted terminal repeat; (2) pAAV44.2 is a packaging plasmid that provides the AAV Rep proteins derived from AAV2 needed for vector replication and the AAVrh.10 viral structural (Cap) proteins VP1, 2 and 3, which define the serotype of the produced AAV vector; and (3) pAdΔF6 is an Ad helper plasmid that provides Ad helper functions of E2, E4 and VA RNA (43). At 72 hr after transfection, the cells were harvested, a crude viral lysate was prepared by four cycles of freeze/thaw and clarified by centrifugation. AAVantiNic was purified by iodixanol gradient and QHP anion exchange chromatography. The purified AAVantiNic was concentrated with an Amicon Ultra-15 100K centrifugal filter device (Millipore) and stored in PBS, pH 7.4, −80°C. The control vector was produced by this method with pAAVantiPA.Mab substituted for pAAVantiNic. Vector genome titers were determined by quantitative TaqMan real-time PCR analysis with a cytomegalovirus promoter-specific primer–probe set (Applied Biosystems).
To assess AAVantiNic-directed expression of the monoclonal antibody in vitro, HEK 293 cells were infected with AAVantiNic at 2 × 105 genome copies per cell (or mock infected), supernatant was harvested 72 hr later and immunoglobulin was purified with protein G sepharose. Anti-nicotine antibody expression was evaluated by coomassie blue stain SDS-PAGE and Western analysis (41) with a sheep anti-mouse IgG heavy chain and light chain secondary antibody (Sigma) and enhanced chemiluminescence reagent (Amersham).
All animal studies were conducted under protocols reviewed and approved by the Weill Cornell Institutional Animal Care and Use Committee. Male C57Bl/6 mice, 4 to 6 weeks old (Taconic) were housed under pathogen-free conditions. At 7 to 9 weeks of age the mice were treated with AAVantiNic at 109, 1010 or 1011 gc by intravenous injection in 100 μl volume.
Blood was obtained by drawing 250 μl of blood from the tail vein at time 0 and at various time points, until 18 weeks. The blood samples were allowed to clot for 1 h, at 23°C, followed by 30 min, at 4°C, and then spun in an Eppendorf microcentrifuge at 10,000 rpm for 20 min to collect serum. The concentration of anti-nicotine antibody were then determined by ELISA. Wells of flat bottomed 96-well EIA/RIA plates (Corning) were coated with 100 μl of 1 mg/ml bovine serum albumin conjugated with AM1, a trans-3′-(hydroxymethyl)nicotine-derived nicotine hapten (30) (ratio of 2:1), in carbonate-buffer overnight at 4°C and then washed with 0.05% Tween 20 in PBS (PBS-Tween) and blocked with 5% dry milk in PBS for 30 min, 23°C. Serial dilutions of sera were incubated for 90 min, 23°C. The plates were washed 4 times with PBS-Tween and 100 μl of 1:2000 diluted goat anti-mouse IgG conjugated to horseradish peroxidase (Santa Cruz) in 1% dry milk in PBS, incubated for 90 min, 23°C. After 4 wash steps, peroxidase substrate (100 μl/well; Bio-Rad) was added to each well, incubated for 15 min at 23°C and the reaction was stopped with addition of 2% oxalic acid (100 μl/well). Absorbance was measured at 415 nm. Anti-nicotine antibody titers were calculated by interpolation of the log(OD)-log(dilution) with a cutoff value equal to twice the absorbance of background and converted to μg/ml based on standard curve with the NIC9D9 antibody.The NIC9D9 antibody was quantified by the bicinchoninic acid assay (Pierce Biotechnology). Antibody specificity for nicotine and metabolites was evaluated by competitive ELISA with sera from AAVantiNic-treated mice to in BSA-AM1 in the presence of increasing concentrations of nicotine and nicotine metabolites nornicotine or cotinine, from 0.1 nmol/l to 0.1 mmol/l. Affinity of the expressed antibody from sera from AAVantiNic treated mice was evaluated by a radioimmunoassay, using [3H] nicotine (10 nm) with increasing concentrations of non-labeled nicotine (1 to 300 nm) as the unlabeled competitor (44).
