A Single Subanesthetic Dose of Ketamine Relieves Depression-like Behaviors Induced by Neuropathic Pain in Rats

• Ketamine reduces depressive symptoms in patients in small reports, but whether this effect is independent of pain relief in patients with chronic pain has not been addressed

• In rats with neuropathic hypersensitivity, low-dose ketamine reduced behavioral measures of depression without altering hypersensitivity

DEPRESSION affects 30–100% of patients with chronic pain and is likely underestimated in postoperative pain.1,–,6Depression leads to additional emotional and cognitive deficits,7and understanding the relationship between pain and depression will help to tailor treatments.8There is evidence that depression alters the threshold of pain, although few studies examined whether depression is an integral affective component of the pain experience.9,– 14 

In rodents, pain is often assessed by sensory hypersensitivity, measured by stereotyped behaviors such as limb withdrawal.15–16Hypersensitivity is an important sensory feature, but does not reveal cognitive and emotional responses to pain, which can be assessed by Morris water maze, elevated plus maze, and open field tests.13,17,–,19In rodents, depression can be measured by forced swim test (FST) and sucrose preference test (SPT).20FST, a test for behavioral despair, was used to examine depression in the spinal nerve ligation model of neuropathic pain. Whereas Kontinen et al.  reported no changes in the FST performance in rats 14 days after surgery,21Suzuki et al.  found rats displayed depressive traits 15 days postoperatively.22SPT measures anhedonia, a key feature of depression.20SPT, however, was rarely used in pain studies.

Ketamine has both analgesic and antidepressant properties and is ideally suited to treat pain-induced depression.23–24Ketamine antagonizes N -methyl-D-aspartic acid receptors in spinal dorsal horn neurons to decrease central sensitization,25–26provides descending monoaminergic inhibition,27and blocks  Na⁺channels and μ-opioid receptors in peripheral fibers.28Ketamine is metabolized within 1 h and is useful as a short-acting analgesic.29Although ketamine may block central sensitization to mediate long-acting analgesia,30this remains to be proven in clinical practices.31In contrast, clinical studies have shown that ketamine provides enduring antidepressant effects.32,–,34,35–36Several proposed mechanisms explain the antidepressant properties of ketamine: increased presynaptic glutamate release and up-regulation of postsynaptic machinery in the prefrontal cortex,37–38and increased brain derived neurotrophic factor expression in the hippocampus.39Thus, whereas short-lived analgesic properties of ketamine are mediated at peripheral and spinal levels, its antidepressant activities involve frontal and limbic structures and are long-lasting. Clinically, depression can be treated at a lower dose than pain.24,33Similarly, in rats, less than 10 mg/kg ketamine provides antidepressant effects but more than 25 mg/kg is needed for analgesia.27,38,40,– 42Therefore, determining whether ketamine can treat depression-like behaviors at a dose (less than 10 mg/kg) that does not treat sensory hypersensitivity in rats may provide a distinction between sensory and depressive components of pain.

We examined the spared nerve injury (SNI) model in rats.43Using a combination of SPT and FST, we found that rats exhibited depression-like behaviors immediately after surgery. Whereas depression-like behaviors induced by sham operation were reversible, depression induced by SNI surgery was chronic. These results suggest that depression may be a key feature of postoperative and chronic pain in rats; hence we favor the routine use of SPT and FST in pain assessments. Moreover, a single subanesthetic dose of ketamine produced rapid and enduring antidepressant effects in rats with chronic pain, without decreasing sensory hypersensitivity, suggesting that depressive symptoms of pain can be modulated independently of sensory symptoms.

All procedures in this study were approved by the New York University School of Medicine Institutional Animal Care and Use Committee as consistent with the National Institutes of Health Guide for the Care and Use of Laboratory Animals  (publication number 85–23) to ensure minimal animal use and discomfort. Male Sprague-Dawley rats were purchased from Taconic Farms, Albany, NY and kept in the New York University Langone Medical Center's Central Animal Facility, with controlled humidity, room temperature, and 12-h (6:30 AM to 6:30 PM) light-dark cycle. Food and water were available ad libitum . Animals arrived at the animal facility at 250–300 g and were given approximately14 days to adjust to the new environment before the onset of any experiments.

