Enhanced blood pressure variability contributes to left ventricular hypertrophy and end-organ damage, even in the absence of hypertension. We hypothesized that the greater number of high-blood pressure episodes associated with enhanced blood pressure variability causes cardiac hypertrophy and dysfunction by activation of mechanosensitive and autocrine pathways. Normotensive mice were subjected to sinoaortic baroreceptor denervation (SAD) or sham surgery. Twelve weeks later, blood pressure variability was doubled in SAD compared with sham-operated mice. Blood pressure did not differ. Cardiac hypertrophy was reflected in greater heart/body weight ratios, larger myocyte cross-sectional areas, and greater left ventricular collagen deposition. Furthermore, left ventricular atrial and brain natriuretic peptide mRNA expression was greater in SAD than in sham-operated mice. SAD had higher left ventricular end-diastolic pressures and lower myocardial contractility indexes, indicating cardiac dysfunction. Cardiac protein content of phosphorylated p125 focal adhesion kinase (p125 FAK) and phosphorylated p38 mitogen-activated protein kinase (p38 MAPK) was greater in SAD than in sham-operated mice, indicating activation of mechanosensitive pathways of cardiac hypertrophy. Furthermore, enhanced cardiac renin and transforming growth factor-β1 (TGFbeta1) protein content indicates activation of autocrine pathways of cardiac hypertrophy. Adrenal tyrosine hydroxylase protein content and the number of renin-positive glomeruli were not different, suggesting that sympathetic activation and the systemic renin-angiotensin system did not contribute to cardiac hypertrophy. In conclusion, more frequent blood pressure rises in subjects with high blood pressure variability activate mechanosensitive and autocrine pathways leading to cardiac hypertrophy and dysfunction even in the absence of hypertension.
- sinoaortic baroreceptor denervation
- transforming growth factor-beta1
- p125 focal adhesion kinase
- p38 mitogen-activated protein kinase
- left ventricular end-diastolic pressure
almost two decades ago, an association between enhanced blood pressure variability and end-organ damage was first reported (23, 31, 53). It was suggested that the degree of variability of blood pressure during a 24-h period bears a relation to organ damage that is independent of the average blood pressure level (9, 22). A more recent study supports this notion by demonstrating that elevated systolic blood pressure variability is an independent risk factor for stroke (33). Further support for the concept of blood pressure variability being a cardiovascular risk factor that is independent of hypertension comes from animal research. Su and coworkers (19, 25, 26, 39) studied the cardiovascular effects of enhanced blood pressure variability in the rat model of sinoaortic denervation that is characterized by markedly enhanced blood pressure variability and no changes in the mean level of arterial blood pressure. Using this model, they reported left ventricular and aortic hypertrophy (25) and end-organ damage in the heart, kidneys, and small arteries (39). In addition, arterial remodeling and enhanced aortic vasoconstrictor responses to norepinephrine were found (26). Because aortic ANG II concentration was increased and plasma ANG II levels were unchanged, the authors suggested that activation of the local tissue renin-angiotensin system mediates vascular remodeling. Finally, the clinical relevance of blood pressure variability is emphasized by a study demonstrating that a 4-mo antihypertensive treatment with the Ca2+-channel blocker nitrendipine reduced end-organ damage in spontaneously hypertensive rats, while the vasodilator hydralazine failed to prevent cardiovascular damage. Both drugs reduced blood pressure similarly, but only nitrendipine reduced blood pressure variability (19).
Although the contribution of blood pressure variability to cardiovascular morbidity and mortality is well established, the mechanisms by which blood pressure variability causes end-organ damage in the absence of hypertension are still unknown. We propose that the same mechanisms that cause cardiovascular damage in hypertension are involved in end-organ damage secondary to enhanced blood pressure variability, even if the average level of blood pressure is not elevated. The rationale for this proposal is that high blood pressure variability is associated with more episodes where blood pressure falls below the average blood pressure level but also with more episodes where blood pressure rises above the average blood pressure level. The greater number of low and high blood pressure episodes will not alter the average of arterial blood pressure. Rather than protecting the heart, the low-pressure episodes may be detrimental due to coronary hypoperfusion (32, 40). Furthermore, the greater number of high-pressure episodes may activate mechanosensitive and autocrine pathways that lead to cardiac hypertrophy and myocardial dysfunction.
The cellular mechanisms that link mechanical stimuli and cardiac hypertrophy are still far from being well understood. However, recent studies have identified several factors involved in the signaling pathway that mediate cardiac hypertrophy in response to mechanical stress. The starting point of the cascade appears to be integrins, mechanosensitive receptors, located at the extracellular matrix (1). Two of the signaling molecules of the mechanosensitive pathway are p38 mitogen-activated protein kinase (p38 MAPK) and p125 focal adhesion kinase (p125 FAK), which have been reported to induce cardiac hypertrophy (1, 29, 38, 48, 52). Another signaling cascade acts as an autocrine pathway and involves activation of tissue factors, such as the local cardiac renin-angiotensin system (7, 27, 50). ANG II has been demonstrated to increase TGF-β1 secretion in neonatal cardiac fibroblasts (8), and TGF-β1 is considered one of the most important growth factors causing cardiac hypertrophy in response to pressure overload (18, 46).
Therefore, we tested the hypothesis that enhanced blood pressure variability causes cardiac hypertrophy and myocardial dysfunction via activation of mechanosensitive pathways depending on p125 FAK and p38 MAPK, as well as activation of autocrine pathways, involving the local cardiac renin-angiotensin system and TGFbeta1. To test this hypothesis we used a murine model of enhanced blood pressure variability induced by sinoaortic baroreceptor denervation (SAD). The cardiac effects of high blood pressure variability were determined by hemodynamic, morphological, immunohistochemical, and molecular biology parameters. Furthermore, a possible contribution of the systemic renin-angiotensin system and of the sympathetic nervous system to blood pressure-induced cardiac hypertrophy and myocardial dysfunction was tested.
MATERIALS AND METHODS
Animals and baroreceptor denervation.
Male C57BL6 mice, 17–19 g of weight (4–5 wk of age) were obtained from Charles River (Sulzfeld, Germany) and housed individually in cages under standard conditions (25 °C, 12:12-h light-dark cycle). The local authority (Landesamt für Arbeitsschutz, Gesundheitsschutz, and technische Sicherheit, Berlin) approved the experimental protocol, which complied with “APS Guiding Principles for Research Involving Animals and Human Beings.” SAD mice (13, 14, 49) (n = 17) were anesthetized with chloral hydrate (4 mg/10 g body wt ip). The common carotid artery was exposed, and the adventitia was removed from the carotid bifurcation and its branches before brushing them with phenol. Medial to the carotid artery, the aortic depressor nerve was sectioned bilaterally. The laryngeal nerve that also contains aortic baroreceptor fibers was not cut. Thus this procedure is consistent with a complete carotid sinus denervation and a partial aortic baroreceptor denervation (49). Sham operation (n = 10) was performed as above, without applying phenol and cutting the nerve. Survival rate in mice with SAD was 65% 12 wk after surgery. By the end of the 12-wk observation period, six mice subjected to SAD had died, but none of the sham-operated animals died.
Twelve weeks after SAD, catheters were inserted into the right femoral artery and exteriorized between scapulae. The tip of the catheter was tapered on a length of 5 mm to ease cannulation. The inner diameter of the nontapered portion was 0.4 mm, ensuring a sufficiently high conductance, as indicated by pulse pressures of >30 mmHg (Table 1). BP recordings were performed 24 h after surgery in conscious unrestrained mice (42). Animals were observed during the recording period, and recording was stopped once an artifact-free recording of at least 5 min was obtained during which conscious mice were resting and not moving around in the cages. Typically, this required 1–2 h of recording. Recordings of acceptable quality were obtained in 8 sham-operated and 8 SAD mice. Then, mice were anesthetized with chloral hydrate (4 mg/10 g body wt ip), and a catheter (outer diameter 0.6 mm) was forwarded into the left ventricle via the right carotid artery. Left ventricular pressure was recorded for 10 min in anesthetized mice. Acceptable left ventricular pressure signals were obtained in six mice of both groups. All hemodynamic data were digitized using a sampling rate of 1,000 Hz. The details of hemodynamic data analysis are provided in the appendix.
Relative cardiovascular risk.
To estimate the cardiovascular risk, exponential relationships between diastolic and systolic blood pressure and the relative risk of death from cardiovascular diseases, such as coronary heart disease and stroke, were used (20, 30). The data for these relationships are based on the Multiple Risk Factor Intervention Trial (MRFIT, 6 years of follow-up, 350,977 male subjects) (20, 21, 41) and The Ohasama Study (5 years of follow-up, 1,542 subjects) (30). Nonlinear exponential fits to the original data (Fig. 1, middle) provided the following exponential equations: relative risk = exp (0.0437·DBP − 3.867), where DBP is diastolic blood pressure (20); and relative risk = exp (0.0517·SBP − 5.452), where SBP is the systolic blood pressure (30).
Using these equations, the relative cardiovascular risks were calculated for every animal (n = 8 for both groups) by the average of the relative risk values calculated from the systolic and diastolic blood pressure values of every heart beat recorded during resting conscious conditions.
Mice were killed by an intravenous application of potassium chloride. The heart, adrenal glands, and kidneys were excised and rinsed in ice-cold physiological saline. Left and right cardiac ventricles and atria were separated and weighed. Then, all tissues were snap frozen in liquid nitrogen and stored at −80°C for further processing.
The myocyte cross-sectional area of the left ventricles (n = 6 in both groups) was determined by planimetry after staining with fluorescein-conjugated wheat germ agglutinin. Left ventricular collagen deposition was determined by immunohistochemistry using an antibody that detects collagen I and collagen III (n = 4 in both groups). Left ventricular protein expression of TGF-β1 and localization of p125 FAK and p38 MAPK was determined using immunohistochemical techniques and Western blot analysis (n = 4 in both groups). The antibodies for p125 FAK and p38 MAPK are specific for the phosphorylated (activated) forms of the proteins.
Renal protein expression of renin was determined by immunohistochemistry (n = 4 in both groups). The number of renin-positive glomeruli and the optical density of the renin staining in the renal sections were assessed. For determination of cardiac renin protein expression, Western blot analyses were performed from cardiac renin samples obtained by immunoprecipitation assays (n = 4 in both groups). Adrenal tyrosine hydroxylase protein expression was also determined by Western blot analysis (n = 8 in both groups).
Cardiac mRNA levels for atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP) were determined by real-time reverse transcriptase-polymerase chain reaction (RT-PCR) in eight animals of both groups.
The details of these analytical techniques are provided in the appendix.
Data are expressed as means ± SE. Differences between the two groups (SAD and sham-operated mice) were compared using unpaired Student's t-test. Statistical significance was assumed if P < 0.05.
Blood pressure variability was significantly increased 12 wk after surgery in SAD mice (Table 1 and Fig. 2). Furthermore, spontaneous baroreceptor reflex sensitivity (BRR gain) in SAD mice was reduced by >50% compared with sham-operated animals (Table 1). The increased blood pressure variability and the reduced baroreceptor reflex gain indicate successful denervation of the baroreceptors. However, baroreceptor denervation most likely was not complete because a considerable number of baroreflex-triggered sequences were found in the 5-min recordings of the denervated animals. Despite the greater blood pressure variability, systolic, mean, and diastolic blood pressures were not significantly altered in SAD animals (Table 1). Left ventricular end-diastolic pressure (LVEDP) and the maximum of the first derivative of the ventricular pressure curve (dP/dtmax) were calculated as indexes of cardiac preload and myocardial contractility. The significant increase in LVEDP and the marked decrease in contractility index dP/dtmax in SAD mice (Table 1) indicate the presence of cardiac dysfunction and impaired myocardial contractility. The relative cardiovascular risk estimated from the diastolic blood pressure values was significantly (P < 0.05) increased in SAD compared with sham-operated mice (Fig. 1, bottom left). When estimated from the systolic blood pressure values, the difference in relative cardiovascular risk between the two groups was just below the level of statistical significance (P = 0.07).
Heart weights and morphology.
SAD resulted in a significant increase in total heart weight to body weight and total heart weight to tibia length ratios, as well as left ventricular weight to body weight and left ventricular weight to tibia length ratios (Table 2). Collagen protein content was markedly elevated in left ventricles of SAD mice compared with sham-operated animals (Fig. 3, A, B, and E). Because the antibody detects both collagen I and collagen III, we cannot differentiate between the two types of collagen. Furthermore, myocyte cross-sectional area was significantly greater in SAD mice than in sham-operated controls (Table 2 and Fig. 3, C and D), indicating hypertrophy of individual cardiomyocytes.
Left ventricular renin protein expression.
Cardiac renin protein expression was determined by immunoprecipitation followed by Western blot analysis. Only one single band at 40 kDa was obtained in the Western blot from the precipitation products, suggesting that no unspecific binding of the antibody occurred. SAD mice had three times greater cardiac renin expression than sham-operated controls (Fig. 3F).
Atrial and brain natriuretic peptides.
As shown in Fig. 4, ANP and BNP mRNA expression was significantly increased in left ventricles of SAD mice compared with sham-operated animals.
Left ventricular p125 FAK, p38 MAPK, and TGF-β1 protein expression.
Western blot analysis revealed that protein content of phosphorylated (activated) p125 FAK, phosphorylated (activated) p38 MAPK, and TGF-β1 in the left ventricles of SAD mice are significantly increased compared with controls (Fig. 5, top). Using immunofluorescence techniques, a greater protein staining for phosphorylated p125 FAK, phosphorylated p38 MAPK, and TGF-β1 was found in left ventricular sections from SAD mice compared with sham-operated control animals (Fig. 5, bottom).
Renal renin protein expression.
Renal renin protein expression was studied by immunohistochemistry. The antibody only stained juxtaglomerular cells, demonstrating the specificity of the antibody for renin. No difference was observed in the number of renin-positive glomeruli in the kidneys from SAD mice compared with kidneys from sham-operated animals. In addition, the optical density of renin staining in fluorescence-labeled slices of kidneys from SAD and sham-operated mice revealed no differences between groups (Fig. 6, A–D).
Adrenal tyrosine hydroxylase protein expression.
To examine whether sympathetic overactivity occurs as a result of SAD, we determined tyrosine hydroxylase protein expression in adrenal glands by Western blotting. Tyrosine hydroxylase is the rate-limiting enzyme of catecholamine synthesis. Ratios of tyrosine hydroxylase to constitutively expressed α-tubulin were not different in SAD animals compared with sham-operated controls (Fig. 6E).
The results of the present study demonstrate that enhanced blood pressure variability causes cardiac hypertrophy and impaired myocardial function and suggest that mechanosensitive and autocrine pathways contribute to these cardiac effects of high blood pressure variability. Blood pressure variability (variance) was doubled in baroreceptor-denervated mice compared with controls. Cardiac hypertrophy was manifested as increased heart weight/body weight ratios, larger cardiac myocyte cross-sectional areas, enhanced myocardial collagen deposition, and greater cardiac ANP and BNP levels. Impaired myocardial function was indicated by a significant increase in left ventricular end-diastolic pressure and decreased left ventricular contractility index dP/dtmax. It is important to note that these alterations occurred in normotensive mice. Western blot analysis and immunohistochemistry revealed induction of p125 FAK and p38 MAPK (Fig. 5), indicating activation of mechanosensitive pathways. Furthermore, cardiac renin and TGF-β1 protein contents were markedly increased in mice with enhanced blood pressure variability (Fig. 3F and Fig. 5), indicating activation of cardiac autocrine pathways. Thus activation of mechanosensitive pathways (via p125 FAK and p38 MAPK) and autocrine pathways (via cardiac renin-angiotensin system and TGF-β1) may contribute to blood pressure variability-induced cardiac hypertrophy and dysfunction.
