NOV 17, 202650 MINS READ
31P-NMR spectroscopy exploits the magnetic properties of phosphorus-31 nuclei to non-invasively detect and quantify phosphorus-containing metabolites in living tissues 2. Unlike 1H-MRI, which suffers from limited chemical shift dispersion (approximately 5 ppm) and significant background interference, 31P-NMR offers a substantially wider chemical shift range (approximately 30 ppm) with negligible background signal, enabling superior resolution of individual metabolite peaks 2. The technique operates by applying radiofrequency (RF) pulses at specific frequencies—typically at magnetic field strengths of 1.5T to 9.4T—and measuring the resulting free induction decay (FID) signals 1,4. Fourier transformation of these time-domain signals yields frequency-domain spectra where distinct resonances correspond to PCr (δ ≈ 0 ppm relative to phosphoric acid), ATP (α-, β-, and γ-phosphate groups at δ ≈ -7.5, -16, and -2.5 ppm respectively), Pi (δ ≈ 5 ppm, pH-dependent), and phosphomonoesters/phosphodiesters 10,11. The chemical shift of Pi is particularly valuable as it serves as an intrinsic pH indicator: δPi = 3.27 + 0.55 × log[(pH - 6.75)/(7.4 - pH)], allowing simultaneous pH determination without exogenous probes 10. Key experimental parameters include: pulse repetition time (TR = 2–30 s to allow T1 relaxation), acquisition time (0.2–0.5 s), spectral width (5–10 kHz), and number of averages (8–256 depending on signal-to-noise requirements) 1,3. Modern implementations employ surface coils or volume coils optimized for 31P detection, with typical spatial resolution of 1–8 cm³ and temporal resolution of 30 s to 10 min 2,11.
Accurate quantification of PCr concentration requires rigorous spectral processing and calibration 3. Following Fourier transformation with exponential line broadening (10–20 Hz), baseline correction, and phase adjustment, peak areas are integrated using Lorentzian or Gaussian fitting algorithms 10. The relationship between peak area (A) and metabolite concentration (C) follows: C = (A / Aref) × Cref × (Vref / Vsample), where Aref and Cref represent the area and known concentration of a reference standard (often phenylphosphonic acid in a capillary insert), and V denotes volumes 3,10. For in vivo studies without external references, relative quantification via metabolite ratios (PCr/ATP, PCr/Pi) is standard practice, with typical healthy skeletal muscle values of PCr/ATP ≈ 3.5–4.5 and PCr/Pi ≈ 4–10 at rest 11,13. Absolute quantification demands correction for: (1) T1 saturation effects (T1,PCr ≈ 4–6 s in muscle at 1.5T), requiring TR ≥ 5×T1 or mathematical correction; (2) nuclear Overhauser enhancement (NOE) from 1H decoupling (typically avoided in quantitative work); and (3) coil loading and B1 inhomogeneity, addressable via phantom calibration or the use of tissue water as an internal standard 1,10. One validated approach employs standard addition: spiking samples with known PCr concentrations (25–500 μM) to generate calibration curves with linear correlation coefficients R² > 0.998 3. For enzyme activity assays, the method measures substrate depletion or product formation over time; for example, phosphodiesterase activity is quantified by monitoring cAMP → AMP conversion, where Δ[cAMP]/Δt directly reflects enzyme velocity (Vmax typically 0.5–5 μmol/min/mg protein) 3.
CEST-based methods provide enhanced sensitivity for detecting PCr-creatine interconversion mediated by creatine kinase (CK) 6,7,9. The technique applies a frequency-selective saturation pulse at the PCr resonance (δ ≈ 0 ppm) for 1–3 s, transferring saturation to creatine (δ ≈ 3.0 ppm for guanidinium protons) via CK-catalyzed exchange 6,9. The CEST ratio, CESTRₓ = (M₀ - Msat)/M₀, where M₀ and Msat are signal intensities without and with saturation, correlates with CK flux: kf = (1/T1,app - 1/T1,int), where T1,app is the apparent relaxation time under saturation and T1,int is the intrinsic T1 12. In cancer tissues overexpressing CK (2–10 fold elevation vs. normal), PCr → creatine conversion rates increase proportionally, yielding CESTR values of 15–35% compared to 5–12% in healthy tissue 6,9. Optimization requires careful selection of saturation power (B1 = 0.5–2.3 μT): higher B1 increases CESTR but reduces specificity due to direct water saturation and magnetization transfer effects 7. The ratiometric CESTR (rCESTR) approach, comparing signals at two B1 levels, improves robustness: rCESTR = log(CESTR₂/CESTR₁), with contrast-to-noise ratios (CNR) peaking at B1a = 0.5 μT and B1b = 2.0 μT 7. This method enables non-invasive CK activity mapping with spatial resolution of 3–5 mm and temporal resolution of 5–10 min, applicable to tumor grading (malignant vs. benign discrimination with 85–92% accuracy) and treatment monitoring 6,9.
Recent advances integrate 31P-NMR with hyperpolarized 129Xe MRI to decouple blood flow from metabolic responses 2. While 31P-chemical shift imaging (CSI) quantifies PCr dynamics reflecting ATP turnover, 129Xe-CSI measures red blood cell (RBC) perfusion efficiency via dissolved-phase xenon signals (δ = 217 ppm for RBC-bound Xe) 2. The blood flow-dependent metabolic parameter is computed as: ΔMetabflow = (ΔPCr/PCrbaseline) / (ΔXeRBC/XeRBC,baseline), where Δ denotes stimulus-induced changes 2. In functional brain studies, visual or motor stimulation increases both cerebral blood flow (CBF, +20–40%) and PCr consumption (-5–15%); normalizing PCr changes by CBF changes isolates oxygen consumption rate (CMRO₂) from hemodynamic confounds 2. Typical acquisition protocols involve: (1) baseline 31P-CSI (TR/TE = 2000/2.3 ms, 8×8 matrix, 2 cm slice, 6 min) and 129Xe-CSI (flip angle = 20°, TR = 15 ms, 1 min); (2) task execution (e.g., 2 Hz visual flicker for 5 min); (3) post-stimulus imaging with identical parameters 2. This dual-nuclear approach revealed that in healthy subjects, ΔMetabflow ≈ 0.3–0.5, indicating tight neurovascular coupling, whereas in cerebrovascular disease, ΔMetabflow > 0.8 suggests metabolic stress despite compensatory hyperemia 2. The method's limitation is the requirement for hyperpolarized 129Xe (polarization > 30%, T1 ≈ 20 s in blood), necessitating on-site polarizers or rapid transport from production facilities 2.
The chemical shift of inorganic phosphate (δPi) is exquisitely sensitive to pH, shifting downfield by approximately 0.55 ppm per pH unit in the physiological range (6.0–7.8) 10. Precise pH calculation employs the Henderson-Hasselbalch-derived equation: pH = 6.75 + log[(δPi - 3.27)/(5.69 - δPi)], calibrated against titrated phantoms (R² = 0.997, pH accuracy ±0.05 units) 10. Simultaneously, free [Mg²⁺] modulates the chemical shifts of ATP resonances: δβ-ATP and δα-ATP shift upfield by 0.3–0.5 ppm as [Mg²⁺] increases from 0 to 5 mM due to Mg-ATP complex formation (Kd,Mg-ATP ≈ 0.05 mM) 10. The peak height ratio hx = Hβ-ATP/Hα-ATP (or Hγ-ATP/Hα-ATP) provides an independent [Mg²⁺] indicator, as β- and γ-phosphates exhibit differential Mg²⁺ binding 10. A dual-parameter fitting algorithm solves: δPi = f₁(pH, [Mg²⁺], [PCr], [ATP]) and hx = f₂(pH, [Mg²⁺], [PCr], [ATP]) using nonlinear least-squares optimization (Mathematica™ or MATLAB) 10. Validation with 88 calibration phantoms spanning pH 6.0–8.0 and [Mg²⁺] 0.25–5.0 mM at 37°C, ionic strength I = 0.25 M, yielded: pH error < 0.08 units and [Mg²⁺] error < 0.15 mM across the physiological range 10. In perfused rat hearts, this method detected acidosis (pH 6.85 ± 0.05) and Mg²⁺ depletion ([Mg²⁺] = 0.4 ± 0.1 mM) during ischemia, compared to normoxic values of pH 7.15 ± 0.03 and [Mg²⁺] = 0.9 ± 0.1 mM 10. The technique's advantages include: no exogenous indicators, real-time monitoring (temporal resolution 1–5 min), and applicability to intact organs or cell suspensions 10.
31P-NMR enables direct, real-time monitoring of enzyme-catalyzed phosphate transfer reactions without radiolabels or chromogenic substrates 3,12. For phosphodiesterase II (PDEII) assays, the method tracks cAMP (δ ≈ -0.5 ppm) hydrolysis to AMP (δ ≈ 4.2 ppm) 3. Protocol: (1) prepare substrate solution (50–500 μM cAMP in 50 mM Tris-HCl, pH 7.5, 5 mM MgCl₂); (2) acquire baseline 31P spectrum (90° pulse, TR = 30 s, 8 scans, 5 min total); (3) add PDEII (0.1–1 μg/mL), initiate reaction at 37°C; (4) acquire spectra every 5 min for 60 min; (5) integrate cAMP and AMP peaks, calculate [cAMP]t and [AMP]t via calibration curves (linear range 25–500 μM, R² > 0.995) 3. Enzyme activity (U/mL) = Δ[AMP]/Δt × Vreaction, with typical values of 0.8–3.5 U/mL for purified PDEII 3. The method's sensitivity (detection limit ≈ 10 μM, 5% CV) and non-destructive nature allow kinetic parameter determination: Km = 85 ± 12 μM, Vmax = 4.2 ± 0.3 μmol/min/mg for PDEII 3. For creatine kinase, magnetization saturation transfer (MST) quantifies the forward (PCr + ADP → Cr + ATP) and reverse (ATP + Cr → PCr + ADP) reaction rates 12. Selective saturation of γ-ATP (3 s, B1 = 50 Hz) reduces PCr signal by 20–40% in CK-active tissues; the pseudo-first-order rate constant kf = (M₀,PCr - Mss,PCr)/(Mss,PCr × T1,PCr), where M₀,PCr and Mss,PCr are equilibrium and steady-state PCr magnetizations 12. In human skeletal muscle, kf = 0.35 ± 0.05 s⁻¹ at rest, increasing to 0.65 ± 0.10 s⁻¹ during moderate exercise (50% VO₂max), reflecting enhanced ATP buffering 12. Pathological states alter these kinetics: in heart failure, kf decreases by 40–60%, correlating with reduced contractile reserve 12.
