Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2013 Jul 22;591(Pt 22):5585–5598. doi: 10.1113/jphysiol.2013.256644

Spinal TNFα is necessary for inactivity-induced phrenic motor facilitation

Oleg Broytman 1, Nathan A Baertsch 1, Tracy L Baker-Herman 1
PMCID: PMC3853497  PMID: 23878370

Abstract

A prolonged reduction in central neural respiratory activity elicits a form of plasticity known as inactivity-induced phrenic motor facilitation (iPMF), a ‘rebound’ increase in phrenic burst amplitude apparent once respiratory neural activity is restored. iPMF requires atypical protein kinase C (aPKC) activity within spinal segments containing the phrenic motor nucleus to stabilize an early transient increase in phrenic burst amplitude and to form long-lasting iPMF following reduced respiratory neural activity. Upstream signal(s) leading to spinal aPKC activation are unknown. We tested the hypothesis that spinal tumour necrosis factor-α (TNFα) is necessary for iPMF via an aPKC-dependent mechanism. Anaesthetized, ventilated rats were exposed to a 30 min neural apnoea; upon resumption of respiratory neural activity, a prolonged increase in phrenic burst amplitude (42 ± 9% baseline; P < 0.05) was apparent, indicating long-lasting iPMF. Pretreatment with recombinant human soluble TNF receptor 1 (sTNFR1) in the intrathecal space at the level of the phrenic motor nucleus prior to neural apnoea blocked long-lasting iPMF (2 ± 8% baseline; P > 0.05). Intrathecal TNFα without neural apnoea was sufficient to elicit long-lasting phrenic motor facilitation (pMF; 62 ± 7% baseline; P < 0.05). Similar to iPMF, TNFα-induced pMF required spinal aPKC activity, as intrathecal delivery of a ζ-pseudosubstrate inhibitory peptide (PKCζ-PS) 35 min following intrathecal TNFα arrested TNFα-induced pMF (28 ± 8% baseline; P < 0.05). These data demonstrate that: (1) spinal TNFα is necessary for iPMF; and (2) spinal TNFα is sufficient to elicit pMF via a similar aPKC-dependent mechanism. These data are consistent with the hypothesis that reduced respiratory neural activity elicits iPMF via a TNFα-dependent increase in spinal aPKC activity.

Key points

  • A central neural apnoea in the absence of hypoxia elicits a form of respiratory plasticity known as inactivity-induced phrenic motor facilitation (iPMF), a rebound increase in phrenic burst amplitude when central respiratory neural activity is restored.

  • iPMF requires spinal atypical protein kinase C (aPKC) activity in spinal segments encompassing the phrenic motor nucleus.

  • Here, we report novel findings that tumour necrosis factor-α (TNFα) signalling in or near the phrenic motor pool is necessary and sufficient for iPMF as: (1) spinal TNFα inhibition inhibits iPMF; and (2) spinal TNFα elicits long-lasting increases in phrenic burst amplitude via an aPKC-dependent mechanism.

  • These data are consistent with the hypothesis that local mechanisms operating within or near the phrenic motor pool sense and respond to reduced respiratory neural activity, and suggest that TNFα-induced activation of aPKC near phrenic motor neurons forms part of the core cellular pathway giving rise to iPMF.

Introduction

A prolonged reduction in central respiratory neural activity elicits a novel form of respiratory plasticity known as inactivity-induced phrenic motor facilitation (iPMF), a ‘rebound’ increase in phrenic burst amplitude apparent when respiratory neural activity is restored (Baker-Herman & Strey, 2011; Mahamed et al. 2011; Strey et al. 2012; Baertsch & Baker-Herman, 2013). Although central neural apnoea results in reduced activity in many respiratory-related neurons throughout the neuraxis, iPMF requires spinal mechanisms, as inhibition of spinal atypical protein kinase C (aPKC) activity (specifically PKCζ and/or PKCι/λ isoforms; abbreviated here as PKCζ/ι for clarity) blocks iPMF (Strey et al. 2012). Thus, we hypothesized that mechanisms operating within the phrenic motor pool sense and respond to reductions in phrenic neural activity (Baker-Herman & Strey, 2011; Strey et al. 2012). Upstream signal(s) leading to spinal PKCζ/ι activation following reduced respiratory neural activity are unknown.

The pro-inflammatory cytokine tumour necrosis factor-α (TNFα) plays a prominent role in inactivity-induced plasticity in the hippocampus (Stellwagen & Malenka, 2006) and visual cortex (Kaneko et al. 2008). For example, a prolonged reduction in hippocampal neuron activity induces a TNFα-dependent increase in AMPA receptor membrane expression and synaptic transmission in vitro (Beattie et al. 2002; Stellwagen & Malenka, 2006). Although a prolonged (>24 h) period of activity deprivation is required to elicit TNFα-dependent increases in hippocampal synaptic transmission (Stellwagen & Malenka, 2006), we hypothesized that inactivity-induced plasticity operates on a much faster time scale in the phrenic motor pool, which must make activity-dependent adjustments within minutes (not days) to sustain life (Baker-Herman & Strey, 2011; Mahamed et al. 2011; Strey et al. 2012; Baertsch & Baker-Herman, 2013). TNFα has the potential to elicit rapid changes in excitatory neurotransmission, as exogenous TNFα increases membrane AMPA receptor expression and synaptic transmission within minutes of application (Beattie et al. 2002; Stellwagen et al. 2005), including within spinal motor neurons (Ferguson et al. 2008; Han & Whelan, 2010). Mechanisms downstream of TNFα that give rise to enhanced synaptic transmission are not well understood.

Here, we tested the hypothesis that TNFα plays a key role in inactivity-induced plasticity within the phrenic motor pool (i.e. iPMF) via spinal aPKC-dependent mechanisms. We show that: (1) spinal TNFα inhibition blocks long-lasting iPMF following neural apnoea; (2) spinal TNFα receptor activation (in the absence of neural apnoea) is sufficient to elicit phrenic motor facilitation (pMF); and (3) spinal TNFα-induced pMF requires spinal aPKC activity. Spinal TNFα inhibition did not block early transient frequency facilitation following neural apnoea, consistent with the hypothesis that frequency facilitation arises from mechanisms distinct from those of iPMF. Together, our data support the hypothesis that reduced respiratory neural activity elicits iPMF via a TNFα-dependent increase in spinal aPKC activity.

Methods

Animals

Experiments were performed on 2–4-month-old male Sprague-Dawley rats (Harlan colony 217). Rats were housed two per cage in a controlled environment (12 h light/12 h dark cycle), with food and water ad libitum. The Institutional Animal Care and Use Committee at the University of Wisconsin, Madison, WI, USA approved all experiments.

Surgical preparation

Isoflurane anaesthesia was induced within a closed chamber, and continued through a nose cone (2.5–3.5% in 50% O2, balance N2). A tracheal catheter was placed to enable mechanical ventilation (tidal volume, 2–3 ml; Rodent Ventilator 683, Harvard Apparatus, Holliston, Massachusetts); isoflurane was then administered through the tracheal cannula. A bilateral vagotomy was performed at the cervical level to prevent entrainment of respiratory frequency with the ventilator. End-tidal CO2 was monitored (Respironics Novametrix, Murrysville, Pennsylvania) and maintained at approximately 45 mmHg throughout surgery by adjusting the ventilator rate and/or adding CO2 to the inspired gas mix. Tracheal pressure was monitored throughout surgery to ensure that rats continued to generate respiratory efforts during surgery and to avoid unintended neural apnoea. The femoral artery and tail vein were catheterized for blood pressure measurement, blood gas sampling and fluid infusion. Rats were then converted to urethane anaesthesia (1.7–1.8 g kg−1, i.v.) and isoflurane was discontinued. The left phrenic nerve was exposed via a dorsal approach, cut distally, desheathed and submerged in mineral oil. A laminectomy was performed at C2, and a small hole was cut in the dura. A soft silicone catheter (2 French, Access Technologies, Skokie, Illinois) connected to a 50 μl Hamilton syringe was placed in the intrathecal space and advanced to spinal segment C4. For experiments in which TNFα and ζ-pseudosubstrate inhibitory peptide (PKCζ-PS) or scrambled ζ-pseudosubstrate inhibitory peptide (scrPKCζ-PS) were delivered (see below), two catheters connected to two different Hamilton syringes were placed. Rats were subjected to neuromuscular blockade with pancuronium bromide (2.5 mg kg−1, i.v.). Fluid infusion of 1–3 ml h−1 of 1: 4 hetastarch (Hespan, 6% hetastarch in 0.9% sodium chloride) and lactated Ringer solution was started, and continued throughout the experiment. Body temperature was measured using a rectal probe and maintained at 37°C with a custom-designed heated table.

Following surgery, the phrenic nerve was placed on a bipolar silver electrode and submerged in mineral oil. Phrenic activity was amplified (×10k) and band-pass filtered (0.3–20 kHz). Raw signals were recorded with a sampling rate of 10,000 s−1 and integrated (time constant, 50 ms) using the PowerLab (AD Instruments, New Zealand) data acquisition system.

