Small particle fluorescence and light scatter calibration using FCM_PASS software.

Abstract

The use of flow cytometry to analyze small particles has been implemented for several decades. More recently small particle analysis has become increasingly utilized owing to the increased sensitivity of conventional and commercially available flow cytometers along with growing interest in other small particles, such as extracellular vesicles. Despite an increase in small particle flow cytometry utilization, the lack of standardization in data reporting has resulted in a growing body of literature regarding extracellular vesicles (EVs) that cannot be easily interpreted, validated, or reproduced. Methods for fluorescence and light scatter standardization are well-established and the reagents to perform these analyses are commercially available. Here we outline FCM_PASS as software package for performing fluorescence and light scatter calibration of small particles while generating standard reports conforming to the MIFlowCyt-EV standard reporting framework. This unit shall cover the workflow of implementing calibration using FCM_PASS as follows: acquisition of fluorescence and light scatter calibration materials, cataloguing the reference materials for use in the software, creating cytometer databases and datasets to associate calibration data and fcs files, importing fcs files for calibration, inputting fluorescence calibration parameter, inputting light scatter calibration parameter, and applying the calibration to fcs files.

INTRODUCTION:

The FCM_PASS software was developed to perform calibration and keep records on multiple instrument platforms. The workflow is intended to be ergonomic for flow cytometry core facilities and researchers with multiple instrument interested in tracking their instrument performance longitudinally and keeping records of their experimental calibration. To enable this, the FCM_PASS software is centered around keeping databases for each cytometer with a dataset added for each calibration (Protocol 5 & 6). For ergonomic use and to maintain high quality reporting outputs for publication, the FCM_PASS utilizes a cataloguing system for reference materials (Protocol 3 & 4). The reference material catalogue enables users to only be required to input complete records on their calibration reference materials once. With this information stored, it can be reference when performing future calibrations and enables completing documenting in the automated calibration reports.

FCM_PASS has been developed to perform fluorescence calibration for any number of fluorescence parameters, with the ability for the user to alter the method of unit scaling for regression, along with defining fluorophore to protein ratios for converting the outputs from molecules of equivalent soluble fluorophore (MESF) to epitope number (Protocol 7). This unit will cover acquisition and gating of fluorescent calibration materials for calibration of sample data (Protocol 2) or cross-calibration of other fluorescence reference materials (Alternative Protocol 2).

Light scatter calibration using the FCM_PASS software is a semi-automated process, whereby the software approximates the collection half-angle to produce a scatter diameter curve (Protocol 8). This unit will cover acquisition and gating of light scatter calibration materials for calibration of sample data (Protocol 1). By default, the software outputs core-shell models based on extracellular vesicle light scattering characteristics. The software does, however, allow users to customize core-shell model outputs along with homogeneous sphere outputs.

Finally, once calibration is performed all calibrated parameters are written to .fcs files which preserve the meta-data from source file (Protocol 9). This allows users to share their data in calibrated units. Upon calibration quality control plots for both light scatter and fluorescence are outputted along with a spreadsheet. The spreadsheet contains a MIFlowCyt-EV report with the fields for calibration automatically completed by the software (Welsh, Van Der Pol, et al., 2020). This spreadsheet also contains the limits of detection for files that used a trigger threshold with a calibrated parameter. All meta-data associated with the fluorescence and light scatter calibration to allow reproducibility is also shared.

BASIC PROTOCOL 1: Acquisition and gating of light scatter calibration materials

This protocol will outline how to prepare, analyze, and gate light scatter calibration materials for use downstream in the FCM_PASS software in order to perform light scatter calibration. The correct use of light scatter reference materials and instrument settings will allow a good approximation of instrument collection half-angle using Mie theory. By approximating the collection half-angle, light scatter calibration can be performed allowing for the determination of instrument sensitivity in standard units if a light scatter trigger threshold is used.

Materials:

Reagents and solutions

DPBS, no calcium, no magnesium (Thermo Fisher Scientific, Cat No. 14190250)

NIST-traceable polystyrene size standards (Thermo Fisher Scientific, 3000 series – 3100A, 3150A, 3200A, 3269A, 3300A, 3350A, 3400A, 3450A, 3500A, 3600A)

NIST-traceable silica size standards (Thermo Fisher Scientific, 8000 series – 8050A, 8070A)

Hardware

Round-Bottom Polystyrene Test Tubes (Falcon, Cat 14–959-2A)

Vortex

Flow cytometer

Protocol steps:

1. Calculate the stock traceable size calibration reference bead particle concentration using percent solids value and particle density provided by the manufacturer and the following formula in equation 1, where ${\mathrm{N}}_{\mathrm{P}}$ is the concentration (particles mL⁻¹), ${\mathrm{W}}_{\mathrm{V}}\mathrm{\%}$, is the percent solids, ${\mathrm{\rho }}_{\mathrm{\rho }}$ is the particle density (g mL⁻¹), and $\mathrm{D}$ is the average diameter (μm).

   $$N_P=\frac{W_V\mathrm{\%},\mathrm{ }·6\times {10}^{12}}{\pi {\rho }_{\rho }{D}^3}$$ Equation 1
   For example, 100 nm polystyrene beads at 1% with 1.05 g mL⁻¹ would be calculated using equation 2:

   $$1.82\times {10}^{13}=\frac{1\mathrm{ }·6\times {10}^{10}}{\pi \times 1.05\mathrm{ }x\mathrm{ }{0.1}^3}$$ Equation 2

2. Thoroughly vortex the traceable size calibration reference bead stock bottles to homogenize the mixtures before dispensing 1 drop (~50 μL) into separate 500 μL low-protein binding Eppendorf.

3. Using the working stock from step 2, make up 500 μL solution at 1×10⁷ particles mL^-1.

   Note: It is recommended that serial dilutions are used and volumes of no less than 10 μL to avoid pipetting errors. The optimal particle concentration at which to run the reference materials will vary depending on several factors, including the flow rate, beam height, and electronic sampling rate. If running for the first time, it is recommended that serial dilutions are performed to determine the optimal concentration for preparation of the beads.

4. On the flow cytometer, set the triggering threshold to the most sensitive light scatter detector and ensure the parameter is using log-scaling (not linear or biexponential).

5. Running DPBS, lower the triggering threshold until the noise floor of the instrument becomes visible. This is most clearly when using a histogram (see Figure 1).

   1. Plotting the trigger-channel height parameter against time and monitoring while running DPBS is a good indication for determining whether an instrument is clean. If the spread of noise (and event rate) decreases over time, it is indicative that the instrument was dirty and is becoming cleaner.
   2. The extent to which the opto-electronic noise of an instrument can be sampled will vary between instruments. Legacy flow cytometers will tolerate a couple of 1000–2000 events/second whilst allowing room to sample desired events, while high-speed jet-in-air sorters are capable of sample 10,000+ events per second.
   3. Triggering using a light scatter parameter on the opto-electronic noise of the instrument has benefits in determining and tracking the lower limit of detection, as well as being informative for buffer + reagent controls where background fluorescence will show clear shifts due to many events being triggered from sampling the noise. The use of this method comes at the cost of having high event rates and therefore larger files. Before utilizing this method the instrument should be validated to determine: 1) its ability to detect and accurately process particles, 2) the event rate at which single dim particles are detected, and 3) the degree to which the opto-electronic noise can be sampled without creating artefacts or reducing the ability to detect genuine events.
   4. On some instruments that utilize peristaltic pumps there can appear to be an increase and decrease of the baseline corresponding to the turnover of the pump. This is a result of the threshold being set close to (but above) the electronic noise, resulting in the increase and decrease in trigger events in light scatter. This can be overcome by lowering the threshold so that the noise is being sampled regardless of the peristaltic pump turnover or increasing the threshold and therefore decreasing the instrument’s limit of sensitivity.
6. Analyze each bead sample at the same acquisition settings until >5000 bead events are recorded.

   1. It is preferable to analyze and store bead populations individually. This will minimize population overlap, aggregates, background noise, and artifacts (see Figure 2).

7. Gate each bead population using the parameter Height vs. Area in a dot-plot to remove doublets/aggregates and then use a histogram on the light scatter parameter (Height) to obtain statistics for each population. The light scatter parameter should use log scaling (see Figure 1 & 2).

8. Obtain the median statistic for each of the bead populations (see Figure 1).

   1. By default, flow cytometers trigger the acquisition of an event using the pulse height parameter. In cases where a trigger threshold is being defined (e.g. SSC), it is recommended that the pulse-height is used so that the limit of detection can be defined in calibrated units. There is no consensus within the small particle community over the use of pulse height vs. area. We recommend that, in general, if the parameter being calibrated was not used as a trigger channel the pulse area statistic should be used due to the tendency for low signal intensities to be linear and therefore a more reliable method for extrapolation.

Figure 1 -. Gating light scatter reference beads.

Each panel shows the gating of polystyrene NIST-traceable reference beads ranging in mean diameter from 100 to 600 nm. The median light scatter statistic of the gated population is given in each panel.

Figure 2 -. Analyzing light scatter reference beads.

A) demonstrates the cumulative distribution of the gated populations from Figure 1. when mixed together. While some populations are clearly distinguished some are not. The areas where bunching of populations occurs is dependent upon the cytometer and is useful in determining the collection angle. B) illustrates overlaid and colored gated bead population from Figure 1.

BASIC PROTOCOL 2: Acquisition and gating of fluorescence calibration materials

This protocol will outline how to prepare, analyze, and gate fluorescence calibration materials for use downstream in the FCM_PASS software in order to perform fluorescence calibration. The correct use of fluorescence reference materials and instrument settings will allow a good approximation of instrument of instrument sensitivity in standard units.

Materials:

Reagents and solutions

DPBS, no calcium, no magnesium (Thermo Fisher Scientific, Cat No. 14190250)

Molecules of equivalent soluble fluorophore calibration beads (Bangs Laboratories, Cat 823)

Hardware

Round-Bottom Polystyrene Test Tubes (Falcon, Cat 14–959-2A)

Vortex

Flow cytometer

Protocol steps:

1. Vortex each fluorescence reference bead bottle before use.

2. Add 1 drop (~50 μL) of each bead population to separate FACS tubes containing 250 μL of DPBS.

   1. Due to the high autofluorescence of the ‘Blank’ beads, their use is not recommended to use as 0 MESF.
   2. Many commercially available fluorescence calibration beads are bright and will require extrapolation instead of interpolation to obtain the dim fluorescence values. The accuracy of the extrapolation will therefore be influenced by several factors including the gating of the populations. While less ergonomic, it is preferable to analyze 1 bead population at a time. This allows for gating on scatter parameters, rather than fluorescence parameters, making the statistics less biased by the gating strategy. Analyzing one bead population at a time will also minimize the subjectivity when gating fluorescence populations that overlap, sometimes causing small peaks, Figure 3.

3. Ensure cytometer fluorescence settings are those used for the assay being calibrated.

4. If the beads are >1 μm in diameter use a forward-scatter trigger threshold.

5. Analyze each beadreap sample at the same acquisition settings until >5000 bead events are recorded.

6. Gate each bead population on FSC-A vs. SSC-A and obtain the median area statistic for the fluorescence parameter being calibrated to move on to the fluorescence calibration protocol (Protocol 7).

   1. By default, flow cytometers trigger the acquisition of an event using the pulse height parameter. In cases where a trigger threshold is being defined e.g. SSC. It is recommended that the pulse-height is used so that the limit of detection can be defined in calibrated units. There is no consensus within the small particle community over the use of pulse height vs. area. We recommend that, in general, if the parameter being calibrated was not used as a trigger channel the pulse area statistic should be used due to it tending to be linear at low signal intensities and therefore a more reliable method for extrapolation.

Figure 3 -. Gating fluorescence reference beads.

A) Gating of bead population using FSC-A and SSC-A. B) Histogram of all four APC MESF bead population in a cumulative distribution. Arrows highlight areas of overlap between beads that may lead to subjectivity on where to manually draw gates. C) Histogram of individual APC MESF bead populations.

ALTERNATE PROTOCOL 1: Cross-calibration of fluorescence reference materials

A more economical method for calibrating fluorescence parameters over time and using high detector settings that many cause some MESF standards to be off scale is to cross-calibrate 8-peak rainbow beads. These beads fluorescence over a wide region of the visible spectrum and range in fluorescence intensity from very bright to very dim. Typically, the dimmest population of an 8-peaks sample has a lower MESF value than dimmest population within MESF calibration beads on the same channel.

Materials:

Reagents and solutions

DPBS, no calcium, no magnesium (Thermo Fisher Scientific, Cat No. 14190250)

APC molecules of equivalent soluble fluorophore calibration beads (Bangs Laboratories, Cat 823)

8-peak rainbow beads (Spherotech, Cat RCP-30–5A)

Hardware

Round-Bottom Polystyrene Test Tubes (Falcon, Cat 14–959-2A)

Vortex

Flow cytometer

Protocol steps

1. Vortex each fluorescence reference bead bottle before use.

2. Add 1 drop (~50 μL) of each MESF bead population to separate FACS tubes containing 250 μL of DPBS.

3. Add 1 drop (~50 μL) of the 8-peak bead population to separate FACS tubes containing 250 μL of DPBS.

4. Ensure cytometer fluorescence settings are those used for assays being calibrated.

5. If the beads are >1 μm in diameter use a forward-scatter trigger threshold.

6. Analyze each bead sample at the same acquisition settings until >2000 bead events are recorded. For the 8-peak beads this will be >16,000 events.

7. Gate each MESF bead population on FSC-A vs. SSC-A and obtain the median statistic for the parameter and perform calibration of the MESF and 8-peak bead files using Protocols 5, 6, 7, 9.

8. Once the 8-peak rainbow beads are calibrated in PE MESF units, gate the population on FSC-A vs. SSC-A to obtain singlets, (see Figure 4). Using the singlet population gate each of the 8-peak populations.

   1. The gating the individual fluorescent bead populations can be done in the parameter which best separates each population. This may be a different fluorescence detector than the calibrated parameter. While the gating of each population does not have to be on the MESF parameter itself, the MESF parameter should be checked to ensure all populations are on scale. In some 3^rd party software the scale limits (minimum and maximum value) will influence the outputted statistic due to how the data is binned.

9. Once each of the 8-peak populations has been gated, see Figure 4, obtain the median MESF value for each of the populations. These values are now the cross-calibrated values for these beads and can be used on the same instrument at different gains.

Figure 4 -. Gating and cross-calibration of fluorescence reference beads.

A) Gating of 8-peak bead population using FSC-A and SSC-A. B) Regression of PE MESF bead reference values vs. acquired arbitrary statistics for each population. C) Histogram of gated (Figure 4A) 8-peak reference beads and gating of each population (red). D) Histogram of gated (Figure 4A) 8-peak reference beads converted to PE MESF units using regression (Figure 4B). E) Histogram of gated PE MESF beads and gating of each population (red). F) Histogram of gated PE MESF beads converted to PE MESF units using regression (Figure 4B).

Note: If an instrument is re-aligned or filters are changed these values will no longer be valid and will require cross-calibration to be performed again. In general, it is good practice to regularly cross-calibrate 8-peak bead reference values, e.g. once a month.

BASIC PROTOCOL 3: Cataloguing light scatter calibration materials

This protocol outlines how to catalogue light scatter calibration reference materials for use in the FCM_PASS software. By cataloguing each reagent, including relevant meta-data such as manufacturer, catalogue numbers and lot numbers, the automated output report can include all relevant information for reporting. By calling upon a catalogue during the calibration input steps within the software, time is saved by the user not having to repeat inputting the same information.

Materials:

Reagents and solutions

NIST-traceable polystyrene size standards (Thermo Fisher Scientific, 3000 series – 3100A, 3150A, 3200A, 3269A, 3300A, 3350A, 3400A, 3450A, 3500A, 3600A)

NIST-traceable silica size standards (Thermo Fisher Scientific, 8000 series – 8050A, 8070A)

Hardware

Computer (Quad-core, 8GB+ RAM, recommended)

FCM_PASS v3+ software (Windows, MacOS compatible, available at: https://nano.ccr.cancer.gov/fcmpass)

Protocol steps

1. Open FCM_PASS software

2. Click ‘Catalogue’ in the top menu bar

3. Under the ‘Light Scatter’ tab entry fields exist for each of the pertinent metadata for reporting with light scatter calibration (Figure 5).

   • a
     Diameter is the mean diameter of the solid bead population provided on the certificate of analysis.
   • b
     Diameter CV is the percent coefficient of variation of the mean diameter provided on the certificate of analysis
   • c
     Refractive Index is the provided refractive index of the bead population on certificate of analysis
   Note: If a refractive index is not available, an approximate guide for polystyrene refractive index is 1.59 at 589 nm. Silica can vary more in refractive index than polystyrene but tends to be ~1.45 at 589 nm.

   • d
     ‘RI Measurement Wavelength’ is the wavelength at which the refractive index was measured and should be provided on the certificate of analysis. This is often 589 nm.
   • e
     Composition can be selected as polystyrene, silica, or other. If polystyrene or silica are selected, changes in detection wavelength e.g. 488 nm to 405 nm are accounted for using the appropriate Sellmeier equations. If ‘Other’ is selected, then the refractive index change is made propositionally to the sheath refractive index.
   • f
     Manufacturer, Catalogue Number, and Lot Number should all be completed appropriately.

4. Once the fields have been completed for a bead population, click ‘Add Bead’. The population should then appear in the table below.

5. Once the relevant beads have been added ‘Bead Sets’ can be created. ‘Bead Sets’ are the bead populations that are used for calibration. Any number of ‘Bead Sets’ and combinations can be made.

   1. In the ‘Selection’ column of the table, check all the bead populations to be included within a bead set.
   2. Click ‘Create Set’, enter a unique Set name, and click ‘OK’.

6. Once your bead set has been defined you will be able to perform light scatter calibration.

   Note: Updates and videos related to this protocol can be found at: dx.doi.org/10.17504/protocols.io.bjcqkivw

Figure 5 -.

Light scatter reference bead catalogue input window.

BASIC PROTOCOL 4: Cataloguing fluorescence calibration materials

This protocol outlines how to catalogue fluorescence calibration reference materials for use in the FCM_PASS software. By cataloguing each reagent, including relevant meta-data such as manufacturer, catalogue numbers and lot numbers, the automated output report can include all relevant information for reporting. By calling upon a catalogue during the calibration input steps within the software, time is saved by the user not having to repeat inputting the same information.

Materials:

Reagents and solutions

FITC molecules of equivalent soluble fluorophore calibration beads (Bangs Laboratories, Cat 555A)

PE molecules of equivalent soluble fluorophore calibration beads (Becton Dickinson, Cat 340495)

Hardware

Computer (Quad-core, 8GB+ RAM, recommended)

FCM_PASS v3+ software (Windows, MacOS compatible, available at: https://nano.ccr.cancer.gov/fcmpass)

Protocol steps

1. Open FCM_PASS software

2. Click ‘Catalogue’ in the top menu bar

3. Under the ‘Fluorescence’ tab entry fields exist for each of the pertinent metadata for reporting with fluorescence calibration, (see Figure 6).

   1. Enter the name of the fluorophore on the fluorescence reference beads.
   2. Enter the manufacturer, catalogue number, and lot number fields appropriately.
   3. In the ‘Bead Ref Values’ field enter each fluorescence beads reference values. This may be in molecules of equivalent soluble fluorophore, equivalent reference fluorophore, or antibody binding capacity.
   4. After each reference value click the ‘+’ button.
   5. Once all fields and reference values have been added click ‘Create Set’. The beads will then appear on the table below and will be available for selection when performing fluorescence calibration.

   Note: Updates and videos related to this protocol can be found at: dx.doi.org/10.17504/protocols.io.bjcqkivw

Figure 6 -.

Fluorescence reference bead catalogue input window.

BASIC PROTOCOL 5: Creating cytometer databases and datasets

This protocol outlines how to create a cytometer database and datasets. Storing the calibration information by dataset for each respective allows an ergonomic means for users who need to perform calibration and track multiple instruments simultaneously.

Materials:

Hardware

Computer (Quad-core, 8GB+ RAM, recommended)

FCM_PASS v3+ software (Windows, MacOS compatible, available at: https://nano.ccr.cancer.gov/fcmpass)

Protocol steps

1. Open FCM_PASS software (see Figure 7)

2. Click the ‘+’ icon next to ‘Cytometer IDs’ list and enter a unique name to identify an instrument.

3. Select the relevant cytometer ID to add the dataset to

4. Click the ‘+’ icon next to the ‘Datasets’ list.

   1. In the window enter the acquisition date of the calibration data and the dataset/experiment name. If there are any notes related to the experiment that are beneficial, they can be entered in the ‘Dataset Notes’ field.

5. Proceed to Protocol 6.

   Note: Updates and videos related to this protocol can be found at: dx.doi.org/10.17504/protocols.io.bjcqkivw

Figure 7 -. Main window of FCM_PASS v3.

Components include cytometer database selection (top left) and cytometer dataset selection (middle left), loaded dataset metadata (right).

BASIC PROTOCOL 6: Importing fcs files

This protocol outlines how to import fcs files into the FCM_PASS software for calibration. Importing the fcs files into the software is the first step in beginning calibration for a dataset. The software at this step can extract the parameters associated with the .fcs file along with the instrument settings used to acquire the data associated with the fcs file.

Materials:

Hardware

Computer (Quad-core, 8GB+ RAM, recommended)

FCM_PASS v3+ software (Windows, MacOS compatible, available at: https://nano.ccr.cancer.gov/fcmpass)

Protocol steps

1. Once a dataset has been created click the ‘Begin Calibration’ button.

2. Import fcs files by selecting the ‘+’ icon next to the ‘Files to calibrate’ table (Figure 8).

   1. In the new window navigate to the folder containing the fcs files you wish to calibrate and select ‘OK’.
3. The fcs files and related metadata will now be imported.

   1. If the folder contains fcs files that you do not wish to be calibrated, select them and click the ‘−’ icon. The metadata related to the remaining files will then be reprocessed.
   2. The parameters e.g. SSC-H, SSC-A that are available in further steps of the software are those that are common to all the loaded fcs files. If files that are loaded do not have any common parameter names a selection will not be available in these steps.

4. Under the ‘Sample Type’ column all loaded files by default are listed as ‘Sample’. For the relevant files these can be adjusted to ‘SSC Calibration’ or ‘FL Calibration’ depending on what the sample was used for.

5. Once completed select ‘Next’.

6. Proceed to Protocol 7.

   Note: Updates and videos related to this protocol can be found at: dx.doi.org/10.17504/protocols.io.bjcqkivw

Figure 8 -.

fcs file import window.

BASIC PROTOCOL TITLE: Fluorescence calibration

This protocol outlines how to input the parameters required to all the FCM_PASS software to perform fluorescence calibration. On the imported fcs files. The use of fluorescence calibration allows uses to report their data in standard units as well as determine their instrument’s limit of sensitivity and dynamic range. The use of fluorescence calibration can also allow users to determine the best instruments and instrument settings for their assay.

Materials:

Hardware

Computer (Quad-core, 8GB+ RAM, recommended)

FCM_PASS v3+ software (Windows, MacOS compatible, available at: https://nano.ccr.cancer.gov/fcmpass)

Fcs file gating software e.g. FlowJo, LLC.

Protocol steps

1. If fluorescence calibration is being performed click the ‘+’ button to add a calibration parameter to the table (see Figure 9). If fluorescence calibration is not required, click ‘Next’.

   1. If you have not yet added the MESF reference bead information that will be used for calibration into the Catalogue’, click ‘Catalogue’ in the top menu bar and complete Protocol 4.

2. Once a parameter is added, double click the ‘Reference Fluorophore’ item and select the bead set used for calibration. The displayed sets are those that have previously been added to the Catalogue in Protocol 4

3. Double click the parameter to select the associated parameter with the correct fluorophore.

4. Double click the relevant cell in the ‘New Parameter Name’ column to adjust how the calibrated parameter’s name will appear once written to the fcs file.

5. The reference bead values for the selected parameter should appear in the ‘Regression Values’ table.

6. Click in the ‘Acquired Value’ box next to each bead reference value and input the acquired statistic

7. Repeat steps 1 to 5 for any further parameters that need to be calibrated. To change the ‘Ref Value’ table to other fluorophores select them in the reference ‘Fluorescence Calibration Parameters’ table.

8. Once completed click ‘Next’.

   Note: The regression plots for the inputted fluorescence calibration parameters can be checked at any time using the ‘Check Regression(s)’ button. The ‘Advanced Settings’ button can be used to specify a fluorophore to protein (F:P) ratio or alter the regression method between linear, log, weighted linear, weighted log.

9. Proceed to Protocol 8.

   Note: Updates and videos related to this protocol can be found at: dx.doi.org/10.17504/protocols.io.bjcqkivw

Figure 9 -.

Fluorescence calibration window.

BASIC PROTOCOL 8: Light scatter calibration

This protocol outlines how to input the parameters required to all the FCM_PASS software to perform light scatter calibration. On the imported fcs files. The use of light scatter calibration allows uses to report their data in standard units as well as determine their instrument’s limit of sensitivity and dynamic range. The use of light scatter calibration can also allow users to determine the best instruments and instrument settings for their assay.

Materials:

Hardware

Computer (Quad-core, 8GB+ RAM, recommended)

FCM_PASS v3+ software (Windows, MacOS compatible, available at: https://nano.ccr.cancer.gov/fcmpass)

Fcs file gating software e.g. FlowJo, LLC.

Protocol steps

1. If light scatter calibration is being performed click the ‘+’ button to add a calibration parameter to the table (see Figure 10). If light scatter calibration is not required, click ‘Next’.

   1. If you have not yet defined the light scatter bead sets in Catalogue’, click ‘Catalogue’ and complete Protocol 3.

2. Double click the ‘Scatter Parameter’ field to change which parameter is being used for light scatter calibration.

3. Alter the ‘Scatter Wavelength (nm)’ to the relevant wavelength for the parameter being used to calibrate light scatter.

   Note: You will see that the ‘Sheath RI’ field will automatically update when this is altered. In the background reference bead, core-shell model, and homogenous sphere model refractive indices will all also be updated.

4. If the selected ‘Scatter Parameter’ was used as a triggering threshold then the ‘Scatter Threshold’ field will automatically update to show the values used as thresholds in the .fcs files loaded. Select a ‘Scatter Threshold’ by double clicking the field and selecting and option from the dropdown menu. A custom entry can also be inputted.

5. Load the light scatter reference beads used by double clicking the ‘Bead Set’ field. Once loaded the beads within the set will populate the bottom table.

6. The ‘Sheath RI’ field automatically accounts for ‘Scatter Wavelength’ but can be updated manually by double clicking the field.

7. In the bottom table enter the median scatter parameter statistic for each population. The acquired CV can optionally also be completed, its use will, however, only be used for plotting purposes and not alter the model calculations.

8. Once complete click ‘Next’.

   Note: Custom core-shell models, solid sphere models, plot data points, modelling parameters, and output settings can be entered or altered by clicking the ‘Advanced Settings’ button. By default, three EV core-shell models relating to high, medium, and low EV refractive indices are calculated. All core-shell models assume a 5 nm shell thickness.

9. Proceed to Protocol 9.

   Note: Updates and videos related to this protocol can be found at: dx.doi.org/10.17504/protocols.io.bjcqkivw

Figure 10 -.

Light scatter calibration window.

BASIC PROTOCOL 9: Performing and reporting fcs file calibration

This protocol outlines how to input the parameters required to all the FCM_PASS software to perform light scatter calibration. On the imported fcs files. The use of light scatter calibration allows uses to report their data in standard units as well as determine their instrument’s limit of sensitivity and dynamic range. The use of light scatter calibration can also allow users to determine the best instruments and instrument settings for their assay.

Materials:

Hardware

Computer (Quad-core, 8GB+ RAM, recommended)

FCM_PASS v3+ software (Windows, MacOS compatible, available at: https://nano.ccr.cancer.gov/fcmpass)

Protocol steps

1. Upon completing Protocol 7 and/or Protocol 8 click the ‘Calibrate’ button.

   Note: If this is the first time using a particle composition, whether it be for a light scatter calibration reference material or a new homogenous sphere or core-shell model, the software will need to generate a database for that composition. The processing time for this can varying depending upon computing processor speed and the number of processor cores. Generally, this will take 5–10 minutes (2Ghz, 4-core processor). Any subsequent calibration using the same light scatter reference materials or custom models will be loaded instantly.

2. The FCM_PASS software will perform fluorescence and light scatter calibration. An FCM_PASS export folder will be created in the directory from which the fcs files were imported. This folder will contain calibrated fcs files, a MIFlowCyt-EV report with fields relevant to fluorescence and light scatter calibration completed and supplementary sheets for reproducing each of the calibrations. A calibration output report file will also be generated that contains the relevant figures to support the fluorescence calibration and light scatter calibration that was performed. For complete and transparent records, all of these files should be kept together when fcs files are shared.

3. The remaining fields within the MIFlowCyt-EV report should be completed as recommended in the associated position paper (Welsh, Van Der Pol, et al., 2020).

   Note: Updates and videos related to this protocol can be found at: dx.doi.org/10.17504/protocols.io.bjcqkivw

REAGENTS AND SOLUTIONS:

The following reagents have been successfully tested by the authors for use as light scatter and fluorescence calibration reagents in the protocols outlined.

60 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

70 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

81 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

100 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

152 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

203 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

240 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

269 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

303 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

345 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

401 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

453 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

508 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

600 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

707 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

1019 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

1587 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 3060)

490 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 8050)

730 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 8070)

990 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 8100)

1570 nm NIST-traceable polystyrene size standards (Thermo Fisher Scientific, Cat. 8150)

R-PE molecules of equivalent soluble fluorophore calibration beads (Becton Dickinson, Cat 340495)

R-PE molecules of equivalent soluble fluorophore calibration beads (Bangs Laboratories, Cat 827)

Alexa Fluor 488 molecules of equivalent soluble fluorophore calibration beads (Bangs Laboratories, Cat 488A)

FITC molecules of equivalent soluble fluorophore calibration beads (Bangs Laboratories, Cat 555A)

APC molecules of equivalent soluble fluorophore calibration beads (Bangs Laboratories, Cat 823A)

Alexa Fluor 647 molecules of equivalent soluble fluorophore calibration beads (Bangs Laboratories, Cat 647)

Anti-human IgG capture calibration beads (Bangs Laboratories, Cat 816)

COMMENTARY

BACKGROUND INFORMATION:

Small particles analysis using flow cytometry has been demonstrated since the late 1970’s(Hercher, Mueller, & Shapiro, 1979), initially with viruses. Today, a variety of small particles are being research for their potential in biomedical sciences, many of which fall under the umbrella term of extracellular vesicles (EVs) (Thery et al., 2018). The definition of EVs continues to evolve, but currently encompasses a variety of particles with overlapping nomenclature, e.g. exomeres, exosomes, microparticles, microvesicles, ectosomes, retroviral-like particles, retroviruses, etc. (Lotvall et al., 2014; Thery et al., 2018). The lack of clarity within the nomenclature is in part due to the lack of ability to separate and characterize all EVs. Today, as flow cytometers become ever more sensitive due to technological advancements, the ability to feasibly describe particles of 100 nm in diameter or smaller on some commercially available instruments is feasible (de Rond et al., 2019; Gasecka et al., 2020; Morales-Kastresana et al., 2019; Stoner et al., 2016; Tian et al., 2020; Tian et al., 2018; Welsh, Jones, & Tang, 2020; Zhu et al., 2014). As the use of small particle flow cytometry increases, it is critical for researchers to understand the particular controls and considerations that are distinctively necessary with small particle flow cytometry, as compared what is necessary with cellular flow cytometry (Nolan, 2015).

Some major hurdles for the field to overcome are utilization of calibration and controls along with clarity in reporting. Researchers with demonstrated history of small particle analysis from the International Society of Extracellular Vesicles (ISEV), International Society of Advancement of Cytometry (ISAC), and International Society for Thrombosis and Haemostasis (ISTH) formed an inter-societal EV flow cytometry working group in 2015. At the time, few controls were agreed upon and only a handful of publications utilized any form of calibration. A key product of this working group has been the development of the MIFlowCyt-EV framework in 2020(Welsh, Van Der Pol, et al., 2020). The MIFlowCyt-EV framework was published as a position paper outlining a reporting framework of key metadata, controls, calibration, and data reporting fields that should to be completed when performing single-EV measurements using flow cytometry, all of which have been demonstrated to be useful in the reproducibility or characterization of small particle data. The implementation of the metrics outlined in this framework will improve transparency and reproducibility in the current literature, where the reported concentration of EVs can differ by several orders of magnitude depending on the equipment or assay being used (Gasecka, Boing, Filipiak, & Nieuwland, 2017; Johnsen, Gudbergsson, Andresen, & Simonsen, 2019).

As the MIFlowCyt-EV framework provided insight into the controls required and a framework for transparent and standardized reporting, the practice of reporting data in calibrated standard units is increasingly encouraged in the EV research community due to the rigor that is required for accurate and reproducible reporting of measurements obtained near the limits of an instrument’s detection. Fluorescence calibration methods have been in use since the late 1990’s, with fluorescence reference materials also commercially available (Gaigalas, Wang, Schwartz, Marti, & Vogt, 2005; Hoffman, 2005; Schwartz et al., 2004; Schwartz et al., 2002; Wang & Hoffman, 2017; Wood, 1998; Wood & Hoffman, 1998). The use fluoresce calibration allows for the conversion of arbitrary units, into standardized units of fluorescence, such as molecules of equivalent soluble fluorophore (MESF) or equivalent number of reference fluorophore (ERF)(Gaigalas et al., 2005; Hoffman, 2005; Schwartz et al., 2004; Schwartz et al., 2002; Wang & Hoffman, 2017; Wood, 1998; Wood & Hoffman, 1998). Light scatter calibration methods were first demonstrated for small particles in 2009 and specifically applied to EVs in 2012 (Fattaccioli et al., 2009; van der Pol, van Gemert, Sturk, Nieuwland, & van Leeuwen, 2012). While reference materials are available for light scatter calibration, the methodology has been relatively inaccessible owing to the lack of free software available and the knowledge required to model small particle light scattering distributions within a flow cytometer. More recently, commercially available kits and software dedicated to small particle flow cytometry calibration have become more readily available (de Rond, Coumans, Nieuwland, van Leeuwen, & van der Pol, 2018).

In 2019, FCM_PASS, a small particle flow cytometer calibration software package for light scatter and fluorescence became available(Welsh, Horak, et al., 2020). FCM_PASS was developed to be compatible with any commercial calibration reference materials for fluorescence and light scattering. The recent development of FCM_PASS v3 now enables instrument tracking for longitudinal performance and alignment for light scatter. The outputs of the FCM_PASS software have also been internationally formatted into the MIFlowCyt-EV framework, so as to semi-automate the completion of the calibration fields and improve the ergonomics of the calibration and reporting process. Here protocols for acquisition of fluorescence and light scatter reference materials, creating reference material catalogues for fluorescence and light scatter reference materials, and performing fluorescence and light scatter calibration are outlined.

CRITICAL PARAMETERS:

In order to approximate the collection half-angle of a flow cytometer in order to perform light scatter calibration, some assumptions have to be made. One of these critical assumptions is that the light scatter calibration beads are the diameter inputted into the software and their refractive index. This is because predict data based on light scatter models using this information will be compared to acquired data. For this reason, assurance that the beads are sized accurately and have reasonable refractive indices are important. NIST-traceable size standards are therefore solely recommended as light scatter calibration reagents. The use of popular fluorescently hard-dyed beads such as Megamix beads (Biocytex) or FluoSpheres (Thermo Fisher Scientific) are not recommended for performing light scatter calibration. This is due their size being an approximation and their fluorescence creating a complex refractive index which is difficult to model accurately. The accuracy of light scatter modelling is also dependent upon the number of bead populations used and their diameters. It is recommended that no less than 5 beads of the same composition are used (more is preferable) with their diameters spanning the smallest bead detectable (preferably 80–100 nm) up to 600 nm. The most accurate application of light scatter calibration using FCM_PASS software is when applied to analyzers with conventional side scatter collection optics i.e. collected perpendicular to the illumination source with a symmetrical collection. Non-conventional cytometer collection optics, such as the Apogee instruments, are currently not supported due to lack of testing availability. While light scatter calibration can be applied to sorters, and has been demonstrated (Morales-Kastresana et al., 2019; Welsh et al., 2018), care must be taken in the stream alignment and the confidence of the model fit will likely be lower. This is due to the use of laser obscuration bars in front of the collection lens that are difficult to account for.

TROUBLESHOOTING:

If the fit confidence (Figure 1A) of the light scatter output is below 80% and the majority of bead populations are not within the ‘Good Fit’ portion of Figure 1C, the light scatter calibration will likely lead to inaccuracies when extrapolating diameter for low refractive index materials such as extracellular vesicles. If this is the case, ensure that the particles used to calibrate the light scatter calibration parameter are NIST-traceable with respect to diameter, that their refractive index was inputted correctly, that the median statistic for each bead was inputted correctly, and the wavelength for modelling is inputted correctly. If all of these statements are true and the instrument is a cytometer with conventional collection optics, it likely that the alignment of the optical fiber collection is not quite central to the light being focused from the side scatter collection lens or the laser beam is not quite perpendicular to the collection lens. Optimal alignment for light scatter modelling should be done using fluorescence so as not to introduce bias from the non-isotropic light scatter distribution of particles $\ge$ 100 nm (Welsh, Horak, et al., 2020). Due to chromatic aberration in the collection optics, the fluorescence signal for optimal alignment of light scatter should be the closest wavelength to one another e.g. a 530/30 signal for 488 nm side scatter alignment. Using this channel, the maximum separation between a dimly fluorescence particle and the instrument noise will provide the highest sensitivity for both this fluorescence channel and the similar wavelength side scatter parameter.

TIME CONSIDERATIONS:

Basic Protocol 1: Acquisition and gating of light scatter calibration materials – 20–40 minutes

Basic Protocol 2: Acquisition and gating of fluorescence calibration materials – 10–20 minutes

Basic Protocol 3: Cataloguing light scatter calibration materials – 10 minutes

Basic Protocol 4: Cataloguing fluorescence calibration materials – 5 minutes

Basic Protocol 5: Creating cytometer databases and datasets – 1 minute

Basic Protocol 6: Importing fcs files – 1–2 minutes

Basic Protocol 7: Fluorescence calibration – 2–5 minutes

Basic Protocol 8: Light scatter calibration – 2–5 minutes

Basic Protocol 9: Performing and reporting fcs file calibration – 2–30 minutes