| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Reproductive Technology |
Drug Discovery Program, Department of Reproductive Biology,3
Pharmaceutical Sciences,4
Virology,5 Parker Hughes Institute, St. Paul, Minnesota 55113
Paradigm Pharmaceuticals, LLC,6 St. Paul, Minnesota 55113
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
AIDS/HIV, assisted reproductive technology, female reproductive tract, intravaginal, microbicide, nonnucleoside inhibitor, rabbit sperm, sperm motility and transport, vagina
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Semen is an important vehicle for cell-associated transmucosal transmission of HIV-1 [5]. Microbicides could provide protection by directly inactivating HIV or preventing HIV from attaching, entering, or replicating in susceptible target cells as well as dissemination from target cells present in semen or the host cells that line the vaginal or rectal wall. Microbicides that are currently being investigated are directed mainly at preventing pregnancy as well as protection against HIV [6–9]. The availability of a nonspermicidal cell permeable anti-HIV microbicide is equally important for 1) sexually active women to allow pregnancy while protecting both mother and child from HIV infection, and 2) for HIV-1 serodiscordant couples who want to have a child by assisted reproductive technologies (ART). Candidate anti-HIV microbicides should have broad-spectrum activity since half of all HIV-infected patients in the United States harbor drug-resistant mutants [10, 11]. In addition, they should not affect sperm function in a way that reduces the chances of conception or pose a health risk for the female partner by producing local inflammation.
The nonnucleoside reverse transcriptase (RT) inhibitors (NNRTIs) of HIV-1 are an inherent ingredient of the drug combination strategies currently used in the treatment of HIV-1 infections [12–15]. Numerous classes of compounds have been described as NNRTIs [16, 17]. However, only three compounds have so far been approved for clinical use: nevirapine, delavirdine, and efavirenz. NNRTIs are notorious for rapidly leading to virus–drug resistance development, primarily based on the emergence of the K103N and Y181C mutations in the HIV-1 RT [18, 19]. NNRTI bind to an allosteric site of HIV-1 RT [20, 21], which is 
10 Å away from the catalytic site [22]. NNRTI binding induces rotamer conformation changes in some residues (Y181 and Y188) and makes the "thumb region" more rigid. Both events alter the substrate binding mode and/or affect the translocation of the double strand, which are critical for the polymerase function, thereby leading to a noncompetitive inhibition of the enzyme. Most mutations conferring resistance to NNRTI are directly in contact with the NNRTI molecule, and thus are associated with changes in the binding of NNRTI to RT [23]. Dozens of mutant strains have been characterized as resistant to NNRTIs, including L1001, K103N, V106A, E138K, Y188I/C, and Y188H. The mutations of these residues lead to the weakening of the inhibitor binding to RT [23, 24].
Our modeling studies for rational drug design involving the construction of a composite NNRTI binding pocket that was generated using high-resolution crystal structure information from nine individual RT-NNI complexes revealed previously unrecognized ligand-derivatization sites for phenethylthiazolylthiourea (PETT) derivatives [25, 26]. Y181C, Y188C, and Y188H mutations in drug-resistant HIV strains were predicted to result in larger unoccupied volume in the binding pocket and a different interaction environment [23]. We hypothesized that preferred thiourea inhibitors should maximize the occupancy in the NNRTI binding site of HIV-1 RT. We used the composite NNRTI binding pocket model to design potent NNRTIs against wild type RT and drug-resistant RT mutants [25–28]. Molecular modeling and score functions were used to analyze how drug-resistant mutations would change the RT binding pocket shape, volume, and chemical make-up, and how these changes could affect NNRTI binding [29]. The results of these studies led to the synthesis of a series of thiophene thiourea compounds that inhibited the replication of the HIV-1 strains HTLVIIIB (NNRTI-sensitive); A17 (NNRTI-resistant, Y181C mutant RT); and A17-Variant (NNRTI-resistant, Y181C + K103N mutant RT) in human peripheral blood mononuclear cells (PBMCs) at nano/submicromolar concentrations [30]. The lead compound, PHI-443 (N'-[2-(2-thiophenylethyl)]-N'-[2-(5-bromopyridyl)]-thiourea), was substantially more potent than the standard NNRTI drugs against the NNRTI-resistant HIV-1 strains harboring Y181C, K103N, or V106A mutations. The documented in vitro potency of PHI-443 against drug-resistant primary clinical HIV-1 isolates is particularly promising for its development as a prophylactic anti-HIV microbicide.
