Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The portal vein functions as the main transfer vessel from the digestive organs to the liver. To effect highly efficient transfer, the portal vein exhibits spontaneous, intermittent rhythmic contractions (Miwa et al., 1997). Dysfunction of this tissue leads not only to reduction in blood supply to the liver but also to serious diseases such as varix mediated by reflux of blood (MacMathuna, 1992). An understanding of the basic mechanisms modulating these spontaneous contractions is a prerequisite for investigation of potential pathophysiological processes. Calcium and potassium channel activity is associated with spontaneous contraction (Loirand et al., 1990; Helliwell and Large, 1997). Regulation of calcium channels is believed to occur via plasma membrane potential and/or intracellular signaling systems. However, the mechanisms involved in increased mechanical activity mediated by receptor stimulation and their related intracellular factors are poorly understood.

We have previously reported that receptor-mediated contractile responses in vascular smooth muscle are altered in diabetes and the mechanisms involved change in phosphatidylinositol (PI) turnover (Nobe et al., 2002). Phosphatidylinositols, diacylglycerol (DG), protein kinase C (PKC), and DG kinase are key elements of PI turnover. Increased contractility in many smooth muscle cell types is dependent on acceleration of PI turnover (Abdel-Latif, 2001). An association with intracellular calcium metabolism was also suggested in experimental diabetic models. It might be mediated by an alteration of calcium influx and/or calcium release from stores (Abebe et al., 1990; Pieper and Gross, 1990; Hattori et al., 1994). In diabetes, it is thought that the change in blood glucose level enhances PI turnover activity (Legan, 1989; Somlyo et al., 1999), which may lead to both high blood pressure and dysfunction in many types of tissue (Begum et al., 2000). Blood glucose levels in the portal vein are directly influenced by food intake and/or lifestyle; consequently, we predicted that spontaneous contraction involves PI turnover as a specific intracellular signaling system. Moreover, the mechanism is influenced by changes in blood glucose level in diabetes. To test this hypothesis, we measured key PI turnover-related parameters associated with TXA₂ analog (U46619)-induced contractile responses, under both normal and high-glucose (HG) conditions.

We determined that U46619-induced changes in the spontaneous contractions were mediated by activation of PI turnover. Furthermore, enhancement of contraction under HG conditions involved acceleration of PI turnover. Both endogenous DG levels and DG kinase activity play important roles in basal and HG-altered portal vein contractility.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Eight-week-old ddY mice (Saitama Jikken Co., Saitama, Japan) were housed at constant room temperature (20 ± 2°C) with a 12-h light/dark cycle. As typical noninsulin-dependent diabetic mellitus model of mice, ob/ob-mice [C57BL/6J-(/)] were purchased from Nippon Clea Co. (Tokyo, Japan). Blood glucose levels were determined with a Tidex glucose analyzer (Bayer-Sankyo, Tokyo, Japan).

Vessel Preparation. Mice were sacrificed with ether. Portal veins were dissected and prepared for analysis as described previously (Lalli et al., 1997). Briefly, vessels were rinsed in cold bicarbonate-buffered physiological salt solution (PSS); additionally, loose fat and connective tissue were removed. PSS contained 188 mmol/l NaCl, 4.73 mmol/l KCl, 1.2 mmol/l MgSO₄, 0.025 mmol/l EDTA, 1.2 mmol/l KH₂PO₄, 2.5 mmol/l CaCl₂, and 11.0 mmol/l glucose (buffering was achieved with 25.0 mmol/l NaHCO₃; pH was 7.4 when bubbled with 95% O₂, 5% CO₂ at 37°C). The efficiency of endothelium removal via this method was confirmed histologically as described previously (Liu et al., 1997) as well as by demonstration of the loss of endothelium-dependent relaxation to acetylcholine. Endothelium removal did not significantly affect the amount of force generated in response to norepinephrine (NE) administration (data not shown).

Portal Vein Force Measurements. Portal veins were mounted on a hook attached to an isometric force transducer (NEC San-ei Instruments Ltd., Tokyo, Japan). Optimal tension was established by adjustment of the length of the vessels to a point where maximum peak-to-peak oscillations of spontaneous isometric contractions were observed. This passive tension was maintained throughout the experiment. Data were obtained using Power Lab hardware and analyzed using Chart Software (AD Instruments Japan, Tokyo, Japan).

