JOURNALOFNEUROPHYSIOLOGY Vol. 68. No. 5, November 1992. Printed

in U.S.A.

Mechanisms Underlying Excitatory Effects of ThyrotropinReleasing Hormone on Rat Hypoglossal Motoneurons In Vitro DOUGLAS Department

A. BAYLISS, FkLIX VIANA, AND ALBERT J. BERGER of Physiology and Biophysics, University of Washington School of Medicine,

SUMMARY

AND

CONCLUSIONS

I. The hypoglossal motor nucleus contains binding sites for the neuropeptide thyrotropin-releasing hormone (TRH) and is innervated by TRfi-containing fibers. Although excitatory effects of TRH on hypoglossal motoneurons (HMs) have been described, the ionic mechanisms by which TRH exerts such effects have not been fully elucidated. Therefore, we investigated the effects of TRH on HMs in transverse slices of rat brainstem with intracellular recording techniques. 2. TRH was applied by perfusion (0.1-10 PM ) or by pressure ejection ( 1.O PM), while HMs were recorded in current or voltage clamp. In all cells tested, TRH caused a depolarization and/or the development of an inward current. These effects were fully reversible, dose dependent, and showed only modest desensitization with long applications. In addition, although TRH increased synaptic activity in many cells, the depolarizing response to TRH was maintained in tetrodotoxin (OS- 1.OpM)-containing or in a nominally Ca2+-free perfusate containing 2 mM Mn 2+. Thus TRH acts directly on HMs to cause the depolarization. 3. Hyperpolarizing current (or voltage) steps superimposed on the TRH-induced depolarization (or inward current) revealed a decreased input conductance. Extrapolated instantaneous current-voltage relationships obtained before and at the peak of the response to TRH intersected (i.e., reversed) at - 10 1 mV, negative to the expectedK+ equilibrium potential (&). When extracellular [K+] wasraisedfrom 3 to 12mM, the reversalpotential wasshiftedin the depolarizingdirection and the magnitudeof the TRH-induced depolarization was diminished. Moreover, the TRH responsewasenhancedin size from depolarizedpotentials (i.e., further from &). Taken together,theseresultsindicate that TRH depolarizesHMs, in part, by decreasinga resting K+ conductance. 4. Similar to TRH, bath-application of 2 mM Ba2+causeda depolarizationassociatedwith decreasedconductance,suggesting that Ba2+alsoblocksa restingK+ conductance.The Ba2+-sensitive and TRH-sensitive resting K+ conductancesare apparently identical; in the presenceof Ba2+, the customary TRH-induced decreasein conductancewasoccluded. 5; It is noteworthy that the TRH-induced inward current (I.&, although diminished, wasnot entirely blocked by Ba2+. This secondBa2’-insensitivecomponent of ImH was not associated with a measurablechange in input conductance. It was especiallyevident during current-clamp recordings,when the diminutive TRH-induced current wasstill capableof causinga sub stantial depolarization. The ionic basisof the residualTRH-inducedinward current remainsto be determined. 6. Weinvestigatedthe functional consequences of thesemechanismsof action of TRH on spike firing behavior of HMs. TRH had little effecton the shapeof the action potential and afterpotentialsbut causeda parallelleftward shift in the relationshipbetween firing frequencyand injected depolarizing current. 7. In conclusion,TRH depolarizesadult HMs by at leasttwo mechanisms.TRH decreases a Ba2+-sensitiverestingK+ conduc-

Seattle, Washington

98195

tance and in addition it inducesan inward current that is Ba2+ resistant.Theseactionsof TRH enhanceHM excitability not only by directly depolarizingthe cell but alsoby decreasingmembrane conductance and thereby lowering the threshold for repetitive firing. INTRODUCTION

Thyrotropin-releasing hormone (TRH) was the first hypophysiotropic hormone to be purified from hypothalamic extracts and characterized (reviewed in Reichlin 1989). Shortly thereafter, upon the discovery of a widespread extrahypothalamic distribution of TRH in the rat central nervous system (Brownstein et al. 1974; Winokur and Utiger 1974)) it was suggested that TRH may have central neural functions distinct from those neuroendocrine effects that prompted its original isolation. Since that time, the TRH neuronal system of rat brain has been extensively characterized (Hokfelt et al. 1975a,b; Lechan et al. 1986; Segerson et al. 1987; Sharif 1989; Tsuruo et al. 1987). Principal among the many targets of the TRH neuronal system appear to be spinal cord and cranial motor nuclei (e.g., the hypoglossal motor nuclei). Thus these nuclei are endowed with a substantial population of TRH-immunoreactive (IR) fibers (Hijkfelt et al. 1975b) and contain numerous TRH binding sites (Manakar and Rizio 1989; Sharif 1989). Importantly, TRH is believed to modulate the activity of motoneurons in these nuclei; the predominant finding has been an enhancement of excitability by TRH (Lacey et al. 1989; Nicoll 1977; Nistri et al. 1990; Rekling 1990; Takahashi 1985; Wang and Dun 1990; White 1985). Indeed this well-documented effect of TRH has led to its therapeutic use in the treatment of muscle weakness associated with amyotrophic lateral sclerosis (motor neuron disease) (reviewed in Brooks 1989). An understanding of the basic ionic mechanisms that mediate the excitatory effects of TRH on motoneurons is, therefore, of fundamental importance. Relatively few studies documenting the excitatory effects of TRH on motoneurons have attempted to determine the ionic mechanisms underlying those effects. In spinal motoneurons of amphibians, TRH causes a depolarization that is associated with either no change or an increase in input conductance ( Lacey et al. 1989; Nicoll 1977). By contrast, in mammalian motoneurons the TRH-induced depolarization is associated with a decrease in input conductance, suggesting involvement of a K+ conductance (Nistri et al. 1990; Rekling 1990; Takahashi 1985; Wang and Dun 1990). This has been directly confirmed only in spinal motoneurons from neonatal rats; in those cells the depolariza-

0022.3077/92 $2.00 Copyright 0 1992 The American Physiological Society Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (018.218.056.169) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

1733

1734

D. A. BAYLISS,

F. VIANA,

tion induced by TRH was diminished in extracellular solutions with raised K+ concentrations, indicating that TRH inhibited a K+ conductance (Takahashi 1985). Subsequent voltage-clamp experiments in neonatal spinal neurons demonstrated that the K+ conductance blocked by TRH was Ba2+ sensitive (Nistri et al. 1990). There have been no comparable studies of the K+ dependence or the Ba2+ sensitivity of TRH effects in adult motoneurons. This is potentially relevant given that intrinsic neuronal properties and the mechanisms by which neurotransmitters modulate those properties may change during the early postnatal period (Fukuda et al. 1989; Fulton and Walton 1986; Haddad et al. 1990). We therefore undertook current- and voltage-clamp experiments to determine mechanisms that mediate excitatory effects of TRH in motoneurons of young adult and adult animals. We chose to study hypoglossal motoneurons (HMs) because an excitatory effect of TRH on these cells had been previously described, but the underlying ionic mechanisms had not been elucidated (Rekling 1990). Furthermore, the modulation by TRH of HM repetitive firing behavior may be of particular importance in the pathophysiology of obstructive sleep apneas. In short, our results indicate that multiple mechanisms contribute to the TRH-induced enhancement of excitability in adult HMs. We found that TRH depolarizes HMs, in part, by inhibiting a Ba2+sensitive, resting K+ conductance. In addition, we discovered a component of the inward current caused by TRH that was Ba2+ insensitive. This additional current was reminiscent of that induced by TRH in amphibian spinal motoneurons inasmuch as it was not associated with a change in conductance. Finally, we correlated the mechanism of action of TRH with its effects on HM repetitive firing behavior. A brief account of some of these results has been presented (Bayliss et al. 199 1).

A. J. BERGER

walledglasscapillary tubescontaining microfilaments( WPI) with a Brown-Flaminghorizontal puller (Sutter). Electricalrecordings were made with an Axoclamp 2A (Axon), operating in either bridge, discontinuouscurrent-clamp (DCC), or single-electrode voltage-clamp(SEVC) mode.The headstageoutput wascontinuously monitored on a separateoscilloscope,and publishedprocedureswere followed to set capacitancecompensationand maximize samplingfrequency (typically 7- 12 kHz) (Finkel and Redman 1985). To reducestray capacitanceto ground and improve samplingrate, a cylindrical steelshaft wasplacedaround the recording electrodeand connectedto the casesocketof the headstage.The extent of “space-clamp” achieved in thesecells with SEVC is uncertain. However, given that membranecurrent wasa smoothand continuous function of membranepotential over the voltage rangestested,it is likely that seriouserrors due to inadequate space-clampwereavoided. Intracellular recordingswere made from neurons within the clearly visibleboundariesof the hypoglossalnuclei. Previouswork from this laboratory with a multiple-labelingtechnique(Viana et al. 1990) and the work of others(Haddad et al. 1990), hasestablishedthat the vast majority (>90%) of neuronswithin the hypoglossalnuclei are indeedmotoneurons.In addition, only neurons with electrophysiologicalproperties characteristic of HMs were studied(Mosfeldt Laursenand Rekling 1989;Viana et al. 1990). We thereforerefer to the cellsstudiedasmotoneurons.Cellswere studiedonly if they had a stablerestingmembranepotential of at least-60 mV and an overshootingaction potential. Membrane current and voltage were monitored during the experiment on a storageoscilloscope(Tektronix) and on a chart recorder (Gould) and were stored for off-line analysison a fourchannel FM tape recorder (Racal) with cutoff frequenciesof either 2.5 or 5 kHz. Selectedsamplesfrom tape recordsweredigitized with a digital oscilloscope(Gould) and transferredto disk storageon a microcomputer(IBM) with softwaredevelopedin the laboratory. Measurementswere madedirectly on the digital oscilloscopewith the aid of on-screencursors.Statisticalanalysisof the effectsof TRH weremadewith Student’st test,with a significance level setat P < 0.05.

