Twenty-eight male and female Sprague-Dawley rats (8 weeks old, 200-250 g) were used. All animal studies were carried out in a semi-pathogen-free barrier zone at Institute for Laboratory Animal Research in accordance with the procedures outlined in the Guide for the Care and Use of Laboratory Animals.

Male or female rats were randomly allocated into either the control group (n = 15) or diabetic group (n = 13). Diabetes was induced by a single intravenous tail vein injection of 60 mg/kg streptozotocin (STZ) dissolved in citrate buffer at pH 4.5. In the age-matched control groups, the same volume of citrate buffer was injected intravenously. All rats were housed in cages and allowed free access to food and water. Three days thereafter, blood glucose levels were measured from a drop of blood taken from the tail using OneTouch Ultra® Blood Glucose Meter (LifeScan, Inc). Rats with blood glucose >300 mg/dl were considered diabetic. All rats were maintained for 4 weeks without intervention.

Four weeks after STZ or citrate buffer injection, blood glucoses levels and body weights were measured. Body weights and blood glucose levels in normal and diabetic rats are shown in Table 1. Anesthesia was induced with sevoflurane-oxygen in a chamber and maintained with 2 vol% sevoflurane. Following anesthesia, rats were positioned in the supine position and four legs were firmly fixed on a tilt board. The tail artery was cannulated with a 24 G catheter for continuous arterial pressure monitoring. The transducer was positioned at the level of the heart. After measuring mean arterial pressure in the supine position (basal MAP), the table was vertically tilted at 45° angle in a head-up direction and MAP at 10, 30, 60, and 120 seconds after tilting were recorded. Changes in MAP related to tilting were compared at each time point between diabetic and control groups.

After the tilting experiment, the rat was euthanized by cervical dislocation under deep anesthesia and the third branch of mesenteric artery was dissected. Arteries were placed in a Petri dish with ice-cold modified Krebs' solution of the following composition (in mM): NaCl, 119; KCl, 4.7; MgSO₄, 1.17; NaHCO₃, 22; CaCl₂, 1.6; HEPES, 8; KH₂PO₄, 1.18; d-glucose, 5. The artery was cleaned of fat and connective tissues under a microscope (Zeiss, Germany). To exclude the effect of endothelium-derived relaxing factor, the endothelium of the arteries was denuded by passing silk through the lumen. Thereafter, each artery was cut into 3 mm long rings for in vitro contractile studies. The remaining vessels were immediately stored at -80℃ for immunofluorescence staining and western blotting.

Each vascular segment, suspended between two triangle-shape tungsten hooks, was mounted in a 5 ml water-jacketed temperature-controlled tissue bath (Harvard Apparatus, Holliston, USA) containing modified Krebs' solution. The lower and upper hooks were fixed to the bottom of the bath chamber and to a force displacement transducer, respectively (NP 72-4493, Harvard Apparatus, Holliston, USA). The tissue bathing solution was equilibrated and suspended with a mixture of 95% O₂ and 5% CO₂ (pH: 7.38-7.42, pCO2: 34-36 mmHg) and maintained at 37℃. Propranolol (1 × 10^-6 M) and yohimbine (1 × 10^-7 M) were always included in the tissue bathing solution to block β- and α₂-adrenoceptors, respectively. Changes in isometric force were recorded through an A/D converter (Powerlab®, ADInstruments, Colorado Springs, USA) and analyzed by using Chart for Window Software® (ADInstruments, Colorado Springs, USA). Following an initial tension of 2.0 g, which provided the optimum concentration-response relationships to KCl 60 mM, the arterial strips were allowed to equilibrate for 60 min. When the resting tension reached a stable baseline, no attempt was made to adjust tissue tension. After the equilibrium, tissues were primed by the addition of phenylephrine (1 × 10^-6 M) to the organ bath. After a steady-state contraction was achieved, bath contents were replaced with modified Krebs' solution several times. Tissues were then allowed to reach baseline tension and the priming procedure was repeated once more before constructing concentration-response curves. This priming procedure reduces tissue variability in response to repeated agonist administration. In addition, the absence of functional endothelium was also confirmed by the loss of the relaxation response to acethylcholine (1 × 10^-6 M) in the pre-contracted rings with a priming dose of phenylephrine (1 × 10^-6 M) during the priming procedure.