The SNI surgery has been previously described in detail.43Briefly, under isoflurane anesthesia (1.5–2%), the skin on the lateral surface of the right thigh of the rat was incised and a section made through the biceps femoris muscle to expose three branches of the sciatic nerve: sural, common peroneal, and tibial nerves. The common peroneal and tibial nerves were tied with nonabsorbent 5.0 silk sutures at the point of trifurcation. The nerves were then cut distal to the knot, and approximately 3–5 mm of the distal ends were removed. In sham surgeries (control), these nerves were dissected but not cut. Muscle and skin layers were then sutured close in distinct layers. Naïve rats underwent no surgery.

Ketamine hydrochloride (Ketaset) was purchased from Fort Dodge Animal Health, Fort Dodge, IA. Ketamine was diluted in saline to a concentration of 1–50 mg/ml (depending on the final dose of ketamine). For the ketamine experiments, rats that underwent SNI surgery were assigned to two subgroups, one receiving ketamine injections and the other saline (control) injections. The rats received 1, 3, 10, 20, or 50 mg/kg ketamine injected intraperitoneally in the ketamine group (0.3–0.5 ml), whereas a similar volume of saline was injected intraperitoneally to the control group. Injections were given at least 14 days after SNI surgeries and were followed by behavioral tests.

A traditional Dixon up-down method with von Frey filaments was used to measure mechanical allodynia.16,44In brief, rats were individually placed into Plexiglass chambers over a mesh table and acclimated for 20 min before the onset of examination. Beginning with 2.55 g, von Frey filaments in a set with logarithmically incremental stiffness (0.45, 0.75, 1.20, 2.55, 4.40, 6.10, 10.50, 15.10 g) were applied to the lateral third of right paws (in the distribution of the sural nerve) of animals before and up to 56 days after SNI or sham surgery and to rats that underwent no surgery but similar handling (naïve group). In addition, the tests were done on the left (uninjured) paws of the SNI animals. The 50% withdrawal threshold was calculated as described previously.16For ketamine experiments, mechanical allodynia tests were done 1 h or 1 day after ketamine (or saline) injection, and observers were blinded to the test conditions (ketamine vs.  saline treatments).

Animals were individually placed into Plexiglass chambers as above and acclimated for 20 min. A drop of acetone was applied to the lateral plantar surface of the paws (in the distribution of the sural nerve). As previously described,15,25the following scoring system was applied. 0: no visible response or startle response lasting less than 0.5 s; 1: paw withdrawal lasting less than 5 s; 2: withdrawal lasting 5–10 s, with or without licking of the paws; 3: prolonged repetitive withdrawal lasting more than 10s. Acetone was applied five times to each paw, and an average score was calculated. Cold allodynia tests were typically done after mechanical allodynia tests on the same day, and observers were blinded to the test conditions (ketamine vs.  saline treatments).

Animals were trained approximately 2 h each day for 7–10 days to drink from two identical bottles, one containing 1% sucrose solution and the other water. In order to avoid side preference, the bottles were placed on alternating sides every day. Before the SNI surgery, baseline preference for sucrose was established. During each test, two bottles (1% sucrose solution vs.  water) were presented to each animal for 30 min, and then the bottles were switched to the opposite side, and the test continued for an additional 30 min. At the end of each test, sucrose preference was calculated as volume of sucrose consumed divided by total liquid consumption for each rat. Based on their baseline preferences, animals were equitably assigned to the naïve, sham surgery, or SNI surgery group. After undergoing surgery, sucrose preference was then tested on postoperative day 2, 7, 14, and 56 for the SNI and sham surgery groups and 2, 7, and 14 for the naïve group. For ketamine experiments, all animals underwent SNI surgery, and animals were assigned to either the ketamine or saline group. Due to the variability of baseline presurgical sucrose preference, animals were assigned to treatment and control groups with the goal of ensuring that the average baseline preference for each group was approximately the same before each experiment. At 14 days after the SNI surgery, ketamine or saline was injected into each animal, and sucrose preference was then tested 1 and 5 days after the injection. Observers were blinded to the test conditions (ketamine vs.  saline treatments).

