Acta Biomaterialia 170 (2023) 543–555 Contents lists available at ScienceDirect Acta Biomaterialia journal homepage: www.elsevier.com/locate/actbio Full length article High-precision targeting and destruction of cancer-associated PDGFR-β + stromal fibroblasts through self-assembling, protein-only nanoparticles Eric Voltà-Durán a,b,c,1, Lorena Alba-Castellón c,d,e,1,∗, Naroa Serna a,b,c,2, Isolda Casanova c,d,e, Hèctor López-Laguna a,b,c, Alberto Gallardo d,g, Alejandro Sánchez-Chardi f,3, Antonio Villaverde a,b,c,∗, Ugutz Unzueta c,d,e, Esther Vázquez a,b,c, Ramón Mangues c,d,e,∗ a Institut de Biotecnologia i de Biomedicina, Universitat Autònoma de Barcelona, Bellaterra, Barcelona 08193, Spain b Departament de Genètica i de Microbiologia, Universitat Autònoma de Barcelona, Bellaterra, Barcelona 08193, Spain c CIBER de Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Spain d Biomedical Research Institute Sant Pau (IIB Sant Pau), Barcelona 08041, Spain e Josep Carreras Leukaemia Research Institute, Barcelona 08025, Spain f Departament de Biologia Evolutiva, Ecologia i Ciències Ambientals, Facultat de Biologia, Universitat de Barcelona, Av. Diagonal 643, Barcelona 08028, Spain g Department of Pathology, Hospital de la Santa Creu i Sant Pau, Barcelona 08025, Spain a r t i c l e i n f o a b s t r a c t Article history: The need for more effective and precision medicines for cancer has pushed the exploration of new mate- Received 25 May 2023 rials appropriate for drug delivery and imaging, and alternative receptors for targeting. Among the most Revised 30 August 2023 promising strategies, finding suitable cell surface receptors and targeting agents for cancer-associated Accepted 1 September 2023 platelet derived growth factor receptor β (PDGFR-β )+ stromal fibroblasts is highly appealing. As a ne- Available online 6 September 2023 glected target, this cell type mechanically and biologically supports the growth, progression, and infiltra- Keywords: tion of solid tumors in non-small cell lung, breast, pancreatic, and colorectal cancers. We have developed Nanoparticles a family of PDGFR-β -targeted nanoparticles based on biofabricated, self-assembling proteins, upon hierar- Protein materials chical and iterative selective processes starting from four initial candidates. The modular protein PDGFD- Cell targeting GFP-H6 is well produced in recombinant bacteria, resulting in structurally robust oligomeric particles that Cancer selectively penetrates into PDGFR-β + stromal fibroblasts in a dose-dependent manner, by means of the Drug delivery PDGFR-β ligand PDGFD. Upon in vivo administration, these GFP-carrying protein nanoparticles precisely Precision medicines Nanomedicine accumulate in tumor tissues and enlighten them for IVIS observation. When GFP is replaced by a micro- bial toxin, selective tumor tissue destruction is observed associated with a significant reduction in tumor volume growth. The presented data validate the PDGFR-β /PDGFD pair as a promising toolbox for tar- geted drug delivery in the tumor microenvironment and oligomeric protein nanoparticles as a powerful instrument to mediate highly selective biosafe targeting in cancer through non-cancer cells. Statement of significance We have developed a transversal platform for nanoparticle-based drug delivery into cancer-associated fibroblasts. This is based on the engineered modular protein PDGFD-GFP-H6 that spontaneously self- assemble and selectively penetrates into PDGFR-β + stromal fibroblasts in a dose-dependent manner, by means of the PDGFR-β ligand PDGFD. In vivo, these protein nanoparticles accumulate in tumor and when incorporating a microbial toxin, they destroy tumor tissues with a significant reduction in tumor volume, in absence of side toxicities. The data presented here validate the PDGFR-β /PDGFD pair as a fully versatile toolbox for targeted drug delivery in the tumor microenvironment intended as a synergistic treatment. © 2023 The Author(s). Published by Elsevier Ltd on behalf of Acta Materialia Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/) ∗ Corresponding authors at: CIBER de Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Spain. E-mail addresses: LAlba@santpau.cat (L. Alba-Castellón), antoni.villaverde@uab.cat (A. Villaverde), rmangues@santpau.cat (R. Mangues). 1 These authors contributed equally to this work. 2 Present address: Nanoligent SL, Edifici Eureka, Universitat Autònoma de Barcelona, Bellaterra, 08193 Barcelona, Spain. 3 Present address: Oficina de Sostenibilitat, Edifici L, Universitat Autònoma de Barcelona, Bellaterra, Barcelona 08193, Spain. https://doi.org/10.1016/j.actbio.2023.09.001 1742-7061/© 2023 The Author(s). Published by Elsevier Ltd on behalf of Acta Materialia Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/) E. Voltà-Durán, L. Alba-Castellón, N. Serna et al. Acta Biomaterialia 170 (2023) 543–555 1. Introduction [48], but appropriate drugs or drug vehicles have not been yet de- veloped and this type of approach has been essentially neglected. Cancer represents a leading cause of death worldwide and an In the present study, we have combined structural concepts unmet clinical problem [1]. It is so despite the substantial eco- from biomaterial design and self-assembling and from cell-targeted nomic and human resources invested in anti-tumor drug discov- cancer nanomedicines based on ligand-receptor pairs, with the po- ery and the diversity of innovative therapeutic approaches and tential use of PDGFR-β -expressing stromal fibroblasts as targets emerging materials intended for more effective drug delivery that for anticancer therapies, either as single agents or in combina- are currently under evaluation [2–6]. The generic lack of selec- tion with additional targeted drugs. Taking the concept of mod- tivity and high toxicity of most chemical drugs, the development ular proteins as versatile and convenient materials for bio-inspired of drug resistance and relapse and the multiple facets that sup- and clinically-oriented uses [49–60], we have generated protein- port cancer development and progression make this goal espe- only nanoparticles based on self-assembling building blocks and cially complex. Increasing drug selectivity for target cells through targeted to cell-surface receptors of these cells. These materials, smarter and selective drug vehicles and identifying complemen- produced by simple and fully scalable biofabrication processes, tary therapeutic targets should both support significant steps to- have been successfully validated here as highly selective targeting wards successful cancer therapies. In the context of these non- agents which allow internalization in PDGFR-β + fibroblasts, both exclusive alternatives, high precision drug delivery is being ex- in cell culture and in vivo, using multiple analytical procedures. By plored mainly by the refinement of nanomedical tools [7–13] and incorporating a potent microbial toxin, the targeted nanoparticles by developing biocompatible materials within the nanoscale as promote a precise and selective destruction of tumor in an animal drug carriers. Their building blocks are to be functionalized with subcutaneous model of colorectal cancer, resulting in reduction of specific ligands of overexpressed tumoral markers for selective cy- tumor volume growth. Altogether, the obtained data fully supports totoxic drug delivery and the consequent cell destruction, in a se- the concept of selective drug carriers based on nanoscale protein lective way [11,14–16]. Based on such functionalization, oligomeric materials to target the main architectonic agents in solid tumors nanoparticles benefit from the resulting cooperativity in receptor for their further development and use in the clinical setting. binding, enhanced endosomal engulfment and higher cell pene- trability compared to monomeric ligands [17–20]. Metastatic can- 2. Experimental section cer stem cells are particularly appealing targets for high precision nanomedicines as they are the origin of metastases, the main cause 2.1. Protein description of patients’ death. So far, despite being theoretically promising, at- tempts of selectivity in cancer nanomedicines have not generically Four peptide ligands of PDGFR-β were selected from literature reached accumulation in target tissues, being over 1-2 % of the and described hereafter. Platelet derived growth factor B (PDGFB) administered material [21,22]. Further, many nanoscale materials, and platelet derived growth factor D (PDGFD) are both human lig- because of their physicochemical properties or their natural inter- ands of PDFGR-β [61]. PDGFRP1 is a small cationic peptide isolated activity with biological receptors are in fact retained in liver or by phage display [62], while Z09591 is an anionic affibody also phagocyted [23–25], precluding the desired biodistribution of the identified by phage display and commonly used for radiolabeling carrier and the payload drug. An interesting exception is the en- and imaging of PDGFR-β expression in different cancers [46,63,64]. gineering of self-assembling tumor-targeted modular proteins, that follow a highly desired tumor biodistribution [26–28] and which 2.2. Protein design, production, and purification in some cases show values of tumor accumulation, upon systemic administration, over 85 % of the total observed signal [29,30]. The sequences of four green fluorescent PDGFR-β targeted pro- On the other hand, the identification of alternative or comple- teins and the PDGFR-β targeted nanotoxin PDGFD-NT-H6 were mentary targets for selective cell killing offers promising routes designed in house as codon-optimized