BDNF and learning: Evidence that instrumental training promotes learning within the spinal cord by up-regulating BDNF expression.

We have previously shown that the spinal cord is capable of learning a sensorimotor task in the absence of supraspinal input. Given the action of brain-derived neurotrophic factor (BDNF) on hippocampal learning, the current studies examined the role of BDNF in spinal learning. BDNF is a strong synaptic facilitator and, in association with other molecular signals (e.g. cAMP-response element binding protein (CREB), calcium/calmodulin activated protein kinase II (CaMKII) and synapsin I), important for learning. Spinally transected rats given shock to one hind leg when the leg extended beyond a selected threshold exhibited a progressive increase in flexion duration that minimized shock exposure, a simple form of instrumental learning. Instrumental learning resulted in elevated mRNA levels of BDNF, CaMKII, CREB, and synapsin I in the lumbar spinal cord region. The increases in BDNF, CREB, and CaMKII were proportional to the learning performance. Prior work has shown that instrumental training facilitates learning when subjects are tested on the contralateral leg with a higher response criterion. Pretreatment with the BDNF inhibitor TrkB-IgG blocked this facilitatory effect, as did the CaMKII inhibitor AIP. Intrathecal administration of BDNF facilitated learning when subjects were tested with a high response criterion. The findings indicate that instrumental training enables learning and elevates BDNF mRNA levels within the lumbar spinal cord. BDNF is both necessary, and sufficient, to produce the enabling effect.

Preemptive analgesia with lidocaine prevents Failed Back Surgery Syndrome.

Failed Back Surgery Syndrome (FBSS) is commonly encountered in pain-treatment settings in the United States. We tested whether potential key factors in this syndrome, such as extracellular concentrations of excitatory amino acids (EAAs), are increased in the dorsal horn by synaptic release due to unintentional stretch and/or deformation/compression/transection of dorsal spinal structures during surgery. We hypothesized that pharmacological nerve block as a form of preemptive analgesia prior to any insult to dorsal root neurons will prevent an abnormally high increase in extracellular concentrations of EAAs in the dorsal horn and ultimately the establishment of central sensitization during back surgery. The L4 and L5 dorsal roots were cut bilaterally near the spinal cord to provide an adequate model to test for preemptive analgesia. Amino acid concentrations were measured by dorsal horn microdialysis sampling; EAAs aspartate and glutamate were significantly increased by 80% and 65% respectively, as were other amino acids compared to sham control values. Topical application of 1% Lidocaine, a voltage-gated Na(+) channel blocker, for 10 min prior to L4 and L5 bilateral dorsal rhizotomy (BDR) significantly attenuated the increase in EAA concentrations such that their values were not different from sham controls. Behavioral tests demonstrated significant hindlimb mechanical allodynia after BDRs that was significantly attenuated by Lidocaine pretreatment. Thus, Lidocaine pretreatment could offer a safe measure for prevention of chronic pain for back surgical procedures if given by intramuscular injection, topical administration onto spinal nerves and/or the dorsal spinal surface during surgical procedures that include nerve entrapment release, intervertebral disc modification and laminectomies.

Nociceptive plasticity inhibits adaptive learning in the spinal cord.

Spinal plasticity is known to play a role in central neurogenic pain. Over the last 100 years researchers have found that the spinal cord is also capable of supporting other forms of plasticity including several forms of learning. To study instrumental (response-outcome) learning in the spinal cord, we use a preparation in which spinally transected rats are given shock to the hind leg when the leg is extended. The spinal cord rapidly learns to hold the leg in a flexed position when given this controllable shock. However, if shock is independent of leg position (uncontrollable shock), subjects fail to learn. Uncontrollable shock also impairs future learning. As little as 6 min of uncontrollable shock to either the leg or the tail generates a learning deficit that lasts up to 48 h. Recent data suggest links between the learning deficit and the sensitization of pain circuits associated with inflammation or injury (central sensitization). Here, we explored whether central sensitization and the spinal learning deficit share pharmacological and behavioral features. Central sensitization enhances reactivity to mechanical stimulation (allodynia) and depends on the N-methyl-d-aspartate receptor (NMDAR). The uncontrollable shock stimulus that generates a learning deficit produced a tactile allodynia (Exp. 1) and administration of the NMDAR antagonist MK-801 blocked induction of the learning deficit (Exp. 2). Finally, a treatment known to induce central sensitization, intradermal carrageenan, produced a spinal learning deficit (Exp. 3). The findings suggest that the induction of central sensitization inhibits selective response modifications.

Upregulation of the phosphorylated form of CREB in spinothalamic tract cells following spinal cord injury: relation to central neuropathic pain.

Spinal cord injury (SCI) often leads to the generation of chronic intractable neuropathic pain. The mechanisms that lead to chronic central neuropathic pain (CNP) following SCI are not well understood, resulting in ineffective treatments for pain relief. Studies have demonstrated persistent hyperexcitability of dorsal horn neurons which may provide a substrate for CNP. We propose a number of similarities between CNP mechanisms and mechanisms that occur in long-term potentiation, in which hippocampal neurons are hyperexcitable. One biochemical similarity may be activation of the transcription factor, cyclic AMP response element-binding protein (CREB), via phosphorylation (pCREB). The current study was designed to examine whether tactile allodynia that develops in segments rostral to SCI (at-level pain) correlates with an increase in CREB phosphorylation in specific neurons known to be involved in allodynia, the spinothalamic tract (STT) cells. This study determined that, in animals experiencing at-level allodynia 35 days after SCI, pCREB was upregulated in the spinal cord segment rostral to the injury. In addition, pCREB was found to be upregulated specifically in STT cells in the rostral segment 35 days after SCI. These findings suggest one mechanism of maintained central neuropathic pain following SCI involves persistent upregulation of pCREB expression within STT cells.

Shock-induced hyperalgesia: IV. Generality.

Brief-moderate shock (3, 0.75 s, 1.0 mA) has opposite effects on different measures of pain, inducing antinociception on the tail-flick test while lowering vocalization thresholds to shock and heat (hyperalgesia) and enhancing fear conditioned by a gridshock unconditioned stimulus (US). This study examined the generality of shock-induced hyperalgesia under a range of conditions and explored parallels to sensitized startle. Reduced vocalization thresholds to shock and antinociception emerged at a similar shock intensity. Severe shocks (3, 25 s, 1.0 mA or 3, 2 s, 3.0 mA) lowered vocalization threshold to shock but increased vocalization and motor thresholds to heat and undermined fear conditioned by a gridshock or a startling tone US. All shock schedules facilitated startle, but only brief-moderate shock inflated fear conditioning. The findings suggest that brief-moderate shock enhances the affective impact of aversive stimuli, whereas severe shocks attenuate pain.

Preserving and restoring behavioral potential within the spinal cord using an instrumental training paradigm.

We have shown that spinal cord neurons can support a simple form of instrumental learning. In a typical experiment, rats are spinalized at the second thoracic vertebra (T(2)) and given shock to one hindleg. One group (master) receives shock whenever the leg is extended. This response-contingent shock causes an increase in response duration that decreases net shock exposure. This instrumental learning is not observed in yoked controls that receive the same amount of shock independent of leg position (noncontingent shock). Interestingly, rats that have received noncontingent shock also fail to learn when they are subsequently exposed to response-contingent shock on either the ipsilateral or contralateral leg. Just 6 min of noncontingent nociceptive stimulation, applied to the leg or tail, undermines behavioral potential for up to 48 h. The present experiments explore whether a behavioral therapy can prevent and/or reverse this deficit. In experiment 1, spinalized rats received 30 min of training with contingent shock, noncontingent shock, or nothing prior to noncontingent tailshock. They were then tested with contingent shock to the contralateral hindleg. Rats that had received noncontingent shock alone failed to learn. Prior exposure to contingent shock had an immunizing effect that prevented the deficit. Experiment 2 examined whether training with contingent shock after noncontingent shock exposure would restore behavioral potential. To facilitate performance during contingent shock training, subjects were given an intrathecal injection of the opioid antagonist naltrexone, a drug treatment that temporarily blocks the expression of the behavioral deficit. Twenty-four hours later subjects were tested with contingent shock on either the ipsilateral or contralateral leg. We found that naltrexone combined with contingent shock therapy restored spinal cord function. Naltrexone alone had no effect. The results suggest that noncontingent nociceptive stimulation can undermine behavioral potential after spinal cord injury and that instrumental training can help preserve, and protect, spinal cord function.

Shock-induced hyperalgesia: III. Role of the bed nucleus of the stria terminalis and amygdaloid nuclei.

Rats exposed to a few moderately intense (1 mA) shocks subsequently exhibit lower vocalization thresholds to shock and thermal stimuli. They also exhibit facilitated learning in a Pavlovian conditioning paradigm. Together, these results suggest that shock exposure can enhance pain (hyperalgesia). The present study examined the role of the amygdala and bed nucleus of the stria terminalis (BNST), 2 systems that have been implicated in the induction and maintenance of negative affective states. Experiment 1 showed that lesions of the central, but not the basolateral, amygdala eliminate shock-induced hyperalgesia as measured by a decrease in vocalization thresholds to shock. Experiment 2 revealed that central nucleus lesions also prevent shock-induced sensitization of the vocalization response to heat. Anterior, but not posterior, BNST lesions had a similar effect.

Shock-induced hyperalgesia: II. Role of the dorsolateral periaqueductal gray.

Exposure to 3 moderately intense (1-mA) tailshocks has been shown to lower vocalization thresholds to both heat and shock. Previous shock exposure also facilitates the acquisition of conditioned fear as measured by freezing. These observations suggest that shock induces hyperalgesia (enhanced pain). This study explored whether shock-induced hyperalgesia depends on neurons within rostral or caudal portions of the dorsolateral periaqueductal gray (dlPAG). Experiment 1 examined the impact of dlPAG lesions on the acquisition of conditioned fear. Sham-operated rats demonstrated enhanced acquisition after shock exposure; both rostral and caudal lesions eliminated this effect. Experiment 2 showed that tailshock lowered vocalization thresholds to heat in sham-operated but not lesioned subjects. These results suggest that the dlPAG plays a critical role in the production of shock-induced hyperalgesia.

Shock-induced hyperalgesia: evidence forebrain systems play an essential role.

Exposure to a few moderately intense (1-mA) tailshocks has opposite effects on two measures of pain reactivity in rats. Tail-withdrawal to radiant heat is inhibited (antinociception) while vocalization thresholds are lowered (hyperalgesia) to both heat and shock (King et al., 1996). Prior work indicates that this hyperalgesia represents an unconditioned response and that it enhances the acquisition of both conditioned freezing and an avoidance response to thermal pain. The present experiments begin to explore the neural mechanisms that underlie hyperalgesia. Experiments 1 and 2 demonstrated that hyperalgesia is eliminated by both decerebration and pentobarbital anesthesia. Lesions limited to the frontal pole had a similar effect (Experiment 3). Experiment 4 showed that lesioning the frontal pole also disrupted the acquisition of conditioned fear.