Evaluation of biological properties of 3,3′,4,4′-benzophenonetetracarboxylic dianhydride derivatives and their ability to inhibit hexokinase activity

Article abstract of DOI:10.1016/j.bmcl.2016.12.055

This investigation has explored the properties of 3,3′,4,4′-benzophenonetetracarboxylic dianhydride (BDTA) derivatives with regard to their being prospective inhibitors of hexokinase II (HKII). A pluripotent embryonic carcinoma cell line P19 (ECC), was used as the biological target for newly generated potential inhibitors of HKII. The results obtained from Virtual High-Throughput Screening (VHTS), molecular modeling and biological activity studies showed BDTA to be a promising leading structure with a good binding score and simplest functionalization. The inhibitory effect was measured after 72?h incubation. Of selected BDTA derivatives, the most active was compound 3b, containing 3-hydroxyphenyl moiety in the para position, being able at 100?μM to decrease the mass of differentiated P19dCs cells by 30%, changing both the mitochondrial transmembrane potential and reactive oxygen species level. Under these conditions, only compound 3b had the ability to decrease hexokinase activity in a dose-dependent manner.

Full text of DOI:10.1016/j.bmcl.2016.12.055

Accepted Manuscript
Evaluation of biological properties of 3,3',4,4'-benzophenonetetracarboxylic di-
anhydride derivatives and their ability to inhibit hexokinase activity
Krzysztof Kochel, Mateusz D. Tomczyk, Rui F. Simões, Tomasz Fraczek,
Adrian Sobon, Paulo J. Oliveira, Krzysztof Z. Walczak, Aneta Koceva-Chyla
PII:
S0960-894X(16)31334-8
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.12.055
BMCL 24544
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
16 August 2016
17 December 2016
20 December 2016
Please cite this article as: Kochel, K., Tomczyk, M.D., Simões, R.F., Fraczek, T., Sobon, A., Oliveira, P.J., Walczak,
K.Z., Koceva-Chyla, A., Evaluation of biological properties of 3,3',4,4'-benzophenonetetracarboxylic dianhydride
derivatives and their ability to inhibit hexokinase activity, Bioorganic & Medicinal Chemistry Letters (2016), doi:
http://dx.doi.org/10.1016/j.bmcl.2016.12.055
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Evaluation of biological properties of 3,3',4,4'-
benzophenonetetracarboxylic dianhydride derivatives and their ability
to inhibit hexokinase activity
Krzysztof Kochela*, Mateusz D. Tomczykb, Rui F. Simõesc, Tomasz Fraczekd, Adrian Sobone,
Paulo J. Oliveirac, Krzysztof Z. Walczakb, Aneta Koceva-Chylaa
aDepartment of Medical Biophysics, Faculty of Biology and Environmental Protection, University of Lodz, Pomorska 141/143, 90-236 Lodz,
Poland
bDepartment of Organic Chemistry, Bioorganic Chemistry and Biotechnology, Silesian University of Technology, Krzywoustego 4, 44-100
Gliwice, Poland
cCNC - Center for Neuroscience and Cell Biology, University of Coimbra, UC-Biotech, Biocant Park, 3060-197 Cantanhede, Portugal
dInstitute of Applied Radiation Chemistry, Faculty of Chemistry, Lodz University of Technology, Zeromskiego 116, 90-924 Lodz, Poland
eDepartment of Industrial Microbiology and Biotechnology, Institute of Microbiology, Biotechnology and Immunology, Faculty of Biology
and Environmental Protection, University of Lodz, Banacha 12/16, 90-237 Lodz, Poland
Abstract
This investigation has explored the properties of 3,3',4,4'-benzophenonetetracarboxylic dianhydride (BDTA) derivatives with regard
to their being prospective inhibitors of hexokinase II (HKII). A pluripotent embryonic carcinoma cell line P19 (ECC), was used as
the biological target for newly generated potential inhibitors of HKII. The results obtained from Virtual High-Throughput Study
(VHTS), molecular modeling and biological activity studies showed BDTA to be a promising leading structure with a good binding
score and simplest functionalization. The inhibitory effect was measured after 72 h incubation. Of selected BDTA derivatives, the
most active was compound 3b, containing 3-hydroxyphenyl moiety in the para position, being able at 100 µM to decrease the mass
of differentiated P19dCs cells by 30%, changing both the mitochondrial transmembrane potential and O2•– level. Under these
conditions, only compound 3b had the ability to decrease hexokinase activity in a dose-dependent manner.
