research communications\(\def\hfill{\hskip 5em}\def\hfil{\hskip 3em}\def\eqno#1{\hfil {#1}}\)

Journal logoSTRUCTURAL BIOLOGY
COMMUNICATIONS
ISSN: 2053-230X

Engineering and crystal structure of a monomeric FLT3 ligand variant

CROSSMARK_Color_square_no_text.svg

aUnit for Structural Biology, Department of Biochemistry and Microbiology, Ghent University, Technologiepark-Zwijnaarde 71, 9052 Zwijnaarde, Belgium, bUnit for Structural Biology, VIB Center for Inflammation Research, Technologiepark-Zwijnaarde 71, 9052 Zwijnaarde, Belgium, and cLaboratory for General Biochemistry and Physical Pharmacy, Ghent University, Ottergemsesteenweg 460, 9000 Gent, Belgium
*Correspondence e-mail: savvas.savvides@ugent.be

Edited by M. J. van Raaij, Centro Nacional de Biotecnología – CSIC, Spain (Received 2 February 2021; accepted 27 March 2021; online 6 April 2021)

The overarching paradigm for the activation of class III and V receptor tyrosine kinases (RTKs) prescribes cytokine-mediated dimerization of the receptor ectodomains and homotypic receptor–receptor interactions. However, structural studies have shown that the hematopoietic receptor FLT3, a class III RTK, does not appear to engage in such receptor–receptor contacts, despite its efficient dimerization by dimeric FLT3 ligand (FL). As part of efforts to better understand the intricacies of FLT3 activation, we sought to engineer a monomeric FL. It was found that a Leu27Asp substitution at the dimer interface of the cytokine led to a stable monomeric cytokine (FLL27D) without abrogation of receptor binding. The crystal structure of FLL27D at 1.65 Å resolution revealed that the introduced point mutation led to shielding of the hydrophobic footprint of the dimerization interface in wild-type FL without affecting the conformation of the FLT3 binding site. Thus, FLL27D can serve as a monomeric FL variant to further interrogate the assembly mechanism of extracellular complexes of FLT3 in physiology and disease.

1. Introduction

Approximately 30% of newly diagnosed patients with acute myeloid leukemia (AML) harbor mutations in FMS-like tyrosine kinase receptor 3 (FLT3), which confer a poor disease prognosis (recently reviewed by Daver et al., 2019[Daver, N., Schlenk, R. F., Russell, N. H. & Levis, M. J. (2019). Leukemia, 33, 299-312.]). While the majority of such cases entail FLT3 with internal tandem duplications (ITDs) in the intracellular juxtamembrane region of the receptor (Nagel et al., 2017[Nagel, G., Weber, D., Fromm, E., Erhardt, S., Lübbert, M., Fiedler, W., Kindler, T., Krauter, J., Brossart, P., Kündgen, A., Salih, H. R., Westermann, J., Wulf, G., Hertenstein, B., Wattad, M., Götze, K., Kraemer, D., Heinicke, T., Girschikofsky, M., Derigs, H. G., Horst, H. A., Rudolph, C., Heuser, M., Göhring, G., Teleanu, V., Bullinger, L., Thol, F., Gaidzik, V. I., Paschka, P., Döhner, K., Ganser, A., Döhner, H., Schlenk, R. F. & German-Austrian AML Study Group (AMLSG) (2017). Ann. Hematol. 96, 1993-2003.]; Tallman et al., 2019[Tallman, M. S., Wang, E. S., Altman, J. K., Appelbaum, F. R., Bhatt, V. R., Bixby, D., Coutre, S. E., De Lima, M., Fathi, A. T., Fiorella, M., Foran, J. M., Hall, A. C., Jacoby, M., Lancet, J., LeBlanc, T. W., Mannis, G., Marcucci, G., Martin, M. G., Mims, A., O'Donnell, M. R., Olin, R., Peker, D., Perl, A., Pollyea, D. A., Pratz, K., Prebet, T., Ravandi, F., Shami, P. J., Stone, R. M., Strickland, S. A., Wieduwilt, M., Gregory, K. M., Hammond, L. & Ogba, N. (2019). J. Natl Compr. Cancer Netw. 17, 721-749.]; Daver et al., 2019[Daver, N., Schlenk, R. F., Russell, N. H. & Levis, M. J. (2019). Leukemia, 33, 299-312.]), somatic mutations in the extracellular and transmembrane domains of FLT3 have also been identified and at least one of them has been confirmed to be a driver mutation (Forbes et al., 2008[Forbes, S. A., Bhamra, G., Bamford, S., Dawson, E., Kok, C., Clements, J., Menzies, A., Teague, J. W., Futreal, P. A. & Stratton, M. R. (2008). Curr. Protoc. Hum. Genet. 57, 10.11.1-10.11.26.]; Fröhling et al., 2007[Fröhling, S., Scholl, C., Levine, R. L., Loriaux, M., Boggon, T. J., Bernard, O., Berger, R., Döhner, H., Döhner, H., Ebert, B. L., Teckie, S., Golub, T. R., Jiang, J., Schittenhelm, M. M., Lee, B. H., Griffin, J. D., Stone, R. M., Heinrich, M. C., Deininger, M. W., Druker, B. J. & Gilliland, D. G. (2007). Cancer Cell, 12, 501-513.]). FLT3 is a transmembrane receptor that is expressed on the surface of early hematopoietic progenitor cells and dendritic cells. The receptor is a member of the class III tyrosine kinase receptors (RTK-IIIs), which include CSF-1R, KIT, PDGFRα and PDGFRβ, which are all characterized by a conserved modular architecture featuring an extracellular domain (ECD) comprising five Ig-like domains, a single membrane-spanning helix (TM) followed by a juxtamembrane (JM) region, and finally an intracellular tyrosine kinase domain (TKD) (Fig. 1[link]a; Lemmon & Schlessinger, 2010[Lemmon, M. A. & Schlessinger, J. (2010). Cell, 141, 1117-1134.]; Verstraete & Savvides, 2012[Verstraete, K. & Savvides, S. N. (2012). Nat. Rev. Cancer, 12, 753-766.]).

