Antagonistic effects of somatostatin and substance P on respiratory regulation in the rat ventrolateral medulla oblongata

Brain Research, 556 (1991) 13-21 (~) 1991 Elsevier Science Publishers B.V. All rights reserved. 0006-8993/91/$03.50 ADONIS 0006899391168843

13

BRES 16884

Antagonistic effects of somatostatin and substance P on respiratory regulation in the rat ventrolateral medulla oblongata Zibin Chen*, GOran Engberg, Thomas Hedner and Jan Hedner Departments of Pharmacology and Clinical Pharmacology, University of G6teborg and Sahlgrenska Hospital, G6teborg (Sweden) (Accepted 26 March 1991)

Key words: Somatostatin; Substance P; Respiratory regulation; Apnoea; Nucleus paragigantocellularis; Ventrolateral medulla

Substance P (SP) in the dose range 0.75-1.5 nmol exerts a potent stimulatory effect on ventilation after mieroinjection into the rat ventrolateral medulla oblongata (VLM; n. retieularis lateralis, n. paragigantocellularis lateralis). A significant but less pronounced effect is also seen in the dorsal medulla (DM; n. tractus solitarius). Somatostatin (0.6-1.8 nmol) inhibited ventilation and induced apnoea after microinjection into the VLM but not the DM. Serial microinjections of the two peptides showed a reciprocal antagonistic action in the VLM but not in the DM. The apnoea-inducing effect of SOM was blunted by SP while SOM reduced the ventilatory stimulation induced by SP. Extracellular single unit recordings were performed following the microiontophoretic application of SP and/or SOM to respiratory-related and non-respiratory-related neurons in the VLM and DM. Although a heterogeneous population of neurons were recorded from, the majority of respiratory-related units in the VLM responded with excitation to SP and inhibitory to SOM. A direct interaction between the peptides was seen in some respiratory-related units. The neurons not responding to either of the peptides were usually non-respiratory. Dorsal to the VLM, the type of response to the two peptides was less likely to be antagonistic and a wider distribution of response types were recorded. The results indicate a direct physiological antagonism between SP and SOM regarding their effects on respiratory regulation elicited in the VLM. INTRODUCTION

The ventrolateral medulla oblongata (VLM) is an important relay station for respiratory control, containing a number of important 'classical' transmitters and neuropeptides. Immunohistochemical mapping has revealed that somatostatin (SOM) and substance P (SP) containing cell bodies and nerve fibers are abundant in the V L M region 11,19. Recent reports also suggest that SOM has important inhibitory actions on respiration 12'14,15"18,3°. In support of the observations of Y a m a m o t o and coworkers 3°, we found that microinjections of S O M into the V L M readily induced apnoea, while administration into the dorsal medulla (DM) was ineffective in this regard 7. SP has been shown to induce basically opposite effects on ventilation with an excitatory effect on tidal volume (Va-) and minute ventilation following administration into the V L M 4'5. The V L M , especially in the region of the nucleus paragigantocellularis (nPGi) has been suggested to have an important role for the integration of ventilatory neuronal drive inputs '°, whereas the DM, especially the neural population of the nucleus of the solitary tract

(nTS), seems to be involved in different respiratory reflexes and control mechanisms influencing the pattern of breathing x°. Thus, experiments in the cat applying focal cooling of the regions of nPGi and n. preolivaris of the V L M have been shown to result in apnoea, while cooling or chemical lesions in the nucleus of the solitary tract of the D M mainly influenced the timing of the respiratory cycle2,23. We have also demonstrated that local application of SOM as well as SP in the nPGi, n. reticularis lateralis (nLRt) and the n. ambiguus (nA) induced profound effects on respiratory regulation 4-7. Thus, microinjection of S O M (0.6-1.8 nmol) decreases VT while the respiratory frequency (f) remains essentially unaltered due to a reduction of inspiratory drive 7. SP administration (0.75-2.25 nmol) into the same area results in an increase in Va- and a decrease in f due to a stimulation of inspiratory drive 4,5. Furthermore, local application of a SP antagonist or S O M in the V L M has been shown to induce apnoea 6,7,3°. D u e to the opposite respiratory effects of S O M and SP, the aim of the present study was to investigate whether a functional physiological or pharmacological interaction exists between these peptides.

* Dr. Z. Chen was on leave from the Department of Physiology, Shanghai Medical University, 138 Yi Xue Yuan Road, Shanghai 200032, People's Republic of China. Correspondence: Z. Chen, Department of Pharmacology, University of G6teborg, Box 33031, S-400 33 G6teborg, Sweden.

14 MATERIALS AND METHODS Experiments were performed on 161 male Sprague-Dawley rats (Anticimex, S6dert~lje, Sweden) weighing 250-350 g.

Animal preparation One to 3 days before experiments, the animals were prepared under pentobarbital (Pentothal, 40 mg/kg, i.p.) anesthesia. The head of the rat was placed into a stereotaxic frame. One or two 26-gange stainless-steel tubes (13.5 mm long) were lowered stereotaxically, unilaterally or bilaterally in the medulla, to a point 1.5 ram dorsal to the intended site(s) of microinjection. The coordinates for the microinjection sites were all chosen and are described according to the atlas of Paxinos and Watson 27. The guide cannula was then anchored with cranioplastic cement to metallic screws placed in the skull and occluded with a stainless-steel stylet. The animals were housed in the department with food and water ad libitum to allow full recovery from the surgical interventions.

