3D theoretical structure of hIL-6 − hIL-6R complex and the key amino acids in hIL-6 identified by hIL-6RTo deeply analyze the intermolecular binding energies and modes between hIL-6 and hIL-6R at the molecular level, this study employed appropriate molecular force field parameters and molecular dynamics optimization techniques. Using the crystal structures of hIL-6 (PDB code: 1il6)25 and hIL-6R (PDB code: 1N26)26 as model, the 3D theoretical structures of hIL-6 and hIL-6R were obtained dealing with Discover_3 program under CVFF force-filed (Insight II 2000, MSI, San Diego) as shown in Fig. 1A, B. Subsequently, the spatial complex structures of hIL-6 and hIL-6R were constructed and optimized using the same program and force field, drawing from the crystal structure of their complex (PDB code: 1p9m)8. According to the 3D theoretical complex structure of the hIL-6 and hIL-6R proteins as shown in Fig. 1C, the key amino acids in the IL-6 binding with IL-6R were predicted using the distance geometry method and computer graphics technology. The distribution of the key residues was located within the complex structure, as shown in Fig. 1D. Based on the van der Waals interactions and coulombic binding activity, the interaction distance between the key amino acid residues of the parent proteins (i.e., hIL-6R and hIL-6) was defined as 6 Å. To further understand the binding between hIL-6R and hIL-6, a local map of the binding between the IL-6R and IL-6 proteins is shown in Fig. 1E. From the docking results, the crucial amino acid residues in hIL-6 that bind to hIL-6R were identified as Q174, R178, and R181. These residues have been highlighted in several previous studies (Table S1)8,16,18,19,21.Figure 1The 3-D modeling structures of hIL-6 and hIL-6R. (A,B) 3-D theoretical structures of hIL-6 and hIL-6R, respectively; (C) 3-D theoretical complex structure of the hIL-6 − hIL-6R; (D) prediction of key amino acids in the hIL-6 binding with hIL-6R; (E) local map of the binding between the hIL-6R and hIL-6; (F) binding energy of hIL-6 and mutants to hIL6-R. 3-D Computer-guided homology modeling and molecular docking were performed using Insight II 2000 software (MSI, San Diego).Mutant design based on the critical amino acids and amino acid propertiesConsidering the properties of the amino acid, acidic or basic amino acids that are prone to form ionic bonds were mutated to alanine. Glutamine, which is prone to forming hydrogen bonds, was also substituted for alanine, which resulted in five single amino acid mutants (R167A, E171A, Q174A, R178A, and R181A). Some special single amino acid mutations were also designed, such as mutant with charge reversal (R178E) and one mutant with a positive charge replaced with a hydrophobic amino acid (R181V). Multiple amino acid mutants in which multiple amino acids were simultaneously replaced with alanine were also generated, which allowed for a comprehensive assessment of the effects of alanine mutations on the elimination of ionic or hydrogen bonds (R167A/E171A/Q174A/R178A/R181A). Furthermore, the binding energies between hIL-6 or its mutants and hIL-6R were calculated and compared, as listed in Fig. 1F. Considering that hIL-6R binds at the site I position of hIL-6, the site I position includes not only the D helix C-terminal end but also the C-terminal end of the AB-loop, according to previous studies27. To evaluate the contribution of the AB-loop amino acids, a D helix deletion mutant (ΔQ155–M183) was simultaneously designed. From the result of the binding energy of hIL-6 to hIL-6R (Fig. 1F), it was found that the IL-6 mutants R167A and E171A had similar binding energy to IL-6R compared to hIL-6 WT. However, the mutants Q174A, R178A, and R181A/V exhibited a higher binding energy than hIL-6 WT. Surprisingly, R178E showed the highest energy, rising from − 118.39 to − 73.99 kCal/mol. This substantial increase was attributed to the loss of electrostatic interactions between the positively charged R178 of hIL-6 and the negatively charged E278 of hIL-6R, as well as the loss of intermolecular hydrogen bonds between R178 and Q281, Y230 of hIL-6R. These findings collectively indicate that the R178E mutant has the lowest affinity for hIL-6R. Moreover, the binding energy of a multipoint mutant (R167A, E171A, Q174A, R178A, and R181A) and a D helix deletion mutant (ΔQ155–M183) could not be calculated, which suggests a loss of binding affinity to IL-6R. These results underscore the critical role of specific amino acid interactions in maintaining the high affinity binding between hIL-6 and hIL-6R.Backbone conformation of WT hIL-6 and its mutantsThe backbone conformation of hIL-6 and its mutants was analyzed through CD spectra that was recorded in the far-UV region (190–260 nm). The CD spectra revealed a characteristic negative band at 206 nm and a smaller peak at around 220 nm, which indicates the presence of antiparallel β-sheets and α-helices in both hIL-6 and most of its mutants (Fig. 2A). Comparing the percentage of α-helices to that of the WT hIL-6 (Fig. 2B), most mutants exhibited a high similarity in α-helical content with the WT hIL-6. However, the multipoint mutant and the D-helix deletion mutant showed slightly lower percentages of α-helices. This observation suggests that the deletion in the D-helix region may lead to structural instability, which results in a reduction in the α-helix content. Overall, the CD spectra of the hIL-6 WT and its mutants displayed similar patterns, with overlapping curves and only minor variations in the mean residue ellipticity [θ] at the characteristic peaks. This indicates that the overall backbone conformation of the mutants does not significantly differ from that of the WT hIL-6 protein.Figure 2Secondary structural features of WT hIL-6 ant its mutants. (A) Far-UV CD spectra of the indicated hIL-6 mutants, illustrating their folding into native secondary structures. The proteins were dissolved in PBS buffer at a concentration of 0.5 mg/mL, pH 7.4, 20 ℃. (B) Percentage of α-helices (%) folded compared to the WT hIL-6.Binding kinetics analysis by ELISA of hIL-6R to hIL-6 and its mutantsThe binding kinetics of hIL-6R to hIL-6 and its mutants were first investigated using ELISA (Fig. 3). The hIL-6 and its mutants was first coated, and the hIL-6R with the Fc tag was serially diluted and incubated before performing the color reaction using HRP-conjugated goat anti-human IgG. The coating concentrations of hIL-6 and its mutants and the concentration of the enzyme-labeled secondary antibody were optimized to allow for the generation of a complete binding curve of hIL-6 to hIL-6R. As shown in Fig. 3A, the inability of the hIL-6 mutants Q174A, R178 A/E, and R181A/V to form an efficient engagement with hIL-6R or have decreased binding capacity, which suggests that Q174, R178, and R181 are critical amino acids for hIL-6R binding to hIL-6, which is consistent with findings from previous studies8,16,18. In addition, multipoint mutants (R167A, E171A, Q174A, R178A, and R181A) also failed to bind to hIL-6R, thus demonstrating the role of these amino acids. The D helix deletion mutant (ΔQ155-M183) also failed to bind to hIL-6R, thus illustrating that although the hIL-6 site I includes the C terminus of the D helix as well as that of the AB-loop, the D helix is more important when binding to hIL-6R, such that after deletion, no efficient binding could be formed.Figure 3Binding affinity of hIL-6R to hIL-6 (mutants) as measured by ELISA. (A) Binding curve of hIL-6R to hIL-6 (mutants); (B) EC50 values of hIL-6 (mutants) binding to hIL-6R. Experiments were performed in triplicate and the error bars denote mean ± SD, n = 3.Binding kinetics analysis by SPR assay of hIL-6R to hIL-6 and its mutantsSPR biosensing is a technique that allows for the label-free detection of various analytes and real-time monitoring of biomolecular interactions28. In this study, we utilized SPR technology to analyze the interaction affinity between hIL-6R and hIL-6 or its mutants. The results, as shown in Fig. 4 and Table 1, were consistent with those obtained from ELISA in Fig. 3. For example, the hIL-6 mutants R167A and E171A exhibited similar affinity for hIL-6R compared to hIL-6 WT. However, the mutants Q174A and R178A showed a significant reduction in their binding affinity for hIL-6R, with a 36- and 41-fold decrease, respectively. This observation was unique and has not been previously reported. On the other hand, the hIL-6 mutants R178E, R181A/V, multipoint mutants, and the D helix deletion mutant showed no binding to hIL-6R in the SPR assay.Figure 4Affinity kinetics between hIL-6 mutants and hIL-6R assayed by SPR. hIL-6R with Fc tag was immobilized onto a CM5 chips via an anti-human Fc antibody and different concentrations of hIL-6 mutants was injected. Affinity kinetics were analyzed by the Langmuir’s 1:1 model fit on seven serial dilutions of hIL-6 mutants from 46.21 nM and flow speed of 30 μl/min. (A) Schematic diagram of binding affinity between hIL-6 mutants and hIL-6R by SPR; (B–K) representative examples of curve fits for the affinity kinetic analysis of IL-6 mutants.Table 1 Affinity of the interaction between hIL-6R and hIL-6/mutants by Biacore SPR.Binding kinetics analysis by SPR assay of gp130 to hIL-6 (or its mutants) − hIL-6R complexUnder expected physiological conditions, hIL-6, hIL-6R, and gp130 form a heterohexameric complex. Typically, hIL-6 preferentially interacts with hIL-6R, and this complex then subsequently binds to gp13029. However, a few previous studies have reported that certain IL-6 mutants and viral IL-6 (vIL-6) are able to directly binding to gp130 without the involvement of hIL-6R10,11. To mimic the classical binding process, monomeric hIL-6R was pre-incubated with different concentrations of hIL-6 and its mutants, and the mixture was flown over a chip sensor with immobilized monomeric gp130 (Fig. 5A). The resulting affinity data for gp130 with the hIL-6 (or its mutants) − hIL-6R complex using SPR are shown in Fig. 5B–K and Table 2. In the WT group, the KD value of gp130 for the hIL-6 − hIL-6R complex was 2.48 nM, which indicates that the SPR detection system was functioning appropriately. Among the IL-6 mutants, R167A and E171A showed a higher affinity to form hIL-6–hIL-6R–gp130 complexes than the hIL-6 WT, while the mutants Q174A, R178A, and R181A had a lower affinity than the WT. The affinity of R181V was about two-fold lower than hIL-6 WT. As expected, the mutants R178E, multipoint mutants, and D helix deletion mutants failed to produce measurable binding signals, which were consistent with hIL-6 for the hIL-6R binding results in SPR and ELISA (Figs. 3,4).Figure 5Affinity kinetics between hIL-6 mutants and hIL-6R − gp130 assayed by SPR. gp130 was immobilized onto a CM5 chips via standard amine coupling and different concentrations of hIL-6 mutants with a saturated concentration of hIL-6R were pre incubated and the mixture were injected. Affinity kinetics were analyzed by the Langmuir’s 1:1 model fit on seven serial dilutions of hIL-6 mutants from 250 nM and flow speed of 30 μL/min. (A) Schematic diagram of binding affinity between hIL-6 mutants and hIL-6R − gp130 by SPR; (B–K) representative examples of curve fits for the affinity kinetic analysis of hIL-6 mutants.Table 2 Affinity of the interaction between hIL-6/mutants and hIL-6R/gp130 by Biacore SPR.Impact of hIL-6 and its mutations on the phosphorylation of STAT3 in HepG2 cellsThe interaction between human interleukin-6 (hIL-6) and its receptor hIL-6R, culminating in the formation of the hIL-6 − hIL-6R − gp130 complex, is pivotal for the activation of the JAK-STAT3 signaling cascade29. Our investigation focused on the functional characteristics of the R167A, E171A, and R178E mutants, analyzing their binding propensities to hIL-6R and the hIL-6R − gp130 complex. We assessed the activation of STAT3 phosphorylation by performing Digital western blot analysis. As shown in Fig. 6, hIL-6 WT significantly increased STAT3 phosphorylation compared to the negative control (without hIL-6). In contrast, the hIL-6 mutant R178E failed to stimulate STAT3 phosphorylation, which is consistent with previous findings that reported its lack of biological activity16. Interestingly, the R167A and E171A mutants all induced higher levels of STAT3 phosphorylation than hIL-6 WT. Furthermore, we examined the impact of the Q174A, R181A, and R181V mutants of hIL-6 on STAT3 activation (Fig. S3). Our results indicate that these mutants exert varying degrees of inhibition on STAT3 phosphorylation, corroborating earlier studies16,18,19,21. These results suggest a clear correlation between the intracellular STAT3 phosphorylation and the affinity of hIL-6 to the hIL-6R − gp130 complex.Figure 6Activation of phosphorylation of STAT3 in HepG2 cells by hIL-6 or its mutants. HepG2 cells were treated with 50 ng/mL hIL-6 or its mutants for 15 min, and cell lysates were analyzed to detect phosphorylation of STAT3. Untreated cells were used as the negative control (NC). (A) Representative, digital western blot images of expression of P-STAT3, STAT3 and β-Actin upon stimulation with hIL-6 WT or R167A, E171A and R178E mutants, respectively; (B) statistical analysis of hIL-6 mutants compared with hIL-6 WT in term of STAT3 phosphorylation levels in HepG2 cells, based on three independent experiments. Data is presented as mean ± SD, n = 3. **p < 0.01; and ***p < 0.001.Effects of hIL-6 and its mutants on STAT3 activation in human leukocytesTo further confirm the physiological role of hIL-6 and its mutants on leukocyte phosphorylation, a flow cytometry assay was performed, as previously reported30. As depicted in Fig. 7, the presence of hIL-6 resulted in a notable increase in STAT3 phosphorylation in leukocytes (10%) compared to the negative control where hIL-6 was absent. Notably, the hIL-6 mutants R167A and E171A both induced significantly higher levels of p-STAT3 (around 20%) compared to hIL-6 WT. On the other hand, the mutant R178E displayed minimal induction of STAT3 activation. These findings are consistent with the results obtained from digital western blot analysis.Figure 7Activate of phosphorylation of STAT3 in Leukocytes by IL-6 mutants. Human leukocytes were preincubated with 50 ng/mL IL-6 mutants for 15 min, followed by permeabilized and stained. Untreated cells were used as the negative control (NC). Phosphorylation of STAT3 was assessed using flow cytometry with PE labeled anti-pSTAT3 (pY705) monoclonal antibody. (A–C) Gating strategy and representative histogram graph of P-STAT3+ cells.; (D) statistical analysis of percentage of P-STAT3+ cells upon stimulation with hIL-6 WT or R167A, E171A and R178E mutants, based on three independent experiments. The data is presented as mean ± SD, n = 3. *p < 0.05; **p < 0.01.
