Yin Xia 1,2 and Raouf A. Khalil 1 1

Transcription

1 Am J Physiol Heart Circ Physiol 31: H1851 H1865, 16. First published May 3, 16; doi:1.1152/ajpheart nancy-associated adaptations in [Ca 2 ] i -dependent and Ca 2 sensitization mechanisms of venous contraction: implications in pregnancy-related venous disorders Yin Xia 1,2 and Raouf A. Khalil 1 1 Vascular Surgery Research Laboratories, Division of Vascular and Endovascular Surgery, Brigham and Women s Hospital, Harvard Medical School, Boston, Massachusetts; and 2 Department of General Surgery, Fuzhou General Hospital, Fuzhou, Fujian, P. R. China Submitted 9 November 15; accepted in final form 26 April 16 Xia Y, Khalil RA. nancy-associated adaptations in [Ca 2 ] i - dependent and Ca 2 sensitization mechanisms of venous contraction: implications in pregnancy-related venous disorders. Am J Physiol Heart Circ Physiol 31: H1851 H1865, 16. First published May 3, 16; doi:1.1152/ajpheart nancy is associated with significant adaptations in the maternal hemodynamics and arterial circulation, but the changes in the venous mechanisms during pregnancy are less clear. We hypothesized that pregnancy is associated with alterations in venous function, intracellular free Ca 2 concentration ([Ca 2 ] i ), and Ca 2 -dependent mechanisms of venous contraction. Circular segments of inferior vena cava (IVC) from virgin and late pregnant (, day 19) Sprague-Dawley rats were suspended between two hooks, labeled with fura-2, and placed in a cuvet inside a spectrofluorometer for simultaneous measurement of contraction and [Ca 2 ] i (fura-2 / ratio). (96 mm), which stimulates Ca 2 influx, caused less contraction ( vs mg/mg tissue) and smaller increases in [Ca 2 ] i (7.12 vs ) in vs. virgin rat IVC. The -adrenergic receptor agonist phenylephrine (Phe; 1 5 M) caused less contraction ( vs mg/mg tissue) and comparable increases in [Ca 2 ] i ( vs. 9.8) in vs. virgin rat IVC. At increasing extracellular Ca 2 concentrations ([Ca 2 ] e ) (.1,.3,.6, 1, and mm), and Phe induced [Ca 2 ] e -contraction and [Ca 2 ] e -[Ca 2 ] i curves that were reduced in vs. virgin IVC, supporting reduced Ca 2 entry mechanisms. The [Ca 2 ] e -contraction and [Ca 2 ] e -[Ca 2 ] i curves were used to construct the [Ca 2 ] i - contraction relationship. Despite reduced contraction and [Ca 2 ] i in IVC, the Phe-induced [Ca 2 ] i -contraction relationship was greater than that of and was enhanced in vs. virgin IVC, suggesting parallel activation of Ca 2 -sensitization pathways. The Ca 2 channel blocker diltiazem, protein kinase C (PKC) inhibitor GF-193X, and Rho-kinase (ROCK) inhibitor Y27632 inhibited - and Phe-induced contraction and abolished the shift in the Phe [Ca 2 ] i -contraction relationship in IVC, suggesting an interplay between the decrease in Ca 2 influx and possible compensatory activation of PKC- and ROCK-mediated Ca 2 -sensitization pathways. The reduced [Ca 2 ] i and [Ca 2 ] i -dependent contraction in rat IVC, despite the parallel rescue activation of Ca 2 -sensitization pathways, suggests that the observed reduction in [Ca 2 ] i -dependent contraction mechanisms is likely underestimated, and that the veins without the rescue Ca 2 -sensitization pathways could be even more prone to dilation during pregnancy. These pregnancy-associated reductions in Ca 2 entry-dependent mechanisms of venous contraction, if occurring in human lower extremity veins and if not adequately compensated by Ca 2 -sensitization pathways, may play a role in pregnancy-related venous disorders. Address for reprint requests and other correspondence: R. A. Khalil, Harvard Medical School, Brigham and Women s Hospital, Division of Vascular Surgery, 75 Francis St., Boston, MA 2115 ( raouf_khalil@hms. harvard.edu). vascular smooth muscle; vein; contraction; calcium NEW & NOTEWORTHY This study demonstrates significant reduction in venous function and Ca 2 entry-dependent mechanisms of venous contraction during pregnancy. The pregnancy-associated reduction in Ca 2 -dependent mechanisms of venous contraction make the veins more prone to venodilation and may play a role in pregnancy-related venous insufficiency disorders. THE VENOUS SYSTEM IS COMPOSED of an intricate network of peripheral and conduit veins, as well as superficial and deep veins. The combined work of unidirectional valves, skeletal muscle contraction, and vein wall constriction ensures antegrade blood flow toward the heart and thereby maintain adequate venous return and cardiac output (51, 69). Vein dysfunction could lead to venous disorders that range in severity from unsightly lower extremity varicose veins to chronic venous insufficiency, deep vein thrombosis, and venous thromboembolism (69). In addition to risk factors such as diabetes, obesity, smoking, and age (3, 76), women may show greater incidence of varicose veins and chronic venous insufficiency than men (3, 5, 8, 82). Also, our experimental studies showed decreased contraction in female vs. male rat inferior vena cava (IVC) (7, 93). Among females, some premenopausal women may suffer from pelvic congestion syndrome, a poorly understood disorder of the pelvic venous circulation characterized by pelvic varicosities and pain worsened by prolonged standing, coitus, and menstruation (7). Importantly, the severity of pelvic congestion syndrome (7) and the risk of varicose veins (2, 3, 8, 83, 96), superficial and deep vein thrombosis, and venous thromboembolism increase during the course of pregnancy (9, 39, 5, 66, 92). In addition to blood hypercoagulability (33), changes in the vein wall mechanics and venous stasis (83) could play a role in pregnancy-related venous disorders. To understand the mechanisms involved in pregnancy-associated venous disorders, it is imperative to understand the cellular mechanisms regulating vascular function during normal pregnancy. Most of our knowledge regarding the cellular mechanisms of vascular contraction and the pregnancy-associated changes in vascular function comes from studies in arteries. Studies in various arterial preparations have suggested that increases in intracellular free Ca 2 concentration ([Ca 2 ] i ) due to initial Ca 2 release from the intracellular stores and maintained Ca 2 entry from the extracellular space are major determinants of /16 Copyright 16 the American Physiological Society H1851

