Psychopharmacology

, Volume 215, Issue 4, pp 689–695

CD-1 and Balb/cJ mice do not show enduring antidepressant-like effects of ketamine in tests of acute antidepressant efficacy

Authors

    • Department of PsychiatryHarvard Medical School—McLean Hospital
  • Karen L. Smith
    • Department of PsychiatryHarvard Medical School—McLean Hospital
  • Catherine S. John
    • Department of PsychiatryHarvard Medical School—McLean Hospital
  • Hannah H. Kang
    • Department of PsychiatryHarvard Medical School—McLean Hospital
  • William A. CarlezonJr.
    • Department of PsychiatryHarvard Medical School—McLean Hospital
  • Bruce M. Cohen
    • Department of PsychiatryHarvard Medical School—McLean Hospital
  • Dost Öngür
    • Department of PsychiatryHarvard Medical School—McLean Hospital
Original Investigation

DOI: 10.1007/s00213-011-2169-8

Cite this article as:
Bechtholt-Gompf, A.J., Smith, K.L., John, C.S. et al. Psychopharmacology (2011) 215: 689. doi:10.1007/s00213-011-2169-8

Abstract

Rationale

In patients, ketamine is a fast-acting antidepressant that can induce long-lasting symptom relief. Similar rapid effects have been reported in rodents, but reports of lasting effects are limited.

Objectives

We sought to extend past findings by examining dose–response curves that overlap with the individual doses previously reported to induce lasting effects in rodents and determining whether effects generalize to the tail suspension test (TST) and Balb/cJ mice.

Methods

Using common tests of antidepressant efficacy we first confirmed our ability to detect the effects of desipramine, a well-characterized antidepressant drug. Next, we sought to determine whether two non-competitive NMDA antagonists, ketamine and MK-801, had long-lasting antidepressant-like effects in CD-1 mice, a strain that has often been used to demonstrate the short-term antidepressant-like effects of ketamine. Finally, we examined the short- and long-term effects of ketamine in a mouse strain that is more sensitive to antidepressant-like effects, Balb/cJ mice.

Results

In CD-1 mice, desipramine treatment yielded significant short-term antidepressant-like effects in the TST and the forced swimming test (FST). However, no significant enduring effects of ketamine or MK-801 were observed 1 week later. Short-term effects of ketamine in the TST were observed in Balb/cJ mice, but lasting effects were absent 1 week later.

Conclusions

Although the TST and FST have been widely used to detect antidepressant-like effects in mice, they do not appear to be sensitive to long-lasting antidepressant-like effects of ketamine in mice and, therefore, do not model the therapeutic effects of ketamine that have been reported in humans with major depression.

Keywords

KetamineDepressionAntidepressantAnimal modelsBalb/cJCD-1Mice

Introduction

It was first reported in 2000 that the NMDA antagonist, ketamine, relieves depressive symptoms in patients experiencing a major depressive episode and that these effects lasted at least 3 days after a single intravenous (i.v.) treatment (Berman et al. 2000). Over recent years, numerous corroborating findings have been reported ranging from case studies to placebo-controlled double blind studies suggesting that the antidepressant effects of ketamine last at least several hours (Machado-Vieira et al. 2009; Phelps et al. 2009; Salvadore et al. 2010) and in many cases, days (Irwin and Iglewicz 2010; Kudoh et al. 2002; Liebrenz et al. 2007; Mathew et al. 2010; Okamoto et al. 2010; Paslakis et al. 2010; Price et al. 2009). Some reports suggest effects persisting for weeks (aan het Rot et al. 2010) to months (Correll and Futter 2006). These findings are exciting for two reasons: (1) the rapid pharmacologic actions of ketamine are entirely distinct from the delayed effects of standard antidepressants, and (2) many of these studies reported relief of symptoms in patients deemed treatment-resistant.

