High-Throughput Phenotyping Toolkit for Characterizing Cellular Models of Hypertrophic Cardiomyopathy In Vitro

Hypertrophic cardiomyopathy (HCM) is a prevalent and complex cardiovascular disease characterised by multifarious hallmarks, a heterogeneous set of clinical manifestations, and several molecular mechanisms. Various disease models have been developed to study this condition, but they often show contradictory results, due to technical constraints and/or model limitations. Therefore, new tools are needed to better investigate pathological features in an unbiased and technically refined approach, towards improving understanding of disease progression. Herein, we describe three simple protocols to phenotype cellular models of HCM in vitro, in a high-throughput manner where technical artefacts are minimized. These are aimed at investigating: (1) Hypertrophy, by measuring cell volume by flow cytometry; (2) HCM molecular features, through the analysis of a hypertrophic marker, multinucleation, and sarcomeric disarray by high-content imaging; and (3) mitochondrial respiration and content via the Seahorse™ platform. Collectively, these protocols comprise straightforward tools to evaluate molecular and functional parameters of HCM phenotypes in cardiomyocytes in vitro. These facilitate greater understanding of HCM and high-throughput drug screening approaches and are accessible to all researchers of cardiac disease modelling. Whilst HCM is used as an exemplar, the approaches described are applicable to other cellular models where the investigation of identical biological changes is paramount.


Introduction
Hypertrophic cardiomyopathy (HCM) is an intricate disease affecting 1:500 individuals where heart dysfunction often leads to sudden cardiac death [1]. The complexity of HCM encompasses a wide range of disease hallmarks as well as clinical manifestations and outcomes, hampering the development of efficient pharmacological treatment options [2]. Modeling HCM in vitro leads to a better understanding of disease progression towards unveiling novel molecular mechanisms and paving the way for targeted drug therapy [3].
Various models based on tissue and cellular explants, skinned myofibrils, and actin/myosin preparations have been generated, greatly contributing to uncover disease hallmarks and facilitating the development of new drugs, with some reaching clinical trials [4]. However, these disease models frequently produce contradictory results due not only to the recognized disease complexity (e.g., mutation-specific effects), but also owing to fundamental differences and limitations between models, such as low availability and demanding logistics of heart explants, over-simplicity of protein preparations, and species-differences relative to animal models [4].

Experimental Design
Accurate phenotyping of cellular models of HCM via this toolkit typically requires dissociation of 2D monolayers or 3D constructs into single-cell suspensions. Efficient isolation of cardiomyocytes from heart tissue biopsies has been the subject of numerous reports and typically consists of enzymatic bulk digestion [30], Langendorff method [31] or mechanical disruption procedures [32]. A recent protocol optimized this process in five main steps: 1) Myocardium dissection into 200 µm-thick slices; 2) slice perfusion with a Ca 2+ -free solution; 3) enzymatic digestion using collagenase II and protease XXIV; 4) filtration of cardiac tissue extract with a 100 µm mesh (to break cell clumps and minimize cell sampling biases [33]); and 5) in vitro culture in 5% foetal bovine serum-containing medium [34]. This method resulted in up to 65% viable cardiomyocyte isolation yield and enabled phenotypic studies of electrophysiology, Ca 2+ imaging, and Seahorse™ analysis. Alternatively, cardiomyocytes can be efficiently sourced in high numbers through cardiac differentiation of hPSCs [35] followed by their dissociation into single cells using a collagenase II digestion protocol [36].
While these cell sources and their culture methods vary between laboratories, baseline conditions such as serum supplementation and time in culture should be kept constant between the different groups being compared (e.g., cell lines or treatments), to minimize technical artefacts.
All cell culture manipulation is performed in a routine type II Biological Safety Cabinet, and cells are cultured in a humidified incubator, at 37 • C and 5% CO 2 . Once a single cell suspension is obtained, each of these 3 protocols can be performed (Figure 1).
While the hypertrophy analysis consists of direct assessment of freshly dissociated cardiomyocytes, the other 2 protocols require further cell culture upon replating. All protocols have built-in quality control steps (e.g., exclusion of cell debris and non-cardiomyocytes) for the cells being used, as well as normalization methods to account for differences in cell numbers.
All the exemplary data reported in this manuscript ( Figure 8) are based on the comparison between a hPSC-CM cell line bearing the R453C-β-myosin heavy-chain mutation (HCM line) and its isogenic wild-type control (healthy). hPSC lines were derived in-house [37] from material harvested under ethical approval from Nottingham Research Ethics Committee 2 (09/H0408/74) with informed patient consent. Workflow showing toolkit for phenotyping hypertrophic cardiomyopathy (HCM) in cellular models. Cardiomyocytes can be obtained from different sources (biopsies or cell lines) and require dissociation into a single cell suspension prior to phenotyping. Evaluation of in vitro hallmarks of HCM is done via: 1) Flow volumetry to investigate hypertrophy; 2) high-content imaging to assess multinucleation, brain natriuretic peptide (BNP) hypertrophic marker expression, and sarcomeric disarray; 3) Seahorse assay and qPCR to assess mitochondrial respiration and content, respectively.  (1) Flow volumetry to investigate hypertrophy; (2) high-content imaging to assess multinucleation, brain natriuretic peptide (BNP) hypertrophic marker expression, and sarcomeric disarray; (3) Seahorse assay and qPCR to assess mitochondrial respiration and content, respectively.    [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].

Hypertrophic Analysis
CRITICAL STEP Prior culture of cardiomyocytes in serum-containing medium may mask hypertrophic phenotypes [17], so exposure to serum should be minimized or fully eliminated.  [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].
CRITICAL STEP Use of 100 µM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].