2 H1852 vascular smooth muscle (VSM) contraction (35, 73). In addition, activation of protein kinase C (PKC) (3, 38, 42, ) or Rho kinase (ROCK) may play a role in the regulation of VSM contraction by increasing the [Ca 2 ] i sensitivity of the contractile proteins (28, 29, 45, 47, 84). Several studies have also examined pregnancy-related changes in vascular reactivity in different arterial preparations from various species (15, 17, 31, 36, 41, 53, ). Our laboratory has shown that aortic contraction is reduced in pregnant () compared with virgin rats (15, 36, 41, 87). Our laboratory has also shown that VSM contraction and [Ca 2 ] i are reduced in renal arterial VSM cells of rats (). Also, endothelin-1-induced vasoconstriction and [Ca 2 ] i are reduced in pressurized mesenteric microvessels of compared with virgin rats (53). Although several studies have examined the pregnancy-related changes in arterial function and contraction mechanism, little is known regarding the changes in venous function during pregnancy. Also, while some studies have examined the mechanisms of VSM contraction in male and female virgin rat veins (67, 68, 72, 93), whether these mechanisms are altered during pregnancy is unclear. The present study was designed to test the hypothesis that venous tissue reactivity is altered during pregnancy, and that the pregnancy-associated changes in venous function reflect underlying changes in [Ca 2 ] i and the Ca 2 - dependent and/or Ca 2 -sensitization mechanisms of venous contraction. METHODS Animals. Age-matched (12 wk) virgin ( g) and (day 19) Sprague-Dawley rats ( g) (Charles River Laboratories, Wilmington, MA) were maintained on ad libitum standard rat chow and tap water in 12:12-h light-dark cycle. rats were studied during estrus to control for reproductive and endocrine confounders. Estrous cycle was determined by taking a vaginal lavage with a Pasteur pipette (55). An estrus lavage primarily contained anucleated cornified squamous cells and was confirmed before all experiments (95). All procedures followed the National Institutes of Health s Guide for the Care of Laboratory Animal Welfare Act and the American Physiological Society and were reviewed and approved by the Harvard Medical Area Standing Committee on Animals. Tissue preparation. On the day of the experiment, rats were euthanized by inhalation of CO 2. The IVC was rapidly excised, placed in Krebs solution, carefully dissected and cleaned of connective tissue under microscopic visualization, and portioned into 5-mm circular segments. One segment of vena cava was studied from each rat. The IVC is very thin and delicate, and extreme care was taken throughout the tissue isolation and dissection to minimize tissue injury. Isometric contraction. Segments of IVC were mounted between two home-designed cuvet-adaptable stainless-steel hooks: one hook was fixed at the bottom of the cuvet, and the other hook was connected to a Grass force transducer (FT3, Astro-Med, West Warwick, RI). Vein segments were stretched under of basal tension and allowed to equilibrate for 45 min in a water-jacketed cuvet filled with 3 ml Krebs solution bubbled with 95% O 2/5% CO 2 at 37 C. Our laboratory has previously constructed the relationship between basal tension and contraction to 96 mm in rat IVC and showed that.5-g tension produced maximal contraction, and further increases in basal tension did not cause any further increases in contraction (71). We also performed basal tension- contraction curve on vena cavae from rats and found that.5-g resting tension gave maximal contraction, and further increases in resting tension did not cause further increases in contraction. The bathing Krebs solution was changed every 1 min. The changes in isometric contraction were recorded using a Grass Polygraph D.C. Driver Amplifier and displayed on a paper chart recorder. To improve the quality of the representative contraction tracings, the recorded contractile responses were scanned, the background chart paper grid lines were removed, the graphs were digitized using UNSCANIT software (Silk Scientific, Orem, UT), and the time course was proportionately scaled to match the time course of the [Ca 2 ] i response. In experiments where the IVC showed an oscillatory contractile response, a line was drawn through the middle of the oscillations to measure the mean steady-state response. Simultaneous measurement of [Ca 2 ] i. The IVC inside the cuvet was placed in a double-excitation double-emission Fluorolog-3 spectrofluorometer (Instruments, Jobin Yvon-SPEX, Edison, NJ). After autofluorescence was measured, the IVC was incubated in Krebs solution containing the cell-permeable Ca 2 indicator fura-2 AM (5 M), bovine serum albumin (3 mg/ml), and the mild detergent cremophor el (.25%) for 3 h, as previously described (43, 81, 93). The IVC was washed three times in Krebs to remove extracellular fura-2 AM and incubated in Krebs for an additional 3 min to allow deesterification of the trapped intracellular fura-2 AM into the Ca 2 - sensitive fura-2. The fura-2-loaded IVC was excited alternately at and nm, and the emitted light was collected at 51 nm every 2 s using DataMax software (Instruments). The -to--nm ratio (/) was calculated and represented the changes in [Ca 2 ] i. The changes in fura-2 / ratio are sensitive to changes in [Ca 2 ] i in the nanomolar range (89). The signal-to-noise ratio was improved by averaging five consecutive / ratio readings. Experimental protocols. IVC segments were first stimulated with 96 mm, and the simultaneous changes in contraction and / (indicative of [Ca 2 ] i) were recorded. Once maximum contraction was reached (within 1 min), the IVC was washed three times in Krebs, 1 min each. IVC was then stimulated with phenylephrine (Phe; 1 5 M), and the simultaneous changes in contraction and [Ca 2 ] i were recorded for 1 min. To investigate pregnancy-associated changes in the Ca 2 release mechanism from the intracellular stores, IVC segments were incubated in Ca 2 -free (2 mm EGTA) Krebs for 5 min and then in nominally Ca 2 