DG Kinase Assay. DG kinase activity in vascular tissue was determined by measurement of the accumulation of [³²P]dioctanoyl-phosphatidic acid ([³²P]diC8-PA) from diC8 in radioactive inorganic phosphate ([³²P]Pi) and diC8-prelabeled tissues. Loading of [³²P]Pi and diC8 and reagent treatments were performed as per our previous report (Nobe et al., 1994). Results were expressed as cpm per milligram of wet weight tissue.

Measurement of Total Mass of DG. Isolated tissues were incubated in PSS containing various compounds at 37°C. The total mass of DG in each tissue was measured as described previously (Nobe et al., 1993). Diorelin was used as a standard. Results were presented as nanograms per milligram of wet weight tissue.

Measurement of Myo-Inositol Incorporation. Measurement of myo-inositol incorporation was conducted using the method reported by Conrad et al. (1991). [³H]Myo-inositol-prelabeled tissues were preincubated in the presence or absence of each reagent for 10 min. Then 100 nM U46619 was subsequently added for 5 min. After termination of the treatment, incorporated [³H]phosphoinositides were analyzed.

Materials. Carrier-free and HCl-free radioactive [³²P]Pi and [³H]myo-inositol were purchased from PerkinElmer Life Sciences (Boston, MA). U46619, tetrodotoxin (TTX), cyclopiazonic acid (CPA), and U73122 were obtained from Sigma-Aldrich (St. Louis, MO). Calphostin C was procured from Calbiochem-Novabiochem (San Diego, CA). SQ29548 was obtained from Cayman Chemicals (Ann Arbor, MI). R59022 was acquired from Janssen Life Science Products (Olen, Belgium). Cochlioquinone A (CA) was extracted and purified from fermented mycelia of Drechslera sacchari as described previously in the laboratory of Mitsubishi Pharma Corp. (Kanagawa, Japan) (Ogawara et al., 1994). All other reagents were of the highest purity and purchased from Sigma-Aldrich except as noted. U46619 was dissolved in ethanol, whereas calphostin C and SQ29548 were dissolved in dimethyl sulfoxide; no effects of vehicle were noted when total vehicle was 0.03% or less.

Data Analysis. Values are displayed as means ± S.E.M. obtained from at least 4 to 16 animals. The significance of differences between the values was assessed by one-way analysis of variance followed by Bonferroni's t test for multiple comparisons.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Alteration of Blood Glucose Levels in Portal Vein and Aorta. It is thought that a portal vein is susceptible to alteration in glucose level. In this study, resting blood glucose level (6 h after fasting) in normal mouse aorta was 97.8 ± 7.5 mg/dl, and the value in portal vein was 172.0 ± 13.4 mg/dl (n = 4) (Fig. 1). After 30 min of feeding (oral glucose trance test), blood glucose levels increased to 124.5 ± 2.8 and 223.8 ± 14.3 mg/dl, respectively. The increase in portal vein was larger than in aorta. As preliminary trials, alteration of the blood glucose levels were examined in typical diabetic mouse models (ob/ob mouse). Resting levels in aorta and portal vein significantly increased compared with control mouse (341.3 ± 17.7 and 397.7 ± 8.5 mg/dl, respectively). After feeding, significant enhancement of blood glucose level was detected only in portal vein (449.3 ± 0.7 mg/dl). From these results, it was indicated that the blood glucose level in portal vein is higher than that in aorta. Enhancement of the level was notable in the values after feeding.

View larger version (54K): [in this window] [in a new window]   Fig. 1.   Blood glucose levels in aorta and portal vein. Local blood at aorta and portal vein were collected from normal (ddY) and diabetic mice (ob/ob diabetic mouse). Blood glucose levels were determined by Tidex glucose analyzer. Each value is the mean ± S.E.M. from at least four independent determinations. *, p < 0.05 versus value in control mouse and #, p < 0.05 versus aorta.

