Perturbations of Vascular Homeostasis and Aortic Valve Abnormalities in Fibulin-4 Deficient Mice
The Fibulins are a 6-member protein family hypothesized to function as intermolecular bridges that stabilize the organization of extracellular matrix structures. Here, we show that reduced expression of Fibulin-4 leads to aneurysm formation, dissection of the aortic wall and cardiac abnormalities. Fibulin-4 knockdown mice with a hypomorphic expression allele arose from targeted disruption of the adjacent Mus81 endonuclease gene. Mice homozygous for the Fibulin-4 reduced expression allele (Fibulin-4R/R) show dilatation of the ascending aorta and a tortuous and stiffened aorta, resulting from disorganized elastic fiber networks. They display thickened aortic valvular leaflets that are associated with aortic valve stenosis and insufficiency. Strikingly, already a modest reduction in expression of Fibulin-4 in the heterozygous Fibulin-4+/R mice occasionally resulted in small aneurysm formation. To get insight into the underlying molecular pathways involved in aneurysm formation and response to aortic failure, we determined the aorta transcriptome of Fibulin-4+/R and Fibulin-4R/R animals and identified distinct and overlapping biological processes that were significantly overrepresented including cytoskeleton organization, cell adhesion, apoptosis and several novel gene targets. Transcriptome and protein expression analysis implicated perturbation of TGF-β signaling in the pathogenesis of aneurysm in fibulin-4 deficient mice. Our results show that the dosage of a single gene can determine the severity of aneurysm formation and imply that disturbed TGF-β signaling underlies multiple aneurysm phenotypes.
Extracellular elastic fibers supply structure and mechanical elasticity to organs such as large arteries, lungs and skin.1 Elastic fibers are assembled through polymerization of tropoelastin monomers and loss of elastin is associated with aneurysmal degeneration of the aorta.2 The 6-member protein family of Fibulins, which are prominently expressed in blood vessels are hypothesized to function as intermolecular bridges that stabilize the organization of extracellular structures such as elastic fibers and basement membranes.3 Fibulin-4 is found in the medial layers of arteries and heart valves.4
Whereas complete Fibulin-4 knockout mice have recently been generated5 and showed embryonic lethality (E12.5), we generated a Fibulin-4 allele with reduced expression by transcriptional interference through placement of a TKneo targeting construct in a downstream gene (Mus81). Mus81 knockout mice have been generated previously.6,7 In the targeting strategy of Dendouga et al, exons 9 to 12 were replaced by a PGKneo marker flanked by loxP sites and subsequently the marker was excised using Cre-recombinase expressing mice. Mus81 knockout mice from which the selectable marker was removed were born at expected Mendelian frequencies and were indistinguishable from WT littermates in terms of development, growth, immune function and fertility.6
To examine a potential role of Fibulin-4 in elastic fiber assembly and cardio-vascular disease we used our mouse model underexpressing Fibulin-4, which revealed its role in vascular homeostasis and cardiac abnormalities. The phenotype of the mice resembles connective tissue disorders such as Marfan Syndrome (MFS),8 Loeys-Dietz syndrome (LDS)9,10 and Arterial Tortuosity Syndrome (ATS).11 Pathological, hemodynamic, microarray gene expression and TGF-β signaling analyses provide a general mechanistic basis for the universal role of TGF-β dysregulation in the formation of not only severe, but also mild forms of aneurysms. In addition, using micro array analysis we identified new genes and pathways implicated in both aneurysm formation (using heterozygous Fibulin-4 mice) as well as in the response to cardiovascular failure (using homozygous Fibulin-4 mice).
