Microwave gallium-68 radiochemistry for kinetically stable bis(thiosemicarbazone) complexes: structural investigations and cellular uptake under hypoxia

Hypoxia selectivity of new gallium-68 bis-(thiosemicarbazones) synthesised via microwave heating were investigated.

until 5% A at 15 minutes, isocratic until 22.5 minutes, reverse gradient from 22.6 minutes 95% A, then hold to 25.5 minutes.

General Radiochemistry Procedures
The zinc complex was prepared as either 1 mg/mL or 2 mg/mL in DMSO Measurements for each mouse were recorded using static scans at time points between 30 minutes and 24 hours. Each scan occurred for between 10 and 45 minutes and contained at least 20 million coincident events. The energy window used was 350-700 keV, with a coincidence timing window of 6 ns. Data were arranged into 2D histograms by Fourier rebinning, and transverse images were reconstructed by filtered back-projection into a 128 x 128 x 63 (0.72 x 0.72 x 1.3 mm 3 ) matrix. Images were normalised to correct for nonuniformity of response of the PET, dead-time count losses, positron branching ratio, and physical decay to the time of injection but no attenuation, scatter, or partial-volume averaging correction was applied. The counting rates of the reconstructed images were converted to activity concentrations (percentage of injected dose [%ID] per gram of tissue) using a calibration factor based on the image of a mouse-sized water-equivalent phantom containing Ga-68. ASIPro VM TM software (Concorde Microsystems) was used to analyse the data.

pH and biomimetic media studies
Buffers from pH 1.1 to pH 10 were made using a Fisher brand Hydrus 600 pH meter in order to investigate the stability of compounds by fluorescence spectroscopy. The following buffer systems were used, following the procedure described in "Buffers: A guide for the preparation and use of buffers in biological systems"; Chandra Mohan, 1997. 6 pH Buffer systems: Weymouth's medium for EMT6 (murine breast carcinoma). The media contained foetal calf serum (FCS) (10% for HeLa, PC-3, EMT6 and MCF-7 and 15% for FEK-4), 0.5% penicillin/streptomycin (10,000 IU mL -1 /10,000 mg mL -1 ) and 200 mM L-Glutamine (5 mL).
All steps were performed in absence of phenol red. Surplus supernatant containing dead cell matter and excess protein was aspirated. The live adherent cells were then washed with 2 x 10 mL aliquots of phosphate buffer saline (PBS) solution to remove any remaining media containing FCS, which inactivates trypsin. Cells were resuspended in solution by incubation in 3 mL of trypsin-PBS solution (0.25% trypsin) for 5 min at 37 • C. After trypsinisation, 5 mL of medium containing 10% serum was added to inactivate the trypsin and the solution was centrifuged for 5 min (1000 rpm, 25 ºC) to remove any remaining dead cell matter. The supernatant liquid was aspirated and 5 mL of cell medium (10% or 15% FCS) was added to the cell matter left behind. Cells were counted using a haemocytometer and then seeded as appropriate. Subsequently, cells were washed three times with PBS and 100 mL of MTT was added (0.5 mg mL -1 , 10% PBS: SFM) followed by a 2 h incubation. Following aspiration, 100 mL of DMSO was added and 96 well plates were read at an ELISA plate reader, Molecular Devices Versa Max (BN02877). Data were obtained from five consistent results and MI 50 was calculated using Origin 8 as half the height of the fitted curve for each compound and for each individual experiment. Due to background absorbance, 100% cell death would not correspond to zero absorbance, therefore the height of the curve was calculated as the highest absorbance of the fit plot minus the minimum absorbance of the curve, at which point death of all cells has been achieved. Where 100% cell death had not been attained, the MI 50 was calculated using the same method in that the minimum absorbance of the fitted curve was subtracted. This value therefore indicates the minimum MI 50 for this compound and is stated in the text/figure legend. Cells were cultured using standard protocols, as described above, in RPMI-1640 (PC-3).
Prior to addition of any commercial co-localisation dye, cells were washed 5 times with PBS.
Protocols adapted from Invitrogen were used throughout. Subsequently, a small volume of SFM was removed (e.g. 10 µL) and compound in DMSO was added in equal volume to that which was removed to obtain a final volume of 1 mL and the desired concentration (e.g. 50 mM in cell medium containing 0.5% DMSO). After 20 min or 1 h incubation with the compound cells were washed 3 times with PBS and fresh serum free medium was added (1 mL) and images were recorded immediately. .2) and specimens illuminated at the microscope stage of a modified Nikon TE2000-U with UV transmitting optics.
The focused laser spot was raster scanned using an XY galvanometer (GSI Lumonics).
Fluorescence emission was collected without de-scanning, bypassing the scanning system and passed through a coloured glass (BG39) filter. The scan was operated in normal mode and line, frame and pixel clock signals were generated and synchronised with an external fast microchannel plate photomultiplier tube used as the detector (R3809-U, Hamamatsu, Japan).

