TAT-P₄-(C5)₂ (YGRKKRRQRRR-PEG₄-(HWLKV)₂) as HCl salt was ordered from WuXI AppTec (Shanghai, China) with > 95% purity, validated by MS and UPLC.

Unless otherwise specified, wild type male C57BL/6NRj mice (Janvier) were used. Animals were allowed at least 7 days of habituation before the start of experiments, which was initiated when the mice were 8 weeks old. Mice were group-housed in a temperature-controlled room, maintained on a 12/12 h light/dark cycle (lights on at 6 A.M.) with free access to standard rodent chow and water. Animal experiments were performed in accordance with guidelines of the Danish Animal Experimentation Inspectorate (permission number 2016-15-0201-00976) in a fully AAALAC-accredited facility under the supervision of the local animal welfare committee.

An injury was induced by injection using an insulin needle (0.3 ml BD Micro-Fine) of 50 μl undiluted Complete Freund’s Adjuvant (CFA; F5881, Sigma) unilaterally in the intraplantar surface of the right hind paw, leaving the contralateral left hind paw as an internal control of the pain threshold of the animal. Intraplantar, as well as intrathecal injections were performed while the animal was under isoflurane anesthesia (2%) for maximum of 60 s.

Following a minimum of 60 min habituation to the room, the animals were placed in PVC plastic boxes (11.5 cm × 14 cm) on a wire mesh and allowed a minimum of 20 min habituation to the equipment. The mechanical pain threshold of the animals was determined by von Frey measurements (Bioseb, France) on both hind paws. The von Frey filaments used were in the range of 0.04–2 g (g = gram-forces; 0.04, 0.07, 0.16, 0.4, 0.6, 1.0, 1.4, 2.0) and applied to the frontocentral plantar surface of the hind paws. Each filament was applied five times (with adequate resting periods between each application) and the withdrawal threshold was determined as the von Frey filament eliciting at least three out of five positive trials in two consecutive filaments. A positive trial was defined as sudden paw withdrawal, flinching, and/or paw licking induced by a von Frey filament. The baseline mechanical threshold of the animals was measured before pain-induction, and pain-relieving effect of TAT-P₄-(C5)₂ was measured on day 2 after pain-induction by either intrathecal (i.t.) or intraplantar (i.pl.) injections of either 20 μM or 200 μM TAT-P₄-(C5)₂, 0.9% saline (B.Braun, Germany) or 100 μg/paw morphine hydrochloride (Copenhagen University Hospital Pharmacy, Denmark) all in a volume of 7 μl.

Animals were habituated to the experimental room for a minimum of 1 h before the initiation of the experiment. At the initiation of the experiment, the mice were injected intraperitoneally (i.p.) with 10 μmol/kg TAT-P₄-(C5)₂ (pH adjusted to approximately 7 with 5 M NaOH) or 0.9% saline (B.Braun, Germany) at a volume of 10 μl/g and left in their home cage for 60 min. Mice were then placed in an open field (40 cm × 40 cm × 80 cm) for 60 min with a video camera above, recording their movements.

Initial perception of the drugs was measured by single-exposure place preference experiments performed in an elongated three-compartment apparatus (67.5 cm × 24 cm) with a biased design of different floor textures and wall patterns and a neutral zone in the middle. We have previously shown that mice of both genders exhibit a strong bias towards the striped compartment and that a single intraperitoneal exposure of the psychostimulants cocaine and/or amphetamine in the gray compartment is sufficient to change the preference towards that compartment (Runegaard et al., 2017, 2019). For further detail, we refer to Runegaard et al. (2017). Experiments lasted 3 days with exposure sessions on days 1 and 2 and a preference test on day 3. On each day, animals were moved from the housing facility to the experimental room and allowed to acclimatize for at least 60 min before initiation of the experiment. Drug (TAT-P₄-(C5)₂/cocaine/gabapentin) was paired with the compartment with holes in the floor and light gray walls, known to be the least preferred compartment (Runegaard et al., 2017, 2019). All mice from a cage were tested at the same time, but not all were given the same treatment. Mice were weighed and injected intraperitoneally with saline, cocaine (10 mg/kg), gabapentin (30 mg/kg) or TAT-P₄-(C5)₂ (10 μmol/kg) in a 10 μl/g body weight volume. Depending on the experiment, mice were either placed immediately in the conditioning apparatus after injection or left in their home cage for 60 min before being placed in the conditioning apparatus. The conditioning protocol was counterbalanced so that each animal received one drug injection and one saline injection on either day 1 or day 2 (with the control group receiving saline injections on both days). For the preference test, the plexiglass^® partitions between the three compartments were removed and the mice were placed in the neutral zone and allowed to freely explore the apparatus for 20 min. The preference test was recorded, and videos were analyzed with video tracking software (Ethovision XT 13, Noldus) to determine the place preference of the animals. Drug-induced place preference was determined by comparing time spent in the drug-paired compartment between groups (drug vs. saline).

