Nanotechnology Applications for Improved Delivery of Antiretroviral Drugs To

P1

2

3

4Q15

6

7891011121314151617181920 Blood-brain barrier21 ATP-binding cassette membrane transporters

22

23

24

25

26

27

28osomes, solid lipid nanoparticles (SLN) and micelles can increase the local drug29cilitate drug transport into the brain via endocytotic pathways and inhibit the30

31signicant increase in the drug bioavailability to32

33

34

35

3637

38

39

4041

42 . . . .43 virus (H44 . . . .45 ith HIV46

47

48

49

50

51

52

53

54

55

56

57

Advanced Drug Delivery Reviews xxx (2009) xxxxxx

ADR-11953; No of Pages 14

Contents lists available at ScienceDirect

Advanced Drug Delivery Reviews

j ourna l homepage: www.e lsev ie r.com/ locate /addr

ARTICLE IN PRESSCOR2. Barriers to antiretroviral (ARV) penetration into the brain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 02.1. BBB and BCSFB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

2.2. ABC transporters at the BBB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 02.2.1. ABC transporters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 02.2.2. Role of ABC-transporters in ARVs delivery to the brain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

2.3. Strategies to improve ARVs penetration across the BBB and BCFSB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 02.3.1. Inhibition of ABC transporters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 02.3.2. Hyper-osmotic opening of the BBB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 02.3.3. Pharmacological disruption of BBB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0UNAbbreviations: ABC transporter, ATP-binding cassettBBB, blood-brain barrier; BCSFB, blood-cerebro spinal antiretroviral therapy; HAD, human immunodeciency vHIVE, human immunodeciency virus encephalitis; LDcrylate; MRP, multidrug resistance-associated proteins;cyanoacryalate); PEG, polyethylene glycol; PIs, HIV prglycolide); SLN, solid lipid nanoparticles; Tat, transcript This review is part of the Advanced Drug Delivery Re Corresponding author. Tel.: +1 416 978 6979.

E-mail address: [email protected] (R. Benday

0169-409X/$ see front matter 2009 Published by Edoi:10.1016/j.addr.2009.11.020

Please cite this article as: H.L. Wong, et al., NDeliv. Rev. (2009), doi:10.1016/j.addr.2009R. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 01.3.1. Pathophysiology . . . .1.3.2. Clinical manifestations .1.3.3. Current treatment and itsContents

1. HIV infection and CNS illnesses .1.1. Human immunodeciency1.2. HIV epidemiology . . . .1.3. Complications associated w EC

TEDthat the specicity and efciency of ARVs delivery can be further enhanced by using nanocarriers withspecic brain targeting, cell penetrating ligands or ABC-transporters inhibitors. Future research should focus

on achieving brain delivery of ARVs in a safe, efcient, and yet cost-effective manner. 2009 Published by Elsevier B.V.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0IV) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0infection of the central nervous system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0e membrane transporter; AIDS, acquired immunodecienuid barrier; CD4, cluster of differentiation 4; CNS, centrirus-associated dementia; hCMEC/D3, human brain micrL, low-density lipoprotein; MCMD, minor cognitive/moNNRT, non-nucleoside reverse transcriptase inhibitors; Notease inhibitors; PIL, PEGylated immunoliposomes; P-ional activator; Vpr, viral protein R.views theme issue on Nanotechnology Solutions for Inf

an).

lsevier B.V.

anotechnology applications for improved del.11.020the brain is expected to be achieved. Recent studies showAntiretroviralNanotechnology ATP-binding cassette (ABC) transporters expressed at the barrier sites. By delivering ARVs with nanocarriers,Brain delivery polymeric nanoparticles, lipconcentration gradients, faKeywords:Human immunodeciency virus (BBB) and blood-cerebrospROOF

Nanotechnology applications for improved delivery of antiretroviral drugs tothe brain

Ho Lun Wong a, Niladri Chattopadhyay b, Xiao Yu Wu b, Reina Bendayan b,a School of Pharmacy, Temple University, 3307 North Broad Street, Philadelphia, Pennsylvania, US 19140, USAb Leslie Dan Faculty of Pharmacy, University of Toronto, 144 College Street, Toronto, Ontario, Canada M5S 2S2

a b s t r a c ta r t i c l e i n f o

Article history:Received 22 June 2009Accepted 14 September 2009Available online xxxx

Human immunodeciency virus (HIV) can gain access to the central nervous system during the early courseof primary infection. Once in the brain compartment the virus actively replicates to form an independentviral reservoir, resulting in debilitating neurological complications, latent infection and drug resistance.Current antiretroviral drugs (ARVs) often fail to effectively reduce the HIV viral load in the brain. This, in part,is due to the poor transport of many ARVs, in particular protease inhibitors, across the blood-brain barrier

inal uid barrier (BCSBF). Studies have shown that nanocarriers includingcy syndrome; apoE, apolipoprotein E; ARVs, antiretroviral drugs;al nervous system; CSF, cerebrospinal uid; HAART, highly activeovessel endothelial cell line; HIV, human immunodeciency virus;tor disorder; MMA-SPM, methylmethacrylate-sulfopropylmetha-RTI, nucleoside reverse transcriptase inhibitor; PBCA, poly(butylgp, P-glycoprotein; PLA, polylactide; PLGA, poly(D,L-lactide-co-

ectious Diseases in Developing Nations.

ivery of antiretroviral drugs to the brain, Adv. Drug

58 .59 .60 .61 .62 .63 .64 ss t65 .66 .67 .68 .69 .70 in t71 .72 .73 eliv74 .75 .76 .77 .78 .79 .80 5.3. Use of better experimental models . . . . . . . . . . . . . .81 5.4. Rene targets and endpoints of ARV delivery . . . . . . . . .82 .

83

84

85

86

87

88

89

90

91

92

93

94 nity [2,3]. As a result, HIV-infected patients are substantially more95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128circulating antibody levels are low (e.g. in brain) [6,7]. Once the virus129fuses with the host cell, DNA is produced from its RNA genome via the130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

2 H.L. Wong et al. / Advanced Drug Delivery Reviews xxx (2009) xxxxxx

ARTICLE IN PRESSUNCORRECvulnerable to opportunistic infections. The abnormal immuneresponses triggered by HIV infection can also result in othercomplications such as neurological illnesses.

1.2. HIV epidemiology

According to the 2007 update by the Joint United Nations Programon HIV/AIDS and World Health Organization [4], every day over 6800individuals become newly infected and over 5700 patients die fromAIDS. It is estimated that 33.2 million persons worldwide are infectedwith HIV-1, and the developing nations continue to be its primaryvictims. Although signs of a decline in the cases of new infection havebeen observed due to better prevention efforts, the sub-SaharanAfrican region remains as the epicenter of the pandemic. An estimated22.5 million people in these countries, equivalent to 5% prevalence,are living with HIV-1 infection. The prevalence is also alarmingly highin the Caribbean Islands (1.0%), Latin America (0.5%), Eastern Europeand Central Asia (0.9%). In East Asia, 92,000 adults and children werefound newly infected with HIV-1 in 2007, representing almost a 20%increase from 2001. Even though the numbers of newHIV-1 infectionshave been relatively stable in the developed nations, this disease isstill an unresolved health issue. In North America only, 1.3 millionpeople are living with HIV-1, equivalent to 0.6% prevalence. Thesedata indicate that the current treatment of HIV-1 still needs signicant

enzyme reverse transcriptase, and then DNA is incorporated into thehost's genome by an integrase enzyme and replicates as a part of thehost DNA [1,8].

HIV is known to invade the central nervous system (CNS) early in thecourse of the infection and primarily targets brain mononuclearmacrophages, perivascular macrophages and microglia. [7,9]. The viruscan enter the CNS compartment from the systemic circulation via tworoutes: i) through the blood-cerebro spinal uid barrier (BCSFB) at thechoroid plexus as cell-free viral particles [10], and/or ii) through theblood-brainbarrier (BBB) in formof infectedmonocytes [11]. The secondroute is known as the Trojan horse approach. In brief, monocytesinfected by HIV-1 are able to cross the BBB between the capillaryendothelial cells in a complex process regulated by the secretion ofchemokines (e.g. MIP-1a/b, MCP-1, RANTES) from glial cells [12]. Thebrain macrophages and microglial cells, upon infection are responsiblefor further production of HIV-1 virus, and can also release viral proteinssuchas glycoprotein 120 (gp120), Tat (transcriptional activator) andVpr(viral protein R) [1316]. These viral proteins have been shown to beneurotoxic in vitro and trigger various harmful events such as activationof apoptotic pathways, cell-cycle arrest of neuronal cells and stimulationof the production of reactive oxidative species, glutamate, cytokines andother inammatory factors from uninfected astrocytes [1719], whichfurther accelerate the neurodegeneration process. Additionally, gp120and Tat can render the BBB leakier which further promotes thepermeability of HIV-infected monocytes [2022]. Other CNS cell typesReferences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. HIV infection and CNS illnesses

1.1. Human immunodeciency virus (HIV)

Human immunodeciency virus (HIV) is a lentivirus from theRetroviridae family responsible for the acquired immunodeciencysyndrome (AIDS). At present, there are two known types of HIV, HIV-1and HIV-2, with HIV-1 being much more virulent, transmittable andprevalent, and the cause of the majority of HIV infections in the world[1]. HIV infection results in compromised immune defense by causingextensive destruction of T-helper cells, macrophages, dendritic cellsand other cellular components associated with cell-mediated immu-2.3.4. Drug modication approach . . . . . . . . . . . .2.3.5. Focused ultrasound and microbubble approach . . .2.3.6. Nanotechnology for ARVs delivery to the brain . . .