Here, we describe the results of a study that examined the potential utility of PHI-443 as a nonspermicidal microbicide for prevention of sexual transmission of HIV. Our goal in this study was to test the effects of PHI-443 on in vivo sperm functions under conditions shown to inactivate HIV-1 in human cells. The microbicidal activity of PHI-443 was evaluated in the humanized severe combined immunodeficient (Hu-SCID) mouse model of vaginally transmitted AIDS using a genotypically and phenotypically drug-resistant HIV-1 isolate. PHI-443 formulated via a self-emulsifying gel was tested for potential in vivo mucosal toxicity and for lack of contraceptive activity in the rabbit model. The results obtained imply that PHI-443 has clinical potential as a prophylactic broad-spectrum anti-HIV microbicide without contraceptive activity.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The following excipients were used to test the solubility profile of PHI-443: Avicel (FMC Corp., Newark, DE); Captex (Abitec Corporation, Janesville, WI); carboxy methyl cellulose and oleic acid (Spectrum Chemical Manufacturing Co., Gardena, CA); Cremophor EL (Badische Anilin and Soda-Fabrik AG [BASF] Corp., Mount Olive, NJ); Gelucire 50/13, Labrafil M, and Labrasol (Gattefosse, Saint Prest, Cedex, France); hydroxy stearate polyethylene glycol 660 (Solutol HS; BASF, Mount Olive, NJ); polyethylene glycol (Union Carbide Chemical and Plastic Co., Inc., Danbury, CT); polysorbate 80 (Croda Inc., Parsippany, NY); Rhodigel 200 (R.T. Vanderbilt Co., Inc., Norwak, CT); sodium benzoate (Cultor Food Science, Ardsley, NY); sorbitol (70% solution; Spectrum); vitamin E and vitamin E DL-TPGS (
-tocopheryl PEG 1000 succinate; Peebock Division of Eastman Company UK Ltd., Llangefni, Anglesey, UK); and Xantural 75 (Pharmaceutical Ingredients, Norristown, PA). Deionized distilled water was purified via the Millipore Milli-Q purification system (Medford, MA).
For high-performance liquid chromatographic (HPLC) analysis of PHI-443, the chemicals used were of analytical reagent grade. HPLC-grade solvents were purchased from Burdick and Jackson (Muskegon, MI). All other chemicals were purchased from Aldrich (Milwaukee, WI); Sigma Chemical Co. (St. Louis, MO); or Fisher Scientific (Pittsburgh, PA) and used without further purification.
Synthesis of PHI-443
PHI-443 (N'-[2-(2-thiophenylethyl)]-N'-[2-(5-bromopyridyl)]-thiourea), a novel heterocyclic thiourea NNRTI (Fig. 1) was synthesized by condensing 2-amino-5-bromopyridine with 1,1-thiocarbonyl di-imidazole to furnish the precursor thiocarbonyl derivative. Further reaction of the product with 2-thiophene ethylamine in dimethyl formamide yielded the target compound in good yields. The product was purified by recrystallization and column chromatography using silica gel 60 (EM Sciences, Gibbstown, NJ) using n-hexane-ethyl acetate as eluents. Purity was determined by proton and carbon nuclear magnetic resonance spectra, infrared and ultraviolet-visible spectra, and mass spectroscopy. Elemental analysis was performed by Atlantic Microlabs (Norcross, GA). The final product with a purity of >99% was used for the preclinical studies.
|
Gel Formulations of PHI-443
A comprehensive preformulation study of PHI-443 was conducted to develop a suspension formulation. Based on the solubility profile of PHI-443 in water, ethanol, oleic acid, 10% vitamin E, 10% Cremophor, Labrasol, and Gelucire, several formulations were prepared. Various excipients of Cremophor EL, Phospholipon 90G, polyethylene glycols (PEG), propylene glycol, Captex 300, Labrasol, vitamin E TPGS, Cremophor RH 60, Gelucire 44/14, Imwitor 742, Polysorbate 80, Labrafilm M2125, oleic acid, and Cremophor RH 35 and combinations thereof were investigated for achieving acceptable solubility of PHI-443 without apparent spermicidal activity. Based on the extensive preformulation studies using molecular complexation with cyclodextrin, microemulsion systems, and use of surfactants and cosurfactants, a self-emulsifying formulation of PHI-443 was formulated. A lipophilic surfactant with hydrophile-lipophile balance values of <16 such as polysorbate 80 was used to promote emulsification and rapid dispersibility. Good rheological properties were achieved by the use of xanthan gum, and microcrystalline cellulose with aqueous polyethylene glycol-sorbitol mixture was used as a carrier.
For determining the amount of PHI-443 in each formulation, a known weight of PHI-443–containing gel was mixed with acetonitrile and extracted over a period of 4 h. This solution was then used for quantitative determination of PHI-443 using HPLC methods described below.
HPLC Analysis of PHI-443
Chromatographic analysis of PHI-443 was carried out using a previously established and validated HPLC method. The HPLC used for these studies was a Hewlett Packard series 1100 instrument equipped with Chemstation software for data analysis (Agilent Technologies, Palo Alto, CA). The analytical column used was a reverse phase Lichrospher 100, RP-18 (5 µm, 4 x 250 mm) column equipped with a 4 x 4 mm Lichrospher 100 and RP-18 (5 µm) guard column, and the detection wavelength was set at 275 nm with a slit width of 4 nm. The method employed an isocratic mobile phase of acetonitrile and 0.1% acetic acid (70:30, v/ v). The column was equilibrated with the mobile phase before data collection. The flow rate was set at 1 ml/mn and the column temperature was set at 20°C. The retention time for PHI-443 was 7.5 min. The concentration of PHI-443 was calculated from an external calibration curve, generated using known concentrations of PHI-443. The standard curves were generated by plotting the peak area against the PHI-443 concentrations tested.
Activity Against Drug-Resistant HIV Strains
The activity of PHI-443 was tested against a drug-sensitive strain (HTLVVIIIB); nonnucleoside RT inhibitor (NNRTI)–resistant strains (A17 and Al7 variant); a multi-drug–resistant HIV-1 strain (RT-MDR); 23 primary clinical isolates harboring two to seven RT gene mutations (M41L, D67N, K20R, K70R, T215Y, K219E, E44D, T69D, T69N, L210W, T69N, K70R, K103N, Y181C, T215F, K219E, F116S, L74V); as well as nine nucleoside RT inhibitor (NRTI)–resistant primary clinical isolates with two to seven RT mutations (M41L, D67N, K20R, K70R, T69N, T215F, T215Y, K219Q, L210W), using the previously established methods [31, 32].
