Share this article on:

Age-Dependent Morphine Tolerance Development in the Rat

Wang, Yan MD; Mitchell, James MD; Moriyama, Kumi MD; Kim, Ki-jun MD, PhD; Sharma, Manohar PhD; Xie, Guo-xi MD, PhD; Palmer, Pamela Pierce MD, PhD

doi: 10.1213/01.ANE.0000152192.23851.40
Pain Medicine: Research Report

In all age groups, the use of opioids to treat chronic pain conditions has increased, yet the impact of age on opioid tolerance development has not been comprehensively addressed. In this study, we investigated age-related differences in morphine tolerance development in rats. Rats aged 3 wk, 3 mo, 6 mo, and 1 yr were used in the study. Morphine (8 mg/kg) was injected subcutaneously twice each day and its analgesic effect assessed by the change in tail-flick latency using a thermal stimulus 5 min before and 30 min after dosing. Tolerance was defined as a 75% reduction in morphine-induced analgesia compared to Day 1. Rats aged 3 wk, 3 mo, 6 mo, and 1 yr developed tolerance on the 4th, 10th, 14th, and 22nd days of morphine treatment, respectively. Plasma levels of morphine and its metabolites showed that pharmacokinetic differences among the groups did not correlate with the differences in tolerance development. This study demonstrates that morphine tolerance occurs more rapidly in younger rats than older rats and is unlikely to be the result of differences in drug metabolism or clearance. Aging may impact molecular processes involved in tolerance development and provide insight into novel therapeutic targets to delay opioid tolerance development.

IMPLICATIONS: Although many aspects of opioid tolerance mechanisms have been studied, the effect of age on the rate of opioid tolerance development has not been adequately addressed. We found that as a rat ages, the rate of tolerance development to morphine is slowed, an effect that cannot be explained by pharmacokinetic differences alone.

Department of Anesthesia and Perioperative Care, University of California, San Francisco

Supported, in part, by NIH grants (AR45570 and NS042593) and a John Severinghaus Radiometer Award (to PPP).

Accepted for publication November 16, 2004.

Address correspondence and reprint requests to Pamela Pierce Palmer, MD, PhD, Department of Anesthesia and Perioperative Care, University of California, San Francisco, 513 Parnassus Ave., Box 0464, Room S-455, San Francisco, CA 94143. Address e-mail to

For years there has been debate in the literature over whether there is a tolerance to opioids in humans and to what extent the presence or absence of nociceptive input, type of opioid, and route of administration has on the development of opioid tolerance. Interestingly, age as it relates to the rate of development of opioid tolerance is not a variable that has been systematically studied. A retrospective review of opioid-dose escalation in patients with chronic nonmalignant pain suggests that the rate of escalation of opioid dosing seems to be more rapid in younger patients than in older patients (1). Previous clinical studies on opioid tolerance development evaluate patients of widely varying ages as a single population and, therefore, it is possible that age as an uncontrolled variable results in the conflicting reports in the literature of opioid tolerance development. In rats, animals of varying ages have been evaluated in single-dose opioid studies to determine the effects of age on pharmacokinetic and analgesic efficacy variables (2). Numerous studies have also evaluated various aspects of opioid tolerance in the prenatal and early postnatal period (3–6), yet no study has systematically evaluated the rate of opioid tolerance development in rats ranging from young adolescents to advanced maturity.

Theoretically, the capacity for enhanced molecular and cellular neuroplasticity in younger neurons would allow for a more rapid development of tolerance to the cellular effects of opioids compared with older neurons. Changes in receptor-G protein interactions, second-messenger system signaling, and movement of cytosolic molecules such as protein kinase C (PKC) to the membrane are reported to occur during the development of tolerance (6). In the elderly, these cellular systems may not be as malleable and, therefore, these aged neurons would continue to respond to the opioid as when initially administered.

There are many studies in the field of synaptic plasticity that demonstrate decreasing neuronal plasticity with advancing age. For example, long-term potentiation (LTP) is a well known phenomenon that is believed to underlie the process of learning and memory. N-methyl-d-aspartate (NMDA) receptors and PKC are two examples of molecular substrates that are critical for the development of LTP. Aging affects both the development of LTP and the function of NMDA receptors and PKC (7–10).

Neuronal tolerance to opioids is also a learned cellular memory that seems to involve NMDA receptors and PKC (11). It is possible that age affects the function of these molecules and, therefore, affects the development of opioid tolerance. The aim of this study was to determine, in rats, whether the rate of development of opioid tolerance is dependent on the age of the rat studied.

Back to Top | Article Outline


Experiments using animals were approved by the Committee on Animal Research of the University of California, San Francisco. Male Sprague-Dawley rats (Bantin and Kingman, Fremont, CA) aged 3 wk, 3 mo, 6 mo, and 1 yr at the start of the experiment were used. On the first day, the body weights of the 4 age groups were 48 ± 2 g, 483 ± 4 g, 625 ± 11 g, and 679 ± 20 g, respectively. The rats were housed at 25°C under controlled lighting conditions (lights on 6:00 am to 6:00 pm) with free access to food and water.

Rats were acclimatized to the test conditions for 2 days before the first dose of morphine. Morphine (Baxter Healthcare Corporation, Deerfield, IL) was administered as a subcutaneous (s.c.) injection of 8 mg/kg at 9:00 am and 3:00 pm each day. A dose of 8 mg/kg was chosen because this dose produced significant analgesia (100%–150% increase from baseline tail-flick latency) while not producing apparent respiratory rate changes or sedation in any age group of rats. Analgesia was assessed by measuring tail-flick latency 5 min before and 30 min after each injection of morphine. It has been shown that repeated tail-flick latency testing with chronic opioid dosing does not produce baseline hyperalgesia, which is why we chose this test over the hot-plate test (12). Rats were placed in a clear acrylic cage and allowed to settle for 5 min. An infrared source of controlled intensity (Ugo Basile, Italy) was placed beneath the clear floor of the cage, 2 cm from the tip of the tail. Time until tail flick or withdrawal was measured. A maximum duration of exposure of 25 s (cutoff time) was set to prevent tissue injury and consequent hyperalgesia. Twice-daily dosing with morphine and testing of analgesia was continued until the post-dose tail-flick latency was not significantly different from the baseline tail-flick latency on Day 1 and loss of 75% of Day 1 analgesia had occurred. This was taken as the end-point for development of tolerance.

To determine whether pharmacokinetic effects influenced differences in tail-flick latencies with time, plasma levels of morphine and its metabolites, morphine-3-glucuronide (M3G) and morphine-6-glucuronide (M6G), were measured in separate groups of rats. Plasma samples were obtained on Days 1, 2, 3, and 4 in 3-wk-old rats, Days 1, 2, 3, 4, and 10 in 3-mo-old rats, Days 1, 4, 10, and 14 in 6-mo-old rats, and Days 1, 4, 10, 14, and 22 in 1-yr-old rats (the last day in each group is the day when rats became tolerant). Thirty minutes after the injection of morphine (8 mg/kg s.c.), a 1-mL specimen of blood was obtained. Blood specimens were taken from 3-wk old rats by aspiration from the heart after general anesthesia with isoflurane in oxygen. Blood specimens from the older rats were obtained by cannulation of a tail vein under general anesthesia and centrifuged at 10,000g for 10 min. The plasma phase (∼500 μL) from each sample was carefully transferred into a polypropylene microcentrifuge tube and stored at −80°C until analysis.

Plasma samples were analyzed for concentrations of morphine, M3G, and M6G by high-performance liquid chromatography/mass spectrometry at the Center for Human Toxicology, Department of Pharmacology and Toxicology, University of Utah Health Sciences Center, Salt Lake City, UT. The lower limit of quantitation for all analyses was 250 pg/mL.

All values are presented as mean ± se of the mean (sem). The end-point for tolerance was defined as a 75% loss of the average Day 1 analgesia as calculated below:



Tolerance was additionally defined as the first time at which the post-dose latency was not significantly greater than Day 1 baseline (P > 0.05). To compare each post-dose latency to the Day 1 baseline tail-flick latency, a one-way analysis of variance followed by a Dunnett test for multiple comparisons to a control were performed (SPSS version 11.0; SPSS, Chicago, IL).

Back to Top | Article Outline


Overall, the rates of tolerance development were significantly different among the four groups of rats (Fig. 1–4). In 1-yr-old rats (n = 6), the mean of pre-dose tail-flick latencies (baseline) was 9.4 ± 0.6 s at the start of Day 1. The pre-dose latencies did not statistically change over time (Day 22 = 9.55 ± 1.3 s), indicating that our experimental protocols did not produce local hyperalgesic conditions. The post-dose tail-flick latencies initially reached cutoff only on the start of Day 1 (25.0 s; 265% of baseline) and gradually decreased until Day 22 when post-dose latency (13.0 ± 2.9 s) was reduced more than 75% of the Day 1 average analgesia and was not significantly different from the Day 1 pre-dose latency (P > 0.05; Fig. 1).

Figure 1

Figure 1

Figure 2

Figure 2

Figure 3

Figure 3

Figure 4

Figure 4

In 6-mo-old rats (n = 10), the mean pre-dose tail-flick latency was 12.0 ± 0.5 s at the start of Day 1 and did not significantly change, being 10.3 ± 0.5 s at the end of Day 14. Over the same period, the post-dose tail-flick latency decreased from 25.0 s (210% of baseline) at the start of Day 1 to a mean value of 11.2 ± 1.4 s at the end of Day 14, a latency that was reduced more than 75% from the Day 1 average analgesia and was not significantly different from the pre-dose latency on Day 1 (P > 0.05; Fig. 2).

In 3-mo-old rats (n = 9), the mean pre-dose tail-flick latency also did not change throughout the testing period, being 8.9 ± 0.5 s at the start of Day 1 and 8.5 ± 0.5 s at the start of Day 10, which was the final day of testing for this group. The post-dose tail-flick latency decreased from 21.9 ± 1.2 s (250% of baseline) at the beginning of Day 1 to a mean value of 13.1 ± 1.7 s at the beginning of Day 10—a latency that was reduced 75% compared with the Day 1 average analgesia and was not significantly different from the pre-dose latency on Day 1 (P > 0.05; Fig. 3).

In 3-wk-old rats (n = 9), the mean pre-dose tail-flick latency was stable with values of 8.9 ± 0.7 s at the start of Day 1 and 8.9 ± 0.4 s by the end of Day 4. The post-dose tail-flick latency was 18.6 ± 1.0 s (210% of baseline) on the start of Day 1, and by Day 4 (after the seventh dose of morphine), the mean post-dose tail-flick latency decreased to 11.5 ± 0.9 s, which was reduced more than 75% of the Day 1 average analgesia and was not significantly different from the mean pre-dose value for Day 1 (P > 0.05; Fig. 4).

An additional group of 3-wk-old rats (n = 3) was tested using a 50% larger dose of morphine (12 mg/kg) to determine whether an increased morphine dose would delay tolerance development. The mean pre-dose tail-flick latency was stable with values of 8.9 ± 0.5 s at the start of Day 1 and 9.5 ± 0.3 s by the end of Day 4. The post-dose tail-flick latency was 25 s (280% of baseline) on the start of Day 1, and by Day 4 (after the eighth dose of morphine), the mean post-dose tail-flick latency decreased to 12.3 ± 1.4 s, which was reduced more than 75% of the Day 1 average analgesia and was not significantly different from the mean pre-dose value for Day 1 (P > 0.05; data not graphed).

Plasma was obtained for analysis from a separate group of rats 30 min after morphine dosing to correlate with the timing of tail-flick latency testing. For 1-yr-, 6-mo-, 3-mo-, and 3-wk-old rats, their mean post-dose plasma levels of morphine on Day 1 were 1043 ± 281, 1120 ± 238, 744 ± 91, and 445 ± 78 ng/mL, respectively. The plasma morphine levels for the four age groups during the period of chronic morphine administration are graphed in Figure 5. Three-week- and 3-mo-old rats had an initial decrease in morphine plasma levels from Days 1 through 4. By Day 10, the 3-mo group’s morphine levels returned to Day 1 levels, although the rats were no longer analgesic to morphine at this time. Similarly, the 6-mo- and 1-yr-old rats were not displaying analgesia to morphine at time points where morphine plasma levels were not significantly different from Day 1 levels.

Figure 5

Figure 5

The mean plasma levels of M3G in the older age groups increased progressively over time (Fig. 5, second row). The levels of M3G in 3-wk-old rats slightly decreased over the 4 days. M6G plasma levels displayed an interesting age-related trend. The M6G levels in the youngest (3-wk-old) and oldest (1-yr-old) rats stayed at low levels from Day 1 until the time of tolerance development. The M6G levels in the middle 2 age groups (3-mo- and 6-mo-old rats) increased significantly in the first few days and then decreased gradually.

Back to Top | Article Outline


The major finding of the present study is that the rate of opioid tolerance development in rats is dependent on age. There was a more than 400% increase in the length of time to tolerance in one-year-old rats as compared to three-week-old rats. Six-month-old and three-month-old rats also demonstrated a 250% and a 150% increase in time to tolerance compared to three-week-old rats, respectively. Whereas most studies on opioid tolerance in rats use young adult rats two to three months of age, there are almost no data on opioid tolerance in older rats, although a few laboratories have studied tolerance development in postnatal rat pups.

There is some debate as to how early a neonatal rat can develop opioid tolerance (Day 9 versus Day 15) (13,14). Similar to our finding of morphine tolerance by Day 4 in three-week-old rats, it has been demonstrated that two-week-old rats develop tolerance to s.c. morphine continuous infusion within 72 hours (15). Furthermore, Van Praag and Frenk (14) demonstrated morphine tolerance to daily intraperitoneal morphine within four days in nine-day-old rats. Most studies in the literature with opioids demonstrate the development of opioid tolerance within 8–10 days in young adult rats (16,17), which is also consistent with our three-month-old rat data demonstrating tolerance by Day 10.

Therefore, individual studies have hinted at age-dependent opioid tolerance, yet there is a large degree of variability in methods of detecting tolerance development in the literature and essentially no study of aged rats and opioid tolerance. Some variables include the opioid used, the route of delivery, the frequency of delivery, and the type of painful stimulus. These variables can result in widely varying rates of tolerance development and make the comparison of studies difficult. We therefore felt it useful to compare different ages of rats under the same experimental conditions. The dramatic differences in the rates of tolerance development suggest that the age of neurons may indeed play a significant role in their ability to alter their response to opioid receptor agonist stimulation. A study by Nozaki et al. (18) measured the rate of opioid tolerance as a function of age using younger rats (up to 12 weeks of age). Also, there was variability in opioid dosing to test for tolerance, and no pharmacokinetic data were reported. Their study suggested that as rats age, the rate of opioid tolerance slows, similar to the findings of our study. With the present trend of increasing use of long-term opioids in the management of chronic pain in humans (19), the importance of determining factors affecting the development of opioid tolerance cannot be overstated.

The possibility of a pharmacokinetic effect, such as a more rapid morphine metabolism or an increase in the antianalgesic metabolite M3G among the different age groups, was also studied. Our data demonstrated that although there was an initial decrease in morphine plasma levels during the first four days in both three-week- and three-month-old rats, the rate of decrease was identical in both groups. Yet by Day 4, the three-week-old rats were completely tolerant to morphine, whereas the three-month-old rats still had a significant analgesic response. Furthermore, a 50% larger dose of morphine (12 mg/kg) given to the second set of three-week-old rats did not delay tolerance development beyond Day 4. Morphine levels returned to baseline by Day 10 in the three-month-old group—the time at which they developed complete tolerance to morphine. Morphine levels remained constant over time in the six-month-old and one-year-old rats, and these levels were not significantly different than the Day 10 levels for the three-month-old rats. Yet, the time for tolerance development among these three groups was vastly different. It therefore seems that differences in plasma morphine levels were not the underlying reason for differences in the development of tolerance among the four age groups of rats.

Most morphine is converted to M3G, which has been described as an antianalgesic based on its stimulating effects on neuronal pathways in numerous studies (20). Significant conversion to M3G occurred initially in the three-week-old group, but these levels decreased over the four-day period. The older age groups had increasing accumulation of M3G up until Days 10–14. Overall, the largest increase in M3G levels occurred in the older rats at a time when significant analgesia was still present. Therefore, changes in M3G plasma levels do not seem to account for the differences in rates of opioid tolerance among the groups.

Only 10% of morphine is converted to M6G, a potent μ-opioid receptor agonist (21). Low plasma levels of this metabolite are usually detected in Sprague-Dawley rats, whereas slightly higher levels are found in humans (21). Human preterm neonates have been shown to produce less M6G compared with older neonates and children (22). Although a highly sensitive assay was used to detect M6G levels for this study, very little M6G could be detected at any time point in the three-week-old and the one-year-old rats. Rats older than one year in age have been shown to have low plasma levels of M6G (2). Both the three-month- and six-month-old rats had no detectable M6G on Day 1 and displayed the highest levels of M6G on Day 4, whereas lower levels were detected at the remaining time points. In our study, when detectable levels of M6G were present, these concentrations were approximately 10- to 30-fold less than the corresponding morphine plasma concentrations. In human clinical studies, although M6G has a demonstrated analgesic efficacy, this metabolite is not substantially more effective in producing analgesia than morphine (23). In rat studies, although M6G delivered intracerebroventricularly or intrathecally displays an analgesic potency much greater than morphine, systemic delivery of these drugs shows that M6G and morphine are equi-effective, with the only difference being that M6G analgesia persists longer than morphine (24). Therefore, the low levels of plasma M6G found in this study would not be expected to have a significant contribution to the analgesic response. Furthermore, rapid tolerance occurs to M6G (24) and, therefore, presence of this metabolite would not be expected to dramatically prolong tolerance development.

Although there are pharmacokinetic differences in these varying age groups, they did not seem to have a major impact on the rates of loss of opioid analgesia among the groups. It is more likely that the loss of opioid analgesia with time is caused by pharmacodynamic tolerance development. Changing pharmacodynamic effects of morphine with age could occur as the result of a variety of mechanisms. Aging affects the function of NMDA receptors and PKC, molecular substrates thought to play critical roles in the development of opioid tolerance. Aged rodents display reductions in the protein expression of multiple subunits of the NMDA receptor (9,25). Aging also seems to diminish the ability of PKC to move from the cytosol to the membrane by impeding the ability of the PKC molecule to effectively anchor to the membrane (8,26).

Many physicians are hesitant to prescribe opioids for chronic nonmalignant pain conditions because of the fear of dose escalation with time. Future clinical studies with patients using opioids for chronic pain will need to include age as an important variable when assessing development of opioid tolerance. Certainly, future laboratory studies are required to evaluate age-dependent opioid tolerance with opioids other than morphine and in models of chronic pain.

We thank Drs. Yilei Xing and Xiaokang Han for technical assistance. We also thank Dr. Diana G. Wilkins and her group at the Center for Human Toxicology University of Utah for their expert help in high-performance liquid chromatography/mass spectrometry analyses.

Back to Top | Article Outline


1. Buntin-Mushock C, Phillip L, Moriyama K, Palmer PP. Age-dependent opioid escalation in chronic pain patients. Anesth Analg 2005;100:1740–5.
2. Jourdan D, Pickering G, Marchand F, et al. Impact of ageing on the antinociceptive effect of reference analgesics in the Lou/c rat. Br J Pharmacol 2002;137:813–20.
3. O’Callaghan JP, Holtzman SG. Prenatal administration of morphine to the rat: tolerance to the analgesic effect of morphine in the offspring. J Pharmacol Exp Ther 1976;197:533–44.
4. Windh RT, Little PJ, Kuhn CM. The ontogeny of mu opiate tolerance and dependence in the rat: antinociceptive and biochemical studies. J Pharmacol Exp Ther 1995;273:1361–74.
5. Zhu H, Barr GA. Ontogeny of NMDA receptor-mediated morphine tolerance in the postnatal rat. Pain 2003;104:437–47.
6. Liu JG, Anand KJ. Protein kinases modulate the cellular adaptations associated with opioid tolerance and dependence. Brain Res Brain Res Rev 2001;38:1–19.
7. Komatsu Y. Age-dependent long-term potentiation of inhibitory synaptic transmission in rat visual cortex. J Neurosci 1994;14:6488–99.
8. Corsini E, Battaini F, Lucchi L, et al. A defective protein kinase C anchoring system underlying age-associated impairment in TNF-alpha production in rat macrophages. J Immunol 1999;163:3468–73.
9. Clayton DA, Browning MD. Deficits in the expression of the NR2B subunit in the hippocampus of aged Fisher 344 rats. Neurobiol Aging 2001;22:165–8.
10. Korzick DH, Holiman DA, Boluyt MO, et al. Diminished alpha1-adrenergic-mediated contraction and translocation of PKC in senescent rat heart. Am J Physiol Heart Circ Physiol 2001;281:H581–9.
11. Mao J, Mayer DJ. Spinal cord neuroplasticity following repeated opioid exposure and its relation to pathological pain. Ann N Y Acad Sci 2001;933:175–84.
12. Gardmark M, Hoglund AU, Hammarlund-Udenaes M. Aspects on tail-flick, hot-plate and electrical stimulation tests for morphine antinociception. Pharmacol Toxicol 1998;83:252–8.
13. Fanselow MS, Cramer CP. The ontogeny of opiate tolerance and withdrawal in infant rats. Pharmacol Biochem Behav 1988;31:431–8.
14. Van Praag H, Frenk H. Evidence for opiate tolerance in newborn rats. Brain Res Dev Brain Res 1991;60:99–102.
15. Thornton SR, Wang AF, Smith FL. Characterization of neonatal rat morphine tolerance and dependence. Eur J Pharmacol 1997;340:161–7.
16. Trujillo KA, Akil H. Inhibition of morphine tolerance and dependence by the NMDA receptor antagonist MK-801. Science 1991;251:85–7.
17. Fan GH, Wang LZ, Qiu HC, et al. Inhibition of calcium/calmodulin-dependent protein kinase II in rat hippocampus attenuates morphine tolerance and dependence. Mol Pharmacol 1999;56:39–45.
18. Nozaki M, Akera T, Lee CY, Brody TM. The effects of age on the development of tolerance to and physical dependence on morphine in rats. J Pharmacol Exp Ther 1975;192:506–12.
19. Bell JR. Australian trends in opioid prescribing for chronic non-cancer pain, 1986–1996. Med J Aust 1997;167:26–9.
20. Baker L, Ratka A. Sex-specific differences in levels of morphine, morphine-3-glucuronide, and morphine antinociception in rats. Pain 2002;95:65–74.
21. Milne RW, Nation RL, Somogyi AA. The disposition of morphine and its 3- and 6-glucuronide metabolites in humans and animals, and the importance of the metabolites to the pharmacological effects of morphine. Drug Metab Rev 1996;28:345–472.
22. Hartley R, Quinn M, Green M, Levene MI. Morphine glucuronidation in premature neonates. Br J Clin Pharmacol 1993;35:314–7.
23. Murthy BR, Pollack GM, Brouwer KL. Contribution of morphine-6-glucuronide to antinociception following intravenous administration of morphine to healthy volunteers. J Clin Pharmacol 2002;42:569–76.
24. Frances B, Gout R, Monsarrat B, et al. Further evidence that morphine-6 beta-glucuronide is a more potent opioid agonist than morphine. J Pharmacol Exp Ther 1992;262:25–31.
25. Magnusson KR, Nelson SE, Young AB. Age-related changes in the protein expression of subunits of the NMDA receptor. Brain Res Mol Brain Res 2002;99:40–5.
26. Pascale A, Govoni S, Battaini F. Age-related alteration of PKC, a key enzyme in memory processes: physiological and pathological examples. Mol Neurobiol 1998;16:49–62.
© 2005 International Anesthesia Research Society