Aerosolized ApoA1 Nanoparticles Synthesized by Microfluidics Cross the Lung Barrier and Modulate Inflammation

(250 words) 18 19 High-density lipoproteins exert vasculoprotective effects, mainly through apolipoprotein A1, 20 which has led to the development of treatments based on apolipoprotein A1 nanoparticles 21 (A1NPs) administered intravenously, mainly for the treatment of cardiovascular diseases. 22 However, their potential as therapy for lung pathologies has not yet been explored. In this work, 23 we produced A1NPs using microfluidics and characterized their therapeutic potential for lung 24 delivery. Their morphology was characterized by dynamic light scattering and transmission 25 electron microscopy. A1NPs toxicity and cellular uptake were performed on both endothelial 26 (HMEC-1) and epithelial (A549) cells and their anti-inflammatory activity was evaluated on 27 TNF-α-stimulated HMEC -1. A1NPs biodistribution was explored in lung mice after 28 aerosolization and their transcytosis was further investigated using A549 air-liquid interface 29 model. Our results demonstrate that the microfluidic synthesis of A1NPs was reproducible and 30 yielded discoidal particles with sizes ranging from 7-12 nm. A1NPs were internalized by both 31 cells without being cytotoxic and significantly reduced IL-6 expression. Aerosolization resulted 32 in homogeneous distribution in lungs, without causing an immunogenic response. A fraction of 33 A1NPs crossed alveolar epithelial cells both in vitro and in vivo, paving the way for future 34 therapeutic strategies targeting not only the lungs, but also other peripheral organs. These 35

are promising for the use of A1NPs as vectors for therapeutic molecules, which could 36 exert synergistic protective effects with Apolipoprotein A1. This is the first study to show the 37 non-invasive administration of A1NP s by aerosolization, which may improve their 38 bioavailability in lungs and appears to be a promising approach for treating lung diseases. 39 40

microfluidic, Apolipoprotein A1, lipid nanoparticles, lung delivery , 41 aerosolization. 42 43 44 45 46 47 48 49 50 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 3 GRAPHICAL ABSTRACT 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 4

77 High-density lipoproteins (HDL s) are complex macromolecules composed of a hydrophobic 78 core with esterified cholesterol and triglycerides, surrounded by phospholipids and proteins. 79 They can adopt a spherical or discoidal conformation depending on their lipid composition 1. 80 Their high density (1.063 to 1.210 g/mL) is mainly due to their high protein content and their 81 mean size is around 5 to 17 nm according to lipids and protein enrichment 2. Indeed, HDL 82 particles are composed of several apolipoproteins, among them apolipoprotein A1 (ApoA1) 83 being the major protein 1. HDLs also contain enzyme s such as Lecithin-Cholesterol Acyl 84 Transferase (LCAT) which play a central role in reverse cholesterol transport 3, giving them 85 anti-atherogenic properties4. In addition, HDLs exert anti-inflammatory and antioxidant effects 86 mainly through ApoA1 but also anti-endotoxic effects, linked to their capacity to bind and 87 eliminate lipopolysaccharides, contributing to their overall vasculoprotective action5–8. 88 89 HDL-cholesterol levels are commonly used as a biomarker of cardiovascular health; a marked 90 decrease being associated with a significant increase in the risk of cardiovascular disease, as 91 initially demonstrated by the Framingham study in 1980 9. On the strength of their beneficial 92 effects, several pharmaceutical companies have developed HDL-mimetic nanoparticles, known 93 as reconstituted HDL (rHDL) or ApoA1 nanoparticles (A1NPs), since they are composed of 94 ApoA1 and phospholipids. To reduce the cholesterol burden in atherosclerotic plaques , CSL 95 Behring has developed CSL -111 and CSL-112 formulations, composed of phospholipids and 96 ApoA1 isolated from human plasma. These nanoparticles have been evaluated in clinical trials 97 such as ERASE and AEGIS -II, studying the impact of A1NPs on reducing atheroma plaque 98 after intravenous administration 10,11. Clinical studies testing the effects of reversing 99 atherosclerosis and limiting the recurrence of cardiovascular events have been disappointing10. 100 However, A1NPs are currently being evaluated for their anti-inflammatory and 101 endothelioprotective effects in the context of sepsis. We demonstrated that intravenous 102 injection of CSL-111 reduced systemic inflammation, notably through their ability to promote 103 lipopolysaccharide (LPS) clearance in the context of bacterial infection in mice 12. Our 104 laboratory also confirmed that CER -001 (A1NPs developed by Abionyx, formerly Cerenis ) 105 injections in a severe COVID-19 patient decreased circulating inflammatory markers 8. These 106 anti-inflammatory properties are mainly mediated by ApoA1. Indeed, ApoA1 interacts with the 107 ABCA1 transporter, involved in cholesterol efflux , and leads to the activation of intracellular 108 molecular pathways 14 leading for example to inhibit the NF -κB signaling pathway in 109 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 5 endothelial cells 15, thereby reducing the production of pro -inflammatory cytokines and 110 chemokines such as IL-6 and MCP-1. 111 112 Although A1NPs have been widely studied in the cardiovascular field, no transposition to 113 pulmonary pathologies has yet been established. To target the lung, aerosolization of A1NPs 114 seems far more appropriate than intravenous injection. Apart from inhalation of lipid 115 nanoparticles16,17, no one has yet evaluated aerosolization of A1NPs. Various methods for 116 producing these nanoparticles have been developed. The conventional method relies on the use 117 of sodium cholate 18,19, but its toxicity has led to the development of alternative approaches, 118 notably via microfluidics 20. This innovative technology enables continuous, rapid production 119 without toxic components, as well as the vectorization of bioactive molecules. 120 121 In this study, we produced A1NPs using microfluidics and confirmed their discoidal 122 morphology in a reproducible manner . These nanoparticles were no-cytotoxic to endothelial 123 (HMEC-1) and lung epithelial (A549) cells, and were efficiently internalized . Notably, this 124 uptake was enhanced in the fraction of ABCA1 pos itive epithelial cells . Their anti -125 inflammatory property has been confirmed by a significant reduction of IL-6 expression 126 induced by TNF-α stimulation in endothelial cells. In vivo , aerosolization of A1NPs led to 127 homogeneous biodistribution throughout the lung, characterized by uptake by type I and II 128 pneumocytes and alveolar endothelial cells. A progressive passage into the systemic circulation 129 from 6 hours post -administration was observed, allowing their biodistribution to peripheral 130 organs, opening up ad ditional therapeutic strategies targeting other pathologies beyond those 131 affecting the lung . Taken together, this study advances our knowledge of the therapeutic 132 potential of A1NPs by opening new perspectives on treatments for respiratory diseases and 133 beyond. The enrichment of A1NPs with therapeutic molecules, particularly hydrophobic, 134 provides insight into new care avenues. 135 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 6

