Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Measuring Comparability of Conformation, Heterogeneity and Aggregation with Circular Dichroism and Analytical Ultracentrifugation

531 views

Published on

"Measuring Comparability of Conformation, Heterogeneity, and Aggregation with Circular Dichroism and Analytical Ultracentrifugation", invited talk, State of the Art Methods for the Characterization of Biological Products and Assessment of Comparability, NIH, June 2003

Published in: Health & Medicine
  • Login to see the comments

Measuring Comparability of Conformation, Heterogeneity and Aggregation with Circular Dichroism and Analytical Ultracentrifugation

  1. 1. Measuring Comparability of Conformation, Heterogeneity, and Aggregation with Circular Dichroism and Analytical Ultracentrifugation John Philo jphilo@ap-lab.com © copyright 2003, Alliance Protein Laboratories Inc.
  2. 2. Outline Circular dichroism---what does it do and how can we apply it for comparability? Introduction to analytical ultracentrifugation: sedimentation velocity and sedimentation equilibrium Sedimentation equilibrium examples: does my protein have the correct quaternary structure? Characterizing protein conformation and aggregation by sedimentation velocity
  3. 3. Circular dichroism (CD) spectroscopy measures the absorbance difference between right-handed and left- handed circularly polarized light sensitive to chirality or asymmetry around chromophores
  4. 4. Far-UV CD ‘far-UV’ protein spectra (190-240 nm) are sensitive to secondary structure by spectral fitting one can estimate percentage α-helix, β-sheet, etc. our view: most useful as comparative spectral fingerprint 190 210 230 250 -40 -20 0 20 40 60 80 ellipticityperresiduex10 -3 (deg·cm/decimole) alpha helix beta sheet random coil wavelength (nm)
  5. 5. Differences in far-UV CD for natural vs. recombinant forms of an enzyme (which also have different enzymatic activity) 210 220 230 240 250 260 -12000 -10000 -8000 -6000 -4000 -2000 0 natural recombinant MolarEllipticity W avelength (nm)
  6. 6. Near-UV CD ‘near-UV’ protein spectra (240-340 nm) are sensitive to local tertiary structure around aromatic residues and disulfide bonds proteins lacking regular tertiary structure show zero near-UV signal (e.g. “molten globules”) signals can be either positive or negative sometimes they nearly cancel strictly a fingerprint---no direct structural interpretation our view: under-utilized usually more sensitive to subtle conformational differences than far-UV
  7. 7. Differences in near-UV circular dichroism between stable and unstable lots of a monoclonal antibody 240 250 260 270 280 290 300 310 320 330 340 -200 -150 -100 -50 0 50MolarEllipticity Wavelength (nm) lot 1 lot 2
  8. 8. Drawbacks to CD Spectral Analysis 1. difficult to quantitate similarity or differences what is “comparable” is often a judgment call 2. like all spectroscopy, if there is heterogeneity all you see is an average unlikely to detect minor components 3. certain buffer components absorb strongly in the far-UV and can cause interference
  9. 9. Measuring thermal stability (thermal unfolding) by CD Another conformational measure than can be used for comparability Very similar uses as DSC can be done at much lower concentrations can tell how the structure is changing, not just that something is unfolding Generally done by monitoring a single wavelength in the far-UV vs. temperature
  10. 10. Quantitative analysis of thermal stability from CD data 30 40 50 60 70 -80 -75 -70 -65 -60 -55 -50 -45 -40 melting end (95%) 60.1 ± 0.5 °C melting onset (5%) 36.6 ± 0.4 °C Tm = 47.94 ± 0.14 °C, ∆H = 51.4 ± 3.2 kcal/mol data fit (2-state model) molarellipticity@220nm Temperature (°C)
  11. 11. Analytical Ultracentrifugation
  12. 12. The modern analytical ultracentrifuge a preparative ultracentrifuge 364371 DALSAINC. CCDImageSensors CL-E1-2048S-134L S/NXXXXXX MadeinCanada + optical systems, special rotors, and sample cells + computerized control, data acquisition, and analysis
  13. 13. 6.0 6.2 6.4 6.6 6.8 7.0 7.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4Absorbance radius (cm) meniscus plateau boundary region of solute depletion 6.0 6.2 6.4 6.6 6.8 7.0 7.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4Absorbance radius (cm) The sedimentation coefficient is determined from the rate of boundary motion. It depends on both molecular weight and shape (conformation). 6.0 6.2 6.4 6.6 6.8 7.0 7.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4Absorbance radius (cm) friction diffusion buoyancysedimentation For each molecular species the boundary position is determined by the balance of sedimentation, buoyancy, and hydrodynamic frictional forces, and the boundary width is determined by its diffusion coefficient. The fundamentals of sedimentation velocity
  14. 14. 6.30 6.35 6.40 6.45 6.50 6.55 6.60 0.0 0.2 0.4 0.6 0.8 1.0 cell base Absorbance radius (cm) diffusion buoyancysedimentation meniscus The fundamentals of sedimentation equilibrium The concentration distribution depends only on molecular weight, independent of shape! ← smaller sample size to reduce time to reach equilibrium →
  15. 15. Both sedimentation methods are “first principle” methods based on fundamental physical laws require no standard molecules for calibration calibration is based only on fundamental units of distance, time, and temperature
  16. 16. Characterizing solution mass by sedimentation equilibrium
  17. 17. Size-exclusion chromatography of a TNF homolog 6 8 10 12 14 16 18 20 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0absorbance(arb.units) elution volume (ml) ovalbumin TNF homolog, elutes exactly as expected for a 17 kDa monomer
  18. 18. Linearized plot of equilibrium data for the TNF homolog -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 -1.9 -1.8 -1.7 -1.6 -1.5 -1.4 -1.3 -1.2 -1.1 -1.0 -0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 Ln(Absorbanceat230nm) (r2 - ro 2 ) / 2 theoretical slope for monomer theoretical slope for trimer
  19. 19. An example of characterizing a protein that self- associates to form dimers. Global analysis of experiments at ~5-250 µg/ml (using multiple wavelengths) gives Kd = 520 +/- 20 nM (∆G = -8570 +/- 25 cal/mol) -1.0-0.9-0.8-0.7-0.6-0.5-0.4-0.3-0.2-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 (R^2 - Ro^2) / 2 (cm^2) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 Absorbance
  20. 20. Using sedimentation velocity to characterize protein conformation and aggregation
  21. 21. How can we interpret these raw data for an antibody? Time-derivative (dc/dt) analysis allows us to convert it into a distribution like a chromatogram 6.0 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 7.0 7.1 Radius (cm) -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 Absorbance the “dc/dt method” uses a group of closely-spaced scans which are subtracted in pairs to approximate the time derivative of the data and thereby derive how much material is sedimenting at various rates
  22. 22. The sedimentation coefficient distribution function for the antibody sample shows it is heterogeneous and contains at least 2 different long-lived aggregates 2 4 6 8 10 12 14 0.0 0.1 0.2 0.3 0.4 g(s*),AU/Svedberg s*, Svedbergs
  23. 23. Comparing lots of antibody from two different purification processes: the conformation of the main peak is identical, but amounts of aggregate differ 2 4 6 8 10 12 0.0 0.1 0.2 0.3 lot B is 94.0% main peak (96.7% by SEC) g(s*)(AU/Svedberg) sedimentation coefficient (Svedbergs) 2 4 6 8 10 12 0.0 0.1 0.2 0.3 0.4 0.5 data single species fit lot A is 98.0% main peak (99.0% by SEC)
  24. 24. How reproducible are sedimentation coefficients for demonstrating comparability? The general rule of thumb is these values should be accurate and reproducible to ±0.5% or better, both run-to-run and even year-to-year precision within the same run should be ~0.1-0.2% Some real data for the same reference standard lot measured 5 different times over 18 months: 6.264 ± 0.005 S 6.268 ± 0.005 S 6.270 ± 0.005 S 6.270 ± 0.005 S 6.265 ± 0.006 S mean 6.267 ± 0.0028 (± 0.04%)
  25. 25. Velocity analysis of two different formulations of an antibody, each analyzed in its own formulation buffer, reveals differences in aggregation 2 4 6 8 10 12 0.0 0.1 0.2 0.3 0.4 0.5 a low ionic strength formulation produces significantly less dimer shift of main peak is due to differences in buffer viscosity g(s*)(AU/Svedberg) sedimentation coefficient (Svedbergs)
  26. 26. High resolution analysis of a highly stressed antibody sample using the c(s) method from Peter Schuck (NIH) 0 2 4 6 8 10 12 14 16 18 20 22 24 0.0 0.2 0.4 0.6 0.8 1.0 heptamer,0.1% hexamer,0.4% pentamer1.4% tetramer5.3% trimer14.6% dimer30.6% main peak (monomer), 45.5% ?HLhalfmolecule,0.8% ?freelightchain,1.4% c(s),normalized(totalarea=1) sedimentation coefficient (Svedbergs)
  27. 27. 0 1 2 3 0.0 0.5 1.0 1.5 0 8 16 24 32 40 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 no salt c(s) +50 mM NaCl c(s) +150 mM NaCl c(s) sedimentation coefficient (Svedbergs) 0 2 4 6 8 10 0.00 0.05 0.10 0.15 20X expanded 0 2 4 6 8 10 0.00 0.05 0.10 0.15 20X expanded Non-covalent aggregates of a ~20 kDa cytokine: monomer to ~100-mer can be measured in a single analysis
  28. 28. Comparability of a monoclonal antibody by the high-resolution method 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 0 1 2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 0.00 0.01 0.02 normalizedc(s) sedimentation coefficient (Svedbergs) 0.42% 0.10% 0.03% 0.30% lot 1 lot 2 X 100 0.95% 0.05% 0.07%
  29. 29. “Aggregates” vs. self-associated oligomers We must be careful to distinguish between rapidly-reversible, self-associated oligomers and irreversible (or at least long- lived) oligomers (“aggregates”) distinction more important now that we are seeing so many high concentration formulations Oligomers which are reversible, but which dissociate only over hours-days, are much more common than most people realize Our View: only the long-lived (hours or more) aggregates are likely to have an impact on immunogenicity, PK, or efficacy oligomers which rapidly dissociate upon dilution in vivo are unlikely to impact safety or efficacy
  30. 30. These same velocity methods can be applied to other biopharmaceuticals besides pure recombinant proteins 1. protein mixtures such as blood products 2. nucleic acids naked vaccines gene therapy ribozymes 3. whole virus vaccines gene therapy vectors 4. drug-polypeptide conjugates toxin:antibody conjugates small-molecule:poly-amino acid conjugates
  31. 31. Summary CD can provide useful spectral fingerprints and thermal unfolding data for measuring conformational comparability Sedimentation coefficients are an excellent and highly quantitative way to demonstrate comparability of conformation Sedimentation velocity is an excellent method to detect and quantify protein aggregates, with resolution and range far exceeding that of SEC Sedimentation equilibrium is a powerful tool for characterizing quaternary structure in solution and protein-protein interactions

×