| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Obstetrics and Gynecology,3
Molecular Genetics,4
Pharmacology and Toxicology5 and
the Research Institutes GROW6 and
CARIM,7 University of Maastricht, 6200 MD Maastricht, The Netherlands
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-actin) and proliferation (Ki-67) immunostaining. We compared the pregnancy-related changes longitudinally and, together with the data on pregnancy outcome, transversally by analysis of variance with Bonferroni correction. During pregnancy, the increases in radius and medial cross sectional area of Nos3–/– UA was significantly less than those of Nos3+/+ UA. Smooth muscle cell dedifferentiation and proliferation were impaired in gravid Nos3–/– mice as deduced from the lack of change in the expression of smoothelin and smooth muscle -actin, and the reduced Ki-67 expression. Until 17 days of gestation, litter size did not differ between both genotypes, but at birth the number of viable newborn pups and their weights were smaller in Nos3–/– than in Nos3+/+ mice. We conclude that absence of eNOS adversely affects UA remodeling in pregnancy, which may explain the impaired pregnancy outcome observed in these mice.
arterial remodeling, female reproductive tract, knockout mice, nitric oxide, pregnancy, uterine artery
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Nitric oxide (NO) increases blood flow by relaxation of the tunica media of the blood vessel wall. Endothelial nitric oxide synthase (eNOS) is the predominant NOS isoform in the vessel wall [4]. Numerous factors modulate eNOS activity. Among these, increases in wall shear stress resulting from increases in blood flow have received considerable attention [5, 6]. Under experimental conditions that reduce blood flow, a decrease in NO synthesis leads to smooth muscle cell (SMC) proliferation, migration, and subsequent neointima formation (defined as the layer of smooth muscle cells between the lamina interna and the endothelium) [7, 8]. The significance of eNOS in arterial remodeling is further strengthened by the effect of NO on the expression of matrix metalloproteinases and their tissue inhibitors (TIMPs) [9].
A number of reports suggest eNOS also to be involved in cardiovascular adaptations during pregnancy. Levels of eNOS and NO are elevated in the UA during pregnancy [10, 11]. It is conceivable that higher circulating levels of estrogen are at least in part responsible for the higher UA expression of eNOS since estrogen has been reported to selectively up-regulate eNOS expression in the reproductive arteries of oophorectomized nonpregnant ewes [12]. In addition, previous microarray data in mice suggest that estrogen regulates eNOS [13] and that estrogen stimulation increases eNOS protein expression in vascular endothelium [14].
Two types of interventional studies have assessed the role of eNOS in arterial remodeling: 1) eNOS overexpression and 2) NOS inhibition. eNOS overexpression, as achieved in transgenic mice or by adenoviral transfection, suppresses vascular lesion formation [15]. The inhibition of the activity of all NOS isoforms by compounds such as NG-nitro-L-arginine methyl ester (L-NAME) also blunts flow-dependent remodeling [16, 17]. Although several inhibitors are available, most of them are not isoenzyme specific and have additional pharmacological effects. To circumvent these drawbacks, eNOS knockout mice have been generated [18]. In these mice, eNOS deficiency enhances neointima formation under various conditions [4, 7]. In addition, hypertension [19, 20] and a poor reproductive outcome, characterized by fetal growth retardation and fetal mortality near term, have been reported [21].
During pregnancy, an increase in UBF coincides with structural changes of the uterine arterial wall including hypertrophy and hyperplasia of the arterial SMC [2, 22]. This remodeling of the UA requires transient dedifferentiation of medial SMCs [3]. In a previous study, we showed that structural changes of the UA take place before the largest increase in UBF. We suggested an estrogen/NO pathway to play a role in UA remodeling [3]. The present study was designed to test the hypothesis that maternal eNOS is an important mediator in UA remodeling during pregnancy. To this end, we determined at specific time points in first pregnancy of wild-type and eNOS-deficient mice the reproductive performance, vessel morphometry, and SMC phenotype.
|
MATERIAL AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animal Preparation
The eNOS mutation was in a C57BL/6 background [19]. Control wild-type mice were aged-matched virgin female C57BL/6 mice (
21 g, n = 6) purchased from Charles River, Maastricht, The Netherlands. All mice were 12 to 14 wk of age at the time of study. Mice had free access to food and water and were maintained on a 12L:12D cycle. Pregnancy was achieved by mating with an experienced wild-type male (age 12–14 wk). For the Nos3–/– females, this crossing yielded pups that were heterozygous for the eNOS mutation. The recovery of a vaginal sperm plug was considered to correspond with Day 1 of pregnancy. Pregnant mice were killed at Gestational Days 5, 11, or 17 (early, mid-, and late pregnancy, respectively) of their first pregnancy. An additional group of mice for each genotype was used to study reproductive outcome at birth. Six animals of both genotypes were studied in each of these gestational age-groups. All animals were weighed before being killed.
