ATP regeneration by a single polyphosphate kinase powers multigram-scale aldehyde synthesisin vitro

Article abstract of DOI:10.1039/d0gc03830j

ATP recycling systems are required to avoid the addition of stoichiometric quantities of cofactor and facilitate industrial implementation of ATP-dependent enzymes. One factor that limits the biocatalytic application of these enzymes is the lack of a scalable AMP to ATP regeneration system. Whole-cells or a combination of purified enzymes are often exploited for ATP regeneration from AMP, whereas cell free systems comprising a single crude enzyme preparation would be preferred. To establish such a system, we focussed on polyphosphate kinases (PPKs) to find a single enzyme that could be used to power ATP-consuming reactions. Screening of some previously reported PPKs revealed limitations of these biocatalysts for scale-up purposes. As such, a panel of novel putative PPK2-III enzymes was constructed and compared to characterised enzymes belonging to the same class. Multidimensional small-scale screening revealed that PPK12 (from an unclassifiedErysipelotrichaceaebacterium) displays enhanced expression levels, ATP formation rates, polyphosphate tolerance and stability under a variety of harsh conditions. The carboxylic acid reductase (CAR) catalysed reduction of carboxylates to aldehydes was chosen as a model reaction to test the applicability of PPK12 as a bifunctional biocatalyst for ATP regeneration from AMP. The implementation of the identified ATP-recycling enzyme provided the first example of cell free multigram-scale aldehyde synthesis employing enzymes and a single PPK2-III, paving the way for affordable scalable ATP regeneration technologies.

Full text of DOI:10.1039/d0gc03830j

View Article Online
View Journal
Green
Chemistry
Cutting-edge research for a greener sustainable future
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: M. Tavanti, J.
Hosford, R. Lloyd and M. Brown, Green Chem., 2021, DOI: 10.1039/D0GC03830J.
Volume 20
Number
May 2018
Pages 1919-2160
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
9
7
Green
Chemistry
Cutting-edge research for a greener sustainable future
rsc.li/greenchem
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 1463-9262
PAPER
Paul T. Anastas et al.
The Green ChemisTREE: 20 years after taking root with the 12
principles
rsc.li/greenchem

Page 1 of 11
Please Gd roe en no tC ha ed mj u iss tt r my argins
View Article Online
DOI: 10.1039/D0GC03830J
ARTICLE
ATP Regeneration by a Single Polyphosphate Kinase Powers
Multigram-Scale Aldehyde Synthesis In Vitro
a
a
b
a
Michele Tavanti,* Joseph Hosford, Richard C. Lloyd and Murray J. B. Brown
Received 00th January 20xx,
Accepted 00th January 20xx
ATP recycling systems are required to avoid the addition of stoichiometric quantities of cofactor and facilitate industrial
implementation of ATP-dependent enzymes. One factor that limits the biocatalytic application of these enzymes is the lack
of a scalable AMP to ATP regeneration system. Whole-cells or a combination of purified enzymes are often exploited for ATP
regeneration from AMP, whereas cell free systems comprising a single crude enzyme preparation would be preferred. To
establish such a system, we focussed on polyphosphate kinases (PPKs) to find a single enzyme that could be used to power
ATP-consuming reactions. Screening of some previously reported PPKs revealed limitations of these biocatalysts for scale-
up purposes. As such, a panel of novel putative PPK2-III enzymes was constructed and compared to characterised enzymes
belonging to the same class. Multidimensional small-scale screening revealed that PPK12 (from an unclassified
Erysipelotrichaceae bacterium) displays enhanced expression levels, ATP formation rates, polyphosphate tolerance and
stability under a variety of harsh conditions. The carboxylic acid reductase (CAR) catalysed reduction of carboxylates to
aldehydes was chosen as a model reaction to test the applicability of PPK12 as a bifunctional biocatalyst for ATP regeneration
from AMP. The implementation of the identified ATP-recycling enzyme provided the first example of cell free multigram-
scale aldehyde synthesis employing enzymes and a single PPK2-III, paving the way for affordable scalable ATP regeneration
technologies.
DOI: 10.1039/x0xx00000x
purposes, as kinetics favour the polyP-driven nucleoside 5’-
diphosphate or nucleoside 5’-triphosphate synthesis rather
than degradation.
Introduction
8
, 9
So far, the vast majority of ATP regeneration systems start
from ADP, while a range of synthetically useful biocatalytic
reactions such as reduction of carboxylates to aldehydes,
ATP-dependent kinases, ligases, transferases and
oxidoreductases are valuable enzymes in the biocatalytic
toolbox. A major factor to be considered when applying these
enzymes is cofactor regeneration, as their stoichiometric
addition brings a number of issues associated with process costs
and sustainability, cofactor by-product inhibition and
1
0
1
1-14
amide bond formation
and cascades for S-
1
5
Adenosylmethionine-dependent alkylations would require
ATP regeneration from AMP. Most AMP to ATP regeneration
systems employ two enzymes, such as adenylate kinase/acetate
1
, 2
downstream processing.
Well-established ATP recycling
1
6
6, 17
18-22
kinase, adenylate kinase/PPK,
two-PPK systems,
or
systems starting from ADP such as acetate kinase/acetyl
more complicated mixtures such as E. coli lysate supplemented
3
4
phosphate, pyruvate kinase/phosphoenolpyruvate and the
2
3, 24
25-27
with acetyl phosphate
and whole cells.
Considering the
5
less exploited creatine kinase/creatine phosphate bring
cost contribution of biocatalysts to the process, the
minimisation of the number of enzymes employed and their
implementation in the crudest form are preferred. As such,
disadvantages concerning cost, stability and availability of the
phosphate donor.6 As such, the polyphosphate kinase
(PPK)/polyphosphate (polyP) ATP regeneration system has
2
8
whole-cell processes are often reported in the literature, even
though the construction of such in vivo cascades can bring some
disadvantages in terms of mass-transfer limitations, product
toxicity to the host organism and complicated downstream
attracted considerable interest, given the stability and low cost
7
of the phosphate donor employed. Two major PPK families
have been described: PPK1 and PPK2. Enzymes belonging to the
latter family are typically employed for ATP regeneration
2
6, 29
processing.
Given the complexity associated with ATP regeneration
from AMP, recently characterised bifunctional PPKs catalysing
ATP formation from AMP have attracted considerable
attention. Phylogenetic analysis led to the characterisation of
three PPK2 subfamilies (classes) with distinct structural and
a. Synthetic Biochemistry, Medicinal Science and Technology, Pharma R&D
GlaxoSmithKline Medicines Research Centre
Gunnels Wood Road, Stevenage SG12NY, U.K.
