Distributable, Metabolic PET Reporting of Tuberculosis

Tuberculosis remains a large global disease burden for which treatment regimens are protracted and monitoring of disease activity difficult. Existing detection methods rely almost exclusively on bacterial culture from sputum which limits sampling to organisms on the pulmonary surface. Advances in monitoring tuberculous lesions have utilized the common glucoside [18F]FDG, yet lack specificity to the causative pathogen Mycobacterium tuberculosis (Mtb) and so do not directly correlate with pathogen viability. Here we show that a close mimic that is also positron-emitting of the non-mammalian Mtb disaccharide trehalose – 2-[18F]fluoro-2-deoxytrehalose ([18F]FDT) – can act as a mechanism-based enzyme reporter in vivo. Use of [18F]FDT in the imaging of Mtb in diverse models of disease, including non-human primates, successfully co-opts Mtb-specific processing of trehalose to allow the specific imaging of TB-associated lesions and to monitor the effects of treatment. A pyrogen-free, direct enzyme-catalyzed process for its radiochemical synthesis allows the ready production of [18F]FDT from the most globally-abundant organic 18F-containing molecule, [18F]FDG. The full, pre-clinical validation of both production method and [18F]FDT now creates a new, bacterium-specific, clinical diagnostic candidate. We anticipate that this distributable technology to generate clinical-grade [18F]FDT directly from the widely-available clinical reagent [18F]FDG, without need for either bespoke radioisotope generation or specialist chemical methods and/or facilities, could now usher in global, democratized access to a TB-specific PET tracer.


Supplementary Materials and Methods
Unless otherwise noted, all chemicals and solvents were purchased from Sigma-Aldrich (Milwaukee, WI, USA), Fisher Scientific (Hanover Park, IL, USA), Sigma-Aldrich UK, Alfa Aesar, Fisher UK, Carbosynth or Acros and used without further purification. Columns and Sep-Pak cartridges used in this synthesis were obtained from Agilent Technologies (Santa Clara, CA, USA) and Waters (Milford, MA, USA), Phenomenex UK respectively. Sep-Pak was conditioned with 5 mL ethanol. Analytical HPLC analyses for radiochemical work were performed on an Agilent 1200 Series instrument equipped with multi-wavelength detectors.
Mass spectra (MS) of decayed [ 18 F]FDT solution were recorded on a 6130 Quadrupole LC/MS, Agilent Technologies instrument equipped with a diode array detector. LC-MS analysis of fluorine-18 labeled trehalose was performed on Agilent 1260 HPLC system coupled to an Advion expression LCMS mass spectrometer with an ESI source. The LC inlet was Agilent 1200 series chromatographic system equipped with 1260 quaternary pump, 1260 Infinity autosampler, 1290 thermostatted column compartment, and radiation detector.
Column flow was split (1:4) between the mass spectrometer and the radiation detector.
Instrument control and data processing performed using Advion's Mass Express and Quant Express Software.  Power Pac Basic (200 V used,50 min.) with using MOPS/MES as buffer.

Preparation of Lysogeny Broth Agar Plates
Lysogeny broth (LB, 6.25 g) and agar (2.5 g) were both added to water (250 mL) and the mixture was autoclaved. After cooled to ~45 ºC, a stock solution of appropriate antibiotic was added and the agar was poured into plates and allowed to settle. Stored at 4 ºC.

Transformation
Plasmid stock (1 µL, 10 ng) and cell stock (20 µL) mixed in 14 mL polyene tube and incubated on ice for 30 min, the mixture was then heat shocked to 42 °C for 45 s then incubated on ice for 2 min. Super optimal broth with catabolic repression (SOC) medium (200 µL) was added and the mixture was incubated at 37 °C for 1h. The mixture was plated on agar plates with the appropriate antibiotic (2 plates, one with 100 µL one with 25 µL). Plates were then incubated overnight (37 °C).

DNA Preparation
A single colony was picked and placed into 5 mL of lysogeny broth (LB) with the appropriate antibiotic. The culture was shaken at 37 °C overnight. Cells were pelleted by centrifugation (4700 rpm, 20 min, 4 °C) then DNA preparation carried out according to the instructions provided with the QIAGEN QIAPrep spin miniprep kit. DNA concentration was determined by absorption on a Nanodrop and Sanger sequencing performed by Source Bioscience.
Sequence analysis was performed by using source bio edit.
The sample was then applied to a well in the gel. The gel was then run at 200 V for 1 h and then visualised by the addition of Coomassie InstantBlue dye (Gentaur) and gentle stirring for 2 h at room temperature.

Western Blots
SDS-PAGE was done as before, but using a Pre-Stained protein ladder and the gel was not stained with InstantBlue dye. Instead, the gel was electroblotted onto a membrane for 7 min using an iBlot machine. The membrane was then washed with 50 mM Tris-HCl, 150 mM NaCl and 0.05% Tween 20, pH 7.6 (TBS-T + 5% BSA w/v , 45 mL) for 1 h. The membrane was further washed with TBS-T + 5% BSA w/v, 45 mL and monoclonal anti-polyHistidinealkaline phosphatase antibody produced in mouse (Sigma, 16 μL). The membrane was then rocked for 4 h at room temperature. After this time, the solution was removed and the membrane washed with TBST buffer (3 × 20 mL). BCIP/NBT alkaline phosphate substrate (Sigma Aldrich, 4 mL) was added and the membrane was gently rocked until protein bands could be seen (~15 min). The membrane was then washed with water.