3. Nanotechnology to improve ARVs delivery to the brain . . . . . . .3.1. General principles of brain delivery using nanocarriers . . . .

3.1.1. Rationale . . . . . . . . . . . . . . . . . . . . .3.1.2. Overview of nanocarrier-mediated drug delivery acro

3.2. Current use of nanocarriers for brain delivery of ARVs . . . .3.2.1. Polymer or dendrimer-based nanocarriers . . . . .3.2.2. Lipid-based nanocarriers . . . . . . . . . . . . . .3.2.3. Micelle-based nanocarriers . . . . . . . . . . . . .

4. Recent trends to optimize nanotechnology use for ARV brain delivery4.1. Optimization of nanocarrier properties to improve passive bra4.2. Development of specic brain targeting strategies . . . . . .4.3. Cell penetrating peptides . . . . . . . . . . . . . . . . . .4.4. Other nanotechnology-based strategies to improve ARVs brain d

4.4.1. Use of macrophages for BBB passage . . . . . . . .4.4.2. Alternative route for nanocarriers administration . .4.4.3. Advanced delivery of ABC-transporter blockers . . .

5. Future perspectives and conclusion . . . . . . . . . . . . . . . .5.1. Improve understanding of the barrier structures . . . . . . .5.2. Tailor-made nano-formulations for brain delivery . . . . . .improvement.

Please cite this article as: H.L. Wong, et al., Nanotechnology applicationsDeliv. Rev. (2009), doi:10.1016/j.addr.2009.11.020TEDPROOF

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0he BBB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0argeting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0ery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

1.3. Complications associated with HIV infection of the central nervoussystem

1.3.1. PathophysiologyHIV and other lentiviruses are unique from other viruses due to

their ability to infect and replicate in non dividing cells includingthose of the monocyte/macrophage lineage. In particular, HIV targetsthe cluster of differentiation 4 positive (CD4+) T lymphocytes andcells of the monocyte-macrophage lineage [5]. CD4 negative cellsmay also be targeted, but these viral strains are highly sensitive toneutralization by host antibodies and are present only at sites where155can also be infected by HIV. Low-grade production of provirus has been

for improved delivery of antiretroviral drugs to the brain, Adv. Drug

156 detected in some cell populations such as astrocytes, and in vitro HIV157 susceptibility in oligodendrocytes and microvascular endothelial cells158 has been observed [23]. Non-CD4 entry pathway(s) may play a role as159 these cells do not express CD4 on their surface [24].

160 1.3.2. Clinical manifestations161 CNS infection by HIV leads to various forms of neurological162 complications. It has been estimated from AIDS patients brain tissue163 obtained at autopsy that the prevalence of neuropathologic abnormal-164 ities can be as high as over 80% [25]. Approximately 10 to 20% treated165 patients demonstrate some forms of overt illnesses [26]. Minor166 cognitive/motor disorder (MCMD), which presents with symptoms167 such as cognitive and motor slowing, poor concentration and impaired168 memory, often occurs during early HIV infection [27]. During the late-169 stage of the infection, a more severe form of neurological complications170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204It has been reported that HAART is, in general, less effective for the205treatment of CNS complications than other AIDS-related illnesses [33].206In the short term, HAART remains fairly effective against the more207severe CNS illnesses such as HAD. Although, the incidence of HAD has208been reduced from over 30% of the AIDS population to around 10% in209post-HAART era [34,35], this is accompanied by a signicant increase in210theprevalenceofHAD since patients now live longer. Relapses ofHAD in211HAART-treated patients are also common [26,35]. The therapeutic value212of HAART for other HIV-related CNS complications is lower. Clinical data213indicate that since theadventofHAART, therehas beena steady increase214in theMCMD incidence [34]. It has been reported that the rate ofMCMD215among adults with symptomatic HIV-1 disease is over 30%, and the216prevalence of MCMD did not signicantly change when the pre-HAART217and post-HAART cohorts of HIV-1 infected homosexual males were218compared. Overall, the data suggest a lack of protection against MCMD219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

t1:1

t1:2t1:3

t1:4 NtRt1:5 fcat1:6 dardt1:7

t1:8 are;t1:9 stat1:10

t1:11

t1:12 pint1:13 fca

e ino in

t1:14

3H.L. Wong et al. / Advanced Drug Delivery Reviews xxx (2009) xxxxxx

ARTICLE IN PRESSUNCORREC

collectively termed HIV-associated dementia (HAD) or AIDS dementiacomplex may develop, usually in patients who have had a CD4 countnadir of b200 cells/mm3 [16,28]. Patients with HAD may present withdiverse symptoms ranging from confusion, behavioral abnormalities,motor dysfunctions, to psychosis and seizure. Without proper treat-ment, the mental conditions of HAD patients can further deteriorate. In5% to 8% of patients, a syndrome known as AIDS mania develops inaddition to HAD [29]. Pediatric HIV patients are particularly vulnerableto HAD. About 50% of untreated children infected with HIV-1 wasestimated to have HAD, and the symptoms are often more severe, withmany of them showing compromised intellectual development [11].Overall, the prognosis of advanced HAD patients is poor, and eventhough less severe, MCMD has been identied as a signicantindependent risk factor for AIDS mortality [27].

1.3.3. Current treatment and its limitationsZidovudine was the rst antiretroviral compound commercially

available for the treatment of AIDS [30]. Since its launch in 1987,intensive research efforts have led to the discovery of several classesof ARVs including: i) nucleoside reverse transcriptase inhibitors(NRTIs), ii) non-nucleoside reverse transcriptase inhibitors (NNRTIs),iii) protease inhibitors (PIs), iv) integrase inhibitors, and v) entryinhibitors [31]. With better understanding of the detailed mechan-isms of HIV replication, single agent ARV therapy has been generallyreplaced by combination therapy. The primary rationale for usingmultiple agents is to disrupt HIV replication at multiple points in thelifecycle. Each of these cocktail regimens often comprises twonucleoside analogues and a PI to achieve potential synergistic effect,sometimes with a secondary PI at low dose (typically ritonavir)included to boost up the bioavailability of the primary PI. Because oftheir higher clinical efcacy in lowering the mortality and morbidityin HIV patients, these therapeutic combinations are referred as thehighly active ARV therapy (HAART). Table 1 lists the major ARVcomponents of HAART recommended by the US Department of Healthand Human Services and International AIDS Society in 2008 [31,32].

Table 1Recommended ARVs in highly active antiretroviral therapy (HAART).

Class of antiretroviral Component Comments

NRTI/NtRTI Tenofovir+emtricitabine Tenofovir is aAbacavir+lamivudine Comparable eZidovudine+lamivudine Previous stanStavudine+lamivudine

NNRTI Efavirenz Standard of cPI Lopinavir Lopinavir new

AtazanavirFosamprenavirDarunavir Superior to loSaquinavir Comparable e

NRTI: nucleoside reverse transcriptase inhibitors; NtRTI: nucleotide reverse transcriptas All PIs require ritonavir-boosting, i.e. co-administration with low dose of ritonavir, t

from Hammer et al, 2008 [32]).

Please cite this article as: H.L. Wong, et al., Nanotechnology applicationsDeliv. Rev. (2009), doi:10.1016/j.addr.2009.11.020TEDPROOFprovided by HAART [26]. Studies also showed that low-gradeinammation frequently persists in patients receiving HAART, suggest-ing the occurrence of ongoing immune activation in the CNS [36,37].This limited efcacy is in part attributable to the inefciency of the

current HAART regimens to eradicate the HIV-1 in the CNS. Thisphenomenonbears a number of therapeutic and pathologic implications.Some studies show that in HIV-1 infection, the severity of neurologicalcomplications is positively correlated with the viral load in thecerebrospinal uid (CSF), and this CSF viral load has been shown fairlyindependent from the plasma viral load [38,39]. Therefore, ARVs willhave little therapeutic value for HIV-related CNS complications unlessthese drugs can efciently reduce the viral load in the CNS compartment.Failure to eliminate such a viral reservoir in the CNS, whereinreplicating virus accumulate and survive with more stable kineticproperties than in the main pool of virus in the peripheral compartment[40], could be responsible for the latent reinfection. In addition, thepresence of CNS viral reservoir may also promote the development ofdrug resistance. Wong et al have shown the presence of different ARVresistance mutations in viruses from the brain as compared with theperipheral sites of infection [39]. Unfortunately, using the current HAARTregimens, most recent studies estimate that it will take up to 7.7 years ofuninterrupted therapy to eliminate this viral reservoir early in the courseof HIV infection [41]. Such long-term use of HAART is highly undesirable.The risk of side effects, including peripheral neuropathy, liver dysfunc-tions, and metabolic complications will signicantly increase, and otherissues such as poor patient compliance, high drug cost, and drug-druginteractions are more likely to arise [42,43].