3.
Run cell samples in the flow cytometer, using the same operating settings as those used for the calibration beads.  (Figure 2A,B). 5. Gate to define FSC-A for each bead used for the calibration. Export FSC-A values relative to PAUSE STEP Once data from flow cytometry is acquired, it can be analysed at any point. In the Kaluza software (or any equivalent flow cytometry analysis software), gate to remove debris and duplets/triplets (exclude events with very low FSC and SSC Area values) (Figure 2A,B).

5.
Gate to define FSC-A for each bead used for the calibration. Export FSC-A values relative to each gate (corresponding to each bead) to an excel file. Ensure there is consistency between the units being used for beads and samples (i.e., linear or log) ( Figure 2C,D). 6.
In the Excel software, plot the FSC-A median values (y axis, obtained from each gate in the Kaluza software) relative to the known bead diameters (x axis). Add trendline of linear regression and display equation (FSCA = slope × diameter + intercept) and R-square value determining the calibration curve ( Figure 2E).

Mitochondrial Respiration and Content
• Seahorse Wave Desktop Software (Agilent, Santa Clara, CA, USA, https://www.agilent.com/en/products/cell-analysis/cell-analysis-software/dataanalysis/wave-desktop-2-6  [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].
CRITICAL STEP If the R-squared value (measure of how much the variation of the FSCA is explained by variation of the bead size) is below 0.95, then the bead solutions must be thoroughly mixed to disrupt clumps. 7.
Gate samples to remove debris and duplets/triples. In cell samples, SSC-width is used in the x axis instead of FSC-A due to higher heterogeneity of the cardiomyocytes relative to the beads ( Figure 2F-H). 8.
Export data from each sample from Kaluza to Excel, as done in 5. 9.
Using the calibration equation determined in 6, convert the FSC-A values measured for each 10. Determine the cell volume by applying the formula of the volume of a sphere: V = Π 6 × d 3 , where d is the cell diameter determined in 9. This assumes the approximation of each cell in suspension to a sphere, which is also applicable to the calibration beads. 11. Import the cell volume values into GraphPad software, as a column table (whereby each column represents a separate sample). 12. Choose "violin plot only" in the "Box and plot" options to graphically represent the data distribution and the main statistical parameters (median and quartiles). Adjust colour as desired ( Figure 8A). 13. Perform statistical analysis (depending on the experimental design) to compare volume distribution between samples. OPTIONAL STEP Plotting such large sample sizes in GraphPad can be computationally demanding. Alternatively, a boxplot can be generated by inputting only 7 cardiomyocyte volume values per condition: Minimum, maximum; 25th percentile, 75th percentile, and the median three times. These values can be easily determined using Excel™ software ( Figure 8B).   Supplement RPMI 1640 medium by adding B27 supplement (RB27, 50x dilution), aliquot volume needed and let it reach RT. OPTIONAL STEP Other extracellular matrix proteins and culture media could be used instead of Vitronectin and RB27, provided they were previously tested for the specific cardiomyocyte source in culture.

4.
Aspirate Vitronectin and add 50 µL of PBS per well. Aspirate PBS and add 75 µL of RB27 per well.

5.
Plate cardiomyocytes at approximately 110,000 cells/cm 2 (35,000 cells per well). OPTIONAL STEP Allow the cells to settle for 0.5 h at room temperature before placing them in the incubator, to minimize edge effects due to the formation of convection currents because of temperature differences between the medium and the incubator. 1. Resuspend freshly dissociated cardiomyocytes in 500 μL Phosphate Buffer Saline (PBS) at 1×10 5 -2.5x10 5 cells per sample and place them on ice prior to flow cytometry analysis. CRITICAL STEP Prior culture of cardiomyocytes in serum-containing medium may mask hypertrophic phenotypes [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].
CRITICAL STEP As controls for immunostaining, some wells should be included for staining with the secondary antibody only to detect unspecific binding (secondary-only control). Alternatively, a cell-type control based on culturing non-cardiomyocytes (e.g., fibroblasts) in the same plate type can also be used and immunostained the same way as cardiomyocytes ( Figure A1). analysis/wave-desktop-2-6).

Hypertrophy Analysis. Time for Completion: 1 day
3.1.1. Data Acquisition. Time for Completion: 2 h 1. Resuspend freshly dissociated cardiomyocytes in 500 μL Phosphate Buffer Saline (PBS) at 1×10 5 -2.5x10 5 cells per sample and place them on ice prior to flow cytometry analysis. CRITICAL STEP Prior culture of cardiomyocytes in serum-containing medium may mask hypertrophic phenotypes [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].
CRITICAL STEP BFA treatment prevents BNP secretion in cardiomyocytes, reaching maximal levels at 15 h post ET1 treatment. ET1 and BOS treatments are required to threshold BNP signal intensity ( Figure A2). 3. Run cell samples in the flow cytometer, using the same operating settings as those used for the calibration beads. PAUSE STEP Once data from flow cytometry is acquired, it can be analysed at any point.  (Figure 2A,B). 5. Gate to define FSC-A for each bead used for the calibration. Export FSC-A values relative to each gate (corresponding to each bead) to an excel file. Ensure there is consistency between the units being used for beads and samples (i.e., linear or log) ( Figure 2C,D). 6. In the Excel software, plot the FSC-A median values (y axis, obtained from each gate in the Kaluza software) relative to the known bead diameters (x axis). Add trendline of linear regression and display equation ( = × + ) and R-square value determining the calibration curve ( Figure 2E).
CRITICAL STEP If the R-squared value (measure of how much the variation of the FSC-A is explained by variation of the bead size) is below 0.95, then the bead solutions must be thoroughly mixed to disrupt clumps. 7. Gate samples to remove debris and duplets/triples. In cell samples, SSC-width is used in the x axis instead of FSC-A due to higher heterogeneity of the cardiomyocytes relative to the beads ( Figure 2 F-H). 8. Export data from each sample from Kaluza to Excel, as done in 5. 10. Determine the cell volume by applying the formula of the volume of a sphere: = × , where d is the cell diameter determined in 9. This assumes the approximation of each cell in suspension to a sphere, which is also applicable to the calibration beads. 11. Import the cell volume values into GraphPad software, as a column table (whereby each column represents a separate sample). 12. Choose "violin plot only" in the "Box and plot" options to graphically represent the data distribution and the main statistical parameters (median and quartiles). Adjust colour as desired ( Figure 8A). 13. Perform statistical analysis (depending on the experimental design) to compare volume distribution between samples. OPTIONAL STEP Plotting such large sample sizes in GraphPad can be computationally demanding. Alternatively, a boxplot can be generated by inputting only 7 cardiomyocyte volume values per condition: Minimum, maximum; 25th percentile, 75th percentile, and the median three times. These values can be easily determined using Excel™ software ( Figure  8B).

