Little is known about the vascular and metabolic adaptations that take place in the fetal heart to maintain cardiac function in response to increased load. Chronic fetal anemia has previously been shown to result in increased ventricular mass, increased myocardial vascularization, and increased myocardial expression of hypoxia-inducible factor-1 (HIF-1) and vascular endothelial growth factor (VEGF). We therefore sought to determine whether chronic fetal anemia induces expression of HIF-1-regulated angiogenic factors and glycolytic enzymes in the fetal myocardium. Anemia was produced in chronically instrumented fetal sheep by daily isovolemic hemorrhage (80–100 ml) for either 3 (n = 4) or 7 days (n = 11) beginning at 134 days of gestation (term 145 days). Catheterized, nonbled twins served as controls. Isovolemic hemorrhage over 7 days resulted in decreased fetal hematocrit (37 ± 1 to 20 ± 1%) and arterial oxygen content (6.5 ± 0.4 to 2.8 ± 0.2 ml O2/dl). Myocardial blood flow and vascularization were significantly increased after 7 days of anemia. Myocardial HIF-1 protein expression and VEGF (left ventricular), VEGF receptor-1 (right ventricular), and VEGF receptor-2 (right ventricular, left ventricular) mRNA levels were elevated (P < 0.05) in 7-day anemic compared with control animals. Myocardial expressions of the glycolytic enzymes aldolase, lactate dehydrogenase A, phosphofructokinase (liver), and phosphoglycerol kinase were also significantly elevated after 7 days of anemia. Despite the absence of a significant increase in myocardial HIF-1α protein in 3-day anemic fetuses, expressions of VEGF, VEGF receptor-1, and the glycolytic enzymes were greater in 3-day compared with 7-day anemic animals. These data suggest that HIF-1 likely participates in the fetal myocardial response to anemia by coordinating an increase in gene expressions that promote capillary growth and anaerobic metabolism. However, factors other than HIF-1 also appear important in the regulation of these genes. We speculate that the return of mRNA levels of angiogenic and glycolytic enzymes toward control levels in the 7-day anemic fetus is explained by a significantly increased resting myocardial blood flow, resulting from coronary vascular growth and increased coronary conductance, and a return to a state of adequate oxygen and nutrient delivery, obviating the need for enhanced transcription of genes encoding angiogenic and glycolytic enzymes.
fetal cardiac stroke volume increases in response to chronic anemia, resulting in increased cardiac work in the face of decreased oxygen carrying capacity of the blood (10). Although alterations in myocardial blood flow, vascularization, substrate delivery, and utilization must occur to support the compensatory increase in cardiac output and work, the mechanisms regulating these adaptive responses are poorly understood. In the ovine fetus, decreased arterial oxygen content has profound effects on myocardial blood flow (9). In response to chronic (5–8 days) hypoxemia and chronic anemia, fetal myocardial blood flow increases severalfold (9, 27). This finding is not unexpected because the heart operates aerobically and coronary flow must increase to provide adequate myocardial oxygen delivery. Capillary proliferation and recruitment of the coronary microcirculation likely contribute to this response (20, 24). Although vascular endothelial growth factor (VEGF) has been suggested to stimulate increased myocardial vascularization in the chronically anemic ovine fetus, other angiogenic agents, including shear-stress-activated factors such as adenosine, are likely involved (2, 24, 42).
Metabolic adaptations must also occur in the myocardium of anemic fetuses to maintain adequate ATP production in the face of increased load. The fetal heart primarily oxidizes glucose and lactate to generate ATP (12). Although significant increases in myocardial glucose or lactate uptake have not been identified in the anemic fetus, studies from our laboratory (26) have demonstrated that myocardial glucose transporter GLUT4 expression is increased and expression of GLUT1 is decreased in chronically anemic fetal sheep. Regulatory mechanisms involving important glycolytic and mitochondrial enzymes perhaps orchestrate the rapid adaptive changes that must occur in response to the increased metabolic requirements of the fetal heart during chronic anemia.
Previous studies in chronically anemic fetal sheep have demonstrated increased myocardial expression of hypoxia-inducible factor 1 (HIF-1), an important transcription factor for many vascular and glycolytic genes (24). We hypothesized that activation of these HIF-1-regulated genes would occur in the fetal heart in response to chronic anemia. Specifically, we examined changes in myocardial blood flow and vascularization along with temporal changes in the expression of select myocardial angiogenic and metabolic genes and proteins that may be involved in fetal cardiac adaptations to chronic anemia.
All procedures were performed within the regulations of the Animal Welfare Act and the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the University of Iowa Animal Care and Use Committee.
Studies were performed in fetuses of pregnant sheep of Dorset and Suffolk mixed breeding, obtained from a local source. The gestational ages were based on the induced ovulation technique as described elsewhere (19). Surgical preparation of the fetus was performed at 128 days of gestation, term being 145 days. Anesthesia and the surgical techniques have been described previously (30). Briefly, after pregnant ewes were fasted for 24 h, general anesthesia was induced with an intravenous infusion of thiopental sodium (12 mg/kg) (Abbott Laboratories, Chicago, IL) and maintained with a mixture of halothane (1%), oxygen (33%), and nitrous oxide (66%). Under sterile conditions, a maternal left flank incision was performed to expose the uterus. The uterus was opened, and polyethylene catheters were inserted into the femoral arteries (PE-160, ID = 1.14 mm, OD = 1.57 mm; Intramedic, Franklin Lakes, NJ) and into the veins (PE-90, ID = 0.86 mm, OD = 1.27 mm) bilaterally. In fetuses undergoing measurement of myocardial blood flow during the 7-day anemia protocol, the fetal thorax was opened and catheters were secured in the left atrial appendage for adenosine infusion and microsphere injection (n = 5). A catheter (PE-160) was also placed in the carotid artery, and the tip was advanced to the proximal ascending aorta. This catheter was used to obtain reference blood samples during microsphere injection. After closure of the chest, the fetus was returned to the uterus. Uterine and maternal incisions were closed in separate layers. All catheters were exteriorized through a subcutaneous tunnel and placed in a cloth pouch on the ewe's flank. At the end of surgery, ampicillin was directly infused into the amniotic cavity (2 g) and administered intramuscularly (2 g) to the ewe. Antibiotic administration was continued for 3 days postoperatively. Butorphenol (0.1 mg/kg iv; Torbugesic, Fort Dodge Animal Health, Fort Dodge, IA) was administered for 24 h postoperatively for analgesia.
After a 72-h recovery period, daily isovolemic hemorrhage was performed by removal of 80–100 ml of blood and simultaneous replacement with warmed 0.9% NaCl for a period of 7 days. During each experiment, fetal mean arterial blood pressure (MABP) and amniotic pressure were recorded daily for 30 min using Statham P23 Db pressure transducers (Spectramed, Critical Care Division, Oxnard, CA) and a Grass 7–24P chart recorder (Grass Instruments, Quincy, MA) and stored online to a personal computer. Fetal MABP was corrected relative to concomitant amniotic pressure. Heart rate was monitored with a cardiotachometer triggered from the arterial pressure pulse wave. Daily measurements of heart rate, blood pressure, arterial blood gases, hematocrit, hemoglobin, arterial oxygen content, and plasma erythropoietin were made before phlebotomy. Serum samples for determination of serum erythropoietin levels were obtained daily and stored at −80°C. Myocardial blood flow was determined in five fetuses using fluorescent microspheres as performed previously and described in detail below (8). The study concluded 24 h after the seventh phlebotomy (day 8 of study) (n = 11). At this time, the ewe was again anesthetized as described above and the fetuses were removed (138 days of gestation). Fetal hearts were removed for determination of total weight and left ventricular (LV) and right ventricular (RV) free wall and septum weights. Tissues from the RV and LV free walls obtained approximately midway between the apex and atrioventricular groove were isolated, and duplicate samples of endocardium and epicardium were sectioned and stored in preweighed glass vials for subsequent blood-flow analysis. Remaining tissue was flash frozen in liquid nitrogen and stored at −80°C (n = 8). In a subgroup of fetuses, the hearts were perfusion fixed in a retrograde manner for later determination of capillary morphometry (n = 3) as described below. Animals of either sex were used. Sham controls (n = 11) were similarly instrumented, although no hemorrhage was performed. Control fetuses underwent similar tissue harvesting (n = 8) and cardiac perfusion (n = 3).
A separate group of fetuses (n = 4 for both experimental and control group) underwent similar surgical preparations and isovolemic hemorrhage as described above, except the experiment was terminated 24 h after the third phlebotomy. No microsphere injections or cardiac perfusions for vascular morphometry were performed in this group of animals.
Arterial blood for pH, Pco2 and Po2, hemoglobin, and oxygen content were collected anaerobically in a heparinized syringe, and measurements were immediately determined at 39.5°C, using a 1302 BGM pH/blood-gas analyzer and an IL 682 cooximeter system (Instrumentation Laboratory, Lexington, MA). Erythropoietin levels were analyzed in duplicate with a specific double-antibody radioimmunoassay that had a sensitivity of 1.2 mU/ml and an intra-assay coefficient of variation of 4% (7).
Microsphere recovery and myocardial blood flow measurement.
Myocardial blood flow was determined with fluorescent microspheres, as described previously (8). Briefly, for each blood flow determination, ∼1.2 × 106 15 μm fluorescent-labeled polystyrene 10 microspheres (Triton Technology, San Diego, CA), suspended in 1 ml of 0.9% NaCl with 0.02% Tween 20 and 0.02% thimersol, were injected into the left atrium and flushed with 2 ml of 0.9% NaCl over a 30-s period. Beginning 25 s before the microsphere injection, reference blood samples were withdrawn from the carotid catheter (tip in the ascending aorta) into heparinized glass syringes for 2.5 min at a rate of 2 ml/min. After a 15-min recovery period, a continuous left atrial infusion of adenosine (150 μg·kg−1·min−1) was started to achieve maximal coronary vasodilation (8). After a 10-min continuous infusion of adenosine, myocardial blood flow measurements (using different fluorescent-colored microspheres) were repeated without interruption of the adenosine administration. This procedure was performed on day 1 (before hemorrhage) and on day 8 of the study. Samples of blood and known weights of myocardium were digested and filtered to recover microspheres with the use of previously published methodologies (Manual for Using Fluorescent Microspheres to Measure Regional Organ Perfusion, Fluorescent Microsphere Resource Center, University of Washington; URL is http://fmrc.pulmcc.washington.edu/frmc/fmrc.html). After digestion, the microspheres were filtered from the solutions with 10-μm filter membranes (Triton Technology). Microspheres were dissolved with 1,000 μl of Cellosolve acetate (Fisher Scientific, Fair Lawn, NJ), 200 μl of solution were transferred to a wellplate, and fluorescence was determined. Fluorescent measurements of experimental and “standard curve” samples were determined with a luminescence spectrophotometer (LS-50B, Perkin-Elmer, Wellesley, MA), using the appropriate excitation/emission wavelengths and slit widths for the given colored sphere. Tissue blood flow (using ascending aorta reference samples) was calculated using the formula (1) where Q̇sample and Q̇ref are the blood flows (in ml/min) in the specific sample and reference sample, respectively, and Fsample and Fref are the fluorescent intensities in the specific sample and reference sample, respectively. Our group has previously found that, in our sheep population, the normal rate of fetal growth between 120 and 138 days of gestation is 2.7%/day (28) and that heart weight-to-body weight ratio is 6.7 ± 0.2 g/kg (31). To correct for heart growth over the study period, the fetal weight on day 1 of study was estimated using the equation W = Woe(0.027 days), where W is the weight at autopsy, Wo is the weight on the first day of study, and “days” is the number of days of the study. In anemic fetuses, the rate of growth is smaller and estimated to be 0.8%/day (9). For these animals, fetal weight was estimated with the equation W = Woe(0.008 days). LV and RV free wall weights were estimated to be 2.24 and 1.99 g/kg estimated fetal weight, respectively (8).
Northern blot analyses.
Isolation of total RNA, preparation of ovine-specific cDNA probes, Northern blot hydridization, and measurements of steady-state mRNA levels were made as described previously (25). Total RNA was isolated from the RV or LV free wall of the anemic or sham control fetuses. Ovine-specific cDNA probes for HIF-1α, VEGF, basic FGF, transforming growth factor (TGF)-β, VEGF receptor (VEGFR)-2, VEGFR-1, aldolase, α-enolase, phosphofructokinase (liver) isoform (PFK-L) and phosphoglycerolkinase (PGK) were prepared to enhance signal-to-noise of the d-32UTP signal of the Northern blots as previously described. Hybridization signals were quantitated with a phosphorimager (Molecular Dynamics). After study with the specific probes, blots were stripped and rehybridized with 32P-labeled probe to the 18S or 28S subunit of ribosomal RNA to normalize for variable RNA loading.
Immunoblots were prepared as described previously (25) using protein isolated from the LV or RV free walls. Monoclonal antibodies to HIF-1α and HIF-2α used for immunoblotting (both raised in rabbits against specific murine proteins; Novus Biologicals, Littleton, CO) have previously been shown by our group (25, 39) to have good specificity in ovine tissues. The secondary antibody used to detect the primary antibodies was conjugated with horseradish peroxidase. We detected the secondary antibody using the Pierce SuperSignal kit, which was exposed with Kodak XAR film at room temperature. Films were digitized, and signals were quantitated with National Institutes of Health (NIH) Image (Wayne Rasband, NIH). Serial protein dilutions were tested with each antibody to ensure quantitated signals were in the linear range for added protein.
Lactate dehydrogenase activity.
Total lactate dehydrogenase (LDH) activity was determined spectrophotometrically; we used a commercial laboratory (ARUP Laboratories, Salt Lake City, UT; William Roberts, M.D., Ph.D., Medical Director) for LDH isoenzyme electrophoresis analysis.
Perfusion fixation and capillary morphometric analysis.
The techniques for fixation and capillary morphometry have previously been described (35, 37). To briefly summarize, hearts were arrested in diastole with a bolus injection of KCl (40 meq), removed, and fixed by retrograde aortic perfusion in vitro at the prevailing MABP for each animal with Locke's solution (500 ml, 37°C) followed by a solution of 1.5% glutaraldehyde, 0.2% paraformaldehyde, 0.1 M cacodylate, and 0.03 M CaCl2, pH 7.4, at room temperature. After they were fixed overnight, transmural blocks were removed from the LV and RV free walls. Samples were embedded in Spurr's plastic and sectioned at a thickness of 1 μm. A Quantimet 720 image analyzer was used for capillary morphometric analysis; capillary length, volume densities, diameter, and surface area were determined as previously described (35, 37). Surface density (Sv) was calculated from the total perimeters (P) of capillary profiles, whereas volume density (Vv) was calculated from profile lumen cross-sectional area (A) using these formulas: (2) where AT is test or planar area of fields. Capillary length density is based on the following calculation: Lv (mm/mm3) = (a/b)NA, where a and b are the long and short axes, respectively, and NA is the number of vessel profiles in the field (numerical density). Vessels with no smooth muscle and diameters <10 μm were noted as capillaries.
All data are expressed as means ± SE. Comparisons of mRNA and protein expression between the chronic anemic and control groups within specific age groups were made by unpaired t-test. Comparisons of mRNA levels between age groups were made by ANOVA, while comparisons of physiological data within an age group were made by repeated-measures ANOVA, factoring for day of study and treatment group. When the ANOVA indicated significant differences by the F statistic, comparison among means was performed by the Duncan multiple-comparison procedure. Significant differences were identified at the P < 0.05 level.
Seven-Day Anemia Studies
Fetal hemodynamics, arterial blood-gas values, and weights.
Baseline MABPs obtained before the first isovolemic hemorrhage (day 1) were similar between the groups (Table 1). MABPs were also not different in the control and 7-day anemic fetuses at the completion of the study (day 8), 24 h after the last hemorrhage. Heart rate was greater in 7-day anemic fetuses at baseline and at the end of the studies compared with controls (Table 1). However, there was no significant change in heart rate over the course of the study in either group. Arterial blood pH and Pco2 were similar among all study groups throughout the study. There was a slight but significant decrease in arterial Po2 in 7-day anemic compared with control fetuses on day 8 of the study.
Daily fetal hematocrit and arterial oxygen content are depicted in Fig. 1. Baseline hematocrit was similar between the groups. The 7-day anemic fetuses had lower hematocrits compared with controls on days 2 through 8 of the study. Minimal change in hematocrit values occurred in the anemic fetuses on days 5 through 8 of the study. As expected, oxygen content also decreased in anemic fetuses after day 1 of study and followed a pattern similar to hematocrit values.
