European Journal of Pharmacology, 35 (1976) 349--360 © North-Holland Publishing Company, Amsterdam - - Printed in The Netherlands

RELEASE OF SMOOTH MUSCLE-CONTRACTING SUBSTANCES FROM ISOLATED PERFUSED LUNGS V A L E R I E A. A L A B A S T E R

and Y.S. B A K H L E

Department of Pharmacology, Instituteof Basic Medical Sciences, Royal College of Surgeons of England, Lincoln's Inn Fields, London, W C 2 A 3PN, U.K. Received 26 June 1975, revised MS received 1 October 1975, accepted 10 October 1975

V.A. A L A B A S T E R and Y.S. BAKHLE, Release of smooth muscle-contracting substances from isolated perfused lungs, European J. Pharmacol. 35 (1976) 349--360. Infusion of tryptamine (1--4 pg/ml) through the pulmonary circulation of rat isolated lung perfused with Krebs solution caused release of a mixture of spasmogens contracting isolated smooth muscle preparations. One component of this mixture had biological activity comparable to E-type prostaglandins. Other components included a slow reacting substance comparable to SRS-A and a rabbit aorta-contracting substance comparable to RCS. Infusions of 5-bydroxytryptamine, acetylcholine and histamine (0.5--2/~g/ml) also caused release. Release induced by the tryptamines but not that by acetylcholine and histamine was prevented by methysergide whereas acetylcholine-induced release was prevented by hyoscine which did not affect tryptamine-induced release. The tryptamines and histamine released spasmogens from dog isolated lungs but only histamine was effective in guinea-pig lungs. We conclude that amine-induced release from isolated lungs is a fairly general phenomenon and that it may represent an endocrine function of lung. Slow reacting substance Pulmonary circulation

Vasoactive amines Isolated lungs

Spasmogen release

Prostaglandins

1. Introduction

2. Materials and methods

During a study of the inactivation of tryptamine in the pulmonary circulation of rat isolated lungs (Alabaster, 1971), biologically active material other than tryptamine was detected in the effluent from lung during infusions of the amine through the pulmonary circulation. We have attempted to identify and quantitate some components of this material and have investigated the ability of 5-hydroxytryptamine (5-HT), histamine and acetylcholine as well as tryptamine to cause release of smooth muscle contracting substances from the isolated lungs of rats, guinea pigs and dogs. Some of these results were demonstrated to the British Pharmacological Society (Alabaster and Bakhle, 1970).

2. I. Isolated lungs and assay tissues

Isolated lungs of the rat, guinea pig and dog perfused with oxygenated Krebs bicarbonate solution at 37 ° C, were prepared as described previously (Bakhle et al.,1969). The perfusion rate was 8 ml/min for rat and guinea-pig lungs and 10 ml/min for lobes of dog lung. The effluent from the lungs supeffused a seriesof up to six assay tissues chosen to detect 5.HT, acetylcholine,prostaglandins (PGs), slow react~ ing substance (SRS), rabbit aorta contracting substance (RCS), angiotensin, bradykinin and catecholamines. The following assay tissues were used: rat stomach strip (Vane, 1957), rat colon (Regoli and Vane, 1964), guinea-pig

350 ileum (Brocklehurst, 1960), chick rectum (Mann and West, 1950), rabbit aorta spiral strip (Furchgott and Bhadrakom, 1953), cat jejunum (Ferreira and Vane, 1967a), cat terminal ileum (Ferreira et al., 1973}. To increase the specificity of the assay tissues, one or more antagonists were infused at 0.1 ml/min over some or all of the assay tissues. The antagonists used and their final concentrations (as base) in the Krebs solution were: methysergide, 200 ng/ml, mepyramine 200 ng/ml, hyoscine 100 ng/ml, phenoxybenzamine 100 ng/ml and propranolol 2/~g/ml. Contractions were recorded by auxotonic levers attached to Harvard isotonic transducers, and displayed on six-channel Watanabe recorders (type WTR 281). Mean pressure in the cannula in the pulmonary artery was detected by a pressure transducer (SE Laboratories) attached to a side arm of the cannula and displayed on the recorder.

2. 2. Extraction procedure and bioassay of prostaglandin-like activity Perfusate from isolated rat lungs was collected during, and for 2 min after, an infusion of laTptamine or 5-HT into the pulmonary circulation, either directly from the lungs or after passing over the assay tissues. The perfusate was acidified to pH 3 with N/3 hydrochloric acid and immediately extracted twice with equal volumes of ethyl acetate (Gilmore et al., 1968). The combined ethyl acetate phases were evaporated to dryness at 50°C under reduced pressure. The residue was dissolved in saline and assayed on rat stomach strip, rat colon and chick rectum superfused in series. Perfusate was also collected during the infusion of k n o w n concentrations of PGE2 to the assay tissues. Extraction and assay of this perfusate gave a 60--70% recovery of the prostaglandin.

