Colorimetric Assessment of Deiodinase 1 Activity in Human Liver Microsomes Using the Sandell-Kolthoff Reaction

Summary

This protocol describes a colorimetric method to reliably assess the influence of chemicals on Deiodinase 1 (DIO1) activity in human liver microsomes using the Sandell-Kolthoff reaction. It includes predefined acceptance criteria and statistical performance metrics and supports ongoing efforts toward OECD test guideline inclusion.

Abstract

This method enables the identification of test substances that inhibit DIO1 activity and may thus interfere with thyroid hormone homeostasis. The protocol presented describes a non-radioactive, colorimetric method for assessing the effects of test substances on the enzymatic activity of Deiodinase 1 (DIO1) in human liver microsomes. The assay is based on the Sandell-Kolthoff (SK) reaction, which quantifies iodide released during the deiodination of reverse triiodothyronine (rT3) to diiodothyronine (T2). The yellow cerium (IV) is reduced to colorless cerium (III) in the presence of iodide, allowing photometric detection at 415 nm. In this standardized setup, the cerium (IV) concentration is set to 40 mM to improve signal stability and robustness under the described conditions. To ensure assay reliability and reproducibility, a comprehensive control setup is included: 6-propyl-2-thiouracil (reference item), aurothioglucose (positive control), 1-thio-β-D-glucose sodium salt (negative control), and 1% dimethyl sulfoxide (solvent control). The protocol includes a range-finding assay to determine appropriate test concentrations, standardization of microsome batch-specific iodide release activity and structured chemical interference testing in the absence of microsomes, with an optional follow-up assessment without dithiothreitol (DTT) to identify DTT-dependent artifacts. The DIO1-SK assay is suitable for medium- to high-throughput screening and mechanistic toxicology studies. Its robustness, scalability, and applicability to human microsomes make it a valuable tool for investigating endocrine disruption. This protocol supports regulatory validation efforts.

Introduction

Endocrine disruptors (EDs) are chemicals that interfere with hormonal regulation and consequently cause adverse effects in humans^1,2. Deiodinases (DIO) are membrane-associated enzymes, with selenocysteine as a key residue in their active sites. This enzyme family comprises three distinct isoforms (DIO1, DIO2, and DIO3), which collectively modulate circulating thyroid hormone (TH) concentrations and govern their tissue-specific availability within cells³. Thyroxine (T4), which serves as a prohormone, and triiodothyronine (T3), the active form, play vital roles in regulating energy expenditure, growth, developmental processes, and thermoregulation in the body⁴. The DIO isoforms differ in the characteristics of their deiodination site specificity, kinetic properties, and substrate preferences³. DIO1 is the main deiodinase isoform in adult liver and exhibits strong substrate affinity for reverse triiodothyronine (rT3) and sulfated iodothyronines, while DIO2 preferentially utilizes T4 and rT3, and DIO3 primarily acts on T4 and T3. Using human liver microsomes and rT3 as substrate allows selective detection of DIO1 activity, minimizing interference from other isoforms. While the precise consequences of substance-induced DIO1 inhibition remain to be fully elucidated, evidence from human gene mutations affecting DIO1 clearly demonstrates that DIO1 can disrupt the thyroid hormone system⁵. As a result, there is a critical need for robust and reproducible methods to investigate DIO1 inhibition.

The goal of the presented method is to quantify DIO1 activity in vitro using a non-radioactive, colorimetric approach based on the Sandell-Kolthoff (SK) reaction. DIO1 catalyzes the conversion of reverse rT3 to T2, a process that results in the release of iodide ions. The SK reaction enables sensitive detection of the released iodide, allowing determination of iodide release activity (IRA) as a direct measure of DIO1 activity. This assay enables the detection of inhibitory effects of test substances on DIO1, supporting mechanistic toxicology, screening, or regulatory decision-making. It is particularly relevant for investigating DIO1 inhibition as a molecular initiating event of adverse outcome pathways leading to thyroid disruption, as outlined by international validation efforts such as those coordinated by EURL ECVAM^6,7. The method was originally developed in 2012^8,9 and further optimized^6,7 in response to the growing need for robust, reproducible, and scalable in vitro assays that can be integrated into test batteries for endocrine screening, especially in the context of thyroid hormone metabolism. Recently, the application of analog SK-based setups for medium-¹⁰ to high-throughput screening¹¹ was published, demonstrating the broad applicability of the SK-based approach.

Even though the DIO1-SK assay indirectly measures DIO1 activity by quantifying IRA rather than TH products, it offers several advantages over alternative techniques. Unlike methods that rely on radiolabeled substrates¹² or mass spectrometry-based quantification of depleted or formed thyroid hormones¹³, the SK reaction provides a simple, scalable, and cost-effective readout by measuring iodide release at 415 nm¹⁴. While LC-MS/MS methods with solid-phase extraction (SPE-LC-MS/MS) offer high specificity and sensitivity for quantifying rT3 and T2, they require expensive instrumentation and are less suitable for higher-throughput screening. In contrast, the DIO1-SK assay can be implemented in standard laboratory settings and has been prevalidated for use with human liver microsomes, demonstrating high reproducibility and predictivity^6,7. In a comparative study, results from the SK reaction, specifically iodide release, were directly compared with quantification of thyroid hormone metabolites (rT3 and T2) using SPE-LC-MS/MS¹⁵. The findings revealed a strong concordance between the two approaches across a range of test substances and inhibition potencies. For all substances tested, changes in iodide release activity (IRA) measured by the SK reaction closely mirrored alterations in rT3 and T2 levels determined by mass spectrometry. This further supported the validity of the SK readout as a surrogate for DIO1 activity. These comparative analyses confirm that the SK assay provides a robust and reliable alternative to more complex analytical techniques, and that iodide release can be measured with confidence to assess DIO1 inhibition in vitro.

However, the SK method has certain limitations that must be considered: It is susceptible to interference from substances that either contain free iodide as an impurity, release iodide non-enzymatically, or react with assay components such as dithiothreitol (DTT). DTT is included as a reducing cofactor to maintain deiodinase catalytic competence by supporting regeneration of the active-site selenocysteine in its reduced state during turnover. Because DTT is a strong reductant, certain test substances (or impurities) may undergo redox reactions that can influence the colorimetric readout and, in some cases, lead to DTT-dependent assay artifacts unrelated to enzymatic iodide release. Such interferences can lead to false-negative results for DIO1 inhibition, as the measured iodide may not originate from the enzymatic conversion of rT3. Thus, while the SK assay is suitable for routine screening and most test substances, LC-MS/MS could be used for confirmation whenever assay interference is suspected or for substances with complex chemical properties.

