This paper explored potential effects of a plasma and serum storage temperature (room temperature, 4°C versus -80°C) and duration, freeze-thaw cycling, viral heat inactivation and tube type and volume on one dimensional (1-D) proton (1H) NMR spectra and Carr-Purcell-Meiboom-Gill (CPMG) spin-echo and lipoprotein plots. NMR spectra of plasma from healthy patients and patients diagnosed with the novel 2019 coronavirus disease (COVID, COVID-19, COVID19) caused by infection with the SARS-CoV-2 virus were compared in specimens before and after heat-inactivation.
Quantitative In-Vitro Diagnostic NMR Spectroscopy for Lipoprotein and Metabolite Measurements in Plasma and Serum: Recommendations for Analytical Artefact Minimization with Special Reference to COVID-19/SARS-CoV-2 Samples.
Loo RL, Lodge S, Kimhofer T, Bong SH, Begum S, Whiley L, Gray N, Lindon JC, Nitschke P, Lawler NG, Schafer H, Spraul M, Richards T, Nicholson JK, Holmes E
Purpose
Conclusion
Inter-individual variance rather than variance due to experimental conditions accounted for the largest portion of the variance in the dataset. As expected differences were observed in the 1D and the CPMG between EDTA plasma, heparin plasma and serum samples and between K2EDTA and K3EDTA plasma, but there were no differences attributed to collection tube type for plasma specimens (plain clot versus gel coated or heparin coated versus heparin gel) or tube volume (4 mL versus 10 mL EDTA tubes). Of the experimental conditions tested, heat-mediated viral inactivation of plasma samples caused the greatest shift in 1D proton NMR spectral plots and CPMG spin-echo plots, while storage of serum at 4°C for 168h resulted in the largest changes to lipoprotein data. While serum storage at 4°C for 168 h and heat inactivation of plasma caused similar directional shifts in both CPMG spectra and lipoprotein data, 1D proton NMR spectral data exhibited a much smaller and orthogonal shift after 4°C storage relative to the changes observed following heat inactivation. Effects observed in serum after three freeze-thaw cycles were comparatively small respective to changes associated with serum storage and heat inactivation, but included shifts in the 1D proton NMR spectral plot and lipoprotein data that were orthogonal to the shift in the CPMG plot and in the opposite direction of the shift post heat inactivation. Importantly, 112 lipoprotein parameters were relatively stable during 48 h of storage at 4°C or room temperature, 3 months of storage at -80°C, and 3 freeze-thaw cycles. Notably, plasma collected from SARS-CoV-2 infected and healthy individuals were differentially affected by heat inactivation (56°C); and, the difference in vector direction between samples collected from healthy and infected individuals indicate that heat inactivation induced differences that were different in SARS-CoV-2 patients degradation. A total of 65 and 57 changes in the lipoprotein parameters were observed after heat inactivation of plasma from healthy and SARS-Cov-2 patients, respectively, but only 30 of these parameters were similarly affected by heat inactivation in specimens from both healthy patients and those with SARS-CoV-2. There were 37 lipoprotein variables that differentiated infected from healthy individuals when not heated, but only 12 of these were unaffected by heat inactivation. Differences were observed in the 1D and the CPMG between EDTA plasma, heparin plasma and serum samples, but no differences were attributed to collection tube type (plain clot versus gel coated or heparin coated versus heparin gel) or tube volume (4 mL versus 10 mL EDTA tubes).
Studies (1)
Purpose
This paper explored potential effects of a plasma and serum storage temperature (room temperature, 4°C versus -80°C) and duration, freeze-thaw cycling, viral heat inactivation and tube type and volume on one dimensional (1-D) proton (1H) NMR spectra and Carr-Purcell-Meiboom-Gill (CPMG) spin-echo and lipoprotein plots. NMR spectra of plasma from healthy patients and patients diagnosed with the novel 2019 coronavirus disease (COVID, COVID-19, COVID19) caused by infection with the SARS-CoV-2 virus were compared in specimens before and after heat-inactivation. Blood was collected from eight healthy individuals and 7 individuals with active SARS-CoV-2 infection and allowed to clot for 30 min at 20°C (except serum silica tubes which were allowed 1 h) prior to isolation of serum and plasma by an unspecified method. Plasma and serum were centrifuged at 13,000 g for 10 minutes at 4°C before NMR analysis. Potential differences in lipoprotein and metabolic endpoints attributable to collection tube were investigated, using blood collected from five healthy volunteers in matched 5 mL SSTII Advance and 10 mL plain serum tubes, 6 mL lithium heparin, 4.5 mL lithium heparin gel PST II, K2EDTA (4 and 10 mL) and 2 mL K3EDTA tubes and storing the resultant plasma/serum specimens at room temperature for 0, 24, 48, 72, and 168 h. Potential effects of serum storage were investigated using blood from five healthy patients (tube type not specified), aliquoted and stored at room temperature for 1 h, 24 h or 48 h, at 4°C for 0, 5, 24, 48, 72 and 168 h, and -80°C for 3 months. To investigate effects of viral heat inactivation, lithium heparin plasma from eight healthy individuals and seven SARS-CoV-2 patients was analyzed before and after incubation at 56°C for 30 min. To investigate effects of freeze-thaw cycling, serum was frozen at -80°C for ≥12 h and thawed at room temperature for 1 h never, twice, or three times and compared to specimens freeze-thaw cycled once. NMR data was collected with a 600 MHz Bruker Advance III HD spectrometer following solvent presaturation and a 1D–Carr-Purcell-Meiboom-Gill (CPMG) spin-echo sequence and a 2D –J-resolved experiment. For all experiments investigating potential effects of pre-analytical handling, data was compared to respective control specimens processed immediately (or within 1 h) without storage or in the freeze-thaw cycling experiment specimens freeze-thaw cycles once.
Summary of Findings
Inter-individual variance among spectral profiles of plasma and serum specimens accounted for 35.9% and 21.3% of the variance in the 1D and CPMG spin-echo datasets, respectively. Differences were observed in both 1D and CPMG spectra generated among EDTA plasma, heparin plasma, and serum samples, but no differences were attributed to collection tube type (plain clot vs gel coated or heparin coated vs heparin gel) or tube volume (4 mL vs 10 mL EDTA tubes). There was a small increase in acetate in specimens collected in K3EDTA rather than K2EDTA tubes. Of the experimental conditions tested, viral inactivation via heating plasma samples caused the greatest shift in the 1D proton NMR spectral plot and CPMG spin-echo plots. The largest change in lipoprotein spectra occurred when serum was stored at 4°C for 168 h compared to controls analyzed immediately, which resulted in shifts in both CPMG spectra and lipoprotein data in the same direction as heating plasma. Notably, the shift in 1D proton NMR spectral data after 4°C storage of serum was comparably small and orthogonal to the shift observed after heat inactivation. Interestingly, when serum samples underwent freeze-thaw cycling the 1D proton NMR spectral plot shifted in the opposite direction relative to the change observed following heat inactivation of plasma, although the shift was comparably less intense. Conversely, the shift observed in the CPMG plot following freeze-thaw cycling of serum specimens was orthogonal to the shift observed after heat inactivation of plasma. Lipoprotein data shifted in the same direction as the shift observed after heat inactivation of plasma after 0 or 2 freeze-thaw cycles but in the opposite direction after 3 cycles. Importantly, the 112 lipoprotein parameters examined were found to be relatively stable during 48 h of storage at 4°C or room temperature, 3 months of storage at -80°C and 3 freeze-thaw cycles.
Overall, viral inactivation via heating serum samples to 56°C for 30 min produced major changes in lipoprotein profiles generated via proton NMR, as well as CPMG and 1D spectral data. Changes included a shift in principal component spacing post-heating relative to controls that did not undergo heat inactivation. However, serum samples from SARS-CoV-2 and healthy individuals were differentially affected by heat inactivation as the net change in vector direction differed among these samples. Importantly, the spectral shifts were also not equivalent in all SARS-CoV-2 patients with one patient showing a shift more similar to that observed in specimens from healthy patients. These observations were supported by a clear difference in concentrations of the lipoprotein panel calculated with the OPLS-DA model that translated to distinct clustering of specimens from infected and noninfected individuals. A total of 65 and 57 significant changes in lipoprotein parameters were observed after the heating inactivation of plasma from healthy and SARS-Cov-2 patients, respectively, but only 30 of these parameters were affected in both healthy and SARS-Cov-2 patients. There were 37 lipoprotein variables that differentiated infected from healthy individuals when not heated, but only 12 of these were unaffected by heat inactivation. A total of 41 parameters differed between healthy and SARS-Cov-2 infected individuals when heat inactivated specimens were analyzed, 27 of which also differed between healthy and SARS-CoV-2 infected individuals when specimens in controls that did not undergo heat inactivation. Specifically, increases in LDL subfractions, H2A1 and TPA1 post heat inactivation were more pronounced in plasma specimens from SARS-CoV-2 patients compared to healthy individuals, while heat inactivation-mediated changes in HDL-L4 and HDL-L6 sub-fractions were more pronounced in plasma specimens from healthy individuals compared to those infected with SARS-CoV-2.
Biospecimens
FluidBloodFluidPlasmaFluidSerumPreservation Types
None (Fresh)FrozenDiagnoses
Pneumonia/Respiratory InfectionNormalPlatforms
Small moleculeNMRLipoproteinNMRPre-analytical Factors
Classification Pre-Analytical Factor Value(s) Preaquisition Diagnosis/ patient condition HealthySARS-CoV-2 infectionBiospecimen Acquisition Anticoagulant Lithium heparinMultiple forms evaluatedPotassium EDTABiospecimen Acquisition Type of collection container/solution 10 mL K2EDTA tubes10 mL plain serum tubes2 mL K3EDTA tubes4 mL K2EDTA tubes4.5 mL lithium heparin gel PST II tubes5 mL SSTII Advance serum tubes6 mL lithium heparin tubesBiospecimen Aliquots and Components Aliquot size/volume 10 mL4 mLStorage Freeze/thaw cycling 0 cycles1 cycle2 cycles3 cyclesStorage Storage duration 0 h168 h24 h3 months48 h5 h72 hStorage Storage temperature -80°C4°C56°CRoom temperature