Updating perspectives on spinal cord function: motor coordination, timing, relational processing, and memory below the brain

Abstract

Those studying neural systems within the brain have historically assumed that lower-level processes in the spinal cord act in a mechanical manner, to relay afferent signals and execute motor commands. From this view, abstracting temporal and environmental relations is the province of the brain. Here we review work conducted over the last 50 years that challenges this perspective, demonstrating that mechanisms within the spinal cord can organize coordinated behavior (stepping), induce a lasting change in how pain (nociceptive) signals are processed, abstract stimulus-stimulus (Pavlovian) and response-outcome (instrumental) relations, and infer whether stimuli occur in a random or regular manner. The mechanisms that underlie these processes depend upon signal pathways (e.g., NMDA receptor mediated plasticity) analogous to those implicated in brain-dependent learning and memory. New data show that spinal cord injury (SCI) can enable plasticity within the spinal cord by reducing the inhibitory effect of GABA. It is suggested that the signals relayed to the brain may contain information about environmental relations and that spinal cord systems can coordinate action in response to descending signals from the brain. We further suggest that the study of stimulus processing, learning, memory, and cognitive-like processing in the spinal cord can inform our views of brain function, providing an attractive model system. Most importantly, the work has revealed new avenues of treatment for those that have suffered a SCI.

Keywords: ionic plasticity; learning; metaplasticity; pain; plasticity; recovery; spinal cord injury.

Figure 1

Anatomy of the spinal cord. (A) Gross anatomy of the spinal cord. Cross-sections of the spinal cord illustrating major structures (B), functional organization (C), and laminae (D) Adapted from Grau et al. (2022). (E) Research suggests that the central pattern generator (CPG) that drives the rhythm of stepping lies in the rostral lumbar region (L1-L2; Cazalets et al., 1995; Magnuson et al., 1999). The structures needed for instrumental learning, and that underlie the development of a learning deficit after uncontrollable stimulation, lie within the lower lumbosacral (L3-S2) spinal cord (Liu et al., 2005).

Figure 2

The release of GABA can have either an inhibitory (hyperpolarizing) or excitatory (depolarizing) effect depending upon the intracellular concentration of Cl⁻. (A) The co-transporters KCC2 and NKCC1 regulate the outward and inward flow of Cl⁻, respectively. In adult animals (right), the outward flow of Cl⁻ through the KCC2 channel maintains a low concentration of the anion within the cell. Under these conditions, engaging the GABA-A receptor allows Cl⁻ to enter the cell, which has a hyperpolarizing effect. Early in development, and after a rostral SCI, the levels of KCC2 are much lower and, as a consequence, there is a rise in the intracellular concentration of Cl⁻. Now, engaging the GABA-A receptor allows Cl⁻ to exit the cell, which has a depolarizing effect. (B) Nociceptive stimulation (input) will engage GABAergic neurons within the spinal cord that regulate neural excitability. In adult uninjured animals, the low intracellular concentration of Cl⁻ will cause GABA to have an inhibitory effect, which will dampen neural excitability. After injury, the reduction in membrane-bound KCC2 would transform how GABA release affects nociceptive circuits, causing it to have a depolarizing [excitatory (+)] effect that could contribute to the development of nociceptive sensitization and spasticity. Excitatory (glutamatergic) transmitters are indicated in blue and inhibitory (GABAergic) transmitters are colored red. Adapted from Grau et al. (2014).

Figure 3

Nociceptive stimulation engages neurons within the spinal cord that release the neurotransmitter glutamate (Glu), engaging signal pathways implicated in plasticity. Akt, protein kinase B; AMPAR, a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor; BDNF, brain-derived neurotrophic factor; CaMKII, calcium/calmodulin activated protein kinase II; ERK, extracellular signal-regulated kinase; GluR2, glutamate receptor 2; IL-1b, interleukin-1 beta; IP3, inositol 1,4,5-trisphosphate; mGluR, metabotropic glutamate receptor; NMDAR, N-methyl-D-aspartate receptor; PKC, protein kinase C; PLC, phospholipase C; TrkB, tropomyosin receptor kinase B; TNF, tumor necrosis factor; TNFR1, TNF receptor 1. Adapted from Grau et al. (2014).