Hu-PBL-SCID Mouse Model
SCID mice The in vivo microbicide efficacy of PHI-443 was evaluated in the Hu-PBL-SCID mouse model of human AIDS [33]. CB0.17 SCID mice 6–8 wk of age were obtained from Taconic Laboratories (Germantown, NY) and maintained in our BL-3 containment facility under specific pathogen-free conditions. Mice were housed in microisolator cages (Allentown Caging Equipment Co., Inc., Allentown, NJ, or Lab Products, Inc., Maywood, NY) containing autoclaved food, water, and bedding. Trimethoprim-sulfamethoxazole (Bactrim) was added to the drinking water once a week. Microbicide efficacy studies in SCID mice were approved by the Parker Hughes Institute Animal Care and Use Committee (PHIACUC), and all animal care procedures conformed to current NIH guidelines.
Hu-PBL-SCID mice were generated by reconstituting CB0.17 SCID mice with an i.p. inoculum of 5 x 107 human peripheral blood leukocytes (PBLs) from an HIV-negative donor [34]. Mice were treated s.c. with 5 mg progestin on the same day to synchronize the estrus cycle and to facilitate viral transmission by the vaginal route. Ten days before HIV infection, PBLs from a different HIV-negative donor were stimulated with 2 µg/ml phytohemagglutinin (PHA) and after 12 h were infected (0.1 MOI) with the primary clinical HIV-1 isolate BR/92/019, a genotypically and phenotypically NRTI-resistant HIV-1 isolate. The RT gene of BR/92/ 019 isolate harbors four mutations, namely D67N, L214F, T215D, and K219Q, which are associated with resistance to NRTIs. Seven days after reconstitution, progestin-treated Hu-PBL-SCID mice were anesthetized with Isoflurane and challenged with an intravaginal inoculum of 2 x 106 HIV-infected PBLs resuspended in tissue culture medium and mixed with human semen (1 h adsorption time before inoculation) with and without pretreatment with 2 µM PHI-443 or vehicle (PEG-400/polysorbate 80). The total volume of the inoculum was 25 µl, of which 5 µl was semen. The inoculum was introduced into the vaginal vault and held in place for 5 min. After vaginal inoculation with HIV-infected PBLs, mice were monitored for 2 wk for overall health and survival and then electively killed to determine if their spleen cells were infected with HIV by PCR amplification of a 115-bp sequence in the gag region of HIV-1 genome and determination of the viral RNA load (log 10 [RNA copies per mg of spleen tissue] using the Organon Teknika's Nuclisens HIV-1 QT assay kit; bio-Merieux, Durham, NC) [34].
Assays of Sperm Function
Sperm motility using computer-assisted sperm analysis Donor semen specimens were obtained after informed consent and in compliance with the guidelines of the Parker Hughes Institute Institutional Review Board. The effect of PHI-443 on human sperm motility was tested using semen and swim-up fractions. Highly motile fraction of sperm was prepared from normospermic semen (n = 5) by discontinuous (90%–45%) gradient centrifugation followed by a swim up of the washed and pelleted fraction in Biggers, Whitten, and Whittingham's medium (BWW; Irvine Scientific, Santa Ana, CA) containing 3% BSA [35]. The supernatant containing highly motile fraction of sperm was recovered in BWW medium with 0.3% BSA and resuspended in the same medium. Motile sperm (>10 x 106/ml) obtained from individual donors were exposed to serial two-fold dilutions of PHI-443 (62.5–1000 µM) in BWW medium. The stock solutions of PHI-443 were prepared in dimethyl sulfoxide (DMSO) and diluted in medium to obtain the desired concentrations in 0.5% DMSO. After a 3-h incubation with PHI-443 at 37°C, the motilities of sperm were compared with those of vehicle-treated (0.5% DMSO) control suspensions of motile sperm.
Sperm kinematic parameters For computer-assisted sperm analysis (CASA), 5-µl of each sperm suspension was loaded into a 20-µm Microcell slide (Conception Technologies, San Diego, CA) in a counting chamber at 37°C. At least 5–8 fields per slide were scanned for analysis using a Hamilton Thorne Integrated Visual Optical System (IVOS), version 10.9i, instrument (Hamilton Thorne Research Inc., Danvers, MA) [36, 37]. Each field was recorded for 30 sec. The computer calibrations were set at 30 frames at a frame rate of 30/sec. Other settings were as follows: minimum contrast, 8; minimum size, 6; low-size gate, 1.0; high-size gate, 2.9; low-intensity gate, 0.6; high-intensity gate, 1.4; phase-contrast illumination; low path velocity, 10 µm/s; threshold straightness, 80%; and magnification factor, 1.95.
The sperm kinematics parameters that were determined included numbers of progressively motile (MOT) sperm; curvilinear velocity (VCL); average path velocity (VAP); straight line velocity (VSL); beat cross frequency (BCF); the amplitude of lateral head displacement (ALH); and the derivatives, straightness (STR = VSL/VAP x 100) and linearity (LIN = VSL/VCL x 100). Data from each individual cell track were recorded and analyzed. At least 200 sperm were analyzed for each aliquot sampled.