Reproductive Outcome
Reproductive outcome was determined in pregnant mice on the basis of litter size (number of uterine implants), number of viable fetuses per litter (pink fetuses in amniotic fluid reacting to mechanical stimuli), and term individual pup weight. Although litter size could be discerned in 11-day pregnant mice, it was not feasible to reliably determine individual pup weight because of size and poor demarcation of the feto-placental units. However, in 17-day pregnant mice, all fetuses could be carefully removed, dried, and weighed after unilateral hysterectomy.
Tissue Preparation
As described previously [3], mice were anesthetized using an intraperitoneal injection of pentobarbital (10 mg/kg). Briefly, a midline incision was made in the abdomen and neck. One uterine and one carotid artery were dissected under a stereomicroscope (Zeiss) and stored at –80°C. Meanwhile, the contralateral uterine and carotid arteries were kept in situ for perfusion and tissue fixation. Frozen arteries were later used for immunohistochemistry.
Histology, Morphometry, and Immunohistochemistry
The remainder of the arterial tree was then perfused at 80–100 mm Hg with phosphate-buffered saline and 4% phosphate-buffered formalin, pH 7.4, both containing 0.1 mg/ml sodium-nitroprusside (Sigma Chemical, St. Louis, MO), through a catheter inserted into the left ventricular apex. This procedure enabled vessel fixation at their maximal diameter for the prevailing in vivo arterial pressure. The remaining uterine and carotid artery, together with the abdominal and thoracic aortae, were dissected, fixed overnight in 4% phosphate-buffered formalin, and stored in ethanol. A segment of each dissected artery (1–2 mm in length) from nonpregnant and pregnant mice was embedded in paraffin and transversely sectioned in 4-µm slices. Cross sections were stained with Lawson's solution (Boom, Meppel, The Netherlands) to visualize the internal and external elastic laminae. Internal and external circumferences, demarcated by the internal and external elastic laminae, were measured (Sigma Scan, Jandel Scientific, Corte Madera, CA). From these values, medial cross-sectional area (CSA), internal radius, and media thickness were calculated for each section. Additional cross sections were stained with eosin and hematoxylin, and the number of SMC nuclei in the medial layer of the vessel wall of each cross section was counted. In addition, cross sections were stained with Sirius red to determine the collagen content in the media of each vessel using standard procedures, while Lawson's solution-stained cross sections were used to determine the medial elastin content in each vessel, using a computerized morphometry system (Sigma Scan, Jandel Scientific).
The frozen uterine and carotid arteries were cross-sectioned on a cryostat (5 µm) and mounted on gelatin-coated slides for immunohistochemistry. To determine the expression of vascular proteins during remodeling, vessels were stained for smooth muscle -actin (-SMA) and smoothelin, a recently described late differentiation marker of vascular SMCs [23]. In addition, to quantify proliferation of smooth muscle cells during the process of remodeling, vessels were stained for the proliferation marker Ki-67 [24]. Briefly, after blocking endogenous peroxidase activity, sections were treated with an avidin/biotin blocking kit (Vector Laboratories) followed by incubation with the biotinylated mouse monoclonal smoothelin antibody (R4A, 1:40). Smooth muscle -actin and Ki-67 reactivity were assessed by incubating the sections with a rabbit polyclonal antibody against either -SMA (1:3000, Sigma) or Ki-67 (1:50, DAKO) followed by incubation with a swine anti-rabbit antibody (1:1000, Amersham Life Sciences). All sections were counterstained with hematoxylin. For negative controls, sections were incubated with the second antibody only and showed no immunoreactivity.