Figure 4

Methods for instituting a Pavlovian (S1→S2) and instrumental (R→O) relation. (A) In rats that have received a rostral (T2) transection, pairing electrical stimulation of the peroneal nerve [S2; the unconditioned stimulus (US)] with weak stimulation of the saphenous nerve [S1; the conditioned stimulus (CS)] amplifies the response elicited by S1 relative to animals that experience S1 and S2 in an unpaired manner (Durkovic, 2001). (B) Electrical stimulation of the motor cortex (S1) can engage surviving descending (corticospinal) fibers after SCI. Pairing S1 with epidural stimulation, which engages sensory afferents, in a Pavlovian manner (open-loop stimulation) enhances motor performance after SCI (Harel and Carmel, 2016). (C) Spinally transected rats (Master) that receive noxious electrical stimulation of the tibialis anterior muscle [the outcome (O)] whenever the leg is extended [the response (R)] exhibit a progressive increase in flexion duration that reduces net exposure to the noxious stimulus. Animals that receive stimulation independent of leg position (Yoked) do not exhibit a change in flexion duration (Grau et al., 1998). (D) An instrumental (R-O) relation can also be established using electrophysiological methods (closed-loop stimulation). For example, after SCI, surviving corticospinal neurons can evoke a small evoked (electrical) muscular response (the R). Stimulating the motor neurons (the O) when a R is detected can strengthen motor performance after SCI (McPherson et al., 2015). Error bars indicate the standard error of the mean.

Figure 5

A neural-functionalist perspective on Pavlovian conditioning. It is assumed that environmental relations can be encoded by multiple mechanisms within the organism, which can be distinguished by their functional properties. It is likewise assumed that a functional mechanism can be neurally encoded in multiple ways and that a particular biological mechanism (e.g., NMDA receptor-mediated plasticity) can be enlisted by multiple processes. Adapted from Grau and Joynes (2005a).

Figure 6

Intermittent stimulation can have distinct effects on spinal cord plasticity depending upon the underlying temporal relation. (A) When the interval between stimuli is randomly varied [variable time (VT)], a learning impairment is observed when animals are tested after 180–900 stimuli (Deficit). If stimuli occur in a regular manner [fixed time (FT)], a learning deficit is observed when animals are exposed to 180–360 stimuli. Exposure to additional stimulation (540–900) has a restorative effect that counters the learning deficit. (B) A restorative effect emerges when the locus of FT stimulation is alternated across regions of the body (e.g., hind leg and tail). (C) Temporally displacing alternating stimuli by a small amount preserves the regularity of stimuli applied at each site. Displacing the stimuli in opposite directions introduces an irregular relation (incoherent) across sites and produces a learning deficit. If the stimuli are displaced in the same direction (coherent), a regular pattern can be abstracted across sites and a restorative effect emerges. Regularity can also be abstracted when the site of stimulation is randomly varied across sites (D) and when half of the stimuli are randomly omitted (E), provided the stimuli remain in phase (FT 50%-Unshifted). Shifting the phase relation after a stimulus is omitted (FT 50%-Shifted) disrupts the abstraction of regularity, causing the same number of stimuli to induce a learning deficit. Adapted from Lee et al. (2015), Lee et al. (2016), and Grau et al. (2022).

Figure 7

A schematic illustrating the intra-spinal processes that mediate instrumental conditioning, timing, and the consequences of uncontrollable stimulation. Evidence suggests that neural processes within the caudal lumbosacral (L3-S2) spinal cord enable sensory-motor integration (lower box). The effect of noxious stimulation appears to be gated by proprioceptive cues related to limb position; if the leg(s) is flexed, stimulation has no impact on spinal function (blue circle). If the leg is not flexed, a biologically prepared circuit enables the rapid detection of a relationship between the current limb position (the R) and the onset of noxious stimulation (the O). If there is a R-O relation, the stimulation is classified as controllable, which fosters the performance of a motor response that reduces net exposure to noxious stimulation. In the absence of a R-O relation (uncontrollable stimulation), a state of over-excitation is induced that enhances reactivity to mechanical stimulation and induces a lasting impairment in relational learning. Conversely, exposure to controllable stimulation has a restorative effect that fosters learning and counters the adverse effect of uncontrollable stimulation. Other work indicates that a central pattern generator exists in the rostral (L1-L2) spinal cord (upper box) that can be entrained by regular stimulation. Evidence suggests that regularity can be abstracted when stimulation is applied to different regions of the lower body and when some stimuli are randomly omitted. Periods of regular stimulation can foster rhythmic behavior, the abstraction of regularity across days (savings), and counter the adverse effects of uncontrollable stimulation (green lines). Exposure to stimuli that occur in a variable (irregular) manner impairs instrumental learning. Further work is needed to determine whether irregular stimulation also interferes with the abstraction of regularity (red?). Research is also needed to determine how sensory-motor integration impacts the central pattern generator. Evidence suggests that noxious stimulation can interfere with CPG function and the generation of rhythmic behavior (Bouffard et al., 2014; Caudle et al., 2015), implying that the dashed red line reflects a bi-directional process. It is not known whether exposure to controllable stimulation fosters the engagement of the CPG. Note that a ‘+’ and ‘–‘indicate how processes affect function, not the nature of neural communication (i.e., whether an excitatory or inhibitory process underlies the effects). The consequences of training that have been shown to have a lasting effect (implying a form of memory) are enclosed with dashed circles. Adapted from Grau et al. (2022).