genes and subcloned into to designing innovative drugs [31]. Looking at those other poten- pET22b plasmids using NdeI and HindIII restriction enzymes. Gen- tial strategies, stroma refers to the tissue that surrounds epithe- eart (ThermoFisher) provided the recombinant plasmids. For fluo- lial cancer cells, including extracellular matrix, endothelial cells, rescent proteins, PDGFR-β targeted domains were placed at the N- immune cells and fibroblasts, which modulate and participate in terminus of each protein, followed by a flexible linker (GGSSRSS), cancer development. Among these elements, fibroblasts, synthe- the green fluorescent protein (GFP) and the hexa-histidine tag H6. size and remodel the extracellular matrix being then responsible The H6 tag allows both protein purification but it also promotes for the cancer tissue architecture and structural properties [32]. Fi- assembling through coordination with divalent cations [65]. For broblasts are also responsible for the production of growth factors the nanotoxin, PDGFD was located at the N-terminus, followed by and cytokines that will be secreted within the tumor [33], support- a furin cleavable site (GNRVRRSV) flanked by two flexible linkers ing its development [32,34]. In particular, a specific subpopulation (GGSSRSS), the catalytic domain of Pseudomonas aeruginosa exo- of this cell type, namely platelet-derived growth factor receptor toxin A (PE24), the H6 tag and the C-terminal subcellular location β (PDGFR-β )-expressing fibroblasts are positively involved in can- signal KDEL. The recombinant pET22b vectors were transformed cer progression and its infiltration in tumors compromises patient into Escherichia coli by heat shock at 42 ºC for 45 s. PDGFRP1-GFP- survival, especially in non-small cell lung, breast, pancreatic and H6 and Z09591-GFP-H6, lacking intramolecular or intermolecular colorectal cancer [35–38]. PDGFR-β is a transmembrane protein, disulfide bonds, were produced in BL21 (DE3), PDGFB-GFP-H6 in member of the family of Tyr kinases that participates in cell dif- Origami B (BL21 DE3, OmpT− , Lon− , TrxB− , Gor− , Novagen), and ferentiation, migration and proliferation processes in several neo- PDGFD-GFP-H6 and PDGFD-NT-H6 in BL21 (DE3) previously engi- plasias [39]. Importantly, its expression levels in cancer-associated neered for the expression of sulfhydryl oxidase and DsbC [66,67] fibroblasts (CAFs) correlates with bad prognosis, relapse and drug (kindly provided by Prof. A. de Marco). resistance in several types of cancer [40–44]. Therefore, stromal Protein production was carried out at the optimal conditions PDGFR-β + fibroblasts are a pivotal, highly interesting cell type for for each candidate. In this sense, PDGFRP1-GFP-H6 and Z09591- imaging, drug delivery or selective destruction in the context of GFP-H6 were produced overnight at 20 ºC and 250 rpm in Luria cancer therapies [45]. Despite the robust data supporting PDGFR- Broth (LB) supplemented with 100 μg mL −1 ampicillin, upon ad- β + fibroblasts as targets in cancer therapies [35,39,42,46,47], the dition of 1×10−4 M of isopropyl-β -d-thiogalactopyronaside (IPTG), modest explorations have been so far limited to interfere signaling at OD550nm 0.6-0.8 units. PDGFB-GFP-H6 was produced overnight 544 E. Voltà-Durán, L. Alba-Castellón, N. Serna et al. Acta Biomaterialia 170 (2023) 543–555 at 16 ºC and 250 rpm in LB supplemented with 100 μg mL −1 GFP-H6. Protein stability in the in vitro culture media (Dulbecco’s ampicillin, 12.5 μg mL −1 tetracycline and 15 μg mL −1 kanamycin, Modified Eagle Medium, DMEM, Gibco; supplemented with fetal upon addition of 1×10−4 M IPTG, at OD550nm 0.6-0.8 units. Cells bovine serum, FBS, Gibco) was measured by incubating the pro- were then harvested by centrifugation (15 min, 50 0 0 g) and stored teins at the maximum concentration used in in vitro experiments at -80 ºC. For protein purification, cells were resuspended in wash (1×10−3 M), for the maximum time of exposure (24 h) at 37 ºC, buffer (2×10−2 M Tris-HCl, 5×10−1 M NaCl, 1×10−2 M imida- and separating the soluble and insoluble fractions by centrifugation zole, pH 8.0) in presence of protease inhibitors (cOmplete EDTA- (15 min, 150 0 0 g). A protein electrophoresis gel, followed by an free, Roche Diagnostics) and disrupted in an EmulsiFlex-C5 system anti-His Western Blot, were finally performed. Size exclusion chro- (Avestin) by 3 rounds at 80 0 0 psi. The soluble fraction was then matography was used to reveal oligomeric status of PDGFB-GFP-H6 collected by centrifugation (45 min at 150 0 0 g) and proteins puri- and PDGFD-GFP-H6. Both proteins were analyzed using Superdex fied by an immobilized metal affinity chromatography (IMAC) us- 20 0 Increase 10-30 0 GL (Cytiva) operated at 4ºC, 0.5 mL/min, and ing a HisTrap HP column (GE Healthcare) in an ÄKTA pure system 0.5 mg/mL (total amount 2.5×104 g). For PDGFD-GFP-H6, an ad- (GE Healthcare). Elution was achieved by a lineal increase of im- ditional analytical size exclusion chromatography was performed idazole concentration (Elution Buffer, 2×10−2 M Tris-HCl, 5×10−1 to discriminate among different populations of nanoparticles. For M NaCl, 5×10−1 M imidazole, pH 8.0). that, Superose 6 Increase 10-300 GL (Cytiva) was used under the For PDGFD-GFP-H6 and PDGFD-NT-H6 protein production, bac- same settings previously described for Superdex 200 Increase 10- teria were grown in LB supplemented with 100 μg mL −1 ampi- 300 GL. cillin and 34 μg mL −1 chloramphenicol at 37 ºC until OD550nm reached 0.4-0.5 units. In that moment, the expression of sulfhydryl 2.4. Ultrastructural characterization oxidase and DsbC was induced by adding 0.5 % (m/v) of L- arabinose and temperature was lowered to 30 ºC. 45 min later, Nanoscale morphometry (size and shape) of the different pro- temperature was lowered to 16 ºC (PDGFD-GFP-H6) or 20 ºC tein nanoparticles was visualized at nearly native state with two (PDGFD-NT-H6) and 1×10−3 M of IPTG was added to induce rapid high-resolution imaging techniques. Drops of 5 μL of sam- overnight expression. In both cases, cells were then harvested by ples diluted at 0.2 μg mL−1 in its stock solution were deposited in centrifugation (15 min 50 0 0 g) and an osmotic shock was per- silicon wafers (Ted Pella) and observed in a field emission scan- formed to remove the metallophores from the periplasmic frac- ning electron microscope (FESEM) Merlin (Zeiss) operating at 1 kV tion [68]. For that, cells were first resuspended in a hypertonic and equipped with an in-lens secondary electron detector. Drops solution (20 % sucrose, 1×10−3 M ethylenediaminetetraacetic acid of samples diluted at 0.2 μg mL−1 in its stock solution were nega- (EDTA), 5×10−2 M 4-(2-hydroxyethyl)-1-piperazineethanesulfonic tively stained with 2 % uranyl acetate (Merck) in 400 mesh carbon- acid (HEPES), pH 7.9) and centrifuged (30 min, 70 0 0 g, 4 ºC). Su- coated copper grids (Electron Microscopy Sciences) and observed pernatant was discarded and cells were then resuspended in a hy- in a transmission electron microscope (TEM) JEM 1400 (Jeol) op- potonic solution (5×10−3 M MgSO4 ), placed 10 min at 4 ºC and erating at 80 kV and equipped with an Orius SC200 CCD camera subsequently centrifuged (15 min, 4500 g) to remove again the su- (Gatan). Representative images of general fields and nanoparticles pernatant. Bacterial cells, now without periplasm, were finally re- details were obtained at three magnifications. suspended in wash buffer in presence of protease inhibitors and stored at -80 ºC. Disruption and purification were carried out as 2.5. Three-dimensional models and visualization stated for the other proteins. For comparison purposes, GFP-H6 was produced and purified as previously described [69]. In silico three-dimensional structure prediction of the four PDGFR-β targeted modular proteins was performed using Al- 2.3. Physicochemical characterization of proteins and protein phaFold Colab. [70] ChimeraX software version 1.2 [71] was used materials then for their 3D structure visualization. GraphPad Prism 8 was used for graphics and statistical tests. After purification, proteins were dialyzed against an appropriate solution, namely sodium carbonate (166 mM NaCO3 H, pH 8.0) for 2.6. Cell lines and cell culture PDGFRP1-GFP-H6 and PDGFD-NT-H6 and sodium carbonate with salt (166 mM NaCO3 H, 333 mM NaCl, pH 8.0) for Z09591-GFP-H6, Mouse Embryonic Fibroblasts (MEFs) and Mesenchymal Stem PDGFB-GFP-H6 and PDGFD-GFP-H6. A particular amount of ZnCl2 Cells (MSCs) were kindly provided by Prof. Antonio Garcia de Her- was added to pure solutions of PDGFRP1-GFP-H6 (3:1 molar ra- reros [72,73]. MEFs, MSCs, and the mouse colorectal cancer cell tio, Zn2+ : Histidine residues in H6) and Z09591-GFP-H6 (1:1 molar line MC-38 (Kerafast) were cultured in Dulbecco’s Modified Eagle’s ratio, Zn2+ : Histidine residues in H6) to promote protein assem- Medium (DMEM) 4.5g/L glucose, supplemented with 10 % Fetal bling and nanoparticle formation. Protein purity and integrity was Bovine Serum (FBS), 100 U/mL penicillin, 100 μg/mL streptomycin determined by sodium dodecyl sulfate polyacrylamide gel elec- and 2 mM glutamine (Life Technologies) and incubated at 37°C and trophoresis (SDS-PAGE), and western blot immunodetection using 5 % CO2 in a humidified atmosphere. an anti-His (Santa Cruz Biotechnology) and an anti-GFP (Santa Cruz Biotechnology) antibody. The same molar amount was used for 2.7. Immunodetection each protein. Protein concentration was determined by the Brad- ford assay (Bio-Rad). Dynamic light scattering (DLS) was used to MEFs, MSCs and MC38 cells were plated in p100 dishes (3×105 determine the volume size distribution of each type of nanopar- cells/p100) and exposed to nanoparticles at different concentration ticle. Measurements were conducted in a Zetasizer Advanced Pro (from 25 nM to 100 nM) for two time periods (1 h, 24 h). When Blue (Malvern Instruments Limited) at 25 ºC and 633 nm (n=3) indicated, cells were treated with 50 ng/mL of platelet derived using a quartz cuvette. Also, the GFP fluorescence was evaluated growth factor B homodimer (PDGF-BB. Prepotech) 30 min prior in a Varian Cary Eclipse Fluorescence Spectrophotometer (Agilent exposure. Cells were washed with PBS, detached from the plate, Technologies). Protein concentration was set to 5 μM, the excita- and trypsinized (1 mg/ml trypsin, Life Technologies) for 15 min at tion wavelength to 488 nm and the emission recorded at 511 nm. 37°C to remove membrane binding of the nanoparticles. RIPA ly- For green fluorescent proteins, the percentage of intrinsic fluores- sis buffer plus phosphatase and protease inhibitors (Roche) was cence was calculated in comparison to the non-targeted control used to obtain whole-cell extracts. Protein extracts were loaded 545 E. Voltà-Durán, L. Alba-Castellón, N. Serna et al. Acta Biomaterialia 170 (2023) 543–555 into 10-12 % acrylamide SDS-PAGE gel and transferred to a nitro- fixed in 4 % formaldehyde solution and paraffin-embedded. Fluo- cellulose membrane to detect GFP (Santa Cruz Technologies, sc- rescence intensity (FLI) data are expressed as average radiant ef- 8334), PDGFR-β (Cell Signaling, 3169) and tubulin (Cell Signaling, ficiency and values have been calculated subtracting the FLI sig- 2148) by western blotting. Bands were visualized using ChemiDoc nal of buffer-treated mice to the FLI signal of nanoparticle treated XRS+ imaging system (Bio-rad). Bands densitometry quantitation animals. For antitumor evaluation, mice with 20-30 mm3 tumors was performed with ImageJ software. All experiments were per- were randomised and intravenously administered with 1 mg/kg formed in triplicate. of PDGFD-NT-H6 nanotoxin or NaHCO3 , daily up to 7 doses. Tu- mors were harvested 72 h after last dose and measured ex vivo to 2.8. Flow cytometry calculate their volume. Tumour samples were paraffin-embedded for histological analysis. Tumor volume was calculated as width2 x Internalization of nanoparticles was quantified by GFP fluores- length x 0.5. Mice were weighed twice a week since tumor implan- cence signal using FACS Calibur (BD Biosciences). MEFs and MSCs tation. Plasma obtained from whole cell blood extracted at the end were exposed at 25, 50, 100, 250, 500 and 1000 nM for 1 h or point was used to detect Uric acid, Albumin, Creatinine (CRE) and 24 h. After treatment, cells were washed with PBS, detached from transaminases (AST, ALT). the plate and trypsinized for 15 min at 37 ºC to eliminate non- internalized nanoparticle. Data were analyzed using the Flow Jo 2.12. Histopathology software and represented as mean fluorescence intensity (MFI) fold change with respect to buffer-treated cells. All experiments were Paraffin-embedded 4 μm tissue sections were used for immuno- performed in triplicate. histochemical analysis in DAKO Autostainer Link48 following the manufacturer’s instructions. Antigen Retrieval was done in PT Link 2.9. Confocal microscopy with pH high solution. The antibodies used were PDGFRβ (1:100, Cell Signaling 3169) and GFP (1:200, Santa Cruz sc-8334). Rep- For fluorescence microscopy, 1×104 cells were seeded on a cov- resentative images were captured using an Olympus DP73 digital erslip in 24-well plate and exposed to 25, 50 or 100 nM nanopar- camera and processed with the Olympus CellD Imaging 3.3 soft- ticles for 1 h. When indicated, 50 ng/mL PDGF-BB (Prepotech) was ware. Hematoxylin-eosin slides were scanned using Pannoramic added 30 min before exposure to assess competition. After in- Scan II (3D-Histech), and necrotic area quantification was per- cubation with nanoparticles, cells were washed, fixed with 4 % formed using the measurement tool of Slide Viewer 2.5 software formaldehyde (ThermoFisher Scientific) and blocked with 3 % BSA. (3D-Histec). The percentage was calculated by dividing the necrotic Then, nuclei were stained with DAPI and coverslips were mounted area by the total tumor area. with ProLongTM Gold Antifade Mountant (ThermoFisher Scientific). Samples were visualized with SP5 Leica Microscopy. 2.13. Statistical analyses 2.10. Cell viability assay Statistical analyses were performed using the GraphPad Prism 5 software (GraphPad Software, San Diego, California USA). Quantita- Cell viability upon exposure to PDGFB-GFP-H6, PDGFD-GFP-H6 tive data were tested for both normal distribution and homogene- and PDGFD-NT-H6 was assessed with the Cell Proliferation Kit II ity of variances, using Kolmogorov-Smirnov and Levene’s tests, re- (XTT) (Roche) according to the manufacturer’s instructions. Briefly, spectively. Then, pairwise comparisons were made with t-Student 1×103 cells (MSCs) or 2×103 cells (MEFs, MC38) were seeded in or U-Mann-Whitney tests. Differences were considered statistically 96-well plates and treated with different concentrations of PDGFB- significant when p-values≤ 0.05. GFP-H6 or PDGFD-GFP-H6 (250, 50 0 and 10 0 0 nM) for 24 or 48h, or PDGFD-NT-H6 (100 nM) for 48 h. XTT reagent was added to 3. Results and discussion the plate and further incubated at 37ºC for 4h, then absorbance, which directly correlates to the number of viable cells, was mea- Four modular GFP-based proteins were designed for recombi- sured using a multi-well spectrophotometer (FLUOstar Optima, nant production, each carrying a specific PDFGR-β ligand, namely BMG Labtech). All experiments were performed in triplicate. PDGFRP1, Z09591, PDGFB, or PDGFD (Fig. 1A). H6 tags were incor- porated into the constructs to promote their self-assembly as reg- 2.11. Mice and procedures for the administration of nanoparticles ular protein-only nanoparticles [65], because of the interaction of the imidazole ring of the His residues with divalent cations [50,78]. All mice and procedures were carried out in accordance with In silico modeling of the constructs did not anticipate any struc- the EU regulations on animal research and approved by Catalonia’s tural constraint for the solvent display of the ligand or any steric Animal ethics committee (reference 9721). Eight-week-old female impediment for the proper folding of the domains (Fig. 1B), as C57BL/6 were purchased from Charles River (France), housed in a they were separated by a peptide linker (gray box, in Fig. 1A). The specific pathogen-free (SPF) environment with sterile food and wa- recombinant production of these constructs in bacteria rendered ter ad libitum. The subcutaneous tumor model was generated by good yields of full-length protein species (between 5 and 120 subcutaneous injection of 1 million MC-38 cells mixed with 0.5 mg/L), which, upon purification, resulted in discrete bands of the million MEFs in the flank of the animal. For biodistribution assay, expected molecular weights (Fig. 1C). The integrity and high level mice bearing tumors around 10 0-20 0 mm3 were administrated in- of purity of all these constructs were further assessed by SDS-PAGE travenously with 200 μg of either PDGFB-GFP-H6 or PDGFD-GFP- and independent Western blot analyses directed against alterna- H6 nanoparticles. This model and associated procedures have been tive epitopes (Fig. S1A). The main biophysical profiles of the lig- previously validated in independent laboratories and shown to be ands and the resulting fusions, as well as the production and opti- robust for preclinical analyses [74–77]. Control animals were ad- mal storage conditions, are depicted in the Table S1. Upon dialysis, ministrated with the stock buffer for nanoparticles. Animals were PDGFB- and PDGFD-based constructs rendered particulate materi- euthanized 2 h post-administration and an ex vivo measurement of als of 16 and 26 nm (with a secondary minor peak of around 100 fluorescence intensity (FLI) of tumors was performed using IVIS® nm in the case of PDGFB-GFP-H6, Fig. 1D, Table S2), as expected for Spectrum 200 (PerkinElmer). Tumors and organs were collected, modular constructs based on H6-tagged GFP proteins with cationic 546 E. Voltà-Durán, L. Alba-Castellón, N. Serna et al. Acta Biomaterialia 170 (2023) 543–555 Fig. 1. Design and production of PDGFR-β targeted-proteins. (A) Modular disposition of each candidate, from the N-terminal (left) to the C-terminal (right) end. Modular proteins consist in a targeting peptide at the N-terminus (colored boxes), a flexible linker (GGSSGGS, grey box), the green fluorescent protein (GFP, green box) and a hexa- histidine tag (H6, orange box) at the C-terminus. The amino acid sequence of each targeting peptide is shown below. (B) Three-dimensional models of each candidate obtained by AlphaFold approach. Color code is as in panel A. (C) Protein gel electrophoresis, showing the obtained protein purity and the proteolytic integrity after affinity purification and further dialysis. P1 refers to PDGFRP1-GFP-H6 (29.7 kDa), Z to Z09591-GFP-H6 (34.6 kDa), B to PDGFB-GFP-H6 (40.5 kDa) and D to PDGFD-GFP-H6 (42.2 kDa). M indicates the marker lane, and the siding numbers the molecular mass of the markers expressed in kDa. (D) Particle size determined by DLS expressed in nm. Numbers indicate the mean peak size for every material. Additional data can be found in Table S2. (E) Representative FESEM images of isolated nanoparticles. The white bars represent 50 nm. Broader fields and TEM images for PDGFB-GFP-H6 and PDGFD-GFP-H6 are shown in the Fig. S2. N-terminal domains [79]. As in the case of related protein con- GFP-H6 emission (Fig. S3A). In addition, the proteins remained pro- structs [80], the assembly was spontaneous and did not require teolytically stable and soluble in complex culture media (Fig. S3B), additional cation supply. PDGFRP1-GFP-H6 and Z09591-GFP-H6 re- indicating that they might be used in further studies. In these anal- sulted in smaller nanoparticles of around 9 nm that were only yses, a first internalization screening of the nanoparticles into two formed when cationic Zn was added to the solution. All these ma- PDGFR-β + mesenchymal cell types, namely mouse embryonic fi- terials were disassembled in the presence of detergent (Fig. S1B), broblasts (MEFs) and mesenchymal stem cells (MSCs), showed a confirming that they were oligomers formed by the recruitment differential uptake (Fig. 2). PDGFB- and PDGFD-based constructs of monomeric building blocks. Size exclusion chromatography data (very especially PDGFD-GFP-H6) showed good cell penetrability at demonstrated the absence of monomeric populations of PDGFD- short exposure times (1 h), as determined by GFP immunodetec- GFP-H6, whereas PDGFB-GFP-H6 presented multiple conformations tion (Fig. 2A) and further densitometry analysis (Fig. 2B), flow cy- (Fig. S1C), compatible with the results obtained by DLS (Fig. 1D). tometry (Fig. 2C), or by confocal microscopy (Fig. 2D). The smaller Interestingly, the analysis of PDGFD-GFP-H6 revealed the presence PDGFRP1-GFP-H6 and Z09591-GFP-H6 particles, in contrast, ren- of protein nanoparticles with different sizes (Fig. S1D), which were dered internalization values indistinguishable from the background hidden in the broad peak obtained by DLS. In this regard, size ex- shown by the parental GFP-H6 (Fig. 2B). The amount of detected clusion chromatography demonstrated the absence of monomeric protein was slightly higher at 1 h than at 24 h (Fig. 2B), proba- populations of PDGFD-GFP-H6, whereas PDGFB-GFP-H6 organized bly because of a combination of fast and efficient cell entry and in multiple conformations with different hydrodynamic sizes (Fig. a moderate lysosomal proteolysis of the engulfed material. In this S1C), compatible with the results obtained by DLS (Fig. 1D). Inter- regard, the dotted distribution of the PDGFD-GFP-H6 fluorescence estingly, the analysis of PDGFD-GFP-H6 revealed the presence of (Fig. 2C) was indicative of an endosomal internalization route. In- protein nanoparticles with different sizes (Fig. S1D), which were terestingly, in none of the taken analyses, the exposed cells showed hidden in the broad peak obtained by DLS. signs of death or cytotoxicity (Fig. 2). For the two internalizing Using electron microscopy, the assembled materials were visu- nanoparticles, namely PDGFB-GFP-H6 and PDGFD-GFP-H6, a dose- alized as discrete entities with sizes compatible to those obtained dependent entry was also demonstrated (Fig. S4), and no plateau by DLS (Fig. 1E). A detailed morphometric exploration of selected was observed, at least up to 1 μM (Fig. S4A, B). This was indica- larger materials in wide and narrow fields revealed a regular, virus- tive of the high internalization capacity of both tested nanoparti- like geometry and a notable structural robustness (Fig. S2). cles and the cell tolerance to the uptake. At 100 nM, the GFP fluo- Envisaging functional analyses of all these materials once ex- rescence of both nanoparticles was widely distributed throughout posed to cells and whole bodies, their fluorescence emission was the cytoplasm of the target cells (Fig. S4C). Again, at high doses of compared to that of the parental GFP-H6 and assessed as a suit- exposed nanoparticles, cells did not exhibit any symptoms of toxi- able monitoring tool. Noteworthy, all the constructs, in the assem- city at different times of exposure to high doses of PDGFB-GFP-H6 bled form, were fluorescent and appropriate for further tracking (Fig. S5A) or PDGFD-GFP-H6 (Fig. S5B). This was consistent with in cell culture and in vivo, in the worst case, reaching 70 % of the the intrinsic biocompatibility of proteins as building blocks of se- 547 E. Voltà-Durán, L. Alba-Castellón, N. Serna et al. Acta Biomaterialia 170 (2023) 543–555 Fig. 2. Internalization of CAFs-directed nanoparticles. Two mesenchymal cell types, mouse embryonic fibroblasts (MEFs) and mesenchymal stem cells (MSCs), were exposed to 100 nM of different nanoparticles. (A) GFP, PDGFR-β and Tubulin immunoblotting of MEFs and MSCs whole cell extracts after 1 h exposure. (B) GFP bands of three independent experiments were quantified to compare internalization between of PDGFB-GFP-H6 and PDGFD-GFP-H6 nanoparticles. (C) Nanoparticle internalization analysed by flow cytometry in MEFs and MSCs upon 1 h or 24 h exposure. Values are expressed as mean fluorescence intensity fold change respect to untreated cells. (D) GFP detection by confocal microscopy upon 1 h exposure, Blue: DAPI. Selected fields are enlarged in the insets for a better visualization. ∗ ∗ p≤0.01; ∗ ∗ ∗ p≤0.001. Bars size: 50 μm. 548 E. Voltà-Durán, L. Alba-Castellón, N. Serna et al. Acta Biomaterialia 170 (2023) 543–555 Fig. 3. PDGFR-β is required for internalization of nanoparticles. MEFs were incubated with 50 ng/mL PDGF-BB for 30 min following 100 nM PDGFB-GFP-H6 or PDGFD-GFP- H6 treatment for 1 h. (A) Fixed cells were stained with DAPI and analysed by confocal microscopy. (B) Whole cell extracts were used for detection of PDGFR and GFP, by immunoblotting. Tubulin was used as loading control. (C) Graphs represents densitometry quantification of 3 independent assays. ∗ p ≤ 0.05; ∗ ∗ p ≤ 0.01; ∗ ∗ ∗ p ≤ 0.001 Bar size: 50 μm. lective nanoscale materials, and encouraged us to further analyze to generate a convenient cancer model, in which PDGFR-β was ex- the performance of these protein materials, focusing on clinically pressed in the tumor stromal compartment, where activated fi- oriented applications. broblasts are located (Fig. 4A). The high homology between hu- In the context of the observed uptake of both tested nanopar- man and mouse PDGFR-β (86 %) and PDGFR-β ligands (89 % for ticles (Fig. 2), a clear confirmation of the specificity of the process PDGFB and 91 % for PDGFD) allowed the testing of the nanoparti- and the requirement of the PDGFR-β as a functional receptor was cles containing human ligands in a mouse model. Two hours after required. In this regard, the soluble version of the ligand, added PDGFB-GFP-H6 or PDGFD-GFP-H6 intravenous administration, and before exposure to nanoparticles, blocked their uptake, as deter- in concordance with the in vitro data, PDGFD-GFP-H6 nanoparti- mined visually by confocal microscopy (Fig. 3A) or analytically, by cles accumulated in the tumor (Fig. 4B,C). Concomitantly with its immunoblotting (Fig. 3B, C). These competition experiments con- better performance in cell culture (Fig. 2), PDGFD-GFP-H6 showed firmed that PDGFB-GFP-H6 and PDGFD-GFP-H6 nanoparticles bind higher retention in target tissues than PDGFB-GFP-H6 (Fig. 4B,C). to and penetrate target cells upon selective attachment to PDGFR- The immunohistochemistry detection of the material in the tumor β . A further confirmation of selectivity was done by immunode- and off-target main organs confirmed the high tumor accumula- tecting GFP in CAFs (MEFs, displaying PDGFRβ ) but not in tumor tion and CAFs tumor-selectivity of PDGFD-GFP-H6, which was ab- (MC38, not displaying the receptor) cell lines exposed to PDGFD- sent in liver and kidneys (Fig. 4D). In contrast, significant amounts GFP-H6 (Fig. S6). of PDGFB-GFP-H6 nanoparticles, whose internalization into PDGFR- To evaluate the in vivo performance of PDGFB-GFP-H6 and β + fibroblasts was moderate (Figs. 2, 4A,B), were unexpectedly re- PDGFD-GFP-H6 nanoparticles, C57BL/6 mice were subcutaneously tained in the kidneys (Fig. 4D). This observation suggested that implanted with the syngeneic colorectal cell line MC38 and MEFs PDGFB-GFP-H6, even when showing a promising selectivity in cell 549 E. Voltà-Durán, L. Alba-Castellón, N. Serna et al. Acta Biomaterialia 170 (2023) 543–555 Fig. 4. Biodistribution and in vivo uptake of PDGFB-GFP-H6 and PDGFD-GFP-H6. (A) Immunohistochemistry detection of PDGFR-β within tumors. (B) GFP fluorescence detection by IVIS-Spectrum in tumors upon 2 h after PDGFB-GFP-H6 or PDGFD-GFP-H6 administration. (C) Quantification of fluorescence in buffer, PDGFB-GFP-H6 or PDGFD- GFP-H6 treated tumors n=4. (D) Immunohistochemistry against GFP to detect CAFs-targeting nanoparticle in tumor, kidney, and liver of treated mice. ∗ p ≤0.05. Bars size: 50 μm (A,D) and 5 mm (B). 