Sperm motility using bovine cervical mucus penetration assay The bovine cervical mucus penetration test was performed using the Penetrak kit (Biochem Immunosystems Inc., Allentown, PA). The test was performed essentially according to the manufacturer's instructions. Briefly, flat capillary tubes were thawed at room temperature for 30 min and snapped above the mucus meniscus. The cut end of two tubes were placed vertically in a vial containing 0.2 ml aliquots of liquefied normospermic donor semen (n = 4) treated with and without 250, 500, and 1000 µM of PHI-443 and incubated at room temperature for 2 h. The capillary tubes were then placed on a calibrated microscope slide and the distance traveled by the vanguard sperm was recorded to the nearest millimeter.
In Vivo Fertility and Mucosal Toxicity Studies in the Rabbit Model
One hundred twelve sexually mature female and 48 male NZW rabbits were obtained from either Charles River Canada (St. Constant, Quebec) or Harlan (Oakwood, MI) and acclimated for a minimum of 3 wk. The bucks (mean age = 9 mo, range = 8–10 months, and mean body weight = 4.6 kg) were trained, for 1–2 months, to provide semen via an artificial vagina [38]. Virgin does were 6- to 7-mo-old (mean body weight = 4.3 kg). Does and bucks were housed individually in stainless steel cages (dimensions, 20 x 25 x 30 inches) in separate rooms under standard conditions (temperature, 20 ± 2°C, 50% ± 10% relative humidity, and 12-h fluorescent light cycle). Food (2031 Global High Fiber Rabbit Diet, Harlan Teklad, Madison, WI) and water were available ad libitum. All animal husbandry operations were conducted under current USDA guidelines. The rabbits were isolated for a minimum of 3–4 wk before the fertility trials. All rabbit procedures were approved by the PHIACUC.
In Vivo Fertility Trials
The preservation of in vivo sperm function following pretreatment of semen with PHI-443 either ex vivo or in vivo was evaluated in the rabbit model. Experiment 1 evaluated the in vivo fertilizing capacity of artificially inseminated sperm pretreated with PHI-443. Twenty-two sexually mature does were divided into two subgroups of 10 and 12: the vehicle group and PHI-443 group. Semen was obtained from trained bucks (n = 24) of proven fertility via a prewarmed (45°C) artificial vagina immediately before use [38]. Sperm count and motility was assessed to ensure that the males were ejaculating good quality semen. Prior to artificial insemination (AI), semen samples without the contamination of urine or gel were pooled, and 0.6 ml (>30 x 107 sperm/ml) aliquots were mixed with PHI-443 to yield a final concentration of 1000 µM in 0.5% DMSO. A corresponding volume of DMSO (0.5%) was added to control semen aliquots. After 15 min, the treated semen (0.5 ml; >15 x 107 sperm; 270 FD50 dose [the dose of sperm resulting in 50% fertility or a litter size half that of a normal litter size]) [39] was deposited into the vagina. Presence of progressively motile sperm in treated semen was confirmed at the time of AI. Ovulation was induced at the time of AI by an intravenous (i.v.) injection of 100 IU of human chorionic gonadotropin (hCG; Sigma) into the marginal ear vein. On Postinsemination Day 8, inseminated does were killed, the uteri and the ovaries were excised, and the total number of embryo implantations in each uterine horn and the number of corpora lutea in each ovary indicative of ovulated eggs were determined. The ratio of embryos to corpora lutea was used as a measure of fertility.
Experiment II evaluated the contraceptive potential of 2% PHI-443-containing gel, which was administered intravaginally at the time of AI. Twenty-six does were divided into two subgroups of 6 and 20: untreated control and 2% PHI-443 groups. Fifteen minutes before AI, 2 ml of appropriate gel was deposited intravaginally at a depth of 8 cm followed by deposition of 0.5 ml semen to a depth of 5–6 cm. Ovulation was induced at the time of AI by an i.v. injection of 100 IU of hCG into the marginal ear vein. Pregnancy was determined by counting the number of embryos 8 days after ovulation induction and AI. Since semen volume during ejaculation is expected to dilute the gel and potentially affect the gel viscosity as well as drug absorption through vaginal mucosa, the gel-based fertility studies were performed with 2 ml of 2% PHI-443-containing gel.
Experiment III evaluated the potential effect of intravaginal administration of PHI-443–containing gel at the time of AI on pregnancy outcome. Forty does and 24 bucks were used. The does were divided into two subgroups of 20: vehicle group and 2% PHI-443 group. Two milliliters of gel with and without 2% PHI-443 was applied intravaginally to a depth of 8 cm. Freshly pooled semen (0.5 ml) was deposited into the vagina to a depth of 5–6 cm 15 min following gel application. After AI and ovulation induction, the does were allowed to complete their pregnancies (31 ± 2 days). Pregnant does were transferred to cages containing nest boxes (16 x 12 x 6 inches). The litter size as well as the weight, length, and condition of each offspring at birth were recorded. Pregnancy rates were calculated as the proportion of does that become pregnant and delivered newborn rabbits. The pups were allowed to remain with the dam until Day 5. The number of pups per litter and the mean weight of the pups surviving on Day 5 were used to measure the perinatal effects of test agents [40].
