Nonadrenergic, Noncholinergic Relaxation of Bovine Iris Sphincter: Role of Endogenous Nitric Oxide

Abstract

purpose. To investigate the role of endogenously generated nitric oxide (NO) in the relaxation of bovine iris sphincter.

methods. Isolated bovine sphincters were mounted on an isometric tension apparatus. Contraction–relaxation response was elicited by electrical field stimulation (ES; 12 Hz, 50-msec duration, 70–80 V). Relaxation was arbitrarily defined as maximal decrease of tension below prestimulation baseline after cessation of ES. We also determined the tissue levels of cyclic guanosine monophosphate (cGMP) by radioimmunoassay.

results. ES produced a biphasic response: contraction followed by relaxation. After cessation of ES, the muscle relaxed to below the initial baseline tension. Tetrodotoxin (TTX) abolished most of the contraction and all the relaxation response. Atropine blocked most of the contraction component, leaving the relaxation component unchanged. Prazosin and bupranolol (α₁-adrenergic and β-adrenergic antagonists, respectively) also did not affect the relaxation component of the response. Neither substance P nor its antagonist (N-acetyl-l-tryptophane 3,5-bis (trifluoromethyl)-benzyl ester; ATTB) inhibited or mimicked the response. The nitric oxide synthase (NOS) inhibitors N^ω-nitro-l-arginine methyl ester (l-NAME), N ^ω-nitro-l-arginine (l-NNA), and aminoguanidine dose-dependently inhibited the relaxation response by 50% to 70%. The free radical scavenger 2-(4-carboxyphenyl)- 4,4,5,5-tetramethyl-imidazoline-1-oxyl-3-oxide (carboxy-PTIO) and the guanylyl cyclase inhibitor methylene blue also abrogated 70% and 45% of the relaxation response, respectively. ES caused an increase in muscle cGMP from 2.3 ± 0.3 to 3.9 ± 0.5 picomoles per muscle. l-NNA or l-NAME significantly decreased the tissue cGMP content (to 1.2 ± 0.1 picomoles per muscle) and prevented the increase caused by ES.

conclusions. The relaxation component of the iris sphincter response to ES is a distinct nonadrenergic, noncholinergic, ES-induced event. Most of the relaxation is mediated by the endogenously generated NO-guanylyl cyclase-cGMP cascade.

Nitric oxide (NO) is an important mediator of smooth muscle relaxation, particularly of the vascular smooth muscle. ^{1 2 3 4 5 6 7} It has been found to mediate nonadrenergic, noncholinergic (NANC) relaxation of the respiratory, ⁸ intestinal, ^{9 10 11} and genitourinary ^{12 13} smooth muscles. Despite the demonstration of nitric oxide synthase (NOS) in ocular tissues, ^{14 15 16 17 18 19 20 21 22 23 24 25} the role of endogenously produced NO in ocular physiology and pathology is still unclear. Recently, several reports have demonstrated that ocular smooth muscles (trabecular meshwork, ciliary, and iris sphincter) respond to exogenous NO donors. ^{26 27 28 29}  

In the present study, we investigated the role of endogenously produced NO. Using specific NOS or guanylyl cyclase inhibitors and monitoring the muscle cyclic guanosine monophosphate (cGMP) content, our data demonstrate that endogenous NO causes an increase in tissue cGMP and thereby mediates approximately 70% of the relaxation of bovine iris sphincter. 

Materials and Methods

Preparation of Bovine Iris Sphincters

Freshly enucleated bovine eyes were kept on ice and dissected immediately before each experiment. The entire sphincter was dissected as described elsewhere. ²⁶ The circular muscle preparation was mounted on an isometric tension apparatus in 15 ml Krebs–Ringer bicarbonate physiological salt solution continuously aerated with an O₂-CO₂ (95:5) gas mixture at 31°C. 