Blood gases were sampled and analysed (Radiometer, Copenhagen, Denmark, ABL 500) at key time points in the protocol (described below) to confirm that Inline graphic remained above 150 mmHg throughout the protocol and that post-neural apnoea Inline graphic levels were within 1.5 mmHg of baseline. Blood pressure was monitored continuously throughout the protocol, and the depth of anaesthesia was assessed periodically by monitoring the pressor responses to paw pad pinch.

Electrophysiological protocols

Approximately 1 h after conversion to urethane anaesthesia, baseline phrenic nerve activity was set using one of two methods. In some rats, apnoeic and recruitment thresholds for phrenic activity were determined by slowly increasing the ventilator rate and/or lowering the inspired CO2 until rhythmic phrenic burst activity ceased (apnoeic threshold); the ventilator rate was then lowered and/or inspired CO2 was increased until phrenic activity resumed (recruitment threshold). End-tidal CO2 was raised 2 mmHg above the recruitment threshold to establish baseline phrenic discharge. In other rats, apnoeic and recruitment thresholds for phrenic bursting were not determined in order to avoid exposure to a neural apnoea prior to the protocol; instead, baseline phrenic nerve activity was set by raising/lowering the ventilator rate and/or inspired CO2 until the phrenic burst frequency was between approximately 45 and 48 bursts min−1. The distributions of rats in which the apnoeic threshold was determined prior to the protocol were as follows: 5/12 rats receiving vehicle prior to neural apnoea, 6/11 rats receiving recombinant human soluble TNF receptor 1 (sTNFR1) prior to neural apnoea and 0/6 sTNFR1 time controls; 5/10 rats receiving intrathecal 25 ng TNFα and 7/7 rats receiving vehicle. No differences in iPMF or pMF magnitude were apparent using either method (P > 0.05), and so the groups were combined. The apnoeic threshold was not determined prior to the protocol for any rats exposed to intrathecal TNFα followed by either scrPKCζ-PS or PKCζ-PS. After 20–30 min of stable phrenic burst amplitude and frequency, an arterial blood sample was drawn; arterial Inline graphic and phrenic burst activity at this time point were considered ‘baseline’ for all subsequent measurements. Rats were then subjected to one of the following experimental series: (1) artificial cerebrospinal fluid (aCSF) prior to neural apnoea vs. sTNFR1 prior to neural apnoea vs. sTNFR1 prior to time control; (2) intrathecal aCSF vs. intrathecal TNFα without neural apnoea; or (3) intrathecal TNFα followed by scrPKCζ-PS vs. intrathecal TNFα followed by PKCζ-PS (see below for further description, Minneapolis, Minnesota).

Compounds were prepared in aCSF (in mm: 120 NaCl, 3 KCl, 2 CaCl2, 2 MgCl2, 23 NaHCO3, 10 glucose bubbled with 95% O2/5% CO2). Rats received sTNFR1 [1.3 μg with 1% bovine serum albumin (BSA); R & D Systems, Minneapolis, Minnesota], recombinant rat TNFα (12.5–100 ng with 1% BSA; Sigma-Aldrich, St. Louis, Missouri), myristoylated PKCζ-PS (20 μg; Tocris Bioscience, Bristol, UK) or myristoylated scrPKCζ-PS (20 μg; Tocris Bioscience, Bristol, UK) in five 2 μl boluses delivered in the C4 intrathecal space over 2 min (total of 10 μl delivered). Vehicle-treated rats received aCSF with 1% BSA.

Experimental series (1): The role of spinal TNFα in iPMF was assessed. Rats received a 10 μl intrathecal injection of vehicle (aCSF with 1% BSA; n= 12) or sTNFR1 (1 μg; n= 11) approximately 15 min prior to the onset of neural apnoea. Neural apnoea was induced by increasing the ventilator rate and/or decreasing the inspired CO2 until phrenic bursting ceased. Inline graphic was then clamped approximately 5 mmHg below the CO2 apnoeic threshold. Neural apnoea was maintained for 30 min, and then phrenic neural activity was resumed by returning Inline graphic to baseline values. An additional ∼3 μl (0.3 μg) of sTNFR1 was delivered approximately 5 min prior to the termination of neural apnoea. As rats were mechanically ventilated, arterial oxygen levels were maintained throughout the neural apnoea. Blood gas analysis confirmed that arterial oxygen levels were above 150 mmHg throughout the protocol, and that Inline graphic returned to within 1.5 mmHg of baseline following the restoration of central respiratory neural activity. To control for potential time-dependent changes in phrenic motor output not related to our treatment, results were compared with time controls that received similar surgery and experimental duration, but no neural apnoea. Time control rats received intrathecal sTNFR1 at corresponding time points (n= 7).

Experimental series (2): The sufficiency of TNFα for pMF was assessed. In the absence of neural apnoea, rats received intrathecal injections of TNFα. In preliminary studies, we tested the response to 12.5 (n= 5), 25 (n= 5), 50 (n= 6) or 100 ng (n= 5) TNFα. A TNFα dose of 25 ng was chosen to complete our studies, and an additional five rats were collected to complete a time course of TNFα-induced changes in phrenic motor output. Rats receiving 25 ng TNFα were compared with rats receiving vehicle (aCSF with 1% BSA; n= 7).

Experimental series (3): The role of aPKC in TNFα-induced pMF was assessed. In the absence of neural apnoea, rats received 25 ng intrathecal TNFα, followed by intrathecal PKCζ-PS (n= 7) 35 min later. These rats were compared with rats receiving control injections of scrPKCζ-PS (n= 6) 35 min following intrathecal TNFα.

Phrenic burst activity was monitored continuously before, during and for 60 min following neural apnoea or intrathecal TNFα, and up to 90 min in rats receiving intrathecal TNFα and PKCζ-PS injections. Arterial blood samples were collected and analysed before, during and 5, 15, 30 and 60 min after neural apnoea, intrathecal TNFα or equivalent duration in time controls. Blood gases were sampled before and 15, 30, 60 and 90 min following intrathecal TNFα injection in rats receiving TNFα and PKCζ-PS or scrPKCζ-PS. Blood gas analysis ensured adequate maintenance of baseline arterial Inline graphic, Inline graphic, standard base excess (SBE) and pH throughout the protocol. At the end of each protocol, a maximal CO2 response (90 < Inline graphic < 100 mmHg) was assessed to ensure that the observed responses were not influenced by deterioration of the preparation. Four rats with a maximal CO2 response <40% baseline were excluded from the analysis. Animals were then killed with a urethane overdose.

Statistical analysis

Integrated phrenic burst amplitude and frequency were analysed in 30–60 s bins before (baseline) and 5, 15, 30 and 60 min (and 90 min in rats receiving TNFα and PKCζ-PS or scrPKCζ-PS injections) following treatments (equivalent duration in time controls). In all rats, the integrated phrenic burst amplitude following the restoration of respiratory neural activity post-neural apnoea (or equivalent duration in time controls) was expressed as a percentage change from baseline (%baseline), whereas burst frequency (breaths min−1) post-neural apnoea was expressed as an absolute change from baseline (Δbaseline). A two-way repeated measures analysis of variance (ANOVA) was used to detect significant differences (Prism 5, GraphPad Software, La Jolla, California) from baseline and between each group. Three experimental series were analysed independently: (1) rats receiving sTNFR1 prior to neural apnoea were compared with rats receiving vehicle prior to neural apnoea and time controls receiving intrathecal sTNFR1; (2) rats receiving intrathecal TNFα in the absence of neural apnoea were compared with rats receiving vehicle; and (3) rats receiving intrathecal TNFα and PKCζ-PS were compared with those receiving intrathecal TNFα and scrPKCζ-PS. When the effects were significant, two-way repeated measures ANOVA was followed by Bonferroni post hoc tests. Differences were considered to be significant if P < 0.05.

Arterial Inline graphic, Inline graphic, pH, SBE and mean arterial pressure (MAP) were analysed using a two-way repeated measures ANOVA (Prism 5, GraphPad Software) and Bonferroni post hoc tests before (baseline) and 15, 30, 60 and 90 min following neural apnoea, TNFα injections or an equivalent duration in time controls [90 min time point was only analysed in rats in experimental series (3)]. Differences were considered to be significant if P < 0.05. Data are presented as mean ± SEM.

Results

Regulation of physiological variables

Table 1 lists the average arterial Inline graphic, Inline graphic, SBE, pH and MAP during baseline and following treatments for each experimental series. No time-dependent changes in Inline graphic or Inline graphic were apparent in any rat group. Importantly, in all rat groups, Inline graphic was maintained above 200 mmHg throughout the protocol, and Inline graphic post-neural apnoea was maintained within ±1.5 mmHg of baseline; thus, changes in phrenic nerve activity as a result of our treatments cannot be attributed to changes in chemoreceptor feedback.

Table 1.

Arterial Inline graphic (mmHg), Inline graphic (mmHg), pH, standard base excess (SBE) and mean arterial pressure (MAP, mmHg) before and after treatment in all rat groups. Data presented as mean ± SEM. No significant differences were detected at any time point between groups within each experimental series

Treatment group Time Inline graphic Inline graphic pH SBE MAP
Spinal TNFα is necessary for iPMF
Neural apnoea + vehicle, n= 12 Baseline 260 ± 14 42.9 ± 1.1 7.39 ± 0.01 0.8 ± 0.4 119 ± 13
15 min 271 ± 10 42.7 ± 1.1 7.38 ± 0.01 −0.0 ± 0.6 109 ± 15*
30 min 269 ± 10 43 ± 1.1 7.37 ± 0.01 −0.3 ± 0.5 103 ± 15*
60 min 272 ± 11 43.2 ± 1.1 7.36 ± 0.01* −0.9 ± 0.6* 97 ± 14*
Neural apnoea + sTNFR1, n= 11 Baseline 254 ± 14 43.3 ± 1.4 7.39 ± 0.01 1.2 ± 0.5 129 ± 11
15 min 262 ± 10 43 ± 1.6 7.39 ± 0.02 0.9 ± 0.6 118 ± 10*
30 min 263 ± 11 43.4 ± 1.5 7.39 ± 0.01 1.3 ± 0.3 118 ± 11*
60 min 264 ± 11 42.9 ± 1.4 7.39 ± 0.01 1.0 ± 0.2 108 ± 12*
Time control + sTNFR1, n= 6 Baseline 251 ± 5 42.2 ± 1.5 7.40 ± 0.01 0.7 ± 0.5 116 ± 10
15 min 250 ± 4 41.7 ± 1.6 7.41 ± 0.01 0.5 ± 0.6 104 ± 9
30 min 247 ± 7 42.2 ± 1.5 7.40 ± 0.01 0.6 ± 0.6 102 ± 8*
60 min 244 ± 2 42.2 ± 1.5 7.40 ± 0.01 0.7 ± 0.5 95 ± 7*
Spinal TNFα is sufficient for pMF
25 ng TNFα, n= 10 Baseline 251 ± 9 44.0 ± 1.6 7.38 ± 0.02 1.0 ± 0.6 125 ± 8
15 min 248 ± 8 44.2 ± 1.6 7.38 ± 0.02 0.5 ± 0.5 119 ± 9
30 min 243 ± 7 44 ± 1.6 7.37 ± 0.02 0.2 ± 0.6 112 ± 10*
60 min 246 ± 8 44.3 ± 1.7 7.37 ± 0.02* 0.2 ± 0.6 108 ± 9*
Vehicle, n= 7 Baseline 244 ± 8 40.4 ± 2.2 7.40 ± 0.01 0.4 ± 0.8 111 ± 14
15 min 243 ± 9 40.6 ± 2.2 7.41 ± 0.01 0.6 ± 0.8 110 ± 12
30 min 241 ± 11 40.4 ± 2.3 7.40 ± 0.02 0.3 ± 0.8 106 ± 14
60 min 238 ± 13 40.4 ± 2.0 7.40 ± 0.02 −0.1 ± 0.8 98 ± 14*
Spinal aPKC is necessary for TNFα-induced pMF
TNFα+ PKCζ-PS, n= 7 Baseline 284 ± 7 42.9 ± 0.9 7.37 ± 0.01 −0.7 ± 0.8 132 ± 7
15 min 287 ± 12 43 ± 1.1 7.36 ± 0.01 −1.2 ± 0.7 134 ± 6
30 min 285 ± 12 43 ± 0.9 7.35 ± 0.01 −1.4 ± 0.6 130 ± 1
60 min 281 ± 13 43 ± 1.0 7.36 ± 0.01 −0.8 ± 0.7 122 ± 3
90 min 268 ± 22 43.0 ± 0.8 7.36 ± 0.01 −0.9 ± 0.6 113 ± 8*
TNFα+ scrPKCζ−PS, n= 6 Baseline 289 ± 4 45.6 ± 1.5 7.36 ± 0.01 0.1 ± 0.4 131 ± 6
15 min 284 ± 5 45.5 ± 1.7 7.35 ± 0.01 −0.9 ± 0.4 128 ± 7
30 min 283 ± 5 45.4 ± 1.8 7.35 ± 0.02 −0.8 ± 0.3 122 ± 8
60 min 278 ± 5 45.1 ± 1.8 7.34 ± 0.02 −1.5 ± 0.3 113 ± 8*
90 min 278 ± 6 46.3 ± 1.4 7.34 ± 0.02* −1.1 ± 0.6 105 ± 10*
Open in a new tab

aPKC, atypical protein kinase C; iPMF, inactivity-induced phrenic motor facilitation; PKCζ-PS, ζ-pseudosubstrate inhibitory peptide; pMF, phrenic motor facilitation; scrPKCζ-PS, scrambled ζ-pseudosubstrate inhibitory peptide; sTNFR1, recombinant human soluble TNF receptor 1; TNFα, tumour necrosis factor-α. *Significantly different from baseline.

Similar to other studies using this experimental preparation (Baker-Herman & Mitchell, 2008; Baertsch & Baker-Herman et al. 2013; Dale & Mitchell, 2013), pH, SBE and MAP decreased significantly from baseline over the course of the protocol in some rat groups (Table 1; P < 0.05), but no significant differences in pH, SBE or MAP were apparent between rat groups (P > 0.05). Time-dependent decreases in pH or SBE were only observed at the last blood sample (i.e. 60 or 90 min), and were not thought to contribute to our results as: (1) iPMF was observed without decreases in pH or SBE at earlier time points (i.e. 15 and 30 min post-neural apnoea); and (2) iPMF at 60 min has been reported in other studies without significant decreases in pH or SBE (Mahamed et al. 2011; Strey et al. 2012). Time-dependent decreases in MAP were also not considered to be physiologically relevant in the context of our study as: (1) all rat groups exhibited a similar decrease in MAP (including time controls); (2) other studies from our laboratory have reported iPMF without a significant change in MAP (Mahamed et al. 2011; Baertsch & Baker-Herman, 2013); and (3) decreases in MAP were small (∼25 mmHg) and changes in MAP of this magnitude have minimal long-lasting effects on respiratory activity in rats (Walker & Jennings, 1998).

Spinal TNFα is necessary for iPMF

To test the hypothesis that spinal TNFα is necessary for iPMF, we delivered sTNFR1 to the intrathecal space at the level of the phrenic motor nucleus approximately 15 min prior to neural apnoea. sTNFR1 is a soluble form of the TNF receptor 1 that scavenges TNFα and prevents it from binding to endogenous TNFα receptors (Gray et al. 1990). Representative compressed phrenic neurogrammes are shown in Fig. 1A, illustrating phrenic burst amplitude during baseline, a 30 min neural apnoea, and for 60 min following restoration of respiratory neural activity, and for a similar duration in time control rats receiving sTNFR1. The average percentage change in phrenic burst amplitude from baseline for 60 min following the restoration of respiratory neural activity after neural apnoea (or equivalent points in time controls) is shown in Fig. 1B. Time control rats expressed no change in phrenic burst amplitude from baseline at any point in the protocol (all P > 0.05; Fig. 1A and B), suggesting that our preparation was stable within the recording period and that sTNFR1 injection did not change baseline phrenic burst amplitude. In rats receiving vehicle injections prior to neural apnoea, phrenic burst amplitude was significantly increased relative to baseline and time controls at all time points following the restoration of respiratory neural activity (all P < 0.05; Fig. 1A and B), indicating iPMF. By contrast, rats receiving intrathecal sTNFR1 prior to neural apnoea did not express significantly increased phrenic burst amplitude relative to time controls at any time point following the restoration of respiratory neural activity (all P > 0.05; Fig. 1A and B); however, a small, transient increase in phrenic burst amplitude relative to baseline was observed shortly after the restoration of respiratory neural activity (5 min; P < 0.05), which was not significantly different from rats receiving vehicle injections prior to neural apnoea (P > 0.05; Fig. 1B). By 15 min post-neural apnoea, phrenic burst amplitude in sTNFR1 rats had returned to baseline (P > 0.05) and was significantly lower than in rats receiving vehicle prior to neural apnoea (P < 0.05), where it remained for the duration of the protocol. These data suggest that spinal TNFα inhibition attenuates early iPMF and eliminates late iPMF.

Figure 1. Spinal tumour necrosis factor-α (TNFα) is necessary for inactivity-induced phrenic motor facilitation (iPMF).

Figure 1

Open in a new tab

A, representative compressed phrenic neurogrammes before, during and 60 min following a 30 min neural apnoea in rats pretreated with intrathecal vehicle (top) or recombinant human soluble TNF receptor 1 (sTNFR1) (middle) or an equivalent duration in time controls (bottom), illustrating that sTNFR1 attenuates iPMF. B, average percentage change in phrenic burst amplitude from baseline for 60 min following the resumption of respiratory neural activity in rats receiving intrathecal vehicle (squares) or sTNFR1 (circles) prior to neural apnoea, or equivalent duration in time controls (diamonds). Pretreatment with sTNFR1 prior to neural apnoea attenuates and shortens iPMF. C, phrenic burst amplitude at 5, 30 and 60 min following the resumption of respiratory neural activity in individual rats pretreated with vehicle (left) or sTNFR1 (middle) prior to neural apnoea, or an equivalent duration in time controls (right). D, average change in phrenic burst frequency from baseline (Δbaseline; in breaths min−1) for 60 min following resumption of respiratory neural activity in rats receiving intrathecal vehicle (squares) or sTNFR1 (circles) prior to neural apnoea, or equivalent duration in time controls (diamonds), illustrating that rats pretreated with vehicle or sTNFR1 prior to neural apnoea expressed a similar increase in phrenic burst frequency immediately following the resumption of respiratory neural activity. Collectively, these data indicate that spinal TNFα inhibition attenuates and shortens iPMF following neural apnoea. Mean values ± SEM. Filled symbols indicate significantly different from baseline; *significantly different from time controls; #significantly different from sTNFR1 rats; P < 0.05.

As reported previously (Mahamed et al. 2011), central neural apnoea elicited an increase in phrenic burst frequency that was apparent immediately upon restoration of respiratory neural activity (5 min; Fig. 1D), relative to baseline and time controls (P < 0.05); by 15 min following resumption of respiratory neural activity, phrenic burst frequency was no longer significantly different from that of time controls, although it remained significantly increased relative to baseline for the duration of the experiment (P > 0.05). Similar frequency facilitation was observed in rats receiving intrathecal sTNFR1 prior to neural apnoea (5 min: P < 0.05, relative to baseline or time controls); by 15 min following resumption of respiratory neural activity, phrenic burst frequency was no longer significantly different from that of time controls (P > 0.05), although it remained significantly increased relative to baseline (P < 0.05). These data suggest that spinal TNFα inhibition does not impair frequency facilitation following neural apnoea.

Spinal TNFα is sufficient for pMF

To determine whether spinal TNFα is sufficient to elicit pMF, a subgroup of rats received recombinant TNFα in the intrathecal space at the level of the phrenic motor pool. The average percentage change in phrenic burst amplitude from baseline at 60 min following TNFα injections in preliminary dose–response experiments is shown in Fig. 2A. As expected, no change in phrenic burst amplitude from baseline was observed in rats receiving intrathecal injections of vehicle. Intrathecal administration of 25 or 50 ng TNFα elicited a significant increase in phrenic burst amplitude relative to baseline (P < 0.05; Fig. 2A); however, only 25 ng TNFα elicited phrenic burst amplitude increases that were significantly higher than in rats receiving vehicle injections (P < 0.05). Following the administration of the lowest (12.5 ng) or highest (100 ng) doses of TNFα, phrenic burst amplitude was not significantly different relative to baseline or rats receiving vehicle (P > 0.05). No significant differences in phrenic burst frequency from baseline or vehicle-treated rats were observed with any dose of TNFα (data not shown; P > 0.05).

Figure 2. Tumour necrosis factor-α (TNFα) is sufficient to induce phrenic motor facilitation (pMF).

Figure 2

Open in a new tab

A, average change in phrenic burst amplitude from baseline 60 min following injections of different doses of TNFα, illustrating that TNFα elicits pMF within a narrow dose range. Black bars indicate significantly different from baseline. B, average percentage change in phrenic burst amplitude from baseline for 60 min following intrathecal TNFα (25 ng; squares) or vehicle (triangles), illustrating that TNFα elicits increases in phrenic burst amplitude from baseline as early as 15 min following infusion. C, phrenic burst amplitude at 15, 30 and 60 min following intrathecal vehicle (left) or TNFα (right) in individual rats. D, average change in phrenic burst frequency from baseline (Δbaseline; in breaths min−1) for 60 min following intrathecal TNFα (squares) or vehicle (triangles). Collectively, these data indicate that spinal TNFα elicits pMF in the absence of neural apnoea. Mean values ± SEM. Filled symbols indicate significantly different from baseline; *significantly different from vehicle; P < 0.05.

The time course of phrenic burst amplitude changes (%baseline) from 5 min to 60 min post-TNFα (25 ng) or vehicle injections is shown in Fig. 2B. Increased phrenic burst amplitude relative to baseline was apparent within 15 min following intrathecal TNFα (P < 0.05; Fig. 2B); by 30 min following intrathecal TNFα, phrenic burst amplitude was significantly increased relative to baseline and rats receiving vehicle injections (P < 0.05), where it remained for the duration of the experiment.

No changes in phrenic burst frequency from baseline were observed in vehicle-treated rats for up to 60 min following vehicle injections (P > 0.05; Fig. 2D). Similarly, phrenic burst frequency was not significantly different from baseline in rats receiving intrathecal 25 ng TNFα at 15 and 30 min following TNFα infusion (P > 0.05), although a small increase relative to baseline was observed at 60 min (P < 0.05). Phrenic burst frequency was not significantly different between rats receiving vehicle or TNFα at any time point following infusion (P > 0.05). Collectively, these data suggest that intrathecal TNFα elicits a dose-dependent pMF that is expressed primarily as an increase in phrenic burst amplitude.

Spinal aPKC is necessary for TNFα-induced pMF

We next sought to determine whether TNFα-induced pMF is similar to iPMF in the requirement for spinal aPKC activity. As spinal aPKC activity is required immediately following a 30 min neural apnoea to transition from a short-term labile form of plasticity to a long-lasting iPMF (Strey et al. 2012), we delivered an aPKC inhibitor 35 min following the induction of TNFα-dependent pMF. In subgroups of rats, intrathecal TNFα (25 ng) was delivered, followed, 35 min later, by intrathecal delivery of cell-permeable PKCζ-PS, which inhibits the catalytic domain of all aPKC isoforms (House & Kemp, 1987; Selbie et al. 1993; Laudanna et al. 1998; Yao et al. 2013), or a non-targeting, scrambled version of the peptide (scrPKCζ-PS). Representative compressed phrenic neurogrammes are shown in Fig. 3A, illustrating phrenic burst amplitude before and for 90 min following intrathecal TNFα in rats treated with intrathecal PKCζ-PS or scrPKCζ-PS. The average percentage change in phrenic burst amplitude for 90 min following intrathecal TNFα is shown in Fig. 3B. As expected, in both groups of rats, phrenic burst amplitude was significantly increased from baseline as early as 15 min following intrathecal TNFα (P < 0.05; Fig. 3B); 35 min following TNFα injections, rats received intrathecal scrPKCζ-PS or PKCζ-PS. As expected, intrathecal scrPKCζ-PS had no effect on TNFα-induced pMF, as phrenic burst amplitude continued to rise over the next hour; indeed, phrenic burst amplitude at 60 and 90 min post-intrathecal TNFα in scrPKCζ-PS-treated rats was significantly increased relative to baseline and the 15 and 30 min time points (prior to scrPKCζ-PS; P < 0.05). By contrast, in rats receiving intrathecal PKCζ-PS, phrenic burst amplitude 60 and 90 min post-intrathecal TNFα was not significantly different from the 30 min time point (immediately prior to PKCζ-PS; P > 0.05), although phrenic burst amplitude remained significantly elevated relative to baseline (P < 0.05). In addition, phrenic burst amplitude 60 and 90 min post-TNFα in rats receiving PKCζ-PS was significantly lower than in rats receiving scrPKCζ-PS (P < 0.05).

Figure 3. Tumour necrosis factor-α (TNFα) elicits phrenic motor facilitation (pMF) via an atypical protein kinase C (aPKC)-dependent mechanism.

Figure 3

Open in a new tab

A, representative compressed phrenic neurogrammes before and for 90 min following intrathecal TNFα, illustrating that a rat receiving a control injection of scrambled ζ-pseudosubstrate inhibitory peptide (scrPKCζ-PS, top) 35 min following intrathecal TNFα continued to express a prolonged increase in phrenic burst amplitude, whereas intrathecal ζ-pseudosubstrate inhibitory peptide (PKCζ-PS, bottom) returned phrenic burst amplitude toward baseline. Grey arrow indicates time point for TNFα delivery (t= 0 min), whereas black arrow indicates time point for scrPKCζ-PS or PKCζ-PS delivery (t= 35 min). B, average percentage change in phrenic burst amplitude from baseline for 90 min following intrathecal TNFα (25 ng) in rats receiving intrathecal scrPKCζ-PS (squares) or PKCζ-PS (circles) 35 min following TNFα, indicating that TNFα-induced pMF was impaired by spinal aPKC inhibition. Black arrow indicates the time point of scrPKCζ-PS or PKCζ-PS delivery (t= 35 min). C, phrenic burst amplitude at 15, 30 and 60 min following intrathecal TNFα in individual rats receiving scrPKCζ-PS (left) or PKCζ-PS (right). D, average change in phrenic burst frequency from baseline (Δbaseline; in breaths min−1) for 90 min following intrathecal TNFα in rats receiving intrathecal scrPKCζ-PS (squares) or PKCζ-PS (circles) 35 min following TNFα delivery. Black arrow indicates the time point of scrPKCζ-PS or PKCζ-PS delivery. Collectively, these data suggest that spinal aPKC inhibition arrests TNFα-induced pMF. Mean values ± SEM. Filled symbols indicate significantly different from baseline; *significantly different from rats receiving scrPKCζ-PS; #significantly different from pre-PKCζ-PS time points; P < 0.05.