Experimental procedures The trachea were cannulated with a Venflow cannula (diameter 2.0 mm Viggo AB, Helsingborg, Sweden) under xylazine (Rompun, 12 mg/kg) and ketamine (Ketalar, 60 mg/kg) anesthesia (i.p.). The ventral tail artery was exposed and cannulated (PP50, polyethylene; Portex, Hythe, Kent, U.K.) for continuous blood pressure monitoring. All animals were then placed in a closed cylinder-formed body plethysmograph (internal diameter 80 mm, length 300 ram) and further anesthesia was maintained with 0.7% halothane (Halothan, Hoechst, E R . G . ) in 0 2 continuously administered via the tracheal cannula by means of a Draeger Vaporizer. The body plethysmograph contained openings at both ends for connections of the arterial and the local injection cannulae to the exterior. The interior of the plethysmograph was connected to a Grass polygraph via a low pressure transducer (Statham P 23 Db). Heart rate (HR) was calculated from an electrocardiogram recorded simultaneously from subcutaneously placed electrodes in the two front limbs and the right hind limb of the rat. Mean arterial pressure (MAP) was calculated from the blood pressure recordings. Stainless steel tubes (32-gauge, 15 mm long) for drug administration were inserted into the cannula guide tube. Ten to 20 min were allowed to pass with the animal in the plethysmograph before obtaining the control values. Tidal volume (VT) and respiratory frequency (f) were continuously recorded by the Grass Polygraph. Minute ventilation (I)'E) was calculated according to the formula V T X f = I;"E. Inspiratory time (TI) and expiratory time (TE) were calculated from the respiratory tracing at high chart speed. The internal temperature of the plethysmograph was continuously registered and rectal temperature was measured with a telethermometer (Opti-lab Instrumentation AB, Sweden). SOM (SRIF; Growth hormone release inhibitory factor) or SP (both from Sigma Chemical Co., St. Louis, MO, U.S.A.) were dissolved in saline to give final concentrations of 2 /zg/pl. A Hamilton microsyringe was used for microinjections and 0.6-3.0 nmol (0.5-2.5/zl) of SOM or 0.75-2.25 nmol (0.5-1.5/~l) of SP was given as a single or combined injection locally at each site to the experimental animals. Control experiments were performed by injection of a corresponding volume of saline into several of the sites tested. At the end of each experiment, during anesthesia, the rats were given an intra-arterial injection of pancuronium bromide (0.4 mg). After cessation of respiratory movements, a stepwise calibration of VT was performed with a graded 2 ml syringe. At the conclusion of the experiments an injection needle, identical to the injection tube was inserted into the guide cannula, and 0.5/zl Fast blue dye was injected. The brain was removed, frozen on solid CO 2 an kept at -20 *C until sectioned in a cryostat. The injector tip placement was plotted according to the atlas of Paxinos and Watson 27.

Single unit recording After a tracheotomy was performed, the animal was mounted in

a stereotaxic apparatus. The ventral surface of the medulla was exposed by partial removal of the basal part of the occipital bone (4 × 6 ram). The dura was carefully removed and a 5-barrelled micropipette with a diameter of 5-6 g m was lowered by means of a hydraulic microdrive into the region of nLRt, nPGi and nA. The central barrel was filled with 2 M NaCl solution saturated with Fast green and was used for recording action potentials. The in vitro impedance was 3-6 M~2, measured in saline at 135 Hz. One of the four side barrels was filled with 4 M NaCi solution and was used for automatic current balancing. The other side barrels contained 10 mM SP or 10 mM SOM. Both peptides were dissolved in sodium acetate (20 mM, pH 4.5). Single unit potentials were passed through a high input-impedance amplifier and filters. The impulses were discriminated from background noise and fed into a digital counter, which was reset at every 1 or 10 s, and finally displayed on a storage oscilloscope, an audiomonitor and a strip chart recorder. The body temperature of the animals was maintained at 37 °C by means of a heating pad. The tip of the electrode was marked at the end of each experiment by iontophoretic ejection of Fast green. The brain was removed and subjected to conventional histological procedures.

Statistics All data were calculated as means _+ S.E.M. Significant differences were established at the P < 0.05 level using one way analysis of variance followed by t-test or Student's t-test. RESULTS