Solutions and reagents

METHODS

Preparation of slices Experimentswere performed with transversebrain stem slices

prepared from 28 Sprague-Dawley rats of either sex (32-240 g; mean = 114.1g). Rats were anesthetized (ketamine, 200 mg/ kg im and xylazine, 14 mg/ kg im), tracheotomized, and ventilated with carbogen(95% O#% COZ). The animal wasrapidly decapi-

tated, and under a steady stream of ice-cold, sucrose-containing solution (see below) the brain stem was exposed,excised,and placedin a dish containing ice-cold sucrosesolution. The tissue waspreparedfor slicing by attaching the rostra1end to a Teflon support with cyanoacrylate glue. The ventral surfacewasstabilized againstan agarblock. With the tissueimmersedin ice-cold sucrosesolution, three or four transversesections(400 pm) at about the level of the obex werecut (Vibroslice, CampdenInstruments). The sliceswere then transferred to a gas-interfacetype chamber,perfusedfirst with a sucrosesolution (20-30 min) and then with a Ringer solution (seebelow), both oxygenatedat 33 & 1°C (mean * SE).

Electrophysiological

AND

recordings

Intracellular recordingswere madewith microelectrodesfilled with 3 M KC1 ( 13-40 MQ) prepared from 1.5mm-OD thin-

Most experimentswere performedwhile the sliceswerebathed in a Ringer solution containing (in mM) 130 NaCl, 3 KCl, 2 MgQ, 2 CaCI,, 1.25 NaH,PO,, 26 NaHCO,, and 10 glucose. The sucrosesolution usedfor the preparationand initial incubation of the sliceswas preparedas described(Ringer solution as above, but with 130 mM NaCl replacedby 260 mM sucrose) (Aghajanian and Rasmussen1989). Nominally Ca2+-freesolutions containing 2 mM Mn2+ or 2 mM Ba2+ were prepared by equimolar substitution of MnCl, or BaCl, for CaCl,; NaH2P0, was omitted from thesesolutionsto avoid precipitation. Ringer solution with elevatedK+ ( 12 mM) waspreparedby equimolar substitution of KC1 for NaCl. Tetrodotoxin (‘ITX), obtained from Calbiochem,waspreparedasa 200.PM stock solution in a modified Ringer solution (without divalent cations or glucose), aliquoted, and frozen. Appropriate volumes were added to the perfusing media to give a final concentration of 0.5-I .OPM. In somecases,a microdroplet of TTX (200 PM) wasalsoadministered directly on the sliceto expediteits effects.TRH (Peninsula) waspreparedasa 10-mM stocksolution in either distilled water or in the modifiedRinger solution (cf. TTX). TRH waseither added to the perfusate(0. l- 10 PM, final concentration) or applied by pressureejectionwith a Picospritzer(General Valve). When pressureejected,the TRH concentration in the pipette was 1.0 PM. The pipette wasplacedinto the bathing solutionat the edgeof the hypoglossalnucleus,ipsilateralto and upstreamof the recording electrode.

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (018.218.056.169) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

TRH EFFECTS ON HYPOGLOSSAL

MOTONEURONS

1735

put resistance. It should be noted, however, that because R, in these cells is voltage-dependent (Mosfeldt Laursen and General Rekling 1989) the apparent increase in R, is somewhat exaggerated. The effect of TRH was tested in a total of 7 1 HMs from A voltage-clamp recording from a different HM is shown 28 animals. The cells had an average resting membrane potential of -69.6 t 5.2 mV (n = 7 1) and input resistance in Fig. 1 B. This cell was voltage clamped at -59 mV, and constant hyperpolarizing voltage steps ( 10 mV) were ap(R,) of 13.5 t 3.6 MQ (n = 47). The predominant effect of TRH in every cell tested was a plied throughout. TRH caused the reversible development of an inward ( negative) current (Jr&, which reached -2.2 depolarization and/or the development of an inward current. Figure 1A shows that 10 PM TRH caused a depo- nA at its peak. The current response to the hyperpolarizing voltage steps during the control period (Fig. 1B, bl ) and at larization of -11 mV, which brought the HM to firing threshold. Rebound firing (truncated) at the offset of the the peak of the TRH response (Fig. 1B, b2) are also shown. hyperpolarizing current pulses is evident near the peak of After the capacitive transient, the voltage step caused an the depolarizing response (also see lower records). After initial current jump followed by the slow development of reaching the peak of the response, the cell slowly recovered an inward current. The magnitude of the initial current toward its original membrane potential. The voltage re- jump was markedly reduced by TRH, indicating that TRH sponses to constant amplitude hyperpolarizing current caused a decrease in input conductance. pulses in the control period followed a trajectory typical for The response of a HM to 1.O PM TRH delivered by presadult HMs ( Fig. 1A, a 1) ; an initial peak was followed by a sure ejection from a pipette placed near the hypoglossal depolarizing sag to a steady voltage level, and a depolarizing nucleus is shown in Fig. 2. The chart record (top) reveals overshoot occurred after the pulse. The same current pulse that shortly after TRH was applied the motoneuron depolarized and reached firing threshold (spikes truncated). at the peak of the TRH-induced depolarization (Fig. IA, a2) caused an electrotonic response generally similar in Full action potential trajectories from the indicated period shape, but larger in amplitude, indicating an increased in- ( 3) are shown at faster time base (bottom right). At the RESULTS

VOLTAGE CLAMP

CURRENT CLAMP TRH (10 pM)

bl

1006 al

a2

bl

b2

I 2,4d

-54 mV *mVT...L...pm

.....

,

,............ b

,.... w ,

0.2 nA

L-J10 mV loo fns

+-YJ

2rlA 2oom

FIG. 1. Thyrotropin-releasing hormone (TRH) causes depolarization and/or development of an inward current in hypoglossal motoneurons. Effect of TRH ( 10 PM), applied by bath-perfusion (for periods indicated by bars), to two different hypoglossal motoneurons ( HMs) held in current clamp (A ) and voltage clamp ( B) is shown. Records of membrane voltage (upper, top traces) and current (upper, bottom traces) demonstrate the slow time course of the response to, and recovery from, application of TRH ( hop). Response to constant-amplitude, hyperpolarizing current (A, 1.OnA) and voltage (B, 10 mV) steps are shown on a faster time base (bottom). A : TRH caused a depolarization to firing threshold in this cell and also caused an increase in synaptic activity (thickening of the trace). Near the peak of the depolarizing response, action potentials (truncated) were also seen on the rebound from the hyperpolarizing pulse. Voltage response to the current pulse was smaller during the control period (al ) than at the peak of the TRH-induced depolarization (a2), indicating that TRH increased the input resistance of the cell. B: from a holding potential of -59 mV, TRH caused the development of an inward current and increased spontaneous synaptic currents. Instantaneous current response to the voltage steps (immediately after the capacitive transient) was larger during the control period (bl ) than at the peak of the TRH-induced inward current, indicating that TRH decreased the input conductance of the cell. Cells were recorded in standard Ringer solution. The celI on the left was at its resting membrane potential (RMP); RMP for the cell on the right was -74 mV.

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (018.218.056.169) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

D. A. BAYLISS, F. VIANA,

1736 TRH (1.0 p.M) ‘I

3

-71 mV

-71 mv

V

100 ms -71 mV -

I

20 mV

-w-m I

50 ms

2. Focal application of TRH by pressure ejection. Depolarizing effect of TRH ( 1.O PM), pressure-ejected from a pipette located near the hypoglossal nucleus, is illustrated. Chart records (top) show that TRH caused a depolarization of the HM and the cell fired repetitively (truncated) at the peak of the response (top, top trace). At the peak of the depolarization, DC current (-0.6 nA) was injected (top, bottom trace) to return the cell to its original membrane potential (-7 1 mV). Voltage response to -OS-nA current pulses (bottom left) was smaller in the control period ( 1) than at the peak of the depolarization (2), indicating that TRH caused an increase in input resistance. After the DC current was removed, the cell again fired repetitively for a period ( 3 ) before recovering. Examples of the full action potential and afterpotential trajectories from that period are shown (3, bottom right). Cell recorded in standard Ringer solution at RMP. FIG.

