Investigating the effect of Ultra-low dose naloxone on spinal BDNF and KCC2 cotransporter in morphine tolerant and hyperalgesia rats

Background: Chronic opioids administration could lead to several side effects including morphine tolerance and opioid induced hyperalgesia (OIH). Tolerance and hyperalgesia to opiates reduce their effectiveness in the treatment of severe pain. Although the mechanisms are unclear. Recently, we have shown that repeated morphine treatments induced increases in spinal PKC (cid:0) and GAT-1 expression. In this study we investigate the BDNF and KCC2 expression through ultra-low dose of naloxone coadministration of morphine. Results: In morphine group, rats received 10 mg i.p. morphine and in treatment group 15ng ultra-low dose of naloxone with morphine for consecutive 8 days. Behavioral tests were performed on day 1 before and after the morphine injection, day 5, 8 (tolerance test) and 10, 48h after last morphine injection (opioid induced hyperalgesia OIH test). A number of rats were sacriced on day 8 and others on day 10, then expression of BDNF and KCC2 were analyzed by western blot and immunohistochemistry techniques, respectively. Behavioral tests suggested that following 8 days of chronic morphine injection tolerance developed. OIH was shown 48 hours after the last morphine injection. Expression of BDNF signicantly was increased and KCC2 downregulated in rats that developed morphine tolerance and OIH respectively. Ultra-low dose of naloxone by decreasing BDNF and increasing KCC2 was able to suppress development of OIH and alleviated morphine tolerance. Conclusions: Our data suggest that BDNF and KCC2 maybe candidate molecules which are involved in tolerance and OIH. Ultra- low dose of naloxone along morphine might be a valuable therapeutic potential for controlling hypersensitivity following chronic morphine administration.


Background
It has been shown that long-term administration of opioids, including morphine, leads to tolerance, dependence, addiction, and hyperalgesia. Among the multiple adverse side effects, tolerance and opioidinduced hyperalgesia are importance. Although morphine remains the best analgesic of all the opioids (1), studies have shown that continuous injection of morphine can increase spinal BDNF release, which is involved in the development of hyperalgesia. This is due to the effect of BDNF, which binds to the tyrosine kinase receptor (TrkB) and then inhibits KCC2 cotransporter activity. KCC2 increases the accumulation of chlorine and potassium ions in the cell and produces sensitivity and excitability (2,3). KCC2 levels downregulate by BDNF release following chronic morphine administration (3)(4)(5). Rivera (1999) presented several pieces of evidence showing that BDNF down-regulates KCC2, leading to impairment of Clextrusion from neurons. Researchers have shown an obvious downregulation of KCC2 mRNA expression by BDNF (6). BDNF-induced hyperexcitability has been proposed to be partly due to effects on Gammaaminobutyric acid (GABA) inhibition (7). The changes in intraneuronal Cl affect the amplitude and polarity of GABAergic responses (8) and are involved in the control of both neuronal and interstitial volume, which play a critical role in the modulation of neuronal excitability (9).
Opioid receptor antagonists, such as naloxone and naltrexone, at mg or µg. doses antagonize opioid functions, but at ng or pg. doses enhance the analgesic effect of opioid agonists (10). Co-administration of morphine with an ultra-low dose of naltrexone or naloxone potentiate the antinociceptive effect of morphine (11). Moreover, it has been shown that ultra-low-dose naloxone modulates spinal neuroin ammation and enhances the antinociceptive effect of morphine (12). These data suggest that ultra-low-dose naloxone might be clinically valuable in pain management (13) as co-infusion of ultra-lowdose naloxone along with morphine maintains the antinociceptive effect of morphine (12). Furthermore, the ultra-low-dose of naloxone is known as an effective adjuvant to suppress opioid tolerance and dependence (14). The mechanism underlying an ultra-low dose of naloxone is not well known. In this study, we examined the effect of co-administration of morphine and ultra-low-dose naloxone on BDNF and KCC2 expression in dorsal horn neurons in morphine-treated rats.

Tail ick test on day 1, 5 and 8 after morphine injection
Four-hour tail-ick test results in the morphine group compared with the sham group showed the antinociceptive effect of morphine. Indeed, morphine created the maximum possible analgesic effects (MPAE). This effect remained signi cant in the sham group, although it decreased gradually from 120 min after injection to 240 min. Also, the results showed that co-administration of the ineffective dose of naloxone along with morphine the rst day sustained the antinociceptive effect of morphine when compared to the sham group for up to three hours but decreased from 180 minutes to 240 min. However, it was not able to sustain the analgesic effect when compared to the morphine group ( Figure 1).
On day 5, the four-hour tail-ick test in the morphine-treated group as compared with the sham group showed the antinociceptive effect of morphine within the rst 60 minutes after injection. This effect gradually decreased from 60 min after injection to 120 min and continued decreasing until 240 min when it became similar to the sham group. The results of the same test showed that in the ineffective dose of co-administration of naloxone along with morphine up to day 5, the analgesic effect of morphine could be maintained for two hours, but after 120 minutes gradually decreased and became similar to the sham group ( Figure 2).
On day 8, the results of the tail-ick test in the morphine-treated group as compared to the sham group showed an analgesic effect within the rst 30 minutes after injection. This effect gradually decreased from 30 min to 240 min when it approached the sham group. Results of the same test showed in the ineffective dose of co-administration of naloxone along with morphine up to day 8 that the antinociceptive effect of morphine was maintained for up to 240 minutes and was signi cant when compared to the sham and morphine groups ( Figure 3).
The results of an ineffective dose of naloxone for 8 days showed no signi cant antinociceptive effect on the experimental days.
2.2. Thermal hyperalgesia between days 1 and 10 using Hargreaves test The results of the withdrawal plantar test, which measure thermal hyperalgesia, showed there was no signi cant difference between experimental groups on the rst day before drug injection. Eight consecutive days of morphine injection showed thermal hyperalgesia on day 10 (48 hours after the last morphine injection) and before drug injection. A withdrawal plantar test on day 10 showed signi cant hyperalgesia in the morphine group, but in the ineffective dose of co-administration of naloxone along with morphine group there was no signi cant paw withdrawal when compared to the sham group. The naloxone groups did not show any signi cant antinociceptive effect ( Figure 4).

BDNF protein expression changes following chronic administration of morphine
Western blot analysis showed the administration of morphine for 8 consecutive days increased BDNF protein expression in spinal dorsal horn neurons on days 8 and 10. Co-administration of an ineffective dose of naloxone along with morphine decreased the BDNF expression. The decreased expression of BDNF on day 8 was greater than day 10. There were no signi cant changes in BDNF expression between the ineffective dose of naloxone and sham groups ( Figure 5).

KCC2 expression Following chronic administration of morphine using immunohistochemical technique
Immunohistochemical analysis of KCC2 in the lumbar section of the spinal cord following chronic administration of morphine for 8 consecutive days showed a signi cant decrease of KCC2 on day 8 (tolerance) and day 10 (hyperalgesia) when compared to the sham group. Co-administration of 10ng daily injection of naloxone along with morphine increased the expression of KCC2 in the spinal dorsal horn neurons on days 8 and 10 when compared to the sham group ( Figure 6). Since both results on days 8 and 10 were similar and there were no signi cant changes of KCC2 in the dorsal horn neurons between days 8 and 10, we indicated day 10 images.

Discussion
Our study showed that intraperitoneal administration of morphine for 8 continuous days led to morphine antinociceptive tolerance and thermal hyperalgesia in rats. Coadministration of an ultra-low dose of naloxone along with morphine showed a controversial role intolerance. In other words, on the rst day that morphine showed maximum antinociceptive effect naloxone didn't have any effect, but gradually by the manifestation of tolerance and hyperalgesia, naloxone was able to improve and reduce tolerance and hyperalgesia by gradually manifesting them especially at day 8 and 10, respectively. It seems that naloxone had no effect until tolerance or hyperalgesia were established. Thus, our results showed the improving effects of coadministration of naloxone along with morphine but not alone. Our results were in agreement with previous studies. It has been previously demonstrated that chronic injection of intraperitoneal, subcutaneous, or intrathecal morphine produce tolerance and hyperalgesia (15,16). However, the development of hyperalgesia depends on the duration and dose of morphine (17).  (18,19). In contrast, Gu et al. (2005) showed that 6 days of i.t. chronic morphine infusion (40 nmol/µL/h) would not develop hyperalgesia (20). In 2015 Zuo et al. observed mechanical and thermal hyperalgesia symptoms in animals after daily intraperitoneal injection of 10 mg/kg morphine (21). Different studies have indicated that the co-administration of opioid antagonists can greatly prevent the side effects of long-term opioids administration. New studies have shown that opioid receptor antagonists, such as naloxone and naltrexone, have anti-in ammatory effects at ultra-low doses and can reduce or reverse the adverse effects of repeated opioid use (22). In 1995 Crain showed that co-administration of a very low dose of naloxone along with morphine reduces tolerance and withdrawal syndrome (23). In contrast, Wang et al. (2005) reported that the analgesic effects of co-administration of morphine and naloxone was reduced after 7 days. Similar to our results, they showed that naloxone alone was ineffective. Many pre-clinical studies have shown that an ultra-low dose of naloxone or naltrexone enhances the analgesic effects of opioids (11,14). In this study, we use an ultra-low dose of naloxone along with morphine and observed improving morphine antinociceptive effect.
Different mechanisms occur in neurons and glia that play an important role in this phenomenon. The molecular mechanisms underlying opioid tolerance and hyperalgesia is thought to overlap considerably with those involved in chronic pain signaling (24). Wang in 2005 reported that treatment with the same ultra-low dose of naloxone alone had no analgesic effect. Another nding of this study is that since the analgesic effects of opioids on receptors are inhibitory and chronic opioid administration switches G protein coupling from Gi/o to Gs, concomitant administration of an ultra-low dose of naloxone decrease Gs coupling and enhances Gi/o binding (14,25). The antinociceptive effects of ultra-low doses of naloxone (ng or pg.) or naltrexone are mediated by various mechanisms such as inhibition of opioid receptor signaling pathways in DRG neurons, the release of enkephalins, and neuro-in ammatory events triggered by microglia. Also, this dose of naloxone speci cally antagonizes the coupling of opioid receptors to Gs proteins, adenylyl cyclase/Gβγ signaling, and the pathway of excitatory signals in the spinal cord, and in turn, increases the tendency to pair with Gi/o thereby maintaining and enhancing morphine analgesia (13,23,25,26). In subsequent studies in 2008 and 2009, Wang et al. stated that naloxone may bind to A laments and prevent G protein switching from inhibitory to excitatory transcriptional pathways (27)(28)(29).
The results of molecular studies in hyperalgesia and tolerant animals showed that administration of morphine for 8 consecutive days resulted in increased expression of BDNF in spinal dorsal horn neurons.
We also observed that daily injection of 15 ng of naloxone plus morphine prevented the increase in BDNF expression, and injection of naloxone at a very low dose without morphine did not affect BDNF expression. Studies have shown that continuous injection of morphine can increase spinal BDNF release, which is itself involved in the development of hyperalgesia. This is also due to the effect that BDNF exerts on neurons by binding to tyrosine kinase receptors (TrkB) so that by inhibiting KCC2 cotransporter activity, it increases the accumulation of chlorine and potassium ions in the cell as well as sensitivity and excitability (2,3). BDNF-induced hyperexcitability has been proposed to be partly due to the effects on GABAergic inhibition, Coull et al. (2005) showed that BDNF has decreased the paw withdrawal threshold and manifested hyperalgesia. In their in vivo study, they added BDNF to the dorsal horn neurons of the spinal cord and observed less negative potential than the control neurons, which could be indicating a shift from inhibitory to excitatory GABA responses (4). In our belief, BDNF may have different roles during morphine administration rather than singular. Therefore, it is necessary to clearly understand the function of BDNF in morphine tolerance and OIH.
In the second part of this study, we used immunohistochemistry to investigate the distribution of kCC2 cotransporter expression in the transverse section of the spinal cord.
In the morphine group, the expression of kCC2 cotransporter showed a signi cant decrement. We also observed increased kCC2 cotransporter activity in the dorsal horn neurons compared to the morphine group in rats that received morphine plus an ultra-low dose of naloxone. There was no signi cant change in kCC2 expression in the ultra-low dose of naloxone saline group. As Kakazu stated in 2000 and Hubner in 2001, the maintenance of hyperpolarizing inotropic responses is dependent on kCC2 (30,31). Rivera (1999) con rmed that kCC2 is the major extruder ion that triggers hyperpolarization postsynaptic inhibition in the CNS (6). Therefore, any dysfunction of this cotransporter results in the accumulation of chloride ions inside the cell, due to the increase of chloride ions inside the cell, causes GABAergic receptors to shift from inhibitory to excitatory. All of these conditions can lead to hyperalgesia behavior in animals. Rivera et al. (2002) showed that following the addition of BDNF to the culture medium, the expression intensity of KCC2 decreased, which subsequently affected neuronal chloride ow. They presented evidence that BDNF regulates KCC2 downregulation and disruption of chlorine ow by acting on the TrkB receptor (7).
In a clinical trial on patients undergoing colorectal surgery in 2015, Xiao et al. showed that the combination of an ultra-low dose of naloxone (0.25 µg/kg/h) with remifentanil decreased the adverse effects of acute morphine administration, tolerance, postoperative pain scores, bowel function recovery time, and length of hospital stay (32). Another clinical trial in 2019 on thoracotomy patients receiving bupivacaine and naloxone (100 ng) compared to bupivacaine alone showed signi cantly longer postoperative pain relief (33). In contrast, a study by Hosseininejad et al. in 2019 on renal colic patients did not report a signi cant analgesic effect between the group receiving naloxone (0.001 mg/kg IV.) plus morphine (0.1 mg/kg) with the group receiving only morphine (34). We suggest that the simultaneous administration of an ultra-low dose of naloxone affects the expression and secretion of BDNF and KCC2 regulation and consequently reduces the tolerance and hyperalgesia induced by chronic morphine administration.
Since opioid receptors mediate the inhibition of adenylyl cyclase, and an ultra-low dose of naloxone antagonize excitatory functions opioid receptors through G protein binding, this pathway is likely to be involved (35). Also, there is a direct relationship between BDNF increasing and KCC2 decreasing (7), since naloxone in ultra-low-dose has an anti-in ammatory effect (22,36,37). Therefore, it is hypothesized that this secondary drug can reduce BDNF secretion and subsequently prevent the downregulation of KCC2, thereby reducing the tolerance and hyperalgesia induced by long-term use of morphine. Equivalent inhibition of the Gs coupling of CREB phosphorylation by naloxone and its binding to pentapeptide lament A seems to be the mechanism of action of a very low dose of naloxone or naltrexone (27)(28)(29).
According to the mentioned studies and our results, there seems to be a strong relationship between the activity of BDNF expression and downregulation of KCC2 cotransporter during chronic morphine administration and neuroin ammation. Since the use of naloxone at very low doses along with morphine has an anti-in ammatory effect and can decrease microglial activity (7, 36), we hypothesize that a ULD of naloxone may affect BDNF, decreasing secretion and subsequently prevent KCC2 downregulation, and thereby prevent the tolerance and hyperalgesia induced by long-term use of morphine. Considering that BDNF increased in morphine treated rats, and the ability of ultra-low dose of naloxone along with morphine to block the rise in BDNF expression, it is clear that decreased BDNF expression plays an important role in modulating KCC2 activity. Increased BDNF expression, in turn, leads to more alterations in nerve cell membranes that could eventually lead to the development of OIH and tolerance. Although it seems that ULD of naloxone affects BDNF, decreasing secretion and subsequently prevent KCC2 downregulation and could improve the analgesic effect of morphine, it should be noted the effects of ultra-low doses of naloxone are mediated by various mechanisms.

Conclusion
Combined with previous study our data suggests that BDNF and KCC2 maybe candidate molecules that are involved in tolerance and OIH. By increasing KCC2, the ultra-low dose of naloxone along with morphine might be a valuable therapeutic potential for controlling hypersensitivity following chronic morphine administration.

Animal and experimental design
All experimental protocols were followed by guidelines on the ethical standard for investigation of morphine tolerance and hyperalgesia in animals by the Animal Animals were allowed to habituate to the housing circumstances for several days before the experiments began. Behavioral studies were performed in a quiet condition. Efforts were made to use the minimum number of animals necessary to achieve statistical signi cance.
Rats were randomly separated into 8 groups with 6 rats in each group and divided to tolerance and OIH groups: i) sham, ii) morphine (10 mg), iii) treatment group: morphine10 mg and naloxone (NAL, 15 ng), and iv) naloxone (15 ng) and saline (SAL) Table 1.
In the sham group, rats received saline, in the morphine group, rats received (IP)10 mg morphine i.p. in the treatment group, rats received 15ng ULD of naloxone i.p., along with morphine, and in the saline group, rats received ULD of naloxone along with saline for 8 consecutive days. Behavioral tests were performed to evaluate tolerance on day 1 before and after the morphine injection, and days 5 and 8 by the tail-ick test (TL) (Borje Sanat Co. Iran); and on day 1 before morphine injection and day 10, 48h after last morphine injection to evaluate OIH by plantar test (Ugo Basile, Italy). Three rats from all groups were euthanized under (2.0-2.5%) Iso urane inhalation anesthesia on day 8 and the Lumbar segments of the spinal cord were collected (to study BDNF expression by the western blotting method). On day 10 three rats from all groups were deeply anesthetized by ketamine (10%) and xylazine (2%) i.p. and transcardially perfused with phosphate-buffered saline. Spinal cords were isolated and lumbar segments were removed (to study KCC2 expression by the immunohistochemistry method). All experiments were done from 8 am until 4 pm.

Behavioral tests
The plantar test (Hargreaves method) was used to assess the development of thermal hyperalgesia. Rats were placed in the plexiglass chambers to adapt to the lab environment 30 minutes before the initiation of tests. A light beam, as the heat source, was focused on the middle surface of the rat's right hind paw. The time interval between the initiation of the light beam and the rapid withdrawal of the hind paw was considered as the nociceptive threshold. A cut off time of 30 seconds was set to prevent tissue damage.
The plantar test was performed on day 1, before the rst injection of morphine and on day 10, 48 hours after the last morphine injection. On each day, 5 trials were made with 5 minutes interval and mean paw withdrawal latency (PWL) was obtained. A signi cant decrease in PWL from baseline in day 1 was de ned as thermal hyperalgesia.
The development of morphine analgesic tolerance was examined using the tail-ick test. Rats were placed on the tail-ick apparatus before the injection of Morphine on days 1, 5, and 8. Radiant heat was administered to the caudal end of the tail. A cut off time of 20 seconds was established to minimize tissue damage. In each session, 3 trials were conducted with 3 minutes in between each trial. The mean tail-ick latencies (TL) of the 3 trials was set as the baseline latency (BL).

Western blotting
On days 8 and 10, three rats from all groups were sacri ced. Western blotting was performed on lumbar segments of spinal cord homogenates to determine BDNF protein expression levels. The whole lumbar segment of the spinal cord was homogenized in lysis buffer, and the total protein extract was obtained by centrifugation at 13,000g for 40 min at 4 °C. Protein concentration was determined by Bradford assay, and equivalent protein amounts of each sample were separated using dodecyl sulfate polyacrylamide gel electrophoresis sodium dodecyl sulfate page (SDS-PAGE). The proteins were transferred onto polyvinylidene di uoride (PVDF) membranes using a Bio-Rad immunoblotting apparatus. Afterward, to block non-speci c binding sites, membranes were incubated in blocking buffer bovine serum albumin (BSA 5%) for 80 minutes. Membranes were then incubated overnight to block non-speci c binding sites; membranes were incubated in blocking buffer (BSA 5%) for 80 minutes. Membranes were then incubated overnight with the BDNF antibody (AB203573,1:5000). The next day, blots were washed three times with 0.1% Tris Buffered Saline Tween (TBST) 20 and Tris Buffered Saline (TBS) each for 10 minutes, and then incubated with horseradish peroxidase-conjugated secondary antibody in blocking buffer (SC-516102,1:10000) for 85 minutes. After washing the blots three times with TBST, protein complexes were visualized using ECL chemiluminescence (Amesrham). After scanning the X-ray lms, the protein expression was quanti ed and analyzed by ImageJ 1.44 NIH software. Data were represented as the ratio of the density of BDNF bands to β-actin bands.