Several potential mechanisms for blood pressure variability-induced myocardial hypertrophy and cardiac dysfunction have been explored in this study. It has been suggested that cardiac norepinephrine release is related to the development of hypertensive left ventricular hypertrophy (36). To investigate whether an enhanced sympathetic nervous system activity also contributes to blood pressure variability-induced cardiac hypertrophy, we determined adrenal tyrosine hydroxylase protein expression. This enzyme is rate limiting for catecholamine synthesis, and its adrenal expression has been demonstrated to reflect preganglionic sympathetic nerve activity to the adrenal glands (5, 28). Furthermore, it correlates well with splanchnic sympathetic nerve activity (10) and, therefore, can be considered a useful marker for overall sympathetic activity. Adrenal tyrosine hydroxylase expression was not different in mice with enhanced blood pressure variability and control animals. This finding suggests that overall sympathetic activation does not contribute to myocardial hypertrophy and cardiac dysfunction in this model of enhanced blood pressure variability. However, we cannot exclude the possibility that a selective increase in cardiac sympathetic nerve activity that would not be reflected in adrenal tyrosine hydroxylase expression, contributed to the cardiac alterations in mice with enhanced blood pressure variability.
An important question is how mechanical signals, such as enhanced blood pressure variability, are transmitted into myocytes to induce cardiac hypertrophy. Although the exact pathways are still not completely understood, the current view is that integrins, which are receptors of the extracellular matrix, sense the mechanical signal and cause activation of the mechanosensitive signaling cascade. It has been suggested that p38 MAPK and p125 FAK are important factors for the signaling cascade of cell growth, differentiation, and transformation in response to mechanical stress (1, 24, 29). Therefore, we studied cardiac activation of p38 MAPK and p125 FAK to test the hypothesis that mechanosensitive pathways contribute to cardiac hypertrophy in this model of enhanced blood pressure variability. Indeed, the phosphorylated (activated) forms of these kinases (p38 MAPK and p125 FAK) were markedly enhanced in the left ventricles of SAD mice compared with sham-operated controls. This finding indicates that enhanced blood pressure variability, indeed, acts via mechanical stress to elicit cardiac hypertrophy. We propose that the mechanical stimuli are the more frequent high-blood pressure episodes that accompany enhanced blood pressure variability (Fig. 1, top). In this regard, it is important to point out, that sinoaortic baroreceptor denervation has long been thought to enhance mainly short-term blood pressure variability. However, more recent data suggest that baroreceptors can also buffer long-term blood pressure variability (47). Thus, in addition to short-term blood pressure variability, long-term blood pressure variability may also have contributed to cardiac hypertrophy and dysfunction in our study, because it is known that long-term blood pressure variability can contribute to organ damage (31, 44).
ANG II has been demonstrated to increase TGF-β1 secretion of neonatal cardiac fibroblasts (8), and ANG II-induced cardiac hypertrophy was absent in TGF-β1 knockout mice (37). Therefore, we tested whether an activated renin-angiotensin system also contributed to the elevated TGF-β1 expression and to cardiac hypertrophy in mice with enhanced blood pressure variability. Immunofluorescence staining of the kidneys for renin revealed no significant differences in the number of renin-positive glomeruli between mice with enhanced blood pressure variability and controls. However, left ventricular renin content determined by immunoprecipitation assay was three times greater in mice with SAD than in control animals (Fig. 3F). Thus our data support the view that enhanced blood pressure variability activates the local cardiac tissue renin-angiotensin system independently from the systemic renin-angiotensin system. One may thus speculate that activation of the cardiac tissue renin-angiotensin system contributed to the elevated expression of TGF-β1 and cardiac hypertrophy in mice with enhanced blood pressure variability.
TGF-β1 is considered one of the most important growth factors causing cardiac hypertrophy in response to pressure overload (18, 46) and humoral factors, such as ANG II (8, 37, 46). Left ventricular TGF-β1 protein expression was significantly elevated in mice with enhanced blood pressure variability. Thus ANG II-induced TGF-β1 stimulation may indeed be involved in blood pressure variability-induced cardiac hypertrophy. This assumption is further substantiated by the localization of TGF-β1 in cardiac fibroblasts and myocytes as detected by immunohistochemistry (Fig. 5). One mechanism by which TGF-β1 causes cardiac hypertrophy and cardiac remodeling is by stimulating the expression of proteins of the extracellular matrix, such as fibronectin and collagen (11, 12). Collagen synthesis occurs mainly in cardiac fibroblasts (4). Interestingly, we found abundant staining for TGF-β1 in cardiac fibroblasts and a marked increase in cardiac collagen I and III deposition (Fig. 3) in mice with enhanced blood pressure variability. Thus TGF-β1, activated by blood pressure variability-induced stimulation of the cardiac renin-angiotensin system, may have elicited collagen synthesis and deposition within the collagen matrix of the heart. The collagen matrix represents a major determinant of myocardial stiffness (45). Thus the elevated left ventricular end-diastolic pressure in mice with enhanced blood pressure variability may be secondary to a greater left ventricular stiffness caused by increased cardiac collagen deposition. Impaired myocardial contractility, as indicated by a reduction in left ventricular dP/dtmax, may relate to slowing of calcium transients that frequently accompanies cardiac fibrosis (45) and may also contribute to the increase in left ventricular end-diastolic pressure.
Twelve weeks after SAD, the calculated cardiovascular risk, estimated based on the well-established relationship between SBP or DBP and the risk of death from cardiovascular endpoints, was significantly (for DBP, P = 0.026) and almost significantly (P = 0.07 for SBP) elevated compared with sham-operated mice. Since the average systolic and diastolic blood pressures were not significantly higher in sinoaortic baroreceptor-denervated mice than in control animals, this result may be surprising. However, the exponential relationship between SBP and DBP and cardiovascular risk (Fig. 1, middle), together with the greater number of high blood pressure episodes associated with elevated blood pressure variability, explains the higher cardiovascular risk in SAD mice. The greater number of low blood pressure episodes (Fig. 1, top, left shaded area) does not reduce cardiovascular risk as much as the high blood pressure episodes (Fig. 1, top, right shaded area) add to it. In fact, low blood pressure episodes may reduce organ perfusion and add to organ damage elicited by high blood pressure variability (32, 40). Therefore, we propose that the greater number of high blood pressure episodes associated with elevated blood pressure variability causes activation of mechanosensitive (p125 FAK and p38 MAPK) and autocrine (cardiac renin and TGF-β1) pathways that ultimately lead to cardiac hypertrophy, fibrosis, and dysfunction.
The relative cardiovascular risk was estimated based on exponential relationships between DBP and SBP and the relative risk for death from cardiovascular diseases (20, 21, 30, 41). Although it may be problematic to apply data obtained in humans to mice, there is no reason not to believe that a similar exponential relationship between arterial blood pressure and cardiovascular endpoints also exists in rodents. The values of the parameters for this exponential relationship are likely to be different in humans and mice. Furthermore, the cardiovascular risk values calculated in this study certainly do not reflect the true cardiovascular risk in the mice of this study. The intention of these calculations was to propose a method that allows estimation of cardiovascular risk based on both the average blood pressure level and the blood pressure variability. This is possible by applying the blood pressure values of every heartbeat to the exponential relationship between blood pressure and cardiovascular risk. This approach will reflect cardiovascular risk more precisely than a calculation based on the average blood pressure level or the blood pressure variability taken alone. This consideration implies that even normotensive subjects may have an enhanced cardiovascular risk if their blood pressure variability is increased. Furthermore, pharmacological treatment aimed at reducing blood pressure variability may be beneficial not only in hypertensive but also in normotensive subjects. In this regard, reducing blood pressure variability has been demonstrated to decrease left ventricular hypertrophy, diminish glomerulosclerosis, and ameliorate vascular lesions in spontaneously hypertensive rats (3, 51).
In conclusion, this study demonstrates myocardial hypertrophy and cardiac dysfunction that is independent of hypertension in mice with enhanced blood pressure variability. Activation of mechanosensitive and autocrine pathways appears to be involved in the mechanisms of cardiac hypertrophy and dysfunction in this model. We propose that these events are triggered by the greater number of high blood pressure episodes that are associated with enhanced blood pressure variability. The greater number of low blood pressure episodes does not prevent myocardial hypertrophy and cardiac dysfunction but prevents a significant increase in the average level of arterial blood pressure and therefore mask hypertension.
Hemodynamic data analysis.
Systolic, mean, and diastolic blood pressure and heart rate were derived for every heartbeat from the 5 min of resting blood pressure recordings obtained in conscious animals (n = 8 in both groups). Blood pressure and heart rate variability were estimated from 2,619 ± 170 heartbeats (corresponding to the 5 min of resting blood pressure recordings) as variance of the beat-by-beat time series for systolic, mean, and diastolic blood pressure and heart rate. Because of the interindividual variability in heart rate, the number of heartbeats was not exactly the same in each mouse. End-diastolic pressure was extracted from the left ventricular pressure signal obtained in anesthetized animals at the time points of the local maxima of the second derivative of the left ventricular pressure signal. The maximum of the first derivative of the left ventricular pressure curve was used as an index for myocardial contractility (dP/dtmax). Spontaneous BRR gain was determined by the sequence method as described previously (2, 16, 43). Briefly, sequences of three or more consecutive blood pressure pulses, where systolic blood pressure and pulse interval changed in the same direction and are positively correlated (r < 0.8), were detected and the slope of these sequences was used as a measure of BRR gain. Only those animals were included in which at least three sequences were detected in the 5-min recording obtained in conscious conditions (n = 7 for SAD animals, n = 5 for sham-operated animals).
Heart morphometry and cardiac and renal immunohistochemistry.
Frozen left ventricles were embedded in water-soluble medium (Cryo-Gel, Instrumedics,), sectioned (10 μm), fixed in 4% paraformaldehyde/PBS, rehydrated in PBS, and stained with fluorescein-conjugated wheat germ agglutinin (Vector Laboratories). Determination of myocyte cross-sectional area (n = 6 for both groups) was done in 750 myocytes per ventricle (15 transverse sections per ventricle, 5 optical fields per section, 10 cells per field, magnification: 400×, ImageJ, National Institutes of Health).
For collagen I/III staining (n = 4 for both groups) sections of the left ventricle were fixed in methanol/acetone and rehydrated with PBS. Endogenous peroxidase activity was quenched with 0.3% H2O2 in PBS and blocked by 10% normal goat serum. Sections were incubated overnight with a rabbit anti-collagen I/III (Calbiochem, San Diego, CA) primary antibody at a 1:500 dilution at 4°C. This antibody detects both collagen I and collagen III (35). After washing, a secondary goat anti-rabbit peroxidase-conjugated antibody was used at a 1:800 dilution. Sections were washed again and visualized by reaction of peroxidase with diaminobenzidine (Sigma-Aldrich, St. Louis, MO) forming a brown precipitate.
For localization of p125 FAK (n = 4 for both groups), p38 MAPK (n = 4 for both groups), and TGF-β1 (n = 4 for both groups), protein expression within the myocardium of the left ventricle, sections were incubated with a rabbit anti-p125 FAK antibody (Calbiochem), a rabbit anti-p38 MAPK antibody (Calbiochem), or a rabbit anti-TGF-β1 antibody (Torrey Pines Biolabs, Houston, TX) at dilutions of 1:100, 1:100, and 1:400, respectively (17). For visualization, a FITC-conjugated secondary antibody (Merck Biosciences, Bad Soden, Germany) was used at a 1:400 dilution. Phalloidin-tetramethylrhodamine isothiocyanate (TRITC; Sigma) background staining was performed to visualize fibrillar structures and to ease identification of cardiac and vascular myocytes.
