Caulerpin (Fig. 1A) of 99% purity was purchased from Yuanye Pharmaceutics (Shanghai, China). A stock solution of metformin was made at a concentration of 100 mM in DMSO and stored at −20°C. KN93, 2DG and 3BP were obtained from Sigma (Shanghai, China). Antibodies against PARP, p-p70S6K, p70S6K, p-S6, S6, p-4EBP1, 4EBP1, AMPKα1, AMPKα2 and β-actin were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). AMPKα, p-AMPKα (Thr172), p-ACC (Ser79), ACC, procaspase-7, procaspase-9 antibodies were purchased from Cell Signaling Technology (Denver, MA, USA). Horseradish peroxidase-labeled secondary antibodies were obtained from Promega (Madison, WI, USA).

The human colorectal cancer cell lines HCT116 and HT29 were obtained from ATCC. LOVO and SW480 were obtained from the Cell Bank of the Shanghai Institute of Biochemistry and Cell Biology, Shanghai, China (http://www.cellbank.org.cn). All cell lines were cultured in DMEM medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin and 100 µg/ml streptomycin and maintained in a humidified atmosphere at 37°C with 5% CO₂. For glucose deprivation experiments, medium was replaced with DMEM containing non-glucose (Invitrogen, Carlsbad, CA, USA).

The effects of drug treatment on the viability of colorectal cell lines were assessed with a cell counting kit-8 from Sigma (17), as described by the manufacturer. Briefly, 2×10⁴ cells were seeded into 96-well plates. After treatment, CCK-8 solution was added to each well and cells were further incubated at 37°C for 2 h. The absorption was measured using a multiskan spectrum microplate reader at 450 nm. All the experiments were repeated at least three times.

Briefly, the cells were treated with caulerpin for 24 h. Cells (1×10⁶) were centrifuged and harvested (18). Next, the cells were suspended in binding buffer, containing Annexin V-FITC and PI and then incubated for 15 min in the dark at room temperature. The stained cells were analyzed using a FACSCanto flow cytometer (BD Biosciences, Flanklin Lakes, NJ, USA).

Briefly, LOVO and SW480 cells were seeded in a 12-well plate at ~80% confluence. After receiving indicated treatment, cells were washed twice with phosphate-buffered saline (PBS) and incubated at 37°C for 30 min in buffer solution containing lipophilic cation JC-1 (KeyGen, Nanjing, China) or DCFDA (Invitrogen), cells were re-suspended and fluorescence was measured using a FACSCalibur (BD Biosciences). Two fluorescence parameters, channel 1 (green fluorescence, FL1) at 530 nm and channel 2 (red fluorescence, FL2) at 590 nm, were used to measure MMP (19). Bright green fluorescence produced by DCFDA means high level of ROS (20).

Cells following different treatments were lysed with RIPA buffer. Protein concentration was determined using a Coomassie blue staining method (21). An equal amount of protein (50 µg) was separated on 10% SDS-PAGE and transferred to PVDF membranes. The membranes were incubated with primary antibodies in 5% BSA at 4°C overnight. The membranes were washed and incubated with HRP-conjugated secondary antibody for 1 h at room temperature. The immunoreactive bands were detected with an ECL kit (Beyotime Institute of Biotechnology, Shanghai, China).

Rat liver mitochondria (0.3 mg protein/ml) and saponin-permeabilized or intact cells (5×10⁶) were incubated with caulerpin for indicated time. The oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) were measured using an XFe-96 Extracellular Flux Analyzer (Seahorse Bioscience, North Billerica, MA, USA). After attachment to XFe cell culture plates, cells were washed with XF assay medium and equilibrated for 1 h. ECAR were determined by the supplement of 2DG. ATP content was examined from OCR differences from basal rate after adding oligomycin. Different substrates were added to evaluate mitochondrial respiration when respiratory complex I to IV was stimulated respectively. Bioenergetic profiling to distinguish the site within the ETC controlled by caulerpin was performed as previously described (22). Measurements were normalized by total protein content with Lowry assay.

Two single guide RNAs (sgRNA) targeting unique sequences of the AMPKα1 gene were designed and cloned into the pSpCas9 (BB)-2A-Puro (PX459) V2.0 (Plasmid #62988; Addgene) plasmid, which contains a puromycin resistance cassette. sgRNAs were as follows: 5′-CACCGTACATTCTGGGTGACACGCT-3′ and 5′-AAACAGCGTGTCACCCAGAATGTAC-3′. After transfection, the cells were cultured in a medium containing puromycin for 48 h (23). Cells resistant to the drug were then selected by FACS for clonal expansion. Gene deficiency in LOVO cells was examined by western blotting with an anti-AMPKα1 antibody.

Scramble siRNA, siRNA targeting AMPKα2 and siRNA transfection reagent were purchased from Invitrogen (Carlsbad, CA, USA). Cells were transfected with Lipofectamine 2000 (Invitrogen) according to the instruction (24).

Four to 6-week-old athymic mice, which were obtained from SLAC Lab Animal Center of Shanghai (Shanghai, China), were housed under standard conditions. This experiment was conducted in accordance with the guidelines issued by the State Food and Drug Administration (SFDA of China). The experiments were approved by the Institutional Animal Care and Use Committee of Ningbo No. 2 Hospital, Ningbo, China. Cells (1×10⁷) in culture medium were injected subcutaneously into the right flank of each mouse (25). Tumor size was monitored by using micrometer calipers every other day for a 30-days period and calculated with a formula: tumor volume (mm³) = width² × length/2. When tumor volume reached ~75 mm³, nude mice were divided into three groups (6 mice/group): saline control, caulerpin (30 mg/kg, in saline with 0.2% CMC-Na), 3BP (15 mg/kg, in saline) or combination group. The oral gavage was executed every other day. It began on day 1, and finished on day 30. At the end of the experiments, all the mice were euthanized. Tumor tissues were used for immunohistochemistry assay.

Paraffin-embedded sections were heat-fixed, deparaffinized, and rehydrated through a graded alcohol series (100, 95, 85 and 75%). Antigen retrieval was performed with citrate buffer and treated with 3% H₂O₂ to block endogenous peroxidase activity prior to the treatment with primary antibodies (anti-PCNA, 1:100; anti-mTOR, 1:200). The tissues were then incubated with the secondary biotinylated antispecies antibody. Counterstaining was performed using hematoxylin (26).

Data are expressed as means ± SD. Statistical significance were performed using one-way ANOVA followed by unpaired Student's t-test. For comparison of multiple samples, the Tukey-Kramer test was used. All data were analyzed with GraphPad PRISM 5 (GraphPad Software, Inc., La Jolla, CA, USA).

In the present study, we examined the anticancer property of caulerpin in various colorectal cancer cell lines (HT29, LOVO, HCT116 and SW480). A growth inhibitory effect was observed in the four colorectal cell lines after 48 h exposure, with IC₅₀ values ranging from 20 to 31 µM (Fig. 1B). Among them, LOVO cells were most sensitive, with 50% inhibition rate at the concentration of 20 µM, while SW480 cells were least sensitive to caulerpin. As shown in Fig. 1C, caulerpin induced apoptosis to colorectal cells, though to a different extent. Caulerpin augmented the activities of PARP-1, caspase-9 and caspase-7 in LOVO cells, as indicated by their cleavage (Fig. 1D), but the broad-range caspase inhibitor z-VAD-fmk only partially prevented cell death, signifying that other caspase-independent signaling was engaged in the action of caulerpin (Fig. 1E).

Previous findings illustrated that caulerpin impaired mitochondrial ETC in cisplatin-resistant C13 cells and human breast cancer T47D cells (13,14), but its effects on oxygen consumption have not been fully addressed. Additionally, whether the emergence of an energy stress correlates with the anticancer activity of caulerpin in colorectal cells needs to be explored. In the present study, oxygen consumption was inhibited by ~60% in SW480 and LOVO cells after caulerpin treatment. The phenomenon was more obvious in permeabilized cells than intact cells (Fig. 2A). LOVO cells pretreated with caulerpin (20 µM, 2 h) were then permeabilized with saponin. Although malate and glutamate (ETC complex I substrates) failed to restore caulerpin-triggered downregulation of mitochondrial respiration in LOVO cells, complex II substrate succinate rescued mitochondrial respiration in LOVO cells pretreated with caulerpin, indicating that caulerpin caused impairment to complex I (Fig. 2B). The inhibitory effect of caulerpin on mitochondria respiration resembled that seen from rotenone, a complex I inhibitor (Fig. 2B). Moreover, in the incubation media containing succinate, respiration was suppressed by supplementation with complex III inhibitor antimycin A, proving that the ETC still works. A combination of TMPD/ascorbate (complex IV substrates) successfully rebooted respiration perturbed by antimycin A (Fig. 2C). In mitochondria isolated from rat liver, similar results were obtained. In the media containing malate and glutamate, oxygen consumption was suppressed due to caulerpin treatment, however, not in the presence of succinate (Fig. 2D). The results demonstrated that caulerpin brought about an intervention to mitochondrial function, via inhibition of mitochondrial complex I without interfering complexes II, III or IV.

Since mitochondrial function was impaired by caulerpin, we then examined mitochondrial membrane potential (ΔΨM) and ROS with flow cytometry, repectively. JC-1 is a cationic dye which accumulates in mitochondria in response to potential change, and the rise in the green/red fluorescence intensity ratio reflects mitochondrial depolarization. Herein, JC-1 fluorescent probes were utilized to analyze ΔΨM. In Fig. 3B, a significant advance in green/red fluorescence ratio (almost 3-fold vs. control) was obtained in both LOVO and SW480 cells treated with caulerpin, pointing to a reduction in ΔΨM. FCCP, an electron transport chain uncoupler, was applied as the positive control. Likewise, intracellular ROS concentration was measured by flow cytometry with the fluorescence dye DCF-DA. In a concentration-dependent manner, caulerpin rapidly upregulated ROS level at 1 h post-induction, and up to 24 h (Fig. 3C).