U46619-Induced Contractile Responses in Mouse Portal Vein. Spontaneous phasic contractile responses were observed in mouse portal vein (Fig. 2A). In the nonstimulated resting state, tension of 1.2 millinewtons (mN) was applied as a minimum basal tone. Absolute peak values of the spontaneous phasic contraction involving basal tone were 1.85 ± 0.14 mN (n = 7). This response was maintained for at least 12 h in our organ bath system (normal PSS, pH 7.4 at 37°C). Treatment of U46619 increased the peak value in a concentration-dependent manner. Stable forces were developed 3 min after U46619 addition. Significant increases in the peak response from the resting state were first detected at 3 nM U46619; the maximal value was obtained at 100 nM U46619 (3.17 ± 0.13 mN; n = 7) (Fig. 2B); the EC₅₀ value was 5.8 nM (n = 7). This contractile response returned to resting levels in about 5 min after removal of U46619 by exchanging the bath contents. After the rinse, identical U46619-induced responses could be elicited (data not shown). Changes in phasic contraction were measured under high-glucose conditions in the presence or absence of U46619 (Fig. 2A). Treatment of portal veins with PSS containing 22.2 mM glucose (HG-PSS), for 30 min did not affect the spontaneous contractions. However, the increase in contractility elicited by U46619 was significantly augmented. Under HG conditions, 100 nM U46619 induced a maximal peak value of 3.70 ± 0.11 mN (n = 7) (Fig. 2B); the EC₅₀ value was 2.9 nM (n = 7). In the preliminary measurements, we confirmed that 2-fold HG-PSS effects were submaximal effects and that was not different from results in 3- and 4-fold HG-PSS (data not shown). Moreover, the HG-PSS effects were time-dependent and 30-min pretreatment period was submaximal. To check an effect of osmotic changes under HG-PSS condition, 11.1 mM sucrose was added to normal PSS. However, this sucrose-added (total 22.2 mM; 11.1 mM glucose + 11.1 mM sucrose) PSS did not affect the U46619-induced responses. The spontaneous contraction without U46619 stimulation in normal, HG-PSS, and sucrose-added PSS was 1.05 ± 0.10, 1.51 ± 0.11, and 1.08 ± 0.07 mN (n = 5), respectively. These results indicated that the 22.2 mM HG-PSS treatment for 30 min was sufficient for considering the effect of high glucose without involving osmotic changes.

View larger version (36K): [in this window] [in a new window]   Fig. 2.   Effects of HG-PSS treatment on U46619-induced contractile responses in mouse portal vein. Representative experimental traces showing the effect of U46619 (1-100 nM) on contraction (A). HG-PSS was pretreated 30 min before U46619 stimulation. Concentration-response curves of the U46619-induced response under normal (open) and HG (closed)-PSS conditions were indicated as percentage of maximal peak response in B. Each point is the mean ± S.E.M. from at least six independent determinations. *, p < 0.05 versus response in normal PSS.

View larger version (42K): [in this window] [in a new window]   Fig. 3.   Basal characteristics of mouse portal vein: effects of SQ29548, TTX, calcium-free-PSS, or CPA treatment on U46619-induced contractile responses in normal and HG-PSS. Tissues were preincubated in the presence or absence of TXA2 receptor antagonist (1 µM SQ29548), neurotransmission inhibitor (500 nM TTX), CaCl2-replaced PSS (Ca2+-free PSS) or sarco- and endoplasmic reticulum calcium ATPase inhibitor (1 µM CPA) for 10 min; 100 nM U46619 was subsequently added. Isometric force developments were measured as described under Materials and Methods. Results were expressed as percentage of peak values induced by 100 nM U46619 in normal PSS. Each value represents the mean ± S.E.M. from at least four independent determinations. *, p < 0.05 versus 100 nM U46619 alone. #, p < 0.05 versus normal PSS.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Effect of enhancement of extracellular calcium concentration on U46619-induced contraction in portal vein. Representative experimental traces showing U46619-induced responses under normal, HG-PSS, and 7.5 mM CaCl2-containing PSS (high calcium). Each condition of PSS was pretreated 30 min before the stimulation.

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Effect of endothelium on U46619-induced contractile responses in mouse portal vein. Concentration-response curves of the U46619-induced response in the absence (open) and presence (closed) of endothelium. Contractile responses were indicated as percentage of maximal peak response. Each point is the mean ± S.E.M. from at least four independent determinations.