PAUSE STEP
Fixed cell plates can be stored at 4 • C for up to 2 months prior to immunostaining and imaging. Wrap plates in parafilm to prevent PBS evaporation and consequent plate drying.

8.
Wash fixed cells with PBS. Permeabilize cells by treating with 0.1% v/v Triton-X and incubating at RT for 15 min. 9.
Wash 3 times with PBS. 10. Add 4% v/v goat serum in PBS (blocking solution). Incubate for 1 h at RT. 11. Replace blocking solution by primary antibodies pertaining each staining: a.

Hypertrophic Analysis
• Kaluza analysis v2.  [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].

CRITICAL STEP
The secondary-only controls should be incubated with blocking solution instead. Cell-type controls should be treated the same way as cardiomyocytes. 12 (Figure 2A,B). 5. Gate to define FSC-A for each bead used for the calibration. Export FSC-A values relative to each gate (corresponding to each bead) to an excel file. Ensure there is consistency between the units being used for beads and samples (i.e., linear or log) ( Figure 2C,D).  Insert "Calculate Morphology Properties" building block. Select Population "Nuclei"; Region: "Nucleus"; Method "Standard", tick "Area" and Roundness" in the inset menu. Define Output Properties as "Nuclei Properties" ( Figure 3F). 4.

5.
Insert "Select Population" building block. Select Population "Nuclei Selected"; Method "Common Filters" and tick "Remove Border Objects" box. Output Population: "Nuclei Final" ( Figure 3H).  [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].
CRITICAL STEP These steps ensure exclusion of debris, cell clumps, and cells in the border of the field from the subsequent analysis. The exact values were optimized for hPSC-CMs and may need to be adapted to different cardiomyocyte source being investigated. 6.
Insert "Calculate Morphology Properties" building block. Select Population "Nuclei Final"; Region: "Nucleus"; Method: "Standard". Tick "Area" and "Roundness" boxes in the inset menu. Output properties: "Nucleus Morphology". This block does the same operation as in 3 but for the population "Nuclei Final". 7.
Insert "Modify Population" building block. Select Population "Nuclei Final"; Region "Nucleus"; Method "Cluster by Distance". Enter in the inset menu Distance "5 µm"; Area > "80 µm 2 ". Output population: "Total population" (Figure 3I).  [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].

