Skip to main content
NCBI home page
Journal of Neurophysiology logoLink to Journal of Neurophysiology
. 2017 Sep 6;118(5):2944–2952. doi: 10.1152/jn.00190.2017

Constitutive activity of 5-HT2C receptors is present after incomplete spinal cord injury but is not modified after chronic SSRI or baclofen treatment

V M Tysseling 1,2,✉, D A Klein 2, R Imhoff-Manuel 2, M Manuel 2, C J Heckman 1,2,3,4, M C Tresch 2,3,4,5
PMCID: PMC5686237  PMID: 28877964

After spinal cord injury (SCI), most people will develop muscle spasms below their level of injury that can severely impact function. In this work, we examine the adaptations that occur within the spinal cord after SCI that contribute to these motor dysfunctions. We also evaluate one hypothesis about how these adaptations develop, which will potentially lead to intervention strategies to improve functional outcomes in persons with SCI.

Keywords: spinal cord injury, serotonin, spasms, motor recovery, motoneuron

Abstract

After spinal cord injury (SCI), reflexes become hyperexcitable, leading to debilitating muscle spasms and compromised motor function. Previous work has described adaptations in spinal systems that might underlie this hyperexcitability, including an increase in constitutively active 5-HT2C receptors in spinal motoneurons. That work, however, examined adaptations following complete transection SCI, whereas SCI in humans is usually anatomically and functionally incomplete. We therefore evaluated whether constitutive activity of 5-HT2C receptors contributes to reflex hyperexcitability in an incomplete compression model of SCI and to spasms in vitro and in vivo. Our results confirm that 5-HT2C receptor constitutive activity contributes to reflex excitability after incomplete SCI. We also evaluated whether constitutive activity could be altered by manipulation of neural activity levels after SCI, testing the hypothesis that it reflects homeostatic processes acting to maintain spinal excitability. We decreased neural activity after SCI by administering baclofen and increased activity by administering the selective serotonin reuptake inhibitor (SSRI) fluoxetine. We found that drug administration produced minimal alterations in in vivo locomotor function or reflex excitability. Similarly, we found that neither baclofen nor fluoxetine altered the contribution of constitutively active 5-HT2C receptors to reflexes after SCI, although the contribution of 5-HT2C receptors to reflex activity was altered after SSRIs. These results confirm the importance of constitutive activity in 5-HT2C receptors to spinal hyperexcitability following SCI in the clinically relevant case of incomplete SCI but suggest that this activity is not driven by homeostatic processes that act to maintain overall levels of spinal excitability.

NEW & NOTEWORTHY After spinal cord injury (SCI), most people will develop muscle spasms below their level of injury that can severely impact function. In this work, we examine the adaptations that occur within the spinal cord after SCI that contribute to these motor dysfunctions. We also evaluate one hypothesis about how these adaptations develop, which will potentially lead to intervention strategies to improve functional outcomes in persons with SCI.

spinal cord injury (SCI) limits function not only through paralysis but also through alterations in spinal reflexes. Reflexes become hyperactive after SCI, so that previously innocuous sensory stimuli elicit potentially debilitating involuntary muscle spasms (Adams and Hicks 2005). Clinical interventions to reduce these spasms are currently limited; the standard treatment to reduce spasms is systemic administration of baclofen, which reduces spinal excitability but also reduces general arousal and activity levels (Bowery 2006). A better understanding of the mechanisms underlying the reflex hyperexcitability that arises after SCI might identify targets for more effective interventions.

Adaptations in motor neuron properties have been shown to contribute to this reflex hyperexcitability. After complete transection at the sacrolumbar junction, sacral motor neurons become hyperexcitable, resulting in muscle spasms and long-lasting reflexes (Bennett et al. 2004). This hyperexcitability is partly produced by an upregulation in the expression of constitutively active 5-HT2C receptors, leading to amplification of synaptic inputs and production of muscle spasms from sensory stimuli (Murray et al. 2010). Blocking the activity in these receptors eliminates the long-lasting reflexes observed in animals with chronic complete transection SCI.

The extent to which these results apply to clinical cases of SCI, however, remains unclear for several reasons. First, the complete transection models of SCI used in those studies differ substantially from the typical cases of incomplete SCI in humans (Chen et al. 2016). The presence of residual axons across the lesion site means that incomplete SCI is not simply a quantitatively “milder” form of SCI but can differ qualitatively from complete SCI. The extent to which mechanisms underlying reflex hyperexcitability identified in complete SCI also contribute to reflex excitability in more clinically relevant incomplete SCI models is therefore unclear. Second, the sacrolumbar site of SCI used in previous studies is clinically uncommon; the extent to which adaptations following injuries at this site generalize to more common rostral injury sites is therefore also not clear. Although previous results suggest that constitutively active 5-HT2C receptors contribute to spasms in humans with complete, but not incomplete, SCI (D’Amico et al. 2013), the necessity of using pharmacologically nonspecific drugs in humans and the potential contribution of descending systems to observed reflexes make it difficult to evaluate alterations in 5-HT2C receptor function in spinal systems after SCI. The main goal of the present study was to examine these issues, evaluating whether constitutively active 5-HT2C receptors contribute to reflex excitability after incomplete thoracic SCI.

Although constitutively active 5-HT2C receptors lead to reflex hyperexcitability, this adaptation is critical for the reexpression of motor function after SCI (Murray et al. 2010). Indeed, one of the common interpretations of reflex hyperexcitability after SCI is that it results from a homeostatic mechanism to restore spinal excitability levels, allowing for effective movements in the absence of descending inputs. This interpretation has been used to explain not only contributions from constitutively active 5-HT2C receptors but also alterations in chloride reversal potentials in motor neurons, both of which contribute to reflex hyperexcitability (Gackière and Vinay 2014). If this homeostatic hypothesis were true, manipulation of spinal excitability levels after SCI should alter these adaptations. The present study evaluates this hypothesis through decreasing spinal excitability levels by administering baclofen or increasing excitability by administering the selective serotonin reuptake inhibitor (SSRI) fluoxetine and then measuring the contribution of constitutively active 5-HT2C receptors to spinal reflexes (Fig. 1A). Although serotonin (5-HT) has both excitatory and inhibitory effects on spinal systems, the net effect of increasing 5-HT levels in the spinal cord after SCI is generally to increase excitability (Leech et al. 2014; Perrier 2016; Perrier and Cotel 2008; Wei et al. 2014).

Fig. 1.

Fig. 1.

