Online Resource Supplementary Information
Richards et al -- A Randomized, Placebo-Controlled Pilot Trial of the Delta Opioid Receptor Agonist AZD2327 in Anxious Depression
Online Resource Supplementary Information
Supplementary Methods
Preclinical Data:Prenatal Stress
All animal experiments were carried out in accordance with the National Institutes of Health (NIH)Guide for the Care and Use of Laboratory Animals.Detailed methods have been previously described(Peters et al. 2011).Briefly, 10pregnant female Sprague Dawley (SD) rats (Charles River Laboratories, Wilmington, MA) arrived at gestation Day 6 (E6) and were single-housed in standard rat cages. After a six-day acclimatization period (E7–E13), the pregnant dams were assigned to either the control group (non-prenatal stress) in which the dams were left alone in the animal housing room, or the prenatal stress (PNS) group,in which the dams were exposed to a novel, variable stress paradigm during the third week of gestation (E14–E20). In this stress paradigm, the pregnant dams were exposed to a different stressor each of the seven days: (1) in the motion stressor, the rat was placed in an empty, clear, 11.5′ × 11.5′ × 7.5′ container attached to a shaker that toggled the container back and forth; the session lasted 30 minutes. (2) In the forced-swim stressor, the pregnant dam wasplaced into a thick, clear, plastic cylinder, 15′ in height, 6.5′ in diameter, about 3/4 full (~ 1.5 l) ofwater at room temperature (~ 23 °C)for 10 minutes. (3) The rat was put in a restraint apparatus for 60 minutes. (4) The wet bedding stressor involved adding 1.5–1.7 l of water to the rat's bedding in which the rat remained overnight. Cages with wet bedding were replaced with clean cages the following day. (5) The rats were kept in a dark, novel room for 36hours. (6) Pregnant dams were placed on a rack and brought outside a closed room that contained barking dogs for six hours. (7) The rat received intermittent puffs of air using a hair dryer set on cool (three five-minute sessions).
Once the pups were born (at approximately prenatal day 22), each litter was culled to four females and six males. The dams and pups were left undisturbed for three weeks, at which point the pups were weaned and group-housed on postnatal day (PND) 21. Behavioral measures were assessed in adult offspring after PND70. The effects of prenatal stress on exploratory behaviors were evaluated using the elevated plus maze (EPM) as previously described (Hudzik et al. 2011). For the forced swim test, animals were placed into water-filled cylindrical containers as described above.Latency to immobility and total immobility time were measured.Twenty rats were in the AZD2327 group, and 31 in the placebo group.
Preclinical Data: BDNF Assays
Subject and Tissue Preparation:Male SD rats were divided into the following experimental groups: 1) intraperitoneal (IP) water for 21 days (n=6); 2) IP tranylcypromine (a monoamine oxidase inhibitor (MAOI)) 7.5 mg/kg for seven days and 10 mg/kg for the next 14 days (n=12); 3) oral (PO) 1% lactic acid in water for seven days (n=6); 4) PO AZD2327, 3.0 mg/kg for seven days (n=12). The dose of tranylcypromine was chosen based on the published literature (Nibuya et al. 1995; Torregrossa et al. 2005). The dose of AZD2327 was chosen based on previous data from a learned helplessness experiment (3 mg/kg was the effective dose)(Hudzik et al. 2011). Animal weight was measured daily. Three hours after the last injection, rats were sacrificed by decapitation and samples of the prefrontal cortex, hippocampus, and plasma were collected. Plasma was isolated using proper test tubes with ethylenediaminetetraacetic acid(EDTA).After extraction, the anterior cortical region of the brain was sliced by hand; the prefrontal cortex and hippocampus were separated from each slice. Brain tissue extracts were prepared using a lysis buffer (50 mMPh=8 Tris-HCL, 500mM NaCl, 0.2% Triton-X-100=0.4ml, 0.1% Sodium azide=0.2g, 2% bovine serum albumin (BSA), 2mM EDTA, 10%Glycerol, and protease inhibitors) and centrifuged for 20 minutes at 15,000rpm at 4°C to obtain a clear lysate. To assess total brain-derived neurotrophic factor (BDNF), samples were processed with 1M HCl acidification and heated for 30minutes at 600C. Subsequently, 1M NaOHwas used for neutralization, according to the manufacturer’s protocol(Okragly and Haak-Frendscho 1997)and samples were diluted with the Block and Sample solution.