AND DISCUSSION 136 137 While therapies based on apolipoprotein A1 nanoparticles (A1NPs) have failed to achieve the 138 expected effects in cardiovascular diseases, their therapeutic potential in lung diseases has not 139 been explored. Here, we have combined the fields of physics and biology to shown that anti-140 inflammatory A1NPs can be homogeneously aerosolized in the lungs, paving the way for new 141 therapeutic strategies for lung diseases. 142 143 Production and physical characterization of A1NPs 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 Figure 1. Physical characterization of Apolipoprotein A1 nanoparticles. (A) A1NPs were 161 produced by microfluidic using a chip with 2 inlets for Apolipoprotein A1 (ApoA1) 162 injection at 0.8 mL/min and 1 inlet for phospholipids (POPC) injection at 0.1 mL/min. (B) 163 Three independent productions of A1NPs were characterized by dynamic light scattering 164 to determine their size (7-12 nm ). (C) E lectron transmission microscopy was used to 165 determine the shape of A1NPs. White triangles indicate stacked disc-like structures, also 166 known as “rolls”. 167 168 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 7 A1NPs were reconstituted using a single-step, self-assembly method in a single layer, 3 -inlet 169 microfluidic device (Figure 1A). As previously shown by Kim et al., this technique allows the 170 production of reproducible and homogeneous batches of A1NPs20. While Kim et al. performed 171 their nanoparticle synthesis in phosphate buffer saline, we opted for reconstitution directly in a 172 buffer designed to preserve A1NPs by limiting their oxidation (TEN buffer). It is also worth 173 noting that we use d a five-fold high er concentration of ApoA 1, enabling us to obtain 174 nanoparticles of the expected majority size without the need for additional purification steps. 175 Dynamic light scattering (DLS) analysis confirmed the reproducibility of these productions . 176 These results indicate that the average size of A1NPs was around 10 nm , similar to that of 177 plasma HDL s1 (Figure 1B). The s tructural organization of A1NPs was observed under a 178 transmission electron microscope (Figure 1C). This observation revealed that A1NPs formed 179 stacked disc-like structures (discoidal shape), also known as “rolls”, similar to those observed 180 for microfluidically synthetized Apo A1 nanoparticles of Kim et al.20 and already described for 181 plasmatic pre-HDL21. The overall morphology of A1NPs was comparable to that of pre-HDL 182 demonstrating the ability of our laboratory to generate biological nanoparticles , as previously 183 done by pharmaceutical groups. CSL-111 and CSL-112 are nanoparticles made from human 184 ApoA1 and soy-derived phospholipids and have been evaluated in clinical trials22, making them 185 a benchmark in terms of morphological and size characteristics. Yet, in the context of coronary 186 artery disease, these nanoparticles did not produce the expected atheromatous plaque reduction 187 effect, underlining the need for optimizations to improve their therapeutic efficacy 22. One 188 proposed solution is to enrich these nanoparticles with bioactive molecules. For instance, 189 Moreno et al. showed that high-density lipoproteins (HDLs) enriched with alpha-1-antitrypsin, 190 significantly reduced neutrophil elastase-induced pulmonary emphysema in mice, compared 191 with native, plasma isolated HDLs23. Thus, the use of A1NPs as vectors for therapeutic 192 molecules represents a promising approach requiring further investigation to optimize their 193 therapeutic potential. Microfluidic offers significant advantages for this type of enrichment, 194 whether of proteins, synthetic molecules or lipids 24. Controlled flow rates and channel 195 dimensions in the micrometer range promote molecule assembly, facilitating the incorporation 196 of therapeutic compounds into nanoparticles, without using additional chemicals. 197 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 8 A1NPs are internalized by cells without cytotoxicity 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 9 Figure 2. Apolipoprotein A1 nanoparticle cytotoxicity and uptake assay. (A,B) MTT assay 232 on endothelial (HMEC-1, A) and alveolar epithelial cell lines (A549, B), 24 hours after 233 A1NP incubation. (n=3-5 independent experiments; ANOVA; Tukey’s multiple 234 comparison test; * p<0.05, ** p<0.01, **** p<0, 0001). (C) A1NPs [0.05mg/ml] are taken 235 up by both cell lines after 6 and 24 hours of incubation. A rabbit anti human ApoA1 236 antibody was used for labelling A1NPs (green) and cell nuclei were stained with DAPI 237 (blue). (D,E) Uptake of A1NPs by ABCA1-positive A549 cells. (D) A dot plot illustration 238 for double staining with DilC18-A1NPs and ABCA1 and (E) the m ean fluorescence 239 intensity of DilC 18 in ABCA1 negative and ABCA1 positive cells. (n=3 independent 240 experiments; Paired t-test; * p < 0.05). 241 242 To assess the safety profile of A1NPs, cell viability tests were conducted across a range of 243 A1NP concentrations. HMEC-1 and A549 cells were incubated with A1NPs at concentrations 244 from 0.015 mg/mL to 1 mg/mL for 24 hours, followed by an MTT assay (Figure 2A ,B). The 245