Antagonistic effects of SOM and SP on ventilation Excitatory effects after SP and inhibitory effects after SOM were seen on ventilation after local administration of the peptides into the medulla oblongata. The experimental findings are summarized in Fig. 1. In the first group of animals (Fig. 1, upper panel, n = 62) SP stimulated ventilation after microinjection (0.75-1.5 nmol) into the ventrolateral medulla, especially in the region of nLRt and nPGi (Figs. 1-4) but also in the region of nA and n. facialis (Fig. 1). A slight but significant stimulation of ventilation after SP administration was also observed in the nTs area of the dorsal medulla (DM) (Fig. 1). Inhibitory ventilatory effects induced by local application of SOM (0.6-1.8 nmol) in the second group of animals (Fig. 1, middle panel, n = 48) were mainly found in the nLRt, nPGi and nA of the VLM, where apnoea occurred within 2-3 min (Figs. 1, 6, 7) after microinjection. The same dose of SOM administered into the n. facialis and nTS induced only a slight or no inhibitory effect (Fig. 1). Local application of 0.75-1.5 nmol SP into the VLM caused increases in Vz (Fig. 3) and central inspiratory drive (VT/TO, as well as a decrease in f due to a lengthening of TE within 3-5 rain (Fig. 4). This ventilatory excitation lasted for 30-40 rain. Administration of SOM into the same injection site during the peak response to SP (Figs. 2-4) caused an almost immediate reversal of SP induced respiratory excitation. Both VT and f returned to the preinjection control level within 2 min. In some animals, a more shallow and faster breathing pattern was also seen, which lasted for about

15 10-20 min before baseline ventilation was resumed. The ventilatory stimulation also reappeared in some animals where the observation period was more than 30-40 min after SP administration (data not shown). In a subgroup of 14 anesthetized rats (Fig. 7), local injection of 0.6-1.8 nmol SOM into the V L M including the nLRi and nPGi caused apnoea after a latency time of approximately 2-3 rain. Usually, depression of ventilation was characterized by a rapid fading of Va-, while f response was much smaller and differed between individuals, i.e. increased transiently followed by a gradual fall or remained at a stable level until apnoea occurred. In 10 rats, where SP (0.75-1.5 nmol) was applied at the same injection site 1-2 min after SOM administration, no animal showed apnoea or even depression of ventilation as seen in the control group (Fig. 6). Furthermore,

SP"

in these SOM pretreated animals, the following SP administration failed to produce any respiratory excitation. Experiments were also designed in the opposite way. As shown in Fig. 7, apnoea occurred within 2 rain after 0.6 nmol SOM injected into the central regions of the nPGi and n L R t (filled circles, n = 5). W h e n 0.6 nmol SOM was administered medial to these nuclei (filled triangles, n = 6), the time to apnoea was considerably longer (mean 6.4 min, no apnoea in one animal) than that after even 1.8 nmol SOM in the central region (0.6 min). A n approximate dose-response curve for the apnoea inducing effect of SOM could be demonstrated. Between 30 and 45 min after the pretreatment with SP (0.75-1.5 nmol) into the V L M , i.e. approaching the end of the SP effect, SOM induced apnoea was only seen after a high dose (1.2 and 1.8 nmol; n -- 4) into the central portion of the V L M nuclei (Fig. 7, open circles) and the latency to apnoea was prolonged. Moreover, the dose-response curve for apnoea induction was markedly shifted to the right in animals injected with S O M into the central part of the nuclei (Fig. 7, open triangles, n = 5) when compared to that obtained from animals without SP pretreatment. When SOM was applied to the same sites as SP in animals pretreated with SP within 30 min, no apnoea was elicited (data not shown). Serial injections of SP and S O M were given in 12 animals. One peptide was given 1-3 min before the other

mmHg BP

8

100[ ~ O" ~JiL~.ii [ ,hi, i!i,L I

,l~li.'.lll!llllll[Jl,

a:~m;,ktZili! hll, lliL~llllllllhllLJlillh !ill li[lllllllllllll

1 nAmb nlJ~t

r~i

n7

Fig. 1. Microinjection of substance P (SP), somatostatin (SOM) or the combined injection of the two peptides into the rat medulla. Injection sites are given in reference to Paxinos and Watson22. Brain sections from the following levels are shown from left to right: interaura1-4.3, -3.3 and -2.3 ram. Each symbol indicates the result from one experiment in one animal. Upper: substance P (SP; 0.75-1.5 nmol). Filled circles indicate sites where VT or l?E increased >30%; hatched circles an increase by <29%; and open circles no effect. Middle: somatostatin (SOM; 0.6-1.8 nmol). Filled triangles indicate sites where apnoea was induced, hatched triangles a decrease in Vv by >30% and open triangles no effect. Lower: somatostatin (SOM; 0.6-1.8 nmol) 7-8 rain after the preceding administration of substance P (SP; 0.75-1.5 nmol). Filled squares indicate sites where SOM reversed the SP induced increase in VT by >30%; hatched squares <29% and open squares no effect. Abbreviations: nTS, nucleus tractus solitarius; nPGi, nucleus paragigantocellularis; nAmb, nucleus ambiguus; n7, nucleus facialis.

A

B

~

- nTs

Fig. 2. Original tracing of blood pressure and respiration during control conditions (A) and 20 rain after local administration of substance P (0.75 nmol) (B) in the VLM. Shown is the inhibitory effect on the ventilatory response by a subsequent injection (indicated by arrow) of somatostatin (0.6 nmol) into the same site. Abbreviations: BP, blood pressure; V-r, tidal volume; nTs, nucleus tractus solitarius; pGi, nucleus paragigantocellularis; VII, nucleus facialis. Time mark = 1 s.

