Heart and Circulatory Physiology


Evidence indicates that gender and sex hormonal status influence cardiovascular physiology and pathophysiology. We recently demonstrated increased L-type voltage-gated Ca2+ current (ICa,L) in coronary arterial smooth muscle (CASM) of male compared with female swine. The promoter region of the L-type voltage-gated Ca2+ channel (VGCC) (Cav1.2) gene contains a hormone response element that is activated by testosterone. Thus the purpose of the present study was to determine whether endogenous testosterone regulates CASM ICa,L through regulation of VGCC expression and activity. Sexually mature male and female Yucatan swine (7–8 mo; 35–45 kg) were obtained from the breeder. Males were left intact (IM, n = 8), castrated (CM, n = 8), or castrated with testosterone replacement (CMT, n = 8; 10 mg/day Androgel). Females remained gonad intact (n = 8). In right coronary arteries, both Cav1.2 mRNA and protein were greater in IM compared with intact females. Cav1.2 mRNA and protein were reduced in CM compared with IM and restored in CMT. In isolated CASM, both peak and steady-state ICa were reduced in CM compared with IM and restored in CMT. In males, a linear relationship was found between serum testosterone levels and ICa. In vitro, both testosterone and the nonaromatizable androgen, dihydrotestosterone, increased Cav1.2 expression. Furthermore, this effect was blocked by the androgen receptor antagonist cyproterone. We conclude that endogenous testosterone is a primary regulator of Cav1.2 expression and activity in coronary arteries of males.

  • voltage clamp
  • vascular
  • voltage-gated calcium channels

epidemiological and clinical evidence indicates that gender and sex hormone status influence cardiovascular physiology and pathophysiology (1, 17, 23, 37). Coronary heart disease (CHD) is a leading cause of mortality in both males and females; however, males show a greater prevalence of CHD compared with premenopausal females (1). Similarly, the incidence of hypertension is greater in males and postmenopausal females compared with premenopausal females (23). These findings led many to conclude that estrogen attenuates and, conversely, testosterone exacerbates the incidence and severity of cardiovascular disease. However, recent clinical studies (17, 37) have failed to support the concept that estrogen is beneficial or that testosterone is detrimental to cardiovascular disease (for review, see Ref. 28), highlighting a fundamental lack of knowledge regarding the interactions of sex hormones on coronary arteries.

Sex hormones exert multifaceted effects on vascular reactivity. Whereas many vascular effects of sex hormones are endothelium mediated (13, 16, 43), considerable evidence implicates vascular smooth muscle as a target. Acute administration of both estrogen and testosterone has been shown to relax arterial preparations in vitro (8, 9, 11). Similarly, acute administration of testosterone has been shown to increase anginal threshold in men with CHD (40). The acute relaxation effect of testosterone is due, in part, to the effects on smooth muscle ion channel activity; in particular, activation of large-conductance Ca2+-activated K+ channels (11, 39, 44) and inhibition of L-type Ca2+ channels (38). Thus ion channels contribute to many acute, presumably nongenomic, effects of sex hormones on vascular smooth muscle.

Fewer and more equivocal studies exist regarding long-term, potentially genomic, effects of testosterone on the vasculature. Khalil and colleagues (9, 10, 34) found that KCl-induced contraction and Ca2+ influx was greater in thoracic aortas of male rats compared with females, suggesting a greater voltage-gated calcium channel (VGCC) activity in males. This sex difference was eliminated by ovariectomy in females and restored with estrogen; in males, castration had no effect. These findings led to the conclusion that estrogen, not testosterone, is the primary sex hormone modulating vascular smooth muscle VGCC activity. However, in swine, chronic exogenous testosterone administration increased KCl-induced contractile responses of coronary arteries in both males and females (12). Similarly, canine coronary artery contractions to U46619 were reported to be greater in males compared with females, a difference which was eliminated by antiandrogens (24). Thus the role of testosterone in modulating vascular smooth muscle VGCCs in large mammals is unclear. Recently, we reported an increased L-type voltage-gated Ca2+ current (ICa,L) in coronary arterial smooth muscle (CASM) of male swine compared with females (3). Sequencing of the 5′-untranslated region of the α-subunit of the L-type VGCC (Cav1.2) gene provides evidence for a hormone response element activated by testosterone (27) suggesting that testosterone could regulate Cav1.2 gene expression and protein levels in vivo and thus contribute to gender differences in VGCC activity. Thus the purpose of the present study was to determine whether endogenous testosterone regulates ICa,L through regulation of VGCC expression and activity in CASM.



Sexually mature male and female Yucatan swine were obtained from the breeder (Sinclair Research Farm; Columbia, MO) and housed in pens at the College of Veterinary Medicine. Animal protocols were approved by the University of Missouri Animal Care and Use Committee in accordance with the “Principles for the Utilization and Care of Vertebrate Animals Used in Testing, Research and Training.”

Castration and hormone replacement.

Castration (orchiectomy) and hormone replacement was performed by the Swine Hormone Core within the Center for Gender Physiology and Environmental Adaptation, University of Missouri. One week after arrival, sexually mature animals (6 to 7 mo of age) were castrated with aseptic techniques under sedation with xylazine (15 mg/kg im) and ketamine (2.5 mg/kg im) and maintained under anesthesia with 1.5–2.0% isoflurane. Males were castrated (CM) via an incision made through the scrotum over each testicle (intact males; IM) and subsequently randomized to receive testosterone replacement (CMT; Androgel, Solvay Pharmaceuticals, 10 mg/day) or vehicle. Testosterone replacement occurred at the time of castration to avoid disruption of hormonal influence. Females remained gonad intact. Animals were euthanized for study 5–6 wk after surgery.

Hormone assays.

Blood samples (5 ml) were collected at the time of surgery before castration and at the time of death. Samples were collected into plain tubes and centrifuged at 1,000 rpm for 5 min, and the serum was decanted and frozen at −80°C until analysis. Testosterone was determined with a commercially available RIA kit (Diagnostic Products; Los Angeles, CA). Essentially, a solid-phase 125I-labeled RIA was utilized with a sensitivity of 4 ng/dl and an interassay and intraassay coefficient of variation of 8% and 6%, respectively. Cross-reactivity studies to the antibody established that the RIA was specific for testosterone with only 19-nortestosterone having 20% and 11-ketotestosterone with 16% cross-reactivity. Parallelism studies were also carried out with pooled boar serum, both at the high and low ends of the standard curve, resulting in good linearity.

Preparation of coronary arteries.

At the time of death, animals were anesthetized with ketamine (30 mg/kg im) and pentobarbital sodium (35 mg/kg iv) and administered heparin. The hearts were removed and placed in Krebs bicarbonate solution (4°C) during vessel isolation. Conduit (>1.0 mm inner diameter) segments of the right coronary artery (RCA) were trimmed of fat and connective tissue in sterile modified Eagle's minimal essential storage media containing 20 mM HEPES on ice.

CASM cell isolation.

Freshly dispersed CASM cells were obtained as described previously (3, 5, 6, 18). RCA were opened longitudinally and pinned, lumen side up, onto a silicone rubber substratum in ∼2 ml of low-Ca2+ solution. Cells were dispersed by incubation in enzyme solution consisting of low-Ca2+ (0.1 mM) physiological solution plus 294 U/ml collagenase (CLS II, Worthington), 6.5 U/ml elastase (Worthington), 2 mg/ml bovine serum albumin (fraction V, Sigma), 1 mg/ml soybean trypsin inhibitor (type I-S, Sigma), and 0.4 mg/ml DNase I (type IV, Sigma) for 60 min in a water bath at 37°C. The enzyme solution was then replaced with enzyme-free low-Ca2+ solution, and isolated single cells were obtained by repeatedly directing a stream of low-Ca2+ solution over the artery via a fire-polished Pasteur pipette. Cell suspensions were stored in low-Ca2+ (0.1 mM) buffer at 4°C until use (0–6 h).

Whole cell voltage clamp.

Whole cell VGCC currents were determined using a standard voltage-clamp technique (3, 5, 6, 18). Cells were initially superfused with physiological saline solution (PSS) containing (in mM) 0.1 CaCl2, 138 NaCl, 1 MgCl2, 5 KCl, 10 HEPES, and 10 glucose, pH 7.4, during gigaseal formation. After whole cell configuration, the superfusate was switched to PSS with tetraethylammonium chloride substitution for NaCl and 10 mM Ba2+ as the charge carrier. The pipette solution contained (in mM) 120 CsCl, 10 tetraethylammonium chloride, 1 MgCl2, 20 HEPES, 5 Na2ATP, 0.5 Tris GTP, and 10 EGTA, pH 7.1. Ionic currents were amplified with an Axopatch 200B patch-clamp amplifier (Axon Instruments). Whole cell currents were low-pass filtered with a cutoff frequency of 1,000 Hz, digitized at 2.5 kHz, and stored on computer. Current densities (pA/pF) were obtained for each cell by normalization of whole cell current to cell capacitance to account for differences in cell membrane surface area. Capacity currents were measured for each cell during 10-ms pulses from a holding potential of −80 mV to a test potential of −75 mV. Capacity currents were filtered at a low-pass cutoff frequency of 5 kHz. Leak subtraction was not performed. Data acquisition and analysis were accomplished with the use of pCLAMP 7.0 software (Axon Instruments). Voltage-dependent activation was determined as g/gmax = peak IBa/[gBa(VErev)], where gBa is the maximal conductance obtained from the linear regression of the positive limb of the current-voltage relationship through the apparent reversal potential (Erev), V is the step potential, and IBa is the peak inward current at the corresponding step potential. Activation data were fit to a conventional Boltzmann distribution equation, 1/{1 + exp[(V0.5V)/k]}, where V0.5 is the membrane potential producing half-maximal activation and k is the slope factor. For voltage-dependent inactivation, peak inward current during a 500-ms step depolarization to +20 mV was determined after a 4-s prepulse at membrane potentials from −80 to +50 mV in 10-mV increments. Relative peak current (I/Imax) data were fit to a conventional Boltzmann distribution equation, 1/{1 + exp[(V0.5V)/k]}. All experiments were conducted at room temperature (22–25°C).


RCA segments (5 mm axial length) were quick frozen in liquid N2 and stored at −80°C until processed. Samples were minced thoroughly in 200 μl of ice-cold buffer (50 mM Tris·HCl, pH 7.4, 0.1 mM EDTA, 0.1 mM EGTA, 0.1% β-mercaptoethanol, 1 mM PMSF, 2 mM leupeptin, 1 mM pepstatin A, 10% vol/vol glycerol) and subjected to 4 × 30 s homogenizations at a speed of 4.0–5.0 using a FASTPREP 120 system (QBiogene) with samples kept on ice for 5 min between each pulse. After a low-speed centrifugation, supernatants were subjected to high-speed centrifugation (100,000 g for 1 h). Crude membrane pellets were dissolved by vortexing in homogenization buffer containing 0.1% Triton X-100 at room temperature. Membrane protein concentrations (3–5 μg/μl) were determined using a Bradford protein assay kit (Bio-Rad). Equal protein amounts (30 μg) of sample were electrophoresed on 5% SDS-polyacrylamide gels and transferred to a polyvinylidene difluoride membrane at 30 V for 14–16 h at 4°C. After transfer, membranes were blocked for 1 h in Tris-buffered saline solution containing 5% dry milk plus 0.1% Tween 20 and probed with anti-Cav1.2 antibody (1:200; Alomone) for 2 h at room temperature. Immunoreactive bands were detected by chemiluminescence (ECL, Amersham) and quantified using NIH Image J software.

Quantitative RT-PCR.

Quantitative RT-PCR for Cav1.2 was performed as previously described (4) with modification. After dissection at 4°C, RCA segments (10 mm axial length) were quick frozen in RNase-free tubes for storage at −80°C until analysis. Frozen samples were powdered under liquid N2, and total RNA was extracted (TRI Reagent, Molecular Research Center). cDNA was transcribed from total RNA using the Superscript II RT kit (Invitrogen) using 200 units of RT in a 20-μl reaction containing 100 ng random hexamers, 5 mM MgCl2, 1 mM dNTPs, and 20 mM DTT. A minus RT reaction was performed to ensure no genomic DNA contamination. Quantitative RT-PCR was performed with the use of Texas red-labeled probes (IDT) on a Smart Cycler (model no. SC1000-1, Cepheid). The 25-μl reaction mixture contained RT buffer (Invitrogen), BHQ-2 quencher, 5 mM MgCl2, 0.8 mM dNTPs, 0.8 μM primer/probe, and 1 unit Taq polymerase (Roche). Each primer set was optimized for cycle time, annealing temperature, and MgCl2 concentration. Each PCR reaction was initiated by a 5-min incubation at 95°C. Cav1.2 was amplified using 40 cycles of 95°C (15 s), 60°C (30 s), and 72°C (30 s). 18S was amplified using 40 cycles of 95°C (15 s) and 60°C (1 min). The primer sets were the following: Cav1.2 sense, 5′-GTG TTC CAG TGT ATC ACC ATG G-3′; antisense, 5′-GTT GAC AGA TTC GGT CTC ACT TG-3′; and probe 5′-GGA CGA GGA GAA GCC CCG AA-3′; 18S sense, 5′-CGG CTA CCA CAT CCA AGG AA-3′; antisense, 5′-AGC TGG AAT TAC CGC GGC-3′; and probe, 5′-TGC TGG CAC CAG ACT TGC CCT C-3′. Nucleotide sequencing confirmed amplification of the expected product and demonstrated complete homology between human and Yucatan porcine Cav1.2 PCR-amplified products. Cav1.2 mRNA expression was normalized to 18S using the 2−ΔΔCMathmethod (29), where the threshold cycle difference (ΔCt) is determined as CtCav12 − Ct18S. Efficiency of the PCR reaction was verified by generating a standard curve-plotting Ct against serial dilutions of cDNA. A replication efficiency near 1.0 was determined by linear regression fit of the standard curve.


Androgen receptor (AR) expressions were determined using a standard immunohistochemistry protocol as used routinely (4). Samples of RCA were fixed in neutral buffered 10% formalin for a minimum of 24 h, embedded routinely in paraffin, and sectioned serially at 5 μm thickness. Sections were floated onto positively charged slides (Fischer), deparaffinized, and steamed in citrate buffer at pH 6.0 (Dako S1699) for 30 min to achieve antigen retrieval and then cooled for 30 min. The slides were stained manually with Tris buffer or water wash steps performed after each step. Sections were incubated with avidin-biotin two-step blocking solution (Vector SP-2001) to inhibit background staining and in 3% hydrogen peroxide to inhibit endogenous peroxidase. Nonserum protein block (Dako X909) was applied to inhibit nonspecific protein binding. Sections were incubated overnight at 4°C with a 1:800 dilution of primary rabbit polyclonal antibodies to AR (Affinity Bioreagents). After the appropriate washing steps were made, sections were incubated with biotinylated anti-rabbit secondary antibody in PBS containing 15 mM sodium azide and peroxidase-labeled streptavidin (Dako LSAB+ kit, peroxidase, K0690). Diaminobenzidine (Dako) was applied for 5 min to visualization of the reaction product. Sections were counterstained with Mayer's hematoxylin stain for 1 min, dehydrated, and coverslipped. Sections were photographed with an Olympus BX40 photomicroscope and Spot Insight Color camera (Diagnostic Instruments).


Data are expressed as means ± SE. Repeated-measures ANOVA was used for comparisons of current-voltage (I-V) relationships between groups. All other comparisons were made by ANOVA. Bonferroni post hoc analyses were applied when significant main or interaction effects were determined. A P value ≤0.05 was set as the criterion for significance in all comparisons.


Serum testosterone.

Animals were of similar age (7.4–7.8 mo) at time of death. Body weights were similar in intact, castrated, and hormone-replaced males at the time of surgery (31.3 ± 1.4, 30.6 ± 1.3, and 29.9 ± 1.5 kg, respectively) and death (31.8 ± 1.4, 31.8 ± 1.4, and 31.0 ± 1.8 kg, respectively). Serum testosterone levels in male and female swine are similar to those found in human males (200–1,300 ng/dl, 7–44 nM) and females (2–75 ng/dl, 0.1–3 nM) (1, 30, 41). Castration reduced circulating testosterone levels >90%, to levels not different from intact females (Fig. 1). Testosterone replacement in castrated males maintained serum testosterone at levels similar to that found in intact males.

Fig. 1.

Serum testosterone (T) level values from intact male (IM; n = 7), castrated male (CM; n = 8), castrated male with testosterone (CMT; n = 9), and intact female (IF; n = 6) at the time of death. Values are means ± SE. *P < 0.05 vs. IM and CMT.

Effect of in vivo testosterone on VGCC in CASM.

Figure 2 shows a representative family of currents obtained in conduit CASM from IM, CM, or CMT. Successive depolarization steps from a holding potential of −80 mV to a final test potential of +50 mV in 10-mV increments produced inward currents showing a peak near +20 mV and slow inactivation (selected test potentials shown). Similar to previous reports (3, 5, 6, 18), the inward currents in CASM from both male and female were dominated by L-type current, with no discernable T-type current present in any group. In the present study, neither castration nor testosterone replacement altered the dominant expression of ICa,L as confirmed by the kinetics of inward currents and the near complete block of inward current by nifedipine (Fig. 2, D and E). Castration produced an overall reduction in inward VGCC current in CASM from males, an effect that was prevented with testosterone (Fig. 3A). Consistent with our previous study (3), the magnitude of both the peak and sustained inward current were lower in intact females compared with intact males (Fig. 3B). In males, castration reduced both inward and sustained currents to levels not significantly different from females, an effect that was prevented with testosterone. These data indicate that endogenous testosterone stimulates CASM ICa and that loss of CASM ICa due to reductions in testosterone can be reversed by exogenous testosterone administration. Testosterone also influenced the voltage dependence of ICa. Castration produced a significant shift in voltage-dependent activation of ICa toward more negative potentials, an effect that was prevented by testosterone (Fig. 3, C and D, left). Neither sex nor testosterone had any effect on voltage-dependent inactivation of ICa (Fig. 3D, right).

Fig. 2.

Effect of testosterone status on voltage-gated Ca2+ channel (VGCC) current in coronary artery smooth muscle (CASM). Representative current traces for VGCC current in CASM from IM (A), CM (B), and CMT (C). Currents were elicited by 500-ms depolarization steps from a holding potential (hp) of −80 mV to test potentials (tp) of −60 to +60 mV in 10-mV increments with 10 mM external Ba2+ as the charge carrier (selected test potentials of −30, −10, and +20 mV shown). Depolarization steps positive to −40 elicited a sustained, slowly inactivating inward current that was maximal near +20 mV. Nifedipine (1 μM; nif) effectively abolished inward current in CM (D) and CMT (E), consistent with a predominance of dihydropyridine-sensitive, L-type current in all groups [nifedipine block in IM and IF shown previously (3, 6)]. Capacitance transients were eliminated for clarity.

Fig. 3.

Effect of testosterone status on the VGCC current-voltage (I-V) relationship. A: I-V relationships for whole cell VGCC current in CASM from IM, CM, and CMT. Current is plotted as the peak inward current measured during a 500 ms step depolarization to the membrane potential (Vm) indicated from a holding potential of −80 mV. Current is normalized to cell membrane capacitance (pA/pF). I-V relationships were significantly attenuated in CM and but not CMT compared with IM. B: group data for peak and sustained (end pulse; SS) current. C: voltage-dependent activation and inactivation. Data are derived as described in materials and methods. D: group data for membrane potential producing half-maximal activation (V0.5) for activation (left) and inactivation (right). Number of cells per group: IM (n = 33), CM (n = 29), CMT (n = 32), and IF (n = 32). *P < 0.05 vs. IM and CMT. All data are means ± SE of 4–5 cells per artery per animal.

The direct, positive relationship between serum testosterone and CASM calcium channel activity is clearly demonstrated in Fig. 4. In males, a significant positive, linear relationship between serum testosterone levels and CASM ICa density was observed (r2 = 0.99). Interestingly, for intact females, ICa was greater than predicted based on serum testosterone, indicating that in females additional factors other than testosterone may influence CASM ICa.

Fig. 4.

Relationship between serum testosterone and CASM Ca2+ current (ICa). Group data for peak ICa density (from Fig. 3B) plotted as a function of serum testosterone (from Fig. 1). In males, linear regression demonstrated a significant direct relationship between serum testosterone and CASM ICa (r2 = 0.99; P < 0.05). Data for IF did not fall on the regression line. Data are means ± SE.

Testosterone, sex, and Cav1.2 expression.

Whole cell ICa is the product of the number (n), unitary conductance, and open probability of active channels. Thus the increase in ICa in CASM due to testosterone could result from either an increase in channel activity, an increase in the number of active channels in the membrane, secondary to increased channel synthesis or both. To determine whether testosterone affects channel synthesis, we examined the influence of sex and testosterone on Cav1.2 mRNA (Fig. 5). Males were found to have ∼50% greater Cav1.2 mRNA compared with intact females. Castration markedly reduced Cav1.2 mRNA levels in males, whereas testosterone prevented the loss of mRNA due to castration. Similarly, Cav1.2 protein levels were greater in males compared with females (Fig. 6A). This sex difference was eliminated by castration and prevented by testosterone. These in vivo effects of testosterone could be replicated in vitro. Coronary arterial segments were incubated in phenol-free RPMI culture media in a 37°C, humidified culture chamber for 18 h in the presence or absence of testosterone or dihydrotestosterone (DHT). Both testosterone and DHT produced similar increases in Cav1.2 protein (Fig. 6B). In addition, both testosterone (100 nM) and DHT (100 nM) produced an approximately twofold increase in Cav1.2 mRNA (2.0 ± 0.4 and 2.5 ± 0.5, respectively; P < 0.05). Dihydrotestosterone is a nonaromatizable androgen produced by conversion from testosterone by 5-α reductase that binds the AR. The similar effect of testosterone and DHT on Cav1.2 protein levels indicates that aromatization of testosterone to estrogen is not necessary for the observed effects of testosterone.

Fig. 5.

Testosterone increases coronary voltage-gated Ca2+ channel 1.2 (Cav1.2) mRNA. A: amplified products from quantitative RT-PCR of Cav1.2 in right coronary artery segments. bp, 123-bp ladder; RT, reverse transcriptase. B: relative Cav1.2 gene expression expressed as Embedded Image, where the threshold cycle difference (ΔCt) was determined as CtCav1.2 − Ct18S for each sample and normalized to average ΔCt for IM. Cav1.2 mRNA was greater in IM compared IF. Castration dramatically reduced Cav1.2 mRNA levels in males, an effect prevented by testosterone. Number of animals per group: IM (n = 6), CM (n = 7), CMT (n = 8), and IF (n = 6). *P < 0.05 vs. all other groups; #P < 0.05 vs. IM.

Fig. 6.

Testosterone increases coronary Cav1.2 protein. A: in vivo effects of densitometric values for each sample were quantified and normalized to the average IM value obtained from the same blot. Inset: representative immunoblot showing right coronary artery membrane fractions from control and HC animals probed with anti-Cav1.2 antibody. Cav1.2-positive band appeared at ∼220 kDa. Equal quantities of crude membrane protein (20 μg) were run in each lane. IM (n = 6), CM (n = 8), CMT (n = 6), and IF (n = 7). *P < 0.05 vs. IM and CMT. B: in vitro effects of 0, 10, and 100 nM testosterone (T; n = 6) and dihydrotestosterone (DHT; n = 5) on coronary Cav1.2 protein levels. Coronary arterial segments were incubated for 18 h in the absence or presence of hormones. *P < 0.05 vs. control (0 nM).

Similar to previous findings (19, 25), immunohistochemistry demonstrated robust AR expression of smooth muscle and endothelium in coronary arteries (Fig. 7A). Interestingly, staining was less intense in castrated males compared with other groups indicating a potential of testosterone to regulate coronary AR expression. Importantly, the AR antagonist cyproterone acetate completely blocks the effect of testosterone (Fig. 7B). Together, these data indicate that endogenous testosterone increases both Cav1.2 mRNA and protein, supporting increased channel synthesis as a mechanism for increasing ICa.

Fig. 7.

Testosterone increases coronary Cav1.2 protein via androgen receptor. A: immunohistochemistry demonstrating androgen receptor expression in CASM and endothelium of porcine coronary arteries from IM, CM, CMT, and IF. B: densitometric analysis of Cav1.2 protein levels in control (C) and testosterone-treated (100 nM) coronary arteries in the absence and presence of androgen receptor antagonist cyproterone (Cy; 1 μM). Coronary arterial segments were incubated in vitro for 18 h (n = 3 per group). *P < 0.05 vs. control.


The present study provides the first evidence that endogenous testosterone regulates the expression of voltage-gated Ca2+ channels in CASM. Specifically, endogenous testosterone increased Cav1.2 mRNA, protein, and ICa,L in CASM. In addition, these data provide a mechanism for greater CASM ICa in males compared with females (3).

Although numerous studies have examined acute effects of sex hormones on vascular function (9, 11, 3840, 44), there is a paucity of information regarding long-term vascular effects of testosterone. Khalil and colleagues (9, 10, 34) provided convincing, yet indirect, evidence that gender differences in vascular reactivity between males and females is due to sex differences in VGCC activity. Crews et al. (10) reported depolarization-induced contraction and Ca2+ influx was greater in thoracic aortas of male rats compared with females. This difference was eliminated by ovariectomy in females and restored with estrogen replacement. Importantly, castration in male rats had no effect. On the basis of these data, it was concluded that estrogen is the primary sex hormone modulating depolarization-induced Ca2+ influx in vascular smooth muscle, in part by reducing VGCC activity. In rodents, this conclusion is supported by reports of increased expression of L-type VGCCs in estrogen-receptor null mice (21). However, our data clearly indicate that in a large mammal model, endogenous testosterone is a potent stimulator of L-type VGCC expression and activity in CASM. Endogenous testosterone increased L-type VGCC mRNA, protein and channel activity, which contributes substantially to sex differences in CASM ICa,L (3). Elimination of gonadal testosterone production in males by castration reduced endogenous serum testosterone, CASM ICa, and Cav1.2 protein to levels similar to that in females, an effect that was prevented by exogenous testosterone. This is in agreement with other large mammal studies, where KCl responses of left anteriod descending coronary artery segments were increased in both male and female swine administered testosterone subcutaneously for 2 wk (12) and a trend for increased KCl-induced contractions in coronary arteries from adult swine compared with prepubescent swine (7). In addition, canine coronary artery contractions to U46619 are increased in males compared with females, a difference eliminated by antiandrogens (24). Thus data obtained in large mammal models support testosterone as a primary mediator of L-type VGCC synthesis in CASM. The long-term genomic effect of circulating testosterone is interesting in light of recent findings that nanomolar levels of testosterone produce acute, rapid and presumably nongenomic inhibition of L-type VGCC activity (38). It is possible that the long-term, genomic upregulation of Cav1.2 is a compensatory mechanism for nongenomic inhibition of L-type channel activity by testosterone. At this point, the net effect of genomic and nongenomic actions of testosterone on calcium influx via L-type Ca2+ channels in vivo is unknown.

While testosterone appears to be a primary determinant of coronary L-VGCC expression and activity in males, factors other than serum testosterone appear to influence L-VGCC activity in females (Fig. 4). This conclusion is based on the finding that, whereas a strong, linear relationship between serum testosterone levels and CASM ICa was observed in males, serum testosterone was not a good predictor of CASM ICa in intact females (Fig. 4). Other factors, e.g., estrogen or progesterone, may influence coronary L-type VGCC activity in females. This is also evident by the Cav1.2 mRNA levels in females, which are higher than predicted based on serum testosterone levels alone.

L-type VGCC are heteromeric complexes minimally composed of three protein subunits, α1, α2/δ, and β. Cav1.2 forms the channel pore (20) and contains the binding sites for dihydropyridine antagonists (42). Testosterone increased Cav1.2 mRNA and protein levels suggesting a stimulation of translational and transcriptional activity of the Cav1.2 gene. The 5′-untranslated region of the α-subunit of the L-type VGCC (Cav1.2) gene contains a putative hormone response element and is activated by testosterone in both cardiac and smooth muscle cells in vitro (27). Similarly, castration has been shown to reduce cardiac Cav1.2 mRNA levels, an effect prevented by testosterone treatment (14, 15). Although we did not measure mRNA synthesis or degradation rates, our quantitative RT-PCR data support a role for testosterone in transcriptional regulation of Cav1.2 in CASM. Interestingly, the relative loss of Cav1.2 mRNA in the CM group was greater than that of Cav1.2 protein or ICa. One plausible explanation may be that the loss of mRNA on testosterone removal is rapid, whereas channel protein turnover is much slower. Others have shown dramatic reductions in cardiac Cav1.2 mRNA 2 wk after castration in rats (14). In the present study, animals were studied 5 wk after castration, a time sufficient for a dramatic reduction in Cav1.2 mRNA. Longer periods of testosterone deficiency may be necessary for channel protein levels to reach a nadir.

Testosterone has been described as exerting both genomic and nongenomic actions. Classic genomic actions are mediated through the intracellular AR, a 110-kDa protein with domains for androgen binding, DNA binding, and transactivation (28). Conversely, rapid, acute, nongenomic actions of testosterone (38) are thought to be mediated by membrane ARs on cell surfaces (28). In addition, testosterone can also be converted to estrogen via vascular aromatase and exert its action through estrogen receptors (33, 35). Vascular smooth muscles, including CASM, express estrogen receptors and ARs (19, 25), a finding confirmed in CASM by this study (Fig. 7A). Thus endogenous testosterone could influence coronary calcium channels through numerous mechanisms. Evidence in this study indicates that testosterone increases Cav1.2 gene expression in CASM through a genomic AR-mediated mechanism. Specifically, 1) testosterone increased Cav1.2 mRNA and protein, 2) the effect of testosterone was blocked by an AR antagonist, and 3) aromatization of testosterone to estrogen was not necessary for these effects as dihydrotestosterone produced responses similar to testosterone.

In conclusion, our data demonstrate that endogenous testosterone is a primary modulator of L-type calcium channel synthesis and activity in CASM. Sex differences in several candidate mechanisms impacting cardiovascular disease have been reported, including nitric oxide synthase content and nitric oxide production (26), smooth muscle proliferation (32), and smooth muscle responsiveness to agonists (34). L-type VGCCs play a central role in regulation of vascular tone (36) and have been proposed to contribute to the incidence and severity of both acute and long-term cardiovascular events, such as vasospasm and coronary heart disease (22, 31). Thus increased expression and activity of L-type VGCCs in coronary arterial smooth muscle by testosterone provides a potential mechanism for sex differences in coronary vasoreactivity and pathology.


This study was supported by a National Aeronautics and Space Administration grant.


The authors are grateful to Cathy Galle, Denise Holiman, April Thompson, Dr. Eric Pope, and Dr. John Dodam for invaluable contributions to this study.


  • 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.


  1. 1.
  2. 3.
  3. 4.
  4. 5.
  5. 6.
  6. 7.
  7. 8.
  8. 9.
  9. 10.
  10. 11.
  11. 12.
  12. 13.
  13. 14.
  14. 15.
  15. 16.
  16. 17.
  17. 18.
  18. 19.
  19. 20.
  20. 21.
  21. 22.
  22. 23.
  23. 24.
  24. 25.
  25. 26.
  26. 27.
  27. 28.
  28. 29.
  29. 30.
  30. 31.
  31. 32.
  32. 33.
  33. 34.
  34. 35.
  35. 36.
  36. 37.
  37. 38.
  38. 39.
  39. 40.
  40. 41.
  41. 42.
  42. 43.
  43. 44.
  44. 45.
View Abstract