Assessment of Spontaneous Contractile Responses in Mouse Portal Vein. Three parameters of the response were defined as follows to assess spontaneous contractile responses under the various conditions (Fig. 6).

View larger version (44K): [in this window] [in a new window]   Fig. 6.   Extraction of parameters from U46619-induced spontaneous phasic contraction in mouse portal vein. Three effective parameters were defined in U46619-induced responses. Typical U46619 (1-100 nM)-induced responses (top) and expanded 6 nM U46619-induced responses (middle) were shown. Amplitude, frequency, and ON-time were calculated using equations displayed (bottom).

Amplitude. The "maximum-minimum" value in each phasic contraction was calculated. Results were expressed as an average in a 3- to 5-min window of stabilized response. Results are expressed in millinewtons.

Frequency. The number of contractile events in a 3- to 5-min window of stabilized response was counted. Threshold consisted of 30% of each spontaneous response. Results were expressed in cycles per minute.

ON-Time. Total seconds in excess of 20% of maximal response induced by 100 to 300 nM U46619 in a 3- to 5-min window of stabilized response were counted. Results were expressed as seconds per minute.

View larger version (17K): [in this window] [in a new window]   Fig. 7.   U46619-induced changes in amplitude (A) and correlation between frequency and ON-time (B). Three parameters of U46619-induced dose-dependent increases in phasic contraction were extracted from data in Fig. 2. Values of the three parameters in normal (open circles) and HG (open squares)-PSS were plotted. Each value represents the mean ± S.E.M. from at least 12 independent determinations. *, p < 0.05 versus normal PSS.

Effect of HG-PSS Treatment on Endogenous DG Level. In normal PSS, the endogenous DG level in portal vein was 176.79 ± 6.53 ng/wet weight tissue (n = 5). U46619 (100 nM) induced a 1.5-fold increase in DG levels (Fig. 8). This U46619-induced response was significantly inhibited by pretreatment with phospholipase C inhibitor U73122 (1 µM, 10 min), without affecting resting levels. In HG-PSS, the resting level of DG increased significantly to 316.55 ± 6.78 ng/wet weight tissue (n = 5). DG further increased to 406.39 ± 7.54 ng/wet weight tissue (n = 5) upon stimulation with 100 nM U46619. Treatment with U73122 in HG-PSS did not significantly influence the U46619-induced increases in the endogenous DG levels.

View larger version (48K): [in this window] [in a new window]   Fig. 8.   U46619-induced changes in total mass of DG in mouse portal vein. Tissues were preincubated in the presence or absence of phospholipase C inhibitor (1 µM U73122) for 10 min. U46619 (100 nM) was subsequently added for 5 min. Response was terminated and total mass of DG was quantified as described under Materials and Methods. Results were expressed as nanograms per milligram of wet weight tissue. Each value represents the mean ± S.E.M. from at least five independent determinations. *, p < 0.05 versus resting level. #, p < 0.05 versus 100 nM U46619 alone. , p < 0.05 versus normal PSS.

Effect of PKC Inhibitor on U46619-Induced Contractile Responses. To investigate the relationship between PKC- and U46619-induced contractile responses, we used the PKC inhibitor calphostin C (1 µM). In normal PSS, calphostin C treatment caused only a slight decrease in phasic contraction without affecting the resting level (Fig. 9A). However, the HG-PSS-induced enhancement of the U46619-induced response was significantly inhibited (Fig. 9B). In these responses, the amplitude was not affected by calphostin C treatment in normal PSS (Fig. 9C). Maximal amplitudes were detected at 10 nM U46619 in the presence or absence of calphostin C. These values were 3.11 ± 0.17 and 3.97 ± 0.06 mN, respectively (n = 8). However, the HG-induced enhancement was suppressed. Furthermore, the response remained similar to those values obtained in normal PSS. Maximal amplitudes in the presence or absence of calphostin C were 4.94 ± 0.26 and 9.55 ± 0.44 mN, respectively (n = 8). With respect to frequency and ON-time, both parameters were reduced by calphostin C pretreatment in normal PSS (Fig. 9D). In HG-PSS, both parameters also decreased. Maximal values for frequency and ON-time were 0.30 ± 0.00 cycles/min and 50.04 ± 0.70 s/min for control responses in HG-PSS, respectively (n = 8). Significant differences between normal and HG-PSS in the presence of calphostin C were not detected.