Email: michele.x.tavanti@gsk.com
Chemical Development, Medicinal Science and Technology, Pharma R&D
b.
GlaxoSmithKline Medicines Research Centre
Gunnels Wood Road, Stevenage SG12NY, U.K.
3
0-32
biochemical properties.
Generally, PPK2-I and PPK2-II
†
Electronic Supplementary Information (ESI) available: supplementary tables and substrate preference leads to nucleoside 5’-triphosphates from
Fig.s. See DOI: 10.1039/x0xx00000x
the corresponding nucleoside 5’-diphosphates and nucleoside
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 2 of 11
ARTICLE
Journal Name
5
’-diphosphates from nucleoside 5’-monophosphates, regeneration system. The preparative scale VsiyewntAhrteiclseisOnlionef
respectively. Remarkably, PPK2-III are bifunctional enzymes aldehydes from carboxylic acids undeDrOI:m10il.d103c9o/Dn0dGitCio0n38s30iJs
forming both nucleoside-5’ diphosphates and nucleoside 5’- significant and provides building blocks for subsequent
1
0
triphosphates (Fig. 1).
functionalisation. Traditional chemical synthesis of aldehydes
Interestingly, Andexer and co-workers demonstrated that is often plagued by poor atom and step economy, harsh
the catalytic activity of some PPK2 enzymes extends beyond the reaction conditions and unwanted side reactions such as
4
0
previously reported substrate preference (Fig. 1). For example, overreduction to the respective alcohol.
Conversely,
PPK2-I and PPK2-II representative enzymes could accept AMP structural and mechanistic features of CARs ensure that
3
3
as substrate for ATP generation. Additionally, PPK2 enzymes substrate overreduction to the corresponding alcohol does not
4
1
displayed adenylate kinase activity (conversion of 2 ADP take place. This desirable feature along with the wide
3
2, 33
molecules to AMP and ATP)
adenosine 5’-polyphosphates synthesis from AMP.
In view of the drawbacks associated with in vivo processes,
and were found to catalyse substrate scope and the benign reaction conditions, make CARs
3
3
42,
promising biocatalysts in the toolbox of the organic chemist.
4
3
However, their successful large scale in vitro application is
in vitro systems comprising purified proteins or crude enzyme hindered by the need of stoichiometric ATP. Reaction
preparations are becoming more popular. Few in vitro intensification with the CAR/PPK system highlighted limitations
biocatalytic transformations incorporating bifunctional PPK2-III of the initially identified ATP regeneration module. In the quest
have been reported, including an enzymatic cascade for cell free for a better enzyme for polyP-driven ATP regeneration starting
protein synthesis,34 amide bond formation reactions via a from AMP, a panel of novel putative PPK2-III enzymes was
3
5
thermophilic CoA ligase or via the adenylation domain of assembled (82 enzymes) and compared with previously
3
6
37
either tyrocidine synthetase or carboxylic acid reductase,
and a challenging carboxylate reduction/amination cascade.
reported enzymes belonging to the same class (9 enzymes).
Eventually, a newly identified bacterial PPK2-III enzyme enabled
3
8
However, successful in vitro ATP regeneration from AMP with the cell-free carboxylate reduction to aldehyde on multigram-
bifunctional PPKs was mainly shown in small scale experiments scale, a significant step towards scalable ATP recycling.
using purified proteins, whereas larger scale reaction
demonstrations with crude enzyme preparations are required
to move towards industrial applicability of this ATP
Experimental section
regeneration system.
Herein, we set out to explore PPK2 enzymes across different
classes to identify an enzyme with desirable activity, stability
For full experimental details please refer to the ESI.
and polyP tolerance when applied as lyophilised cell-free Results and discussion
extract for AMP to ATP synthesis and ATP recycling. The
characterisation of an enzyme with these features would open
up the door to a range of valuable biocatalytic cascades, whose
large-scale demonstration can lead to a step change in chemical
manufacturing.39 Screening of a small collection of reported
Initial PPK2 Enzymes Collection: Enzyme Selection and Screening
An initial panel of 11 PPK2 enzymes including characterised
and uncharacterised sequences covering all three PPK2 classes
was assembled (Table S1). In particular, putative PPK2
sequences were obtained from extremophile organisms to
increase the likelihood of identifying protein backbones
adapted to challenging reaction conditions or stable proteins
PPK2
enzymes
including
polyphosphate:AMP
phosphotransferases (PAP) led to the selection of PPK2-II from
Acinetobacter johnsonii (ajPAP) as the most suitable known
single biocatalyst for ATP regeneration.
4
4-46
amenable to protein engineering.
Even though our specific
aim was to find a single biocatalyst for AMP to ATP
regeneration, we decided to include enzymes other than PPK2-
III biocatalysts as it was recently shown that PPK2-I and PPK-II
Carboxylate to aldehyde reduction at the expense of ATP
and NADPH catalysed by carboxylic acid reductases (CARs) was
chosen as a model reaction to test the selected AMP to ATP
3
3
enzymes can phosphorylate AMP to ATP via ADP.
PPK2-III
PPK2-III
PPK2-II
PPK2-I
Fig. 1 PPK2 subfamilies and preferred catalysed reaction(s) related to ATP regeneration.
2
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 11
Please Gd roe en no tC ha ed mj u iss tt r my argins
View Article Online
DOI: 10.1039/D0GC03830J
ARTICLE
Enzymes were produced in E. coli and tested, unless
otherwise specified, as lyophilised cell free extract as the
preferred biocatalyst formulation due to the ease of
time. Thus, this first experiment demonstrated that the
polyP-driven AMP to ATP synthesis using a single enzyme is
possible with a variety of PPK2 enzymes belonging to
different classes.
4
7, 48
preparation, application and storage.
Considering the
biocatalyst production method, host background activities
supporting ATP regeneration were also expected. For initial
screening, substrate depletion and product formation were
followed for 180 minutes in reactions containing 0.1 mg/mL
lyophilised cell free extract, 1 mM AMP and 50 mM polyP
Subsequent screening reactions were designed to mimic
the envisioned process conditions, especially with respect to
PPK and polyP loadings. Reactions powered by polyP-based
ATP regeneration systems typically employ a stoichiometric
excess of polyP with respect to the substrate to be converted
by the ATP-consuming enzyme in order to increase product
6
(based on phosphate units). Not only PPK2-III, but also PPK2-
1
9, 35, 49-51
I and PPK2-II enzymes were capable of AMP phosphorylation
to ATP (Fig. 2). Low activity and low expression levels (Fig.