Procedure
CRITICAL STEP These steps enable the identification of several nuclei belonging to the same cell, other than being part of the different cells, by inferring on their proximity. The values pertaining "Distance" and "Area" may need to be changed when using cardiomyocyte sources other than hPSC-CMs (i.e., displaying different morphology properties). 8.
Insert "Calculate Properties" building block. Population "Total population"; Method "By Related Population". In the inset menu: Related Population "Nuclei final"; tick box: "Number of Nuclei Final". Output Properties: "per Cell". This step will effectively calculate the number of nuclei belonging to each cell. 9.
Insert  11. In the default "Define Results" building block select Method "List of Outputs" in the drop down menu, then tick the box "Number of Objects" for the options: "Population: Total population", "Population: Mononucleated", Population: Binucleated", and "Population: Multinucleated". 12. In the same drop-down menu within "Define Results", define three sections of "Method-Formula Output". In each of these sections in the inset menu, enter "Formula: (a/b)×100; Variable A: "Mononucleated-Number of Objects", "Binucleated-Number of Objects", or "Multinucleated-Number of Objects". Enter as Variable B "Total population-Number of Objects". The Output Names for each section are "%mononucleated", "%binucleated", and "%multinucleated", respectively. 13. Copy the percentages pertaining to each proportion of mono-, bi-, or multinucleated cells to a GraphPad "Grouped" Table, by inserting "% mononucleated", "% binucleated", and "% Insert "Select Population" building block. Population "Total population"; Method "Filter by Property". Inset menu: "Number of Nuclei Final -per Cell > 2". Output Population "Multinucleated" ( Figure 3L). 11. In the default "Define Results" building block select Method "List of Outputs" in the drop down menu, then tick the box "Number of Objects" for the options: "Population: Total population", "Population: Mononucleated", Population: Binucleated", and "Population: Multinucleated". 12. In the same drop-down menu within "Define Results", define three sections of " Method-Formula Output". In each of these sections in the inset menu, enter "Formula: (a/b)×100; Variable A: "Mononucleated-Number of Objects", "Binucleated-Number of Objects", or "Multinucleated-Number of Objects". Enter as Variable B "Total population-Number of Objects". The Output Names for each section are "%mononucleated", "%binucleated", and "%multinucleated", respectively. 13. Copy the percentages pertaining to each proportion of mono-, bi-, or multinucleated cells to a GraphPad "Grouped" Table, by inserting "% mononucleated", "% binucleated", and "% multinucleated" in columns and different cell lines/conditions in rows. 14. Create a "Stacked bars" Graph Type. Colour each bar and perform statistical analysis according to the experimental design ( Figure 8C). OPTIONAL STEP As a quality control measure, the cardiomyocyte purity in the cell population being analysed can be determined in a separate analysis script, as follows ( Figure 4): 15. After repeating building blocks in steps 1-5 of Section 3.2.5.1, insert "Find Cytoplasm" building block. Channel "DRAQ5" (same excitation/emission parameters as HCS Cell Mask DeepRed); Nuclei "Nuclei Final"; Method "F". In the inset menu, define Membrane Channel "DRAQ5"; Individual Threshold: 0.04. This step delineates the cytoplasm of cardiomyocytes by using the Cell Mask counterstain ( Figure 4A-C). 16. Insert "Calculate Intensity Properties" building block. Channel "Alexa 488"; Population: "Nuclei Final"; Region "Cytoplasm"; Method "Standard", tick box "Mean" in the inset menu. Output Properties "Intensity alpha-actinin". 17. Insert "Select Population" building block. Population "Nuclei Final"; Method "Filter by Property".
In the inset menu, insert "Intensity alpha-actinin Mean" and set it > the value of the secondary only or cell-type control ( Figures 4D and A1). Output population "Cardiomyocytes". Confirm if the highlighted cells display positive α-actinin signal ( Figure 4E). 18. Insert "Select Population" building block. Population "Nuclei Final"; Method "Filter by Property".
In the inset menu, insert "Intensity alpha-actinin Mean" and set it ≤ the value of the secondary only or cell-type control. Output population "Non-Cardiomyocytes". Confirm if the highlighted cells display negative alpha-actinin signal ( Figure 4F). 19. In the default "Define Results" building block, select Method "List of Outputs", then tick the box "Number of Objects" for the options "Population: Nuclei Final" and "Population: Cardiomyocytes". 20. In the same drop-down menu within "Define Results", define "Method-Formula Output", and in the inset menu, enter "Formula: (a/b)*100; Variable A: "Cardiomyocytes-Number of Objects"; Variable B "Nuclei Final-Number of Objects". Output Name "% Cardiomyocyte Purity". 21. Import purity values to GraphPad (e.g., "Column" table, "Scatter plot" graph) and perform statistical analysis according to the experimental design ( Figure 8D). Cardiomyocyte purities should typically be >90% using directed differentiation protocols from hPSCs [35].  Figure 5F). The sarcomeric signal is ubiquitous in cardiomyocytes so can be used to define cytoplasm, similar to the Cell Mask. 4. Insert "Calculate Intensity Properties" building block. Select Population "Nuclei Final"; Region: "Cytoplasm"; Method "Standard", tick "Mean" and "Standard Deviation" in the inset menu. Define Output Properties as "Intensity cardiac Troponin T". 5. Insert "Select Population" building block. Population "Nuclei Final"; Method "Filter by Property". In the inset menu, insert "Intensity cardiac Troponin T" and set it > the value of the secondary only or cell-type control (e.g., fibroblasts). Output population "Cardiomyocytes". Confirm if the highlighted cells display positive cardiac Troponin T signal. This is important so the subsequent analysis is performed only in cardiomyocytes. It can also be used for calculating purity (Figures A1 and 5G). 6. Insert "Select Region" building block. Select Population "Nuclei Final"; Region "Nucleus"; Method "Resize Region [%]". In the inset menu, enter "-75%" in the section Outer Border. Leave "Inner Border" section empty. Output Population: "Cytoplasmic ring". This step defines a perinuclear ring where BNP expression is maximal ( Figure 5H).  7. Insert "Calculate Intensity Properties" building block. Select Channel "Alexa 488"; Population "Cardiomyocytes"; Region: "Cytoplasmic Ring"; Method "Standard", tick "Mean" and "Standard Deviation" in the inset menu. Define Output Properties as "Intensity BNP perinuclear". 8. Insert "Select Population" building block. Population "Cardiomyocytes"; Method "Filter by Property". In the inset menu, insert "Intensity BNP perinuclear" and set it > the value of the bosentan-treated cardiomyocytes ( Figure A2). Output population "BNP positive cardiomyocytes". Confirm if the highlighted cells display positive BNP signal. This signal should always be higher than that of secondary-only or cell-type controls ( Figure 5I). 9. Insert "Select Population" building block. Population "Cardiomyocytes"; Method "Filter by Property". In the inset menu, insert "Intensity BNP perinuclear" and set it ≤ the value of the bosentan-treated cardiomyocytes (same value as in 8). Output population "BNP negative cardiomyocytes". Confirm if the highlighted cells do not display positive BNP signal. OPTIONAL STEP Further division of positive BNP signal into medium vs. high intensity can be accurately done provided ET1 and BOS treatment of cardiomyocytes has been performed, as follows. CRITICAL STEP Prior culture of cardiomyocytes in serum-containing medium may mask hypertrophic phenotypes [17], so exposure to serum should be minimized or fully eliminated.  E) and uses the cardiac troponin T (cTnT) staining to define cell cytoplasm (F). The cTnT signal intensity is quantified and a positive signal threshold is determined by comparing with cell-type control enabling exclusion of non-cardiomyocytes from the analysis (G). A perinuclear ring is defined in cardiomyocytes (H) and BNP intensity is quantified in this region. Comparison of BNP signal intensity to that of ET1or BOS-treated cardiomyocytes enables distinction between high vs. medium BNP-expressing cells (I-L). BNP-brain natriuretic peptide; cTnT-cardiac troponin T; scale bar = 100 µm.
Insert "Select Population" building block. Population "Cardiomyocytes"; Method "Filter by Property". In the inset menu, insert "Intensity BNP perinuclear" and set it > the value of the bosentan-treated cardiomyocytes ( Figure A2). Output population "BNP positive cardiomyocytes". Confirm if the highlighted cells display positive BNP signal. This signal should always be higher than that of secondary-only or cell-type controls ( Figure 5I). 9.
Insert "Select Population" building block. Population "Cardiomyocytes"; Method "Filter by Property". In the inset menu, insert "Intensity BNP perinuclear" and set it ≤ the value of the bosentan-treated cardiomyocytes (same value as in 8  [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].
CRITICAL STEP Prior culture of cardiomyocytes in serum-containing medium may mask hypertrophic phenotypes [17], so exposure to serum should be minimized or fully eliminated. 10. Insert "Select Population" building block. Population "Cardiomyocytes"; Method "Filter by Property". In the inset menu, insert "Intensity BNP perinuclear" and set it > the value of the ET1-treated cardiomyocytes. Output population "BNP high cardiomyocytes". Confirm if the highlighted cells display positive and high BNP signal ( Figure 5J). 11. Insert "Select Population" building block. Population "Cardiomyocytes"; Method "Filter by Property". In the inset menu, insert "Intensity BNP perinuclear" and set it ≤ the value of the endothelin-treated cardiomyocytes. Add an additional row and set "Intensity BNP perinuclear" to ≥ the value of bosentan-treated cardiomyocytes. Output population "BNP medium cardiomyocytes". Confirm if the highlighted cells display positive BNP signal ( Figure 5K).