550 E. Voltà-Durán, L. Alba-Castellón, N. Serna et al. Acta Biomaterialia 170 (2023) 543–555 Fig. 5. Antitumoral activity of a PDGFR-β targeted protein nanotoxin. (A) Modular disposition of the nanotoxin PDGFD-NT-H6, from N-terminal (left) to C-terminal (right), based on the catalytic domain of Pseudomonas aeruginosa exotoxin A (PE24), whose sequence is depicted in blue. The modular protein also contains the targeting domain PDGFD, flexible linkers (grey boxes), a furin cleavable site (FCS, in black), a H6 tag (orange box) and a subcellular location signal (KDEL tetrapeptide). (B) Three-dimensional model of PDGFD-NT-H6, obtained by AlphaFold approach. Color code as in panel A. (C) Particle size of PDGFD-NT-H6 determined by DLS. Number indicates the mean peak size. Additional data can be found in Table S2. (D) Viability of MSCs and MC38 exposed to 100 nM PDGFD-NT-H6 for 48 h. Untreated cells viability was set as 100 %. (E) Schematic representation of experimental design to assess antitumor activity in vivo. (F) In vivo measurements of tumor volume throughout the experiment. (G) Tumor volume measured ex vivo at the end point after cleaning. (H) Haematoxylin-eosin staining of representative images of necrotic area (dotted line). The scale bars represent 100 μm. (I) Quantification of the necrotic area observed in the tumors of both experimental groups, shown at (H). (J). Mice weight as recorded throughout the experiment. Values are presented as the mean ± SEM. Statistical analysis was performed using the Student’s t-test (D) or Mann-Whitney (F,G) ∗ p<0.05, ∗ ∗ ∗ p ≤0.001. Tumor volume over time was analysed comparing areas under the curves (AUC). culture (Fig. 3) might be unsuitable as a PDGFR-β + CAFs-targeting and retention (EPR) effect [89]. The assembled PDGFD-NT-H6 pro- agent since its biodistribution was not satisfactory. Although it moted a significant decrease in the viability of cultured MSCs (but could be not fully discarded that oligomerization might impair re- not MC-38 tumor cells) upon incubation for 48 h at 1×10−7 M ceptor binding, this fact has been not observed in any of the simi- (Fig. 5D). Therefore, an in vivo experiment was designed to assess lar modular constructs previously developed for tumor cell target- the antitumor activity of the construct (Fig. 5E). C57BL/6 mice were ing in vivo, in which the ligand is displayed at the solvent-exposed subcutaneously implanted with the syngeneic colorectal cell line, N-terminus of GFP [81]. However, it has been indeed described MC38, and MEFs to generate a convenient cancer model. The drug that the size of the oligomers or the oligomerization status might prototype was intravenously administered to 1 mg/kg of PDGFD- deeply affect biodistribution of protein materials with properly dis- NT-H6, daily for up to 7 doses (Fig. 5E). Interestingly, the tumor played cell surface ligands [82]. No signs of systemic toxicity were growth was significantly reduced (Fig. 5F,G), which was associated observed in any of the cases (data not shown). with induction of necrosis (Fig. 5H,I). The histopathological anal- To test the ability of PDGFD to deliver antitumoral functional ysis revealed an increase in the necrotic area within the tumor of domains to PDGFR-β + CAFs, a targeted PDGFD-based nanotoxin PDGFD-NT-H6-treated animals as compared with the buffer-treated was designed (PDGFD-NT-H6, Fig. 5A), containing the catalytic do- control set (Fig. 5H). Interestingly, no signs of side toxicities were main of Pseudomonas aeruginosa exotoxin A (PE24). This protein observed and in agreement, the body weight of the treated animals segment is highly cytotoxic and it has been previously used with evolved as the control group (Fig. 5J). In agreement, the histological success in other types of targeted antitumor protein nanoparti- examination of liver and kidneys in treated animals revealed again cles directed to cancer cells [83,84]. In fact, PE24 has been clini- absence of undesired acute toxicities (Fig. S7A), a fact that was fur- cally approved for the treatment of hairy cell leukemia (moxetu- ther confirmed by unaffected levels of relevant blood markers (Fig. momab pasudotox) [85]. Upon recombinant production and purifi- S7B). These data robustly demonstrate the antitumor capability of cation performed as for PDGFD-GFP-H6 (Supplementary Table 1), PDGFD-NT-H6 through the targeting and destruction of PDGFR-β + the toxin-based multi-domain protein (Fig. 5B) rendered nanopar- CAFs in vivo, that was highly selective as demonstrated by the cor- ticles of about 44 nm in diameter (Fig. 5C), coincident with the rect mouse weight evolution. value (around 50 nm) determined as optimal for endosomal for- In summary, by addressing the need to develop non-toxic, non- mation and receptor-mediated endocytosis [86–88] and within the xenobiotic, and highly selective materials for delivery into tumor- size range of the materials suited for the enhanced permeability associated PDGFR-β + CAFs, we have constructed and characterized 551 E. Voltà-Durán, L. Alba-Castellón, N. Serna et al. Acta Biomaterialia 170 (2023) 543–555 a set of 4 modular proteins, targeting such receptor via alterna- spectrum of therapeutic possibilities among which the use of fused tive peptidic ligands that organize as stable and regular oligomeric protein drugs has been demonstrated here. In particular, chemi- nanoparticles (Fig. 1). Ligand-carrying protein oligomers benefit cally attaching conventional small molecular weight drugs to cell- from multiple surface presentations of the ligand in a virus-like targeted protein-only nanoparticles (based on modular constructs fashion [90] and from the cooperativity in the receptor binding similar to those developed here) is fully feasible and it has ren- and endosome formation [91–93], that results in the final cyto- dered interesting prototypes under preclinical development for dif- plasmic delivery of the engulfed material. Being protein-based de- ferent types of cancers [27,99,101]. Here, nanoparticles, apart from velopment of cell-targeted nanoparticles a matter of assay and er- providing targeting, enlarge the drug size over the renal cut-off ror, PDGFRP1-GFP-H6 and Z09591-GFP-H6 particles failed in an ef- and prevent a fast clearance through the kidneys [26]. Importantly, ficient and selective cell binding (Fig. 2). Their small size around the selective presence of PDGFD-GFP-H6 in tumor (Fig. 4) and spe- 9 nm (Figs. 1, S2) probably contributed to this issue, as larger cially, the absence of side toxicities observed when administering nanoparticles ranging from ∼ 15 to 80 nm have been repeat- the highly cytotoxic PDGFD-NT-H6 (Figs. 5, S7) suggest a high level edly observed to be as optimal for receptor-dependent cell up- of accumulation. Although experimental data regarding this issue take [8,9,94–97]. Noteworthy, the introduction of these ligands did is still missing, similar protein constructs targeted to metastasis not promote detectable toxicity upon exposure to target cells (Fig. cancer stem cells resulted in accumulation levels between 60 and S5) or off-target toxicity in vivo (Figs. 5J, S7). Both PDGFB-GFP-H6 86 % of the material detectable in the body [29,30], much over the and PDGFD-GFP-H6 nanoparticles, ranging from 16 to 26 nm, se- conventional 1-2 % commonly observed in targeted nanomedicines lectively targeted and entered PDGFR-β + CAFs (Fig. 3), maintain- [21]. Although needing experimental confirmation, the results pre- ing the fluorescence of the integrated GFP (Figs. 2–4). However, sented here could be in the line of such high values. In addi- PDGFB-GFP-H6, upon in vivo administration in a convenient mouse tion, the set of results generated in the study totally supports the cancer model, did not show a good biodistribution, being mainly generic concept of proteins as editable, versatile and biodegradable found in the kidney (Fig. 4). Despite its good receptor targeting and building blocks suited to construct nanoscale oligomers for clinical selectivity in cell culture (Fig. 3), the protein was unable to per- purposes. Among other materials under exploration such as lipids, form well in vivo regarding the desired biodistribution. This might carbon nanotubes, ceramics, metals, and non-protein polymers, be attributed to its slight structural instability, as determined by that might pose concerns regarding cell toxicity and permanence DLS and SEC, as this protein construct is separated into populations in the media [102–108], proteins, as main biological molecules, are of distinct sizes (Table S2, Fig. S3C). In contrast, PDGFD-GFP-H6 building blocks convenient for the biofabrication of nanomaterials. resulted in nanoparticles with higher stability and relatively low polydispersion, with a convenient size of around 26 nm and high 4. Conclusion potency for receptor-dependent specific cell uptake (Figs. 2, 3). Al- though a small fraction of this protein tends to form small soluble Taking concepts from molecular interactomics and protein aggregates under specific in vitro analytical conditions (Fig. S3D), materials, the modular proteins PDGFD-GFP-H6 and PDGFD-NT- this fact did not impair protein functionality in cell culture or in H6 have been designed to self-assemble as protein-only, homo- vivo. oligomeric nanoparticles. These materials specifically bind and in- In fact, this nanoparticle was the candidate most efficiently in- ternalize PDGFR-β + stromal fibroblasts, the non-cancer cell type ternalized by target cells (Fig. 2), showing an exquisite biodistribu- that supports growth, progression and dissemination of high- tion in mice (Fig. 4), with a very selective accumulation in tumoral indicence solid tumors. The delivered GFP or NT (the microbial tissues. This construct was able to target, penetrate and selectively toxin PE24) are functional upon internalization and either en- make PDGFR-β + CAFs fluorescent (Fig. 4) because of the integrated lighten or destroy target cells depending on the functional protein GFP (Fig. 1), that remained bioactive despite the multiple biolog- incoreporated to the building block. The toxin-mediated cell de- ical barriers that the construct surpassed after their intravenous struction in vivo results in significant tumor growth reduction and administration. The substitution of GFP by a cytotoxic protein drug necrosis, validating the developed platform as promising in oncol- approved in clinics (Fig. 5A,B) equally resulted into stable nanopar- ogy. The clinical relevance of targeting cancer-associated fibroblasts ticles (Fig. 5C) that upon in vivo administration, promoted an im- and the previous unavailability of molecular tools in this regard, portant reduction in tumor growth and induction of necrosis inthe support the proposed concepts as applicable in the molecular de- target tissue (Fig. 5F–G), thus validating the developed platform livery, imaging, therapy and theragnostics of cancer by a so far ne- as a potent and original approach for the generation of anticancer glected and unconventional strategy. drugs. According to the results shown in the present study, PDGFD- Declaration of Competing Interest GFP-H6 (and the derived cytotoxic construct) has been fully vali- dated, among the set of starting candidates, as a potent prototype The authors declare the following financial interests/personal for molecular delivery, imaging and theranostics in cancer through relationships which may be considered as potential competing in- the so far unexplored targeting of CAFs. When a cytotoxic protein terests: Antonio Villaverde has patent #NANOCONJUGATES CON- is incorporated into the constructs (Fig. 5A), treated animals un- TAINING PDGFR-beta LIGANDS AND USES THEREOF. EP22383103 dergo a significant reduction of tumor growth (Fig. 5F) linked to pending to no. induction of necrosis (Fig. 5H,I). This must be attributed to a selec- tive destruction of target CAFs, since the tumor cells used to gen- CRediT authorship contribution statement erate the animal model (MC38) do not internalize the nanoparticle (Figs. 5D, S6). Eric Voltà-Durán: Conceptualization, Methodology, Formal Importantly, this is the first report of the bacterial produc- analysis, Investigation, Writing – review & editing. Lorena Alba- tion and use of PDGFD as a targeting ligand, a fact that opens Castellón: Conceptualization, Methodology, Formal analysis, Inves- a spectrum of opportunities for clinical actions. In this context, tigation, Writing – review & editing. Naroa Serna: Methodology, it was worthy to stress that protein nanoparticles resulting from Formal analysis, Investigation. Isolda Casanova: Methodology, For- oligomeric self-assembly can be functionalized through intrinsic, mal analysis, Investigation. Hèctor López-Laguna: Methodology, biologically active protein domains [98], by chemically coupling Formal analysis, Investigation. Alberto Gallardo: Methodology, For- drugs [99], or through combining both strategies [100], opening a mal analysis, Investigation. Alejandro Sánchez-Chardi: Methodol- 552 E. Voltà-Durán, L. Alba-Castellón, N. Serna et al. Acta Biomaterialia 170 (2023) 543–555 ogy, Formal analysis, Investigation. Antonio Villaverde: Project ad- [7] I. Casanova, U. Unzueta, I. Arroyo-Solera, M.V. Cespedes, A. Villaverde, ministration, Supervision, Resources, Writing – original draft, Writ- R. Mangues, E. Vazquez, Protein-driven nanomedicines in oncotherapy, Curr. Opin. Pharmacol. 47 (2019) 1–7. ing – review & editing. Ugutz Unzueta: Conceptualization, Super- [8] M.J. Mitchell, M.M. Billingsley, R.M. Haley, M.E. Wechsler, N.A. Peppas, vision, Resources, Writing – review & editing. Esther Vázquez: R. Langer, Engineering precision nanoparticles for drug delivery, Nat. Rev. Conceptualization, Project administration, Supervision, Resources, Drug Discov. 20 (2021) 101–124. [9] S. Raj, S. Khurana, R. Choudhari, K.K. Kesari, M.A. Kamal, N. Garg, J. Ruoko- Writing – review & editing. Ramón Mangues: Conceptualization, lainen, B.C. Das, D. Kumar, Specific targeting cancer cells with nanoparticles Project administration, Supervision, Resources, Writing – review & and drug delivery in cancer therapy, Semin. Cancer Biol. 69 (2019) 166–177. editing. [10] M.S. Lee, E.C. Dees, A.Z. Wang, Nanoparticle-delivered chemotherapy: old drugs in new packages, Oncology 31 (2017) 198–208. [11] A. David, Peptide ligand-modified nanomedicines for targeting cells at the tu- Data availability mor microenvironment, Adv. Drug Deliv. Rev. 119 (2017) 120–142. [12] J. Shi, P.W. Kantoff, R. Wooster, O.C. Farokhzad, Cancer nanomedicine: Raw data can be found at https://doi.org/10.34810/data585. progress, challenges and opportunities, Nat. Rev. Cancer 17 (2017) 20–37. [13] S. Tran, P.J. DeGiovanni, B. Piel, P. Rai, Cancer nanomedicine: a review of re- cent success in drug delivery, Clin. Transl. Med. 6 (2017) 44. Acknowledgements [14] P. Pradeep, P. Kumar, Y.E. Choonara, V. Pillay, Targeted nanotechnologies for cancer intervention: a patent review (2010-2016), Expert Opin. Ther. Pat. 27 (2017) 1005–1019. The authors are indebted to Agencia Española de In- [15] O. Cano-Garrido, P. Álamo, L. Sánchez-García, A. Falgàs, A. Sánchez-Chardi, vestigación (AEI) for granting projects on the construction N. Serna, E. Parladé, U. Unzueta, M. Roldán, E. Voltà-Durán, I. Casanova, A. Villaverde, R. Mangues, E. Vázquez, Biparatopic protein nanoparticles for of protein materials of clinical interest (PID2019-105416RB- the precision therapy of CXCR4+ cancers, Cancers 13 (2021) 2929. I00/AEI/10.13039/501100011033 and PDC2022-133858I00/ [16] M. Srinivasarao, C.V. Galliford, P.S. Low, Principles in the design of ligand– AEI/10.13039/50110 0 011033/ Unión Europea NextGenera- targeted cancer therapeutics and imaging agents, Nat. Rev. Drug Discov. 14 tionEU/PRTR to EV, and PID2020-116174RB-I00 to AV), to AGAUR (2015) 203–219. [17] K.C. Tjandra, P. Thordarson, Multivalency in drug delivery–when is it too for projects SGR2021 0 0 092 to AV and SGR2021 01140 to RM, much of a good thing? Bioconjug. Chem. 30 (2019) 503–514. and to ISCIII for projects P21/00150 to RM and PI20/0 040 0 [18] Z. Xu, U. Unzueta, M. Roldán, R. Mangues, A. Sánchez-Chardi, N. Ferrer-Mi- to UU co-funded by European Regional Development Fund ralles, A. Villaverde, E. Vázquez, Formulating tumor-homing peptides as reg- ular nanoparticles enhances receptor-mediated cell penetrability, Mater. Lett. (ERDF, “a way to make Europe”). We also appreciate the sup- 154 (2015) 140–143. port from CIBER -Consorcio Centro de Investigación Biomédica [19] J. Cossu, F. Thoreau, D. Boturyn, Multimeric RGD-based strategies for selective en Red- (CB06/01/0014), Instituto de Salud Carlos III, Ministe- drug delivery to tumor tissues, Pharmaceutics 15 (2023) 525. [20] N. Brabez, K. Saunders, K.L. Nguyen, T. Jayasundera, C. Weber, R.M. Lynch, rio de Ciencia e Innovación and CERCA Programme (Generalitat G. Chassaing, S. Lavielle, V.J. Hruby, Multivalent interactions: synthesis and de Catalunya). Nanoparticle biodistribution was assessed in evaluation of melanotropin multimers–tools for melanoma targeting, ACS the Nanotoxicology Unit of the ICTS-141007 Nanbiosis Platform Med. Chem. Lett. 4 (2013) 98–102. [21] R. Duncan, R. Gaspar, Nanomedicine(s) under the microscope, Mol. Pharm. 8 (http://www.nanbiosis.es/portfolio/u18-nanotoxicology-unit/). (2011) 2101–2141. U.U. was supported by a Miguel Servet contract (CP19/0 0 028) [22] S. Wilhelm, A.J. Tavares, Q. Dai, S. Ohta, J. Audet, H.F. Dvorak, W.C.W. Chan, from ISCIII co-funded by European Social Fund (ESF investing Analysis of nanoparticle delivery to tumours, Nat. Rev. Mater. 1 (2016) 16014. [23] E. Blanco, H. Shen, M. Ferrari, Principles of nanoparticle design for overcom- in your future). A.V. received an ICREA ACADEMIA award. E.V.D ing biological barriers to drug delivery, Nat. Biotechnol. 33 (2015) 941–951. was supported by a predoctoral fellowship from Ministerio de [24] L. Yildirimer, N.T. Thanh, M. Loizidou, A.M. Seifalian, Toxicology and clinical Ciencia e Innovación, Spain (FPU18/04615), L.A-C by a Postodoc- potential of nanoparticles, Nano Today 6 (2011) 585–607. toral fellowship from AECC (Spanish Association Against Cancer, [25] N. Lewinski, V. Colvin, R. Drezek, Cytotoxicity of Nanoparticles, Small 4 (2008) 26–49. POSTD20070ALBA) and H.L.L by a predoctoral fellowship from [26] M.V. Cespedes, U. Unzueta, W. Tatkiewicz, A. Sanchez-Chardi, O. Conchillo– AGAUR (2019FI_B00352). Molecular graphics and analyses per- Sole, P. Alamo, Z. Xu, I. Casanova, J.L. Corchero, M. Pesarrodona, J. Cedano, formed with UCSF ChimeraX, developed by the Resource for X. Daura, I. Ratera, J. Veciana, N. Ferrer-Miralles, E. Vazquez, A. Villaverde, R. Mangues, In vivo architectonic stability of fully de novo designed pro- Biocomputing, Visualization, and Informatics at the University of tein-only nanoparticles, ACS Nano 8 (2014) 4166–4176. California, San Francisco, with support from National Institutes of [27] M.V. Céspedes, U. Unzueta, A. Aviñó, A. Gallardo, P. Álamo, R. Sala, Health R01-GM129325 and the Office of Cyber Infrastructure and A. Sánchez-Chardi, I. Casanova, M.A. Mangues, A. Lopez-Pousa, R. Eritja, A. Villaverde, E. Vázquez, R. Mangues, Selective depletion of metastatic stem Computational Biology, National Institute of Allergy and Infectious cells as therapy for human colorectal cancer, EMBO Mol. Med. 10 (2018) Diseases. The authors want to thank Clara Seira for her technical e8772. assistance. [28] U. Unzueta, M.V. Céspedes, N. Ferrer-Miralles, I. Casanova, J. Cedano, J.L. Corchero, J. Domingo-Espín, A. Villaverde, R. Mangues, E. Vázquez, Intra- cellular CXCR4+ cell targeting with T22-empowered protein-only nanoparti- Supplementary materials cles, Int. J. Nanomed. 7 (2012) 4533–4544. [29] A. Falgàs, V. Pallarès, U. Unzueta, M.V. Céspedes, I. Arroyo-Solera, M.J. Moreno, J. Sierra, A. Gallardo, M.A. Mangues, E. Vázquez, A. Villaverde, R. Mangues, Supplementary material associated with this article can be I. Casanova, A CXCR4-targeted nanocarrier achieves highly selective tumor found, in the online version, at doi:10.1016/j.actbio.2023.09.001. uptake in diffuse large B-cell lymphoma mouse models, Haematologica 105 (2020) 741–753. [30] E. Medina-Gutiérrez, M.V. Céspedes, A. Gallardo, E. Rioja-Blanco, M.À. Pavón, References L. Asensio-Puig, L. Farré, L. Alba-Castellón, U. Unzueta, A. Villaverde, E. Vázquez, I. Casanova, R. Mangues, Novel endometrial cancer models using [1] H. Sung, J. Ferlay, R.L. Siegel, M. Laversanne, I. Soerjomataram, A. Jemal, sensitive metastasis tracing for CXCR4-targeted therapy in advanced disease, F. Bray, Global Cancer Statistics 2020: GLOBOCAN estimates of incidence and Biomedicines 10 (2022) 1680. mortality worldwide for 36 cancers in 185 countries, CA Cancer J. Clin. 71 [31] N. Serna, L. Sanchez-Garcia, U. Unzueta, R. Diaz, E. Vazquez, R. Mangues, (2021) 209–249. A. Villaverde, Protein-based therapeutic killing for cancer therapies, Trends [2] A. Kamb, S. Wee, C. Lengauer, Why is cancer drug discovery so difficult? Nat. Biotechnol. 36 (2018) 318–335. Rev. Drug Discov. 6 (2007) 115–120. [32] A. Santi, F.G. Kugeratski, S. Zanivan, Cancer associated fibroblasts: the archi- [3] A.F. Costa, D. Campos, C.A. Reis, C. Gomes, Targeting glycosylation: a new road tects of stroma remodeling, Proteomics 18 (2018) 1700167. for cancer drug discovery, Trends Cancer 6 (2020) 757–766. [33] D.C. Hinshaw, L.A. Shevde, The tumor microenvironment innately modulates [4] B.A. Carneiro, W.S. El-Deiry, Targeting apoptosis in cancer therapy, Nat. Rev. cancer progression, Cancer Res. 79 (2019) 4557–4566. Clin. Oncol. 17 (2020) 395–417. [34] R. Kalluri, The biology and function of fibroblasts in cancer, Nat. Rev. Cancer [5] F. McCormick, Cancer gene therapy: fringe or cutting edge? Nat. Rev. Cancer 16 (2016) 582–598. 1 (2001) 130–141. [35] G. Hu, L. Huang, K. Zhong, L. Meng, F. Xu, S. Wang, T. Zhang, PDGFR-beta(+) [6] B.B.S. Zhou, H. Zhang, M. Damelin, K.G. Geles, J.C. Grindley, P.B. Dirks, Tu- fibroblasts deteriorate survival in human solid tumors: a meta-analysis, Aging mour-initiating cells: challenges and opportunities for anticancer drug dis- 13 (2021) 13693–13707. covery, Nat. Rev. Drug Discov. 8 (2009) 806–823. 553 E. Voltà-Durán, L. Alba-Castellón, N. Serna et al. Acta Biomaterialia 170 (2023) 543–555 [36] P.J. Asif, C. Longobardi, M. Hahne, J.P. Medema, The role of cancer-associated [62] V. Askoxylakis, A. Marr, A. Altmann, A. Markert, W. Mier, J. Debus, P.E. Huber, fibroblasts in cancer invasion and metastasis, Cancers 13 (2021) 4720. U. Haberkorn, Peptide-based targeting of the platelet-derived growth factor [37] S. Fujino, N. Miyoshi, M. Ohue, Y. Takahashi, M. Yasui, T. Hata, C. Matsuda, receptor beta, Mol. Imaging Biol. 15 (2013) 212–221. T. Mizushima, Y. Doki, M. Mori, Platelet–derived growth factor receptor–β [63] V. Tolmachev, Z. Varasteh, H. Honarvar, S.J. Hosseinimehr, O. Eriksson, gene expression relates to recurrence in colorectal cancer, Oncol. Rep. 39 P. Jonasson, F.Y. Frejd, L. Abrahmsen, A. Orlova, Imaging of platelet-derived (2018) 2178–2184. growth factor receptor beta expression in glioblastoma xenografts using affi- [38] E.J. Steller, D.A. Raats, J. Koster, B. Rutten, K.M. Govaert, B.L. Emmink, N. Sno- body molecule 111In-DOTA-Z09591, J. Nucl. Med. 55 (2014) 294–300 official eren, S.R. van Hooff, F.C. Holstege, C. Maas, I.H. Borel Rinkes, O. Kranenburg, publication, Society of Nuclear Medicine. PDGFRB promotes liver metastasis formation of mesenchymal-like colorectal [64] J. Strand, Z. Varasteh, O. Eriksson, L. Abrahmsen, A. Orlova, V. Tolmachev, Gal- tumor cells, Neoplasia 15 (2013) 204–217. lium-68-labeled affibody molecule for PET imaging of PDGFRβ expression in [39] A.A. Jitariu, M. Raica, A.M. Cîmpean, S.C. Suciu, The role of vivo, Mol. Pharm. 11 (2014) 3957–3964. PDGF-B/PDGFR-BETA axis in the normal development and carcinogene- [65] H. López-Laguna, L. Sánchez-García, N. Serna, E. Voltà-Durán, J.M. Sánchez, sis of the breast, Crit. Rev. Oncol. Hematol. 131 (2018) 46–52. A. Sánchez-Chardi, U. Unzueta, M. Łoś, A. Villaverde, E. Vázquez, Engineering [40] S. Avril, Y. Dincer, K. Malinowsky, C. Wolff, S. Gundisch, A. Hapfelmeier, protein nanoparticles out from components of the human microbiome, Small M. Boxberg, H. Bronger, K.F. Becker, B. Schmalfeldt, Increased PDGFR-beta 16 (2020) e2001885. and VEGFR-2 protein levels are associated with resistance to platinum-based [66] V.D. Nguyen, F. Hatahet, K.E. Salo, E. Enlund, C. Zhang, L.W. Ruddock, Pre-ex- chemotherapy and adverse outcome of ovarian cancer patients, Oncotarget 8 pression of a sulfhydryl oxidase significantly increases the yields of eukary- (2017) 97851–97861. otic disulfide bond containing proteins expressed in the cytoplasm of E.coli, [41] L. Lu, X. Fu, Z. Li, Y. Qiu, W. Li, Z. Zhou, W. Xue, Y. Wang, M. Jin, M. Zhang, Microb. Cell Fact. 10 (2011) 1. Platelet-derived growth factor receptor alpha (PDGFRα ) is overexpressed in [67] G. Veggiani, A. de Marco, Improved quantitative and qualitative production of NK/T-cell lymphoma and mediates cell survival, Biochem. Biophys. Res. Com- single-domain intrabodies mediated by the co-expression of Erv1p sulfhydryl mun. 504 (2018) 525–531. oxidase, Protein Expr. Purif. 79 (2011) 111–114. [42] Y. Nordby, E. Richardsen, M. Rakaee, N. Ness, T. Donnem, H.R.H. Patel, [68] A. Magnusdottir, I. Johansson, L.G. Dahlgren, P. Nordlund, H. Berglund, En- L.T. Busund, R.M. Bremnes, S. Andersen, High expression of PDGFR-β in abling IMAC purification of low abundance recombinant proteins from E. coli prostate cancer stroma is independently associated with clinical and bio- lysates, Nat. Methods 6 (2009) 477–478. chemical prostate cancer recurrence, Sci. Rep. 7 (2017) 43378. [69] E. Voltà-Durán, C.G. Olivia, N. Serna, H. López-Laguna, L. Sánchez-García, [43] S. Szubert, R. Moszynski, D. Szpurek, B. Romaniuk, S. Sajdak, M. Now- M. Pesarrodona, A. Sánchez-Chardi, R. Mangues, A. Villaverde, E. Vázquez, icki, S. Michalak, The expression of Platelet-derived Growth factor receptors U. Unzueta, Controlling self-assembling and tumor cell-targeting of pro- (PDGFRs) and their correlation with overall survival of patients with ovarian tein-only nanoparticles through modular protein engineering, Sic. China cancer, Ginekol. Pol. 90 (2019) 242–249. Mater. 63 (2019) 147–156. [44] A. Östman, PDGF receptors in tumor stroma: biological effects and associa- [70] J. Jumper, R. Evans, A. Pritzel, T. Green, M. Figurnov, O. Ronneberger, K. Tun- tions with prognosis and response to treatment, Adv. Drug Deliv. Rev. 121 yasuvunakool, R. Bates, A. Žídek, A. Potapenko, A. Bridgland, C. Meyer, (2017) 117–123. S.A.A. Kohl, A.J. Ballard, A. Cowie, B. Romera-Paredes, S. Nikolov, R. Jain, [45] K.C. Valkenburg, A.E. de Groot, K.J. Pienta, Targeting the tumour stroma to J. Adler, T. Back, S. Petersen, D. Reiman, E. Clancy, M. Zielinski, M. Steinegger, improve cancer therapy, Nat. Rev. Clin. Oncol. 15 (2018) 366–381. M. Pacholska, T. Berghammer, S. Bodenstein, D. Silver, O. Vinyals, A.W. Senior, [46] N. Effendi, K. Mishiro, K. Shiba, S. Kinuya, K. Ogawa, Development of radiogal- K. Kavukcuoglu, P. Kohli, D. Hassabis, Highly accurate protein structure pre- lium-labeled peptides for platelet-derived growth factor receptor β (PDGFRβ ) diction with AlphaFold, Nature 596 (2021) 583–589. imaging: influence of different linkers, Molecules 26 (2021) 41. [71] E.F. Pettersen, T.D. Goddard, C.C. Huang, E.C. Meng, G.S. Couch, T.I. Croll, [47] Y. Yang, Y. Deng, X. Chen, J. Zhang, Y. Chen, H. Li, Q. Wu, Z. Yang, L. Zhang, J.H. Morris, T.E. Ferrin, UCSF ChimeraX: structure visualization for researchers, B. Liu, Inhibition of PDGFR by CP-673451 induces apoptosis and increases cis- educators, and developers, Protein Sci. 30 (2021) 70–82. platin cytotoxicity in NSCLC cells via inhibiting the Nrf2-mediated defense [72] R. Batlle, L. Alba-Castellón, J. Loubat-Casanovas, E. Armenteros, C. Francí, mechanism, Toxicol. Lett. 295 (2018) 88–98. J. Stanisavljevic, R. Banderas, J. Martin-Caballero, F. Bonilla, J. Baulida, [48] N. Papadopoulos, J. Lennartsson, The PDGF/PDGFR pathway as a drug target, J.I. Casal, T. Gridley, A. García de Herreros, Snail1 controls TGF-β respon- Mol. Aspects Med. 62 (2018) 75–88. siveness and differentiation of mesenchymal stem cells, Oncogene 32 (2013) [49] N. Serna, V. Pallarès, U. Unzueta, A. Garcia-Leon, E. Voltà-Durán, 3381–3389. A. Sánchez-Chardi, E. Parladé, A. Rueda, I. Casanova, A. Falgàs, L. Alba-Castel- [73] L. Alba-Castellón, R. Olivera-Salguero, A. Mestre-Farrera, R. Peña, M. Her- lón, J. Sierra, A. Villaverde, E. Vázquez, R. Mangues, Engineering non-antibody rera, F. Bonilla, J.I. Casal, J. Baulida, C. Peña, A. García de Herreros, human proteins as efficient scaffolds for selective, receptor-targeted drug Snail1-dependent activation of cancer-associated fibroblast controls epithelial delivery, J. Control. Release 343 (2022) 277–287. tumor cell invasion and metastasis, Cancer Res. 76 (2016) 6205–6217. [50] H. Lopez-Laguna, E. Volta-Duran, E. Parlade, A. Villaverde, E. Vazquez, U. Un- [74] B. Grum-Schwensen, J. Klingelhofer, C.H. Berg, C. El-Naaman, M. Grigorian, zueta, Insights on the emerging biotechnology of histidine-rich peptides, E. Lukanidin, N. Ambartsumian, Suppression of tumor development and Biotechnol. Adv. 54 (2022) 107817. metastasis formation in mice lacking the S100A4(mts1) gene, Cancer Res. 65 [51] E. Parladé, E. Voltà-Durán, O. Cano-Garrido, J.M. Sánchez, U. Unzueta, (2005) 3772–3780. H. López-Laguna, N. Serna, M. Cano, M. Rodríguez-Mariscal, E. Vazquez, [75] A. Noel, M.C. De Pauw-Gillet, G. Purnell, B. Nusgens, C.M. Lapiere, J.M. Foidart, A. Villaverde, An in silico methodology that facilitates decision making in the Enhancement of tumorigenicity of human breast adenocarcinoma cells in engineering of nanoscale protein Materials, Int. J. Mol. Sci. 23 (2022) 4958. nude mice by matrigel and fibroblasts, Br. J. Cancer 68 (1993) 909–915. [52] S.T. Wang, B. Minevich, J. Liu, H. Zhang, D. Nykypanchuk, J. Byrnes, W. Liu, [76] D. Jelinek, E.R. Zhang, A. Ambrus, E. Haley, E. Guinn, A. Vo, P. Le, A.E. Kesaf, L. Bershadsky, Q. Liu, T. Wang, G. Ren, O. Gang, Designed and biologically J. Nguyen, L. Guo, D. Frederick, Z. Sun, N. Guo, P. Sevier, E. Bilotta, K. Atai, active protein lattices, Nat. Commun. 12 (2021) 3702. L. Voisin, H.A. Coller, A mouse model to investigate the role of cancer-associ- [53] H. López-Laguna, E. Voltà-Durán, E. Parladé, A. Villaverde, E. Vázquez, U. Un- ated fibroblasts in tumor growth, J. Vis. Exp. JoVE 166 (2020) e61883. zueta, Insights on the emerging biotechnology of histidine-rich peptides, [77] A. Herrera, M. Herrera, L. Alba-Castellon, J. Silva, V. Garcia, J. Loubat– Biotechnol. Adv. 54 (2021) 107817. Casanovas, A. Alvarez-Cienfuegos, J. Miguel Garcia, R. Rodriguez, B. Gil, C. Ma [54] H. Lopez-Laguna, J.M. Sanchez, J.V. Carratala, M. Rojas-Pena, L. Sanchez-Gar- Jesus, L. Ma Jesus, J. Ignacio Casal, A.G. de Herreros, F. Bonilla, C. Pena, Protu- cia, E. Parlade, A. Sanchez-Chardi, E. Volta-Duran, N. Serna, O. Cano-Gar- morigenic effects of Snail-expression fibroblasts on colon cancer cells, Int. J. rido, S. Flores, N. Ferrer-Miralles, V. Nolan, A. de Marco, N. Roher, U. Un- Cancer 134 (2014) 2984–2990. zueta, E. Vazquez, A. Villaverde, Biofabrication of functional protein nanopar- [78] H. López-Laguna, J. Sánchez, U. Unzueta, R. Mangues, E. Vázquez, ticles through simple His-tag engineering, ACS Sustain. Chem. Eng. 9 (2021) A. Villaverde, Divalent cations: a molecular glue for protein materials, Trends 12341–12354. Biochem. Sci. 45 (2020) 992–1003. [55] S. Gopalakrishnan, J. Xu, F. Zhong, V.M. Rotello, Strategies for fabricating pro- [79] U. Unzueta, N. Ferrer-Miralles, J. Cedano, X. Zikung, M. Pesarrodona, P. Sac- tein films for biomaterial applications, Adv. Sustain. Syst. 5 (2021) 20 0 0167. cardo, E. Garcia-Fruitos, J. Domingo-Espin, P. Kumar, K.C. Gupta, R. Mangues, [56] J. Shim, C. Zhou, T. Gong, D.A. Iserlis, H.A. Linjawi, M. Wong, T. Pan, C. Tan, A. Villaverde, E. Vazquez, Non-amyloidogenic peptide tags for the regulat- Building protein networks in synthetic systems from the bottom-up, Biotech- able self-assembling of protein-only nanoparticles, Biomaterials 33 (2012) nol. Adv. 49 (2021) 107753. 8714–8722. [57] A. Korpi, E. Anaya-Plaza, S. Valimaki, M. Kostiainen, Highly ordered protein [80] E. Voltà-Durán, J.M. Sánchez, L. Hèctor López, E. Parladé, L. Sánchez-García, cage assemblies: a toolkit for new materials, Wiley Interdiscip. Rev. Nanomed. A. Sánchez-Chardi, A. de Marco, U. Unzueta, E. Vázquez, A. Villaverde, The Nanobiotechnol. 12 (2020) e1578. spectrum of building block conformers sustains the biophysical properties of [58] R.V. Ulijn, A. Lampel, Order/disorder in protein and peptide-based biomateri- clinically-oriented self-assembling protein nanoparticles, Sci. China Mater. 65 als, Isr. J. Chem. 60 (2020) 1129. (2022) 1662–1670. [59] W.A. Hansen, S.D. Khare, Recent progress in designing protein-based [81] F. Rueda, M.V. Céspedes, O. Conchillo-Solé, A. Sánchez-Chardi, J. Seras-Fran- supramolecular assemblies, Curr. Opin. Struct. Biol. 63 (2020) 106–114. zoso, R. Cubarsi, A. Gallardo, M. Pesarrodona, N. Ferrer-Miralles, X. Daura, [60] X. Xu, X. Chen, J. Li, Natural protein bioinspired materials for regeneration of E. Vázquez, E. García-Fruitós, R. Mangues, U. Unzueta, A. Villaverde, Bot- hard tissues, J. Mater. Chem. B 8 (2020) 2199–2215. tom-up instructive quality control in the biofabrication of smart protein ma- [61] A.J. Najy, J.J. Won, L.S. Movilla, H.R. Kim, Differential tumorigenic potential terials, Adv. Mater. 27 (2015) 7816–7822 (Deerfield Beach, Fla). and matriptase activation between PDGF B versus PDGF D in prostate cancer, [82] H. López-Laguna, R. Sala, J.M. Sánchez, P. Álamo, U. Unzueta, Mol. Cancer Res. MCR 10 (2012) 1087–1097. A. Sánchez-Chardi, N. Serna, L. Sánchez-García, E. Voltà-Durán, R. Mangues, 554 E. Voltà-Durán, L. Alba-Castellón, N. Serna et al. Acta Biomaterialia 170 (2023) 543–555 A. Villaverde, E. Vázquez, Nanostructure empowers active tumor targeting in [96] D. Verma, N. Gulati, S. Kaul, S. Mukherjee, U. Nagaich, Protein based nanos- ligand-based molecular delivery, Part. Part. Syst. Charact. 36 (2019) 1900304. tructures for drug delivery, J. Pharm. 2018 (2018) 9285854. [83] Y. Núñez, A. Garcia-León, A. Falgàs, N. Serna, L. Sánchez-García, A. Garrido, [97] L. Shang, K. Nienhaus, G.U. Nienhaus, Engineered nanoparticles interacting J. Sierra, A. Gallardo, U. Unzueta, E. Vázquez, A. Villaverde, R. Mangues, with cells: size matters, J. Nanobiotechnol. 12 (2014) 5. I. Casanova, T22-PE24-H6 nanotoxin selectively kills CXCR4-high expressing [98] E. Rioja-Blanco, I. Arroyo-Solera, P. Alamo, I. Casanova, A. Gallardo, U. Un- AML patient cells in vitro and potently blocks dissemination in vivo, Pharma- zueta, N. Serna, L. Sanchez-Garcia, M. Quer, A. Villaverde, E. Vazquez, X. Leon, ceutics 15 (2023) 727. L. Alba-Castellon, R. Mangues, CXCR4-targeted nanotoxins induce GSDME-de- [84] Z. Zhao, Y. Huang, J. Wang, H. Lin, F. Cao, S. Li, Y. Li, Z. Li, X. Liu, A self-assem- pendent pyroptosis in head and neck squamous cell carcinoma, J. Exp. Clin. bling CXCR4-targeted pyroptosis nanotoxin for melanoma therapy, Biomater. Cancer res. CR 41 (2022) 49. Sci. 11 (2023) 2200–2210. [99] A. Falgàs, V. Pallarès, U. Unzueta, Y. Núñez, J. Sierra, A. Gallardo, L. Alba– [85] C.F. Nobre, M.J. Newman, A. DeLisa, P. Newman, Moxetumomab pasudo- Castellón, M.A. Mangues, P. Álamo, A. Villaverde, E. Vázquez, R. Mangues, tox-tdfk for relapsed/refractory hairy cell leukemia: a review of clinical con- I. Casanova, Specific cytotoxic effect of an auristatin nanoconjugate towards siderations, Cancer Chemother. Pharmacol. 84 (2019) 255–263. CXCR4(+) diffuse large B-cell lymphoma Cells, Int. J. Nanomed. 16 (2021) [86] S. Zhang, J. Li, G. Lykotrafitis, G. Bao, S. Suresh, Size-dependent endocytosis of 1869–1888. nanoparticles, Adv. Mater. 21 (2009) 419–424 (Deerfield Beach, Fla). [100] E. Volta-Duran, N. Serna, L. Sanchez-Garcia, A. Avino, J.M. Sanchez, [87] H. Yuan, S. Zhang, Effects of particle size and ligand density on the kinetics H. Lopez-Laguna, O. Cano-Garrido, I. Casanova, R. Mangues, R. Eritja, of receptor-mediated endocytosis of nanoparticles, Appl. Phys. Lett. 96 (2010) E. Vazquez, A. Villaverde, U. Unzueta, Design and engineering of tumor-tar- 033704. geted, dual-acting cytotoxic nanoparticles, Acta Biomater. 119 (2021) 312–322. [88] Banerjee A., Berzhkovskii A., Nossal R. Efficiency of cellular uptake [101] V. Pallarès, U. Unzueta, A. Falgàs, A. Aviñó, Y. Núñez, A. García-León, of nanoparticles via receptor-mediated endocytosis. arXiv preprint L. Sánchez-García, N. Serna, A. Gallardo, L. Alba-Castellón, P. Álamo, J. Sierra, arXiv:14117348 2014. L. Cedó, R. Eritja, A. Villaverde, E. Vázquez, I. Casanova, R. Mangues, A mul- [89] M. Ikeda-Imafuku, L.L.W. Wang, D. Rodrigues, S. Shaha, Z. Zhao, S. Mitragotri, tivalent Ara-C-prodrug nanoconjugate achieves selective ablation of leukemic Strategies to improve the EPR effect: a mechanistic perspective and clinical cells in an acute myeloid leukemia mouse model, Biomaterials 280 (2022) translation, J. Control. Release 345 (2022) 512–536. 121258. [90] U. Unzueta, M.V. Cespedes, E. Vazquez, N. Ferrer-Miralles, R. Mangues, [102] W. Yang, L. Wang, E.M. Mettenbrink, P.L. DeAngelis, S. Wilhelm, Nanoparticle A. Villaverde, Towards protein-based viral mimetics for cancer therapies, toxicology, Annu. Rev. Pharmacol. Toxicol. 61 (2021) 269–289. Trends Biotechnol. 33 (2015) 253–258. [103] X. Yuan, X. Zhang, L. Sun, Y. Wei, X. Wei, Cellular toxicity and immunological [91] J.F. Stefanick, D.T. Omstead, T. Kiziltepe, B. Bilgicer, Dual-receptor targeted effects of carbon-based nanomaterials, Part. Fibre Toxicol. 16 (2019) 18. strategy in nanoparticle design achieves tumor cell selectivity through coop- [104] J.B. Raftis, M.R. Miller, Nanoparticle translocation and multi-organ toxicity: a erativity, Nanoscale (2019). particularly small problem, Nano Today 26 (2019) 8–12. [92] J. Reichenwallner, A. Thomas, T. Steinbach, J. Eisermann, C.E.H. Schmelzer, [105] J. Du, J. Tang, S. Xu, J. Ge, Y. Dong, H. Li, M. Jin, A review on silver nanoparti- F. Wurm, D. Hinderberger, Ligand-binding cooperativity effects in poly- cles-induced ecotoxicity and the underlying toxicity mechanisms, Regul. Tox- mer-protein conjugation, Biomacromolecules 20 (2019) 1118–1131. icol. Pharm. 98 (2018) 231–239. [93] Y. Li, Y. Wang, G. Huang, J. Gao, Cooperativity principles in self-assembled [106] X. Li, W. Liu, L. Sun, K.E. Aifantis, B. Yu, Y. Fan, Q. Feng, F. Cui, F. Watari, Effects nanomedicine, Chem. Rev. 118 (2018) 5359–5391. of physicochemical properties of nanomaterials on their toxicity, J. Biomed. [94] R.V. Kalaydina, K. Bajwa, B. Qorri, A. Decarlo, M.R. Szewczuk, Recent advances Mater. Res. Part A 103 (2015) 2499–2507. in "smart" delivery systems for extended drug release in cancer therapy, Int. [107] M. Palombo, M. Deshmukh, D. Myers, J. Gao, Z. Szekely, P.J. Sinko, Pharmaceu- J. Nanomed. 13 (2018) 4727–4745. tical and toxicological properties of engineered nanomaterials for drug deliv- [95] P. Debie, C. Lafont, M. Defrise, I. Hansen, D.M. van Willigen, F.W.B. van ery, Annu. Rev. Pharmacol. Toxicol. 54 (2014) 581–598. Leeuwen, R. Gijsbers, M. D’Huyvetter, N. Devoogdt, T. Lahoutte, P. Mollard, [108] A. Dhawan, V. Sharma, Toxicity assessment of nanomaterials: methods and S. Hernot, Size and affinity kinetics of nanobodies influence targeting and challenges, Anal. Bioanal. Chem. 398 (2010) 589–605. penetration of solid tumours, J. Control. Release 317 (2020) 34–42. 555