Pharmacokinetics, Biodistribution, and Toxicity of Folic Acid-Coated Antiretroviral Nanoformulations

Abstract

The drug delivery platform for folic acid (FA)-coated nanoformulated ritonavir (RTV)-boosted atazanavir (FA-nanoATV/r) using poloxamer 407 was developed to enhance cell and tissue targeting for a range of antiretroviral drugs. Such formulations would serve to extend the drug half-life while improving the pharmacokinetic profile and biodistribution to reservoirs of human immunodeficiency virus (HIV) infection. To this end, we now report enhanced pharmacokinetics and drug biodistribution with limited local and systemic toxicities of this novel nanoformulation. The use of FA as a targeting ligand for nanoATV/r resulted in plasma and tissue drug concentrations up to 200-fold higher compared to equimolar doses of native drug. In addition, ATV and RTV concentrations in plasma from mice on a folate-deficient diet were up to 23-fold higher for mice administered FA-nanoATV/r than for mice on a normal diet. Compared to earlier nanoATV/r formulations, FA-nanoATV/r resulted in enhanced and sustained plasma and tissue ATV concentrations. In a drug interaction study, ATV plasma and tissue concentrations were up to 5-fold higher in mice treated with FA-nanoATV/r than in mice treated with FA-nanoATV alone. As observed in mice, enhanced and sustained plasma concentrations of ATV were observed in monkeys. NanoATV/r was associated with transient local inflammation at the site of injection. There were no systemic adverse reactions associated with up to 10 weeks of chronic exposure of mice or monkeys to FA-nanoATV/r.

INTRODUCTION

Remarkable progress was realized in the global effort to defeat human immunodeficiency virus (HIV) infection through effective antiretroviral therapy (ART). ART has transformed HIV disease into a long-term manageable disorder (1, 2). However, treatment failure and related drug resistance remain an issue with ART, which can contribute to suboptimal drug efficacy, variable pharmacokinetics, poor adherence to lifelong therapy, comorbid conditions, substance abuse, and inherent ART toxicities (3–5). These, we posit, can be addressed by developing long-acting and targeted ART.

Long-acting nanoformulated ART (nanoART) can achieve enhanced and sustained steady-state drug concentrations with infrequent dosing in mice (6–9). NanoART treatment resulted in marked improvements in the pharmacokinetic (PK) and pharmacodynamic (PD) profiles in normal and humanized mouse models (9–11). This effect is mediated by mononuclear phagocytes (MP; monocytes, macrophages, and dendritic cells), which can act as reservoirs and transporters of HIV-1 and could also potentially facilitate drug uptake, transport, and release of nanoART (8, 12–14). We have demonstrated that the improved PK profile of nanoART over those of unformulated drugs, at least in part, is the result of the sustained release of protease inhibitors from the site of drug injection and from macrophages present in tissues (9). However, our first nanoART formulation, using poloxamer 188 (P188)-encased drug, demonstrated shortcomings that may hamper clinical translation. These included high dose requirements (up to 250 mg/kg of body weight) and local toxicities, including those at the site of injection. To overcome these limitations, further efforts were made to develop novel nanoART formulations using specific macrophage-targeting ligands. Such ligands were placed onto the drug particles in order to enhance nanoART uptake and their carriage in macrophages. Macrophages express folic acid (FA) receptors on their surface, which are upregulated following immune activation (15, 16). As immune activation is seen during HIV-1 disease (17–19) and following nanoART-macrophage engagements (20), macrophage uptake of nanoART is enhanced by folic acid coating due to the induction and activation of folate receptor β on the surface of macrophages (P. Puligujja, S. Balkundi, L. Kendrick, H. Baldridge, J. Hilaire, A. N. Bade, P. K. Dash, G. Zhang, L. Poluektova, S. Gorantla, X. M. Liu, T. Ying, Y. Feng, Y. Wang, D. S. Dimitrov, J. M. McMillan, and H. E. Gendelman, submitted for publication). These studies demonstrated that nanoART prepared with FA-modified poloxamer 407 (P407) showed enhanced uptake and retention by macrophages as well as improved antiretroviral efficacy without altering cell viability (21). Preliminary PK and biodistribution data demonstrated up to 5-fold-enhanced plasma and tissue drug concentrations 14 days after a single dose of FA-coated nanoformulated ritonavir (RTV)-boosted atazanavir (FA-nanoATV/r) in mice (21). Such encouraging data prompted efforts to fully characterize the PK, biodistribution, and toxicity of FA-nanoATV/r in mice and in monkeys after intramuscular (i.m.) administration. For this promising therapeutic approach, use of relevant preclinical models becomes paramount to the development of effective products for human use. It was reported previously that high serum folate concentrations, like those found in the serum of laboratory rodents, could potentially block the binding of FA-drug conjugates to the folate receptors in tissues (22). Because of that, a mouse model treated with a folate-deficient diet was developed for the purpose of lowering endogenous mouse serum folate concentrations toward the normal human range (23). Therefore, in this study mice on both a folate-deficient diet and a normal diet were used. Results showed clear improvements in the PK profile over those of native (unformulated) drug and previous poloxamer (P188)-coated nanoART. These preclinical studies may further enable the development of FA-nanoATV/r for clinical intervention (24, 25). Such formulations would provide advantages in PK properties and patient compliance over what is now established by conventional native drug regimens (26).

MATERIALS AND METHODS

Chemicals.

Free-based RTV was obtained from Shengda Pharmaceutical Co. (Zhejiang, China). ATV-sulfate was purchased from Gyma Laboratories of America Inc. (Westbury, NY, USA) and free-based using triethylamine. Lopinavir (LPV) was purchased from Toronto Research Chemicals Inc. (North York, Ontario, Canada). High-performance liquid chromatography (HPLC)-grade methanol, acetonitrile, ammonium acetate, acetic acid, propylene glycol, and 1× phosphate-buffered saline (PBS) were obtained from Fisher Scientific (Fair Lawn, NJ, USA). Isoflurane was obtained from Halocarbon Product Corporation (River Edge, NJ, USA). Cremophor EL and poloxamer 407 (P407) were obtained from Sigma-Aldrich (St. Louis, MO, USA), and ethyl alcohol was obtained from Acros Organics (NJ, USA). Folate-modified P407 (FA-P407) was prepared as described previously (21).

