A glycoconjugate of Haemophilus influenzae Type b capsular polysaccharide with tetanus toxoid protein: hydrodynamic properties mainly influenced by the carbohydrate

Abstract

Three important physical properties which may affect the performance of glycoconjugate vaccines against serious disease are molar mass (molecular weight), heterogeneity (polydispersity), and conformational flexibility in solution. The dilute solution behaviour of native and activated capsular polyribosylribitol (PRP) polysaccharides extracted from Haemophilus influenzae type b (Hib), and the corresponding glycoconjugate made by conjugating this with the tetanus toxoid (TT) protein have been characterized and compared using a combination of sedimentation equilibrium and sedimentation velocity in the analytical ultracentrifuge with viscometry. The weight average molar mass of the activated material was considerably reduced (Mw ~ 0.24 × 10(6) g.mol(-1)) compared to the native (Mw ~ 1.2 × 10(6) g.mol(-1)). Conjugation with the TT protein yielded large polydisperse structures (of Mw ~ 7.4 × 10(6) g.mol(-1)), but which retained the high degree of flexibility of the native and activated polysaccharide, with frictional ratio, intrinsic viscosity, sedimentation conformation zoning behaviour and persistence length all commensurate with highly flexible coil behaviour and unlike the previously characterised tetanus toxoid protein (slightly extended and hydrodynamically compact structure with an aspect ratio of ~3). This non-protein like behaviour clearly indicates that it is the carbohydrate component which mainly influences the physical behaviour of the glycoconjugate in solution.

Figure 1

Sedimentation coefficient distributions, g*(s) vs s profiles, at different concentrations for (a) Hib PRP-native capsular polysaccharide (b) Hib PRP-ADG (c) Hib PRP-TT conjugate. The apparent sharpening of the peaks as the concentration increases is due to “hypersharpening” through the combined effects of polydispersity and non-ideality: the faster moving species in a distribution are slowed down by having to sediment through a solution of the slower ones. These effects diminish as the concentration is reduced.

Figure 2. Concentration dependence (reciprocal) sedimentation coefficient plot for Hib PRP-native, Hib PRP-ADH and Hib PRP-TT, to remove the effects on non-ideality.

Sedimentation coefficients measured in the phosphate chloride buffer (pH = 6.8, I = 0.10) had been normalized to standard conditions (the viscosity and density of water at a temperature of 20.0 °C).

Figure 3. SEDFIT-MSTAR output for analysis of Hib PRP-TT conjugate at a loading concentration of 0.3 mg.mL⁻¹ to find the (apparent) weight average molar mass M_w,app over the whole distribution.

(a) The operational point average molar mass M*(r) plotted as a function of radial position from the centre of rotation r. M_w,app, the (apparent) weight average molar mass for the whole distribution being measured = M* extrapolated to the radial position at the cell base. Retrieved M_w,app from this extrapolation = (7.3 ± 0.4) × 10⁶ g.mol⁻¹. (b) Plot of the “point” or “local” apparent average molar mass M_w,app(r) at radial positions r, as a function of local concentration c(r) in the ultracentrifuge cell. The “hinge point” corresponds to the radial position where the c(r) = the initial loading concentration. At this hinge point M_w,app(r) = (7.3 ± 0.5) × 10⁶ g.mol⁻¹. Although not as precise a way of estimating M_w,app from the sedimentation equilibrium records it does provide an internal check for consistency.

Figure 4. Molar mass distribution f(M) profile from sedimentation velocity for Hib PRP-TT conjugate.

Obtained by transforming the sedimentation coefficient distribution of Fig. 1c by the Extended Fujita method, using the weight average sedimentation coefficient with the weight average M_w,app (from Fig. 3) molar mass and two different plausible values of the conformation parameter b. The broad distribution is completely different for the sharp monomer-dimer distribution we observed earlier for TT by itself .

Figure 5. Plot of mass per unit length M_L versus persistence length L_p evaluation using the multi-HYDFIT procedure of Ortega and Garcia de la Torre.

(a) the Hib PRP-native polysaccharide. The plot yields L_p ∼ 7.0 × 10⁻⁷ (cm) and M_L ∼ 3.8 × 10⁹ (g.mol⁻¹.cm⁻¹) at the minimum target (error) function (indicated by the white cross). (b) Hib PRP-ADH, L_p ∼ 7.0 × 10⁻⁷ (cm) and M_L∼ 6.0 × 10⁹ (g. mol⁻¹. cm⁻¹); (c) Hib PRP-TT L_p ∼ 4.5 × 10⁻⁷ (cm) and M_L ∼ 10.7 × 10⁹ (g. mol⁻¹. cm⁻¹).

Figure 6. Conformation Zoning plot, Hib PRP-native, PRP-ADH and PRP-TT conjugate (with spacer) all have very flexible structures in the “Zone D” region close to Zone C Zone A: Rigid rod with no flexibility; Zone B: Rigid rod with some flexibility; Zone C: Semi-flexible coil; Zone D: Random coil; Zone E: Globular or heavily branched structures. See ref. .