Assays of Mucosal Toxicity
Genital epithelial cell viability The potential cytotoxicity of PHI-443 against normal human vaginal and endocervical epithelial cells was measured metabolically using the MTT (3-[4, 5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide)–based colorimetric assay [41, 42]. Normal human vaginal and endocervical epithelial cells (Clonetics Corporation, San Diego, CA) were dispensed into 96-well microtiter plates at 104/0.1 ml in modified small airway epithelial cell basal medium [7, 43]. The cells were then exposed to serial two-fold dilutions of PHI-443 or nonoxynol-9 (N-9 or IGEPAL CO-630; Rhone Poulenc, Cranbury, NJ) ranging from 7.8 to 1000 µM in triplicate for up to 24 h at 37°C. The optical density (at 540 nm) values obtained were converted to percent cell survival and expressed as mean IC50 values [43]. Nonlinear regression analysis was used to find IC50[MTT] values from the concentration-effect curves defined as the concentration required for 50% reduction in cell survival.
Rabbit vaginal irritation test For the vaginal irritation study, 24 female rabbits in subgroups of six were administered intravaginally with 1-ml of a gel with and without (placebo control) 0.5%, 1%, and 2% PHI-443 for 14 consecutive days. The formulations were applied to a depth of >8 cm to ensure adequate spreadability and retention, as well as to prevent any leakage. Rabbits were individually observed daily for overt clinical signs (genital swelling, redness, and vaginal discharge including bleeding). Animals were killed on Day 15, 24 h after the last gel application, and the genital tract was examined for swelling and redness, as well as bleeding. The vaginal wall was excised and parts of the upper (cervico-vagina), middle (midvagina), and lower (uro-vagina) regions of each vagina were fixed in 10% neutral-buffered formalin.
Fixed vaginal tissues were embedded in paraffin, sectioned at a thickness of 4–6 µm, stained with hematoxylin-eosin (H&E), and examined under x200 and x400 magnification using a Leica light microscope (Milton Keynes, Buckinghamshire, UK) interfaced with an image analysis system. Each of the three regions of vagina was examined for epithelial ulceration, leukocyte infiltration, vascular congestion, and edema. The histopathological scoring was performed in a blinded fashion by a board-certified veterinary pathologist. The irritation scores were assigned based on the semiquantitative scoring system of Eckstein et al. [44], which was as follows: individual score, 0 = none, 1 = minimal, 2 = mild, 3 = moderate, and 4 = intense irritation. The scoring system correlates to human irritation potential, which is as follows: total scores of 0–8 are acceptable, scores of 9–10 indicate borderline irritation potential, and scores of 11 and above are indicative of significant irritation potential. Results were expressed as the mean values and compared with those observed for does (n = 3) given a 4% N-9 containing gel under identical test conditions [45].
The systemic absorption from intravaginally applied 2% PHI-443–containing gel (10 mg/kg) was monitored by analytical HPLC of plasma extracts. Rabbits in subgroups of three per time point were administered intravaginally 2 ml of a 2% PHI-443–containing gel and blood samples were obtained at 0, 5, 15, 30, 60, 120, 180, and 240 min. Aliquot (150 µl) of each plasma sample was mixed 1:4 with acetone (600 µl) and vortexed for at least 30 sec. Following centrifugation at 7000 x g for 5 min, the supernatant was transferred into a clean tube and dried under nitrogen. A 50 µl solution of 90% methanol was used to reconstitute the extraction residue, and 40 µl of the reconstituted sample were subjected to analytical HPLC. The HPLC chromatograms were compared with control plasma extracts spiked with known amounts of PHI-443. The inter- and intraassay coefficients of variation in plasma were <4%. The lowest limit of detection of PHI-443 was 0.1 µM/0.1 ml plasma at a signal-to-noise ratio of 2. The overall accuracy of this method was >98%.
Evaluation of Cell Proliferation by Proliferating Cell Nuclear Antigen (PCNA) Staining
PCNA immunostaining was performed to evaluate the proliferative activity of cells in the vaginal tissues of rabbits from the 0.5%, 1%, and 2% PHI-443 group as well as the placebo control group. Immunostaining of sections (4 µm) with negative and positive controls was performed using the Zymed PCNA kit (Zymed Laboratories, South San Francisco, CA) as described previously [45]. The nature of staining and the distribution of PCNA immunoreactivity were determined by scoring a minimum of 300 cells in the vaginal epithelium and the stromal region in several random fields, and the percentage of PCNA-positive cells for each region of the tissue section was calculated.
Statistical Analysis
The statistical significance of the treated group mean with that of the control group with respect to sperm function parameters, mean number of corpora lutea, and embryos was analyzed by a one-way analysis of variance (ANOVA) followed by Dunnett multiple comparison test using GraphPad Instat version 2.03 software (San Diego, CA). Fisher exact probability test was used to compare the proportion of pregnant rabbits or the number of mice infected between test and control groups. The unweighted linear regression analysis of PHI-443 standard curves was performed by using the CA-Cricket Graph III version 1.5.3 software program (Computer Associates International, Islandia, NY). Nonlinear regression analysis was used to find IC50 values from the concentration-effect curves using the GraphPad PRISM version 4.0a software program. Differences were considered statistically significant if P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
PHI-443 is a broad-spectrum anti-HIV agent Table 1 summarizes the anti-HIV activity profile of PHI-443. PHI-443 effectively inhibited the replication of the HIV-1 strain HTLVIIIB in human PBMC with an IC50 value of 0.03 µM. It was almost as potent against the NNRTI-resistant HIV-1 strain A17 with a Y181C mutation (IC50, 0.04 µM vs. 0.03 µM) and inhibited the replication of the NNRTI-resistant HIV-1 strain, A17 variant, containing the Y181C+K103N RT mutations with an IC50 value of 3.2 µM, whereas the IC50 values of trovirdine, nevirapine, and delavirdine were all >100 µM. PHI-443 was 10 times more potent against the multidrug-resistant HIV-1 strain RT-MDR with a V106A mutation as well as additional mutations involving RT residues 74V, 41L, and 215Y relative to its activity against the HTLVIIIB strain containing wild-type RT. Notably, PHI-443 was active against genotypically and/or phenotypically NRTI/NNRTI-resistant 23 primary clinical HIV-1 isolates (subtypes A, B, F, and G carrying two to seven thymidine analogue mutations [TAMs]) tested with a mean IC50 value of 0.02 µM (range 0.24–0.00004 µM). In addition, PHI-443 inhibited the infectivity of nine NRTI-resistant primary clinical HIV-1 isolates (carrying two to seven TAMs) tested in a syncytial focus (plaque) formation assay using the CD4-expressing HeLa cell line HT4-6C with a mean IC50 value of 0.03 µM (range 0.18–0.005 µM).
|
PHI-443 inhibits vaginal transmission of HIV-1 in the Hu-PBL-SCID mouse model In preliminary studies, PHI-443 exhibited remarkable microbicidal activity against BR/ 92/019, a genotypically and phenotypically NRTI-resistant HIV-1 isolate, in the Hu-PBL-SCID mouse model of sexually transmitted AIDS. Splenocytes from 5 of 10 vehicle control SCID mice became HIV-PCR positive 2 wk following intravaginal inoculation with HIV-1–infected human PBMCs suspended in 20% human semen (Table 2). In contrast, splenocytes from none of the seven mice intravaginally inoculated with HIV-1 infected PBMCs in the presence of PHI-443 plus semen showed PCR evidence of HIV infection (P < 0.05, Fisher exact probability test).
|
Sperm Function
Treating human sperm with PHI-443 has no effect on sperm motility and kinematics Exposure of a highly motile fraction of sperm to PHI-443, which inhibited HIV-1 replication in human PBLs with an IC50 value of 0.03 µM for p24 viral antigen production and 0.8 µM for RT activity, did not significantly affect sperm motility even at concentrations as high as 500 µM (Table 3). Further, sperm motion kinematics using CASA confirmed that PHI-443 treatment (>3 h) did not significantly alter (P = 0.20–0.94) the sperm motion parameters, such as the mean progressive motility (MOT), track speed (VCL), path velocity (VAP), straight line velocity (VSL), straightness of the swimming pattern (STR), linearity of the sperm tracks (LIN), beat-cross frequency (BCF), and the amplitude of lateral sperm-head displacement (ALH; Table 3).
|
Treating human sperm with PHI-443 has no effect on functional motility in cervical mucus The bovine cervical mucus penetration test has been used to evaluate the in vitro fertilizing capacity of human sperm [46]. Therefore, we tested the ability of PHI-443–treated sperm to penetrate standardized midcycle bovine cervical mucus (Penetrak). Treatment of human semen with increasing concentrations of PHI-443 had no significant inhibitory effect in terms of the vanguard sperm penetration distance (Fig. 2). The mean penetration distance of the vanguard sperm from fresh vehicle-treated specimens in bovine cervical mucus was 34.2 ± 5.9 mm after 2 h of migration, whereas the corresponding distances for sperm treated with 250, 500, and 1000 µM of PHI-443 were 35.4 ± 7.8 mm, 31.2 ± 6.2 mm, and 32.2 ± 7.2 mm, respectively.
|
Treating rabbit sperm ex vivo with PHI-443 has no effect on in vivo fertility Since PHI-443 had no effect on the functional motility of sperm in the in vitro tests, we tested the in vivo fertilizing capacity of artificially inseminated semen pretreated with PHI-443 in the rabbit model. The cumulative pregnancy rates and implantation data are summarized in Table 4. Treatment of semen with 1000 µM PHI-443, which is nearly 30 000 times higher than its anti-HIV IC50 value, had no statistically significant effect on reproductive indices as assessed by the mean number of embryos or the percent conceptus based on number of embryos to corpora lutea. The mean numbers of ovarian corpora lutea on postovulation/insemination Day 8 were similar in both subgroups (Table 4). In addition, the preimplantation losses in rabbits inseminated with PHI-443–pretreated semen were not greater than those of vehicle control group (17% vs. 18%, respectively).
|
Formulation and stability of PHI-443 Since PHI-443 is a lipophilic molecule (log P = 4.39), a self-emulsifying nonspermicidal gel composed of microcrystalline cellulose, xanthan gum, and sorbitol as suspending agents along with polyethylene glycol 400 and water as a carrier and polysorbate-80 as the surfactant was developed. The formulation was a fine dispersion of the active ingredient in the formulation matrix. PHI-443 was stable at 5°–70°C when tested for three months, and its solubility was consistent at pH ranges of 4–8 (data not shown).
Exposure of rabbit semen in vivo to PHI-443–containing gel has no effect on fertility Since PHI-443 treatment of semen did not impede in vivo fertility, we conducted further studies to test the potential lack of contraceptive activity of PHI-443–containing gel administered at the time of AI. Intravaginal administration of gel with and without 2% PHI-443 before AI had no effect on reproductive indices such that the percentage of embryos based on the number of embryos to corpora lutea. The mean number of ovarian corpora lutea and mean number of embryos on Postovulation/Insemination Day 8 were similar in both subgroups (Table 4). In addition, the preimplantation losses in rabbits inseminated with 2% PHI-443–pretreated semen were not greater than those of the vehicle control group (38% vs. 32%, respectively).
The noncontraceptive nature and safety of intravaginally applied gel with and without 2% PHI-443 was confirmed in the rabbit model. Ovulated NZW rabbits in subgroups of 20 received an intravaginal application of 2 ml of gel with and without 2% PHI-443 minutes before AI and allowed to complete term pregnancy. Exposure of sperm to 2% PHI-443–containing gel in vivo had no effect on subsequent fertility in the rabbit model. When compared with placebo control, intravaginal administration of a 2% PHI-443 gel had no effect on the pregnancy rate, litter size, or the number of viable litters (Table 5). Rabbits that delivered litters following an intravaginal application of 2% PHI-443–containing gel before AI had healthy offspring with no apparent peri- or postnatal repercussions. Placebo control and 2% PHI-443–treated females that sired pups had no statistically significant differences in the mean pup weight and pup survival rate on Lactation Days 1 and 5, respectively (Table 5).
|
Mucosal Safety Studies
PHI-443 is not cytotoxic to genital tract epithelial cells The MTT cell viability assay was used to test the potential in vitro cytotoxic effect of PHI-443 against confluent monolayers of normal human vaginal and endocervical epithelial cells. Cells were exposed to PHI-443 at concentrations ranging from 7.8 to 1000 µM for up to 24 h. When compared with N-9, which was used as a positive control, PHI-443 had no deleterious effect on the viability of normal human vaginal and cervical epithelial cells even at 1000 µM (selectivity index [SI] = >30 000; Fig. 3). In contrast, the mean IC50[MTT] values calculated from the concentration-dependent cell survival curves for N-9–treated vaginal and cervical epithelial cells were <40 µM (SI = <18).
|
Gel formulation of PHI-443 lacks mucosal toxicity in the rabbit model
A rabbit vaginal irritation test was performed to assess the potential for PHI-443–containing gel to cause mucosal toxicity. Histological evaluation of three different vaginal regions (cervico-vagina, midvagina, and uro-vagina) after daily intravaginal application of gel with and without 0.5%, 1%, and 2% PHI-443 for 14 consecutive days did not cause vaginal irritation in any of the 18 rabbits evaluated (mean individual scores 0–2). The total scores for the gel formulation containing 0, 0.5%, 1%, and 2% PHI-443 were 3–4, 4, 4–5, and 6, respectively (Table 6). The corresponding total score for the 4% N-9 gel observed under identical conditions were 8–11. None of the PHI-443–treated rabbit vaginal tissues had total scores of >7 (range 4–6). Thus, the irritation potential of PHI-443–containing gel even at a concentration 2 x 106 times higher than its in vitro anti-HIV IC50 was in the acceptable range (total score 
8) for clinical trial.
|
Intravaginal PHI-443 has no effect on the number of cycling cells Table 7 provides the percentages of PCNA-positive cells, a marker for cellular hyperplasia, in the vaginal epithelium and stroma of tissue sections from rabbits given gel formulation with and without 0.5%, 1%, or 2% PHI-443 for 14 consecutive days. Nearly 60% of vaginal epithelial cells and 38% of stromal cells showed nuclear staining for PCNA in control tissues. No significant differences in PCNA positivity were noted in the epithelium or stromal cells of placebo control and PHI-443–exposed tissues.
|
Lack of vaginal absorption of PHI-443 The time course of potential systemic absorption of PHI-443 following intravaginal administration was studied in NZW rabbits. PHI-443 added to control plasma was clearly visualized using established HPLC conditions as a single peak with a retention time of 7.5 min (Fig. 4, A and B). In contrast, no peak at 7.5 min was detected in the HPLC chromatograms of plasma extracts prepared at 15 min (Fig. 4C) and 240 min (Fig. 4D) after intravaginal administration of 2% PHI-443– containing gel. All plasma samples had undetectable PHI-443 throughout the 4-h sampling period.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Through an integrated effort involving synthesis, docking studies, and biological evaluation, we identified thiophene thiourea (PHI-443) with unprecedented potency against NNRTI-resistant HIV-1 strains harboring Y181C, K103N, or V106A mutations [25, 26]. As predicted, PHI-443 was substantially more potent than the standard NNRTI drugs, nevirapine and delavirdine, against the NNRTI-resistant HIV-1 strains. PHI-443 was five times more potent than trovirdine, 100 times more potent than delavirdine, and 1250 times more potent than nevirapine against multidrug-resistant HIV-1 strain RT-MDR with mutations involving RT residues V106A, 74V, 41L, and 215Y. Similarly, PHI-443 was more effective than trovirdine, nevirapine, and delavirdine against the problematic NNRTI-resistant HIV-1 strain A17 variant with both Y181C and K103N mutations in RT. These findings established PHI-443 as a potent inhibitor of drug-sensitive and multidrug-resistant strains of HIV-1.
Our docking studies with PHI-443 using the computer-generated model of the NNRTI binding pocket indicated that the thiophene ring situated in the Wing 2 region of the NNI binding pocket provide better contact with RT residues including Y181. The thiophene group of PHI-443 was found to be located in close proximity of the Y181 residue [30]. In this docked position, the sulfur (S) atom of the thiophene ring is only 4.4 Å away from the carbon atom of the Y181 residue, which is mutated to an S atom in the RT Y181C mutant strains (A17 and A17 variant). The S atom of the thiophene group is more compatible with the S-containing cysteine 181 residue than the pyridyl group of trovirdine. Accordingly, PHI-443 exhibited superior anti-HIV activity especially against drug-resistant HIV-1 primary clinical isolates.
