Sex dependence and temporal dependence of the left ventricular genomic response to pressure overload
Ellen O. Weinberg*,,
Maria Mirotsou*,,
Joseph Gannon,
Victor J. Dzau,
Richard T. Lee and
Richard E. Pratt
Cardiovascular Research, Department of Medicine, Brigham and Womens Hospital, Harvard Medical School, Boston, Massachusetts 02115
 |
ABSTRACT
|
---|
To characterize responses of the left ventricle (LV) to pressure overload at the genomic level, we performed high-density microarray analysis on individual mouse LVs. Male and female mice underwent transverse aortic constriction. At 1 day and 30 wk, the LV free wall was harvested and RNA isolated from 27 individual ventricles was analyzed on Mu74Av2 GeneChips, which contain
12,483 distinct genes. Interestingly, a greater number of genes was regulated in response to acute overload than in response to chronic overload. Hierarchical cluster analysis revealed the presence of several distinct expression profiles. Of these clusters, the majority contained genes that were regulated either in response to acute overload or both acute and chronic overload. In addition, clusters revealing sex-specific responses to overload were detected. In summary, the acute and chronic genomic responses to pressure overload are distinct. Moreover, sex modifies these responses. Furthermore, these studies have uncovered several novel and potentially important genes that are regulated in response to overload and may open unrecognized avenues for further functional analysis.
genomics; hypertrophy; myocyte; gene expression; pressure overload; sex differences
 |
INTRODUCTION
|
---|
THE HEART RESPONDS TO AN INCREASE in load, such as that produced by aortic constriction or hypertension, through cardiac hypertrophy (14). Several genes that are associated with the hypertrophic response have been identified (12). Despite this, heart failure secondary to cardiac hypertrophy remains a significant clinical problem, suggesting that the major pathways involved in the process remain to be elucidated. Furthermore, studies have documented sex differences in the adaptive response of the heart to load (8, 23, 15). The molecular basis for and the extent of these differences across a large proportion of the genome remains incompletely defined.
Cardiac myocytes respond to acute biomechanical load within minutes, and a variety of signaling mechanisms can be biomechanically activated (27, 28). On this basis, we sought to characterize the acute response of the LV to pressure overload on a genomic level. We speculated whether these acutely regulated genes persist, or whether an additional set of genes is involved in the maintenance of chronic pressure overload hypertrophy. To this aim, we used the well-established transverse aortic constriction model of pressure overload hypertrophy in mice (9). We performed DNA microarray analysis on individual LV samples in each group to obtain statistically based insight into gene expression. Furthermore, we compared gene expression in male and female animals to yield additional information on genome-wide sex-dimorphic responses to pressure overload. These results demonstrate temporal- and sex-dependent components of the genomic response to pressure overload that open unrecognized avenues for further functional analysis.
 |
MATERIALS AND METHODS
|
---|
Animal surgery.
All animal procedures were approved by the Harvard Standing Committee on Animals. Male and female FVB mice (Jackson Laboratories) underwent transverse aortic constriction (9) at 8 wk of age under ketamine and xylazine anesthesia. Additional mice underwent the identical surgical procedure without placement of the aortic band to serve as age- and sex-matched sham-operated controls. Mice were randomized to acute (1 day after banding) and chronic (30 wk after banding) pressure overload or sham groups.
Echocardiography.
To assess left ventricular (LV) mass and geometry (17), echocardiographic analysis was performed with spontaneous respiration under light anesthesia using intraperitoneal pentobarbital titrated to maintain a heart rate >400 beats/min. Short-axis two-dimensional images using an 812 MHz transducer placed at the midpapillary levels of the LV were stored as digital loops.
RNA extraction.
The heart was removed, and left and right ventricle were dissected and weighed. The free wall apical portion of the LV was isolated and frozen in liquid nitrogen for RNA extraction. RNA was extracted using TriReagent (Sigma). RNA concentration was determined by optical density at 260 nm, and quality was confirmed by denaturing agarose gel electrophoresis and visualization of distinct 28S and 18S ribosomal RNA. Eight micrograms of total RNA from individual hearts was used for microarray analysis.
A total of 27 LV samples were used for microarray analysis. Each of the 27 LV samples was hybridized as an individual sample to a single microarray chip. LV samples from male and female mice that underwent transverse aorta constriction (banded) or sham operation (sham) were collected 1 day after surgery (acute pressure overload) [male sham (MS), n = 2; male banded (MB), n = 4; female sham (FS), n = 3; female banded (FB), n = 3] and 30 wk after surgery to represent chronic compensatory LV hypertrophy (MS, n = 6; MB, n = 3; FS, n = 3; and FB, n = 3).
Oligonucleotide arrays.
