Immobilized enzyme reactors based on nucleoside phosphorylases and 2′-deoxyribosyltransferase for the in-flow synthesis of pharmaceutically relevant nucleoside analogues

Highlights

• An in-flow enzymatic system based on immobilized NPs and NDT was developed.

• LrNDT and CpUP/AhPNP were immobilized on monolithic silica columns.

• The two immobilized enzyme reactors (IMERs) catalyzed continuous reactions in flow.

• Pure nucleoside analogues were synthesized in suitable amounts for drug discovery.

• Both bioreactors are promising alternatives to nucleoside chemical synthesis.

Abstract

In this work, a mono- and a bi-enzymatic analytical immobilized enzyme reactors (IMERs) were developed as prototypes for biosynthetic purposes and their performances in the in-flow synthesis of nucleoside analogues of pharmaceutical interest were evaluated. Two biocatalytic routes based on nucleoside 2′-deoxyribosyltransferase from Lactobacillus reuteri (LrNDT) and uridine phosphorylase from Clostridium perfrigens (CpUP)/purine nucleoside phosphorylase from Aeromonas hydrophila (AhPNP) were investigated in the synthesis of 2′-deoxy, 2′,3′-dideoxy and arabinonucleoside derivatives. LrNDT-IMER catalyzed the synthesis of 5-fluoro-2′-deoxyuridine and 5-iodo-2′-deoxyuridine in 65–59% conversion yield, while CpUP/AhPNP-IMER provided the best results for the preparation of arabinosyladenine (60% conversion yield).

Both IMERs proved to be promising alternatives to chemical routes for the synthesis of nucleoside analogues. The developed in-flow system represents a powerful tool for the fast production on analytical scale of nucleosides for preliminary biological tests.

Introduction

The use of enzymes for the synthesis of active pharmaceutical ingredients (APIs) represents an interesting alternative to classical chemical routes. In particular, the intrinsic selectivity of enzymes allows the reduction of synthetic steps and their natural ability to work under mild conditions makes the production bioprocess eco-friendly (Busacca et al., 2011, Pollard and Woodley, 2007, Truppo, 2017, Woodley, 2008).

In addition, the immobilization of the enzyme to a solid support allows its stabilization and, consequently, the exploitation of more severe conditions (pH, temperature, solvents), as well as the reuse of the biocatalyst for multiple cycles (Bernal et al., 2018, Sheldon and Woodley, 2018).

Immobilized enzymes can be used in batch or in a system operating under continuous flow conditions (in-flow). In-flow reactions offer advantages compared to those performed in batch, resulting in an increase in productivity. Moreover, flow reactors can be prepared on different scales (from analytical to production scale). For the development of bioreactors on analytical scale, enzyme immobilization is generally performed directly in the column, thus creating an immobilized enzyme reactor (IMER) (Fang et al., 2012). IMERs can be used for rapid preparation of small quantities (microscale) of new products as required for drug discovery (Britton et al., 2018, Fang et al., 2012, Girelli and Mattei, 2005, Tamborini et al., 2018).

The choice of the immobilization carrier is an important issue to preserve enzymatic activity. In this context, monoliths emerged as interesting carriers for the preparation of IMERs due to their attractive features: high permeability and low back-pressure; accessibility of the immobilized macromolecule thanks to the high porosity; possibility to use different immobilization methods and chemistries; stability (Calleri et al., 2012, Vlakh and Tennikova, 2013a, Vlakh and Tennikova, 2013b). For these reasons, monolithic supports represent an advancement compared to more common beads and particle columns.

In recent years, in-flow IMERs have been applied to the synthesis of different classes of APIs or pharmaceutical intermediates (Britton et al., 2018, Naldi et al., 2018, Tamborini et al., 2018). Among APIs, nucleoside analogues represent intriguing targets since their biocatalyzed synthesis can overcome different drawbacks of the classical chemical process used for their production (Ding et al., 2010, Fresco-Taboada et al., 2013, Mikhailopulo, 2007). Nucleoside phosphorylases (NPs) have proven their potential for the synthesis of modified nucleosides, which find applications as antiviral and antitumor agents. NPs can act on ribo- or 2′-deoxyribosylnucleosides by a reversible phosphorolytic reaction, resulting in the formation of the corresponding nucleobase and glycosyl moiety. The NP-catalyzed transfer of the sugar residue to a second nucleobase leads to the production of a new nucleoside (transglycosylation reaction) (Cattaneo et al., 2017, Fresco-Taboada et al., 2013). NPs are classified into families that possess different substrate specificity (for purine or pyrimidine nucleosides). Pyrimidine nucleoside phosphorylases have been successfully employed in the pyrimidine-pyrimidine transglycosylation reaction (Serra et al., 2013a, Serra et al., 2011), while the coupling of two NPs is often necessary to obtain the desired nucleoside when the transglycosylation occurs between a purine and a pyrimidine base. Recently, using in-flow bioreactors, the purine-purine transglycosylation catalyzed by a purine nucleoside phosphorylase from Aeromonas hydrophila (AhPNP) has been investigated (Calleri et al., 2015). Moreover, the in-flow bi-enzymatic tranglycosylation has also been described by connecting in series two bioreactors containing a uridine phosphorylase from Clostridium perfrigens (CpUP) and AhPNP (Cattaneo et al., 2017).

A different class of enzymes, 2′-deoxyribosyltransferases (NDTs), has also been used in batch for the synthesis of modified nucleosides (Fernández-Lucas et al., 2013, Fernández-Lucas et al., 2012, Fernández-Lucas et al., 2011, Fernández-Lucas et al., 2010). Reactions catalyzed by NDTs consist in the exchange between a nucleobase of a 2′-deoxyribonucleoside and a free nucleobase in one step, with regio- and stereoselectivity (Fernández-Lucas et al., 2010, Fresco-Taboada et al., 2013, Huang et al., 1983). Two classes of NDTs can be distinguished: NDT type I (PDT), which catalyzes the deoxyribose group transfer exclusively between purines, and NDT type II (NDT), which catalyzes the transfer between purines and/or pyrimidines (Becker and Brendel, 1996, Holguin and Cardinaud, 1975, Kaminski, 2002).

The mechanism of NDTs shows to be similar to retaining glycoside hydrolases. Glycosyltransfer reaction occurs via double-displacement mechanism involving the formation of a covalently bound glycosyl-enzyme intermediate. In addition, NDTs perform acid/base catalysis similarly to glycosidases (Danzin and Cardinaud, 1974, Del Arco et al., 2019, Fresco-Taboada et al., 2018, Fresco-Taboada et al., 2013, Huang et al., 1983, Porter et al., 1995, Short et al., 1996).

Among the described NDTs, nucleoside 2′-deoxyribosyltransferase from Lactobacillus reuteri (LrNDT) has been revealed as suitable biocatalyst for the synthesis of nucleoside analogues with therapeutic activity, showing the highest specific activity as well as catalytic efficiency (Fernández-Lucas et al., 2011, Fernández-Lucas et al., 2010). NDTs proceed through a ping-pong bi-bi mechanism. In the first step, the catalytic Glu residue attacks the anomeric C1′ releasing the nucleobase, which leads to the glycosylated intermediate. In a second step, the second nucleobase attacks the glycosylated intermediate (transglycosylation) to generate the corresponding nucleoside. Interestingly, a hydrolase function was also described in absence of nucleobase acceptors (Smar et al., 1991) or at long reaction times (Huang et al., 1983).

In this work, the preparation of an LrNDT-IMER on a monolithic epoxy silica column is described for the first time. The obtained bioreactor has been applied to the synthesis of selected nucleosides of pharmaceutical interest. In addition, the co-immobilization of CpUP and AhPNP has also been investigated using a monolithic aminopropyl silica carrier. The two different bioreactors were included in a chromatographic system for in-flow reactions and tested in the preparation of different nucleosides on analytical scale in order to find the most adequate system for the synthesis of analogues modified at the base and/or at the sugar moiety.