No changes in phrenic burst frequency from baseline were observed for up to 90 min following TNFα injection in rats receiving PKCζ-PS (P > 0.05; Fig. 3D). Similar to TNFα injection alone (Fig. 2D), a slight, but significant, increase in phrenic burst frequency from baseline was observed at 60 and 90 min following TNFα injection in rats receiving scrPKCζ-PS (P < 0.05). Phrenic burst frequency was not significantly different between rats receiving TNFα and PKCζ-PS or TNFα and scrPKCζ-PS at any time point (P > 0.05). Collectively, these data suggest that spinal aPKC inhibition arrests TNFα-induced pMF.

Discussion

Here, we report a key role for TNFα in the mechanism underlying iPMF, a novel form of plasticity in phrenic burst amplitude induced by a reduction in respiratory neural activity. Inhibition of spinal TNFα signalling prior to neural apnoea blocked long-lasting iPMF, with only a transient, attenuated increase in phrenic burst amplitude remaining following the restoration of respiratory neural activity. Spinal TNFα inhibition did not block frequency facilitation following neural apnoea, suggesting that, similar to other forms of respiratory plasticity (Powell et al. 1998; Blitz & Ramirez, 2002; Baker-Herman & Mitchell, 2008), mechanisms giving rise to iPMF are distinct from those giving rise to burst frequency plasticity. Collectively, these results suggest that iPMF occurs via spinal mechanisms and that iPMF consists of at least two phases: (1) an early, labile phase; and (2) a late, TNFα-dependent phase. Exogenous spinal TNFα increased phrenic burst amplitude in the absence of neural apnoea (i.e. pMF), demonstrating that TNF receptor activation is sufficient to elicit pMF. Finally, TNFα-induced pMF required spinal aPKC activity, consistent with the interpretation that a common aPKC-dependent mechanism underlies both TNFα-induced pMF and iPMF (Strey et al. 2012).

Inactivity-induced plasticity in the phrenic motor system

Phrenic motor neurons receive rhythmic excitatory drive from neurons in the medulla throughout life. The inhibition of these descending inputs elicits plasticity within the phrenic motor system (Castro-Moure & Goshgarian, 1996, 1997). For example, a transient unilateral C2 axon conduction block disrupts descending inputs to the ipsilateral phrenic pool; following C2 axon conduction block reversal, diaphragm electromyogram activity is significantly increased above baseline in the hemidiaphragm ipsilateral (but not contralateral) to the conduction block (Castro-Moure & Goshgarian, 1996), suggesting that the reduction of synaptic inputs elicits plasticity locally within phrenic/diaphragm motor circuits. Although it is not known whether these changes result from inactivity-induced plasticity in phrenic spinal circuits versus diaphragm neuromuscular junctions, rapid morphological changes are observed in the ipsilateral phrenic motor nucleus, consistent with a spinal synaptic mechanism (Castro-Moure & Goshgarian, 1997).

The reduction of descending respiratory neural activity via a central neural apnoea also elicits pMF (i.e. iPMF), consistent with a spinal (not neuromuscular junction) mechanism (Strey et al. 2012). As similar iPMF can be elicited following multiple models of central neural apnoea with different mechanisms of action (hyperventilation, vagal inhibitory feedback, isoflurane-induced respiratory depression; Mahamed et al. 2011), we suggested that iPMF arises from a common factor: reduced respiratory motor activity (Baker-Herman and Strey, 2011; Mahamed et al. 2011). Indeed, reversible blocking of spinal synaptic inputs to some phrenic motor neurons via spinal axon conduction block with procaine elicits phenotypically similar iPMF that requires spinal TNFα and aPKC activity (Strey and Baker-Herman, 2012). Collectively, these data suggest that reduced synaptic inputs to phrenic motor neurons elicit a rapid form of plasticity that serves to strengthen phrenic motor output. The mechanisms whereby local activity is sensed within the phrenic motor pool are not yet understood.

Neural apnoea also elicits a small, transient increase in respiratory burst frequency (Mahamed et al. 2011; Baertsch & Baker-Herman, 2013), which we suggest arises at a different locus from iPMF in the CNS. In many other models of respiratory plasticity, burst frequency plasticity is suggested to arise from brainstem respiratory rhythm generating neurons (Powell et al. 1998; Blitz & Ramirez, 2002; Baker-Herman & Mitchell, 2008), whereas phrenic burst amplitude plasticity arises largely from processes in the spinal cord (Baker-Herman & Mitchell, 2002; Baker-Herman et al. 2004; Bocchiaro & Feldman, 2004; Wilkerson et al. 2008; MacFarlane & Mitchell, 2009; Tadjalli et al. 2010; Dale-Nagle et al., 2011; Dale et al., 2012; Hoffman et al. 2012; Nichols et al. 2012; Strey et al. 2012). Our data suggest a similar separation between mechanisms giving rise to respiratory frequency plasticity versus iPMF. Indeed, spinal TNFα inhibition did not impair the early, transient respiratory frequency facilitation following the resumption of respiratory neural activity. Similarly, the respiratory burst frequency following exogenous application of TNFα to the spinal cord was not significantly different from the response following intrathecal vehicle. However, a small, late increase in burst frequency from baseline (i.e. burst frequency prior to TNFα) was observed in rats receiving intrathecal TNFα or intrathecal TNFα and scrPKCζ-PC; this slight increase in burst frequency from baseline may be characteristic of the preparation (anaesthetized, vagotomized rat subjected to neuromuscular blockade), as even time controls that do not receive any treatment per se will often show burst frequency rising slowly over time (see Figs 1D and 2D; for a review, see Baker-Herman & Mitchell, 2008). Nevertheless, it remains possible that spinal TNFα may elicit modest effects on burst frequency via indirect effects on respiratory rhythm generation (Baker-Herman & Mitchell, 2002).

Role of TNFα in inactivity-induced plasticity

TNFα is a pro-inflammatory cytokine that plays key roles in normal CNS function and CNS pathology (Park & Bowers, 2010; Santello & Volterra, 2012). These seemingly contradictory roles of TNFα appear to be a function of concentration, receptor subtype activated and/or the specific array of adaptor proteins and signal transduction molecules expressed within the affected cells (Santello & Volterra, 2012). Among the many roles for TNFα in the healthy CNS, TNFα modulates neuronal excitability via bidirectional regulation of synaptic AMPA receptor expression (Beattie et al. 2002; Ogoshi et al. 2005; Stellwagen et al. 2005; Stellwagen & Malenka, 2006). As such, blockade of endogenous TNFα in hippocampal cultures decreases surface AMPA receptor expression and AMPA receptor currents, whereas exogenous TNFα application rapidly (within minutes) increases surface AMPA receptor expression and AMPA receptor currents, thereby increasing synaptic strength (Beattie et al. 2002).

Activity-dependent changes in TNFα release may form part of the core cellular response that gives rise to homeostatic adjustments of synaptic strength in response to reduced activity (Stellwagen & Malenka, 2006; Kaneko et al. 2008). Such ‘homeostatic synaptic plasticity’ uses negative feedback mechanisms to adjust synaptic properties and/or cellular excitability up or down to maintain target neuronal activity levels during prolonged or excessive excitation or inhibition (Turrigiano, 2008, 2011). For example, glia release TNFα in response to prolonged reductions in hippocampal neural activity, activating neuronal TNF receptors and increasing AMPA receptor surface expression and synaptic transmission (Stellwagen & Malenka, 2006). However, some evidence suggests that TNFα may not function as an instructive signal that actively induces inactivity-induced synaptic plasticity, but rather enables synaptic plasticity by controlling the expression of several key scaffolding molecules (Steinmetz & Turrigiano, 2010). Although currently available evidence suggests that glia increase TNFα release only after prolonged reductions in neuronal activity (>24 h), we hypothesize that a shorter time constant for TNFα release and subsequent upregulation of excitatory neurotransmission with reduced neuronal activity may be characteristic of the respiratory control system, where reduced neural activity threatens life within minutes.

The phrenic burst amplitude response to exogenous TNFα in dose–response experiments was complex: enhanced phrenic burst amplitude from baseline was observed with 25 and 50 ng TNFα, but lower or higher doses did not change phrenic burst amplitude. These data are consistent with other reports finding that the effect of TNFα on cellular or synaptic properties exhibits a bell-shaped curve (Hua et al. 1996; Sorkin & Doom, 2000); for example, low to moderate doses of TNFα are more effective at increasing spinal nociceptive thresholds (Bianchi et al. 1992), AMPA-induced Ca2+ currents (De et al. 2003), NF-kB activation (Kaltschmidt et al. 1999) and CREB-binding protein expression (Saha et al. 2009), whereas high TNFα doses have no effect (Bianchi et al. 1992; Kaltschmidt et al. 1999; De et al. 2003). The mechanisms that give rise to the narrow window of concentrations in which we and others have observed TNFα-induced effects are unknown.

Although the signal transduction pathway whereby TNFα increases phrenic motor output is not yet fully understood, our data demonstrate a prominent role for aPKCs. TNFα activates several pathways, including phosphoinositide 3-kinase (PI3K; Pastorino et al. 1999; Stellwagen et al. 2005; Yin et al. 2012) and aPKC (Sanz et al. 1999; Liang et al. 2007). A key role for PI3K activity may be indicated, as TNFα-induced increases in surface AMPA receptor expression in hippocampal neurons are blocked following application of a PI3K inhibitor (Stellwagen et al. 2005; Yin et al. 2012). Here, we tested the hypothesis that spinal aPKC activity is necessary for TNFα-induced pMF as spinal aPKC is necessary for iPMF (Strey et al. 2012). Atypical PKC isoforms include PKCζ and PKCι/λ, and a constitutively active form of PKCζ that lacks a regulatory subunit known as PKMζ (Reyland, 2009). iPMF requires PKCζ and/or PKCι/λ (referred to here as PKCζ/ι for clarity) activity to transition from an early, labile form of plasticity to a long-lasting iPMF, an effect that may be mediated through interactions between PKCζ/ι and the scaffolding protein p62/ZIP (Strey et al. 2012). p62/ZIP is a scaffolding/adaptor protein that binds to PKCζ/ι, relocating and anchoring the activated kinase to a context-specific signalling complex (Mochly-Rosen, 1995; Samuels et al. 2001), including signalling complexes associated with TNF receptors (Sanz et al. 1999, 2000). Here, we found that, similar to iPMF, inhibition of spinal aPKC activity during induction of TNFα-induced pMF arrests further development of pMF. Thus, we hypothesize that p62/ZIP is an important intermediary, linking TNF receptors and PKCζ/ι to downstream targets. As PI3K is an important activator of PKCζ/ι (Hirai & Chida, 2003), future studies on the role of PI3K on TNFα-induced pMF and iPMF are warranted.

Methodological considerations

To assess the role of TNFα in iPMF, rats were anaesthetized and mechanically ventilated to enable the investigation of the effects of reduced respiratory neural activity without concomitant hypoxia and hypercapnia that would normally accompany a neural apnoea. In addition, afferent feedback was minimized to prevent entrainment of respiratory neural drive with the ventilator, and to enable the investigation of the effects of reduced central respiratory neural activity on phrenic burst frequency. Afferent feedback from the vagus (Golder & Martinez, 2008) and phrenic (Sandhu et al. 2010) nerves, as well as anaesthesia (Cao & Ling, 2010), have been shown to modulate the expression of plasticity; nevertheless, our findings demonstrate that neural networks underlying breathing sense and respond rapidly to reduced respiratory neural activity, independent of the effects of hypoxia and hypercapnia. The conditions under which these mechanisms manifest as a phenotypic physiological response are of clear clinical interest.

Significance of iPMF

iPMF consists of at least two mechanistically distinct phases: an early, labile phase that requires spinal PKCι/λ activity to transition to a late, stable phase (Strey et al. 2012). The data presented here suggest that TNFα signalling is also necessary for the development of long-lasting iPMF, but may not be completely necessary for early, labile increases in phrenic burst amplitude following a prolonged neural apnoea, as a small, transient increase in phrenic burst amplitude from baseline continued to be observed following the resumption of respiratory neural activity. Our working model of iPMF suggests that reduced neural activity within the phrenic motor pool stimulates local TNFα release, resulting in the formation of an activated PKCζ/ι-ZIP/p62 complex within or near phrenic motor neurons to stabilize a long-lasting increase in phrenic motor output (i.e. iPMF). The cell types releasing TNFα in response to inactivity are not known, but glia are probable candidates (Stellwagen & Malenka, 2006).

The significance of iPMF in the neural control of breathing is not well understood. We hypothesize that iPMF-like mechanisms may compensate for decreases in synaptic inputs to phrenic motor output that may occur during normal and/or pathophysiological conditions. In this sense, iPMF may represent one component in a continuum of ‘homeostatic plasticity’ mechanisms (Turrigiano, 2008, 2011) that enable critical adjustments in phrenic neural activity during conditions that cause excessive inhibition (Hengen et al. 2012) or result in a prolonged change in motor neuron firing properties, such as development (Mantilla & Sieck, 2008). Further, as brief (∼1 min) intermittent neural apnoea is more efficient at eliciting iPMF than a single sustained bout (Baertsch & Baker-Herman, 2013), iPMF-like mechanisms may play a prominent role in conditions that result in repeated fluctuations in respiratory neural activity, such as central sleep apnoea (Strey et al. 2013).

iPMF-like mechanisms may also play a prominent role following pathological conditions that impair descending respiratory drive to phrenic motor neurons. For example, high cervical spinal injuries that damage descending respiratory pathways immediately decrease ipsilateral phrenic activity, followed by a slow, spontaneous partial recovery (Nantwi et al. 1999; Golder et al. 2001, 2003; Goshgarian, 2003, 2009; Golder and Mitchell, 2005; Fuller et al. 2006; Vinit et al. 2007; Lane et al. 2008, 2009). The mechanisms giving rise to this spontaneous recovery are unknown. Of interest, spinal injury is associated with a rapid increase in TNFα (Yune et al. 2003) and aPKC activity (Guenther et al. 2012), and spinal injury-induced increases in TNFα rapidly increase surface AMPA receptor expression in spinal motor neurons (Ferguson et al. 2008). Further studies are necessary to determine whether impaired iPMF diminishes spontaneous recovery of respiratory neural activity following cervical spinal injury.

Acknowledgments

We would like to thank Gordon S. Mitchell and Kristi A. Strey for their insightful comments on the manuscript.

Glossary

aCSF

artificial cerebrospinal fluid

ANOVA

analysis of variance

aPKC

atypical protein kinase C

BSA

bovine serum albumin

iPMF

inactivity-induced phrenic motor facilitation

MAP

mean arterial pressure

PI3K

phosphoinositide 3-kinase

PKC

protein kinase C

PKCζ-PS

ζ-pseudosubstrate inhibitory peptide

pMF

phrenic motor facilitation

SBE

standard base excess

scrPKCζ-PS

scrambled ζ-pseudosubstrate inhibitory peptide

sTNFR1

recombinant human soluble TNF receptor 1

TNFα

tumour necrosis factor-α

Additional information

Competing interests

The authors declare no conflicts of interest.

Author contributions

O.B. collected, analysed and interpreted the data, conceived and designed the experiments, and drafted/revised the manuscript. N.A.B. collected, analysed and interpreted the data, conceived and designed the experiments, and drafted/revised the manuscript. T.L.B-H. conceived and designed the experiments, analysed and interpreted the data, and drafted/revised the manuscript. All authors approved the final version of this manuscript.

Funding

This work was supported by a grant from the National Institutes of Health (HL105511).

References

  1. Baertsch NA, Baker-Herman TL. Inactivity-induced phrenic and hypoglossal motor facilitation are differentially expressed following intermittent versus sustained neural apnea. J Appl Physiol. 2013;114:1388–1395. doi: 10.1152/japplphysiol.00018.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Baker-Herman TL, Fuller DD, Bavis RW, Zabka AG, Golder FJ, Doperalski NJ, Johnson RA, Watters JJ, Mitchell GS. BDNF is necessary and sufficient for spinal respiratory plasticity following intermittent hypoxia. Nat Neurosci. 2004;7:48–55. doi: 10.1038/nn1166. [DOI] [PubMed] [Google Scholar]
  3. Baker-Herman TL, Mitchell GS. Phrenic long-term facilitation requires spinal serotonin receptor activation and protein synthesis. J Neurosci. 2002;22:6239–6246. doi: 10.1523/JNEUROSCI.22-14-06239.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Baker-Herman TL, Mitchell GS. Determinants of frequency long-term facilitation following acute intermittent hypoxia in vagotomized rats. Respir Physiol Neurobiol. 2008;162:8–17. doi: 10.1016/j.resp.2008.03.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Baker-Herman TL, Strey KA. Similarities and differences in mechanisms of phrenic and hypoglossal motor facilitation. Respir Physiol Neurobiol. 2011;179:48–56. doi: 10.1016/j.resp.2011.06.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Beattie EC, Stellwagen D, Morishita W, Bresnahan JC, Ha BK, Von Zastrow M, Beattie MS, Malenka RC. Control of synaptic strength by glial TNFalpha. Science. 2002;295:2282–2285. doi: 10.1126/science.1067859. [DOI] [PubMed] [Google Scholar]
  7. Bianchi M, Sacerdote P, Ricciardi-Castagnoli P, Mantegazza P, Panerai AE. Central effects of tumor necrosis factor alpha and interleukin-1 alpha on nociceptive thresholds and spontaneous locomotor activity. Neurosci Lett. 1992;148:76–80. doi: 10.1016/0304-3940(92)90808-k. [DOI] [PubMed] [Google Scholar]
  8. Blitz DM, Ramirez JM. Long-term modulation of respiratory network activity following anoxia in vitro. J Neurophysiol. 2002;87:2964–2971. doi: 10.1152/jn.2002.87.6.2964. [DOI] [PubMed] [Google Scholar]
  9. Bocchiaro CM, Feldman JL. Synaptic activity-independent persistent plasticity in endogenously active mammalian motoneurons. Proc Natl Acad Sci U S A. 2004;101:4292–4295. doi: 10.1073/pnas.0305712101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cao Y, Ling L. Urethane inhibits genioglossal long-term facilitation in un-paralyzed anaesthetized rats. Neurosci Lett. 2010;477:124–128. doi: 10.1016/j.neulet.2010.04.047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Castro-Moure F, Goshgarian HG. Reversible cervical hemispinalization of the rat spinal cord by a cooling device. Exp Neurol. 1996;141:102–112. doi: 10.1006/exnr.1996.0143. [DOI] [PubMed] [Google Scholar]
  12. Castro-Moure F, Goshgarian HG. Morphological plasticity induced in the phrenic nucleus following cervical cold block of descending respiratory drive. Exp Neurol. 1997;147:299–310. doi: 10.1006/exnr.1997.6615. [DOI] [PubMed] [Google Scholar]
  13. Dale EA, Mitchell GS. Spinal vascular endothelial growth factor (VEGF) and erythropoietin (EPO) induced phrenic motor facilitation after repetitive acute intermittent hypoxia. Respir Physiol Neurobiol. 2013;185:481–488. doi: 10.1016/j.resp.2012.10.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dale-Nagle EA, Satriotomo I, Mitchell GS. Spinal vascular endothelial growth factor induces phrenic motor facilitation via extracellular signal-regulated kinase and Akt signalling. J Neurosci. 2011;31:7682–7690. doi: 10.1523/JNEUROSCI.0239-11.2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dale EA, Satriotomo I, Mitchell GS. Cervical spinal erythropoietin induces phrenic motor facilitation via extracellular signal-regulated protein kinase and Akt signalling. J Neurosci. 2012;32:5973–5983. doi: 10.1523/JNEUROSCI.3873-11.2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. De A, Krueger JM, Simasko SM. Tumor necrosis factor alpha increases cytosolic calcium responses to AMPA and KCl in primary cultures of rat hippocampal neurons. Brain Res. 2003;981:133–142. doi: 10.1016/s0006-8993(03)02997-4. [DOI] [PubMed] [Google Scholar]
  17. Ferguson AR, Christensen RN, Gensel JC, Miller BA, Sun F, Beattie EC, Bresnahan JC, Beattie MS. Cell death after spinal cord injury is exacerbated by rapid TNF alpha-induced trafficking of GluR2-lacking AMPARs to the plasma membrane. J Neurosci. 2008;28:11391–11400. doi: 10.1523/JNEUROSCI.3708-08.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Fuller DD, Golder FJ, Olson EB, Jr, Mitchell GS. Recovery of phrenic activity and ventilation after cervical spinal hemisection in rats. J Appl Physiol. 2006;100:800–806. doi: 10.1152/japplphysiol.00960.2005. [DOI] [PubMed] [Google Scholar]
  19. Golder FJ, Fuller DD, Davenport PW, Johnson RD, Reier PJ, Bolser DC. Respiratory motor recovery after unilateral spinal cord injury: eliminating crossed phrenic activity decreases tidal volume and increases contralateral respiratory motor output. J Neurosci. 2003;23:2494–2501. doi: 10.1523/JNEUROSCI.23-06-02494.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Golder FJ, Martinez SD. Bilateral vagotomy differentially alters the magnitude of hypoglossal and phrenic long-term facilitation in anaesthetized mechanically ventilated rats. Neurosci Lett. 2008;442:213–218. doi: 10.1016/j.neulet.2008.07.034. [DOI] [PubMed] [Google Scholar]
  21. Golder FJ, Mitchell GS. Spinal synaptic enhancement with acute intermittent hypoxia improves respiratory function after chronic cervical spinal cord injury. J Neurosci. 2005;25:2925–2932. doi: 10.1523/JNEUROSCI.0148-05.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Golder FJ, Reier PJ, Davenport PW, Bolser DC. Cervical spinal cord injury alters the pattern of breathing in anaesthetized rats. J Appl Physiol. 2001;91:2451–2458. doi: 10.1152/jappl.2001.91.6.2451. [DOI] [PubMed] [Google Scholar]
  23. Goshgarian HG( The crossed phrenic phenomenon: a model for plasticity in the respiratory pathways following spinal cord injury. J Appl Physiol. 2003;94:795–810. doi: 10.1152/japplphysiol.00847.2002. [DOI] [PubMed] [Google Scholar]
  24. Goshgarian HG. The crossed phrenic phenomenon and recovery of function following spinal cord injury. Respir Physiol Neurobiol. 2009;169:85–93. doi: 10.1016/j.resp.2009.06.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Gray PW, Barrett K, Chantry D, Turner M, Feldmann M. Cloning of human tumour necrosis factor (TNF) receptor cDNA and expression of recombinant soluble TNF-binding protein. Proc Natl Acad Sci U S A. 1990;87:7380–7384. doi: 10.1073/pnas.87.19.7380. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Guenther CH, Windelborn JA, Tubon TC, Jr, Yin JC, Mitchell GS. Increased atypical PKC expression and activity in the phrenic motor nucleus following cervical spinal injury. Exp Neurol. 2012;234:513–520. doi: 10.1016/j.expneurol.2012.01.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Han P, Whelan PJ. Tumour necrosis factor alpha enhances glutamatergic transmission onto spinal motoneurons. J Neurotrauma. 2010;27:287–292. doi: 10.1089/neu.2009.1016. [DOI] [PubMed] [Google Scholar]
  28. Hengen KB, Nelson NF, Stang KM, Johnson SM, Crader SM, Watters JJ, Mitchell GS, Behan M. Increased GABA(A) receptor ɛ-subunit expression on ventral respiratory column neurons protects breathing during pregnancy. PLoS. 2012;7:e30608. doi: 10.1371/journal.pone.0030608. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Hirai T, Chida K. Protein kinase Czeta (PKCζeta): activation mechanisms and cellular functions. J Biochem. 2003;133:1–7. doi: 10.1093/jb/mvg017. [DOI] [PubMed] [Google Scholar]
  30. Hoffman MS, Nichols NL, Macfarlane PM, Mitchell GS. Phrenic long-term facilitation after acute intermittent hypoxia requires spinal ERK activation but not TrkB synthesis. J Appl Physiol. 2012;113:1184–1193. doi: 10.1152/japplphysiol.00098.2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. House C, Kemp BE. Protein kinase C contains a pseudosubstrate prototope in its regulatory domain. Science. 1987;238:1726–1728. doi: 10.1126/science.3686012. [DOI] [PubMed] [Google Scholar]
  32. Hua XY, Chen P, Fox A, Myers RR. Involvement of cytokines in lipopolysaccharide-induced facilitation of CGRP release from capsaicin-sensitive nerves in the trachea: studies with interleukin-1beta and tumor necrosis factor-alpha. J Neurosci. 1996;16:4742–4748. doi: 10.1523/JNEUROSCI.16-15-04742.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Kaltschmidt B, Uherek M, Wellmann H, Volk B, Kaltschmidt C. Inhibition of NF-kappaB potentiates amyloid beta-mediated neuronal apoptosis. Proc Natl Acad Sci U S A. 1999;96:9409–9414. doi: 10.1073/pnas.96.16.9409. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Kaneko M, Stellwagen D, Malenka RC, Stryker MP. Tumour necrosis factor-alpha mediates one component of competitive, experience-dependent plasticity in developing visual cortex. Neuron. 2008;58:673–678. doi: 10.1016/j.neuron.2008.04.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Laudanna C, Mochly-Rosen D, Liron T, Constantin G, Butcher EC. Evidence of zeta protein kinase C involvement in polymorphonuclear neutrophil integrin-dependent adhesion and chemotaxis. J Biol Chem. 1998;273:30306–30315. doi: 10.1074/jbc.273.46.30306. [DOI] [PubMed] [Google Scholar]
  36. Liang H, Baudouin C, Behar-Cohen F, Crisanti P, Omri B. Protein kinase C-zeta mediates retinal degeneration in response to TNF. J Neuroimmunol. 2007;183:104–110. doi: 10.1016/j.jneuroim.2006.11.028. [DOI] [PubMed] [Google Scholar]
  37. Lane MA, Fuller DD, White TE, Reier PJ. Respiratory neuroplasticity and cervical spinal cord injury: translational perspectives. Trends Neurosci. 2008;31:538–547. doi: 10.1016/j.tins.2008.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Lane MA, Lee KZ, Fuller DD, Reier PJ. Spinal circuitry and respiratory recovery following spinal cord injury. Respir Physiol Neurobiol. 2009;169:123–132. doi: 10.1016/j.resp.2009.08.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. MacFarlane PM, Mitchell GS. Episodic spinal serotonin receptor activation elicits long-lasting phrenic motor facilitation by an NADPH oxidase-dependent mechanism. J Physiol. 2009;587:5469–5481. doi: 10.1113/jphysiol.2009.176982. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Mahamed S, Strey KA, Mitchell GS, Baker-Herman TL. Reduced respiratory neural activity elicits phrenic motor facilitation. Respir Physiol Neurobiol. 2011;175:303–309. doi: 10.1016/j.resp.2010.12.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Mantilla CB, Sieck GC. Key aspects of phrenic motoneuron and diaphragm muscle development during the perinatal period. J Appl Physiol. 2008;104:1818–1827. doi: 10.1152/japplphysiol.01192.2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Mochly-Rosen D. Localization of protein kinases by anchoring proteins: a theme in signal transduction. Science. 1995;268:247–251. doi: 10.1126/science.7716516. [DOI] [PubMed] [Google Scholar]
  43. Nantwi KD, El-Bohy AA, Reier PJ, Goshgarian HG. Spontaneous functional recovery in a paralyzed hemidiaphragm following upper cervical spinal cord injury in adult rats. Neurorehabil Neural Repair. 1999;13:225–234. [Google Scholar]
  44. Nichols NL, Dale EA, Mitchell GS. Severe acute intermittent hypoxia elicits phrenic long-term facilitation by a novel adenosine-dependent mechanism. J Appl Physiol. 2012;112:1678–1688. doi: 10.1152/japplphysiol.00060.2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Ogoshi F, Yin HZ, Kuppumbatti Y, Song B, Amindari S, Weiss JH. Tumour necrosis-factor-alpha (TNF-alpha) induces rapid insertion of Ca2+-permeable alpha-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA)/kainate (Ca-A/K) channels in a subset of hippocampal pyramidal neurons. Exp Neurol. 2005;193:384–393. doi: 10.1016/j.expneurol.2004.12.026. [DOI] [PubMed] [Google Scholar]
  46. Park KM, Bowers WJ. Tumour necrosis factor-alpha mediated signalling in neuronal homeostasis and dysfunction. Cell Signal. 2010;22:977–983. doi: 10.1016/j.cellsig.2010.01.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Pastorino JG, Tafani M, Farber JL. Tumour necrosis factor induces phosphorylation and translocation of BAD through a phosphatidylinositide-3-OH kinase-dependent pathway. J Biol Chem. 1999;274:19411–19416. doi: 10.1074/jbc.274.27.19411. [DOI] [PubMed] [Google Scholar]
  48. Powell FL, Milsom WK, Mitchell GS. Time domains of the hypoxic ventilatory response. Respir Physiol. 1998;112:123–134. doi: 10.1016/s0034-5687(98)00026-7. [DOI] [PubMed] [Google Scholar]
  49. Reyland ME. Protein kinase C isoforms: multi-functional regulators of cell life and death. Front Biosci. 2009;14:2386–2399. doi: 10.2741/3385. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Saha RN, Ghosh A, Palencia CA, Fung YK, Dudek SM, Pahan K. TNF-alpha preconditioning protects neurons via neuron-specific up-regulation of CREB-binding protein. J Immunol. 2009;183:2068–2078. doi: 10.4049/jimmunol.0801892. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Samuels IS, Seibenhener ML, Neidigh KB, Wooten MW. Nerve growth factor stimulates the interaction of ZIP/p62 with atypical protein kinase C and target endosomal localization: evidence for regulation of nerve growth factor-induced differentiation. J Cell Biochem. 2001;82:452–466. doi: 10.1002/jcb.1177. [DOI] [PubMed] [Google Scholar]
  52. Sandhu MS, Lee KZ, Fregosi RF, Fuller DD. Phrenicotomy alters phrenic long-term facilitation following intermittent hypoxia in anaesthetized rats. J Appl Physiol. 2010;109:279–287. doi: 10.1152/japplphysiol.01422.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Santello M, Volterra A. TNFα in synaptic function: switching gears. Trends Neurosci. 2012;35:638–647. doi: 10.1016/j.tins.2012.06.001. [DOI] [PubMed] [Google Scholar]
  54. Sanz L, Diaz-Meco MT, Nakano H, Moscat J. The atypical PKC-interacting protein p62 channels NF-kappaB activation by the IL-1-TRAF6 pathway. EMBO J. 2000;19:1576–1586. doi: 10.1093/emboj/19.7.1576. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Sanz L, Sanchez P, Lallena MJ, Diaz-Meco MT, Moscat J. The interaction of p62 with RIP links the atypical PKCs to NF-kappaB activation. EMBO J. 1999;18:3044–3053. doi: 10.1093/emboj/18.11.3044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Selbie LA, Schmitz-Peiffer C, Sheng Y, Biden TJ. Molecular cloning and characterization of PKC iota, an atypical isoform of protein kinase C derived from insulin-secreting cells. J Biol Chem. 1993;268:24296–24302. [PubMed] [Google Scholar]
  57. Sorkin LS, Doom CM. Epineurial application of TNF elicits an acute mechanical hyperalgesia in the awake rat. J Periph Nerv Syst. 2000;5:96–100. doi: 10.1046/j.1529-8027.2000.00012.x. [DOI] [PubMed] [Google Scholar]
  58. Steinmetz CC, Turrigiano GG. Tumour necrosis factor-α signalling maintains the ability of cortical synapses to express synaptic scaling. J Neurosci. 2010;30:14685–14690. doi: 10.1523/JNEUROSCI.2210-10.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Stellwagen D, Beattie EC, Seo JY, Malenka RC. Differential regulation of AMPA receptor and GABA receptor trafficking by tumour necrosis factor-alpha. J Neurosci. 2005;25:3219–3228. doi: 10.1523/JNEUROSCI.4486-04.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Stellwagen D, Malenka RC. Synaptic scaling mediated by glial TNF-alpha. Nature. 2006;440:1054–1059. doi: 10.1038/nature04671. [DOI] [PubMed] [Google Scholar]
  61. Strey KA, Baertsch NA, Baker-Herman TL. Inactivity-induced respiratory plasticity: protecting the drive to breathe in disorders that reduce respiratory neural activity. Respir Physiol Neurobiol. 2013 doi: 10.1016/j.resp.2013.06.023. doi: pii: S1569-9048(13)00229-2. doi: 10.1016/j.resp.2013.06.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Strey KA, Baker-Herman TL. Local reductions in synaptic inputs to the phrenic motor pool elicits atypical Protein Kinase C (aPKC) and Tumor Necrosis Factor alpha (TNFa) dependent inactivity-induced phrenic motor facilitation. FASEB J. 2012;26:704.12. [Google Scholar]
  63. Strey KA, Nichols NL, Baertsch NA, Broytman O, Baker-Herman TL. Spinal atypical protein kinase C activity is necessary to stabilize inactivity-induced phrenic motor facilitation. J Neurosci. 2012;32:16510–16520. doi: 10.1523/JNEUROSCI.2631-12.2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Tadjalli A, Duffin J, Peever J. Identification of a novel form of noradrenergic-dependent respiratory motor plasticity triggered by vagal feedback. J Neurosci. 2010;30:16886–16895. doi: 10.1523/JNEUROSCI.3394-10.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Turrigiano GG. The self-tuning neuron: synaptic scaling of excitatory synapses. Cell. 2008;135:422–435. doi: 10.1016/j.cell.2008.10.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Turrigiano G. Too many cooks? Intrinsic and synaptic homeostatic mechanisms in cortical circuit refinement. Annu Rev Neurosci. 2011;34:89–103. doi: 10.1146/annurev-neuro-060909-153238. [DOI] [PubMed] [Google Scholar]
  67. Vinit S, Stamegna JC, Boulenguez P, Gauthier P, Kastner A. Restorative respiratory pathways after partial cervical spinal cord injury: role of ipsilateral phrenic afferents. Eur J Neurosci. 2007;25:3551–3560. doi: 10.1111/j.1460-9568.2007.05619.x. [DOI] [PubMed] [Google Scholar]
  68. Walker JK, Jennings DB. Respiratory effects of pressor and depressor agents in conscious rats. Can J Physiol Pharmacol. 1998;76:707–714. doi: 10.1139/cjpp-76-7-8-707. [DOI] [PubMed] [Google Scholar]
  69. Wilkerson JE, Satriotomo I, Baker-Herman TL, Watters JJ, Mitchell GS. Okadaic acid-sensitive protein phosphatases constrain phrenic long-term facilitation after sustained hypoxia. J Neurosci. 2008;28:2949–2958. doi: 10.1523/JNEUROSCI.5539-07.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Yao Y, Shao C, Jothianandan D, Tcherepanov A, Shouval H, Sacktor TC. Matching biochemical and functional efficacies confirm ZIP as a potent competitive inhibitor of PKMζ in neurons. Neuropharmacology. 2013;64:37–44. doi: 10.1016/j.neuropharm.2012.07.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Yin HZ, Hsu CI, Yu S, Rao SD, Sorkin LS, Weiss JH. TNF-α triggers rapid membrane insertion of Ca(2+) permeable AMPA receptors into adult motor neurons and enhances their susceptibility to slow excitotoxic injury. Exp Neurol. 2012;238:93–102. doi: 10.1016/j.expneurol.2012.08.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Yune TY, Chang MJ, Kim SJ, Lee YB, Shin SW, Rhim H, Kim YC, Shin ML, Oh YJ, Han CT, Markelonis GJ, Oh TH. Increased production of tumour necrosis factor-alpha induces apoptosis after traumatic spinal cord injury in rats. J Neurotrauma. 2003;20:207–219. doi: 10.1089/08977150360547116. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES