Vascular smooth muscle (VSM) maturation is developmentally regulated and differs between vascular beds. The maturation and contribution of VSM function to tissue blood flow and blood pressure regulation during early gestation are unknown. The carotid artery (CA) contributes to fetal cerebral blood flow regulation and well being. We studied CA VSM contractility, protein contents, and phenotype beginning in the midthird of ovine development. CAs were collected from early (88–101 day of gestation) and late (138–150 day; term = day 150) fetal (n = 14), newborn (6–8 day old; n = 7), and adult (n = 5) sheep to measure forces in endothelium-denuded rings with KCl, phenylephrine, and ANG II; changes in cellular proteins, including total and soluble protein, actin and myosin, myosin heavy chain isoforms (MHC), filamin, and proliferating cell nuclear antigen; and vascular remodeling. KCl and phenylephrine elicited age- and dose-dependent contraction responses (P < 0.001) at all ages except early fetal, which were unresponsive. In contrast, ANG II elicited dose responses only in adults, with contractility increasing greater than fivefold vs. that shown in fetal or neonatal animals (P < 0.001). Increased contractility paralleled age-dependent increases (P < 0.01) in soluble protein, actin and myosin, filamin, adult smooth muscle MHC-2 (SM2) and medial wall thickness and reciprocal decreases (P < 0.001) in nonmuscle MHC-B, proliferating cell nuclear antigen and medial cellular density. VSM nonreceptor- and receptor-mediated contractions are absent or markedly attenuated in midgestation and increase age dependently, paralleling the transition from synthetic to contractile VSM phenotype and, in the case of ANG II, paralleling the switch to the AT1 receptor. The mechanisms regulating VSM maturation and thus blood pressure and tissue perfusion in early development remain to be determined.
- myosin heavy chain isoforms
- nonmuscle myosin
- fetal development
- receptor and nonreceptor function
- smooth muscle growth
- angiotensin II
smooth muscle development normally proceeds in a well-orchestrated manner before and after birth (5, 6, 11, 14, 41, 53). These changes occur in three phases: cellular differentiation, functional maturation, and growth (42). During differentiation, progenitor cells derived from the mesenchyme are transformed into immature smooth muscle cells (SMC) localized to either visceral or vascular sites. The subsequent maturational changes result in developmentally regulated improvements in specific organ or vascular function essential for the well being and growth of the developing fetus and newborn. For example, functional maturation of the ovine bladder and umbilical artery smooth muscle occurs early in development (4, 6). The former is required for maintenance of fetal fluid balance and establishment of amniotic fluid volume, which permit normal lung development. The latter is essential in the regulation of fetal oxygen and nutrient uptake from the maternal placental circulation and probably blood pressure (4, 32). In contrast, uterine smooth muscle is not essential until sexual maturation; therefore, maturation is delayed long after birth (5). Thus the pattern and timing of SMC development vary with the organ or vascular bed studied.
Tissue blood flows and systemic blood pressure increase throughout development, modulating growth and adaptation through changes in vascular smooth muscle (VSM) function. However, the timing of these changes during development and the mechanisms regulating them are incompletely understood (10, 26, 30, 44, 48). Furthermore, it is evident that this differs between and within various vascular beds (4, 43, 53, 55). For example, VSM obtained in late ovine gestation from several peripheral vascular beds, including the femoral artery, demonstrates diminished contractile capacity, whereas umbilical artery VSM function resembles that shown in the adult (4). Thus umbilical artery VSM is considered to demonstrate precocious maturation compared with other peripheral vascular beds. It also has precocious expression of contractile proteins and the adult ANG II AT1-receptor (AT1R) subtype (4, 14). Thus observations of VSM function and maturation during development are likely to be specific to the vascular bed under study, and results cannot be generalized. Furthermore, this has not been thoroughly examined early in development.
Maturational changes in VSM function are associated with changes in cellular protein expression in near term and newborn animals (6, 11, 17, 43); however, this is not described in midgestation, e.g., <100 days gestation in the sheep and ≤28 wk in the human. Recent clinical advances have resulted in survival of neonates born at ≤28 wk (33). However, our understanding of the mechanisms that regulate tissue blood flow and blood pressure in these neonates is unclear (20), including the status of receptor expression and function, cell signaling, and biochemical parameters. Thus the contributions of the peripheral circulation to adaptive responses (9, 29, 34) and the maintenance of blood pressure are unclear (15). Characterizing the functional, phenotypic, and morphological changes in arterial VSM in early development will provide a basis for understanding the mechanisms that contribute to the normal and pathological regulation of tissue blood flow and blood pressure in extremely preterm neonates and the alterations in normal VSM programming that may later contribute to the genesis of adult-onset cardiovascular disease (25, 41, 47, 53).
The carotid artery (CA) contributes to fetal and neonatal cerebral blood-flow regulation (22, 24). Maintenance of cerebral perfusion is essential to normal development. Although the function of the CA has been extensively studied in near-term animals (19, 22, 43), there is no information about CA VSM characteristics and function earlier in development. Therefore, we sought to examine age-dependent changes in CA VSM beginning in the mid-third of ovine gestation and to characterize the changes in cellular proteins and vessel wall morphology that might contribute to or explain developmental changes in function. Because differences in nonreceptor- and receptor-mediated function may exist (4, 13), we studied VSM responses to KCl, the α1-agonist phenylephrine (PE), and ANG II, whose receptor ontogeny in CA VSM is well described (13, 14). We hypothesize that, in early development, VSM function is attenuated and responses to PE and ANG II differ because of changes in AT1R subtype expression.
Twenty-six sheep were included in these studies. Animals were divided into four groups: early (EF; n = 7, 88–101 days of gestation) and late (LF; n = 7, 138–150 days of gestation; term = 150 days) fetal, newborn (n = 7, 6–8 days old), and adult (n = 5) sheep. There were two twin gestations in the EF group, none in LF, and one in the newborn. Animals were euthanized with an intravenous bolus infusion of pentobarbital sodium (50 mg/kg) into the mother or newborn. The left and right common CAs were rapidly removed, and a segment 2.5–3.0 cm in length was placed in physiological salt solution (in mM: 120.5 NaCl, 4.8 KCl, 1.2 MgSO4, 1.2 NaH2PO4, 20.4 NaHCO3, 1.6 CaCl2, 10 dextrose, and 1.0 pyruvate) at room temperature and transported within 30 min to the laboratory for contraction studies. The remaining CAs were placed in iced PBS (8 mM Na2HPO4, 137 mM NaCl, pH 7.5), and the adventitia were removed by blunt and sharp dissection as previously described (4, 11). Half of these samples were opened along the long axis of the artery, and the endothelium was removed with a soft cotton swab (4). The endothelium-denuded samples were cut into strips, blotted dry to remove excess fluid, frozen in liquid nitrogen, and stored at −80°C until studied. The protocols used in these studies were approved by the Institutional Animal Care and Use Committee at the University of Texas Southwestern Medical Center.
Measurements of contractility.
Samples transported to the laboratory for study of contractile function had the adventitia carefully removed by sharp dissection using a dissecting microscope, rings 4–5 mm in length were cut, and the endothelium was carefully removed by turning a small forceps tip in the lumen. Endothelium removal was verified histologically, and the basal lamina and underlying VSM were intact. Each ring was placed on a stirrup attached to a transducer to measure force generation in a 25-ml volume chamber. After a 30-min equilibration period, rings were progressively stretched to obtain optimal length (L0), determining forces with 65 mM KCl (2). Utilizing L0, a dose-response curve was constructed for each ring with cumulative doses of KCl (10–120 mM) to determine nonreceptor-mediated responses and PE (10−8 to 10−5 M), a pure α1-agonist, to study adrenergic-receptor-mediated responses. ANG II responses were examined with the use of one dose per ring (10−8 to 10−5 M), since tachyphylaxis occurs with subsequent doses. Data were recorded on an electronic data-acquisition system (ACQuire, Gould Systems, Valley View, OH) in grams of force generated at L0. Because maturational differences occurred in several parameters (see results), each of which could potentially modify contractile responsiveness during development, data from each experiment and individual rings were normalized to the maximum adult response to 65 mM KCl and expressed as a percentage of the adult responses to 65 mM KCl. This permits comparisons between agonists and developmental groups.
Protein analysis and content.
Samples of frozen endothelium-denuded arteries (9.0–32.3 mg) were weighed and homogenized in 40× volumes of SDS buffer containing 2% SDS, 20% sucrose, and 0.4 M Tris (pH 6.8). The homogenates were divided into two aliquots. The first was used to determine the total homogenate protein content, which includes cellular and noncellular protein. The second was centrifuged at 10,000 g for 2 min, and the supernatant was removed to determine soluble or cellular protein contents in each sample. Aliquots were analyzed for protein concentrations using a bicinchoninic acid protein assay reagent kit (Pierce, Rockford, IL). Bromophenol blue and 2-mercaptoethanol were added to aliquots of the supernatant from each CA, and 20 μg of soluble protein were subjected to SDS-PAGE using 4–20% polyacrylamide minigels to determine the actual contents of total actin, myosin, and filamin as previously reported (4–6, 11). Gels contained molecular mass standards to confirm relative mobility and were subjected to electrophoresis at 150 V until the dye front had run off the gel for 10 min. Gels were stained overnight with Coomassie brilliant blue and destained to remove background staining. The fractions of Coomassie blue-stained protein accounted for by actin (42 kDa), total myosin (200–204 kDa), and filamin (250 kDa) were scanned and analyzed with the TotalLab software package (Biosystematica, Sarnau, Wales, UK). The fractions of stained protein were converted to micrograms and expressed as micrograms per milligram of wet weight (4, 11).
Myosin heavy chain (MHC) in VSM consists of at least three isoforms (4–7, 11, 18). Smooth muscle myosin is expressed as two isoforms with molecular weights of 204 (SM1) and 200 (SM2) kDa, which are splice variants from the same gene differing at the COOH-terminal end and considered to be the predominant MHC isoforms in the contractile VSM phenotype (7). The nonmuscle isoform is derived from another gene and is also 200 kDa (MHC-B); its presence is considered to denote the presence of the synthetic VSM phenotype during development (35, 36). Furthermore, reciprocal changes in MHC-B and SM2 have been described in developing VSM, the former decreasing with age (4, 6). To assess the CA VSM phenotype during ovine development, we measured the relative density of the 200-kDa isoforms of SM2 and MHC-B with immunoblot analyses, running all samples from each age group on the same gel after loading equal quantities of soluble protein, which permits a comparison between age groups. The antibodies and the methods used have been described in detail (4, 11).
Because MHC-B denotes the presence of the synthetic VSM phenotype, it also is considered a biomarker for SMC proliferation (4, 11); however, this has not been confirmed using another marker of SMC proliferation. Therefore, we measured levels of proliferating cell nuclear antigen (PCNA) or cyclin, a 36-kDa protein that associates with the cell nucleus during transcription and cell division (38, 46, 55). Immunoblotting was performed with a polyclonal anti-PCNA antibody (Santa Cruz Biotechnology, Santa Cruz, CA) at a dilution of 1:200 on samples of CA VSM soluble protein that also were probed for MHC-B.
Assessment of CA growth.
Samples of CA obtained at each stage in development were fixed in 4% formaldehyde for 6 h at room temperature and embedded in paraffin as previously described (14). Sections were cut, mounted on slides, and stained with hematoxylin and eosin. To examine vascular remodeling, we measured changes in medial wall thickness in the unstressed state by superimposing a micrometer ruler over the microscopic field at ×10 and measuring the medial width of each artery (Fig. 1). Ten randomly selected areas were measured for each artery studied. To estimate smooth muscle cellular density, cells were counted in 20 randomly selected microscopic fields at ×100 magnification with immersion oil using an indexed 10 × 10 square grid graticule as previously described (23). CA from two randomly selected animals at each age group were used for both measurements. Measurements of cellular density and wall thickness were independently performed by two observers who were blinded, and the differences were statistically assessed.
Intact samples of the arteries used to assess vascular remodeling were also prepared for immunohistochemistry to detect changes in MHC-B and SM2 immunostaining and localization within the vessel wall during development and in the adult. Sections were mounted on slides, deparaffinized, hydrated, incubated with avidin-biotin blocking agent for 30 min, and incubated overnight at room temperature with polyclonal antibodies to MHC-B and SM2 (1:600) described above. After endogenous peroxidases were quenched with 3% H2O2 in H2O for 30 min, immunostaining was detected with standard streptavidin-biotin-horseradish peroxidase and hematoxylin counterstaining.
We used one-way ANOVA for multiple groups to examine differences between the age groups and two-way ANOVA with repeated measures to assess dose-dependent changes in force generation and the effects of age. When significance was observed by ANOVA at P < 0.05, a multiple-comparison procedure was used to isolates groups and to determine differences between groups. Nonsignificant observations had a P > 0.1 in all instances. Data across development obtained for protein contents were also analyzed with linear or polynomial regression analysis. Differences in arterial wall morphology were determined by nonpaired Student's t-test. Data are presented as means ± SE.
Measurements of contractile function.
To compare fetal and neonatal responses with each other and the adult, as well as responses to each agonist, the contraction responses were normalized to the maximum adult response observed with 65 mM KCl and are presented as percentage of adult response (see methods). Endothelium-denuded CA rings demonstrated both dose- and age-dependent contractile responses to KCl (Fig. 2; P < 0.001). Although there were dose-dependent responses in LF, newborn, and adult CA rings (P < 0.001, ANOVA), rings from EF animals were essentially unresponsive at all doses studied (P > 0.4). Furthermore, by pairwise multiple-comparison procedure, the responses in fetal and newborn CA were different from those in adult, and newborn animals differed from both fetal groups (P < 0.001), thus demonstrating progressive increases in age-dependent nonreceptor-mediated contractility. There also was a significant difference in the maximum responses in LF, newborn, and adult CAs at KCl doses ≥50 mM (P < 0.001); the magnitude of the response in LF sheep was ∼50% of the newborn and ∼20% of the adult. When the EC50 was examined, the value was greater in LF (47.0 mM) than in newborn (32.9 mM) and adult (38.5 mM) animals. Representative responses at 65 mM are shown in Fig. 2, differing at P < 0.01 (ANOVA).
There were also dose-dependent increases in the normalized responses to the α1-agonist PE (Fig. 3; P = 0.001, ANOVA) at each age group except the EF group, which again was unresponsive to all doses studied (P > 0.8, ANOVA). The other three age groups demonstrated dose-dependent contraction responses beginning at 10−7 M PE (P < 0.001). There also was an age-dependent change in the dose-response curves between the four groups (P < 0.001), but there was no difference between the LF and newborn at any dose of PE. The maximum responses at 10−5 M had a similar pattern. Notably, maximum responses to PE in the LF, newborn, and adult animals were less than those elicited by KCl. The EC50 was 8.9 × 10−7 M for EF, 8.8 × 10−7 M for newborn, and 3.3 × 10−6 M for adult animals.
Because the patterns of ANG II-receptor subtype expression and binding in CA VSM during ovine development have been described in detail (13, 14), we examined the contractile responses to ANG II. There was an age-dependent increase in the responses to ANG II across development (P < 0.001); however, similar to that shown with KCl and PE, the EF group was unresponsive at all ANG II doses studied (Fig. 4; P > 0.1). Furthermore, the pattern of responses during development differed from that observed with KCl and PE; i.e., the LF and newborn CAs demonstrated modest contractile responses that did not differ and were not dose dependent (P > 0.1). Therefore, the EC50 could not be determined. In contrast, the adult CA, as anticipated, demonstrated a dose-response curve for ANG II (P < 0.001), and, as with KCl and PE, the adult was more responsive at the highest doses of ANG II (P < 0.001) than fetal and newborn animals.
Total protein in CA VSM rose throughout the period of ovine development studied (r = 0.71, P < 0.001, 2nd-order regression). To determine when in development the changes occurred, we performed an ANOVA using the age groups described earlier. Total protein rose ∼50% between EF and LF, increasing from 82.0 ± 3.5 to 125 ± 3.8 μg/mg wet wt (Fig. 5A; P < 0.001, ANOVA), respectively. Levels rose an additional 25% between LF and newborn animals (156 ± 10.6 μg/mg wet wt); however, levels in newborn and adult animals were similar (160 ± 8.5 μg/mg wet wt).
Total protein represents the sum of cellular and noncellular proteins. To measure cellular protein contents, we measured soluble protein. Soluble protein also rose throughout development (Fig. 5B; r = 0.82, P < 0.001). With the use of ANOVA to compare age groups, we observed an ∼90% rise in cellular protein between EF and LF periods, with levels increasing from 35.2 ± 1.9 to 65.5 ± 2.4 μg/mg wet wt, respectively. This was followed by a 27% increase from LF to newborn (83.1 ± 6.3 μg/mg wet wt) and another 35% rise between newborn and adult (106 ± 4.1 μg/mg wet wt; P < 0.001, ANOVA). Thus cellular protein contents rose approximately threefold during development.
To determine whether proteins involved in the contractile apparatus in VSM contributed to the rise in cellular protein (48), we measured the contents of actin and total myosin by SDS-PAGE and determined their patterns of expression. Actin contents increased in parallel with the rise in soluble protein (r = 0.94, P < 0.001), increasing from 4.7 ± 0.3 μg/mg wet wt in EF to 12.7 ± 0.5 μg/mg wet wt in LF (∼3-fold rise, Fig. 6A), to 20.1 ± 2.6 μg/mg wet wt in newborn, and to 30.7 ± 1.2 μg/mg wet wt in adult animals. Thus VSM actin increased greater than sixfold during the period of development studied (P < 0.001, ANOVA), with a large component occurring before birth.
Total myosin contents also rose (r = 0.88, P < 0.001) throughout ovine development. Between EF and adult, myosin increased >4.5-fold, from 1.3 ± 0.4 in EF to 6.1 ± 0.2 μg/mg wet wt in adult animals (Fig. 6B; P < 0.001, ANOVA). However, as with actin, the greatest rise occurred between EF and LF, with values increasing ∼2.7-fold. There was no change in actin-to-myosin ratios during the period of development examined (P = 0.6, ANOVA), demonstrating that there were proportionate increases in these contractile proteins.
To assess changes in noncontractile and contractile proteins related to MHC and SMC function, we used immunoblot analysis to examine nonmuscle MHC-B and SM2, which are both 200 kDa. This also provided insight into the change in SMC phenotype. There was a marked fall in MHC-B expression throughout development, decreasing 38% at term, 66% at 1 wk postnatal, and 95% in adult animals (Fig. 7A; P < 0.001, ANOVA). Albeit small, MHC-B expression persisted in the adult. The fall in MHC-B was associated with a reciprocal rise in SM2 (Fig. 7B), with levels rising 1.5-fold by 1 wk postnatal and 2.7-fold in the adult CA VSM (P < 0.001, ANOVA). In view of the dramatic fall in MHC-B expression during development and persistence in neonatal and adult animals, we examined the intensity and cellular localization of MHC-B immunostaining within the medium in intact CA. At 97 days gestation, there was diffuse immunostaining for MHC-B throughout the medium with an occasional cell noted in the adventitia (Fig. 8B). Consistent with the immunoblot analysis, MHC-B immunostaining decreased substantially at 1 wk postnatal and was essentially absent in adult animals (Fig. 8, C–F). There was no immunostaining of adventitial cells identified after birth or in the adult. We also performed immunohistology with the SM2 antisera. Although immunostaining increased modestly with age, this was not as impressive as the fall in MHC-B, likely because of the relatively low proportion of SM2 vs. total myosin; there were no isolated populations of cells with immunostaining (data not shown).
Although MHC-B is considered a marker of the VSM synthetic phenotype, its expression during development has not been compared with another marker of SMC proliferation. Therefore, we examined PCNA or cyclin expression as another marker of proliferation. Using the samples probed for MHC-B, we found that PCNA expression fell throughout development (Fig. 9; r = 0.97, n = 12, P < 0.001), decreasing 30% by LF, 47% at 1 wk postnatal, and >70% by adulthood. These changes parallel the fall in MHC-B. As with MHC-B, there was persistence of PCNA in the adult samples.
To determine whether proteins in the cytoskeletal compartment of the VSM also change during development and how this relates to changes in functional aspects of CA VSM, we measured filamin, an intermediate filament protein within the cytoskeleton, by SDS-PAGE. Filamin contents in CA VSM did not differ significantly in EF, LF, and newborn, averaging ∼1 μg/mg wet wt (Fig. 10; P < 0.01, ANOVA). However, filamin contents rose more than twofold between the newborn and adult, increasing to 3.5 ± 0.4 μg/mg wet wt (P = 0.001, ANOVA).
CA wall thickness and cellular density.
Vascular growth and remodeling occur during development and may contribute to the age-dependent increases in contractile capacity. Therefore, we measured medial width at five time points in ovine development, examining two CAs at each age in the unstressed state. Each CA was independently measured by two individuals, and the results were blinded (see methods). There were no differences in their measurements (P > 0.1). As anticipated, there was not only an increase in vessel diameter and circumference but also in medial thickness throughout development (Fig. 1). When values were quantified, the medial width increased ∼2.4-fold (P < 0.001, ANOVA) between <90 days and term gestation (77.5 ± 1.8 to 187.5 ± 4.1 μm) and 7.6-fold by adulthood (585.7 ± 11 μm) (Fig. 11A). To assess the contribution of SMC density to this growth, the same evaluators independently measured medial SMC density in the same CA using methods described earlier (23). There were no significant differences in their blinded measurements (P > 0.1). SMC density was unchanged between <90 days and 90–101 days of gestation (Fig. 11B; 12.6 ± 1.5 and 13.1 ± 1.4 VSM cells/field) but decreased 12% in LF (11.6 ± 0.8 cells/field), remained stable at 1 wk postnatal (11.0 ± 1.6 cells/field), and decreased an additional 48% in adult samples (6.4 ± 0.8 cells/field; P < 0.001, ANOVA). These changes in SMC density are illustrated in the insets in Fig. 1, B and F.
The mechanisms that regulate tissue blood flow and blood pressure during development are complex and incompletely understood, especially early in development. They are likely to include a variety of maturational changes in VSM, e.g., alterations in contractile proteins, ion channels, endothelial cell function, vessel wall growth and receptor expression, and non-VSM functions such as reflexes, endocrine changes, and autocrine and paracrine mechanisms. Our group (4, 13, 14, 32) and others (42, 53) have reported tissue-specific differences in the pattern of vascular development in sheep and other species; thus it is not possible to make global conclusions from studies of a single vessel. Neonatal survival has increased 40–50% in the midthird of pregnancy; however, the status of vascular function at this time is unclear. Because cerebral perfusion is essential to fetal and neonatal well being, we examined comprehensive changes in CA VSM maturation between midgestation and 1 wk postnatal and compared the observations to those shown in adult animals. In these studies, neither nonreceptor- nor receptor-mediated agonists elicited contraction responses in EF sheep, and, except for ANG II, increases in VSM contractile capacity paralleled increases in contractile protein contents, decreases in proliferative capacity, and remodeling of the vessel wall. Thus we provide biochemical and functional evidence of a switch from a predominantly synthetic to contractile VSM phenotype in the last third of ovine pregnancy and parallel changes in vessel growth and function. Notably, these changes are associated with a switch from AT2R to AT1R expression (13, 14), suggesting that they may contribute to the modulation of VSM maturation. These observations provide a basis for further assessment of the regulatory mechanisms involved in SMC maturation in an animal model that permits invasive studies during development.
Smooth muscle maturation dictates the development and function of several organs and tissues essential to fetal growth and the adaptive responses of the fetus and newborn. In adults, SMC are highly specialized and designed for contraction. Although plasticity exists, the contractile phenotype predominates (18, 21, 41, 53, 54). The fetal SMC phenotype varies with the organ or vascular bed studied. The developing bladder demonstrates precocious SMC maturation and predominance of the contractile SMC phenotype early in development (6); thus the bladder is capable of emptying in the first third of gestation, thereby establishing and maintaining amniotic fluid volume, which allows for lung growth. The umbilical artery also develops precociously and resembles the adult vasculature (4), i.e., predominance of the contractile VSM phenotype and adult AT1R, demonstrating its importance in regulating fetal nutrient and oxygen uptake and blood pressure (4, 6, 13, 14, 32). In contrast, VSM maturation is delayed in the femoral artery and aorta (4), which is commonly studied in small mammals. Thus SMC of different embryological origins and from different vascular beds may have specific developmental patterns (27, 53, 55). Similar to the study of Hai et al. (25), which began at 106–108 days of gestation, most studies of VSM development and function have been limited to the last third of gestation, i.e., >100 days of gestation, and have been predominantly near term (17, 45).
In the present report, we studied CA development starting at the midthird of gestation. Nonreceptor- or receptor-mediated contractile responses were not evident with any agonist in EF sheep, and the “maturational” pattern for contractility was similar for each agonist; i.e., responses seen at term and postnatal week 1 differed little, although these responses were significantly less than those shown for adult animals. Thus vascular function improves in the late perinatal period but does not achieve adult responsiveness until some time after the first month of age or later (50, 51). Although CA contractile function is delayed compared with the umbilical artery (4), it precedes aortic and femoral artery maturation, which also exhibit delayed AT1R expression, another marker of maturation (6, 14). Unlike KCl and PE, CA responses to ANG II were not dose dependent until adulthood. This is consistent with persistent AT2R expression and binding in ovine CA VSM until >3–4 wk postnatal (13, 14). Although the thoracic aorta of near-term mice have an intact ANG II response, comparisons are difficult because the relative magnitude of these responses was not reported (45). The present data suggest that the renin-angiotensin system (RAS) may not directly contribute to the regulation of fetal and neonatal cerebral blood flow, which is consistent with observations by Kaiser et al. (32) and Velaphi et al. (50, 51). However, contractile maturation in CA VSM occurs early compared with other vascular beds, and the CA likely contributes to the regulation of cerebral blood flow and oxygen delivery. The mechanisms regulating these tissue-specific differences in VSM maturation are unknown. It is notable, however, that the umbilical arteries and CA are high-flow vessels, receiving ∼40% and ∼15% of cardiac output, respectively; thus the magnitude of blood flow may contribute to the regulation of gene expression and vascular modeling (31). It is possible that the pattern of VSM maturation also differs in the distal resistance arteries (43, 55), but this has not been well studied.
The transition from immature to mature SMC is associated with changes in cellular proteins (4–6, 11, 40), including increased contents of actin and myosin, which implies maturation of the contractile apparatus (49). There was a nearly threefold rise in actin and total myosin contents by term, paralleling the increased responsiveness to all three agonists, and a further increase in the first postnatal week that is consistent with studies in other arteries (6) and paralleled the additional increases in KCl sensitivity. Notably, the actin-to-myosin ratios were unchanged, suggesting that similar mechanisms regulate their expression, e.g., increases in blood flow or perfusion pressure. However, this is unclear (53). The values reported were not corrected for the modest increase in water content observed in fetal and neonatal CAs (∼5%) vs. that observed in the adult (43). This difference is unlikely to alter the observed pattern of change because the age-dependent increases in protein greatly exceeded the change in water content. Nonetheless, the increases in contractile proteins are likely to contribute to nonreceptor- and receptor-mediated responses at term and after birth (2, 3).
Filamin, an important cytoskeletal protein, demonstrated a different expression pattern. Values were unchanged until after birth and beyond the first week postnatal. It is unclear what this signifies; interestingly, it parallels the switch from AT2R to AT1R in CA VSM (13, 14). It has been suggested that these receptors contribute to the regulation of SMC protein expression during development, with AT2R acting as a positive or negative regulator (28, 45). The inverse relationship between AT2R expression (13, 14) and increases in cellular protein suggests that it may be a negative regulator. This is supported by recent studies in growth-restricted fetal sheep that have a precocious switch in subtype (47).
The switch from the proliferative to the adult contractile VSM phenotype is an important maturational event (42). The timing of this event is unclear and may be species specific; e.g., in rodents it occurs soon after birth (53), whereas it begins before birth in sheep (4, 6, 11). To address this, we examined MHC isoform expression using immunoblots and immunohistochemistry (1, 11). The contractile SMC phenotype expresses SM1, a 204-kDa protein that is found in early development and increases over time, and SM2, a 200-kDa splice variant whose expression occurs late in development and increases near term (11, 36, 39). MHC-B is a nonmuscle myosin, also 200 kDa, and is considered a marker of the synthetic SMC phenotype (35). Thus decreases in MHC-B and reciprocal increases in SM2 confer the contractile phenotype (4, 6, 11). MHC-B was predominate in EF sheep, falling >65% in the last third of gestation and after birth; however, SM2 rose. This transition gradually occurred throughout the last third of ovine pregnancy, consistent with earlier observations (11), and the immunohistology demonstrated diffuse immunostaining of both proteins within the media that also changed gradually. Interestingly, it is unknown whether apoptosis contributes to this transition. The switch in 200-kDa proteins occurred as VSM function increased and proliferation decreased, providing additional evidence of emergence of the adult VSM phenotype. The contribution of the emerging SM2 to the enhanced contractility is unclear. Notably, MHC-B persisted in the adult CA, suggesting the presence of SMC with an inherent capacity for replication in the event of vessel damage (1, 40, 53). We also observed a parallel fall in PCNA expression, another marker of cellular proliferation (38, 46, 55), providing additional evidence of a gradual change in the SMC phenotype and that MHC-B is a reliable marker of the synthetic phenotype. Notably, MHC-B immunostaining was absent in adventitial cells after birth and in adult animals, suggesting smooth muscle progenitor cells may reside in the CA media (53).
Another component of vascular development is growth and remodeling. Vascular growth allows for body and organ growth. We studied sheep weighing 800–1,000 g at midgestation, ∼3,500 g at term, 5–6 kg at 1 wk postnatal, and finally 60 kg at adulthood, weights that are strikingly similar to humans at comparable ages. This growth was associated with increases in total and soluble protein contents, suggesting increases in noncellular proteins, e.g., collagen and elastin (8), and SMC replication. Remodeling was evidenced by increased vessel diameter, circumference, medial thickness, and lumen size, but decreased SMC density, consistent with observations by Pearce and colleagues in older fetal and newborn sheep (19, 43). It is important to note that medial wall thickness changes in vivo as blood pressure rises, which normally increases during development and may serve as a means of regulating vascular remodeling and increases in tissue blood flow. The decrease in SMC density is associated with modest decreases in cellular water and the DNA-to-unit wet weight ratio, interpreted as evidence of SMC hypertrophy (19). Because the decrease in SMC density was associated with increases in cellular protein, SMC hypertrophy is likely occurring (8). However, medial width increases until adulthood (40); thus cellular proliferation must also continue well beyond the perinatal period, explaining the persistent expression of MHC-B and PCNA. Thus CA growth includes VSM proliferation, especially in mid- and late gestation, and hypertrophy, illustrated by proportional increases in extracellular matrix and a stable soluble-to-total protein ratio. As noted earlier, changes in ATR subtypes have been implicated in vessel remodeling and SMC growth. For example, a switch from AT2R to AT1R is associated with decreased VSM proliferation and increases in cellular hypertrophy (28, 45). Notably, the decreases in CA SMC proliferation parallel decreases in AT2R expression (14), whereas the increase in AT1R occurs in the presence of SMC hypertrophy. The mechanisms responsible for these associations can be examined in this model.
We have examined the maturational changes in ovine CA VSM at midgestation, which is comparable to the 24- to 26-wk gestation period in humans. Contractile responses were absent in EF animals, implying that CA may not contribute to the regulation of brain blood flow at this time. If the peripheral vasculature is similarly developed, cardiac output and umbilical blood flow, which have precocious maturation and account for 40–50% of cardiac output (4, 48, 52), may regulate perfusion pressure and tissue blood flows in utero (16, 32). Unlike rodents, changes in ovine VSM phenotype and contractile proteins begin early in utero and parallel improved vascular function at term. In humans, this is associated with age-dependent increases in blood pressure, implying prenatal changes resembling those in the developing sheep (20, 56). Notably, the changes in MHC-B and SM2 correlate with a fall in AT2R and rise in AT1R expression (14), supporting evidence that RAS contributes to vascular development (47, 48, 53). Finally, our observations support reports of SMC within the medial wall of the adult CA with synthetic potential (1, 37, 53). These observations, however, do not clarify the mechanisms that govern SMC maturation and could include alterations in tissue flow, local autocrine and paracrine factors, and RAS; rather, they demonstrate the need for further studies of blood pressure regulation in early development.
These studies were funded in part by National Institute of Child Health and Human Development Grant HD-08783 and the George L. MacGregor Professorship in Pediatrics.
C. Hutanu was a postdoctoral trainee in Perinatal Medicine. B. E. Cox played an essential role in initiating these studies, which were completed after his untimely death.
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.
- Copyright © 2007 the American Physiological Society