Detachment of affinity-captured bioparticles by elastic deformation of a macroporous hydrogel

  1. Maria B. Dainiak*,†,
  2. Ashok Kumar*,†,‡,
  3. Igor Yu. Galaev*, and
  4. Bo Mattiasson*,§
  1. *Department of Biotechnology, Center for Chemistry and Chemical Engineering, Lund University, P.O. Box 124, SE-221 00 Lund, Sweden; †Protista Biotechnology AB, Ideon, SE-223 70 Lund, Sweden; and ‡Department of Biological Sciences and Bioengineering, Indian Institute of Technology, Kanpur-208016, India

  1. Edited by Alexander M. Klibanov, Massachusetts Institute of Technology, Cambridge, MA, and approved December 5, 2005 (received for review October 6, 2005)

 
Next Section

Abstract

Adsorption of bioparticles to affinity surfaces involves polyvalent interactions, complicating greatly the recovery of the adsorbed material. A unique system for the efficient binding and release of different cells and particles is described. Affinity-bound bioparticles and synthetic particles are detached from the macroporous hydrogel matrix, a so-called cryogel, when the cryogel undergoes elastic deformation. The particle detachment upon elastic deformation is believed to be due to breaking of many of the multipoint attachments between the particles and the affinity matrix and the change in the distance between affinity ligands when the matrix is deformed. However, no release of affinity-bound protein occurred upon elastic deformation. The phenomenon of particle detachment upon elastic deformation is believed to be of a generic nature, because it was demonstrated for a variety of bioparticles of different sizes and for synthetic particles, for different ligand–receptor pairs (IgG–protein A, sugar–ConA, metal ion–chelating ligand), and when the deformation was caused by either external forces (mechanical deformation) or internal forces (the shrinkage of thermosensitive, macroporous hydrogel upon an increase in temperature). The elasticity of cryogel monoliths ensures high recovery of captured cells under mild conditions, with highly retained viability. This property, along with their continuous porous structure makes cryogel monoliths very attractive for applications in affinity cell separation.

The interactions between bioparticles (bacteria, organelles, cells, viruses, and inclusion bodies) and liquid–solid interfaces occurring in many biological systems and the adsorption of cells to surfaces in affinity-based separation have a polyvalent nature. Polyvalent interactions are characterized by the simultaneous binding of multiple receptors on the surface of bioparticles to multiple ligands on another surface and can be collectively much stronger than corresponding monovalent interactions (1).

The difficulty in disrupting multivalent interactions is one of the main problems in designing affinity techniques for cell separation. Theoretical studies have shown that, for situations where the number of interactions is >10, it is unlikely that reasonable concentrations of a soluble monovalent competitor (i.e., a biospecific eluent) can displace the binding equilibrium (2). Under typical chromatographic conditions (1010 to 1012 ligands and receptors per cm2 and 10–10 to 10–8 cm2 contact area), the number of specific binding interactions can be between 1 and 10,000 (3). Thus, in most cases, an external force affecting the entire bioparticle or the matrix in an integral way is required to simultaneously disrupt multiple bonds and detach specifically adsorbed bioparticles (4). Leading to cooperative effects, the retention of cells on affinity surfaces is correlated to the affinity of individual receptor–ligand binding (5) and the concentration of surface ligands (6) and is sensitive to the presence of competitive binding inhibitors (7). In approaches currently used for cell release, the detachment forces are generated by the passage of air–liquid interfaces (8, 3) or by using flow-induced shear forces (9, 10). The latter leads to a high degree of dilution of the eluted cells and involves the risk of cell damage.

Because of the high heterogeneity of the cell surface, there may be other factors (e.g., hydrophobic and electrostatic interactions and van der Waals attraction), along with affinity interactions, controlling cell behavior at cell–surface interfaces. Until recently, studies of cell adhesion have been focused mainly on cellular response to surface chemistry (11–13) and topography (14, 15) and on microbial adherence to stiff supports, such as polystyrene, Teflon, and glass (9, 16–18). In biological systems, cells often come into contact with soft surfaces, e.g., tissue or an extracellular matrix, which can undergo changes in elasticity (e.g., wound healing). In vivo, the primary step of any bacterial or virus infection is the adsorption of the pathogen onto an elastic, hydrophilic matrix, namely the tissue of a living organism (19). Only recently have systematic studies of the effect of substrate mechanics on cell adhesion been carried out, and softness and elasticity of the surface were shown to be important for cell–surface interactions (20–22). Investigations of the relationship between different types of cells and the elasticity of polyacrylamide- and alginate-based surfaces have revealed a reduction of cell spreading and weakening of cell–surface interactions after a decrease in substrate stiffness (20, 23, 24). These trends are independent of the adhesive ligand. An important implication of such a finding is that the use of soft materials in cell-affinity separation may help to avoid or to decrease nonspecific cell–surface interactions.