ELISA:BDNF was detected with an ELISA kit (Promega, USA). Briefly, 96-well plates were coated with anti-BDNF monoclonal antibody and incubated at 4°C for 18hours. The plates were incubated in a blocking buffer for onehour at room temperature; samples were then added. To assess total BDNF, samples were processed with 1M HCl acidification and subsequent 1M NaOH neutralization and then were diluted with the Block and Sample solution. The samples and BDNF standards were incubated for twohours, followed by washing with the appropriate buffer. Plates were incubated with anti-human BDNF polyclonal antibody at room temperature for twohours, washed, and then incubated with anti-IgG antibody conjugated to horseradish peroxidase for onehour at room temperature. Finally, the plates were incubated in peroxidase substrate and tetramethylbenzidine solution to produce color reaction. The reaction was stopped with 1M HCl; absorbance was measured at 450nm using a microplate reader.BDNF changes among each treatment group were analyzed with one-way ANOVA followed by Bonferroni’s post-hoc tests.
Clinical Studies
Pharmacokinetic, Pharmacodynamic, and Safety Evaluation
The pharmacokinetic data indicated that AZD2327 has a highly variable intersubject pharmacokinetic profile. The compound was rapidly absorbed at each dose level (reaching Cmax between 1.5 to 2.5 hours); mean Cmax values ranged from 0.77 (1 mg) to 19.8 (15 mg) ng/mL while area under the curve (AUC) ranged from 6.10 (1 mg) to 224 (15 mg) hr·ng/mL (AstraZeneca, internal communication).AZD2327 then displayed a distribution phase followed by elimination, with a mean terminal half-life between 5.39 (1 mg) to 21 hours (15 mg).
Pharmacokinetic samples to determine plasma levels of AZD2327 and its major metabolite, AZ12311418, were collected before treatment, and at 1, 3, 7, and 12 hours post-dose.Pharmacokinetic samples were also collected on Day 7 before treatment and at each outpatient visit (visits four to six).The total blood drawn for each of these assessments was 4 ml, with a total of 36 ml drawn during the entire study.In addition, blood was drawn up to five times for additional biomarker analysis (approximately 15-25 ml each time).
A linear mixed model with restricted maximum likelihood estimation and a compound symmetry covariance structure was used to examine drug levels over time.The mixed model was used to allow estimates in places with missing values.Analyses were performed separately for AZD2327 and AZ12311418.Given skewed distributions, each value was log transformed prior to analysis.Samples were collected immediately afterdrug administration as well as at 1, 3, 7, and 12 hours post-dose and on Days 7, 14, 21, and 28.The weekly values were collected one hour after drug administration.
Given the potential side effects of AZD2327 in phase I studies, additional precautions were implemented for convulsions and syncope.We excluded use of drugs or substances known to cause abnormal electrocardiogram(ECG) and electroencephalogram (EEG) activity.In addition to the baseline measures noted above, EEG recordings were also taken for three hours on Day 1 starting one hour prior to dosing and continuing until two hours post-dose.Additional EEG readings were also taken for 90 minutes on Day 28.EEGs were also used to capture potential pharmacodynamic endpoints.ECGs were also conducted pre-dose (within 30 minutes of dose) and at 30 minutes, 45 minutes, two hours, and 3.5 hours post-dose.Additional orthostatic blood pressures and pulse were taken two hours post-dose, and semi-recumbent blood pressure and pulse were taken two hours pre-dose and 30 minutes, one hour, and 3.5 hours post-dose.Serial orthostatic blood pressures were obtained throughout the inpatient and outpatient periods of the study.