indicate no significant difference between stimulated and unstimulated cells, suggesting 246 A1NPs have no impact on the viability of either cell type. To visualize A1NPs internalization 247 into cells, confocal microscopy was performed on both HMEC -1 and A549 cells using anti-248 ApoA1 antibodies. Observations show that after 6 hours, cells have internalized the A1NPs, 249 with an increase in labeling intensity observed after 24 hours (Figure 2C). In addition to MTT 250 assay, these findings indicate that the nanoparticles do not alter cell morphology and integrity. 251 It should be emphasized that nanoparticles not only bind to the cell surface, but are also 252 internalized. Indeed, Silver et al. demonstrated that hepatocytes incubated with HDL s at 4°C 253 only led to binding. At 37°C, results indicate an active uptake process25,26. Given that the uptake 254 of HDLs is mediated by the scavenger receptor class B type I (SR-BI) and the ATP binding 255 cassette subfamily A member 1 (ABCA1)27, one may argue that A1NPs internalization is also 256 dependent on these receptors. Alveolar epithelial cells (type I and type II) as well as A549 cells 257 do express ABCA1 28,29 but whether A1NP -mediated uptake is similar to that of endothelial 258 cells27,30 is unclear. To clarify the potential role of ABCA1 in A1NP uptake by A549 cells, we 259 incubated cells with fluorescent A1NPs (DilC18 staining) for 6 hours and analyzed cells by flow 260 cytometry. We observed that about 95% of A549 cells were positive for DilC 18-A1NP. 261 However, only approximately 27% of A549 cells were positive for ABCA1, but this population 262 captured a higher amount of DilC18-A1NP, as evidenced by a significant increase in the mean 263 fluorescence intensity of DilC 18 compared to ABCA1 -negative A549 cells (Figure 2D,E and 264 S1). This result confirms that a fraction of A549 cells do express ABCA1 in unstimulated 265 condition and that ABCA1 is, at least in part, involved in A1NP uptake. Other receptors, also 266 expressed by A549 cells, such as SRB -131, could also be involved in A1NPs uptake, thereby 267 explaining that 95% of cells are able to internalize these nanoparticles . Interestingly, the dual 268 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 10 role of ABCA1 in repressing inflammation while maintaining cholesterol homeostasis 269 represents a promising therapeutic target for inflammatory lung diseases in the future29. 270 271 A1NPs display anti-inflammatory properties 272 Endothelial cells were stimulated with 2.5 ng/mL of TNF-α to induce an inflammatory response 273 and co-stimulated with A1NPs. Following 6 hours of incubation, we assessed the gene 274 expression of the pro -inflammatory mediator IL -6 (Figure 3A). After 16 hours, IL -6 protein 275 levels in the culture medium were quantified by ELISA (Figure 3B). Both IL -6 mRNA an d 276 protein levels increased about 3-fold under inflammatory conditions compared to untreated 277 controls. However, treatment with A1NPs significantly reduced IL -6 expression at both the 278 transcriptional and protein levels, indicating an anti -inflammatory effect . These findings 279 demonstrate that A1NPs are biologically functional and exhibit anti-inflammatory activity in 280 HMEC-1 cells under TNF-α-induced inflammatory conditions. This assay on endothelial cells 281 is a classical hallmark to appreciate the anti-inflammatory properties of HDLs and mimetics32. 282 It has been shown that ApoA1 binding to ABCA1 may trigger the expression of tristetraprolin, 283 which subsequently promotes the degradation of inflammatory cytokine mRNA in response to 284 LPS, including IL-6, through its 3’-UTR AREs33. 285 286 287 288 289 290 291 292 293 294 295 296 Figure 3. Anti-inflammatory properties of A1NPs. A1NPs [0.05 mg/ml] significantly 297 decrease TNF-α-induced IL-6 at mRNA (A) and protein levels (B) in HMEC-1 cells. (n=3-298 5 independent experiments; ANOVA; Tukey’s multiple comparison test; * p<0.05, ** 299 p<0.01, **** p<0,0001). 300 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 11 A1NPs are homogeneously distributed in the lung after aerosolization 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 Figure 4. Biodistribution of A1NPs after aerosolization. ( A) Experimental design for 321 studying the biodistribution of A1NPs administered intratracheally by aerosolization. (B) 322 Determination of human ApoA1 concentration in mouse plasma at different time points 323 after aerosolization: 0h, 3h, 6h, 12h and 24h (n=6 for PBS, n=10 for A1NPs). (C) A1NPs 324 labeled with DilC18 dye (red) reach the left and right lungs 6 hours after aerosolization 325 and persist in both lungs for up to 24 hours. Representative illustration of 6 mice treated 326 with PBS and 10 mice with A1NPs-DilC18. 327 328 To investigate the biodistribution of A1NPs following their administration by aerosolization, 329 A1NPs were first fluorescently labelled with DilC 18 (Figure 4A). Since the lung is closely 330 linked to the capillary network , we also quantified the passage of A1NPs in the bloodstream. 331 A1NPs reached a peak in plasma 6 hours after administration, before decreasing 12 hours later, 332 but their detection persisted after 24 hours (Figure 4B). At 6 h ours post-administration, 333 homogeneous red fluorescence was detected throughout the lung parenchyma of mice that had 334 received DilC18-A1NPs while no fluorescence was observed in control mice aerosolized with 335 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 12 PBS (Figure 4C). This uniform biodistribution of A1NPs persisted at 24 hours , which is also 336 consistent with A1NP plasma kinetics (Figure 4B). Interestingly, red fluorescence was also 337 detected in the liver and kidneys at 24 hours (Figure S 2). These observations suggest that 338 A1NPs-DilC18 behave similarly to HDL particles, with elimination via hepatic and renal 339 pathways. No significant variation in the body weight of the mice was observed throughout the 340 experiment, supporting the absence of in vivo nanoparticle toxicity (Figure S3A). Moreover, 341 additional experiments have demonstrated the absence of immunogenicity in mice given 342 A1NPs on days 0, 1 and 12 (Figure S3B). Surprisingly, other organs, such as the brain and the 343 spleen, were also enriched in A1NP s after their passage into the bloodstream (Figure S 4), 344 opening up therapeutic prospects target ing these organs. Although the lung remains the main 345 organ targeted by aerosoliz ation, this non -invasive route could also be considered for the 346 treatment of pathologies characterized by chronic inflammation in peripheral organs. 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 Figure 5. Cellular localization of A1NPs after aerosolization. Immunofluorescence of lung 365 sections from mice 6 hours after A1NP administration. ApoA1 appears in green and cell 366 nuclei are stained with DAPI (blue). (A) AGER (red), specific to type I pneumocytes. (B) 367 SFTPC (orange), marker for type II pneumocytes. (C) EMCN (red), characteristic of 368 pulmonary endothelial cells. Representative illustration of 10 mice with A1NPs-DilC18. 369 370 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 13 To further evaluate the precise localization of A1NPs , we performed co -labeling between 371 ApoA1 and different cell types specific to lung tissue. We observed co -localization of ApoA1 372 with cell-specific markers (AGER: type I pneumocytes; SFPTC: type II pneumocytes; EMCN: 373 vascular endothelial cells) (Figures 5 and S5). These different structural lung cell types were 374 able to internalize A1NPs, opening up interesting therapeutic perspectives. These include 375 intracellular application of A1NPs potentially enriched with therapeutic molecules that act 376 intracellularly, such as siRNAs. This A1NP's broad spectrum of pulmonary penetration makes 377 it the vector of choice for lung diseases. 378 379 A1NPs pass through an epithelium grown at an air-liquid interface. 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 14 Figure 6. Passage of A1NPs through A549 epithelial cells grown in air-liquid interface. 405 (A) Schema showing the experimental setup used to study A1NPs migration through lung 406 epithelial cells cultured in air/liquid interface (ALI). (B) Quantification of A1NPs 407 transcytosis under ALI conditions, measured by ELISA specific for human ApoA1, in the 408 basolateral medium. The cumulative concentration of A1NPs in the basolateral medium 409 is determined at different incubation times: 30 min, 1h, 2h, 4h and 6h. (C) 410 Immunofluorescence analysis of ALI membrane after 6h of transcytosis with 0.5 mg/mL 411 A1NPs. Phalloidin (green) labels the actin cytoskeleton, cell nuclei are stained blue, and 412 ApoA1 appear in red. (D) Three -dimensional visualization of the ALI membrane under 413 different experimental conditions. 414 415 Previous results in vivo suggest a progressive transfer of A1NPs into the bloodstream. The 416 nanoparticles may cross the alveolar barrier in the lung, being initially internalized by alveolar 417 epithelial cells before reaching endothelial cells, allowing access to the vascular compartment. 418 To further explore the passage of A1NPs through an epithelium, we set up an air-liquid interface 419 (ALI) model of lung epithelial cells using inserts (Figure 6A). According to previous 420 characterization of ALI culture of A549 alveolar epithelial cells , this model reconstitutes 421 epithelial layers with the expression of markers of both alveolar epithelial type I and type II 422 cells34. After apical addition of A1NPs, the inserts were incubated for 6 hours. A progressive 423 passage of A1NPs was observed over time in these experimental conditions (Figure 6B). In 424 another set of experiments, we made sure that after the assay, the permeability of both A1NPs 425 and without A1NPs epithelium was the same, ruling out the possibility that stimulation may 426 alter this parameter (Figure S6). Confocal microscopy confirmed the presence of ApoA1 in the 427 cytoplasm of epithelial cells (Figure 6C,D and S7). A deeper understanding of the mechanisms 428 involved in the transepithelial passage of A1NPs would be relevant, to determine whether this 429 is a process of transcytosis or other alternative mechanisms. ABCA1 is known to facilitate the 430 interaction and internalizati on of pre -β HDL particles. This mechanism has been extensively 431 characterized in endothelial cells 30. Given that A1NPs exhibit structural and functional 432 similarities to pre-β HDL, one may argue that their cellular internalization is also mediated by 433 ABCA1. Our flow cytometry results on ABCA1 expression showing an increase uptake in 434 ABCA1 positive cells support a model in which ABCA1 partially mediates A1NP 435 internalization, although additional receptors may contribute to uptake through other specific 436 mechanisms. Consistent with existing literature, discoidal ApoA1 particles have been shown to 437 preferentially interact with ABCA1 to facilitate lipid acquisition 30. In contrast, the SR -B1 438 receptor is known to recognize lipid-rich spherical HDL particles. The ABCG1 transporter also 439 plays a role in lipid efflux and may be implicated in HDL trafficking 15. It is interesting to note 440 that Moreno et al. observed that i ntravenous injection of HDL in mice with elastase -induced 441 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 15 emphysema led to increased HDL recruitment in the lungs compared to control mice 23. This 442 study suggests that A1NP may be preferentially recruited to inflamed tissues. Collectively, our 443 data suggest tha t the microfluidically produce d A1NPs mimic the biological behavior of 444 circulating discoidal HDL particles and may use similar uptake pathways. 445 446

447 In this study, we report the successful development of apolipoprotein A1 -based nanoparticles 448 (A1NPs) produced by microfluidics and designed for pulmonary delivery. Our results 449 demonstrate that A1NPs possess favorable physicochemical characteristics, includ ing a 450 reproducible discoidal morphology and nanoscale dimensions (7 –12 nm), comparable to 451 clinically evaluated HDL mimetics. In vitro , A1NPs were efficiently internalized by both 452 endothelial and alveolar epithelial cells, with uptake enhanced in ABCA1-positive populations, 453 and displayed no cytotoxicity across a range of concentrations. Importantly, A1NPs retained 454 biological functionality, as evidenced by their significant anti-inflammatory effects on TNF-α-455 stimulated endothelial cells, with a reduction in I L-6 expression at both transcript and protein 456 levels. 457 Upon aerosolization in mice, A1NPs exhibited homogeneous pulmonary biodistribution, 458 reaching both lobes and persisting in lung tissue for up to 24 hours without triggering detectable 459 immunogenicity. Fur thermore, we confirmed their transcytosis across alveolar epithelial 460 barriers both in vitro and in vivo, with a progressive passage through the epithelium. This dual 461 capacity for local action and systemic translocation highlights the versatility of A1NPs not only 462 as lung -targeted agents but also as systemic drug delivery vehicles following non -invasive 463 administration. 464 Together, these findings position A1NPs as a promising nanoplatform for the treatment of 465 respiratory diseases, with the potential to be enric hed with bioactive molecules for synergistic 466 therapeutic effects. Future investigations will focus on optimizing A1NPs for targeted delivery 467 of anti-inflammatory or antioxidant agents, and evaluating their efficacy in relevant models of 468 acute and chronic lung inflammation. This work opens new perspectives for the translation of 469 HDL-mimetic nanoparticles into pulmonary medicine. 470 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 16