Electrical Field Stimulation

After a 90- to 180-minute initial equilibration period with frequent solution changes, the tension of the preparation, measured with a force transducer (model FT03C; Grass Instrument, Quincy, MA) was adjusted to 0.5 to 1.0 g and maintained at this level throughout the experiment. Electrical stimulation (ES) was applied through two platinum wires parallel to the preparation. The tissue was stimulated for 24 seconds at 12 Hz, 80 V, and 50-msec pulse duration. These conditions have been found to produce optimal response (Fig. 1) . An additional stimulation was performed when the tension returned to baseline, but at no interval closer than minutes. Tension was measured with a polygraph amplifier-recorder (model 79C; Grass). Drugs were added directly to the organ bath from 100- to 1000-fold concentrated stock solutions. 

cGMP Determination

Entire iris sphincters were mounted on the isometric tension apparatus, and the desired experiments performed as described above and in the Results section. Immediately after the experiment, the muscle was quickly removed and frozen in liquid nitrogen. The frozen tissue was homogenized (Ultra-Turrax, Janke-Kunkel AG, Staufen, Germany; at setting 8, 10 seconds) in 3 ml cold 6% trichloroacetic acid. The homogenate was centrifuged 15 minutes at 2000g at 4°C. The supernatant was extracted four times with five volumes of water-saturated ether. Residual ether was removed by vacuum. After acetylation, 50-μl samples were assayed in duplicate for cGMP content by radioimmunoassay (RIA), using a cGMP determination kit (Biotrak RPA 525; Amersham, Arlington Heights, IL). Because of the difficulty of protein determination in the highly pigmented tissue, results were presented as picomoles cGMP per entire muscle. 

Materials

N ^ω-nitro-l-arginine, l-NNA, N ^ω-nitro-l-arginine methyl ester (l-NAME), l-arginine, atropine methylnitrate, aminoguanidine, carbamylcholine, methylene blue, substance P, N-acetyl-l-tryptophane 3,5-bis (trifluoromethyl)-benzyl ester (ATTB), and prazosin were purchased from Sigma (St. Louis, MO). 2-(4-Carboxyphenyl)- 4,4,5,5-tetramethyl-imidazoline-1-oxyl-3-oxide (carboxy-PTIO) was produced by Calbiochem (La Jolla, CA). Bupranolol was a gift of Schwartz Pharma, Monheim, Germany. Tetrodotoxin (TTX) was purchased from Alomone Laboratories (Jerusalem, Israel). All other chemicals were of analytical grade. The composition of the Krebs–Ringer bicarbonate solution was (in mM): 118 NaCl, 4.7 KCl, 1.2 MgCl₂, 2.5 CaCl₂, 25 NaHCO₃, 1.2 KH₂PO₄, and 11 glucose (pH 7.4). 

Statistics

All experiments were repeated at least four times on sphincters obtained from different animals. Results are presented as mean ± SE. Student’s t-test was used to evaluate statistical significance at P < 0.05. For calculation of IC₅₀ values, the normalized data (expressed as a percentage of maximal relaxation) were fitted to Michaelis–Menten kinetics using the GraFit program (Erithacus Software Ltd, Staines, UK). 

Results

ES resulted in a complex contractile response of the iris sphincter muscle. The response consisted of a rapid contraction followed by a variable phase. In some preparations the tonus of the muscle was either maintained throughout the ES period or decreased slightly. In many preparations, however, the contraction component was followed by a relaxation component of variable rate. After cessation of ES there was a rapid relaxation, which usually brought the muscle to a tension below the initial baseline. The magnitude of the decrease in muscle tension below that of the prestimulation baseline was arbitrarily defined by us as a measure of the relaxation component of the response. Typical recordings are shown in Figure 2A . For sphincters from the same batch analyzed on the same day, the relaxation response was reproducible. For example, analysis of nine sphincters yielded a mean relaxation value of 0.20 ± 0.02 g. In a minority of preparations (<10%), the tension returned to the baseline value after the cessation of ES. These preparations were not included in the study, because they had no distinct relaxation component. These results suggest that the relaxation component of the response resulted from the action of a neurotransmitter released by ES. Moreover, it appears to be a distinct phenomenon and not an absence of contraction. 

