Small Artery Remodeling Depends on Tissue-Type Transglutaminase
http://www.100md.com
《循环研究杂志》
From the Department of Medical Physics (E.N.T.P.B., J.A.E.S., J.P., A.G., T.M.R., O.S., E.V.B.) and Cardiovascular Research Institute Amsterdam, Academic Medical Center, Amsterdam, the Netherlands; and the Department of Pharmacology (C.L.B., L.H.B., M.J.M.), University of Aarhus, Aarhus, Denmark.
Remodeling of small arteries is essential in the long-term regulation of blood pressure and blood flow to specific organs or tissues. A large part of the change in vessel diameter may occur through non–growth-related reorganization of vessel wall components. The hypothesis was tested that tissue-type transglutaminase (tTG), a cross-linking enzyme, contributes to the inward remodeling of small arteries. The in vivo inward remodeling of rat mesenteric arteries, induced by low blood flow, was attenuated by inhibition of tTG. Rat skeletal muscle arteries expressed tTG, as identified by Western blot and immunostaining. In vitro, activation of these arteries with endothelin-1 resulted in inward remodeling, which was blocked by tTG inhibitors. Small arteries obtained from rats and pigs both showed inward remodeling after exposure to exogenous transglutaminase, which was inhibited by addition of a nitric oxide donor. Enhanced expression of tTG, induced by retinoic acid, increased inward remodeling of porcine coronary arteries kept in organ culture for 3 days. The activity of tTG was dependent on pressure. Inhibition of tTG reversed remodeling, causing a substantial increase in vessel diameter. In a collagen gel contraction assay, tTG determined the compaction of collagen by smooth muscle cells. Collectively, these data show that small artery remodeling associated with chronic vasoconstriction depends on tissue-type transglutaminase. This mechanism may reveal a novel therapeutic target for pathologies associated with inward remodeling of the resistance arteries.
Key Words: blood flow tissue transglutaminase vascular remodeling vasoconstriction
Chronic alteration in the hemodynamic profile is associated with arterial remodeling. Both large and small arteries adapt to a reduction in blood flow with a decrease in lumen diameter,1,2 and in several forms of hypertension, the wall-to-lumen ratio of arteries is increased.3,4 Although hypertrophy of the vessel wall may contribute to this remodeling in larger arteries, in resistance arteries, it mainly involves a geometrical reorganization of wall components around a smaller lumen.5 Thus, in essential hypertension, the reduction in lumen size of resistance arteries appears to be eutrophic, ie, without a change in the amount of wall material.3 In the process of inward remodeling, the reorganization of smooth muscle cells, induced by chronic vasoconstriction, may be an early event.6 Whereas inward remodeling is identified as an important risk factor for cardiovascular events,7 the mechanisms that control blood vessel caliber under physiological and pathological conditions are incompletely understood.
Tissue-type transglutaminase (tTG), also called transglutaminase type 2, belongs to a family of enzymes that includes coagulation factor XIII. tTG is ubiquitously expressed and present both within the cells and at the cell surface, where it associates with integrins.8 The enzyme catalyzes the formation of an N (-glutamyl)lysine cross-link, a bond between a glutamine residue and the primary amino group of either a peptide-bound lysine or a polyamine. Many intracellular and extracellular matrix proteins therefore form a substrate. The activity of tTG is dependent on calcium, whereas nitric oxide and GTP inhibit its activity.9,10 The physiological role of the enzyme is still being uncovered, and includes cell adhesion, wound healing, apoptosis, and matrix reorganization.11
Based on its potential to modify matrix proteins by the formation of a chemically and mechanically stable bond, we hypothesized that tTG plays a crucial role in the inward remodeling of resistance arteries. The causal relationship between tTG and remodeling was studied using several experimental models. First, an arterial ligation model2 was used to study remodeling in vivo. Second, pressurized isolated arteries were studied using a previously established organoid culture model.12–14 Third, a collagen contraction model15 was used to directly study cell-matrix interaction. Collectively, the results show that inward remodeling in these models is the result of a combination of vasoconstriction and the cross-linking activity of tTG.
Ligation Model
The ligation model has been described previously.2 This model allows in vivo studies on vascular remodeling in response to changes in blood flow. Ligation of mesenteric arcading arteries creates a situation of low blood flow through one artery and a compensatory increase in blood flow through the adjacent artery, thereby maintaining blood flow to the intestine. Rats, 6 to 8 weeks old (n=8; Harlan Scandinavia Aps, Aller?d, Denmark), were anesthetized with a combination of fentanyl (0.1 mg/kg), fluanisone (2.8 mg/kg), and midazolam (1.4 mg/kg). Postoperative pain was treated with buprenorphine (0.1 mg/kg) twice a day for the first day. Two first-order mesenteric arteries were ligated in each rat. The ligated arteries were separated by 4 arteries, of which the middle two were used as control. The animals were allowed to recover for 2 days, during which they either received two doses of 10 mg/kg cystamine (transglutaminase blocker) a day, or solvent (IP). Animals were then euthanized and arteries were harvested. Passive pressure-diameter relationships were established in calcium-free physiological saline solution, supplemented with 10–4 mol/L papaverine. Measurements from both two control arteries and two ligated arteries from each animal were averaged.
Isolated Vessel Studies
Rats were decapitated, and the cremaster muscle was excised. Small arteries from cremaster muscle were mounted in an organ culture setup as described previously.12 To determine the role of tTG, a number of inhibitors were used with different modes of action: cystamine, a substrate and active-site inhibitor of tTG,16 and either 5-(biotinamido)pentylamide (BPA) or monodansyl cadaverine (cadaverine), competitive inhibitors of tTG.17 In the first series, arteries were stimulated with endothelin-1 (10–8 mol/L) for 20 hours at 50 mm Hg. Cystamine (10–4 mol/L) or 5-(biotinamido)pentylamide (BPA; 10–4 mol/L) were used to inhibit tTG during activation with endothelin-1. In the second series, arteries were kept at 3 mm Hg and exposed to guinea pig liver transglutaminase (100 mU/mL) for 24 hours. Cadaverine (10–4 mol/L) was added to some experiments. The passive pressure-diameter relationship (3 to 120 mm Hg) of the arteries was measured after full dilation with papaverine (10–4 mol/L) on day 0 and day 1. Rat experiments were performed according to Danish legislation.
Yorkshire female farm pigs (20 to 25 kg; from a local farm in Lelystad, the Netherlands) were sedated by intramuscular injection of ketamine (25 mg/kg) and midazolam (1 mg/kg). Pigs were then intubated and ventilated (O2/NO2, 1:2, supplemented with 0.2% isoflurane), while heart rate, PCO2, and PO2 in the ventilation gasses were monitored. After midsternal thoracotomy and exposure of the heart, heparin (0.1 mg/kg) was injected through the central ear vein. The heart was fibrillated with a 9V battery, excised, and placed in cold MOPS buffer (composition in mol/L: NaCl 145, KCl 4.7, NaH2PO4 1.2, MgSO4 1.2, CaCl2 2, 3-(N-morpholino) propane sulfonic acid 3, glucose 5, and pyruvate 2; pH 7.4). Small subepicardial arteries (250 μm) were excised from the right ventricle and mounted in an organoid culture setup as described previously.14 Cannulated arteries were kept in Leibovitz culture medium at 37°C, under conditions of low flow (shear stress <1 dyne/cm2). In the first series, arteries were kept at 3 mm Hg during a 24-hour incubation period. Arteries were left either untreated (control; n=7), incubated with guinea pig liver transglutaminase (100 mU/mL) for 24 hours (n=9), or incubated with a combination of transglutaminase (100 mU/mL) and sodium nitroprusside (10–4 mol/L; n=6). Passive pressure-diameter relationships of the arteries were established before and after the incubation period, under conditions of full dilation with 10–4 mol/L papaverine. In the second series, arteries were kept in organoid culture for 3 days at a pressure level of 60 mm Hg. Arteries were left untreated (control; n=7), kept in the presence of all-trans-Retinoic acid (10–7 mol/L; n=6), or treated with cadaverine (10–4 mol/L; n=6). Both the active diameter during culture and the passive pressure-diameter relationships before and after the incubation period were measured. Arteries were then further processed for RT-PCR or confocal microscopy. Confocal images were made using the same settings for all arteries. Pig experiments were approved by the local experiment committee of the Academic Medical Center, Amsterdam.
RT-PCR
Total RNA was extracted from porcine coronary arteries using Tri Reagent (Sigma). Reverse transcription (RT)–PCRs were performed using the Omniscript RT kit 505111 (Qiagen) and oligo(dT)-15 (Promega). The primers used for the amplification of tTG were as follows: 5'-gaacatgggcagcgactt-3' (forward) and 5'-tccagggagaagttgagcag-3' (reverse). Using the Taq PCR Core kit 201233, the following PCR program was used: 5 minutes at 94°C, followed by 35 cycles of 1 minute 94°C, 1 minute 59°C, 2 minutes 72°C, followed by 10 minutes at 72°C. PCR products were then resolved on a 1.5% agarose gel and stained with ethidium bromide. Sequence analysis of the PCR product showed a positive identification for tTG.
