p-html.html

Journal Pre-proof

Oral Milk Exosome-PLGA Nanoparticles Enhance Anti-Tuberculosis Efficacy of

PBTZ169 and Bedaquiline

Eryue Liu, Chennan Liu, Yangxue Ye, Zimo Wang, Weiyan Zhang, Lei Fu, Bin Wang,

Yujin Wang, Yu Lu
PII:

S2589-0042(26)01016-3

DOI:

https://doi.org/10.1016/j.isci.2026.115641

Reference:

ISCI 115641

To appear in:

iScience

Received Date: 18 September 2025
Revised Date: 26 January 2026
Accepted Date: 4 April 2026

Please cite this article as: Liu, E., Liu, C., Ye, Y., Wang, Z., Zhang, W., Fu, L., Wang, B., Wang, Y.,

Lu, Y., Oral Milk Exosome-PLGA Nanoparticles Enhance Anti-Tuberculosis Efficacy of PBTZ169 and

Bedaquiline,

iScience

(2026), doi:

https://doi.org/10.1016/j.isci.2026.115641

This is a PDF of an article that has undergone enhancements after acceptance, such as the addition

of a cover page and metadata, and formatting for readability. This version will undergo additional

copyediting, typesetting and review before it is published in its final form. As such, this version is no

longer the Accepted Manuscript, but it is not yet the definitive Version of Record; we are providing

this early version to give early visibility of the article. Please note that Elsevier’s sharing policy for the

Published Journal Article applies to this version, see:

https://www.elsevier.com/about/policies-and-

standards/sharing#4-published-journal-article

. Please also note that, during the production process,

errors may be discovered which could affect the content, and all legal disclaimers that apply to the

journal pertain.

Oral Milk Exosome-PLGA Nanoparticles Enhance

Anti-Tuberculosis Efficacy of PBTZ169 and

Bedaquiline

Eryue Liu

, Chennan Liu

, Yangxue Ye

, Zimo Wang

, Weiyan Zhang

, Lei Fu

，

Bin Wang

Yujin Wang

, Yu Lu

1 Beijing Key Laboratory of Drug Resistance Tuberculosis Research, Beijing Chest Hospital,

Capital Medical University, Beijing Tuberculosis and Thoracic Tumor Research Institute,

Beijing 101149

Lead contact: Yu Lu (luyu4876@hotmail.com)

Corresponding author: Yu Lu (luyu4876@hotmail.com)

Keywords

: Exosome, Oral drug delivery, Tuberculosis, Murine Model, Bedaquiline, PBTZ169

Journal Pre-proof

Summary

: The oral delivery of next-generation anti-tuberculosis drugs, such as PBTZ169 and

bedaquiline, is hindered by poor solubility and low bioavailability. We have developed a

bioinspired delivery platform that combines milk exosomes with PLGA nanoparticles to deliver

the two drugs separately. This nano-delivery system leverages the gastrointestinal stability and

mucosal penetration of exosomes, along with the high encapsulation efficiency of PLGA,

significantly enhancing drug hydrophilicity and stability. In murine models, the exosome-coated

nanoparticles increased plasma bioavailability by 2.5

–

4.9-fold compared to free drugs and

achieved superior accumulation in target organs, while mitigating the cardiotoxicity risk associated

with BDQ. This synergistic strategy, integrating synthetic and natural carriers, overcomes key

pharmacological barriers, offering a promising approach to developing effective and patient-

compliant oral therapies for tuberculosis and potentially other infectious diseases.

Introduction

Tuberculosis (TB), a chronic infectious disease caused by

Mycobacterium tuberculosis

(MTB),

remains a formidable global health challenge, persistently ranking among the top ten causes of

mortality worldwide. According to the World Health Organization (WHO) 2024 Global

Tuberculosis Report and relevant academic studies, there were an estimated 10.8 million new TB

cases globally in 2023 alone, highlighting the disease's continuing threat to public health

. While

significant progress has been made in anti-TB drug development, current therapeutic regimens

continue to suffer from critical limitations, including protracted treatment durations, suboptimal

efficacy, and severe adverse effects, all contributing to poor patient compliance

. These challenges

necessitate the urgent development of novel treatment strategies characterized by shorter duration,

reduced toxicity, and enhanced therapeutic outcomes.

Journal Pre-proof

Recent years have witnessed notable advances in next-generation anti-TB agents, yet several

pharmacological hurdles persist. Macozinone (PBTZ169), a frontrunner in the benzothiazinone

(BTZ) class of DprE1 inhibitors, demonstrates exceptional

in vitro

potency with a remarkably low

minimum inhibitory concentration (MIC) of 0.2 ng/mL

—

significantly surpassing that of existing

therapeutics, including bedaquiline (BDQ)

3,4

. However, its clinical translation is hindered by poor

drug-like properties stemming from high lipophilicity (ClogP = 5.06), which manifests in low oral

bioavailability and dose-limiting hepatotoxicity and gastrointestinal complications

5,6

. Similarly,

bedaquiline

—

while mechanistically innovative

—

faces substantial delivery challenges due to its

extreme hydrophobicity (logP = 6.37, aqueous solubility = 0.000193 mg/mL; values from the

DrugBank public property profile)

–

. These physicochemical challenges are compounded by

safety considerations that require careful management in the clinical setting. While BDQ has

demonstrated an acceptable overall safety profile, its known QTc interval prolongation effect

necessitates monitoring during treatment, as it may increase the risk of Torsades de Pointes (TdP),

a potentially life-threatening arrhythmia, particularly when co-administered with other QTc-

prolonging drugs. Although hyperuricemia has been reported in some studies, it is not considered

a primary or common adverse reaction associated with BDQ therapy

–

To circumvent these limitations, contemporary TB drug development has increasingly focused

on advanced oral delivery platforms. Compared to parenteral administration, oral formulations

offer superior patient compliance and clinical practicality.

Given the need to overcome the

physicochemical and pharmacokinetic hurdles of promising candidates like BDQ and PBTZ169

while optimizing their therapeutic safety window, the development of novel delivery systems is of

significant value. Among emerging strategies, nanocarrier-based delivery systems have shown

particular promise. Functionalized poly(lactic-co-glycolic acid) (PLGA) nanoparticles, for

Journal Pre-proof

instance, have demonstrated enhanced therapeutic profiles for rifampicin and isoniazid in

preclinical models, outperforming conventional formulations at equivalent doses

. As an FDA-

approved biomaterial, PLGA nanoparticles provide distinct advantages, including high

encapsulation efficiency, controlled biodegradability, excellent biocompatibility, and versatile

surface engineering potential

. Nevertheless, their clinical utility as oral delivery vehicles remains

limited by premature gastrointestinal degradation and off-target biodistribution

Extracellular vesicles (EVs), particularly exosomes (30

–

150 nm), have recently emerged as

revolutionary biological delivery vectors. Their innate phospholipid bilayer architecture, coupled

with exceptional biocompatibility, prolonged circulation kinetics, and unique biological barrier-

traversing capabilities, positions them as ideal drug carriers

–

. However, intrinsic limitations in

production scalability and drug loading efficiency continue to hinder their therapeutic application

–

. In addition, milk-derived exosomes are highly resistant to harsh gastrointestinal tract

environments

–

. This has spurred innovative hybrid approaches that combine synthetic

nanoparticles with exosomal components, creating "bioinspired nanovesicles" that synergize

artificial design with natural biological activity.

Building upon these research breakthroughs, our study innovatively developed an oral delivery

platform that ingeniously integrates PLGA nanoparticles with milk-derived exosomes to achieve

separate delivery of the anti-tuberculosis drugs PBTZ169 and bedaquiline. This biomimetic system

achieves dual optimization: (1) dramatically improving drug solubility and pharmacokinetic

profiles through nanoformulation, while (2) harnessing exosomes' innate targeting specificity and

mucosal penetration capabilities for enhanced drug delivery efficiency. The incorporation of MEs

further enhances the biocompatibility and physiological stability of the system. By addressing

multiple pharmacological challenges simultaneously

—

from bioavailability enhancement to

Journal Pre-proof

toxicity mitigation

—

this innovative approach represents a paradigm shift in TB therapeutics,

potentially initiating a novel phase of patient-centric, high-efficiency oral treatment modalities.

Results

Preparation and Characterization of ME-PLGA-PBTZ169 and ME-PLGA-BDQ

Nanoparticles

In this study, ME-PLGA-PBTZ169/BDQ NPs were synthesized through a three-step process: (1)

isolation ofME, (2) preparation of PLGA-PBTZ169/BDQ NPs, and (3) co-incubation of ME with

PLGA-

PBTZ169/BDQ NPs. The resulting nanoparticles were lyophilized and stored at −80°C for

further use.

Transmission electron microscopy (TEM) imaging confirmed the characteristic cup-shaped

morphology of ME (Fig. 1A). Dynamic light scattering (DLS, Table 1) analysis revealed a

hydrodynamic diameter of 104.50±4.86 nm for ME, with a size distribution ranging from 30 to

150 nm, consistent with typical exosome dimensions. The relatively low polydispersity index (PDI,

0.16±0.02) and high yield (BCA protein concentration: 3.29 ± 0.20 mg/mL) suggest that both the

extraction method and the milk source used in this study were suitable for efficient ME isolation.

The morphological characterization of ME-PLGA-PBTZ169 NPs and ME-PLGA-BDQ NPs

(Figure 1B-C) and particle size/potential measurements (Table 1) demonstrated that the

nanoparticle diameter increased correspondingly with the addition of the bilayer cell membrane

coating on the nanoparticle core. Furthermore, zeta potential results indicated surface potentials of

-20.28 ± 0.45 mV and -18.65 ± 0.83 mV for PLGA-PBTZ169 NPs and ME-PLGA-PBTZ169 NPs,

100

respectively, due to the charge shielding effect from the membrane coating. Similarly, PLGA-

101

BDQ NPs and ME-PLGA-BDQ NPs showed surface potentials of -20.44 ± 0.39 mV and -18.26 ±

102

0.47 mV, respectively. In this study, zeta potential measurements revealed an increase in surface

103

Journal Pre-proof

charge of PLGA after membrane coating. Compared to PLGA NPs, ME-PLGA NPs exhibited

104

increased diameter.

We evaluated the storage stability of ME-PLGA NPs at -80°C for one month.

105

As shown in Figure 2, no significant differences were observed in the average particle size,

106

polydispersity index, or zeta potential of the nanoparticles after reconstitution following

107

lyophilization. This indicates that the nanoparticles maintained good physical stability under long-

108

term low-temperature storage conditions, making them suitable for subsequent experimental use.

109

The encapsulation efficiency (EE) and drug loading capacity (DL) of ME-PLGA-PBTZ169 NPs

110

and ME-PLGA-BDQ NPs were determined. Both drug-loaded nanoparticles exhibited high

111

encapsulation efficiency (63.3% and 68.52%, respectively) and drug loading (15.90% and 18.13%,

112

respectively) (Table 2). Western Blot (WB) analysis confirmed the presence of exosomal marker

113

proteins in both ME and drug-loaded ME-PLGA NPs (Figure 1). Due to dilution during the

114

preparation process, the ME-PLGA NPs exhibited weaker band intensities in WB. However,

115

quantitative analysis demonstrated that the procedure successfully preserved over 85% of the

116

membrane proteins (Figure 1), confirming its efficiency. In conclusion, we successfully prepared

117

ME-PLGA NPs.

