Fig 1.
Three mutually exclusive binding modes.
The three binding modes aligned together to the same CD44 template. Blue surface represents HA in the crystallographic mode (B-form, as in Ref. [10]), green represents the parallel mode, and red describes the upright binding mode (as in Ref. [4]). These orientations represent the situation in the beginning of the gathering simulations (Table A in S1 File).
Fig 2.
Molecular details of the HA–CD44 binding modes.
The three binding modes found in this study. Tan surface represents the CD44 HABD domain, the light green spheres depict R41 (important binding residue), and the multicolored rod is the HA16 oligomer. The images shown at the top are snapshots from the starting frames of the gathering simulations (Table A in S1 File). The figures at the bottom describe the most important binding residues in each binding mode. The more reddish the surface becomes, the more contacts it holds, so that red color corresponds to 350 or more contacts on average.
Table 1.
HA binding residues of CD44-HABD identified in the literature.
Fig 3.
Contacts between HA and the arginines of CD44.
Average number of contacts of HA with each arginine residue in the CD44 HABD, calculated from the three replica systems (3 × 1000 ns) per each binding mode and the A-form simulations spanning 300 ns (Gathering simulations in Table A in S1 File). The error bars represent standard errors.
Fig 4.
Stability of each binding mode.
Fluctuations observed in the HA structure for each binding mode. The tan surface represents the protein in the first frame of the trajectory. The green spheres are the arginine residues. The multicolored sticks represent the HA conformation, plotted here every 50 nanoseconds of the trajectory. The frames have been taken from the first replica system of each Gathering simulation (Table A in S1 File) (data based on the other replicas were consistent with the data shown here).
Fig 5.
HA–CD44 free energy ΔG (kJ mol−1) as a function of distance ξ to the binding site. Pull groups are the centre of mass of HA and the centre of mass of the HABD residues 75–79. The error bars, based on weighted histogram analysis, are shaded accordingly.
Table 2.
In silico electrophoresis as qualitative measurement of stability.
Fig 6.
Time-evolution of CD44–HA interaction.
Contour plots describing the contacts of HA residues with two CD44 proteins as they diffuse along HA. Data are shown for two independent replica systems (proteins P1 and P2 in replica 1, and proteins P3 and P4 in replica 2). Vertical axis runs through the HA residues, horizontal axis depicts time through the 3 μs simulation time span, and the contours highlight the number of HA-CD44 contacts in terms of color code shown on the right. The dashed lines guide the eye to visualize the motion of the proteins. The data are calculated from the ‘Clustering’ simulations in Table A in S1 File.