INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ADULT MAMMALIAN CARDIAC MUSCLE expresses two genes encoding for myosin heavy chains (MHCs), which have been designated - and -MHC (28). The - and -MHC genes are members of the MHC multigene family, in which each of the genes is expressed and highly regulated in a muscle-type-specific fashion (28, 30). Although -MHC is expressed only in the heart, -MHC is expressed in the heart and is also the major myosin isoform expressed in slow-twitch skeletal muscle (30).

In the myocardium, the MHC isoform composition affects the physiodynamics and energetics of the working heart, which are of great physiological significance to cardiac performance. The -MHC isoform is characterized by a higher ATPase activity and faster shortening speed than the -MHC isoform (2, 21). Thus hearts rich in the -isoform have a high intrinsic contractility, whereas those rich in the -isoform have a lower contractility but a higher economy of tension development (2). In the adult rodent, the -MHC is the predominant isoform expressed in the ventricles, accounting for ~85-90% of the total MHC protein pool, whereas the -MHC accounts for the remaining 10-15% (18). In cardiac cells of different mammalian species, the expression of the two MHCs is developmentally regulated (28) and can be altered by a variety of pathophysiological conditions, including abnormal thyroid status (14, 24, 28, 38), diabetes (5), and hemodynamic overload (22, 29).

Chronic hemodynamic overload, as occurs in hypertension, is a complex physiological stimulus that triggers significant changes in myocardial structure and function. The visible changes include cardiac hypertrophy, which is expressed as an increased heart weight-to-body weight ratio, while changes on the molecular level include altered phenotype expression of specific cardiac genes that include several contractile and regulatory proteins (4, 25, 31, 39). For example, in the hypertensive rodent heart, -MHC gene expression is significantly increased relative to normal control values. The upregulation of -MHC gene expression in hypertensive adult mammalian hearts has been well documented (16, 22, 29); however, the molecular mechanisms driving this upregulation are poorly understood. On the basis of nuclear run-on assays, it is established that the transcription of the -MHC gene is significantly increased above control level (40). Most previous studies have attempted to understand cardiac gene transcription regulation by using transient transfections in cultured isolated neonatal cardiac myocytes (7, 8, 11, 26, 27, 41). When neonatal cardiomyocytes in culture are subjected to stretch or treated with various growth factors, vasoactive substances (endothelin and ANG II), or ₁-adrenergic agonist, the cells hypertrophy and undergo molecular changes similar to that observed in the hypertensive heart (20, 27, 32, 36). However, this culture model of isolated cells is not able to mimic the complex circulatory and hemodynamic effects on the intact heart. Furthermore, the response of mature adult myocytes might differ from that of neonatal immature myocytes, which are used in most in vitro studies. Thus it is critical to evaluate the molecular mechanisms in response to pathophysiological changes in vivo, at the intact organ level. This issue is further emphasized by the fact that some discrepancy in the results has been found between in vivo and in vitro studies (9).

In the present study, we used the direct gene transfer approach to study the transcriptional activity of the -MHC gene promoter in the intact heart in vivo in the context of its regulation in abdominal aortic constriction (AbCon)-induced overload hypertension. -MHC promoter fragments of various lengths linked to the firefly luciferase (FLuc) reporter were injected into the left ventricular apex of adult rats at an early stage of systemic hypertension (5 days after surgically induced hypertension), and reporter expression was studied 7 days after injection. Our objective was to characterize the role of various -MHC promoter sequences in the regulation of -MHC gene transcription in the pressure-overloaded (hypertensive) rodent heart. More specifically, we tested the full-length (3500-bp) -MHC promoter activity in the AbCon heart and compared its regulation with that of the endogenous gene. We also tested the responsiveness of different-length -MHC promoters in the AbCon heart to define possible involvement of a specific region or elements of the promoter in the upregulation of the gene. Finally, we introduced a series of mutations in specific regulatory elements of the promoter and tested them for their ability to blunt the response observed in the wild-type promoter. Our results show that some previously defined, specific key cis-elements in the proximal promoter are involved in maintaining basal promoter activity; however, none of the tested key regulatory elements, individually or collectively, play a regulatory role in mediating the increased transcriptional activity of the -MHC gene promoter in response to AbCon.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animal Model and Experimental Design

DNA Injection Procedure

Tissue Processing

Blood Pressure Measurements

View this table: [in this window] [in a new window]   Table 1.   Mean blood pressure measurements, ventricular weight-to-body weight ratios, and endogenous -MHC mRNA expression in NC vs. AbCon hearts after 6 and 12 days

Plasmid Constructs

The 3500-, 408-, and 215-bp -MHC-pGL3 constructs are the same as those used previously (44). The 299-, 171-, and 71-bp promoter constructs were generated by PCR using the 408-bp construct as a template and Pfu high-fidelity DNA polymerase (Stratagene). The upstream sense primers were -MHC promoter sequences from 299 to 280 bp for the 299-bp fragment amplification, from 171 to 152 bp for the 171-bp fragment amplification, and from 71 to 52 bp for the 71-bp fragment amplification; the upstream primers were designed to contain an NheI site at their 5' ends to allow digestion and ligation in specific sites of the pGL3 vector. The downstream antisense primer was the same for all amplifications, and its sequence was complementary to that linking the -MHC promoter at the +34 position to the pGL3 sequence, spanning 26 bp and containing the HindIII site in its center. Subsequent to the amplifications, the PCR products were purified by gel electrophoresis and extraction (Qiagen gel extraction kit) and subjected to NheI-HindIII restriction digest. The digested fragments were purified by gel electrophoresis and extraction (Qiagen) and ligated into the pGL3 basic vector at the NheI-HindIII multicloning site. All promoter constructs were sequenced and examined for possible unwanted mutations due to the PCR amplification procedure.

To account for differences in DNA uptake, a fixed amount (8.03 µg) of the -MHC-pRL plasmid, expressing the RLuc reporter gene, was coinjected with the test -MHC-pGL3 plasmid in all experiments, and its activity was used to correct for variation in gene transfer efficiency. Thus -MHC promoter activity was expressed relative to -MHC-pRL by the FLuc-to-RLuc ratio, as described previously (44). It had been demonstrated previously by direct gene transfer into AbCon hearts (20, 34) that the -MHC gene transcription remains unchanged in the AbCon vs. NC hearts. This and the fact that the -MHC gene is a cardiac muscle-specific gene that is coexpressed with the -MHC gene in the same myocytes (37) make the -MHC-pRL promoter construct the ideal control reference for all direct gene transfer experiments involving the AbCon model. The -MHC promoter 2936 to +420 bp was a kind gift from Dr. Eugene Morkin (University of Arizona, Tucson, AZ) (42).

Plasmids were amplified in Escherichia coli cultures and purified by anion-exchange chromatography using disposable columns (Endofree Maxiprep, Qiagen). Plasmids were suspended in sterile PBS, and the concentration was determined by ultraviolet absorbance at 260 nm, using the conversion factor of 50 µg/ml per optical density (OD) unit. Plasmid preparations were examined by ethidium bromide staining after agarose gel electrophoresis to verify their supercoiled nature and their freedom from genomic DNA and cellular RNA.

Site-Directed Mutagenesis

A 408-bp e1 mutant containing a 4-bp mutation introduced into the promoter sequence of the 408-bp -MHC construct within the e1 regulatory element. This mutation consisted of changing GGTGG to CATAT (see Table 2 for mutagenic oligonucleotide sequence).

A 408-bp e2/e3 double mutant containing two consecutive 3-bp mutations introduced into the promoter sequence of the 408-bp -MHC construct within the e2 and e3 regulatory elements. This mutation consisted of changing GTG to TGT in the e2 element and changing ACC to CGG in the e3 element (see Table 2 for mutagenic oligonucleotide sequence).

A 408-bp e2/GATA/e3 triple mutant consisting of a 4-bp mutation introduced into the promoter sequence of the 408-bp e2/e3 double mutant at the site of the GATA sequence adjacent to the e2 element. This mutation consisted of changing GATAT to ACTCG (see Table 2 for mutagenic oligonucleotide sequence).

A 215-bp e3 mutant containing a 3-bp mutation introduced into the promoter sequence of the 215-bp -MHC construct within the e3 regulatory element. This mutation was identical to the e3 mutation in the 408-bp e2/e3 double-mutant construct described above.

