Nitric oxide synthase-mediated blood pressure regulation in obese melanocortin-4 receptor-deficient pregnant rats

Abstract

Although obesity increases the risk for hypertension in pregnancy, the mechanisms responsible are unknown. Increased nitric oxide (NO) production results in vasodilation and reduced blood pressure during normal pregnancy in lean rats; however, the role of NO is less clear during obese pregnancies. We examined the impact of obesity on NO synthase (NOS)-mediated regulation of blood pressure during pregnancy by testing the hypothesis that NOS activity, expression, and regulation of vascular tone and blood pressure are reduced in obese pregnant rats. At gestational day 19, melanocortin-4 receptor (MC4R)-deficient obese rats (MC4R) had greater body weight and fat mass with elevated blood pressure and circulating sFlt-1 levels compared with MC4R pregnant rats. MC4R pregnant rats also had less circulating cGMP levels and reduced total NOS enzymatic activity and expression in mesenteric arteries. Despite decreased biochemical measures of NO/NOS in MC4R rats, NOS inhibition enhanced vasoconstriction only in mesenteric arteries from MC4R rats, suggesting greater NOS-mediated tone. To examine the role of NOS on blood pressure regulation in obese pregnant rats, MC4R and MC4R pregnant rats were administered the nonselective NOS inhibitor N^G-nitro-l-arginine methyl ester (l-NAME, 100 mg/l) from gestational day 14 to 19 in drinking water. The degree by which l-NAME raised blood pressure was similar between obese and lean pregnant rats. Although MC4R obese pregnant rats had elevated blood pressure associated with reduced total NOS activity and expression, they had enhanced NOS-mediated attenuation of vasoconstriction, with no evidence of alterations in NOS-mediated regulation of blood pressure.

hypertensive disorders of pregnancy are a significant killer of pregnant women in the United States. According to the Centers for Disease Control, between the years 2011 and 2012, hypertensive disorders of pregnancy contributed to 7.6% of pregnancy-related deaths in the US (Pregnancy Mortality Surveillance System, 26 July 2016, retrieved from http://www.cdc.gov/reproductivehealth/maternalinfanthealth/pmss.html). Preeclampsia is one of these dangerous hypertensive disorders of pregnancy, which manifests in the latter half of gestation as new-onset maternal hypertension alongside an onslaught of maternal cardiovascular, cerebral, or renal abnormalities (16). The incidence of preeclampsia is on the rise, and maternal obesity is hypothesized to be a significant cause for this increase. Indeed, epidemiological studies have repeatedly found a positive relationship between increasing maternal body mass index and the rates of maternal hypertension (5, 15). However, less is known about the mechanisms responsible.

Maternal bioavailability of nitric oxide (NO), which is a vasodilator, a buffer to vasoconstriction, and a pronatriuretic molecule, is thought to be a major target of soluble placental and immune cell factors in the pathogenesis of hypertension during pregnancy. One such factor is the antiangiogenic molecule soluble fms-like tyrosine kinase (sFlt)-1. It elicits many of the characteristics of preeclampsia when infused into once normotensive pregnant rats and mice. It increases blood pressure with carotid artery endothelial dysfunction (2) and reductions in glomerular production of NO (18). This sFlt-1-induced hypertension is prevented by supplementation with l-arginine, a cofactor required for NO synthase (NOS) activity. Moreover, plasma from placental ischemic rats reduced endothelial-dependent vasorelaxation in small mesenteric arteries from normal pregnant rats via reductions in NOS function (26). These data support that soluble factors promote vascular dysfunction in preeclamptic pregnancies, at least in part, by impairing the NO/NOS pathway. Quenching of NO promotes the development of hypertension in pregnancy (11), therefore, an intact NOS system is critical for blood pressure regulation during normal pregnancy.

In humans, it is evident that there are reductions in NOS in preeclampsia, including lower circulating levels of the surrogate measure of NO bioavailability, cGMP, especially in those with fetal intrauterine growth restriction (22). Far less is known about NOS biology in obese, hypertensive pregnancies. In the present study, we directly examined NOS activity, the impact of NOS on the regulation of vascular tone, and the role of NOS in blood pressure control during pregnancy using nonselective NOS inhibition in obese and lean rats. Previously, we characterized the melanocortin-4 receptor (MC4R)-deficient rat as model of obese pregnancy exhibiting increased body weight, total body fat mass, visceral white adipose tissue mass, and blood pressure at the end of pregnancy compared with wild-type controls. We tested the hypothesis that obese pregnant rats have increased circulating sFlt-1, reduced NOS activity, and NOS-dependent regulation of vascular tone in small arteries. Furthermore, we determined whether there is a reduced ability of NOS inhibition to increase blood pressure, as an indicator of reduced NOS regulation of blood pressure in obesity.

MATERIALS AND METHODS

Animals and treatments.

All animal experiments were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals with all animal-use protocols approved by The University of Mississippi Medical Center’s Institutional Animal Care and Use Committee. MC4R^+/+ wild-type and MC4R^+/− heterozygous female rats were obtained from a colony maintained at UMMC from founders purchased from Charles River Laboratories (Wilmington, MA) and originally developed by Mul and colleagues (17). Our rat colony was maintained by non-brother-sister mating of heterozygous rats, as ad libitum-fed MC4R^−/− female rats do not readily mate. MC4R^+/+ rats included littermate controls with additional animals produced by mating wild-type rats. Experimental animals were produced from at least three different breeding pairs from first and second pregnancies. Breeding pairs for generation of experimental animals were maintained on standard chow diet from Harlan Laboratories (unautoclaved, Teklad 7013/NIH31; Indianapolis, IN) and water ad libitum. At weaning (3 wk), all offspring were placed on NIH31 and tail biopsies were collected for genotyping.

At ~13 wk old, females were mated with genotyped-matched males, which had been maintained on NIH31 diet, for the generation of timed-pregnant experimental rats. The observation of sperm in vaginal smears was indicative of gestational day (GD) 0. From GD 14 to 19, pregnant rats were started on tap water supplemented with or without N^G-nitro-l-arginine methyl ester (l-NAME, 100 mg/l). This resulted in four groups of pregnant rats: untreated MC4R^+/+; l-NAME-treated MC4R^+/+; untreated MC4R^+/− and l-NAME-treated MC4R^+/−. On GD 18, total body fat mass was quantitated using an Echo-MRI-700 (Echo Medical Systems, Houston, TX).

Hemodynamic measurements.

On GD 18, rats were placed under isoflurane anesthesia (Butler Schein Animal Health, Dublin, OH), and indwelling catheters were implanted in the left carotid artery and exposed at the nape of the neck using aseptic techniques. Catheters consisted of V/1 tubing attached to V/3 tubing (Scientific Commondities, Lake Havasu City, AZ). Approximately 2.5 cm of the V/3 end of the catheter was inserted into the carotid. Catheters were filled with sterile heparin-0.9% saline solution (300 mg/ml; Pfizer, New York City, NY) and stoppered with a stainless steel catheter plug (SP22/12; Instech Laboratories, Plymouth Meeting, PA) to maintain patency. On GD 19, rats were placed in restraint cages and catheters connected to pressure transducers (MLT0699; ADInstruments, Colorado Springs, CO) coupled to a computerized data acquisition system (PowerLab, ADInstruments). Once hemodynamic readings stabilized (~40 min), mean arterial blood pressure data were collected. A subset of these rats was included for glomerular filtration rate (GFR) measurements, as previously described in our laboratory (23). These hemodynamic measurements are presented as means and Δ ± SE. Δs were calculated by subtracting the individual values in the obese group from the average of the lean groups.

Tissue harvest.

On GD 19, rats were placed under isoflurane anesthesia, a midline incision was made, and uterine horns with fetuses were exteriorized. Blood was collected from the abdominal aorta into Vacutainer K₂EDTA tubes (BD, Franklin Lakes, NJ) and spun at 2,500 rpm for 12 min at 4°C, and plasma was separated and then stored at −20°C. The number of viable and reabsorbed fetuses in each animal was recorded along with individual fetus and placenta weights. Data presented are the combined averages from these measurements for each pregnant rat and total fetal and placental weights. Placental sufficiency was calculated as fetal weight divided by placental weight for each fetal-placental unit and then averaged for each pregnant rat; this is similar to the calculation for this measure in humans (20). Visceral adipose tissue, including the fat from around the kidneys, adrenals, and the retroperitoneal fat, was weighed. Whole mesenteric artery arcades were snap-frozen in liquid N₂ and stored at −80°C until processed with a section saved for vasoconstriction studies.

Mesenteric artery tissue processing.

In a subset of each group of rats, mesenteric arteries were used for Western blot analysis as previously described (25). Briefly, membranes were probed using antibodies for anti-NOS3 (1:500; catalog no. 610297, BD Biosciences, San Jose, CA) and anti-β-actin (1:20,000; catalog no. A2066, Sigma). Secondary antibodies were used to detect the NOS3 antibody (goat anti-mouse 1:1,000; catalog no. A11375, Invitrogen, Carlsbad, CA) and the β-actin antibody (goat anti-rabbit 1:10,000; catalog no. A21038, Invitrogen). When analyzed, expression was normalized to corresponding β-actin densities. The same artery homogenates were used for the l-[³H]arginine-to-l-[³H]citrulline conversion assay for examination of total NOS activity, as previously described (25).

Vasoconstriction studies.

Third-order mesenteric arteries were cleaned of perivascular adipose tissue for vascular function studies. Vascular rings of ~2.5 mm in length were mounted on chucks in a wire myograph (model 620M, Danish Myo Technology A/S, Aarhus, Denmark) containing 5 ml PSS (concentration in mmol/l: 118.3 NaCl, 4.7 KCl, 2.5 CaCl₂, 1.2 MgSO₄, 1.2 KH₂PO₄, 25 NaHCO₃, and 11.1 dextrose) (Sigma) warmed to 37°C and bubbled with 95% O₂-5% CO₂. A preload of 4 mN was placed on the arterial rings. Blood vessel integrity was examined with a bolus dose of phenylephrine (Phe, Sigma) to produce vasoconstriction followed by a bolus of acetylcholine (ACh, Sigma) to induce vasorelaxation, ensuring proper endothelial function. Only those arterial segments that reached 100% relaxation were included in our myography studies. Arterial segments were washed with PSS, equilibrated for 15 min, and then incubated in the presence or absence of a nonselective NOS inhibitor l-NAME (100 μM, Sigma) for 15 min. Cumulative concentration-response curves were generated to increasing concentrations of Phe (1E-8 to 3E-3).

Plasma biochemistry.

Circulating levels of cGMP (Cayman Chemicals, Ann Arbor, MI), sFlt-1 (R&D Systems, Minneapolis, MN), and leptin (R&D Systems) were quantified using ELISAs per manufacturers’ instructions. Total cholesterol (Wako Diagnostics, Osaka, Japan) and free fatty acids (Zen-bio, Durham, NC) were assessed by colorimetric assays.

Statistical analysis.

All data are expressed as means ± SE. Data were graphed and analyzed using GraphPad Prism 5 (La Jolla, CA). Data were analyzed using a Student’s t-test or two-way ANOVA with a Bonferroni’s multiple comparison post-hoc tests, where necessary. Statistical significance was set at P < 0.05.

RESULTS

Body weight and fat mass.

Maternal body weight (Table 1) was greater in MC4R^+/− versus MC4R^+/+ pregnant rats on GD 19. This was accompanied by increased amounts of total body fat (Table 1) and visceral adipose tissue (Table 1) masses when assessed at the end of pregnancy. Lean body mass was not different between MC4R^+/− and MC4R^+/+ pregnant rats (Table 1). Cumulative food intake from GD 14–19 was similar in the obese MC4R^+/− versus lean MC4R^+/+ pregnant rats (Table 1).

Table 1.

Maternal body weight (with fetal and placental weights subtracted), EchoMRI total body fat mass, and visceral adipose tissue and total plasma cholesterol, leptin, and sFlt-1 measured at the end of pregnancy in lean MC4R^+/+ and obese MC4R^+/− pregnant rats

Parameter	MC4R+/+	MC4R+/−
Maternal body weight, g	337 ± 8 (19)	366 ± 10 (14)*
Total body fat mass, g	60 ± 5 (17)	96 ± 5 (8)*
Total body lean mass, g	284 ± 5 (11)	278 ± 8 (7)
Cumulative food intake, GD14-19, g	104 ± 6 (6)	98 ± 7 (9)
Visceral adipose tissue mass, g	3.6 ± 0.3 (19)	6.1 ± 0.5 (10)*
Plasma total cholesterol, mg/dl	93 ± 5 (10)	123 ± 5 (11)*
Plasma leptin, ng/ml	3.6 ± 0.3 (10)	5.9 ± 0.6 (11)*
Plasma sFlt-1, pg/ml	34 ± 6 (16)	121 ± 57 (11)*

Values are means ± SE; number of rats in parentheses. *P < 0.05 vs. MC4R^+/+ pregnant rats.

Circulating obesity-related metabolic factors and sFlt-1.

The greater amounts of fat mass found in the MC4R^+/− over MC4R^+/+ pregnant rats were accompanied with increased circulating levels of total cholesterol (Table 1) and the adipokine leptin (Table 1) at GD 19. However, there was no significant difference in free fatty acid levels for MC4R^+/− 2,751 ± 272 pmol, N = 10) versus MC4R^+/+ (2,904 ± 491 pmol, N = 9) pregnant rats.

Plasma levels of the antiangiogenic factor sFlt-1 were higher in the obese versus lean pregnant rats (Table 1).

Circulating cGMP levels.

To get a grasp on the levels of NO bioavailability in the cardiovascular system, a surrogate measure of NO signaling was examined in plasma cGMP. Its levels were significantly reduced in the obese MC4R^+/− versus lean MC4R^+/+ pregnant rats at GD 19 (Fig. 1A).

Fig. 1.

Maternal plasma cGMP levels (A) at gestational day (GD) 19 in lean MC4R^+/+ (n = 12) and obese MC4R^+/− (n = 11) pregnant rats and total nitric oxide synthase (NOS) activity levels (B) and endothelial NOS isoform (NOS3) protein expression (C) in whole mesenteric arterial bed at GD 19 in lean MC4R^+/+ (n = 3–8) and obese MC4R^+/− (n = 3–8) pregnant rats. Representative Western blot is presented above the quantification in B. *P < 0.05 vs. MC4R^+/+ pregnant rats.

Small artery NOS activity and expression.

To directly evaluate vascular levels of NOS, total NOS enzymatic activity was determined in small mesenteric arteries. Consistent with the reduced circulating measure of NO bioavailability, cGMP, total NOS activity was significantly less in the small arteries isolated from obese MC4R^+/− versus lean MC4R^+/+ pregnant rats (Fig. 1B). This was accompanied by reduced protein expression of NOS3 in the obese versus lean pregnant group in this small artery bed (Fig. 1C).

NOS regulation of vasoconstriction in small mesenteric arteries.

NOS-mediated regulation of vascular tone was examined by performing cumulative-concentration response curves to the adrenergic vasoconstrictor Phe in third-order mesenteric arteries incubated ex vivo with or without l-NAME. Foremost, obese MC4R^+/− pregnant rats had significantly reduced sensitivity to Phe-induced vasoconstriction. However, incubation with l-NAME normalized the contractile response to wild-type levels, indicating that the blunted constriction in MC4R^+/− pregnant rats was mediated by NOS (Fig. 2, A and B). l-NAME did not alter Phe-induced contraction in MC4R^+/+ pregnant rats (Fig. 2, A and B). Furthermore, the sensitivity (EC₅₀) of KCl-induced vasoconstriction was also significantly (P = 0.05) attenuated in obese (57 ± 3 mM, n = 6) versus lean (48 ± 2 mM, n = 5) pregnant rats.

Fig. 2.

Vascular reactivity curves (A) and sensitivity in logEC₅₀ (B) to phenylephrine (Phe)-induced constriction, calculated as a percentage of response at the final dose, in third-order mesenteric artery rings isolated from lean MC4R^+/+ (n = 5–9) and obese MC4R^+/− (n = 5–7) pregnant rats at GD 19. Rings were treated ex vivo in organ baths with N^G-nitro-l-arginine methyl ester (l-NAME, 100 μM). *P < 0.05 vs. untreated rings from MC4R^+/+ pregnant rats; †P < 0.05 vs. untreated rings from MC4R^+/− pregnant rats.

Effects of obesity and NOS inhibition on maternal hemodynamics.

Mean arterial pressure (MAP) was measured during late pregnancy in MC4R^+/+ and MC4R^+/− rats. Two-way ANOVA revealed that MAP was significantly elevated at GD 19 in the obese versus lean pregnant rats (Fig. 3A).

Fig. 3.

Mean arterial blood pressure (MAP, A) and glomerular filtration rate (GFR, B) at GD 19 in lean MC4R^+/+ (n = 13–16) and obese MC4R^+/− (n = 8–12) pregnant rats. Inset: results from the two-way ANOVA, where P_S represents statistics comparing rat genotypes as a whole, regardless of l-NAME; P_L represents statistics comparing differences between l-NAME treatment as a whole, regardless of rat genotype; and P_S*L represents statistics examining whether the differences found are dependent on both rat strain and l-NAME treatment.

To determine whether NOS-mediated regulation of blood pressure was reduced in obese pregnant rats, obese and lean pregnant rats were administered the nonselective NOS inhibitor l-NAME in their drinking water from GD 14 to 19. l-NAME treatment almost completely abolished total NOS activity in mesenteric arteries from both MC4R^+/− (38 ± 10 vs. 5 ± 1 pmol/mg per 30 min, P < 0.05, N = 8–9/group) and MC4R^+/+ (49 ± 8 vs. 9 ± 1 pmol/mg per 30 min, P < 0.05, N = 6–8/group) pregnant rats. Western blot analyses revealed that NOS3/β-actin expression in untreated versus l-NAME-treated obese MC4R^+/− pregnant rats was 0.68 ± 0.13 vs. 0.53 ± 0.12, P = 0.92, N = 4/group, respectively, and lean MC4R^+/+ pregnant rats 1.07 ± 0.15 vs. 0.77 ± 0.23, P = 0.66, N = 3–4/group), respectively.

The l-NAME treatment significantly increased MAP in lean MC4R^+/+ pregnant rats, and this increase was similar in obese MC4R^+/− pregnant rats (Fig. 3A). Here, the Δ for the blood pressure rise in MC4R^+/+ pregnant rats was 31 mmHg, whereas it was 29 mmHg in MC4R^+/− pregnant rats. The two-way ANOVA revealed that GFR was not different between untreated obese or lean pregnant rats or after l-NAME treatment, even though this chronic l-NAME decreased this value by 18% in lean and 10% in obese pregnant rats (Fig. 3B).

Pregnancy biometrics following NOS inhibition.

Average fetal weight (Fig. 4A) was not different between obese MC4R^+/− and lean MC4R^+/+ pregnant rats. The only effect of l-NAME treatment on pregnancy biometrics was that it reduced average fetal weight in the obese pregnant rat group (Fig. 4A). Average placental weight (Fig. 4B) was less in the obese pregnant group. When total fetal (Fig. 4C) and total placental (Fig. 4D) weights were determined, both of these measures were lower in the obese versus lean pregnant groups. This seemed to be a result of the greater fetal reabsorption rate (Fig. 4E) and reduced placental sufficiency (Fig. 4F) found in the obese pregnant group.

Fig. 4.

Average fetal weight (A), average placental weight (B), total fetal weight (C), total placental weight (D), fetal reabsorption rate (E), and placental sufficiency (F) GD 19 in lean MC4R^+/+ (n = 13–16) and obese MC4R^+/− (n = 8–12) pregnant rats. *P < 0.05 vs. untreated MC4R^+/− pregnant rats; †P < 0.05 for MC4R^+/− vs. MC4R^+/+ pregnant rats.

DISCUSSION

We report that there is no evidence of alterations in NOS-mediated regulation of blood pressure in MC4R^+/− obese pregnant rats. This is even though they had increased body weight, total body fat mass, visceral white adipose tissue, and blood pressure and that the higher blood pressure was accompanied by increase sFlt-1 levels, reductions in circulating cGMP levels, and reduced total NOS enzymatic activity and expression of the endothelial NOS isoform (NOS3) in small mesenteric arteries. In contrast, ex vivo studies demonstrated an increase in the ability of NOS to control vascular tone in the obese, hypertensive pregnant rats. Indeed, incubating third-order mesenteric artery rings with a NOS inhibitor increased the vasoconstriction response to phenylephrine in obese, but not lean pregnant rats. However, there was no difference in the degree by which l-NAME raised blood pressure between obese and lean pregnant rats. The efficacy of l-NAME to inhibit this enzyme was evident by the almost complete inhibition of small artery NOS activity in both treated obese and lean pregnant rats.

Although this has been well documented that obesity increases the risk for hypertensive disorders of pregnancy (5, 15), the mechanisms responsible are unknown. In the human literature, it is evident that there are reductions in NOS in preeclampsia, including lower circulating levels of the surrogate measure of NO bioavailability, cGMP, especially in those with fetal intrauterine growth restriction (22), but the effects of obesity on this were not examined. Suggesting that obese pregnancies are at increased risk for reduced NO bioavailability and hypertension, Mayret-Mesquiti et al. (14) showed that exogenous infusion of triglycerides and cholesterol reduced circulating nitrite in lean, healthy pregnant women. To address this important unanswered question, we utilized the MC4R^+/− rat model that presents with significantly increased body weight and fat mass in pregnancy to test the hypothesis that there is increased blood pressure due to reduced NOS-mediated regulation of blood pressure in obese pregnant rats.

The rationale for testing this hypothesis was bolstered by the finding of increased circulating sFlt-1 levels in obese pregnant rats. Our laboratory has previously shown that sFlt-1 targets the maternal endothelium where it reduces NO production (18). The understanding of this mechanism is important, as normal pregnancy is associated with increasing blood pressure dependence of NOS. This was demonstrated by the greater l-NAME-induced hypertension in pregnant versus nonpregnant Sprague-Dawley rats (9). In the present study, we found that l-NAME administration into MC4R^+/+ pregnant rats, from GD 14–19, dramatically increased their blood pressure. The timing of this administration was based on the finding that many disorders of hypertension in pregnancy present during the third trimester (21). Based on this knowledge and the increased circulating sFlt-1 levels in obese pregnant rats, we examined whether the obese MC4R^+/− pregnant rats had reduced vascular expression and activity levels of NOS and reduced blood pressure control by this system.

Most studies conducted to examine the mechanisms mediating the pathogenesis of obesity-induced hypertension have been conducted in male animals showing reduced NO bioavailability. This has been found in high-fat diet models (6, 10) and Zucker rats (8, 12). Less is known in females, especially during pregnancy, and whether this could contribute to increased rates of hypertension in obese pregnancies. We found increased circulating levels of sFlt-1, which is known to inhibit NO production. Indeed, there were reduced levels of the surrogate measure of NO, cGMP, in the obese versus lean pregnant rats. To more directly examine vascular levels of this system, total NOS activity was examined in a resistance artery bed. Complementary to the circulating cGMP measure, NOS activity and expression were reduced in the mesenteric arteries isolated from obese pregnant group compared with wild-type controls.

At this point, it was suspected that NOS control of vascular tone would be reduced in obese pregnant rats having increased blood pressure and reduced vascular NOS signaling. This would be detected by increased responsiveness to vasoconstrictors via reduced NOS-mediated buffering of vasoconstriction. However, contrary to our hypothesis, these obese pregnant rats had reduced sensitivity to Phe-induced vasoconstriction, which was in fact due to increased NOS-mediated buffering of vasoconstriction. It is not yet understood if the reduced circulating cGMP levels are of vascular origin. However, these vascular function data suggested that the reduced expression of NOS was compensated by increased NOS function to control vascular tone. Thus the increased blood pressure found at GD 19 in the untreated, obese pregnant rats may not be due to NOS dysfunction, and if anything, there may not be a difference in the ability to raise blood pressure between lean and obese pregnant rats. Indeed, our in vivo studies using l-NAME administration demonstrated that there was no difference in the degree of blood pressure rise following NOS inhibition between obese and lean pregnant rats.

The conserved NOS regulation of blood pressure and the increased NOS buffering of vasoconstriction, in light of reduced biomarkers of NO signaling, in our obese pregnant rats supports our previous study. There we reported in vascular reactivity experiments that the sensitivity to the NO donor SNP was enhanced in third-order mesenteric arteries from the obese MC4R^+/− over lean MC4R^+/+ pregnant rats, at GD 19. Thus, although there are reduced biomarkers of NOS and the circulating marker of NO signaling cGMP, that at the vascular level, there is enhanced local responsiveness of the smooth muscle to NO. Therefore, we collectively conclude that, although there is a deficit in biochemical indices of NO/NOS in obese pregnancies, the regulation of blood pressure and vascular function is still dependent on, and able to respond to, NO.

Although these obese pregnancies were accompanied by reduced NOS expression and activity and increased blood pressure, we actually found an increased role for NOS in modulating vasoconstriction in isolated arteries. This is similar to what has been found in angiotensin II-induced hypertension in male Sprague-Dawley rats where there was reduced basal and acetylcholine-stimulated cGMP release from mesenteric arteries (7). Even so, there was a tremendous upregulation in the ability of NOS to mediate vasorelaxation that helped to largely preserve maximal endothelial-dependent vasorelaxation. However, this was not due to NOS-derived NO, but to NOS-derived hydrogen peroxide (H₂O₂). Follow-up studies should examine the role H₂O₂ in regulation of vascular tone in our obese pregnant rats, and whether this mechanism helps to maintain the NOS-mediated control of blood pressure in them. In addition, the neuronal NOS isoform NOS1 has been demonstrated to put a brake on vasoconstriction in mesenteric arteries, although studied in males (24). It would be interesting to study the degree of its contribution to the increased NOS control of vascular tone in obese pregnant rats. Future studies should also determine whether sFlt-1, and even angiotensin II, in combination with the increased circulating obesity-related metabolic factors induce such a shift in NOS activity and function in obese pregnancies.

The exact mechanisms behind the increased blood pressure of the untreated obese pregnant rats remain unclear. Furthermore, we did not examine which of the four hypertensive disorders of pregnancy are modeled by our obese pregnant rats. We only present that they have increased blood pressure at the end of pregnancy. Future telemetry-based studies will determine whether our obese female rats have increased blood pressure before pregnancy and/or whether pregnancy induces hypertension in the same obese rats.

These obese rats also are heterozygous deficient for the MC4R. This obese rat model was generated by a pharmacological-induced mutation of this receptor (17). It is known, at least in males, that MC4R homozygous-deficient rats have extreme obesity but are normotensive and lack a response to hyperleptinemia-induced hypertension (4). Furthermore, l-NAME-induced hypertension was blunted by intracerebroventricular inhibition of this receptor in male mice (3). This was not examined in females. However, a recent study reported that nonpregnant female MC4R^−/− rats have increased blood pressure (13). In contrast, male MC4R^−/− rats have similar blood pressure when compared with sex-matched MC4R^+/+ control rats (4). Finally, we believe that obesity and obesity-related metabolic factors (independent of MC4R activation) may lead to elevations in blood pressure via alterations in placental release of vasoactive factors or increased vascular sensitivity to placental-derived factors that we have previously shown to be involved in pregnancy-induced hypertension.

Perspectives and Significance

Our data support that obesity, regardless of MC4R deficiency, is associated with increased blood pressure during pregnancy. Interestingly, this does not seem to be dependent on reductions in the ability of NOS to regulate blood pressure. This suggests that other prohypertensive mechanisms are at play. Here, we found that the antiangiogenic factor sFlt-1 was greater in the obese versus lean pregnant rats. We have previously found that sFlt-1 infusion increases blood pressure via endothelin-1, a potent vasoconstrictor (19). Moreover, others have also found that sFlt-1 infusion, although studied in males, increased blood pressure via a thromboxane-dependent mechanism (1). These pathways should be examined in our obese pregnant rat model. Furthermore, based on our findings, we propose that the hypertensive response to surgically induced placental ischemia and hypoxia utilizing the reduced uterine perfusion pressure model maybe exaggerated in the presence of high-fat diet feeding or obesity. Placental ischemia is known to drive increases not only in circulating sFlt-1 levels, but also numerous proinflammatory cytokines that promote this maternal hypertensive response. This is clinically relevant as placental ischemia-reperfusion is a major theory in the pathogenesis of preeclampsia. An exaggerated blood pressure response to this insult under obesogenic conditions may explain how obese women are at increased risk for preeclampsia.

GRANTS

Funding for this study includes T32HL105324-01 (to F. T. Spradley), P20GM104357 (to F. T. Spradley and J. M. Sasser), P01HL051971 (to J. P. Granger), R01HL108618 (to J. P. Granger), and K01DK095018 (to J. M. Sasser).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

F.T.S., J.M.S., J.B.M., J.C.S., and J.P.G. conception and design of research; F.T.S., J.M.S., and J.B.M. performed experiments; F.T.S., J.M.S., J.B.M., and J.C.S. analyzed data; F.T.S., J.M.S., J.C.S., and J.P.G. interpreted results of experiments; F.T.S. prepared figures; F.T.S. drafted manuscript; F.T.S., J.M.S., J.C.S., and J.P.G. edited and revised manuscript; F.T.S., J.M.S., J.B.M., J.C.S., and J.P.G. approved final version of manuscript.