INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CARDIOVASCULAR HOMEOSTASIS is dramatically altered in pregnancy. There is a 50-70% increase in maternal plasma volume (1), accompanied by an increase in cardiac output and decreased arterial pressure (3, 23). The pressor response to ANG II is blunted during pregnancy (11), as is the renal response to volume load (17). It is not well understood how these changes occur, but the hormonal changes during pregnancy likely play an important role.

Plasma levels of the progesterone metabolite 5-pregnan-3-ol-20-one (pregnan) increase during pregnancy in the human and in the rat (6, 8, 10). Pregnan is a potent modulator of GABA type A (GABA_A) receptors that bind GABA, the major inhibitory neurotransmitter in the brain (19). It has been shown that pregnan administration to virgin female rats results in attenuation of the reflex increase in renal sympathetic outflow normally observed during a hypotensive challenge (13). Pregnan also influences central control of blood volume. Atrial distention, which occurs during a volume load, normally causes diuresis and natriuresis (14) and increased c-fos expression in the paraventricular nucleus in the brain of virgin female rats (9, 17). These responses are blocked in the pregnant rat (6, 14) and in pregnan-treated virgin female rats (6). Pregnan is therefore able to mimic pregnancy with respect to the reflex responses to changes in both mean arterial pressure (MAP) and blood volume.

Nitric oxide (NO), which is important for maintaining normal vascular tone in humans (4) and rats (14), is increased during pregnancy (7). It is probably responsible, in part, for the pregnancy-induced vasodilation and plasma volume expansion, because NO synthase inhibition in gravid animals leads to increased blood pressure and reduced plasma volume (25, 26). It is not known what stimulates NO biosynthesis to increase during pregnancy. We believe that pregnan, which is also increased during pregnancy, may possibly be involved in this increase. Because pregnan has been shown to affect the central control of blood pressure and blood volume, we were interested to see whether it might also influence NO biosynthesis and plasma volume in rats. Specifically, we hypothesized that pregnan would increase both NO biosynthesis and plasma volume.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animals. Virgin female Long-Evans rats (200-250 g) were obtained from Charles River Canada (St. Foy, Quebec). They were held for several weeks before surgical preparation in a temperature- and humidity-controlled room with a 12:12-h light-dark cycle (light 0700-1900). After surgery, the rats were allowed to recover to their preoperative weights before the experiments began. Rats from which urine samples were taken for NO biosynthesis were maintained on a low nitrate/nitrite diet (Harlan Teklad amino acid diet TD 86529, Madison, WI). Otherwise, they were maintained on a 0.28% sodium diet (PMI Feeds, St. Louis, MO). Water was offered ad libitum.

Surgery. The rats were prepared under pentobarbital sodium anesthesia (0.65 mg/kg body wt; MTC Pharmaceuticals) with Silastic cannulas (0.05-cm inner diameter, 0.09-cm outer diameter; Dow Corning, Midland, MI) implanted nonocclusively into the inferior vena cava (IVC) and the jugular vein. The cannulas were exteriorized by tunneling them subcutaneously to the back of the neck and connecting them to stainless steel tubing secured to the interscapular region. Two additional groups of rats were prepared with telemetric pressure-recording devices (PA-C40, Data Sciences International) implanted nonocclusively in the abdominal aorta. MAP was recorded continuously before and after 2 days of treatment with pregnan (or vehicle) in conscious unrestrained rats using the PhysioTel Telemetry System (Data Sciences International).

Preparation of drugs. 5-Pregnan-3-ol-20-one (Sigma Chemical, Oakville, ON, Canada) was dissolved in 20% 2-hydroxypropyl--cyclodextrin (Sigma) in sterile distilled water. Each treatment consisted of 500 µg pregnan in 400 µl cyclodextrin plus 200 µl sunflower oil, or vehicle alone (cyclodextrin and oil), given subcutaneously. Animals were given the treatment once daily for 2 days.

Vaginal smears. The rat estrus cycle lasts 4 days. Rats are in proestrus for ~12 h, estrus for ~12 h, and metestrus for ~21 h. The rest of the time, 50-55 h, they are in diestrus. These stages are easily identified by examining the cells obtained from vaginal lavage. Vaginal smears were taken at the end of the 2-day treatment with pregnan and 2 days later to confirm that the rats were not pseudopregnant.

Nitrate reductase assay. The stable metabolites of NO are nitrate and nitrite (NO_x). Urinary NO_x is a reliable marker for NO biosynthesis. Urine samples were diluted (1/40) to a final volume of 100 µl and mixed with 5 µl each of 50 µM potassium dihydrogenphosphate at pH 7.5 (Sigma), 50 µM NADPH (Boehringer Mannheim, Laval, QB, Canada), and 5 µM FAD (Boehringer). Last, 10 µl of nitrate reductase from aspergillus (2 U/ml; Boehringer) were added to reduce nitrate into nitrite. After incubation at 37°C for 20 min, lactate dehydrogenase (10 U/ml; Boehringer) and 10 mM sodium pyruvate (Sigma) were added to oxidize NADPH to avoid any interference with the following nitrite determination. Griess' reagent was then added to the reaction to quantify the NO_x concentration by spectrophotometry, using a microplate reader (Bio-rad 550; Hercules, CA). A standard curve was constructed using standards of NaNO₃ (BDH Chemicals, Toronto, ON, Canada) at 1, 3, 10, 30, and 100 µM. All urine samples were analyzed in duplicate, and a pooled sample of rat urine was used for each assay to demonstrate reliability of the assay.

Plasma and blood volume determination. Evans blue dye (Fisher Scientific, Whitby, ON, Canada) was injected into the jugular cannula (0.3 ml of a 0.5% wt/vol solution in sterile 0.9% sodium saline). Blood samples (0.15 ml) were taken from the IVC line at 10, 20, 30, 40, and 60 min after the dye was injected. Blood was transferred to Caraway Micro Blood Collecting tubes (Fisher) and centrifuged at 11,700 rpm (Clay Adams Micro-Hematocrit II). The tubes were cut, and plasma (50 µl) was diluted in 950 µl of 0.9% sodium saline. Dye concentration was read at 605 nm (LKB Novaspec), and plasma and blood volume were determined by comparing the degree of dilution of dye from time 0. Details of the blood-sampling procedure and volume determination have previously been described (16).

Experimental protocol for NO biosynthesis. Three days after surgery, rats were fed a low nitrate/nitrite diet and continued on this diet until the end of the experiment. When preoperative weight was reached (7-10 days after surgery), rats were placed in Nalgene metabolism cages. Urine samples were taken from all animals before treatment. Rats were then randomly chosen to receive either pregnan or vehicle injections for 2 days, after which a second urine sample was taken. Before each 24-h urine collection, the cages were cleaned with antibacterial soap and wiped with 70% ethanol to prevent nitrate and/or nitrite formation from bacteria. Urine samples were centrifuged and frozen at 40°C until assayed for nitrite using the nitrate reductase assay.

Experimental protocol for plasma volume analysis. Rats were placed in Nalgene metabolism cages after preoperative weight was reached. The next day, initial plasma and blood volumes were measured using the Evans blue dye dilution technique. Plasma and blood volumes were measured again after treatment with pregnan or vehicle. To ensure adequate and standardized hydration levels before dye injection and blood sampling for each experiment, food and water were removed and animals were infused with 0.9% sodium saline through the IVC cannula, over a period of 60 min at a rate of 2 ml/h using an LKB (Bromma) Micro Perpex pump.

Statistics. Student's t-test was used to determine the significance of changes in blood pressure, plasma volume, blood volume, and nitrate/nitrite levels between pregnan-treated and vehicle-treated rats. Student's t-test for paired data was used to compare any significant changes within groups.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

It was confirmed that administration of pregnan did not induce pseudopregnancy in the rats, i.e., they continued to cycle. Nor did pregnan alter MAP [experimental group: initial MAP 96 ± 4 mmHg, final MAP 93 ± 5 mmHg (n = 6); control group: initial MAP 99 ± 4 mmHg, final MAP 99 ± 3 mmHg (n = 6)].

Urinary nitrate/nitrite levels. The means ± SE interassay nitrite concentration for pooled rat urine was 0.6 ± 0.1 µM across four assays. The amount of nitrate/nitrite in the low amino acid diet fed to rats was negligible compared with the amount of nitrate/nitrite found in the urine samples, as previously confirmed (7). Figure 1 shows the change in mean NO_x content in 24-h urine collections from baseline in vehicle-treated control and pregnan-treated rats. Mean baseline NO_x levels in controls were 4.8 ± 0.5 µmol (n = 10) and 4.5 ± 0.4 µmol (n = 10) in pregnan-treated rats. The mean increase of NO_x levels from baseline was significantly greater in rats treated with pregnan (1.9 ± 0.8 µmol, n = 10) compared with vehicle-treated controls (0.3 ± 0.4 µmol, n = 10, P < 0.05).

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Change in urinary nitrate/nitrite (NOx) levels per 24-h urine collection from baseline for vehicle-treated control (n = 10, solid bar) and 5-pregnan-3-ol-20-one (pregnan)-treated (n = 10, open bar) rats. Vertical bars delineate means ± SE. *Significant increase in urinary NOx levels after pregnan treatment compared with vehicle treatment, P < 0.05.

Plasma and blood volume determination. Mean plasma volume (in ml/100 g body wt) increased significantly from baseline in both vehicle-treated control and pregnan-treated rats. Mean baseline plasma volume levels in controls were 4.9 ± 0.2 ml/100 g (n = 11) and 5.0 ± 0.1 ml/100 g (n = 11) in pregnan-treated rats. However, the mean increase of plasma volume in pregnan-treated rats (0.7 ± 0.2 ml/100 g, n = 11) was significantly greater than the vehicle-treated control group (0.2 ± 0.1 ml/100 g, n = 11, P < 0.05) (Fig. 2).

View larger version (9K): [in this window] [in a new window]   Fig. 2.   Change in plasma volume levels in ml/100 g of body wt from baseline in vehicle-treated control (n = 11, solid bar) and pregnan-treated (n = 11, open bar) rats. Vertical bars delineate means ± SE. *Significant increase in plasma volume after pregnan treatment compared with vehicle treatment, P < 0.05.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

We have shown that pregnan increases NO biosynthesis in virgin female rats and that the increase in NO is paralleled by an increase in plasma volume. We propose that this progesterone metabolite, plasma levels of which are elevated during pregnancy (6, 8, 10), may thus be an important factor mediating the increase in plasma volume that is essential for a normal pregnancy.

In the rat, plasma pregnan levels increase during pregnancy, from basal levels of ~10 ng/ml to a maximum of ~40 ng/ml just before parturition (6). It is known that exogenous pregnan (5 mg · kg¹ · day¹) causes an increase in plasma levels to ~40 ng/ml (5). Thus we estimate that our dose of 2.5 mg · kg¹ · day¹ would have increased plasma pregnan to a level characteristic of midpregnancy, i.e., ~20 ng/ml. Furthermore, acute administration of pregnan at a dose estimated to achieve a similar plasma level (~22 ng/ml) mimics pregnancy in reducing MAP and potentiating arterial baroreflex sympathoinhibition (18).

Unlike Laiprasert et al. (18), we did not find that pregnan altered basal blood pressure. This is consistent with our previous study showing there to be no significant fall in MAP at early or midpregnancy, but only at late pregnancy (22). This apparent discrepancy may be readily explained by the fact that our pregnant and pregnan-treated animals were all conscious, as opposed to the chloralose-anesthetized rats used by Laiprasert et al. (18). Credence is lent to this explanation by the fact that when Masilamani and Heesch (20) studied baroreflex control of blood pressure in conscious rats, they also found that pregnan did not alter basal MAP.

Our baseline NO_x values are comparable to levels previously reported for virgin female rats (7). After pregnan administration, we found an increase in urinary NO_x of 1.9 µmol (per 24 h urine). By comparison, in pregnant rats, Conrad et. al. (7) detected an increase of ~1.1 µmol on day 6 of pregnancy (not significantly greater than prepregnant levels) and a maximal increase of ~3.1 µmol on day 13 of pregnancy (significantly different from prepregnant levels) (4). Thus our value of 1.9 µmol falls into the range of increase of urinary NO_x normally seen during midpregnancy in rats. This suggests that exogenous pregnan can mimic pregnancy in causing an increase in NO biosynthesis. Although pseudopregnancy has also been associated with increased NO biosynthesis (7), this was not a factor in our experiments because the rats continued to cycle during and after the pregnan treatment.

In the pregnan-treated rats, plasma volume increased by 13.2% from baseline values. By comparison, in pregnant rats, the increase in plasma volume from nonpregnant levels is reported to be ~7% on day 6 of pregnancy, 14% on day 9, 32% on day 13, and 68% on day 20 (1-3). Our increase of 13.2% in plasma volume after pregnan treatment thus signifies a rise again comparable to midpregnancy. We suggest that the small increase in plasma volume (4%) found in the control animals, which was not paralleled by a significant increase in blood volume, was not physiologically significant.

The mechanism by which pregnan causes an increase in NO and plasma volume is uncertain, but previous work in our laboratory suggests that pregnan may act to stimulate adrenomedullin (ADM) release. Plasma levels of ADM, a 54-amino acid peptide, are increased in pregnant rats and in virgin female rats treated with pregnan (15). ADM is known to induce NO biosynthesis (12). We propose therefore that pregnan stimulates ADM secretion, which then induces NO biosynthesis. In turn, NO is known to increase plasma volume (25, 26). In addition, pregnan has the ability to alter central control of blood volume, probably through modulation of GABA_A-inhibitory pathways; specifically, pregnan reduces the central response to atrial distension (9).

Perspectives

In conclusion, we have found that chronic administration of the progesterone metabolite pregnan, at a dose designed to achieve plasma levels characteristic of midpregnancy, causes increases in NO biosynthesis and plasma volume of a magnitude typically observed during midpregnancy. The ability of pregnan to alter central control of blood pressure (13) and blood volume (9), to increase NO biosynthesis, and to induce plasma volume expansion in virgin animals suggests that pregnan may be responsible, at least in part, for the cardiovascular changes that normally occur during pregnancy.