Materials and Methods
Construction of Targeting Vectors and Fibulin-4 Transgenic Mice
Genomic fragments hybridizing to the Mus81 cDNA were subcloned in pBluescript II (Stratagene). The location of the intron-exon borders was determined by DNA sequence data from Celera. A targeting vector was made by inserting a cassette with the neomycin resistance gene driven by the TK promoter in the BglII sites. E14 ES cells (subclone IB10) were cultured in BRL-conditioned medium supplemented with 1000 U/mL leukemia inhibitory factor. A 10 μg portion of the NotI and SalI linearized targeting vector was electroporated into approximately 107 ES cells in 400 μL culture medium. Selection with 200 μg/mL G-418 was started 24 hours after electroporation. After 8 to 10 days, G418-resistant clones were isolated. Screening for homologous recombinants was performed using DNA blot analysis of KpnI-digested DNA with a 300 bp 5′ external probe and confirmed using DNA blot analysis of BamHI-digested DNA with 1 kb 3′ external probe. One clone identified as correctly targeted was injected into C57-BL/6 blastocysts. Chimeras identified on the basis of agouti pigmentation in the coat were backcrossed to C57Bl/6 mice and the agouti offspring were genotyped by Southern blot analysis and PCR. Heterozygous mutant progeny were intercrossed to produce the animals analyzed in this study. Mice were earmarked and genotyped 4 weeks after birth to avoid stress-related vascular injury.
Quantitative Real-Time PCR Expression Analysis
Total RNA was isolated using a Total RNA isolation kit (Qiagen) as described by the manufacturer. Quantitative PCR (Q-PCR) was performed with a DNA Engine Opticon device according to the instructions of the manufacturer (MJ Research). Primer pairs, designed to generate intron-spanning products of 180 to 210bp for Fibulin-4, Mus81 and Cfl-1 were as follows: Fibulin-4: 5′-GGGTTATTTGTGTCTGCCTCG-3′ and 5′-TGGTAGGAGCCAGGAAGGTT-3′, for Mus81: 5′-CAAAGCCTTCCACAAACCC-3′ and 5′-TCATAAGCAGCCAGGAGACT-3′, for Cfl-1: 5′-CCAGAAGAAGTGAAGAAACGC-3′ and 5′-GAAGATGAACACCAGGTCCT-3′. The generation of specific PCR products was confirmed by melting curve analysis (which measures product specificity by the decrease in fluorescence signal when the PCR product is denatured) and gel electrophoresis (using Roche Agarose MS for analyzing small PCR products). Each primer pair was tested with a logarithmic dilution of a cDNA mix to generate a linear standard curve (crossing point (CP) plotted versus log of template concentration), which was used to calculate the primer pair efficiency (E=10(-1/slope)). Hypoxanthine guanine phosphoribosyltransferase1 (Hprt-1) mRNA was used as an external standard. For data analysis, the second derivative maximum method was applied: (E1gene of interest ΔCP (cDNA of wt mice - cDNA of Fibulin-4+/R or Fibulin-4R/R) gene of interest)/ (Ehprt-1 ΔCP (cDNA wt mice- cDNA of Fibulin-4+/R or Fibulin-4R/R) hprt-1).
Mice were killed (10 days old by cervical dislocation or by perfusion with buffered formalin) and autopsied according to standard protocols (eg, Eumorphia protocol). Organ weights were determined and macroscopic abnormalities noted. Organs and tissues were fixed in formalin. Heart, aorta, and lungs of Fibulin-4+/+, Fibulin-4+/R and Fibulin-4R/R were isolated and fixed in 4% buffered formalin. After fixation, macroscopical images were taken using the stereoscope. Then aortas and heart valves were paraffin embedded and 4 μm sections were stained for elastin (Verhoeff-van Giesson, cartilage (alcian blue) and calcium-phosphate (von Kossa). BrdU staining was performed according the protocol of the manufacturer (Roche, Basel). Immunohistochemistry for pSmad2, CTGF and pSmad1/5/8 was performed using reagents and methods that were previously described.10
Echocardiography, Hemodynamic Measurements, and Data Analysis
Mice (15 to 20 weeks old) were weighed, sedated with 4% isoflurane and intubated as previously described.12 The mice were ventilated with a mixture of O2 and N2O (1/2, vol/vol) with a pressure controlled ventilator (CWE, SAR-830/P) to which 2.5% isoflurane was added for anesthesia. Ventilation rate was set at 90 strokes/min with a peak inspiration pressure of 18 cmH2O and a positive end expiration pressure (PEEP) of 4 cmH2O. The mice were placed on a heating pad to maintain body temperature at 37°C.