Affymetrix Oligonucleotide Murine Genome U74A Array version 2 (MG-U74Av2) array chips, which contain
12,483 known genes and expressed sequence tags (ESTs), were used in these studies. Each gene or EST is represented on the chip by 16 probe pairs. Each probe pair consists of a 25-bp oligonucleotide sequence that is a perfect match to the gene of interest and a 25-bp oligonucleotide containing a single base mismatch at bp 13. Affymetrix GeneChip analysis was performed according to recommended procedures. Briefly, total RNA was reverse-transcribed to double-stranded cDNA (Life Technologies), and biotinylated cRNA was generated by an in vitro transcription reaction (Enzo). Labeled cRNA was purified (Qiagen), fragmented by alkaline treatment, and hybridized to a GeneChip array overnight at 45°C. The array was washed, stained with streptavidin-phycoerythrin, and scanned. Each array was scanned twice, and an average intensity for each probe pair was generated. Data were first analyzed using Affymetrix Microarray Suite (MAS 4.01) to assess quality of RNA and hybridization. Further analysis was performed using dChip (Wong, http://www.biostat.harvard.edu/complab/) (13) and GeneSpring software (Silicon Genetics).
dChip was used to evaluate performance of probe pairs. dChip uses a probe sensitivity index to describe the performance of each probe pair and estimates a model-based probe expression index to determine possible cross-hybridizing probes and other artifacts. In addition, it provides standard errors for the model-based expression as a measurement of accuracy. These standard errors are also used to estimate confidence intervals of the fold changes and for clustering. A running median smooth line method is applied for normalization between different individual arrays.
Data analysis.
All genes identified as "absent" in all 27 arrays were eliminated. To estimate fold changes, all negative values were brought to 0. The genes that were differentially expressed between each experimental group (banded, sham, acute, chronic, male, female) were identified by querying for transcripts that were present in at least one of the samples compared and had a 90% probability of changing at least 1.2-fold (90% lower bound cut off). Hierarchical clustering was performed after three additional filtering criteria were applied. 1) A gene had to be present in at least 2 of the 27 samples. 2) The ratio of the standard deviation to the mean of expression of a gene across all samples was greater than 0.3 and less than 10; this allowed us to exclude genes that showed little variation across samples. 3) The variation within replicate array was greater than 0 and less than 0.5; this filtering excluded genes that had inconsistent expression values within replicate arrays.
Before clustering, the expression values across all samples were standardized to a mean of 0 and standard deviation of 1. These standardized values were used to estimate correlations between the samples and to define the clustering nodes. For initial clustering, each sample was treated individually to obtain sample distribution. Subsequently, replicate arrays from each experimental group were pooled. To assess the reliability of the clusters generated, each cluster was resampled 3040 times. A gray-scale bar, with black representing genes that belonged to the same cluster in all resamplings and white representing genes that were irreproducible upon resampling, accompanies each cluster.
Real-time RT-PCR amplification.
Real-time RT-PCR reactions were carried out in iCycler IQ Real-Time Detection Systems (Bio-Rad). SuperScript One-step RT-PCR with Platinum Taq kits (Invitrogen) were used for all real-time RT-PCR amplification in a total volume of 50 µl, which contained 200 ng total RNA, 5 mM MgSO4, 500 nM forward and reverse primers, and 200 nM fluorogenic probe. Real-time RT-PCR amplification for each RNA sample was performed in triplicate. For the actin-
1 and four-and-a-half domain genes, one RNA sample of each group was assessed with this method. For the osteoblast-specific factor and orosomucoid 1 genes, two independent RNA samples from each group were used for validation. One no-RT (without reverse transcriptase) control for each RNA sample and one no-RNA (substitute RNA with distilled H2O) control for each primer and probe set were also performed. The one-step real-time RT-PCR condition is as follows: 15 min at 50°C and 5 min at 95°C, followed by a total of 45 two-temperature cycles (15 s at 95°C and 1 min at 60°C). Relative gene expression analysis is carried out using the standard curve method (11, 26).
Northern blotting.
Five micrograms of total RNA from individual LV samples was size-fractionated by electrophoresis in 1.2% agarose-formaldehyde gel and transferred to a nitrocellulose membrane (Schleicher and Schuell). The membrane was prehybridized for 10 min and hybridized with specific probes for 1 h in QuikHyb solution (Stratagene) at 68°C. After hybridization, the membrane was washed and exposed to autoradiography film. The membrane was stripped (0.1x SSC, 0.1% SDS) for sequential hybridization. Stripping was verified by autoradiography. Probes were generated by PCR as full-length cDNAs for each specific gene. The cDNA fragments were radiolabeled with [
-32P]deoxycytidine triphosphate (New England Nuclear/Dupont) using random priming.
 |
RESULTS
|
---|
Body weights were similar in mice randomized to acute vs. chronic banding for each sex at the time of aortic banding.
Characteristics of mice 30 wk after aortic banding (Table 1).
