ProTide generated long-acting abacavir nanoformulations

Article abstract of DOI:10.1039/c8cc04708a

Abacavir pronucleotide nanoformulations (NM3ABC) were prepared as a novel long acting slow effective release antiretroviral therapy. Single NM3ABC treatment of human monocyte-derived macrophages produced sustained intracellular carbovir-triphosphate and antiretroviral activities for up to 30 days.

Full text of DOI:10.1039/c8cc04708a

ChemComm
View Article Online
View Journal
COMMUNICATION
ProTide generated long-acting abacavir
nanoformulations†
Cite this:DOI: 10.1039/c8cc04708a
Zhiyi Lin,ab Nagsen Gautam,b Yazen Alnouti,b JoEllyn McMillan,a Aditya N. Bade,a
ab
and Benson Edagwa*a
Received 14th June 2018,
Accepted 2nd July 2018
Howard E. Gendelman
*
DOI: 10.1039/c8cc04708a
rsc.li/chemcomm
Abacavir pronucleotide nanoformulations (NM3ABC) were prepared demonstrated improved plasma drug stability, membrane penetra-
as a novel long acting slow effective release antiretroviral therapy. tion and encapsulation. The synthesis and antiretroviral profiles of
Single NM3ABC treatment of human monocyte-derived macrophages three ABC ProTides are described. One, M3ABC, was defined by
produced sustained intracellular carbovir-triphosphate and antiretroviral efficient encasement into nanoparticles, production of high intra-
activities for up to 30 days.
cellular carbovir-triphosphate (CBV-TP) metabolites and superior
monocyte-macrophage depot formation that enhanced antiretroviral
The translation of long acting slow effective release antiretro- activities.
viral therapy (LASER ART) from laboratory research into clinical
Prior studies demonstrated that L-alanine and L-phenylalanine
practice could improve human immunodeficiency virus (HIV) ester phosphoramidates provide potent ProTides.10,11 Thus, we
prevention and treatment and inevitably speed up viral eradication created M1ABC and M2ABC bearing alanine and phenylalanine
efforts. The advantages of LASER ART over other antiretroviral methyl ester residues (Scheme 1 and Fig. 1A). M3ABC was synthe-
regimens are defined by high antiretroviral drug (ARV) penetrance sized by substituting the methyl ester in M2ABC for a docosyl ester
into cell and tissue viral reservoirs and infrequent dosing require- promoiety. A docosanol masking ester motif was selected based
ments. Both affect treatment outcomes by maximally suppressing
viral growth through facilitated ARV entry into viral target cells.1,2
Regimen adherence can be improved by reducing the dosing
frequency. Indeed, recent data from our laboratories demonstrate
that conversion of existing ARVs into prodrugs extends their
apparent half-lives and reduces systemic drug toxicities.3–6 However,
while prodrugs represent 410% of all small molecules approved for
human use, few have entered into HIV treatment regimens and
none have appeared as part of long-acting ARV therapies.7 None-
theless, their advantages in facilitating drug biodistribution and
extending drug half-lives is already known.8
Herein, ABC prodrugs were successfully developed by PROdrug
and nucleoTIDE (ProTide) technology. While the idea was first
introduced more than two and a half decades ago, the application
of ProTide technology to abacavir (ABC) has focused on improving
drug potency9,10 not on extending its apparent half-life or on
affecting viral reservoir biodistribution. To these ends, we
implemented strategies to mask parent drug monophosphates
with cleavable hydrophobic lipids creating ProTides that
a Department of Pharmacology and Experimental Neuroscience,
University of Nebraska Medical Center, Omaha, NE 68198-5880, USA.
E-mail: hegendel@unmc.edu, benson.edagwa@unmc.edu
b Department of Pharmaceutical Science, University of Nebraska Medical Center,
Omaha, NE 68198-5880, USA
Scheme 1 Synthesis of ABC ProTides. Reagent and conditions: Step 1(a)
1-docosanol, HATU, imidazole, Et3N, DMF/CHCl3, 45 1C, 48 h; Step 1(b) Pd/C,
Et3SiH, MeOH/CHCl3 (1 : 1 v/v), rt, 16 h. Step 2(a) phenyl dichlorophosphate, Et3N,
CH2Cl2, À78 1C–rt, 16 h. Step 3(a) ABC, tert-BuMgCl, THF, À78 1C–rt, 48–90 h.
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8cc04708a
This journal is ©The Royal Society of Chemistry 2018
Chem. Commun.

View Article Online
Communication
ChemComm
and M3ABC were 177, 35650, 33135 and 541 fmol per million cells,
respectively. At 48 h, the CBV-TP levels were 33, 789, 5025 and
230 fmol per million cells for ABC, M1ABC, M2ABC and M3ABC. The
differences in active metabolite levels further suggest that the three
ProTides are rapidly taken up by MDMs but metabolized at rates
dependent on amino acid ester hydrolysis. Notably, at 48 h, the
CBV-TP levels for M3ABC declined only by 57% compared to 81, 97
and 84% in the amount of active metabolite for ABC, M1ABC and
M2ABC (Fig. S5, ESI†). We posit that enhanced lipophilicity and
slower hydrolysis of M3ABC provides controlled and extended
release of CBV-TP.
As intracellular nucleoside triphosphate levels have been
implicated in drug toxicities,19 we assessed the mitochondrial
dehydrogenase activity of the ProTides using a CCK-8 assay. As
shown in Fig. 1D, neither ABC, M1ABC, M2ABC nor M3ABC affected
cell viability at drug concentrations of up to 300 mM. However,
differences in mitochondrial activity were seen at 400 mM with
M1ABC and M2ABC compared to ABC. No significant differences
were observed between M3ABC and ABC across all tested drug
concentrations. These data indicated that the sustained release of
Fig. 1 The structure and biological characterization of ABC and its ProTides. (A)
M1ABC, M2ABC and M3ABC were synthesized based on ProTide technology.
(B) EC50 of ABC, M1ABC, M2ABC and M3ABC against HIV-1ADA were
determined in MDMs. (C) Intracellular CBV-TP levels were measured after
MDMs were treated with 10 mM free ABC, M1ABC, M2ABC or M3ABC for
various times. (D) Cytotoxicity of ABC, M1ABC, M2ABC and M3ABC in MDMs.
Results are shown as percentage of cell viability as compared to untreated
MDMs. Data are presented as mean Æ SD for n = 3 samples per group.
on its lipophilicity, antiviral activities and inherent synergistic effect an effective inhibitory concentration of CBV-TP from M3ABC was
on nucleoside analogs.12,13 ProTide hydrolysis was reasoned to not detrimental to cell viability and suggested that M3ABC could be
release two pharmacophores that inhibit viral reverse transcriptase translated for human use. Previous studies have shown that in vivo
activity.14 We reasoned that improving ABC physicochemical accumulation of nucleoside triphosphates inside cells could be
features would facilitate nanoformulation preparation, enhance facilitated by prodrugs that exhibit greater stability in plasma.20,21
intracellular ABC levels and extend the drugs’ apparent half-life. To evaluate prodrug stability, we incubated three ProTides in human
The first step in the synthesis of ABC ProTides required preparation serum. After 24 h, 460% of M1ABC and M2ABC was hydrolyzed.
of aryl amino ester phosphorochloridates from appropriate amino Nineteen percent of M3ABC, in contrast, was degraded after 24 h.
acid esters and commercially available phenyl phosphorodichlori-
LASER ART was developed, in our laboratories, to improve
date.15 Phenylalanine and alanine methyl esters were purchased therapeutic and prophylactic outcomes for HIV/AIDS. In the
while phenylalanine docosyl ester was prepared according to step 1 first phase this was accomplished by formulating myristoylated
of Scheme 1. The aryl amino ester phosphorochloridates were then prodrugs encased in a poloxamer surfactant.3–6 For ABC, such
synthesized by coupling phenyl phosphorodichloridates to amino modifications provided only low levels of CBV-TP for less than two
acid esters in the presence of triethylamine (Scheme 1 Step 2).10 weeks.5,22 We reasoned that ABC ProTide nanoformulations would
Each of the amino ester phosphorochloridates was then reacted produce higher levels of intracellular CBV-TP at substantially
with ABC in the presence of tert-butylmagnesium chloride and reduced dosing frequency. The less hydrophobic M1ABC was
purified by flash column chromatography. This generated high encapsulated into lipid-coated poly lactic-co-glycolic acid (PLGA)
yields of M1ABC, M2ABC and M3ABC (Scheme 1 Step 3, Fig. 1A). nanoparticles (NM1ABC) by a modified single emulsion solvent
Successful synthesis of ABC ProTides was confirmed by 1H, 13C and evaporation method while hydrophobic and lipophilic M2ABC
31P NMR spectra (Fig. S1–S3, ESI†). As previously reported for and M3ABC ProTides were stabilized into poloxamer coated drug
similar compounds, splitting of the peaks in the spectra of ProTides nanoparticles (NM2ABC and NM3ABC) by high-pressure homo-
was associated with the co-existence of two stereoisomers at the genization.23–25 The directive of both schemes was to produce
phosphorus atom.16
controlled release drug delivery systems. The drug loadings for the
We next evaluated the ProTides for antiretroviral activity. As NM1ABC, NM2ABC and NM3ABC nanoparticles were 3, 43, and
shown in Fig. 1B, M1ABC, M2ABC and M3ABC exhibited a 6- to 47%, respectively, underlying the differences between PLGA and
200-fold increase in antiretroviral activity over ABC in human poloxamer drug encasement systems. Indeed, unlike high-pressure
monocyte-derived macrophages (MDMs). M2ABC demonstrated the homogenization that produces nanoparticles stabilized by surfac-
lowest EC50 of 0.2 nM. The EC50 of M1ABC and M3ABC were 1.8 nM tants with high drug loading, PLGA nanoparticles exhibit low drug
and 7.0 nM, respectively. The differences were likely linked to rapid loading.23,26 Effective diameter (Deff), polydispersity index (PdI),
prodrug uptake and metabolism of M2ABC to form CBV-TP17,18 in and z-potential were determined by dynamic light scattering (DLS)
MDM. Activation of ABC ProTides requires intracellular processing (Table S1, ESI†) while transmission electron microscopy (TEM)
to generate CBV-TP.10 To better explain the differences in EC50 assessed nanoparticle morphologies (Fig. S6, ESI†). The nanoparticle
between native ProTides and ABC, MDMs were exposed to equivalent morphologies for NM1ABC and NM3ABC were spherical with
concentrations of ABC, M1ABC, M2ABC or M3ABC followed by particle sizes of 135 nm and 230 nm, respectively, while NM2ABC
quantitation of CBV-TP levels at multiple time points.9 As shown particles were rod-shaped with a size of 265 nm. Importantly,
in Fig. 1C, the maximum CBV-TP levels for ABC, M1ABC, M2ABC high drug loading in NM2ABC and NM3ABC nanoformulations
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2018

View Article Online
ChemComm
Communication
can translate into reduced dosage volumes and minimal injection NM2ABC displayed the highest CBV-TP concentration of 39 015 fmol
site reactions.25 per 106 cells at 4 h followed by a 43% decrease (22 274 fmol per
Macrophage-drug interactions are demonstrated by LASER ART 106 cells) at 8 h. Even though lower concentrations of CBV-TP were
technologies. These have been shown to improve ART pharmaco- measured for NM3ABC compared to NM1ABC or NM2ABC, the
kinetics and pharmacodynamics by facilitating cell and tissue drug active metabolite levels were sustained over 8 h, ranging from 4626
depots in the reticuloendothelial system.27–29 Macrophages are to 8601 fmol per 106 cells versus 215 and 811 fmol per 106 cells for
outstanding targets for LASER ART due to their large storage ABC. Conversion of M3ABC to CBV-TP demonstrates sustained drug
capacity, their high mobility that allows entry to sites of infection release and could lead to improved patient adherence to therapeutic
and inflammation, and their role as viral reservoirs.30,31 To affirm ARVs. To explore the potential of ProTide formulations to extend the
such observations, NM1ABC, NM2ABC and NM3ABC were adminis- apparent half-life of ABC, MDMs were exposed to formulations for
tered to MDMs at concentrations of 100 mM. The intracellular 8 h. The intracellular ProTide and CBV-TP levels were measured over
prodrug concentrations were subsequently evaluated over 8 h. As a 30-day time period. As shown in Fig. 2C, the amount of ProTide
shown in Fig. 2A, the highest drug nanoparticle MDM uptake was retained by MDM was undetectable for NM1ABC and NM2ABC,
observed with NM3ABC. At 8 h, the intracellular prodrug concen- while NM3ABC exhibited drug concentrations of 95.3 and 48.5 nmol
tration for NM3ABC was 118.5 nmol per 106 cells, a 56- and 108-fold per 106 cells at days 1 and 30, respectively. The retention of CBV-TP
higher concentration than that observed for NM1ABC (2.1 nmol per in MDMs (Fig. 2D) was sustained until day 30 for NM1ABC, NM2ABC
106 cells) or NM2ABC (1.1 nmol per 106 cells), respectively. Visualiza- and NM3ABC but only detected to day 5 for ABC. Notably, a single
tion of nanoparticles within MDMs by TEM revealed greater accu- treatment of MDMs with NM3ABC provided steady and sustained
mulation of NM3ABC nanoparticles in the cytoplasmic vesicles CBV-TP levels greater than 100 fmol per 106 cells over the entire
compared to NM1ABC and NM2ABC (Fig. 2E). These observations
likely reflect nanoparticle stability.32 To further confirm the release of
ProTides from the nanoparticles and their conversion to active
metabolites, we quantified the intracellular CBV-TP levels in the
MDMs after a single exposure to ABC, NM1ABC, NM2ABC or
NM3ABC. As shown in Fig. 2B, the CBV-TP levels for ABC increased
over time and peaked at 4 h (811 fmol per 106 cells) before declining
by 37% (510 fmol per 106 cells) at 8 h. For NM1ABC, the maximum
CBV-TP levels were observed at 8 h (39 533 fmol per 106 cells) while
Fig. 3 ABC ProTide nanoformulations protect MDMs from HIV-1 infection
over time. MDMs were treated with nanoformulations containing 100 mM ABC
Fig. 2 Uptake and retention of the nanoformulations in MDMs. (A) and (B) For
cell uptake, MDMs were treated with 100 mM ABC or prodrug nanoformulations
for 1–8 h. (B) and (D) Drug retention in MDMs was determined after an 8 h drug
loading followed by half media exchanges every other day for up to 30 days.
Intracellular drug concentrations and CBV-TP levels were determined. Data are
expressed as mean Æ SD for n = 3 samples per group. (E) Transmission electron
microscopy (TEM) was performed to visualize the morphologies of the
formulation-loaded MDMs after 8 h incubation with nanoformulations.
equivalent for 8 h. At predetermined time points, the MDMs were challenged
with HIV-1ADA. Samples were collected for antiretroviral activity assessment ten
days after viral challenge by (A) measuring HIV RT activities and (C) staining HIV-
1p24 antigen. (B) and (D) MDMs were treated with 1, 10, 25, 50 and 100 mM
NM3ABC for 8 h. At day 30 post treatment, the MDMs were challenged with
HIV-1ADA. Ten days after infection, the culture media were collected for RT assay
and the cells were fixed and stained for HIV-1p24 antigen. Data are expressed as
mean Æ SD for n = 3 samples per group.
This journal is ©The Royal Society of Chemistry 2018
Chem. Commun.

View Article Online
Communication
ChemComm
3 B. Sillman, A. N. Bade, P. K. Dash, B. Bhargavan, T. Kocher, S. Mathews,
H. Su, G. D. Kanmogne, L. Y. Poluektova and S. Gorantla, Nat. Commun.,
2018, 9, 443.
4 T. Zhou, H. Su, P. Dash, Z. Lin, B. L. D. Shetty, T. Kocher, A. Szlachetka,
B. Lamberty, H. S. Fox and L. Poluektova, Biomaterials, 2018, 151, 53–65.
5 D. Singh, J. McMillan, J. Hilaire, N. Gautam, D. Palandri, Y. Alnouti,
H. E. Gendelman and B. Edagwa, Nanomedicine, 2016, 11, 1913–1927.
30 day period. Such sustained release formulations could minimize
variant pharmacokinetic profiles and maintain effective drug con-
centrations at cellular and tissue reservoirs of infection.
To determine whether sustained intracellular CBV-TP from
ProTide formulations would translate into improved antiretroviral
activity, MDMs were challenged with HIV-1ADA for up to 30 days
after a single 8 h treatment with 100 mM of ABC equivalent. HIV-1
reverse transcriptase (RT) activity and p24 antigen expression were
assessed in infectious supernatants and adherent MDMs on day 10
post-infection. As shown in Fig. 3A and C, complete viral inhibition
was observed for up to 15 days for NM1ABC. At day 20, 91% viral
inhibition was recorded that gradually decreased to 83% inhibition
at day 30. HIV-1p24 antigen staining showed that NM2ABC
protected MDM from infection with viral breakthrough at day 30.
Of significance, full viral inhibition was observed for 30 days after
treatment with NM3ABC. In contrast, minimal protection against
viral infection was observed with native ABC at all time points.
Comparisons of antiviral efficacy of NM3ABC were then assessed to
determine the lowest drug concentration required for long-term
MDM protection against viral infection. As shown in Fig. 3B and D,
50 mM NM3ABC afforded complete viral inhibition for up to 30 days,
while a single 8 h treatment with 25 mM or 10 mM NM3ABC inhibited
viral replication by greater than 90% at day 30 post drug treatment.
These results paralleled prolonged high intracellular CBV-TP levels
for NM3ABC compared to ABC, NM1ABC or NM2ABC.
´
6 D. Guo, T. Zhou, M. Araınga, D. Palandri, N. Gautam, T. Bronich,
Y. Alnouti, J. McMillan, B. Edagwa and H. E. Gendelman, JAIDS,
J. Acquired Immune Defic. Syndr., 2017, 74, e75–e83.
7 J. Rautio, N. A. Meanwell, L. Di and M. J. Hageman, Nat. Rev. Drug
Discovery, 2018, DOI: 10.1038/nrd.2018.46 [Epub ahead of print].
8 K. M. Huttunen, H. Raunio and J. Rautio, Pharmacol. Rev., 2011, 63, 750–771.
9 J. Balzarini, S. Aquaro, A. Hassan-Abdallah, S. M. Daluge, C. F. Perno
and C. McGuigan, FEBS Lett., 2004, 573, 38–44.
10 C. McGuigan, S. A. Harris, S. M. Daluge, K. S. Gudmundsson, E. W.
McLean, T. C. Burnette, H. Marr, R. Hazen, L. D. Condreay, L. Johnson,
E. De Clercq and J. Balzarini, J. Med. Chem., 2005, 48, 3504–3515.
11 C. McGuigan, K. G. Devine, T. J. O’Connor and D. Kinchington,
Antiviral Res., 1991, 15, 255–263.
12 D. H. Katz, J. F. Marcelletti, L. E. Pope, M. H. Khalil, L. R. Katz and
R. McFadden, Ann. N. Y. Acad. Sci., 1994, 724, 472–488.
13 J. F. Marcelletti, Antiviral Res., 2002, 56, 153–166.
14 C. Piantadosi, C. J. Marasco, Jr., S. L. Morris-Natschke, K. L. Meyer,
F. Gumus, J. R. Surles, K. S. Ishaq, L. S. Kucera, N. Iyer and C. A. Wallen,
et al., J. Med. Chem., 1991, 34, 1408–1414.
15 D. Cahard, C. McGuigan and J. Balzarini, Mini-Rev. Med. Chem.,
2004, 4, 371–381.
16 S. Kandil, J. Balzarini, S. Rat, A. Brancale, A. D. Westwell and
C. McGuigan, Bioorg. Med. Chem. Lett., 2016, 26, 5618–5623.
17 R. D. Hanson, P. A. Hohn, N. C. Popescu and T. J. Ley, Proc. Natl.
Acad. Sci. U. S. A., 1990, 87, 960–963.
18 G. Birkus, N. Kutty, G. X. He, A. Mulato, W. Lee, M. McDermott and
T. Cihlar, Mol. Pharmacol., 2008, 74, 92–100.
A pilot in vivo study in rats was conducted with NM3ABC.
Peripheral blood mononuclear cells were recovered from whole
blood and the intracellular CBV-TP levels measured at day 7
(Table S2, ESI†). Reduced CBV-TP levels are explained, in part, by the
rapid degradation of ProTides or slow cleavage of phosphoramide
intermediates in rodents that precedes triphosphate formation.33
In conclusion, we posit that LASER ART can overcome
challenges of ARV adherence and biodistribution. In support of
this idea, M1ABC, M2ABC and M3ABC ProTides were synthesized.
NM3ABC nanoparticles exhibited improved MDM drug uptake,
sustained retention and antiretroviral activities for up to one month.
We posit that this work is a significant step forward for the
management and prevention of HIV-1 infection.
19 T. N. Kakuda, Clin. Ther., 2000, 22, 685–708.
20 W. A. Lee, G. X. He, E. Eisenberg, T. Cihlar, S. Swaminathan, A. Mulato
and K. C. Cundy, Antimicrob. Agents Chemother., 2005, 49, 1898–1906.
21 R. L. Mackman, A. S. Ray, H. C. Hui, L. Zhang, G. Birkus, C. G. Boojamra,
M. C. Desai, J. L. Douglas, Y. Gao, D. Grant, G. Laflamme, K. Y. Lin, D. Y.
Markevitch, R. Mishra, M. McDermott, R. Pakdaman, O. V. Petrakovsky,
J. E. Vela and T. Cihlar, Bioorg. Med. Chem., 2010, 18, 3606–3617.
22 N. Gautam, Z. Lin, M. G. Banoub, N. A. Smith, A. Maayah, J. McMillan,
H. E. Gendelman and Y. Alnouti, J. Pharm. Biomed. Anal., 2018, 153,
248–259.
23 T. Zhou, Z. Lin, P. Puligujja, D. Palandri, J. Hilaire, M. Arainga,
N. Smith, N. Gautam, J. McMillan, Y. Alnouti, X. Liu, B. Edagwa and
H. E. Gendelman, Nanomedicine, 2018, 13, 871–885.
24 Y. Liu, K. Li, J. Pan, B. Liu and S. S. Feng, Biomaterials, 2010, 31, 330–338.
25 Z. Lin and H. E. Gendelman, in Encyclopedia of AIDS, ed. T. J. Hope,
M. Stevenson and D. Richman, Springer New York, New York, NY,
2016, pp. 1–10, DOI: 10.1007/978-1-4614-9610-6_220-1.
26 J. Xu, S. Zhang, A. Machado, S. Lecommandoux, O. Sandre, F. Gu
and A. Colin, Sci. Rep., 2017, 7, 4794.
The research was supported by the University of Nebraska
Foundation and the National Institutes of Health grants R01
MH104147, P01 DA028555, R01 NS36126, P01 NS31492, 2R01
NS034239, P01 MH64570, P30 MH062261, P30 AI078498, and
R01 AG043540.
27 D. P. Gnanadhas, P. K. Dash, B. Sillman, A. N. Bade, Z. Lin,
D. L. Palandri, N. Gautam, Y. Alnouti, H. A. Gelbard, J. McMillan,
R. L. Mosley, B. Edagwa, H. E. Gendelman and S. Gorantla, J. Clin.
Invest., 2017, 127, 857–873.
28 P. Puligujja, S. S. Balkundi, L. M. Kendrick, H. M. Baldridge, J. R. Hilaire,
A. N. Bade, P. K. Dash, G. Zhang, L. Y. Poluektova, S. Gorantla, X. M. Liu,
T. Ying, Y. Feng, Y. Wang, D. S. Dimitrov, J. M. McMillan and
H. E. Gendelman, Biomaterials, 2015, 41, 141–150.
29 B. J. Edagwa and H. E. Gendelman, Nat. Mater., 2018, 17, 114–116.
30 M. Arainga, B. Edagwa, R. L. Mosley, L. Y. Poluektova, S. Gorantla
and H. E. Gendelman, Retrovirology, 2017, 14, 17.
Conflicts of interest
There are no conflicts to declare.
31 M. Arainga, D. Guo, J. Wiederin, P. Ciborowski, J. McMillan and
H. E. Gendelman, Retrovirology, 2015, 12, 5.
32 S. Kumar, A. C. Anselmo, A. Banerjee, M. Zakrewsky and S. Mitragotri,
J. Controlled Release, 2015, 220, 141–148.
33 M. Slusarczyk, M. H. Lopez, J. Balzarini, M. Mason, W. G. Jiang,
S. Blagden, E. Thompson, E. Ghazaly and C. McGuigan, J. Med.
Chem., 2014, 57, 1531–1542.
Notes and references
1 B. Edagwa, J. McMillan, B. Sillman and H. E. Gendelman, Expert
Opin. Drug Delivery, 2017, 14, 1281–1291.
2 R. J. Landovitz, R. Kofron and M. McCauley, Curr. Opin. HIV AIDS,
2016, 11, 122–128.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2018