Antidepressant-like effects of ketamine have also been reported in rodents, especially using a standard test of antidepressant efficacy, the forced swimming test (FST). Many of these reports have demonstrated effects of ketamine across a wide range of doses (2.5–50 mg/kg) within 30–60 min of treatment (Sprague–Dawley rats; Engin et al. 2009; Wistar rats; Garcia et al. 2008; mice; Maeng et al. 2008; CD-1 mice; Mantovani et al. 2003; CD-1 mice; Popik et al. 2008; CD-1 mice; Rosa et al. 2003)—a timeframe in which traditional antidepressants are also effective in the same tests (Cryan et al. 2005a, b). The acute antidepressant-like effects of ketamine have proven robust across species and can be differentiated from the acute activating effects of the drug (Sprague–Dawley rats; Engin et al. 2009; CD-1 mice; Hayase et al. 2006). It has been more difficult to document persistent antidepressant-like effects of ketamine in rodents, although three reports using single doses of ketamine suggest antidepressant-like effects lasting 24 h (Sprague–Dawley rats; 10 mg/kg; Li et al. 2010), 10 days (Wistar rats; 160 mg/kg; Yilmaz et al. 2002) or 2 weeks (mice; 2.5 mg/kg; Maeng et al. 2008). In contrast, one report noted that the antidepressant-like effects of ketamine were not persistent in mice (C57/Bl/Han and Swiss) or rats (Wistar and Sprague–Dawley) across doses ranging from 1.25 to 160 mg/kg (Popik et al. 2008).

The mechanisms through which ketamine can elicit long-lasting antidepressant effects in patients are largely unknown and rodent models that are sensitive to these effects could facilitate the development of improved treatments. On the other hand, reports of lasting effects in rodents have been limited. Therefore, we sought to develop a reliable method to detect these effects using the tail suspension test (TST) and the FST in mice. After validating these tests with desipramine, a well-characterized standard antidepressant, we generated dose–response curves for lasting effects of ketamine with doses ranging from 0.5 to 160 mg/kg in CD-1 mice. We began looking for lasting effects in this strain because these mice have been commonly used in studies of ketamine’s short-term antidepressant-like effects (Hayase et al. 2006; Mantovani et al. 2003; Popik et al. 2008; Rosa et al. 2003). In parallel, we also tested MK-801, a drug that shares ketamine’s NMDA receptor antagonist effects. Next, we examined the short- and long-term effects of ketamine in the TST with doses ranging from 40 to 320 mg/kg in Balb/cJ mice, a strain that is reportedly vulnerable to depressive-like symptoms (Belzung and Griebel 2001; Dulawa et al. 2004; Popa et al. 2008; 2010; Trullas et al. 1989) and is particularly sensitive to the effects of some antidepressants (Crowley et al. 2005, 2006; Dulawa et al. 2004).

Materials and methods

Animals

Male CD-1 (also known as Swiss or ICR; n = 242) and Balb/cJ (n = 79) mice were obtained from Charles River Laboratories (Wilmington, MA) and Jackson Laboratory (Bar Harbor, ME), respectively. Eight-week-old mice were housed four per polycarbonate cage and maintained on a 12-h light–dark cycle with ad libitum access to food and water. Procedures were conducted with the approval of the McLean Hospital Institutional Animal Care and Use Committee and within the guidelines of The National Research Council’s Guide for the Care and Use of Laboratory Animals (1996).

Drugs

Desipramine HCl, ketamine HCl and (+)-MK-801 hydrogen maleate were acquired from Sigma-Aldrich (St. Louis, MO), dissolved in distilled water, and administered intraperitoneally (i.p.) at a concentration appropriate for a 10.0 ml/kg injection volume.

Tail suspension test

The TST was conducted as previously described (Crowley et al. 2004; Steru et al. 1985). Thirty minutes, 1 h, 3 h, or 7 days after injection, mice were secured to a vertical steel bar (bar size: 12.5 × 2.5 × 0.3 cm) with adhesive tape such that the base of the tail was aligned with the bottom of the bar leaving the body hanging freely below the bar with the head down. The 6-min sessions were videotaped and subsequently scored by an observer kept unaware of treatment conditions for the number of seconds spent immobile during the entire 6-min session.