The DIO1-SK assay is currently being readied for inclusion as an OECD test guideline in an international PEPPER (Public-private platform for the validation of endocrine disruptors characterization methods) project, which highlights its growing importance for regulatory applications and the international harmonization of endocrine disruptor testing strategies. The assay may be integrated into comprehensive in vitro test batteries, supporting both chemical safety assessment and mechanistic investigations into endocrine disruption¹⁶. When combined with assays addressing additional key events along the thyroid adverse outcome pathway, it contributes to a more complete understanding of thyroid hormone disruption mechanisms.

In addition, the protocol’s scalability and high reproducibility make it well-suited for medium- to high-throughput screening. This is essential for evaluating large numbers of substances during drug development and in academic settings, as recently demonstrated¹¹. Researchers interested in DIO1 inhibition, whether for screening, mechanistic toxicology, or regulatory decision-making, will benefit from a robust, accessible, and versatile assay. The validation of this method will make it a standard tool for research on thyroid hormone system disruption and for evaluating chemical safety.

Protocol

NOTE: Before performing the main assay, a series of preparatory steps is essential to ensure the accuracy, reproducibility, and validity of the DIO1-SK assay. These preassay activities include casting ion-exchange, resin-filled filter plates, which are critical for effective separation and minimizing background interference in the SK reaction. The Sandell-Kolthoff reaction is standardized regularly using iodide standard curves to monitor for systematic changes and maintain assay quality. The activity of human liver microsomes is measured to determine batch-specific iodide release capacity, ensuring consistent enzymatic performance across experiments. Together, these steps provide a robust foundation for the subsequent main assay work and data evaluation, supporting the generation of reproducible and interpretable results.

1. Preassay work

1. Preparation of buffers and reagents
   1. Prepare all buffers and stock solutions in advance as described in Table 1. Store aliquots as indicated to ensure stability and reproducibility.
   2. On the day of the assay, prepare the substrate mix freshly for each 96-well plate. For one 96-well plate, combine 5.75 mL of HEPES/EDTA buffer (50 mM HEPES, 2.16 mM EDTA, pH 7.0), 0.5 mL of thawed 1 M DTT, and 10 µL of 6 mM rT3 (dissolved in DMSO). Mix gently and keep the substrate mix on ice until use.
2. Preparation of reference
   1. Prepare a 10^-1 M stock solution of reference control 6-PTU in an appropriate solvent (e.g., DMSO) freshly on the day of performing the assay.
   2. Prepare a 1:10 dilution of this stock in diH₂O (RI-C8) to achieve a DMSO concentration of 10% and prepare a series of dilutions from this stock in 10 % DMSO in diH₂O as described in Table 2.
3. Preparation of the positive control
   1. Prepare a 10^-2 M stock solution of the positive control Aurothioglucose in an appropriate solvent (e.g., DMSO) and store it for up to 6 months at 4 °C.
   2. On the day of the assay, dilute the stock to 10^-3 M in diH₂O to achieve a DMSO concentration of 10% DMSO.
4. Preparation of the negative control
   1. Prepare a 10^-2 M stock solution of the negative control 1-Thio-β-D-glucose sodium salt freshly on the day of performing the assay.
   2. On the day of the assay, dilute the stock to 10^-3 M in diH₂O to achieve a DMSO concentration of 10% DMSO.
      NOTE: Always use appropriate safety precautions, especially when handling hazardous chemicals such as sodium arsenite and sulfuric acid. All reagents should be prepared with deionized water and stored according to the manufacturer's recommendations. The substrate mix, once prepared, and the thawed DTT aliquot are intended for single use on the day of the assay and should not be refrozen.
5. Casting of ion exchange resin
   1. Use an appropriate rectangular transparent container for the preparation.
   2. Add about 250 g (or the desired quantity) of ion exchange resin to a large container and wash. Wash by adding 10% acetic acid; let the resin suspension rest for 10 min.
   3. After letting the suspension settle for 10 min, slightly tilt and aspirate the supernatant. Wash at least 5x in total or until no more color is leaking into the solvent.
   4. After the last washing step, add 10% acetic acid so that it makes up approximately 50% of the total volume.
   5. Tilt the container so that the resin is moving to one side and stabilize the container in the tilted position. Wait for ~10 min for the phases to separate.
   6. Place a 96-well filter plate on top of a 96-deep well plate and add 100 µL of 10% acetic acid into each well of the 96-well filter plate.
   7. Cut the tips used for resin casting at a height of ~1 cm to widen the opening at the end and cast 300 µL of ion exchange resin into each well of the 96-well filter plate.
      NOTE: The tips used in our work have a maximum volume of 1,200 µL and can be attached to multichannel pipettes.
   8. Elute the acetic acid by centrifuging into a 96-deep well plate with 100–200 × g in a centrifuge with a swing-out rotor for microtiter plates for 1 min. Check whether the resin is distributed evenly across the plate.
   9. Seal the plate with an impermeable sheet of plastic and store at 4 °C for a maximum of 6 months.
      NOTE: Ion exchange resin constituents are known to affect the SK reaction if not washed out properly. The removed supernatant can be used directly in the SK reaction (use 50 µL of supernatant sample, add 50 µL of cerium, and finally, 50 µL of arsenite solution) and can be compared against a 10% acetic acid sample (blank control) to ensure complete washing out of the components.
6. Standardization of the Sandell-Kolthoff reaction
   1. Prepare iodide dilutions from a respective iodide source (e.g., iodide standard solution) with recommended concentrations of 1,500, 1,000, 750, 500, 400, 300, 200, 100, 50, 25, 10, 5, and 1 nM iodide using a respective iodide standard in diH₂O.
   2. Add 50 µL of pure diH₂O as a control and 50 µL of the prepared iodide dilutions to a 96-well plate in triplicate as described in the layout in Figure 1.
   3. Add 50 µL of 40 mM cerium solution to all wells and start the reaction by adding 50 µL of 25 mM arsenite solution to all wells.
   4. As soon as possible after adding the arsenite solution, measure the absorption using a plate reader at 415 nm (±2 nm) by recording the optical density (OD) every minute for 21 min, as well as recording the initial OD.
   5. Evaluate the results as described in section 4.
      NOTE: Usually, a background of ΔOD_21min > 0.3 in the pure diH₂O control will need attention and further investigation of underlying causes (e.g., contamination in the As or Ce-batch, low water quality).
7. Assessing microsome activity
   1. Prepare the highest concentration (RI-C8) of the reference item (6-PTU) as described in step 1.2.
   2. Prepare the six microsome dilutions as described in Table 3 and the substrate mix as described in Table 1.
   3. Add 10 µL of a 10% (v/v) solvent dilution in diH₂O (e.g., 10% DMSO in diH₂O) and 10 µL of 10^-2 M 6-PTU (RI-C8) into a 96-well plate in triplicate as described in the plate layout in Figure 2.
   4. Add 40 µL of microsome dilutions in diH₂O (resulting in 20, 10, 5, 2.5, 1.25, 0.68, and 0 µg microsomal protein per well) to the 96-well plate as shown in the plate layout.
      NOTE: Suppliers typically provide the stock concentration of microsomal protein. If not, standard protein quantification methods may be used, though we did not apply them in this study.
   5. Add 50 µL of freshly prepared substrate mix to each well and seal the plate with an impermeable sheet of plastic.
   6. Place the 96-well plate on a shaker in an incubator (37 °C at 850 rpm) and incubate for 2 h.
   7. After the incubation time, put the plate on ice to stop the reaction.
   8. Conduct the ion exchange analogous to step 2.2.
   9. Transfer appropriate amounts of sample from the deep well obtained in the ion exchange step to a fresh 96-well plate to measure undiluted (50 µL of sample), 1:2 (25 µL of sample + 25 µL of 10% acetic acid), and 1:4 (12.5 µL of sample + 37.5 µL of 10% acetic acid) dilutions, keeping the plate layout unchanged.
   10. Evaluate the results as described in section 4 and determine the dilution factor for the samples in 10% acetic acid as well as the microsomal protein concentration that still leads to the highest possible ΔOD_21min -BG values in the Sandell-Kolthoff reaction without reaching the plateau of saturation.
       NOTE: Usually, an iodide release of ΔOD_21min > 0.5 is easily achievable in the 1:4 or 1:2 acetic acid dilutions of the 20 µg of microsomal protein per well sample and usually ranges around a ΔOD_21min of 1. Higher microsomal protein concentrations are not recommended for use, as they increase the background and nonspecific binding during the assay. It is recommended to repeat the measurement of a 6-PTU dilution series using the determined microsome concentration and dilution factor to confirm the result.
8. Solubility assessment of test items
   1. Dissolve the test item at 100 mM or 200 mg/mL in a suitable solvent (e.g., DMSO) at room temperature. Mix or vortex and check dissolution visually (microscope).
   2. If undissolved, sonicate for up to 5 min and check again. If still undissolved, heat at 37 °C for up to 60 min and recheck.
   3. If insoluble, dilute further or prepare a new stock at a lower concentration and repeat the procedure. If DMSO is not suitable, try water or ethanol as alternative solvents.
   4. Once dissolved, prepare a 1:10 dilution in diH₂O (10% solvent) and repeat the assessment (mixing, sonication, heating as needed).
   5. To simulate assay conditions, prepare a 1% test item dilution in assay buffer and repeat the solubility check.
9. Test item preparation and range finding
   1. Prepare the test item stock solution at the highest soluble concentration in a suitable solvent (preferably DMSO), as determined during solubility assessment in step 1.8.
   2. For the range finding assay, prepare eight test item dilutions by consecutive 1:10 (v/v) dilution steps using the stock solution and 10% DMSO in diH₂O.
   3. Conduct the range finding assay to estimate the inhibition profile of the test item by performing the steps described in section 2. If the test item shows ≥25% inhibition at any concentration, adjust the concentration range for the final assay runs to include at least three concentrations in the linear inhibition range and two concentrations below 10% IRA, if applicable.
   4. If the test item shows <25% inhibition in all tested concentrations, reuse the same dilution series for the final assay runs. In this case, the range finding assay may count as one of the three required independent assay runs.
      NOTE: Perform at least two, ideally three, independent experiments for each test item to ensure reproducibility and reliability of the results. If another dilution is chosen to better cover the activity range, a deviation from the 1:10 dilution ratio is allowed. The 25% cutoff in this testing strategy (see steps 1.9.3 and 1.9.4) distinguishes test item effects from negative controls, accounting for baseline activity variability while providing an additional safety margin. This threshold ensures reliable classification of inhibitory profiles.

Figure 1: Plate layout for iodide standard curve. This figure shows the 96-well plate layout used for generating the iodide standard curve. Wells contain serial dilutions of iodide ranging from 1 nM to 1,500 nM, as well as diH₂O-only controls.  Please click here to view a larger version of this figure.

Figure 2: Plate layout for microsomal protein titration and reference item testing. This figure displays the 96-well plate layout for microsomal protein titration and reference inhibitor (6-PTU) testing. Rows contain increasing amounts of microsomal protein per well, with and without 6-PTU.  Please click here to view a larger version of this figure.

2. Performing the DIO1–SK assay

1. Microsome incubation with test item
   1. Prepare the reference item as described in step 1.2, as well as the positive and negative control, as described in steps 1.3 and 1.4. Prepare the test items with suitable concentrations following the protocol described in steps 1.8 and 1.9.
   2. Prepare the substrate mix as described in Table 1 and prepare a microsome suspension in the predefined protein concentration according to step 1.7 by diluting carefully thawed, aliquoted microsome solution in diH₂O. Keep the microsome suspension on ice until needed for incubation.
   3. Following the plate layouts shown in Figure 3 and Figure 4 (if more than two test items are evaluated simultaneously) add 10 µL of reference item concentrations, positive and negative control, solvent control (10% (v/v) solvent in diH₂O), and test items concentrations to the wells.
   4. Add 40 µL of the microsome suspension to all wells. Add 50 µL of the freshly prepared substrate mix to the samples.
   5. Seal the plate with an impermeable sheet of plastic and place it on a shaker in an incubator (37 °C at 850 rpm) for 2 h; place the plate after the incubation on ice to stop the reaction.
   6. Place the plate on ice after incubation to stop the reaction.
      NOTE: The plate can be stored on ice until measurement to preserve sample integrity. The first assay plate contains eight concentrations of the reference item, while each additional plate on the same day contains only the highest concentration of the reference item, leaving more space for additional test items if needed.
2. Separation via ion exchange resin-filled 96-well filter plates
   1. Put a prepared ion exchange resin-filled 96-well filter plate, as prepared in step 1.5, on top of a 96-deep well plate and add 150 µL of 10% acetic acid to each well of the ion exchange resin-filled 96-well filter plate to wet the resin in the columns.
   2. Elute the acetic acid by centrifuging into the used 96-deep well plate with 100–200 × g in a centrifuge with a swing-out rotor for microtiter plates for 1 min.
   3. Replace the used 96-deep well plate with a new 96-deep well plate.
   4. Add 133 µL of 10% acetic acid to each well of the incubated 96-well plate from step 2.1.6 and shake the plate briefly to mix the solutions.
   5. Transfer 175 µL of the samples prepared in the previous step onto the ion exchange resin-filled 96-well filter plate, maintaining the initial plate layout.
   6. Pass the samples through the resin and into the 96-deep well plate by centrifuging (100–200 × g) in a centrifuge with a swing-out rotor for microtiter plates for 1 min. Remove the ion exchange resin-filled 96-well filter plate.
      NOTE: The 96-deep well plate with samples can be sealed with an impermeable sheet of plastic and stored at 4 °C for at least 3 months. This allows additional measurements in case of manual/technical errors or changes of the dilution factor in the Sandell-Kolthoff reaction or measurement on the following days.
3. Sandell-Kolthoff reaction
   1. Depending on the determined dilution factor of the samples in 10% acetic acid for the microsome batch used (see step 1.7), add 50 µL of the undiluted or appropriately diluted sample to a novel 96-well plate.
   2. Add 50 µL of 40 mM cerium solution (prepared as described in Table 1) to all wells and start the reaction by adding 50 µL of 25 mM arsenite solution (prepared as described in Table 1) to all wells.
   3. As soon as possible after adding the arsenite solution, measure the absorption using a plate reader at 415 nm (±2 nm) by recording the optical density (OD) every minute for 21 min, as well as recording the initial OD.

Figure 3: Plate layout for parallel testing of two test items. 96-well plate layout for testing two test items in triplicate across multiple concentrations, along with replicates of negative, positive, solvent controls, and the reference item. Used as the first assay plate of the day. Please click here to view a larger version of this figure.

Figure 4: Plate layout for parallel testing of three test items. 96-well plate layout for testing three test items in triplicate across multiple concentrations, with replicates of negative, positive, and solvent controls. The reference item is included only at its highest concentration, as it is fully tested on the first plate (see Figure 3). Please click here to view a larger version of this figure.

3. Testing for interference (± microsomes and ± DTT)

1. Prepare the highest concentration (RI C8) of the reference item (6 PTU) as described in step 1.2 and prepare the test item stock solution and dilutions (highest test item concentration included).
2. Prepare the microsome suspension at the defined concentration per well (see step 1.7) by diluting carefully thawed, aliquoted microsomes in diH₂O. Keep the microsome suspension on ice until use.
3. Prepare two substrate mixes: the standard substrate mix with DTT as described in Table 1, and an otherwise identical substrate mix without DTT (omit DTT only).
4. Dispense into the designated wells of a 96 well plate according to Figure 5: 10 µL of RI C8, 10 µL of 10% (v/v) DMSO in diH₂O (solvent control), and 10 µL of each test item dilution.
5. Add 40 µL of microsome suspension to wells designated “with microsomes” and 40 µL of diH₂O to wells designated “without microsomes”, according to Figure 5.
6. On ice, add 50 µL of substrate mix with DTT to the wells designated “ with DTT” and 50 µL of the substrate mix without DTT to the wells designated “without DTT” (Figure 5).
7. Seal the plate with an impermeable sheet of plastic and incubate on a shaker in an incubator (37 °C, 850 rpm) for 2 h.
8. After incubation, place the plate on ice to stop the reaction.
9. Perform ion exchange separation and iodide quantification via the SK reaction as described in steps 2.2 and 2.3, and evaluate results as described in Section 4.
   NOTE: This single plate setup (Figure 5) tests the highest test item concentration in parallel with vs. without microsomes and with versus without DTT to distinguish enzymatic DIO1 activity from assay artifacts. An IRA increase ≥20% in wells without microsomes indicates SK interference (iodide release/readout independent of DIO1), and the substance is classified as “not applicable in the DIO1 SK assay” and excluded from analysis. Differences between with DTT and without DTT conditions help identify DTT-dependent iodide release as an additional interference mechanism.

Figure 5: Plate layout for interference testing. This figure shows the 96-well plate layout for interference testing, where test items, reference item, and solvent control are tested both with and without microsomes in the presence of DTT as well as without microsomes and without the presence of DTT. Please click here to view a larger version of this figure.

4. Data evaluation

1. Evaluation of DIO1 activity
   1. Determine the ΔOD_21min values via subtraction of the 21 min values of all samples from the initial measured (0 min) values:
      
   2. Determine the ΔOD-BG values by subtracting the mean of ΔOD_21min values of the inhibited 10 ³ M 6-PTU (RI) controls from the ΔOD_21min values of all samples:
      
   3. Normalize the values of the test item to the respective solvent control values via division of the test item(s) (TI) ΔOD_21min-BG_TI values by the mean of the ΔOD_21min -BG_SC values of the respective solvent control (SC), generating iodide release activity (IRA) values in %.
      
   4. Plot the IRA values of the different concentrations of the test items in a statistics software package, with the IRA values on the y-axis (linear) and the test item concentrations on the x-axis (logarithmic).
   5. Use a curve-fit algorithm to visualize a concentration-response relationship (e.g., “[Inhibitor] versus response -- Variable slope (four parameters)” in statistical software of choice. “Top” represents the maximal response, “Bottom” represents the lowest response, and “HillSlope” describes the steepness of the curve:
      
   6. Determine the IC₅₀ of the reference item and, if applicable, of the test item.
   7. Calculate the coefficient of variance (CV) of log IC₅₀ estimate of the reference item in %:
      
   8. Calculate the IRA of the negative control normalized to the solvent control in %:
      
   9. Calculate the IRA of the positive control normalized to the solvent control in %:
      
   10. Calculate the z’-factor:
       
   11. To check the validity of the experiment, compare the calculated values to the acceptance criteria in Table 4.
       NOTE: All criteria must be reached to obtain a valid assay run.
   12. Classify the test items as full inhibitors, partial inhibitors, or non-inhibitors based on their maximum inhibition and IC₅₀ values, as summarized in Table 5.

Results

Figure 6 provides an overview of the DIO1‑SK assay workflow and the key standardization steps required to successfully establish the method in a laboratory. The workflow (Figure 6A) summarizes the sequential phases of the assay, including the use of human liver microsomes, incubation with test items and controls, iodide extraction via ion‑exchange separation, and the colorimetric iodide quantification using the Sandell–Kolthoff (SK) reaction. Figure 6B highlights the standardization strategy that underpins robust performance and valid results, including (i) standardization of the SK reaction, (ii) batch‑specific microsome activity testing, (iii) solubility assessment and interference testing, (iv) a range‑finding assay, and (v) adjustment of test item concentrations based on the first valid assay run.

A central element of this strategy is batch‑specific microsomal activity testing, because human liver microsome batches can differ substantially in their rT3 deiodination activity, necessitating the determination of an appropriate enzyme concentration for each new batch prior to routine substance testing.

This staged testing strategy is critical for implementation: only after the assay’s baseline performance (controls, SK reaction behavior, and microsome activity) is verified and the assay conditions are standardized can substance testing be expected to yield valid, interpretable inhibition profiles.

Figure 6: DIO1 SK assay workflow and standardization steps. (A) The workflow illustrates the usage of human liver microsomes, setup of the 96-well plate, incubation with test items and controls, iodide extraction, and colorimetric quantification via the SK reaction. Each step is depicted sequentially to highlight the main phases of the assay. (B) The standardization steps include standardization of the Sandell-Kolthoff reaction, batch-specific microsome activity testing, solubility assessment and interference testing, range finding assay, and adjustment of test item concentrations based on the first valid assay run. Please click here to view a larger version of this figure.

Building on the workflow and standardization framework shown in Figure 6, Figure 7 illustrates how laboratories determine an appropriate microsomal protein concentration (and post‑ion‑exchange dilution) to achieve a sufficient dynamic range while maintaining a sigmoidal inhibition response for the reference inhibitor. In Figure 7A, ΔOD21min‑BG increases with increasing microsomal protein and depends on the post‑ion‑exchange dilution (undiluted, 1:2, 1:4). The dilution that yields the highest ΔOD21min‑BG while maintaining a sigmoidal response is selected for subsequent experiments (example shown: 1:2 dilution). In Figure 7B, 6‑PTU inhibition curves are compared across microsomal protein amounts to identify the lowest protein concentration that still yields a sigmoidal curve (example shown: 5 µg of protein/well), thereby balancing sensitivity with background.

Importantly, this optimization step must be performed by each laboratory when implementing the method, because the required protein concentration depends on the activity of the microsome batch used, necessitating batch‑specific standardization of enzyme concentration.

Figure 7: Microsomal activity and inhibition by 6-PTU. (A) The graph shows the increase in activity as the background (BG)‑corrected Sandell-Kolthoff signal (ΔOD_{21 min} n- BG) with increasing amounts of microsomal protein per well (log10[µg protein/well]), tested at undiluted, 1:2, and 1:4 dilutions. The highest dilution with the highest ΔOD_{21 min}– BG value, as well as with a sigmoidal shape, will be used for future experiments. In this case, it is the 1:2 dilution. (B) Concentration–response curve of the reference inhibitor 6‑propyl‑2‑thiouracil (6‑PTU) showing iodide release activity (IRA in %) plotted against log10[concentration (M)] at different microsomal protein loads (5, 7.5, and 10 µg protein/well). IRA values are normalized to the solvent control. Data points of both graphs represent mean ± SD of three technical replicates (wells) per condition from one representative microsome activity test. Curves were fitted by nonlinear regression using a four‑parameter logistic (“variable slope”) model (inhibitor vs. response). The smallest protein concentration with a sigmoidal curve will be used for future experiments; in this case, 5 µg/well. Please click here to view a larger version of this figure.

Once the workflow and standardization steps (Figure 6) have been completed and an appropriate microsomal protein concentration has been established for the selected microsome batch (Figure 7), the core assay parameters are in place to proceed with routine substance testing. At this stage, test items can be evaluated alongside the plate control set (reference item, solvent control, and positive/negative controls) in the standardized plate format, enabling robust normalization and consistent interpretation of inhibition profiles.

Figure 8 demonstrates assay discrimination based on concentration–response behavior in iodide release activity (IRA). The reference inhibitor 6-PTU (Figure 8A) shows a concentration-dependent decrease in IRA, providing the expected sigmoidal inhibition profile, as expected for assay performance verification. Tannic acid (Figure 8B) displays an inhibitory profile that is comparable to the reference inhibitor (maximum inhibition > 90%) under the shown conditions. The three highest concentrations of the test substance exhibited negative IRA values. IRA values <0 % do not reflect a biologically meaningful ‘negative iodide release’ but arise from the calculation procedure (background subtraction and normalization to the solvent control), which can propagate small fluctuations in the solvent control into negative normalized values. Accordingly, IRA values below 0% are considered equivalent to 0% (baseline) for result interpretation. Dibutylphthalate (Figure 8C) exhibits no inhibition (max. inhibition below 25%) across the tested concentration range, representing a non‑inhibitor profile. Together, these profiles illustrate that the assay can differentiate inhibitory from non‑inhibitory behavior under standardized conditions. The test system can also identify partial inhibitors, which show maximum inhibition levels between 25% and 90% (data not shown). Further details about how these are categorized can be found in Table 5.

Figure 8: Dose-response behavior of the DIO1–SK assay distinguishes inhibitors from non‑inhibitors. Dose–response curves show iodide release activity (IRA in %) plotted against log10 concentration (M) for (A) the reference inhibitor 6‑PTU, (B) a representative inhibitor (tannic acid), and (C) a representative non‑inhibitor (dibutyl phthalate). IRA values were calculated after background subtraction and normalized to the solvent control. Symbols represent mean ± SD (n = 3) from one independent experiment, each performed with three technical replicates (wells) per concentration. Curves were fitted by nonlinear regression using a four‑parameter logistic (“variable slope”) model (inhibitor vs. response). IRA values below 0% are considered equivalent to 0% (baseline) for result interpretation. These data demonstrate that the assay yields a sigmoidal inhibition profile for the reference inhibitor and can confirm whether a concentration-response curve is inhibitory or non-inhibitory under standardized conditions. Please click here to view a larger version of this figure.

In a valid assay run, solvent controls (SC) show a clear ΔOD_21min signal separation from the fully inhibited reference condition (6‑PTU, RI‑C8). In an example dataset (10 independent runs), each plate included n = 9 technical replicates for SC and the fully inhibited reference (6‑PTU at 10⁻³ M). Across plates, the plate‑mean SC ΔOD_21min values ranged from 1.383 to 1.819 (mean ± SD: 1.565 ± 0.114), whereas the plate‑mean ΔOD_21min values for the inhibited 6‑PTU condition ranged from 0.258 to 0.771 (mean ± SD: 0.473 ± 0.087). The calculation of ΔOD in this assay is based on the difference between OD_0min and OD_21min, and assay validity is commonly supported by a z′‑factor threshold of >0.5.

Across 10 independent assay runs, 6‑PTU produced a sigmoidal concentration-response curve with IC₅₀ values ranging 1.20E-06 M to 5.02E-06 M (mean ± SD: 2.74E-06 M ± 1.13E-06 M). The acceptance criteria for the IC₅₀ values in the reference range of 1.00E-06 to 1.00E-05 are based on data from previous validation studies. However, the IC₅₀ values are significantly affected by the specific microsome batch used and may exhibit considerable variability. Nonetheless, these variations are expected to remain within the same order of magnitude.

Assay quality can be summarized using the z′‑factor, which reflects both the signal window and the variability of the controls. In the described dataset, six reference curves were associated with z′‑factors ranging from 0.544 to 0.774 (mean ± SD: 0.658 ± 0.062), consistent with the commonly used acceptance target of z′ > 0.5.

In practice, suboptimal runs can be identified by deviations in assay background and/or failure to meet predefined acceptance criteria. For example, a high background in the Sandell-Kolthoff reaction is indicated when the pure diH₂O blank shows ΔOD > 0.3, which warrants investigation (e.g., reagent contamination or water quality issues). Such background increases can reduce the assay’s dynamic range and elevate control variability, thereby contributing to failed acceptance criteria, including a z′‑factor below 0.5. Wells showing obvious technical artifacts (e.g., precipitation or abnormal discoloration) are excluded as outliers from evaluation. Outliers determined via boxplot outlier testing are also excluded. If acceptance criteria are not met after removing the outliers, the assay run is considered non‑valid and must be repeated.

Reagent/Buffer	Information of preparation
Acidic ammonium cerium solution (40 mM (NH4)4Ce(SO4)4*2H2O, 0.5 M H2SO4) (250 mL)	Add 6.32 g of (NH4)4Ce(SO4)4*2H2O and 125 mL of diH2O to a 250 mL volumetric flask. Add 125 mL 1 M H2SO4 to reach a final volume of 250 mL. Store at room temperature up to 6 months.
DTT (Aliquots of 1 M)	Aliquot a prepared or supplied 1 M DTT solution in H2O as 0.5 mL aliquots into 1.5 mL microcentrifuge tubes and store at -20°C for up to 6 months.
HEPES (50 mM) / EDTA (2.16 mM) buffer (pH 7.0)	Using a 250 mL volumetric flask, add 12.5 mL 1 M HEPES solution, 201 mg Ethylenediaminetetraacetic acid (EDTA) and fill up to 250 mL with diH2O. Store at  4°C for up to 3 months.
rT3 Solution (“substrate mix tubes”)	Dissolve rT3 in DMSO to a final concentration of 6 mM, aliquot them in 10 μL portions in 15 mL centrifuge tubes and store them at -20°C for up to 6 months.
Sodium arsenite solution (25 mM NaAsO2, 0.8 M NaCl, 0.5 M H2SO4) (250 ml)	Add 0.81 g of NaAsO2, 11.7 g of NaCl and 125 mL of diH2O to a 250 mL volumetric flask. Add 125 mL 1 M H2SO4 to reach a final volume of 250 mL. Store at room temperature up to 6 months.

Table 1: Preparation of buffers and reagents. This table details the preparation steps and storage instructions for all buffers and stock solutions used in the DIO1 SK assay, including HEPES/EDTA buffer, DTT, rT3 solution, acidic ammonium cerium solution, and sodium arsenite solution.

Name of the reference item dilution	Reference item dilution concentration [M]	10% DMSO [µL]	Reference item	Final concentration of reference item in the assay [M]
RI-C8	1.00E-02	450	50 µL of 10-1 M reference item stock solution	1.00E-03
RI-C7	1.00E-03	450	50 µL of RI-C8	1.00E-04
RI-C6	1.00E-04	450	50 µL of RI-C7	1.00E-05
RI-C5	3.16E-05	342	158 µL of RI-C6	3.16E-06
RI-C4	1.00E-05	450	50 µL of RI-C6	1.00E-06
RI-C3	3.16E-06	450	50 µL of RI-C5	3.16E-07
RI-C2	1.00E-06	450	50 µL of RI-C4	1.00E-07
RI-C1	1.00E-07	450	50 µL of RI-C2	1.00E-08

Table 2: Preparation of reference item dilutions. This table summarizes the preparation of serial dilutions for the reference item (6-PTU), including concentrations, volumes of DMSO, and final concentrations used in the assay.

Microsome per well [µg]	diH2O [µL]	Microsome dilution [µL]	Final enzyme concentration in the assay [µg/mL]
20	780	20 µL of 20 mg/mL microsome stock solution	200
10	400	400 µL of 20 µg microsome per well dilution	100
5	400	400 µL of 10 µg microsome per well dilution	50
2.5	400	400 µL of 5 µg microsome per well dilution	25
1.25	400	400 µL of 2.5 µg microsome per well dilution	12.5
0.68	400	400 µL of 1.25 µg microsome per well dilution	6.8

Table 3: Preparation of microsome dilutions. This table provides the dilution scheme for microsomal protein per well, including the required volumes of diH₂O and microsome stock solution to achieve the final microsomal protein concentrations for the assay.

Acceptance criteria	Suggested cut-off value
Numeric
IC50 of reference item [M]	1*10-05 - 1*10-06
CV of log IC50 estimate of reference item [%]	< 3
IRA negative control [%]	80 - 120
IRA positive control [%]	< 20
z’-Factor	> 0.5
Binary
Shape of reference item (sigmoidal?)	yes
The final concentration-response curve of the reference item is composed of minimum six concentrations from three replicates	yes
The final concentration-response curve of the test item is composed of minimum six concentrations from three replicates	yes

Table 4: Acceptance criteria for DIO1 SK assay. This table lists the acceptance criteria for assay validity such as IC₅₀ range, coefficient of variation, IRA values for controls, z’-factor, and requirements for concentration-response curves. The acceptance windows were derived from previously reported DIO1‑SK validation studies⁶, including a reproducibility assessment based on five independent runs that informed the reference IC₅₀ window. In an example dataset of 10 independent runs, z′‑factor performance ranged from 0.544 to 0.774 (mean ± SD: 0.658 ± 0.062), supporting the z′ acceptance threshold used here.

Category (by efficacy)	Subcategory (by potency)	Inhibition activity	Threshold
Category 1: full inhibitor	A: Potent full inhibitor	Fully inhibits DIO1 activity at a concentration comparable to the reference item 6-PTU	Max. inhibition greater 90% and IC50 at or below upper range IC50 of 6-PTU
	B: Weak full inhibitor	Fully inhibits DIO1 activity at higher concentrations than the reference item 6-PTU	Max. inhibition greater 90% and IC50 above upper range IC50 of 6-PTU
Category 2: partial inhibitor	-	No full inhibition, but greater than negative and solvent control	Max. inhibition between 25% and 90%
Category 3: not inhibitor	-	Does not inhibit DIO1 (same as negative and solvent control)	Max inhibition below 25%

Table 5: Classification of test items based on DIO1 inhibition potency and efficacy. Test items are categorized into three efficacy groups (full, partial, or non-inhibitor) based on their maximum inhibition of DIO1 activity. Full inhibitors are further subcategorized by potency, depending on their IC₅₀ relative to the reference item 6-PTU.

Discussion

The DIO1-SK assay provides a non-radioactive, cost-effective, and scalable alternative to traditional methods for assessing DIO1 activity, such as radiolabeled assays or mass spectrometry. Its prevalidation for human liver microsomes and strong concordance with SPE-LC-MS/MS measurements of thyroid hormone levels confirm its reliability and regulatory relevance, as reflected by its inclusion in the OECD test guideline workplan. The simplicity of the Sandell-Kolthoff (SK) reaction makes the assay accessible to laboratories without advanced instrumentation.

In the present protocol, several practical aspects were specified in greater detail compared with previously published DIO1-SK procedures to further support robustness and reproducibility in a routine 96-well format. These clarifications focus on the rationale for the control concept (solvent, negative and positive controls), the decision to use freshly prepared ion-exchange resin plates rather than re-using previously cast plates, and SK reaction conditions that improve handling robustness in multi-well plate measurements, while keeping the biological interpretation consistent with prior standardization work^6,7.

Despite its strengths, several critical steps must be addressed to accurately quantify the iodide released during the DIO1-mediated conversion of rT3 to T2. This is achieved via the Sandell-Kolthoff reaction, which requires precise timing and consistent reagent handling to ensure reproducibility. The SK reaction is a rapid catalytic process in which iodide accelerates the reduction of yellow cerium (IV) to colorless cerium (III) by arsenite; the rate of color change is directly proportional to the iodide concentration. Because the reaction proceeds quickly once the arsenic solution is added, it is essential that the measurement is started without delay. In a 96-well plate format, slow or inconsistent addition of the arsenic solution can introduce significant variability between wells. Therefore, the use of a multichannel pipette or an automated dispenser is strongly recommended to ensure uniform and rapid reagent addition across all wells.

To maximize comparability across wells and between plates, the reagent addition order in the SK reaction should be kept consistent with the described workflow: Cerium is present as the chromogenic substrate, and the reaction is initiated by arsenite addition, which defines the start time and therefore requires rapid, synchronized dispensing. Alternative reagent addition orders were not evaluated in this setup; thus, changing the order is not recommended when reproducing the protocol. In addition, the cerium concentration can influence the practical performance characteristics of the SK readout (e.g., signal window and robustness). While 25 mM cerium was defined as a validated standard condition during earlier assay standardization, higher concentrations such as 40 mM were shown to increase signal window and z′-factor without affecting IC₅₀ values or biological interpretation, as already described in previous work⁶. In this protocol, 40 mM cerium is used to ensure complete and rapid iodide complexation under the assay conditions and to improve robustness and signal stability in routine plate-based measurements. The acceptance criteria (e.g., z′-factor, IC₅₀ values of the reference inhibitor) are based on historical performance observed in prior validation studies and is intended as a routine assay performance check.

The use of human liver microsomes introduces biological variability, making the standardization of microsome batch-specific iodide release activity essential. Careful preparation of ion exchange resin-filled filter plates is necessary to avoid uneven resin distribution across the filter plate and to ensure complete separation of iodide from potentially interfering substances. With respect to the resin workflow, freshly casting ion exchange resin-filled filter plates was chosen to minimize the potential risk of additional variability arising from resin saturation, iodide or protein carryover, and potential loss of binding efficiency upon reuse. Re-using resin and filter plates is technically feasible, however might harbour some risks that were not systematically evaluated in this study. It may introduce batch-to-batch drift if resin binding capacity is reduced or if residual matrix components remain in the resin bed after prior runs. For laboratories considering recycling, a pragmatic recommendation is to validate resin performance after reuse (e.g., by monitoring background and standard curve behavior and ensuring consistent iodide capture) and to discontinue reuse if increased background, reduced dynamic range, or inconsistent separation is observed. It is recommended to utilize resin-filled filter plates for single use, as this protocol has been specifically developed based on that methodology.

Additionally, the solubility of each test item must be determined in an appropriate vehicle, as the maximal concentration tested is limited by solubility (typically up to 1 mM with 1% DMSO). Because all test compounds are applied in DMSO as the solvent, the solvent control defines the baseline DIO1 activity under assay-relevant conditions and serves as the most appropriate reference condition for normalization and comparison. In contrast, 1-thio-β-D-glucose sodium salt was included as a negative control because it is structurally similar to the positive control aurothioglucose (ATG) but chemically inert with respect to iodothyronine deiodination and does not interfere with microsomal redox systems or DIO1 catalytic activity. Its inclusion helps distinguish true DIO1-specific effects from nonspecific effects related to compound addition, handling, or assay matrix interactions. A no-treatment/no-vehicle control was not used as the primary reference condition because it would not reflect the actual assay environment in which DMSO is present in all compound-treated wells; therefore, the solvent control is the relevant baseline for assay interpretation.

Troubleshooting focuses on identifying interferences in the SK reaction. Iodine-containing, chelating, or strongly oxidizing substances may artificially affect the colorimetric readout^14,17,18. To address this, the protocol includes interference testing without microsomes. Furthermore, the use of dithiothreitol (DTT) as a cofactor can also lead to non-enzymatic iodide release from certain test substances. In such cases, orthogonal methods like SPE-LC-MS/MS are recommended to confirm DIO1 inhibition. Nonspecific interference via general denaturing effects (e.g., surfactants) can generate false positives, so a general strategy to identify frequent hitters, using in silico tools or integrated testing approaches, can be valuable.

The assay has several limitations. It is susceptible to interference from test substances that react with assay components or release iodide independently of DIO1 activity. The assay does not distinguish between specific inhibition of DIO1 and unspecific interference with enzyme function or structural integrity. To improve specificity, inhibition of additional microsomal enzymes can be analyzed in parallel, and comparing inhibitory potencies across different enzyme assays may provide further evidence.

A further strategy for identifying possible artifacts is to include integrity or viability tests, as commonly performed in cell-based, in vitro assays, to define the maximum test concentration. While a direct integrity test is lacking for the DIO1-SK assay, its use within an in vitro test battery allows integration of cytotoxicity data from other cell-based assays to inform concentration selection. If such cross-assay information is not available, the upper concentration in the DIO1-SK assay is limited by solubility, which may not reflect relevant in vivo exposures. However, toxicokinetic approaches can be used to better define and interpret in vitro test concentrations.

The DIO1-SK assay is particularly valuable in toxicological screening and mechanistic studies of endocrine disruption. It addresses a key event in thyroid hormone homeostasis and can be integrated into in vitro test batteries. Its robustness, reproducibility, and adaptability to different laboratory settings and throughput needs make it a valuable tool for chemical safety assessment and regulatory validation. The assay provides a robust in vitro measure of DIO1 inhibition under standardized conditions and supports classification of test substances relative to a well‑characterized reference inhibitor (6‑PTU). However, in vivo data on DIO1 inhibition are scarce, and direct correlation of in vitro DIO1 inhibition outcomes with in vivo effects is therefore not currently applicable for estimating in vivo predictivity. Accordingly, assay results should be interpreted as mechanistic evidence for interference with one key event in thyroid hormone metabolism and integrated with additional lines of evidence (e.g., other thyroid MoA assays and orthogonal measurements where needed). Importantly, the DIO1–SK assay has been evaluated in multiple validation-oriented studies (optimization/standardization, predictivity assessment, and robustness/accuracy analysis)^6,7,15, and as the method progresses toward OECD implementation, the present article provides a level of step-by-step procedural detail intended to facilitate transfer and consistent implementation across additional laboratories.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
1-Thio-β-D-glucose sodium salt	Sigma-Aldrich	T6375-1G	CAS 10593-29-0
3,3’,5’-triiodothyronine (rT3)	Santa Cruz Biotechnology	sc-209692	CAS 5817-39-0
6-Propyl-2-thiouracil (6PTU)	Supelco	P3755-10G	CAS 51-52-5
Acetic acid	Thermo Fisher	9526-33	CAS 64-19-7
Analytical balance	NA	NA	capable of accurately weighing up to 30 g with 0.1 mg readability
Arsenic sodium oxide (NaAsO2)	Thermo Fisher	041533.AP	CAS 7784-46-5
Assay plates (96-well format)	Merck	Z707902-108EA	e.g. tissue culture plates, 96-well plate, flat bottom, polystyrene, 0.34 cm2, sterile, 108/cs, TPP
Aurothioglucose (ATG)	Sigma-Aldrich	A0606-5MG	CAS 12192-57-3
Centrifuge Tubes 15 and 50 mL	Merck	Z707724-800EA, Z707716-320EA	e.g. TPP centrifuge tubes, volume 15 and 50 mL, polypropylene,
Centrifuge with swing-out rotor for microtiter plates	NA	NA	Should be high enough to fit a 96-deep well plate with 96-well filter plate on top (at least about 6 cm high)
Cerium (IV) ammonium sulphate (Ce(NH4)4(SO4)4)	Sigma-Aldrich	22269-100G-F	CAS 10378-47-9
CO2 Incubator	NA	NA	capable of keeping temperatures of 37°C, 5% CO2 and ≥90% humidity
Deep well plates (96-well format)	Supelco	575653-U	e.g. SPE 96-Deep Square Well Collection Plate, well volume 2 mL, polypropylene
Dimethyl sulfoxide (DMSO)	AppliChem	A3672,0250	CAS 67-68-5
Dithiothreitol (DTT)	Sigma-Aldrich	D9779-1G	CAS 3483-12-3
Ethylenediaminetetraacetic acid (EDTA)	Sigma-Aldrich	APO456787707-500G	CAS 6381-92-6
Filter plates (96-well format)	Merck	WHA77002810	e.g. Micro-Plates, 96-well, clear polystyrene, 800 µL, DNA Binding, Whatman
Gas-tight plate sealers	Thermo Fisher	232698	e.g. Sealing tape, polyester, sterile, Sealing tape, polyester, sterile, Nunc
HEPES	Pan Biotech	P05-01100	CAS 7365-45-9
Human liver microsomes	Biovit	X008070	e.g. INVITROCYP 150-Donor Pooled Human Liver Liver Microsomes
Iodide (IC standard)	Supelco	41271-100ML	e.g. Iodide standard for IC, 1000 mg/L in water, Sigma-Aldrich
Ion exchange resin like Dowex 50WX2	Roth	334.4	CAS 12612-37-2; Mesh 100-200
Microcentrifuge tubes 1.5 mL	Merck	EP0030120086-1PAK	e.g. Eppendorf® Safe-Lock microcentrifuge tubes, volume 1.5 mL, natural
Multichannel dispenser capable of delivering 50 to 300 µL per step	NA	NA	NA
Multichannel pipette capable of delivering 10 to 100 µL per step	NA	NA	NA
pH meter with electrode and calibration buffers	NA	NA	capable of reading +/- 0.1 pH units
Photometer for absorbance measurement	Tecan Trading AG	NA	e.g. Sunrise Absorbance Reader, INSTSUN-3
Pipets capable of delivering 1 to 10 µL	NA	NA	NA
Pipets capable of delivering 10 to 100 µL	NA	NA	NA
Pipets capable of delivering 100 to 1000 µL	NA	NA	NA
Pipets for higher volumes	NA	NA	serological pipettes, e.g. 10, 25, 50 mL
Plate shaker	Thermo Fisher	NA	e.g. Thermo Scientific H+P MONOSHAKE VORTEXER microtiter plate, directly controlled
Repeater pipette	NA	NA	NA
Sodium chloride (NaCl)	Sigma-Aldrich	S9888-25G	CAS: 7647-14-5, e.g. sodium chloride, ACS reagent, ≥99.0 %
Statistics software	NA	NA	Able to perform regression analysis that reflect assay characteristics and able to calculate inhibitory concentrations e.g. GraphPad Prims 8, GraphPad
Sulfuric acid (H2SO4)	Supelco	1007311000	CAS 7664-93-9
Volumetric flask	NA	NA	certified with defined volume