View larger version (44K): [in this window] [in a new window]   Fig. 9.   Effect of protein kinase C inhibitor on U46619-induced contraction in mouse portal vein. Representative experimental traces showing the effect of protein kinase C inhibitor (1 µM calphostin C for 10 min) on U46619-induced contraction in normal (A) and HG (B)-PSS. HG-PSS was pretreated 30 min before U46619 stimulation. Three parameters of U46619-induced contraction were extracted as described in Fig. 6. Changes in amplitude (C) and relation between frequency and ON-time (D) in the presence (closed) or absence (open) of calphostin C under normal (circles) and HG (squares) conditions were plotted. Each value represents the mean ± S.E.M. from at least five independent determinations. *, p < 0.05 versus HG-PSS. #, p < 0.05 versus normal PSS.

Effects of DG Kinase Inhibitors on U46619-Induced DG Kinase Activation and Contraction. DG kinase activity was measured as an accumulation of [³²P]diC8-PA in [³²P]Pi and diC8-prelabeled tissues. In the portal vein, the resting level of DG kinase activity was 28.04 ± 0.74 cpm/mg wet weight tissue (n = 7) in normal PSS (Fig. 10). This activity was significantly increased by treatment with 100 nM U46619 for 5 min. Maximal values were 105.79 ± 11.30 cpm/mg wet weight tissue (n = 7). Under HG-PSS conditions, both resting and 100 nM U46619-treated DG kinase activities were significantly elevated (53.11 ± 4.13 and 182.80 ± 8.29 cpm/mg wet weight tissue, respectively; n = 7). This U46619-induced activation of DG kinase was inhibited by Ca²⁺-free normal and HG-PSS; DG values were 34.28 ± 2.62 and 61.51 ± 6.30 cpm/mg wet weight tissue (n = 7), respectively. There were no significant differences between normal and HG-PSS.

View larger version (39K): [in this window] [in a new window]   Fig. 10.   Effect of Ca2+-free PSS or DG kinase inhibitors on U46619-induced DG kinase activation in mouse portal vein. [32P]- and diC8-loaded tissues were preincubated in the presence or absence of Ca2+-free PSS or DG kinase inhibitors (7 µM R59022 or 7 µM CA) for 10 min. U46619 (100 nM) was subsequently added for 5 min. Reactions were terminated and [32P]diC8-PA accumulation was quantified as described under Materials and Methods. Results were expressed as cpm per milligram of wet weight tissue. Each value represents the mean ± S.E.M. from at least seven independent determinations. *, p < 0.05 versus resting level. #, p < 0.05 versus 100 nM U46619 alone.

View larger version (42K): [in this window] [in a new window]   Fig. 11.   Effect of DG kinase inhibitor on U46619-induced contraction in mouse portal vein. Representative experimental traces showing the effect of DG kinase inhibitor (7 µM CA for 10 min) on U46619-induced contraction in normal (A) and HG (B)-PSS. HG-PSS was pretreated 30 min before U46619 stimulation. Three parameters of U46619-induced contraction were extracted as described in Fig. 6. Changes in amplitude (C) and relation between frequency and ON-time (D) in the presence (closed) or absence (open) of CA under normal (circles) and HG (squares) conditions were plotted. Each value represents the mean ± S.E.M. from at least eight independent determinations. *, p < 0.05 versus HG-PSS. #, p < 0.05 versus normal PSS.

Effects of U46619 Stimulation on [³H]Myo-Inositol Incorporation. We determined the total activity of PI turnover, incorporation of [³H]myo-inositol in mouse portal vein (Fig. 12). Tissues were incubated with [³H]myo-inositol under several conditions for 5 min and intracellular [³H]inositolphospholipids were subsequently analyzed.

View larger version (46K): [in this window] [in a new window]   Fig. 12.   Effect of PKC or DG kinase inhibitor on U46619-induced myo-inositol incorporation in mouse portal vein. [3H]Myo-inositol-loaded tissues were preincubated in the presence or absence of 1 µM calphostin C or 7 µM CA for 10 min. U46619 (100 nM) was then added for 5 min. Reactions were terminated and [3H]phosphoinositides were quantified as described under Materials and Methods. Results were expressed as cpm per milligram of wet weight tissue. Each value represents the mean ± S.E.M. from at least seven independent determinations. *, p < 0.05 versus resting level. #, p < 0.05 versus 100 nM U46619 alone.

