EXPERIMENTALLY INDUCED DISEASE - University of Edinburgh



EXPERIMENTALLY INDUCED DISEASEShort Title: Canine Mitral Valve Tissue EngineeringDevelopment and Evaluation of a Tissue-Engineered Fibrin-based Canine Mitral ValveThree-dimensional Cell Culture SystemM.-M. Liu*, T. C. Flanagan?, S. Jockenhoevel?, A. Black§, C.-C. Lu*, A. T. French*,D. J. Argyle* and B. M. Corcoran**Royal (Dick) School of Veterinary Studies, The Roslin Institute, the University of Edinburgh, Easter Bush, Roslin, Mid-Lothian, Scotland, UK, ?School of Medicine, Health Sciences Centre, University College Dublin, Belfield, Dublin, Ireland, ?Department of Tissue Engineering and Textile Implants, AME – Helmholtz Institute for Biomedical Engineering, RWTH Aachen University, Aachen, Germany and §Department of Anatomy, National University of Ireland Galway, Galway, Ireland.Correspondence to: B. Corcoran (e-mail: Brendan.Corcoran@ed.ac.uk). SummaryMyxomatous mitral valve disease is the most common cardiac disease of the dog, but examination of the associated cellular and molecular events has relied on the use of cadaveric valve tissue, in which functional studies cannot be undertaken. The aim of this study was to develop a three-dimensional (3D) cell co-culture model as an experimental platform to examine disease pathogenesis. Mitral valve interstitial (VIC) and endothelial (VEC) cells were cultured from normal and diseased canine (VIC only) valves. VICs were embedded in a fibrin-based hydrogel matrix and one surface was lined with VECs. The 3D static cultures (constructs) were examined qualitatively and semiquantitatively by light microscopy, immunofluorescence microscopy and protein immunoblotting. Some constructs were manipulated and the endothelium damaged, and the response examined. The construct gross morphology and histology demonstrated native tissue-like features and comparable expression patterns of cellular (α-smooth muscle actin [SMA] and embryonic smooth muscle myosin heavy chain [SMemb]) and extracellular matrix associated markers (matrix metalloproteinase [MMP]-1 and MMP-3), reminiscent of diseased valves. There were no differences between constructs containing normal valve VICs and VECs (type 1) and those containing diseased valve VICs and normal valve VECs (type 2). Mechanical manipulation and endothelial damage (type 3) tended to decrease α-SMA and SMemb expression, suggesting reversal of VIC activation, but with retention of SMemb+ cells adjacent to the wounded endothelium consistent with response to injury. Fibrin-based 3D mitral valve constructs can be produced using primary cell cultures derived from canine mitral valves, and show a phenotype reminiscent of diseased valves. The constructs demonstrate a response to endothelial damage indicating their utility as experimental platforms.Keywords: tissue engineering; valve interstitial cell; valve endothelial cell; myxomatous mitral valve disease; dogIntroduction Myxomatous mitral valve disease (MMVD) is the most common cardiac disease in dogs (Whitney, 1974; Buchanan, 1977; Beardow and Buchanan, 1993; Connell et al., 2012). Most studies, to date, have examined the cellular and molecular features of the disease using native valve tissue collected at the time of necropsy examination. However, tissue availability can be variable and unpredictable, and there can be heterogeneity in samples obtained. There is, therefore, a need for cell culture models to provide consistent and reproducible experimental platforms for studies that are not readily achievable using valve tissue alone.In MMVD, affected leaflets become thickened, distorted and less transparent. Valve endothelium denuding is common and there is evidence of endothelial-to-mesenchymal transition (Han et al., 2010, 2013; Lu et al., 2015). Extracellular matrix (ECM) changes can be severe, but are located mainly in the distal third of the leaflet up to the free edge, and include collagen loss, production of fibrillar collagen that fails to mature, and increased proteoglycan and glycosaminoglycan synthesis (Cole et al., 1984; Tamura et al., 1995; Aupperle et al., 2009a; Gupta et al., 2009; Hadian et al., 2010; Han et al., 2010). The pathogenesis of MMVD is only partially understood, but may in part be a result of valve leaflet maladaptive responses to long-term shear stress damage of the valve endothelium, triggering subendothelial valve interstitial cell (VIC) activation and ECM remodelling (Corcoran et al., 2004; Gotlieb et al., 2002; Pedersen and Haggstrom, 2000). VICs, the major source of valve matrix, change from a quiescent interstitial phenotype to an activated myofibroblast, and then congregate in the subendothelium (Stein et al., 1989; Mow and Pedersen, 1999; Corcoran et al., 2004; Barth et al., 2005; Black et al., 2005; Han et al., 2008). Activated VICs are characterized by increased expression of alpha smooth muscle actin (α-SMA) and/or the embryonic form of non-muscle myosin heavy chain (SMemb) and have heightened wound repair and ECM remodelling capability (Tamura et al., 2000; Rabkin