Analytical ultracentrifugation with fluorescence detection system reveals differences in complex formation between recombinant human TNF and different biological TNF antagonists in various environments

Figures & data

Figure 1. UV-SV with 2 μM adalimumab (a), infliximab (b), and etanercept (c). The c(s) distributions in the absence of TNF (black) and of the 1:1 molar mixtures of the respective antagonist:TNF incubated at 20°C for 2 hours (blue) or at 37°C overnight (red) are shown.

Figure 2. FDS-SV with 25 nM adalimumab (a), infliximab (b), and etanercept (c) in the presence of varying concentrations of TNF in PBS. The c(s) distributions in the absence of TNF (black) and of the 10:1 (blue), 2:1 (green), and 1:1 (red) molar mixtures of the respective antagonist:TNF are shown.

Table 1. TNF-antagonist complexes detected using native MS and AUC FDS-SV.

	Native MS
	MW (Da)			AUC FDS-SV
TNF antagonist	measured	calculated	Stoichiometry	s20,w (S)	MWapp (kDa)	Stoichiometry
adalimumab	203,667	203,707	1:1	8.2	222	1:1
	352,252	351,854	2:1	11.9	383	2:1
	408,047	407,415	2:2	14.4	508	3:1
	556,656	555,561	3:2	17.4	652	3:2
infliximab	204,402	204,408	1:1	8.0	190	1:1
	353,474	353,297	2:1	12.6	377	2:1
	409,362	408,816	2:2	15.2	502	3:1
	559,262	557,664	3:2	18.2	654	3:2
etanercept	190,306	189,307	1:1	7.0	163	1:1

Table 2. The weight-average sedimentation coefficients calculated for each experimental condition.

		s20,w (S)
TNF antagonist	TNF concertation (nM)	PBS	human serum^1
adalimumab	0	6.8	7.6
	2.5	7.8	9.2
	12.5	15.7	23.8
	25	15.5	20.2
infliximab	0	7.1	9.4
	2.5	8.9	12.7
	12.5	18.4	36.0
	25	19.0	47.0
etanercept	0	5.2	4.6
	2.5	5.4	4.8
	12.5	6.4	5.8
	25	6.8	6.4

¹ The signal originating from the HSA-bilirubin complex as measured in SV experiments performed with human serum alone was subtracted from the total fluorescent signal obtained by integrating the area under the c(s) distribution of antagonist-TNF mixtures.

Figure 3. Mass spectra of the TNF-adalimumab complexes. Mass spectra of the mixtures containing trimeric TNF:adalimumab at molar ratios of 1:0 (black), 1:1 (orange), 1:1.25 (green), 1:1.5 (red), 1:2 (cyan), and 0:1.5 (purple) are shown.

Figure 4. ITC analysis of direct titration of 39.8 μM trimeric TNF into 9.8 μM adalimumab-Fab. Raw heat changes (top panel), normalized heat changes with best-fit values (solid line) (middle panel), and residuals of the fit (bottom panel) are shown. Non-linear regression analysis of the curves using A + B + B + B <-> AB + B + B <-> ABB + B <-> ABBB binding model yielded the K_d of 11.6 ± 1.7 nM and ΔH values of 11.6 ± 1.7 nM and -6.4 ± 0.1 kcal/mol, respectively, as determined from triplicate measurements. Representative profile from triplicate measurements is shown.

Figure 5. Simulation analysis of the populations of each AbAg complex in an Ab-Ag mixture. Simulations were performed for the mixtures containing 25 nM Ab and 2.5–100 nM Ag using K_d of 11.6 ± 1.7 nM derived from the ITC analysis of direct titration of TNF into an adalimumab-Fab solution.

Figure 6. FDS-SV with 25 nM adalimumab (a), infliximab (b), and etanercept (c) in the presence of varying concentrations of TNF in human serum. The c(s) distributions in the absence of TNF (black) and of the 10:1 (blue), 2:1 (green), and 1:1 (red) molar mixtures of the respective antagonist:TNF are shown.

Figure 7. FDS-SV analysis in serial dilutions of HSA in PBS in the absence (a) or presence of 5 mg/mL human immunoglobulin (b). The c(s) distributions of 25 nM adalimumab in 0 (black), 1 (purple), 2.5 (blue), 5 (magenta), 10 (green), and 18 mg/mL (red) HSA are shown. For the clarity of presentation, the derived c(s) distributions were normalized against the height of the main peak.

Figure 8. FcγR reporter cell assay. Activation of FcγRIIa (a) and FcγRIIIa (b) by adalimumab (green), infliximab (red), and etanercept (blue) in the presence of TNF was measured.