Protein Mass Spectrometry
Various machines used for protein LC-MS. In all cases solvent A was water and solvent B was acetonitrile with both containing 0.1% formic acid.
Set-up 1: Waters LCT Classic coupled to a Shimadzu 20 Series HPLC using a Thermo Proswift (250 x 4.6 mm x 5 µm) column. Electrospray source parameters were as follows: capillary voltage 3000 V; cone voltage 25 V. Spectra were calibrated to at least 17 matched peaks of the multiply charged ion series of equine myoglobin run under equivalent conditions. The gradient program is shown below; a flow rate of 0.4 ml/min was used throughout (see also Supplementary Table S3).
Set-up 2: Waters Xevo G2-QS QToF MS coupled to a Water Acquity UPLC using a Thermo Proswift (250 x 4.6 mm x 5 µm) column. Electrospray source parameters were as follows: capillary voltage 3000 V; cone voltage 20 V. Spectra were calibrated through use of an internal lock-spray. The gradient program is shown below; a flow rate of 0.3 ml/min was used throughout (see also Supplementary Table S4).
Nitrogen was used as the nebulizer and desolvation gas at a total flow of 600 l hr -1 . Spectra were calibrated using a calibration curve constructed from a minimum of 16 matched peaks from the multiply charged ion series of equine myoglobin obtained at a cone voltage of 35 V.
Data was processed using MassLynx software (v. 4.1 from Waters) according to the manufacturer's instructions.

Circular Dichroism (CD) Spectroscopy
For CD measurement, protein was buffer exchanged into phosphate buffer (10 mM potassium phosphate, pH 7.5). Measurements were made at room temperature using a Chirascan CDspectrometer (Applied Photophysics) using 0.1 cm quartz cuvette. Background readings were taken with using 0.2 mL of buffer only, followed by protein samples measured at concentrations of 0.2 mg/mL. Scans were run using 0.5 s per time point and 0.5 nm increments, scanning from 185 to 280 nm. Data reported is the mean average of 3 results.

Liquid Chromatography-mass spectrometry for Small Molecule Analysis
Liquid chromatography-mass spectrometry (LC-MS) for quantification of trehalose and glucose was performed on the LC-MS Quattro instrument coupled with agilent technologies HPLC system, using a TOSOH TSKgel Amide-80 column (4.6 mm x 250 mm, 5µM, 80 Å). Solvent A (CH3CN) and (H2O) with a volume ratio of 70:30 were used as the isocratic mobile phase at a flow rate of 1.0 mL / min. The MS file was programme as 20 min for each sample.
The electrospray source was operated with a capillary voltage of 3.0 kV and a cone voltage of 20 V. Nitrogen was used as the nebulizer and desolvation gas at a total flow of 500 Lh -1 . The data was acquired by single ion recording (SIR) of m/z 341 for trehalose and m/z 179 for glucose in negative mode and processed using MassLynx software (v. 4.1 from Waters) according to the manufacturer's instructions. See Supplementary Table S9 for the LC-MS methods employed in the detection of FDT.

High Performance Liquid Chromatography (HPLC) for Enzyme Reaction Analysis
All of the enzyme reactions were analysed by HPLC for UDP-glucose consumption and UDP release. For the analysis, few different stationary and mobile phases were used to optimise the separation between UDP and UDP-glucose (summarised in Supplementary Table S5).
Method 3 was selected as a method of choice because it gave the best separation between the two phosphate sugars.

Expression, Purification and Characterisation of OtsA E. coli
pET-22b(+) C-Terminal-His vector containing gene for OtsA was transformed into BL21 (DE3) cells using standard transformation protocol. The transformed cells were used to inoculate an overnight culture in 100 mL of freshly prepared LB medium containing 100 μL of Ampicillin stock solution. The culture was incubated overnight at 37 ˚C. The overnight culture was used to isolate plasmid DNA using the standard QIAGEN miniprep kit protocol.
The gene construct was confirmed by sequencing.
In a typical expression experiment performed on a 800 mL LB medium scale, transformed cells were grown overnight at 37 ˚C in 100 mL of freshly prepared LB containing 100 μL of Ampicillin stock solution (100 mg/mL). 4 mL of overnight culture was used to inoculate 800 mL of freshly prepared LB media containing 800 μL of Amp stock solution. The culture was incubated at 30 ˚C with shaking. OD600 was measured from time to time. Once the OD ~ 0.45, protein expression was induced by adding 4 mL of 1 M IPTG solution to a final concentration of 0.5 M. The culture was left to grow overnight at 30 ˚C for 18 hours. The culture was harvested by centrifugation (8000 rpm, 10 minutes, 4 ˚C). Cell pellet was suspended in binding buffer (50 mM Hepes, 200 mM NaCl, 10 mM imidazole, pH 7.5) and one tablet of Complete protease inhibitor cocktail (Roche), DNAase and lysozyme was added.
The mixture was stirred vigorously at 4 ˚C for 2 hours. The cells were lysed via sonication.
The suspension was then centrifuged (14000 rpm, 30 min) to remove cell debris. The supernatant was then filtered using a 0.45 μm syringe filter. The filtered supernatant was applied to a pre-equilibrated (using binding Buffer) GE Life Sciences 5mL HisTrap Ni-NTA column. Purification was done at 4 ˚C. The column was washed by 15 column volumes of binding buffer, followed by a linear gradient to 100 % Elution Buffer (50 mM Hepes, 200 mM NaCl, 500 mM Imidazole, pH 7.5) over 16 column volumes. Eluted protein fraction were analysed using SDS-PAGE analysis.

