Hyaluronic acid-nimesulide conjugates as anticancer drugs against CD44-overexpressing HT-29 colorectal cancer in vitro and in vivo

Abstract

Carrier-mediated drug delivery systems are promising therapeutics for targeted delivery and improved efficacy and safety of potent cytotoxic drugs. Nimesulide is a multifactorial cyclooxygenase 2 nonsteroidal anti-inflammatory drug with analgesic, antipyretic and potent anticancer properties; however, the low solubility of nimesulide limits its applications. Drugs conjugated with hyaluronic acid (HA) are innovative carrier-mediated drug delivery systems characterized by CD44-mediated endocytosis of HA and intracellular drug release. In this study, hydrophobic nimesulide was conjugated to HA of two different molecular weights (360 kDa as HA with high molecular weight [HAH] and 43kDa as HA with low molecular weight [HAL]) to improve its tumor-targeting ability and hydrophilicity. Our results showed that hydrogenated nimesulide (N-[4-amino-2-phenoxyphenyl]methanesulfonamide) was successfully conjugated with both HA types by carbodiimide coupling and the degree of substitution of nimesulide was 1%, which was characterized by ¹H nuclear magnetic resonance 400 MHz and total correlation spectroscopy. Both Alexa Fluor^® 647 labeled HAH and HAL could selectively accumulate in CD44-overexpressing HT-29 colorectal tumor area in vivo, as observed by in vivo imaging system. In the in vitro cytotoxic test, HA-nimesulide conjugate displayed >46% cell killing ability at a nimesulide concentration of 400 µM in HT-29 cells, whereas exiguous cytotoxic effects were observed on HCT-15 cells, indicating that HA-nimesulide causes cell death in CD44-overexpressing HT-29 cells. Regarding in vivo antitumor study, both HAL-nimesulide and HAH-nimesulide caused rapid tumor shrinkage within 3 days and successfully inhibited tumor growth, which reached 82.3% and 76.4% at day 24 through apoptotic mechanism in HT-29 xenografted mice, without noticeable morphologic differences in the liver or kidney, respectively. These results indicated that HA-nimesulide with improved selectivity through HA/CD44 receptor interactions has the potential to enhance the therapeutic efficacy and safety of nimesulide for cancer treatment.

Keywords: CD44; COX-2 inhibitor; colorectal cancer; hyaluronic acid; nimesulide.

Figure 1

Schematic illustration of the synthesis of (A) NiNH₂ and (B) HA–nimesulide conjugates. Abbreviations: EA, ethyl acetate; EDC, 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide; HA, hyaluronic acid; h, hours; NHS, N-hydroxysuccinimide; NiNH₂, N-(4-amino-2-phenoxyphenyl)methanesulfonamide.

Figure 2

¹H NMR spectra of NiNO₂ (nimesulide) and NiNH₂. Abbreviations: NiNH₂, N-(4-amino-2-phenoxyphenyl)methanesulfonamide; NMR, nuclear magnetic resonance.

Figure 3

¹H NMR spectra of (A) NiNH₂, (B) HAL–nimesulide-1%, (C) HAH–nimesulide-1% and (D) free HAH. Abbreviations: NiNH₂, N-(4-amino-2-phenoxyphenyl)methanesulfonamide; NMR, nuclear magnetic resonance; HA, hyaluronic acid; HAH, HA with high molecular weight; HAL, HA with low molecular weight.

Figure 4

TOCSY (bold line) results for HA–nimesulide conjugate. Abbreviations: HA, hyaluronic acid; TOCSY, total correlation spectroscopy.

Figure 5

(A) Flow cytometry analysis of CD44 expression in HT-29 and HCT-15 cells; (B) immunocytochemistry of CD44 expression in HT-29 or HCT-15 cells assessed by confocal microscopy at a concentration of 1 µM CD44-FITC or 1 mg/mL HA-dye. Intracellular distribution of HAH-ADH-FITC at 3, 6 and 24 h posttreatment in (C) HT-29 and (D) HCT-15 cells. Hoechst 33342 (shown in blue) is used for nuclei labeling. CD44-FITC or HAH-ADH-FITC is shown in green. Scale bar: 10 µm. Abbreviations: ADH, adipic acid dihydrazide; FITC, fluorescein isothiocyanate; HA, hyaluronic acid; HAH, HA with high molecular weight.