In vivo trans thoracic echocardiography of the left ventricle (LV) was performed with an ALOKA echo-machine (Pro Sound, SSD-4000; Japan) using a 13-MHz linear interfaced array transducer.12 M-mode echocardiograms were captured from short-axis 2D views of the LV at midpapillary level with simultaneous ECG (ECG). LV diameters at end diastole and end systole (LVEDD and LVESD) were measured from the M-mode images using SigmaScan Pro 5 Image Analysis software (SPSS Inc, Chicago, Ill). Three cardiac cycles for each animal were analyzed by a blinded observer. Following echocardiography, the mice were instrumented with a fluid filled polyethylene catheter (PE 10) inserted into the left carotid artery and advanced into the aortic arch for measuring mean aortic pressure. A stretched PE 50 catheter was introduced into the left external jugular vein and advanced into the superior caval vein for infusion of Hemaccel (Hoechst Marion Roussel B.V., Hoevelaken, The Netherlands), to maintain fluid-balance. A 1.4F Millar Instruments pressure transducer catheter (SPR-671, Millar Instruments, Houston, Texas, USA; calibrated before each experiment with a mercury manometer) was inserted in the right carotid artery, for measurement of systolic and diastolic aortic blood pressure, and advanced into the LV for measuring LV pressure. Hemodynamic data were recorded and digitized using an online 4-channel data acquisition program (ATCODAS, Dataq Instruments, Akron, Ohio, USA), for later analysis with a program written in MatLab. Ten consecutive beats were selected for determination of heart rate, systolic, diastolic and mean aortic pressure, LV systolic pressure and LV end diastolic pressure.
At the conclusion of each experiment, the heart was excised, the atria removed and the right ventricle (RV) and LV (including septum) were separated and weighed.
Aortic Pressure-Diameter Relation
Mice were anesthetized with a mixture of ketamine (25 mg/kg i.p.) and xylazine (5 mg/kg i.p.) and the thoracic aortas were excised. To determine aortic distensibility, the excised aorta was cannulated at both ends with polyethylene catheter (PE50), tested for leaks, and placed in a 37°C bath filled with MOPS-buffered Ringer’s solution. Each cannula was supplied by a reservoir of MOPS buffer. Each reservoir was separately pressurized by a computer-driven pressure interfase (Danish Myo Technology, Model 110P) and the aorta was visualized on an inverted microscope connected to a camera, allowing continuous real-time measurement of the dimensions of the outer thoracic aorta. Intra-aortic chamber pressure was increased from 20 to 220 mm Hg in 20 mm Hg steps, and the external dimensions of the thoracic aorta were determined (Myoview, Danish Myo Technology, Denmark).
Micro Array Hybridizations
Standard procedures were used to obtain total RNA (Qiagen) from the aorta of 4 WT, 2 Fibulin-4+/R and 2 Fibulin-4R/R 10-days-old mice. Synthesis of double stranded cDNA and biotin labeled cRNA was performed according to the instructions of the manufacturer (Affymetrix). Fragmented cRNA preparations were hybridized to full mouse genome oligonucleotide arrays (Affymetrix, mouse expression 430 V2.0 arrays), using a hybridization Oven 640 (Affymetrix), washed, and subsequently scanned on a GeneChip Scanner 3000 (Affymetrix). Initial data extraction and normalization within each array was performed by means of the GCOS software (Affymetrix). Data intensities were Log transformed and normalized within and between arrays by means of the quantile normalization method as previously described.13 Two-tailed pair wise analysis of variance was used by means of the Spotfire Decision Site software package 7.2 v10.0 (Spotfire Inc, Mass) to extract the statistically significant data from each of the 4 individual microarrays obtained for each genotype. The criteria for significance were set at P≤0.03 and a positive or negative 1.2-fold change.
Gene Ontology classification and network analysis of all significant gene entries were subjected to GO classification (http://www.geneontology.org). Significant over-representation of GO-classified biological processes was assessed by comparing the number of pertinent genes in a given biological process to the total number of the relevant genes printed on the array for that particular biological process (Fisher exact test, P≤0.05, False detection rate (FDR) ≤0.1) using the publicly accessible software Ease.14 Network analysis was performed through the use of Ingenuity Pathways Analysis, “a web-delivered application that enables biologists to discover, visualize and explore therapeutically relevant networks significant to their experimental results, such as gene expression array data sets (www.ingenuity.com)”. Microarray experiments complied with the regulations for Minimum Information of Microarray Experiments (MIAME) and can be retrieved from ArrayExpress (www.ebi.ac.uk/arrayexpress/, accession code: E-MEXP-840).
Pathology of Fibulin-4 Mutant Mice
We decreased the murine Fibulin-4 expression through transcriptional interference by targeted integration of a neomycin selectable marker (see online data supplement and supplemental Figure I of the online supplement available at http://circres.ahajournals.org.). A reduction in Fibulin-4 (Efemp-2) RNA expression levels was confirmed by real-time quantitative PCR on RNA isolated from cells, hearts, lungs and aortas of these mice and microarray expression analysis of aortas (see supplemental Figure I, II, and Table IV in the online data supplement available at http://circres.ahajournals.org). Heterozygous (Fibulin-4+/R) and homozygous (Fibulin-4R/R) Fibulin-4 mutant cells and tissues contain approximately 2- and 4- fold less Fibulin-4 RNA, respectively, compared with wild type (WT) cells and tissues. Decreased expression of Fibulin-4 did not impair embryonic viability because both Fibulin-4+/R and Fibulin-4R/R mutant mice were born at expected Mendelian frequencies. Furthermore, Fibulin-4+/R and Fibulin-4R/R mice did not show gross abnormalities and were morphologically indistinguishable from WT littermates in terms of development, growth and appearance. However, several Fibulin-4R/R mice died suddenly of cardiovascular complications before reaching weaning age.
Necroscopy was performed on 2 homozygous mutant animals that had died suddenly at 9 or 10 days of age. Both showed evidence of vascular compromise with hemopericardium causing cardiac tamponade. Serial sectioning of the aorta showed dissection of the thoracic aorta. In addition, the thoracic aorta was elongated and tortuous. Interestingly, there were no abnormalities in the lungs and skin. We compared the morphology of the aorta of newborn WT, Fibulin-4+/R and Fibulin-4R/R mice (Figure 1). All homozygous newborn Fibulin-4R/R mice showed dramatic dilatation of the ascending aorta resulting in an aorta with an at least 2-fold enlarged diameter (Figure 1C). Ten day-old old Fibulin-4R/R mice showed similar aortic dilatation and an increased heart size because of an enlarged left ventricle (not shown). Moreover, all Fibulin-4R/R mice analyzed had a tortuous aorta, similar to what has been reported for Fibulin-5 knockout mice.15,16 Whereas defective elastic fiber formation in the Fibulin-5 null mice results in a broad spectrum of moderate phenotypes, also including severe emphysema and loose skin similar to cutis laxa patients,17 impaired elastogenesis in our Fibulin-4R/R mice was specific to, and extremely severe in, the cardiovascular system. We also compared 14-week-old Fibulin-4+/R and Fibulin-4R/R mice. Although these mice show no differences in gross appearance compared with their WT littermates, aortas of Fibulin-4R/R mice showed severely enlarged and tortuous aortas and signs of intramural bleedings (Figure 1H). Detailed examination of heterozygous Fibulin+/R mice revealed subtle changes in the aorta of these mice. Some of the newborn Fibulin-4+/R mice showed abnormal ballooning of the proximal right subclavian artery, indicative of aneurysm (Figure 1B). In comparison with WT mice the aortas of Fibulin+/R mice were less translucent, corresponding with minor changes in the architecture of the vessel wall. Thus, even a modest reduction in Fibulin-4 expression results in aberrations of the aortic wall.
To obtain insight into the cause of the aortic abnormalities in Fibulin-4R/R mice, we performed histological examinations of aortas from these mice (Figure 2). There were significant changes in the architecture of the elastic laminae of the aorta, consisting of a granular appearance of elastin in the outer layers (adventitial side) of the aorta. In addition, the inner layers showed fragmentation and disarray of elastic fibers and the accumulation of amorphous matrix between fibers leading to aortic wall thickening. Whereas the integrity of elastic fibers in Fibulin-4+/R mice was preserved, excessive deposition of amorphous matrix was observed, indicating a gene dosage effect (Figure 2E). Elastin staining of longitudinal sections of the ascending aorta of Fibulin-4R/R mice showed regions with relatively well-organized elastic laminae, but also regions with dramatically affected laminae (Figure 2 J–O). Aortas of Fibulin-4R/R mice showed increased vascular smooth muscle cell proliferation, as evidenced by an increase in BrdU labeling (Figure 2G, 2H, and 2I). The major changes in cellular proliferation visualized by BrdU staining are seen in the tunica adventitia of the aorta. Like in Fibrillin-1 mice, this phenotypic alteration of vascular smooth muscle cells possibly precedes elastolysis.18 Disruption of the elastic laminae was already evident as early as postnatal day 1. This suggests that the defect seen in adult aortas of Fibulin-4R/R mice was not a result of late degradation of intact elastic laminae by activated inflammatory cells, but rather the consequence of an underlying developmental defect in the organization of the elastic fibers.
Because the Fibulin-4 deficiency evaluated is a serendipitous hypomorphic expression allele arising from targeted disruption of the adjacent Mus81 endonuclease gene, we also analyzed a possible contribution of the Mus81 endonuclease to aortic disease progression. To this end we histologically analyzed Mus81 mice from which the selectable marker was removed6 for aorta and heart valve abnormalities and found no indications for a role for Mus81 in the aortic diseases observed in our Fibulin-4 knockdown mice. Hematoxylin and elastin staining of cross sections of the Mus81 aortas of the Dendouga et al mice did not reveal any change in the vascular wall or make-up of the elastin layers (data not shown). Therefore we assign causality of the aortic phenotype to the Fibulin-4 hypomorfic allele.
Left Ventricular and Aortic Dimensions and Function
Body weight, heart rate and mean aortic pressure were not significantly different among Fibulin-4+/+, Fibulin-4+/R and surviving Fibulin-4R/R mice (Table). However, aortic pulse pressure (difference between systolic and diastolic pressure) was 2- to 3-fold higher in Fibulin-4R/R mice, which appeared to be the result of an increased aortic stiffness (Figure 3A), in conjunction with an increased stroke volume because of aortic insufficiency (Figure 3B). The latter resulted in an increase in left ventricular (LV) end-diastolic diameter and pressure. Fibulin-4R/R mice also demonstrated a small increase in systolic pressure-gradient across the aortic valve (LV systolic pressure minus aortic systolic pressure; Table), and transvalvular systolic blood flow velocity (Figure 3B), suggestive of a mild aortic valvular stenosis. Although a valvular stenosis could have resulted from thickening of the aortic valve leaflets and reactive cartilage and bone formation at the attachment site of the valve leaflets (Figure 4), it is important to note that the increases in pressure gradient and velocity are small and could be explained by the increase in stroke volume.
The combined LV pressure overload (stiffening of the aorta) and volume overload (aortic regurgitation) resulted in significant LV hypertrophy, reflected by a 50% increase in relative LV mass.
Transcriptome Analysis of Fibulin-4+/R and FibulinR/R Mice
Having established the relevance of the Fibulin-4+/R and Fibulin-4R/R mouse model to elastin-deficiency related diseases, we next sought to compare the transcriptomes of Fibulin-4+/R and Fibulin-4+/+ (n=6) aortas, thereby obtaining an unbiased, holistic insight into those molecular pathways most relevant to the early development of aneurysm formation. In addition, the comparison of the Fibulin-4R/R to that of Fibulin-4+/+ (n=6) aorta transcriptomes would also allow the identification of genes with a central role in the response to aortic failure. A 2-tailed, pair-wise analysis of variance of Affymetrix full mouse genome arrays revealed 553 probe sets with significant expression changes between the Fibulin-4+/R and Fibulin-4+/+ (P≤0.03, 1.2 fold change up- or downregulated, supplemental Tables I and III). To avoid any initial gene preselection and thus introduction of bias, we implemented a rigorous methodology14 to identify, among this set of genes, those biological processes -classified by the Gene Ontology classification- with a statistically significant disproportionate number of responsive genes relative to those printed on microarrays (False detection rate ≤0.10, Figure 5A). This unbiased approach revealed genes associated with cell adhesion (eg, Vcam1, Clecsf10, Stab2, Itgal, Cd72, Fcer2a, Zyx, Lamb2, Adrm1, Bcar1 and Epdm2), cytoskeleton organization and muscle development (eg, Myh1, Acta1, Mylpf, Myh2, Myl1, Myo1f, Mylk, Actc1, Myh7, Myh6, Myl7, Tnni2, Tpm2, Pva, Tnnt3, Tnnt2, Smpx, Tncc, Tnni3) and immune responses (eg, Ccl8, Cxcl9, H2-Q7, Cxcl10, Ly86, H2-T23, amhd1, Ifi205, Ifi205, H2-Aa, H2-Ab1, H2-Dma, Igh-6, H2-D1, H2-Aa, Cxcl13, H2-Ea, H2-DMb2, Cd8b, Cd8a, H2-Oa, Cd8a) to constitute the most significantly over-represented processes in Fibulin-4+/R mouse aortas (Figure 5A and supplemental Table I). The mild, yet detectable abnormalities in the elastic fiber network of Fibulin-4+/R aortas suggests that these identified processes constitute an early, adaptive response to the improper assembly of elastic fibers.
Next, to analyze the late events in the response to aortic failure, we then compared the transcriptomes of Fibulin-4R/R and Fibulin-4+/+ aortas under the same selection criteria. This approach revealed 653 probe sets that were significantly associated with apoptosis and cell death (eg, Tradd, Sulf1, Bok, Sulf1, Bnip2, Ripk2, Siva, Malt1, Nfkb1, Bcl2, Zfp346, Clu), intracellular signaling (eg, Tyk2, Stat1, Tec, Rasa3, Stam, Sec23b, Rgl1, Pdlim1, Prkcb, Sh2d3c, Gga1, Pld2, Stx6) and cell growth and metabolism (Rela, Nras, Cbl, Erbb2, Igfbp4, Rras, Extl3) (Figure 5A and supplemental Tables II and IV) likely reflecting the severity and end stage pathology of elastic fiber network resulting in the dissection of Fibulin-4R/R mouse aortas. Quantitative real-time PCR of several genes confirmed the validity of microarray data (online data supplement and supplemental Figure II). To identify those pathways most likely to be perturbed by the Fibulin-4 deficiency, we subsequently performed genome-scaled network analysis (Ingenuity) including all genes with significant expression changes. This approach revealed those pathways perturbed by Fibulin-4 deficiency and pointed to a central role for the cytokine TGF-β (Figure 5B). Apart from TGF-β as the central cytokine upregulated, also other signaling pathways such as BMP signaling were affected since we found chondroid metaplasia in the aorta valves and upregulation of Smad5 (a downstream effector of the BMP signaling pathway) in Fibulin-4 mice. Analysis of aortic tissue from Fibulin-4+/R and Fibulin-4R/R mice revealed a graded increase in TGFβ signaling in comparison to WT animals, as evidenced by increased phosphorylation and nuclear translocation of Smad2, increased expression of connective tissue growth factor CTGF, and increased collagen deposition (Figure 6). The prominent perturbation of normal elastin lamellar structure may arise from induction of TGF-β regulated MMP9, because we found MMP9 upregulated in both Fibulin-4+/R mice (1.3-fold probability value 0.0134) and FibulinR/R mice (2.2-fold, probability value 0.3388). Evaluation of phospho- Smad1/5/8 accumulation by immunohistochemistry showed additional evidence for dysregulation of BMP signaling in Fibulin-4 mice (Figure 6D).
In summary, these Fibulin-4 mice proved to be a relevant model to follow the pathogenic sequence of aneurysm at the physiological and functional genomics level. Our results indicate that the expression level of a single gene can determine the extent of maturation of elastic fibers in vivo in the aorta and aortic valves. The Fibulin-4 deficiency-induced loss of elastic fiber structure spontaneously results in severe aneurysm formation and aortic stiffening, followed by aortic dissection. Furthermore, we found thickened aortic valvular leaflets that are associated with aortic valve stenosis and insufficiency. Interestingly, Hucthagower et al have recently reported a missense mutation resulting in reduced expression of fibulin-4 in a patient with cutis laxa syndrome.19 This patient was reported to have vascular tortuosity and ascending aortic aneurysm. The highly organ specific and clear abnormalities seen in all the Fibulin-4R/R mice possibly resemble unidentified aorta associated syndromes and Fibulin-4R/R mice provide a new entry for the dissection of the pathways involved in elastic laminae formation. Our data support the notion that a deficiency of elastic fibers is a marker of a more critical event in the pathogenesis of aneurysm. Excessive TGF-β signaling emerges as a strong candidate. Interestingly, increased TGF-β signaling has previously been documented in the context of other well-established aortic aneurysm syndromes, including MFS, LDS and ATS.8,20 The mechanism by which Fibulin-4 deficiency initiates excessive TGF-β signaling is unknown. In theory, this could reflect disruption of other elastic fiber components known to regulate TGF-β such as Fibrillin-1 or Emilin.21,22 Alternatively, if analogies can be drawn to Fibulin-5, a molecule required for elastogenesis that bridges elastic fibers and cells through interaction with both elastin and integrins, it is possible that Fibulin-4 has a direct influence on integrin-mediated intracellular signaling events that modulate TGF-β activity. Here, it is informative to compare the Fibulin-4 and Fibulin-5 deficient phenotypes. Mice deficient for either Fibulin show arterial tortuosity, a finding also observed with other causes of a severe and early defect in elastogenesis in both mouse models and patients including LDS, ATS and primary elastin deficiency. However, Fibulin-5 deficient mice and people do not develop aneurysms. These data challenge a common perception that a primary or acquired structural deficiency of the elastic matrix is necessary and sufficient to initiate aneurysm formation and progression.
If validated, this model suggests that therapeutic strategies aimed to target TGF-β signaling for the treatment of MFS and other disorders, including the FDA-approved drug losartan,23 will find relevance for the treatment of other aneurysm phenotypes. In addition, the milder phenotype of heterozygous Fibulin-4+/R mice could be invaluable in providing an informative aneurysm model system to delineate the early events in the pathogenetic sequence that culminate in aneurysm, since the subtle manifestations of aberrant elastin formation in the Fibulin-4+/R mice, but clear adaptive response seen in the microarray expression data and TGF-β signaling might in fact resemble those patients experiencing“sporadic” and barely detectable forms of aneurysms. Therefore these Fibulin-4+/R mice provide the opportunity to unravel the biological processes within early onset of aortic degeneration and identify serum markers for early diagnosis of aortic failure.
We would like to thank Theo Gorgels and Wim Vermeulen for helpful discussion, Bert van der Horst for providing valuable reagents, D. Moechars and J. Vialard for Mus81 control mice, and Ingrid van der Pluijm and Wim Vletter for technical assistance.
Sources of Funding
H.C. Dietz is funded by the Howard Hughes Medical Institute, the National Institutes of Health (USA, AR-41135 and AR-049698), the William S. Smilow Center for Marfan Syndrome Research and the National Marfan Foundation (USA).
This manuscript was sent to Donald D. Heistad, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
Current affiliation for K.H., Cancer Research UK Laboratories, Weatherall Institute of Molecular Medicine, University of Oxford, John Radcliffe Hospital, Oxford, UK.
Original received June 8, 2006; resubmission received November 6, 2006; revised resubmission received January 19, 2007 accepted January 25, 2007.
Kielty CM, Sherratt MJ, Shuttleworth CA. Elastic fibres. J Cell Sci. 2002; 115: 2817–2828.
McLaughlin PJ, Chen Q, Horiguchi M, Starcher BC, Stanton JB, Broekelmann TJ, Marmorstein AD, McKay B, Mecham R, Nakamura T, Marmorstein LY. Targeted disruption of fibulin-4 abolishes elastogenesis and causes perinatal lethality in mice. Mol Cell Biol. 2006; 26: 1700–1709.
Dendouga N, Gao H, Moechars D, Janicot M, Vialard J, McGowan CH. Disruption of murine Mus81 increases genomic instability and DNA damage sensitivity but does not promote tumorigenesis. Mol Cell Biol. 2005; 25: 7569–7579.
McPherson JP, Lemmers B, Chahwan R, Pamidi A, Migon E, Matysiak-Zablocki E, Moynahan ME, Essers J, Hanada K, Poonepalli A, Sanchez-Sweatman O, Khokha R, Kanaar R, Jasin M, Hande MP, Hakem R. Involvement of mammalian Mus81 in genome integrity and tumor suppression. Science. 2004; 304: 1822–1826.
Mizuguchi T, Collod-Beroud G, Akiyama T, Abifadel M, Harada N, Morisaki T, Allard D, Varret M, Claustres M, Morisaki H, Ihara M, Kinoshita A, Yoshiura K, Junien C, Kajii T, Jondeau G, Ohta T, Kishino T, Furukawa Y, Nakamura Y, Niikawa N, Boileau C, Matsumoto N. Heterozygous TGFBR2 mutations in Marfan syndrome. Nat Genet. 2004; 36: 855–860.
Loeys BL, Chen J, Neptune ER, Judge DP, Podowski M, Holm T, Meyers J, Leitch CC, Katsanis N, Sharifi N, Xu FL, Myers LA, Spevak PJ, Cameron DE, De Backer J, Hellemans J, Chen Y, Davis EC, Webb CL, Kress W, Coucke P, Rifkin DB, De Paepe AM, Dietz HC. A syndrome of altered cardiovascular, craniofacial, neurocognitive and skeletal development caused by mutations in TGFBR1 or TGFBR2. Nat Genet. 2005; 37: 275–281.
van den Bos EJ, Mees BM, de Waard MC, de Crom R, Duncker DJ. A novel model of cryoinjury-induced myocardial infarction in the mouse: a comparison with coronary artery ligation. Am J Physiol Heart Circ Physiol. 2005; 289: H1291–H1300.
Peeters PJ, Fierens FL, van den Wyngaert I, Goehlmann HW, Swagemakers SM, Kass SU, Langlois X, Pullan S, Stenzel-Poore MP, Steckler T. Gene expression profiles highlight adaptive brain mechanisms in corticotropin releasing factor overexpressing mice. Brain Res Mol Brain Res. 2004; 129: 135–150.
Loeys B, Van Maldergem L, Mortier G, Coucke P, Gerniers S, Naeyaert JM, De Paepe A. Homozygosity for a missense mutation in fibulin-5 (FBLN5) results in a severe form of cutis laxa. Hum Mol Genet. 2002; 11: 2113–2118.
Bunton TE, Biery NJ, Myers L, Gayraud B, Ramirez F, Dietz HC. Phenotypic alteration of vascular smooth muscle cells precedes elastolysis in a mouse model of Marfan syndrome. Circ Res. 2001; 88: 37–43.
Coucke PJ, Willaert A, Wessels MW, Callewaert B, Zoppi N, De Backer J, Fox JE, Mancini GM, Kambouris M, Gardella R, Facchetti F, Willems PJ, Forsyth R, Dietz HC, Barlati S, Colombi M, Loeys B, De Paepe A. Mutations in the facilitative glucose transporter GLUT10 alter angiogenesis and cause arterial tortuosity syndrome. Nat Genet. 2006; 38: 452–457.
Zacchigna L, Vecchione C, Notte A, Cordenonsi M, Dupont S, Maretto S, Cifelli G, Ferrari A, Maffei A, Fabbro C, Braghetta P, Marino G, Selvetella G, Aretini A, Colonnese C, Bettarini U, Russo G, Soligo S, Adorno M, Bonaldo P, Volpin D, Piccolo S, Lembo G, Bressan GM. Emilin1 links TGF-beta maturation to blood pressure homeostasis. Cell. 2006; 124: 929–942.
Habashi JP, Judge DP, Holm TM, Cohn RD, Loeys BL, Cooper TK, Myers L, Klein EC, Liu G, Calvi C, Podowski M, Neptune ER, Halushka MK, Bedja D, Gabrielson K, Rifkin DB, Carta L, Ramirez F, Huso DL, Dietz HC. Losartan, an AT1 antagonist, prevents aortic aneurysm in a mouse model of Marfan syndrome. Science. 2006; 312: 117–121.