[Figure 1]
Figure 1
(a) FLT3 belongs to the class III receptor tyrosine kinase family, the members of which are characterized by a conserved modular build and activation mechanism. For all RTK-IIIs, cytokine ligands simultaneously bind to the membrane-distal domains (yellow; D1, D2 and/or D3) of two cognate receptors. Although this interaction has been shown to facilitate homotypic interactions between membrane-proximal domains (blue; D4 and/or D5) of almost all RTK-IIIs, this has not yet been demonstrated for the FL–FLT3 complex. The generation of such a ternary complex, possibly involving interactions of the transmembrane domains (TM), invokes a transphosphorylation of the inhibitory juxtamembrane (JM) domain, eventually resulting in fully activated kinase activity. (b) The dimeric interface of FL is centered around Leu27. A cartoon representation of FL (PDB entry 1ete; Savvides et al., 2000[Savvides, S. N., Boone, T. & Karplus, P. A. (2000). Nature Struct. Biol. 7, 486-491.]) is shown with the constituting protomers colored green and sand yellow. Coloring according to the Eisenberg hydrophobicity scale (inset, surface representation; red is more hydrophobic) illustrates how Leu27 from each protomer (blue) is inserted into the hydrophobic interior of the other one.

Due to their highly similar build and the dimeric nature of their cognate cytokine ligands, RTK-IIIs are thought to be activated by similar mechanisms (Verstraete & Savvides, 2012[Verstraete, K. & Savvides, S. N. (2012). Nat. Rev. Cancer, 12, 753-766.]). The binding of a dimeric cytokine to an RTK-III induces receptor dimerization that results in transactivation of the auto-inhibited tyrosine kinase domains and the activation of downstream signaling pathways (Fig. 1[link]a). While the intracellular activation mechanism of RTK-III is conserved in all RTK-IIIs, it has been shown that ligand binding to the membrane-distal domains takes place by homotypic receptor–receptor contacts that are mediated by the membrane-proximal Ig-like domains D4 and/or D5. Although such ligand-induced extracellular homotypic receptor interactions have been shown to be present in most RTK-IIIs (Elegheert et al., 2011[Elegheert, J., Desfosses, A., Shkumatov, A. V., Wu, X., Bracke, N., Verstraete, K., Van Craenenbroeck, K., Brooks, B. R., Svergun, D. I., Vergauwen, B., Gutsche, I. & Savvides, S. N. (2011). Structure, 19, 1762-1772.]; Felix et al., 2013[Felix, J., Elegheert, J., Gutsche, I., Shkumatov, A. V., Wen, Y., Bracke, N., Pannecoucke, E., Vandenberghe, I., Devreese, B., Svergun, D. I., Pauwels, E., Vergauwen, B. & Savvides, S. N. (2013). Structure, 21, 428-439.], 2015[Felix, J., De Munck, S., Verstraete, K., Meuris, L., Callewaert, N., Elegheert, J. & Savvides, S. N. (2015). Structure, 23, 1621-1631.]; Yuzawa et al., 2007[Yuzawa, S., Opatowsky, Y., Zhang, Z., Mandiyan, V., Lax, I. & Schlessinger, J. (2007). Cell, 130, 323-334.]; Chen et al., 2015[Chen, P., Unger, V. & He, X. (2015). J. Mol. Biol. 427, 3921-3934.]), they are absent in FLT3, as supported by the `open horseshoe' structures of its complexes revealed via X-ray crystallography and negative-stain electron microscopy (Verstraete, Vandriessche et al., 2011[Verstraete, K., Vandriessche, G., Januar, M., Elegheert, J., Shkumatov, A. V., Desfosses, A., Van Craenenbroeck, K., Svergun, D. I., Gutsche, I., Vergauwen, B. & Savvides, S. N. (2011). Blood, 118, 60-68.]). Furthermore, the removal of two membrane-proximal domains of FLT3 resulted in only a moderate change in affinity for the ligand as determined by isothermal titration calorimetry, suggesting that these domains do not contribute significantly to the overall affinity of FLT3 for its cytokine (Verstraete, Vandriessche et al., 2011[Verstraete, K., Vandriessche, G., Januar, M., Elegheert, J., Shkumatov, A. V., Desfosses, A., Van Craenenbroeck, K., Svergun, D. I., Gutsche, I., Vergauwen, B. & Savvides, S. N. (2011). Blood, 118, 60-68.]).

As a consequence of the clear correlation between AML and mutations in FLT3, therapeutic targeting of FLT3 has been intensely pursued in industry and academia for over two decades (Badar et al., 2015[Badar, T., Kantarjian, H. M., Nogueras-Gonzalez, G. M., Borthakur, G., Garcia Manero, G., Andreeff, M., Konopleva, M., Kadia, T. M., Daver, N., Wierda, W. G., Luthra, R., Patel, K., Oran, B., Champlin, R., Ravandi, F. & Cortes, J. E. (2015). Am. J. Hematol. 90, 1065-1070.]; Leick & Levis, 2017[Leick, M. B. & Levis, M. J. (2017). Curr. Hematol. Malig. Rep. 12, 153-167.]). With a few notable exceptions, most efforts have focused on the development of tyrosine kinase inhibitors addressing the intra­cellular domains of FLT3. With the advent of driver somatic mutations in the extracellular regions of FLT3, we hypothesized that more in-depth insights into the basic principles underlying FLT3 receptor activation could possibly reveal novel approaches to tackle constitutively activated oncogenic receptor variants. Indeed, despite having crystallographic models of FL and its complex with the ectodomains of FLT3 (Verstraete, Vandriessche et al., 2011[Verstraete, K., Vandriessche, G., Januar, M., Elegheert, J., Shkumatov, A. V., Desfosses, A., Van Craenenbroeck, K., Svergun, D. I., Gutsche, I., Vergauwen, B. & Savvides, S. N. (2011). Blood, 118, 60-68.]), the absence of structural insights into the possible inactive conformation of FLT3 and the conformational changes required to transition to an activated receptor–cytokine complex render our understanding of the extracellular complex principles incomplete.

