Mestrenova 10

UNAG-4P was dissolved in 99.9% D₂O at a concentration of 6 mg/ml and spectra were recorded at 25 °C using a Varian 500 NMR system spectrometer (Agilent). NMR experiments were processed and analyzed using the MestReNova 10.0 (MestreLab Research) software. Atom names and assignment is given in Supplementary Fig. 5.
Signal assignment of ¹H NMR experiments (500 MHz, ¹H-¹H-COSY, D₂O): δ = 7.98 (d, 1H, ³J_HH = 8.2 Hz, H-U6), 6.0–5.98 (m, 2H, H-U5, H-1′), 5.53 (dd, 1H, ³J_HH = 7.3 Hz, ³J_HP = 3.2 Hz, H-1), 4.40–4.37 (m, 2H, H-2′, H-3′), 4.30 (m, 1H, H-4′), 4.28–4.25 (m, 1H, H-5′a), 4.22–4.18 (m, 1H, H-5′b), 4.08–3.97 (m, 4H, H-2, H-3, H-4, H-5), 3.90–3.84 (m, 2H, H-6a, H-6b), 2.09 (s, 3H, H-8) ppm.
Assignment of ¹³C NMR experiments (152.7 MHz, APT, HSQC, D₂O): δ = 174.63 (C, C=O), 166.14 (C, C=O), 151.72 (C, C=O), 141.55 (CH, C-U6), 102.59 (CH, C-U5), 94.03 (CH, J_CP = 6.2 Hz, C-1), 88.38 (CH, C-1′), 83.16 (CH, J_CP = 9.0 Hz, C-4′), 73.71 (CH, C-3′), 73.25 (CH, J_CP = 4.7 Hz, C-5), 72.03 (CH, J_CP = 5.7 Hz, C-4), 70.41 (CH, C-3), 69.57 (CH, C-2′), 64.90 (CH₂, J_CP = 6.2 Hz, C-5′), 60.11 (CH₂, C-6), 53.35 (CH, J_CP = 8.5 Hz, C-2), 22.00 (CH₃, C-8) ppm.
Assignment of ³¹P NMR decoupling experiments (202.4 MHz, D₂O): δ = 0.91 (s, P-4), −11.46 (d, J_PP = 19.7 Hz, P-5′), −13.16 (d, J_PP = 20.0 Hz, P-1) ppm.

NMR experiments were performed on a Varian Direct Drive 600 MHz spectrometer (California, USA) at 25°C. For each experiment samples were freshly prepared in deuterium oxide at a protein concentration of 100 μM in 5 mM sodium phosphate buffer, 0.05% sodium azide, pH 7.2, containing the desired SDS concentration. Diffusion-ordered spectroscopy (DOSY) spectra were obtained using a gradient length of 3 ms and a diffusion delay of 250 ms. Gradient strength was previously calibrated using the HDO signal in a doped D₂O standard. The DOSY data sets were composed by 30 gradient strengths with 128 scans every increment. The DOSY data were processed using MestreNova 10.0 (Mestrelab Research S.L, Spain). Diffusion coefficients were obtained by fitting the intensity decays versus the gradient strength as described elsewhere [46 ]. The hydrodynamic radii were calculated with the Stokes-Einstein equation, R_h = k_BT/6πηD, where k_B is the Boltzmann constant, T is the absolute temperature, D is the diffusion coefficient and η is the viscosity of D₂O at 298 K.

All ¹H ¹D NMR experiments were performed at 310 K on a Bruker Avance III 500 MHz spectrophotometer equipped with a TXI probe (Bruker Daltonics, Bremen, Germany). A 1D CPMG pulse sequencer (Carr-Purcell-Meiboom-Gill) with cpmgpr1d parameters, 73,728 points in F1, 12,019.230 Hz spectral width, 2048 pass, with a cyclic delay of 4 secs, was used for the measurement, with water suppression using presaturation. We used the reference chemical database Madison-Qingdao Metabolomics Consortium Database to analyze the obtained metabolites^{103 (link)} and all determined metabolites were compared with reference compounds. We acquired, processed, and evaluated the results using Bruker Topspin 3.1, and MestReNova 10.0 software (Mestrelab Research, Santiago de Compostela, Spain). We related the concentration of metabolites to the TSP standard. We performed measurements in triplicate for each monitored condition.