Letter Cite This: ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
pubs.acs.org/chemneuro
6‑Monoacetylmorphine (6-MAM), Not Morphine, Is Responsible for the Rapid Neural Effects Induced by Intravenous Heroin David Perekopskiy and Eugene A. Kiyatkin*
Downloaded via UNIV OF SOUTHERN INDIANA on July 18, 2019 at 22:32:10 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
Behavioral Neuroscience Branch, National Institute on Drug AbuseIntramural Research Program, National Institutes of Health, DHHS, 333 Cassell Drive, Baltimore, Maryland 21224, United States ABSTRACT: Heroin rapidly enters the CNS but is quickly metabolized into 6monoacetylmorphine (6-MAM) and then morphine. Although morphine is often thought to mediate heroin’s neural effects, pharmacokinetic data question this view. To further understand the effects of heroin and its metabolites, oxygen sensors were used to examine changes in nucleus accumbens (NAc) oxygen levels. Heroin, 6-MAM, and morphine were all administered intravenously at two human-relevant doses (0.25 μmol/kg and 0.98 μmol/kg) in freely moving rats. Intravenous heroin induced a biphasic change in NAc oxygen, with a decrease resulting from respiratory depression and an increase resulting from cerebral vasodilation. 6-MAM caused similar but more rapid and slightly weaker effects than heroin. The stronger response to heroin can be primarily attributed to heroin’s permeability and metabolism resulting in more 6-MAM in the brain. Morphine only induced weak increases in NAc oxygen. Therefore, it appears that 6-MAM is the major contributor to acute neural effects induced by iv heroin. KEYWORDS: opioid, metabolism, voltammetry, vasoconstriction, vasodilation, brain hypoxia, respiratory depression, diacetylmorphine
H
eroin, due to high lipophilicity, rapidly enters the CNS1 but is quickly broken down into metabolites and byproducts. While heroin is sequentially metabolized into 6monoacetylmorphine (6-MAM) and then to morphine,2,3 it is often believed that morphine is responsible for the neural effects of heroin.4,5 However, recent assessments of pharmacokinetics of heroin and its primary metabolites in the blood and brain extracellular space questioned this view. When 3 μmol/kg of heroin was injected intravenously, brain levels of heroin assessed by microdialysis in awake rats rapidly peak (1− 2 min, 1.5 μM) and quickly decrease with a t1/2 of 0.9 min, becoming undetectable at 5 min postinjection. 6 The pharmacokinetics of 6-MAM mimicked that of heroin, with equally rapid entry into brain tissue peaking at 4.3 min, at a concentration four times as high as heroin (∼6 μM). In contrast, the levels of morphine after intravenous (iv) heroin injection increased slowly, peaking at lower levels (0.8 μM) and much later times (21 min). Timing and concentration values are critical because the most profound psychoemotional (intense rush) and physiological effects of heroin occur within minutes after its iv administration.7−9 Among these rapid and potentially life-threatening effects of heroin is respiratory depression, which could lead to significant brain hypoxia. As shown by using oxygen sensors coupled with amperometry, iv heroin at low, behaviorally relevant doses (0.1 mg/kg) decreased brain oxygen levels. This effect greatly progressed with increases in dose.10 The decrease in oxygen levels occurred with short latencies and peaked between 1.5 and 5 min postinjection. If we are to compare the effects of heroin to pharmacokinetic data, 6-MAM seems to correlate This article not subject to U.S. Copyright. Published XXXX by the American Chemical Society
best with these effects. Therefore, 6-MAM, not morphine, appears to be responsible for the initial, rapid neural effects of heroin. This study was designed to test this hypothesis by directly assessing the pattern and time course of brain oxygen responses induced by iv heroin and its two pharmacologically active metabolites, 6-MAM and morphine, injected at equimolar doses in freely moving rats. Similar to our previous studies,10,11 oxygen recordings were conducted in the nucleus accumbens (NAc), a critical structure for sensorimotor integration and a part of motivation−reinforcement circuit.12,13 Heroin was injected at two human relevant doses (0.1 and 0.4 mg/kg or 0.25 and 0.98 μmol/kg), which served as molar standards for 6-MAM and morphine. Heroin responses were then compared to 6-MAM and morphine to further elucidate mechanisms of heroin’s neural effects. To capture the rapid oxygen dynamics, data were analyzed with both slow, minutescale and rapid, second-scale time resolution.
■
RESULTS AND DISCUSSION In contrast to most psychoactive drugs, which induce their effects via direct interaction with centrally and peripherally located neural substrates, heroin is rapidly metabolized in vivo into 6-MAM and then to morphine, both of which have high affinity to μ-opioid receptors.14 While morphine is often Received: May 24, 2019 Accepted: July 3, 2019 Published: July 3, 2019 A
DOI: 10.1021/acschemneuro.9b00305 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
ACS Chemical Neuroscience
Letter
Figure 1. Mean (±SEM) changes in NAc oxygen levels (in percent vs preinjection baseline, which is set to 100%) induced by iv heroin (A, at 0.1 mg/kg; B, at 0.4 mg/kg) in awake, freely moving rats. Left graphs show oxygen changes analyzed with low, 1-min time resolution, and right graphs show oxygen changes analyzed with high, 4-s time resolution. Filled symbols show values significantly different from initial baseline.
netics due to the rapid metabolism of heroin in vivo. Though our experiments were conducted in rats, it appears that kinetics of heroin metabolism in humans share a basic similarity to that found in rats.19 Therefore, our approach provides highresolution analysis of functionally and clinically important neural effects of heroin and its two metabolites under human relevant conditions. Data below show the changes in NAc oxygen levels induced by iv heroin, 6-MAM, and morphine in two equimolar doses (0.25 μmol/kg and 0.98 μmol/kg) in awake, freely moving rats. Data were analyzed with both slow, 1-min, and rapid, 4-s, time resolution. These data were obtained in 8 rats (24 recording sessions), in which the locations of sensor tips were confirmed histologically. Basal levels of NAc oxygen varied in different rats between 12.28 and 36.62 μM (mean ± SEM = 24.80 ± 0.91 μM) and typically maintained at relatively stable levels during long-term recording. The Effects of Heroin (Control). Consistent with our previous findings,10 heroin injected at a low, behaviorally relevant dose (0.25 μmol/kg) induced a biphasic effect on NAc oxygen levels (F15,1365 = 11.42, p < 0.0001; Figure 1A). Oxygen levels rapidly decreased after the injection, reached nadir at the second minute (93% of baseline), and then increased above baseline, peaking at 8 min postinjection (120% of baseline). A biphasic oxygen response also occurred after the higher dose heroin injection (F7,637 = 17.42, p < 0.0001; Figure 1B). In this case, NAc oxygen levels decreased equally rapidly and reached nadir at the same time as with a smaller dose, but the amplitude of decrease and its duration were much larger (peaked at 48% of baseline and only crossed the baseline at ∼11 min postinjection). Similar to that with a smaller dose, oxygen levels then increased above baseline, peaking at 24 min
viewed as the primary heroin metabolite responsible for its psychoactive and physiological effects, an alternative hypothesis suggests the critical role of 6-MAM. This hypothesis was first introduced many years ago based on measurements of heroin and its metabolites in animal carcasses and various organs,15 and it received substantial support in experiments with microdialysis measurements of heroin and its metabolites in blood and brain tissue.6,16,17 This study was designed to test this hypothesis by measuring NAc oxygen responses induced by heroin and its two primary active metabolites at equimolar doses. A few aspects of this study make it novel and add to the knowledge about heroin’s effects. First is our use of oxygen sensors coupled with amperometry. Brain oxygen is an important homeostatic parameter that could be valuable for assessing respiratory depression, a prominent rapid effect typical to opioid drugs.18 As shown previously10 and confirmed in this study, the NAc oxygen response induced by iv heroin is principally biphasic, depending on the balance between two opposite factors: decrease in blood oxygen levels due to respiratory depression and increase in cerebral blood flow due to the drug’s central effects. With this technology, electrochemical measurements can be conducted with second-scale temporal resolution, thus allowing us to assess the dynamics of oxygen fluctuations in real time. This method, unlike locomotion, makes it possible to analyze the effects caused by heroin and its metabolites in a more direct way. Second, with our experimental protocol we tried to mimic human conditions, by using awake, freely moving rats and iv drug delivery at low, human-relevant doses under stress- or cue-free conditions. In contrast to iv drug delivery, subcutaneous injections provide different rates of metabolism and diffusion that will affect drug pharmacokiB
DOI: 10.1021/acschemneuro.9b00305 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
ACS Chemical Neuroscience
Letter
Figure 2. Mean (±SEM) changes in NAc oxygen levels (in percent vs preinjection baseline, which is set to 100%) induced by iv 6-MAM (A, at 0.103 mg/kg; B, at 0.411 mg/kg) in awake, freely moving rats. Left graphs show oxygen changes analyzed with low, 1-min time resolution, and right graphs show oxygen changes analyzed with high, 4-s time resolution. Filled symbols show values significantly different from initial baseline. Black solid lines in each graph show changes induced by iv heroin at equimolar dose.
doses (F15,1140 = 4.041 and F6,456 = 13.78, both p < 0.0001; Figure 2). After injection, 6-MAM oxygen levels began to decrease below 100% of baseline at 22 and 8 s for the small and large dose, respectively (54 and 24 s for heroin at small and large doses). Though heroin is metabolized quickly, shorter latencies would be expected if 6-MAM was the primary chemical to induce the rapid neural effects of heroin. Also, NAc oxygen responses to 6-MAM are qualitatively similar to that of heroin, but 6-MAM was less effective to decrease NAc oxygen levels than heroin at the same dose. The Effects of Morphine and Its Differences versus Heroin. The effects of morphine, the second metabolite of heroin, differed drastically from those of heroin. In both doses tested, morphine only slightly increased NAc oxygen levels (F15,1365 = 2.77, p < 0.01, and F7,637 = 0.70, NS; Figure 3). These effects were weak in terms of magnitude and transient, with a significant increase within a couple of minutes following the injection. When data were recalculated with high time resolution, the effect was stronger (F15,1140 = 3.02, p < 0.0001; F7,532 = 1.88, p < 0.05; Figure 3) at both doses, but only lowmagnitude increases were seen. These increases occurred rapidly after the injection, and they were approximately equal in magnitude at both doses. Therefore, the NAc oxygen response induced by morphine is very weak and different than that induced by heroin at the same equimolar dose. When analyzing the data, one glaring issue arises: why is the effect of 6-MAM weaker than that of heroin? If 6-MAM is the primary metabolite inducing rapid neural effects, it could be assumed that its effects should be greater or equivalent to those induced by heroin. There can be a few reasons that could explain these unexpected differences. First, it may be thought
(120% of baseline) and slowly decaying toward baseline thereafter. Since heroin-induced changes were rapid and the most profound changes occurred immediately after the injection, to capture these rapid dynamics data were analyzed with high, 4-s time resolution for the first 5 min postinjection (Figure 1). At each dose, heroin induced a significant decrease in oxygen levels (0.1 mg/kg, F15,1140 = 29.95, and 0.4 mg/kg, F7,532 = 28.80, both p < 0.0001). In both cases, the decrease occurred with definite onset latency, which was longer for the low-dose injection (∼56 s) than for the high-dose injection (∼24 s). While the nadir of oxygen decrease was similar for both doses (92 and 104 s), the amplitude of decrease was clearly larger for the higher heroin dose (37% of baseline vs 91%). At this higher dose, the decrease was sustained for a much longer time, while oxygen began to increase quicker at the low dose. The Effects of 6-MAM and Its Differences versus Heroin. 6-MAM, the first metabolite of heroin, injected at the lower dose increased NAc oxygen levels (F15,1365 = 11.61, p < 0.0001; Figure 2A), peaking at 6 min (118% of baseline) and slowly descending to baseline. Though there was no initial oxygen drop, the concentration curves of 6-MAM and heroin were almost superimposed, except for the few minutes following the injection. At a higher dose, 6-MAM induced a biphasic oxygen change (F6,546 = 10.96, p < 0.0001; Figure 2B) with an initial decrease that was weaker (70% of baseline) than that for heroin (48% of baseline). One large dose 6-MAM recording was not included in the data set because of electrical artifacts. When data were analyzed with high temporal resolution, the effect of 6-MAM on NAc oxygen levels was significant at both C
DOI: 10.1021/acschemneuro.9b00305 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
ACS Chemical Neuroscience
Letter
Figure 3. Mean (±SEM) changes in NAc oxygen levels (in percent vs preinjection baseline, which is set to 100%) induced by iv morphine (A, at 0.094 mg/kg; B, at 0.378 mg/kg) in awake, freely moving rats. Left graphs show oxygen changes analyzed with low, 1-min time resolution, and right graphs show oxygen changes analyzed with high, 4-s time resolution. Filled symbols show values significantly different from initial baseline. Black solid lines in each graph show changes induced by iv heroin at equimolar dose.
tissue into 6-MAM. The combination of diffusion and metabolism results in a higher concentration of 6-MAM in the brain after heroin injection than after the injection of 6MAM alone. This then induces larger neural effects such as brain hypoxia as shown by our data. This conclusion is also supported by a study where rats continued to self-administer heroin even with 6-MAM antibodies injected systemically.21 Though pharmacokinetic data revealed that heroin is primarily metabolized in the periphery, this data was obtained with subcutaneous injections which would exhibit different kinetics of metabolism and diffusion than iv injections.22 Although this is our central reasoning, ideas such as 6-MAM metabolizes to morphine, which has weaker BBB permeability,1 could also affect the neural responses. Lastly, though heroin is rapidly metabolized and has a 6-fold weaker affinity to opioid receptors, it may still play a part in rapid neural effects of this drug before its degradation. It is true that the BBB permeability of morphine is weaker than that of heroin and 6-MAM,1 and this could explain the much weaker oxygen responses induced by iv morphine injection in this study. While morphine has been shown to induce respiratory depression, this effect occurs at much higher drug doses.23 We do not discount that these effects could be much stronger if morphine appears in the brain as a product of heroin’s metabolism. However, microdialysis data argue against the idea that morphine induces rapid neural effects such as brain hypoxia. When the levels of 6-MAM peaked at ∼4 min postinjection at 6 μM, morphine levels were at ∼0.5 μM, that is, 12 times less than 6-MAM.6 Since 6-MAM and morphine have been found to have equivalent affinities to μ-opioid
that heroin itself is the primary effector inducing heroin’s rapid neural effects. This possibility seems very unlikely because the affinity of heroin to μ-opioid receptors is 6-fold lower than that for 6-MAM.14 Also, the metabolism of iv-delivered heroin is too fast, peaking in the brain’s extracellular space at 1−2 min at ∼4-fold lower concentration than that of 6-MAM.6 Another possibility is that morphine is the primary effector of rapid neural effects. Though 6-MAM and morphine have the same affinity to μ-opioid receptors, microdialysis data seems to disprove this variant because of the slow appearance of this substance in the brain tissue, also peaking at concentrations 7fold lower than those of 6-MAM. Rapid neural effects such as brain hypoxia are seen to peak around 1.5−5 min depending on dose. Only 6-MAM demonstrates both high affinity to opioid receptors and rapid and strong increases in concentrations in brain tissue, which correlate with the rapid onset of brain hypoxia caused by heroin. Additionally, 6-MAM at both doses induced shorter latencies than heroin to decrease oxygen levels. This would be expected if 6-MAM induced these rapid neural effects. These considerations bring us to the reasons why we believe 6-MAM has a weaker oxygen response than heroin. The study of Gottas17 may provide a clue for this unexpected result. In this study, heroin and its two metabolites were injected iv into rats at equimolar doses, and their levels were measured in the brain. They found that in heroin-injected rats, 6-MAM levels peaked at a 40% higher level than in the 6-MAM-injected rats. Heroin is a more lipophilic drug than 6-MAM;20 thus this allows it to cross the blood−brain barrier (BBB) more easily than 6-MAM. Therefore, it appears that heroin after its iv injection crosses the BBB and then is metabolized in brain D
DOI: 10.1021/acschemneuro.9b00305 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
ACS Chemical Neuroscience
Letter
receptors,14 even if morphine has an effect, it would exert a significantly smaller influence than that of 6-MAM. In conclusion, our data support the idea of heroin’s first metabolite, 6-MAM, being primarily responsible for rapid neural effects of this drug. While the NAc oxygen response was a parameter examined in this study, this conclusion could possibly be applicable to other rapid physiological and psychoemotional effects of heroin (including acute euphoria or rush), which occur almost immediately after iv heroin injection. While the NAc oxygen response induced by heroin also depends on neuronal effects and subsequent changes in the tone of cerebral vessels, rapid, dose-dependent decreases in oxygen levels reflect respiratory depression, one of the deadliest effects of heroin and other highly potent opioids. This effect appeared within 0.5−1.0 min and peaked between 1.5 and 5 min after the injection. 6-MAM when injected at equimolar doses induced similar responses, which were more rapid but slightly weaker than those of heroin. The stronger response of heroin can be primarily attributed to heroin’s permeability and metabolism resulting in larger doses of 6MAM in the brain. Morphine, when injected at equimolar doses, failed to decrease brain oxygen levels, inducing only small transient increases. While all these data lead to the conclusion that 6-MAM, not morphine, is responsible for rapid neural effects induced by iv heroin, this does not deny the role of morphine derived from heroin in mediating the delayed and relatively prolonged sedative, euphoric (“high”), and analgesic effects of heroin.
■
recorded at 1 s intervals, using PAL software (version 1.5.0, Pinnacle Technology). Oxygen sensors were calibrated at 37 °C by the manufacturer (Pinnacle Technology) according to a standard protocol described elsewhere.25 The sensors produced linear current changes with increases in oxygen concentrations within the wide range of previously reported brain oxygen concentrations (0−40 μM). Substrate sensitivity of oxygen sensors varied from 0.584 to 1.778 nA/1 μM (mean = 1.03 nA/1 μM). Oxygen sensors were also tested by the manufacturer for their selectivity toward other electroactive substances, including dopamine (0.4 μM) and ascorbate (250 μM), none of which had significant effects on reduction currents. Experimental Procedures. At the beginning of each experiment, rats were lightly anesthetized (