Western Blot
After culture and storage at –80°C arteries were homogenized in lysis buffer, centrifuged, and mixed with gel sample buffer.18 The protease inhibitors were 1% of a cocktail (P 8340) obtained from Sigma-Aldrich. The samples were run on 10% polyacrylamide gels (Bio-Rad) and blotted to polyvinylidene diflouride membranes (Immobilon-P, Millipore) followed by incubation of the membrane in tris-buffered saline with 0.1% Tween20 and 5% fat-free dry-milk. Primary antibodies were mouse monoclonal antibody to tissue transglutaminase (1 μg/mL, Transglutaminase II Ab-2, clone TG100, NeoMarkers) and mouse monoclonal antibody to ?-actin (1:5000, ab6276, Abcam). Horseradish peroxidase conjugated secondary antibody (Santa Cruz Biotechnology) was used at a dilution of 1:4000. Enhanced chemiluminescence reagents (Amersham Biosciences UK) and the Storm860 phosphor imager (Amersham Biosciences UK) were used for visualization.
Immunostaining
A small part of the rat cremaster muscle was dissected free, placed in TISSUE-TEK mounting medium (Sakura Fintek Europe), and frozen in liquid nitrogen. Cross-sections (8 μm) were cut in a cryostat and mounted on SuperFrostPlus slides (Menzel-Gl?ser). The slides were submerged in cold acetone for 10 minutes, dried, and stored at freezing temperatures. Slides were washed in phosphate buffered saline (PBS) or water and incubation with primary antibodies was done in the presence of 1% bovine serum albumin. The slides were pretreated for 10 minutes with 3% hydrogen peroxide to block endogenous peroxidase. The primary antibodies, anti-tissue transglutaminase (2 μg/mL, Transglutaminase II Ab-2, clone TG100, NeoMarkers) and a negative control IgG1 antibody (catalogue number X0931; DAKO), were present during a 30-minute incubation. For further processing the DAKO LSAB2 KIT (catalogue no. K609, DAKO) and the chromogen diaminobezidine were used. The sections were counterstained with a short treatment with Meyer’s hematoxylin solution.
Collagen Gel Contraction
Smooth muscle cells were isolated from porcine coronary artery by digestion with papain and dithioerythritiol, followed by digestion with elastase, collagenase, and soybean trypsin inhibitor. Cells were grown in Leibovitz medium with 10% FCS as described previously.13 Cells from passage 1 to 5 were used for collagen gel contraction experiments. Collagen gels were constituted from solubilized type I collagen (1 mg/mL) in Leibovitz medium, with 10% FCS to promote cell attachment. Each well of a 12-well plate was populated with 105 cells, gently dispersed over the gel. Cells were allowed to contract the collagen gel over a 24-hour period. Some wells were incubated with papaverine (10–4 mol/L), cadaverine (10–4 mol/L), or guinea pig liver transglutaminase (100 mU/mL).
Statistics
Data are represented as mean±SEM, with n indicating the number of independent experiments. To eliminate differences in initial vessel size, diameters were normalized to the maximal diameter of each vessel at 120 mm Hg on day 0. In the ligation experiments, measurements were made only at the end of the experiment. As both ligated and normal flow arteries in treated rats were exposed to cystamine, diameters were normalized to the maximal diameter at 120 mm Hg of the paired, normal-flow artery of the untreated rat. ANOVA or repeated measurements ANOVA followed by Dunnetts t test or a paired Students t test was used to analyze data. Statistical significance was assumed if P<0.05.
Transglutaminase and In Vivo Remodeling
To address the role of tTG in an in vivo model of vascular remodeling, rats were subjected to a blood flow–modifying protocol. After surgical ligation of first order mesenteric arteries, rats were treated with either the tTG inhibitor cystamine or vehicle solution. The ligated and control arteries, exposed to normal flow, were harvested after 2 days of recovery. A pressure-diameter relationship was established in vitro under conditions of full dilation. Comparison of diameters from ligated and control arteries showed inward remodeling associated with low-flow. This reduction was most evident at higher pressure levels. Cystamine did not alter the passive properties of arteries exposed to normal blood flow. However, cystamine treatment significantly attenuated the inward remodeling associated with low flow (Figure 1).
Figure 1. Inhibition of tTG during in vivo remodeling. Pressure-diameter relationship of maximally dilated rat mesenteric arteries. Arteries were exposed to normal flow (NF, n=8), or to low-flow (LF, n=8), obtained 2 days after arterial ligation from nontreated rats (n=4) and rats treated with cystamine (Cyst; n=4). Low-flow is associated with significant inward remodeling at higher pressure levels (P<0.05; P<0.01 of LF vs NF). Cystamine treatment significantly attenuated inward remodeling (#P<0.05 of LF vs LF+cystamine).
Remodeling Induced by Exogenous Transglutaminase
To directly test the effect of transglutaminase on vessel properties, cannulated small arteries from rats and pigs were incubated with guinea pig liver transglutaminase. Control arteries from rat cremaster muscle (169±3 μm; n=18) did not show a change in maximal diameter after 24 hours (Figure 2a). However, a reduction in their maximal diameter was found after 24-hour incubation with transglutaminase. This effect of transglutaminase could be blocked by the competitive inhibitor cadaverine. A similar protocol was applied to porcine coronary arteries (209±15 μm; n=23). Control arteries did not show a change in maximal diameter (Figure 2b). However, transglutaminase induced a reduction in the maximal diameter. Additional experiments with nitroprusside showed that this effect of transglutaminase could be fully inhibited by simultaneous addition of the nitric oxide donor (Figure 2b).
Figure 2. Exogenous transglutaminase induces vascular remodeling in vitro. Diameters were measured at 60 mm Hg after full dilation, before and after exposure to transglutaminase. Arteries were kept at 3 mm Hg when exposed to transglutaminase to prevent remodeling by endogenous mechanisms. Data are shown as the percentage change in maximal diameter. a, Rat cremaster arteries did not show a change in diameter after 24 hours (control; n=6). Arteries exposed to transglutaminase (TG) for 24 hours (n=6) showed a reduction in the maximal diameter. This remodeling was blocked by cadaverine (TG+Cad; n=6). b, Porcine coronary arteries did not show a change in the passive diameter after 24 hours (control; n=7). Exposure to transglutaminase (TG) induced a significant reduction in the maximal diameter (n=9). In the presence of nitroprusside (TG+NP), remodeling is absent (n=7). ,P<0.01, P<0.001.
Remodeling During Organ Culture
Cannulated, pressurized rat cremaster arterioles (162±4 μm; n=19) were activated with endothelin-1 to induce remodeling. Arteries developed spontaneous tone after mounting in the setup. Activation with endothelin-1 induced an additional vasoconstriction, which only partially subsided during the subsequent 20-hour incubation period. The presence of tTG inhibitors cystamine or BPA did not affect vascular tone (Figure 3a). Comparison of the maximal diameter after 20 hours of activation with endothelin-1 revealed inward remodeling: a 6±1% reduction in the maximal diameter, measured at 60 mm Hg, was found. However, vessels incubated with endothelin-1 and either cystamine or BPA showed no inward remodeling (Figure 3b).
Figure 3. Remodeling induced by endothelin-1. a, Rat cremaster arteries developed spontaneous tone in vitro at 60 mm Hg. Pa(0) indicates the normalized passive diameter at day 0; Sp(0), spontaneous tone at day 0; E, endothelin-1 at day 0 and day 1. Endothelin-1 enhanced vasoconstriction during a 20-hour period (n=7). Addition of neither cystamine (n=6) nor 5-(biotinamido)pentylamine (BPA; n=6) affected tone throughout the experiment. b, The maximal diameter at 60 mm Hg was significantly reduced after activation with endothelin-1. Both cystamine and BPA completely prevented this inward remodeling. P<0.001. c, Western blot showing the presence of tTG (top band) and ?-actin (bottom band, 42 kDa). Protein was obtained from single rat cremaster muscle arteries. d, Immunostaining of tTG protein in intact rat cremaster muscle. Positive staining (brown) is observed throughout the arterial wall, including the endothelium, and in the fibrous tissue and capillaries in between skeletal muscle fibers.
The presence of tTG protein in these vessels was studied using Western blot and immunostaining. Tissue-type transglutaminase could be detected in single arteries from cremaster muscle in a Western blot (Figure 3c). Sections of the rat cremaster muscle revealed the presence of tTG protein throughout the vessel wall, with particularly strong staining in the endothelium (Figure 3d). Positive staining was also observed in capillaries and fibrous tissue surrounding muscle fibers.
To manipulate endogenous tTG expression and activity, porcine coronary arteries (n=19) were kept in organoid culture during a 3-day experimental period. Vessels gradually developed an increasing level of vasoconstriction (Figure 4a). To increase the expression of tTG, all-trans-Retinoic acid (10–7 mol/L) was added to the culture medium.19 This concentration increased the level of tTG mRNA 3-fold, as indicated by RT-PCR (Figure 4b). Although retinoic acid did not acutely alter vascular tone, these arteries showed enhanced vasoconstriction as compared with controls during subsequent organoid culture (Figure 4a). Other arteries were incubated with the inhibitor cadaverine, which was added after 24 hours at a concentration of 10–4 mol/L. This compound induced vasodilation, which was maintained during the remaining of the experiment (Figure 4a). Measured under a condition of maximal vasodilation, the pressure-diameter relationship of control arteries showed a tendency to inward remodeling after 3 days of organoid culture (P=0.07; Figure 4c). Arteries that were incubated with retinoic acid showed a pronounced and significant inward remodeling (Figure 4c). In contrast, remodeling was reversed in arteries exposed to cadaverine and substantial outward remodeling was found (Figure 4c). Remodeling occurred without a change in the wall cross sectional area in all groups. At day 3, the cross sectional area was 99±5%, 104±8%, and 100±2% as percentage of day 0 for control, cadaverine, and retinoic acid groups.
Figure 4. Manipulation of endogenous tTG expression and activity. a, Diameter of coronary arteries during culture when left untreated (control; n=7), in the presence of retinoic acid (n=6) or the transglutaminase inhibitor cadaverine (n=6). Retinoic acid causes a small, but significant increase in vasoconstriction; cadaverine causes significant vasodilation during organoid culture. P<0.05, P<0.001. b, Retinoic acid enhances the expression of tTG in coronary artery (P<0.05). Left panel shows GAPDH mRNA; right panel, tTG mRNA. Representative of 4 experiments. c, Pressure-diameter relationship after full dilation. Arteries kept in organoid culture in the presence of retinoic acid show a significant reduction in the maximal diameter from day 0 to 3. Remodeling is reversed to outward remodeling after inhibition of transglutaminase with cadaverine (P<0.001).
Transglutaminase Activity
The activity of tTG in the vessel wall is reflected by the incorporation of fluorescent cadaverine, the competitive inhibitor of tTG used in the present study. After the organoid culture period, segments that were incubated with cadaverine were unmounted and viewed with a confocal microscope (Figure 5a). These arteries were compared with nonpressurized arteries that were kept in a culture dish in the presence of cadaverine for 3 days (Figure 5b) and vessels that were pressurized but not incubated with cadaverine (Figure 5c). Only in the artery that was pressurized and incubated with cadaverine a distinct pattern of small subcellular patches of high fluorescent intensity was observed. Nonpressurized vessels showed only autofluorescence, as is evident from a comparison of Figure 5b with 5c.
Figure 5. Incorporation of cadaverine reflects transglutaminase activity. a, Patches of high fluorescent intensity (green) reflect incorporation of fluorescent cadaverine in a porcine coronary artery undergoing vascular remodeling. b, An unpressurized coronary artery maintained in culture for 3 days incorporates little fluorescent cadaverine (green). c, Green autofluorescence from fibers is observed in an artery cultured in the absence of cadaverine. Ethidium bromide was used to stain nuclei (red). Pictures are representative of 4 experiments. Images were obtained using the same settings of the confocal microscope.
Transglutaminase Determines Collagen Gel Contraction
The collagen gel contraction assay is a model to study matrix reorganization by cultured cells.15 We used this model to more closely study the interaction of smooth muscle cells with the extracellular matrix component collagen. Cultured smooth muscle cells, obtained from porcine coronary artery, were seeded on a collagen gel. The cells compacted the collagen over a 24-hour period. The process of compaction depended on transglutaminase, as addition of transglutaminase markedly enhanced collagen compaction, whereas it was fully inhibited by cadaverine (Figure 6). In addition, the smooth muscle cell contractile machinery is involved, as compaction was also completely inhibited by the vasodilator papaverine (Figure 6).
Figure 6. Collagen contraction by smooth muscle cells. Smooth muscle cells cultured from porcine coronary artery compact a collagen gel (control; n=27). Compaction, shown as the percent surface area of the well occupied by collagen gel after 24 hours is significantly enhanced by transglutaminase (TG; n=8). Compaction is inhibited by both 0.1 mol/L cadaverine (cad; n=10) and the vasodilator 0.1 mol/L papaverine (pap; n=8). P<0.001 as compared with control.
In the present study, we used several approaches to establish the causal relationship between inward remodeling and tTG activity. In a first set of experiments, we demonstrated that inhibition of tTG activity by cystamine blocks the inward remodeling seen after reduction of blood flow in the mesenteric circulation of the rat. The rapid inward remodeling after flow reduction in this model was shown previously.2 A similar relation between experimental flow reduction and inward remodeling was found in carotid arteries.1,20 However, the currently found inhibition of this inward remodeling by a blocker of tTG is novel.
Several blockers of tTG, with different modes of action, were used in the present study and provided consistent evidence for a role of tTG in inward remodeling of small arteries. However, the use of potentially more selective molecular approaches such as siRNA or tTG knockout mice could provide additional evidence for the role of tTG. Interestingly, the lack of tTG expression in tTG knockout mice is not associated with a clear phenotype.21,22 Thus, limited information on tTG knockout mice indicates normal cardiovascular function. This suggests that either tTG is not important during development and normal function or redundant mechanisms operate. It remains to be determined, however, how these mice respond when challenged with a vascular remodeling stimulus.
If inward remodeling of blood vessels depends on tTG, application of this enzyme could stimulate such remodeling. We indeed observed this in rat cremaster and porcine coronary vessels. The specificity of the action of transglutaminase was supported by the inhibition by cadaverine, whereas the inhibiting effect of the NO donor SNP likely represents direct inhibition of this enzyme through nitrosilation.12 It is important to realize that these experiments were done on cannulated vessels kept at low pressure (3 to 4 mm Hg). At this pressure, the arteries are maintained at a small resting diameter. However, they have no intrinsic tendency to remodel at this pressure, as was previously shown in rat cremaster arteries.13 The current study confirmed this in porcine coronary vessels. Thus, in the absence of exogenous transglutaminase, the control vessels had identical pressure-diameter relations before and after culture (Figure 2). The remodeling induced by exogenous transglutaminase became apparent when vessels were pressurized to 60 mm Hg. We suggest that this inward remodeling represents cross-linking of collagen, because at this testing pressure, the diameter of the passive vessels is set by the collagen backbone.
Based on the in vivo effects of tTG inhibition and the in vitro effects of exogenous transglutaminase, we aimed to test whether manipulation of tTG expression or activity would influence the inward remodeling in isolated blood vessels. This was tested in two models using organoid culture: first, cannulated rat cremaster arterioles were maintained at physiological pressure in the presence of ET-1. In this model, we previously reported inward remodeling after 3 days.12 However, in the current study, a 24-hour exposure to ET-1 was sufficient to induce inward remodeling. The full inhibition of inward remodeling by cystamine and BPA in this model indeed supports the role of tTG. In a second model using porcine coronary arteries, involvement of tTG in inward remodeling was supported by the inhibition by cadaverine and the enhanced inward remodeling on upregulation of tTG expression by retinoic acid. Surprisingly, inhibition of tTG not only suppressed the inward remodeling of these vessels but actually resulted in a marked (20%) outward remodeling in three days of organoid culture. These data therefore show that coronary arterial structure is highly dynamic. It remains to be established whether in the in vivo coronary bed similar fast remodeling can occur and whether tTG plays the same dominant role there.
Our previous work12–14 and the available literature, suggest that persistent changes in vascular tone drive remodeling in physiological and pathological conditions. Smooth muscle reorganization during chronic vasoconstriction could be an early event herein.6 Evidence shows that the inward remodeling in hypertension is also related to vascular tone, because vasodilation, but not hypotension, corrects vascular structure.23 In the coronary circulation, inward remodeling is found at sites of local vasospasm,24 or associated with microvascular spasm distal to sites of injured large arteries.25 Based on this association of tone and remodeling, we suggest that inward remodeling requires two conditions: a long-lasting state of vasoconstriction and the activity of cross-linking enzymes to fixate the vessel in this state.
In the current study, remodeling was induced by either reduced blood flow in vivo or vasoconstriction in vitro. Evidence for the role of tTG in remodeling was provided for both models, but direct evidence for a long-lasting state of vasoconstriction was shown only for the in vitro experiments. However, it is likely that prolonged vasoconstriction also was induced in the in vivo model. Both mouse20 and rabbit27 carotid artery show a vasoconstrictor response after a surgically imposed reduction in blood flow, which gradually becomes irreversible. Conversely, the vasodilatory effect of flow is well established in small arteries, including those from the rat mesenteric circulation.26 Therefore, it is to be expected that chronic vasoconstriction is a common feature for both models. Thus, whereas differences with respect to proliferation and apoptosis may exist in different situations, we believe that vasoconstriction and fixation by tTG play a major role in diameter reduction both in vivo and in vitro.
We previously observed that only active vasoconstriction, and not passive reduction in diameter by decreasing the distending pressure of a relaxed vessel, results in remodeling.13 Those data suggested that activity of the cross-linking enzyme is strongly related to smooth muscle activation. Indeed, the enzyme that we now find to drive remodeling, tTG, has this relation: its activity is dependent on pressure, as reflected by the incorporation of fluorescent substrate (Figure 5), and on calcium.9 Interestingly, the vasodilator molecule nitric oxide has been reported to inhibit transglutaminase activity,12 and thus may be a physiologically relevant inhibitor of inward remodeling. We therefore incubated arteries with transglutaminase in the presence of the nitric oxide donor nitroprusside. Indeed, under these conditions exogenous transglutaminase did not alter the passive properties of the arteries.
The role of tTG in the close relationship of tone and remodeling is further substantiated by the results from the organ culture experiments. These results show that increased expression of tTG is associated with enhanced tone and inward remodeling of porcine coronary arteries. Conversely, inhibition of tTG with cadaverine caused vasodilation and reversed the remodeling to substantial outward remodeling. The vasodilatory effect may be explained by the role of tTG as a G-protein in transmembrane signaling through phospholipase C.28 Alternatively, the transamidation reactions catalyzed by tTG, which activate targets such as the monomeric G-protein RhoA,29 followed by activation of Rho-kinase and inhibition of myosin light chain phosphatase may underlie this relationship with vascular tone. The vasoconstriction induced by endothelin-1 in rat cremaster arteries was not affected by the inhibitors cystamine and BPA. This suggests that the role of tTG in vascular tone may differ in various preparations and mechanisms of vasoconstriction.
Although vascular remodeling may be initiated by vascular tone, remodeling is evident from the change in diameter under fully dilated conditions.30 Therefore, smooth muscle activation subsequently needs to be linked to rearrangement of the extracellular matrix. Integrins play an important role in the physical connection and signaling between smooth muscle cells and extracellular matrix. Interestingly, remodeled resistance arteries from spontaneously hypertensive rats show an abnormal expression of specific integrins.31 We recently showed that integrins participate in the process of remodeling in rat resistance arteries.12 Because part of the tTG is localized at the cell surface in association with integrins and fibronectin,8 it can be hypothesized that tTG acts in concert with integrins in the reorganization of matrix components. The collagen gel experiments support this view. We previously found that the reorganization of collagen in this gel depends on integrins and now showed that also tTG plays an essential role in this process. Although collagen is only one of many substrates of tTG, it is a relevant molecule to study in this context, because it is believed that collagen is the major determinant of the passive vessel diameter at higher pressure levels.32,33
In conclusion, a crucial role for tTG in the process of arterial remodeling is uncovered in the present study. This enzyme, that cross-links extracellular matrix proteins, may provide the crucial link between functional and structural modulation of arterial diameter.
References
Langille BL, O’Donnell F. Reductions in arterial diameter produced by chronic decreases in blood flow are endothelium-dependent. Science. 1986; 231: 405–407.
Buus CL, Pourageaud F, Fazzi GE, Janssen G, Mulvany MJ, De Mey JGR. Smooth muscle cell changes during flow-related remodeling of rat mesenteric resistance arteries. Circ Res. 2001; 89: 180–186.
Heagerty AM, Aalkjaer C, Bund SJ, Korsgaard N, Mulvany MJ. Small artery structure in hypertension: dual processes of remodeling and growth. Hypertension. 1993; 21: 391–397.
Intengan HD, Schiffrin EL. Vascular remodeling in hypertension: roles of apoptosis, inflammation, and fibrosis. Hypertension. 2001; 38: 581–587.
Prewitt RL, Rice DC, Dobrian AD: Adaptation of resistance arteries to increases in pressure. Microcirculation. 2003; 9: 295–304.
Martinez-Lemus LA, Hill MA, Bolz SS, Pohl U, Meininger GA. Acute mechanoadaptation of vascular smooth muscle cells in response to continuous arteriolar vasoconstriction: implications for functional remodeling. FASEB J. 2004; 18: 708–710.
Rizzoni D, Porteri E, Boari GE, De CC, Sleiman I, Muiesan ML, Castellano M, Miclini M, Agabiti-Rosei E. Prognostic significance of small-artery structure in hypertension. Circulation. 2003; 108: 2230–2235.
Akimov SS, Krylov D, Fleischman LF, Belkin AM. Tissue transglutaminase is an integrin-binding adhesion coreceptor for fibronectin. J Cell Biol. 2000; 148: 825–838.
Smethurst PA, Griffin M. Measurement of tissue transglutaminase activity in a permeabilized cell system: its regulation by Ca2+ and nucleotides. Biochem J. 1996; 313: 803–808.
Lai TS, Hausladen A, Slaughter TF, Eu JP, Stamler JS, Greenberg CS. Calcium regulates S-nitrosylation, denitrosylation, and activity of tissue transglutaminase. Biochemistry. 2001; 40: 4904–4910.
Lorand L, Graham RM. Transglutaminases: crosslinking enzymes with pleiotropic functions. Nat Rev Mol Cell Biol. 2003; 4: 140–156.
Bakker EN, Buus CL, VanBavel E, Mulvany MJ. Activation of resistance arteries with endothelin-1: from vasoconstriction to functional adaptation and remodeling. J Vasc Res. 2004; 41: 174–182.
Bakker EN, van der Meulen ET, van den Berg BM, Everts V, Spaan JA, VanBavel E. Inward remodeling follows chronic vasoconstriction in isolated resistance arteries. J Vasc Res. 2002; 39: 12–20.
Bakker ENTP, van der Meulen ET, Spaan JAE, VanBavel E. Organoid culture of cannulated rat resistance arteries: effect of serum factors on vasoactivity and remodeling. Am J Physiol. 2000; 278: H1233–H1240.
Guidry C, Grinnell F. Studies on the mechanism of hydrated collagen gel reorganization by human skin fibroblasts. J Cell Sci. 1985; 79: 67–81.
Siefring GEJ, Apostol AB, Velasco PT, Lorand L. Enzymatic basis for the Ca2+-induced cross-linking of membrane proteins in intact human erythrocytes. Biochemistry. 1978; 17: 2598–2604.
Stenberg P, Curtis CG, Wing D, Tong YS, Credo RB, Gray A, Lorand L. Transamidase kinetics. Amide formation in the enzymic reactions of thiol esters with amines. Biochem J. 1975; 147: 153–163.
Eskildsen-Helmond YE, Mulvany MJ. Pressure-induced activation of extracellular signal-regulated kinase 1/2 in small arteries. Hypertension. 2003; 41: 891–897.
Ou H, Haendeler J, Aebly MR, Kelly LA, Cholewa BC, Koike G, Kwitek-Black A, Jacob HJ, Berk BC, Miano JM. Retinoic acid-induced tissue transglutaminase and apoptosis in vascular smooth muscle cells. Circ Res. 2000; 87: 881–887.
Rudic RD, Bucci M, Fulton D, Segal SS, Sessa WC. Temporal events underlying arterial remodeling after chronic flow reduction in mice: correlation of structural changes with a deficit in basal nitric oxide synthesis. Circ Res. 2000; 86: 1160–1166.
De Laurenzi V, Melino G: Gene disruption of tissue transglutaminase. Mol Cell Biol. 2001; 21: 148–155
Nanda N, Iismaa SE, Owens WA, Husain A, Mackay F, Graham RM. Targeted inactivation of Gh/tissue transglutaminase II. J Biol Chem. 2001; 276: 20673–20678.
Christensen KL, Mulvany MJ. Vasodilatation, not hypotension, improves resistance vessel design during treatment of essential hypertension: a literature survey. J Hypertens. 2001; 19: 1001–1006.
Hong MK, Park SW, Lee CW, Ko JY, Kang DH, Song JK, Kim JJ, Mintz GS, Park SJ. Intravascular ultrasound findings of negative arterial remodeling at sites of focal coronary spasm in patients with vasospastic angina. Am Heart J. 2000; 140: 395–401.
Saitoh S, Onogi F, Aikawa K, Muto M, Saito T, Maehara K, Maruyama Y. Multiple endothelial injury in epicardial coronary artery induces downstream microvascular spasm as well as remodeling partly via thromboxane A2. J Am Coll Cardiol. 2001; 37: 308–315.
Sun D, Huang A, Koller A, Kaley G. Adaptation of flow-induced dilation of arterioles to daily exercise. Microvasc Res. 1998; 56: 54–61.
Langille BL, Bendeck MP, Keeley FW. Adaptations of carotid arteries of young and mature rabbits to reduced carotid blood flow. Am J Physiol. 1989; 256: H931–H939.
Im MJ, Russell MA, Feng JF. Transglutaminase II: a new class of GTP-binding protein with new biological functions. Cell Signal. 1997; 9: 477–482.
Singh US, Kunar MT, Kao YL, Baker KM. Role of transglutaminase II in retinoic acid-induced activation of RhoA-associated kinase-2. EMBO J. 2001; 20: 2413–2423.
Bund SJ, Lee RM. Arterial structural changes in hypertension: a consideration of methodology, terminology and functional consequence. J Vasc Res. 2003; 40: 547–557.
Intengan HD, Thibault G, Li JS, Schiffrin EL. Resistance artery mechanics, structure, and extracellular components in spontaneously hypertensive rats: effects of angiotensin receptor antagonism and converting enzyme inhibition. Circulation. 1999; 100: 2267–2275.
Dobrin PB. Mechanical properties of arteries. Physiol Rev. 1978; 58: 397–460.
VanBavel E, Siersma P, Spaan JA. Elasticity of passive blood vessels: a new concept. Am J Physiol. 2003; 285: H1986–H2000.(Erik N.T.P. Bakker, Carst)
Remodeling of small arteries is essential in the long-term regulation of blood pressure and blood flow to specific organs or tissues. A large part of the change in vessel diameter may occur through non–growth-related reorganization of vessel wall components. The hypothesis was tested that tissue-type transglutaminase (tTG), a cross-linking enzyme, contributes to the inward remodeling of small arteries. The in vivo inward remodeling of rat mesenteric arteries, induced by low blood flow, was attenuated by inhibition of tTG. Rat skeletal muscle arteries expressed tTG, as identified by Western blot and immunostaining. In vitro, activation of these arteries with endothelin-1 resulted in inward remodeling, which was blocked by tTG inhibitors. Small arteries obtained from rats and pigs both showed inward remodeling after exposure to exogenous transglutaminase, which was inhibited by addition of a nitric oxide donor. Enhanced expression of tTG, induced by retinoic acid, increased inward remodeling of porcine coronary arteries kept in organ culture for 3 days. The activity of tTG was dependent on pressure. Inhibition of tTG reversed remodeling, causing a substantial increase in vessel diameter. In a collagen gel contraction assay, tTG determined the compaction of collagen by smooth muscle cells. Collectively, these data show that small artery remodeling associated with chronic vasoconstriction depends on tissue-type transglutaminase. This mechanism may reveal a novel therapeutic target for pathologies associated with inward remodeling of the resistance arteries.
Key Words: blood flow tissue transglutaminase vascular remodeling vasoconstriction
Chronic alteration in the hemodynamic profile is associated with arterial remodeling. Both large and small arteries adapt to a reduction in blood flow with a decrease in lumen diameter,1,2 and in several forms of hypertension, the wall-to-lumen ratio of arteries is increased.3,4 Although hypertrophy of the vessel wall may contribute to this remodeling in larger arteries, in resistance arteries, it mainly involves a geometrical reorganization of wall components around a smaller lumen.5 Thus, in essential hypertension, the reduction in lumen size of resistance arteries appears to be eutrophic, ie, without a change in the amount of wall material.3 In the process of inward remodeling, the reorganization of smooth muscle cells, induced by chronic vasoconstriction, may be an early event.6 Whereas inward remodeling is identified as an important risk factor for cardiovascular events,7 the mechanisms that control blood vessel caliber under physiological and pathological conditions are incompletely understood.
Tissue-type transglutaminase (tTG), also called transglutaminase type 2, belongs to a family of enzymes that includes coagulation factor XIII. tTG is ubiquitously expressed and present both within the cells and at the cell surface, where it associates with integrins.8 The enzyme catalyzes the formation of an N (-glutamyl)lysine cross-link, a bond between a glutamine residue and the primary amino group of either a peptide-bound lysine or a polyamine. Many intracellular and extracellular matrix proteins therefore form a substrate. The activity of tTG is dependent on calcium, whereas nitric oxide and GTP inhibit its activity.9,10 The physiological role of the enzyme is still being uncovered, and includes cell adhesion, wound healing, apoptosis, and matrix reorganization.11
Based on its potential to modify matrix proteins by the formation of a chemically and mechanically stable bond, we hypothesized that tTG plays a crucial role in the inward remodeling of resistance arteries. The causal relationship between tTG and remodeling was studied using several experimental models. First, an arterial ligation model2 was used to study remodeling in vivo. Second, pressurized isolated arteries were studied using a previously established organoid culture model.12–14 Third, a collagen contraction model15 was used to directly study cell-matrix interaction. Collectively, the results show that inward remodeling in these models is the result of a combination of vasoconstriction and the cross-linking activity of tTG.
Ligation Model
The ligation model has been described previously.2 This model allows in vivo studies on vascular remodeling in response to changes in blood flow. Ligation of mesenteric arcading arteries creates a situation of low blood flow through one artery and a compensatory increase in blood flow through the adjacent artery, thereby maintaining blood flow to the intestine. Rats, 6 to 8 weeks old (n=8; Harlan Scandinavia Aps, Aller?d, Denmark), were anesthetized with a combination of fentanyl (0.1 mg/kg), fluanisone (2.8 mg/kg), and midazolam (1.4 mg/kg). Postoperative pain was treated with buprenorphine (0.1 mg/kg) twice a day for the first day. Two first-order mesenteric arteries were ligated in each rat. The ligated arteries were separated by 4 arteries, of which the middle two were used as control. The animals were allowed to recover for 2 days, during which they either received two doses of 10 mg/kg cystamine (transglutaminase blocker) a day, or solvent (IP). Animals were then euthanized and arteries were harvested. Passive pressure-diameter relationships were established in calcium-free physiological saline solution, supplemented with 10–4 mol/L papaverine. Measurements from both two control arteries and two ligated arteries from each animal were averaged.
Isolated Vessel Studies
Rats were decapitated, and the cremaster muscle was excised. Small arteries from cremaster muscle were mounted in an organ culture setup as described previously.12 To determine the role of tTG, a number of inhibitors were used with different modes of action: cystamine, a substrate and active-site inhibitor of tTG,16 and either 5-(biotinamido)pentylamide (BPA) or monodansyl cadaverine (cadaverine), competitive inhibitors of tTG.17 In the first series, arteries were stimulated with endothelin-1 (10–8 mol/L) for 20 hours at 50 mm Hg. Cystamine (10–4 mol/L) or 5-(biotinamido)pentylamide (BPA; 10–4 mol/L) were used to inhibit tTG during activation with endothelin-1. In the second series, arteries were kept at 3 mm Hg and exposed to guinea pig liver transglutaminase (100 mU/mL) for 24 hours. Cadaverine (10–4 mol/L) was added to some experiments. The passive pressure-diameter relationship (3 to 120 mm Hg) of the arteries was measured after full dilation with papaverine (10–4 mol/L) on day 0 and day 1. Rat experiments were performed according to Danish legislation.
Yorkshire female farm pigs (20 to 25 kg; from a local farm in Lelystad, the Netherlands) were sedated by intramuscular injection of ketamine (25 mg/kg) and midazolam (1 mg/kg). Pigs were then intubated and ventilated (O2/NO2, 1:2, supplemented with 0.2% isoflurane), while heart rate, PCO2, and PO2 in the ventilation gasses were monitored. After midsternal thoracotomy and exposure of the heart, heparin (0.1 mg/kg) was injected through the central ear vein. The heart was fibrillated with a 9V battery, excised, and placed in cold MOPS buffer (composition in mol/L: NaCl 145, KCl 4.7, NaH2PO4 1.2, MgSO4 1.2, CaCl2 2, 3-(N-morpholino) propane sulfonic acid 3, glucose 5, and pyruvate 2; pH 7.4). Small subepicardial arteries (250 μm) were excised from the right ventricle and mounted in an organoid culture setup as described previously.14 Cannulated arteries were kept in Leibovitz culture medium at 37°C, under conditions of low flow (shear stress <1 dyne/cm2). In the first series, arteries were kept at 3 mm Hg during a 24-hour incubation period. Arteries were left either untreated (control; n=7), incubated with guinea pig liver transglutaminase (100 mU/mL) for 24 hours (n=9), or incubated with a combination of transglutaminase (100 mU/mL) and sodium nitroprusside (10–4 mol/L; n=6). Passive pressure-diameter relationships of the arteries were established before and after the incubation period, under conditions of full dilation with 10–4 mol/L papaverine. In the second series, arteries were kept in organoid culture for 3 days at a pressure level of 60 mm Hg. Arteries were left untreated (control; n=7), kept in the presence of all-trans-Retinoic acid (10–7 mol/L; n=6), or treated with cadaverine (10–4 mol/L; n=6). Both the active diameter during culture and the passive pressure-diameter relationships before and after the incubation period were measured. Arteries were then further processed for RT-PCR or confocal microscopy. Confocal images were made using the same settings for all arteries. Pig experiments were approved by the local experiment committee of the Academic Medical Center, Amsterdam.
RT-PCR
Total RNA was extracted from porcine coronary arteries using Tri Reagent (Sigma). Reverse transcription (RT)–PCRs were performed using the Omniscript RT kit 505111 (Qiagen) and oligo(dT)-15 (Promega). The primers used for the amplification of tTG were as follows: 5'-gaacatgggcagcgactt-3' (forward) and 5'-tccagggagaagttgagcag-3' (reverse). Using the Taq PCR Core kit 201233, the following PCR program was used: 5 minutes at 94°C, followed by 35 cycles of 1 minute 94°C, 1 minute 59°C, 2 minutes 72°C, followed by 10 minutes at 72°C. PCR products were then resolved on a 1.5% agarose gel and stained with ethidium bromide. Sequence analysis of the PCR product showed a positive identification for tTG.
Western Blot
After culture and storage at –80°C arteries were homogenized in lysis buffer, centrifuged, and mixed with gel sample buffer.18 The protease inhibitors were 1% of a cocktail (P 8340) obtained from Sigma-Aldrich. The samples were run on 10% polyacrylamide gels (Bio-Rad) and blotted to polyvinylidene diflouride membranes (Immobilon-P, Millipore) followed by incubation of the membrane in tris-buffered saline with 0.1% Tween20 and 5% fat-free dry-milk. Primary antibodies were mouse monoclonal antibody to tissue transglutaminase (1 μg/mL, Transglutaminase II Ab-2, clone TG100, NeoMarkers) and mouse monoclonal antibody to ?-actin (1:5000, ab6276, Abcam). Horseradish peroxidase conjugated secondary antibody (Santa Cruz Biotechnology) was used at a dilution of 1:4000. Enhanced chemiluminescence reagents (Amersham Biosciences UK) and the Storm860 phosphor imager (Amersham Biosciences UK) were used for visualization.
Immunostaining
A small part of the rat cremaster muscle was dissected free, placed in TISSUE-TEK mounting medium (Sakura Fintek Europe), and frozen in liquid nitrogen. Cross-sections (8 μm) were cut in a cryostat and mounted on SuperFrostPlus slides (Menzel-Gl?ser). The slides were submerged in cold acetone for 10 minutes, dried, and stored at freezing temperatures. Slides were washed in phosphate buffered saline (PBS) or water and incubation with primary antibodies was done in the presence of 1% bovine serum albumin. The slides were pretreated for 10 minutes with 3% hydrogen peroxide to block endogenous peroxidase. The primary antibodies, anti-tissue transglutaminase (2 μg/mL, Transglutaminase II Ab-2, clone TG100, NeoMarkers) and a negative control IgG1 antibody (catalogue number X0931; DAKO), were present during a 30-minute incubation. For further processing the DAKO LSAB2 KIT (catalogue no. K609, DAKO) and the chromogen diaminobezidine were used. The sections were counterstained with a short treatment with Meyer’s hematoxylin solution.
Collagen Gel Contraction
Smooth muscle cells were isolated from porcine coronary artery by digestion with papain and dithioerythritiol, followed by digestion with elastase, collagenase, and soybean trypsin inhibitor. Cells were grown in Leibovitz medium with 10% FCS as described previously.13 Cells from passage 1 to 5 were used for collagen gel contraction experiments. Collagen gels were constituted from solubilized type I collagen (1 mg/mL) in Leibovitz medium, with 10% FCS to promote cell attachment. Each well of a 12-well plate was populated with 105 cells, gently dispersed over the gel. Cells were allowed to contract the collagen gel over a 24-hour period. Some wells were incubated with papaverine (10–4 mol/L), cadaverine (10–4 mol/L), or guinea pig liver transglutaminase (100 mU/mL).
Statistics
Data are represented as mean±SEM, with n indicating the number of independent experiments. To eliminate differences in initial vessel size, diameters were normalized to the maximal diameter of each vessel at 120 mm Hg on day 0. In the ligation experiments, measurements were made only at the end of the experiment. As both ligated and normal flow arteries in treated rats were exposed to cystamine, diameters were normalized to the maximal diameter at 120 mm Hg of the paired, normal-flow artery of the untreated rat. ANOVA or repeated measurements ANOVA followed by Dunnetts t test or a paired Students t test was used to analyze data. Statistical significance was assumed if P<0.05.
Transglutaminase and In Vivo Remodeling
To address the role of tTG in an in vivo model of vascular remodeling, rats were subjected to a blood flow–modifying protocol. After surgical ligation of first order mesenteric arteries, rats were treated with either the tTG inhibitor cystamine or vehicle solution. The ligated and control arteries, exposed to normal flow, were harvested after 2 days of recovery. A pressure-diameter relationship was established in vitro under conditions of full dilation. Comparison of diameters from ligated and control arteries showed inward remodeling associated with low-flow. This reduction was most evident at higher pressure levels. Cystamine did not alter the passive properties of arteries exposed to normal blood flow. However, cystamine treatment significantly attenuated the inward remodeling associated with low flow (Figure 1).
Figure 1. Inhibition of tTG during in vivo remodeling. Pressure-diameter relationship of maximally dilated rat mesenteric arteries. Arteries were exposed to normal flow (NF, n=8), or to low-flow (LF, n=8), obtained 2 days after arterial ligation from nontreated rats (n=4) and rats treated with cystamine (Cyst; n=4). Low-flow is associated with significant inward remodeling at higher pressure levels (P<0.05; P<0.01 of LF vs NF). Cystamine treatment significantly attenuated inward remodeling (#P<0.05 of LF vs LF+cystamine).
Remodeling Induced by Exogenous Transglutaminase
To directly test the effect of transglutaminase on vessel properties, cannulated small arteries from rats and pigs were incubated with guinea pig liver transglutaminase. Control arteries from rat cremaster muscle (169±3 μm; n=18) did not show a change in maximal diameter after 24 hours (Figure 2a). However, a reduction in their maximal diameter was found after 24-hour incubation with transglutaminase. This effect of transglutaminase could be blocked by the competitive inhibitor cadaverine. A similar protocol was applied to porcine coronary arteries (209±15 μm; n=23). Control arteries did not show a change in maximal diameter (Figure 2b). However, transglutaminase induced a reduction in the maximal diameter. Additional experiments with nitroprusside showed that this effect of transglutaminase could be fully inhibited by simultaneous addition of the nitric oxide donor (Figure 2b).
Figure 2. Exogenous transglutaminase induces vascular remodeling in vitro. Diameters were measured at 60 mm Hg after full dilation, before and after exposure to transglutaminase. Arteries were kept at 3 mm Hg when exposed to transglutaminase to prevent remodeling by endogenous mechanisms. Data are shown as the percentage change in maximal diameter. a, Rat cremaster arteries did not show a change in diameter after 24 hours (control; n=6). Arteries exposed to transglutaminase (TG) for 24 hours (n=6) showed a reduction in the maximal diameter. This remodeling was blocked by cadaverine (TG+Cad; n=6). b, Porcine coronary arteries did not show a change in the passive diameter after 24 hours (control; n=7). Exposure to transglutaminase (TG) induced a significant reduction in the maximal diameter (n=9). In the presence of nitroprusside (TG+NP), remodeling is absent (n=7). ,P<0.01, P<0.001.
Remodeling During Organ Culture
Cannulated, pressurized rat cremaster arterioles (162±4 μm; n=19) were activated with endothelin-1 to induce remodeling. Arteries developed spontaneous tone after mounting in the setup. Activation with endothelin-1 induced an additional vasoconstriction, which only partially subsided during the subsequent 20-hour incubation period. The presence of tTG inhibitors cystamine or BPA did not affect vascular tone (Figure 3a). Comparison of the maximal diameter after 20 hours of activation with endothelin-1 revealed inward remodeling: a 6±1% reduction in the maximal diameter, measured at 60 mm Hg, was found. However, vessels incubated with endothelin-1 and either cystamine or BPA showed no inward remodeling (Figure 3b).
Figure 3. Remodeling induced by endothelin-1. a, Rat cremaster arteries developed spontaneous tone in vitro at 60 mm Hg. Pa(0) indicates the normalized passive diameter at day 0; Sp(0), spontaneous tone at day 0; E, endothelin-1 at day 0 and day 1. Endothelin-1 enhanced vasoconstriction during a 20-hour period (n=7). Addition of neither cystamine (n=6) nor 5-(biotinamido)pentylamine (BPA; n=6) affected tone throughout the experiment. b, The maximal diameter at 60 mm Hg was significantly reduced after activation with endothelin-1. Both cystamine and BPA completely prevented this inward remodeling. P<0.001. c, Western blot showing the presence of tTG (top band) and ?-actin (bottom band, 42 kDa). Protein was obtained from single rat cremaster muscle arteries. d, Immunostaining of tTG protein in intact rat cremaster muscle. Positive staining (brown) is observed throughout the arterial wall, including the endothelium, and in the fibrous tissue and capillaries in between skeletal muscle fibers.
The presence of tTG protein in these vessels was studied using Western blot and immunostaining. Tissue-type transglutaminase could be detected in single arteries from cremaster muscle in a Western blot (Figure 3c). Sections of the rat cremaster muscle revealed the presence of tTG protein throughout the vessel wall, with particularly strong staining in the endothelium (Figure 3d). Positive staining was also observed in capillaries and fibrous tissue surrounding muscle fibers.
To manipulate endogenous tTG expression and activity, porcine coronary arteries (n=19) were kept in organoid culture during a 3-day experimental period. Vessels gradually developed an increasing level of vasoconstriction (Figure 4a). To increase the expression of tTG, all-trans-Retinoic acid (10–7 mol/L) was added to the culture medium.19 This concentration increased the level of tTG mRNA 3-fold, as indicated by RT-PCR (Figure 4b). Although retinoic acid did not acutely alter vascular tone, these arteries showed enhanced vasoconstriction as compared with controls during subsequent organoid culture (Figure 4a). Other arteries were incubated with the inhibitor cadaverine, which was added after 24 hours at a concentration of 10–4 mol/L. This compound induced vasodilation, which was maintained during the remaining of the experiment (Figure 4a). Measured under a condition of maximal vasodilation, the pressure-diameter relationship of control arteries showed a tendency to inward remodeling after 3 days of organoid culture (P=0.07; Figure 4c). Arteries that were incubated with retinoic acid showed a pronounced and significant inward remodeling (Figure 4c). In contrast, remodeling was reversed in arteries exposed to cadaverine and substantial outward remodeling was found (Figure 4c). Remodeling occurred without a change in the wall cross sectional area in all groups. At day 3, the cross sectional area was 99±5%, 104±8%, and 100±2% as percentage of day 0 for control, cadaverine, and retinoic acid groups.
Figure 4. Manipulation of endogenous tTG expression and activity. a, Diameter of coronary arteries during culture when left untreated (control; n=7), in the presence of retinoic acid (n=6) or the transglutaminase inhibitor cadaverine (n=6). Retinoic acid causes a small, but significant increase in vasoconstriction; cadaverine causes significant vasodilation during organoid culture. P<0.05, P<0.001. b, Retinoic acid enhances the expression of tTG in coronary artery (P<0.05). Left panel shows GAPDH mRNA; right panel, tTG mRNA. Representative of 4 experiments. c, Pressure-diameter relationship after full dilation. Arteries kept in organoid culture in the presence of retinoic acid show a significant reduction in the maximal diameter from day 0 to 3. Remodeling is reversed to outward remodeling after inhibition of transglutaminase with cadaverine (P<0.001).
Transglutaminase Activity
The activity of tTG in the vessel wall is reflected by the incorporation of fluorescent cadaverine, the competitive inhibitor of tTG used in the present study. After the organoid culture period, segments that were incubated with cadaverine were unmounted and viewed with a confocal microscope (Figure 5a). These arteries were compared with nonpressurized arteries that were kept in a culture dish in the presence of cadaverine for 3 days (Figure 5b) and vessels that were pressurized but not incubated with cadaverine (Figure 5c). Only in the artery that was pressurized and incubated with cadaverine a distinct pattern of small subcellular patches of high fluorescent intensity was observed. Nonpressurized vessels showed only autofluorescence, as is evident from a comparison of Figure 5b with 5c.
Figure 5. Incorporation of cadaverine reflects transglutaminase activity. a, Patches of high fluorescent intensity (green) reflect incorporation of fluorescent cadaverine in a porcine coronary artery undergoing vascular remodeling. b, An unpressurized coronary artery maintained in culture for 3 days incorporates little fluorescent cadaverine (green). c, Green autofluorescence from fibers is observed in an artery cultured in the absence of cadaverine. Ethidium bromide was used to stain nuclei (red). Pictures are representative of 4 experiments. Images were obtained using the same settings of the confocal microscope.
Transglutaminase Determines Collagen Gel Contraction
The collagen gel contraction assay is a model to study matrix reorganization by cultured cells.15 We used this model to more closely study the interaction of smooth muscle cells with the extracellular matrix component collagen. Cultured smooth muscle cells, obtained from porcine coronary artery, were seeded on a collagen gel. The cells compacted the collagen over a 24-hour period. The process of compaction depended on transglutaminase, as addition of transglutaminase markedly enhanced collagen compaction, whereas it was fully inhibited by cadaverine (Figure 6). In addition, the smooth muscle cell contractile machinery is involved, as compaction was also completely inhibited by the vasodilator papaverine (Figure 6).
Figure 6. Collagen contraction by smooth muscle cells. Smooth muscle cells cultured from porcine coronary artery compact a collagen gel (control; n=27). Compaction, shown as the percent surface area of the well occupied by collagen gel after 24 hours is significantly enhanced by transglutaminase (TG; n=8). Compaction is inhibited by both 0.1 mol/L cadaverine (cad; n=10) and the vasodilator 0.1 mol/L papaverine (pap; n=8). P<0.001 as compared with control.
In the present study, we used several approaches to establish the causal relationship between inward remodeling and tTG activity. In a first set of experiments, we demonstrated that inhibition of tTG activity by cystamine blocks the inward remodeling seen after reduction of blood flow in the mesenteric circulation of the rat. The rapid inward remodeling after flow reduction in this model was shown previously.2 A similar relation between experimental flow reduction and inward remodeling was found in carotid arteries.1,20 However, the currently found inhibition of this inward remodeling by a blocker of tTG is novel.
Several blockers of tTG, with different modes of action, were used in the present study and provided consistent evidence for a role of tTG in inward remodeling of small arteries. However, the use of potentially more selective molecular approaches such as siRNA or tTG knockout mice could provide additional evidence for the role of tTG. Interestingly, the lack of tTG expression in tTG knockout mice is not associated with a clear phenotype.21,22 Thus, limited information on tTG knockout mice indicates normal cardiovascular function. This suggests that either tTG is not important during development and normal function or redundant mechanisms operate. It remains to be determined, however, how these mice respond when challenged with a vascular remodeling stimulus.
If inward remodeling of blood vessels depends on tTG, application of this enzyme could stimulate such remodeling. We indeed observed this in rat cremaster and porcine coronary vessels. The specificity of the action of transglutaminase was supported by the inhibition by cadaverine, whereas the inhibiting effect of the NO donor SNP likely represents direct inhibition of this enzyme through nitrosilation.12 It is important to realize that these experiments were done on cannulated vessels kept at low pressure (3 to 4 mm Hg). At this pressure, the arteries are maintained at a small resting diameter. However, they have no intrinsic tendency to remodel at this pressure, as was previously shown in rat cremaster arteries.13 The current study confirmed this in porcine coronary vessels. Thus, in the absence of exogenous transglutaminase, the control vessels had identical pressure-diameter relations before and after culture (Figure 2). The remodeling induced by exogenous transglutaminase became apparent when vessels were pressurized to 60 mm Hg. We suggest that this inward remodeling represents cross-linking of collagen, because at this testing pressure, the diameter of the passive vessels is set by the collagen backbone.
Based on the in vivo effects of tTG inhibition and the in vitro effects of exogenous transglutaminase, we aimed to test whether manipulation of tTG expression or activity would influence the inward remodeling in isolated blood vessels. This was tested in two models using organoid culture: first, cannulated rat cremaster arterioles were maintained at physiological pressure in the presence of ET-1. In this model, we previously reported inward remodeling after 3 days.12 However, in the current study, a 24-hour exposure to ET-1 was sufficient to induce inward remodeling. The full inhibition of inward remodeling by cystamine and BPA in this model indeed supports the role of tTG. In a second model using porcine coronary arteries, involvement of tTG in inward remodeling was supported by the inhibition by cadaverine and the enhanced inward remodeling on upregulation of tTG expression by retinoic acid. Surprisingly, inhibition of tTG not only suppressed the inward remodeling of these vessels but actually resulted in a marked (20%) outward remodeling in three days of organoid culture. These data therefore show that coronary arterial structure is highly dynamic. It remains to be established whether in the in vivo coronary bed similar fast remodeling can occur and whether tTG plays the same dominant role there.
Our previous work12–14 and the available literature, suggest that persistent changes in vascular tone drive remodeling in physiological and pathological conditions. Smooth muscle reorganization during chronic vasoconstriction could be an early event herein.6 Evidence shows that the inward remodeling in hypertension is also related to vascular tone, because vasodilation, but not hypotension, corrects vascular structure.23 In the coronary circulation, inward remodeling is found at sites of local vasospasm,24 or associated with microvascular spasm distal to sites of injured large arteries.25 Based on this association of tone and remodeling, we suggest that inward remodeling requires two conditions: a long-lasting state of vasoconstriction and the activity of cross-linking enzymes to fixate the vessel in this state.
In the current study, remodeling was induced by either reduced blood flow in vivo or vasoconstriction in vitro. Evidence for the role of tTG in remodeling was provided for both models, but direct evidence for a long-lasting state of vasoconstriction was shown only for the in vitro experiments. However, it is likely that prolonged vasoconstriction also was induced in the in vivo model. Both mouse20 and rabbit27 carotid artery show a vasoconstrictor response after a surgically imposed reduction in blood flow, which gradually becomes irreversible. Conversely, the vasodilatory effect of flow is well established in small arteries, including those from the rat mesenteric circulation.26 Therefore, it is to be expected that chronic vasoconstriction is a common feature for both models. Thus, whereas differences with respect to proliferation and apoptosis may exist in different situations, we believe that vasoconstriction and fixation by tTG play a major role in diameter reduction both in vivo and in vitro.
We previously observed that only active vasoconstriction, and not passive reduction in diameter by decreasing the distending pressure of a relaxed vessel, results in remodeling.13 Those data suggested that activity of the cross-linking enzyme is strongly related to smooth muscle activation. Indeed, the enzyme that we now find to drive remodeling, tTG, has this relation: its activity is dependent on pressure, as reflected by the incorporation of fluorescent substrate (Figure 5), and on calcium.9 Interestingly, the vasodilator molecule nitric oxide has been reported to inhibit transglutaminase activity,12 and thus may be a physiologically relevant inhibitor of inward remodeling. We therefore incubated arteries with transglutaminase in the presence of the nitric oxide donor nitroprusside. Indeed, under these conditions exogenous transglutaminase did not alter the passive properties of the arteries.
