Communication www.advancedscience.com Artificial Inclusion Bodies for Clinical Development Julieta M. Sánchez, Hèctor López-Laguna, Patricia Álamo, Naroa Serna, Alejandro Sánchez-Chardi, Verónica Nolan, Olivia Cano-Garrido, Isolda Casanova, Ugutz Unzueta, Esther Vazquez,* Ramon Mangues,* and Antonio Villaverde* protein product plus a diversity of residual Bacterial inclusion bodies (IBs) are mechanically stable protein particles in macromolecules from bacterial cells, the microscale, which behave as robust, slow-protein-releasing amyloids. including nucleic acids, carbohydrates, Upon exposure to cultured cells or upon subcutaneous or intratumor injec- proteins, and cell wall components. Their tion, these protein materials secrete functional IB polypeptides, function- morphology is defined by mechanical lim- itations imposed by the bacterial cell wall, ally mimicking the endocrine release of peptide hormones from secretory usually at the edge of sub-micrometer and amyloid granules. Being appealing as delivery systems for prolonged protein micrometer sizes. The biological activity drug release, the development of IBs toward clinical applications is, how- associated to IB proteins,[4] together with ever, severely constrained by their bacterial origin and by the undefined and the mechanical stability and high porosity protein-to-protein, batch-to-batch variable composition. In this context, the de of these protein particles has pushed to re-evaluate them as unconventional func- novo fabrication of artificial IBs (ArtIBs) by simple, cell-free physicochemical tional materials with a wide spectrum of methods, using pure components at defined amounts is proposed here. By applications in biotechnology and bio- this, the resulting functional protein microparticles are intriguing, chemically medicine.[5] Any functional polypeptide defined biomimetic materials that replicate relevant functionalities of natural suitable for production in bacteria can IBs, including mammalian cell penetration and local or remote release of be engineered to be packaged in form of functional ArtIB-forming protein. In default of severe regulatory issues, the IBs.[6] This is because IBs show a com- plex structure with a dual organization concept of ArtIBs is proposed as a novel exploitable category of biomaterials of the forming polypeptide chains; while for biotechnological and biomedical applications, resulting from simple fabri- around 40% of the IB protein generates cation and envisaging soft developmental routes to clinics. a mechanically stable fibril network, the remaining fraction represents functional or quasi-functional species embedded Bacterial inclusion bodies (IBs) are water-insoluble proteina- in such stable structure.[7] This protein population is properly ceous inclusions generated in the cytoplasm of recombinant folded, nearly soluble, releasable under physiological condi- bacteria,[1,2] stabilized by an amyloid fibril architecture that con- tions, and responsible for the biological activity of IBs.[1] As fers mechanical robustness.[3] They are formed by the transgene self-immobilized catalysts, enzyme-based IBs show high level Dr. J. M. Sánchez, H. López-Laguna, Dr. N. Serna, Dr. O. Cano-Garrido,[+] H. López-Laguna, Dr. N. Serna, Dr. O. Cano-Garrido, Dr. E. Vazquez, Dr. E. Vazquez, Prof. A. Villaverde Prof. A. Villaverde Institut de Biotecnologia i de Biomedicina Departament de Genètica i de Microbiologia Universitat Autònoma de Barcelona Universitat Autònoma de Barcelona Bellaterra, 08193 Barcelona, Spain Bellaterra, 08193 Barcelona, Spain E-mail: Esther.Vazquez@uab.es; antoni.villaverde@uab.cat H. López-Laguna, Dr. P. Álamo, Dr. N. Serna, Dr. I. Casanova, Dr. J. M. Sánchez, Dr. V. Nolan Dr. U. Unzueta, Dr. E. Vazquez, Prof. R. Mangues, Prof. A. Villaverde Instituto de Investigaciones Biológicas y Tecnológicas (IIBYT) CIBER de Bioingeniería (CONICET-Universidad Nacional de Córdoba) Biomateriales y Nanomedicina (CIBER-BBN) ICTA & Cátedra de Química Biológica 28029, Madrid, Spain Departamento de Química E-mail: rmangues@santpau.cat FCEFyN, UNC. Av. Velez Sarsfield 1611, X 5016GCA, Córdoba, Argentina Dr. P. Álamo, Dr. I. Casanova, Dr. U. Unzueta, Prof. R. Mangues The ORCID identification number(s) for the author(s) of this article Biomedical Research Institute Sant Pau (IIB-Sant Pau) and Josep can be found under https://doi.org/10.1002/advs.201902420. Carreras Research Institute Hospital de la Santa Creu i Sant Pau © 2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, 08041, Barcelona, Spain Weinheim. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and repro- Dr. A. Sánchez-Chardi duction in any medium, provided the original work is properly cited. Servei de Microscòpia Universitat Autònoma de Barcelona [+]Presentaddress: Nanoligent SL, Edifici Eureka, Universitat Autònoma Bellaterra, 08193 Barcelona, Spain de Barcelona, Bellaterra, 08193 Barcelona, Spain DOI: 10.1002/advs.201902420 Adv. Sci. 2020, 7, 1902420 1902420 (1 of 7) © 2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com www.advancedscience.com of operational stability and reusability.[8] In tissue engineering, To explore the fabrication of synthetic IBs, common labo- they have been adapted as nontoxic topologies that provide a ratory enzymes that form functional IBs during recombinant combination of mechanical and biological stimuli for controlled production, namely, β-galactosidase (β-Gal) and alkaline phos- cell proliferation.[9] In a different context, IBs have been tailored phatase (AP),[13] were selected as models for two alternative as unexpected drug delivery systems or “nanopills,”[10] that approaches to ArtIB fabrication. In one (Figure 1a), soluble mimic the functionalities of the secretory granules of the mam- protein was salted out plus thermally aggregated to generate malian endocrine system[11] for the intracellular, local or remote amyloidal networks, further used as IB-like seeds to recruit and delivery of functional (either untargeted or receptor-targeted) entrap homologous soluble protein versions. This multiple step IB protein.[12] Therefore, IBs, as functional biomaterials, show procedure (ms), was developed to imitate the dual, sponge-like promise in protein replacement therapies and in any clinical networks in natural IBs.[7] Lipids, common component of bac- uses aimed to systemic, local or precision protein delivery. terial IBs were also incorporated. In a simpler single-step (ss) The enormous clinical potential of bacterial IBs is, however, approach, divalent cations (Zn, in form of ZnCl2), involved in darkened by their heterogeneous and undefined composition.[6] amyloid formation[14] and generically, in protein–protein con- Trapping an indeterminate catalogue of bacterial cell materials, tacts,[15] were added to a protein solution (Figure 1a). The appli- clinically oriented IBs would hardly overcome the severe reg- cation of these procedures resulted in mechanically stable, dis- ulatory constraints imposed by medicament agencies. In this crete, and moderately disperse protein particles sizing around context, we wondered whether chemically pure IB mimetic par- 1–2 µm (AP) and 2–6 µm (β-Gal) (Figure 1b,c), whose surface ticles (artificial IBs, ArtIBs) could be fabricated in vitro. This rugosity, size variability, and amorphous appearance remem- would imply the production of particulate microscale protein bered those of natural IBs (Figure 1b). Both enzymes, in this materials, made of pure and chemically controlled components, packaged form, were enzymatically active (Figure 1d). On the which would replicate those functionalities of natural IBs that other hand, cross-β-sheet amyloid like structure (ALS) was are relevant to protein delivery. These properties are mechan- detected by attenuated total reflectance (ATR) in AP ArtIBs at ical stability, absence of intrinsic cytotoxicity, mammalian cell proportions (31–37%, ss and ms, respectively) matching those penetrability, and the ability to release functional IB protein found in IBs.[7] upon physiological conditions. Pure protein composition would Further, we constructed new ArtIBs (Figure 2a) formed by be added to such functionally rich profile. the self-assembling modular proteins T22-GFP-H6[16] and Figure 1. Fabrication of ArtIBs. a) Multiple (ms) and single (ss) step procedures for ArtIB fabrication from soluble pure protein are summarized, indi- cating the main operational steps (arrows). OS is organic solvent. Precise details can be found in the Experimental Section. Final products are framed. b) Representative field-emission scanning electron microscopy (FESEM) images of AP and β-Gal ArtIBs. Magnifications are equivalent in all images. c) Dynamic light scattering (DLS) size analyses of