AJP - Heart Watch the video to learn how APS reaches out to developing nations.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 294: H659-H667, 2008. First published November 21, 2007; doi:10.1152/ajpheart.01147.2007
0363-6135/08 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
294/2/H659    most recent
01147.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Xiao, H. D.
Right arrow Articles by Bernstein, K. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xiao, H. D.
Right arrow Articles by Bernstein, K. E.

Mice expressing ACE only in the heart show that increased cardiac angiotensin II is not associated with cardiac hypertrophy

Hong D. Xiao,1 Sebastien Fuchs,1 Ellen A. Bernstein,1 Ping Li,1 Duncan J. Campbell,2 and Kenneth E. Bernstein1

1Department of Pathology and Laboratory Medicine, Emory University, Atlanta, Georgia; and 2Saint Vincent's Institute of Medical Research and the Department of Medicine, University of Melbourne, Saint Vincent's Hospital, Fitzroy, Victoria, Australia

Submitted 3 October 2007 ; accepted in final form 14 November 2007


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
In the heart, angiotensin II has been suggested to regulate cardiac remodeling and promote cardiac hypertrophy. To examine this, we studied compound heterozygous mice, called angiotensin-converting enzyme (ACE) 1/8, in which one ACE allele is null, whereas the other ACE allele (the 8 allele) targets expression to the heart. In this model, cardiac ACE levels are about 15 times those of wild-type mice, and ACE expression is reduced or eliminated in other tissues. ACE 1/8 mice have 58% the cardiac ACE of a previous model, called ACE 8/8, but both ACE 1/8 and ACE 8/8 mice have ventricular angiotensin II levels about twofold those of wild-type controls. Despite equivalent levels of cardiac angiotensin II, ACE 1/8 mice do not develop the marked atrial enlargement or the conduction defects previously reported in the ACE 8/8 mice. Six-month-old ACE 1/8 mice have normal cardiac function, as determined by echocardiography and left ventricular catheterization, despite the elevated levels of angiotensin II. ACE 1/8 mice also have normal levels of connexin 43. Both wild-type and ACE 1/8 mice develop similar degrees of cardiac hypertrophy after aortic banding. These data suggest that a moderate increase of local angiotensin II production in the heart does not produce cardiac dysfunction, at least under basal conditions, and that, in response to aortic banding, cardiac hypertrophy is not augmented by a twofold increase of cardiac angiotensin II.

connexin; gene targeting; blood pressure; genetically modified mice

THE RENIN-ANGIOTENSIN SYSTEM (RAS) plays an important role in the regulation of cardiovascular function. A central component of the RAS is angiotensin-converting enzyme (ACE), which converts angiotensin I to angiotensin II, the major effector molecule of the RAS (4). The main site of ACE expression is vascular endothelium, producing angiotensin II adjacent to vascular smooth muscle. ACE is produced by many tissues, but high levels are found in the lung, kidney, and testes. The elimination of ACE in genetically engineered mice leads to a marked reduction of blood pressure, underlining the vital role of ACE in the normal control of blood pressure (7). ACE inhibitors are clinically used as antihypertensive agents. ACE inhibitors also improve disease in heart failure patients and patients with diabetic nephropathy (8, 12). These effects may not be explained simply by blood pressure reduction, since other antihypertensive drugs do not appear to give the same results. In recent years, the paradigm of a locally functioning RAS has gained significant attention, since essential components of the RAS were identified in organs such as the heart and kidney (1). This paradigm posits that angiotensin II may have direct effects on tissues apart from changes in blood pressure.

In the heart, angiotensin II has been suggested to regulate cardiac remodeling and promote cardiac hypertrophy (6). A direct role of angiotensin II is supported by in vitro studies showing that angiotensin II can stimulate the growth of cardiomyocytes in cell culture (9, 24, 25). In vivo experiments also show that the infusion of a subpressor dose of angiotensin II was sufficient to induce cardiac hypertrophy in the absence of reported blood pressure changes (18, 22). However, even with a subpressor dose, a systemic effect of angiotensin II cannot be completely ruled out. We have reported a different approach to evaluate the local function of angiotensin II in the heart by creating a line of mice, called ACE 8/8, in which ACE was overexpressed in cardiomyocytes but simultaneously eliminated from most other tissues (28). This was achieved by targeted homologous recombination in stem cells to position the {alpha}-myosin heavy chain ({alpha}-MHC) promoter to control ACE expression. Because of this genetic change, ACE 8/8 mice produce very high levels of ACE in atria and ventricles but lack ACE expression by endothelium, vascular adventitia, and the kidney. Thus ACE 8/8 mice differ from other transgenic models that overexpress ACE or the angiotensin II type 1 receptor on a background of normal RAS protein expression. Because ACE is localized in the myocardium, ACE 8/8 mice produce higher levels of angiotensin II in the heart without a systemic increase of angiotensin II production. As a result, these mice had a nearly normal blood pressure, normal ventricular size, and normal ventricular contraction velocity. Instead of ventricular hypertrophy, ACE 8/8 mice developed enlarged atria and cardiac electrical abnormalities, such as prolonged atrial-ventricular conduction and a markedly reduced QRS amplitude (28).