et al., 2001; Gotlieb et al., 2002; Black et al., 2005; Disatian et al., 2008; Han et al., 2008). Similar effects have been seen with endothelial damage in an ovine mitral valve organ culture model, suggesting that endothelial changes likely contribute to MMVD pathogenesis (Lester et al., 1992, 1993). The capability to investigate complex interactions between VICs and valvular endothelial cells (VECs) and the effect of VECs on VIC function, and by extension valve remodelling, is limited using native valve tissue. Tissue engineering (TE) has been used widely to model cardiovascular diseases and to examine cell behaviour, function and pathophysiological responses, and should be applicable to MMVD (Butcher et al., 2004; Butcher and Nerem, 2006; Gupta et al., 2008a, b). In that context, three-dimensional (3D) cultures and TE models are superior to traditional two-dimensional (2D) cultures as they more resemble the in-vivo arrangement of cell distribution and orientation, cell signalling and communication, and ECM production and organisation (Mueller-Klieser, 1997). The aims of the present study, therefore, were to use primary canine mitral valve cells, derived from both normal and diseased dogs, to develop a fibrin-based VEC/VIC 3D co-culture system (construct), and to partially assess its utility as a platform to investigate the pathogenesis of MMVD, by mechanical manipulation and endothelial damage. Materials and MethodsCell Isolation, Culture and Characterization Canine mitral VECs and VICs were isolated and cultured as described previously (Liu et al., 2015). VECs were derived from healthy mitral valves only, while VICs were derived from both healthy and diseased valves (Whitney grade 1 and 2; mild to moderate disease). Dogs ranged in age from 1.5–5 years, with one elderly (i.e. >7 years) dog, from which VICs only were harvested. VECs were harvested at passage 2 or 3 and VICs at passage 4 to 7. All harvesting of valve tissues and cells was undertaken with full owner permission and with ethical approval from the Veterinary Ethics Research Committee of the Royal (Dick) School of Veterinary Studies, the University of Edinburgh. Constructs were then produced using cells from different dogs. Cell phenotypes were confirmed by immunofluorescence microscopy examining for the expression of the general mesenchymal marker, vimentin and the VEC-specific marker, CD31 (platelet and endothelial cell adhesion molecule [PECAM]-1). Each construct was divided into quarters for immunofluorescene microscopy and western blotting (WB), with the remainder archived for future studies.Native Mitral Valve TissueMitral valves were examined to demonstrate patterns of marker distribution for healthy and diseased valves. Sections were obtained from archived valves as previously reported (Lu et al., 2016). Fabrication of Fibrin-based 3D Mitral Valve Constructs Constructs were developed based on previously described techniques with minor modification (Flanagan et al., 2006; Cholewinski et al., 2009). Each construct was fabricated in one well of a sterile 24-well culture plate by mixing 175?l VIC suspension in pH 7.4 Tris buffered saline (TBS) (1 × 106 cells/gel), 37.5?l 40 IU/ml thrombin (Sigma, Poole, UK) 37.5?l 50mM CaCl2 (BDH, UK) and 250?l of 10mg/ml bovine plasma fibrinogen (Sigma). The hydrogel was left to polymerize at 37°C in 5% CO2 in an incubator for 40–60min. VECs (1.8 × 104/ml) were re-suspended in a customized bioreactor culture medium consisting of advanced DMEM/F-12 medium (Life Technologies), 1 mM L-ascorbic acid 2-phosphate (Sigma), 10% fetal bovine serum (Life Technologies), 1% penicillin G and streptomycin (Invitrogen), 1% L-glutamine (Gibco) and 0.0001% tranexamic acid (Sigma). Once the fibrin/VIC hydrogel had completely polymerized, 1 ml of VEC suspension was seeded onto the top of each construct. The constructs were then maintained under standard tissue culture conditions, with daily media changes for 14 days. Two types of constructs were developed to mimic healthy (type 1; normal VIC and VEC), diseased (type 2; diseased VIC and normal VEC) valves, and their characteristics were compared. Additionally, type 1 constructs underwent endothelial damage (type 3). Construct Manipulation and Endothelial DamageType 1 healthy VEC/VIC constructs (n = 12) were fabricated and maintained under static culture conditions for 14 days in one 24-well culture plate. On day 14, culture medium was removed, the constructs were rinsed with sterile phosphate buffered saline (PBS) and divided into intact sham control and endothelial-damaged groups (type 3). The endothelial surface of type 3 constructs was damaged by drawing multiple linear scratches in a hatched fashion (five lines each direction at right angles), using the tip of a pair of surgical forceps, across the construct surface. All constructs were manually detached from the culture plate and rinsed in sterile PBS to remove any detached cells before adding fresh culture medium. Two pairs of constructs (sham and wounded) were collected at days 0 (within 5 min), 2 and 6 for examination. Light Microscopy Constructs were fixed in 4% paraformaldehyde, processed routinely and embedded in paraffin wax. Sections (5 m) were stained with haematoxylin and eosin (HE) and for general connective tissue detection, a modified Russell–Movat pentachrome method was used (KTRMP, American MasterTech, USA). All sections were examined under bright field illumination using a Leica DMRB microscope (Leica, Germany).Immunofluorescence MicroscopyConstructs and samples of healthy canine mitral valves were processed and sectioned as before and examined for the presence of activated VICs and ECM products using immunofluorescence microscopy. For the mitral valve sections, antigen retrieval included incubating with pH 9.0 Tris-EDTA buffer or with pH 7.8 0.05% trypsin/0.1% CaCl2, depending on antigen type. All sections were incubated with Image-iT TM FX Signal Enhancer (Invitrogen) for 30 min then blocked with 10% goat serum in 0.5% PBS-Tween for 30 min at room temperature. Sections were probed with primary antibodies examining for cell phenotype and proliferation, and ECM synthetic activity in a humid chamber overnight at 4°C (Table 1). To control for non-specific binding the primary antibody was omitted and the same volume of diluted buffer was applied. Sections were incubated with fluorescein-conjugated secondary antibodies (Table 1) in a dark humidified chamber at room temperature for 1 h, before counterstaining with 4, 6-diamidino-2-phenylindole (DAPI) and observing using a Leica fluorescence microscope (Leica). Endothelium Labelling Acetylated low density lipoprotein conjugated to the fluorescent dye 1, 1’-dioctadecyl-3, 3, 3’, 3’-tetramethyl-indocarbocauanine perchlorate (DiI-Ac-LDL) was used to confirm endothelium formation on day 14 constructs. Fresh constructs were transferred to a chamber slide and rinsed once with sterile PBS. DiI-Ac-LDL reagent (Biomedical Technologies, UK) was diluted to 5g/ml with culture medium and 0.5ml was applied to each construct for 4 h at 37°C in 5% CO2 in a dark room. Constructs were rinsed once with PBS and transferred to an inverted glass chamber slide with the endothelial surface facing down, and viewed using a LSM710 confocal microscope (Carl Zeiss, Germany).Image Capture and ProcessingFor light microscopy and immunofluorescence studies, images were captured and processed using Leica Firecam imaging and ImageJ (National Institute Health, Bethesda, Maryland, USA). For immunofluorescence, the original antibody and DAPI channel colour images were merged using Adobe Photoshop CS6 software (Adobe System, USA). For DiI-Ac-LDL labelling, images were processed with ZEN software (Carl Zeiss).Western Blotting Western blotting was used to semiquantify a set of disease-related markers involved in cell activation and matrix metalloproteinase (MMP) synthesis. Constructs from the same batch were pooled for protein extraction (50 mg total). Samples (5–30g, depending on target antigens) were separated by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) and blotted onto Protran? Nitrocellulose Transfer Membrane (Whatman GmbH, Germany). The membranes were blocked in 5% milk/PBS-0.1%Tween 20 for 1h at room temperature, and incubated with primary antibodies (Table 1) overnight at 4°C or for 1h at room temperature, followed by incubation with horseradish peroxidase-conjugated secondary antibodies at room temperature for 1h. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) antibody was used as an internal control. Commercial pre-stained protein markers (Bio-Rad and GE Healthcare, UK) were used to gauge the size of the protein bands. Membranes were ‘developed’ with AmershamTM ECLTM Western Blot Detection Reagent (GE Healthcare). Protein bands were visualised by developing radiographic films (GE Healthcare) in a SRX-101A radiographic processor (Konica Minolta, UK), scanned and the protein band intensity was quantified using ImageJ. For the comparison of type 1 and type 2 constructs, data were initially semiquantified by normalizing band intensity of interest to the internal control, then normalized further by calculating the mean type 2/type 1 ratio for each marker. Type 3 constructs were further normalized to the day 0 control.Statistical Analysis Statistical analysis was carried out on type 1 and type 2 western blot semiquantitative data only. Data were checked for normal distribution using the Anderson–Darling normality test prior to further statistical analysis and the data were found to be normally distributed. One way analysis of variance (ANOVA) was performed with Minitab software (Minitab) to determine the difference between the two groups of data with significance set at P <0.05. ResultsMacroscopical and Microscopical Features of Fibrin Based ConstructsAfter 14 days, the constructs exhibited some features reminiscent of native tissue morphology (Fig. 1). All constructs shrunk in size and were less transparent than controls. HE-stained cells were found to be evenly distributed throughout the construct. There was a cell/matrix