Effects of Halothane and Enflurane Anesthesia on Sympathetic β-adrenoreceptor–mediated Pulmonary Vasodilation in Chronically Instrumented Dogs

All surgical procedures and experimental protocols were approved by the Cleveland Clinic Foundation Institutional Animal Care and Use Committee, Cleveland, Ohio.

Eighteen conditioned mongrel dogs (weight, 28 ± 1 kg) were used in this study. The surgery for chronic instrumentation has been described in detail. 7All dogs were premedicated with intramuscular morphine sulfate, 10 mg, and were anesthetized with intravenous pentobarbital sodium, 20 mg/kg, and intravenous fentanyl citrate, 15 μg/kg. After tracheal intubation, the lungs were mechanically ventilated. Anesthesia was maintained with halothane (approximately 1.2% end-tidal). A left thoracotomy was performed via  the fifth intercostal space using sterile surgical technique, and the pericardium was incised ventral to the phrenic nerve. Heparin-filled Tygon catheters (1.02 mm ID, Norton, Akron, OH) were inserted into the descending thoracic aorta, left and right atrium, and main pulmonary artery and were secured with purse-string sutures. After careful dissection and isolation, a hydraulic occluder (18 mm ID, In Vivo Metric, Healdsburg, CA) was loosely positioned around the right main pulmonary artery, and an electromagnetic flow probe (10 mm ID, Zepeda, Seattle, WA) was placed around the left main pulmonary artery. After loose apposition of the pericardial edges, the free ends of the catheters, occluder, and flow probe were threaded through the chest wall and were tunneled subcutaneously to a final position between the scapulae. A chest tube placed in the left thorax before closure was removed on the first postoperative day. Intramuscular morphine sulfate, 10 mg, was administered postoperatively for pain as required. Ampicillin, 1 g, cefazolin, 1 g, and gentamicin, 80 mg, were administered intraoperatively and on a daily basis for 10 days postoperatively. The dogs were allowed to recover for at least 2 weeks before experimentation.

As described previously, 7vascular pressures were measured by attaching the fluid-filled catheters to strain-gauge manometers (Quest Medical, Allen, TX) and were referenced to atmospheric pressure with the transducers positioned at midchest at the level of the spine. Heart rate (HR) was calculated from the phasic aortic pressure trace. Left pulmonary blood flow (LQ) was measured by connecting the flow probe to an electromagnetic flowmeter (SWF-5RD, Zepeda, Seattle, WA). The flow probe was calibrated in vivo  on a weekly basis by the thermal dilution technique. 7Values for LQ were referenced to body weight (ml·min^{− 1}·kg⁻¹). The aortic and pulmonary artery catheters were used to obtain blood samples to measure systemic arterial and mixed venous blood gases, respectively. Systemic arterial and mixed venous p  H, carbon dioxide tension (Pco²), and oxygen tension (Po²) were measured with an ABL-600 (Radiometer, Copenhagen, Denmark). Oxyhemoglobin saturation (So²) was measured with a Hemoximeter OSM-3 (Radiometer).

All solutions were prepared on the day of the experiment. Isoproterenol (Sanofi Winthrop Pharmaceuticals, New York, NY), dibutyryl cAMP (N⁶, 2′-O-dibutyryl adenosine 3′,5′-cyclic monophosphate, Sigma Chemical, St. Louis, MO), and the thromboxane analog U46619 (9,11-dideoxy-11α,9α -epoxymethano-prostaglandin F^2α, Sigma Chemical) were diluted in 0.9% saline.

All experiments were performed with each healthy, chronically instrumented dog lying on its right side in a quiet laboratory environment. Conscious dogs were not sedated. LPQ plots were used to assess the effects of the various interventions on the pulmonary circulation. LPQ plots were constructed by continuously measuring the pulmonary vascular pressure gradient (pulmonary arterial pressure–left atrial pressure [PAP-LAP]) and LQ during gradual (approximately 1 min) inflation of the hydraulic occluder implanted around the right main pulmonary artery. Measuring the pulmonary vascular pressure gradient takes into account the possibility that an experimental intervention could alter the driving pressure (PAP) or the effective downstream pressure (LAP). This technique to measure the LPQ relationship is highly reproducible and has little or no effect on systemic hemodynamics, blood gases, or the zonal condition of the lung. 7 

We investigated the effect of halothane anesthesia on the pulmonary vascular response to cumulative doses of the sympathetic β-adrenoreceptor agonist isoproterenol after preconstriction with the thromboxane analog U46619. Preconstriction is necessary because the normal pulmonary circulation is essentially maximally dilated at rest. A baseline LPQ plot was first obtained in each conscious dog (n = 8). U46619 was then administered (0.12 ± 0.01 μ g·kg⁻¹· min⁻¹) intravenously to preconstrict the pulmonary circulation before the administration of isoproterenol. LPQ plots were obtained during U46619 preconstriction alone and again with each dose of intravenous isoproterenol (0.01, 0.02, 0.05, and 0.10 μg·kg⁻¹· min⁻¹) during its incremental administration (≈15 min at each dose) and while the infusion of U46619 was continued. We have verified that pulmonary vasoconstriction induced by U46619 is stable over the time course of this protocol. 8On a separate day, this protocol was repeated in the same dogs during halothane anesthesia. Anesthesia with halothane was induced by mask and was supplemented with a subanesthetic dose of intravenous thiopental sodium, 3 mg/kg, to minimize excitatory behavior. The trachea was intubated (9 mm ID), and ventilation was controlled with a mechanical ventilator with zero end-expiratory pressure. Muscle relaxants were not used in this study. Immediately after intubation, halothane was delivered via  a vaporizer (Fluotec MK III, Ohmeda, Austell, GA). Tidal volume was fixed at 15 ml/kg. Systemic arterial blood gas values were matched to values measured in the conscious state by administering supplemental oxygen (fractional inspiratory oxygen = 0.22) and by adjusting the respiratory rate to 10–20 breaths/min. End-tidal carbon dioxide and halothane concentrations were continuously measured at the adapter end of the endotracheal tube throughout the experiment (Solar 7000, Random Access Mass Spec, Marquette Electronics, Milwaukee, WI). After induction, halothane was allowed to equilibrate for at least 1 h to achieve steady-state conditions. At this time, the end-tidal halothane concentration was 1.2–1.4%. During halothane anesthesia, the dose of intravenous U46619 (0.08 ± 0.01 μ g·kg⁻¹· min⁻¹was titrated to achieve the same degree of preconstriction induced in the conscious state (i.e. , an approximate doubling of PAP-LAP at any given value of LQ). This technique allowed us to assess the pulmonary vasodilator response to isoproterenol at the same level of vasomotor tone in the conscious and halothane-anesthetized states.

We investigated the effect of enflurane anesthesia on the pulmonary vascular response to cumulative doses of isoproterenol in the presence of U46619 preconstriction. For each conscious dog (n = 8), LPQ plots were obtained in the baseline condition, during intravenous U46619 preconstriction (0.11 ± 0.01 μg· kg⁻¹·min^{− 1}), and during the incremental administration of isoproterenol as described in protocol 1. On a separate day, this protocol was repeated in the same dogs during enflurane anesthesia. Anesthesia with enflurane was induced as described in protocol 1, and the end-tidal concentration was maintained at 3.0–3.3%. During enflurane anesthesia, the dose of intravenous U46619 (0.06 ± 0.01 μg·kg⁻¹· min⁻¹) was titrated to achieve the same degree of preconstriction induced in the conscious state.

We investigated the effect of halothane anesthesia on the pulmonary vascular response to cumulative doses of dibutyryl cAMP, the membrane-permeable cyclic AMP analog, after preconstriction with U46619. A baseline LPQ plot was first obtained in each conscious dog (n = 8). Intravenous U46619 was then administered (0.12 ± 0.02 μg·kg^{− 1}·min⁻¹) to preconstrict the pulmonary circulation before the administration of dibutyryl cAMP. LPQ plots were obtained during U46619 preconstriction alone and then again with each intravenous dose of dibutyryl cAMP (150, 300, and 600 μg·kg^{− 1}·min⁻¹) during its incremental administration (approximately 15 min at each dose) while the infusion of U46619 was continued. On a separate day, this protocol was repeated in the same dogs during halothane anesthesia. Anesthesia with halothane was induced as described in protocol 1, and the end-tidal concentration was maintained at 1.2–1.4%. During halothane anesthesia, the dose of intravenous U46619 (0.08 ± 0.01 μg·kg⁻¹· min⁻¹) was titrated to achieve the same degree of preconstriction induced in the conscious state.

