Glutathione and Glutathione-Like Sequences of Opioid and Aminergic Receptors Bind Ascorbic Acid, Adrenergic and Opioid Drugs Mediating Antioxidant Function: Relevance for Anesthesia and Abuse

Opioids and their antagonists alter vitamin C metabolism. Morphine binds to glutathione (l-γ-glutamyl-l-cysteinyl-glycine), an intracellular ascorbic acid recycling molecule with a wide range of additional activities. The morphine metabolite morphinone reacts with glutathione to form a covalent adduct that is then excreted in urine. Morphine also binds to adrenergic and histaminergic receptors in their extracellular loop regions, enhancing aminergic agonist activity. The first and second extracellular loops of adrenergic and histaminergic receptors are, like glutathione, characterized by the presence of cysteines and/or methionines, and recycle ascorbic acid with similar efficiency. Conversely, adrenergic drugs bind to extracellular loops of opioid receptors, enhancing their activity. These observations suggest functional interactions among opioids and amines, their receptors, and glutathione. We therefore explored the relative binding affinities of ascorbic acid, dehydroascorbic acid, opioid and adrenergic compounds, as well as various control compounds, to glutathione and glutathione-like peptides derived from the extracellular loop regions of the human beta 2-adrenergic, dopamine D1, histamine H1, and mu opioid receptors, as well as controls. Some cysteine-containing peptides derived from these receptors do bind ascorbic acid and/or dehydroascorbic acid and the same peptides generally bind opioid compounds. Glutathione binds not only morphine but also naloxone, methadone, and methionine enkephalin. Some adrenergic drugs also bind to glutathione and glutathione-like receptor regions. These sets of interactions provide a novel basis for understanding some ways that adrenergic, opioid and antioxidant systems interact during anesthesia and drug abuse and may have utility for understanding drug interactions.


Introduction
The purpose of this study is to investigate the possibility that opioids and adrenergic compounds bind directly to glutathione and to glutathione-like regions of opioid and adrenergic receptors, accounting for some of the observed negative effects of these compounds on antioxidant functions related to anesthesia and drug abuse. Disparate sets of observations link adrenergic, opioid, ascorbate and glutathione functions through a network of interactions that include common effects on ascorbate recycling; the formation of covalently bonded drug-protein adducts involving glutathione-like cysteine-containing sequences; drug binding to highly conserved glutathione-like regions on G

Results
Eighteen peptides ( Figure 1) were tested for their ability to bind ascorbic acid, dehydroascorbic acid, adrenergic compounds, and opioid compounds as well as a number of control compounds such as glucose and various neurotransmitters. The method used was ultraviolet spectrophotometry, which is a sensitive way to test for the binding of pairs of molecules. Binding data obtained using this method have previously been validated by direct comparison to results of nuclear magnetic resonance, circular dichroism, capillary electrophoresis, and other physicochemical methods (e.g., [51,[87][88][89][90]).
Oxidized glutathione (GSSH) has a significantly lower affinity for most of the compounds tested than reduced glutathione (GSH), presumably because the disulfide bond that forms GSSH results in decreased affinity for the sulfide side chain of the cysteine (Figure 3; Tables 1 and 2). Table 2. Summary of ultraviolet spectroscopy studies of the binding of various compounds to peptides: Glut RED = reduced glutathione; Glut Ox = oxidized glutathione; B2AR = beta 2 adrenergic receptor; D1DR = dopamine 1 receptor; H1HR = histamine 1 receptor; Dehydro Asc = dehydroascorbic acid; L-DOPA = L-3,4-dihydroxyphenylalanine.   Figure 2 demonstrates that binding affinities are generally lower to oxidized glutathione than to reduced glutathione (see Tables 1 and 2 for additional comparisons).  Figure 2 demonstrates that binding affinities are generally lower to oxidized glutathione than to reduced glutathione (see Tables 1 and 2 for additional comparisons).