AND A. J. BERGER

peak of the depolarizing response, DC current was injected (equal to ZTRH) to manually clamp the cell at the original membrane potential for comparison of the electrotonic response to the current pulses. The voltage responses to the current pulses obtained at the peak of the TRH response (2, bottom left) were clearly augmented relative to those from the control period ( 1, bottom lefi), reflecting the increase in R,. The overall characteristics of this response were, therefore, indistinguishable from those obtained by bath-application (cf. Fig. I). The general characteristics of the response to TRH were examined in detail in 36 cells. There was no immediately obvious difference in the response of HMs from ventral or dorsal regions of the nucleus. TRH ( 10 PM by bath) depolarized HMs by 13.4 t 0.7 mV (n = 8) and induced an inward current of 1.2 t 0.2 nA from an average holding potential of -62.7 t 1.7 mV (n = 15). The input conductance decreased to 64.6 t 3.2% of control (n = 15). Similar responses, although smaller in size, were recorded with lower concentrations of TRH in the perfusate (0.1 and 1.O CONTROL

TRH

1OOms

TRH (10 @A)

lr

8 min -

/

0

-67 mV

CONTROL

l

3

-1 i

v TRH

TRH (10 pU)

-5 14

-68 mV

--

I

10 mV

2 min

3. TRH-induced depolarization shows only modest desensitization and tachyphylaxis with long or repeated applications. Effect of longduration exposure to TRH was determined. TRH ( 10 PM) was applied to the slice for the periods indicated ( - 15 min, top trace; -9 min, bottom trace). TRH-induced depolarization reached a peak of 6 mV after -4 min. Response exhibited limited desensitization over the period that the HM was exposed to TRH; after 15 min (top trace) or 9 min (bottom trace), the cell was still 4 mV depolarized from control levels. The slice was returned to normal perfusate (i.e., without TRH), and after a latency roughly equivalent to the time-to-peak ( -4 min), the cell began to recover slightly more rapidly to its original potential. TRH application periods were separated by 36 min. The 2 depolarizing responses (5th and 6th for the cell) were essentially identical in both time course and amplitude. Thus TRH-induced depolarizations are maintained for long periods and are reproducible. HM recorded at RMP in normal Ringer solution containing 1.O PM tetrodotoxin (TTX). Negative deflections in membrane potential are responses to constant amplitude hyperpolarizing current pulses ( 1.O nA). For abbreviations, see Fig. 1 legend. FIG.

6

E rw=-99 mV I

A MEMBRANE

I

-80

POTENTIAL

-60

(mv)

4. TRH-induced inward current reverses near potassium equilib rium potential (&). Reversal potential of TRH-induced inward current ( lTRH) was determined from instantaneous current-voltage relationships obtained in single-electrode voltage clamp (SEVC) before and during exposure to TRH ( 10 PM). The raw records represent approximately the first -300 ms of current responses (top) to 2-s voltage steps (bottom) applied in control conditions (top /e#) and at the peak of the TRH response (top right). TRH caused development of an inward current (i.e., negative shift in holding current) and a decreased conductance (i.e., decreased magnitude of the instantaneous current jumps). Instantaneous current response to each of the series of voltage steps was measured imme diately after the capacitive transient, but preceding the development of the slow inward current (marked by appropriate symbol), and plotted as a function of the voltage measured at the same time point (bottom). Datapoints from control ( l ) and from TRH (A ) were well fitted by linear regression. TRH caused a negative shift and a decreased slope of the I-v relationship. Regression lines intersected at -99 mV, slightly negative to the presumed &. RMP of the HM was -76 mV. The cell was held at -59 mV with 2.9.nA holding current in control; all other current measurements are plotted with reference to this holding current. For abbreviations, see Fig. 1 legend. FIG.

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (018.218.056.169) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

TRH EFFECTS ON HYPOGLOSSAL

MOTONEURONS

CONTROL

B

A

1737

TRH l

0

CONTROL

. v

v

TRH

- 60 mV

12 mM K+ v - 60 mV

WASH , -100

-80

MEMBRANE

-60

- 60 mV

POTENTIAL (mV)

FIG. 5. High extracellular [K’] shifts the reversal potential of I TM and attenuates the TRH-induced depolarization. A : effect of high extracellular [K+) on the reversal potential of ITRH was determined. Instantaneous current-voltage (I- V) relationships were obtained from SEVC records in control ( l ) and at the peak of the response to TRH ( A ), first in normal Ringer (3 mM K+) and then in Ringer with elevated [K+] ( 12 mM K+). Reversal potential ( Eccv)was determined from the intersection of regression lines fitted to instantaneous I- Y relationships (cf. Fig. 4). Emvwas shifted from - 100 mV in 3 mM K+ to -77 mV in 12 mM K+. Resting membrane potential ( RMP) and holding potential were -79 mV (i.e., holding current was 0) in 3 mM K+. Currents in 3 mM-K+ were referenced to this holding current. In 12 mM K+ the cell was held at -65 mV with +0,2 nA; currents in 12 mM K+ were referenced to this holding current. B: effect of high extracellular [K+] on the size of the TRH-induced depolarization was determined. A representative example from a single HM is shown. The cell was held at -60 mV throughout the experiment. TRH ( 1.O PM) was applied (v) from a pipette located near the hypoglossal nucleus by pressure ejection. TRH-induced depolarization was determined when the slice was bathed in normal Ringer solution containing 3 mM K+ (top, CONTROL), then in a Ringer solution containing 12 mM K+ for ~36 min (middle, 12 mM K+), and finally 22 min after washing the slice in normal Ringer solution (bottom, WASH ) . The magnitude of the depolarization was decreased in the high K+ solution and recovered toward control levels after the wash period. RMP for the cell was -76 mV. These data together are consistent with an effect of TRH on a K+ conductance. Cells were recorded in OS-PM tetrodotoxin (TTX). For abbreviations, see Fig. 1 legend.

PM). It is important to note that the actual concentration of TRH in the vicinity of the cell could not be determined, but given our protocol for bath-application (brief bolus), we suggest that it was substantially lower than the concentration in the perfusate. This view is supported by our experiments in which small volumes of a lower concentration of TRH ( 1.O PM) were pressure ejected from a micropipette close to the HM. Under these conditions, the concentration at the cell under study must have been 4.0 PM. Yet, in some cases the magnitude of the response to TRH applied in this manner was as large or larger than that measured with an order of magnitude more TRH in the perfusate (cf. Figs. 1 and 2). Pressure ejection of 1.O PM TRH caused a depolarization of 8.8 t 0.9 mV (n = 13), an inward current of 0.7 t 0.1 nA (n = 17), and a decrease in input conductance to 72.1 t 1.6% of control (n = 17) from -70.5 t 0.9 mV (n = 2 I ). We found there was a slow but complete recovery from the response to TRH (cf. Figs. 1 and 2). The ability of the cell to recover fully appeared to be related to the length of exposure and the concentration of TRH. The response to TRH desensitized only very slowly with continued application of TRH, as demonstrated in Fig. 3. In this experiment (top trace), 10 PM TRH was bath applied for an extended period ( - 15 min); the cell depolarized from -67 to -6 1 mV at the peak and after 15 min in TRH was still depolarized (-63 mV) relative to the control period. On returning to normal perfusate (i.e., without

TRH ) the cell recovered more rapidly to its original potential. Thus in contrast to an earlier report (Rekling 1990), we found that HMs can maintain a response even during extended periods of exposure to TRH. A modest tachyphylaxis was sometimes noticed with multiple applications. The most pronounced tachyphylaxis appeared to occur between the first and second applications, whereas subsequent applications produced remarkably consistent responses. For example, two consecutive depolarizing responses to TRH (the 5th and 6th for the cell) are also shown in Fig. 3. The later response (bottom trace) is essentially identical to the response recorded 36 min earlier (top trace) in both time course and magnitude. Responses to brief application of TRH by pressure ejection recovered more readily and could be reproduced with higher fidelity than those elicited by longer bath applications. TRH eflects on Hills are direct

It is noteworthy that TRH often, but not always, caused an increase in synaptic activity recorded in HMs. This is evident in the chart records of Fig. 1 as a thickening of the baseline, which is especially prominent around the peak of the response. This suggested the possibility that TRH was exerting its effects on HMs solely via an indirect mechanism. This possibility was ruled out by our findings that the TRH effect (i.e., depolarization with decreased input conductance) was maintained in a Ringer solution containing

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (018.218.056.169) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

1738

D. A. BAYLISS, F. VIANA,

AND A. J. BERGER

TRH (1.0 FM) -63 mV

-68 mV

v

-79 mV v

-90

-89 mV

I “‘llllll~‘llllll”lllf’llli~llll~”’l~~~’lil~ll~~~~~~l’~~~ll~~~’l

20 mV

-

-80

-70

-60

-50

MEMBRANEPOTENTIAL(mv)

