|
|
||||||||
Divisions of Cardiovascular Medicine and Pulmonary and Critical Care Medicine, University of California-Davis, Davis, California
Submitted 7 September 2005 ; accepted in final form 18 November 2005
| ABSTRACT |
|---|
|
|
|---|
-opioid antagonist naloxone was only able to partially restore estrogens attenuating effect on the pressor response to static contraction suggested that estrogen affected an additional pathway, involving the dorsal root ganglion (DRG). Estrogen has been described to stimulate transcription within 10 min of its application to the DRG, raising the possibility that rapid genomic effects on neurotransmitter production may have contributed to estrogens effect on the exercise pressor reflex. This prompted us to test the hypothesis that estrogen modulated the pressor response to static contraction by influencing gene expression of the neurotransmitters released by the thin-fiber muscle afferents that evoke the exercise pressor reflex. We confirmed in decerebrated female rats that topical application of estrogen (0.01 µg/ml) to the lumbosacral spinal cord attenuated the pressor response to static muscle contraction (from 10 ± 3 to 1 ± 1 mmHg; P < 0.05). DRG were then harvested postmortem, and changes in mRNA expression were analyzed. GeneChip analysis revealed that neither estrogen nor contraction alone changed the mRNA expression of substance P, the neurokinin-1 receptor, CGRP, NGF, the P2X3 receptor, GABAA and GABAB, the 5-HT3A and 5-HT3B receptor, N-methyl-D-aspartate and non-N-methyl-D-aspartate receptors, opioid receptors, and opioid-like receptor. Surprisingly, however, contraction stimulated the expression of neuropeptide Y in the DRG in the presence and absence of estrogen. We conclude that estrogen does not attenuate the exercise pressor reflex through a genomic effect in the DRG. neural control of the circulation; static contraction; spinal cord; gender; neuropeptides; rats
Estrogen may have attenuated the exercise pressor reflex by a genomic mechanism, in addition to its effect on opioid signaling pathways. For example, within 10 min of its application to the DRG, estrogen is known to activate cyclic AMP response-element binding protein (35), which alters the transcription of multiple genes (1). Moreover, estrogen exposure for
6 h has been reported to regulate the synthesis of neuropeptides such as substance P (23, 29) and calcitonin gene-regulated peptide (CGRP) (9, 21, 22, 28) in the DRG of rats. Substance P has been shown to play a role in the spinal cord in evoking the exercise pressor reflex (16, 17, 20, 51), and both substance P and CGRP are found in many of the thin-fiber afferents evoking the reflex (18, 46).
In the present study, we asked the question of whether estrogen, topically applied to the spinal cord, attenuated the exercise pressor reflex through a rapid activation of genomic pathways in the DRG. We examined changes in the expression of mRNA encoding enzymes and proteins that regulate the neurotransmitters and neuromodulators implicated in evoking the exercise pressor reflex. We performed our experiments on female rats and confirmed that the attenuating effect of spinally applied estrogen on the exercise pressor reflex holds true for rats and is not limited to one species, namely the cat.
| METHODS |
|---|
|
|
|---|
A laminectomy was performed to expose the lumbar and sacral spinal cord. The laminectomy was extended laterally to partially expose the DRG. The dura and arachnoid membranes were incised longitudinally over the length of exposed spinal cord and reflected onto the surface of the remaining vertebral bones. This incision allowed access to the subarachnoid cavity, which enveloped the DRG, and was contiguous with the surface of the spinal cord. The L4 or L5 ventral root was identified and cut. A well of vinyl polysiloxane (VPS, Jeneric/Pentron) was formed on top of the reflected dura and remaining vertebral bones in a manner that enclosed the L4L5 dorsal roots and their respective ganglia. The method has been described in detail elsewhere (2, 50). To ensure its integrity, the well was filled with saline and checked for leakage. The well was then filled either with saline or drugs dissolved in saline throughout the experiment. The skin of the back was used to form a space for a pool around the exposed parts of the spinal cord, including the VPS well, and was filled with warm mineral oil (37°C). The saline-containing well, therefore, was seated in a mineral oil pool; this allowed us to monitor the well for leaks. Only data derived from rats in which no leakage was detected were analyzed. The musculature of the right hindlimb was exposed. The calcaneal bone was severed, and its tendon was attached to a force transducer (Grass, model FT 10). The knee was clamped in place. The rat was removed from the ventilator and allowed to breathe room air spontaneously.