16 TABLE I

Effect of substance P (SP) and somatostatin (SOM) on single unit activity in the medulla oblongata Shown are the results from 137 units tested. + depicts an excitatory, - an inhibitory and 0 no response. Ventrolateral medulla units were located within n. reticularis lateralis and n. paragigantoceilularis lateralis. Dorsal medulla units were located dorsal to these nuclei. The distribution between respiratory and non-respiratory units as well as inspiratory and expiratory units is shown. In units where the actual interaction between the two peptides was tested, the frequency of positive interaction out of the number tested is given by numbers within brackets.

Response pattern SP

Ventrolateral medulla SOM

Dorsal medulla

Respiratory

Non-respiratory

lnsp.

Exp.

Respiratory Insp.

Non-respiratory Exp.

+

-

11 (9/9)

3 (3/3)

9 (5/5)

2* (1/2)

0

-

+

1 (1/1)

1 (0/1)

3 (2/3)

0

1"

3 (1/2) 0

+

+

0

0

0

0

0

0

-

-

1

0

1

1"*

0

3

+ 0

0 +

3 0

0 0

2 1

4 0

0 0

6 4

-

0

0

0

0

2**

0

0

0 0

0

2 0

2 2

4 25

2** + 3* 5**

2* 0

7 21

18

8

45

3

44

Sum

19

* Indicates units located within the n. ambiguus and ** units located within the n. tractus solitarius.

into the same injection sites as described above. A functional antagonism was only found between SP and SOM in the VLM, mainly in the nPGI and nLRt (Fig. 1, bottom panel).

2.6

.

I->

.

.

.

*

*

.

e

,; ..........

0.25

E "-" 6 . 0

1.8

o II-

I,.-

5.0

1.4

"I 1.0

0.30

'~ 7.0

n=lO

.....

,

2.2 v

*

8.0

~

soM

SP

4.0 I

70 E "~ ¢.. 60

I

0.20

I

I

I 0.15

1.2

50

0 W

k40

0.8

®

E 100

0.4

.~ ..........

0

E W >

6

r-

80 60

OI 0

I 5

I 10 TIME

I 15

I 20

( min )

Fig. 3. Tidal volume ( V T ) , respiratory frequency 0") and minute volume (I2~) after substance P (SP) alone (circles, 0.75 nmol, n = 10) or SP followed by somatostatin (SOM) administration (squares 0.6 nmol, n = 12) into the same brain site (ventrolaterai medulla oblongata). Peptide injections indicated by vertical arrows. Shown are the means + S.E.M. from 10 or 12 animals. Significance indicated versus respective baseline control value; *P < 0.05.

8P i 0

480M

I 4

I 8

Time

I 12

J 16

(rain)

Fig. 4. Effects of substance P (SP; 0.75 nmol) and the subsequent administration of somatostatin (SOM; 0.6 nmol) on inspiratory time (TI), expiratory time (TE) (lower), 'inspiratory drive' (Va4TI) and 'respiratory duty cycle' (TI/TToT) (upper). Timepoints for peptide injections indicated by vertical arrows. Shown are the means + S.E.M. of experiments from 12 animals where SP and SOM was administered into the VLM.

17 140 120 100 80 I-.

>

-'~ _ ~

n=lO

~

±

•

±

±

~

~

,

~

60 40 20

0

.,~

75

¢:

E m .1: @ 1.

1001-

,

t [

~

1

,

vt.

120 100

80 60 40

20 0

roL

SOM

SP I

I

0

5

I:

I

10 Time

130 !20 100 E 80 60 Ill 40 > 20

I

]

15

20

(min)

Fig. 5. Tidal volume (VT), respiratory frequency (f) and minute volume (f'E) after substance P (SP; 0.75 nmol) followed by somatostatin (SOM; 0.6 nmol) into the same brain site (dorsal medulla oblongata). Timepoints for peptide injections indicated by vertical arrows. Shown are the means + S.E.M. from 5 animals. Significance indicated versus respective baseline control value; *P < 0.05.

A less pronounced but gradual increase in VT was induced by local injection of SP into the DM (Fig. 5). This effect was not antagonized by SOM (Fig. 5).

Effect of SOM and SP on single neuron activity in the medulla oblongata Extracellular single unit recordings (n = 138 cells) were performed before and after microiontophoretic application of SOM and/or SP (Table I). Among 71 units tested in the nLRt and nPGi of the VLM, 26 were respiratory-related units. The majority of these units (92%) responded to either or both of the peptides. An excitatory response to SP as well as an inhibitory response to SOM was seen in 54% of the respiratoryrelated units. A direct interaction could be demonstrated in 13 out of 14 units tested, i.e. the SP or SOM effects could be reversed by concomitant administration of the other peptide (Fig. 8). Of the 45 non-respiratory units, 25 (56%) were not sensitive to either SP or SOM. A minor part of these, 9 units, reacted with an excitation to SP and inhibition to SOM, and in 5 of the units, a direct interaction could be demonstrated between SP and SOM. Some of the non-respiratory units responded to SP and SOM in the opposite way, and 7 were only sensitive to either peptide (Table I). A total of 67 units were tested in a region dorsal to the

m

A

SOM I

I

i

r

I

I

I

I

I

I

I

I

0

1

2

3 4 Time

5

6

7

8

9

10

11

(min)

Fig. 6. Effects of substance P (SP) on somatostatin (SOM) induced ventilatory depression. Shown is mean + S.E.M. of tidal volume (VT), respiratory frequency (D and minute volume (f'E) in 14 animals after SOM (0.6 nmol) and in 10 animals where SP (0.75 nmol) was given within the first minute after the preceding administration of SOM (0.6 nmol) into the same brain site (ventrolateral medulla oblongata). Timepoints for peptide injections indicated by vertical arrows. Injection sites for SOM alone (filled circles) are shown in the lower left part of the figure and for SOM and SP (filled squares) in the lower right part.

nLRt and nPGi (Table I). Of these, 22 were respiratoryrelated units and located in the nA and nTS. Only 2 units, both located in the nA, were excited and inhibited following application of SP and SOM, respectively. A direct interaction between the peptides could only be shown in one unit. Half of the respiratory related units in the nA and nTS (11 neurons) responded only to one of the peptides, i.e. inhibition by SOM or excitation by SP (Fig. 9). The other half showed a complex response or no response at all. Of the 44 non-respiratory units outside the nLRt and nPGi, only 3 were excited by SP and inhibited by SOM. The major part was unresponsive or responded to only one of the peptides.

18

U ¢)

No spnea induced > 10 rain a f t e ¢ 8OM 10

~-

\

o:

,

--

........

I

o

100sec

4

! -o

J

0L

SP SaM

~ a

1°°I

z~

, 2

-Arnb

|

•

8

/,

I,--

0

0

• 0,6

I 1.2 Ooae of 80M

~ 1.8

I 2.4

J 3.0

~, 2O

(nmol)

Fig. 7. Time to apnoea after local administration of somatostatin (SOM; 0.6-3.0 nmol) into the ventrolateral medulla oblongata. Injection sites are shown in the brainstem sections (interaura1-4.3 and -3.3 mm). Filled symbols indicate injections of SOM made during control conditions and open symbols injections made 30-45 min after the preceding microinjection of SP (0.75-1.5 nmol). Circles indicate injections made within the central region of the nuclei. Borderline injections medial to these nuclei are indicated by triangles. Superimposed lines represent a rough estimation of the dose-response relationship for SOM-induced apnoea with or without SP.

o~

(2.

0 SP SOM

~nTS

' 100see

Fig. 9. Single unit recordings from an inspiratory-related unit in the n. ambiguus (Amb; upper tracing) and a non-respiratory-related unit in the n. tractus solitarius region (nTS; lower tracing). Shown is the integrated response (spikes/10 s) after microiontophoretic application of substance P (SP) and/or somatostatin (SOM). Peptide administration indicated by vertical bars below the integrated response tracing. Ejection current of both peptides was 10 hA. Note the excitatory response to SP in the nTS unit where SOM is ineffective and the inhibitory response to SOM in the n. ambiguus unit where SP is ineffective. There was no antagonistic effect when combined administration of the peptides was given.

\

SP

SOM

~ - n P G i

loosec

DISCUSSION

y~

Both SP ~ 6 and S O M 7'12'14'15'18"30 s e e m to play an

\

SOM SP loosec

Fig. 8. Single unit recordings from neurons in the ventrolateral part of the medulla. Shown is the integrated response (spikes/s) after microiontophoretic application of substance P (SP) and somatostatin (SOM) to an inspiratory-related unit in the n. paragigantocellularis lateralis (nPGi; upper tracing) and a non-respiratoryrelated unit in the n. reticularis lateralis (nLR; lower tracing). Peptide administration indicated by vertical bars below the integrated response tracing. Ejection current of both peptides was 10 nA. Note the excitatory action of SP and the inhibitory action of SOM, as well as the antagonistic effect of SP on the SOM-induced inhibitory action in both units.

important role in the regulation of respiration at the level of the medulla oblongata. In a g r e e m e n t with our previous studies, SP induced a p o t e n t stimulatory effect on ventilation after microinjections in the ventral portion of the m e d u l l a o b l o n g a t a (n. paragigantocellularis, n. reticularis lateralis, n. ambiguus) as well as into the dorsal region of the m e d u l l a o b i o n g a t a (n. tractus solitarius) 4'5. Iontophoretically a p p l i e d SP has been shown by MorinSurun et al. to induce excitatory effects in respiration related neurons in the n. tractus solitarius region of the cat 24. Both the D M and the V L M are intimately involved in the regulation o f respiration by m o d u l a t i o n o f afferent input as well as inspiratory and expiratory activity and contain an a b u n d a n c e of cell bodies with activity associated with the respiratory cycle 1°. M o r e o v e r , a dense network of SP containing nerve cell bodies and fibers has