To expand our molecular toolkit towards broadening the structural and mechanistic insights into FL–FLT3 assembly, we sought to engineer a monomeric variant of FL. The rationale behind such an endeavor was manifold. Firstly, a monomeric variant could be of use for the dissection of cytokine-mediated activation principles, as has been shown for CSF-1 (Elegheert et al., 2012[Elegheert, J., Bracke, N., Pouliot, P., Gutsche, I., Shkumatov, A. V., Tarbouriech, N., Verstraete, K., Bekaert, A., Burmeister, W. P., Svergun, D. I., Lambrecht, B. N., Vergauwen, B. & Savvides, S. N. (2012). Nat. Struct. Mol. Biol. 19, 938-947.]). Indeed, it has been suggested that Ig domain 1 of FLT3 could be involved in an inhibitory cis interaction with the membrane-proximal domains of the extracellular region (Verstraete & Savvides, 2012[Verstraete, K. & Savvides, S. N. (2012). Nat. Rev. Cancer, 12, 753-766.]). Secondly, a monomeric ligand could be a starting point for the further engineering of tools to dissect receptor-activation principles in cellular assays, as previously illustrated for stem-cell factor variants (Ho et al., 2017[Ho, C. C. M., Chhabra, A., Starkl, P., Schnorr, P. J., Wilmes, S., Moraga, I., Kwon, H. S., Gaudenzio, N., Sibilano, R., Wehrman, T. S., Gakovic, M., Sockolosky, J. T., Tiffany, M. R., Ring, A. M., Piehler, J., Weissman, I. L., Galli, S. J., Shizuru, J. A. & Garcia, K. C. (2017). Cell, 168, 1041-1052.]; Tilayov et al., 2020[Tilayov, T., Hingaly, T., Greenshpan, Y., Cohen, S., Akabayov, B., Gazit, R. & Papo, N. (2020). Molecules, 25, 4850.]) or as an in vitro binding probe that allows the discrimination of properly folded, binding-prone receptor species. Finally, we hypothesized that a non-activating, albeit receptor-binding-competent, variant of FL could lead to the stabilization of mechanistically relevant conformational states of FLT3.

2. Materials and methods

2.1. Macromolecule production

2.1.1. Production of recombinant FLT3 ligands in Escherichia coli

Recombinant wild-type FL (FLWT) and its Leu27Asp mutant (FLL27D) were produced according to previously published methods (Table 1[link]; Verstraete et al., 2009[Verstraete, K., Koch, S., Ertugrul, S., Vandenberghe, I., Aerts, M., Vandriessche, G., Thiede, C. & Savvides, S. N. (2009). Protein J. 28, 57-65.]). Briefly, both proteins were expressed in an E. coli Rosetta-gami(DE3) bacterial strain (Novagen) as inclusion bodies. The harvested cell pellets were resuspended in lysis buffer (50 mM Tris pH 8.0, 100 mM NaCl, 1% Triton X-100, 1 mM EDTA) and lysed by sonication. Inclusion bodies were isolated, washed and solubilized in guanidine buffer (6 M guanidinium hydrochloride, 100 mM NaH2PO4, 10 mM Tris, 10 mM 2-mercaptoethanol pH 8.0) by gentle stirring at 40°C, followed by the strict application of previously published protocols (Verstraete et al., 2009[Verstraete, K., Koch, S., Ertugrul, S., Vandenberghe, I., Aerts, M., Vandriessche, G., Thiede, C. & Savvides, S. N. (2009). Protein J. 28, 57-65.]).

Table 1
Macromolecule-production information

Forward primer CGGCAGCCATATGACCCAGGACTGCTCCTTCC
Reverse primer CGGATCCTTAGGGCTGACACTGCAGCTCCAGGC
Expression vector pET-15b
Expression host E. coli Rosetta-gami
Complete amino-acid sequence of the produced FLL27D protein MGSSHHHHHHSSGLVPRGSHMTQDCSFQHSPISSDFAVKIRELSDYLDQDYPVTVASNLQDEELCGGLWRLVLAQRWMERLKTVAGSKMQGLLERVNTEIHFVTKCAFQPPPSCLRFVQTNISRLLQETSEQLVALKPWITRQNFSRCLELQCQP
Amino-acid sequence of the thrombin-digested FLL27D protein GSHMTQDCSFQHSPISSDFAVKIRELSDYLDQDYPVTVASNLQDEELCGGLWRLVLAQRWMERLKTVAGSKMQGLLERVNTEIHFVTKCAFQPPPSCLRFVQTNISRLLQETSEQLVALKPWITRQNFSRCLELQCQP
2.1.2. Expression of recombinant proteins in mammalian cells and purification

The cDNA sequence coding for human FLT3 domains 1–5 (FLT3D1–D5; residues Met1–Asp541) was obtained from Verstraete et al. (2009[Verstraete, K., Koch, S., Ertugrul, S., Vandenberghe, I., Aerts, M., Vandriessche, G., Thiede, C. & Savvides, S. N. (2009). Protein J. 28, 57-65.]) and Verstraete, Remmerie et al. (2011[Verstraete, K., Remmerie, B., Elegheert, J., Lintermans, B., Haegeman, G., Vanhoenacker, P., Van Craenenbroeck, K. & Savvides, S. N. (2011). Acta Cryst. F67, 325-331.]). Constructs for transient mammalian expression of secreted proteins carrying a C-terminal thrombin-cleavable AviTag followed by a hexahistidine sequence were cloned in the pHLsec vector (Aricescu et al., 2006[Aricescu, A. R., Lu, W. & Jones, E. Y. (2006). Acta Cryst. D62, 1243-1250.]). For the generation of stable cell lines, similar constructs were generated in the pcDNA4/TO vector (Thermo Fisher Scientific).

