Actu Physiul Scund 1991, 141, 531-539
ADONIS
0001677291000778
Relaxation of sheep urethral muscle induced by electrical stimulation of nerves : involvement of nitric oxide A. GARCIA-PASCUAL"?, G. COSTA", A. GARCIA-SACRISTAN" and K.-E. A N D E R S S O N t " Department of Physiology, Faculty of Veterinary, Complutense University, Madrid, Spain and Department of Clinical Pharmacology, University Hospital, Lund, Sweden
t
GARCIA-PASCUAL, A,, COSTA,G., GARCIA-SACRISTAN, A. & ANDERSON,K.-E. 1991. Relaxation of sheep urethral smooth muscle induced by electrical stimulation of nerves. Actu Physiol Scund 141, 531-539. Received 22 October 1990, accepted 6 December 1990. ISSN 0001-6772. Department of Physiology, Faculty of Veterinary, Complutense University, Madrid, Spain, and Department of Clinical Pharmacology, University Hospital, Lund, Sweden. Isolated smooth muscle preparations from the sheep urethra responded to electrical field stimulation with contraction when basal tension was low (5-6 mN), but with relaxation when the preparations were contracted with noradrenaline (NA), clonidine, or prostaglandin F2&.No relaxant response could be elicited in high K+ (124 mM) contracted preparations. Electrically induced relaxations had a threshold of less than 1 Hz and a maximum at 8 Hz. Both contractant and relaxant responses were abolished by tetrodotoxin, indicating that they were caused by transmitters released from nerves. The amplitude of the relaxant responses showed a highly significant correlation to the tension induced by noradrenaline. A coefficient (R/T) was calculated relating relaxation to noradrenaline-induced tension. In this way it is possible to separate the effect of drugs on muscle tension (non-specific effect) from their action on the electrically induced relaxation (specific effect). Chemical sympathectomy with 6-OHDA did not significantly modify the relaxant response to 6 Hz in noradrenaline contracted strips, as evaluated by the R / T coefficient. The electrically induced relaxation was not affected by hexamethonium, propranolol, phentolamine, muscarinic receptor blockade, cocaine, indomethacin, or methysergide. Both nifedipine and Bay K 8644 inhibited significantly the response induced by electrical stimulation, decreasing its maximum. Nifedipine, but not Bay K 8644, significantly reduced the level of tension induced by noradrenaline, and its effect, evaluated by the R / T coefficient, was an increase in the electrically induced relaxation, whereas Bay K 8644 had a significant inhibitory effect. Pre-treatment with N"-nitro-L-arginine (L-NOARG) for 30 minutes did not significantly change resting tension, but concentration-dependently reduced the relaxant responses, and at the highest L-NOARG concentration used, relaxation was changed into a contraction. NO (present in acidified solution of NaNO,) still produced relaxation. N'-nitro-D-arginine (D-NOARG) had no effects. It is suggested that in contracted sheep urethral muscle, the relaxation produced by electrical stimulation of nerves is mediated by NO.
Key wurds ; sheep, urethra, smooth muscle, relaxation, transmitter, nitric oxide, electrical stimulation.
Correspondence : K.-E. Andersson, MD, Department of Clinical Pharmacology, University Hospital of Lund, S-221 85 Lund, Sweden.
53 1
532
A . Garcia-Pascual et al. a
2.5 mN
L
1
2
4
6
8
12Hz
2 rnin
NA ( 1 0 - 5 ~ )
b
15 min
NA ( 1 0 - 5 ~ )
TTX (10-6M)
Fig. 1. (a) Relaxant responses to electrical field stimulation of sheep urethral preparations M). (b) The responses to electrical field stimulation are contracted by noradrenaline (NA, M). blocked by tetrodotoxin (TTX,
INTRODUCTION In isolated urethral smooth muscle from rabbit, pig and man, transmural stimulation of nerves has previously been shown to evoke a relaxant response, particularly in preparations with a high resting tension (Anderson et al. 1983, Klarskov et al. 1983). The relaxation had a maximum at low frequencies of stimulation, and was abolished by tetrodotoxin (TTX) suggesting that it was produced by transmitter(s) released from nerves. However, the nature and the localization of this relaxation producing transmitter and the mechanism of its effect are not known. It is well established that the urethra receives adrenergic, cholinergic as well as a non-adrenergic, non-cholinergic (NANC) innervation (Anderson & Sjogren 1982), but so far the role of the latter two types of innervation has not been established. The endothelium derived relaxing factor (EDRF) has been shown to be an important factor for vascular relaxation (Furchgott 1990), and there is evidence that EDRF is nitric oxide (NO). NG-monomethyl-L-arginine (L-NMMA) prevents the synthesis of N O from L-arginine in several cell types, including the endothelium of
vascular smooth muscle, macrophages, neutrophils, and brain cells (Moncada et al. 1989). However, also in non-vascular smooth muscle, N O may have a relaxant function. Thus, in the rat (Gillespie et al. 1989, Hobbs & Gibson 1990) and mouse (Gibson et al. 1990) anococcygeus muscle, and also in the canine ileocolic junction (Bult et al. 1990) and guinea-pig trachea (Tucker et al. 1990), L-NMMA and another analogue, L-NG-nitrO arginine (L-NOARG), inhibited NANC-mediated relaxation. I n the present study, the relaxant response to transmural nerve stimulation has been further studied in isolated sheep urethral muscle. As the NANC-mediated relaxant response in the rabbit was recently reported to be blocked by LNOARG (Anderson et al. 1990), special attention was given to the effects of this drug. MATERIALS AND METHODS Dissection and preparation. Lower urinary tracts from female lambs were collected at the local slaughterhouse and transported to the laboratory in ice-cold Krebs solution within 30 min. Specimens were transferred to a Petri dish containing Krebs solution at room temperature. The urethra was removed and opened longitudinally. The mucous membrane was discarded and transversal
Urethral relaxation and nitric oxide 100
C
._ c
: 0
50
0
0
8
Frequency (Hz) 0
I
C
._ c
-
50
a n:
i
i*
8
\
100
Y'
Fig. 2. Frequency-response relationship for the contractant and relaxant responses of electrically stimulated sheep urethral preparations. The contractant responses have been obtained in preparations at a low basal tension, and the relaxant responses in noradrenaline ( M) contracted strips. Results are expressed as per cent of the maximum response. Mean+ S.E.M. ( n = 59-61).