Data and Statistical Analysis
All sections were evaluated blindly by four independent observers, and sections for -SMA and smoothelin were scored semiquantitatively according to four levels of intensities: from 0 if no staining was detected to 4 if the entire medial layer was stained. Proliferating SMCs, identified by Ki-67 staining, displayed dark brown nuclei, while nondividing cells displayed blue nuclei because of counterstaining with hematoxylin. SMC nuclei were counted in the media of each artery, and the total number of proliferating and nonproliferating cells could be quantified. From these values, the ratio of proliferating/total SMC in each artery was calculated. All data are expressed as mean ± SD. To test differences of mean values between nonpregnant and pregnant animals in each genotype (longitudinal) and between genotype (transversal), statistical significance of differences was evaluated by analysis of variance with Bonferroni correction. Values were considered significantly different when P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Increase in maternal body weight during pregnancy was larger in Nos3+/+ than in Nos3–/– mice (62% vs. 41%, respectively; Table 1). Although litter size was comparable between both genotypes during the course of pregnancy, there were significantly less viable pups in the litters of Nos3–/– than in those of Nos3+/+ mice at birth (4.4 ± 0.6 vs. 7.2 ± 0.5, respectively; P < 0.001). Moreover, mean fetal weight was lower in Nos3–/– than in Nos3+/+ dams, both at Pregnancy Day 17 (0.47 ± 0.05 g vs. 0.59 ± 0.05 g, respectively; P < 0.05; Table 1) and at birth (1.25 ± 0.06 g vs. 1.37 ± 0.05 g, respectively; P < 0.05; Table 1). In addition, mean placental weight was reduced in Nos3–/– mice as compared to Nos3+/+ mice at Pregnancy Day 17 (0.12 ± 0.02 g vs. 0.14 ± 0.03 g, respectively; P < 0.05). At term pregnancy, we did not observe any delay or differences in parturition between both genotypes.
|
Differences Between UA of Nonpregnant Nos3+/+ and Nos3–/– Mice
Lumen radius, medial CSA, medial thickness, medial SMC number, and the densities of collagen and elastin did not differ significantly between UA of nonpregnant Nos3+/+ and Nos3–/– mice (Figs. 1–4). The same applied to smoothelin (Fig. 1, panels G and J, and Fig. 3C) and -SMA immunoreactivity (Fig. 1, panels M and P, and Fig. 3D).
|
|
|
Pregnancy-Related UA Morphometrical Changes
Differences between Nos3–/– and Nos3+/+ mice were already observed as early as Day 5 of pregnancy (Fig. 2). During pregnancy, the radius and the CSA of the UA of Nos3+/+ mice increased gradually and were significantly larger on Days 11 and 17 as compared to nonpregnant mice (Fig. 1, panels A, B, and C, and Fig. 2). However, in Nos3–/– mice, pregnancy induced only a modest increase in the radius, medial CSA, and medial thickness of the UA as compared to nonpregnant Nos3–/– mice (Fig. 1, panels D–F, and Fig. 2). While structural changes reached statistical significance by 11 days of gestation in Nos3+/+ mice, statistical significance of structural changes was reached only at 17 days of gestation in Nos3–/– mice.
|
Pregnancy-Related UA Cytological and Extracellular Changes
In Nos3+/+ mice, pregnancy induced a number of cytological changes in UA medial SMC that were not observed in Nos3–/– mice. The SMC of Nos3+/+ mice displayed a transient decrease in smoothelin (Fig. 1, panel H, and Fig. 3C) and a progressive decrease in -SMA (Fig. 1, panels N and O, and Fig. 3D), coinciding with a transient increase in the number of Ki-67 positive cells by Day 11 of pregnancy (Fig. 3B). Also, the number of medial SMC increases in the course of pregnancy (Fig. 3A). The observed changes are indicative for dedifferentiation and proliferation of the medial SMC. In Nos3–/– mice, pregnancy did not induce changes in the number of medial SMCs and in Ki-67 expression (Fig. 3, A and B, respectively). In addition, pregnancy was not accompanied by changes in the expression of smoothelin and -SMA in Nos3–/– mice (Fig. 1, panels K, L and Q, R, respectively, and Fig. 3D). Neither Nos3+/+ nor Nos3–/– responded to pregnancy with a change in the UA density of the extracellular matrix components collagen and elastin (Fig. 4). Nevertheless, because of the larger increase in medial CSA in Nos3+/+ UA, absolute collagen and elastin synthesis was larger in the latter genotype.
Pregnancy-Related Changes in Nonuterine Vessels
Pregnancy had no appreciable effect on the structural and cellular properties of the carotid arteries and abdominal and thoracic aortae of Nos3+/+ and Nos3–/– mice (data not shown).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The endothelium has been demonstrated to perceive changes in blood flow [26] and subsequently orchestrate the regulation of vascular tone [27, 28]. Chronic increase in blood flow increases eNOS expression and subsequent NO production. The latter compound induces vasodilation [29, 30]. While eNOS is expressed in endothelial cells, neuronal NOS (nNOS) expression is abundant in vascular SMC [31]. As suppression of NO synthesis with L-NAME may [16] or may not [32] inhibit flow-induced outward remodeling, the relationship between synthesis of and vasodilation by eNOS/nNOS on the one hand and arterial remodeling on the other remains unclear [32, 33].