There are at least 10 distinct group M (major) subtypes of HIV-1 strains that are endemic to distinct geographical sites [47–49]. Currently available anti-HIV agents have been traditionally developed against subtype B HIV-1 strains that are the predominant strains in the United States and Europe, although worldwide the majority of HIV-infected individuals are infected with nonsubtype B strains, and the vast majority of new infections are caused by nonsubtype B HIV-1 strains [48–50]. Therefore, there is an urgent need to identify anti-HIV agents with potent activity against nonsubtype B HIV-1. Notably, PHI-443 was active against the primary nonsubtype B HIV-1 isolates originating from South America, Asia, and sub-Saharan Africa with nano/submicromolar IC50 values. This is particularly relevant because a high percentage of newly infected individuals in the United States harbor NRTI/NNRTI-resistant mutants with increased incidence of HIV subtypes [51]. As a potential microbicide, PHI-443 is likely to be used in a prophylactic manner primarily by noninfected individuals, and there is little opportunity of HIV-1 to remain under constant drug pressure in these cases. It is also unlikely that the use of NNRTI-containing microbicides will allow the transmission of NNRTI-resistant strains to naive individuals. The concentration of PHI-443 readily attainable in the lipophilic gel is substantial, much higher than the IC90 for even the most resistant virus.
Although semen is an important vehicle for sexual transmission of HIV-1 [6], the blood viral burden appears to be the main determinant of the risk of heterosexual transmission of HIV-1, and transmission is rare from infected individuals with levels of less than 1500 copies of HIV-1 RNA per milliliter [52]. Furthermore, intrauterine inseminations performed among HIV-1 serodiscordant couples in Europe have clearly shown that HIV-1-seropositive men can have their own children without infecting their female partners or fetus [53, 54]. Effective antiretroviral therapy before ART procedure can reduce the viral burden in both blood and genital secretions, thereby reducing the likelihood of sexual transmission of HIV-1 [55]. However, recent studies indicate that the evolution of HIV-1 is independent in the blood and genital tract compartments [56]. Consequently, the genital tract fluids in which HIV-1 can reside may contain HIV-1 variants with genotypic resistance. Not all men on antiretroviral therapy have complete suppression of HIV-1 replication in the genital tract; thus, they shed resistant HIV-1 strains. Inasmuch as HIV-1 can still be cultured from the genital secretions of some patients who are receiving antiretroviral therapy and who have undetectable levels of HIV-1 RNA in their blood [51], pretreatment of semen with PHI-443 before ART procedures is likely to further reduce the risk of sexual transmission of HIV-1. The results of our present study illustrate the preservation of sperm function following pretreatment of sperm with PHI-443 as a prophylactic antiviral agent.
Because PHI-443 is lipophilic, we formulated a novel nonspermicidal self-emulsifying gel as a vaginal drug delivery vehicle for PHI-443. The gel components identified are pharmaceutically, pharmacologically, and pharmacokinetically acceptable nontoxic excipients used in the preparation of a variety of topical and/or oral medications [57– 59]. Microcrystalline cellulose is a widely used pharmaceutical and food additive. PEG is commonly used as a base for cream, gel, and ointment preparations because of its physical characteristics and the versatile consistencies. Polysorbate 80 is the least cytotoxic among the various surfactants currently used in cosmetic products [60, 61]. Xanthan gum was preferred as a gel base because of its safety, bioadhesive properties, and wide acceptability [62]. Xanthan gum exhibits pseudoplastic (shear thinning) behavior in solution that is higher than other commercial hydrocolloids. It is extremely stable in both acidic and alkaline solutions over a broad pH range of 2–12 and over a wide range of temperatures (up to 60°C). Because the constituents of the self-emulsifying gel are nontoxic drug solubilizers and polymers, in the 14-day rabbit vaginal tolerance test PHI-443 lacked mucosal toxicity even at concentrations as high as 2 x 106 times its in vitro anti-HIV IC50.
The in vivo toxicity, pharmacokinetic features, tissue distribution, and metabolism of PHI-443 have been examined in CD-1 mice using a sensitive HPLC-based quantitative detection method. PHI-443 was nontoxic to CD-1 mice at dose levels ranging from 10 to 100 mg/kg and exhibited favorable pharmacokinetic behavior when administered intravenously or intraperitoneally. Pharmacokinetic studies following both oral and intravaginal administration of PHI-443 (350 and 10 mg/kg, respectively) indicated very low capacity to be absorbed through the oral (3.4%) or vaginal mucosal epithelium. PHI-443 did not show significant changes in solubility as a function of pH. This is consistent with its chemical structure due to lack of easily ionizable functional groups.
In conclusion, the broad-spectrum anti-HIV agent PHI-443 shows potential as a vaginal microbicide for the prevention of the sexual transmission of HIV, which is thought to be transmitted primarily by infected cells present in semen. The demonstrated lack of adverse effects on fertility and vaginal mucosa indicates that PHI-443 displays many of the properties required of an ideal nonspermicidal microbicide and warrants further preclinical research and development.
|
FOOTNOTES |
|---|
2 Correspondence: Osmond J. D'Cruz, Parker Hughes Institute, 2657 Patton Road, St. Paul, MN 55113. FAX: 651 628 9891; odcruz{at}ih.org
Received: 6 June 2004.
First decision: 30 June 2004.
Accepted: 11 August 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