2 min FIG. 6. TRH-induced depolarization is voltage-dependent. The effect of membrane potential on the size of the TRH-induced depolarization was determined. Left: example from a single HM. Constant DC current injection was used to hold the cell at a number of different potentials; TRH ( 1.0 PM) was applied via pressure ejection from a pipette located near the hypoglossal nucleus (v ). From the most depolarized potential, TRH caused the HM to depolarize beyond the threshold for repetitive firing ( top trace). From progressively more hyperpolarized potentials (lower traces), the size of the TRH-induced depolarization was progressively reduced in amplitude. Negative deflections in the voltage trace represent the voltage responses to constant amplitude current pulses (- 1.O nA). RMP for the HM was -73 mV. Right:scatter plot of data from cells studied at different holding potentials ( y1= 7 ) . Data from individual cells are indicated by a common symbol. Data were well fitted with linear regression (-; t95% confidence interval, - - -). Magnitude of the TRH-induced depolarization was increased from depolarized potentials (P < 0.05 ) . If TRH caused repetitive firing, the magnitude of the depolarization could not be determined and the data point was not included in the scatter plot. Most cells ( 6 / 7 ) , therefore, were recorded in 0.5 PM tetrodotoxin ( TTX ) . For abbreviations, see Fig. 1 legend.

2 mM Mn2+ (substituted equimolar for Ca”‘) to inhibit chemical synaptic transmission (n = 3; not shown). The efficacy of the 2 mM Mn2+/ 0 Ca2+ solution was assured before TRH application by demonstrating a substantial reduction of the Ca2+ -dependent, medium duration spike after-hyperpolarization (m-AHP) . Furthermore, we found that TRH effects were maintained after blockade of action potential-dependent synaptic transmission with TTX (cf. Figs. 3 and 5, n = 34). Thus it is certain that TRH acts directly on HMs to cause depolarization and decreased input conductance; indirect effects of TRH were evident as increased synaptic activity but were not analyzed further.

current. In addition, there was a decreased input conductance, seen as a decrease in the magnitude of the initial current jumps. The currents associated with each voltage step were measured immediately after the capacitive transient, but before activation of the slow inward current (marked by appropriate symbol), and plotted as a function of membrane potential. Compared with the Z-V relationship in the control period (a), that measured in TRH was shifted down and had a reduced slope ( A ) , reflecting both the TRH-induced development of inward current and the decreased input conductance. Both I- v relationships were well fitted by linear regressions, which intersected at -99 mV. The averaged Emv of ZTRH was - 10 1.1 t 1.O mV (n = 8)) which is slightly hyperpolarized to the presumed EK TRH inhibits a resting K+ conductance under our experimental conditions ( - -9 1 mV; Jiang and The fact that TRH-induced depolarization was asso- Haddad 199 1). Involvement of a Cl- conductance in the response to TRH is unlikely because E,, would be shifted to ciated with decreased conductance suggested involvement of a K+ conductance. If this were the case, the reversal po- more depolarized potentials by the high KC1 concentration tential ( Erev) for ZTRHshould approximate the estimated K+ in our electrodes. Thus ZTRHreverses close to EK, suggesting equilibrium potential (&). We tested this possibility by that a K+ conductance is involved in the TRH response. determining the intersection of instantaneous current We next performed more direct tests of the hypothesis (I)-voltage (V) relationships obtained before and at the that a K+ current contributes to ZTRH. First, we determined peak of the TRH response; the point of intersection is Erev the effect of raising extracellular [K+] on the reversal of An example of such an experiment is shown in Fig. ZTRH (Fig. 5A ) . A protocol essentially identical to that de4. The SEVC records used to construct the I- vplots, repre- scribed above (cf. Fig. 4) was performed, first in control sent approximately the first 300 ms of 2-s voltage steps (top Ringer solution (3 mM K+) and then in a Ringer solution traces). At the peak of the response to TRH, a steady in- with [K+] raised to 12 mM. In 3 mM K+, the Eev of ZTRH ward current was evident as a negative shift in holding was - 100 mV. However, in 12 mM K+ the point of inter0fh-I

l

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (018.218.056.169) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

TRH EFFECTS ON HYPOGLOSSAL

2 mM BARIUM

CONTROL

I

40 mV

[

l CONTROL v 2 mM BAFWM

-3 -100

-80 MEMBRANE

I -40

-60 POTENTIAL

(mv)

FIG. 7. The Ba2+-induced inward current reverses near Ek. Reversal potential of the Ba2%nduced inward current was determined from instantaneous I- Vrelationships obtained in SEVC before and during exposure to 2 mM Ba2+. Raw records show current responses (top) to the voltage steps (bottom) applied in control conditions (top I&) and 12 min after bathing the slice in a solution containing 2 mM Ba2+ ( top right). Ba2+ caused the development of an inward current and a decreased input conductance. Instantaneous current response to each of the series of voltage steps was measured immediately after the capacitive transient but preceding the development of the slow inward current (marked by appropriate symbol) and plotted as a function of voltage measured at the same time point (bottom). Data points from control ( l ) and from Ba2+ (A ) were well fitted by linear regression; Ba2+ caused a negative shift and a decreased slope of the I-V relationship. Regression lines intersected at -89 mV, close to the presumed Ek. Inward current induced by 2 mM Ba2+ is thus very similar to that induced by TRH (cf. Fig. 4). RMP of the HM was -72 mV. The cell was held at -52 mV with 2. I-nA holding current in control; all other current measurements are plotted with reference to this holding current. For abbreviations, see Fig. 1 legend.

MOTONEURONS

1739

portantly, this effect cannot be attributed to tachyphylaxis because the size of the depolarizing response returned to near control levels (7 mV) after washing the slice in 3 mM K+ (bottom). With this protocol we found that the TRHinduced depolarization was attenuated in solutions with 12 mM K+ by 70% (P < 0.0 1; n = 4). These results, together with those in which raised extracellular [K+] shifted Erev of ZTRH, provide convincing evidence that reduction of a K+ conductance contributes to ZTRH in HMs. It is apparent from the Z-V relationships presented (cf. Fig. 4) that ZTRH is larger at more depolarized potentials, presumably by virtue of the increased driving force for K+ at potentials farther from EK. The functional consequence of this property is illustrated in Fig. 6, where the effect of TRH was tested at different membrane potentials. The TRH-induced depolarization at -63 mV was of sufficient magnitude to cause the cell to reach firing threshold (left, top trace). As the cell was held at ever more hyperpolarized potentials, the size of the response was progressively diminished (Ieft). The relationship between the magnitude of the depolarizing response and holding potential is shown for all cells tested using this protocol (n = 7). A regression line fitted to the scatter plot predicted a 2.5-mV increase in TRH-induced depolarization for a 10-mV depolarizing shift in membrane potential (P < 0.0 1) and predicted a null effect at -89 mV. Thus the effect of TRH on HMs is more pronounced at depolarized potentials, a finding again consistent with K+ current involvement. A component of ITRH is Ba2+ sensitive

Barium blocks various types of K+ channels, including resting “leak” K+ channels (Hille 199 1; Jones 1989). In addition, it was recently demonstrated in neonatal spinal motoneurons that the K+ conductance blocked by TRH is Ba2+ sensitive (Nistri et al. 1990). We therefore tested I ) whether adult HMs possess a Ba2+-sensitive resting K+ conductance and 2) if inhibition of such a conductance section was shifted +23 mV so that ZTRH reversed at -77 could account for all of the effects of TRH. mV. Therefore Z&V of ZTRH was shifted in the depolarizing We found that when the media perfusing the slice was direction by raised extracellular [K+] as expected for a K+ switched from normal Ringer to one containing 2 mM current. The magnitude of the shift was, however, smaller Ba2+, HMs underwent a slow depolarization, which like than predicted by the Nemst equation (23 vs. 37 mV). It is that caused by TRH was associated with a decrease in mempossible that the slice had not completely equilibrated with brane input conductance. The effect of Ba2+ on a HM rethe high [K+] solution at the time of the TRH application. corded in SEVC is shown in Fig. 7. The top traces show Nevertheless, the depolarizing shift in Erev is good evidence current responses to a family of voltage steps obtained bethat ZTRH results from inhibition of a K+ current. Second, it fore (Ieft) and during perfusion of the slice with 2 mM Ba2+ is apparent from Fig. 5A that when Erev of ZTRH is shifted in ( right). In the presence of Ba2+, a steady inward current is the depolarized direction by high extracellular [K+], the apparent (negative shift in holding current), as is a deTRH current (i.e., the difference between TRH and control creased input conductance (decreased magnitude of the iniZ-I+) around the resting membrane potential is substan- tial current jumps). Instantaneous I- V relationships (bottially reduced. So, if a K+ current is involved, raising extra- tom) obtained from these records in control (0) and in 2 cellular K+ should diminish the TRH-induced depolarizamM Ba2+ ( A) were well fitted by linear regression and intion. An example from a representative experiment in tersected at -89 mV very near the presumed EK. In another which this prediction was tested is shown in Fig. 5 B. In this HM tested in this way, the Ba2+-induced inward current case, TRH was pressure ejected from a micropipette near a again reversed at -89 mV. Thus 2 mM Ba2+ apparently HM that was held in DCC at -60 mV. In the control period depolarized HMs by decreasing a resting K+ conductance. (top), TRH caused a depolarization of 9 mV. During expo- The effects of Ba2+ were remarkably similar to those of sure to 12 mM K+ (middle) the TRH-induced depolarizaTRH (cf. Fig. 4)) consistent with the hypothesis that TRH tion was markedly diminished, measuring only 2 mV. Im- and Ba2+ affect the same conductance.