ArtIBs, indicating the mode (in nm) and the polydispersion index (pdi). d) Specific activity of both AP and β-Gal ArtIBs, compared to that of commercial soluble protein counterpart. Asterisks indicate statistically different from the specific activity of the soluble protein (p < 0.001, Holme–Sidak test). Adv. Sci. 2020, 7, 1902420 1902420 (2 of 7) © 2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com www.advancedscience.com Figure 2. Characterization of ArtIBs formed by modular proteins. a) FESEM images of CXCR4-targeted ArtIBs, all recorded at the same magnification. At the bottom of each image, specific fluorescence decay (SFD), hydrodynamic size peak (pdi ± s.e.m.) and percentage of ALS are shown. b) Internali- zation of T22-GFP-H6 ArtIBs in cultured HeLa cells, recorded at different times after exposure through intracellular green fluorescent protein (GFP) fluorescence (top). Bottom: AMD3100-mediated inhibition of ArtIB internalization. c) Viability of cultured HeLa cells upon 96 h of T22-GFP-H6 and T22-PE24-H6 ArtIB exposure in presence or absence of AMD3100. d) Stain-free protein detection of released soluble protein (r) from ArtIBs, 7 days after incubation in physiological buffer. In the plot, kinetics of soluble protein release from T22-GFP-H6 ssArtIBs. e) Hydrodynamic mode size peak of T22-GFP-H6 nanoparticles released from ssArtIBs, compared to equivalent soluble nanoparticles after purification from recombinant bacteria (those used for ArtIB fabrication). In the inset, an FESEM image of those nanoparticles released from ArtIBs. f) AMD3100-mediated inhibition of HeLa cell internalization of recombinant soluble and ArtIBs-released nanoparticles. Symbols indicate significant differences to the control (*, p < 0.05, Tukey test) and between samples with or without AMD3100 (—, p < 0.05, two tail, t-test). T22-PE24-H6,[17] that are targeted to the cell-surface cytokine in a CXCR4-dependent way (Figure 2f). This fact unveiled a receptor CXCR4[18] through the N-terminal tumor homing pep- potential of ArtIBs as chemically homogenous protein reser- tide T22.[19] The size of these materials was smaller than that voirs for prolonged in vivo delivery of tumor-targeted, nano- of those formed by the previously tested enzymes (Figure 1), structured protein drugs. a fact that, considering potential medical applications, could In this context, different categories of T22-GFP-H6 ArtIBs generally favor molecular and cell interactivity and protein were implanted subcutaneously (SC) in a CXCR4+ colorectal release, by increasing the surface/volume ratio. When exposed cancer mouse model, releasing fluorescent material from the to cultured CXCR4+ Hela cells, T22-GFP-H6 ArtIBs internal- implantation point, followed by selective uptake by a remote ized very efficiently as in the case of IB-based nanopills,[10] CXCR4+ tumor, with specific kinetics for each ArtIB type. A by a CXCR4-dependent route that is inhibited by the CXCR4 preliminary screening of T22-GFP-H6 msArtIBs and T22-GFP- antagonist AMD3100[20] (Figure 2b). Cell viability was not H6 ssArtIBs (Zn2+, at 100:1 ratio of zinc to protein) showed affected by T22-GFP-H6 ArtIBs (Figure 2c), but it was instead slow release and negligible or a small amount of material accu- dramatically compromised, in a CXCR4-dependent fashion, mulated in the tumor by 21 days (Figure 3a). Lowering the by the Pseudomonas aeruginosa exotoxin (PE24) contained in proportion of Zn2+ (50:1) or using alternative cations to induce T22-PE24-H6 ArtIBs. As in the case of IBs, ArtIBs steadily ssArtIBs formation, improved protein release and tumor released a fraction of the forming protein in soluble form when uptake. In particular, ss ArtIBs formed by Ca+2 were more effi- incubated in physiological buffer, at least for 7 days (Figure 2d). cient than ArtIBs (Zn+2 50:1) in maintaining a faster and pro- T22-GFP-H6 solubilized in vitro from ArtIBs was fluorescent gressive protein release from the SC injection site leading to (1039.83 AU mg−1), it assembled as 13 nm nanoparticles indis- a higher accumulation and longer residence time (starting at tinguishable in size from soluble T22-GFP-H6 (Figure 2e), and day 3 and at least until day 10) in the remote CXCR4+ tumor this material was equally able to penetrate cultured HeLa cells (Figure 3b). Further, T22-PE24-H6 ArtIBs Ca2+, containing Adv. Sci. 2020, 7, 1902420 