Chemiluminescence-based Assays for Detection of Nitric Oxide and its Derivatives from Autoxidation and Nitrosated Compounds

Summary

Here, we present protocols for detecting nitric oxide and its biologically relevant derivatives using chemiluminescence-based assays with high sensitivity.

Abstract

Nitric oxide (NO) activity in vivo is the combined results of its direct effects, the action of its derivatives generated from NO autoxidation, and the effects of nitrosated compounds. Measuring NO metabolites is essential to studying NO activity both at vascular levels and in other tissues, especially in the experimental settings where exogenous NO is administered. Ozone-based chemiluminescence assays allow precise measurements of NO and NO metabolites in both fluids (including plasma, tissue homogenates, cell cultures) and gas mixtures (e.g., exhaled breath). NO reacts with ozone to generate nitrogen dioxide in an excited state. The consequent light emission allows photodetection and the generation of an electric signal reflecting the NO content of the sample. Aliquots from the same sample can be used to measure specific NO metabolites, such as nitrate, nitrite, S-nitrosothiols, and iron-nitrosyl complexes. In addition, NO consumed by cell-free hemoglobin is also quantified with chemiluminescence analysis. An illustration of all these techniques is provided.

Introduction

Since Salvador Moncada and Nobel laureates Robert Furchgott, Louis Ignarro, and Ferid Murad identified nitric oxide (NO) as the previously known endothelial-derived relaxation factor (EDRF), the central role of NO has been established in several key mechanisms spanning throughout vascular biology, neurosciences, metabolism, and host response^{1,2,3,4,5,6,7}. Exogenous administration of NO gas has become an established treatment for respiratory failure due to pulmonary hypertension in the newborn⁸. Nitric oxide gas has also been investigated for treatment of respiratory syncytial virus (RSV) infection, malaria and other infective diseases, ischemia-reperfusion injury, and for prevention of acute kidney injury in patients undergoing cardiac surgery^9,10,11,12. The need for precise measurement techniques to assess the levels of NO, its metabolites, and those of its target proteins and compounds arises from both mechanistic and interventional studies.

Due to its high reactivity, NO may undergo different reactions depending on the biological matrix in which it is produced and/or released. In the absence of hemoglobin (Hb) or other oxy-hemoproteins, NO is oxidized almost completely to nitrite (NO₂^-).

2NO + O₂ → 2NO₂

NO₂ + NO → N₂O₃

N₂O₃ + H₂O → NO₂^- + H⁺

NO first undergoes autoxidation with molecular oxygen (O₂) to yield nitrogen dioxide (NO₂) and reacts with NO₂ itself to generate dinitrogen trioxide (N₂O₃). One molecule of N₂O₃ reacts with water (H₂O) to form two molecules of NO₂^- and a proton (H⁺)¹³. Within whole blood^14,15, NO and NO₂^- are rapidly converted to nitrate (NO₃^-) as these molecules react avidly with the oxidized heme groups of Hb [Hb-Fe²⁺-O₂ or oxyhemoglobin (oxyHb)] to yield NO₃^-. This reaction is coupled with the transition of the heme group to the ferric state [Hb-Fe³⁺ or methemoglobin (metHb)]:

Hb-Fe²⁺-O₂ + NO → Hb-Fe³⁺ + NO₃^-

The red blood cell (RBC) barrier and the space immediately adjacent to the endothelium are the main factors limiting this reaction and allowing a small portion of the NO released by the endothelium to act as EDRF^16,17. In fact, cell-free Hb in the circulation is known to disrupt vasodilation in experimental and clinical settings^17,18. Within the RBC, depending on oxygenation and NO₂^- concentration, a portion of NO reacts with deoxyhemoglobin (Hb-Fe²⁺) to form iron-nitrosyl Hb (Hb-Fe²⁺-NO or HbNO):

Hb-Fe²⁺ + NO → Hb-Fe²⁺-NO

In the RBC^15,17, NO₂^- can form Hb-Fe³⁺ by reducing Hb-Fe²⁺ leading to the release of NO, which in turn binds Hb-Fe²⁺-O₂ (preferentially) or Hb-Fe²⁺.

The generation of NO-derivatives should not be considered strictly unidirectional as NO can be regenerated from NO₂^- and NO₃^- in various tissues and by different enzymes (e.g., by intestinal bacteria or within mitochondria, particularly under hypoxic conditions)^19,20.

A variable amount of NO produced (or administered) leads to the downstream generation of S-nitrosothiols, mainly by thiol transnitrosation from N₂O₃ in the presence of a nucleophile creating a NO⁺ donor intermediate (Nuc-NO⁺-NO₂^-):

N₂O₃ + RS^- → RS-NO + NO₂^-

Another possibility for S-nitrosothiols generation is nitrosylation of oxidized thiols (NO reacting with an oxidized thiol):

RS• + NO → RS-NO

This mechanism and direct thiol oxidation by NO₂ might be possible only in very specific conditions which are described elsewhere²¹. S-nitrosothiols range from light molecules like S-nitrosoglutathione to large thiol-containing proteins. S-nitrosohemoglobin (S-NO-Hb) is formed by nitrosation of a thiol group of a conserved cysteine residue in the β-chain (β93C)²².

The generation and metabolism of S-nitrosothiols are part of important regulatory mechanisms. Examples include regulation of glutathione, caspases, N-methyl-D-Aspartate (NMDA), and ryanodine receptors^{23,24,25,26,27,28}. Previously considered as a major mediator of NO biology in vivo, nitrosated albumin (S-nitroso-albumin) seems to be a NO/NO⁺ transporter without any specific additional biological activity²⁹.

When measuring the concentration of NO and its derivatives from a specific biological sample within a biological matrix, it is important to consider characteristics such as acidity, oxygenation, temperature, and the presence of reagents. Examples include administered exogenous NO donors and, in the setting of acute inflammation, hydrogen peroxide (H₂O₂) reacting with NO₂ leading to generation of supernormal concentration of free radicals like peroxynitrite (ONOO^-)²¹. In addition to the analytical method that is employed, the preanalytical phase of sample preparation and storage is fundamental. Downstream reactions that do not represent the in vivo NO activity shall be predicted, considered, and blocked. A valid example is the instability of S-NO-Hb, requiring a dedicated treatment of blood samples when it is targeted for measurement²².

Chemiluminescence-based assays are the gold standard for detecting the levels of NO and its main metabolites [NO₂^-, NO₃^-, S-NO and iron-nitrosyl complexes (Fe-NO)] in any biological fluid, including tissue homogenates^30,31. These methods rely upon the chemiluminescence detector (CLD), a device that houses the reaction of NO with ozone (O₃), generating NO₂ in an excited state (NO₂•). Relaxation of NO₂• causes emission of a photon of light that is detected by a photomultiplier tube, generating an electric signal that is directly proportional to the NO content of the sampled gas mixture³². A simplified schematic of the CLD is represented.

Figure 1: Simplified schematic of a chemiluminescence detector for nitric oxide gas. Chemiluminescence-based detection of nitric oxide (NO) is the stoichiometric generation of one photon per NO gas molecule that is introduced in the chemiluminescence detector (CLD). The chemiluminescence reaction is obtained in a designated chamber supplied with ozone (O3) from an internal generator, which is kept at negative pressure by connection with an external pump, allowing continuous and constant inflow of sample gas. The generation of O₃ requires diatomic oxygen (O₂) that is supplied by a dedicated O₂ tank connected with the CLD (other manufacturers provide CLDs operating with ambient air). Within the reaction chamber, each molecule of NO gas contained in the sampled gas reacts with oxygen to yield one molecule of nitrogen dioxide in the activated state (NO₂*). By returning to its ground state, each NO₂* molecule emits one photon that is detected by a photomultiplier tube (PMT) located adjacent to the reaction chamber. The PMT with the associated amplifier and central processing unit produces a signal proportional to the photon count and the number of NO molecules in the reaction chamber. Please click here to view a larger version of this figure.

The sample inlet of the CLD can be connected to a glassware system containing a reaction chamber for liquid samples. The system is continuously purged with an inert gas such as nitrogen, helium, or argon, transferring NO from the reaction chamber to the CLD. Liquid-phase samples are injected through a dedicated membrane into the purge vessel.

Figure 2: Structure of a purge vessel for chemiluminescence-based detection of nitric oxide gas The purge vessel (right) allows for the detection of nitric oxide (NO) gas or any other compound that can be readily converted to NO gas when released from a liquid phase reagent. The inert gas inlet is connected to a source (tank) of an inert gas such as Argon, Xeon, or diatomic nitrogen (N₂). The needle valve (opens to the left) is used for pressure control within the purge vessel and can be completely removed to clean the vessel. The injection port is covered by a cap with a membrane septum for sample injection. The membrane should be replaced often. A heated jacket surrounds the reaction chamber and is connected to a hot water bath to perform the VCl₃ in HCl assay. The purge vessel outlet is connected to the chemiluminescence detector (CLD) or to the NaOH trap (required for VCl₃ in HCl assays). To drain the reaction chamber content, first close the stopcocks at the inert gas inlet and the purge vessel outlet, close the needle valve, remove the cap at the injection port and finally open the stopcock at the drain. The NaOH trap (left) is required to be placed inline between the purge vessel and the CLD if the VCl₃ in HCl assay is performed because of the corrosivity of HCl. The connection to the CLD always requires an intense field dielectric (IFD) filter to be placed between the CLD and the output of the purge vessel (or the NaOH trap, if used). The IFD filter removes airborne particles and stops liquid from passing through the purge vessel. PTFE = polytetrafluoroethylene. Please click here to view a larger version of this figure.

As a consequence, any compound that can be converted to NO through a specific and controlled chemical reaction can be detected with high sensitivity in any biological fluid and tissue homogenate²⁴. Direct measurement of gas-phase NO through chemiluminescence is performed in both experimental and clinical settings. These techniques are extensively described elsewhere^33,34,35. Measurement of NO₂^-, S-nitrosothiols, S-nitrosated proteins, and Fe-NOs can be performed by adding samples in a reaction mixture with triiodide (I₃^-), which stoichiometrically releases NO gas from all these compounds:

I₃^- → I₂ + I^-

2NO₂⁻ +2I⁻ +4H⁺ → 2NO + I₂ +2H₂O

I₃⁻ + 2RS-NO → 3I⁻ + RSSR + 2NO⁺

2NO⁺ + 2I⁻ → 2NO + I₂

while I₃^- does not react with NO₃^-15. Precise measurements of each compound are made possible by pre-treatment of sample aliquots with acidified sulfanilamide (AS) with or without mercuric chloride (HgCl₂). Specifically, pre-treatment with AS removes all NO₂^- content. As a consequence, the NO content measured by the CLD only reflects the sum of S-NOs and Fe-NOs concentration. Injection of HgCl₂ in a sample aliquot before AS injection causes NO₂^- to be released by S-NO. Treatment with AS (leading to NO₂^- removal) ensures that the measured NO content only reflects the concentration of Fe-NOs. A series of subtractions between the assessments allow to calculate the precise concentration of the three NO derivatives²².

Figure 3: Steps in sample preparation for the I₃^- in acetic acid chemiluminescence assay. The sequential steps for preparation of the I₃^- in acetic acid chemiluminescence assay are illustrated. Use of light-protected centrifuge tubes is required. Tubes 1, 2, and 3 are those used to prepare for the assay. Another sample aliquot (tube 4) is needed for the VCl₃ in HCl assay if the measurement of nitrate (NO₃^-) is required. Steps are indicated by numbers in red. Prefill (Step 1) as indicated with phosphate buffer saline (PBS) or HgCl₂ before adding the sample volume. Add the sample volume (2) as indicated, vortex, and incubate for 2 min at room temperature (RT). Add (3) PBS or acidified sulfanilamide (AS) as indicated,vortex, and incubate for 3 min at RT. Run the assay (4). The concentration measured by the assay is the sum of the concentration of the compounds reported under each tube. Tube number 1 will allow measurements of nitrite (NO₂^-), S-nitrosothiols (S-NO), and iron-nitrosyl complexes (Fe-NOs) as a single signal. For nitrate (NO₃^-) measurement, samples shall be run with both I₃^- in acetic acid and VCl₃ in HCl assays, and the value obtained from tube 1 should be subtracted from the one obtained from tube 4. *suggested quantities to be used for Hb analysis for determination of residual NO₂^-, S-nitrosohemoglobin and iron-nitrosyl-hemoglobin. Please click here to view a larger version of this figure.

For NO₃^- measurement, Vanadium (III) chloride (VCl₃) in hydrochloric acid (HCl) is used for conversion of NO₃^- to NO in the purge vessel in order to measure NO₃^- stoichiometrically with the CLD:

2 NO₃^-+ 3V⁺³ + 2H₂O → 2NO + 3VO₂⁺ + 4H⁺

To achieve a sufficiently fast conversion, the reaction needs to be performed at 90-95 °C. Reduction from NO₃^- to NO₂^- is coupled with reduction of NO₂^- to NO by HCl. Vanadium metal also reduces S-NOs liberating their NO moiety^22,36. The final concentration obtained by CLD with VCl₃ in HCl reflects the aggregate concentration of NO₃^-, NO₂, and other nitrosated compounds. Subtraction of the latter value from the concentration yielded with CLD with I₃^- allows for the calculation of NO₃^- concentration^36,37 (Figure 3).

In the NO consumption assay, the continuous release of NO in the purge vessel by NO donors like (Z)-1-[2-(2-aminoethyl)-N-(2-ammonioethyl)amino]diazen-1-ium-1,2-diolate (DETA-NONOate) generates a stable signal allowing quantifying cell-free oxyHb in the injected samples. The amount of NO consumed in the purge vessel is in a stoichiometric relationship with the amount of oxyHb in the sample³⁸.

Protocols for measurement of NO₂^-, NO₃^-, S-nitrosothiols, iron-nitrosyl complexes, and NO consumption by cell-free Hb in plasma samples are illustrated. Studies on NO in the RBC environment require specific sample treatment followed by exclusion chromatography to measure extremely fragile S-NO-Hb and Hb-NO coupled with the determination of total Hb concentration^15,22. Sample preparation is instrumental in correcting measurement. Pre-existence of NO₂^- in H₂O and release of NO₂^- during the assay can lead to measurement of artificially higher concentrations of NO derivatives such as S-NO-Hb^14,39. Important aspects of sample preparation are also presented.

