Heme oxygenase-1 in the spinal cord plays crucial roles in the analgesic effects of pregabalin and gabapentin in a spared nerve-injury mouse model

INTRODUCTION
Neuropathic pain remains one of the most intractable types of pain; although calcium channel α2δ ligands, such as pregabalin and gabapentin, are classified as first-line drugs, they have only modest efficacy. Heme oxygenase-1 (HO-1) signaling attenuates glial activation during neuropathic pain. Thus, this study aimed to investigate the effects of the blood-brain barrier (BBB)-permeable HO-1 inhibitor, tin protoporphyrin IX (SnPP), or the BBB-impermeable HO-1 inhibitor, zinc (II) protoporphyrin IX (ZnPP), on the analgesic efficacy of pregabalin and gabapentin. Additionally, we examined the effects of co-administration of SnPP with pregabalin or gabapentin on the expression of glial markers or other genes.


METHODS
Neuropathic pain was induced by spared nerve injury (SNI) of the sciatic nerve. The mechanical threshold was tested using the von Frey filaments. The expression of spinal glial markers or other genes was examined using reverse transcription polymerase chain reaction.


RESULTS
Systemic HO-1 inhibition reversed the mechanical antiallodynic effects of pregabalin and gabapentin, although peripheral HO-1 inhibition did not alter the mechanical antiallodynic effects of either pregabalin or gabapentin. Intrathecal injection of SnPP or ZnPP abolished the mechanical antiallodynic effects of pregabalin and gabapentin. Pregabalin and gabapentin increased HO-1, arginase-1, and endogenous opioid precursor preproenkephalin gene expression and decreased the expression of glial markers, interleukin-1β, and inducible nitric oxide synthase.


CONCLUSIONS
This study suggests that spinal HO-1 plays a crucial role in the analgesic effects of calcium channel α2δ ligands through the attenuation of glial activation and endogenous opioid release.


Introduction
Neuropathic pain is one of the most intractable types of pain [1] and is estimated to affect 3%-17% of the general population [2]. Calcium channel α 2 δ ligands attenuate central sensitization and excitatory neurotransmitter release in the spinal dorsal horn [3], exerting analgesic effects peripherally (dorsal root ganglia) and centrally (spinal dorsal horn) [4,5]. Thus, calcium channel α 2 δ ligands are classified as first-line drugs for neuropathic pain [6].
The effects of HO-1 signaling and the analgesic effects of calcium channel α 2 δ ligands occur peripherally and centrally [17,18]. However, the location of the interaction between HO-1 signaling and calcium channel α 2 δ ligands remains to be elucidated. In this study, we investigated 1) the effects of SnPP and ZnPP on the analgesic efficacy of pregabalin and gabapentin and 2) effects of the co-administration of SnPP with pregabalin or gabapentin on the expression of glial markers or other genes.

Animals
Male C57BL6 mice aged 8-10 weeks were obtained from Japan SLC, Inc. (Hamamatsu, Japan). The Animal Research Committee of Kagoshima University approved all experimental procedures, which were implemented according to the guidelines of the National Institutes of Health and International Association for the Study of Pain (approved number, MD17071) [19]. Four mice weighing 20 to 25 g were housed in each cage (floor area 54 square inch, height 5 in.) for approximately 7 days before the beginning of the study. The mice were provided free access to food and water and were on a 12:12 light/dark schedule at 21 • C and 60% humidity.

Neuropathic pain model
Neuropathic pain was induced by spared nerve injury (SNI) of the sciatic nerve [20]. The mice were deeply anesthetized by inhalation of 1.5%-2.0% isoflurane (Abbott, Tokyo, Japan) via a nose cone. An incision was made at the mid-thigh level, and a section was made through the biceps femoris. The tibial and common peroneal nerves were ligated and transected using a 6-0 silk suture. A 1-to 2-mm section of the two nerves was removed. The procedure was performed carefully, avoiding any damage to the sural nerve. The muscles and skin were sutured using two 6-0 silk sutures. We have previously reported the development of ipsilateral mechanical allodynia 1 to 8 weeks after inducing SNI [10]. As observed in the previous study, all mice developed neuropathic pain in the present study [20]. We confirmed reduction of the mechanical threshold after surgery. There was no adverse event such as paralysis. All mice used in the study received SNI surgery. Since neuropathic pain is established on day 7 after SNI surgery, all the experiments (both behavioral and gene expression tests) were performed on the day.

Mechanical threshold
The mechanical threshold was determined using calibrated von Frey filaments (0.008-2.0 g; Aesthesio® Precise Tactile Sensory Evaluator; Danmic Global, San Jose, CA, USA) introduced serially to the hind paw in ascending order of strength. Animals were placed in non-transparent plastic cubicles on a mesh floor for an acclimatization period of at least 30 min on the morning of the test day. A positive response was defined as rapid withdrawal and/or licking of the paw immediately after the application of the stimulus. Filaments were tested five times per paw, and the paw withdrawal threshold was defined as the filament for which three or more withdrawals in five trials were observed [21]. The operators who conducted the tests were blinded to the treatment. Antinociception induced by the experimental drugs was expressed as the percentage of the maximal possible effect calculated according to the following equation [10]: The baseline values were obtained before test drugs administration on the day 7 after SNI surgery.

Experimental protocol
The doses of SnPP or ZnPP combined with pregabalin or gabapentin in this study were selected as those that produced a relevant effect in accordance with previous studies [10,18,23].
In the first experiment, we evaluated effects of the systemic (intraperitoneal) administration of SnPP on the mechanical antiallodynic effects produced by pregabalin or gabapentin (n = 8 animals per group). In the second set of experiments, we evaluated effects of the peripheral (intra-planter in the left hind paw) administration of ZnPP on the mechanical antiallodynic effects produced by pregabalin or gabapentin (n = 7-8 animals per group). In the third set of experiments, we evaluated effects of the intrathecal administration of SnPP or ZnPP on the mechanical antiallodynic effects produced by pregabalin or gabapentin (n = 7-8 animals per group).
Finally, in the fourth set of experiments, we evaluated effects of the intrathecal administration of SnPP combined with intraperitoneal administration of pregabalin or gabapentin on the expression of HO-1, a microglial marker (ionized calcium-binding adapter molecule-1, Iba-1), an M1 microglial marker (inducible nitric oxide synthase, iNOS), an M2 microglial marker (arginase-1), the pro-inflammatory cytokine interleukin (IL)-1β, an astrocyte marker (glial fibrillary acidic protein, GFAP), and the endogenous opioid precursor, preproenkephalin. SNIaffected mice treated with intrathecal administration of dimethyl sulfoxide (DMSO) combined with intraperitoneal administration of saline were used as controls (n = 4-6 samples per group).

Drugs
Pregabalin and gabapentin were purchased from Sigma-Aldrich (St. Louis, MO, USA) and were dissolved in saline solution (0.9% NaCl). In all four sets of experiments, pregabalin and gabapentin were administered intraperitoneally 60 min before the experiment. SnPP and ZnPP purchased from Enzo Life Sciences, Inc. (Farmingdale, NY, USA) were dissolved in DMSO (1% solution in saline). All drugs were freshly prepared before use. In the first experiment, 10 mg/kg of SnPP or 1% DMSO was mixed with pregabalin or gabapentin before intraperitoneal administration. The total amount of solution injected intraperitoneally was 300 μL. In the second experiment, 400 nmol of ZnPP or 1% DMSO was injected 30 min before the behavioral tests. The total amount of solution injected into the hind paws was 20 μL/paw for intra-planter injection. In the third and fourth experiments, percutaneous intrathecal injections between L5 and L6 were performed in unanesthetized mice according to the method described by Hylden and Wilcox, using a 10-mL Hamilton syringe and disposable 30-gauge 0.5-inch needles [24]. The appropriate needle placement for injection was confirmed by the elicitation of a strong tail-flick reflex. The total amount of intrathecally injected solution was 4 μL.

Statistical analysis
Data are expressed as mean ± standard error of the mean. To detect a value for sigma 20%-40%, with α = 0.0125-0.05 and power of 80%-99%, we needed a sample size of 5-8 for behavioral testing and 4-6 for gene expression studies, based on our previous publications and preliminary data (R version 4.1.1, R Foundation for Statistical Computing, Vienna, Austria) [10]. Statistical analysis was performed using Graph-Pad Prism 7.0 (GraphPad Software, La Jolla, CA, USA). All comparisons were two-tailed. The effects of the experimental drugs on gene expression in the ipsilateral spinal dorsal horn were compared using one-way analysis of variance (ANOVA) followed by Tukey's multiple comparison test. Statistical significance was set at P < 0.05.

Effects of systemic SnPP administration on the mechanical antiallodynic efficacy of pregabalin and gabapentin
We assessed effects of the intraperitoneal administration of 10 mg/kg of SnPP or vehicle (1% DMSO) on the mechanical antiallodynic effects produced by intraperitoneal administration of pregabalin (30 mg/kg) or gabapentin (30 mg/kg) in mice with SNI. Co-administration of pregabalin with SnPP completely reversed the mechanical antiallodynic effects of pregabalin administered alone (Fig. 1A). Similarly, coadministration of gabapentin with SnPP completely reversed the mechanical antiallodynic effects produced by gabapentin administered alone (Fig. 1B).

Effects of peripheral ZnPP administration on the mechanical antiallodynic efficacy of pregabalin and gabapentin
We also investigated effects of intra-planter (left hind paw) administration of 400 nmol of ZnPP or vehicle (1% DMSO) on the mechanical anti-allodynic effects produced by intraperitoneal administration of pregabalin (30 mg/kg) or gabapentin (30 mg/kg) in mice with SNI. The intra-plantar injection of ZnPP did not affect the mechanical Peripheral ZnPP administration does not alter the effects produced by (C) pregabalin or (D) gabapentin (one-way ANOVA followed by Tukey's multiple comparison test, n = 7-8). ANOVA, analysis of variance; DMSO, dimethyl sulfoxide; ip, intraperitoneal; ipl, intra-planter; SnPP, tin protoporphyrin IX; ZnPP, Zinc (II) Protoporphyrin IX. antiallodynic effects produced by pregabalin (Fig. 1C) or gabapentin (Fig. 1D).

Effects of the intrathecal administration of SnPP or ZnPP on the mechanical antiallodynic efficacy of pregabalin and gabapentin
We assessed effects of the intrathecal administration of 100 nmol/kg of SnPP or vehicle (1% DMSO) on the mechanical antiallodynic effects produced by intraperitoneal administration of pregabalin (30 mg/kg) or gabapentin (30 mg/kg) in mice with SNI. Coadministration of pregabalin with SnPP completely reversed the mechanical antiallodynic effects of pregabalin administered alone ( Fig. 2A). Similarly, coadministration of gabapentin with SnPP completely reversed the mechanical antiallodynic effects produced by gabapentin administered alone (Fig. 2B).
Furthermore, we assessed effects of the intrathecal administration of 100 nmol/kg of ZnPP or vehicle (1% DMSO) on the mechanical antiallodynic effects produced by intraperitoneal administration of pregabalin (30 mg/kg) or gabapentin (30 mg/kg) in mice with SNI. Coadministration of pregabalin with ZnPP completely reversed the mechanical antiallodynic effects of pregabalin administered alone (Fig. 2C). Similarly, co-administration of gabapentin with ZnPP completely reversed the mechanical antiallodynic effects produced by gabapentin administered alone (Fig. 2D). Figure 3 shows the mRNA levels of HO-1, Iba-1, GFAP, iNOS, arginase-1, IL-1b, and preproenkephalin in the ipsilateral spinal dorsal horn of mice treated with vehicle, pregabalin alone, or pregabalin combined with intrathecal SnPP. In mice treated with pregabalin alone, increased expression of HO-1, arginase-1, and preproenkephalin and decreased expression of Iba-1, iNOS, IL-1b, and GFAP were noted. The changes in the expression of HO-1, arginase-1, preproenkephalin, IL-1b, and GFAP were reversed by coadministration of SnPP. The decrease in the expression of Iba-1 and iNOS was not altered by coadministration of SnPP. Figure 4 shows the mRNA levels of HO-1, Iba-1, GFAP, iNOS, arginase-1, IL-1b, and preproenkephalin in the ipsilateral spinal dorsal horn of mice treated with vehicle, gabapentin alone, or gabapentin combined with intrathecal SnPP. In mice treated with gabapentin alone, increased expression of HO-1, arginase-1, and preproenkephalin and decreased expression of IL-1b, and GFAP were noted. The changes in the expression of HO-1, arginase-1, and preproenkephalin were partially

Discussion
In this study, we demonstrated in SNI model mice that the systemic administration of a BBB-permeable HO-1 inhibitor, SnPP, reversed the analgesic effects of pregabalin or gabapentin, while peripheral administration of the BBB-impermeable HO-1 inhibitor ZnPP did not alter these analgesic effects. However, the analgesic effects of pregabalin or gabapentin were reversed by the intrathecal administration of either SnPP or ZnPP. Pregabalin administration increased the expression of HO-1, arginase-1, and preproenkephalin and decreased expression of Iba-1, iNOS, IL-1b, and GFAP. Similarly, gabapentin administration increased expression of HO-1, arginase-1, and preproenkephalin and decreased expression of IL-1b and GFAP. However, the pregabalininduced alterations in the expression of HO-1, arginase-1, preproenkephalin, IL-1b, and GFAP, as well as the gabapentin-induced alterations in the expression of HO-1, arginase-1, and preproenkephalin were prevented by intrathecal administration of SnPP. Taken together, these results suggest that spinal HO-1 plays a crucial role in the analgesic effects of calcium channel α 2 δ ligands through the modulation of glial activation and endogenous opioid release.
SnPP and ZnPP are metal-based HO-1 inhibitors [11] that are frequently used in pain studies [10,18,23,25]. Some studies have reported that SnPP does not pass through the BBB [26,27]. However, we and other researchers have shown that the intraperitoneal administration of SnPP significantly inhibits HO-1 in the brain and spinal cord [10,23]. Conversely, the HO-1 inducers CoPP or CORM-2 increase HO-1 expression in the spinal cord, which consequently reduces microglial/ astrocytic activation and iNOS levels [28]. We have previously reported that HO-1 induction potentiates analgesic effects of pregabalin/gabapentin in neuropathic pain [10]. Although pregabalin and gabapentin have similar pharmacologic properties, we used both drugs. The two drugs showed similar responses, which strengthens our results' generalization.
Microglial and astrocytic activation is a key component of neuropathic pain [12,13]; there is clinical evidence of glial activation in cases Fig. 3. Effects of intrathecal administration of 100 nmol/kg of SnPP combined with intraperitoneal administration of 30 mg/kg of pregabalin on the expression of glial markers, IL-1b, and preproenkephalin. Pregabalin significantly increased the expression of (A) HO-1, (D) Arg1, or (G) Penk and significantly decreased the mRNA levels of (B) Iba-1, (C) iNOS, (E) IL-1b, or (F) GFAP in the ipsilateral spinal dorsal horn. Coadministration of SnPP partially reverses the effects (one-way ANOVA followed by Tukey's multiple comparison test, n = 5-6). ANOVA, analysis of variance; Arg-1, arginase-1; DMSO, dimethyl sulfoxide; GFAP, glial fibrillary acidic protein; HO, heme oxygenase; Iba-1, ionized calcium-binding adapter molecule-1; IL-1β, interleukin-1β; iNOS, inducible nitric oxide synthase; ip, intraperitoneal; it, intrathecal; SnPP, tin protoporphyrin IX; Penk, preproenkephalin. of neuropathic pain [29]. SNI increases expression of Iba-1 and GFAP in the spinal cord compared to sham operated mice [30]. HO-1 promotes macrophage/microglia polarization towards the anti-inflammatory M2 type [16], and endogenous opioids are secreted by leukocytes, including M2 macrophages [31]. Ahmad et al. showed that intrathecal administration of calcium channel α 2 δ ligands increased the expression of antiinflammatory cytokine IL-10 and β -endorphin in microglia, but not in neurons or astrocytes [32]. However, no study has shown that microglia secrete enkephalin, an endogenous opioid [33]. Enkephalin is expressed by astrocytes and GABA-ergic interneurons [34]. Spinal astrocytes are activated in the spinal dorsal horn of patients with human immunodeficiency virus (HIV) infection who also have chronic pain [35], but not in that of HIV-infected patients without chronic pain. Herein, pregabalin attenuated astrogliosis, and SnPP reversed these effects; therefore, astrocytes may be the source of increased enkephalin. Furthermore, pregabalin and gabapentin increase glutamate-induced intracellular calcium concentrations in astrocytes to stimulate descending inhibition [36,37]. The endoplasmic reticulum and mitochondria are important cytosolic calcium controllers; endoplasmic-reticulum stress and mitochondrial oxidative stress are involved in neuropathic pain induced by nerve injury [38,39].
This study has limitations. First, because we did not modulate glial activity, we could identify the cells (microglia, astrocytes, or other cells) that secrete inflammatory cytokines or endogenous opioids. Second, although we observed gene expression induced by calcium channel α 2 δ ligand, it is unclear if the changes in gene expression are a result of or occur simultaneous to glial modulation. Calcium channel α 2 δ ligandinduced modulation in each glial cell needs to be explored in future studies.
In conclusion, spinal HO-1 plays a crucial role in the analgesic effects of calcium channel α 2 δ ligands through the modulation of glial activation and endogenous opioid release. The current results provide important information for understanding precise mechanisms of calcium channel α 2 δ ligands.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.