Insights into the chromatography of proteins provided by mathematical modeling

Current Opinion in Biotechnology

Volumn 7, Issue 2, 1996, Pages 210-218

  Jungbauer, Alois ^a  

^a UNIVERSITY OF AGRICULTURAL SCIENCES   (Austria)

Author keywords

[No Author keywords available]

Indexed keywords

ADSORPTION CHROMATOGRAPHY; ALGORITHM; COLUMN CHROMATOGRAPHY; MATHEMATICAL MODEL; PRIORITY JOURNAL; PROTEIN ANALYSIS; PROTEIN PROTEIN INTERACTION; REVIEW; THERMODYNAMICS;

EID: 0029864343     PISSN: 09581669     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0958-1669(96)80015-2     Document Type: Article

References (51)

1. Smit JC, Smit HC, De Jager EM: Computer implementation of simulalion models for non-linear, non-ideal chromatography. Part 1: fundamental mathematical considerations. Anal Chim Acta 1980, 122:1-26.
2. Morton DW, Young CL: Analysis of peak profiles using statistical moments. J Chromatogr Sci 1995, 33:514-524.
3. Jungbauer A, Kaltenbrunner O: Fundamental questions in optimizing protein-chromatography using computer-aided process design. Biotechnol Bioeng 1996, in press.
4. Li Q, Grandmaison EW, Hsu CC, Talyor D, Goosen MFA: Interparticle and intraparticle mass transfer in chromatographic separation. Bioseparation 1995, 5:189-202. Intraparticle mass transfer is reviewed with respect to intraparticle convection in chromatography beads with gigapores. In addition, a comprehensive discussion is provided of the plate and rate theory is given.
5. Mao QM, Prince IG, Hearn MTW: High-performance chromatography of amino acids, peptides and proteins: CXXXIX. Impact of operating parameters in large-scale chromatography of proteins. J Chromatogr A 1995, 691:273-283. The sectional model, a discontinuous flow model, is employed to study the processing rate, which is dependent on the flow rate, purity and column utilization. In this model, the chromatography column is assumed to be divided into a series of sections. Each section is then treated as a well mixed tank. The difference between the series of sections in the tank model and the sectional model is the flow. In the tank model, the flow is continuous, whereas in the sectional model, the flow is discontinuous. After a certain time, the liquid phase is transferred into the next section and is mixed again for a certain time period. Compared with the continuous and discontinuous flow models, the advantage of this model is that the time required for adsorption is taken into consideration. It is not obligatory for equilibrium to be attained in each section.
6. Jungbauer A, Hackl S, Yamamoto S: Calculation of peak profiles in preparative chromatography of biomolecules. J Chromatogr A 1994, 658:399-406.
7. Woodbury CP Jr: A stochastic model of chromatography. J Chromatogr Sci 1994, 32:339-348. Reports the stochastic model of chromatography as an extension of the early statistical theory of chromatography by Giddings and Eyring (Giddings JC, Eyring H: A molecular dynamic theory of chromatography. J Phys Chem 1955, 59:416-421). Each solute is considered to have a residence time that follows a random distribution. The model is developed for the linear situation.
8. Woodbury CP Jr: A stochastic model of chromatography. J Chromatogr Sci 1994, 32:339-348. Reports the stochastic model of chromatography as an extension of the early statistical theory of chromatography by Giddings and Eyring (Giddings JC, Eyring H: A molecular dynamic theory of chromatography. J Phys Chem 1955, 59:416-421). Each solute is considered to have a residence time that follows a random distribution. The model is developed for the linear situation.
9. Golshan-Shirazi S, Guiochon G: Modeling of preparative liquid chromatography. J Chromatogr A 1994, 658:149-171. Provides a discussion of the ideal model and the equilibrium dispersive model of preparative chromatography for the single and multi-component situation. Guidelines are given for mathematical treatment of the different models and a general description of each model is provided. The paper does not consider proteins, however.