Open in a new tab

A: logic of experiments for improving motor function by administering SSRIs in incomplete spinal cord injury (SCI). After an incomplete SCI, both ionotropic and neuromodulatory descending projections to the spinal cord below the lesion are compromised (top), but local spinal interneuronal circuits are preserved. Constitutive activity in motoneurons (MNs) is present after incomplete SCI, indicated by arrows. Administration of SSRIs (middle) will prolong the availability of synaptically released 5-HT from residual raphespinal systems, increasing descending neuromodulatory drive and motoneuron excitability. This increased excitability might suppress adaptations underlying hyperexcitable reflexes such as constitutively active 5-HT2C receptors. Alternatively, decreasing overall excitability by application of baclofen would potentially enhance constitutive activity (bottom). IN, interneurons. B: typical placement of stimulating electrodes for measuring in vivo flexor withdrawal reflexes after SCI. C: experimental setup of in vitro sacral cord preparation. All sacral dorsal roots are placed on a stimulating electrode, while the individual ventral roots are placed on separate recording electrodes. Stimulation was applied to all dorsal roots simultaneously, and the responses evoked on each separate ventral root were recorded.

The use of SSRIs is especially relevant for incomplete SCI, since there are residual raphespinal serotonergic terminals remaining below the injury site. In principle, SSRIs might increase the efficacy of endogenously released 5-HT, restoring normal modulation of motor neuron excitability across activity levels and potentially facilitating effective movements by residual ionotropic systems (Fig. 1A). We therefore hypothesized that SSRI treatment would be an effective strategy to improve motor function, allowing for effective motor output while also reducing development of reflex hyperexcitability.

MATERIALS AND METHODS

Spinal cord injury.

All procedures were approved by the Institutional Animal Care and Use Committee at Northwestern University. Adult female CD-1 mice (18–28 g) were anesthetized (2–3% isoflurane) and prepared for aseptic surgery. A laminectomy was performed at T10 to expose the T12–13 spinal segments. The clip compression injury model was used to create the SCI at spinal segment T13 (Joshi and Fehlings 2002a, 2002b; Tysseling et al. 2010, 2011; Tysseling-Mattiace et al. 2008). This method uses calibrated clips to compress the spinal cord, with the amount of force applied by the clips allowing for different degrees of SCI. We used the moderate clip (8 g compression for 1 min) in order to produce an incomplete spinal injury, partially sparing the raphespinal, reticulospinal, corticospinal, vestibulospinal, and rubrospinal tracts (Joshi and Fehlings 2002b). As documented by Joshi and Fehlings (2002a, 2002b), this injury causes substantial reduction in motor function; of particular relevance to this study, raphespinal projections to the lumbar spinal cord are reduced by ~35%. After spinal injury, the wound was closed with wound clips. Animals were given meloxicam (0.3 mg/kg sc) for 2 days after SCI, and bladders were expressed manually until bladder function recovered (~1 wk).

Implantation of EMG electrodes.

We implanted electromyograph (EMG) electrodes as described previously (Tysseling et al. 2013). Briefly, electrode arrays were constructed with multistranded stainless steel wire (A-M Systems) attached to a connector (Omnetics, Nano Series) embedded in a nylon mesh. A pair of wires was implanted in four muscles in each leg: tibialis anterior (TA, ankle flexor), lateral gastrocnemius (LG, ankle extensor), biceps femoris posterior (BFp, hip extensor, knee flexor), and vastus lateralis (VL, knee extensor). The wires were tunneled subcutaneously to the back where the connector was implanted. Electrode exposures (~0.5 mm) were placed within the muscle belly and secured with knots on either side of the muscle. A reference electrode was placed subcutaneously on the back. Animals were monitored after recovery from the surgery in case they removed implanted electrodes (Tysseling et al. 2013), and electrodes were reimplanted if possible.

In vivo drug administration.

Immediately after SCI, animals were placed in one of three drug treatment groups: water control, fluoxetine (Sigma F132), or baclofen (Gallipot 104622). All drugs were administered through the drinking water of the mice. The drugs were added to water bottles for a final concentration of 50 mg/l based on preliminary measurements of average water consumption (5 ml) and average adult weight (22.5 g) to target delivery of 10 mg·kg body wt−1·day−1 for each animal. The water was replaced every other week to reduce the effects of drug instability. These drug concentrations were chosen because they are similar to concentrations used clinically in humans (Balu et al. 2009; Núñez et al. 2006; Zhang et al. 2000). Previous work has shown that this concentration of fluoxetine results in elevation of 5-HT in the CNS and effective modulation of neural systems (Koschnitzky et al. 2014). Similarly, previous work has shown that the concentration of baclofen used in this study reduces neural excitability (Spano et al. 2007; Yajima et al. 2000).

Locomotor function.

The mice underwent open-field behavioral testing and were scored with the Basso Mouse Scale for Locomotion (BMS) (Basso et al. 2006). Behavioral scoring was initiated immediately after SCI before the first administration of drugs and then continued biweekly for at least 2 mo. We measured locomotor function biweekly, since we were mainly interested in how these drugs affected the plateau level of functional recovery after SCI.

In vivo evaluation of flexion reflexes.

After 2 mo of SCI, we evaluated the excitability of flexion reflexes in animals in each treatment group. The flexor withdrawal test is an in vivo test of hyperreflexia, or muscle spasms. These experiments were performed in awake, unanesthetized animals. We placed silver ball electrodes on the plantar side of the hind paw (Fig. 1B) to evoke flexion withdrawal reflexes, choosing the leg that had the largest number of good-quality EMG electrodes. The amplitude of stimulation (square pulse, 4-ms duration) was varied to determine the threshold for visible movement of the stimulated leg. We then applied suprathreshold stimulation and measured the evoked EMGs. We waited at least 30 s between stimulation trials. Only trials for which there was no substantial EMG activity in all muscles before the stimulation were included for subsequent analyses. In one subset of animals, responses to a range of stimulation strengths were examined; in the rest, only responses to strong stimulation were examined. To increase statistical power, we combined both groups together and only examined the response to the strongest stimulation (>5 T). EMGs from the leg ipsilateral to the stimulation electrode were amplified (A-M Systems, model 1700; 10,000×) and band-pass filtered (30–1,000 Hz) before being recorded to computer (5,000 Hz) for off-line data analysis. EMG activity was digitally filtered (high-pass filter, 50 Hz) before being rectified. Reflex magnitude was calculated as a percentage above baseline activity: reflex magnitude (%) = 100 × [(EMG integrated for 2 s after stimulus)/(integrated EMG measured on prestimulus baseline) − 1]. A reflex magnitude of 0 indicates that there was no response above baseline. All responses were inspected visually for the absence of artifact and for minimal prestimulus baseline activity.

In vitro sacral cord preparation.

After all in vivo behavioral assessments were performed, we prepared animals for in vitro experiments examining the state of isolated sacral spinal cords in each treatment group. We followed previously published in vitro sacral cord preparation procedures (Jiang and Heckman 2006; Li et al. 2004). Briefly, the spinal cord was removed under anesthesia, exposing the spinal cord by dorsal laminectomy and isolating the sacral spinal cord with intact dorsal and ventral roots. Ventral and dorsal roots were wrapped around mounted silver wire electrodes and isolated with a mixture of Vaseline and mineral oil (Fig. 1C). A single electrode was used to stimulate all sacral dorsal roots simultaneously. Because the initial SCI in these animals was incomplete, the process of isolating the sacral cord for these in vitro experiments involved transecting residual descending axons to the sacral cord. As a result, reflex excitability in isolated sacral spinal cords was generally low because of the induced spinal shock caused by cutting these residual fibers. We therefore applied low doses of strychnine (0.2–0.4 μM) to the recording bath and waited at least 2 h before evaluating reflexes. Electrical stimulation (0.1 ms, 1-min interval) of dorsal roots began at ~5 μA and increased incrementally until a reliable reflex response was evoked on at least one ventral root; this intensity was defined as threshold for that experiment. Thereafter, all stimulations were at three times threshold intensity and applied every minute. After baseline reflex amplitude was established over 15 min, either 3 μM SB206553 (SB206, inverse agonist of 5-HT2C; Sigma) or 10 μM SB242084 (SB242, neutral antagonist of 5-HT2C; Sigma) was washed in, following the logic described by Murray et al. (2010). The inverse agonist SB206 will block constitutively active 5-HT2C receptors and prevent binding of 5-HT at these receptors. The neutral antagonist SB242 will only prevent binding of 5-HT. A larger effect of SB206 on evoked reflexes compared with SB242 implies that constitutively active 5-HT2C receptors contribute to observed reflexes. These drugs have been shown to be effective in mice (Dalton et al. 2006; Griebel et al. 1997). The effect of these drugs on reflexes was recorded up to 120 min after application. Electroneurograms (ENGs) were amplified (1,000×) and band-pass filtered (300–10,000 Hz) before being digitized (20 kHz) and saved for off-line analysis. Off-line processing of recorded ENGs and quantification of reflex magnitude were the same as described above for EMGs.

Statistical analyses.

We examined whether drug treatment affected locomotor function by analyzing changes in BMS scores over 2 mo after incomplete SCI. We used a linear mixed model for this statistical test, with fixed effects of treatment condition (untreated water, baclofen, fluoxetine) and week after SCI (0, 2, 4, 6, and 8 wk) and random effect of animal. To account for the plateauing of BMS scores over 2 mo we also included a quadratic week term (after centering the week variable). All analyses were performed in MATLAB (“fitlme” with parameters of maximum-likelihood model fitting and effects comparisons, “anova” with Satterthwaite estimation of degrees of freedom). We confirmed that results obtained with MATLAB were the same as results obtained with other statistical software packages (SAS, R). Data distributions were inspected for normality, and the quality of model fits was evaluated by inspection of residuals. For nonnegative data (e.g., reflex amplitudes, described below) data were log transformed before statistical analysis. Post hoc Bonferroni tests were performed with significance values adjusted according to the number of comparisons made (with “coeftest” in MATLAB).

We examined whether chronic drug treatment affected the magnitude of reflexes evoked from high-threshold stimulation in vivo. We used a linear mixed model with fixed effects of treatment condition (untreated water, baclofen, fluoxetine) and stimulation time and random effect of animal. The stimulation time was used to account for potential increasing [e.g., because of windup (Mendell and Wall 1965)] or decreasing [e.g., because of reflex adaptation (Spencer et al. 1966)]. Because reflexes were always observed as activity above baseline, reflex magnitudes were nonnegative and the overall distribution of magnitudes was highly skewed and non-Gaussian. We therefore log transformed all magnitude values before statistical analysis.

We evaluated whether constitutively active 5-HT2C receptors contributed to hyperexcitable reflexes after incomplete SCI. For this analysis, we examined only control animals treated with water and compared the effects of SB206 and SB242 on reflex amplitudes evoked from dorsal root stimulation and recorded on ventral roots. Reflex magnitudes were normalized to the magnitude observed before drug application. We used a linear mixed model with fixed effect of applied drug (SB206 or SB242). To account for the pseudoreplications of measuring multiple ventral roots from the same animal, we included random effect terms for animal and root.

We also evaluated whether chronic drug treatment affected the development of constitutively active 5-HT2C receptors and their contribution to spinal reflexes. For this analysis we included fixed effects of chronic treatment condition (untreated water, SSRI, baclofen) and drug (SB206 or SB242) and random effects of animal and root. Post hoc Bonferroni tests were performed with significance values adjusted according to the number of comparisons made.

RESULTS

Constitutively active 5-HT2C receptors contributed to reflex excitability after incomplete SCI.

Previous experiments have demonstrated that constitutively active 5-HT2C receptors contribute to the prolonged spinal reflexes observed after complete SCI (Murray et al. 2010). We first examined whether a similar contribution could be observed after incomplete SCI. Figure 2 shows examples of reflexes in two in vitro sacral cord preparations, showing the activity evoked on ventral roots in response to a single pulse of dorsal root stimulation. As seen in the figure, dorsal root stimulation evoked long-lasting ventral root reflexes lasting several seconds.

Fig. 2.

Fig. 2.

Open in a new tab

Constitutively active 5-HT2C receptors contribute to reflex excitability after incomplete SCI. A: the reflex evoked by dorsal root stimulation and recorded on a ventral root in the in vitro sacral spinal cord preparation before (left) and after (right) application of the inverse agonist SB206. SB206 greatly reduced the evoked reflex. B: a recording in a different preparation before (left) and after (right) application of the neutral antagonist SB242. SB242 did not significantly reduce the evoked reflex. C: quantification of these experiments, quantifying the magnitude of the evoked reflexes before and after application of SB206 or SB242. Reflex magnitude is expressed as % above the prestimulation baseline; a magnitude of 0 means that there was no evoked reflex. Data are means ± SD. All animals n = 13; all roots n = 39.

We then examined whether constitutively active 5-HT2C receptors contributed to these long-lasting reflexes, using the same procedures described by Murray et al., comparing the effects of the 5-HT2C inverse agonist (SB206) to the effects of the neutral antagonist (SB242) (Murray et al. 2010). The inverse agonist SB206 will block both constitutively active 5-HT2C receptors and binding of 5-HT at these receptors, while the neutral antagonist SB242 will only prevent binding of 5-HT. If constitutively active 5-HT2C receptors contributed to the observed responses, we should observe a preferential reduction in the reflex amplitude after SB206 compared with SB242. Figure 2 illustrates that this was the case for these examples; whereas SB206 strongly reduced the reflex amplitude, SB242 did not but instead, in this case, caused a small increase.