Expression, Purification and Characterisation of His-MBP-OtsB from E. coli
In a routine restriction digestion reaction for vectors, 2 μL of each restriction enzyme (NdeI & XhoI), 3 μL of NEB buffer 4 (10x), 1 μg of vector (pDB-His-MBP) were taken in an eppendorf tube. dd H2O was added to reaction mixture to make reaction volume 30 μL. The reaction mixture was incubated at 37 ˚C for 4 h. For insert, similar reaction was set-up with 3 μL of OtsB gene in pET-29b(+) vector. Digested vector and insert was loaded onto 1% Agarose gel. The gel was run at 110 V for 40 minutes.
Restriction-digested products were extracted from agarose gel using the standard QIAquick gel extraction kit protocol. Under UV transilluminator, gel fragments containing restrictiondigested products were excised using a clean, sharp scalpel. Gel were transferred to eppendorf tubes, weight of gel was taken. 3 volumes of QG buffer were added to 1 volume of gel. The sample was incubated at 50 ˚C for 10 minutes with occasional vortexing. 1 gel volume of isopropanol was added and was mixed by inverting the eppendorf tube 4-6 times. The sample was decanted into a spin column and was centrifuged for 1 minute. The flow through was discarded. 500 μL of QG buffer was added to the column and centrifuged for 1 minute. The supernatant was discarded and 750 μL of buffer PE was added to the column. The spin column was centrifuged for 1 minute. The supernatant was discarded, to remove the residual wash buffer; column was centrifuged further for 1 minute. The spin column was placed in a new Eppendorf tube and 50 μL of dd H2O was placed at the centre of column. The column was left to stand for 4-5 minutes, after which it was centrifuged for 1 minute. The DNA was stored at -20 o C for further manipulation. Concentration of products was calculated using Nanodrop.
For ligation, 50 ng of vector was combined with a 3-fold molar excess of PCR product. The volume was adjusted to 20 μL with dd H2O. 2 μL of 10 x T4 ligation buffer was added and mixed thoroughly. 1 μL of Quick T4 DNA Ligase was added and mixed thoroughly. Reaction mixture was centrifuged briefly and was incubated at 25 C for 5 minutes. The ligated product was stored at -20 o C. Transformation of the ligated product was done in XL10 Gold competent cells through standard transformation protocol using 2 μL of ligated reaction mixture. Transformed cells were plated on LB-Agar/Kan plates and incubated at 37 ˚C overnight.
Single colonies of transformed cell were used to inoculate overnight culture. Plasmid DNA was isolated from the overnight culture using standard miniprep protocol. The successful subcloning was verified by sequencing.
pDB-His-MBP vector containing gene for OtsB was transformed into BL21 (DE3) cells using standard transformation protocol. A single transformed cell was used to inoculate a 100 mL starter culture of freshly prepared LB medium containing 100 μL of Kanamycin stock solution (50 mg/mL). The culture was incubated at 30 ˚C for 16 hours. 8 mL of starter culture was used to inoculate 800 mL of LB medium containing 800 μL of Kanamycin stock solution.
The culture was incubated at 37 ˚C with shaking. OD600 was measured from time to time.
Once OD600 reaches close to 0.6. Expression was induced by adding 800 μL of 1 M IPTG solution. The culture was incubated at 30 ˚C with shaking for 16 hours.
The culture was harvested by centrifugation (8000 rpm, 10 minutes, 4 o C). Cell pellet was suspended in binding buffer (50 mM Tris, 200 mM NaCl, 20 mM Imidazole, pH 7.5) containing Complete protease inhibitor cocktail (Roche), DNAase and lysozyme. The mixture was stirred vigorously at 4 ˚C for 2 hours. The cells were lysed via sonication using program P9 consisting of 0.5 sec on, 0.5 sec off pulses for 30 sec followed by 1 minute of cooling at 40% amplitude. The sonication was done for 5 cycles. The suspension was centrifuged (20000 rpm, 40 min) to remove cell debris. The supernatant was then filtered using a 0.45 μm syringe filter. The filtered supernatant was applied to a pre-equilibrated (Binding Buffer) GE Life Sciences 5ml HisTrap HP column. Purification was done at 4 ˚C. The column was washed by 10 column volumes of binding buffer, followed by a linear gradient to 100 % Elution Buffer (50 mM Tris, 200 mM NaCl, 500 mM imidazole, pH 7.5) over 15 column volumes. Purification was done at a flow rate of 5 mL/min. Eluted protein fraction were analysed using SDS-PAGE analysis. Buffer exchange was done using dialysis tubing with MWCO 10,000 Da. Fractions containing the ca. 72 KDa protein were pooled and dialyzed thrice against 4 liters of 50 mM Tris, pH 7.5 and stored at 4 ˚C. Protein were concentrated using Vivaspin with MWCO 30,000 Da to a final concentration of 12.55 mg/mL. The protein was further stored at -20 o C and characterised by SDS-PAGE.

Expression, Purification and Characterisation of TreT P. horikoshii Enzyme
The DNA plasmid from group plasmid bank was verified by sequencing from source bioscience. The plasmid encoded a C-terminal 6 x Histidine tag (His-tag) on the protein. The plasmid was transformed into E.coli (BL21 DE3) cells and plated on LB agar containing 50 mg/mL kanamycin. One of the resulting grown colony was inoculated into 10 mL LB media containing 10 µL of kanamycin (50 mg/mL) solution and left incubating overnight (37 °C, 200 rpm).
To express the TreT protein, 10 mL grown culture was inoculated into 1 L of LB media containing 1 mL of kanamycin solution (50 mg/mL) and left incubating (37 °C, 200 rpm).
OD600 was measured every one hour and once the inoculated culture reached the mid-log phase, it was induced with 1 mL of IPTG solution (final concentration = 1 mM) and grown in shaking incubator overnight (37 °C, 200 rpm).
Cells were then harvested by centrifugation (4700 rpm, 20 min, 4 °C) and the pellet was resuspended in 30 mL lysis buffer (300 mM NaCl, 50 mM NaH2PO4, 20 mM imidazole). One protein inhibitor tablet, 25 mg lysozyme and 5 mg of DNase was also added into the lysate and lysis reaction was carried out on rotating platform (60 min, 4 °C). Lysed cells were sonicated (5 x 30 s, 30 s pulse on, 60 s pulse off, 75% amplitude). Sonicated cells were centrifuged (21000 x g, 20 min, 4 °C) to clarify the lysate. Supernatant was then filtered (0.45 µm and 0.2 µm filter). Next supernatant was loaded on a 50 mL FPLC loop and purified by Ni-NTA chromatography using 5 mL His-trap HP column and eluted with 15 column volumes of elution buffer (300 mM NaCl, 50 mM NaH2PO4, 250 mM imidazole). Protein samples were concentrated using Vivaspin at 10 kDa MWCO (Sartorius). Elution buffer was exchanged for 300 mM NaCl, 50 mM HEPES, pH 7.5. Protein samples were then characterised by SDS-PAGE to verify protein purity and LC-MS to confirm the mass of the protein.