Immunohistochemistry analysis
The expression of KCC2 in the lumbar segments of the spinal cord was assessed by immunohistochemistry analysis. On day 10 three rats from all groups were deeply anesthetized by ketamine (10%) and xylazine (2%) and trans cordially perfused with phosphate-buffered saline (PBS), followed by freshly prepared 4% phosphate-buffered paraformaldehyde (pH 7.4). Spinal cords were isolated and lumbar segments transected after cardiac perfusion with 4% paraformaldehyde. Next, the lumbar segments of the spinal cords were removed and post-xed in 4% paraformaldehyde for 48 hours. Lateral sections with a 5 µm thickness from lumbar segments of the spinal cords were prepared using a microtome rotatory apparatus. Immunohistochemical analysis of KCC2 was performed on para nembedded sections. First, sections were placed in the oven for 20 minutes, then depara nized in xylene and rehydrated. Next, sections were paci ed in 3% hydrogen peroxide (H2O2) and then exposed to heat mediated antigen retrieval using citrate buffer (pH 6) for 50 minutes. Subsequently, sections were blocked to prevent non-speci c binding, prior to incubating them overnight with primary antibody (MAB=16982, 1/300) at 4 °C. The sections were then washed with wash buffer and incubated with a ready-to-use biotinylated secondary antibody (PAB0096), followed by the addition of a streptavidinconjugated enzyme. After immunoreactions, sections were washed with wash buffer and then color development performed using 3,3'-diaminobenzidine (DAB) solution for 10 minutes. Slides were then counterstained with hematoxylin and dehydrated using graded alcohols and xylene. Finally, sections were mounted onto slides, following coverslip application. The images were taken using an optical Olympus AX70 microscope (Japan) with a DP11 digital camera. The immunoreaction intensity was analyzed with NIH Image J software. The values were obtained by the densitometric analysis performed by ImageJ software. The background intensity was subtracted from the intensity in the areas of interest. Average intensities were obtained from 3 animals in each group.

Statistical Analysis
In this study, data were expressed as mean ± standard (S.E.M.). GraphPad Prism (version 6.0; GraphPad Prism Software Inc., San Diego, CA, USA) was used for statistical analysis. Data from behavioral tests were examined between groups by two-way ANOVA analysis followed by Tukey's post-hoc test for multiple comparisons. Western Blot analysis were compared by one-way ANOVA followed by Tukey's post-hoc test. Quantitative analysis of immunohistochemical images was performed by t-test. P-values less than 0.05 were considered as statistically signi cant. Maximum antinociceptive effect (MPAE%) after morphine injection on the rst day. Tail ick test was measured at 30, 60, 120, 180 and 240 minutes after drug injection. In the MOR group, the antinociceptive effect of morphine decreased gradually from 120 min after injection to 240 min but remained signi cant when compared to the sham group. An ineffective dose of naloxone did not produce the antinociceptive effect in comparison to the sham group. NLX + MOR was able to produce a signi cant antinociceptive affect up to 180 minutes after injection when compared to the sham but gradually decreased after three hours. However, it was not able to sustain the analgesic effect when compared to the morphine group.

Figure 2
Maximum antinociceptive effect (MPAE%) after morphine injection on day 5. Tail ick test was measured at 30, 60, 120, 180 and 240 minutes after drug injection. The four-hour tail-ick test in the morphinetreated group compared with the sham showed the maximum antinociceptive effect of morphine within the rst 60 minutes after injection. It was gradually decreased from 60 min after injection to 120 min and continued until 240 min and became similar to the sham group. The results of the same test showed that in the ineffective dose of co-administration of naloxone along with morphine up to day 5, the antinociceptive effect of morphine could maintain for two hours, but after 120 minutes gradually decreased and became similar to the sham group. The effect of analgesia has shown as a percentage.
P<0.05 was considered a signi cant difference. (P <0.01 **, *** P <0.001, **** P <0.0001)  The results of the withdrawal plantar test which measure thermal hyperalgesia showed that there was no signi cant difference between experimental groups on the rst day before drug injection. Consecutive 8 days of morphine injection showed thermal hyperalgesia on day 10 (48 hours after the last morphine injection) and before drug injection. Withdrawal plantar test on day 10 showed signi cant hyperalgesia in the morphine group but in the ineffective dose of co-administration of naloxone along with morphine group, there was no signi cant paw withdrawal when compared to sham. The naloxone groups did not show any signi cant antinociceptive effect. P <0.05 was considered as the least signi cant difference. P <0.0001 **** = Comparison of thermal hyperalgesia between morphine and sham groups