Prostaglandin F₂ (PGF₂) and NE-Induced Contractile Responses in Portal Vein. To reveal whether the enhancement of spontaneous contraction in portal vein under the HG condition was unique in the U46619 stimulation, PGF_2a and NE treatments were performed as typical vasocontractile agonists (Fig. 13). After confirm the U46619-induced responses, PGF_2a and NE were treated cumulatively. These agonists induced dose-dependent contraction in portal vein. Although each maximal response was not over the 100 nM U46619-induced responses, patterns of change were similar. Moreover, enhancement effects under HG conditions were also detected.

View larger version (42K): [in this window] [in a new window]   Fig. 13.   PGF2 and NE-induced spontaneous contraction in portal vein. Representative experimental traces showing the PGF2 (A) and NE (B) stimulation under normal and HG-PSS conditions. HG-PSS was pretreated 30 min before the stimulation. U46619 responses were collected as positive responses.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our results indicated the involvement of PI turnover in U46619-induced spontaneous phasic contraction in mouse portal vein. Enhanced contraction under HG conditions was also associated with an acceleration of PI turnover mediated by elevated endogenous DG, PKC and DG kinase activities. It has been reported that treatment with NE (El Sayah et al., 2000), prostaglandin E₁ (Miwa et al., 1997), PGF₂ (Shirane et al., 1992), or N-methyl-D-aspartate (Rossetti et al., 2000) increased spontaneous contraction in rabbit and rat portal vein. These receptor-mediated responses involve calcium influx via calcium channels. Additionally, we detected similar responses induced by PGF₂ and NE in mouse portal vein (Fig. 13). In this study, a TXA₂ analog, U46619, was selected as an agonist (Nobe et al., 2001). U46619-induced elevations in the peak levels of the spontaneous phasic contractions in a dose-dependent manner (Fig. 2, A and B). Significant increases were detected with 1 to 100 nM U46619; moreover, this range was lower than that in NE responses (Fig. 13B). The U46619-induced contractions were detected in the presence or absence of endothelium in portal vein (Fig. 5). However, alteration of PGF_2a-induced aortic contraction mediated by endothelium dysfunction was reported in diabetes (Kamata et al., 1995). To detect an alteration of smooth muscle function in portal vein with out involving the multiple regulation from endothelium, an endothelium-removed portal vein was used in this study.

It was of interest to establish whether U46619-induced responses were influenced by HG conditions. A level 2- to 5-fold greater than typical blood glucose levels (>400 mg/dl) was previously detected in a streptozotocin-induced diabetic rat (insulin-dependent diabetes mellitus model) (Nobe et al., 1998). In addition, the Otsuka Long Evans Tokushima Fatty rat (noninsulin-dependent diabetes mellitus model) also showed a 2-fold increase in glucose level. We also confirmed that the glucose level in portal vein was higher than that in aorta and was susceptible of food intake (Fig. 1). In the U46619-induced response in portal vein, similar alterations under HG conditions were detected in preliminary trials of this study under 2- to 4-fold increase in glucose conditions (data not shown). To minimize effects of increased osmotic, a 2-fold elevation in glucose (22.2 mM) PSS was used as HG condition. Treatment with HG-PSS for 30 min increased U46619-induced peak responses (Fig. 2B). This result indicates the possibility that portal vein contractility is enhanced in diabetes. In contrast, sucrose-added PSS had little effect (data not shown). Therefore, the increased contractile responses in HG-PSS were specific effects of glucose. Because our results and those of others show that these phasic contractions are totally dependent on calcium influx, HG-PSS-enhanced U46619-induced contractions are unlikely to affect intracellular calcium handling. Results of the contraction in high-calcium PSS (Fig. 4) were also supported our consideration.