Radioactive cell uptake investigation
Cells were seeded as 3 x10 5 cells per well in a 6 well plate and incubated for ca. 12 h.
The cell medium was aspirated and replaced with serum free medium containing the Ga-68 radiolabelled complex (following the radiochemical procedure above). This was incubated at 20.7% O 2 and 5% CO 2 at 37  C for the normoxic sample. Hypoxic conditions were obtained by pre-incubating the cells for 20 minutes at 1% O 2 and 5% CO 2 at 37  C, followed by incubation under the same conditions with the compound for the time course of the study.
Cell plates were put on ice, washed 3 times with ice-cold PBS and lysed using 0.2 mL RIPA buffer for 10 minutes (Thermo Fisher Scientific Inc., Rockford, IL, USA). PBS (0.5 mL) was added to each well and cell lysates were transferred to counting tubes, with measurements of decay-corrected radioactivity performed using a gamma counter (Cobra II Auto-Gamma counter, Packard Biosciences Co, Pangbourne, UK). Aliquots were snap-frozen and subsequently protein determination was carried out using a BCA 96-well plate assay (Thermo Fisher Scientific Inc., Rockford, IL, USA). Decay corrected counts were corrected to protein concentration, with data presented as percent of total radioactivity per mg of protein,   Acenaphthenequinone (0.25g, 0.137mmol) and 4-allyl-3-thiosemicarbazide (0.54 g, 0.411 mmol) were suspended in 40 mL ethanol and were refluxed for 4 hours. 10 drops of conc.

Synthesis of the ligands
HCl were added upon reflux. The solid was isolated by filtration whilst hot, resuspended in hot methanol (10 mL) and stirred for 15 minutes before filtering and washing with further methanol. The resultant yellow solid (415 mg, 74%), was dried under vacuum.
Microwave synthesis: carried out according to general method B, (0.402 g, 94 %)  Compounds generally emitted between 520 and 675 nm, whereas excitation mainly occurred between 250 and 570 nm. All compounds displayed intrinsic fluorescence.  em-max was found to vary for gallium complexes depending on the functionalities: 510 nm, 400 nm, 450 nm and 520 nm for Ga (3) and Ga(4) complexes respectively.

Fluorescence Spectroscopy
Quantum yields were calculated using the standard solution of [Ru(bipy) 3 ](PF 6 ) in water as a reference and utilising the equation:

Laser scanning confocal microscopy
Experiments were carried out to ascertain if the weak fluorescence of the ligand precursors was sufficient to be observed in vitro. Cells were cultured using standard protocols as described in 1.6.
Cell      Interestingly, complex Ga(4) when incubated for 60 minutes, punctuation was also observed during the process of finding a representative area to image. A further study was carried out incubating HeLa cells with Ga(4) for 3 hours at 100 µM, 1% DMSO to determine if this effect occurs without irradiation when incubation is longer and more compound is present.
This effect was not observable in absence of irradiation indicating that this process does not occur with time alone. Additionally, the study was carried out at 37 °C and 4 °C, to investigate if the uptake occurred by passive diffusion alone, since mechanisms such as endocytosis do not occur at low temperatures. Ga(4) entered cells both at 4 ºC and 37 ºC, implying that cell uptake occurs by passive diffusion alone in fixed cells with the nucleus stained with DAPI. Nuclear uptake was observed at 37 ºC, but not at 4 ºC, signifying that the process did not occur by passive diffusion.

Two-photon excitation and fluorescence lifetime imaging microscopy
Two-photon absorption occurs when the combined energy of two simultaneous photons results in molecular excitation. The first photon to interact generates what is known as a virtual state, which has no classical analogue and only exists transiently (generally for femtoseconds) in which time a photon can travel ca. 1 μm. Arrival of the second photon must occur before the virtual state decays (dephases), which will occur if the laser intensity is high and focussed. The probability of single photon absorption is proportional to the intensity of light and is in contrast to two-photon absorption, which is dependent on both spatial and temporal coincidence (Equations 2.1 and 2.2). 7 Therefore, where NA x is the number of photons absorbed per second, I is the intensity, σ x is the cross-section and x is 1 or 2 photon absorption.
Single-photon absorption: Two-photon absorption: An advantage of two-photon excitation therefore, is that imaging occurs only from the focal plane. This is contrasting with single-photon excitation microscopy, in which the image is often distorted as it may occur outside the focal plane. 7 Furthermore, despite the requirement of high-powered lasers, two-photon microscopy has been found to cause less damage to biological cells than single-photon microscopy, as well as to decrease photo-bleaching and augment imaging depth. 8 Fluorescence lifetime is often considered as a means of distinguishing between different fluorophores, which possess sufficiently different lifetimes and can also be achieved using spectrally resolved microscopy.
FLIM is frequently used to separate different portions of the same fluorophore, for purposes such as quenching of luminescence by intracellular ions (e.g. Ca 2+ or Cl -) or by oxygen.
Furthermore, this technique can be used to monitor binding of and even distance from a fluorophore, with a conformational change is likely to result in a different rate of internal non-radiative decay and, therefore, a modification of the fluorescence lifetime. 9, 10 pH effects can also be studied, since protonated and non-protonated forms of a molecule may possess different lifetimes. In addition investigations of aggregation, viscosity, proximity to metal surfaces and nanoparticles (due to their long luminescent lifetimes) are carried out. 9 There has, however, been little use of FLIM for the purpose of assessing the stability of sodium 11 and magnesium 12   The full width at half maximum (FWHM), calculated from the lifetime distribution curve within the focal area was used to assess the error. The percentage of components τ 1 (major) and τ 2 (minor) in cells was from the respective amplitudes a 1 and a 2 calculated using SPCImage software, which models the data for each individual pixel to Equation 2.3, where F is fluorescence, a 0 is background and t is time: fluorescent lifetime, which could suggest that the mono(substituted) proligand geometry was less variable than its bis(substituted) counterpart. 9 Interestingly, although consistency of data was observed in cells, the respective short, major component is slightly shorter (ca. 0.2 ns), which suggests that the complexity of the cellular environment does have impact on the lifetime. There appeared to be a more limited effect on the mono(substituted) than the bis(substituted) ligand precursor, which displayed greater variation of lifetime within cells.

For each bis(substituted) ligand precursor tested by Time-Correlated
Notably, the fluorescence of the mono(substituted) ligand precursors was very weak and comparable to the bis(substituted) ligands.

Kinetic stability testing in an aqueous environment
Gallium complex Ga(4) had an unexpected intensity increase with time up to 8 h. The compounds remained fluorescent at 24 h, but at a lower intensity than at 8 h. Preliminary assays with media in the absence of serum displayed a gradual decrease in fluorescence with time, which is consistent with the zinc complexes Zn(4) in serum and serum free media. The difference between these two assay types is protein content, indicating that the gallium compounds likely interact with the protein.
Whilst changes in the UV/Vis spectra for the gallium complex Ga (   It can be observed that there is a much higher fluorescence intensity for Ga(4) at 24 h than for 4 at 15 minutes, indicating that although most complex may have converted to free ligand, it is likely that not all of the complex has. In contrast to the data acquired for L-cysteine, Lmethionine and L-histidine, in which Ga(4) appeared much more stable.
In the presence of 2 eq. of DFO, it was observed that the UV-vis spectrum of Ga (3)