Graphical abstract
Abbreviations:3,3',4,4'-benzophenonetetracarboxylic dianhydride (BDTA); Embryonal carcinoma cells (ECC);
Glukokinase (GK); Glucose 6-phosphate (G6P); Glucose 6-phosphate dehydrogenase (G6PD); Mitochondrial
Membrane Potential (MMP); Mitochondrial Permeability Transition Pore (mPTP); Superoxide Anion (O2•–);
Sulforhodamine B (SRB); Tetramethylrhodamine (TMRE); Tricarboxylic acid (TCA); Virtual High-Throughput Study
(VHTS); Voltage Dependent Anion Channel (VDAC).

Glycolysis, a metabolic process convertining glucose into pyruvate, takes place in the cytosol and can
proceed under both aerobic and anaerobic conditions. In eukaryotic cells, pyruvate enters the mitochondria matrix
where it undergoes oxidation in the TCA cycle and contributions to oxidative phosphorylation and ATP synthesis.
While normal cells almost exclusively use oxidative phosphorylation for the ATP synthesis, cancer cells, especially
those in the rapidly growing tumors with high demand for glucose, increase the level of aerobic glycolysis, from
respiration towards fermentation that generates lactate even in aerobic conditions1. This specific metabolic pattern
(Warburg effect2) ensures that substrates and more ATP are produced for rapidly proliferating cancer cells1. High level
of aerobic glycolysis aids malignant cell survival in a hypoxic or acidosis microenvironment during the development
of a cancer. Many normal cells, ranging from microbes to lymphocytes, also use aerobic glycolysis during increased
proliferation, suggesting that this process meets the metabolic requirements of rapidly dividing cells and support their
growth33.
Glucose metabolism starts with phosphorylation of its 6-hydroxyl group4, trapping G6P inside the cell. The
process is catalyzed by hexokinases, a family with 4 isoforms (HKI, HKII, HKIII and HKIV) showing cell- and tissue-
specific expression, e.g. HKII mRNA expression can be up to 100-fold higher in cancer cells than normal cells5. HKs
I-III are widely distributed in mammalian tissues, whereas HKIV, a specific isoform of hexokinase (also known as
GK) is mainly expressed in liver and pancreatic β-cells 4,6. GK structure contains only one 50 kDa domain, whereas
other HKs have two 50 kDa domains. Unlike other hexokinase isoforms, capable of phosphorylating several hexoses,
GK have a 50-fold lower affinity for its substrate, which is only glucose7.The two most common isoforms HKI and
HKII overlap in tissue expression, but they have different subcellular distributions; their N-terminal domains allow
them to attach to the outer membrane of mitochondria4.. Whereas HKI is mainly associated with mitochondria, HKII is
associated with both mitochondrial and cytoplasmic compartments. It has been postulated that the differences in the
subcellular distribution of HKI and HKII isoforms might result in different metabolic roles in cells. Mitochondrially-
bound HKs may direct the catabolism of glucose (G6P formation towards glycolysis), whereas cytoplasmic HKII may
control its anabolism (glycogen synthesis)8. Indeed, the catabolic and anabolic fates of glucose are dynamically
regulated by its extracellular fraction.
Regulation of HKII activity and its subcellular translocation involves the participation of intracellular
glucose, G6P and Akt, all acting as signaling molecules. The activity of HKI is much less sensitive to control by these
factors, and this HK isoform is mainly committed to glycolysis. These processes could be important in HKs allowing
cells to keep their energy balance, to adapt to changing metabolic conditions and in avoiding injury8.
Mitochondrial HKII is necessary for the survival of cancer cells5. In almost 70% of cells9, once HKII has
been phosphorylated by Akt kinase, it binds to VDAC, a protein located in the outer mitochondrial membrane. HKII
binding to VDAC stabilizes mPTP and inhibits apoptosis by blocking the pro-apoptotic activity of Bax and Bak
proteins. The dual role of HKII in both processes, aerobic glycolysis and cell survival, makes this HK isoform an
attractive subject for investigation9,10 Research to date has resulted in the discovery of different molecules that can
cleave mitochondrial HKII11, found that it interacts with VDAC12, and allosterically inhibits the cytosolic fraction of
this enzyme13, thereby promoting cancer cell death. Glucosamine derivatives can also interact with the active center of
HKII14.
