Ketamine and its metabolite, (2R,6R)-HNK, restore hippocampal LTP and long-term spatial memory in the Wistar-Kyoto rat model of depression

Abstract

Accumulating evidence implicates dysregulation of hippocampal synaptic plasticity in the pathophysiology of depression. However, the effects of ketamine on synaptic plasticity and their contribution to its mechanism of action as an antidepressant, are still unclear. We investigated ketamine's effects on in vivo dorsal hippocampal (dHPC) synaptic plasticity and their role in mediating aspects of antidepressant activity in the Wistar-Kyoto (WKY) model of depression. dHPC long-term potentiation (LTP) was significantly impaired in WKY rats compared to Wistar controls. Importantly, a single low dose (5 mg/kg, ip) of ketamine or its metabolite, (2R,6R)-HNK, rescued the LTP deficit in WKY rats at 3.5 h but not 30 min following injection, with residual effects at 24 h, indicating a delayed, sustained facilitatory effect on dHPC synaptic plasticity. Consistent with the observed dHPC LTP deficit, WKY rats exhibited impaired hippocampal-dependent long-term spatial memory as measured by the novel object location recognition test (NOLRT), which was effectively restored by pre-treatment with both ketamine or (2R,6R)-HNK. In contrast, in WKYs, which display abnormal stress coping, ketamine, but not (2R,6R)-HNK, had rapid and sustained effects in the forced swim test (FST), a commonly used preclinical screen for antidepressant-like activity. The differential effects of (2R,6R)-HNK observed here reveal a dissociation between drug effects on FST immobility and dHPC synaptic plasticity. Therefore, in the WKY rat model, restoring dHPC LTP was not correlated with ketamine's effects in FST, but importantly, may have contributed to the reversal of hippocampal-dependent cognitive deficits, which are critical features of clinical depression. Our findings support the theory that ketamine may reverse the stress-induced loss of connectivity in key neural circuits by engaging synaptic plasticity processes to "reset the system".

Keywords: Antidepressant; HNK; Hippocampus; Ketamine; LTP; Memory and cognition; Model of depression; Synaptic plasticity; Wistar-Kyoto rat.

Fig. 1

Ketamine decreases FST immobility in WKY and WIS rats without affecting locomotor activity. (A, B) Average day2 FST immobility for saline/ketamine treated WKY (n = 14–16/group) and WIS (n = 6–7/group) rats. Stress-susceptible WKYs exhibited dramatic baseline immobility compared to WIS controls (p < 0.0001). a Ketamine significantly decreased FST immobility in WKY rats at 30 min and 24 h after injection compared to saline (***p < 0.0001), with effects being comparable at 30 min and 24 h. b Ketamine also significantly decreased FST immobility in WIS rats at 30 min and 24 h after injection compared to saline (**p < 0.0073 and **p = 0.0063), with effects being comparable at 30 min and 24 h. c, d Average total distance travelled in the OFT for saline and ketamine treated WKY and WIS rats (n = 6/group). WKY rats exhibited less general locomotor activity compared to WIS controls (p < 0.0001). Ketamine had no effects on general locomotor activity at 30 min or 24 h post-injection in (c) WKY or (d) WIS rats. * vs. SAL; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001