One of the concerns in analyzing the ACE 8/8 mice was to discriminate effects due to the localized production of cardiac angiotensin II versus any potential side effects induced by a very high local production of ACE protein. To address this and to further understand the role of angiotensin II in the heart, we studied a line of compound heterozygous mice created by breeding ACE 8/8 mice with a line of ACE knockout mice called ACE 1/1. The ACE allele termed ACE 1 is a classic null allele; compound heterozygous mice, designated 1/8, produce half the amount of ACE in cardiac tissues compared with ACE 8/8 mice but maintain the same ACE tissue distribution pattern. Surprisingly, ACE 1/8 mice do not show the cardiac defects observed in ACE 8/8 mice, even though both lines of mice have similarly elevated levels of cardiac angiotensin II. In addition, induction of cardiac hypertrophy by aortic banding generated equivalent levels of cardiac hypertrophy in ACE 1/8 mice, compared with wild-type controls, suggesting that the local increase in cardiac angiotensin II production did not augment cardiac hypertrophy beyond the hemodynamic changes induced by banding. The ACE 1/8 model strongly suggests that it is not the overexpression of angiotensin II in cardiac tissues that induces the cardiac changes observed in ACE 8/8 mice. This view is reinforced by studying ACE 8/8 mice lacking all production of angiotensinogen. Thus our studies of the ACE 1/8 mice suggest that elevated cardiac angiotensin II production is not deleterious to cardiac morphology or function under both basal conditions and the pathological changes induced by aortic banding.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Creation and genotyping of ACE 1/8 mice. To produce ACE 1/8 compound heterozygote mice, ACE 1/wt mice were mated with ACE 8/wt mice, where wt is the wild-type ACE allele. Both the ACE 1/1 and the ACE 8/8 lines were created using targeted homologous recombination (7, 28). These mice have a mixed genetic background of 129/SVx129/SvJ (129) and C57BL/6. Specifically, the ACE 1/wt mice used in matings were in the seventh generation of backcrossing to C57Bl/6, whereas the ACE 8/8 mice were from the F2 generation. Thus the ACE 1/8 mice were also a mixed background of 129 and C57BL/6. Age- and sex-matched littermate controls were used in all studies. Animal procedures were approved by the Institutional Animal Care and Use Committee and were supervised by the Emory University Division of Animal Research.

Genomic DNA was isolated from tail clippings, and mice were genotyped by PCR. Four primers were used for PCR genotyping. Two primers, 5'-AGCAGGCATATGGGATGGGA and 5'-TTCTCCTCCGTGATGTTGGT, amplify a 334-bp fragment of the ACE 8 mutated allele. Another two primers, 5'-GGATCTTGCTGCCCTCTATG and 5'-GGTTGTTCAGACTACAATCTGACC, amplify a fragment of about 600 bp from the ACE 1-mutated allele. If neither fragment was amplified, a second PCR amplification was performed to confirm that the DNA samples contained the wild-type DNA allele. The second PCR used the same primer set as those used for ACE 8/8 genotyping (28).

ACE activity assay. Heart, lung, and kidney samples were briefly homogenized at low speed in ACE homogenization buffer containing 50 mmol/l HEPES (pH 7.4), 150 mmol/l NaCl, 25 mol/l ZnCl2, and 1 mmol/l PMSF. These homogenates were centrifuged at 10,000 g, and the supernatant was discarded to remove blood from tissue. Tissue pellets were then resuspended in ACE homogenization buffer containing 0.5% Triton X-100 and rehomogenized at high speed. The resulting tissue homogenates were again spun at 10,000 g, and supernatants were used for ACE activity assay. Plasma samples were directly diluted 1:10 in ACE buffer containing Triton X-100. ACE activity was measured using the ACE-REA kit from American Laboratory Products, (Alpco, Windham, NH) following the manufacturer's instructions. Protein assay was performed using BCA Protein Assay Reagent kit (Pierce, Rockford, IL). Tissue ACE activity was calculated as ACE units per microgram protein. Plasma ACE activity was calculated as ACE units per microliter plasma.

Blood and tissue angiotensin peptide levels. Mice were anesthetized by intraperitoneal injection of a mixture of ketamine and xylazine (125 and 12.5 mg/kg body wt, respectively). Blood was collected from the inferior vena cava directly into a syringe containing 5 ml of 4 mol/l guanidine thiocyanate using a 25-gauge needle. Ventricles and the left kidneys were then rapidly removed and homogenized in 5 ml guanidine thiocyanate. The blood and tissue homogenates were frozen and stored at –80°C until shipped on dry ice to St. Vincent's Institute of Medical Research, where peptide measurements were performed. Angiotensin I and II peptides were measured using HPLC-based radioimmunoassays as previously described (28). The method analyzed both angiotensin I and II in the same sample during a single HPLC run, thereby reducing the variance of the peptide ratio. Two data points were removed from the blood peptide measurements because of very low blood angiotensin I values. Both points were more than three standard deviations from the means and would have significantly distorted the angiotensin II/angiotensin I data if they were kept.

Blood pressure and urine osmolality. Systolic blood pressure and urine osmolality were measured as previously described (28). Briefly, blood pressure was measured in conscious mice using a Visitech Systems BP2000-automated tail-cuff system (Apex). Approximately 80 measurements, collected over at least 4 days, were averaged to calculate the pressure. Urine osmolality was determined using a Wescat 5500 Vapor Pressure Osmometer (Wescor, Logan, UT). Urine samples were collected before and after 24 h of water deprivation.

Transverse aortic banding. Mice (12–16 wk old) were anesthetized by intraperitoneal injection of ketamine and xylazine. An endotracheal tube was inserted through the mouth, and the mice were ventilated with room air using a minivent rodent ventilator (Harvard Apparatus, Holliston, MA). The left thorax was opened at the second intercostal space, and a 7-0 silk-suture ligature was tied around the transverse aorta between the right brachiocephalic trunk and the left carotid artery against a 27-gauge needle. The needle was then withdrawn. The chest wall, muscle, and skin were sequentially closed using 6-0 silk sutures. Sham-operated controls went through the same procedure but without banding.

Left ventricular catheterization and in vivo hemodynamic measurement. Left ventricular (LV) pressure recordings were performed as previously described (28). Under ketamine and xylazine anesthesia, a 1.4-Fr Millar high-fidelity pressure catheter (SPR-671, AD Instruments) was inserted into the right carotid artery and then advanced into the left ventricle. Data were recorded using the PowerLab system and Chart 5 software (AD Instruments). Zero pressure was established at the end of the experiment with the probe in an open pool of blood. LV systolic pressure and LV end-diastolic pressure were calculated directly from the LV pressure wave forms. The maximum and minimum first-degree differential of the LV pressure (LV dP/dtmax and LV dP/dtmin, respectively) were obtained. For right carotid pressure measurement, a 1.4-Fr Millar probe was inserted into the artery without advancing into the left ventricle.

ECG monitoring. Surface ECG was recorded in mice anesthetized with an intraperitoneal injection of ketamine and xylazine (125 and 12.5 mg/kg, respectively). Three surface probes were inserted into the subcutaneous space, and the probes were placed following a lead II configuration. ECG data were also recorded using the PowerLab system and Chart 5 software (AD Instruments). Both QRS amplitude and PR intervals were measure using Chart 5 software. QRS amplitude was defined as the distance between the highest and lowest point of a QRS complex, and data from five consecutive QRS complexes were averaged. For the PR interval, we measured the distance between the highest point of a P wave and the highest point of the next R wave. Data were averaged from three measurements from well-defined P waves.

Western blot analysis. Tissue homogenates were prepared as described for the ACE activity assay, except that 50 mM NaF and 1 mM Na3VO4 were added into the homogenization buffer. Protein samples were separated on a 10% SDS gel and transferred to a nitrocellulose membrane. The membrane was blotted with a 1:250 dilution of anti-connexin 43 antibody (Zymed, South San Francisco, CA). An anti-GAPDH antibody at a concentration of 1:1,000 (Santa Cruz Biotechnology, Santa Cruz, CA) was used as a loading control. Donkey anti-rabbit antibody (Amersham, Piscataway, NJ) was used at a concentration of 1:10,000. The membranes were exposed to X-ray film using the enhanced chemiluminescence method.

Statistical analysis. All data were expressed as means ± SE. The significance of the difference between two groups was obtained by an unpaired Student's t-test. The significance of the difference among multiple groups was obtained using analysis of variance (ANOVA) and the Tukey honestly significant difference test.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Creation of ACE 1/8 mice. The ACE 1/8 compound heterozygous mice were created by crossing two strains of genetically manipulated mice, ACE 1/1 and ACE 8/8. Both these models were previously described and contain mutations produced by targeted homologous recombination in embryonic stem cells (7, 28). ACE 1/8 mice were created by breeding heterozygous ACE 1/wt and ACE 8/wt mice. The resulting offspring were ACE 1/8, ACE 1/wt, ACE 8/wt, and ACE wt/wt, respectively. Of 347 pups genotyped, 105 were ACE 1/8, 90 were ACE 1/wt, 80 were ACE 8/wt, and 72 were ACE wt/wt. This pattern of births is not significantly different from Mendelian distribution as analyzed by the {chi}-square test, indicating that ACE 1/8 mice do not have increased mortality compared with wild-type mice before weaning. These mice carry a mixed genetic background of C57BL/6 and 129, similar to the genetic background of the ACE 8/8 mice.

ACE and angiotensin peptide levels. ACE expression levels in ACE 1/8, ACE 8/wt, and littermate wild-type mice were examined by ACE activity assay (Fig. 1). In the ventricles of the heart, ACE levels were 59.5 ± 0.8 ACE U/µg protein in the ACE 1/8 mice. This compares with ACE levels of 61.5 ± 0.4 U/µg protein in ACE 8/wt mice and 4.1 ± 0.7 ACE U/µg protein in wild-type mice. These data show that cardiac ACE levels in ACE 1/8 mice were about 58% of the previously reported level of 104.5 U/µg protein in the ventricles of ACE 8/8 mice (28). The data are consistent with the known effect of the ACE 1 allele, which is a traditional null allele without any ACE expression.


Figure 1
View larger version (10K):
[in this window]
[in a new window]

  		Fig. 1. Angiotensin-converting enzyme (ACE) activity. Littermate wild-type (WT, black bars), ACE 8/wt (white bars), and ACE 1/8 (gray bars) mice, where wt is the WT ACE allele, were euthanized, and ACE activity was measured in organ homogenates and plasma. ACE activity is expressed as units per microgram solubilized protein in tissues and as units per microliter in plasma samples. The data are the average of 5 (ACE 1/8) or 6 individual mice. Data are the group means ± SE. <DL, less than the detection limit. * P < 0.05, **P < 0.01, and ***P < 0.001 compared with WT.

 
Previous analysis of the ACE 8/8 model documented lung ACE levels that were 43% those of wild-type mice. This was due in part to high ACE expression in the vascular smooth muscle of pulmonary artery branches accompanying the bronchial tree and reflected the demonstrated activity of the {alpha}-MHC promoter in pulmonary vascular smooth muscle. In ACE 1/8 mice, lung ACE levels were 30% those of wild-type mice and the histological distribution was similar to ACE 8/8 (data not shown). ACE 8/wt mice demonstrated 79% wild-type ACE levels, due to the presence of one wild-type ACE allele. In the kidney, ACE 1/8 mice do not express measurable levels of ACE, similar to the results found in ACE 8/8 mice. Thus an analysis of ACE activity shows that ACE 1/8 mice have the same pattern of tissue ACE expression found in ACE 8/8 mice, although the amount of ACE expression in the heart is substantially reduced.

In plasma, ACE 1/8 mice have 23% the ACE activity of wild-type mice. This was a little less than half the 56% previously measured in ACE 8/8 mice. Plasma ACE activity in ACE 8/wt mice was 75% that of wild-type mice.

Radioimmunoassay was used to quantify angiotensin II and I levels, as well as the ratio of angiotensin II to angiotensin I, in the ventricles of wild-type, ACE 8/wt, ACE 1/8, and ACE 8/8 mice (Fig. 2). Both the ACE 1/8 and the ACE 8/8 mice have ~2.1-fold the angiotensin II concentration found in ventricles from wild-type mice. In contrast, the angiotensin II concentration in the ACE 8/wt ventricles was not significantly higher than wild-type (wild-type, 53.2 ± 6.6; ACE 8/wt, 64.7 ± 8.9; ACE 1/8, 112.1 ± 14.4; and ACE 8/8, 111.2 ± 13.1 fmol/g). We also measured ventricular angiotensin I levels, but here no difference was detected between the four groups. The angiotensin II-to-angiotensin I ratio nicely demonstrates the similarity between these mice, because both models have a ratio of approximately twofold higher than that seen in either wild-type or ACE 8/wt mice.


Figure 2
View larger version (22K):
[in this window]
[in a new window]

  		Fig. 2. Angiotensin (ANG) peptide levels. ANG II and ANG I peptide levels were measured in individual samples of heart (ventricles), blood, and kidney by an HPLC-based method, as previously described (28). Data are also presented as the ratio of ANG II to ANG I. Shown here are WT (n = 15), ACE 8/wt (n = 13), ACE 1/8 (n = 11), and ACE 8/8 (n = 6) mice. All data are means ± SE. **P < 0.01 and ***P < 0.001 compared with WT. In heart, the ANG II levels and ANG II-to-ANG I ratio are not different between ACE 1/8 and ACE 8/8 mice.

 
It should be noted that we previously reported that angiotensin II levels in the ACE 8/8 heart were about fourfold that of the wild-type levels (28). This is different from the new set of data that showed only a 2.1-fold increase of angiotensin II in both ACE 1/8 and ACE 8/8 mice compared with wild-type mice. This difference may be attributed to sample variations between animals and the variation of the peptide assay itself.

We also measured angiotensin peptide levels in blood and in renal tissue (Fig. 2). In the blood, there was no significant difference between angiotensin II levels in the four groups analyzed. However, ACE 1/8 mice, having only one active ACE allele, demonstrated approximately a threefold increase in blood concentration of angiotensin I. This is also reflected in the ratio of angiotensin II to angiotensin I, where the ACE 1/8 group showed a significant reduction from either wild-type, ACE 8/wt, or ACE 8/8 mice. These data are consistent with our previous conclusion that even a single null ACE allele results in a compensatory change in angiotensin I levels and that this is sufficient to reestablish physiological levels of angiotensin II (3). Angiotensin I levels in the ACE 1/8 kidney are increased compared with wild-type mice. However, this difference did not reach statistical significance by ANOVA analysis (P = 0.02 between ACE 1/8 and wild-type by t-test, P = 0.056 by ANOVA after the Tukey post hoc test). Renal angiotensin II levels were not significantly different between the four groups studied.

Blood pressure and renal function of ACE 1/8 mice. The upregulation of plasma angiotensin I in ACE 1/8 mice suggests a partial activation of the RAS. To evaluate how this affected overall physiological function, the systolic blood pressure and renal concentrating ability of both ACE 1/8 and wild-type mice were evaluated. Systolic blood pressure was measured in conscious mice using the tail-cuff method and showed no difference between the two groups (wild-type, 108.4 ± 4.7 mmHg, n = 6; and ACE 1/8, 111.3 ± 2.7 mmHg, n = 11). Renal concentrating ability was evaluated by measuring urine osmolality before and after 24 h of water deprivation (Fig. 3). We also measured the percentage of body weight reduction observed after the 24 h of water deprivation. Although both groups of mice markedly increased their urine osmolality to >3,000 mosmol/kgH2O, the levels reached by the ACE 1/8 mice were slightly less than observed in wild-type mice, though these values were not significantly different. Furthermore, the analysis of weight loss after water deprivation showed no significant difference between the two groups (wild-type, –9.2 ± 0.5% body wt, n = 13; and ACE 1/8, –10.1 ± 0.6% body wt, n = 17). Thus these data suggest that the combination of one ACE null allele and one allele targeting ACE expression to the heart does not substantially disadvantage a mouse exposed to the stress of dehydration.


Figure 3
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 3. Renal concentrating ability. Spot urine was collected from WT (n = 13) or ACE 1/8 mice (n = 17) before or after 24 h of water deprivation, and urine osmolality was measured. Data for individual mice are shown, as well as for the group (means ± SE). There is no statistical difference between WT and ACE 1/8 mice before or after 24 h of water deprivation.

 
Cardiac function of ACE 1/8 mice. The major morphological change we observed in the ACE 8/8 mice was marked atrial enlargement with an atrial weight three times that of the wild-type mice (28). In ACE 1/8 mice, no gross heart morphological changes were identified by visual examination. However, the ACE 1/8 atria did show a slightly increased weight as measured relative to total body weight (Fig. 4A : wild-type, 0.20 ± 0.00; ACE 8/wt, 0.23 ± 0.01; and ACE 1/8, 0.24 ± 0.01 mg/g body wt, n = 15–17, P < 0.05). We should note that no significant difference in body weight was noted between the three groups (wild-type, 26.2 ± 0.8; ACE 8/wt, 25.0 ± 0.9; and ACE 1/8, 24.0 ± 0.9 g). For the ventricular weight, both ACE 8/wt and ACE 1/8 mice showed a 10–13% increase of weight compared with wild-type mice (Fig. 4B: wild-type, 3.85 ± 0.07; and ACE, 1/8: 4.25 ± 0.10 mg/g body wt, n = 9, P < 0.01). We also noted that both ACE 1/8 and ACE 8/wt mice showed similar changes. Since ACE 8/wt mice have normal cardiac levels of angiotensin II (see Fig. 2), the increase of atrial and ventricular weight is unlikely to be due to differences in cardiac expression of angiotensin II.


Figure 4
View larger version (9K):
[in this window]
[in a new window]

  		Fig. 4. Atrial and ventricular weights. After collection, the atria (A) and ventricles (B) from WT, ACE 8/wt, and ACE 1/8 mice were separated under a dissecting microscope. Tissues were blotted dry and weighed. Tissue weight is plotted relative to total body weight (BW). Data for individual mice are shown, as well as for the group (means ± SE); n = at least 8 for each group. *P < 0.05, **P < 0.01, and ***P < 0.001 compared with WT.

 
Previous analysis of ACE2 knockout mice showed cardiac deterioration appearing at 6 mo of age (5). This was attributed, in part, to increased angiotensin II levels that were less than twofold those of wild-type mice. To determine whether increased angiotensin II levels in the ACE 1/8 heart may result in similar changes, we evaluated cardiac function in 6-mo-old ACE 1/8 mice using both echocardiography and LV catheterization. Echocardiography showed no difference in ventricular ejection fraction and LV wall thickness between the ACE 1/8 and wild-type mice (n = 7, data not shown). Six-month-old ACE 1/8 mice and wild-type controls were also studied using LV catheterization. Heart rate, LV dP/dt during both systole and diastole, as well as LV systolic and end-diastolic pressure were evaluated (Table 1). These showed no significant differences between the two groups of mice. Our data agree with data published by Gurley et al. (10) on another strain of ACE2 knockout mice that did not show cardiac dysfunction, even in older mice.


