Kinetics and interaction studies of anti-tetraspanin antibodies and ICAM-1 with extracellular vesicle subpopulations using continuous flow quartz crystal microbalance biosensor.

Continuous flow quartz crystal microbalance (QCM) was utilized to study binding kinetics between EV subpopulations (exomere- and exosome-sized EVs) and four affinity ligands: monoclonal antibodies against tetraspanins (anti-CD9, anti-CD63, and anti-CD81) and recombinant intercellular adhesion molecule-1 (ICAM-1) or CD54 protein). High purity CD9+, CD63+, and CD81+ EV subpopulations of <50 nm exomeres and 50-80 nm exosomes were isolated and fractionated using our recently developed on-line coupled immunoaffinity chromatography - asymmetric flow field-flow fractionation system. Adaptive Interaction Distribution Algorithm (AIDA), specifically designed for the analysis of complex biological interactions, was used with a four-step procedure for reliable estimation of the degree of heterogeneity in rate constant distributions. Interactions between exomere-sized EVs and anti-tetraspanin antibodies demonstrated two interaction sites with comparable binding kinetics and estimated dissociation constants Kd ranging from nM to fM. Exomeres exhibited slightly higher affinity compared to exosomes. The highest affinity with anti-tetraspanin antibodies was achieved with CD63+ EVs. The interaction of EV subpopulations with ICAM-1 involved in cell internalization of EVs was also investigated. EV - ICAM-1 interaction was also of high affinity (nM to pM range) with overall lower affinity compared to the interactions of anti-tetraspanin antibodies and EVs. Our findings proved that QCM is a valuable label-free tool for kinetic studies with limited sample concentration, and that advanced algorithms, such as AIDA, are crucial for proper determination of kinetic heterogeneity. To the best of our knowledge, this is the first kinetic study on the interaction between plasma-derived EV subpopulations and anti-tetraspanin antibodies and ICAM-1.

Biosensor-Enabled Deconvolution of the Avidity-Induced Affinity Enhancement for the SARS-CoV-2 Spike Protein and ACE2 Interaction.

Avidity is an effective and frequent phenomenon employed by nature to achieve extremely high-affinity interactions. As more drug discovery efforts aim to disrupt protein-protein interactions, it is becoming increasingly common to encounter systems that utilize avidity effects and to study these systems using surface-based technologies, such as surface plasmon resonance (SPR) or biolayer interferometry. However, heterogeneity introduced from multivalent binding interactions complicates the analysis of the resulting sensorgram. A frequently applied practice is to fit the data based on a 1:1 binding model, and if the fit does not describe the data adequately, then the experimental setup is changed to favor a 1:1 binding interaction. This reductionistic approach is informative but not always biologically relevant. Therefore, we aimed to develop an SPR-based assay that would reduce the heterogeneity to enable the determination of the kinetic rate constants for multivalent binding interactions using the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein and the human receptor angiotensin-converting enzyme 2 (ACE2) as a model system. We employed a combinatorial approach to generate a sensor surface that could distinguish between monovalent and multivalent interactions. Using advanced data analysis algorithms to analyze the resulting sensorgrams, we found that controlling the surface heterogeneity enabled the deconvolution of the avidity-induced affinity enhancement for the SARS-CoV-2 spike protein and ACE2 interaction.

Advanced Analysis of Biosensor Data for SARS-CoV-2 RBD and ACE2 Interactions.

The traditional approach for analyzing interaction data from biosensors instruments is based on the simplified assumption that also larger biomolecules interactions are homogeneous. It was recently reported that the human receptor angiotensin-converting enzyme 2 (ACE2) plays a key role for capturing SARS-CoV-2 into the human target body, and binding studies were performed using biosensors techniques based on surface plasmon resonance and bio-layer interferometry. The published affinity constants for the interactions, derived using the traditional approach, described a single interaction between ACE2 and the SARS-CoV-2 receptor binding domain (RBD). We reanalyzed these data sets using our advanced four-step approach based on an adaptive interaction distribution algorithm (AIDA) that accounts for the great complexity of larger biomolecules and gives a two-dimensional distribution of association and dissociation rate constants. Our results showed that in both cases the standard assumption about a single interaction was erroneous, and in one of the cases, the value of the affinity constant KD differed more than 300% between the reported value and our calculation. This information can prove very useful in providing mechanistic information and insights about the mechanism of interactions between ACE2 and SARS-CoV-2 RBD or similar systems.

Systematic investigations of peak distortions due to additives in supercritical fluid chromatography.

The impact of eluent components added to improve separation performance in supercritical fluid chromatography was systematically, and fundamentally, investigated. The model system comprised basic pharmaceuticals as solutes and eluents containing an amine (i.e., triethylamine, diethylamine, or isopropylamine) as additive with MeOH as the co-solvent. First, an analytical-scale study was performed, systematically investigating the impact of the additives/co-solvent on solute peak shapes and retentions, using a design of experiments approach; here, the total additive concentration in the eluent ranged between 0.021 and 0.105 % (v/v) and the MeOH fraction in the eluent between 16 and 26 % (v/v). The co-solvent fraction was found to be the most efficient tool for adjusting retentions, whereas the additive fraction was the prime tool for improving column efficiency and peak analytical performance. Next, the impacts of the amine additives on the shapes of the so-called overloaded solute elution profiles were investigated. Two principal types of preparative peak deformations appeared and were investigated in depth, analyzed using computer simulation with mechanistic modeling. The first type of deformation was due to the solute eluting too close to the additive perturbation peak, resulting in severe peak deformation caused by co-elution. The second type of deformation was also due to additive-solute interactions, but here the amine additives acted as kosmotropic agents, promoting the multilayer adsorption to the stationary phase of solutes with bulkier aryl groups.

Rapid affinity chromatographic isolation method for LDL in human plasma by immobilized chondroitin-6-sulfate and anti-apoB-100 antibody monolithic disks in tandem.

Low-density lipoprotein (LDL) is considered the major risk factor for the development of atherosclerotic cardiovascular diseases (ASCVDs). A novel and rapid method for the isolation of LDL from human plasma was developed utilising affinity chromatography with monolithic stationary supports. The isolation method consisted of two polymeric monolithic disk columns, one immobilized with chondroitin-6-sulfate (C6S) and the other with apolipoprotein B-100 monoclonal antibody (anti-apoB-100 mAb). The first disk with C6S was targeted to remove chylomicrons, very-low-density lipoprotein (VLDL) particles, and their remnants including intermediate-density lipoprotein (IDL) particles, thus allowing the remaining major lipoprotein species, i.e. LDL, lipoprotein(a) (Lp(a)), and high-density lipoprotein (HDL) to flow to the anti-apoB-100 disk. The second disk captured LDL particles via the anti-apoB-100 mAb attached on the disk surface in a highly specific manner, permitting the selective LDL isolation. The success of LDL isolation was confirmed by different techniques including quartz crystal microbalance. In addition, the method developed gave comparable results with ultracentrifugation, conventionally used as a standard method. The reliable results achieved together with a short isolation time (less than 30 min) suggest the method to be suitable for clinically relevant LDL functional assays.

Reliable Strategy for Analysis of Complex Biosensor Data.

When using biosensors, analyte biomolecules of several different concentrations are percolated over a chip with immobilized ligand molecules that form complexes with analytes. However, in many cases of biological interest, e.g., in antibody interactions, complex formation steady-state is not reached. The data measured are so-called sensorgram, one for each analyte concentration, with total complex concentration vs time. Here we present a new four-step strategy for more reliable processing of this complex kinetic binding data and compare it with the standard global fitting procedure. In our strategy, we first calculate a dissociation graph to reveal if there are any heterogeneous interactions. Thereafter, a new numerical algorithm, AIDA, is used to get the number of different complex formation reactions for each analyte concentration level. This information is then used to estimate the corresponding complex formation rate constants by fitting to the measured sensorgram one by one. Finally, all estimated rate constants are plotted and clustered, where each cluster represents a complex formation. Synthetic and experimental data obtained from three different QCM biosensor experimental systems having fast (close to steady-state), moderate, and slow kinetics (far from steady-state) were evaluated using the four-step strategy and standard global fitting. The new strategy allowed us to more reliably estimate the number of different complex formations, especially for cases of complex and slow dissociation kinetics. Moreover, the new strategy proved to be more robust as it enables one to handle system drift, i.e., data from biosensor chips that deteriorate over time.

