Chemical Synthesis of the PAX Protein Inhibitor EG1 and Its Ability to Slow the Growth of Human Colorectal Carcinoma Cells

Colorectal cancer is primarily a disease of the developed world. The incidence rate has continued to increase over time, reflecting both demographic and lifestyle changes, which have resulted in genomic and epigenomic modifications. Many of the epigenetic modifications occur in genes known to be closely associated with embryonic development and cellular growth. In particular, the paired box (PAX) transcription factors are crucial for correct tissue development during embryogenesis due to their role in regulating genes involved in proliferation and cellular maintenance. In a number of cancers, including colorectal cancer, the PAX transcription factors are aberrantly expressed, driving proliferation and thus increased tumour growth. Here we have synthesized and used a small molecule PAX inhibitor, EG1, to inhibit PAX transcription factors in HCT116 colorectal cell cultures which resulted in reduced proliferation after three days of treatment. These results highlight PAX transcription factors as playing an important role in the proliferation of HCT116 colorectal cancer cells, suggesting there may be a potential therapeutic role for inhibition of PAX in limiting cancer cell growth.

Mass spectra were recorded on a Bruker micrOTOF-Q II mass spectrometer by electrospray ionisation in positive and negative mode. High-resolution mass spectra (HRMS) were obtained with a nominal resolution of 5,000 to 10,000.

LC-MS analysis
Mass spectrometry grade acetonitrile (CH 3 CN) and water (H 2 O) were purchased from Thermofisher Scientific. Mass spectrometry grade formic acid was purchased from Sigma-Aldrich. RP-LC-MS analyses were conducted on an analytical RP-HPLC (Shimadzu LC-20AD equipped with an SPD-20A UV detector [210 and 254 nm] and a Shimadzu LC-MS-2020 Liquid Chromatograph Mass Spectrometer operating in positive or negative ion mode) using a Phenomenex Prodigy column (C-18, 5 μm, 3.00 × 250 mm) at 0.5 mL/min and heated to 40 °C. The solvent system for LC purposes was a mixture of A (0.1% formic acid in H 2 O) and B (0.1% formic acid in CH 3 CN).

4ꞌ-((2-(Methoxycarbonyl)phenyl)carbamoyl)benzenaminium acetate (4)
Palladium on charcoal (10 % weight, 177 mg, 0.17 mmol) was added to a suspension of 3 (0.5 g, 1.67 mmol) in a mixture of AcOH:CH 2 Cl 2 :EtOAc (1:10:30, 41 mL) and flushed with N 2 gas for 10 minutes. H 2 gas (1 atm) was then introduced and the reaction mixture heated to 35-40 ˚C. After 22 hours the reaction mixture was flushed with N 2 gas for 10 minutes and the mixture was filtered through Celite. The Celite was washed with CH 2 Cl 2 (75 mL) and the organic filtrate concentrated in vacuo to obtain the title compound 4 as a white solid (432 mg,

Methyl 2-(4ꞌ-(2ꞌꞌ-methoxybenzamido)benzamido)benzoate (EG1 methyl ester 7)
Oxalyl chloride (190 μL, 2.27 mmol) was added to a solution of 2-methoxybenzoic acid 5 (240 mg, 1.54 mmol) in anhydrous CH 2 Cl 2 (8 mL) at rt. A catalytic quantity of DMF (two drops) was added to the reaction mixture at rt to result in effervescence. The reaction mixture was stirred at rt for 1.5 hours and then concentrated in vacuo to afford acid chloride 6 as a yellow oil. Without further purification, acid chloride 6 was immediately diluted in anhydrous CH 2 Cl 2 (7 mL) and cooled to 0 ˚C. In a separate round bottom flask, a solution of amine acetate salt 4 (503 mg, 1.52 mmol) was prepared by the subsequent addition of anhydrous DMF (3 mL), DIPEA (620 μL, 3.56 mmol) and then anhydrous CH 2 Cl 2 (12 mL). The resulting solution of amine acetate 4 was then added slowly over five minutes to the cooled solution of acid chloride 6. The reaction mixture was then allowed to warm to rt and stirred for 69 hours and then concentrated in vacuo. The residue was taken up in saturated NH 4 Cl (20 mL) and the aqueous phase extracted with CH 2 Cl 2 (3 x 50 mL). The combined organic extracts were washed with H 2 O (80 mL), then dried over Na 2 SO 4 and concentrated in vacuo to obtain a yellow oil. Purification was performed first with flash chromatography (CH 2 Cl 2 load, 0%, then 35%, then 40 % EtOAc in petroleum ether, then 50% EtOAc in CH 2 Cl 2 ). The fractions containing the product were combined and concentrated in vacuo to afford a yellow solid. This was purified further by trituration from CH 2 Cl 2 (20 mL) and petroleum ether (40 mL) to yield the title compound 7 as a pale orange-white powder (358 mg, 60%

Standard curve for EG1 by LC-MS using Selective Ion Monitoring (SIM)
Each data point obtained from the average of 2 injections per standard.

TIC of SIM m/z values 389 and 779 in negative ionisation mode for EG1 standards
2 µL injection of EG1 standards. The label on the chromatogram is the retention time in minutes.

STR profiling reports of the non-transformed HCT116 and the transformed HCT116/FUCCI cell lines
The HCT116 cell line was short tandem repeat (STR) profiled to ensure that the culture used for this study has retained the same STR markers as the reference HCT116 cell line. The profiling was performed by DNA diagnostics using the applied biosystems AmpFLSTR identifier or SeqStudio Genetic Analyser to measure STR abundance. The profiling showed that our culture has an 89% similarity to the reference with the only differences occuring in D3S1358, vWA and FGA, therefore we are confident that our cells are of colorectal cancer descent.
Identical to our HCT116 cell culture report the HCT116/FUCCI cells have shown an 89% similarity to the reference with differences in D3S1358, vWA and FGA. Therefore, STR profiling of the HCT116/FUCCI cell line has shown that regardless of the construct these cell's have been verified as HCT116's and thus are expected to behave in the same way as the original non-transformed cell line.

All PAX gene expression in HCT116, UACC62 and PC-3
Expression of the nine annotated PAX isoforms in HCT116, UACC62 and PC-3 cell lines have been measured by RT-qPCR to compare HCT116 isoform expression with two other cancer cell lines. We found that the highest PAX expression in HCT116 cells was PAX2, PAX6 and PAX9 overall, however these cells had higher expression of PAX1 and PAX2 than the other cell lines. Additionally, HCT116 has similar levels of PAX8 and PAX9 as PC-3 and UACC62 respectively.

Flow cytometry single stain data
An example of single stain flow cytometry data used to adjust the voltages and set the viability and apoptosis gates for the control and EG1 treated samples. A) Cell trace violet in the unstained cells shows that without staining the cells are unable to be detected and so will reliable produce results in the treatment populations. B) A full stain cell trace violet control was used to identify the upper limit of staining in cells which have not yet divided. C) Propidium iodide staining of a mixed population of living and heat treated dead cells allowed us to gate on the viable cells to ensure that only these cells were used in the proliferation plot. D) Positive Annexin V 647 staining of an apoptosis induced population was used to determine the fluorescence intensity of apoptosing cells thus measuring the extent of apoptosis with EG1 treatment. E) Propidium iodide versus Annexin V in the unstained population was unable to identify cells in the dead or apoptotic quadrants when the gates were determined by the single stained plots. F) Staining of both Propidium iodide and Annexin V produced cell populations in the three main quadrants; dead cells, living cells and apoptosed cells.

Flow cytometry proliferation calculation and analysis
FACS proliferation plots 12.11 Preliminary PC-3 proliferation and apoptosis data shows minimal change between the control and EG1 treated cultures.

Proliferation
Apoptosis EG1 has previously been described as primarily being a PAX2 inhibitor (Grimley et al. 2017) and so to determine if this compound preferentially inhibits PAX2 over other structurally similar PAX proteins we measured proliferation and apoptosis in the PC-3 cell line. PC-3 cells were used as an indicator of this as they express low levels of PAX2 and higher levels of PAX6 and PAX8 compared to the HCT116's. Thus, analysis of proliferation in these cells has shown that at 48 hours there were 47% more control cells than EG1 treated that had undergone two divisions, however at 72 hours 30% more EG1 treated cells than controls had progressed through three divisions. Finally, at 96 hours there was minimal difference between the control and EG1 treated populations, suggesting that EG1 has more impact on PAX2 function over other PAX proteins. Additionally, there was minimal apoptosis in the EG1 treated or control cell populations, with 1.96% in the EG1 treated population at 24 hours being the highest rate overall, and the largest difference being 0.38% more apoptosis in the EG1 treated cells after 24 hours when compared to the controls.

Cell cycling flow cytometry visualization of data and analysis
The scatter plots produced by propidium iodide (PI) versus forward scatter area for each population were used to identify the distribution of cells within and between the cell cycle phases. By using cell density this plot gave a better representation of population shift than the histogram and was able to show that by using PI we could identify the movement of HCT116's through the cell cycle.
Histograms of cell count versus PI were produced during cell collection to identify changes in population percentages per phase over 96 hours of treatment in the EG1 treated compared to the controls. The percentage of cells in each phase was determined by the gates set using the full stain control for each time point by identifying the fluorescent peak boundaries which represent the G0/G1, S and G2/M phases. The percentage of cells which lie within the gates for each phase was then used to determine a change in cell cycling with EG1 treatment.