You are on page 1of 17

Research Article

Molecular Pain
Volume 13: 1–16
! The Author(s) 2017
Pharmacology, pharmacokinetics, Reprints and permissions:
and metabolism of the DNA-decoy DOI: 10.1177/1744806917703112
AYX1 for the prevention of acute
and chronic post-surgical pain

Julien Mamet1, Scott Harris1, Michael Klukinov2,
David C Yeomans2, Renee R Donahue3, Brad K Taylor3,
Kelly Eddinger4, Tony Yaksh4 and Donald C Manning1

Background: AYX1 is an unmodified DNA-decoy designed to reduce acute post-surgical pain and its chronification with a
single intrathecal dose at the time of surgery. AYX1 inhibits the transcription factor early growth response protein 1, which is
transiently induced at the time of injury and triggers gene regulation in the dorsal root ganglia and spinal cord that leads to
long-term sensitization and pain. This work characterizes the AYX1 dose-response profile in rats and the link to AYX1
pharmacokinetics and metabolism in the cerebrospinal fluid, dorsal root ganglia, and spinal cord.
Results: The effects of ascending dose-levels of AYX1 on mechanical hypersensitivity were measured in the spared nerve
injury model of chronic pain and in a plantar incision model of acute post-surgical pain. AYX1 dose-response profile shows
that efficacy rapidly increases from a minimum effective dose of  0.5 mg to a peak maximum effective dose of  1 mg. With
further dose escalation, the efficacy paradoxically appears to decrease by  30% and then returns to full efficacy at the
maximum feasible dose of  4 mg. The reduction of efficacy is associated to doses triggering a near-saturation of AYX1
metabolism by nucleases in the cerebrospinal fluid and a paradoxical reduction of AYX1 exposure during the period of early
growth response protein 1 induction. This effect is overcome at higher doses that compensate for the effect of metabolism.
Discussion: AYX1 is a competitive antagonist of early growth response protein 1, which is consistent with the overall
increased efficacy observed as dose-levels initially escalate. Chemically, AYX1 is unprotected against degradation by nucle-
ases. The sensitivity to nucleases is reflected in a paradoxical reduction of efficacy in the dose-response curve.
Conclusions: These findings point to the importance of the nuclease environment of the cerebrospinal fluid to the research
and development of AYX1 and other intrathecal nucleotide-based therapeutics.

Chronic pain, early growth response protein 1, oligonucleotide, nucleases, cerebrospinal fluid

Date received: 31 December 2016; revised: 21 February 2017; accepted: 28 February 2017

Pain following surgery remains a major public health
issue with 80% of surgery patients suffering acute 1
Adynxx, Inc., CA, USA
pain and  10% to 50% developing chronic pain.1–4 2
Department of Anesthesia, Stanford University, CA, USA
The transition to chronic pain reflects the long-term sen- 3
Department of Physiology, University of Kentucky, KY, USA
sitization of the dorsal root ganglia (DRG) and spinal Department of Anesthesiology and Pharmacology, University of California,
cord network triggered by surgical trauma.5 Upon such San Diego, CA, USA
trauma, the transcription factor early growth response
Corresponding author:
protein 1 (EGR1) is transiently induced in the DRG Julien Mamet, Adynxx, Inc., 100 Pine Street, Suite 500, San Francisco,
and spinal cord and locally initiates genomic regulations CA, USA.
that establish long-lasting neuronal sensitization.6–10 Email:

Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-
NonCommercial 3.0 License ( which permits non-commercial use, reproduction and
distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://

gov identifier ascending dose-volume scheme similar to that of NCT02081703). Experiments 720 min following IT administration. ing IT injection of DNA-decoys as a class of molecules. 0–35 We characterized AYX1 local pharmacokinetics (PK) total) was recorded as a response.11 IT injection of chronic pain. the animal head and forepaws excluded from the study due to autotomy and euthanasia both taped to the nose mask to secure the animal and following surgery).05 mg AYX1 group were flurane using a nose mask. rats using the complementary spared nerve injury and plantar incision models of pain. and the oligonucleotide unprotected from nuclease-based codes were only revealed after the entire testing was metabolism. For the PK experiments. lumbar cerebrospinal fluid (LCSF). 10. 240. and Laboratories).. and the number of paw withdrawals (0–5 per level. a one-time administration provides up to 80% GCCCACGCATAC-30 ) or vehicle were administered reduction of mechanical hypersensitivity over controls. DRG. ical formulation (110 mg/mL) was injected in 10 identifier We characterized the AYX1 dose-response profile in NCT02081703).e.. 4. 10. A 30G needle was inserted directly . by EGR1 and specifically inhibits its activity. 60.13 The pharmacokinetic before testing. antisense strand: 50 -CTACGCCCACC briefly. and 8 g. The time interval and metabolism features driving AYX1 exposure in the between each testing level for a given rat was 3 min. tissue incision. beginning to be understood14–17 and to the best of our 26.2 mg in 20 mL.1 mg of AYX1 clin- models of complementary etiologies (e. allow the experimenter to gently apply traction to the tail cology study (no exclusion).11 AYX1 is under active clinical 2. and spinal For each study. spinal cord/vertebra. 1.11 It is delivered via a single intrathecal (IT) bolus injection IT administration around the time of surgery to reduce acute pain and prevent its chronification. 250 to 350 g were used. at which point pain was resolving in volume was 20 mL in the ADY-SNI2 and ADY-INC5 controls in our hands) and is observed across pain studies. 42 rats for the ADY-INC5 pharma.11 intrathecally under anesthesia as a percutaneous bolus The preventive effect can lasts for over a month (i. 1. 75 rats for the in vivo PK during collection. For each testing level.11 Rats were habitu- from 30 to 60 min up to at least 12 h following a nox. each tested cohort of animals included cord during that timeframe. 120. Methods Tissue sampling For AYX1 concentration measurement. Testing was conducted by a single experimenter show that the AYX1 dose-response pattern is consistent at each testing site in a blinded fashion: blinded test and with the PK and metabolism features of AYX1 as an control article vials were sent to the testing sites. The collection were carried out according to animal care protocols was facilitated by pressing the neck and applying mech- approved by the respective Institutional Animal Care anical traction to the tail while holding the head of the and Use Committees of each testing site and were animal and/or inclining the animal’s position to an designed to minimize the amount of animals utilized. or DNA-decoy. the von Frey hair knowledge. Prior studies in the spinal The spared nerve injury and plantar incision models were cord have shown that EGR1 induction is detectable performed as described elsewhere. AYX1 clinical trials (clinicaltrial.2 Molecular Pain AYX1 is an unmodified double-stranded 23-bp and metabolism work (no exclusion). clinicaltrial. was applied five times consecutively around the incision. anesthetized under 1% to 3% iso- cle group and one rat in the 1. 4 rats for the spinal deoxyoligonucleotide. with a sequence homogenate metabolism work (no exclusion). Calibrated von Frey hairs were applied properties of single strand oligonucleotides are just with the following pseudo-random pattern: 6. AYX1 Behavioral testing must be present in the DRG-spinal cord network at suf- ficient levels to inhibit EGR1. 360. and 4 rats that mimics the natural genomic DNA sequence bound for the histology presented here (no exclusion).12. completed. These combined data controls. LCSF was col- Animals lected percutaneously once at the L3/4 vertebral level in Sprague Dawley rats (Harlan industry or Charles River separate groups of rats at 30. and 3.85 mg in 35 mL to follow an development (ADYX-004 trial. the at the L4/5 or L5/6 vertebral level at the time of surgery longest tested period in the rat spared nerve injury model as previously described in literature. bone.g. no information is publicly available regard. ated to cages with mesh wire floors for at least 1 h ious or sensitizing stimulus. AYX1 efficacy in animal AYX1 (sense strand: 50 -GTATGCGTGGGCGGTG pain models has been described in a prior publication: GGCGTAG-30 . approximate 45 angle as follow: rats were placed over The following number of rats were used: 35 rats for a stainless steel bowl to curve the lumbar portion of the the ADY-SNI2 pharmacology study (one rat in the vehi. To be effective. or nerve injury).

