| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Vascular Biology |
From the Center for Human and Clinical Genetics (S.H., N.C., I.S.L.L., M.H.B., D.J.M.P.), the Department of Anatomy and Embryology (J.C.V.M., M.C.D.), Leiden University Medical Center, and the Gaubius Laboratory (N.V.L., R.H.), TNO BioSciences, Leiden, The Netherlands.
Correspondence to Dr D.J.M. Peters, Department of Human Genetics, Center for Human and Clinical Genetics, Leiden University Medical Center, Postal zone: S-04-P, P.O. Box 9600, 2300 RC, Leiden, The Netherlands. E-mail d.j.m.peters{at}lumc.nl
| Abstract |
|---|
|
|
|---|
Methods and Results— Quantitative real-time polymerase chain reaction revealed that in the aorta the expression of normal Pkd1 is reduced to approximately 26%. Using (immuno)histochemistry we have characterized the pathogenetic sequence for dissecting aneurysm formation. The aorta shows regions with accumulation of matrix components between the elastin lamellae. This is followed by increased numbers of smooth muscle cells and locally weakening of the media. In the intima, accumulation of matrix components and detachment of endothelial cells from the elastin lamellae results in a tear. The combination of weak media and a tear in the intima leads to rupture of the vessel wall resulting in intramural bleeding.
Conclusions— The Pkd1nl/nl mouse reveals that polycystin1 is implicated in maintenance of the vessel wall structural integrity, and it is a useful model for dissecting aneurysm formation studies.
Autosomal Dominant Polycystic Kidney Disease (ADPKD) is a common genetic disease associated with dissecting aneurysms. We characterized dissecting aneurysm formation in a Pkd1 mouse model. It reveals the importance of polycystin1 in the maintenance of the vessel wall integrity, and it is a useful model to study dissections in ADPKD.
Key Words: dissecting aneurysm polycystic kidney disease aorta extracellular matrix Pkd1
| Introduction |
|---|
|
|
|---|
Several Pkd1 and Pkd2 knock-out mouse models have been generated, however knock-out embryos die at embryonic days 13.5 to 14.5 from a primary cardiovascular defect.9–12 These embryos show edema, focal vascular leaks, and hemorrhage. The fact that these mice die in utero withholds the possibility to study aneurysm formation in ADPKD. Recently, we have generated a Pkd1 hypomorphic mouse model, Pkd1nl/nl, with reduced Pkd1 transcripts in the kidneys.13 Pkd1nl/nl mice are viable, and the majority of the animals die within 1 to 2 months after birth. These mice show dissecting aneurysms and cysts in kidneys, liver, and pancreas, which is in line with the pathogenic features found in the human ADPKD phenotype. Pkd1nl/nl mice are the first viable ADPKD mice with spontaneous dissecting aneurysm formation in the aorta enabling studies on vascular remodeling in ADPKD. In this study we have carefully characterized dissecting aneurysm formation in these hypomorphic mice. Reduced levels of Pkd1 transcripts measured in the aorta indicate a direct association between Pkd1 and vessel wall integrity.
| Materials and Methods |
|---|
|
|
|---|
Antibodies
SM-
-actin (1A4) 1:2000 (Sigma), Collagen I and III 1:200 (Sanbio), Fibrillin-1 1: 500 (a generous gift from R. Mecham, Washington), Fibronectin 1:4000 (a generous gift from B.N. Bachera, Leiden, The Netherlands), Isolectin 1:20 (Sigma-Aldrich Chemie), Osteopontin 1:100 (a generous gift from L.W. Fisher, Bethesda, Md), Perlecan, Versican alpha and beta 1:20 (a generous gift from J.R. Hassell, Pittsburgh, Pa), Rabbit-anti-mouse Macrophage (AIA 31240) 1:200 (Accurate Chemical and scientific), matrix metalloproteinase (MMP)-2 Ab-4 (75 to 7F7) 1:4 (Calbiochem), MMP-9 11A-5 (11A-5) 1: 100 (Neo Markers), Ki-67 1:3000 (Novocastra), terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) 1:14 (Roche). Secondary antibodies: Rabbit-anti-mouse HRP 1:200, Swine-anti-rabbit HRP 1:200 (DAKO), Goat-anti-rabbit 1:50, Rabbit-peroxidase-anti-peroxidase 1:250 (Nordic Immunologic Laboratories), mouse-biotinylated (ABC-kit vectastain; Brunschwig Chemie).