pH study
The stability of complexes at biologically relevant pH is an important factor in this study.
Cancer cells are slightly more acid than non-cancerous cells it is important that the molecular probes are stable at lower pH. 17 All fluorescent scans were carried out with an excitation wavelength of 400 nm and at a concentration of 100 M, with a 1:1 DMSO:buffer solution.
An initial preliminary study was carried out using fluorescence spectroscopy for the gallium complexes.  Therefore an intial study where citric acid was used was carried out (prior to the assays described above) followed by another investigation that did not include citric acid as part of the buffer systems, a means to monitor if the nature of the buffer had a significant effect.
These gallium complexes are therefore very pH sensitive.
Furthermore all gallium complexes appeared most stable at either pH 2 or pH 3 after 15 minutes or less, which is also in agreement with data acquired where citric acid was a component of the buffer solutions. By UV-visible spectrometry compound Ga(4) seems to decompose completely after 24 hours when incubated in a pH 1.1 buffer, yet appears to remain intact at 15 minutes, as its spectrum has formed the shape characteristic of 4 in this time, fluorescence however appears to be fully quenched at 15 minutes. It is likely that stability of the gallium complex is poor at this pH. In pH buffers 2.0, 3.0 and 5.0 there is complex presence indicated by both absorption and emission spectra at 15 minutes, however a clear decrease in fluorescence intensity and absorbance including a blue shift for pH 3.0 and pH 5.0 buffers are indicative of instability of this compound at longer incubations.
As was observed with Zn(4), Ga(4) spectra closely resembled those of compound 4 when incubated with a pH 7.0 buffer, however due to the high stability observed in the 5% FCS assay the apparent lack of stability could be explained by interaction with the components of the buffer rather than the pH itself.

Preliminary In vivo imaging tests by microPET
Normoxic PC-3 xenografts were grown on the right shoulder of nude mice in Memorial Sloan-Kettering Cancer Center, New York. MicroPET was carried out for 68 Ga complex Ga(3), good renal clearance was observed, indicating suitability of these probes. Furthermore, no tumour uptake under these normoxic conditions was observed for all cases, with some limited uptake in the lungs, liver and spleen and clear bladder localisation/excretion.
This represents a promising result for probes designed, signifying that it would be beneficial for an in vitro hypoxia uptake focused future study to be carried out to reveal the exact biodistribution.

a) b) c) d)
O 2 , 5% CO 2 at 37 C before complex addition. Following addition of the compound cells were incubated for a further 20 minutes and subsequently washed three times with PBS before being returned to serum free media and imaged immediately. To investigate this issue a preliminary radioactive cell uptake experiment was carried out, showing that 68 Ga uptake was greater in hypoxic cells than normoxic cells (64% and 49% higher at 30 and 60 minutes respectively). This therefore indicates that the most likely cause of the reduction in fluorescence is due to the combination of increase in uptake as well as conversion to free ligand, signifying that the gallium complexes possesses selectivity for cells under hypoxic conditions. Figure below shows relatively high uptake of the allyl-68 Ga-BTSC compound in the EMT6 cells, especially when compared to uptake with 68 GaCl3 (Figure 2).
Total activity counted (% total tracer/mg of protein) is higher under hypoxia than normoxia (need at least another experiment to do appropriate statistics and test significance). This difference observed is highest at 2 hours (2-fold increase in uptake under hypoxia).

Cyclic voltammetry experiment of bis(thiosemicarbazonato) complexes and ligands
Cyclic voltammetry measurements were carried out in degassed dimethylformamide using glassy carbon as working electrode, platinum as supporting electrode and a silver wire as pseudo reference electrode. Compounds 3, 4 and complexes Zn (3), Zn(4), Ga (3) and Ga (4) were present at a 1 mM concentration, ferrocene was used as a reference at a 1 mM concentratrion and tetrabutylammonium tetrafluoroborate was added as supporting electrolyte in a 0.1 M concentration.
Initial and final potential was -1.6 V and scan rate was set at 100 mV/s. Five cyclic voltammograms were acquired for each compound.