1902420 (3 of 7) © 2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com www.advancedscience.com Figure 3. ArtIBs material release, tumor uptake and antitumor activity in a CXCR4+ colorectal cancer model. a) Preliminary screening of released material and tumor uptake after subcutaneous implantation of T22-GFP-H6 msArtIBs, T22-GFP-H6 ssArtIBs (Zn2+ 100:1) or PBS. b) Representa- tive FLI images obtained at the injection point (IP) and at the remote tumor (T), along time (day 0, 3, 6, and 10) after T22-GFP-H6 ssArtIBs Ca2+, T22-GFP-H6 ssArtIBs Zn2+ (1:50) or buffer SC administration. c) Antitumor effect, measured as bioluminiscence emission by cancer cells along time, in the CXCR4+ SW1417-luci tumor model, after SC injection of 1 mg dose per mouse of T22-GFP-H6 Ca2+ArtIBs, T22-PE24-H6 Ca2+ ArtIBs or control PBS (*, p < 0.05, Tukey test). Fluorescence in (a) and (b), or bioluminescence in (c) intensity were measured using IVIS Spectrum and expressed as x- ± SE of average radiant efficiency. the CXCR4-targeted cytotoxic polypeptide PE, induced an immune concerns, or to enhance any putative immune reac- (p = 0.083) inhibition of tumor growth (n = 3, 1.0 ± 0.2 × 108) tion over those associated to the repeated, conventional stronger than T22-GFP-H6 Ca2+ ArtIBs (n = 3, 1.5 ± 0.7 × 108), administration regimens of soluble protein drugs (such as as compared to the control buffer-treated group (n = 2, insulin, interferons, and many others). In particular, the sim- 2.6 ± 1.0 × 108) (Figure 3c). This fact occurred in absence of sys- pler ss fabrication method allows engineering the strength of temic toxicity (lack of histopathological alterations in hematox- protein–protein interactions in the material by means of the ylin and eosin (H&E)-stained liver and kidney at the end of the stoichiometric control of metal or nonmetal divalent cations. In experiment, not shown). These observations fully confirmed contrast to other excellent and biocompatible materials devel- both the secretion-like prolonged protein release and the pre- oped as micro- or nanoparticles for the slow and sustained cise cell targeting of functional materials through the blood drug release,[22] such as those based on PLA or PLGA,[23] ArtIBs stream, from a remote location. are chemically homogeneous and show no chemical distinc- In summary, ArtIBs can be fabricated in vitro as a new type tion between carrier and cargo, thus acting as self-contained, of biomimetic material, from pure protein and by simple phys- self-released drug materials. Then, the protein drug itself acts, icochemical methods. These protein particles reproduce IB in addition, as a scaffold material, what results in intriguing, properties that are relevant to potential uses in biomedicine, totally novel and chemically homogenous drug delivery systems especially protein release, but their potential use would fully with simpler fabrication processes as opposite to hybrid plat- prevent the immunotoxic reactions potentially associated to the forms. ArtIBs might not only replace IBs as functional protein administration of IBs, that contain bacterial debries at variable reservoirs and offer homogeneous materials for drug-oriented but significant proportions. On the other hand, since protein development, but they will enable, in addition, the packaging of drugs used in clinics have a human origin,[21] their adminis- glycosylated proteins of mammalian cell origin as IB-like mate- tration in an ArtIBs format is not expected to pose significant rials. Since these proteins would be never produced in bacteria Adv. Sci. 2020, 7, 1902420 1902420 (4 of 7) © 2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com www.advancedscience.com in functional forms, ArtIBs will then expand, as a universal 1 kV and equipped with a high resolution secondary electron detector. platform, the catalogue of enzymes or protein drugs that could Representative images of a general fields and nanoparticle detail were be formulated as pure microscale biocatalysts or as secretory captured at magnifications ranging between 5500× and 8500× and a working distance of 3.5 mm. protein granules. Attenuated Total Reflectance: The most suitable concentration of ArtIBs was placed and dried with a continuous N2 flow on spectroscopic crystal surfaces. Total reflectance spectroscopy was detected 15 times as spectra by using a scan rate of 50 