Modeling Downstream Process

1. Modeling Downstream Process Lecture 2 Mahesh Bule

2. Downstream process

3. Downstream process

4. Downstream process

5. Purification techniques • Filtration/Centrifugation • Precipitation • Liquid-liquid two-phase separation • Chromatography • Size exclusion (gel filtration) • Ion-exchange • Hydrophobic interaction • Reverse- Phase • Affinity (protein A,G etc, dyes, metal chelates, lectins etc…)

6. Centrifugation Theory and Practice • Routine centrifuge rotors • Calculation of g-force • Differential centrifugation • Density gradient theory

7. axis of rotation Swinging-bucket g At rest Spinning g Fixed-angle Centrifuge rotors

8. axis of rotation rmax rav rmin a rmin rmin rav rav rmax rmax b c Sedimentation path length Geometry of rotors

9. k’-factor of rotors • The k’-factor is a measure of the time taken for a particle to sediment through a sucrose gradient • The most efficient rotors which operate at a high RCF and have a low sedimentation path length therefore have the lowest k’-factors • The centrifugation times (t) and k’-factors for two different rotors (1 and 2) are related by:

10. Calculation of RCF and Q RCF = Relative Centrifugal Force (g-force) Q = rpm; r = radius in cm

11. RCF in swinging-bucket and fixed-angle rotors at 40,000 rpm • Beckman SW41 swinging-bucket (13 ml) • gmin = 119,850g; gav = 196,770g; • gmax = 273,690g • Beckman 70.1Ti fixed-angle rotor (13 ml) • gmin = 72,450g; gav = 109,120g; • gmax = 146,680g

12. v = velocity of sedimentation d = diameter of particle l = density of liquid p = density of particle  = viscosity of liquid g = centrifugal force Velocity of sedimentation of a particle

13. Differential centrifugation • Density of liquid is uniform • Density of liquid << Density of particles • Viscosity of the liquid is low • Consequence: Rate of particle sedimentation depends mainly on its size and the applied g-force.

14. Size of major cell organelles • Nucleus 4-12 m • Plasma membrane sheets 3-20 m • Golgi tubules 1-2 m • Mitochondria 0.4-2.5 m • Lysosomes/peroxisomes 0.4-0.8 m • Microsomal vesicles 0.05-0.3m

15. Decant supernatant 1000g/10 min 3000g/10 min etc. Differential centrifugation of a tissue homogenate (I)

16. Differential centrifugation of a tissue homogenate (II) • Homogenate – 1000g for 10 min • Supernatant from 1 – 3000g for 10 min • Supernatant from 2 – 15,000g for 15 min • Supernatant from 3 – 100,000g for 45 min • Pellet 1 – nuclear • Pellet 2 – “heavy” mitochondrial • Pellet 3 – “light” mitochondrial • Pellet 4 – microsomal

17. Differential centrifugation (III)Expected content of pellets • 1000g pellet: nuclei, plasma membrane sheets • 3000g pellet: large mitochondria, Golgi tubules • 15,000g pellet: small mitochondria, lysosomes, peroxisomes • 100,000g pellet: microsomes

18. Differential centrifugation (IV) • Poor resolution and recovery because of: • Particle size heterogeneity • Particles starting out at rmin have furthest to travel but initially experience lowest RCF • Smaller particles close to rmax have only a short distance to travel and experience the highest RCF

19. Swinging-bucket rotor: Long sedimentation path length gmax >>> gmin Fixed-angle rotor: Shorter sedimentation path length gmax > gmin Differential centrifugation (V)

20. Differential centrifugation (VI) • Rate of sedimentation can be modulated by particle density • Nuclei have an unusually rapid sedimentation rate because of their size AND high density • Golgi tubules do not sediment at 3000g, in spite of their size: they have an unusually low sedimentation rate because of their very low density: (p - l) becomes rate limiting.

21. Density Barrier Discontinuous Continuous Density gradient centrifugation

22. Least dense Most dense How does a gradient separate different particles?

23. Predictions from equation (I) When p > l : v is +ve When p = l : v is 0

24. Predictions from equation (II) When p < l : v is -ve

25. Summary of previous slides • A particle will sediment through a solution if particle density > solution density • If particle density < solution density, particle will float through solution • When particle density = solution density the particle stop sedimenting or floating

26. 1 2 3 4 5 Buoyant density banding Equilibrium density bandingIsopycnic banding

27. 1 2 3 3 Formats for separation of particles accordingto their density When density of particle < density of liquid V is -ve

28. Density Barrier Discontinuous Continuous I II Resolution of density gradients

29. Problems with top loading

30. Separation of particles according to size p >> l : v is +ve for all particles throughout the gradient

31. Cell Disruption http://www.biologics-inc.com/sd-models.htm • Disruption: the cell envelope is physically broken, releasing all intracellular components into the surrounding medium • Methods: Mechanical and non mechanical • Mechanical • - Ultrasonication (sonicators) • bacteria, virus and spores • suspensions at lab-scale • Electronic generator→ultrasonic waves • →mechanical oscillation • by a titanium probe immersed • in a cell disruption.

32. Cell Disruption Dyno-Mill (liquid) http://www.cbmills.com/Products/horizontalmills.htm • Mechanical • Milling: continuous operation, • Algae, bacteria and fungi • Large scale, up to 2000kg/h • liquid and solid • Principle of operation: • A grinding chamber filled with about 80% beads. • A shaft with designed discs or impellers is within the chamber. • The shift rotates at high speeds, high shearing and impact forces from the beads break the cell wall.

33. http://www.unitednuclear.com/mills.htm Cell Disruption • Mechanical • Ball Mill: solid • Frozen cell paste, cells attached to or within a solid matrix. • Large scale

34. Theory of grinding Kick’s law Rittinger’s law

35. Product release from disrupted cells C Time Single pass Multi pass

36. Cell Disruption • Mechanical • Homogenization: suspension, large scale • To pump a slurry (up to 1500 bar) through a restricted orifice valve. • The cells disrupt as they are extruded through the valve to atmosphere pressure by • - high liquid shear in the orifice • - sudden pressure drop upon discharge • i.e. French press & Gaulin-Manton: lab scale • Rannie high-pressure • Homogenizer (large scale) High pressure orifice

37. Cell Disruption • Nonmechanical • - Chemicals: use chemicals to solubilise the components in the cell walls to release the product. • Chemical requirements: • - products are insensitive to the used chemicals. • - the chemicals must be easily separable. • Types of chemicals: • - surfactants (solubilising lipids): sodium sulfonate, sodium dodecylsulfate. • - Alkali: sodium hydroxide, harsh • - Organic solvents: penetrating the lipids and swelling the cells. e.g. toluene. • e.g. Bacteria were treated with acetone followed by sodium dodecyl sulfate extraction of cellular proteins.

38. Cell Disruption • Nonmechanical • - Enzymes: to lyse cell walls to release the product. • gentle, but high cost • i.e. lysozyme (carbohydrase) to lyse the cell walls of bacteria. • - Osmotic shock • Osmosis is the transport of water molecules from high- to a low-concentration region when these two phases are separated by a selective membrane. • Water is easier to pass the membrane than other components. • When cells are dumped into pure water, cells can swell and burst due to the osmotic flow of water into the cells.

39. Cell Disruption • Challenge: Damage to the product • Heat denaturation • Oxidation of the product • Unhindered release of all intracellular products

40. Separation of Soluble Products • Liquid-liquid extraction: • Difference of solubility in two immiscible liquid • Applicable: separate inhibitory fermentation products such as ethanol and acetone-butanol from fermentation broth. • antibiotics (i.e. solvent amylacetate) • Requirements of liquid extractants : • nontoxic, selective, inexpensive, immiscible with fermentation broth and • high distribution coefficient: KD=YL/XH • YL and XH are concentrations of the solute in light and heavy phases, respectively. • The light phase is the organic solvent and the heavy phase is the fermentation broth. e.x. Penicillin is extracted from a fermentation broth using isoamylacetate. KD could reach 50. Light, YL Heavy, XH

41. Light, i, j Heavy, I, j Separation of Soluble Products • Liquid-liquid extraction: • When fermentation broth contains more than one component, then the selectivity coefficient (β) is important. • βil= KD,,i/KD,j • KD,,I and KD,j are distribution coefficients of component i and j. • The higher the valueof βilis, the easier the separation of i from j. • pH effect • e.g. at low pH <4, Penicillin can be • separated from other impurities such as • acetic acid from the fermentation broth to • organic phase amylacetate.

42. Separation of Soluble Products Precipitation Reduce the product solubility in the fermentation broth by adding chemicals. Applicable: separate proteins or antibiotics from fermentation broth.

43. Separation of Soluble Products • Precipitation • Methods: • - salting-out by adding inorganic salts such as ammonium sulfate, or sodium sulfate to increase high ionic strength (factors: pH, temperature) • e.g. The solubility of hemoglobin is reduced with increased amount of ammonium sulfate. • - added salts interact more stronger with water so that the proteins precipitate. • - inexpensive • - Isoelectric (IE) precipitation: • Precipitate a protein at its isoelectric point. E.g. The IE of cytochrome cM (without histidine tag) is 5.6 (Cho, et.al., 2000,Eur. J. Biochem. 267, 1068±1074).

44. Separation of Soluble Products • Adsorption • Adsorb soluble product from fermentation broth onto solids. • Approaches: physical adsorption (activated carbon), ion exchange (carboxylic acid cation exchange resin for recovering streptomycin) • Adsorption capacity: mass of solute adsorbed per unit mass of adsorbent • Affected by properties of adsorbents: functional groups and their numbers, surface properties by properties of solution: solutes, pH, ionic strength and temperature • Difference of Affinity of product in the solid and liquid phase. • Applicable: soluble products from dilute fermentation

45. Precipitation Introduction Protein solubility Structure and size Charge Solvent Precipitate Formation Phenomena Initial Mixing Nucleation Growth governed by diffusion and Growth governed by fluid motion Precipitate Breakage Precipitate Aging Methods of Precipitation Design of Precipitation System Summary

46. Introduction - Precipitation widely used for the recovery of bulk proteins can be applied to fractionate proteins (separate different types) or as a volume reduction method For example: all the proteins in a stream might be precipitated and redissolved in a smaller volume or a fractional precipitation might be carried out to precipitate the protein interest and leave many of contaminating proteins in the mother liqour Precipitation is usually induced by addition of a salt or an organic solvent, or by changing the pH to alter the nature of the solution. the primary advantages: relatively inexpensive, can be carried out with simple equipment, can be done continuously and leads to a form of the protein that is often stable in long-term storage Keys Problem Are the solvents and salts used on a small scale the best choices at larger scale? How can we carry out the precipitation at a larger scale, for example, in a 5000 liter tank?

47. Protein Solubility The most important factors affecting the solubility of proteins are structure and size, protein charge, and the solvent. Explanations follow for each of these factors. Structure and Size In the native state, a protein in an aqueous environment assumes a structure that minimizes the contact of the hydrophobic amino acid residues with the water solvent molecules and maximizes the contact of the polar and charged residues with the water. The major forces acting to stabilize a protein in its native state are hydrogen bonding, van derWaals interactions, and solvophobic interactions (driven forces of folding protein). In aqueous solution, these forces tend to push the hydrophobic residues into the interior of the protein and the polar and charged residues on the protein’s surface.

48. For example, one study of 36 globular proteins - shown that 95% of the ionizable groups are solvent accessible. In other studies of 69 proteins, the average solvent-(water-) accessible atomic surface was found to be 57% nonpolar, 25% polar, and 19% charged. Thus, in spite of the forces operating to force hydrophobic residues to the proteins interior, the surface of proteins usually contains a significant fraction of non polar atoms. The forces acting on a protein lead to the achievement of a minimum Gibbs free energy. For a protein in its native configuration, the net Gibbs free energy is on the order of only 10 to 20 kcal/mol. This is a relatively small net free energy, which means that the native structure is only marginally stable and can be destabilized by relatively small environmental changes Water molecules bind to the surface of the protein molecule because of association of charged and polar groups and immobilization by nonpolar groups.

49. For example, a study of the hydration of human serum albumin found two layers of water around the protein. • These hydration layers are thought to promote solubility of the protein by maintaining a distance between the surfaces of protein molecules. This phenomenon is illustrated in Figure. Schematic diagram of the limit of approach of two protein molecules to each other because of the hydration layers on each molecular surface

50. The size of a protein becomes important with respect to solubility when the protein is excluded from part of the solvent- happen when nonionic polymers - are added to the solution result in steric exclusion of protein molecules from the volume of solution occupied by the polymer. Juckes developed a model for this phenomenon based on the protein molecule being in the form of a solid sphere and the polymer molecule in the form of a rod- gave the following equation for S, the solubility of the protein: rsand rr= the radius of the protein solute and polymer rod, respectively, = the partial specific volume of the polymer, cp = the polymer concentration, and β’ = a constant.