Measuring naloxone antagonism of discriminative opioid stimulus

The numerous studies of opioids as discriminative stimuli, beginning in 1971, have shown specificity, similarity of several opioids, differences in potency (fentanyl greater than heroin greater methadone greater than morphine), and antagonism by naloxone and naltrexone. The discriminative opioid stimulus is differentiated from those of other classes of drugs, such as sedatives and anxiolytics. Greater potency of the opioid stimulus has been found in rats after subcutaneous (s.c.) than intraperitoneal administration. The discriminative opioid stimulus and its antagonism by naloxone or naltrexone have been demonstrated in rats, squirrel monkeys, gerbils, and pigeons. A few studies have quantified the competitive agonist-antagonist interaction at the receptor by calculating the pA2, which reflects the dose of the antagonist that requires doubling the agonist dose to obtain the original agonist response. The pA2 for naloxone is the same in groups of rats trained to discriminate different doses of morphine (1, 2, or 4 mg/kg s.c.) from saline. Higher pA2 values in tests after fentanyl and methadone than after heroin and morphine in rats trained to discriminate fentanyl (0.04 mg/kg s.c.) from saline reflect greater susceptibility of the synthetic than the natural exogenous opioids to antagonism by naloxone. Different pA2 values are usually interpreted as indicating differences among populations of receptors.     

Enhancement of opiate-induced depressions in serum luteinizing hormone levels by the prior administration of naloxone in the male rat

The effects of naloxone pretreatment on opiate agonist-induced depressions in serum luteinizing hormone (LH) levels were examined in male rats. Our results demonstrated a pronounced enhancement of morphine's actions 6 hours after the administration of naloxone (0.5 mg/kg). This effect was characterized by a 10 fold reduction in the ED50 (1.26 mg/kg versus 0.13 mg/kg in saline- and naloxone-pretreated rats, respectively) and much greater depressions in serum LH levels at each dose of morphine. The actions of naloxone were not confined to morphine, since similar increased potencies were found for opioid agonists with selectivity for a variety of opioid receptor subtypes. Because naloxone did not alter the uptake of subsequently administered morphine into brain, our results cannot be explained on the basis of an increased availability of the agonist. Rather, it appears that naloxone pretreatment induces a change in the sensitivity of those receptors involved in the effects of opioid agonists on LH.     

Blockade of morphine supersensitivity by an antisense oligodeoxynucleotide targeting the delta opioid receptor (DOR-1)

An antisense oligodeoxynucleotide (ODN) targeting 20 bases of the coding sequence of the cloned delta opioid receptor (DOR-1), a mismatched ODN (different from the antisense ODN at 4 bases) or saline was administered to 3 groups of CD-1 mice implanted with naltrexone pellets (7.5 mg) for 7 days. Morphine supersensitivity (i.e., increased potency as defined by decreased morphine ED₅₀ values) was observed 24 h after pellet removal (day 8) in mice treated with saline or mismatch ODN, but not in antisense ODN treated mice. Antisense ODN alone had no effect on basal nociceptive thresholds or morphine analgesia but reduced the analgesic potency of the delta₂ opioid agonist [D-Ala²]deltorphin II. These data suggest that the delta₂ opioid receptor system participates in the adaptive changes contributing to increased morphine potency following chronic naltrexone treatment.     

Dissociation of morphine withdrawal diarrhea and jumping in mice by the peripherally selective opioid antagonist SR 58002 C

Mice were rendered physically dependent by repeated administration of morphine, 25 mg/kg s.c., 5 times daily for 4 days, and on the 5th day, 2 h after the last morphine dose, they were challenged with a s.c. injection of either naloxone, 25 mg/kg, or the peripherally selective opioid antagonist SR 58002 C (N-methyl levallorphan mesilate), 75 mg/kg. Naloxone provoked jumping and diarrhea in all the animals; mice challenged with SR 58002 C presented no significant jumping but a high frequency of withdrawal diarrhea. When naloxone, 12 mg/kg, or SR 58002 C, 50 mg/kg, were given s.c. in combination with repeated morphine as above, mice which had received naloxone with morphine presented virtually no diarrhea or jumping upon naloxone challenge; those repeatedly treated with morphine plus SR 58002 C were substantially protected from naloxone-precipitated diarrhea, but not jumping. These results further support the remarkable selectivity for peripheral opioid receptors of SR 58002 C, even after repeated high-dose treatment, and are strongly consistent with the primary role of a local intestinal mechanism in the development and expression of opioid withdrawal diarrhea in mice. The in vivo dissociation of central and peripheral components of dependence on morphine is illustrated, apparently for the first time.     