Because we have previously observed that isoflurane potentiates the pulmonary vasodilator response to isoproterenol, 6we investigated the effect of isoflurane anesthesia on the pulmonary vascular response to cumulative doses of dibutyryl cAMP in the presence of U46619 preconstriction. For each conscious dog (n = 8), LPQ plots were obtained in the baseline condition, during intravenous U46619 preconstriction (0.12 ± 0.02 μg· kg⁻¹·min^{− 1}), and during the incremental administration of dibutyryl cAMP as described in protocol 3. On a separate day, this protocol was repeated in the same dogs during isoflurane anesthesia. Anesthesia with isoflurane was induced as described in protocol 1, and the end-tidal concentration was maintained at 1.6–1.7%. During isoflurane anesthesia, the dose of intravenous U46619 (0.08 ± 0.01 μg·kg⁻¹· min⁻¹) was titrated to achieve the same degree of preconstriction induced in the conscious state.

Phasic and mean vascular pressures and LQ were displayed continuously on an eight-channel strip-chart recorder (Gould, Eastlake, OH). Mean pressures and LQ, measured at end-expiration, were obtained with the use of passive electronic filters with a 2-s time constant. All vascular pressures were referenced to atmospheric pressure before and after each LPQ plot. The analog pressure and LQ signals were digitally converted and multiplexed (Medical Systems, PCM-8, Greenvale, NY) and stored on videotape (Videocassette recorder AG-1260, Panasonic, Secaucus, NJ) for later playback and analysis. As reported in previous studies, 9,10the LPQ relationship was linear by inspection over the empirically measured range of LQ. Therefore, linear regression analysis was used to calculate the slope and intercept for PAP-LAP (or PAP-0 if LAP was ≤ 0 mmHg) as a function of LQ in each experiment. The correlation coefficient for the LPQ relationship in each protocol averaged 0.98 or higher. The composite LPQ plots summarized in the figures were generated using the regression parameters from each continuously measured LPQ plot to calculate PAP-LAP at 10-ml· min⁻¹·kg^{− 1}intervals of LQ over the empirically measured range of LQ. The minimum and maximum values of LQ in each composite LPQ plot represent the average minimum and maximum values of LQ for the dogs studied in that protocol. One-way analysis of variance followed by Student t  test for paired comparisons was used to assess the pulmonary vascular effects of isoproterenol and dibutyryl cAMP within each group. Two-way analysis of variance followed by Student t  test for paired comparisons was used to assess the effects of the anesthetic agents on steady-state hemodynamics and blood gases. Two-way analysis of variance was also used to assess the effects of the anesthetics on the pulmonary vascular responses to isoproterenol and dibutyryl cAMP compared with responses measured in the conscious state. The pulmonary vasodilator response to isoproterenol (or dibutyryl cAMP) is expressed as the percentage decrease in U46619 preconstriction and was calculated with the following formula:2,4,5,7,11 

Thus, an isoproterenol-induced (or dibutyryl cAMP-induced) decrease in PAP-LAP of 100% represents a complete reversal of U46619 preconstriction and a full return to the baseline LPQ relationship. A “floating” value of LQ was used to calculate the percent reversal of U46619 preconstriction because isoproterenol and dibutyryl cAMP increased LQ in a dose-dependent manner. This allowed us to calculate the agonist-induced percent reversal of U46619 preconstriction at mid-range values of empirically measured LQ. For isoproterenol, LQ values of 80, 100, 120, and 140 ml · min⁻¹· kg⁻¹were used at concentrations of 0.01, 0.02, 0.05, and 0.10 μg · kg^{− 1}· min⁻¹, respectively, in the conscious and anesthetized states. For dibutyryl cAMP, LQ values of 80, 100, and 120 ml · min⁻¹· kg⁻¹were used at concentrations of 150, 300, and 600 μg · kg^{− 1}· min⁻¹, respectively. All values are presented as mean ± SEM.

