ACTBIO 3590
No. of Pages 11, Model 5G
17 February 2015 Acta Biomaterialia xxx (2015) xxx–xxx 1
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat 5 6
4
Precise engineering of dapivirine-loaded nanoparticles for the development of anti-HIV vaginal microbicides
7
José das Neves ⇑, Bruno Sarmento
8 9
INEB – Instituto de Engenharia Biomédica, University of Porto, Porto, Portugal CESPU, Instituto de Investigação e Formação Avançada em Ciências e Tecnologias da Saúde, Gandra PRD, Portugal
3
10 11 1 2 3 6 14 15 16 17 18 19 20 21 22 23 24 25
a r t i c l e
i n f o
Article history: Received 23 October 2014 Received in revised form 27 December 2014 Accepted 10 February 2015 Available online xxxx Keywords: Response surface methodology Poly(D,L-lactic-co-glycolic acid) Cytotoxicity Mucin Vaginal drug delivery
a b s t r a c t Polymeric nanoparticles (NPs) have the potential to provide effective and safe delivery of antiretroviral drugs in the context of prophylactic anti-HIV vaginal microbicides. Dapivirine-loaded poly(D,L-lactic-coglycolic acid) (PLGA) NPs were produced by an emulsion-solvent evaporation method, optimized for colloidal properties using a 3-factor, 3-level Box–Behnken experimental design, and characterized for drug loading, production yield, morphology, thermal behavior, drug release, in vitro cellular uptake, cytotoxicity and pro-inflammatory potential. Also, drug permeability/membrane retention in well-established HEC-1-A and CaSki cell monolayer models as mediated by NPs was assessed in the absence or presence of mucin. Box–Behnken design allowed optimizing monodisperse 170 nm drug-loaded NPs. Drug release experiments showed an initial burst effect up to 4 h, followed by sustained 24 h release at pH 4.2 and 7.4. NPs were readily taken up by different genital and macrophage cell lines as assessed by fluorescence microscopy. Drug-loaded NPs presented lower or at least similar cytotoxicity as compared to the free drug, with up to around one-log increase in half-maximal cytotoxic concentration values. In all cases, no relevant changes in cell pro-inflammatory cytokine/chemokine production were observed. Dapivirine transport across cell monolayers was significantly decreased when mucin was present at the apical side with either NPs or the free drug, thus evidencing the influence of this natural glycoprotein in membrane permeability. Moreover, drug retention in cell monolayers was significantly higher for NPs in comparison with the free drug. Overall, obtained dapivirine-loaded PLGA NPs possess interesting technological and biological features that may contribute to their use as novel safe and effective vaginal microbicides. Ó 2015 Published by Elsevier Ltd. on behalf of Acta Materialia Inc.
27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
49 50
1. Introduction
51
Recent years have seen an increasing interest in vaginal microbicides as a potential prevention modality for sexual HIV transmission. This strategy, also commonly referred to as topical preexposure prophylaxis (PrEP), comprises the use of vaginal products containing anti-HIV compounds prior to or around the time of sexual intercourse in order to avoid early transmission events at the mucosal level [1]. Proof-of-concept for vaginal microbicides was recently achieved in a Phase 2b clinical trial testing a gel containing 1% of the nucleotide reverse transcriptase inhibitor (NtRTI) tenofovir [2]. Most microbicide products developed and tested so far in human clinical trials have relied in semi-solid dosage forms,
52 53 54 55 56 57 58 59 60 61
⇑ Corresponding author at: INEB – Instituto de Engenharia Biomédica, Rua do Campo Alegre, 823, 4150-180 Porto, Portugal. Tel.: +351 226 074 900; fax: +351 226 094 567. E-mail address: [email protected] (J. das Neves).
particularly gels. These are coitally-dependent systems (i.e., require administration shortly before/after intercourse) that frequently lead to poor user adherence and, thus, potentially reduced ability to protect against transmission [3]. However, modified and sustained drug delivery systems, such as vaginal rings, may be preferable and are receiving a lot of the current attention as these can provide coitally-independent options for protection [4]. In particular, one vaginal ring containing the non-nucleoside reverse transcriptase inhibitor (NNRTI) dapivirine is currently undergoing two Phase 3 clinical trials, with results scheduled to be announced by late 2015 [5]. While waiting for these and other clinical results, alternative possibilities for advancing microbicides have been proposed, namely the use of nanotechnology-based carriers for delivering potent antiretroviral drugs. For example, nanocarriers may provide protective drug levels at cervicovaginal tissues and HIVtarget cells for longer time-frames [6,7]. This in turn may potentially increase protection against infection while allowing obtaining coitally-independent microbicides.
http://dx.doi.org/10.1016/j.actbio.2015.02.007 1742-7061/Ó 2015 Published by Elsevier Ltd. on behalf of Acta Materialia Inc.
Please cite this article in press as: das Neves J, Sarmento B. Precise engineering of dapivirine-loaded nanoparticles for the development of anti-HIV vaginal microbicides. Acta Biomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.02.007
62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79
ACTBIO 3590
No. of Pages 11, Model 5G
17 February 2015 2
J. das Neves, B. Sarmento / Acta Biomaterialia xxx (2015) xxx–xxx
104
Different research groups have been developing polymeric nanocarriers over the last years for the vaginal delivery of antiHIV microbicide compounds. These systems can be designed in order to modulate the interaction with cervicovaginal mucus fluids [8], respond to physiological changes occurring upon sexual intercourse (e.g., pH changes) [9], enhance drug targeting [10], promote drug penetration and accumulation at genital tissues [11], increase intracellular drug delivery [12], and improve drug activity [13]. Further, drug-loaded polymeric nanoparticles (NPs) may beneficially affect local genital pharmacokinetics (PK) and, therefore, potentially enhance the ability of antiretroviral compounds to protect against infection in a coitally-independent fashion [14]. However, the particularities of the vaginal route and dynamics of sexual intercourse require precise engineering of drug nanocarriers in order to allow optimal performance. Size and surface properties, in particular, play important roles in microbicide NP performance [15,16]. This work describes the systematic development of dapivirine-loaded poly(D,L-lactic-co-glycolic acid) (PLGA)-based NPs using a Box–Behnken factorial design. This particular statistical design approach has been proven useful previously in the characterization and optimization of drug nanocarriers with different applications [17,18], including vaginal HIV transmission prophylaxis [19]. Additionally, we detail on the physicochemical and biological characterization of an optimized dapivirine nanocarrier as relevant for vaginal anti-HIV microbicides.
105
2. Materials and methods
106
2.1. Materials
107
116
Dapivirine was provided by the International Partnership for Microbicides (Silver Spring, MD, USA). Acid-terminated PLGA (PurasorbÒ PDLG 5004A, 50:50 D,L-lactide:glycolide ratio, 0.4 dL g1 inherent viscosity) was an offer from Purac Biomaterials (Gorinchem, The Netherlands). Polyvinyl alcohol (PVA, 87–90% hydrolyzed, 30–70 kDa) and thiazolyl blue tetrazolium bromide (MTT) were acquired from Sigma-Aldrich (St. Louis, MO, USA), Hoechst 33342 from Invitrogen (Carlsbad, CA, USA), and poloxamer 407 from BASF (Ludwigshafen, Germany). All other chemicals and solvents were of analytic grade or equivalent.
117
2.2. Cell lines and culture conditions
118
129
HeLa human cervical cells (passages 34–36), CaSki human cervical cells (passages 11–13), HEC-1-A human endometrial cells (passages 12–14), and RAW 264.7 murine monocytes/macrophages (passages 37–39) were obtained from ATCC (Manassas, VA, USA). HeLa and RAW 264.7 cells were maintained in Dulbecco’s Modified Eagle medium with Ultraglutamine 1 (Lonza, Verviers, Belgium), CaSki cells in RPMI-1640 medium (Invitrogen), and HEC-1-A cells in McCoy’s 5A modified medium (Invitrogen). All media were supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin and 100 lg/mL streptomycin, all from Invitrogen. Cells were maintained/incubated at 37 °C, 5% CO2 and 95% humidity, and media renewed every 2–3 days.
130
2.3. Preparation and optimization of nanoparticles
131
Dapivirine-loaded NPs were prepared by a single oil-in-water emulsion-solvent evaporation method. PLGA (50 mg) and dapivirine (0–10 mg) were dissolved in ethyl acetate (2–6 mL) and mixed with 10 mL of a 1–5% (w/v) PVA solution using a Vibra-Cell™ VCX 130 ultrasonic processor equipped with a standard 6 113 mm probe (Sonics & Materials, Inc., Newtown, CT, USA) at 60–90% intensity for one minute. The emulsion was further diluted with
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103
108 109 110 111 112 113 114 115
119 120 121 122 123 124 125 126 127 128
132 133 134 135 136 137
20 mL of 0.2% (w/v) PVA solution and stirred overnight. Finally, NPs were centrifuged (45,000g, 30 min, at 4 °C) and washed twice with water. Freeze-drying of NPs was performed whenever required. Fluorescent NPs were prepared in the same way except 10% in weight of PLGA was substituted by the same amount of rhodamine-123-labeled PLGA (Supplementary materials, S1). The preparation method of dapivirine-loaded NPs was optimized using a 3-factor, 3-level Box–Behnken statistical design. Factors (or variables) such as the volume of ethyl acetate, the concentration of PVA solution and the intensity of sonication were studied for their influence in the properties of NPs. Evaluated responses included particle size (hydrodynamic diameter), polydispersity index (PdI), zeta potential, drug association efficiency (AE%), drug loading (DL%) and particle yield. MinitabÒ software (v. 15.1.20.0; Minitab Inc., State College, PA, USA) was used for generating the experimental design and result analysis. Optimized parameters for NP production were determined using the Response Optimizer function of the software in order to obtain particle size in the range of 150–200 nm, the lowest PdI, and the highest AE% and DL% values.
138
2.4. Characterization of nanoparticles
158
NPs dispersed in water to a final concentration of around 0.1 mg/mL were characterized for hydrodynamic diameter, PdI and zeta potential using a ZetaSizer Nano ZS (Malvern, Worcestershire, UK). Yield was determined as the percentage ratio between obtained and theoretical NP weight after freeze-drying. AE% and DL% of NPs were assessed indirectly by assaying the amount of drug recovered in supernatants resulting from centrifugation/ washing steps during particle production using a previously optimized and validated HPLC-UV method [20], according to the following equations:
159
139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157
160 161 162 163 164 165 166 167 168
169
Initial Dapivirine Recovered Dapivirine AE% ¼ 100 Initial Dapivirine
ð1Þ
171 172
Initial Dapivirine Recovered Dapivirine DL% ¼ 100 NPs Weight
ð2Þ
Scanning electron microscopy (SEM) of freeze-dried NPs vacuum-coated with a gold-palladium alloy was performed at an acceleration voltage of 5.0 kV using a FEI Quanta 400 FEG microscope (FEI Company, Hillsboro, OR, USA). Differential scanning calorimetry (DSC) was performed using a DSC 200 F3 MaiaÒ calorimeter (NETZSCH-Gerätebau, Selb, Germany). Analysis of freeze-dried NPs was carried out in order to assess the physical state of the matrix-forming polymer and other components of NPs, and potential interactions among these last. Samples (10 mg) were placed in lid-pierced aluminum crucibles and thermograms obtained in the range of 30–300 °C, at a heating rate of 10 °C/min, and under nitrogen gas flow (40 mL/min). Drug release from NPs was determined in phosphate buffered saline (PBS, pH 7.4) and a simulated vaginal fluid (SVF, pH 4.2) adapted from Owen & Katz [21]. SVF contained glucose (0.5%), sodium chloride (0.351%), lactic acid (0.2%), potassium hydroxide (0.14%), acetic acid (0.1%), urea (0.04%), calcium hydroxide (0.0222%), glycerol (0.016%), hydrochloride acid (eq for pH 4.2), and water (eq for 100%). Polysorbate 80 at a concentration of 1% (w/v) was added to both media in order to overcome the poor water solubility of dapivirine (