19 been identified in respiration related areas of the DM and VLM by radioimmunoassay and immunohistochemical mapping 1,s,13. SOM appears to exert potent inhibitory actions on ventilation as we 7, in support of the findings by Yamamoto and coworkers 3°, have shown that the peptide readily induced apnoea after microinjection into the VLM. The SOM induced apnoea in the VLM was due to a depression of inspiratory drive, while the respiratory timing remained constant. Administration into the dorsal medulla (DM), was, in contrast, ineffective in eliciting respiratory effects7. SOM exhibits a widespread anatomical distribution throughout the central nervous system n. Moreover, there is an abundant distribution of SOM cell bodies and nerve terminals within the DM and the region along the lateral ventral surface of the medulla has recently been shown to contain SOM-like immunoreactivity22. While the DM, in particular the neural population of the nucleus of the solitary tract, may be involved in different respiratory reflexes and control mechanisms influencing the pattern of breathing 1°, the VLM, in particular the nPGi region, has been suggested to be an important site for the integration of ventilatory drive inputs 1°. Focal reversible cooling in the nPGi and n. preolivaris region of the cat is a potent apnoea inducing stimulus, while in the nucleus of the solitary tract of the DM reversible cooling or chemical lesions mainly influenced the timing of the respiratory cycle2,23. In view of the pronounced effects of SP and SOM on the respiratory drive, we therefore considered it of particular interest to focus extracellular single unit recording studies to the VLM region. Not only SOM, but also the substance P antagonist, [D-Pro2,D-Trp7-9]Sp, induced apnoea in the spontaneously breathing halothane anesthetized rat after microinjection into the VLM 6. Although there is some controversy regarding the specificity of this antagonist for central nervous system SP receptors 24, this finding may have several implications. Not only may a tonic SP influence play a modulatory role in the maintenance of respiratory drive, but SP and SOM may in fact have a reciprocal influence on respiratory drive mechanisms by effects elicited in the VLM. In the present microinjection studies both peptides were found to exclusively affect ventilatory drive in different directions. The increased ventilatory drive after SP was rapidly reversed by SOM in the VLM. A SOM dose of 0.6 nmol in the unpretreated animal induced apnoea in 8/11 cases. After SP, however, this dose reestablished ventilation to the pre-SP level within 3-5 min but failed to induce apnoea. These findings may, however, reflect a pharmacological rather than a physiological interaction in view of the opposite

effects induced on ventilation by SP and SOM. In order to further study this possibility extracellular single unit recordings were employed involving microiontophoretic application of the peptides to respiratory-related units in the VLM. The vast majority of the respiratory-related units in the nLRt and nPGi of the VLM showed a reciprocal response to the two peptides. In fact there was a far higher proportion of respiratory than non-respiratory units responding to either or both of the peptides, indicating a certain selectivity in the action of these peptides to mechanisms associated with the regulation of respiration. Moreover, there was a proportionally higher homogeneity of neuronal responses within the nLRt and nPGi compared to single units located dorsal to these nuclei. Whenever there was a response to SOM, this was almost exclusively of inhibitory character. Thus, the present findings strongly indicate a direct antagonistic action of SP and SOM in respiratory regulating mechanisms in the VLM. SOM has in previous electrophysiological studies been shown to induce both excitatory25 and inhibitory actions z6 after iontophoretic application in different areas of the brain. In the medulla oblongata, SOM has, however, been shown to induce a hyperpolarization of vagal motor nucleus neurons 26. Only a small part of the units studied in the present study, mainly non-respiratory units, showed an excitatory response pattern to SOM indicating that the main neurophysiological effects of the peptide are indeed inhibitory. The SP antagonist, [o-Pro2,D-TrpT-9]Sp, was not applied to neurons of the VLM in the present electrophysiological studies. Morin-Surun and coworkers 24 failed to show any antagonistic effect of this SP analogue or [D-Pro4,D-Trp7'9'10]Sp on respiratory related neurons in the n. tractus solitarius in artificially ventilated cats. However, there appears to be conflicting findings after administration of these SP analogues into other brain regions as well. Engberg et al. 9 could block SP induced excitation in locus coeruleus neurons in vivo while this was not the case in an in vitro study by Cheeseman and coworkers 3. Obviously, although the specificity of the antagonist may be questioned, these discrepant findings may also relate to species differences, to the brain region studied or to whether the SP neuronal pool is excited or not. If central SP activity is inhibited, the apnoea seen after microinjection of the SP antagonist may simply be due to an attenuation of tonic SP activity in the VLM. Alternatively, however, in view of the interactive effects of SP and SOM, the apnoea after [D-Pro2,t~-TrpT-9]Sp may relate to an increased release of the inhibitory transmitter, i.e. an increased endogenous SOM activity. Further studies involving the administration of a SOM antagonist are called upon to clarify these mechanisms. Both SP and SOM have been shown to coexist with

20 other neuropeptides or classical neurotransmitters in the medulla oblongata. The colocalization reported for SP with epinephrine 21 or serotonin 16,29 and T R H 17 and for SOM with enkephalin-like immunoreactivity z2, raises the possibility that local modulation of the activity in these neuronal systems may occur at the synaptic level which ultimately influences the expression of modified SP and SOM activity on respiratory regulation. The ventilatory stimulation after microinjection of SP as well as the reversal of this effect and the induction of apnoea after microinjected SOM appeared to have a latency of between 1 and 5 min. In the single unit recording studies, however, effects appeared momentarily after administration of the peptides. This delay in the pharmacological response after microinjections deserves further comment. In view of the injection volumes used (usually 0.5 but up to 1.5/A), the peptides would be expected to rapidly diffuse within the target area. A concentration gradient is created and the biological response is presumably elicited as a local threshold concentration of the peptide is reached. A delayed response would therefore indicate that either a relatively widespread population of receptors has to be activated before the particular response is elicited or that the local pharmacokinetic or pharmacodynamic factors may constitute a limiting factor for the neuropeptide to elicit a response. The latter explanation is supported by the observation that the apnoea elicited by microinjection of [D-PrOe,D-TrpT-9]Sp, has a considerably shorter latency compared to that seen after SOM 6. Moreover, specific effects on ventilation may be induced by neuropeptides in well demarcated areas of the medulla oblongata whereas injections into immediately adjacent areas may be devoid of effects 4'7. The latency of the SOM induced apnoea appears to relate to the ventilatory drive input. Ventilation during combined hypoxic and hypercapnic conditions has been reported to shorten the latency to apnoea onset after

