Expanded-Bed Adsorption Chromatography
`
`UNIT 8.8
`
`All protein purification processes can be divided into three main steps (UNIT 8.1). The
`first is the so-called capture step, since it involves immobilization of the target protein
`on an adsorptive surface. This step can also be viewed as a combination of clarification,
`concentration, stabilization, and initial purification. Most downstream processes require
`that the feed stream be free of particulate material, and so the initial capture phase
`of most processes includes some sort of particle reduction operation. If clarification
`can be achieved on a chemically active adsorptive surface, such as a chromatographic
`medium or a membrane, then particulate removal can be accomplished simultaneously
`with concentration, stabilization, and initial purification. Since the target molecule is
`immobilized along with only a few closely related molecules, the target is concentrated
`and somewhat purified after the capture step. It is also known that when a protein is
`immobilized, it becomes stabilized, because it is somewhat resistant to protease attack.
`Therefore, the most successful capture steps will result in some degree of concentration,
`purification, and stability enhancement, along with particulate removal.
`
`The second step in a purification process is called intermediate purification. Inthis step,
`most of the remaining contaminants are removed from the product. Chromatography is
`the most useful tool in this phase, providing high selectivity and a high rate of product
`recovery. The most useful and commonly used techniques are hydrophobic-interaction
`chromatography (HIC; UNIT 8.4), affinity chromatography (Chapter 9), and ion-exchange
`chromatography (UNIT 8.2).
`
`The last phase of a purification scheme, the polishing step, also typically involves chro-
`matography. Every contaminant that has been purified along with the product to this
`point is very closely related to the product, and very high selectivity and efficiency are
`needed to separate the desired product from these closely related structural variants.
`The most useful and commonly employed techniques are gel-filtration chromatography
`(UNIT 8.3), hydrophobic-interaction chromatography (UNIT 8.4), ion-exchange chromatog-
`raphy (UNIT 8.2), and reversed-phase high-performance liquid chromatography (UNIT 8.7).
`More thorough discussions of the design of a separation protocol can be found in Sofer
`and Hagel (1997) and in other units in this chapter.
`
`The capture phase of the separation can be made more efficient if the separation power of
`chromatography can be combined with an operation that permits the use of particulate-
`laden (i.e., nonclarified) feed. In recent years, expanded beds have been successfully used
`to combine clarification and concentration. The expanded bed also accomplishes an initial
`purification of the product. The references cited in the following three reviews provide
`a useful summary of these applications: Chase (1994); Hjorth (1997); and Shiloach and
`Kennedy (2000). In addition, two special issues of the journal Bioseparation (Volume 8,
`Iss. 1-5, 1999; Volume 10, Iss. 1-3, 2001) are dedicated to expanded-bed adsorption
`chromatography.
`
`Although expanded-bed adsorption (EBA) is a scalable technique, this unit focuses on
`EBA at the bench scale. EBA is useful in three protein purification/capture scenarios.
`The first is the recovery of molecules from cell cultures in which the recombinant cell
`line is very shear-sensitive. In these cases, clarification of the cell harvest by membrane
`filtration and/or centrifugation causes product degradation, because the cell lysis that
`results from these techniques leads to the release of proteases into the harvest broth. This
`scenario is common in the use of engineered CHO cell lines. The second scenario applies
`to situations in which the recombinant protein is attached to or strongly associated with
`
`Contributed by Robert M. Kennedy
`Current Protocols in Protein Science (2005) 8.8.1-8.8.25
`Copyright C(cid:1) 2005 by John Wiley & Sons, Inc.
`
`Conventional
`Chromatographic
`Separations
`
`8.8.1
`
`Supplement 40
`
`KASHIV EXHIBIT 1033
`IPR2019-00797
`
`Page 1
`
`