Fig. 2

WKYs exhibit a significant SC-CA1 LTP deficit, with normal basal synaptic transmission and LTD. a Representative fEPSP signals and average input-output (I/O) curves of fEPSP slope for WKY and WIS rats (n = 19–21 rats × 2 hemispheres /strain). SC-CA1 basal synaptic transmission was comparable between the two strains. b LTD recordings following LFS (3 Hz, 900pulses, 5 min (solid black line)) by strain (WKY n = 9, WIS: n = 14). Although fEPSP slope was transiently reduced 5 min post-LFS in WKY and WIS rats (p = 0.14, n.s. and p = 0.04), it effectively returned to baseline within 30 min in both strains. c, d Weak LTP recordings following HSF (1 × 100 Hz, 1 s (single black arrow)) by (c) strain and (d) time bin (WKY n = 21, WIS: n = 26). While there was a significant increase in fEPSP slope 5 min post-HFS in both strains (***p = 0.0006 and ***p < 0.0001), wLTP induction was significantly compromised in WKY compared to WIS rats (# # #p = 0.0007), and despite some decay in both strains (‡p = 0.03 and ‡p = 0.02), significant wLTP was still observed 30 min later in control WIS (***p < 0.0001) but not WKY rats (p = 0.41, n.s.; vs. WIS # # #p = 0.0005). e, f Saturated LTP recordings following HSF (4 × 100 Hz, 1 s, 5 min apart (4 black arrows)) by (e) strain and (f) time bin (WKY n = 12, WIS: n = 14). Although fEPSP slope increased immediately post-HFS in both strains (***p < 0.0001), significant sLTP was still observed 90 min later in control WIS (***p < 0.0001) but not WKY rats (p = 0.76, n.s.; due to significant sLTP decay ‡ ‡ ‡p = 0.0006; vs. WIS # # #p = 0.001). * vs. [1] = potentiation effect, ‡ vs. [2] = decay effect, # vs. WIS = strain effect; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001

Fig. 3

Ketamine restores SC-CA1 sLTP in WKY rats at 3.5 h but not 30 min after injection, with residual effects at 24 h. a, b WKY saturated LTP at 30 min post-injection by (a) drug treatment and (b) time bin (SAL: n = 12, KET: n = 5). Ketamine administration had no significant acute effects on WKY basal synaptic transmission. Although sLTP induction only reached significance in saline (***p < 0.0001) but not ketamine (p = 0.14, n.s.) treated rats, there were no significant differences between the two groups at any time point, and no significant sLTP was observed at 90 min as before, regardless of the treatment group (statistically significant decay in SAL rats only, ‡ ‡ ‡p = 0.0008). c, d WKY sLTP at 3.5 h post-injection by (c) drug treatment and (d) time bin (SAL: n = 19, KET: n = 18). Initially, significant sLTP was observed in both groups (***p < 0.0001); however, while potentiation again completely decayed in saline-treated WKYs (‡ ‡ ‡p < 0.0001), robust sLTP was still present at 90 min post-HFS in the ketamine group following only partial decay (***p < 0.0001; ‡ ‡p = 0.003; vs. SAL # # #p = 0.0003). e, f WKY sLTP at 24 h post-injection by (e) drug treatment and (f) time bin (SAL: n = 12, KET: n = 10). Significant sLTP was induced in both saline (***p = 0.0004) and ketamine (***p = 0.0009) treated WKYs; however, while the potentiation again completely decayed over the 90 min in the saline group (‡ ‡p = 0.0024), only partial decay was observed following ketamine (p = 0.17, n.s.), so that some sLTP was still present 90 min post-HFS (p = 0.097, n.s.). (G) WKY I/O curves at 24 h (SAL: n = 12, KET: n = 10, same rats as in (e, f)). fEPSP slope across all current intensities was consistently higher 24 h following ketamine compared to saline treatment, with the stimulation magnitude evoking ~ 50% of the maximal response effectively shifted leftward (from 100 μA to 60 μA) in ketamine-treated rats. h WIS sLTP at 3.5 h post-injection by drug treatment (SAL: n = 6, KET: n = 3). Ketamine had no significant acute effects on WIS basal synaptic transmission. Significant sLTP was observed in both saline p = 0.0003) and ketamine p < 0.0001) treated WIS rats, and independent of the treatment group, robust sLTP was still present 90 min later (SAL: p = 0.028 and KET p = 0.013). Single big arrow = drug injection, 4 small arrows = strong HSF protocol; * vs. [1] = potentiation effect, ‡ vs. [2] = decay effect, # vs. SAL = treatment effect; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001

