Overexpression of Hyaluronan in the Tunica Media Promotes the Development of Atherosclerosis
http://www.100md.com
循环研究杂志 2005年第3期
the Research Laboratory for Biochemical Pathology (S.C., Q.C., T.L., L.W.), Electron Microscopy and Stereological Research Laboratory (J.R.N.)
the Department of Clinical Biochemistry (L.M.R.), Aarhus University Hospital, Denmark
the Institute of Anatomy (C.C.D.) and the Institute of Molecular and Structural Biology (P.H.), University of Aarhus, Denmark
The Burnham Institute (Y.Y.), La Jolla, Calif.
Abstract
The arterial content of hyaluronan (HA) undergoes diffuse changes as part of the diabetic macroangiopathy. Because HA influences the phenotype of vascular cells in vitro such as proliferation, migration, and secretion, it is tempting to speculate that diabetes-induced hastened cardiovascular disease may be linked to the increased amount of HA. To explore the pathophysiological role of altered HA content in the arterial wall in vivo, we created transgenic (Tg) mice with HA overexpression in smooth muscle cells (SMCs) in large and small vessels, targeted by the smooth-muscle-cell-actin (SMA) promoter fused to the human hyaluronan synthase 2 (hHAS2) cDNA. RT-PCR demonstrated hHAS2 mRNA expression in the tunica media of large and small vessels. In situ hybridization confirmed that hHAS2 mRNA was targeted to the SMCs. The aortic HA content was elevated in the Tg mice, and by immunohistochemistry, it was seen that HA accumulated in the tunica media. The secretory profile of high- and low-molecular HA was similar in wild-type and Tg animals. Overproduction of HA in the aorta resulted in thinning of the elastic lamellae in Tg mice. Our data suggest that this may lead to increased mechanical stiffness and strength, as determined by controlled stretching until failure. Finally, overproduction of HA on the genetic background of the ApoE-deficient mouse strain promoted atherosclerosis development in the aorta. These results indicate that a single component of the diabetic macroangiopathy, diffuse accumulation of HA, accelerates the progression of atherosclerosis.
Key Words: hyaluronan biomechanics atherosclerosis diabetes transgenic
Introduction
Hyaluronan (HA) is a large, nonsulfated glycosaminoglycan (GAG) produced in the vasculature by smooth muscle cells (SMCs) and endothelial cells (ECs). It is synthesized at the inner face of the cell membrane by hyaluronan synthases 1 to 3 (HAS 1 to 3), followed by translocation to the outer surface and the intercellular space.1 Recent in vitro investigations have shown that high (HMW) as well as low molecular (LMW) forms of HA influence important cellular functions, such as proliferation, migration, and secretory capabilities.1eC4 There seems to exist a delicately regulated balance between production, sizing, and removal of HA that is central to its biological functions during normal conditions. If this balance is disturbed, it has been hypothesized that disease may develop.5 The notion that HA is likely to take part in the development of vascular pathologies has been nourished by a number of observations, which show altered HA amounts in several arterial disease entities such as diabetic angiopathy, atherosclerosis, and restenosis.6eC8 During the build-up of the atherosclerotic plaques in humans, some areas with local accumulation of HA appear.9eC11 This is in contrast to the situation in patients with diabetes in whom the HA accumulation is disseminated in tunica media and is regarded as one important element in a series of diffuse matrix changes in the vessels.6,12 The diabetic angiopathy deserves some specific comments. First, a very high and equal frequency of cardiovascular disease (CVD) exists among men and women with diabetes for as yet unexplained reasons.13eC15 In line with this, it is interesting that in subjects with long-term diabetes the increasing mortality from CVD is independent of any well-known risk factors such as hypertension, dyslipidemia, and microalbuminuria.16,17 Second, the vessel wall thickness is increased in juvenile patients with a diabetes duration of 5 years.18,19 Third, histochemical and biochemical records demonstrate that atheromatosis free arterial wall segments from patients with diabetes contain excessive amounts of HA, type IV collagen, and fibronectin in the tunica media.6,12 Finally, in vitro studies reveal that factors from the diabetic metabolism affect the release of HA from SMCs.20 In concert these studies indicate that matrix alterations in the arterial wall in diabetes may be part of a generalized angiopathy. We hypothesize that these changes make the vessel wall more susceptible to atherogenic stimuli, thus contributing to the increased prevalence of CVD in diabetic patients. To address the biological effects of increased HA content in the arteries in vivo, we created transgenic mice with disseminated overexpression of HA in the tunica media of large and small vessels. We investigated the ultrastructural changes in the arterial wall, the stiffness and strength of aortic rings, and the influence on the development of atherosclerosis in ApoE-deficient mice.
Materials and Methods
Production of hHAS2 Transgenic Mice
A restriction fragment containing the mouse SMC-actin 5' flanking region SMP8 (eC1074 bp, 63 bp of 5'UT and 2.5 kb of intron 1), the human hyaluronan synthase 2 (hHAS2) cDNA, and a SV40 polyA terminator sequence21,22 was gel-purified using GenecleanII (BIO101) and microinjected into pronuclei of fertilized eggs from F1 (C57BL/6JxDBA/2J) females crossed with F1 (C57BL/6JxDBA/2J) males (Taconic Europe, Ry, Denmark). The progeny from backcrosses to C57BL/6J were identified as transgenic (Tg) or nontransgenic (WT) by polymerase chain reaction (PCR) with primers complementary to sequences in the SV40 fragment. hHAS2 transgenic animals (>3-generation C57BL/6J) were crossed with ApoE-deficient (ApoEeC/eC) mice (>10-generation C57BL/6J) (Bomholtgaard Breeding and Research Center) to generate animals that were hHAS2+ApoEeC/eC or hHAS2eCApoEeC/eC (>3 to 4 generation). Unless otherwise stated, mice were euthanized by cervical dislocation. For details, including legal approvals, see the expanded Material and Methods in the online data supplement available at http://circres.ahajournals.org.
RNA Isolation and RT-PCR
Total RNA was reversed transcribed to gain cDNA. Amplification of cDNA was performed using the following primers: hHAS2, 5'-GCATCACACCTCATCATCC-3' and 5'-GACTTCTCTTT-TTCCACCCC-3'. Murine -actin (m-actin), 5'-CTACAAT-GAGCTGCGTGTGGC-3' and 5'-GTCCAGACGCAGGATGG-CATG-3' and accomplished as described online. The PCR products were separated by electrophoresis in a 2% agarose gel with ethidium bromide. The intensity of the visualized bands was measured in arbitrary absorbance units with a BIO-RAD UV-gel camera. Results are presented as the ratio between the density of hHAS2 and m-actin. For details, see expanded Material and Methods.
In Situ Hybridization
Hybridization of deparaffinized sections from selected organs was performed with 35S-UTPeClabeled sense or antisense RNA probes (6x105 cpm/section) according to standard methods (see expanded Material and Methods).
HA Immunoassay
Tissues were excised, and the aorta was dissected into intimamedia. The tissues were weighed, homogenized in 2 mL of 0.01 mmol/L Tris-HCl and 0.2 mmol/L EDTA pursued by digestion with protease from Streptomyces Griseus (1 mg/mL, Sigma) overnight at 37°C and followed by heat inactivation (100°C for 30 minutes). After centrifugation (30 000g for 10 minutes), the HA content in the supernatants was measured by an HA ELISA (Chugai Diagnostics Science Co). The detection limit of the assay was 10 ng/mL and the intra-assay CV% was 3.6% to 4.7%. The plasma HA concentration was measured using the same ELISA.
Histochemistry
We stained paraffin or cryosections by standard methods with orcein, biotinylated HABP (1:250, Seikagaku Co), and antibodies against SMC-actin (1:150, NeoMarkers) and CD11b/CD18 (Mac1) (1:150, Chemicon) (see expanded Material and Methods).
Determination of HA Molecular Weight by High-Performance Liquid Chromatography
The aorta (Tg: n=6, WT: n=4) was dissected and chopped into small pieces that were distributed in two wells (96-well plate) and incubated at 37°C for 24 hours in MEM containing 40 e藽i D-[6-3H]-glucosamine-hydrochloride (Amersham Biosciences, UK). Tissue from two mice were pooled, homogenized, and extracted with guanidinium-chloride. Diluted supernatants were passed through an anion exchange cellulose column (DEAE-52, Whitman). The HA-containing elute underwent dialyze and freeze-drying. The samples were resuspended in 0.1 mol/L CH3COONH4, pH 7.2. One part was treated with Hyaluronidase (Streptomyces hyaluronlyticus) (15 U/e蘈, Calbiochem, VWR Int. GmbH), the other sample was left untreated. All samples (Tg, n=3; WT, n=2), were separated by high-performance liquid chromatography (HPLC) (Superose 6 HR 10/30 column) (PharmaciaBiotech). Fractions were collected and counted. For details, see expanded Material and Methods.
Electron Microscopy
Aortic ring profiles were embedded in Epon for electron microscopy and 50-nm sections were examined ultrastructurally with a digitalized Philips CM10 electron microscope (SIS 3.0). All vascular profiles were scored according to the amount of the intercellular ground substance (clear areas) and the degree of elastin fragmentation (0=absent/low, 1=moderate, 2=severe). The width of 4 to 7 elastic lamellae, including membrana elastica interna and externa, and the distance between the lamellae, were measured at two or four positions chosen at random. Mean values of the parameters were obtained for each mouse before group means were calculated. All measurements were performed in a blinded fashion. For details, see expanded Materials and Methods.
Mechanical Testing and Hydroxyproline Content
Freshly excised aortas were stored at eC20°C until analysis. The lower thoracic aorta was cleaned of adhering fat and loose connective tissue. Aortic ring specimens (1-mm high; in average 3.5 specimens per aorta) were cut between intercostal arteries and mounted in a materials testing machine (Alwetron TCT5, Lorentzen & Wettre). Then the specimen was stretched (10 mm/min) until failure by pulling the wires apart in a direction perpendicular to the wires. Thereafter, the collagen content in the ring specimen was estimated by hydroxyproline determination. From the recorded load-deformation values, the load-strain curve was calculated and maximal load and maximal stiffness (ie, maximal slope of the load strain curve) were derived. In calculation of strain, the original circumference of the ring specimen was defined as the circumference at which the specimen attained a minimal load (1% of maximum). Mean values of the parameters were obtained for each mouse before group means were calculated.
Analysis of Atherosclerosis in hHAS2+ApoEeC/eC and hHAS2eCApoEeC/eC Transgenic Mice
The heart and entire aorta were removed from hHAS2+ApoEeC/eC and hHAS2eCApoEeC/eC littermates after 16 to 20 weeks on normal diet. The processing of the root is described in the expanded Material and Methods. The average atherosclerotic lesion area in three levels was determined using Olympus BX50 microscope and analysis software. The extent of atherosclerosis in the aortic arch and thoracic aorta was measured en face by visualizing cholesterol deposits with Oil Red O. The surface area of the lesion within the arch was expressed as percentage of the arch covered by lesion calculated by means of analysis software including multiple-image alignment. The area with positive staining in the thoracic aorta was calculated by using a frame (175 349.1 e2/144 crosses; CAST, Olympus) that was moved stepwise, from side to side all along the aorta. All analyses were made in a coded fashion by the same investigator.
Statistics
Statistical analysis between two groups of data were performed by Student t test after exclusion of variance in-homogeneity (tested by F-test). Otherwise, we used the ManneCWhitney rank sum test. Statistical analysis between more than two groups was performed by one-way ANOVA with post hoc Bonferroni. In the case of variance in-homogeneity as tested by the Bartlett test, the KruskalleCWallis test was used followed by the ManneCWhitney rank sum test, which was corrected for the number of comparisons. Data were tested for outliners. Differences were considered statistically significant at a level of P0.05.
Results
Expression of hHAS2 RNA in the Tunica Media of Large and Small Vessels (RT-PCR and ISH)
Four of 26 pups were Tg. One died before transmission of the transgene. Thus, three lines were propagated for investigations (5986, 5989, and 5990), of which all expressed hHAS2 RNA in aorta, kidney, and bladder (5986, n=6; 5989, n=4; 5990, n=6) (Figure 1). The hHAS2 RNA amount was higher in the bladder versus the aorta in all lines (all probability values <0.01). The level of hHAS2 RNA in the aorta was approximately two times higher in line 5990 versus line 5986 (P<0.05). Line 5989 showed the lowest hHAS2 expression. hHAS2 RNA was absent in WT mice (n=6) (Figure 1). Targeting of hHAS2 RNA production to the SMCs of the media of the aorta and smaller renal vessels and of the bladder was confirmed by in situ hybridization (ISH) in Tg mice of line 5986 and 5990 (n=2 for each line) (Figure 2A). Few signals were found in the media of large and small vessels in Tg mice of line 5989 (n=2) and none were seen in WT mice (n=2) (Figure 2A). At this stage, line 5989 was terminated.
Increased Aortic HA Content
The HA content (ELISA) in selected organs from 8- to 14-month-old mice are given in the Table. The HA amount in the aorta was higher in Tg mice of line 5986 (n=12) and 5990 (n=4) versus WT animals (n=9) (both P<0.05). The HA content in the kidney only reached statistical significance in mice of line 5986 (P<0.05). Both lines exhibited elevated HA levels in the bladder; however, there were no statistically significant differences versus WT mice (Table). The HA content in the selected organs was similar when comparing the two lines of transgenic mice (Table). A similar pattern of HA production in the kidney and bladder was seen in younger, 4- to 5-month-old Tg mice (line 5986) (Bladder, 44.9±11.2 ng/mg; Kidney, 1.31±0.64 ng/mg; n=6) versus WT mice (Bladder, 46±11.2 ng/mg; Kidney, 0.35±0.05 ng/mg; n=5) (P>0.05 and P=0.01, respectively). Immunohistochemistry (IHC) demonstrated that HA was localized in the tunica media of the aorta (n=4 for both line 5986 and 5990), and the smaller vessels in the kidney (n=4 for both line 5986 and 5990) and retina (n=9, line 5986) of Tg mice (Figure 2B). On the other hand, WT mice exhibited weak HA expression (for each organ n=4 to 7). In the bladder, HA was present in the connective tissue layer of the mucosa and the muscularis in all tested mice; however, it was difficult to detect any differences in the staining pattern in Tg (n=4 for both line 5986 and 5990) versus WT mice (n=4) (data not shown). The mean plasma HA concentration was similar among Tg mice of line 5986 (705±156 ng/mL; n=8), line 5990 (621±98 ng/mL; n=8), and WT mice (650±80 ng/mL; n=9) (P>0.05).
Molecular Weight Characterization
The distribution of HMW and LMW HA forms produced in vitro from cultured slices of aorta exhibited a similar pattern in Tg (n=3) versus WT mice (n=2) (Figure 3). In all incidents, the HA isolated and purified from overnight incubated aorta slices were eluted as two major molecular size species G1 (fractions 8 to 10) [Tg, 55±5.2% (n=3); WT, 53±6% (n=2); G2 (fractions 16 to 20) Tg, 40±5% (n=3); WT, 44±6% (n=2)]. The ratio between the peaks was 1.5±0.29 in Tg (n=3) versus 1.3±0.31 in WT (n=2) mice. The first peak corresponds to the void volume equivalent to a molecular size of >1x106 Da. The second peak corresponds to molecular size species a little larger than 1.1x103 Da. Both peaks represent HA, because they disappeared after treatment of parallel handled samples with Streptomyces hyaluronlyticus before HPLC (data not shown).
Thinning of the Elastic Membranes in the Aorta of Transgenic Mice
Orcein staining did not reveal any obvious differences in the vascular morphology between WT and Tg mice (Figure 4). By electron microscopy we found, however, that the average width of the elastic lamellae in Tg (n=8) and WT (n=7) mice was 0.763±0.097 and 0.926±0.083 e, respectively (P<0.001). The distance between the elastic lamellae (Tg, 3.831±1.015 e; WT, 3.861±1.142 e) and the total media thickness (Tg, 26.7±4.26 e; WT, 25.4±7.9 e) were similar in the two groups. More ground substance was visible in the space between the elastic lamellae and the SMCs in Tg versus WT mice (P<0.01). Debris from the elastic fibers was more often seen in Tg mice than in normal animals, and fragmentation of the elastic membranes was clearly more evident in Tg mice versus WT controls (P<0.01) (Figure 4).
Alterations of the Biomechanical Properties of the Aorta in Transgenic Mice
The maximal load and the maximal stiffness were significantly different among the groups of both the 3-month-old (P=0.029 and P=0.028, respectively) and 6-month-old mice (P=0.009 and P=0.061, respectively) (Figure 5). At 3 months of age, both the maximal load and the maximal stiffness were higher in Tg mice of line 5986 (n=6) versus age-matched WT mice (n=10) (both P<0.05), whereas the values of line 5990 (n=6) were in between. At 6 months of age, the maximal load was higher in Tg mice of line 5986 (n=6) versus age-matched WT mice (n=8) and mice of line 5990 (n=6) (both P<0.05), whereas the maximal stiffness, with the highest value in mice of line 5986, tended to be different among the groups (P=0.061) (Figure 5). In all tested groups of mice, both the aortic extensibility and the total collagen content expressed as mg per mm of aortic circumference were similar (data not shown).
Promotion of Atherosclerosis in hHAS2+ApoEeC/eC Transgenic Mice
We backcrossed hHAS2 transgenic mice (line 5986) with ApoEeC/eC animals to obtain hHAS2+ApoEeC/eC and hHAS2eCApoEeC/eC animals. The HA staining was profound in the tunica media of the aortic root (n=9), the aortic arch (n=6), and the thoracic aorta (n=1) in hHAS2+ApoEeC/eC, whereas it was weak in hHAS2eCApoEeC/eC animals (root, n=9, aortic arch, n=7, and thoracic aorta, n=1) (Figure 6A). The HA content was higher in the aortic arch, thoracic aorta, and the bladder in hHAS2+ApoEeC/eC versus hHAS2eCApoEeC/eC mice (all P<0.05) (Table). hHAS2+ApoEeC/eC mice exhibited only marginally elevated HA levels in the kidney (P=0.28) (Table). At 16 to 20 weeks of age, the extent of the atherosclerotic lesions in the aortic root were higher 80 and 160 e from the valve cusps in hHAS2+ApoEeC/eC (n=15) versus hHAS2eCApoEeC/eC (n=9) (both P<0.05) (Figure 7A). The area (%) of lipid deposits in the aortic arch was increased in hHAS2+ApoEeC/eC (n=17) versus hHAS2eCApoEeC/eC (n=10) (P<0.001) (Figure 7B). This difference was also present in the descending aorta (hHAS2+ApoEeC/eC (n=15) and hHAS2eCApoEeC/eC (n=10) (all P<0.05) (Figure 7C). The SMCs were seen throughout the tunica media (Figure 6B). The distribution of SMC-positive cells in the plaques in the aortic root was similar in hHAS2+ApoEeC/eC (n=5) versus hHAS2eCApoEeC/eC (n=8). They were primarily seen under the fibrous capsule and scattered around in the central area of the plaques (Figure 6B). The presence of HA was obvious in areas with SMCs in hHAS2+ApoEeC/eC but not in hHAS2eCApoEeC/eC (Figure 6B). Both genotypes exhibited HA staining in areas with cholesterol crystals. This may, however, be due to background staining (Figure 6B). Also the distribution of CD11b/CD18-positive cells in the plaques in the aortic root was similar in hHAS2+ApoEeC/eC (n=8) versus hHAS2eCApoEeC/eC (n=7). The macrophages were primarily seen central in the lesion and in the underlying tunica media and adventitia (Figure 6B).
Discussion
This study clearly indicates that raised HA levels in the aortic wall significantly transform the property of the vessel, leading to reduced thickness of the elastic lamellae, altered biomechanical properties, and most importantly, an increased susceptibility to atherogenic insults. This is in line with observations reported by Renard et al23 who suggest that GAG accumulation is one of the first steps in initiation of diabetes-induced atherosclerosis.
Our investigations used transgenic mice in which hHAS2 expression, driven by the murine SMC-actin promoter, was able to generate arterial hHAS2 RNA, leading to raised HA levels with a normal distribution of HMW versus LMW forms.5 We found a relatively high HA content in WT aorta, which at a first glance seems contradictory to the almost absent HA staining in the tunica media in the WT mice. It should be underlined, however, that the vessel extracts probably include some components from the tunica adventitia that normally contain HA. Clearly, our data show that accumulation of HA occurs especially in the tunica media. The circulating HA levels were similar in transgenic and WT mice. This suggests that HA in our model acts in an autocrine or paracrine fashion. It is generally believed that HA throughout the body is cleared by receptor-mediated uptake mediated by the hyaluronan receptor for endocytosis (HARE), which is found in lymph node, liver, and spleen sinusoidal endothelial cells.24 Tissue HA is mainly removed by lymph and passed to the lymph nodes in which it is degraded. Approximately 15% of HA exits the lymph nodes, enters the blood, and is removed by the liver sinusoidal endothelial cells,25 which has a rather high clearance capacity.26 Taken together, it is not surprising that the HA plasma levels are normal in Tg mice.
The higher HA content tend to increase the maximal load and the maximal stiffness of the vessel in hHAS2 transgenic mice. This observation was manifest in one of our transgenic lines (5986) at 3 and 6 months of age. Furthermore, a trend toward increased maximal load and maximal stiffness was seen in 3-month-old mice of line 5990. However, this trend disappeared with age. Our observations suggest that HA may influence the biomechanical properties of the aorta. Hyaluronan promotes the shrinkage of collagen gels by aortic SMCs in vitro.27 This observation is believed to be associated with changes in collagen reorganization and cell shape and partly mediated by activation of the CD44 signaling pathway.27 In line with this is the observation that blocking of RHAMM (receptor of HA-mediated motility) is linked to attenuated collagen gel contraction by SMCs isolated from the bladder exposed to stretch injury.28 Furthermore, in accordance with the minimal strength and stiffness of elastin compared with collagen,29 the thinning and fragmentation of the elastic membranes in the Tg mice were not found to be reflected in neither reduced mechanical strength, stiffness, and extensibility of the aorta or changes in the toe-part of the load-strain curves corresponding to the normal in vivo range of loading (data not shown). Therefore, the observed changes of the elastic membranes more likely contribute to collagen reorganization resulting in increased strength and stiffness of the aortic wall in Tg mice. This is in line with observations in a mouse model of Marfan syndrome and in experiments evaluating the impact of elastases on aortic elasticity, in which fragmentation of the elastic membranes is associated with increased stiffness of the vessel wall.30,31
The most prominent observation in the present study is that ApoE-deficient animals with increased HA levels in the vessel wall develop accelerated atherosclerosis. It should be underlined that atherosclerosis is absent in our hHAS2 mice on the normal C57 genetic background. Thus, the presence of one element of the diabetic macroangiopathy in combination with an atherogenic milieu, but without hyperglycemia, accelerates the development of atherosclerotic lesions. Several observations have implicated HA in a variety of the biological processes involved in atherogenesis,9eC11 but our data show for the first time, that disseminated accumulation of HA accelerates the atherogenic process in vivo. Several mechanisms are likely to play a role in the augmented atherogenesis. First, a recent report has described an attenuated development of atherosclerosis in mice genetically engineered to lack CD44.32 Because HA is the principal ligand for CD44, the observation is compatible with our results. Recruitment of macrophages was decreased and SMCs were in a quiescent state in the CD44-deficient mice. These are in line with in vitro data showing that HA accumulation may be important for macrophages/foam cell migration and activity through HA-CD44 receptor interactions.33,34 Likewise, SMC proliferation and migration is modulated by HA.2 Thus, a HA-rich environment is believed to promote SMCs as well as macrophage secretory capabilities, migration, and proliferation. Clearly, both alterations in the phenotypic stage of SMCs and macrophages may play a role in the increased atherogenesis. Presently, we can only say that there is no overwhelming difference in the pattern of macrophage infiltration of the atherosclerotic plaques in hHAS+ApoEeC/eC versus hHAS2eCApoEeC/eC mice. However, it is as yet unknown whether the functional status of the infiltrating macrophages and SMCs are altered. Secondly, augmented atherogenesis may relate to the observed biomechanical alterations, because it has been hypothesized that increased arterial stiffness may give rise to an atherosclerosis-prone vessel wall. This proposal is based both on observational data about relations between arterial stiffness and atherosclerosis35 and on experimental results, linking increased arterial stiffness with marked alterations in endothelial cell function, putatively leading to vascular damage.36 Finally, it is possible that lipoprotein transport may be altered due to modifications in the arterial wall, and because HA binds lipoproteins, this may accelerate atherogenesis in hHAS2+ApoEeC/eC mice.37 It is intriguing that our findings are in agreement with a recent publication by Sussmann et al.9 They report that upregulation of HAS2 expression via EP2 and IP receptors may contribute to HA accumulation during the development of the atherosclerotic plaques, thereby mediating proatherogenic functions of COX2. On the other hand, downregulation of HAS2 expression may mediate the antiatherogenic properties of COX2 inhibitors.9 The exact mechanisms behind the HA-induced atherogenesis, however, needs further exploration.
In summary, our in vivo observations suggest that accumulation of HA in the aortic tunica media is followed by an increased mechanical stiffness and strength of the aorta. Furthermore, the accumulation of HA stimulates the development of atherosclerosis in ApoEeC/eC mice. Based on the observations reported in this study, we propose that increased levels of HA in the vascular wall promote the progression of atherosclerotic lesions. Moreover, the obtained knowledge of disseminated HA accumulation as a promoter of atherogenesis may have implications for the understanding of diabetic angiopathy. In diabetes, the amount of HA, and other extracellular matrix components in the tunica media, is increased in atheromatosis-free areas.6,12 Our data are compatible with the notion that raised levels of HA in the tunica media, as part of the diabetic macroangiopathy, may facilitate the development of atherosclerosis among patients with diabetes, with HA being one missing factor linking diabetes and diabetes-induced atherosclerosis.
Acknowledgments
The project was supported by the Danish Medical Research Council (grants nos. 9601759, 9802621, and 9600822; Aarhus University, Novo Nordisk Center for Research in Growth and Regeneration), Fougner-Hartmanns Foundation, the Novo Nordisk Foundation, The Danish Diabetes Association, and Clinical Institute, University of Aarhus, Denmark. We are indebted to Lotte Arentoft, Anita Morell, Karin Vestergaard, Lone Lysgaard, Annette Berg, and Eva K. Mikkelsen for their technical assistance. Carsten Hjskov is thanked for helping with the HPLC.
