Biocatalysis in Polymersomes: Improving Multienzyme Cascades with

Apr 26, 2017 - Compartmentalized Aqueous-Organic Emulsion for Efficient Biocatalysis. Qingcai Zhao , Marion B. Ansorge-Schumacher , Rainer Haag ...
0 downloads 0 Views 1MB Size
Subscriber access provided by HACETTEPE UNIVERSITESI KUTUPHANESI

Letter

Biocatalysis in Polymersomes: Improving Multienzyme Cascades with Incompatible Reaction Steps by Compartmentalization Ludwig Klermund, Sarah T Poschenrieder, and Kathrin Castiglione ACS Catal., Just Accepted Manuscript • Publication Date (Web): 26 Apr 2017 Downloaded from http://pubs.acs.org on April 26, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Ludwig Klermund, Sarah T. Poschenrieder and Kathrin Castiglione* Institute of Biochemical Engineering, Technical University of Munich, Boltzmannstr. 15, 85748 Garching. ABSTRACT: Incompatibilities encountered in multienzyme syntheses often arise from inhibition or inactivation of individual enzymes by low molecular mass compounds. Polymersomes have the postulated, yet unproven potential to enhance the performance of cascade reactions by spatial separation of enzymes from the respective source of incompatibility. A main challenge is the requirement to reduce mass transport limitations across the polymer membrane with sufficient selectivity to maintain the compartmentalization. We demonstrate that cross-inhibitions in cascade reactions can be avoided by reconstituting highly selective channel proteins into the membrane. Thus, the three-step synthesis of CMP-N-acetylneuraminic acid was improved 2.2-fold compared to the non-compartmentalized reaction.

KEYWORDS biocatalysis, cascade reaction, compartmentalization, incompatible reactions, membrane protein, multienzyme, nanoreactor, polymersome Enzymatic cascade reactions, i.e. the combination of several enzyme reactions in one pot without isolation of intermediates, have great potential for the establishment of sustainable chemical processes. However, many cascade reactions suffer from incompatibilities such as cross-inhibitions or -inactivations by components of the reaction system.1-3 The implementation of multienzyme syntheses in polymer vesicles formed from block copolymers that self-assemble in aqueous solution, so-called polymersomes, has been widely discussed.4-7 Especially the greater mechanical stability and a generally lower membrane permeability of polymersomes compared to liposomes 4 have led to the proposition of polymersomes as nano-scale enzyme membrane reactors (nano-EMRs). Due to the vast variety of different polymers that can assemble to polymersomes, the permeability can be tuned to match the desired characteristics. By equipping polymersomes with the necessary enzymes, these nanoreactors are capable of compartmentalizing reaction spaces. As such, polymersomes may overcome limits to multienzyme reactions which encounter cross-inhibitions or inactivations. The beneficial characteristics of polymersomes over liposomes are essentially important to control the compartmentalization, especially when employing polymersomes in biotechnological applications over extended reaction times. As a prerequisite, however, a highly selective mass transport across the membrane must be provided, enabling a spatial separation not only of the enzymes but also of low molecular mass compounds, which usually are the source of incompatibility. So far, three approaches have been followed to selectively reduce the diffusion barrier of the polymer membrane: 1) the use of porous membranes,8-10 2) the use of stimuli-responsive membranes that swell upon external stimulus and become more permeable,11-12 and 3) the use of low permeable membranes with reconstituted membrane channel proteins. 5, 13 In the first two cases, the physical principles enabling a spatial separation of molecules are restricted to (nonadjustable) size-

exclusion. For both membrane types, the molecular mass cutoffs (MMCO) have not been defined so far, but molecules with molecular masses in the range of 515 – 740 Da have been shown to diffuse readily.11, 14 As the majority of reactants involved in preparative biocatalytic reactions is significantly smaller, these membranes are not well suited to improve cascade reactions with incompatibility issues by compartmentalization. This is reflected in that only compatible reaction cascades have been implemented in polymersomes so far,7 with the exception of polymersomes used for protecting enzymes from proteolytic degradation.8-9 As a consequence, the spacetime-yields were not improved as the reaction cascade can at best perform as good as the same reaction free in solution. On the contrary, the incorporation of natural or engineered channel proteins in polymer membranes holds great potential for establishing highly selective mass transport.15 However, up to now, the available transport protein repertoire that has been successfully inserted in polymer membranes is rather limited and most of the studies have been focusing on unspecific porins with high MMCOs such as the Outer Membrane Protein F (OmpF) from Escherichia coli (MMCO: 600 Da). Thus, the potential of polymersomes with reconstituted channel proteins for the improvement of reaction cascades with incompatible reaction steps is fully unexploited.

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 10

Figure 1. (A) Three-step synthesis of CMP-N-acetylneuraminic acid (CMP-Neu5Ac). (B) Compartmentalized reaction scheme in polymersomes. The AGE is separated from the incompatible component CTP by encapsulation in polymersomes. The NAL and the CSS are immobilized on the surface to form a single biocatalytic entity. Channel proteins enable a selective mass transport across the polymer membrane.

Herein, we describe the applicability of porin-functionalized polymersomes to not only perform but enhance multienzyme reactions with cross-inhibitions on the exemplary three-step synthesis of CMP-N-acetylneuraminic acid (CMP-Neu5Ac) from N-acetylglucosamine (GlcNAc), pyruvate and cytidine triphosphate (CTP) (Figure 1A). Furthermore, we demonstrate that a highly selective membrane permeability is crucial for reaction improvement. The model reaction is composed of three enzymatic reactions. In the first reaction step, GlcNAc is converted to Nacetylmannosamine (ManNAc) by an N-acyl-D-glucosamine 2-epimerase (AGE). The AGE requires adenosine triphosphate (ATP) as allosteric activator without consuming it. In a second step, ManNAc reacts with pyruvate to N-acetylneuraminic acid (Neu5Ac) in an aldol condensation reaction catalyzed by an N-acetylneuraminate lyase (NAL). Neu5Ac is then activated with CTP to form CMP-Neu5Ac by a CMP-sialic acid synthetase (CSS). The reaction cascade in one pot is favored over isolated reaction steps because both the AGE and the NAL reaction exhibit an unfavourable reaction equilibrium, which lies on the substrate side (Keq,AGE = 0.2616; Keq,NAL = 2.1 L mol-1).17 In turn, the quasi-irreversible CSS reaction can pull the cascade to completion.18-21 However, the implemented system suffers from incompatibilities that mainly arise from a

strong inhibition of the substrate CTP of the third reaction on the enzyme AGE of the first reaction (inhibition constant = 1 mM, Figure S1). This incompatibility between reaction 1 and 3 essentially requires the spatial separation of the AGE from CTP. To separate the incompatible AGE and CSS reactions in one pot, the AGE was encapsulated together with 1 mM ATP in the lumen of poly(methyloxazoline) 15poly(dimethylsiloxane)68-poly(methyloxazoline)15 (PMOXAPDMS-PMOXA) polymersomes (Figure 1B) during vesicle formation. Vesicle formation was performed according to a polymersome production method described by Poschenrieder et al.22 The NAL and the CSS were immobilized on the outer surface using hydrophobic peptide anchors as recently described (Figures S2-S3).23 Whereas the kinetic parameters of the NAL were not altered upon immobilization, the CSS activity was reduced by 68 % when immobilized on the polymersome surface with minor differences in the half-saturation constants (Table S2-S3). However, the reduced activity of the CSS was counterbalanced by a 24-fold stabilization of the CSS, which increased from a half-life of 1.5 h to 36 h at 30°C. The nano-EMRs had an average size of 110 nm in diameter as determined by dynamic light scattering.

ACS Paragon Plus Environment

Page 3 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

The polymer membrane was essentially not permeable toward the charged molecules pyruvate, Neu5Ac, CTP and ATP (permeability coefficients Pe below 10-13 cm s-1, Table S1) but showed low permeability toward the uncharged GlcNAc and ManNAc (Pe: 10-10 cm s-1). Thus, the polymer membrane poses a naturally selective barrier discerning between AGE substrates and AGE inhibitors and is able to introduce the required compartmentalization of the reaction cascade. However, the low permeability of GlcNAc and ManNAc across the membrane reduced the effective AGE activity by 80 % in intact polymersomes compared to its maximum activity under saturated substrate concentrations, demonstrating the major drawbacks of an insufficient exchange of substrates. This phenomenon is often described when using polymersomes in biocatalysis7, 15 and highlights the need for a reduction of mass transport limitations. Due to the pronounced mass transport limitations and the present incompatibilities, a highly selective mass transfer must be introduced to reduce the mass transport limitations but at the same time must be able to maintain the compartmentalization. To increase diffusion of GlcNAc and ManNAc into and out of the polymersomes and at the same time maintaining the compartmentalization of ATP and CTP, we chose to incorporate the channel protein OmpF G119D into the membrane. This OmpF mutant exhibits a reduced pore size with a MMCO of approximately 300 Da.24-25 Through the exchange of the glycine at position 119 to aspartate, an additional negative charge is introduced in the channel cavity which increases the cation selectivity, making it ideally suitable for selective transport of the polar GlcNAc and ManNAc (221 Da) and exclusion of the negatively charged Neu5Ac (309 Da), CTP (483 Da) and ATP (507 Da), which was also demonstrated experimentally (Figure S4). Whereas the exclusion of CTP is required to circumvent the cross-inhibition, the exclusion of ATP is beneficial in terms of retaining ATP in sufficiently high concentrations within the polymersomes to activate the AGE. Thus, ATP was encapsulated inside the polymersomes at a concentration of 1 mM (10 times the halfsaturation constant KM of the AGE for ATP; KM = 95 µM). Since ATP was stable under process conditions (97 % of ATP still present after 100 h) and was used in a more than 10-fold excess both in terms of the molar ratio (1 mM ATP vs. 25 µM AGE) and in terms of KM of the AGE for ATP, ATP was sufficiently available within the polymersomes. The small but negatively charged pyruvate was able to slowly diffuse through OmpF G119D although its diffusion was greatly impaired compared to the wildtype OmpF. For the reaction cascade, a total of 2.4∙1015 polymersomes per liter (0.13 % w/v of polymer) were employed with total enzyme concentrations of 0.56 mg L-1 AGE (8.77 U L-1), 0.31 mg L-1 NAL (4.77 U L-1) and 0.10 mg L-1 CSS (5.37 U L-1). This amounts to approximately 4 AGE molecules trapped inside the polymersomes (encapsulation efficiency26: 0.36 %, statistical encapsulation efficiency27: 86 %, Table S4), 2 NAL molecules and 1 CSS molecule immobilized on the polymersome surface (immobilization efficiency28: 12.5 and 2.7 %, Table S4), and approximately 1 OmpF G119D trimeric channel in 10 polymersomes (integration efficiency29: 1 %). Despite the low amount of OmpF G119D, the nano-EMRs were functional, producing 0.87 mM CMP-Neu5Ac in 93 h from 128 mM GlcNAc, 80 mM pyruvate and 50 mM CTP. To validate that CTP was successfully excluded from the polymersome lumen by OmpF G119D, the reaction cascade was performed in the nano-EMRs at different inhibitor con-

centrations of 3 mM and 50 mM CTP (Figure 2). To compare the two reactions and because the reaction at 3 mM CTP should proceed faster than the reaction at 50 mM CTP due to reduced inhibitory effects, the progress of the reaction is given in relative CMP-Neu5Ac concentration with respect to the final amount of CMP-Neu5Ac that was produced after 93 h with an initial 3 mM CTP. The inhibitory effect of CTP on the AGE is demonstrated when performing the reaction cascade in a non-compartmentalized set-up. The reaction cascade was significantly slowed down at 50 mM CTP reaching a product concentration of approximately 34 % compared to the same reaction performed with an initial 3 mM CTP. This is in good agreement with the expected inhibition of the AGE by CTP, which reduces the AGE activity to 32 % at 50 mM CTP compared to 3 mM CTP (Figure S5). In contrast, the reaction rate of the compartmentalized reaction was successfully uncoupled from the CTP concentration, reaching the same final CMPNeu5Ac concentration with an initial 3 mM and 50 mM CTP. This uncoupling of the reaction rate from the CTP concentration demonstrates a successful elimination of the crossinhibition and is essential for improving the reaction cascade.

Figure 2. Relative CMP-Neu5Ac concentration (Relative product conc.) with 3 mM (white) and 50 mM (black) initial CTP. (A) Non-compartmentalized reaction cascade with immobilized NAL and CSS. (B) Compartmentalized reaction cascade with encapsulated AGE and reconstituted OmpF G119D. Enzyme concentrations were 0.56 mg L-1 AGE, 0.31 mg L-1 NAL and 0.10 mg L-1 CSS. Schematic representations of the respective systems are shown on the right (Blue: AGE; light green: NAL; dark green: CSS; red: OmpF G119D). For each system, the CMP-Neu5Ac concentration that was reached with an initial 3 mM CTP after 93 h was used as 100 % relative product concentration. All data points are means of triplicate measurements.

To assess the functionality of the nano-EMRs with reconstituted OmpF G119D, we compared their activity to the noncompartmentalized reaction cascade and to nano-EMRs without membrane channel and with unselective mass transport through wildtype OmpF (Figure 3). Without compartmentalization and with all three enzymes free in solution, no CMPNeu5Ac was formed in 93 h which was in part due to the strong cross-inhibition of the AGE by CTP and in part due to

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the low stability of the CSS in solution at 30°C with a half-life of 1.5 h (Figure S6). By immobilizing the CSS on the polymersome surface and its concomitant stabilization, polymersomes without encapsulated AGE but with immobilized NAL and CSS and externally added AGE rendered a non-compartmentalized system which was solely governed by the inhibitory effects of CTP on the AGE and resulted in a final product concentration of 0.4 ± 0.1 mM. By encapsulating the AGE but without reconstituting a channel protein, product formation was reduced to 0.22 ± 0.06 mM. Although reasonable product formation was observed, which was due to the selective permeability of the polymer membrane itself, mass transport limitations of GlcNAc and ManNAc across the membrane reduced the product formation compared to the non-compartmentalized reaction. Thus, although the required selectivity was accomplished without additional channel proteins, the mass transport limitations across the membrane had a stronger impact on the AGE activity than did the CTP inhibition. By reconstituting wildtype OmpF into the membrane (2 OmpF in 3 polymersomes; integration efficiency: 7%), we completely alleviated mass transport limitations (Figure S7), however, the reintroduction of the cross-inhibition as well as a loss of the allosteric activator ATP reduced the effectiveness of the nano-EMRs leading to a product formation which was not significantly increased compared to the noncompartmentalized system. In contrast, although statistically only one trimeric channel was incorporated into every tenth polymersome, the diffusion of GlcNAc and ManNAc through the OmpF G119D was sufficient to lower mass transport limitations, increasing the product yield compared to polymersomes without membrane channel by 3.9-fold (paired t-test with two-tailed p < 0.001) to 0.87 ± 0.05 mM. In combination with maintaining the compartmentalization with respect to CTP, product formation could be increased by 2.2-fold compared to the non-compartmentalized system (p = 0.021). Thus, by selectively retaining ATP within the polymersomes while excluding CTP from the lumen, the functionalized nanoEMRs improved the cascade reaction by implementing selective mass transport across the compartment boundaries. Given the fact that the reaction theoretically proceeds to a 100 % yield, the relatively low yields obtained in this study were due to low concentrations of nano-EMRs and, thus, low biocatalyst concentrations in the samples. By increasing the nano-EMR concentration from 0.13 % w/v to 0.24 % w/v, the final CMPNeu5Ac concentration was increased to 1.5 ± 0.2 mM, linearly increasing the space-time-yield by increasing the biocatalyst concentrations.

Page 4 of 10

Figure 3. CMP-Neu5Ac concentration at 50 mM initial CTP after 93 h without compartmentalization with free and immobilized enzymes and with compartmentalized systems. “Soluble” denotes the three enzyme cascade with soluble enzymes and “immobilized” with immobilized NAL and CSS. Statistical significance is given by asterisks (*: p ≤ 0.05; **: p ≤ 0.01; ***: p ≤ 0.001).

We demonstrate that the use of polymersomes as nanoEMRs can be extended to intrinsically incompatible reactions that suffer from cross-inhibitions by compartmentalizing the reaction space. Crucial for a functional system is a selective mass transport across the polymer membrane. Whereas unselective mass transpo rt is able to alleviate mass transport limitations, an enhancement of the reaction is not possible. However, since every conceivable cascade reaction will require a different membrane selectivity depending on the substrates and the inhibitors, PMOXA-PDMS-PMOXA polymersomes are especially suitable due to their generally low permeability and thus the ability to modulate the membrane permeability via membrane channels. Thus, the main limitations we see in this approach are the requirement for specialized and highly selective channels and satisfactory integration efficiencies. Recently, the reconstitution of DNA channels into polymersomes has added an interesting alternative, especially due to the ease of engineering DNA channels.30 However, information on integration efficiencies and channels per polymersome are generally scarce and often phenomenological, stating only the ratio of channel protein to polymer applied. Since the integration efficiency is usually not 100 %, this does not necessarily reflect the amount of channel proteins that is present per polymersome. Integration efficiencies of up to 22.5 % with 19 molecules per polymersome with 110 nm diameter have been reported.31 Habel et al. calculated 2.87 aquaporin Z channels per polymersome.32 Recent unpublished data, however, suggests that especially monomeric protein channels show good integration efficiencies compared to multimeric channel proteins, reaching more than 20 and up to 160 channels per polymersomes.33 This indicates that low channel integration into the polymer membrane is not a general problem and high degrees of functionalization can be obtained. Furthermore, monomerization of multimeric channels can improve the distribution of channel proteins and lead to an easier fine tuning of the membrane permeability, considering that few channels are able to drastically reduce mass transport limitations. In other words, three out of ten active polymersomes with one monomeric OmpF G119D channel may perform better than one highly active polymersome with one OmpF G119D trimer and nine polymersomes with low activity. Thus, the characterization and availability of highly selective membrane channels as well as investigations into increasing the integration efficiency of membrane channels, for example by rationally engineering multimeric channels into monomeric channels34 or by adjusting the length of the transmembrane domains,35 needs to be addressed to generate a toolbox for cascade reactions that can greatly benefit from compartmentalization. In ongoing work we have started investigating several membrane channels, such as the hydrophobic channel AlkL, to extend the scope of polymersomes as nanoEMRs to other incompatible reaction cascades.

Supporting Information. Materials, Methods and Supplementary Results.

ACS Paragon Plus Environment

Page 5 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

This material is available free of charge via the Internet at http://pubs.acs.org.

* Dr. Kathrin Castiglione Address: Institute of Biochemical Engineering, Technical University of Munich, Boltzmannstr. 15, 85748 Garching E-mail: [email protected]

LK led the experiments and drafted the manuscript. SP was responsible for polymersome production and handling. KC conceived of the study, and participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.

The authors declare no competing financial interest.

We thank the BMBF (German Federal Ministry of Education and Research) Grant-No. 031A178 for funding. We thank Tom S. Schwarzer and Florian Sedlmaier for their assistance in the laboratory. The support of LK and STP by the TUM Graduate School is acknowledged. The authors gratefully acknowledge the support of this work by Prof. Dirk Weuster-Botz (Institute of Biochemical Engineering, Technical University of Munich, Garching, Germany).

AGE, N-acyl-D-glucosamine 2-epimerase; ATP, adenosine triphosphate; CMP-Neu5Ac, CMP-N-acetylneuraminic acid; CSS, CMP-sialic acid synthetase; CTP, cytidine triphosphate; GlcNAc, N-acetylglucosamine; ManNAc, N-acetylmannosamine; MMCO, molecular mass cut-off; NAL, N-acetylneuraminate lyase; nanoEMRs, nano-scale enzyme membrane reactors; Neu5Ac, Nacetylneuraminic acid; OmpF, outer membrane protein; PDMS, poly(dimethylsiloxane); PMOXA, poly(2-methyloxazoline).

(1) Sauerzapfe, B.; Elling, L. In Multi-Step Enzyme Catalysis: Biotransformations and Chemoenzymatic Synthesis, Garcia-Junceda, E., Ed.; Wiley-VCH: Weinheim, 2008; pp 83-107. (2) Busto, E.; Simon, R. C.; Richter, N.; Kroutil, W. ACS Catal. 2016, 6, 2393-2397. (3) Denard, C. A.; Hartwig, J. F.; Zhao, H. ACS Catal. 2013, 3, 2856-2864. (4) Discher, D. E.; Eisenberg, A. Science 2002, 297, 967-973. (5) Nardin, C.; Thoeni, S.; Widmer, J.; Winterhalter, M.; Meier, W. Chem. Commun. 2000, 1433-1434. (6) Ranquin, A.; Versees, W.; Meier, W.; Steyaert, J.; Van Gelder, P. Nano Lett. 2005, 5, 2220-2224. (7) Schoonen, L.; van Hest, J. C. M. Adv. Mater. 2016, 28, 11091128. (8) Spulber, M.; Baumann, P.; Saxer, S. S.; Pieles, U.; Meier, W.; Bruns, N. Biomacromolecules 2014, 15, 1469-1475. (9) Peters, R. J. R. W.; Marguet, M.; Marais, S.; Fraaije, M. W.; van Hest, J. C. M.; Lecommandoux, S. Angew. Chem. Int. Ed. 2014, 53, 146-150. (10) Gaitzsch, J.; Appelhans, D.; Wang, L.; Battaglia, G.; Voit, B. Angew. Chem. Int. Ed. 2012, 51, 4448-4451. (11) Gräfe, D.; Gaitzsch, J.; Appelhans, D.; Voit, B. Nanoscale 2014, 6, 10752-10761.

(12) Kim, K. T.; Cornelissen, J. J. L. M.; Nolte, R. J. M.; van Hest, J. C. M. Adv. Mater. 2009, 21, 2787-2791. (13) Siti, W.; de Hoog, H. P. M.; Fischer, O.; Shan, W. Y.; Tomczak, N.; Nallani, M.; Liedberg, B. J. Mater. Chem. B 2014, 2, 2733-2737. (14) Meeuwissen, S. A.; Rioz-Martinez, A.; de Gonzalo, G.; Fraaije, M. W.; Gotor, V.; van Hest, J. C. M. J. Mater. Chem. 2011, 21, 18923-18926. (15) Schmitt, C.; Lippert, A. H.; Bonakdar, N.; Sandoghdar, V.; Voll, L. M. Front. Bioeng. Biotechnol. 2016, 4, 1-12. (16) Klermund, L.; Groher, A.; Castiglione, K. J. Biotechnol. 2013, 168, 256-263. (17) Groher, A.; Hoelsch, K. J. Mol. Catal. B: Enzym. 2012, 83, 17. (18) Warren, L.; Blacklow, R. S. J. Biol. Chem. 1962, 237, 35273534. (19) Knorst, M.; Fessner, W.-D. Adv. Synth. Catal. 2001, 343, 698-710. (20) Yu, H.; Yu, H.; Karpel, R.; Chen, X. Bioorg. Med. Chem. 2004, 12, 6427-6435. (21) Kean, E. L.; Roseman, S. J. Biol. Chem. 1966, 241, 56435650. (22) Poschenrieder, S. T.; Wagner, S. G.; Castiglione, K. J. Appl. Polym. Sci. 2016, 133, 43274-43280. (23) Klermund, L.; Poschenrieder, S. T.; Castiglione, K. J. Nanobiotechnol. 2016, 14:48, 1-12. (24) Jeanteur, D.; Schirmer, T.; Fourel, D.; Simonet, V.; Rummel, G.; Widmer, C.; Rosenbusch, J. P.; Pattus, F.; Pages, J. M. Proc. Natl. Acad. Sci. U. S. A. 1994, 91, 10675-10679. (25) Saint, N.; Lou, K. L.; Widmer, C.; Luckey, M.; Schirmer, T.; Rosenbusch, J. P. J. Biol. Chem. 1996, 271, 20676-20680. (26) The encapsulation efficiency is the mass ratio of protein encapsulated in polymersomes to protein added during vesicle formation. (27) The statistical encapsulation efficiency is the amount of protein encapsulated compared to the amount of protein that can be maximally encapsulated at the given concentration. (28) The immobilization efficiency is the mass ratio of protein immobilized on polymersomes to total protein added during protein immobilization. (29) The integration efficiency is the mass ratio of protein integrated into the polymer membrane to protein added during vesicle formation. (30) Messager, L.; Burns, J. R.; Kim, J.; Cecchin, D.; Hindley, J.; Pyne, A. L.; Gaitzsch, J.; Battaglia, G.; Howorka, S. Angew. Chem. Int. Ed. 2016, 55, 11106-11109. (31) Itel, F.; Najer, A.; Palivan, C. G.; Meier, W. Nano Lett. 2015, 15, 3871-3878. (32) Habel, J.; Hansen, M.; Kynde, S.; Larsen, N.; Midtgaard, S.; Jensen, G.; Bomholt, J.; Ogbonna, A.; Almdal, K.; Schulz, A.; HélixNielsen, C. Membranes 2015, 5, 307-351. (33) Unpublished data. (34) Naveed, H.; Jimenez-Morales, D.; Tian, J.; Pasupuleti, V.; Kenney, L. J.; Liang, J. J. Mol. Biol. 2012, 419, 89-101. (35) Muhammad, N.; Dworeck, T.; Fioroni, M.; Schwaneberg, U. J. Nanobiotechnol. 2011, 9:8, 1-9.

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Paragon Plus Environment

Page 6 of 10

6

Page 7 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Figure 1. (A) Three-step synthesis of CMP-N-acetylneuraminic acid (CMP-Neu5Ac). (B) Compartmentalized reaction scheme in polymer-somes. The AGE is separated from the incompatible component CTP by encapsulation in polymersomes. The NAL and the CSS are im-mobilized on the surface to form a single biocatalytic entity. Channel proteins enable a selective mass transport across the polymer membrane. 126x95mm (300 x 300 DPI)

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Relative CMP-Neu5Ac concentration (Relative product conc.) with 3 mM (white) and 50 mM (black) initial CTP. (A) Non-compartmentalized reaction cascade with immobilized NAL and CSS. (B) Compartmentalized reaction cascade with encapsulated AGE and reconstituted OmpF G119D. Enzyme concentrations were 0.56 mg L-1 AGE, 0.31 mg L-1 NAL and 0.10 mg L-1 CSS. Schematic representations of the respective systems are shown on the right (Blue: AGE; light green: NAL; dark green: CSS; red: OmpF G119D). For each system, the CMP-Neu5Ac concentration that was reached with an initial 3 mM CTP after 93 h was used as 100 % relative product concentration. All data points are means of triplicate measurements. 82x82mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 8 of 10

Page 9 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

CMP-Neu5Ac concentration at 50 mM initial CTP after 93 h without compartmentalization with free and immobilized enzymes and with compartmentalized systems. “Soluble” denotes the three enzyme cascade with soluble enzymes and “immobilized” with immobilized NAL and CSS. Statistical significance is given by asterisks (*: p ≤ 0.05; **: p ≤ 0.01; ***: p ≤ 0.001). 55x37mm (300 x 300 DPI)

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

288x249mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 10 of 10