EEG, VEGF, and BDNF
EEG and serum samples for BDNF and vascular endothelial growth factor (VEGF) analysis were collected as putative biomarkers of drug effects and response.Briefly, artifact-free waking EEG (30 sec from C3,4, P3,4 derivations) from supine subjects with eyes closed were collected using PolysmithAcquisition and Review software (v. 4.0.25.0; Nihon Kohden, USA)at three time points before and after drug administration:1) Day 1 [0-20minutes before drug (BL)], 2) Day 1 [1.5-2 hours after drug (D1)], and 3) Day 28 [1.5-2 hours after drug (D28)].Gamma and control band EEG power were calculated using EEGLab 10.2.5.8b and Matlab 7.8.0 (2009).
Blood samples were centrifuged at 216 gat 4°C for five minutes and stored at 80°C until assayed.To determine VEGF and BDNF concentrations, anti-VEGF and anti-BDNF sandwich ELISAkits (R&D Systems, Minneapolis, MN and Millipore, Billerica, MA, respectively) wereused accordingto the manufacturer’s instructions. The plasma was diluted with a samplebuffer and the assay performed in duplicate.In orderto create the standard curve for VEGF and BDNF levels, astandard solution was provided by the manufacturer.Streptavidinenzyme, substrate, and stop solution were added andVEGF and BDNF levels were then determined by absorbance in a spectrophotometer at 450 nm.The total amount of blood drawn during the study was less than 200ml.Experiments were carried out in duplicate and blind to clinical information.
SupplementaryResults
Preclinical Studies
The effects of AZD2327 on the behavior ofPNS rats are shown in Figure S1.In the forced swim test, IP administration of AZD2327 (0.03-3 mg/kg) significantly increased latency to immobility compared to vehicle at doses ≥ 1 mg/kg(Figure S1).Pretreatment with AZD2327 (3mg/kg, ip) 30 minutes prior to testing significantly increasedthe percentage of time spent in the open arms of the EPM (t(49)=6.120, p<0.0001(Figure S1), the total number of entries (t(49)=2.834, p<0.0067), and the percentage of entries into the open arms (t(49)=4.397, p<0.0001).AZD2327 reversed the highly anxious phenotype produced by PNS, as evidenced by increased activity into the open arms of the EPM. In this model, in which selective serotonin reuptake inhibitors (SSRIs) are ineffective(Maciag, personal observations), positive effects could be due to strong anxiolytic effects or to a combination of antidepressant and anxiolytic effects.
Like tranylcypromine, AZD2327 increased BDNF expression in hippocampus but, unlike tranylcypromine, failed to do so in frontal cortex, suggesting some differences in its potential spectrum of activity from monoamine oxidase inhibitors (Figure S2).Neither drug altered plasma BDNF.
Clinical Studies
Adverse Effects
Adverse events were recorded daily for the first week then weekly for the following three weeks.In all, 88 symptoms were rated on the Adverse Event Record as “none”, “mild”, “moderate”, or “severe” for each time period (Table 2).
After receiving the first dose of 3 mg of active AZD2327 at 10:45 AM on Day 1 of the study, a 32 year-old female patient reported dizziness while lying down at 11:15 AM.Upon standing at 12:50 PM, the patient had orthostatic hypotension (standing blood pressure: 79/54) and a brief syncopal episode.After other symptoms were largely resolved, the patient vomited 30 minutes after eating dinner and later decided to drop out of the study.The drug levels for this patient were: less than 0.5nM immediately following drug administration, 1.620 at one hour, 6.360 at three hours, 3.420 at seven hours, and 1.070 at 12 hours.The symptoms were deemed most likely a drug reaction by the treating physician.
