In view of the continuous increase in the number of new biologics arriving on the market, there is a pressing need for efficient process development technologies in high-throughput formats. Even now, purification of biopharmaceuticals is dominated by sequential liquid chromatography in a packed column bed.
This methodology is difficult, time-consuming, and very expensive. Repligen’s OPUS® RoboColumn® technology is a resin-screening tool that uses miniature chromatography columns (MCCs), and applies the packed bed principle of the traditional chromatography on a microliter scale.
Operating MCCs together with an automated liquid handling system allows simultaneous study of various experimental conditions, and enables more information to be obtained per run in less time by using less feed material.
MCCs have a wide range of possible applications such as sample preparation, dynamic binding capacity determination, establishing cleaning regimens, and optimization of step elution conditions[2, 3, 4, 5].
This article describes how Merck KGaA in Darmstadt, Germany, has implemented automated resin screening on the Freedom EVO® liquid handling platform. It also includes example results from the optimization of protein separation in MCCs employing a pseudo-linear gradient elution.
On the basis of the concept of discretization, linear gradient elution was converted into a multi-step gradient protocol. The influence of flow rate, column scale, multi-step gradient, and fractionation scheme on separation efficiency was studied.
The optimal settings for effective column performance were also defined. The results obtained demonstrate that automated chromatography on a miniaturized scale reliably predicts separation performance on a lab scale with conventional chromatography systems, such as ÄKTA™ (GE Healthcare Life Sciences), and also helps speed up the resin screening process.
Material and methods
The process was automated on a Tecan Freedom EVO® 200 workstation, as shown in Figure 1, which was fitted with a robotic manipulator (RoMa) arm for fully automated plate and lid handling, an eight-channel liquid handling (LiHa) arm, a Te-Chrom™ Shuttle for collection of the eluates in microplates, and a Te-Chrom module and waste tray to house OPUS RoboColumns.
Integrated hotels provided on-deck storage for 18 deep-well and 18 UV-transparent microplates, for preparing elution buffers and fraction collection, respectively. The system was also provided with microplate carriers and many different tubes and troughs for liquid storage. A Tecan Infinite® M200 PRO was used to measure the absorbance of the individual eluted fractions, and the measurements were analyzed with Magellan™ data reduction software.
Figure 1. Freedom EVO 200 workstation equipped with LiHa and RoMa Arms, a Te-Chrom module and an Infinite® M200 PRO microplate reader.
Proteins, buffers and chromatography columns
A three-component mixture of α-chymotrypsinogen A, cytochome C, and lysozyme was employed as the model feed in order to enable reliable and simple assessment of the separation performance by determination of the resolution values between adjacent peaks.
Column chromatography was performed with Eshmuno® CEX prototype (Merck KGaA), a strong cation exchange resin, prepacked in OPUS RoboColumns with column volumes of either 0.6 ml (3 cm bed height) or 0.2 ml (1 cm bed height). An ÄKTA chromatography system with reusable 5 mm ID Superformance® columns (Götec-Labortechnik GmbH) was used to perform reference runs.
Pseudo-linear gradient formation on the Freedom EVO
The LiHa Arm was used to apply a finely graduated multi-step gradient to the MCCs, mimicking linear gradient elution (Figure 2). A group of gradient elution buffers with increasing salt concentration was prepared from three stock solutions.
The stock solutions were: buffer A, the equilibration buffer without NaCl; buffer B, the final elution buffer with 1 M or 0.5 M NaCl; and buffer C, a 1:1 mixture of buffers A and B. Deep-well plates were used to pipette the stock solutions, which were then stored without lids at room temperature.
Three stock buffers were employed, so that liquid levels do not fall below a minimum pipetting volume of 20 μl. The required buffer composition of each step and the related buffer volumes were measured in Microsoft Excel®.
A pipetting worklist was collected and imported into the robotic script. Low and high salt concentration buffer stock solutions were pipetted individually into deep-well plates such that a minimum final volume of 950 μl is obtained.
The deep-well plates were sealed with an adhesive foil and manually mixed. The final elution buffers were sequentially applied to the columns, so that a multi-step gradient is formed. Parallel pipetting with the LiHa Arm enables parallel preparation of eight elution buffers, thus allowing an option to use a different buffer salt, pH, or salt gradient for each one.
Figure 2. Step gradient definition
Freedom EVO liquid classes and pipetting parameters
The elution buffers were prepared using the Freedom EVO’s standard liquid class for water. An optimized liquid class was defined for dispensing of buffers and samples into OPUS RoboColumns, so that sample dilution is prevented, the sample is separated from the system liquid, and a small partitioning volume (50 μl extra sample plus 20 μl sample excess volume) is aspirated between two air gaps (Figure 3).
Figure 3. An optimized liquid class for dispensing of samples and buffers
Gradient elution with the Freedom EVO
OPUS RoboColumns were equilibrated with 5 column volumes of 20 mM sodium phosphate buffer for all automated liquid handling experiments. The column was loaded with 10 mg of total protein per ml of resin, which was washed with 2.6 column volumes of equilibration buffer before elution.
Next, pseudo-linear gradient protocols with varying gradient slopes and elution flow rates were tested (refer Results and Discussion). The flow velocity for all steps, barring elution, was 100 cm/hour, which corresponded to a residence time of 1.8 minutes. The Infinite® M200 PRO was used to measure flow-through absorbance at A528 (cytochrome C), A280 (total protein), and A975-900 (path length correction).
Gradient elution with ÄKTApurifier™
20 mM phosphate with pH 6.0 was used as the equilibration buffer, while a pH 6 buffer consisting of 20 mM phosphate + 1 M NaCl was used to perform elution. The column load was configured to 10 mg of total protein per ml of packed resin volume.
Linear gradient elution, from 0 to 100% elution buffer, was carried out after sample application. The flow velocity during loading and re-equilibration was determined to be a constant 150 cm/hour.
Results and discussion
Several factors impacting the quality of automated protein separation in MCCs were studied to determine the optimal parameter settings to achieve high operational efficiency.
A comparison was made of the separation quality obtained using MCCs on the Freedom EVO with the separation achieved on the lab scale ÄKTA system. Lastly, the two resin screening techniques were compared with regard to time demands.
Flow velocity (residence time)
Chromatograms obtained from separation runs performed on 600 μl OPUS RoboColumns at elution flow velocities ranging between 25 cm/hour and 300 cm/hour are shown in Figure 4. Step height (10.4 mM NaCl), step length (300 μl), gradient slope (GA, 10.2 mM/cm), and fraction size (300 μl) were identical for all runs.
The separation at 300 cm/hour elution flow velocity gave rise to overlapping peaks, thus demonstrating only partial separation (Figure 4A). A substantial improvement was achieved by reducing the elution flow velocity to 100 cm/hour (Figure 4B). Additional gain in peak resolution, albeit to a smaller extent, was obtained by reducing the elution flow velocity to 25 cm/hour (Figure 4C).
Figure 4. Peak separation on 600 μl OPUS® RoboColumns® at three different flow velocities. Protein elution order: chymotrypsinogen A (1), cytochrome C (2), lysozyme (3)
Peak resolution values as a function of residence time are shown in Figure 5. The regression demonstrates an expected hyperbolic relationship between peak resolution and residence time. In order to obtain a high resolution using 600 μl format OPUS RoboColumns, elution flow velocity must not be more than 100 cm/hour, corresponding to a minimum residence time of 1.8 minutes.
Residence times of a similar order of magnitude, ranging between 3 minutes and 4 minutes, are generally applied in normal scale column operation.
Figure 5. Results from separation runs with 600 μl OPUS RoboColumns showing resolution values calculated for adjacent peaks as a function of residence time. RS1/2 refers to the resolution of chymotrypsinogen A (1) and cytochrome C (2), RS2/3 to the resolution of cytochrome C (2) and lysozyme (3).
Peak separation is also impacted by the slope of the elution gradient, as shown in Figure 6. By adjusting the elution step volume, the slope was made to vary between 3.4 mM/cm and 27.3 mM/cm. There was no change in the elution flow velocity (100 cm/hour) and step height (10.4 mM NaCl).
The fraction volume was equal to the step length, but not more than 300 μl so as to ensure a minimum of one data point per step. The data (not shown) suggests that peak separation is enhanced when there is a reduction in the gradient slope.
Expectedly, flattening of the gradient improves the separation performance, although the effect was less under the conditions tested; a 2.7-fold reduction of the gradient slope, from 27.3 mM/cm to 10.3 mM/cm, resulted in only a slight increase in the resolution. The effect was more pronounced, from 27.3 mM/cm to 3.4 mM/cm, when the slope was changed by a factor of 8.
Figure 6. Impact of gradient slope on peak resolution using a 600 μl MCC operated in a pseudo-linear gradient elution mode with a liquid handling system. RS1/2 refers to the resolution of chymotrypsinogen A (1) and cytochrome C (2), RS2/3 to the resolution of cytochrome C (2) and lysozyme (3).
Column length had no influence on resolution when a (see Reference) direct comparison of separations performed with 200 μl and 600 μl MCCs was made.
However, the choice of MCC format should be based on resolution requirements as well as analytical constraints. A more comprehensive characterization by several analytical methods is enabled by the larger quantity of sample material obtained from the 600 μl MCC format.
Assay run time and labware consumption
Other than the quality of peak separation, labware consumption and method run time must also be considered when designing a high-throughput protocol for resin screening. A protocol employing a 25 cm/hour elution flow rate and 10.2 mM/cm gradient slope (Table 1, MCC run 7) achieved separation within a period of 6.3 hours, by using only nine fraction collection plates.
However, a protocol with a 100 cm/hour elution flow velocity and gradient slope of 3.4 mM/cm (MMC run 11) needed a longer method run time of 7.6 hours, incurred a greater fractional plate consumption (17 plates), and gave slightly poorer resolution.
Table 1. Comparison of assay run time and labware consumption using different elution flow velocities and gradient slopes.
Comparison of the automated Freedom EVO and ÄKTA resin screening approaches
The combination of a liquid handling system and a pseudo-linear gradient elution approach offers great potential for the improvement of resin selectivity screening. Figure 7 provides a comparison of the time demands for screening experiments performed by employing the Freedom EVO platform or an ÄKTA system, based on linear gradient elution performed in 3 cm bed height columns.
In the two cases, separations were carried out at 100 cm/hour linear flow velocity (throughout loading, washing, and elution steps) and a gradient slope of 3.4 mM/cm, with a model protein feed of 10 mg total protein per ml of packed resin.
A 600 μl step length was employed for the Freedom EVO protocol. A breakdown of the individual steps of the technique shows that extra time is needed for offline UV reading, offline elution buffer mixing, plate handling, and holding times caused by pump flow interruptions.
However, the longer method run time is offset by the capability to carry out parallel pipetting, enabling up to eight columns to be tested in less than eight hours. If eight columns run sequentially on an ÄKTA system, it would take over 24 hours (Figure 7).
Figure 7. Time requirements for the Freedom EVO and ÄKTA methods, using 30 x 5 mm ID miniature chromatography columns. Bar diagram compares the total time necessary for performing single and eight column runs using the following protocol: 0.5 column volume (CV) load at 3 minutes, 3 CV wash at 3 minutes, linear gradient elution from 0-1 M NaCl in 97 CV at 1.8 minutes, 5 CV re-equilibration at 3 minutes, UV measurement.
Regardless of its different operational mode and design, the Freedom EVO exhibited very similar separation efficiency to the ÄKTA system, as shown Figure 8. Peak resolution and conductivity values at the maximum peak height were almost the same.
The two systems were equally suited for the operation of MCCs as they delivered comparable results. However, a significant advantage was the higher throughput and automated sample handing offered by the Freedom EVO, making this technique superior for resin screening.
Figure 8. Separation of a ternary protein mixture with Eshmuno® CEX prototype resin in a 30 x 5 mm ID OPUS RoboColumn on a Freedom EVO® (black lines) and 30 x 5 mm ID Superformance column on an ÄKTA system (red lines). UV absorbance at 280 nm and the salt gradient are shown. With similar miniature column formats, the same separation performance and chromatographic results were obtained.
- Fully automated step gradient elution can be carried out by using the Freedom EVO liquid handling workstation.
- Gradient slope settings and optimal flow velocity are vital for the effective operation of OPUS RoboColumns on a liquid handling system.
- Comparable data is obtained by automation of OPUS RoboColumn chromatography on a Freedom EVO workstation to lab scale chromatography on the ÄKTA system. This data can be used to reliably predict protein elution conditions on large-scale formats.
- Automation of OPUS RoboColumn protocols on the Freedom EVO enables improved chromatography resin screening, increased throughput, and reduced feed requirements.
We sincerely thank André Kiesewetter for providing us with data and interesting scientific discussions. For more details on optimizing gradient elution on the Freedom EVO, please also refer to the publication Kiesewetter A, Menstell P, Peeck LH, Stein A. Development of pseudo-linear gradient elution for high-throughput resin selectivity screening in RoboColumn Format. Biotechnol Prog. DOI: 10.1002/btpr.2363
- Atoll GmbH Germany. MediaScout® RoboColumn® Instruction Manual. Automated Parallel Chromatography No. 01-08R-E02 TS. 2010.
- Yoo D, Provchy J, Park C, Schulz C, Walker K. Automated high-throughput protein purification using an ÄKTApurifier and a CETAC autosampler. J. Chromatogr. A 2014, 1344: 23–30.
- Osberghaus A, Drechsel K, Hansen S, Hepbildikler SK, Nath S, Haindl M, von Lieres E, Hubbuch J. Model-integrated process development demonstrated on the optimization of a robotic cation exchange step. Chem. Eng. Sci. 2012; 76: 129–139.
- Susanto A, Knieps-Grünhagen E, von Lieres E, Hubbuch J. High Throughput Screening for the Design and Optimization of Chromatographic Processes: Assessment of Model Parameter Determination from High Throughput Compatible Data. Chem. Eng. Technol. 2008; 31(12): 1846–1855.
- Elich T, Iskra T, Daniels W, Morrison CJ. High Throughput Determination of Cleaning Solutions to Prevent the Fouling of an Anion Exchange Resin. Biotechnol. Bioeng. 2015; 9999: 1–9.
- Kiesewetter A, Menstell P, Peeck LH, Stein A. Development of pseudo-linear gradient elution for high-throughput resin selectivity screening in RoboColumn format. Biotechnol Prog. DOI: 10.1002/btpr.2363 factor.
Tecan is a leading global provider of automated laboratory instruments and solutions. Their systems and components help people working in clinical diagnostics, basic and translational research and drug discovery bring their science to life.
In particular, they develop, produce, market and support automated workflow solutions that empower laboratories to achieve more. Their Cavro branded instrument components are chosen by leading instrumentation suppliers across multiple disciplines.
They work side by side with a range of clients, including diagnostic laboratories, pharmaceutical and biotechnology companies and university research centers. Their expertise extends to developing and manufacturing OEM instruments and components, marketed by their partner companies. Whatever the project – large or small, simple or complex – helping their clients to achieve their goals comes first.
They hold a leading position in all the sectors they work in and have changed the way things are done in research and development labs around the world. In diagnostics, for instance, they have raised the bar when it comes to the reproducibility and throughput of testing.
In under four decades Tecan has grown from a Swiss family business to a brand that is well established on the global stage of life sciences. From pioneering days on a farm to the leading role our business assumes today – empowering research, diagnostics and many applied markets around the world
Sponsored Content Policy: News-Medical.net publishes articles and related content that may be derived from sources where we have existing commercial relationships, provided such content adds value to the core editorial ethos of News-Medical.Net which is to educate and inform site visitors interested in medical research, science, medical devices and treatments.