Hydrophobic αF-helix serves as a central scaffold.

(A). αF-helix is very hydrophobic and creates an interface with multiple motifs including the Catalytic machinery (colored in tan) and tethering sites (in sand) of PKA C-subunit. Two key charged residues, D220 and E230, sit at the two ends of αF-helix. (B). αF-helix is a central scaffold for assembly of the entire molecule. D220 forms two H-bonds to Y164 and R165 of YRD motif, and E230 salt-bridges to P-2 Arg of PKI. (C). Sequence alignment of αF-helix segment of PKA with other kinases. All share a very hydrophobic helix, the highly conserved residues, D220 are colored in red and E230 in red box.

Hydrophobic interface anchors the Catalytic machinery to the αE-helix and αF-helix.

(A). I150 (in red) from αE-helix plays an important role by docking to αF-helix and Catalytic loop. Three residues from β7, L172, L173, and I174 (in dark tan) assemble the hydrophobic surface from αE-helix to ATP pocket. L173 and I174 are part of C-Spine. R-spine are also shown in red. D220 bridges H158 from αE-helix to YRD motif. (B). The logo of spines shows how important those hydrophobic residues are. The sequence of β6-β9 segment is also shown.

The buried surface of αC-β4 loop.

(A). αC-β4 loop (in teal) in PKA links the N-lobe to its C-lobe. The hydrophobic residues on the loop docks to αE-helix, F18 and L19 from αA-helix and Y306 from C-tail are also part of this surface. F100 is colored in red. (B). The H-bond network of αC-β4 loop. Y156 from αE-helix H-bonds to N99 backbone amide. Three water molecules also help to nucleus the network. (C). Sequence alignment of αC-β4 loop of PKA with other kinases. The highly conserved residues are highlighted, E91 colored in red, L95 and L106 in red box.

αC-β4 loop is a very stable element.

(A). Hydrophobic surface of αE-helix which anchors to αC-β4 loop, αF-helix and tethering sites. (B). Superimposition of αC-β4 loop in structures of PKA and other kinases. αC-β4 loop is very stable, not only in different conformations of PKA, but also in other kinase structures including active and inactive Src. (C). Sequence alignment of αE-helix of PKA with other kinases. Y156, in red box, is highly conserved.

Hydrophobic residues play key roles in PKA.

(A). High-affinity binding of PKI is mediated by the hydrophobic surface of an amphipathic helix and P+1 inhibitor site, both highlighted in red. The sequence of PKI is also shown. (B). Hydrophobic interface between the αE and αF-helices. One side of αF-helix (in red) is shown in tan, and another side in sand, which is the same color coding as Figure 1A. (C). Hydrophobic pocket surrounding ATP. L172, L173, and I174 from β7 anchor this ATP pocket to αE-helix. F327 from C-tail is highlighted in red.

Local Spatial Pattern (LSP) alignment method.

(A). Spatial patterns detected by the LSP-alignment are formed by Cα-Cβ vectors. Both mutual positions and orientation in space of these vectors are taken into account. (B). Two major centralities that characterize graphs. Degree centrality (DC) is the sum of connections for each node. The highlighted node in the middle has the highest DC value of 8. Residues with high level of DC are local “hubs”. Betweenness Centrality (BC) of a node is the number of shortest paths between all other pairs of nodes that pass through this node. Two highlighted nodes are the main connectors in the graph with BC equal to 6. Residues with high level of BC are global connectors between local hubs. (C) Visualization of Protein Residue Networks (PRNs) of PKA before and after the mutation laid out by Gephi software package using ForceAtlase2 algorithm. The diameter of nodes is proportional to the residues DC. Nodes with higher BC have darker color. Four residues from the αC-β4 loop (103–106) are highlighted by oval. The Hinge region is indicated by the dashed oval.

Distribution of Betweenness centrality vs. Degree centrality for PKA residues: wt C-subunit (A) and F100A (B).

Residues with Degree centrality (DC) are local “hubs”, representing the most stable parts of the molecule. Residues with high Betweenness centrality (BC) are global connectors and are “bottlenecks” between densely interconnected “hubs”. Residues of the central αF-helix are shown as brown circles. Their high levels of DC and BC don’t change upon the F100A mutation, meaning that the helix remains to be a central structural element in the PKA mutant. Five residues from the αC-β4 loop are highlighted as blue dots. The significant increase in BC and a certain increase in DC shows that in the mutant this group of residues becomes a significant point of connectivity between the kinase lobes. On the contrary, the highly conserved residues highlighted as red circles show a drastic decrease in BC. Error bars represent standard error calculated for five 10 ns trajectories.

Changes of DC and BC in PKA upon F100A mutation.

The middle graphs represent changes in the corresponding parameters. Positive changes are mapped on the PKA structure (left) with dark red color corresponding to the maximum changes. On the right the negative values of the changes are mapped on the structure to illustrate their distribution. Dark blue color corresponds to the most negative values. Error bars represent standard error calculated for five 10 ns trajectories. The secondary structure of PKA is shown on top of the sequence axis for reference.

Dynamic feature of K105 on αC-β4 loop.

(A). In the wt PKA:ATP:Mg:PKI complex structure, the side chain of K105 is likely interact with E107 and E121, and its main chain carbonyl forms a H-bond to backbone amine of E121. (B). In the wt Apo PKA structure, the K105 side chain interacts with β-turn, whereas its main chain carbonyl still H-bonds to backbone amines of E107 and E121. The H-bonds are shown as dash line in black, and the non H-bond are colored in green and the distance are labeled.

Energy landscapes of wt C-subunit and F100A mutant in binary complex.

The free energy landscapes (FEL) were generated based on the principal component analysis (PCA) for the wt C-subunit (A) and F100A mutant (B). The H-bonds are shown in the black dash line and the non H-bond in green along with the distance in Angstroms (Å). The probability distribution plots for specific side chain dynamics are also shown; (C) K105 NZ to E107 OE2, (D) K105 NZ to E121 OE2, (E) K105 NZ to N99 OD1, and (F) N99 N to Y156 OH. The probability density function (PDF) is a relative measure of how densely data points are distributed along the x-axis.

Energy landscapes of wt C-subunit and F100A mutant in ternary complex.

The free energy landscapes (FEL) were generated based on the principal component analysis (PCA) for the wt C-subunit with PKI (A) and F100A mutant with PKI (B). The H-bonds are shown in the black dash line and the non H-bond in green along with the distance in Angstroms (Å). The probability distribution plots for specific side chain dynamics are also shown; (C) K105 NZ to E107 OE2, (D) K105 NZ to E121OE2, (E) K105 NZ to N99 OD1, and (F). N99 N to Y156 OH. (G) the enhanced side chain dynamics of K105 in the F100A ternary complex: the side chain of K105 toggles to its neighboring residues, E121 (white, global energy minima), N99 (green, first secondary minima), and E107 (sand, second secondary minima). The probability density function (PDF) is a relative measure of how densely data points are distributed along the x-axis.

High-affinity binding of ATP and PKI converge at the αC-β4 loop.

(A). V104 in αC-β4 loop (in teal) is hydrophobically anchored to the adenine ring of ATP. The main chain carbonyl of K105 forms an H-bond to the backbone amide of E121, while its main chain carbonyl hydrogen bonds to ATP. The distance between K105 and E121 side chains is strengthened in the F100A mutant. Two spine residues, L95 and L106, in this loop are also shown. The linker that joins the N- and C-lobes (red) is flanked by E121 and E127. (B). Hydrophobic capping of the adenine ring of ATP is mediated mostly by N-lobe residues including V104 in the αC-β4 loop, as well as F327 in the C-terminal tail. In contrast, the P-3 to P+1 peptide is anchored to the catalytic machinery of the C-lobe. By binding to L173 in the C- lobe, the adenine ring completes the C-Spine, and thus fuses the adenine capping motif in the N-lobe with the extensive hydrophobic core architecture of the C-lobe. In the fully closed conformation, the side chain of Y330, also in the C-terminal tail, is anchored to the ribose ring of ATP. The only direct contact of the peptide/catalytic machinery with the N-lobe is mediated by the P-3 arginine which binds to the ribose ring of ATP and to E127 in the linker. In the fully closed conformation the P-3 arginine also binds to the side chain of Y330 in the C-terminal tail. Some of the mutations that disrupt the synergistic high-affinity binding of ATP and peptide/protein (E230Q, Y204A, F327A, L173A, and E31V) are highlighted. The hydrophobic residues in the amphipathic helix and P+1 inhibitor site of PKI are shown in red.