layer (endothelial layer) overlying the construct surface in contact with the culture medium. There was ECM synthesis with collagen fibres detected, predominantly in the construct stroma, and with mucin expression localized mainly to the loose connective tissue zone adjacent to the endothelial layer. Endothelialization was confirmed by positive labelling for DiI-Ac-LDL, confined to the surface in contact with the culture medium. Cell and ECM marker expression patterns in the constructs were, in part, comparable with those seen in the native valve (Fig. 2). Cells positive for vimentin and SMemb were distributed through the entire construct and native valve, but not at the same density. α-SMA expression was limited, localized mainly to the atrialis in native valves, and the equivalent area in the constructs (i.e. beneath the endothelial layer), and to a lesser extent in the construct stroma. Collagen I and III labelling was mainly located in the native valve fibrosa layer and primarily located adjacent to stromal cells in the constructs, and adjacent to the endothelial layer in some. There was expression of MMP-1 in the construct- and native valve stroma, but minimal MMP-3 expression in both. Biglycan labelling showed similar distribution throughout the native valve and construct stroma with some cellular expression. Decorin was present in native valves, with similar pattern to biglycan, but detection was weak in the construct. The proliferation marker Ki67 was expressed by most cells in both native valve and parison of MMVD Markers in Type 1 and Type 2 ConstructsQualitatively and semiquantitatively, using immunofluorescence microscopy and western blotting, there were no differences found when comparing type 1 and 2 constructs for MMP-1, MMP-3, SMemb, α-SMA and vimentin expression (Fig. 3). CD31, as assessed by western blotting, was detected in all construct samples, but varied between different batches of constructs.Construct Response to Manipulation (Tension Release) and Endothelial Injury The cellular response to a combination of manual manipulation by freeing the construct from the bottom of the plate and so releasing tension, both under test and sham conditions, and endothelial wounding, tended to decrease cell activity over a 6-day period. Constructs contracted after detachment more in the first 48 h than non-detached constructs and morphology remained stable thereafter.Vimentin-positive cells were distributed throughout the entire construct thickness from days 0 to 6 in both type 3 and sham controls, and the labelling pattern did not differ appreciably between each of these two time points; however, there was some loss of vimentin expression in some of the older control constructs (Fig. 4). Type 3 constructs exhibited similar α-SMA expression to the sham controls at days 0, 2 and 6. In both groups, there was decreased α-SMA+ cells by days 2 and 6 compared with day 0; mainly localized to the construct stroma (Fig. 4). Stromal expression of SMemb+ cells remained static at the three time points in both sham and type 3 constructs. In the sham constructs at days 2 and 6, there was reduced expression of SMemb in the endothelial layer compared with day 0. For the majority of type 3 constructs (6/8), SMemb+ cells were still visible on or close to the wounded surface at both days 2 and 6 (Fig. 4). Pooled construct protein semiquantitative results showed a trend for reduced expression of α-SMA, SMemb and vimentin on days 2 and 6 compared with day 0, but there were no apparent differences between type 3 and the sham controls. Discussion This study reports the development of a fibrin-based 3D static VIC/VEC canine mitral co-culture (construct) using healthy and diseased canine mitral valve cells. The constructs were found to exhibit several native canine valve features, including tissue-like morphology, expression of native valve cell phenotypic markers and ECM synthesis and degradation capability. The constructs retained their endothelial covering, but these formed into a layer several cells thick, not seen in native valves. These results are similar to that seen with porcine mitral VECs and VICs grown on a collagen–glycosaminoglycan hydrogel scaffold, which synthesized a native valve-like ECM profile (Flanagan et al., 2006). Mitral valve VECs and VICs, in addition to demonstrating synthetic capability, have good cell replication capacity in vitro, and cells derived from a single medium-size dog mitral valve could potentially generate hundreds of static 3D constructs (Liu et al., 2015). When diseased VICs were used in the constructs, unexpectedly, differences in construct cell phenotype or structure were not identified. VICs in diseased valves were altered to an activated myofibroblast phenotype, as shown by increased expression of α-SMA and SMemb, and expressed excess ECM catabolic enzymes, including the collagenase MMP-1 and stromelysin MMP-3 (Rabkin et al., 2001; Disatian et al., 2008; Han et al., 2008; Aupperle et al., 2009b). Overall, and in contrast to what was seen in healthy