A dual-mode approach to the selective separation of antibodies and their fragments

A novel chromatography method for the separation of antibodies is described. The adsorption of antibodies on the solid phase involves interaction with a ligand that combines mild hydrophobic characteristics and some degree of molecular recognition with a derivative of pyridine. This combined effect results in the adsorption of antibodies in the absence of lyotropic salts. When environmental pH is changed, the ligand becomes ionically charged, allowing the desorption of antibodies. The mechanism of adsorption, involving hydrophobic associations and ionic related interaction, is here qualified as dual-mode. Studies on the determination of the apparent dissociation constant for immunoglobulins G are presented. Adsorption of antibodies from crude feedstocks typically occurs without adjustment of pH or ionic strength. The sorbent is then washed with a buffer to eliminate protein impurities and, when lowering the environmental pH, antibodies are desorbed. The solid-phase material is used for the separation of antibodies from an ascites fluid and from a cell culture supernatant, followed by a polishing step on an hydroxyapatite column. Preliminary studies, related to the ability of the solid phase to separate antibody fragments, are also reported. In these studies, it has been demonstrated that both Fab and Fc fragments from polyclonal IgG are adsorbed to the solid phase under typical binding conditions. Under other defined physico-chemical conditions (ionic strength and pH), separation of both fragments in a single step has been achieved.     

Preparation of DNA topoisomers by RP-18 high-performance liquid chromatography.

A method for the separation of superhelical DNA on the basis of superhelical density by reverse-phase HPLC on RP-18 columns is described. The technique can be used to prepare superhelical DNA in milligram amounts and narrow topoisomer distributions in 0.1 mg amounts. We show example separations of the plasmids pUC18 (2687 bp) and pi AN13 (895 bp). While the best separation for pUC18 yields topoisomer distributions of two or three major components, the small plasmid can be separated into single topoisomer fractions. The basis of the separation is probably an interaction of partially opened bases with the hydrophobic column matrix. This hypothesis is supported by the elution behavior of DNA fragments on this column: DNA fragments with sticky ends, even at a length of several hundred base pairs, elute at much higher methanol concentrations than blunt-ended fragments.     

Structure, force, and energy of a double-stranded DNA oligonucleotide under tensile loads

The end-to-end stretching of a duplex DNA oligonucleotide has been studied using potential of mean force (PMF) calculations based on molecular dynamics (MD) simulations and atomic force microscopy (AFM) experiments. Near quantitative agreement between the calculations and experiments was obtained for both the extension length and forces associated with strand separation. The PMF calculations show that the oligonucleotide extends without a significant energetic barrier from a length shorter than A-DNA to a length 2.4 times the contour length of B-DNA at which the barrier to strand separation is encountered. Calculated forces associated with the barrier are 0.09±0.03 nN, based on assumptions concerning tip and thermal-activated barrier crossing contributions to the forces. Direct AFM measurements show the oligonucleotide strands separating at 2.6±0.8 contour lengths with a force of 0.13±0.05 nN. Analysis of the energies from the MD simulations during extension reveals compensation between increases in the DNA-self energy and decreases in the DNA-solvent interaction energy, allowing for the barrierless extension of DNA beyond the canonical B form. The barrier to strand separation occurs when unfavorable DNA interstrand repulsion cannot be compensated for by favorable DNA-solvent interactions. The present combination of single molecule theoretical and experimental approaches produces a comprehensive picture of the free energy surface of biological macromolecular structural transitions. Received: 2 June 1998 / Revised version: 25 January 1999 / Accepted: 11 February 1999     

Bioavailability of Sorbed 3-Chlorodibenzofuran

One of the main factors impeding the bioremediation of polluted soils, sediments, and aquifers is the low bioavailability of chemicals which are sorbed by organic matter. To obtain more insight into the factors that control the degradation of sorbed compounds, we used a defined model system in which 3-chlorodibenzofuran (3CDF) was the organic contaminant, porous Teflon granules were the sorbent, and Sphingomonas sp. strain HH19k was the test organism. The sorption of 3CDF to Teflon reached equilibrium within 150 min. The curved shape of the sorption isotherm, the extent of sorption, and the desorption kinetics suggested that there was a surface interaction (adsorption) between 3CDF and Teflon which took place mainly inside the pores of the granules. The kinetics of desorption could be ascribed to sorption-retarded radial diffusion inside the granules since the desorption rate not only was correlated with the sorbed-phase concentration, but also depended on the equilibration status of sorption, since (i) the high initial desorption rate sharply declined because of the depletion of 3CDF in the outermost parts of the granules, but high rates were observed again after the system had been given time to reequilibrate, and (ii) the initial desorption rate was higher when the preceding contact time between sorbate and sorbent was shorter (i.e., most 3CDF was still located in the exterior parts of the granules). These characteristics were observed irrespective of whether the desorption was driven by percolating water through the sorbent or by attaching active bacteria to the sorbent. 3CDF consumption by attached cells drove 3CDF desorption to a considerable extent. The attached cells were thus efficiently supplied with desorbing 3CDF. On the basis of our results, we propose that the rate at which a sorbed substrate becomes available for organisms is influenced by (i) the specific affinity of the degrading organisms (i.e., their ability to reduce the aqueous substrate concentration) and (ii) the tendency of the organisms to adhere to the sorbent.     

Low-energy (5-25 eV) electron damage to homo-oligonucleotides

Radiation-induced damage to homo-oligonucleotides is investigated by electron-stimulated desorption of neutral fragments from chemisorbed organic films. Six and 12 mers of cytidine phosphate (poly dCs) and thymidine phosphate (poly dTs) are chemisorbed from various solutions onto a crystalline gold substrate by a thiol modification at the 3' end and are irradiated under ultra-high vacuum conditions with 5-25 eV electrons. The mass selected neutral desorption yields consist mainly of fragments of the DNA bases, i.e. CN and OCN (and/or H2NCN for poly dCs) from both poly dCs and poly dTs, indicating that the electrons interact specifically via fragmentation of the aromatic ring of either of the bases. Other heavier fragments are also detected such as H3CC-CO from poly dTs. The yields generally possess a threshold near 5 eV and a broad maximum around 12-13 eV incident electron energy. Dissociative electron attachment as well as electronically excited neutral or cation states are believed to be responsible for the various desorption yields. The latter yields are consistently larger for oligos chemisorbed from water and acetone solutions, compared to methanol solution. The invariance of the fragment yield intensities with oligo length suggests that the molecules are likely to adsorb almost parallel to the surface.     

Energetics of the strand separation transition in superhelical DNA.

In this paper the values of three free energy parameters governing the superhelical strand separation transition are determined by analysis of available experimental data. These are the free energy, a, needed to initiate a run of separation, the torsional stiffness, C, associated with interstrand winding of the two single strands comprising a separated site and the coefficient, K, of the quadratic free energy associated to residual linking. The experimental data used in this analysis are the locations and relative amounts of strand separation occurring in the pBR322 DNA molecule and the measured residual linking, both evaluated over a range of negative linking differences. The analytic method used treats strand separation as a heteropolymeric, co-operative, two-state transition to a torsionally deformable alternative conformation, which takes place in a circular DNA molecule constrained by the constancy of its linking number. The values determined for these parameters under the experimental conditions (T = 310 K, pH = 7.0, monovalent cation concentration = 0.01 M) are a = 10.84(+/- 0.2) kcal/mol, C = 2.5(+/- 0.3) x 10(-13) erg/rad2 and K = 2350(+/- 80) RT/N, where N is the molecular length in base-pairs. In order to assess the accuracy of the author's theoretical methods, these free energy parameters are incorporated into the analysis of superhelical strand separation in different molecules and under other conditions than those used in their evaluation. First, the temperature dependence of transition is treated, then superhelical strand separation is analyzed in a series of DNA molecules having systematic sequence modifications, and the results of these theoretical analyses are compared with those from experiments. In all molecules, transition is predicted in the range of linking differences where it is seen experimentally. Moreover, it occurs at the specific sequence locations that the analysis predicts, and with approximately the predicted relative amounts of transition at each location. The known sensitivities of this transition to changes of temperature and to small sequence modifications are predicted in a quantitatively precise manner by the theoretical results. The demonstrated high-level precision of these theoretical methods provides a tool for the screening of DNA sequences for sites susceptible to superhelical strand separation, some of which may have regulatory or other biological significance.     

The main role of the sequence-dependent DNA elasticity in determining the free energy of nucleosome formation on telomeric DNAs

Using a competitive reconstitution assay, we measured the free energy spent in nucleosome formation of eight telomeric DNAs, differing in sequence and/or in length. The obtained values are in satisfactorily good agreement with those derived from a theoretical model that allows the calculation of the free energy of nucleosome formation on the basis of sequence-dependent DNA elasticity, using a statistical thermodynamic approach. Both theoretical and experimental evaluations show that telomeres are characterized by the highest free energies of nucleosome formation among all the DNA sequences so far studied. The free energy of nucleosome formation varies according to the different telomeric sequences and the length of the fragments. Theoretical analysis and experimental mapping by lambda exonuclease show that telomeric nucleosomes occupy multiple positions spaced every telomeric repeat. Sequence-dependent DNA elasticity appears as the main determinant of the stability of telomeric nucleosomes and their multiple translational positioning.     

Time-resolved synchrotron radiation X-ray solution scattering study of DNA melting

Time resolved x-ray solution scattering measurements were made during thermal denaturation of DNA from various sources in the temperature range of 20-90 degrees C. Preliminary results on the influence of fragment length, ionic strength, and origin of the DNA on the time course of the scattering are described. Interpretation is based on model calculations of the scattering patterns. The results indicate that, for long DNA fragments at very low ionic strength, the melting process is a continuous phenomenon over the whole temperature range. It is accompanied by a progressive decrease of the radius of gyration of the cross section and of the mass per unit length. For short fragments of 146 base pair nucleosomal core DNA, stiffening of the DNA appears to precede a sharp melting transition.     

Mechanism of spontaneous DNA-DNA interaction of homologous linear duplexes

Previously, we demonstrated the interaction of homologous linear duplexes with formation of four-way DNA structures on the model of five PCR products. We propose that homologous duplex interaction is initiated by the nucleation of several dissociated base pairs of the complementary ends of two fragments with Holliday junction formation, in which cross point migration occurs via spooling of DNA strands from one duplex to the other one, finally resulting in complete resolution into new or previously existing duplexes. To confirm that DNA-DNA interaction involves formation of four-way DNA structures with strand exchange at the cross point, we have demonstrated the strand exchange process between identical duplexes using homologous fragments, harboring either biotin label or (32)P-label. Incubation of the mixture resulted in the addition of (32)P-label to biotin-labeled fragments, and the intensity of (32)P-labeling of biotinylated fragments was dependent upon the incubation duration. DNA-DNA interaction is not based on surface-dependent denaturing, as Triton X-100 does not decrease the formation of complexes between DNA duplexes. The equilibrium concentration of Holliday junctions depends on the sequences of the fragment ends and the incubation temperature. The free energy of Holliday junction formation by the fragments with GC and AT ends differed by 0.6 kcal/mol. Electron microscopic analysis demonstrated that the majority of Holliday junctions harbor the cross point within a 300 base pair region of the fragment ends. This insight into the mechanism of homologous duplex interaction extends our understanding of different DNA rearrangements. Understanding of DNA-DNA interaction is of practical use for better interpretation and optimization of PCR-based analyses.     

Two-point fluorescence detection and automated fraction collection applied to constant denaturant capillary electrophoresis

Constant denaturant capillary electrophoresis (CDCE) has been shown to be a sensitive method to detect point mutations in DNA sequences of 100-bp lengths. Here, we report a significant modifications for the instrumental setup that allows a highly accurate prediction of the elution time of DNA fragments from the capillary and an efficient collection of separated fractions. Fluorescently labeled DNA fragments of TP53 exon 8 wild-type and two mutants (base pair number 14480 and 14525) are detected at two separate points of the same capillary. This permits the precise calculation of the fragment velocity after separation in the heated zone because, at room temperature, all DNA fragments of the same length have the same velocity. Such precision permits the selective collection of separated fragments using an automated fraction collector for additional CDCE analysis or sequencing. Also, the two-point detection allows one to rapidly distinguish between double-stranded and single-stranded DNA fragments of the same length, a process that cannot be achieved with a one-point detection system alone. Both modifications greatly improve the procedure to detect novel mutations by means of CDCE.     

Structure and dynamics of double helices in solution: modes of DNA bending

The long range structure of DNA restriction fragments has been analysed by electro-optical measurements. The overall rotation time constants observed in a low salt buffer with monovalent ions is shown to decrease upon addition of Mg2+ or spermine. Since the circular dichroism and also the limiting value of the linear dichroism remain almost constant under these conditions, the effect is attributed to a change of the long range structure. According to a weakly bending rod model, the persistence length decreases from about 600 A in the absence of Mg2+ or spermine to about 350 A in the presence of these ions. The persistence length measured in the presence of Mg2+ is almost independent of temperature in the range of 10 to 40 degrees C. The nature of DNA bending is analysed by measurements of bending amplitudes and time constants from dichroism decay curves. The observed absence of changes in the bending amplitudes upon addition of Mg2+ or spermine, even though addition induces changes of the persistence length by a factor of 2, is hardly consistent with simple thermal bending. The combined results, including the remarkably small temperature dependence of persistence length and bending amplitude, can be explained by the existence of two bending effects: inherent curvature of DNA dominates at low temperature, whereas thermal bending prevails at high temperature. Analysis of bending amplitudes from dichroism decay curves according to an arc model provides an approximate measure for the degree of bending in restriction fragments. The model is consistent with the observed chain length dependence of bending amplitudes and provides an approximate curvature corresponding to a radius of about 400 A. Thus the curvature observed in restriction fragments is similar to that observed for high molecular DNA condensed into toroids by addition of ions like spermine. Particularly strong bending of DNA is induced by [Co(NH3)6]3+, indicated by an apparent persistence length of 200 A and an increased bending amplitude together with a reduced limit value of the linear dichroism. This effect is attributed to the high charge density of this ion and potential site binding.     

Influence of the energy of H-bond protons on the electron transfer rate in photosynthetic reaction centers

We present here a theoretical interpretation of the temperature dependence of the rate of dark recombination between a primary quinone (Q_A) and a bacteriochlorophyll dimer in the reaction center of Rhodobacter sphaeroides. We were able to describe qualitatively the nonmonotonous character of this dependence using the energy of interaction between an excess electron and H-bond protons. We considered a molecular model of Q_A and two reaction center fragments that make H-bonds with Q_A: His(M219) and Asn(M259)-Ala(M260). We used the two-center approach with regard for electron-phonon interaction in order to calculate the characteristic time of electron tunneling during the recombination reaction. The energy of the phonon emitted/ absorbed during the electron tunneling was determined by the relative shift of donor and acceptor energy levels, the detuning of levels. The detuning was shown to depend on temperature nonmonotonously for H-bonds with double-well potential energy surface. The characteristic time (or the reaction rate) depended on temperature parametrically. The computed dependence was in qualitative agreement with the experimental one.     

Two-dimensional conformation-dependent electrophoresis (2D-CDE) to separate DNA fragments containing unmatched bulge from complex DNA samples

DNA fragments containing mispaired and modified bases, bulges, lesions and specific sequences have altered conformation. Methods for separating complex samples of DNA fragments based on conformation but independent of length have many applications, including (i) separation of mismatched or unmatched DNA fragments from those perfectly matched; (ii) simultaneous, diagnostic, mismatch scanning of multiple fragments; (iii) isolation of damaged DNA fragments from undamaged fragments; and (iv) estimation of reannealing efficiency of complex DNA samples. We developed a two-dimensional conformation-dependent electrophoresis (2D-CDE) method for separating DNA fragments based on length and conformation in the first dimension and only on length in the second dimension. Differences in migration velocity due to conformation were minimized during second dimension electrophoresis by introducing an intercalator. To test the method, we constructed 298 bp DNA fragments containing cytosine bulges ranging from 1 to 5 nt. Bulge-containing DNA fragments had reduced migration velocity in the first dimension due to altered conformation. After 2D-CDE, bulge-containing DNA fragments had migrated in front of an arc comprising heterogeneous fragments with regular conformation. This simple and robust method could be used in both analytical and preparative applications involving complex DNA samples.     

Sequence-Dependent DNA Separation by Anion-Exchange High-Performance Liquid Chromatography

High-performance liquid chromatography (HPLC) system with a new nonporous anion-exchange resin, DNA–NPR, made it possible to rapidly separate DNA fragments up to 20 kbp with high resolution. In order to further characterize this chromatographic DNA separation system, we prepared a mixture of double-stranded DNAs of constant length carrying a fully degenerated 50-bp region and analyzed their chromatographic behavior on the DNA–NPR column. The results indicated that the separation of DNA fragments on the anion-exchange HPLC was governed not only by size, but also by nucleotide sequence: even DNA fragments with the same size and the same base content could be separated on this column. Taking advantage of this characteristic feature of the anion-exchange HPLC, we could readily fractionate human cDNAs with practically acceptable recovery and high resolution. Furthermore, the combination of HPLC and gel electrophoresis realized separation of a mixture of DNA fragments in a two-dimensional pattern.     

DNA sequencing using biotinylated dideoxynucleotides and mass spectrometry

Matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MS) has been explored widely for DNA sequencing. The major requirement for this method is that the DNA sequencing fragments must be free from alkaline and alkaline earth salts as well as other contaminants for accurately measuring the masses of the DNA fragments. We report here the development of a novel MS DNA sequencing method that generates Sanger-sequencing fragments in one tube using biotinylated dideoxynucleotides. The DNA sequencing fragments that carry a biotin at the 3′-end are made free from salts and other components in the sequencing reaction by capture with streptavidin-coated magnetic beads. Only correctly terminated biotinylated DNA fragments are subsequently released and loaded onto a mass spectrometer to obtain accurate DNA sequencing data. Compared with gel electrophoresis-based sequencing systems, MS produces a very high resolution of DNA-sequencing fragments, fast separation on microsecond time scales, and completely eliminates the compressions associated with gel electrophoresis. The high resolution of MS allows accurate mutation and heterozygote detection. This optimized solid-phase DNA-sequencing chemistry plus future improvements in detector sensitivity for large DNA fragments in MS instrumentation will further improve MS for DNA sequencing.     

Analysis of brush-particle interactions using self-consistent-field theory

Non-specific protein adsorption can be reduced by attaching polymer chains by one end to a sorbent surface. End-grafted polymer modified surfaces have also found application in size-based chromatographic bioseparations. To better understand how to tailor surfaces for these applications, a numerical SCF model has been used to calculate theoretical results for the polymer density distribution of interacting polymer chains around a solute particle positioned at a fixed distance from a surface. In addition, the excess energy required to move the particle into the polymer chains (interaction energy) is calculated using a statistical mechanical treatment of the lattice model. The effect of system variables such as particle size, chain length, surface density and Flory interaction parameters on density distributions and interaction energies is also studied. Calculations for the interaction of a solute particle with a surface covered by many polymer chains (a brush) show that the polymer segments will fill in behind the particle quite rapidly as it moves toward the surface. When there is no strong energetic attraction between the polymer and solute we predict that the interaction energy will be purely repulsive upon compression due to losses in conformational entropy of the polymer chains. Above a critical chain length, which depends upon particle size, a maximum in the force required to move the particle toward the surface is observed due to an engulfment of the particle as chains attempt to access the free volume behind the particle. If an attraction exists between the polymer and solute, such that a minimum in the interaction energy is seen, the optimum conditions for solute repulsion occur at the highest surface density attainable. Long chain length can lead to increased solute concentration within the polymer layer due to the fact that an increased number of favourable polymer–solute contacts are able to occur than with short chains at a similar entropic penalty.     

Electric dichroism and bending amplitudes of DNA fragments according to a simple orientation function for weakly bent rods

The linear dichroism is calculated for DNA fragments in their thermal bending equilibrium. These calculations are given for relatively short fragments, where bent molecules can be described by an arc model. Using the measured value of 350 A for the persistence length, the limit dichroism (corresponding to complete alignment) decreases due to thermal bending, e.g., for a fragment with 100 base pairs to 80% of the value expected for straight molecules. Thermal bending should lead to a strong continuous decrease of the dichroism with increasing chain length, which is not observed, however, in electric dichroism experiments due to electric stretching. The influence of the electric field on the bending equilibrium is described by a contribution to the bending energy, which is calculated from the movement of charge equivalents against the potential gradient upon bending. The charge equivalents, which are assigned to the helix ends, are derived from the dipole moments causing the stationary degree of orientation. By this procedure the energy term inducing DNA stretching is given for induced, permanent, and saturating induced dipole models without introduction of any additional parameter. The stationary dichroism at a given electric field strength is then calculated according to an arc model by integration over all angles of orientation of helix axes or chords with respect to the field vector, and at each of these angles the contribution to the dichroism is calculated by integration over all helices with different degrees of bending. Orientation functions obtained by this procedure are fitted to dichroism data measured for various restriction fragments. Optimal fits are found for an induced dipole model with saturation of the polarizability. The difference between orientation functions with and without electric stretching is used to evaluate dichroism bending amplitudes. Both chain length and field strength dependence of bending amplitudes are consistent with experimental amplitudes derived from the dichroism decay in low salt buffers containing multivalent ions like Mg2+, spermine, or [CoNH3)6]3+. Bending amplitudes can be used to evaluate the persistence length from electrooptical data obtained for a single DNA restriction fragment. Bending and stretching effects are considerable already at relatively low chain length, and thus should not be neglected in any quantitative evaluation of experimental data.     

Aggregation of melted DNA by divalent metal ion-mediated cross-linking.

In an accompanying paper we reported the use of differential scanning calorimetry and optical densitometry to characterize the melting and aggregation of 160 bp fragments of calf thymus DNA during heating in the presence of divalent metal cations. Aggregation is observed as thermal denaturation begins and becomes more extensive with increasing temperature until the melting temperature Tm is reached, after which the aggregates dissolve extensively. The order of effectiveness of the metals in inducing aggregation is generally consistent with their ability to induce melting: Cd > Ni > Co > Mn approximately Ca > Mg. Under our experimental conditions (50 mg/ml DNA, 100 mM MCl2, [metal]/[DNA phosphate] approximately 0.6), no measurable aggregates were observed for BaDNA or SrDNA. In this paper we show that the Shibata-Schurr theory of aggregation in the thermal denaturation region provides a good model for our observations. Free energies of cross-linking, induced by the divalent cations, are estimated to be between 34% and 38% of the free energies of base stacking. The ability of a divalent metal cation to induce DNA aggregation can be attributed to its ability to disrupt DNA base pairing and simultaneously to link two different DNA sites.     

Size‐selective DNA separation: Recovery spectra help determine the sodium chloride (NaCl) and polyethylene glycol (PEG) concentrations required

In the presence of sodium chloride (NaCl), DNA fragments can be size‐selectively separated by varying the final concentration of polyethylene glycol (PEG). This separation strategy in combination with the use of paramagnetic particles provides a valuable platform for achieving the desired DNA size interval, which is important in automated library preparation for high‐throughput DNA sequencing. Here, we report the establishment of recovery spectra of DNA fragments that enable the determination of suitable NaCl and PEG concentrations for size‐selective separation. Firstly, at a given NaCl concentration, the recovery equation was obtained by fitting the DNA recovery ratios versus the PEG concentrations using the logistic function to determine the required parameters. Secondly, the slope function of the recovery equation was achieved by deducing its first derivative. Therefore, the recovery spectrum can be generated using the slope function based on those parameters. According to the recovery spectra of different length DNA fragments, suitable NaCl and PEG concentrations can be determined, respectively, by calculating their resolution values and recovery ratios. The strategy was effectively applied to the size‐selective separation of 532‐, 400‐, and 307‐bp fragments at the selected reagent concentrations with recoveries of 96.9, 64.7, and 85.9%, respectively. Our method enables good predictions of NaCl and PEG concentrations for size‐selective DNA separation.     

Study of the mechanism of interaction of antibody (IgG) on two mixed mode sorbents

Purification of target proteins from a crude biological mixture containing proteins, peptides and other biomolecules is the chromatographic challenge. Mixed mode chromatography offers additional selectivities to improve the overall productivity of commercial bioprocesses with novel chromatographic sorbents being introduced to overcome the problem. HEA HyperCel? (n-hexyl amine) and PPA HyperCel? (phenyl propyl amine) are industry scalable mixed mode chromatography sorbents where both hydrophobic and electrostatic interactions are predominant. Our study focuses on understanding the underlying mechanism of interaction of protein with the sorbent. Parameters like buffer conditions, pH and temperature were tuned to study the adsorption and desorption conditions of the protein. Dynamic binding capacity of HEA HyperCel? and PPA HyperCel? sorbents was studied with human IgG as a model protein. Our study shows that, in HEA the interaction of IgG to the sorbent is predominantly hydrophobic as the binding is enhanced (50–60 mg/ml of sorbent) by presence of salt in buffer and increase in temperature. Binding capacity of PPA is 50–60 mg/ml of sorbent irrespective of temperature effect and/or the presence of salt. The chromatographic experiments show that the interaction could be hydrophobic or ionic or some charge transfer mechanism depending upon the buffer conditions.