On the first session of the test, each animal was placed into a standard clear Porsolt (Stoelting Co., Wood Dale, Illinois) chamber with water at 25°C filled to 25 cm for 15 min. Afterward, the animal was taken out of the chamber, dried, and put back in its home cage. Twenty-four hours later, the animal was again placed into the Porsolt chamber under the same conditions for 5 min. Both sessions were videotaped, but only the second session was analyzed. Immobility was defined as a lack of movement of the hind paws lasting longer than 1 s. Two independent observers examined and graded the total time of immobility for each rat, and the average grade was presented for each animal. FST was conducted 14 days after surgeries. For the ketamine experiments, animals were randomly assigned to either treatment (ketamine) or control group (saline). FST was conducted 1 day after ketamine/saline injection, and observers were blinded to the test conditions (ketamine vs.  saline treatments).

Blood samples were procured from the jugular vein of each rat after decapitation. Blood samples then were spun down at ×1000 g for 10 min, and serum was then extracted from the sample and immediately frozen at −80°C. A corticosterone rat/mouse enzyme-linked immunosorbent assay kit and the manufacturer's recommended protocol (Immuno-Biologic Laboratories, Inc., Minneapolis, MN) was used to determine corticosterone absorbance levels within each serum sample using plate reader detection. Corresponding corticosterone concentrations were determined using a standard curve, where the standards were provided by the manufacturer. EnVision plate reader software (PerkinElmer, Inc., Waltham, MA) was used to determine the standard curve. Absorbance levels of each sample were plugged into the standard curves equation generated by the EnVision software and the corresponding corticosterone concentrations were generated.

The results of behavioral experiments were given as mean ± SEM. For mechanical allodynia, a two-way ANOVA with post hoc  multiple pairwise comparison Bonferroni correction tests was used to compare the 50% withdrawal threshold of the right and left legs of the SNI animals, and right legs of sham-operated and naïve animals. Cold allodynia was analyzed using the Kruskal-Wallis test with post hoc  Dunn multiple pairwise comparison tests at each time point. A one-way ANOVA with post hoc  multiple pairwise comparison Tukey post hoc  tests was used to analyze mechanical allodynia of SNI-treated animals that received higher doses of ketamine (20 or 50 mg/kg) or saline, and a Kruskal-Wallis test with post hoc  Dunn multiple pairwise comparison tests was used to analyze the cold allodynia data. Weight gain was analyzed using a two-way ANOVA with post hoc  multiple pairwise comparison Bonferroni correction tests. For the sucrose preference test, a two-way ANOVA with post hoc  Bonferroni correction tests was used to compare the preference at different time points of the SNI, sham-operated, and naïve groups, and to compare the preference at different time points of rats injected with ketamine or saline. In the dose-response experiment, a two-way ANOVA with post hoc  Bonferroni correction tests was used to compare sucrose preference for different doses of ketamine (1, 3, 10, and 20 mg/kg) versus  saline control. For the FST, a unpaired two-tailed Student t  test was used to compare the performances of sham-operated and SNI groups as well as to compare the performances of ketamine and saline groups. For corticosterone levels and total fluid consumption during the SPT, a one-way ANOVA was conducted. For all tests, a P  value less than 0.05 was considered statistically significant. All data were analyzed using GraphPad Prism Version 4 software (GraphPad, La Jolla, CA). We did not perform a power analysis a priori  for our experiments. The numbers of subjects used in our experiments were based on our experience with these behavioral tests and in agreement with standard literature. Same or similar numbers of subjects were used in control and test conditions. Thus, if the test condition provided statistical significance but the control did not, we would conclude that the control condition resulted in nonmeaningful findings.

We studied the SNI model for neuropathic pain. Compatible with previous reports,43we found that SNI surgery produced significant mechanical and cold hypersensitivity in the (spared) sural nerve distribution acutely, on postoperative day 1, and these effects persisted for up to 2 months (fig. 1A, 1B). In contrast, rats that underwent sham surgery or no surgical manipulation (naïve rats) did not demonstrate any sensory hypersensitivity in the spared nerve distribution. In addition, we observed no hypersensitivity in the uninjured paw of an animal after the SNI surgery. SNI results in neuropathy, which may produce chronic stress to complicate behavioral tests. To rule out this possibility, we used weight gain as a measure for the overall health and well-being of rats and found no difference among SNI-treated rats, naïve (nonoperated) rats or sham-operated rats (fig. 1C). Furthermore, we measured levels of the stress hormone corticosterone of rats 14 days after surgeries and found no differences among rats in the SNI, sham-operated, and naïve groups (fig. 1D).

Next, we applied standard tests of depression to rats in chronic pain. The first test we used was the FST. Increased time of immobility (instead of swimming) in a water tank is considered a measurement of behavioral despair. We performed our tests 14 days after the surgery – a standard time for the development of chronic neuropathic pain – and found that compared with sham-operated rats, rats after treatment for SNI developed significantly increased time of immobility (fig. 2A). We confirmed this expression of depressive behavior with the SPT, which assesses anhedonia, measured by the preference of sucrose solution to water. At baseline, sham-operated and SNI-operated rats demonstrated approximately more than 90% preference for sucrose (1%) solution over water. However, 14 days after surgery, SNI-operated rats demonstrated significantly decreased preference for sucrose compared with sham-operated rats, and this difference lasted up to 2 months (fig. 2B), concurrent with the time course of chronic sensory hypersensitivity demonstrated by SNI animals (fig. 1A and B). Next, we turned our attention to the early postoperative period (postoperative days 1 through 14). Because we were concerned about postoperative incisional pain associated with sham operations, we added an additional control group–naïve rats, who underwent no surgeries. Rats in both SNI and sham surgery groups developed significantly decreased preference for sucrose in the immediate postoperative period (fig. 2C, postoperative day 2), compared with the naïve group. This decrease in sucrose preference was transient in the sham surgery group, lasting less than 7 days. In contrast, the decrease in the SNI group was stable. To ensure that neuropathy resulting from SNI did not interfere with drinking behaviors, we measured the total fluid consumption and did not find any differences among the three groups of animals (fig. 2D).

If SNI causes depressive behaviors, then an antidepressant should be able to treat these behaviors. We applied a single dose of ketamine (10 mg/kg) known to treat depression38 versus  saline (control) intraperitoneally to animals that had undergone SNI surgery 14 days previously. One day after drug administration, we tested the sucrose preference for these animals. Although animals that received saline injection remained anhedonic, animals that received ketamine injections demonstrated a return to baseline presurgical sucrose preference (fig. 3A). Although ketamine improved sucrose preference, it did not affect total volume of fluid consumption (fig. 3B). We repeated these tests 5 days after the injection to determine whether the antidepressant effect of ketamine was enduring. We found animals that received ketamine 5 days previously continued to demonstrate improvements in anhedonia (fig. 3A). We next used the FST to confirm the antidepressant effects of ketamine in animals after SNI surgeries. One day after ketamine administration, animals with neuropathy showed a significant improvement in their immobility score, compared with those animals that received saline injections (fig. 4). In fact, the immobility score for SNI-operated animals after ketamine treatment was comparable with that of sham-operated animals.

Because ketamine also carries antinociceptive properties, we next asked whether the antidepressant effect of ketamine that we found at this dose (10 mg/kg) caused improvements in sensory hypersensitivity. We administered a single dose of ketamine (10 mg/kg) intraperitoneally to animals that had undergone SNI surgery 14 days previously. Because the half-life of ketamine is less than 1 h and its analgesic effects are short acting, we tested its analgesic effects 1 h and 1 day after injection. At this dose, we did not observe any changes in mechanical or cold hypersensitivity compared with saline (control) either 1 h after the injection, or 1 day after injection (fig. 5A through D). To further verify the lack of long-acting (more than 1 h) antinociceptive effects of ketamine, we also tested sensory hypersensitivities at two higher doses, 20 mg/kg and 50 mg/kg, 1 day after administration (we did not perform these tests 1 h after drug administration because sedative side effects of ketamine interfered with behavior tests). At these doses, we still did not observe any changes in mechanical or cold hypersensitivity compared with saline (fig. 5E and F). This lack of long-lasting analgesia contrasted with the enduring antidepressant effects that we observed up to 5 days after ketamine injection (fig. 3A). These results suggested that the antidepressant effects of ketamine we observed are independent of the antinociceptive effects of ketamine at low subanesthetic doses.

Next we sought to determine the dosage required to establish antidepressant effects for ketamine. We applied a series of doses – 1 mg/kg, 3 mg/kg, 10 mg/kg, and 20 mg/kg – and compared the antidepressant effects of these doses of ketamine with saline control using the SPT. As shown in figure 6, at 10 or 20 mg/kg, antidepressant effects were fully established statistically (P < 0.01 for 10 mg/kg, P < 0.05 for 20 mg/kg).

Depression and pain are important comorbidities.1,– 4In the current study, we used a rat neuropathic pain (SNI) model to test if pain causes depression-like behaviors. Our study had three key findings. First, both SNI and sham surgeries rapidly induced depression-like behaviors. Second, depression induced by SNI was chronic and concurrent with sensory hypersensitivity. Third, a single subanesthetic dose of ketamine provided quick and enduring treatment for neuropathy-induced depression without affecting sensory hypersensitivity.

Using behavioral despair (in FST) and anhedonia (in SPT) as indices for depression and mechanical and cold hypersensitivity as indices for pain, we found that both depression-like behaviors and pain quickly developed after SNI surgery and persisted up to 2 months. Surprisingly, sham-operated rats also developed reversible postoperative depression. Although postoperative depression has been described clinically,5–6,45–46the underlying mechanisms remain unknown. One possible cause is pain. In our model, it was difficult to test incisional pain after sham surgery using sensory tests, but the time course of depression that we observed (less than 7 days after surgery) correlated with the time course of acute postoperative pain in other models.47,–,49Other possible causes of depression include surgical stress and inflammation, but these were unlikely causes in our experiments as weight gain and stress hormone levels were unchanged among the test groups. In the future, it will be instructive to test depression in a postoperative pain model such as the paw incision model. At the same time, our results provided an explanation for studies by Kontinen et al.  21and Suzuki et al. ,22where affective changes such as anxiety and depression were not detected until 14 days after the spinal nerve ligation surgery. Thus, in these studies, the control group–sham-operated rats – may have also experienced affective changes in the initial postoperative period, and the difference in FST performance between the test group (spinal nerve ligation) and the control group (sham) may have been obscured as a result.

In rodent studies, pain has traditionally been tested by sensory hypersensitivity. Pain, however, also has cognitive and affective components. Previous studies on the depression-like response to pain have focused on the FST,21–22but the FST requires some motor competence. We observed that rats in the SNI group used their noninjured legs more often than the injured legs during the FST. This deviation did not affect our conclusions because we measured time of immobility, which did not require motor coordination; and ketamine did not treat neuropathy but reduced immobility associated with the SNI surgery. Increased immobility in the FST could be caused by spontaneous pain, which may make rats less likely to move, independent of depression. We cannot rule out this possibility, but we think this interpretation is unlikely. First, for spontaneous pain to increase the time of immobility in the FST, all SNI-treated rats would have had to experience spontaneous pain within the 5-min interval of the FST (fig. 2). Second, ketamine at high doses (50 mg/kg) did not provide relief for evoked pain (fig. 5), suggesting that at lower doses (10 mg/kg) it may not provide relief for spontaneous pain either. In contrast, ketamine (10 mg/kg) improved immobility in the FST (fig. 4). Therefore, we believe the increased immobility observed in the FST was unlikely caused by spontaneous pain, but rather, reflected depression-like behaviors. To further confirm this observation, we used SPT, another commonly used test for depression in rodents. Compared with FST, SPT has minimal requirements for motor competence, and findings are less likely confounded by motor deficits or spontaneous pain. Combining the SPT with the FST, we are the first group to show that depressive behaviors develop rapidly after surgery and persist in a chronic fashion with neuropathic pain. We argue that depressive behaviors are a key feature of pain in rats, and hence SPT and FST should be used routinely in pain assessments.

The molecular mechanisms for how pain causes depression are not well studied. However, pain has been shown to cause altered synaptic connectivity at the prefrontal cortex50and hippocampus,51as well as altered dopamine signaling from the ventral tegmental area.52–53These biochemical changes have been known to trigger negative symptoms of depression and may form the link between pain and depression.20,54,– 60 

An alternative explanation for our data is that surgeries induced depression and pain independently. Neuropathy from SNI may create chronic stress that causes depression. However, this is unlikely in our study. First, we did not observe any weight differences–a good measure for general well-being–among SNI-treated, sham-treated and naïve rats. Second, we did not detect differences in levels of the stress hormone corticosterone among the three groups. A number of studies have examined stress levels in neuropathic pain models and found no association between pain and chronic stress.13,22Another possible cause of depression from SNI is neuroinflammation. However, in the sham surgery, nerves were not dissected and we would not expect any neuroinflammation, but sham surgery still caused reversible depression. In our study of neuropathic pain, it was difficult to test the hypothesis that relieving pain would treat depression, because drugs used to treat neuropathic pain (opioids, antidepressants, or anticonvulsants) can alter performances in the FST and the SPT independently. In the future, it will be interesting to test depression using a reversible pain model, such as the acute postoperative pain or the inflammatory pain model.

Ketamine treats both depression and pain. Clinically, ketamine has not consistently demonstrated long-lasting analgesic effects at low doses. Rodent studies have shown more than 25 mg/kg is necessary for antinociception.40,–,42The duration of antinociception is also variable, but most often limited to within 24 h.27,41It is possible that ketamine relieved spontaneous pain at low doses, but higher doses are required to treat evoked pain (tested by mechanical and cold hypersensitivities). However, even at higher doses of ketamine (20 and 50 mg/kg), there was no relief of evoked pain 24 h after administration (fig. 5). Thus, ketamine did not likely provide long-lasting analgesia for evoked or spontaneous pain at subanesthetic doses. In contrast, both clinical and animal studies have shown that ketamine provides antidepressant effects that last from days to weeks.32,–,34At 10 mg/kg ketamine is known to treat depression38without causing psychomotor side effects.27,41We observed statistically significant relief of depression beginning 1 day after ketamine administration and lasting 5 days. Therefore, at this dose ketamine appeared to illustrate a distinction between the sensory and affective responses to pain. This result has threefold significance. First, it further validates the use of FST and SPT in pain studies. A concern with higher order behavioral tests is that their results may be confounded by spontaneous or movement-associated pain. Ketamine reduced immobility without changing hypersensitivity, implying that immobility in our study was not likely caused by the sensation of painful movement. Rather, it more likely reflected behavioral despair, an index for depression. The same is true for the SPT. The second significance of this finding is that the effective dosage (10 mg/kg) and time course (at least 5 days) of ketamine therapy in our model was highly compatible with other models of depression.32,38,61This suggests that the underlying mechanism of ketamine's antidepressant function may be conserved among different models of depression. Whereas analgesic properties of ketamine are mediated at spinal and peripheral level,26,–,28its antidepressant effects have been shown to be mediated at cortical and limbic areas.37–38Our results argue that ketamine likely ameliorates pain-induced depression at the supraspinal level. Third, our result shows that we can selectively target the depressive symptoms of pain without altering the sensory component. This concept has high clinical relevance, especially when the sensory component of pain is difficult to treat, such as the case with complex regional pain syndrome or phantom limb pain. In clinical studies, ketamine has been shown to improve mood in patients with phantom limb pain long after its analgesia effects have worn off.62Mood improvements are critical for physical rehabilitation during the postoperative period and for functional recovery among patients with chronic pain.46,63,– 65Ketamine may have an important role in this area of pain treatment, given its quick onset of action and enduring activity. Our preclinical data, therefore, advocate further investigation into the clinical use of ketamine as an effective and specific antidepressant in pain patients.

In conclusion, we found that pain rapidly and stably induces depression, but pain-induced depression can be treated by ketamine at low doses without treating the sensory input. These data argue that standard tests of depression such as the SPT and the FST should be applied to the battery of pain tests in rat studies. They also suggest that the affective component of pain can be selectively targeted successfully. Clinically, our results encourage careful evaluation and aggressive treatment of depression in both acute and chronic pain patients; they also show that low-dose ketamine may have a unique role in treating the depressive symptoms of neuropathic pain.