Preparation and characterization of FA-nanoATV/r.

Folate-coated nanoformulated ATV and RTV (FA-nanoATV/r) were prepared with polymer excipients (60% P407/40% FA-P407) by high-pressure homogenization and lyophilized as described previously (21). Nanoformulations were characterized for size, surface charge, and polydispersity by dynamic light scattering using a Malvern Zetasizer Nano Series Nano-ZS system (Malvern Instruments, Westborough, MA, USA). Drug loading of lyophilized particles was determined by high-performance liquid chromatography (HPLC-UV) and by ultraperformance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) (27). The formulations were screened for cell uptake, retention, release, and antiretroviral activity using human monocyte-derived macrophages as described previously (21).

Mouse studies.

Eight-week-old male BALB/c mice were purchased from Charles River Laboratories (Wilmington, MA). One set of mice was maintained on a folate-deficient diet (Harlan Teklad TD.00434; Harlan Laboratories, Inc., Indianapolis, IN, USA) beginning 2 weeks prior to FA-nanoATV/r administration and throughout the study. Another set of mice was maintained on a normal sterilized 7012 Teklad diet (Harlan, Madison, WI). Food and water were provided ad libitum. Mice were housed in the University of Nebraska Medical Center (UNMC) laboratory animal facility according to Association for Assessment and Accreditation of Laboratory Animal Care guidance. All procedures were approved by the Institutional Animal Care and Use Committee at UNMC as set forth by the National Institutes of Health (NIH).

(i) Single-dose administration.

Single-dose PK studies were performed at 10 and 20 mg/kg of either native (unformulated) or FA-nanoATV/r. Injection solutions of FA-nanoATV/r were prepared from individual lyophilized ART formulations that were combined in the same suspension. Two groups of mice were dosed with FA-nanoATV/r, and two groups were dosed with native ATV/r. There were six mice in every group; three mice were sacrificed on day 14, and three mice were sacrificed on day 42. All mice were maintained on a folate-deficient diet starting 2 weeks prior to dose administration. FA-nanoATV/r doses were suspended in water, and native drug doses were suspended in a mixture of ethanol-Cremophor EL-propylene glycol-water (43/5/20/32, vol/vol). The injection volume was 40 μl for both FA-nanoATV/r and native drugs, and injection was by the intramuscular (i.m.) route. Blood samples were collected at days 1, 3, 7, 14, 28, and 42 for drug analysis. Tissue samples of liver, kidneys, spleen, lungs, brain, and muscle from the site of injection (∼100 mg) were collected on days 14 and 42.

(ii) Multiple-dose administration.

Multiple-dose pharmacokinetic studies were performed at a dose of 20 mg/kg FA-nanoATV/r or native drugs. The experimental design is summarized in Table 1. Four groups of mice (n = 10) were dosed with either FA-nanoATV/r or native drugs, two groups each. One group from each arm was maintained on a folate-deficient diet, and the other group was maintained on a normal diet. All groups were dosed by i.m. injection (40 μl) on days 0, 3, 7, 14, 21, and 28 (3 doses in the first week, followed by weekly dosing for another 3 weeks) over 4 weeks. Blood samples were collected on days 1, 3, 7, 8, 14, 15, 21, 22, 28, and 29. On day 35, half the mice in each group (n = 5) were sacrificed and tissues and blood samples were collected. The remaining mice were dosed biweekly on days 42 and 56. Blood samples were collected on days 42, 43, 49, 56, 57, and 63, and on day 70 mice were sacrificed and blood and tissue samples were collected. One additional group of mice (n = 5) maintained on a folate-deficient diet was dosed, as described above, on days 0, 3, 7, 14, 21, and 28 (3 doses in the first week, followed by weekly dosing for another 3 weeks), over a period of 4 weeks. Mice then received an additional dose on day 56 (monthly dosing group), and animals were sacrificed on day 70, when the tissue and blood samples were collected.

TABLE 1.

Experimental design of multiple-dose administration in mice

Formulation	Mouse diet	Dose schedule	Day(s) of procedure:
			Injection	Blood (plasma) collection (n = 5)	Tissue collection (n = 5)
FA-nanoATV/r	Folate deficient	Weekly	0, 3, 7, 14, 21, 28	1, 3, 7, 8, 14, 15, 21, 35	35
		Weekly and biweekly	0, 3, 7, 14, 21, 28, 42, 56	22, 28, 43, 49, 56, 57, 63, 70	70
		Weekly and monthly	0, 3, 7, 14, 21, 28, 56	29, 42, 49, 56, 57, 63, 70	70
	Normal	Weekly	0, 3, 7, 14, 21, 28	1, 3, 7, 8, 14, 15, 21, 28, 35	35
		Weekly and biweekly	0, 3, 7, 14, 21, 28, 42, 56	22, 29, 42, 43, 49, 56, 57, 63, 70	70
Native dose	Folate deficient	Weekly	0, 3, 7, 14, 21, 28	1, 3, 7, 8, 14, 15, 21, 28, 35	35
		Weekly and biweekly	0, 3, 7, 14, 21, 28, 42, 56	22, 29, 42, 43, 49, 56, 57, 63, 70	70
	Normal	Weekly	0, 3, 7, 14, 21, 28	1, 3, 7, 8, 14, 15, 21, 28, 35	35
		Weekly and biweekly	0, 3, 7, 14, 21, 28, 42, 56	22, 29, 42, 43, 49, 56, 57, 63, 70	70

(iii) Drug-drug interaction study.

