Enhanced antitumor activity of combinations of free and HPMA copolymer-bound drugs

Abstract

The synergism in anticancer effect toward human renal carcinoma A498 cells by binary combinations of free and N-(2-hydroxypropyl)methacrylamide (HPMA) copolymer-bound anticancer drugs, SOS thiophene (SOS), doxorubicin (DOX), and mesochlorin e₆ monoethylenediamine (Mce₆), was evaluated. The combination index (CI) analysis was used to quantify the synergism, antagonism, and additive effects. Both free drugs and HPMA copolymer conjugates, when used as single agents or in combination, exhibited cytotoxic activities against A498 cells, as determined using a modified MTT assay. As single agents, SOS and P-GFLG-SOS (HPMA copolymer conjugates containing SOS bound via glycylphenylalanylleucylglycine [GFLG] spacer) were significantly more effective than the other agents evaluated. The synergistic effects ranked in the order SOS+DOX>P-GFLG-DOX+P-GFLG-Mce₆≈DOX+Mce₆>P-GFLG-SOS+P-GFLG-DOX≈SOS+Mce₆>P-GFLG-SOS+P-GFLG-Mce₆. The combination of SOS+DOX proved to be synergistic over all cell growth inhibition levels. All other combinations exhibited synergism in a wide range of drug effect levels. The SOS+Mce₆ and P-GFLG-SOS+P-GFLG-Mce₆ combinations displayed synergism up to drug affected fraction (f_a) values of about 0.8 and reached slight antagonism and nearly additivity at f_a = 0.95, respectively. However, all other combinations were synergistic up to f_a < 0.9 and were additive at higher f_a values. The observations that most combinations produced synergistic effects will be important for clinical translation.

1. Introduction

Renal cell carcinoma accounts for about 3% of adult malignancies and is by far the most frequent neoplasm arising from the kidney. The prognosis for advanced stage renal carcinoma is poor. The low efficiency of chemotherapeutic agents is due to high levels of p-glycoprotein expression in normal renal proximal tubules and renal carcinoma cells (Mickisch et al., 1990; Kim et al., 1996). Moreover, about one third of patients develop metastatic disease after nephrectomy (Sachdeva, 2006). Evidently, novel approaches in the treatment of renal carcinoma are needed.

It is well known that macromolecular drug delivery systems have the potential to overcome multidrug resistance. The exclusion of the polymer-drug conjugates from the cytoplasm of the cell, due to the fact that intracellular trafficking occurs within membrane-limited organelles, renders efflux pumps ineffective (Kopeèek et al., 2000). Experimental data on sensitive (A2780) and resistant (A2780/AD) human ovarian carcinoma cells showed that, in contrast to free doxorubicin (DOX), N-(2-hydroxypropyl)methacrylamide (HPMA) copolymer-bound DOX (P-GFLG-DOX; where P is the HPMA copolymer backbone) overcame pre-existing MDR1-gene-encoded multidrug resistance, and did not induce it de novo after acute or chronic exposure in vitro (Minko et al., 1998, 1999). Similar results were obtained in solid tumor mice models of DOX sensitive and resistant human ovarian carcinoma. Free DOX was effective only in sensitive tumors, while P-GFLG-DOX was effective in both sensitive and drug-resistant tumors (Minko et al., 2000).

To improve the therapeutic outcome and reduce the toxicity of anticancer agents, combination treatments have been developed and studied (Frei et al., 1965). For example, combinations of chemotherapies using agents with different mechanisms of action (Harris et al., 2005; Shanks et al., 2005); chemotherapy in combination with photodynamic therapy (PDT) (Nahabedian et al., 1988; Jin et al., 1992; Peterson et al., 1995); chemotherapy and immunotherapy (Welt et al., 2003); PDT and radiotherapy (Colasanti et al., 2004); and radioimmunotherapy with radiotherapy (Buchegger et al., 2000) have been studied.

Recently, a novel concept of using combination therapy with water-soluble polymer bound drugs was developed (Kopeèek and Krinick, 1993). In vivo combination chemotherapy and PDT studies on two cancer models, Neuro 2A neuroblastoma induced in A/J mice (Krinick et al., 1994) and human ovarian carcinoma heterotransplanted in nude mice (Peterson et al., 1996; Shiah et al., 2000, 2001b), demonstrated that combination therapy with HPMA copolymer-bound DOX and HPMA copolymer-bound Mce₆ (mesochlorin e₆ monoethylene diamine) produced tumor cures which could not be obtained with either chemotherapy or PDT alone. Additionally, significantly lower nonspecific toxicities were observed when compared to low-molecular weight drugs.

The most commonly used method to evaluate drug combinations in vitro is median-effect analysis, based on the Loewe additivity model, as originally proposed by Chou and Talalay (Chou and Talalay, 1984; Zhao et al., 2004; Chou, 2006). The median-effect method assesses the drug-drug interaction by a term called the “combination index” (CI), which is based on the concentration-response relationship. Numerous anticancer agent combinations had been analyzed using the median-effect method. Examples include: combination therapy for chronic myelogenous leukemia with imatinib and γ-irradiation or alkylating agents (busulfan and treosulfan) (Topaly et al., 2002); the combination treatment of lung adenocarcinoma cell line using perifosine and 7-hydroxystaurosporine (Dasmahapatra et al., 2004); and combination treatment using irofulven and 5-fluorouracil or cisplatin against human colon and ovarian carcinoma cells (Poindessous et al., 2003).

The novel anticancer agent 2,5-bis(6-hydroxymethyl-2-thienyl) furan (SOS, NSC 652287) is a dithiophene compound possessing a potent and selective activity against human renal cancer cell lines. The mechanism of action of SOS has been shown to block the p53–HDM-2 (human double minute-2) interaction in vitro and in vivo in various tumor cells lines expressing wild-type p53, such as the A498 cell line. In SOS treated cells, p53 levels increase, inducing cell cycle arrest and apoptosis (Nieves-Neira et al., 1999; Rivera et al., 1999; Fischer and Lane, 2004; Issaeva et al., 2004).

The anthracycline antitumor antibiotic DOX is one of the most effective chemotherapeutic agents. DOX inhibits the activity of topoisomerase II, produces non-protein-associated DNA strand breaks, and generates free radicals, creating DNA damage within the cells, cellular membrane damage, and ultimately cell death (Fornari et al., 1994).

Mce₆ is the second-generation synthetic photosensitizer. When activated by light, it interacts with molecular oxygen to produce highly reactive singlet oxygen, causing irreversible photodamage to cells resulting in cell death (Hopper, 2000).

In this study, SOS, DOX, and Mce₆ were chosen as anticancer agents; these represent low-molecular weight compounds possessing different sites and/or mechanisms of action. The interactions between free and HPMA copolymer-bound SOS, DOX, and Mce₆ in binary combination against the A498 renal carcinoma cell line in vitro were evaluated using median-effect analysis. We hypothesized that a combination of these agents may produce synergistic effects and thereby reduce effective doses, compared to the doses required for each agent alone to produce a given drug effect level.

2. Materials and methods

2.1. Abbreviations

CI, combination index; D_m, median-effect dose; DI, deionized; DIPEA, N,N-diisopropylethylamine; DMAP, 4-dimethylaminopyridine; DMF, N,N-dimethylformamide; DMSO, dimethylsulfoxide; DOX, doxorubicin hydrochloride; DRI, dose-reduction index; EPR, enhanced permeability and retention; GFLG, glycylphenylalanylleucylglycine; HPMA, N-(2-hydroxypropyl)methacrylamide; IC₅₀, concentration that inhibited cell growth by 50% as compared with control cell growth; MA, methacryloyl; M_n, number average molecular weight; M_w, weight average molecular weight; Mce₆, mesochlorin e₆ monoethylenediamine disodium salt; MRP, multidrug resistance-related protein; MTT assay, modified 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide assay; ONp, p-nitrophenoxy; P, HPMA copolymer backbone; PDT, photodynamic therapy; P-GFLG-DOX, HPMA copolymer-DOX conjugate; P-GFLG-Mce₆, HPMA copolymer-Mce₆ conjugate; P-GFLG-ONp, HPMA copolymer precursor containing reactive p-nitrophenyl ester groups at side chain termini; P-GFLG-SOS, HPMA copolymer-SOS conjugate; SOS, 2,5-bis(5-hydroxymethyl-2-thienyl)furan.

2.2. Materials

SOS was kindly supplied by the National Cancer Institute. DOX was a gift from Dr. A. Suarato, Pfizer, Milano, Italy. Mce₆ was purchased from Porphyrin Products (Logan, UT). All other chemicals were purchased from Sigma Chemical Co. (St. Louis, MO).

2.3. Cell lines

The human renal carcinoma cell line A498 was purchased from American Type Culture Collection (ATCC). Cells were grown as monolayer cultures in EMEM medium (ATCC, Manassas, VA) supplemented with 10% fetal bovine serum (HyClone Laboratories, Logan, UT), at 37°C in a humidified atmosphere of 5% CO₂ (v/v).

2.4. Synthesis of HPMA copolymer-drug conjugates

2.4.1. HPMA copolymer-SOS conjugate (P-GFLG-SOS)

P-GFLG-SOS was synthesized using a polymer analogous reaction in two steps. First, the polymer precursor (P-GFLG-ONp) was prepared by radical precipitation copolymerization of HPMA and N-methacryloyl glycylphenylalanylleucylglycine p-nitrophenyl ester (MA-GFLG-ONp; Kopeèek et al., 1991; Ulbrich et al., 2000). The polymer precursor contained 4.15 mol% of active ester groups; molecular weight (M_w) = 23 kDa; polydispersity (M_w/M_n) = 1.3. Second, SOS was bound to P-GFLG-ONp by an ester linkage. The reactions are shown in Scheme 1. P-GFLG-ONp (450mg, 0.116 mmol ONp groups) was dissolved in N,N-dimethylformamide (DMF; ~4 ml). SOS (71 mg, 0.243 mmol) was dissolved in DMF (~1 ml). Polymer precursor solution was then mixed with the SOS solution. 4-Dimethylaminopyridine (DMAP; 40 mg, 0.327 mmol) was added into the reaction mixture. The reaction solution was bubbled with N₂ and allowed to proceed for 72 h in the dark at room temperature. The DMF was then partially evaporated to ~2 ml under reduced pressure. The product was precipitated in a mixture of acetone: ether (3:1 (v/v), ~550 ml). The precipitate was collected by filtration, washed with acetone (~50 ml) and ether (~50 ml), and dried under vacuum. To purify the product was dissolved in methanol (~3 ml) and applied to a Sephadex LH-20 column with methanol as the mobile phase. The polymer band was collected, concentrated under reduced pressure and re-precipitated. The product was a yellowish powder, with a yield of 340 mg (61%). The characterization of the polymer conjugates is shown in Table 1.

Scheme 1.

Synthesis and chemical structure of N-(2-hydroxypropyl)methacrylamide (HPMA) copolymer-SOS/-DOX/-Mce₆ conjugates (P-GFLG-SOS, P-GFLG-DOX, and P-GFLG-Mce₆, respectively) containing glycylphenylalanylleucylglycine spacers.

Table 1.

Characterization of HPMA copolymer-drug conjugates.

Conjugates	Mol% of druga	mmol drug per g polymer conjugate	Number of drug molecules per macromolecule	Apparent Mw (kDa)b	Polydispersityc
P-GFLG-SOS	1.62	0.106	3.4 (SOS)	32	1.50
P-GFLG-DOX	1.93	0.121	3.0 (DOX)	25	1.24
P-GFLG-Mce6	2.04	0.125	2.9 (Mce6)	23	1.34

^aDetermined by UV spectrophotometry (extinction coefficient at 358 nm (ε₃₅₈) = 33 000 M⁻¹ cm⁻¹ in methanol for SOS, ε₄₈₈ = 11 000 M⁻¹ cm⁻¹ in DI water for DOX, and ε₃₉₅ = 158 000 M⁻¹ cm⁻¹ in methanol for Mce₆).

^bApparent molecular weight (M_w) of polymers was estimated by size exclusion chromatography using a fast performance liquid chromatography (FPLC) system equipped with a Superose 6 column, calibrated with polyHPMA fractions. Acetate buffer pH 5.5 + 30% (v) acetonitrile was used for P-GFLG-SOS. PBS buffer pH 7.3 + 30% (v) acetonitrile was used for P-GFLG-DOX and P-GFLG-Mce₆.

^cPolydispersity = the ratio of weight-average to number-average molecular weight

2.4.2. HPMA copolymer-DOX conjugate (P-GFLG-DOX)

P-GFLG-DOX was prepared as described previously (Kopeèek et al., 1991; Krinick et al., 1994). Briefly, the conjugate was synthesized by conjugating DOX to P-GFLG-ONp precursor, containing 5.5 mol% of active ester groups; M_w = 22 kDa; M_w/M_n = 1.2. The procedure is shown in Scheme 1. DOX (127 mg, 0.220 mmol) and P-GFLG-ONp (1 g, 0.330 mmol ONp groups) were dissolved in dimethylsulfoxide (DMSO; ~4 ml). The reaction mixture was stirred to complete dissolution for 20 min. N,N-diisopropylethylamine (DIPEA; 48 μl, 0.275 mmol) was added while stirring. The reaction mixture was stirred overnight in the dark at room temperature. Then, 1-amino-2-propanol (~15 μl) was added under stirring. The reaction mixture was precipitated into a mixture of acetone: ether (3:1 (v/v), ~500 ml). The precipitate was collected by filtration, washed with a mixture of acetone: ether (3:1 (v/v)) and ether, and dried under vacuum. The product was dissolved in methanol (~10 ml) and purified twice on a Sephadex LH-20 column with methanol/0.5% acetic acid as the mobile phase. The polymer band was then collected and evaporated to dryness. The product was dissolved in deionized (DI) water (~20 ml), dialyzed overnight against DI water and lyophilized. The product yield was 820 mg (68%).

2.4.3. HPMA copolymer-Mce₆ conjugate (P-GFLG-Mce₆)

P-GFLG-Mce₆ was also prepared as described previously (Kopeèek et al., 1991; Krinick et al., 1994). The conjugate was synthesized by conjugating Mce₆ to P-GFLG-ONp precursor, containing 5.1 mol% of active ester groups; M_w = 22 kDa; M_w/M_n = 1.2. The procedure is shown in Scheme 1. Mce₆ (160 mg, 0.234 mmol) was dissolved in DMF (~3 ml). P-GFLG-ONp (750 mg, 0.230 mmol ONp groups) was dissolved in DMF (~2.5 ml). The suspension of Mce₆ was added dropwise into polymer precursor solution while stirring. The reaction mixture was stirred for 1 h until the mixture was completely dissolved. DIPEA (40 μl, 0.229 mmol) was added under stirring. The solution was stirred overnight in the dark at room temperature. The solution was precipitated into a mixture of acetone: ether (3:2 (v/v), ~500 ml). The precipitate was collected by filtration and dried under vacuum. The product was dissolved in methanol (~10 ml) and purified twice on a Sephadex LH-20 column with methanol/0.5% acetic acid as the elution solvent. The polymer band was collected and evaporated to dryness. The product was dissolved in DI water (~20 ml), dialyzed overnight and lyophilized. The product yield was 670 mg (69%).

2.5. Drug stock solution preparations

SOS was dissolved in PBS containing cyclodextrin (5% (w/v) cyclodextrin in PBS/1 mg of SOS) to enhance the solubility of SOS (Alley et al., 2004). Other samples (DOX, Mce₆, P-GFLG-SOS, P-GFLG-DOX, and P-GFLG-Mce₆) were prepared in DI water. All stock solutions were filtered using a 0.22 μm sterile filter and kept in sterilized Eppendorf tubes. Drug contents were determined by UV spectrophotometry. All stock solutions were freshly prepared and gradually diluted with EMEM culture medium before use.

2.6. In vitro growth inhibition bioassays

The drug concentration that inhibited cell growth by 50% compared with control cells (IC₅₀) was determined using a modified 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide (MTT) assay (Hansen et al., 1989). Cells were seeded in 96-well flat bottom microplates at a density of 5,000 cells/well in 200 μl of EMEM medium and allowed to grow for 36 h. The cells were then exposed to various concentrations of each free drug alone (SOS, DOX, and Mce₆), each copolymer conjugate (P-GFLG-SOS, P-GFLG-DOX, and P-GFLG-Mce₆), or their binary combinations (n=6). After 16 h of exposure, the drugs were removed, cells were washed with warm PBS and the medium (300 μl) replaced. For the cell growth inhibition studies using Mce₆ or P-GFLG-Mce₆ (alone or in combinations), the cells were irradiated with three tungsten halogen lamps through a 650 nm band pass filter at 3.0 mW/cm² for 30 min. After an additional 3 days in culture, medium was removed and replaced with 100 μl of fresh medium and 10 μl of sterile-filtered MTT solution (5mg/ml in PBS). After incubating for 24 h, 150 μl of 20% (w/v) sodium dodecyl sulfate in water was added to each well and incubated overnight. The following day, the absorbance of each well was read at 570 nm with a reference wavelength at 630 nm. Untreated cells served as a 100% cell viability control and the media served as background reference. Growth inhibition was expressed as the growth of drug-treated cells related to that of untreated control cells.

2.7. Dose-effect analysis and determination of combination index (CI)

In binary combination treatment studies, A498 cells were treated with a dose range of SOS, DOX, Mce₆, P-GFLG-SOS, P-GFLG-DOX, and P-GFLG-Mce₆ simultaneously (n=6). Drug interactions and CI values were analyzed using median-effect principle according to the method of Chou and Talalay (Chou and Talalay, 1984; Chou, 2006).

The median-effect equation describes dose-effect relationships, which is described by

$$\frac{f_a}{f_u}={\left(\frac{D}{D_m}\right)}^m$$ (1)

where f_a and f_u are the fraction affected [1 − (absorbance of treatment well - average of absorbance of blanks)/(average of absorbance of untreated cell wells - average of absorbance of blanks)] and unaffected (f_u = 1 − f_a) by the dose or concentration D, D_m is the median-effect dose (IC₅₀) that inhibits the cell growth by 50%, and m is the coefficient signifying the shape of the dose-effect relationship.

Based on the logarithmic conversion of Eq. (1) (Chou, 1976),

$$log\left(\frac{f_a}{f_u}\right)=m\hspace{0.17em}log\hspace{0.17em}(D)-m\hspace{0.17em}log\hspace{0.17em}(D_m)$$ (2)

where m is the slope and D_m is the anti-log of the x-intercept. The plot of x = log(D) versus y = log(f_a/f_u) is called the median-effect plot. An example is shown in Figure 1 for the SOS+DOX combination, including the dose-effect curves and the median-effect plots.

Figure 1.

(A) Dose-effect curves and (B) the median-effect plots of SOS (○), DOX (□), and SOS+DOX combination (△) of A498 cells, according to Eq. (1) and Eq. (2).

The combination index (CI) describes the interaction between two drugs and quantitates the synergism, antagonism or additive effects. The CI is determined by the equation:

$$\text{CI}=\frac{{(D)}_1}{{(D_x)}_1}+\frac{{(D)}_2}{{(D_x)}_2}$$ (3)

where (D_x)₁ and (D_x)₂ are the doses of drug 1 alone and drug 2 alone that inhibit the cell growth x%, respectively. (D)₁ and (D)₂ are for doses in combination that also inhibit x%.

The CI values were calculated by solving Eq. (3) for different values of f_a and plotting the CI values as a function of f_a values, using CompuSyn software (ComboSyn Inc, Paramus, NJ). In the Fa-CI plot, if CI < 1, = 1, and > 1 indicate synergism, additivity, and antagonism, respectively.

The dose-reduction index (DRI) (Chou and Martin, 2006) is a determination of how many times the dose of each drug in a synergistic combination may be reduced at a given effect level compared with the doses of each drug alone. The dose reduction is an indication of how toxicity is reduced toward normal cells or host while the desired therapeutic effect is maintained. The DRI value for each corresponding drug was calculated by the following simple equation:

$$\mathit{DRI}=\frac{{(D_x)}_1}{{(D)}_1}+\frac{{(D_x)}_2}{{(D)}_2}$$ (4)

2.8. Statistical Analysis

All mean values are presented as means ± standard deviation (SEM). The Student’s t-test (two tailed) was used to evaluate the statistical significance of any differences in mean values in the experimental groups. The ANOVA test was used to assess the differences in means between single agents and combination treatments. In all statistical analyses, p-values < 0.05 were considered to indicate statistical significance.

3. Results and discussion

3.1. Characteristics of HPMA copolymer -SOS/-DOX/-Mce₆ conjugates

P-GFLG-SOS, P-GFLG-DOX and P-GFLG-Mce₆ conjugates were synthesized as shown in Scheme 1. The GFLG oligopeptide sequence was chosen as the drug attachment/release site and incorporated in all three conjugates. This tetrapeptidyl linker was designed to be stable in blood plasma, but susceptible to cleavage by cathepsin B (lysosomal cysteine proteinase) within the lysosomal compartment (Kopeèek and Rejmanová, 1983; Rejmanová et al., 1983; Kopeèek, 1984).

The characteristics of the conjugates, including the drug content in mole% and mole/gram of polymer conjugate, average amount of drug per macromolecule, molecular weight, and polydispersity, are summarized in Table 1. The drug content in all conjugates was similar. P-GFLG-SOS, P-GFLG-DOX and P-GFLG-Mce₆ conjugates contained 3.4, 3.0, and 2.9 drug molecules per macromolecule, respectively. The apparent M_w of polymer conjugates were estimated by size exclusion chromatography using ÄKTA system, equipped with a Superose 6 HR 10/30 column calibrated with poly(HPMA) samples. The apparent M_w of the polymer conjugates were between 23 and 32 kDa, whereas the polydispersity ranged from 1.2 to 1.5.