In the long-term retention experiment, mice were 8–12 weeks old at initiation of the experiment and in the reversal learning experiment all mice were 8 weeks of age at the beginning of the experiment. On each day, the animals were moved from the housing facility to the experimental room and allowed to acclimatize for at least 60 min before initiation of the experiment. All drugs used (0.9% saline or 10 μmol/kg TAT-P₄-(C5)₂) were injected intraperitoneally in a 10 μl/g body weight volume. The experiment was divided into Barnes maze habituation, spatial acquisition (training), and probe trials. During habituation to the Barnes maze, animals were placed under a cylinder in the middle of the Barnes maze for 10 s and then gently guided to the escape box and left there for 2 min. During spatial acquisition (training), each mouse was placed under a cylinder at the middle of the Barnes maze for 10 s followed by up to 3 min of free exploration of the Barnes maze. Once the animal entered the escape cage, it was left there for 1 min before being transferred back to its home cage for 15–20 min. This training was performed four times per day per mouse. During probe trials, the reference memory of the animal was tested by placing each mouse under a cylinder at the middle of the Barnes maze for 10 s followed by 90 s of free exploration of the Barnes maze with no escape cage.

All data were analyzed using GraphPad Prism 9.1.1. and presented as mean ±SEM with the significance level set to p < 0.05. For locomotion, single exposure place preference, and memory experiments, Ethovision XT 13 (Noldus) was used to track the distance traveled by each animal. For within group analysis (i.e., von Frey experiments) two-way ANOVA followed by Dunnett’s post hoc test was used. For between group analysis of multiple time points (i.e., locomotion experiments and training sessions in Barnes maze experiments) two-way ANOVA was used with no post hoc test due to lack of treatment-significance in the ANOVA test. When analyzing histograms of only two groups (i.e., saline vs. TAT-P₄-(C5)₂ in the learning and memory experiments and locomotion experiment) unpaired t-tests were used when a Gaussian distribution was observed. Otherwise, the unparametric Mann–Whitney test was used. Drug-induced place preference was determined by comparing time spent in the drug-paired compartment between groups (saline vs. drug) with pairwise comparison using a one-way ANOVA followed by Sidak post hoc test, if significant ANOVA. The experimenter was blinded to treatment throughout all experiments.

Previously, we have shown that TAT-P₄-(C5)₂ provides full relief of mechanical allodynia in the spared nerve injury (SNI) model of neuropathic pain following i.t. administration, but not following i.pl. administration (Christensen et al., 2020). In the present study, we tested the effect of TAT-P₄-(C5)₂ in the Complete Freund’s adjuvants (CFA)-induced model of inflammatory pain following both i.t. (Figure 1B) or i.pl. (Figure 1C) administration. On day 0, the mechanical paw withdrawal threshold (PWT) was established (PWT of all mice was between 1 and 2 g forces) and the mice were randomly assigned into two (Figure 1B) or four (Figure 1C) different groups. After baseline testing, all animals were injected with 50 μl CFA into the plantar surface of their right hind paw and returned to their home cages (Figure 1A). On day 2, the PWT of the animals was measured and consistent with previous studies (Atianjoh et al., 2010; Aoki et al., 2014), i.pl. injection of CFA led to mechanical allodynia of the ipsilateral paw, with no effect on the contralateral paw; i.e., the gram forces needed to induce paw withdrawal of the ipsilateral hind paw, was significantly lower after CFA injection compared to baseline (Figure 1B; Saline; p < 0.0001, TAT-P₄-(C5)₂; p < 0.0001 ; Figure 1C; Saline; p < 0.0001, 20 μM TAT-P₄-(C5)₂; p < 0.0001, 200 μM TAT-P₄-(C5)₂; p < 0.0001, morphine; p < 0.0001, two-way ANOVA followed by Dunnett post hoc test of baseline vs. day 2), which is a behavioral indication of mechanical allodynia. On day 2 we assessed the effect of a single dose of TAT-P₄-(C5)₂ administered i.t. under anesthesia (2% isoflurane). Interestingly, 20 μM TAT-P₄-(C5)₂ induced a significant increase in the mechanical pain threshold of the injured paw up to 5 h after injection compared to before injection with no effect on PWT of the contralateral paw or of saline injection (TAT-P₄-(C5)_{2_ipsi}; p_{1 h} = 0.0434, p_{5 h} = 0.0033; Figure 1B). Next, we tested the peripheral effects of two different concentrations of TAT-P₄-(C5)₂ by injecting it straight into the inflamed paw and measuring the PWT of both hind paws 1, 5, and 24 h after injection (Figure 1C). As a positive control, we injected 100 μg/paw morphine (Rodrigues and Duarte, 2000), and as a negative control, we injected 0.9% saline solution. Both doses of TAT-P₄-(C5)₂ revealed significant reversal of the CFA-induced mechanical allodynia at 1 h post administration, with the highest dose showing relief of mechanical allodynia at 5 h post injection with no effect on the contralateral paw at any time point of either concentration (20 μM TAT-P₄-(C5)_{2_ipsi}; p_{1 h} = 0.0021, 200 μM TAT-P₄-(C5)_{2_ipsi}; p_{1 h} < 0.0001, p_{5 h} = 0.0002; Figure 1C). The positive control (morphine), showed relief of mechanical allodynia at 1 h post injection with no effect of the contralateral paw, and the negative control (saline), showed no effect at any time point on either paw (100 μg/paw morphine_{_ipsi}; p_{1 h} = 0.0011; Figure 1C). Importantly, since we previously demonstrated that i.t. injection but not i.pl. injections of 20 μM nor 200 μM TAT-P₄-(C5)₂ significantly relieved mechanical allodynia in the SNI model, we can firmly conclude that the action of the i.pl. injection of TAT-P₄-(C5)₂ in the CFA model is indeed peripheral.

PICK1 plays an important role in AMPAR related synaptic transmission as well as dopamine homeostasis (Madsen et al., 2005; Xu and Xia, 2006; Jensen et al., 2018). Therefore, we tested if TAT-P₄-(C5)₂ affected general locomotion in a classic open field test (Figure 2). Since we know the pain-relieving effects of TAT-P₄-(C5)₂ to be efficient at one hour after injection (Christensen et al., 2020; Figure 1), we tested the locomotion of the animals at this time point. Animals were injected i.p. with either 0.9% saline or 10 μmol/kg TAT-P₄-(C5)₂ and placed in their home cage for 1 h before being placed in the open field box and their locomotion was monitored for a further 60 min (Figure 2). Although a tendency was seen, locomotion was not reduced (Figure 2B). Also, the total distance traveled during 60 min did not show any difference between treatment groups (Figure 2C).

Due to the high abuse liability of current treatments for chronic pain, and in particular opioids (Smith et al., 2016; Gostin et al., 2017), we wanted to assess the abuse liability of TAT-P₄-(C5)₂. The initial sensitivity to the rewarding properties of drugs is believed to be an important endophenotype in relation to the vulnerability to addiction (Lambert et al., 2006). The experiment was performed in a three-compartment apparatus with a striped and gray compartment separated by a neutral white zone (Figure 3A). Here, we injected the mice with either the psychostimulant cocaine, our peptide TAT-P₄-(C5)₂, or gabapentin (a first-line treatment for chronic pain) and investigated their ability to induce place-preference in this single-exposure set-up (Figure 3A). As expected, cocaine led to a significant increase in time spent in the paired compartment compared to the saline group (Figure 3B; p = 0.02) during the preference test on day 3. TAT-P₄-(C5)₂ had no effect on the preference (Figure 1B; p = 0.85), but surprisingly gabapentin showed aversive effects decreasing the preference for the paired compartment even further (Figure 3B; p = 0.049). Since the pharmacokinetics of TAT-P₄-(C5)₂ is likely slower than that of cocaine, we repeated the experiment with an hour’s delay between drug injection and placing the animals in the conditioning compartments (Figure 3C). Neither TAT-P₄-(C5)₂ nor gabapentin showed any effect on place preference in this setup (Figure 1D).

PICK1 is required for hippocampal long-term potentiation (LTP) and long-term depression (LTD), cerebellar LTD, Ca²⁺-permeable AMPAR plasticity, and mGluR LTD in the perirhinal cortex (Xia et al., 1999; Gardner et al., 2005; Liu and Cull-Candy, 2005; Steinberg et al., 2006; Terashima et al., 2008; Clem and Huganir, 2010; Thorsen et al., 2010; Volk et al., 2010).

Consequently, we tested spatial learning and memory performance of mice following intraperitoneal administration of 10 μmol/kg TAT-P₄-(C5)₂ (Figures 4, 5) using a Barnes maze (Barnes, 1979; Bach et al., 1995). We used a standard design with a circular platform top of 92 cm in diameter with 20 holes equally spaced around the perimeter edge and a detachable escape cage under one of the 20 holes (Figures 4A, 5A).

In the first set-up, we investigated whether TAT-P₄-(C5)₂ affects long-term memory retention (Figure 4). The mice were divided into two groups and went through 4 days of training, learning the location of the escape cage (Figure 4C). Both groups showed a significant decrease in latency to enter the escape cage over the four days, reflecting their learning and memory of the escape cage location (F_{time(2.316,60.22)} = 107.8, p < 0.0001, F_{subject(26,78)} = 3.756, p < 0.0001, two-way ANOVA), with no difference between groups. On day 5, the mice were injected with TAT-P₄-(C5)₂, and a probe trial was performed (Figures 4D–H). Again, no difference was observed between groups neither in relation to nose pokes (Figure 4D), total distance moved (Figure 4E), mean speed of the animals (Figure 4F), time spent in target zone (Figure 4G), nor in their latency to reach the target hole (Figure 4H). On day 12, their long-term retention was assessed through a second probe trial, and again no difference was observed between the two groups on any of the investigated parameters; nose pokes (Figure 4I), total distance moved (Figure 4J), mean speed of the animals (Figure 4K), time spent in target zone (Figure 4L), or their latency to reach the target hole (Figure 4M).

In the second set-up, we investigated whether TAT-P₄-(C5)₂ affects reversal learning (Figure 5) by rotating the location of the escape cage 90 degrees counter-clockwise after the first probe trial, followed by three additional training sessions and ending with a second probe trial (Figure 5B). This time, the mice were injected with TAT-P₄-(C5)₂ on days 6, 7 and 8 before initiation of the training session of those days (Figure 5B).

The mice were divided into two groups and went through 4 days of training, learning the initial location of the escape cage (Figure 5C). Both groups showed a significant decrease in latency to enter the escape cage over the 4 days, reflecting their learning and memory of the escape cage location (F_{time(1.799,37.78)} = 55.11, p < 0.0001, F_{subject(21,63)} = 2.493, p = 0.0028), with no difference between treatment groups. On day 5, the probe trial was performed (Figures 5E–H). As expected, no difference was observed between groups, neither in relation to latency to reach the target hole (Figure 5E), time spent in target zone (Figure 5F), total distance moved (Figure 5G), or the mean speed of the animals (Figure 5H). After rotating the escape cage 90 degrees counter-clockwise, the mice underwent daily injections of TAT-P₄-(C5)₂ followed by new training sessions, for three consecutive days. During the training sessions, TAT-P₄-(C5)₂ did not affect the reversal learning of the mice compared to the saline group (Figure 5D). Both groups showed a significant decrease in latency to enter the rotated escape cage over the 3 days, reflecting their ability to learn and memorize the new escape cage location (F_{time(1.462,30.70)} = 28.09, p < 0.0001, F_{subject(21,42)} = 4.743, p < 0.0001), with no difference between groups. On day 9, the second probe trial was performed (Figures 5I–L). Again, no difference was observed between groups neither in relation to latency to reach the target hole (Figure 5I), time spent in the target zone (Figure 5J), total distance moved (Figure 5K), or the mean speed of the animals (Figure 5L). Since the animals were injected with TAT-P₄-(C5)₂ for 3 days in a row, and PICK1 is involved in metabolism by affecting insulin and growth hormone storage and release (Cao et al., 2013; Holst et al., 2013; Herlo et al., 2018; Li et al., 2018), we monitored their weight (Figure 5M). Weight measurements were performed in the morning, before initiating the experiment on that day, and no effect on either time nor treatment over the duration of the experiment was found.

These results show that TAT-P₄-(C5)₂ can be administered systemically to alleviate both peripheral and central sensitization to mechanical allodynia involving inflammatory aspects, without putative on-target side effects on locomotion and memory performance and without apparent abuse liability.