Structure-based design of prefusion-stabilized human metapneumovirus fusion proteins

Abstract

The human metapneumovirus (hMPV) fusion (F) protein is essential for viral entry and is a key target of neutralizing antibodies and vaccine development. The prefusion conformation is thought to be the optimal vaccine antigen, but previously described prefusion F proteins expressed poorly and were not well stabilized. Here, we use structures of hMPV F to guide the design of 42 variants containing stabilizing substitutions. Through combinatorial addition of disulfide bonds, cavity-filling substitutions, and improved electrostatic interactions, we describe a prefusion-stabilized F protein (DS-CavEs2) that expresses at 15 mg/L and has a melting temperature of 71.9 °C. Crystal structures of two prefusion-stabilized hMPV F variants reveal that antigenic surfaces are largely unperturbed. Importantly, immunization of mice with DS-CavEs2 elicits significantly higher neutralizing antibody titers against hMPV A1 and B1 viruses than postfusion F. The improved properties of DS-CavEs2 will advance the development of hMPV vaccines and the isolation of therapeutic antibodies.

Fig. 1. Beneficial substitutions for hMPV F stabilization.

a Schematic of the hMPV F ectodomain. Disulfide bonds and N-linked glycosylation sites are highlighted. The residue numbers indicating the beginning and the end of the F1 and F2 subunits are shown. b Side view of the trimeric hMPV F ectodomain in a prefusion conformation (PDBID: 5WB0). One protomer is shown as a ribbon diagram, and the other two are shown as a white molecular surface. Insets highlight the position of select stabilizing substitutions. Side chains in each inset are shown as dark red sticks with sulfur atoms in yellow, nitrogen atoms in blue and oxygen atoms in red. In both panels, regions undergoing conformational changes during the pre-to-postfusion transition are colored blue and the regions remaining static are colored green.

Fig. 2. Characterization of individual hMPV F variants.

a Relative expression of purified hMPV F variants, calculated from the area under the curve (AUC) of size-exclusion chromatography peaks (SEC). Variants are colored by design. b SEC of purified F variants, grouped by design (proline, polar, cavity filling, and disulfide). A vertical dotted line indicates the peak retention volume for the hMPV F base construct. Molecular weights of protein standards in kDa are indicated at the top. c Differential scanning fluorimetry (DSF) analysis of thermostability of disulfide variants. The vertical dotted line indicates the melting temperature for the base construct. d Reducing SDS-PAGE of hMPV F base construct and individual F variants. Each variant and base construct were analyzed twice independently via reducing SDS-PAGE, and the additional image is shown in the Source Data file. The molecular weight standards in kDa are indicated at the left. The positions of uncleaved F₀, furin-cleaved F₁ and F₂ are indicated at the right. T127C/N153C SEC and DSF were run independently and normalized to the other samples. Source data are provided as a Source Data file.

Fig. 3. Characterization of multiple-substitution hMPV F variants.

a–c SEC of purified multiple-substitution hMPV F variants. A vertical dotted line indicates the peak retention volume for the hMPV F base construct. d DSF analysis of thermostability of multiple-substitution F variants. The vertical dotted line indicates the melting temperature for the base construct. e SEC trace of DS-CavEs2 purified from a 1 L culture of FreeStyle 293-F cells. f Binding of heat-treated DS-CavEs2 or DS-CavEs2 stored for 2.5 months at 4 °C to MPE8 Fab measured by biolayer interferometry. A vertical dotted line indicates the end of the association step. Untreated DS-CavEs2 was included as a control. Source data are provided as a Source Data file for panels a–f.

Fig. 4. Crystal structure of engineered hMPV F variant DSx2 bound to antibody MPE8.

a Side view of the atomic model of hMPV F variant DSx2 bound to MPE8 Fab, shown as ribbons. The F protomer is colored green, the heavy chain of MPE8 is colored purple and the light chain of MPE8 is colored white. The constant region of MPE8 Fab is omitted for clarity. Side chains of two disulfide substitutions in DSx2 are highlighted as sticks. b Zoomed view of the binding interface of the MPE8 light chain CDRs and F antigenic sites II and V. c Zoomed view of the binding interface of the MPE8 heavy chain CDRs and F. Main chain atoms of H-CDR3 pack against antigenic site III, which is highlighted as a transparent surface. Key residues that form polar interactions are shown as sticks.

Fig. 5. Crystal structure of hMPV F DS-CavEs2 exhibits a prefusion conformation.

a Side view of the atomic model of apo hMPV F variant (DS-CavEs2) in a prefusion conformation. The model (red ribbon) is superimposed with a previously determined hMPV F structure (white ribbon, PDB ID: 5WB0). The side chains of the introduced substitutions are highlighted as sticks. Each inset corresponds to the regions where the superimposition is performed. b Representative negative stain EM 2D class averages of DS-CavEs2 complexed with MPE8 Fab.

Fig. 6. Prefusion-stabilized hMPV F variants elicit neutralizing antibodies in mice.

a 8-week-old male BALB/c mice (n = 6/group) were immunized at weeks 0 and 3 with 1 μg of prefusion-stabilized hMPV F variants or heat-treated postfusion hMPV F, all adjuvanted with CpG. b, c 10 days after the week 3 injection, mice were bled for analysis of serum neutralization titers against hMPV b A1 and c B1 viruses. Each point represents an individual mouse (blue circle, postfusion; orange square, prefusion base construct, BaseCon; green triangle, DSx2; purple inverted triangle, DS-CavEs2). All box plots show mean as a plus sign, median as a central line, 25 and 75% as lower and upper box limits, and minimum to maximum values as whiskers. Each experimental group was compared with the postfusion-immunized group using one-way ANOVA. * = p-value < 0.05, ** = p-value < 0.01. For b, postfusion vs. BaseCon, p = 0.0029; postfusion vs. DSx2, p = 0.0055; postfusion vs. DS-CavEs2, p = 0.0062. For c, postfusion vs. BaseCon, p = 0.0051; postfusion vs. DSx2, p = 0.0454; Postfusion vs. DS-CavEs2, p = 0.0169. d, e The sera from immunized mice were depleted with 1 μg of postfusion or DS-CavEs2 proteins prior to the analysis of serum neutralization titers against hMPV d A1 and e B1 viruses. Serum depletion groups were compared with the non-depletion group using one-way ANOVA. ns = no significant, * = p-value < 0.05, *** = p-value < 0.005. For d and postfusion immunized sera, non-depletion vs. postfusion, p = 0.0004; non-depletion vs. DS-CavEs2, p = 0.0179. For d and DS-CavEs2 immunized sera, non-depletion vs. postfusion, p = 0.2461; non-depletion vs. DS-CavEs2, p = 0.0004. For e and postfusion immunized sera, non-depletion vs. postfusion, p = 0.0006; non-depletion vs. DS-CavEs2, p = 0.3489. For e and DS-CavEs2 immunized sera, non-depletion vs. postfusion, p = 0.2742; non-depletion vs. DS-CavEs2, p = 0.0005. Source data are provided as a Source Data file for panels b–e.