Male C57BL/6 mice (7 weeks of age) were purchased from Taconic Farms (Hudson, NY). Animals were placed on a folate-deficient diet (Harlan Teklad TD.00434) for 2 weeks prior to drug treatment and maintained on the diet throughout the study. Mice were dosed by i.m. injection (40 μl) on days 0, 3, 7, 14, 21, and 28 with FA-nanoATV (20 or 50 mg/kg) with or without coadministration of an equal dose of FA-nanoRTV (n = 5 per treatment group). Blood samples were collected on days 1, 3, 7, 14, 21, and 28. On day 35, the mice were sacrificed and blood and tissue samples were collected for drug determinations.

Rhesus macaque study.

Rhesus macaques were purchased from PrimGen (Hines, IL) and tested negative for simian immunodeficiency virus (SIV), simian retrovirus (SRV) type D, and cercopithecine herpesvirus 1. All protocols and procedures were performed under approval of the Institutional Animal Care and Use Committee of UNMC according to NIH guidelines. Animals were anesthetized with 10 mg/kg of ketamine, administered intramuscularly, prior to experimental procedures and bleeding. Blood was drawn from the femoral vein, and plasma was obtained by centrifugation of EDTA-treated blood. PK studies were performed using FA-nanoATV/r at doses of 20 mg/kg ATV and 7 mg/kg RTV in animals weighing 5.0 ± 0.5 kg (n = 3). Prior to injection, 80 mg/ml FA-nanoATV and 27.8 mg/ml FA-nanoRTV were suspended together in PBS. FA-nanoATV/r was injected as an i.m. bolus. Three monkeys were dosed with FA-nanoATV on days 0, 3, 7, 14, 21, and 28 (3 doses in the first week, followed by weekly dosing for another 3 weeks), while FA-nanoRTV was coadministered to the same animals on days 3, 7, 14, 21, and 28. Blood samples were collected on days 1, 3, 7, 14, 21, 28, 35, 42, 56, 63, 70, 77, and 112 for plasma drug concentrations, metabolic panels, and complete blood count (CBC). Metabolic and hematologic panels were determined by the UNMC clinical pathology laboratory.

Blood and tissue collection.

For mice, blood samples (100 μl) were collected from the facial vein using a sterile 0.5-mm Goldenrod animal lancet (MEDIpoint, Inc., Mineola, NY). Blood was collected into plasma separator tubes (Microvette 300 LH; Sarstedt, Newton, NC). Plasma was separated by centrifugation of blood samples at 1,500 × g for 10 min at 4°C within 1 h of sample collection and stored at −80°C until analysis. Tissue samples were stored at −80°C until analyzed by LC-MS/MS. For monkeys, blood samples (2 ml) were collected from the femoral vein using a 21-gauge syringe, into an EDTA-treated tube. Plasma was separated by centrifugation at room temperature for 20 min at 900 × g and stored at −80°C until analysis by LC-MS/MS.

Sample preparation and analysis.

Plasma and tissue sample preparation and analyses were performed as described previously (27). Briefly, about 100 mg of tissues of interest was homogenized in deionized H₂O (1:4 [wt/vol]). One milliliter of ice-cold acetonitrile was added to 100 μl plasma or tissue homogenate samples prespiked with 10 μl internal standard (IS; 2.0 μg/ml lopinavir, 200 ng/ml final concentration). Samples were vortexed for 3 min, shaken continuously for 15 min, and centrifuged at 16,000 × g for 10 min. The supernatant was aspirated, evaporated under vacuum at room temperature, reconstituted in 100 μl of 50% methanol in H₂O, and sonicated for 5 min. After centrifugation at 16,000 × g for 10 min, 10 μl of each sample was used for LC-MS/MS analysis using a Waters Acquity UPLC (Waters, Milford, MA) coupled to an Applied Biosystems 4000 Qtrap quadrupole ion trap hybrid mass spectrometer (Applied Biosystems, Foster City, CA).

Pharmacokinetic analysis.

Mean plasma drug concentrations were calculated per treatment group for different doses. The pharmacokinetic parameters were derived using noncompartmental analysis of averaged plasma concentration-versus-time profiles, using WinNonlin Professional software (version 5.1). The elimination t_0.5 was obtained from the formula of 0.693/K. The area under the curve (AUC_0–∞) was estimated using the linear trapezoidal method from 0–t_last and extrapolation from t_last to infinity based on the observed concentration at the last time point divided by the terminal elimination rate constant (λ). The mean residence time (MRT) was calculated as AUMC_0–∞/AUC0–[r][inf]∞. Mean tissue concentrations were calculated and expressed as ng/g tissues.

Folate receptor expression.

Male BALB/cJ mice were maintained on a folate-deficient diet for 2 weeks and then treated with a 40-μl volume of PBS intramuscularly. Thirty-six hours later, mice were euthanized and tissues were collected and stored at −80°C. Folate receptor expression in tissues was determined by Western blotting assays. Tissues were dissociated by brief sonication in 3 volumes of ice-cold homogenization buffer (0.05 M Tris, pH 7.4, containing 1.15% KCl, 20% glycerol, and 10 μl/ml Halt protease inhibitor cocktail [Thermo Scientific Inc., Rockford, IL, USA]) and then centrifuged at 10,000 × g for 20 min at 4°C. The protein content of the supernatant was determined using the Pierce bicinchoninic acid (BCA) assay (Thermo Scientific). For folate receptor detection, 37.5 μg of supernatant protein was separated using a 4 to 12% NuPAGE Bis-Tris gel and transferred to a polyvinylidene fluoride membrane. The membrane was blocked with 5% powdered milk-5% bovine serum albumin (BSA) in phosphate-buffered saline–Tween 20 (PBST) and probed with anti-folate receptor antibody (1:200; Santa Cruz) followed by secondary antibody (horseradish peroxidase [HRP]-conjugated goat anti-rabbit IgG [1:5,000; Novex Life Technologies]). β-Actin was detected using rabbit anti-mouse anti-β-actin primary antibody (1:500; Abcam, Cambridge, MA, USA) and HRP-conjugated goat anti-rabbit IgG secondary antibody. All proteins were visualized using SuperSignal West Pico chemiluminescent substrate (Thermo Scientific). Folate receptor expression was normalized to β-actin using ImageJ software.

