Development of a candidate stabilizing formulation for bulk storage of a double mutant heat labile toxin (dmLT) protein based adjuvant

Abstract

This work describes the formulation design and development of a novel protein based adjuvant, a double mutant of heat labile toxin (dmLT), based on knowledge of the protein's structural integrity and physicochemical degradation pathways. Various classes of pharmaceutical excipients were screened for their stabilizing effect on dmLT during exposure to thermal and agitation stresses as monitored by high throughput analytical assays for dmLT degradation. Sucrose, phosphate, sodium chloride, methionine and polysorbate-80 were identified as potential stabilizers that protected dmLT against either conformational destabilization, aggregation/particle formation or chemical degradation (e.g., Met oxidation and Lys glycation). Different combinations and concentrations of the selected stabilizers were then evaluated to further optimize dmLT stability while maintaining pharmaceutically acceptable ranges of solution pH and osmolality. The effect of multiple freeze-thaw (FT) cycles on the physical stability of candidate bulk formulations was also examined. Increasing the polysorbate-80 concentration to 0.1% in the lead candidate bulk formulation mitigated the loss of protein mass during FT. This formulation development study enabled the design of a new bulk formulation of the dmLT adjuvant and provides flexibility for future use in combination with a variety of different vaccine dosage forms with different antigens.

Keywords: Adjuvant; Aggregation; Excipient; Formulation; Stability; dmLT.

Fig. 1

OD₃₅₀ studies of aggregation propensity (delta temperature values) of dmLT containing solutions after thermal stress in presence of different excipients. Average delta temperature value at which the OD₃₅₀ value reaches 0.1 absorbance unit for dmLT (0.15 mg/mL) in control buffer plus excipient vs. dmLT in control buffer alone (10 mM sodium phosphate, 150 mM NaCl, pH 6.0 with no additional excipient; highlighted box). The dmLT samples are shown in order of highest to lowest OD₃₅₀ (indicating highest to lowest stability in terms of aggregation behavior). Error bars represent the standard deviation from triplicate experiments. The inset shows the representative OD₃₅₀ thermal melt experiment of dmLT formulated in control buffer alone and control buffer in the presence of glycerol, mannitol and sorbitol. Excipients in green, yellow and red showed a large increase, moderate increase and low/no increase in stability, respectively. ^*For 15% glycerol, OD₃₅₀ value did not reached 0.1 absorbance unit and hence delta temperature could not be calculated.

Fig. 2

MFI studies of subvisible particle formation in dmLT containing solutions after agitation stress as a function of excipient addition. Total increase in the sub-visible particle concentration (after 4 h of agitation minus time zero results from the same solution) is shown for each of the dmLT samples in order of decreasing particle concentration. The control (dmLT (0.15 mg/mL) in 10 mM sodium phosphate, 150 mM NaCl, pH 6.0 with no additional excipient) is indicated by black bar and is included for reference. Error bars represent the standard deviation from triplicate experiments. Excipients added to dmLT in control buffer in green, yellow and red showed a large increase, moderate increase and low/no increase in number of sub-visible particles, respectively.

Fig. 3

Effect of sodium chloride concentration on dmLT physical stability profile at 0.15 mg/mL in a base buffer containing 10 mM phosphate buffer, ±10% w/v sucrose, pH 6.0. (A) Thermal stress as monitored by OD₃₅₀ temperature values of dmLT as a function of salt concentration. Average temperature value at which the OD₃₅₀ value reaches 0.1 of different concentrations of salts is shown, and (B) agitation stress as measured by MFI in terms of total increase in sub-visible particle concentration (agitation for 4 h minus time zero for same solution) is shown for each of the dmLT samples. Error bars represent the standard deviation from triplicate experiments.

Fig. 4

Effect of phosphate buffer concentration and pH on the thermal stability of dmLT at pH 6.0 and pH 7.4 in two different candidate formulations (see text for formulation conditions). (A) Temperature to reach 0.1 absorbance unit as measured by OD₃₅₀ thermal melts assay with dmLT in different solutions, and (B) Tm values for dmLT in two different candidate formulations (pH 6.0 and pH 7.4) containing an additional 10 and 50 mM phosphate ion as measured by DSC. Error bars represent the standard deviation from triplicate experiments.

Fig. 5

Effect of PS-80 concentration (0.01, 0.05 and 0.1% w/v) on freeze-thaw (0, 1 and 5 freeze-thaw cycles) stability of dmLT compared to dmLT in the current formulation. (A) Absorbance at 280 nm showing protein loss with increasing freeze thaw cycles, (B) % of native dmLT (AB₅ complex) as a function of freeze-thaw cycles as measured by HIC, and (C) radar plot analysis of the number and size distribution of sub-visible particles formed upon freeze-thaw as measured by MFI. The dmLT protein concentration was 0.4 mg/mL in the four formulations namely F1: 50 mM sodium phosphate, 50 mM NaCl, 10% w/v sucrose, 5 mM methionine, 0.01% v/v PS-80 pH 7.4; F2: 50 mM sodium phosphate, 50 mM NaCl, 10% w/v sucrose, 5 mM methionine, 0.05% v/v PS-80 pH 7.4; F3: 50 mM sodium phosphate, 50 mM NaCl, 10% w/v sucrose, 5 mM methionine, 0.1% v/v PS-80 pH 7.4 and F4: 42.7 mM sodium phosphate, 10.7 mM potassium phosphate, 82 mM NaCl, 5% lactose, pH7.4 (current formulation buffer). Error bars indicate standard deviation of triplicate samples.

Supplementary Figure 1

Supplementary Figure 2

Supplementary Figure 3