To verify these hypotheses, we performed two experiments. We blocked most of the contraction component with 0.3 μM atropine, a muscarinic antagonist. Atropine inhibited the contraction component of the response by more than 80%, but the relaxation component of the response was unaffected (Fig. 2B) . We also subjected sphincters to ES in the presence of TTX, a blocker of fast sodium channels. TTX prevents depolarization of axons and thus blocks release of neurotransmitters. TTX at 0.4 μM blocked the relaxation component of the response and almost completely abolished the contractile component within 1 minute of its addition. Prolonged washout (60 minutes) of TTX resulted in a partial recovery of the original response (Fig. 3) . 

Our results with atropine suggested that the generation of NO is not dependent on the stimulation of muscarinic receptors. We have also studied the possible involvement of the α₁- andβ -adrenergic receptors. Because our previous studies ³⁰ showed that there is an involvement of β₃-adrenergic receptors in the relaxation of bovine iris sphincters, we used high-concentration (1 μM) bupranolol, which inhibits allβ -adrenergic receptor subtypes. ³¹ Neither atropine, prazosin, or bupranolol affected the relaxation component of the response. Moreover, the proportion of the response that could be inhibited by 0.1 mM l-NAME did not change (Fig. 4) . We concluded, therefore, that the relaxation component reflects release of NANC transmitter(s). 

It has been reported that substance P elicits contractile responses in bovine sphincter. ³² We tried, therefore, to test whether substance P receptors participate in the relaxation response. Substance P alone had a small contractile effect (24% ± 5% of that produced by ES), even at concentrations as high as 1 μM (not shown). Substance P (0.2–1.0 μM) potentiated the ES-induced relaxation response by 14%± 7%. ATTB, an antagonist of neurokinin-1 receptors, alone had no effect on sphincter tension (not shown). At very high concentration (1μ M), it potentiated both contraction and relaxation by approximately 40% to 50%. This potentiation, however, was observed with the vehicle alone (1% dimethyl sulfoxide). These results indicate that substance P is not the neurotransmitter responsible for ES-induced relaxation. 

To investigate the possible contribution of the endogenously generated NO to the relaxation component of the response, we used the specific NOS inhibitors l-NAME, l-NNA, and aminoguanidine. l-NNA (0.1 mM) inhibited relaxation by 65%± 11% (n = 9, P < 0.02, Fig. 6 ). l-NAME dose-dependently inhibited the relaxation component of the response with an IC₅₀ of 60 nM and maximal effect (68% inhibition) at more than 10 μM. Similar results were obtained with the inhibitor of inducible NOS, aminoguanidine. The IC₅₀ was, however, 11.5 μM, suggesting that the enzyme participating in the relaxation response was constitutive NOS. These results are shown in Figure 5 . 

NOS inhibitors often increased the slope of the contraction component, partially or fully prevented the decrease in tension during the ES period, and decreased the slope of the relaxation on cessation of ES. These results (not shown) suggest that relaxation proceeds independently of and in parallel with the contraction component from the onset of ES. 

To further verify the involvement of endogenously generated NO in the relaxation response, we used the free radical scavenger carboxy-PTIO. In four experiments, 0.5 mM carboxy-PTIO inhibited the relaxation component by 70% ± 7% (P < 0.02). Similarly, the guanylyl cyclase inhibitor methylene blue inhibited the relaxation response by 45% ± 3% (n = 5, P < 0.02). These results are shown in Figure 6 . Therefore, pharmacologic investigation strongly suggests that at least 70% of the relaxation component of the response to ES is mediated by endogenously generated NO. 

To complement our study, we have attempted to potentiate the relaxation component of the response by adding the NOS substrate l-arginine. Although we observed variable potentiation of the relaxation in some experiments, usually l-arginine had no effect on the response. This suggests that endogenous arginine concentration is usually sufficient to support maximal relaxation. l-Arginine, at 1.0 mM or higher, was inhibitory to both the contraction and the relaxation components of the response, suggesting nonspecific effects. 

To confirm the hypothesis that ES causes an increase of endogenously generated NO and produces relaxation through the activation of guanylyl cyclase, we assayed the content of iris sphincter cGMP under a variety of conditions. ES for 24 seconds produced an increase in the cGMP content (from 2.3 ± 0.3 to 3.9 ± 0.5 picomoles per sphincter; n = 8 and 7, respectively; P < 0.02). Incubation of the sphincters with 10μ M l-NNA or l-NAME produced a modest but significant decrease in cGMP content (to 1.2 ± 0.1 picomoles per sphincter; n = 5; P < 0.02), which was not affected by ES (1.5 ± 0.1 picomoles per sphincter; n = 7; not significant relative to l-NNA or l-NAME alone). These results (Fig. 7) are consistent with ES-induced increase in tissue NO, which activated guanylyl cyclase. The increased cGMP caused the relaxation of the iris sphincter muscle. 

Discussion

Several studies have documented existence of NOS or reduced nicotinamide adenine dinucleotide phosphate (NADPH) diaphorase in ocular tissues, among which are the conjunctiva, ³³ iris vasculature, ³⁴ iris sphincter, ^{26 33 35} ciliary processes and ciliary muscle, ^{36 37} trabecular meshwork, ^{36 37} choroid, ³⁶ and retina. ^{33 36 38 39 40} In the present report, we suggest a physiological function of this enzyme in the normal bovine eye. Our results demonstrate a distinct mechanism for the relaxation of the iris sphincter. Similar to the mechanisms operating in the vascular smooth muscle, relaxation is an independent process rather than simply a cessation of contraction. Endogenously generated NO appears to serve as the mediator of relaxation according to a number of pharmacologic criteria. Relaxation due to exogenously administered NO donors has been shown previously. ^{10 26 27 28 41 42 43} Here, we demonstrated that inhibitors of NOS abolished most of the relaxation response. The same effect was observed when the effector enzyme guanylyl cyclase was inhibited by methylene blue and when NO was eliminated by the free radical scavenger carboxy-PTIO. The specific inhibitors of constitutive NOS were much more potent than aminoguanidine, an inhibitor of the inducible enzyme. ⁴⁴ Therefore, the relaxation of iris sphincter appears to be largely mediated by NO synthesized by constitutive NOS. 

The accepted pathway of NO signal transduction includes the activation of guanylyl cyclase and an increase in cellular cGMP. The results of cGMP determination are consistent with this mechanism. ES increased sphincter cGMP content almost twofold, whereas l-NNA or l-NAME decreased the basal level to 50% of the control values. Moreover, NOS inhibitors prevented the increase in cGMP due to ES. 

The basal levels of cGMP were determined per intact sphincter muscle. The wet weight of bovine sphincter averages approximately 0.2 g, with a conservative estimate of protein content of approximately 20 mg per muscle. These rough assumptions yield cGMP content of the resting tissue of approximately 0.1 picomole per milligram protein. This was similar to the values found for the human lower esophageal sphincter ⁴³ and the rat ileum ¹¹ and somewhat lower than that reported for rabbit urethra strips. ¹³ In other tissues, much higher values were reported. In pig gastric fundus, ¹⁰ cat adrenals, ⁴² and rabbit iris sphincter ^{26 27} the values vary between 0.7 and 3.0 picomoles per milligram. In the few reports of the effect of ES on cGMP content (range between 50% and 150% in unstimulated control samples ^{10 11 12 41 43} ), the increases brought about by ES were comparable to those found here. Moro et al. ⁴² alone have reported up to a 15-fold increase in cGMP content after ES in cat adrenals. The tissue cGMP can be markedly elevated by exogenous NO donors (e.g., sodium nitroprusside) in a variety of smooth muscles, which explains their potency as relaxants. ^{26 29 41 42 43} However, our results and those of other laboratories ^{10 11 12 41 43} indicate that a twofold increase is sufficient to obtain the maximal physiological relaxation. 

The extent of the participation of NO in the relaxation process was estimated conservatively as approximately 70%. We based this estimate on the maximal inhibition of relaxation by l-NAME, l-NNA, aminoguanidine, or carboxy-PTIO. This may be interpreted in terms of two mechanisms of relaxation: one that is NO-mediated and an additional unknown mechanism that mediates the remaining 30% of the response. 

Independent of these considerations, it is obvious that a NANC pathway activates the relaxation of the bovine iris sphincter. Despite the existence of neurokinin-1 receptors and a demonstration that substance P may elicit a contractile response in the isolated sphincter, ³² our data rule out this transmitter as a cause of sphincter relaxation. Relaxation may be due to a direct nitrergic innervation (as is demonstrated in Reference 45) or to a release of another transmitter that, in turn, activates the enzyme in the sphincter muscle. Our data demonstrate the feasibility of sphincter relaxation through the nitrergic pathway. Further experiments in vivo are required to assess the importance of this pathway in normal physiology. 

 Supported by a Binational Science Foundation grant (YO) and the Foglanst Grant for Eye Research (OG).

 Submitted for publication November 18, 1998; revised April 7, 1999; accepted May 4, 1999.

 Commercial relationships policy: N.

 

Figure 1.

 

ES-induced relaxation as a function of stimulus frequency or duration. Bovine sphincter was mounted on an isometric tension apparatus. Response was elicited by ES of 80 V, 50-msec pulse duration, and variable frequency (A) or at 12 Hz and variable pulse duration (B). Results are expressed as a percentage of maximal relaxation amplitude.

Figure 2.

 

Representative contraction–relaxation response. (A) Sphincter muscle mounted on an isometric tension apparatus was stimulated for 24 seconds at 80 V, 12 Hz, and 50-msec pulse duration (horizontal bar). (B) Stimulation of the same muscle, 10 minutes after the addition of 0.3 μM atropine.

Figure 3.

 

The effect of TTX on ES-induced response. A contraction–relaxation response was elicited in an isolated bovine sphincter as described in Figure 2 . After three consecutive responses, 0.4 μM TTX was added and responses elicited by periodic stimulation. When maximal effect of TTX was observed, the tissue was repeatedly washed and subjected to a further series of stimulations. (A) Control response before the addition of TTX; (B) response obtained 1 minute after the addition of TTX; (C) partial recovery of the response obtained 60 minutes after removal of TTX and repeated washing. Similar results were obtained in three additional experiments.

Figure 4.

 

The effects of sympathetic and parasympathetic antagonists on relaxation. Sphincters were treated as described in the Methods section and in Figure 2 . Drugs were added at least 10 minutes before ES. The results are presented as a percentage of control relaxation (without drugs). Each point represents the mean ± SE of 4 to 12 independent experiments. The concentrations of the drugs were atropine (ATR) 0.3 μM, prazosin (PRZ) 0.1 μM, and bupranolol (BUP) 1.0μ M.

Figure 5.

 

Inhibition of relaxation: Dose response to NOS inhibitors. Control relaxation was determined on each sphincter by obtaining three similar consecutive contraction–relaxation complexes. The ES was thereafter repeated at 3-minute intervals in the presence of increasing doses of the drugs, starting 10 minutes after the drugs’ addition. Relaxation was calculated as a percentage of the control response (i.e., before the addition of the drugs). Each point represents the mean ± SE of five to six or three to four independent assays for l-NAME and aminoguanidine, respectively.

Figure 6.

 

Inhibition of relaxation by various drugs. The experiments were conducted essentially as described in the Methods section and in Figure 1 . After three to four control contraction–relaxation responseswere obtained, drugs were added at the following concentrations: aminoguanidine (AG) and l-NAME 10 μM, l-NNA 100 μM; methylene blue (MB) 200 μM, and carboxy-PTIO (c-PTIO) 500 μM. After at least 10 minutes, additional ES was applied. Results are presented as inhibition of control relaxation (before the addition of any drug). Each column represents the mean ± SE of four to nine experiments.

Figure 7.

 

The effects of ES and drugs on sphincter cGMP content. Sphincters were mounted on the isometric tension apparatus and stimulated as described in the Methods section and Figure 1 . The tissue was removed from the bath at the point of maximal relaxation and rapidly frozen in liquid nitrogen. cGMP was assayed as described in the Methods section. Drugs, when present, were added at least 10 minutes before the tissue was frozen. Each column represents the mean ± SE of seven to eight determinations in different sphincters. Because the effects of NOS inhibitors (either l-NNA or l-NAME, 10 μM) were similar, the results were pooled. C, control (without ES); INH, NOS inhibitors (either l-NNA or l-NAME, 10μ M).