A monoclonal stable HEK293S MGAT1−/− TR+ cell line (Reeves et al., 2002[Reeves, P. J., Callewaert, N., Contreras, R. & Khorana, H. G. (2002). Proc. Natl Acad. Sci. USA, 99, 13419-13424.]) was generated and grown to 90% confluence in the presence of 50 µg ml−1 zeocin (Verstraete, Remmerie et al., 2011[Verstraete, K., Remmerie, B., Elegheert, J., Lintermans, B., Haegeman, G., Vanhoenacker, P., Van Craenenbroeck, K. & Savvides, S. N. (2011). Acta Cryst. F67, 325-331.]). To induce expression, the growth medium was replaced by serum-free medium supplemented with 2 µg ml−1 tetracycline and 3.6 mM valproic acid. After 4–5 days of transient or tetracycline-induced expression, the conditioned medium was harvested, cleared of cellular debris by centrifugation and filtered through a 22 µm cutoff bottle-top filter. Recombinant hexahistine-tagged proteins were captured from the conditioned medium by IMAC purification using a cOmplete His-tag purification column (Roche). After elution with 500 mM imidazole, the eluate was concentrated and further purified by size-exclusion chromatography using HiLoad 16/60 Superdex 75/200 columns (GE Healthcare) with HBS buffer (20 mM HEPES pH 7.4, 150 mM NaCl) as the running buffer. Protein purity was assessed by SDS–PAGE.

2.1.3. Size-exclusion chromatography coupled to multi-angle laser light scattering (SEC-MALLS)

Recombinant proteins and complexes thereof were concentrated to 1 mg ml−1 and injected onto a Superdex 200 Increase column (GE Healthcare) inline with an ultraviolet detector (Shimadzu), a multi-angle laser light scattering miniDAWN TREOS (Wyatt) and an Optilab T-rEX refractometer (Wyatt) at 25°C. HBS was used as the running buffer at a flow of 0.5 ml min−1. Data were analyzed using the ASTRA6 software (Wyatt). For the analysis of glycosylated protein species, conjugate analysis was performed using theoretical protein extinction coefficients and a dn/dc value of 0.16 for the glycan modifier (Bloch et al., 2018[Bloch, Y., Bouchareychas, L., Merceron, R., Składanowska, K., Van den Bossche, L., Detry, S., Govindarajan, S., Elewaut, D., Haerynck, F., Dullaers, M., Adamopoulos, I. E. & Savvides, S. N. (2018). Immunity, 48, 45-58.]).

2.2. Crystallization

Recombinant FLL27D was treated with 1 U µg−1 thrombin (New England Biolabs) overnight to remove the hexahistidine purification tag. Subsequently, thrombin and the cleaved peptide tag were separated from FLL27D by size-exclusion chromatography. Sitting-drop vapor-diffusion crystallization trials were set up using a Mosquito crystallization robot (SPT Labtech) in nanolitre-scale Swissci 96-well triple-drop plates (Table 2[link]). The protein plates were incubated at 293 K. Commercially available sitting-drop crystallization screens from Molecular Dimensions and Hampton Research were used to screen for conditions allowing crystal nucleation and growth. Seeding of crystallization conditions was performed using the Seed Bead Kit (Hampton Research) following the contemporary protocol.

Table 2
Crystallization

Method Microseeding in combination with vapor diffusion
Plate type Swissci 96-well 3-drop plates
Temperature (K) 293
Protein concentration (mg ml−1) 22
Buffer composition of protein solution 20 mM HEPES, 150 mM NaCl pH 7.4
Composition of reservoir solution 2.0 M ammonium sulfate, 0.1 M HEPES pH 5.0
Volume and ratio of drop 1:1
Volume of reservoir (µl) 45 
Cryoprotectant None

2.3. Data collection and processing

X-ray diffraction experiments were performed at 100 K on the PROXIMA-1 beamline at SOLEIL, Gif-sur-Yvette, France. Two wedges of diffraction data (1–90° and 120–180°) were indexed, integrated and scaled into a single data set using the XDS suite (Kabsch, 2010[Kabsch, W. (2010). Acta Cryst. D66, 125-132.]).

2.4. Structure solution and refinement

The initial phases were obtained by maximum-likelihood molecular replacement in Phaser as implemented in the CCP4 package (McCoy et al., 2007[McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C. & Read, R. J. (2007). J. Appl. Cryst. 40, 658-674.]; Winn et al., 2011[Winn, M. D., Ballard, C. C., Cowtan, K. D., Dodson, E. J., Emsley, P., Evans, P. R., Keegan, R. M., Krissinel, E. B., Leslie, A. G. W., McCoy, A., McNicholas, S. J., Murshudov, G. N., Pannu, N. S., Potterton, E. A., Powell, H. R., Read, R. J., Vagin, A. & Wilson, K. S. (2011). Acta Cryst. D67, 235-242.]) using a search model generated from the X-ray structure of FL (PDB entry 1ete; Savvides et al., 2000[Savvides, S. N., Boone, T. & Karplus, P. A. (2000). Nature Struct. Biol. 7, 486-491.]). Structure building and refinement were performed iteratively using Coot (Emsley et al., 2010[Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486-501.]), Phenix (Liebschner et al., 2019[Liebschner, D., Afonine, P. V., Baker, M. L., Bunkóczi, G., Chen, V. B., Croll, T. I., Hintze, B., Hung, L.-W., Jain, S., McCoy, A. J., Moriarty, N. W., Oeffner, R. D., Poon, B. K., Prisant, M. G., Read, R. J., Richardson, J. S., Richardson, D. C., Sammito, M. D., Sobolev, O. V., Stockwell, D. H., Terwilliger, T. C., Urzhumtsev, A. G., Videau, L. L., Williams, C. J. & Adams, P. D. (2019). Acta Cryst. D75, 861-877.]) and BUSTER (Bricogne et al., 2011[Bricogne, G., Blanc, E., Brandl, M., Flensbueg, C., Keller, P., Paciorek, W., Roversi, P., Sharff, A., Smart, O. S., Vonrhein, C. & Womack, T. O. (2011). BUSTER. Global Phasing Ltd, Cambridge, UK.]). Final refinement was performed using BUSTER 2.10.3.

3. Results and discussion

3.1. Engineering strategy to monomerize FL

Mature wild-type FL belongs to the structural family of short-chain four-helical bundle cytokines and consequently exhibits a noncovalently linked homodimeric structure, in which the two protomers interact head to head (Savvides et al., 2000[Savvides, S. N., Boone, T. & Karplus, P. A. (2000). Nature Struct. Biol. 7, 486-491.]). The availability of several crystallographic models of FL, both unbound (PDB entry 1ete) and in complex with its receptor (PDB entries 3qs7 and 3qs9), provided a structural basis for the development of a strategy to disrupt the dimeric interface of FL without introducing significant changes in the receptor-binding epitope (Savvides et al., 2000[Savvides, S. N., Boone, T. & Karplus, P. A. (2000). Nature Struct. Biol. 7, 486-491.]; Verstraete, Vandriessche et al., 2011[Verstraete, K., Vandriessche, G., Januar, M., Elegheert, J., Shkumatov, A. V., Desfosses, A., Van Craenenbroeck, K., Svergun, D. I., Gutsche, I., Vergauwen, B. & Savvides, S. N. (2011). Blood, 118, 60-68.]). Following the strategy used to monomerize CSF-1 (Elegheert et al., 2012[Elegheert, J., Bracke, N., Pouliot, P., Gutsche, I., Shkumatov, A. V., Tarbouriech, N., Verstraete, K., Bekaert, A., Burmeister, W. P., Svergun, D. I., Lambrecht, B. N., Vergauwen, B. & Savvides, S. N. (2012). Nat. Struct. Mol. Biol. 19, 938-947.]), several constructs were generated with a tandem duplication of the dimer epitope (residues 18–30), some of which had one or multiple point mutations at sites playing a key role at the dimeric interface. However, despite extensive optimization of the purification protocols, we did not succeed in purifying a monomeric species that was stable in solution. Therefore, we resorted to a more targeted approach by introducing a single point mutation targeting Leu27 at the heart of the dimeric interface (Fig. 1[link]b). In each protomer, Leu27 is located at the tip of a loop formed by residues Leu26–Gln29, protruding into the hydrophobic interior of the four-helical bundle of the accompanying protomer. By mutating Leu27 to an aspartate, we hypothesized that the entropic penalty for burying a charged residue in the hydrophobic interior of the second protomer would be detrimental for any dimerization event to occur. Interestingly, previous work by Graddis et al. (1998[Graddis, T. J., Brasel, K., Friend, D., Srinivasan, S., Wee, S., Lyman, S. D., March, C. J. & McGrew, J. T. (1998). J. Biol. Chem. 273, 17626-17633.]) identified a Leu27-to-proline mutation in FL, based on random mutagenesis, that was deficient in dimerization at low protein concentrations.

3.2. FLL27D is monomeric and engages in a 1:1 stoichiometric complex with FLT3

The expression of FLL27D in E. coli followed by in vitro refolding of FLL27D (Verstraete et al., 2009[Verstraete, K., Koch, S., Ertugrul, S., Vandenberghe, I., Aerts, M., Vandriessche, G., Thiede, C. & Savvides, S. N. (2009). Protein J. 28, 57-65.]) led to a stable and monodisperse protein that eluted in a size-exclusion chromatography (SEC) experiment as a protein with a substantially reduced hydrodynamic radius (Rhyd) compared with wild-type FL (FLWT; Fig. 2[link], green and gray curves). Multi-angle laser light scattering (SEC-MALLS) analysis of these proteins during elution from SEC led to molecular-weight determinations of 35 and 17 kDa for FLWT and FLL27D, respectively (Fig. 2[link]b). Importantly, no concentration-dependent dimerization could be detected even at concentrations as high as 95.83 µM (1.7 mg ml−1). We concluded that these experimentally determined values are in excellent agreement with their theoretical molecular weights and confirmed that FLL27D is indeed a monomeric species in solution.

[Figure 2]
Figure 2
FLL27D is a stable monomer capable of binding only one FLT3 molecule. (a) SEC-MALLS characterization of FLWT, FLL27D and receptor complexes thereof. Elution profile monitored by the forward and right-angle laser detector (left axis) plotted against the SEC retention volume and overlaid with the measured molecular weight (right axis). FLWT (green) is able to recruit two FLT3 molecules (yellow) into complex formation (blue), whereas FLL27D (gray) binds FLT3 in an equimolar fashion (red). (b) Summary of the predicted molecular weights, based on the amino-acid sequence, and the MALLS-measured molecular weights. Further glycoprotein conjugate analysis of the latter allowed part of the mass to be attributed to the glycan content.

To assess whether the monomer-inducing point mutation at position 27 affected the FLT3 binding site, which is localized close to the N-terminal region of FL (residues 6–13), we investigated its ability to form a complex with the extracellular region of recombinant human FLT3 comprising domains 1–5 (FLT3D1–D5; Verstraete, Remmerie et al., 2011[Verstraete, K., Remmerie, B., Elegheert, J., Lintermans, B., Haegeman, G., Vanhoenacker, P., Van Craenenbroeck, K. & Savvides, S. N. (2011). Acta Cryst. F67, 325-331.]). The titration of a 20% molar excess of FLL27D against FLT3D1–D5 and subsequent SEC-MALLS analysis resulted in a predominantly monodisperse species with an Rhyd exceeding that of both molecules alone (Fig. 2[link]a, red curve). With only an excess of FLL27D detected, this shift indicates that despite its monomeric nature, FLL27D was still able to recruit all available receptor molecules into complex formation. The molecular species has a molecular mass of 70 kDa as determined by SEC-MALLS, which is well below that of an FL-mediated receptor complex (152 kDa; Fig. 2[link]b) and therefore allowed us to infer that the apparent FLL27D–FLT3 complex consists of one molecule of FLL27D and one molecule of FLT3.