Protocols.
The hindlimb muscles were contracted statically by electrically stimulating the L4 and L5 ventral roots at two to three times motor threshold (0.1 ms, 3040 Hz). Resting tension of the triceps surae muscles was set at 0.1 kg. The contraction period was 30 s. Each rat underwent two to three static contractions. The time between subsequent electrical stimulations was 15 min. In seven rats, an estrogen solution was administered after the contractions. 17
-Estradiol, which was dissolved in saline, was placed in the spinal VPS well in concentrations of 0.01 µg/ml. The volume was 100 µl. In a previous study in rats (2), the tissue concentration of neurokinin A (molecular weight 1,133) in the underlying superficial laminae of the spinal cord (at a depth of 500 µm) was 2570 times lower than the concentration in the superfusate in the well. Peak concentration was reached after 30 min of perfusion. The tissue concentration in the deeper laminae (depth of 0.751.5 mm) was found to be lower than that at the superficial laminae (2). Therefore, we chose our dosages to be in the range of 70 times the physiological estrogen concentration in blood during peak estrus. In a previous study in female cats (39), we found that application of estrogen in this way was sufficient in blocking the pressor response to static contraction of the triceps surae muscles.
The solution remained in the well for 60 min, and the responses to the exercise pressor reflex were tested 30, 45, and 60 min after the estrogen solution had been topically applied. At the end of the protocol, rats were killed with pentobarbital sodium (75 mg/kg iv). The L4L5 dorsal roots and DRG of the left and right sides were harvested and stored in RNAlater (Ambion) for further analysis of their mRNA profile.
RNA extraction and preparation of biotin-labeled RNA for GeneChip analysis. Total RNA was extracted from pooled DRG dissected from five female rats that underwent no surgery before death (control), from six female rats that performed static contraction but had no estrogen applied to their spinal cords, and from seven female rats that performed static contraction and had estrogen applied to the spinal cords. Equivalent material was collected from the contralateral DRG of rats that underwent the same surgical procedures as those described above (i.e., decerebration and laminectomy), but in which contraction was not elicited; these rats served as a control (sham operated) for each treatment group. Each pooled RNA sample, which comprised three or four DRG, was subjected to GeneChip analysis.
Total RNA was extracted as described previously (12). Briefly, total RNA was extracted from L4L5 DRG from three to four rats in each treatment group by homogenization of the tissue in Trizol reagent. RNA was purified and quantified according to the manufacturers (Invitrogen) protocol. An aliquot (20 µg) of pooled RNA solution was used for preparation of biotin-labeled RNA for hybridization to high-density oligonucleotide arrays (Mu74Av2), as described in the Affymetrix (Santa Clara, CA) protocol for sample preparation (12). Biotin-labeled RNA samples were hybridized to rat GeneChips (RAE 230 2.0). The scanned images of hybridization signals were analyzed with the Affymetrix Microarray Analysis Suite 5.0. The data were searched for the expression and changes in the expression of genes implicated in the neurotransmission and neuromodulation of the exercise pressor reflex.
Analysis of physiological data. Data reported in this study were obtained from female rats in which the presence of ovaries was confirmed postmortem. Values for mean arterial pressure (MAP) and HR are expressed as means ± SE. Baseline MAP and HR were recorded immediately before the muscle contraction; peak values represent the highest level reached during the muscle contraction. Statistical significance was determined by one- and two-way repeated-measures ANOVA, followed by Dunnetts and Tukeys post hoc tests when applicable. The criterion for statistical significance was P < 0.05.