Naive control or AAVantiNic-treated mice were anesthetized by intraperitoneal injection of ketamine (100 mg/kg) and xylazine (10 mg/kg) 3 min before intravenous administration of 0.8 μg nicotine (equivalent to a 2 cigarette human dose) containing 1.0 μCi [3H] nicotine (11,24). One min later, mice were killed, and the brain and blood were collected separately. Brain tissue was homogenized in PBS, and 300 μl of brain homogenate and separately 50 μl of serum were added to 5 ml liquid scintillant (Ultima Gold, PerkinElmer), which was assessed for radioactivity, and the data were normalized with a standard quenching curve. For the blood compartment, nicotine was normalized to serum volume (ng/ml), and brain nicotine was normalized to brain wet weight (ng/g).
To assess whether the treatment abrogated the cardiovascular effects of nicotine, mice were implanted with radio-telemeters (Data Sciences International, PA-C10 model) 7 weeks after administration of AAVantiNic or AAVcontrol vector as described (45). One week after telemeter implantation, baseline measurements of mean arterial blood pressure and heart rate were collected for 30 min in conscious mice. Subsequently, nicotine (25.0 μg, 1.0 mg/kg) or PBS was administered subcutaneously, and cardiovascular parameters were recorded for 180 min with the DSI Dataquest A.R.T. system (Data Sciences International).
We recorded mouse locomotor behavior with infrared beam-equipped activity chambers (20 × 20 cm chamber, AccuScan Instruments) at 4 weeks after AAVantiNic treatment. Mice were allowed to habituate to the room for 1 hr before each test. Mice were placed in the chamber for 15 min to record pre-challenge behavior, then removed, injected with PBS or nicotine (12.5 μg, 0.5 mg/kg) subcutaneously and returned to the chamber for 15 min. Locomotor assays were repeated over a 3 week period for a total of 10 nicotine challenges. Chambers recorded ambulatory distance traveled and vertical movement.
Dose response and pharmacokinetic data are expressed as means ± SEM, and comparisons between groups were conducted by a two-tailed unpaired t-test. For cardiovascular studies, mean arterial blood pressure and heart rate time point data over the 210 min course were compared by two-way repeated measures ANOVA with AAVantiNic treatment as the between subjects variable and time as the within subjects variable and Bonferroni post-hoc comparison was performed at each time point. The two-way repeated measures ANOVA was used for locomotor studies with AAVantiNic treatment as the between subjects variable and nicotine challenge day as the within subjects variable. Comparisons of cumulative distance and vertical activity between groups were performed with the Kolmogorov–Smirnov test.
We thank M. Staudt and A. Krause for helpful discussions, and N. Mohamed and D.N. McCarthy for help in preparing this manuscript. Funding: These studies were supported, in part, by R01 DA025305 (SMK, KDJ subcontract) and R01 HL63887 (RLD) and TRDRP 20XT-0156m (KDJ). MH is supported, in part, by T32HL094284. JR is supported, in part, by The National Foundation for Cancer Research, and The Malcolm Hewitt Wiener Foundation. CNY is supported by an American Physiological Society postdoctoral fellowship.
Author contributions: Study concept and design: RGC, RLD, NRH,SMK,SW,JP,BPD,JBR, KDJ, MJH; Experimental procedures: BPD,JBR, OP,MJH; Data analysis: JBR,MJH,CY; Manuscript preparation: MJH,JP,JBR, NRH, SMK, RGC, KDJ; Literature search: MJH, NM,RGC,BPD.
Competing interests: The antibody gene NIC9D9 is available from The Scripps Research Institute, and an MTA must be obtained before it can be used. Weill Cornell Medical College has filed a patent based on the work described in this paper: Disrupted adenovirus-based vaccine against drugs of abuse Authors: Ronald Crystal, Bishnu De, Martin Hicks, Jonathan Rosenberg, Stephen M. Kaminsky The other authors declare that they have no competing interests.