Mapping N- to C-terminal allosteric coupling through disruption of a putative CD74 activation site in D-dopachrome tautomerase

The macrophage migration inhibitory factor (MIF) protein family consists of MIF and D-dopachrome tautomerase (also known as MIF-2). These homologs share 34% sequence identity while maintaining nearly indistinguishable tertiary and quaternary structure, which is likely a major contributor to their overlapping functions, including the binding and activation of the cluster of differentiation 74 (CD74) receptor to mediate inflammation. Previously, we investigated a novel allosteric site, Tyr99, that modulated N-terminal catalytic activity in MIF through a "pathway" of dynamically coupled residues. In a comparative study, we revealed an analogous allosteric pathway in MIF-2 despite its unique primary sequence. Disruptions of the MIF and MIF-2 N termini also diminished CD74 activation at the C terminus, though the receptor activation site is not fully defined in MIF-2. In this study, we use site-directed mutagenesis, NMR spectroscopy, molecular simulations, in vitro and in vivo biochemistry to explore the putative CD74 activation region of MIF-2 based on homology to MIF. We also confirm its reciprocal structural coupling to the MIF-2 allosteric site and N-terminal enzymatic site. Thus, we provide further insight into the CD74 activation site of MIF-2 and its allosteric coupling for immunoregulation. The macrophage migration inhibitory factor (MIF) protein family consists of MIF and D-dopachrome tautomerase (also known as MIF-2). These homologs share 34% sequence identity while maintaining nearly indistinguishable tertiary and quaternary structure, which is likely a major contributor to their overlapping functions, including the binding and activation of the cluster of differentiation 74 (CD74) receptor to mediate inflammation. Previously, we investigated a novel allosteric site, Tyr99, that modulated N-terminal catalytic activity in MIF through a "pathway" of dynamically coupled residues. In a comparative study, we revealed an analogous allosteric pathway in MIF-2 despite its unique primary sequence. Disruptions of the MIF and MIF-2 N termini also diminished CD74 activation at the C terminus, though the receptor activation site is not fully defined in MIF-2. In this study, we use site-directed mutagenesis, NMR spectroscopy, molecular simulations, in vitro and in vivo biochemistry to explore the putative CD74 activation region of MIF-2 based on homology to MIF. We also confirm its reciprocal structural coupling to the MIF-2 allosteric site and N-terminal enzymatic site. Thus, we provide further insight into the CD74 activation site of MIF-2 and its allosteric coupling for immunoregulation. The human cytokine-like enzymes macrophage migration inhibitory factor (MIF) and D-dopachrome tautomerase (also called MIF-2) are broadly expressed immunoregulators (1Merk M. Zierow S. Leng L. Das R. Du X. Schulte W. et al.The D-dopachrome tautomerase (DDT) gene product is a cytokine and functional homolog of macrophage migration inhibitory factor (MIF).Proc. Natl. Acad. Sci. U. S. A. 2011; 108: E577-E585Crossref PubMed Scopus (163) Google Scholar). Elevated levels of MIF family proteins are implicated in inflammatory diseases such as asthma, acute respiratory distress syndrome, and arthritis, while antibody neutralization of MIF proteins attenuates inflammatory symptoms in animal models (2Rajasekaran D. Zierow S. Syed M. Bucala R. Bhandari V. Lolis E.J. Targeting distinct tautomerase sites of D-DT and MIF with a single molecule for inhibition of neutrophil lung recruitment.FASEB J. 2014; 28: 4961-4971Crossref PubMed Scopus (58) Google Scholar). Current research surrounding the therapeutic targeting of the MIF superfamily is limited by the fact that wholesale inhibition is undesirable due to their protective roles in innate immunity and host microbial responses. New efforts to locate and characterize allosteric sites within this superfamily is a promising avenue to regulate their disease-state function. Such an undertaking relies on detailed structural and dynamic information and is bolstered by the fact that several intervention points for MIF and MIF-2 are shared, including the transcriptional activation of proinflammatory factors upon binding to cluster of differentiation 74 (CD74), the recruitment of signaling subunit CD44 to the MIF-CD74 complex, the activation of extracellular signal-regulated kinases, and the recruitment of leukocytes (3Leng L. Metz C.N. Fang Y. Xu J. Donnelly S. Baugh J. et al.MIF signal transduction initiated by binding to CD74.J. Exp. Med. 2003; 197: 1467-1476Crossref PubMed Scopus (833) Google Scholar, 4Shi X. Leng L. Wang T. Wang W. Du X. Li J. et al.CD44 is the signaling component of the macrophage migration inhibitory factor-CD74 receptor complex.Immunity. 2006; 25: 595-606Abstract Full Text Full Text PDF PubMed Scopus (480) Google Scholar, 5Bernhagen J. Krohn R. Lue H. Gregory J.L. Zernecke A. Koenen R.R. et al.MIF is a noncognate ligand of CXC chemokine receptors in inflammatory and atherogenic cell recruitment.Nat. Med. 2007; 13: 587-596Crossref PubMed Scopus (947) Google Scholar). These conserved functions are thought to be attributed to the shared tertiary and quaternary structure of MIF and MIF-2. An evolutionarily conserved proline at the N terminus acts as the catalytic base for an enol-keto tautomerase activity of unknown biological significance in both MIF and MIF-2. However, the literature overwhelmingly supports a role for the MIF enzymatic site in its "cytokine activity," namely the activation of CD74. Indeed, mutation of Pro1 abolishes enzymatic activity and diminishes CD74 activation in vivo (6Pantouris G. Syed M.A. Fan C. Rajasekaran D. Cho T.Y. Rosenberg Jr., E.M. et al.An analysis of MIF structural features that control functional activation of CD74.Chem. Biol. 2015; 22: 1197-1205Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar, 7Fingerle-Rowson G. Kaleswarapu D.R. Schlander C. Kabgani N. Brocks T. Reinart N. et al.A tautomerase-null macrophage migration-inhibitory factor (MIF) gene knock-in mouse model reveals that protein interactions and not enzymatic activity mediate MIF-dependent growth regulation.Mol. Cell. Biol. 2009; 29: 1922-1932Crossref PubMed Scopus (110) Google Scholar). The ability of MIF and MIF-2 to toggle their activities has been attributed to allostery; the regulation of chemical function from spatially distant sites within the proteins. These remote effects may occur through intramolecular clusters of residues (i.e., allosteric pathways) or through changes in shape mediated by alterations to intermolecular forces and may contribute separately or synergistically to changes in function. MIF-2 maintains classical hallmarks of allostery, including a multidomain structure with several ligand docking sites and a conformational equilibrium that is stabilized in discrete states by its binding partners. We previously demonstrated that a novel allosteric residue, Tyr99, modulated N-terminal catalytic activity in MIF through a pathway of dynamic crosstalk comprising intramolecular and intermolecular correlations (8Pantouris G. Khurana L. Ma A. Skeens E. Reiss K. Batista V.S. et al.Regulation of MIF enzymatic activity by an allosteric site at the central solvent channel.Cell Chem. Biol. 2020; 27: 740-750.e745Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar). Disruption of the MIF N terminus revealed that allosteric residues in this region also diminished CD74 activation, which occurs at the C terminus. Additionally, we showed that N-terminal mutations in the MIF-2 homolog caused widespread nuclear magnetic resonance (NMR) perturbations and line broadening at its C terminus with a concomitant decrease of CD74 activation in vivo (9Chen E. Reiss K. Shah D. Manjula R. Allen B. Murphy E.L. et al.A structurally preserved allosteric site in the MIF superfamily affects enzymatic activity and CD74 activation in D-dopachrome tautomerase.J. Biol. Chem. 2021; 297101061Abstract Full Text Full Text PDF Scopus (8) Google Scholar), though the CD74 activation site of MIF-2 has not been extensively mapped. At present, it is clear that the C terminus of MIF-2 "senses" the structural and dynamic effects of mutations at the N terminus, but whether allosteric crosstalk is reciprocal based on perturbations at the C terminus or whether the C terminus is a functional hub in MIF-2, is unknown and explored here. We introduced a series of mutations at the MIF-2 C terminus, located