Morphine dependence in single enteric neurons from the mouse colon requires deletion of β-arrestin2

Abstract

Chronic administration of morphine results in the development of tolerance to the analgesic effects and to inhibition of upper gastrointestinal motility but not to colonic motility, resulting in persistent constipation. In this study we examined the effect of chronic morphine in myenteric neurons from the adult mouse colon. Similar to the ileum, distinct neuronal populations exhibiting afterhyperpolarization (AHP)-positive and AHP-negative neurons were identified in the colon. Acute morphine (3 μM) decreased the number of action potentials, and increased the threshold for action potential generation indicative of reduced excitability in AHP-positive neurons. In neurons from the ileum of mice that were rendered antinociceptive tolerant by morphine-pellet implantation for 5 days, the opioid antagonist naloxone precipitated withdrawal as evidenced by increased neuronal excitability. Overnight incubation of ileum neurons with morphine also resulted in enhanced excitability to naloxone. Colonic neurons exposed to long-term morphine, remained unresponsive to naloxone suggesting that precipitated withdrawal does not occur in colonic neurons. However, morphine-treated colonic neurons from β-arrestin2 knockout mice demonstrated increased excitability upon treatment with naloxone as assessed by change in rheobase, number of action potentials and input resistance. These data suggest that similar to the ileum, acute exposure to morphine in colonic neurons results in reduced excitability due to inhibition of sodium currents. However, unlike the ileum, dependence to chronic exposure of morphine develops in colonic neurons from the β-arrestin2 knockout mice. These studies corroborate the in-vivo findings of the differential role of neuronal β-arrestin2 in the development of morphine tolerance/dependence in the ileum and colon.

Keywords: Colon; enteric neurons; morphine.

Figure 1.

Neurons and glia cells from the colonic longitudinal muscle/myenteric plexus (LMMP) preparation. Confocal microscopy revealed neuronal‐specific β III‐tubulin (first column; Abcam rabbit, 1:1500) staining and staining by the glial marker GFAP (second column, Chemicon, mouse 1:500) in cells isolated from the LMMP preparation of the colon (A). (B) at higher magnification, the close proximity of neurons and glia is evident. Antibodies were visualized with appropriate goat secondary antibody Alexa 488 (green, Molecular probes, 1:1000) or Alexa 594 (red, Molecular Probes, 1:1000). No staining was seen when primary antibody was omitted (data not shown).

Figure 2.

Cultured neurons from the adult mouse colon consist of two electrophysiologically distinct populations. In current clamp mode (A, B), all neurons exhibit action potentials upon current injection of 0.09 nA. At the end of current pulse, neurons either returned to their original resting membrane potential (A), or dipped below baseline in a slow after‐hyperpolarization (AHP, arrow, B). AHPs have an average magnitude of −7.06 ± 1.09 mV, an initial duration of 198.23 ± 14.8 ms, and a τ = 109.7 ± 29.2. In voltage‐clamp mode (C), inward Na⁺ currents follow by a sustained outward K⁺ current are readily apparent (insert). Current density–voltage relationships of Na⁺ and K⁺ currents in both AHP‐negative and AHP‐positive neurons showed that AHP‐positive neurons had significantly greater current densities as determined by two‐way ANOVA (*P > 0.05).

Figure 3.

Morphine decreases neuronal excitability in AHP‐positive neurons of the adult mouse colon without changing the resting membrane potential. In current clamp mode (A), colonic neurons fire multiple action potentials in the absence of drug (left panel). The red trace refers to the rheobase at which action potentials are first elicited. Following 3 μmol/L morphine treatment (right panel), action potentials are noticeably blunted, only single action potentials fire, and the neuron no longer fires at the previous rheobase. Statistical analysis (B) shows a significant increase in AP height, and a significant increase in rheobase without an alteration in the resting membrane potential of the neuron, determined by paired t‐test, significance at P < 0.05.

Figure 4.

Morphine decreases inward Na⁺ current density without affecting large sustained outward K⁺ current density. In voltage‐clamp mode (A), net currents were measured for duration of 300 ms demonstrating inward Na⁺ and outward K⁺ currents. Morphine (3 μmol/L) significantly reduced peak inward currents but not outward currents. (B) I–V curves for peak inward and end of pulse outward currents in the absence and presence of morphine. Two‐way ANOVA of current density and voltage (*P < 0.05).

Figure 5.

Precipitated withdrawal caused hyper excitability in morphine‐tolerant AHP‐positive enteric neurons from the ileum. Enteric neurons were isolated from 5‐day morphine‐pelleted mice and maintained in cell culture containing 3 μmol/L morphine. In current clamp mode (A), the neuron from tolerant mouse ileum (left panel) fires robust action potentials at 0.01 nA (red line). Following 1 μmol/L naloxone treatment (right panel), significantly greater number of action potentials were elicited at the rheobase. (B) Bar graph illustrating number of action potentials at 0.01 nA, the input resistance and the resting potential in the absence and presence of naloxone from morphine‐tolerant mouse ileum. (C) Current–voltage relationship for peak inward Na⁺ and outward K⁺ currents in the absence and presence of naloxone. Two‐way ANOVA of voltage and current density, significance at P < 0.05.

Figure 6.

Precipitated withdrawal does not occur in the absence of morphine tolerance, or in AHP‐negative neurons. Raw traces demonstrating lack of enhanced cell excitability in response to naloxone from (A) AHP‐positive neuron that was not pretreated with morphine, (B) an AHP‐negative neuron obtained from ileum of morphine‐pelleted mouse, and (C) an AHP‐negative neuron from the ileum that was pretreated with morphine (overnight).

Figure 7.

Genetic ablation of β‐arrestin two leads to the development of morphine dependence in AHP‐positive enteric neurons from the colon. Raw traces from enteric neurons preincubated with morphine for 48 h. (A) Ileum neuron in the presence of naloxone fired multiple action potentials at rheobase and demonstrated increased input resistance. (B) AHP‐positive colon neuron showed on increase in number of action potentials or increased input resistance in the presence of naloxone. (C) Colon neuron from β‐arrestin2 knockout mice responded to naloxone with an increased number of action potentials and enhanced input resistance.