Quantifying Flavin mononucleotide: an internationally validated methodological approach for enhanced decision making in organ transplantation

Abstract

Background: Increasing donor risk, particularly in liver transplantation, where organs are often marginal, has made dynamic organ preservation techniques and viability assessment essential to safely improve organ quality and increase utilisation. However, existing viability parameters are based on routine clinical assessment in patients with acute liver failure, trauma, or liver resections. These parameters often do not correlate with clinically relevant post-transplant outcomes.

Methods: This article presents a detailed protocol for the spectrophotometric quantification of Flavin mononucleotide (FMN), a marker of mitochondrial injury. FMN release from mitochondrial complex I was described many decades ago as the initial sign of ischaemia-reperfusion injury, i.e. when oxygen is reintroduced in ischaemic tissues during organ transplantation or machine perfusion. This study describes the detailed FMN quantification in donor plasma and various fluids obtained during machine perfusion, and discusses confounders, challenges, and the role of individual test components.

Findings: FMN quantification was identified as an immediate organ assessment tool, demonstrating a strong correlation with graft survival and other relevant complications after human liver transplantation.

Interpretation: The results highlight FMN quantification as a reliable and standardized method for assessing organ viability, offering significant potential for improving organ selection and better utilisation. This method could provide better a predictive value for transplant outcomes compared to existing parameters currently in use.

Funding: This research received no external funding but was supported by the Catalyst grant No. CCG0280 at Cleveland Clinic Ohio, U.S. dedicated to A.S.

Keywords: Flavin mononucleotide; Fluorescence spectroscopy; Ischaemia-reperfusion injury; Machine perfusion; Organ transplantation; Viability assessment.

Fig. 1

Required material for FMN quantification with fluorescence spectroscopy: a) Succinate triggers uncoordinated electron flow and induces ROS and FMNH₂ release from complex I, that can be quantified by fluorescence spectroscopy when fully reduced to FMN. Required equipment for our technology: b) NaCl 0.9% (Saline for preparation of stock solutions and standard and sample dilution); c) set of pipettes (P1000, P200 and P20); d) FMN standard: Riboflavin 5′Phosphate (USP™, 100 mg), light protected; e) 1.5 ml Eppendorf tubes; f) Available 96 well plates (Thermo Scientific™); g) Synergy H1 M−SN (Spectrometer Biotek®).

Fig. 2

FMN stock solution and standard preparation: Step 1: Create Stock Solution A (1 mg/ml)–Dilute and mix Riboflavin 5′-Phosphate powder with 0.9% NaCl at 1:1 ratio. FMN is light sensitive; the powder and the FMN standard solutions must be protected from light and kept cold (i.e. wrap the tubes with aluminium foil and put them on ice). Step 2 and 3: Prepare HOPE, NMP and Plasma/Serum stock solution: Step 2: For HOPE Standard 1 Solution, dilute Stock Solution A (1 mg/ml) directly in 0.9% NaCl at 1:2000 ratio (i.e. 10 μl Stock solution A with 19′990 μl 0.9% NaCl). For NMP and Plasma/Serum Stock Solution, a Stock Solution B (50 μg/ml) will be created by diluting Stock Solution A in 0.9% NaCl in a 1:20 ratio (i.e. 1 ml Stock solution A with 19 ml 0.9% NaCl). Step 3: prepares the final NMP and Plasma/Serum Standard 1 by diluting Stock Solution B (50 μg/ml). Dilute Stock Solution B with 0.9% NaCl at a 1:8 ratio for NMP Standard 1 (FMN concentration 6.25 μg/ml). Dilute Stock Solution B 1:4 with 0.9% NaCl to receive Plasma/Serum Standard 1 (FMN concentration of 25 μg/ml). Step 4 includes the serial dilution of HOPE, NMP, Plasma/Serum Standard 1 with 0.9% NaCl at 1:1 to receive Standard 2, 3, 4, 5 and 6. In general, the concentration of the next higher (“diluted”) standard is always half of the previous standard (i.e. NMP St.1 = 6.25 μg/ml, St.2 = 6.25/2 μg/ml, St. 3 = 6.25/4 μg/ml etc.). Standard 7 is pure 0.9% NaCl (FMN concentration: 0 μg/ml) for all standards.