at the hypothesized site of CD74 activation and allosteric regulation. We used NMR spectroscopy, molecular dynamics (MD) simulations, and in vitro and in vivo functional assays to investigate whether the C terminus of MIF-2 indeed contributes to the allosteric pathway partially mapped in this system and explored previously in MIF. In our prior study of N-terminal mutations in MIF-2, we noted strong structural and dynamic coupling between the N terminus, solvent cavity at the trimer symmetry axis, Phe100 allosteric site (which is analogous to the Tyr99 allosteric site in MIF), and C terminus via NMR chemical shifts and relaxation parameters. We used this as a basis for our current work to (1) confirm allosteric reciprocity at the MIF-2 C terminus and (2) map the CD74 activation site in MIF-2, which has not been defined. We systematically mutated MIF-2 residues Arg98, Ile107, Gly108, Lys109, Ile110, and Thr112 that appear in the same spatial positions as those determined to regulate CD74 activation in MIF (Asn97, Val106, Gly107, Trp108, Asn109, and Ser111, Fig. 1) (6Pantouris G. Syed M.A. Fan C. Rajasekaran D. Cho T.Y. Rosenberg Jr., E.M. et al.An analysis of MIF structural features that control functional activation of CD74.Chem. Biol. 2015; 22: 1197-1205Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar, 10Pantouris G. Ho J. Shah D. Syed M.A. Leng L. Bhandari V. et al.Nanosecond dynamics regulate the MIF-induced activity of CD74.Angew. Chem. Int. Ed Engl. 2018; 57: 7116-7119Crossref PubMed Scopus (27) Google Scholar). Mutation of these C-terminal residues divided the six variants into two resultant categories; those with wildtype (wt)-like enzymatic activity and those with strongly attenuated activity, where the latter demonstrate allosteric reciprocity to the N terminus. The keto-enol tautomerase function of the MIF superfamily is well-studied and presents a straightforward functional handle to assess the impact of structural perturbation on MIF-2 in vitro. Variants K109A and I110A comprise the "wt-like" group, retaining 60% and 83% of wt-MIF-2 enzymatic activity, respectively. Variants R98A, I107A, G108A, and T112A fall into the latter group, having at most ∼30% of the activity of wt-MIF-2 (Fig. 2). Though the thermal stabilities of two variants, R98A and I107A, are below those of wt-MF-2 and the other variants, all denaturation profiles are of similar shape and the secondary structures of all MIF-2 variants are consistent (Fig. S1), suggesting no significant structural decomposition from the mutations. Due to the number of variants used in this study, we discuss in detail a single representative MIF-2 variant of each category going forward: I110A and I107A of the wt-like and non-wt-like groups, respectively. To investigate whether the C-terminal residues of MIF-2 sustain the allosteric relay from Phe100 and explain the variation in enzymatic activity, we used solution NMR to determine changes in local structure, dynamics, and inter-residue communication. 1H15N TROSY-HSQC NMR spectra of wt-MIF-2 compared to C-terminal variants (Fig. 2) revealed a negative correlation between enzymatic activity and the degree of structural and dynamic perturbation in MIF-2, in agreement with prior reports (8Pantouris G. Khurana L. Ma A. Skeens E. Reiss K. Batista V.S. et al.Regulation of MIF enzymatic activity by an allosteric site at the central solvent channel.Cell Chem. Biol. 2020; 27: 740-750.e745Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar, 9Chen E. Reiss K. Shah D. Manjula R. Allen B. Murphy E.L. et al.A structurally preserved allosteric site in the MIF superfamily affects enzymatic activity and CD74 activation in D-dopachrome tautomerase.J. Biol. Chem. 2021; 297101061Abstract Full Text Full Text PDF Scopus (8) Google Scholar). The I107A variant perturbed regions distal to the mutation site via chemical shift perturbations and the greatest extent of line broadening of any variant, indicative of heightened dynamics across the entirety of the protein and supporting the involvement of Ile107 in intramolecular communication. The observed spectral perturbations correlate very well with regions of allosteric functional significance from our prior studies of MIF and MIF-2 (8Pantouris G. Khurana L. Ma A. Skeens E. Reiss K. Batista V.S. et al.Regulation of MIF enzymatic activity by an allosteric site at the central solvent channel.Cell Chem. Biol. 2020; 27: 740-750.e745Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar, 9Chen E. Reiss K. Shah D. Manjula R. Allen B. Murphy E.L. et al.A structurally preserved allosteric site in the MIF superfamily affects enzymatic activity and CD74 activation in D-dopachrome tautomerase.J. Biol. Chem. 2021; 297101061Abstract Full Text Full Text PDF Scopus (8) Google Scholar, 11Parkins A. Skeens E. McCallum C.M. Lisi G.P. Pantouris G. The N-terminus of MIF regulates the dynamic profile of residues involved in CD74 activation.Biophys. J. 2021; 120: 3893-3900Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar), including monomer interfaces in X-ray crystal structures (β2: residues 38–42, β3: 46–49, β6,7: 105–112) as well as residues involved in hydrophobic interactions between the core β-sheets and α-helices (Leu3, Leu22, Ile59, Phe80, Phe81, Fig. 2). Other affected sites include N-terminal Phe2 and Leu3, which along with the global alterations caused by the I107A mutation, likely drive the 82% decrease in enzymatic activity from wt-MIF-2. Other variants with strongly attenuated catalytic activity—R98A, G108A, and T112A—exhibit similar responses to mutation, namely widespread chemical shift perturbations and line broadening. R98A MIF-2, the most structurally perturbative C-terminal variant of those tested (based on NMR chemical shifts), is most affected at the hydrophobic core (Leu3, Leu18, Leu22, Ala26, Val37, Val39, Val41, Ile61, Phe80, Phe81, Phe83, and Ile95). G108A and T112A MIF-2 variants each have fewer hydrophobic core perturbations, possibly due to the more solvent exposed positions of those residues at the C terminus. The total number of chemical shift perturbations and line broadened resonances follows a trend of R98A > I107A > T112A > G108A > K109A > I110A that directly correlates with the percent decrease in enzymatic activity (Figs. 2; S2 and S3). In contrast, the I110A mutation causes modest structural and dynamic effects on MIF-2. Chemical shift perturbations and line broadening occur only at sites surrounding the mutation, at flexible loop regions, and at the ends of β-strands (Fig. 2). Remarkably, the affected residues include other "allosteric pathway" connections, such as Ile107 and Thr112 that lie at the monomer–monomer interface, several of which are at least 7 Å away from Ile110, demonstrating that relatively small structural effects can be sensed by functionally coupled distal sites. It is worth noting that the side chain of Ile110 points outward into the solvent and the replacement of the bulky Ile side chain with an alanine methyl group should not affect the neighboring amino acids as strongly, explaining the smaller structural perturbations and near-wt enzymatic activity. Mutation of Lys109 perturbs the structure more strongly than Ile110 when examining NMR chemical shifts, though most still cluster around the mutation site (Fig. S2) and both variants induce similarly low levels of NMR line broadening. The stronger effect of K109A on the local chemical environment may then be due to the side chain orientation of Lys109, which points toward the monomer interface, in contrast to Ile110. Conformational fluctuations on multiple timescales can affect protein function and influence ligand binding events (12Tzeng S.R. Kalodimos C.G. Protein activity regulation by conformational entropy.Nature. 2012; 488: 236-240Crossref PubMed Scopus (392) Google Scholar, 13Farber P.J. Mittermaier A. Concerted dynamics link allosteric sites in the PBX homeodomain.J. Mol. Biol. 2011; 405: 819-830Crossref PubMed Scopus (24) Google Scholar, 14Moschen T. Grutsch S. Juen M.A. Wunderlich C.H. Kreutz C. Tollinger M. Measurement of ligand-target residence times by (1)H relaxation dispersion NMR spectroscopy.J. Med. Chem. 2016; 59: 10788-10793Crossref PubMed Scopus (20) Google Scholar). These include ps-ns bond vector motions that organize ligand binding sites and routes of chemical signaling or report on global tumbling, as well as higher energy barrier μs-ms processes that affect folding and catalytic mechanisms (15Lisi G.P. Loria J.P. Solution NMR spectroscopy for the study of enzyme allostery.Chem. Rev. 2016; 116: 6323-6369Crossref PubMed Scopus (89) Google Scholar). To determine how (or if) altered structural dynamics of MIF-2 variants contribute to the measured catalytic activities, we used T1, T2, and heteronuclear 1H-[15N] NOE spin relaxation experiments. T1 and T2 values for wt-MIF-2 and MIF-2 variants suggest similar rotational correlation times (τc) and structural compactness. The average T1 and T2 across all residues of wt-MIF-2 were 1154 ± 57 ms and 51 ± 5.4 ms, respectively, in line with values expected of a compact ∼35 kDa protein at 600 MHz and 30 °C and consistent with a τc ≈ 20 ns (16Su C. Jergic S. Ozawa K. Burns N.D. Dixon N.E. Otting G. Measurement of dissociation constants of high-molecular weight protein-protein complexes by transferred 15N-relaxation.J. Biomol. NMR. 2007; 38: 65-72Crossref PubMed Scopus (17) Google Scholar). Given that MIF-2 is a trimer of 37 kDa, its behavior likely differs from a globular monomeric protein, accounting for slight variations in the expected T1, T2, and τc for a monomer of similar molecular weight. The average T1 and T2 values across all MIF-2 variants were determined to be nearly identical to those of wt-MIF-2, at 1199 ± 64 ms and 51 ± 1.2 ms, respectively. Thus, despite local fluctuations in per-residue plots of T1/T2 (Fig. S4), especially for variants with dramatically attenuated catalytic function, overall molecular tumbling rates are unaffected by the mutations. Representative R1 and R2 relaxation rates are shown as difference plots (mutant – wt) to quantify how the variants diverge from wt-MIF-2 on a single amino acid basis (Fig. 3, Tables S1–S6 and Figs. S5–S7). Unsurprisingly, I110A does not strongly perturb the dynamics of wt-MIF-2. A modest, but global, decrease in R1 and R2 is observed, with the greatest effect in the latter third of the protein (residues 70–117). In this region, the average R1 has over a 2-fold difference compared to the average R1 of residues 1 to 60. The effect on R2 is modest, except for decreases in residues Thr69 and allosteric pathway residue Ile107 (Fig. 3). Changes in relaxation parameters are small in I107A MIF-2 as well, but they appear in a different pattern. The R1 values of most residues are moderately decreased, but unlike I110A, these effects generally occur consistently across the whole protein. This effect is particularly pronounced in residues 1 to 60, where I110A has minimal effect. Some residues with noticeable elevations in R1 are Glu4, Ser63, and Val66, suggesting only small enhancements of fast timescale fluctuations. Residues with elevated R1 parameters lie closest to Ile107 of the same monomer. An Ile107 residue from an adjacent monomer would be located on the other side of the β-sheet structure, far from the perturbed sites. Upon mutation to alanine, the Ala107 side chain is unlikely to directly interact with residues displaying elevated R1, especially since both R1 and R2 parameters hint at a more rigid structure around the site of mutation, relative to wt-MIF-2 (Fig. 3). Ala107 from the adjacent monomer may transmit dynamic perturbations to Pro1 through the allosteric network, which then propagates to Ser63 and Val66. Though numerous MIF-2 residues are not observed in relaxation studies of I107A due to extensive line broadening, many sites with quantifiable changes in relaxation parameters are in close proximity to Ser63 and Val66 and are directly next to Glu4, in the middle of a β-strand at the enzymatic site. R2 values of the I107A variant are mostly elevated, in contrast to the global depression caused by I110A (Fig. 3) and consistent with prior findings that enhanced protein motion contributes to diminished enzymatic activities of MIF and MIF-2. Local changes in relaxation rates are more clearly visualized in heat maps of MIF-2 variants, where I107A has a larger number and magnitude of quantified change across the protein, particularly at the N terminus (rightmost β-sheets and α-helix) when compared to I110A (Fig. 3). Other C-terminal variants that strongly attenuate MIF-2 enzymatic activity (R98A, G108A, T112A) have global effects on R1 and R2, similarly to I107A (Figs. S5 and S6). The lack of an obvious pattern in relaxation difference plots when comparing these variants to wt-MIF-2 suggests that each mutated residue has a unique contribution to MIF-2 dynamics. Generally, heightened flexibility of the C-terminal variants is a diagnostic of poor enzymatic function, and one observable trend is that R1 gradually decreases following the sequence R98A, I107A, G108A, K109A, I110A, and T112A, culminating in a near protein-wide depression of R1 for T112A. This inversely follows the trend of enzymatic activity, which increases in the same order (Fig. 2). When visualized on heat maps of ΔR1, the core β-sheet structure in MIF-2 is quite red (positive values) for R98A and trends toward increasingly blue shades (negative values) in sequential order of mutation to T112A (Fig. S8). These data suggest that the location of the amino acid at the C terminus and its unique effect on the MIF-2 allosteric network has site-specific influence over fast timescale dynamics that contribute to proper MIF-2 function (9Chen E. Reiss K. Shah D. Manjula R. Allen B. Murphy E.L. et al.A structurally preserved allosteric site in the MIF superfamily affects enzymatic activity and CD74 activation in D-dopachrome tautomerase.J. Biol. Chem. 2021; 297101061Abstract Full Text Full Text PDF Scopus (8) Google Scholar). The 1H-[15N] NOE (σNOE) was also used to evaluate fluctuations of the MIF-2 backbone and revealed a consistent pattern in the variants. Despite the fact that all MIF-2 variants maintain an average σNOE that is similar to wt-MIF-2 when considering the entire structure, per-residue σNOE values more tightly cluster around those of wt-MIF-2 in variants with wt-like activity, as seen in K109A and I110A. In contrast, I107A has a noticeably larger σNOE range, with many negative ΔNOE across the protein, suggesting heightened flexibility in this variant (Figs. 3, S9 and S10). Other variants with depressed catalytic activity and high flexibility via widespread negative ΔNOE values (R98A, I107A, G108A, and T112A) display heat maps with greater blue density (Fig. S8). These data echo the pattern observed in analysis of NMR chemical shift perturbations and line broadening, namely that variants with greater conformational plasticity on fast-to-intermediate timescales most strongly attenuate MIF-2 enzymatic activity. The per-residue variance of σNOE and overall flexibilities of the I107A and I110A case studies are echoed in order parameters (S2) determined from model-free analysis of multifield relaxation measurements (Fig. S11). Although line broadening caused by the I107A mutation precluded model-free analysis of nearly 30 resonances, plots of S2 and ΔS2 reveal fluctuations in backbone bond vectors (i.e., large negative ΔS2) in residues 20 to 30, which show highly depressed σNOE, substantial structural fluctuations in MD simulations (vide infra), and are implicated in ligand binding. Also affected are residues 60 to 80, which contain hotspots of mutation-induced allosteric crosstalk determined by MD simulations and are bracketed by large clusters of exchange broadening. In contrast, S2 values of I110A are almost identical to those of wt-MIF-2, consistent with relaxation parameters discussed earlier. We analyzed the change in mutation-induced flexibility by computing the change in root-mean-squared fluctuation (ΔRMSF) between wt-MIF-2 and the C-terminal variants (Fig. 4). Consistent with NMR data, we find that low-activity variants, most notably G108A and T112A, increase mobility of the C terminus significantly, which propagates to residues 28 to 31, 33 to 36, 50 to 54, and 65 to 68 via more moderate gains in flexibility that likely hinder catalytic activation. To further understand the mutation-induced structural changes to MIF-2, we computed the per-residue secondary structure and observed the six MIF-2 variants to be similar. MIF-2 variants, in general, display modest secondary structure fluctuations across the protein sequence, consistent with a stable symmetric trimer. However, there is a prominent helix-to-coil conversion of the C terminus due to the G108A and T112A mutations, commensurate with the increased RMSF and heightened ps-ms dynamics observed by NMR spin relaxation and line broadening (Fig. 4). We gained insight into MIF-2 allosteric communication via an electrostatic eigenvector centrality (EEC) metric by computing the correlation of electrostatic energy (calculated from the Kabsch–Sander formalism) between residues throughout an MD trajectory, allowing us to pinpoint those residues most important to the electrostatic network. By taking the difference between wt-MIF-2 and the variants, we identified residues whose importance to the network changes significantly upon mutation (Figs. 5 and S12). EEC has been shown to correlate with NMR chemical shifts and therefore is complementary to our experiments aimed at understanding the dynamics of regions that experience significant changes in their chemical environment due to a perturbation such as a point mutation. Again taking G108A MIF-2 as an example, this mutation induces a strong increase in the C-terminal ΔEEC, suggesting it becomes more strongly coupled to the MIF-2 allosteric network, commensurate with both the mutation-induced increase in RMSF as well as the calculated helix-to-coil conversion measured for this region of the protein. While the C-terminal effect is not as strong for other variants, substantial differences are observed near residues 60 to 70, a region that was shown to be flexible by NMR order parameters and 1H-[15N] NOEs and critical for allosteric crosstalk in an earlier study of MIF-2 (9Chen E. Reiss K. Shah D. Manjula R. Allen B. Murphy E.L. et al.A structurally preserved allosteric site in the MIF superfamily affects enzymatic activity and CD74 activation in D-dopachrome tautomerase.J. Biol. Chem. 2021; 297101061Abstract Full Text Full Text PDF Scopus (8) Google Scholar). Through molecular dynamics simulations, the C-terminal region of MIF has been previously demonstrated to be critical for CD74 receptor activity (17Meza-Romero R. Benedek G. Jordan K. Leng L. Pantouris G. Lolis E. et al.Modeling of both shared and distinct interactions between MIF and its homologue D-DT with their common receptor CD74.Cytokine. 2016; 88: 62-70Crossref PubMed Scopus (18) Google Scholar). Mutations in the MIF C terminus resulted in diminished neutrophil recruitment in murine lungs in vivo (6Pantouris G. Syed M.A. Fan C. Rajasekaran D. Cho T.Y. Rosenberg Jr., E.M. et al.An analysis of MIF structural features that control functional activation of CD74.Chem. Biol. 2015; 22: 1197-1205Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar, 10Pantouris G. Ho J. Shah D. Syed M.A. Leng L. Bhandari V. et al.Nanosecond dynamics regulate the MIF-induced activity of CD74.Angew. Chem. Int. Ed Engl. 2018; 57: 7116-7119Crossref PubMed Scopus (27) Google Scholar), which is the established assay for CD74 activation by MIF proteins (18Takahashi K. Koga K. Linge H.M. Zhang Y. Lin X. Metz C.N. et al.Macrophage CD74 contributes to MIF-induced pulmonary inflammation.Respir. Res. 2009; 10: 33Crossref PubMed Scopus (89) Google Scholar). However, the functional significance of comparable MIF-2 mutations on murine lung inflammation was unknown. To test this effect, saline control, wt-MIF-2, and C-terminal variants of MIF-2 (R98A, I107A, G108A, K109A, and T112A) were delivered intratracheally into murine lungs, and the percentage of neutrophils and total protein (a marker for pulmonary edema) were measured from the bronchoalveolar lavage (BAL) fluid (Fig. 6). Mutation of MIF-2 C-terminal residues affected neutrophil recruitment in a manner that suggests Arg98, Ile107, Gly108, Lys109, and Ile110 may not be involved in binding and stabilization of CD74 but are instead modulators of CD74 receptor activity (Fig. 6A). This speculation is driven primarily by the observation that these variants increase neutrophil recruitment, where amino acids comprising the direct binding site for CD74 would be more likely to strongly attenuate neutrophil recruitment if mutated. Interestingly, neutrophil recruitment is only diminished by a T112A mutation. To further rationalize the functional consequences of the MIF-2 variants and connect to allosteric signaling and C-terminal flexibility, we used the previously described EEC to assess the ability of the C-terminal residues to propagate electrostatic couplings throughout the protein, serving as a proxy for allosteric communication. The clearest example for a biophysical rationale appears when focusing on the representative G108A and T112A variants, which have opposite im