3.2. In vitro inhibition of A498 cell growth by drugs as single agents

Renal cell carcinoma is the predominant form of kidney cancer and highly refractory to chemotherapy due to the multidrug resistance. This is due to gp-170, a membranous glycoprotein encoded by the MDR1 gene (Vugrin, 1987). The A498 cell line is a primary renal carcinoma cell line that expresses a moderate level of gp-170 (Yu et al., 1998). Moreover, A498 cell has been reported to express a functional wild-type p53. The transcriptional activity of p53 in renal cell carcinoma is significantly regulated by MDM-2 (mouse double minute-2) (Warburton et al., 2005).

SOS has a pronounced antitumor activity toward tumor cells lines expressing wild-type p53, especially the A498 human renal carcinoma cell line. SOS molecule interrupts the p53–HDM-2 interaction (Issaeva et al., 2004), induces DNA damage and generates DNA-protein and DNA-DNA crosslinks (Nieves-Neira et al., 1999). These effects lead to p53 stabilization and activation, resulting in G₀–G₁ and G₂-M cell cycle arrest, thereby strongly inducing apoptosis. In vivo studies of SOS in mice with xenografts derived from different drug-sensitive cell lines, including renal A498, renal CAKI-1, melanoma UACC-257, ovarian OVCAR-5, colon HCC-2998 (Rivera et al., 1999), colon HCT116 expressing wild-type p53 and its derivative p53-null HCT116 (Issaeva et al., 2004), showed that SOS possessed strong antitumor activity. It produced complete tumor regression of A498 tumor xenografts, and possessed a moderate or minimal antitumor activity to other cell lines, but had no effect on p53-null HCT116 xenografts.

DOX is one of the most widely used chemotherapeutic agents with multiple mechanisms of action, including: intercalating DNA (Neidle, 1979), inhibiting DNA-topoisomerase II interaction (Tewey et al., 1984), and inducing the generation of singlet oxygen and other oxygen radicals (Sinha et al., 1987). These mechanisms lead to DNA strand breaks, DNA damage, and damage to cellular membranes, resulting in G₁ and G₂ cell cycle arrest and cell death (Fornari et al., 1994; Blagosklonny, 2002; Larsen et al., 2003). In vitro studies of DOX cytotoxicity toward renal carcinoma cell lines have shown that they are sensitive to DOX, especially the A498 cell line (Mertins et al., 2001; Yu et al., 1998). However, the usefulness of DOX is limited by side effects, particularly cumulative dose-dependent cardiotoxicity (Shan et al., 1996). Previous studies in mice have demonstrated that the severe cardiotoxicity of DOX can be reduced by conjugating it to HPMA copolymers (Seymour et al., 1990). Similar results have been observed in clinical trials (Vasey et al., 1999).

Photodynamic therapy is a three-component therapy consisting of a photosensitizer, visible light and oxygen. Mce₆, a second generation photosensitizer, is activated by a characteristic wavelength of light to an excited singlet state, which rapidly decays to the triplet state. Both singlet and triplet state species can transfer energy to surrounding oxygen (in ground triplet state) to produce singlet oxygen. Singlet oxygen is highly reactive and will oxidize biomolecules, resulting in irreversible photodamage to the cells, cell destruction, and cell death (Hopper, 2000; Dolmans et al., 2003). Common side effects are prolonged cutaneous and systemic photosensitivities, which remain a problem in patients treated with low-molecular weight photosensitizers (Dougherty et al., 1990; Chiarello, 2004). Binding Mce₆ to HPMA copolymers containing a targeting moiety can also limit its distribution in the body and reduce side effects (Krinick et al., 1990; Goff et al., 1991; Shiah et al., 2001b).

The inhibitory effects of SOS, DOX, Mce₆, P-GFLG-SOS, P-GFLG-DOX, and P-GFLG-Mce₆ as single agents on the growth of A498 cells were evaluated after 16 h of drug exposure using the MTT assay. Values collected using untreated control cells corresponded to 100% cell viability. The IC₅₀ values for the free drugs and the HPMA copolymer conjugates are shown in Table 2. The A498 cell line was highly susceptible to SOS and P-GFLG-SOS with IC₅₀ values of 3 nM and 23 nM, respectively. The IC₅₀ of SOS in this study was in agreement with that measured by Rivera et al., who reported an IC₅₀ of 2 nM in the A498 cell line after a 48 h exposure to SOS (Rivera et al., 1999). In all experiments the linear correlation coefficient of the median-effect plot (r) was > 0.98, providing a reliable basis for further calculations.

Table 2.

Cell proliferation IC₅₀ values for SOS, DOX, Mce₆, and their HPMA copolymer conjugates against A498 cells

Drug	IC50Dm or (μM) a
SOS	0.003 ± 0.0003 b
DOX	0.24 ± 0.02 b
Mce6	3.55 ± 0.08 b
P-GFLG-SOS	0.023 ± 0.0004 b
P-GFLG-DOX	26.8 ± 3.4 b
P-GFLG-Mce6	7.32 ± 0.12 b

^aD_m is the median-effect dose that inhibits the cell growth by 50%. IC₅₀ or D_m values are the means ± SEM (n=6).

^bThe differences of means against control were statistically significant at p-value < 0.05 (Student’s t-test).

SOS was 80 and 1170 times more effective than DOX and Mce₆, respectively; P-GFLG-SOS was 1170 and 320 times more effective than P-GFLG-DOX and P-GFLG-Mce₆, respectively (Table 2). The IC₅₀ values of P-GFLG-SOS, P-GFLG-DOX and P-GFLG-Mce₆ were higher than those of free SOS, DOX, and Mce₆, respectively. These results reflect the different mechanisms of cellular internalization of free drugs vs. copolymer conjugates. Copolymer conjugates containing hydrophobic drugs/moieties are internalized by fluid-phase pinocytosis and adsorptive pinocytosis concurrently, which is slower than the diffusion process of free drugs (Duncan et al., 1981). The relatively low disparity between the IC₅₀ doses of Mce₆ and P-GFLG-Mce₆ (3.55 vs. 7.32 μM) reflects the fact that it is not necessary for Mce₆ molecules to be cleaved from copolymer backbone to generate a photodynamic effect. Polymer-bound Mce₆ can also produce singlet oxygen, albeit at a lower quantum yield than free Mce₆ (Krinick et al., 1990).

The advantages of polymer-bound drugs (when compared to low-molecular weight drugs) are (reviewed in Putnam and Kopeèek, 1995; Kopeèek et al., 2000; Duncan, 2003): a) active uptake by fluid-phase pinocytosis (non-targeted polymer-bound drug) or receptor-mediated endocytosis (targeted polymer-bound drug), b) increased active accumulation of the drug at the tumor site using targeting (Lu et al., 1999; Shiah et al., 2001b) c) increased passive accumulation of the drug at the tumor site by the enhanced permeability and retention (EPR) effect (Shiah et al., 2001a), d) long-lasting circulation in the bloodstream (Seymour et al., 1987, 1990), e) decreased non-specific toxicity of the conjugated drug (Øíhová et al., 1988), f) decreased immunogenicity of the targeting moiety (Øíhová et al., 1988), f) immunoprotecting and immunomobilizing activities (Øíhová et al., 2001), g) modulation of the cell signaling and apoptotic pathways (Minko et al., 1999, 2000, 2001; Nishiyama et al., 2003; Malugin et al., 2006), h) enhanced solubility of hydrophobic drugs, and i) the potential to overcome efflux pump-mediated mechanism of drug resistance (Minko et al., 1998, 1999, 2000).

3.3. In vitro growth inhibition of agents in combination

Experiments investigating the cytotoxicity potential of binary combinations of SOS, DOX, Mce₆, P-GFLG-SOS, P-GFLG-DOX, P-GFLG-Mce₆ against A498 cells were evaluated by exposing cells to combinations of free drugs/conjugates at ratios based on their respective IC₅₀ concentrations (Table 2). Figure 2 shows the composite dose-response curves of A498 cells indicating the anti-proliferative effects of single agents and their combinations. The dose-response curves for combined treatment were obtained by plotting cell viability (y) vs. the combined dose of two single agents (x). All of the binary combination treatments showed anti-proliferative activities toward the A498 cell line. The dose ratio, IC₅₀ or D_m values of the combination treatment, and dose of each drug/conjugate combination that inhibit cell growth by 50% are shown in Table 3. The dose of each drug/conjugate in combination was substantially lower than the IC₅₀ doses of the drugs as single agents (compare Tables 2 and 3). These results clearly indicate that all of the combination treatments were effective against A498 cells.

Figure 2.

Dose-response curves of A498 cells treated with SOS, DOX, Mce₆, P-GFLG-SOS, P-GFLG-DOX, and P-GFLG-Mce₆ as single agents and binary combinations at constant ratios of their respective IC₅₀ concentrations; bars represent the standard error (n=6).

Table 3.

Dose ratios and IC₅₀ doses in combination treatments of free drugs (SOS+DOX, SOS+Mce₆, DOX+Mce₆) and their copolymer conjugates (P-GFLG-SOS+P-GFLG-DOX, P-GFLG-SOS+P-GFLG-Mce₆, P-GFLG-DOX+P-GFLG-Mce₆) in A498 cells

Drug combination				D (μM) b
Drug A	Drug B	Dose ratio	Dm (μM) a	Drug A	Drug B
SOS	DOX	1:80	0.0213 ± 0.002	0.00027	0.021
SOS	Mce6	1:1330	0.270 ± 0.028	0.00020	0.27
DOX	Mce6	1:16.7	0.392 ± 0.018	0.022	0.37
P-GFLG-SOS	P-GFLG-DOX	1:1200	1.37 ± 0.052	0.0011	1.37
P-GFLG-SOS	P-GFLG-Mce6	1:280	1.48 ± 0.291	0.0053	1.47
P-GFLG-DOX	P-GFLG-Mce6	1:0.23	3.71 ± 0.337	3.01	0.70

^aD_m represent the means ± SEM (n=6). The statistical comparisons in D_m between single agent treatments (Table 2) and combination treatments (Table 3) were significant at p-value < 0.05 (ANOVA).

^bD is the dose of each drug/conjugate in a combination that inhibits the cell growth by 50%.

To evaluate potential synergy of the combinations of free drugs and copolymer conjugates towards A498 cells in vitro, the combination index (CI) analysis was used (Chou and Talalay, 1984; Chou, 2006). In the CI analysis, values of CI < 1, CI = 1, and CI > 1 indicate synergy, additivity, and antagonism, respectively. Figure 3 and Table 4 show a graphic summary of the CI analyses over all levels of effect (f_a = 0.05–0.95 or 5–95% of inhibition effect) and CI values at 25%, 50%, 75% and 95% of growth inhibition effects in A498 cells, respectively. It should be noted that at very high and at very low drug effect levels, the method is less accurate due to logarithmic transformation (Zhao et al., 2004). Consequently, we have chosen f_a from 0.05 to 0.95 for evaluation. The most synergistic binary combination was SOS+DOX. Simultaneous addition of SOS and DOX to A498 cells in a monolayer culture yielded CI values lower than 1 over the entire range of cytotoxicity, indicating a strong synergistic to moderate synergistic effect. The SOS+Mce₆ and P-GFLG-SOS+P-GFLG-Mce₆ combinations both displayed synergism for f_a values up to about 0.8, but showed slight antagonism and near additivity at f_a = 0.95, respectively. However, all other combinations were synergistic to varying degrees up to f_a < 0.9, and were additive at higher f_a values. The experimental data were also analyzed using the isobologram method (Loewe, 1953). The result comparison (data not shown) indicated that both the CI and isobologram analyses produced very similar results.

Figure 3.

Combination index plots (Fa-CI plots) obtained from median-effect analysis. (A) free drug combinations, and (B) copolymer conjugate combinations. Chemotherapeutic drugs and their copolymer conjugates were graually diluted at the ratio of their IC₅₀ values as a series of two-fold dilutions from 8 to 0.03125 times IC₅₀ and A498 cells exposed to drugs simultaneously for 16 h as described in Materials and methods. CI <1, =1 and >1 indicates synergism, additive effect and antagonism, rspectively. The vertical bars indicate the 95% confidence intervals based on Sequential Deletion Analysis (SDA) and can be generated by using CompuSyn software.

Table 4.

Combination index values at different effect levels for combination treatments of A498 cells with free and HPMA copolymer-bound drugs

Drug combination		Combination index (CI) a at
Drug A	Drug B	IC25	IC50	IC75	IC95
SOS	DOX	0.16 ± 0.02	0.18 ± 0.01	0.25 ± 0.02	0.60 ± 0.17
SOS	Mce6	0.34 ± 0.08	0.45 ± 0.08	0.64 ± 0.11	1.33 ± 0.38
DOX	Mce6	0.23 ± 0.04	0.20 ± 0.03	0.27 ± 0.03	0.77 ± 0.16
P-GFLG-SOS	P-GFLG-DOX	0.46 ± 0.07	0.54 ± 0.06	0.66 ± 0.06	0.97 ± 0.15
P-GFLG-SOS	P-GFLG-Mce6	0.69 ± 0.05	0.77 ± 0.05	0.88 ± 0.07	1.19 ± 0.13
P-GFLG-DOX	P-GFLG-Mce6	0.20 ± 0.06	0.21 ± 0.04	0.31 ± 0.06	1.02 ± 0.66

^aCombination index (CI) values <1, =1, and >1 characterize synergism, additive effect, and antagonism, respectively.

CI values shown are mean ± SEM (n=6).

The synergistic effects of combinations may depend on the cytotoxic mechanism of each agent. SOS, DOX, and Mce₆ have different sites of action, but produce similar cytotoxic intermediates and outcomes including: DNA damage, cellular damage, cell cycle arrest, and cell apoptosis. DOX+Mce₆ and P-GFLG-DOX+P-GFLG-Mce₆ combinations showed better synergistic effects compared to other combinations, probably due to the multiple mechanisms of action of DOX, including generation of reactive oxygen species (Sinha et al., 1987), which is analogous to the activity of Mce₆. The mechanism of this synergistic effect is the potentiation of PDT by DOX and vice versa. In the cell, DOX can be reduced by NADPH cytochrome P450 reductase to semiquinone-free radicals. In the presence of molecular oxygen, the semiquinone radicals are capable of enhancing superoxide production (Berlin and Haseltine, 1981; Bartoszek, 2002), which can cause damage to biological molecules. This suggests that DOX+Mce₆ and P-GFLG-DOX+P-GFLG-Mce₆ combinations produced similar cytotoxic intermediates with a concomitant enhancement of efficacy. Enhanced efficacy resulting from the combination of a photosensitizer and DOX has been demonstrated on various cell lines, such as Walker 256 carcinosarcoma cells (Edell and Cortese, 1988), H-MESO-1 human malignant mesothelioma cells (Brophy and Keller, 1992), murine L929 cells (Lanks et al., 1994), and murine hepatoma MH-22A (Kirveliene et al., 2006).

SOS+DOX and P-GFLG-SOS+P-GFLG-DOX combinations also showed synergistic effects. Both drugs act on DNA: SOS induces DNA-protein and DNA-DNA crosslinks with no detectable DNA strand breaks (Nieves-Neira et al., 1999), but DOX can intercalate into DNA strands and also produce non-protein-associated and protein-associated DNA strand breaks (Fornari et al., 1994). Regardless of the detailed mechanism, both SOS and DOX induce DNA damage, cell cycle arrest at G₁ and G₂, and, ultimately cell apoptosis.

Furthermore, many in vivo studies have described enhanced antitumor activity of DOX in combination with other chemotherapeutic agents. Examples include: the combination of immunoconjugate BR96-DOX and paclitaxel against MCF7 human breast carcinoma, L2987 human lung carcinoma, RCA and LS174T human colon carcinomas in athymic mice (Trail et al., 1999), doxorubicin with anti-fetal liver kinase 1 monoclonal antibody in human SKLMS-1 leiomysarcoma and RD rhabdomyosacoma xenografts in SCID mice (Zhang et al., 2002), pegylated liposomal doxorubicin administered with nanoliposomal topotecan for treatment of intracranial brain tumor xenografts (Yamashita et al., 2007), and the combination of liposomal doxorubicin and topotecan, docetaxel, gemcitabine, capecitabine, or celecoxib for treatment of OVCAR-3 and ES-2 human ovarian carcinoma xenografts (Saucier et al., 2007).

SOS+Mce₆ and P-GFLG-SOS+P-GFLG-Mce₆ combinations showed weaker synergistic effects than the other combinations. It is not clear how the facts, that SOS and Mce₆ have different mechanisms of action and produce different cytotoxic intermediates, relate to this phenomenon.

The DRI values in Table 5 indicate the fraction that the drug concentrations can be decreased by to achieve the IC₅₀. For example, the IC₅₀ value was 0.003 μM SOS or 0.24 μM DOX in single agent treatments (Table 2), but a 1:80 combination of SOS+DOX can inhibit 50% of cell growth using 0.00027 μM SOS and 0.021 μM DOX. This represents an 11-fold decrease for both SOS and DOX concentrations. All combinations produced IC₅₀ values with DRIs ranging from 4.5 to 20.5. These results indicated that the median effect dose of each agent could be reduced when used in combination and consequently reduce the non-specific side effect of each agent.

Table 5.

Dose-reduction index values at 50% effect levels of combined treatments of free drugs and their copolymer conjugates in A498 cells

Drug combination		Dose-reduction index (DRI) a for
Drug A	Drug B	Drug A	Drug B
SOS	DOX	11.16 ± 1.38	11.12 ± 0.42
SOS	Mce6	14.96 ± 0.53	13.36 ± 1.73
DOX	Mce6	10.74 ± 1.28	9.62 ± 1.30
P-GFLG-SOS	P-GFLG-DOX	20.50 ± 0.93	19.52 ± 1.37
P-GFLG-SOS	P-GFLG-Mce6	4.46 ± 0.05	4.97 ± 0.42
P-GFLG-DOX	P-GFLG-Mce6	8.90 ± 0.50	10.42 ± 1.61

^aDose-reduction index (DRI) values at IC₅₀ (or D_m) were determined using DRI = (D_x)/(D) (see Materials and methods).

DRI values represent means ± SEM (n=6).

Previously, in vitro studies of the interaction between free DOX and Mce₆ using the isobologram method (Peterson et al., 1995) and the cooperation between free and HPMA copolymer -DOX and HPMA copolymer -Mce₆ conjugates (Lu et al., 1999) in human ovarian OVCAR-3 carcinoma cells demonstrated that the combination DOX+Mce₆ decreased the percentage of viable cells and displayed synergistic-to-additive effects in the dose range tested. P-GFLG-DOX improved the efficacy of P-GFLG-Mce₆ when the variable dose of P-GFLG-DOX was simultaneously added to an effective dose of P-GFLG-Mce₆. By contrast, P-GFLG-Mce₆ did not significantly improve the efficacy profile of P-GFLG-DOX when the variable dose of P-GFLG-Mce₆ was simultaneously added to an effective dose of P-GFLG-DOX (Lu et al., 1999).

Recently, Vicent and collaborators (Greco et al., 2005; Vicent et al., 2005) have synthesized an HPMA copolymer conjugate containing both DOX and the aromatase inhibitor aminoglutethimide (AGM) attached to one macromolecule and evaluated its biological activity of toward MCF-7 and MCF-7ca breast cancer cells. Their results showed that HPMA copolymer bearing both DOX and AGM on one macromolecule showed increased cytotoxicity in vitro, while a combination of P-GFLG-DOX and HPMA copolymer-bound AGM (P-GFLG-AGM) exhibited low toxicity. The mechanism is unknown, but is probably related to differences in the subcellular pharmacokinetics of one conjugate vs. the combination of two conjugates possessing one drug.

Recently, many researchers have focused on drug combinations in vivo and in clinical trials to determine the synergism or antagonism of combined drugs in order to lower dosages and reduce side effects. A recently published review of clinical studies of combination treatments of renal cell carcinoma (Amato, 2005) showed five different combination regimens with 5% to 71% antitumor response along with acceptable toxicity profiles. Hainsworth et al. studied the clinical efficacy and toxicity of combined treatment of metastatic renal cell carcinoma with bevacizumab and erlotinib in a phase II trial. It was demonstrated that both targeted agents in combination were effective and well-tolerated (Hainsworth et al., 2005). Using drug combinations with different drug exposure schedules, or simultaneous vs. consecutive exposure, may result in different antitumor activities. For example, the combination of a bis-phenazine (XR5944) with 5-fluorouracil (5-FU) and irinotecan, simultaneously or sequentially, against colon cancer HT29 cell line showed that simultaneous exposure of cells to XR5944 and 5-FU or irinotecan exhibited antagonism, while sequential exposure to either order of these drugs displayed additive effects or better (Harris et al., 2005).

4. Conclusions

Combinations of free and HPMA copolymer-bound SOS, DOX, and Mce₆ produced synergistic effects in the treatment of A498 renal carcinoma cells. This bodes well for further development of macromolecular nanomedicines based on HPMA copolymers. Several HPMA copolymer drug conjugates underwent clinical testing, including DOX (Vasey et al., 1999) and platinates (Rademaker-Lakhai et al., 2004). The advantages of combination therapy using HPMA copolymer conjugates have been demonstrated on animal models of ovarian carcinoma (Krinick et al., 1994; Peterson et al., 1996; Shiah et al., 1999, 2000, 2001b). The results presented here highlight the potential applications of synergistic combinations and dose reduction for combination therapy for renal carcinoma. These observations may prove useful in the development of in vivo combination study protocols for the treatment of renal cancer and could be further confirmed in future clinical applications.