We have focused our investigation on BDTA derivatives as prospective inhibitors of HKs, which to the best
of our knowledge has not been previously undertaken in the search for potential anticancer agents.
The chemical structures of the compounds, we selected for our investigation, have been examined by VHTS.
The AutoDock Vina15, a chemical database downloaded from the Commercial Compound Collection16, was used for
this purpose. The open form of HKII17 was chosen as a receptor model. From the results of the final analysis,
compound 1 (Scheme 1) was selected as the most worthy structure for further investigation.
Scheme 1. Chemical structure of compound 1.
Since compound 1 contains only hydrophobic naphthalene groups, we synthesized other derivatives, named
3a-g, which include a hydroxyl group as a common structural element along with different structural variations, such
as heterocyclic rings and flexible alkyl chains. The condensation reaction involved a 2-fold excess of amino
compounds with respect to BTDA in polar aprotic solvent DMF (20 ml). The reaction was carried out in reflux, its
progress being monitored by TLC (MeOH:CHCl3, 1:1, v/v). When TLC indicated complete consumption of BTDA,
the mixture was cooled and poured into the water. The precipitant was filtered off, rinsed with cold water and
recrystallized twice from glacial acetic acid. The structure of synthesized compounds was based on analytical data
from 1H NMR, 13C NMR, IR and MALDI TOF MS18.

Scheme 2. Synthesis of BDTA derivatives 3a-g18.
To predict the binding mode of compounds 3a-g with an open form of HKII17,a more detailed study was
carried out by a molecular docking technique. The protein model was prepared in Protein Preparation Wizard. All
small molecules (water, ATP and ions) were removed for clarity. Water was removed since there are no available data
indicating its role in ligand binding within the enzyme. ATP was removed as it does not affect docking of our ligands.
The receptor grid was centered at Asp657, the size of the receptor box was set at 10 Å. To model small changes in the
receptor caused by ligand binding, van der Waals radii of amino acid residues were scaled down by 0.9. The results are
presented in Table 1. The best docking score was with compound 3b, and the weakest one was compound 3c.
Interestingly, the most active compound (3b) and the inactive compound (3e) had similar e-model value (one of the
scoring functions used by Glide).
Table 1. Glide docking results for examined compounds
Compound
Docking score [kcal/mol]
e-model
-61.7
-54.2
-53.2
-56.1
-61.4
-48.5
-52.2
-6.3
-5.7
-5.6
-5.2
-4.9
-4.9
-4.7
3b
3g
3f
3a
3e
3d
3c
The predicted binding mode of 3b is shown in Image 1. The ligand binds along the active site located in the
deep cleft of the enzyme (1a). Compound 3b forms 4 hydrogen bonds with Thr658, Asp861, Ser682 and Trp709, and
has a π-cation interaction with Lys510. There was also some expected interactions between π electrons of one of
phthalimide moieties with the carboxyl group of Asp657. The predicted binding mode of 3e was different from that of
3b, however, both compounds occupy roughly the same site on the enzyme. Moreover, 3b and 3e form a hydrogen
bond with Asp861. In contrast, the cytotoxic compound 3d is not so deeply buried in the binding site than the
previously discussed compounds, possibly explaining its lack of activity with hexokinase (1c)19–22
.

Image 1. BDTA derivative 3b docked to HKII: a) whole enzyme, b) close-up of docking site. Overlay of predicted binding modes of BDTA
derivatives (panel c): 3d (blue) 3b (red) and 3e (green).
Cytotoxicity was assayed using P19 cells (CRL-1825; Manassas, VA, USA), which can be cultured in the
undifferentiated (P19SCs) or differentiated state (P19dCs). P19SCs cells are highly glycolytic for energy production,
have immature, poorly polarized and inactive mitochondria23 and compared with P19dCs cells also have twice the
level of mitochondrially-bound HKII, much higher lactate production and use 30% more glucose23. Mitochondria of
P19dCs cells are more polarized and filamentous23.
Table 2. Effect of incubation time and concentration of BDTA derivatives on cell mass and metabolic activity. Data are means ±
SD from at least 4 independent measurements in 6 repeats of each. Bold values are statistically significant compared with the
control (p<0.05). Undifferentiated and differentiated P19 cells are described as P19SCs or P19dCs, respectively..