cm−1 min−1 and a nominal resolution Experimental Section of 2 cm−1 in a Tensor 27 Bruker spectrometer coupled to a Specac Fabrication of ArtIBs: To produce msArtIBs, 1 mg of pure soluble protein Golden Gate ATR accessory. All measurements were performed at was denatured and concomitantly precipitated by heating at 100 °C in 25 °C, the absorbance obtained was corrected against the background NaCl2 (500 × 10−3 m), ZnCl2 (26.4 × 10−3 m), and MgCl2 (18.4 × 10−3 m) and the PBS buffer signal was subtracted. Fourier deconvolution of in distilled H2O. The precipitate was centrifuged at 15 000 × g for 15 min the spectra and the second derivative allow the identification of the at 4 °C, isolated from the soluble fraction and resuspended with 1 mg of different band components. Fitting of the components to the original phosphatidylcholine with chloroform/methanol 2:1 v/v in a final volume of (not deconvolved) spectrum was essentially performed according to 300 µL. The excess of organic solvent (OS) was removed by a continuous a described procedure.[24] Peak height, band width, and peak position N2 flow inducing the formation of a protein–lipid film phase that acted of the components were allowed to vary one at a time in this order. A as scaffold. The scaffold was afterwards resuspended in 1 mg mL−1 of Gaussian shape was assumed. previous soluble protein diluted in phosphate buffered saline (PBS) at Cell Culture: CXCR4+ cervical cancer cell lines (HeLa ATCC-CCL-2) 4 °C overnight. Finally, the sample was centrifuged, and soluble fraction were used to study the performance of ArtIBs in vitro. Cells were discarded. The manufacturing of ssArtIBs was approached by diluting pure routinely cultured in Eagle’s minimum essential medium (Gibco), soluble protein in distilled H2O at a final concentration of 2 mg mL−1 and supplemented with 10% fetal bovine serum (FBS, Gibco) and incubated final volume of 200 µL. Protein samples (0.196 × 10−3 m) were subsequently in a humidified atmosphere at 37 °C and 5% of CO2. mixed with ZnCl2, at a 100:1 ratio of zinc to protein. After 10 min of Protein Internalization: HeLa CXCR4+ cells were cultured in 24-well incubation at room temperature, samples were centrifuged at 15 000 × g plates in MEM Alpha 1× GlutaMAX medium (Gibco) supplemented for 15 min and soluble fraction discarded to obtain the final product. with foetal bovine serum (FBS) at 37 °C in a 5% CO2 humidified Alternatively, zinc at a ratio 50:1 and calcium at a ratio 300:1 (in form of atmosphere until 70% of confluence was reached. The medium was CaCl2) were used for the in vivo experimental. β-Gal [EC3.2.1.23] and AP then exchanged for serum free OptiPro medium (Gibco) before the [EC3.1.3.1], both from Escherichia coli, were purchased from Sigma-Aldrich. addition of the protein. Protein uptake was determined at different times T22-GFP-H6 and T22-PE24-H6 were produced as recombinant proteins ranging from 10 min to 24 h at a final concentration of 2.5 µg. Cells and purified by single step chromatography as reported.[17] were detached, and external hooked protein removed by Trypsin-EDTA Determination of Enzymatic Activity: Between 3.7 and 21.3 ng of (Gibco) at 1 mg mL−1 exposure for 15 min at 37 °C. Intracellular protein pure soluble β-Gal protein or β-Gal ArtIBs were mixed with 5 × 10−3 m fluorescence was detected by flow cytometry using a FACS-Canto system of ortho-nitrophenylgalactopiranoside in a final volume of 500 µL of PBS. (Becton Dickinson) with an air-cooled argon ion laser (15 mW) exciting The mixture was incubated for 15 min at 37 °C, the reaction stopped by at 488 nm and a D detector (530/30 nm as band pass filter). In addition, adding 200 µL of Na2CO3 (2.8 m) and the product amount determined the internalization specificity through CXCR4 receptor was tested by by measuring absorbance at 420 nm (ε420 = 4530 m−1 cm−1) in a UV–vis exposing cells to the CXCR4 antagonist AMD3100[25] 1 h prior protein spectrophotometer (Ultrospec 1000E, Pharmacia Biotech). On the other incubation at (protein/AMD3100) 1:10 ratio. hand, between 3.9 and 92 ng of pure soluble AP protein or AP ArtIBs Cell Viability: HeLa (ATCC-CCL-2) cell line was cultured in opaque- were mixed with 20 × 10−3 m of para-nitrophenylphosphate (pNPP) in a walled 96-well plates at a final concentration of 6000 cells per well for final volume of 500 µL of PBS. The mixture was incubated for 15 min at 24 h. MEM Alpha GlutaMAX medium (Gibco) supplemented with FBS 37 °C, the reaction stopped by adding 200 µL of NaOH (1 m) and activity was used at 37 °C in a 5% CO2 humidified atmosphere, until 70% determined by measuring p-nitrophenyl phosphate absorbance at 405 nm of confluence was reached. ArtIBs were incubated at 1 × 10−6 m for (ε405 = 18 000 m−1 cm−1) in a UV–vis spectrophotometer (Ultrospec 96 h using MEM Alpha GlutaMAX medium (Gibco). Cell viability was 1000E, Pharmacia Biotech). measured by CellTiter-Glo Luminescent Cell Viability Assay (Promega) in Determination of Specific Fluorescence: Pure soluble T22-GFP-H6 or a Multilabel Plater Reader Victor3 (Perkin Elmer). ArtIB versions were diluted in PBS at concentrations ranging from 0.2 Soluble Protein Release from Artificial IBs: ArtIBs were resuspended to 1 mg mL−1. The excitation wavelength (λex) was set at 488 nm and in 1 mL of PBS 1× reaching a final concentration of 1 mg mL−1 and the emission (λem) at 510 nm, meanwhile the excitation slit was set at incubated at 37 °C without agitation. 100 µL were taken from each 2.5 nm and the emission slit at 5 nm. Fluorescence was measured in a sample at different times ranging from 0 to 7 days and centrifuged for Cary Eclipse Fluorescence Spectrophotometer (Agilent Technologies) by 15 min at 15 000 × g at 4 °C to isolate soluble and insoluble fractions. using a quartz cell with a 10 mm path of light. The intrinsic fluorescence Soluble protein was then stain-free detected by TGX (TGX FastCast of each sample was then represented referred to protein concentration, Acrylamide Kit) and subsequently quantified by ImageLab software to defining the SFD mathematically represented as a slope. The % of SFD determine the % of released protein. (%SFD) represents the relationship of the parameter with the SFD of In Vivo Release of Fluorescent Material by Subcutaneously Implanted soluble T22-GFP-H6 protein. ArtIBs and Their Tumor Uptake: Four-week-old female mice of the Size Distribution Analysis: Volume size distribution of all Swiss nude strain, in the 18–20 g body weight range (Charles River, nanostructures was determined at 633 nm and 25 °C in a Zetasizer L-Abreslle, France), maintained in pathogen-free conditions, were used Nano ZS (Malvern Instruments Limited) by using ZEN2112 3 mm quartz for the in vivo experiments. All experimental procedures were approved batch cuvettes. Protein samples dissolved in PBS from 0.2 to 1 mg mL−1 by the Hospital de Sant Pau Animal Ethics Committee and performed were measured in triplicate and mode size peak and polydispersion according to European Council directives. To generate the CXCR4+ index (pdi ± s.e.m.) obtained. SW1417 CRC cancer model, a 10 mg aliquot of SW1417-luci tumor Electron Microscopy: Ultrastructural morphometry (size and shape) tissue was obtained from donor animals and deposited in the anterior of ArtIBs was characterized at nearly native state with field emission or posterior flank subcutis of the animals. When tumors reached ≈120– scanning electron microscopy (FESEM). Drops of 20 µL of each sample 200 mm3 volume, animals were randomly allocated and implanted in diluted at 0.3 mg mL−1 in their respective buffers were directly deposited the subcutis of the mouse lumbar region with a pellet of T22-GFP-H6 on silicon wafers (Ted Pella Inc.) for 30 s and immediately observed msArtIBs or T22-GFP-H6 ssArtIBs Zn2+ in a preliminary study, at a single without coating with an FESEM Zeiss Merlin (Zeiss) operating at dose injection of 1 mg per mouse, suspended in a 150 µL PBS buffer, Adv. Sci. 2020, 7, 1902420 1902420 (5 of 7) © 2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com www.advancedscience.com whereas in a second study, T22-GFP-H6 Zn2+ ArtIBs or of T22-GFP-H6 (http://www.nanbiosis.es/portfolio/u18-nanotoxicology-unit/). Electron Ca2+ ArtIBs were implanted at the same dose. Control buffer injection microscopy studies were performed by the Servei de Microscòpia at was used as a negative control. The ArtIBs IP was selected to position the UAB. A.V. received an ICREA ACADEMIA award. H.L.L. received it as far away as possible from the tumor in the same mouse, being a predoctoral fellowship from AGAUR (2019 FI_B 00352) and U.U. located either in the anterior or posterior flanks. was supported by PERIS program from the Health Department of la After ArtIBs