Analysis of GeneChip data. Absolute mRNA expression (present or absent) and differential (ipsilateral vs contralateral sides) mRNA expression analysis were obtained using the Microarray Analysis Suite 5.0 software. When a P value for the detection signal was <0.039 (P = 0.00020.039), the expression of the mRNA was classified as present. All mRNAs with a P value for detection of >0.039 were considered absent. The signal intensities for transcripts classified as present ranged from 5 to 7,000 units.
To identify "contraction-sensitive" genes, data obtained from the "contraction and noncontraction" sides of rats were compared. The difference was expressed as a fold change for each mRNA. Data that showed either an increase or a decrease of >1.5-fold were classified as contraction-sensitive genes. Similarly, data obtained from the contraction and noncontraction sides of rats, in which estrogen was applied to the spinal cord, were compared. If there was no change in gene expression in response to contraction alone, "estrogen sensitivity" was defined as a >1.5-fold change in mRNA expression following estrogen treatment. If there was a change in gene expression in response to contraction, estrogen sensitivity in contraction-sensitive genes was determined as the fold change in mRNA expression observed with estrogen treatment weighted against the change in mRNA expression evoked by contraction. Genes, whose increase or decrease in expression was two times larger in response to estrogen than that in response to contraction, were classified as estrogen-sensitive genes. Genes for which a discrepancy between control (no surgery) and sham-operated data were detected (by >1.5-fold) were excluded from further analysis based on the assumption that changes in expression were reflective of the surgical procedure.
Results and discussions focus primarily on differentially (>1.5-fold) expressed mRNAs that were detected with a high level of confidence (P < 0.04) in at least one treatment group. This conservative analysis excluded a large number of genes that were identified as significantly different between the groups but whose detection was of low confidence (P > 0.04).
Our laboratory (12, 49) and others (24, 44) have shown that the changes in the expression of mRNAs selected by a conservative analysis of hybridization data, as described above, could be confirmed by independent analysis such as Northern blot, PCR, real-time PCR, and in some cases immunoblot analysis of the encoded proteins (12). Therefore, further analysis of the changes in the expression of mRNAs described below was not performed.
| RESULTS |
|---|
|
|
|---|
|
-estradiol on the pressor and cardioaccelerator response to static contraction.
17
-Estradiol applied topically to the spinal cord significantly attenuated the pressor response to static contraction (Table 1). The pressor response was of a similar magnitude as that in female rats whose hindlimbs were contracted but did not subsequently receive estrogen (Table 1). Thirty minutes after spinal application of 17
-estradiol, the pressor response to contraction was significantly decreased from 10 ± 3 to 1 ± 1 mmHg (P < 0.05; Table 1). Baseline MAP did not vary over time (Fig. 1). The developed muscle tension was consistent throughout the protocol (Table 1).
|
GeneChip analyses.
The mRNAs for estrogen receptors were expressed only marginally in the L4L5 DRG (Table 2). In contrast, the mRNA for tachykinin, the precursor to substance P, and CGRP were highly expressed in the L4L5 DRG (Table 3). The mRNA for the neurokinin-1 receptor (tachykinin receptor 1) and a CGRP receptor component protein were also detectable. Although the mRNA for NGF was not expressed in the DRG, the mRNA for its receptors (TNFRSF16 and Ntrk1) was present in the DRG. Additionally, the mRNAs for the purinergic receptor P2X3, the N-methyl-D-aspartate (NMDA) receptor, the non-NMDA glutamate receptor, and the GABAA and GABAB receptors were detected in the DRG. The mRNAs for serotonergic receptors subtypes 3a and 3b were also present in the DRG. In contrast to other opioid receptor subtypes, the mRNAs for the µ and
subtypes were not expressed in the DRG (Table 3). The mRNA for neuropeptide Y was expressed in high amounts in the DRG. Static contraction increased the mRNA expression of neuropeptide Y; this was found to be the case, regardless of whether estradiol was applied topically to the cord (Table 3). However, there was no apparent difference in the mRNA expression for neuropeptide Y between the rats that served in the no-surgery control group and the rats that served in the sham-operated group.
|
|
|
| DISCUSSION |
|---|
|
|
|---|
-estradiol to the lumbosacral spinal cord of decerebrated female rats attenuated the pressor response to static contraction, a finding that is consistent with our previous findings in decerebrated female cats (39, 41). The cardioaccelerator response to contraction was modest in our experiments; nevertheless, the effect was abolished by estrogen application.
We also found that estrogen, topically applied to the spinal cord, had little or no effect on gene expression in the DRG. The simplest interpretation of these findings is that the estrogen-induced attenuation of the pressor response to contraction was not caused by a genomic effect in the DRG. Specifically, estrogen did not affect the expression of genes important in the synthesis and catabolism of neurotransmitters, neuromodulators, and their corresponding receptors that are found in the DRG and are thought to be involved in the spinal processing of the exercise pressor reflex. Included in the neurotransmitters, neuromodulators, and receptors examined were substance P, CGRP, glutamate, GABA, the neurokinin-1 receptor, µ- and
-opioid receptors, GABA receptors, the P2X3 receptor, serotonergic receptors, and NMDA and non-NMDA receptors.
Previously, our laboratory found that estrogens attenuating effect on the pressor response to static contraction was in part opioid dependent in female cats (41). In the present study, we found that the mRNAs of µ- and
-opioid receptors were not expressed in a significant amount in the DRG of female rats and, furthermore, that their expression was not affected by topical application of estrogen to the spinal cord. Our findings suggest that the opioid-dependent influence of spinally applied estrogen on the exercise pressor reflex was not caused by a genomic effect on primary afferent fibers. In addition, the absence of a significant amount of µ- and
-opioid receptor mRNA in the female DRG implies that the major site of action of estrogen on the exercise pressor reflex was the dorsal horn. Our findings contrast with previous reports that measured the mRNA of µ- and
-opioid receptors in the DRG of male rats (25, 26, 38). One explanation for the difference between our findings and those of other investigators (25, 26, 38) might be sex. A second explanation for the difference might be that we identified mRNA with microarrays, whereas these other investigators identified mRNA with either in situ hybridization (25, 26) or a ribonuclease protection assay (38). We cannot exclude the possibility that these latter techniques may have been more sensitive to low levels of mRNA than was the microarray technique used in the present study.
In our laboratorys previous study using only female cats, (41), we found that naloxone, a µ- and
-opioid receptor antagonist, only partially restored the estrogen-induced attenuation of the exercise pressor reflex. This finding raised the possibility that another nongenomic pathway contributed to the attenuating effect of estrogen. Recently, 17
-estradiol was found to inhibit an ATP-induced calcium influx in cultured rat DRG cells, an effect that was caused by the activation of P2X receptors and voltage-gated Ca2+ channels (5). In addition, application of a P2X receptor antagonist into the dorsal horn was found to attenuate the exercise pressor reflex (10). These reports lead to the speculation that estrogen might have attenuated the exercise pressor reflex by blocking or desensitizing P2X receptors on dorsal horn cells participating in this reflex arc.
Surprisingly, we found that the mRNA for neuropeptide Y was expressed in the DRG of female rats. We also found that static contraction further increased its expression in both untreated and estrogen-treated rats. Previous measurements of either NPY or its mRNA in L4L5 rat DRG have yielded no constitutive levels of either. These studies, however, were conducted either in male rats (19, 30, 33, 47) or in rats in which the sex was not specified (11). The possibility exists that the NPY expression found in our study was caused by the trauma inflicted by the extensive surgical preparation required (4). This possibility, however, was not supported by our finding that there was no difference in NPY expression between the control (no surgery) and the sham-operated rats. The significance of the increase in NPY expression evoked by static contraction is unknown, but it is interesting to speculate that the release of NPY by thin-fiber muscle afferents feeds back onto these primary afferents to inhibit the release of neurotransmitters, such as glutamate, and neuromodulators, such as substance P (3, 6, 48).
In our study, 5-HT3A and 5-HT3B receptors were found to be the most dominant serotonergic receptors expressed in the DRG. Nevertheless, neither static contraction nor topical application of estrogen to the cord reliably altered their mRNA expression. DRG neurons have been described to express 5-HT3A and 5-HT3B receptors in small-, medium-, and large-diameter cell bodies (14, 27, 31). Small- and medium-diameter cell bodies are presumed to transduce nociceptive information, whereas large-diameter cell bodies are presumed to transduce nonnoxious mechanical information. Intrathecal application of a 5-HT3 receptor agonist has been shown to inhibit nociception in the rat hind paw (37). Acting presynaptically, 5-HT3A and 5-HT3B receptors may function to inhibit the discharge of the thin-fiber muscle afferents responsive to contraction. A presynaptic action of serotonin may account for its inhibitory effect on the exercise pressor reflex when this monoamine is injected intrathecally (15, 32).
Previously, the mRNA for the GABAB was shown to be highly expressed in the rat DRG (45). We confirmed and extended this finding by showing that the gene expression for the GABAB receptor was not altered by either static contraction or application of estrogen to the spinal cord. GABAB receptors are believed to be located on or near the terminals of dorsal root fibers; when stimulated, they are thought to have an inhibitory effect on the discharge of these primary afferents (52). For example, activation of GABAB receptors in chick DRG neurons was found to inhibit calcium channels in both a voltage-dependent and a voltage-independent manner (36). Moreover, application of GABA to DRG cells has been shown to produce a slow inhibition of P2X3 receptors, an effect that was mediated by GABAB receptors (43). These findings, considered together, raise the possibility that the high expression of the mRNA for the GABAB receptors in the DRG results in this receptor being responsible, at least in part, for the presynaptic inhibition of the thin-fiber muscle afferent input causing the exercise pressor reflex.
The relevance to humans of our present and previous findings in cats and rats (3941) is uncertain, but there are some interesting parallels to be drawn. For example, in humans, differences in the magnitudes of the pressor and muscle sympathetic components of the exercise pressor reflex have been attributed to gender. Specifically, the pressor and muscle sympathetic nerve responses to static exercise were less in premenopausal women than in either postmenopausal women or men (7, 8). Moreover, these components fluctuated with the ovarian cycle, being less during the follicular phase than during the menstrual phase (8). Furthermore, oral administration of estrogen to postmenopausal women was found to attenuate the cardiovascular responses to exercise (34), a finding that also might be partly attributed to the effect of this sex hormone on the exercise pressor reflex.
In summary, we found that 17
-estradiol, topically applied to the lumbosacral spinal cord of decerebrated female rats, did not attenuate the pressor response to static contraction through rapid activation of genomic pathways in the DRG. This finding leads us to speculate that the attenuating action of estrogen on the exercise pressor reflex is not attributable to a decrease in the release of neurotransmitters and neuromodulators by thin-fiber muscle afferents. We further speculate that this attenuation occurred in the dorsal horn, may have been nongenomic, and was caused in part by opioid signaling pathways (41).
| GRANTS |
|---|
|
|
|---|
| FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
| REFERENCES |
|---|
|
|
|---|
-preprotachykinin mRNA levels in uninjured dorsal root ganglion neurons, but not in axotomized neurons. Exp Neurol 170: 101108, 2001.[CrossRef][ISI][Medline]
-estradiol attenuate the exercise pressor reflex in male cats. J Appl Physiol 94: 14311436, 2003.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |