Ovarian steroids alter mu opioid receptor trafficking in hippocampal parvalbumin GABAergic interneurons

Abstract

The endogenous hippocampal opioid systems are implicated in learning associated with drug use. Recently, we showed that ovarian hormones regulate enkephalin levels in the mossy fiber pathway. This pathway overlaps with parvalbumin (PARV)-basket interneurons that contain the enkephalin-activated mu opioid receptors (MORs) and are important for controlling the "temporal timing" of granule cells. Here, we evaluated the influence of ovarian steroids on the trafficking of MORs in PARV interneurons. Two groups of female rats were analyzed: cycling rats in proestrus (relatively high estrogens) or diestrus; and ovariectomized rats euthanized 6, 24 or 72 h after estradiol benzoate (10 microg, s.c.) administration. Dorsal hippocampal sections were dually immunolabeled for MOR and PARV and examined by light and electron microscopy. As in males, in females MOR-immunoreactivity (-ir) was in numerous PARV-labeled perikarya, dendrites and terminals in the dentate hilar region. Variation in ovarian steroid levels altered the subcellular distribution of MORs in PARV-labeled dendrites but not terminals. In normal cycling rats, MOR-gold particles on the plasma membrane of small PARV-labeled dendrites (area <1 microm2) had higher density in proestrus rats than in diestrus rats. Likewise, in ovariectomized rats MORs showed higher density on the plasma membrane of small PARV-labeled dendrites 72 h after estradiol exposure. The number of PARV-labeled cells was not affected by estrous cycle phase or estrogen levels. These results demonstrate that estrogen levels positively regulate the availability of MORs on GABAergic interneurons in the dentate gyrus, suggesting cooperative interaction between opioids and estrogens in modulating principal cell excitability.

Figure 1

By light microscopy, MOR-ir is contained in numerous PARV-labeled neurons in the hilus of the dentate gyrus. Confocal micrographs show MOR alone (A), PARV alone (B) and a merged Z-stack reconstruction image (C). Arrows indicate dual labeled cells. Bars, 50 μm.

Figure 2

By electron microscopy, MOR SIG particles are found in PARV-labeled dendrites and terminals in the female rat dentate gyrus. (A, B) PARV peroxidase labeled dendrites (den) contain MOR SIG particles in the cytoplasm (white arrowheads), and at the plasma membrane (regular arrow) from a proestrus rat (A) and a diestrus rat (B). In A, the PARV-labeled dendrite contains two MOR SIG particles at the plasma membrane and one of the particles is in close apposition to an asymmetric synapse (solid black arrow) with an unlabeled terminal (ut). (C, D) PARV peroxidase labeled small terminals (ter) containing MOR-SIG particles in the cytoplasm or in the plasma membrane from a proestrus rat (C) and a diestrus rat (D). In C, the small terminal has one SIG particle in the plasma membrane (regular arrowhead); the terminal is next to a MOR SIG labeled soma. Both terminals contain numerous small synaptic vesicles. Bars, 500 nm.

Figure 3

By electron microscopy, few MOR SIG particles are found in PARV-labeled perikarya in the female rat dentate gyrus. Only two SIG particles (white arrows) are observed. Note that the lack of MOR-SIG labeling is not due to antibody penetration since this picture was taken from a superficial sample as noted by surface-tissue interface (asterisk). This dual labeled soma is in close apposition to a PARV-SIG MOR dual labeled terminal (ter). Another single labeled PARV terminal is observed on the field.

Figure 4

The distribution of MOR-SIG particles in small PARV-labeled dendrites changes in normal cycling female rats. (A) The number of MOR-SIG particles in small PARV dendrites (i.e., area < 1 μm² and perimeter < 5 μm) is increased on the perimeter and decreased in the cytoplasm of proestrus compared to diestrus rats. (B) In proestrus compared to diestrus rats, small PARV dendrites show a higher ratio of plasma membrane:total MOR-SIG particles and lower ratio of cytoplasmic:total MOR-SIG particles. * represents significantly different from diestrus group (p< 0.05) on a t-test.

Figure 5

The distribution of MOR-SIG particles in small PARV dendrites is altered in OVXed rats administered pulsatile EB over a 72 hr period (72 hr EB group). (A) The number of MOR-SIG particles is increased on the perimeter and tends to decrease in the cytoplasm of small PARV dendrites from the 72 hr EB group. (B) In the 72 hr EB group, the ratio of plasma membrane:total MOR-SIG particles is increased in small PARV dendrites. # represents significantly different from control and 24 hrs group. Smaller inserts on the right show a comparison of control and 72 hrs EB group only compared using Student's t-test. * represents significantly different from control group (p< 0.05).

Figure 6

The distribution of MOR-SIG particles near the plasma membrane of large PARV-labeled dendrites is decreased in OVXed rats in the 24 and 72 hr EB groups. (A) The number of MOR-SIG particles that lie near (i.e., within 0.1 μm but not touching) the plasma membrane of large (area bigger than 1 μm² and a perimeter bigger than 5 μm) PARV-labeled dendrites are decreased in the 24 and 72 hr EB groups. (B) The ratio of cytoplasmic:total MOR-SIG particles is increased in large PARV-labeled dendrites in the 24 hr EB group. * represents significantly different from control and 6 hr EB group and ^ represents significantly different from 6 hr EB group only.

Figure 7

Schematic diagram summarizing the observed effect on the redistribution of MORs in PARV interneurons in the female rat dentate gyrus. When estrogen levels are high (either at proestrus or 72 hr after EB administration) enkephalin/dynorphin levels are elevated in the mossy fibers and more MORs are on the small dendrites plasma membrane of PARV GABAergic interneurons. However, when estrogen levels are low (either at diestrus or in OVX control rats), less MORs are on the small dendrites plasma membrane and more on the cytoplasm. Together, these results suggest that during periods of high estrogen levels, more MORs are available to be activated and this can lead to greater disinhibition (e.g., more excitation) of pyramidal cells.