Toxicology.

Alanine aminotransferase, alkaline phosphatase, creatinine, blood urea nitrogen, total bilirubin, and total serum calcium and potassium were determined in mouse blood collected by cardiac puncture at 70 days after initial treatment using a VetScan comprehensive diagnostic profile disc (Abaxis Veterinary Diagnostics, Union City, CA) and a VetScan VS-2 instrument. For histopathological analysis of muscle at the site of injection, mice were treated with a single dose of FA-nanoATV/r (50 mg/kg in 40 μl) and muscle was collected 14 days later, placed in 10% neutral buffered formalin, and embedded in paraffin. Sections (5 μm thick) were cut and mounted on glass slides and then stained with hematoxylin and eosin. Muscle histopathological assessment was conducted in accordance with the guidelines of the Society of Toxicologic Pathology.

RESULTS

Single-dose administration.

ATV and RTV plasma concentration-versus-time profiles following i.m. administration of FA-nanoATV/r and native drug from the single-dose studies are shown in Fig. 1. By 3 days after native ATV/r single-dose administration at 10 and 20 mg/kg, plasma ATV concentrations decreased 60- to 85-fold, whereas by day 3 after single-dose FA-nanoATV/r administration, plasma ATV concentrations decreased by only 13- to 15-fold (Fig. 1A and B). For RTV, plasma concentrations decreased 10- to 15-fold by 3 days after both native and FA-nanoATV/r administration of 10 and 20 mg/kg (Fig. 1C and D). Native ATV was detected in the plasma until day 14 after 10-mg/kg and day 42 after 20-mg/kg native drug administration, whereas FA-nanoATV was detected in the plasma until day 42 at both doses. RTV plasma concentrations after both native and FA-nanoATV/r administration were detected until day 14 at 10-mg/kg and day 28 after 20-mg/kg dose administration.

FIG 1.

Plasma concentration-versus-time profiles after single-dose administration of FA-nanoATV/r and native administration of 10 mg/kg ATV (A), 20 mg/kg ATV (B), 10 mg/kg RTV (C), and 20 mg/kg RTV (D) in mice (n = 5, means ± standard errors of the means).

Pharmacokinetic parameters of FA-nanoATV/r and native ATV/r at 10- and 20-mg/kg single doses are shown in Table 2. The t_0.5 was 50% longer, while clearance was 1.6- to 4.5-fold lower, for FA-nanoATV than for native ATV. In addition, AUC and mean residence time (MRT) were 1.6- to 4.5-fold and 3-fold higher, respectively, for FA-nanoATV than for native ATV. For RTV, t_0.5 was either similar or shorter, while clearance was 3.3- to 6.8-fold lower, for FA-nanoATV/r than for native ATV/r. In addition, RTV AUC was 3.3- to 6.8-fold higher, while MRT was lower, for FA-nanoATV/r than for native ATV/r (Table 2).

TABLE 2.

Pharmacokinetic parameters of FA-nanoATV/r and native ATV and RTV after single 10- and 20-mg/kg dose administration in mice (n = 6)

Dose and formulation	PK parameter for drug:
	ATV					RTV
	t0.5 (h)	AUClast (ng · h/ml)	AUC0–∞ (ng · h/ml)	CLa (liter/h/kg)	MRT0–∞ (h)	t0.5 (h)	AUClast (ng · h/ml)	AUC0–∞ (ng · h/ml)	CL (liter/h/kg)	MRT0–∞ (h)
10 mg/kg
Native	128.4	5,839.6	5,922.9	1.7	38.3	105.2	880.8	926.3	10.8	76.4
FA-nanoformulation	197.6	26,365.2	26,650.3	0.4	107.4	54.4	6,288.0	6,311.5	1.6	37.5
20 mg/kg
Native	172.8	36,908.4	37,033.1	0.5	50.8	212.3	4,742.4	4,864.9	4.1	98.4
FA-nanoformulation	252.7	57,652.8	59,001.9	0.3	147.1	167.2	15,630.0	15,847.1	1.3	71.2

^aCL, clearance.

Multiple-dose administration.

Plasma concentration-versus-time values for ATV and RTV after multiple-dose administrations of FA-nanoATV/r or native ATV/r are shown in Fig. 2. Mice in the weekly-dosing group were administered three doses in the first week (days 0, 3, and 7) followed by weekly injections for an additional 3 weeks, i.e., a total of 4 weeks, and were sacrificed on day 35. For mice on the folate-deficient diet, plasma concentrations of ATV declined 12- to 33-fold and 9- to 14-fold, compared to day 1, by the end of each dosing interval (1 week) of the native and FA-nanoATV/r administration, respectively (Fig. 2A). In addition, ATV concentrations in plasma from mice on a folate-deficient diet 1 day after each dose administration were 1- to 2-fold higher, and by the end of each dosing interval 2- to 5-fold higher, after FA-nanoATV/r administration compared to those treated with equimolar doses of the native drugs. In contrast, ATV concentrations in mice on a normal diet declined only 1- to 3-fold by the end of each dosing interval, compared to day 1 values for both native and FA-nanoATV/r administration (Fig. 2A). In addition, ATV concentrations in plasma from mice on a normal diet 1 day after dose administration were only 1- to 3-fold higher and did not increase over time in mice treated with FA-nanoATV/r compared with those treated with native drug. ATV concentrations in plasma from mice on a folate-deficient diet, 1 day after each dose administration, were on average 14- to 23-fold higher for both FA-nanoATV/r and native drugs than for mice on a normal diet. This diet effect decreased over time, and by the end of each dosing interval, for the nanoformulated drug, ATV plasma concentrations were 2.5- to 3.5-fold higher in the folate-deficient mice than in those on a normal diet. These differences between the diet groups were not observed when native ATV/r was administered (Fig. 2A).