No. of Pages 11, Model 5G
17 February 2015 Acta Biomaterialia xxx (2015) xxx–xxx 1
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat 5 6
4
Precise engineering of dapivirine-loaded nanoparticles for the development of anti-HIV vaginal microbicides
7
José das Neves ⇑, Bruno Sarmento
8 9
INEB – Instituto de Engenharia Biomédica, University of Porto, Porto, Portugal CESPU, Instituto de Investigação e Formação Avançada em Ciências e Tecnologias da Saúde, Gandra PRD, Portugal
3
10 11 1 2 3 6 14 15 16 17 18 19 20 21 22 23 24 25
a r t i c l e
i n f o
Article history: Received 23 October 2014 Received in revised form 27 December 2014 Accepted 10 February 2015 Available online xxxx Keywords: Response surface methodology Poly(D,L-lactic-co-glycolic acid) Cytotoxicity Mucin Vaginal drug delivery
a b s t r a c t Polymeric nanoparticles (NPs) have the potential to provide effective and safe delivery of antiretroviral drugs in the context of prophylactic anti-HIV vaginal microbicides. Dapivirine-loaded poly(D,L-lactic-coglycolic acid) (PLGA) NPs were produced by an emulsion-solvent evaporation method, optimized for colloidal properties using a 3-factor, 3-level Box–Behnken experimental design, and characterized for drug loading, production yield, morphology, thermal behavior, drug release, in vitro cellular uptake, cytotoxicity and pro-inflammatory potential. Also, drug permeability/membrane retention in well-established HEC-1-A and CaSki cell monolayer models as mediated by NPs was assessed in the absence or presence of mucin. Box–Behnken design allowed optimizing monodisperse 170 nm drug-loaded NPs. Drug release experiments showed an initial burst effect up to 4 h, followed by sustained 24 h release at pH 4.2 and 7.4. NPs were readily taken up by different genital and macrophage cell lines as assessed by fluorescence microscopy. Drug-loaded NPs presented lower or at least similar cytotoxicity as compared to the free drug, with up to around one-log increase in half-maximal cytotoxic concentration values. In all cases, no relevant changes in cell pro-inflammatory cytokine/chemokine production were observed. Dapivirine transport across cell monolayers was significantly decreased when mucin was present at the apical side with either NPs or the free drug, thus evidencing the influence of this natural glycoprotein in membrane permeability. Moreover, drug retention in cell monolayers was significantly higher for NPs in comparison with the free drug. Overall, obtained dapivirine-loaded PLGA NPs possess interesting technological and biological features that may contribute to their use as novel safe and effective vaginal microbicides. Ó 2015 Published by Elsevier Ltd. on behalf of Acta Materialia Inc.
27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
49 50
1. Introduction
51
Recent years have seen an increasing interest in vaginal microbicides as a potential prevention modality for sexual HIV transmission. This strategy, also commonly referred to as topical preexposure prophylaxis (PrEP), comprises the use of vaginal products containing anti-HIV compounds prior to or around the time of sexual intercourse in order to avoid early transmission events at the mucosal level [1]. Proof-of-concept for vaginal microbicides was recently achieved in a Phase 2b clinical trial testing a gel containing 1% of the nucleotide reverse transcriptase inhibitor (NtRTI) tenofovir [2]. Most microbicide products developed and tested so far in human clinical trials have relied in semi-solid dosage forms,
52 53 54 55 56 57 58 59 60 61
⇑ Corresponding author at: INEB – Instituto de Engenharia Biomédica, Rua do Campo Alegre, 823, 4150-180 Porto, Portugal. Tel.: +351 226 074 900; fax: +351 226 094 567. E-mail address: [email protected] (J. das Neves).
particularly gels. These are coitally-dependent systems (i.e., require administration shortly before/after intercourse) that frequently lead to poor user adherence and, thus, potentially reduced ability to protect against transmission [3]. However, modified and sustained drug delivery systems, such as vaginal rings, may be preferable and are receiving a lot of the current attention as these can provide coitally-independent options for protection [4]. In particular, one vaginal ring containing the non-nucleoside reverse transcriptase inhibitor (NNRTI) dapivirine is currently undergoing two Phase 3 clinical trials, with results scheduled to be announced by late 2015 [5]. While waiting for these and other clinical results, alternative possibilities for advancing microbicides have been proposed, namely the use of nanotechnology-based carriers for delivering potent antiretroviral drugs. For example, nanocarriers may provide protective drug levels at cervicovaginal tissues and HIVtarget cells for longer time-frames [6,7]. This in turn may potentially increase protection against infection while allowing obtaining coitally-independent microbicides.
http://dx.doi.org/10.1016/j.actbio.2015.02.007 1742-7061/Ó 2015 Published by Elsevier Ltd. on behalf of Acta Materialia Inc.
Please cite this article in press as: das Neves J, Sarmento B. Precise engineering of dapivirine-loaded nanoparticles for the development of anti-HIV vaginal microbicides. Acta Biomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.02.007
62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79
ACTBIO 3590
No. of Pages 11, Model 5G
17 February 2015 2
J. das Neves, B. Sarmento / Acta Biomaterialia xxx (2015) xxx–xxx
104
Different research groups have been developing polymeric nanocarriers over the last years for the vaginal delivery of antiHIV microbicide compounds. These systems can be designed in order to modulate the interaction with cervicovaginal mucus fluids [8], respond to physiological changes occurring upon sexual intercourse (e.g., pH changes) [9], enhance drug targeting [10], promote drug penetration and accumulation at genital tissues [11], increase intracellular drug delivery [12], and improve drug activity [13]. Further, drug-loaded polymeric nanoparticles (NPs) may beneficially affect local genital pharmacokinetics (PK) and, therefore, potentially enhance the ability of antiretroviral compounds to protect against infection in a coitally-independent fashion [14]. However, the particularities of the vaginal route and dynamics of sexual intercourse require precise engineering of drug nanocarriers in order to allow optimal performance. Size and surface properties, in particular, play important roles in microbicide NP performance [15,16]. This work describes the systematic development of dapivirine-loaded poly(D,L-lactic-co-glycolic acid) (PLGA)-based NPs using a Box–Behnken factorial design. This particular statistical design approach has been proven useful previously in the characterization and optimization of drug nanocarriers with different applications [17,18], including vaginal HIV transmission prophylaxis [19]. Additionally, we detail on the physicochemical and biological characterization of an optimized dapivirine nanocarrier as relevant for vaginal anti-HIV microbicides.
105
2. Materials and methods
106
2.1. Materials
107
116
Dapivirine was provided by the International Partnership for Microbicides (Silver Spring, MD, USA). Acid-terminated PLGA (PurasorbÒ PDLG 5004A, 50:50 D,L-lactide:glycolide ratio, 0.4 dL g1 inherent viscosity) was an offer from Purac Biomaterials (Gorinchem, The Netherlands). Polyvinyl alcohol (PVA, 87–90% hydrolyzed, 30–70 kDa) and thiazolyl blue tetrazolium bromide (MTT) were acquired from Sigma-Aldrich (St. Louis, MO, USA), Hoechst 33342 from Invitrogen (Carlsbad, CA, USA), and poloxamer 407 from BASF (Ludwigshafen, Germany). All other chemicals and solvents were of analytic grade or equivalent.
117
2.2. Cell lines and culture conditions
118
129
HeLa human cervical cells (passages 34–36), CaSki human cervical cells (passages 11–13), HEC-1-A human endometrial cells (passages 12–14), and RAW 264.7 murine monocytes/macrophages (passages 37–39) were obtained from ATCC (Manassas, VA, USA). HeLa and RAW 264.7 cells were maintained in Dulbecco’s Modified Eagle medium with Ultraglutamine 1 (Lonza, Verviers, Belgium), CaSki cells in RPMI-1640 medium (Invitrogen), and HEC-1-A cells in McCoy’s 5A modified medium (Invitrogen). All media were supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin and 100 lg/mL streptomycin, all from Invitrogen. Cells were maintained/incubated at 37 °C, 5% CO2 and 95% humidity, and media renewed every 2–3 days.
130
2.3. Preparation and optimization of nanoparticles
131
Dapivirine-loaded NPs were prepared by a single oil-in-water emulsion-solvent evaporation method. PLGA (50 mg) and dapivirine (0–10 mg) were dissolved in ethyl acetate (2–6 mL) and mixed with 10 mL of a 1–5% (w/v) PVA solution using a Vibra-Cell™ VCX 130 ultrasonic processor equipped with a standard 6 113 mm probe (Sonics & Materials, Inc., Newtown, CT, USA) at 60–90% intensity for one minute. The emulsion was further diluted with
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103
108 109 110 111 112 113 114 115
119 120 121 122 123 124 125 126 127 128
132 133 134 135 136 137
20 mL of 0.2% (w/v) PVA solution and stirred overnight. Finally, NPs were centrifuged (45,000g, 30 min, at 4 °C) and washed twice with water. Freeze-drying of NPs was performed whenever required. Fluorescent NPs were prepared in the same way except 10% in weight of PLGA was substituted by the same amount of rhodamine-123-labeled PLGA (Supplementary materials, S1). The preparation method of dapivirine-loaded NPs was optimized using a 3-factor, 3-level Box–Behnken statistical design. Factors (or variables) such as the volume of ethyl acetate, the concentration of PVA solution and the intensity of sonication were studied for their influence in the properties of NPs. Evaluated responses included particle size (hydrodynamic diameter), polydispersity index (PdI), zeta potential, drug association efficiency (AE%), drug loading (DL%) and particle yield. MinitabÒ software (v. 15.1.20.0; Minitab Inc., State College, PA, USA) was used for generating the experimental design and result analysis. Optimized parameters for NP production were determined using the Response Optimizer function of the software in order to obtain particle size in the range of 150–200 nm, the lowest PdI, and the highest AE% and DL% values.
138
2.4. Characterization of nanoparticles
158
NPs dispersed in water to a final concentration of around 0.1 mg/mL were characterized for hydrodynamic diameter, PdI and zeta potential using a ZetaSizer Nano ZS (Malvern, Worcestershire, UK). Yield was determined as the percentage ratio between obtained and theoretical NP weight after freeze-drying. AE% and DL% of NPs were assessed indirectly by assaying the amount of drug recovered in supernatants resulting from centrifugation/ washing steps during particle production using a previously optimized and validated HPLC-UV method [20], according to the following equations:
159
139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157
160 161 162 163 164 165 166 167 168
169
Initial Dapivirine Recovered Dapivirine AE% ¼ 100 Initial Dapivirine
ð1Þ
171 172
Initial Dapivirine Recovered Dapivirine DL% ¼ 100 NPs Weight
ð2Þ
Scanning electron microscopy (SEM) of freeze-dried NPs vacuum-coated with a gold-palladium alloy was performed at an acceleration voltage of 5.0 kV using a FEI Quanta 400 FEG microscope (FEI Company, Hillsboro, OR, USA). Differential scanning calorimetry (DSC) was performed using a DSC 200 F3 MaiaÒ calorimeter (NETZSCH-Gerätebau, Selb, Germany). Analysis of freeze-dried NPs was carried out in order to assess the physical state of the matrix-forming polymer and other components of NPs, and potential interactions among these last. Samples (10 mg) were placed in lid-pierced aluminum crucibles and thermograms obtained in the range of 30–300 °C, at a heating rate of 10 °C/min, and under nitrogen gas flow (40 mL/min). Drug release from NPs was determined in phosphate buffered saline (PBS, pH 7.4) and a simulated vaginal fluid (SVF, pH 4.2) adapted from Owen & Katz [21]. SVF contained glucose (0.5%), sodium chloride (0.351%), lactic acid (0.2%), potassium hydroxide (0.14%), acetic acid (0.1%), urea (0.04%), calcium hydroxide (0.0222%), glycerol (0.016%), hydrochloride acid (eq for pH 4.2), and water (eq for 100%). Polysorbate 80 at a concentration of 1% (w/v) was added to both media in order to overcome the poor water solubility of dapivirine (