Effects of acute and chronic morphine and narcotic antagonists on brain energy metabolism

Morphine sulphate (4 mg/kg to 32 mg/kg) produced a dose-dependent decrease in brain malate as antinociception increased. Decreased brain malate persisted 72 hours after implantation of morphine pellets by which time mice had become tolerant to antinociception. This finding suggests that malate decrease, unlike changes of other metabolites in other studies, might not be simply a result of general metabolic changes. Malate change as well as antinociception was prevented by prior injection of naloxone (3.0 mg/kg) or naltrexone (0.6 mg/kg) in acute experiments. Malate decrease in pelleted mice was no longer present if withdrawal was produced by naloxone or naltrexone in mice implanted with morphine pellets for 72 hours. Brain P-creatine was elevated in all mice implanted with morphine pellets even after withdrawal, thus, apparently, representing a more generalized effect than malate change.     

Morphine intake and the effects of naltrexone and buprenorphine on the acquisition of methamphetamine intake

Some common genetic factors appear to influence risk for drug dependence across multiple drugs of abuse. In previous research, mice that were selectively bred for higher amounts of methamphetamine consumption, using a two‐bottle choice methamphetamine drinking procedure, were found to be less sensitive to the locomotor stimulant effects of morphine and of the more selective μ‐opioid receptor agonist fentanyl, compared to mice that were bred for low methamphetamine consumption. This suggested that μ‐opioid receptor‐mediated pathways may influence genetic risk for methamphetamine consumption. We hypothesized that these differences in opioid sensitivity would impact opioid intake in the methamphetamine drinking lines and that drugs with μ‐opioid receptor activity would impact methamphetamine intake. Consumption of morphine was examined in 2, two‐bottle choice studies, one that compared morphine to quinine consumption and another that used a saccharin fading procedure. Next, naltrexone (0, 0.5, 1, 2, 5, 10 and 20 mg/kg), a μ‐opioid receptor antagonist, and buprenorphine (0, 1, 2 or 4 mg/kg), a μ‐opioid receptor partial agonist, were each examined for their effects on the acquisition of methamphetamine consumption. Low methamphetamine drinking mice consumed more morphine compared to high methamphetamine drinking mice. Naltrexone did not alter methamphetamine consumption in either selected line; however, buprenorphine reduced methamphetamine intake in the high methamphetamine drinking line. These data show that greater sensitivity to opioids is associated with greater opioid intake and warrant further investigation of drugs with μ‐opioid receptor‐specific agonist activity in genetically determined differences in methamphetamine consumption.     

In vivo characterization of the opioid antagonist nalmefene in mice

AimsThe current study assessed the in vivo antagonist properties of nalmefene using procedures previously used to characterize the opioid antagonists naloxone, naltrexone, 6β-naltrexol and nalbuphine.Main methodsICR mice were used to generate antagonist dose–response curves with intraperitoneal (i.p.) nalmefene against fixed A₉₀ doses of morphine in models of morphine-stimulated hyperlocomotion and antinociception. Additional dose–response curves for antagonist precipitated opioid withdrawal were run in mice treated acutely (100 mg/kg, s.c., ? 4 h) or chronically (75 mg pellet, s.c., ? 72 h) with morphine. Comparisons were made between antagonist potency and degree of precipitated withdrawal.Key findingsNalmefene produced dose- and time-related antagonism of morphine-induced increases in locomotor activity with a calculated ID₅₀ (and 95% confidence interval) of 0.014 (0.007–0.027) mg/kg. Nalmefene produced rapid reversal of morphine-induced locomotor activity (5.1 min for 50% reduction in morphine effect). A 0.32 mg/kg dose of nalmefene produced blockade of morphine-induced antinociception in the 55 °C tail-flick test that lasted approximately 2 h. Nalmefene was able to potently precipitate withdrawal in mice treated acutely or chronically with morphine.SignificanceThese results demonstrate that nalmefene is similar to naloxone and naltrexone with respect to its in vivo pharmacology in mice. Specifically, nalmefene produces potent antagonism of morphine agonist effects while precipitating severe withdrawal. The compound has a slower onset and longer duration of action compared to naloxone and naltrexone. The data allows for a more complete preclinical comparison of nalmefene against other opioid antagonists including the putative opioid neutral antagonist 6β-naltrexol.     

Teratogenic and postnatal developmental studies of morphine in Sprague-Dawley rats

The teratogenic and postnatal developmental effects of morphine exposure during pregnancy were studied in Sprague-Dawley rats in three separate experiments using chronically implanted osmotic minipumps in order to avoid respiratory depression. In the first experiment, the teratogenic effects of three different morphine dosages were studied: a low dose (10 mg/kg/day), an intermediate dose (35 mg/kg/day), and a high dose (70 mg/kg/day). On day 5 of gestation, osmotic minipumps that deliver their contents at a constant rate for 15 days were implanted subcutaneously on the back of the rats. On day 20 of gestation, cesarean sections were performed, reproductive indices were determined, and fetuses were examined externally and then preserved for subsequent visceral and skeletal examinations. The pregnancy rate was significantly reduced at the intermediate and high doses to 57% and 6%, respectively (control, 83%). No teratogenic effects were observed at any dosage, but growth retardation was present in the intermediate-dose group. In the second experiment, postnatal survival of the offspring of dams treated with either normal saline, morphine (35 mg/kg/day), or the synthetic opioid, fentanyl (500 micrograms/kg/day) were studied. Offspring of morphine-treated dams had a significantly higher mortality rate, which peaked at 56% within 2 days. No effect was seen after fentanyl treatment. In the third experiment, pups of morphine-treated dams were cross-fostered by saline-treated dams; the postnatal mortality in offspring of morphine-treated dams remained high (62%). Our results indicate that doses of morphine up to 35 mg/kg/day delivered by osmotic minipumps are not teratogenic in rats but cause other adverse fetal effects that result in increased postnatal mortality.     

Modification of the effects of naloxone in morphine-dependent mice

Mice were rendered dependent on morphine by mixing morphine with their food (2 mg/g) for three days. Increasing doses of naloxone precipitated dose-dependent withdrawal reactions such as weight loss and jumping. These withdrawal reactions were antagonized by morphine pretreatment. Effects of morphine, such as increased locomotor activity, inhibition of intestinal transport, and analgesia were antagonized by naloxone in both non-dependent and dependent subjects. The antagonist actions of naloxone were increased in dependent subjects; lower doses of naloxone were sufficient to antagonize effects of morphine. The present results confirm earlier studies indicating that precipitation of withdrawal can be antagonized by morphine pretreatment suggesting that withdrawal reactions are due to actions of naloxone at the same receptor at which opioid agonists act. The increased antagonist potency of naloxone in dependent subjects extends earlier results obtained with analgesic effects to several other agonist effects of morphine and is consistent with the interpretation that exposure to an opioid agonist induces a change in the conformation of opioid receptors.     

Effects of opioid antagonists and their quaternary analogs on temperature changes in morphine-dependent rats

Subcutaneous administration of three opioid antagonists; naloxone, naltrexone and nalmefene, produced a significant rise in tail skin temperature and a subsequent fall in rectal temperature in morphine dependent rats. However, subcutaneous administration of equimolar concentrations of the quaternary derivatives of these opioid antagonists (naloxone methobromide, naltrexone methobromide and n-methylnalmefenium iodide) failed to produce any significant alterations in either tail skin or rectal temperatures in the morphine dependent rat. At doses of naloxone methobromide 6 to 9 times greater than naloxone, there was a slight reduction of rectal temperatures with no significant elevation of skin temperature. However, the fall in rectal temperature was still significantly less than that achieved with administration of naloxone. When each of these six agents were administered centrally (20 micrograms/5 microliter, icv) in the morphine dependent rat, similar increases in tail skin temperature and decreases in rectal temperature were observed. These temperature changes were similar to those observed following systemic administration of the opioid antagonist. Previously, we have suggested that acute withdrawal in the morphine-dependent rat may serve as an animal model for the mechanism of the menopausal hot flush. Collectively, these results suggest that the temperature changes associated with morphine-withdrawal in our rat model for studying the mechanisms of the menopausal hot flush are centrally mediated.     

Analgesic activity of spiradoline mesylate (U-62,066E), a kappa opioid agonist in mice

The present study was undertaken to evaluate the analgesic potency of spiradoline mesylate, a k(kappa) opioid agonist, in comparison with that of morphine, by hot plate, tail-pinch and acetic acid-induced writhing assay. The ED50 values of spiradoline in hot plate, tail-pinch and acetic acid-induced writhing assay were 0.46, 0.26 and 0.20 mg/kg, respectively. The analgesic potency of spiradoline was 1.5-7.0 times higher than that of morphine. Repeated treatment with spiradoline as well as morphine developed tolerance to the analgesic effect in hot plate assay. In mice developed tolerance to one analgesic, response to the other analgesic did not alter compared to saline-treated mice. Single administration of spiradoline (1.5 and 3 mg/kg, s.c.) did not inhibit morphine-induced analgesia. These results suggest that spiradoline has more potent analgesic activity than morphine, presumably mediated through stimulation of receptors different from morphine.     

Regulation of opioid peptides on the release of arginine vasotocin in the hen

Arginine vasotocin (AVT), an avian neurohypophysial hormone, is released during osmotic stimulation and oviposition. In the present study, the role of opioid peptides on AVT release was studied by examining the effects of an opioid agonist and antagonist on osmotic- and oviposition-induced secretion of AVT. The administration of hypertonic saline (1.5 M NaCl) induced an increase in the plasma levels of AVT. The simultaneous administration of morphine, an opioid receptor agonist, inhibited the osmotically induced increase in plasma levels of AVT in a dose-dependent manner. On the other hand, the co-administration of morphine with naloxone, an opioid receptor antagonist, attenuated the inhibitory effect of morphine. Moreover, injection of naloxone alone enhanced the osmotically induced increase in plasma levels of AVT. However, the administration of morphine did not inhibit the oviposition-induced increase in plasma levels of AVT. These results suggest that osmotic-induced release of AVT may be under opioid regulation, while oviposition-induced release of AVT may be controlled by a different mechanism. J. Exp. Zool. 286:481-486, 2000.     

Central, stereoselective receptors mediate the acute effects of opiate antagonists on luteinizing hormone secretion

Although a central site of acute opiate action in regulating luteinizing hormone (LH) secretion has been suggested by the ability of centrally implanted opiate antagonists to increase LH levels, opiate antagonists are lipophilic and could influence the pituitary in situ. Also, the physiological significance of opiate receptor blockade with antagonists rests on the assumed, but untested, stereoselectivity of these receptors. Therefore, a lipophobic quaternized derivative of naltrexone (MRZ 2663-Naltrexone methobromide) and dextro- (+) and levo- (-) stereoisomers of naloxone were used to study the site- and stereoselectivity of gonadotropin responses to opiate antagonists in vivo. Male rats were injected intracerebroventricularly (icv) or intravenously (iv) with the quaternary or tertiary congeners of naltrexone and subcutaneously (sc) with (-) or (+)-naloxone. Rats injected icv with 20 ug of quaternary naltrexone displayed significant increases in serum luteinizing hormone (LH). The onset of the response was rapid with serum LH levels being significantly elevated 15 minutes after the injection and returning to basal levels 30 minutes later. Rats injected iv with 10 mg/kg of quaternary naltrexone failed to show significant LH responses. Rats injected either centrally or periphally with equivalent doses of tertiary naltrexone showed LH responses that were similar to those found in animals injected icv with quaternary naltrexone. As little as 0.5 mg/kg of (-)-naloxone resulted in significant elevations in serum LH that were higher than those elicited by up to 10 mg/kg of (+)-naloxone, indicating that this effect of naloxone is stereoselective. These data support the argument that opioids can acutely modulate LH secretion through actions at stereoselective opioid receptors in the central nervous system.     

Opioid receptor agonists in the rabbit colon: comparison of in vivo and in vitro studies

Myoelectrical activity was recorded in the proximal and distal colon of rabbits using chronically implanted electrodes. The motility in both the proximal and distal colon was inhibited by the intravenous (IV) administration of the following opioid agonists for mu receptors: morphine and fentanyl, kappa receptors: ethylketazocine (EKC) and U 50 488 H, and delta receptors: D-Ala2 D-Leu5-enkephalin (DADLE) and D-Ser2 Leu-enkephalin-Thr6 (DSLET). In contrast, the myoelectric activity in the distal colon was increased during the infusion of an endogenous kappa opioid agonist, dynorphin (DYN). All of these effects were prevented by naloxone pretreatment. During in vitro studies using extraluminal force transducers, fentanyl, U 50 488 H and DSLET inhibited spontaneous contractions of the proximal colon, but U 50 488 H and DSLET caused a substantial increase in the motility of the distal colon. The observed motor responses in the proximal and distal colon following opioid agonist administration indicate that the control of these two intestinal segments may be different. It is suggested that the stimulatory effect of dynorphin on the distal colon is peripherally-mediated while inhibition of the whole colon by opioid agonists regardless of subtypes seems to be centrally-mediated.     

Effect of methamphetamine on the development of acute morphine tolerance and dependence in mice

The effect of methamphetamine on morphine analgesia (tail-flick assay) was studied in non-tolerant mice and in mice made acutely tolerant to morphine following a single injection of 100 mg/kg morphine. The analgesic potency of morphine was increased in non-tolerant and tolerant mice to the same extent by 3.2 mg/kg methamphetamine (3.3 and 4.4 fold increases, respectively). In contrast, the ED50's for morphine analgesia and naloxone-precipitated jumping in mice pretreated with either 100 mg/kg morphine or both morphine and 3.2 mg/kg methamphetamine were not significantly different, indicating that methamphetamine had no effect on the development of acute morphine tolerance and dependence. Although methamphetamine had no effect on the development of acute tolerance to morphine, 4-day pretreatment with methamphetamine produced cross-tolerance to morphine analgesia. However, cross-tolerance to morphine was not accompanied by enchanced sensitivity to naloxone.     

Modulatory effects of Gs-coupled excitatory opioid receptor functions on opioid analgesia,tolerance, and dependence

Electrophysiologic studies of opioid effects on nociceptive types of dorsal root ganglion (DRG) neurons in organotypic cultures have shown that morphine and mostμ, δ, and κ opioid agonists can elicit bimodal excitatory as well as inhibitory modulation of the action potential duration (APD) of these cells. Excitatory opioid effects have been shown to be mediated by opioid receptors that are coupled via Gs to cyclic AMP-dependent ionic conductances that prolong the APD, whereas inhibitory opioid effects are mediated by opioid receptors coupled via Gi/Go to ionic conductuances that shorten the APD. Selective blockade of excitatory opioid receptor functions by low (ca. pM) concentrations of naloxone, naltrexone, etorphine and other specific agents markedly increases the inhibitory potency of morphine or other bimodally acting agonists and attenuates development of tolerance/dependence. These in vitro studies have been confirmed by tail-flick assays showin that acute co-treatment of mice with morphine plus ultra-low-dose naltrexone or etorphine remarkably enhances the antinociceptive potency of morphine whereas chronic co-treatment attenuates development of tolerance and naloxone-precipitated withdrawal-jumping symptoms. Special issue dedicated to Dr. Eric J. Simon.     

Repeated administration of naltrexone and diprenorphine decreases food intake and body weight in squirrel monkeys

Although chronic administration of naloxone has been reported to reduce food intake and body weight in rats, there have been no comparable investigations using a nonhuman primate. We examined the effects of repeated injections of two long acting opiate antagonists - naltrexone and diprenorphine - on the ad libitum intake of a nutritional complete liquid diet and on body weight in squirrel monkeys. Naltrexone binds with highest affinity to the mu opioid receptor whereas diprenorphine binds with equally high affinity to several subtypes of opioid receptor. Diprenorphine (ED50 = 0.01 mg/kg) was 22 times more potent than naltrexone (ED50 = 0.22 mg/kg) in decreasing 2 h food intake, suggesting that more than one opioid receptor subtype may be involved in the anorectic effects of opiate antagonists. A 1.0 mg/kg dose of drug reduced 24 h food intake by 50% and was associated with a weekly reduction in body weight of 4 and 5% for naltrexone and diprenorphine, respectively. Thus, in contrast with shorter time intervals, 24 h food intakes were similar for the two drugs, and this was associated with comparable body weight profiles. The decreases in food intake and body weight remained constant over the period of drug administration. Some monkeys showed profuse salivation and "wet dog shakes" after 4 days of treatment with the 1.0 mg/kg dose but not after 1 day. Therefore, opiate antagonists given chronically to monkeys reduced food intake and body weight in a dose-dependent manner with no evidence of tolerance to these effects.     

Cardiovascular effects of naloxone,naltrexone and morphine in the squirrel monkey

Heart rate (HR) and mean arterial blood pressure (BP) were recorded from conscious, chair-restrained squirrel monkeys surgically prepared with chronically indwelling arterial and venous catheters to determine the effects of acute intravenous injections of two opiate antagonists and an agonist. Naloxone (0.3–10.0 mg/kg) or naltrexone (0.3–10.0 mg/kg) had little effect on HR or BP during a 30-minute post-injection period. Morphine (3.0–5.6 mg/kg) produced biphasic effects comprising an initial decrease followed by an increase in HR, and an increase followed by a decrease in BP. Lower morphine doses had lesser effects during a 100-minute post-injection period. Pre-treatment with 0.03 mg/kg naloxone attenuated the depressive effect of morphine on HR and BP, but increases in HR and BP due to morphine were enhanced. Pretreatment with 0.3 mg/kg naloxone prevented morphine-induced decreases in HR and BP, yet increases in HR and BP persisted. In previous behavioral studies, morphine in combination with naloxone similarly increased rates of responding in the squirrel monkey. Together, these data suggest an effect of naloxone that goes beyond mere pharmacological antagonism of the effects of morphine.     

Reversal of morphine, methadone and heroin induced effects in mice by naloxone methiodide

Opioid overdose, which is commonly associated with opioid induced respiratory depression, is a problem with both therapeutic and illicit opioid use. While the central mechanisms involved in the effects of opioids are well described, it has also been suggested that a peripheral component may contribute to the effects observed. This study aimed to further characterise the effects of the peripherally acting naloxone methiodide on the respiratory, analgesic and withdrawal effects produced by various opioid agonists. A comparison of the respiratory and analgesic effects of morphine, methadone and heroin in male Swiss-Albino mice was conducted and respiratory depressive ED(80) doses of each opioid determined. These doses (morphine 9 mg/kg i.p., methadone 7 mg/kg i.p., and heroin 17 mg/kg i.p.) were then used to show that both naloxone (3 mg/kg i.p.) and naloxone methiodide (30-100 mg/kg i.p.) could reverse the respiratory and analgesic effects of these opioid agonists, but only naloxone precipitated withdrawal. Further investigation in female C57BL/6J mice using barometric plethysmography found that both opioid antagonists could reverse methadone induced decreases in respiratory rate and increases in tidal volume. Its effects do not appear to be strain or sex dependent. It was concluded that naloxone methiodide can reverse the respiratory and analgesic actions of a variety of opioid agonists, without inducing opioid withdrawal.     

mu-1 opioid receptor stimulation decreases body temperature in conscious,unrestrained neonatal rats

The influence of mu-selective opioid agonists on neonatal thermoregulatory mechanisms has received little attention. Opioid treatment in adult subjects can cause either hyper- or hypothermia, depending on the experimental conditions, the strain of rat used, and the dose and route of administration of the drug. The present study assessed the effect of two mu opioid agonists on body temperature in neonatal Wistar rats aged 2 to 13 days. Rat pups were administered either saline or one of the two mu-selective opioid agonists, dermorphin (0.4 mg/kg) or fentanyl (0.06 mg/kg), by subcutaneous injection. Continuous rectal temperatures were measured both prior to and following drug or saline injection in freely moving, conscious animals. Ambient temperature in a plethysmograph chamber was maintained within or close to the thermoneutral zone for pups (32 degrees C). To distinguish between mu-1 and mu-2 effects, all animals received either saline or 10 mg/kg of the irreversible mu-1 antagonist naloxonazine (NALZ) 1 day prior to agonist administration. NALZ on its own had no effect on body temperature. Dermorphin and fentanyl both caused a fall in body temperature in pups of all age groups. The temperature decreases ranged from 0.8 degrees -2.2 degrees C. These opioid-induced changes were inhibited by NALZ pretreatment. Although there was no evidence for endogenous mu-1 opioid activity, this study indicated that stimulation of mu-1 opioid receptors causes a decrease in body temperature in conscious, unrestrained neonatal rats under or close to thermoneutral conditions.