Thirty weeks after aortic banding, body weights were similar between female banded and female sham mice and between male banded and male sham mice, but were significantly greater in male mice compared with female mice (two-factor ANOVA P = 0.0001). Transverse aortic constriction for 30 wk resulted in significant increases in LV weight in both banded females (t-test P = 0.01) and males (t-test P = 0.002) compared with their sex-matched sham controls. The increase in LV weight in female banded mice was similar to the increase in male banded mice [two-factor interaction term P = not significant (NS)]. Similar effects were observed for LV wet weight/body weight (LV wet wt/BW). Fractional shortening, a measure of systolic function, was similar among all groups, indicating that systolic function was preserved at this stage of compensatory hypertrophy. End diastolic dimension (EDD), a measure of LV cavity dimension relative to LV wall thickness, was slightly but not significantly increased in female or male banded mice vs. their sex-matched sham controls as analyzed by t-test. According to two-factor ANOVA, EDD was significantly increased in banded groups (P = 0.01), but there was no significant effect of sex, and the interaction of sex and banding was not significant. Anterior LV wall thickness (AWT) was significantly increased in female banded mice compared with female sham mice (t-test P = 0.01), and there was a strong trend for an increase in AWT in male banded vs. male sham mice (t-test P = 0.06). By two-factor ANOVA, there was a significant effect of banding on AWT (P = 0.006), but no significant effect of sex or interaction (P = NS). Posterior LV wall thickness (PWT) was significantly increased in both female and male banded mice compared with their sex-matched sham mice (t-test, both P = 0.02). By two-factor ANOVA, there was no significant effect of sex on PWT, a highly significant effect of banding (0.0003), and a significant interaction between sex and banding on PWT (P = 0.048). This suggests that the magnitude of the increase in posterior wall thickness in response to aortic banding was greater in female compared with male mice, although the magnitude of hypertrophy was similar in males and females. LV weight calculated by echocardiography (Echo LV weight) was significantly increased in female banded vs. female sham mice (t-test P = 0.005) and in male banded vs. male sham mice (t-test P = 0.02). Two-factor ANOVA revealed a significant effect of banding on Echo LV weight (P = 0.0001), but no significant effect of sex and no significant interaction (P = NS). The correlation between post mortem LV weight and echocardiography calculated LV weight was 0.911.
In summary, these results demonstrate significant LV hypertrophy due to transverse aortic constriction for 30 wk, which was similar in magnitude between male and female banded mice. LV systolic function (fractional shortening) was similar among all groups, demonstrating preservation in contractile function at this stage of compensatory hypertrophy. A subtle difference in geometric remodeling between male and female mice was revealed through the observation of a significantly greater magnitude of increase in posterior wall thickness in female compared with male banded mice.
Gene expression quality control.
The results from the Affymetrix arrays indicated consistently high quality. Out of 12,483 probes, 3,360 genes and ESTs were expressed per array (mean percentage = 27.1 ± 0.7). The percentage of single outliers (due to artifacts in a single array) or percentage of array outliers (due to cross hybridization) was below the 5% threshold that is considered a warning of sample contamination or hybridization problems (Table 2), (13). Median signal intensities were similar among individual arrays, indicating good reproducibility (mean median signal intensity = 87.5 ± 2.1, Table 2). Representative linear correlation matrices of transcripts expressed in at least one of two arrays for sham and banded samples at day 1 are shown in Fig. 1. Within-group comparisons (banded vs. banded) exhibited a tighter correlation compared with between-group comparisons (banded vs. control sham). These figures are representative of all inter- and intra-group comparisons (data not shown).

View larger version (47K):
[in this window]
[in a new window]
|
Fig. 1. Quality control for microarray analysis. Scatter correlation plots of gene expression in representative samples. Black dots indicate probes that were expressed in both arrays, whereas yellow dots represent probes that were expressed in only one of the two arrays or had marginal expression. Note that intra-group correlation (top) is tighter than the intergroup correlation. Similar scatter plots were observed in all other samples (data not shown).
|
|
Expression profiles in response to pressure overload (Table 3).
Initial analysis using the 1.2-fold, 90% lower bound cutoff identified the number of transcripts whose level of expression changed when compared among the experimental groups. When t-tests were applied, all transcripts passed the P < 0.1 cutoff criterion with
80% of those also passing the P < 0.05 cutoff criterion (data not shown).
The number of genes changing in response to acute pressure overload (male banded acute vs. male sham acute, 269; female banded acute vs. female sham acute, 110) was greater than the number of genes changing in response to chronic pressure overload (male banded chronic vs. male sham chronic, 53; female banded chronic vs. female sham chronic, 56). In both males and females, upregulated genes were predominant (95% in males and 84% in females) during acute pressure overload. Upregulated genes also predominated in response to chronic (30 wk) pressure overload, although this effect was less pronounced than in response to acute load, and the percentage was similar between males and females (60% in males and 59% in females). Venn diagrams (Fig. 2, A and B) demonstrate that the number of genes common to both acute and chronic samples (males, 10; females, 12) is small, indicating distinct, nonoverlapping expression profiles during acute vs. chronic overload. The known hypertrophic marker genes, such as brain natriuretic peptide and skeletal actin, were upregulated in both males and females in response to both acute and chronic pressure overload. Notably, all of the genes, with the exception of cyritestin, exhibited the same directional change (either up- or downregulation) during acute and chronic overload. Cyritestin, an ADAMs family member not previously reported as expressed in the heart (19), was upregulated following acute overload and downregulated following chronic overload.

View larger version (12K):
[in this window]
[in a new window]
|
Fig. 2. Comparisons of differentially regulated genes. Venn diagrams were produced comparing the number of genes altered in comparisons between banded and sham samples in response to acute or chronic pressure overload in males and females. A: genes differentially regulated between male banded mice (MB) vs. male sham mice (MS) acutely, chronically, or both. Note that only a small number of genes were regulated in response to both acute and chronic pressure overload in male mice. B: genes differentially regulated between female banded mice (FB) vs. female sham mice (FS) acutely, chronically, or both. Note that only a small number of genes were regulated in response to both acute and chronic pressure overload in female mice. C: genes differentially regulated between banded mice vs. sham mice in response to acute pressure overload in male mice, female mice, or both. Note that only a small number of genes were regulated in response to acute pressure overload in both sexes. D: genes differentially regulated between banded mice vs. sham mice chronically in males, females, or both. Note that only a small number of genes were regulated in response to chronic pressure overload in both sexes.
|
|
Effects of sex differences on expression profiles.
The acute response to pressure overload in males, as determined by the number of differentially regulated genes, was greater than that in females (269 vs. 110 genes); whereas by the same criterion the response to chronic pressure overload was similar between males and females (53 vs. 56). Upregulated genes predominated during acute and chronic overload in both sexes. Venn diagrams (Fig. 2, C and D) demonstrate that sex differences markedly influenced the response to pressure overload, with only 48 genes common to males and females during acute overload and 14 common to males and females during chronic overload. During chronic overload, atrial natriuretic peptide, brain natriuretic peptide, and connective tissue growth factor were among the genes coregulated in males and females. Interestingly, ninjurin, an extracellular matrix gene that maps to a chromosomal locus (9q22) associated with familial dilated cardiomyopathy (3), was increased in both males and females during chronic pressure overload.
Distinct gene clusters during acute vs. chronic pressure overload.
The 930 genes differentially regulated in at least one of the comparisons shown in Table 3 were subjected to further filtering (described in MATERIALS AND METHODS) and analyzed by hierarchical clustering. A total of 263 genes passed these filtering criteria. Initially, both genes and individual samples were clustered to reveal sample distribution. The 1 day (acute) and 30 wk (chronic) time points formed distinct clusters (Fig. 3). Furthermore, within both time points the individual aortic-banded samples were separated from the sham samples. These cluster distributions strongly supported the small intergroup variance of the physiological parameters of hypertrophy, as well as the high quality of the data, and validated our filtering criteria. This enabled us to pool replicate samples from each group, and in this manner genes with similar patterns of expression were identified.

View larger version (55K):
[in this window]
[in a new window]
|
Fig. 3. Hierarchical clustering of pressure overload gene expression data. Hierarchical clustering of genes and samples was performed. Note that the individual samples were clustered appropriately, i.e., acute samples clustered together and the chronic samples banded together. Moreover, within the acute and chronic samples, the sham animals and the banded animals were placed in distinct clusters. Red indicates high relative level of expression, and green represents low relative level of expression. The color bar at bottom represents calibrated levels of expression.
|
|
Cluster analysis revealed that the two largest clusters consisted of genes upregulated in response to acute and chronic LV pressure overload. The cluster shown in Fig. 4C comprised 77 genes that were upregulated only during acute pressure overload in both males and females. The cluster shown in Fig. 4B comprised 53 genes that were upregulated during acute pressure overload and further upregulated during chronic pressure overload in both males and females. Interestingly, no cluster was found to contain genes that were upregulated during chronic, but not acute pressure overload, although individual genes displayed this pattern (data not shown). The genes forming the two identified unique clusters are grouped according to function and are shown in Tables 4 and 5. According to function, immunity/inflammation/stress response genes, extracellular matrix, and cytoskeletal genes are well-represented during the acute response, and transcription regulation, extracellular matrix, and cytoskeletal genes are well-represented in the cluster of genes upregulated following acute overload that remain elevated in chronic overload. Twenty-nine unassigned genes were upregulated during acute overload and 10 unassigned genes were upregulated during acute overload that remained elevated during chronic overload.

View larger version (41K):
[in this window]
[in a new window]
|
Fig. 4. Distinct clusters of response to acute or chronic pressure overload. Hierarchical clustering was performed on the average signal within each group. A: hierarchical clustering of gene expression data in pooled replicate arrays. B: magnified cluster and average expression plots representing genes upregulated in response to both acute and chronic pressure overload. C: magnified cluster and average expression plots representing genes upregulated in response to acute but not chronic pressure overload. MB, male banded; MS, male sham; FB, female banded; FS, female sham. Red indicates high relative level of expression, and green represents low relative level of expression. Each cluster is accompanied by a gray-scale bar, with black representing genes that always belong to the cluster and white representing genes that do not belong to the cluster upon 40 resamplings.
|
|
Sex dimorphic responses to pressure overload.
Sex dimorphic gene clusters were identified and are shown in Fig. 5 and Table 6. Figure 5A shows a cluster of genes that was upregulated in males but downregulated in females during acute pressure overload. Figure 5B shows a cluster of genes that was upregulated in females and not changed in males during acute pressure overload. Figure 5C shows a cluster of genes that was upregulated in males and not changed in females during acute pressure overload. Immunity/inflammation/stress response genes are strongly represented in these clusters and appear to be sex dimorphic. In addition, nine unassigned genes were identified as responsive to pressure overload in a sex-dimorphic manner.

View larger version (48K):
[in this window]
[in a new window]
|
Fig. 5. Sex-specific differences in the response to pressure overload. As in Fig. 4, hierarchical clustering was performed on the average signal within each group. The resultant plot was examined for genes exhibiting sex-related patterns. A: genes upregulated in male mice but downregulated in female mice in response to acute pressure overload. B: genes upregulated only in female mice in response to acute pressure overload. C: genes that are upregulated only in male mice in response to acute pressure.
|
|
Validation by Northern blotting and real-time RT-PCR.
Validation of our microarray data was performed on nine genes. Northern blotting (Fig. 6A) demonstrated that lipocalin and serine protease inhibitor-2 were increased in response to acute LV pressure overload, and cardiac ankyrin repeat protein and connective tissue growth factor (FISP-2) were increased in response to both acute and chronic pressure overload in agreement with the microarray data. Real-time RT-PCR demonstrated that four-and-a-half LIM domains was upregulated in response to acute pressure overload, and skeletal
-actin and osteoblast-specific factor-2 were upregulated in response to both acute and chronic LV pressure overload, in agreement with the microarray data. Orosomucoid 1 was increased in males and decreased in females in response to acute pressure overload, as was observed by microarray analysis. The fold changes observed by real-time RT-PCR were similar in magnitude and direction to the fold changes observed by microarray analysis (Fig. 6B). One additional gene, Down syndrome critical region-1, was validated by our group to be increased in chronic LV pressure overload in vivo and was recently published (24).

View larger version (42K):
[in this window]
[in a new window]
|
Fig. 6. Validation by Northern blotting and real-time RT-PCR. A: validation of lipocalin-2 and serine protease inhibitor (upregulated in response to acute pressure overload) and cardiac ankyrin repeat protein (CARP) and FISP-12 (connective tissue growth factor) (upregulated in response to acute and chronic pressure overload) by Northern blotting. B: validation of four-and-a-half LIM domains (upregulated in response to acute pressure overload), skeletal -actin and osteoblast-specific factor 2 (upregulated in response to acute and chronic pressure overload), and orosomucoid (acutely increased in males; decreased in females) by real-time RT-PCR. *Real-time RT-PCR was conducted in duplicate for two independent RNA samples from each group. The average fold changes are given. For the remaining samples, real-time RT-PCR was conducted in triplicate from one RNA sample from each group.
|
|
 |
DISCUSSION
|
---|
We employed DNA microarray hybridization to evaluate the patterns of expression that occur in response to acute and chronic cardiac pressure overload in vivo. Transverse aortic constriction produced pressure overload as a stimulus for the development of concentric hypertrophy resembling the clinical scenario observed in patients with hypertension or aortic stenosis. In this study, we used multiple samples per group and individual hearts were sampled on arrays. The quality of hybridization was confirmed when experimental analysis demonstrated that the percentages of expressed genes were similar among the arrays and the number of probe outliers due to cross-hybridization and artifacts extremely low. Importantly, following filtering for cluster analysis, samples fell into distribution according to time point and experimental procedure (aortic banded vs. sham).
The microarray data presented in the present study were validated for nine genes using alternate techniques. Using Northern blotting, we detected a similar increase in DSCR1 in response to acute and chronic pressure overload, and this finding was recently published by our group (24). Original data for the validation of lipocalin-2, serine protease inhibitor-2, cardiac ankyrin repeat protein, connective tissue growth factor, four-and-a-half LIM domains, skeletal
-actin, osteoblast-specific factor 2, and orosomucoid 1 are presented in the present study. Thus the consistency of microarray results to the validation results obtained using alternate techniques in this study as well as to data in published literature provide confidence in these results.
Of 12,483 probes, 930 genes were load-responsive under one or more conditions. Hierarchical clustering revealed several clusters of genes that were upregulated in response to acute pressure overload that either remained elevated or that returned to baseline during chronic overload. Surprisingly, we did not observe gene clusters that were regulated in response to chronic pressure overload alone. These data indicate that the acute molecular response to hypertrophy may be an important determinant of later remodeling, prior to the clinical, pathophysiological manifestations of hypertrophy and failure. We documented through echocardiography that aortic-banded hearts were at a stage of compensatory hypertrophy and not in failure. The number of genes differentially regulated during chronic compensatory hypertrophy (slightly greater than 50 in both males and females) demonstrates that compensation is an active process of up- and downregulation of gene expression. The scope of this active compensatory process has not previously been appreciated prior to the application of the genomics approach. Further studies are needed to determine the changes in expression during the transition to cardiac failure.
The present study confirms the consistent finding of increased atrial natriuretic peptide and ß-actin in cardiac hypertrophy (10) as well as following myocardial infarction (20). We observed increased expression of genes in response to chronic pressure overload that were recently shown to be increased in human heart failure (21), including connective tissue growth factor, thrombospondin, osteoblast-specific factor 2, and collagen-
type 1. Thus these genes may likely not be considered markers of cardiac failure, as we have documented the absence of LV decompensation revealed by similar fraction shortening (systolic function) in all groups.
We also observed increased expression of transcripts recently reported to be increased following myocardial infarction using transcriptional profiling (20), including calcium binding protein, collagens, fibronectin, cathepsin, kinesin, transforming growth factor-induced transcript, osteoblast-specific factor 2, caveolin, and ankyrin-like repeat protein, suggesting some conservation in the myocardial response to injury and to hemodynamic overload. Enhanced expression of genes from functional categories comprising the extracellular matrix, transcription factors and gene regulation, receptor and cell signaling, defense, and metabolism as well as several nonannotated transcripts were revealed to be responsive to load. In addition, our analysis identified genes not previously associated with cardiac hypertrophy. The ankyrin repeat domain 2 protein (Arpp), for example, is a homolog of the known cardiac marker, cardiac ankyrin repeat domain (Carp). Arpp has been shown to be expressed in the heart ventricle but has thus far only been associated with skeletal muscle hypertrophy (22). In addition, the protein inhibitor of activated T cells, STAT3, was found to be upregulated both acutely and chronically. STAT3 had been recently shown to contribute to cardiac adaptation by regulating vascular function under conditions of stress and has been associated with controlling the transition to failure (18, 29). Here we demonstrate its expression before the development of failure.
Numerous in vivo and in vitro studies indicate that the cellular mechanisms responsible for the integrated response to pressure overload are complex (4), and the influence of sex difference in these pathways is unknown. Clinical and experimental studies have documented that sex influences the response to pressure overload (8, 23, 15). Clinical studies have documented that female patients with a similar degree of aortic stenosis frequently demonstrate thick-walled chambers (i.e., more favorable geometric remodeling) compared with male patients (1, 2). Studies focusing on small numbers of genes have shown intrinsic differences in the molecular adaptation to pressure overload independent of factors extrinsic to the myocardium (25). The present study addresses this extensively using transcriptional profiling. Several genes showed differing degrees of upregulated expression between males and females (Tables 4 and 5). Among these were serine protease inhibitor, lipocalin, LIM protein, initiation factor eIF-4AI, thrombospondin, catalase, Drosophila enabled homolog, WD repeat domain protein, programmed cell death 6 interacting protein, Sp1 transcription factor, and caveolin, as well as several transcripts not currently annotated.
Of additional interest are the three identified sex dimorphic clusters, which contain genes with differential patterns of expression between males and females. Interpretation of sex-related changes in gene reprogramming in response to pressure overload must take into consideration the magnitude of LV hypertrophy in male and female banded animals of identical age. A potential limiting factor in this model is the comparability of the systolic pressure overload, because a similar size occlusion was placed on the aortas of male and female mice of differing body weights. We have not determined the pressure gradient at the time of banding in male vs. female mice, and therefore we cannot rule out a difference; however, in response to this systolic pressure overload, the resulting magnitude of LV hypertrophy was similar in male and female aortic-banded mice, making any difference unlikely. The magnitude of changes in end-diastolic dimensions and anterior wall thickness were also similar in male and female banded mice. Interestingly, a subtle difference in geometric remodeling between male and female mice was observed: the magnitude of increase in posterior wall thickness was significantly higher in female compared with male banded mice. We cannot determine from the results in this study whether the sex-specific differences in gene expression contribute to or are the result of this difference in geometric remodeling, or whether these sex differences persist during cardiac failure. The transcript for the estrogen receptor has positively been identified in cardiac myocytes from both male and female animals (25), demonstrating the potential for estrogen receptor signaling in this cell type, although estrogen receptor signaling would presumably be much lower in males vs. females, due to an order of magnitude lower level of serum estrogen in males (16). Estrogen has been shown to regulate certain myosin isoforms and other structural proteins (16, 25). Further studies to determine the potential regulation by estrogen or other mechanisms involved in the sex-specific response to heart are needed. Little is known about the effects of testosterone signaling on gene expression in the myocardium. Many of the sex-specific genes identified in this study are novel, nonannotated genes or genes not previously shown to be expressed in the heart. One of these, orosomucoid (
1-acid glycoprotein, AGP), was validated as a differentially regulated gene in this study using real-time RT-PCR. Orosomucoid 1 has not previously been shown to be expressed in the heart. Orosomucoid is excreted in the urine, and its urinary excretion rate was increased in patients with type II diabetes and found to predict all-cause and cardiovascular mortality in these patients (5). Its myocardial expression level in heart failure is not known. In addition, FK506 binding protein 5 (51 kDa) was found to be exclusively upregulated acutely in females. This protein is a potent inhibitor of the glucocorticoid receptor (7) and is involved in controlling steroid receptor localization and transport (6). Further studies are warranted to uncover novel pathways and mechanisms associated with these genes to understand the molecular basis for sex differences in cardiac hypertrophy and failure.
 |
ACKNOWLEDGMENTS
|
---|
These studies were supported by grants from the National Heart Lung and Blood Institute.
Editor M. Stoll served as the review editor for this manuscript submitted by Editors R. E. Pratt and V. J. Dzau.
 |
FOOTNOTES
|
---|
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: R. E. Pratt, Brigham and Womens Hospital, Thorn 13, 75 Francis St., Boston, MA 02115 (E-mail: rpratt{at}rics.bwh.harvard.edu).
10.1152/physiolgenomics.00046.2002.
* E. O. Weinberg and M. Mirotsou contributed equally to this research. 
 |
References
|
---|
- Aurigemma GP and Gaasch WH. Gender differences in older patients with pressure-overload hypertrophy of the left ventricle. Cardiology 86: 310317, 1995.[ISI][Medline]
- Carroll JD, Carroll EP, Feldman T, Ward DM, Lang RM, McGaughey D, and Karp RB. Sex-associated differences in left ventricular function in aortic stenosis of the elderly. Circulation 86: 10991107, 1992.[Abstract]
- Chadwick BP, Heath SK, Williamson J, Obermayr F, Patel L, Sheer D, and Frischauf AM. The human homologue of the ninjurin gene maps to the candidate region of hereditary sensory neuropathy type I (HSNI). Genomics 47: 5863, 1998.[ISI][Medline]
- Chien KR. Genomic circuits and the integrative biology of cardiac diseases. Nature 407: 227232, 2000.[ISI][Medline]
- Christiansen MS, Hommel E, Magid E, and Feldt-Rasmussen B. Orosomucoid in urine predicts cardiovascular and over-all mortality in patients with type II diabetes. Diabetologia 45: 115120, 2002.[ISI][Medline]
- Davies TH, Ning YM, and Sanchez ER. A new first step in activation of steroid receptors: hormone-induced switching of FKBP51 and FKBP52 immunophilins. J Biol Chem 277: 45974600, 2002.[Abstract/Free Full Text]
- Denny WB, Valentine DL, Reynolds PD, Smith DF, and Scammell JG. Squirrel monkey immunophilin FKBP51 is a potent inhibitor of glucocorticoid receptor binding. Endocrinology 141: 41074113, 2000.[Abstract/Free Full Text]
- Douglas PS, Katz SE, Weinberg EO, Chen MH, Bishop SP, and Lorell BH. Hypertrophic remodeling: gender differences in the early response to left ventricular pressure overload. J Am Coll Cardiol 32: 11181125, 1998.[ISI][Medline]
- Esposito G, Prasad SV, Rapacciuolo A, Mao L, Koch WJ, and Rockman HA. Cardiac overexpression of a G(q) inhibitor blocks induction of extracellular signal-regulated kinase and c-Jun NH(2)-terminal kinase activity in in vivo pressure overload. Circulation 103: 14531458, 2001.[Abstract/Free Full Text]
- Friddle CJ, Koga T, Rubin EM, and Bristow J. Expression profiling reveals distinct sets of genes altered during induction and regression of cardiac hypertrophy. Proc Natl Acad Sci USA 97: 67456750, 2000.[Abstract/Free Full Text]
- Heid CA, Stevens J, Livak KJ, and Williams PM. Real time quantitative PCR. Genome Res 6: 986994, 1996.[Abstract]
- Hwang DM, Dempsey AA, Lee CY, and Liew CC. Identification of differentially expressed genes in cardiac hypertrophy by analysis of expressed sequence tags. Genomics 66: 114, 2000.[ISI][Medline]
- Li C and Hung Wong W. Model-based analysis of oligonucleotide arrays: model validation, design issues and standard error application. Genome Biol 2: research0032.10032. 11, 2001.
- Lorell BH and Carabello BA. Left ventricular hypertrophy: pathogenesis, detection, and prognosis. Circulation 102: 470479, 2000.[Free Full Text]
- Luchner A, Brockel U, Muscholl M, Hense HW, Doring A, Riegger GA, and Schunkert H. Gender-specific differences of cardiac remodeling in subjects with left ventricular dysfunction: a population-based study. Cardiovasc Res 53: 720727, 2002.[ISI][Medline]
- Malhotra A, Buttrick P, and Scheuer J. Effects of sex hormones on development of physiological and pathological cardiac hypertrophy in male and female rats. Am J Physiol Heart Circ Physiol 259: H866H871, 1990.[Abstract/Free Full Text]
- Nakamura A, Rokosh DG, Paccanaro M, Yee RR, Simpson PC, Grossman W, and Foster E. LV systolic performance improves with development of hypertrophy after transverse aortic constriction in mice. Am J Physiol Heart Circ Physiol 281: H1104H1112, 2001.[Abstract/Free Full Text]
- Osugi T, Oshima Y, Fujio Y, Funamoto M, Yamashita A, Negoro S, Kunisada K, Izumi M, Nakaoka Y, Hirota H, Okabe M, Yamauchi-Takihara K, Kawase I, and Kishimoto T. Cardiac-specific activation of signal transducer and activator of transcription 3 promotes vascular formation in the heart. J Biol Chem 277: 66766681, 2002.[Abstract/Free Full Text]
- Shamsadin R, Adham IM, Nayernia K, Heinlein UA, Oberwinkler H, and Engel W. Male mice deficient for germ-cell cyritestin are infertile. Biol Reprod 61: 14451451, 1999.[Abstract/Free Full Text]
- Stanton LW, Garrard LJ, Damm D, Garrick BL, Lam A, Kapoun AM, Zheng Q, Protter AA, Schreiner GF, and White RT. Altered patterns of gene expression in response to myocardial infarction. Circ Res 86: 939945, 2000.[Abstract/Free Full Text]
- Tan FL, Moravec CS, Li J, Apperson-Hansen C, McCarthy PM, Young JB, and Bond M. The gene expression fingerprint of human heart failure. Proc Natl Acad Sci USA 99: 1138711392, 2002.[Abstract/Free Full Text]
- Tsukamoto Y, Senda T, Nakano T, Nakada C, Hida T, Ishiguro N, Kondo G, Baba T, Sato K, Osaki M, Mori S, Ito H, and Moriyama M. Arpp, a new homolog of carp, is preferentially expressed in type 1 skeletal muscle fibers and is markedly induced by denervation. Lab Invest 82: 645655, 2002.[ISI][Medline]
- Wallen WJ, Cserti C, Belanger MP, and Wittnich C. Gender-differences in myocardial adaptation to afterload in normotensive and hypertensive rats. Hypertension 36: 774779, 2000.[Abstract/Free Full Text]
- Wang Y, De Keulenaer GW, Weinberg EO, Muangman S, Gualberto A, Landschulz KT, Turi TG, Thompson JF, and Lee RT. Direct biomechanical induction of endogenous calcineurin inhibitor Down syndrome critical region-1 in cardiac myocytes. Am J Physiol Heart Circ Physiol 283: H533H539, 2002.[Abstract/Free Full Text]
- Weinberg EO, Thienelt CD, Katz SE, Bartunek J, Tajima M, Rohrbach S, Douglas PS, and Lorell BH. Gender differences in molecular remodeling in pressure overload hypertrophy. J Am Coll Cardiol 34: 264273, 1999.[ISI][Medline]
- Winer J, Jung CK, Shackel I, and Williams PM. Development and validation of real-time quantitative reverse transcriptase-polymerase chain reaction for monitoring gene expression in cardiac myocytes in vitro. Anal Biochem 270: 4149, 1999.[ISI][Medline]
- Yamamoto K, Dang QN, Kennedy SP, Osathanondh R, Kelly RA, and Lee RT. Induction of tenascin-C in cardiac myocytes by mechanical deformation. Role of reactive oxygen species. J Biol Chem 274: 2184021846, 1999.[Abstract/Free Full Text]
- Yamamoto K, Dang QN, Maeda Y, Huang H, Kelly RA, and Lee RT. Regulation of cardiomyocyte mechanotransduction by the cardiac cycle. Circulation 103: 14591464, 2001.[Abstract/Free Full Text]
- Yamauchi-Takihara K and Kishimoto T. A novel role for STAT3 in cardiac remodeling. Trends Cardiovasc Med 10: 298303, 2000.[ISI][Medline]