SRB assay
Compound
P19SCs
48 h
P19dCs
48 h
24 h
72 h
24 h
72 h
120.4 ± 22.5
103.6 ± 1.9
103.3 ± 12.4
90.2 ± 6.6
99.6 ± 4.1
102.0± 7.9
79.9 ± 14.1
81.8 ± 5.2
77.6 ± 9.4
99.4 ± 6.8
92.4 ± 9.3
101.5 ± 2.0
103.8 ± 8.0
106.5 ± 10.2
58.9 ± 16.0
86.8 ± 2.6
90.5 ± 13.7
95.6 ± 16.4
100.0 ± 19.8
86.7 ± 22.1
93.3 ± 6.1
93.6 ± 7.5
97.4 ± 9.5
79.8 ± 4.2
71.9 ± 3.3
30.9 ± 4.3
82.0 ± 12.1
90.8 ± 18.6
80.5 ± 15.4
94.7 ± 3.2
2
93.0 ± 17.8
102.2 ± 27.2
48.5 ± 13.0
96.0 ± 2.2
3a
3b
3c
3d
3e
3f
81.2 ± 9.3
45.5 ± 15.8
97.4 ± 11.1
106.7 ± 7.9
100.0 ± 2.5
115.2 ± 11.2
83.9 ± 15.8
106.0± 9.8
95.8 ± 19.4
96.2 ± 7.3
88.2 ± 15.1
108.4 ± 7.4
97.9 ± 7.4
110.6 ± 0.7
3g
Resazurin assay
100.1 ± 6.0
103.5 ± 4.6
102.6 ± 8.2
88.5±4.4
91.2 ± 10.1
99.7 ± 5.4
76.5 ± 5.7
74.4± 2.7
94.0±4.4
101.3 ± 4.1
101.2 ± 6.9
59.8 ± 17.5
79.9 ± 4.0
53.2 ± 23.7
109.6±10.3
113.8±10.2
100.0±2.3
90.5 ± 1.4
105.0 ± 9.0
100.0 ± 11.3
96.2±2.6
101.7 ± 3.9
105.3 ± 2.9
97.5 ± 15.2
87.4 ± 2.3
94.4 ± 7.5
99.4 ± 3.2
91.5 ± 7.2
93.8 ± 2.2
86.5 ± 29.7
105.0 ± 12.0
99.4 ± 14.5
98.1 ± 1.3
2
3a
3b
3c
3d
3e
3f
92.7±19.6
113.8±10.6
101.5±1.2
105.0±4.8
114.0±18.4
116.3±15.1
113.3±14.3
100.1±4.0
96.1 ± 21.0
105.3 ± 7.2
103.7 ± 5.3
101.1 ± 0.4
103.2±10.3
107.8 ±6.4
98.1±4.1
3g

Cytotoxicity in both types of cells was estimated after incubationd with 1, 10, or 100 µM of 3a-g for 24, 48
or 72 h in culture, and subsequent analysis involved SRB24 and resazurin25 assays. The metabolic activity of cells was
measured by the resazurin assay, and their mass by the SRB technique. None of the compounds were cytotoxic at 1
and 10 µM concentrations. At 100 µM the highest activity was with compounds 3b and 3d, and the longer incubation
time of 72 h increased cytotoxicity assayed with SRB method. Both compounds acted more potently on P19SCs cells
than P19dCs cells. In contrast, compound 3c containing a hydroxyl group in the para position of the benzene ring was
more effective on the differentiated cells, decreasing their mass by 70%. Compounds 3f and 3g with an aliphatic chain
at the nitrogen atom of imide skeleton instead of a phenol group were inactive (Table 2).
Based on cytotoxicity results, we selected compounds 3a, 3b and 3c to examine the changes in the MMP and
•–
mitochondrial O2 level in P19 cells, usingTMRE26 and MitoSox27 fluorescent dyes. Cells were incubated for 24 h or
72 h with 1 or 100 µM of each compound and the results are given in Table 3.
Compound 3a was inactive on both cell types. Compound 3b decreased metabolic activity and mass of both cell
lines, but only after the longer incubation time (72 h). Literature data show that HKII can regulate MMP and ROS
production28. Thus, we decided to investigate whether selected BDTA derivatives influence on MMP and ROS level.
Compound 3b increased MMP in P19SCs cells, but had the opposite effect on P19dCs cells. A 24 h incubation of the
undifferentiated cells with compound 3c, irrespective of its concentration, significantly increased MMP. Under the
same conditions, differentiated cells behaved differently, but a decrease in the MMP occurred in cells incubated at 100
µM.
•–
The mitochondrial O2 level decreased in P19SCs incubated with 100 µM 3b for 24 h, but increased in cells
incubated with 1 or 100 µM 3a for 72 h. P19dCs cells incubated with 1 or 100 µM 3c for 72 h showed an increase in
the MMP.
Table 3. Effect of incubation time and concentration of 3b, 3c and 3d on MMP and mitochondrial O2•– . Data are means ± SD from 4
independent measurements in at least 4 repeats of each. Bold values are statistically significant compared with the control (p<0.05). P19
undifferentiated and differentiated cells are called P19SCs or P19dCs, respectively.
MMP
Compound
P19SCs
P19dCs
24 h
100 µM
72 h
24 h
100 µM
72 h
1 µM
1 µM
100 µM
1 µM
1 µM
100 µM
3a
3b
3c
89.6±19.1
91.0±21.0
206.6±15.2
81.9±21.9
110.4±12.6
93.0±8.7
81.6±14.9
131.1±18.9
116.7±17.2
117.8±12.7
94.5±18.6
110.6±12.2
112.0±10.2
96.4±15.2
65.8±15.6
89.40±16.3
87.8±8.0
93.3±11.1
50.9±16.4
136.6±11.1
82.2±13.6
86.0±18.7
176.1±16.3
144.7±29.1
Mitochondrial O2•–level
3a
3b
3c
108.7±37.0
83.2±36.9
72.1±32.0
108.3±39.8
55.1±39.8
106.6±24.6
111.1±39.2
168.0±39.5
117.4±40.9
113.2±27.6
97.8±35.0
105.3±16.7
138.0±22.3
86.0±24.2
149.1±38.4
118.8±22.2
115.4±33.6
106.0±22.1
78.4±36.4
129.9±24.0
108.3±34.7
72.5±39.7
112.2±39.5
75.8±33.0
The inhibitory effect of these compounds on hexokinase activity was measured spectrophotometrically in cell
lysates by following an increase in NADPH absorbance at 340 nm29. As a positive control a pure enzyme (HKII, 6308-
100 from BioVision) was used. Parameters of its kinetic are given in Table 4. Results for Km are different from those
reported in the literature. Ardehali et al.,30,31 got Km values of 0.34 ± 0.06 and 0.36 ± 0.10, whereas Ahn et al.,32 got
0.37±0.12. These differences may result from slightly different methods and enzyme sources used to determine Km.
The first values were determined by measuring the phosphorylation of [U-14C] -labeled glucose as a function of
glucose added to the reaction mixture. The second method was based on spectrophotometric measurement in which
G6P formation is coupled with the NADPH production in the presence of an excess of G6PD. Under these conditions
only compound 3b decreased hexokinase activity in a dose-dependent manner (Figs 1 and 2).
Table 4. Kinetic parameters of HKII enzyme monitored at 340 nm. The results, shown as means ± SD, are
from 4 independent experiments in at least 4 repeats of each.
Km (glucose)
[mM]
Vmax
[mM min-1]
kcat
[sec-1]
Kcat/Km
[M-1s-1]
0.42±0.07
0.016±0.004
93.39±23.38
0.22*106

Fig. 1. Effect of 3b and 2-DG on hexokinase activity in supernatant of freshly prepared lysates of P19SCs cells. The results, shown as means
± SD, are from 4 independent experiments in at least 4 repeats of each. *p<0.05 - statistically significant in relation to control.
Fig. 2. Effect of 3b and 2-DG on the activity of HKII (Recombinant human hexokinase II protein, 6308-100 from BioVision). The results,
shown as means ± SD, are from 4 independent experiments in at least 4 repeats of each. *p<0.05 - statistically significant in relation to
control.
Summarizing, in examining the biological activity of BDTA derivatives and their ability to inhibit hexokinase
activity, our results indicate that the compounds with a hydroxybenzene ring were cytotoxic to mouse embryonic
carcinoma cells, but only at relatively high concentrations. Among newly synthesized BDTA derivatives only 3b
compound was able to partially inhibit a HK activity at a concentration at least 500 times lower (100 µM) than 2-DG
(50 mM). This demonstrates that compound 3b can be considered as a core structure in the design and development of
novel and effective hexokinase inhibitors.