In this work, the results of studies on the interaction of bioparticles and their detachment from elastic affinity cryogel monoliths are reported. Monolithic macroporous polyacrylamide gels, so-called cryogels, have recently been developed for applications in bioseparation (25–28) and are characterized by high porosity and elasticity. Because of the size (10–100 μm) and interconnected structure of the pores and the absence of nonspecific interactions with the adsorbent, cells pass freely through native cryogels without affinity ligands (28). Unlike traditional polyacrylamide gels, which are rather brittle, polyacrylamide-based cryogels are elastic, soft, sponge-like materials that can withstand large deformations and can be easily compressed 4- to 6-fold without being mechanically damaged. The compressed monoliths reswell and adopt their initial shape upon the addition of more liquid (29).

Previous SectionNext Section

Results and Discussion

In contrast to rigid, hydrophobic materials traditionally studied with respect to cell separation, biological tissues are soft, elastic, and hydrophilic. Thanks to their elasticity, biological tissues are subject to pronounced deformation (e.g., muscle contraction–relaxation). To the best of our knowledge, the effect of the deformation of a soft matrix on the retention of specifically bound cells has not been addressed in detail.

Cryogels represent a hydrophilic, highly porous (Fig. 1), and highly elastic matrix providing an ample solid–liquid interface for specific attachment of bioparticles. Affinity cryogels were chosen as a model system to study the effect of mechanical deformation of the matrix on the retention of specifically bound bioparticles (bacterial, mammalian, and yeast cells and antibody-labeled inclusion bodies) and synthetic particles (microgels of cross-linked poly(N-isopropyl acrylamide-co-N-vinylimidazole). The characteristics of the analyzed particles and corresponding affinity cryogel monoliths are presented in Table 1.

Fig. 1.
View larger version:
Fig. 1.

Scanning electron micrograph of cryogel monolith.


View this table:
Table 1. Particles and corresponding affinity cryogel monoliths studied

To estimate the strength of attachment between different types of target particles and affinity cryogel monoliths and its effect on the release of the adsorbed material, suspensions of yeast cells and microgel particles were incubated for different periods of time within ConA– and Cu(II)–IDA–cryogel monoliths (inserted into the column 20 × 7.0 mm i.d.), respectively. Fresh affinity cryogel monoliths were used in each test. Synthetic microgels (≈0.4 μm in diameter) and yeast cells (≈8 μm in diameter) differ significantly from one another in size and physicochemical surface properties and were chosen as two model systems for comparison. After the binding step, affinity cryogel monoliths were washed with 12 column volumes of the corresponding running buffers at a flow rate of 21 cm/h to remove unbound particles. Incubation was required with the ConA–cryogel monolith for efficient binding of yeast cells, whereas the amount of microgel particles bound to the Cu(II)–IDA–cryogel monolith was independent of the time of contact (Table 2). An increase in the amount of yeast cells or microgel particles applied to ConA– or Cu(II)–IDA–cryogel monoliths did not lead to an increase in binding; the excess of applied particles was found in the flow-through (data not shown).

View this table:
Table 2. Release of affinity-bound particles by flow-induced detachment shear force followed by mechanical compression of cryogel monoliths

Two different strategies were used for the release of bound particles: application of shear forces by passing pulses of buffer and a corresponding eluent through the column at a velocity of 430 cm/h (flow-induced detachment) and by mechanical compression of cryogel monoliths (Fig. 2A). The column was disconnected from the pump before the compression. The duration of the pulses of buffer and eluent during flow-induced detachment was optimized, and the next detachment step was applied when no more cells could be recovered by the previous step.

Fig. 2.
View larger version:
Fig. 2.

Schematic illustration of the procedure used for the release of bound particles by mechanical compression of affinity cryogel monoliths (A) and of the mechanism of detachment of bound particles induced by compression (B).


The duration of contact between yeast cells and the ConA–cryogel monolith preceding the washing step had a pronounced effect on the strength of interaction and, as a result, on the efficiency of flow-induced detachment (Table 2). In the test performed without the incubation step, 29% and 36% of the bound yeast cells were detached by applied flow, without and in the presence of 0.3 M α-d-manno-pyranoside, respectively. After 30 min incubation in the affinity adsorbent, only 9% of bound cells were released by the first pulse and 9% by applied shear in the presence of specific eluent. These results show that there are several fractions of bound yeast cells characterized by different binding strengths to the affinity matrix. The fraction of strongly bound cells that cannot be detached by flow increases with increasing contact time between the cells and affinity adsorbent. A time-dependent effect on the strength of interaction between yeast cells and a ConA-coated surface has also been reported (7). The stabilization phenomenon has been explained by the possibility for further formation of bonds in a time-dependent manner after the initial anchoring of the cell by only a few affinity ligands on the surface (30). It has been observed that there is a distribution of attachment strengths because of the distribution of the number of bonds formed between a population of cells and an affinity surface (31). However, the compression of ConA–cryogel monoliths containing the fraction of yeast cells that was not released by shear forces, even in the presence of specific eluent, resulted in total recovery of that fraction of cells (Table 2).

No more than 12% of microgel particles captured by Cu(II)–IDA–cryogel monoliths, with or without the preincubation step, were removed from the column with a pulse of 0.3 M imidazole applied at a flow velocity of 430 cm/h (Table 2). About 60% of the captured particles were detached by compressing the Cu(II)–IDA–cryogel monolith.

Presumably, the mechanical compression of cryogel monoliths results in the disruption of multiple bonds between bound particles and the surface of affinity cryogel and the recovery of the detached particles in the liquid expressed (Fig. 2B). It has been demonstrated by scanning electron-microscopy studies that cells captured by affinity cryogel monoliths are bound to the plain “flat” parts of the pore walls and are not entrapped in “dead flow” zones (25, 32). Possible reasons for the disruption of affinity bonds may be the deformation of the surface and/or the damage to affinity ligands caused by mechanical compression. The possibility of ligand damage as the result of cryogel compression was studied in the adsorption tests by using affinity cryogel monoliths regenerated after the squeezing procedure. In cycle I, bound particles were released by compression in the presence of the corresponding eluent. Flow-induced detachment was not carried out before compression. After the detachment step, the column was regenerated by washing with 12 column volumes of the corresponding running buffer, and the cycle was repeated. The compression did not have a pronounced effect on the binding properties of ConA– and Cu(II)–IDA–cryogel monoliths or on the detachment efficiency in cycle II, indicating that it is unlikely that the ligands were damaged in cycle I. Thus, the main mechanisms involved in the compression-induced detachment of bound particles from the surface are probably the physical dislodging of cells by microscopic deformation of the surface carrying affinity ligands and the removal of dislodged particles by the flow of liquid squeezed out. The presence of specific eluent may contribute to detachment by decreasing the equilibrium number of bonds and preventing readsorption of detached particles on their way out of the column.

The observed phenomenon of cell release by mechanical deformation of the affinity surface was further studied by analyzing the effect of different parameters, such as concentration and nature of specific eluent, density and affinity of binding groups at the surface of bioparticles, different size and geometry of the particles, and rigidity of cryogel pore walls on the efficiency of the squeezing procedure. For this purpose, monoliths with various immobilized affinity ligands (Table 1) were inserted into the open-ended wells of a 96-well microtiter plate. Capillary forces keep liquid inside the cryogel monoliths, making them “drainage protected,” i.e., application of a certain volume of liquid on top of the column results in the displacement of exactly the same volume at the bottom. The drainage-protecting properties of cryogel monoliths make them suitable for the application in the multiwell format, which allows parallel analysis of a large number of samples (29).

“Dense” (6%) cryogel monoliths are characterized by a significantly higher density of the pore walls than “soft” (5%) cryogel monoliths. The thickness of the pore walls increases, whereas the pore size decreases with increasing concentration of monomers in the initial reaction mixture (33). The difference in the structure of the pore walls results in different elastic modulus of the monoliths, 0.016 and 0.065 MPa for “soft” and “dense” cryogel monoliths, respectively. Mechanical properties of the monoliths may also be modulated by changing the degree of cross-linking (34).

In this series of experiments, two additional model systems were studied: the adsorption and release of recombinant Escherichia coli cells on Ni(II)–IDA–cryogel monoliths, and of IgG-labeled inclusion bodies (the amino and carboxylic ends of the target protein are located at the surface of the inclusion bodies and are recognized by specific antibodies) (35) on protein A–cryogel monoliths (Table 1).

The strategy of flow-induced detachment was not applicable in the 96-well format. Therefore, detachment of bound particles was carried out by conventional elution, i.e., by passing 3 column volumes of the appropriate eluent through the wells or by compressing the adsorbent equilibrated with different concentrations of the eluent. No more than 20% and 38% of the yeast cells captured on ConA–cryogel monoliths were eluted with 0.3 M solutions of glucose and α-d-manno-pyranoside, respectively. Only 10–20% of captured His6-E.coli cells were recovered from Ni(II)–IDA–cryogel monoliths by elution with 50 mM EDTA (Fig. 3), whereas microgel particles were not detached at all by conventional elution, even at high concentrations of EDTA. The fact that microgel particles do not interact with native cryogel monoliths, whereas microgels captured on Cu(II)–IDA–cryogel monoliths remain bound to the adsorbent after EDTA treatment, which removes Cu(II) ions, indicates that specific binding promotes further nonspecific adhesion of microgel particles to the adsorbent.

Fig. 3.
View larger version:
Fig. 3.

Release of bound recombinant E. coli cells by conventional elution and by compression of Ni(II)–IDA–cryogel monoliths equilibrated with different concentrations of EDTA. The amount of bound cells was assumed to be 100%.


Compression, even in the absence of the eluent in the equilibration buffer, detached 40–80% of bound cells. Manual compression with a glass rod takes ≈1–2 seconds and results in the expression of 365–375 μl of liquid. Thus, detached cells enter the flow at a high velocity, which prevent most of cells from readsorbing. The rate of cell release is faster than the kinetics of bond formation between the particles and the affinity matrix. Because immobilized affinity ligands are not damaged by compression, it is probable that the bonds may reform when the cells are left in contact with the affinity monolith once compression is complete.

The amount of cells released increased with increasing concentration of the specific eluent in the running buffer. The effect of compression was especially pronounced in the case of soft cryogel monoliths, which have higher porosity and elasticity than dense cryogel monoliths. Practically all bound yeast cells and recombinant E. coli cells were released by compressing soft ConA– and Ni(II)–IDA–cryogel monoliths in the presence of 10 mM α-d-mannopyranoside and 3 mM EDTA, respectively. A quantitative release of yeast cells by compressing the matrix was achieved in the presence of 40–60 mM glucose. Glucose is an eluent with a lower affinity for ConA ligands, and, therefore, higher concentrations were required than in experiments with α-d-manno-pyranoside. Thus, the recovery of affinity-bound cells depends on both the concentration and the nature of the specific eluent. Quantitative recovery of yeast cells by compressing dense ConA–cryogel monoliths was observed at 0.5 M α-d-manno-pyranoside, and 80% was recovered in the presence of 0.7 M glucose. Recombinant E. coli cells recovered by squeezing retained their viability and grew on chloramphenicol-containing agar plates.

The concentration of the eluent in the equilibration buffer affected the amount of microgel particles released by compression. About 40% of the microgel particles remained bound to the adsorbent after compression in the presence of 0.3 M imidazole or 20 mM EDTA. Further increase in the eluent concentration did not result in improved recovery, probably because of the nonspecific interactions mentioned above. The density of the pore walls and porosity of the affinity cryogel monolith had a less-significant effect on the efficiency of recovery of microgel particles by compressing the matrix than in the case of yeast and His6-E. coli cells. The most pronounced effect of the structure of the cryogel monolith on compression was observed in the case of large particles, i.e., yeast cells, probably because of increased mechanical entrapment of bigger cells within dense monoliths with decreased elasticity and pore size.

Binding between protein A and a variety of mammalian IgG molecules is strong, and harsh conditions (e.g., pH 2.0–3.0) are commonly used for the elution of IgG bound to protein A adsorbents. By using two types of antibodies, against 15- and against 17-aa residues at the N and C termini (anti-A15-IgG and anti-B17-IgG) of the protein produced as inclusion bodies during the fermentation of recombinant E. coli cells, it was possible to bind antibody-labeled inclusion bodies on protein A–cryogel. About 60% and 40% of bound anti-A15-labeled and anti-B17-labeled inclusion bodies, respectively, was released by compressing the monoliths under very mild conditions at pH 7.0. Anti-B17-IgG has a higher affinity for the antigens on the surface of inclusion bodies than does anti-A15-IgG. The preparations of IgG-labeled inclusion bodies obtained under identical conditions but with different IgGs are characterized by different densities of the respective IgG on the surface of inclusion bodies. Stronger binding was observed in the case of anti-B17-IgG-labeled inclusion bodies, i.e., fewer anti-B17-IgG-labeled particles were released by compression. Thus, the density of the “receptors” on the surface of the bound particles affected the efficiency of the compressing procedure. The lowest recovery yield (18%) was observed in the case of inclusion bodies labeled in the presence of higher amounts of anti-B17-IgG, i.e., these particles had more receptors on the surface available for binding with protein A–cryogel monoliths.

The effect of compression to release the target macromolecule, His6-LDH, bound by single-site interactions on a Cu(II)–IDA–cryogel monolith was studied as a control experiment. Efficient desorption of protein from Cu(II)–IDA–cryogel monoliths was achieved by conventional elution with 0.3 M imidazole. Compressing the Cu(II)–IDA–cryogel monoliths did not result in any increase in the recovery of bound protein at any concentration of imidazole used in the study.

The results presented above clearly demonstrate that compression of elastic cryogel monoliths significantly improves the detachment of bound bioparticles, regardless of the type of individual ligand–receptor pair: metal ion–chelating ligand, Fc fragment of IgG–protein A or Con A–sugar ligand. Because the detachment conditions are very mild, the viability of cells that were bound and later released was retained. It was interesting to compare the efficiency of detachment and retained viability of cells eluted by compression with that of cells detached by shear force rather than by deformation of the matrix, e.g., when vortexing the beads with bound cells. As a model, we chose fragile mammalian cells, namely, human acute myeloid leukemia cells (KG-1). The cells have CD34 receptors on the surface and, after labeling with anti-CD34 antibodies, were bound either to protein A–cryogel monoliths or poly(vinyl alcohol) (PVA) beads with coupled protein A. The porous structure and elastic properties of PVA beads are rather similar to those of cryogel monoliths; however, the pore size (0.1–1 μm) of PVA beads is too small for KG-1 cells to penetrate inside the pores, so the cells are bound only to the outer surface and are fully exposed to the shear force during vortexing.

Again, compression of cryogel monoliths significantly increased the recovery of cells (Table 3). Moreover, both the efficiency of cell detachment by compression and the viability of the detached cells were significantly higher than when vortexing beads with cells bound to the bead surface.

View this table:
Table 3. Recovery of antibody-labeled CD34 human acute myeloid leukemia (KG–1) cells from protein A–cryogel monoliths and protein A PVA beads

These results illustrate the interaction of cells with an elastic affinity surface. Studies on affinity-mediated binding and recovery of bioparticles under various conditions by using affinity-cryogel monoliths can serve as a model mimicking the interactions of bacteria, viruses, macrophages, etc., with elastic tissues in biological systems. Indeed, it was possible to simulate the contraction of biological tissue triggered by conformational transitions of polymers forming the tissue matrix. The cryogel monoliths were prepared from a so-called thermoresponsive polymer, poly(N-isopropylacrylamide) (polyNIPAAm). PolyNIPAAm gels shrink dramatically as the temperature is increased above a certain critical transition temperature Tc , which is ≈32°C, depending on the conditions and the structure of the gel. The reason for the gel shrinking is believed to be the coil-to-globule transition of poly-NIPAAm macromolecules (36). Macroporous polyNIPAAm-based cryogel monoliths shrank in a way similar to ordinary polyNIPAAm gels when the temperature was increased >Tc. When bearing Cu(II)–IDA ligands, polyNIPAAm-based cryogel monoliths bound E. coli cells in a way similar to that of polyacrylamide-based Cu(II)–IDA–cryogel monoliths. Compression of the poly-NIPAAm-based cryogel monoliths at 25°C in the absence of imidazole detached 56% of the bound cells. The bound cells were eluted with 65% efficiency by using 0.2 M imidazole buffer at 25°C, i.e., <Tc. However, when elution was carried out with the same buffer at 40°C, i.e., >Tc , the polyNIPAAm-based cryogel monolith shrank almost instantaneously upon contact with the “warm” buffer, resulting in the release of 85% of the bound cells. The cells eluted at the elevated temperature retained their viability. Hence, the improved elution of bound bioparticles when the matrix underwent elastic deformation has been demonstrated for different types of deformation, as a result of either external forces (mechanical compression) or internal forces (shrinking of the matrix because of the coil-to-globule transition of the polymer forming the matrix).

The phenomenon of detachment of specifically bound bioparticles from the matrix when undergoing elastic deformation is believed to be of generic character, because it was demonstrated for a variety of bioparticles of different sizes (inclusion bodies, microbial cells, yeasts, and mammalian cells) as well as for synthetic particles, for different ligand–receptor pairs (IgG–protein A, sugar–ConA, metal ion–chelating ligand) and when the deformation was caused by either external or internal forces.

From the practical point of view, the release of specifically bound cells, especially fragile mammalian cells, by mechanical compression of elastic macroporous monoliths is a unique and efficient method of cell detachment. Mechanical compression may be scaled up by using simple mechanical devices. However, homogeneous increase of temperature in a large volume to recover cells from temperature-sensitive materials may be difficult to realize. The elasticity of cryogel monoliths ensures high recovery of bound cells under mild conditions, thus ensuring retained viability. This detachment strategy employing continuous porous structure makes cryogel monoliths very attractive for applications in affinity cell separation.

Previous SectionNext Section

Materials and Methods

Materials. Monolithic epoxy-activated polyacrylamide cryogels (rods; 0.5 ml) (cryogel monoliths) produced by using 5% and 6% solutions of comonomers in the reaction mixture (soft and dense monoliths, respectively) and poly(N-isopropylpolyacrylamide)-based (polyNIPAAm) cryogels were provided by Protista Biotechnology. The monoliths had a size (12.5 × 7.1-mm diameter) that fitted into the well (7-mm diameter) of a 96-well plate. The preparation of PVA beads (elastic particles of 0.5–1 mm) and coupling of protein A to the PVA beads has been described in ref. 32. Anti-CD34 monoclonal antibodies, recognizing the cell surface receptors of KG-1 cells, were produced as ascites by injecting human stem-cell antigen (HSCA) hybridoma cells (Kaneka Chemical Industries, Tokyo) into mice and purified by using a protein G column (PerSeptive Biosystems, Foster City, CA). A human acute myeloid leukemia cell line (KG-1) expressing CD34 surface antigens was obtained from The Global Bioresource Center (American Type Culture Collection, CCL-246.1). The recombinant strain of E. coli TG1 cells expressing a thermostable lactate dehydrogenase (LDH) (from thermophilic Bacillus stearothermophilus) carrying a tag of six histidine residues (His6-LDH) was a gift from Leif Bülow (Lund University). E. coli K12 strain pop 6510 (thr, leu, tonB, thi, lacY1, recA, dex5, metA, supE, and dex5) with plasmid pLH2 encoding the hybrid LamB-His (two 6× His) monomers (His6-E. coli) (37) was generously provided by Victor de Lorenzo (Centro Nacional de Biotecnologia-Consejo Superior de Investigaciones Cientificas, Madrid, Spain). Microgels were prepared by radical copolymerization of N-isopropylacrylamide, vinylimidazole, and N,N′-methylene-bis(acrylamide) at 60°C until a uniform colloidal suspension of microgels had formed.

A 33-kDa model protein fermented as inclusion bodies in recombinant E. coli and IgG anti-A15 and anti-B17 against 15 aa on the amino end and 17 aa on the carboxylic end of the 33-kDa target protein, respectively (35), were generously provided by Josefin Ahlqvist and Gunnar Hörnsten (Swedish Institute for Food and Biotechnology, Ideon Science Park, Lund, Sweden). Baker's yeast was purchased from a local store.

Uniaxial Compression of Cryogel Monoliths. A glass tube (20 × 7 mm i.d.) containing a monolith (12.5 × 7.1 mm) was placed on a digital balance (1264 MP; Sartorius). A load was applied by placing a glass rod vertically on top of the cryogel monolith. The compressional force acting on the monolith was calculated from the reading of the balance. The resulting deformation was measured with a ruler. The compression modulus of cryogel monoliths was estimated by using the equation E = F/A × (Δl/l)–1, where E is the elastic modulus, F is the applied force, A is the cross-sectional area of the test sample, l is the initial length of the test sample, and Δl is the change in length under the compressive force. The measurements were performed in triplicate with different samples. The Δl was proportional to F.

Affinity Cryogel Monoliths. Cu(II)/Ni(II)–IDA–, Con A– and protein A–cryogel monoliths were prepared, and the amount of immobilized ligands was determined as described in refs. 25 and 26.

Cell Culture and Protein Expression. The E. coli K12 strain with the plasmid pLH2 encoding the hybrid LamB-His6 monomers (His6-E. coli) and the E. coli TG1 strain producing His6-LDH (E. coli TG1) were cultured as described in refs. 25 and 37. The cell pellet was kept on ice and suspended in 20 mM Hepes/200 mM NaCl, pH 7.0, before adsorption tests. The cells were used within 1–2 days after cultivation. The assay of LDH activity was performed as described in ref. 28.

Binding and Recovery of Microbial Cells and Microgel Particles by Using Affinity Cryogel Monoliths. The following running buffers were used in the adsorption tests: 0.1 M Tris with 150 mM NaCl, 5 mM CaCl2, and 5 mM MgCl2, pH 7.4, for yeast cells and ConA–cryogel monoliths and 20 mM Hepes with 0.2 M NaCl, pH 7.0, in the immobilized metal affinity chromatography (IMAC) tests. Yeast cells, His6-E. coli cells, and microgel particles were suspended in the appropriate running buffer to an OD600 of 0.855 and an OD450 of 0.727 and 0.674, respectively. Aliquots (200 μl) of suspensions of yeast cells, His6-E. coli cells and microgel particles were applied to ConA–, Ni(II)–IDA–, and Cu(II–IDA–cryogel monoliths, respectively, equilibrated with the corresponding running buffer. After 10–15 min of incubation, cryogel monoliths were washed with 3.5 ml of the appropriate buffer to remove unbound particles. The effluents were analyzed by measuring the absorbance at 600 nm, in the case of yeast cells, and at 450 nm in the cases of His6-E. coli cells, and microgel particles (Cu(II)/EDTA at the concentrations used do not absorb light at this wavelength). The amount of bound nanoparticles was calculated as the difference between the amounts of applied and unbound nanoparticles.

The desorption of nanoparticles from affinity cryogel monoliths was carried out by elution with 1.5 ml of the appropriate eluent: α-d-manno-pyranoside (or glucose) in the case of yeast cells bound to ConA–cryogel monoliths and EDTA (or imidazole) in IMAC test. Alternatively, the particles were recovered by compressing the cryogel monoliths by using a glass rod. The monoliths were reswollen by adding a new portion (0.5 ml) of the eluent and compressed once again. The liquid expressed (total volume 0.70–0.75 ml) was collected and analyzed by measuring the absorbance at 600 or 450 nm.

When polyNIPAAm-based cryogel monoliths were used, E. coli (1-ml cell suspension; OD450 of 0.717) was applied to the column at a flow rate of 1.0 ml/min by using 50 mM PBS with 0.2 M NaCl. The column was washed thoroughly with the binding buffer, and elution was then carried out by passing 50 mM PBS, pH 7.2, containing 0.4 M NaCl at a constant temperature of 35–40°C. When the warm elution buffer was passed through the column, the cryogel shrank. In a separate experiment, recovery was also carried out by using specific eluent, 0.2 M imidazole buffer, pH 7.4, containing 0.4 M NaCl at 25°C.

Binding and Recovery of Inclusion Bodies by Using Protein A–Cryogel Monoliths. The preparation of IgG-labeled inclusion bodies and the binding were conducted as described in ref. 35. The recovery of bound inclusion bodies was carried out by compressing the protein A–cryogel monoliths as described above.

Binding of Mammalian Cells to PVA Beads and Cryogel Monoliths with Coupled Protein A. PVA beads with coupled protein A (0.1 ml) were washed three times with 1 ml of binding buffer (20 mM Hepes/0.2 M NaCl, pH 7.0). The beads were then suspended in 500 μl of binding buffer. Lymphocytes (1 ml, 2.0–4.0 × 107 cells per ml) were first treated with 10 μl (1 μg/μl) of goat anti-human IgG by incubating at 4°C for 15 min. Cells were centrifuged at 200 × g for 5 min and suspended in 1 ml of balanced salt solution (0.145 M Tris·HCl, pH 7.6, containing 0.1% glucose, 5.4 mM KCl, and 14 mM NaCl). Antibody-treated lymphocytes (0.5 ml) were applied to the beads in batch mode and incubated at 4°C for 20 min with gentle shaking. The unbound lymphocytes were collected and kept on ice for subsequent flow-cytometric analysis. The same volume of protein A beads was tested in packed-bed mode. Microcolumns (0.6 ml) containing protein A beads were washed with binding buffer, and anti-human-IgG-labeled lymphocytes were loaded. The cell-loaded affinity columns were maintained at 4°C for 20 min by closing the outlet. The unbound cells were then removed by opening the outlet and washing the column with buffer.

The human acute myeloid leukemia (KG-1) cells (1 ml; 1 × 107 cells per ml) were first treated with 10 μl (1 μg/μl) of monoclonal antibodies, anti-CD34, by incubating at 4°C for 15 min. Cells were centrifuged at 200 × g for 5 min and suspended in 1 ml of balanced salt solution. The labeled cells were then applied to protein A beads as described above.

Cell binding on the protein A–cryogel monoliths was carried out as described in refs. 26 and 33. The cell-recovery rate was investigated with different methods. To provide milder desorption conditions, the cells were recovered by displacement with IgG solution. First, the monolithic column or the beads with adsorbed cells were incubated at 37°C for 30 min. To the column, 2 ml of 40 mg/ml human IgG solution was applied, and the column was incubated at room temperature without buffer flow for 10 min. Then, 20 ml of buffer was run through the column at a flow rate of 1.5 ml/min. Similarly, bound cells on the beads were recovered by treating 0.5 ml of beads with bound cells with 0.5 ml of 40 mg/ml IgG solution and gently vortexing.

The cells were also recovered mechanically by compressing the monolith. The monolith was pressed gently in the syringe column with the syringe piston, either in the presence of IgG solution or without it. Compression was repeated after reswelling of the gel in buffer to achieve maximum recovery of the cells. In the case of beads, the cells were recovered by vortexing the beads in the presence or absence of IgG solution.

All experiments described above were performed in triplicate Cryogel samples for microscopy were prepared according to ref. 25.

Previous SectionNext Section

Acknowledgments

This work was supported by Protista Biotechnology AB, the Swedish Foundation for Strategic Research (area of Chemistry and Life Sciences), and the Swedish Research Council/Swedish International Development Cooperation Agency-research link project.

Previous SectionNext Section

Footnotes

  • ↵ § To whom correspondence should be addressed. E-mail: bo.mattiasson{at}biotek.lu.se.

  • Conflict of interest statement: No conflicts declared.

  • This paper was submitted directly (Track II) to the PNAS office.

  • Abbreviations: IDA, iminodiacetate; LDH, lactate dehydrogenase; PVA, poly(vinyl alcohol).

Previous Section
 

References

  1. ↵
    Mammen, M., Choi, S.-K. & Whitesides G. M. (1998) Angew. Chem. Int. Ed. 37 , 2754–2794.
    CrossRef
  2. ↵
    Hubble, J. (1997) Immunol. Today 18 , 305.pmid:9190118
    Medline
  3. ↵
    Cao, X., Eisenthal, R. & Hubble, J. (2002) Enzyme Microb. Technol. 31 , 153–160.
    CrossRef
  4. ↵
    Bell, G. I. (1978) Science 200 , 618–627.pmid:347575
    Abstract/FREE Full Text
  5. ↵
    Kuo, S. C. & Lauffenburger, D. A. (1993) Biophys. J. 65 , 2191–2200.pmid:8298043
    Abstract/FREE Full Text
  6. ↵
    Ward, M. D., Dembo, M. & Hammer, D. A. (1995) Ann. Biomed. Eng. 23 , 322–331.pmid:7631985
    MedlineISI
  7. ↵
    Lam, A., Cao, X., Eisenthal, R. & Hubble, J. (2001) Enzyme Microb. Technol. 29 , 28–33.pmid:11427232
    Medline
  8. ↵
    Gomez-Suarez, C., Busscher, H. J. & Van der Mei, H. C. (2001) Appl. Environ. Microbiol. 67 , 2531–2537.pmid:11375160
    Abstract/FREE Full Text
  9. ↵
    Ming, F., Whish, W. J. D. & Hubble, J. (1998) Enzyme Microb. Technol. 22 , 94–99.
    CrossRef
  10. ↵
    Cozens-Roberts, C., Quinn, J. A. & Lauffenburger, D. A. (1990) Biophys. J. 58 , 857–872.pmid:2174272
    Abstract/FREE Full Text
  11. ↵
    Dainiak, M. B., Galaev, I. Yu. & Mattiasson, B. (2002) J. Chromatogr. A 942 , 123–131.pmid:11822378
    Medline
  12. Castner, D. G. & Ratner, B. D. (2002) Surf. Sci. 500 , 28–60.
    CrossRef
  13. ↵
    Nath, N., Hyun, J., Ma, H. & Chilkoti, A. (2004) Surf. Sci. 570 , 98–110.
    CrossRef
  14. ↵
    Curtis, A. & Wilkinson, C. (1997) Biomaterials 18 , 1573–1583.pmid:9613804
    CrossRefMedlineISI
  15. ↵
    Li, X., Liu, T. & Chen, Y. (2004) Biochem. Eng. J. 22 , 11–17.
    CrossRef
  16. ↵
    Rijnaarts, H. H. M., Norde, W., Bouwer, E. J., Lyklema, J. & Zehnder, A. J. B. (1995) Colloids Surf. B 4 , 5–22.
  17. Jucker, B. A., Harms, H. & Zehnder, A. J. B. (1996) J. Bacteriol. 178 , 5472–5479.pmid:8808938
    Abstract/FREE Full Text
  18. ↵
    Parent, M. E. & Velegol, D. (2004) Colloids Surf. B 39 , 45–51.
    CrossRef
  19. ↵
    Madigan, M. T., Martinko, J. M. & Parker, J. (2000) in Brock Biology of Microorganisms, (Prentice–Hall, Upper Saddle River, NJ), 9th Ed., pp. 773–799
  20. ↵
    Pelham, R. J. & Wang, Y. L. (1997) Proc. Natl. Acad. Sci. USA 94 , 13661–13665.pmid:9391082
    Abstract/FREE Full Text
  21. Engler, A. J., Richert, L., Wong, J. Y., Picart, C. & Discher, D. E. (2004) Surf. Sci. 570 , 142–154.
    CrossRef
  22. ↵
    Wong, J. Y., Leach, J. B. & Brown, X. Q. (2004) Surf. Sci. 570 , 119–133.
    CrossRef
  23. ↵
    Engler, A., Bacakova, L., Newman, C., Hategan, A., Griffin, M. & Discher, D. (2004) Biophys. J. 86 , 617–628.pmid:14695306
    Abstract/FREE Full Text
  24. ↵
    Genes, N. G., Rowley, J. A., Mooney, D. J. & Bonassar, L. J. (2004) Arch. Biochem. Biophys. 422 , 161–167.pmid:14759603
    Medline
  25. ↵
    Arvidsson, P., Plieva, F. M., Lozinsky, V. I., Galaev, I. Yu. & Mattiasson, B. (2003) J. Chromatogr. A 986 , 275–290.pmid:12597634
    Medline
  26. ↵
    Kumar, A., Plieva, F. M., Galaev, I. Yu. & Mattiasson, B. (2003) J. Immunol. Methods 283 , 185–194.pmid:14659910
    Medline
  27. Dainiak, M. B., Kumar, A., Plieva, F. M., Galaev, I. Yu. & Mattiasson, B. (2004) J. Chromatogr. A 1045 , 93–98.pmid:15378883
    Medline
  28. ↵
    Arvidsson, P., Plieva, F. M., Savina, I. N., Lozinsky, V. I., Fexby, S., Bulow, L., Galaev, I. Yu. & Mattiasson, B. (2002) J. Chromatogr. A 977 , 27–38.pmid:12456093
    CrossRefMedline
  29. ↵
    Galaev, I. Yu., Dainiak, M. B., Plieva, F. M., Hatti-Kaul, R. & Mattiasson, B. (2005) J. Chromatogr. A 1065 , 169–175.pmid:15782962
    Medline
  30. ↵
    Hubble, J. (1997) Trends Biotechnol. 15 , 249–255.
  31. ↵
    Ming, F., Eisenthal, R., Whish, W. J. D. & Hubble, J. (2000) Enzyme Microb. Technol. 26 , 216–221.pmid:10689080
    Medline
  32. ↵
    Kumar, A., Rodríguez-Caballero, A., Plieva, F. M., Galaev, I. Yu., Nandakumar, K. S., Kamihira, M., Holmdahl, R., Orfao, A. & Mattiasson, B. (2005) J. Mol. Rec. 18 , 84–93.
    CrossRef
  33. ↵
    Plieva, F. M., Karlsson, M., Aguilar, M.-R., Gomez, D., Mikhalovsky, S. & Galaev, I. Yu. (2005) Soft Matter 1 , 303–309.
    CrossRef
  34. ↵
    Peppas, N. A., Bures, P., Leobandung, W. & Ichikawa, H. (2000) Eur. J. Pharm. Biopharm. 50 , 27–46.pmid:10840191
    CrossRefMedline
  35. ↵
    Ahlqvist, J., Kumar, A., Ledung, E., Sundström, H., Hörnsten, G. & Mattiasson, B. J. Biotechnol., in press.
  36. ↵
    Shibayama, M. & Tanaka, T. (1993) in Responsive Gels: Volume Transitions I, ed. Dusek, K. (Springer, Berlin), pp. 1–62.
  37. ↵
    Sousa, C., Cebolla, A. & DeLorenzo, V. (1996) Nat. Biotechnol. 14 , 1017–1020.pmid:9631043
    CrossRefMedline
« Previous | Next Article »Table of Contents

This Article

  1. Published online before print January 17, 2006, doi: 10.1073/pnas.0508432103 PNAS January 24, 2006 vol. 103 no. 4 849-854
  1. AbstractFree
  2. Figures Only
  3. » Full Text
  4. Full Text (PDF)

Classifications

Navigate This Article

About Our New Site Design >>
Link: Advanced Search

This Week's Issue

  1. August 19, 2008, 105 (33)
  1. Current Issue
  1. Alert me to new issues of PNAS

Most