& METHODS 471 472 A1NPs production and characterization . A1NPs are synthesized from plasma 473 apolipoprotein A1 (ApoA1) (plasma was obtained from the French blood national agency, EFS-474 LR agreement number #2018001378) and commercial phospholipids (2-oleoyl-1-palmitoyl-sn-475 glycero-3-phosphocholine; 42773-500MG, Sigma). The production is carried out on a 476 microfluidic chip with three input channels. The central inlet is for phospholipids prepared at a 477 concentration of 6 mg/mL in ethanol at an injection flow rate of 0.1 mL/min (Darwin 478 Microfluidics, SeryngeONE, Connection Kit 01). The other two inlets are used to inject ApoA1 479 at a concentration of 1 mg/mL in TEN buffer (10 mM TRIS, 1mM EDTA, 150 mM NaCl) at 480 an injection rate of 0.8 mL/min. The solution obtained at the microfluidic chip outlet is 481 centrifuged at 12,000 g for 15 minutes to sediment the aggregates. A1NPs are then concentrated 482 and washed with a TEN buffer using a 10 kDa cut-off concentrator (Corning). 483 Dynamic light scattering characterization . A1NPs were characterized by dynamic 484 light scattering (DLS) spectroscopy as described previously35. 485 Electron microscopy characterization. Transmission electron microscopy analyses 486 were conducted at the Center for Quantitative Imaging Lyon East (CIQLE, University of Lyon 487 1, Lyon, France). The morphological characteristics and size of the nanoparticles were assessed 488 using negatively stained samples, imaged with a Gatan Orius 600 CCD camera (Gatan, USA) 489 on a LaB6 JEOL JEM -1400 transmission electron microscope (JEOL, Japan) operating at an 490 accelerating voltage of 120 kV. For sample preparation, 300-mesh copper grids coated with a 491 carbon film (Delta Microscopies, France) were glow-discharged for 30 seconds (Balzers SCD 492 040, Liechtenstein) to render the carbon surface hydrophilic, facilitating sample adhesion. 493 Nanoparticle suspensions were subsequently deposited onto the treated grids, followed by 494 negative staining using a 2% aqueous solution of uranyl acetate. After complete air drying, the 495 grids were mounted onto a single -tilt holder and introduced into a JEOL JEM -2100 496 transmission electron microscope (JEOL, Japan), equipped with a cryo pole piece and operated 497 at 120 kV. Image acquisition was performed using a Gatan SC600A CCD camera (Gatan, 498 USA). 499 Cell Culture and stimulation. HMEC-1 cells, a human microvascular endothelial cell 500 line (ATCC# CRL -3243), were cultured in MCDB 131 medium (P04 -80057, Pan Biotech) 501 supplemented with 10% fetal bovine serum ( FBS) (ST30-3302, Pan Biotech ), 100 units/mL 502 Penicillin- 100µg/mL Streptomycin (P06-07100, Pan Biotech ), 250µg/mL Amphotericin-B 503 (P06-01100, Pan Biotech), 10 ng/mL Epidermal Growth Factor (EGF) (E9644; Sigma-Aldrich, 504 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 17 USA), 1 µg/mL Hydrocortisone (Sigma), and 10 mM L-Glutamine (Pan Biotech). A549 cells, 505 a human alveolar epithelial cell line (ATCC# CCL-185), were routinely cultured in RPMI 1640 506 medium (P04-22100, Pan Biotech) supplemented with 10% heat-inactivated FBS (ST30-3302, 507 Pan Biotech) and 100 units/mL Penicillin- 100µg/mL Streptomycin (P06-07100, Pan 508 Biotech). Both cell lines were maintained in a humidified incubator at 37°C with 5% CO₂ and 509 subcultured when they reached 90% confluence. During stimulation with A1NPs, the respective 510 media were supplemented with 10% delipidated FBS for HMEC-1 and 5% delipidated FBS for 511 A549. 512 A1NP Cytotoxicity. The cytotoxicity of A1NPs on each cell line was evaluated using 513 the 3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay. 96-well plate 514 were seeded with 10,000 cells/well. After 24hours of incubation, the medium was removed and 515 the cells were stimulated with 100 µL of A1NPs at concentrations ranging from 1 mg/mL to 516 0.015 mg/mL for 24 hours. 10µL of MTT reagent (M2128-5G, Sigma-Aldrich) diluted in PBS 517 was added to each well for five hours, resulting in a f inal MTT concentration of 0.5 mg/mL. 518 The plate was then centrifuged at 500g for 4 minutes at 25°C. The medium was removed and 519 replaced with 200 µL of dimethyl sulfoxide (DMSO) (UD8050-B, Euromedex) to dissolve the 520 formazan crystals. Absorbance was measured at 560 nm using a CLARIOStar Plus plate reader 521 (BMG Labtech). 522 A1NP uptake by endothelial and epithelial cells. 8-well Labtek chamber slides were 523 seeded with 50,000 cells/well for each cell line studied. Once cell confluence reached 80%, the 524 wells were washed with PBS solution and were serum -deprived for 3 hours. Cells were 525 stimulated with A1NPs at a concentration of 0.05 mg/mL, or with their respective media for 6 526 hours and 24 hours. Cells were washed three times with PBS, then fixed with 4% 527 paraformaldehyde (PFA) for 15 minutes. Slides were washed once with PBS for 5 minutes, 528 then twice with PBS containing 0.05% Triton for 10 minutes. Cells were then blocked for 1 529 hour in PBS 0.05% triton 2% BSA. Primary anti -human ApoA1 antibody (Calbiochem) was 530 diluted 1:500 in PBS 0.05% triton 0.2% BSA and incubated with the cells overnight at 4°C. 531 After 3x 10-minute washes, the cells were incubated with DAPI (1μg/mL, Sigma) mixed with 532 an Alexa 488 goat anti -rabbit secondary antibody at 1:1000 dilution for 1 hour at room 533 temperature. After 5x-10-minute washes, the slides were mounted with fluorescent medium and 534 images were captured with a confocal microscope (Nikon Eclipse Ti2). 535 Flow Cytometry. A total of 200,000 A549 epithelial cells were seeded in 6-well plates. 536 After 24 hours, the culture medium was replaced with a lipid-depleted medium for an additional 537 24-hour period. Subsequently, cells were either stimulated or not with DilC 18 A1NPs at a 538 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 18 concentration of 0.05 mg/mL for 4 hours, under lipid -free conditions. Following stimulation, 539 cells were detached using Accutase (25-058-CI, Corning) and immediately placed on ice. Cell 540 suspension was centrifuged at 300g for 5 minutes. The supernatant containing Accutase was 541 discarded, and the resulting cell pellet was resuspended in FACS buffer (PBS supplemented 542 with 0.1% BSA). After a second centrifugation, the supernatant was removed, and Fc receptor 543 blocking reagent (anti-CD16/CD32; 553141, BD Bioscience) was added at a concentration of 544 1 µg per 10⁶ cells. The suspension was incubated for 10 minutes at 4°C. Cells were then washed 545 with FACS buffer and incubated with the human ABCA1 Alexa Fluor 488 conjugated antibody 546 (NB100-2068G, NovusBio) at 1 µg per 10⁶ cells, for 30 minutes in the dark at 4°C. Following 547 a final wash with PBS, the cells were resuspended and analyzed by flow cytometry (Cytoflex, 548 Beckman Coulter). 