Impact of column and stationary phase properties on the productivity in chiral preparative LC.

By generating 1500 random chiral separation systems, assuming two-site Langmuir interactions, we investigated numerically how the maximal productivity (PR,max ) was affected by changes in stationary phase adsorption properties. The relative change in PR,max , when one adsorption property changed 10%, was determined for each system and for each studied parameter the corresponding productivity change distribution of the systems was analyzed. We could conclude that there is no reason to have columns with more than 500 theoretical plates and larger selectivity than 3. More specifically, we found that changes in selectivity have a major impact on PR,max if it is below ∼2 and, interestingly, increasing selectivity when it is above ∼3 decreases PR,max . Increase in relative saturation capacity will have a major impact on PR,max if it is below ∼40%, but only modest above this percent. Increasing total monolayer saturation capacity, or decreasing the first eluting enantiomer's retention factor, will have a modest effect on PR,max and increased efficiency will have almost no effect at all on PR,max unless it is below ∼500 theoretical plates. Finally, we showed that chiral columns with superior analytic performance might have inferior preparative performance, or vice versa. It is, therefore, not possible to assess columns based on their analytical performance alone.

Thermodynamic and kinetic approaches for evaluation of monoclonal antibody - Lipoprotein interactions.

Two complementary instrumental techniques were used, and the data generated was processed with advanced numerical tools to investigate the interactions between anti-human apoB-100 monoclonal antibody (anti-apoB-100 Mab) and apoB-100 containing lipoproteins. Partial Filling Affinity Capillary Electrophoresis (PF-ACE) combined with Adsorption Energy Distribution (AED) calculations provided information on the heterogeneity of the interactions without any a priori model assumptions. The AED calculations evidenced a homogenous binding site distribution for the interactions. Quartz Crystal Microbalance (QCM) studies were used to evaluate thermodynamics and kinetics of the Low-Density Lipoprotein (LDL) and anti-apoB-100 Mab interactions. High affinity and selectivity were observed, and the emerging data sets were analysed with so called Interaction Maps. In thermodynamic studies, the interaction between LDL and anti-apoB-100 Mab was found to be predominantly enthalpy driven. Both techniques were also used to study antibody interactions with Intermediate-Density (IDL) and Very Low-Density (VLDL) Lipoproteins. By screening affinity constants for IDL-VLDL sample in a single injection we were able to distinguish affinity constants for both subpopulations using the numerical Interaction Map tool.

A model free method for estimation of complicated adsorption isotherms in liquid chromatography.

Here we show that even extremely small variations in the adsorption isotherm can have a tremendous effect on the shape of the overloaded elution profiles and that the earlier in the adsorption isotherms the variation take place, the larger its impact on the shape of the elution profile. These variations are so small that they can be "hidden" by the discretization and in the general experimental noise when using traditional experimental methods, such as frontal analysis, to measure adsorption isotherms. But as the effects of these variations are more clearly visible in the elution profiles, the Inverse Method (IM) of adsorption isotherm estimation is an option. However, IM usually requires that one selects an adsorption isotherm model prior to the estimation process. Here we show that even complicated models might not be able to estimate the adsorption isotherms with multiple inflection points that small variations might give rise to. We therefore developed a modified IM that, instead of fixed adsorption isotherm models, uses monotone piecewise interpolation. We first validated the method with synthetic data and showed that it can be used to estimate an adsorption isotherm, which accurately predicts an extremely "strange" elution profile. For this case it was impossible to estimate the adsorption isotherm using IM with a fixed adsorption model. Finally, we will give an example of a real chromatographic system where adsorption isotherm with inflection points is estimated by the modified IM.

Investigation of plateau methods for adsorption isotherm determination in supercritical fluid chromatography.

The Perturbation Peak (PP) method and Frontal analysis (FA) are considered as the most accurate methods for adsorption isotherms determination in liquid chromatography. In this study we investigate and explain why this is not the case in Supercritical Fluid Chromatography (SFC), where the PP method does not work at all, using a modern analytical system. The main reason was found to be that the solute to be studied must be dissolved in the MeOH reservoir before it is mixed with CO2. Since the solute occupies a certain partial volume in the reservoir, the larger the solute content the larger this fractional volume will be, and the final MeOH fraction in the mobile phase will then be smaller compared to the bulk mobile phase without solute in the modifier. If the retention of small injections on the concentration plateaus, i.e., "analytical-size" perturbation peaks, is sensitive to small variations of MeOH in the eluent, this will seriously decrease the accuracy of the PP method. This effect was verified and compensated for and we also demonstrated that the same problem will occur in frontal analysis, another concentration plateau method.

Relative importance of column and adsorption parameters on the productivity in preparative liquid chromatography II: Investigation of separation systems with competitive Langmuir adsorption isotherms.

In this study we investigated how the maximum productivity for commonly used, realistic separation system with a competitive Langmuir adsorption isotherm is affected by changes in column length, packing particle size, mobile phase viscosity, maximum allowed column pressure, column efficiency, sample concentration/solubility, selectivity, monolayer saturation capacity and retention factor of the first eluting compound. The study was performed by generating 1000 random separation systems whose optimal injection volume was determined, i.e., the injection volume that gives the largest achievable productivity. The relative changes in largest achievable productivity when one of the parameters above changes was then studied for each system and the productivity changes for all systems were presented as distributions. We found that it is almost always beneficial to use shorter columns with high pressure drops over the column and that the selectivity should be greater than 2. However, the sample concentration and column efficiency have very limited effect on the maximum productivity. The effect of packing particle size depends on the flow rate limiting factor. If the pumps maximum flow rate is the limiting factor use smaller packing, but if the pressure of the system is the limiting factor use larger packing up to about 40µm.

Sample conditions to avoid pH distortion in RP-LC.

Band deformations might take place for acids and bases in preparative applications and adsorption studies where it is necessary to use overloaded injections. In this study, we focus on how deformations can be prevented without using highly concentrated buffers that may precipitate in the eluent. We have systematically investigated how the elution zones depend on which protolytic form the analyte has when it is dissolved. Basic and acidic model compounds are investigated using eluents with different pH values and the resulting elution profiles are compared when the analytes are dissolved in their protonated and deprotonated form, i.e., in uncharged form or as different kinds of salts. Depending on the analyte's protolytic form, a sample zone is created at the column inlet whose pH deviates, more or less, from the bulk eluent's. If the local adsorption strength in this sample zone is greater than the bulk eluent's, the elution profiles are compressed. Under opposite conditions, the eluted bands are more or less deformed and may even be split; completely different deformations can even take place for different kinds of salt combinations. Explanations of these, and other, effects, together with detailed guidelines for proper sample preparation to avoid peak deformations, are given.

Determination of adsorption isotherms in supercritical fluid chromatography.

In this study we will demonstrate the potential of modern integrated commercial analytical SFC-systems for rapid and reliable acquisition of thermodynamic data. This will be done by transferring the following adsorption isotherm determination methods from liquid chromatography (LC) to supercritical fluid chromatography (SFC): Elution by Characteristic Points (ECP), the Retention Time Method (RTM), the Inverse Method (IM) and the Perturbation Peak (PP) method. In order to transfer these methods to SFC in a reliable, reproducible way we will demonstrate that careful system verification using external sensors of mass flow, temperature and pressure are needed first. The adsorption isotherm data generated by the different methods were analyzed and compared and the adsorption isotherms ability to predict new experimental elution profiles was verified by comparing experiments with simulations. It was found that adsorption isotherm data determined based on elution profiles, i.e., ECP, IM and RTM, were able to accurately predict overloaded experimental elution profiles while the more tedious and time-consuming PP method, based on small injections on concentration plateaus, failed in doing so.

Enhanced interpretation of adsorption data generated by liquid chromatography and by modern biosensors.

In this study we demonstrate the importance of proper data processing in adsorption isotherm estimations. This was done by investigating and reprocessing data from five cases on two closely related platforms: liquid chromatography (LC) and biosensors. The previously acquired adsorption data were reevaluated and reprocessed using a three-step numerical procedure: (i) preprocessing of adsorption data, (ii) adsorption data analysis and (iii) final rival model fit. For each case, we will discuss what we really measure and what additional information can be obtained by numerical processing of the data. These cases clearly demonstrate that numerical processing of LC and biosensor data can be used to gain deeper understanding of molecular interactions with adsorption media. This is important because adsorption data, especially from biosensors, is often processed using old and simplified methods.

Relative importance of column and adsorption parameters on the productivity in preparative liquid chromatography. I: investigation of a chiral separation system.

Starting out from an experimental chiral separation system we have used computer simulations for a systematic investigation on how the maximum productivity depends on changes in column length, packing particle size, column efficiency, back pressure, sample concentration/solubility, selectivity, retention factor of the first eluting component and monolayer saturation capacity. The study was performed by changing these parameters, one at a time, and then calculating the corresponding change in maximum productivity. The three most important parameters for maximum production rate was found to be (i) the selectivity (ii) the retention factor of the first eluting component and (iii) the column length. Surprisingly, the column efficiency and sample concentration/solubility were of minor importance. These findings can be used as rough guidelines for column selection, e.g. a low-efficiency column are more likely perform better, in terms of productivity, than a high-efficiency column that have higher retention factor for the first eluting component.

Optimization strategies accounting for the additive in preparative chiral liquid chromatography.

This study is an in-depth investigation on how numerical optimization strategies that also account for the additive type and concentration, in preparative batch chromatography, should be performed. As a model system, the separation of Z-(R,S)-2-aminobutyric acid enantiomers on a quinidine carbamate-based chiral stationary phase in polar organic mode was used, with different additive strengths of acetic acid or hexanoic acid in methanol. The inverse method was used to determine the competitive adsorption isotherm parameters for the enantiomers and the additives. Three different optimization strategies were examined: (1) injection volume optimization, (2) optimization of injection volume and additive concentration, and (3) full optimization including injection volume, additive concentration, sample concentration and flow rate. It was concluded that (i) it is important to incorporate the additive concentration in the optimization procedure to achieve the highest production rates, (ii) the full optimization strategy had the overall best results, and (iii) the selection of additive is very important (here acetic acid additive was superior to the hexanoic acid additive). By including the additive in the adsorption model and in the numerical optimization it is not only possible to achieve higher production rates but also to properly select the additive that is most advantageous for the specific separation problem.

Enantioseparation of omeprazole--effect of different packing particle size on productivity.

Enantiomeric separation of omeprazole has been extensively studied regarding both product analysis and preparation using several different chiral stationary phases. In this study, the preparative chiral separation of omeprazole is optimized for productivity using three different columns packed with amylose tris (3,5-dimethyl phenyl carbamate) coated macroporous silica (5, 10 and 25 µm) with a maximum allowed pressure drop ranging from 50 to 400 bar. This pressure range both covers low pressure process systems (50-100 bar) and investigates the potential for allowing higher pressure limits in preparative applications in a future. The process optimization clearly show that the larger 25 µm packing material show higher productivity at low pressure drops whereas with increasing pressure drops the smaller packing materials have substantially higher productivity. Interestingly, at all pressure drops, the smaller packing material result in lower solvent consumption (L solvent/kg product); the higher the accepted pressure drop, the larger the gain in reduced solvent consumption. The experimental adsorption isotherms were not identical for the different packing material sizes; therefore all calculations were recalculated and reevaluated assuming identical adsorption isotherms (with the 10 µm isotherm as reference) which confirmed the trends regarding productivity and solvent consumption.

Why ultra high performance liquid chromatography produces more tailing peaks than high performance liquid chromatography, why it does not matter and how it can be addressed.

The purpose of this study is to demonstrate, with experiments and with computer simulations based on a firm chromatographic theory, that the wide spread perception of that the United States Pharmacopeia tailing factor must be lower than 2 (T(f)<2) is questionable when using the latest generation of LC equipment. It is shown that highly efficient LC separations like those obtained with sub-2 µm porous and 2.7 µm superficially porous particles (UHPLC) produce significantly higher T(f)-values than the corresponding separation based on 3 µm porous particles (HPLC) when the same amount of sample is injected. Still UHPLC separations provide a better resolution to adjacent peaks. Expressions have been derived that describe how the T(f)-value changes with particle size or number of theoretical plates. Expressions have also been derived that can be used to scale the injection volume based on particle size or number of theoretical plates to maintain the T(f)-value when translating a HPLC separation to the corresponding UHPLC separation. An aspect that has been ignored in previous publications. Finally, data obtained from columns with different age/condition indicate that T(f)-values should be complemented by a peak width measure to provide a more objective quality measure.

Injection profiles in liquid chromatography II: predicting accurate injection-profiles for computer-assisted preparative optimizations.

In computer assisted optimization of liquid chromatography it has been known for some years that it is important to use experimental injection profiles, instead of rectangular ones, in order to calculate accurate elution bands. However, the incorrectly assumed rectangular profiles are still mostly used especially in numerical optimizations. The reason is that the acquisition of injection profiles, for each injection volume and each flow rate considered in a computer-assisted optimization requires a too large number of experiments. In this article a new function is proposed, which enables highly accurate predictions of the injection profiles and thus more accurate computer optimizations, with a minimum experimental effort. To model the injection profiles for any injection volume at a constant flow rate, as few as two experimental injection profiles are required. If it is desirable to also take the effect of flow rate on the injection profiles into account, then just two additional experiments are required. The overlap between fitted and experimental injection profiles at different flow rates and different injection volumes were excellent, more than 90%, using experimental injection profiles from just four different injection volumes at two different flow rates. Moreover, it was demonstrated that the flow rate has a minor influence on the injection profiles and that the injection volume is the main parameter that needs to be accounted for.

A systematic investigation of algorithm impact in preparative chromatography with experimental verifications.

Computer-assisted optimization of chromatographic separations requires finding the numerical solution of the Equilibrium-Dispersive (ED) mass balance equation. Furthermore, the competitive adsorption isotherms needed for optimization are often estimated numerically using the inverse method that also solves the ED equations. This means that the accuracy of the estimated adsorption isotherm parameters explicitly depends on the numerical accuracy of the algorithm that is used to solve the ED equations. The fast and commonly used algorithm for this purpose, the Rouchon Finite Difference (RFD) algorithm, has often been reported not to be able to accurately solve the ED equations for all practical preparative experimental conditions, but its limitations has never been completely and systematically investigated. In this study, we thoroughly investigate three different algorithms used to solve the ED equations: the RFD algorithm, the Orthogonal Collocation on Finite Elements (OCFE) method and a Central Difference Method (CDM) algorithm, both for increased theoretical understanding and for real cases of industrial interest. We identified discrepancies between the conventional RFD algorithm and the more accurate OCFE and CDM algorithms for several conditions, such as low efficiency, increasing number of simulated components and components present at different concentrations. Given high enough efficiency, we experimentally demonstrate good prediction of experimental data of a quaternary separation problem using either algorithm, but better prediction using OCFE/CDM for a binary low efficiency separation problem or separations when the compounds have different efficiency. Our conclusion is to use the RFD algorithm with caution when such conditions are present and that the rule of thumb that the number of theoretical plates should be greater than 1000 for application of the RFD algorithm is underestimated in many cases.