`

`the cell membrane. Due to the high charge density of the resin, the resin will out-compete
`the cell for the protein, yielding very high recovery rates. (The chromatography resin
`has a charge density on the order of mmol/ml, which is very high compared with the
`charge density associated with the cell surface or cell surface debris; charged molecules
`are drawn to this higher charge density, similar to the way in which a stronger battery
`will make a bulb glow brighter than a weaker battery will.) This is the case in certain
`Pichia expression systems. The third scenario in which EBA is particularly useful is
`when exposure of the target protein to proteases must be avoided. In this scenario, the
`usual strategy is to work in the cold, as is the case in the recovery of some enzymes from
`natural sources.
`
`STRATEGIC PLANNING
`The experimental strategy for the development of an expanded-bed process involves four
`major steps: (1) method scouting, (2) method optimization, (3) process verification, and
`(4) production. Most investigators can accomplish the first two steps and be satisfied that
`they have a reproducible method. Industrial applications typically require completion of
`all four steps if the process is to be scaled up for use in a production environment.
`
`The purpose of method scouting is to define the most suitable adsorbent and the optimal
`conditions for binding and elution. Method scouting is performed in both packed-bed
`mode (with a clarified feedstock sample) and expanded-bed mode. The alternate protocols
`in this unit describe initial method scouting procedures—namely, determining conditions
`for ion-exchange chromatography, HIC, and affinity chromatography. Employing these
`protocols using a clarified feed will only provide an approximate starting condition,
`however. Development of a method using only clarified feed does not take into account
`the charge carried by the biomass in the crude feedstock. This can lead to erroneous
`results, poor fluidization, and lowered capacity.
`
`Method scouting with clarified feed
`The adsorbent is selected using the same principles that apply to packed-bed chromatog-
`raphy. The medium of choice will exhibit the strongest binding to the target protein while
`retaining only a minimum amount of contaminating proteins. To establish the chemical
`parameters of the separation, initial method scouting runs should be performed using feed
`that has been clarified. Later steps will be performed with unclarified feed to establish
`the physical parameters of the separation. It is critical to keep the characteristics of the
`feedstock in mind when designing a process.
`
`For method scouting, the bed height is typically 15 cm in a 2.5- or 2.6-cm-diameter
`column. This bed height allows the target protein sufficient time to bind at normal flow
`velocities of ∼300 cm/hr. The flow velocity used should be similar to what will subse-
`quently be used in expanded-bed mode. It should be noted that the use of flow velocity (in
`cm/hr) instead of flow rate (in ml/min) throughout this unit is intentional, as this allows
`the flow to be normalized for any column diameter. The first step in the conversion from
`cm/hr to ml/min is to multiply the cross-sectional area of the column being used (πr2,
`where r is the column radius in cm) by the specified velocity in cm/hr. This gives a result
`in units of cm3/hr, which is equivalent to ml/hr. From there, the result can be divided by
`60 to give the desired flow rate in units of ml/min. For a given flow velocity, actual flow
`rates will increase as a function of r2. Knowledge of flow velocities is of practical utility
`when comparing runs between different columns, or when scaling a successful run to a
`column of larger diameter.
`
`Elution can be performed in stepwise fashion or by applying a gradient. Information
`from elution studies is used to optimize selectivity for the target molecule and to avoid
`the binding of contaminants that bind less avidly to the resin.
`
`Current Protocols in Protein Science
`
`Expanded Bed
`Adsorption
`Chromatography
`
`8.8.2
`
`Supplement 40
`
`Page 2
`
`

`

`Figure 8.8.1 Breakthrough curve for BSA on the STREAMLINE DEAE resin as determined by
`frontal analysis, in which the sample (BSA) is continuously applied to the column and C/Co, the
`ratio of BSA concentration in the flowthrough (C) toBSA concentration in the feedstock (C o), is
`measured as a function of the amount of BSA applied. The resin becomes saturated (i.e., C/C0
`approaches 1) as more and more BSA is applied to the column. As can be seen from the graph,
`the chromatographic behavior of the resin in expanded-bed mode is the same as in packed-bed
`format. Reproduced with permission from Amersham Biosciences (Expanded-Bed Adsorption:
`Principles and Methods; see Internet Resources) and Academic Press (Shiloach and Kennedy,
`2000).
`
`Once selectivity has been optimized, maximum dynamic binding capacity is deter-
`mined from breakthrough tests conducted using the chosen binding conditions. Figure
`8.8.1 shows breakthrough capacity curves in packed and expanded modes and in two
`STREAMLINE columns (Amersham Biosciences), representing laboratory scale and
`pilot scale. Note the similar shape of the breakthrough curves for packed-bed mode
`and expanded-bed mode, as well as for the different-sized STREAMLINE columns
`(STREAMLINE 50 versus STREAMLINE 200; numeric specification refers to the inner
`diameter of the column in mm). Thus, adsorption performance is the same in expanded-
`bed and packed-bed modes, and STREAMLINE columns can be scaled up while main-
`taining the same performance.
`
`When defining a suitable maximum loading capacity for EBA chromatography, it is
`necessary to apply a safety margin to compensate for sources of variability in the sample
`that could affect binding capacity. The maximum loading amount should be ∼75% of
`breakthrough capacity.
`
`Method scouting with feedstock containing biomass
`A second method scouting run should be performed starting with unclarified biomass. Two
`simple experiments that require no specialized equipment can be conducted to measure the
`interaction of the biomass with the resin (Kula et al., 2001; A. Liten and K. Kinealy, unpub.
`observ.). The first involves simply comparing the amount of biomass present in the sam-
`ple before and after passage through the column by measuring light scattering at 600 nm
`(Fig. 8.8.2). This experiment can be done with flowthrough monitors and peak in-
`tegration, but it can also be accomplished off-line using cuvettes and a spectropho-
`tometer. (The reader is directed to UNIT 5.3, for instructions on measuring biomass
`with a spectrophotometer; briefly, for E. coli, anoptical density reading of 0.5 at
`
`Current Protocols in Protein Science
`
`Conventional
`Chromatographic
`Separations
`
`8.8.3
`
`Supplement 40
`
`Page 3
`
`

`

`Figure 8.8.2 Schematic diagram of an experiment measuring gross interaction between the
`expanded-bed resin and the biomass in a feedstock mixture. A pulse of biomass is injected, and
`its light-scattering profile is measured at 600 nm before and after passage through the column.
`The resulting peaks are then integrated, with a change in peak area after passage of the biomass
`through the column indicating that the sample has interacted with the resin. Reproduced with
`permission from A. Liten.
`
`600 nm will correspond to a concentration of ∼5 × 108 cells/ml.) The relevant cal-
`culation is a simple mass balance of biomass before and after passage through the resin,
`and the experiment should be followed by inspection of the resin for aggregation. Resin
`aggregation can be seen with the naked eye and is mostly due to interaction of the resin
`with the biomass. If aggregation has occurred, clumps of resin, rather than a uniform bed,
`will be evident. If aggregation of the resin is seen and appears to affect bed expansion
`or loading capacity for the target molecule, then dilution of the biomass, changes in salt
`concentration, or the use of a different salt may be necessary. If aggregation persists, and
`loading capacity and target recovery are unacceptably low, it may be the case that EBA is
`not an appropriate technique for harvesting the target molecule from the feedstock being
`used. In general, choose conditions that result in the lowest degree of sample interaction
`with the resin and in little or no aggregation.
`
`The second experiment involves estimating the equilibrium ratio of target protein to resin
`(see Support Protocol 1). The purpose of this procedure is to determine the maximum
`amount of sample that should be loaded onto a column (usually 30% of the equilibrium
`target protein binding capacity).
`
`The most important considerations in initial EBA method scouting are 1) to never choose
`conditions under which the resin is observed to aggregate; 2) to select conditions under
`which there is minimal interaction between the biomass and the resin, but under which
`the target protein still binds; and 3) to set the maximum amount of sample that can be
`loaded onto the column at 30% of the equilibrium capacity for the chosen conditions.
`
`After method scouting, the remaining three steps in the development of an expanded-bed
`process are performed in expanded-bed mode.
`
`EBA method optimization
`The next step in the development of an expanded-bed process is optimization. The
`expanded-bed method is first optimized on a small scale (e.g., a 2.5-cm column) us-
`ing crude, unclarified feed. The purpose of this optimization is to examine the effect of
`
`Current Protocols in Protein Science
`
`Expanded Bed
`Adsorption
`Chromatography
`
`8.8.4
`
`Supplement 40
`
`Page 4
`
`

`

`the crude feed on the stability of the expanded bed and on chromatographic performance
`and to make adjustments to the flow velocity if necessary.
`
`During method optimization, the process should be challenged by applying a sample load
`close to the maximum sample load as defined in the breakthrough study performed on the
`packed bed. In addition, a breakthrough capacity study should be performed to determine
`the effect of the crude feed on the binding capacity of the resin.
`
`Elution, wash, and re-equilibration volumes should also be determined as a part of method
`optimization, and finally, a suitable cleaning-in-place (CIP) procedure should be chosen.
`(Development of the CIP method should be initiated in the method scouting step.) CIP
`procedures for expanded-bed media are complicated by the fact that the debris load can be
`challenging to clean up. That is, unlike chromatographic media, expanded-bed media are
`typically exposed to a significant amount of contamination and particulates. CIP protocols
`for expanded-bed media are usually straightforward, but their identification requires an
`investment of time and effort. An expanded bed can be characterized by a variety of tracer
`techniques that can be used to monitor bed performance after CIP. See Batt et al. (1995)
`for a description of bed characterization.
`
`Process verification
`Process verification, or pilot-scale work, is done using a pilot-scale column (e.g., a 20-cm
`column). For some applications, this column may also be suitable for the final production
`scale. The usual bed height of 15 cm provides a sedimented bed volume of 4.7 liters
`in a 20-cm column. Scale-up is performed by increasing the column diameter while
`maintaining the sedimented bed height, flow velocity, and expanded bed height.
`
`STREAMLINE EBA columns
`Two main features characterize a STREAMLINE column. First, the design of the liquid
`distribution system at the base of the column ensures the formation of a stable expanded
`bed. Second, a movable adaptor allows the column height to be altered during the different
`stages of an expanded-bed adsorption cycle.
`
`three stages of process
`There are STREAMLINE columns for each of the last
`development—i.e., laboratory-scale, pilot-scale, and production-scale columns. These
`three stages of process development are also phases in the development of an optimized
`STREAMLINE expanded-bed unit operation. For the majority of applications, method
`optimization is done on the laboratory scale using STREAMLINE 25 or STREAMLINE
`50 columns, process verification is done on the pilot scale using STREAMLINE 100 or
`STREAMLINE 200 columns, and finally, manufacture is carried out using production-
`scale columns that range in size from 400 to 1200 mm (inner diameter).
`
`This unit focuses on laboratory-scale EBA. The appropriate column for laboratory-scale
`EBA is the 25-mm-i.d. column, STREAMLINE 25. The STREAMLINE 50 column (i.d.,
`50 mm) can be used for laboratory-scale work as well, but it is top-heavy and not as user-
`friendly as the STREAMLINE 25 column. The STREAMLINE 25 column is available
`in two formats: manual and hydraulic. The equipment setup diagrams in this unit are for
`the hydraulic version. The manual version is set up in exactly the same manner, except
`that there is no need for the hydraulic pump or the hydraulic valve after the hydraulic
`pump. If only one peristaltic pump is available, the manual setup is more appropriate.
`
`Resins
`Gel media for EBA are commercially available (Table 8.1.1). The fluidization properties of
`several types of media used in protein recovery are reported by Thommes (1997). Of these,
`
`Current Protocols in Protein Science
`
`Conventional
`Chromatographic
`Separations
`
`8.8.5
`
`Supplement 40
`
`Page 5
`
`

`

`STREAMLINE media are the most widely used. Commercially available resins include
`STREAMLINE SP, STREAMLINE DEAE, STREAMLINE SP XL, STREAMLINE Q
`XL, STREAMLINE rProtein A, STREAMLINE chelating resin, and STREAMLINE
`heparin resin (Amersham Biosciences); CM Hyper Z resin (Ciphergen); and Prosep A
`and Prosep G resins (Millipore).
`
`STREAMLINE adsorbents exhibit Gaussian-like distributions with respect to particle size
`and particle density. This particle polydispersity is an important factor that contributes to
`the stability of the expanded bed. The size and density gradients present cause the beads
`to be positioned at specific heights in the expanded bed depending on the sedimentation
`velocity of the individual adsorbent particles. The mean particle size of STREAMLINE
`beads is 200 µm, compared with ∼90 µm for packed-bed chromatography media. This
`larger diameter results from modifications made to increase particle density, which is
`∼1.2 g/ml. In the bead population, smaller, lighter particles move to positions near the
`top of the expanded bed, while larger, heavier particles move to the bottom, resulting
`in a stable, uniform expansion. In other words, the beads find their ideal position in
`the column, which explains the low degree of axial dispersion in EBA. The sizes and
`densities of STREAMLINE adsorbents have been defined to allow optimal expansion at
`flow velocities up to 500 cm/hr.
`
`STREAMLINE Q XL is an anion-exchange resin based on highly cross-linked 6% agarose
`modified by the inclusion of an inert quartz core to produce the desired density. Long
`molecules of dextran are coupled to the agarose matrix, and ion-exchange groups are
`then attached to these dextran chains. This construction increases the effective interaction
`volume as well as the availability of ligands for adsorption of the target. The overall result
`is a large increase in binding capacity. The binding capacity of the STREAMLINE Q XL
`resin is >110 mg BSA per ml adsorbent, compared with 40 mg/ml for the normal DEAE
`Fast Flow ion exchanger (Amersham Biosciences).
`
`STREAMLINE SP XL is a cation exchanger that is also based on highly cross-linked
`6% agarose modified by the inclusion of an inert quartz core to produce the desired
`density. As in the STREAMLINE Q XL resin, long molecules of dextran are coupled to
`the agarose matrix, and ion-exchange groups are then attached to these dextran chains.
`The binding capacity of the STREAMLINE SP XL resin is >140 mg lysozyme per
`ml adsorbent, compared with 60 mg/ml for the normal SP Fast Flow ion exchanger
`(Amersham Biosciences).
`
`The degree of expansion (calculated as the ratio of expanded bed height to sedimented
`bed height) for both of the above media is 2 to 3 at a flow rate of 400 cm/hr. The degree
`of expansion is a quick and useful measure of bed stability, although it is a less accurate
`measure compared with the number of theoretical plates (UNIT 8.1; also see Basic Protocol,
`steps 20 to 23).
`
`STREAMLINE rProtein A is an affinity adsorbent for purifying monoclonal and poly-
`clonal antibodies. It is based on highly cross-linked 4% agarose modified by the inclusion
`of an inert metal alloy core to provide the desired density. The binding capacity of the
`STREAMLINE rProtein A resin is ∼50 mg human IgG per ml of gel, and the degree of
`expansion is between 2.2 and 3 at 300 cm/hr. The ligand used in this medium is recom-
`binant protein A, which has been engineered to include a C-terminal cysteine residue.
`The epoxy chemistry is controlled to favor a thioether coupling, providing single-point
`attachment of the protein A. This oriented coupling enhances the binding of IgG. The re-
`combinant protein A has no mitogenic activity in human lymphocytes in vitro. (Mitogenic
`activity has sometimes been a concern when using native protein A from Staphylococcus
`aureus.) Leakage of protein A from the STREAMLINE rProtein A resin is generally
`low. In pharmaceutical production processes, protein A must be removed from the final
`
`Current Protocols in Protein Science
`
`Expanded Bed
`Adsorption
`Chromatography
`
`8.8.6
`
`Supplement 40
`
`Page 6
`
`

`

`product; leached recombinant protein A can be removed efficiently from an IgG-
`containing fraction by gel filtration or ion-exchange chromatography.
`
`STREAMLINE chelating resin is used for immobilized metal affinity chromatography
`(IMAC; UNIT 9.4). This technique exploits the different affinities of proteins for metal ions
`bound to chelating groups on the adsorbent. The chelating support can be primed with
`metal ions by passing a salt solution through the column. The chelating group consists
`of iminodiacetic moieties on a spacer arm coupled to the matrix via stable covalent
`linkages.
`
`STREAMLINE heparin resin consists of heparin immobilized to the matrix. This adsor-
`bent can be used in two modes: (1) as an affinity matrix for coagulation factors, or (2) as
`a high-capacity cation exchanger, due to the large number of anionic sulfate groups on
`the ligand. For example, when interacting with nucleic acid binding proteins, the ligand
`mimics a nucleic acid in terms of its polyanionic structure.
`
`STREAMLINE phenyl resin consists of phenyl groups immobilized to the matrix. This
`adsorbent is used for hydrophobic-interaction chromatography, or HIC (UNIT 8.4), a tech-
`nique that separates proteins and peptides based on the strength of their hydrophobic
`interactions with hydrophobic groups attached to an uncharged matrix. It is estimated
`that as much as 30% to 40% of the accessible surface area of water-soluble proteins is
`nonpolar, which explains why HIC is such a widely applicable technique.
`
`STREAMLINE chelating resin, STREAMLINE heparin resin, and STREAMLINE
`phenyl resin, along with the DEAE and SP ion exchangers mentioned above, have average
`densities of 1.2 g/ml and expand to 2.5 times their original bed height in water at a flow
`rate of 300 cm/hr.
`
`CM Hyper Z is a weak cation exchanger in which ion exchange is mediated by car-
`boxymethyl groups; the base material is zirconium oxide. The resin has a dynamic bind-
`ing capacity of 30 mg IgG per ml at a flow rate of 300 cm/hr and is designed to function in
`a high-salt environment (15 to 20 millisiemens per cm), which is typical of fermentation
`conditions.
`
`Prosep A and Prosep G resins contain immobilized recombinant protein A and immobi-
`lized recombinant protein G, respectively. Both have very high densities, because their
`base material is 75- to 125-µm controlled-pore glass. These resins have a dynamic ca-
`pacity of 25 mg IgG per ml at a flow rate of 750 cm/hr, and they support flow velocities
`of up to 1500 cm/hr for regeneration and equilibration.
`
`EXPANDED-BED ADSORPTION CHROMATOGRAPHY
`This protocol covers the preparation of the EBA resin and the setup of the apparatus,
`as well as the typical steps in a chromatography experiment (equilibration of the resin,
`evaluation of the column, application of the sample, and elution of the target molecule).
`Cleaning and storage are described at the end of the protocol. The selection of buffers for
`equilibration and elution will be guided by the mode of chromatography being performed.
`Typically, the standard buffer systems for bimolecular purification are used.
`
`As mentioned before, the expanded-bed column can be operated manually or by a hy-
`draulic pump. The figures related to this protocol show the hydraulic system, because
`it is the more complex of the two. In order to use the hydraulic system, two peristaltic
`pumps that can be dedicated to expanded-bed chromatography are needed. If only one
`peristaltic pump is available, then it will be necessary to use the manual system. Aside
`from making it possible to move the column flow adaptor hydraulically, the hydraulic
`
`Current Protocols in Protein Science
`
`BASIC
`PROTOCOL
`
`Conventional
`Chromatographic
`Separations
`
`8.8.7
`
`Supplement 40
`
`Page 7
`
`

`

`system is no different from the manual system. Wherever necessary, separate instructions
`are given for manual and hydraulic columns.
`For methods development,∼1 liter of feedstock is needed. Once the method is established,
`the volume of feedstock used will be determined by the amount of time required for feed
`application and washout.
`
`The protocol below is applicable for all of the resin types that can be used in EBA (see
`Alternate Protocol 1, Alternate Protocol 2, and Alternate Protocol 3).
`
`Materials
`Chromatography resin (see Strategic Planning)
`Start buffer
`Equilibration buffer
`0.25% (v/v) acetone in equilibration buffer
`Feedstock under constant stirring (e.g., by a magnetic stir plate or overhead mixer)
`Wash buffer
`Elution buffer
`Cleaning-in-place (CIP) solution
`20% (v/v) ethanol (optional)
`Sintered glass funnel (optional)
`Water aspirator (optional)
`Chromatography column (see Strategic Planning) with or without hydraulic pump
`Chromatography apparatus (see Fig. 8.8.3, Fig. 8.8.4, Fig. 8.8.6, Fig. 8.8.7, and
`Fig. 8.8.8 for various schematics) consisting of the following:
`Peristaltic pump(s) (e.g., Cole-Parmer Masterflex tubing pump; do not use
`peristaltic pumps designed for laboratory chromatography, such as P1,
`Microperpex, or Rabbit pumps, as these pumps do not produce sufficiently
`high flow rates) attached to Masterflex size 16 silicon pump tubing
`Two-channel, four-port valve for flow reversal
`Single-channel, three- or four-port valves
`Pressure gauge
`Flowthrough UV monitor with S2-type or 6-mm-i.d. flow cell (do not use flow
`cells from analytical instruments, since cell debris may clog them)
`Chart recorder
`Flowthrough conductivity meter (optional)
`Flowthrough pH meter (optional)
`Wash bottle
`Spirit level
`Label tape
`Additional reagents and equipment for CIP (see Support Protocol 2)
`
`NOTE: See Expanded Bed Adsorption: Principles and Methods (Amersham Biosciences;
`see Internet Resources) for additional details regarding setup of the chromatography
`apparatus.
`
`Prepare the resin for use
`NOTE: Since EBA resins are typically a mixture of particles of various sizes and densities,
`they, unlike other chromatographic resins, cannot simply be poured into a column. Instead,
`most EBA resins must be specially prepared to ensure an appropriate mixture of particle
`sizes and densities in the column. All particles in Hyper Z and Prosep media, however,
`have the same density, and so there is no need to perform the sampling procedure described
`in steps 2 to 6 if either of these media is being used; instead, simply pour the slurry supplied
`by the manufacturer to the desired bed height in step 8.
`
`Current Protocols in Protein Science
`
`Expanded Bed
`Adsorption
`Chromatography
`
`8.8.8
`
`Supplement 40
`
`Page 8
`
`

`

`1. Shake the resin container until the resin is completely slurried.
`
`2. Immediately pour the entire contents of the container into a sintered glass funnel or
`(if the volume of the resin container is larger than is convenient for a glass funnel)
`into a beaker or bucket, and allow the resin to settle completely.
`
`3a. If using a funnel: Allow the liquid component of the resin slurry to drain through the
`funnel.
`
`3b. If using a beaker or bucket: Remove the liquid from the top of the resin cake by
`aspiration through a hose connected to a water aspirator, taking care not to aspirate
`the resin along with the liquid.
`
`4. Using the following formula, calculate the volume of resin needed to obtain the
`desired sedimented bed height: V = πr2h, where V is the volume of resin needed, r
`is the column radius, and h is the desired sedimented bed height.
`If column radius and bed height are entered into the formula in units of cm, then the
`calculated volume will be in units of cm3, or ml.
`The column for initial method scouting should have a sedimented bed height of ∼15 cm.
`5. Calculate the amount (in grams) of resin needed to obtain the desired bed height by
`multiplying the volume calculated in step 4 by the average density of the resin (in
`g/ml).
`The average density of the rProtein A STREAMLINE resin is 1.3 g/ml. The average density
`of all other STREAMLINE media is 1.2 g/ml.
`
`6. Cut a wedge-shaped slice of the settled resin from the funnel, beaker, or bucket, and
`weigh it. Trim the slice or add more resin (also cut in wedges from the resin cake
`in the funnel, beaker, or bucket) as needed to obtain the desired amount of resin (as
`determined in step 5). Transfer the desired amount of resin to a beaker, and suspend
`in start buffer to obtain an ∼50% (w/v) slurry.
`Cutting wedge-shaped pieces ensures that a representative sample (in terms of particle
`sizes and densities) is obtained from the resin cake.
`
`The exact volume of start buffer used to make the slurry is not critical. The purpose of
`making a slurry in starting buffer is simply to make it easier to pour the resin into the
`column.
`
`Set up the column for use
`7. Connect the column to the chromatography apparatus. Set each valve as indicated in
`Figure 8.8.3. Remove the lid and adaptor from the column, and then use pump 1 to
`fill the column to approximately two-thirds of its capacity with start buffer via the
`bottom valve (valve 4).
`For columns with i.d. >25 mm, it may be necessary to remove air bubbles trapped under
`the column’s bottom net; this is done by drawing air out of the net using tubing connected
`to a water aspirator.
`
`8. Make sure that the resin/start buffer slurry is uniform and, without letting the resin
`settle, quickly pour the slurry into the column. Use additional start buffer to rinse the
`sides of the container from which the slurry was poured, and pour this rinse liquid
`into the column as well. Use a wash bottle filled with start buffer to dislodge any
`resin stuck to the sides of the column.
`
`9. After the resin has been poured into the column, add start buffer until the column is
`full to capacity.
`At this point, a tray or a tote can be placed under the column to catch any buffer that
`might spill out in subsequent steps.
`
`Current Protocols in Protein Science
`
`Conventional
`Chromatographic
`Separations
`
`8.8.9
`
`Supplement 40
`
`Page 9
`
`

`

`Figure 8.8.3 Schematic diagram showing the tubing connections, valve placement, pump config-
`uration, and monitoring options for expanded-bed adsorption chromatography. Valves that control
`column operation are labeled V1, V2, V3, and V4. This diagram shows a hydraulic column; adaptor
`movement is controlled by pump 2. In a manual column, pump 2 and V3 are not present, and the
`column lid does not have a connector to the hydraulic line, but all other connections are the same.
`Buffer 1 and buffer 2 are typically equilibration buffer and elution buffer, respectively. P1 and P2 are
`fraction collection lines. CIP, cleaning-in-place solution (see Support Protocol 2); UV, flowthrough
`UV detector; cond., flowthrough conductivity meter; pH, flow-through pH meter. Reproduced with
`permission from Amersham Biosciences (Expanded-Bed Adsorption: Principles and Methods; see
`Internet Resources).
`
`angle with respect to the
`10. Tip the adaptor such that its bottom makes at least a 45
`surface of the buffer in the column, and then insert it into the column under the
`buffer, taking care not to introduce air bubbles. As soon as the entire adaptor O-ring
`is just inside the column, straighten the adaptor so that its bottom is parallel to the
`surface of the buffer, and then put the lid on the column and secure it with the screws
`provided by the manufacturer.
`
`◦
`
`IMPORTANT NOTE: No resistance should be felt when inserting the adaptor into the
`column. If resistance is