Forced swimming test

The FST was conducted as previously described (Crowley et al. 2004) with minor modifications. Thirty minutes or 7 days after injection, mice were individually placed in polycarbonate cylinders (25.3 cm height × 22.2 cm diameter) filled with 15 cm of water (21°C). The sessions were videotaped and scored automatically for the number of seconds spent immobile during the 6-min session. Automated scoring was achieved using Noldus Ethovision software (Leesburg, VA), which scored immobility as movement below the velocity of 3 cm/s.

Procedures

Experiments 1–4 were conducted with CD-1 mice. CD-1 mice were selected for these studies because this strain was used in most of the published reports assessing antidepressant-like effects of ketamine in mice (Hayase et al. 2006; Mantovani et al. 2003; Popik et al. 2008; Rosa et al. 2003). In experiment 1, mice were injected with desipramine and subjected to the TST (0.0 or 5.0 mg/kg) or FST (0.0 or 20.0 mg/kg) 30 min later. In experiments 2 (covering lower ketamine doses: 0.0, 0.5, 2.5, or 12.5 mg/kg) and 3 (covering higher ketamine doses: 0.0, 40.0, 80.0, or 160.0 mg/kg), CD-1 mice were injected with ketamine and subjected to the TST or FST 7 days later. In experiment 4, mice were injected with MK801 (0.125, 0.25, 0.5 mg/kg) and subjected to the TST 7 days later. Mice were tested 1 week after treatment to capture the possible lasting effects of ketamine that have been previously reported in mice (Maeng et al. 2008) and rats (Yilmaz et al. 2002).

Experiments 5 and 6 were conducted in Balb/cJ mice because they exhibit high immobility (Trullas et al. 1989) and high responsiveness to antidepressants in the TST (Crowley et al. 2005; 2006; Dulawa et al. 2004) and because they have been characterized as a “highly emotional” strain (Belzung and Griebel 2001; Dulawa et al. 2004; Popa et al. 2008; 2010). In Experiment 5, mice were injected with ketamine (0.0, 40.0, 80.0, or 160.0 mg/kg) and tested in the TST 1 h and 7 days later. Testing at both proximal and distal time points was performed to mimic published reports using the FST where repeated testing was employed (Maeng et al. 2008; Yilmaz et al. 2002) and therefore to test the hypothesis that performance in an initial test could influence performance in a later test. Because there was some suggestion in experiment 5 that 160 mg/kg of ketamine had the potential to produce long-lasting antidepressant-like effects, mice in experiment 6 were injected with this and higher doses of ketamine (0.0, 160.0, 240.0, or 320.0 mg/kg) and tested in the TST 3 h and 7 days later. To prevent the likelihood of our findings being confounded by nonspecific changes in locomotion, the proximal times for experiments 5 and 6 were chosen, based on initial observations, to coincide with normalization of ketamine’s effects on gross motor activity. Specifically, individuals experienced in the observation of mouse behavior passively observed mice at 10-min intervals after vehicle or ketamine treatment and determined that activity was normalized to vehicle levels after 1 h for the anesthetic dose range and 3 h for the higher dose range.

Statistical procedures

Data from experiment 1 (desipramine) were analyzed using a t test. Experiments 2–4 (ketamine, MK-801, CD-1 mice) were analyzed using one-way ANOVAs for the between subjects effects of dose. Experiments 5 and 6 (ketamine, Balb/cJ mice) were analyzed using mixed two-way ANOVAs with dose as the between subjects factor and time as the within subjects factor. Significant ANOVAs were followed-up with Bonferroni corrected pair-wise comparisons.