5 mg up to the maximum 37 C. No impact of anesthesia was observed on mated by log-linear regression of 3þ points in the ter- LCSF collection. WI.Mamet et al. fixed in efficacy of each tested dose-level was measured as a per- 4% paraformaldehyde at 4 C. an increase of effi- Histology cacy was expressed as an increase in magnitude of pain AYX1 was conjugated on the sense strand to an suppression compared to controls and as an increase in ALEXA488 tag (50 ALEXA Fluor 488. mals regardless of whether pain was due to last a few days or several weeks (see literature. Data are pre- phenol-chloroform and alcohol precipitation and hybri.8. Across dose-levels.11 Figure 1(a) and (b)). and homogenized with a dounce hom. Thirty minutes following the injection. using capillary electrophoresis with laser-induced fluor- escence and back-calculated to a matrix standard curve. spinal erature. For the EXOIII nuclease experiment. a 20 mL IT injection at the time of an injury.4. sented throughout the article as Mean followed by dized to a fluorescent-labeled probe specific to AYX1. La Jolla Hybridization (CGEH). hypersensitivity in rat models of acute and chronic pain was first described in Mamet et al. pooled into 500 mL of ice-cold the five studies described in Mamet et al. AUC was estimated by was then collected using a sterile insulin syringe and the trapezoid method. www. meta-analysis dose-responses across studies and pain . To build up AYX1 dose-response in given washed in ice cold 1X PBS. cord and DRG were collected in a dark room. Animals were then sacrificed. Values AYX1 dose-response profile reported as ‘‘total AYX1’’ include full length plus short. at maximal effi- and stopped with 20 mM ethylene diamine tetraacetic cacy. 30 to 50 mL of LCSF were collected minal phase. catalogue # M1811) of injury and administration and lasted. cut in pieces. and spinal cord from the T12 to T13 vertebral level (cor- responding to the L1–L2 spinal level)18 to the sacral tip were collected.0a Software. The assay allows detecting and quantifying N-1 to N-6 Results shortmer metabolites or shorter versions of AYX1 pro- duced by nuclease-based nucleotide removal. cryoprotected in sucrose. seven independent stu- Spinal cord homogenate incubation dies were conducted to measure the effect of ascending Rats were anesthetized. Tissue Frey responses compared to controls from the time of IT sections of 12 mm were made and observed under fluor. the effect of sub-optimal IDTDNA. including weighed. until hypersensitivity was resolving in control ani- acid. terminal slope and T1/2 were esti- needle. 3 into the IT space to allow the LCSF to rise naturally by PK analysis differential pressure between the LCSF and the air out. injection and injury until the last day of testing. NHS Ester. The studies were performed at several US laboratories ogenizer. exsanguinated with ice cold 1X doses of AYX1 on mechanical hypersensitivity following PBS. AYX1-homogenate mixes were incubated at and covered doses from 0. Dose- AYX1 analytical assay dependent relationships were analyzed using analysis of AYX1 was quantified by Capillary Gel Electrophoresis. the washed in saline. followed by a T-Welsh analysis (DRG were pooled for each animal). and DRG of regression R2 for the slope estimation was <0.1). and volume of IT injection. Samples were extracted using California USA. washed in 1X phosphate buffered saline Statistical analysis (PBS). and mounted for cryostat sectioning.graphpad. embedded. IL) and was injected IT as described doses did not last until pain resolved in controls (see lit- above. incu. pH 8). Reactions were stopped with proteinase K feasible dose of  4 mg based on AYX1 solubility limit (100 mg/mL). The LCSF that rose in the needle hub ment methods (no modeling). for which we nuclease buffer (100 mM tris-HCL and 1 mM magnesium provide here a meta-analysis (Table 1).e. The escent microscope. variance (GraphPad Prism 7.11 During AYX1 preclinical development. the duration of that effect.11 Figure 1(a) and (b) and Table 1). acetate. and frozen for subsequent AYX1 tissue uptake A non-parametric Student T-test. either a standard error (SE) or a standard error to the The probe-analyte species were separated and detected mean. frozen on centage of reduction of area-under-the curve of total von dry ice. The AYX1 efficacy profile for reducing mechanical mer metabolites.. analysis for uneven variance correction. was used to ana- lyze individual conditions and whole data distribution between experimental conditions (Excel 14. spinal cords harvested. AYX1 and shortmers were analyzed by non-compart- side the needle. The AYX1 efficacy profile bations were made using the reaction buffer provided was similar across studies: efficacy appeared within 24 h with the enzyme (Promega. and T1/2 were not reported if the coefficient per animal. i.

001 0. Efficacy normalized on vehicle and on maximum efficacy within a given study: 0 ¼ no difference from vehicle.12/4.05 <0. Both normalized and non.32 1. Dose-response across AYX1 pharmacology studies.88 4.80 0.56 0.83 1.40 0. To allow for a sensitive analysis. of AZ Mamet et al. Examining individual complementary pain models and separate testing labora- studies.2: 0. the overall dose-response pattern and exponential fit sured for that study.48 1. AZ: Arizona.39 0.05 0.65 1. p Spared nerve SNI1 Stanford Uni.80 0.2: 0.00 0.58 veh/2.. KY: Kentucky.75 1.00 SNI2 Uni.11 0.52 0.60 1.20 0.07 0.001 1.00 SNI3 Uni. ANOVA was used to assess the strength of the dose-response in each study including the vehicle groups.04 1.25 0. 2014 10 days 0. 2014 72 h 0.13 0.40 0. (c)). Analysis of Pain model (ADY-) Testing site Study description duration dose (mg) vehicle efficacy variance.42 INC5 Uni.67 0.33 0.39 0.052 1.80 0.40 0. data were produced a lower efficacy than predicted.12 0. 2014 4 weeks 0.47 0. (Figure 1(b) and (c) and Table 1).14 <0. IA: Iowa.00 SNI4 Uni.004 0.70 0.57 0. Uni: University. of IA Mamet et al.24 0. -SNI4.08 2. of KY Current publication 48 h 0.40 0.67 0.85 1.68 0.35 all groups :0. random effect.21 0.40 0.42 1.84 0..40 veh/1.80 0.8/4.70 0.05 0.84 2.36 0..2: 0. 1 ¼ 100% suppression of mechanical hypersensitivity compared to vehicle.73 0.04 0. in four out of the seven studies cacy profile observed across those studies and pain (ADY-SNI2.72 veh/1.8/4. -INC5. of KY Mamet et al. 2014 72 h 1.22 4.70 0.12 0. Individual dose-efficacy levels for each pharmacology study of the AYX1 dose-response meta-analysis.80 0. Mamet et al.49 model 2.03 incisional 1.40 0.12 0. models were justified by the similarity of the AYX1 effi.001 injury model 1. -DOD1). Mamet et al.23 0.29 1.20 0.38 Plantar INC2 Uni.00 Note: veh: vehicle. Interestingly. of KY Current publication 10 days 0.12 0.00 DOD1 Stanford Uni.28 0. This feature is not only observed in indi- suggesting an exponential association relationship vidual studies but also conserved in the meta-analysis of followed by a plateau of efficacy (Figure 1(b) and the combined seven studies which delineates the .78 1.0001 1.18 0. Efficacy Study Efficacy normed name Study AYX1 normed on on max.83 1.00 2.4 Molecular Pain Table 1.44 0..55 1.12/2. 2014 4 weeks 0.84 0. The occurrence of normalized efficacy values for each tested dose and this observation across more than half of the studies in study are presented in Table 1. the magnitude of AYX1 efficacy rapidly tories supports a real feature over the likelihood of a increases with dose until maximum efficacy is reached. one dose-level models.56 0.33 0.50 0.40 0.23 0. an outlier in normalized for each study on the maximum efficacy mea.56 0.00 1.40 0..56 0.84 0.25 0.55 1.90 2.

values are presented as mean þ SEM. 1. and CSF.05 mg (white squares).4 0.02.Mamet et al.8 (white circles).5 mg. we characterized AYX1 metab- the dose is further increased. 1. efficacy) 1.4 mg 6 * * * * * 0. and 4.6 2.6 0. Total responses (number of paw withdrawals) to repetitive von Frey stimulation in animals treated with vehicle (black circles).0 4.2 0. 0. data distribution over the testing period: p < 0.02. spinal cord. Connecting curves are presented as dotted lines.8 mg 1.8 0.40. (a) Effect of ascending AYX1 dose-levels on the development of pain measured as mechanical hypersensitivity in the spared nerve injury model of chronic neuropathic pain in the ADY-SNI2 study.12 (black triangles).8 0.8 1.97 and from 0.2 4.4 mg (white circles).2 mg (white squares) of AYX1 groups are displayed as representative responses observed during the study. excluding the outlier dose shifts the coefficient of regression R2 of the exponential fit from 0. with the exception of the homogenates using concentrations from 0.2 4. levels around 1 mg and as pattern of dose-response.2 0. (c) and (d) Dose-response patterns observed in the ADY-SNI2 and ADY-INC5 studies.4 3. average. the vehicle (black circles).0 0. data distribution over the testing period.44 to 0.12. following pattern: efficacy rapidly increases from a min.8 AYX1 dose (mg) AYX1 dose (mg) Figure 1. values are presented as mean þ SEM.7 mg 4 * * 4 * * * 2.8 0.001 for all tested doses. For clarity in light of the amount of tested groups. excluding the outlier dose (circled). efficacy reduces by 30% in olism and PK in the lumbar DRG. To understand this imum efficacy dose of 0.85 in the ADY-INC5 study. T-test followed by a T-Welsh analysis: *p ¼ 0. limited range of efficacy reduction.0 0. vehicle versus 4.6 2. The magnitude of effects of all tested doses is presented in Table 1.1 mg 12 * ** 8 4.0 0.05. and full efficacy can be recovered for higher doses up to the maximum feasible dose of 4 mg AYX1 metabolism in spinal cord homogenates (Figure 2). 2. respectively.0 4. 5 (a) total (b) total von Frey responses von Frey responses 24 vehicle 18 20 16 14 vehicle 16 12 * 2. n ¼ 4 to 5 per group. (b) Effect of ascending AYX1 dose-levels on the development of pain measured as mechanical hypersensitivity in the plantar incisional model of acute pain in the ADY-INC5 study. The effect of AYX1 0. or 2.0 0.6 0. efficacy) (normed on max. T-test.2 mg 8 * 1.8 mg 2 * * * 0 * 0 0 1 2 3 4 5 6 7 8 9 10 0 24 48 (c) days post-surgery (d) hours post-surgery AYX1 efficacy AYX1 efficacy (normed on max.8 mg (white lozenges) AYX1 are presented. 1.75 to 0.0 0.8 1. Vehicle or AYX1 were administered once IT at the time of surgery. 1.07. 1. The general shape of the AYX1 dose-pattern is consistent with its mechanism of action as a competitive AYX1 metabolism was characterized in fresh spinal cord antagonist of EGR1 activity.7 mg (black squares).001 fold to . **p ¼ 0.2 mg: p ¼ 0.56.84. and 4.4 0.80.20 mg was tested against vehicle. The magnitude of effect for each dose-level normalized on the maximum efficacy measured within each study is presented (black triangles). For the ADY-SNI2 study.0 1. 2. different from vehicle at a given time-point: *p < 0. Vehicle or AYX1 were administered once IT at the time of surgery.4 3. Data are fitted with an exponential association fit (dashed line).05 mg * 10 1. n ¼ 6 per group. 0.

85 mg of AYX1. and -DOD1 (Mamet et al. A systematic series of potential dose- response fits were tested. INC5. 2014).0 0.6 Molecular Pain (a) AYX1 efficacy (normed on max. or 3.2 3.8 0. and data fitted with a polynomial equation of the fourth order are presented as a dashed line.1. Data are fitted with a sigmoidal function (dotted line). including one-phase exponential or hyperbolic fits.0 2..6 4.106 2.106 0.4 2.106 3.2 1.2.0 4. The connecting curve is presented with a dotted line.8 1.35 for the polynomial fit above and equivalent or lower for various association equations. 2. efficacy) 1. n ¼ 5 per dose-level.8 3.. 2014).8 1.. it was selected as the closest representation of AYX1 dose-response over a hyperbolic or exponential plateauing curve.106 4.2 1. 2014) studies testing AYX1 efficacy in either the spared nerve injury of neuropathic pain or in the incisional model of post-surgical pain. lumbar CSF) 5.2 3. (a) Meta-analysis of AYX1 dose-response profile including results from the ADY-SNI2.2-4 (Mamet et al. Data are normalized internally for each study against the maximum observed efficacy relative to vehicle-treated animals and presented as mean  SEM. (b) Concentration of AYX1 shortmers in the LCSF 30 min following injection of 1. The highest coefficients of regression were R2  0.6 0.4 0.0 2.0 0.4 0. .106 1.0 0.2 AYX1 (mg) (b) AYX1 shortmer metabolites (ng/mL.4 2.2 1.6 4.2 0.6 2. In absence of a robust coefficient of regression and considering that the polynomial curve fits the meta-analysis and the results of the majority of individual studies. SN1.8 3.6 2. results are presented as mean  SEM for each dose-level.0 AYX1 (mg) Figure 2. -INC25 (Mamet et al.4 -0.4 0.

2 1.01 30 5365 3111 3.08 51.22.730 11. 30.85/35 106.100 1.370 1.350 Total 2. This phenomenon EGR1 induction is detectable within 30 to 60 min is common for oligonucleotides19–21 and has been following a noxious stimulation.Mamet et al. Individual AYX1 and shortmer values are presented in Tables 3 and 4.36 207.2/20 24.6 30 836 110.1/10 39.770 3. 5 to 10 fold the maximum spinal cord concentration of metabolism similar to homogenates.24 161.700 69.2/20 6469 NR 30 230.2/20 46.1/10 4539 148. defined as a plateau of metabolism rate despite an tion of 25 mg AYX1/mg. DRG. LCSF: lumbar cerebrospinal fluid.140 Note: SE: standard error.1/10 1430 205. NR: not reported.7 3.000 3.416. the mg/mL units in the LCSF listed in the Cmax.86 30 3632 2936 3.37 30 14. When appropriate.46 30 4546 2747 3.1 60 2743 2847 shortmers 2.2 1. where it can compete with the newly . Mean. and 8921 ng/ rate of AYX1 metabolism. and tissue. SE. N ¼ 4 to 5 rat per dose.2/20 10300 NR 30 463.85/35 13720 NR 30 803.74 30 1784 868. or min.03 30 1305 609.63 NR 30 651 441.668 NR 30 414 243 AYX1 2. DRG.060 9761 Lumbar spinal cord 1.6 60 370.33 30 4364 3299 3.300 309.1/10 3109 10. respectively.4 120 10.300 86. Saturation of metab.1 39.85/35 35.85/35 73. Further.760 12. increase in AYX1 concentration.48 NR 30 879 65.13 The histologic attributed to either a masking of oligonucleotides by visualization of a fluorescent conjugate of AYX1 in endogenous protein and/or by autoretardation due to the lumbar spinal cord and DRG 30 min following an end-product inhibition. and/or Median columns correspond to mg/mg of tissue units for the DRG and spinal cord tissues.85/35 26.000 46.3 30 3035 1906 Total 2. and spinal cord pharmacokinetic parameters following a single IT administration of ascending dose-levels of AYX1. time-point. appeared as a linear function of 60 min with the EXOIII compared to 92.2/20 27.309.72 108. and spinal pharmacokinetic parameters of AYX1 and shortmers metabolites.190 1.85/35 9328 9.800 193. AYX1 rate of metabolism remained AYX1 lumbar exposure and in vivo metabolism slowed down over time (Figure 3(a)).32 135. 19. DRG: dorsal root ganglia.064.800 194. Comparison of LCSF.1 30 101.85/35 82.200 shortmers 2.12.258. measured after 5.1/10 35.4 AYX1 2.2 30 2384 2348 shortmers 2.68 30 159.1/10 7.1/10 26.57 3.1/10 10.65 54.94 104. AYX1 (mg)/vol Cmax T1/2 AUC (0-T) AUC 60–720 min Tissue Variable (uL) (mg/mL) (min) Tmax (min) (min*mg/mL) (min*mg/mL) LCSF 1.2/20 3830 NR 30 225. see Table 2). AUC.23 Incubation of AYX1 IT injection shows that AYX1 is already present in cell with a recombinant EXOIII nuclease showed a decrease nuclei (Figure 4). 7 Table 2. degradation rates were measured at 5.810 AYX1 2.4 55.72 74. The specific protein binding: 86.900 3.73 189.66 101. AYX1 concentration (Figure 3(a)).000 Lumbar DRG 1.750 10. suggesting AYX1 observed in vivo following a maximum feasible IT dose metabolism slows down via autoretardation versus non- (10–20 mg AYX1/mg of spinal cord.2 30 12.7 30 260.46 47.592 30 432. 17.1 60 3157 2604 Total 2.1/10 37.56 52. and 60 min of incubation.400 156.2/20 71.19. LCSF. was not attained even at the highest tested concentration of 100 mg/mg of spinal cord.09 NR 120 12.2/20 84.100 1.85/35 89. 30.85/35 4688 686. and 9518 ng/min with the equivalent homogenate condi- olism.2/20 57.000 141.400 502.

561 ng/ml (2. . LCSF. and 3.366. and time. and spinal cord was olites was measured using a CGEH method. LCSF: cerebrospinal fluid.0 0. 2. 360.1 mg dose) 0.1 mg dose) 0. The presence of full- illustrated in the spinal cord. Values are presented as mean þ SEM. Linear regression for each data set are presented. for at least 720 min following the initial stimulation.4 1.2 7.8 0.8 0.401 ng/mg (3.772 ng/ml (1.6 2.2 mg dose) 0. 30.0 0.8 0. and the using doses across the broad range of AYX1 efficacious corresponding individual data-points are shown in dose-levels: 1.313 ng/ml (3.6 892 ng/mg (1.2 mg dose) 97. 120.101 ng/mg (2.85 mg. data are presented as mean þ SEM.6 104.4 2.0 FL N-1 N-2 N-3 N-4/6 FL N-1 N-2 N-3 N-4/6 Figure 3.2 0.12.4 60. The rate of AYX1 metabolism is presented as the amount of full-length AYX1 degraded per minute as a function of AYX1 concentration introduced in the homogenates.0 amount 1. intact AYX1 and N-1 to N-6 shortmer metab- metabolism in the LCSF. 240.2 0. FL ¼ full-length AYX1.0 0. DRG.228 ng/mg (2. EGR1 induction is known to continue spinal cord were collected at 30. (a) Rate of AYX1 metabolism as a function of AYX1 concentration in fresh spinal cord homogenates.0 0.2 amount 1.8 0.1 mg dose) 0. compartment.85 mg dose) 0.2 amount 1. DRG: dorsal root ganglia.228 ng/mg (3.1 mg dose) 25.0 amount 1.85 mg dose) 0. 30 (square). Each curve represents the metabolite pattern observed for a dose-level.106.2.558 ng/mg (1. N-x ¼ metabolite species. n ¼ 2 to 4 per condition. The amounts of N-4 to N-6 metabolites were pooled together due to the frequent fusion of their corresponding peaks in the CGHE assay.8 Molecular Pain (c) LCSF 60 min (a) AYX1 metabolism rate (μg/min) 450 400 350 300 250 200 LCSF 240 min 150 100 50 0 0 10 20 30 40 50 60 70 80 90 100 AYX1 concentration (μg/mg of spinal cord homogenate) LCSF 60 min DRG 60 min (b) Relative 1. (c) Illustration of electropherograms from the CGEH analytical method from the LCSF at 60 and 240 min following injection. (b) Metabolic patterns of full-length AYX1 and individual N-1 to N-6 metabolites measured with the CGHE assay.2 0.0 FL N-1 N-2 N-3 N-4/6 FL N-1 N-2 N-3 N-4/6 LCSF 240 min DRG 240 min Relative 1.99 for 5. as and 720 min post-injection.210 ng/ml (3.6 0. coefficient of linear regression R2 is 0. For each analyzed sample. Right arrows show full-length AYX1 and the left arrows the extremity of the analyzable area.4 0. The result- studied during the overall period of EGR1 induction ing PK parameters are shown in Table 2.2 Relative 1. 60.338 ng/ml (1. and Tables 3 and 4. induced EGR1. and 60 (triangle) min.2 mg dose) 0.85 mg dose) 1. the relative amount of AYX1 or individual metabolite species was normalized on the species found in largest amount. DRG.13 AYX1 exposure and length.2 Relative 1.727321 ng/ml (2. Data are presented for each measured time-point: 5 (circle).1. and 60 min. Peaks in between the two arrows represent AYX1 shortmers from N-1 to N-6.2 mg dose) 16.85 mg dose) 86. and the corresponding concentration of total AYX1 is listed in the legend.

and of intact AYX1. This is illustrated by the estimated T1/2. initial AYX1 concentra- increased with doses and were observed at 30 min.7 mg of AYX1 conjugated to an ALEX488 tag (green). Top panel scale bar ¼ 100 mm. The maximal concentrations of AYX1 in the LCSF following injection. Areas identified in the top panels with white rectangles are magnified in the bottom panels. and AYX1 was still pre- injection. 9 Figure 4. the tions in the DRG and spinal cord after injection were first tested time-point and initial AYX1 concentrations approximately three to four orders of magnitude lower dropped by approximately 50-fold 120 min following than the LCSF concentrations. as measured by shortmer concentrations. limit. This observation is consistent with independent experiments showing the presence of decoys in the nucleus and/or cytoplasm of cells following various administration methods47–49. defined as N-minus metabolite species and their concentrations largely dropping over the 120 min relative amounts. shortmer concentrations diminished and 60 min period. DRG. values which were longer overall for the shortmers versus AYX1 (Table 2). One example of AYX1-positive cell nuclei is pointed out for the DRG and spinal cord with a vertical arrow. of metabolism in the LCSF during the first 30 to Subsequently. AYX1 presence in lumbar DRG (a) and spinal cord (b) cells was observed 30 min following an IT injection of 1. Table 4. respectively (Table 4 and Figure 2(b)). Specificity of AYX1 metabolite patterns The AYX1 exposure pattern in the DRG and spinal The comparison of oligonucleotide shortmer metabolite cord was similar to that observed in the LCSF. bottom panel scale bar ¼ 25 mm. This is The total exposure of AYX1 increased with doses illustrated by the similar and maximal concentrations when calculated from the first measured time-point of of AYX1 shortmer metabolites measured at 30 min for 30 min until the last time-point of 720 min. spinal cord during the 60 min to 720 min timeframe Maximal levels of shortmers further remained at (Table 2). However. with patterns. Tissue auto-fluorescence was controlled using non-injected rats (c). provides important clues as to the . leveled off while AYX1 concentrations continued to while the low dose only produced a low metabolism rate drop in the low ng/mL range or below detection (Figure 2(b) and Tables 3 and 4). and one example is pointed out in both tissues by a horizontal arrow. A few cells show AYX1 present in the cytoplasm.Mamet et al. the mid and high AYX1 doses: 4393 and 3829 mg/ the exposure at the mid dose was approximately half mL of shortmers in presence of 6468 and 9328 mg/mL that of the low dose exposure in the LCSF. and n ¼ 2 rats per condition. to a period of elevated AYX1 metabolism. This observation correlates to the fact that 60 min for the high dose: 4688 mg/mL in presence the mid dose of AYX1 engages a near-saturating rate of 2678 mg/mL of intact AYX1. However. This timeframe of rapid decrease corresponds sent at 1 to 30 ng/mg of tissue after 720 min (Table 3).