3.3. Structural differences between FLL27D and FLWT are limited to the dimerization-interface region

To further validate that mutation of Leu27 to aspartate does not compromise the overall fold of the molecule, we pursued structural characterization of FLL27D by X-ray crystallography. Initial crystallization trials resulted in the identification of multiple crystallization conditions across a wide pH range, all characterized by a high concentration (>1.8 M) of ammonium sulfate. Subsequent optimization of these initial hits led to crystals that diffracted synchrotron X-rays to high resolution, although all diffraction patterns showed signs of multiple crystal lattices (Fig. 3[link]). Nevertheless, we were able to index at least one crystal into a single crystal lattice in space group P1 and used the obtained data to determine the crystal structure to 1.65 Å resolution (Tables 3[link] and 4[link], Fig. 4[link]).

Table 3
Data collection and processing

Values in parentheses are for the outer shell.

Diffraction source PROXIMA-1, SOLEIL, France
Wavelength (Å) 0.97625
Temperature (K) 100
Detector PILATUS 6M
Crystal-to-detector distance (mm) 321.8
Rotation range per image (°) 0.1
Total rotation range (°) 180
Exposure time per image (s) 0.2
Space group P1
a, b, c (Å) 28.30, 43.49, 46.36
α, β, γ (°) 82.82, 85.41, 85.10
Mosaicity (°) 0.105
Resolution range (Å) 18.42–1.65 (1.709–1.650)
Total No. of reflections 70278 (4053)
No. of unique reflections 24967 (1712)
Completeness (%) 94.9 (89.2)
Multiplicity 2.81 (2.37)
I/σ(I)〉 10.6 (2.33)
Overall B factor from Wilson plot (Å2) 16.95
Rmeas (%) 7.7 (56.7)
CC1/2 (%) 99.6 (72.8)

Table 4
Structure refinement

Values in parentheses are for the outer shell.

Resolution range (Å) 18.42–1.65 (1.709–1.65)
No. of reflections, working set 24910 (2403)
No. of reflections, test set 1246 (120)
Final Rcryst 0.1643
Final Rfree 0.2026
No. of non-H atoms
 Total 2436
 Protein 2193
 Ligand 25
 Water 218
No. of protein residues 269
R.m.s.d., bond lengths (Å) 0.017
R.m.s.d., angles (°) 1.49
Ramachandran favored (%) 98.11
Ramachandran allowed (%) 1.89
Ramachandran outliers (%) 0.00
Rotamer outliers (%) 0.00
Clashscore 8.42
Average B factors (Å2)
 Overall 21.57
 Protein 20.48
 Ligands 50.19
 Solvent 29.2
No. of TLS groups 1
[Figure 3]
Figure 3
Representative crystal morphologies and corresponding X-ray diffraction from crystals of FLL27D. (a) Representative image of a crystallization drop containing crystals of FLL27D displaying macroscopic crystal-growth pathologies. (b) Test X-ray diffraction image from the crystal that resulted in the data set used for obtaining the structure of the monomeric FLL27D variant reported here. Resolution shells are displayed as circles. A close-up of the diffraction image (inset) reveals severe diffraction pathologies, including multiple lattices.
[Figure 4]
Figure 4
Structural differences between FLL27D and FLWT are limited to the dimerization-interface region. (a) Superimposition of FLL27D (gray) and FLWT (green). Crystallographic models of the ligands are shown in cartoon representation with indication of the twofold-symmetry axis (inset) or as ribbon diagrams (main panel); the side chain of Asp27 in FLL27D is shown as sticks and FLT3 is shown in surface representation. With the exception of the αB–βA loop, the main chain of both FLL27D molecules superimposes very well (average Cα r.m.s.d. of 0.85 Å) with the main chain of all four FLWT copies (PDB entry 1ete). (b) The asymmetric unit of FLL27D crystals features a top-to-top packing of molecules. This topology is distinct from the twofold-symmetry axis within one FLWT molecule and supports the L27D mutation preventing dimerization even in the context of crystal packing. (c) Detail of the superimposed αB–βA loop of FLL27D (gray) and FLWT (green). Loop residues are shown as sticks. Hydrogen bonds are indicated by dashed lines. (d) Detail of the superimposed αB–βA loops of FLL27D and FLWT, as viewed from the second FLWT protomer. FLWT is colored according the Eisenberg hydrophobicity scale (red is more hydrophobic); key residues of FLL27D are shown as sticks. Hydrogen bonds are indicated by dashed lines.

The obtained crystal structure of FLL27D superimposes very well with a single protomer of FLWT (Fig. 4[link]a). Indeed, not taking the αB–βA loop (residues 25–30) into account, the average root-mean-square deviation (r.m.s.d.) with FLWT (PDB entry 1ete; Savvides et al., 2000[Savvides, S. N., Boone, T. & Karplus, P. A. (2000). Nature Struct. Biol. 7, 486-491.]) is only 0.851 Å, indicating no large structural changes in the overall conformation of FLL27D. Given the observation that FLL27D still binds FLT3, it comes as no surprise that the absence of structural deviation from FLWT remains valid for residues 6–13, which are all key players in the largest interaction site of the FL–FLT3 epitope (Verstraete, Vandriessche et al., 2011[Verstraete, K., Vandriessche, G., Januar, M., Elegheert, J., Shkumatov, A. V., Desfosses, A., Van Craenenbroeck, K., Svergun, D. I., Gutsche, I., Vergauwen, B. & Savvides, S. N. (2011). Blood, 118, 60-68.]). Importantly, although the triclinic unit cell contains two copies of FLL27D (Fig. 4[link]b) with apparent twofold rotational symmetry, the observed apparent symmetry axis is dramatically distinct in direction and context from the twofold-symmetry axis in dimeric FLWT (Fig. 4[link]a, inset). Likewise, no combination of symmetry relations can reconstitute the head-to-head dimer resembling FLWT, despite the fact that the loop containing Asp27 is located near tightly packed crystal lattice contacts.

