Elsevier

Chemical Engineering Science

Volume 63, Issue 21, November 2008, Pages 5347-5355
Chemical Engineering Science

Separation of amino acids by simulated moving bed under solvent constrained conditions for the integration of continuous chromatography and biotransformation

https://doi.org/10.1016/j.ces.2008.07.019Get rights and content

Abstract

A large number of promising enzyme reactions, such as diastereospecific carbon–carbon bond formation by aldolases, suffers from an unfavorable position of the reaction equilibrium. Combining continuous chromatography and enzymatic reactions should allow for a new in situ or online product recovery process to achieve high reaction productivity and yield and make these biocatalysts economically more attractive. The integration imposes a series of constraints on the chromatographic separation, mainly on the applicable solvent, which is at the same time the reaction medium for the enzymatic reaction. We exploit this concept for a model process, the integrated biocatalytic production of L-allo-threonine from glycine and acetaldehyde. Of crucial importance for this process is the separation of the two physicochemically similar amino acids glycine and threonine, in particular in the presence of additional compounds such as the second starting material and enzyme cofactors. This separation was first investigated on a lab-scale simulated moving bed (SMB) unit under enzyme compatible conditions. After triangle theory-based identification of SMB operating points, the two amino acids could be efficiently separated, applying aqueous eluents with minor content of organic co-solvent at neutral pH on a weak cation exchanger resin. Remarkably, the separation performance with respect to the two amino acids was only slightly reduced by coupling the SMB to a continuously operated enzyme membrane reactor, whose efflux contained, in addition to the amino acids, acetaldehyde and the cofactor pyridoxal-5-phosphate. This represents an important step to the future design of even further integrated biocatalytic reaction–separation schemes.

Introduction

Simulated moving bed (SMB) is a continuous chromatographic multi-column process that emerged in the 1960s in the oil industry (Broughton, 1968, Broughton et al., 1970) and has made its way since into the pharmaceutical and fine chemicals industry (Juza et al., 2000). SMB technology provides some advantages over batch chromatography, like the efficient continuous operation mode that leads to reduced volumes of stationary and mobile phase and results in considerable cost savings (Pynnonen, 1998). The continuous mode of operation is also more permissive to integration with additional equipment in a more complex process design (Nicoud, 2000). Furthermore, SMB-operation is easy to scale up (Nicoud, 1999).

The online integration of SMB and continuous biotransformations (Bechtold et al., 2006) enables a variety of applications, such as driving thermodynamically limited reactions to a theoretical yield of 100%. Previous processes that integrated SMB technology and biotransformation mostly applied the concept of an SMB reactor (SMBR) and addressed product inhibition (Akintoye et al., 1991, Azevedo and Rodrigues, 2001, Shieh and Barker, 1996). An SMBR is basically operated the same way as a conventional SMB, with the catalytic enzyme simply added throughout the unit, either in soluble form to the mobile phase or in immobilized form to the stationary phase. The SMBR concept was furthermore applied to overcome thermodynamic limitations in lipase-catalyzed esterifications in organic solvents, where removal of the by-product water allowed increasing reaction yield (Meissner and Carta, 2002, Mensah and Carta, 1999). In these cases, the enzyme was part of the stationary phase while one substrate was added with the feed and the other substrate with the eluent. We proposed previously another concept of integration, separated but coupled SMB and reactor (cSMB&R) (Bechtold et al., 2006), for which we here investigate a promising SMB separation.

We are interested in the stereoselective formation of carbon–carbon bonds by aldolases, a topic of prime importance in (bio)catalysis (Breuer and Hauer, 2003). Though of high chemical potential, many biotransformations in this group are characterized either by the dependence on expensive starting materials (Samland and Sprenger, 2006) or by the limitation of product yield due to unfavorable reaction thermodynamics (Kimura et al., 1997, Mahmoudian et al., 1997). Specifically, we investigate the formation of L-allo-threonine from glycine and acetaldehyde catalyzed by the serine hydroxymethyltransferase GlyA from Escherichia coli as a model reaction (Fig. 1). The enzyme has been previously thoroughly characterized for process application and applied for continuous reactions in an enzyme membrane reactor (EMR) (Makart et al., 2007a).

The integration of the enzyme reactor and continuous chromatography defines a series of constraints for process conditions. First of all, it restricts the range of useful eluents, which at the same time constitute the reaction medium for the biotransformation. This reduces the choice for most enzymes effectively to aqueous buffers with minor content of organic co-solvents (Makart et al., 2007a). Furthermore, the chosen model reaction produces a complex reactor efflux that needs to be separated on the SMB. It contains glycine as the cheap starting material provided in excess to drive the reaction to the product side, the second starting material acetaldehyde, the product L-allo-threonine, and—though in small concentrations only—the cofactor pyridoxal-5-phosphate (PLP) that is required for the functioning of the enzyme.

In this model reaction system, the prime separation task is the separation of the two amino acids glycine and threonine to remove the product in pure form from the system and to return the excess of glycine to the reaction system, and this in the face of the very similar physicochemical characteristics the two amino acids have. Amino acids are commonly separated on sulfonated cation exchangers for analytical purposes (Weiss, 2004) or on cation exchangers on production scale, as in the case of purification of lysine from cell-free fermentation broth (Hermann, 2003). These separations are performed under strongly acidic conditions, under which the carboxyl group is no longer charged and the amine group is protonated, favoring the adsorption to the cation exchanger. These conditions are usually not feasible in an integrated process with biotransformation. Furthermore, reports in the literature on the separation of amino acids using SMB focus mainly on the model compounds phenylalanine and tryptophan (Hur et al., 2007, Wu et al., 1998, Xie et al., 2000, Yang et al., 2004) in an unbuffered system with water as eluent. Consequently, we investigated a novel SMB-process for the integration with aldolase reactions.

Section snippets

Adsorption isotherms

The determination of the adsorption isotherms for the compounds that are involved in the separation is fundamental for SMB-design. For the applied low feed concentrations of up to 500 mM amino acid, linear adsorption isotherms can be used for threonine and glycine:qi=Hi·ciwith qi as the adsorbed solid phase concentration, ci as the mobile phase concentration and Hi as Henry's constant of compound i. In the case of linear isotherms, the shape of the peaks remains symmetric and the Henry constants

Chemicals

All chemicals were either from Sigma-Aldrich (Buchs, Switzerland) or Roth (Reinach, Switzerland) unless mentioned otherwise. Acetonitrile (HPLC grade) was purchased from Chemie Brunschwig (Basel, Switzerland). The suppliers for the tested resin materials are listed in the supplementary material. The resin material used for the SMB, Amberlite CG-50 II, was purchased from Sigma-Aldrich.

All percentages of liquid mixtures in this text refer to a volume to volume ratio.

Resin screening

The amino acids used for

Choice of stationary phase

From an extensive qualitative screening of commercially available resins (see supplementary material), we identified the weak cation exchanger Amberlite CG-50 II as the most promising stationary phase for an integrated process for the enzymatic production of L-allo-threonine. This material showed the highest selectivity and band broadening was not as pronounced as for other materials. With the material used in the reported SMB separations of amino acids, a poly-4-vinylpyridine resin (Wu et al.,

Conclusions

The system of coupled SMB and reactor has great potential as an in situ product removal tool able to cope with complex reactions and/or separations in one continuous process. On the other hand, the integration of the two steps—reaction and separation—in one process reduces the range of possible operation points for the reactor and the SMB, first of all due to the restrictions of useful eluents (Bechtold et al., 2006).

We could show the efficient separation of amino acids, glycine and threonine,

Notation

ciconcentration of species i, gL-1 or, mM as specified in the text
D,F,E,Rdesorbent, feed, extract and raffinate streams
HiHenry's constant of species i
jsection index
kglobal mass transfer coefficient, s-1
mjflow rate ratio in section j
ncolumn number
qeqsolid phase concentration at equilibrium
qiadsorbed phase concentration of species i, gL-1
Qvolumetric flow rate, mLmin-1
ttime, s
t0residence time of an unretained compound, s
tiRretention time for species i, s
t*switch time, s
usolid phase velocity, cmmin-

Acknowledgments

The authors wish to thank Dr. Joachim Stadler (GE Healthcare, Freiburg, Germany) for advice in the installation of the SMB unit and G. Caviezel and D. Strässler for their help in the resin screening. This work was supported by the ETH with a grant to S. Makart and by the Swiss National Science Foundation with a grant to M. Bechtold.

References (37)

  • S. Makart et al.

    Towards preparative asymmetric synthesis of β-hydroxy-α-amino acids: L-allo-threonine formation from glycine and acetaldehyde using recombinant GlyA

    Journal of Biotechnology

    (2007)
  • M. Mazzotti et al.

    Optimal operation of simulated moving bed units for nonlinear chromatographic separations

    Journal of Chromatography A

    (1997)
  • M. Pedeferri et al.

    Experimental analysis of a chiral separation through simulated moving bed chromatography

    Chemical Engineering Science

    (1999)
  • B. Pynnonen

    Simulated moving bed processing: escape from the high-cost box

    Journal of Chromatography A

    (1998)
  • M.D. Truppo et al.

    Production of (S)-γ-fluoroleucine ethyl ester by enzyme mediated dynamic kinetic resolution: comparison of batch and fed batch stirred tank processes to a packed bed column reactor

    Chemical Engineering Science

    (2008)
  • Q. Yu et al.

    Multicomponent ion-exchange chromatography for separating amino acid mixtures

    Reactive Polymers

    (1987)
  • A. Akintoye et al.

    The inversion of sucrose on a semicontinuous counter-current chromatographic bioreactor–separator

    Transactions of the Institution of Chemical Engineer

    (1991)
  • Y.A. Beste et al.

    Side-stream simulated moving-bed chromatography for multicomponent separation

    Chemical Engineering Technology

    (2002)
  • Cited by (0)

    View full text