View this table:
[in this window]
[in a new window]

  		Table 1. LV function in 6-mo-old mice by ventricular catheterization

 
ECG and connexin 43 in ACE 1/8 mice. In ACE 8/8 mice, we observed prominent ECG changes characterized by a severe reduction of QRS amplitude and various degrees of heart block. These changes were so consistent and obvious in the ACE 8/8 model that an adult ACE 8/8 animal could easily be identified based solely on ECG findings. To examine whether similar changes occurred in ACE 1/8 mice, surface ECG was recorded in these mice and wild-type controls while under anesthesia. Similar to the morphological appearance of the heart, no gross differences between mice were noted. However, the pooled data showed a moderate reduction of QRS amplitude in ACE 1/8 mice compared with wild-type controls (Fig. 5A). The degree of QRS amplitude reduction in ACE 1/8 mice was similar to the ACE 8/wt mice. Unlike ACE 8/8 mice, all animals examined here showed discernable P waves. The PR intervals were not different between ACE 1/8 and control mice (data not shown). ACE 8/8 mice showed significantly increased mortality starting at 3 wk of age such that by 10 wk, only ~64% of animals survived (28). In contrast, ACE 1/8 mice showed mortality rates equivalent to that of wild-type animals. Connexin 43 levels were studied by Western blot analysis in the ventricles of ACE 1/8 and wild-type mice (Fig. 5B). This study showed no differences between the two groups of mice and contrasts to our findings in ACE 8/8 animals, which demonstrated a marked reduction of connexin 43 levels (17).


Figure 5
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 5. A: ECG QRS amplitude. Surface ECG was recorded in anesthetized mice, and the QRS amplitude was measured for lead II. QRS amplitude was significantly decreased in the ACE 1/8 mice and the ACE 8/wt mice compared with WT mice. Data from individual mice are shown, as well as from the group (means ± SE); n = 12 for WT, n = 7 for ACE 8/wt, and n = 20 for ACE 1/8 mice. *P < 0.05 compared with WT. B: connexin (Cx)43 levels were measured in the ventricles of ACE 1/8 and WT mice by Western blot analysis and quantified by densitometry. Ventricular GAPDH levels were used as internal controls. Cx43 levels, as normalized with GAPDH, were not significantly different between ACE 1/8 and WT mice (n = 6 for WT and n = 5 for ACE 1/8 mice).

 
Cardiac abnormalities in ACE 8/8-angiotensinogen knockout mice. As discussed above, the measurement of a variety of parameters in ACE 1/8 mice showed either no difference from wild-type mice or differences that were much reduced from those noted in the ACE 8/8 model. For example, the gross cardiac abnormalities present in ACE 8/8 mice were not observed in ACE 1/8 mice. Because the cardiac angiotensin II levels in ACE 1/8 and 8/8 mice are similar, we questioned whether cardiac angiotensin II is a major factor contributing to the defects observed in the ACE 8/8 model. To examine this, we completely eliminated angiotensin II production in ACE 8/8 mice by creating a strain that was both ACE 8/8 and angiotensinogen null (angiotensinogen knockout). If angiotensin II was the major cause of atrial enlargement, these mice should have atria near the size of wild-type mice. However, even with all angiotensin II production genetically eliminated, these mice still developed marked atrial enlargement. Specifically, atrial weight in ACE 8/8-angiotensinogen knockout mice was 0.52 ± 0.05 (n = 4) versus 0.45 ± 0.04 mg/g body wt for ACE 8/8 mice that carried the wild-type alleles for angiotensinogen (n = 6). ECG of ACE 8/8-angiotensinogen knockout mice also showed a clear reduction of QRS amplitude and an absence of discernable P waves, similar to the original findings of the ACE 8/8 animals (data not shown). These results suggest that an increase of cardiac angiotensin II itself was not the major factor in producing the atrial enlargement and the electrical abnormalities previously documented in the ACE 8/8 model. However, we did observe that by eliminating angiotensinogen expression, the mortality of ACE 8/8 mice was significantly reduced (Fig. 6). It is unclear whether this protection was due to a reduction of cardiac angiotensin II levels or the hemodynamic changes associated with angiotensinogen elimination, since the blood pressure of angiotensinogen knockout mice is significantly reduced (angiotensinogen wild-type: 101.5 ± 1.9, n = 10; angiotensinogen heterozygous: 94.1 ± 6.9, n = 6; angiotensinogen knockout: 65.8 ± 2.2 mmHg, n = 12, P < 0.001). This is true regardless of ACE genotype since, among the 12 mice that were angiotensinogen knockouts, four were ACE 8/8, three were ACE wild-type, and five were ACE 8/wt; all mice had an equivalently low blood pressure. Also, the ventricular weight of ACE 8/8-angiotensinogen knockout mice was 3.10 ± 0/18 (n = 4) versus 4.31 ± 0.09 mg/g body wt for wild-type mice (n = 6, P < 0.001), probably due to the lower blood pressures. With the use of Kaplan-Meier analysis, there was no significant difference in survival between ACE 8/8 mice that carries the wild-type and heterozygous angiotensinogen alleles.


Figure 6
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 6. Kaplan-Meier plot of survival. The survival of ACE 8/8 mice having a WT, heterozygous (HZ), or knockout (KO) genotype for the angiotensinogen gene was plotted using the Kaplan-Meier method. The number of mice was 29, 39, and 14 for WT, HZ, and KO, respectively. Removal of functional angiotensinogen genes showed a protective effect on the incidence of early death noted in ACE 8/8 mice with no ACE 8/8-angiotensinogen double knockout mice dying during the duration of the study (P < 0.01 using Kaplan-Meier analysis). P > 0.05 in survival between angiotensinogen WT and HZ mice using Kaplan-Meier analysis.