85/35 360 5 0 0 0 3.85/35 30 5 26.884383132 22.004046355 0.1/10 30 5 10.90767435 30.9501099 1464.01096018 0.77196 255.10 Molecular Pain Table 3.517023367 0.2/20 30 5 24.85/35 30 5 9328.005950316 3.1/10 240 4 0.1/10 360 4 0 0 0 1.0126 0 1.836208772 11.5476851 2908.003778745 0.1/10 360 4 0.1/10 240 4 0 0 0 1.7795 3. Tissue AYX1 (mg)/vol (mL) Time (min) N Mean (mg/mL) SE (mg/mL) Median (mg/mL) LCSF 1.06257237 346.1/10 120 4 41.036514832 2.6482 1.2146364 653.25506945 114.1/10 720 4 0. DRG.587725 19.004594702 3.1/10 120 4 0.005724716 0.63809148 3.1/10 240 4 0.180800152 0.2/20 360 5 0 0 0 2.2/20 120 4 0.069117621 1.009110736 0.60045935 1.1/10 720 4 0 0 0 2.3966 259.02144515 2.2/20 120 4 0.1/10 360 4 0 0 0 1.463704766 2.568782709 6.3905 2.1/10 720 4 0.2/20 30 5 27.000818852 0.47877045 6.009575 0.6169484 3023.85/35 720 3 0.280427501 1.20848275 0.217325199 0.1/10 60 4 1384. and spinal AYX1 concentrations for each AYX1 dose-level and at each measured time-point.003963512 0.0126 0.85/35 240 4 0.2/20 60 4 0.2/20 30 5 6468.63232418 1.2/20 720 4 0 0 0 3.85/35 360 4 0 0 0 3.2/20 240 4 0 0 0 (continued) .089622801 7.57879113 3.013328716 2.3156 460.85/35 60 4 2678.020483504 0.12447694 2.3316 2.4782 1351.002656131 1.006678209 1.92505 2.93761384 2.92784122 34.755526589 1.1/10 60 4 7.74928 1180.1/10 30 5 7.2/20 120 4 118.85/35 60 4 7.2/20 240 4 0.034634755 0.818188664 2.017725 253.2/20 720 3 0.908725 19. LCSF.1875 3.71598671 3.9908 3.2/20 60 4 720.124114177 22.532875 59.29521 6613.77076212 2.284382244 1.85/35 720 4 0 0 0 DRG 1.1/10 30 5 3108.022397921 2.11685 8870.2/20 360 4 0 0 0 2.2/20 60 4 1.001809653 0 1.5843 1.000818852 0 2.85/35 120 4 2.002292878 0.548 1.007363318 0.01096018 0 Spinal cord 1.85/35 120 4 360.005836822 1.001809653 0.667657941 1.1/10 120 4 0.163416839 0.018598467 7.004430156 0.00765 3.017266029 4.2/20 240 3 0 0 0 2.055901619 1.006860178 0.002597334 0.300554058 0.001988417 0.85/35 240 4 0 0 0 3.588309906 1.920219497 1.149641096 4.462844897 1.1/10 60 4 3.753794535 0.

5283 3.6167 2.20149944 95.2/20 120 4 100.6962055 3382.1/10 60 4 722.338137454 1.88145 3.2/20 60 4 2007.1/10 720 4 0.7562697 997.85/35 120 4 116.2/20 360 5 0. Continued Tissue AYX1 (mg)/vol (mL) Time (min) N Mean (mg/mL) SE (mg/mL) Median (mg/mL) 2.85/35 360 5 0 0 0 3. Table 4.732933554 1.2/20 720 3 0.46068157 54.31426526 7.85/35 720 4 0 0 0 DRG 1.2657 1.2008 14.828536182 0.85/35 120 4 2.4167534 637.3306593 4636.1/10 60 4 35.1/10 30 5 1430.66970954 14.2/20 30 5 57.00011489 16.585075 5.645387362 9.915116254 1.99725 518.85/35 360 4 91.68192968 0.08379438 1.00295173 0.1/10 120 4 2.65144157 4.68185 5.99885 2.001746952 0.85/35 30 5 56.076550053 0.85/35 240 4 0 0 0 3.001916241 0.807271349 3.09900073 2.0878 3.299263821 3.2/20 30 5 3829.27906899 2.2/20 240 4 0.1/10 360 4 51.1/10 360 4 0.073519484 2.2/20 360 5 0.058065924 0.66248569 24.8372 3.2/20 120 4 12.85/35 60 4 4687. Individual AYX1 concentration values over time following a single IT administration of ascending dose-levels of AYX1.897248086 0.2/20 360 4 71.2/20 240 3 60.1989776 1944.1/10 240 4 0.030442652 34.8175 10. 11 Table 3.9528 2.2484032 4279.85/35 240 4 97.22149446 5.002808648 3.01635 11.30448375 3.85/35 30 5 4393.2/20 60 4 30.85/35 30 5 35.841458069 100.1/10 120 4 115.2/20 720 4 0 0 0 3.55538 420.365275 22.05737843 113.342176242 2.90782032 92. Tissue AYX1 (mg)/vol (mL) Time (min) N Mean (mg/mL) SE (mg/mL) Median (mg/mL) LCSF 1.85/35 720 3 0.1/10 720 4 0 0 0 2.853425 6. When appropriate.898455803 101.265015581 1.31611509 45.85/35 60 4 10.22032106 2.36510465 1.2347 1.884681372 62.08351415 4.491545206 1.072593559 19.64975 3.375275 138.689466667 2.1/10 30 5 22.249071141 0.4557 2.03844 228.32217034 3.69677409 1.324925 5.001916241 0 Note: LCSF: lumbar cerebrospinal fluid. the mg/mL units in the LCSF section correspond to mg/mg of tissue units for the DRG and spinal cord tissues sections.658890882 1. DRG: dorsal root ganglia.1/10 240 4 104.026227013 0. LCSF.75351275 81.163014365 33.2/20 720 3 0 0 0 3. DRG.9664 1.56093333 2.939187865 0.47892 457. and spinal shortmers concentrations for each AYX1 dose-level and at each measured time-point of AYX1.169358434 0.63588399 3.6510067 119.012175677 0 2.1125 1.55215 2.243203582 57.Mamet et al.61231783 (continued) .486917073 0.303575 675.53445 1.012175677 0.41389241 11.

