Abstract
Background/Aim: Hypoxia-activated pro-drugs, such as TH-302, may kill hypoxic treatment-resistant tumor cells, but have failed in clinical trials. This may be related to variable levels of drug-activating reductases. Compounds such as bacteria-derived BE-43547, which target hypoxic cells independently of reductases, may be beneficial. This study characterized the in vitro potency and hypoxia selectivity of BE-43547 and TH-302. Materials and Methods: Tumor cells were exposed to different oxygenation levels in the presence/absence of drug, and survival was quantified using total cell number (BE-43547) or clonogenic survival (BE-43547 and TH-302) assays. Half-maximal inhibitory concentration (IC50) values and the hypoxia-cytotoxicity-ratio (HCR: normoxic IC50/hypoxic IC50) were determined from dose-response curves. Finally, both drugs were tested in spheroids exposed to 20% or 0% O2 for 24 h followed by assessment of clonogenic survival. Results: BE-43547 was highly potent and displayed little inter-cell line variability. Strongly enhanced cytotoxicity was observed under oxygen-restricted conditions with HCR’s of ~100 and ~20 after 24 h of treatment with 0 or 0.5% O2, respectively. Reducing treatment time somewhat reduced hypoxia selectivity. Hypoxia selectivity was observed regardless of whether the drug was added before or during the hypoxic challenge. TH-302 IC50 values varied 10-fold under oxic conditions, whereas those of the anoxic-to-normoxic HCR varied from 15 to 88. Both BE-43547 and TH-302 were unable to completely sterilize anoxic incubated spheroids. Conclusion: BE-43547 is highly hypoxia-selective, and unlike TH-302, displayed minimal variability between cell lines, suggesting that BE-43547 targets a fundamental feature/target that is only present, or of survival importance, during hypoxia. Spheroid experiments suggested inadequate tissue penetrability, which may be overcome by designing novel drug analogs.
Tumors often contain hypoxic, yet viable, cells, due to a poorly organized and unstable blood supply (1). Hypoxic cells are relatively resistant to radiotherapy (2). This effect gradually becomes significant at pO2 levels below ~10 mmHg, and at near anoxic levels, a 2-3 fold higher radiation dose is required, to obtain a given cell kill than under well-oxygenated conditions (2). Furthermore, the efficacy of chemotherapy is often lowered in hypoxic cells (3, 4) due to poor drug availability and because many drugs specifically target proliferating cells, and severely hypoxic cells are often quiescent. Accordingly, tumor hypoxia results in a poor prognosis in a variety of cancer diseases (5, 6). Importantly, it has been shown that selective targeting of hypoxic cells in patients with head and neck squamous cell carcinoma using the radiosensitizer nimorazole improves loco-regional control after radiotherapy, strongly suggesting that surviving hypoxic stem cells per se, are responsible for post-treatment tumor regrowth (6-9). Clinical proof that selective targeting of hypoxic cells in non-squamous tumor types improves outcome is missing. Still, doranidazole, another radiosensitizer, improved 3-year survival in patients with pancreatic cancer (10), and the bioreductive cytotoxin TH-302 in combination with gemcitabine resulted in a non-significant trend towards improved survival in patients with pancreatic cancer (11). In addition, numerous preclinical studies in tumor-bearing mice suggest that selective targeting of hypoxic cells may be beneficial in a broader selection of tumors (12-14). However, radiosensitizers are only of therapeutic relevance when combined with radiotherapy and other drugs that selectively target and kill hypoxic tumor cells are warranted.
Hypoxia-activated pro-drugs that are converted into cytotoxic compounds by reductases only under hypoxic conditions may overcome treatment resistance (11). Tirapazamine has shown promise preclinically (11, 15), but failed to meet the primary endpoints in clinical trials (11). This may be partly due to lack of appropriate quality control of radiotherapy (16) and the use of unselected patients with unknown extent of tumor hypoxia (17). Other problems include poor drug penetrability into the most hypoxic cell population (18), partly due to the metabolism of tirapazamine by more proximal cells existing at intermediate pO2 (19), and variable hypoxia selectivity. Drugs with improved tissue penetrability and hypoxia selectivity have subsequently been developed, including TH-302 (evofosfamide) (20, 21). TH-302 showed promise preclinically (22); however, clinical trials have been disappointing (23, 24). This may relate to insufficient levels of drug-activating cellular reductases or other cellular components/organelles (e.g., respiratory chain) that are involved in drug activation (20, 25, 26). A personalized treatment approach, that includes biomarkers for expression/activity of reductases has thus been proposed (25, 26), but the validity of such a strategy was later questioned (27), suggesting that our understanding of bioreductive cytotoxins remains incomplete. After decades of intense research, a bioreductive cytotoxin has yet to be integrated into daily clinical practice. Thus, the search for, and testing of, novel drugs that selectively target hypoxic tumor cells independently of cellular reductase activity are warranted. Interestingly, the macrocyclic depsipeptide BE-43547, which is produced by some bacteria in hypoxic niches (unpublished observations), possibly to kill competing microorganisms, shows selective cytotoxicity against hypoxic cells (28). The specific mechanism is under investigation, however, the compound was recently shown to covalently target eukaryotic translation elongation factor 1 alpha 1 (eEF1A1) under oxic conditions (29). The same group subsequently showed that the hypoxia-selectivity of BE-43547 involves disruption of the association between eEF1A1 and FoxO1 [unpublished work discussed in (30)]. Under hypoxia, BE-43547 induces necrotic cell death involving rapid loss of mitochondrial function and integrity with phenotypic similarities to ferroptosis (31), yet lacking the key markers of this type of cell death (32). The hypoxia-selective toxicity of BE-43547 is shared with the structurally related rakicidin-family of natural products (33, 34); however, BE-43547 displayed enhanced hypoxia selectivity when tested in pancreatic PANC-1 cells. Furthermore, the full potential of this drug remains uncertain including benchmarking against a lead bioreductive agent. The purpose of this study was therefore to provide a thorough investigation of BE-43547 that included several cell lines, and different treatment times and oxygenation levels. In addition, a spheroid drug penetrability assay was performed and for selected assays, a head-to-head comparison with TH-302 was conducted.
Materials and Methods
The study included the squamous cell carcinoma cell lines FaDuDD (head and neck), SiHa and Caski (cervix) and the adenocarcinoma cell lines MDA-MB-231 (mammary), DU-145 and PC-3 (prostate). SiHa was cultured in MEM, whereas the remaining cell lines were cultured in HEPES-buffered (10 mM) D-MEM. Media were supplemented with 10% fetal bovine serum (Cytiva, Marlborough, MA, USA), pyruvate (1 mM) and non-essential amino acids (Gibco, Life Technologies, Roskilde, Denmark). Experiments were conducted in the same medium type as that used for culturing. BE-43547 was isolated as a mixture of congeners (A1-C2) from Micromonospora sp. RV43 cultivated in artificial seawater-based ISP-2 (glucose 4 g/l, malt extract 10 g/l, yeast extract 4 g/l) over 7-10 days at 28°C. Extraction and purification was adapted from a previous study (28). The study consisted of five experiments (I-V).
Experiment I: 24 h treatment using traditional polystyrene substrata. Since earlier work on BE-43547 (28, 32) was performed on polystyrene (PS) substrate, the drug was initially tested in cells grown in PS Petri dishes, using a previously developed cell number assay (35). Briefly, 1,000-2,000 cells (cell line dependent) were seeded in 50 mm Petri dishes in discrete islets, by adding multiple 75 μl droplets of cells. Following cell attachment (>6 h), dishes were flooded with medium. Two days after seeding, cells were treated with vehicle (DMSO) or different drug concentrations and exposed to normoxia (~20% O2/cell incubator) or a humidified anoxic (95% N2, 5% CO2) gas for 24 h using gas tight chambers from Billups-Rothenberg (San Diego, CA, USA). Chambers were flushed at 5 l/min for one hour and then sealed. To validate that near-anoxic conditions were reached, anaerobic indicator strips (Merck, Darmstadt, Germany) were included. Following drug exposure, cells were washed and grown in fresh medium for 72 h, fixed in methanol, stained using toluidine blue, and rinsed. Stained cells were scanned at a resolution of 1,200 DPI using a document scanner (HP Scanjet 8300, Hewlett Packard, Palo Alto, CA, USA). Images were segmented, and stained area was quantified.
Experiment II: 24 h treatment using glass substrata. Next, we characterized the oxygenation-dependent cytotoxicity of BE-43547 using the assay described above, but modified it to allow cells to be grown on glass substrata. In short, 75 μl cell suspension was seeded on 9 mm circular glass cover slips in compartmentalized (free fluid movement between compartments) glass Petri dishes, with four compartments each holding one cover slip (36). All treatment and post-treatment experimental details were as described above, except that a third oxygenation level of 0.5% O2 was also included.
Experiment III: Acute 4 h treatment. Cells were prepared as above, and pre-incubated for 2 h under 20% O2 or N2 in gas-tight chambers, and then treated with BE-43547 for 4 h. The compound was added through small sealable holes to ensure maintenance of the defined gas atmosphere. Without convection, adequate gas equilibrium between cell medium and the flushing gas may take hours (37). Therefore, in these short experiments, cells were subjected to gentle orbital shaking for the entire gassing and treatment period. To allow compound addition, and further shortening of the equilibration time, experiments were conducted without Petri dish lids. Post-treatment processing were as in experiment I.
Experiment IV: Clonogenic assay. Five hundred cells were seeded in glass Petri dishes. The next day, cells were treated with BE-43547 or TH-302 for 24 h under 20% O2/N2 as in experiment I, washed, and grown in fresh medium for 10 days. Finally, cells were fixed, stained, and scanned as described above. The size of a typical colony with 50 cells was defined from manual microscopic evaluation and converted to number of pixels, which were used to count colonies in segmented binary images using ImageJ software.
Experiment V: Spheroid assay. Spheroids were established from FaDuDD cells using Corning spheroid well plates (Corning, NY, USA). The remaining cell lines only formed loosely aggregated and fragile spheroids. Briefly, 10,000 cells were added per well in 300 μl medium and allowed to grow for five days, which resulted in dense and mechanically robust spheroids with a diameter of ~650 μm. Spheroids were transferred to Glass Petri dishes (4 spheroids per dish) and treated with BE-43547 or TH-302 for 24 h under 20% O2 or N2. A drug dose well above anoxic IC50, yet considerably below normoxic IC50, was chosen based on results from the monolayer studies. A higher dose was also tested. After treatment, spheroids were centrifuged and collected in 0.15 ml of medium and 1.85 ml of trypsin was added. Trypsin was allowed to work for 10 min, intermixed with mechanical separation using vigorous pipetting to obtain a homogenous single cell suspension. To ensure an appropriate colony density (no overgrowth and sufficient colonies for reliable quantification), different cell suspension volumes were transferred to PS Petri dishes and allowed to grow for 10 days. Petri dishes with less than 400 colonies were analyzed for colonies as in experiment IV. As a control, 5,000 cells were seeded in Glass Petri dishes one day prior to treatment initiation and treated and processed as in experiment IV.
Statistics/calculations. IC50 values were calculated by fitting data for cell/colony number with a 4-parameter Hill logistic curve. HCR’s were calculated as the ratio between normoxic and hypoxic IC50 values. When relevant, a single factor ANOVA followed by a Holm-Sidak test for multiple pairwise comparisons were used to test for cell line differences.
Results
Initially, we tested the hypoxia-selectivity of BE-43547 using conventional polystyrene (PS), since PS was used in a previous study (28). Examples of individual experiments are shown in Figure 1 and summarized in Table I. There were no significant differences between cell lines in neither normoxic (505-593 nM) or anoxic IC50 values (6.1-11.5 nM). HCR values, when calculated from the ratio between mean IC50 values, were typically around 80. Of note, dose response curves were very steep around the IC50, showing that the dose range from no effect to full cell killing is narrow (Figure 1).
Hypoxia-selective cytotoxicity of BE-43547 in cells grown as discrete droplets in conventional polystyrene Petri dishes. Upper panel: an example of stained Petri Dishes for the cell line MDA-MB-231 treated with BE-43547 at different doses (in nM) under normoxic or anoxic conditions for 24 h. After treatment, cells were allowed to grow for 72 h in fresh drug-free medium and fixed, stained, and digitalized for further processing. The plots show typical examples of dose-response curves for four different cell lines, quantified as relative stained area normalized to respective controls (DMSO). The hypoxia cytotoxicity ratio (HCR) was calculated as the ratio between normoxic and anoxic IC50, for the plotted example (for HCR based on means of all experiments, as well as IC50 values see Table I).
Normoxic and hypoxic IC50 values for different cell lines grown in polystyrene dishes as discrete droplets or on separate glass cover slips and treated for 24 h.
Since PS contains dissolved releasable oxygen, which may compromise accurate oxygenation control in studies where near-anoxic conditions are required (38), we next tested BE-43547 in SiHa, DU-145, and MDA-MB-231 cells grown on glass. BE-43547 was highly hypoxia-selective with HCR values >100 when comparing normoxic and anoxic cells (Figure 2). This hypoxia-selectivity was somewhat higher than that observed in cells grown on PS, mainly due to a minor decrease in anoxic IC50 values on glass vs. PS. Raising the oxygenation level to 0.5% O2 reduced drug potency, but considerable hypoxia-selectivity was maintained, with HCR ranging from 17 to 35 (Table I) or from 17 to 42 (Figure 2), depending on the calculation method.
Hypoxia-selective cytotoxicity of BE-43547 in cells grown on discrete glass cover slips with four cover slips per condition. Upper panel: an example of stained cells on separate cover slips, for the cell line MDA-MB-231 treated with different drug doses (in nM) under normoxic (20% O2), 0.5% O2 or anoxic (0% O2) conditions for 24 h. Occasionally, a cover slip was lost during handling, staining, and other processing, resulting in only three glasses per treatment. After treatment, cells were allowed to grow for 72 h in fresh drug-free medium and fixed and stained, and relative cell number was calculated as average stained area for all cover glasses. The plots in the left column, show dose-response curves for three different cell lines, treated for 24 h at three different oxygen levels, quantified as relative stained area normalized to control (DMSO). Plots in the right column show cells treated for 4 h under normoxic or anoxic conditions, where cells were pre-equilibrated prior to drug treatment. HCR was calculated as the ratio between IC50 values for the plotted curves of averaged data. For additional data, including IC50 values, see Table I. Data are from two independent experiments and are expressed as mean±SD.
To assess whether reduced treatment time and pre-treatment oxygenation status affects drug potency and selectivity, cells were pre-equilibrated to 0 or 20% O2 for 2 h and treated for 4 h. Pre-equilibration did not compromise selectivity, but the reduced treatment time lowered HCR mainly due to an increase in anoxic IC50 to (mean±SD) 13.6±2.0 nM, 8.4±0.1 nM, and 17.3±0.7 nM in SiHa, DU-145, and MDA-MB-231, respectively. Reduced treatment time only marginally increased normoxic IC50 values to 702±147 nM, 535±12 nM, and 700±47 nM in SiHa, DU-145, and MDA-MB-231, resulting in HCR’s of 52, 64, and 41, respectively, when calculated as the ratio between these mean IC50 values. Figure 2 shows plots for pooled data, including HCR values calculated from these plots, which were largely similar to HCR values calculated as ratios between means of IC50 values from individual experiments, as stated in the text above.
Since delayed onset of cell death or loss of proliferative potential may be missed by the assays applied above, we conducted a supplementary clonogenic survival assay using glass as substrate (Figure 3). Results were largely in accordance with cell number assays, with excellent hypoxia selectivity and similar normoxic and anoxic IC50 values across three cell lines. For comparison, we included TH-302 in this experiment. Interestingly, the hypoxia selectivity of TH-302 was inferior compared to BE-43547 and there was substantial inter-cell line variability, especially due to differences in drug cytotoxicity against well-oxygenated cells, with a 10-fold difference in IC50 values across cell lines. Results are summarized in Figure 3.
Clonogenic survival assay in cells treated with BE-43547 or TH-302 in the presence or absence of oxygen. Five hundred (500) cells were seeded in glass Petri dishes and allowed to attach overnight. Cells were treated with BE-43547 or TH-302 for 24 h under normoxic or anoxic conditions and allowed to grow for an additional 10 days in fresh medium. Number of cell colonies (>50 cells) were determined following fixation and staining and normalized to the number of colonies in untreated (DMSO) cells, as exemplified in the upper row. The line graphs in the middle row: survival/dose data from three independent experiments were averaged, plotted, and fitted, resulting in a single normoxic and anoxic IC50 value and a single HCR value for each cell line. In contrast, the bar charts provide a summary of the average IC50 values, derived from each of the three independent experiments. HCR values calculated as the ratios between these means are included. The two different approaches were largely in agreement. Triangles: SiHa; Circles: MDA-MB-231; Squares: FaDuDD. It is evident that TH-302 is inferior to BE-43547 due to lowered hypoxia-selectivity and substantial, and highly significant, inter cell line variability, in normoxic IC50 values. Data are from three independent experiments and are shown as mean±SD.
Poor tissue drug penetration may limit treatment efficacy. Therefore, we tested the efficacy of BE-43547 and TH-302 in FaDuDD spheroids incubated under normoxic or anoxic conditions (Figure 4). BE-43547 was unable to sterilize spheroids at a dose of 50 nM (~10-fold above anoxic IC50 and ~10-fold below normoxic IC50), whereas there were no surviving colonies when 5,000 cells were seeded in monolayer in Petri dishes and run in parallel (not shown). A higher concentration of 200 nM was also ineffective. TH-302 performed better, but still a dose of 100 nM (~5-fold above anoxic IC50) which ensured 100% cell death/proliferative arrest in monolayer (not shown), could not fully sterilize spheroids.
Spheroid-based in vitro modeling of tissue penetration barriers of BE-43547 and TH-302. FaDuDD spheroids were established in ultra-low attachment well plates and exposed to normoxic or anoxic conditions for 24 h in the absence/presence of 50 nM BE-43547 or 100 nM TH-302, which are doses well above what is required for sterilization of cells grown in a monolayer. A higher dose was also tested in anoxic spheroids. Following treatment, spheroids were trypsinized and seeded to assess clonogenic survival. The dishes in the figure were not those used for quantification (se materials and methods), since they are partly overgrown, but were chosen for better graphic presentation. Data are mean±SD of 3-4 independent experiments, except for the high concentration TH-302 where n=2.
Discussion
Previous work revealed that BE-43547 displays a distinct cytotoxicity against hypoxic cells (28, 32, 34). However, further studies are justified, since 1) earlier work was limited to one cell line and treatment time; 2) cytotoxicity was only tested under fully oxygenated or anoxic conditions and only with drug administration prior to the gas challenge; 3) a direct comparison with a lead bioreductive cytotoxin, was not provided; 4) testing was done using traditional polystyrene well-plates, which are known to contain releasable oxygen that may confound results, and 5) tissue penetrability ability issues were not addressed.
Of note, we showed that BE-43547 displays excellent hypoxia selectivity and little inter-cell line variability, compared to the bioreductive cytotoxin TH-302 in monolayer cell cultures. Our spheroid experiments suggest inappropriate tissue penetrability, which may be overcome by designing novel drugs. Structural modification of this compound class, despite its significant complexity, is feasible (34, 39, 40).
Polystyrene contains releasable oxygen, which may affect radiosensitivity (38) and lower the activation of bioreductive drugs (41) and retention of hypoxia-selective PET tracers (36); however, since previous work used PS as substrate we first tested cells grown on PS using a previously developed cell number assay, where cells can be grown in discrete areas. This set-up allows reliable quantification of cell number based on stained area, while maintaining a low total cell number-to-medium volume ratio, which lowers the risk of drug depletion (if consumed/removed), and other unwanted changes in medium composition over time that may influence results. HCR values, calculated as the ratio between oxic and anoxic mean IC50 values, reached impressive values of ~80 (see Figure 1 for examples and Table I for summarized data). There was little cell-to-cell line variability, with no significant differences in normoxic or anoxic IC50 values across cell lines. HCR ratios were higher, and IC50 values lower than those reported in a previous study in the pancreatic cancer cell line PANC-1 (28). Differences may relate to cell line differences, or more likely the methodology since ¨PANC-1 cell survival was quantified immediately after treatment, which may miss delayed-onset cell death, which indeed corroborates with markedly higher IC50 values reported in that study. In addition, PANC-1 cells were treated in well-plates, which may further prolong the time to reach near-anoxic conditions.
Next, we characterized the hypoxia selectivity of BE-43547 in cells grown on glass. Our results revealed a further increase in hypoxia-selectivity with HCR’s above 100 and anoxic activity in the single-digit nanomolar range (Figure 2, Table I). As expected, the increase in HCR was mainly due to a modest decrease in anoxic IC50 values, suggesting that releasable oxygen dissolved in polystyrene is of significance even during prolonged gassing/treatment. Importantly, the variation between cell lines was small, suggesting that a specific drug dose that is effective across a range of diverse tumors may be defined.
An ideal cytotoxicity profile as a function of pO2 for a hypoxia-selective drug cannot be defined, since it is a compromise between a wish to target all treatment resistant cells, and the need to spare normal tissues, such as the skin and pericentral liver regions (42), where low levels of oxygen may occur. The typical in vitro Km value where half radio-sensitization is observed is around 0.5% O2 (~3.5 mmHg) (2) and tumor cells at such intermediate oxygenation levels pose a potential treatment barrier in radiotherapy (43), thus arguably a cytotoxicity profile where some, but not full, activity is maintained at 0.5% O2 is a reasonable aim. Increasing O2 from 0% to 0.5%, lowered drug potency, but substantial hypoxia selectivity was maintained, suggesting that BE-43547 also has potential to target intermediate tumor hypoxia (Figure 2, Table I).
The in vivo pharmacokinetics of BE-43547 have not been investigated, and it is unclear how the drug should ideally be administered. Of special interest is therefore to what extent drug potency and specificity is maintained when drug exposure time is shortened. Reduced treatment time (Figure 2) where cells were pre-equilibrated to the gas challenge prior to treatment, increased anoxic IC50 values, whereas normoxic IC50 values were largely unaffected, thus reducing HCR’s to ~40-70. Thus, a chronic-type low dose treatment may further broaden the therapeutic window. This experiment also addressed another important question. Treatment with BE-43547 leads to a fast collapse of mitochondrial integrity and function in hypoxic cells leading to cell death with necrotic characteristics (28, 32). Recently, it was shown that BE-43547 targets eEF1A1 under oxic conditions (29) and that the same target plays a role in the hypoxia-selective mechanism [discussed in (30)]. Regardless of the exact working mechanism, the hypoxia-selective toxicity may arise from interference with crucial morphological, biochemical and/or translational adaptive changes occurring in the transition phase from well-oxygenated to oxygen deprived conditions, which would limit the usability of such a drug against chronic hypoxic cells. This possibility, however, was ruled out since HCR values, as discussed above, remained high.