References
Lee JY, Spicer AP. Hyaluronan: a multifunctional, megaDalton, stealth molecule. Curr Opin Cell Biol. 2000; 12: 581eC586.
Evanko SP, Angello JC, Wight TN. Formation of hyaluronan- and versican-rich pericellular matrix is required for proliferation and migration of vascular smooth muscle cells. Arterioscler Thromb Vasc Biol. 1999; 19: 1004eC1013.
Savani RC, Wang C, Yang B, Zhang S, Kinsella MG, Wight TN, Stern R, Nance DM, Turley EA. Migration of bovine aortic smooth muscle cells after wounding injury: the role of hyaluronan and RHAMM. J Clin Invest. 1995; 95: 1158eC1168.
Lokeshwar VB, Selzer MG. Differences in hyaluronic acid-mediated functions and signaling in arterial, microvessel, and vein-derived human endothelial cells. J Biol Chem. 2000; 275: 27641eC27649.
Tammi MI, Day AJ, Turley EA. Hyaluronan and homeostasis: a balancing act. J Biol Chem. 2002; 277: 4581eC4584.
Heickendorff L, Ledet T, Rasmussen LM. Glycosaminoglycans in the human aorta in diabetes mellitus: a study of tunica media from areas with and without atherosclerotic plaque. Diabetologia. 1994; 37: 286eC292.
Riessen R, Wight TN, Pastore C, Henley C, Isner JM. Distribution of hyaluronan during extracellular matrix remodeling in human restenotic arteries and balloon-injured rat carotid arteries. Circulation. 1996; 93: 1141eC1147.
Turley EA. Extracellular matrix remodeling: multiple paradigms in vascular disease. Circ Res. 2001; 88: 2eC4.
Sussmann M, Sarbia M, Meyer-Kirchrath J, Nusing RM, Schror K, Fischer JW. Induction of hyaluronic acid synthase 2 (HAS2) in human vascular smooth muscle cells by vasodilatory prostaglandins. Circ Res. 2004; 94: 592eC600.
Papakonstantinou E, Roth M, Block LH, Mirtsou-Fidani V, Argiriadis P, Karakiulakis G. The differential distribution of hyaluronic acid in the layers of human atheromatic aortas is associated with vascular smooth muscle cell proliferation and migration. Atherosclerosis. 1998; 138: 79eC89.
Levesque H, Girard N, Maingonnat C, Delpech A, Chauzy C, Tayot J, Courtois H, Delpech B. Localization and solubilization of hyaluronan and of the hyaluronan-binding protein hyaluronectin in human normal and arteriosclerotic arterial walls. Atherosclerosis. 1994; 105: 51eC62.
Andresen JL, Rasmussen LM, Ledet T. Diabetic macroangiopathy and atherosclerosis. Diabetes. 1996; 45: S91eCS94.
Kannel WB, McGee DL. Diabetes and cardiovascular disease. The Framingham study. JAMA. 1979; 241: 2035eC2038.
Lloyd CE, Kuller LH, Ellis D, Becker DJ, Wing RR, Orchard TJ. Coronary artery disease in IDDM. Gender differences in risk factors but not risk. Arterioscler Thromb Vasc Biol. 1996; 16: 720eC726.
Howard BV, Hsia J, Ouyang P, Van Voorhees L, Lindsay J, Silverman A, Alderman EL, Tripputi M, Waters DD. Postmenopausal hormone therapy is associated with atherosclerosis progression in women with abnormal glucose tolerance. Circulation. 2004; 110: 201eC206.
Kuusisto J, Mykkanen L, Pyorala K, Laakso M. NIDDM and its metabolic control predict coronary heart disease in elderly subjects. Diabetes. 1994; 43: 960eC967.
Spijkerman AM, Dekker JM, Nijpels G, Jager A, Kostense PJ, van Hinsbergh VW, Bouter LM, Heine RJ, Stehouwer CD. Impact of diabetes duration and cardiovascular risk factors on mortality in type 2 diabetes: the Hoorn Study. Eur J Clin Invest. 2002; 32: 924eC930.
Jarvisalo MJ, Jartti L, Nanto S, Irjala K, Ronnemaa T, Hartiala JJ, Celermajer DS, Raitakari OT. Increased aortic intima-media thickness: a marker of preclinical atherosclerosis in high-risk children. Circulation. 2001; 104: 2943eC2947.
Jarvisalo MJ, Putto L, Jartti L, Lehtimaki T, Solakivi T, Ronnemaa T, Raitakari OT. Carotid artery intima-media thickness in children with type 1 diabetes. Diabetes. 2002; 51: 493eC498.
Erikstrup C, Pedersen LM, Heickendorff L, Ledet T, Rasmussen LM. Production of hyaluronan and chondroitin sulphate proteoglycans from human arterial smooth muscle - the effect of glucose, insulin, IGF-I or growth hormone. Eur J Endocrinol. 2001; 145: 193eC198.
Min BH, Foster DN, Strauch AR. The 5'-flanking region of the mouse vascular smooth muscle alpha-actin gene contains evolutionarily conserved sequence motifs within a functional promoter. J Biol Chem. 1990; 265: 16667eC16675.
Watanabe K, Yamaguchi Y. Molecular identification of a putative human hyaluronan synthase. J Biol Chem. 1996; 271: 22945eC22948.
Renard CB, Kramer F, Johansson F, Lamharzi N, Tannock LR, von Herrath MG, Chait A, Bornfeldt KE. Diabetes and diabetes-associated lipid abnormalities have distinct effects on initiation and progression of atherosclerotic lesions. J Clin Invest. 2004; 114: 659eC668.
Zhou B, McGary CT, Weigel JA, Saxena A, Weigel PH. Purification and molecular identification of the human hyaluronan receptor for endocytosis. Glycobiology. 2003; 13: 339eC349.
Weigel JA, Raymond RC, McGary C, Singh A, Weigel PH. A blocking antibody to the hyaluronan receptor for endocytosis (HARE) inhibits hyaluronan clearance by perfused liver. J Biol Chem. 2003; 278: 9808eC9812.
Suzuki M, Takeuchi H, Kakita T, Unno M, Katayose Y, Matsuno S. The involvement of the intracellular superoxide production system in hepatic ischemia-reperfusion injury. In vivo and in vitro experiments using transgenic mice manifesting excessive CuZn-SOD activity. Free Radic Biol Med. 2000; 29: 756eC763.
Travis JA, Hughes MG, Wong JM, Wagner WD, Geary RL. Hyaluronan enhances contraction of collagen by smooth muscle cells and adventitial fibroblasts: Role of CD44 and implications for constrictive remodeling. Circ Res. 2001; 88: 77eC83.
Bagli DJ, Joyner BD, Mahoney SR, McCulloch L. The hyaluronic acid receptor RHAMM is induced by stretch injury of rat bladder in vivo and influences smooth muscle cell contraction in vitro [corrected]. J Urol. 1999; 162: 832eC840.
Shadwick RE. Mechanical design in arteries. J Exp Biol. 1999; 202: 3305eC3313.
Marque V, Kieffer P, Gayraud B, Lartaud-Idjouadiene I, Ramirez F, Atkinson J. Aortic wall mechanics and composition in a transgenic mouse model of Marfan syndrome. Arterioscler Thromb Vasc Biol. 2001; 21: 1184eC1189.
Dobrin PB, Baker WH, Gley WC. Elastolytic and collagenolytic studies of arteries. Implications for the mechanical properties of aneurysms. Arch Surg. 1984; 119: 405eC409.
Cuff CA, Kothapalli D, Azonobi I, Chun S, Zhang Y, Belkin R, Yeh C, Secreto A, Assoian RK, Rader DJ, Pure E. The adhesion receptor CD44 promotes atherosclerosis by mediating inflammatory cell recruitment and vascular cell activation. J Clin Invest. 2001; 108: 1031eC1040.
McKee CM, Penno MB, Cowman M, Burdick MD, Strieter RM, Bao C, Noble PW. Hyaluronan (HA) fragments induce chemokine gene expression in alveolar macrophages. The role of HA size and CD44. J Clin Invest. 1996; 98: 2403eC2413.
McKee CM, Lowenstein CJ, Horton MR, Wu J, Bao C, Chin BY, Choi AM, Noble PW. Hyaluronan fragments induce nitric-oxide synthase in murine macrophages through a nuclear factor kappaB-dependent mechanism. J Biol Chem. 1997; 272: 8013eC8018.
Weber T, Auer J, O‘Rourke MF, Kvas E, Lassnig E, Berent R, Eber B. Arterial stiffness, wave reflections, and the risk of coronary artery disease. Circulation. 2004; 109: 184eC189.
Peng X, Haldar S, Deshpande S, Irani K, Kass DA. Wall stiffness suppresses Akt/eNOS and cytoprotection in pulse-perfused endothelium. Hypertension. 2003; 41: 378eC381.
Srinivasan SR, Yost K, Radhakrishnamurthy B, Dalferes ER, Jr., Berenson GS. Lipoprotein-hyaluronate associations in human aorta fibrous plaque lesions. Atherosclerosis. 1980; 36: 25eC37.(Song Chai, Qing Chai, Car)
the Department of Clinical Biochemistry (L.M.R.), Aarhus University Hospital, Denmark
the Institute of Anatomy (C.C.D.) and the Institute of Molecular and Structural Biology (P.H.), University of Aarhus, Denmark
The Burnham Institute (Y.Y.), La Jolla, Calif.
Abstract
The arterial content of hyaluronan (HA) undergoes diffuse changes as part of the diabetic macroangiopathy. Because HA influences the phenotype of vascular cells in vitro such as proliferation, migration, and secretion, it is tempting to speculate that diabetes-induced hastened cardiovascular disease may be linked to the increased amount of HA. To explore the pathophysiological role of altered HA content in the arterial wall in vivo, we created transgenic (Tg) mice with HA overexpression in smooth muscle cells (SMCs) in large and small vessels, targeted by the smooth-muscle-cell-actin (SMA) promoter fused to the human hyaluronan synthase 2 (hHAS2) cDNA. RT-PCR demonstrated hHAS2 mRNA expression in the tunica media of large and small vessels. In situ hybridization confirmed that hHAS2 mRNA was targeted to the SMCs. The aortic HA content was elevated in the Tg mice, and by immunohistochemistry, it was seen that HA accumulated in the tunica media. The secretory profile of high- and low-molecular HA was similar in wild-type and Tg animals. Overproduction of HA in the aorta resulted in thinning of the elastic lamellae in Tg mice. Our data suggest that this may lead to increased mechanical stiffness and strength, as determined by controlled stretching until failure. Finally, overproduction of HA on the genetic background of the ApoE-deficient mouse strain promoted atherosclerosis development in the aorta. These results indicate that a single component of the diabetic macroangiopathy, diffuse accumulation of HA, accelerates the progression of atherosclerosis.
Key Words: hyaluronan biomechanics atherosclerosis diabetes transgenic
Introduction
Hyaluronan (HA) is a large, nonsulfated glycosaminoglycan (GAG) produced in the vasculature by smooth muscle cells (SMCs) and endothelial cells (ECs). It is synthesized at the inner face of the cell membrane by hyaluronan synthases 1 to 3 (HAS 1 to 3), followed by translocation to the outer surface and the intercellular space.1 Recent in vitro investigations have shown that high (HMW) as well as low molecular (LMW) forms of HA influence important cellular functions, such as proliferation, migration, and secretory capabilities.1eC4 There seems to exist a delicately regulated balance between production, sizing, and removal of HA that is central to its biological functions during normal conditions. If this balance is disturbed, it has been hypothesized that disease may develop.5 The notion that HA is likely to take part in the development of vascular pathologies has been nourished by a number of observations, which show altered HA amounts in several arterial disease entities such as diabetic angiopathy, atherosclerosis, and restenosis.6eC8 During the build-up of the atherosclerotic plaques in humans, some areas with local accumulation of HA appear.9eC11 This is in contrast to the situation in patients with diabetes in whom the HA accumulation is disseminated in tunica media and is regarded as one important element in a series of diffuse matrix changes in the vessels.6,12 The diabetic angiopathy deserves some specific comments. First, a very high and equal frequency of cardiovascular disease (CVD) exists among men and women with diabetes for as yet unexplained reasons.13eC15 In line with this, it is interesting that in subjects with long-term diabetes the increasing mortality from CVD is independent of any well-known risk factors such as hypertension, dyslipidemia, and microalbuminuria.16,17 Second, the vessel wall thickness is increased in juvenile patients with a diabetes duration of 5 years.18,19 Third, histochemical and biochemical records demonstrate that atheromatosis free arterial wall segments from patients with diabetes contain excessive amounts of HA, type IV collagen, and fibronectin in the tunica media.6,12 Finally, in vitro studies reveal that factors from the diabetic metabolism affect the release of HA from SMCs.20 In concert these studies indicate that matrix alterations in the arterial wall in diabetes may be part of a generalized angiopathy. We hypothesize that these changes make the vessel wall more susceptible to atherogenic stimuli, thus contributing to the increased prevalence of CVD in diabetic patients. To address the biological effects of increased HA content in the arteries in vivo, we created transgenic mice with disseminated overexpression of HA in the tunica media of large and small vessels. We investigated the ultrastructural changes in the arterial wall, the stiffness and strength of aortic rings, and the influence on the development of atherosclerosis in ApoE-deficient mice.
Materials and Methods
Production of hHAS2 Transgenic Mice
A restriction fragment containing the mouse SMC-actin 5' flanking region SMP8 (eC1074 bp, 63 bp of 5'UT and 2.5 kb of intron 1), the human hyaluronan synthase 2 (hHAS2) cDNA, and a SV40 polyA terminator sequence21,22 was gel-purified using GenecleanII (BIO101) and microinjected into pronuclei of fertilized eggs from F1 (C57BL/6JxDBA/2J) females crossed with F1 (C57BL/6JxDBA/2J) males (Taconic Europe, Ry, Denmark). The progeny from backcrosses to C57BL/6J were identified as transgenic (Tg) or nontransgenic (WT) by polymerase chain reaction (PCR) with primers complementary to sequences in the SV40 fragment. hHAS2 transgenic animals (>3-generation C57BL/6J) were crossed with ApoE-deficient (ApoEeC/eC) mice (>10-generation C57BL/6J) (Bomholtgaard Breeding and Research Center) to generate animals that were hHAS2+ApoEeC/eC or hHAS2eCApoEeC/eC (>3 to 4 generation). Unless otherwise stated, mice were euthanized by cervical dislocation. For details, including legal approvals, see the expanded Material and Methods in the online data supplement available at http://circres.ahajournals.org.
RNA Isolation and RT-PCR
Total RNA was reversed transcribed to gain cDNA. Amplification of cDNA was performed using the following primers: hHAS2, 5'-GCATCACACCTCATCATCC-3' and 5'-GACTTCTCTTT-TTCCACCCC-3'. Murine -actin (m-actin), 5'-CTACAAT-GAGCTGCGTGTGGC-3' and 5'-GTCCAGACGCAGGATGG-CATG-3' and accomplished as described online. The PCR products were separated by electrophoresis in a 2% agarose gel with ethidium bromide. The intensity of the visualized bands was measured in arbitrary absorbance units with a BIO-RAD UV-gel camera. Results are presented as the ratio between the density of hHAS2 and m-actin. For details, see expanded Material and Methods.
In Situ Hybridization
Hybridization of deparaffinized sections from selected organs was performed with 35S-UTPeClabeled sense or antisense RNA probes (6x105 cpm/section) according to standard methods (see expanded Material and Methods).
HA Immunoassay
Tissues were excised, and the aorta was dissected into intimamedia. The tissues were weighed, homogenized in 2 mL of 0.01 mmol/L Tris-HCl and 0.2 mmol/L EDTA pursued by digestion with protease from Streptomyces Griseus (1 mg/mL, Sigma) overnight at 37°C and followed by heat inactivation (100°C for 30 minutes). After centrifugation (30 000g for 10 minutes), the HA content in the supernatants was measured by an HA ELISA (Chugai Diagnostics Science Co). The detection limit of the assay was 10 ng/mL and the intra-assay CV% was 3.6% to 4.7%. The plasma HA concentration was measured using the same ELISA.
Histochemistry
We stained paraffin or cryosections by standard methods with orcein, biotinylated HABP (1:250, Seikagaku Co), and antibodies against SMC-actin (1:150, NeoMarkers) and CD11b/CD18 (Mac1) (1:150, Chemicon) (see expanded Material and Methods).
Determination of HA Molecular Weight by High-Performance Liquid Chromatography
The aorta (Tg: n=6, WT: n=4) was dissected and chopped into small pieces that were distributed in two wells (96-well plate) and incubated at 37°C for 24 hours in MEM containing 40 e藽i D-[6-3H]-glucosamine-hydrochloride (Amersham Biosciences, UK). Tissue from two mice were pooled, homogenized, and extracted with guanidinium-chloride. Diluted supernatants were passed through an anion exchange cellulose column (DEAE-52, Whitman). The HA-containing elute underwent dialyze and freeze-drying. The samples were resuspended in 0.1 mol/L CH3COONH4, pH 7.2. One part was treated with Hyaluronidase (Streptomyces hyaluronlyticus) (15 U/e蘈, Calbiochem, VWR Int. GmbH), the other sample was left untreated. All samples (Tg, n=3; WT, n=2), were separated by high-performance liquid chromatography (HPLC) (Superose 6 HR 10/30 column) (PharmaciaBiotech). Fractions were collected and counted. For details, see expanded Material and Methods.