In the present study, we investigated the effect of pregnancy on structural and cellular characteristics of the SMC of the UA in Nos3–/– mice. Previously, we showed that during pregnancy the UA displayed outward hypertrophic remodeling of the vessel wall [3]. This was accompanied by dedifferentiation and proliferation of the SMC. In Nos3–/– mice, dedifferentiation and proliferation of medial SMCs and subsequent remodeling of the UA are impaired, and the growth and survival of the fetuses are reduced. During pregnancy, lumen diameter and medial mass of the UA increase substantially in Nos3+/+ mice [3, 25] but not in Nos3–/– mice. Dedifferentiation of SMCs as deduced from a reduced expression of smoothelin and smooth muscle -actin and considered a prerequisite for proliferation [34] appears absent in SMCs of Nos3–/– mice. This corroborates with a stable number of UA SMCs in general and Ki-67 positive cells in particular. In Nos3+/+ mice, both parameters display a significant increase during pregnancy. Although the increase in medial CSA and radius of the UA is impaired in Nos3–/– mice as compared with Nos3+/+ mice, a slight increase for both morphometric parameters was found in the Nos3–/– mice at the end of pregnancy. This increase has to be contributed to hypertrophy of the medial SMC since neither the number of these cells nor the proportional volume of extracellular matrix components, such as collagen and elastin, had changed. Thus, absence of eNOS results in a delayed and impaired UA remodeling. Blockade of NO synthase and eNOS deficiency increase blood pressure [19, 35] and prevent the fall in blood pressure during pregnancy in rodents [20, 36]. Our present data do not allow one to fully dissociate between direct and indirect mechanisms that might underlie blunted UA remodeling in Nos3–/– mice.
In studies in which shear stress was shown to induce vessel remodeling, eNOS was demonstrated to modulate the remodeling process. The focus of these studies was inward remodeling in response to vessel manipulation [7, 8, 37, 38]. However, the mechanism of the latter may differ from that of outward hypertrophic remodeling, which takes place in the UA during the first half of pregnancy and is characterized by growth of the CSA and medial thickness without concomitant lumen widening (rise in wall-to-lumen ratio) [3]. We suggested that paracrine and/or hormonal factors such as estrogens might initiate UA remodeling since estrogens have been shown to affect vascular remodeling [39, 40]. Previous reports show that sex steroids modulate NO production. Protein expression of inducible NO synthase (iNOS) in the uterus and cervix is elevated in third-trimester pregnancy [41, 42] but has not been found involved in outward hypertrophic remodeling. In addition, microarray data in mice suggest that estrogen regulates eNOS [13], and there is in vitro evidence of a possible involvement of estrogen in the pregnancy-associated increase in eNOS [43]. The results of the present study indicate that during early pregnancy, a possible role of estrogens in UA remodeling could be mediated by eNOS. The rapidly elevated levels of estrogens that take place during pregnancy induce a selective up-regulation of eNOS and nNOS in the UA [12, 44]. In gravid Nos3–/– mice, the increase in CSA and lumen observed in the second half of pregnancy might be induced by estrogen and mediated by nNOS. Recent reports show that nNOS is present in vascular SMC [31] and is implicated in responses to changes in blood flow or vascular injury [12, 45]. In these investigations, increase of nNOS levels are reported to occur a week or more after treatment [31, 38]. This delay is in conformity with the onset of changes observed in the second half of pregnancy in Nos3–/– mice. Thus, the role of eNOS in outward remodeling of the UA during pregnancy differs from that induced by ligation or denudation-induced inward remodeling.