The metabolic profile of colorectal cancer cells was further examined with the Seahorse XF-e96 analyzer. Caulerpin treatment inhibited both maximal and basal respiration (Fig. 4A), accompanied by an increment in AMP/ATP ratio, which could be observed after 5 min and 1 h incubation with caulerpin in LOVO cells (Fig. 4B). AMPK is a crucial energetic sensor that is activated either by decreased cellular ATP levels or by ROS elevation (27). Activated AMPK will further suppress mTOR signaling for cellular adaptation in a stress condition (28). In the present study, caulerpin-triggered phosphorylation of AMPK and the phosphorylation of the downstream target ACC were observed in both cell lines, accompanied with inhibition of the downstream mTORC1 targets 4E-BP1 and S6 (Fig. 4C).

As a kinase responsive to oxidative activation upstream of AMPK, the essential role of CaMKII was explored to uncover the mechanism of caulerpin-elicited AMPK activation (29). As shown in Fig. 4D, the CaMKII inhibitor KN93 reduced caulerpin-induced ATP levels in colorectal cells, and induced a remarkable dephosphorylation of AMPK and ACC. These results signified that caulerpin-stimulated AMPK activation was CaMKII-dependent.

To further dissect the effect of AMPK isoforms on the viability of cells treated with caulerpin, the CRISPR-cas9 system was applied to knock out AMPKα1. crspAMPKα1 attenuated caulerpin-induced inactivation of mTOR pathway. Loss of AMPKα1 generated a compensatory growth in AMPKα2 expression (Fig. 5A). CCK-8 assay displayed that knock out of AMPKα1 protected cells against caulerpin-induced cytotoxicity (2 and 8 h), while siAMPKα2 exhibited no significant influence on cell death, which was induced by caulerpin treatment (Fig. 5B).

As a direct measure of ECAR, lactic acid was secreted by cells and thus serving as a hint for glycolytic activity. The baseline OCR/ECAR ratio was low in the SW480 cells and high in the LOVO cells, indicating a higher intrinsic glycolytic rate in SW480 cells than LOVO cells, which generate ATP primarily through OXPHOS (Fig. 6A).

We further examined whether caulerpin-triggered OXPHOS attenuation altered cell glucose metabolism. After 8 h treatment with caulerpin, glycolysis in LOVO cells was strengthened to compensate ATP loss, as revealed by significant increases in glucose uptake and lactate production (Fig. 6B and C). Moreover, upregulation of GLUT1, HKII, and 6-phosphofructo-2-kinase (PFKFB3) protein expression was also observed (Fig. 6D). Nevertheless, a prolonged exposure to caulerpin, unexpectedly, weakened glucose uptake in LOVO cells, as evidenced by downregulation of GLUT1 level after 48 h of caulerpin incubation (Fig. 6B and D). In the highly glycolytic SW480 cells, 8 h treatment with caulerpin exerted no significant effect on glucose metabolism; on the other hand, a remarkable decline of GLUT1 protein expression was observed after 24 h and continued after 48 h. Therefore, in spite of the initial attempt to adapt to the energy stress, long-term incubation with caulerpin eventually elicited a notable impairment to glucose metabolism.

To further characterize the biochemical events elicited by caulerpin and to examine its combination with the glycolytic inhibitor, we investigated nutrient signaling events and the AMPK kinase pathway. Glucose withdrawal promoted AMPK activation in both cell lines. Interestingly, levels of p-AMPK is higher in cells co-treated with caulerpin, 3BP (25 µM) and glucose than cells treated with glucose only. These data emphasized the key role of the glycolytic switch for compensating ATP generation during caulerpin incubation (Fig. 7A). In Fig. 7B, in vitro CCK-8 assays were applied to examine the cell viability. The IC₅₀ of 3BP is 41.2±1.6 and 38.9±1.6 µM for LOVO and SW480 cells, respectively. As can be seen from the surviving factions, the viability of cells treated with 3BP (25 µM) + caulerpin (60 µM) was significantly lower than cells treated with 3BP or caulerpin alone. The combination index of 3BP (25 µM) + caulerpin (60 µM) is 0.67 for LOVO cells and 0.21 for SW480 cells, suggesting a synergistic effect. When CI is >1, it means antagonism. So we concluded that caulerpin plus 3BP may act synergistically in inhibiting cellular proliferation in vitro (Table I).

To further assess the antitumor efficiency of the combination in vivo, athymic nude mouse model bearing SW480 implanted xenografts were treated with control, 3BP, caulerpin, or a combination of both agents. Tumor growth was slower in caulerpin monotherapy group than the vehicle group. No significant difference was observed in tumor size between mice treated with saline and 3BP. However, the combination of 3BP with caulerpin displayed remarkable tumor regression (Fig. 7C). Tumor xenograft sections were subjected to immunohistochemical analysis for p-mTOR and PCNA, which is an indicator for cell proliferation. PCNA and p-mTOR was significantly inhibited in combined treatment of 3BP with caulerpin (Fig. 7F).