Total body weight was similar in anemic and control fetuses at the completion of the study (Table 2). The LV free wall-to-body weight ratio was significantly increased in the 7-day anemic fetuses relative to control (Table 2). There was a tendency for RV free wall-to-body weight ratio to also be increased in this group, although significance was not achieved (P = 0.09). The LV/RV weight ratio was also similar in both groups of fetuses.
Myocardial blood flow and coronary reserve.
Determinations of myocardial blood flow were made in four fetuses subjected to 7 days of anemia (Fig. 2). As expected, the development of anemia resulted in significantly increased LV and RV endocardial blood flows relative to baseline values (before anemia). Infusion of adenosine significantly increased myocardial blood flow on both day 1, before the development of anemia, and after the development of anemia (day 8), suggesting the presence of coronary artery reserve in both states. The increases in endocardial blood flow in response to adenosine and anemia were similar in both ventricles. Measurements of epicardial blood flow (data not shown) displayed a pattern similar to that for endocardial flow.
Myocardial capillary morphometry.
To examine the contribution of coronary vascular remodeling to the increase in myocardial blood flow, capillary morphometry was performed on control and 7-day anemic fetal hearts that were perfused at a prevailing mean pressure of 45 mmHg (n = 3 for each group). Capillary volume density was significantly increased in both the RV (+24%) and LV (+40%) of anemic animals compared with controls (Fig. 3). Average capillary diameter was significantly increased in the right ventricle but not left ventricle of anemic fetuses. RV and LV capillary length density and average surface area were similar in 7-day anemic and control animals.
Effects of fetal anemia on myocardial gene expression.
Because previous studies in this animal model demonstrated significant increases in myocardial HIF-1 expression in response to chronic anemia, we chose to examine the expression of several vasculogenic and metabolic genes with known hypoxia response elements. Consistent with the findings of Martin et al. (24), HIF-1α protein levels were increased in the LV and RV of 7-day anemic fetuses compared with control values (Fig. 4). In contrast, myocardial levels of HIF-1α mRNA (Fig. 4) and HIF-2α protein (Fig. 5) were similar in anemic and control fetuses. By Northern blot analysis, we found that steady-state mRNA levels of VEGF were significantly increased in the LV but not RV of 7-day anemic fetuses (LV data; Fig. 6). mRNA levels of the VEGF receptors were also altered by chronic anemia. Levels of VEGFR-1, which is known to be regulated by hypoxia, were significantly (P < 0.05) increased in the RV (23 ± 5%) but not LV (data not shown), whereas mRNA levels of VEGFR-2 increased to similar extents in both the LV and RV of 7-day anemic fetuses (LV data; Fig. 6).
A number of other angiogenic factors have been shown to be expressed in the developing heart, including basic FGF and TGF-β (1, 38). However, no significant changes in basic FGF or TGF-β mRNA expression were detected in either ventricle from anemic compared with control fetuses (n = 8 for each group, data not shown). Because endothelial nitric oxide synthase (eNOS) plays an important role in myocardial angiogenesis and capillary development in part by altering VEGF expression (16), changes in the myocardial expression of eNOS in response to chronic anemia were quantitated. In both ventricles, myocardial eNOS protein expression did not appear to be altered by the development of chronic anemia (n = 8 for each group, data not shown).
Transcription of several genes encoding glycolytic enzymes have been shown to be regulated by HIF-1 through an upstream binding site (32, 33). Consistent with an increase in HIF-1α protein levels, steady-state mRNA levels of aldolase A, PFK-L, and PGK were significantly increased in both the RV and LV of anemic fetuses (LV data, Fig. 7). Expression of LDH A mRNA was also increased in the hearts of anemic fetuses (LV data; Fig. 7), although no differences in total LDH or isoenzyme specific activity were detected (data not shown). The expression of α-enolase mRNA also remained unchanged in ventricles of both anemic and control fetuses.
Because the fetal cardiac mitochondrial function is likely altered in response to increased work, resulting from fetal anemia, we also postulated that expression of nuclear regulatory factor 1 (NRF-1), a member of an important family of transcription factors that appear to regulate gene expression of several nuclear encoded mitochondrial proteins, as well as cytochrome oxidase (COX) subunit Va and carnitine palmitoyltransferase I (CPT I) would be increased in anemic fetuses. However, by Northern blot analysis, steady-state mRNA levels of NRF-1, CPT-I, and COX subunit Va showed no changes in response to anemia (data not shown).
Three-Day Anemia Studies
Because significant decreases in arterial oxygen content did not occur beyond the fourth day of the study despite continued isovolemic hemorrhage, we speculated that, by day 8 of study, expression of many HIF-1-regulated genes might have returned toward the baseline level of expression. Consistent with this hypothesis was the finding that serum erythropoietin levels, which were obtained as part of a concurrent study in these same animals, were maximum on day 4 of the study and returned toward baseline on days 7 and 8 of the study (serum erythropoietin levels: 34 ± 16, 348 ± 145, and 176 ± 70 mU/ml on days 1, 4, and 8 of study, respectively; n = 10). To better investigate the time course of gene expression, we studied a second group of anemic fetuses in which the animals were euthanized on day 4 of the study, 24 h after the third hemorrhage.
Fetal heart rate values were greater on day 4 of the study in anemic compared with control animals, although values did not change over time within either group (Table 1). MABP was, however, lower in anemic fetuses at the end of the 3-day study period relative to baseline values (day 1) and control animals. Similar to 7-day anemic fetuses, there was a slight but significant decrease in arterial Po2 in 3-day animals at the end of the study relative to baseline values. Importantly, the 3-day anemic fetuses displayed similar decreases in hematocrit, hemoglobin, and oxygen content as did the 7-day anemic fetuses over the initial period of study (Fig. 1). No significant differences were detected in total heart or ventricular weights relative to body weight in 3-day anemic compared with control animals (Table 2). Despite the fact that myocardial HIF-1α and HIF-2α protein levels were unchanged with anemia, expressions of RV steady-state mRNA levels of VEGF, VEGFR-1, VEGFR-2, aldolase, enolase, PFK, LDH-A, and PGK were all significantly increased in anemic fetuses relative to controls (P < 0.05). Furthermore, the increase in expression of these genes was greater after 3 days of anemia relative to the 7-day anemic animals (RV data, Fig. 8).
The observations of this study complement previous findings in the chronically anemic fetal sheep, demonstrating that chronic anemia results in increased myocardial mass, blood flow, vascularization, and expression of the HIF-1 and VEGF genes (24). Additionally, we observed that the induction of a number of vascular and metabolic genes, for which HIF-1 is an important transcription factor, occurs in response to chronic anemia. Furthermore, steady-state mRNA levels of these myocardial vascular, glycolytic, and mitochondrial oxidative genes increase with the initial phase of anemia, in the absence of a detectable increase in HIF-1 or HIF-2 levels. Expression of the angiogenic and glycolytic genes returns toward baseline after 7 days of anemia despite continued low arterial oxygen content and increased HIF-1 protein levels. This return toward baseline expression likely occurs as the heart adapts to the increased workload resulting from anemia by increasing myocardial blood flow and remodeling myocardial vasculature to maintain oxygen and nutrient delivery despite decreased hematocrit and oxygen contents.
Previous studies found that HIF-1α protein and VEGF mRNA expression were increased three- to fourfold in the hearts of chronically anemic animals, whereas no more than twofold changes in expression were detected in our animals (24). It is possible that variations in the severity and pattern of decreased arterial oxygen content may have contributed to these differences in myocardial HIF-1α protein and VEGF mRNA expression. Animals in our study did not achieve fetal hematocrit or arterial oxygen content values as low as those reported by others (9, 24). In addition, animals in our study reached a nadir of arterial oxygen content by day 5 of the study, remaining at a relative “steady state” through day 8, whereas fetal oxygen content in other studies progressively decreased throughout the period of anemia. Martin et al. (24) surmised that increased HIF-1 expression indicated the presence of myocardial hypoxia from decreased oxygen delivery and increased oxygen demand, resulting from increased cardiac work and tissue mass. However, in a previous study (10) of chronically anemic fetuses, although LV oxygen consumption increased compared with controls, no changes in oxygen extraction occurred, as would be expected if consumption increased disproportionately to delivery. Furthermore, myocardial glucose and lactate consumption remained unchanged and there was no net lactate production, suggesting aerobic metabolism was preserved. Our group (26) similarly found in a subgroup of anemic fetuses described in the present study that, although serum lactate increased, suggesting systemic hypoxemia and a shift toward anaerobic metabolism, there was no net myocardial lactate production. Thus the role of in vivo tissue hypoxia in regulating myocardial expression of HIF-1 protein and, in turn, VEGF mRNA is not clear.
Despite the less severe fetal anemia and hypoxemia in this study compared with previous studies, the myocardial blood flow and capillary morphometric findings are similar to those previously described (10). Specifically, we observed a three- to fourfold increase in resting myocardial blood flow with chronic anemia. Myocardial blood flow in anemic fetuses could also be increased with adenosine, suggesting coronary flow reserve is maintained even in the presence of severe anemia. Findings from Davis et al. (10) demonstrated that the increase in fetal myocardial blood flow, likely required because of increased ventricular oxygen consumption, resulted from decreased blood viscosity and increased myocardial vascularity. Our data demonstrating increased capillary volume density in both the LV and RV of anemic animals compared with controls suggest that anemia did not stimulate significant neovascularization but rather remodeling of the capillary tree. Because the hearts used for this analysis were fixed by vascular perfusion at the MABP for each respective animal, the observed differences in capillary morphometry between the control and anemic groups represent anatomic changes present under “baseline” conditions.
Studies in a number of animals models demonstrate that VEGF plays a vital role in coronary vasculogenesis and angiogenesis (18, 36). In vitro and in vivo studies have demonstrated that, in addition to mechanical load, tissue hypoxia is a strong inducer of VEGF mRNA expression (6, 11). Increased levels of VEGF mRNA in response to hypoxia may be mediated by activation of HIF-1, which in turn increases VEGF mRNA transcription and stability (15, 22). There is also evidence to suggest that expression of myocardial VEGF may be predominantly regulated by HIF-2 relative to HIF-1 (41). Hypoxia also appears to upregulate expression of VEGFR-1 but not VEGFR-2, although results have been conflicting (5, 23).
Studies in fetal and newborn animals suggest that increased cardiac mass resulting from a variety of factors, including pulmonary artery and/or ascending aorta banding or chronic isovolemic anemia, is accompanied by increased myocardial vascularization (13, 14, 24). In young pigs, pulmonary artery banding results in rapid induction (2 h) of genes important for vascular growth, including VEGF and VEGFR-2 mRNA (3). In the present study, we examined the effects of chronic anemia on expression of several previously identified regulators of vasculogenesis. Relative levels of VEGF, VEGFR-2, and VEGFR-1 were increased in the hearts of anemic fetuses, particularly early in the development of anemia (3-day time point) when vascular growth is likely maximal.
Glycolytic Enzyme Expression in the Fetal Heart
In contrast to the adult heart, in which fatty acids contribute the majority of the fuel source and the contribution from glycolysis to energy production is negligible, the fetal heart oxidizes primarily glucose and lactate via glycolysis for energy production (12). LDH, which normally catalyzes the last step of glycolysis, allows the fetal myocardium to obtain additional energy from the oxidation of lactate. LDH is a tetrameric protein, derived from two different 35-kDa polypeptide chains, A and B, each encoded by separate genes (29) and resulting in five isoenzymes. The LDH-A gene exhibits an hypoxia response element, and transcription of the gene is upregulated in low oxygen states by HIF-1, whereas the LDH-B gene is not regulated by oxygen tension (32). We previously demonstrated that myocardial expression of LDH-A mRNA is developmentally regulated in sheep and parallels the postnatal decline in HIF-1 protein levels (25). In the present study, we found that myocardial LDH-A mRNA levels were greater in 7-day anemic compared with control fetuses, consistent with increased expression of HIF-1α protein in these tissues. Despite the increase in LDH A mRNA levels, total LDH activity was not increased in the myocardium of anemic fetuses relative to controls. Furthermore, no changes in LDH isoenzyme composition were detected in the hearts of anemic fetuses. The lack of correlation between LDH mRNA abundance and LDH-specific activity has been described in a number of tissues and likely relates to influences on LDH-A mRNA stability (29).
