Fig. S5.

Nucleotide-dependent polymerization of hGBP1_F. (A and B) Absorbance of the sample with different concentrations of hGBP1_F (A) and hGBP1_NF (B) after addition of 1 mM GTP (t = 0 s). (C and D) Analysis of polymer formation of 20 µM hGBP1_F (C) and hGBP1_NF (D) by dynamic light scattering. The particle size, given as the hydrodynamic radius (R_H) in nanometers without nucleotide (blue) and in the presence of 1 mM GTP (gray) or GDP·AlFx (orange) (300 µM AlCl₃, 10 mM NaF, and 200 µM GDP) was monitored over 250–450 s at 25 °C. Farnesylated hGBP1 transiently forms very large polymers during GTP hydrolysis with a radius around 1,000 nm. In contrast, the addition of GDP·AlFx leads to an irreversible formation of polymers with an average R_H of 25 nm. Nucleotide-free hGBP1_NF shows a R_H of 5 nm. Complexation with GDP·AlFx leads to a slight irreversible increase to 6.5 nm and addition of GTP to a transient increase to a maximum value of 7.2 nm. (E) Reversibility of GTP-induced hGBP1_F polymers investigated by sedimentation. During a typical turbidity experiment (see A, 10 µM hGBP1_F and 1 mM GTP added at t = 0 s), samples were taken at indicated time points (arrows), spun down, and the fraction of protein in pellet (P) and supernatant (S), respectively, was analyzed by SDS/PAGE (100-µL samples were taken, centrifuged at 30,000 × g, 4 °C and for 1 min. After removal of the supernatant, the pellet was resuspended in the same volume of buffer, and corresponding fractions of protein in pellet (P) and supernatant (S) were analyzed by SDS/PAGE. Considering that the proportion of protein in the pellet represents protein complexes large enough to sediment, e.g., polymers, SDS/PAGE analysis reflected the same observation as already monitored by time-trace of absorbance: The GTP-induced increase and decrease of the absorbance considered as reversible assembly and disassembly of hGBP1_F polymers was accompanied by increase and decrease of protein share in the pellet.