Fig. 4

Ketamine restores NOLRT long-term spatial memory in WKY rats, without affecting performance in WIS controls. a In the NOLRT, following 3 days of habituation (H1–3), rats received 2 training sessions (T1 and T2, 10 min each), with two identical objects placed in two opposing arena corners. During the testing session (T3, 1 h or 24 h later for short or long -term memory), one object stayed at the familiar location (FL), while the other was moved to a new location (NL), where an NL preference (NL/NL + FL) of 60% or more indicates strong hippocampal-dependent spatial memory. In experiments involving drug treatment (purple arrows), saline, ketamine or (2R,6R)-HNK were injected 3.5 h before T2, i.e. 27.5 h before the 24 h testing session (T3). b Total NOLRT T3 exploration time (NL + FL) at 1 h or 24 h (n = 8–12/group) was comparable between the two strains. c % NL preference for the NOLRT test session for drug-free (1 h or 24 h) and drug-treated (saline or ketamine) WKY and WIS rats. Under drug-free conditions (1 h/24 h drug-naïve: n = 8–12/strain, same rats as in (b)), short-term memory (at 1 h) was equivalent between strains; however, long-term spatial memory (at 24 h) was significantly impaired in WKY compared to WIS rats (#p = 0.027; WKY 1 h vs. 24 h ‡ ‡p = 0.0037). In drug-treated rats (24 h SAL/KET WKY: n = 25/group and WIS: n = 10/group), while NOLRT location recognition memory at 24 h remained impaired in saline treated WKY compared to WIS rats (#p = 0.05), ketamine administration significantly facilitated long-term spatial memory compared to saline-treated WKYs (***p = 0.0008), effectively restoring performance to control WIS levels (~ 60% at 24 h). Ketamine had no effect on NOLRT performance in WIS controls. # vs. WIS = strain effect, ‡ drug-naïve 24 h vs. 1 h = time delay effect, * vs. SAL = treatment effect; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001

Fig. 5

(2R,6R)-HNK restores SC-CA1 sLTP and NOLRT long-term spatial memory without affecting FST immobility in WKY rats. a, b WKY sLTP at 3.5 h post-injection by (a) drug treatment and (b) time bin (SAL: n = 19, HNK: n = 8, +KET: n = 18 in (b) only). (2R,6R)-HNK had no significant acute effects on WKY basal synaptic transmission. Significant sLTP was initially present in all groups (***p < 0.0001); however, potentiation immediately post-HFS was significantly higher in HNK treated rats compared to saline # #p = 0.0015) but also ketamine (†p = 0.017). In addition, while significant decay was seen in all groups (p ≤ 0.003, not shown), robust sLTP was still present 90 min later in the HNK-treated group (***p < 0.0001, vs. SAL # # #p < 0.0001), which is now similar in magnitude to that following ketamine, with no sLTP in the saline group as expected. Single big arrow = drug injection, 4 small arrows = strong HSF protocol; * vs. [1] = potentiation effect, # vs. SAL and † vs. KET = treatment effects; c WKY % NL preference for the NOLRT test session (1 h/24 h drug-naïve: n = 9–12/group; SAL: n = 25, KET: n = 25 and HNK: n = 15). The WKY NOLRT long-term memory deficit (drug-naïve 24 h vs. 1 h, ‡ ‡p = 0.0037) was significantly reversed by (2R,6R)-HNK (vs. SAL *p = 0.012), as previously reported with ketamine (**p = 0.0017). ‡ drug-naïve 24 h vs. 1 h = time delay effect, * vs. SAL = treatment effect; d Average day2 FST immobility for saline (n = 7/group) and (2R,6R)-HNK (n = 9/group) treated WKY rats at 30 min and 24 h post-injection, indicating no effects of HNK in the FST. e Average total distance travelled in the OFT for saline and (2R,6R)-HNK treated WKY rats (n = 6/group) at 30 min or 24 h post-injection, where HNK had no effects on WKY general locomotor activity. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001