Effects of fibrillin mutations on the behavior of heart muscle cells in Marfan syndrome

Abstract

Marfan syndrome (MFS) is a systemic disorder of connective tissue caused by pathogenic variants in the fibrillin-1 (FBN1) gene. Myocardial dysfunction has been demonstrated in MFS patients and mouse models, but little is known about the intrinsic effect on the cardiomyocytes (CMs). In this study, both induced pluripotent stem cells derived from a MFS-patient and the line with the corrected FBN1 mutation were differentiated to CMs. Several functional analyses are performed on this model to study MFS related cardiomyopathy. Atomic force microscopy revealed that MFS CMs are stiffer compared to corrected CMs. The contraction amplitude of MFS CMs is decreased compared to corrected CMs. Under normal culture conditions, MFS CMs show a lower beat-to-beat variability compared to corrected CMs using multi electrode array. Isoproterenol-induced stress or cyclic strain demonstrates lack of support from the matrix in MFS CMs. This study reports the first cardiac cell culture model for MFS, revealing abnormalities in the behavior of MFS CMs that are related to matrix defects. Based on these results, we postulate that impaired support from the extracellular environment plays a key role in the improper functioning of CMs in MFS.

Figure 1

Phase contrast or immunofluorescent pictures from cell cultures of corrected (A,C,E,G) and MFS (B,D,F,H). hiPSC cultures have similar growth characteristics (A,B) and show expression of pluripotency markers OCT4 (green) and SOX2 (red) and HOECHST (blue) (C,D). Both CMs from corrected and MFS are growing as a layer on top of a basal layer of cells (E,F). The area of CMs in (E) and (F) is highlighted by the dashed line. Passaged CMs show better recovery in corrected (G) compared to MFS CMs (H) as indicated by the more brighter areas in corrected CMs. Scalebars indicate 500 µm. Graph (I) shows the ratio between TNNI3 and TNNI1 as a ratiometric marker for maturation of the CMs at day 15 and day 24, n = 3. Graph (J) represents the relative gene expression for TNNT2 in the cardiac model at day 24 for corrected (black) and MFS (red), n = 6.

Figure 2

Immunohistochemical staining for fibrillin-1 on CM cell culture for corrected and MFS. Fibrillin-1 in red and nuclei stained with HOECHST in blue after 25 days after the start of cardiac differentiation. Fibrillin-1 microfibrils or punctate assemblies are observed in corrected or MFS cultures, respectively. The average fluorescent intensity of fibrillin-1 in this image is respectively 26.65 and 19.51 for corrected and MFS cell culture. Orthogonal view shows the z-section of the area, indicated with the arrow in the overlay image. Bar indicates 100 µm.

Figure 3

Atomic Force Microscopy of 25-day-old cardiomyocytes: phase-contrast images of corrected (A) and MFS (B) cell cultures the mechanical stiffness mapping for corrected (C) and MFS (D) and the topography mapping of the (height) for corrected (E) and MFS (F) (colours: red is stiff, blue is more elastic) for corrected (C) and MFS (D) cells. Annotation of cell border (green line) and nuclear area (blue) in (E) and (F). Mechanical stiffness mapping is overlaid with topography mapping for corrected (G) and MFS (H). Dimensions are indicated in µm. AFM was performed on three independent cultures for each cell line. Large scale measurements of the complete CM cell cultures show a higher average Young’s modulus for the corrected (black) cell culture in comparison to MFS (red) (p < 0.0001) (I). The CMs from MFS show a higher Young’s modulus compared to corrected CMs when only the 25% highest areas, composed of CMs, were measured (p < 0.0001) (J). The bars in (I) and (J) indicate the standard deviation. The level of significance is indicated by asterisks: p values less than 0.001 are indicated with three asterisks. Bar in panel (A) and (B) indicates 500 µm.

Figure 4

Video analysis of the beating of cardiomyocytes using Musclemotion was performed on corrected (n = 3, black) and MFS cell cultures (n = 3, red). The representative plots show differences in contraction amplitude. The bar graph shows the average contraction amplitude of three independent experiments per cell line and shows a significant difference (p = 0.0078), error bars represent the standard error in the bar graph. The level of significance is indicated by asterisks: p values less than 0.01 indicated with two asterisks.

Figure 5

MEA measurements of the extracellular field potentials for the control line (H9), the corrected CMs and the MFS CMs after 20–25 days after start of cardiac differentiation. (A) The baseline measurements without the addition of isoproterenol (ISO). The extracellular field potentials are shown for 10 nM (B) and 100 nM ISO (C). The bar plots show the differences between control, corrected and MFS CMs for RR interval (D), SDNN (E), SDSD (F) and coefficient of variance (G). Corrected CMs (n = 9, black), MFS CMs (n = 8, red). Poincaré plot from three representative replicates of the control (blue), corrected (black) and MFS CMs (red) (H). The three clusters of MFS are magnified in the right corner of figure (H). The level of significance is indicated by asterisks: p values less than 0.001 are indicated with three asterisks and p values less than 0.01 indicated with two asterisks.

Figure 6

MEA measurements of control, corrected and MFS CMs after treatment with serial isoproterenol (ISO) concentrations of 5, 10, 25, 50 and 100 nM. Bar plots show differences between the three lines at each ISO concentration for RR interval (A), SDNN (B), SDSD (C) and the coefficient of variance presented as factor of the baseline (D). The Spearman’s rank correlation is indicated in panel A. Corrected CMs (n = 6, black), MFS CMs (n = 7, red), H9 control (n = 5, blue).

Figure 7

Cell-deposition of fibronectin. Chronic isoproterenol (ISO) treatment (1 µM) for 7 days for both MFS and corrected cell cultures is compared with no treatment using immunohistochemistry staining for cTnT and fibronectin, nuclei visualised with HOECHST. Bar indicates 100 µm. Bar graphs show average fluorescent intensity of cTnT and fibronectin for corrected (n = 3, black) and MFS (n = 3, red) with ISO treatment and no treatment. The level of significance is indicated by asterisks: p values less than 0.05 are indicated with one asterisk.

Figure 8

Stretching of in vitro cells to analyse structural integrity of the MFS CMs. Fluorescent images of immunohistochemistry staining for cTnT (green) after 6 h and 40 h of stretching using Flexcell and unstretched control (on membrane) for both MFS and corrected CMs. Bar plot shows the average intensity of cTnT in corrected (black) and MFS (red) CMs and includes the standard deviation. Bar = 100 µm.

Figure 9

RR interval of corrected (black) and MFS (red) CMs measured on different substrate stiffnesses: glass (≈10 GPa) n = 8, plastic (≈0.1 GPa) n = 3 and on a flexible Flexcell membrane (≈150 kPa) n = 6. Error bars represent the standard error in the bar graphs. The level of significance is indicated by asterisks: p values less than 0.001 are indicated with three asterisks and p values less than 0.01 indicated with two asterisks.