mitral valve tissue, VICs predominately had an activated phenotype expressing a combination of vimentin, α-SMA and/or SMemb. ECM remodelling, as evidenced by expression of MMP-1 and MMP-3, was also present in both construct types, suggesting that the constructs were more closely modelling MMVD than the normal valve. VICs derived from mildly diseased canine mitral valves have been shown previously to express both vimentin and α-SMA on 3D collagen hydrogels (Waxman et al., 2012). However, human aortic VICs also express α-SMA when cultured in medium containing 10% serum, but do not in low (2%) serum media, and we would recommend this latter approach is taken in future construct manufacture (Latif et al., 2015). Manipulation of the constructs, by detaching from the culture plate surface and so reducing tension forces, decreased cell activity, reduced α-SMA expression and increased construct shrinkage. α-SMA expression is upregulated by exogenous mechanical forces or high culture substrate stiffness in aortic VICs (Stephens et al., 2011; Kural and Billiar, 2016). It is likely that the decreased tension on the constructs moderated the activated myofibroblast phenotype to a more quiescent and/or synthetic phenotype. The same findings have been observed in a porcine aortic VIC 3D construct model in response to changes in mechanical tension (Kural and Billiar, 2016). Furthermore, applying cyclic strain to a collagen hydrogel model has been shown to modify activated VICs from canine myxomatous diseased mitral valves to a more quiescent phenotype (Waxman et al., 2012). Reversible phenotypic plasticity of VICs in response to disease and physiological remodelling has been suggested, with cells returning to a quiescent phenotype once a steady-state equilibrium has been achieved, and the data from the canine constructs and other studies would tend to support this hypothesis (Rabkin-Aikawa et al., 2004; Rabkin-Aikawa et al., 2005). Endothelial damage resulted in retention of SMemb+ cells close to and within the endothelial layer compared with the sham controls, but had no differential effects on construct morphology, ECM or synthetic activity, and in both test and sham constructs α-SMA and SMemb expression declined. This endothelial localization is somewhat reminiscent of that seen with diseased valves where activated VICs accumulate close to the endothelial surface, with a mixture of α-SMA+, Smemb+ and α-SMA/Smemb+ cells being found (Disatian et al., 2008; Han et al., 2008; Lu et al., 2016). Mitral VICs have been shown to contribute to valve surface wound repair in cultured tissue explants, with cell proliferation and migration, increased cytokine expression and ECM synthesis (Lester et al., 1992, 1993; Gotlieb et al., 2002). In contrast to α-SMA, SMemb is considered in diseased mitral valves as a marker of mesenchymal cell activation, where there is increased synthetic activity (Rabkin-Aikawa et al., 2004; Hayek et al., 2005; Disatian et al., 2008; Lu et al., 2016). The fibrin hydrogel used in this study is favourable for cell distribution, cell communication and the synthesis and accumulation of ECM within tissue-engineered constructs, but is also affected by cell-mediated contraction and shrinkage (Ye et al., 2000; Jockenhoevel et al., 2001; Flanagan et al., 2007). VICs do have contractile and force generation properties and are also sensitive and responsive to the mechanical properties of the surrounding matrix (Stephens et al., 2010, 2011). Human aortic VICs activate spontaneously when grown on mechanically soft methacrylated gelatine hydrogels, while activated myofibroblasts have heightened contractility and force generation capability, compared with fibroblasts, when grown on collagen gels (Stephens et al., 2010). However, fibrin is not a valve ECM component and needs to be removed by VICs as they generate new ECM. MMP-3 can cleave fibrin cross links in vitro, is also important in valve ECM remodelling, and was expressed in the canine constructs, but at minimal levels (Bini et al., 1996). An interesting feature of static 3D cultures is development of a cell-dense layer on the side closest to the culture medium, as was the case in this study (Carrier et al., 1999). Dynamic conditioning (flow and pressure) of a tissue-engineered valve construct can overcome this problem, improving cell proliferation and viability and enhance ECM production and organization (Flanagan et al., 2007). In conclusion, fibrin-based 3D co-culture constructs were successfully generated using canine primary mitral valve cells. Under static culture conditions, constructs assembled into a form more reminiscent of MMVD than a healthy valve, regardless of using VICs derived from healthy or diseased valves. External mechanical manipulation modulated VIC behaviour in this model and endothelial damage affected the adjacent cell phenotype. This construct is relatively easy to manufacture and can be used as a partial experimental model for canine MMVD. AcknowledgmentsWe thank R. Muirhead of the Roslin Institute for help and training in cell culture systems. The current address for A.T. French is University of Glasgow, School of Veterinary Medicine, Bearsden Road, Glasgow, Scotland, UK. The findings of this study were presented at the 7th Society of Heart Valve Disease Biennial Meeting, Venice, Italy, September 2013. M.-M. Liu was funded by a China State PhD Scholarship.Conflict of Interest StatementThe authors declare no conflicts of interest with respect to the publication of this manuscript.ReferencesAupperle H, Marz I, Thielebein J, Kiefer B, Dinges G et al. (2009a) Distribution of extracellular matrix components in normal and degenerated canine tricuspid valve leaflets. Journal of Comparative Pathology, 141, 41-51.Aupperle H, Marz I, Thielebein J, Kiefer B, Kappe A et al. (2009b) Immunohistochemical characterization of the extracellular matrix in normal mitral valves and in chronic valve disease (endocardiosis) in dogs. Research in Veterinary Science, 87, 277-283.Barth PJ, Koster H, Moosdorf R (2005) CD34+ fibrocytes in normal mitral valves and myxomatous mitral valve degeneration. Pathology Research and Practice, 201, 301-304.Beardow AW, Buchanan JW (1993) Chronic mitral valve disease in cavalier King Charles spaniels: 95 cases (1987–1991). Journal of the American Veterinary Medical Association, 203, 1023-1029.Bini A, Itoh Y, Kudryk BJ, Nagase H (1996) Degradation of cross-linked fibrin by matrix metalloproteinase 3 (stromelysin 1): hydrolysis of the gamma Gly 404-Ala 405 peptide bond. Biochemistry, 35, 13056-13063.Black A, French AT, Dukes-McEwan J, Corcoran BM (2005) Ultrastructural morphologic evaluation of the phenotype of valvular interstitial cells in dogs with myxomatous degeneration of the mitral valve. American Journal of Veterinary Research, 66, 1408-1414.Buchanan JW (1977) Chronic valvular disease (endocardiosis) in dogs. Advances in Veterinary Science and Comparative Medicine, 21, 75-106.Butcher JT, Nerem RM (2006) Valvular endothelial cells regulate the phenotype of interstitial cells in co-culture: effects of steady shear stress. Tissue Engineering, 12, 905-915.Butcher JT, Penrod AM, Garcia AJ, Nerem RM (2004) Unique morphology and focal adhesion development of valvular endothelial cells in static and fluid flow environments. Arteriosclerosis, Thrombosis and Vascular Biology, 24, 1429-1434.Carrier RL, Papadaki M, Rupnick M, Schoen FJ, Bursac N et al. (1999) Cardiac tissue engineering: cell seeding, cultivation parameters, and tissue construct characterization. Biotechnology and Bioengineering, 64, 580-589.Cholewinski E, Dietrich M, Flanagan TC, Schmitz-Rode T, Jockenhoevel S (2009) Tranexamic acid – an alternative to aprotinin in fibrin-based cardiovascular tissue engineering. Tissue Engineering Part A, 15, 3645-3653.Cole WG, Chan D, Hickey AJ, Wilcken DE (1984) Collagen composition of normal and myxomatous human mitral heart valves. Biochemical Journal, 219, 451-460.Connell PS, Han RI, Grande-Allen KJ (2012) Differentiating the aging of the mitral valve from human and canine myxomatous degeneration. Journal of Veterinary Cardiology, 14, 31-45.Corcoran BM, Black A, Anderson H, McEwan JD, French A et al. (2004) Identification of surface morphologic changes in the mitral valve leaflets and chordae tendineae of dogs with myxomatous degeneration. American Journal of Veterinary Research, 65, 198-206.Disatian S, Ehrhart EJ 3rd, Zimmerman S, Orton EC (2008) Interstitial cells from dogs with naturally occurring myxomatous mitral valve disease undergo phenotype transformation. Journal of Heart Valve Disease, 17, 402-412.Flanagan TC, Cornelissen C, Koch S, Tschoeke B, Sachweh JS et al. (2007) The in-vitro development of autologous fibrin-based tissue-engineered heart valves through optimised dynamic conditioning. Biomaterials, 28, 3388-3397.Flanagan TC, Wilkins B, Black A, Jockenhoevel S, Smith TJ et al. (2006) A collagen-glycosaminoglycan co-culture model for heart valve tissue engineering applications. Biomaterials, 27, 2233-2246.Gotlieb AI, Rosenthal A, Kazemian P (2002) Fibroblast growth factor 2 regulation of mitral valve interstitial cell repair in vitro. Journal of Thoracic Cardiovascular Surgery, 124, 591-597.Gupta V, Barzilla JE, Mendez JS, Stephens EH, Lee EL et al. (2009) Abundance and location of proteoglycans and hyaluronan within normal and myxomatous mitral valves. Cardiovascular Pathology, 18, 191-197.Gupta V, Werdenberg JA, Lawrence BD, Mendez JS, Stephens EH et al. (2008a) Reversible secretion of glycosaminoglycans and proteoglycans by cyclically stretched valvular cells in 3D culture. Annals of Biomedical Engineering, 36, 1092-1103.Gupta V, Werdenberg JA, Mendez JS, Grande-Allen KJ (2008b) Influence of strain on proteoglycan synthesis by valvular interstitial cells in three-dimensional culture. Acta Biomaterealia, 4, 88-96.Hadian M, Corcoran BM, Bradshaw JP (2010) Molecular changes in fibrillar collagen in myxomatous mitral valve disease. Cardiovascular Pathology, 19, e141-e148.Han RI, Black A, Culshaw G, French AT, Corcoran BM (2010) Structural and cellular changes in canine myxomatous mitral valve disease: an image analysis study. Journal of Heart Valve Disease, 19, 60-70.Han RI, Black A, Culshaw GJ, French AT, Else RW et al. (2008) Distribution of myofibroblasts, smooth muscle-like cells, macrophages, and mast cells in mitral valve leaflets of dogs with myxomatous mitral valve disease. American Journal of Veterinary Research, 69, 763-769.Han RI, Clark CH, Black A, French A, Culshaw GJ et al. (2013) Morphological changes to endothelial and interstitial cells and to the extra-cellular matrix in canine myxomatous mitral valve disease (endocardiosis). Veterinary Journal, 197, 388-394.Hayek E, Gring CN, Griffin BP (2005) Mitral valve prolapse. Lancet, 365, 507-518.Jockenhoevel S, Zund G, Hoerstrup SP, Chalabi K, Sachweh JS et al. (2001) Fibrin gel – advantages of a new scaffold in cardiovascular tissue engineering. European Journal of Cardiothoracic Surgery, 19, 424-430.Kural MH, Billiar KL (2016) Myofibroblast persistence with real-time changes in boundary stiffness. Acta Biomaterealia, 32, 223-230.Latif N, Quillon A, Sarathchandra P, McCormack A, Lozanoski A et al. (2015). Modulation of human valve interstitial cell phenotype and function using a fibroblast growth factor 2 formulation. PLoS One, 10, e0127844.Lester WM, Damji AA, Gedeon I, Tanaka M (1993) Interstitial cells from the atrial and ventricular sides of the bovine mitral valve respond differently to denuding endocardial injury. In Vitro Cell and Developmental Biology, 29A, 41-50.Lester WM, Damji AA, Tanaka M, Gedeon I (1992) Bovine mitral valve organ culture: role of interstitial cells in repair of valvular injury. Journal of Molecular and Cellular Cardiology, 24, 43-53.Liu MM, Flanagan TC, Lu CC, French AT, Argyle DJ et al. (2015) Culture and characterisation of canine mitral valve interstitial and endothelial cells. Veterinary Journal, 204, 32-39.Lu CC, Liu MM, Clinton M, Culshaw G, Argyle DJ et al. (2015) Developmental pathways and endothelial to mesenchymal transition in canine myxomatous mitral valve disease. Veterinary Journal.Lu CC, Liu MM, Culshaw G, French A, Corcoran B (2016) Comparison of cellular changes in Cavalier King Charles spaniel and mixed breed dogs with myxomatous mitral valve disease. Journal of Veterinary Cardiology, 18, 100-109.Mow T, Pedersen HD (1999) No expression of angiotensin II receptors and angiotensin-converting enzyme in myxomatous canine mitral valve leaflets. An autoradiographic study. Zentralbl Veterinarmed A, 46, 465-472.Mueller-Klieser W (1997) Three-dimensional cell cultures: from molecular mechanisms to clinical applications. American Journal of Physiology, 273, C1109-C1123.Pedersen HD, Haggstrom J (2000) Mitral valve prolapse in the dog: a model of mitral valve prolapse in man. Cardiovascular Research, 47, 234-243.Rabkin-Aikawa E, Farber M, Aikawa M, Schoen FJ (2004) Dynamic and reversible changes of interstitial cell phenotype during remodeling of cardiac valves. Journal of Heart Valve Disease, 13, 841-847.Rabkin-Aikawa E, Mayer JE Jr, Schoen FJ (2005) Heart valve regeneration. Advances in Biochemical Engineering and Biotechnology, 94, 141-179.Rabkin E, Aikawa M, Stone JR, Fukumoto Y, Libby P et al. (2001) Activated interstitial myofibroblasts express catabolic enzymes and mediate matrix remodeling in myxomatous heart valves. Circulation, 104, 2525-2532.Stein PD, Wang CH, Riddle JM, Sabbah HN, Magilligan DJ Jr. et al. (1989) Scanning electron microscopy of operatively excised severely regurgitant floppy mitral valves. American Journal of Cardiology, 64, 392-394.Stephens EH, Carroll JL, Post AD, Kuo JJ, Grande-Allen KJ (2010) Functional characterization of fibronectin-separated valve interstitial cell subpopulations in three-dimensional culture. Journal of Heart Valve Disease, 19, 759-765.Stephens EH, Durst CA, West JL, Grande-Allen KJ (2011) Mitral valvular interstitial cell responses to substrate stiffness depend on age and anatomic region. Acta Biomaterealia, 7, 75-82.Tamura K, Fukuda Y, Ishizaki M, Masuda Y, Yamanaka N et al. (1995) Abnormalities in elastic fibers and other connective-tissue components of floppy mitral valve. American Heart Journal, 129, 1149-1158.Tamura K, Jones M, Yamada I, Ferrans VJ (2000) Wound healing in the mitral valve. Journal of Heart Valve Disease, 9, 53-63.Waxman AS, Kornreich BG, Gould RA, Moise NS, Butcher JT (2012) Interactions between TGFbeta1 and cyclic strain in modulation of myofibroblastic differentiation of canine mitral valve interstitial cells in 3D culture. Journal of Veterinary Cardiology, 14, 211-221.Whitney JC (1974) Observations on the effect of age on the severity of heart valve lesions in the dog. Journal of Small Animal Practice, 15, 511-522.Ye Q, Zund G, Jockenhoevel S, Schoeberlein A, Hoerstrup SP et al. (2000) Scaffold precoating with human autologous extracellular matrix for improved cell attachment in cardiovascular tissue engineering. Asaio Journal, 46, 730-733.Received, October 23rd, 2017 Accepted,Figure Legends Fig. 1. Morphology of fibrin-based canine VEC/VIC 3D co-culture constructs. (A) Shrinkage of construct (dashed arrow) after 14 d in culture compared with a cell-free fibrin control (solid arrow). Bar, 1cm. (B) A construct showing cells distributed within the fibrin-based stroma and a dense cell/matrix layer at the construct medium contact surface (endothelium) (arrow). HE. (C) Movat pentachrome staining showing de novo ECM synthesis; collagen fibres in stroma (yellow) and mucin in subendothelial area (blue–green). (D) DiI-Ac-LDL labelling of endothelial cells on construct surface. Immunofluorescence microscopy. ×200.Fig. 2. Assessment of expression of cellular and ECM markers comparing normal mitral valve and type I constructs (normal VIC/VEC) using immunofluorescence. Left panel (A): vimentin- (green A and B) and SMemb- (red C and D) positive cells are evenly distributed in both the construct and native valve. α-SMA expression (green E and F) is minimal in the native valve, but evident beneath the dense cell layer of the construct (arrow). Centre panel (B): collagen I and III (red A and B), MMP-1 (red C and D) are detected in both native valve and the constructs. MMP-3 (red E and F) is minimally expressed in native valve and partially detected in constructs. Right panel (C): biglycan (green A and B) is extensively distributed in all layers in both native valve and constructs. Decorin (red C and D) is expressed in a similar pattern to biglycan in native valve, more so in the fibrosa layer, but expression is minimal in the constructs (D). Ki67 expression (cell proliferation marker; red E and F) is identified in the majority of cells in the native valve and constructs. DAPI (blue) counterstained cell nuclei. Bars, 100 ?m.Fig. 3. Comparison of type 1 (normal VIC/VEC) and 2 (diseased VIC/normal VEC) constructs using immunofluorescence and western blotting. (A) MMVD related marker expression assessed by immunofluorescence microscopy. No apparent differences are appreciated in expression patterns for MMP-1 (red A and G), MMP-3 (red B and H), SMemb (red C and I), α-SMA (green D and J) and vimentin (green E and K). DAPI (blue) counterstained cell nuclei. Bars, 100 ?m. (B) MMVD related marker expression assessed by western blotting. Left panel: representative protein band patterns of two (1 and 2) experiments for MMP-1, MMP-3, Smemb, α-SMA, vimentin and GAPDH (loading control) show similar expression for the two construct types. Right panel: semiquantified protein band intensity (normalized to GAPDH and then by calculating the mean value of type 2/type 1 ratio for each marker) found no significant difference (ANOVA) between the two construct types. (C) Endothelial marker CD31 expression assessed by western blotting (n = 4 each type). There is detectable, but variable expression, of CD31 when comparing different constructs. Native valve (MV) as positive control. Fig. 4. Examination of construct response to endothelial damage (type 3 versus sham control) using immunofluorescence microscopy. Endothelial loss after damage is clearly seen in all type 3 constructs, with loss of cells in the dense cell layer. (A) From days 0 to 6, vimentin expression is consistent in both sham (A, C and E) and type 3 constructs (B, D and F). (B) At day 0, α-SMA (green) positive cells are evident in both sham (A) and type 3 constructs (B). By days 2 and 6, α-SMA positive cells have decreased markedly in all sham (C and E) and type 3 constructs (D and F). (C) At day 0, Smemb (red) positive cells are evident in both sham and type 3 constructs (A and B). By days 2 and 6, in sham controls SMemb expression is primarily observed in the construct stroma (C and E), while in the majority of type 3 constructs, in addition to stromal expression, SMemb-positive cells are found on or close to the wounded surface (arrows in D and F). DAPI (blue) counter-stained cell nuclei. Bars, 200 ?m. Bars, 100 ?m for inset images. Table 1Primary and secondary antibodies used in immunofluorescence and western blot analysisPrimary antibodyspecificityProduct informationConcentrationImmunofluorescenceWestern blotVimentin (MM)V6389, Sigma, St. Louis, Missouri, USA1 in 1,6001 in 2,000SMemb (RP)Ab24761, Abcam, Cambridge, UK1 in 1,0001 in 1,000α-SMA (MM)A2547, Sigma1 in 2001 in 1,000Collagen I and III (RP)Ab24137, Abcam1 in 10MMP-1 (RP)Ab38929, Abcam1 in 2001 in 5,000MMP-3 (RP)Ab53015, Abcam1 in 1001 in 500GAPDH (MM)CB1001, Calbiochem, Finland1 in 1,000Biglycan (MM)Ab54855, Abcam1 in 100Decorin (RP)Ab137508, Abcam 1 in 25Ki67 (RP)Ab15580, Abcam1 in 100Secondary antibodiesGoat anti-rabbit IgG (H+L) Alexa Fluor568A11011, Invitrogen, UK1 in 500Goat anti-mouse IgG (H+L) Alexa Fluor488A10667, Invitrogen1 in 500Polyclonal rabbit anti-mouse immunoglobulins/HRPP0260, Dako, Glostrup, Denmark1 in 1,000Polyclonal swine anti-rabbit immunoglobulins/HRPP0217, Dako1 in 3,000SMemb, embryonic smooth muscle myosin heavy chain; ?-SMA, alpha smooth muscle actin; MMP, matrix metalloproteinase; GAPDH, glyceraldehyde phosphate dehydrogenase; MM, mouse monoclonal; RP, rabbit polyclonal ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download