10. Fellinger A, Guiochon G: Rapid simulation of chromatographic band profiles on personal computers. J Chromatogr A 1994, 658:511-515. These authors compare the computational time needed by different approaches for the simulation of the separation of a binary mixture.
11. Talbot J, Jin X, Wang NHL: New equations for multicomponent adsorption kinetics. Langmuir 1994, 10:1663-1666.
12. Jin X, Talbot J, Wang NHL: Materials of steric hindrance: effects on adsorption kinetics and equlibria. Am Inst Chem Eng J 1994, 40:1685-1696. The RSA theory on a continuous surface is extended to proteins, leading to a RRA algorithm on a continuous surface. The model takes into consideration the accessible sites and predicts breakthrough curves at very high surface coverage. The asymmetry of a breakthrough curve can be described, which is a big advantage compared with other models such as the Thomas solution.
13. Subrammanian A, Van Cott KE, Milbrath DS, Velander WH: Role of local antibody density effects on immunosorbents efficiency. J Chromatogr A 1994, 672:11-24.
14. Dowd V, Yon RJ: Further examination of a concerted cluster model of multivalent affinity - heterogeneous adsorption of lactate-dehydrogenase to cibachrom-blue immobilized on cellulose. J Chromatogr A 1995, 693:15-21.
15. Renard J, Vidalmadjar C, Sebille B, Lapresle C: Chromatographic kinetic measurements of human serum-albumin adsorption on monoclonal-antibodies. J Mol Recognit 1995, 8:85-89.
16. Suen S, Etzel MR: Sorption kinetics and breakthrough curves for pepsin and chymosin using pepstatin A affinity membranes. J Chromatogr A 1994, 686:179-192.
17. Li Y, Pinto NG: Influence of lateral interactions on preparative protein chromatography. J Chromatogr A 1994, 658:445-457.
18. Li Y, Pinto NG: Model for ion-exchange equilibria of macromolecules in preparative chromatography. J Chromatogr A 1995, 702:113-123. This paper demonstrates the significance of lateral interaction in ion-exchange chromatography. Unfortunately, the authors use a weak mathematical model for the simulation of peak profiles that neglects axial dispersion and extra column effect. Thus, the influence of lateral adsorption cannot be estimated out by superimposing the experimental and simulated peak profile. Despite its flaws, the description of adsorption isotherms by this model is of a very high predictive value.
19. Tongata A, Liapis AI, Siehr DJ: Equilibrium and kinetic parameter of adsorption of α-chymotrypsinogen A onto hydrophobic porous adsorbent particles. J Chromatogr A 1994, 686:21-29.
20. Xie J, Aguilar M, Hearn MTW: High-performance liquid chromatography of amino acids, peptides and proteins: adsorption of horse heart cytochrome C onto a tentacle-type cation-exchanger. J Chromatogr A 1995, 691:263-271.
21. Gallant SR, Kundu A, Cramer SM: Modeling non-linear elution of proteins in ion-exchange chromatography. J Chromatogr A 1995, 702:125-143. The SMA model is used for the prediction of peak profiles. The model does not allow an analytical solution of the adsorption isotherm; therefore, an implicit solution is employed. A shielding factor, which is a unique property of a protein interacting with a particular ion-exchanger at certain conditions, compensates for the influence of salt shielding.
22. Patrickios CS, Patrickios ES: Stoichiometric mass action model, explicit isotherm for mono-, di-, tri-, and tetrameric ions. J Chromatogr A 1995, 694:480-485. The explicit expression of the isotherm is shown to be possible for exact numbers of interaction sites. The same algorithm is not applicable when the number of interaction sites is very high, which is the case for proteins.
23. Brooks CA, Cramer SM: Investigation of displacer equilibrium properties and mobile phase operating conditions in ion-exchange displacement chromatography. J Chromatogr A 1995, 693:187-196.
24. Gallant SR, Kundu A, Cramer SM: Optimization of step gradient separations: considerations of nonlinear adsoption. Biotechnol Bioeng 1995, 47:355-372. The SMA model is used to optimize resolution in preparative ion-exchange chromatography. A finite difference grid is employed for numerical simulation of the peak profiles. This model takes into consideration the different axial dispersion properties for each protein, which may be of importance when proteins with different molecular masses are employed. The model also integrates the solubility of a protein, which is an important constraint in protein chromatography.
25. Le Borgne S, Graber M, Condoret JS: Experimental and theoretical analysis of the chromatographic behaviour of protein purification fusions carrying charged tails. Bioseparation 1995, 5:53-64. Polyglutamic acid tails are fused to β-galactosidase by recombinant DNA technology. Both the wild-type protein and the recombinant protein are purified by ion-exchange chromatography. Using the SDM formalism, the binding constants are calculated and found to be higher for the modified proteins.
26. Frey DD, Barnes A. Strong J: Numerical studies of multicomponent chromatography using pH gradients. Am Inst Chem Eng J 1995, 41:1171-1183. A model, which takes several buffer species into consideration, is developed for elution using pH gradients. The interaction of proteins with the ion-exchanger is thought to take place in a stoichiometric manner. The model provides good predictions of changes in pH. The non-linear isotherm situation is not incorporated.
27. Hashim MA, Chu KH, Tsan PS: Equilibrium isotherm for protein adsorption in a bisolute system. Trans Inst Chem Engineers 1995, 73:22-26.
28. Yoshida H, Nishihara H, Kataoka T: Adsorption of BSA on strongly basic chitosan - equilibria. Biotechnol Bioeng 1994, 43:1087-1093.
29. Kaltenbrunner O, Jungbauer A: Adsorption isotherms in protein chromatography, combined influence of protein and salt concentration on adsorption isotherm. J Chromatogr A 1996, in press.
30. Boschetti E: Advanced sorbents for preparative protein separation purposes. J Chromatogr A 1994, 658:207-223.
31. Bülow M, Micke A: Determination of transport coefficients in microporous solids. Adsorption 1995, 1:29-39. A review describing macroscopic transient methods with respect to their applicability to the investigation of molecular transport in microporous adsorption systems. Although the focus is on the adsorption of low molecular weight substances, several processes are identical to the adsorption of proteins.
32. Frey DO, Schweinheim E, Horváth C: Effect of intraparticle convection on the chromatography of biomacromolecules. Biotechnol Prog 1993, 9:273-284.
33. Liapis AI, McCoy MA: Perfusion chromatography - effect of micropore diffusion on column performance in systems utilizing perfusive adsorbent particles with a bidisperse porous structure. J Chromatog A 1994, 660:85-96. This paper provides a perfusion chromatography model that considers a chromatography medium with a bidisperse pore structure. The model is based on the assumption that convection through the pore (perfusion) is unidirectional and diffusion is multi-directional, The contours of isoconcentration of adsorbed solute may provide a good measure for perfusion. As yet, the theoretical model has not been supported by experimental data, perhaps because the experimental determination of isoconcentration contours is not a simple task.
34. Rodrigues AE, Loureiro JM, Chenou C, Rendueles de la Vega M: Bioseparation with permeable particles. J Chromatogr A 1995, 664:233-240. These authors apply the theory of constant normalized breakthrough curves to media such as gel in a shell media (Q-HyperD; BioSepra Inc. Massachusetts, USA). They observe a weak dependence of the HETP equation on flow rate, which is a characteristic property of perfusion chromatography. In addition, they suggest perfusion also occurs in Q-HyperD media, although it is physically impossible because the bead does not possess through-pores. Under their experimental conditions, Q-HyperD excludes the model protein and behaves like a non-porous medium. This may be the explanation for the small change in the HETP equation with flow rate.
35. Carta G: Linear driving force approximation for intraparticle diffusion and convection in permeable supports. Chem Eng Sci 1995, 50:887-889.
36. Weaver LE, Carta G: Protein adsorption on carbon exchangers: comparison of macroporous and gel composite media. Biotechnol Prog 1996, in press. Provides an explanation of the adsorption mechanisms of a gel in a shell medium.
37. Heeter GA, Liapis AI: Perfusion chromatography: performance of periodic counter-current column operation and its comparison with fixed-bed operation. J Chromatogr A 1995, 711:3-21.
38. Luo RG, Hsu JT: Intraparticle protein diffusion effect on gradient station chromatography. Separ Technol 1993, 3:221-229.
39. Lightfoot EN, Athalye AM, Coffman JL, Roper DK, Root TW: Nuclear magnetic resonance and the design of chromatographic separations. J Chromatogr A 1995, 707:45-55. These authors employ nuclear magnetic resonance to measure axial dispersion in a chromatography column.
40. Ma Z, Whitley RD, Wang NHL: Comparison of pore and surface diffusion in multicomponent adsorption and liquid chromatography systems. Am Inst Chem Eng J 1996, in press, The influence of pore and surface diffusion on peak broadening and peak tailing is investigated.
41. McCoy B: Chromatographic separation of continuous mixtures in distributed pores. J Chromatogr A 1995, 697:533-540.
42. Martys NS: Numerical simulation of hydrodynamic dispersion in a random manner. J Chromatogr A 1995, 707:35-43.
43. Yoshida H, Yoshikawa M, Kataoka T: Parallel transport of BSA by surface and pore diffusion in strongly basic chitosan. Am Inst Chem Eng J 1994, 40:3034-2044.
44. Edwards DA, Davis AM: Diffusion and convective dispersion through arrays of spheres with surface adsorption, diffusion, and unequal solute partitioning. Chem Eng Sci 1993, 50:1441-1454.
45. Mao QM, Heam MTG: Optimization of affinity and ion-exchange chromatographic process for the purification of proteins. Biotechol Bioeng 1906, in press. Presents a comparison of the sectional model and the non-porous particle adsorption model. The mathematical equations are derived very precisely.
46. Wiblin DJ, Simon DR Myhill GR: Computer aided desk-top scale-up and optimization of chromatographic processes. J Chromatogr A 1995, 702:81-87. Conventional methodology is used to set up the equations for the mass balance in order to calculate the peak profiles. A moving grid model is used for numerical simulation of the chromatographic process. The grid moves down the column in tandem with the protein. Therefore, computation is carried out only in the area of the column where the protein is present. Computation time is significantly reduced by this approach.
47. Jennings LS, Teo KL, Wang FY, Yu O: Optimal protein separation. Comput Chem Eng 1995, 19:567-573. Optimization algorithms for analytical chromatography are employed for the preparative chromatography of proteins. Objective functions, such as the chromatography response function and chromatography optimization function, are defined and it is shown that each of the problems can be solved by standard min-max optimization problems.
48. Lettner HP, Kaltenbrunner O, Jungbauer A: HETP in process ion-exchange chromatography. J Chromatogr Sci 1995, 33:451-457.
49. Sievers W, Muller G: Dynamische simulation der adsorption an polymeren adsorbentien. Chem Ing Tech 1995, 67:897-901. [Title translation: Dynamic simulation of adsorption on polymeric sorbents.] The computer program SIAD (simulation adsorptiver trennprozesse) allows a very efficient simulation of breakthrough curves through the use of a finite bath model. The computation time with this computer program is remarkably short.
50. Guiochon G, Sarker M: Consolidation of the packing material in chromatographic columns under dynamic axial compression. I. Fundamental study. J Chromatogr A 1995, 704:247-268.
51. Freitag R, Frey D, Horváth C: Effect of bed compression on high-performance liquid chromatography columns with gigaporous polymeric packings. J Chromatogr A 1995, 686:155-177. The behavior of chromatographic packings with gigapores is investigated. Below a certain packing pressure, the theoretical relationship between porosity, permeability and efficiency can be confirmed experimentally. The increased efficiency at higher packing pressure is explained by a higher intra-particle flow due to lower interstitial porosity.