Hypertrophy Analysis. Time for Completion: 1 day
3.1.1. Data Acquisition. Time for Completion: 2 h 1. Resuspend freshly dissociated cardiomyocytes in 500 μL Phosphate Buffer Saline (PBS) at 1×10 5 -2.5x10 5 cells per sample and place them on ice prior to flow cytometry analysis. CRITICAL STEP Prior culture of cardiomyocytes in serum-containing medium may mask hypertrophic phenotypes [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].
CRITICAL STEP Defining the threshold for positive BNP signal should be done carefully and according to the controls used. The control where cardiomyocytes were treated with ET1 should typically display >90% BNP-positive cardiomyocytes, so the intensity of BNP signal should be higher in those wells (this should be selected as the BNP-medium/high signal threshold). In contrast, >90% BOS-treated cardiomyocytes should display negative BNP signal, so the signal threshold should distinguish between negative and medium/positive BNP signal ( Figure A2). The threshold for medium/high signal intensity is typically 4 times higher than that of negative/medium signal intensity ( Figure 5L). 12. In the default "Define Results" building block, select Method "List of Outputs" in the drop down menu, then tick the box "Number of Objects" for the options: "Population: Cardiomyocytes", "Population: Nuclei Final", "Population: BNP positive Cardiomyocytes", "Population: BNP negative Cardiomyocytes", "Population: BNP medium Cardiomyocytes", and "Population: BNP high Cardiomyocytes". 13. In the same drop-down menu within "Define Results", define five sections of " Method-Formula Output". In each of these sections in the inset menu, enter "Formula: (a/b)*100. 14. In the first Menu define Variable A: "Cardiomyocytes-Number of Objects", and Variable B: "Nuclei Final-Number of Objects". Output Name: "% Cardiomyocyte Purity". 15. In the second menu: Variable A: "BNP positive cardiomyocytes-Number of Objects", and Variable B "Cardiomyocytes-Number of Objects". Output Name: "%BNP positive cardiomyocytes". 16. In the third menu: Variable A: "BNP high cardiomyocytes-Number of Objects", and Variable B "Cardiomyocytes-Number of Objects". Output Name: "%BNP high cardiomyocytes". 17. In the fourth menu: Variable A: "BNP medium cardiomyocytes-Number of Objects", and Variable B "Cardiomyocytes-Number of Objects". Output Name: "%BNP medium cardiomyocytes". 18. In the fifth menu: Variable A: "BNP negative cardiomyocytes-Number of Objects", and Variable B "Cardiomyocytes-Number of Objects". Output Name: "%BNP negative cardiomyocytes". 19. Check that the percentages pertaining to each Output add up. If not, confirm that the intensity thresholds between building blocks are consistent and that the comparison operators are complementary.
%BNP positive cardiomyocytes + %BNP negative cardiomyocytes = 100% %BNP positive cardiomyocytes = %BNP high cardiomyocytes + %BNP medium cardiomyocytes % BNP high cardiomyocytes + %BNP medium cardiomyocytes + %BNP negative cardiomyocytes = 100% 20. Import purity and %BNP positive values to GraphPad and perform statistical analysis according to the experimental design. 21. Copy the percentages pertaining to each proportion of BNP high, BNP medium, or BNP negative cardiomyocytes cells to a GraphPad "Grouped" Table, by inserting "% high", "% medium", and "% negative" in columns and different cell lines/conditions in rows. 22. Create a "Stacked bars" Graph Type. Colour each bar and perform statistical analysis according to the experimental design ( Figure 8E).