2. 3. Drugs The following drugs were used: acetylcholine perchlorate (BDH), angiotensin II amide (Hypertensin, Ciba), bradykinin (a gift from

V.A. ALABASTER, Y.S. BAKHLE Dr. H.O.J. Collier), histamine acid phosphate (BDH), histamine liberator-- c o m p o u n d 48/80 (Wellcome Research Laboratories}, hyoscine h y d r o b r o m i d e (BDH), 5-hydroxytryptamine creatinine sulphate (May and Baker), mepyramine maleate (May and Baker), methysergide bimaleate (Sandoz), (--)-noradrenaline (BDH), phenoxybenzamine hydrochloride (SKF), propranolol hydrochloride (ICI), prostagiandin El, E2, and F2~ (gifts from Dr. J. Pike, Upjohn), tryptamine hydrochloride (BDH). All drugs were dissolved in saline :(0.9% NaC1 w/v) except phenoxybenzamine which was dissolved in a small volume of ethanol and then diluted with saline. Doses are given as the weight of the base/ml and refer to the final concentration of the substances.

3. Results

The basic observation of tryptamine-induced release is illustrated in fig. 1. I t s h o w s the responses of t w o rabbit aortic strips, superfused with the effluent from rat lungs, the lower strip being treated with a constant infusion of methysergide (100 ng/ml). This concentration of methysergide attenuates the response of the lower strip to tryptamine (0.25, 0.5 and 1.0 pg/ml) infused directly. The first infusion of tryptamine (1.0 pg/ml) through the lung causes a contraction of the upper strip smaller than that caused by the same concentration infused directly, whereas on the lower strip the contraction to the infusion through the lung is greater than that to the direct infusion. This result is indicative of the release of a smooth muscle-contracting material other than tryptamine from the lung during the infusion of tryptamine through the lung. The release was induced by infusions of tryptamine at concentrations of 1--4/ag/ml and a concentration of 2 pg/ml was selected as the standard infusion. The nature of the contractor substance(s) released by tryptamine from rat lungs was investigated further using assay tissues treated with the combined antagonists (see Materials

AMINE-INDUCED

SPASMOGEN

RELEASE

FROM

LUNGS

351

~0m,p

RbA

RbA + METI'ffSF-RC,IOIE

-

-

iii

--:

p ~

p~,~

0~syg 0.syg

U'g

Fig. 1. Removal of tryptamine and large doses of 5-HT in rat lungs showing release of a rabbit aorta contracting substance. The upper records show" contractions of 2 rabbit aortas (RbA), one untreated and one treated with methysergide, superfused in series with effluent from rat lung perfused with Krebs solution. 5-HT or tryptamine (T) were infused directly to the tissues (DIR) or into the pulmonary artery cannula (AL). 8 8 % of an infusion of 5-HT (50 ng/ml) and 3 0 % of an infusion of tryptamine (1 ~g/ml) were removed by the lung. However, these values do not reflect the true degree of removal since the infusions released a substance from the lung which contracted the methysergide-treated rabbit aorta. The perfusion pressure was increased during the A L infusions of 5-HT and tryptamine.

TABLE 1 Effects of vasoactive amines on rat isolated lungs; release of PO-like activity and increase in perfusion pressure. Amine (concentration o f base)

Tryptamine (2/~g/ml) 5-Hydroxytryptamine (0.5 ~g/ml) Histamine (2/~g/ml) Acetylcholine (0.5 pg/ml)

PG-like activity released as PGE2 (ng equivalents/ml) RSS*

CR

Perfusion pressure increase (ram Hg)

No. of expts.

1.4 -+ 0.25**

0.7 + 0.13

11.8 -+ 1.3

25

0.5 -+ 0.08

0.3 -+ 0.07

10.2 + 1.6

9

1.0-+ 0.46

***

8

-+ 2.8

6

1.0 -+ 0.14

0.7 -+ 0.09

2

-+ 0.6

12

* RSS, CR = rat s t o m a c h strip, chick rectum. The PG-like activity did n o t assay identically o n these two tissues (see text). ** Mean -+ S.E. *** The chick r e c t u m c o n t r a c t e d to this c o n c e n t r a t i o n of histamine in the presence o f antagonist and could not be used to assay released spasmogens.

352

V.A. ALABASTER, Y.S. BAKHLE ,lo_~ J

PERFUSlON PRESSURE

,

-

} ~ '

,

I I I

I I

RNS." , I I

~ :

- •

~m~ o Hg

q t

'i

'

I i I

i I I

RC. "

I

I5¢m

T p e r ml.

AL

~ ,, ~

~

4~

4~

Fig. 2. Release of active substances by tryptamine in rat isolated lungs. The top record shows lung perfusion pressure and in descending order the other records show contractions of a rat stomach strip (RSS), a rat colon (RC), a chick rectum (CR) and a guinea-pig ileum (GPI) superfused in series with the effluent from lungs perfused with Krebs solution. The assay tissues were treated with combined antagonists as indicated in this and other figures by an asterisk*. Infusions of tryptamine (T) given directly to the tissues had no effect but the same infusion given into the pulmonary arterial cannula (at arrows) released a prostaglandin-like substance and an SRS from the lungs. Repeated tryptamine infusions through the lungs released smaller amounts of these substances and caused smaller increases in perfusion pressure. The contractions of the assay tissues were compared with responses produced by infusions of prostag!,andin E2 (PGE2), prostaglandin F2a (PGF2a), angiotensin (Ang) and histamine (H) infused directly.

mine released a substance from rat lungs which cont ract ed these three tissues (fig. 2). The relative sizes of the c o n t r a c t i o n indicated the PG-like material released was mainly of the E-type, for the rat colon is more sensitive to PGF2 ~ than PGE2 while the rat stomach strip is more sensitive to PGE2 (fig. 2). The a m o u n t of PG (in terms of E2 ) released was greater as measured by the rat stomach strip (1.4 ng/ml) than that measured by the chick rectum (0.7 ng/ml) (table 1, fig. 2). These; results also suggested t h a t substances ot her than PGs were released by the lungs, and c o n t r i b u t e d t o the c o n t r a c t i o n of the rat stomach strip but n o t to t h a t o f the chick rectum. This was confi rm ed by ext ract i on in 3 experiments. The lung perfusate was collected during a 4 min infusion of t r y p t a m i n e and for 2 min after, and the perfusate e x t r a c t e d with acidified et hyl acetate. The e x t r a c t e d material was taken up in saline and assayed on the rat stomach strip, chick r e c t u m and rat colon. The responses o f these tissues suggested the presence of mainly E-type PG. The activity measured by the rat stomach strip and the chick rect um before and after extraction is shown in table 2. Allowing for the loss of activity during the ext ract i on procedure, only 7 0 - 8 0 % of the original c o n t r a c t o r activity measured by the rat stomach strip could be recovered while 80-100% of t hat measured by the chick rect um was recovered. Thus the a m o u n t of PG-like material released is more reliably measured by assay on the chick rectum.

3.1.2. SRS-like activity and methods). These antagonists made the tissues insensitive t o the tryptamines, histamine, acetylcholine and catecholamines. The contractions o f the assay tissues indicated release o f substances with the biological activity o f PGs, SRS and RCS (see fig. 2).

3.1.1. Prostaglandin-like activity In th e presence o f c o m b i n e d antagonists, the rat stomach strip, rat colon and chick r e c t um are selectively c o n t r a c t e d by PGs and trypta-

T r y p t a m i n e also released f r o m rat lungs a substance which c o n t r a c t e d the guinea-pig ileum treated with the com bi ned antagonigts (which include m epyram i ne) (fig. 2). The guinea-pig ileum, which was no longer responding to PGE2, started to c o n t r a c t 2--3 min after t r y p t a m i n e infusion to the lungs and reached a peak 1--2 min after the end of the 4 rain infusion. The ileum then relaxed, but so slowly that in m ost experiments relaxation was still incomplete at the end of the experiment. A purified preparation of SRS was n o t available

AMINE-INDUCED

SPASMOGEN

RELEASE

FROM

LUNGS

TABLE 2 Assay of PG-like activity in lung perfusate before and after extraction.

The perfusate from 3 separate rat lungs during release of PG-like activity in response to an infusion of tryptamine (2 p.g/ml) was collected. The amount of PG-like activity was assayed before and after extraction on rat stomach strip, rat colon and chick rectum superfused in series. Experiment number

% recovery of PGE2 standard

A m o u n t of activity (ng PGE2 equivalents) in perfusate as measured by Rat stomach strip

Chick rectum

Before

After* Before After extrac- extrac- extrac- extraction tion tion tion

1 2 3

70 70 62

6.4 25.6 7.5

4.3 21.4 5.5

16 5

15 5

* Corrected for recovery of P G E 2 standard.

to use as a standard, so the a m o u n t of SRS released in these experiments could n o t be quantitated.

3.1.3. Rabbit aorta contracting substance (RCS) A substance (RCS) released from guinea-pig lungs during anaphylaxis which contracts the rabbit aorta in the presence of combined antagonists, is unstable and is unlike any previously described substance in terms of its biological activity (Piper and Vane, 1969). A similar substance was released from rat lungs during infusions of tryptamine (fig. 1 : 9 expts.). The contraction of the rabbit aorta was rapid in onset and in some experiments n o t maintained for the duration of the tryptamine infusion. When the active lung perfusate was collected and retested on the aortic strips 20 min later, no contraction was produced. As there was no stable standard preparation of RCS it was difficult to quantitate release or to calibrate the rabbit aortas for their sensitivity to RCS before the experiment. It was not possible, therefore, to deduce whether some

353

lungs did not release R C S in response to tryptamine infusions or whether R C S was released and the aortas used were not sensitiveenough to detect it. In two experiments, isolatedlungs from guinea pigs sensitizedto egg albumin were challenged by injection of 10 m g egg albumin into the pulmonary artery. The lung effluent was superfused over assay tissues;the resulting contraction of the rabbit aortas was greater than that produced during an infusion of trypt~ amine through rat lungs.

3.1.4. The presence o f the following substances was not detected 3.1.4.1. R CS-releasing factor (R CS-RF). Perfusate from shocked guinea-pig lungs contains a stable RCS-RF (Piper and Vane, 1969). The perfusate from rat lungs, collected during an infusion of tryptamine (2 /zg/ml), did n o t contain any detectable RCS-RF (5 experiments). 3.1.4. 2. Angiotensin. The experiment shown in fig. 2 compares the response of the assay tissues, treated with combined antagonists, to tryptamine infusions through the rat lung with the response produced by an infusion of angiotensin (2 ng/ml) given directly to the tissues. It is clear that the contractor activity in lung perfusate after tryptamine could n o t be due to angiotensin. Similar results were obtained in 4 experiments. 3.1.4.3. Histamine and bradykinin. A strip of cat terminal ileum was treated with methysergide (200 ng/ml). This tissue is sensitive to histamine (1--10 ng/ml) b u t n o t to tryptamine (2/~g/ml). It is also contracted by kinins (0.5--2 ng/ml) so a second strip treated with both methysergide and mepyramine was used to detect kinins. Tryptamine (2 pg/ml, 7 expts.) infused through rat lungs did n o t release histamine or bradykinin. In 2 experiments, tryptamine was infused at a concentration of 10 ~g/ml and also single injections of up to 500/~g were given into the pulmonary artery b u t no histamine or kinin release was detected. This lack of release is illustrated in fig. 4 b u t with 5-HT as the infused amine. To show that the lungs were capable of

354

V.A. ALABASTER, Y.S. BAKHLE

14E1HYSERGIOE

I I I

,, ~.TI

i 5CM

M!cPYR~INE

I

/ \

I

1 ,,

GPI

AL

p~"ml

~

ng .g n g g g ~ pg

ng JJg 2~{~

ng

ug

Fig. 3. Release of active substances by tryptamine and compound 48]80 in rat isolated lung: lack of histamine release by tryptamine. The record shows contraction of assay tissues superfused in series with

the effluent from rat lungs perfused with Krebs solution. In descending order, the assay tissues shown are a cat terminal ileum (CTI) treated with methysergide (200 ng/ml), a CTI treated with methysergide and mepyramine (100 ng/ml) and a rat stomach strip (RSS) treated 'with combined antagonists. Tryptamine (T) and compound 48/80 were infused directly to the assay tissues (DIR) or through the lungs (AL). Responses were also obtained to prostaglandin E2 (PGE2) and histamine (H) infused directly. Tryptamine infused through the lungs released an SRS and a PG-like substance but no histamine, while compound 48/80 released histamine and a small amount of PG-like substance.

releasing histamine, c o m p o u n d 48/ 80 (1 pg/ml) was infused through them. There was an immediate release of histamine (20--100 ng/ml) which, since we had selected tissues with high sensitivity to histamine, generally exceeded th e calibration range of the tissue. An e x p e r i m e n t comparing the effects of tryptamine and c o m p o u n d 48/ 80 on rat lung is shown in fig. 3. The lung perfusate ba t he d t w o pieces

of cat terminal ileum, one treated with mepyramine (100 ng/ml), and a rat stomach strip and guinea-pig ileum treated with com bi ned antagonists. T r y p t a m i n e (2 pg/ml) had no direct effect on the tissues, but when infused through the lungs p r o d u c e d a large c o n t r a c t i o n of the rat stomach strip and guinea-pig ileum w i t h o u t affecting the cat terminal ileums. Thus, despite a release of SRS-like and PG-like material, no histamine was detected. In contrast, 48/80 (2 pg/ml) infused through the sam e lungs produced an immediate release of histamine (over 50 ng/ml) and a very small release of SRS-like and PG-like substances. 3.1.4.4. Catecholamines. The chick rect um is relaxed by adrenaline but n o t by noradrenaline whereas b o t h catecholamines relax the rat stomach strip and the rat colon. F u r t h e r m o r e , an infusion o f noradrenaline (1 ng/ml) diminished the c o n t r a c t o r response of the rat stomach strip to an infusion of PGE2 (1 ng/ml). This effect would n o t be seen under our standard conditions as the com bi ned antagonists include p h e n o x y b e n z a m i n e and propranolol. However, the s m o o t h muscle c o n t r a c t o r activity released f r o m the lungs by t r y p t a m i n e was unchanged when propranolol and p h e n o x y b e n z a m i n e were o m i t t e d from the com bi ned antagonists infused over the assay tissues.

3. 2. Changes in perfusion pressure Concentrations of t r y p t a m i n e s which released spasmogens from rat lungs also increased the perfusion pressure (8--22 m m Hg). The standard t r y p t a m i n e infusions (2 pg/ml) increased pressure by nearly 12 m m Hg (table 1).

3. 3. Other releasing agents Infusions of 5-HT, histamine and acetylcholine through the p u l m o n a r y circulation of rat isolated lungs also released additional biological activity into the lung effluent; these substances also increased perfusion pressure (table 1). The effects of each agonist amine are described below.

AMINE-INDUCED

SPASMOGEN

RELEASE

FROM

LUNGS

rial from the lungs and increased perfusion pressure (table I). The presence of an SRS and RCS in the lung perfusate was not investigated.

lOmln

1 GPI* PERFUSION

, ./~'~--~___

PRESSURE

flM~ Hg

1

J 0

DIR per ml

BK 2ng

h h

~

H 5HT 48*80 lOng 0"SJJg 1)Jg

h

H 50rig

355

BK 1rig

3. 3. 3. Acetylcholine Acetylcholine (500 ng/ml) released a PG-like substance and an SRS from the rat lungs. No RCS was detected in the lung perfusate in the 2 experiments in which rabbit aortas were used. Acetylcholine had little or no effect on perfusion pressure. A comparison of the release of PG-like substances by tryptamine, histamine and acetylcholine from rat lung is shown in fig. 5. Tryptamine and histamine produced a sub-

h

AL per ml

5HT 4S-80 0-SJJe l~g lOmin

Fig. 4. Lack of histamine and bradykinin release from rat isolated lung induced by 5-HT. The record shows contractions of a cat terminal ileum treated with methysergide (CTI top trace), a cat terminal ileum treated with methysergide and mepyramine (middle), and a guinea-pig ileum (GPI) treated with combined antagonists (bottom). The assay tissues were superfused in series with the effluent from a rat lung perfused with Krebs solution. The lowest record shows lung perfusion pressure. 5-HT infused directly to the tissue (DIR) had no effect but the same infusion given through the lung (AL) released an SRS but no histamine or bradykinin. In comparison, compound 48/80 infused through the lung released a l a r g e amount (>50 ng/ml) of histamine. The top CTI was sensitive to 10 and 50 ng/ml histamine (H) infused directly and both CTI contracted to bradykinin (BK), 1 and 2 ng/rnl infused directly.

3.3.1.5-HT Routinely, 500 ng 5-HT/ml were used, although 50 ng/ml were effective in a few experiments (see fig. 1). The material released behaved similarly to that released by tryptamine, though there was less of the PG-like substance (table 1). Release was accompanied by increases in perfusion pressure equivalent to those produced by tryptamine (table 1). No release of histamine, bradykinin (see fig. 4), catecholamines or RCS-RF was detected.

3.2.2. Histamine Histamine (2 pg/ml) released PG-like mate-

i

RSS*

i

i

1

J i

I

I

I

i

i

5¢m

CR*

2O

PERFUSION PRESSURE

,

] 10 m m Hg I

r

|

~

POE t DiR!~rmlJO.Sng

' ,

L~I

H 2~g

F"

AL per ml J~

PGF.~ lng

i AC H 0.SjJg

I

L, Ti

I

2~g I

%

H

2~g

T

2.ug

I I

% ACH 0.5~g

Fig. 5. Release of active substances by tryptamine, histamine and acetylcholine in rat isolated lung. The upper records show contractions of a rat stomach strip (RSS) and a chick rectum (CR) superfused in series with the effluent from rat lungs perfused with Krebs solution. The assay tissues were treated with combined antagonists. The lowest record shows lung perfusion pressure. Tryptamine (T), histamine (H) and acetylcholine (ACH) infused directly to the tissues (DIR) had no effect, but when infused into the pulmonary artery cannula (AL) they released substances from the lung which contracted the assay tissues. The contractions were compared with those produced by prostaglandin E2 (PGE2) infused directly. Infusions of tryptamine and histamine through the lung increased lung perfusion pressure but acetylcholine had no marked effect.

356

V.A. ALABASTER, Y.S. BAKHLE

TABLE 3 Effect of antagonists on amine-induced release of spasmogen and increase of perfusion pressure in rat isolated lungs.

Antagonist

Agonist amine Tryptamine

5-HT

Histamine

Acetylcholine

Blocked (5)*

Blocked (2)

No effect (2)

No effect (3)

No effect (2)

No effect (3)

Methysergide (200 ng/ml) Hyoscine (100 ng/ml)

Blocked (4)

* Number of experiments.

stantial rise in perfusion pressure but acetylcholine did not.

while the tryptamines were still effective (table 3).

3.4. Effects o f methysergide and hyoscine on release from rat lung

3. 5. Release of active substances from guineapig perfused lungs

The antagonists were infused through the lungs for 10 min before, and during, the infusion of agonist amine. Methysergide (200 ng/ml) abolished both the release of PG-like substances and the perfusion pressure rise induced by the tryptamines, although histamine and acetylcholine still induced release and pressure rises in the same rat lungs. Hyoscine (100 ng/ml) prevented the effects of acetylcholine

The effects of amines on guinea-pig isolated lungs are summarized in table 4. Of the amines used, only histamine induced release of spasmogens and increases in perfusion pressure. There was no evidence of an RCS-RF (3 expts.) or catecholamines (2 expts.) being present in the perfusate. Mepyramine (100 ng/ml) prevented the increase in perfusion pressure (produced by hista-

TABLE 4 Release of spasmogens and perfusion pressure increase in guinea-pig isolated perfused lungs induced by amines (-- = no effect; + = effect detected).

Amine

Type of spasmogen released PG

Tryptamine (2//g/ml) 5-HT

.

SRS .

.

No. of expts.

RCS

. .

Perfusion pressure increase

.

6

.

.

4

(0.5--1 pg/ml) Histamine (2 pg/ml) Compound

48/80

(I--I0 pg/ml)

+ .

+ .

+ .

.

+

7 3

AMINE-INDUCED SPASMOGEN RELEASE FROM LUNGS tOmln

RSS*

RC*

J

L

~

~

RbA*

~

]scm '

--

357

contracted. Mepyramine (100 ng/ml) infused through the lung blocked the increase in perfusion pressure and the release of active substance produced by histamine. The responses of the chick rectum to histamine infused directly were not markedly different from responses to histamine infused through the lungs, indicating that the pulmonary circulation of guinea-pig lung was not removing significant amounts of histamine.

3.6. Release of active substances from dog perfused lung

CR* 10 ZO m m H g tO 0

PERFUSION PRESSURE DIR per ml

PGEz lng

PGF2~ H 2rig 2fig

PGE1 OSng H

AL per ml

2ug

In dog lungs, tryptamine (2/~g/ml, 2 expts.), 5-HT (0.5 #g/ml, 1 expt.) and histamine (2 /~g/ml, 2 expts.) caused a rise in perfusion pressure and a release of activity which behaved like a mixture of PGs, SRS and RCS.

H

2~g MEPYRAMINE 1OOnllml AL

Fig. 6. Release of active substances by histamine in guinea-pig isolated lungs and effect of mepyramine. The records show contractions of a rat stomach strip (RSS), a rat colon (RC), a rabbit aorta (RbA) and a chick rectum (CR) superfused in series with the effluent from guinea pig isolated lungs perfused with Krebs solution. The assay tissues were treated with combined antagonists. The b o t t o m record shows lung perfusion pressure. Histamine (H) infused directly to the tissues (DIR) produced only a contraction of the CR but the same infusion given into the pulmonary artery cannula (AL) released substances from the lung which contracted all the assay tissues and increased perfusion pressure. The contractions of the tissues were compared to contractions produced by infusions of prostaglandin E2 (PGE2) and F2c~ (PGF2a) infused directly. Mepyramine infused through the lungs as indicated, prevented the histamine-induced release of substances from-the lung.

mine) and prevented the release of spasmogens from the lungs (2 expts.). In the experiment illustrated in fig. 6, histamine (2 pg/rnl) had a direct effect only on the chick rectum, which contracted in response to histamine despite the presence of an antihistamine in the combined antagonists. When histamine was infused through the lung, there was an increase in perfusion pressure and all the assay tissues

4. Discussion Tryptamine released smooth muscle contracting substances from rat lungs. We have tried to analyse the substances released and to find how general this phenomenon of amineinduced release was in terms of inducing amines and species. To do this, the concentration of releasing amine was fixed, generally at the highest concentration which did not interfere with bioassay of the substances released. The released substances included PG-like ~ material, an SRS and an RCS. The component most consistently present was PG-like material. Acid--lipid extraction with ethyl acetate and bioassay of the extracted material showed that this component had physical and pharmacological characteristics comparable with those of the E-type PGs. PGEI, PGE2 and PGF~ a are more than 90% inactivated in the pulmonary circulation in vivo and in vitro (Ferreira and Vane, 1967b; Piper et al., 1970). Thus the PGs that appear in the effluent from the lung are either released at a site downstream from the inactivating enzymes or represent the severely attenuated remainder of a much larger release. The activity released was consistently higher

358

when assayed on the rat stomach strip (in terms of PGE2 ) than that measured by the other t w o tissues (see table 1) but, after acid--lipid extraction, the discrepancy disappeared (table 2). These results suggested that other substances active only on the stomach strip were being released. Partially purified SRS-A will contract the stomach strip w i t h o u t affecting rat colon or chick rectum (Piper and Vane, 1969), and those authors (unpublished) have also found that RCS contracts rat stomach strip, but not chick rectum. The presence of an SRS (as defined by Brocklehurst, 1960) and RCS was demonstrated directly in several experiments, although they were never detected in the absence of PG-like material. The SRS caused a contraction of the ileum whose latency and duration were comparable to those described for SRS-A by Brocklehurst (1960). This c o m p o n e n t was released by the lungs of all three species used. SRS-like materials are formed in a number of tissues from many species in response to stimuli other than anaphylaxis {Chakravarty et al., 1959; Brocklehurst, 1960; Marquis and Smith, 1963). RCS was originally defined by Piper and Vane {1969) as a material contracting the rabbit aorta, spontaneously inactivated within 20 min and distinguishable from 5-HT, histamine, bradykinin, catecholamines, PGE2, angiotensin, potassium ion, and other known substances. The release of RCS from dog or rat lungs has n o t previously been described and, apart from satisfying the criterialaid down, we do n o t k n o w if rat and dog RCS are the same as guinea-pig RCS. An RCS is also released from guinea-pig lungs by arachidonic acid (Vargaftig and Dao Hal, 1971) and another formed during prostaglandin synthesis from arachidonic acid catalysed by dog chopped spleen (Gryglewski and Vane, 1971). Whether or n o t all these types of RCS are identical cannot be decided until RCS has been stabilized. Active substances including PGs, SRS and RCS are released from guinea-pig lungs during anaphylaxis (Piper and Vane, 1969), b u t these are different from the present experiments in

V.A. ALABASTER, Y.S. BAKHLE

two respects. First, the amounts of active materials released in our experiments are much smaller than those released during anaphylaxis. For example, PG release in anaphylaxis is between 20--100 ng/ml, whereas we found only about 0.5--4 ng/ml. We may also have been close to the threshold of detection of RCS and SRS and, since calibration to these two substances was n o t possible, lack of detection of SRS and RCS in an experiment may have been due to relatively insensitive assay tissues. The second difference is that we could detect no histamine release (except by c o m p o u n d 48/80), whereas this amine was always released by anaphylaxis. Histamine can be released by the tryptamines from skin and muscle (Feldberg and Smith, 1953) and from dog lungs (Moore et al., 1963) although at least 5 0 0 p g of the tryptamine is needed. In all these cases, large amounts of histamine are released {0.5--10 pg/ml), well above the threshold of response of our assay tissues (1--10 ng/ml). Although we failed to detect histamine release by the tryptamines, we easily demonstrated histamine release from rat lungs by 48/80. This contrasts with the results of Moore et al. (1963) who found some preparations of dog lungs were refractory to both 5-HT and c o m p o u n d 48/8O. A mixture of PGE2 and other unidentified active substances similar to the one we describe was released from guinea-pig isolated lungs (Berry et al., 1972) during positive pressure ventilation (32 strokes/min for 5--15 min). In this material, no histamine was detected and the amounts of PG released were small (0.5-3.0 ng PGE2/ml). The release was not correlated with the uptake and metabolism of the vasoactive amines. 5-HT, tryptamine and histamine all caused release, b u t both 5-HT and tryptamine were readily inactivated in the pulmonary circulation of all three species, whereas histamine was not inactivated either in isolated lung (see Bakhle and Vane, 1974 for references) or in vivo (Ferreira et al., 1973). The release of PG-like substances may be secondary to the contraction of vascular

AMINE-INDUCED SPASMOGEN RELEASE FROM LUNGS

smooth muscle and this concept is supported by the finding that antagonists inhibit perfusion pressure increase and release simultaneously. However, release was not correlated with the magnitude of the increase as acetylcholine caused release but little increase in perfusion pressure. Furthermore, the observed pressure change may be the sum of the effects of the inducing amine and those of the released Spasmogens. It is unlikely that the bronchial smooth muscle is involved in this release as the bronchial circulation would not be perfused in our preparation. At the biochemical level, the crucial event in release may be activation of phospholipase A. Activation of this enzyme has been linked with the release of SRS during anaphylaxis (HSgberg and Uvn~, 1960) and phospholipase A can also increase PG synthesis in perfused lungs (Vogt et al., 1969}. Thus activation of endogenous phospholipase A could give rise to both SRS and PGs and, since PG biosynthesis can be accompanied by RCS formation (Gryglewski and Vane, 1971), also to RCS. The amount, localisation or type of phospholipase A would have to be enough only to generate SRS, PGs and RCS and not enough to liberate histamine (Marquis and Smith, 1963). The link between the presence of agonist amine (5-HT, acetylcholine, etc.), and the activation of phospholipase A is unproven but could perhaps involve the requirement of this enzyme for calcium ions and the rise in intracellular calcium during activation of smooth muscle. The release of active substances from lungs is induced by a number of vasoactive amines in the isolated lungs of different species. This release phenomenon seems therefore to be a fairly general response although the components of the released mixture of spasmogens may vary. Many aspects of the release phenomenon remain to be elucidated and it is being investigated further. Since many of the pharmacokinetic properties of the pulmonary circulation in isolated lungs are also present in the pulmonary circulation in vivo (Bakhle and Vane, 1974), it is possible that this release may also occur in vivo (Sandier, 1972), adding an

359

endocrine function to the other, well-established functions of the lungs.

Acknowledgements We thank Dr. J.R. Vane for his valuable criticisms. One of us (VAA) thanks the Wellcome Trust for generous financial support. This work was supported by the M.R.C.

References Alabaster, Valerie, A., 1971, Metabolism of vasoactive substances by the lung, Ph.D. Thesis, London University. Alabaster, Valerie, A. and Y.S. Bakhle, 1970, The release of biologically active substances from isolated lungs by 5-hydroxytryptamine and tryptamine, Brit. J. Pharmacol. 40, 582P. Bakhle, Y.S., A.M. Reynard and J.R. Vane, 1969, Metabolism of the angiotensins in isolated perfused tissues, Nature (London) 222, 956. Bakhle, Y.S. and J.R. Vane, 1974, Pharmacokinetic function of the pulmonary circulation, Physiological Rev. 54, 1007. Berry, E.M., J.F. Edmonds and J.H. Wyllie, 1972, Release of prostaglandin E2 and unidentified factors from ventilated lungs, Brit. J. Surgery 58, 189. Brocklehurst, W.E., 1960, The release of histamine and formation of a slow reacting substance (SRSA) during anaphylactic shock, J. Physiol. (London) 151,416. Chakravarty, N., B. HSgberg and B. Uvn~is, 1959, Mechanism of the release by compound 48/80 of histamine and of a lipid soluble smooth-musclestimulating principle ('SRS'), Acta Physiol. Scand. 45, 255. Feldberg, W. and A.N. Smith, 1953, Release of histamine by tryptamine and 5-hydroxytryptamine, Brit. J. Pharmacol. Chemotherap. 8, 406. Ferreira, S.H. and J.R. Vane, 1967a, The disappearance of bradykinin and eledoisin in the circulation and vascular beds of the cat, Brit. J. Pharmacol. Chemotherap. 30, 417. Ferreira, S.H. and J.R. Vane, 1967b, Prostaglandins; their disappearance from and release into the circulation, Nature (London) 216, 868. Ferreira, S.H., K.K.F. Ng and J.R. Vane, 1973, The continuous bioassay of the release and disappearance of histamine in the circulation, Brit. J. Pharmacol. 49, 543. Furchgott, R.F. and S. Bhadrakom, 1953, Reactions

360 of strips of rabbit aorta to epinephrine, isopropylarterenol, sodium nitrite and other drugs, J. Pharmacol. Exptl. Therap. 108, 128. Gilmore, N., J.R. Vane and J.I-L Wyllie, 1968, Prostaglandins released by the spleen, Nature (London) 219, 740. Gryglewski, R. and J.R. Vane, 1971, Rabbit aorta contracting substance (RCS) may be a prostaglandin precursor, Brit. J. Pharmacol. 43, 420P. HSgberg, B. and B. Uvn~is, 1960, Further observations on the disruption of rat mesentery mast cells caused by compound 48/80, antigen-antibody reaction, Lecithinase A and Decylamine, Acta Physiol. Scand. 48, 133. Mann, M. and G.B. West, 1950, The nature of hepatic and splenic sympathin, Brit. J. Pharmacol. Chemotherap. 5, 173. Marquis, V.O. and W.G. Smith, 1963, The effects of histamine release on the lipid content of the isolated perfused lung of sensitised guinea-pigs, J. Pharm. Pharmacol. 15, 652. Moore, T.C., L. Normell and B. Eiseman, 1963, Effect of serotonin loading on histamine release and blood flow of isolated perfused liver and lung, Arch. Surgery (Chicago) 87, 42.

V.A. ALABASTER, Y.S. BAKHLE Piper, Priscilla J. and J.R. Vane, 1969, Release of additional factors in anaphyiaxis and its antagonism by anti-inflammatory drugs, Nature (London) 223, 28. Piper, Priscilla J., J.R. Vane and J.H. Wyllie, 1970, Inactivation of prostaglandins by the lungs, Nature (London) 225, 600. Regoli, D. and J.R. Vane, 1964, A sensitive method for the assay of angiotensin, Brit. J. Pharmacol. 23, 351. Sandler, M., 1972, Migraine -- a pulmonary disease? Lancet i, 618. Vane, J.R., 1957, A sensitive method for the assay of 5-hydroxytryptamine, Brit. J. Pharmacol. 12, 344. Vargaftig, B.B. and N. Dao Hai, 1971, Release of vasoactive substances from guinea-pig lungs by slow-reacting substance C and arachidonic acid, Pharmacol. (Basel) 6, 99. Vogt, W., U. Meyer, H. Kunze, E. Lufft and S. Babilli, 1969, Entstehung von SRS-C in der durchstrSmten Meerschweinchenlunge durch Phospholipase A. Identifizierung mit Prostaglandin, Arch. Exptl. Pathol. Pharmakol. 262, 124.