Cell number assays that quantify cell number/viability immediately after treatment or after only a few doubling times could underestimate the true drug cytotoxicity, since late-onset proliferative arrest or cell death is missed. Therefore, the clonogenic survival assay is considered a “gold standard”, as it considers both early and late events. We therefore performed a clonogenic survival assay in three cell lines and included TH-302 for direct comparison. A new cell line, FaDuDD, was included, since this cell line was useful for establishing spheroids, used in the last experiment. For BE-43547, the results were very similar to the more crude cell number assay, with HCR values exceeding 80. Importantly, the bioreductive lead compound TH-302 performed poorly in comparison, with lower and more variable HCR values ranging from 15-88 or 19-91, depending on the calculation method (Figure 3), which was mainly due to differences in oxic IC50 values. Large variability in TH-302 potency and hypoxia selectivity were also observed in a previous study (21), where anoxic IC50 values varied 900-fold and oxic IC50 values 35-fold between the least and most sensitive cell line. Such differences in drug sensitivity may limit the broad applicability of TH-302, and possibly other bioreductive cytotoxins, and may partly be responsible for disappointing clinical results.
Inadequate drug tissue penetrability may limit the usefulness of hypoxia-selective drugs. The diffusion distance from the capillaries to near anoxic, and thus the most treatment-resistant cell population may reach 100-200 μm, equaling 10-20 cell layers. Spheroids are a valuable in vitro model for initial assessment of potential tissue diffusion barriers. Of several cell lines tested, only FaDuDD, formed dense and mechanically robust spheroids useful for testing. Interestingly, 24 h treatment with 50 nM (~10 fold above anoxic IC50 in monolayer studies) was unable to sterilize anoxically incubated spheroids, and this inability could not be overcome by increasing the dose to 200 nM. TH-302, performed better, but still a very high dose ~3 times above oxic IC50 did not fully sterilize anoxic spheroids, suggesting that both drugs experience significant tissue penetrability barriers. Mitochondria are rapidly affected by treatment with BE-43547 and fluorophore-tagged derivatives of the related compound rakicidin distributes intracellularly (24), observations which strongly support transmembrane drug diffusion. It is therefore unclear why BE-43547 is unable to reach, or alternatively shows little efficacy, against deeper cell layers, but this critical question should be addressed in future studies. BE-43547 contains a long lipid anchoring motif, which may negatively affect the rate of diffusion through multiple cell layers. Alternatively, unwanted metabolism of BE-43547 could deplete drug availability in deeper cell layers. Clearly, further delineation of the precise mode-of-action, possible off-targets, and metabolism of BE-43547 is crucial, and could pave the way for development of drug analogs with improved tissue penetrability. Such a drug could be a gamechanger and would have great potential as a novel approach to target hypoxic treatment-resistant cells.
Footnotes
Authors’ Contributions
M.B. designed the study and performed all experiments. T.T. and K.M.J. supplied BE-43547. P.P.E. performed the data analysis of the clonogenic assay. M.R.H, J.O. and T.B.P supervised the study. All Authors contributed to the study design, and manuscript writing.
Conflicts of Interest
Thomas B. Poulsen, Thomas Tørring and Kristian M. Jacobsen are together the inventors of a series of synthetic analogs of the natural products. These analogs are covered by a patent application that is in the PCT phase.
Funding
This work was supported by a grant from the Danish Cancer Society:R231-A14066.
- Received October 4, 2023.
- Revision received October 26, 2023.
- Accepted October 27, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.