Immunohistochemistry
After deparaffinization, sections were stained using standard procedures.14 The antibody complexes were visualized using 0.04% diaminobenzidine-tetra-hydrochloride (DAB; Sigma). Antigen retrieval was applied for fibronectin and ki67 staining and for monoclonal antibody 1A4 (SM-
-actin) a complex with the secondary antibody was prepared before incubation. The following histological stainings were performed: Hematoxylin Eosin (HE), Resorcin Fuchsin (RF) for elastin lamellae, Alcian blue for staining glycosaminoglycans and thereby the proteoglycans, von Kossa and alizarin red for calcium deposition and Sirius red for collagen. For details please see the supplemental materials, available online at http://atvb.ahajournals.org.
RNA Analysis
Complete aortas and kidneys of 5 control and 6 hypomorphic mice at the age of 1 month were dissected. Total RNA was isolated and used in a quantitative real-time reverse transcriptase PCR. For detailed protocol please see the supplemental materials.
Aortic Volume Measurements
To measure the volumes of aortic walls (including the dissection) and aortic lumens, we used the volume counting method according to Cavalieri. For details please see the supplemental materials.
Zymography
Whole aortas of 4 mutant and 3 wild-type mice were powdered in liquid nitrogen, and a tissue extract was made in 50 mmol/L Tris pH 7.7, 1.5 mmol/L NaCl and 0.25% Triton X100. MMPs expression was analyzed by zymography. For details please see the supplemental materials.
| Results |
|---|
|
|
|---|
|
Characterization of Aortic Pathology in Pkd1nl/nl Mice
Eight of the 9 examined Pkd1nl/nl mice showed 1 or more abnormalities with different severity at different regions in the descending thoracic and abdominal aorta. Dissections with intramural bleeding were detected in 5 mice. Five mice had progressive dissecting aneurysms showing pseudo lumens all over the descending aorta (3 mice; Figure 2A) or caudal mesenteric artery (2 mice). Two mice showed a local intramural bleeding (Figure 2B). The dissections in the caudal mesenteric arteries were ruptured leading to retroperitoneal bleeding. The volume of the vessel wall in the areas of these progressive dissections was significantly increased in both the thoracic and abdominal aorta compared with the wild-type aortas of the same regions. However, the inner vessel lumen was not changed (Figure 3A and 3B).
|
|
The ascending aorta and segment B, which is located between the left carotid and the left subclavia in the aortic arch, were normal. One mouse showed media thickening in the whole area of the aortic arch and descending thoracic aorta except for segment B and the ascending aorta (Figure 2C). In these 2 normal regions SMCs seemed to keep their connection with the matrix, whereas the surrounding region of the aorta showed media thickening with disjoined elastin lamellae. One mouse showed local progressive medial thickening in the abdominal aorta without intramural bleeding.
In addition to dissecting aneurysms, all descending aortas studied contained normal vessel wall structure and incomplete dissections. Medial thickening was found in the thoracic aorta of 6 mice, in the abdominal aorta of 2 mice and intima thickening was found in the thoracic aorta of 1 mouse and in the abdominal aorta of 2 mice. The abnormalities found, were of varying pathological severity. This allowed us to describe the sequential order of events occurring in the media and intima, resulting in a dissecting aneurysm.
Media Thickening in Pkd1nl/nl Aortas
The media of the aorta of wild-type mice consists of 5 to 6 layers of SMCs alternated with concentric sheets of elastin lamellae (Figure 2D, 2E, and 2H). Surrounding the SMCs extracellular matrix molecules like proteoglycans and fibronectin are present (Figure 2F and 2G). Examination of elastin lamellae in Pkd1nl/nl aortas by Resorcin Fuchsin staining showed an increase in the space between the lamellae without any fragmentation of the lamellae (Figure 2I). The expression levels of proteoglycans and fibronectin were both notably increased in these regions (Figure 2J and 2K) compared with wild-type aortas (Figure 2F and 2G). The proteoglycans versican and perlecan have been shown to be increased in cultures of medial smooth muscle cells from tissue affected by abdominal aortic aneurysm.15 In our mouse model perlecan was increased whereas versican was absent in both wild-type and mutant mice. Using immunohistochemistry no MMP-2 and 9 expression was detected in the wild-type or Pkd1nl/nl aortas (Figure 2Q). Zymography, which is a more sensitive method than immunohistochemistry, revealed none to a very low and similar MMP-2 and MMP-9 expression in both, mutant and wild-type, aortas (supplemental Figure I). At other more progressive regions in the aortas, the elastin lamellae were more disjoined with an increased number of smooth muscle cells in between (Figure 2L). However, the elastin lamellae were still intact (Figure 2M). The SMCs were disarranged with a low SM–
-actin expression (Figure 2O). Few SMCs had condensed nuclei, however these cells did not stain positive for TUNEL (Figure 2R). Macrophage staining did not reveal any immune-infiltrate in this region (Table). No accumulation of fibronectin was detected in this region. However, proteoglycans were still increased (Figure 2N). The distribution of collagen and fibrillin 1 in the vessel wall of Pkd1nl/nl mice was comparable to wild-type mice. No calcification or expression of osteopontin was detected in all vessels analyzed (Table).
|
Intima Thickening in Pkd1nl/nl Aortas
The wild-type intima consists of a single layer of flattened endothelial cells on a basal membrane and a subjacent elastin lamellae known as internal elastic lamina. Histological analysis of the intima of Pkd1nl/nl by Hematoxylin Eosin showed detachment of the ECs from the internal elastic lamina (Figure 4A and 4B). At these regions isolectin staining was decreased indicating a decreased number of ECs or decreased expression of this marker on ECs (Figure 4C and 4D). In other regions the subendothelial region was filled with proteoglycans and the cell numbers were increased (Figure 4E). However, the accumulation of proteoglycans is neither attributable to an increase in versican or perlecan nor did it coincide with increased fibronectin expression as observed in the media thickening (Table). One of the mice showed a tear in the intima next to the degenerative changes within the media resulting into a progressive dissection all over the aorta (Figure 4F). At the entrance of dissection, fragmentation of the elastin lamellae was found, allowing the blood to spread through the aortic media. Despite substantial effort, we could not locate the exact site of rupture in the other animals.
|
Dissecting Aneurysm
Abnormalities in the media and intima were identified in different regions of the aorta, indicating that the reduction of Pkd1 in both SMCs and ECs contributes to the formation of dissecting aneurysms by 2 processes, 1 in the media and the other in the intima. It is not clear whether the initial event of a dissecting aneurysm occurs in the intima or media. We hypothesize that on reducing the expression of polycystin1, the stability of the contact between SMCs and extracellular matrix molecules is reduced. This may disturb some cellular events in the SMCs resulting in a phenotypic change toward a more synthetic phenotype, which means that they produce more matrix components, which accumulate between the elastin lamellae, and start to hyperproliferate (Figure 5B). This results in an increased thickening of the media, which will become weaker (Figure 5C). In the intima the endothelial cells detach from the elastin lamellae (Figure 5D). The number of intima cells increases and the matrix components accumulate in the subendothelial region (Figure 5E). The final event allowing the blood to spread through the aortic media is the tear of the intima (Figure 5F and 5G). When present, next to the degenerative changes within the media, the resistance of the aortic wall to hemodynamic stress reduces, leading to dissection (Figure 5H).
|
| Discussion |
|---|
|
|
|---|
A direct correlation between MMP-mediated elastin degradation and elastin calcification has been reported. Perivascular administration of CaCl2 to abdominal aortas induced calcification and degradation of elastin fibers.17 Focal linear calcifications were also detected in a mouse model with reduction of fibrillin-1 showing fragmentation of elastin lamellae and MMP-9 expression.18 Our data fit into this correlation. In our mouse model MMP-2 and MMP-9 expression was very low, comparable with wild-type expression, and no elastin calcification or elastin degradation were found in the media thickening. In addition, we did not observe influx of inflammatory cells which are the main source of MMPs secretion. These properties of dissecting aneurysm formation in Pkd1nl/nl mice, therefore, seem to be distinct from other models.
Polycystin1 has been found to be a component of large multiprotein complexes associated with cell-matrix contact at focal adhesions in cultures of renal epithelial cells. It is also associated with adherence junctions and desmosomes in confluent cultures of the renal epithelial cell line Madin-Darby canine kidney (MDCK).19–21 Polycystin1 is also implicated in SMC-elastin contact. The expression of Pkd1 is found in dense plaques, which are analogous to focal adhesions in epithelial cells and attach the intracellular contractile filaments of SMCs to the elastin lamellae.22,23 In our Pkd1nl/nl mice a widened space between the elastin lamellae was found, suggesting loss of contact between SMCs and elastin or other extra cellular matrix structures. Loss of physical contacts, and hence signals from these interactions, is associated with a phenotypic change toward a more synthetic phenotype of the SMCs.24 Furthermore, loss of the SMC-elastin interaction may increase SMC proliferation as elastin has a regulatory function controlling SMC proliferation.25 All these features are found in our Pkd1nl/nl mouse model, confirming this regulatory mechanism. Although we did not detect ki67-positive cells in the sections available for analysis, eg, very early stages and dissections (Figure 2P), cell proliferation should have occurred because increased cell layers were present at certain stages of dissection formation (Figure 2L) and these cells did not stain for inflammatory cells. The decreased expression of SM-
-actin and increased expression of proteoglycans and fibronectin in the Pkd1nl/nl mouse model indicate a response toward a more synthetic phenotype. Polycystin1 might be involved in a SMC-matrix adhesion molecule complex, which maintains a stable SMC phenotype and thereby maintaining the structural vessel wall integrity.
Notably, no media thickening was found in segment B of the aortic arch of our mutant mice. The areas adjacent to segment B showed a progressive media thickening, suggesting a normal SMC-matrix contact in this segment (Figure 2C). In wild-type mice the media of this segment shows characteristics in which it differs from the rest of the aorta. It is negative for SM-
-actin and has extensive medial innervations. These SMCs have also a different embryonic origin compared with the descending aorta as it is derived from the fourth pharyngeal artery in the embryonic thoracic arterial tree.14,26 Interestingly, lowering the Pkd1 expression did not show any influence on the structure of this vessel segment. Pkd1 expression may not be critical in maintaining the structural vessel wall integrity in the aortic B-segment.
Both polycystin1 and 2 are probably involved in mechanosensing as they are localized in the primary cilia in renal cells.27 Polycystin1 is also concentrated around centrioles and in particles along primary cilia of human umbilical vein endothelial cells (HUVECs).28 Recently, monocilia were demonstrated on a subset of endothelial cells in low-shear regions.29 In addition, polycystin1 has been shown to colocalize with platelet-endothelial cell adhesion molecule-1 (PECAM-1), a cell adhesion molecule also found to be involved in mechanosensing mechanism on the cell membrane.30,31 PECAM-1 also maintains adherens junction integrity and permeability. The latter function may be influenced by Pkd1 expression, as Pkd1 knock-out embryos showed increased microvascular permeability at endothelium cell-cell junctions, edema, and hemorrhages.10,11 All these factors may weaken the intima and alert the shear stress response, making the intima more prone to rupture.
Thus, lowering of Pkd1 transcription levels induces degenerative alterations in both the intima and media resulting into a dissecting aneurysm. However, intact regions of the aorta were still present in the Pkd1nl/nl mice and the overall construction of the aorta with regard to the elastin lamellae is normal. Furthermore, we did not found any abnormalities in these mice at the age of 1 week (data not shown). The elastin assembly may be independent of Pkd1 expression during development. Matrix components that are involved in elastin organization, like versican, and fibrillin-1, were also not altered compared with the wild-type aortas.32 This indicates that polycystin1 is probably not implicated in vessel wall development but rather maintains the structural integrity of the vasculature.
| Acknowledgments |
|---|
Sources of Funding
This research was funded by the Netherlands organization for scientific research (NWO/ZonMw 016.036.353) and the Dutch Heart foundation (NHS 2002B177).
Disclosures
None.
| Footnotes |
|---|
Original received May 30, 2006; final version accepted June 25, 2007.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
I. S. Lantinga-van Leeuwen, W. N. Leonhard, A. van der Wal, M. H. Breuning, E. de Heer, and D. J.M. Peters Kidney-specific inactivation of the Pkd1 gene induces rapid cyst formation in developing kidneys and a slow onset of disease in adult mice Hum. Mol. Genet., December 15, 2007; 16(24): 3188 - 3196. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2007 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |