FOLATE-TARGETED POLYFORMULATIONS OF CYTOTOXIC NUCLEOSIDE TRIPHOSPHATES AND PACLITAXEL

Introduction

Increasing efficacy of anticancer treatment and reversal of drug resistance are mainstream approaches besides the new drug discovery. Many cytotoxic nucleoside analogs (NA) belong to the first-line drugs in cancer chemotherapy. These drugs are activated in cancer cells into 5′-mono-, di- and triphosphates (NATP) that affect different cellular chains and induce apoptosis and cell death. Cellular delivery and activation of NA determine the drug efficacy and are the main mechanisms involved in the development of drug resistance. Recently, we have demonstrated that NATP could be efficiently formulated and delivered in cationic nanogels^1-3. Direct delivery of NATP results in the fast accumulation of the activated drug in cytoplasm of cancer cells, bypassing near all mechanisms of drug resistance. NATP-nanogel formulations demonstrated an equivalent or, in certain examples, even higher cytotoxicity than parental NA. In this paper we report the synthesis of cationic micellar and cross-linked Poloxamer networks for encapsulation of cytotoxic NATP decorated with multiple folate moieties for targeting cancer cells. NATP have been synthesized by optimized one-pot method from unprotected NA, mercaptoguanosine and 5-fluoro-2’-deoxyuridine, and tris(imidazolyl)phosphate reagent on millimolar scale⁴. Cationic polymer dispersions mixed with NATP produced stable nanoparticles with high drug content and diameters below 200 nm. Furthermore, another cytotoxic drug, paclitaxel, could be encapsulated in hydrophobic cores of nanocarriers. Application of both NATP and paclitaxel in one nanoformulation resulted in synergistic effect on cytotoxicity.

Experimental

Materials

All reagents were the highest grade available. All used solvents were of anhydrous grade and stored over activated molecular sieves 4A. Folic acid, spermine hydrochloride, water-soluble carbodiimide (EDC), 1,1’-carbonyldiimidazole (CDI), paclitaxel, (−)3-aminopropanediol-1,2 (APD), ethylenediamine (EDA) and cyanoborohydride were purchased from Aldrich Chemical Co. Poloxamer/Pluronic^R were obtained as gift from BASF Co. Mercaptoguanosine (MG) and 2’-deoxy-5-fluorouridine (FdU) were purchased from SynQuest Laboratories Co. DTBP reagent was purchased from Pierce Co. NAP-10 and 25 columns were from GE Healthcare Co. Dialysis membrane (MWCO 2000 Da) tubes were from Fisher Scientific Co.

Instrumentation

Buchi rotor evaporator with dried air inlet was used in all syntheses to remove traces of water from reactants. Folate content in polymer conjugates was calculated by UV absorbance at 363 nm (ε 6500). Amine content was measured by the 2,4,6-trinitrobenzenesulphonic acid (TNBS) method using ethanolamine as a standard. Elemental analysis was performed by M-W-C Laboratories in Phoenix, AZ. ¹H-, ¹³C- and ³¹P-spectra were measured on Bruckner NMR spectrometer in d₃-chloroform using 5 mm NMR tubes.

Synthesis of EDA-Poloxamer-EDA and APD-Poloxamer-APD conjugates

Dried Poloxamers F127, P104 and P103 (1 g each) were treated with the 6-fold molar excess of CDI in 10 ml of anhydrous acetonitrile for 4 h at 25°C. Separate ethanolic solutions of EDA and APD at 20-fold molar excess were then added to F127 and P104/P103, respectively, and reaction was continued at stirring for additional 4 h at 25°C. Bis-urethane Poloxamer conjugates were desalted by dialysis against water (3 changes × 17, 3 and 3 h), frozen and lyophilized. Mass yields were near quantitative. The products were analyzed by TLC on silicagel in dichloromethane-methanol (7:3), using iodine vapor and ninhydrine staining.

Synthesis of spermine-Poloxamer-spermine conjugates

Both APD-Poloxamer-APD conjugates (100 μmol) were dissolved in 10 ml of water and equal volume of 50 mM aqueous solution of sodium periodate was added at 4°C. The periodate oxidation was continued for 2 h in the dark , then drop of glycerol was added to stop reaction and bis-aldehyde products were desalted by dialysis against water (2 changes × 3 h). Desalted polymeric reagents were poured into stirring solutions of spermine hydrochloride (8-fold molar excess over the Poloxamers). Dry cyanoborohydride (60 mg) was immediately added to each mixture and reaction was continued overnight at 25°C. Bis-urethane Poloxamer conjugates were desalted by dialysis against water (3 changes × 3, 3 and 18 h), frozen and lyophilized. The products were analyzed by TLC on silicagel in dichloromethane-methanol (7:3), using ninhydrine staining, and by the colorimetric TNBS test.

Synthesis of folate-Poloxamer-folate conjugates

Folic acid (120 μmol) was dissolved in water by adjusting pH to 6.5 using 0.5N sodium hydroxide. Water-soluble carbodiimide EDC (120 μmol) was added to the folic acid at stirring, then, slowly, the aqueous solution of EDA-Poloxamer-EDA (30 μmol) with pH adjusted to 6.5 using 0.5N hydrochloric acid. The reaction was continued for additional 2 h at 25°C and the mixture was desalted by dialysis against water (3 changes × 3, 3 and 18 h), frozen and lyophilized. The product was analyzed by TLC on silicagel in dichloromethane-methanol (7:3), using iodine vapor staining, and by the UV spectrophotometry.

One-pot synthesis of NATP using tris(imidazolyl)phosphate reagent

Dried NA (1 mmol) was suspended in 2 ml of anhydrous DMF and treated with solid tris(imidazolyl)phosphate (2 mmol) in ice bath for 60 min. The mixture was treated with 0.1 ml of anhydrous methanol and added dropwise to the stirring solution of tributylammonium (TBA) salt of pyrophosphate (4 mmol) in 6 ml of anhydrous DMF. Reaction was continued for 30 min and then treated with 10 ml of 0.2M TBA acetate buffer, pH 7, for 30 min at room temperature. The solution was diluted with water to the final volume of 40 ml and loaded into the short column with Silicagel C18. NATP in the form of TBA salt was eluted with methanol gradient (0 to 70%), concentrated in vacuo and precipitated in 1% sodium perchlorate in acetone to isolate NATP sodium salt with yields of 65−80%. Products were usually ca.90% pure by spectrophotometry and ion-pair reverse phase HPLC analysis.

Results and Discussion

Design and synthesis of bis-folate and bis-spermine Poloxamer conjugates

The choice of these polymeric conjugates was based on four hypotheses: (1) bis-spermine Poloxamers will efficiently bind two molecules of NATP, (2) two Poloxamers with the identical and large size (M.W. >3000) of hydrophobic polyoxypropylene (PPO) part will be able to form mixed micelles, (3) bis-folate Poloxamer of higher M.w. will create PEG envelope with tumor targeting moieties around the NATP-bound internal core, and (4) large hydrophobic core will readily accommodate the second hydrophobic anticancer drug such as paclitaxel. Poloxamers P103 and P104 have the PPO block of M.w. 3300, but different hydrophilic-lipophilic balance, and will carry cationic moieties for binding NATP. Poloxamer F127 has a large PPO block of M.w. 4000, long PEG arms, and will provide cancer cell targeting, protecting of NATP complexes and encapsulation of hydrophobic drugs.

Synthesis of Pluronic-spermine conjugates is depicted in Figure 1. Modification of Poloxamer F127 by folate involves two steps: (1) modification of both ends with CDI and EDA⁵, and (2) reaction of the modified Poloxamer with EDC-activated folic acid. Folate content was ca.1.7 mol per conjugate. The measured amine content values were close to calculated ones at more than 90% substitution rate.

Figure 1.

Scheme of the synthesis of bis-spermine Pluronic conjugates.

Relatively simple synthesis of NATP was performed starting from NA by using solid tris(imidazolyl)phosphate phosphorylation reagent prepared by reaction of N-trimethylsilylimidazole with POCl₃. As shown in Figure 2, the first step was the formation of 5’-bis(imidazolide)phosphate of NA. The second step, without isolation of this product, was the reaction with pyrophosphate and formation of cyclic trimetaphosphate⁶. This intermediate rapidly hydrolyzed of into 5’-triphosphate with high yield.

Figure 2.

Preparation of tris(imidazolyl)phosphate and the one-pot synthesis of NATP.

Cytotoxicity of MGTP polyformulations in leukemic cells

Cationic layer in nanoformulations could bind polyanion molecules and form relatively stable in physiological conditions complexes. Total drug loading reached 15% of polyformulation weight (Figure 3). Folate-modified polyformulations were much less cytotoxic than identical formulations without folate decoration. MOLT-4 and KG-1 leukemic cells have different expression levels of the folate receptor alpha (FRalpha). Both type of cells demonstrated elevated sensitivity to MGTP polyformulations compared to MGTP alone, but in FRalpha-rich KG-1 cells this difference was 3.5-fold, while in FRalpha-low MOLT-4 cells the difference was only 1.5-fold (Table 1).

Figure 3.

Formation of nanoformulations of bis-spermine and bis-folate Poloxamer conjugates with NATP and paclitaxel.

Table 1.

Cytotoxicity of MGTP, Cationic and Folate Nanoformulations in KG-1 and MOLT-4 Leukemic Cells (IC₅₀ Values Following 72 h-Treatment).

Formulation	KG-1	MOLT-4
MGTP	0.07	0.03
Fol-F127	1.3	0.8
Sp-P104	0.7	0.3
Sp-P104/MGTP	0.13	0.15
Fol-F127/Sp-P104/MGTP	0.1	0.21
Fol-F127/Sp-P104(cl)/MGTP	0.1	0.11
Sp-P103	0.3	0.17
Sp-P103/MGTP	0.01	0.04
Fol-F127/Sp-P103/MGTP	0.02	0.02

Cytotoxicity of FdUTP polyformulations in MCF-7 carcinoma cells

We found an increased cytotoxic effect of polyformulations also in human breast carcinoma cells. These cells showed an increased sensitivity to FdUTP and paclitaxel nanoformulations. Maximum loading of paclitaxel was ca. 10% that did not affect stability/aggregation of polyformulations. Additional drug inclusion resulted in the enhanced cytotoxicity, although therapeutic input of paclitaxel was evidently higher. Poloxamer P103 formulations were more cytotoxic than P104 formulations, the difference between the inherent cytotoxicities of the carrier and formulated drug was 2.4-fold for P104 and 9-fold for P103 polyformulations (Table 2). The reduced therapeutic concentrations of micellar formulations mean that drug-loaded micelles could dissociate early following injection and dilution in the bloodstream. Therefore, we assayed effect of stabilizing these formulations by DTBP cross-linking (cl)⁷. This reagent does not reduce the total amount of charged amines following modification. We have found that cross-linking has low effect on cytotoxicity of studied polyformulation, so cross-linked can be used for in vivo experiments in animal tumor models.

Table 2.

Cytotoxicity of FdUTP and Paclitaxel (Pxl) Nanoformulations in Breast Carcinoma MCF-7 Cells (IC₅₀ Values Following 72 H-Treatment).

Formulation	MCF-7
FdUTP	0.3
Pxl	0.05
Sp-P104	0.12
Sp-P104/Pxl	0.07
Sp-P104/Pxl/FdUTP	0.05
Sp-P104(cl)/Pxl/FdUTP	0.08
Sp-P103	0.09
Sp-P103/Pxl	0.05
Sp-P103/FdUTP/Pxl	0.01

Conclusions

The present study demonstrated successful application of novel folate-decorated anticancer polyformulations of cytotoxic NATP. These polymeric carriers were nontoxic, while their NATP formulations were much more cytotoxic to various cancer cells than parental NA. Inclusion of the second cytotoxic agent (paclitaxel) with different mode of action was capable to additionally elevate the potency of these formulations. These cytotoxic drug polyformulations are novel promising chemotherapeutic compositions of high affinity and potential applications against drug-resistant tumors.