Krebs for 5 min and were then stimulated with Phe (1 5 M), and the changes in contraction and [Ca 2 ] i were measured. To investigate pregnancy-associated changes in Ca 2 entry, tissues were first stimulated with Ca 2 96 mm solution or with Phe 1 5 MinCa 2 Krebs, then increasing extracellular CaCl 2 concentrations ([Ca 2 ] e) (.1,.3,.6, 1, mm) were added, and the [Ca 2 ] e-contraction curve and [Ca 2 ] e-[ca 2 ] i curve were constructed. To test for pregnancy-associated changes in the [Ca 2 ] i sensitivity of the contractile proteins, the - and Phe-induced [Ca 2 ] e-contraction and [Ca 2 ] e-[ca 2 ] i curves were used to construct the [Ca 2 ] i-contraction relationship in IVC of virgin and rats. To test the contribution of Ca 2 entry, PKC, and ROCK to pregnancy-associated changes in IVC contraction, the effects of the Ca 2 channel blocker diltiazem (1 5 M), PKC inhibitor GF- 193X (1 5 M), and ROCK inhibitor Y27632 (1 5 M) on and Phe-induced contraction, [Ca 2 ] i, and [Ca 2 ] i-contraction relationship were tested in IVC from virgin and rats. The selected concentrations of diltiazem, GF-193X, and Y27632 were based on their reported IC 5 and previous studies demonstrating that these concentrations were specific and efficacious in inhibiting Ca 2 influx (21, 22), PKC (24, 88), and ROCK (32, 64), respectively. Solutions, drugs, and chemicals. Normal Krebs solution contained the following (in mm): NaCl 1, 5.9, NaHCO 3 25, NaH 2PO 4, dextrose 1, CaCl 2, and MgCl 2. Krebs was bubbled with 95% O 2 and 5% CO 2 for 3 min, at an adjusted ph 7.4. For nominally Ca 2 Krebs, CaCl 2 was omitted. For Ca 2 -free Krebs, CaCl 2 was omitted, and 2 mm EGTA (Sigma) was added. (96 mm) was prepared as normal Krebs but with equimolar substitution of NaCl with. Stock solutions of Phe (1 1 M, Sigma), diltiazem (1 2 M, EMD Millipore), and Y (1 2 M, EMD Millipore) were

3 H1853 prepared in deionized water. Stock solution of GF-193X (1 2 M, Enzo Life Sciences) was prepared in dimethyl sulfoxide. The final concentration of dimethyl sulfoxide in experimental solution was.1%. All other chemicals were of reagent grade or better. Statistical analysis. Experiments were conducted on one IVC segment from each rat, and cumulative data from different rats were presented as means SE, with the n value representing the number of rats. Data were first analyzed using ANOVA. When a statistical difference was observed, the data were further analyzed using Student-Newman-Keuls post hoc test for multiple comparisons and Student s t-test for comparison of two means. Differences were considered statistically significant if P.5. To facilitate comparison of the [Ca 2 ] i-contraction relationship in IVC of vs. virgin rats, data points were best fitted using three-order nonlinear polynomial regression and.3 units trend line forward prediction (Microsoft Excel). RESULTS In rat IVC loaded with fura-2 and incubated in normal Krebs ( mm Ca 2 ), the basal [Ca 2 ] i corresponded to a / ratio of , which was significantly less than that in virgin rat IVC (8.4, P.48). Effect of. High is generally known to cause membrane depolarization and to induce Ca 2 influx in VSM (44, 62). High (96 mm) caused significant contraction in IVC of both virgin (Fig. 1A) and rats (Fig. 1B). -induced contraction was rapid in onset, reached steady state within 5 to 7 min, and was maintained for at least 1 min. Also, in IVC of both virgin and rats, caused an increase in the fura-2 -nm fluorescence signal, a decrease in the -nm fluorescence signal (Fig. 1, C and D), and an increase in the / A B C (96 mm) D (96 mm) E F [Ca 2+ ] I (/ Rato) 2.8 Fig. 1. Effect of (96 mm) on contraction and [Ca 2 ] i in virgin and rat IVC. Circular IVC segments isolated from virgin (A, C, and E) and rats (B, D, and F) and loaded with fura-2 were stimulated with (96 mm), and the simultaneous changes in contraction (A and B), - and -nm fluorescence signals (C and D), and / ratio as representative of [Ca 2 ] i (E and F) were recorded. Cumulative time course graphs represent means SE of -induced contraction (G) and [Ca 2 ] i measurements (H) in IVC segments from virgin and rats (n 18 25). Curve is significantly different (P.5) in vs. virgin rats. G H -Induced Contraction Time (min) -Induced [Ca 2+ ] I (/ Ratio) Time (min)

4 H1854 ratio (Fig. 1, E and F), indicating simultaneous increase in [Ca 2 ] i during contraction. Cumulative data indicated that caused less contraction ( vs mg/mg tissue) (Fig. 1G) and smaller increases in [Ca 2 ] i (7.12 vs ) (Fig. 1H) in vs. virgin rat IVC, suggesting reduced Ca 2 entry mechanisms of IVC contraction in rats. Effect of Phe. In IVC of virgin rats, the -adrenergic receptor agonist Phe (1 5 M) caused a significant contraction (Fig. 2A) that reached steady state in 5 min. In rat IVC, Phe induced a smaller contraction that reached steady state in 5 min (Fig. 2B). Also, in IVC of virgin and rats, Pheinduced contraction was associated with simultaneous increase in the fura-2 -nm fluorescence signal, a decrease in the -nm fluorescence signal (Fig. 2, C and D), and an increase in the / ratio (Fig. 2, E and F), indicating simultaneous increase in [Ca 2 ] i during Phe-induced contraction. Cumulative data indicated that Phe-induced steady-state contraction was reduced in vs. virgin IVC ( vs mg/mg tissue) (Fig. 2G). Phe-induced steady-state [Ca 2 ] i was not significantly different in vs. virgin rat IVC ( vs. 9.8) (Fig. 2H). -induced [Ca 2 ] e -contraction and [Ca 2 ] e -[Ca 2 ] i curves. In virgin rat IVC incubated in Ca 2 -free (2 mm EGTA) Krebs for 5 min, then nominal Ca 2 Krebs for another 5 min, stimulation with Ca 2 (96 mm) caused minimal contraction ( mg/mg tissue) (Fig. 3A) and little change in the - and -nm fluorescence signal (Fig. 3C) and the / ratio, or [Ca 2 ] i from 8.3 to 3.11 (Fig. 3E). In rat IVC, Ca 2 (96 mm) also caused minimal contraction ( mg/mg tissue) (Fig. 3B) and little change in - and -nm signal (Fig. 3D) and [Ca 2 ] i from 1.11 to 1.12 (Fig. 3F), which were not significantly different from the levels A B Fig. 2. Effect of Phe (1 5 M) on contraction and [Ca 2 ] i in virgin and rat IVC. IVC segments from virgin (A, C, and E) and rats (B, D, and F) were loaded with fura-2 and then stimulated with Phe (1 5 M), and the simultaneous changes in contraction (A and B), - and -nm fluorescence signals (C and D), and the / ratio as indicative of [Ca 2 ] i (E and F) were recorded. Cumulative time course graphs represent means SE of Phe-induced contraction (G) and underlying [Ca 2 ] i (H) in 15 IVC segments from virgin and rats (n 15 ). Curve is significantly different (P.5) in vs. virgin rats. C E G [Ca 2+ ] i (/39 Ratio) Phe (1-5 M) D F H Phe (1-5 M) Phe-Induced Contraction Time (min) Phe-Induced [Ca 2+ ] I (/ Ratio) 2.2 Virgi n Time (min)