pellet injection, the IVIS Spectrum equipment Generalitat de Catalunya. (PerkinElmer Inc.) was used to monitor the GFP-emitted fluorescence by the SC implants in whole-body mouse by registering immediately (0 h) and at specific time points (3, 6, and 10 days) after the administration to determine the fluorescence remaining in the Conflict of Interest subcutaneous ArtIBs implants, as well as the fluorescent material that E.V., R.M., and A.V. are cofounders of NANOLIGENT, devoted to develop reached the remote tumor along time, in each mouse. Fluorescent antitumoral drugs based on proteins. J.M.S., H.L-L., P.A., E.V., R.M. and signal was digitalized, displayed as pseudocolor overlay, and expressed A.V. are co-inventors in a patent covering the use of ArtIBs. as radiant efficiency. The fluorescence intensity (FLI) ratio was calculated dividing the signal from the IBs-treated mice by the FLI autofluorescent signal of buffer-administered control mice either in the injection point or in the tumor. Keywords In Vivo Antitumor Activity of SC Implanted ArtIBs: The CXCR4+ SW1417 CRC cancer model used to test antitumor activity was generated as biomimetic materials, cancer, drug release, microparticles, recombinant described above. The expression of luciferase by cancer cells in this proteins model allowed for the noninvasive follow-up of tumor growth along time. A week before the deposition of the tumor aliquot in the mouse Received: September 5, 2019 subcutis, mice were randomly allocated to be SC administered in the Revised: October 30, 2019 mouse lumbar region with 1 mg per mouse dose of T22-GFP-H6 Ca2+ Published online: November 27, 2019 ArtIBs or T22-PE24-H6 Ca2+ ArtIBs suspended in a 150 µL of PBS buffer or buffer-treated control mice. After IBs administration, mouse body weight was recorded, and bioluminescent image intensity in the tumor, measured using the IVIS [1] U. Rinas, E. Garcia-Fruitos, J. L. Corchero, E. Vazquez, Spectrum equipment (PerkinElmer Inc.), was digitalized and expressed J. Seras-Franzoso, A. Villaverde, Trends Biochem. Sci. 2017, 42, 726. as radiant efficiency. Tumor tissue, liver, and kidney were formalin-fixed [2] A. de Marco, N. Ferrer-Miralles, E. Garcia-Fruitos, A. Mitraki, and paraffin-embedded for histology. To that aim, four-micrometer- S. Peternel, U. Rinas, M. A. Trujillo-Roldan, N. A. Valdez-Cruz, thick sections were stained with H&E, and analyzed for possible E. Vazquez, A. Villaverde, FEMS Microbiol. Rev. 2019, 43, 53. histological alterations by two independent observers. Representative [3] N. S. de Groot, R. Sabate, S. Ventura, Trends Biochem. Sci. 2009, 34, images were taken using Cell^B software (Olympus Soft Imaging 408. v. 3.3). Statistical Analysis: All analyses were performed with SPSS versus [4] N. Gonzalez-Montalban, E. Garcia-Fruitos, A. Villaverde, Nat. Bio- version 11.0 (IBM) software. One-way ANOVA and t-tests were technol. 2007, 25, 718. performed to assess differences in assays with a minimum n = 3. The [5] C. Slouka, J. Kopp, O. Spadiut, C. Herwig, Appl. Microbiol. Bio- Holme–Sidak test was applied for equal variance and Tukey or Mann technol. 2019, 103, 1143. Whitney U test for unequal variance (indicated in the figure legend). Two [6] C. Slouka, J. Kopp, S. Hutwimmer, M. Strahammer, D. Strohmer, tail t-test was also used for individual comparisons. Data were presented E. Eitenberger, A. Schwaighofer, C. Herwig, Microb. Cell Fact. 2018, as means ± standard error of the mean (s.e.m). Differences between the 17, 148. protein samples were considered significant at p ≤ 0.05. [7] O. Cano-Garrido, E. Rodriguez-Carmona, C. Diez-Gil, E. Vazquez, E. Elizondo, R. Cubarsi, J. Seras-Franzoso, J. L. Corchero, U. Rinas, I. Ratera, N. Ventosa, J. Veciana, A. Villaverde, E. Garcia-Fruitos, Acta Biomater. 2013, 9, 6134. Acknowledgements [8] E. Hrabarova, L. Achbergerova, J. Nahalka, Insoluble Proteins: J.M.S., H.L.L., and P.A. contributed equally to this work. The authors Methods Protoc. 2015, 1258, 411. are indebted to Agencia Estatal de Investigación (AEI) and to Fondo [9] Y. Loo, M. Goktas, A. B. Tekinay, M. O. Guler, C. A. Hauser, Europeo de Desarrollo Regional (FEDER) (Grant No. BIO2016- A. Mitraki, Adv. Healthcare Mater. 2015, 4, 2557. 