AZD2327 Plasma levels
AZD2327 levels increased through three hours before declining slightly through 12 hours.Levels were higher at Day 7 through 28, reaching their peak at Day 14.Hamilton Depression Rating Scale (HAM-D) responders did not have significantly different AZD2327 levels than non-responders (response: F=1.52, df=1,12, p=.24, d=0.72; response by time: F=0.93, df=8,73, p=.49).Hamilton Anxiety Rating Scale (HAM-A) responders did not have significantly different AZD2327 levels than non-responders (response: F=0.75, df=1,12, p=.40, d=0.51; response by time: F=0.77, df=8,73, p=.63).
For AZ12311418, levels increased through three hours before declining slightly through 12 hours.Levels were higher at Day 7 through 28, reaching their peak at Day 28.HAM-D responders did not have significantly higher AZ12311418levels than non-responders (response: F=3.36, df=1,11, p=.09, d=1.09; response by time: F=1.38, df=8,71, p=.22).However, HAM-A responders had significantly higher AZ12311418levels than non-responders (response: F=5.99, df=1,11, p=.03, d=1.49; response by time: F=1.42, df=8,71, p=.202).The difference between responders and non-responders appeared to begin at Day 7, and lasted through the end of the study.
EEG Analysis
Patients receiving AZD2327 had significantly larger (p=0.02) gamma EEG power than those receivingplacebo, both at Day 1 post-drug and at Day 28. However, in control band (2-30Hz), EEG power did not significantly differ from placebo patients at either time point (Figure S2).While there was no significant difference in gamma EEG power between HAM-D or HAM-Aresponders and non-responders, there was a trend (p<.06) for HAM-A responders to have higher EEG gamma power than non-responders (Figure S2). Correlations in all patients between Day 1 serum AZ12311418 levels and gamma power, both elevated in HAM-A responders, were not significant.
Online ResourceFigure Legends
Figure S1. Brain-derived neurotrophic factor (BDNF) levels in the hippocampus, prefrontal cortex, and plasma of animals treated with AZD2327.Compared to controls, BDNF expression was significantly increased in (A)the hippocampus (F(2,33)=19.8, P<0.001), but not in (B) the prefrontal cortex.(C)BDNF plasma levels did not significantly differ in animals treated with AZD2327 and tranylcypromine compared to controls (F(2,33)=0.8, P>0.05); n=4 animals per group.
Figure S2: Gamma electroencephalogram (EEG) power on AZD2327.No significant difference was noted in gamma EEG power between Hamilton Depression Rating Scale (HAM-D) or Hamilton Anxiety Rating Scale (HAM-A)responders and non-responders, but there was a trend (p<.06) for HAM-A responders to have higher EEG gamma power than non-responders.
References
Hudzik TJ, Maciag C, Smith MA, Caccese R, Pietras MR, Bui KH, Coupal M, Adam L, Payza K, Griffin A, Smagin G, Song D, Swedberg MD, Brown W (2011) Preclinical pharmacology of AZD2327: a highly selective agonist of the delta-opioid receptor. J Pharmacol Exp Ther 338: 195-204.
Nibuya M, Morinobu S, Duman RS (1995) Regulation of BDNF and trkB mRNA in rat brain by chronic electroconvulsive seizure and antidepressant drug treatments. J Neurosci 15: 7539-7547.
Okragly AJ, Haak-Frendscho M (1997) An acid-treatment method for the enhanced detection of GDNF in biological samples. Exp Neurol 145: 592-6.
Peters MF, Zacco A, Gordon J, Maciag CM, Litwin LC, Thompson C, Schroeder P, Sygowski LA, Piser TM, Brugel TA (2011) Identification of short-acting kappa-opioid receptor antagonists with anxiolytic-like activity. Eur J Pharmacol 661: 27-34.
Torregrossa MM, Folk JE, Rice KC, Watson SJ, Woods JH (2005) Chronic administration of the delta opioid receptor agonist (+)BW373U86 and antidepressants on behavior in the forced swim test and BDNF mRNA expression in rats. Psychopharmacology (Berl) 183: 31-40.
1