Expression, Purification and Characterisation of TreT T. tenax
Enzyme 2 x 1 L cultures of TreT T. tenax protein were expressed in BL21 (DE3) cells. Expression was carried out by inoculating 10 mL cultured BL21 (DE3) transformed with plasmid DNA into 1 L of LB media. 1 mL of Kanamycin sulphate solution (50 mg / mL) was used as the resistance antibiotic. Both the flasks were shaken at 200 rpm and 37 °C for 3 hours when the OD600 value reached to 0.6. At this stage protein induction was carried out by using 1.0 mL of IPTG solution (final concentration = 1 mM), post induction both flasks were then kept overnight in the shaking incubator. Cells were then harvested by centrifugation (4700 rpm, 20 min, 4 °C), and the pellet was resuspended in 30 mL lysis buffer (300 mM NaCl, 50 mM NaH2PO4, 20 mM imidazole). One protein inhibitor tablet, 25 mg lysozyme and 5 mg of DNase was also added into the lysate and lysis reaction was carried out on rotating platform (60 min, 4 °C). Lysed cells were sonicated (5 x 30 s, 30 s pulse on, 60 s pulse off, 75% amplitude). Sonicated cells were centrifuged (21000 x g, 20 min, 4 °C) to clarify the lysate. Supernatant was then filtered (0.45 µm and 0.2 µm filter). Next supernatant was loaded on a 50 mL FPLC loop and purified by Ni-NTA chromatography using 5 mL His-trap HP column and eluted with 15 column volumes of elution buffer (300 mM NaCl, 50 mM NaH2PO4, 250 mM imidazole).
FPLC fractions were then analysed by SDS-PAGE. Fraction containing potential protein were concentrated using Vivaspin at 10 kDa MWCO (Sartorius) and centrifuging at 8000 x g.
Protein concentration was determined by nanodrop and characterisation was done with mass spectrometry and SDS-PAGE.

Plasmid Design and Trial Expression of OtsAB Fusion Enzyme
N-terminus His-tagged OtsAB fusion protein plasmid DNA was transformed into BL21 (DE3) cells. The grown colonies were inoculated into the LB media (10 mL) containing 10 µL of ampicillin (100 mg/mL) solution incubated overnight in the shaking incubator (37 °C, 200 rpm). Small scale expression trial was carried out by pipetting 250 µL of grown culture in 20 mL of LB media in 6 different 50 mL falcon tube. 20 µL of ampicillin solution was also added. Tube 1 & 2 were used as control (0.4 OD600 and 0.6 OD600); tube 3 & 4 were induced with 0.5 mM and 1 mM IPTG solution when OD600 value was 0.4 and 0.6 respectively. A similar process was repeated for tube 5 & 6. 1 mL of sample was taken from each tube at 0 min, before adding IPTG solution, 4 h, 16 h, 24 h and 48 h post IPTG induction.
Cells were then harvested from liquid culture by centrifugation (10,000 x g, 10 min, 4 °C).
The cell solution was decanted and the pellet was allowed to drain completely. 100 µL of bugbuster reagent was added into each of the sample and the cell pellet was completely resuspended by vortexing slowly. Cells were then incubated at a shaking platform for 20 min at room temperature. Insoluble cell debris was removed by centrifugation (16000 x g, 20 min, 4 °C). Supernatant was transferred into a fresh tube and analysed by SDS-PAGE.

Large Scale Expression of OtsAB Fusion Protein
N-terminus His-tagged OtsAB fusion protein plasmid DNA (2 µL) was transformed into BL21 (DE3) cells (20 µL). The grown colonies were inoculated into the LB media (10 mL FPLC fractions were then visualised by SDS-PAGE method and fraction containing potential protein were concentrated through Vivaspin (5 mL, 30 kDa MWCO) by centrifugation (5000 x g, 15 min, 4 °C). Protein samples were then characterised by LC-MS.
Reaction was started by adding the enzyme solution and stopped by placing on ice and quenched by adding 20 µL acetonitrile solution. Reaction was carried out on the thermoshaker (37 °C, 700 rpm) and continued for 24 h.

Detection of Trehalose by LC-MS for TreT and OtsAB Fusion enzyme
In order to detect the trehalose LC-MS method was used as described in the general experimental condition. Standard solutions for glucose and trehalose were analysed along with the samples. For TreT reaction UDP-glucose was used in excess and 8 mM glucose was used as a substrate. Enzyme concentration, metal ion concentration and buffer amounts were kept the same as in the case UDP-Glucose and UDP reaction.

Reaction Comparison between TreT homologues
To demonstrate the yield of FDT from TreT enzyme reactions, we tested the reaction for both homologues. Equimolar amount of starting substrate and enzyme concentration was used (see Supplementary Table S6). All the other parameters and reagents concentration was kept constant so that yield for FDT formation from each enzyme system can be compared.
In summary reaction volume was maintained to 500 μL and HEPES buffer (50 mM HEPES, pH 7.5, 200 mM NaCl and 10 mM MgCl2) was used to make up the required volume.
Reactions were initiated by adding the enzyme and then incubated at 37 o C, 700 rpm on the thermoshaker. 150 μL of reaction samples were taken after 30 min, 60 min and 120 min respectively from each reaction mixture. Reactions were stopped by incubating the reaction mixture at 100 o C for 5 min and further diluted to 500 μL with D2O solution, centrifuged and stored for the analysis.
Sample analysis was carried out using 500 MHz NMR spectrometer (AVB500) and 19 F-NMR signal for the formation of FDT were sought as compared to FDG peaks in all the reaction mixtures.

Evaluation of Reaction Reversibilty with TreT P. horikoshii
To test the reversibility of TreT P. horikoshii, we set-up small scale reactions from 1-10 mM Based on the bacmid DNAs, the recombinant baculovirus of OtsA and OtsB were made by transfection of Sf9 cells. Culture and preparation of Sf9 cells were performed according to the protocol provided by manufacturer. A Sf9 cell stock with a viability >95% and a density of 1.5 × 10 6 -2.5 × 10 6 cells/mL was prepared before proceeding to transfection.

Experimental procedures for generating P1 baculovirus for OtsA and OtsB
The P1 baculovirus stocks of OtsA and OtsB were generated using Sf9 cells in 6-well culture plates. Briefly, 2 ml of Sf9 cells were added to each well of two 6-well plates with the cell number of 8 × 10 5 in each well. Cells were allowed to attach for 15 minutes at room

Expression and Purification of C-terminally tagged, recombinant OtsA
Insect cells were grown in serum-free medium to a density of 2 × 10 6 cells/mL and a cell viability of greater than 95%. The cell culture (1 L) was transfected with the viral stock, 2.8 x 10 6 cells/mL. At 72 h post-infection, the cells were harvested by centrifugation. After freeze/thaw cycle, the pellet was resuspended in lysis buffer (100 mM HEPES, 200 mM NaCl, 10 mM imidazole, 10 mM MgCl2, 1 mM βME). DNase 3 mg and 1x Complete® protease inhibitor cocktail was also added into the protein pellet and lysed using a homogenizer. Cell debris was removed by centrifugation and the cell free extract passed through a 0.45 μm membrane filter. The filtrate was applied to a Ni-NTA column equilibrated in lysis buffer. After removal of the cell debris by centrifugation, the protein of interest was purified with a linear gradient of 10-300 mM imidazole in 100 mM HEPES, 200 mM NaCl, 10 mM MgCl2, 1 mM βME, pH 7.5, using Ni-NTA. Protein samples were exchanged into 100 mM Hepes, 150 mM NaCl, 10 mM MgCl2, pH 7.5, using a Hiload TM 26/60 Superdex TM 200 desalting column. The sample was passed through an Acrodisc Mustang E membrane (Pall #MSTG25E3) to reduce endotoxin to 1.9 EU/mL. After the addition of 10% glycerol, the aliquots containing the recombinant enzyme were flash frozen, and stored at −80 °C.

Expression and purification of recombinant OtsB
Insect cells were grown in serum-free medium to a density of 2 × 10 6 cells/mL and a cell viability of greater than 95%. The cell culture (1 L) was transfected with the viral stock, 2.0x10 6 cells/mL. At 72 h post-infection, the cells were harvested by centrifugation. After freeze/thaw cycle, the pellet was resuspended in lysis buffer (50 mM Tris-HCl, 200 mM NaCl, 5 mM MgCl2, 30 mM imidazole, 1 mM βME). DNase 3 mg, 1x Complete® protease inhibitor cocktail was also added in the pellet and lysed using a homogenizer. Cell debris was removed by centrifugation and the cell free extract passed through a 0.45 μm membrane filter.
The filtrate was applied to a Ni-NTA column equilibrated in lysis buffer. After removal of the cell debris by centrifugation, the protein of interest was purified with a linear gradient of 10-300mM imidazole in 50 mM Tris-HCl, 200 mM NaCl, 5 mM MgCl2, 1 mM βME, pH 7.5, using Ni-NTA. The fractions with potential protein were concentrated using Vivaspin (5 MWCO) and then desalted on Hiload TM 26/60 Superdex TM 200 with 50 mM Tris-HCl, 5 mM MgCl2, 100 mM NaCl, pH 7.5. The sample was passed through an Acrodisc Mustang E membrane (Pall #MSTG25E3) to reduce endotoxin to 1.9 EU/mL. After the addition of 10% glycerol, the aliquots containing the recombinant enzyme were flash frozen, and stored at −80 °C. SDS-PAGE analysis revealed the presence of a protein with the correct molecular weight.

Small-scale synthesis of [ 19 F]FDT using 3-Enzyme, One-pot System
To mimic 'hot' FDT synthesis, we tested 'cold' FDT synthesis from 0.138 mmol -5.520 mmol scale. Reagent amounts are detailed in Supplementary Table S8. In summary, in a 1.5 mL Eppendorf tube were added the following: FDG (various concentrations, see Supplementary Table S8

General Chemistry Experimental Procedures
Proton nuclear magnetic resonance spectra ( 1 H NMR) were recorded on a Bruker AVX 500 (500 MHZ), a Bruker AVB 500 (500 MHz) or a Bruker AVB 400 (400 MHZ) spectrometer as indicated. Carbon nuclear magnetic resonance ( 13 C NMR) spectra were recorded on the same spectrometer as described above. 19 F-NMR spectra were recorded on a Bruker AVX500, AVB500, AV600 and AVB400. NMR spectra were fully assigned using COSY and HSQC.
All chemical shifts are quoted on the δ scale in ppm using residual solvents as the internal standards. Coupling constants (J) are reported in Hz with the following splitting abbreviations: s = singlet, d = doublet, t = triplet, q = quartet, and m = multiplet. Infrared (IR) Spectra were recorded on a Bruker Tensor 27 Fourier Transform Spectrometer using KBr discs for solids and crystals. Absorption maxima (νmax) are reported in wave number (cm -1 ) and classified as strong (s) or broad (br).
Low-resolution mass spectrometry (LRMS) were recorded using an agilent 6120 Quadrupole spectrometry using electrospray ionisation (ESI) in either positive or negative mode. High resolution mass spectra (HRMS) were recorded on a Thermo Orbitrap Exactive mass spectrometer. The instrument is calibrated with a standard calibration mix from thermo before Thin layer chromatography (TLC) was carried out using aluminium backed sheets coated with 60F254 silica gel (Merck). Visualization of the silica plates was achieved using a UV lamp (λmax = 254 nm), and/or ammonium molybdate 5% in 2M H2SO4. Column chromatography was carried out using Geduran ® Si 60 silica gel (Merck). Mobile phases are reported in ratio of solvents (e.g. 4:1 petrol/ethyl acetate).

Radio-HPLC Method
Radio-HPLC analysis was carried out in a Bioscan radio-HPLC system (flow count) coupled with agilent 1200 HPLC system. Two analytical methods were used for the analysis of the depletion of starting material and formation of the product. In method 1, initially isocratic elution of 75:25 acetonitrile/water mobile phase combination was used with TSK gel normal phase HILIC column (phase amide -80, 10 cm x 4.6 mm x 5 μm) as stationary phase. The flow rate was maintained at 0.4 mL/min. 2.7 μm) was used as the stationary phase. The mobile phase was the gradient of two solvent system; solvent A= 50 mM ammonium formate, pH 4.5 and solvent B= acetonitrile. Flow rate was maintained at 0.5 mL/min and the total run time was 20 min.

Radio-TLC Method
Radio-TLC analysis was carried out in a Bioscan AR-2000 radio-TLC scanner. Silica gel plates 150 Å Silica Gel HL 250 cm 10 x 20 cm were used as the stationary phase and the 75:25 acetonitrile/water combination was used as the mobile phase. Once the strip developed to the 10 cm mark, was scanned by the radio-TLC scanner and analysed.

Luna-NH2 Purification Method
[ 18 F]FDT was purified by Sep-Pak cartridges. Sep-Pak Aminopropyl (NH2) plus short cartridge (Waters, WAT020535, 360 mg) used as the stationary phase. The cartridge was prerinsed with 1 mL EtOH and then twice with 3 mL per wash sterile water for injection. A 10 mL syringe was attached to the cartridge with plunger removed to the 5 μm filter. 1 mL of EtOH was loaded on to the syringe.
The reaction mixture was diluted with 2 mL of EtOH. This mixture was poured into the above syringe with filter. Reaction vial was rinsed with further 1 mL of EtOH and transferred to the syringe. The contents of the syringe were filtered through 5 μm syringe tip filter to remove precipitate. Eluent was loaded onto the NH2 cartridge and eluted without washing at a very low flow rate. Mixture was then concentrated first by evaporation and then by heating at 60 °C with a nitrogen flush. Finally the mixture was diluted with sterile water and filtered through sterile filter and drawn into the syringe.

OtsA Enzyme kinetic assay
The OtsA enzyme kinetic assay were measured through a NADH-linked GT-continued spectrophotometric assay. The decrease in absorbance of NADH at 340 nm (ε340 = 6220 M -1 cm -1 ) was measured at 25 °C using a spectrophotometer equipped with a thermostat. To

TreT P. horikoshii Enzyme Kinetic Assay
TreT P. horikoshii enzyme kinetic assay were measured through a NADH-linked GTcontinued spectrophotometric assay. The decrease in absorbance of NADH at 340 nm (ε340 = 6220 M -1 cm -1 ) was measured at 37 °C using a spectrophotometer equipped with a thermostat.
To determine the kinetic parameters of UDP-Glc, various UDP-Glc cocentrations (

TreT T. tenax Enzyme Kinetic Assay
In order to determine kinetic parameters (kcat, Km) of TreT thermoproteus tenax enzyme, UDP-Glc concentrations were varied from 0.03125 mM -7 mM In 1.5 mL Eppendorf tube were added glucose 50 µl (50 mM), MgCl2 20 µl (10 mM), TreT enzyme 4 µl, Hepes buffer varied volumes (50 mM, pH 7.5) to make up the total volume to 100 µl. Different volumes of UDP-glucose were added to the reaction mixture so that the final concentration of UDP-glucose in the reaction mixture was 0.01325, 0.06125, 0.125, 0.50, 1.00, 2.00, 3.00, 4.00, 5.00, 6.00 and 7.00 mM. Reactions were started by adding 4 µl enzyme and immediately placing on the thermoshaker at 70 °C . Each reaction was heated for 10 min and reactions were stopped by placing on ice and quenched by adding 20 µl acetonitrile to precipitate the protein. Reactions were then flash-frozen with liquid nitrogen and stored at -20 °C . Prior to HPLC analysis each reaction mixture was centrifuged at 13000 rpm for 20 min.
HPLC analysis was carried out by HPLC method 3 as specified in Supplementary Table S5.
The kinetic parameters of TreT T.tenax were also determined based on the variable acceptor substrate concentrations by GT-continued assay as described for TreT P. horikoshii under the similar conditions. The amount of UDP-Glc was fixed at 20 mM and Glc concentration varied (0.5 mM, 1 mM, 2 mM, 4 mM and 10 mM). The reactions were incubated at 37 °C for 5 min and then enzyme (662 nM) added and reaction rates determined.

OtsAB Enzyme Kinetic Assay
OtsAB kinetic parameters were determined using HPLC method 3 as specified in Supplementary Table S5

OtsB Enzyme Kinetics
For OtsB sf enzyme kinetics we used 1 H NMR spectroscopy employing the 1 H anomeric resonances as primary signals that are well resolved and with low background. Initially, we analyzed standard samples of trehalose-6-phosphate (T6P) and trehalose. For kinetic assay, we incubated several concentrations of T6P (0.5 mM, 1.0 mM, 1.5 mM, 2.0 mM, 2.5 mM, 3.0 mM, 3.5 mM, 4.0 mM and 5.0 mM). Reaction was initiated by adding 10 µL of OtsB and incubated at 37 °C for 10 min. After 10 min, reactions were stopped by boiling at 95 °C for 5 min and then analysed by 1 H NMR using D2O as a solvent. A standard calibration curve for trehalose was also generated and, using this calibration curve, concentrations of resulting trehalose from enzymatic reaction were obtained and analyzed through non-linear regression in GraphPad prism to give a Michaelis-Menten curve.
The activity of OtsB E. coli was also tested using commercially available Abcam Phosphate assay kit. By determining the amount of inorganic phosphate released from T6P by action of OtsB, enzymatic activity was to calculated. First, 10 μL of 10 mM phosphate standard was added to 900 μL of buffer (50 mM Tris, 5 mM MgCl2, pH 7.5) to generate 100 μM standard phosphate solution. 0,10, 20,30, 40 and 50 μL of 100 μM standard phosphate solution were added to individual wells of a 96-well plate. The volume was adjusted to 200 μL with dH2O to generate phosphate standards. 30 μL of working reagent was added to all wells and was mixed well. The plate was incubated at 37 °C for 30 minutes. Absorbance was read at 650 nm. A calibration curve was thus generated. As a practical note, in earlier experiments, reaction aliquots were mixed with 30 μL to quench the reaction and then volume was made up to 200 μL; it was found that doing so leads to precipitation. Therefore, in later experiments, reaction aliquots were diluted in the well using the reaction buffer, followed by addition of working reagent. In this way, no precipitate was observed. Next, enzymatic reactions were conducted in a final volume of 200 μL containing the following components: 5 mM MgCl2, 50 mM Tris buffer, pH 7.5, OtsB and T6P. 1.720 μM of enzyme was used in each case. T6P concentration was varied from 0.5 mM to 3 mM with an increment of 0.5 mM. Reaction aliquots were taken from the reaction at time, t = 0.5 min, 1 min, 2 min, 2.5 min, 3 min, 3.5 min, 4 min, 5 min, 6 min, 8 min and 10 minutes. The volume of reaction aliquots was taken in such a way that when diluted with reaction buffer, the maximum possible amount of phosphate remained within the linear range of the assay. 30 μL of working reagent was added after dilution. Samples were incubated at 37 °C for 30 min. Absorbance was read at 650 nm.
Using the calibration curve, the rate of formation of Pi was calculated and plotted against time to give rate of reaction at a particular substrate concentration. Data for different substrate concentrations were analyzed using non-linear regression in GraphPad prism to give a Michaelis-Menten curve.

Chemoenzymatic Reaction Optimisation and Scale-up
During reaction optimisation we tested the effect of ATP concentrations to the conversion of FDG to FDG6P initially at 1 mg scale. We set-up four different reactions in 1.5 mL Eppendorf tubes as shown in Supplementary Table S7. Reactions were incubated at 37 °C for 1 h at 400 rpm in a thermoshaker and then analysed by 19 F-NMR spectroscopy to observe the conversion.
Reactions were incubated at 37 °C at 400 rpm for 1 h and then analysed by 19 F-NMR spectroscopy.
In order to further optimise the enzymatic reaction conditions using 3 enzyme synthesis method, we tested various substrate concentrations i.e. with fixed enzyme concentrations. In summary in a 50 mL falcon tube were added FDG (various FDG concentrations i.e. 6.25 mg, 12.5 mg, 25 mg and 50 mg), ATP with equivalent or slight excess molar concentration to FDG, hexokinase 5 mg (680 U), UDP-Glc (37 mg with final concentration of 30 mM), OtsA enzyme (2 mg, 18 µM), OtsB (0.4 mg, 6.2 µM) and KCl (100 mM). Reaction volume was made up to 2 mL by HEPES buffer solution (HEPES 50 mM, pH 7.5, 100 mM NaCl, MgCl2 10 mM). Reactions were incubated at 37 °C and samples were taken after 15 min, 30 min, 60 min, 120 min, 200 min and 690 min, diluted with D2O and analysed by 19 F-NMR spectroscopy. 50 mg reaction was further optimised with additional UDP-Glc (168 mg, 138 mM) and analysed by 19 F-NMR.

Chemoenzymatic Synthesis of FDT
In a 50 mL sterile falcon tube were added; FDG (100 mg, 367 mM), ATP (320 mg, 368 mM) and MgCl2 (5 mg). The contents were dissolved in HEPES buffer solution containing 100 mM NaCl (100 mM, 1 mL, pH 7.5). The pH was then adjusted to 8.0 with NaOH. Hexokinase (0.6 mL) was added to the reaction mixture to initiate the reaction. The reaction mixture was incubated at 30 °C, 50 rpm for 1-2 hours. Phosphorylation of FDG to FDG6P was monitored by TLC and 19 F-NMR spectroscopy. Full conversion to FDG6P was observed within 2 hours for each batch synthesized.
Within the same pot, UDP-Glc (200 mg, 110 mM), KCl (100 mM) and OtsA enzyme (26 μM) was added. The total reaction was then adjusted to 3 mL. The reaction mixture was incubated to similar conditions as above for 24 h to allow appropriate conversion to FDT6P. The conversion of FDG6P to FDT6P was monitored by 19 F-NMR spectroscopy. The average maximum conversion after 24 h was between 65-70 %. After 24 h, in the same reaction mixture OtsB enzyme (6.23 μM) was added. The pH was adjusted to 8.0 and reaction mixture was incubated at 37 °C, 50 rpm. Reaction progress was monitored by TLC and 19 F-NMR. The reaction was usually complete within 2 hours.

Purification and Characterisation of FDT
The enzymes from crude reaction mixture were precipitated by either heating the mixture at 95 °C for 10 min or adding 2 mL EtOH and isolated by centrifugation (Vivaspin, 10,000 MWCO) and/or filtration. The resulting filtrate was lyophilized and the crude mixture was

Stability Testing of FDT Solution in HEPES Buffer
In order to establish the storage conditions of the final formulation of an injection, small scale stability testing of the final formulation was carried out. In summary, FDT sample was dissolved in HEPES Buffer solution (50 mM, pH 7.5) containing 7.5 mM MgCl2. One portion of the dissolved solution was stored at room temperature (20 -25 °C) and the other portion was stored between 2-8 °C. Samples were taken at day 0, 3, 5, 7 and 10 and analysed by 19 F NMR spectroscopy in triplicate. The means of the obtained integrals are then plotted against the day of testing. In addition, spectras were observed for any additional peaks as well.

Stability Testing of FDT Solid
In order to ensure that the compound will be stable throughout the testing period, on-going stability testing of the FDT sample from solid was also investigated over nine months' period.
Material for testing was stored at -20 °C and periodic testing was carried out by dissolving ~ 8 mg of sample in D2O and then subsequently analysed by 1 H NMR spectroscopy.

Determination of Purity by 1 H NMR
Purity of the final compound was determined by 1 H NMR spectroscopy by determining relative concentration of the substance as compared to impurities. Known quantity of FDT was dissolved in D2O and quantified by 1 H NMR spectroscopy. Maleic acid was used as an internal standard.

Na-NMR Quantification
In order to determine the amount of salt (NaCl) in the FDT compound we used 23 Na-NMR in AVX500 instrument. We generated the calibration curve of NaCl solution of known concentration and then run the FDT samples using the same parameters as NaCl standards.
The amount of Na was calculated from the standard calibration curve as an unknown and then from molar ratio of Na:FDT the amount of salt then determined. The reaction was started by incubating at 37 °C and monitored by radio-HPLC. The reaction did not work under the conditions described above even after 90 min of incubation.
In an another attempt, higher enzyme concentrations were added into the reaction mixture and incubated for 2 h at 37 o C but there was no conversion at all.  The eluent was then concentrated in vacuo. The resulting solution was filter-sterilized into a sterile vial for delivery. Identity of the compound was confirmed by LC-MS.

Specific Activity Analysis of [ 18 F]FDT
The radioactivity of the final products, [ 18 F]FDT, was obtained using a radiation detector. To quantify the mass of the decayed [ 18 F]FDT in the form of [ 18 O] trehalose, a modified version of the trehalose quantification enzymatic assay was performed (46) . A trehalose standard calibration curve (linear fit, R 2 = 0.9984) was generated using known concentrations of trehalose. The 'cold equivalent' masses of [ 18 F]FDT from syntheses were calculated based on this calibration curve. Then, the specific activity of the final product was calculated, following the definition, radioactivity at the end of the synthesis / unit mass of compound.

Chemical synthesis of [ 19 F]-3-FDG
In a 100 mL flask was dissolved 0.50 g of diacetone allofuranose (1.92 mmol, 1 eq) and 0.50g of N,N-diaminopyridine (4.1 mmol, 2.1 eq) in 20 mL of dry dichloromethane. The mixture was cooled down to -20°C using an acetone/dry ice bath then 0.5 mL of diethylaminosulfur trifluoride (DAST, 3.8 mmol, 2.0 eq) was added slowly, then the solution was allowed to return slowly to room temperature while stirring overnight. TLC (95:5 CH2Cl2:MeOH) showed a single compound (Rf 0.77). The mixture was cooled to 0 °C and 5mL of methanol was added slowly.
The mixture was stirred 1h and evaporated to dryness to give a yellow oil that was purified by silica column chromatography (49:1 CH2Cl2:EtOAc then 24:1 CH2Cl2:EtOAc) to give 416 mg of a colourless oil, that was used in the next step.
In a 50mL flask was dissolved 416mg of diacetone 3-FDG (1.58 mmol) in 20mL of water and 4 mL of ethanol. Ca. 6 mL of activated DOWEX 50W-H8 was added and the mixture was
Hexokinase (5.0 mg/mL final concentration) was added to initiate the reaction. Reaction mixture was incubated at 30 ˚C and monitored by ESI-MS at every 10 minute interval.
Alternatively, aliquots were withdrawn at every 30 minutes. Withdrawn aliquots were frozen on liquid nitrogen and boiled for 1 minute. Proteins were filtered off using a Vivaspin (Sartorius, 10K NMWCO) and each aliquot was analyzed by HPLC (Dionex UltiMate 3000) with the column, Phenomenex Luna NH2 (4.6 ´ 250 µm). The column was eluted with isocratic 7/3 acetonitrile/water. The maximum concentration of OtsB that could be used was 3.4 mg/mL. Above this concentration, OtsB precipitated while the reaction did not proceed to a meaningful extent at lower than this concentration of OtsB. By ESI/MS, it was found that the reaction did not go completion in 2 hours, after which precipitation formed. Based on peak intensity it was estimated that a mixture 7:1 3FTre:3FGlc was formed. HPLC analysis also revealed poor separation.       Heteronuclear Single Quantum Coherence experiment was used to determine proton-carbon single bond correlations. In the above spectrum 13 C NMR spectrum is on the Y-axis whereas 1 H NMR spectrum is shown on the X-axis. The sample was run on a Thermo Orbitrap Exactive mass spectrometer calibrated with standard calibration solutions that included MRFA (L-methionyl-arginyl-phenylalanylalanine acetate.xH2O), Ultramark 1621, caffeine and glacial acetic acid.

mixture and purified product of [ 19 F]-FDT.
Step 1 involves the conversion of FDG to FDG6P; Step 2 represents conversion of FDG6P to FDT6P; Step 3 shows the conversion of FDT6P to [ 19 F]-FDT.
Step 4  Supplementary Tables   Table S1. Endotoxin test for the one-pot reaction mixture containing OtsA and OtsB from E. coli at each purification step.
Protein concentration at each purification step was also carried out at the same time using BCA method (Thermo Scientific