76063-R, AEI/FEDER, UE) to A.V., AGAUR (2017SGR-229) to A.V., and [10] E. Vazquez, J. L. Corchero, J. F. Burgueno, J. Seras-Franzoso, 2017SGR-865 GRC to R.M. CIBER-BBN (project NANOPROTHER) A. Kosoy, R. Bosser, R. Mendoza, J. M. Martinez-Lainez, U. Rinas, granted to A.V. and CIBER-BBN project 4NanoMets to R.M. ISCIII E. Fernandez, L. Ruiz-Avila, E. Garcia-Fruitos, A. Villaverde, Adv. (PI15/00272 cofounding FEDER) to E.V. and ISCIII (cofounding Mater. 2012, 24, 1742. FEDER) PIE15//00028 and PI18/00650 to R.M., and to EU COST [11] S. K. Maji, M. H. Perrin, M. R. Sawaya, S. Jessberger, K. Vadodaria, Action CA 17140. The authors are also indebted to the Networking R. A. Rissman, P. S. Singru, K. P. Nilsson, R. Simon, D. Schubert, Research Center on Bioengineering, Biomaterials and Nanomedicine D. Eisenberg, J. Rivier, P. Sawchenko, W. Vale, R. Riek, Science 2009, (CIBER-BBN) that is an initiative funded by the VI National R&D&I Plan 325, 328. 2008–2011, Iniciativa Ingenio 2010, Consolider Program, CIBER Actions [12] a) M. V. Cespedes, Y. Fernandez, U. Unzueta, R. Mendoza, and financed by the Instituto de Salud Carlos III, with assistance from J. Seras-Franzoso, A. Sanchez-Chardi, P. Alamo, V. Toledo-Rubio, the European Regional Development Fund. Protein production was performed in part by the ICTS “NANBIOSIS,” more specifically by the N. Ferrer-Miralles, E. Vazquez, S. Schwartz, I. Abasolo, Protein Production Platform of CIBER in Bioengineering, Biomaterials J. L. Corchero, R. Mangues, A. Villaverde, Sci. Rep. 2016, 6, 35765; & Nanomedicine (CIBER-BBN)/IBB, at the UAB sePBioEs scientific- b) M. Pesarrodona, T. Jauset, Z. V. Díaz-Riascos, A. Sánchez-Chardi, technical service (http://www.nanbiosis.es/portfolio/u1-protein- M. E. Beaulieu, J. Seras-Franzoso, L. Sánchez-García, production-platform-ppp/) and the nanoparticle size analysis by the R. Baltà-Foix, S. Mancilla, Y. Fernández, U. Rinas, S. Schwartz, Biomaterial Processing and Nanostructuring Unit. Biodistribution L. Soucek, A. Villaverde, I. Abasolo, E. Vázquez, Adv. Sci. 2019, studies were performed by the ICTS “NANBIOSIS,” Nantoxicology Unit 6, 1900849; c) U. Unzueta, M. V. Cespedes, R. Sala, P. Alamo, Adv. Sci. 2020, 7, 1902420 1902420 (6 of 7) © 2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com www.advancedscience.com A. Sanchez-Chardi, M. Pesarrodona, L. Sanchez-Garcia, [19] H. Tamamura, M. Imai, T. Ishihara, M. Masuda, H. Funakoshi, O. Cano-Garrido, A. Villaverde, E. Vazquez, R. Mangues, H. Oyake, T. Murakami, R. Arakaki, H. Nakashima, A. Otaka, J. Seras-Franzoso, J. Controlled Release 2018, 279, 29. T. Ibuka, M. Waki, A. Matsumoto, N. Yamamoto, N. Fujii, Bioorg. [13] E. Garcia-Fruitos, N. Gonzalez-Montalban, M. Morell, A. Vera, Med. Chem. 1998, 6, 1033. R. M. Ferraz, A. Aris, S. Ventura, A. Villaverde, Microb. Cell Fact. [20] J. Kim, K. L. Connelly, E. M. Unterwald, S. M. Rawls, Brain, Behav., 2005, 4, 27. Immun. 2016, 62, 30. [14] R. S. Jacob, S. Das, S. Ghosh, A. Anoop, N. N. Jha, T. Khan, [21] L. Sanchez-Garcia, L. Martin, R. Mangues, N. Ferrer-Miralles, P. Singru, A. Kumar, S. K. Maji, Sci. Rep. 2016, 6, 23370. E. Vazquez, A. Villaverde, Microb. Cell Fact. 2016, 15, 33. [15] Q. Luo, C. Hou, Y. Bai, R. Wang, J. Liu, Chem. Rev. 2016, 116, 13571. [22] a) P. Davoodi, L. Y. Lee, Q. Xu, V. Sunil, Y. Sun, S. Soh, [16] F. Rueda, M. V. Cespedes, O. Conchillo-Sole, A. Sanchez-Chardi, C. H. Wang, Adv. Drug Delivery Rev. 2018, 132, 104; b) C. Helary, J. Seras-Franzoso, R. Cubarsi, A. Gallardo, M. Pesarrodona, M. F. Desimone, Curr. Pharm. Biotechnol. 2015, 16, 635; N. Ferrer-Miralles, X. Daura, E. Vazquez, E. Garcia-Fruitos, c) O. S. Fenton, K. N. Olafson, P. S. Pillai, M. J. Mitchell, R. Langer, R. Mangues, U. Unzueta, A. Villaverde, Adv. Mater. 2015, 27, Adv. Mater. 2018, 30, 1705328. 7816. [23] a) D. da Silva, M. Kaduri, M. Poley, O. Adir, N. Krinsky, J. Shainsky- [17] L. Sanchez-Garcia, N. Serna, P. Alamo, R. Sala, M. V. Cespedes, Roitman, A. Schroeder, Chem. Eng. J. 2018, 340, 9; b) D. Ding, M. Roldan, A. Sanchez-Chardi, U. Unzueta, I. Casanova, Q. Zhu, Mater. Sci. Eng., C 2018, 92, 1041. R. Mangues, E. Vazquez, A. Villaverde, J. Controlled Release 2018, [24] V. Nolan, J. M. Sanchez, M. A. Perillo, Colloids Surf., B 2015, 136, 1202. 274, 81. [25] Y. H. Jung, D. Y. Lee, W. Cha, B. H. Kim, M. W. Sung, K. H. Kim, [18] F. Balkwill, Semin. Cancer Biol. 2004, 14, 171. S. H. Ahn, Head Neck 2016, 38, 1479. Adv. Sci. 2020, 7, 1902420 1902420 (7 of 7) © 2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim