Indospicine combined with arginine deprivation triggers cancer cell death via caspase-dependent apoptosis

Reagents

Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS), dialyzed FBS, gentamicin, l-arginine monohydrochloride (Arg), l-methionine, l-leucine, l-lysine monohydrochloride, rapamycin (Rapa), and cycloheximide (CHX) were purchased from Sigma-Aldrich (St. Louis, MO, USA). l-canavanine sulfate (Cav) was obtained from PAN Biotech GmbH (Aidenbach, Germany). Indospicine (Isp) was kindly provided by Associate Professor Fletcher M.T. Recombinant human arginase I (rhARG) used in this work was expressed as a secretory protein in the methylotrophic yeast Ogataea polymorpha and affinity-purified (our unpublished data).

Cell lines and culture conditions

Human colorectal carcinoma HCT-116, SW480, and SW620 as well as human immortalized keratinocyte cell lines HaCaT were obtained from the ATCC (Manassas, VA, USA). Cell cultures were routinely tested negative for mycoplasma contamination and were genetically validated at the Institute of Legal Medicine (TU Dresden, Germany) as previously detailed (Kurlishchuk et al., 2016) prior to use.

Cells were routinely incubated in DMEM supplemented with 50 mg/L gentamicin and 10% FBS at 37°C in a humidified atmosphere with 5% CO₂. Arg deprivation was achieved by specially formulated DMEM (HyClone Laboratories, South Logan, UT, USA), lacking the amino acid Arg or by adding to the Arg-sufficient complete medium (CM) Arg degrading enzyme rhARG at a concentration of 2 U/mL. Arg-free medium (AFM) and the respective Arg-sufficient CM were supplemented with 10% dialyzed FBS devoid of small molecules including amino acids (<10 kDa cut off). Both means of selective Arg deprivation were shown previously to be identically effective in lowering intracellular Arg level and triggering cell cycle arrest in various tumor cell lines in vitro (Vynnytska-Myronovska et al., 2012, 2016; Hinrichs et al., 2018).

Assay of arginase activity

The activity of rhARG was assayed in phosphate-buffered saline (PBS) with 0.5 mM Arg and/or 0.5 mM Isp, or Cav and monitored via the level of urea production. For inhibition experiments, the reaction mixture was supplemented with 0.5 mM Arg analogues and pre-incubated for 2 h. After 1 h incubation of the reaction mixture at 37°C the resulting urea concentration was assessed spectrophotometrically at 520 nm after the addition of a color-producing reagent and heating at 100°C for 10 min, as described before (Marsh et al., 1965). The standard urea solution (Simko, Ukraine) and PBS were used as positive and negative controls, respectively.

Reverse-transcription polymerase chain reaction (RT-PCR)

Total RNA was isolated using the GeneMATRIX Universal RNA Purification Kit (EURx, Gdansk, Poland) according to the manufacturer's instructions. Complementary DNA was synthesized from 2 μg of total RNA using the NG dART RT kit (EURx). Primer pairs for genes of interest were as follows: CHOP (forward, F: 5′-ATGGCAGCTGAGTCATTGCC-3′, reverse, R: 5′-AGCTGTGCCACTTTCCTTTC-3′); for GRP78 (Forward, F: 5′-TGTGCTCTCTGGTGATCAAG-3′, reverse, R: 5′-CGATTTCTTCAGGTGTCAGG-3′); MYC (forward, F: 5′-GGATTCTCTGCTCTCCTCGAC-3′, reverse, R: 5′-GCTGTGAGGAGGTTTGCTG-3′); ACTB (forward, F: 5′-TGCGTCTGGACCTGGCTG-3′, reverse, R: 5′-CTGCTGGAAGGTGGACAG-3′) was used as an internal control. PCR fragments were separated on 1.5% agarose gels and visualized by ethidium bromide staining. Relative intensity of the bands was quantified with Gel-Pro Analyzer version 3.2 and presented after normalization to the reference gene (ACTB).

Western blotting

Cells were homogenized on ice in lysis buffer as previously described (Vynnytska et al., 2011). Equal amounts of total proteins were resolved on 8–12% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride membrane (Millipore Corporation, Bedford, MA, USA). Membranes were incubated with primary antibodies overnight at 4°C. Antibodies against the following antigens were used: phospho-Akt (Ser473) (#4060), Akt (#9272), phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204) (#14227), p44/42 MAPK (Erk1/2) (#4695), phospho-p38 MAPK (Thr180/Tyr182) (#9211), p38 MAPK (#9212), phospho-p70 S6 kinase (Thr389) (#9234), p70 S6 kinase (#2708), phospho-S6 ribosomal protein (Ser240/244) (#2215), S6 ribosomal protein (#2217), c-Myc (#5605), cleaved caspase-9 (Asp330) (#9501), cleaved caspase-3 (Asp175) (#9661), cleaved poly (ADP-ribose) polymerase (PARP) (Asp214) (#5625), all from Cell Signaling Technology, Danvers, MA, USA; GRP78 BiP (ab21685) from Abcam, Cambridge, UK; β-actin (A5441) from Sigma-Aldrich. After 1 h of exposure to horseradish peroxidase-conjugated goat anti-mouse (AP308P) or anti-rabbit (AP307P) IgG secondary antibodies (Millipore) at room temperature, immunoreactive proteins were visualized with Western Blotting Luminol Reagent (sc-2048; Santa Cruz Biotechnology, Heidelberg, Germany). All experiments were reproduced at least twice and the duplicate protein samples were subjected to a minimum of three independent western blot detections.

Cell viability detection by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay

MTT reagent was purchased from Sigma-Aldrich. The assay was performed as described in our previous report (Vynnytska et al., 2011). Cell viability is presented in percent from the levels of the untreated controls at 0 h.

Statistical analysis

Mean values ± standard deviation from repeated experiments are shown and the number of independent experiments is indicated in the figure legends. Statistical analysis was performed using the Student's t test. Differences of the means were considered significant when the calculated P < 0.05.

Indospicine does not inhibit arginase activity and is not an arginase substrate

It could not be excluded that Isp, due to its structural similarity to the proteinogenic amino acid arginine (Figure 1), may potentially inhibit arginase enzymatic activity, thus rendering a proposed combined treatment approach of Arg starvation with Isp inefficient. To address this question, we measured the specific activity of produced in yeasts rhARG in PBS in the absence or presence of Isp at equimolar to Arg concentration, as described in Materials and Methods. By monitoring urea production it was revealed that Isp at a 0.5 mM concentration had no significant inhibitory effect on rhARG activity even after previous pretreatment with Isp (Table 1). To the contrary, increased urea production was found when another Arg analogue Сav was present at the same concentration. This data also suggests, that, unlike Cav, Isp is not a catalytic substrate of rhARG.

Arginine starvation profoundly increases indospicine cytotoxicity

We compared the cytotoxic effect of Isp in three human colorectal cancer (CRC) cell lines (HCT-116, SW480, and SW620) under Arg-rich and Arg-deficient culture conditions. For this purpose, the cells were treated with Isp at a broad range of concentrations (0, 0.025, 0.05, 0.1, and 1 mM) in both complete medium with 0.4 mM Arg (CM) and Arg-free medium (AFM). In addition, CM pre-incubated with rhARG (2 U/mL) for 24 h before cell exposure was applied as Arg-depleted condition. Cell viability was determined after 24, 48, and 72 h of incubation via the MTT assay. The amount of Isp required to kill 50% of the cells in culture was defined as IC₅₀ (inhibitory concentration 50%).

No cytotoxic effect of Isp was observed in HCT-116 cells incubated in CM medium at concentrations of up to 1 mM. The same was true when cells were exposed to AFM for 24 h. However, at 48 h in AFM, Isp at concentrations of 0.025 mM led to 50% cell death in HCT-116 cultures; the effect became more pronounced in further time-course of the treatment. Accordingly, this Isp IC₅₀ concentration for HCT-116 cells was applied in subsequent experiments (Figure 2A).

SW480 and SW620 colorectal carcinoma cells were also sensitive to the cytotoxic effect of Isp in AFM, much more pronounced for the latter cell line. In SW620 cultures, some dose-dependent decrease in the number of viable cells was observed even in Isp-supplemented CM (Figure 2B). Overall, the obtained results suggest that Arg deprivation profoundly sensitizes human colon cancer cells to the cytotoxic effects of Isp. However, non-transformed human keratinocyte HaCaT cells appeared to be much less sensitive to Isp action under Arg deprivation, exhibiting IC₅₀ of 1 mM, that is, at least 40 times higher than all tested CRC cells (Figure S1A).

Impact of combined treatment of arginine deprivation and indospicine on cell signaling and apoptosis

HCT-116 cells were incubated in AFM alone or in AFM supplemented with 0.025 mM Isp for 24 and 72 h to evaluate the cumulative effects of Isp and Arg deprivation on cell signaling regulating survival and proliferative potential. For the analysis of MAPK and mammalian target of rapamycin (mTOR) pathways, c-Myc, Akt, markers of endoplasmic reticulum (ER) stress and apoptosis were assessed by western blot analysis and/or RT-PCR as described in the Materials and Methods section.

We first analyzed the level of phosphorylated Akt protein in Arg-starved HCT-116 cells in the presence and absence of Isp. Indeed, Arg deprivation alone caused a drastic increase in Akt phosphorylation in comparison to CM condition, whereas Isp treatment under AFM for 72 h resulted in a pronounced dephosphorylation of Akt and decrease in its protein level (Figure 3A).

We also observed an increase in the phosphorylation of ERK1/2 and p38 MAPK proteins under arginine deprived conditions, which, in contrast to Akt, was further enhanced by Isp treatment (Figure 3A). Apparently, Isp evokes a strong activation of the MAPK/ERK axis in Arg-deprived cells.

The transcription factor c-Myc is often overexpressed in cancer cells and stimulates metabolic changes that are implicated in malignant transformation (Miller et al., 2012). We observed a pronounced specific decrease in c-Myc messenger RNA (mRNA) and protein levels in HCT-116 cells incubated for 72 h with Isp in AFM but not in cells exposed to AFM alone or CM + Isp, as demonstrated by semi-quantitative RT-PCR (Figure 3B) and western blot analyses (Figure 3C).

In line with previous reports (Vynnytska-Myronovska et al., 2016), long-term incubation of the cancer cells in AFM downregulated mTOR activity as evidenced by a more or less pronounced decrease in the phosphorylation levels of its downstream substrates p70S6 kinase (p70S6K) and S6 protein. This trend was reversed in the presence of Isp, indicating a transient activation of mTOR and of total protein translation in the cells by this Arg analogue (Figure 3D).

We reported previously that Arg starvation induces ER stress in solid cancer cells (Bobak et al., 2016). Therefore, ER stress markers GRP78 and CHOP were also monitored in Isp-treated HCT-116 cells. As expected, Arg deprivation alone significantly increased GRP78 and CHOP mRNAs relative to untreated cells. Co-treatment with Isp led to a certain increase in CHOP but not GRP78 expression (Figure 4A). However, western blot analysis revealed that GRP78 protein level also gradually increased at 72 h of the co-treatment (Figure 4B).

The alterations in PI3K-Akt-mTOR, MAPK/ERK signaling, as well as ER stress described above could directly affect apoptosis. Accordingly, HCT-116 cells treated with Isp in AFM for 72 h showed clear caspase 3 activation and PARP cleavage, whereas AFM mono-treatment did not induce apoptosis (Figure 4C). These results were essentially confirmed by cells Hoechst/PI staining demonstrating profound drop in the number of viable cells and induction of apoptosis under combined treatment with Isp (Figure S2).

The observed effects (Figures 3 and 4) tentatively indicate that Isp is misrecognized by cellular receptors and may be incorporated into nascent proteins in place of Arg, thus leading to the observed elevated ER stress and apoptosis.

Inhibition of protein synthesis rescues cells viability upon combined AFM and Isp treatment

To evaluate whether Isp is a proteinogenic amino acid in human cancer cells and whether this contributes to its cytotoxic effects, HCT-116 cells were incubated in AFM with or without Isp in the presence of the translation inhibitors rapamycin (Rapa, 0.04 µM; an inhibitor of mTOR signaling) and cycloheximide (CHX, 0.01 mM; inhibitor of peptidyltransferase). Cell viability was monitored after 72 h of incubation by the MTT test.

As shown in Figure 5A, both Rapa and CHX exhibited minor cytotoxic effects under AFM conditions, but at the same time significantly elevated cell viability in Isp-co-treated cells. The observed protective effect of CHX was concomitant with the inhibition of caspase-dependent apoptosis as evidenced by the decrease in the levels of cleaved forms of caspase-9, caspase-3, and PARP (Figure 5B).

Thus, the observed cytotoxic effects of Isp towards Arg-deprived human CRC cells are most probably due to its misincorporation into nascent proteins, together with its impact on amino acid signaling.