S1) were observed for tePPK and the previously
uncharacterised mhPAP and hrPPK, with a maximum 30%
AMP conversion to ADP observed after 180 minutes. AMP
conversion was 50% or higher for other enzymes in the panel
after 15 minutes reaction. Notably, ~70% conversion to ATP
was obtained after 180 minutes reaction with ajPAP, rmPPK,
rpPPK, drPPK and mrPPK, with about 30-40% conversion to
ATP for ajPAP, rpPPK and drPPK within 15 minutes reaction
yield.
However, excessive polyP was shown to
3
5, 36, 51,
inhibit PPKs and/or the coupled ATP-utilising enzyme.
5
2
As such, higher polyP loadings were employed in the
following screening experiments to highlight potential
limitations. Biocatalyst loading was initially increased to 1
mg/mL and a variety of AMP concentrations were
investigated to compare ATP formation rates.
The best performing enzymes identified after the initial
AMP conversion experiment were tested at different AMP
loadings (20, 40 and 100 mM) with 200 mM polyP.
Fig 2. Reaction time course for AMP phosphorylation with different PPKs. Reactions (1 mL biotransformation volume, 30 °C, 800 rpm) were carried out in 50 mM
potassium phosphate (KPi) buffer pH 7.5 and contained 0.1 mg/mL lyophilised cell free extracts, 1 mM AMP, 50 mM polyP (based on phosphate units) and 100 mM
MgCl2.
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 4 of 11
View Article Online
DOI: 10.1039/D0GC03830J
ARTICLE
Gratifyingly, the selected enzymes tolerated the polyP loading average ATP formation rate observed in these initial screening
applied under these reaction conditions: ajPAP displayed the reactions could be considered as an estimate of the product
highest ATP formation rates across different AMP formation rate achievable with a coupled ATP-utilising enzyme
concentrations, generating up to 35 mM ATP within 180 for an ideal system that is not affected by kinetic, stability
minutes reaction time starting from 100 mM AMP (Fig. S2). and/or mass-transfer limitations. The extrapolated value for
Under the same reaction conditions and reaction time, drPPK product formation rate (0.18 mM/min) is only 2.5-fold lower
generated 25 mM ATP, whereas rmPPK and rpPPK gave than the desired value (0.46 mM/min), a promising result
approximately 10 mM ATP. Additional experiments were considering the ambitious biocatalyst loading target that was
performed to investigate the impact of increasing polyP set for these calculations. However, these estimates should be
loadings. Sequential polyP dosing was adopted by first intent to interpreted with caution, as a number of assumptions were
5
1
minimise the risk of substrate inhibition. The relative made to gauge biocatalyst performance.
performance of the selected biocatalysts mirrored the results
obtained in the preceding experiment (Fig. 3). As a result of the
Carboxylate Reduction to Aldehyde Enabled by ATP recycling
from AMP
At this stage, the utility of the selected ATP production
system for ATP recycling purposes remained to be clarified as
the rate of formation of high concentrations of ATP from a high
concentration of AMP is not directly related to the rate of
recycling of ATP from a low fixed concentration of AMP. To
answer this question, we focussed on carboxylic acid reductases
increased polyP and enzyme loading (10 mg/mL lyophilised cell
free extracts), up to approximately 80 mM ATP could be formed
within 180 minutes by ajPAP and the obtained ATP yield
increased noticeably when additional polyP was supplied (Fig.
S3). Overall, this set of experiments led to the selection of ajPAP
as a single biocatalyst for the polyP driven ATP synthesis starting
from AMP.
Approximate threshold values for process metrics can be
used to evaluate enzyme performance and to guide the
(CARs) as a testbed to investigate polyP-based ATP recycling.
5
3, 54
optimisation of biocatalytic processes (Table S2).
The
Fig 3. Reaction time course for intensified AMP phosphorylation reactions with different PPKs. Reactions (1 mL biotransformation volume, 30 °C, 800 rpm) were carried
out in 50 mM KPi buffer pH 7.5 and contained 10 mg/mL lyophilised cell free extracts, 200 mM AMP, 100 mM MgCl and 500 mM polyP (added over three doses at 15,
0 and 60 minutes, see arrows).
2
3
Please do not adjust margins

Page 5 of 11
Please Gd roe en no tC ha ed mj u iss tt r my argins
View Article Online
DOI: 10.1039/D0GC03830J
ARTICLE
The intent of these experiments was to demonstrate were observed at higher conversion which ultimately led to
scalability and highlight limitations of polyP-based AMP to ATP lower product yield when less ajPAP was employed. This
regeneration systems employing a single PPK, rather than trying reaction trajectory suggests that limited ajPAP stability affects
6
2
to outperform chemical methods for carboxylate reduction.
the performance of the biocatalytic system.
These promising results represent an unprecedented
Recently, small scale in vitro CAR reactions with dual
cofactor recycling have been shown. The well-established intensification and scale for an in vitro ATP regeneration system
glucose dehydrogenase (GDH)/glucose and formate comprising a single enzyme, however further work was required
dehydrogenase/formate systems were typically employed for to find a suitable PPK with better activity and stability compared
NADPH regeneration, whereas two-enzymes systems were the to ajPAP in order to allow the implementation of the polyP-
2
2, 27, 50, 55-59
preferred option for ATP recycling.
Having selected ajPAP as single biocatalyst for in vitro ATP industrial use.
recycling, the reduction of 4-methoxybenzoic acid 1 to 4-
methoxybenzaldehyde 2 by a carboxylic acid reductase from an PPK2-III Panel: Identification and Screening
in-house CAR panel (CAR33) was chosen as model reaction To identify further enzymes with high ATP synthetic activity
Scheme 1). Glucose dehydrogenase was employed for NADPH from AMP, an enzyme acquisition was centred on PPK2-III
driven recycling system at lower biocatalyst loading suitable for
.
(
recycling Similarly to the evaluated PPKs, CAR33 was employed enzymes. Identified PPK2 containing proteins (28000
as a lyophilised cell free extract for ease of use and a sequences, InterPro IPR022488) were used as starting point for
carboxylate/polyP feeding approach was implemented. Small phylogenetic analysis (Fig. S8). The differential clustering of
3
0
scale reactions were carried out at different substrate and polyP PPK2 sequences belonging to different classes was exploited
loadings. The polyP:substrate molar ratio was maintained at 5:1 to retrieve putative PPK2-III encoding sequences (4004
in initial experiments to ensure that the system was not limited sequences) from which a panel of novel putative PPK2-III
by polyP availability.4 A maximum concentration of 47.5 mM sequences was generated. Genes were obtained from Bacteria
products with 1 mM adenine nucleotides was obtained for the (27 sequences), Archaea (26 sequences) and metagenomes (29
polyP driven process (Fig. S4). These encouraging results sequences), with a preference towards sequences originating
demonstrated that ajPAP can be used not only to synthesise from extremophiles. Additionally, previously reported PPK2-III
ATP from AMP but also to enable multiple AMP to ATP recycles enzymes were included in the panel to have a comprehensive
9
under moderately intense substrate loading conditions
comparison between these biocatalysts (Table S3).
Substantial aldehyde overreduction has been observed in Primary screening conditions and selection were designed to
2
7,
processes employing growing E. coli cells overexpressing CAR
identify enzymes with good expression levels and AMP-to-ATP
and up to 2-8% alcohol was detected in CAR reactions using phosphorylation activity without significant assistance from
purified proteins in combination with a crude GDH preparation host proteins to allow for lower lysate loadings.
6
0
5
0
under unoptimized conditions. Host background enzymatic
activities rather than CARs are responsible for reduction to the
6
1
unwanted alcohol (Fig. S5 and Fig. S6). A considerable amount
of aldehyde overreduction was also observed in our
experiments, as we deliberately employed crude enzyme
preparations. As such, subsequent reaction optimisation
focussed on minimising the amount of aldehyde overreduction.
Dramatic improvements in alcohol:aldehyde ratio were
achieved after implementation of a biphasic reaction system
comprising toluene to facilitate in situ product removal and
minimise aldehyde exposure to host enzymes (Fig. S4).
Eventually, a maximum concentration of 43 mM products was
achieved and notably only 5% alcohol overreduction product
was detected at the highest ajPAP loading tested (Fig. S7).
Follow-up catalyst loading studies conducted by varying
ajPAP concentration at fixed CAR/GDH loadings suggested that
the ATP recycling system was not rate limiting during the first 2
Gluconic Acid
Glucose
GDH
O
NADP + H+
+
NADPH
O
OH
O
O
CAR33
PPK
1
2
ATP
AMP + PPi
polyP
Scheme 1 Biocatalytic aldehyde preparation with dual cofactor recycling. NADPH
is recycled through glucose oxidation by glucose dehydrogenase (GDH), whereas
h of reaction (Fig. S7). However, differences in reaction rate AMP is converted back to ATP with a single PPK. Pyrophosphate (PPi) is generated
as by product of the CAR reaction.
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 6 of 11
ARTICLE
Journal Name
Thus, enzymes were produced in E. coli and high-throughput data obtained after pre-incubation at pH 7.5 witVhieworArtwiclietOhnoliunet
protein purification was performed. Successful protein organic solvents gives an indication of thDeOaIp: 1p0l.i1c0a3b9i/lDit0yGoCf0t3h8e3s0eJ
purification was observed for 58% of the proteins in the panel enzymes in the presence of DMSO or toluene. Generally, the
(Fig. S9). The purified proteins were employed in conversion addition of DMSO was well tolerated by ajPAP, chPPK, drPPK
experiments with 5 mM AMP and product formation monitored and PPK11, whereas the inclusion of 30% v/v DMSO had a
by Ultra-Performance Liquid Chromatography (UPLC, Fig. S10). destabilising on PPK12 after 4 h incubation. Pre-incubation in a
A considerable fraction of previously uncharacterised enzymes two-phase system comprising toluene (50% v/v) did not lead to
(65%) showed more than 20% AMP conversion to ATP after 22 a noticeable loss of PPK12 and drPPK activity, while the addition
h, whereas 16% enzymes (13 novel enzymes) showed higher of a toluene overlay had a stabilising effect on chPPK. Notably,
than 50% AMP conversion ATP in just 15 minutes (Fig. S11). a more significant fraction of ajPAP activity was lost after 4 h
Additionally, unknown peaks (maximum 30% conversion) were pre-incubation in the presence of toluene compared to pre-
detected after prolonged incubation times with 10 of the incubation in aqueous conditions at pH 7.5.
enzymes in the panel. These peaks were tentatively assigned to
adenosine 5’-tetraphosphate and adenosine
Overall, PPK12 displayed more desirable activity and
5’- stability traits, suggesting it may make a more versatile
pentaphosphate based on retention times and recent biocatalyst compared to other enzymes in the panel. The gene
3
3
observations made by Andexer and coworkers. Follow-up for PPK12 (UniProt: A0A3D5XRJ5) is from an unclassified
experiments with reduced purified enzyme loadings led to the Erysipelotrichaceae bacterium, and sequence alignment to
selection of 7 enzymes for subsequent work based on ATP characterised PPK2-III considered in this study (Fig. S15, ~40%
formation rates and AMP conversion to ATP after 22 h (Fig. S13 sequence identity) reveals that the enzyme sequence shows
and Table S5). Peaks corresponding to unknown species were motifs involved in nucleotide and polyP binding (Walker A and
not detected under these reaction conditions.
Walker B domains), conservation of charged and aromatic
A second screening round was designed to find versatile amino acids in the region corresponding to the flexible lid
enzymes usable under a variety of conditions and tolerant of module, as well as the PPK2-III signature residue
high polyP loadings while retaining good AMP to ATP (Glu126PPK12).31, 32
phosphorylation activity. The aim was to identify a stable and
active biocatalyst for ATP recycling that can function under a
Table 1 %ATP obtained with selected PPK2 enzymes at different polyP loadings after 120
range different reaction conditions. Biocatalyst production was
minutes reactions (Table 1A) and %ATP obtained after 15 minutes reactions with
enzymes pre-incubated for 24h under different conditions (Table 1B). Reactions (0.4 mL
conducted in shake flasks and lyophilised cell free extracts
prepared for subsequent application. Firstly, activity was biotransformation volume, 30 °C, 800 rpm) were carried out in 50 mM Tris buffer pH 7.5
assayed with 5 mM AMP and different polyP loadings (Table 1A and contained 50 μg/mL lyophilised cell free extracts, 5 mM AMP, 50 mM MgCl
2
and 100-
00 mM polyP (100 mM polyP for stability assays, Table 1B). %ATP lower than 0.3% were
below the limit of quantitation for this assay. n.d. = not detected.
4
and Fig. S14, top). ATP formation was observed for all enzymes
tested with 100 mM polyP, with chPPK, PPK11 and PPK12
converting ~70% AMP to ATP within 120 minutes. The relative
performance of these enzymes as lyophilised cell-free extracts
aligned well with the purified proteins within 15 minutes
reaction time (Fig. S13). Generally, ATP formation rates dropped
at higher polyP loadings: AMP to ATP conversion fell from 74%
with 100 mM polyP to 16% with 400 mM polyP with chPPK,
whereas AMP conversion decrease was less significant for
PPK11 and PPK12. Specifically, PPK12 displayed superior polyP
tolerance, with AMP to ATP conversion falling from 72% to 45%
moving from 100 mM polyP to 400 mM polyP.
A
polyP (mM)
200
PPK
100
400
aaPPK
chPPK
drPPK
PPK11
PPK12
PPK62
PPK82
49.7
74.0
46.3
74.2
72.1
54.5
59.1
28.2
63.7
20.1
63.8
64.4
39.4
25.8
0.7
16.3
n.d.
34.1
45.2
4.4
0.9
Stability for the selected biocatalysts including ajPAP was
evaluated by measuring AMP conversion after pre-incubating
the enzymes (2-48 h, 30 °C) at different pHs or in the presence
of organic solvents (Table 1B and Fig. S14, bottom). Most
enzymes displayed poor half-lives at 30 °C and pH 7.5, with little
or no AMP conversion observed for aaPPK, chPPK, PPK11,
PPK62 and PPK82 after 2 h pre-incubation. Conversely, PPK12
lost little or no activity after 48 h pre-incubation under the same
conditions. Pre-incubation at pH 5 and pH 9 caused a more
significant activity drop for ajPAP and chPPK compared to pre-
incubation at neutral pH, whereas aaPPK and PPK82 displayed
improved stability at pH 9. Again, PPK12 retained its ATP
synthetic activity for prolonged incubation times at pH 5 and pH
B
2
0%
30%
DMSO
PPK
pH 7.5
pH 5.0
pH 9.0
Toluene
DMSO
aaPPK
ajPAP
chPPK
drPPK
PPK11
PPK12
PPK62
PPK82
< 0.3
1.3
< 0.3
< 0.3
< 0.3
1.8
< 0.3
0.5
< 0.3
0.9
< 0.3
1.6
< 0.3
0.8
< 0.3
6.3
< 0.3
7.2
< 0.3
6.5
0.3
35.4
5.5
6.0
< 0.3
34.4
< 0.3
< 0.3
< 0.3
25.2
< 0.3
< 0.3
< 0.3
27.3
< 0.3
< 0.3
< 0.3
23.6
< 0.3
< 0.3
1.0
< 0.3
29.7
< 0.3
< 0.3
7.3
< 0.3
< 0.3
9
, even though incubation at pH 5 led to a moderate activity loss
compared to incubation at neutral pH. A comparison between
6
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 11
Please Gd roe en no tC ha ed mj u iss tt r my argins
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/D0GC03830J
Fig 4. Panel enzymes and ATP formation rates from AMP in phylogenetic context. Sequence alignment and phylogenetic tree were generated with MAFFT63 using the entire set of
investigated proteins. ATP formation rates calculated from experiments with purified proteins are indicated by blue bars. Activity levels are indicated by rings around the phylogenetic
tree. This visualisation was produced using iTOL.64
Phylogenetic analysis of panel enzymes showing ATP also includes PPK9, which displayed the highest ATP formation
-1
synthetic activity starting from AMP, revealed that PPK12 falls rate in purified protein experiments (~20 s ). However, capillary
into a distinct clade (Fig. 4). Within this set of proteins PPK8 and electrophoresis analysis after high-throughput protein
PPK10 showed poor ATP formation rates. Interestingly, the production showed that PPK9 expressed poorly compared to
PPK8 and PPK10 sequences feature non-conservative amino selected PPK2-III enzymes for subsequent screening
acid substitutions of lysine residues poised for polyP binding in (approximately 20-fold lower protein concentration), and this is
chPPK
chPPK
chPPK
homologous proteins (Lys86
, Lys217
and Lys258
,
also reflected by the reduced protein yield observed after high-
Fig. S16). As replacement of these residues with alanine in throughput protein purification (Fig. S9). Therefore, PPK9 was
3
2
chPPK produced almost inactive proteins, it is reasonable to not initially considered for subsequent studies but it could
hypothesise that PPK8 and PPK10 substrate binding pockets are become a valuable biocatalyst after additional work to improve
not as finely tuned as those of other PPK2-III enzymes for the its expression levels.
double phosphorylation of AMP towards ATP. The same clade
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 8 of 11
ARTICLE
Journal Name
Table 2 Scale-up overview for the CAR catalysed 4-methoxybenzaldVeiheywdeAr2ticsleynOthnelinsies
Multigram-scale Carboxylate Reduction to Aldehyde Enabled
by ATP recycling
with ATP recycling supported by either ajPAP or PPK12.
DOI: 10.1039/D0GC03830J
To validate the improved utility of PPK12 for ATP recycling
purposes, the CAR33 catalysed 4-methoxybenzoic acid 1
reduction was re-evaluated. Similarly to the ajPAP case,
reactions were initially carried out on a 30 mL-scale and at
different ATP regeneration enzyme loadings (Fig. 5). More than
ajPAP
4.8
PPK12
4.0
Substrate Loading (g)
Substrate Concentration (g/L)
Overall Conversion (%)a
PPK Loading (% w/w)
Process Time (h)
10.7
10.0
95.0
>99.0
7.7
5
0 mM products were formed when PPK12 concentration was
just 1 mg/mL as opposed to 43 mM products generated with a
0-fold higher ajPAP loading. A maximum concentration of 62
32.6
44
22
1
Isolated Yield (g) (%)
3.0 (70.3)
7.7
2.6 (73.4)
5.6
mM products was formed with 10 mg/mL PPK12. Additionally,
no alcohol overreduction product was found at the end of the
reactions at the lower loading (Fig. S17).
These promising results prompted additional experiments to
evaluate the impact of polyP loading (Fig. S18), toluene (Fig.
S19) and 4-methoxybenzoic acid/polyP feeding strategy (Fig.
S20). Ultimately, it was found that best results were obtained at
b
Alcohol/Aldehyde (%)
a
b
UPLC data. calculated by NMR of the isolated products (Fig. S21).
could be dropped more than 4-fold compared to the value for
the ajPAP enabled carboxylate reduction process. About 85%
substrate (10 g/L) was converted in 5 h, with full conversion
achieved within 22 h (Fig. S22, top). The implementation of a
more active and stable PPK2 enzyme resulted in a lowered ATP-
recycling enzyme loading that approaches desired values for
this process metric.
3
00 mM polyP loading (polyP:product formed molar ratio 5:1)
in the presence of toluene to limit aldehyde overreduction.
To evaluate the scalability of the polyP-driven ATP recycling,
reactions were carried out on a multigram-scale (Table 2). In
investigating the ajPAP powered aldehyde synthesis, reaction
optimisation was aimed at reducing the loading of the PPK to
Notably, the pH regulation profile reveals that 55 mmol
NaOH were needed from the start of the reaction to maintain
pH 7.5 (55 mL of a 1 M NaOH solution, Fig. S22, bottom).
Acidification would only be expected from the generation of
gluconic acid as a result of NADPH recycling (mol gluconic acid
generated = mol products formed based on Scheme 1). The
production of approximately 26 mmol gluconic acid can be
estimated based on final product concentration and reaction
volume (65 mM 4-methoxybenzaldehyde 2 in 0.4 L). Therefore,
it appears that an excessive amount of base was needed to
maintain the pH at the set value. This trend applies to the rest
of pH-controlled reactions discussed in this work (data not
shown). There are only few examples in the literature reporting
the intensification of fully in vitro polyP-driven ATP-dependent
biocatalytic processes with pH control, but a quantitative
~10% w/w while obtaining a feasible multigram-scale process.
Ultimately, the reaction was scaled to 0.45 L (4.8 g of 4-
methoxybenzoic acid 1) and 4-methoxybenzaldehyde 2 was
obtained with 95% conversion and 70.3% isolated yield (3.0 g).
However, ajPAP loading was considerably higher than the
desired target value and the process entailed multiple doses of
reaction components and enzymes over 44h (Table 2). Such an
extensive optimisation was not required for PPK12
implementation: substrate acid and polyP were able to be
provided at the beginning of the reaction as the higher
concentrations were tolerated by PPK12 leading to a
considerably simpler batch process compared to the fed-batch
process required with ajPAP. The ATP recycling enzyme loading
5
0, 65
measure of base additions was not provided.
As polyP
hydrolysis has been linked to proton release and pH
6
5-67
homeostasis,
we speculate that hydrolysis of polyP
anhydride bonds can cause the acidification of the reaction
mixture at this polyP concentration, leading to the addition of
base equivalents. However, acid-generating background
activities from the host-cell background cannot be excluded.
Conclusions
There are a number of biocatalysts of interest to industry
such as carboxylic acid reductases, amide bond synthetases and
methyl transferases whose potential cannot be realised without
technically and economically feasible ATP recycling. Various
aspects need to be considered when developing an industrially
Fig. 5. Time course for the CAR-catalysed 4-methoxybenzoic acid reduction with dual
cofactor recycling on a 30 mL scale. Reactions (30 mL initial volume, 30°C, 400 rpm, 21
h) contained 10 mg/mL CAR33, 1 or 10 mg/mL ATP recycling enzyme (see legend), 1
mg/mL GDH-CDX-901, 150 mM 4-methoxybenzoic acid, 7% v/v DMSO, 300 mM polyP, relevant ATP recycling system, including the number of
+
4
0 mM MgCl2, 1 mM NADP , 300 mM D-Glucose, 1 mM ATP, toluene (50% v/v). 4-
enzymes employed (to reduce the number of fermentations
required), the overall loading required (to reduce direct costs
and issues with downstream processing), tolerance of high
methoxybenzoic acid and polyP were added over 5 h. Final concentrations take base
additions into account. The overall yield is the sum of aldehyde and alcohol yields.
8
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 9 of 11
Please Gd roe en no tC ha ed mj u iss tt r my argins
Journal Name
ARTICLE
6
7
8
S. M. Resnick and A. J. Zehnder, Appl Environ MiVcireowbAiroticl,le2O0n0lin0e,
6, 2045-2051.
N. N. Rao, M. R. Gomez-Garcia and A. Kornberg, Annu Rev
Biochem, 2009, 78, 605-647.
H. Zhang, K. Ishige and A. Kornberg, Proc Natl Acad Sci U S A,
2002, 99, 16678-16683.
concentrations of phosphate donor (to allow intensified
reactions), and tolerance of a broad range of reaction
conditions (to allow operation with a wide range of primary
biocatalysts). Initially we investigated the capability of reported
polyP kinases to support a model, moderately intensified, ATP
dependent biocatalyst. Notably, ajPAP, a PPK2-II enzyme, was
able to recycle AMP to ATP as a single heterologously expressed
lyophilised enzyme powder formulation supported by
undefined host cell enzymes. However, intensification of
reactions towards industrially relevant objectives highlighted
that greater activity and stability were required. This might be
achieved by enzyme engineering but the reliance upon
unspecified host enzymes would limit the extent to which
improvements could be made. We therefore sought to identify
a PPK2-III enzyme capable of unsupported regeneration of ATP
from AMP with superior properties which might serve as a basis
for future engineering efforts. Importantly, we provide the first
comprehensive comparison between new sequences and
characterised PPK2-III enzymes. More than 50 novel enzymes
that can be used for ATP regeneration from AMP were added to
the limited list of characterised PPK2 enzymes. In seeking a
more active and stable biocatalyst a tiered screening system
monitoring ATP synthetic activity, polyP tolerance and enzyme
stability were adopted for enzyme selection. Ultimately, the
novel PPK12 from an unclassified Erysipelotrichaceae bacterium
powered a multigram-scale cell free aldehyde synthesis at
significantly lower PPK-loading compared to the process
implementing ajPAP in a much simplified process. Overall, this
work contributes significantly to the industrial applicability of
ATP dependent enzymes.
6
DOI: 10.1039/D0GC03830J
9
1
K. Ahn and A. Kornberg, J Biol Chem, 1990, 265, 11734-11739.
0 S. R. Derrington, N. J. Turner and S. P. France, J Biotechnol,
2
019, 304, 78-88.
1
1 M. Petchey, A. Cuetos, B. Rowlinson, S. Dannevald, A. Frese,
P. W. Sutton, S. Lovelock, R. C. Lloyd, I. J. S. Fairlamb and G.
Grogan, Angew Chem Int Ed Engl, 2018, 57, 11584-11588.
2 H. K. Philpott, P. J. Thomas, D. Tew, D. E. Fuerst and S. L.
Lovelock, Green Chemistry, 2018, 20, 3426-3431.
1
1
3 A. J. L. Wood, N. J. Weise, J. D. Frampton, M. S. Dunstan, M.
A. Hollas, S. R. Derrington, R. C. Lloyd, D. Quaglia, F.
Parmeggiani, D. Leys, N. J. Turner and S. L. Flitsch, Angew
Chem Int Ed Engl, 2017, 56, 14498-14501.
4 M. Winn, S. M. Richardson, D. J. Campopiano and J.
Micklefield, Curr Opin Chem Biol, 2020, 55, 77-85.
1
1
1
5 M. R. Bennett, S. A. Shepherd, V. A. Cronin and J. Micklefield,
Curr Opin Chem Biol, 2017, 37, 97-106.
6 16. R. L. Baughn, O. Adalsteinsson and G. M. Whitesides,
Journal of the American Chemical Society, 1978, 100, 304-
3
06.
7 S. Krauser, T. Hoffmann and E. Heinzle, ACS Catalysis, 2015, 5,
407-1413.
1
1
1
8 A. Kameda, T. Shiba, Y. Kawazoe, Y. Satoh, Y. Ihara, M.
Munekata, K. Ishige and T. Noguchi, Journal of Bioscience and
Bioengineering, 2001, 91, 557-563.
1
2
2
2
9 S. Mordhorst, J. Siegrist, M. Muller, M. Richter and J. N.
Andexer, Angew Chem Int Ed Engl, 2017, 56, 4037-4041.
0 M. R. Petchey, B. Rowlinson, R. C. Lloyd, I. J. S. Fairlamb and
G. Grogan, ACS Catal, 2020, 10, 4659-4663.
1 S. Mordhorst, A. Maurer, D. Popadić, J. Brech and J. N.
Andexer, ChemCatChem, 2017, 9, 4164-4168.
2 A. Schwarz, S. Hecko, F. Rudroff, J. T. Kohrt, R. M. Howard and
M.
Winkler,
Biotechnol
J,
2020,
DOI:
1
0.1002/biot.202000315, e2000315.
Conflicts of interest
There are no conflicts to declare.
2
2
2
2
2
2
2
3
3
3 A. Alissandratos, K. Caron, T. D. Loan, J. E. Hennessy and C. J.
Easton, ACS Chem Biol, 2016, 11, 3289-3293.
4 T. D. Loan, C. J. Easton and A. Alissandratos, N Biotechnol,
2
019, 49, 104-111.
5 T. Endo and S. Koizumi, Advanced Synthesis & Catalysis, 2001,
43, 521-526.
Acknowledgements
This project has received funding from the European Union’s
Horizon 2020 research and innovation programme under grant
agreement No [722361].
We are grateful to Lois Butterwith and Kitty Clouston for cloning
and producing enzymes for the initial PPK panel. The authors
also thank Dr. Douglas E. Fuerst for critically reviewing this
manuscript.
3
6 S. P. France, L. J. Hepworth, N. J. Turner and S. L. Flitsch, ACS
Catalysis, 2016, 7, 710-724.
7 M. Horvat and M. Winkler, ChemCatChem, 2020, DOI:
1
0.1002/cctc.202000895.
8 A. J. Straathof, S. Panke and A. Schmid, Curr Opin Biotechnol,
2002, 13, 548-556.
9 P. Tufvesson, W. Fu, J. S. Jensen and J. M. Woodley, Food and
Bioproducts Processing, 2010, 88, 3-11.
0 K. Motomura, R. Hirota, M. Okada, T. Ikeda, T. Ishida and A.
Kuroda, Appl Environ Microbiol, 2014, 80, 2602-2608.
1 A. E. Parnell, S. Mordhorst, F. Kemper, M. Giurrandino, J. P.
Prince, N. J. Schwarzer, A. Hofer, D. Wohlwend, H. J. Jessen, S.
Gerhardt, O. Einsle, P. C. F. Oyston, J. N. Andexer and P. L.
Roach, Proc Natl Acad Sci U S A, 2018, 115, 3350-3355.
2 B. P. Nocek, A. N. Khusnutdinova, M. Ruszkowski, R. Flick, M.
Burda, K. Batyrova, G. Brown, A. Mucha, A. Joachimiak, Ł.
Berlicki and A. F. Yakunin, ACS Catalysis, 2018, 8, 10746-
Notes and references
1
2
3
4
5
J. N. Andexer and M. Richter, Chembiochem, 2015, 16, 380-
86.
S. Mordhorst and J. N. Andexer, Nat Prod Rep, 2020, DOI:
0.1039/d0np00004c.
A. Pollak, R. L. Baughn and G. M. Whitesides, Journal of the
American Chemical Society, 1977, 99, 2366-2367.
J. Bolte and G. M. Whitesides, Bioorganic Chemistry, 1984, 12,
3
3
1
1
0760.
3
3
3 S. Mordhorst, J. Singh, M. K. F. Mohr, R. Hinkelmann, M.
Keppler, H. J. Jessen and J. N. Andexer, Chembiochem, 2019,
2
0, 1019-1022.
1
70-175.
4 P. H. Wang, K. Fujishima, S. Berhanu, Y. Kuruma, T. Z. Jia, A. N.
J. Zhang, B. Wu, Y. Zhang, P. Kowal and P. G. Wang, Org Lett,
003, 5, 2583-2586.
2
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 10 of 11
ARTICLE
Journal Name
Khusnutdinova, A. F. Yakunin and S. E. McGlynn, ACS Synth 61 A. M. Kunjapur, Y. Tarasova and K. L. Prather, J AVmiewCAhrteicmle OSnolince,
Biol, 2020, 9, 36-42. 2014, 136, 11644-11654.
DOI: 10.1039/D0GC03830J
3
3
3
5 C. M. Lelièvre, M. Balandras, J. L. Petit, C. Vergne-Vaxelaire 62 M. Nordblad, M. D. Gomes, M. P. Meissner, H. Ramesh and J.
and A. Zaparucha, ChemCatChem, 2020, 12, 1184-1189.
M. Woodley, Organic Process Research & Development, 2018,
22, 1101-1114.
63 K. Katoh, J. Rozewicki and K. D. Yamada, Brief Bioinform, 2019,
20, 1160-1166.
6 S. Suzuki, R. Hara and K. Kino, J Biosci Bioeng, 2018, 125, 644-
6
48.
7 M. Lubberink, C. Schnepel, J. Citoler, S. R. Derrington, W.
Finnigan, M. A. Hayes, N. J. Turner and S. L. Flitsch, ACS 64 I. Letunic and P. Bork, Nucleic Acids Res, 2016, 44, W242-245.
Catalysis, 2020, 10, 10005-10009.
8 T. P. Fedorchuk, A. N. Khusnutdinova, E. Evdokimova, R. Flick,
65 J. Yao, J. Li, D. Xiong, Y. Qiu, G. Shi, J. M. Jin, Y. Tao and S. Y.
Tang, Appl Microbiol Biotechnol, 2020, 104, 3417-3431.
3
3
R. Di Leo, P. Stogios, A. Savchenko and A. F. Yakunin, J Am 66 U. Pick and M. Weiss, Plant Physiol, 1991, 97, 1234-1240.
Chem Soc, 2020, 142, 1038-1048.
67 R. Docampo, W. de Souza, K. Miranda, P. Rohloff and S. N.
Moreno, Nat Rev Microbiol, 2005, 3, 251-261.
9 M. A. Huffman, A. Fryszkowska, O. Alvizo, M. Borra-Garske, K.
R. Campos, K. A. Canada, P. N. Devine, D. Duan, J. H. Forstater,
S. T. Grosser, H. M. Halsey, G. J. Hughes, J. Jo, L. A. Joyce, J. N.
Kolev, J. Liang, K. M. Maloney, B. F. Mann, N. M. Marshall, M.
McLaughlin, J. C. Moore, G. S. Murphy, C. C. Nawrat, J. Nazor,
S. Novick, N. R. Patel, A. Rodriguez-Granillo, S. A. Robaire, E.
C. Sherer, M. D. Truppo, A. M. Whittaker, D. Verma, L. Xiao, Y.
Xu and H. Yang, Science, 2019, 366, 1255-1259.
4
4
0 G. Qu, J. Guo, D. Yang and Z. Sun, Green Chemistry, 2018, 20,
7
77-792.
1 D. Gahloth, M. S. Dunstan, D. Quaglia, E. Klumbys, M. P.
Lockhart-Cairns, A. M. Hill, S. R. Derrington, N. S. Scrutton, N.
J. Turner and D. Leys, Nat Chem Biol, 2017, 13, 975-981.
2 M. Winkler, Curr Opin Chem Biol, 2018, 43, 23-29.
4
4
3 A. Fryszkowska and P. N. Devine, Curr Opin Chem Biol, 2020,
5
5, 151-160.
4
4
4 B. van den Burg, Current Opinion in Microbiology, 2003, 6,
13-218.
5 J. D. Bloom, S. T. Labthavikul, C. R. Otey and F. H. Arnold, Proc
Natl Acad Sci U S A, 2006, 103, 5869-5874.
2
4
4
6 N. Tokuriki and D. S. Tawfik, Science, 2009, 324, 203-207.
7 N. C. Goodwin, J. P. Morrison, D. E. Fuerst and T. Hadi, ACS
Med Chem Lett, 2019, 10, 1363-1366.
4
8 J. P. Adams, M. J. B. Brown, A. Diaz-Rodriguez, R. C. Lloyd and
G. D. Roiban, Advanced Synthesis & Catalysis, 2019, DOI:
1
0.1002/adsc.201900424.
4
5
5
9 G. A. Strohmeier, A. Schwarz, J. N. Andexer and M. Winkler, J
Biotechnol, 2020, 307, 202-207.
0 G. A. Strohmeier, I. C. Eiteljorg, A. Schwarz and M. Winkler,
Chemistry, 2019, 25, 6119-6123.
1 E. Restiawaty, Y. Iwasa, S. Maya, K. Honda, T. Omasa, R.
Hirota, A. Kuroda and H. Ohtake, Process Biochemistry, 2011,
4
6, 1747-1752.
5
5
2 S. T. Kulmer, A. Gutmann, M. Lemmerer and B. Nidetzky,
Advanced Synthesis & Catalysis, 2017, 359, 292-301.
3 P. Tufvesson, J. Lima-Ramos, N. A. Haque, K. V. Gernaey and J.
M. Woodley, Organic Process Research & Development, 2013,
1
7, 1233-1238.
4 S. Luetz, L. Giver and J. Lalonde, Biotechnol Bioeng, 2008, 101,
47-653.
5
5
6
5 A. N. Khusnutdinova, R. Flick, A. Popovic, G. Brown, A.
Tchigvintsev, B. Nocek, K. Correia, J. C. Joo, R. Mahadevan and
A. F. Yakunin, Biotechnol J, 2017, 12.
5
6 W. Finnigan, R. Cutlan, R. Snajdrova, J. P. Adams, J. A.
Littlechild and N. J. Harmer, ChemCatChem, 2019, 11, 3474-
3
489.
5
5
7 J. Citoler, S. R. Derrington, J. L. Galman, H. Bevinakatti and N.
J. Turner, Green Chemistry, 2019, 21, 4932-4935.
8 J. I. Ramsden, R. S. Heath, S. R. Derrington, S. L. Montgomery,
J. Mangas-Sanchez, K. R. Mulholland and N. J. Turner, J Am
Chem Soc, 2019, 141, 1201-1206.
5
6
9 M. P. Thompson, S. R. Derrington, R. S. Heath, J. L. Porter, J.
Mangas-Sanchez, P. N. Devine, M. D. Truppo and N. J. Turner,
Tetrahedron, 2019, 75, 327-334.
0 S. Maurer, H. Schewe, J. Schrader and M. Buchhaupt, J
Biotechnol, 2019, 305, 11-17.
1
0 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 11 of 11
Green Chemistry
View Article Online
DOI: 10.1039/D0GC03830J
GDH
CAR
NADPH
NADP
+
ATP
AMP
Novel PPK2-III
P
P
P
P
P
P
P
…
Towards scalable ATP recycling: a newly identified PPK2-III
biocatalyst unlocked fully in vitro multigram-scale
aldehyde synthesis employing a carboxylic acid reductase.