Sulfamate inhibitor S4 influences carbonic anhydrase IX ectodomain shedding in colorectal carcinoma cells

Abstract

Carbonic anhydrase IX (CAIX) is a pivotal pH regulator under hypoxia, which by its tumor-specific expression represents an attractive target for cancer therapy. Here, we report on effects of the sulfamate CAIX inhibitor S4 (4-(3′-(3″,5″-dimethylphenyl)ureido)phenyl sulfamate) in colorectal carcinoma cell lines. S4 was administered under experimental hypoxia or normoxia to HT29, KM20L2 and HCT116 cells. Effects on survival, proliferation, pH, lactate extrusion and CAIX protein expression were evaluated. S4 treatment resulted in attenuated hypoxia-induced extracellular acidification and reduced clonogenic survival under hypoxia in HT29 cells. The pH effects were present only in a -free buffer system and were accompanied by decreased lactate extrusion. The main finding of this work was that S4 treatment caused alterations in CAIX ectodomain shedding. This merits further investigation to understand how sulfamates influence CAIX activity and how such drugs may be of use in cancer treatment.

• Carbonic anhydrase IX
• colorectal cancer
• ectodomain shedding
• extracellular pH
• hypoxia
• intracellular pH
• sulfamate inhibitor

Introduction

Tumor hypoxia is associated with therapy resistance and is recognized as a driver of the metastatic process. Hence, specific targeting of hypoxic tumor cells would in theory represent an ideal strategy to complement standard cytotoxic treatment1^,2. Tumor hypoxia is a particular challenge in large, heterogeneous and locally advanced tumors with hypoxic tumor components, where the risk of metastasis development is high after curatively intended local treatment. Colorectal cancer (CRC) illustrates this concept and represents a good model system for investigating such effects3.

Hypoxia causes a metabolic shift toward glycolysis that leads to increased production of acidic metabolites4. To avoid intracellular acidosis, which is incompatible with cell survival, tumor cells are dependent on activating mechanisms to increase extrusion of acids and/or increase uptake of neutralizing bases5. A key response effector in hypoxia, the hypoxia inducible factor-1α (HIF-1α), induces the expression of a number of genes involved in pH regulation in tumor cells, among these; the transmembrane enzyme carbonic anhydrase IX (CAIX)6^,7. CAIX promotes pH homeostasis by catalyzing the reversible hydration of CO₂ to and H⁺8. Through its catalytic activity CAIX makes available for cellular uptake while simultaneously stimulating diffusion of the weak acid CO₂ out of the cell9^,10. In addition to maintaining a favorable intracellular pH (pHi) to promote cell survival, CAIX activity can contribute to acidification of the extracellular environment, which is associated with increased invasion and development of metastasis11–13.

CAIX is highly expressed in a wide variety of human tumors (including malignancies of the breast, head-and-neck, lung, colon and rectum, uterine cervix and brain), but rarely in normal tissues14–16. High-tumor CAIX expression is generally associated with poor prognosis16. In addition, a soluble version of CAIX has been detected in body fluids, including serum and urine17. High levels of circulating CAIX have been found in cancer patients, by us18 and others, although its clinical significance remains unclear19^,20.

The pivotal role of CAIX as a pH regulating enzyme in tumors under hypoxia together with its tissue-specific expression have evoked interest in CAIX as a potential target for cancer therapy. A broad range of small molecular inhibitors specifically inhibiting CAIX has been developed over the past few years, including the sulfonamide inhibitors and its isoesters (sulfamates and sulfamides)21. The aim of this study was to investigate the effects of one such inhibitor, (4-(3′-(3″,5″-dimethylphenyl)ureido)phenyl sulfamate) (S4), in CRC cell lines under normoxia and hypoxia.

Methods

Cell culture

The human CRC cell lines HCT116 and HT29 (ATCC, Manassas, VA) and KM20L2 (kindly provided by Dr. M. R. Boyd, National Cancer Institute, Frederick, MD) were maintained in RPMI 1640 medium supplemented with 2.0 mM glutamax, 1 mM Hepes and 9% fetal bovine serum at 37 °C in air with 5% CO₂ (all the reagents from Sigma, St. Louis, MO). The cell lines were free from mycoplasma infection, and cell line identity was validated by short tandem repeat analysis. Cells were exposed to hypoxia (1% and 0.2% O₂) in a hypoxic chamber (Invivo2 200, Ruskinn Technologies, Bridgend, UK) in parallel to normoxia (21% O₂). Incubations were performed either in -buffered media (RPMI-1640, Sigma) containing 22 mM and 5% CO₂, or in -free media (modified RPMI-1640, Sigma) containing 20 mM Hepes, 0 mM and 0% CO₂.

Treatment with the CAIX inhibitor S4

The CAIX inhibitor S4 (developed by C. Supuran, University of Florence, Italy)22 was dissolved in dimethyl sulfoxide (DMSO, Sigma) and diluted in cell culture media to a final concentration of 100 µM. Control cells were given vehicle containing equivalent amount of DMSO. Cells (1 × 10⁶) were seeded in T25 cell culture flasks. Culture media was replaced with media containing the inhibitor or vehicle after 24 h and incubated in hypoxia or normoxia for 24 h before cells and cell culture media were harvested for analysis. Cell culture media were centrifuged (1000 rpm, 10 min) to remove cell debris, upon which samples were stored at −80 °C until analysis.

Preparation of protein lysates

Harvested cells were washed twice with ice-cold phosphate-buffered saline (PBS, Sigma) and collected using a cell scraper. Cells were lysed in mammalian protein extraction reagent supplemented with protease and phosphatase inhibitors (Thermo Fisher Scientific, Waltham, MA). Lysates were centrifuged for 15 min (14 000 × g, 4 °C) to remove cell debris, and the supernatant was collected and stored at −80 °C.

Western immunoblot analysis

Whole cell lysates (7.5 µg/ml protein) or cell culture media were separated on 4–12% NuPAGE® Novex Bis–Tris Gels in NuPAGE® MES SDS running buffer (Life Technologies, Carlsbad, CA) and transferred to Immobilon-P membranes (Millipore, Bedford, MA). Membranes were blocked for 1 h at room temperature in Tris-buffered saline with 0.1% Tween-20 (TBST) and 5% non-fat dry milk and incubated overnight at 4 °C with mouse anti-CAIX antibody (M75; a kind gift from Silvia Pastorekova), mouse anti-HIF-1α antibody (BD Transduction Laboratories, Franklin Lakes, NJ), goat anti-actin antibody (Santa Cruz Biotechnology, Santa Cruz, CA) or rabbit anti-GAPDH antibody (Cell Signaling Technology, Beverly, MA). After washing in TBST, the membranes were incubated for 1 h at room temperature with appropriate horseradish peroxidase conjugated secondary antibody (DAKO, Glastrup, Denmark). Bands were visualized using Super Signal West Dura Extended Duration Substrate (Thermo Fisher Scientific).

Measurement of pH and lactate in cell culture media

Extracellular pH (pHe) was measured in cell culture media using a pH-meter (SevenEasy™ instruments, Mettler Toledo, Columbus, OH) immediately after treatment. Analysis of lactate in harvested cell culture media was performed with an in vitro blood gas analyzer (GEM Premier 4000, Instrumentation Laboratory, Bedford, MA).

Cell proliferation

After 24 h of treatment with 100 μM S4 under normoxic and hypoxic conditions, cells were washed once with PBS before being detached with trypsin. Detached cells were mixed well in pure cell culture media and an aliquot was counted using an automatic cell counter (Countess automated cell counter, Life Technologies).

Clonogenic survival

Single cells were plated and left to attach for 6 h before treatment with 100 µM S4 for 18 h under normoxic and hypoxic conditions. Fresh culture media without inhibitor was then added and cells were incubated under normoxic conditions for 10–14 days. Colonies were fixed in ice-cold methanol and stained with 0.05% crystal violet. Colonies containing >50 cells were counted to calculate surviving fractions. At least three parallel samples were scored in three independent biological experiments for each treatment condition. Surviving fractions are given as mean ± SEM. Plating efficiency determined from control experiments was 0.61 ± 0.03 in HT29 cells and 0.40 ± 0.02 in HCT116 cells.

Intracellular pH measurement

Intracellular pH was measured using the pH sensitive fluorophore, carboxy SNARF-1 AM acetate (SNARF-1, Life Technologies). Cells were trypsinized, centrifuged and resuspended in PBS before dye-loading with 5 µM SNARF-1 for 15 min at room temperature. Stained cells were centrifuged and PBS added to a final concentration of 1 × 10⁶ cells/ml, and fluorescence intensity was measured instantly on a flow cytometer (BD LSR II, BD Bioscience, San Jose, CA). The nigericin calibration method was used to determine pHi from SNARF-1 fluorescence intensity23.

Flow cytometry

Cells were fixed in ice-cold methanol and stored at −20 °C until analysis. On the day of analysis, cells were washed with ice-cold PBS and centrifuged at 2000 rpm for 5 min. A second washing step was performed with PBS containing 1% human serum albumin (HSA), before 30-min incubation with 100 µl of primary antibody (M75) against CAIX in room temperature. Unbound antibody was removed by washing with PBS containing 1% HSA before 30-min incubation with the secondary FITC-anti mouse antibody (DAKO). Finally, cells were washed and dissolved in PBS, before measuring fluorescence intensity on a flow cytometer (BD LCR II). Median fluorescence intensity was determined in each sample.

Enzyme-linked immunosorbent assay

Carbonic anhydrase IX was quantified in cell culture media using a commercially available enzyme-linked immunosorbent assay (ELISA) (Quantikine human CAIX, R&D systems, Minneapolis, MN) according to the manufacturer’s manual. Briefly, 50 µl of cell culture medium was added to a micro titer plate pre-coated with anti-CAIX antibody and incubated on an orbital shaker for 2 h. After washing, an enzyme-linked polyclonal antibody against CAIX was added followed by 2-h incubation. Following another washing step, a substrate solution was added for 30 min before signal intensity was measured at 450 nm on a microplate reader (Modulus™ Microplate Multimode Reader, Turner BioSystem, Sunnyvale, CA). To correct for optical imperfections, plate readings at 540 nm were subtracted. Finally, the concentration of CAIX was calculated from a standard curve. All the samples were analyzed in duplicate.

Statistical analysis

Differences between groups were analyzed using the two-sided Student’s t test under conditions of normality and a non-parametric test (Mann–Whitney U) under other conditions. p Values <0.05 were considered statistically significant.

Results

CAIX expression under various culturing conditions

Western immunoblot analysis revealed induction of HIF-1α in response to hypoxia in HT29, KM20L2 and HCT116 cells, however CAIX expression levels differed between the three cell lines. HT29 cells showed relatively high CAIX expression under normoxia, which increased further under hypoxia. In KM20L2 cells, CAIX expression was highly induced under hypoxia, from relatively low levels under normoxia, whereas only a slight induction was observed in HCT116 cells under hypoxia from undetectable expression under normoxia (Figure 1A). Flow cytometric analysis of CAIX surface expression was in accordance with expression levels seen in whole cell lysate from HT29 and KM20L2 cells (Figure 1A and B). However, in HCT116 cells, CAIX surface expression was also observed under normoxia, and no further induction could be detected under hypoxia as seen in whole cell lysate (Figure 1A and B). Both -buffered and -free media were used in experiments and the influence of buffer composition on CAIX expression was analyzed in HT29 cells. CAIX expression was similar under normoxia and the induction in response to hypoxia was almost identical for both buffer compositions (Supplementary Figure S1).

Figure 1. Carbonic anhydrase IX (CAIX) expression in HT29, KM20L2 and HCT116 cells. (A) Western immunoblot analysis of CAIX expression in whole cell lysate under normoxia (21% O₂) and hypoxia (1% and 0.2% O₂) after 24-h incubation in -buffered medium. Actin was used as loading control. (B) Flow cytometric analysis of CAIX surface expression under normoxia (21% O₂) and hypoxia (0.2% O₂). The values represent the surface expression as a ratio of the median fluorescence relative to unstained control cells and is the mean of minimum three independent experiments (**p < 0.001).

S4 counteracted hypoxia-induced extracellular acidification under –free conditions

A well-known effect of CAIX activity is acidification of the extracellular environment. The ability of S4 to modify this effect was investigated by measuring pHe in cell culture media under normoxic and hypoxic conditions (Figure 2). As expected, incubation under hypoxia resulted in acidification of the media for the three cell lines. The hypoxia-induced decline in pHe was 0.4–0.5 pH units relative to the normoxic control condition and was similar for all three cell lines and buffer systems, although baseline pHe values varied. In -free medium, treatment of HT29 and KM20L2 cells with 100 µM S4 significantly reduced the hypoxia-induced acidification (Figure 2B). A similar trend was observed under the same conditions for HCT116 cells, although not statistically significant. In -buffered medium, S4 did not inhibit acidification under hypoxia in neither HT29 nor HCT116 cells (Figure 2A). Under normoxia, S4 treatment resulted in acidification of the cell culture media. This change was statistically significant in HT29 cells when grown in -free medium only (Figure 2B), although a similar trend was seen for HCT116 cells under -buffered conditions (Figure 2A). Since the effects of S4 on pHe were most apparent in -free media, this incubation condition was chosen for further experiments. Hypoxia-induced metabolic shift toward glycolysis is known to result in increased production of lactic acid. Hypoxia alone resulted in a doubling of lactate medium concentration in HT29 cells, and S4 treatment partially counteracted this effect. In contrast, S4 treatment resulted in significantly increased medium lactate concentration under normoxia (Figure 3).

Figure 2. Extracellular pH (pHe) measurements in cell culture media from HT29, KM20L2 and HCT116 cells, after 24-h treatment with 100 µM S4 under either normoxic (21% O₂) or hypoxic (0.2% O₂) conditions. Cells were incubated in either -buffered media (A) or -free media (B). Data represent the mean ± SD of the mean of minimum three independent experiments (*p < 0.05 and **p < 0.001).

Figure 3. Lactate measurements (mmol/l) in cell culture media from HT29 cells after treatment with 100 µM S4 for 24 h under normoxia (21% O₂) and hypoxia (0.2% O₂). Data represent the mean ± SD of the mean of minimum three independent experiments (*p < 0.05).

Effects of S4 treatment on cell proliferation and clonogenic survival

One explanation for the observed S4 effects on pHe and lactate in media might be that S4 influences proliferation, and this was assessed by counting cells after S4 treatment under normoxic and hypoxic conditions. In HT29 and KM20L2, hypoxia neither influenced cell proliferation, nor did S4 treatment. In contrast, in HCT116 cells, both hypoxia and S4 treatment significantly reduced cell proliferation. However, the reduced proliferation did not influence pHe (Figure 4A). Observing no direct effect of S4 on cell proliferation in HT29 cells, while pHe effects were apparent, we questioned whether S4 treatment might influence clonogenicity. Following exposure to S4, the surviving fraction of hypoxic HT29 cells was significantly lowered (0.80 ± 0.02). In contrast, no change in clonogenicity was seen after normoxic S4 treatment (1.05 ± 0.08). In HCT116 cells, no effect on clonogenicity was observed under either normoxic (1.05 ± 0.09) or hypoxic (1.16 ± 0.09) S4 treatment (Figure 4B).

Figure 4. Cell proliferation and clonogenic survival after S4 treatment of HT29, KM20L2 and HCT116 cells with 100 µM S4 under normoxia (21% O₂) and hypoxia (0.2% O₂). (A) Cell proliferation; bars represent the number of cells/ml after 24-h incubation. (B) Clonogenic survival; bars represent the surviving fraction. The data are illustrated as the mean ± SD of the mean of minimum three independent experiments (*p < 0.05).

Effects of S4 treatment on intracellular pH

Since maintenance of stable pHi is essential for cell survival, the ability of S4 to influence pHi was investigated under normoxic and hypoxic conditions in HT29 cells (Figure 5). Hypoxia caused an increase in pHi by 0.1 units with no further effects observed upon addition of S4. Under normoxia, a slight increase of pHi was detected upon S4 treatment.

Figure 5. Measurements of intracellular pH (pHi) in HT29 cells by flow cytometry, using the pH sensitive fluorophore, Carboxy SNARF-1 AM acetate. A calibration curve was made for each experiment to relate the fluorescence intensity to pHi. The bars illustrate the difference in pHi (ΔpHi = pHi_Treatment − pHi_{Normoxic control}) and represent the mean ± SD of the mean from minimum three independent experiments (*p < 0.05).

S4 treatment influenced CAIX expression by alterations in ectodomain shedding

Flow cytometric analysis revealed alterations in CAIX surface expression in response to S4 treatment, and the response differed between the three cell lines. In HT29 cells, S4 treatment induced a decrease in CAIX protein levels on the cell surface after both normoxic and hypoxic incubation (Figure 6A). The observed decrease in protein levels was confirmed on western immunoblot analysis of whole cell lysates from HT29 cells (Supplementary Figure S2). The reduction in CAIX surface expression was of the same magnitude under normoxia and hypoxia (20% and 21% decrease, respectively). In contrast, S4 treatment significantly increased surface expression in KM20L2 under hypoxia (62% increase), while a reduction was seen under normoxia (22% decrease). In HCT116, no changes in CAIX expression levels could be detected upon different treatment conditions. CAIX surface expression might be influenced by shedding of the CAIX ectodomain from the cell surface, and CAIX in culture media was therefore measured by ELISA (Figure 6B). The amount of CAIX detected in the cell culture media in the three cell lines varied according to their cellular expression levels. HT29 cells, which express high cellular levels of CAIX, displayed high concentration in media; 1091 ± 101 and 865 ± 80 pg/ml under normoxia and hypoxia, respectively. KM20L2 cells, expressing low cellular CAIX levels under normoxia, had relatively low CAIX concentration in media (115 ± 22 pg/ml), which increased under hypoxia (316 ± 30 pg/ml). In HCT116, CAIX concentration was undetectable in media under normoxia, whereas very low levels were detected under hypoxia (15 ± 0.50 pg/ml). Western immunoblot analysis was performed comparing the molecular size of CAIX detected in cell culture media and whole cell lysates. In cell culture media a band pair with a decreased size (∼4 kDa), consistent with the predicted size of the CAIX ectodomain, was detected from HT29 and KM20L2 cells, whereas the CAIX concentration was too low to be detected from HCT116 cells (Figure 6C). In HT29 cells, S4 treatment resulted in increased CAIX shedding both under normoxia and hypoxia (29% and 76% increase, respectively). In KM20L2 cells, the opposite effect was observed; S4 treatment resulted in decreased CAIX medium concentrations under both normoxia and hypoxia (16% and 32% decrease, respectively). In HCT116 cell culture media, CAIX could not be detected under normoxia, but S4 treatment resulted in decreased levels under hypoxia (38% decrease) (Figure 6D).

Figure 6. Carbonic anhydrase IX (CAIX) expression in HT29, KM20L2 and HCT116 cells after 24-h treatment with 100 µM S4 under normoxic and hypoxic conditions. (A) Flow cytometric analysis of CAIX surface expression. The values represent the surface expression as a ratio of the median fluorescence relative to the normoxic control (*p < 0.05, **p < 0.001). (B) CAIX concentrations in cell culture media were measured by enzyme-linked immunosorbent assay (ELISA). The bars represent the concentrations (pg/ml) and are illustrated as the mean ± SD of the mean from minimum three independent experiments (**p < 0.001). (C) Western immunoblot analysis of total cell lysate from HT29 cells incubated under hypoxia and cell culture media from HT29, KM20L2 and HCT116 cells after hypoxic incubation comparing the size of intact CAIX to the shorter version consistent with the CAIX ectodomain (TCL, total cell lysate; M, cell culture medium). (D) CAIX concentration in cell culture media was measured by ELISA. The bars represent the concentration as a ratio relative to the normoxic control for HT29 and KM20L2 cells and a ratio relative to hypoxic control for HCT116 cells and are illustrated as the mean ± SD of the mean from minimum three independent experiments (*p < 0.05, **p < 0.001).

Discussion

Carbonic anhydrase IX catalyzes the reversible hydration of CO₂ to and H⁺ and its activity is influenced by the amount of available CO₂ and 24. Therefore, levels present in the cell culture medium may be of importance when investigating CAIX activity12. In our study, the buffer system (-buffered or -free medium) did not influence CAIX expression, but was crucial when investigating the influence of S4 treatment on pHe. S4-induced changes in pHe were only detectable under -free conditions. In order for CAIX to catalyze , the reaction must be driven out of equilibrium, hence in a perfectly balanced system its role becomes redundant, which might explain why S4 did not influence CAIX activity under -buffered conditions25. Although the role of the buffer system was relatively clear in our experiments, varying results have been obtained by others, which illustrate the experimental complexity and underline the importance of controlling the experimental conditions11^,12^,26.

Hypoxia creates a shift in metabolism toward glycolysis, where the final product is lactic acid27. Lactic acid contributes to extracellular acidification by two mechanisms, either by direct transport out of the cell or by donating H⁺ intracellularly to generate CO₂, which diffuses out of the cell to become a substrate for acidification through CAIX activity25. The best known effect of CAIX’s contribution to extracellular acidification is through the hydration of CO₂11. However, CAIX has also been shown to facilitate lactic acid extrusion, suggesting that CAIX inhibition can influence both mechanisms of lactic acid contribution to extracellular acidification28^,29. To investigate how S4 inhibited the hypoxia-induced acidification, the contribution from direct lactic acid extrusion was determined by measuring lactate in the cell culture media of HT29 cells. In accordance with the literature, the extracellular lactate concentration increased under hypoxia30. Interestingly, S4 treatment reduced hypoxia-induced lactate accumulation in the media, suggesting that S4 inhibited CAIX-dependent extrusion of lactic acid. However, there have also been reports showing that lactate extrusion can be independent of CAIX expression levels, thus S4 might influence this process through a CAIX-independent mechanism11^,26.

S4 has previously shown to inhibit cell proliferation in colorectal and breast cancer models in vitro, both under normoxic and hypoxic conditions22^,31. In our study, S4 treatment did not influence proliferation when analyzed immediately after 24-h incubation (0.2% O₂) in HT29 and KM20L2 cells, whereas a slight reduction in proliferation was seen after S4 treatment under normoxia in HCT116 cells. Others have reported reduced proliferation when studied 4 days after 24 h of S4 treatment under anoxic conditions (0% O₂) in both HT29 cells and HCT116 cells (90% and 60% reduction in proliferation, respectively). When studying clonogenicity, which is a more robust survival endpoint, S4 treatment resulted in reduced survival in HT29 cells under hypoxia only, whereas no effect was seen in HCT116 cells. Thus, it seems that the method and the timing used for measuring survival and proliferation is critical for the end results. The mechanisms for inhibiting cell survival by CAIX inhibitors are not well understood, but have previously been related to CAIX’s role in maintaining a stable pHi in acidic environments32–34. To investigate if the reduced clonogenicity observed in HT29 cells was a result of such interference, pHi was measured. Hypoxia alone resulted in a weak pHi increase by ∼0.1 units, which was in agreement with another report32, illustrating that HT29 cells were able to maintain a relatively stable pHi under hypoxia and thus more acidic conditions. Furthermore, no alterations in pHi were observed upon addition of S4 under hypoxia, and pHi change can therefore not explain the reduced clonogenicity seen under hypoxia in HT29 cells.

Under normoxia, very different S4-induced effects were observed compared to treatment under hypoxia. For instance, a decline of pHe was detected upon S4 treatment in HT29 cells as opposed to the increase seen under hypoxia, a decline that corresponded with an increase in extracellular lactate. Additionally, S4 treatment caused a slight increase in pHi under normoxia, but no effects were seen on clonogenicity. This highlights that the cellular effects caused by CAIX inhibition depends on cellular oxygenation, and although increasing evidence supports CAIX’s role in assisting tumor cell survival under hypoxic and acidic conditions, its role under normoxic and neutral conditions remains controversial. This might be of particular concern in a potential therapeutic context of inhibiting CAIX in heterogeneous tumors consisting of components with variable degree of oxygenation.

S4 treatment counteracted hypoxia-induced extracellular acidification in a CAIX-dependent manner and selectively inhibited clonogenicity under hypoxia in HT29 cells. The sulfonamide-based compounds are thought to inhibit CAIX by binding to the zinc ion in the catalytic site of the enzyme6, but previous studies have also shown that sulfonamide-based inhibitors can influence both cellular CAIX levels and enzyme-regulated proteolytic cleavage of membrane-bound CAIX, i.e. ectodomain shedding32^,34–36. In this study, S4 treatment influenced CAIX expression levels, which could be explained by alterations in CAIX ectodomain shedding. In HT29 cells, decreased CAIX levels were detected in both total cell lysates and on the cell surface, and the effect was independent of oxygen levels. Interestingly, the decline in CAIX surface expression on HT29 cells in response to S4 treatment was accompanied by a substantial increase of CAIX in the cell culture medium, suggesting that shedding of CAIX from the surface could explain this observation. Western immunoblot analysis confirmed that CAIX detected in medium was indeed the truncated form of the protein previously reported to represent the CAIX ectodomain17. In KM20L2 cells, S4 treatment had the opposite effect; leading to increased CAIX surface expression under hypoxia, accompanied by decreased ectodomain shedding. In the CAIX low-expressing HCT116 cells, S4 treatment also resulted in decreased CAIX levels in the cell culture medium under hypoxia, although no alterations in CAIX expression were observed on the cell surface. Membrane-bound CAIX undergoes constitutive shedding in a regulated process involving metalloproteinases37. However, CAIX ectodomain release can also be activated in response to a range of different stimuli. For instance, enhanced shedding has been demonstrated in response to cytotoxic therapy and immunotherapy, and has been suggested to be an effect of cell death37^,38. The increased shedding seen in response to S4 treatment in HT29 cells could be such a cellular death response. However, even though the largest increase in shedding was observed under hypoxic S4 treatment, where clonogenicity was reduced, increased shedding was also observed under normoxia and can therefore not solely explain this effect.

In a previous study, S4 treatment decreased the rate of ectodomain shedding detected in blood of a laryngeal tumor mouse model, which is in accordance with the decreased shedding observed in response to S4 treatment in KM20L2 and HCT116 cells36. Interestingly, sulfonamide-based inhibitors have previously shown to efficiently inhibit metalloproteinases; the enzymes involved in cleaving CAIX from the cell surface39. It is therefore tempting to speculate that the observed decrease in CAIX shedding could be an effect of inhibiting the action of metalloproteinases involved in CAIX cleavage.

The different effects seen in response to S4 treatment on ectodomain shedding could be a result of varying mechanisms causing ectodomain shedding, which might be cell type dependent. If S4 and related compounds are to be used in a clinical setting it is important to address the implications of CAIX shedding, and further investigate the mechanism that causes the different responses to shedding observed here.

Conclusions

Treatment with the CAIX inhibitor S4 counteracted hypoxia-induced extracellular acidification in CRC cell lines and specifically inhibited clonogenicity under hypoxia in HT29 cells. Under normoxia, effects of S4 were opposite or absent compared to those observed under hypoxia, underlining the importance of investigating the effects of CAIX inhibition under different culture conditions. Importantly, treatment with S4 brought about alterations in ectodomain shedding of CAIX in a cell-dependent manner, representing a possible mechanism for how sulfamate inhibitors may modify CAIX. The detailed mechanisms for effects on CAIX shedding are unclear, as is the impact of the oxygenation level on CAIX activity, both of which merit investigation to understand more completely how sulfamates and related compounds influence CAIX activity and how such drugs may be of use in cancer treatment.

Supplementary material available online

Supplementary Figures S1-S2

Acknowledgements

The authors thank Prof. Silvia Pastorekova for the generous gift of the CAIX antibody (M75) and Prof. Claudiu Supuran for providing the CAIX inhibitor (S4).

Declaration of interest

The authors report no declarations of interest. This work was supported by the European Union 7th Framework Programme Grant 222741-METOXIA (to A.H.R. and K.F.) and Akershus University Hospital Grant No. 2014011 (to A.H.R.). H.H.H. is Research Fellow at University of Oslo. K.R.R. is Postdoctoral Research Fellow supported by South-Eastern Norway Regional Health Authority Grant No. 2012002 (to A.H.R.).