We observed reduced maternal body weight gain, fetal growth retardation, reduced placental weight, and increased perinatal mortality in eNOS-deficient mice. This resulted in a decline in the number of viable pups per litter at term. Whereas other studies have dealt with offspring that were homozygous for the eNOS deletion [19, 20], this study has used a breeding scheme where Nos3–/– females were crossed with Nos3+/+ males and thus yielded pups that were heterozygous for the eNOS deletion. Heterozygous eNOS mice have been shown to be physiologically similar to wild-type mice [19, 46, 47]. As a consequence, the differences observed in vascular remodeling of the maternal uterine bed and pregnancy outcome between Nos3+/+ and Nos3–/– mice are attributable not to the genotype of the offspring but solely to the maternal genotype. The delayed increase of the UA lumen, as a consequence of impaired remodeling, may enable a minimal rise in UBF in the first half of pregnancy. However, this small flow rise may be sufficient to allow normal placentation and early embryonic development. Similar litter size and number of viable pups throughout pregnancy support this concept. The impact on pregnancy outcome of diminished UA remodeling can be expected to be largest in late pregnancy, when UBF demands increase substantially. Various studies have shown that inhibition of NO synthesis during pregnancy causes decreased placental and fetal perfusion resulting in fetal growth restriction [48, 49]. We therefore suggest that maternal eNOS deficiency contributes to impaired UA remodeling and limits the rise of utero-placental perfusion, in particular during the third trimester of pregnancy, thereby compromising growth and survival of the fetuses. We did not observe delayed or impaired parturition. All three NOS isoforms (endothelial, neuronal, and inducible) are involved in the onset of labor [41, 42]. A deficiency in one of the three isoforms may delay or impair parturition. However, the inducible isoform of NOS (iNOS) seems more important in the onset of parturition than the endothelial or neuronal NOS isoforms [41, 42]. A moderate compensatory up-regulation of other NOS isoforms could therefore account for the unimpaired process of parturition observed in the present study.
Our findings in young Nos3–/– mice resemble closely those in delayed first pregnancy in wild-type mice. In 40-wk-old mice compared to 12-wk-old mice, changes in UA diameter, CSA, SMC dedifferentiation and proliferation, and fetal survival are all reduced [3]. From this study, it can be deduced that the impaired remodeling appears (in part) to be related to reduced levels of eNOS/NO since aging is associated with general endothelial dysfunction [50–52].
In conclusion, our findings indicate that the structural and cellular changes characteristic of UA remodeling during pregnancy are markedly reduced in eNOS-deficient mice, contributing to a poorer pregnancy outcome. Thus, eNOS plays a critical role in the adaptive processes seen in the UA in response to pregnancy.
|
FOOTNOTES |
|---|
2 Correspondence: G.J.J.M. van Eys, Department of Molecular Genetics, University of Maastricht, P.O. Box 616, 6200 MD Maastricht, The Netherlands. FAX: 31 43 388 4574; g.vaneys{at}gen.unimaas.nl
Received: 21 July 2004.
First decision: 6 August 2004.
Accepted: 10 January 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Gassmann, A. Manini, T. Stallmach, B. Saam, G. Kuhn, B. Grenacher, A. Y. Bogdanova, and J. Vogel Abortion in Mice with Excessive Erythrocytosis Is Due to Impaired Arteriogenesis of the Uterine Arcade Biol Reprod, June 1, 2008; 78(6): 1049 - 1057. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Pallares and A. Gonzalez-Bulnes Intrauterine Growth Retardation in Endothelial Nitric Oxide Synthase-Deficient Mice Is Established from Early Stages of Pregnancy Biol Reprod, June 1, 2008; 78(6): 1002 - 1006. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. N. Van Vliet and L. L. Chafe Maternal Endothelial Nitric Oxide Synthase Genotype Influences Offspring Blood Pressure and Activity in Mice Hypertension, March 1, 2007; 49(3): 556 - 562. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Kulandavelu, D. Qu, and S. L. Adamson Cardiovascular Function in Mice During Normal Pregnancy and in the Absence of Endothelial NO Synthase Hypertension, June 1, 2006; 47(6): 1175 - 1182. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D.-b. Chen, S. Jia, A. G. King, A. Barker, S.-m. Li, E. Mata-Greenwood, J. Zheng, and R. R. Magness Global Protein Expression Profiling Underlines Reciprocal Regulation of Caveolin 1 and Endothelial Nitric Oxide Synthase Expression in Ovariectomized Sheep Uterine Artery by Estrogen/Progesterone Replacement Therapy Biol Reprod, May 1, 2006; 74(5): 832 - 838. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. W.H. van der Heijden, Y. P.G. Essers, M. E.A. Spaanderman, J. G.R. De Mey, G. J.J.M. van Eys, and L. L.H. Peeters Uterine Artery Remodeling in Pseudopregnancy Is Comparable to That in Early Pregnancy Biol Reprod, December 1, 2005; 73(6): 1289 - 1293. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