Mitochondrial Respiration and Content
• 1. Resuspend freshly dissociated cardiomyocytes in 500 μL Phosphate Buffer Saline (PBS) at 1×10 5 -2.5x10 5 cells per sample and place them on ice prior to flow cytometry analysis. CRITICAL STEP Prior culture of cardiomyocytes in serum-containing medium may mask hypertrophic phenotypes [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].
CRITICAL STEP Steps 7-10 are performed in order for the Harmony software to detect differences in morphology and texture between cardiomyocytes. Texture information is extracted from each image by using different filtering strategies. The parameters for each approach were optimized for hPSC-CMs and should be adapted for other cardiomyocyte sources. 12. Provide a training set in the option "Training" by picking cells belonging to each class (either organized or disarrayed), in independent fields, wells, or plates. Select only cells that belong to one of these categories clearly ( Figure 6J).

Hypertrophy Analysis. Time for Completion: 1 day
3.1.1. Data Acquisition. Time for Completion: 2 h 1. Resuspend freshly dissociated cardiomyocytes in 500 μL Phosphate Buffer Saline (PBS) at 1×10 5 -2.5x10 5 cells per sample and place them on ice prior to flow cytometry analysis. CRITICAL STEP Prior culture of cardiomyocytes in serum-containing medium may mask hypertrophic phenotypes [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].
CRITICAL STEP Due to the heterogeneity of hPSC-CMs, a large training set of at least 100 cells for each category is desired. Phenotypic controls such as cardiomyocytes with disrupted sarcomeres (genetic knockouts or pharmacologically induced) greatly support this distinction. 13. Evaluate the validity of the training performed by checking that the "Goodness" Parameter is >1, indicating a good separation between the 2 classes, and which texture or morphology properties more accurately define those differences ( Figures A3 and 6K,L).
Methods Protoc. 2019, 2, 83 17 of 26 11. Insert "Select Population" building block. Population: "Cardiomyocytes"; Method "Linear Classifier". In the inset menu, set Number of Classes to 2 and tick all the boxes below. Output Population A: "Organised"; Output Population B: "Disarrayed". 12. Provide a training set in the option "Training" by picking cells belonging to each class (either organized or disarrayed), in independent fields, wells, or plates. Select only cells that belong to one of these categories clearly ( Figure 6J). CRITICAL STEP Due to the heterogeneity of hPSC-CMs, a large training set of at least 100 cells for each category is desired. Phenotypic controls such as cardiomyocytes with disrupted sarcomeres (genetic knockouts or pharmacologically induced) greatly support this distinction. 13. Evaluate the validity of the training performed by checking that the "Goodness" Parameter is >1, indicating a good separation between the 2 classes, and which texture or morphology properties more accurately define those differences ( Figure A3 and 6K,L). The cell mask channel is used to define cell cytoplasm (G) and the α-actinin signal intensity is quantified and compared against that of a cell-type control to exclude non-cardiomyocytes from the analysis (H). Morphologic and texture-based information is extracted from the sharpened α-actinin signal displayed by cardiomyocytes (I), followed by a machine-learning procedure where the user manually selects cardiomyocytes displaying organised vs. disarrayed sarcomeres (J). The software then applies this algorithm to the rest of the images being analysed to distinguish organised vs. disarrayed cardiomyocytes (K,L). Scale bar = 100 μm, except for L where it is 20 μm.
14. In the default "Define Results" building block, select Method "List of Outputs" in the drop down menu, then tick the box "Number of Objects" for the options: "Population: Cardiomyocytes", "Population: Nuclei Final". 15. In the same drop-down menu within "Define Results", define two menus of "Method-Formula Output", and in the inset menu, enter "Formula: (a/b)×100. Figure 6. Sarcomeric disarray analysis sequence. Representative micrographs of image analysis of hPSC-CMs. After inputting the image (A-C), the software uses mathematical transformation algorithms to sharpen sarcomeric signal (D), followed by exclusion of debris and cell clumps (E,F). The cell mask channel is used to define cell cytoplasm (G) and the α-actinin signal intensity is quantified and compared against that of a cell-type control to exclude non-cardiomyocytes from the analysis (H). Morphologic and texture-based information is extracted from the sharpened α-actinin signal displayed by cardiomyocytes (I), followed by a machine-learning procedure where the user manually selects cardiomyocytes displaying organised vs. disarrayed sarcomeres (J). The software then applies this algorithm to the rest of the images being analysed to distinguish organised vs. disarrayed cardiomyocytes (K,L). Scale bar = 100 µm, except for L where it is 20 µm.
14. In the default "Define Results" building block, select Method "List of Outputs" in the drop down menu, then tick the box "Number of Objects" for the options: "Population: Cardiomyocytes", "Population: Nuclei Final". 15. In the same drop-down menu within "Define Results", define two menus of "Method-Formula Output", and in the inset menu, enter "Formula: (a/b)×100. 16. In the first Menu define Variable A: "Cardiomyocytes-Number of Objects"; Variable B "Nuclei Final-Number of Objects". Output Name "% Cardiomyocyte Purity". 17. In the second Menu, define Variable A: "Disarrayed-Number of Objects" and Variable B: "Cardiomyocytes-Number of Objects. Output Name "% Cardiomyocyte Disarray".
18. In the third Menu, define Variable A: "Organized-Number of Objects" and Variable B: "Cardiomyocytes-Number of Objects. Output Name "% Cardiomyocyte Organized". Confirm that % organized + % disarrayed = 100% 19. Import percentages to a GraphPad "Column" Table, create a "Box-plot" graph, and perform statistical analysis according to experimental design ( Figure 8F). Plate cardiomyocytes at approximately 5000 cells/mm 2 (50,000 cells per well).

High-Content Imaging
• Harmony High-Content Imaging and Analysis Software (PerkinElmer, Waltham, MA, USA; cat. No.: HH17000001), with PhenoLOGIC™ machine-learning technology (required only for sarcomeric disarray analysis). • GraphPad Prism v8.2.0 (La Jolla, CA, USA).  [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].

Mitochondrial Respiration and Content
CRITICAL STEP This cell density is aimed at covering the well completely without forming clumps or cell aggregates (crucial for an accurate Seahorse assay and subsequent normalization). It may need to be adjusted to different cardiomyocyte sources (e.g., tissue-derived or resulting from various differentiation protocols that produce cells with varying sizes).  [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].
CRITICAL STEP Prior culture of cardiomyocytes in serum-containing medium may mask hypertrophic phenotypes [17], so exposure to serum should be minimized or fully eliminated. OPTIONAL STEP Allow the cells to settle for 0.5 h at room temperature before placing them in the incubator, to minimize edge effects due to the formation of convection currents because of temperature differences between the medium and the incubator.  [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].
CRITICAL STEP These concentrations have been optimized for hPSC-CMs and may need to be adjusted to different cardiomyocyte sources 8.
Load the injection compounds (25 µL/port) into the cartridge plate ports using the ultrafine 200 µL tips and the corresponding port guides (port A-oligomycin; port B-FCCP; port C-rotenone; port D-XF assay medium).

9.
Initiate Seahorse analyser and Wave software and set up experimental protocol based on 3 basal rate measurements prior to the first injection, followed by 3 measurements after each injection. Mix-Wait-Measure times per measurement are 3 min-0 min-3 min ( Figure 7A). 10. Insert cartridge into the XF analyser for calibration. When prompted by the software, replace utility plate with cell plate, and run the assay to determine oxygen consumption rate (OCR). Save analysis file upon completion.  (Figure 2A,B). 5. Gate to define FSC-A for each bead used for the calibration. Export FSC-A values relative to each gate (corresponding to each bead) to an excel file. Ensure there is consistency between the units being used for beads and samples (i.e., linear or log) ( Figure 2C,D). 6. In the Excel software, plot the FSC-A median values (y axis, obtained from each gate in the Kaluza software) relative to the known bead diameters (x axis). Add trendline of linear regression and display equation ( = × + ) and R-square value determining the calibration curve ( Figure 2E).
CRITICAL STEP If the R-squared value (measure of how much the variation of the FSC-A is explained by variation of the bead size) is below 0.95, then the bead solutions must be thoroughly mixed to disrupt clumps. 7. Gate samples to remove debris and duplets/triples. In cell samples, SSC-width is used in the x axis instead of FSC-A due to higher heterogeneity of the cardiomyocytes relative to the beads ( Figure 2 F-H). 8. Export data from each sample from Kaluza to Excel, as done in 5. , where d is the cell diameter determined in 9. This assumes the approximation of each cell in suspension to a sphere, which is also applicable to the calibration beads. 11. Import the cell volume values into GraphPad software, as a column table (whereby each column represents a separate sample). 12. Choose "violin plot only" in the "Box and plot" options to graphically represent the data distribution and the main statistical parameters (median and quartiles). Adjust colour as desired ( Figure 8A). 13. Perform statistical analysis (depending on the experimental design) to compare volume distribution between samples. OPTIONAL STEP Plotting such large sample sizes in GraphPad can be computationally demanding. Alternatively, a boxplot can be generated by inputting only 7 cardiomyocyte volume values per condition: Minimum, maximum; 25th percentile, 75th percentile, and the median three times. These values can be easily determined using Excel™ software ( Figure  8B).
PAUSE STEP Fixed cells can be stored at 4 • C for up to 2 months with plates wrapped in parafilm, prior to staining and imaging. 12. Image plates in Cellavista plate reader by optimizing Hoechst signal intensity and focus in a single field followed by scanning the whole plate across the entire well ( Figure 7B). 13. Quantify number of cell nuclei in the "Evaluation" mode. Export data to an Excel file.  [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].
CRITICAL STEP Other normalization techniques based on protein content should be avoided as a hypertrophic phenotype characteristic of HCM is typically associated with increased protein/DNA ratio [38]. OPTIONAL STEP Using Cellavista ensures high-throughput quantification of cell nuclei but other instruments (automated or not) can be used for the same purpose.  (to enable comparisons between experimental replicates and with the literature) ( Figure 7C). OPTIONAL STEP Alternatively, input OCR and cell number values into the Excel Macro provided by Agilent for the analysis (https://www.agilent.com/en/products/cell-analysis/cellanalysis-software/data-analysis/seahorse-xf-cell-mito-stress-test-report-generators). 16. Calculate parameters of mitochondrial respiration as follows ( Figure 7C):  17. Import data into GraphPad as an "XY Table". Plot and amend the colour of the lines/bar graphs according to the experimental design and indicate in the graph when compounds were added ( Figure 8G-J [17], so exposure to serum should be minimized or fully eliminated. 2. Add 2 drops of bead solution of each size to a flow cytometer tube, mix well by flicking it, and measure FSC and side scatter (SSC) values (area, width, and height) at 488 nm laser wavelength (area and height). CRITICAL STEP Use of 100 μM flow cytometer nozzle size is highly recommended to minimize sampling biases [33].
CRITICAL STEP Reverse-pipetting is to ensure high accuracy of pipetting technique, due to the very high sensitivity of the qPCR technique. 8.
Add 1 µL of DNA template to each respective well of the 96-well plate. 9.
Stick the MicroAmp™ Optical Adhesive Film to the 96-well plate. 10. Initiate the 7500 Real-Time PCR equipment and its respective software. Define experimental setup as: Instrument "7500 Fast (96 Wells); Type of Experiment "Quantitative -Comparative C T (∆∆C T ); Reagents "TaqMan ® Reagents"; Ramp Speed "Standard (~2 h to complete a run). 11. Enter plate setup according to experimental design. Set Run Method as:

Expected Results
HCM is a complex and intractable disease that requires further characterization to facilitate therapeutic intervention [2]. Several cellular and molecular mechanisms of disease have been described [38], but they have shown to be mutation-specific and/or restricted to particular patient backgrounds [10]. Thus, uncovering novel molecular mechanisms of HCM progression in cellular models where the genetic causation has been elucidated is a promising strategy to address the lack of effective treatment in order to restore cardiac function [4,40].
So as to address the complex in vitro characterization of HCM, we developed a set of phenotypic assays that can be used to quantify disease hallmarks. Our toolkit consists of three straightforward protocols with varying degrees of logistic requirement (e.g., equipment and cost) that are not only accessible to any research groups investigating HCM, but also applicable to other fields evaluating identical changes in cellular responses. The data generated from this characterization encompass analysis of (i) hypertrophy, (ii) HCM molecular features, and (iii) mitochondrial respiration and content ( Figure 8).
While cardiomyocytes cultured in 2D conditions do not fully replicate pathophysiological disease features [29,41], the information acquired from such a multi-parametric evaluation produces a significant level of understanding of HCM characteristics with directly quantifiable variables. Remarkably, these can be harnessed in high-throughput drug screening approaches aimed at identifying new compounds with therapeutic potential, which are integrated in early stages of drug discovery pipelines where fast and inexpensive assays are paramount [6]. In this regard, this toolkit can be used to detect attenuation or rescue of HCM phenotypes upon pharmacological of genetic intervention. This is useful to exclude drug candidates early on, before more complex, lower-throughput, and often time-consuming and labour-intensive assays are required [5]. This greatly expedites subsequent functional assaying of cardiomyocytes (e.g., investigating calcium handling and contractile force), typically performed in 3D organoids or engineered heart tissues [26][27][28][42][43][44].
Overall, it is expected that disseminating this toolkit will greatly support disease modelling research fields, accelerating the discovery of new mechanisms of disease and ultimately paving the way for more efficient therapeutics. showing (K) similar mitochondrial DNA content (determined by ratiometric mito/nuclear DNA qPCR analysis). Data represent mean +/-SEM of five independent experiments comparing R453C-β-myosin heavy-chain hPSC-CMs (HCM) to isogenic wild-type control (healthy), with 100,000-250,000 cells analysed per sample. Statistical analysis was performed by unpaired parametric Student's t test between HCM and Healthy conditions, with the exception of Panel E, where unpaired one-way ANOVA test was used with Dunnett's post hoc test for correction of multiple comparisons relative to Healthy condition (n. s., non-significant; *P < 0.05; **P < 0.01; ***P < 0.005; ****P < 0.0001, colour-coded by inter-compared category-black asterisks apply to all categories). Figure A1. Cell-type control for α-actinin staining. Comparing signal intensities of α-actinin between fibroblasts or any cell type that does not express the cardiac marker (cell-type control) with cardiomyocytes enables determination of positive signal threshold required for high-content imaging. Scale bar = 100 μm. Figure A2. Controls for BNP signal thresholding. In order to categorize signal intensities of cardiomyocytes expressing BNP, endothelin-1 treatment can be used to determine high-intensity threshold as it maximises BNP expression relative to untreated cells, whereas bosentan treatment has the opposite effect (defining low/nonexistent signal). Scale bar = 100 μm. Figure A1. Cell-type control for α-actinin staining. Comparing signal intensities of α-actinin between fibroblasts or any cell type that does not express the cardiac marker (cell-type control) with cardiomyocytes enables determination of positive signal threshold required for high-content imaging. Scale bar = 100 µm.
Methods Protoc. 2019, 2, 83 23 of 26 Figure A1. Cell-type control for α-actinin staining. Comparing signal intensities of α-actinin between fibroblasts or any cell type that does not express the cardiac marker (cell-type control) with cardiomyocytes enables determination of positive signal threshold required for high-content imaging. Scale bar = 100 μm. Figure A2. Controls for BNP signal thresholding. In order to categorize signal intensities of cardiomyocytes expressing BNP, endothelin-1 treatment can be used to determine high-intensity threshold as it maximises BNP expression relative to untreated cells, whereas bosentan treatment has the opposite effect (defining low/nonexistent signal). Scale bar = 100 μm. Figure A2. Controls for BNP signal thresholding. In order to categorize signal intensities of cardiomyocytes expressing BNP, endothelin-1 treatment can be used to determine high-intensity threshold as it maximises BNP expression relative to untreated cells, whereas bosentan treatment has the opposite effect (defining low/nonexistent signal). Scale bar = 100 µm. Figure A3. Parameters for machine learning step in the sarcomeric disarray analysis sequence. The PhenoLOGIC™ training relies on manual assignment of cardiomyocytes to two different categories (A, organised; B, disarrayed). The parameters of algorithm using the training set provided by the user indicate goodness of fit factor (>1) as well as the texture/ morphological properties evaluated by the software that differ the most between the two categories. Figure A3. Parameters for machine learning step in the sarcomeric disarray analysis sequence. The PhenoLOGIC™ training relies on manual assignment of cardiomyocytes to two different categories (A, organised; B, disarrayed). The parameters of algorithm using the training set provided by the user indicate goodness of fit factor (>1) as well as the texture/ morphological properties evaluated by the software that differ the most between the two categories.