Kd
Significant binding was observed between adrenergic compounds such as phenylephrine and propranolol, and even adrenergic precursors such as tyrosine and phenylalalanine, and the cysteine-containing adrenergic receptor peptide 183-185 (NCY) ( Figure 4; Table 3). No binding was observed between this peptide and various control compounds such as glucose, acetylcholine, glycine, and glutamate. Notably, this peptide had no observable affinity for opioid compounds either but did bind ascorbic acid ( Table 2).
Significant binding was observed between adrenergic compounds such as phenylephrine and propranolol, and even adrenergic precursors such as tyrosine and phenylalalanine, and the cysteinecontaining adrenergic receptor peptide 183-185 (NCY) ( Figure 4; Table 3). No binding was observed between this peptide and various control compounds such as glucose, acetylcholine, glycine, and glutamate. Notably, this peptide had no observable affinity for opioid compounds either but did bind ascorbic acid (Table 2).  Table 2 for additional data. Table 3. Overall patterns of binding between classes of compounds and receptor types. ^ Kd < or = 60 µM but >15 µM/Kd < or = 15 µM. Classes of compounds that bound to a receptor type at both cutoff values are shown bolded on a medium gray background; classes of compounds that bound to a receptor type only between 16 and 60 µM are bolded on a light gray background, while those that met neither cutoff value are normal type on a white background.   Table 2 for additional data. Binding of test compounds to the mu opioid receptor (muOPR) region 111-122, which lacks a cysteine residue and is therefore significantly unlike glutathione, was minimal or unobservable ( Figure 5). Such minimal binding was also observed for the other peptides lacking cysteine that were tested, such as muOPR 121-131 and the insulin receptor peptides (Table 1). The observed lack of significant binding was typical of all of the peptides lacking cysteine residues (see Figure 1 and Table 1).
In contrast to the relative lack of binding to the muOPR peptides 111-122 and 121-131, muOPR 132-143 displayed significant binding to ascorbic acid but not dehydroascorbic acid, and adrenergic compounds such as phenylephrine and L-DOPA ( Figure 6; see also Table 2). Notably, this peptide also bound opioid compounds such as methadone ( Figure 6) as well as morphine, naloxone and metenkephalin (Table 2). The observed lack of significant binding was typical of all of the peptides lacking cysteine residues (see Figure 1 and Table 1).
In contrast to the relative lack of binding to the muOPR peptides 111-122 and 121-131, muOPR 132-143 displayed significant binding to ascorbic acid but not dehydroascorbic acid, and adrenergic compounds such as phenylephrine and L-DOPA ( Figure 6; see also Table 2). Notably, this peptide also bound opioid compounds such as methadone ( Figure 6) as well as morphine, naloxone and met-enkephalin (Table 2).
In contrast to the relative lack of binding to the muOPR peptides 111-122 and 121-131, muOPR 132-143 displayed significant binding to ascorbic acid but not dehydroascorbic acid, and adrenergic compounds such as phenylephrine and L-DOPA ( Figure 6; see also Table 2). Notably, this peptide also bound opioid compounds such as methadone ( Figure 6) as well as morphine, naloxone and metenkephalin (Table 2).  Table 1.
Each of the drugs tested in these experiments yielded a diversity of binding curves to the various receptor peptides (Tables 1 and 2). Figure 7 provides a typical example, in this case of the binding of propranolol to peptides derived from the various classes of receptors tested here.  Table 1.