intracisternally applied SOM ~5. In view of the present results with reciprocal effects between the two peptides, a shortening of the latency may be due to increased SOM or decreased SP turnover in the V L M during hypoxia and hypercapnia. However, recent experiments employing the microdialysis technique have shown that hypoxia constitutes a potent stimulus to increase SP-like immunoreactivity in the D M of artificially ventilated cats 2°. Indeed, SP microinjected into the V L M or locally applied to the ventral brainstem surface appears to be an important modulator of the hypercapnic and hypoxic ventilatory responses in the spontaneously ventilating halothane anesthetized rat 5. Neuromodulatory systems in the V L M with a possible direct influence on ventilatory drive induced by either hypoxia or hypercapnia are of specific clinical interest. Hypoventilation and apnoea are in themselves potent stimuli to activate ventilatory drive, The ventilatory drive may consequently be regulated by the delicate balance between excitatory and inhibitory neuronal modulators. If the activity of such modulators is sensitive to factors such as hypoxia or hypercapnia, they may constitute the basis for pathologic changes in different hypoventilation states. Although this is an interesting perspective in the pathogenesis of such disorders, there are at present no studies evaluating the SP or SOM turnover in patients with chronic hypoventilation or apnoea. In conclusion, we suggest that SP and SOM exert physiologically antagonistic effects on respiratory regulation in the VLM. These two peptidergic systems may have a direct reciprocal physiologic influence on ventilatory drive mechanisms as judged by their actions in the nLRt and nPGi of the rat.

REFERENCES

6 Chen, Z., Hedner, T. and Hedner, J., Hypoventilation and apnea induced by the substance P antagonist [I~PrO2,D-TrpT-9]-SP in the ventrolateral rat medulla, Acta Physiol. Scand., 134 (1988) 153-154. 7 Chen, Z., Hedner, T. and Hedner, J., Local appfication of somatostatin in the ventrolateral brain medulla induces apnea in the rat, J. Appl. Physiol., in press. 8 Cooper, P.E., Fernstrom, M.H., Rorstad, O.P., Leeman, S.E. and Martin, J.B., The regional distribution of somatostatin, substance P and neurotensin in human brain, Brain Research, 218 (1981) 219-232. 9 Engberg, G., Svensson, T.H., Rosell, S. and Folkers, K., A synthetic peptide antagonist of substance P, Nature, 293 (1981) 222-223. 10 Euler, C. von, Brainstem mechanisms for generation and control of the breathing pattern. In N.S. Cherniack and J.G. Widdicombe (Eds.), Handbook of Physiology, The Respiratory System, Vol. 12, Control of Breathing, Bethesda, MD, Physiol. Soc.,

1 Brownstein, M.J., Mroz, E.A., Kizer, E.A., Palkovits, M. and Leeman, S.E., Regional distribution of substance P in the brain of the rat, Brain Research, 116 (1976) 299-305. 2 Budzinska, K., Euler C. yon, Kao, EE, Pantaleo T. and Yamamoto, Y., Effect of graded focal cold block in rostral areas of the medulla, Acta Physiol. Scand., 124 (1985) 329-340. 3 Cheeseman, H.E, Pinnock, R.D. and Henderson, G., Substance P excitation of rat locus coeruleus neurones, Eur. J. Pharmacol., 94 (1983) 93-99. 4 Chen, Z., Hedner, J. and Hedner, T., Local effects of substance P on respiratory regulation in the rat medulla oblongata, J. Appl. Physiol., 68 (1989) 693-699. 5 Chen, Z., Hedner, J. and Hedner, T., Substance P in the ventrolateral medulla oblongata regulates hypoxic, hypercapnic and somatic ventilatory responses in the rat, J. Appl. Physiol., 68 (1990) 2631-2639.

Acknowledgements. This project was supported by the Swedish Medical Research Council (proj nos. 2464, 2862, 7484 and 8642). Dr. Z. Chen was supported by the Swedish Institute and the Swedish Foundation for Clinical Pharmacology.

