Carfilzomib-Loaded Ternary Polypeptide Nanoparticles Stabilized by Polycationic Complexation

Abstract

Carfilzomib (CFZ) is a second-generation proteasome inhibitor showing great efficacy in multiple myeloma treatment, yet its clinical applications for other diseases such as solid cancers are limited due to low aqueous solubility and poor biostability. Ternary polypeptide nanoparticles (tPNPs) are drug carriers that we previously reported to overcome these pharmaceutical limitations by entrapping CFZ in the core of the nanoparticles and protecting the drugs from degradation in biological media. However, preclinical studies revealed that tPNPs would require further improvement in particle stability to suppress initial burst drug release and thus achieve prolonged inhibition of proteasome activity with CFZ against tumor cells in vivo. In this study, CFZ-loaded tPNPs are stabilized by polycations which have varying pKa values and thus differently modulate nanoparticle stability in response to solution pH. Through polyion complexation, the polycations appeared to stabilize the core of tPNPs entrapping CFZ-cyclodextrin inclusion complexes while allowing for uniform particle size before and after freeze drying. Interestingly, CFZ-loaded tPNPs (CFZ/tPNPs) showed pH-dependent drug release kinetics, which accelerated CFZ release as solution acidity increased (pH < 6) without compromising particle stability at the physiological condition (pH 7.4). In vitro cytotoxicity and proteasome activity assays confirmed that tPNPs stabilized with cationic polymers improved bioactivity of CFZ against CFZ-resistant cancer cells, which would be greatly beneficial in combination with pH-dependent drug release for treatment of solid cancers with drug resistance and tumor microenvironment acidosis by using CFZ and other proteasome inhibitors.

Introduction

Carfilzomib (CFZ) is an epoxomicin analog that has demonstrated great efficacy in treating patients with relapsed or refractory multiple myeloma who had at least two prior therapies,^1,2 and it is the second-in-class proteasome inhibitor approved by the FDA in 2012.^3,4 Unlike first-generation proteasome inhibitors such as Bortezomib, CFZ binds selectively and irreversibly to the 20S proteasome to inhibit its chymotrypsin-like activity.^5–7 Through this mechanism, CFZ prevents the proteasome from tagging proteins with ubiquitin for cellular degradation, which intracellularly builds up defective proteins that induces cell stress, cell cycle arrest, apoptosis and ultimately inhibits the growth of cancer cells.⁶ Furthermore, CFZ has been effective in reducing dose limiting toxicity associated with Bortezomib among patients that developed resistance to the first-generation proteasome inhibitor.⁸

The great efficacy and potential benefits of CFZ have garnered attention to treat other cancers such as advanced solid and metastatic cancers using the drug, but there have been limited success in the clinic due to poor biostability of the tetrapeptide (Phe-Leu-Phe-Leu) with an epoxyketone ring that undergoes protonation and subsequently ring opening catalyzed under acidic microenvironment to make the pharmacophore inactive in tumors with acidosis.^9–11 The amide bonds linking the peptides were found also susceptible to enzymatic degradation in vivo.¹² In addition to poor biostability, CFZ has other pharmaceutical limitations such as low water solubility and efflux by P-glycoprotein (P-gp),^13–16 which justify the development of injection formulations for CFZ.

To mitigate the pharmaceutical limitations and maximize therapeutic potential of CFZ, we previously developed ternary polypeptide nanoparticles (tPNPs) that consist of drug-cyclodextrin inclusion complexes enveloped by biocompatible polymers, which are further stabilized through polyion complexation with small molecule organic acids such as citric acid and lactic acid.^17–25 Preclinical studies revealed that CFZ-loaded tPNPs (CFZ/tPNPs) significantly improved biostability of the tetrapeptide in vitro, yet tPNPs would require further improvement in particle stability to suppress initial burst drug release and thus achieve prolonged inhibition of proteasome activity with CFZ against tumor cells in vivo.

In this study, we designed tPNPs comprising amine-substituted β-cyclodextrins (CD) and biocompatible poly(ethylene glycol)-poly (L-glutamic acid) (PEG-PLE) that self-assemble into nanoparticles in solution while polycations with varying pKa values were used to stabilize the assemblies through polyion complexation (Fig. 1). We hypothesized that cationic polymers would stabilize the core of tPNPs by neutralizing charges between positive CD with amines and negative PEG-PLE with carboxylic acids to ensure a well-defined structure of a nanoparticle with uniform size, high drug loading and stability in aqueous solutions. The cationic polymers were also expected to modulate the release of CFZ from CFZ/tPNPs in a pH-dependent manner due to their pKa values determining the dissociation of tPNPs through disruption of charges in the nanoparticle core. Such pH-sensitive nanoparticles hold great promise to develop smart drug delivery systems that remain stable in the physiological pH 7.4 yet trigger drug release when exposed to acidic environment such as tumors with acidosis (pH 6.8−7.2) or endosomes and lysosomes in the cell (pH 5 – 6.8).^26–28 Thus, tPNPs may provide a unique option to deliver CFZ to tumor cells in the body while increasing therapeutic efficacy and reducing toxicity of the proteasome inhibitor.

Fig. 1.

CFZ-loaded tPNPs stabilized by polycations with varying pKa values.

Based on the background and rationale, we tested three polycations with varying pKa values as stabilizing agents for CFZ/tPNPs: poly(ethylene glycol)-poly(L-lysine) (PLL, pKa ≈ 10), poly(L-histidine) (PLH, pKa ≈ 6), and linear polyethylene imine (PEI, pKa ≈ 7, where pKa1 = 4.5 and pKa2 = 10). PLL, PLH, and PEI are polycations that are widely used as building components for various drug and gene delivery systems due to their biocompatibility and safety.^29–40 Thus, with ensured safety of the selected polycations, this study focuses on investigating if CFZ/tPNPs stabilized by polycations would show uniform physiochemical properties such as particle size, surface charge and drug loading while improving drug release profiles to enhance proteasome inhibition for CFZ.

Materials and Methods

Materials

Potassium biphthalate sodium hydroxide buffer and HPLC grade water were obtained from Fisher Scientific (Fair-lawn, NJ). Slide-A-Lyzer Mini Dialysis Units (20,000 MWCO) were obtained from Thermo Scientific (Rockford, IL). Heptakis (6-amino-6-deoxy)-β-CD heptahydrochloride (CD) was purchased from Cyclolab R&D Laboratory Ltd. (Hungary). Citric acid was from EM Science (Gibbstown, NJ). CFZ was purchased from LC laboratories (Woburn, MA). Methoxy-poly(ethylene glycol)-poly(L-glutamic acid) (PEG-PLE), methoxy-poly(ethylene glycol)-poly(L-lysine) (PLL) was purchased from Alamanda Polymers (Huntsville, AL). α-Ω-bis-(amino)-terminated poly(L-histidine) (PLH) was purchased from Polymer Source Inc (Canada). Linear polyethyleneimine (PEI) was purchased from Polysciences Inc (Warrington, PA).

Cell Culture

Human colorectal adenocarcinoma cell (DLD1) was from ATCC and cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum, 1% sodium pyruvate and 1% penicillin-streptomycin at 37 °C in a humidified incubator with 5% CO₂. The cells were maintained in an exponential growth phase by periodic sub-cultivation. For CFZ-resistant DLD1 cells, parent DLD1 cells were treated with increasing concentration of CFZ for about 4 months until cells became > 95% survival at a concentration of 1 μM CFZ.

Preparation of tPNPs

tPNPs were prepared by using a solvent evaporation method as reported previously with slight modification.^17,18 Briefly, CFZ in ethanol was mixed with CD dissolved in deionized water at a 1:4 mass ratio in a 50 mL round bottom flask. Citric acid was added to the mixture to facilitate complexation between CFZ and CD while preventing the inclusion complexes from precipitating. After sonicating the solution for 2 minutes, ethanol was removed by rotary evaporation at 60 °C with reduced pressure. Following the removal of ethanol, PEG-PLE was added to the aqueous solution of CFZ/CD inclusion complexes and the solution was sonicated for 5 minutes to prepare CFZ/tPNPs in the presence of PLL, PLH or PEI at mixing ratios to neutralize charges in the nanoparticle core. CFZ/tPNPs were then frozen in dry ice and lyophilized overnight (Labconco Freezone^®) using 2% (w/v) sucrose as a cryoprotectant to obtain a fine powder of nanoparticles. The primary and secondary drying was performed at −50 °C and 25 ° C for 24 h, followed by storage at −20 °C prior to use.