Each of the drugs tested in these experiments yielded a diversity of binding curves to the various receptor peptides (Tables 1 and 2). Overall, receptor peptides tended to fall into classes depending on their affinities for ascorbic acid, opioids and adrenergic compounds (Tables 1 and 2). Reduced glutathione had significant affinity for ascorbic acid, opioids and adrenergic compounds as well as serotoninergic compounds, but not for other neurotransmitters or glucose. Oxidized glutathione tended to have less affinity for these compounds except for dehydroascorbic acid and some of the adrenergic compounds. Mu opioid receptor peptides that contained cysteines tended to mimic the binding profile of reduced  Tables 1 and 2. Overall, receptor peptides tended to fall into classes depending on their affinities for ascorbic acid, opioids and adrenergic compounds (Tables 1 and 2). Reduced glutathione had significant affinity for ascorbic acid, opioids and adrenergic compounds as well as serotoninergic compounds, but not for other neurotransmitters or glucose. Oxidized glutathione tended to have less affinity for these compounds except for dehydroascorbic acid and some of the adrenergic compounds. Mu opioid receptor peptides that contained cysteines tended to mimic the binding profile of reduced glutathione with the exception of having little affinity for dehydroascorbic acid. Opioid receptor sequences lacking a cysteine had little affinity for any of the tested compounds, which was also typical of the insulin receptor sequences tested. Aminergic receptors tended to have one of two profiles: those that bound opioids tended to bind serotoninergic compounds but not adrenergic compounds; those that bound adrenergic compounds tended to have lower affinities for opioid compounds and to lack affinity for serotoninergic compounds. Additionally, longer peptides tended to display a higher degree of discrimination among the various ligands than did shorter compounds, suggesting that conformation becomes a discriminatory factor with increased peptide length. Notably, however, none of the peptides tested had observable affinity for glucose, glycine, acetylcholine, glutamate, or (with one histamine receptor peptide) histamine. Thus, despite the wide range of adrenergic and opioid compounds that demonstrated affinity for the adrenergic and opioid receptor peptides tested, their affinity is nonetheless class specific and similar to the affinities observed for glutathione.

Discussion
Our data suggest that a wider range of pharmacologically important interactions occur between glutathione and adrenergic, histaminergic and opioid drugs than anyone has reported before and we report here, for the first time, similar binding of these classes of drugs to glutathione-like regions of aminergic receptors that are involved in extracellular ascorbic acid recycling. These data are, however, limited to binding to isolated extracellular loop GPCR peptides, not to intact receptors, and must be interpreted with caution. We note, though, that many experiments demonstrating in vitro and in vivo effects of these drugs have been carried out (as summarized in the Introduction) and that glutathione-like activity (including recycling of ascorbic acid and the transformation of dehydroascorbic acid back into ascorbic acid) has been demonstrated for both intact GPCR and for many of the peptides tested here [48][49][50][51][52].
The results summarized in Table 1 demonstrate that ascorbic acid binds to reduced glutathione with a binding constant of approximately 60 µM and to oxidized glutathione at approximately 20 µM (with lower-affinity binding at 310 µM). Dehydroascorbic acid binds similarly. Opioids also bind with similar affinities to reduced glutathione (ca. 60 µM) but with significantly lower affinity for oxidized glutathione. Several adrenergic compounds, including phenylephrine, propranolol, amphetamine, and epinephrine, bind with even higher affinity to reduced glutathione (5 to 20 µM) as well as to oxidized glutathione. These results suggest that many of these compounds may be able to participate in the formation of the types of glutathione adducts described in the Introduction above. Other sugars (e.g., glucose) and bioactive small molecules (histamine, glycine, glutamate, etc.) do not bind appreciably to either reduced or oxidized glutathione with the exception of serotonin and to a lesser extent melatonin.
The results summarized in Table 1 further demonstrate that several of the adrenergic compounds, especially epinephrine, norepinephrine and amphetamine, have high affinity for extracellular opioid receptor peptides, with binding constants in the 1-5 µM range, but do not bind significantly to a transmembrane region (OPR 121-131). Notably, opioid binding to these peptides is significantly less than the adrenergic compounds, suggesting that the function of these extracellular regions of the opioid receptor has been optimized to facilitate allosteric control of the receptor by adrenergic compounds rather than to attract opioids to the receptor. Previous studies have demonstrated that binding of adrenergic drugs to intact OPR at these concentrations results in increased opioid binding [67,68]. Ascorbic acid has high affinity (5 µM) for one of the five OPR peptides tested (muOPR 132-143), and moderate affinity (ca. 70 µM) for another two (muOPR 38-51 and 211-226); this finding may indicate that ascorbate can also function as an allosteric modulator of the OPR, which would be consistent with the fact that it shares some common structural binding motifs with opioid compounds [66][67][68] as well as shared GPCR enhancement [49][50][51][52][58][59][60][61][62][63]65,[69][70][71][72][73][74][75][76][77]. Dehydroascorbic acid and glucose had no measurable affinity for the OPR peptides nor did any of the other control compounds except, once again, serotonin and melatonin.
The results summarized in Table 2 demonstrate that ascorbic acid and dehydroascorbic acid bind to select aminergic receptor peptides, with affinities ranging from 7 to 65 µM, which is within the physiological range for these compounds in blood serum and equivalent to, or better than, their affinity for reduced and oxidized glutathione. Opioids tend to bind to the same set of adrenergic receptor peptides as ascorbic acid and with similar binding constants, suggesting a similarity between the effects of the two sets of compounds. Notably, the binding of adrenergic compounds to these aminergic receptor peptides tends to be the inverse of the ascorbic acid-opioid binding, so that peptides that bind one set of compounds do not bind the other, and vice versa. To some extent, serotonin, melatonin and histamine mimicked the behavior of the adrenergic compounds, suggesting that these extracellular regions of the aminergic receptors are fairly non-specific in their attraction for amines in general. The other control compounds did not bind to any of the aminergic receptor peptides with measurable affinities.
In sum (Table 3), the binding of opioids and adrenergic compounds to extracellular glutathione-like peptides derived from aminergic and opioid receptors generally mimics both the specificity and the affinity that they have for glutathione itself. The receptor peptides have, however, evolved to discriminate better than glutathione between opioid and adrenergic compounds, tending to be optimized for one set or the other although significant overlaps still exist. All of the receptor peptides mimic glutathione in discriminating clearly between opioids or amines and other types of bioactive small molecules such as glucose, glycine, glutamate, histamine, etc. Notably, however, serotonin and melatonin share with the adrenergic compounds significant affinity for many of the same receptor peptides and may therefore have similar effects on opioid function.
The binding studies reported here are relevant to understanding antioxidant metabolism. Ascorbic acid is present in 50-110 µM concentrations in the blood plasma of normal human beings, who also have approximately 5-20 µM dehydroascorbic acid present [91,92]. The binding constants reported here for ascorbic acid binding to receptors range from approximately 10 to 300 µM, which translates to anything from nearly complete saturation of these binding sites on the receptors to approximately one-third saturation. Dehydroascorbate has significantly less affinity for most of the receptor peptides tested so that very little of it will be bound to the receptors under normal physiological conditions. However, while ascorbic acid has not been shown to produce adducts with glutathione, dehydroascorbic acid can [93,94]. A reasonable inference is that dehydroascorbic acid may also undergo a covalent reaction with aminergic receptor loops as well, degrading receptor function over time if there is a significant amount of ascorbate oxidation occurring. This prediction is experimentally testable and of possible clinical significance in situations in which overproduction of dehydroascorbate might be expected, such as under conditions in which ascorbic acid is overwhelmed by reactive oxygen species (ROS). ROS, in other words, may indirectly antagonize aminergic and opioid receptors by producing covalent adducts that interfere with ligand binding.
Binding of bioactive amines and aminergic drugs to glutathione itself has been known since the 1960s but little studied. Studies in the 1960s [95,96] demonstrated in vitro, in both aqueous solution and also in human blood samples, that epinephrine binds directly to glutathione and when oxidized to its adrenochrome form, epinephrine can undergo a covalent reaction producing an adduct that involves the glutathione sulfhydryl group. More recently [97], it was found that such adducts are also formed in vivo under chronic high-stress conditions (and, presumably, if a patient were treated with epinephrine pharmacologically). Amphetamines also form adducts with glutathione under physiological conditions [98][99][100] as do dopamine-related catechols [101,102], which can also form adducts with other thiol-containing proteins. The damage to the glutathione system by amphetamines, which includes depletion in dopamine/serotonin nerve rich areas of the brain [103], can be reduced or blocked by ascorbyl compounds [104], as would be predicted by the mutual complementarity of both ascorbate and opioids for amphetamine.
Since the binding of adrenergic compounds (Tables 1 and 2) to some of the cysteine-containing aminergic receptor peptides (Figure 1 and Tables 1 and 2) is on a par with their binding to glutathione, it can again be predicted that similar adducts would be formed under conditions of chronic stress or exposure to pharmacological concentrations of adrenochrome compounds. Higher-affinity binding of aminergic compounds to some of the cysteine-containing opioid receptor peptides (Table 1) suggests that these peptide regions may be more susceptible to such adduct formation in the presence of adrenochrome compounds than would be the aminergic receptors-another testable prediction. Notably, glutathione transferases catalyze the detoxication of the o-quinones aminochrome, dopachrome, adrenochrome and noradrenochrome, derived from dopamine, dopa, adrenaline and noradrenaline respectively, protecting cells from the intracellular effects of these reactive compounds [105], but there appears to be no equivalent system at work extracellularly to protect the receptors studied here. The binding of opioid compounds to some of the aminergic receptor peptides tested above also suggests the possibility of functional interactions. For example, Yang et al. [106] reported that naloxone is capable of blocking norepinephrine-induced antinociception. Such an observation makes sense only in a model in which naloxone can interact directly with adrenergic receptors, as has been demonstrated here. Such an aminergic-receptor-mediated mechanism is also consistent with enhancement of adrenergic receptor activity by naloxone and naltrexone [57][58][59][60][61][62][63][64][65].
Adduct formation may also be a consequence of opioid binding to glutathione-like regions of aminergic receptors. While adduct formation between glutathione and opioids has mainly been found to be carried out enzymatically in the liver (see Introduction), Misra and Woods [107] found that morphine can bind directly to glutathione in vitro and, in the presence of ferrous sulfate (a common iron supplement and food antioxidant), produce a covalent adduct. Thus, the binding characterized here can potentially have significant physiological consequences not just for liver and other tissues containing relevant enzymes but also throughout the body. Plasma levels of opiate drugs and their antagonists can reach concentrations of 0.5 to 5 µM or 0.1 to 1 mg/L) under therapeutic and drug use conditions [108]. Since some of the aminergic receptor peptides bound opioids with binding constants between 1 and 30 µM, some binding would be expected to most aminergic receptors and, in the case of the beta 2 adrenergic receptor, it is likely that the opioid binding region would be saturated at these opioid concentrations. Not only would such binding maximize receptor enhancement but, in the presence of any free iron, create the conditions for adduct formation, potentially dysregulating receptor function.
The observation that enkephalins bind to glutathione and glutathione-like peptides may also have important functional implications. While enkephalins are present free in plasma only at 30-60 pM or 60 ng/L [109], their functional concentration in neuronal synaptosomes has been measured to be approximately 2 µM [110] and the intrasynaptic concentration of other neurotransmitters has been calculated to range from a minimum of 2 µM to a maximum of 1.5 mM [111][112][113]. At the binding constants measured in Tables 1 and 2 above, these enkephalin concentrations translate to anything between one-fifth saturation to complete saturation of the binding sites on the aminergic receptors tested. Such binding is once again theoretically sufficient, given enough time and the presence free iron, to result in the formation of adducts with the glutathione-like sequences of the receptors present in synapses. The same reaction will be catalyzed more efficiently in the liver as discussed in the Introduction.
Binding, without adduct formation, may also modify enkephalin activity within synapses. Ogita et al. [114] have demonstrated that 100 µM met-or leu-enkephalin is able to profoundly diminish glutathione binding to various tissues in a naloxone-independent fashion. Presumably, the mechanism of this antagonism is direct binding of the enkephalins to glutathione producing an inactive complex.
Conversely, addition of glutathione to opioid preparations (including enkephalins) antagonizes mu opioid receptor activation [115]. Moreover, glutathione binds with high affinity to N-methyl-D-aspartic acid (NMDA) receptors [116], suggesting a novel mode of NMDA regulation by enkephalins and other opioids using glutathione as an intermediary.
Binding of opioid drugs to adrenergic receptors has, of course, been observed previously, as noted in the Introduction, but the site of the binding has not previously been characterized or associated with glutathione-like regions of aminergic receptors generally. Generalized binding of opioids to non-opioid receptors was observed as soon as attempts were made to identify the first opioid receptors, creating significant technical obstacles to their isolation [117]. For example, Benyhe et al. [118] documented Met5-enkephalin-Arg 6-Phe7 binding to non-opioid receptors as did Webster et al. [119] for benzomorphan. Munro et al. [120] identified one of these non-opioid receptor targets as being adrenergic receptors, observing that three kappa opioid receptor agonists bound in the mid-to high-nanomolar range to alpha-1a adrenoceptors, one acting as a potent functional enhancer. The data generated by this study verify that the enhancer site is the same as has previously been characterized for ascorbic acid enhancement of adrenergic receptors [67,68] and involves glutathione-like sequences. Thus, this study indicates that a mechanistic basis exists for the networking of opioid function and ascorbic acid/antioxidant function.
The implications of these findings for anesthesia, analgesia and substance abuse should be obvious and has recently been reviewed [121]. Any chronic use of opioids and/or adrenergic drugs is likely to deplete glutathione and impair opioid and aminergic receptor function. Fortunately, as was pointed out in the Introduction for opioids, supplemental glutathione or other cysteinyl compounds is able to prevent the deleterious effects of these drugs and are likely to be able to prevent the deleterious effects of adrenergic drugs as well. Ascorbic acid is another potentially valuable compound that may protect both glutathione and the glutathione-like regions of opioid and adrenergic receptors [122]. Indeed, oddly enough, ascorbic acid has been found to have antinociceptive effects at high doses [123][124][125], to be an opiate-sparing adjunctive addition to opioid analgesics [126][127][128], and to moderate some of the effects of withdrawal in opioid addicts [129][130][131]. These effects make sense from the perspective of ascorbic acid and opioids mimicking each other's binding patterns to glutathione and glutathione-like regions of opioid and aminergic receptors. The heterodimerization of these receptors [78,79] links their activity and the fact that ascorbic acid can bind to both creates a further mechanism for mediating their crosstalk [121].
Whether the impact of anesthesia followed by pain management has the same effects on the antioxidant system as abuse of opioids, amphetamines and other drugs affecting these systems is an issue that needs further research since such effects may be assumed to be a function of drug dosage, means of delivery, length of use and the degree to which tolerance develops. In general, it may be assumed that the use of such drugs for anesthetic purposes will have fewer effects on the antioxidant system than their chronic abuse. For example, typical opioid dosages for anesthesia result in blood plasma concentrations ranging from approximately 1 to 100 nM in blood plasma following oral or intravenous delivery and are typically kept at these levels for only a few days or weeks [132,133]. However, intrathecal delivery of morphine results in transient cerebrospinal fluid concentrations of approximately 25 µM [134], which might result in significant antioxidant effects and covalent reactions despite the brief tissue exposure. Surprisingly, occasional abusers of oral morphine achieve blood concentrations not dissimilar to anesthetized patients, ranging from approximately 6 to 30 nM in blood plasma [135], though their exposure to the drug may last longer and may produce more significant effects on their antioxidant function. Treatment-resistant, chronic opiate abusers are characterized by much higher plasma concentrations ranging from 120 nM to 1 µM [136] and may be presumed to have correspondingly more serious antioxidant deficits. Opioid abusers who must undergo surgery need to achieve concentrations of opioids in the same range as treatment-resistant, chronic abusers [137], underscoring the fact that patients with different drug histories may have different physiological responses to identical drug dosages.
In sum, the antioxidant system, consisting of both intracellular glutathione and extracellular glutathione-like opioid and aminergic receptor regions, plays a fundamental role in mediating opioid and adrenergic drug activity. Ascorbic acid (vitamin C), through its binding to and recycling by glutathione and glutathione-like receptor regions on GPCR, plays a mediating role in all aspects of that opioid and aminergic activity.

Opioid Receptor Peptide Synthesis and Preparation
The peptides listed in Figure 1 were synthesized to at least 95% purity (as determined by mass spectrometry) by RS Synthesis (Louisville, KY, USA). Each of these receptor peptides were made into individual stock solutions with a concentration of 1 mg/mL in pH 7.00 phosphate-buffered saline solution (Fischer Scientific, Hampton, NH, USA). For each individual test, the concentration of the stock solution was calculated and then diluted to a 10 µM solution. Each of these peptide solutions was tested for binding with the various opioid, adrenergic and ascorbate-like compounds and controls such as histamine and acetylcholine (Sigma-Aldrich, St. Louis, MO, USA), as indicated in Tables 1 and 2. The opioid, adrenergic, ascorbyl and control compounds were made up at 1.0 mM solutions and then diluted to 5 µM concentrations for use in the peptide binding experiments.

Peptide Binding Test Methods
After the solutions were made, a 96-well quartz crystal plate was prepared to be run through the spectrophotometer at room temperature (ca. 24 • C). The plate was set up to have the absorbance of each of the adrenergic compound dilutions measured on their own, and with each receptor peptide. The absorbance of each receptor peptide without the presence of the adrenergic compound was measured as well. The absorbance of each well was measured at every 10 nm increment from 190-260 nm. The maximum absorbance that can be measured was set to be 4. Each well had 200 µL of solution, so if the absorbance of one component was being measured, it was diluted by 1 2 with phosphate buffer. Thus, the final concentration of receptor peptide in each well varied from 0.5 mM to 0.05 pM while the other compound was held fixed at 5 µM.
Spectrophotometry (SPECTRAmax plus scanning spectrophotometer with the SOFTmax PRO program) was used to measure the binding between opioid receptor peptides and opioid, adrenergic and control compounds. The method of calculating binding constants was to measure the absorbance (minus the absorbance of the buffer) of 100 µL of each compound by itself diluted by 100 µL of buffer at a given wavelength (in this case at 200 nm) at the range of concentrations utilized in the experiments. The average of the resulting values is calculated. (If the experiments are run properly, there is very little variance-a couple of percent. If there was more variance than that, the experiment was re-run.) An "expected" value is calculated following Beer's law, by adding the absorbances of each pair of compounds (peptide plus small molecule). The actual chemical combinations of these molecules was then run under identical conditions.

Data Analysis
The absorbance for the phosphate buffer is subtracted for each well before any calculations are performed. The difference between the "expected" and actual values was determined at each concentration of the varied compound. This difference was then plotted against the concentration of the varied compound. If there was significant binding, an S-shaped curve resulted. The binding constant was approximated by using the inflection point of the binding curve (i.e., the point at which the dissociation and association of the two molecules was equal). If there was no significant binding, no curve resulted or the curve did not inflect over the range of concentrations utilized in the experiments, in which case the binding constant was greater than could be calculated from the experimental conditions. All data were analyzed using Microsoft Excel.
As noted above, all combinations were in duplicate and every combination was repeated in independent experiments at least twice, yielding the same Kd. Because the resulting absorbances are the differences between the independently measured absorbances, it is not possible to calculate error bars or to perform statistics on the resulting values.

Conflicts of Interest:
The authors declare no conflict of interest.