Staining for renin was performed in eight cryosections per kidney (n = 4 for both groups). The rabbit anti-renin antibody was a generous gift from Prof. T. Inagami, Vanderbilt University and was used at a 1:300 dilution (34). A secondary FITC-conjugated antibody (Merck Biosciences) was used at a 1:500 dilution. Again, phalloidin-TRITC (Sigma) background staining was performed to visualize fibrillar structures and to ease identification of blood vessels and glomeruli. A total of four independent optical fields were analyzed per section by counting renin-positive glomeruli. In addition, renin staining was analyzed in all eight sections from each kidney using an image processing software (ImageJ).
Left ventricles (n = 4 for both groups) were homogenized, and samples containing 10 μg total protein were incubated with a rabbit anti-renin antibody (34) in a 1:100 dilution. Incubation was done overnight at 4°C. Subsequently, 50 μl of protein G-sepharose (Sigma) was added to each sample followed by a 2-h incubation period at room temperature. Then, the samples were centrifuged (5 min, 10,000 g), and the sediment was washed with PBS. This washing step was repeated three times. A Laemmli sample buffer was added, and the samples were heated to 90°C for 5 min. and centrifuged (5 min, 10,000 g). The supernatants containing the renin antigen were submitted to SDS-PAGE gel electrophoresis and Western blotting.
Left ventricles and adrenal glands were homogenized in liquid nitrogen, and samples were prepared as described previously (15). Fifty micrograms of total protein from each sample and from cardiac renin immunoprecipitates were separated by 7.5% and 10% SDS-PAGE gel electrophoresis, respectively, and transferred onto nitrocellulose membranes [Hybond enhanced chemiluminescence (ECL), Amersham Pharmacia Biotech, Little Chalfont, United Kingdom] or Protran membrane for cardiac renin samples. After blocking, rabbit anti-p125 FAK antibody (1:1,000 dilution), rabbit anti-p38 MAPK antibody (1:1,000 dilution), rabbit anti-TGF-β1 antibody (1:2,000 dilution) (17), mouse anti-tyrosine hydroxylase monoclonal antibody (1:5,000 dilution, Calbiochem), or rabbit anti-renin antibody (1:5,000 dilution) (34) were applied 150 min on membranes with samples from the left ventricles (n = 4 for both groups), adrenal glands (n = 8 for both groups), or cardiac renin precipitation products (n = 4 for both groups). After appropriate washing, the membranes were incubated for 60 min with a horseradish peroxidase-linked secondary antibody (Amersham Pharmacia Biotech) diluted 1:5,000 in Tween-20 Tris-base sodium (1:7,000 for cardiac renin samples). The same membranes were used for p125 FAK, p38 MAPK, and TGF-β1 by stripping and reprobing. Immunolabeled proteins were visualized by chemiluminescence using a commercially available ECL Western blotting kit (Amersham Pharmacia Biotech). To control for protein loading, membranes were stripped, and the whole protocol was repeated using a mouse anti-α-tubulin monoclonal antibody (1:3,000, Santa Cruz Biotechnologies, Santa Cruz, CA). Except for cardiac renin (immunoprecipitation), protein content was expressed as the ratio of optical densities of the specific spots for the respective proteins and spots corresponding to constitutively expressed α-tubulin from the same sample.
For mRNA quantification, left ventricular myocardial samples were analyzed. Total RNA preparation, complete deoxyribonuclease digestion, and reverse transcription were performed as described previously (6). The mRNA levels for ANP (n = 8 for both groups), BNP (n = 8 for both groups) and the reference gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were determined by a “hot start” PCR procedure with SYBR Green detection. This procedure was validated with respect to reproducibility and linearity within the measurement range and performed in duplicate with the TaqMan (7,700) instrument (Applied Biosystems, Foster City, CA). A calibration curve was used for each PCR reaction to estimate mRNA amount within each sample. To correct for potential variations between samples in mRNA extraction or in reverse transcription efficiency, the mRNA content of the target gene was normalized to the expression of the constantly expressed reference gene GAPDH in the same sample. PCR products were identified by sequencing. Results are expressed as ratio of ANP or BNP/GAPDH. The following primer sequences were used:
This study was supported by the German Research Foundation (GRK 754/1–01).
We thank J. Thomas and B. Hannack for excellent technical assistance and N. Wagner and P. Krenek for much technical advice.
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