Given that the hydrophobic cavity that sheltered Leu27 of the accompanying FLWT protomer would remain solvent-exposed after the L27D monomerization event, we wondered how FLL27D would structurally compensate for this. When analyzing the conformational changes in the αB–βA loop (Fig. 4[link]c), we noticed that Asp27 is able to recruit Tyr30 via an intramolecular hydrogen bond, thus stabilizing the rotamer conformation of the latter such that it effectively shields the hydrophobic cavity that otherwise mediates dimeric FLWT (Fig. 4[link]d). Thus, we have shown that FL can display structural plasticity in this region, which may open additional possibilities to engineer this part of FL for structure–function purposes.

Supporting information


Acknowledgements

We thank the SOLEIL synchrotron, Saint-Aubin, France for beam-time allocation and the staff of the PROXIMA-1 beamline for excellent technical support.

Funding information

Funding for this research was provided by: Vlaams Agentschap Innovatie en Ondernemen (award to Erwin Pannecoucke); Herculesstichting; Kom op tegen Kanker; Geconcerteerde Onderzoeksacties; Ghent University; Vlaams Instituut voor Biotechnologie.

References

First citationAricescu, A. R., Lu, W. & Jones, E. Y. (2006). Acta Cryst. D62, 1243–1250.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationBadar, T., Kantarjian, H. M., Nogueras-Gonzalez, G. M., Borthakur, G., Garcia Manero, G., Andreeff, M., Konopleva, M., Kadia, T. M., Daver, N., Wierda, W. G., Luthra, R., Patel, K., Oran, B., Champlin, R., Ravandi, F. & Cortes, J. E. (2015). Am. J. Hematol. 90, 1065–1070.  CrossRef CAS PubMed Google Scholar
First citationBloch, Y., Bouchareychas, L., Merceron, R., Składanowska, K., Van den Bossche, L., Detry, S., Govindarajan, S., Elewaut, D., Haerynck, F., Dullaers, M., Adamopoulos, I. E. & Savvides, S. N. (2018). Immunity, 48, 45–58.  Web of Science CrossRef CAS PubMed Google Scholar
First citationBricogne, G., Blanc, E., Brandl, M., Flensbueg, C., Keller, P., Paciorek, W., Roversi, P., Sharff, A., Smart, O. S., Vonrhein, C. & Womack, T. O. (2011). BUSTER. Global Phasing Ltd, Cambridge, UK.  Google Scholar
First citationChen, P., Unger, V. & He, X. (2015). J. Mol. Biol. 427, 3921–3934.  CrossRef CAS PubMed Google Scholar
First citationDaver, N., Schlenk, R. F., Russell, N. H. & Levis, M. J. (2019). Leukemia, 33, 299–312.  CrossRef CAS PubMed Google Scholar
First citationElegheert, J., Bracke, N., Pouliot, P., Gutsche, I., Shkumatov, A. V., Tarbouriech, N., Verstraete, K., Bekaert, A., Burmeister, W. P., Svergun, D. I., Lambrecht, B. N., Vergauwen, B. & Savvides, S. N. (2012). Nat. Struct. Mol. Biol. 19, 938–947.  Web of Science CrossRef CAS PubMed Google Scholar
First citationElegheert, J., Desfosses, A., Shkumatov, A. V., Wu, X., Bracke, N., Verstraete, K., Van Craenenbroeck, K., Brooks, B. R., Svergun, D. I., Vergauwen, B., Gutsche, I. & Savvides, S. N. (2011). Structure, 19, 1762–1772.  Web of Science CrossRef CAS PubMed Google Scholar
First citationEmsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486–501.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationFelix, J., De Munck, S., Verstraete, K., Meuris, L., Callewaert, N., Elegheert, J. & Savvides, S. N. (2015). Structure, 23, 1621–1631.  Web of Science CrossRef CAS PubMed Google Scholar
First citationFelix, J., Elegheert, J., Gutsche, I., Shkumatov, A. V., Wen, Y., Bracke, N., Pannecoucke, E., Vandenberghe, I., Devreese, B., Svergun, D. I., Pauwels, E., Vergauwen, B. & Savvides, S. N. (2013). Structure, 21, 428–439.  CrossRef Google Scholar
First citationForbes, S. A., Bhamra, G., Bamford, S., Dawson, E., Kok, C., Clements, J., Menzies, A., Teague, J. W., Futreal, P. A. & Stratton, M. R. (2008). Curr. Protoc. Hum. Genet. 57, 10.11.1–10.11.26.  Google Scholar
First citationFröhling, S., Scholl, C., Levine, R. L., Loriaux, M., Boggon, T. J., Bernard, O., Berger, R., Döhner, H., Döhner, H., Ebert, B. L., Teckie, S., Golub, T. R., Jiang, J., Schittenhelm, M. M., Lee, B. H., Griffin, J. D., Stone, R. M., Heinrich, M. C., Deininger, M. W., Druker, B. J. & Gilliland, D. G. (2007). Cancer Cell, 12, 501–513.  PubMed Google Scholar
First citationGraddis, T. J., Brasel, K., Friend, D., Srinivasan, S., Wee, S., Lyman, S. D., March, C. J. & McGrew, J. T. (1998). J. Biol. Chem. 273, 17626–17633.  CrossRef CAS PubMed Google Scholar
First citationHo, C. C. M., Chhabra, A., Starkl, P., Schnorr, P. J., Wilmes, S., Moraga, I., Kwon, H. S., Gaudenzio, N., Sibilano, R., Wehrman, T. S., Gakovic, M., Sockolosky, J. T., Tiffany, M. R., Ring, A. M., Piehler, J., Weissman, I. L., Galli, S. J., Shizuru, J. A. & Garcia, K. C. (2017). Cell, 168, 1041–1052.  CrossRef CAS PubMed Google Scholar
First citationKabsch, W. (2010). Acta Cryst. D66, 125–132.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationLeick, M. B. & Levis, M. J. (2017). Curr. Hematol. Malig. Rep. 12, 153–167.  CrossRef PubMed Google Scholar
First citationLemmon, M. A. & Schlessinger, J. (2010). Cell, 141, 1117–1134.  Web of Science CrossRef CAS PubMed Google Scholar
First citationLiebschner, D., Afonine, P. V., Baker, M. L., Bunkóczi, G., Chen, V. B., Croll, T. I., Hintze, B., Hung, L.-W., Jain, S., McCoy, A. J., Moriarty, N. W., Oeffner, R. D., Poon, B. K., Prisant, M. G., Read, R. J., Richardson, J. S., Richardson, D. C., Sammito, M. D., Sobolev, O. V., Stockwell, D. H., Terwilliger, T. C., Urzhumtsev, A. G., Videau, L. L., Williams, C. J. & Adams, P. D. (2019). Acta Cryst. D75, 861–877.  Web of Science CrossRef IUCr Journals Google Scholar
First citationMcCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C. & Read, R. J. (2007). J. Appl. Cryst. 40, 658–674.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationNagel, G., Weber, D., Fromm, E., Erhardt, S., Lübbert, M., Fiedler, W., Kindler, T., Krauter, J., Brossart, P., Kündgen, A., Salih, H. R., Westermann, J., Wulf, G., Hertenstein, B., Wattad, M., Götze, K., Kraemer, D., Heinicke, T., Girschikofsky, M., Derigs, H. G., Horst, H. A., Rudolph, C., Heuser, M., Göhring, G., Teleanu, V., Bullinger, L., Thol, F., Gaidzik, V. I., Paschka, P., Döhner, K., Ganser, A., Döhner, H., Schlenk, R. F. & German–Austrian AML Study Group (AMLSG) (2017). Ann. Hematol. 96, 1993–2003.  Google Scholar
First citationReeves, P. J., Callewaert, N., Contreras, R. & Khorana, H. G. (2002). Proc. Natl Acad. Sci. USA, 99, 13419–13424.  Web of Science CrossRef PubMed CAS Google Scholar
First citationSavvides, S. N., Boone, T. & Karplus, P. A. (2000). Nature Struct. Biol. 7, 486–491.  Web of Science PubMed CAS Google Scholar
First citationTallman, M. S., Wang, E. S., Altman, J. K., Appelbaum, F. R., Bhatt, V. R., Bixby, D., Coutre, S. E., De Lima, M., Fathi, A. T., Fiorella, M., Foran, J. M., Hall, A. C., Jacoby, M., Lancet, J., LeBlanc, T. W., Mannis, G., Marcucci, G., Martin, M. G., Mims, A., O'Donnell, M. R., Olin, R., Peker, D., Perl, A., Pollyea, D. A., Pratz, K., Prebet, T., Ravandi, F., Shami, P. J., Stone, R. M., Strickland, S. A., Wieduwilt, M., Gregory, K. M., Hammond, L. & Ogba, N. (2019). J. Natl Compr. Cancer Netw. 17, 721–749.  CrossRef CAS Google Scholar
First citationTilayov, T., Hingaly, T., Greenshpan, Y., Cohen, S., Akabayov, B., Gazit, R. & Papo, N. (2020). Molecules, 25, 4850.  CrossRef Google Scholar
First citationVerstraete, K., Koch, S., Ertugrul, S., Vandenberghe, I., Aerts, M., Vandriessche, G., Thiede, C. & Savvides, S. N. (2009). Protein J. 28, 57–65.  Web of Science CrossRef PubMed CAS Google Scholar
First citationVerstraete, K., Remmerie, B., Elegheert, J., Lintermans, B., Haegeman, G., Vanhoenacker, P., Van Craenenbroeck, K. & Savvides, S. N. (2011). Acta Cryst. F67, 325–331.  CrossRef IUCr Journals Google Scholar
First citationVerstraete, K. & Savvides, S. N. (2012). Nat. Rev. Cancer, 12, 753–766.  Web of Science CrossRef CAS PubMed Google Scholar
First citationVerstraete, K., Vandriessche, G., Januar, M., Elegheert, J., Shkumatov, A. V., Desfosses, A., Van Craenenbroeck, K., Svergun, D. I., Gutsche, I., Vergauwen, B. & Savvides, S. N. (2011). Blood, 118, 60–68.  CrossRef CAS PubMed Google Scholar
First citationWinn, M. D., Ballard, C. C., Cowtan, K. D., Dodson, E. J., Emsley, P., Evans, P. R., Keegan, R. M., Krissinel, E. B., Leslie, A. G. W., McCoy, A., McNicholas, S. J., Murshudov, G. N., Pannu, N. S., Potterton, E. A., Powell, H. R., Read, R. J., Vagin, A. & Wilson, K. S. (2011). Acta Cryst. D67, 235–242.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationYuzawa, S., Opatowsky, Y., Zhang, Z., Mandiyan, V., Lax, I. & Schlessinger, J. (2007). Cell, 130, 323–334.  Web of Science CrossRef PubMed CAS Google Scholar

This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.

Journal logoSTRUCTURAL BIOLOGY
COMMUNICATIONS
ISSN: 2053-230X
Follow Acta Cryst. F
Sign up for e-alerts
Follow Acta Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds