Modulating Pharmacokinetics, Tumor Uptake and Biodistribution by Engineered Nanoparticles

Background Inorganic nanoparticles provide promising tools for biomedical applications including detection, diagnosis and therapy. While surface properties such as charge are expected to play an important role in their in vivo behavior, very little is known how the surface chemistry of nanoparticles influences their pharmacokinetics, tumor uptake, and biodistribution. Method/Principal Findings Using a family of structurally homologous nanoparticles we have investigated how pharmacological properties including tumor uptake and biodistribution are influenced by surface charge using neutral (TEGOH), zwitterionic (Tzwit), negative (TCOOH) and positive (TTMA) nanoparticles. Nanoparticles were injected into mice (normal and athymic) either in the tail vein or into the peritoneum. Conclusion Neutral and zwitterionic nanoparticles demonstrated longer circulation time via both IP and IV administration, whereas negatively and positively charged nanoparticles possessed relatively short half-lives. These pharmacological characteristics were reflected on the tumor uptake and biodistribution of the respective nanoparticles, with enhanced tumor uptake by neutral and zwitterionic nanoparticles via passive targeting.


Introduction
The limited solubility, stability, and short circulation time of traditional therapeutic agents can introduce side effects and unwanted accumulation of therapeutics into non-diseased tissue [1]. Multimodal nanoparticles, polymer-drug conjugates, monoclonal antibodies and immunoconjugates provide a means of increasing therapeutic efficacy [2], while also providing new therapeutic and imaging modalities [3]. The diversity of available structures and core materials [4,5] coupled with sub-cellular size and biocompatibility have generated numerous applications of these materials in fields ranging from delivery [6][7][8] and imaging [9], to photothermal ablation of tumors [10,11].
Key issues for the creation of effective nanotherapeutics include (i) overcoming biological barriers, (ii) specific accumulation of the therapeutic at the target site (i.e, targeting), and (iii) preventing rapid clearance [12]. While there have been a number of studies on the effects of particle size on biodistribution [13], little is known about the effect of surface charge on biodistribution and pharmacokinetics of nanoparticles, as these studies have generally used neutral poly(ethylene glycol) (PEG) functionalization [14]. This ''stealth'' coverage decreases the rate of opsonization, providing more efficient transport to the target tissue, e.g. tumor. The neutral coating provided by PEG coverage provides one route to enhanced bioavailability. Charged systems including zwitterionic surfaces [15], however, provide an alternative means of dictating the bioavailability of nanomaterials. To date, there has not been a detailed statistically robust investigation on how the surface charge of nanosystems impact their biodistribution, pharmacokinetics and tumor uptake.
In the current study, we systematically investigate the role of the surface charge of engineered gold nanoparticles in dictating the pharmacokinetics, and hence the biodistribution and tumor uptake of these nanomaterials. As the biodistribution of drug carriers is dependent upon the route of administration [16], we administered these particles using both intravenous (IV) and intraperitoneal (IP) routes [17]. We first assessed how surface charge impacts pharmacokinetics (PK) and biodistribution of the particles in a mouse model, demonstrating that surface charge determines circulation time of gold nanoparticles in the blood and tumor uptake through passive targeting. Through these studies, we reveal that surface charge and mode of administration has tremendous consequences on nanoparticle behavior in vivo, thus providing insight for the improved design in nanoparticle therapeutics.

Nanoparticle Fabrication
Our studies used gold nanoparticles (AuNP) featuring 2 nm diameter cores, with overall hydrodynamic diameters of ,9-10 nm. As shown in Figure 1, all nanoparticles contain a hydrophobic interior that confers stability [18] and a tetra(ethylene glycol) functionality to provide compatibility and solubility in cell culture media and other biological fluids. The interactions of these particles are dictated by the choice of headgroups, facilitating the determination of structure-activity correlations [19]. These particles are also exceptionally stable in biofluids [20], rendering them useful as drug delivery vehicles [21] and in sensing [22]. To provide a concise study of the role of charge in bioavailability, AuNPs featuring anionic, cationic and zwitterionic particle surfaces were fabricated using ligands having the appropriate headgroups [23]. These particles have been shown to be stable against aggregation in serum and have relatively low toxicities in both in vitro cell assays [24] and in fish [25]. Central to our studies, the common structure shared by these particles allows direct assessment of the role of charge on in vivo behavior (see Figure S1).