Particle Size and Zeta Potential Determination

The particle size and surface charge of tPNPs with and without CFZ were determined by dynamic light scattering (DLS) and zetapotential measurements (Zetasizer Nano ZS, Malvern, UK). Freeze-dried CFZ/tPNPs were resuspended and sonicated in normal saline at 2 mg/mL. The resultant suspension was placed in a folded capillary cuvette and DLS was measured with 173° backscatter at 25 °C. Data including the polydispersity index (PDI) of particle size were also calculated with the accompanying software.

Drug Loading and Encapsulation Efficiency

Drug loading of CFZ/tPNPs was determined by high performance liquid chromatography (HPLC, Agilent XDB-C18). The mobile phase consisted of water and acetonitrile (ACN) with 0.1% formic acid. Using an isocratic method of 55:45 of ACN:H₂O at a flow rate of 0.5 mL/min, CFZ was eluted at a retention time of 3.6 minutes. Drug concentrations were measured by integrating peaks at a wavelength of 210 nm and the peak area was compared to a calibration curve established with a serial dilution of CFZ. Drug loading efficiency was defined as the percentage of CFZ encapsulated to the weight of nanoparticles, and encapsulation efficiency was defined as the percentage of drug encapsulated to drug added. Data are represented as mean § standard deviation (SD).

Drug Release Confirmation

CFZ release kinetics was determined as previously reported.⁴¹ Briefly, freeze-dried CFZ/PLL-tPNP, CFZ/PLH-tPNP, and CFZ/PEI-tPNP were dissolved in PBS (pH 7.4, 10 mM) at 10 mg/mL and placed in dialysis cups (MWCO 20 kDa). The dialysis cups were then placed in 4 L buffers at pH 7.4, 6 or 5 at 37 °C to simulate conditions in blood stream, acidic tumor microenvironment, and intracellular lysosomes. Nanoparticle solutions were retrieved at 0, 0.5, 1, 3, 6, 24, 48 and 72 h, and drug remaining was measured using the HPLC method described above. Percent drug remaining at each time point was calculated as a normalized value of drug remaining compared to drug concentration at t = 0 h. The drug release profiles were fitted to an exponential decay with GraphPad Prism 9 software.

Cell Viability Determination

CFZ-resistant DLD1 cells were cultured in drug-free media for 9 days prior to the cell viability assay. DLD1 cells were seeded in a 96-well plate at 5,000 cells per well and incubated overnight for attachment. The cells were treated with serial dilutions of free CFZ in DMSO and CFZ/tPNPs in aqueous solutions. After treating the cells for 96 h, 20 μL of MTS dyes was added to each well and the plates were placed in an incubator at 37 °C and 5% CO₂ for 2 h to determine cell viability. Absorbance was measured at 490 nm using a plate reader (Spectra Max M5). The MTS assay (Promega) was used to determine half-maximal inhibitory concentration (IC₅₀) values for CFZ/PLL-tPNP, CFZ/PLH-tPNP, and CFZ/PEI-tPNP, which were calculated by a non-linear regression method using GraphPad Prism 9 software.

Proteasome Activity Assay

Proteasome activity was determined by measuring the chymotrypsin-like activity in CFZ-resistant DLD1 cells treated with free CFZ, CFZ/PLL-tPNP, CFZ/PLH-tPNP or CFZ/PEI-tPNP. The cells were cultured for 9 days without CFZ and then seeded in a 6-well plate at 150,000 cells/well for additional 24 h. The cells were incubated with free CFZ or CFZ/tPNPs at CFZ equivalent concentrations (0.5 μM) at 37 °C with 5% CO₂ for 6, 48 and 96 h. At each time point, cell pellets were washed with DPBS and harvested for storage at −20 °C until they were lysed with 1x Passive Lysis Buffer. The total protein concentrations were determined by the Bradford assay. The proteasome activity was measured by treating 5 μg of protein collected from each sample with 100 μM of Suc-LLVY-AMC substrate. Fluorescence was quantified with a plate reader (Spectra Max M5) at a rate of 1 reading per minute with excitation and emission wavelengths of 360 and 460 nm, respectively. The percent of proteasome activity remaining was determined as the relative fluorescence between treated and untreated cells. We performed multiple unpaired t-tests for statistical analysis to compare the means of groups for a time point. Statistical significance was determined with a p value < 0.01 or < 0.05 with respect to cells treated with free CFZ using GraphPad Prism 9 software.

Results

Physiochemical Properties of tPNPs

PLL, PLH and PEI were used in this study because they induced no precipitation of tPNPs after CFZ loading in preliminary experiments. All freeze-dried CFZ/tPNPs readily dissolved in aqueous solutions with simple vortexing and showed a unimodal particle size distribution (Fig. 2). The particle size of empty and CFZ/tPNPs was less than 100 nm and zeta potentials were ± 5 mV (Table 1).

Fig. 2.

Particle size distribution of empty and CFZ-loaded tPNPs determined by dynamic light scattering measurement. The He-Ne laser (λ = 633 nm) was used to measure the speed of nanoparticles undergoing Brownian motion.

Table 1.

Physiochemical properties of tPNPs.

Formulation	Particle size (nm)	PDI	Zeta potential (mV)	Drug loading capacity (%)	Entrapment efficiency (%)
Empty PLL-tPNP	34.9 ± 3.0	0.09 ± 0.03	−0.69 ± 0.40	−
CFZ/PLL-tPNP	54.2 ± 12.5	0.69 ± 0.01	−1.02 ± 0.43	11.44 ± 0.48	95.3 ± 2.3
Empty PLH-tPNP	41.8 ± 3.5	0.26 ± 0.01	−1.63 ± 0.40	−
CFZ/PLH-tPNP	91.9 ± 13.5	0.63 ± 0.05	−1.59 ± 0.29	9.80 ± 0.98	91.0 ± 3.2
Empty PEI-tPNP	34.0 ± 3.3	0.20 ± 0.03	−1.07 ± 0.37	−
CFZ/PEI-tPNP	56.52 ± 9.8	0.48 ± 0.04	2.86 ± 0.49	8.95 ± 0.92	80.3 ± 4.5

Drug Loading and Entrapment Efficiency

The amounts of CFZ loaded in tPNPs were calculated using the formulas described in the methods section. CFZ/PLL-tPNP, CFZ/PLH-tPNP and CFZ/PEI-tPNP had drug loading of 11.44%, 9.80% and 8.95% respectively. The encapsulation efficiency for CFZ/PLL-tPNP, CFZ/PLH-tPNP, CFZ/PEI-tPNP was 95.3%, 91.0% and 80.3% respectively. These results confirmed that polycations showed no adverse effect on loading and entrapment of CFZ in tPNPs.

Drug Release of CFZ from tPNPs

Drug release experiments were performed at 37 °C in buffers with three pH values mimicking the conditions in blood stream (pH 7.4), acidic tumor microenvironment (pH 6.0), and intracellular lysosomes (pH 5.0). At pH 7.4, none of tPNPs showed burst drug release, which suggests the particle stabilizing effect of polycations. CFZ/tPNPs, however, accelerated drug release as pH decreased below 6.0 while complete disposal of drug was observed at pH 5.0 for CFZ/PEI-tPNP after 48 h incubation and CFZ/PLH-tPNP after 3 h incubation (Fig. 3). Drug release rates were further analyzed by determining drug release half-life with biphasic curve fitting (Table 2). For all CFZ/tPNPs, drug release underwent short transitions from initial fast drug release (t_1/2 fast < 1 h) to the second phase of modest drug release (t_1/2 slow > 30 h) at pH 7.4, while t_1/2 slow values significantly dropped as drug release was accelerated at pH 6.0 and 5.0. Interestingly, at pH 7.4, CFZ/PLL-tPNP showed no t_1/2 slow yet maintained a sustained release over 72 h while CFZ/PEI-tPNP and CFZ/PLH-tPNP demonstrated biphasic drug release profiles. These results showed that all CFZ/tPNPs stabilized with polycations were stable at pH 7.4 with drug release half-life greater than 30 h while accelerating drug release in acidic conditions (pH 6.0) in the order of PEI-tPNP (fastest), PLH-tPNP and PLL-tPNP (slowest). Only PLH induced complete dissociation of tPNPs at pH 5.0 in the time frame of 72 h.

Fig. 3.

Effects of pH on the release of CFZ from tPNPs. Drug release experiments were performed at 37 °C in buffers with pH values mimicking the conditions in blood stream (pH 7.4), acidic tumor microenvironment (pH 6.0), and intracellular lysosomes (pH 5.0), respectively.

Table 2.

Drug release kinetic soft PNPs.

PIC	pH	t1/2fast(h)	t1/2slow(h)
CFZ/PLL-tPNP	7.4	N.D.	44.04 ± 5.25
	6.0	0.54 ± 0.12	25.27 ± 2.70
	5.0	0.63 ± 0.08	19.98 ± 11.5
CFZ/PLH-tPNP	7.4	0.30 ± 0.01	48.73 ± 1.01
	6.0	0.32 ± 0.121	7.64 ± 4.54
	5.0	0.34 ± 0.11	0.34 ± 0.06
CFZ/PEI-tPNP	7.4	0.39 ± 0.23	30.72 ± 5.63
	6.0	0.49 ± 0.05	16.09 ± 1.31
	5.0	0.45 ± 0.17	7.54 ± 1.63

Cell Viability

Fig. 4 shows the results of cytotoxicity assays in CFZ-resistant human colorectal cancer cells. Drug resistance of DLD1 cells was confirmed by > 10% of cell viability at 100 μM of free CFZ in DMSO. The cell viability curve shows that CFZ/PLL-tPNP effectively inhibited the growth of CFZ-resistant cells in comparison to free CFZ. CFZ/PLH-tPNP was as equally effective as free CFZ yet CFZ/PEI-tPNP failed to suppress more than 50% of cell viability even at the maximum aqueous solubility achieved for CFZ using tPNPs. Among tPNPs, CFZ/PLL-tPNP had the lowest IC₅₀ value (3.9 ± 0.2 μM), which was approximately 1.8 and >3.6 times more effective than CFZ/PLH-tPNP (7.0 ± 0.5 μM) and CFZ/PEI-tPNP (> 14 μM), respectively.

Fig. 4.

Cytotoxicity of CFZ-loaded tPNPs in comparison to free drug. Viability of CFZ-resistant DLD1 cells was measured after treating the cells with serial dilutions of free CFZ in DMSO or CFZ/tPNPs in aqueous media.

Proteasome Activity

Free CFZ used as a positive control induced a rapid reduction in proteasome activity in CFZ-resistant DLD1 cells following 6 h of treatment, but the proteasome activity recovered after 48 post incubation (Fig. 5). CFZ/PLL-tPNP inhibited the proteasome activity by 50% as effectively as free CFZ while CFZ/PLL-tPNP and CFZ/PLH-tPNP showed mediocre proteasome activity inhibition (< 30%) at 6 h post incubation. Interestingly, all tPNPs showed proteasome activity inhibition greater than free CFZ at 48 h, although proteasome activity recovered in the next 48 h. CFZ/PLL-tPNP was the only formulation that remained to suppress proteasome activity more than 50% at 96 h post incubation. These results revealed that CFZ/PLL-tPNP induced the best combination of effective inhibition and prolonged suppression of proteasome activity.

Fig. 5.

Confirmation of proteasome activity inhibition by CFZ-loaded tPNPs. Proteasome activity remaining was measured in CFZ-resistant DLD1 cells after treating the cells with free CFZ or CFZ-loaded tPNPs for 6, 48, or 96 h. Data was analyzed to determine statistical significance in rapid (< 6 h) and prolonged (> 48 h) inhibition of proteasome activity for CFZ/tPNPs with respect to free CFZ (*: p < 0.01, **: p < 0.05).

Discussion

CFZ has demonstrated great therapeutic efficacy in treating multiple myeloma yet achieved limited success in treating solid cancers due to pharmaceutical limitations such as low water solubility and poor biostability.^42–46 To address these problems and maximize therapeutic potential of CFZ, we have developed injection formulations that control the delivery and release of the proteasome inhibitor to solid tumors by using tPNPs.^17–25 tPNPs are biocompatible nanoparticles with an average diameter < 50 nm, which is suitable for tumor-specific drug delivery following intravenous injection.⁴⁷ Our previous studies suggest that tPNPs have the advantage of low immunogenicity, better stability, versatile structures, and higher drug loading compared to other nano formulations like liposomes and micelles.^17,18,20,48 When designed with optimized particle size and surface characteristics, tPNPs can increase in vivo blood circulation time, stealth, and drug accumulation in tumors. However, CFZ-loaded tPNPs still need improvement in suppressing initial burst drug release, which appeared critical to maximize therapeutic outcomes for CFZ against solid cancers.

In this study, we developed a self-assembled pH sensitive tPNPs with a high CFZ loading capacity and encapsulation by using polycations as stabilizers (Fig. 1). tPNPs entrap CFZ-CD inclusion complexes in the core through polyion complexation, which is further stabilized with polycations that neutralize the charge in the nanoparticle core. We found that the inclusion complex ensures high loading of CFZ while polyion complexation stabilized with polycations modulates the electrostatic interactions between the positively charged CFZ-CD inclusion complexes with the negatively charged PEG-PLE to finely tune drug release while improving particle stability without compromising the uniformity of tPNPs (Fig. 2). All tPNPs exhibited particle size < 100 nm and neutral surface charge even in the high loading capacity of CFZ (Table 1). The sub-100 nm particles are expected to facilitate the delivery of CFZ to tumors by taking advantage of the leaky vasculature and poor lymphatic drainage in tumor tissue to improve nanoparticle accumulation in tumors, which is also known as the enhanced permeation and retention effect.^49–51 It must be noted that the surface charge of CFZ/tPNPs was neutral (± 5 mV) regardless of the polycation used, which is favorable for prolonged retention of the nanoparticles in biological systems by avoiding non-specific binding to plasma proteins or cellular uptake by the mononuclear phagocytic system.^52,53 In addition, tPNPs showed a high encapsulation efficiency of > 80%, which can deliver a large quantify of CFZ to tumors for better efficacy. This is a clear difference from many other nanoparticle formulations with low drug encapsulation that often show insufficient accumulation of cytotoxic agents in disease sites and thus poor therapeutic outcomes.^54,55

Drug release experiments revealed that tPNPs released CFZ at a slow rate without showing burst drug release at the physiological pH 7.4, which may prolong drug exposure to tumor cells for better efficacy (Fig. 3). The slow drug release was probably due to the core of tPNPs stabilized by polycations that neutralized the charges between the CFZ-CD inclusion complexes and PEG-PLE polymers and thus made it difficult for the hydrophobic drugs to escape tPNPs. Interestingly, polycations appeared to trigger dissociation of tPNPs to accelerate drug release depending on the degree of ionization as a function of pKa that induces charge repulsion in the core of tPNPs. PLL (pKa ≈ 10) indeed maintained particle stability and sustained drug release in a broad pH range (7.4−5.0), whereas PLH (pKa ≈ 6) and PEI (pKa ≈ 7) triggered fast drug release as pH decreased below 6.0. In comparison to PLH that abruptly dissociated at pH 5.0 within 3 h post incubation, PEI showed better stability in acidic conditions up to 24 h post incubation, which was probably due to the buffering effect of the polymer with two pKa values (pKa1 = 4.5 and pKa2 = 10). Thus, the unique two-phase decay of CFZ/tPNPs in acidic conditions might be attributed to secondary assembly formation between anionic PLE and cationic CFZ-CD complexes as polycations move around during particle dissociation, although further investigation on the inter- and intra-molecular interactions among polymers within the tPNP particle is needed.

The acid-sensitive drug release profiles are potentially beneficial for tPNPs to deliver CFZ to solid tumors because most tumor sites have acidic microenvironment compared to normal tissue.^56–59 To this end, tPNPs may stably travel through the bloodstream under normal physiological conditions (pH 7.4) while accelerating CFZ release in solid tumors with acidosis (pH < 6.8) or endosomes/lysosomes in cancer cells (pH 5–6.5).^60,61 Such a drug release scenario is partially supported by cytotoxicity assays shown in Fig. 4. Although three tPNP formulations showed similar particle sizes, drug entrapment yields, and drug release half-life at pH 7.4, they showed distinctive cytotoxic effects against CFZ-resistant DLD1 cells. CFZ/PLL-tPNP showed the highest cytotoxicity as the formulation showed the greatest difference in drug release at varying pH conditions. CFZ/PLL-tPNP with the greatest drug remaining at 72 h post incubation appeared to effectively kill DLD1 cells that have a doubling time of approximately 20 h, which is consistent with our previous findings that drug release rates synchronized with cell growth would be most effective to suppress the growth of tumor cells in vitro and in vivo.^62–64 In addition, tPNPs showed a clear pH dependent drug release profile. Even CFZ/PLL-tPNPs that showed no burst release at physiological pH exhibited a sustained release with half-life of about 44 h (Table 2). When the pH was reduced to 6 and 5, CFZ/PLL-tPNPs showed a biphasic release profile with the initial drug release half-life less than one hour for both pH values followed by a slow-release half-life of about 25 h at 6.0 and 20 h at 5.0. These results suggest that PNPs would release CFZ fast in endosomes or lysosomes and increase intracellular drug accumulation for improved inhibition of proteasome activity.

Related to the observed drug release and cytotoxicity patterns, tPNPs demonstrated intriguing effects in proteasome activity inhibition as shown in Fig. 5. At early time points of 6 h post incubation, free CFZ and CFZ/PLL-tPNP were equally effective in suppressing proteasome activity in DLD1 cells. However, CFZ/PLH-tPNP and CFZ/PEI-tPNP had mediocre effects on proteasome activity inhibition, which might be attributed to their drug release patterns discussed above. CFZ/PLL-tPNPs showed no biphasic drug release at pH 7.4, which could allow for tPNPs to internalize inside the cell without being recognized by P-gp as we previously observed.^65,66 For CFZ/PLH-tPNPs and CFZ/PEI-tPNP, the initial release might have facilitated P-gp activation to reduce intracellular drug concentrations. Over longer period (>48 h), all three tPNPs suppressed proteasome activity, which is a clear improvement over free CFZ. Targeting the ubiquitin-proteasome pathway has drawn more attention recently as a promising pathway for cancer treatment.⁶⁷ Suppression of proteasome activity has been effective for cancer treatment by inducing cell cycle arrest and subsequently apoptosis.⁶⁸ We postulate that CFZ/PLL-tPNPs killed CFZ-resistant DLD1 cells and suppressed proteasome activity for a prolonged period of time better than free CFZ as they modulated drug release in a pH-dependent manner and avoid drug efflux mediated by P-gp. From this perspective, tPNPs have great potential to improve the efficacy and toxicity of CFZ in future preclinical applications.

Conclusions

We have developed CFZ-loaded tPNPs stabilized by cationic polymers that improve biostability and modulate the release of the proteasome inhibitor from the nanoparticles in response to pH conditions, which would be beneficial to increase drug concentrations in solid tumors of acidosis or inside tumor cells overexpressing P-gp drug efflux pumps to greatly improve therapeutic outcomes of CFZ or other proteasome inhibitors.