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (018.218.056.169) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

D. A. BAYLISS,

1740

A 2 mM Ca*‘/

F. VIANA,

AND

0 Ba*+

l

A. J. BERGER

B 2 mM Ba*‘/

CONTROL

0 Ca”

-4

v a c v-6

4 -6 t.

- ,

-100

-80

l

TOTAL I,,, a Ba2+-INSENSITIVE

l

I

-100

I -80

l -80

1

D Ba2+-SENSITIVE

C TRH CURRENTS

-2

’

-100

-60

TRH

-2 I -60

MEMBRANE

4

It

I

-100

POTENTIAL

I I I : Ere”= -88 mV I 1 -80

l -60

I

I,,,

I -60

I

(mv)

8. Ba2+ blocks a component of TRH-sensitive current ( ITRH). An occlusion paradigm was used to determine if the inward currents induced by TRH and Ba2+ are mediated by the same K+ conductance. Instantaneous I-V relationships generated in control ( l , o ) and in 10 PM TRH (v, v ) were used to determine the TRH-sensitive currents ( cf. Fig. 4)) first in normal Ringer solution (A, l , v) and then in Ringer containing 2 mM Bazf /O Ca2’ (B, o, v ). In normal Ringer (A), TRH ( 10 PM) caused the customary inward current and decreased input conductance. Data were well fitted with linear regression; the regression lines intersected at -98 mV. In 2 mM Ba2+/0 Ca2+ (B), TRH again caused the development of an inward current. Data were well fitted with linear regression, but the slopes of the regression lines were essentially identical, indicating no change in input conductance. The JTRHwere calculated as difference currents from TRH and control I-Vs, and plotted with respect to membrane potential (C). Total I TRH (m) had a voltage dependence similar to what we described previously (i.e., JTRHis greater at depolarized potentials and reverses at -98 mV). By contrast, &H measured in 2 mM Ba2+ ( Ba2+-insensitive current, o) displayed essentially no voltage dependence and did not reverse over the voltage range tested. The &a 2+-sensitive ITRHwas determined by subtracting the Ba2’-insensitive &H from the total ITRH and plotted against membrane potential (D) . This current component was voltage dependent and reversed at - 88 mV, very close to the presumed EK . Thus the voltage dependence of the Ba2%ensitive component of&H was very similar to that of the current induced by BaZ’ alone, indicating that Ba2+ and TRH interact with the same K+ conductance. However, these data also indicate that TRH induces an additional inward current component that is not occluded by Ba2+. RMP of the HM was -70 mV. A and B: holding potential, -60 mV, holding currents, 1.7 and 0.3 nA, respectively. All other currents in A and Bare referenced to the respective holding currents. Cell was recorded in 0.5 PM tetrodotoxin (TTX). FIG.

We tested directly the possibility that TRH and Ba*+ act on the same conductance in HMs with an occlusion paradigm. An example of such an experiment performed in SEVC is shown in Fig. 8. Instantaneous Z-V relationships were used to define Zm, first in normal Ringer (2 mM Ca*+/O Ba*‘) and then in Ringer containing 2 mM BaZ+/O Ca*+ . The I- V relationships in normal Ringer ( Fig. 8A ) have the typical shape described previously, reflecting the inward current and decreased conductance caused by TRH. However, the response in 2 mM Ba*‘/O Ca*’ ( Fig. 8B) was very different. Although TRH caused the development of a small inward current, there was no measurable effect on input conductance (i.e., the slopes of the two Z-V relationships were essentially identical). Averaged data (n = 4) revealed that in the presence of Ba*+, input conductance of HMs in 10 PM TRH was essentially unchanged ( 10 1.3 t 2.6% ofcontrol), whereas ZmH was -0.6 t 0.1 nA. The TRH-sensitive currents for the HM in Fig. 8, obtained

by subtracting control from TRH I-‘vs, are shown in Fig. 8C The Ba*+-insensitive ZTRH (Fig. 8C; 0) displayed esse&ally no voltage dependence, at least .over the voltage range tested, and was quite distinct from the total ZTRH, which was voltage dependent and reversed near - 100 mV (Fig. 8C; H). The Ba*+ -sensitive TRH current, shown in Fig. 8 D, represents the difference between the total ZmH and the Ba*+ -insensitive ZmH. This component displays a voltage-dependence and reversal potential ( -88 mV) that is very reminiscent of the inward current induced by Ba*+ alone (cf. Fig. 7). Therefore Ba*+ blocked a component of ZTRH and blocked the effect of TRH on input conductance. The Ba*+-sensitive component of ZTRH was similar to the current induced by Ba*+ alone, indicating that Ba*+ and TRH interact with the same resting K+ conductance in HMs. These data also show that TRH induces a current component that is not blocked by Ba*+. The existence of a residual Ba2+-insensitive TRH-in-

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (018.218.056.169) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

TRH EFFECTS ON HYPOGLOSSAL

1741

MOTONEURONS

A CONTROL -53mV -66mV

4

al I

I

20 mV 2nA

B 2 mM Ba2+ TRH

bl

.. . .. . ..

-64 mV

3 min 100 ms

b2

-52 mV bb?

.$-

I 9. Ba2+ occludes the conductance decrease but not the depolarization induced by TRH. Effect of 2 mM Ba2+ on the TRH-induced depolarization was determined. A : TRH ( 10 PM) was applied to HM for the period indicated while the slice was perfused with standard Ringer. The cell depolarized 13 mV at the peak of the response (a 1) before recovering to its original potential. After the recovery, the cell was depolarized to the peak potential (-53 mV) with DC current injection (a2). Comparison of the voltage responses to constant amplitude (-0.5 nA) current pulses on a faster time base (far right) revealed that the input resistance was higher at the peak of the TRH effect (a 1) than in the recovery period (a2). B: after perfusing the slice in Ringer solution containing 2 mM Ba2+/0 Ca2+ and 0.5 PM tetrodotoxin (TTX), TRH ( 10 PM) was again applied to the HM. The depolarizing response to TRH under these conditions was very similar in time course and magnitude ( 12 mV, b 1) to the previous response ( cf. Fig. 1OA ) . After recovery from the TRH-induced depolarization, the cell was depolarized to the peak potential ( -52 mV) by DC current injection (b2). Voltage responses to the current pulses (far right) at the peak of the response to TRH (b 1) and in the recovery period (b2) were not different. Thus, in contrast to the previous response, TRH had no effect on the input conductance of the HM despite causing a depolarization of similar magnitude. These data are consistent with the voltage-clamp records that revealed a Ba2+ -resistant inward current induced by TRH that was not associated with a change in input conductance (cf. Fig. 9). Responses recorded in discontinuous current clamp (DCC) from RMP. For abbreviations, see Fig. 1 legend. FIG.

duced current was particularly obvious during currentclamp recordings, when this current component caused a substantial depolarization often as large as that recorded in normal Ringer solution. An example of this is shown in Fig. 9. In normal Ringer, current pulses of constant amplitude were delivered to a HM as it depolarized during a brief exposure to TRH. After it had recovered, DC current injection was used to return the cell to the peak potential for an isopotential comparison of the electrotonic response to the current pulse. The records of those voltage responses show the typical increase in input resistance at the peak of the TRH response (right). The identical protocol repeated in the same cell while the slice was perfused with 2 mM Ba2+ produced very different results. The TRH-induced depolarization was essentially as large as in the control condition; yet, examination of the voltage responses to the current pulses revealed that there had not been the customary increase in input resistance. Thus Ba2+ did not block the TRH-induced depolarization despite the fact that it effectively inhibited the effect of TRH on HM input resistance. Presumably, the reduction in ZTRH by Ba2+ is offset somewhat by the Ba2+ -induced increase in membrane resistance, leaving the TRH-induced depolarization relatively unaffected by Ba2+.

TRH shifts the input-output relationship of HMs The results presented to this point indicate that the predominant effect of TRH on HMs is a depolarization result-

ing not only from the inhibition of a Ba2+-sensitive resting K+ conductance by TRH but also from a Ba2+-resistant inward current induced by TRH. There are at least two functional consequences of these effects on HMs. First, and most obviously, the depolarization moves the cell closer to the threshold for generation of action potentials. Indeed, under our experimental conditions TRH often depolarized cells to beyond firing threshold (cf. Fig. 2). The second functional consequence, a result of the effect of TRH on input conductance, is demonstrated in Fig. 10. Here we analyzed the effect of TRH on the input-output characteristics of the cell. The top records show the spike-firing response of a HM to injection of long (1 s) depolarizing current steps, before and during application of TRH. The response in TRH was recorded while DC current was applied to overcome the TRH-induced depolarization. The cell responded to the current pulse with far more spikes in the presence of TRH. This is shown graphically (bottom left); instantaneous firing frequency in TRH was substantially higher throughout the pulse. The averaged steadystate firing frequency (final 250 ms) was determined from each of a family of such depolarizing current steps under the two conditions and plotted as a function of injected current (bottom right). This steady-state f-Z relationship was shifted to the left by TRH, indicating a reduction in threshold for minimal repetitive firing. The threshold for minimal repetitive firing was shifted from 1.4 t 0.2 to 0.8 t 0.1 nA by TRH (n = 6; P < 0.01). Further examination of the f-Z curves revealed that TRH had no effect on the slope

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (018.218.056.169) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

D. A. BAYLISS, Fe VIANA,

1742

AND A. J. BERGER

TRH

CONTROL

//

-65 mV m

7

-65 mV 250 ms

I

INSTANTANEOUS

STEADY

STATE

I

TIME (ms)

CURRENT

FIG. 10. TRH shifts the input-output relationship of HMs. We determined the effect of TRH on repetitive firing properties of HMs. The spike-firing response to a 1.4-nA depolarizing current pulse ( 1 s) was recorded in control conditions (top /eJ) and then while the slice was bathed in 10 PM TRH (top righr ). After TRH had caused the cell to depolarize, DC current was injected to return the HM to the same membrane potential as in the control period (-65 mV). In the presence of TRH, the same magnitude current pulse from the same membrane potential caused the HM to discharge many more action potentials. This is illustrated graphically (bottom lefr) as the instantaneous firing frequency (calculated from interspike intervals) was higher throughout the full duration of the pulse. Steady-state firing frequency (average of the instantaneous firing frequencies over the final 250 ms) was determined from a family of depolarizing current pulses in control and in TRH and plotted with respect to injected current ( f-1 curve, hottom right). The f-I relationship was shifted to the left by TRH, indicating that smaller current inputs were necessary to reach the threshold for minimal repetitive firing. TRH had no effect on the slope of the f-1 curve. RMP was -70 mV. For abbreviations, see Fig. 1 legend.

(nA)

( 17.7 t 2.2 Hz/nA in control vs. 16.9 t 2.2 Hz/nA in TRH; NS), suggesting that action potentials and afterpotentials were relatively unaffected. This was confirmed by analysis of the effects of TRH on action potentials generated by brief (2 ms) depolarizing current pulses (n = 5). There were no substantial differences in spike duration (measured 10 mV above inflection; 553 t 20 vs. 520 t 8 ps), or in the amplitude (-7.3 t 0.5 vs. -7.7 t 0.4 mV) or duration (76.8 t 4.9 vs. 74.9 t 4.2 ms) of the m-AHP. DISCUSSION

Electrophysiological effects of TRH on adult HMs were determined with intracellular recording techniques. We confirmed an earlier finding that TRH acts directly to decrease the input conductance of HMs and causes them to depolarize (Rekling 1990) and have extended those findings to include an analysis of the underlying ionic conductances. We found two mechanisms that contribute to the TRH-induced depolarization. First, we showed that TRH inhibited a resting K+ conductance that is Ba*+ sensitive. In this respect, our findings on cranial motoneurons from adult rats are similar to those reported recently to explain the effects of TRH on neonatal rat spinal motoneurons (Nistri et al. 1990) and indicate that this may be a general mechanism by which TRH depolarizes rat motoneurons. Second, we identified a previously undescribed component of ZTRH that was not blocked by Ba*+ and not associated with a measurable change in input conductance, which also contributed to the depolarization. Furthermore, we have examined the functional consequences of the mechanism of action of TRH on spike-firing behavior of HMs. TRHinduced depolarization brought the cell closer to firing threshold, whereas the accompanying decrease in conductance promoted more effective transduction of input (current) to output (spike-firing).

The TRH- and Ba2+-sensitiveK+ conductance We found that the TRH-sensitive current reversed at - 10 1 mV. This reversal potential is close to but slightly negative of & calculated with the Nernst equation taking values of intracellular K+ concentration of HMs measured with ion-selective microelectrodes (approximately -9 1 mV) (Jiang and Haddad 199 1). Consistent with the predicted value of E,, we found that the inward current induced by 2 mM Ba*+, which was presumably an effect solely on a K+ conductance, reversed at very near the predicted EK (-89 mV). The slight discrepancy between the measured EreVof ZTRHand the calculated EK is, in fact, consistent with our demonstration that ZTRHrepresents the sum of two separate current components (cf. Fig. 8 ). One component of ZTRH, the Ba*+ -sensitive component, represents the inward current that resulted from inhibition of a K+ conductance by TRH. Indeed, this ZTRH component reversed very near the predicted EK (-88 mV), similar to the Ba*+-induced current. However, we found a second component of ZTRHthat was Ba*+ resistant and not associated with a measurable change in input conductance. Addition of this component to the Ba*+ -sensitive component caused a parallel downward shift in the relationship between the total ZTRH and membrane potential, and shifted the measured to a point negative of EK. The measured reverErev Of ITRH sal potential was thus determined by contributions from the two components of ZTRH. The resting TRH-sensitive K+ conductance in adult rat HMs that we describe was similar to that reported by Nistri et al. ( 1990) in neonatal rat spinal motoneurons inasmuch as it was blocked by barium. There were, however, some differences in the characteristics of the current. In neonatal spinal motoneurons, the current was found to reverse at far more hyperpolarized potentials ( - 120 mV). This disparity is not a function of the different extracellular K+ concentrations used in the two studies; the higher [K+] used in the

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (018.218.056.169) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

TRH

EFFECTS

ON

HYPOGLOSSAL

study of neonatal spinal motoneurons (4.5 vs. 3 mM) would actually favor a more depolarized reversal of ZTRH. The difference in reversal potentials may, however, be explained by the way in which measurements were made in the two studies. In our experiments, instantaneous Z- Vrelationships were used to estimate Z& of Z-rRH. This protocol provided a measure of the effects of TRH on conductances active at holding potentials where minimal activation of voltage-dependent conductances would be expected. In contrast, Nistri et al. ( 1990) used steady-state I- Vs for their estimates of E,,“. Steady-state Z-Ys reflect not only the effects of TRH on conductances active at the holding potential but also any effects that TRH might exert on voltageand time-dependent conductances activated by the voltage step. One such conductance expressed in neonatal rat spinal motoneurons underlies a hyperpolarization-activated inward current ( Zh) (Takahashi 1990). In fact, we have found by analysis of steady-state I- Vs that Zh was enhanced in some cells by TRH, and as predicted, reversal of the steady-state I’fRH occurred at more hyperpolarized potentials than reversal of instantaneous ZTRH in those cells (unpublished observations). This could explain the rather negative value of ZTRHreversal reported by Nistri et al. ( 1990). It remains to be clarified whether the enhancement of Zh by TRH observed in some cells represents an artifact of improved “space-clamp” or a true biological effect. In the earlier report on neonatal spinal motoneurons, the TRH-sensitive conductance ( GTRH) was reported to display a peculiar bell-shaped voltage-dependence (Nistri et al. 1990). Because the voltage-dependence was unlike that of any other known K+ conductances, the authors suggested that TRH was affecting a “novel” K+ conductance. However, their estimate of GTRH and its voltage-dependence was based on the measurement of ZTRHat steady state and therefore may be confounded by contributions from other voltage-activated currents (see above). In addition, their estimate of GTRH probably included a contribution from the Ba2+-insensitive current component that we have described. Thus the peculiar voltage-dependence of GTRH, attributed entirely to an effect on a novel resting K+ conductance, may actually have resulted from the combinatorial effects of TRH on a resting K+ current and on other currents. We have, therefore, considered whether any previously described Ba2+-sensitive K+ currents could represent the TRH-sensitive K+ current. A Ba2+-sensitive, K+ conductance that is blocked by muscarine (M-current) and modulated by a number of neuropeptides [e.g., luteinizing hormone-releasing hormone, somatostatin] has been described in many peripheral and central neurons (reviewed in Brown 1988). Because we recorded TRH-induced depolarizations at potentials that would be at the extreme of the reported range for M-current activation (Adams et al. 1982), we consider it unlikely that an M-like current, at least one with its classic voltage-dependence, is responsible for ZTRH. In clonal pituitary cells, TRH was reported to decrease a Ba2+ - and Cs+-sensitive, inwardly rectifying K+ current that is partially activated at rest (Bauer et al. 1990). The absence of any alinearities in the instantaneous Z-V relationships before or during TRH (cf. Figs. 4 and 5) sug-

MOTONEURONS

1743

gest that such a current probably is not involved in the response to TRH in HMs. A Ba2+-sensitive, resting K+ conductance that has little voltage sensitivity and is relatively unaffected by tetraethylammonium ions (TEA) has been described in frog sympathetic neurons (Jones 1989). Interestingly, a K+ conductance with similar pharmacologic properties has also been expressed in Xenopus oocytes after injection of mRNA isolated from rat brain (Parker et al. 1990). This expressed conductance is active at rest and modulated by agonist occupation of receptors that act via activation of the phospholipase C second-messenger pathway (Parker et al. 1990); the same pathway is activated by TRH in pituitary cells (Gershengorn 1989) and in Xenopus oocytes injected with TRH receptor mRNA (Straub et al. 1990). It is possible that such a conductance mediates the effects of TRH on HMs. Ba2+-insensitive

component of I,,

We described a component of Z-rRH that was resistant to occlusion by Ba2+ and that caused a substantial depolarization in the absence of any significant change in conductance. This component was revealed when the dominant effect of TRH (inhibition of a K+ conductance) was blocked by Ba2+. The contribution of this current under control conditions, although not immediately apparent, was evident in the slightly hyperpolarized position of the reversal potential of ZTRH (see above and Fig. 8). The charge carrier for the Ba2+ -resistant ZTRH has not been determined; the fact that the current is not associated with a measurable conductance change suggests that an ionic species with an equilibrium potential far from rest is involved. In this regard, it is possible that the residual current component represents the activation of a mechanism similar to that described in frog motoneurons. In those cells, TRH induces a small inward current, perhaps carried by Na+, which is associated with either no change or a slight increase in conductance (Lacey et al. 1989; Nicoll 1977). Alternatively, the Ba2+-resistant current may result from the activation of a voltage-independent, dihydropyridine-sensitive, divalent cation conductance, similar to that activated by TRH in bovine lactotrophs (Mason et al. 1988). To our knowledge, such a conductance has not been described in motoneurons. Interestingly, we have recently found that serotonin [ 5-hydroxytryptamine ( 5-HT)] , whichis colocalized with TRH (Johansson et al. 198 1), induces a current with properties similar to the Ba2+ -resistant component of Z-rRH in whole-cell recordings from neonatal HMs (Berger et al. 1992). Physiological

sign@ance

of TRH effects on H.Ms

We were able to demonstrate directly that TRH enhances the excitability of HMs. This effect of TRH was a result of I) the TRH-induced depolarization of the cell and 2) the decreased input conductance caused by TRH. The latter effect of TRH can account for the leftward shift we observed in the input-output relationship of the motoneurons. Thus in the presence of TRH, less excitatory synaptic current would be required to elicit a threshold response, and a greater response would be evoked for any given su-

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (018.218.056.169) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

1744

D. A. BAYLISS, F. VIANA,

prathreshold synaptic current. In contrast to the effects of 5-HT on neonatal HMs (Berger et al. 1992), we found no effect of TRH on the m-AHP or on the slope of the f-1 relationship. It is important to note that, because of the decreased membrane conductance, inhibitory synaptic currents would also be more effective in causing hyperpolarization in cells exposed to TRH (cf. Fig. 2 in which hyperpolarizing current pulses caused a larger electrotonic response during exposure to TRH). TRH had essentially the same effect in all HMs tested regardless of their location within the nucleus, a finding consistent with the relatively uniform distribution of TRH binding sites within the nucleus (Manakar and Rizio 1989). Thus TRH acts both on ventral hypoglossal motoneurons controlling tongue protrusion and dorsal motoneurons controlling tongue retraction. If TRH acts in a parasynaptic fashion, as suggested for other neuropeptides (Zieglg%nsberger et al. 199 1 ), it would act simultaneously on dorsal and ventral motoneuron pools that activate functionally antagonistic muscle groups. Because TRH potentiates excitatory and inhibitory synaptic inputs (see above and see Wang and Dun 1990), such a parasynaptic effect would enhance the contrast between activation and inhibition of motoneurons acting on antagonistic m uscle groups. Further, this effect could be maintained over long periods as the response was long-lasting and desensitized only slightly during prolonged TRH applications ( cf. Figs. l-3 ) . The effects of TRH on HMs may be particularly relevant to the control of upper airway patency and to the pathophysiology of obstructive sleep apneas. Many hypoglossal motoneurons receive phasic synaptic inputs tightly coupling their activity, as well as the activity of the genioglossus muscle (a tongue protruder), to the respiratory rhythm (Remmers et al. 1978; Withington-Wray et al. 1988). The inspiratory-related burst of genioglossus activity stabilizes the tongue to prevent it from collapsing on the upper airway during inspiration (Remmers et al. 1978). The respiratoryrelated activity of genioglossus is graded over the sleepwake cycle, being highest during active waking, intermediate during nonrapid eye movement (NREM) sleep and lowest during rapid eye movement (REM) sleep (Lowe 1990). In fact, obstructive sleep apneas associated with an abrupt decrease in respiratory-related genioglossus activity tend to occur during phasic REM sleep (Wiegand et al. 199 1). The TRH innervation of the dorsal medulla oblongata is derived predominantly, perhaps exclusively, from the midline raphe complex of the medulla (Bayliss et al. 1992; Lynn et al. 199 1; Palkovits et al. 1986). Activity of raphe neurons is also graded in a state-dependent manner, much like that of genioglossus; raphe cells are most active during waking, less active during NREM sleep and show very low levels of activity during REM sleep (Fomal et al. 1985). Therefore, one could speculate that high levels of TRH release by raphe neurons while they are most active (e.g., during waking) would enhance the responsiveness of HMs to phasic respiratory-related synaptic inputs, and the HMs would provide the necessary motor output to maintain airway patency. Conversely, low levels of TRH-release during periods of relative quiescence in raphe neurons (e.g., during REM sleep) would decrease the excitability of HMs.

AND A. J. BERGER

In this case, the identical respiratory-related be rendered less effective and the likelihood struction would be increased.

inputs would of airway ob-

The authors express their appreciation to Dr. William Satterthwaite for technical assistance, especially in the development of data acquisition software and to H. Atkins for her help in the preparation of the paper. The work was supported by a Javits Neuroscience Research Award (NS14857) to A. J. Berger. D. A. Bayliss was supported by a Parker B. Francis Fellowship. Address for reprint requests: D. A. Bayliss, Dept. of Physiology and Biophysics, SJ-40, Univ. of Washington School of Medicine, Seattle WA 98195. Received 13 March 1992; accepted in final form 25 June 1992. REFERENCES P. R., BROWN, D. A., AND CONSTANTI, A. M-currents and other potassium currents in bullfrog sympathetic neurons. J. physiol. Lund. 330: 537-572, 1982. AGHAJANIAN, G. K. AND RASMUSSEN, K. Intracellular studies in the facial nucleus illustrating a simple new method for obtaining viable motoneucons in adult rat brain slices. Synapse NY 3: 33 l-338, 1989. BAUER, C. K., MEYERHOF, W., AND S~HWARZ, J. R. An inward-rectifying K+ current in clonal rat pituitary cells and its modulation by thyrotrophin-releasing hormone. J. Physiol. Lond. 429: 169- 189, 1990. BAYLISS, D. A., CZYZYK-KRZESKA, M. F., SZYMECZEK, C. L., ERICKSON, J. T., SEROOGY, K. B., AND MILLHORN, D. E. Specific mRNAs in neurons identified by projection or neurochemical content: in situ hybridization combined with retrograde tract-tracing and immunohistochemistry. In: In Vitro-In Vivo Autoradiography and Correlative Imaging, edited by W. E. Stumpf and J. F. Solomon. New York: Academic. In press. BAYLISS, D. A., VIANA, F., AND BERGER, A. J. Thyrotropin-releasing hormone decreases a potassium conductance in hypoglossal motoneurons in vitro. Sot. Neurosci. Abstr. 17: 464, 199 1. BERGER, A. J., BAYLISS, D. A., AND VIANA, F. Modulation of neonatal rat hypoglossal motoneuron excitability by serotonin. Neurosci. Lett. 143: 164-168, 1992. BROOKS, B. R. A summary of the current position of TRH in ALS therapy. Ann. NYAcad. Sci. 553: 431-461, 1989. BROWN, D. M-currents: an update. Trends Neurosci. 11: 294-299, 1988. BROWNSTEIN, M. J., PALKOVITS, M., SAAVEDRA, J. M., BASSIRI, R. M., AND UTIGER, R. D. Thyrotropin-releasing hormone in specific nuclei of rat brain. Science Wash. DC 185: 267-269, 1974. FINKEL, A. S. AND REDMAN, S. J. Optimal voltage clamping with single microelectrodes. In: Voltage and Patch Clamping with Microelectrodes, edited by T. G. Smith, Jr., H. Lecar, S. J. Redman, and P. W. Gage. Bethesda, MD: Am. Physiol. Sot., 1985, p. 95-120. FORNAL, C., AUERBACH, S., AND JACOBS,B. L. Activity of serotonin-containing neurons in nucleus raphe magnus in freely moving cats. Exp. Neural. 88: 590-608, 1985. FUKUDA, A., NABEKURA, J., ITO, C., PLATA-SALAMAN, C. R., AND OoMURA, Y. Developmentally different onset of CY l- and cu2-adrenergic responses in neonatal rat dorsal motor nucleus of the vagus in vitro. Brain Res. 493: 357-36 1, 1989. FULTON, B. P. AND WALTON, K. Electrophysiological properties of neonatal rat motoneurones studied in vitro. J. Physiol. Land. 370: 65 l-678, 1986. GERSHENGORN, M. C. Mechanism of signal transduction by TRH. Ann. NYAcad. Sci. 553: 191-196, 1989. HADDAD, G. G., DONNELLY, D. F., AND GETTING, P. A. Biophysical prop erties of hypoglossal neurons in vitro: intracellular studies in adult and neonatal rats. J. Appl. Physiol. 69: 1509- 15 17, 1990. HILLE, B. Ionic Channels ofExcitable Membrane. Sunderland, MA: Sinauer, 199 1. H&FELT, T., FUXE,K.,JOHANSSON,O., JEFFCOATE,~., AND WHITE,N. Distribution of thyrotropin-releasing hormone (TRH) in the central nervous system as revealed by immunohistochemistry. Eur. J. Pharmacol. 34: 389-392, 1975a. H~KFELT, T., FUXE, K., JOHANSSON,O.,JEFFCOATE,S.,AND WHITE,N. Thyrotropin releasing hormone (TRH)-containing nerve terminals in ADAMS,

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (018.218.056.169) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

TRH EFFECTS ON HYPOGLOSSAL certain brain stem nuclei and in the spinal cord. Neurosci. Lett. 1: 133-139, 1975b. JIANG, C. AND HADDAD, G. G. Effect of anoxia on intracellular and extracellular potassium activity in hypoglossal neurons in vitro. J. Neurophysiol. 66: 103-l 11, 1991. JOHANSSON, O., H&FELT, T., PERNOW, B., J-ATE, S. L., WHITE, N., STEINB~~~H, H. W. M., VERHOFSTAD, A. A. J., EMSON, P. C., AND SPINDEL, E. Immunohistochemical support for three putative transmitters in one neuron: coexistence of 5-hydroxytryptamine, substance Pand thyrotropin releasing hormone-like immunoreactivity in medullary neurons projecting to the spinal cord. Neuroscience 6: 1857- 188 1, 1981. JONES,S. W. On the resting potential of isolated frog sympathetic neurons. Neuron 3: 153-161, 1989. LA-, G., NISTRI, A., AND RHYS-MAITLAND, E. R. Large enhancement of excitatory postsynaptic potentials and currents by thyrotropin-releasing hormone (TRH) in frog spinal motoneurons. Brain Res. 488: 8088, 1989. LECHAN, R. M., WV, P., AND JACKSON, I. M. D. Immunolocalization of the thyrotropin-releasing hormone prohormone in the rat central nervous system. Endocrinology 119: 12 1O- 12 16, 1986. LQVVE, A. A. The tongue and airway. Otolaryngol. Clin. North Am. 23: 677-698, 1990. LYNN, R. B., KREIDER, M. S., AND MISELIS, R. R. Thyrotropin-releasing hormone-immunoreactive projections to the dorsal motor nucleus and the nucleus of the solitary tract of the rat. J. Comp. Neurol. 3 11: 27 l288, 1991. h&umKER, S., AND RIZIO, G. Autoradiographic localization of thyrotropin-releasing hormone and substance P receptors in the rat dorsal vagal complex. J. Comp. Neurol. 290: 5 16-526, 1989. MASON, W. T., RAWLINGS, S. R., COBBET~, P., SIKDAR, S. K., ZOREC, R., AKERMAN, S. N., BENHAM, C. D., BERRIDGE, M. J., CHEEK, T., AND MORETON, R. B. Control of secretion in anterior pituitary cells-linking ion channels, messengers and exocytosis. J. Exp. Biol. 139: 287-3 16, 1988. MOSFELDT LAURSEN, A. AND REKLING, J. C. Electrophysiological properties of hypoglossal motoneurons of guinea-pigs studied in vitro. Neuroscience,30: 6 19-637, 1989. NI~~LL, R. A. Excitatory action of TRH on spinal motoneurons. Nature Lond. 265: 242-243, 1977. NISTRI, A., FISHER, N. D., AND GURNELL, M. Block by the neuropeptide TRH of an apparently novel K+ conductance of rat motoneurons. Neurosci. Lett. 120: 25-30, 1990. PALKO~, M., MEZEY, E., E~KAY, R. L., AND BROWNSTEIN, M. J. Innervation of the nucleus of the solitary tract and the dorsal vagal nucleus by thyrotropin-releasing hormone-containing raphe neurons. Brain Res. 373: 246-25 1, 1986. PARKER, I., PANICKER, M. M., AND MILEDI, R. Serotonin receptors ex-

MOTONEURONS

1745

pressed in Xenopus oocytes by mRNA from brain mediate a closing of IV membrane channels. Mol. Brain Res. 7: 3 l-38, 1990. REICHLIN, S. TRH: historical aspects. Ann. NY Acad. Sci. 553: l-6, 1989. REKLING, J. C. Excitatory effects of thyrotropin-releasing hormone (TRH) in hypoglossal motoneurons. Brain Res. 5 10: 175-l 79, 1990. REMMERS, J. E., DEGROOT, W. J., SAUERLAND, E. K., AND ANCH, A. M. Pathogenesis of upper airway occlusion during sleep. J. Appl. Physiol. Respir. Environ. Exercise Physiol. 44: 93 l-938, 1978. SEGERSON, T. P., HOEFLER, H., CHILDERS, H., WOLFE, H. J., WV, P., JACKSON, I. M. D., AND LECHAN, R. M. Localization of thyrotropin-releasing hormone prohormone messenger ribonucleic acid in rat brain by in situ hybridization. Endocrinology 121: 98-107, 1987. SHARIF, N. A. Quantitative autoradiography of TRH receptors in discrete brain regions of different mammalian species. Ann. NY Acad. Sci. 53: 147-175, 1989. STRAUB, R. W., FRECH, G. C., JOHO, R. H., AND GERSHENGORN, M. C. Expression cloning of a cDNA encoding the mouse pituitary thyrotropin-releasing hormone receptor. Proc. Natl. Acad. Sci. USA 87: 95 149518, 1990. TAKAHASHI, T. Thyrotropin-releasing hormone mimics descending slow synaptic potentials in rat spinal motoneurons. Proc. R. Sot. Lond. B Biol. Sci. 225: 39 l-398, 1985. TAKAHASHI, T. Inward rectification in neonatal rat spinal motoneurons. J. Physiol. Lond. 423: 47-62, 1990. TSURUO, Y., H&FELT, T., AND VISSER, T. Thyrotropin-releasing hormone (TRH)-immunoreactive cell groups in the rat central nervous system. Exp. Brain Res. 68: 2 13-217, 1987. VIANA, F., GIBBS, L., AND BERGER, A. J. Double- and triple-labeling of functionally characterized central neurons projecting to peripheral targets studied in vitro. Neuroscience 38: 829-84 1, 1990. WANG, M. Y. AND DUN, N. J. Direct and indirect actions of thyrotropinreleasing hormone on neonatal rat motoneurons in vitro. Neurosci. Lett. 113: 349-354, 1990. WHITE, S. R. A comparison of the effects of serotonin, substance P and thyrotropin-releasing hormone on excitability of rat spinal motoneurons in vivo. Brain Res. 335: 63-70, 1985. WIEGAND, L., ZWILLICH, C. W., WIEGAND, D., AND WHITE, D. P. Changes in upper airway muscle activation and ventilation during phasic REM sleep in normal men. J. Appl. Physiol. 7 1: 488-497, 199 1. WINOKUR, A. AND UTIGER, R. D. Thyrotropin-releasing hormone: regional distribution in rat brain. Science Wash. DC 185: 265-267, 1974. WITHINGTON-WRAY, D. J., MIFFLIN, S. W., AND SPYER, K. M. Intracellular analysis of respiratory-modulated hypoglossal motoneurons in the cat. Neuroscience 25: 1041-1051, 1988. ZIEGLG~NSBERGER, W., DODT, H.-U., DEISZ, R., AND PAWELZIK, H. Peptides in neurotransmission. In: Volume Transmission in the Brain: Novel Mechanisms for Neural Transmission, edited by K. Fuxe and L. F. Agnati. New York: Raven, 199 1, p. 37 l-380.

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (018.218.056.169) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.