Not surprisingly, the steady-state mRNA levels of several other enzymes in the glycolytic reaction sequence that contain cis-acting hypoxia response elements that bind HIF-1, including aldolase A, and PFK-L were increased in the hearts from anemic fetuses relative to controls. However, levels of α-enolase, which by genome analysis and in vitro assays also contains a hypoxia response element in the promoter region (32), were unchanged in response to 3 or 7 days of anemia. In sheep, myocardial α-enolase mRNA levels are high during fetal life and decrease postnatally, whereas levels of β-enolase increase after birth (25). A significant postnatal decline in α-enolase subunit at both the mRNA and protein level has been found in rats (21). In adult rats with cardiac hypertrophy resulting from aortic stenosis, β-enolase gene expression is decreased, whereas α-enolase remains unaltered (21). We were unable to detect an increase in α-enolase mRNA levels in the myocardium of anemic fetuses despite two concurrent conditions, namely, oxygen deprivation and increased cardiac workload, which would be expected to increase α-enolase gene expression. The expression of β-enolase was not examined because levels are already low in the fetal sheep heart (25).
An important observation of this study is the significant upregulation of several myocardial glycolytic enzyme mRNAs after 3 days of anemia, despite the lack of an increase in HIF-1α protein levels, an important transcription factor for these genes. Expression of these genes also returned toward baseline after 7 days of anemia, despite the relative hypoxemic in utero environment. These findings indicate that the metabolic response of the myocardium to chronic anemia is to stimulate transcription of enzymes that are important in providing energy to the myocardium. The stimuli for this response appear to be independent of myocardial HIF-1 expression, suggesting other factors are involved. These factors likely relate to the increased myocardial workload (increased stroke volume and cardiac output) associated with fetal chronic anemia and limited nutrient delivery. The return of mRNA levels of these glycolytic enzymes toward control levels in the 7-day anemic fetus may be explained by significantly increased resting myocardial blood flow. After coronary vascular growth occurs and coronary conductance increases, myocardial blood flow is increased and a state of adequate oxygen and nutrient delivery is restored, obviating the need for enhanced transcription of genes encoding glycolytic enzymes.
In the adult hypertrophied heart, the attendant increases in mitochondrial oxidative capacity, due to increased mitochondrial number and glycolytic capacity, has been extensively investigated (17, 34). Increasing mitochondrial content poses a complex regulatory problem because both mitochondrial and nuclear genes need to be activated concurrently to produce a complete mitochondrion. NRF-1 appears to play a key role in coordinating the mitochondrial and nuclear gene activities needed to increase mitochondrial oxidative capacity in the hypertrophied heart (40, 43). We found NRF-1, CPT I, and COX subunit Va mRNA levels were unchanged by chronic anemia. Thus the effects of fetal chronic anemia on mitochondria in the heart may be minimal. Changes in mitochondrial volume density in fetal cardiac myocytes in response to anemia were not examined.
There are several limitations to this study. The sample size for the 3-day anemia studies is small; thus analyses of mRNA and protein levels in these animals may be prone to type II errors. Concern regarding the “housekeeping” genes used for the determination of relative mRNA levels may also exist. The expression of “housekeeping” genes may vary considerably in certain biological samples, depending on the study conditions. Although we did not detect any statistical differences in 28S or 18S rRNA levels between control and anemic myocardial samples, we cannot absolutely rule out the possibility that changes in myocardial rRNA levels occurred in the anemic animals.
The results from these studies demonstrate that the adaptive response of the fetal heart to chronic anemia includes a coordinated induction of a number of myocardial vasculogenic and glycolytic mRNAs that may in part be regulated by HIF-1. The expressions of these mRNAs are likely vital for proper adaptation of the fetal heart to increased work in the face of limited oxygen supply. Furthermore, the expression of these genes in response to acute anemia appears to be tightly regulated, displaying a rapid (3 day), yet nonsustained (7 day) response. The long-term implications of transient stress and accompanying alterations in gene expression in the fetal heart are unknown. Recent studies suggest that a period of fetal anemia results in altered contractile responses to hypoxic stress in the mature adult (4). Detailed understanding of the pathways involved in regulating fetal myocardial vascular and metabolic responses to alterations in the in utero environment and the long-term consequences remain to be elucidated.
This work was supported by National Heart, Lung, and Blood Institute Grants T32-HL-07413 (J. C. Ralphe and A. L. Olson), R01-HL-075446 (R. J. Tomanek), K02-HL-04495 (T. D. Scholz), and R01-HL-64770 (J. L. Segar).
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 © 2005 the American Physiological Society