Protocol

The procedures indicated in this protocol are in accordance with the review board of Massachusetts General Hospital. The blood samples that were used had been collected during a previous study and were de-identified for the current purpose¹⁸.

NOTE: See manufacturer's instructions for specific guidance regarding optimal connections between tubing and glassware constituting the purge vessel, washing, and general maintenance. Connections need to be firm and carefully made not to damage the glassware. Identify the components of the glass purge vessel: gas inlet line, purge vessel with heating jacket and condenser, sodium hydroxide (NaOH) gas bubbler trap, connection line between the purge vessel and the bubbler, purge vessel outlet gas line (to CLD) endowed with an intense field dielectric (IFD) filter. An IFD filter line between the purge vessel and the sample inlet of the CLD must be in place every time NO metabolites in liquid form (plasma, cell cultures, tissue homogenates) are measured (all the assays presented in the protocol). Sample preparation depends on the fluid or tissue that is analyzed and on the compounds of interest. Important aspects of the preanalytical phase are covered in section 1 and 2. Specific preparation steps for specific assays are included in sections 3-5. Sections 6-8 apply to all assays.

1. Preparation of dedicated reagents

NOTE: For more details, refer to previous publications^15,22.

1. Prepare 5% (290 mM) acidified sulfanilamide (AS) solution for NO₂^- removal by dissolving 500 mg of sulfanilamide into 10 mL of 1N HCl. This solution is stable for months.
2. Prepare 50 mM mercuric chloride (HgCl₂) solution for NO₂^- release from S-NOs by dissolving 67.9 mg of HgCl₂ into 5 mL of PBS. Protect the stock solution from light.
3. Prepare NO₂^- blocking solution with 800 mM ferricyanide [K₃Fe(CN)₆] to oxidize Hb together with 100 mM N-ethylmaleimide (NEM) to block thiol-groups and (OPTIONAL) 10% nonylphenyl-polyethylene glycol solution (Nonidet p-40) to solubilize red cell membranes.
   1. Add K₃Fe(CN)₆ to deionized and distilled water (ddH₂O, 263.5 g of powder per liter) to obtain a final concentration of 800 mM.
   2. Add NEM to the 800 mM K₃Fe(CN)₆ in ddH₂O solution (12.5 g of powder per liter to have 100 mM concentration) and mix the solution to dissolve all crystals.
   3. Add one part of 10% NP-40 to nine parts of the 800 mM NEM K₃Fe(CN)₆ 100 mM NEM solution (111 mL per liter) and mix well (mandatory step for whole blood analysis).
4. Prepare the S-NO-Hb stabilization solution containing 12 mM K₃Fe(CN)₆, 10 mM NEM, 100 µM diethylenetriaminepentaacetic acid (DTPA, for metal chelation), and 1% Nonidet p-40 detergent from their stock solutions.
   1. Prepare a 200 mM NEM solution by adding NEM crystals to PBS (25 mg/mL, e.g., 250 mg in 10 mL PBS) and mix the solution until all crystals are dissolved (to be made on experiment day).
   2. Prepare a 800 mM K₃Fe(CN)₆ solution by adding K₃Fe(CN)₆ to ddH₂O (263.5 mg/mL e.g. 1.32 g in 5 mL of ddH₂O to obtain final concentration of 800 mM) (to be made on experiment day).
   3. Prepare a 10 mM DTPA stock solution by adding 786 mg of DTPA to 200 mL of ddH₂O and adjust pH to 7.0 with 5N NaOH to fully solubilize DTPA.
   4. Add 5 mL of 200 mM NEM solution, 1.5 mL of 800 mM K₃Fe(CN)₆ solution and 1 mL of DTPA stock solution to 81.5 mL of PBS at 7.2 pH, and finally, add 11 mL of 10% NP-40 at last to bring final volume to 100 mL.

2. Sample collection

NOTE: For more details on sample collection, refer to previously published works^15,22,40.

1. Collect whole blood
   1. Collect blood in heparin-coated tubes preferring venous over arterial vessel (unless specifically required) and preferring catheter placement over venipuncture (if possible) with catheter or needle bore of at least 20 G or larger to minimize hemolysis.
   2. Immediately add the NO₂^- blocking solution (1 part of the solution to 4 parts of whole blood), process (sections 3 or 4), or freeze and store at -80 °C.
2. Collect plasma and red blood cells (RBCs)
   1. Collect blood in heparin-coated tubes preferring venous over arterial blood (unless specifically required) and needles of at least 20 G or larger to minimize hemolysis and centrifuge immediately for 5 min at 4000 x g at 4 °C.
   2. Mix the supernatant (plasma) with the NO₂^- blocking solution (1 part of the solution to 4 parts of the supernatant) in a new tube and process (sections 3 or 4), or freeze and store at -80 °C.
      NOTE: For the NO consumption assay, the NO₂^- blocking solution cannot be used. The assay can be performed without plasma pre-treatment.
   3. Resuspend the RBC pellet from the bottom to a new tube pre-filled with S-NO stabilization solution (see step 1.4) (1 mL of pellet to 9 mL of solution) and incubate for 5 min.
   4. Pass the RBC lysate in a sizing column with G-25 Sephadex polymer that has been previously rinsed with ddH₂O for exclusion chromatography
   5. Collect the Hb fraction to process (section 4) and to measure Hb concentration using Drabkin's reagent (for Hb measurement, refer to previously published work²²).
      NOTE: To prepare and sample a specific tissue/organ, identify its hilum, and surgically isolate it. Incise the vein, puncture the artery, and inject with heparinized saline (10 U/mL) via the artery. Excise the tissue when saline starts to backflow at the venous incision. Homogenize the tissue with a mechanic homogenizer while adding 1 part of NO₂^- blocking solution to 4 parts of homogenized tissue.

3. VCl₃ in HCl assay preparation

NOTE: For more details on VCl₃ in HCl assay preparation, refer to previously published works^37,41.

CAUTION: The CLD will be damaged if the NaOH trap is not properly in place when performing this assay. This is due to the corrosivity of HCl.

1. Prepare standard solutions of NO₃^- for standard curve
   1. Dissolve 85 mg of NaNO₃ in 10 mL of ddH₂O to obtain 0.1 M NaNO₃^- (this solution remains stable for a few weeks).
   2. Use the stock solution to prepare standards by dilution in ddH₂O to obtain concentrations of 5 µM, 10 µM, 20 µM, 40 µM, 80 µM, 200 µM NaNO₃ in order to perform a calibration curve for plasma or urine samples (use lower concentrations if working with cell cultures).
2. Prepare VCl₃ (vanadium chloride) saturated solution for NO₃^-​ reduction in the purge vessel
   CAUTION: The reaction of water and VCl₃ is exothermic. Pay attention to high glassware temperature when adding acid and when rinsing the glassware at the end of the experiment.
   1. Dissolve 1.6 g of VCl₃ in 200 mL of 1 M HCl by first adding VCl₃ in a clean flask then adding 200 mL of 1 M HCl.
   2. Vacuum filter the solution through filter paper (such as 11 μm filter paper, but any filter paper can be used).
      NOTE: The filtered solution shall turn to clear blue, while the unfiltered VCl₃ solution is brown because of undissolved particles.
   3. Keep the saturated solution covered with either aluminum foil or polytetrafluoroethylene (PTFE) tape as the compound is light sensitive.
3. Prepare the circulating water bath
   1. Connect a circulating water bath device to the water jacket of the purge vessel. Make sure that the lines are dry before priming.
   2. Start the water bath at 95 °C and verify the absence of leaks on the water lines by applying (nonadherent) paper towels around the lines.
4. Set up the gas bubbler trap
   1. Verify that the PTFE sleeve of the bubbler is in place and that it is not damaged.
   2. Open the gas bubbler and inject 15 mL of 1 M NaOH into the bubbler base.
   3. Reposition the gas bubbler and seal the connection tightly by pressing the bottom towards the top and slightly twisting the two parts. The impossibility of turning the top of the bubbler without applying force indicates a correct seal.
   4. Connect the outlet of the purge vessel to the inlet of the gas bubbler trap.

4. I₃^- in acetic acid assay preparation

NOTE: For more details on I₃^- in acetic acid assay preparation, refer to previously published works^{15,22,38,41,42}.

1. Prepare standard curve for NO₂^-
   1. Prepare a stock solution by dissolving 69 mg of sodium nitrite (NaNO₂) in 10 mL of ddH₂O to obtain 100 mM solution. This solution is stable if stored in an airtight container, refrigerated, and protected from light.
   2. Serially dilute the stock solution into 1.5 mL microcentrifuge tube pre-filled with 900 μL of ddH₂O: Add 100 μL of the stock solution to the first centrifuge tube, mix, label, and use 100 μL of the tube for the second tube, then repeat resulting in 10 mM, 1 mM, then 100 μM.
   3. Further dilute with ddH₂O to obtain 50 μM, 25 μM, 10 μM, 1 μM and 500 nM NaNO₂ aliquots to be used in the calibration curve.
2. Prepare the I₃^- in acetic acid for the purge vessel (can be stored at room temperature (RT) for 1 week)²²
   1. Add 2 g of potassium iodide (KI) and 1.3 g of iodine (I₂) to 40 mL of ddH₂O and 140 mL of acetic acid.
   2. Mix thoroughly by stirring the mixture for at least 30 min.
3. Prepare the samples for differential determination of NO₂^-, S-nitrosothiols (S-NO-Hb, if Hb is collected) and iron-nitrosyl complexes (Hb-NO if Hb is collected) (Figure 3)
   1. Divide each sample in 3 aliquots of 270 μL (900 μL of Hb if measuring S-NO-Hb and Hb-NO) in light-protected microcentrifuge tubes, 2 of them pre-filled with 30 μL of 1x PBS (100 μL if measuring S-NO-Hb and Hb-NO) and the third tube with 30 μL of HgCl₂ (100 μL if measuring S-NO-Hb and Hb-NO), vortex and incubate at RT for 2 min (Figure 3).
   2. Add 30 μL of 5% AS to the sample with HgCl₂ to measure Fe-NOs, (100 μL for Hb-NO) and to one with added PBS to measure S-NOs and Fe-NOs (100 μL for S-NO-Hb and Hb-NO) and add 30 μL of PBS to the third one already pre-filled with 1x PBS to measure NO₂^-, S-NOs and Fe-NOs (100 μL for residual NO₂^- from Hb collection, S-NO and Hb-NO). Vortex and incubate at RT for 3 min (Figure 3).

5. NO consumption by cell-free Hb setup

NOTE: For more details, refer to the previously published work³⁸.

1. Prepare standard oxyHb solutions from a purified stock Hb solution with a known concentration
   1. Serially dilute the stock solution into 1.5 mL microcentrifuge tubes by addition of ddH₂O to obtain the solutions that will be used for the calibration curve: 62 μM, 50 μM, 25 μM, 12.5 μM, 6.25 μM, 3.125 μM, 1.56 μM.
2. Prepare the DETA-NONOate solution
   1. Add 10 mg of DETA-NONOate to 610 µL of 10 µM NaOH in pH 7.4 PBS to generate 100 mM of DETA-NONOate and keep it on ice.

6. Start the chemiluminescence detector (CLD) and prepare the purge vessel

NOTE: For the preparation of the purge vessel, refer to the previously published work⁴³.

1. Verify the main connections to and from the CLD
   1. Connect the oxygen line to the CLD and open the oxygen tank at a pressure that is in agreement with the CLD's manufacturer.
   2. Make sure that the Intense Field Dielectric (IFD) filter line is connected to the CLD but not to the purge vessel or the NaOH trap
2. Start the CLD
   1. On the CLD interface, start running the detection program for liquid phase assays.
   2. Verify that the oxygen supply is adequate. If this is the case, the CLD will successfully start sampling from its inlet and indicate detection by a signal in millivolts (0-5 mV). Otherwise, the CLD will prompt a negative diagnostic signal.
3. Prepare the purge vessel
   1. Close the purge vessel on all three ports: fully screw the needle valve to the right, close the inlet and outlet stopcocks.
   2. Remove the cap from the purge vessel and add a sufficient quantity of the reagent specific to the planned assay to the reaction chamber (Table 1) so that the syringe needle used to inject the samples can reach the fluid column.
   3. Verify the presence of a stable desired baseline (Table 1).
4. Start the purge gas flow
   1. Ensure that the inert gas tank (e.g., N₂) is equipped with a two-stage regulator and connect the inert gas tank with the gas inlet of the vessel.
   2. Open the gas with an outlet pressure at the regulator of 1-5 psi, open the inlet of the purge vessel and slowly open the needle valve of the purge vessel to allow inflow of gas. Verify bubbling within the purge vessel.
5. Adjust the gas flow
   1. Record the cell pressure measured by the CLD with the IFD filter line sampling ambient air.
   2. Reposition the cap on the purge vessel, connect the IFD filter line to the purge vessel (or to the NaOH trap in the VCl₃ in HCl assay), and open the outlet of the purge vessel.
   3. Use the needle valve to reach the same cell pressure at the CLD level that is recorded in ambient air.

7. Experiment

NOTE: For more details regarding the experiment, refer to the previously published work⁴³.

1. Start the chemiluminescence signal acquisition program
   1. Connect the serial port of the CLD to the computer's serial port into which the acquisition program has been installed.
   2. Run the analysis program.
   3. Click on Acquire, select the folder to save the .data file, type in the file name, and click on Save.
      NOTE: Notice the preset run time on screen, as the recording stops automatically when the preset time elapses. If needed, the preset running time can be increased.
2. Prepare for repeated sample injections
   1. Adjust the voltage scale on the screen to have control over the targeted baseline by clicking on the Minimum and/or Maximum buttons and then entering the desired value.
   2. Have a 20- or 50-mL tube filled with ddH₂O to rinse the syringe between samples.
   3. Have a box of delicate task wipes readily available.
3. Sample injection
   NOTE: Start from the standard solutions for the calibration curve (inject from the least concentrated to the most concentrated samples), then proceed to the experiment samples (consider doing so in duplicates or triplicates).
   1. Rinse the syringe at least twice or more with ddH₂O before withdrawing each sample (and after each injection) and verify every time unobstructed water ejection on a task wipe.
   2. Insert the syringe in the sample tube while holding both the syringe and the tube at a close distance, pull up the plunger to the desired volume while ensuring no air bubble and/or non-homogenized solid parts are trapped.
   3. Clean the tip of the syringe with a task wiper, then insert the syringe into the septa cap at the injection port and inject after verifying that the tip of the syringe is within the liquid phase in the reaction chamber.
4. Mark the injection in the software program and wait
   1. Verify that the injection causes an upwards change in the signal (Supplemental Figure 1) (downwards in the NO consumption by cell-free Hb assay) and type the sample name by clicking the grey box under Sample Names, then click on Mark Injection.
      NOTE: Suspect syringe obstruction if the sample injection does not generate a signal.
   2. Wait for the electric signal to reach baseline again (this usually takes 3-4 min). This time can be used to perform step 7.3.1.
5. Repeat all steps indicated in steps 7.3 and 7.4 during and after every injection until the end of the experiment. Remember to run a sample of the preservation solution (if used)
6. Stop the experiment
   1. Click on STOP to interrupt the signal acquisition, stop the CLD and turn off the water bath (if NO₃^- is measured).
   2. Interrupt the gas flow, open the needle valve, remove the cap from the purge vessel, place a waste container under the drain and open the draining stopcock.
      ​NOTE: If the experiment requires data acquisition for longer than 60 min, it is necessary to restart the acquisition after 60 min of running time (repeat step 7.1.3) and make a new file.

8. Measurements and calculations

NOTE: Measurements and calculations are made offline and can be performed at a different time.

1. Start the chemiluminescence acquisition program for offline data analysis
   1. Start the program and click on Process.
   2. Select the experiment file, then click on Open.
2. Calculate the area under the curve for each administration
   1. The software automatically graphs on screen the baseline (Supplemental Figure 2A, horizontal yellow line) and the peak axis of each wave generated by each sample administration (vertical yellow lines): verify their correct position (or adjust by clicking on each line and moving it with the mouse or the arrows) and click on Threshold OK (Supplemental Figure 2B).
      NOTE: In the NO consumption by cell-free Hb measurement assay, the software does not typically manage to correctly capture the waveform generated by sample injection. By zooming on each waveform, the operator can easily assist the software in the area calculation (Supplemental Figure 2).
   2. The software automatically graphs the beginning (vertical green line) and the end (vertical red line) of each peak caused by each sample administration: verify their correct position (or adjust by clicking on each line and moving it with the mouse or the arrows) and click on Integrate (Supplemental Figure 2C).
      NOTE: Some areas in the trace may be mistakenly defined as injections at this point, and some peaks may be automatically counted twice. Both mistakes can be re-identified and removed during step 8.2.2
   3. The software automatically matches every signal area following an injection marked during the experiment and its assigned name: navigate each peak (indicated by a yellow vertical line) with the assigned name by clicking on Next Peak and Previous Peak, then click on the button All OK to finally obtain the calculation for all areas on the screen.
   4. In order to correct all naming or matching mistakes made by the user or by the program, use as needed the buttons indicated in Supplemental File 1.
3. Transfer the calibration curve values to a spreadsheet and generate a linear regression equation (Supplemental Figure 3)
   1. Transfer the data from the CLD program to a new spreadsheet through copy-paste. Arrange the two columns on the datasheet as Sample Concentration and Area Under the Curve, and add a matched value of zero on both columns.
   2. Select the two columns, click on Insert > Scatter, then under the Chart Design menu, select Add Chart Element > Trendline > Linear.
   3. Right-click on the generated trendline, click on Format Trendline, then click the options Display Equations on Chart and Display R-squared value on Chart on the Format Trendline menu to obtain a simple linear calibration equation.
4. Transfer the calculated area of each sample to calculate its concentration (Supplemental Figure 3)
   1. Report every value on the spreadsheet. In the next column, apply the equation obtained from step 8.3.3 to obtain the concentration of each injected sample, where y is the concentration (value of the new column) and x is the area under the curve measured after injection.
      NOTE: Remember to take into account the concentration measured in the preservation solution (if used) and subtract the values accordingly.

Representative Results

The NO-consumption by cell-free Hb assay was used in samples containing known concentrations of cell-free oxyHb (Figure 4). As one heme of oxyHb stoichiometrically releases one NO molecule in the assay, purified cell-free oxyHb is used to build the calibration curve for the assay (Supplemental Figure 3).

The dose-response relationship between cell-free Hb (measured with a colorimetric assay) and NO consumption in patients coming off cardiopulmonary bypass (CPB) during cardiac surgery is shown in Figure 5. It can be assumed that after CPB there is a higher concentration of heme groups in the Hb-Fe²⁺-O₂ status (Figure 5, red) compared to patients receiving NO during CPB from which only a minority of heme groups of cell-free Hb consumes NO (Figure 5, green).

The protocol for assessing NO consumption by cell-free Hb was previously applied to test plasma samples from 50 patients undergoing cardiac surgery and requiring CPB. Blood was collected at baseline and 15 min, 4 h, and 12 h after CPB. The purpose of the study was to investigate the existing relationship between both pulmonary and systemic vascular resistance and the increase in hemolysis observed after the CPB. Cell-free Hb concentration was assessed with a Hb colorimetric assay. Hemolysis peaked 15 min after CPB. The accumulated free Hb was slowly eliminated from plasma within 12 h¹⁸ (Figure 6A). NO consumption, instead, peaked at 15 min and reached baseline values in just 4 h (Figure 6B). Linear regression curves of NO consumption by cell-free Hb exhibited a stoichiometric relationship at the 15 min post-CPB timepoint as opposed to baseline, 4 h, and 12 h post-CPB (Figure 6C). The explanation for the observed difference in kinetics lies in the abundance of newly released free Hb that has not yet encountered NO molecules being released by the patient's endothelium. The concentration of oxyHb (scavenging NO and resulting in consumption in the assay) was, in fact, higher at 15 min than at any other timepoint. Plasma collected at baseline and on other timepoints had a relatively higher component of metHb, which had already been oxidized by NO, did not bind NO, and did not cause consumption in the assay. Pulmonary and systemic vascular resistances were measured invasively and peaked at 15 min. For the first time in this specific patient population, NO consumption by free Hb was observed to be temporally coupled with vasoconstriction, which confirmed previous observations in animal models and in patients with sickle cell disease¹⁸.

The administration of NO gas during CPB and after surgery increases the relative quantity of circulating metHb vs. oxyHb. When patients undergoing cardiac surgery were treated with NO-gas intra-operatively, and post-operatively, the observed increase of free Hb after CPB (Figure 7A) was not coupled with any increase in NO consumption, as measured with the aforementioned protocol (Figure 7B). The exogenously administered NO gas converts most oxyHb into metHb, which in turn does not scavenge NO in vivo nor consumes NO in vitro (unpublished data).

Figure 4: Nitric oxide consumption by purified cell-free oxyhemoglobin. Nitric oxide consumption assay performed by injecting samples of known concentrations of purified cell-free oxyhemoglobin. Regression line is shown in red. Please click here to view a larger version of this figure.

Figure 5. Nitric oxide consumption and cell-free hemoglobin concentration from patients undergoing cardiac surgery with cardiopulmonary bypass. Each patient's cell-free hemoglobin level, as measured with a colorimetric kit, was plotted with the nitric oxide (NO) consumption measured through chemiluminescence. Blood samples were drawn 15 min after the end of CPB. Regression lines are shown for cardiac surgery patients not receiving (in red) or receiving (in green) NO during cardiopulmonary bypass (CPB). Hemoglobin is expressed in μM of heme groups. Please click here to view a larger version of this figure.

Figure 6: Nitric oxide consumption and cell-free hemoglobin in cardiac surgery patients. (A) Cell-free hemoglobin (Hb) concentration in the plasma of patients undergoing cardiac surgery with cardiopulmonary bypass (CPB) (N = 50) was determined at different time points with the use of a commercially available colorimetric determination kit. Data were analyzed by fitting a mixed model with evidence of a fixed effect between time points. Plasma concentration at different time points was compared to baseline with Tukey's multiple comparison test. (B) Nitric oxide (NO) consumption by plasma obtained from the same samples used for quantification of free Hb. Data were analyzed by fitting a mixed model with evidence of a fixed effect between time points. Plasma concentration at different time points was compared to baseline with Dunnett's multiple comparison test. (C) Regression lines for cell-free Hb plasma content matched with NO consumption. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, ^ns not significant. Data are presented as median ± interquartile range. Please click here to view a larger version of this figure.

Figure 7: Effects of nitric oxide gas administration on nitric oxide consumption in patients undergoing cardiac surgery. (A) Cell-free hemoglobin concentration in the plasma of patients undergoing cardiac surgery with cardiopulmonary bypass (CPB) either receiving a placebo (N = 28) or nitric oxide (NO) gas (N = 22) for 24 h since the beginning of CPB. The concentration was determined at different time points with the use of a commercially available colorimetric kit. Data were analyzed by Friedman's test, indicating the presence of an effect between time points in both groups. Plasma concentration at different time points was compared to baseline with Dunn's multiple comparison test. (B) NO consumption by plasma obtained from the same samples used for quantification of cell-free Hb. Data were analyzed by Friedman's test, indicating the presence of an effect between time points in both groups. Plasma concentration at different time points was compared to baseline with Dunn's multiple comparison test. ** p < 0.01, *** p < 0.001. Data are presented as median ± interquartile range. Please click here to view a larger version of this figure.

Assay	Reagent Mixture	Target Baseline (mV)
VCl3 in HCl	5 mL of VCl3 in HCl saturated solution + 100 μL of antifoam agent	0–5 mV*
I3– in acetic acid	5–7 mL of I3– reagent	0–5 mV*
NO consumption by cell-free Hemoglobin	5–7 mL of PBS at pH 7.4 + 100 μL of antifoam agent + 20 μL of expanded DETA-NONOate solution	70–100 mV**
*ideal value: baseline might be higher due to the quality of the air in the experiment environment
**ensure stability for 20 min before starting the experiment.

Table 1: Purge vessel reagent composition and target baseline for various chemiluminescence assays. Purge vessel reagent composition for each described assay. NO = Nitric Oxide; PBS = Phosphate Buffer Saline; DETA-NONOate: (Z)-1-[2-(2-aminoethyl)-N-(2-ammonioethyl)amino]diazen-1-ium-1,2-diolate. *ideal value: baseline might be higher due to the quality of the air in the experiment environment. **ensure stability for 20 min before starting the experiment. To perform the NO consumption assay, the scale should be changed on the software to capture values between 70 mV and 100 mV (e.g., 0-100 mV), which is the targeted baseline voltage. This adjustment is suggested to have full control on the sample injection, which enables to document the signal moving from baseline following the injection of a sample and to observe the signal returning to baseline. Data recording is not influenced by nor limited to the visualized scale.

Supplemental Figure 1: Peaks following NO detection in the VCl₃ in HCl assay. The peaks which follow sample administration in the VCl₃ in HCl are shown on the CLD bundle software. Calculation of the area under the peak for each injection is shown for convenience and explained in Supplemental Figure 2. In this experiment, standards of different concentrations have been administered for the calibration curve, followed by plasma samples. Dilution is a useful solution to calculate metabolites from samples that yield NO at the high end or even outside of the calibration curve. Please click here to download this File.

Supplemental Figure 2: User-assisted automated calculation of the area under the peak. When performing the NO consumption assay by cell-free hemoglobin, it may be necessary to set peaks manually in order to provide the software with the most precise limits to measure the area under the peak generated by each sample injection. (A) Representative full view of a recorded experiment. It is reached by clicking on Process in the main menu and by selecting and opening the recorded file of interest. The yellow line represents the signal threshold (mV) that constitutes the straight segment of the area under the peak. By clicking on it, the user can drag it towards the signal area. The x and y-axis can be changed to zoom on the peak of interest for precise measurement by either clicking on the lower left buttons or by typing in the desired minimum and maximum value for each. (B) Threshold positioning. The user can select the optimal baseline threshold by dragging the yellow line to the signal midpoint right prior to the signal downslide caused by the sample injection. To continue limiting the area, the user shall click on the Set Peaks Manually button. (C) Start and End Cursors positioning. By clicking on the Set Start Cursor button, a green line appears on the screen. The green line should be placed (by dragging and releasing) at the very beginning of the downslide caused by the sample injection. The same button changes its label to Set End Cursor by clicking on which a red line appears on the screen. The red line should be placed at the point where the signal reaches the baseline again after the deflection caused by NO consumption from the injected sample. Altogether, the threshold line (yellow, B), the green line, and the red line precisely outline the area generated by the sample injection. (D) Yielding the area under the peak. By clicking on the Integrate button (C), the calculated area under the peak is displayed on the top right of the screen. Please click here to download this File.

Supplemental Figure 3: Representative spreadsheet with calibration curve and results. The calculated areas under the peaks generated from nitric oxide (NO) consumption by injection of samples with a known concentration of cell-free hemoglobin (expressed as [µM] of heme) are shown on the top left (on yellow background). The values are plotted to build the calibration curve and its resulting linear equation y= mx + q. The slope (m) is the number of oxyHb heme groups (µM) required to decrease the signal detected by the photomultiplier tube by 1 mV. Calculated values from triplicate injections (Run N 1,2,3) of samples from one patient obtained at four different time points (marked by different background colors) were entered in the spreadsheet. The average of each triplicate was used to yield the value of NO consumption caused by each sample by solving the linear equation: Please click here to download this File.

Supplemental File 1: Reviewing the peaks using the bundled software. This file includes a screenshot taken from the bundled software and the actions allowed for each button. For every peak (including false peaks caused by artifacts), the user can use the buttons indicated in the file to confirm, or disregard, or delete, or name a specific peak, as needed. The user can also record a peak that was not previously detected. Please click here to download this File.

Discussion

Due to the high sensitivity, chemiluminescence-based assays for the determination of NO and related compounds have a high risk of NO₂^- contamination. Each reagent (especially the NO₂^- blocking solution) and dilutant (including ddH₂O) used in the experiment should be tested for its NO₂^- content to correct for background signal. Nitrite is extremely reactive with a half-life in whole blood around 10 min and rapidly generates NO₃^-. The time that elapses between blood collection and centrifugation, or NO₂^- blocking, may therefore cause preanalytical errors and should be minimized¹⁵. In an assay using plasma samples, collect blood preferably in heparin tubes and centrifuge at 750 x g for 5 min at 4 °C after thiol group blockade with 40 μL of 200 nM NEM produced by dissolving 250 mg of NEM into 10 mL of phosphate-buffered saline (PBS). The stock solution can be stored at 4 °C. Compounds such as nitroarginine, nitroprusside, and nitroglycerin that inhibit NO synthase (NOS) undergo slow non-quantitative conversion to NO, resulting in a baseline elevation. If the use of NOS inhibitors is part of the investigation, it is advised that blockers that do not contain nitro groups be used⁴³. The inlet pressure in the purge vessel should match the continuous negative pressure made by the CLD for sampling. If the inlet pressure is low, the vacuum distillation of the medium may lead to contaminate gas purging lines and chambers, resulting in signal distortion. Furthermore, small volumes of air may enter the purge vessel and react with the solution to generate excessive NO. On the contrary, excessive pressure caused by a flow rate that is higher than sample withdrawal flow might release NO into the atmosphere leading to an underestimation of the signal¹⁵. In addition, both the VCl₃ in HCl and the I₃^- in acetic acid protocol require multiple dilutions during preparation and multiple calculations to yield sample concentration. Every step should be carefully recorded as dilutions and calculations are recognized by many groups as the main source of error⁴⁴.

Addition of NO₂^- blocking solution (step 2.1.2) causes red blood cells to lyse, reducing all OxyHb to MetHb. It avoids the rapid reduction of iron-containing proteins (primarily OxyHb) by NO₂^- with subsequent generation of NO₃^-. Treatment of samples with this solution upon collection is required to correctly measure NO₂^- and/or NO₃^-. In contrast, the solution should not be used when planning the NO consumption by cell-free Hb assay. Aliquots of the NO₂^- blocking solution should be stored together with the samples and used to remove the portion of signal caused by the solution itself. Finally, the dilution factors should be taken into account for calculating the final concentration of the measured metabolite. In the I₃^- in acetic acid assay, the NO₂^- concentration is obtained by subtracting the concentration of the aliquot containing AS without HgCl₂ (S-NO and Fe-NO) from that of aliquot containing only PBS (NO₂^-, S-NO, Fe-NO). To calculate the concentration of S-NO (mercury-labile proteins), the concentration of the aliquot treated with AS with HgCl₂ (Fe-NOs or mercury-stable nitroso-proteins) should be subtracted from that of the aliquot treated with AS without HgCl₂. Before subtracting, the user should recall multiplying each aliquot's concentration by 0.8181 (270/330) to take into account the dilution factor used during sample preparations (Figure 3). The same concept applies for S-NO-Hb and Hb-NO measurement, where absolute concentrations are a fraction of measured Hb²². Depending on the type of sample, a precise count of the NO₃^- concentration may require running both the VCl₃ in HCl protocol and the I₃^- in acetic acid protocol to finally subtract the concentration of NO₂^- from those of NO₃^- and NO₂^-. This is not always necessary in whole blood and plasma samples, for example, where NO₃^- typically far exceeds NO₂^-.

It is very important to remind that the CLD will be damaged if the NaOH trap is not in proper place when performing a VCl₃ in HCl assay due to the corrosivity of HCl. The 1 M NaOH solution used to fill the NaOH trap can be purchased or prepared either by dissolving 4 g of NaOH salt in 100 mL of ddH₂O or by diluting 5 mL of 50% NaOH in 100 mL of ddH₂O. When further diluted to 10 µM with ddH₂O, the NaOH is very useful to remove protein debris by injection in the purge vessel.

The baseline signal detected by the CLD should be stable and within 0-5 mV. A higher baseline signal could be caused by air pollution (by NO and derivatives), and this can be verified by leaving the inlet in the open air. If a higher voltage is observed only in the purge vessel, while contamination of the tank with inert gas should be considered, it is most often due to the persistence of organic remnants or the presence of NO₂^- in the tube. Washing the purge vessel multiple times with ddH₂O can help minimize the baseline voltage. If unsuccessful, the incubation of the purge vessel with 100 mM NaOH for 24-48 h followed by multiple washes with ddH₂O helps eliminate pollutants. In the NO consumption by cell-free Hb assay, the required CLD signal is 70-100 mV for at least 20-30 min. If the signal is not stable (i.e., voltage decreases), an extra dose of 10 μL of DETA NONOate can be added to the purge vessel. This can be repeated as needed. It is essential to remember that both unused powder and already expanded DETA NONOate should be kept at -80 °C. This reagent decays rapidly despite optimal storage and maybe underperforming when not freshly prepared, requiring multiple injections into the purge vessel and leading to a less stable isoelectric signal. More importantly, it should be expanded in PBS with pH 7.4 as this optimizes its half-life (56 h at pH 7.4 at room temperature vs. quasi-instant dissociation at pH of 5.0)⁴⁵. The syringe used for injection of DETA NONOate should be exclusively dedicated to that action and not used for injection of samples.

If the injection of a sample generates a peak that is higher than the upper point of the calibration curve, especially if it generates a voltage higher than 700 mv, it is advised to dilute the sample (2x or 4x) to minimize error. This also applies to the NO consumption assay if the dip caused by sample injection is deeper than the one caused by the injection of the most concentrated sample used in the calibration curve. The evidence of a peak (or a through) in signal after injection that is less than expected or absent may indicate clogging of the precision syringe used for injection. Due to the small volumes used (10-20 μL/sample), particularly if plasma or urine assays are performed, it may be difficult to distinguish between a filled syringe and the absence of the sample due to clogging of the syringe. It is advised to pay attention to the resistance made by the plunger. After confirmation by inspection of the syringe, removal of the obstruction may be attempted by repeatedly washing the syringe with ddH₂O and by rubbing the syringe on a delicate task wipe if the obstruction is externally visible. When in doubt, it is advised to eject the sample volume on a task wipe and go back to wash the syringe with ddH₂O and try to repeat sample administration. Foaming is a common complication and forces the experiment to be terminated once the height of the liquid column exceeds that of the reaction column, causing signal instability. Exceeding the antifoaming agent dose indicated in the protocol may lead to an increase in the baseline signal and should be avoided. The number of samples causing the formation of excessive foaming becomes predictable when repeated assays are performed, a phenomenon that should inform how the experiment can be optimally planned. An optional step to decrease foaming is to mix plasma or tissue supernatant with cold methanol (ratio 1:1) and centrifuge for 15 min, 13000 x g at 4 °C. It should be considered that methanol causes proteins to precipitate. Only light S-NOs and Fe-NOs that are not proteic in nature are measured. Thereby, this option cannot be adopted in studies on whole blood or in any other case where there is interest in S-NO and Fe-NO proteins.

More reagent solutions other than VCl₃ in HCl and the I₃^- in acetic acid have been described and successfully adopted. Ascorbic acid specifically reduces NO₂^-, does not react with NO₃^- nor S-NOs, and can be used to reduce the bulk of injected samples and calculations in situations where NO₂^- is the only molecule of interest. The main pitfall is the limited stability of the reagent, requiring more frequent changes of the reagent solution and, thereby, repeated calibrations⁴⁶. An assay using carbon monoxide and cuprous chloride (CuCl)-saturated cysteine (3C method) can specifically detect S-NOs with a sensitivity that is comparable with I₃^- but still requires treatment with and without HgCl₂ to avoid non-specific reactions⁴⁷. To achieve higher precision in signal analysis, experiments are recorded in the .data format so that offline waveform analysis can be completed with different software other than the bundle program shown in the protocol.

Once the first injection is made, the experiment cannot be stopped. If a pause is needed and/or if any condition is changed (composition of the purge vessel medium, purge gas flow, baseline voltage), the experiment is valid only for the samples injected up to that point. A new standard curve should be performed before measuring more samples. The precision allowed by chemiluminescence-based assays comes with a price. The assays are time-consuming for a number of reasons: 1) automation is not possible as manual sample injection is required in duplicates; 2) most reagents that are used cannot be stocked for long periods; 3) each experiment needs to be started by running standard samples at known concentrations for calibration and cannot be paused unless calibration is repeated; 4) each sample is divided into aliquots containing different mixtures of reagents to block specific reactions in order to finally calculate the concentration of specific metabolites by subtracting others.

This work is focused on the device Zysense NOA 280i, which has a detection threshold of <1 ppb NO in the gas phase (corresponding to up to 1 pM of NO in the liquid phase) and a sampling flow of 10-300 mL/min. Other CLDs are available with a bundle tubing system for the measurement of metabolites in the liquid phase. Ecophysics CLDs have a higher sensitivity threshold, 1 ppb in their most sensitive model but the advantage of not requiring an oxygen tank for ozone generation. On the other hand, their reported sampling flow is 1000 mL/min, which implies a higher consumption of the inert gas tank used for the purge vessel. Other CLD machines are dedicated to exhaled NO analysis for clinical use.

Chemiluminescence is the most sensitive and validated method to specifically assess NO consumption by cell-free Hb in the field of hemolysis. Other techniques are available to measure NO, NO₂^-, NO₃^-, S-NOs and mercury-labile nitroso compounds. The direct measurement of NO gas can be obtained either by chemiluminescence or with the use of dedicated electrodes that range from the measurement of the exhaled gas to single-cell electrodes. These measure only NO gas, are less expensive (yet more perishable), require calibration more often, and are less precise compared to a CLD³¹. Nitrites and NO₃^- have been historically measured with the Griess reaction, a colorimetric technique that in its original version detects both molecules without the possibility of distinguishing their relative concentration in the sample. Modern variants of the technique involve reverse phase chromatography employing separation of the sample in two columns, allowing two independent reactions with NO₃^- and NO₂^-. The sensitivity of those methods is micromolar or slightly less, as compared with 1 nM with CLD. The main pitfalls are time consumption and incompatibility with samples pretreated with mercury and/or ferricyanide¹⁴. S-NOs can be detected with colorimetric and fluorometric techniques as well with a sensitivity up to 0.5 µM compared to 100 fM described by using chemiluminescence as shown in the I₃^- in acetic acid protocol^14,41,42. The biotin-switch assay, despite failing to detect S-NOs in the picomolar range, has the advantage of generating biotin-labeled proteins that can be purified and identified with proteomic analysis⁴⁸. When looked at in detail, each technique shares one common pitfall, which is the extreme reactivity of NO and its derivatives mandating blockade of reactions for the purpose of in vitro measurement. Pre-treatment of samples is critical, and answering the question of whether a certain pre-treatment completely blocks a reaction or does so better than another one has been so far a driving force towards an increase in methodological studies. While the extreme reactivity in vivo of NO has certainly been a drive to research sensitive methods for detection of its metabolites, direct measurement of NO has met significant progress in the past three decades since the first amperometric detector was developed in 1990. More than twenty different probes have been described since then, including nanosensors used to study single-cell NO content⁴⁹. Electrochemical sensors are the main tools that can detect NO both in vivo and in real-time. In pulmonary medicine, multiple studies have compared electrode-based machines with CLDs. Measurement of exhaled NO is more simple than other assays for metabolite detection, requiring either direct connection of the gas line (via IFD filter) to a closed-system respiratory adjunct or a few breaths in a bag for offline analysis, yet the cost and poorer portability of CLDs standstill. A number of electrochemical devices currently meet the criteria for clinical use, although they have lower sensitivity and reproducibility when compared to CLDs. Furthermore, the concordance in terms of absolute measured values is suboptimal between devices, so it is recommended that patients should be always followed up with the same electrochemical device⁵⁰. The use of CLDs in settings such as measurement nasal exhaled NO and in research settings like therapeutic low and high-dose inhaled NO is still necessary.

Precise measurements of NO, its derivatives, and NO-binding proteins are necessary at many levels to understand an overlooked and still mostly unknown mechanism of signaling. The fields of application span biology and pathology at both preclinical and clinical levels with specific target areas, including immunobiology, neurosciences, cardiovascular sciences, and infective diseases. Preclinical studies and clinical trials are investigating the use of inhaled NO gas as a therapeutic agent. Deeper knowledge on how exogenous NO gas influences the concentration of NO metabolites and NO-bound proteins at every level, from plasma to cell compartments, will be required. Implementation of high-yield proteomics may likely increase the number of known mediators that are influenced by NO and its derivatives and may mandate new mechanistic investigations requiring precise measurements of NO metabolites.

Materials

Name	Company	Catalog Number	Comments
Acetic Acid	Sigma	45754	500 mL - liquid
Antifoam B Emulsion	Sigma	A5757	250 mL - liquid
DETA NONOate	Cayman	82120	10 mg
Gibco DPBS (1x) no calcium, no magnesium	ThermoFisher	14190144	500 mL
Hydroochloric Acid 37% (1 M)	Sigma	258148	500 mL - liquid
Iodine	SAFC	207772	100 g - solid
Kimwipes	Kimtech	34155
Mercury (II) Chloride ACS reagent> 99.5%	Sigma	215465	100 g - solid (dissolve in water)
Mili-Q Water Purification System	Millipore
Model 705 RN 50 μL syringe	Hamilton	80530	Microliter syringe
Model 802 N 25 μL Syringe	Hamilton	84854	Microliter syringe
N-ethylmaleimide	Sigma	4260	25 g - crystalline
Nitric Oxide Analyzer + Bundle Software - Purge Vessel	Zysense	NOA 280i	Chemiluminescence Detector
Nonidet p-40  (NP-40)	ThermoFisher	85125	10% - 500 mL
Potassium hexacyanoferrate (III) ACS reagent≥ 99%	Sigma	244023	100 g - powder
Potassium Iodide ACS reagent> 99%	Sigma	221945	100 g - solid
Potassium Nitrite cryst. For analysis EMSURE ACS	Supelco	105067	250 g - crystalline
PowerGen 125	Fisher Scientific	14-359-251	Mechanic Homogenizer
RV3 Two Stage Rotary Vane Pump	Edwards	A65201906	Vacuum Pump - Bundled with analyzer
Sodium Heparin - BD Hemogard Clo	BD Biosciences	BD367871	75 USP Units
Sodium hydroxide anhydrous ACS reagent ≥ 97%	Sigma	795429	1 kg - pelletts
Sodium Nitrate ACS reagent ≥ 99%	Sigma	221341	500 g - powder
Sodium nitrite ≥ 99%	Sigma	S2252	500 g - crystalline
Sulfanilamide ≥ 98%	Sigma	S9251	100 g - solid
Vanadium (III) Chloride	Sigma	112393	25 g - solid - Caution (exothermic)
Whatman 1 Filter Paper	Sigma	WHA10010155