References and notes
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Lincet, H.; Icard, P. Oncogene 2015, 34, 3751–3759.
Warburg, O. Science (80-. ). 1956, 123, 309–314.
Lunt, S. Y.; Vander Heiden, M. G. Annu. Rev. Cell Dev. Biol. 2011, 27, 441–464.
Wilson, J. E. J. Exp. Biol. 2003, 206, 2049–2057.
Mathupala, S. P.; Ko, Y. H.; Pedersen, P. L. Semin. Cancer Biol. 2009, 19, 17–24.
Meglasson, M. D.; Matschinsky, F. M. Diabetes / Metab. Rev. 1986, 2, 163–214.
Printz, R. L.; Magnuson, M. A.; Granner, D. K. Annu. Rev. Nutr. 1993, 13, 463–496.
John, S.; Weiss, J. N.; Ribalet, B. PLoS One 2011, 6, e17674.
Shoshan-Barmatz, V.; Zakar, M.; Rosenthal, K.; Abu-Hamad, S. Biochim. Biophys. Acta - Bioenerg. 2009, 1787, 421–430.
Majewski, N.; Nogueira, V.; Bhaskar, P.; Coy, P. E.; Skeen, J. E.; Gottlob, K.; Chandel, N. S.; Thompson, C. B.; Robey, R. B.;
Hay, N. Mol. Cell 2004, 16, 819–830.
11.
Goldin, N.; Arzoine, L.; Heyfets, A.; Israelson, A.; Zaslavsky, Z.; Bravman, T.; Bronner, V.; Notcovich, A.; Shoshan-Barmatz, V.;
Flescher, E. Oncogene 2008, 27, 4636–4643.
12.
13.
14.
Li, Y. C.; Fung, K. P.; Kwok, T. T.; Lee, C. Y.; Suen, Y. K.; Kong, S. K. Life Sci. 2002, 71, 2729–2740.
Pelicano, H.; Martin, D. S.; Xu, R.-H.; Huang, P. Oncogene 2006, 25, 4633–46.
Lin, H.; Zeng, J.; Xie, R.; Schulz, M. J.; Tedesco, R.; Qu, J.; Erhard, K. F.; Mack, J. F.; Raha, K.; Rendina, A. R.; Szewczuk, L.
M.; Kratz, P. M.; Jurewicz, A. J.; Cecconie, T.; Martens, S.; McDevitt, P. J.; Martin, J. D.; Chen, S. B.; Jiang, Y.; Nickels, L.;
Schwartz, B. J.; Smallwood, A.; Zhao, B.; Campobasso, N.; Qian, Y.; Briand, J.; Rominger, C. M.; Oleykowski, C.; Hardwicke, M.
A.; Luengo, J. I. ACS Med. Chem. Lett. 2016, 7, 217–222.
15.
16.
17.
18.
Trott, O.; Olson, A. J. J. Comput. Chem. 2010, 31, 455–461.
Del Rio, A.; Barbosa, A. J. M.; Caporuscio, F.; Mangiatordi, G. F. Mol. Biosyst. 2010, 6, 2122.
Salani, B.; Rio, A. D.; Marini, C.; Sambuceti, G.; Cordera, R.; Maggi, D. Endocr. Relat. Cancer 2014, 21, R461–R471.
All reagents and solvents were of analytical grade, obtained from commercial suppliers and used without further purification
except for DMF, which was distilled prior to use and dried through storage over activated 4A molecular sieves. The progress of
reaction was monitored by TLC using silica-gel-coated aluminium plates with a fluorescence indicator (SiO2 60, F254) and
visualized by UV light. Column chromatography was performed using silica gelpacked columns (particle size 0.040-0.063 mm,
Merck). 1H NMR spectra were recorded on a Varian 600 MHz System or Agilent 400 MHz Spectrometer; 13C NMR spectra were
recorded at 150 MHz or 100 MHz, respectively. Chemical shifts were measured relative to residual non-deuterated solvent
resonances. Melting points were determined using a Boethius M HMK hot-stage apparatus. IR spectra (ATR technique) were
recorded on Nicolet 6700 FT-IR Spectrometer (Thermo Scientific). MALDI TOF MS spectra were performed using a QTRAP
4500 Ab Sciex. N,N'-bis(2-hydroxyphenyl)benzophenone-3,3’,4,4’-tetracarboxylic diimide (3a)As light yellow solid in 89%. m.p.
>300 °C. 1H NMR (400 MHz, DMSO-d6): δ 6.94-6.96 (m, 2H), 7.00-7.03 (m, 2H), 7.28-7.36 (m, 4H), 8.17-8.21 (m, 4H), 8.27-
8.29 (m, 2H), 9.89 (s, 2H). FT-IR (cm-1): 3370 (m, br), 1780 (w), 1699 (s), 1389 (s), 717 (m). MALDI-TOF/MS (m/z): calc. for
C29H16N2O7 504.5, found 505.4 [M+H]+. N,N'-bis(3-hydroxyphenyl)benzophenone-3,3’,4,4’-tetracarboxylic diimide (3b)As
light yellow solid in 90%. m.p. >300 °C. 1H NMR (400 MHz, DMSO-d6): δ 6.85-6.87 (m, 6H), 7.32-7.35 (m, 2H), 8.15-8.17 (m,
4H), 8.25-8.26 (m, 2H), 9.78 (s, 2H). FT-IR (cm-1): 3300 (m, br), 1780 (w), 1718 (s), 1391 (s), 715 (m). MALDI-TOF/MS (m/z):
calc. for C29H16N2O7 504.5, found 505.4 [M+H]+. N,N'-bis(4-hydroxyphenyl)benzophenone-3,3’,4,4’-tetracarboxylic diimide
(3c)As light yellow solid in 94%. m.p. >300 °C. 1H NMR (400 MHz, DMSO-d6): δ 6.88-6.90 (d, 4H), 7.23-7.25 (d, 4H), 8.12-
8.15 (m, 4H), 8.23-8.25 (m, 2H), 9.78 (s, 2H). FT-IR (cm-1): 3451 (m, br), 1779 (w), 1709 (s), 1390 (s), 715 (m). MALDI-
TOF/MS (m/z): calc. for C29H16N2O7 504.5, found 505.3 [M+H]+. N,N'-bis(1,2,4-triazol-3-yl)benzophenonete-3,3’,4,4’-
tracarboxylic diimide (3d)As grey solid in 71%. m.p. >300 °C. 1H NMR (400 MHz, DMSO-d6): δ 8.19-8.21 (m, 2H), 8.23-8.25
(m, 2H), 8.29-8.31 (m, 2H), 8.77 (s, 2H). FT-IR (cm-1): 3392 (w, br), 1790 (w), 1730 (s), 1374 (s), 719 (m). MALDI-TOF/MS
(m/z): calc. for C21H10N8O5 454.4, found 455.2 [M+H]+. N,N'-bis(uracil-5-yl)benzophenone-3,3’,4,4’-tetracarboxylic diimide
(3e)As pale yellow solid in 91%. m.p. >300 °C. 1H NMR (400 MHz, DMSO-d6): δ 7.89 (s, 2H), 8.17-8.21 (m, 4H), 8.27-8.29 (m,
2H), 11.40 (br, 2H), 11.63 (br, 2H). FT-IR (cm-1): 3180 (s), 3160 (s), 1770 (w), 1715 (s), 1385 (s), 745 (m). MALDI-TOF/MS
(m/z): calc. for C21H16N2O7 540.4, found 541.1 [M+H]+.N,N'-bis(2-hydroxyethyl)benzophenone-3,3’,4,4’-tetracarboxylic
diimide (3f)As pale grey solid in 89%. m. p. 208 °C. 1H NMR (400 MHz, DMSO-d6): δ 3.63 (q, J = 6.6 Hz, 4H), 4.13 (t, J = 6.6
Hz, 4H), 4.81 (br s, 2H), 8.17-8.20 (m, 4H), 8.27-8.29 (m, 2H), FT-IR (cm-1): 3250 (w, br), 2954 (w), 1780 (w), 1714 (s), 1388 (s),
740 (m). MALDI-TOF/MS (m/z): calc. for C21H16N2O7 408.4, found 409.5 [M+H]+. N,N'-bis(3-hydroxyprop-1-
yl)benzophenone-3,3’,4,4’-tetracarboxylic diimide (3g)As pale grey solid in 90%. m. p. 170 °C. 1H NMR (400 MHz, DMSO-d6):
δ 1.76 (qt, J= 6,6 Hz, 4H), 3.46 (q, J = 6.6 Hz, 4H), 3.68 (t, J = 6.6 Hz, 4H), 4.51 (br s, 2H), 8.04-8.07 (m, 4H), 8.16-8.19 (m, 2H);
FT-IR (cm-1): 3242 (w, br), 2947 (w), 1779 (w), 1698 (s), 1393 (s), 743 (m). MALDI-TOF/MS (m/z): calc. for C21H16N2O7
436.4, found 437.3 [M+H]+.