2. Barriers to antiretroviral (ARV) penetration into the brain

2.1. BBB and BCSFB

To effectively treat HIV-associated CNS complications, it is highlycritical to improve the efciency of CNS penetration by ARVs. Anumber of obstacles have to be overcome to achieve this goal. As

TI. Effective and well-tolerated; new standard of NRTI/NtRTI components in HAARTcy to tenofovir+emtricitabine in patients with low to moderate viral loads of NRTI components

available as a once-daily xed dose with tenofovir+emtricitabinendard of care. Atazanavir and fosamprenavir have comparable efcacy

avir in patients with viral load 100,000 HIV RNA copies/mLcy, but more frequent dosing required

hibitor; NNRTI: non-nucleoside reverse transcriptase inhibitors; PI: protease inhibitors.crease the plasma concentration and area-under-the-curve of the ARVs. (data obtainedfor improved delivery of antiretroviral drugs to the brain, Adv. Drug

C251 reected by the pharmacokinetic parameters (shown in Table 2),252 ARVs often exhibit high non-specic binding, and do not reside long in253 the blood plasma [4447]. The share of the administered dose of the254 drug that can reach the brain is consequently quite limited. The CNS255 penetration by ARVs is further compromised by the presence of the256 BBB and BCSFB [48]. This is reected by the low values of CSF-plasma257 ratio observed in most ARVs, especially the PIs (Table 2).258 Both BBB and BCSFB are equipped with specialized anatomical259 structures which dramatically prevent access of several exogenous260 compounds to the CNS compartment [49,50]. The BBB is formedmainly261 by the brain capillary endothelium. It serves as the primary interface262 between the CNS and the peripheral circulation, separating the brain263 parenchyma from the bloodstream [51,52]. There are a number of264 structural features unique to brain capillaries when compared to other265 blood capillaries such as a lack of fenestration and minor pinocytosis.266 The tight junctions between these cells are extensive, continuous and267 provide a very high electrical resistance (over 1500 cm2) across the268 brain capillary endothelium. Together, these properties signicantly269 limit the paracellular transport of hydrophilic molecules [53]. To allow270 access of the brain by hydrophilic compounds such as glucose, amino271 acids and proteins that are essential for brain functioning, specic cell272 surface processes such as solute transporters and receptor-mediated273 endocytosis are present at this barrier [50,51].274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

t2:1

t2:2t2:3

t2:4

t2:5

t2:6

t2:7

t2:8

t2:9

t2:10

t2:11

t2:12

t2:13

t2:14

t2:15

t2:16

t2:17

t2:18

t2:19

t2:20

t2:21

t2:22

t2:23


ARTICLE IN PRESSUNCORRE

The BCSFB serves as the secondary barrier against drug penetra-tion into the CNS. It is formed by the choroid plexus epithelial cellswhich like the brain microvessel endothelial cells present tightjunctions [49,53]. These epithelial cells have well-developed apicalbrush border and basolateral interdigitations, as well as numerousmitochondria, all of which may be important in uid and solutetransport [50]. The choroid plexus consists of a single layer of cuboidalepithelial cells surrounding a rich vascular network from which theepithelial cells are separated by a loose stroma. CSF secretion is about350 l/min in an adult male which represents a turnover rate of about0.4%/min [54]. The production of CSF serves to keep the concentrationof compounds that passively diffuse into the brain lower than thatfound in plasma, a phenomenon known as the CSF sink effect [54].

The penetration of a drug molecule across these two barriersdepends on a number of its physicochemical properties includinglipophilicity, size, and degree of ionization, Although in principle,uncharged lipophilic compounds smaller than 400600 Da shouldeasily cross these barriers by passive diffusion across the cellmembranes, many of these compounds are restricted from access into

Table 2Clinical pharmacokinetic properties of commonly prescribed ARVs.

Antiretroviraldrugs

Plasma proteinbinding (%)

Eliminationhalf-life (h)

CSF-plasma ratio

Abacavir 50 1.54 0.18-0.33Didanosine b5 1.5 0.21Lamivudine b36 1.42 0.060.31Emtricitabine b4 810 UnknownStavudine Negligible 1.21.4 0.2Zalcitabline Negligible 1.22 0.10.37Zidovudine b38 1.1 0.170.60Delavirdine 98 5.8 0.004Efavirenz 96-99 52-76 0.02-0.1Nevirapine 60 45 0.45Amprenavir 90 7.110.6 b0.01Atazanavir 86 5.28 0.00210.0226Indinavir 60 1.8 0.15Lopinavir 98-99 5.6 NegligibleNelnavir N98 3.5-5 NegligibleRitonavir 98-99 3-5 0.010.05Saquinavir 97 2.5 0.010.02

Zalcitabine was discontinued in 2006. Amprenavir was discontinued in 2004; a prodrug version (fosamprenavir) iscurrently available.Data obtained from: Oldeld and Plosker, 2006; Perry et al., 2005; Swainston and Scott,

2005; Wynn et al., 2002. [4447].


theCNS. For examples, even thoughPIs typically exhibit a highdegree oflipophilicity (log10P = 2.95.2) [55], their CSF levels are extremely low(Table 2). While the large (typically 600750 Da) molecular weight ofthese PIs likely contributes to their low CNS bioavailability, it must bepointed out that smaller, neutral PIs such as amprenavir (505.6 Da) stillexhibit a poor CSF-plasma ratio. It is obvious that in addition to theanatomical features, the BBB and BCSFB have other mechanisms tocontrol passage of drugs into the CNS. Indeed, there are severalmembrane transporters located at these two barriers which mediatetheir efux from the CNS compartment back to the blood. Most of thesetransporters belong to the superfamily of ATP-binding Cassette (ABC)membrane transporters [52]. Furthermore, cerebral blood ow anddegree of local inammation can also affect drug CNS permeability.

2.2. ABC transporters at the BBB

2.2.1. ABC transportersThe ABC family is among the most ubiquitously expressed and

largest membrane-associated protein superfamily known to date. ABCmembers are involved in the translocation of both endogenous andexogenous substrates and metabolites against their concentrationgradient [56]. The energy to transport substrates is provided byhydrolysis of ATP at the nucleotide binding domains. In humans, 50ABC genes have been identied and are classied according to sevensubfamilies based on the organization and sequence of their ATP-binding domain(s) [57]. ABC-transporters can also be classied intofull or half transporters. A full ABC-transporter consists of twotransmembrane domains and two ATP-binding domains, whereas ahalf transporter consists of only one of each [58].

Several ABC transporters, specically P-glycoprotein (P-gp), multi-drug resistance-associated proteins (MRP) isoforms, and ABCG2 areknown to be involved in the cellular extrusion of a broad range of drugmolecules. P-gp, a full transporter also known as MDR1 protein, ABCB1or CD243, is probably themost studied and characterized ABCmember.It was rst found as a 170-kDa ATP dependent membrane glycoproteinthat acts as a drug efuxpump [59]. Up todate, a diversity of structurallyand functionally distinct biomolecules and chemical compounds havebeen identied as the substrates of P-gp [60]. These substrates aretypically hydrophobic, amphipathic compounds, with many of thempossessing weakly basic or cationic groups. MRPs belong to the ABCCsubfamily [60,61]. Multiple isoforms of MRP (MRP-1 to MRP-8) havebeen discovered [62]. They are full transporters normally expressed inthe canalicular part of the hepatocyte where they play a crucial role inbiliary transport [63]. Many MRP substrates are also lipophilic, basic orcationic compounds, but unlike P-gp, MRP substrates also includeneutral or mildly anionic molecules [63,64] . Many of these substratesare drug conjugates of lipophilic anions (e.g. glucuronate or glutathioneconjugates). ABCG2, also known as breast cancer resistance protein ormitoxantrone-resistance protein, is a half ABC-transporter rst identi-ed in breast cancer [65]. It was later found in other normal tissues,including placenta, liver canaliculi, small intestine, colon, the bronchialepithelial layer in the lung, and brain capillary endothelial cells [66].Many P-gp substrates are also substrates of ABCG2.

2.2.2. Role of ABC-transporters in ARVs delivery to the brainMany ARVs are large, lipophilic compounds with fairly high

molecular weights. They are therefore likely candidates to be ABCtransporter substrates. Several studies in polarized and P-gp over-expressing cell lines have indeed shown that various PIs (e.g.amprenavir, indinavir, lopinavir, nelnavir, ritonavir, saquinavir) aresubstrates of P-gp [6771]. As shown by a number of in vitro studies,many PIs are also substrates of MRPs, particularly MRP1 and MRP2[67,68,72,73] . Furthermore, PIs have been shown to be competitiveblockers of P-gp and MRPs, as shown by their inhibitory activitiesagainst the transport of established P-gp and MRP substrates [7479].355Studies in human embryonic kidney cells and in MDCKII cells stably


F356 transfected with ABCG2 have shown that while PIs are probably not357 ABCG2 substrates, they are good inhibitors of ABCG2-mediated358 mitoxantrone or pheophorbide A transport [80,81]. A few recent359 studies have further demonstrated that NRTIs and NNRTIs are also360 capable of interacting with ABC membrane transporters. For361 instances, delavirdine is a P-gp blocker, whereas abacavir and362 stavudine are substrates of MRP4 and MRP5, respectively [8284].363 There are other lines of evidence supporting the frequent interactions364 of ABC transporters and ARV compounds. For further detail, please365 refer to the reviews by Dallas et al and Ronaldson et al [62,85].366 P-gp, MRP isoforms and ABCG2 have been identied at the BBB and367 BCFSB and found to play a signicant role in regulating the levels of368 ARVs, most notably PIs, in the brain compartment [85]. Fig. 1 presents369 the proposed distribution of these transporters at the BBB and BCSFB.370 Studies in isolated animal brain microvessels have shown the potent371 inhibition of P-gp and MRP-mediated transport by saquinavir and372 ritonavir [86]. Our group has also shown that indinavir, saquinavir and373 ritonavir can inhibit the accumulation of digoxin, a P-gp substrate, in374 an immortalized rat astrocyte cell line system (CTX TNA2), a rat brain375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398after treatment of mice with a P-gp inhibitor LY-335979 [89]. Multiple-399fold increases in the brain accumulation of saquinavir were similarly400observed in animals treated with GF120918 and MK571, a P-gp/ABCG2401blocker and MRP inhibitor, respectively [90]. In another study, a 9-fold402increase in nelnavir concentration in the brain of mdr1a/1b (+/+)403wild-type mice versus P-gp knockout mdr1a/b (/) mice was404obtainedwhenGF120918was co-administered [91]. However, it should405be noted that the nelnavir levels in other vital organs were also406increased by multiple-fold. Considering the ubiquitous presence of ABC407transporters, these blockers are unlikely to be CNS-specic. The408resulting overhaul of the global pharmacokinetics is generally undesir-409able as it can easily lead to elevated risks of drug toxicity and410unpredictable drug interactions.

4112.3.2. Hyper-osmotic opening of the BBB412It is known that an hypertonic solution of mannitol or urea can413shrink the capillary endothelial cells by inducing water efux and414subsequently opening the tight junction network momentarily [92,93].415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

rs at



microvessel endothelial cell line (RBE4) and primary cultures of ratastrocytes [75,87]. Strong interactions between PIs and P-gp andMRPs were observed. Further evidence is provided from in vivostudies using mdr1a (/) knockout mice. Signicant enhancement(436 fold) in brain accumulation of indinavir, nelnavir, ritonavir,and saquinavir has been shown in the knockout mice compared to themdr1a (+/+) wild-type controls [71]. In comparison, the role ofABCG2 in BBB transport of ARVs is more controversial. Studies usingABCG2-decient mice indicate that ABCG2 does not play a signicantrole in limiting the CNS distribution of zidovudine and abacavir [88].Further work is required to clarify the role of this transporter in vivo.

2.3. Strategies to improve ARVs penetration across the BBB and BCFSB

Knowing the anatomical and molecular characteristics of the BBBand BCFSB, a number of strategies have been proposed to improve thepermeability of ARVs across these barriers. Each of these strategiespresents strengths and limitations.

2.3.1. Inhibition of ABC transportersIn the last decade, a number of chemical entities capable of

blocking specic ABC-transporters have been developed. Somesuccess in using these blockers to improve ARVs availability to theCNS has been shown in various animal studies using different types ofblockers. For example, the CNS levels of several PIs (i.e., amprenavir,indinavir, nelnavir, saquinavir) in mice were signicantly enhanced

Fig. 1. Proposed localization of major ATP-binding cassette (ABC) membrane transporte

side; BL: basolateral side; CSF: cerebrospinal uid. (Adapted from Ronaldson et al., 2008).

Please cite this article as: H.L. Wong, et al., Nanotechnology applicationsDeliv. Rev. (2009), doi:10.1016/j.addr.2009.11.020TEDPROO

As a result, the paracellular ow can be considerably increased allowingmore efcient BBB passage by drug compounds. This strategy has beenapplied in animals with some success for increasing the BBBpermeability [92]. Unfortunately, it is a risky procedure. Seizures wereobserved in some subjects and unpredictable long-term neurologicalcomplications can occur. This strategy is therefore reserved as the lastresort [94].

2.3.3. Pharmacological disruption of BBBCytotoxic agents, especially alkylating agents such as etoposide

and cisplatin, may disrupt tight junctions and create openingsbetween the endothelial cells [95]. Similarly, vasoactive agents suchas bradykinin, peptidase inhibitors and angiotensin II may also renderthe BBB permeable temporarily [9698]. Improved brain accumula-tion of drug molecules and even drug carriers has been observed afteradministration of these agents. However, many of these compoundsare very toxic [99101]. Long-term use of this strategy for braindelivery of ARVs is likely not appropriate.

2.3.4. Drug modication approachMolecules with good lipophilicity can cross the cell membrane of

the endothelial cells by passive diffusion. Improved BBB passage canthus be achieved by conjugating ARV molecules with suitable sidebranch or functional group(s) to form prodrugswithmore favorablelipophilicity. These prodrugs, after gaining access to the endothelialcells, can be hydrolyzed and release the parent ARVs [102]. Since aprodrug is often ofcially recognized as a distinct chemical entity,

the blood-brain barrier (BBB) and blood-cerebrospinal uid barrier (BCSFB). AP: apicalfor improved delivery of antiretroviral drugs to the brain, Adv. Drug

C440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560



substantially more drug purication steps, screening tests and clinicalstudies are expected, which is usually not cost-effective.

2.3.5. Focused ultrasound and microbubble approachMicrobubbles are ne gas bubbles of less than 50 m in diameter.

Whenexposed toultrasound, thesemicrobubbles serveas the cavitationnuclei to focus and transduce the acoustic energy into mechanicalpower [103105]. Studies have shown that this combinational approachwas able to induce transient disruption of the BBB [103,105] This maylead to enhanced delivery of therapeutic compounds into the braincompartment. However, there are several concerns regarding the safetyof this strategy [105]. More work is required to establish the optimalconditions for extensive use of this method.

2.3.6. Nanotechnology for ARVs delivery to the brainARVs can be effectively delivered to the brain using drug carriers of

nanometer or submicron scale [106109]. Depending on the type ofnanocarriers, chemical modications of ARVs are often not required forefcient loading and delivery. There is a broad range of nanocarriersavailable, including liposomes, polymeric systems, nanoparticles, andmicelles, many of them clinically used before. Their versatility allowsthem to carrydiverseARVs. The following sectionswill discuss theuse ofnanocarriers for ARVs delivery to the brain in details.

3. Nanotechnology to improve ARVs delivery to the brain

3.1. General principles of brain delivery using nanocarriers

3.1.1. RationaleThe use of nanocarriers can improve brain delivery of ARVs in

several ways. The availability of ARVs to the CNS compartment can beimproved. Numerous studies using distinctively different nanocar-riers for delivering a broad range of therapeutic or diagnostic agentshave generally reported enhanced in vitro and in vivo BBB perme-ability and drug accumulation in the brain [106109]. Table 3 providessome examples of brain delivery of various therapeutic compoundsusing diverse types of nanocarriers. Many of the agents delivered arewell-established substrates of ABC transporters. These include P-gpsubstrates, e.g. doxorubicin, digoxin, rhodamine, vinblastine, andMRPsubstrates, e.g. methotrexate, uorescein [60,62]. Although theyusually present poor BBB permeability, with the use of nanocarriersthese compounds were able to achieve the desired therapeutic levelsin the CNS [see details and references in Table 3].

It is believed that with signicantly higher levels of ARVs reachingthe CNS, it is possible to reduce their doses and shorten the length oftherapy. This may translate into reduced risks of peripheral adversedrug effects. In addition, nanocarrier systems are known for theirexibility and versatility. They can bemade of different biocompatiblematerials, and most carriers can be engineered to obtain moredesirable pharmacokinetic and biodistribution proles for optimaltreatment of the CNS [106]. For example, the dosing frequency can bereduced by using a carrier that releases the ARV in a sustainedmanner. The circulation time can be prolonged and non-specic tissuebinding reduced by coating a nanocarrier with polyethylene glycol(PEG) [110]. Because only the carrier itself is engineered without theneed to modify the drug molecules, dramatic alterations of the drugpharmacology can often be avoided.

3.1.2. Overview of nanocarrier-mediated drug delivery across the BBBFig. 2 presents a proposed scheme depicting how nanocarriers can

be used to improve drug transport across the BBB. Overall, nanocarrierscan enhance brain delivery by three major pathways, which include:i) increasing the local drug gradient at the BBB by passive targeting,ii) allowing drug-trafcking by non-specic or receptor-mediated

endocytosis and iii) blocking drug efux transporters at the BBB.


In brief, by carefully choosing the biomaterials and adjusting theformulation parameters of the nanocarriers, these can be preparedwithphysicochemical properties desirable for interaction with the barrierstructures of the CNS (further details in Section 4.1). This is sometimesknown as passive targeting. As a result, a high local level of drug-loadednanocarriers can accumulate at the brain capillary endothelium, pro-ducing a high local concentration gradient to drive the drug penetrationrate by passive diffusion.

In addition to simply staying on the endothelium surface to releasethe loaded drug, some nanocarriers can enter cells by endocytosis, apathway that allows drug-trafcking [111]. This can occur via non-receptor mediated or receptor-mediated mechanisms. One example ofthe non-receptor mediated endocytosis is macropinocytosis. Macropi-nocytosis is a relatively non-specic processwhich allows cellular uptakeof large particles up to the micron size range [112]. This is likely a usefulpathway for nanocarriers such as solid lipid nanoparticles, which arefrequently 200 to 300 nm indiameter.Macropinocytosis is also related tothe uptake of Tat-peptides [113]. Receptor-mediated endocytosis, on theother hand, is triggered by receptorligand interaction. By choosing areceptor that is strongly and specically expressed on the surface of thecells to be targeted, and tagging the nanocarrier surface with the ligandmolecules thatmatch the receptor type, thedelivery process canbemademore selective and efcient [111]. Many receptor-mediated endocytoticpathways involve the formation of clathrin-coated pits, which envelopthe nanocarriers to be transported and eventually form vesicles, detachfrom the cell surface and carry the nanocarriers into the cytosoliccompartment [114]. This is a highly regulated and energy-dependentprocess, but may allow the whole nanocarrier and the loaded drug to gothrough the BBB, even bypassing the drug efux transporters.

Finally, some efux transporters expressed at the barrier structurescan be inhibited by the nanocarrier itself or the inhibitor blocking agentloaded into the nanocarrier (see Sections 3.2.3 and 4.4.3). This results inlocal inhibition of the drug efux and opens up a window where ARVmolecules can permeate.

3.2. Current use of nanocarriers for brain delivery of ARVs

So far, the research onnanocarrier-based delivery of ARVs to the CNSlargely remains at the experimental or pre-clinical stage. A number ofnanocarrier systems have been studied in in vitro or animal models[106109,115,116]. These nanocarriers generally fall into a few broadcategories: polymer/dendrimer-based, lipid-based or micelle-based.Despite the vast number of biomaterials eligible for nanocarrierpreparation, only a few of them are suitable for brain delivery. Thecomplexity of the CNS calls for conservative choices of biomaterialswithsolid track records of safety, particularly considering the durationof ARVtherapy, which usually takes years. Nanocarriers must be non-toxic andfully biodegradable, producing well-characterized and harmless degra-dationproducts only.Many lipids exist physiologically and are relativelynon-toxic. Polymers that t these criteria include the acrylic polymersand polyesters that contain lactide units. Pluronic block co-polymers/surfactants have also been widely used for preparing brain-targetingmicellar systems [108].

3.2.1. Polymer or dendrimer-based nanocarriersPoly(butyl cyanoacryalate) (PBCA), an acrylic polymer, has so far

been the most studied polymer for brain delivery. PBCA nanoparticleshave demonstrated good accumulation in both brain tissues andcerebrospinal uid without physical disruption of the BBB integrity[117]. PBCA is biodegradable and its lipophilicity facilitates efcientencapsulation of diverse types of neutral and weak base compoundssuch as dalargin, loperamide, amitriptyline, methotrexate and doxoru-bicin [117119]. However, their in vivo biodegradation is generally toofast and potentially harmful formaldehyde by-products can be formedduring the degradation [120]. Their loading capacity is also relatively561low, particularly for polar or ionic compounds. Charged acrylic co-


OOF

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

Table 3t3:1Examples of nanocarrier brain delivery systems for therapeutic compounds.

t3:2t3:3 Type Materials Therapeutic agents Surface coating Comments

t3:4 Polymericnanoparticles

PBCA Methotrexate [118] Polysorbate 80 Signicant increase in methotrexate levels in brain and cerebrospinaluid. Size b 100 nm penetrated BBB better.

t3:5 MMA-SPM; PCBA Zidovudine (AZT)t3:6 Lamivudine (3TC) [121] PBCA np increase BBB permeability of AZT & 3TC 820 and 1018 folds,

respectively; MMA-SPM np increase BBB permeability of both drugsby 100%; MMA-SPM loads AZT better

t3:7 PBCA Rhodamine [117] 20-fold increase in uptake by brain endothelial cells after Tween-80coating

t3:8 PBCA Dalargin [119] 3-fold increase in dalargin BBB penetration; dalargin has to be pre-adsorbed on to PBCA np for enhanced BBB penetration

t3:9 PLA Vasoactive intestinal peptide [125] PEG, agglutinin Nasal administration of np led to 5.67.7-fold increase in the brainaccumulation with agglutinin coating

t3:10 PLGA Dexamethasone [126] Embedded in alginate PLGA np embedded in alginate matrices were administered from neuralelectrode; dexamethasone released slowly in 2 weeks to reduceinammation of surrounding glial cells

t3:11 albumin Loperamide [157] Apo-lipoprotein E Loperamide-loaded np induced antinociceptive effects after iv injection;interaction with lipoprotein receptors required

t3:12 Chitosan 99 m-Technetium [193] Polysorbate 80 5-fold increase in brain concentration in micet3:13 Liposomes Phospholipids Phenytoin [136] N/A Improved local action against epilepsyt3:14 GABA [136] Decreased penicillin induced epileptic activityt3:15 Horseradish peroxidase [194] Transferrin Increased in vitro passage across BCEC culturet3:16 Amphotericin [140] PEG-RMP-7 Multiple fold increase in brain uptaket3:17 Micelle Pluronic P85 DPDPE, biphalin, morphine [146,148] N/A P85 enhanced the analgesic prole of biphalin, DPDPE, and morphine,

both above and below the critical micelle concentration.t3:18 Doxorubicin, digoxin, ritonavir, Drug permeability in monolayered BBB model by P-gp substrates

-Peo-glyt3:19


ARTICLE IN PRESSpolymers suchasmethylmethacrylate-sulfopropylmethacrylate (MMA-SPM) were therefore studied as a substitute. Their negative chargesgrant them a higher loading capacity for polar compounds includingzidovudine when compared to PBCA [121]. This MMA-SPM nanocarriersystem was able to increase the permeability of zidovudine and

taxol, vinblastine,rhodamine 123 [147,149]

BBB: blood-brain barrier; BCEC: brain capillary endothelial cells; DPDPE: [D-Pen2,Dsulfopropylmethacrylate; np: nanoparticles; PLA: polylactide; PLGA: poly(D,L-lactide-cUNCORREC

lamivudine across an in vitro BBB model of bovine brain-microvascularendothelial cells by 820 and 1018 folds, respectively [121]. Further invivo studies will help to establish whether this increase in brainpermeability will improve delivery to infected brain cells such as brainmononuclear macrophages.

Fig. 2.Major pathways used by nanocarrier systems to improve antiretroviral penetration actargeting, (2a) allowing drug-trafcking by endocytosis (non-specic or receptor-mediated

Please cite this article as: H.L. Wong, et al., Nanotechnology applicationsDeliv. Rev. (2009), doi:10.1016/j.addr.2009.11.020DPR

Polyesters such as poly(D,L-lactide-co-glycolide) (PLGA) and poly-lactide (PLA) have several qualities that make them appealing for braindelivery. Their degradationproducts (e.g.water and carbon dioxide) aremetabolic by-products and relatively non-toxic [122]. Because of theirsafety proles, PLGA and PLA are two of the few polymers ofcially

increased from 1.6 to 19.0 fold; no statistical difference in digoxinconcentration in the brain of mdr1 knockout mice with theaddition of Pluronic P85

n5]-enkephalin; PBCA: Poly(butyl cyanoacryalate), MMA-SPM methylmethacrylate-colide).TE 577approved for clinical use. In addition, these polyesters are known for 578their versatility. Their molecular weights, hydrophilicity, degradation579rate and hence the release kinetics can be conveniently tailored by580adjusting the composition [123]. They also easily form hydrolysable581bonds with diverse therapeutic molecules [122124] and targeting

ross the blood-brain barrier. (1) increasing the local drug gradient at the BBB by passive), (2b) blocking drug efux transporters. (): inhibitory effect.


CR

582

583

584

585

586

587

588

589

590

591

592

593

594

595

596

597

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613 3.2.2. Lipid-based nanocarriers614

615

616

617

618

619

620

621

622

623

624

625

626

627

628

629

630

631

632

633

634

635

636

637

638

639

640

641

642

643

644

645

646

647

648

649

650

651

652

653

654

655

656

657

658

659

660

661

662

663

664

665

666

667

668

669

670

671environment, the loaded drugs can bemore adequately protected from672degradation. Our group has investigated the use of SLN for atazanavir673

674

675

676

677

678

679

680

681

682

683

684

685

686

687

688

689

690

691

692

693

694

695

696

697

698

699

700

701

702

703

704

705

706

707

708



Lipid-based nanocarriers hold strong promise for delivery of ARVs tothe CNS. There are a wide range of physiological lipids and phospho-lipids available for lipid nanocarriers [please see reviews [131133].These materials are by nature biocompatible and biodegradable. Anumber of lipid-based formulations (e.g. liposome, lipoplex) are alreadycommercially available and all of themhave solid track record of clinicalsafety. The technology of their production on industrial scale has alsobeen well-established. Because lipophilic materials have the naturaltendency to target the BBB, it is expected that lipid-based nanocarrierswill be useful for CNS delivery of ARVs. There are several classes of lipid-based nanocarriers available, including liposomes, micro- or nanoemul-sion and solid lipid nanoparticles (SLN).