549 A1NP anti-inflammatory properties. The anti-inflammatory activity of A1NPs was 550 assessed in the TNF-α-stimulated HMEC-1 cell line by RT-qPCR and ELISA. 551 RT-qPCR. 50,000 cells were seeded in a 24 -well plate. When the cells reached 70% 552 confluence, they were washed and serum-deprived (0% FBS) for 3 hours. Cells were stimulated 553 with A1NPs at a concentration of 0.05 mg/mL for 6 hours, with or without addition of TNF -α 554 (2.5 ng/mL). After stimulation, cells were lysed and RNA was extracted with RNeasy Plus Mini 555 kit (74136, Qiagen) and quantif ied by Nanodrop ( BMG Labtech). Reverse transcription was 556 carried out using the NxGen M -MuLV reverse transcriptase (30222 -1, Lucigen) according to 557 the manufacturer’s standard protocol. Quantitative PCR was performed using the Blastaq Green 558 2x qPCR MasterMix (G892, Abm). The transcript ion levels of IL-6 were measured using the 559 following primers: forward 5’ -ACCCCCAGGAGAAGATTCCA-3’, reverse 5’ -560 GCCTCTTTGCTGCTTTCACA-3’. The data were normalized against GAPDH (forward 5’ -561 AGCCACATCGCTCAGACAC-3’, reverse 5’-GCCCAATACGACCAAATCC-3’) and RNA 562 polymerase II (RNApol2) (forward 5’ -CGAGAAGGTCTCATTGACACAG-3’, reverse 5’ -563 ACCACCTGGTTGATGGAGTTCC-3’). 564 ELISA. 70,000 cells/wells were seeded in a 12-well plate. When the cells reached 70% 565 confluence, they were washed and serum -deprived (0% FBS) for 3 hours. Next, cells were 566 stimulated with A1NPs at a concentration of 0.05 mg/mL for 16 hours, with or without the 567 addition of TNF -α (2.5 ng/m L). Supernatant was collected for IL -6 quantification using a 568 human IL-6 ELISA Ready-SET-Go assay (Thermofisher). Absorbance was measured using a 569 CLARIOstar plate reader (BMG Labtech) at 450-570 nm. 570 571 572 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 19 Aerosolization in mice. C57BL/6 mice (8-10 weeks; male female ratio 1:1 ) were fed 573 ad libitum with standard laboratory chow and water. All animal experiments were approved by 574 the local ethics committee and Ministry of Higher Education and Research (APAFIS #53302-575 2025011413372952 v4). To visualize A1NPs in the lung, the nanoparticles were first incubated 576 with 200 μg of DilC18 at 37°C overnight with agitation at 250 RPM. Free DilC18 was removed 577 by ultracentrifugation. The density of the A1NPs -DilC18 was adjusted to 1.23 g/mL, and the 578 particles were then recovered by a first layer consisting in KBr at a density of 1.21 g/mL and a 579 second layer at 1.063 g/mL. Following ultracentrifugation at 252,000g for 18 hours at 4°C, the 580 A1NPs-DilC18 were collected. ApoA1 concentration in A1NPs was d etermined by BCA 581 protein assay ( Sigma). A1NPs -DilC18 were prepared at 8mg/m L. For intratracheal 582 aerosolization of A1NP s, mice were anesthetized with isoflurane (2.5%), followed by an 583 intraperitoneal (i.p. ) injection of ketamine (90 mg/kg) and xylazine (4.5 mg/kg) to deepen 584 anesthesia. They were positioned on a panel inclined at 45° for intratracheal instillation, with a 585 light source to visualize the tracheal orifice. Using a microsprayer aerosolizer, model YAN 586 30012 (Yuyan Instruments) , a volume of 25 µL was administered between the vocal cords. 587 Mice received either PBS solution or A1NPs -DilC18 (8 mg/ml). Blood was collected before 588 and after aerosolization at 3, 6, 12 and 24 hours. 4 mice (1 PBS and 3 A1NPs-DilC18) were 589 sacrificed at 12 hours and 4 mice (1 PBS and 3 A1NPs-DilC18) were sacrificed at 24 590 hours. After transcardiac perfusion with PBS and then 4% PFA, the lungs were then fixed for 591 24 hours in 4% PFA, kept in 30% sucrose solution overnight, and then frozen in OCT at -80°C. 592 Immunostaining. Frozen sections (10 µm) were obtained using a cryostat ( Leica 593 CM1520; Leica Biosystems). OCT was eliminated with PBS and tissue section were incubated 594 with DAPI (1 μg/mL) at RT for 20 minutes. Ibidi mounting medium was used to see 595 fluorescence and images were obtained unsing a Nanozoomer S60 digital slide scanner 596 (Hamamatsu). 597 To assess the cell types capable of internalizing A1NPs, co -labeling involving ApoA1 and 598 specific markers was performed. Tissue sections were first subjected to antigen unmasking in 599 sodium citrate (pH 6), held at 80°C for 30 minutes. Once cooled to room temperature, the slides 600 were washed with PBS, then blocked for 90 minutes in PBS 0.1% Triton - 2% BSA. Then, 601 slides were incubated for a further 30 minutes in PBS, 0.1% Triton - 0.2% BSA with Fc Block 602 at 0.025 mg/mL. Slides were then incubated overnight at 4°C with specific primary antibodies. 603 For ApoA1 detection, a mouse anti -human ApoA1 primary antibody diluted 1:200 was used. 604 In parallel, specific antibodies were applied: a mouse anti-AGER rat antibody diluted 1:100 for 605 type I pneumocytes (MAB 1179-500; R&D Systems), a mouse anti -SFTPC rabbit antibody 606 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 20 diluted 1:200 for type II pneumocytes (10774-1AP; Proteintech), and a mouse anti-EMCN goat 607 antibody diluted 1:100 for endothelial cells (AF4666-SP; R&D Systems). After five minutes 608 washes with PBS 0.1% Triton, the slides were incubated with secondary antibodies coupled to 609 suitable fluorochromes, including a donkey anti-mouse Alexa Fluor 488, a goat anti-rat Alexa 610 Fluor 594, goat anti-rabbit Alexa Fluor 647, and donkey anti-goat Alexa Fluor 594, all diluted 611 at 1:1000. After this step, the slides were washed three times with PBS 0.1% Triton , then 612 incubated with DAPI at 1 µg/mL for 15 minutes for nuclei staining. 613 Finally, after 3 washes in PBS 0.1% Triton, slides were mounted with IBIDI medium. 614 Observations and acquisitions were made using a confocal microscope (Nikon Eclipse Ti2). 615 616 Air/liquid interface model . A total of 500,000 A549 cells were seeded into 617 polycarbonate cell culture inserts with pore-size of 0.4 μm (PIHP01250; Millipore) pre-coated 618 with 70 µg/mL of type I rat tail collagen (354236, Corning, USA). After 24 hours, the medium 619 in the apical compartment was removed and the medium at the basolateral side was replaced 3 620 times per week for 2 weeks34. 621 Transcytosis. To assess the potential of A1NPs to cross a reconstituted epithelial 622 barrier, the inserts were exposed to 0.05 mg/mL of A1NPs . Basolateral media were collected 623 after 30 minutes, 1 hour, 2 hours, 4 hours, and 6 hours and analyzed with a human ApoA1 624 ELISA assay (3710 -1HP-2, Mabtech). Absorbance was measured using a CLARIOstar plate 625 reader (BMG Labtech) at 450 -570 nm. In another set of experiments, A1NPs-exposed inserts 626 were washed with PBS (top and bottom), then fixed with 4% paraformaldehyde (PFA) for 15 627 minutes on the top and the bottom. Inserts were washed twice with PBS and could either be 628 stored at 4°C for 1 week or used directly. Membranes of insert were cut and washed with PBS 629 0.5% Triton for 5 minutes. Cells were blocked with PBS triton 0.5% BSA 4% for 30 minutes. 630 Primary anti-human ApoA1 antibody ( 178422; Calbiochem) was diluted 1:500 in PBS triton 631 0.5% BSA 1% and incubated for 45 minutes at RT. After 2 washes with PBS, membranes were 632 incubated with Alexa 594 goat anti-rabbit secondary antibody at 1:1000 dilution for 45 minutes 633 at RT. After 2 washes with PBS, membranes were incubated with Alexa Fluor 488 phalloidin 634 at dilution 1:2000 and DAPI (1μg/mL) for 45 minutes at RT. Membranes were mounted with 635 fluorescent medium and images were captured with a confocal microscope (Nikon Eclipse Ti2). 636 Permeability assay. Following 6 hours of transcytosis, the inserts were retrieved and 637 transferred into 300 µL of complete culture medium without phenol red (P04-16516, Pan 638 Biotech). Subsequently, 100 µL of dextran labeled with fluorescein isothiocyanate (FITC-639 dextran) 70 kDa (Sigma) at a concentration of 1 mg/mL was applied to the apical compartment. 640 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 21 The inserts were incubated at 37 °C for 40 minutes. Post -incubation, the basolateral medium 641 was collected for each experimental condition. The fluorescence intensity was quantified using 642 a spectrophotometer with excitation/emission wavelengths set at 490 ± 15 nm and 530 ± 30 nm 643 (BMG Labtech). The concentration of FITC -dextran (70 kDa) in the basolateral compartment 644 was determined using a standard calibration curve ranging from 0.25 mg/mL to 0.008 mg/mL. 645

was expressed as the relative concentration, calculated as the ratio between the initial 646 concentration applied at T₀ and the concentration measured at T₄₀ minutes (Ct40/Ct0). 647 Statistics. All statistical tests were performed on Graphpad Prism 5 software (Graphpad 648 Software, San Diego, CA). Results were displayed as mean ± SEM values of repeated 649 independent experiments. Statistical tests used were ordinary one-way ANOVA with Tukey’s 650 multiple comparisons test or paired T test. Results were considered statistically significant when 651 p<0.05. 652 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 22

653 (1) Kontush, A.; Lindahl, M.; Lhomme, M.; Calabresi, L.; Chapman, M. J.; Davidson, W. S. 654 Structure of HDL: Particle Subclasses and Molecular Components. Handb Exp Pharmacol 655 2015, 224, 3–51. https://doi.org/10.1007/978-3-319-09665-0_1. 656 (2) Rached, F. H.; Chapman, M. J.; Kontush, A. HDL Particle Subpopulatio ns: Focus on 657 Biological Function. Biofactors 2015, 41 (2), 67–77. https://doi.org/10.1002/biof.1202. 658 (3) Calabresi, L.; Franceschini, G. Lecithin:Cholesterol Acyltransferase, High -Density 659 Lipoproteins, and Atheroprotection in Humans. Trends Cardiovasc Med 2010, 20 (2), 50–660 53. https://doi.org/10.1016/j.tcm.2010.03.007. 661 (4) Rye, K.-A.; Bursill, C. A.; Lambert, G.; Tabet, F.; Barter, P. J. The Metabolism and Anti-662 Atherogenic Properties of HDL. J Lipid Res 2009, 50 Suppl (Suppl), S195 -200. 663 https://doi.org/10.1194/jlr.R800034-JLR200. 664 (5) Camont, L.; Chapman, M. J.; Kontush, A. Biological Activities of HDL Subpopulations 665 and Their Relevance to Cardiovascular Disease. Trends Mol Med 2011, 17 (10), 594–603. 666 https://doi.org/10.1016/j.molmed.2011.05.013. 667 (6) Barter, P. J.; Nicholls, S.; Rye, K.-A.; Anantharamaiah, G. M.; Navab, M.; Fogelman, A. 668 M. Antiinflammatory Properties of HDL. Circ Res 2004, 95 (8), 764 –772. 669 https://doi.org/10.1161/01.RES.0000146094.59640.13. 670 (7) Levels, J. H.; Abraham, P. R.; van den Ende, A.; v an Deventer, S. J. Distribution and 671 Kinetics of Lipoprotein -Bound Endotoxin. Infect Immun 2001, 69 (5), 2821 –2828. 672 https://doi.org/10.1128/IAI.69.5.2821-2828.2001. 673 (8) Soran, H.; Schofield, J. D.; Durrington, P. N. Antioxidant Properties of HDL. Front 674 Pharmacol 2015, 6, 222. https://doi.org/10.3389/fphar.2015.00222. 675 (9) Wilson, P. W.; Garrison, R. J.; Castelli, W. P.; Feinleib, M.; McNamara, P. M.; Kannel, 676 W. B. Prevalence of Coronary Heart Disease in the Framingham Offspring Study: Role of 677 Lipoprotein Chol esterols. Am J Cardiol 1980, 46 (4), 649 –654. 678 https://doi.org/10.1016/0002-9149(80)90516-0. 679 (10) Tardif, J.-C.; Grégoire, J.; L’Allier, P. L.; Ibrahim, R.; Lespérance, J.; Heinonen, T. M.; 680 Kouz, S.; Berry, C.; Basser, R.; Lavoie, M. -A.; Guertin, M.-C.; Rodés-Cabau, J.; Effect 681 of rHDL on Atherosclerosis -Safety and Efficacy (ERASE) Investigators. Effects of 682 Reconstituted High -Density Lipoprotein Infusions on Coronary Atherosclerosis: A 683 Randomized Controlled Trial. JAMA 2007, 297 (15), 1675 –1682. 684 https://doi.org/10.1001/jama.297.15.jpc70004. 685 (11) Michael Gibson, C.; Korjian, S.; Tricoci, P.; Daaboul, Y.; Yee, M.; Jain, P.; Alexander, J. 686 H.; Steg, P. G.; Lincoff, A. M.; Kastelein, J. J. P.; Mehran, R.; D’Andrea, D. M.; 687 Deckelbaum, L. I.; Merkely, B.; Zarebinski, M.; Ophuis, T. O.; Harrington, R. A. Safety 688 and Tolerability of CSL112, a Reconstituted, Infusible, Plasma -Derived Apolipoprotein 689 A-I, After Acute Myocardial Infarction: The AEGIS -I Trial (ApoA-I Event Reducing in 690 Ischemic S yndromes I). Circulation 2016, 134 (24), 1918 –1930. 691 https://doi.org/10.1161/CIRCULATIONAHA.116.025687. 692 (12) Tanaka, S.; Genève, C.; Zappella, N.; Yong -Sang, J.; Planesse, C.; Louedec, L.; 693 Viranaïcken, W.; Bringart, M.; Montravers, P.; Denamur, E.; Durantea u, J.; Couret, D.; 694 Meilhac, O. Reconstituted High -Density Lipoprotein Therapy Improves Survival in 695 Mouse Models of Sepsis. Anesthesiology 2020, 132 (4), 825 –838. 696 https://doi.org/10.1097/ALN.0000000000003155. 697 (13) Tanaka, S.; Begue, F.; Veeren, B.; Tran-Dinh, A.; Robert, T.; Tashk, P.; Lortat-Jacob, B.; 698 Faille, D.; de Chaisemartin, L.; Zappella, N.; Atchade, E.; Kramer, L.; Montravers, P.; 699 Meilhac, O. First Recombinant High -Density Lipoprotein Particles Administration in a 700 Severe ICU COVID-19 Patient, a Mult i-Omics Exploratory Investigation. Biomedicines 701 2022, 10 (4), 754. https://doi.org/10.3390/biomedicines10040754. 702 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 23 (14) Remaley, A. T.; Stonik, J. A.; Demosky, S. J.; Neufeld, E. B.; Bocharov, A. V.; 703 Vishnyakova, T. G.; Eggerman, T. L.; Patterson, A. P.; Duverger, N. J.; Santamarina-Fojo, 704 S.; Brewer, H. B. Apolipoprotein Specificity for Lipid Efflux by the Human ABCAI 705 Transporter. Biochem Biophys Res Commun 2001, 280 (3), 818 –823. 