Quantitation of the phasic contractions was an important consideration. Generally, contractile responses in portal vein have been evaluated using peak values (Maeda et al., 1999). This approach is effective when the baseline force is maintained. However, at levels greater than 10 nM U46619 the baseline increased (Fig. 2A). Moreover, dependent on the U46619 concentration, a reduction of the spontaneous response, as well as increase in baseline, was detected. Thus, we attempted to evaluate U46619-induced contractile responses three parameters (Fig. 6). The amplitude of the spontaneous contraction indicates a volume of blood flow in each phasic contraction, frequency indicates timing of the spontaneous contraction, and ON-time indicates blood pressure at the portal vein via vascular tonus. In normal PSS, the amplitude was transiently increased and the maximal value was detected at 10 nM U46619 (Fig. 7A). At levels in excess of an optimal amplitude, a reduction in blood flow might occur. This pattern of changes was enhanced under HG conditions. Significant increases were detected at most U46619 concentrations (1-100 nM); however, the maximal amplitude was detected at 10 nM. These results indicated HG-PSS sensitized the amplitude without affecting the relation between U46619 and TXA₂ receptor. On the other hand, the correlation between frequency and ON-time changed in a dose-dependent manner, similar to a reverse-clockwise loop (Fig. 7B). This loop indicated that cumulative addition of U46619 preferentially induced increases in frequency. Subsequently, ON-time attained maximal values. Under HG-PSS conditions, significant increases in ON-time were observed; however, frequency was unaffected. The loop of the relationship also shifted to a high ON-time aspect. These results suggested that decreases in amplitude with enhancement of baseline and/or submaximal increase in ON-time value could lead to a reduced blood flow rate with high blood pressure in the portal vein. Moreover, inhibition of the blood flow rate may be further exaggerated in diabetes mediated by HG conditions.

In the present experiments, U46619 induced increases in spontaneous phasic contraction and HG enhanced these effects. We previously reported that an endogenous DG- and DG kinase-mediated alteration of PI turnover was involved in vascular dysfunction in diabetes (Nobe et al., 2002). In this study, we hypothesized that alteration of PI turnover activity was caused by HG-induced enhancement in portal vein contraction. To test this hypothesis, several indispensable factors in PI turnover were evaluated under normal and HG conditions.

DG is an important second messenger in PI turnover functioning as an endogenous PKC activator (Nishizuka, 1995). The DG level was increased by HG-PSS (Fig. 8). The mechanism governing this increase is not clear. However, Inoghchi et al. (2000) indicated that HG induced an increase in glucose incorporation in cultured vascular cells. Similar results were documented in diabetic rat aorta (Sandirasegarane et al., 1994). These reports suggested that the incorporated glucose was converted to DG via a de novo synthesis pathway. Similar changes in endogenous DG levels were detected in the present investigation. In the portal vein, U46619-induced increases in DG levels in normal PSS were significantly inhibited by pretreatment with U73122. However, the inhibitory effect could not be detected under HG conditions. These findings indicated that endogenous DG was derived from phosphatidylinositols by phospholipase C in normal PSS; however, DG was not present in HG-PSS during U46619 stimulation. These results supported the possibility that increases in glucose incorporation and de novo synthesis activity led to enhancement of endogenous DG levels.

Treatment with calphostin C suppressed the HG-induced enhancement of the U46619 response; in contrast, this suppression was absent in normal PSS (Fig. 9, A and B). In the presence of calphostin C, HG-induced increases in amplitude and ON-time were inhibited and attained values observed in normal PSS (Fig. 9, C and D). This indicated that PKC activation was involved in HG-induced increases in contraction. Because prostanoid receptor-mediated aortic vascular smooth muscle contraction involves PKC activity (Heaslip and Sickels, 1989), it is likely that this activation of PKC arises from an increase in endogenous DG levels in HG-PSS.

We next focused on identification of a target of the activated PKC. It is generally accepted that PKC can affect ion channels (Satoh and Sperelakis, 1995), contractile proteins (Buus et al., 1998), calcium sensitizers (Kitazawa et al., 2000), and/or enzymes. Previously, we determined that PKC regulates DG kinase as a feedback mechanism in guinea pig taenia coli (Nobe et al., 1995). We suggested that activated PKC may regulate endogenous DG levels via DG kinase. Direct interaction via phosphorylation between PKC and DG kinase has also been reported (Kanoh et al., 1989). In mouse portal vein, the possibility exists that PKC functions as a DG kinase regulator, leading to activation of PI turnover. To investigate this possibility the effects of DG kinase inhibition on U46619-induced responses in normal and HG-PSS were examined. Two types of DG kinase inhibitors were used, R59022 and CA. R59022 is used in many cell types (Igal et al., 2001; Oprins et al., 2001). Inhibitory effects were established; additionally, nonspecific effects were also noted (Lai and el-Fakahany, 1990).

