Angiotensin II-induced hypertension differentially affects estrogen and progestin receptors in central autonomic regulatory areas of female rats

Abstract

Estrogen receptor (ER) activation in central autonomic nuclei modulates arterial blood pressure (ABP) and counteracts the deleterious effect of hypertension. We tested the hypothesis that hypertension, in turn, influences the expression and trafficking of gonadal steroid receptors in central cardiovascular circuits. Thus, we examined whether ER- and progestin receptor (PR)-immunoreactivity (ir) are altered in medullary and hypothalamic autonomic areas of cycling rats following chronic infusion of the hypertensive agent, angiotensin II (AngII). After 1 week AngII-infusion, systolic ABP was elevated from 103+/-4 to 172+/-8 mmHg (p<0.05; N=8/group) and all rats were in diestrus (low estrogen). In AngII-infused rats the number of PR-immunoreactive nuclei was reduced (-72%) compared to saline-infused controls also in diestrus (p<0.05). Furthermore, the intensity of ERalpha-ir increased selectively in nuclei (16%) and cytoplasm (21%) of cells in the commissural nucleus of the solitary tract (cNTS; p<0.05) while neither the number nor intensity of ERbeta-labeled cells changed (p>0.05). Following chronic AngII-infusion, electron microscopy showed a higher cytoplasmic-to-nuclear ratio of ERalpha-labeling selectively in tyrosine hydroxylase (TH)-labeled neurons in the cNTS. Furthermore, AngII-infusion increased ERalpha-ir in the cytosol of TH- and non-TH neuronal perikarya and increased the amount of ERalpha-ir associated with endoplasmic reticulum only in TH-containing perikarya. The data suggest that hypertension modulates the expression and subcellular distribution of ERalpha and PR in central autonomic regions involved in blood pressure control. Considering that ERalpha counteracts the central and peripheral effects of AngII, these receptor changes may underlie adaptive responses that protect females from the deleterious effects of hypertension.

Fig. 1. Schematic drawings through the medulla at the levels analyzed for TH and nuclear gonadal steroid receptor dual labeling (adapted from Swanson (Swanson, 1992), plates 61, 69 and 71)

A. The anterior portion of the RVLM (aRVLM; about 12.2mm caudal to bregma) and slightly more posterior portions of RVLM (not shown) were analyzed. B. At the level of the area postrema (AP), the medial (mNTS), the dorsomedial NTS (dmNTS), and A1 region (A1) were examined (about 13.76 mm from bregma). C. More caudally, the commissural NTS (cNTS) was analyzed (about 14.4 mm from bregma). Abbreviations: AMB, nucleus ambiguous; CU, cuneate nucleus; DMX, dorsal motor nucleus vagus nerve; ECU, external cuneate nucleus; GR, gracile nucleus; icp, inferior cerebral peduncle; IO, inferior olivary complex; LRN, lateral reticular nucleus; mlf, medial longitudinal fascicle; ml, medial lemniscus; MV, medial vestibular nuc; py, pyramidal tract; RO, nucleus raphe obscurus; SPV, spinal nucleus of trigeminal; spV spinal tract trigeminal; ts, tractus solitarius; XII, hypoglossal nucleus. Bar, 1 mm.

Fig. 2. Levels of nuclear immunoreactivity for PRs and ERαs fluctuate across the estrous cycle

A. The number of nuclei with detectable ERα-immunoperoxidase labeling is significantly higher in diestrus (DiE) than in proestrus (ProE) in the cNTS, is marginally higher in the paraventricular nucleus of the hypothalamus (PVN), and does not change in the other regions examined. B. The number of nuclei with detectable PR-immunoperoxidase labeling is significantly higher in proestrus animals in the cNTS, periventricular hypothalamus (PVH) and arcuate hypothalamus (ArH). The number of PR-labeled nuclei in the aRVLM and PVN tended to be higher at proestrus compared to diestrus, but this was not significant. C. The number of cells with ERβ-ir was not significantly different over the estrous cycle in the medial region of the NTS, aRVLM and the PVN. n = 10 sections/region/condition **, p < 0.01; *, p < 0.05; ^, p = 0.053 (ANOVA).

Fig. 3. Chronic AngII treatment decreases the detectable number of PR-immunoreactive nuclei, but not ERα or ERβ, in cNTS

A. In saline-treated controls, PR-labeled nuclei are scattered in cNTS. B. Following AngII treatment that increased ABP, PR-labeled nuclei in cNTS appear lighter and fewer than in controls. Bars A & B, 50 µm. C. In cNTS, the number of PR-labeled nuclei is significantly lower with AngII administration. D & E. In all regions examined, the number of ERα-labeled nuclei (D) or ERβ-labeled cells (E) is not significantly different with AngII administration. **, p < 0.001 (ANOVA); n =13 /region/condition for ERα; n = 10/region/condition for ERβ; n = 18/region/condition for PR.

Fig. 4. Chronic Ang II treatment increases the intensity of ERα-labeling in cNTS

A. ERα labeling in cytoplasm and nuclei of neurons in a saline-treated rat in diestrus. B. ERα-ir neurons in cNTS of a diestrus rat with Ang-II-induced increases in ABP. In A and B, nuclei are shown with feathered arrows while cytoplasm is designated with plain arrows. C. Following Ang II administration, the optical density (OD) of ERα-labeling significantly increased in nuclei and cytoplasm of neurons in cNTS. **, p < 0.001; *, p < 0.05 (ANOVA) n = 37 cells/condition. Bar = 50 µm.

Fig. 5. About half of the catecholaminergic neurons in the cNTS contain ERα- and PR-immunoreactivities in their nuclei

A. TH (green) and nuclear ERα (red) labeling in cNTS. Some neurons colocalize both labels (arrows) while other only have TH (arrowhead) or only ERα (asterisk). B. Nuclear ERα is in varying proportions of TH-ir neurons in NTS and RVLM subfields (n=6–8 sections from 4 rats). The greatest degree of colocalization is in cNTS, while no colocalization is seen in dmNTS or posterior RVLM (pRVLM). C. PR-ir (red) and TH-ir (green) are colocalized (arrows) in cNTS cells. A single-labeled TH-ir neuron (arrowhead) is shown for comparison. D. About half of the PR-ir nuclei in cNTS are in TH-ir neurons, but other NTS subregions have very little PR-ir and no observed colocalization with TH-ir. In RVLM, no TH-ir neurons have detectable PR-ir, although the rRVLM has relatively more PR-ir than other areas examined (n = 12 sections for NTS, 6 sections for RVLM). Bar A & C, 50 µm.

Fig. 6. At the ultrastructural level, ERα-ir is detected in the nuclei and cytoplasm of TH-labeled neurons in the cNTS

A & B. Examples of two TH-labeled perikarya from control rats containing ERα-immunogold labeling (ERa/TH-P) in their nuclei (black particles) and cytoplasm. The density of ERα-labeled gold particles is the nucleus is less in cell A (0.4 particles/µm²) than in cell B (3.3 particles/µm²). Within the cytoplasm, ERα-labeled gold particles can be affiliated with the plasma membrane (pm), endoplasmic reticulum (er), mitochondria (m) or lack any organelle affiliations in the plane of section, i.e., “free” (f). Bar, 2 µm.

Fig. 7. In both control and AngII-exposed rats, ERα-ir is found in non-TH labeled neurons and in glial cells in the cNTS

A. In this non-TH labeled perikaryon (Era-P), ERα immunogold particles are found mostly in the nucleus (N; 0.8 particles/µm²) and to a lesser extent in the cytoplasm (0.5 particles/µm²). A TH-labeled perikaryon (TH-P) is nearby. B. High magnification of the cytoplasm of an unlabeled perikaryon showing examples of ERα immunogold particles (arrow) affiliated with a saccule of endoplasmic reticula (er) adjacent to a mitochondrion (m) and near (arrowhead) a Golgi apparatus (G). C. ERα immunogold particles (black) are found in the perikaryon (Era-P) of an astrocytic cell body identified by the bundle of fibrils (gf) in its cytoplasm. ERα-labeled gold particles are found in the nucleus (N) and affiliated with a mitochondrion (m) in the cytoplasm. Bar A & C, 2 µm; B, 500 nm.

Fig. 8. Chronic AngII-infusion affects the subcellular distribution of ERα-ir in the cytoplasm of TH-labeled neurons in the cNTS

A. Example of a TH-labeled perikaryon from an AngII-infused rat containing ERα-immunogold labeling (ERa/TH-P) in the nucleus (black particles; 2.7 particles/µm²) and cytoplasm. In this example ERα-immunogold labeling is affiliated with several mitochondria (m) and endoplasmic reticula (er). B & C. High magnification of two different TH-labeled perikarya showing examples of ERα-immunogold labeling affiliated with mitochondria (m), endoplamsic reticula (er) or “free” (f) in the cytosol. Bar A, 2 µm; Bar B & C, 500 nm.

Fig. 9. Chronic AngII infusion produces re-distribution of ERα in cNTS

A. With Ang II infusion, the ratio of cytoplasmic to nuclear ERα-gold particles significantly increased only in TH-labeled cells and was unchanged in neurons lacking TH or in glia. The number of cells assessed (n) is shown in the bars. B. Within TH-labeled neurons of rats chronically infused with AngII, significantly more “free” ERα-gold particles were found in the cytosol. In a separate analysis of ERα-gold particles associated with cytoplasmic organelles, significantly more were associated with the endoplasmic reticulum (er). Although not significant, more ERα-gold particles tended to be associated with mitochondria (mito). C. Within non-TH-containing neurons of rats chronically infused with AngII, significantly more ERα-gold particles were free in the cytosol; however, no differences in the subcellular distribution relative to specific organelles were seen. **, p < 0.001; *, p < 0.05; ^, p = 0.02 (ANOVA).