19.
20.
Madhavi Sastry, G.; Adzhigirey, M.; Day, T.; Annabhimoju, R.; Sherman, W. J. Comput. Aided. Mol. Des. 2013, 27, 221–234.
Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.; Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.;
Perry, J. K.; Shaw, D. E.; Francis, P.; Shenkin, P. S. J. Med. Chem. 2004, 47, 1739–1749.
21.
22.
23.
24.
Friesner, R. A.; Murphy, R. B.; Repasky, M. P.; Frye, L. L.; Greenwood, J. R.; Halgren, T. A.; Sanschagrin, P. C.; Mainz, D. T. J.
Med. Chem. 2006, 49, 6177–6196.
Halgren, T. A.; Murphy, R. B.; Friesner, R. A.; Beard, H. S.; Frye, L. L.; Pollard, W. T.; Banks, J. L. J. Med. Chem. 2004, 47,
1750–1759.
Vega-Naredo, I.; Loureiro, R.; Mesquita, K. A.; Barbosa, I. A.; Tavares, L. C.; Branco, A. F.; Erickson, J. R.; Holy, J.; Perkins, E.
L.; Carvalho, R. A.; Oliveira, P. J. Cell Death Differ. 2014, 21, 1560–1574.
Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMahon, J.; Vistica, D.; Warren, J. T.; Bokesch, H.; Kenney, S.; Boyd, M. R.
JNCI J. Natl. Cancer Inst. 1990, 82, 1107–1112.
25.
26.
27.
28.
O’Brien, J.; Wilson, I.; Orton, T.; Pognan, F. Eur. J. Biochem. 2000, 267, 5421–5426.
Rodrigues, E. T.; Pardal, M.; Laizé, V.; Cancela, M. L.; Oliveira, P. J.; Serafim, T. L. Environ. Pollut. 2015, 206, 619–626.
Johnson-Cadwell, L. I.; Jekabsons, M. B.; Wang, A.; Polster, B. M.; Nicholls, D. G. J. Neurochem. 2007, 101, 1619–1631.
da-Silva, W. S.; Gomez-Puyou, A.; de Gomez-Puyou, M. T.; Moreno-Sanchez, R.; De Felice, F. G.; de Meis, L.; Oliveira, M. F.;
Galina, A. J. Biol. Chem. 2004, 279, 39846–39855.
29.
The 10 µl of supernatant of freshly prepared and centrifuged cell lysate or solution of HKII (Recombinant human Hexokinase II
protein, 6308-100 from BioVision), was added to 88 µl of 100 mM Tris-HCl (pH 8.0) buffer containing 0.5 mM EDTA, 10 mM
ATP, 10 mM MgCl2, 2 mM glucose, 0.1 mM NADP (Sigma-Aldrich) , and 0.1 U/ml of G6PD (Alfa-Aesar). HK activity was
determined spectrophotometrically at 340 nm tracking the G6P-dependent conversion of NADP to NADPH. The reaction was
initialized with 10 mM of ATP added to the reaction mixture after 10 min-preincubation of microplate at 37°C.

30.
31.
32.
Ardehali, H.; Printz, R. L.; Whitesell, R. R.; May, J. M.; Granner, D. K. J. Biol. Chem. 1999, 274, 15986–15989.
Ardehali, H.; Yano, Y.; Printz, R. L.; Koch, S.; Whitesell, R. R.; May, J. M.; Granner, D. K. J. Biol. Chem. 1996, 271, 1849–1852.
Ahn, K.-J.; Kim, J.; Yun, M.; Park, J.-H.; Lee, J.-D. BMB Rep. 2009, 42, 350–355.