T-type Ca2+ channels are known to be important regulators of pain transmission in primary afferent sensory neurons and the spinal cord [1]. Among the three isoforms of T-type Ca2+ channels that are expressed in the mammalian genome, the Cav3.2 channel isoform appears to be the predominant T-type channel subtype involved in this process [2]. It is expressed in a subpopulation of primary afferent fibers and the spinal dorsal horn [3], and its expression is enhanced in these tissues in a wide range of chronic pain conditions in rodents [1]. Consequently, systemic or intrathecal delivery of T-type channel inhibitors mediates analgesia (for review see [1, 4]. T-type channels can be inhibited by different types of endocannabinoids [5], terpenes [6] and phytocannabinoids such as cannabidiol and Δ9-THC [7, 8]. In particular, Δ9-THC mediates strongly state dependent inhibition of Cav3.2 channels with a preference for binding to inactivated channels [7, 8]. It is known that spinally delivered Δ9-THC inhibits mechanical and cold allodynia in models of neuropathic pain [9], and analgesia exerted by Δ9-THC delivered to the brain involves modulation of both CB1 and CB2 receptors [10]. However, it is unclear whether the spinal actions of Δ9-THC involve T-type channels, cannabinoid receptors, or a combination thereof. Thus, the present study was designed to investigate contributions of spinal CB receptor subtypes and Cav3.2 channels on the antihyperalgesic effect of spinally delivered Δ9-THC. All experiments were carried out with approval of an animal protocol by the Institutional Animal Care and Use Committee, and all efforts were made to minimize animal suffering according to the policies and recommendations of the International Association for the Study of Pain. Δ9-THC was delivered by intrathecal injection as described before [6, 8] into male and female C57BL/6J (wild-type), or male CB1 null, CB2 null, or Cav3.2 null mice (20 − 25 g, 8–10 weeks; Jackson Laboratories). We first assessed the analgesic action of spinally delivered Δ9-THC in the acute nociceptive (phase 1) and inflammatory pain (phase 2) phases of a standard formalin test [11]. Intrathecally delivered Δ9-THC, 20 min before testing, significantly and dose-dependently reduced the duration of nocifensive responses in the first (Fig. 1a) and second (Fig. 1b) phases of the formalin test. Next, we verified whether spinal Δ9-THC was also able to inhibit mechanical hyperalgesia caused by chronic neuropathy 21 days after partial sciatic nerve injury which was performed as described by us before [11]. Mechanical hyperalgesia was measured using a Dynamic Plantar Aesthesiometer (Ugo Basile, Varese, Italy). When compared to the neuropathic control group, treatment of mice with Δ9-THC (10.0 µg/i.t.), but not vehicle (PBS, 10 µl/i.t.) produced marked anti-hyperalgesia when evaluated 45 min after treatment (Fig. 1c). These data show that Δ9-THC mediates robust analgesia in wild type mice. Next, we investigated the effect of Δ9-THC in a model of persistent inflammatory pain. 20 μl of Complete Freund's Adjuvant (CFA) were given intraplantarly (i.pl.) in the ventral surface of the right hindpaw, whereas sham groups received 20 μl of PBS. Thermal hyperalgesia was examined by measuring the latency to withdrawal of ipsilateral hind paws in response to a focused beam of radiant heat (IR = 30) using a plantar test apparatus (UgoBasile, Varese, Italy). Two days after CFA injection, intrathecal treatment with Δ9-THC (10 µg/i.t) but not with vehicle (10 µl/i.t.) resulted in anti-hyperalgesia that remained significant up to 3 h (Fig. 1d). Δ9-THC was also effective in increasing paw withdrawal latencies when delivered to female mice (tested 45 min after its spinal delivery, Fig. 1e). To determine whether the analgesic effects observed for spinally delivered Δ9-THC were mediated by cannabinoid CB1 or CB2 receptors, we repeated the CFA model using male CB1 (Fig. 1f) and male CB2 (Fig. 1g) null mice in comparison with male wild-type mice that were simultaneously tested. For this purpose, mice were injected intrathecally with either vehicle (control) or Δ9-THC (10.0 µg/i.t.) and tested 45 min later. Similar to wild type animals, Δ9-THC produced significant analgesic effects, indicating that neither of these two receptors are essential for the observed analgesia even though this compound is an agonist of both receptor types [12].

Fig. 1
figure 1

Δ9-THC produces spinal analgesia in mice that is Cav3.2 channel-dependent. Dose response action of Δ9-THC (delivered 20 min before formalin) in the a first and b second phases of the formalin test. Each bar represents the mean of 5 animals, error bars denote S.E.M. Data are representative of 2 independent sets of experiments. Statistical analyses were performed by two-way ANOVA followed by Tukey's test. Asterisks denote a significant difference, **P < 0.01 and ***P < 0.001 when compared with the control groups. c Mechanical threshold of PSNI mice 45 min after treatment with Δ9-THC (10 ug/i.t.). Bars represent the mean of 7 animals, error bars denote S.E.M. Data are representative of 2 independent sets of experiments. Two-way ANOVA followed by Tukey's test revealed significance, ###P < 0.001 and ****P < 0.0001 when compared with the control groups. d Time-course of the effect of Δ9-THC (10 μg/i.t.) on thermal withdrawal latencies of CFA-injected male mice. e Effect on CFA-treated female mice when evaluated 45 min following treatment. In d and e, error bars are S.E.M. Data are representative of 2 independent sets of experiments Two-way ANOVA followed by Tukey's test revealed statistical differences, *P < 0.05 **P < 0.01 or ***P < 0.001 when the CFA + treated group is compared with the CFA + vehicle control group, and ###P < 0.001 when the PBS group is compared with the  control groups. f, g Comparison of the effect of 10 μg/i.t. Δ9-THC on CFA-injected wild type and f CB1, g CB2, and h Cav3.2 knockout mice. Each bar represents the mean of 6–10 mice, error bars are S.E.M. Data are representative of 2 independent sets of experiments. Two-way ANOVA followed by Tukey's test revealed statistical differences, *P < 0.05 or ****P < 0.0001 when compared with the control group

Full size image

We then tested the analgesic effect of spinal Δ9-THC in Cav3.2 null mice. These mice develop CFA-induced hypersensitivity despite the absence of Cav3.2 channels [11], most likely due to compensatory mechanisms that are not fully understood. As shown in Fig. 1h, Δ9-THC lost its analgesic effects when delivered to Cav3.2 null mice, indicating that the key biological target for spinally delivered Δ9-THC are T-type channels.

There is considerable evidence that CB1 receptor activation mediates analgesia [13], however there are also reports that the analgesic activity of Δ9-THC is lost in CB receptor null mice [10, 14]. We do not challenge a possible involvement of these receptors when Δ9-THC is delivered systemically. Our focus was to specifically isolate a spinal effect, and this can be cleanly accomplished by the intrathecal route of delivery used in our study (Additional file 1: Fig S1). What we do not know is the overall contribution of the spinal action to the overall analgesic properties of Δ9-THC. We attempted testing the effect of systemically delivered Δ9-THC in Cav3.2 null mice, however, we found that these mice became lethargic, thus confounding the types of pain behavioral measurements that we typically perform. Finally, our laboratory has previously reported that the analgesic effect of intrathecally delivered mixed CB receptor/Cav3.2 ligands are abolished in Cav3.2 null mice, but they retain activity upon blocking CB1 receptors with AM-281 [15]. Interestingly, inhibition of CB2 receptors with AM-630 did attenuate the analgesic effects of these compounds and we concluded that although CB2 receptors may be involved in their actions, this may be due to CB2 receptor modulation of Cav3.2 channel activity. Hence, we cannot rule out the possibility that Δ9-THC might activate spinal CB2 receptors which may in turn inhibit Cav3.2 in addition to the direct inhibitory actions of Δ9-THC on these channels. We note that CB1 receptors do not functionally inhibit Cav3.2 in heterologous systems [8] but we could at that time not explore such coupling for CB2 receptors for technical reasons. Nonetheless, even if CB2 receptors augment direct inhibition of Cav3.2 channels, our results clearly implicate Cav3.2 channels as an essential target of Δ9-THC in the actions of spinal Δ9-THC as an analgesic, whereas CB receptors are not required.