706 https://doi.org/10.1006/bbrc.2000.4219. 707 (15) Tran-Dinh, A.; Diallo, D.; Delbosc , S.; Varela-Perez, L. M.; Dang, Q. B.; Lapergue, B.; 708 Burillo, E.; Michel, J. B.; Levoye, A.; Martin -Ventura, J. L.; Meilhac, O. HDL and 709 Endothelial Protection. Br J Pharmacol 2013, 169 (3), 493 –511. 710 https://doi.org/10.1111/bph.12174. 711 (16) Lokugamage, M. P.; Vanover, D.; Beyersdorf, J.; Hatit, M. Z. C.; Rotolo, L.; Echeverri, 712 E. S.; Peck, H. E.; Ni, H.; Yoon, J. -K.; Kim, Y.; Santangelo, P. J.; Dahlman, J. E. 713 Optimization of Lipid Nanoparticles for the Delivery of Nebulized Therapeutic mRNA to 714 the Lungs. Nat Biomed Eng 2021, 5 (9), 1059–1068. https://doi.org/10.1038/s41551-021-715 00786-x. 716 (17) Leong, E. W. X.; Ge, R. Lipid Nanoparticles as Delivery Vehicles for Inhaled 717 Therapeutics. Biomedicines 2022, 10 (9), 2179. 718 https://doi.org/10.3390/biomedicines10092179. 719 (18) Matz, C. E.; Jonas, A. Micellar Complexes of Human Apolipoprotein A -I with 720 Phosphatidylcholines and Cholesterol Prepared from Cholate -Lipid Dispersions. J Biol 721 Chem 1982, 257 (8), 4535–4540. 722 (19) B Uribe, K.; Benito -Vicente, A.; Martin, C.; Blanco -Vaca, F.; Rotllan, N. (R)HDL in 723 Theranostics: How Do We Apply HDL’s Biology for Precision Medicine in 724 Atherosclerosis Management? Biomater Sci 2021, 9 (9), 3185 –3208. 725 https://doi.org/10.1039/d0bm01838d. 726 (20) Kim, Y.; Fay, F.; Cormode, D. P.; Sanchez-Gaytan, B. L.; Tang, J.; Hennessy, E. J.; Ma, 727 M.; Moore, K.; Farokhzad, O. C.; Fisher, E. A.; Mulder, W. J. M.; Langer, R.; Fayad, Z. 728 A. Single Step Reconstitution of Multifunctional High -Density Lipoprotein -Derived 729 Nanomaterials Using Microfluidics. ACS Nano 2013, 7 (11), 9975 –9983. 730 https://doi.org/10.1021/nn4039063. 731 (21) Zhang, L.; Song, J.; Newhouse, Y.; Zhang, S.; Weisgraber, K. H.; Ren, G. An Optimized 732 Negative-Staining Protocol of Electron Microscopy for apoE4 POPC Lipoprotein. J Lipid 733 Res 2010, 51 (5), 1228–1236. https://doi.org/10.1194/jlr.D002493. 734 (22) Ortega-Paz, L.; Giordano, S.; Capodanno, D.; Mehran, R.; Gibson, C. M.; Angiolillo, D. 735 J. Clinical Pharmacokinetics and Pharmacodynamics of CSL112. Clin Pharmacokinet 736 2023, 62 (4), 541–558. https://doi.org/10.1007/s40262-023-01224-8. 737 (23) Moreno, J. -A.; Ortega -Gomez, A.; Rubio -Navarro, A.; Louedec, L.; Ho -Tin-Noé, B.; 738 Caligiuri, G.; Nicoletti, A.; Levoye, A.; Plantier, L.; Meilhac, O. High -Density 739 Lipoproteins Potentiate Α1 -Antitrypsin Therapy in Elastase -Induced Pulmonary 740 Emphysema. Am J Respir Cell Mol Biol 2014, 51 (4), 536 –549. 741 https://doi.org/10.1165/rcmb.2013-0103OC. 742 (24) Kim, J.; Dey, A.; Malhotra, A.; Liu, J.; Ahn, S. I.; Sei, Y. J.; Kenney, A. M.; MacDonald, 743 T. J.; Kim, Y. Engineered Biomimetic Nanoparticle for Dual Targeting of the Cancer 744 Stem-like Cell Population in Sonic Hedgehog Medulloblastoma. Proc Natl Acad Sci U S 745 A 2020, 117 (39), 24205–24212. https://doi.org/10.1073/pnas.1911229117. 746 (25) Silver, D. L.; Wang, N.; Tall, A. R. Defective HDL Particle Uptake in Ob/Ob Hepatocytes 747 Causes Decreased Recycling, Degradation, and Selective Lipid Uptake. J Clin Invest 748 2000, 105 (2), 151–159. https://doi.org/10.1172/JCI8087. 749 (26) Silver, D. L.; Wang, N. ; Xiao, X.; Tall, A. R. High Density Lipoprotein (HDL) Particle 750 Uptake Mediated by Scavenger Receptor Class B Type 1 Results in Selective Sorting of 751 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint 24 HDL Cholesterol from Protein and Polarized Cholesterol Secretion. J Biol Chem 2001, 752 276 (27), 25287–25293. https://doi.org/10.1074/jbc.M101726200. 753 (27) Rohrer, L.; Ohnsorg, P. M.; Lehner, M.; Landolt, F.; Rinninger, F.; von Eckardstein, A. 754 High-Density Lipoprotein Transport through Aortic Endothelial Cells Involves Scavenger 755 Receptor BI and ATP-Binding Cassette Transporter G1. Circ Res 2009, 104 (10), 1142–756 1150. https://doi.org/10.1161/CIRCRESAHA.108.190587. 757 (28) He, P.; Smith, A.; Gelissen, I. C.; Ammit, A. J. The Effect of Statins and the Synthetic 758 LXR Agonist T0901317 on Expression of ABCA1 Transporter Protei n in Human Lung 759 Epithelial Cell Lines in Vitro. Pharmacol Rep 2019, 71 (6), 1219 –1226. 760 https://doi.org/10.1016/j.pharep.2019.08.006. 761 (29) He, P.; Gelissen, I. C.; Ammit, A. J. Regulation of ATP Binding Cassette Transporter A1 762 (ABCA1) Expression: Cholestero l-Dependent and – Independent Signaling Pathways 763 with Relevance to Inflammatory Lung Disease. Respir Res 2020, 21 (1), 250. 764 https://doi.org/10.1186/s12931-020-01515-9. 765 (30) Cavelier, C.; Rohrer, L.; von Eckardstein, A. ATP -Binding Cassette Transporter A1 766 Modulates Apolipoprotein A -I Transcytosis through Aortic Endothelial Cells. Circ Res 767 2006, 99 (10), 1060–1066. https://doi.org/10.1161/01.RES.0000250567.17569.b3. 768 (31) Hrzenjak, A.; Reicher, H.; Wintersperger, A.; Steinecker -Frohnwieser, B.; Sedlmayr, P.; 769 Schmidt, H.; Nakamura, T.; Malle, E.; Sattler, W. Inhibition of Lung Carcinoma Cell 770 Growth by High Density Lipoprotein-Associated Alpha-Tocopheryl-Succinate. Cell Mol 771 Life Sci 2004, 61 (12), 1520–1531. https://doi.org/10.1007/s00018-004-4101-4. 772 (32) Denimal, D. Antioxidant and Anti -Inflammatory Functions of High-Density Lipoprotein 773 in Type 1 and Type 2 Diabetes. Antioxidants (Basel) 2023, 13 (1), 57. 774 https://doi.org/10.3390/antiox13010057. 775 (33) Yin, K.; Deng, X.; Mo, Z. -C.; Zhao, G.-J.; Jiang, J.; Cui, L. -B.; Tan, C.-Z.; Wen, G.-B.; 776 Fu, Y.; Tang, C. -K. Tristetraprolin -Dependent Post -Transcriptional Regulation of 777 Inflammatory Cytokine mRNA Expression by Apolipoprotein A-I: Role of ATP-Binding 778 Membrane Cassette Transporter A1 and Signal Transducer and Activator of Transcription 779 3. J Biol Chem 2011, 286 (16), 13834–13845. https://doi.org/10.1074/jbc.M110.202275. 780 (34) Wu, J.; Wang, Y.; Liu, G.; Jia, Y.; Yang, J.; Shi, J.; Dong, J.; Wei, J.; Liu, X. 781 Characterization of Air-Liquid Interface Culture of A549 Alveolar Epithelial Cells. Braz 782 J Med Biol Res 2017, 51 (2), e6950. https://doi.org/10.1590/1414-431X20176950. 783 (35) Rosanaly, S.; Apalama, M. L.; Bringart, M.; Giraud, P.; Allard, B.; Veeren, B.; Meilhac, 784 O.; Couprie, J.; Rondeau, P. Production, Characterization an d Biodistribution of 785 Therapeutic High-Density Lipoprotein-like Nanoparticles Reconstituted with or without 786 Histidine-Tagged Recombinant ApoA1. Biochim Biophys Acta Mol Cell Biol Lipids 2025, 787 1870 (3), 159606. https://doi.org/10.1016/j.bbalip.2025.159606. 788 789 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663869doi: bioRxiv preprint