Thus the most significant gains in Tm due to protein deuteration are only observed at temperatures around 50 K and below. Unlike the 1/Tm temperature dependence, the spin longitudinal relaxation rate (1/T1) does not show any major difference between the non-deuterated and all-deuterated samples, which indicates that within this temperature range, the nuclear spins

do not play a significant role in the spin–lattice relaxation mechanism. For both samples 1/T1 shows slight temperature dependence, and during the observed temperature range BYL719 it does not approach the value of 1/Tm suggesting that T1 processes do not have a significant effect on the electron spin echo dephasing [3]. We have shown the strong effect of protein deuteration on Tm. However as Tm is extended it becomes more sensitive to other effects like instantaneous diffusion and electron spin–spin diffusion [17]. The electron spin echo dephasing observations, in which the histone octamer was increasingly segmentally deuterated, showed, in addition to strong ESEEM modulations, an oscillation resulting from the electron dipole–dipole interaction between the two spin labels present on the protein (see Fig. 2). Such distance dependent dipolar PARP assay interactions were only observed in the case of H3-D/H4-D and All-D samples due to their long Tm. In Fig. 2 we can see that the longer the Tm, the more pronounced the electron dipole–dipole interaction.

The observation that the 2 pulse ESE experiment is capable of detecting electron spin–spin interactions in biradicals has been made previously [16]. In a two-pulse echo experiment, when the second pulse is applied and flips two dipole-coupled spins simultaneously, the dipole interaction does not refocus and leads ID-8 to a dephasing of spin pairs, this effect is known as instantaneous diffusion. Two cases

can be envisaged. One situation is where the spin pairs are randomly distributed and so there will be a wide distribution of dipolar interactions, D, and therefore the echo oscillations, occurring at a range of frequencies, average out, leaving only an exponential-like echo decay contribution. In the second case, when the spin pairs have a defined dipole interaction, D, the echo decay will be modulated by the dipole–dipole frequency, y(τ) ∼ cos(Dτ) [16]. The H3-D/H4-D and All-D histone octamer constructs clearly fulfil the requirements for a dipolar interaction to be observed in a 2 pulses ESE experiment since they are double spin labeled, with a defined dipole–dipole interaction, and have long Tm. The Fourier transform of the ESE decay yielded a dipolar coupling which is in good agreement with the PELDOR data (see supplementary material Fig. S4a and b). In order to get more insight into the effect of deuteration, we have also studied the concentration dependence of Tm ( Fig. 5) for the fully deuterated sample at 50 K.