5 H1855 A B [Ca 2+ ] e (mm) C (96 mm) D (96 mm) E G -Induced Contraction F H -Induced [Ca 2+ ] i (/ Ratio) Fig. 3. -induced [Ca 2 ] e-contraction and [Ca 2 ] e-[ca 2 ] i curves in virgin and rat IVC. IVC segments from virgin (A, C, and E) and rats (B, D, and F) were incubated in Ca 2 -free (2 mm EGTA) Krebs for 5 min and then nominally Ca 2 Krebs for 5 min. The bathing solution was changed to Ca 2 96 mm, and the contractile response (A and B), - and -nm fluorescence signals (C and D), and / ratio as an estimate of [Ca 2 ] i (E and F) were recorded. Increasing concentrations of extracellular CaCl 2 (.1,.3,.6, 1, mm) were added, and the -induced [Ca 2 ] e-contraction curve (G) and [Ca 2 ] e-[ca 2 ] i curve (H) were constructed and compared in virgin and rat IVC. Values are means SE of measurements in 8 1 IVC segments from virgin and rats (n 8 1). Curve is significantly different (P.5) in vs. virgin rats [Ca 2+ ] e (mm) [Ca 2+ ] e (mm) observed in virgin rat IVC. In virgin rat IVC, stepwise addition of extracellular Ca 2 (.1,.3,.6, 1, mm) caused corresponding increases in -induced contraction (Fig. 3A), simultaneous increases in the -nm signal and decreases in the -nm signal (Fig. 3C), and increases in the / ratio or [Ca 2 ] i (Fig. 3E) that reached a maximum at mm [Ca 2 ] e. In rat IVC, stepwise addition of [Ca 2 ] e caused smaller increases in -induced contraction (Fig. 3B) and changes in the - and -nm fluorescence signals (Fig. 3D) and [Ca 2 ] i (Fig. 3F). Cumulative -induced [Ca 2 ] e -contraction curve (Fig. 3G) and [Ca 2 ] e -[Ca 2 ] i curve (Fig. 3H) were reduced in vs. virgin IVC, supporting pregnancy-associated reduction in Ca 2 entry mechanisms. Phe-induced [Ca 2 ] e -contraction and [Ca 2 ] e -[Ca 2 ] i curves. In virgin rat IVC incubated in Ca 2 -free (2 mm EGTA) Krebs for 5 min and then nominal Ca 2 Krebs for another 5 min, Phe (1 5 M) caused a very small contraction (8.67 mg/mg tissue) (Fig. 4A), no detectable change in the - and -nm fluorescence signals (Fig. 4C), and almost no change in the / ratio or [Ca 2 ] i from 5.15 to 5.13 (Fig. 4E), suggesting little Ca 2 release from the vein intracellular stores. In rat IVC incubated in Ca 2 Krebs, Phe also caused minimal contraction ( mg/mg tissue) (Fig. 4B), little changes in the - and -nm signals (Fig. 4B), and insignificant change in the / ratio or [Ca 2 ] i from to 1..9 (Fig. 4F), which were not significantly different from the levels observed in

6 H1856 A B [Ca 2+ ] e (mm) Fig. 4. Phe-induced [Ca 2 ] e-contraction and [Ca 2 ] e-[ca 2 ] i curves in virgin and rat IVC. IVC segments from virgin (A, C, and E) and rats (B, D, and F) were incubated in Ca 2 -free (2 mm EGTA) Krebs for 5 min and then nominally Ca 2 Krebs for 5 min. The tissues were stimulated with Phe (1 5 M), and the initial contraction (A and B), - and -nm fluorescence signals (C and D), and / ratio as an estimate of [Ca 2 ] i (E and F) were recorded. Increasing concentrations of extracellular CaCl 2 (.1,.3,.6, 1, mm) were added, and the Phe-induced [Ca 2 ] e-contraction curve (G) and [Ca 2 ] e-[ca 2 ] i curve (H) were constructed and compared in virgin and rat IVC. Values are means SE of measurements in 5 7 IVC segments from virgin and rats (n 5 7). Curve is significantly different (P.5) in vs. virgin rats. C E G Phe-Induced Contraction [Ca 2+ ] I (/ Ratio) Phe (1-5 M) Phe (1-5 M) [Ca 2+ ] e (mm) D F H Phe-Induced [Ca 2+ ] I (/ Ratio) 5 3 [Ca 2+ ] I (/39 Ratio) [Ca 2+ ] e (mm) virgin rat IVC, suggesting little contribution and no significant differences in the intracellular Ca 2 release mechanism in vs. virgin IVC. In virgin rat IVC, stepwise addition of extracellular Ca 2 caused corresponding increases in Phe-induced contraction (Fig. 4A), simultaneous increases in the -nm fluorescence signal and decreases in the -nm fluorescence signal (Fig. 4C), and increases in the / ratio or [Ca 2 ] i (Fig. 4E), which reached a maximum at mm [Ca 2 ] e and then started to decline. In rat IVC, stepwise addition of [Ca 2 ] e caused smaller increases in Phe-induced contraction (Fig. 4B) and changes in the fura-2 fluorescence signal particularly at nm (Fig. 4D) and [Ca 2 ] i (Fig. 4F). The cumulative Phe-induced [Ca 2 ] e -contraction curve (Fig. 4G) and [Ca 2 ] e -[Ca 2 ] i curve (Fig. 4H) were reduced in vs. virgin rat IVC. vs. Phe [Ca 2 ] i -contraction relationship. The - and Phe-induced [Ca 2 ] e -contraction and [Ca 2 ] e -[Ca 2 ] i curves illustrated in Figs. 3 and 4 were used to construct the [Ca 2 ] i - venocontraction relationship in IVC of virgin (Fig. 5A) and rats (Fig. 5B). The - and Phe-induced [Ca 2 ] i -contraction relationships almost overlapped in virgin rat IVC (Fig. 5A). In contrast, trend analysis showed that the Phe-induced [Ca 2 ] i -contraction relationship was enhanced and shifted to the left of that of in rat IVC (Fig. 5B), suggesting changes in other mechanisms of venous contraction in addition to [Ca 2 ] i. Effect of Ca 2 channel, PKC, and ROCK inhibitors. To further delineate the mechanisms involved in the pregnancyassociated changes in IVC contraction, we tested the effects of Ca 2 channel blockers and PKC and ROCK inhibitors on and Phe-induced contraction. In IVC of virgin and rats precontracted with, the Ca 2 channel blocker diltiazem (1 5 M) caused rapid inhibition of contraction (Fig. 6, A and B) and decrease in [Ca 2 ] i (Fig. 6, E and F). Cumulative data

7 A Contraction (mg/mg Tissue) 3 Phe B Contraction (mg/mg Tissue) 3 Phe H1857 Fig. 5. [Ca 2 ] i-contraction relationship in virgin and rat IVC. The - and Pheinduced [Ca 2 ] e-contraction and [Ca 2 ] e- [Ca 2 ] i curves illustrated in Figs. 3 and 4 were used to construct and compare the Phe- vs. -induced [Ca 2 ] i-venocontraction relationship in IVC of virgin (A) and rats (B). Values are means SE of measurements in 5 1 IVC segments from virgin and rats (n 5 1). To facilitate comparison, data points were best fitted using third-order nonlinear polynomial regression and.3 units trend line forward prediction (Microsoft Excel). showed that diltiazem caused rapid and complete inhibition of -contraction (Fig. 6I) and significantly greater decrease in [Ca 2 ] i (Fig. 6J) in vs. virgin rats. In IVC precontracted with Phe, diltiazem caused rapid inhibition of contraction (Fig. 6, C and D) with small decrease in [Ca 2 ] i (Fig. 6, G and H) in IVC of virgin and rats. Cumulative data showed that diltiazem caused more rapid and complete inhibition of Phe contraction (Fig. 6K) with greater decrease in [Ca 2 ] i (Fig. 6L) in IVC of vs. virgin rats. We tested the role of PKC and observed that, in IVC precontracted with, the PKC inhibitor GF-193X (1 5 M) caused partial inhibition of contraction (Fig. 7, A, B), with no change in [Ca 2 ] i (Fig. 7, E and F) in IVC of virgin and rats. Cumulative data showed that GF-193X caused Effect of Ca 2+ Channel Blockade Responses Phe Responses A B C D Diltiazem (1-5 M).25 g Diltiazem.25 g Diltiazem.25 g Diltiazem.25 g E 1. (96 mm) Phe (1-5 M) Phe F 1. G 1. H 1. I J K L -Induced Contraction Time in Diltiazem (min) -Induced [Ca 2+ ] i (/ Ratio) Time in Diltiazem (min) Phel-Induced Contraction Time in Diltiazem (min) Phe-Induced [Ca 2+ ] i (/ Ratio) Time in Diltiazem (min) Fig. 6. Effect of the Ca 2 channel blocker diltiazem on - and Phe-induced contraction and [Ca 2 ] i in virgin and rat IVC. IVC segments isolated from virgin (A, C, E, and G) and rats (B, D, F, and H) and loaded with fura-2 were stimulated with (96 mm; A, B, E, and F) or Phe (1 5 M; C, D, G, and H), then diltiazem (1 5 M) was added, and the simultaneous changes in contraction (A D) and / ratio as indicative of [Ca 2 ] i (E H) were recorded. Cumulative time course graphs represent means SE of the effects of diltiazem on - (I and J) and Phe-induced (K and L) contraction (I and K) and [Ca 2 ] i (J and L) in 6 1 IVC segments from vs. virgin rats (n 6 1). Curve is significantly different (P.5) in vs. virgin rats.

8 H1858 Effect of PKC Inhibition Responses Phe Responses A B C D GFX (1-5 M) GFX GFX GFX.25 g.25 g.25 g.25 g E F G H 1. (96 mm) Phe (1-5 M) Phe 1. I J K L Induced Contraction Time in GFX (min) -Induced [Ca 2+ ] i (/ Ratio) Time in GFX (min) Phe-Induced Contraction Time in GFX (min) Phe-Induced [Ca 2+ ] i (/ Ratio) Time in GFX (min) Fig. 7. Effect of the PKC inhibitor GF-193X (GFX) on - and Phe-induced contraction and [Ca 2 ] i in virgin and rat IVC. IVC segments isolated from virgin (A, C, E, and G) and rats (B, D, F, and H) and loaded with fura-2 were stimulated with (96 mm; A, B, E, and F) or Phe (1 5 M; C, D, G, and H), then GF-193X (1 5 M) was added, and the simultaneous changes in contraction (A D) and / ratio as indicative of [Ca 2 ] i (E H) were recorded. Cumulative time course graphs represent means SEM of the effects of GF-193X on - (I and J) and Phe-induced (K and L) contraction (I and K) and [Ca 2 ] i (J and L) in 5 8 IVC segments from vs. virgin rats (n 5 8). similar inhibition of contraction (Fig. 7I), with no significant change in [Ca 2 ] i (Fig. 7J) in vs. virgin rats. In comparison, in Phe precontracted IVC, GF-193X caused complete inhibition of contraction (Fig. 7, C and D) with little change in [Ca 2 ] i (Fig. 7, G and H) in IVC of virgin and rats. Cumulative data showed that GF-193X caused complete inhibition of Phe contraction (Fig. 7K) with no significant change in [Ca 2 ] i (Fig. 7L) in IVC of and virgin rats. We tested the role of ROCK and found that the ROCK inhibitor Y27632 (1 5 M) caused complete inhibition of contraction (Fig. 8, A and B) with almost no change in [Ca 2 ] i (Fig. 8, E and F) in IVC of virgin and rats. Cumulative data showed that Y27632 caused similar inhibition of contraction (Fig. 8I) with no change in [Ca 2 ] i (Fig. 8J) in and virgin rats. Similarly, Y27632 caused rapid inhibition of Phe-induced contraction (Fig. 8, C and D) with little changes in [Ca 2 ] i (Fig. 8, G and H) in IVC of virgin and rats. Cumulative data showed that Y27632 caused complete inhibition of Phe contraction (Fig. 8K) with no change in [Ca 2 ] i (Fig. 8L) in IVC of vs. virgin rats. vs. virgin IVC [Ca 2 ] i -contraction relationship. The - and Phe-induced [Ca 2 ] e -contraction and [Ca 2 ] e - [Ca 2 ] i curves illustrated in Figs. 3 and 4 were also used to compare the [Ca 2 ] i -venocontraction relationship in IVC of vs. virgin rats during activation by (Fig. 9A) and Phe (Fig. 9B). During activation, the [Ca 2 ] i -venocontraction relationship was slightly shifted to the right in vs. virgin rats (Fig. 9A), suggesting that the same amount of [Ca 2 ] i would cause less contraction in vs. virgin rat IVC. In contrast, trend analysis suggests that the Phe [Ca 2 ] i -venocontraction relationship was shifted to the left or enhanced in vs. virgin rats (Fig, 9B), i.e., for the same amount of [Ca 2 ] i, Phe is capable of producing more contraction in vs. virgin rats. The -associated shifts in the or Phe-induced [Ca 2 ] i -venocontraction relationship were abolished in IVC pretreated with the Ca 2 channel blocker diltiazem (Fig. 9, C and D), PKC inhibitor GF-193X (Fig. 9, E and F), or ROCK inhibitor Y27632 (Fig. 9, G and H). DISCUSSION The main findings are as follows. 1) nancy is associated with decreased - and Phe-induced IVC contraction, and [Ca 2 ] i and Ca 2 -dependent contraction. 2) The Phe [Ca 2 ] i -

9 H1859 Effect of ROCK Inhibition Responses Phe Responses A B C D Y27632 (1-5 M) Y27632 Y27632 Y g.25 g.25 g.25 g (96 mm) Phe (1-5 M) Phe E F G H [Ca 2+ ] i Ratio (/) I J K L Induced Contraction Time in Y27632 (min) -Induced [Ca 2+ ] i (/ Ratio) Time in Y27632 (min) Phe-Induced Contraction Time in Y27632 (min) Phe-Induced [Ca 2+ ] i (/ Ratio) Time in Y27632 (min) Fig. 8. Effect of the ROCK inhibitor Y27632 on - and Phe-induced contraction and [Ca 2 ] i in virgin and rat IVC. IVC segments isolated from virgin (A, C, E, and G) and rats (B, D, F, and H) and loaded with fura-2 were stimulated with (96 mm; A, B, E, and F) or Phe (1 5 M; C, D, G, and H), then Y27632 (1 5 M) was added, and the simultaneous changes in contraction (A D) and / ratio as indicative of [Ca 2 ] i (E H) were recorded. Cumulative time course graphs represent means SE of the effects of Y27632 on - (I and J) and Phe-induced (K and L) contraction (I and K) and [Ca 2 ] i (J and L) in 5 8 IVC segments from vs. virgin rats (n 5 8). venocontraction relationship is enhanced compared with that of in but not virgin rats. 3) The [Ca 2 ] i -venocontraction relationship was slightly reduced during activation by but enhanced during activation by Phe in vs. virgin IVC, suggesting hidden increases in Ca 2 sensitivity. 4) IVC contraction appears to involve Ca 2 entry-dependent and PKCand ROCK-mediated Ca 2 sensitization pathways, as inhibitors of these pathways inhibited contraction in IVC of both virgin and rats and abolished the dormant changes in Ca 2 sensitivity in IVC of rats. [Ca 2 ] i is a major determinant of vascular contraction (, 43, 44, 73). Ca 2 binds calmodulin to form a Ca 2 -calmodulin complex, which activates myosin light chain (MLC) kinase, causes MLC phosphorylation, initiates actin-myosin interaction, and produces VSM contraction. Membrane depolarization by high is often used to measure Ca 2 influx through L-type voltage-gated Ca 2 channels (Ca v ) (43, 44). Previous studies have shown that causes simultaneous increases in contraction and [Ca 2 ] i in fura-2-loaded rat aorta (81), male rabbit IVC (43), and male and female virgin rat IVC (93). The present observation that caused parallel increases in contraction and [Ca 2 ] i in rat IVC is consistent with previous reports and suggests activation of Ca 2 entry through voltagegated channels. Although is thought to mainly stimulate Ca 2 entry, we observed a small IVC contraction to in Ca 2 medium. This is likely due to some residual Ca 2 that remains bound to VSM surface membrane, despite prior incubation in Ca 2 -free Krebs. The membrane depolarization by Ca 2 would then allow the residual membrane-bound Ca 2 to enter VSM, where it may activate Ca 2 -induced Ca 2 release from the intracellular stores (, ). However, we could not detect significant increases in IVC [Ca 2 ] i in response to in Ca 2 medium, suggesting that the contribution of these mechanisms is too small to be detected by our detection method. On the other hand, the reduced maintained contraction and [Ca 2 ] i in vs. virgin rat IVC suggest reduction in Ca 2 entry through voltage-gated channels. In numerous arterial preparations, -adrenergic receptor agonists stimulate contraction with underlying changes in [Ca 2 ] i, although the [Ca 2 ] i profile markedly varies (12, 18, 81). For instance, in rat aorta loaded with the fluorescent Ca 2 indicator fura-2, norepinephrine produced parallel increases in contraction and [Ca 2 ] i (81), while in ferret aorta loaded with the luminescent Ca 2 indicator aequorin, Phe-induced contrac-

10 H18 Effect of Ca 2+ Channel Blockade and PKC or ROCK Inhibition on [Ca 2+ ] i -Contraction Relationship Fig. 9. Effect of Ca 2 channel blockade and PKC and ROCK inhibitors on [Ca 2 ] i-contraction relationship in virgin and rat IVC. The - and Phe-induced [Ca 2 ] e-contraction and [Ca 2 ] e-[ca 2 ] i curves illustrated in Figs. 3 and 4 were used to construct and compare the [Ca 2 ] i-venocontraction relationship in control IVC of vs. virgin rat IVC during activation by (A) and Phe (B). In other experiments, the IVC was pretreated with diltiazem (1 5 M; C and D), GF- 193X (GFX; 1 5 M; E and F), or Y27632 (1 5 M; G and H) to determine their effect on the - (A, C, E, and G) and Phe-induced (B, D, F, and H) [Ca 2 ] i-venocontraction relationship in vs. virgin rat IVC. Values are means SE of measurements in 5 1 IVC segments from vs. virgin rats (n 5 1). To facilitate comparison, data points were best fitted using third-order nonlinear polynomial regression and.3 units trend line forward prediction (Microsoft Excel). A C E Contraction Contraction Contraction (mg/mg Tissue) Control Diltiazem GFX B D F Phe Contraction Phe Contraction Phe Contraction (mg/mg Tissue) Phe Control Diltiazem GFX G Contraction 3 Y H Phe Contraction 3 Y Phe tion is associated with a large [Ca 2 ] i transient, followed by a very small maintained increase in [Ca 2 ] i (18). The differences in the [Ca 2 ] i profile could be related to the Ca 2 indicator used, or the thickness of the aortic tissue and inherent difficulties in its loading with Ca 2 indicators. Studies in thinner veins, such as the ferret portal vein and rabbit IVC, have shown Phe-induced maintained contraction and detectable increases in [Ca 2 ] i (43, 59). The present findings in virgin and rat IVC demonstrate that Phe causes measurable increases in contraction and [Ca 2 ] i. In VSM, -adrenergic receptor agonists activate phospholipase C and increase the breakdown of phosphatidylinositol 4,5-bisphosphate into inositol 1,4,5-trisphosphate (IP 3 ) and diacylglycerol (6, 56, 57, 63). IP 3 diffuses into the cytosol and stimulates Ca 2 release from the intracellular stores. The observed Phe-induced small contraction in Ca 2 Krebs is consistent with activation of Ca 2 release from the intracellular stores. Because the intracellular stores are limited, the maintained Phe contraction is largely due to Ca 2 influx from the extracellular space. Although some studies suggest that -ad-

11 renergic receptor agonists stimulate Ca 2 entry through voltage-gated channels (65), other studies have suggested activation of receptor-operated, ligand-gated, store-operated, or transient receptor potential channels (4,, 48, 49, 94). We have previously shown sex differences in - and Phe-induced contraction of rat IVC, with the contraction in females less than that in males (93). The present study showed that both and Phe contraction were reduced in vs. virgin rat IVC. These observations are consistent with ultrasonography studies in pregnant women, which have shown increases in the diameter and area of the maternal IVC as the gestation progresses, reflecting blood volume expansion (78). We examined the potential mechanisms underlying the pregnancy-associated changes in IVC contraction. The pregnancyassociated decrease in IVC contraction may not be due to changes in the intracellular Ca 2 release mechanisms, because Phe-induced contraction in Ca 2 Krebs was not different in vs. virgin IVC. This is consistent with previous reports demonstrating lack of pregnancy-related changes in the Ca 2 release mechanisms in rat aorta (14, 62). The pregnancyassociated decrease in IVC contraction could be due to differences in voltage-gated Ca 2 channels because of the following. 1) caused parallel increases in contraction and [Ca 2 ] i, and both responses were reduced in vs. virgin IVC. 2) The -induced [Ca 2 ] e -contraction and [Ca 2 ] e -[Ca 2 ] i curves where reduced in vs. virgin IVC. 3) The -induced [Ca 2 ] i -contraction relationship in IVC was superimposed on that of virgin IVC. 4) Diltiazem caused rapid and complete inhibition of -contraction and significantly greater decrease in [Ca 2 ] i in vs. virgin rats, suggesting pregnancy-associated increase in sensitivity to Ca 2 channel blockers. The present results in rat IVC are consistent with previous reports that the contraction, Ca 2 influx, and [Ca 2 ] i are reduced in aortic strips and isolated aortic VSM cells of vs. virgin rats (15, ). The reduced Ca 2 entry in rat IVC could be due to decreased amount or activity of voltagegated channels in VSM and should be further examined in future biochemical and electrophysiological studies. In addition to the possible involvement of voltage-gated Ca 2 channels, the pregnancy-associated reduction in IVC contraction may involve differences in other receptor-operated, ligand-gated, store-operated, or transient receptor potential channels because of the following: 1) the pregnancy-related decrease in Phe contraction was associated with decreased [Ca 2 ] i ; and 2) the Phe-induced [Ca 2 ] e -contraction curve and [Ca 2 ] e -[Ca 2 ] i curve were reduced in vs. virgin rat IVC. Interestingly, while Phe contraction was reduced in rat IVC, the Phe [Ca 2 ] i -contraction relationship appeared to be enhanced compared with that of in rats. Assuming that contraction is due to Ca 2 entry through voltagegated channels, the enhanced Phe [Ca 2 ] i -contraction relationship could be due to activation of Ca 2 channels other than voltage-gated channels. Its is likely that other channels with different opening properties could be activated during pregnancy because of the following: 1) diltiazem caused significant inhibition of -induced contraction and [Ca 2 ] i, but caused inhibition of Phe contraction with little change in [Ca 2 ] i in IVC of rat; and 2) the [Ca 2 ] i -contraction relationship during activation by was reduced, while that during activation by Phe was enhanced in compared with virgin rat IVC, and IVC treatment with diltiazem neutralized these H1861 differences. The enhanced Phe [Ca 2 ] i -contraction relationship compared with that of in rat IVC could also be due to differences in Ca 2 handling and buffering mechanisms (11, 9) or other venous contraction pathways, in addition to increases in [Ca 2 ] i. In addition to Ca 2 -calmodulin and MLC kinase, PKC may play a role in the regulation of VSM contraction by increasing the [Ca 2 ] i sensitivity of the contractile proteins (3, 38, 42, ). Our laboratory has previously shown sex-related distinctions in PKC activity and decreased PKC activity in female rat aortic VSM (37). The present data support a role of PKCmediated [Ca 2 ] i sensitization pathway in IVC contraction because of the following. 1) The Phe-induced [Ca 2 ] i -contraction relationship was enhanced compared with that of in rat IVC. 2) The PKC inhibitor GF-193X partially inhibited contraction with no change in [Ca 2 ] i in both virgin and rats. Although mainly stimulates Ca 2 entry, some reports showed that GF-193X partially inhibited contraction in rabbit femoral artery (13). 3) GF- 193X completely inhibited Phe contraction with no change in [Ca 2 ] i in virgin and rats. 4) GF-193X neutralized any differences in the - or Phe-induced [Ca 2 ] i -contraction relationship in vs. virgin rats. PKC is a family of Ca 2 - dependent, Ca 2 -independent, and atypical isoforms. Because both and Phe stimulated Ca 2 entry, the PKC isoforms involved in IVC contraction could be Ca 2 dependent. On the other hand, the complete inhibition of Phe contraction and reversal of the enhanced Phe [Ca 2 ] i -contraction relationship in rat IVC suggest activation of additional Ca 2 -independent PKC isoforms that need to be identified in future experiments. ROCK has also been suggested to play a role in Ca 2 sensitization by inhibiting MLC phosphatase and enhancing MLC phosphorylation and VSM contraction (28, 29, 45, 47, 84). The present study suggests a role of ROCK in IVC contraction that changes during pregnancy because of the following. 1) Y27632 completely inhibited contraction with no change in [Ca 2 ] i in both virgin and rats. 2) Y27632 completely inhibited Phe contraction with no change in [Ca 2 ] i in virgin and rats. 3) Y27632 neutralized any differences in the - or Phe-induced [Ca 2 ] i -contraction relationship in vs. virgin rats. Although is generally thought to mainly stimulate Ca 2 entry, the observation that Y27632 completely inhibited contraction should not be surprising, as several reports suggested that -induced tonic arterial VSM contraction could involve activation of ROCK (), and that contraction in airway smooth muscle might involve translocation of ROCK1 from noncaveolar to caveolar regions in the plasmalemma (85). Studies have also shown that Y27632 and other ROCK inhibitors inhibit -induced contraction, Ca 2 influx, and MLC phosphorylation in rat aorta and mesenteric artery (25, 97), rat tail artery (58), rabbit femoral artery (13), and porcine airway smooth muscle (34). ROCK is a family of Ca 2 -dependent and Ca 2 -independent isoforms (46, 79, 91). While Ca 2 -dependent ROCK could contribute to contraction, additional perhaps Ca 2 -independent ROCK could play a role in Phe contraction and the enhanced Phe-induced [Ca 2 ] i -contraction relationship during pregnancy, and the changes in ROCK isoforms need to be further investigated in future experiments.

12 H1862 The observation that Ca 2 channel blockade as well as PKC and ROCK inhibition inhibited - and Phe-induced contraction and abolished the shift in the Phe [Ca 2 ] i -contraction relationship in IVC suggests interplay between the decreased Ca 2 influx and possible compensatory activation of PKC- and ROCK-mediated Ca 2 -sensitization pathways. The reduced [Ca 2 ] i and [Ca 2 ] i -dependent contraction in rat IVC, despite the latent rescue activation of Ca 2 -sensitization pathways, suggests that the observed reduced [Ca 2 ] i -dependent contraction mechanisms are likely underestimated, and that, in the absence of these rescue Ca 2 -sensitization pathways, the veins could be even more prone to dilation during pregnancy. An important question is what causes the changes in Ca 2 entry-dependent mechanisms of venous contraction during pregnancy. nancy is associated with increased plasma levels of sex hormones, such as estrogen (E 2 ) and progesterone (23, 75). Our laboratory has previously shown that the expression of E 2 receptor (ER)-, ER-, and G protein-coupled ER is greater in female than male rats IVC, and that E 2 and ER subtype-specific agonists cause concentration-dependent relaxation of female rat IVC (7). Our laboratory and others have also shown increased E 2 -mediated vasodilation and ER-, ER-, and G protein-coupled ER expression in uterine and systemic arteries of pregnant rat and sheep (1, 16, 52, 77). Our laboratory has also shown that E 2 and ER-mediated relaxation largely involve inhibition of Ca 2 entry from the extracellular space (52, 54, 61, 74). Whether E 2 and ER-mediated vasodilation and inhibition of Ca 2 entry are augmented in veins during pregnancy should be examined in future studies. Limitations. Other points that need to be considered are as follows. 1) The present study was performed on endotheliumintact IVC, and the pregnancy-associated changes in Phe and contraction and [Ca 2 ] i can be due to effects of endogenous E 2 on both endothelium-dependent and -independent mechanisms. Although the rat IVC is a very thin and delicate tissue, our laboratory has been able to preserve the endothelium during the dissection procedure and to measure significant acetylcholine-induced relaxation (7). Our attempts to remove the endothelium by rubbing the vein interior around the tip of forceps or with filter paper showed that these harsh methods significantly injure the VSM layer and make the responses to Phe or almost undetectable. Another approach to minimize the role of endothelium is to block endothelium-derived NO, prostacyclin, and hyperpolarizing factor using a cocktail of N G -nitro-l-arginine methyl ester, indomethacin, and K channel blockers, respectively. Future studies should examine whether such a multifaceted intervention would alter the Phe and contractions in rat veins, suggesting a role of endothelium-derived factors or that the responses would be similar in veins treated with blockers of endothelium-derived vasodilators and nontreated veins, suggesting mainly endotheliumindependent mechanisms. 2) The enhanced Phe [Ca 2 ] i -contraction relationship during pregnancy and the abolishment of this enhancement by PKC and ROCK inhibitors support activation of [Ca 2 ] i sensitization pathways. Future studies in -toxin permeabilized IVC would allow manipulation of intracellular Ca 2 levels and further test for changes in Ca 2 sensitivity of the contractile myofilaments in pregnant compared with virgin rat veins. 3) The present study was performed on rat IVC. Although the rat is a four-legged animal, the rat is a consistent breed, and studies on rat veins can avoid the variability related to age, body weight, and other confounding factors that are encountered in studies on human veins. Also, our laboratory s studies on the rat IVC have produced consistent contractile response (71, 72), making it a reliable model to perform mechanistic studies on venous tissue. Whether similar pregnancy-associated changes in the mechanisms of venous contraction occur in other veins, such as the iliac and femoral veins, needs to be examined. 4) The gestational period in rats is 21 days. We selected day 19 rats to demonstrate the maximal changes in venous function just before parturition. Future studies should examine the time course of the changes in venous function at different stages of pregnancy. Future studies should also test whether the pregnancy-associated changes in venous function are reversed completely to virgin levels postpregnancy. 5) Multiparity or repeated pregnancy is associated with increased risk of cardiovascular disease. Repeatedly bred rats show increases in the pressor response to Phe and to acute stress partly due to changes in tone of the splanchnic arterial vasculature (19). Multiparity could also affect venous tone and compliance. Compared with virgin rats, repeatedly bred rats show greater increase in mean arterial pressure after volume loading, greater mean circulatory filling pressure (an index of venomotor tone) in response to norepinephrine, and less pressure-induced increases in passive diameter and greater reactivity to norepinephrine in isolated mesenteric veins. These observations have led to the suggestion that repeated pregnancy could cause reduction in splanchnic venous compliance and augment splanchnic venous reactivity and sympathetic control of venous tone. This could, in turn, compromise the ability of the capacitance venous system to accommodate volume overload or changes in cardiac preload (19). Therefore, the effects of multiple pregnancies on the Ca 2 -dependent and Ca 2 -sensitization mechanisms of venous contraction should be examined in future studies. Perspective. The venous compartment plays an important role in the regulation of venous return and cardiac output, maintains the balance between circulating and noncirculating blood volumes, and regulates the amount of reserve blood stored in the splanchnic venous bed (27). nancy is associated with hemodynamic adaptations and several changes in the venous circulation. Doppler sonography studies have shown gestational adaptations in the maternal venous compartment (27), and ultrasonography studies in pregnant women have shown increases in the size of the maternal IVC as gestation progresses (78). Prolonged relaxation of venous wall tone and increased lower extremity venous pressure could lead to spider veins (telangiectasias) and varicose veins. nancy is also associated with a hypercoagulable state and, coupled with the relative stasis resulting from pelvic venous compression by the uterus and the decreased venous tone, could increase the risk of deep vein thrombosis (83). The present study demonstrated that pregnancy is associated with changes in venous function and decreased Ca 2 entry mechanisms of venous contraction. The pregnancy-associated decrease in venous contraction Ca 2 entry mechanisms, if also occurring in human lower extremity veins, could render the veins more prone to dilation and cause an increase in lower extremity venous pressure, leading to a recalcitrant cycle, progressive venous dilation, and varicose vein formation. Latent PKC- and ROCK-mediated [Ca 2 ] i -sensitization pathway appear to com-