The role of tTG in the close relationship of tone and remodeling is further substantiated by the results from the organ culture experiments. These results show that increased expression of tTG is associated with enhanced tone and inward remodeling of porcine coronary arteries. Conversely, inhibition of tTG with cadaverine caused vasodilation and reversed the remodeling to substantial outward remodeling. The vasodilatory effect may be explained by the role of tTG as a G-protein in transmembrane signaling through phospholipase C.28 Alternatively, the transamidation reactions catalyzed by tTG, which activate targets such as the monomeric G-protein RhoA,29 followed by activation of Rho-kinase and inhibition of myosin light chain phosphatase may underlie this relationship with vascular tone. The vasoconstriction induced by endothelin-1 in rat cremaster arteries was not affected by the inhibitors cystamine and BPA. This suggests that the role of tTG in vascular tone may differ in various preparations and mechanisms of vasoconstriction.
Although vascular remodeling may be initiated by vascular tone, remodeling is evident from the change in diameter under fully dilated conditions.30 Therefore, smooth muscle activation subsequently needs to be linked to rearrangement of the extracellular matrix. Integrins play an important role in the physical connection and signaling between smooth muscle cells and extracellular matrix. Interestingly, remodeled resistance arteries from spontaneously hypertensive rats show an abnormal expression of specific integrins.31 We recently showed that integrins participate in the process of remodeling in rat resistance arteries.12 Because part of the tTG is localized at the cell surface in association with integrins and fibronectin,8 it can be hypothesized that tTG acts in concert with integrins in the reorganization of matrix components. The collagen gel experiments support this view. We previously found that the reorganization of collagen in this gel depends on integrins and now showed that also tTG plays an essential role in this process. Although collagen is only one of many substrates of tTG, it is a relevant molecule to study in this context, because it is believed that collagen is the major determinant of the passive vessel diameter at higher pressure levels.32,33
In conclusion, a crucial role for tTG in the process of arterial remodeling is uncovered in the present study. This enzyme, that cross-links extracellular matrix proteins, may provide the crucial link between functional and structural modulation of arterial diameter.
References
Langille BL, O’Donnell F. Reductions in arterial diameter produced by chronic decreases in blood flow are endothelium-dependent. Science. 1986; 231: 405–407.
Buus CL, Pourageaud F, Fazzi GE, Janssen G, Mulvany MJ, De Mey JGR. Smooth muscle cell changes during flow-related remodeling of rat mesenteric resistance arteries. Circ Res. 2001; 89: 180–186.
Heagerty AM, Aalkjaer C, Bund SJ, Korsgaard N, Mulvany MJ. Small artery structure in hypertension: dual processes of remodeling and growth. Hypertension. 1993; 21: 391–397.
Intengan HD, Schiffrin EL. Vascular remodeling in hypertension: roles of apoptosis, inflammation, and fibrosis. Hypertension. 2001; 38: 581–587.
Prewitt RL, Rice DC, Dobrian AD: Adaptation of resistance arteries to increases in pressure. Microcirculation. 2003; 9: 295–304.
Martinez-Lemus LA, Hill MA, Bolz SS, Pohl U, Meininger GA. Acute mechanoadaptation of vascular smooth muscle cells in response to continuous arteriolar vasoconstriction: implications for functional remodeling. FASEB J. 2004; 18: 708–710.
Rizzoni D, Porteri E, Boari GE, De CC, Sleiman I, Muiesan ML, Castellano M, Miclini M, Agabiti-Rosei E. Prognostic significance of small-artery structure in hypertension. Circulation. 2003; 108: 2230–2235.
Akimov SS, Krylov D, Fleischman LF, Belkin AM. Tissue transglutaminase is an integrin-binding adhesion coreceptor for fibronectin. J Cell Biol. 2000; 148: 825–838.
Smethurst PA, Griffin M. Measurement of tissue transglutaminase activity in a permeabilized cell system: its regulation by Ca2+ and nucleotides. Biochem J. 1996; 313: 803–808.
Lai TS, Hausladen A, Slaughter TF, Eu JP, Stamler JS, Greenberg CS. Calcium regulates S-nitrosylation, denitrosylation, and activity of tissue transglutaminase. Biochemistry. 2001; 40: 4904–4910.
Lorand L, Graham RM. Transglutaminases: crosslinking enzymes with pleiotropic functions. Nat Rev Mol Cell Biol. 2003; 4: 140–156.
Bakker EN, Buus CL, VanBavel E, Mulvany MJ. Activation of resistance arteries with endothelin-1: from vasoconstriction to functional adaptation and remodeling. J Vasc Res. 2004; 41: 174–182.
Bakker EN, van der Meulen ET, van den Berg BM, Everts V, Spaan JA, VanBavel E. Inward remodeling follows chronic vasoconstriction in isolated resistance arteries. J Vasc Res. 2002; 39: 12–20.
Bakker ENTP, van der Meulen ET, Spaan JAE, VanBavel E. Organoid culture of cannulated rat resistance arteries: effect of serum factors on vasoactivity and remodeling. Am J Physiol. 2000; 278: H1233–H1240.
Guidry C, Grinnell F. Studies on the mechanism of hydrated collagen gel reorganization by human skin fibroblasts. J Cell Sci. 1985; 79: 67–81.
Siefring GEJ, Apostol AB, Velasco PT, Lorand L. Enzymatic basis for the Ca2+-induced cross-linking of membrane proteins in intact human erythrocytes. Biochemistry. 1978; 17: 2598–2604.
Stenberg P, Curtis CG, Wing D, Tong YS, Credo RB, Gray A, Lorand L. Transamidase kinetics. Amide formation in the enzymic reactions of thiol esters with amines. Biochem J. 1975; 147: 153–163.
Eskildsen-Helmond YE, Mulvany MJ. Pressure-induced activation of extracellular signal-regulated kinase 1/2 in small arteries. Hypertension. 2003; 41: 891–897.
Ou H, Haendeler J, Aebly MR, Kelly LA, Cholewa BC, Koike G, Kwitek-Black A, Jacob HJ, Berk BC, Miano JM. Retinoic acid-induced tissue transglutaminase and apoptosis in vascular smooth muscle cells. Circ Res. 2000; 87: 881–887.
Rudic RD, Bucci M, Fulton D, Segal SS, Sessa WC. Temporal events underlying arterial remodeling after chronic flow reduction in mice: correlation of structural changes with a deficit in basal nitric oxide synthesis. Circ Res. 2000; 86: 1160–1166.
De Laurenzi V, Melino G: Gene disruption of tissue transglutaminase. Mol Cell Biol. 2001; 21: 148–155
Nanda N, Iismaa SE, Owens WA, Husain A, Mackay F, Graham RM. Targeted inactivation of Gh/tissue transglutaminase II. J Biol Chem. 2001; 276: 20673–20678.
Christensen KL, Mulvany MJ. Vasodilatation, not hypotension, improves resistance vessel design during treatment of essential hypertension: a literature survey. J Hypertens. 2001; 19: 1001–1006.
Hong MK, Park SW, Lee CW, Ko JY, Kang DH, Song JK, Kim JJ, Mintz GS, Park SJ. Intravascular ultrasound findings of negative arterial remodeling at sites of focal coronary spasm in patients with vasospastic angina. Am Heart J. 2000; 140: 395–401.
Saitoh S, Onogi F, Aikawa K, Muto M, Saito T, Maehara K, Maruyama Y. Multiple endothelial injury in epicardial coronary artery induces downstream microvascular spasm as well as remodeling partly via thromboxane A2. J Am Coll Cardiol. 2001; 37: 308–315.
Sun D, Huang A, Koller A, Kaley G. Adaptation of flow-induced dilation of arterioles to daily exercise. Microvasc Res. 1998; 56: 54–61.
Langille BL, Bendeck MP, Keeley FW. Adaptations of carotid arteries of young and mature rabbits to reduced carotid blood flow. Am J Physiol. 1989; 256: H931–H939.
Im MJ, Russell MA, Feng JF. Transglutaminase II: a new class of GTP-binding protein with new biological functions. Cell Signal. 1997; 9: 477–482.
Singh US, Kunar MT, Kao YL, Baker KM. Role of transglutaminase II in retinoic acid-induced activation of RhoA-associated kinase-2. EMBO J. 2001; 20: 2413–2423.
Bund SJ, Lee RM. Arterial structural changes in hypertension: a consideration of methodology, terminology and functional consequence. J Vasc Res. 2003; 40: 547–557.
Intengan HD, Thibault G, Li JS, Schiffrin EL. Resistance artery mechanics, structure, and extracellular components in spontaneously hypertensive rats: effects of angiotensin receptor antagonism and converting enzyme inhibition. Circulation. 1999; 100: 2267–2275.
Dobrin PB. Mechanical properties of arteries. Physiol Rev. 1978; 58: 397–460.
VanBavel E, Siersma P, Spaan JA. Elasticity of passive blood vessels: a new concept. Am J Physiol. 2003; 285: H1986–H2000.(Erik N.T.P. Bakker, Carst)