Reporter Gene Assays

MHC mRNA Analysis

Analyses of MHC mRNA isoforms utilized a modification of a previously used RT-PCR technique designed to quantitate relative amounts of MHC mRNAs representing the various MHC isoforms ( and ), as described in detail previously for skeletal muscle MHC isoforms (6).

RT. For each sample, 1 µg of total RNA was reverse transcribed using the SuperScript II RT (GIBCO BRL) and oligo(dT) primers according to the provided protocol. At the end of the RT reaction, the tubes were heated at 85°C for 5 min to stop the reaction and stored frozen at 80°C until they were used in the PCR.

PCR primers and internal control fragment. The 5'-upstream primer was designed from a highly conserved region in all known rat MHC genes located at ~500-550 bp upstream from the stop codon. All known seven MHC mRNA isoforms are identical in this region spanning ~35 nt. A 20-nt oligonucleotide of the following common sequence was chosen: 5'-AGAAGGAGCAGGACACCAGC-3'. It is the same 5' primer used to amplify skeletal MHC mRNA as described previously (1). The 3'--MHC isoform-specific oligonucleotides used in the PCR was the same as that used previously (6, 43). The 3'--MHC-specific primer used in the amplification is the same as that used for Northern hybridization and is of the following sequence: 5'-GTGGGATAGCAACAGCGAGGC-3'.

PCRs. Each RT reaction was diluted 40-fold with nuclease-free water and mixed with an equal volume of the control fragment at the appropriate dilution (~1 amol/µl). Two microliters of this mixture were used for 25-µl PCRs, which were carried out as described previously (1, 6). PCR products were separated on a 2% agarose gel by electrophoresis, and signal quantification was done as reported previously (6, 43). This method was used to calculate the expression of each specific MHC mRNA isoform relative to the total MHC mRNA pool.

Gel Mobility Shift Assays

Statistical Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Validation of the Hypertension Model

Our results show that in the AbCon groups the mean blood pressure increased by 45 and 56% after 6 and 12 days, respectively (P < 0.05), compared with the NC animals (Table 1). Also the hearts were significantly enlarged by 26% at 6 days and by 35% at 12 days on the basis of the ventricular weight-to-body weight ratio (Table 1). Endogenous -MHC mRNA analysis shows that, after 12 days of AbCon, the -MHC mRNA proportion was significantly increased from 12 ± 2% in the NC hearts to 38 ± 2% in the AbCon hearts (Table 1). These findings confirm that 6-12 days after the AbCon surgery the heart was significantly altered and underwent changes at the molecular level that resulted in a phenotype change. Therefore, this model was most suitable to study transcriptional events affecting the activity of the -MHC gene promoter.

Transcriptional Activity of the 3500-bp -MHC Promoter in the AbCon Heart: Validation of the Gene Injection Protocol

There is strong evidence that the -MHC promoter is a good reference promoter in interpreting the transcriptional activity of the -MHC promoter under normal conditions. However, when studies involve experimental manipulations, for correct interpretation of the test promoter activity, the control reference promoter activity (used to normalize the response) should ideally not change in response to these manipulations. Also, it is important to ensure that the manipulation does not affect differentially the posttranscriptional handling of the reporter genes (FLuc and RLuc), such as mRNA and protein synthesis and degradation rates.

Little is known about the transcriptional activity of the -MHC promoter in AbCon hearts, especially in the early stage of the hypertensive response. However, Ojamaa et al. (34) and Herzig et al. (20) found that the -MHC promoter activity was not altered in the AbCon model under conditions similar to our experimental model, although the -MHC mRNA expression was slightly but significantly decreased. This was attributed to posttranscriptional control of the -MHC gene expression in the AbCon hearts (20, 34).

Our results show that when the 3500-bp -MHC-pGL3 and -MHC-pRL were cotransfected in NC and AbCon hearts, the -MHC promoter activity relative to  (-to- ratio) was induced by ~385% over the NC expression level (Fig. 1A). Analysis of the net -MHC promoter activity in the heart of NC and AbCon rats without any correction shows that -MHC activity was significantly increased by ~160% (P < 0.01) in the AbCon group (Fig. 1B). Along with these changes, the net activity of the -MHC promoter was decreased by ~33% (Fig. 1C), but this decrease was not statistically significant (P = 0.2). We speculate that this apparent 33% decrease in the -MHC activity is the result of a reduced gene uptake. In studying the various -MHC promoter constructs tested in this study (total of 10 different constructs), it was found that, in all -MHC promoter-containing constructs, the net -MHC activity was consistently higher in the AbCon than in the NC hearts, while the -MHC promoter activity, on average, was not different between NC and AbCon hearts (Fig. 1D). For example, in these separate experiments, a fixed amount of the -MHC-pRL (8.03 µg/apex) was coinjected with the various tested -MHC promoter constructs in the NC vs. AbCon model. When the -MHC promoter net activity was averaged across all 10 experiments, the -MHC activity was similar in NC (n = 97) and AbCon (n = 142) hearts (Fig. 1D), thus validating the use of -MHC promoter as a reference control to correct for plasmid uptake in the AbCon model.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Promoter activity of 3500-bp -myosin heavy chain (MHC) and -MHC in normal control (NC) hearts and hearts with abdominal aortic constriction (AbCon) 7 days after plasmid injection and 12 days after AbCon. A: 3500-bp -MHC promoter activity as normalized to -MHC and expressed as -MHC firefly luciferase (FLuc)-to--MHC-pRL ratio in NC and AbCon hearts. B: -MHC promoter activity expressed as net FLuc activity in relative light units (RLU; i.e., the difference between the actual sample and the uninjected heart sample). C: -MHC promoter Renilla luciferase (RLuc) activity expressed as net RLU. D: average of pooled -MHC activity in a total of 10 separate experiments done in this study, which include the deletion and mutational analysis of the -MHC promoter, results of which are reported in Figs. 2-4 (n = 97 for NC and 142 for AbCon). Injected plasmid mixtures consisted of 10 µg of the 3500-bp -MHC promoter-pGL3 and 8.03 µg of the -MHC promoter-pRL (equimolar to 10 µg of 3500-bp -MHC-pGL3) dissolved into 40 µl of sterile PBS. Values are means ± SE; n = 10 NC and 14 AbCon. NS, not significant. *P < 0.05, AbCon vs. NC. RLU in 5 µl of tissue extract was determined using a luminometer (Monolight 2010C) and integrated over a 10-s interval.

Furthermore, the data in Fig. 1 show that the exogenous 3500-bp -MHC promoter is active in the normal rodent heart, and this activity is induced ~3.8-fold in response to AbCon. However, this induction in reporter activity ratio (FLuc/RLuc) could be attributed (at least in part) to differential effects of pressure overload on posttranscriptional handling of the reporter genes such as mRNA and protein stability. For example, preferentially, accumulation of FLuc protein over RLuc protein in the AbCon hearts would result in a higher FLuc/RLuc without necessarily a higher transcription rate of the FLuc gene. To test the effects of AbCon on FLuc/RLuc, the promoterless pGL3 plasmid was coinjected with the -MHC-pRL plasmid in NC and AbCon hearts in the same way as was done for all the tested -MHC promoter constructs (see MATERIALS AND METHODS). Hearts injected with the promoterless pGL3 basic plasmid have a low background level of FLuc expression that can be accurately determined via the reporter assay system (see MATERIALS AND METHODS). The background activity resulting from promoterless plasmid injection was consistently >20-fold higher that of uninjected tissue extract activity. Thus the promoterless pGL3 plasmid injection is a good approach to test the effects of the AbCon model on FLuc stability without the confounding effects from transcription. The results of these injections are reported in Fig. 2. The data show that the net activity levels of FLuc (Fig. 2A) or RLuc (Fig. 2B) were unchanged in response to AbCon. More importantly, RLuc/FLuc was not different in the AbCon hearts (Fig. 2C), although the endogenous -MHC mRNA levels were induced to 36 ± 3%, and heart weight-to-body weight ratio increased by 34 ± 2% (Table 3). These results indicate that any induction in this FLuc/RLuc (as observed in Fig. 1A) in the AbCon hearts must indeed result from a change in the -MHC promoter activity upstream to the FLuc gene.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Luciferase activities in NC and AbCon hearts 7 days after promoterless pGL3 basic plasmid and -MHC-pRL coinjection and 12 days after AbCon. A: FLuc activity in RLU. Activity in 5 µl of tissue extract was determined using a luminometer (Monolight 2010C) and integrated over 10-s intervals. B: RLuc activity in RLU. For FLuc, activity in the same 5 µl of tissue extract was determined using the dual-luciferase assay reagents (Promega) and a Monolight 1010C luminometer. C: FLuc-to-RLuc ratio in NC and AbCon hearts. RLU reflects the difference between the actual sample and the uninjected heart sample. Injected plasmid mixtures consisted of 5.08 µg of promoterless (basic) pGL3 and 8.03 µg of the -MHC promoter-pRL (both are equimolar to 10 µg of 3500-bp -MHC-pGL3) dissolved into 40 µl of sterile PBS. Values are means ± SE; n = 10 NC and 14 AbCon.

View this table: [in this window] [in a new window]   Table 3.   Relative increase in ventricular weight-to-body weight ratios and percent -MHC mRNA expression in AbCon groups in which the specified -MHC promoters were studied

Thus we have confirmed that the changes in -MHC gene expression due to AbCon are mainly regulated at the transcriptional level and that a deletion analysis of the promoter would enable us to delineate important sequences of the promoter that are responsible for regulating the hypertension-induced increase in -MHC gene transcription.

Deletion Analysis of the -MHC Promoter in NC and AbCon Hearts

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Deletion analysis of the -MHC-pGL3 promoter-reporter construct: responsiveness to AbCon. A: schematic diagram of the different-length -MHC promoters tested in NC vs. AbCon hearts. The enhancer region between 2900 and 3500 bp of the promoter, as defined in previous experiments (44), is depicted. Also shown are several cis-regulatory elements on the 408-bp promoter based on previous work: e1, e2, e3, C-rich (41), TRE/E box (8), GATA (19), and nuclear factor of activated T cells (NF-AT)-like binding site based on sequence comparison to NF-AT-binding site consensus (35). All these promoters extend to +34 bp 3' to the transcription start site (TSS) of the gene and are inserted in the multicloning site of the pGL3 vector (Promega) just upstream to the FLuc reporter gene. B: activities of -MHC promoter deletions in NC hearts; activities are normalized to the coinjected -MHC-pRL activity and expressed as FLuc-to-RLuc ratio. The 3500 construct has the highest activity, and the 408-, 299-, and 215-bp constructs have similar activities in NC hearts but at a lower level than the 3500-bp construct. The 171-bp fragment shows an overall lower level of activity (P < 0.05) than the 408-, 299-, and 215-bp fragments. The 71-bp and the promoterless (0) constructs have similar activities, i.e., 0.0026 ± 0.0002 and 0.0027 ± 0.0002, respectively, which are ~3% of the 171-bp construct activity. C: fold increase of the -MHC promoter activity (-to- ratio) in AbCon hearts relative to NC hearts. Data show the response for all the tested deletions and the promoterless (0) construct. Each bar is the difference in activity between AbCon and NC normalized to NC [(AbCon  NC)/NC]. Different-length -MHC promoter constructs were injected in equal molar amount to 10 µg of 3500-bp -MHC pGL3. The -MHC-pRL was always injected at 8.03 µg. n = 10 for each NC group and as reported in Table 3 for the AbCon groups. *P < 0.05 vs. all others; **P < 0.05 vs. all others but not 0 vs. 71; #P < 0.05 vs. 408 and 299.

When the same constructs were studied in AbCon hearts, all the tested deletions, including the short 71-bp construct, of the -MHC promoter showed significant upregulation of the -to- reporter ratio. The 3500-bp full-length promoter showed maximal responsiveness: an approximately fourfold increase in the -to- reporter ratio compared with the NC hearts (Fig. 3C). The 408- and 299-bp -to- reporter ratios were induced to a similar level by 155 and 157%, respectively, in response to AbCon, but this responsiveness was 60% lower than that of the 3500-bp promoter (P < 0.05; Fig. 3C). The 215- and 171-bp promoter activities were induced by 67 and 87%, respectively (Fig. 3C). Interestingly, although the 71-bp promoter construct had lost almost all of its activity, upregulation of reporter gene activity in hypertension was the same as in the 171-bp construct (87%), indicating that the promoter sequence located between 171 and 71 bp is crucial for basal expression of the gene but is not involved in mediating the hypertension response. Unlike the 71-bp -MHC promoter construct, the promoterless construct (pGL3 basic) reporter activity did not show responsiveness to hypertension (Figs. 2 and 3C), suggesting that the upregulation observed in the 71-bp -MHC promoter is specific to the -promoter and not a generalized response to AbCon. This strongly indicates that a regulatory element that is involved in hypertension regulation of the gene is located between the TSS and position 71 of the -MHC promoter.

These results suggest that full regulation of -MHC gene expression in the AbCon model requires the full-length promoter (i.e., 3500 to +34 bp). However, because regulation is also significant in the shortest promoter studied (71 bp), which is lacking all the known positive regulatory elements, the hypertension response appears to be regulated through several elements located in different regions of the promoter. The level of induction of the 299-bp promoter fragment was identical to that of the 408-bp fragment, which suggests that the e1 repressor element or other promoter sequences located between 408 and 299 bp (Fig. 3) are not critically involved with the upregulation of the -MHC gene expression in the AbCon model. The significant decrease in responsiveness when the 299-bp -MHC response is compared with the 215-bp -MHC response (Fig. 3C) suggests that some sequences located between 299 and 215 bp relative to TSS are responsible for the -MHC upregulation in response to hypertension. In this region, three regulatory elements are defined: the e2 element, GATA (AT rich), and C-rich regions (Fig. 3A). As more sequences were removed between 215 and 71 bp, including the loss of the e3 element (41) and a nuclear factor of activated T cells (NF-AT)-like binding site (35), no further reduction in the responsiveness to AbCon was observed (Fig. 3C), indicating that no hypertension response elements are located between 215 and 71 bp. In conclusion, at least three regions of the -MHC promoter appear to be involved in hypertension regulation of the gene: 1) upstream sequences between 408 and 3500 bp, 2) sequences between 299 and 215 bp, and 3) sequences between 71 bp and the TSS.

It is important to note that the differences in responsiveness of these promoter fragments to AbCon cannot be attributed to differences in the endogenous response. On the basis of ventricular weight-to-body weight ratios and endogenous -MHC mRNA expression (Table 3), the AbCon response was relatively homogeneous.

Mutation Analysis of the 408-bp -MHC Promoter in NC and AbCon Hearts

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Mutation analysis of the 408-bp -MHC promoter in NC hearts (A) and responsiveness to AbCon (B). A: promoter activity (-to- ratios) of the 408-bp wild-type (WT) and mutated constructs in NC hearts. Note the significant increase in -MHC promoter activity when the e1 repressor element was mutated. Simultaneous mutations of the e2 and e3 elements significantly reduced the promoter activity relative to WT activity in NC hearts; however, the GATA mutation in addition to the mutation of the e2-e3 elements had no further effect on the promoter activity. B: fold increase of the promoter activity in the AbCon hearts: effects of specific mutations on the 408-bp -MHC promoter responsiveness to AbCon. There was no significant difference in the increase of ventricle-to-body weight ratio or in the increased levels of the endogenous -MHC mRNA expression in the AbCon hearts (Table 4). Values are means ± SE; n = 10 for each NC group and as reported in Table 4 for the AbCon groups. *P < 0.05 vs. WT.

Simultaneous mutations of the e2 and e3 elements on the 408-bp -MHC promoter did not alter the degree of responsiveness to AbCon compared with the wild-type mutation (Fig. 4B); however, the same mutations reduced the promoter activity in NC hearts by 39% (Fig. 4A), indicating that the e2 and e3 elements are positive regulatory elements but are not involved in the transcriptional regulation in the AbCon response in the heart. Furthermore, mutation of a GATA-like sequence, in addition to e2 and e3 mutations on the 408-bp -MHC promoter, had no effect on NC expression or responsiveness to AbCon (Fig. 4A). This result raises a question concerning the importance of this GATA-like site in regulating -MHC promoter activity. At the same time, these results confirm in the in vivo model the role of e1, e2, and e3 regulatory elements in maintaining the normal expression of the -MHC gene but do not support their involvement in the upregulation response to AbCon when studied in the 408-bp promoter fragment. In the AbCon groups described above, the endogenous response of the heart to AbCon was relatively homogeneous in all AbCon groups that were studied. There was no significant difference in the AbCon-induced increase in ventricle weight-to body weight ratio or in the increased levels of the endogenous -MHC mRNA expression among the various AbCon groups (Table 4). Thus any difference in the reporter gene expression cannot be attributed to a lack of homogeneity in the endogenous response to AbCon among the various groups.

View this table: [in this window] [in a new window]   Table 4.   Relative increase in ventricular weight-to-body weight ratios and percent -MHC mRNA expression in AbCon groups after injection with the specified 408-bp -MHC-pGL3 promoter construct

Effect of a e3 Mutation on the 215-bp -MHC Promoter: Activity in NC Hearts and Responsiveness to AbCon

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Effect of e3 mutation on the 215-bp -MHC promoter: activity in NC hearts (A) and responsiveness to AbCon (B). The e3 mutation on the 215-bp fragment was associated with a significant decrease in -MHC promoter activity relative to the wild-type activity in the NC heart (A). However, its responsiveness to AbCon was not reduced (B). Thus the e3 element is a positive regulatory element in the normal expression of the -MHC gene but does not play a role in responsiveness of the gene to AbCon. Values are means ± SE; n = 10 for each NC group and as reported in Table 5 for the AbCon groups. *P < 0.05 vs. WT.

Again the heart relative mass (ventricle weight-to- body weight ratio) and percent -MHC mRNA expression were not different among the NC groups or among the AbCon groups. A consistent increase was observed in both AbCon groups relative to their corresponding controls (Table 5).

View this table: [in this window] [in a new window]   Table 5.   Relative increase in ventricular weight-to-body weight ratios and percent -MHC mRNA expression in AbCon groups after injection with the specified 215-bp -MHC-pGL3 promoter-reporter construct

Nuclear Extract Interaction With Specific Cis-Elements on the -MHC Promoter

View larger version (40K): [in this window] [in a new window]   Fig. 6.   Cardiac nuclear extract (NE) binding to e2, GATA, and e3 cis-regulatory elements using DNA gel mobility shift assays. A: e2 binding. Lane 1, no binding was detected after NE interaction with 32P-labeled mutant double-stranded e2 oligonucleotide. Lanes 2-5, NE interaction with wild-type e2; binding resulted in formation of 3 specific complexes (C1, C2, and C3) that were competed off with 100-fold molar excess of cold self (lane 3); competition failed with 100-fold excess of mutant (Mut) double-stranded e2 (lane 4) or palindromic thyroid response element (TRE) double-stranded oligonucleotide (lane 5), which is of similar size but unrelated to the e2 sequence. Lanes 6 and 7, e2 binding with NE from NC and AbCon (AC) hearts, respectively. B: GATA binding. Lanes 1-4, NE interaction with the GATA-like element. Binding resulted in formation of 2 specific complexes (C1 and C2) that were competed off with 100-fold molar excess of cold self (lane 2); competition was not as effective with 100-fold excess of mutant double-stranded GATA (lane 3) or palindromic TRE sequence (lane 4), which is of similar size but unrelated to the GATA sequence. Lanes 5 and 6, GATA binding with NE from NC and AbCon hearts, respectively. C: e3 binding. Lanes 1-4, NE interaction with the e3 element. Binding resulted in formation of 2 specific complexes (C1 and C2) that were effectively competed off with 100-fold molar excess of cold self (lane 2); competition was not as effective with 100-fold excess of mutant double-stranded e3 (lane 3) or palindromic TRE sequence (lane 4), which is of similar size but unrelated to the e3 sequence. Lanes 5 and 6, e3 binding with NE from NC and AbCon hearts, respectively. In gels, density of the shifted bands was determined by scanning densitometry (ImageQuant, Molecular Dynamics). Bar graph represents data (means ± SE) from 6 independent observations per group, whereby each n consisted of a pool from heart tissue of 5 rats. For all competition assays, a mix of NC and AbCon cardiac NE was used for bindings. SU, arbitrary scanning unit; Comp, competitor cold double-stranded oligonucleotides used at 100-fold molar excess to the 32P-labeled probe. See Table 2 for oligonucleotide sequences.

In addition to e2, GATA, and e3, we also examined the binding to the e1 repressor element, as well as to the TATA box, and these results are reported in Fig. 7. NC and AbCon heart nuclear extracts formed shifted complexes on the gel. The specific binding to the e1 sense strand increased by ~26% under AbCon conditions (P < 0.05). e1 is a repressor element, as defined previously (7, 11), which is confirmed by our data showing that the in vivo promoter activity increased in the NC heart when this e1 element was mutated in the 408-bp -MHC promoter construct (Fig. 4). When this element was deleted, as in the 299-bp -MHC promoter (Fig. 3), or mutated, as in the 408-bp -MHC promoter (Fig. 4), the responsiveness to AbCon was still similar to that of the wild-type promoter, thus eliminating a potential role played by e1 in the AbCon induction of the promoter activity. If e1 were a player in the AbCon-induced -MHC promoter upregulation, one might expect to find a decreased repressor activity/nuclear extract-e1 interaction in AbCon hearts, but that was not the case. In contrast, a trend for the opposite was observed, i.e., increased e1 binding (Fig. 7). The significance of this observation remains unclear; thus more research is needed in this area.

View larger version (45K): [in this window] [in a new window]   Fig. 7.   Cardiac NE binding to e1 and -MHC TATA cis-regulatory elements using DNA gel mobility shift assays. A: e1 sense strand (ss) binding. Lanes 1-5, NE interaction with the single-stranded e1 ss. Binding resulted in formation of 2 specific complexes (C1 and C2) that were effectively competed off with 100-fold molar excess of cold self (lane 2); competition was not as effective with use of 100-fold excess of mutant e1 ss (lane 3), double-stranded (ds) e1 wild type (lane 4), or double-stranded palindromic TRE (lane 5). Lanes 5 and 6, e1 ss binding to NE from NC and AbCon hearts, respectively. B: -MHC TATA binding. Lanes 1-3, NE interaction with the -MHC TATA box. Binding resulted in formation of 1 shifted band (C1) that was effectively competed off with 100-fold molar excess of cold self (lane 2); competition was not as effective with use of 100-fold excess of double-stranded e2 element (lane 3), a TATA-unrelated sequence. Lanes 4 and 5, -MHC TATA binding to NE from NC and AbCon hearts, respectively. In gels, density of the shifted bands was determined by scanning densitometry (ImageQuant, Molecular Dynamics). Bar graph represents data (means ± SE) from 6 separate n per group, where n consisted of a pool from heart tissue of 5 rats. For all competition assays, a mix of NC and AbCon cardiac NE was used for the binding. See Table 2 for oligonucleotide sequences.

Because promoter activity of the shortest tested fragment, 71 bp, was also upregulated in response to AbCon, one could conclude that the basal promoter activity is induced. The basal promoter consists of downstream sequences located within 100 bp at the 3' end of the promoter, and these include the TATA box. Several nuclear proteins interact at the basal promoter region to form a multimeric complex required for transcription initiation. To establish whether AbCon hearts are associated with increased binding activity to the TATA element in the basal -MHC promoter, a gel mobility shift assay was used to test the interaction between the -MHC TATA element and nuclear extract from normal vs. AbCon hearts. Our results reported in Fig. 7 show no significant difference between NC and AbCon hearts' binding activities to TATA. Thus increased TATA binding protein, in and of itself, is not an explanation for the increased promoter activity in the AbCon hearts. One possible explanation is that other factors in the basal promoter machinery are activated, or other regulatory transcription factors are involved. Further deletions and studies of the activity of shorter fragments are required to characterize the -MHC promoter response to AbCon (see below).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The heart responds to systemic hypertension by altering the expression of various genes that are important for economizing the contractile properties of the heart. The -MHC gene is known to be upregulated in the hypertensive rodent heart. However, the mechanism by which this upregulation in gene expression is mediated is not well understood. Previously, tissue culture studies have attempted to delineate regions on the -MHC promoter that might be involved in mediating the hypertension response of this gene. Because of the complexity of the stimulus (mechanical, hormonal, neuronal), it is very difficult to create a useful tissue culture model of hypertension. Furthermore, it is important to study gene regulation in vivo under physiological conditions, because it has been shown that the results from in vitro studies of the gene can be in discordance with the results from in vivo studies. For example, Edwards and Ghaleh (9) tested -MHC promoter deletion constructs containing the rat or human e2 element under in vitro and in vivo conditions. When promoter/reporter constructs were transfected into cultured fetal rat cardiomyocytes, the human -reporter was expressed more than threefold above the equivalent rat construct. However, when these same -MHC promoter constructs were injected into adult rat hearts (in vivo), the levels of reporter