FIG 2.

Plasma concentration-versus-time profiles in mice on folate-deficient diet and normal diet. ATV (A) and RTV (B) after multiple-dose administration of FA-nanoATV/r and native drugs, at 20 mg/kg (n = 6, means ± standard errors of the means). ↓, days of dose administration.

Similarly to ATV, RTV concentrations in plasma from mice on a folate-deficient diet declined 34- to 68-fold, and 14- to 25-fold, compared to day 1, by the end of each dosing interval of the native and FA-nanoATV/r administration, respectively (Fig. 2B). In addition, RTV concentrations in plasma from mice on a folate-deficient diet, 1 day after each dose administration, were 1.5- to 4-fold higher, and by the end of each dosing interval 4- to 8-fold higher, after FA-nanoATV/r administration than after administration of the native drugs. In contrast, concentrations of RTV in mice on a normal diet declined only 14- to 34-fold and 7- to 10-fold by the end of each dosing interval of the native and nanoformulated drug administrations, respectively (Fig. 2B). In addition, RTV concentrations in plasma from mice on a normal diet were only 1- to 3-fold higher 1 day after dose administration and did not increase over time in mice treated with FA-nanoATV/r compared with those treated with native drugs.

The biweekly-dose group mice were dosed weekly until day 28 and then biweekly on days 42 and 56 and were sacrificed on day 70. Compared to day 1, plasma concentrations of ATV in mice on the folate-deficient diet declined 45- to 75-fold and 19- to 22-fold by the end of each dosing interval (2 weeks) of the native and FA-nanoATV/r administration, respectively (Fig. 2A). In addition, ATV concentrations in plasma from mice on the folate-deficient diet were 1.5- to 5-fold higher throughout the experiment after FA-nanoATV/r administration than after administration of the native drugs. In contrast, concentrations of ATV in mice on a normal diet declined only 3- to 4.5-fold by the end of each dosing interval for both the native and FA-nanoATV/r administrations (Fig. 2A). In addition, ATV concentrations in plasma from mice on the normal diet were 1- to 2-fold higher throughout the experiment in mice treated with FA-nanoATV/r than in mice treated with native drug. Similarly, RTV concentrations in plasma from mice on the folate-deficient diet declined 66- to 232-fold and 119- to 207-fold by the end of each dosing interval of the native and FA-nanoATV/r administrations, respectively (Fig. 2B). In contrast, concentrations of RTV in mice on the normal diet declined only 56- to 497-fold and 33- to 97-fold by the end of each dosing interval of the native and nanoformulated drug administrations, respectively (Fig. 2B).

Tissue concentrations of ATV and RTV, after multiple doses of FA-nanoATV/r and native drug, are shown in Table 3. ATV and RTV concentrations in tissues from mice on the folate-deficient diet were 4- to 6-fold and 10- to 200-fold higher, respectively, after FA-nanoATV/r administration than after treatment with equimolar doses of the native drug (Table 3). In contrast, both ATV and RTV concentrations in tissues from mice on a normal diet were 1- to 3-fold higher after FA-nanoATV/r administration than after treatment with equimolar doses of the native drug (data not shown). In addition, ATV concentrations in tissues from mice on a folate-deficient diet were 2- to 15-fold and 1.5- to 3-fold higher for FA-nanoATV/r and native drug, respectively, compared with mice on the normal diet. Similarly, RTV concentrations in tissues from mice on a folate-deficient diet were 1.5- to 4.5-fold higher for FA-nanoATV/r compared with normal mice and there was no effect for native RTV.

TABLE 3.

Tissue concentrations of ATV and RTV, after multiple doses of FA-nanoATV/r and native-drug administration at 20 mg/kg in mice (n = 5)

Tissue	Sample collectiond	Mean concn (ng/g or μg/ga) in tissue ± SEM
		FA-nanoATV	Native ATV	FA-nanoRTV	Native RTV
Liver	Day 35, special dietb	1,314.2 ± 121.7	219.6 ± 35.1	108.9 ± 19.2	0.5 ± 0.3
	Day 70, biweekly special dietc	260.1 ± 40.6	55.1 ± 11.6	6.9 ± 2.3	0.5 ± 0.3
	Day 35, normal diet	87.7 ± 7.3	72.3 ± 10.7	44.1 ± 4.2	13.4 ± 5.8
	Day 70, biweekly normal diet	49.0 ± 14.5	18.2 ± 3.1	9.9 ± 3.6	6.5 ± 6.5
Spleen	Day 35, special diet	62.9 ± 11.2	15.2 ± 3.0	75 ± 17.6	4.2 ± 1.0
	Day 70, biweekly special diet	9.1 ± 0.4	4.5 ± 0.5	4.3 ± 1.1	2.1 ± 0.3
	Day 35, normal diet	11.6 ± 0.7	10.6 ± 0.9	33.2 ± 3.1	14.4 ± 6.3
	Day 70, biweekly normal diet	5.8 ± 1.0	2.2 ± 0.2	5.9 ± 1.4	5.3 ± 4.1
Kidney	Day 35, special diet	126.5 ± 14.4	31.5 ± 2.8	103.2 ± 11.5	3.7 ± 0.6
	Day 70, biweekly special diet	33.0 ± 4.1	11 ± 1.3	9.7 ± 3.0	3.2 ± 0.9
	Day 35, normal diet	17.5 ± 1.7	23.2 ± 7.3	69.4 ± 7.1	21.3 ± 7.5
	Day 70, biweekly normal diet	7.6 ± 0.9	5.9 ± 0.5	10.8 ± 3.2	8.1 ± 6.5
Lung	Day 35, special diet	103.2 ± 35.7	28 ± 8.8	31.3 ± 10.5	2.2 ± 1.0
	Day 70, biweekly special diet	21.7 ± 2.7	6.9 ± 1.2	2.3 ± 0.7	0.9 ± 0.2
	Day 35, normal diet	15.7 ± 2.3	13.8 ± 2.2	15.9 ± 2.3	4.9 ± 2.0
	Day 70, biweekly normal diet	65.8 ± 47.5	3.1 ± 0.4	2.4 ± 0.7	1.8 ± 1.3
Lymph node	Day 35, special diet	36.1 ± 3.5	21.8 ± 2.7	23.6 ± 7.2	4.3 ± 1.5
	Day 70, biweekly special diet	26.0 ± 1.7	20.8 ± 11.6	1.9 ± 0.7	1.2 ± 0.3
	Day 35, normal diet	17.8 ± 3.6	16.6 ± 11.1	25.4 ± 8.7	9.1 ± 6.4
	Day 70, biweekly normal diet	21.8 ± 12.9	19.5 ± 8.7	10 ± 6.3	3.1 ± 1.1
Brain	Day 35, special diet	2.78 ± 0.75	2.15 ± 0.5	0.8 ± 0.27	—e
	Day 70, biweekly special diet	1.14 ± 0.26	—	—	—
	Day 35, normal diet	0.29 ± 0.07	0.68 ±	0.4 ± 0.08	—
	Day 70, biweekly normal diet	—	—	—	—
Site of injection	Day 35, special diet	1,145.0 ± 127.8	454.4 ± 32.7	51.0 ± 17.7	9.0 ± 0.5
	Day 70, biweekly special diet	345.8 ± 69.4	563.3 ± 191.4	0.1 ± 0.0	11.6 ± 3.4
	Day 35, normal diet	847.0 ± 278.0	286 ± 50.5	11.8 ± 5.9	5.4 ± 0.7
	Day 70, biweekly normal diet	918.5 ± 161	302.6 ± 123.5	6.9 ± 5.9	4.8 ± 2.1

^aValues for site of injection are in micrograms per gram.

^bTissue samples were collected on day 35, i.e., 7 days after the last dose in the weekly-dosing group.

^cTissue samples were collected on day 70, i.e., 14 days after the last dose in the biweekly-dosing group.

^dSpecial diet is the folate-deficient diet.

^e—, values are less than the limit of quantitation (i.e., 0.5 ng/ml).

Drug-drug interaction study.

Low-dose RTV is included in combination ART treatments to inhibit CYP3A4-mediated metabolism of other antiretrovirals such as ATV (28). To determine whether FA-nanoRTV coadministration would indeed affect plasma and tissue FA-nanoATV concentrations, C57BL/6 mice were treated i.m. with FA-nanoATV (20 or 50 mg/kg) with or without the coadministration of equal doses of FA-nanoRTV on days 0, 3, 7, 14, 21, and 28. Plasma samples were collected at various time points, mice were sacrificed 1 week after the last dose, and tissue drug concentrations were determined (Fig. 3). Plasma and tissue ATV concentrations were 2- to 5-fold higher at all time points in mice treated with 50 mg/kg FA-nanoATV/r than in mice treated with the same dose of FA-nanoATV alone. Little to no difference in plasma ATV concentration was observed in mice treated with 20-mg/kg FA-nanoATV/r in comparison to FA-nanoATV alone; however, liver and spleen ATV concentrations were 4.7- and 2.7-fold greater in FA-nanoATV/r-treated mice (Fig. 3).

FIG 3.

Plasma concentration-time profile and tissue concentrations of ATV in wild-type mice administered intramuscular FA-nanoATV/RTV and FA-nanoATV alone at 20 or 50 mg/kg (n = 5, means ± standard errors of the means). ↓, days of dose administration.

Folate receptor distribution.

Expression of folate receptors (α, β, and γ) was determined in spleen, liver, kidney, and lung from BALB/cJ mice treated with PBS. As shown in Fig. 4, folate receptors were detected in all tissues, with the highest concentrations in kidney and liver when expression was normalized to β-actin. Expression of folate receptors was not significantly altered by treatment with FA-nanoATV/r.

FIG 4.

Folate receptor (FOLR) expression in tissues. (A) FOLR was determined by Western blotting in tissue homogenates from control mice, on the folate-deficient diet, treated with PBS 36 h prior to collection of tissues. (B) Receptor expression was normalized to β-actin levels.

Toxicology.

At the end of the 70-day multiple-dose PK study, a panel of serum metabolites and enzymes was examined and all values were within the normal range, indicating no long-term adverse reactions to the multiple-dose FA-nanoATV/r administration in mice (Table 4). Histopathology at the site of i.m. injection showed focal areas of macrophages, but no fibrosis, 14 days after a single dose of 50-mg/kg FA-nanoATV/r (Fig. 5).

TABLE 4.

Serum chemistry profiles in mice treated with FA-nanoATV/r (20 mg/kg, intramuscularly) weekly and biweekly for 70 days (n = 5)

Serum component (U)a	Normal rangeb	Mean ± SEM for FA-nanoATV/r
AMY (U/liter)	483–3,240	841.0 ± 100.4
ALT (U/liter)	17–77	34.9 ± 8.3
ALP (U/liter)	35–222	86.7 ± 17.9
Total bilirubin (phosphate units in mg/dl)	0.0–0.9	1.0 ± 0.1
Glucose (phosphate units in mg/dl)	140–263	203.6 ± 35.5
Total protein (g/dl)	3.9–6.4	6.4 ± 1.2
Creatinine (phosphate units in mg/dl)	0.20–0.90	0.8 ± 0.2
Calcium (phosphate units in mg/dl)	6.0–13.0	11.6 ± 1.7
BUN (phosphate units in mg/dl)	9–33	14.6 ± 2.1
Albumin (g/dl)	2.5–4.6	3.0 ± 1.0
Sodium (mmol/liter)	110–195	126.3 ± 3.9
Potassium (mmol/liter)	4.0–10.5	6.5 ± 0.4
Serum globulin (g/dl)		1.9 ± 1.1

^aAbbreviations: AMY, amylase; ALT, alanine aminotransferase; ALP, alkaline phosphatase; BUN, blood urea nitrogen.

^bValues obtained from Jackson Labs (http://phenome.jax.org/db/q?rtn=projects/details&sym=Eumorphia1&mlistmode=strainvalues&reqstrainid=5).

FIG 5.

Histopathology at the site of i.m. injection. Mice were given a single injection of PBS or 50 mg/kg FA-nanoATV/r. Muscle was collected from the site of injection 14 days later and stained with hematoxylin and eosin. Arrowheads indicate a focal area of macrophage infiltration (magnification, ×200). NanoATV/r injection elicited an inflammatory response at the site of injection without formation of granuloma or fibrosis. This consisted of increased infiltration of blood leukocytes; the majority of cells, based on morphology and size, were mononuclear phagocytes, with a smaller number of lymphocytes. Normal muscle morphology was observed at the site of injection in PBS-treated animals.

Rhesus macaque study.

Monkeys were administered three FA-nanoATV/r or native drug doses in the first week (days 0, 3, and 7) followed by weekly injections for an additional 3 weeks, i.e., a total of 4 weeks, and blood samples were collected for 112 days. ATV and RTV plasma concentration-versus-time profiles in monkeys are shown in Fig. 6. Plasma concentrations of ATV were nearly constant starting at day 7 and through day 35 and then gradually decreased after the last dose on day 28 until day 112. In contrast, RTV plasma concentrations gradually decreased throughout the study and were undetectable after day 49. There was no observed adverse reaction at the site of injection, and metabolic panels and complete blood count (CBC) tests were within the normal range (see Fig. S1 and Table S1 in the supplemental material).

FIG 6.

Plasma concentration-versus-time profiles after multiple-dose administration of FA-nanoATV/r for 20 mg/kg ATV (A) and 7 mg/kg RTV (B) in monkeys (n = 3, means ± standard errors of the means). ↓, days of dose administration.

DISCUSSION

We previously demonstrated enhanced and sustained PK of ATV/r in mice and monkeys with untargeted nanoformulations (9). While a first step toward the development of long-acting antiretrovirals, there were several shortcomings. First and foremost were the high dose requirements of up to 250 mg/kg to achieve therapeutically effective plasma concentrations. Second were the local toxicities at the site of injection. Indeed, granuloma formation was observed at the site of subcutaneous (s.c.) injection for earlier nanoATV/r formulations (11). Thus, further development was needed toward product translation. To this end, a new formulation of ART loaded into nanoparticles made from FA-coated polymer (P407) that specifically targeted monocytes/macrophages was sought. This novel nanoformulation resulted in encouraging in vitro and in vivo anti-HIV responses (21), which prompted efforts to fully characterize its pharmacokinetics and biodistribution in vivo. To such ends, we found that FA-nanoATV/r resulted in enhanced and sustained plasma and tissue concentrations of ATV and at lower doses compared to older untargeted formulations. ATV plasma concentrations were 3- to 8-fold higher 7 days after every dose, and tissue concentrations were 2.5- to 6-fold higher on day 35 (7 days after last dose of weekly-dose administration) of FA-nanoATV/r in i.m. 20-mg/kg PK studies, compared to 2.5-fold (50 mg/kg) the dose of subcutaneous (s.c.) untargeted nanoATV/r (9). In contrast, FA-nanoATV/r did not demonstrate substantive improvements in the pharmacokinetic profile of RTV over its untargeted counterpart. This may be due to differences in drug crystal configuration, ligand binding, or drug bioavailabilities. Current research activities in our laboratories are designed to uncover such limitations.

Plasma concentrations of ATV after a single dose at 10 and 20 mg/kg were up to 20-fold higher in mice treated with FA-nanoATV/r than in those treated with equimolar doses of the native ATV/r. Similarly, RTV plasma concentrations were up to 8.4-fold higher in mice treated with FA-nanoATV/r than in those treated with native drug (Fig. 1). The FA-nanoATV/r effect on ATV was also demonstrated through the calculation of PK parameters, where t_0.5, MRT, and AUC values were 1.5- to 4.5-fold higher and clearance was 1.5- to 4.5-fold lower for FA-nanoATV compared to native ATV. Moreover, RTV AUC increased by 3- to 7-fold and clearance decreased by 3- to 7-fold, while no improvement was observed in t_0.5 and MRT values (9). Similarly to plasma drug concentrations, FA-nanoATV/r also resulted in markedly higher tissue concentrations, up to 25-fold, of both ATV and RTV (data not shown). Such data, taken together, underscore the improvements in PK through such targeted nanoformulation schemes in drug delivery.

In the multiple-dose PK study, ATV and RTV concentrations in plasma from mice on the folate-deficient diet, 1 day after each dose administration, were 1.5- to 4-fold higher after FA-nanoATV/r administration compared to concentrations from those treated with equimolar doses of the native drug. This FA-nanoATV/r effect increased over time, and by the end of each dosing interval, ATV and RTV plasma concentrations were up to 2- to 8-fold higher after administration of FA-nanoATV/r compared to native drugs. The FA-nanoATV/r effect on ATV and RTV was less in the normal-diet group, where plasma concentrations were only 1- to 3-fold higher after FA-nanoATV/r administration compared to native drug (Fig. 2).

BALB/c mice have serum folate concentrations of approximately 250 nM, compared to 25 nM in humans (23). Therefore, mice were kept on a folate-deficient diet for 2 weeks prior to any treatments to bring down their serum folate to concentrations comparable to those in humans. The biological significance of lowering serum folate concentrations through dietary modulation was established previously (22, 23), where steady-state human equivalent serum and red blood cell (RBC) folate concentrations were observed in mice on a folate-deficient diet for up to 18 weeks (23). Reduced serum folate concentrations will decrease any competition between FA-nanoATV/r and free folate to bind to the folate receptor. We observed that the folate-deficient diet itself increased plasma concentrations of both native and FA-nanoATV/r. ATV plasma concentrations in mice on a folate-deficient diet were increased 14- to 23-fold 1 day after dose administration for both FA-nanoATV/r and native drug compared to mice on a normal diet. This diet effect diminished over time, where ATV plasma concentrations were only 2.5- to 3.5-fold higher, and RTV concentrations were not different, in the folate-deficient-diet mice compared to the normal-diet mice by day 7 of every dose (Fig. 2). Similar to the PK analysis of the single dose, in multiple-dose PK analysis t_0.5, MRT, and AUC values were 1.3- to 1.6-fold higher, and clearance was 1.6-fold lower, for FA-nanoATV than for native ATV. In addition, RTV PK parameter AUC increased by 4-fold, and clearance decreased by 4-fold, while no improvement was observed in t_0.5 and MRT values (data not shown).

ATV and RTV concentrations in tissues from mice on a folate-deficient diet were up to 200-fold higher after FA-nanoATV/r administration compared to the native drug (Table 3). The FA-nanoATV/r effect on enhancing tissue concentrations was weakened in the normal-diet group, where both ATV and RTV concentrations were only up to 3-fold higher in mice receiving FA-nanoATV/r compared to native drug. Similar to plasma, the folate-deficient diet itself increased tissue concentrations of both native and FA-nanoATV/r. ATV concentrations in tissues from mice on the folate-deficient diet were up to 15-fold higher for FA-nanoATV/r, and only up to 3-fold higher for native drug, compared to mice on the normal diet. Similarly, RTV concentrations in tissues from mice on the folate-deficient diet were up to 4.5-fold higher than normal-diet mice in the FA-nanoATV/r groups, while no diet effect existed in the native groups (Fig. 2) One important limitation of antiretroviral therapy (ART) is the lack of penetration into viral reservoirs (29); however, our targeted nanoformulation therapy demonstrates improved penetration of protease inhibitors into HIV reservoirs such as lymph nodes and spleen. Enhanced delivery of anti-HIV drugs into these latent reservoirs will increase the chances of eradicating the virus from these hidden sanctuaries. In this study, for example, we observed up to 5.4-fold-higher concentrations of FA-nanoATV/r than of native ATV/r in lymph nodes, which are known latent reservoirs for HIV.

Prior studies of FA-nanoATV/r showed a significant increase in uptake (>2-fold) compared to the nontargeted nanoformulations through binding to folate receptor (FOLR2) on macrophages. In addition, in recent studies (21) FA-nanoATV/r showed greater antiretroviral activity than nontargeted nanoATV/r as shown by HIV-1p24 staining and reverse transcriptase (RT) activity. Also, plasma and tissue concentrations were up to 5-fold higher 14 days after a single dose of FA-coated nanoformulated ATV/r (FA-nanoATV/r) compared to uncoated nanoATV/r in mice (21). The advantages of targeted delivery of FA-nanoATV/r were supported in this current PK study. We detected expression of folate receptors in liver, kidney, lung, and spleen, with the highest concentrations in kidney and the lowest in spleen (Fig. 4). It has been previously reported that kidney has the highest concentration of folate receptor expression (on tubular epithelial cells), and the liver has been identified as the storage site for folate and shows very high receptor-mediated uptake of folate (23). Therefore, the tissue targeting that we are observing with nanoART could be related to the high expression of folate receptor in all these tissues and the liver in particular. The implications of these findings rest in the idea that macrophages can serve as reservoirs and transporters for FA-nanoATV/r.

The primary role of RTV in combinational ART regimens is to improve the pharmacokinetics of other drugs such as RTV via CYP3A4 inhibition (28). It was also reported that RTV inhibits CYP3A and drug transporters in mice, which increases the bioavailability of other drugs (30). This study was performed to investigate the effect of FA-nanoRTV on the pharmacokinetics of FA-nanoATV. A boosting effect of RTV was observed in this study: ATV plasma and tissue concentrations were up to 5-fold higher in mice treated with FA-nanoATV/r than in mice treated with FA-nanoATV alone (Fig. 3). However, the RTV boosting effect on ATV was not observed in monkeys (data not shown).

Perhaps even more importantly, the improved PK profile of ATV/r was not associated with any signs of toxicity after chronic exposure of mice and monkeys to FA-nanoATV/r for up to 10 weeks. Serum chemistry and metabolic panels were within the normal ranges in both mice and monkeys, indicating the lack of systemic toxicities (Table 4; see also Fig. S1 and Table S1 in the supplemental material). Furthermore, histopathology examinations of the sites of repeated injections indicated the lack of fibrosis in mice (Fig. 5). NanoATV/r injection elicited an inflammatory response at the site of injection without formation of granuloma or fibrosis. This consisted of increased infiltration of blood leukocytes; the majority of cells, based on morphology and size, were mononuclear phagocytes, with a lower number of lymphocytes. Normal muscle morphology was observed at the site of injection in PBS-treated animals.

Similarly to mice, the enhanced and sustained plasma accumulation of ATV as well as the lack of systemic and local toxicities associated with FA-nanoATV/r was observed in monkeys. In monkeys, ATV plasma concentrations were detected up to 112 days after the last dose administration on day 28 (Fig. 6).

In summary, folic acid-coated nanoformulation (FA-nanoATV/r) showed sustained and enhanced pharmacokinetics and drug distribution with limited local and systemic toxicities in both mice and monkeys. The use of FA as a targeting ligand for nanoATV/r resulted in enhanced and sustained plasma and tissue concentrations compared to equimolar doses of native drugs. In addition, compared to previous untargeted nanoATV/r formulations, FA-nanoATV/r improved the PK and toxicity profiles, which allowed about a 5-fold reduction in dose requirements.