45905314 13.29570307 23.1/10 30 5 26.022585718 1.07303081 42. a reduction of efficacy followed by a recov- same pattern of metabolites is now observed: a low rela. injury (four studies) models of acute and chronic pain.82376256 2. with a minimum efficacy of 32%.20 The quantification of N-1 to Discussion N-6 AYX1 metabolites in the lumbar DRG. Its pharmacology for reducing mech- level of total AYX1 concentrations while LCSF patterns anical hypersensitivity was characterized in multiple stu- varied from those with concentrations.06033945 62.294443044 1.077281886 0.099909523 2.069128587 1.54314857 46.1/10 240 4 0.2/20 30 5 46. Continued Tissue AYX1 (mg)/vol (mL) Time (min) N Mean (mg/mL) SE (mg/mL) Median (mg/mL) 3.2/20 120 4 6.85/35 30 5 70.1/10 720 4 0. and CSF at 60 and 240 min highlights specific pat- AYX1 dose-response profile terns across overlapping concentrations of intact AYX1 AYX1 is a DNA-decoy administered once by the IT plus shortmer metabolites. Individual AYX1 shortmer metabolites concentration values over time following a single IT administration of ascending dose-levels of AYX1.78052733 2.123270579 17.61584808 9.243865375 25. a low relative presence of AYX1 with increasing and a maximum efficacy of 80% (Table 1). DRG and route to prevent pain and its chronification following spinal cord patterns were similar at each measured surgery or trauma.393005601 1.85/35 240 4 9. DRG: dorsal root ganglia. the average maximum effect of vidual metabolites of shorter length is observed in the AYX1 measured as a reduction of area-under-the-curve LCSF at 60 min with elevated concentrations of total during the entire study period compared to controls was AYX1.234685714 2.85/35 120 4 89. At the same time-point in the DRG and spinal 65%  7.68070379 19.29563893 3.82752407 3.94835537 3.1/10 120 4 4. DRG and LCSF dies in the incisional (three studies) and spared nerve patterns are illustrated Figure 3(b) and (c).729129583 0.38740579 2.933503571 3.052341893 0 Spinal cord 1.0491 3.2/20 60 4 33.85/35 360 5 1. or total AYX1.627302761 5.224379814 0.144297059 3. At 240 min.96980246 92.306484997 0.186606863 0.85/35 240 4 1. nuclease environments. When appropriate.117330072 1.2/20 360 5 0.66395684 21.182547597 0. ery phase are observed in the majority of the individual tive presence of AYX1 with elevated N-2 and N-3 studies (Table 1) as well as in their meta-analyses.85/35 60 4 46.435677476 0. The asso- amounts of shorter length metabolites is observed for ciated dose-response pattern initially shows a rapid concentrations of total AYX1 that are approximately increase of efficacy from the minimally efficacious dose two orders of magnitude lower.85/35 60 4 32.615999167 2. spinal cord.07231587 57.94819926 9. Across those studies.052341893 0. but the feasible dose.85/35 120 4 73. cord.9% (SE).294673064 32. A high rela.363497293 3.68212152 1.04840256 3.85150049 1.896101046 0. total up to a maximum efficacy with only a two-fold dose AYX1 concentrations are still two orders of magnitude increase.084782621 0.1/10 60 4 21.134327203 0.85/35 720 3 0.08738192 14. The metabolites.354115198 0.2/20 720 3 0 0 0 3.872365 1.059888095 3.27566035 1.125419048 8. As doses further increase toward the maximum higher in the LCSF compared to local tissues.85/35 720 3 0 0 0 Note: LCSF: lumbar cerebrospinal fluid.035981222 0. the mg/mL units in the LCSF section correspond to mg/mg of tissue units for the DRG and spinal cord tissues sections.00893874 26.43102964 19. average peak dose found to reduce efficacy across studies .912439333 0.12 Molecular Pain Table 4. tive presence of AYX1 with decreasing amounts of indi.038355346 0.72917591 1.59521031 3.1/10 360 4 0.086580168 1.2/20 240 4 0.85/35 360 5 0.

metal ions.44 mg (SE). The number of copies of a transcription factor efficacy in the AYX1 dose-response pattern occurs for varies per cell from 5000 to 4. a phenomenon in saturation of AYX1 metabolism while the 1 mg dose enzymology referring to enzymes which kinetics only triggered a slow. Note that in one study (ADY-SNI4). esis of the nucleases in the CSF. responds slowly to rapid changes of substrate AYX1 concentration from the 2 mg dose (720 mg/ concentrations.g. At that time-point...13 An increase of local AYX1 exposure dur. Based on an average mea- metabolism during the initial period following dosing. peak dose on the high end of that range at 2 mg..52 mg  0.79 mg (i. nucleases metabolizes AYX1 as a function of concentra- tides. Also.00033 and there are . weight of a DRG the ‘‘excess’’ of AYX1 over the 2 mg dose appeared to used in the PK work described above). By direct efficacious doses were in the low ng/mg of tissue range. Concurrently. can estimate the number of AYX1 molecules per DRG ectly transferred to the DRG and spinal cord where cell relative to the estimated number of EGR1 molecules AYX1 exerts its pharmacological activity. During most of the EGR1 induction period.5 to 2 mg AYX1. This may To be effective. and is rapidly LCSF across animals and studies. saturable. DRG.11 ferent environments (e. the polyno.000. diffusion from the LCSF to the local tissues. AYX1 dose. The CSF nuclease environment influences the AYX1 tion. Of note. Our data suggest that it is a viscosity at this concentration (i. Indeed.28–30 The data from this doses 1 mg. hence with water). sured DRG weight of 1. based on AYX1 solubility limit and volume of IT injec.. pH. AYX1 must inhibit EGR1 activity in reflect the significant influence of the composition of dif- the DRG and/or spinal cord during EGR1 induction. a reduction of effi- Cerebrospinal fluid (CSF) nuclease environment cacy was also observed at the maximum feasible dose.g. cells per neuron per DRG and a negligible extracellular Those combined data suggest that the reduction in space).28.e. AYX1 con- ciated with a similar and basal metabolism rate that centrations in the DRG and spinal cord after injection of maintained their relative exposures onward. with dose-response curves that often span within a tion.e. This is similar to antisense oligonucleo. and spinal cord the CSF. 3–4 mg/mL).24 ites in the CSF and tissues at 240 min while total AYX1 concentrations are 2 order of magnitude higher sug- gests similar nucleases metabolize AYX1 but that in AYX1 exposure in the LCSF.000 neurons32 and 10 satellite 60 min. The exposure of 2 mg AYX1 is two fold work provide a possible link between those apparently lower in the LCSF and four times lower in the DRG and contradictory observations: the CSF contains active spinal cord than that of 1 mg dose during that same nucleases but their activity was not readily detected in period.25–27 lation.Mamet et al. which could lead to a variable distribution in the a low affinity of CSF nucleases for AYX1. suggesting that both doses were now asso. nuclease Km. one positioning of exposure of the two doses appeared dir.6 million AYX1 molecules exposure remained above that of the 1 mg dose after per cell (assuming 15.31 mL) was lower in the LCSF at 60 min compared to the 1 mg dose (1400 mg/mL).1 mg/mL of CSF) compared to efficacy observed in the dose-response curve. AYX1 (i. 0. the relative To understand the meaning of such concentrations. For the and EGR1 binding sites in the genome for an AYX1 higher 4 mg dose. This likely results in a lower opposition to EGR1 prior studies likely due the low oligonucleotide concen- action and could explain the paradoxical reduction of trations used (e. This result could be explained by AYX1 increased dose-response profile. near. which also triggers a saturating concentration 10 ng/mg. the presence of the same pattern of metabol- only 10-fold factor. Prior studies applying oligonucleotides to the CSF ing that timeframe is anticipated to increase efficacy up reported an absence or a low nuclease activity in the to an exposure level that exceeds what is required for a CSF while the analysis of the CSF proteome confirms complete inhibition of EGR1..e. 10 ng AYX1/mg offset an abrupt impact of the metabolism and its of DRG corresponds to 4. their affinity for AYX1 is lower. 40 times higher than multi-nuclease system with an elevated Km. the range of doses triggering near-saturation of metab- mial fit of the dose-response meta-analysis places that olism in the LCSF upon injection: 1.30 Lastly. the detection of several AYX1 response and its landmark features occur over a narrow metabolic patterns suggests that an increasing range of range of doses.12. which seems to occur for the presence of several nucleases. and salts) on from 30 to 60 min up to 12 h after a noxious stimu. basal metabolism. the fact that the The paradoxical reduction in exposure appears to be metabolism remains saturated at 60 min for the high driven by the level of nuclease activity engaged in the dose in presence of concentrations of full-length AYX1 LCSF during the first 30 to 60 min following injection: lower than metabolite concentrations suggests an hyster- the 2 mg dose was sufficient to trigger a rapid. 13 was 1. AYX1 tissue concentrations the concentrations of shortmers were 100 mg/mL for both doses. Consequently.

We show that AYX1 efficacy increases 1. Johansen A. range can be in excess of the number of genomic EGR1 This hypothesis would require experimental confirm- binding sites and potential EGR1 molecules. Since oligonucleotide 3.46 it is possible 1618–1625. EGR1 expression profile and AYX1 PK sup. glial. 367: high oligonucleotide concentrations. at AYX1 site of action can be reduced and associated to 2. Mehta SS. interest with respect to the research. The micro. Michael Klukinov. thermal or mechanical noxious sti. The authors thank Giuseppina Iacono at Charles River muli. Pain 2012. and Tony Yaksh are consultants AYX1 likely enters these cells via endocytic pathways.12. Brad Taylor. or immune cells to carry out complemen. further supporting a reduction of efficacy.34 Thus. in the DRG and spinal cord appears specific to neurons rather than glial cells as observed after peripheral nerve Acknowledgment injury. inflammation. it is assumed that AYX1 enters neurons but Declaration of Conflicting Interests an entry in non-neural cells cannot be excluded. Jensen TS and Woolf CJ. Donald Manning. DRG to modulate pain signals in neuropathic pain situ.38 While macrophages Jerry Moore at Pacific Biodevelopment for the AYX1 PK ana- and other immune cells penetrate the spinal cord and lysis. AYX1 is a DNA-decoy drug candidate that inhibits the transcription factor EGR1 in the DRG-spinal cord net- References work at the time of injury to reduce acute pain and its chronification. and electrically induced-long term potenti. and its chemistry (an neurons. however. Finally. Dr. For doses that trigger a near-satura. et al. Persistent postsurgical transport into cells is saturable and can be inhibited by pain: risk factors and prevention. unmodified oligonucleotide sensitive to nuclease tary functions.35–37 This expression pattern directly ment. and Dr. Kehlet H. authorship.14 Molecular Pain 1000 to 1500 EGR1 binding sites per genome. DRG.45 Funding The author(s) disclosed receipt of the following financial sup- Conclusion port for the research. Tony Yaksh and Kelly Eddinger guided pharmacology study designs. Author Contributions Altogether. and spinal cord samples. authorship. David rons in the DRG and spinal cord rather than an action in Yeomans. that AYX1 penetrates cells rapidly. esti. and spinal cord). experience: results from a national survey suggest post- operative pain continues to be undermanaged. 97: 534–540. the authors are grateful to Dr. metabolism could also be at play. ation. and Scott Harris designed port AYX1 focused action within EGR1-expressing neu. and/or publication of this article: This work was funded by Adynxx. analysis of the LCSF. Persistent a lowering of efficacy. Postoperative pain overall with dose. Julien Mamet.43 In terms of uptake mechanism. David Yeomans. cellular uptake while the level of full-length AYX1 is mated amount of AYX1 molecules per cell within that already low. post-injury. This The author(s) declared the following potential conflicts of latter possibility is illustrated with the finding that anti. Renee Donahue. . Inc. these data show that AYX1 pharma- cology is directly tied to its mechanism of action (inhibiting EGR1 over a period of time following AYX1. 153: 1390–1396. Inc. Additional mechanisms related to postsurgical pain in a general population: prevalence and the excess of metabolites associated with near-saturating predictors in the Tromso study. and/or publi- cation of this article: Julien Mamet. All cells nuclei 30 min following an IT administration shows authors read and approved the final manuscript. and monitored this work. pharmacology experiments. In a nociceptive context. term neuronal sensitization. and neurons trauma).12. William K.44. EGR1.13.39–42 which is past the time of residence of AYX1 in the LCSF and local neuronal tissues. which are common as well as saturable mechanisms of entry for oligonucleotides. Lancet 2006. et al. Eric Marcusson ations. Montreal for performing the AYX1 pharmacokinetic experi- ation. Chen C. local exposure Analg 2003. Altogether. While the type of cells taking up AYX1 cannot be determined from this experiment. Michael Klukinov also performed scopic visualization of an AYX1-fluorescent conjugate in the spinal cord homogenate experiments. Apfelbaum JL. Nielsen CS. that the excess of rapidly produced AYX1 metabolites AYX1 concentrations in the ng/mg range represent bio. and Brad Taylor performed other cell types that can also express EGR1. Donald Manning. Jim Timmins at Helix Diagnostics for the CGEH correlates to EGR1 core function of inducer of long.45. this happens significantly after one or more days for his input to the article. Romundstad L. and/or stockholders of Adynxx. DRG. and sense oligonucleotides can penetrate both neurons and Scott Harris are employees and stockholder of Adynxx. glia in the spinal cord. Anesth tion of metabolism in the LCSF. Schmidt for his overall guidance and to Dr. route of administration (local exposure in the EGR1 can be expressed in a variety of cells including LCSF. associated with the mid range doses competitively inhibit logically active concentrations as the corresponding. EGR1 induction degradation).

physiology of inflammatory versus neuropathic pain. is expressed in activated microglia of mice and nonhuman primates. FOS. 29. Natelson BH. Fos and Krox. Hydrolysis of chains carrying a teins in spinal neurons following noxious transsynaptic monoesterified phosphate on carbon 5’. Smisek DL. 142: 57–61. Boca Raton: CRC Press. 277: the nervous system. Differential of antisense oligonucleotides in rat liver homogenates. Dev Cell 2011. 10. et al. Leah J. Single intrathecal human DNase I and its hyperactive variants. Library of gene expression by Jun. Pfenning AR. Ensor E. 4: 35. Klukinov M. proteome of normal human cerebrospinal fluid. Neurochem Int 1997. Egr transcription factors in tide analogs in mice. dren with spinal muscular atrophy. The initial reaction 2005. BMC Neurosci 2007. The effect of divalent cat- of Egr1 (zif/268) to persistent inflammatory pain. Stability. 31. et al. products produced from covalently closed circular DNA. Kinetic aspects of regulation of metabolic pro- 14. and KROX-24 pro. Boston: proteins. 8: 20. Vadakkan KI. Suzuki R. Ko SW. Modulating gene expression by antisense scription factors in the mammalian nervous system: control oligonucleotides to understand neural functioning. and disposition of intraventricularly adminis- serotonergic controls regulate inflammation-induced mech. Graham MJ. 2010. but not immediate-early gene ified survival of motor neuron splicing oligonucleotide in encoded proteins. Kluwer Academic. 86: 32. Chavany C. Studies of the specificity of and spatial distribution of JUN. p. Campbell VW and Jackson DA. Fiallos-Estrada C. Chun SJ. Darras BT. Rahman OI. et al. et al. 31: 477–510. et al. 43: fluids. scription factor CREB. Brar-Rai A. 2008. et al. J Biol Chem 1970. Opin Anaesthesiol 2011. Kendall G. tome. stimulation. 10: e0123604. et al. a pain. on c-Fos or Zif/268 expression in the rat spinal dorsal 20. 155: 322–333. Mamet J. the rat. NGFI-A (Egr-1) and NGFI-B (nur 77) in adult rat 18. Xu Q and Yaksh TL. phosphorothioate oligonucleotide targeting ribonucleotide 5. Vanecko S and Laskowski M. 1995. 24. 27. 9. prevents acute and chronic pain after incisional. The vertebral level of origin of dorsal root ganglion neurons. Pharmacokinetic properties of several novel oligonucleo- 6. Zuckerman JE. Crooke RM. Swoboda KJ. et al. first-in-man study. Establishing the diate early gene protein expression in the spinal cord fol. 245: 5788–5799. Selective contribution 25. 21: 611–626. quantitative continua. 8. randomised. Schmid W. Chun S. 171. A comparative a phase 1. 1994. 90: 4665–4669. Yaksh TL. Brain Res Mol Brain Res spinal nerves in the rat. Rigo F. 19. J effects of NMDA and AMPA/KA receptor antagonists Pharmacol Exp Ther 2000. Herdegen T and Leah JD. Drug Metab Dispos 2008. David W. Geselowitz D. deoxyribonuclease I. Kinetics of phos. J Pharmacol Exp Ther 1996. et al. PloS One lowing conditioning stimulation of the sciatic nerve in rats. Inducible and constitutive tran. 25: 3615–3620. Metabolism 37. 17. 13. Mol Pain Acad Sci U S A 1993. Specific temporal 23. Belenky A. Terayama R.Mamet et al. application within the central nervous system. Whitesell L. Pan CQ and Lazarus RA. Local and descend. Schwartz R and Barth AL. horn induced by noxious thermal or mechanical stimula- phorothioate oligonucleotide metabolism in biological tion. inflamma. Srinivasan D. Lavand’homme P. Schmalbruch H. et al. patients with SOD1 familial amyotrophic lateral sclerosis: 34. Delivery strategies for antisense oligonucleotide tory. Chiriboga CA. Protein Sci administration of the transcription factor decoy AYX1 1999. Lancet Neurol genomics approach to identifying the plasticity transcrip- 2013. 36: 2227–2233. from a phase 1 study of nusinersen (ISIS-SMNRx) in chil. Neurology 2016. or formalin injection. PloS One 2015. The hysteretic enzyme concept. Winer L. et al. Proc Natl phorylation in the rat superficial dorsal horn. 12: 435–442. reductase. Time course of imme. JAMA Neurol 2013. Dai G. ing circuits regulate long-term potentiation and zif268 26. J Biol Chem 1961. 24: 400–407. J Comp Neurol 1991. Single-strand-specific nucleases. Panja D. Schutzer SE. tered oligodeoxynucleotides: implications for therapeutic anical sensitivity and methyl-CpG-binding protein 2 phos. 16. Herdegen T. and CREB/ATF of Congress Cataloging-in-Publication Data. Gilar M. Martin MJ. Gelderd JB and Chopin SF. J Biol Chem 1980. 219: 315–322. III. Ao H. Neurosci Lett 1992. Nerve growth spinal fluid as a pharmacodynamic marker for antisense factor induces expression of immediate-early genes oligonucleotide therapy. Wei X. Curr Opin Anaesthesiol 2011. McCarthy MM. Frieden C. Pharmacology of a 35. et al. 6: 12–20. Anat Rec 1977. expression in spinal neurons. J Pain ions on the mode of action of DNase I. Geranton SM. et al. Liu Z. Enzyme kinetics of GTI-2040. lumbar spinal cord following sciatic nerve transection in 350: 46–55. Rygh LJ. Rahman W. Beckmann AM and Wilce PA. J Pharmacol Exp Ther 2014. 15 4. or neuropathic injury. Brain Res Brain Res Rev 1998. Pestronk A. Bittins M. Crooke ST. 188: 45–47. 24: FEMS Microbiol Rev 2003. Pain 2014. 5: e10980. 24: 545–550. 1998. A brief comparison of the patho. Ikeda T. 70: 201–207. Miller TM. Biggin MD. Tochiki KK. SOD1 in cerebral 36. Nucleic Acids Res 1997. Kovary K. 28. Anat Rec 1987. Fratto V. et al. 26: 457–491. 28: 370–490. Curr 22. Liu T. Akhtar S. Bojovic O. Results cesses. Desai NA and Shankar V. 7. The number of neurons in dorsal root 890–897. Norris DA. Ca2þ-dependent activity of 11. 12. 25: 73–79. 255: 3726–3735. 33. 15. The tran- central nervous system delivered 2’-O-methoxyethyl-mod. Herdegen T. 292: 140–149. 8: 1780–1788. Neurosci Res 2002. An antisense oligo. ganglia L4-L6 of the rat. et al. et al. therapeutics. 236: 3312–3316. 30. Graham MJ. . Eur J Neurosci 2006. 313: 178–191. 923–937. Descending clearance. 389–399. et al. 761–772. The progression from acute to chronic 21. Animal transcription networks as highly con- nucleotide against SOD1 delivered intrathecally for nected. et al. et al.

et al. Contribution of macro. Brain Res 2011. Miller TM.16 Molecular Pain 38. Detailed characterization oligonucleotide accumulation in hepatocytes. et al. et al. 16: 734–740. Sci U S A 1989. Errington ML. in mice. et al. French PJ. Kim CF and Moalem-Taylor G. oligonucleotide transport into living cells. Piirainen S. Eur J Immunol 2008. Nucleic of neuro-immune responses following neuropathic injury Acids Res 2011. 58: 107–117. CNS-infiltrating CD4þ T lympho. 48. and activators of transcription 6 (STAT6)-binding site 43. Akeda K. Sumi K. A requirement 44. Brain Behav tion of a cis element ‘decoy’ against signal transducers Immun 2016. 33: 15095–15108. 116: 2290–2296. 1405: 95–108. Nishimura A. Chappell A. Yamanaka K. Proc Natl Acad cytes contribute to murine spinal nerve transection. Human Gene Ther 2005. et al. Loke SL. nucleotide transfer facilitated by a polymer-coated stent. Vascular oligo- 42. Kwon MJ. 86: 3474–3478. Stein CA. glial and macrophage activation traits determine the differ. Bioconjug Chem 2012. Zhang XH. Gene Ther 2004. of NF-kappaB decoy oligodeoxynucleotide suppresses pul- phages to enhanced regenerative capacity of dorsal root monary metastasis by murine osteosarcoma. Vincent TM. Jones MW. J atopic dermatitis mouse model. Radke PW. Characterization of 39. 47. Ther 2011. Antisense ameliorates IgE-mediated late-phase reaction in an oligonucleotide therapy for neurodegenerative disease. 4: nucleotides. In vivo transfec- minocycline in chronic neuropathic pain. Ming X and Nakagawa O. Transfection 41. Spinal versus brain micro. 38: 46. Nat Neurosci 2001. . Kivela A. Smith RA. 1753–1762. J Neurosci 2013. Cao L and DeLeo JA. Li Z. Wu MH. et al. 18: 250–259. Griesenbach U. Cellular uptake for the immediate early gene Zif268 in the expression of and intracellular trafficking of antisense and siRNA oligo- late LTP and long-term memories. ential neuroinflammatory responses and analgesic effect of 49. 23: 147–157. Mechanisms of 448–458. et al. Wei H. single-stranded phosphorothioate modified antisense 40. induced neuropathic pain. Juliano RL. Shin H. Kim J. Matsubara T. Cancer Gene ganglia sensory neurons by conditioning injury. Yokozeki H. et al. 39: 4795–4807. Koller E. 289–296. 11: Clin Invest 2006. et al. 45.

Under the CCAL.BioMed Central publishes under the Creative Commons Attribution License (CCAL). . authors retain copyright to the article but users are allowed to download. as long as the original work is properly cited. distribute and /or copy articles in BioMed Central journals. reprint.