Fig. 3

Standard and sample distribution into the 96-well plate for FMN quantification. A clear, flat bottom 96-well plate (ThermoScientific®) is used with a conventional fluorescence spectrometer. The total fluid volume added to each well for the measurement is 200 μl. Steps 6 and 7 describe the required additional dilutions for standard and samples based on the sample origin. Any number of samples can be aliquoted onto the plate in duplicates or triplicates, depending on the experimental settings. For HOPE perfusates, there is no need for additional sample dilutions prior to distribution in the plate. Fill the subsequent plate wells with 150 μl 0.9% NaCl and distribute 50 μl of HOPE standards and HOPE samples to those wells. For NMP, dilute the sample and NMP standards in a separate Eppendorf tube with 0.9% NaCl. NMP samples (perfusate and bile) and NMP standard 1–6 will be diluted in a 1:51 ratio with 0.9% NaCl in a new empty 1.5 ml Eppendorf tube (i.e. mix 500 μl 0.9% NaCl and 10 μl of sample). Vortex and distribute 200 μl from those Eppendorf vials in empty plate wells. For Plasma/Serum analyses, dilute the Sample and Standard, required to avoid overflow (the signal has exceeded the maximal capacity of the system used to detect it). Serum/Plasma FMN standards and samples are diluted with 0.9% NaCl in a ratio of 1:101 in an empty 1.5 ml Eppendorf tube (mix 1000 μl 0.9% NaCl and 10 μl of sample). Vortex the mixture well before distributing 200 μl in the 96-well plate. Standard 7 equals 200 μl NaCl 0.9% without FMN, which will be added to the corresponding plate wells for all three fluid types. Spectrometer settings are shown for FMN quantification (Step 8). Settings for Fluorescence Spectroscopy include excitation wavelength: 485 nm; emission wavelength: 528 nm; Gain: 130%. FMN, Flavin mononucleotide; HOPE, Hypothermic oxygenated perfusion; NMP, normothermic machine perfusion.

Fig. 4

Calculation of FMN concentrations in different media obtained with fluorescence spectroscopy (Excitation: 485 nm; Emission: 528 nm; Gain: 130%). The measured fluorescence signal (A.U.) of FMN standards is plotted with the corresponding FMN concentration (μg/ml) in a linear function a) r = 0.994; b) r = 0.997; c) r = 1.0). Each A.U. value corresponds to a specific FMN concentration in μg/ml. Seven digits are considered for calculation. Sample FMN concentrations are listed for each of the different fluids (a, b &c). Based on the same dilution ratio of all standards and samples for each fluid type (i.e. HOPE perfusate, NMP perfusate, Serum/Plasma) additional dilution factors do not need to be considered for value calculation. The FMN concentration prior to final standard and sample dilution (HOPE 1:4; NMP 1:51; Plasma/Serum 1:101) can be considered for the calculation instead. The increasing FMN concentration from NMP Standard 7–1 can be observed in the fluorescence spectrum (450 nm–700 nm) by steadily increasing the peak A.U. The lowest detected FMN concentration in this series was 0.0034905910 μg/ml; d) Example formula and calculation of FMN concentration. Based on the calibration curves in a, b and c, the corresponding sample concentrations were determined using this formula. e) The spectrum peak is seen at an excitation wavelength of 485 nm within an enlarged emission range of 300–700 nm showing that our settings result in the ideal dynamic range of FMN, which means the FMN signal can be adequately detected. f) Multiple calibration curves were produced by 6 different operators with our method with identical signal intensity, as shown with the example of the NMP standard (example month June 2024, n = 70). Large volume stock solution preparation with stock solution aliquot and standard freezing contributes further to the high reproducibility.

Fig. 5

Fluorescence spectroscopy comparing FMN NMP standard and sample readings using different 96 well plates (Excitation: 485 nm; Emission: 528 nm; Gain: 130%). a–d) Black and most of the transparent plates (a, c & d) do not affect the FMN concentration because they are calibrated with the standard despite the different fluorescence intensities and absolute A.U. values. However, among the different transparent plates, material differences may affect the FMN results. Plate b seems to reduce the absolute A.U. and sample FMN concentration. Black plate compared with clear plate (Excitation: 485 nm; Emission: 528 nm; Gain: 130%). e) Concentrations read using the ThermoScientific® clear plate (as used in all tests and in a) correlated well with the available ThermoScientific® black plate (as used in test d, r = 0.9980). f) Using our standard powder with standard dilutions and spectrometer settings as described we experienced only a very minimal photobleaching effect. Repeat measurements of the same standards in the same plate (1st–15th measurement) showed almost identical results for FMN concentrations.

Fig. 6

FMN quantification in Standard and NMP perfusate using different solutions for dilution, measured with different spectrometer settings (Excitation: 485 nm; Emission: 528 nm; Gain: 130%; n = 2 each measurement). The different slope of the FMN standard line shows that different solutions chosen for dilution can affect the signal intensity of the FMN. However, the correlation between the fluorescence signal of FMN Standards and their concentration remains linear. The most common solutions, such as 0.9% NaCl and Belzer MPS, behave similarly (a). Next, NMP perfusate samples were diluted in different fluids including our routine 0.9% NaCl, Belzer MPS, distilled water and Ringers. Such samples were analysed for FMN concentrations using our standard spectrometer settings (i.e. Excitation 485 nm; Emission 528 nm; Gain 130%) (b) and with reduced Gain of 100% (c). Given the spectrometer settings and standard preparations are kept as per routine and as described (dilution in 0.9% NaCl) (d–g), gain modifications from 130% to 100% lead to the same FMN concentrations (r = 0.9988) (h). Similarly, applying our standard spectrometer settings in comparison with the settings presented by Wang et al. and used by the team from Vienna (Austria) leads to comparable FMN concentrations (r = 0.9919) (i). A gain reduction to 120% (using the autogain option, see also Fig. 4, step 9, template 6b) results in the same FMN concentration but prevents “sample overflow” with too high A.U. and resulting in no FMN concentration. This might be relevant in fluids with very high FMN levels, such as bile released during NMP from livers with high risk (j).

Fig. 7

Fluorescence spectrum signal of NMP and HOPE standards measured with different spectrometers and the role of FMN standard powders and temperature effects (Excitation 485 nm; Emission 528 nm; Gain 130%; duplicates for each measurement). Despite the use of different Spectrometers, NMP FMN standards behave linearly to the FMN concentration (a). FMN concentrations measured with the Synergy H1 (our own established reader) and the Cytation 5 were comparable and correlated well (r = 0.9877). Old (established) and new Synergy H1 readers quantified FMN with exact the same concentrations (r = 0.9998) (b). The absolute FMN concentration of HOPE standard remains the same despite using five different Spectrometers in two different centres (US and the Netherlands) (c), given the same wavelength was used for excitation and emission with all devices. The parameter gain had to be either adjusted or accepted as not adjustable. Despite the different devices and the various gain (i.e. Spectromax ID3 with autogain, Clariostar with 1000 or 1500, Cytofluor with 50 and 90), FMN concentrations were identical. Such analyses confirm the high relevance and reproducibility of our results with various spectrometers and settings using the same settings and standard dilutions. d) Repeat fluorescence spectroscopy with the same spectrometer and standard settings (Excitation: 485 nm; Emission: 528 nm; Gain: 130%) were done using standard FMN powder (Riboflavin 5′-phosphate, USPTM, 1535700, CAS: 6184-17-4) compared to Riboflavin 5′-phosphate provided by Sigma (Sigma Aldrich Fine Chemicals Biosciences (Riboflavin 5′-phosphate sodium slat hydrate, MFCD00150992, R777425G). Such two providers guarantee a >70% purity. “Prescribed For Life” (PFL) provides Riboflavin 5′-phosphate Sodium USP39 through Amazon (Lot—V702-2204001-R, net weight 57 g). Riboflavin 5′- phosphate provided by USP and Sigma has high purity and subsequently higher and identical A.U. signals when tested with our routine standard concentrations 1–7. e) The powder provided by PFL however seems to have a lower purity, triggering reduced signals as seen with an A.U. of 24′304.5 for the highest standard 1 with 6.25 μg FMN/ml. f) Our standard method was applied reading NMP FMN standards at different temperatures. Within a range of 22–37 °C results were identical.