Figure 3 shows the results of all animals (n = 13) and all roots (n = 39). We used a mixed model statistical analysis to leverage the full set of observed responses while controlling for concerns about multiple observations from the same animal (Littell et al. 1998). We normalized all responses to the predrug (SB206 or SB242) magnitude, so that changes were expressed as a percent change from baseline reflex magnitude. We found that SB206 significantly decreased reflex amplitude (P < 0.001). SB242 reduced reflex amplitudes at a level approaching significance (P = 0.022; post hoc Bonferroni correction means the significance level is 0.05/3 = 0.0167). This reduction by the neutral antagonist SB242 suggests an effect of residual 5-HT, potentially released by spared raphespinal terminals, on the reflexes produced in these animals with incomplete SCI. Critically, we found that SB206 caused a significantly larger reduction in reflex amplitude than SB242 (P < 0.001) (see also Fig. 6). Note also the consistency of the reflex reduction for SB206 compared with that for SB242, as seen in Fig. 3; 19 of 20 responses were reduced by SB206, whereas 13 of 19 responses were reduced by SB242. These results demonstrate that, similar to what was observed in complete transection SCI, constitutively active 5-HT2C receptors contribute to reflex excitability in animals after the more clinically relevant case of incomplete SCI.

Fig. 3.

Fig. 3.

Open in a new tab

Effects of SB206 and SB242 after incomplete SCI for all experiments. Each connected pair of dots shows the reflex magnitude before and after application of SB206 (left) and SB242 (right) on the same ventral root. All animals n = 13; all roots n = 39.

Fig. 6.

Fig. 6.

Open in a new tab

Effect of chronic administration of baclofen, fluoxetine, and untreated water on the contribution of 5-HT2C receptors to reflex magnitude measured in the isolated sacral spinal cord. Effects of SB206 and SB242 are shown here as % change of the reflex magnitude before and after drug application. A: the effects measured for each individual root recording. B: the average effect for each condition. Data are means ± SD. Animals treated with fluoxetine (28 root recordings in 8 animals), baclofen (25 root recordings in 7 animals), or water (39 root recordings in 13 animals). *P = 0.025, **P = 0.037.

Chronic administration of SSRIs and baclofen had minimal effect on gross locomotor ability or withdrawal reflexes in vivo.

The contribution of constitutively active 5-HT2C receptors to spinal reflexes after SCI might reflect a homeostatic process, such that the loss of inputs from descending systems is countered by an increased intrinsic excitability of spinal systems. To test this hypothesis we evaluated whether chronic manipulation of spinal excitability altered locomotor performance, reflex excitability, and the contribution of constitutively active 5-HT2C receptors to long-lasting reflexes. To decrease spinal excitability after incomplete SCI, we administered baclofen; to increase spinal excitability, we administered the SSRI fluoxetine.

We evaluated gross locomotor ability of mice in each condition, using the BMS and Basso, Beattie, Bresnahan Locomotor Rating Scale scoring systems as we have used previously (Basso et al. 2006; Tysseling et al. 2010, 2011, 2013; Tysseling-Mattiace et al. 2008). We report here analysis of BMS scores, although similar results were obtained with either scoring system. Figure 4 shows the BMS scores for animals given SSRIs (n = 45), baclofen (n = 15), or water control (n = 77) starting immediately after an incomplete SCI and evaluated every other week thereafter. There was a significant interaction between treatment condition and week after SCI (P = 0.02). Although post hoc tests found no significant differences between treatment groups at any time point after SCI (P > 0.05), it appeared that this difference was due to a smaller improvement in functional performance by the baclofen group. This possibility was supported by repeating the analysis with only two treatment conditions at a time. The significant effect was maintained when water and baclofen were tested together (P = 0.006) and was eliminated when water and SSRI treatment groups were tested together (P = 0.25). It should be noted, however, that the initial BMS score for baclofen-treated animals was slightly higher immediately after SCI and so the interaction term could have reflected this initial difference. To evaluate this possibility, we repeated the analysis but omitted the first time point and found that there was no significant effect of treatment condition (P = 0.82) and no interaction between treatment condition and week after SCI (P = 0.44). These results are therefore consistent with the possibility that all three treatment groups recovered to indistinguishable functional levels 2 mo after SCI.

Fig. 4.

Fig. 4.

Open in a new tab

Effect of chronic administration of baclofen, fluoxetine, and untreated water on locomotor function after incomplete SCI. The BMS score of animals was measured every other week over 2 mo after SCI. Week 0 indicates the week when the injury was made. Data are means ± SE. SSRIs n = 45; baclofen n = 15; water control n = 77.

We also evaluated whether treatment condition affected reflex excitability by examining the magnitude of reflexes evoked by electrical stimulation of the hind paw foot pad in animals in each of the treatment conditions. Figure 5A shows an example of the EMGs recorded in VL, BFp, TA, and LG in response to a single pulse of high-intensity electrical stimulation (>5 T). In general, the evoked responses were dominated by activation of TA, although other animals showed coactivation of muscles or more complicated temporal modulation of activity. Figure 5B shows the reflex amplitude produced by each muscle in response to high-threshold stimulation applied to the hindlimb foot pad, combined across all animals and separated for each treatment condition. We found no consistent effect of treatment condition on these evoked reflex amplitudes, suggesting that chronic treatment with SSRIs or baclofen did not significantly modulate reflex amplitudes in vivo (VL: P = 0.55, n = 11/11/12 animals for baclofen/SSRIs/H2O; BFp: P = 0.55, n = 11/9/11; TA: P = 0.53, n = 11/7/9; LG: P = 0.99, n = 11/10/10).

Fig. 5.

Fig. 5.

Open in a new tab

Effect of chronic administration of baclofen, fluoxetine, and untreated water on reflexes evoked from high-threshold stimulation of the plantar surface of the hind paw. A: example of EMGs evoked by a single pulse of electrical stimulation applied to the foot. B: effect of drug treatment on reflex magnitude for each muscle, averaged across animals. Data are means ± SD. n for baclofen/SSRIs/H2O, respectively: VL, n = 11/11/12; BFp, n = 11/9/11; TA, n = 11/7/9; LG, n = 11/10/10.

Chronic administration of SSRIs altered serotonergic systems in isolated spinal cord but not constitutively active 5-HT2C receptors.

The in vivo results described above suggested that chronic treatment with SSRIs or baclofen did not substantially alter functional outcomes after SCI. However, because animals had an incomplete SCI, these in vivo measures reflect contributions from both spinal and supraspinal systems. Any changes in function, or lack thereof, following different treatment conditions might reflect a combination of adaptations in both spinal and supraspinal systems. We therefore examined whether treatments affected adaptations in spinal systems following SCI, evaluating whether there were differences in the contribution of constitutively active 5-HT2C receptors to spinal reflexes.

Figure 6 shows the main results of these experiments, illustrating the change in reflex amplitudes caused by SB206 and SB242 in animals treated with fluoxetine (28 root recordings in 8 animals), baclofen (25 root recordings in 7 animals), or water (39 root recordings in 13 animals). The control animals treated with normal water are the same as those shown in Fig. 3. Across all three treatment conditions SB206 produced a greater reduction in reflex amplitude than SB242 (P = 0.013). There was also a significant difference between treatment groups (P = 0.037), with animals treated with fluoxetine being significantly different from animals treated with either water or baclofen (post hoc comparison, P = 0.025). For the animals treated with fluoxetine, there appeared to be a shift in the effects of SB206 and SB242 compared with control animals, with SB206 only minimally affecting the evoked reflex and SB242 producing on average an increase in reflex amplitude. Importantly, however, there was no significant interaction between treatment condition and drug, suggesting that the relative effects of SB206 and SB242 were not altered by treatment conditions (P = 0.80), indicating that chronic drug treatments did not alter the expression of constitutively active 5-HT2C receptors.

In the analysis described above, we quantified changes in reflexes caused by SB206/SB242 as a percent change of the initial reflex magnitude. We chose this approach to reduce variability due to differences in initial reflex amplitude; e.g., a large reflex that is reduced by 50% will be treated the same as a small reflex that is reduced by the same amount. However, this approach might overemphasize the importance of initially small reflexes that might be unreliable or insensitive to further reductions in amplitude. We evaluated this possibility by excluding all reflexes that had very small initial magnitudes (a response <10% above baseline) and repeating the analysis. Although this criterion excluded 20 of 92 root responses from the analysis, we found a similar pattern of results: there was a main effect of drug applied (SB206 vs. SB242) (P = 0.026) and treatment condition (P = 0.030), with SSRI treatment different from water (P = 0.019), and no interaction between treatment condition and drug applied (P = 0.71).

We also examined whether chronic drug treatment affected reflex magnitudes in the isolated sacral cord preparation before application of either SB206 or SB242. We found no significant differences in the amplitude of reflexes across drug treatment group (P = 0.74), consistent with the lack of effect of this drug treatment on reflex amplitude measured in vivo, as described above.

DISCUSSION

There were two main results of our experiments. First, we found that constitutively active 5-HT2C receptors made substantial contributions to spinal reflex excitability after incomplete SCI. Second, we found that chronic manipulation of spinal excitability by application of SSRIs or baclofen had minimal effects on behavioral outcomes after incomplete SCI or on the development of constitutive activity in 5-HT2C receptors but that SSRIs altered serotonergic function in spinal systems.

Constitutive activity in 5-HT2C receptors after incomplete SCI.

Our finding showing that expression of constitutively active 5-HT2C receptors contributes to reflex hyperexcitability after SCI replicates and extends previous work (D’Amico et al. 2013; Li et al. 2007; Murray et al. 2010, 2011). Our results show that observing this constitutive activity does not depend on the species examined (mice vs. rats), the level of SCI (thoracic vs. sacral), or the severity of the SCI (incomplete vs complete). This last point is important, since most cases of SCI in humans are incomplete (NSCISC 2016). Our results therefore suggest that similar mechanisms might underlie increased spasticity in humans with SCI (D’Amico et al. 2013).

Although we showed that constitutive activity of these receptors contributes to reflex excitability after incomplete injury, this contribution might be less than that observed after complete injury. A direct comparison of our results to those in previous work with complete SCI is difficult, however. It is likely that the process of isolating the sacral spinal cord in these experiments with incomplete SCI resulted in a greater reduction of spinal excitability than in previous work with complete SCI. With incomplete SCI, isolation of the sacral cord requires cutting residual connections to the cord and thereby causing spinal shock; with complete SCI, there are no residual connections to be cut and therefore less reduction in spinal excitability. It is therefore likely that isolating the spinal cord in this study with incomplete SCI reduced spinal excitability to a greater extent than in previous studies using complete SCI. Our need to apply low doses of strychnine to increase excitability is consistent with this possibility. Similarly, it is likely that even larger doses of strychnine or bicuculline would be necessary in animals with acutely isolated spinal cord, making it difficult to compare contributions of constitutively active receptors to spinal reflexes in those animals to the animals studied here or in Murray et al. (2010). Furthermore, the thoracic injury used here might cause degrees of adaptations in neural systems below the lesion different from the sacral injuries used in other studies. More generally, recovery of function after incomplete SCI involves adaptations in neural systems on both sides of the injury, and so incomplete SCI is not simply a “milder” form of SCI compared with complete transection (Bareyre et al. 2004; Courtine et al. 2008; Raineteau and Schwab 2001).

Effects of drug treatment on functional outcomes after incomplete SCI.

We found that chronic administration of SSRIs or baclofen had minimal effects on locomotor performance or reflex magnitudes after incomplete SCI. The lack of effect of SSRIs here is in contrast to other work showing that fluoxetine improves recovery after SCI (Scali et al. 2013). There are several potential reasons for the difference in outcomes between these studies, including a difference in the animals studied (mice in this study vs. rats), injury type (thoracic compression injury vs. cervical dorsal horn injury), treatment protocol (SSRIs started at time of injury vs. several weeks before injury), and SSRI dosage level (10 mg·kg−1·day−1 vs. 16 mg·kg−1·day−1). It should be noted that although we considered the role of SSRIs here in the context of their effects on serotonergic function, the functional improvement described by Scali et al. (2013) was attributed to facilitation of plasticity and corticospinal axon sprouting by fluoxetine consistent with fluoxetine’s demonstrated role in improving neurogenesis, synaptic plasticity, and functional recovery in a variety of systems (Ampuero et al. 2010; Kobayashi et al. 2010; Maya Vetencourt et al. 2008; Southwell et al. 2010). The lack of obvious effect of SSRI treatment might also reflect the heterogeneous effects of 5-HT on spinal systems (Perrier 2016). It will be important in future work to evaluate the differences in outcomes between these studies and the mechanisms through which SSRIs act to alter functional recovery.

We found a small impairment in recovery for animals treated with baclofen, although this difference appeared to be mainly driven by a slightly higher motor function immediately after SCI. There was no significant difference in reflex magnitudes in baclofen-treated animals. The absence of reduction in reflexes might reflect development of tolerance to baclofen, as is commonly observed in patients with SCI (Veerakumar et al. 2015), so that by the 2 mo time point at which we evaluated reflexes the effects of baclofen were minimal.

Alterations in spinal serotonergic systems after chronic SSRI treatment.

One of the goals of this study was to evaluate whether adaptations in spinal systems reflect homeostatic processes. We found that chronic SSRI administration altered serotonergic systems in the spinal cord, as demonstrated by the altered response of spinal reflexes to SB206 and SB242. The reduced effect of SB206 after SSRI treatment compared with water-treated control animals (Fig. 6) is consistent with reduction of constitutively active 5-HT2C receptors. However, SSRI treatment also apparently altered the effects of SB242; whereas SB242 caused a slight reduction in reflex amplitude in control animals (and in animals treated with baclofen), it caused a slight increase in reflex amplitude after SSRI treatment. Because both SB206 and SB242 block 5-HT binding to 5-HT2C receptors while SB206 additionally blocks constitutively active 5-HT2C receptors (Murray et al. 2010), the simplest interpretation of these results is that SSRI treatment altered the consequence of 5-HT binding at 5-HT2C receptors. In this interpretation, the amount of constitutively active 5-HT2C receptors was unaltered after SSRI treatment, but there was a development of an inhibitory effect of 5-HT binding to 5-HT2C receptors on evoked reflexes.

The apparent development of an inhibitory effect of 5-HT binding on 5-HT2C receptors in spinal systems after SSRI treatment was unexpected, especially since 5-HT2C receptors are thought to activate an excitatory Gq-coupled pathway (Di Giovanni et al. 2011). One possible explanation for this finding would be an increased expression in 5-HT2C receptors on inhibitory neurons, mostly likely on GABAergic neurons since the present experiments were performed with low doses of strychnine. In other systems, 5-HT2C receptors are expressed on GABAergic neurons and are increased after stress (Martin et al. 2014; Spoida et al. 2014). Although 5-HT2C receptor expression has been shown to increase after SCI (Husch et al. 2012), it is not clear whether this increase is expressed selectively in specific neuronal subtypes or how SSRI administration might alter this expression. It is also important to note that there was considerable overlap between the changes in reflex amplitudes for animals treated with SSRIs compared with control or baclofen treatment (Fig. 6A) and so the effects of SSRIs on 5-HT receptor function after SCI, although statistically significant, were not necessarily substantial.

These observations, along with the lack of alteration due to baclofen administration, do not support the hypothesis that spinal adaptations underlying reflex hyperexcitability are due to homeostatic processes maintaining neural excitability. Recent work has suggested that the increase in constitutively active 5-HT2C receptors might not be triggered by decreased activity levels but by an initial inflammatory response after SCI, inducing mRNA editing of the 5-HT2C receptor, among others (Di Narzo et al. 2015). Alternatively, the reduction of 5-HT per se after SCI, independent of its effects on neural activity, might drive adaptations in 5-HT receptors. Although our results suggest that constitutive activity in 5-HT2C receptors is not driven by homeostatic processes, it remains possible that homeostatic processes are involved in other adaptations after SCI, such as alterations in chloride transporters (Gackière and Vinay 2014).

One limitation of these results is that systemic administration of drugs will affect neural systems both below and above the site of SCI. Drug administration might therefore affect spinal systems not only directly but also indirectly, by altering activity in residual descending systems. Experiments using intrathecal application of drugs below the site of injury would help evaluate the contribution of such indirect effects on spinal adaptations.

Implications for interventions after SCI.

The results described here provide several potential insights to guide development of interventions to improve function after SCI. First, the fact that changes in constitutively active 5-HT2C receptors contribute to reflex excitability in the clinically relevant case of incomplete SCI suggests that interventions targeting this constitutive activity might help reduce spasms, though potentially at the expense of reducing motor recovery (Murray et al. 2010). Our results also suggest that the factors driving development of hyperexcitable reflexes in incomplete SCI are more complicated than simply reflecting homeostatic mechanisms, although this conclusion will require further experiments to fully establish. If homeostatic processes were responsible for spinal adaptations after SCI, any intervention that maintained spinal excitability after SCI might be effective at preventing the development of hyperexcitable reflexes. Instead, if homeostatic processes are not involved, as suggested by the results described here, interventions might need to target other aspects of the spinal injury (e.g., inflammation) that bring about the adaptations underlying hyperexcitable reflexes. Development of interventions to improve function in people with SCI will benefit from a better understanding of the processes involved in triggering and regulating expression of constitutively active 5-HT2C receptors and their contributions to reflex excitability and motor function after SCI.

GRANTS

This work was supported by the Craig H. Neilsen Foundation Postdoctoral Fellowship (V. M. Tysseling) and National Institutes of Health Grants 5K01 HD-084672 (V. M. Tysseling), R01 NS-089313 (C. J. Heckman), and R01 AR-053608 (M. C. Tresch).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

V.M.T. and M.C.T. conceived and designed research; V.M.T., D.A.K., R.I.-M., M.M., and M.C.T. performed experiments; V.M.T., D.A.K., R.I.-M., and M.C.T. analyzed data; V.M.T., D.A.K., R.I.-M., C.J.H., and M.C.T. interpreted results of experiments; V.M.T. and M.C.T. prepared figures; V.M.T. and M.C.T. drafted manuscript; V.M.T., R.I.-M., M.M., C.J.H., and M.C.T. edited and revised manuscript; V.M.T., D.A.K., R.I.-M., M.M., C.J.H., and M.C.T. approved final version of manuscript.

REFERENCES

  1. Adams MM, Hicks AL. Spasticity after spinal cord injury. Spinal Cord 43: 577–586, 2005. doi: 10.1038/sj.sc.3101757. [DOI] [PubMed] [Google Scholar]
  2. Ampuero E, Rubio FJ, Falcon R, Sandoval M, Diaz-Veliz G, Gonzalez RE, Earle N, Dagnino-Subiabre A, Aboitiz F, Orrego F, Wyneken U. Chronic fluoxetine treatment induces structural plasticity and selective changes in glutamate receptor subunits in the rat cerebral cortex. Neuroscience 169: 98–108, 2010. doi: 10.1016/j.neuroscience.2010.04.035. [DOI] [PubMed] [Google Scholar]
  3. Balu DT, Hodes GE, Anderson BT, Lucki I. Enhanced sensitivity of the MRL/MpJ mouse to the neuroplastic and behavioral effects of chronic antidepressant treatments. Neuropsychopharmacology 34: 1764–1773, 2009. doi: 10.1038/npp.2008.234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bareyre FM, Kerschensteiner M, Raineteau O, Mettenleiter TC, Weinmann O, Schwab ME. The injured spinal cord spontaneously forms a new intraspinal circuit in adult rats. Nat Neurosci 7: 269–277, 2004. doi: 10.1038/nn1195. [DOI] [PubMed] [Google Scholar]
  5. Basso DM, Fisher LC, Anderson AJ, Jakeman LB, McTigue DM, Popovich PG. Basso Mouse Scale for locomotion detects differences in recovery after spinal cord injury in five common mouse strains. J Neurotrauma 23: 635–659, 2006. doi: 10.1089/neu.2006.23.635. [DOI] [PubMed] [Google Scholar]
  6. Bennett DJ, Sanelli L, Cooke CL, Harvey PJ, Gorassini MA. Spastic long-lasting reflexes in the awake rat after sacral spinal cord injury. J Neurophysiol 91: 2247–2258, 2004. doi: 10.1152/jn.00946.2003. [DOI] [PubMed] [Google Scholar]
  7. Bowery NG. GABAB receptor: a site of therapeutic benefit. Curr Opin Pharmacol 6: 37–43, 2006. doi: 10.1016/j.coph.2005.10.002. [DOI] [PubMed] [Google Scholar]
  8. Chen Y, He Y, DeVivo MJ. Changing demographics and injury profile of new traumatic spinal cord injuries in the United States, 1972–2014. Arch Phys Med Rehabil 97: 1610–1619, 2016. doi: 10.1016/j.apmr.2016.03.017. [DOI] [PubMed] [Google Scholar]
  9. Courtine G, Song B, Roy RR, Zhong H, Herrmann JE, Ao Y, Qi J, Edgerton VR, Sofroniew MV. Recovery of supraspinal control of stepping via indirect propriospinal relay connections after spinal cord injury. Nat Med 14: 69–74, 2008. doi: 10.1038/nm1682. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. D’Amico JM, Murray KC, Li Y, Chan KM, Finlay MG, Bennett DJ, Gorassini MA. Constitutively active 5-HT2/α1 receptors facilitate muscle spasms after human spinal cord injury. J Neurophysiol 109: 1473–1484, 2013. doi: 10.1152/jn.00821.2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dalton GL, Lee MD, Kennett GA, Dourish CT, Clifton PG. Serotonin 1B and 2C receptor interactions in the modulation of feeding behaviour in the mouse. Psychopharmacology (Berl) 185: 45–57, 2006. doi: 10.1007/s00213-005-0212-3. [DOI] [PubMed] [Google Scholar]
  12. Di Giovanni G, Esposito E, Di Matteo V, editors. 5-HT2C Receptors in the Pathophysiology of CNS Disease. New York: Humana, 2011, p. xi. doi: 10.1007/978-1-60761-941-3. [DOI] [Google Scholar]
  13. Di Narzo AF, Kozlenkov A, Ge Y, Zhang B, Sanelli L, May Z, Li Y, Fouad K, Cardozo C, Koonin EV, Bennett DJ, Dracheva S. Decrease of mRNA editing after spinal cord injury is caused by down-regulation of ADAR2 that is triggered by inflammatory response. Sci Rep 5: 12615, 2015. doi: 10.1038/srep12615. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gackière F, Vinay L. Serotonergic modulation of post-synaptic inhibition and locomotor alternating pattern in the spinal cord. Front Neural Circuits 8: 102, 2014. doi: 10.3389/fncir.2014.00102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Griebel G, Perrault G, Sanger DJ. A comparative study of the effects of selective and non-selective 5-HT2 receptor subtype antagonists in rat and mouse models of anxiety. Neuropharmacology 36: 793–802, 1997. doi: 10.1016/S0028-3908(97)00034-8. [DOI] [PubMed] [Google Scholar]
  16. Husch A, Van Patten GN, Hong DN, Scaperotti MM, Cramer N, Harris-Warrick RM. Spinal cord injury induces serotonin supersensitivity without increasing intrinsic excitability of mouse V2a interneurons. J Neurosci 32: 13145–13154, 2012. doi: 10.1523/JNEUROSCI.2995-12.2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Jiang MC, Heckman CJ. In vitro sacral cord preparation and motoneuron recording from adult mice. J Neurosci Methods 156: 31–36, 2006. doi: 10.1016/j.jneumeth.2006.02.002. [DOI] [PubMed] [Google Scholar]
  18. Joshi M, Fehlings MG. Development and characterization of a novel, graded model of clip compressive spinal cord injury in the mouse: Part 1. Clip design, behavioral outcomes, and histopathology. J Neurotrauma 19: 175–190, 2002a. doi: 10.1089/08977150252806947. [DOI] [PubMed] [Google Scholar]
  19. Joshi M, Fehlings MG. Development and characterization of a novel, graded model of clip compressive spinal cord injury in the mouse: Part 2. Quantitative neuroanatomical assessment and analysis of the relationships between axonal tracts, residual tissue, and locomotor recovery. J Neurotrauma 19: 191–203, 2002b. doi: 10.1089/08977150252806956. [DOI] [PubMed] [Google Scholar]
  20. Kobayashi K, Ikeda Y, Sakai A, Yamasaki N, Haneda E, Miyakawa T, Suzuki H. Reversal of hippocampal neuronal maturation by serotonergic antidepressants. Proc Natl Acad Sci USA 107: 8434–8439, 2010. doi: 10.1073/pnas.0912690107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Koschnitzky JE, Quinlan KA, Lukas TJ, Kajtaz E, Kocevar EJ, Mayers WF, Siddique T, Heckman CJ. Effect of fluoxetine on disease progression in a mouse model of ALS. J Neurophysiol 111: 2164–2176, 2014. doi: 10.1152/jn.00425.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Leech KA, Kinnaird CR, Hornby TG. Effects of serotonergic medications on locomotor performance in humans with incomplete spinal cord injury. J Neurotrauma 31: 1334–1342, 2014. doi: 10.1089/neu.2013.3206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Li X, Murray K, Harvey PJ, Ballou EW, Bennett DJ. Serotonin facilitates a persistent calcium current in motoneurons of rats with and without chronic spinal cord injury. J Neurophysiol 97: 1236–1246, 2007. doi: 10.1152/jn.00995.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Li Y, Harvey PJ, Li X, Bennett DJ. Spastic long-lasting reflexes of the chronic spinal rat studied in vitro. J Neurophysiol 91: 2236–2246, 2004. doi: 10.1152/jn.01010.2003. [DOI] [PubMed] [Google Scholar]