Halothane had no net effect on the baseline LPQ relationship compared with the conscious state (fig. 1). Matched preconstriction of the pulmonary circulation in conscious and halothane-anesthetized dogs was achieved with U46619 (fig. 1). A lower dose (P < 0.05) of U46619 was required during halothane to match the same degree of preconstriction achieved in the conscious state. Figure 2summarizes the effects of one dose of isoproterenol on the LPQ relationship in conscious and halothane-anesthetized dogs. In the presence of U46619 preconstriction, the intravenous infusion of isoproterenol (0.02 μg·kg⁻¹· min⁻¹) induced pulmonary vasodilation in the conscious state and during halothane anesthesia. The pulmonary vascular dose–response relationship for isoproterenol in the conscious and halothane-anesthetized states is summarized in figure 3. After U46619 preconstriction, isoproterenol induced pulmonary vasodilation at all doses in conscious and halothane-anesthetized dogs. However, the magnitude of the pulmonary vasodilator response to isoproterenol was markedly enhanced during halothane anesthesia.

Steady-state hemodynamics and blood gases for protocol 1 are summarized in tables 1 and 2. Compared with the conscious state, baseline mean systemic arterial pressure (SAP) and LQ were decreased and LAP was increased during halothane anesthesia. U46619 increased SAP and PAP in the conscious state and during halothane anesthesia and decreased HR in the conscious state. Isoproterenol decreased SAP and increased PAP during halothane anesthesia and increased LQ and HR in both conditions. Baseline systemic arterial Pco²and mixed venous Pco², Po², and So²were slightly decreased during halothane compared with the conscious state. U46619 decreased systemic arterial So², and mixed venous p  H, Po²and So²in both conditions. Isoproterenol, 0.10 μg· kg⁻¹·min^{− 1}, decreased systemic arterial and mixed venous p  H during halothane and mixed venous Pco²in both conditions. Isoproterenol, 0.10 μg·kg^{− 1}·min⁻¹, increased mixed venous Po²and So²in both conditions.

Enflurane had no net effect on the baseline LPQ relationship compared with the conscious state (fig. 4). A lower dose (P < 0.05) of U46619 was required during enflurane anesthesia to match the same degree of preconstriction achieved in the conscious state (fig. 4). The pulmonary vascular dose–response relationship for isoproterenol in the conscious and enflurane-anesthetized states is summarized in figure 5. Although the magnitude of the pulmonary vasodilator response to isoproterenol tended to be slightly greater during enflurane anesthesia compared with the conscious state, this effect did not achieve statistical significance.

Steady-state hemodynamics and blood gases for protocol 2 are summarized in tables 3 and 4. Compared with the conscious state, the effects of enflurane on steady-state hemodynamics were similar to changes observed during halothane anesthesia. Baseline blood gases were similar in conscious and enflurane-anesthetized dogs. Changes in systemic arterial and mixed venous blood gases in response to U46619 and isoproterenol were similar to responses observed in protocol 1.

A smaller dose (P < 0.05) of U46619 was required during halothane anesthesia to match the same degree of preconstriction achieved in the conscious state. The pulmonary vascular dose–response relationship for dibutyryl cAMP in the conscious and halothane-anesthetized states is summarized in figure 6. After U46619 preconstriction, the magnitude of dibutyryl cAMP-induced pulmonary vasodilation was similar in the conscious state and during halothane anesthesia.

Steady-state hemodynamics and blood gases for protocol 3 are summarized in tables 5 and 6. Dibutyryl cAMP, 600 μg·kg⁻¹· min⁻¹, decreased SAP during halothane anesthesia and decreased LAP and increased LQ and HR in both conditions. Dibutyryl cAMP decreased systemic arterial and mixed venous Pco²and increased systemic arterial and mixed venous Po²and mixed venous So²in both conditions.

Because we previously observed that the magnitude of isoproterenol-induced pulmonary vasodilation is potentiated during isoflurane anesthesia, 6we assessed the effects of dibutyryl cAMP on the LPQ relationship in conscious and isoflurane-anesthetized dogs. A smaller dose (P < 0.05) of U46619 was required during isoflurane anesthesia to match the same degree of preconstriction achieved in the conscious state. The pulmonary vascular dose–response relationship for dibutyryl cAMP in the conscious and isoflurane-anesthetized states is summarized in figure 7. After U46619 preconstriction, the magnitude of dibutyryl cAMP-induced pulmonary vasodilation was similar in the conscious state and during isoflurane anesthesia.

Changes in steady-state hemodynamics and blood gases in response to dibutyryl cAMP during isoflurane anesthesia were similar to the changes observed during halothane anesthesia (tables 5 and 6).

This in vivo  study demonstrated that the magnitude of the pulmonary vasodilator response to the β-adrenoreceptor agonist isoproterenol was potentiated during halothane anesthesia compared with the conscious state but not during enflurane anesthesia. Moreover, the pulmonary vasodilator response to dibutyryl cAMP was similar in conscious, halothane-anesthetized, and isoflurane-anesthetized dogs.

These studies used dogs that were chronically instrumented to measure the LPQ relationship. There are several advantages to this experimental model. First, this model avoids the effects of background anesthetic agents and acute surgical trauma. Second, we were able to study the same dogs in the conscious and anesthetized states. And third, this model avoids the limitations that are inherent in the interpretation of single-point calculations of pulmonary vascular resistance. 12 

Neither halothane nor enflurane had an effect on the baseline LPQ relationship, which confirms a previous finding from our laboratory. 4A lower concentration of U46619 was required during halothane, enflurane, and isoflurane anesthesia to match the same degree of pulmonary vasoconstriction observed in the conscious state. This effect is likely the result of an attenuating influence of inhalational anesthetics on endothelium-derived relaxing factors, 1–3which normally modulate the pulmonary vascular response to a vasoconstrictor influence. 7One may expect the anesthetic agents to shift the conscious baseline LPQ relationship because of this effect. However, we have previously reported that the endothelium-derived relaxing factor nitric oxide does not tonically regulate the baseline LPQ relationship in conscious dogs. 7 

Recent evidence suggests that there are at least three possible mechanisms by which β-adrenoreceptor activation can result in vasodilation. Classically, the signal transduction pathway for β-adrenoreceptor–mediated vasodilation involves stimulation of vascular smooth muscle adenylyl cyclase and an increase in the intracellular concentration of cAMP. 13Consistent with this, isoproterenol-induced increases in cAMP have been observed in endothelium-denuded isolated canine pulmonary artery 14and rat aorta. 15A second mechanism for isoproterenol-induced vasodilation appears to be mediated by the endothelium-dependent release of nitric oxide, resulting in an increase in vascular smooth muscle cyclic guanosine monophosphate (cGMP). In rat aorta, removal of the endothelium or pretreatment with a nitric oxide synthase inhibitor markedly attenuated the vasorelaxant response to isoproterenol. 16,17A direct role for nitric oxide in isoproterenol-induced vasorelaxation has also been reported in rat mesenteric resistance arteries 18and canine coronary resistance vessels 19but not in large canine coronary arteries. 20We previously reported that isoproterenol caused an endothelium-dependent increase in cGMP in isolated canine pulmonary artery. 14A third mechanism for isoproterenol-induced vasodilation appears to involve activation of K^+ATPchannels. 21Inhibition of K^+ATPchannels attenuated the vasorelaxant response to isoproterenol in isolated rat pulmonary artery rings, whereas activation of K^+ATPchannels potentiated isoproterenol-induced vasorelaxation. 21These effects required activation of β adrenoreceptors and were not observed in response to direct activation of adenylyl cyclase with forskolin. 21Thus, there are at least three vasodilator pathways that could mediate the enhanced pulmonary vasodilator response to isoproterenol during halothane anesthesia observed in the present study.

There is a surprising paucity of information in the literature concerning the effects of inhalational anesthetic agents on β-adrenoreceptor–mediated vasodilation. Halothane and isoflurane were found to inhibit the vasorelaxation response to isoproterenol in endothelium-denuded rat aortic strips. 15Isoflurane also attenuated isoproterenol-induced relaxation in denuded rat coronary microvessels. 22In contrast, halothane had no effect on the vasorelaxant response to isoproterenol in endothelium-intact rat aortic rings. 16As noted earlier, we have previously reported that isoflurane anesthesia actually potentiated the pulmonary vasodilator response to isoproterenol. 6In the present study, halothane, but not enflurane, also potentiated the pulmonary vasodilator response to isoproterenol.

What is the mechanism by which halothane (and isoflurane) potentiated the pulmonary vasodilator response to isoproterenol? It seems unlikely that this effect could be mediated by an enhanced nitric oxide-mediated component of isoproterenol-induced vasodilation. Halothane 1and isoflurane 2have been shown to attenuate the pulmonary vasodilator response to endothelium-dependent cGMP-mediated agonists. It also appears unlikely that the potentiated response could be the result of an enhanced K^+ATP-mediated component of isoproterenol-induced vasodilation. Halothane 4and isoflurane 5have been shown to attenuate the pulmonary vasodilator response to K^+ATPchannel activation in chronically instrumented dogs. Thus, it appears most likely that the potentiated response is the result of an effect of the anesthetic agents on the β-adrenoreceptor–cAMP signaling pathway. Inhalational anesthetic agents have been shown to alter intracellular cAMP concentration. For example, isoflurane and halothane increased cAMP in rat aortic strips. 23In rat uterine homogenate, halothane increased cAMP 24and enhanced the stimulation of adenylyl cyclase by isoproterenol. 25These effects of halothane and isoflurane on cAMP accumulation were not altered by β-adrenoreceptor inhibition. 23–25Moreover, halothane had no effect on β-adrenoreceptor affinity for isoproterenol and may have even reduced β-adrenoreceptor density. 26,27Taken together, these in vitro  studies suggest that inhalational anesthetic agents may enhance the action of β -adrenoreceptor agonists by activating adenylyl cyclase in a receptor-independent manner to increase cAMP.

Not all in vitro  studies support the aforementioned conclusion. Tanaka and Tsuchida reported an inhibitory effect of halothane and isoflurane on isoproterenol-induced vasorelaxation in endothelium-denuded rat aortic strips that was associated with a concomitant reduction in cAMP concentration. 15Moreover, these investigators observed that the anesthetic agents had no effect on vasorelaxation or cAMP accumulation in response to direct activation of adenylyl cyclase with forskolin, nor did they observe an effect of the anesthetics on the binding characteristics of β adrenoreceptors. These investigators concluded that the anesthetic agents exerted their inhibitory effects at a point in the signaling pathway distal to β-receptor activation but proximal to adenylyl cyclase activation. However, Iranami et al.  16did not observe an inhibitory effect of halothane on either isoproterenol-induced vasorelaxation or cAMP accumulation in endothelium-intact rat aortic rings. The differential findings in these in vitro  studies remain unresolved.

In vivo  studies provide an opportunity to assess the integrative response of an intervention on the measurement of interest. Our results clearly indicate that halothane potentiates the pulmonary vasodilator response to isoproterenol compared with the response measured in the same animal in the conscious state. This effect could be beneficial in the clinical setting in instances where pulmonary vasodilation is desired, particularly because a lower concentration of the β agonist may be efficacious. A limitation of in vivo  studies is that it is more difficult to identify the mechanism of action of inhalational anesthetic agents on isoproterenol-induced vasodilation. The potentiated pulmonary vasodilator response to isoproterenol is not likely the result of differential changes in systemic hemodynamics or blood gases because these changes were similar in halothane- and enflurane-anesthetized dogs, and yet the potentiated vasodilator response was only observed during halothane. The fact that the pulmonary vasodilator response to dibutyryl cAMP was not altered during either halothane or isoflurane anesthesia would suggest that the effects of the anesthetic agents are not mediated at sites distal to adenylyl cyclase activation (e.g. , inhibitory effect on cAMP phosphodiesterase activity). Although dibutyryl cAMP is widely used to investigate the β-adrenoreceptor signaling pathway, a possible confounding variable is the associated release of butyrate, which has been shown to mediate some of the actions of dibutyryl cAMP in some cell systems. 28This possibility aside, it seems reasonable to postulate that the anesthetic agents are likely to exert their effects on the β adrenoreceptors, the guanine nucleotide binding proteins (G-proteins), or the interactions between β adrenoreceptors and G-proteins.

Clinically relevant concentrations of the anesthetic agents were used in this study. Because muscle relaxants were not used, spontaneous respiration precluded a systematic study of subanesthetic concentrations of the inhalational anesthetic agents. Similarly, supraanesthetic concentrations would only have exacerbated anesthesia-induced systemic hypotension.

In summary, compared with the conscious state, the pulmonary vasodilator response to β-adrenoreceptor activation was potentiated during halothane, but not during enflurane, anesthesia. The normal pulmonary vasodilator response to dibutyrl cAMP in halothane- and isoflurane-anesthetized dogs indicates that the site in the signaling pathway for the anesthesia-induced potentiation is proximal to cAMP accumulation.