This is the accepted version of the journal article: Voltà-Durán, Eric; Parladé Molist, Eloi; Serna, Naroa; [et al.]. «Endosomal es- cape for cell-targeted proteins. Going out after going in». Biotechnology advances, Vol. 63 (March-April 2023), art. 108103. DOI 10.1016/j.biotechadv.2023.108103 This version is available at https://ddd.uab.cat/record/283745 under the terms of the license 1 Endosomal escape for cell-targeted proteins. Going out 2 after going in 3 Eric Voltà-Durán 1,2,3 , Eloi Parladé 1,2,3 , Naroa Serna 1,2,3,&, Antonio Villaverde 1,2,3, 4 *, Esther Vazquez 1,2,3, * and Ugutz Unzueta 2,3,4,5,* 5 1 6 Institut de Biotecnologia i de Biomedicina, Universitat Autònoma de Barcelona, 08193 7 Cerdanyola del Vallès (Barcelona), Spain. 2 8 CIBER de Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Instituto de Salud Carlos 9 III, 08193 Cerdanyola del Vallès (Barcelona), Spain. 3 10 Departament de Genètica i de Microbiologia, Universitat Autònoma de Barcelona, 08193 11 Cerdanyola del Vallès (Barcelona), Spain. 4 12 Biomedical Research Institute Sant Pau (IIB Sant Pau), 08025 Barcelona, Spain. 5 13 Josep Carreras Leukaemia Research Institute, 08916 Badalona (Barcelona), Spain. & 14 Present address: Nanoligent SL. Edifici Eureka, Universitat Autònoma de Barcelona, Bellaterra, 15 08193 Barcelona, Spain. * 16 Corresponding Authors: 17 AV antonio.villaverde@uab.cat 18 EV esther.vazquez@uab.cat 19 UU uunzueta@santpau.cat 20 21 22 23 24 25 26 27 28 29 30 1 31 Abstract: 32 Protein-based nanocarriers are versatile and biocompatible drug delivery systems. They 33 are of particular interest in nanomedicine as they can recruit multiple functions in a 34 single modular polypeptide. Many cell-targeting peptides or protein domains can 35 promote cell uptake when included in these nanoparticles through receptor-mediated 36 endocytosis. In that way, targeting drugs to specific cell receptors allows a selective 37 intracellular delivery process, avoiding potential side effects of the payload. However, 38 once internalized, the endo-lysosomal route taken by the engulfed material usually 39 results in full degradation, preventing their adequate subcellular localization, 40 bioavailability and subsequent therapeutic effect. Thus, entrapment into endo- 41 lysosomes is a main bottleneck in the efficacy of protein-drug nanomedicines. 42 Promoting endosomal escape and preventing lysosomal degradation would make this 43 therapeutic approach clinically plausible. In this review, we discuss the mechanisms 44 intended to evade lysosomal degradation of proteins, with the most relevant examples 45 and associated strategies, and the methods available to measure that effect. In addition, 46 based on the increasing catalogue of peptide domains tailored to face this challenge as 47 components of protein nanocarriers, we emphasize how their particular mechanisms of 48 action can potentially alter the functionality of accompanying protein materials, 49 especially in terms of targeting and specificity in the delivery process. 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Keywords: 66 Protein nanocarriers, Protein engineering, Targeting, Endosomal Escape. 67 2 68 1. Introduction: 69 Drug targeting is an urgent and unmet need in molecular medicine and especially in 70 oncology, where most of the administered drugs display potent cytotoxic effects (Bazak 71 et al., 2015). In this sense, cell-specific drug delivery is usually explored through the 72 development of nanoscale drug carriers (Raj et al., 2021), that are functionalized with 73 ligands for specific cell surface molecules overexpressed in target cells (Jiang et al., 74 2019). For that, a nanoscale size reached through the self-assembly of building blocks 75 into homo-oligomers is particularly convenient to favor the intended biodistribution, to 76 exploit the enhanced retention and permeability effect (EPR), and to allow multiple 77 display of the cell ligand for cooperative receptor binding, resulting into highly selective 78 cell penetrability through endosomal routes (Duncan and Gaspar, 2011). 79 In this context, cell-targeted delivery systems, in form of functional nanoparticles, are 80 being developed as promising nanomedical tools. Among the diversity of materials 81 explored to construct suitable nanoparticles, proteins represent an emerging and very 82 useful category of biomaterials. This is because of their full biocompatibility, 83 biodegradability and easy production in cell factories(Corchero et al., 2014), together 84 with their capability to tailor many of the functions valuable in nanomedicine such as 85 self-assembly or specific molecular recognition and binding. In addition, the 86 extraordinary plasticity of protein materials allows the recruitment of different 87 functional domains such as targeting ligands, architectonic tags or nuclear localization 88 signals, in a single multifunctional polypeptide by simple genetic fusion (Ferrer-Miralles 89 et al., 2015; Vazquez et al., 2009). In this regard, protein drugs can be also incorporated 90 to such constructs resulting in unique species of protein-only, self-organized nanosized 91 therapeutics (Casanova et al., 2019; Serna et al., 2018). 92 Cell-targeted protein delivery usually relies on receptor-mediated endocytic processes, 93 in which early endosome vesicles rapidly mature into late endosomes and finally fuse 94 with lysosomes. This fusion leads to the complete degradation of the engulfed protein 95 material by lysosomal enzymes and acidic conditions (Bareford and Swaan, 2007; 96 Huotari and Helenius, 2011). Since most of delivered drugs, including proteins, do not 97 perform their activities in lysosomes but in the cell cytoplasm or in other subcellular 98 compartments, internalized nanomaterials usually need to be efficiently released from 99 endosomes to reach their intracellular therapeutic target (Lee et al., 2019). Thus, the 100 entrapment of protein-based therapeutics into endo-lysosomes is a main reason for 101 their limited efficacy (Erickson et al., 2006; LeCher et al., 2017; Pirie et al., 2011). Some 102 strategies as the incorporation of non-natural amino acids are intended to delay protein 103 degradation within the lysosomes (Kato et al., 2016; Yamashita et al., 2014) . However, 104 these strategies could complement but not replace the incorporation of efficient 105 endosomal escape (EE) agents, including peptides or protein domains, which are 106 mandatory to protect their final functionality. To such end, peptides with endosome- 107 disrupting capacities have been recruited into protein constructs through genetic fusion 108 to act in cis, fitting the modular setting (Ahmad et al., 2019; Varkouhi et al., 2011). 109 However, most of the described EE peptides act through fusogenic or pore-forming 110 activities, that being receptor-independent usually promote unspecific cell interactions 111 and penetration, which can potentially impair the highly desired specific cell binding 112 capacities of targeted constructs (Sánchez-García et al., 2017; Serna et al., 2019; 113 Varkouhi et al., 2011). Therefore, although current nanomedical approaches have 3 114 already developed a huge armamentarium of EE peptides, it is critical to evaluate which 115 of them are potentially suitable to be incorporated into targeted delivery platforms, 116 without affecting their cell specificity throughout the delivery process. 117 Taking all these points into consideration, the rational design of protein nanocarriers 118 devoted to targeted intracellular delivery of protein-based and chemical drugs or 119 imaging agents, that are sensitive to lysosomal degradation, should look forward to the 120 incorporation of efficient EE peptides that do not interfere into receptor-mediated 121 internalization. In addition, the fine evaluation of EE also requires very accurate 122 analytical tools especially refined for such purpose. Altogether, this review aims to 123 discuss the different mechanisms of action employed by currently described EE 124 peptides, focusing on their potential interference over cell-surface receptor targeting. 125 Moreover, we will also discuss the recent advances in the methods used to track these 126 processes, highlighting the EE protein domains that have already been empirically tested 127 for receptor-targeted delivery approaches as well as the ones predicted to be especially 128 suited for that. 129 2. Mechanisms of endosomal escape 130 Understanding the particular mechanisms used by different peptides and protein 131 domains to control the EE of proteins is a key element for the proper rational design of 132 protein-based nanomedicines (Mejia et al., 2022; Pei, 2022). For that, it is important to 133 note that in the context of cell-targeted proteins, EE peptides are usually recruited 134 within the same modular polypeptide (in cis) in order to induce the EE of the whole 135 protein after receptor-mediated endocytosis. In this sense, although some particular 136 peptides are supposed to induce the EE of accompanying proteins in trans, they could 137 also potentially induce the release of their neighbor internalized polypeptides when 138 incorporated in cis in a fusion protein. In some cases, the incorporation of a furin 139 cleavable site between the different protein modules is also mandatory to allow the 140 endosomal cleavage of the EE domain from the polypeptide to further induce the EE of 141 the cargo protein in trans. All together, the most studied approaches involve the use of 142 membrane fusion peptides (Ahmad et al., 2019) or pore-forming peptides (LeCher et al., 143 2017) among other more sophisticated strategies. 144 2.1 Pore formation. 145 Highly cationic peptides are widely known to induce pore formation in lipid bilayers. 146 They act through three main mechanisms accounted by the barrel stave, carpet or 147 toroidal models (Figure 1A) (Matsuzaki, 2019; Priyadarshini et al., 2022). The barrel- 148 stave model involves the self-assembly of a pore-forming peptide on the surface of the 149 membrane for the subsequent internalization of the protein or oligomer through the 150 hydrophobic region of the lipid bilayer, creating a channel (Figure 1A, i). The carpet 151 model, however, implies a dual-step mechanism in which the oligomerization of the 152 protein is not necessary to create the pore. Instead, the pore-forming peptide binds to 153 lipids on the membrane through electrostatic interactions, covering the surface. 154 Eventually, upon reaching an adequate concentration, a consequent micellization occurs 155 and leads to the membrane dissolution (Figure 1A, ii). Finally, the toroidal mechanism 156 also involves a dual step process in which the pore-forming peptide is first oriented in 157 parallel to the lipid bilayer to be then reoriented perpendicularly to penetrate the 158 hydrophobic region of the membrane. Thus, along with some lipid molecules, they form 4 159 lipid-lined pores and contiguous membrane leaflets that result in an irreversible rupture 160 of the membrane (Figure 1A, iii). 161 2.1.1 Pore-forming peptides in endosomal escape. 162 Some authors have claimed the use of pore-forming peptides as EE promoters. 163 However, it is important to note that being highly cationic, most of them can also 164 interact with extracellular membranes. Arginine-rich cell penetrating peptides (CPP) 165 such as TAT, Penetratin or 8R are clear examples of this kind of pore-forming motifs, 166 which show interesting endosomolytic activities and can also induce unspecific cell 167 uptake via micropinocytosis (Lönn and Dowdy, 2015). In this regard, the dimeric 168 fluorescent TAT (dfTAT) has been engineered as a potent and non-toxic endosomolytic 169 agent that efficiently induces the cytosolic leakage of accompanying molecules in late 170 endosomes (Erazo-Oliveras et al., 2016, 2014). However, dfTAT was proven to also 171 penetrate cells majorly through micropinocytosis (Allen et al., 2018), which makes it a 172 poor candidate as a partner domain in targeted delivery settings. 173 Some other pore-forming peptides have exhibited intrinsic cytotoxic effects, which 174 makes them also unsuited for general drug delivery purposes. This set includes the 175 antimicrobial peptide (AMP) GWH1 and other bacterial peptides such as the 176 pneumococcal pneumolysin (Browne et al., 1999), streptococcal streptolysin O (SLO) 177 (Browne et al., 1999) or the Listeria listeriolysin O (LLO) (Plaza-GA et al., 2019). Similar 178 problems were reported for the bee venom-derived peptide melittin, whose cell 179 membrane-disrupting activities produced high cell toxicity at neutral pH (Kandil et al., 180 2019; Paray et al., 2021). However, the engineering of novel peptide analogs with 181 reduced cell cytotoxicity and enhanced endosome-selective pore-forming activity has 182 recently generated growing interest (Paray et al., 2021). In this regard, the substitution 183 of basic residues for glutamic acids in melittin and two other membrane-disrupting 184 AMPs (LL-37 and bombolitin V) produced pH-sensitive analogs with selective membrane 185 disrupting activity at low pH (pH 5.0), with no evidence of cell membrane permeability 186 or toxicity at neutral pHs (Ahmad et al., 2015). The same occurs with the novel synthetic 187 leucine zipper endosomolytic peptide (LZEP), whose pore-forming ability is obtained 188 only under acidic conditions(Ahmad et al., 2021). In a similar way, Aurein 1.2 is an AMP 189 of animal origin that selectively disrupts endosomal membranes via the carpet 190 mechanism (Fernandez et al., 2012). Thus, it enhances the EE of cargo proteins in vitro 191 and in vivo with no toxicity over mammalian cells (Li et al., 2015). Recently, Aurein 1.2 192 has been successfully conjugated to an EGFR-targeted fluorescent protein, promoting 193 improved intracellular protein accumulation without affecting its specificity or 194 functionality (Lieser et al., 2022). 195 Related to that, in an attempt to selectively target other pore-forming peptides to 196 cancer cells, the AMP Ceratotoxin A (CtxA) was also recently conjugated to folic acid and 197 tested over folate-receptor overexpressing KB cancer cells (Mayer et al., 2019). In this 198 sense, CtxA is a 36-aa cationic peptide that efficiently forms pores into cell membranes 199 following the barrel stave model. However, and opposed to expected, folic-acid targeted 200 CtxA (CtxA-FA) showed no difference over KB cancer cells compared to wild type CtxA 201 (Mayer et al., 2019). Similar results were also reported for a Her2-targeted lysteriolysin 202 O (LLO) (Fotoohi-Ardakani et al., 2019). In contrast, recent work targeting LLO to LHRH- 203 receptor showed some extend of target-cell selectivity (Kheirandish et al., 2019). 5 204 Nevertheless, all these AMPs exerted their pore-forming activity over external cell 205 membranes, showing no specificity for endosomes. 206 In summary, despite being quite useful to promote the cytosolic delivery of some 207 accompanying proteins, cationic pore-forming peptides still display important cell 208 penetrating properties that are generally accompanied with unspecific targeting and cell 209 membrane-lysis associated toxicities. Therefore, unless adapted to exhibit endosome- 210 selective activities, they may not be the most appropriate choice when dealing with 211 targeted delivery approaches. Related to that, the described attempts to selectively 212 target pore-forming peptides to cancer cells were in general not very encouraging either 213 (Fotoohi-Ardakani et al., 2019; Kheirandish et al., 2019; Mayer et al., 2019). 214 2.2. Fusion with lipid membranes. 215 Some peptides are able to induce the release of the endosomal content by driving the 216 fusion of lipid membranes they are embedded in with the endosomes. This class of 217 peptides are usually rich in bulky hydrophobic residues and show large proportions of 218 glycine and alanine residues (Pei and Buyanova, 2019). Thus, a change in the pH during 219 the endosome maturation process induces a conformational shift in the peptide, which 220 exposes the hydrophobic residues that interact then with the endosomal membranes. 221 This particular mechanism of action is extensively present in viral proteins (Pal, 2021; 222 Somiya and Kuroda, 2020). As a representative example (Figure 1B), the well-known 223 fusogenic protein from the influenza virus hemagglutinin-2 (HA2) undergoes a 224 conformational change at acidic pH due to the protonation of glutamate and aspartate 225 residues. Consequently, the hydrophobic residues of the peptide become exposed and 226 trigger the fusion of the free HA2 end to the endosomal membrane. Thus, both ends of 227 the same HA2 polypeptide become embedded in two parallel (viral and endosomal) 228 membranes. Finally, the protein folds into a hairpin-like conformation that approaches 229 both protein ends leading to a close contact of the two membranes which eventually 230 fuse (Worch, 2014). 231 2.2.1 Fusogenic peptides in endosomal escape 232 Most of the EE peptides described in the literature act as fusogenic peptides. This 233 category includes peptides with viral (HA2, L2, gp41, gpH, WNV-ENV, INF-7 or DEN), 234 animal (B18), synthetic (KALA, GALA or GALA3) or even human (S19) origin (Varkouhi et 235 al., 2011). As stated, the HA2 peptide and its derivatives (as INF-7) are the most 236 representative members of this group(Lin et al., 2014; Sung et al., 2013). However, 237 although having shown potential in targeted delivery approaches (Liu et al., 2017; Sala 238 et al., 2019) there are still important concerns about whether the peptides remain 239 attached to endosomal membranes after promoting their lysis (Lee et al., 2011) and the 240 influence of these agents over the selectivity of proteins intended to be delivered in a 241 targeted way (Lee et al., 2011; Lieser et al., 2022; Sánchez-García et al., 2017). 242 Similar worries arise with the use of GALA, a synthetic amphipathic peptide consisting 243 in EALA (glutamic acid, alanine, leucine, alanine) repetitions, whose activity is based on 244 glutamic acid protonation at acidic pH (Nakase et al., 2011). Thus, the incorporation of 245 specific acidic amino acids (Glu) in the sequence has served to generate a pH-sensitive 246 peptide, whose charge completely changes from physiological to acidic pH, resulting in 247 a pH-selective membrane interaction activity (Ahmad et al., 2019). In this sense, despite 6 248 maintaining some specificity in a targeted approach(Lieser et al., 2022), GALA has failed 249 to induce the cytosolic release of some proteins due to its high hydrophobicity and 250 tendency to aggregate on the plasma membrane (Li et al., 2020). A rationally-modified 251 version (GALA3) with reduced repeated units of EALA has successfully solved this 252 problem and has efficiently induced the EE of a non-targeted fusion protein (Li et al., 253 2020). However, further exploration is required to determine the suitability of this new 254 version for targeted delivery settings. 255 Related to that, the PreS2-domain of the hepatitis-B virus surface antigen is one of the 256 first reported fusogenic peptides that despite being described as a cell permeable motif 257 (Oess and Hildt, 2000) has been used to promote the cytosolic delivery of an EGFR- 258 targeted protein (Saporin), reducing its side effects and enhancing its antitumor activity 259 in vivo (Fuchs et al., 2007). The 19-aa peptide DEN (from the Dengue virus envelope 260 protein) is another recently reported fusogenic peptide (Huang et al., 2010) that, by 261 showing selective affinity for specific lipids present in the late endosomes (Zaitseva et 262 al., 2010), has also been successfully used in targeted delivery approaches (Kiesgen et 263 al., 2014). In this sense, the addition of DEN efficiently promoted the in vitro cytotoxic 264 activity of an EGFR-targeted Ranpirnase (RNase) without affecting its cells selectivity 265 (Kiesgen et al., 2014). Therefore, enlightened by the promising applicability of these 266 peptides in targeted delivery, their potential immune reactivity, derived from their viral 267 origin, needs to be still addressed. 268 A similar concern was raised regarding the use of the 18-aa fusogenic peptide (B18) from 269 the sea urchin fertilization protein (animal origin) that was successfully used for the 270 intracellular delivery a of TAT-guided GFP protein (Niikura et al., 2015) and in an EGFR- 271 targeted approach (Niikura et al., 2016). To solve this problem, a potentially less 272 immunogenic 19-aa fusogenic peptide (S19) derived from the human protein syncytin 1 273 has been recently described by the same group and successfully used to enhance the 274 intracellular delivery of different proteins including GFP, SNAP-tag or β-galactosidase 275 into mammalian cells in absence of cytoxicity (Sudo et al., 2017). However, although the 276 reported assays suggested an endosome-specific affinity of S19, cell uptake was still 277 induced by TAT-mediated micropinocytosis in the tested proteins. Therefore, its real 278 influence over a targeted setting is yet to be determined. 279 All in all, the use of fusogenic peptides that undergo conformational changes at acidic 280 pH and that therefore become active at late endosomes, is an attractive strategy 281 because of their small size, presumable endosome-specificity and lack of cell toxicity. 282 Nevertheless, it is still unclear how peptides that promote fusion between lipidic 283 membranes (which makes them extremely appealing for membrane-based delivery 284 approaches such as liposomes) can assist the release of proteins without a lipidic 285 envelope from the endosomal lumen to the cytosol. Therefore, although some 286 promising peptides as B18, DEN or S19 have already been reported, further exploration 287 is needed to better understand their particular mechanism of action and determine if 288 their general use in targeted delivery settings is a near reality or not. 289 2.3 Proton-sponge mechanism 290 Basic amino acids are hydrophilic, polar and cationic in nature. In this sense, histidine is 291 the only basic amino acid that is found deprotonated at physiological pH and becomes 292 protonated at slightly lower pH (pKa = 6) (He et al., 2020). During the acidification 7 293 process of endosomes, the imidazole ring of histidine gets protonated, accepting 294 hydrogen molecules from the medium and neutralizing the endosomal pH. This fact 295 results in an increased active transport of hydrogen protons inside the endosomes along 296 with an influx of chloride ions to reach the desired pH. In consequence, a passive 297 transport of water molecules into the endosomes promotes their osmotic swelling and 298 subsequent lysis. This event is known as the proton sponge effect (Figure 1C). Based on 299 such principle, the presence of histidine residues in a given protein sequences has 300 already been reported to assist its endo-lysosomal escape (Liu et al., 2022; Váňová et 301 al., 2022). 302 2.3.1 Histidine-rich peptides in endosomal escape 303 Genetic fusion of histidine tags, mainly hexahistidine, is a very common procedure in 304 biotechnology, not only to allow immobilized metal affinity chromatography (IMAC) of 305 recombinant proteins but also, as an architectonic tag for protein assembly as nanoscale 306 entities (López-Laguna et al., 2022). In this context, the role of polyhistidines as EE 307 peptides has gained increasing interest and has been extensively studied, especially in 308 targeted drug delivery, as such peptides do not interfere with the targeting properties 309 of the protein (Dhankher et al., 2021; Yu et al., 2021). Interestingly, and as an additional 310 feature, the incorporation of hexahistidine tags in a carrier protein allows its direct 311 release into the cytosol through the proton sponge effect, avoiding any need to interact 312 with lipid bilayers. Being then hexahistidine peptides appealing in drug delivery, their 313 potential immunogenicity has been recently addressed by the generation of humanized 314 histidine-rich variants that retain endosomolytic properties while still allowing an 315 efficient purification via IMAC (López-Laguna et al., 2020). 316 Other histidine-rich peptides have also been used to induce the cytosolic release of 317 proteins intended for targeted delivery. As an example, the peptide ppTG21 successfully 318 promotes the endosomal escape of CRISPR-Cas ribonucleoproteins in trans in a cell- 319 targeted system (Rouet and Christ, 2019). Histidine residues have also been 320 incorporated into previously described pore-forming peptides to improve their 321 endosomolytic properties. In this regard, EB1 is a synthetic analog of penetratin in which 322 certain residues have been substituted for histidines to yield an amphipathic alpha helix 323 upon protonation at acidic pH (Lundberg et al., 2007). 324 Therefore, the use of histidine-rich peptides appears as a very promising strategy to 325 induce the cytosolic release of targeted fusion proteins. In this regard, the major 326 challenge related to the use of this type of EE peptides is determining the minimal 327 amount of histidine residues that need to be incorporated to promote an effective 328 proton sponge effect within the endosome. 329 2.4 Protein translocation to the cytosol 330 This particular EE mechanism has been mainly described in the diphtheria toxin (DT) 331 from Corynebacterium diphtheriae (Ladokhin et al., 2021). DT is a potent toxin 332 composed of three different domains including a receptor (R domain), a translocation (T 333 domain) and a catalytic domain (C domain). Thus, the R domain servers as a ligand for 334 the receptor-mediated endocytosis of the toxin. Then, a furin cleavable site located 335 between the T and C domains, drives the separation of both domains at early 336 endosomes, which remain attached through an intermolecular disulfide bond. Later, at 8 337 acidified endosomes, the T domain undergoes a conformational change that results in 338 its endosomal membrane insertion, forming a small pore through which the C domain is 339 specifically transported to the cytosol. There, the intermolecular disulfide bond is 340 immediately cleaved by cytosolic thioredoxin reductase and the C domain is finally 341 refolded back into its active conformation (assisted by cytosolic Hsp90 chaperone) to 342 catalyze the ADP-ribosylation of the elongation factor 2 and promote cell death (Figure 343 1D) (Schuster et al., 2017). 344 A similar strategy was reported for the anthrax toxin from Bacillus anthracis. However, 345 here, the protective antigen (PA) is both the receptor-binding and the pore-forming 346 subunit (Friebe et al., 2016). Thus, once internalized by receptor-mediated endocytosis, 347 the PA undergoes a conformational change at early endosomes that leads to pore 348 formation and subsequent translocation of the enzymatic subunits across the 349 membrane (Friebe et al., 2016). In this regard, although PA has already served for the 350 translocation of some proteins (Bachran et al., 2013; Liao et al., 2014), its use in a 351 targeted setting is not indicated because the PA itself shares both, intrinsic cell-targeting 352 and EE properties. 353 2.4.1 Diphtheria-toxin assisted translocation in endosomal escape 354 Fused to the N-terminal catalytic domain of the toxin, several proteins of multiple sizes 355 and structures including Sumo, α-amylase or m-Cherry, have been efficiently delivered 356 to the cytoplasm of cells, hijacking the translocation process of the toxin (Auger et al., 357 2015). However, this strategy requires the use of both T (21.6 kDa) and C (20.4 kDa) 358 domains of the toxin in order to transport other accompanying proteins to the cytosol. 359 The T domain alone has also been reported to be sufficient to promote EE of targeted 360 proteins (Gilyazova et al., 2006). Nonetheless, it is still unclear how the T domain might 361 assist the translocation of other proteins in absence of their disulfide linking. In this 362 regard, a small region of the catalytic domain that includes its target cysteine to allow 363 the disulfide-bond formation has been also incorporated to some protein designs (Aullo 364 et al., 1993). Going still further, direct disulfide-conjugation to the diphtheria T-domain 365 has also allowed the successful cytosolic delivery of the catalytic domain from the toxin 366 ricin (Sundan et al., 1982). However, a partial interference of the T Domain over the 367 receptor-specificity of a tumor targeted oligomeric protein nanoparticle has been also 368 recently reported (Voltà-Durán et al., 2022). 369 Therefore, the diphtheria toxin T domain appears as an attractive tool in targeted 370 proteins delivery since it shows high protein specificity, exclusively transporting the 371 tagged protein without disrupting endosomes. However, its potential interference over 372 the cell-targeting of some protein constructs needs to be closely watched. Also, the 373 minimal domain segment required for such process needs to be still fully determined 374 and its potential immunogenicity, derived from its large size and bacterial origin, needs 375 to be also addressed. 376 2.5 Lysosomal bypassing 377 Some pathogens and many plant and bacterial toxins such as Pseudomonas aeruginosa 378 exotoxin A, ricin, cholera toxin, Shiga toxin, pertussis toxin or cytolethal distending 379 toxins, hijack the Endoplasmic Reticulum-Associated Degradation (ERAD) mechanism to 380 reach the cell cytosol and avoid lysosomal degradation (Morito and Nagata, 2015; 9 381 Nowakowska-Gołacka et al., 2019). This protein clearance mechanism allows misfolded, 382 misassembled, or metabolically regulated proteins to be specifically dislocated from the 383 endothelial reticulum into the cytosol to be degraded. Thus, once internalized by 384 receptor-mediated endocytosis, toxins are retrogradely transported to the endoplasmic 385 reticulum (ER), where the enzymatically active subunit is disassembled from the 386 holotoxin and delivered to the cytosol (Figure 1E) (Nowakowska-Gołacka et al., 2019). It 387 is important to note that the process of retrograde transport, extraction from the ER 388 and refolding in the cytosol is particular for each toxin and it involves complex 389 procedures in which different receptors, chaperones and other proteins take part 390 (Mantis, 2012; Sowa-Rogozińska et al., 2019; Tsai et al., 2001). 391 2.5.1 Lysosomal bypassing in endosomal escape 392 Taking advantage of the ERAD pathway, some protein fragments from the above- 393 mentioned toxins have been recently used to drive the lysosomal bypassing of cargo 394 proteins. For example, the translocation domain of P. aeruginosa exotoxin A has allowed 395 the cytosolic delivery of different proteins including GFP and monobodies (Mohammed 396 et al., 2012; Schmit et al., 2019). However, there is still a need to understand which is 397 the exact mechanism by which fused proteins are efficiently translocated from the ER 398 to the cytosol and then properly refolded to avoid their degradation. 399 In an attempt to reduce the toxin segment needed for this process, a simpler approach 400 that just includes the furin cleavable site (FCS) and the C-terminal signal tetrapeptide 401 KDEL has been successfully used to induce the cytosolic delivery of CXCR4-targeted 402 catalytic segment of exotoxin A and ricin toxin (Díaz et al., 2018; Sánchez-García et al., 403 2018). Of note, both constructions lacked the full translocation domain, but also, both 404 toxins originally use the ERAD mechanism in their natural intoxication process (Inoue et 405 al., 2011; Michalska and Wolf, 2015). Therefore, it is important to know if this simpler 406 strategy can be transversally used for the lysosomal bypassing of other cargo proteins 407 or is strictly restricted to specific toxins. 408 Taking all presented EE mechanisms into consideration, the stronger candidates to be 409 potentially included in targeted delivery protein-based nanocarriers have been 410 summarized in Table 1. Here, pore-forming peptides are presented as probably the less 411 suitable candidates since their cell penetrating properties can strongly interfere with 412 protein targeting. In consequence, just few engineered pH-sensitive analogs derived 413 from antimicrobial peptides (Mellitin, Bombilitin V, LL-37) and especially the animal 414 peptide Aurein 1.2 has raised interest for a targeted-delivery setting. Fusogenic peptides 415 could represent a more sophisticated option since their pH-selective activity better 416 restricts their fusogenic action to endosomes. Moreover, some of the candidates (DEN, 417 preS2, B18) have already been validated in EGFR-targeted delivery approaches, and 418 human-derived peptides have also been described (S19). Nevertheless, the precise 419 mechanism by which peptides that promote membrane fusion are able to induce the 420 transportation of proteins across the endosomal membrane is still unclear. Histidine- 421 rich peptides are presented in this review as one of the most promising candidates for 422 cell-targeted delivery settings. This is because the proton sponge effect is endosome- 423 exclusive and although they display milder endosomolytic potency, they show no 424 interference with extracellular protein targeting. Moreover, humanized candidates such 425 as H5E (HEHEHEHEHE) have been also successfully developed(López-Laguna et al., 10 426 2020). Finally, the use of particular fragments of toxin proteins for direct endosomal 427 translocation (Diphteria toxin T Domain) or lysosomal by-passing (Exotoxin A T Domain, 428 FCS + KDEL) has raised substantial interest but also important concerns at the same time. 429 In this sense, they appear as attractive strategies because of their high protein specificity 430 and general compatibility with protein targeting. However, their exact mechanism of 431 action is yet to be revealed and in some cases their use may be also restricted to certain 432 proteins or show partial interferences with targeting. Additionally, toxin domains are 433 generally large segments within bacterial proteins (up to 21 kDa). Thus, their potential 434 immunogenicity requires proper assessment if they are to be used in a clinical setting. 435 436 3. Analytical evaluation of the endosomal escape: 437 Appropriate tools for EE evaluation are important to properly determine the adequate 438 performance of the EE elements incorporated to drug delivery platforms. In this sense, 439 many methodologies, based on diverse protein properties, have been proposed to 440 follow-up their intracellular fate. In general, all these strategies can be classified in two 441 main groups that include basic and more simple assays performed using artificial 442 membranes (ex cellulo) or complete and more complex assays performed in real cells (in 443 cellulo). 444 3.1 Ex cellulo assays 445 Membrane models such as liposomes have been frequently used as a preliminary test 446 for EE of peptides and protein domains. Although representing an easy model for 447 membrane interaction studies, liposomes do not completely mimic endosomal 448 membranes. In this sense, they lack protein content and lipid asymmetry, but more 449 importantly, they also lack the cellular microenvironment. To improve their similarity, 450 liposomes containing mixtures of zwitterionic 2-oleoyl-1-pamlitoyl-sn-glyecro-3- 451 phosphocholine (POPC) and negatively charged phospholipids have been prepared 452 (Akishiba et al., 2017; Madani et al., 2011). Among them, formulations that include the 453 anionic lipid Bis(monoacylglycerol)phosphate (BMP) are frequent as working tools (Allen 454 et al., 2019; Sudo et al., 2017; Yang et al., 2010) since this particular lipid is highly 455 enriched in the intraluminal side components of late endosomes (Kobayashi et al., 456 2001). Related to that, pH is also an important parameter to be considered when trying 457 to establish a liposome model, especially when testing EE proteins devoted to 458 discriminate endosomes (acidic micro-environment) from extracellular membranes 459 (neutral pHs). For that, liposomes are usually tested against the EE agent in a buffer at 460 the selected pH. However, some authors have also elaborated a more sophisticated 461 method where the EE molecule is encapsulated within a vesicle that is enriched with 462 bacteriorhodopsin (Madani et al., 2011), a transmembrane protein that allows the 463 control of intraluminal pH. Finally, other membrane models such as isolated endosomes 464 (Ahmad et al., 2015; Evans et al., 2013; Liu et al., 2017; Prchla et al., 1995) or 465 erythrocytes in hemolysis assay (Ahmad et al., 2015; Evans et al., 2013; Liu et al., 2017) 466 are also usually used since they show higher similarity with the cellular environment. 467 Spectrofluorimetry is one of the preferred technologies for the ex cellulo evaluation of 468 pore-forming and fusogenic EE peptides (Martens et al., 2014; Selby et al., 2017). This 469 methodology generally relies on the encapsulation of a quenched fluorophore within a 11 470 liposome, being then released upon the action of the EE promoter. Here we can 471 distinguish between the use of self-quenched tracers such as calcein (Ahmad et al., 472 2015; Oude Blenke et al., 2017; Plaza-GA et al., 2019), sulforhodamine B (Burks et al., 473 2015) or carboxyfluorescein (Kakimoto et al., 2009) from the co-encapsulation of 474 fluorophore/quencher pairs such as ANTS/DPX (8-Aminonaphthalene-1,3,6-trisulfonic 475 acid disodium salt/p-xylene-bis-pyridinium bromide) (Akishiba et al., 2017; Yang et al., 476 2010) or HPTS/DPX (8-Hydroxypyrene-1,3,6-trisulfonic acid trisodium salt/p-xylene-bis- 477 pyridinium bromide) (Yang et al., 2010). In both cases, the release of encapsulated 478 fluorophores and their subsequent dilution in the media results in a reduced quenching 479 activity and consequent increase of fluorescence, which is easily detected in a 480 spectrofluorometer. Although this methodology provides relevant data about the 481 membrane integrity of liposomes and the permeability of small molecules, it does not 482 necessarily predict their capacity to release larger molecules such as proteins (Smith et 483 al., 2019). To overcome this limitation, alternative approaches for the direct detection 484 of encapsulated proteins have also been proposed. In this regard, radiolabeled proteins 485 (van Rossenberg et al., 2002) or fluorescent-labelled proteins (encapsulated with their 486 quencher) (Madani et al., 2013) can be directly detected in the extra-liposomal media 487 by agarose gel electrophoresis or spectrofluorimetry, respectively. 488 In general, experiments with membrane models represent a simple and efficient tool for 489 the EE evaluation of fusogenic or pore-forming candidates. However, they show 490 important limitations when dealing with more complex mechanisms such as the proton 491 sponge effect, the direct translocation, or the retrograde transport, which require the 492 cellular microenvironment and intracellular trafficking machinery. Because of this, 493 several in cellulo evaluation methods have been developed. 494 3.2 In cellulo assays 495 Two important considerations are needed when designing EE evaluation methods using 496 living cells (in cellulo). First, a precise comprehension of the information that can be 497 extracted from each particular technique is required. Second, different approaches must 498 be usually combined in order to allow a precise and reliable interpretation of the 499 obtained data (Selby et al., 2017). 500 3.2.1 Fluorescence measurement: 501 Fluorescence evaluation, generally based on flow cytometry or fluorescence 502 microscopy, is the most common way to detect EE (Martens et al., 2014). In this sense, 503 flow cytometry provides high-throughput and quantitative data about the total 504 fluorescence intensity within cells. On the other hand, fluorescence microscopy and 505 especially confocal laser microscopy provide qualitative data about the intracellular 506 localization of this signal, also known as intracellular fluorescence pattern (IFP) 507 (Salomone et al., 2012). Thus, a diffuse fluorescence signal within the cytoplasm would 508 be indicative of a cytosolic distribution, while a dotted signal, would suggest an 509 entrapped fluorophore (probably into endo-lysosomes). Therefore, the combination of 510 both methodologies, flow cytometry and IFP analysis, represents a potent tool for EE 511 evaluation. 512 12 513 3.2.1.1 Independent fluorescent tracers: 514 The use of self-quenched tracers is an extended method for the evaluation of endosomal 515 membrane integrity, as seen in ex cellulo approaches. In this sense, calcein is a widely 516 used membrane-impermeable fluorophore that is loaded into endosomes by passive 517 micropinocytosis (Smith et al., 2019). Calcein is weakly fluorescent inside the 518 endosomes due to its self-quenching activity and low pH. Thus, when endosomes are 519 disrupted, the release and consequent dilution of calcein into the cytosol results in an 520 increase of fluorescence (Hu et al., 2007; Kim et al., 2018; J.-S. Kim et al., 2016; Salomone 521 et al., 2012; Wong et al., 2015) (Figure 2A). In a similar approach, acridine orange (AO), 522 forms red fluorescent dimers at high concentrations and acidic environment. Then, it 523 shifts to green fluorescence when released into the cytosol (Kandil et al., 2019) (Figure 524 2B). 525 This kind of tracers have been proven very useful in the evaluation of endosomal 526 permeability to small molecules but have shown some limitation when dealing with 527 proteins. In this regard, an alternative approach takes advantage of the lysosomal 528 degradation of DQ-ovalbumin, a high MW protein (45 KDa) conjugated with a self- 529 quenched green fluorescent tracer. Here, the self-quenched tracer becomes highly 530 fluorescent upon protein degradation in lysosomes (Quarta et al., 2017) (Figure 2C). 531 Thus, in this innovative approach, the absence of fluorescence signal directly correlates 532 with protein release. 533 3.2.1.2 Direct protein detection: 534 Not all the analytical strategies are based on the indirect detection of independent 535 tracers. Some methods directly track the intracellular fate of proteins through their 536 intrinsic fluorescence (Auger et al., 2015; Niikura et al., 2015; Sánchez-García et al., 537 2017; Sudo et al., 2017) or by labelling them with a fluorescent tracer (Ahmed et al., 538 2016; A. Kim et al., 2016; Mellert et al., 2012). Here, fluorescence microscopy is the 539 preferred tool, allowing the qualitative evaluation of IFP (Figure 2D). Some protocols, 540 however, have also tried to stablish a quantitative evaluation of IFP using microscopy. 541 For that, they measure the median fluorescence value (Mellert et al., 2012) or the 542 absolute fluorescence value at determined regions of interest (Basha et al., 2011). Going 543 still further, Lee et al. reported an innovative approach based on a fluorescence 544 resonance energy transfer (FRET) assay. For that, a green fluorescent tracer with 545 capacity to perform FRET was attached to mCherry together with its quencher molecule, 546 the last one attached by a disulfide bond. Thus, in case of EE, the disulfide bond is 547 reduced into the cytosol and the quencher molecule released. In consequence, upon 548 excitation of the green tracer, fluorescent energy is transferred to mCherry (FRET) and 549 red fluorescence is detected into the cytosol (Lee et al., 2008) (Figure 2E). 550 3.2.1.3 Co-localization studies: 551 The use of fluorescent labels, usually highly hydrophobic, have raised some concerns 552 when dealing with direct protein tracking since they can interfere with the EE process 553 (Mellert et al., 2012). Similarly, the use of acidotropic dye molecules such as 554 Lysotracker® or Lysoview®, which are widely used in organelle co-localization studies 555 (Ahmed et al., 2016; Allen et al., 2019; J.-S. Kim et al., 2016; Lieser et al., 2022; Sakurai 556 et al., 2011), might influence the acidification process of endosomes due to their weakly 13 557 basic nature (Selby et al., 2017). The use of such molecules can be avoided by the direct 558 detection of lysosomal-associated membrane proteins such as LAMP-1 by 559 immunostaining (Basha et al., 2011; J.-S. Kim et al., 2016; Wensley et al., 2020) or by its 560 genetic fusion with a fluorescent protein (Allen et al., 2019; Wong et al., 2015). The 561 lysosomal protein LAMP-2 (Huang et al., 2010) or the endosomal protein EEA1 (Huang 562 et al., 2010; Wensley et al., 2020) have been also used in this regard. Altogether, 563 although co-localization studies are mainly qualitative, some efforts have been done 564 towards turning them quantitative (Bolte and Cordelières, 2006; Vercauteren et al., 565 2011). 566 3.2.1.4 Fluorescence accumulation pattern: 567 Flow cytometry is the quantitative partner of fluorescence microscopy. Here, the 568 intracellular accumulation of fluorescent proteins, in combination with endosomal 569 maturation inhibitors such as chloroquine, has served as a lysosomal degradation 570 indicator. Thus, a larger increase of intracellular fluorescence in presence of chloroquine 571 has been correlated with a lower ability of EE (López-Laguna et al., 2020; Sánchez-García 572 et al., 2017; Serna et al., 2019). However, the use of chloroquine is still controversial 573 since its precise role in EE is unclear yet (Hajimolaali et al., 2021; Wensley et al., 2020). 574 Alternatively, other endosomal acidification blockers such as the v-type proton-pump 575 ATPase inhibitor Bafilomycin A1 have been also applied in the same sense (Martens et 576 al., 2014). 577 Recently, an innovative flow cytometry methodology based on the pulse shape analysis 578 of accumulated fluorescent signal has been proposed (Wensley et al., 2020). In this 579 approach, when the fluorescence homogenously distributes in the cytosol (endosomal 580 escape), the width of the pulse signal increases. In contrast, when the fluorescence is 581 trapped into the endosomes (dotted pattern), the height of the pulse signal increases 582 while the width decreases. This assay was found to be more sensitive than confocal 583 microscopy and represents an elegant tool for EE quantification of fluorescent proteins. 584 3.2.1.5 Protein complementation assays: 585 Protein complementation assays are also very useful for EE evaluation in fluorescent 586 proteins. In such assays, one region of the fluorescent protein is constitutively expressed 587 in the cytosol of the target cell. Then, the other complementing region is extracellularly 588 administered in form of a fusion protein that include the EE enhancer and the targeting 589 moiety for cell internalization. Thus, in case of efficient EE, the full protein is re- 590 assembled into the cytosol and restores its fluorescence (Figure 2F). In this regard, GFP- 591 complementation has been the most explored strategy (Kim et al., 2018, 2015; Lönn et 592 al., 2016). 593 3.2.2 Biological activity measurements: 594 Measuring the action of specific proteins that perform their biological activity into the 595 cytosol or other subcellular organelles can be also an indirect way to evaluate their EE 596 (Figure 3A). However, it is important to note that this process is highly dependent on 597 the intracellular trafficking of particular cell lines or types, which can somehow narrow 598 the transversal value of data interpretation (Deprey et al., 2019). Thus, the detection of 599 apoptosis-associated markers is frequently used when studying the action of proteins 14 600 with cytotoxic activity such as targeted toxins or proapoptotic proteins (Díaz et al., 2018; 601 Hetzel et al., 2008; Kiesgen et al., 2014; Wensley et al., 2020). Other approaches 602 combine EE peptides with enzymes such as α-amylase (Auger et al., 2015), β- 603 galactosidase (Maier et al., 2012), luciferase (Jiang et al., 2020) or β-lactamase (Garcia- 604 Castillo et al., 2015) that show easily measurable enzymatic activities into the cytosol. 605 Another more sophisticated method measures the cytosolic activity of Cre recombinase 606 in a cell line that has been previously transfected with a plasmid susceptible of gene 607 recombination. Thus, if EE takes place, Cre recombinase promotes the recombination 608 event, which can then be measured through the expression and activity of the encoded 609 reporter protein (Akishiba et al., 2017; Li et al., 2015; Wadia et al., 2004) . The use of 610 antigens and the subsequent analysis of the immune response by assessing the 611 production of IL-2 (Yuba et al., 2010) or MHC I presentation kinetics (Vasdekis et al., 612 2012) have also served as a proof of cytosolic delivery. 613 Finally, the EE of targeted proteins has been also evaluated by measuring the biological 614 activity of a cargo nucleic acid in the target cell. However, the need of reaching the cell 615 nucleus when transporting DNA introduces an additional barrier that significantly 616 complicates the interpretation of data resulting from this approach (Selby et al., 2017). 617 To avoid this problem, the delivery of cargo siRNAs for endogenous (e.g. syntaxin 5) 618 (Dyer et al., 2015) or reporter (e.g. luciferase) (Hatakeyama et al., 2009; Sakurai et al., 619 2011) gene knockdown have been more extensively used, as they directly perform their 620 action in the cytosol. Still, this methodology may present some important limitations 621 such as RNA degradation or intracellular dissociation during the process. 622 3.2.3 Cellular fractionation: 623 Cellular fractionation and subsequent protein localization is a very interesting approach 624 for EE evaluation. Although this methodology has mainly been explored for subcellular 625 localization of nucleic acids using qPCR (Akita et al., 2004; Hama et al., 2006), its basis 626 can also be applied to protein localization assays via fluorescence measurements or 627 Western Blot immunodetection (Lee et al., 2010; Niikura et al., 2016) (Figure 3B). The 628 major advantage of this particular approach relays in its suitability for in vivo EE 629 determination, as cellular fractionation can also be performed from tissue lysates 630 (Richardson et al., 2010). Nevertheless, the methodology is laborious and cross- 631 contamination between different subcellular fractions must be prevented during the 632 process (Lee et al., 2010). 633 3.2.4 Cytosolic enzyme modifications: 634 Several research groups have evaluated the EE of proteins by measuring specific 635 modifications generated into the cytosol by endogenous (farnesyltransferase, de- 636 ubiquitinase) (Falnes et al., 1995; Loison et al., 2005), or recombinant (β-galactosidase 637 or Biotin ligase) (Chao and Raines, 2013; Verdurmen et al., 2017) enzymes. These 638 modifications, that usually result in a change in the protein molecular weight, are easily 639 detected by SDS-PAGE of Western Blot (Figure 3C). Although these approaches are 640 versatile and compartment-specific, the addition of tags in the protein sequence may 641 potentially affect their behavior or generate artifacts due to protein degradation (Serna 642 et al., 2019). 643 15 644 3.2.5 Electron microscopy: 645 Visualization through electron microscopy has been proved useful for the localization of 646 intracellular inorganic particles, which showing high electrodensity, require no labelling 647 for detection (Martens et al., 2014). However, this methodology appears to be more 648 challenging when dealing with protein materials or reporter proteins, since their low 649 electrodensity prevents their easy distinction from other subcellular structures and 650 proteins. In this regard, the subcellular localization of specific proteins can be easily 651 detected in electron microscopy by immunogold approaches (gold-labeled protein- 652 specific antibodies) (Chanoca and Otegui, 2014; Guzmán et al., 2010; Peters et al., 2006, 653 1991)(Figure 3D). 654 Altogether, many different strategies have been developed for EE evaluation, being in 655 cellulo assays the ones providing the most accurate information about specific proteins 656 intracellular localization. However, an adequate combination of these approaches is still 657 recommended to overcome the main drawback presented in each method. In this 658 regard, different in cellulo methods have been summarized in Table 2, making emphasis 659 in their key advantages and major drawbacks. It is important to note that strategies 660 requiring protein labeling are the ones raising most important concerns regarding 661 targeted proteins, as the labeling could potentially interfere with protein targeting and 662 its subsequent intracellular trafficking. 663 664 4. Future perspectives 665 Poor or moderate EE is a main bottleneck in receptor-mediated drug delivery, especially 666 when the drug and/or the cargo are protein-based. The rapidly increasing number of 667 protein drugs, protein-based cell-targeting agents and protein materials used as carriers, 668 combined with the urgent need of receptor-targeted medicines make this limitation, 669 long time neglected, very evident. Then, while the catalogue of specific ligands or cell 670 surface receptors is increasing, the number of EE protein agents remains limited. More 671 importantly, their mechanisms of actions are diverse, controversial and in general poorly 672 understood. Therefore, the mere incorporation of any among of such agents to a protein 673 drug or carrier does not guarantee a significant increase in the cytoplasmic release of 674 active molecules. Importantly, the background cell penetrating activities of some of the 675 suspected EE peptides might interfere with the receptor-targeting agents present in the 676 protein-drug nanomedicine. Because of the relevance of specific cell-targeting, 677 especially in oncology but also in other applications of precision nanomedicines, the 678 potential loss of specificity should be a matter of attention when designing a 679 multifunctional construct with EE activities. In fact, the progressive elucidation of the EE 680 mechanics for specific peptides combined with a detailed design and tailoring of the 681 protein constructs is expected to provide highly selective drugs with relevant endosomal 682 escape abilities. 683 In this context, methodologies and approaches to finely analyze EE emerge as new and 684 extremely valuable tools, necessary for a proper functional design of drugs for 685 cytoplasmic activities. Despite the intrinsic complexity of the EE determination, since it 686 involves procedures that interfere with the cell biology, enormous advances in the field 687 are paving the way for a better comprehension of the diversity of EE mechanisms, and 16 688 for their functional and structural adaptation to specific drug delivery platforms. 689 Effective nanoscale carriers that combine high receptor targeting and efficient 690 endosomal escape are expected to be developed in the next years. This will allow a 691 quantitative and significant jump in the effectiveness and performance of protein-based 692 drugs. After 40 years of recombinant protein drug production, the design of a new 693 generation of highly efficient biopharmaceuticals should finally fulfill the current, 694 pushing and unpostponable demands of innovative medicine. 695 696 697 698 699 700 701 17 702 Table 1. EE agents with potential interest in targeted delivery approaches 703 Specificity Cis delivery Targeted delivery Origin Refs Amino acid sequence mechanism tested tested Pore-forming peptides Mellitin analogue CGIGAVLEVLTTGLPALISWIEEEEQQ Animal* CC at acidic pH No No (Ahmad et al., 2015) Bombollitin V analogue CINVLGILGLLGEALSEL Animal* CC at acidic pH No No (Ahmad et al., 2015) CLLGDFFEESEEEIGEEFEEIVQEIEDFLE LL-37 analogue Human* CC at acidic pH No No (Ahmad et al., 2015) NLVPETES LZEP CLELLEELLELLEELLELLEEL Synthetic CC at acidic pH No No (Ahmad et al., 2021) (Li et al., 2015; Lieser et Aurein 1.2 GLFDIIKKIAESF Animal Affinity for AL Yes Yes al., 2022) Fusogenic peptides (Lin et al., 2014; Liu et al., INF-7 GLFEAIEGFIENGWEGMIWDYG Synthetic CC at acidic pH Yes Yes 2017; Sung et al., 2013) GALA3 LAEALAEALEALAA Synthetic CC at acidic pH Yes No (Li et al., 2020) (Fuchs et al., 2007; Oess Pre-S2 domain PLSSIFSRIGDP Viral Unclear Yes Yes and Hildt, 2000) CC at acidic pH + DEN MVDRGWGNGCGLFGKGGIV Viral Yes Yes (Kiesgen et al., 2014) affinity for AL B18 PYDLGLLLRHLRHHSNLLANI Animal Unclear Yes Yes (Niikura et al., 2016, 2015) Affinity for LE S19 PFVIGAGVLGALGTGIGGI Human Yes No (Sudo et al., 2017) membranes Histidine-rich peptides HHHHHH Protonation at Polyhistidine tags HHHHHHHH Synthetic Yes Yes (Dhankher et al., 2021) acidic pH HHHHHHHHHH HAAHAH Protonation at Humanized histidine tags HTHTHTHTH Human* Yes Yes (López-Laguna et al., 2020) acidic pH HEHEHEHEH 1 Protonation at ppTG21 GLFHALLHLLHSLWHLLLHA Synthetic No Yes (Rouet and Christ, 2019) acidic pH Toxins-derived peptides GNRVRRSVGSSLSCINLDWDVIRDKTK TKIESLKEHGPIKNKMSESPNKTVSEEK AKQYLEEFHQTALEHPELSELKTVTGT NPVFAGANYAAWAVNVAQVIDSETA Diphtheria toxin T domain DNLEKTTAALSILPGIGSVMGIADGAV Bacterial CC at acidic pH Yes Yes (Gilyazova et al., 2006) HHNTEEIVAQSIALSSLMVAQAIPLVG ELVDIGFAAYNFVESIINLFQVVHNSYN RPAYSPGHKT GQDAMYEYMAQACAGNRVRRSVGS SLSCINLDWDVIRDKTKTKIESLKEHGPI KNKMSESPNKTVSEEKAKQYLEEFHQT ALEHPELSELKTVTGTNPVFAGANYAA Diphtheria toxin C domain WAVNVAQVIDSETADNLEKTTAALSIL Bacterial CC at acidic pH Yes No (Aullo et al., 1993) C-ter region + T domain PGIGSVMGIADGAVHHNTEEIVAQSIA LSSLMVAQAIPLVGELVDIGFAAYNFV ESIINLFQVVHNSYNRPAYSPGHKT GGSLAALTAHQACHLPLETFTRHRQP RGWEQLEQCGYPVQRLVALYLAARLS WNQVDQVIRNALASPGSGGDLGEAIR EQPEQARLALTLAAAESERFVRQGTG (Mohammed et al., 2012; Exotoxin A T domain Bacterial ERAD pathway Yes Yes NDEAGAANADVVSLTCPVAAGECAG Schmit et al., 2019) PADSGDALLERNYPTGAEFLGDGGDV SFSTR GNRVRRSV + cargo + C-terminal Exotoxin A FCS + KDEL Bacterial ERAD pathway Yes Yes (Díaz et al., 2018) KDEL 2 704 705 * Modified from the original sequence 706 CC: Conformational Change. 707 LE: Late Endosomes 708 AL: Anionic Lipids 709 Furin cleavable sites (FCS) have been highlighted in bold 3 710 Table 2. In celullo methods for EE evaluation Category Subcategory Technique Key advantages Major drawbacks Ref Possibility of false positives due to membrane Easy (Kim et al., 2018; J.-S. Kim et al., 2016; Calcein leakage assay destabilization Salomone et al., 2012; Wong et al., 2015) pH-specific Not useful for translocations Easy Tracer may affect endosomal maturation process Possibility of false positives due to membrane Acridine orange assay (Kandil et al., 2019) Require the pH-specific destabilization use of tracers Not useful for translocations Measures degradation Labelling requirement DQ-ovalbumin assay (Quarta et al., 2017) Measures protein escape Not useful for translocations pH-specific Tracers may affect endosomal maturation process (Ahmed et al., 2016; Allen et al., 2019; J.-S. Co-localization studies High-levels of background Kim et al., 2016; Sakurai et al., 2011; Selby with tracers Useful for all mechanisms Difficult to quantify et al., 2017) Fluorescence measurements Co-localization studies pH-specific Requires stable cell line expressing exogenous gene (Allen et al., 2019; Huang et al., 2010; J.-S. Detection by in stable cell lines Useful for all mechanisms Difficult to quantify Kim et al., 2016; Wensley et al., 2020) fluorescence Measures protein escape Double labelling requirement microscopy Possibility of false positives due to spontaneous Cytosolic specific (mostly Cytosolic specific FRET reduction of S-S bond (Lee et al., 2008) qualitative) Possibility of false negatives due to spontaneous Useful for all mechanisms and/or flow unlabelling cytometry Requires the Possibility of false negatives due to fluorophore (mostly use of Easy Fluorescence dilution (López-Laguna et al., 2020; Sánchez- quantitative) inherent accumulation pattern García et al., 2017; Serna et al., 2019) fluorescent Useful for all mechanisms Controversy in the use of some agents proteins or fluorophore- Easy More data is required conjugated Useful for all mechanisms proteins Do not rely on the use of any additional Pulse-shape analysis (Wensley et al., 2020) compound Endosomal distribution in used cell line affects the Higher sensibility than confocal sensiblity microscopy Useful for all mechanisms Requires stable cell line expressing exogenous gene Complementation assay Do not rely on the use of any additional (Kim et al., 2018, 2015; Lönn et al., 2016) compound Subunits must find each other in the cytosol No labelling Easy Possibility of false negatives due to subsequent Applicable to therapeutic candidates Biological Measure the protein trafficking Useful for all mechanisms (Díaz et al., 2018; Hetzel et al., 2008; activity activity of Cell viability assay No labelling Kiesgen et al., 2014; Wensley et al., 2020) measurements proteins Possibility of false negatives due to the effect of High sensibility endosomal escape promoter to the protein Compartment-specific 1 Possibility of false negatives due to subsequent Applicable to therapeutic candidates protein trafficking (Auger et al., 2015; Garcia-Castillo et al., Enzymatic assays Useful for all mechanisms 2015; Jiang et al., 2020; Maier et al., 2012) Possibility of false negatives due to the effect of No labelling endosomal escape promoter to the protein Compartment-specific Applicable to therapeutic candidates Useful for all mechanisms Possibility of false negatives due to subsequent Immunologic assays (Vasdekis et al., 2012; Yuba et al., 2010) No labelling protein trafficking Compartment-specific Possibility of false negatives due to subsequent cargo Applicable to therapeutic candidates Measure the trafficking Gene silenciation (Dyer et al., 2015; Hatakeyama et al., 2009; activity of Cytosolic specific RNA labelling requirement assays Sakurai et al., 2011) nucleic acids For silencing exogenous genes, it requires stable cell Useful for all mechanisms line expressing exogenous gene Fluorescence Suitable for in vivo assays Labor-intensiveness Measure measurements or No labelling Risk of contamination between subcellular locations (Lee et al., 2010; Niikura et al., 2016) protein Western Blot Useful for all mechanisms Macromolecules may alter vesicle density Cellular Labor-intensiveness fractionation Suitable for in vivo assays Measure Risk of contamination between subcellular locations qPCR (Akita et al., 2004; Hama et al., 2006) nucleic acids Macromolecules may alter vesicle density Useful for all mechanisms Labelling requirement Require Cytosolic specific Requires stable cell line expressing exogenous gene expression of Beta-galactosidase and (Chao and Raines, 2013; Verdurmen et al., exogenous biotin ligase assay Useful for all mechanisms Labelling requirement 2017) Cytosolic enzymes enzyme Do not Cytosolic specific Requires subsequent cellular treatment modifications require the Farnseylation assay CaaX motif C-ter might promote protein cytosolic (Falnes et al., 1995) Useful for all mechanisms expression of degradation exogenous Cytosolic specific Requires subsequent cellular treatment De-ubiquitination assay (Loison et al., 2005) enzymes Useful for all mechanisms Labelling requirement Compartment-specific Electron Cell fixation requirement (Chanoca and Otegui, 2014; Guzmán et al., Immuno-TEM Direct visualization microscopy Labelling requirement 2010; Peters et al., 2006, 1991) Useful for all mechanisms 711 2 712 Figure Captions: 713 Figure 1. Endosomal escape mechanisms. A. Pore-forming peptides can create channels in 714 endosomes membranes following the barrel-stave (i), carpet (ii) or toroidal (iii) models. B. Viral 715 fusogenic peptides (in blue) drive the fusion between viral and endosomal membranes due to a 716 pH-dependent conformation change that induce their insertion into the endosomal membranes. 717 C. Histidine-rich peptides in proteins (in red) become protonated at acidic pH. This fact promotes 718 the entrance of protons (in yellow) in the endosomal lumen, which is followed by an entrance 719 of chloride ions (in green) to compensate charges. The consequent increase in the osmotic 720 pressure ends up in endosome lysis and in the release of its content to the cytosol (proton 721 sponge effect). D. Direct protein translocation from the endosome lumen to the cytosol 722 mediated by a toxin translocation domain (in yellow). A pH-mediated conformational 723 rearrangement of the translocation domain initiates this process at late-endosomes. E. 724 Lysosomal bypassing of proteins (in blue) following the ERAD recycling pathway. Specific toxin 725 translocation domains drive the cargo protein to the endoplasmic reticulum, by which they are 726 retrogradely transported to the cytosol. E: Endosome, L: Lysosome, ER: Endoplasmatic 727 Reticulum. 728 729 Figure 2. Fluorescence based methods: A. Endosomal escape evaluation via self-quenched 730 fluorescent tracers such as calcein. Endosomal disruption dilutes the fluorophore into the 731 cytoplasm ending its quenching and triggering an increase in green fluorescence. B. Acridine 732 orange is used similarly to calcein to detect endosomal escape. However, in acidic endosomes it 733 assumes the form of red-fluorescent dimers, allowing a clear differentiation of endosomal 734 entrapment or escape situations. C. Use of DQ-ovoalbumin, a self-quenched fluorogenic 735 substrate for proteases that emits a strong fluorescent signal upon degradation to small labelled 736 peptides inside the lysosome. D. Use of labelled or intrinsically fluorescent proteins to directly 737 evaluate their endosomal fate via fluorescence microscopy. E. Use of a quenched FRET- 738 dependent fluorescent protein that emits red fluorescence upon release into the cytosol from 739 the endosomes. Reduction of the disulfide-bound quencher allows the green fluorescent moiety 740 to excite the red fluorophore. F. Protein complementation assay where the endosomal escape 741 peptide is fused to the complementing domain of a cytosol-expressed truncated fluorescent 742 protein. When endosomal escape occurs, the truncated protein in the cytosol is complimented, 743 restoring its fluorescence emission. 744 745 Figure 3. Other analytical methods: A. Overview of the main detection systems to evaluate 746 endosomal escape based on biological functions or the alteration of these. B. Subcellular 747 fractionation and subsequent antigenic or fluorescence detection in order to elucidate the fate 748 of a given protein candidate with an endosomal escape domain. C. Endosomal escape is 749 evaluated via determination of the candidate protein molecular weight. The studied protein is 750 susceptible to cytosolic modifications that can increase or decrease its mass. D. Use of electron 751 microscopy combined with gold immunolabeling to explore the subcellular location of the 752 protein of interest. 753 754 755 1 756 Acknowledgements: 757 The authors are indebted to Agencia Española de Investigación for granting projects on 758 the construction of protein materials of clinical interest (PID2019-105416RB- 759 I00/AEI/10.13039/501100011033 to EV and PID2020-116174RB-I00 to AV), to AGAUR 760 for project 2017 SGR- 229 to AV, and to ISCIII for project PI20/00400 to UU co-funded 761 by European Regional Development Fund (ERDF, “a way to make Europe”). We also 762 appreciate the funding from the CIBER -Consorcio Centro de Investigación Biomédica en 763 Red- (CB06/01/0014), Instituto de Salud Carlos III, Ministerio de Ciencia e Innovación, 764 through projects (Nanoremote, Nanopills, Venom4Cancer, Nanoprother II, Nanoscape 765 and Nanolink). U.U. was supported by a Miguel Servet contract (CP19/00028) from ISCIII 766 co-funded by European Social Fund (ESF investing in your future). E.V-D was supported 767 by a predoctoral fellowship from Ministerio de Ciencia, Innovación y Universidades 768 (FPU18/04615). A.V. received an ICREA ACADEMIA award. 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 2 784 References: 785 Ahmad, A., Khan, J.M., Haque, S., 2019. Strategies in the design of endosomolytic agents for 786 facilitating endosomal escape in nanoparticles. Biochimie 160, 61–75. 787 https://doi.org/10.1016/j.biochi.2019.02.012 788 Ahmad, A., Ranjan, S., Zhang, W., Zou, J., Pyykkö, I., Kinnunen, P.K.J., 2015. Novel 789 endosomolytic peptides for enhancing gene delivery in nanoparticles. Biochimica et 790 Biophysica Acta (BBA) - Biomembranes 1848, 544–553. 791 https://doi.org/10.1016/j.bbamem.2014.11.008 792 Ahmad, A., Rilla, K., Zou, J., Zhang, W., Pyykkö, I., Kinnunen, P., Ranjan, S., 2021. Enhanced 793 gene expression by a novel designed leucine zipper endosomolytic peptide. Int J Pharm 794 601, 120556. https://doi.org/10.1016/j.ijpharm.2021.120556 795 Ahmed, S., Fujita, S., Matsumura, K., 2016. Enhanced protein internalization and efficient 796 endosomal escape using polyampholyte-modified liposomes and freeze concentration. 797 Nanoscale 8, 15888–15901. https://doi.org/10.1039/C6NR03940E 798 Akishiba, M., Takeuchi, T., Kawaguchi, Y., Sakamoto, K., Yu, H.-H., Nakase, I., Takatani-Nakase, 799 T., Madani, F., Gräslund, A., Futaki, S., 2017. Cytosolic antibody delivery by lipid-sensitive 800 endosomolytic peptide. Nat Chem 9, 751–761. https://doi.org/10.1038/nchem.2779 801 Akita, H., Ito, R., Khalil, I.A., Futaki, S., Harashima, H., 2004. Quantitative Three-Dimensional 802 Analysis of the Intracellular Trafficking of Plasmid DNA Transfected by a Nonviral Gene 803 Delivery System Using Confocal Laser Scanning Microscopy. Molecular Therapy 9, 443– 804 451. https://doi.org/10.1016/j.ymthe.2004.01.005 805 Allen, J., Brock, D., Kondow-McConaghy, H., Pellois, J.-P., 2018. Efficient Delivery of 806 Macromolecules into Human Cells by Improving the Endosomal Escape Activity of Cell- 807 Penetrating Peptides: Lessons Learned from dfTAT and its Analogs. Biomolecules 8, 50. 808 https://doi.org/10.3390/biom8030050 809 Allen, J., Najjar, K., Erazo-Oliveras, A., Kondow-McConaghy, H.M., Brock, D.J., Graham, K., 810 Hager, E.C., Marschall, A.L.J., Dübel, S., Juliano, R.L., Pellois, J.-P., 2019. Cytosolic Delivery 811 of Macromolecules in Live Human Cells Using the Combined Endosomal Escape Activities 812 of a Small Molecule and Cell Penetrating Peptides. ACS Chem Biol 14, 2641–2651. 813 https://doi.org/10.1021/acschembio.9b00585 814 Auger, A., Park, M., Nitschke, F., Minassian, L.M., Beilhartz, G.L., Minassian, B.A., Melnyk, R.A., 815 2015. Efficient Delivery of Structurally Diverse Protein Cargo into Mammalian Cells by a 816 Bacterial Toxin. Mol Pharm 12, 2962–2971. 817 https://doi.org/10.1021/acs.molpharmaceut.5b00233 818 Aullo, P., Giry, M., Olsnes, S., Popoff, M.R., Kocks, C., Boquet, P., 1993. A chimeric toxin to 819 study the role of the 21 kDa GTP binding protein rho in the control of actin microfilament 820 assembly. EMBO J 12, 921–931. https://doi.org/10.1002/j.1460-2075.1993.tb05733.x 821 Bachran, C., Morley, T., Abdelazim, S., Fattah, R.J., Liu, S., Leppla, S.H., 2013. Anthrax Toxin- 822 Mediated Delivery of the Pseudomonas Exotoxin A Enzymatic Domain to the Cytosol of 823 Tumor Cells via Cleavable Ubiquitin Fusions. mBio 4. 824 https://doi.org/10.1128/mBio.00201-13 3 825 BAREFORD, L., SWAAN, P., 2007. Endocytic mechanisms for targeted drug delivery☆. Adv Drug 826 Deliv Rev 59, 748–758. https://doi.org/10.1016/j.addr.2007.06.008 827 Basha, G., Novobrantseva, T.I., Rosin, N., Tam, Y.Y.C., Hafez, I.M., Wong, M.K., Sugo, T., Ruda, 828 V.M., Qin, J., Klebanov, B., Ciufolini, M., Akinc, A., Tam, Y.K., Hope, M.J., Cullis, P.R., 2011. 829 Influence of Cationic Lipid Composition on Gene Silencing Properties of Lipid 830 Nanoparticle Formulations of siRNA in Antigen-Presenting Cells. Molecular Therapy 19, 831 2186–2200. https://doi.org/10.1038/mt.2011.190 832 Bazak, R., Houri, M., el Achy, S., Kamel, S., Refaat, T., 2015. Cancer active targeting by 833 nanoparticles: a comprehensive review of literature. J Cancer Res Clin Oncol 141, 769– 834 784. https://doi.org/10.1007/s00432-014-1767-3 835 BOLTE, S., CORDELIÈRES, F.P., 2006. A guided tour into subcellular colocalization analysis in 836 light microscopy. J Microsc 224, 213–232. https://doi.org/10.1111/j.1365- 837 2818.2006.01706.x 838 Browne, K.A., Blink, E., Sutton, V.R., Froelich, C.J., Jans, D.A., Trapani, J.A., 1999. Cytosolic 839 Delivery of Granzyme B by Bacterial Toxins: Evidence that Endosomal Disruption, in 840 Addition to Transmembrane Pore Formation, Is an Important Function of Perforin. Mol 841 Cell Biol 19, 8604–8615. https://doi.org/10.1128/MCB.19.12.8604 842 Burks, S.R., Legenzov, E.A., Martin, E.W., Li, C., Lu, W., Kao, J.P.Y., 2015. Co-Encapsulating the 843 Fusogenic Peptide INF7 and Molecular Imaging Probes in Liposomes Increases 844 Intracellular Signal and Probe Retention. PLoS One 10, e0120982. 845 https://doi.org/10.1371/journal.pone.0120982 846 Casanova, I., Unzueta, U., Arroyo-Solera, I., Céspedes, M.V., Villaverde, A., Mangues, R., 847 Vazquez, E., 2019. Protein-driven nanomedicines in oncotherapy. Curr Opin Pharmacol 848 47, 1–7. https://doi.org/10.1016/j.coph.2018.12.004 849 Chanoca, A., Otegui, M.S., 2014. Immunogold Labeling and Electron Tomography of Plant 850 Endosomes. pp. 63–80. https://doi.org/10.1007/978-1-4939-1420-3_5 851 Chao, T.-Y., Raines, R.T., 2013. Fluorogenic label to quantify the cytosolic delivery of 852 macromolecules. Mol Biosyst 9, 339. https://doi.org/10.1039/c3mb25552b 853 Corchero, J.L., Vázquez, E., García-Fruitós, E., Ferrer-Miralles, N., Villaverde, A., 2014. 854 Recombinant protein materials for bioengineering and nanomedicine. Nanomedicine 9, 855 2817–2828. https://doi.org/10.2217/nnm.14.153 856 Deprey, K., Becker, L., Kritzer, J., Plückthun, A., 2019. Trapped! A Critical Evaluation of Methods 857 for Measuring Total Cellular Uptake versus Cytosolic Localization. Bioconjug Chem 30, 858 1006–1027. https://doi.org/10.1021/acs.bioconjchem.9b00112 859 Dhankher, A., Lv, W., Studstill, W.T., Champion, J.A., 2021. Coiled coil exposure and histidine 860 tags drive function of an intracellular protein drug carrier. Journal of Controlled Release 861 339, 248–258. https://doi.org/10.1016/j.jconrel.2021.09.026 862 Díaz, R., Pallarès, V., Cano-Garrido, O., Serna, N., Sánchez-García, L., Falgàs, A., Pesarrodona, 863 M., Unzueta, U., Sánchez-Chardi, A., Sánchez, J.M., Casanova, I., Vázquez, E., Mangues, R., 864 Villaverde, A., 2018. Selective CXCR4 Cancer Cell Targeting and Potent Antineoplastic 4 865 Effect by a Nanostructured Version of Recombinant Ricin. Small 14, 1800665. 866 https://doi.org/10.1002/smll.201800665 867 Duncan, R., Gaspar, R., 2011. Nanomedicine(s) under the Microscope. Mol Pharm 8, 2101– 868 2141. https://doi.org/10.1021/mp200394t 869 Dyer, P.D.R., Shepherd, T.R., Gollings, A.S., Shorter, S.A., Gorringe-Pattrick, M.A.M., Tang, C.-K., 870 Cattoz, B.N., Baillie, L., Griffiths, P.C., Richardson, S.C.W., 2015. Disarmed anthrax toxin 871 delivers antisense oligonucleotides and siRNA with high efficiency and low toxicity. 872 Journal of Controlled Release 220, 316–328. 873 https://doi.org/10.1016/j.jconrel.2015.10.054 874 Erazo-Oliveras, A., Najjar, K., Dayani, L., Wang, T.-Y., Johnson, G.A., Pellois, J.-P., 2014. Protein 875 delivery into live cells by incubation with an endosomolytic agent. Nat Methods 11, 861– 876 867. https://doi.org/10.1038/nmeth.2998 877 Erazo-Oliveras, A., Najjar, K., Truong, D., Wang, T.-Y., Brock, D.J., Prater, A.R., Pellois, J.-P., 878 2016. The Late Endosome and Its Lipid BMP Act as Gateways for Efficient Cytosolic Access 879 of the Delivery Agent dfTAT and Its Macromolecular Cargos. Cell Chem Biol 23, 598–607. 880 https://doi.org/10.1016/j.chembiol.2016.03.016 881 Erickson, H.A., Jund, M.D., Pennell, C.A., 2006. Cytotoxicity of human RNase-based 882 immunotoxins requires cytosolic access and resistance to ribonuclease inhibition. Protein 883 Engineering, Design and Selection 19, 37–45. https://doi.org/10.1093/protein/gzi073 884 Evans, B.C., Nelson, C.E., Yu, S.S., Beavers, K.R., Kim, A.J., Li, H., Nelson, H.M., Giorgio, T.D., 885 Duvall, C.L., 2013. Ex Vivo Red Blood Cell Hemolysis Assay for the Evaluation of pH- 886 responsive Endosomolytic Agents for Cytosolic Delivery of Biomacromolecular Drugs. 887 Journal of Visualized Experiments. https://doi.org/10.3791/50166 888 Falnes, P., Wiedlocha, A., Rapak, A., Olsnes, S., 1995. Farnesylation of CaaX-Tagged Diphtheria 889 Toxin A-Fragment as a Measure of Transfer to the Cytosol. Biochemistry 34, 11152– 890 11159. https://doi.org/10.1021/bi00035a021 891 Fernandez, D.I., le Brun, A.P., Whitwell, T.C., Sani, M.-A., James, M., Separovic, F., 2012. The 892 antimicrobial peptide aurein 1.2 disrupts model membranes via the carpet mechanism. 893 Physical Chemistry Chemical Physics 14, 15739. https://doi.org/10.1039/c2cp43099a 894 Ferrer-Miralles, N., Rodríguez-Carmona, E., Corchero, J.L., García-Fruitós, E., Vázquez, E., 895 Villaverde, A., 2015. Engineering protein self-assembling in protein-based nanomedicines 896 for drug delivery and gene therapy. Crit Rev Biotechnol 35, 209–221. 897 https://doi.org/10.3109/07388551.2013.833163 898 Fotoohi-Ardakani, G., Kheirollahi, M., Zarei Jaliani, H., Noorian, M., Ansariniyia, H., 2019. 899 Targeting MCF-7 Cell Line by Listeriolysin O Pore Forming Toxin Fusion with AHNP 900 Targeted Peptide. Adv Biomed Res 8, 33. https://doi.org/10.4103/abr.abr_18_19 901 Friebe, S., van der Goot, F., Bürgi, J., 2016. The Ins and Outs of Anthrax Toxin. Toxins (Basel) 8, 902 69. https://doi.org/10.3390/toxins8030069 903 Fuchs, H., Bachran, C., Li, T., Heisler, I., Dürkop, H., Sutherland, M., 2007. A cleavable molecular 904 adapter reduces side effects and concomitantly enhances efficacy in tumor treatment by 5 905 targeted toxins in mice. Journal of Controlled Release 117, 342–350. 906 https://doi.org/10.1016/j.jconrel.2006.11.019 907 Garcia-Castillo, M.D., Tran, T., Bobard, A., Renard, H.-F., Rathjen, S.J., Dransart, E., Stechmann, 908 B., Lamaze, C., Lord, M., Cintrat, J.-C., Enninga, J., Tartour, E., Johannes, L., 2015. 909 Retrograde transport is not required for cytosolic translocation of the B-subunit of Shiga 910 toxin. J Cell Sci 128, 2373–2387. https://doi.org/10.1242/jcs.169383 911 Gilyazova, D.G., Rosenkranz, A.A., Gulak, P. v., Lunin, V.G., Sergienko, O. v., Khramtsov, Y. v., 912 Timofeyev, K.N., Grin, M.A., Mironov, A.F., Rubin, A.B., Georgiev, G.P., Sobolev, A.S., 913 2006. Targeting Cancer Cells by Novel Engineered Modular Transporters. Cancer Res 66, 914 10534–10540. https://doi.org/10.1158/0008-5472.CAN-06-2393 915 Guzmán, F.A., Arneodo, J.D., Saavedra Pons, A.B., Truol, G.A., Luque, A. v., Conci, L.R., 2010. 916 Immunodetection and subcellular localization of Mal de Río Cuarto virus P9-1 protein in 917 infected plant and insect host cells. Virus Genes 41, 111–117. 918 https://doi.org/10.1007/s11262-010-0480-9 919 Hajimolaali, M., Mohammadian, H., Torabi, A., Shirini, A., Khalife Shal, M., Barazandeh Nezhad, 920 H., Iranpour, S., Baradaran Eftekhari, R., Dorkoosh, F., 2021. Application of chloroquine as 921 an endosomal escape enhancing agent: new frontiers for an old drug. Expert Opin Drug 922 Deliv 18, 877–889. https://doi.org/10.1080/17425247.2021.1873272 923 Hama, S., Akita, H., Ito, R., Mizuguchi, H., Hayakawa, T., Harashima, H., 2006. Quantitative 924 Comparison of Intracellular Trafficking and Nuclear Transcription between Adenoviral 925 and Lipoplex Systems. Molecular Therapy 13, 786–794. 926 https://doi.org/10.1016/j.ymthe.2005.10.007 927 Hatakeyama, H., Ito, E., Akita, H., Oishi, M., Nagasaki, Y., Futaki, S., Harashima, H., 2009. A pH- 928 sensitive fusogenic peptide facilitates endosomal escape and greatly enhances the gene 929 silencing of siRNA-containing nanoparticles in vitro and in vivo. Journal of Controlled 930 Release 139, 127–132. https://doi.org/10.1016/j.jconrel.2009.06.008 931 He, J., Xu, S., Mixson, A.J., 2020. The Multifaceted Histidine-Based Carriers for Nucleic Acid 932 Delivery: Advances and Challenges. Pharmaceutics 12, 774. 933 https://doi.org/10.3390/pharmaceutics12080774 934 Hetzel, C., Bachran, C., Fischer, R., Fuchs, H., Barth, S., Stöcker, M., 2008. Small Cleavable 935 Adapters Enhance the Specific Cytotoxicity of a Humanized Immunotoxin Directed 936 Against CD64-positive Cells. Journal of Immunotherapy 31, 370–376. 937 https://doi.org/10.1097/CJI.0b013e31816a2d23 938 Hu, Y., Litwin, T., Nagaraja, A.R., Kwong, B., Katz, J., Watson, N., Irvine, D.J., 2007. Cytosolic 939 Delivery of Membrane-Impermeable Molecules in Dendritic Cells Using pH-Responsive 940 Core−Shell Nanoparticles. Nano Lett 7, 3056–3064. https://doi.org/10.1021/nl071542i 941 Huang, C.Y.-H., Butrapet, S., Moss, K.J., Childers, T., Erb, S.M., Calvert, A.E., Silengo, S.J., 942 Kinney, R.M., Blair, C.D., Roehrig, J.T., 2010. The dengue virus type 2 envelope protein 943 fusion peptide is essential for membrane fusion. Virology 396, 305–315. 944 https://doi.org/10.1016/j.virol.2009.10.027 945 Huotari, J., Helenius, A., 2011. Endosome maturation. EMBO J 30, 3481–3500. 946 https://doi.org/10.1038/emboj.2011.286 6 947 Inoue, T., Moore, P., Tsai, B., 2011. How Viruses and Toxins Disassemble to Enter Host Cells. 948 Annu Rev Microbiol 65, 287–305. https://doi.org/10.1146/annurev-micro-090110- 949 102855 950 Jiang, Y., Lu, Q., Wang, Yongheng, Xu, E., Ho, A., Singh, P., Wang, Yifei, Jiang, Z., Yang, F., 951 Tietjen, G.T., Cresswell, P., Saltzman, W.M., 2020. Quantitating Endosomal Escape of a 952 Library of Polymers for mRNA Delivery. Nano Lett 20, 1117–1123. 953 https://doi.org/10.1021/acs.nanolett.9b04426 954 Jiang, Z., Guan, J., Qian, J., Zhan, C., 2019. Peptide ligand-mediated targeted drug delivery of 955 nanomedicines. Biomater Sci 7, 461–471. https://doi.org/10.1039/C8BM01340C 956 Kakimoto, S., Hamada, T., Komatsu, Y., Takagi, M., Tanabe, T., Azuma, H., Shinkai, S., Nagasaki, 957 T., 2009. The conjugation of diphtheria toxin T domain to poly(ethylenimine) based 958 vectors for enhanced endosomal escape during gene transfection. Biomaterials 30, 402– 959 408. https://doi.org/10.1016/j.biomaterials.2008.09.042 960 Kandil, R., Xie, Y., Heermann, R., Isert, L., Jung, K., Mehta, A., Merkel, O.M., 2019. Coming in 961 and Finding Out: Blending Receptor‐Targeted Delivery and Efficient Endosomal Escape in 962 a Novel Bio‐Responsive siRNA Delivery System for Gene Knockdown in Pulmonary T Cells. 963 Adv Ther (Weinh) 1900047. https://doi.org/10.1002/adtp.201900047 964 Kato, T., Yamashita, H., Misawa, T., Nishida, K., Kurihara, M., Tanaka, M., Demizu, Y., Oba, M., 965 2016. Plasmid DNA delivery by arginine-rich cell-penetrating peptides containing 966 unnatural amino acids. Bioorg Med Chem 24, 2681–2687. 967 https://doi.org/10.1016/j.bmc.2016.04.031 968 Kheirandish, M.H., Jaliani, H.Z., Rahmani, B., Nikukar, H., 2019. Specific targeting of a pore- 969 forming toxin (listeriolysin O) to LHRH-positive cancer cells using LHRH targeting peptide. 970 Toxicon 164, 82–86. https://doi.org/10.1016/j.toxicon.2019.04.005 971 Kiesgen, S., Liebers, N., Cremer, M., Arnold, U., Weber, T., Keller, A., Herold-Mende, C., 972 Dyckhoff, G., Jager, D., Kontermann, R.E., Arndt, M.A.E., Krauss, J., 2014. A fusogenic 973 dengue virus-derived peptide enhances antitumor efficacy of an antibody-ribonuclease 974 fusion protein targeting the EGF receptor. Protein Engineering Design and Selection 27, 975 331–338. https://doi.org/10.1093/protein/gzu040 976 Kim, A., Miura, Y., Ishii, T., Mutaf, O.F., Nishiyama, N., Cabral, H., Kataoka, K., 2016. 977 Intracellular Delivery of Charge-Converted Monoclonal Antibodies by Combinatorial 978 Design of Block/Homo Polyion Complex Micelles. Biomacromolecules 17, 446–453. 979 https://doi.org/10.1021/acs.biomac.5b01335 980 Kim, J., Choi, D.-K., Park, S., Shin, S.-M., Bae, J., Kim, D.-M., Yoo, T.H., Kim, Y.-S., 2015. 981 Quantitative assessment of cellular uptake and cytosolic access of antibody in living cells 982 by an enhanced split GFP complementation assay. Biochem Biophys Res Commun 467, 983 771–777. https://doi.org/10.1016/j.bbrc.2015.10.066 984 Kim, J., Park, J.-Y., Shin, S.-M., Park, S.-W., Jun, S.-Y., Hong, J.-S., Choi, D.-K., Kim, Y.-S., 2018. 985 Engineering of a tumor cell–specific, cytosol-penetrating antibody with high endosomal 986 escape efficacy. Biochem Biophys Res Commun 503, 2510–2516. 987 https://doi.org/10.1016/j.bbrc.2018.07.008 7 988 Kim, J.-S., Choi, D.-K., Shin, J.-Y., Shin, S.-M., Park, S.-W., Cho, H.-S., Kim, Y.-S., 2016. Endosomal 989 acidic pH-induced conformational changes of a cytosol-penetrating antibody mediate 990 endosomal escape. Journal of Controlled Release 235, 165–175. 991 https://doi.org/10.1016/j.jconrel.2016.05.066 992 Kobayashi, T., Startchev, K., Whitney, A.J., Gruenberg, J., 2001. Localization of 993 Lysobisphosphatidic Acid-Rich Membrane Domains in Late Endosomes. Biol Chem 382. 994 https://doi.org/10.1515/BC.2001.059 995 Ladokhin, A.S., Kyrychenko, A., Rodnin, M. v., Vasquez-Montes, V., 2021. Conformational 996 switching, refolding and membrane insertion of the diphtheria toxin translocation 997 domain. pp. 341–370. https://doi.org/10.1016/bs.mie.2020.12.016 998 LeCher, J.C., Nowak, S.J., McMurry, J.L., 2017. Breaking in and busting out: cell-penetrating 999 peptides and the endosomal escape problem. Biomol Concepts 8, 131–141. 1000 https://doi.org/10.1515/bmc-2017-0023 1001 Lee, Y.H., Tan, H.T., Chung, M.C.M., 2010. Subcellular fractionation methods and strategies for 1002 proteomics. Proteomics 10, 3935–3956. https://doi.org/10.1002/pmic.201000289 1003 Lee, Y.-J., Datta, S., Pellois, J.-P., 2008. Real-Time Fluorescence Detection of Protein 1004 Transduction into Live Cells. J Am Chem Soc 130, 2398–2399. 1005 https://doi.org/10.1021/ja7102026 1006 Lee, Y.-J., Johnson, G., Peltier, G.C., Pellois, J.-P., 2011. A HA2-Fusion tag limits the endosomal 1007 release of its protein cargo despite causing endosomal lysis. Biochimica et Biophysica 1008 Acta (BBA) - General Subjects 1810, 752–758. 1009 https://doi.org/10.1016/j.bbagen.2011.05.013 1010 Lee, Y.-W., Luther, D.C., Kretzmann, J.A., Burden, A., Jeon, T., Zhai, S., Rotello, V.M., 2019. 1011 Protein Delivery into the Cell Cytosol using Non-Viral Nanocarriers. Theranostics 9, 3280– 1012 3292. https://doi.org/10.7150/thno.34412 1013 Li, C., Cao, X.-W., Zhao, J., Wang, F.-J., 2020. Effective Therapeutic Drug Delivery by GALA3, an 1014 Endosomal Escape Peptide with Reduced Hydrophobicity. J Membr Biol 253, 139–152. 1015 https://doi.org/10.1007/s00232-020-00109-2 1016 Li, M., Tao, Y., Shu, Y., LaRochelle, J.R., Steinauer, A., Thompson, D., Schepartz, A., Chen, Z.-Y., 1017 Liu, D.R., 2015. Discovery and Characterization of a Peptide That Enhances Endosomal 1018 Escape of Delivered Proteins in Vitro and in Vivo. J Am Chem Soc 137, 14084–14093. 1019 https://doi.org/10.1021/jacs.5b05694 1020 Liao, X., Rabideau, A.E., Pentelute, B.L., 2014. Delivery of Antibody Mimics into Mammalian 1021 Cells via Anthrax Toxin Protective Antigen. ChemBioChem 15, 2458–2466. 1022 https://doi.org/10.1002/cbic.201402290 1023 Lieser, R.M., Li, Q., Chen, W., Sullivan, M.O., 2022. Incorporation of Endosomolytic Peptides 1024 with Varying Disruption Mechanisms into EGFR-Targeted Protein Conjugates: The Effect 1025 on Intracellular Protein Delivery and EGFR Specificity in Breast Cancer Cells. Mol Pharm 1026 19, 661–673. https://doi.org/10.1021/acs.molpharmaceut.1c00788 8 1027 Lin, N., Zheng, W., Li, L., Liu, H., Wang, T., Wang, P., Ma, X., 2014. A novel system enhancing 1028 the endosomal escapes of peptides promotes Bak BH3 peptide inducing apoptosis in lung 1029 cancer A549 cells. Target Oncol 9, 163–170. https://doi.org/10.1007/s11523-013-0282-9 1030 Liu, S., Wang, B., Sheng, Y., Dong, S., Liu, G., 2022. Rational Design of Self‐Assembled 1031 Mitochondria‐Targeting Lytic Peptide Conjugates with Enhanced Tumor Selectivity. 1032 Chemistry – A European Journal 28. https://doi.org/10.1002/chem.202103517 1033 Liu, X., Zhang, P., Rödl, W., Maier, K., Lächelt, U., Wagner, E., 2017. Toward Artificial 1034 Immunotoxins: Traceless Reversible Conjugation of RNase A with Receptor Targeting and 1035 Endosomal Escape Domains. Mol Pharm 14, 1439–1449. 1036 https://doi.org/10.1021/acs.molpharmaceut.6b00701 1037 Loison, F., Nizard, P., Sourisseau, T., le Goff, P., Debure, L., le Drean, Y., Michel, D., 2005. A 1038 ubiquitin-based assay for the cytosolic uptake of protein transduction domains. 1039 Molecular Therapy 11, 205–214. https://doi.org/10.1016/j.ymthe.2004.10.010 1040 Lönn, P., Dowdy, S.F., 2015. Cationic PTD/CPP-mediated macromolecular delivery: charging 1041 into the cell. Expert Opin Drug Deliv 12, 1627–1636. 1042 https://doi.org/10.1517/17425247.2015.1046431 1043 Lönn, P., Kacsinta, A.D., Cui, X.-S., Hamil, A.S., Kaulich, M., Gogoi, K., Dowdy, S.F., 2016. 1044 Enhancing Endosomal Escape for Intracellular Delivery of Macromolecular Biologic 1045 Therapeutics. Sci Rep 6, 32301. https://doi.org/10.1038/srep32301 1046 López-Laguna, H., Cubarsi, R., Unzueta, U., Mangues, R., Vázquez, E., Villaverde, A., 2020. 1047 Endosomal escape of protein nanoparticles engineered through humanized histidine-rich 1048 peptides. Sci China Mater 63, 644–653. https://doi.org/10.1007/s40843-019-1231-y 1049 López-Laguna, H., Voltà-Durán, E., Parladé, E., Villaverde, A., Vázquez, E., Unzueta, U., 2022. 1050 Insights on the emerging biotechnology of histidine-rich peptides. Biotechnol Adv 54, 1051 107817. https://doi.org/10.1016/j.biotechadv.2021.107817 1052 Lundberg, P., El‐Andaloussi, S., Sütlü, T., Johansson, H., Langel, Ü., 2007. Delivery of short 1053 interfering RNA using endosomolytic cell‐penetrating peptides. The FASEB Journal 21, 1054 2664–2671. https://doi.org/10.1096/fj.06-6502com 1055 Madani, F., Abdo, R., Lindberg, S., Hirose, H., Futaki, S., Langel, Ü., Gräslund, A., 2013. 1056 Modeling the endosomal escape of cell-penetrating peptides using a transmembrane pH 1057 gradient. Biochimica et Biophysica Acta (BBA) - Biomembranes 1828, 1198–1204. 1058 https://doi.org/10.1016/j.bbamem.2012.12.008 1059 Madani, F., Perálvarez-Marín, A., Gräslund, A., 2011. Liposome Model Systems to Study the 1060 Endosomal Escape of Cell-Penetrating Peptides: Transport across Phospholipid 1061 Membranes Induced by a Proton Gradient. J Drug Deliv 2011, 1–7. 1062 https://doi.org/10.1155/2011/897592 1063 Maier, K., Martin, I., Wagner, E., 2012. Sequence Defined Disulfide-Linked Shuttle for Strongly 1064 Enhanced Intracellular Protein Delivery. Mol Pharm 9, 3560–3568. 1065 https://doi.org/10.1021/mp300404d 1066 Mantis, N. (Ed.), 2012. Ricin and Shiga Toxins. Springer Berlin Heidelberg, Berlin, Heidelberg. 1067 https://doi.org/10.1007/978-3-642-27470-1 9 1068 Martens, T.F., Remaut, K., Demeester, J., de Smedt, S.C., Braeckmans, K., 2014. Intracellular 1069 delivery of nanomaterials: How to catch endosomal escape in the act. Nano Today 9, 1070 344–364. https://doi.org/10.1016/j.nantod.2014.04.011 1071 Matsuzaki, K., 2019. Membrane Permeabilization Mechanisms. pp. 9–16. 1072 https://doi.org/10.1007/978-981-13-3588-4_2 1073 Mayer, S.F., Ducrey, J., Dupasquier, J., Haeni, L., Rothen-Rutishauser, B., Yang, J., Fennouri, A., 1074 Mayer, M., 2019. Targeting specific membranes with an azide derivative of the pore- 1075 forming peptide ceratotoxin A. Biochimica et Biophysica Acta (BBA) - Biomembranes 1076 1861, 183023. https://doi.org/10.1016/j.bbamem.2019.07.011 1077 Mejia, F., Khan, S., Omstead, D.T., Minetos, C., Bilgicer, B., 2022. Identification and 1078 optimization of tunable endosomal escape parameters for enhanced efficacy in peptide- 1079 targeted prodrug-loaded nanoparticles. Nanoscale 14, 1226–1240. 1080 https://doi.org/10.1039/D1NR05357D 1081 Mellert, K., Lamla, M., Scheffzek, K., Wittig, R., Kaufmann, D., 2012. Enhancing Endosomal 1082 Escape of Transduced Proteins by Photochemical Internalisation. PLoS One 7, e52473. 1083 https://doi.org/10.1371/journal.pone.0052473 1084 Michalska, M., Wolf, P., 2015. Pseudomonas Exotoxin A: optimized by evolution for effective 1085 killing. Front Microbiol 6. https://doi.org/10.3389/fmicb.2015.00963 1086 Mohammed, A.F., Abdul-Wahid, A., Huang, E.H.-B., Bolewska-Pedyczak, E., Cydzik, M., Broad, 1087 A.E., Gariépy, J., 2012. The Pseudomonas aeruginosa exotoxin A translocation domain 1088 facilitates the routing of CPP–protein cargos to the cytosol of eukaryotic cells. Journal of 1089 Controlled Release 164, 58–64. https://doi.org/10.1016/j.jconrel.2012.10.006 1090 Morito, D., Nagata, K., 2015. Pathogenic Hijacking of ER-Associated Degradation: Is ERAD 1091 Flexible? Mol Cell 59, 335–344. https://doi.org/10.1016/j.molcel.2015.06.010 1092 Nakase, I., Kogure, K., Harashima, H., Futaki, S., 2011. Application of a Fusiogenic Peptide GALA 1093 for Intracellular Delivery. pp. 525–533. https://doi.org/10.1007/978-1-60761-919-2_37 1094 Niikura, K., Horisawa, K., Doi, N., 2016. Endosomal escape efficiency of fusogenic B18 and B55 1095 peptides fused with anti-EGFR single chain Fv as estimated by nuclear translocation. J 1096 Biochem 159, 123–132. https://doi.org/10.1093/jb/mvv083 1097 Niikura, K., Horisawa, K., Doi, N., 2015. A fusogenic peptide from a sea urchin fertilization 1098 protein promotes intracellular delivery of biomacromolecules by facilitating endosomal 1099 escape. J Control Release 212, 85–93. https://doi.org/10.1016/j.jconrel.2015.06.020 1100 Nowakowska-Gołacka, J., Sominka, H., Sowa-Rogozińska, N., Słomińska-Wojewódzka, M., 2019. 1101 Toxins Utilize the Endoplasmic Reticulum-Associated Protein Degradation Pathway in 1102 Their Intoxication Process. Int J Mol Sci 20, 1307. https://doi.org/10.3390/ijms20061307 1103 Oess, S., Hildt, E., 2000. Novel cell permeable motif derived from the PreS2-domain of 1104 hepatitis-B virus surface antigens. Gene Ther 7, 750–758. 1105 https://doi.org/10.1038/sj.gt.3301154 1106 Oude Blenke, E., Sleszynska, M., Evers, M.J.W., Storm, G., Martin, N.I., Mastrobattista, E., 2017. 1107 Strategies for the Activation and Release of the Membranolytic Peptide Melittin from 10 1108 Liposomes Using Endosomal pH as a Trigger. Bioconjug Chem 28, 574–582. 1109 https://doi.org/10.1021/acs.bioconjchem.6b00677 1110 Pal, D., 2021. Spike protein fusion loop controls SARS-CoV-2 fusogenicity and infectivity. J 1111 Struct Biol 213, 107713. https://doi.org/10.1016/j.jsb.2021.107713 1112 Paray, B.A., Ahmad, A., Khan, J.M., Taufiq, F., Pathan, A., Malik, A., Ahmed, M.Z., 2021. The 1113 role of the multifunctional antimicrobial peptide melittin in gene delivery. Drug Discov 1114 Today 26, 1053–1059. https://doi.org/10.1016/j.drudis.2021.01.004 1115 Pei, D., 2022. How Do Biomolecules Cross the Cell Membrane? Acc Chem Res 55, 309–318. 1116 https://doi.org/10.1021/acs.accounts.1c00560 1117 Pei, D., Buyanova, M., 2019. Overcoming Endosomal Entrapment in Drug Delivery. Bioconjug 1118 Chem 30, 273–283. https://doi.org/10.1021/acs.bioconjchem.8b00778 1119 Peters, P.J., Borst, J., Oorschot, V., Fukuda, M., Krähenbühl, O., Tschopp, J., Slot, J.W., Geuze, 1120 H.J., 1991. Cytotoxic T lymphocyte granules are secretory lysosomes, containing both 1121 perforin and granzymes. Journal of Experimental Medicine 173, 1099–1109. 1122 https://doi.org/10.1084/jem.173.5.1099 1123 Peters, P.J., Bos, E., Griekspoor, A., 2006. Cryo‐Immunogold Electron Microscopy. Curr Protoc 1124 Cell Biol 30. https://doi.org/10.1002/0471143030.cb0407s30 1125 Pirie, C.M., Hackel, B.J., Rosenblum, M.G., Wittrup, K.D., 2011. Convergent Potency of 1126 Internalized Gelonin Immunotoxins across Varied Cell Lines, Antigens, and Targeting 1127 Moieties. Journal of Biological Chemistry 286, 4165–4172. 1128 https://doi.org/10.1074/jbc.M110.186973 1129 Plaza-GA, I., Manzaneda-González, V., Kisovec, M., Almendro-Vedia, V., Muñoz-Úbeda, M., 1130 Anderluh, G., Guerrero-Martínez, A., Natale, P., López Montero, I., 2019. pH-triggered 1131 endosomal escape of pore-forming Listeriolysin O toxin-coated gold nanoparticles. J 1132 Nanobiotechnology 17, 108. https://doi.org/10.1186/s12951-019-0543-6 1133 Prchla, E., Plank, C., Wagner, E., Blaas, D., Fuchs, R., 1995. Virus-mediated release of 1134 endosomal content in vitro: different behavior of adenovirus and rhinovirus serotype 2. 1135 Journal of Cell Biology 131, 111–123. https://doi.org/10.1083/jcb.131.1.111 1136 Priyadarshini, D., Ivica, J., Separovic, F., de Planque, M.R.R., 2022. Characterisation of cell 1137 membrane interaction mechanisms of antimicrobial peptides by electrical bilayer 1138 recording. Biophys Chem 281, 106721. https://doi.org/10.1016/j.bpc.2021.106721 1139 Quarta, A., Rodio, M., Cassani, M., Gigli, G., Pellegrino, T., del Mercato, L.L., 2017. Multilayered 1140 Magnetic Nanobeads for the Delivery of Peptides Molecules Triggered by Intracellular 1141 Proteases. ACS Appl Mater Interfaces 9, 35095–35104. 1142 https://doi.org/10.1021/acsami.7b05709 1143 Raj, S., Khurana, S., Choudhari, R., Kesari, K.K., Kamal, M.A., Garg, N., Ruokolainen, J., Das, B.C., 1144 Kumar, D., 2021. Specific targeting cancer cells with nanoparticles and drug delivery in 1145 cancer therapy. Semin Cancer Biol 69, 166–177. 1146 https://doi.org/10.1016/j.semcancer.2019.11.002 11 1147 Richardson, S.C.W., Pattrick, N.G., Lavignac, N., Ferruti, P., Duncan, R., 2010. Intracellular fate 1148 of bioresponsive poly(amidoamine)s in vitro and in vivo. Journal of Controlled Release 1149 142, 78–88. https://doi.org/10.1016/j.jconrel.2009.09.025 1150 Rouet, R., Christ, D., 2019. Efficient Intracellular Delivery of CRISPR-Cas Ribonucleoproteins 1151 through Receptor Mediated Endocytosis. ACS Chem Biol 14, 554–561. 1152 https://doi.org/10.1021/acschembio.9b00116 1153 Sakurai, Y., Hatakeyama, H., Sato, Y., Akita, H., Takayama, K., Kobayashi, S., Futaki, S., 1154 Harashima, H., 2011. Endosomal escape and the knockdown efficiency of liposomal- 1155 siRNA by the fusogenic peptide shGALA. Biomaterials 32, 5733–5742. 1156 https://doi.org/10.1016/j.biomaterials.2011.04.047 1157 Sala, R., Sánchez-García, L., Serna, N., Céspedes, M.V., Casanova, I., Roldán, M., Sánchez- 1158 Chardi, A., Unzueta, U., Vázquez, E., Mangues, R., Villaverde, A., 2019. Collaborative 1159 membrane activity and receptor-dependent tumor cell targeting for precise nanoparticle 1160 delivery in CXCR4+ colorectal cancer. Acta Biomater 99, 426–432. 1161 https://doi.org/10.1016/j.actbio.2019.09.002 1162 Salomone, F., Cardarelli, F., di Luca, M., Boccardi, C., Nifosì, R., Bardi, G., di Bari, L., Serresi, M., 1163 Beltram, F., 2012. A novel chimeric cell-penetrating peptide with membrane-disruptive 1164 properties for efficient endosomal escape. Journal of Controlled Release 163, 293–303. 1165 https://doi.org/10.1016/j.jconrel.2012.09.019 1166 Sánchez-García, L., Serna, N., Álamo, P., Sala, R., Céspedes, M.V., Roldan, M., Sánchez-Chardi, 1167 A., Unzueta, U., Casanova, I., Mangues, R., Vázquez, E., Villaverde, A., 2018. Self- 1168 assembling toxin-based nanoparticles as self-delivered antitumoral drugs. Journal of 1169 Controlled Release 274, 81–92. https://doi.org/10.1016/j.jconrel.2018.01.031 1170 Sánchez-García, L., Serna, N., Mattanovich, M., Cazzanelli, P., Sánchez-Chardi, A., Conchillo- 1171 Solé, O., Cortés, F., Daura, X., Unzueta, U., Mangues, R., Villaverde, A., Vázquez, E., 2017. 1172 The fusogenic peptide HA2 impairs selectivity of CXCR4-targeted protein nanoparticles. 1173 Chemical Communications 53, 4565–4568. https://doi.org/10.1039/C6CC09900A 1174 Schmit, N.E., Neopane, K., Hantschel, O., 2019. Targeted Protein Degradation through Cytosolic 1175 Delivery of Monobody Binders Using Bacterial Toxins. ACS Chem Biol 14, 916–924. 1176 https://doi.org/10.1021/acschembio.9b00113 1177 Schuster, M., Schnell, L., Feigl, P., Birkhofer, C., Mohr, K., Roeder, M., Carle, S., Langer, S., 1178 Tippel, F., Buchner, J., Fischer, G., Hausch, F., Frick, M., Schwan, C., Aktories, K., Schiene- 1179 Fischer, C., Barth, H., 2017. The Hsp90 machinery facilitates the transport of diphtheria 1180 toxin into human cells. Sci Rep 7, 613. https://doi.org/10.1038/s41598-017-00780-x 1181 Selby, L.I., Cortez‐Jugo, C.M., Such, G.K., Johnston, A.P.R., 2017. Nanoescapology: progress 1182 toward understanding the endosomal escape of polymeric nanoparticles. WIREs 1183 Nanomedicine and Nanobiotechnology 9. https://doi.org/10.1002/wnan.1452 1184 Serna, N., Sánchez, J.M., Unzueta, U., Sánchez-García, L., Sánchez-Chardi, A., Mangues, R., 1185 Vázquez, E., Villaverde, A., 2019. Recruiting potent membrane penetrability in tumor cell- 1186 targeted protein-only nanoparticles. Nanotechnology 30, 115101. 1187 https://doi.org/10.1088/1361-6528/aaf959 12 1188 Serna, N., Sánchez-García, L., Unzueta, U., Díaz, R., Vázquez, E., Mangues, R., Villaverde, A., 1189 2018. Protein-Based Therapeutic Killing for Cancer Therapies. Trends Biotechnol 36, 318– 1190 335. https://doi.org/10.1016/j.tibtech.2017.11.007 1191 Smith, S.A., Selby, L.I., Johnston, A.P.R., Such, G.K., 2019. The Endosomal Escape of 1192 Nanoparticles: Toward More Efficient Cellular Delivery. Bioconjug Chem 30, 263–272. 1193 https://doi.org/10.1021/acs.bioconjchem.8b00732 1194 Somiya, M., Kuroda, S., 2020. DDS Nanocarriers Mimicking Early Infection Machinery of 1195 Viruses. YAKUGAKU ZASSHI 140, 147–152. https://doi.org/10.1248/yakushi.19-00187-2 1196 Sowa-Rogozińska, N., Sominka, H., Nowakowska-Gołacka, J., Sandvig, K., Słomińska- 1197 Wojewódzka, M., 2019. Intracellular Transport and Cytotoxicity of the Protein Toxin Ricin. 1198 Toxins (Basel) 11, 350. https://doi.org/10.3390/toxins11060350 1199 Sudo, K., Niikura, K., Iwaki, K., Kohyama, S., Fujiwara, K., Doi, N., 2017. Human-derived 1200 fusogenic peptides for the intracellular delivery of proteins. J Control Release 255, 1–11. 1201 https://doi.org/10.1016/j.jconrel.2017.03.398 1202 Sundan, A., Olsnes, S., Sandvig, K., Pihl, A., 1982. Preparation and properties of chimeric toxins 1203 prepared from the constituent polypeptides of diphtheria toxin and ricin. Evidence for 1204 entry of ricin A-chain via the diphtheria toxin pathway. J Biol Chem 257, 9733–9. 1205 Sung, M.S., Mun, J.-Y., Kwon, O., Kwon, K.-S., Oh, D.-B., 2013. Efficient myogenic differentiation 1206 of human adipose-derived stem cells by the transduction of engineered MyoD protein. 1207 Biochem Biophys Res Commun 437, 156–161. 1208 https://doi.org/10.1016/j.bbrc.2013.06.058 1209 Tsai, B., Rodighiero, C., Lencer, W.I., Rapoport, T.A., 2001. Protein Disulfide Isomerase Acts as a 1210 Redox-Dependent Chaperone to Unfold Cholera Toxin. Cell 104, 937–948. 1211 https://doi.org/10.1016/S0092-8674(01)00289-6 1212 van Rossenberg, S.M.W., Sliedregt-Bol, K.M., Meeuwenoord, N.J., van Berkel, T.J.C., van Boom, 1213 J.H., van der Marel, G.A., Biessen, E.A.L., 2002. Targeted Lysosome Disruptive Elements 1214 for Improvement of Parenchymal Liver Cell-specific Gene Delivery. Journal of Biological 1215 Chemistry 277, 45803–45810. https://doi.org/10.1074/jbc.M203510200 1216 Váňová, J., Číhařová, B., Hejtmánková, A., Epperla, C.P., Škvára, P., Forstová, J., Hubálek 1217 Kalbáčová, M., Španielová, H., 2022. VirPorters: Insights into the action of cationic and 1218 histidine-rich cell-penetrating peptides. Int J Pharm 611, 121308. 1219 https://doi.org/10.1016/j.ijpharm.2021.121308 1220 Varkouhi, A.K., Scholte, M., Storm, G., Haisma, H.J., 2011. Endosomal escape pathways for 1221 delivery of biologicals. Journal of Controlled Release 151, 220–228. 1222 https://doi.org/10.1016/j.jconrel.2010.11.004 1223 Vasdekis, A.E., Scott, E.A., O’Neil, C.P., Psaltis, D., Hubbell, Jeffrey.A., 2012. Precision 1224 Intracellular Delivery Based on Optofluidic Polymersome Rupture. ACS Nano 6, 7850– 1225 7857. https://doi.org/10.1021/nn302122h 1226 Vazquez, E., Ferrer-Miralles, N., Mangues, R., Corchero, J., Schwartz Jr., S., Villaverde, A., 2009. 1227 Modular Protein Engineering in Emerging Cancer Therapies. Curr Pharm Des 15, 893–916. 1228 https://doi.org/10.2174/138161209787582084 13 1229 Vercauteren, D., Deschout, H., Remaut, K., Engbersen, J.F.J., Jones, A.T., Demeester, J., de 1230 Smedt, S.C., Braeckmans, K., 2011. Dynamic Colocalization Microscopy To Characterize 1231 Intracellular Trafficking of Nanomedicines. ACS Nano 5, 7874–7884. 1232 https://doi.org/10.1021/nn2020858 1233 Verdurmen, W.P.R., Mazlami, M., Plückthun, A., 2017. A Biotin Ligase-Based Assay for the 1234 Quantification of the Cytosolic Delivery of Therapeutic Proteins. pp. 223–236. 1235 https://doi.org/10.1007/978-1-4939-6857-2_14 1236 Voltà-Durán, E., Sánchez, J.M., Parladé, E., Serna, N., Vazquez, E., Unzueta, U., Villaverde, A., 1237 2022. The Diphtheria Toxin Translocation Domain Impairs Receptor Selectivity in Cancer 1238 Cell-Targeted Protein Nanoparticles. Pharmaceutics 14, 2644. 1239 https://doi.org/10.3390/pharmaceutics14122644 1240 Wadia, J.S., Stan, R. v, Dowdy, S.F., 2004. Transducible TAT-HA fusogenic peptide enhances 1241 escape of TAT-fusion proteins after lipid raft macropinocytosis. Nat Med 10, 310–315. 1242 https://doi.org/10.1038/nm996 1243 Wensley, H.J., Johnston, D.A., Smith, W.S., Holmes, S.E., Flavell, S.U., Flavell, D.J., 2020. A Flow 1244 Cytometric Method to Quantify the Endosomal Escape of a Protein Toxin to the Cytosol 1245 of Target Cells. Pharm Res 37, 16. https://doi.org/10.1007/s11095-019-2725-1 1246 Wong, A.S.M., Mann, S.K., Czuba, E., Sahut, A., Liu, H., Suekama, T.C., Bickerton, T., Johnston, 1247 A.P.R., Such, G.K., 2015. Self-assembling dual component nanoparticles with endosomal 1248 escape capability. Soft Matter 11, 2993–3002. https://doi.org/10.1039/C5SM00082C 1249 Worch, R., 2014. Structural biology of the influenza virus fusion peptide. Acta Biochim Pol 61, 1250 421–6. 1251 Yamashita, H., Demizu, Y., Shoda, T., Sato, Y., Oba, M., Tanaka, M., Kurihara, M., 2014. 1252 Amphipathic short helix-stabilized peptides with cell-membrane penetrating ability. 1253 Bioorg Med Chem 22, 2403–2408. https://doi.org/10.1016/j.bmc.2014.03.005 1254 Yang, S.-T., Zaitseva, E., Chernomordik, L. v., Melikov, K., 2010. Cell-Penetrating Peptide 1255 Induces Leaky Fusion of Liposomes Containing Late Endosome-Specific Anionic Lipid. 1256 Biophys J 99, 2525–2533. https://doi.org/10.1016/j.bpj.2010.08.029 1257 Yu, X., Xue, L., Zhao, J., Zhao, S., Wu, D., Liu, H.Y., 2021. Non-Cationic RGD-Containing Protein 1258 Nanocarrier for Tumor-Targeted siRNA Delivery. Pharmaceutics 13, 2182. 1259 https://doi.org/10.3390/pharmaceutics13122182 1260 Yuba, E., Kojima, C., Harada, A., Tana, Watarai, S., Kono, K., 2010. pH-Sensitive fusogenic 1261 polymer-modified liposomes as a carrier of antigenic proteins for activation of cellular 1262 immunity. Biomaterials 31, 943–951. https://doi.org/10.1016/j.biomaterials.2009.10.006 1263 Zaitseva, E., Yang, S.-T., Melikov, K., Pourmal, S., Chernomordik, L. v., 2010. Dengue Virus 1264 Ensures Its Fusion in Late Endosomes Using Compartment-Specific Lipids. PLoS Pathog 6, 1265 e1001131. https://doi.org/10.1371/journal.ppat.1001131 1266 1267 1268 14 1269 1270 1271 1272 15