Liposomes are vesicles made of one or more phospholipids bilayers.They are probably the most studied and used lipid-based nanocarriers[132]. In fact, a number of liposomal systems have been developed andevaluated for the treatment of various brain illnesses, such as cerebralischemia by citicholine [134], brain tumors by cisplatin [135], andepilepsy by phenytoin [136]. Overall, signicant improvement in braindrug levels were observed in these studies. Although there are a fewliposomal formulations for delivery of ARVs, e.g. stavudine andzidovudine [137,138], relatively few of them are designed for HIV-associated CNS illnesses. Foscarnet is an antiviral used as a salvagetherapy for late-stageHIVpatientswithmultidrug resistance. Liposomalfoscarnet was able to increase the drug level in rat brains by 13-foldwhen compared to the free foscarnet solution [139]. Another drug,amphotericin B, is commonly used to treat the opportunistic fungalinfections in HIV patients. However, amphotericin B does not cross theBBB. The use of liposomes coupled with brain targeting peptides foramphotericin B delivery signicantly increased the drug transportacross the simulated BBB model formed by rat brain endothelial cells[140]. These studies further support the use of liposomes for moremoieties such as lectin [124]. Drug loading into PLGA/PLA nanocarriersandmodication of these systems for brain targeting are therefore quiteconvenient. Multiple-fold increases in brain drug concentration wereindeedobserved in PLGA/PLA systems (e.g. vasoactive intestinal peptideon PLA, dexamethasone on PLGA) [125,126], both administered viaintranasal route. Even large molecules such as peptides have beenshown capable of crossing the BBB in animalmodels [125]. However, upto date the use of PLGA/PLA based nanocarrier specically for ARVsdelivery to brain has not been reported. This is clearly an area for furtherstudy.

Like regular polymers, dendrimers also consist of repeatingmonomer units, but dendrimers are characterized by their repeatedlybranchedmolecular structures. These highly branchedmolecules, whenprecisely engineered, can form monodispersed globular or spheroidalnanostructures of 1 to over 10 nm in diameter [127]. Some of thesenanostructures may contain internal void spaces or surface functionalgroups for encapsulation or conjugation of drug molecules, and can beused as nanocarriers for drug delivery. They have been shown toincrease the BBB permeability of therapeutic agents such as DNA andmethotrexate [128,129]. Recently dendrimers have been evaluated forCNS delivery of ARVs. Polyamidoamine dendrimers loaded withlamivudine, a NRTI commonly used in HIV treatment, were evaluatedfor their in vitro antiviral activity inMT2 cells infectedwithHIV-1.Whenloaded on dendrimeric nanocarriers, a 21-fold increase in cellularlamivudine uptake and 2.6-fold reduction in the viral p24 levels wereobserved when compared to the group treated with free drug solution[130]. Despite these promising results, it should be noted that the drugrelease kinetics of dendrimers are sometimes inconsistent, and theirlong-term safety proles are relatively less established than polymerslike PLGA. More in vivo data are needed to further validate the use ofdendrimeric nanosystems for ARVs delivery to the CNS.effective treatment of HIV-associated CNS complications.

Please cite this article as: H.L. Wong, et al., Nanotechnology applicationsDeliv. Rev. (2009), doi:10.1016/j.addr.2009.11.020TEDPdelivery [144]. Using a human brain microvessel endothelial cell line(hCMEC/D3) representative of the BBB, a signicantly improved

accumulation of [3H]-atazanavir was obtained when the drug wasdelivered by SLN. Cytotoxicity experiments indicate that SLN exhibit notoxicity in hCMEC/D3 cells up to a concentration corresponding to200 nM of atazanavir. It was also noted that rhodamine-123, a well-established P-gp substrate, delivered by the same system also resultedin higher cellular accumulation, demonstrating that the P-gp efuxactivity at brain endothelial cells can be bypassed using SLN formula-tions. SLN evidently hold strong promise for brain delivery of ARVs,especially PIs which are mostly lipid soluble.

3.2.3. Micelle-based nanocarriersA micelle is an aggregate formed by typically 50100 amphiphilic

molecules (e.g. surfactants, block-copolymers) when dispersed in aliquid phase [145]. In aqueous solution the amphiphilic moleculesaggregate and expose their hydrophilic heads outside and hide theirhydrophobic segments in the inner core region. This structure facilitatessolubilization of hydrophobic drug compounds within the micelle core.The size of a micelle usually falls in the range of 5 to 20 nm in diameter.The small size and good drug solubilization properties make micellespotentially valuable nanocarriers.

Pluronicmicelleswere shownhighly effective for BBBdrug transportenhancement in vitro and in vivo [145147]. Using bovine brainmicrovessel endothelial cell monolayer model, the effect of PluronicP85 on the permeability of a broad range of structurally unrelatedcompounds was examined by Kabanov's group [147,148]. Increases inthe drug permeability up to 19-fold were detected. This permeabilityenhancement was particularly strong with P-gp substrates such aspaclitaxel, vinblastine as well as ritonavir [148], a PI frequently used inHAART regimens as a booster. In animalmodels, Pluronics increased thedrug delivery to the brain of wild-type mice, but the same benet wasnot observed in mdr1a/b knockout mice, indicating that the drugpermeability enhancingeffect by Pluronics ismediated at least in part byP-gp inhibition at the BBB [145]. It was suggested that this P-gpsuppressive effect could be mediated by ATP depletion, membraneuidization by the co-polymer, or a combination of both mechanismsOOF

Nanoemulsions and microemulsions are usually oil-in-waterformulations in which the oil phase is highly dispersed to dropletsof submicron size and stabilized by surfactants and co-surfactants.This type of formulations is especially suitable for highly lipophilicHIV drugs such as PIs. Recently, saquinavir, the rst PI marketed forHIV treatment, was evaluated for brain delivery in an oral formulationof axseed oil-based nanoemulsion [141]. The average oil droplet sizewas around 100 to 200 nm in diameter. Use of saquinavir nanoemul-sion instead of its free drug solution resulted in a three-fold increasein the saquinavir concentration in the systemic circulation and three-and ve-fold increase in the area-under-the curve (AUC) values andmaximum saquinavir concentration in the brain, respectively, of malebalb/c mice. This study shows that in addition to enhancement of BBBpermeability, the small size of the nanoemulsionmay also help bypassother barriers such as the gastrointestinal tract when used as oralformulations. There is clearly untapped potential in this nanocarrierclass.

SLN are a relatively new class of lipid-based nanocarriers [142]. Theyare made of one or more lipids with melting points higher than bodytemperature, so the carriers remain in solid state after administration.The low solubility of nanocarrier biomaterials probably contributes tothe high tolerability of this formulation. A study showed that SLN in factcaused less non-specic cell toxicity even compared to nanoparticlesmade of PLGA [143], which has long been the standard for biocompat-ible materials. In addition, by getting immobilized within a lipophilic709[147149].


710

711

712

713

714

715

716

717

718

719

720

721

722

723

724

725

726

727

728

729

730

731

732

733

734

735

736

737

738

739

740

741

742

743

744

745

746

747

748

749

750

751

752

753

754

755

756

757

758

759

760

761

762

763

764

765

766

767

768

769

770

771

772

773

774

775

776

777

778

779

780

781

782

783

784

785

786

787

788

789

790

791

792

793

794

795

796

797

798

799

800

801

802

803

804

805

806

807

808

809

810

811

812

813

814

815

816

817

818

819

820

821

822

823

824

825

826

827

828

829

830

831

832

833

834

835

836



4. Recent trends to optimize nanotechnology use for ARVbrain delivery

Despite promising data, the use of nanocarriers for ARVs deliveryto the brain remains at the experimental stage. Different strategieshave been proposed to further improve the various aspects of thisnovel therapeutic approach, from efciency, safety and specicity.

4.1. Optimization of nanocarrier properties to improve passive braintargeting

The physicochemical parameters of nanocarriers such as their size,surface charge and hydrophilicity can be optimized to favor the non-specic, passive form of brain targeting. Using methotrexate-loadedPBCA nanoparticles with sizes of 70, 170, 220, 345 nm, respectively, itwas shown that the 70 nm nanoparticles achieve signicantly higherpeak drug levels in the cerebrum, cerebellum and cerebrospinal uidthan the particles of larger sizes [129]. The authors suggested thatparticles smaller than 100 nm could better mimic the membranereceptors (e.g. low density lipoprotein receptors) and be moreefciently transcytosed via receptor-mediated pathways. In anotherstudy, using liposomes administered by the convection enhanceddelivery technique, not only the smaller liposomes (40 nm and80 nm) crossed the BBB more easily than the larger ones (200 nm) toachieve higher overall brain levels but they also penetrated deeper inthe brain tissues (0.79 mm average radius of penetration for the 80 nmliposomes vs. 0.64 mm for the 200 nm liposomes) [150]. Overall,carriers of size smaller than 100 nm are likely suitable for BBB passage.

The effect of carrier surface charge on brain delivery is lessconclusive. BBB is inherently negative in charge, so in theory cationiccarriers should lead to largest extent of brain delivery. However, astudy by Lockman et al., performed in rodents found that cationic SLNadministered by in situ brain perfusion did not effectively permeatethe BBB [151]. It is possible that the strong cell binding may preventthe carriers from penetrating. In the same study it was noticed thatanionic SLN were able to permeate the BBB about 1 to 2-fold betterthan the neutral or cationic formulations even at low SLN concentra-tion range, and there was no sign of damage to the endothelial tightjunctions [151]. This increase in the BBB permeability was thereforemore likely a result of improved cell internalization rather thanleakage via the paracellular pathway. It was suggested that thisunexpected behaviour could be derived from the strong binding of theanionic SLN to the low-density lipoprotein (LDL) receptors at the BBB,which induced receptor-mediated endocytosis in both in vitro and invivo models [151,152]. However, it should be noted that neutralcarriers were found to exhibit the strongest in vivo stability. Overall,carriers with neutral or anionic charge will probably result in optimalin vivo brain targeting.

A drug can passively diffuse through the BBB in a more efcientmanner after it is converted into a more lipophilic prodrug. The sameprinciple can be applied to brain targeting by delivering drugs onnanocarriers with enhanced lipophilicity. Fenart et al showed thatwhen polysaccharide nanoparticles were coated with a lipid bilayer, a3 to 4-fold improvement in brain uptake without disruption of theBBB integrity was observed [152]. The study also demonstrated a 27-fold increase in the uptake of albumin when it was coated with thesame lipid bilayer. However, nanocarriers with high surface lipophi-licity may favor non-specic tissue binding, and are also likely tobecome a target of the reticuloendothelial system (RES) so they willbe eliminated from the circulation before it reaches the brain [153]. Toreduce the interactions with the phagocytic cells in the RES locatedmainly in the spleen, liver and lymph nodes, many researchers coattheir colloidal carriers with PEG. It is well reported that PEG coating onnanocarriers may minimize their removal by the phagocytic cells,thereby prolonging their circulation time to typically 8 to 10h [153].

This will create a larger time window for the drugs to reach the CNS.


4.2. Development of specic brain targeting strategies

To further improve the efciency and specicity for brain delivery,various biomolecules expressed at the BBB can be targeted. Overall,these measures typically fall into two categories indirect targetingand direct receptor targeting.With indirect targeting, nanocarriers aremade of materials that bind to specic molecules in human body,which have high afnity with the receptors at the BBB. With directtargeting, nanocarriers are surface-grafted with ligand molecules thatspecically target those receptors at the BBB.

Polysorbates (also known as Tweens) are a commonly used class ofnon-ionic surfactants. They have very low toxicity and are ofciallyapproved for intravenous use. Polysorbates may serve as micellarnanocarriers when used alone, and can also form the surface coating ofother nanocarrier systems to confer these systems the BBB permeabiliz-ing properties [154]. It was found that the brain targeting properties ofthe aforementioned PBCA nanoparticlesweremostly derived from theirpolysorbate coating [155]. Studies revealed that polysorbate, particu-larly polysorbate-80, can increase the concentration of apolipoprotein E(apoE) adsorbed on the nanoparticle surface, and these apoE-enrichednanoparticles probably exploit the LDL receptor-mediated endocytoticpathway at the brain endothelial cells [155]. More evidence supportingthis hypothesis was later provided by Kreuter et al, which showed thatthe drug has to be loaded in the nanoparticles to gain passage across theBBB, and therefore a nanoparticle-mediated drug transport processmust be involved [156].

LDL-receptor targeting can also be achieved by the directapproach. Instead of using polysorbate as the linker moiety to adsorbapoE, a recent study directly conjugated the apoE molecules to theiralbumin-based nanoparticles by covalent linkages [156]. Signicantlyimproved delivery to the mouse brains was achieved. The same studycompared this direct approach (covalently linked apoE) to theindirect approach (apoE adsorbed by polysorbate-80), and showedmoderately longer therapeutic effect in the former group (157).

Other receptors have been studied for brain targeting by immuno-liposomes. Huwlyer et al. [158] have developed immunoliposomes fordelivery of the antineoplastic agent daunomycin to the rat brain. Themonoclonal antibody (mAb) used in these studies is the OX26 mAb tothe rat transferrin receptor [159], which in vivo is selectively enriched inthe brainmicrovascular endothelium [160]. Signicant improvement inbrain uptake of [3H]daunomycin was achieved when the drug wasdelivered usingOX26 immunoliposomes versus the standard liposomeswithout OX26 mAb. Brain targeting of immunoliposomes was notobserved when immunoliposomes were conjugated with a mouse IgG(2) isotype control. This technology has been extensively studied[161164] with development of a monoclonal antibody to the humaninsulin receptor [165] and could be extended to immunoliposomesloaded with ARVs for targeted delivery to the brain.

The specic receptor targeting strategy is particularly needed forPEG-coated nanocarriers. The hydrophilicity of PEG molecules canreduce the cell-carrier interaction and may be counterproductive tobrain targeting. Researchers developed PEGylated immunoliposomes(PIL) to solve this issue [166,167]. In PIL, approximately 1 to 2% of PEGmolecules are conjugated to targeting peptidomimetic monoclonalantibodies. This helps trigger receptor-mediated transcytosis of thePIL across the BBB. Using PIL, the therapeutic agents that normallycannot enter the brain, such as antisense oligomers and peptide drugs,can be made available to the brain [166,167].

It was found that single domain antibodies alone, without attachingto a nanoparticle or liposome, can also serve as a vector to delivertherapeutic peptides across the BBB. For instance, novel single domainantibodies such as FC5 were shown able to transmigrate across humancerebral microvessel endothelial cells in vitro and the BBB in vivo [168].By attaching horseradish peroxidase-tagged IgG to FC5 their transmi-gration across cerebral endothelial cells was signicantly enhanced837[168]. Pretreatment of human brain endothelial cells with wheat germ


C838

839

840

841

842

843

844

845

846

847

848

849

850

851

852

853

854

855

856

857

858

859

860

861

862

863

864

865

866

867

868

869

870

871

872

873

874

875

876

877

878

879

880

881

882

883

884

885

886

887

888

889

890

891

892

893

894

895

896

897

898

899

900

901

902

903

904

905

906

907

908

909

910

911

912

913

914

915

916

917

918

919

920

921

922

923

924

925

926

927

928

929

930

931

932

933

934

935

936

937

938

939

940

941

942

943

944

945

946

947

948

949

950

951

952

953

954

955

956

957



agglutinin, sialic acid, alpha (2,3)-neuraminidase or Maackia amurensisagglutinin which recognize alpha (2,3)-sialoglycoprotein receptorsignicantly reduced FC5 transcytosis. The data suggest that FC5 bindsluminal alpha(2,3)-sialoglycoprotein receptor which triggers clathrin-mediated endocytosis. This antibody-based vector may provide a newbrain-targeting drug delivery platform for HIV treatment [169].

Overall, the potential drawback of this class of specic receptor-targeting platforms lies in their cost. To obtain large quantity ofmonoclonal antibodies of clinically useful grade at sufciently low costposes a major obstacle.

4.3. Cell penetrating peptides

Trans-activator of transcription (Tat) is a peptide derived from HIV-1. It is able to substantially promote the level of transcription of the HIVDNA [15,170]. Infected cells can produce Tat to activate the uninfectedcells to initiate the HIV gene production. Tat contains a basic regionconsisting of six arginine and two lysine residues [171]. Their strongcationic charges facilitate interaction with the normally negativelycharged cell surface, trigger permeabilization of the cellmembrane via areceptor/transporter independent pathway which results in endocyto-sis of the sequence [172]. It was found that by tagging particulatesubjects with Tat, these particles can gain entrance into cells using thesame uptake mechanism. For instance, Tat-conjugated fused proteinswere able to efciently bypass the BBB [173]. These ndings led to anumber of studies on Tat-based brain targeting systems for treatment ofHIV-associated CNS complications [174176]. Nanoparticles taggedwith intact or scrambled Tat sequences can interact better with cellmembrane of vascular endothelial cells, and signicantly increase thebioavailability of ritonavir to the brain when compared to thenanoparticles without Tat. The potency of Tat is quite high. Use ofnanograms of Tat was sufcient to achieve a therapeutic level ofritonavir in the CSF.

Other cell penetrating peptides such as antennapedia and penetratinhave also been studied [177,178]. Itmust benoted the long-term risks oftoxicity or immune responses associated with these products have notbeen fully established. Their target specicity is also inconclusive [179].Regardless, considering thepotential return this novel strategymay leadto, it is still an exciting area to further explore.

4.4. Other nanotechnology-based strategies to improve ARVs brain delivery

There are experimental strategies based on nanotechnology thathave not been extensively tested for ARV delivery to brain, but arehighly novel and show enormous potential.

4.4.1. Use of macrophages for BBB passageA novel therapeutic strategy was developed based on the nding

that the BBB of HIV-infected patients can be easily penetrated bymacrophages by the aforementioned Trojan horse effect. Indinavir washomogenized into verynenanocrystals and coatedwithphospholipids[180]. These nanocrystals were allowed to be internalized into bone-marrow-derived macrophages and the indinavir-loaded macrophageswere injected into amousemodel of HIV-infection. Itwas found that themacrophages migrated into the brain and delivered therapeutic dose ofindinavir. The antiretroviral effects, as determined by the decrease inHIV-1 p24 viral levels, were sustained for weeks after a single injectionof the formulation [180].

4.4.2. Alternative route for nanocarriers administrationRecently, the use of intranasal administration for brain targeting has

raised signicant interest. Extensive studieshavebeen conductedon thetransport of large molecules from the nasal cavity to the brain. Fordetails please refer to the reviews by Illum L [181]. This approachwas infact inspired by the brain entrance mechanisms of a number of viruses,

which enter the olfactory lobe of the brain and the cerebrospinal uid


via the nasal passages. This implies that particulate matters fromproteins to viruses should be able to use this shortcut route to reach thebrain. The major barrier is in the passage of the carriers through thenasal mucosal membrane. Because this membrane has plenty of lectinreceptors, PLA-PEG nanoparticles targeting lectin receptors wererecently studied [182], and 2-fold increase in brain uptakewas observedwhen compared with the nanoparticles without the lectin coating. Useof this route to deliver anti-HIV peptides was also tested [183],improvement of cognitive functions in tested subjects was observed.

Convection enhanced delivery is a novel delivery technique tobypass the BBB and administer therapeutic agents directly into targetedbrain parenchyma or tissue. This technique involves one or morecatheters tobe stereotacticallyplaced throughcranial burr holes into thebrain. Therapeutic agents such as liposomes are subsequently admin-istered by microinfusion pump [150]. Although this method is toodrastic to be used in most HIV-infected patients, it may be reserved forlate-stage patients when the worsening of mental conditions is out ofcontrol.

4.4.3. Advanced delivery of ABC-transporter blockersTheuse ofABC-transporter blockers to enhance CNSdeliveryof ARVs

is limited by their risks of side effects and drug interactions. Theselimitations can possibly be overcome by encapsulating and deliveringthese agents with drug carriers. By choosing nanocarriers with highbrain afnity, strong localized blocking effect on the ABC-transportersmay be obtained. This approach has beenrst adopted for the treatmentof cancer expressing P-gp [184,185]. Polymer-lipid hybrid nanoparticlesco-loaded with doxorubicin (cytotoxic compound) and verapamil orGF120918 (P-gp blockers) were prepared. It was found that this co-loaded system resulted in signicant accumulation of doxorubicin in theP-gp overexpressing MDA435/MDR1 breast cancer cell line. Thedoxorubicin-mediated cell kill was improved by nearly a log10 order.This same strategy for BBB targetinghas been evaluated in a few studies.For example, PSC833, a non-immunosuppresise, cyclosporine-A analog,P-gp blocker, was encapsulated in liposomes made of intralipids for invivo inhibition of P-gp at the BBB in primates. The AUCbrain/AUCbloodratio of 11C-verapamil (a P-gp substrate) radioactivity was increased2.3-fold with the addition of PSC833-liposomes [186]. As discussed inthe previous sections, Pluronics and polysorbates do not just serve asmicelles to carry drugs, these amphiphilic molecules also possessintrinsic ABC-transporter blocking activities. In the future, it is expectedthat more nanocarrier systems will incorporate biomaterials or asecondary drug with ABC-transporter blocking properties to improvethe delivery of ARVs to the CNS.

5. Future perspectives and conclusion

Since the advent of HAART, it is no longer unrealistic to achievereasonable control of the HIV viral load in the periphery. However,without effective measures to deliver the ARVs to the brain, the viralload that persists at this site can lead to various debilitating neurologicalcomplications and pose a strong threat to drug resistance and latentreinfection. Current use of HAART regimens to clear the CNS viral load isnot only ineffective and risky, but also too costly for widespread use inthe developing nations. Studies have demonstrated that nanocarrierscan signicantly increase the CNS penetration of several ARVs. Althoughmost studies remain at the experimental stage, they have alreadyestablished the feasibility of this approach. The next step will be tooptimize this general strategy, in terms of efciency, safety andspecicity. To achieve this goal, future research in this eld shouldfocus on the following issues.

5.1. Improve understanding of the barrier structures

The knowledge of the various biomolecular events that occur at

958the BBB allows us to develop the more active and specic forms of


959

960

961

962

963

964

965

966

967

968

969

970

971

972

973

974

975

976

977

978

979

980

981

982

983

984

985

986

987

988

989

990

991

992

993

994

995

996

997

998

999

1000

1001

1002

1003

1004

1005

1006

1007

1008

1009

1010

1011

1012

1013

1014

1015

1016

1017

1018

1019

1020

1021

1022

1023

1024

1025

1026

1027

1028

1029

1030

1031

1032

1033

1034

1035

1036

1037

1038

1039

1040

104110421043104410451046104710481049105010511052105310541055105610571058105910601061106210631064immunohistochemistry, AIDS 10 (1996) 573585.1065[10] H. Chen, C. Wood, C.K. Petito, Comparisons of HIV-1 viral sequences in brain,10661067106810691070107110721073107410751076107710781079108010811082108310841085108610871088108910901091109210931094



brain targeting strategy. It is expected that more and more receptorswill be identied for specic brain targeting. In fact, receptors such astransferrin receptor and insulin receptor have been targeted andimproved BBB passages were demonstrated [187,188]. The discoveryof cell penetrating peptides such as Tat also opens up excitingopportunities, although issues like specicity and safety need to beaddressed in a more conclusive manner. Development of nanocarriersfor the delivery of ABC-transporter blockers is another valuableoption. The main goal should be to minimize excessive interference ofthe ABC-transporters in other organs and tissues.

5.2. Tailor-made nano-formulations for brain delivery

With a better understanding of the architectures of the BBB andBCFSB, nanocarriers should be tailor-made with the suitable physico-chemical properties that will allow at least adequate passive braintargeting. This means careful choice of the biomaterials andformulation parameters. Lipophilic nanocarriers below or near100 nm in diameter will probably be most useful. In addition, thebiomaterials selected need to possess very low toxicity and are fullybiodegradable to avoid damages to the CNS. Overall, the moreestablished materials such as lipids, phospholipids, PLGA and a fewselected non-ionic surfactants are likely good candidates to buildthese nanocarrier platforms.

5.3. Use of better experimental models

The lack of appropriate in vivo models to simulate the changes inBBB integrity in HIV infection is a major obstacle to furtherdevelopment of brain delivery strategies. There has been increasingevidence that indicate structural and functional alterations of the BBBduring HIV infection [189,190]. In particular, key membrane proteins(e.g. occludin and zona occludens-1) forming the tight junctions ofbrain capillary endothelium are signicantly downregulated in HIVE.For detailed mechanisms of the BBB disruption in HIV-infection,please see the review by Toborek et al., [191]. It has been welldocumented that in cancer nanotechnology, increased tumor pene-tration of nanoparticles can be caused by the leaky vasculature andpoor lymphatic drainage in the solid tumors [192]. Similarly it couldbe hypothesized that alterations in the BBB integrity could alloweasier passage of nanocarriers across the BBB and thereby permittingenhanced delivery of ARVs. With more realistic in vivo models, novelstrategies based on therapeutically or pathophysiologically inducedchanges in the drug permeability of the BBB can be more thoroughlystudied to maximize delivery of ARVs using nanocarriers.

When invitromodels are used, their limitations should be consideredin order to avoid unsubstantiated conclusions. For example, while ananocarrier system can be efciently taken up by brain microvesselendothelial cells, this does not necessarily translate into high BBBpermeability. Trapping of nanocarriers in endosomes/lysosomes canfrequently occur followingendocytoticprocesses (e.g. clathrin-mediatedendocytosis) [50]. Unless the loaded ARVs can be released or thenanocarrier can trancytosed, thedrugswill likely accumulate in thebraincapillary endothelium. Furthermore, successful passage of ARVs across asimulated BBBmodel is not fully predictive of its therapeutic effect. Thedrugs still need to access the HIV-infected cells in the brain and remainbiologically active. Caution should be exercised in using the datagenerated from these models.

5.4. Rene targets and endpoints of ARV delivery

It must be noted that in many of the previously discussed in vivoanimal studies, the whole brain was often considered as a single,homogeneous target. Drug levels in different brain cell types wereseldom differentiated. As a result, uninfected cell types such as the brain

capillary endothelial cells may accumulate high ARV levels, whereas the

Please cite this article as: H.L. Wong, et al., Nanotechnology applicationsDeliv. Rev. (2009), doi:10.1016/j.addr.2009.11.020choroid plexus and spleen: Potential role of choroid plexus in the pathogenesis ofHIV encephalitis, J. NeuroVirol. 6 (2000) 498506.

[11] F. Gonzlez-Scarano, J. Martn-Garca, The neuropathogenesis of AIDS, NatureRev. Immunol. 5 (2005) 6981.

[12] V.C. Asensio, I.L. Campbell, Chemokines in the CNS: plurifunctional mediators indiverse states, Trends Neurosci. 22 (1999) 504512.

[13] M. Kaul, S.A. Lipton, Chemokin

Nanotechnology Applications for Improved Delivery of Antiretroviral Drugs To

Documents