Reproducible cumulative concentration-response curves were constructed by exposing each artery to increasing concentrations of phenlyephrine (1 × 10^-9 to 1 × 10^-4 M at 0.5 log unit increments) until a sustained maximum response (Emax) was observed. Concentration-response curves to phenylephrine were then obtained (each at 60 min apart) in the presence of three selective α₁-adrenoceptor antagonists to each artery and the procedure was repeated with a high concentration of antagonists in the same preparation. The incubation time for selective α₁-adrenoceptor antagonists, prazosin (α₁; 1, 3, and 10 × 10^-9 M), 5-methylurapidil (5-MU; α_1A; 3, 10, and 30 × 10^-8 M), and BMY 7378 (α_1D; 3, 10, and 30 × 10^-8 M) was 30 min. Propranolol (1 × 10^-6 M) and yohimbine (1 × 10^-7 M) were always included in the tissue bath solutions to block β - and α₂-adrenoceptors, respectively.

After removing vessels and cleaning extraneous tissues, Hematoxylin and eosin (H&E) staining was used to assess the morphologic changes. Arteries were cut into 2-mm rings, embedded in OCT compound (Tissue-Tek® OCT™ Compound, Sakura Finetek, Japan) and quickly frozen (-50℃ in isopentane cooled with dry ice). Frozen blocks were cut into 5-µm sections and mounted on a slide. All sections were stored at -80℃ until use. After warming at room temperature for 30 minutes, the tissue sections were fixed in ice cold acetone for 5 minutes. After rinsing the sections in PBS, a blocking solution with 10% normal donkey serum was placed on the sections and then incubated for 30 minutes to block non-specific binding sites of immunoglobulin. After removal of the blocking solution, the tissue section was incubated with 1 : 100-diluted the primary antibody (α_1D-AR cat. no. s.c.-1475, α_1A-AR cat. no. s.c.-1477; Santa Cruz Biotechnology, Santa Cruz, USA) for 60 minutes at room temperature. After washing the primary antibody in PBS, fluoresceinisothiocyanate (FITC)-conjugated donkey anti-goat IgG (1 : 100 dilution, no. s.c.-2024; Santa Cruz Biotechnology) secondary antibody was placed on the sections for 60 minutes in the dark at room temperature. After removing secondary antibody by washing with PBS, mounting media was placed on the slide and a glass coverslip was adhered. To reduce the technical error, staining was performed on as same slide for each type of artery from each group. Three independent investigators observed the expression of α₁-adrenoceptor subtypes of each artery with a fluorescence microscope (Olympus VENOX-AHBT3, Tokyo, Japan).

Isolated vascular segments were homogenized in lysis buffer (1% Nonidet-40, 20 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1 mM Ethylenediaminetetracetic acid [EDTA], 0.1% SDS and 2 mM Phenylmethylsulfonyl fluoride [PMSF]). The sample was centrifuged at 12,000 g for 15 min at 4℃ and the supernatant was extracted. Protein concentrations were determined using the method of Sambrook and Russel [11]. Samples containing 180 ug of protein were boiled for 5 min to extract membrane proteins, loaded and electrophoresed on SDS-Polyacrolamide gels (PAGE), transferred to PVDF and then blocked for 1 h with 5% nonfat dry milk in tris-buffered saline with tween (TBST®; 20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.05% Tween 20) for 1 hr at room temperature. Commercially available polyclonal antibodies to α_1A- and α_1D-adrenergic receptors (α_1A-AR cat. no. s.c.-1477, α_1D-AR cat. no. s.c.-1475; Santa Cruz Biotechnology) were applied to membranes (1 : 500 in TBST) overnight at 4℃. The membranes were washed with TBST® and secondary antibody diluted to 1 : 4,000 (Horse-radish peroxidase conjugated donkey anti-gout IgG; no. s.c.-2020; Santa Cruz Biotechnology) was incubated in room temperature with the membranes. After washing with TBST®, binding of the antibodies was detected with Chemiluminescent (ECL®, Amersham, Little Chalfont, Buckinghamshire, UK).

Data are presented as means ± SD, means ± SEM, number of rats, or number of arterial segments. In each concentration-response curve, the changes in tension related to increasing concentrations of phenylephrine are expressed as the difference of tension (g) between the basal values and the measured value with various concentrations of phenylephrine. Curves were fitted to all the data by non-linear regression using GraphPad Prism software (GraphPAD software, San Diego, USA) to calculate the negative logarithm of the concentration of agonist that provokes a half-maximal response (pEC50 = agonist potency). A pA₂ value and Schild slopes were calculated by the method of Arunlakshana and Schild [12]. Using three different concentrations of each antagonist, the ratios between the half-maximal concentrations (EC₅₀) of phenylephrine obtained with the specific antagonist concentration and that obtained in its absence (concentration ratio, R) were calculated when the maximal amplitudes between concentration-response curves in those two conditions were similar. Data were plotted as negative log-concentrations of the antagonist on the x-axis vs. log (R-1) on y-axis and fitted as a regression line. For calculation purposes, the slope parameter was set to -1.0 when statistically not different from unity. The x-intercept of the fitted regression line was an estimated pA₂. The value of pA₂ presented a logarithmic measurement of the potency of the specific antagonist and also the dose of antagonist that required a 2-fold increase in agonist concentration. Data were compared by two-way ANOVA, followed by unpaired Student t test with the Bonferroni correction for multiple comparisons, to determine which comparisons were statistically significant. Data were considered significantly different at values of P < 0.05.

Diabetic rats showed significantly higher blood glucose concentrations and lower body weight than age-matched normal rats (Table 1, all P < 0.001). After a 45° tilting in the head-up direction, recovery of MAP was significantly slower in diabetic rats than in normal rats and recovery in normal female rats were significantly slower than in normal male rats (Fig. 1, all P < 0.05).

The concentration-response curves for phenylephrine-induced contraction of normal mesenteric artery showed that the Emax and -logEC₅₀ in female rats were significantly lower than those in male rats (Table 2). Diabetic rats had the significantly lower Emax and -logEC₅₀ than normal rats. Further, the Emax in diabetic female group was lower than that in diabetic male group but there was no significant difference in the -logEC₅₀ (Table 2).

Regardless of arterial type, prazocin, 5-MU, and BMY 7378 shifted the concentration-response curve according to phenylephrine to the right in both diabetic and normal rats. In addition, pA₂ values for prazocin, 5-MU, and BMY 7378 for inhibiting the contractile response to phenylephrine decreased by a similar order of magnitude in distal mesenteric arteries from diabetic rats (all P < 0.001, Table 3). In normal female rats, the pA₂ value for BMY 7378 was significantly smaller than in normal male rats.

H & E staining images in male and female mesenteric arteries from normal and streptozotocin-induced diabetic rats showed that diabetic mesenteric arteries were characterized by a decrease in density of muscle fiber at the media of the vessel wall (Fig. 2). Immunofluorescence images of α_1A- and α_1D-adrenoceptor subtypes in mesenteric arteries of diabetic rats were characterized by a decreased fluorescent intensity (Fig. 3 and 4).

The α_1D-adrenoceptor subtype was expressed less in female normal rats than in male normal rats (Fig. 5). In diabetic rats, the expressions of the α_1A- and α_1D-adrenoceptor subtypes were decreased compared with those in the gender-matched normal rats (all P < 0.05). In particular, in female diabetic rats, the expression of alpha_1D - receptors was not detectable above background.