  25. Littell RC, Henry PR, Ammerman CB. Statistical analysis of repeated measures data using SAS procedures. J Anim Sci 76: 1216–1231, 1998. doi: 10.2527/1998.7641216x. [DOI] [PubMed] [Google Scholar]
  26. Martin CB, Gassmann M, Chevarin C, Hamon M, Rudolph U, Bettler B, Lanfumey L, Mongeau R. Effect of genetic and pharmacological blockade of GABA receptors on the 5-HT2C receptor function during stress. J Neurochem 131: 566–572, 2014. doi: 10.1111/jnc.12929. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Maya Vetencourt JF, Sale A, Viegi A, Baroncelli L, De Pasquale R, O’Leary OF, Castrén E, Maffei L. The antidepressant fluoxetine restores plasticity in the adult visual cortex. Science 320: 385–388, 2008. doi: 10.1126/science.1150516. [DOI] [PubMed] [Google Scholar]
  28. Mendell LM, Wall PD. Responses of single dorsal cord cells to peripheral cutaneous unmyelinated fibres. Nature 206: 97–99, 1965. doi: 10.1038/206097a0. [DOI] [PubMed] [Google Scholar]
  29. Murray KC, Nakae A, Stephens MJ, Rank M, D’Amico J, Harvey PJ, Li X, Harris RL, Ballou EW, Anelli R, Heckman CJ, Mashimo T, Vavrek R, Sanelli L, Gorassini MA, Bennett DJ, Fouad K. Recovery of motoneuron and locomotor function after spinal cord injury depends on constitutive activity in 5-HT2C receptors. Nat Med 16: 694–700, 2010. doi: 10.1038/nm.2160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Murray KC, Stephens MJ, Ballou EW, Heckman CJ, Bennett DJ. Motoneuron excitability and muscle spasms are regulated by 5-HT2B and 5-HT2C receptor activity. J Neurophysiol 105: 731–748, 2011. doi: 10.1152/jn.00774.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. NSCISC Spinal cord injury (SCI) 2016 facts and figures at a glance. J Spinal Cord Med 39: 493–494, 2016. doi: 10.1080/10790268.2016.1210925. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Núñez MJ, Balboa J, Rodrigo E, Brenlla J, González-Peteiro M, Freire-Garabal M. Effects of fluoxetine on cellular immune response in stressed mice. Neurosci Lett 396: 247–251, 2006. doi: 10.1016/j.neulet.2005.11.042. [DOI] [PubMed] [Google Scholar]
  33. Perrier JF. Modulation of motoneuron activity by serotonin. Dan Med J 63: B5204, 2016. [PubMed] [Google Scholar]
  34. Perrier JF, Cotel F. Serotonin differentially modulates the intrinsic properties of spinal motoneurons from the adult turtle. J Physiol 586: 1233–1238, 2008. doi: 10.1113/jphysiol.2007.145706. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Raineteau O, Schwab ME. Plasticity of motor systems after incomplete spinal cord injury. Nat Rev Neurosci 2: 263–273, 2001. doi: 10.1038/35067570. [DOI] [PubMed] [Google Scholar]
  36. Scali M, Begenisic T, Mainardi M, Milanese M, Bonifacino T, Bonanno G, Sale A, Maffei L. Fluoxetine treatment promotes functional recovery in a rat model of cervical spinal cord injury. Sci Rep 3: 2217, 2013. doi: 10.1038/srep02217. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Southwell DG, Froemke RC, Alvarez-Buylla A, Stryker MP, Gandhi SP. Cortical plasticity induced by inhibitory neuron transplantation. Science 327: 1145–1148, 2010. doi: 10.1126/science.1183962. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Spano MS, Fattore L, Fratta W, Fadda P. The GABAB receptor agonist baclofen prevents heroin-induced reinstatement of heroin-seeking behavior in rats. Neuropharmacology 52: 1555–1562, 2007. doi: 10.1016/j.neuropharm.2007.02.012. [DOI] [PubMed] [Google Scholar]
  39. Spencer WA, Thompson RF, Neilson DR Jr. Response decrement of the flexion reflex in the acute spinal cat and transient restoration by strong stimuli. J Neurophysiol 29: 221–239, 1966. [DOI] [PubMed] [Google Scholar]
  40. Spoida K, Masseck OA, Deneris ES, Herlitze S. Gq/5-HT2c receptor signals activate a local GABAergic inhibitory feedback circuit to modulate serotonergic firing and anxiety in mice. Proc Natl Acad Sci USA 111: 6479–6484, 2014. doi: 10.1073/pnas.1321576111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Tysseling VM, Janes L, Imhoff R, Quinlan KA, Lookabaugh B, Ramalingam S, Heckman CJ, Tresch MC. Design and evaluation of a chronic EMG multichannel detection system for long-term recordings of hindlimb muscles in behaving mice. J Electromyogr Kinesiol 23: 531–539, 2013. doi: 10.1016/j.jelekin.2012.11.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Tysseling VM, Mithal D, Sahni V, Birch D, Jung H, Miller RJ, Kessler JA. SDF1 in the dorsal corticospinal tract promotes CXCR4+ cell migration after spinal cord injury. J Neuroinflammation 8: 16, 2011. (Erratum. J Neuroinflammation 14: 35, 2017). doi: 10.1186/1742-2094-8-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Tysseling VM, Sahni V, Pashuck ET, Birch D, Hebert A, Czeisler C, Stupp SI, Kessler JA. Self-assembling peptide amphiphile promotes plasticity of serotonergic fibers following spinal cord injury. J Neurosci Res 88: 3161–3170, 2010. doi: 10.1002/jnr.22472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Tysseling-Mattiace VM, Sahni V, Niece KL, Birch D, Czeisler C, Fehlings MG, Stupp SI, Kessler JA. Self-assembling nanofibers inhibit glial scar formation and promote axon elongation after spinal cord injury. J Neurosci 28: 3814–3823, 2008. doi: 10.1523/JNEUROSCI.0143-08.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Veerakumar A, Cheng JJ, Sunshine A, Ye X, Zorowitz RD, Anderson WS. Baclofen dosage after traumatic spinal cord injury: a multi-decade retrospective analysis. Clin Neurol Neurosurg 129: 50–56, 2015. doi: 10.1016/j.clineuro.2014.11.023. [DOI] [PubMed] [Google Scholar]
  46. Wei K, Glaser JI, Deng L, Thompson CK, Stevenson IH, Wang Q, Hornby TG, Heckman CJ, Kording KP. Serotonin affects movement gain control in the spinal cord. J Neurosci 34: 12690–12700, 2014. doi: 10.1523/JNEUROSCI.1855-14.2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Yajima Y, Narita M, Tsuda M, Imai S, Kamei J, Nagase H, Suzuki T. Modulation of NMDA- and (+)TAN-67-induced nociception by GABAB receptors in the mouse spinal cord. Life Sci 68: 719–725, 2000. doi: 10.1016/S0024-3205(00)00975-9. [DOI] [PubMed] [Google Scholar]
  48. Zhang Y, Raap DK, Garcia F, Serres F, Ma Q, Battaglia G, Van de Kar LD. Long-term fluoxetine produces behavioral anxiolytic effects without inhibiting neuroendocrine responses to conditioned stress in rats. Brain Res 855: 58–66, 2000. doi: 10.1016/S0006-8993(99)02289-1. [DOI] [PubMed] [Google Scholar]

Articles from Journal of Neurophysiology are provided here courtesy of American Physiological Society

RESOURCES