 
Cardiac hypertrophy in ACE 1/8 mice after aortic banding. ACE 1/8 mice do not have the significant increase of mortality observed in the ACE 8/8 model. Furthermore, ACE 1/8 mice produce increased levels of cardiac angiotensin II but without a systemic change of blood pressure. Thus these mice may be an ideal model to study the effects of increased local cardiac angiotensin II in cardiac pathology. Increased local cardiac angiotensin II has often been speculated to play an important role in cardiac hypertrophy. Thus, to investigate cardiac hypertrophy in ACE 1/8 mice, we used a transverse aortic banding model. Both ACE 1/8 and wild-type mice were banded at the level of the transverse aorta, between the right brachiocephalic trunk and the left carotid artery. Separate groups of mice were sham-operated without aortic banding. Right carotid pressure and heart weight were measured 4 wk after surgery. Systolic pressure in the right carotid artery was equivalent to LV systolic pressure and was measured using a Millar probe. Both the ACE 1/8 and wild-type mice showed a significant increase of right carotid pressure after banding (Fig. 7A). Also, hearts from both groups of mice developed significant cardiac hypertrophy after banding (Fig. 7B). However, the degree of hypertrophy measured by heart weight was not significantly different between the two banded groups, suggesting that in this model, the local production of angiotensin II was not a major determining factor in the development of cardiac hypertrophy.


Figure 7
View larger version (10K):
[in this window]
[in a new window]

  		Fig. 7. Cardiac hypertrophy by aortic banding. A Millar probe was used to measure right carotid systolic pressures in WT and ACE 1/8 mice treated with aortic banding (band) or with a sham operation (sham) (A). An elevated right carotid pressure indicates increased left ventricular systolic pressure after successful transverse aortic banding. B: heart weights, normalized to BW, are shown. Data for individual mice are shown, as well as for the group (means ± SE). A and B: P > 0.05 comparing WT-banded mice to ACE 1/8-banded mice by ANOVA after the Tukey post hoc test (n = at least 6 for each group).

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
A locally functional RAS in cardiac tissue has drawn considerable interest and discussion. It was reported that the effect of ACE inhibitors in reducing cardiac hypertrophy in some hypertensive rat models correlated better with the reduction of local angiotensin II levels than the reduction of systemic blood pressure or plasma angiotensin II (20). In addition, some investigators have published that the infusion of a subpressor dose of angiotensin II induced cardiac hypertrophy without blood pressure changes (18, 22). In vitro studies also suggested that angiotensin II directly stimulated cardiomyocyte growth (9, 24, 25). Several transgenic mouse models have been developed to examine this question through the overexpression of individual components of the RAS within the heart (13–15, 19, 21, 26, 27). As previously reviewed, these studies were somewhat contradictory, with some animal models developing ventricular hypertrophy or fibrosis, whereas others showed no apparent pathology (23). In all these transgenic models, the RAS was overexpressed in the heart on top of the intrinsic RAS of the animal.

In many ways the mice termed ACE 8/wt are similar to some of these previous models. Here, cardiac ACE expression is increased about 15-fold (the same level as in ACE 1/8), but these mice still express endothelial and renal ACE. ACE 8/wt mice demonstrate that even a very marked increase of ACE levels in the heart, when not associated with other changes in the mouse, does not result in increased cardiac levels of angiotensin II.

To increase cardiac angiotensin II, we created ACE 8/8 mice. By replacing the ACE promoter with an {alpha}-MHC promoter, ACE was overexpressed in the heart but on a background of reduced ACE expression in other tissues. The result was cardiac defects, including atrial enlargement, atrial ventricular conduction block, and sudden death. Although these effects undoubtedly resulted from the local elevation of ACE expression in the heart, we were not certain of the biochemical abnormality downstream of ACE that directly mediated this phenotype. The current model, ACE 1/8, shares many characteristics with ACE 8/8 mice; both models overexpress ACE in the heart and lack ACE expression in other tissues. Given the nearly exclusive expression of ACE in the heart, it is not surprising that both models present with elevated cardiac angiotensin II, and, in fact, the level of this peptide appears similar in the hearts of both ACE 8/8 and ACE 1/8 mice. The major difference is that cardiac ACE is reduced by about half in ACE 1/8 compared with ACE 8/8 mice.

Given the comparable cardiac angiotensin II levels in ACE 1/8 and ACE 8/8 hearts, it is remarkable that the ACE 8/8 cardiac phenotype was not reproduced in the ACE 1/8 animals. This suggested that increased cardiac angiotensin II was not the major cause of the ACE 8/8 phenotype. This idea was further tested by creating ACE 8/8 mice that also genetically lacked the ability to produce angiotensinogen (angiotensinogen knockout mice). Mice that were both ACE 8/8 and angiotensinogen null maintained a phenotype similar to the ACE 8/8 mice, with atrial enlargement and a markedly abnormal ECG. Thus it seems safe to conclude that a majority of the cardiac abnormalities observed in the ACE 8/8 mice were not due to an overproduction of cardiac angiotensin II. This is different from what we originally proposed based on the information first available with the ACE 8/8 mice (28). Whatever induces the phenotype in ACE 8/8 mice has to be explained in the setting of a model that contains roughly 25-fold normal quantities of cardiac ACE but is not seen in a model (ACE 1/8) in which there is 15-fold overexpression of ACE. ACE is a somewhat nonspecific peptidase and can hydrolyze several peptides besides angiotensin I. Thus we cannot formally eliminate the possibility that some other peptide product of ACE, apart from angiotensin II (and bradykinin), is responsible for the pathology observed in the ACE 8/8 model. However, another possibility concerns the physical presence of the ACE protein within the cardiomyocyte membrane. It may be that the increased cardiac ACE noted in ACE 8/8 mice may physically disrupt the cardiomyocyte membrane in a fashion that is significantly more pathological than occurs with the cardiac ACE levels noted in ACE 1/8 mice. Large amounts of transgenic protein expression driven by the {alpha}-MHC promoter have occasionally been reported to generate artifacts. For example, the cardiac overexpressions of nonmammalian proteins such as Cre recombinase, green fluorescent protein, and the yeast transcriptional activator Gal 4 have all caused cardiac abnormalities, including dilated cardiomyopathy (2, 11, 16). Interestingly, the ACE inhibitor captopril improved survival of the Cre recombinase overexpressing mice, even though RAS activation was not indicated in this model (2). Probably, this is due to the effect of captopril on blood pressure. This phenomenon makes it hard to assess whether increased ACE activity or the amount of expressed ACE protein directly caused the ACE 8/8 phenotype.

In many ways, the ACE 1/8 model has exactly the phenotype that we hoped to achieve in creating the ACE 8/8 mice. Our original idea was to use targeted homologous recombination to focus ACE expression in the heart and thus force this organ to become the major locus for the generation of angiotensin II. The ACE 1/8 model, a model with normal blood pressure, normal cardiac histology, and near normal basal cardiac physiology, was the ideal situation in which to examine the role of elevated cardiac angiotensin II levels in cardiac pathology. This is an important question since many investigators continue to debate whether the major role of ACE inhibitors is to reduce blood pressure and cardiac afterload or to reduce local cardiac formation of angiotensin II. Although our studies are not definitive, they provide evidence supporting the primacy of blood pressure reduction in explaining the effects of ACE inhibition. Our data show that cardiac angiotensin II levels, double those of wild-type mice, did not induce cardiac fibrosis or basal cardiac dysfunction. Also, a model of cardiac hypertrophy, induced by aortic banding, showed no difference between the ACE 1/8 model and wild-type mice. As recently reviewed by Reudelhuber et al. (23), these data are consistent with other experimental findings challenging the hypothesis that local angiotensin II levels are deleterious to cardiac function. Thus the ACE 1/8 model suggests a clear separation between local cardiac angiotensin II production and a systematic overload of angiotensin II. This supports the idea that cardiac hypertrophy is much more dependent on hemodynamic changes than on local angiotensin II levels.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants R01-DK-039777 and R01-DK-051445; American Heart Association (AHA) Scientist Development Grant 0530136N (to H. D. Xiao); AHA Beginning Grant-in-Aid 0665176B0 (to S. Fuchs); National Heart, Lung, and Blood Institute Pathway to Independence Award K99-HL-088000 (to S. Fuchs); and National Health and Medical Research Council of Australia Senior Research Fellow Grant 395508 (to D. J. Campbell).


   ACKNOWLEDGMENTS
 
We thank Dr. Jooyoung Julia Shin (Albert Einstein College of Medicine) for providing cardiac echocardiography measurements and Seth Lesch for help with blood pressure measurements and mouse genotyping.


   FOOTNOTES
 

Address for reprint requests and other correspondence: H. D. Xiao, 101 Woodruff Cir., Rm. 7006, Emory Univ., Dept. of Pathology, Atlanta, GA 30322 (e-mail: hxiao2{at}emory.edu)

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.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Bader M. Role of the local renin-angiotensin system in cardiac damage: a mini-review focusing on transgenic animal models. J Mol Cell Cardiol 34: 1455–1462, 2002.[CrossRef][Web of Science][Medline]
  2. Buerger A, Rozhitskaya O, Sherwood MC, Dorfman AL, Bisping E, Abel ED, Pu WT, Izumo S, Jay PY. Dilated cardiomyopathy resulting from high-level myocardial expression of Cre-recombinase. J Card Fail 12: 392–398, 2006.[CrossRef][Web of Science][Medline]
  3. Cole JM, Khokhlova N, Sutliff RL, Adams JW, Disher KM, Zhao H, Capecchi MR, Corvol P, Bernstein KE. Mice lacking endothelial ACE: normal blood pressure with elevated angiotensin II. Hypertension 41: 313–321, 2003.[Abstract/Free Full Text]
  4. Corvol P, Williams TA, Soubrier F. Peptidyl dipeptidase A: angiotensin I-converting enzyme. Methods Enzymol 248: 283–305, 1995.[Web of Science][Medline]
  5. Crackower MA, Sarao R, Oudit GY, Yagil C, Kozieradzki I, Scanga SE, Oliveira-dos-Santos AJ, da Costa J, Zhang L, Pei Y, Scholey J, Ferrario CM, Manoukian AS, Chappell MC, Backx PH, Yagil Y, Penninger JM. Angiotensin-converting enzyme 2 is an essential regulator of heart function. Nature 417: 822–828, 2002.[CrossRef][Medline]
  6. Dahlöf B. Effect of angiotensin II blockade on cardiac hypertrophy and remodelling: a review. J Hum Hypertens 9, Suppl 5: S37–S44, 1995.
  7. Esther CR Jr, Howard TE, Marino EM, Goddard JM, Capecchi MR, Bernstein KE. Mice lacking angiotensin-converting enzyme have low blood pressure, renal pathology, and reduced male fertility. Lab Invest 74: 953–965, 1996.[Web of Science][Medline]
  8. Garg R, Yusuf S. Overview of randomized trials of angiotensin-converting enzyme inhibitors on mortality and morbidity in patients with heart failure: Collaborative Group on ACE Inhibitor Trials. JAMA 273: 1450–1456, 1995.[Abstract/Free Full Text]
  9. Geisterfer AA, Peach MJ, Owens GK. Angiotensin II induces hypertrophy, not hyperplasia, of cultured rat aortic smooth muscle cells. Circ Res 62: 749–756, 1988.[Abstract/Free Full Text]
  10. Gurley SB, Allred A, Le TH, Griffiths R, Mao L, Philip N, Haystead TA, Donoghue M, Breitbart RE, Acton SL, Rockman HA, Coffman TM. Altered blood pressure responses and normal cardiac phenotype in ACE2-null mice. J Clin Invest 116: 2218–2225, 2006.[CrossRef][Web of Science][Medline]
  11. Habets PE, Clout DE, Lekanne Deprez RH, van Roon MA, Moorman AF, Christoffels VM. Cardiac expression of Gal4 causes cardiomyopathy in a dose-dependent manner. J Muscle Res Cell Motil 24: 205–209, 2003.[CrossRef][Web of Science][Medline]
  12. Hasslacher C, Kempe HP, Bostedt-Kiesel A. ACE inhibitors and diabetic nephropathy: clinical and experimental findings. Clin Investig 71, Suppl 5: S20–S24, 1993.[Web of Science][Medline]
  13. Hein L, Stevens ME, Barsh GS, Pratt RE, Kobilka BK, Dzau VJ. Overexpression of angiotensin AT1 receptor transgene in the mouse myocardium produces a lethal phenotype associated with myocyte hyperplasia and heart block. Proc Natl Acad Sci USA 94: 6391–6396, 1997.[Abstract/Free Full Text]
  14. Higaki J, Aoki M, Morishita R, Kida I, Taniyama Y, Tomita N, Yamamoto K, Moriguchi A, Kaneda Y, Ogihara T. In vivo evidence of the importance of cardiac angiotensin-converting enzyme in the pathogenesis of cardiac hypertrophy. Arterioscler Thromb Vasc Biol 20: 428–434, 2000.[Abstract/Free Full Text]
  15. Hoffmann S, Krause T, van Geel PP, Willenbrock R, Pagel I, Pinto YM, Buikema H, van Gilst WH, Lindschau C, Paul M, Inagami T, Ganten D, Urata H. Overexpression of the human angiotensin II type 1 receptor in the rat heart augments load induced cardiac hypertrophy. J Mol Med 79: 601–608, 2001.[CrossRef][Web of Science][Medline]
  16. Huang WY, Aramburu J, Douglas PS, Izumo S. Transgenic expression of green fluorescence protein can cause dilated cardiomyopathy. Nat Med 6: 482–483, 2000.[CrossRef][Web of Science][Medline]
  17. Kasi VS, Xiao HD, Shang LL, Iravanian S, Langberg J, Witham EA, Jiao Z, Gallego CJ, Bernstein KE, Dudley SC. Cardiac restricted angiotensin converting enzyme overexpression causes conduction defects and connexin dysregulation. Am J Physiol Heart Circ Physiol 293: H182–H192, 2007.[Abstract/Free Full Text]
  18. Kim S, Ohta K, Hamaguchi A, Yukimura T, Miura K, Iwao H. Angiotensin II induces cardiac phenotypic modulation and remodeling in vivo in rats. Hypertension 25: 1252–1259, 1995.[Abstract/Free Full Text]
  19. Mazzolai L, Nussberger J, Aubert JF, Brunner DB, Gabbiani G, Brunner HR, Pedrazzini T. Blood pressure-independent cardiac hypertrophy induced by locally activated renin-angiotensin system. Hypertension 31: 1324–1330, 1998.[Abstract/Free Full Text]
  20. Nagano M, Higaki J, Mikami H, Nakamaru M, Higashimori K, Katahira K, Tabuchi Y, Moriguchi A, Nakamura F, Ogihara T. Converting enzyme inhibitors regressed cardiac hypertrophy and reduced tissue angiotensin II in spontaneously hypertensive rats. J Hypertens 9: 595–599, 1991.[CrossRef][Web of Science][Medline]
  21. Paradis P, Dali-Youcef N, Paradis FW, Thibault G, Nemer M. Overexpression of angiotensin II type I receptor in cardiomyocytes induces cardiac hypertrophy and remodeling. Proc Natl Acad Sci USA 97: 931–936, 2000.[Abstract/Free Full Text]
  22. Pillai JB, Gupta M, Rajamohan SB, Lang R, Raman J, Gupta MP. Poly(ADP-ribose) polymerase-1-deficient mice are protected from angiotensin II-induced cardiac hypertrophy. Am J Physiol Heart Circ Physiol 291: H1545–H1553, 2006.[Abstract/Free Full Text]
  23. Reudelhuber TL, Bernstein KE, Delafontaine P. Is angiotensin II a direct mediator of left ventricular hypertrophy? Hypertension 49: 1196–1201, 2007.[Free Full Text]
  24. Ritchie RH, Schiebinger RJ, LaPointe MC, Marsh JD. Angiotensin II-induced hypertrophy of adult rat cardiomyocytes is blocked by nitric oxide. Am J Physiol Heart Circ Physiol 275: H1370–H1374, 1998.[Abstract/Free Full Text]
  25. Sadoshima J, Izumo S. Molecular characterization of angiotensin II-induced hypertrophy of cardiac myocytes and hyperplasia of cardiac fibroblasts. Critical role of the AT1 receptor subtype. Circ Res 73: 413–423, 1993.[Abstract/Free Full Text]
  26. Tian XL, Pinto YM, Costerousse O, Franz WM, Lippoldt A, Hoffmann S, Unger T, Paul M. Over-expression of angiotensin converting enzyme-1 augments cardiac hypertrophy in transgenic rats. Hum Mol Genet 13: 1441–1450, 2004.[Abstract/Free Full Text]
  27. Van Kats JP, Methot D, Paradis P, Silversides DW, Reudelhuber TL. Use of a biological peptide pump to study chronic peptide hormone action in transgenic mice: direct and indirect effects of angiotensin II on the heart. J Biol Chem 276: 44012–44017, 2001.[Abstract/Free Full Text]
  28. Xiao HD, Fuchs S, Campbell DJ, Lewis W, Dudley SC Jr, Kasi VS, Hoit BD, Keshelava G, Zhao H, Capecchi MR, Bernstein KE. Mice with cardiac-restricted angiotensin-converting enzyme (ACE) have atrial enlargement, cardiac arrhythmia, and sudden death. Am J Pathol 165: 1019–1032, 2004.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Am. J. Physiol. Regul. Integr. Comp. Physiol.Home page
V. G. Barauna, F. C. Magalhaes, J. E. Krieger, and E. M. Oliveira
AT1 receptor participates in the cardiac hypertrophy induced by resistance training in rats
Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2008; 295(2): R381 - R387.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
294/2/H659    most recent
01147.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Xiao, H. D.
Right arrow Articles by Bernstein, K. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xiao, H. D.
Right arrow Articles by Bernstein, K. E.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2008 by the American Physiological Society.