Plasma Pharmacokinetics is affected by nanoparticle surface charge
The plasma dispositions of the AuNPs were characterized in male CD1 mice after IV and IP administration. Plasma concentration-time profiles for each AuNP are illustrated in Figure 2, with the pharmacokinetic parameters are summarized in Table 1. High peak plasma concentrations (Cmax) .80 mg/ml were achieved with IV injection of the negative (TCOOH), neutral (TEGOH) and zwitterionic (TZwit) particles, while a 10-fold lower peak plasma concentration was achieved with the positive (TTMA) particles (Table 1). Following IV injection, high plasma clearance was observed for the negative (TCOOH) and positive (TTMA) charged AuNPs (0.0739 mg/ml/min and 0.170 mg/ml/ min, respectively) such that the plasma concentrations fell below 1 mg/ml within 15 minutes after injection ( Figure 1A). In contrast, plasma clearance for the neutral (TEGOH) and zwitterionic (TZwit) particles was substantially lower (0.00605 mg/ml/min and 0.00561 mg/ml/min, respectively) such that plasma concentrations remained above 1 mg/ml 24 hours  after injection ( Figure 1A). After IP injection, low concentrations (,1 mg/ml) of the negative (TCOOH) and positive (TTMA) AuNPs were detected in the circulation ( Figure 2B). In marked contrast, the neutral (TEGOH) and zwitterionic (TZwit) AuNPs rapidly entered into circulation, with peak concentrations above 10 mg/ml achieved 1.5 and 3 hours, respectively. As with the IV injection, plasma concentrations of the neutral (TEGOH) and zwitterionic (TZwit) AuNPs remained above 1 mg/ml 24 hours after injection ( Figure 2B) and bioavailability values for these AuNP were high (77% and 70%, respectively). From these results, it is apparent that nanoconjugates with neutral and zwitterionic properties maximize the circulation time.

Tumor Uptake of AuNPs is related to circulation resident time
Passive targeting of nanomaterials via the enhanced permeability and retention (EPR) effect is dependent on their blood resident time [26]. Our hypothesis was that the neutral and zwitterionic nanoconjugates would exhibit increased tumor uptake relative to negative and positively charged analogs. As expected, the nanoparticles with a long retention time in circulation (TEGOH and TZwit) accumulated more efficiently into the tumor ( Figure 3). After 24 hours post-injection (200 mg per mouse via tail vein or intraperitoneal), subcutaneously implanted ovarian tumor-bearing athymic nude mice were euthanized and the tumors were analyzed for gold content. The TZwit and TEGOH nanoparticles had substantially higher tumor uptake irrespective of the mode of administration. When injected into the peritoneum, a minute amount of the charged particles accumulated in the tumor, as expected from our pharmacokinetic studies. However, the negative nanoparticle TCOOH accumulated in the tumor after IV injection whereas the positive particle TTMA did not. These results are consistent with the pharmacokinetics of the IV administered nanoparticles. Thus, as a result of improved pharmacokinetics, neutral and zwitterionic particles demonstrated increased tumor accumulation with both routes of administration as compared to the other nanoparticles through the EPR (Enhanced Permeability and Retention) effect [27]. Our studies are in agreement with previous reports indicating that long plasma halflife can lead to increased tumor uptake [28], an important goal in imaging, diagnostics and therapeutics [29].

Ligand end group influences biodistribution of AuNPs
To determine the role of particle charge in regulating the biodistribution of these engineered nanomaterials, we compared the organ distribution of the AuNPs in immunocompetent mice (IV vs. IP, Table 2). After IV injection the nanoparticles were predominantly localized in the liver and spleen ( Figure 4B), with little particle found in the brain, kidneys, or lungs. In contrast, 24 hours after IP injection the concentration of gold was the highest in the pancreas ( Figure 4A) for all four particles. This localization may arise from intraperitoneal circulation and altered lymphatic clearance [30]; nanoparticles are removed from systemic circulation through the permeable vascular endothelium in lymph nodes [27,31]. The difference in localization demonstrates that the mode of administration of AuNPs affects the level of particle uptake in different tissues. Moreover, the levels of AuNPs found in the liver and spleen indicate that RES is the dominant mode of clearance for these particles.

Discussion
In recent years, inorganic nanoparticles have emerged as drug delivery systems [32], imaging agents [29] and diagnostic biosensors [33]. However, the biological fate and effect of nanoparticles in living systems needs to be defined in order to improve therapeutic engineering for in vivo applications, including  passively [34] and actively targeted vectors [35]. Previous work in this field has primarily focused on the biodistribution of inorganic nanoparticles based on their size. In our studies we maintained a constant particle size while varying the surface properties of the nanoparticle (see Figure S1, S2, and S3). Using pegylated gold nanoparticles, we investigated the pharmacokinetics, tumor uptake, and biodistribution of 2 nm diameter core/10 nm overall ( Figure 1) particles, focusing on the effect of surface charge on in vivo behavior. Blood sampling revealed that the neutral and zwitterionic particles TEGOH and TZwit AuNPs provided the highest concentrations and IP bioavailability. These concentrations were sustained over the 24-hour study period. The particle pharmacokinetics after IP injection was likewise strongly dependent upon particle surface charge. Plasma clearance was very low in comparison to the charged AuNPs ( Figure 2). This marked difference in bioavailability could be primarily due to opsonization of the nanoparticles with antibodies for recognition by resident macrophages. With the high concentration of phagocytes in the RES, opsonized particles are cleared out within minutes. Previous reports have also indicated that positively charged polymeric and hybrid nanoparticles accumulate in the liver and spleen in spite of PEG functionalization [36,37]. However, we show by modifying the nanoparticle surface charge, the pharmacokinetics of the engineered nanoparticles can be tuned. One of the major issues facing pharmacological agents is tissue selectivity [38]. Since the window of delivery of therapeutic agents is often limited it is imperative that penetration into highly vascular tumors is efficient. Accumulation in the tumor environment determines the therapeutic efficacy of antitumor therapeutics. However, this accumulation is diminished by non-specific tissue uptake and clearance by the RES. By utilizing the EPR effect, it is possible for long-circulating nanoparticles to reach and accumulate in the tumor [27]. In our studies, the more bioavailable TEGOH and TZwit nanoparticles exhibited a significantly tumor uptake over the TTMA particles after IV injection. Futhermore, with IV injections, TCOOH along with the TTMA and TEGOH nanoparticles had increased tumor uptake compared to the TTMA nanoparticle. Previous reports have also observed that a long plasma half-life can lead to increased tumor uptake [14], enabling biomedical applications such as imaging, diagnostic and therapeutics.
While all of the nanoparticles used in this study had a pronounced uptake in the liver following IP injection, it is interesting to note that when injected through the tail vein, TCOOH and TZwit uptake in the liver was considerable compared to the other nanoparticles. Previous reports have shown that b-2 glycoprotein and apolipoprotein B are known to bind to negatively-charged surfaces [39]. These proteins attach to foreign substances forming lipoproteins and transport them though the body via the lymphatic system or blood. Since both proteins are hepatic, it is possible that the TCOOH and TZwit nanoparticles are being sequestered to the liver due to the negative entity of their headgroup. Another interesting point is that IP injection resulted in high levels of particles in the pancreas. This localization may arise from intraperitoneal circulation and altered lymphatic Table 2. Biodistribution Study.
In conclusion, charge is a key determinant of the interactions of biological and synthetic materials with biosystems. We have quantified the role of surface charge on pharmacokinetics, tumor uptake and biodistribution using a structurally consistent family of neutral, zwitterionic, negative and positive gold nanoparticles (2 nm core, 10 nm overall diameter, see Figures S1 and S2). Neutral and zwitterionic particles provide high systemic exposure and low clearance when administered through intravenous administration and are rapidly absorbed in the circulation after intraperitoneal administration. Negative particles provide moderate systemic exposure while positive particles are rapidly cleared. Both positive and negative particles are poorly absorbed in the circulation after intraperitoneal administration, indicating the inability of these particles to cross the peritoneal barrier. Low plasma clearance for both administration routes is reflected in the increased tumor uptake of the neutral and zwitterionic nanoparticles in a subcutaneously implanted xenograft model of ovarian cancer. Biodistribution studies in different mouse strains (immunodeficient vs. immunocompetent) demonstrate that surface charge of the nanoparticles and their modes of systemic administration uniquely alter their pharmacokinetics, organ distribution and tumor uptake. The ability of surface charge to dictate bioavailability provides critical information to improve the design of nanotherapeutics for enhanced tumor uptake by both passive and active targeting strategies.

Gold nanoparticle synthesis and characterization
The Brust-Schiffrin two-phase synthesis method [40,41] was used for synthesis of AuNPs with core diameters around 2 nm ( Figure S1). After that, Murray place-exchange method was used to obtain functionalized AuNPs ( Figure S3) [22,42].

Animals
All experiments were done under protocols approved by the Mayo Clinic Institutional Animal Care and Use Committee (Protocol No A14108). Male HEJ/C3H mice, 4-6 weeks old and athymic mice (4-6 week old) were obtained from NCI Repository (Fredrick, MD). Male CD1 mice (,20 g) were received from Charles River.

Pharmacokinetics
AuNPs (100 mL of 40 mM) were administered to male CD1 mice (20 g) intravenously (IV) via the lateral tail vein and intraperitoneally (IP) in the right side of the stomach using a tuberculin syringe fitted with a 27-gauge needle. Blood samples were collected 5, 15, 30, 60, 90, 210, 480 and 1440 min postinjection. Mice were anesthetized under isoflurane vapors and blood samples collected by cardiac puncture using a 10% heparin in citrate phosphate dextrose solution anticoagulant (150 mL anticoagulant/ml whole blood), transferred to silanized amber microcentrifuge tubes and immediately chilled on ice. After separation by centrifugation (10,000 rpm 63 min at 4uC) plasma was transferred to silanized amber microcentrifuge tubes and immediately frozen. Samples were stored at 270uC until analysis. AuNP plasma concentration-time data were analyzed by standard noncompartmental methods using the program WinNonlin Pro (Pharsight Corp, Mountain View, CA).

Cell Culture
CP-70 cells were grown and maintained in RPMI medium supplemented with 10% fetal bovine serum and 1% antibioticantimycotic. Cultures were maintained at 37uC and 5% CO 2 atmosphere.

Tumor Xenograft
The animal use protocol was approved by the Mayo Clinic Animal Care and Use Committee (IACUC). CP-70 cells were lifted from tissue culture dishes using trypsin (16) and pelleted at 1500 rpm. Cells were resuspended in HBSS and counted using a hemocytometer. A 100 mL volume containing 2610 6 cells was injected subcutaneously on the right flank of 4-6 week old male athymic mice. Tumor growth was monitored until the mass reached 0.5 cm in length in any direction.

Biodistribution Study
The mice were randomized prior to nanoparticle injection. Mice were injected with 200 mg of AuNPs via IP or IV injections. After 24 hours, the mice were euthanized by CO 2 asphyxiation and tissue was collected for analysis. Total uptake of gold was analyzed using Instrumental Neutron Activation Analysis (INAA) [43].