Electron Microscopy
Aortic ring profiles were embedded in Epon for electron microscopy and 50-nm sections were examined ultrastructurally with a digitalized Philips CM10 electron microscope (SIS 3.0). All vascular profiles were scored according to the amount of the intercellular ground substance (clear areas) and the degree of elastin fragmentation (0=absent/low, 1=moderate, 2=severe). The width of 4 to 7 elastic lamellae, including membrana elastica interna and externa, and the distance between the lamellae, were measured at two or four positions chosen at random. Mean values of the parameters were obtained for each mouse before group means were calculated. All measurements were performed in a blinded fashion. For details, see expanded Materials and Methods.
Mechanical Testing and Hydroxyproline Content
Freshly excised aortas were stored at eC20°C until analysis. The lower thoracic aorta was cleaned of adhering fat and loose connective tissue. Aortic ring specimens (1-mm high; in average 3.5 specimens per aorta) were cut between intercostal arteries and mounted in a materials testing machine (Alwetron TCT5, Lorentzen & Wettre). Then the specimen was stretched (10 mm/min) until failure by pulling the wires apart in a direction perpendicular to the wires. Thereafter, the collagen content in the ring specimen was estimated by hydroxyproline determination. From the recorded load-deformation values, the load-strain curve was calculated and maximal load and maximal stiffness (ie, maximal slope of the load strain curve) were derived. In calculation of strain, the original circumference of the ring specimen was defined as the circumference at which the specimen attained a minimal load (1% of maximum). Mean values of the parameters were obtained for each mouse before group means were calculated.
Analysis of Atherosclerosis in hHAS2+ApoEeC/eC and hHAS2eCApoEeC/eC Transgenic Mice
The heart and entire aorta were removed from hHAS2+ApoEeC/eC and hHAS2eCApoEeC/eC littermates after 16 to 20 weeks on normal diet. The processing of the root is described in the expanded Material and Methods. The average atherosclerotic lesion area in three levels was determined using Olympus BX50 microscope and analysis software. The extent of atherosclerosis in the aortic arch and thoracic aorta was measured en face by visualizing cholesterol deposits with Oil Red O. The surface area of the lesion within the arch was expressed as percentage of the arch covered by lesion calculated by means of analysis software including multiple-image alignment. The area with positive staining in the thoracic aorta was calculated by using a frame (175 349.1 e2/144 crosses; CAST, Olympus) that was moved stepwise, from side to side all along the aorta. All analyses were made in a coded fashion by the same investigator.
Statistics
Statistical analysis between two groups of data were performed by Student t test after exclusion of variance in-homogeneity (tested by F-test). Otherwise, we used the ManneCWhitney rank sum test. Statistical analysis between more than two groups was performed by one-way ANOVA with post hoc Bonferroni. In the case of variance in-homogeneity as tested by the Bartlett test, the KruskalleCWallis test was used followed by the ManneCWhitney rank sum test, which was corrected for the number of comparisons. Data were tested for outliners. Differences were considered statistically significant at a level of P0.05.
Results
Expression of hHAS2 RNA in the Tunica Media of Large and Small Vessels (RT-PCR and ISH)
Four of 26 pups were Tg. One died before transmission of the transgene. Thus, three lines were propagated for investigations (5986, 5989, and 5990), of which all expressed hHAS2 RNA in aorta, kidney, and bladder (5986, n=6; 5989, n=4; 5990, n=6) (Figure 1). The hHAS2 RNA amount was higher in the bladder versus the aorta in all lines (all probability values <0.01). The level of hHAS2 RNA in the aorta was approximately two times higher in line 5990 versus line 5986 (P<0.05). Line 5989 showed the lowest hHAS2 expression. hHAS2 RNA was absent in WT mice (n=6) (Figure 1). Targeting of hHAS2 RNA production to the SMCs of the media of the aorta and smaller renal vessels and of the bladder was confirmed by in situ hybridization (ISH) in Tg mice of line 5986 and 5990 (n=2 for each line) (Figure 2A). Few signals were found in the media of large and small vessels in Tg mice of line 5989 (n=2) and none were seen in WT mice (n=2) (Figure 2A). At this stage, line 5989 was terminated.
Increased Aortic HA Content
The HA content (ELISA) in selected organs from 8- to 14-month-old mice are given in the Table. The HA amount in the aorta was higher in Tg mice of line 5986 (n=12) and 5990 (n=4) versus WT animals (n=9) (both P<0.05). The HA content in the kidney only reached statistical significance in mice of line 5986 (P<0.05). Both lines exhibited elevated HA levels in the bladder; however, there were no statistically significant differences versus WT mice (Table). The HA content in the selected organs was similar when comparing the two lines of transgenic mice (Table). A similar pattern of HA production in the kidney and bladder was seen in younger, 4- to 5-month-old Tg mice (line 5986) (Bladder, 44.9±11.2 ng/mg; Kidney, 1.31±0.64 ng/mg; n=6) versus WT mice (Bladder, 46±11.2 ng/mg; Kidney, 0.35±0.05 ng/mg; n=5) (P>0.05 and P=0.01, respectively). Immunohistochemistry (IHC) demonstrated that HA was localized in the tunica media of the aorta (n=4 for both line 5986 and 5990), and the smaller vessels in the kidney (n=4 for both line 5986 and 5990) and retina (n=9, line 5986) of Tg mice (Figure 2B). On the other hand, WT mice exhibited weak HA expression (for each organ n=4 to 7). In the bladder, HA was present in the connective tissue layer of the mucosa and the muscularis in all tested mice; however, it was difficult to detect any differences in the staining pattern in Tg (n=4 for both line 5986 and 5990) versus WT mice (n=4) (data not shown). The mean plasma HA concentration was similar among Tg mice of line 5986 (705±156 ng/mL; n=8), line 5990 (621±98 ng/mL; n=8), and WT mice (650±80 ng/mL; n=9) (P>0.05).
Molecular Weight Characterization
The distribution of HMW and LMW HA forms produced in vitro from cultured slices of aorta exhibited a similar pattern in Tg (n=3) versus WT mice (n=2) (Figure 3). In all incidents, the HA isolated and purified from overnight incubated aorta slices were eluted as two major molecular size species G1 (fractions 8 to 10) [Tg, 55±5.2% (n=3); WT, 53±6% (n=2); G2 (fractions 16 to 20) Tg, 40±5% (n=3); WT, 44±6% (n=2)]. The ratio between the peaks was 1.5±0.29 in Tg (n=3) versus 1.3±0.31 in WT (n=2) mice. The first peak corresponds to the void volume equivalent to a molecular size of >1x106 Da. The second peak corresponds to molecular size species a little larger than 1.1x103 Da. Both peaks represent HA, because they disappeared after treatment of parallel handled samples with Streptomyces hyaluronlyticus before HPLC (data not shown).
Thinning of the Elastic Membranes in the Aorta of Transgenic Mice
Orcein staining did not reveal any obvious differences in the vascular morphology between WT and Tg mice (Figure 4). By electron microscopy we found, however, that the average width of the elastic lamellae in Tg (n=8) and WT (n=7) mice was 0.763±0.097 and 0.926±0.083 e, respectively (P<0.001). The distance between the elastic lamellae (Tg, 3.831±1.015 e; WT, 3.861±1.142 e) and the total media thickness (Tg, 26.7±4.26 e; WT, 25.4±7.9 e) were similar in the two groups. More ground substance was visible in the space between the elastic lamellae and the SMCs in Tg versus WT mice (P<0.01). Debris from the elastic fibers was more often seen in Tg mice than in normal animals, and fragmentation of the elastic membranes was clearly more evident in Tg mice versus WT controls (P<0.01) (Figure 4).
Alterations of the Biomechanical Properties of the Aorta in Transgenic Mice
The maximal load and the maximal stiffness were significantly different among the groups of both the 3-month-old (P=0.029 and P=0.028, respectively) and 6-month-old mice (P=0.009 and P=0.061, respectively) (Figure 5). At 3 months of age, both the maximal load and the maximal stiffness were higher in Tg mice of line 5986 (n=6) versus age-matched WT mice (n=10) (both P<0.05), whereas the values of line 5990 (n=6) were in between. At 6 months of age, the maximal load was higher in Tg mice of line 5986 (n=6) versus age-matched WT mice (n=8) and mice of line 5990 (n=6) (both P<0.05), whereas the maximal stiffness, with the highest value in mice of line 5986, tended to be different among the groups (P=0.061) (Figure 5). In all tested groups of mice, both the aortic extensibility and the total collagen content expressed as mg per mm of aortic circumference were similar (data not shown).
Promotion of Atherosclerosis in hHAS2+ApoEeC/eC Transgenic Mice
We backcrossed hHAS2 transgenic mice (line 5986) with ApoEeC/eC animals to obtain hHAS2+ApoEeC/eC and hHAS2eCApoEeC/eC animals. The HA staining was profound in the tunica media of the aortic root (n=9), the aortic arch (n=6), and the thoracic aorta (n=1) in hHAS2+ApoEeC/eC, whereas it was weak in hHAS2eCApoEeC/eC animals (root, n=9, aortic arch, n=7, and thoracic aorta, n=1) (Figure 6A). The HA content was higher in the aortic arch, thoracic aorta, and the bladder in hHAS2+ApoEeC/eC versus hHAS2eCApoEeC/eC mice (all P<0.05) (Table). hHAS2+ApoEeC/eC mice exhibited only marginally elevated HA levels in the kidney (P=0.28) (Table). At 16 to 20 weeks of age, the extent of the atherosclerotic lesions in the aortic root were higher 80 and 160 e from the valve cusps in hHAS2+ApoEeC/eC (n=15) versus hHAS2eCApoEeC/eC (n=9) (both P<0.05) (Figure 7A). The area (%) of lipid deposits in the aortic arch was increased in hHAS2+ApoEeC/eC (n=17) versus hHAS2eCApoEeC/eC (n=10) (P<0.001) (Figure 7B). This difference was also present in the descending aorta (hHAS2+ApoEeC/eC (n=15) and hHAS2eCApoEeC/eC (n=10) (all P<0.05) (Figure 7C). The SMCs were seen throughout the tunica media (Figure 6B). The distribution of SMC-positive cells in the plaques in the aortic root was similar in hHAS2+ApoEeC/eC (n=5) versus hHAS2eCApoEeC/eC (n=8). They were primarily seen under the fibrous capsule and scattered around in the central area of the plaques (Figure 6B). The presence of HA was obvious in areas with SMCs in hHAS2+ApoEeC/eC but not in hHAS2eCApoEeC/eC (Figure 6B). Both genotypes exhibited HA staining in areas with cholesterol crystals. This may, however, be due to background staining (Figure 6B). Also the distribution of CD11b/CD18-positive cells in the plaques in the aortic root was similar in hHAS2+ApoEeC/eC (n=8) versus hHAS2eCApoEeC/eC (n=7). The macrophages were primarily seen central in the lesion and in the underlying tunica media and adventitia (Figure 6B).
Discussion
This study clearly indicates that raised HA levels in the aortic wall significantly transform the property of the vessel, leading to reduced thickness of the elastic lamellae, altered biomechanical properties, and most importantly, an increased susceptibility to atherogenic insults. This is in line with observations reported by Renard et al23 who suggest that GAG accumulation is one of the first steps in initiation of diabetes-induced atherosclerosis.
Our investigations used transgenic mice in which hHAS2 expression, driven by the murine SMC-actin promoter, was able to generate arterial hHAS2 RNA, leading to raised HA levels with a normal distribution of HMW versus LMW forms.5 We found a relatively high HA content in WT aorta, which at a first glance seems contradictory to the almost absent HA staining in the tunica media in the WT mice. It should be underlined, however, that the vessel extracts probably include some components from the tunica adventitia that normally contain HA. Clearly, our data show that accumulation of HA occurs especially in the tunica media. The circulating HA levels were similar in transgenic and WT mice. This suggests that HA in our model acts in an autocrine or paracrine fashion. It is generally believed that HA throughout the body is cleared by receptor-mediated uptake mediated by the hyaluronan receptor for endocytosis (HARE), which is found in lymph node, liver, and spleen sinusoidal endothelial cells.24 Tissue HA is mainly removed by lymph and passed to the lymph nodes in which it is degraded. Approximately 15% of HA exits the lymph nodes, enters the blood, and is removed by the liver sinusoidal endothelial cells,25 which has a rather high clearance capacity.26 Taken together, it is not surprising that the HA plasma levels are normal in Tg mice.
The higher HA content tend to increase the maximal load and the maximal stiffness of the vessel in hHAS2 transgenic mice. This observation was manifest in one of our transgenic lines (5986) at 3 and 6 months of age. Furthermore, a trend toward increased maximal load and maximal stiffness was seen in 3-month-old mice of line 5990. However, this trend disappeared with age. Our observations suggest that HA may influence the biomechanical properties of the aorta. Hyaluronan promotes the shrinkage of collagen gels by aortic SMCs in vitro.27 This observation is believed to be associated with changes in collagen reorganization and cell shape and partly mediated by activation of the CD44 signaling pathway.27 In line with this is the observation that blocking of RHAMM (receptor of HA-mediated motility) is linked to attenuated collagen gel contraction by SMCs isolated from the bladder exposed to stretch injury.28 Furthermore, in accordance with the minimal strength and stiffness of elastin compared with collagen,29 the thinning and fragmentation of the elastic membranes in the Tg mice were not found to be reflected in neither reduced mechanical strength, stiffness, and extensibility of the aorta or changes in the toe-part of the load-strain curves corresponding to the normal in vivo range of loading (data not shown). Therefore, the observed changes of the elastic membranes more likely contribute to collagen reorganization resulting in increased strength and stiffness of the aortic wall in Tg mice. This is in line with observations in a mouse model of Marfan syndrome and in experiments evaluating the impact of elastases on aortic elasticity, in which fragmentation of the elastic membranes is associated with increased stiffness of the vessel wall.30,31
The most prominent observation in the present study is that ApoE-deficient animals with increased HA levels in the vessel wall develop accelerated atherosclerosis. It should be underlined that atherosclerosis is absent in our hHAS2 mice on the normal C57 genetic background. Thus, the presence of one element of the diabetic macroangiopathy in combination with an atherogenic milieu, but without hyperglycemia, accelerates the development of atherosclerotic lesions. Several observations have implicated HA in a variety of the biological processes involved in atherogenesis,9eC11 but our data show for the first time, that disseminated accumulation of HA accelerates the atherogenic process in vivo. Several mechanisms are likely to play a role in the augmented atherogenesis. First, a recent report has described an attenuated development of atherosclerosis in mice genetically engineered to lack CD44.32 Because HA is the principal ligand for CD44, the observation is compatible with our results. Recruitment of macrophages was decreased and SMCs were in a quiescent state in the CD44-deficient mice. These are in line with in vitro data showing that HA accumulation may be important for macrophages/foam cell migration and activity through HA-CD44 receptor interactions.33,34 Likewise, SMC proliferation and migration is modulated by HA.2 Thus, a HA-rich environment is believed to promote SMCs as well as macrophage secretory capabilities, migration, and proliferation. Clearly, both alterations in the phenotypic stage of SMCs and macrophages may play a role in the increased atherogenesis. Presently, we can only say that there is no overwhelming difference in the pattern of macrophage infiltration of the atherosclerotic plaques in hHAS+ApoEeC/eC versus hHAS2eCApoEeC/eC mice. However, it is as yet unknown whether the functional status of the infiltrating macrophages and SMCs are altered. Secondly, augmented atherogenesis may relate to the observed biomechanical alterations, because it has been hypothesized that increased arterial stiffness may give rise to an atherosclerosis-prone vessel wall. This proposal is based both on observational data about relations between arterial stiffness and atherosclerosis35 and on experimental results, linking increased arterial stiffness with marked alterations in endothelial cell function, putatively leading to vascular damage.36 Finally, it is possible that lipoprotein transport may be altered due to modifications in the arterial wall, and because HA binds lipoproteins, this may accelerate atherogenesis in hHAS2+ApoEeC/eC mice.37 It is intriguing that our findings are in agreement with a recent publication by Sussmann et al.9 They report that upregulation of HAS2 expression via EP2 and IP receptors may contribute to HA accumulation during the development of the atherosclerotic plaques, thereby mediating proatherogenic functions of COX2. On the other hand, downregulation of HAS2 expression may mediate the antiatherogenic properties of COX2 inhibitors.9 The exact mechanisms behind the HA-induced atherogenesis, however, needs further exploration.
In summary, our in vivo observations suggest that accumulation of HA in the aortic tunica media is followed by an increased mechanical stiffness and strength of the aorta. Furthermore, the accumulation of HA stimulates the development of atherosclerosis in ApoEeC/eC mice. Based on the observations reported in this study, we propose that increased levels of HA in the vascular wall promote the progression of atherosclerotic lesions. Moreover, the obtained knowledge of disseminated HA accumulation as a promoter of atherogenesis may have implications for the understanding of diabetic angiopathy. In diabetes, the amount of HA, and other extracellular matrix components in the tunica media, is increased in atheromatosis-free areas.6,12 Our data are compatible with the notion that raised levels of HA in the tunica media, as part of the diabetic macroangiopathy, may facilitate the development of atherosclerosis among patients with diabetes, with HA being one missing factor linking diabetes and diabetes-induced atherosclerosis.
Acknowledgments
The project was supported by the Danish Medical Research Council (grants nos. 9601759, 9802621, and 9600822; Aarhus University, Novo Nordisk Center for Research in Growth and Regeneration), Fougner-Hartmanns Foundation, the Novo Nordisk Foundation, The Danish Diabetes Association, and Clinical Institute, University of Aarhus, Denmark. We are indebted to Lotte Arentoft, Anita Morell, Karin Vestergaard, Lone Lysgaard, Annette Berg, and Eva K. Mikkelsen for their technical assistance. Carsten Hjskov is thanked for helping with the HPLC.
References
Lee JY, Spicer AP. Hyaluronan: a multifunctional, megaDalton, stealth molecule. Curr Opin Cell Biol. 2000; 12: 581eC586.
Evanko SP, Angello JC, Wight TN. Formation of hyaluronan- and versican-rich pericellular matrix is required for proliferation and migration of vascular smooth muscle cells. Arterioscler Thromb Vasc Biol. 1999; 19: 1004eC1013.
Savani RC, Wang C, Yang B, Zhang S, Kinsella MG, Wight TN, Stern R, Nance DM, Turley EA. Migration of bovine aortic smooth muscle cells after wounding injury: the role of hyaluronan and RHAMM. J Clin Invest. 1995; 95: 1158eC1168.
Lokeshwar VB, Selzer MG. Differences in hyaluronic acid-mediated functions and signaling in arterial, microvessel, and vein-derived human endothelial cells. J Biol Chem. 2000; 275: 27641eC27649.
Tammi MI, Day AJ, Turley EA. Hyaluronan and homeostasis: a balancing act. J Biol Chem. 2002; 277: 4581eC4584.
Heickendorff L, Ledet T, Rasmussen LM. Glycosaminoglycans in the human aorta in diabetes mellitus: a study of tunica media from areas with and without atherosclerotic plaque. Diabetologia. 1994; 37: 286eC292.
Riessen R, Wight TN, Pastore C, Henley C, Isner JM. Distribution of hyaluronan during extracellular matrix remodeling in human restenotic arteries and balloon-injured rat carotid arteries. Circulation. 1996; 93: 1141eC1147.
Turley EA. Extracellular matrix remodeling: multiple paradigms in vascular disease. Circ Res. 2001; 88: 2eC4.
Sussmann M, Sarbia M, Meyer-Kirchrath J, Nusing RM, Schror K, Fischer JW. Induction of hyaluronic acid synthase 2 (HAS2) in human vascular smooth muscle cells by vasodilatory prostaglandins. Circ Res. 2004; 94: 592eC600.
Papakonstantinou E, Roth M, Block LH, Mirtsou-Fidani V, Argiriadis P, Karakiulakis G. The differential distribution of hyaluronic acid in the layers of human atheromatic aortas is associated with vascular smooth muscle cell proliferation and migration. Atherosclerosis. 1998; 138: 79eC89.
Levesque H, Girard N, Maingonnat C, Delpech A, Chauzy C, Tayot J, Courtois H, Delpech B. Localization and solubilization of hyaluronan and of the hyaluronan-binding protein hyaluronectin in human normal and arteriosclerotic arterial walls. Atherosclerosis. 1994; 105: 51eC62.
Andresen JL, Rasmussen LM, Ledet T. Diabetic macroangiopathy and atherosclerosis. Diabetes. 1996; 45: S91eCS94.
Kannel WB, McGee DL. Diabetes and cardiovascular disease. The Framingham study. JAMA. 1979; 241: 2035eC2038.
Lloyd CE, Kuller LH, Ellis D, Becker DJ, Wing RR, Orchard TJ. Coronary artery disease in IDDM. Gender differences in risk factors but not risk. Arterioscler Thromb Vasc Biol. 1996; 16: 720eC726.
Howard BV, Hsia J, Ouyang P, Van Voorhees L, Lindsay J, Silverman A, Alderman EL, Tripputi M, Waters DD. Postmenopausal hormone therapy is associated with atherosclerosis progression in women with abnormal glucose tolerance. Circulation. 2004; 110: 201eC206.
Kuusisto J, Mykkanen L, Pyorala K, Laakso M. NIDDM and its metabolic control predict coronary heart disease in elderly subjects. Diabetes. 1994; 43: 960eC967.
Spijkerman AM, Dekker JM, Nijpels G, Jager A, Kostense PJ, van Hinsbergh VW, Bouter LM, Heine RJ, Stehouwer CD. Impact of diabetes duration and cardiovascular risk factors on mortality in type 2 diabetes: the Hoorn Study. Eur J Clin Invest. 2002; 32: 924eC930.
Jarvisalo MJ, Jartti L, Nanto S, Irjala K, Ronnemaa T, Hartiala JJ, Celermajer DS, Raitakari OT. Increased aortic intima-media thickness: a marker of preclinical atherosclerosis in high-risk children. Circulation. 2001; 104: 2943eC2947.
Jarvisalo MJ, Putto L, Jartti L, Lehtimaki T, Solakivi T, Ronnemaa T, Raitakari OT. Carotid artery intima-media thickness in children with type 1 diabetes. Diabetes. 2002; 51: 493eC498.
Erikstrup C, Pedersen LM, Heickendorff L, Ledet T, Rasmussen LM. Production of hyaluronan and chondroitin sulphate proteoglycans from human arterial smooth muscle - the effect of glucose, insulin, IGF-I or growth hormone. Eur J Endocrinol. 2001; 145: 193eC198.
Min BH, Foster DN, Strauch AR. The 5'-flanking region of the mouse vascular smooth muscle alpha-actin gene contains evolutionarily conserved sequence motifs within a functional promoter. J Biol Chem. 1990; 265: 16667eC16675.
Watanabe K, Yamaguchi Y. Molecular identification of a putative human hyaluronan synthase. J Biol Chem. 1996; 271: 22945eC22948.
Renard CB, Kramer F, Johansson F, Lamharzi N, Tannock LR, von Herrath MG, Chait A, Bornfeldt KE. Diabetes and diabetes-associated lipid abnormalities have distinct effects on initiation and progression of atherosclerotic lesions. J Clin Invest. 2004; 114: 659eC668.
Zhou B, McGary CT, Weigel JA, Saxena A, Weigel PH. Purification and molecular identification of the human hyaluronan receptor for endocytosis. Glycobiology. 2003; 13: 339eC349.
Weigel JA, Raymond RC, McGary C, Singh A, Weigel PH. A blocking antibody to the hyaluronan receptor for endocytosis (HARE) inhibits hyaluronan clearance by perfused liver. J Biol Chem. 2003; 278: 9808eC9812.
Suzuki M, Takeuchi H, Kakita T, Unno M, Katayose Y, Matsuno S. The involvement of the intracellular superoxide production system in hepatic ischemia-reperfusion injury. In vivo and in vitro experiments using transgenic mice manifesting excessive CuZn-SOD activity. Free Radic Biol Med. 2000; 29: 756eC763.
Travis JA, Hughes MG, Wong JM, Wagner WD, Geary RL. Hyaluronan enhances contraction of collagen by smooth muscle cells and adventitial fibroblasts: Role of CD44 and implications for constrictive remodeling. Circ Res. 2001; 88: 77eC83.
Bagli DJ, Joyner BD, Mahoney SR, McCulloch L. The hyaluronic acid receptor RHAMM is induced by stretch injury of rat bladder in vivo and influences smooth muscle cell contraction in vitro [corrected]. J Urol. 1999; 162: 832eC840.
Shadwick RE. Mechanical design in arteries. J Exp Biol. 1999; 202: 3305eC3313.
Marque V, Kieffer P, Gayraud B, Lartaud-Idjouadiene I, Ramirez F, Atkinson J. Aortic wall mechanics and composition in a transgenic mouse model of Marfan syndrome. Arterioscler Thromb Vasc Biol. 2001; 21: 1184eC1189.
Dobrin PB, Baker WH, Gley WC. Elastolytic and collagenolytic studies of arteries. Implications for the mechanical properties of aneurysms. Arch Surg. 1984; 119: 405eC409.
Cuff CA, Kothapalli D, Azonobi I, Chun S, Zhang Y, Belkin R, Yeh C, Secreto A, Assoian RK, Rader DJ, Pure E. The adhesion receptor CD44 promotes atherosclerosis by mediating inflammatory cell recruitment and vascular cell activation. J Clin Invest. 2001; 108: 1031eC1040.
McKee CM, Penno MB, Cowman M, Burdick MD, Strieter RM, Bao C, Noble PW. Hyaluronan (HA) fragments induce chemokine gene expression in alveolar macrophages. The role of HA size and CD44. J Clin Invest. 1996; 98: 2403eC2413.
McKee CM, Lowenstein CJ, Horton MR, Wu J, Bao C, Chin BY, Choi AM, Noble PW. Hyaluronan fragments induce nitric-oxide synthase in murine macrophages through a nuclear factor kappaB-dependent mechanism. J Biol Chem. 1997; 272: 8013eC8018.
Weber T, Auer J, O‘Rourke MF, Kvas E, Lassnig E, Berent R, Eber B. Arterial stiffness, wave reflections, and the risk of coronary artery disease. Circulation. 2004; 109: 184eC189.
Peng X, Haldar S, Deshpande S, Irani K, Kass DA. Wall stiffness suppresses Akt/eNOS and cytoprotection in pulse-perfused endothelium. Hypertension. 2003; 41: 378eC381.
Srinivasan SR, Yost K, Radhakrishnamurthy B, Dalferes ER, Jr., Berenson GS. Lipoprotein-hyaluronate associations in human aorta fibrous plaque lesions. Atherosclerosis. 1980; 36: 25eC37.(Song Chai, Qing Chai, Car)