118

119

Surface Hydrophilicity of ME-PLGA Nanoparticles Compared to Raw APIs

120

The surface wettability characteristics of PBTZ169, BDQ, PLGA-PBTZ169 NPs, PLGA-BDQ

121

NPs, ME-PLGA-PBTZ169 NPs, and ME-PLGA-BDQ NPs were evaluated via contact angle

122

measurements after lyophilization and tablet compression. The contact angle (θ),

defined as the

123

angle formed between the tangent to the liquid-vapor interface and the solid-liquid interface at the

124

three-phase contact point, serves as a key parameter for assessing surface wettability. A contact

125

Journal Pre-proof

angle θ < 90° indicates a hydrophilic surface with good liquid spreading, while θ > 90° suggests a

126

hydrophobic surface with poor wettability.

127

Marked differences in surface wetting behavior were observed experimentally: when water

128

droplets were placed on the surfaces of the raw APIs (PBTZ169 and BDQ), the droplets exhibited

129

a distinct spherical shape with contact angles exceeding 90°, demonstrating typical hydrophobic

130

characteristics with minimal spreading. In contrast, on the surfaces of ME-PLGA NPs and PLGA

131

NPs, the droplets spread rapidly, with contact angles all below 90°, reflecting the wetting behavior

132

characteristic of hydrophilic surfaces. As shown in Figure 3, both ME-PLGA NPs and PLGA NPs

133

displayed contact angles < 90°, confirming their hydrophilic nature, whereas the raw APIs showed

134

signi

ficantly higher contact angles (θ > 90°), indicating strong hydrophobicity.

135

These results demonstrate that the ME-PLGA encapsulation strategy can significantly improve

136

the surface properties of the original drug compounds, effectively enhancing their hydrophilicity

137

and solubility.

138

139

In Vitro Anti-mycobacterial Activity of ME-PLGA-PBTZ169 and ME-PLGA-BDQ

140

Nanoparticles

141

The anti-tubercular activity of ME-PLGA-PBTZ169 NPs and ME-PLGA-BDQ NPs was

142

evaluated using the microplate Alamar Blue assay (MABA) against

Mycobacterium tuberculosis

143

The minimum inhibitory concentrations (MICs) were determined for both nanoformulations and

144

their corresponding free drugs.

Free PBTZ169 and BDQ exhibited MIC values of 0.000625

μg/mL

145

and 0.0625

μg/mL, respectively.

Drug-loaded PLGA NPs and ME-PLGA NPs showed comparable

146

MIC values to free drugs (

> 0.05), indicating no significant loss of antimicrobial potency after

147

nanoencapsulation. Blank PLGA and ME-PLGA NPs (10 mg/mL) displayed no inherent

148

Journal Pre-proof

antibacterial activity, confirming that the observed effects were drug-dependent. Notably, while

149

free drugs required DMSO for solubilization, the nanoformulations were readily dispersible in

150

0.9% saline, demonstrating significantly improved aqueous compatibility.

Notably, identical MIC

151

values were obtained for H37Ra compared to H37Rv across all tested samples (see Supplementary

152

Table S1), demonstrating equivalent susceptibility of both reference strains to the compounds and

153

nanoformulations.”

154

155

In Vitro Cytotoxicity of ME-PLGA-PBTZ169 and ME-PLGA-BDQ Nanoparticles

156

The cytotoxicity evaluation was conducted on multiple cell lines, including J774A.1, HepG2, Vero,

157

Caco-2, and Hela, revealed that both ME-PLGA-PBTZ169 nanoparticles and ME-PLGA-BDQ

158

nanoparticles exhibited toxicity profiles comparable to their corresponding free drugs (PBTZ169

159

and BDQ, respectively). No statistically significant differences (

> 0.05) in cell viability were

160

generally observed across all tested concentration ranges. Notably, the blank PLGA and ME-PLGA

161

nanoparticle carriers demonstrated excellent biocompatibility in all tested cell lines, maintaining

162

cell viability above 90% even at the extremely high concentration of 10 mg/mL. These

163

comprehensive findings collectively indicate that: (1) compared to free drug formulations, the

164

nanoencapsulation process largely did not introduce additional cytotoxic effects; (2) the

165

nanoparticle delivery system itself possesses favorable in vitro safety characteristics; and (3) the

166

developed nanoformulations generally retained the favorable toxicity profiles of the original drugs

167

while offering the advantages of nanoparticle delivery.

168

169

In Vitro Release Profiles of ME-PLGA Nanoparticles in Simulated Gastrointestinal

170

Media

171

Journal Pre-proof

The present study systematically compared the sustained-release characteristics of three drug

172

delivery systems, demonstrating that ME-PLGA NPs exhibit superior sustained-release

173

performance, significantly outperforming both PLGA NPs and free drugs. Analysis of release

174

behaviors under different pH conditions revealed that both PLGA NPs and ME-PLGA NPs showed

175

(Figure 5) higher release rates in simulated intestinal fluid compared to simulated gastric fluid,

176

while ME-PLGA NPs displayed more stable pH-independent release properties. This ME-PLGA

177

composite system not only maintains high stability in gastric environments but also achieves

178

optimized release kinetics in intestinal conditions. Its unique sustained-release characteristics

179

effectively prolong the duration of drug action and reduce the dosing frequency, providing

180

significant advantages for clinical therapy. Particularly in tuberculosis treatment, this delivery

181

system demonstrates remarkable clinical application value by maintaining stable plasma drug

182

concentrations, reducing dosing frequency, and improving patient compliance. Building on these

183

findings, we further evaluated the system's pharmacodynamic and pharmacokinetic profiles in

184

tuberculosis animal models and clinical studies. These translational medical investigations will

185

provide crucial foundations for developing more effective tuberculosis treatment regimens.

186

187

Pharmacokinetic and Bioavailability Analysis of ME-Coated Formulations

188

The pharmacokinetic evaluation demonstrated that the nanoformulations, particularly the milk

189

exosome-coated PLGA nanoparticles (ME-PLGA-NPs), significantly enhanced the oral

190

bioavailability and altered the biodistribution profiles of both PBTZ169 and BDQ. Both drugs

191

exhibited substantially increased systemic exposure after encapsulation, with notable elevations in

192

plasma AUC and C

max

values compared to their free drug counterparts. The terminal half-life (

1/2

)

193

Journal Pre-proof

was prolonged in plasma for both nanoformulations, indicating sustained release

194

characteristics( Figure 6-7, Table S2-7).

195

Critically, the nanoformulations markedly altered the in vivo distribution behavior of the drugs,

196

leading to significantly enhanced accumulation in disease-relevant tissues. Pulmonary drug

197

concentrations were substantially elevated, particularly for ME-PLGA-PBTZ169 NPs, which

198

achieved an approximately 10-fold increase in AUC

–∞

compared to the free drug. This substantial

199

enrichment in the lungs

—

the primary site of TB infection

—

holds significant therapeutic relevance.

200

Similarly, drug exposure in the spleen was significantly increased, which could facilitate the

201

eradication of latent bacterial reservoirs. The ME coating effectively moderated the excessively

202

prolonged pulmonary retention observed with PLGA-BDQ NPs (reducing

1/2

from 71.89 h to 33.77

203

h), potentially reducing toxicity risks associated with drug accumulation and suggesting possibly

204

improved biocompatibility and clearance profiles.

205

These results collectively demonstrate that the ME-PLGA nano-platform successfully enhances

206

drug solubility, achieves optimized drug release properties, and promotes favorable redistribution

207

of drugs to therapeutically important organs. This biodistribution-shifting effect offers a promising

208

and broadly applicable strategy for improving the therapeutic efficacy of anti-tuberculosis agents.

209

210

In Vivo Antimicrobial Efficacy of ME-PLGA-PBTZ169 and ME-PLGA-BDQ

211

Nanoparticles

212

After two weeks of treatment at equivalent drug doses, a comparative evaluation of antibacterial

213

activity revealed significantly enhanced therapeutic efficacy for both nanoformulations in both

214

lung and spleen tissues.

215

Journal Pre-proof

In lung tissue (Figure 8-E), the ME-PLGA-PBTZ169 NPs group demonstrated a reduction in

216

bacterial load exceeding 1 log10 compared to the untreated group, performing markedly better

217

than both the free drug group (0.4 log10 reduction) and the exosome-free PLGA NPs group (0.6

218

log10 reduction). The ME-PLGA-BDQ NPs group exhibited even stronger antibacterial efficacy,

219

achieving a reduction exceeding 2.5 log10, which surpassed not only the free BDQ drug group (1

220

log10 reduction) but also the plain PLGA NPs group (1.8 log10 reduction).

221

In spleen tissue (Figure 8-F), the ME-PLGA-PBTZ169 NPs group showed a 0.85 log10

222

reduction compared to the untreated group, while both the free drug and PLGA NPs groups

223

demonstrated minimal therapeutic effects. The ME-PLGA-BDQ NPs group achieved a more than

224

2 log10 reduction, significantly outperforming both the free BDQ group (approximately 0.9 log10

225

reduction) and the plain PLGA NPs group (approximately 1.7 log10 reduction).

226

This enhanced antibacterial activity correlates with our pharmacokinetic findings, where both

227

nanoformulations showed: (1) improved drug solubility in physiological conditions, (2) enhanced

228

intestinal absorption, and (3) significantly increased bioavailability. Supporting evidence was

229

obtained through drug concentration measurements in blood and target organs (lungs and spleen)

230

at T

max

, demonstrating substantially higher drug accumulation in ME-PLGA NP-treated groups

231

(

<0.05). Notably, after two weeks of treatment, both blood and tissue drug concentrations

232

remained significantly elevated in nanoparticle-treated groups compared to free drug

233

administration (Figure 8-A-D), confirming the sustained-release characteristics and improved

234

biodistribution profile of the nanoformulations. These results collectively demonstrate that the

235

ME-PLGA platform not only enhances initial drug absorption but also maintains therapeutic drug

236

levels over extended periods, leading to superior in vivo antibacterial outcomes.

237

238

Journal Pre-proof

Effect of Formulation Modification on the Cardiotoxicity of BDQ in vivo

239

Tg(myl7:GFP) is a transgenic zebrafish line in which GFP is expressed in cardiomyocytes, and

240

it was used to evaluate the impact of formulation modification of BDQ on its inherent

241

cardiotoxicity. Zebrafish larvae were treated with BDQ, PLGA-BDQ NPs, and ME-PLGA-BDQ

242

NPs. Cardiac function, including stroke volume (SV), fractional shortening (%FS), cardiac output

243

(CO), and heart rate (HR), was monitored through video recordings and images of zebrafish

244

embryonic hearts.

245

Figure 9 shows that PLGA-BDQ NPs exhibited a certain degree of cardioprotective effect, while

246

ME-PLGA-BDQ NPs demonstrated the optimal cardiac safety profile among all tested groups,

247

with heart rates and cardiac function parameters closest to those of the normal control group. In

248

contrast, free BDQ induced cardiac dysfunction, showing significantly reduced SV, %FS, CO, and

249

HR compared to the control group. Compared to the free BDQ treatment, the formulation

250

modifications were able to mitigate the inherent cardiovascular toxicity effects of BDQ in this

251

zebrafish model.

252

253

Discussion

254

Mycobacterium tuberculosis (M. tb), a recalcitrant pathogen latent in nearly a quarter of the

255

global population, necessitates the urgent development of robust oral delivery systems for anti-

256

tuberculosis therapeutics

. While the piperazine-containing benzothiazinone PBTZ169

257

(macozinone) offers significant therapeutic potential

28,29

and bedaquiline remains pivotal for

258

managing drug-resistant strains, the latter is increasingly compromised by resistance mechanisms

259

involving Rv0678 mutations and subsequent MmpS5L5 efflux pump overexpression

260

Furthermore, despite their potent anti-TB efficacy, nitrobenzothiazinones (BTZs) such as BTZ043

261

Journal Pre-proof

and PBTZ169 are constrained by unfavorable physicochemical properties, specifically high

262

lipophilicity and limited aqueous solubility

. Addressing these multifaceted pharmacological

263

hurdles, this study presents a groundbreaking biomimetic platform that synergizes synthetic PLGA

264

nanoparticles with natural milk-derived exosomes (ME) to facilitate the co-delivery of PBTZ169

265

and bedaquiline.

266

MEs represent a promising vector for the oral delivery of protein and peptide therapeutics,

267

primarily attributed to their exceptional ability to traverse epithelial barriers. However, their utility

268

is constrained by intrinsic drawbacks, including suboptimal drug loading efficiency, limited mucus

269

penetration, and vulnerability to membrane protein loss

. In contrast, biodegradable microspheres

270

based on Poly(lactic-co-glycolic acid) (PLGA) are widely established in medical applications due

271

to their superior biocompatibility, low toxicity, and capacity to deliver a broad spectrum of drug

272

molecules via multiple pathways

. To synergize these complementary attributes, this study

273

engineers a novel bioinspired drug delivery system that integrates PLGA microspheres with ME

274

carriers, thereby addressing the individual limitations of each component while enhancing overall

275

therapeutic efficacy.

276

Key physicochemical determinants of oral drug bioavailability encompass lipophilicity,

277

hydrophilicity, and gastrointestinal barrier permeability

33,34

. Our characterization confirms the

278

successful engineering of ME-PLGA hybrid nanoparticles optimized for oral delivery. Contact

279

angle assays (Figure 3) reveal that the ME coating substantially enhances the hydrophilicity of

280

both PBTZ169 and BDQ, effectively mitigating the inherent constraints of these highly lipophilic

281

agents (ClogP values: 5.06 and 6.37). This surface modification is pivotal for gastrointestinal

282

absorption, corroborated by superior dissolution profiles in simulated intestinal fluid (Figure 5).

283

Notably, the ME-PLGA system exhibits pH-independent release kinetics, maintaining gastric

284

Journal Pre-proof

stability while enabling controlled intestinal release

—

a capability absent in conventional PLGA

285

nanoparticles

. Mechanistic analysis indicates that ME-coated particles remain intact in gastric

286

fluids but facilitate enhanced drug release in intestinal environments. This behavior likely stems

287

from the exosomal membrane providing physical and enzymatic shielding against gastric

288

degradation while promoting core exposure to intestinal factors such as bile salts, lipases,

289

hydration, and endosomal acidification

–

. Collectively, these factors accelerate water ingress,

290

PLGA hydrolysis (including autocatalysis), and drug solubilization, thereby minimizing gastric

291

loss and maximizing intestinal release compared to uncoated PLGA or free lipophilic drugs

–

292

Pharmacokinetic evaluations highlight distinct advantages of this platform. While prior studies

293

established synergistic efficacy between PBTZ169 and bedaquiline

, our most significant finding

294

is that ME-PLGA nanoparticles increased the relative bioavailability of both drugs by at least

295

twofold compared to free administration (Table S2-7), further amplifying therapeutic outcomes.

296

This enhancement arises from three synergistic mechanisms: (1) exosomal membrane components

297

facilitating mucosal penetration via intrinsic biological trafficking; (2) the PLGA core protecting

298

against gastric degradation; and (3) the nanoformulation dramatically improving aqueous

299

dispersibility of these hydrophobic compounds. Extended half-lives (t

1/2

) confirm the system

’

300

sustained-release capacity. These results align with established strategies employing PLGA

301

nanoparticles to boost oral bioavailability of hydrophobic drugs; for instance, previous work

302

demonstrated a ~4.9-fold increase in ergosterol bioavailability attributed to protective and

303

solubilizing effects

. Our study further validates nanoencapsulation as an effective strategy for

304

delivering poorly soluble therapeutics.

305

From a translational standpoint, milk-derived exosomes (MEs) offer distinct advantages in

306

scalability and cost-efficiency. Industrial-scale manufacturing substantially curtails expenses,

307

Journal Pre-proof

positioning MEs favorably against human- or stem cell-derived counterparts regarding production

308

scalability, industrialization barriers, and commercialization timelines

44,45

. These benefits stem

309

from three primary factors: first, the abundance and low cost of raw materials, as milk

—

a product

310

of large-scale agriculture

—

ensures stable supply at fractions of the cost associated with complex

311

human cell cultures; second, highly optimized production workflows, where industrial techniques

312

like tangential flow filtration (TFF) and size-exclusion chromatography facilitate continuous,

313

automated processing of vast volumes (tens to hundreds of liters), drastically reducing extraction

314

time and labor resources; and third, significant economies of scale, wherein unit costs decline

315

progressively as production volume expands.

316

Consequently, pending successful clinical safety validation, the deployment of the ME

‑

PLGA

‑

317

BDQ/PBTZ169 nanoformulation will necessitate a multi-stakeholder collaborative framework.

318

International procurement entities (e.g., the Global Fund) and pharmaceutical manufacturers must

319

secure accessibility in low- and middle-income countries via tiered pricing structures.

320

Concurrently, promoting voluntary licensing and technology transfer will enable regionalized

321

production to guarantee long-term supply continuity. Initially positioned as an optimized regimen

322

for drug-resistant tuberculosis, this formulation could be integrated into national strategic reserves.

323

Thus, realizing equitable patient access demands that international organizations, industry partners,

324

governments, and public health institutions collectively establish a sustainable translational

325

pathway.

326

In pharmacodynamic assessments, ME-PLGA nanoparticles elicited a substantial bacterial load

327

reduction of 0.85

–

2.5 log units relative to free drugs (Figure 8), markedly surpassing the efficacy

328

of uncoated PLGA counterparts. This enhanced therapeutic performance exhibits a strong

329

correlation with observed pharmacokinetic advancements, notably a 3- to 5-fold elevation in drug

330

Journal Pre-proof

accumulation within lung tissue, the primary site of tuberculosis infection. Parallel trends in

331

splenic drug concentrations further indicate superior systemic distribution capabilities.

332

Several key aspects of our findings warrant emphasis. First, consistent with prior reports

46,47

, the

333

nanoencapsulation process preserved the intrinsic antimicrobial potency of both drugs, as

334

evidenced by unchanged minimum inhibitory concentration (MIC) values (Table 3). Second, the

335

platform demonstrated excellent biocompatibility, confirming its suitability as an oral nanocarrier.

336

Furthermore, milk-derived exosome (ME) components not only enhanced natural drug

337

accumulation in target organs but also improved the stability of lyophilized formulations during

338

storage, underscoring their substantial translational value.Notably, this delivery platform achieved

339

over threefold higher drug concentrations in target organs at equivalent doses compared to free

340

drugs, a feature that may help mitigate dose-limiting toxicities. The potential to reduce dosing

341

frequency while maintaining therapeutic levels could significantly improve patient adherence to

342

tuberculosis treatment regimens. Economically, for established agents like bedaquiline, this

343

bioavailability-enhancing approach offers the prospect of superior therapeutic outcomes at reduced

344

overall treatment costs, thereby alleviating the financial burden on patients who currently face

345

expensive regimens. Moreover, compared to mammalian cell-derived exosomes, MEs offer

346

distinct advantages in scalable production. As an abundant and accessible source, milk endows

347

this platform with unique properties that extend its potential applications beyond tuberculosis to

348

include gastrointestinal disorders, skin ulcers, and cancer therapeutics

349

350

In conclusion, the ME-PLGA hybrid delivery platform represents a potential paradigm shift in

351

tuberculosis (TB) therapeutics, successfully synergizing pharmaceutical nanotechnology with

352

biological vector engineering. By simultaneously overcoming critical barriers

—

specifically

353

Journal Pre-proof

solubility limitations (evidenced by a ~60% reduction in solid-liquid contact angle), bioavailability

354

challenges (demonstrated by a 3- to 10-fold increase in target organ accumulation), and toxicity

355

concerns

—

this platform paves the way for developing patient-friendly, highly efficient oral anti-

356

TB regimens. Moreover, the versatility of this approach suggests broad applicability beyond TB,

357

offering a promising strategy for treating other infectious diseases.

358

359

Limitations of the study

360

While this study demonstrates promising improvements in drug delivery efficiency, several

361

limitations warrant careful consideration. First, although preliminary assessments using a zebrafish

362

model indicated that ME-PLGA-BDQ nanoparticles exhibited the most favorable cardiac safety

363

profile among tested groups

—

with heart rate and function parameters closely mirroring normal

364

controls

—

and that uncoated PLGA-BDQ also showed some cardioprotective effects (Figure 9),

365

these findings remain preliminary. The current evaluation lacks a systematic toxicity analysis of

366

the modified compounds, particularly regarding long-term safety validation in advanced

367

mammalian models. While these results suggest the delivery system may mitigate bedaquiline

’

368

(BDQ) inherent cardiovascular toxicity by altering release kinetics or biodistribution, definitive

369

confirmation is pending. Furthermore, the precise mechanisms governing intestinal absorption and

370

tissue targeting require further elucidation. Although the high lipophilicity of both drugs suggests

371

passive diffusion as a primary transport pathway, structural modifications within our delivery

372

system may also facilitate receptor-mediated transport processes. To address these gaps, future

373

research should prioritize: (1) Systematic Toxicity Assessment: Comprehensive evaluation in

374

mammalian models to assess potential reductions in hepatotoxicity, nephrotoxicity, and

375

cardiotoxicity. (2) Mechanistic Studies: Clarification of absorption pathways linked to

376

Journal Pre-proof

physicochemical properties and the specific mechanisms underlying the observed reduction in

377

cardiotoxicity. (3) Therapeutic Index Validation: Confirmation of efficacy and safety profiles in

378

multidrug-resistant tuberculosis (MDR-TB) models to support clinical translation. Crucially, given

379

the well-documented correlation between BDQ

’

s cardiotoxicity and its high lipophilicity, it

380

remains to be confirmed whether our delivery system effectively mitigates these adverse effects

381

through controlled release or altered biodistribution using comprehensive in vivo safety data.

382

Consequently, during the subsequent preclinical development phase supporting an Investigational

383

New Drug (IND) application, it is imperative to conduct systematic pharmacokinetic-toxicokinetic

384

assessments

—

including detection of the M2 metabolite

—

in higher-order mammalian models (e.g.,

385

canines or non-human primates) to fully elucidate the safety profile.

386

387

RESOURCE AVAILABILITY

388

The data that support the findings of this study are available from the corresponding author upon

389

reasonable request.

390

Lead Contact

: Further information and requests for resources and reagents should be directed

391

to and will be fulfilled by the Lead Contact, Yu Lu (luyu4876@hotmail.com).

392

Materials Availability

This study did not generate new unique reagents.

393

Data and Code Availability

Data (Data reported in this paper will be shared by the lead contact

394

upon request), Code (This paper does not report original code) and other items (Any additional

395

information required to reanalyse the data reported in this paper are available from the lead contact

396

upon request).

397

398

Author contributions

399

Journal Pre-proof

Authors

400

Eryue Liu (yueer1015087099@163.com

)- Beijing Key Laboratory of Drug Resistance

401

Tuberculosis Research, Beijing Chest Hospital, Capital Medical University, Beijing Tuberculosis

402

and Thoracic Tumor Research Institute, Beijing 101149

403

Chennan Liu

- Beijing Key Laboratory of Drug Resistance Tuberculosis Research, Beijing Chest

404

Hospital, Capital Medical University, Beijing Tuberculosis and Thoracic Tumor Research

405

Institute, Beijing 101149

406

Yangxue Ye

- Beijing Key Laboratory of Drug Resistance Tuberculosis Research, Beijing Chest

407

Hospital, Capital Medical University, Beijing Tuberculosis and Thoracic Tumor Research

408

Institute, Beijing 101149

409

Zimo Wang

- Beijing Key Laboratory of Drug Resistance Tuberculosis Research, Beijing Chest

410

Hospital, Capital Medical University, Beijing Tuberculosis and Thoracic Tumor Research

411

Institute, Beijing 101149

412

Weiyan Zhang

- Beijing Key Laboratory of Drug Resistance Tuberculosis Research, Beijing Chest

413

Hospital, Capital Medical University, Beijing Tuberculosis and Thoracic Tumor Research

414

Institute, Beijing 101149

415

Lei Fu

- Beijing Key Laboratory of Drug Resistance Tuberculosis Research, Beijing Chest

416

Hospital, Capital Medical University, Beijing Tuberculosis and Thoracic Tumor Research

417

Institute, Beijing 101149

418

Journal Pre-proof

Bin Wang

- Beijing Key Laboratory of Drug Resistance Tuberculosis Research, Beijing Chest

419

Hospital, Capital Medical University, Beijing Tuberculosis and Thoracic Tumor Research

420

Institute, Beijing 101149

421

Yujin Wang

- Beijing Key Laboratory of Drug Resistance Tuberculosis Research, Beijing Chest

422

Hospital, Capital Medical University, Beijing Tuberculosis and Thoracic Tumor Research

423

Institute, Beijing 101149

424

Yu Lu* (Email: luyu4876@hotmail.com

) - Beijing Key Laboratory of Drug Resistance

425

Tuberculosis Research, Beijing Chest Hospital, Capital Medical University, Beijing Tuberculosis

426

and Thoracic Tumor Research Institute, Beijing 101149

427

CRediT authorship contribution statement

428

Eryue Liu: Writing

–

review & editing, Writing

–

original draft, Methodology, Investigation,

429

Software, Data curation, Conceptualization, Formal analysis

430

Chennan Liu: Methodology, Investigation

431

Yangxue Ye: Investigation

432

Zimo Wang: Investigation

433

Weiyan Zhang: Investigation

434

Lei Fu: Validation, Supervision

435

Bin Wang: Investigation, Project administration

436

Yujin Wang: Validation, Supervision

437

Yu Lu: Writing

–

review & editing, Funding acquisition, Formal analysis

438

439

Acknowledgment

440

Journal Pre-proof

The authors acknowledge the following sources of support for this work:

441

Technical Assistance

: We thank Apeng Wang from the Institute of Biotechnology, Chinese

442

Academy of Medical Sciences, for providing the rotary evaporator for sample preparation.

443

Advice and Feedback

: We are deeply grateful to Wenjing Liu for her expert insights on sample

444

synthesis.

445

These contributions do not constitute a competing interest. Funding Sources

: This work was

446

supported by the National Natural Science Foundation of China (82173862) and the Beijing

447

Municipal Administration of Hospitals’ Ascent Plan (DFL20221402).

448

449

Declaration of interests

450

The authors declare no competing financial interest.

451

452

ABBREVIATIONS

453

TB: Tuberculosis; PBTZ169, MCZ: macozinone; BDQ: bedaquiline; MEs: milk exosomes; PLGA:

454

Poly lactic-co-glycolic acid; M. tb: Mycobacterium tuberculosis; WHO: World Health

455

Organization; MIC: minimum inhibitory concentration; EVs: Extracellular vesicles; DMEM:

456

Dulbecco's Modified Eagle Medium; FBS: fetal bovine serum; CCK-8: Cell Counting Kit-8;

457

EDTA: Ethylene Diamine Tetraacetic Acid; NPs: nanoparticles; PVA: polyvinyl alcohol; DLS:

458

dynamic light scattering; SGF: simulated gastric; SIF: intestinal fluid; MABA: microplate alamar

459

blue assay; TEM: Transmission electron microscopy; PDI: polydispersity index; WB: Western

460

Blot; BTZs: Nitrobenzothiazinones;

461

462

Journal Pre-proof

463

STAR

★

Methods

464

465

Animals and Ethics Statement

: Female BALB/c mice (6-8 weeks old, specific pathogen-free)

466

were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. The animal

467

study protocol was reviewed and approved by the Animal Ethics Committee of Beijing Chest

468

Hospital, Capital Medical University (Approval No. XK2024-206, 2024). All experiments were

469

conducted in strict accordance with the NIH Guide for the Care and Use of Laboratory Animals,

470

and every effort was made to minimize animal suffering and the number of animals used. The

471

study is reported in accordance with the ARRIVE guidelines.

472

473

EXPERIMENTAL MODEL AND STUDY PARTICIPANT DETAILS

474

Bacterial Strains: Mycobacterium tuberculosis H37Rv (ATCC 27294) and H37Ra (ATCC

475

25177) were used for in vitro antimicrobial susceptibility testing. Bacteria were cultured in

476

Middlebrook 7H9 broth (supplemented with 10% OADC and 0.05% Tween 80) or on 7H10 agar

477

plates.

478

Cell Lines: The following mammalian cell lines were used for cytotoxicity studies: HepG2

479

(human liver carcinoma), Vero (African green monkey kidney epithelial), Caco-2 (human

480

colorectal adenocarcinoma), HeLa (human cervical adenocarcinoma), and J774A.1 (murine

481

macrophage). Cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS) at

482

37°C in a 5% CO₂ atmosphere. Cell lines were obtained

from commercial sources and

483

authenticated by the suppliers.

484

Animal Models:

485

Journal Pre-proof

BALB/c Mice: 72 SPF-grade female BALB/c mice, aged 6

–

8 weeks and weighing 18

–

20 g,

486

were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. for the in vivo

487

pharmacodynamic evaluation of different formulations of PBTZ169 and BDQ. The mice were

488

housed in an air-conditioned room designed for infected animal models, with six animals per cage.

489

The temperature was maintained at 21

C, humidity was kept at 55%

15%, and a 12-hour

490

light/dark cycle was implemented. The animals had free access to sterilized feed and filtered water.

491

Before the experiment, the mice were allowed to acclimatize for 5 days.

492

An infection model was established by injecting a suspension of

Mycobacterium tuberculosis

493

H37Ra into the tail vein. Cultures in the logarithmic growth phase (OD600 value of 1.00

–

1.20)

494

were diluted with sterile saline to an OD600 of 0.1 prior to inoculation, achieving a bacterial load

495

of 4.0

–

5.0 log10 CFU in the lungs.

496

Treatment began 2 weeks post-infection and lasted for 2 weeks (5 days per week). The mice

497

were randomly divided into the following groups, with 9 mice per group: (1) Active

498

Pharmaceutical Ingredient (API) group (PBTZ169/BDQ suspended in 0.5% sodium

499

carboxymethyl cellulose solution), (2) PLGA NPs group (PLGA-PBTZ169 NPs/PLGA-BDQ NPs

500

dispersed in saline solution), (3) ME-PLGA NPs group (ME-PLGA-PBTZ169 NPs/ME-PLGA-

501

BDQ NPs dispersed in saline solution), (4) Vehicle control group (0.5% sodium carboxymethyl

502

cellulose solution and saline group), and (5) Untreated control group. The dose for all active

503

treatment groups was 10 mg/kg (based on the mass of the active ingredient), with an equivalent

504

dose defined as equal mass administration of the active pharmaceutical ingredient across different

505

formulations.

506

6 mice from each group were euthanized on day 1 post-infection (baseline), on the day treatment

507

began (day 0), and on the last day of treatment (day 14). Lung and spleen tissues were collected.

508

Journal Pre-proof

After homogenization, tissue samples were plated onto 7H10 solid medium containing 0.4%

509

charcoal and incubated at 37°C with 5% CO₂ for 4 weeks for colony counting. The remaining three

510

mice in each group were used for pharmacokinetic analysis. Lung and spleen samples were

511

collected at the peak drug concentration time point on day 14 to measure drug concentrations.

512

Zebrafish: Adult Tg(myl7:GFP) zebrafish (CZRC ID: CZ56) were obtained from the China

513

Zebrafish Resource Center and maintained under standard conditions (28

C, 14-hour light/10-

514

hour dark cycle). Embryos were generated by natural crossing and cultured in E3 medium with 0.2

515

mM N-phenylthiourea (PTU) added from 1 day post-fertilization (dpf) to inhibit pigmentation.

516

Larvae of mixed sexes were used for cardiotoxicity assessment at 2 dpf.

517

The animal study protocol was reviewed and approved by the Animal Ethics Committee of

518

Beijing Chest Hospital, Capital Medical University (Approval No. XK2024-206, 2024). All

519

experiments were conducted in strict accordance with the NIH Guide for the Care and Use of

520

Laboratory Animals, and every effort was made to minimize animal suffering and the number of

521

animals used. The study is reported in accordance with the ARRIVE guidelines.

522

523

Method DETAILS

524

Isolation of MEs

525

The classical differential centrifugation (ultracentrifugation) method was used to extract MEs

526

To remove somatic cells and debris, pasteurized bovine skim milk was centrifuged for 30 min at

527

4 °C at 4000 ×g, followed by 13,000 ×g for 30 min. To precipitate milk casein, the supernatant

528

was mixed with 0.5 M Ethylene Diamine Tetraacetic Acid (EDTA) solution (3:1, v/v) and stirred

529

on ice for 30 min. Subsequently, the suspension underwent ultracentrifugation at a speed of

530

100,000 ×g (Optima TM L-100 XP, Beckman, USA) for 60 min, precipitating microvesicles and

531

Journal Pre-proof

residual proteins. In the end, pellets of exosomes were obtained after centrifuging at 140,000 ×g

532

for 90 min. The bicinchoninic acid (BCA) assay kit (Beyotime, P0010) was used to determine the

533

exosome pellets' total protein concentration after being resuspended in 10 mM phosphate-buffered

534

saline (PBS, pH 7.4). 20 μL of each sample or protein standard was mixed with 200 μL of BCA

535

working solution

(reagent A: regent B=50: 1, v/v)

in a 96-well plate. After incubation at 37 °C for

536

30 min, the absorbance was measured at 562 nm using a microplate reader. Sample protein

537

concentrations were calculated based on the standard curve. Any ME samples with a total protein

538

concentration below 5 mg/mL were stored at −80 °C until needed

539

Fabrication of PLGA-PBTZ169/BDQ Nanoparticles and Preparation of ME-Coated

540

Nanoparticles

541

PBTZ169/BDQ-loaded PLGA nanoparticles (PLGA-PBTZ169/BDQ NPs) were prepared using

542

the oil/water (O/W) single-emulsion solvent evaporation method. 10 mg of PBTZ169/BDQ and

543

100 mg of poly(D,l-lactide-

-glycolide)(50:50)-b-poly(ethylene glycol) (PLGA; Sigma-Aldrich,

544

USA) were dissolved in 10 mL of dichloromethane (DCM; Sigma-Aldrich, USA). The PLGA-

545

PBTZ169/BDQ NPs mixture was then added to 50 mL of 5 % polyvinyl alcohol (PVA) solution

546

(Sigma-Aldrich, USA) in distilled water, which served as the surfactant. The solution was then

547

subjected to ultrasonication (SONICS, USA) at 40 W and 4 °C for 5 min 30 s, followed by solvent

548

evaporation using a rotary evaporator at 25 °C. The PLGA-PBTZ169/BDQ NPs were collected by

549

ultracentrifugation (Beckman Coulter Optima L-90 K) at 8000 ×

for 30 min. The pellets

550

containing PLGA-PBTZ169/BDQ NPs were washed three times with distilled water. After

551

lyophilization, the PLGA-PBTZ169/BDQ NPs were stored at 4°C until further use. The

552

encapsulation efficiency (EE) and drug loading (DL) of PLGA-PBTZ169/BDQ NPs were

553

quantified by UV-Vis spectrophotometry.

554

Journal Pre-proof

The DL and EE of PLGA-PBTZ169/BDQ NPs can be calculated using the following formulas:

555

𝐷𝐿(%) =

𝑊

Drug

𝑊

All

× 100%

556

𝐸𝐸(%) =

𝑊

Drug

𝑊

Total

× 100%

557

Where

𝑊

All

represents the combined mass of the carrier and the drug,

𝑊

Total

refers to the total

558

amount of drug initially added in the experiment.

559

The milk-exosome was mixed with the PLGA core for cell membrane coating at a polymer-to-

560

membrane protein weight ratio of 1:1. The mixture was then sonicated with a bath sonicator (Fisher

561

Scientific FS30D) for 3 min. After that, excess MEs were removed by centrifugation at 12,000 ×

562

for 30 min. The pellets containing ME-PLGA-PBTZ169/BDQ NPs were washed three times with

563

distilled water, lyophilized, and stored at −

80°C until further use.

To evaluate storage stability, the

564

nanoparticles were preserved at −80 °C for 1 month, a

nd their particle size and zeta potential were

565

measured on the first day and the thirtieth day.

566

Physicochemical Characterization of Nanoparticles

567

The size (diameter, nm) and surface zeta potential (mV) of the nanoparticles were measured by

568

using dynamic light scattering (DLS, NanoBrook 90plus PALS). For studying morphology, the

569

nanoparticle samples were adsorbed onto carbon-coated copper grids (400 mesh, Electron

570

Microscopy Sciences) and stained with 0.2 wt% uranyl acetate (Electron Microscopy Sciences).

571

The grids were imaged on a Hitachi HT-7800 transmission electron microscope. We performed

572

Western blot analysis to verify the integrity of isolated ME and their successful encapsulation in

573

PLGA nanoparticles (ME-PLGA-PBTZ169 NPs and ME-PLGA-BDQ NPs). Samples were

574

prepared using RIPA lysis buffer containing protease inhibitors. β

-actin serves as a loading control.

575

The primary antibodies used in this study include anti-CD63 (abs123229, absin), anti-CD81

576

Journal Pre-proof

(abs126599, absin), anti-TSG101 (ba-1365R, bioss). Secondary antibodies used in Western blot

577

analysis include goat anti-rabbit IgG-HRP (S0001, Affinity) and goat anti-mouse IgG-HRP (S0002,

578

Affinity).

579

Measurement of Surface Static Water Contact Angle

580

The wettability characteristics of the test samples were evaluated through static contact angle

581

measurements using a standard sessile drop technique (Model JY-82C, Chengde Dingsheng

582

Testing Equipment). Measurements were conducted under ambient laboratory conditions (25 ±

583

1 °C) for: Pure drug compounds (PBTZ169 / BDQ), PLGA-PBTZ169 / PLGA-BDQ NPs, ME-

584

PLGA-PBTZ169 / ME-PLGA-BDQ NPs. Three replicate measurements were obtained for each

585

sample, with results expressed as mean ± standard deviation. Before testing, all samples were

586

equilibrated under measurement conditions for 30 min to ensure thermal stability.

587

In Vitro Release Kinetics of Drug Formulations in Simulated Gastrointestinal Media

588

The release profiles of PBTZ169/BDQ in three formulations (free drug, PLGA nanoparticles,

589

and ME-PLGA nanoparticles) were assessed in simulated gastric fluid (SGF:

pH=1.2

) and

590

simulated intestinal fluid (SIF: pH=6.8). For each formulation, 5 mg samples were suspended in 5

591

mL of either SGF or SIF, followed by incubation at 37 °C with continuous agitation (100 rpm) for

592

36 h. Aliquots were collected at 0, 2, 4, 6, 8, 10, 12, 24, and 36 h, and drug concentrations were

593

determined by UV-

vis spectroscopy at λ

max

values of 346 nm for PBTZ169 and 333 nm for BDQ,

594

respectively.

595

Assessment of In Vitro Antibacterial Activity of Drug-Loaded Formulations

596

The antibacterial activity of different drug formulations was evaluated by determining their

597

minimum inhibitory concentrations (MICs) against the H37Rv and H37Ra reference strain using

598

the Microplate Alamar Blue Assay (MABA). Five experimental groups were included: (1) free

599

Journal Pre-proof

PBTZ169/BDQ, (2) PLGA-PBTZ169/BDQ NPs, (3) ME-PLGA-PBTZ169/BDQ NPs, (4) PLGA-

600

NPs, and (5) ME group.

601

The assay was performed using two-fold serial dilutions of each formulation in 96-well plates.

602

Each well received 100 μL of bacterial

suspension (2×10

CFU/mL) in Middlebrook 7H9 broth,

603

bringing the total volume to 200 μL. Following incubation at 37

°C for 7 days, 12.5 μL of 20%

604

Tween 80 and 20 μL of alamarBlue reagent were added to each well.

After an additional 24 h

605

incubation, fluorescence was measured (excitation 530 nm/emission 590 nm) using a microplate

606

reader. The MIC was defined as the lowest drug concentration that inhibited

≥

90% of bacterial

607

growth compared to drug-free controls, with

M. tuberculosis

H37Rv and H37Ra serving as the

608

drug-susceptible control strain.

609

Evaluation of In Vitro Cytotoxicity

610

Cytotoxicity assessment of drug-loaded nanoparticle formulations (PBTZ169/BDQ, PLGA-

611

PBTZ169/BDQ NPs, and ME-PLGA-PBTZ169/BDQ NPs) and two blank carrier materials

612

(PLGA-NPs and ME-PLGA-NPs) performed in HepG2, Vero, Caco-2, HeLa, and J774A.1 cells

613

using the CCK-8 assay. Cells were plated in 96-well plates at a density of 10,000 cells per well

614

and allowed to adhere during a 24 h incubation period in DMEM at 37 °C. After the initial culture,

615

the growth medium was replaced with fresh medium containing the test formulations at specified

616

concentrations: blank carrier materials (10 mg/mL) or drug-loaded nanoparticle formulations (drug

617

concentration gradient: 8, 16, 32, 64, 128 μg/mL).

Cells were then incubated with these

618

formulations for 48 hours. After the 48-hour treatment period, the medium in each well was

619

carefully aspirated and replaced with 100

L of fresh medium containing 10% CCK-8 reagent.

620

Wells containing medium plus CCK-8 reagent without cells served as blank controls, and wells

621

with cells in medium without any treatment served as cell-only controls (100% viability). The

622

Journal Pre-proof

absorbance of each well was measured at 450 nm using a microplate reader. Cell viability was

623

calculated as a percentage relative to the cell-only control group using the following formula:Cell

624

Viability (%) = [(OD

sample

-OD

blank

)/(OD

control

-OD

blank

)] *100%

625

sample

represents the OD

450

value of wells containing cells treated with either drug-loaded

626

formulations or blank carrier materials. OD

blank

represents the OD

450

values of wells containing

627

culture medium only (blank control). OD

control

represents the OD

450

value of wells containing

628

untreated cells (negative control).

629

Investigation of Pharmacokinetics and Bioavailability of PBTZ169 Formulations in

630

BALB/c Mice

631

All experimental procedures involving animals were approved by the Beijing Chest Hospital,

632

affiliated with Capital Medical University Committee on Animal Use and Care. All Mice (BALB/c,

633

female, 6-8 weeks, 18-20g) were obtained from Beijing Vital River Laboratory Animal

634

Technology Co., Ltd. Mice were randomly assigned to each group.

The pharmacokinetic profile

635

and relative bioavailability of different PBTZ169 formulations were evaluated following single-

636

dose administration in BALB/c mice (n = 3 per group) after 12 hours of fasting. Three formulations

637

- free PBTZ169 (suspended in 0.5% carboxymethyl cellulose), PLGA-PBTZ169 nanoparticles (in

638

normal saline), and ME-PLGA-PBTZ169 (in normal saline) - were administered orally at 10

639

mg/kg PBTZ169. Blood and tissue samples were collected at 0.25, 0.5, 1, 3, 8, and 24 h post-

640

dosing. Blood samples were centrifuged within 1 h of collection to obtain plasma, which was

641

subsequently stored at -80°C. Tissue samples (lung and spleen) were homogenized in ice-cold

642

methanol, followed by protein precipitation with acetonitrile containing an internal standard.

643

The concentrations of PBTZ169 in plasma and tissue homogenates were quantified using liquid

644

chromatography-tandem mass spectrometry (LC-MS/MS). Pharmacokinetic parameters (AUC

0-inf

645

Journal Pre-proof

1/2

, C

max

, T

max

) were derived using the non-compartmental analysis module of WinNonlin®

646

(v6.2.1), with relative bioavailability determined through AUC comparisons across formulations.

647

This comprehensive analysis demonstrated distinct pharmacokinetic behaviors and bioavailability

648

characteristics among the different PBTZ169 delivery systems.

649

Investigation of Pharmacokinetics and Bioavailability of BDQ Formulations in BALB/c

650

Mice

651

The pharmacokinetic profile and relative bioavailability of different BDQ formulations were

652

evaluated following single-dose administration in BALB/c mice (n=3 per group) after 12 hours of

653

fasting.

Mice were purchased and randomly assigned to each group.

Three formulations - free

654

BDQ (suspended in 0.5% carboxymethyl cellulose), PLGA-BDQ nanoparticles (in normal saline),

655

and ME-PLGA-BDQ (in normal saline) - were administered orally at 10 mg/kg BDQ. Blood and

656

tissue samples collected at 1, 3, 8, 24, 48, and 72 h post-dosing were processed to obtain plasma

657

(centrifuged within 1 h and stored at -80°C) and tissue homogenates (lung/spleen in ice-cold

658

methanol, precipitated with acetonitrile containing internal standard). The concentrations of BDQ

659

in plasma and tissue homogenates were quantified using LC-MS/MS. Pharmacokinetic parameters

660

(AUC

0-inf

, T

1/2

, C

max

, T

max

) were derived using the non-compartmental analysis module of

661

WinNonlin® (v6.2.1), with relative bioavailability determined through AUC comparisons across

662

formulations. This comprehensive analysis demonstrated distinct pharmacokinetic behaviors and

663

bioavailability characteristics among the different BDQ delivery systems.

664

10.

Pharmacodynamic Studies

665

Mice were purchased and acclimatized in a Biosafety Level 2 (BSL-2) facility.

A total of 100

666

female BALB/c mice (6-8 weeks, 18

–

20g) were intravenously inoculated with

Mycobacterium

667

tuberculosis

H37Ra via the tail vein to establish an infection model. To achieve a pulmonary

668

Journal Pre-proof

bacterial load of 4.0

–

5.0 log

CFU, mid-log phase cultures (OD

600

1.00

–

1.20) were diluted to

669

600

of 0.1 in sterile saline prior to administration. Following a 2-week infection period,

670

treatment was initiated and maintained for 2 weeks (5 days/week) with the following groups

671

(n=9/group, mice were randomly assigned): (1) API group (PBTZ169/BDQ in 0.5% CMC

672

solution), (2) PLGA NPs group (PLGA-PBTZ169 NPs/PLGA-BDQ NPs in normal saline), (3)

673

ME-PLGA NPs group (ME-PLGA-PBTZ169 NPs/ME-PLGA-BDQ NPs in normal saline), (4)

674

Vehicle controls (0.5% CMC and normal saline groups), and (5) Untreated control group. All

675

active treatment groups were administered at a dose of 10 mg/kg (based on the mass equivalence

676

of the active pharmaceutical ingredient). The equivalent dose is defined as the equal mass of the

677

active drug in different formulations.

We euthanized mice through CO

inhalation.

For sample

678

collection and analysis, six mice per group were sacrificed at baseline (1 day post-infection, D-

679

13), treatment initiation (Day 0, D0), and study endpoint (Day 14 post-infection, D14). Lung and

680

spleen homogenates were plated on 0.4% (w/v) charcoal-supplemented 7H10 agar and incubated

681

at 37°C under 5%

CO₂ for 4 weeks before CFU enumeration. The remaining three mice per group

682

were used for pharmacokinetic analysis, with plasma, lung, and spleen samples collected at each

683

drug’s T

max

on D14 for drug concentration determination.

684

11.

Zebrafish maintenance, husbandry, and Cardiotoxicity evaluation

685

This experimental procedure was approved by the Beijing Chest Hospital affiliated Capital

686

Medical University Committee on Animal Use and Care.

Adult zebrafish were purchased from the

687

China Zebrafish Resource Center (CZRC). The zebrafish and embryo cultures followed standard

688

procedures. They were maintained under a 14-hour light/10-hour dark cycle. Embryos were

689

generated by crossing adult zebrafish aged 2 months to 1 year. The Tg(myl7:GFP) zebrafish line

690

(CZRC ID: CZ56) was used in this study, with individuals of mixed sexes. Embryos and larvae

691

Journal Pre-proof

were cultured at 28 °C in E3 medium (5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2, 0.33 mM

692

MgSO4; Deepspace Bio, China). From 1 day post-fertilization (dpf) onward, 0.2 mM N-

693

phenylthiourea (PTU; Rhawn, China) was added to prevent pigment formation and avoid

694

interference with imaging.

695

Tg(myl7:GFP) zebrafish larvae were used for cardiovascular function studies. Healthy 2 dpf

696

zebrafish larvae were employed and randomly placed into 12-well microplates, with 10

–

15 larvae

697

per well, and treated with 1 mL of either free drug BDQ (72 μM) or an equivalent mass

698

concentration of PLGA-BDQ NPs and ME-PLGA-BDQ NPs, respectively.

699

After a 2-day incubation period, zebrafish embryos were embedded in 1% (wt%) low-melting-

700

point agarose to fix their orientation and position and to restrict movement. At the end of the

701

incubation, a 15s video of the beating heart of individual larvae was recorded using a Leica

702

fluorescence stereomicroscope M165 FC at room temperature to capture heart morphology and

703

enable quantitative assessment of ventricular function. Ventricular function was evaluated based

704

on heart rate (HR), stroke volume (SV), cardiac output (CO), and fractional shortening (FS).

705

Heart rate was determined by counting the number of heartbeats within a 15s interval. Video

706

images were used to measure the longitudinal axis length (a) and lateral axis length (b) between

707

myocardial borders of the ventricles at end-diastole and end-systole, respectively. Ventricular end-

708

diastolic volume (EDV) and end-systolic volume (ESV) in larvae were calculated from heart

709

dimensions using the formula for a prolate spheroid: V = 4/3πab². SV, CO, and %FS were

710

calculated as follows: SV = (EDV − ESV), CO = SV × HR, %FS = (Diastolic diameter − Systolic

711

diameter) / (Systolic diameter) × 100%.

712

The average value of each parameter obtained from three independent biological replicates was

713

used to assess the cardiac function of zebrafish larvae.

714

Journal Pre-proof

QUANTIFICATION AND STATISTICAL ANALYSIS

715

CFU counts were log

-transformed before analysis. To compare two groups, a two-tailed

716

Student

’

s t-test was utilized, while for comparisons involving multiple groups, one-way analysis

717

of variance (ANOVA) followed by the Newman

–

Keuls post hoc test was employed. All analyses

718

were conducted using GraphPad Prism v5 (GraphPad Software, San Diego, CA), with

0.05

719

considered statistically significant.

Here, all statistical details of the experiment can be found in

720

the figure legends, where n represents the sample size.

721

ADDITIONAL RESOURCES

722

This study did not generate or utilize any additional resources, such as dedicated websites or

723

databases.

724

725

Journal Pre-proof

726

Reference

727

1. Yang, H., Ruan, X., Li, W., Xiong, J., and Zheng, Y. (2024). Global, regional, and national

728

burden of tuberculosis and attributable risk factors for 204 countries and territories, 1990-

729

2021: a systematic analysis for the Global Burden of Diseases 2021 study. BMC Public Health

730

, 3111. https://doi.org/10.1186/s12889-024-20664-w.

731

2. Saukkonen, J.J., Duarte, R., Munsiff, S.S., Winston, C.A., Mammen, M.J., Abubakar, I.,

732

Acuña-Villaorduña, C., Barry, P.M., Bastos, M.L., Carr, W., et al. (2025). Updates on the

733

Treatment of Drug-Susceptible and Drug-Resistant Tuberculosis: An Official

734

ATS/CDC/ERS/IDSA Clinical Practice Guideline. Am. J. Respir. Crit. Care Med.

211

, 15

–

33.

735

https://doi.org/10.1164/rccm.202410-2096ST.

736

3. Shi, J., Lu, J., Wen, S., Zong, Z., Huo, F., Luo, J., Liang, Q., Li, Y., Huang, H., and Pang, Y.

737

(2018). In Vitro Activity of PBTZ169 against Multiple Mycobacterium Species. Antimicrob.

738

Agents Chemother.

. https://doi.org/10.1128/AAC.01314-18.

739

4. Makarov, V., Lechartier, B., Zhang, M., Neres, J., Sar, A.M., Raadsen, S.A., Hartkoorn, R.C.,

740

Ryabova, O.B., Vocat, A., Decosterd, L.A., et al. (2014). Towards a new combination therapy

741

for tuberculosis with next generation benzothiazinones. EMBO Mol. Med.

, 372.

742

https://doi.org/10.1002/emmm.201303575.

743

Fan, D., Wang, B., Stelitano, G., Savková, K., Shi, R., Huszár, S., Han, Q., Mikušová, K.,

744

Chiarelli, L.R., Lu, Y., et al. (2021). Structural and Activity Relationships of 6-Sulfonyl-8-

745

Nitrobenzothiazinones

Antitubercular

Agents.

Med.

Chem.

746

https://doi.org/10.1021/acs.jmedchem.1c01049.

747

6. Lv, K., You, X., Wang, B., Wei, Z., Chai, Y., Wang, B., Wang, A., Huang, G., Liu, M., and

748

Lu, Y. (2017). Identification of Better Pharmacokinetic Benzothiazinone Derivatives as New

749

Antitubercular

Agents.

ACS

Med.

Chem.

Lett.

636.

750

https://doi.org/10.1021/acsmedchemlett.7b00106.

751

7. McLeay, S.C., Vis, P., Heeswijk, R.P.G. van, and Green, B. (2014). Population

752

Pharmacokinetics of Bedaquiline (TMC207), a Novel Antituberculosis Drug. Antimicrob.

753

Agents Chemother.

, 5315. https://doi.org/10.1128/AAC.01418-13.

754

8. Okezue, M.A., Byrn, S.J., and Clase, K.L. (2022). Determining the solubilities for benzoate,

755

nicotinate, hydrochloride, and malonate salts of bedaquiline. Int. J. Pharm.

627

, 122229.

756

https://doi.org/10.1016/j.ijpharm.2022.122229.

757

9. Supramolecular strategy for reducing the cardiotoxicity of bedaquiline without compromising

758

its antimycobacterial efficacy (2018). Food Chem. Toxicol.

119

, 425

–

429.

759

https://doi.org/10.1016/j.fct.2017.12.022.

760

Journal Pre-proof

10. Field, S.K. (2015). Bedaquiline for the treatment of multidrug-resistant tuberculosis: great

761

promise

disappointment?

Ther.

Adv.

Chronic

Dis.

170.

762

https://doi.org/10.1177/2040622315582325.

763

11. Tong, A.S.T., Choi, P.J., Blaser, A., Sutherland, H.S., Tsang, S.K.Y., Guillemont, J., Motte,

764

M., Cooper, C.B., Andries, K., Broeck, W.V. den, et al. (2017). 6-Cyano Analogues of

765

Bedaquiline as Less Lipophilic and Potentially Safer Diarylquinolines for Tuberculosis. ACS

766

Med. Chem. Lett.

, 1019. https://doi.org/10.1021/acsmedchemlett.7b00196.

767

12. Khatib, R., Sabir, F.R.N., Omari, C., Pepper, C., and Tayebjee, M.H. (2021). Managing drug-

768

induced QT prolongation in clinical practice. Postgrad. Med. J.

, 452

–

458.

769

https://doi.org/10.1136/postgradmedj-2020-138661.

770

13. Sharma, A., Sharma, S., and Khuller, G.K. (2004). Lectin-functionalized poly (lactide-co-

771

glycolide) nanoparticles as oral/aerosolized antitubercular drug carriers for treatment of

772

tuberculosis. J. Antimicrob. Chemother.

, 761

–

766. https://doi.org/10.1093/jac/dkh411.

773

14. Danhier, F., Ansorena, E., Silva, J.M., Coco, R., Le Breton, A., and Préat, V. (2012). PLGA-

774

based nanoparticles: An overview of biomedical applications. J. Controlled Release

161

, 505

–

775

522. https://doi.org/10.1016/j.jconrel.2012.01.043.

776

15. He, Y., Chen, Q.-W., Yu, J.-X., Qin, S.-Y., Liu, W.-L., Ma, Y.-H., Chen, X.-S., Zhang, A.-Q.,

777

Zhang, X.-Z., and Cheng, Y.-J. (2023). Yeast cell membrane-camouflaged PLGA nanoparticle

778

platform for enhanced cancer therapy. J. Controlled Release

359

, 347

–

358.

779

https://doi.org/10.1016/j.jconrel.2023.05.043.

780

16. Pozo-Acebo, L. del, Hazas, M.-C.L. de las, Tomé-Carneiro, J., Gil-Cabrerizo, P., San-

781

Cristobal, R., Busto, R., García-Ruiz, A., and Dávalos, A. (2021). Bovine Milk-Derived

782

Exosomes as a Drug Delivery Vehicle for miRNA-Based Therapy. Int. J. Mol. Sci.

, 1105.

783

https://doi.org/10.3390/ijms22031105.

784

17. Munagala, R., Aqil, F., Jeyabalan, J., and Gupta, R.C. (2015). Bovine milk-derived exosomes

785

for drug delivery. Cancer Lett.

371

, 48. https://doi.org/10.1016/j.canlet.2015.10.020.

786

18. Kim, M.S., Haney, M.J., Zhao, Y., Mahajan, V., Deygen, I., Klyachko, N.L., Inskoe, E.,

787

Piroyan, A., Sokolsky, M., Okolie, O., et al. (2016). Development of exosome-encapsulated

788

paclitaxel to overcome MDR in cancer cells. Nanomedicine Nanotechnol. Biol. Med.

, 655

–

789

664. https://doi.org/10.1016/j.nano.2015.10.012.

790

19. Li, Y., Xing, L., Wang, L., Liu, X., Wu, L., Ni, M., Zhou, Z., Li, L., Liu, X., and Huang, Y.

791

(2023). Milk-derived exosomes as a promising vehicle for oral delivery of hydrophilic

792

biomacromolecule

drugs.

Asian

Pharm.

Sci.

100797.

793

https://doi.org/10.1016/j.ajps.2023.100797.

794

20. Xu, Q., Yang, S., Zhang, K., Liu, Y., Li, L., and Qu, S. (2024). Enhanced antibacterial activity

795

of bovine milk exosome-based drug formulation against bacterial pathogens. Food Chem.

447

796

139034. https://doi.org/10.1016/j.foodchem.2024.139034.

797

Journal Pre-proof

21. Xiao, P., Wang, H., Liu, H., Yuan, H., Guo, C., Feng, Y., Qi, P., Yin, T., Zhang, Y., He, H.,

798

et al. (2024). Milk Exosome-Liposome Hybrid Vesicles with Self-Adapting Surface Properties

799

Overcome the Sequential Absorption Barriers for Oral Delivery of Peptides. ACS Nano

800

21091

–

21111. https://doi.org/10.1021/acsnano.4c02560.

801

22. Betker, J.L., Angle, B.M., Graner, M.W., and Anchordoquy, T.J. (2019). The Potential of

802

Exosomes From Cow Milk for Oral Delivery. J. Pharm. Sci.

108

, 1496

–

1505.

803

https://doi.org/10.1016/j.xphs.2018.11.022.

804

23. Rani, P., Vashisht, M., Golla, N., Shandilya, S., Onteru, S.K., and Singh, D. (2017). Milk

805

miRNAs encapsulated in exosomes are stable to human digestion and permeable to intestinal

806

barrier in vitro. J. Funct. FOODS

, 431

–

439. https://doi.org/10.1016/j.jff.2017.05.009.

807

24. Yan, C., Chen, J., Wang, C., Yuan, M., Kang, Y., Wu, Z., Li, W., Zhang, G., Machens, H.-G.,

808

Rinkevich, Y., et al. (2022). Milk exosomes-mediated miR-31-5p delivery accelerates diabetic

809

wound healing through promoting angiogenesis. Drug Deliv.

, 214

–

228.

810

https://doi.org/10.1080/10717544.2021.2023699.

811

25. Qu, S., Han, Y., Liu, Y., Zhu, J., Acaroz, U., Shen, J., and Zhu, K. (2022). Milk Exosomes

812

Facilitate Oral Delivery of Drugs against Intestinal Bacterial Infections. J. Agric. Food Chem.

813

, 16069

–

16079. https://doi.org/10.1021/acs.jafc.2c04971.

814

26. Cui, Z., Amevor, F.K., Zhao, X., Mou, C., Pang, J., Peng, X., Liu, A., Lan, X., and Liu, L.

815

(2023). Potential therapeutic effects of milk-derived exosomes on intestinal diseases. J.

816

NANOBIOTECHNOLOGY

. https://doi.org/10.1186/s12951-023-02176-8.

817

27. Alsayed, S.S.R., and Gunosewoyo, H. (2023). Tuberculosis: Pathogenesis, Current Treatment

818

Regimens and New Drug Targets. Int. J. Mol. Sci.

. https://doi.org/10.3390/ijms24065202.

819

28. Lupien, A., Vocat, A., Foo, C.S.-Y., Blattes, E., Gillon, J.-Y., Makarov, V., and Cole, S.T.

820

(2018). Optimized Background Regimen for Treatment of Active Tuberculosis with the Next-

821

Generation Benzothiazinone Macozinone (PBTZ169). Antimicrob. AGENTS Chemother.

822

https://doi.org/10.1128/AAC.00840-18.

823

29. Diab, A., Dickerson, H., and Al Musaimi, O. (2025). Targeting the Heart of Mycobacterium:

824

Advances in Anti-Tubercular Agents Disrupting Cell Wall Biosynthesis. Pharm. Basel Switz.

825

. https://doi.org/10.3390/ph18010070.

826

30. Fountain, A.J., Waller, N.J.E., Cheung, C.-Y., Jowsey, W., Chrisp, M.T., Troll, M., Edelstein,

827

P.H., Cook, G.M., Mcneil, M.B., and Ramakrishnan, L. (2025). Verapamil and its metabolite

828

norverapamil inhibit the Mycobacterium tuberculosis MmpS5L5 efflux pump to increase

829

bedaquiline

activity.

Proc.

Natl.

Acad.

Sci.

122

830

https://doi.org/10.1073/pnas.2426827122.

831

31. Zhong, X., Wu, J., Du, N., Zhou, S., Ma, C., Xue, T., Wei, M., Gong, J., Wang, B., Liu, M.,

832

et al. (2024). Design, synthesis and antimycobacterial activity of novel benzothiazinones with

833

improved

water

solubility.

Eur.

Med.

Chem.

279

116829.

834

https://doi.org/10.1016/j.ejmech.2024.116829.

835

Journal Pre-proof

32. Su, Y., Zhang, B., Sun, R., Liu, W., Zhu, Q., Zhang, X., Wang, R., and Chen, C. (2021).

836

PLGA-based biodegradable microspheres in drug delivery: recent advances in research and

837

application. Drug Deliv.

, 1397

–

1418. https://doi.org/10.1080/10717544.2021.1938756.

838

33. Di, L., Artursson, P., Avdeef, A., Benet, L.Z., Houston, J.B., Kansy, M., Kerns, E.H.,

839

Lennernäs, H., Smith, D.A., and Sugano, K. (2020). The Critical Role of Passive Permeability

840

Designing

Successful

Drugs.

ChemMedChem

1862

–

1874.

841

https://doi.org/10.1002/cmdc.202000419.

842

34. Luo, Z., Paunovi

ć, N., and Leroux, J.

-C. (2021). Physical methods for enhancing drug

843

absorption from the gastrointestinal tract. Adv. Drug Deliv. Rev.

175

, 113814.

844

https://doi.org/10.1016/j.addr.2021.05.024.

845

35. Kambayashi, A., and Shirasaka, Y. (2023). Food effects on gastrointestinal physiology and

846

drug

absorption.

Drug

Metab.

Pharmacokinet.

100488.

847

https://doi.org/10.1016/j.dmpk.2022.100488.

848

36. Wu, L., Wang, L., Liu, X., Bai, Y., Wu, R., Li, X., Mao, Y., Zhang, L., Zheng, Y., Gong, T.,

849

et al. (2022). Milk-derived exosomes exhibit versatile effects for improved oral drug delivery.

850

Acta Pharm. Sin. B

, 2029

–

2042. https://doi.org/10.1016/j.apsb.2021.12.015.

851

37. Liao, Y., Du, X., Li, J., and Lönnerdal, B. (2017). Human milk exosomes and their microRNAs

852

survive digestion in vitro and are taken up by human intestinal cells. Mol. Nutr. Food Res.

853

https://doi.org/10.1002/mnfr.201700082.

854

38. Mayer, J.B., Patil, S.M., Shin, S.-H., Yoo, J., and Won, Y.-Y. (2025). Internal Water-Induced

855

Acceleration, Chemical Pathways, and Contributing Factors in the Degradation of Poly(lactic-

856

co-glycolic acid) (PLGA) Microparticles and Devices. ACS Biomater. Sci. Eng.

, 3932

–

857

3948. https://doi.org/10.1021/acsbiomaterials.5c00419.

858

39. Siepmann, J., Elkharraz, K., Siepmann, F., and Klose, D. (2005). How autocatalysis

859

accelerates drug release from PLGA-based microparticles: a quantitative treatment.

860

Biomacromolecules

, 2312

–

2319. https://doi.org/10.1021/bm050228k.

861

40. Aburahma, M.H. (2016). Bile salts-containing vesicles: promising pharmaceutical carriers for

862

oral delivery of poorly water-soluble drugs and peptide/protein-based therapeutics or

863

vaccines. Drug Deliv.

, 1847

–

1867. https://doi.org/10.3109/10717544.2014.976892.

864

41. Agrawal, A.K., Aqil, F., Jeyabalan, J., Spencer, W.A., Beck, J., Gachuki, B.W., Alhakeem,

865

S.S., Oben, K., Munagala, R., Bondada, S., et al. (2017). Milk-derived exosomes for oral

866

delivery of paclitaxel. Nanomedicine Nanotechnol. Biol. Med.

, 1627

–

1636.

867

https://doi.org/10.1016/j.nano.2017.03.001.

868

42. Qu, S., Han, Y., Liu, Y., Zhu, J., Acaroz, U., Shen, J., and Zhu, K. (2022). Milk Exosomes

869

Facilitate Oral Delivery of Drugs against Intestinal Bacterial Infections. J. Agric. Food Chem.

870

, 16069

–

16079. https://doi.org/10.1021/acs.jafc.2c04971.

871

Journal Pre-proof

43. Zhang, H.-Y., Firempong, C.K., Wang, Y.-W., Xu, W.-Q., Wang, M.-M., Cao, X., Zhu, Y.,

872

Tong, S.-S., Yu, J.-N., and Xu, X.-M. (2016). Ergosterol-loaded poly(lactide-co-glycolide)

873

nanoparticles with enhanced in vitro antitumor activity and oral bioavailability. Acta

874

Pharmacol. Sin.

, 834

–

844. https://doi.org/10.1038/aps.2016.37.

875

44. Turner, R., Joseph, A., Titchener-Hooker, N., and Bender, J. (2018). Manufacturing of

876

Proteins and Antibodies: Chapter Downstream Processing Technologies. Adv. Biochem. Eng.

877

Biotechnol.

165

, 95

–

114. https://doi.org/10.1007/10_2016_54.

878

45. Yamauchi, M., Shimizu, K., Rahman, M., Ishikawa, H., Takase, H., Ugawa, S., Okada, A.,

879

and Inoshima, Y. (2019). Efficient method for isolation of exosomes from raw bovine milk.

880

Drug Dev. Ind. Pharm.

, 359

–

364. https://doi.org/10.1080/03639045.2018.1539743.

881

46. Ritsema, J.A.S., Herschberg, E.M.A., Borgos, S.E., Løvmo, C., Schmid, R., Te Welscher,

882

Y.M., Storm, G., and van Nostrum, C.F. (2018). Relationship between polarities of antibiotic

883

and polymer matrix on nanoparticle formulations based on aliphatic polyesters. Int. J. Pharm.

884

548

, 730

–

739. https://doi.org/10.1016/j.ijpharm.2017.11.017.

885

47. Patil, S.M., Diorio, A.M., Kommarajula, P., and Kunda, N.K. (2024). A quality-by-design

886

strategic approach for the development of bedaquiline-pretomanid nanoparticles as inhalable

887

dry

powders

for

treatment.

Int.

Pharm.

653

123920.

888

https://doi.org/10.1016/j.ijpharm.2024.123920.

889

48. Rashidi, M., Bijari, S., Khazaei, A.H., Shojaei-Ghahrizjani, F., and Rezakhani, L. (2022). The

890

role of milk-derived exosomes in the treatment of diseases. Front. Genet.

, 1009338.

891

https://doi.org/10.3389/fgene.2022.1009338.

892

893

894

Journal Pre-proof

Figure 1. TEM images and membrane protein concentration of ME, blank ME-PLGA NPs,

895

and ME-PLGA-PBTZ169/BDQ NPs.

(A) TEM images of natural MEs and magnified views.

896

Scale bars: 200 nm & 100 nm. (B) TEM images of ME-PLGA-PBTZ169 NPs and (D) ME-PLGA-

897

BDQ NPs with magnified views. Scale bars: 1 μm & 200 nm.

(C) WB results of characteristic ME

898

proteins and membrane protein retention rates during the preparation process (Mean ± SD, n = 3),

899

and statistical analysis was performed by a two-

tailed Student’s t

-test. No significant (NS)

900

differences were observed in the three specific proteins before and after encapsulation.

901

Figure 2 Physical stability assessment of ME-PLGA nanoparticles after storage at -80°C for

902

one month.

(A) Changes in the average particle size of nanoparticles; (B) Changes in zeta potential;

903

(C) Changes in polydispersity index. Data are presented as Mean ± SD (n=3), and statistical

904

analysis was performed by a two-

tailed Student’s t

-test. No statistically significant differences

905

were observed between groups (

> 0.05), indicating that the nanoparticles remained stable after

906

lyophilization and reconstitution under long-term low-temperature storage conditions.

907

Figure 3

. Contact angle measurements of raw drug PBTZ169, Blank PLGA NPs, and PLGA-

908

PBTZ169 NPs

(Mean ± SD, n = 3), and statistical analysis was performed by one-way analysis of

909

variance (ANOVA) tests. Compared with the PBTZ169 group, **

<0.01; Compared with the

910

BDQ group,

△△

<0.01.

911

Figure 4. Cytotoxicity evaluation of BDQ/PBTZ169, PLGA-BDQ/PBTZ169 nanoparticles,

912

ME-PLGA-BDQ/PBTZ169 nanoparticles, and two blank carrier materials in HepG2, Vero,

913

Caco-2, HeLa, and J774A.1 cells.

The figure shows the cell viability of two blank carrier

914

materials (10 mg/mL) and three formulations (PBTZ169 and BDQ) at various drug concentration

915

gradients (8, 16, 32, 64, 128 μg/mL), comparing the in vitro safety of different formulations at the

916

same drug concentration. (Mean ± SD, n = 3, and statistical analysis was performed by one-way

917

Journal Pre-proof

analysis of variance (ANOVA) tests) Compared with the PBTZ169 group, *

< 0.05; compared

918

with the PLGA-BDQ nanoparticles group, #P < 0.05. Statistical analysis was performed for

919

different drug formulations at the same concentration, and no significant differences were observed

920

except for the groups indicated in the figure.

921

Figure 5. Drug release profiles of three formulations (BDQ/PBTZ169, PLGA-

922

BDQ/PBTZ169 NPs, and ME-PLGA-BDQ/PBTZ169 NPs) under simulated gastrointestinal

923

conditions.

Simulated Gastric Fluid (SGF, pH=1.2) and Simulated Intestinal Fluid (SIF, pH=6.8)

924

(

A) and (B) represent the drug release profiles of three formulations of PBTZ169 and three

925

formulations of BDQ in gastric and intestinal fluids over 36 hours (sampling at 0, 2, 4, 6, 8, 10,

926

12, 24, and 36 hours). Mean ± SD, n = 3.

927

Figure 6. The plasma concentration-time profiles following single-dose administration of

928

three different formulations (BDQ/PBTZ169, PLGA-BDQ/PBTZ169 NPs, and ME-PLGA-

929

BDQ/PBTZ169 NPs) in mice.

Figures A, B, and C show the drug concentration-time curves of

930

BDQ in plasma (A), lung (B), and spleen (C), respectively; Figures D, E, and F show the drug

931

concentration-time curves of PBTZ169 in plasma (D), lung (E), and spleen (F), respectively.

932

(Mean ± SD, n = 3)

933

Figure 7. Comparative analysis of AUC values following single-dose administration of three

934

different formulations (BDQ/PBTZ169, PLGA-BDQ/PBTZ169 NPs, and ME-PLGA-

935

BDQ/PBTZ169 NPs) in mice

, Mean ± SD, n = 3, and statistical analysis was performed by one-

936

way analysis of variance (ANOVA) tests. Compared with the PBTZ169/BDQ group, *

< 0.05,

937

<0.01

938

Journal Pre-proof

Figure 8. Analysis of drug concentrations in vivo and bacterial loads in lungs/spleens of mice

939

after two weeks of administration of different nanoformulations.

(A) Drug concentrations in

940

lung tissues of three different PBTZ169 formulations at the peak time after the last dose; (B) Drug

941

concentrations in lung tissues of three different BDQ formulations after two weeks of

942

administration; (C) Drug concentrations in spleen tissues of three different PBTZ169 formulations

943

at the peak time after the last dose; (D) Drug concentrations in spleen tissues of three different

944

BDQ formulations after two weeks of administration; (E) Bacterial colony (CFU) counts in lung

945

tissues after two weeks of treatment with different formulations; (F) Bacterial colony (CFU) counts

946

in spleen tissues after two weeks of treatment with different formulations; (G) Schematic timeline

947

of infection, treatment, and organ collection in mice (experimental details are provided in Section

948

3.7). Data are presented as Mean ± SD (n=6), and statistical analysis was performed by one-way

949

analysis of variance (ANOVA) tests. *

< 0.05, **

< 0.01, ***

< 0.001, ****

< 0.0001.

950

is defined as no statistical significance.

951

Figure 9.

Cardiac function evaluation of zebrafish (2 dpf) exposed to free BDQ (72 μM), ME

952

PLGA-

BDQ NPs (72 μM), and PLGA

-BDQ NPs

(72 μM)

for 2 days.

The figure shows the

953

heart rate (A), fractional shortening (B), stroke volume (C), and cardiac output (D) of zebrafish

954

hearts. Data are presented as Mean ± S.E.M (n = 10), and statistical analysis was performed by

955

one-way analysis of variance (ANOVA) tests. All experiments were independently repeated three

956

times. * denotes comparison with the control group; # denotes comparison with the free BDQ

957

group; */#

<0.05, **/##

<0.01, ***/###

<0.001, ****/####

<0.0001.

NS is defined as no

958

statistical significance.

959

960

Journal Pre-proof

Table 1. Particle size, polydispersity index (PDI), and Zeta potential of ME, blank ME-PLGA

961

NPs, and ME-PLGA-PBTZ169/BDQ NPs (Mean ± SD, n = 3).

962

Sample name

Particle size

（

）

Polydispersity

index (PDI)

Zeta

potential

（

）

104.50±4.86

0.16±0.02

-9.95±0.12

PLGA-BDQ NPs

232.13±7.24

0.15±0.01

-20.44±0.39

PLGA-PBTZ169 NPs

216.87±7.76

0.14±0.01

-20.28±0.45

ME-PLGA-BDQ NPs

252.33±2.91

0.15±0.02

-18.26±0.47

ME-PLGA-PBTZ169 NPs

252.50±7.56

0.15±0.02

-18.65±0.83

963

Table 2. Encapsulation efficiency (EE) and drug loading (DL) of ME-PLGA-PBTZ169/BDQ

964

NPs (Mean ± SD, n = 3).

965

Sample name

DL(%)

EE(%)

ME-PLGA- PBTZ169 NPs

15.90±1.32%

63.30±5.02%

ME-PLGA- BDQ NPs

18.13±2.92%

68.52±7.13%

966

Journal Pre-proof

Table 3. Minimum inhibitory concentrations (MICs) of BDQ/PBTZ169, PLGA-

967

BDQ/PBTZ169 NPs, ME-PLGA-BDQ/PBTZ169 NPs, and two blank carrier materials

968

against Mycobacterium tuberculosis reference strain H37Rv in vitro

969

Sample name

MIC

（

μg/mL

）

PLGA-Blank NPs

>10

ME-PLGA-Blank NPs

>10

PBTZ169

0.000625

PLGA-PBTZ169 NPs

0.000625

ME-PLGA-PBTZ169 NPs

0.000625

BDQ

0.0625

PLGA-BDQ NPs

0.0625

ME-PLGA-BDQ NPs

0.0625

970

971

Journal Pre-proof

Key resources table

REAGENT or

RESOURCE

SOURCE

IDENTIFIER

Antibiotics

CD63

Absin

abs123229

CD81

Absin

abs126599

TSG101

bioss

bs-1365R

goat anti-rabbit IgG-HRP

Affinity

S0001

goat anti-mouse IgG-

HRP

Affinity

S0002

Bacterial and virus strains

M. tuberculosis

H37Rv

Lab stock

ATCC 27294

M. tuberculosis

H37Ra

Lab stock

ATCC 25177

Cell lines

HepG2 cells (Male)

Lab stock originally

from ATCC

ATCC HB-8065

Vero cells (Female)

Lab stock originally

from ATCC

ATCC CCL-81

Caco-2 cells (Male)

Lab stock originally

from ATCC

ATCC HTB-37

HeLa cells (Female)

Lab stock originally

from ATCC

ATCC CCL-2

Journal Pre-proof

J774A.1 cells (Female)

Lab stock originally

from ATCC

ATCC TIB-67

Experimental models: Organisms/strains

BALB/c Mice (Female)

Beijing Vital River

Laboratory Animal

Technology Co., Ltd.

RRID: MGI:2161072

Zebrafish (larvae, sex not

determined)

China Zebrafish Resource

Center(CZRC)

CZ56

Chemicals, peptides, and recombinant proteins

bovine skim milk

Mengniu Dairy Co.,

Ltd. (Hohhot, China)

Provided commercially for this study

poly(D,l-lactide-

glycolide)(50:50)-b-

poly(ethylene glycol)

(PLGA)

Aladdin Scientific

(Shanghai, China)

34346-01-5

PBTZ169

Shanghai Hanxiang

Biotechnology Co., Ltd.

20210924

BDQ

Shanghai Hanxiang

Biotechnology Co., Ltd.

20210318

Software and algorithms

WinNonlin® (v6.2.1)

Certara, USA

Version 6.2.1

Journal Pre-proof