21 1986, pp. 1-67. 11 Finley, J.C.W., Maderdrut, J.L., Roger, L.J. and Pertrusz, P., The immunocytochemical localization of somatostatin-containing neurons in the rat central nervous system, Neuroscience, 6 (1981) 2173-2192. 12 Fuxe, K., Agnati, L.E, Hirfstrand, A., Mutt, V., Andersson, K., H6kfelt, T., Vale, W., Brown, M. and Rivier, J., Cardiovascular and respiratory actions of somatostatin peptides following intracisternal injections into the a-chloralose anaesthetized rat, Neurosci. Lett., Suppl. 10 (1982) 189. 13 Gingras, J.L., Brunner, S.L. and McNamara, M.C., Developmental characteristics of substance P immunoreactivity within specific rabbit brainstem nuclei, Regul. Peptides, 23 (1988) 183-192. 14 Hiirfstrand, A., Fuxe, K., Kalia, M. and Agnati, F., Somatostatin induced apnoea: prevention by central and peripheral administration of the opiate receptor blocking agent naloxone, Acta Physiol. Scand., 125 (1985) 91-95. 15 Hiirfstrand, A., Kalia, M., Fuxe, K., Kaijser, L. and Agnati, L.F., Somatostatin-induced apnoea: interaction with hypoxia and hypercapnia in the rat, Neurosci. Lett., 50 (1984) 37-42. 16 H6kfelt, T., Holets, V.R., Staines, W., Meister, B., Melander, T., Schalling, M., Schultzberg, M., Freedman, J., Bj6rklund, H., Olson, L., Lindh, G., Elfvin, L.-G., Lundberg, J.M., Lindgren, J./~., Samuelsson, B., Pernow, B., Terenius, L., Post, C., Everitt, B. and Goldstein, M., Coexistence of neuronal messengers - - an overview. In T. HOkfelt, K. Fuxe and B. Pernow (Eds.), Progress in Brain Research, Vol. 68, Elsevier, Amsterdam, 1986, pp. 33-70. 17 Johansson, O., HOkfelt, T., Pernow, B., Jeffcoate, S.L., White, S., Steinbush, H.W.M., Verhofstad, A.A.J., Emson, P.C. and Spindel, E., Immunohistochemical support for three putative transmitters in one neuron: coe~dstence of 5-hydroxytryptamine-, substance P-, and thyrotropin releasing hormone-like immunoreactivity in medullary neurons projecting to the spinal cord, Neuroscience, 6 (1981) 1857-1881. 18 Kalia, M., Fuxe, K., Agnati, L.E, HSkfelt, T. and Hiirfstrand, A., Somatostatin produces apnoea and is localized in medullary respiratory nucleus: a possible role in apnoic syndromes, Brain Research, 236 (1984) 339-344. 19 Leibstein, A.G., Dermitzel, R., Willenberg, I.M. and Pauschert, R., Mapping of different neuropeptides in the lower

brainstem of the rat: with special reference to the ventral surface, J. Auton. Nerv. Syst., 14 (1985) 299-313. 20 Lindefors, N., Yamamoto, Y., Pantaleo, T., Lagercrantz, H., Brodin, E. and Ungerstedt, U., In vivo release of substance P in the nucleus tractus solitarius increases during hypoxia, Neurosci. Lea., 69 (1986) 94-97. 21 Lorenz, R.G., Saper, C.G., Wong, D.L., Ciaranello, R.D. and Loewy, A.D., Co-localization of substance P- and phenylethanolamine-N-methyltransferase-like immunoreactivity in neurons of ventrolateral medulla that project to the spinal cord: potential role in control of vasomotor tone, Neurosci. Lett., 55 (1985) 255-260. 22 Millhorn, D.E., H6kfelt, T., Terenius, L., Buchan, A. and Brown, J.C., Somatostatin- and enkephalin-like immunoreactivities are frequently colocalized in neurons in the caudal brain stem of rat, Exp. Brain Res., 67 (1987) 420-428. 23 Morin-Surun, M.P., Champagnat, J., Boudinot, E. and DenavitSaubie, M., Differentiation of two respiratory areas in the cat medulla using kainic acid, Respir. Physiol., 58 (1984) 323-334. 24 Morin-Surun, M.P., Jordan, D., Champagnat, J., Spyer, K.M. and Denavit-Saubie, M., Excitatory effects of iontophoretically applied substance P on neurons in the nucleus tractus solitarius of the cat: lack of interaction with opiates and opioids, Brain Research, 307 (1984) 388-392. 25 Olpe, H.R., Balcar, V.J., Bittiger, H., Rink, H. and Sieber, P., Central actions of somatostatin, Eur. J. Pharmacol., 63 (1981) 127-133. 26 Oomura, Y. and Mizuno, Y., Effect of somatostatin on the vagal motor neuron in the rat, Brain Res. Bull., 17 (1986) 397-401. 27 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic, New York, 1982. 28 Pittman, Q.J. and Siggins, G.R., Somatostatin hyperpolarizes hippocampai pyramidal cells in vitro, Brain Research, 221 (1981) 402-408. 29 Thor, K.B. and Helke, C.J., Serotonin and substance P colocalization in medullary projections to the nucleus tractus solitarius: dual-color immunohistochemistry combined with retrograde tracing, J. Chem. Neuroanat., 2 (1989) 139-148. 30 Yamamoto, Y., Runoid, M., Prabhakar, M., Pantaleo, T. and Lagercrantz, H., Somatostatin in the control of respiration, Acta Physiol. Scand., 134 (1988) 529-533.