Paclitaxel was from Sigma–Aldrich Inc. (St. Louis, MO, USA). Antibodies against p21 and p53 were from Abcam (Burlingame, CA, USA). HK1, PHGDH, GLS, SLC7A11, SLC1A5, and GDH were from Absin Bioscience Inc. (Shanghai, China). β-actin was from Sigma–Aldrich Inc. (St. Louis, MO, USA).

Human colorectal carcinoma cell line HCT116 was from the Cell Collection of the Chinese Academy of Sciences (Shanghai, China) and cultured in DMEM-F-12 medium (HyClone, Logan, UT, USA) supplemented with heat-inactivated 10% fetal bovine serum (Gibco, Carlsbad, CA, USA) and 100 IU/mL penicillin (Solarbio, Beijing, China). The cells were seeded into a gelatin-coated 75-cm² flask and cultured in 10 mL of medium at 37°C in a humidified atmosphere of 5% CO₂ in air.

Exponentially growing HCT116 cells were transferred to 12-well plates and cultured at 37°C in a 5% CO₂ atmosphere. Cells were treated with 0.1 and 0.3 nM PTX for 1–3 days. When the cells were at 60 to 70% confluence, they were rinsed twice with PBS, and the supernatant was discarded. Then, images were taken using an OLYMPUS IX 71 microscope (10 × 10) (OLYMPUS, Tokyo, Japan).

Cells in the logarithmic growth phase were digested into a single-cell suspension with a trypsin-EDTA (Solarbio, Tongzhou District, Beijing, China) solution, and then 10 mL of the cell suspension was seeded onto 9 cm culture plates (NEST, Wuxi, Jiangsu, China) at a density of 20 cells/mL. After adherence, these cells were treated with PTX (0.1 and 0.3 nM) for 72 h and then cultured for 15 days. Thereafter, the cells were fixed in formaldehyde and stained with 10% Giemsa stain (Solarbio, Tongzhou District, Beijing, China). After multiple washes, the plates were air dried and imaged. Individual clones were counted, and statistical analyses were performed.

The 3- (4, 5 - dimethylthiazol -2 - yl) - 2, 5 - diphenyltetrazolium bromide (MTT) colorimetric assay was used to determine cytotoxicity of PTX. HCT116 cells were plated at a density of 1000 and 2000 cells/well in 96-well plates and were incubated either with or without PTX at 0.1 or 0.3 nM for 3 days. Then, 50 μL of MTT 1 mg/mL (Sigma–Aldrich) tetrazolium substrate was added to each well, and plates were incubated for an additional 4 h at 37°C. The resulting violet formazan precipitate was solubilized by adding 100 μL of 50% N,N-dimethylformamide. All of these plates were shaken for 5 min and read immediately at 578 nm using a model 550 Micro Plate Reader (Bio-Rad, Richmond, CA, USA).

The cells were plated into 6-well plates and incubated for 1 and 3 days with nutrient solution containing various concentrations of PTX (0, 0.1, and 0.3 nM). Briefly, the cells were collected and fixed in 0°C 70% ethanol and stored at -20°C. The cells were then washed and resuspended in cold PBS and incubated at 37°C for 30 min with 10 mg/mL RNase and 1 mg/mL propidium iodide (Sigma–Aldrich, USA). DNA content analysis was performed by flow cytometry (BD, San Diego, CA, USA). The percentage of cells in different cell cycle phases was determined with Cell Quest acquisition software (BD Biosciences Pharmingen, USA).

Lactate generation was analyzed, following the manufacturer’s instructions. In brief, HCT116 cells were treated with low-dose PTX (0.1 and 0.3 nM) for 0, 1, 2, or 3 days. Then, the culture media supernatant was collected for the detection of lactate generation using a Lactate Assay Kit purchased from Sigma–Aldrich, Inc. (MAK064, St. Louis, MO, USA) and an EnVision Multilabel Reader. Lactate generated was calculated as follows: Lactate generation (ng/μL) = (amount of lactic acid in unknown sample from standard curve/sample volume added into the wells) ^∗ lactate molecular weight.

Culture media samples treated with PTX were collected at the indicated time points as described above to directly measure their pH with a meter from Sartorius (PB-10, Göttingen, Germany).

HCT116 cells were treated with 0.1 and 0.3 nM PTX for 0, 1, 2, or 3 days. Total RNA was isolated using TRIpure reagent (BioTek, China), according to the manufacturer’s instructions. The cDNA was synthesized from total RNA using a PrimeScript RT reagent Kit with g Eraser (TaKaRa, Japan). The obtained cDNA was used as a template in SYBR green-based quantitative-polymerase chain reaction (q-PCR) (CFX-96, Bio-Rad, USA). The mRNA expression levels of glycolysis-related genes and glutaminolysis-related genes were assessed by q-PCR. β-actin was used for normalization. Primers are listed in Table 1.

After HCT116 cells were treated with PTX at the indicated times and concentrations, cells were washed twice with PBS and collected. Then, total protein concentrations of cell lysates were determined with a BCA Protein Assay kit (Beyotime, Shanghai, China). Protein samples (total protein loading of 100 μg) were separated by 10% SDS–PAGE and transferred onto PVDF membranes. These membranes were incubated for 30 min in 5% BSA buffer (Solarbio, Beijing, China) with gentle shaking to block non-specific binding before incubation with the diluted primary antibody (p53: 1:200, p21: 1:200, β-actin: 1:5000) overnight at 4°C. Subsequently, membranes were incubated with 5000-fold diluted secondary antibody (Santa Cruz, CA, USA) for 90 min at room temperature. The membrane was washed three times in PBS, for 10 min each time. Then, the membrane was treated for 3 min in the dark with reagent from an Easysee Western Blot Kit (Transgene, Alsace, France).

Statistical comparisons were performed using Student’s t-test. Quantitative data were expressed as the means ± SD. p < 0.05 was considered significant. ^∗p < 0.05 and ^∗∗p < 0.01 versus controls.

Our previous studies demonstrated that 1 and 3 nM PTX had an impact on morphology and viability of colorectal carcinoma HCT116 cells (Li et al., unpublished). Here, we further investigated whether lower doses of PTX had similar effects. Here we found that the low-dose PTX did not alter morphology of HCT116 cells (Figure 1A), however, it significantly decreased colony-forming ability of these cells (Figures 1B,C). Furthermore, the MTT assay further indicated that low-dosage PTX significantly inhibited viability of HCT116 cells in a dose-independent manner (Figure 1D). These findings indicate that low-dose PTX exerts a significant influence on the proliferation of HCT116 cells.

Our previous results indicated that 1 and 3 nM PTX blocked the cell cycle at the G₀/G₁ phase, which inhibited colorectal carcinoma cell proliferation (Li et al., unpublished). To investigate effects of low-dose PTX on the cell cycle in HCT116 cells, we also performed an analysis of the cell cycle using flow cytometry (Figure 2A). We found that the sub-G₁, G₁, S and G₂ phases of HCT-116 cells were not significantly changed by treatment at indicated concentrations and times (Figure 2B).

Interestingly, these PTX treatments increased levels of p21 protein by 5.87- and 7.34-fold, in a dose-dependent manner (Figures 2C,D). As an inhibitor of cyclin D/cdk complexes, p21 can influence the cell cycle (Orlando et al., 2015), but our results indicated that the cycle of treated cells did not change. Based on the above findings, we surmised that low-dose PTX acts through another mechanism, other than blocking the cell cycle, to inhibit colorectal carcinoma cell growth.

When HCT116 cells were treated as described Figure 3A, we found that the culture media gradually turned yellow through time. This media color indicated higher acidity. We thus collected the culture media as described above and determined its pH value. Culture media pH decreased through time (Figure 3B).

Tumor cell proliferation is susceptible to alterations in microenvironment pH (Bolzoni et al., 2016). To explore causes of reduced pH, we measured its lactate concentration, because lactate is a major source of acidity in cell metabolism (Seheult et al., 2017). We found that low-dose PTX increased lactate production with dose and time (Figure 3C), consistent with the lowered pH of the medium. In glucose metabolism, lactate production is primarily affected by the lactate dehydrogenase enzyme (LDH). However, the mRNA expression level of LDHA was not different in cells treated with or without PTX (Figure 3D), which suggests that PTX induces lactate production through a mechanism other than increased expression of LDHA.

Lactate is final product of glycolysis, and this pathway is regulated by HK1, PHGDH, GLUT1 and PDK1 among others (Soga, 2013; Wu et al., 2014). Thus, to determine whether lactate production was related to glycolysis, we measured expression levels of glycolysis-related genes in treated cells using qPCR and Western blot. We found that the low-dose PTX significantly decreased mRNA expression of glycolysis-related genes HK1 and PHGDH in time-dependent manners (Figures 4A,D), however, the protein expression of these genes did not change (Figures 4B,C,E,F). Moreover, the mRNA expression of glycolysis-related genes PDK1 and GLUT1 did not change (Supplementary Figure S1). Therefore, pathways other than glycolysis appear to be involved in the PTX-induced lactate production.

To maintain a robust increase in nutrient uptake, in addition to increasing glycolysis, tumor cells commonly activate glutaminolysis. GLS, GDH, SLC7A11 and SLC1A5 genes play an important role in glutamate transport and metabolism (Locasale, 2013). To further explore whether glutaminolysis was related to lactate production, we also treated HCT116 cells as described above and found that low-dose PTX significantly decreased the mRNA of glutaminolysis-related genes GLS and SLC1A5 in a time-dependent manner, SLC7A11 for 3 days at 0.3 nM PTX (Figures 5A,D,G). And the protein expression level of GLS, SLC1A5 and SLC7A11 genes significantly decreased in dose-dependent manner (Figures 5B,C,E,F,H,I). In addition, the mRNA expression of GDH significantly decreased in a time-dependent manner, whereas the protein expression had no change (Figures 5J–L). Our results showed that the low-dose PTX significantly decreased the expression level of glutaminolysis-related genes, promoting lactate production.

Published reports show that PTX inhibits tumor cell growth through up-regulation of tumor-suppressive p53 (Choi and Yoo, 2012). In addition, based on the above finding of an increase in the protein expression level of p21, which is a downstream gene of p53, we further verified that low-dose PTX had an impact on cell growth through regulation of p53 in PTX-treated HCT116 cells. We detected the protein expression level of p53 in treated cells by Western blot and found that low-dose PTX significantly increased expression level of p53, by 2.34- to 2.88-fold, in a dose-dependent manner (Figures 6A,B).