31P-NMR coupled with ergometry enables precise correlation of mechanical work with metabolic state 11,13. The experimental setup positions a limb (typically forearm) within a 2.0T magnet bore, with a hand lever connected to a Cybex dynamometer outside the magnet via a non-magnetic linkage 11,13. Subjects perform rhythmic isometric contractions (e.g., 1 contraction every 6 s, 50% maximal voluntary contraction, 10 min duration) while 31P spectra are acquired continuously (TR = 6 s, 1 spectrum per contraction cycle) 13. Key metabolic indices include: (1) PCr/Pi ratio, declining from 4.5 ± 0.8 at rest to 1.2 ± 0.3 at steady-state exercise, reflecting the balance between ATP synthesis (oxidative phosphorylation) and consumption (actomyosin ATPase) 11,13; (2) PCr recovery time constant (τPCr) post-exercise, a direct measure of mitochondrial oxidative capacity (τPCr = 30–60 s in healthy muscle, >90 s in mitochondrial myopathies) 13; (3) pH decline, from 7.05 ± 0.05 to 6.70 ± 0.10 during intense exercise due to glycolytic lactate production 11. Plotting work rate (J/s) versus Pi/PCr yields a linear relationship (R² = 0.92–0.97) with slope proportional to mitochondrial density 13. In patients with McArdle disease (myophosphorylase deficiency), PCr depletion is accelerated (PCr/Pi < 0.5 within 2 min) despite low work rates, and pH remains stable (no lactate accumulation), providing diagnostic
| Org | Application Scenarios | Product/Project | Technical Outcomes |
|---|---|---|---|
| Innovation Academy for Precision Measurement Science and Technology CAS | Brain energy metabolism evaluation in neurovascular diseases, functional brain studies during visual/motor stimulation, and cerebrovascular disease diagnosis where metabolic stress detection requires decoupling from hemodynamic responses. | Multinuclear MRI System for Blood Flow-Dependent Metabolism | Simultaneous acquisition of 31P-CSI for phosphocreatine metabolism and 129Xe-CSI for red blood cell perfusion, enabling blood flow-normalized metabolic assessment with spatial resolution of 3-5mm and temporal resolution of 5-10min. |
| Ningbo University | Enzyme activity assays in biological research, phosphate-modifying enzyme screening, cAMP hydrolysis monitoring, and drug discovery applications requiring label-free, real-time enzymatic reaction tracking. | 31P-NMR Phosphodiesterase Activity Assay Platform | Real-time enzyme activity quantification with detection limit of 10μM, 5% coefficient of variation, linear quantification range 25-500μM (R²>0.998), enabling non-destructive kinetic parameter determination (Km=85±12μM, Vmax=4.2±0.3μmol/min/mg for PDEII). |
| Sidra Medical and Research Center | Cancer diagnosis and grading, tumor malignancy assessment, treatment efficacy monitoring, and cancer prognosis prediction in breast, prostate, ovarian, brain, and other cancerous tissues overexpressing creatine kinase enzyme. | CK-CEST Cancer Imaging Technology | Non-invasive creatine kinase activity mapping via chemical exchange saturation transfer with CESTR values of 15-35% in cancer tissues versus 5-12% in healthy tissue, achieving 85-92% accuracy in malignant versus benign tumor discrimination. |
| The General Hospital Corporation | Intracellular pH monitoring in cancer tissues, metabolic acidosis detection, ischemia assessment, and physiological pH mapping in brain, muscle, and other tissues requiring non-invasive pH quantification. | Ratiometric CEST pH Measurement System | Quantitative pH mapping using dual-saturation power levels (B1=0.5μT and 2.0μT) with enhanced contrast-to-noise ratio, providing pH accuracy of ±0.05 units and simultaneous creatine concentration determination in tissue. |
| Laboratory Corporation of America Holdings | Clinical diagnostics for kidney stone risk assessment, renal function monitoring, biomarker detection in urine and blood samples, and personalized medicine applications requiring accurate metabolite concentration measurements. | NMR Biomarker Analysis Platform | High-sensitivity NMR spectroscopy for citrate and creatinine quantification in biological samples, enabling personalized treatment pathway determination for kidney stone formation and renal disease assessment. |