Results

Experiment 1: Proximal effects of desipramine in the TST and FST in CD-1 mice

As expected, desipramine treatment significantly reduced immobility in CD-1 mice in the TST and FST when given 30 min prior to testing. This conclusion was supported by significant t tests [TST: t(18) = 2.34, p < 0.05; FST: t(22) = 2.24, p < 0.05] demonstrating that vehicle-treated mice displayed greater immobility compared to desipramine treated mice (Table 1).
Table 1

Effects of desipramine, ketamine and MK801 in CD-1 mice

TST

Desipramine (proximal: treatment 30 min prior to test)

Dose (mg/kg)

Vehicle

5

  

Immobility (s)

103.80

51.90a

  

SEM

18.00

13.00

  

Low-dose ketamine (distal: treatment 7 days prior to test)

Dose (mg/kg)

Vehicle

0.5

2.5

12.5

Immobility (s)

96.50

81.33

94.90

98.80

SEM

15.51

12.72

5.59

13.74

Anesthetic-dose ketamine (distal: treatment 7 days prior to test)

Dose (mg/kg)

Vehicle

40

80

160

Immobility (s)

110.90

80.80

109.40

138.64b

SEM

10.90

12.37

14.31

17.16

MK-801 (distal: treatment 7 days prior to test)

Dose (mg/kg)

Vehicle

0.125

0.25

0.5

Immobility (s)

109.95

133.57

123.96

113.82

SEM

11.97

20.29

18.68

14.62

FST

Desipramine (proximal: treatment 30 min prior to test)

Dose (mg/kg)

Vehicle

20

  

Immobility (s)

132.66

108.91a

  

SEM

9.84

3.96

  

Low-dose ketamine (distal: treatment 7 days prior to test)

Dose (mg/kg)

Vehicle

0.5

2.5

12.5

Immobility (s)

126.54

129.61

124.23

113.67

SEM

11.07

16.31

14.39

20.44

Anesthetic-dose ketamine (distal: treatment 7 days prior to test)

Dose (mg/kg)

Vehicle

40

80

160

Immobility (s)

108.07

119.02

108.24

141.32

SEM

13.50

12.66

13.07

13.66

Significance for a and b are both p < 0.05

aDifferent from vehicle

bDifferent from 40 mg/kg

Experiment 2: Distal effects of low (subanesthetic) doses of ketamine in the TST and FST in CD-1 mice

As analyzed by a one-way ANOVA, low doses of ketamine failed to induce long-lasting antidepressant effects in the TST or FST. Specifically, ketamine failed to decrease immobility at any dose relative to vehicle treated animals 1 week after treatment (Table 1).

Experiment 3: Distal effects of ketamine doses within the anesthetic range in the TST and FST in CD-1 mice

As shown in Fig. 1, doses of ketamine within the anesthetic range did not significantly reduce immobility in the TST 1 week after treatment. However, a one-way ANOVA revealed a significant effect of dose [F(3,40) = 0.38, p < 0.05]. Follow-up Bonferroni comparisons demonstrated that the 160 mg/kg ketamine dose group exhibited higher immobility than the 40-mg/kg-dose group 1 week after treatment, but no other significant differences were observed. Analysis of ketamine’s effects in the anesthetic range in the FST revealed a similar, though non-significant, trend for the highest dose to induce greater immobility compared to vehicle treated mice.
https://static-content.springer.com/image/art%3A10.1007%2Fs00213-011-2169-8/MediaObjects/213_2011_2169_Fig1_HTML.gif
Fig. 1

Ketamine doses within the anesthetic range did not systematically affect immobility in the TST 7 days after treatment in CD-1 mice. Bars represent mean (+SEM) time immobile during the 6-min test. Follow-ups to the significant one-way ANOVA revealed that mice treated with 160 mg/kg demonstrated significantly greater immobility compared to mice treated with 40 mg/kg (*p < 0.05)

Experiment 4: Distal effects of MK-801 in the TST in CD-1 mice

Like ketamine, the non-competitive NMDA antagonist, MK-801 failed to elicit long lasting reductions in immobility in the TST 1 week after treatment. This conclusion was based on a non-significant one-way ANOVA for MK-801 dose (Table 1).

Experiment 5: Proximal and distal effects of ketamine doses within the anesthetic range in the TST in Balb/cJ mice

Figure 2 shows that ketamine treatment elicited an antidepressant-like effect 1 h after treatment, but did not elicit significant effects 1 week later. A two-way ANOVA revealed significant main effects of time [F(1,36) = 7.44, p < 0.01], dose [F(3,36) = 12.38, p < 0.001], and their interaction [F(3,36) = 5.43, p < 0.005]. Follow-up Bonferroni comparisons revealed that the 160-mg/kg-dose group displayed significantly lower immobility 1 h after treatment, but no other significant effects. Although there was a tendency toward the 160-mg/kg-dose group showing less immobility at 7 days, this comparison did not approach significance (p = 0.29). These findings suggest that ketamine within this range may have short-term antidepressant-like effects in the TST, but that these effects are not long lasting.
https://static-content.springer.com/image/art%3A10.1007%2Fs00213-011-2169-8/MediaObjects/213_2011_2169_Fig2_HTML.gif
Fig. 2

Ketamine doses within the anesthetic range reduced immobility in the TST at 1 h, but not at 7 days after treatment in Balb/cJ mice. Bars represent mean (+SEM) time immobile during the 6-min test. Follow-ups to the significant two-way interaction revealed that mice treated with 160 mg/kg demonstrated significantly less immobility at 1 h compared to all other groups at the 1 h time-point (*p < 0.001)

Experiment 6: Proximal and distal effects of high doses of ketamine in the TST in Balb/cJ mice

Ketamine doses in a much higher range, likewise, failed to elicit short- or long-term antidepressant like effects. This suggestion was supported by a two-way ANOVA, which yielded no significant main effects or interactions (Table 2).
Table 2

Effects of ketamine on TST behavior in Balb/cJ mice tested at both proximal and distal time-points relative to ketamine treatment in the same mice

Anesthetic-dose ketamine (proximal: treatment 1 h prior to test)

Dose (mg/kg)

Vehicle

40

80

160

Immobility (s)

142.10

114.40

124.70

44.40a

SEM

6.33

9.06

11.90

11.06

Anesthetic-dose ketamine (distal: treatment 7 days prior to test)

Dose (mg/kg)

Vehicle

40

80

160

Immobility (s)

133.50

134.70

124.50

102.40

SEM

10.85

11.40

9.27

11.37

High-dose ketamine (proximal: treatment 3 h prior to test)

Dose (mg/kg)

Vehicle

160

240

320

Immobility (s)

106.40

92.60

97.50

90.22

SEM

9.85

10.30

9.38

10.43

High-dose ketamine (distal: treatment 7 days prior to test)

Dose (mg/kg)

Vehicle

160

240

320

Immobility (s)

88.10

107.60

126.80

97.67

SEM

15.21

8.66

8.16

18.06

Significance for a is p < 0.001

aDifferent from all other groups at the 1 h time-point

Discussion

Here, we confirm the short-term antidepressant-like effects of ketamine using the TST in mice, but were unable to demonstrate long-lasting antidepressant-like effects 7 days after ketamine treatment in the TST or FST. These studies included doses of ketamine ranging from 0.5 to 360 mg/kg, tested in two tests of antidepressant efficacy and in two mouse strains.

The findings presented here are in agreement with numerous reports that ketamine can produce short-term antidepressant effects in humans (Machado-Vieira et al. 2009; Phelps et al. 2009; Salvadore et al. 2010) and rodents (Engin et al. 2009; Garcia et al. 2008; Maeng et al. 2008; Mantovani et al. 2003; Popik et al. 2008; Rosa et al. 2003). Based on initial observations of motor behavior that was normalized by the time of the proximal test, it is likely that this effect is independent of the acute activating effects of ketamine, which was demonstrated by Engin et al. (2009). However, the antidepressant-like effect of ketamine observed at 1 h after administration in experiment 5 was not replicated in experiment 6 where the time between ketamine treatment and testing in the TST was extended to 3 h, suggesting that the antidepressant-like effects of ketamine in the TST may be very short-lived.

The current findings are in agreement with a recent report (Popik et al. 2008), which suggested that ketamine does not have lasting effects in the mouse TST or the mouse or rat FST, but differ from two reports indicating antidepressant-like effects of ketamine in the FST persisting 2 weeks after administration in mice (Maeng et al. 2008) and at 3, 7, and 10 days in rats (Yilmaz et al. 2002). Although much of the work presented here are from the TST, our lack of agreement with some rodent studies is not explained entirely by the test that we used, since we also did not detect lasting ketamine effects in the FST.

It is possible that the lasting antidepressant-like effects of ketamine are specific to certain mouse strains. In conducting these studies, we began with CD-1 mice because they were the strain that had been most commonly used to examine the short-term antidepressant-like effects of ketamine (Hayase et al. 2006; Mantovani et al. 2003; Popik et al. 2008; Rosa et al. 2003) and the single report of long-term antidepressant-like effects in mice did not indicate the mouse strain that was used (Maeng et al. 2008). C57 mice were used in that study (H. Manji, personal communication, August 18, 2010), so it is possible that this apparent discrepancy in results is explained by strain differences. It is not uncommon for individual mouse strains to be more or less sensitive to antidepressants with different mechanisms of action (for example Crowley et al. 2005; 2006). Thus, the reported lasting effects of ketamine may be specific to the C57 versus the CD-1 or Balb/cJ mice strains. It is worth highlighting that a lasting effect of a single dose of ketamine was also reported in rats (Yilmaz et al. 2002). Still, it is possible that effects generalize across species (rat and mouse), but not across closely related inbred laboratory mouse strains (C57 and Balb). The fact that long-term antidepressant effects of ketamine are not always detectable raises the possibility that other, more subtle factors may be involved in producing and reversing depressive-like behaviors in mice and rats, particularly in long-term studies.

Although numerous, reports of ketamine’s effects in humans have been noted to be limited to single cases or partially controlled studies. Some (e.g., Popik et al. 2008) have suggested that the profound psychoactive effects of ketamine preclude truly controlled clinical studies. Accordingly, it is possible that at least some of the effects reported in humans are not dependent upon ketamine’s pharmacologic antidepressant activity but upon nonspecific psychological placebo effects. Future clinical work including a psychoactive placebo that is not expected to produce antidepressant effects might address this issue.

Alternatively, these data may represent a disassociation between effects observed in humans and laboratory mice, which is based on the limitations of the standard tests of antidepressant efficacy or on species differences in metabolism or drug effect. The TST and FST are sometimes criticized with the suggestion that they are not likely to detect novel antidepressants that act through mechanisms that are different from those in current use, because these tests were established based on the effects of known antidepressants. Indeed, it has been shown that false negatives are possible with the FST (Cryan et al. 2005b) and this explanation cannot be ruled out. Metabolic species differences could also underlie this apparent dissociation. For example, the half-life of ketamine is just 13 min in a variety of mouse lines (Maxwell et al. 2006) compared to 1.4–3 h in humans (McLean et al. 1996). Methods like those employed by Majewski-Tiedeken et al. (2008) to sustain the acute effects of ketamine may be required to reliably mimic effective human treatments in mice.

In summary, the present studies confirm that ketamine has short-term antidepressant-like effects in the TST, but were unable to identify long-term effects of ketamine in CD-1 mice in the TST or FST nor the TST in Balb/cJ mice. While negative results cannot be considered ultimate proof, these systematic data contribute to the growing body of data suggesting that rodent tests of antidepressant efficacy do not invariably detect the lasting antidepressant effects of ketamine that are reported in humans with depression.

Copyright information

© Springer-Verlag 2011