CA was present in fermented mycelia of Drechslera sacchari in 1994 (Ogawara et al., 1994). The compound exhibited both specific inhibitory effects on DG kinase and sufficient cell permeability. Effective inhibition of DG kinase activity by CA in vascular smooth muscle was also observed (data not shown). In a DG kinase assay, U46619-induced DG kinase activation was detected; moreover, activation was enhanced under HG conditions (Fig. 10). Furthermore, a calcium dependence of DG kinase activation was also confirmed. Activation was concurrent with the changes in contraction (Fig. 2A). These observations suggested that DG kinase-mediated PI turnover plays a role in portal vein contraction. CA suppressed HG-induced enhancement of activity in the portal vein.

In this study, CA demonstrated selectivity in the HG-induced enhancement of DG kinase activation. The bases remain unclear; however, it is likely that CA displays differential selectivity among DG kinase isoforms. At least nine types of DG kinase isoforms have been identified (Ohanian and Ohanian, 2001). In addition, it has been suggested that the different isoforms are expressed in small vessels, e.g., portal vein, rather than in large arteries, e.g., aorta. DG kinase-, -, and - were detected in small arteries. The inhibitory effects of CA to portal vein response may be due to the selectivity of CA to these isoforms. The HG-induced enhancement of contraction (Fig. 2A) and DG kinase activation (Fig. 10) was dependent on calcium concentration, suggesting that a DG kinase isoform inhibited by CA possesses calcium sensitivity. Calcium-binding sites (EF-hand) were detected exclusively in DG kinase- among DG kinase isoforms occurring in small vessels (Yamada et al., 1997); consequently, this finding suggested that CA might display selectivity to the isoform.

In both normal and HG-PSS, U46619-induced increases in contractility were significantly inhibited by CA (Fig. 11, A and B). CA inhibited both amplitude and ON-time (Fig. 11, C and D); however, frequency values were not affected. With CA treatment, all differences in parameters between normal and HG-PSS disappeared. These results indicated that DG kinase is associated with regulation of amplitude and ON-time in portal vein contraction, as well as with the HG-induced enhancement. To confirm that the HG-enhanced of portal vein contraction is mediated by PI turnover accelerated via DG kinase activation, myo-inositol incorporation was measured as the total PI turnover activity (Fig. 12). In resting and U46619-treated tissue, increases in incorporation were detected; moreover, readings were enhanced under HG conditions. These findings indicated that PI turnover was accelerated not only by U46619 stimulation but also by HG-induced increases in endogenous DG, PKC, and DG kinase levels. These results confirmed the acceleration of PI turnover under HG conditions, supporting our hypothesis. In other types of tissue (aorta and mesangial cells), decreased and uninfluenced PI turnover activities in diabetes were reported (Legan, 1989; Seal et al., 1995). These differences were not clearly understood. Although portal vein needs a lot of energy for spontaneous contraction, the glucose metabolism in this tissue may be distinguished from that of other tissue. Although glucose utilization was different, similar contractile responses were detected in not only U46619 or TXA₂ but also in PGF₂ and NE (Fig. 13). There is a possibility that these vasocontractile factors also mediated similar intracellular signaling pathway.

In this investigation, we determined that U46619-induced increases in amplitude and ON-time values of spontaneous phasic contraction in portal vein were enhanced under HG conditions. Our data suggest that the mechanism consists of the following steps: 1) accumulation of endogenous DG by incorporation of glucose and de novo synthesis, 2) accumulated DG activation of PKC, 3) PKC induction of DG kinase activation, and 4) PI turnover acceleration attributable to DG kinase activation. This acceleration may lead to an increase in spontaneous contraction. Under HG conditions such as diabetes, enhancement of spontaneous contraction may induce alteration of blood flow rate in the portal vein.