Figure 6

In vivo fluorescence imaging of HA-NIR dye with different molecular weight of HA. Notes: (A) Time-dependent distribution of HA-NIR was measured by the IVIS Xenogen imaging system after HA-NIR dye administration intravenously (HA-NIR dye: 20 mg/mL, 200 mg/kg). (B) Quantification of fluorescence signal on the tumor site (n=3). (C) Fluorescence spectra of HAH-NIR, HAL-NIR and dye. (D) The fluorescence images of each organ and (E) fluorescence quantification of the tumor images of mice treated with HAL-NIR, HAH-NIR or free dye at 48 h postinjection (n=3). Abbreviations: HA, hyaluronic acid; HAH, HA with high molecular weight; HAL, HA with low molecular weight; IVIS, in vivo imaging system; min, minimum, max, maximum; ROI, region of interest.

Figure 6

In vivo fluorescence imaging of HA-NIR dye with different molecular weight of HA. Notes: (A) Time-dependent distribution of HA-NIR was measured by the IVIS Xenogen imaging system after HA-NIR dye administration intravenously (HA-NIR dye: 20 mg/mL, 200 mg/kg). (B) Quantification of fluorescence signal on the tumor site (n=3). (C) Fluorescence spectra of HAH-NIR, HAL-NIR and dye. (D) The fluorescence images of each organ and (E) fluorescence quantification of the tumor images of mice treated with HAL-NIR, HAH-NIR or free dye at 48 h postinjection (n=3). Abbreviations: HA, hyaluronic acid; HAH, HA with high molecular weight; HAL, HA with low molecular weight; IVIS, in vivo imaging system; min, minimum, max, maximum; ROI, region of interest.

Figure 7

(A) Cell viability assessment of NiNH₂ and NiNO₂ (nimesulide) in HT-29 and HCT-15 cells. (B) Viability of HT-29 cells in HAL–nimesulide-1% or HAH–nimesulide-1% for 48 h and (C) cell viability of HT-29 and HCT-15 cells treated with HAH–nimesulide-1% for 48 h. Abbreviations: NiNH₂, N-(4-amino-2-phenoxyphenyl)methanesulfonamide; HA, hyaluronic acid; HAH, HA with high molecular weight; HAL, HA with low molecular weight.

Figure 8

Apoptotic analysis of HT-29 cells using Annexin V-FITC and PI double staining. Notes: (A) Negative control group; (B) cells treated with H₂O₂ at 10 mM as positive control group and (C) cells treated with HAL–nimesulide at 200 µM nimesulide concentration for 48 h. The X-axis represents the density of Annexin V-FITC, whereas the Y-axis represents the density of PI. Abbreviations: FITC, fluorescein isothiocyanate; PI, propidium iodide; HA, hyaluronic acid; HAL, HA with low molecular weight.

Figure 9

In vivo antitumor effects of HA–nimesulide in nude mice bearing HT-29 xenografts. Notes: Nude mice were implanted subcutaneously with HT-29 cells and treated three times weekly with 0.2 mL PBS, nimesulide (1.5 mg/kg), HA–nimesulide (1.5 mg/kg equivalent nimesulide concentration) or 5-FU (50 mg/kg) once per week. (A) Tumor size and (B) body weight of mice after treatment with drugs. (C) Organ morphology, tumor weight and (D) TGI % after treatment with drugs. Statistical significances were determined using the One-way ANOVA followed by Bonferroni tests, *P<0.05. Abbreviations: 5-FU, 5-fluorouracil; ANOVA, analysis of variance; HA, hyaluronic acid; HAH, HA with high molecular weight; HAL, HA with low molecular weight; PBS, phosphate-buffered saline; TGI, tumor growth inhibition.

Figure 10

Histologic assessment of liver, kidney and tumor after treatment with PBS, 5-FU, nimesulide or HA–nimesulide in nude mice bearing HT-29 xenografts (400×). Abbreviations: 5-FU, 5-fluorouracil; HA, hyaluronic acid; PBS, phosphate-buffered saline.

Figure 11

Detection of DNA strand breaks by TUNEL assay in nude mice bearing HT-29 xenografts treated with PBS, nimesulide or HAH–nimesulide. Abbreviations: HA, hyaluronic acid; HAH, HA with high molecular weight; PBS, phosphate-buffered saline; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end-labeling.