strips were dissected approximately 5 mm long and with a diameter of 2 mm. All preparations were taken from the proximal urethra. Only one preparation was obtained from each urethra. All preparations were investigated the day the animals were sacrificed. Recording of mechanical activity. Urethral preparations were attached by silk ties between two L-formed hooks in organ baths (Hogestatt et al. 1983) with a volume of 5 ml. Mechanical activity was recorded isometrically by means of a Grass Instruments FT03C force-displacement transducer, connected to one of the two hooks. The signal from the transducer was displayed on a Grass Instruments model 7D polygraph. The other hook was attached to a displacement unit, which could be manipulated by a micrometer screw for precise adjustment of the resting tension. The bath solution was bubbled with 95 yo 0, and 5 yo CO,, maintaining pH at 7.4. During an equilibration period of approximately 60 min, the resting tension was adjusted to 5 mN. Electrical stimulation. Preparations were stimulated with a Cibertec CS20 or Grass S88 stimulator
533
connected to a pair of platinum electrodes placed in parallel to the preparation. Square-wave pulses of 0.8 ms duration and with a frequency of 1-50 Hz were applied for periods of 5 s at a supramaximum voltage. Relaxant responses induced by electrical stimulation were investigated on urethral preparations contracted with NA M), clonidine M), or PGF,, (lo-' M), producing stable contractions, 7&80y0 of the maximal responses. The effects of investigated substances were compared with control experiments run in parallel. In vitro denervation with 6-hydroxydopamine (6OHDA). Urethral strips were treated with 6-OHDA according to the method previously described by Aprigliano & Hersmeyer (1976). 6-OHDA is extremely susceptible to oxidation at neutral and alkaline pH. In order to reduce its rate of oxidation, an unbuffered solution containing glutathione (0.02 mM, Sigma) and bubbled with 100yoN, (pH 4.9), was used as vehicle. 6-OHDA at a concentration of 0.45 mg x ml-' was applied for 15 min, and the bath solution was changed every two min during this period. At all other times and during the recording of contractile responses, tissues were bathed in standard Krebs solution of pH 7.4. The effects of 6-OHDA were investigated at the frequency producing a relaxation amounting to 8&95y0 of the maximum response. The mean of three consecutive reproducible responses was regarded as representative. Control preparations run in parallel showed highly reproducible responses (variation < 10%). Solutions and drugs. Standard Krebs solution contained (mM): 119 NaCI, 4.6 KC1, 1.5 CaCI,, 1.2 MgCI,, 24.9 NaHCO,, 1.4 KH,PO,, 0.07 EDTA (ethylenediaminetetraacetic acid, disodium salt) and 11 glucose. Analytical grade chemicals and redistilled water were used for all solutions. The following drugs were used : adenosine (Serva), adenosine-5'-triphosphate (ATP, Serva), Bay K 8644 (Bayer), cocaine hydrochloride (ACO), D-arginine, hexamethonium bromide, 6-hydroxydopamine, indomethacin, L-arginine (Sigma), methysergide maleate (Sandoz], NaNO,, nifedipine, NG-nitro-L-arginine (L-NOARG), NGnitroe-arginine (D-NOARG) (Sigma), dl-noradrenaline hydrochloride (Serva), phentolamine hydrochloride (Ciba-Geigy), propranolol hydrochloride (ICI-Farma), prostaglandin (PG) E,, PGE, (Sigma), PGF,, (Dinoprost, Upjohn), scopolamine hydrochloride, serotonin (Sigma), tetrodotoxin (Sigma). Drugs were dissolved in redistilled water except nifedipine, PGE,, PGE,, and indomethacin, which were dissolved in ethanol (99%), and methysergide which was dissolved in methanol. PGF,, was provided in ampoules (Dinoprost, Upjohn) and further dilutions were made in water. Acidified solutions of NaNO, (pH 2) were prepared as described by Furchgott
534
A . Garcia-Pascual et al.
(1988). Nifedipine and Bay K 8644 were kept in the dark to avoid light-induced degradation. T h e solvents did not affect the preparations. All substances were added directly to the organ bath in volumes of 5-50 , ~ l . Treatment of dnta. Statistical analysis was performed by using Student's t-test for paired and unpaired data. A probability level less than 5% was regarded as significant. T h e results shown in figures and text are expressed as mean valuefS.E.M., where n denotes the number of preparations examined (all from
...
20-
-E
./.
I
0
::
10-
-
.. ..*
L 0)
0-
0
10
20
30
4'0
Sb
Contraction (mm)
Table 1. Effects of different drugs on electrically induced relaxation of isolated sheep urethral preparations
ControlCd TTXd' Hexamethoniumd' Atropined' Scopolamined' Propranolold' Cocained' Indomethacind' Methysergided' Control" Phentolamine"' Nifedipined' Ray K 8644'' L-NOARGds
10-5 M 10-4 M
13 6 5 5 6 5 5 6 5 5 5 5 6
2.4 f0.4
4.8 f0.3"
87.6 f6.7 5.2f0.5""" 80.7 f4.5 73.9 f4.3 88.9 f6.4 97.8 f7.5 79.5 f 4 . 3 83.6 f5.4 71.8 f6.2 88.4f 3.2 84.253.3 34.5 f5.9""" 55.8k9.8"
9 9
0 0
28.0 k 6.4""" 0
4
2.0 f0.5
61.5k7.5
6
1.7 f0.2
117f4.9"
0 3.7 f0.3 3.1 f 2 . 2 3.3 f2.9 2.6 k 0.7 1.9 f0.9 2.2 f0.4 4.1 f0.9" 0.4 i0.01 0.6 i0.07
0
D-NOARGd8
10-4 M L-arginine
10-3 M
a The frequency giving 50% of the maximum response. 0 denotes values that did not achieve 50% relaxation. Maximum relaxant response expressed as yo of the maximum relaxation in the control curve of each experiment. Values used as controls belong to the second frequency-response curve in experiments run in parallel. d , Preparations were contracted by noradrenaline M (d) or prostaglandin F,, M (e). ' Drugs were added 15 min before the second frequency-response curve at M. L Drugs were added 30min before the second frequency-response curve. Mean fS.E.M., n indicates the number of preparations examined. " = P < 0.05, *"*: = P < 0.001.
Fig. 3. Linear correlation between the amplitude of the relaxation induced by electrical field stimulation of sheep urethral preparations with the degree of M), muscular tension elicited by noradrenaline ( as revealed by regression analysis. y = 0.127f0.31x, r = 0.62 ( P < 0.001).
different animals). EF,, values, i.e., the frequency of electrical stimulation producing half maximum effect, were determined graphically for each curve by linear interpolation. T h e method of least squares was used for calculating linear regressions.
RESULTS At resting tension, electrical stimulation within the frequency range 1-50 Hz evoked contractile responses. The amplitude of the contraction elicited at 50 Hz amounted to about 60% of the maximal response to NA (reached at 3 x M). In preparations where tension was increased by addition ofNA M, electrical field stimulation induced a frequency-dependent relaxant response (Figs. 1 & 2). The threshold frequency was less than 1 Hz and the maximum was reached at 8 Hz (EF,, = 1.86f0.37 Hz, n = 59). The maximum relaxant response amounted to 62 f3 % ( n = 59) of the contraction induced by NA M). There was a highly significant correlation ( r = 0.62, P < 0.001, n = 36) between the increase in tension induced by NA and the amplitude of the relaxant response (Fig. 3). A coefficient (R/T) was therefore calculated, relating the amplitude of the relaxant response with the degree of muscle tension induced by NA. In this way it was possible to separate the effect of drugs on muscle tension (non-specific effect) from their action on the electrically induced relaxation (specific effect).
Urethral relaxation and nitric oxide a
200
i
535
b
C
C
0 .c 0 ([I
s C
100
C
0 .+-
0"
([I
X
-a,
8
([I
a 8
1
*
a 0
c
1oc
*1
C 0 .c ([I
-1 B
100
8
* 200 Fig. 4. The effect of 6-hydroxydopamine (6-OHDA, filled bars) pretreatment on the relaxant response induced by 6 Hz electrical field stimulation (a and b), and on the contraction induced by noradrenaline 10-' M (c) in sheep urethral preparations. Results are expressed: (a) as per cent of the control relaxation obtained before treatment with 6-OHDA. (b) as per cent of the R / T coefficient. 100% denotes the values obtained before treatment. (c) as per cent of the control contraction induced by noradrenaline at the beginning of the experiment. Mean S.E.M. (n = 5). *; P < 0.05, when compared with control experiments run in parallel (open bars). Blockade of nerve transmission with tetroM) abolished the relaxation (Fig. 1 dotoxin ( & Table 1) without affecting the contractile response to NA. However, ganglionic blockade M), or addition of with hexamethonium ( M), cocaine M), atropine propranolol M), or scopolamine (lo-' M), did not significantly affect the relaxant frequencyresponse relationship (n = 6) (Table 1). Electrical stimulation induced relaxation also M in preparations contracted by clonidine (n = 6), or PGF,, lo-' M (n = lo), showing
Fig. 5. Maximum relaxation induced by electrical field stimulation in sheep urethral preparations contracted by noradrenaline (lo-' M) n the absence (0) and presence of nifedipine M (m) or Bay K 8644 M (a).Drugs were added 15 min before the second frequency-response curve. Results are expressed (a) as per cent of control relaxation before treatment. (b) as per cent of the R / T coefficient. Mean+S.E.M. (n = 5). * P < 0.05, **P < 0.01, ***P< 0.001.
that this relaxation was independent on the stimulation of a specific receptor type. However, in preparations contracted by high K' concentrations (124 mM), no relaxant response could be evoked (n = 6). In PGF,, contracted preparations, phentolamine M did not influence the relaxant response (n = 6) (Table 1). After chemical sympathectomy with 6O H D A , the relaxant response to 6 Hz in NA contracted strips, as evaluated by the R / T coefficient (Fig. 4) was not significantly modified. T h e contractile response to electrical stimulation was reduced to less than 5% (not shown). However, the contractile response to NA was increased by 6-OHDA treatment, and consequently, the amplitude of the electrically induced relaxation, in absolute values, was also increased (Fig. 4). ATP and adenosine (10-9-10-4 M) had effects neither on preparations at basal tension, nor on NA-contracted strips (n = 4-8). At the basal
536
A . Garcia-Pascual et al.
Control
ES ( Hz ) 1
2
4
6
8 12 16
ES ( Hz )
NQj (log M) -55-4 -45 -3 -35
m
NO, (log M)
J
I t L-NOARG (104hn)
Fig. 6. Differential effects of NG-nitro-L-arginine (L-NOARG) on the relaxation caused by electrical field stimulation (ES) and on the response to exogenously added NO (NO,-) in isolated sheep urethral muscle. The preparations were contracted by noradrenaline (NA) M. tension level, both PGE, and PGE, caused concentration dependent contractions. However, in preparations contracted by NA, PGE, induced contraction and PGE, relaxation. The maximum relaxation induced by PGE, was 48 f 14% (n = 8) of the tension induced by NA M). Inhibition of the prostaglandin synthesis with indomethacin ( M) did not significantly affect the relaxation induced by electrical stimulation (n = 5) (Table 1). Serotonin (10-9-10-4 M) had weak and inconsistent effects. Blockade of serotonin receptors with methysergide ( M) had a depressant action on the electrically-induced relaxations, increasing the EF,, value (Table 1). However, this agent also induced a 20% decrease in the level of tension produced by NA and its effect on the R / T coefficient was not significant (n = 6). Both nifedipine M) and Bay K 8644 ( M) inhibited significantly the response induced by electrical stimulation decreasing its maximum (Fig. 5 & Table 1). Nifedipine, but not Bay K 8644 significantly reduced the level of tension induced by NA to 31 & 5% of control values (P < 0.001, n = 5 ) , and its effect, evalu-
ated by the R / T coefficient, was an increase in the electrically induced relaxation, whereas Bay K 8644 had a significant inhibitory effect (Fig. 5). Pre-treatment with L-NOARG ( 10-5-10-4 M) for 30 min did not significantly change resting tension, but concentration-dependently reduced the relaxant responses (Figs. 6 & 7, Table 1). At the highest L-NOARG concentration used, M (n = 6), relaxation was changed into a contraction (Fig. 6). N O (present in acidified solution of NaNO,) still produced relaxation (Fig. 6). D-NOARG had no effects (Fig. 7 & Table 1). Electrically induced relaxation was slightly (18+ 5yo, n = 6, P < 0.05) enhanced by Larginine M (Fig. 7). L-arginine M did not reverse the inhibition caused by L-NOARG lo-, M. However pre-treatment with L-arginine M prevented the inhibitory effect of LNOARG lo-, M on relaxation. In contrast, in the presence of D-arginine M, electrically induced relaxation was abolished by L-NOARG M in 15 min (n = 4). N O (present in acidified solutions of NaNO,
Urethral relaxation and nitric oxide
--
537
0-
g
. I -
c
8
c
0
8 50v
C
.-0
.I-
l 3
-maX, [r
100-
A
0
1
2
4
6
8
Frequency ( Hz )
l
-4
I
a
I
-3 NOi(log M)
Fig. 7. Frequency- and concentration-response curves for the relaxation induced by electrical field stimulation and exogenously added NO, respectively, in the absence (0) and presence of NGnitro-L-arginine (L-NOARG), M (m) and M (a),L-arginine M (*), or NG-nitroM, and M (A). Preparations were contracted by noradrenaline D-arginine (D-NOARG) pre-treated for 30 min with L-NOARG, D-NOARG or L-arginine. Results are expressed as per cent of maximum control relaxation before treatment, and given as mean S.E.M. (n = 4-9).
(10-4-3 x M) caused transient and concentration-dependent relaxations which were not affected by L-NOARG or L-arginine (Fig. 7). DISCUSSION I n urethral smooth muscle from rabbits (Andersson et al. 1983; Ito & Kimoto, 1985), humans (Andersson et al. 1983) and pigs (Klarskov et al. 1983), the existence of a nonadrenergic, non-cholinergic (NANC) inhibitory innervation has previously been described. T h e results obtained in the present study support the view that also in the sheep urethra, electrical stimulation induces the release of an inhibitory neurotransmitter from nerves. The relaxant response is dependent on the amount of prevailing tension, but seems to be independent of the mechanism by which tension is increased. Only in preparations contracted by high K + solutions, field stimulation did not cause any response, showing that the release and/or the action of the transmitter need the polarized state of the nerve and/or muscle membrane. T h e relaxant response was blocked by T T X and unaffected by hexamethonium, suggesting
that the electrical stimulus acts mainly on intrinsic nerve terminals. However, the lack of effect of a- and /3-adrenoceptor and muscarinic receptor antagonists as well as of cocaine suggests that the inhibitory transmitter is of NANCnerve origin. It now seems clear that most nerves contain a mixture of different neurotransmitter substances that vary in proportion in different tissues and species and probably during development and disease (Burnstock 1986). The frequencyresponse relationship for the relaxation in the sheep urethra was demonstrated at a lower range of frequencies than that of the contractile activation. This suggests the possible coexistence of NA with a NANC inhibitory transmitter in the sympathetic nerves, which could be released by different frequencies of stimulation. However, chemical sympathectomy with 6-OHDA failed to relaxation, suggesting that the relaxationproducing transmitter has its origin in nonadrenergic nerves. ATP has been proposed to be an excitatory NANC transmitter in the urinary bladder (Burnstock 1972). In the sheep urethral preparations, ATP and adenosine failed to induce either
538
A . Garcia-Pascual et al.
a contractile or a relaxant response. Anderson et al. (1983) and Hills et al. (1984) described in the rabbit and pig urethras slowly developing and incomplete relaxations induced by very high concentrations of ATP and adenosine. Therefore, ATP and adenosine are unlikely contenders for the NANC inhibitory transmitter role. Lower urinary tract tissues are able to synthesize and release different PGs of E- and Ftype (Downie & Karmazyn 1984, Jeremy et al. 1986). Prostanoid synthesis can be produced spontaneously and also induced by different stimuli, e.g. nerve stimulation. In several species, PGs of the E-type, especially PGE,, show a dual effect, contractant on bladder and relaxant on urethra, suggesting its involvement in bladder emptying. Thus, reduction of intraurethral pressure by PGs of E-type has been demonstrated in dogs (Mutoh et al. 1983), cats (Person 1976) and humans (Anderson et al. 1977). The results obtained in the present study showed dual effects of PGE, on sheep urethral smooth muscle, depending the prevailing muscle tension. However, the lack of effect of indomethacin on relaxation induced by electrical stimulation, suggests that PGs are not involved in this response. Serotonin has been considered as a potential neurotransmitter in the urinary tract smooth muscles (Creed & Tulloch 1978, Klarskov & Hclrby-Petersen 1986). Serotonin causes contraction of the detrusor muscle (Creed & Tulloch 1978) and rabbit and dog urethras (Creed & Tulloch 1978, Anderson et al. 1983). However, pig urethral smooth muscle was shown to be relaxed by serotonin (Hills et al. 1984, Klarskov & Herby-Petersen 1986), and Hills et al. (1984) suggested its participation in the NANC relaxant response of the pig urethra. I n the present experiments, the responses to serotonin were weak and inconsistent, and methysergide had no significant effect on the R / T relationship, suggesting that serotonin is not involved in the nerve-mediated relaxant response in sheep urethra. In the present study, nifedipine increased the electrically induced relaxation in preparations contracted by NA (evaluated by the R / T relationship), whereas Bay K 8644 produced a significant inhibition. This suggests that the relaxant response is not dependent on inhibition of voltage operated calcium channels in urethral smooth muscle, but rather that an increased
Ca2+-influx may compete with the postjunctional mechanisms causing relaxation. Whether or not dihydropyridine sensitive calcium channels are involved in the synthesis or release of N O (see below) cannot be decided from these experiments. L-NOARG and analogues prevent the synthesis of N O from L-arginine in several cell types (Moncada et al. 1989), including the endothelium of vascular smooth muscle, where NO has been identified as an important relaxant factor. Also in the present experiments on sheep urethral muscle, L-NOARG inhibited the electrically evoked relaxations, and the relaxant responses were enhanced by L-arginine, whereas no effects were obtained by the D-analOgUeS. In addition, N O (present in acidified NaNO, solution) relaxed the preparation, even in the presence of L-NOARG. This is in agreement with preliminary findings in the rabbit urethra (Anderson et al. 1990). T h e present results thus suggest that NO, or a N O containing compound, is the main mediator of the electrically induced urethral relaxation. Even if it has been shown that neuronal structures have the capacity to generate NO (Knowles et al. 1989), it cannot be decided from the present experiments whether NO is released from nerves directly, or whether an unknown nerve-released transmitter causes the generation of NO. This project was supported by the Swedish Medical Research Council (project no. 06837), and by the Spanish Ministry of Science and Education (grant no. PF90 7215117).
REFERENCES ANDERSON K.-E., EK, A. & PERSON,C.G.A. 1977. Effects of prostaglandins on the isolated human bladder and urethra. Acta Physiol Scand 100, 165-171. ANDERSON, K.-E., GARCIA-PASCUAL, A., FORMAN, A. & TOTTRUP, A. 1990. Non-adrenergic, non-cholin-
ergic nerve mediated relaxation of rabbit urethra is caused by nitric oxide. Acta Physiol Scand 141, 133-134. ANDERSON,K.-E., MATTIASSON, A. & SJOGREN, C. 1983. Electrically induced relaxation of the noradrenaline contracted isolated urethra from rabbit and man. 3 Urol 129, 210-213. ANDERSON, K.-E. & SJOGREN, C. 1982. Aspects on the physiology and pharmacology of the bladder and urethra. Progr Neurobiol 19, 71-89.
Urethral relaxation and nitric oxide
539
APRIGLIANO, 0. & HERSMEYER, K. 1976. In vitro nerve-mediated relaxation of the pig bladder neck : denervation of the portal vein and caudal artery of an examination of possible neurotransmitter candidates. Eur J Pharmacol 99, 287-293. the rat. 3 Pharmacol Exp Ther 198, 568-577. A. 1990. L-NG-nitro-arginine BULT, H., BOECKXSTAENS, G.E., PELCKMANS, P.A., HOBBS,A.J. & GIBSON, Y.M. & HERMAN, and its methyl ester are potent inhibitors of nonJORDAENS, F.H., VAN MAERCKE, adrenergic, non-cholinergic transmission in the rat A.G. 1990. Nitric oxide as an inhibitory nonBY 3 Pharmacol 100, 749-752. anococcygeus. adrenergic non-cholinergic neurotransmitter. HOGESTATT, E.D., ANDERSSON,K.-E. & EDVINSSON, Nature 345, 356-347. L. 1983. Mechanical properties of rat cerebral BURNSTOCK, G. 1972. Purinergic nerves. Pharmacol arteries as studied by a sensitive device for recording Rev 24, 509-581. of mechanical activity in isolates small blood vessels. BURNSTOCK, G. 1986. Nonadrenergic neurotransActa Physiol Scand 117, 49-61. mitters in relation to sympathetic nervous control ITO, Y. & KIMOTO,Y. 1985. The neural and nonof the lower urinary tract. Clin Sci 70 (suppl. 14), neural mechanisms involved in urethral activity in 15s-20s. rabbits. 3 Physiol 367, 57-72. CREED, K.E. & TULLOCH, A.G.S. 1978. The effect of JEREMY, J.Y., MIKHAILIDIS, D.P. & DANDONA,P. pelvic nerve stimulation and some drugs on the 1986. Prostanoid synthesis by the rat urinary urethra and bladder of the dog. Br 3 Urol 50, bladder : evidence for stimulation through mus398405. carinic receptor-linked calcium channels. Naunyn DOWNIE, J.W. & KARMAZYN, M. 1984. Mechanical Schmiedeberg’s Arch Pharmacol 334, 463467. trauma to bladder epithelium liberates prostanoids KLARSKOV, P., GERSTENBERG, T., RAMIREZ, D. & HALD,T. 1983. Non-cholinergic, non-adrenergic which modulate neurotransmission in rabbit denerve mediated relaxation of trigone, bladder neck trusor muscle. JPharmacol Exp Ther 230,445449. and urethral smooth muscle in vttro. J Urol 129, FURCHGOTT, R.F. 1990. Studies on endothelium848-8 5 0. dependent vasodilation and the endothelium deP. & HORBY-PETERSEN, J. 1986. Influence rived relaxing factor. Acta Physiol Scand 139, KLARSKOV, on lower urinary tract smooth muscle of serotonin 257-270. in vitro. Br 3 Urol 58, 507-513. FURCHGOTT, R.F. 1988. Studies on the relaxation of R.G., PALACIOS, M., PALMER, R.M.J. & rabbit aorta by sodium nitrite: the basis for the KNOWLES, MONCADA, S. 1989. Formation of nitric oxide from proposal that the acid activatable inhibitory factor L-arginine in the central nervous system : a from bovine retractor penis is inorganic nitrite and transduction mechanism for stimulation of the the endothelium-derived relaxing actor is nitric soluble guanylate cyclase. Proc Natl Acad Sci oxide. In: P.M. Vanhoutte (ed) Vasodilation : VasU.S.A. 86, 5159-5162. cular smooth muscle, Peptides, Autonomic nerves, MONCADA, S., PALMER, R.M.J. & HIGGS, E.A. 1989. and Endothelium, pp. 401414. Raven Press, New Biosynthesis of nitric oxide from L-arginine. York. Biochem Pharmacol38, 1709-1715. GIBSON, A,, MIRZAZADEH, S., HOBBS,A.J. & MOORE, MUTOH,S., UEDA,S., YANO,S., IKEGANI, K. & P.K. 1990. L-NG-monomethyl arginine and L - N ~ SAKANISHI, M. 1983. Effects of some prostaglandins nitro arginine inhibit non-adrenergic, non-cholinon urinary bladder and urethra isolated from the ergic relaxation of the mouse anococcygeus muscle. dog. Urol Int 38, 219-222. PERSSON, C.G.A. 1976. Inhibitory effect at the bladder Br 3 Pharmacol99, 602-606. urethral junction. Acta Physiol Scand 97, 139-141. GILLESPIE, J.S., LIU, X. & MARTIN, W. 1989. The J.F., BRAVE,S.R., CHARALAMBOUS, L., effects of L-arginine and NG-monomethyl L- TUCKER, HOBBS, A.J. & GIBSON,A. 1990. L-NG-nitroarginine on the response of the rat anococcygeus arginine inhibits non-adrenergic, non-cholinergic muscle to NANC nerve stimulation. B r 3 Pharmacol relaxations of guinea-pig isolated tracheal smooth 98, 108&1082. HILLS,J., MELDRUM, L., KLARSKOV, P. & BURNSTOCK, muscle. Br _7 Pharmacol 100, 663-664. G. 1984. A novel non-adrenergic non-cholinergic
ADONIS
0001677291000778
Relaxation of sheep urethral muscle induced by electrical stimulation of nerves : involvement of nitric oxide A. GARCIA-PASCUAL"?, G. COSTA", A. GARCIA-SACRISTAN" and K.-E. A N D E R S S O N t " Department of Physiology, Faculty of Veterinary, Complutense University, Madrid, Spain and Department of Clinical Pharmacology, University Hospital, Lund, Sweden
t
GARCIA-PASCUAL, A,, COSTA,G., GARCIA-SACRISTAN, A. & ANDERSON,K.-E. 1991. Relaxation of sheep urethral smooth muscle induced by electrical stimulation of nerves. Actu Physiol Scund 141, 531-539. Received 22 October 1990, accepted 6 December 1990. ISSN 0001-6772. Department of Physiology, Faculty of Veterinary, Complutense University, Madrid, Spain, and Department of Clinical Pharmacology, University Hospital, Lund, Sweden. Isolated smooth muscle preparations from the sheep urethra responded to electrical field stimulation with contraction when basal tension was low (5-6 mN), but with relaxation when the preparations were contracted with noradrenaline (NA), clonidine, or prostaglandin F2&.No relaxant response could be elicited in high K+ (124 mM) contracted preparations. Electrically induced relaxations had a threshold of less than 1 Hz and a maximum at 8 Hz. Both contractant and relaxant responses were abolished by tetrodotoxin, indicating that they were caused by transmitters released from nerves. The amplitude of the relaxant responses showed a highly significant correlation to the tension induced by noradrenaline. A coefficient (R/T) was calculated relating relaxation to noradrenaline-induced tension. In this way it is possible to separate the effect of drugs on muscle tension (non-specific effect) from their action on the electrically induced relaxation (specific effect). Chemical sympathectomy with 6-OHDA did not significantly modify the relaxant response to 6 Hz in noradrenaline contracted strips, as evaluated by the R / T coefficient. The electrically induced relaxation was not affected by hexamethonium, propranolol, phentolamine, muscarinic receptor blockade, cocaine, indomethacin, or methysergide. Both nifedipine and Bay K 8644 inhibited significantly the response induced by electrical stimulation, decreasing its maximum. Nifedipine, but not Bay K 8644, significantly reduced the level of tension induced by noradrenaline, and its effect, evaluated by the R / T coefficient, was an increase in the electrically induced relaxation, whereas Bay K 8644 had a significant inhibitory effect. Pre-treatment with N"-nitro-L-arginine (L-NOARG) for 30 minutes did not significantly change resting tension, but concentration-dependently reduced the relaxant responses, and at the highest L-NOARG concentration used, relaxation was changed into a contraction. NO (present in acidified solution of NaNO,) still produced relaxation. N'-nitro-D-arginine (D-NOARG) had no effects. It is suggested that in contracted sheep urethral muscle, the relaxation produced by electrical stimulation of nerves is mediated by NO.
Key wurds ; sheep, urethra, smooth muscle, relaxation, transmitter, nitric oxide, electrical stimulation.
Correspondence : K.-E. Andersson, MD, Department of Clinical Pharmacology, University Hospital of Lund, S-221 85 Lund, Sweden.
53 1
532
A . Garcia-Pascual et al. a
2.5 mN
L
1
2
4
6
8
12Hz
2 rnin
NA ( 1 0 - 5 ~ )
b
15 min
NA ( 1 0 - 5 ~ )
TTX (10-6M)
Fig. 1. (a) Relaxant responses to electrical field stimulation of sheep urethral preparations M). (b) The responses to electrical field stimulation are contracted by noradrenaline (NA, M). blocked by tetrodotoxin (TTX,
INTRODUCTION In isolated urethral smooth muscle from rabbit, pig and man, transmural stimulation of nerves has previously been shown to evoke a relaxant response, particularly in preparations with a high resting tension (Anderson et al. 1983, Klarskov et al. 1983). The relaxation had a maximum at low frequencies of stimulation, and was abolished by tetrodotoxin (TTX) suggesting that it was produced by transmitter(s) released from nerves. However, the nature and the localization of this relaxation producing transmitter and the mechanism of its effect are not known. It is well established that the urethra receives adrenergic, cholinergic as well as a non-adrenergic, non-cholinergic (NANC) innervation (Anderson & Sjogren 1982), but so far the role of the latter two types of innervation has not been established. The endothelium derived relaxing factor (EDRF) has been shown to be an important factor for vascular relaxation (Furchgott 1990), and there is evidence that EDRF is nitric oxide (NO). NG-monomethyl-L-arginine (L-NMMA) prevents the synthesis of N O from L-arginine in several cell types, including the endothelium of
vascular smooth muscle, macrophages, neutrophils, and brain cells (Moncada et al. 1989). However, also in non-vascular smooth muscle, N O may have a relaxant function. Thus, in the rat (Gillespie et al. 1989, Hobbs & Gibson 1990) and mouse (Gibson et al. 1990) anococcygeus muscle, and also in the canine ileocolic junction (Bult et al. 1990) and guinea-pig trachea (Tucker et al. 1990), L-NMMA and another analogue, L-NG-nitrO arginine (L-NOARG), inhibited NANC-mediated relaxation. I n the present study, the relaxant response to transmural nerve stimulation has been further studied in isolated sheep urethral muscle. As the NANC-mediated relaxant response in the rabbit was recently reported to be blocked by LNOARG (Anderson et al. 1990), special attention was given to the effects of this drug. MATERIALS AND METHODS Dissection and preparation. Lower urinary tracts from female lambs were collected at the local slaughterhouse and transported to the laboratory in ice-cold Krebs solution within 30 min. Specimens were transferred to a Petri dish containing Krebs solution at room temperature. The urethra was removed and opened longitudinally. The mucous membrane was discarded and transversal
Urethral relaxation and nitric oxide 100
C
._ c
: 0
50
0
0
8
Frequency (Hz) 0
I
C
._ c
-
50
a n:
i
i*
8
\
100
Y'
Fig. 2. Frequency-response relationship for the contractant and relaxant responses of electrically stimulated sheep urethral preparations. The contractant responses have been obtained in preparations at a low basal tension, and the relaxant responses in noradrenaline ( M) contracted strips. Results are expressed as per cent of the maximum response. Mean+ S.E.M. ( n = 59-61).
strips were dissected approximately 5 mm long and with a diameter of 2 mm. All preparations were taken from the proximal urethra. Only one preparation was obtained from each urethra. All preparations were investigated the day the animals were sacrificed. Recording of mechanical activity. Urethral preparations were attached by silk ties between two L-formed hooks in organ baths (Hogestatt et al. 1983) with a volume of 5 ml. Mechanical activity was recorded isometrically by means of a Grass Instruments FT03C force-displacement transducer, connected to one of the two hooks. The signal from the transducer was displayed on a Grass Instruments model 7D polygraph. The other hook was attached to a displacement unit, which could be manipulated by a micrometer screw for precise adjustment of the resting tension. The bath solution was bubbled with 95 yo 0, and 5 yo CO,, maintaining pH at 7.4. During an equilibration period of approximately 60 min, the resting tension was adjusted to 5 mN. Electrical stimulation. Preparations were stimulated with a Cibertec CS20 or Grass S88 stimulator
533
connected to a pair of platinum electrodes placed in parallel to the preparation. Square-wave pulses of 0.8 ms duration and with a frequency of 1-50 Hz were applied for periods of 5 s at a supramaximum voltage. Relaxant responses induced by electrical stimulation were investigated on urethral preparations contracted with NA M), clonidine M), or PGF,, (lo-' M), producing stable contractions, 7&80y0 of the maximal responses. The effects of investigated substances were compared with control experiments run in parallel. In vitro denervation with 6-hydroxydopamine (6OHDA). Urethral strips were treated with 6-OHDA according to the method previously described by Aprigliano & Hersmeyer (1976). 6-OHDA is extremely susceptible to oxidation at neutral and alkaline pH. In order to reduce its rate of oxidation, an unbuffered solution containing glutathione (0.02 mM, Sigma) and bubbled with 100yoN, (pH 4.9), was used as vehicle. 6-OHDA at a concentration of 0.45 mg x ml-' was applied for 15 min, and the bath solution was changed every two min during this period. At all other times and during the recording of contractile responses, tissues were bathed in standard Krebs solution of pH 7.4. The effects of 6-OHDA were investigated at the frequency producing a relaxation amounting to 8&95y0 of the maximum response. The mean of three consecutive reproducible responses was regarded as representative. Control preparations run in parallel showed highly reproducible responses (variation < 10%). Solutions and drugs. Standard Krebs solution contained (mM): 119 NaCI, 4.6 KC1, 1.5 CaCI,, 1.2 MgCI,, 24.9 NaHCO,, 1.4 KH,PO,, 0.07 EDTA (ethylenediaminetetraacetic acid, disodium salt) and 11 glucose. Analytical grade chemicals and redistilled water were used for all solutions. The following drugs were used : adenosine (Serva), adenosine-5'-triphosphate (ATP, Serva), Bay K 8644 (Bayer), cocaine hydrochloride (ACO), D-arginine, hexamethonium bromide, 6-hydroxydopamine, indomethacin, L-arginine (Sigma), methysergide maleate (Sandoz], NaNO,, nifedipine, NG-nitro-L-arginine (L-NOARG), NGnitroe-arginine (D-NOARG) (Sigma), dl-noradrenaline hydrochloride (Serva), phentolamine hydrochloride (Ciba-Geigy), propranolol hydrochloride (ICI-Farma), prostaglandin (PG) E,, PGE, (Sigma), PGF,, (Dinoprost, Upjohn), scopolamine hydrochloride, serotonin (Sigma), tetrodotoxin (Sigma). Drugs were dissolved in redistilled water except nifedipine, PGE,, PGE,, and indomethacin, which were dissolved in ethanol (99%), and methysergide which was dissolved in methanol. PGF,, was provided in ampoules (Dinoprost, Upjohn) and further dilutions were made in water. Acidified solutions of NaNO, (pH 2) were prepared as described by Furchgott
534
A . Garcia-Pascual et al.
(1988). Nifedipine and Bay K 8644 were kept in the dark to avoid light-induced degradation. T h e solvents did not affect the preparations. All substances were added directly to the organ bath in volumes of 5-50 , ~ l . Treatment of dnta. Statistical analysis was performed by using Student's t-test for paired and unpaired data. A probability level less than 5% was regarded as significant. T h e results shown in figures and text are expressed as mean valuefS.E.M., where n denotes the number of preparations examined (all from
...
20-
-E
./.
I
0
::
10-
-
.. ..*
L 0)
0-
0
10
20
30
4'0
Sb
Contraction (mm)
Table 1. Effects of different drugs on electrically induced relaxation of isolated sheep urethral preparations
ControlCd TTXd' Hexamethoniumd' Atropined' Scopolamined' Propranolold' Cocained' Indomethacind' Methysergided' Control" Phentolamine"' Nifedipined' Ray K 8644'' L-NOARGds
10-5 M 10-4 M
13 6 5 5 6 5 5 6 5 5 5 5 6
2.4 f0.4
4.8 f0.3"
87.6 f6.7 5.2f0.5""" 80.7 f4.5 73.9 f4.3 88.9 f6.4 97.8 f7.5 79.5 f 4 . 3 83.6 f5.4 71.8 f6.2 88.4f 3.2 84.253.3 34.5 f5.9""" 55.8k9.8"
9 9
0 0
28.0 k 6.4""" 0
4
2.0 f0.5
61.5k7.5
6
1.7 f0.2
117f4.9"
0 3.7 f0.3 3.1 f 2 . 2 3.3 f2.9 2.6 k 0.7 1.9 f0.9 2.2 f0.4 4.1 f0.9" 0.4 i0.01 0.6 i0.07
0
D-NOARGd8
10-4 M L-arginine
10-3 M
a The frequency giving 50% of the maximum response. 0 denotes values that did not achieve 50% relaxation. Maximum relaxant response expressed as yo of the maximum relaxation in the control curve of each experiment. Values used as controls belong to the second frequency-response curve in experiments run in parallel. d , Preparations were contracted by noradrenaline M (d) or prostaglandin F,, M (e). ' Drugs were added 15 min before the second frequency-response curve at M. L Drugs were added 30min before the second frequency-response curve. Mean fS.E.M., n indicates the number of preparations examined. " = P < 0.05, *"*: = P < 0.001.
Fig. 3. Linear correlation between the amplitude of the relaxation induced by electrical field stimulation of sheep urethral preparations with the degree of M), muscular tension elicited by noradrenaline ( as revealed by regression analysis. y = 0.127f0.31x, r = 0.62 ( P < 0.001).
different animals). EF,, values, i.e., the frequency of electrical stimulation producing half maximum effect, were determined graphically for each curve by linear interpolation. T h e method of least squares was used for calculating linear regressions.
RESULTS At resting tension, electrical stimulation within the frequency range 1-50 Hz evoked contractile responses. The amplitude of the contraction elicited at 50 Hz amounted to about 60% of the maximal response to NA (reached at 3 x M). In preparations where tension was increased by addition ofNA M, electrical field stimulation induced a frequency-dependent relaxant response (Figs. 1 & 2). The threshold frequency was less than 1 Hz and the maximum was reached at 8 Hz (EF,, = 1.86f0.37 Hz, n = 59). The maximum relaxant response amounted to 62 f3 % ( n = 59) of the contraction induced by NA M). There was a highly significant correlation ( r = 0.62, P < 0.001, n = 36) between the increase in tension induced by NA and the amplitude of the relaxant response (Fig. 3). A coefficient (R/T) was therefore calculated, relating the amplitude of the relaxant response with the degree of muscle tension induced by NA. In this way it was possible to separate the effect of drugs on muscle tension (non-specific effect) from their action on the electrically induced relaxation (specific effect).
Urethral relaxation and nitric oxide a
200
i
535
b
C
C
0 .c 0 ([I
s C
100
C
0 .+-
0"
([I
X
-a,
8
([I
a 8
1
*
a 0
c
1oc
*1
C 0 .c ([I
-1 B
100
8
* 200 Fig. 4. The effect of 6-hydroxydopamine (6-OHDA, filled bars) pretreatment on the relaxant response induced by 6 Hz electrical field stimulation (a and b), and on the contraction induced by noradrenaline 10-' M (c) in sheep urethral preparations. Results are expressed: (a) as per cent of the control relaxation obtained before treatment with 6-OHDA. (b) as per cent of the R / T coefficient. 100% denotes the values obtained before treatment. (c) as per cent of the control contraction induced by noradrenaline at the beginning of the experiment. Mean S.E.M. (n = 5). *; P < 0.05, when compared with control experiments run in parallel (open bars). Blockade of nerve transmission with tetroM) abolished the relaxation (Fig. 1 dotoxin ( & Table 1) without affecting the contractile response to NA. However, ganglionic blockade M), or addition of with hexamethonium ( M), cocaine M), atropine propranolol M), or scopolamine (lo-' M), did not significantly affect the relaxant frequencyresponse relationship (n = 6) (Table 1). Electrical stimulation induced relaxation also M in preparations contracted by clonidine (n = 6), or PGF,, lo-' M (n = lo), showing
Fig. 5. Maximum relaxation induced by electrical field stimulation in sheep urethral preparations contracted by noradrenaline (lo-' M) n the absence (0) and presence of nifedipine M (m) or Bay K 8644 M (a).Drugs were added 15 min before the second frequency-response curve. Results are expressed (a) as per cent of control relaxation before treatment. (b) as per cent of the R / T coefficient. Mean+S.E.M. (n = 5). * P < 0.05, **P < 0.01, ***P< 0.001.
that this relaxation was independent on the stimulation of a specific receptor type. However, in preparations contracted by high K' concentrations (124 mM), no relaxant response could be evoked (n = 6). In PGF,, contracted preparations, phentolamine M did not influence the relaxant response (n = 6) (Table 1). After chemical sympathectomy with 6O H D A , the relaxant response to 6 Hz in NA contracted strips, as evaluated by the R / T coefficient (Fig. 4) was not significantly modified. T h e contractile response to electrical stimulation was reduced to less than 5% (not shown). However, the contractile response to NA was increased by 6-OHDA treatment, and consequently, the amplitude of the electrically induced relaxation, in absolute values, was also increased (Fig. 4). ATP and adenosine (10-9-10-4 M) had effects neither on preparations at basal tension, nor on NA-contracted strips (n = 4-8). At the basal
536
A . Garcia-Pascual et al.
Control
ES ( Hz ) 1
2
4
6
8 12 16
ES ( Hz )
NQj (log M) -55-4 -45 -3 -35
m
NO, (log M)
J
I t L-NOARG (104hn)
Fig. 6. Differential effects of NG-nitro-L-arginine (L-NOARG) on the relaxation caused by electrical field stimulation (ES) and on the response to exogenously added NO (NO,-) in isolated sheep urethral muscle. The preparations were contracted by noradrenaline (NA) M. tension level, both PGE, and PGE, caused concentration dependent contractions. However, in preparations contracted by NA, PGE, induced contraction and PGE, relaxation. The maximum relaxation induced by PGE, was 48 f 14% (n = 8) of the tension induced by NA M). Inhibition of the prostaglandin synthesis with indomethacin ( M) did not significantly affect the relaxation induced by electrical stimulation (n = 5) (Table 1). Serotonin (10-9-10-4 M) had weak and inconsistent effects. Blockade of serotonin receptors with methysergide ( M) had a depressant action on the electrically-induced relaxations, increasing the EF,, value (Table 1). However, this agent also induced a 20% decrease in the level of tension produced by NA and its effect on the R / T coefficient was not significant (n = 6). Both nifedipine M) and Bay K 8644 ( M) inhibited significantly the response induced by electrical stimulation decreasing its maximum (Fig. 5 & Table 1). Nifedipine, but not Bay K 8644 significantly reduced the level of tension induced by NA to 31 & 5% of control values (P < 0.001, n = 5 ) , and its effect, evalu-
ated by the R / T coefficient, was an increase in the electrically induced relaxation, whereas Bay K 8644 had a significant inhibitory effect (Fig. 5). Pre-treatment with L-NOARG ( 10-5-10-4 M) for 30 min did not significantly change resting tension, but concentration-dependently reduced the relaxant responses (Figs. 6 & 7, Table 1). At the highest L-NOARG concentration used, M (n = 6), relaxation was changed into a contraction (Fig. 6). N O (present in acidified solution of NaNO,) still produced relaxation (Fig. 6). D-NOARG had no effects (Fig. 7 & Table 1). Electrically induced relaxation was slightly (18+ 5yo, n = 6, P < 0.05) enhanced by Larginine M (Fig. 7). L-arginine M did not reverse the inhibition caused by L-NOARG lo-, M. However pre-treatment with L-arginine M prevented the inhibitory effect of LNOARG lo-, M on relaxation. In contrast, in the presence of D-arginine M, electrically induced relaxation was abolished by L-NOARG M in 15 min (n = 4). N O (present in acidified solutions of NaNO,
Urethral relaxation and nitric oxide
--
537
0-
g
. I -
c
8
c
0
8 50v
C
.-0
.I-
l 3
-maX, [r
100-
A
0
1
2
4
6
8
Frequency ( Hz )
l
-4
I
a
I
-3 NOi(log M)
Fig. 7. Frequency- and concentration-response curves for the relaxation induced by electrical field stimulation and exogenously added NO, respectively, in the absence (0) and presence of NGnitro-L-arginine (L-NOARG), M (m) and M (a),L-arginine M (*), or NG-nitroM, and M (A). Preparations were contracted by noradrenaline D-arginine (D-NOARG) pre-treated for 30 min with L-NOARG, D-NOARG or L-arginine. Results are expressed as per cent of maximum control relaxation before treatment, and given as mean S.E.M. (n = 4-9).
(10-4-3 x M) caused transient and concentration-dependent relaxations which were not affected by L-NOARG or L-arginine (Fig. 7). DISCUSSION I n urethral smooth muscle from rabbits (Andersson et al. 1983; Ito & Kimoto, 1985), humans (Andersson et al. 1983) and pigs (Klarskov et al. 1983), the existence of a nonadrenergic, non-cholinergic (NANC) inhibitory innervation has previously been described. T h e results obtained in the present study support the view that also in the sheep urethra, electrical stimulation induces the release of an inhibitory neurotransmitter from nerves. The relaxant response is dependent on the amount of prevailing tension, but seems to be independent of the mechanism by which tension is increased. Only in preparations contracted by high K + solutions, field stimulation did not cause any response, showing that the release and/or the action of the transmitter need the polarized state of the nerve and/or muscle membrane. T h e relaxant response was blocked by T T X and unaffected by hexamethonium, suggesting
that the electrical stimulus acts mainly on intrinsic nerve terminals. However, the lack of effect of a- and /3-adrenoceptor and muscarinic receptor antagonists as well as of cocaine suggests that the inhibitory transmitter is of NANCnerve origin. It now seems clear that most nerves contain a mixture of different neurotransmitter substances that vary in proportion in different tissues and species and probably during development and disease (Burnstock 1986). The frequencyresponse relationship for the relaxation in the sheep urethra was demonstrated at a lower range of frequencies than that of the contractile activation. This suggests the possible coexistence of NA with a NANC inhibitory transmitter in the sympathetic nerves, which could be released by different frequencies of stimulation. However, chemical sympathectomy with 6-OHDA failed to relaxation, suggesting that the relaxationproducing transmitter has its origin in nonadrenergic nerves. ATP has been proposed to be an excitatory NANC transmitter in the urinary bladder (Burnstock 1972). In the sheep urethral preparations, ATP and adenosine failed to induce either
538
A . Garcia-Pascual et al.
a contractile or a relaxant response. Anderson et al. (1983) and Hills et al. (1984) described in the rabbit and pig urethras slowly developing and incomplete relaxations induced by very high concentrations of ATP and adenosine. Therefore, ATP and adenosine are unlikely contenders for the NANC inhibitory transmitter role. Lower urinary tract tissues are able to synthesize and release different PGs of E- and Ftype (Downie & Karmazyn 1984, Jeremy et al. 1986). Prostanoid synthesis can be produced spontaneously and also induced by different stimuli, e.g. nerve stimulation. In several species, PGs of the E-type, especially PGE,, show a dual effect, contractant on bladder and relaxant on urethra, suggesting its involvement in bladder emptying. Thus, reduction of intraurethral pressure by PGs of E-type has been demonstrated in dogs (Mutoh et al. 1983), cats (Person 1976) and humans (Anderson et al. 1977). The results obtained in the present study showed dual effects of PGE, on sheep urethral smooth muscle, depending the prevailing muscle tension. However, the lack of effect of indomethacin on relaxation induced by electrical stimulation, suggests that PGs are not involved in this response. Serotonin has been considered as a potential neurotransmitter in the urinary tract smooth muscles (Creed & Tulloch 1978, Klarskov & Hclrby-Petersen 1986). Serotonin causes contraction of the detrusor muscle (Creed & Tulloch 1978) and rabbit and dog urethras (Creed & Tulloch 1978, Anderson et al. 1983). However, pig urethral smooth muscle was shown to be relaxed by serotonin (Hills et al. 1984, Klarskov & Herby-Petersen 1986), and Hills et al. (1984) suggested its participation in the NANC relaxant response of the pig urethra. I n the present experiments, the responses to serotonin were weak and inconsistent, and methysergide had no significant effect on the R / T relationship, suggesting that serotonin is not involved in the nerve-mediated relaxant response in sheep urethra. In the present study, nifedipine increased the electrically induced relaxation in preparations contracted by NA (evaluated by the R / T relationship), whereas Bay K 8644 produced a significant inhibition. This suggests that the relaxant response is not dependent on inhibition of voltage operated calcium channels in urethral smooth muscle, but rather that an increased
Ca2+-influx may compete with the postjunctional mechanisms causing relaxation. Whether or not dihydropyridine sensitive calcium channels are involved in the synthesis or release of N O (see below) cannot be decided from these experiments. L-NOARG and analogues prevent the synthesis of N O from L-arginine in several cell types (Moncada et al. 1989), including the endothelium of vascular smooth muscle, where NO has been identified as an important relaxant factor. Also in the present experiments on sheep urethral muscle, L-NOARG inhibited the electrically evoked relaxations, and the relaxant responses were enhanced by L-arginine, whereas no effects were obtained by the D-analOgUeS. In addition, N O (present in acidified NaNO, solution) relaxed the preparation, even in the presence of L-NOARG. This is in agreement with preliminary findings in the rabbit urethra (Anderson et al. 1990). T h e present results thus suggest that NO, or a N O containing compound, is the main mediator of the electrically induced urethral relaxation. Even if it has been shown that neuronal structures have the capacity to generate NO (Knowles et al. 1989), it cannot be decided from the present experiments whether NO is released from nerves directly, or whether an unknown nerve-released transmitter causes the generation of NO. This project was supported by the Swedish Medical Research Council (project no. 06837), and by the Spanish Ministry of Science and Education (grant no. PF90 7215117).
REFERENCES ANDERSON K.-E., EK, A. & PERSON,C.G.A. 1977. Effects of prostaglandins on the isolated human bladder and urethra. Acta Physiol Scand 100, 165-171. ANDERSON, K.-E., GARCIA-PASCUAL, A., FORMAN, A. & TOTTRUP, A. 1990. Non-adrenergic, non-cholin-
ergic nerve mediated relaxation of rabbit urethra is caused by nitric oxide. Acta Physiol Scand 141, 133-134. ANDERSON,K.-E., MATTIASSON, A. & SJOGREN, C. 1983. Electrically induced relaxation of the noradrenaline contracted isolated urethra from rabbit and man. 3 Urol 129, 210-213. ANDERSON, K.-E. & SJOGREN, C. 1982. Aspects on the physiology and pharmacology of the bladder and urethra. Progr Neurobiol 19, 71-89.
Urethral relaxation and nitric oxide
539
APRIGLIANO, 0. & HERSMEYER, K. 1976. In vitro nerve-mediated relaxation of the pig bladder neck : denervation of the portal vein and caudal artery of an examination of possible neurotransmitter candidates. Eur J Pharmacol 99, 287-293. the rat. 3 Pharmacol Exp Ther 198, 568-577. A. 1990. L-NG-nitro-arginine BULT, H., BOECKXSTAENS, G.E., PELCKMANS, P.A., HOBBS,A.J. & GIBSON, Y.M. & HERMAN, and its methyl ester are potent inhibitors of nonJORDAENS, F.H., VAN MAERCKE, adrenergic, non-cholinergic transmission in the rat A.G. 1990. Nitric oxide as an inhibitory nonBY 3 Pharmacol 100, 749-752. anococcygeus. adrenergic non-cholinergic neurotransmitter. HOGESTATT, E.D., ANDERSSON,K.-E. & EDVINSSON, Nature 345, 356-347. L. 1983. Mechanical properties of rat cerebral BURNSTOCK, G. 1972. Purinergic nerves. Pharmacol arteries as studied by a sensitive device for recording Rev 24, 509-581. of mechanical activity in isolates small blood vessels. BURNSTOCK, G. 1986. Nonadrenergic neurotransActa Physiol Scand 117, 49-61. mitters in relation to sympathetic nervous control ITO, Y. & KIMOTO,Y. 1985. The neural and nonof the lower urinary tract. Clin Sci 70 (suppl. 14), neural mechanisms involved in urethral activity in 15s-20s. rabbits. 3 Physiol 367, 57-72. CREED, K.E. & TULLOCH, A.G.S. 1978. The effect of JEREMY, J.Y., MIKHAILIDIS, D.P. & DANDONA,P. pelvic nerve stimulation and some drugs on the 1986. Prostanoid synthesis by the rat urinary urethra and bladder of the dog. Br 3 Urol 50, bladder : evidence for stimulation through mus398405. carinic receptor-linked calcium channels. Naunyn DOWNIE, J.W. & KARMAZYN, M. 1984. Mechanical Schmiedeberg’s Arch Pharmacol 334, 463467. trauma to bladder epithelium liberates prostanoids KLARSKOV, P., GERSTENBERG, T., RAMIREZ, D. & HALD,T. 1983. Non-cholinergic, non-adrenergic which modulate neurotransmission in rabbit denerve mediated relaxation of trigone, bladder neck trusor muscle. JPharmacol Exp Ther 230,445449. and urethral smooth muscle in vttro. J Urol 129, FURCHGOTT, R.F. 1990. Studies on endothelium848-8 5 0. dependent vasodilation and the endothelium deP. & HORBY-PETERSEN, J. 1986. Influence rived relaxing factor. Acta Physiol Scand 139, KLARSKOV, on lower urinary tract smooth muscle of serotonin 257-270. in vitro. Br 3 Urol 58, 507-513. FURCHGOTT, R.F. 1988. Studies on the relaxation of R.G., PALACIOS, M., PALMER, R.M.J. & rabbit aorta by sodium nitrite: the basis for the KNOWLES, MONCADA, S. 1989. Formation of nitric oxide from proposal that the acid activatable inhibitory factor L-arginine in the central nervous system : a from bovine retractor penis is inorganic nitrite and transduction mechanism for stimulation of the the endothelium-derived relaxing actor is nitric soluble guanylate cyclase. Proc Natl Acad Sci oxide. In: P.M. Vanhoutte (ed) Vasodilation : VasU.S.A. 86, 5159-5162. cular smooth muscle, Peptides, Autonomic nerves, MONCADA, S., PALMER, R.M.J. & HIGGS, E.A. 1989. and Endothelium, pp. 401414. Raven Press, New Biosynthesis of nitric oxide from L-arginine. York. Biochem Pharmacol38, 1709-1715. GIBSON, A,, MIRZAZADEH, S., HOBBS,A.J. & MOORE, MUTOH,S., UEDA,S., YANO,S., IKEGANI, K. & P.K. 1990. L-NG-monomethyl arginine and L - N ~ SAKANISHI, M. 1983. Effects of some prostaglandins nitro arginine inhibit non-adrenergic, non-cholinon urinary bladder and urethra isolated from the ergic relaxation of the mouse anococcygeus muscle. dog. Urol Int 38, 219-222. PERSSON, C.G.A. 1976. Inhibitory effect at the bladder Br 3 Pharmacol99, 602-606. urethral junction. Acta Physiol Scand 97, 139-141. GILLESPIE, J.S., LIU, X. & MARTIN, W. 1989. The J.F., BRAVE,S.R., CHARALAMBOUS, L., effects of L-arginine and NG-monomethyl L- TUCKER, HOBBS, A.J. & GIBSON,A. 1990. L-NG-nitroarginine on the response of the rat anococcygeus arginine inhibits non-adrenergic, non-cholinergic muscle to NANC nerve stimulation. B r 3 Pharmacol relaxations of guinea-pig isolated tracheal smooth 98, 108&1082. HILLS,J., MELDRUM, L., KLARSKOV, P. & BURNSTOCK, muscle. Br _7 Pharmacol 100, 663-664. G. 1984. A novel non-adrenergic non-cholinergic
