This is the accepted version of the article: De Marco, Ario; Ferrer Miralles, Neus; García Fruitós, Elena; [et al.]. «Bacterial inclusion bodies are industrially exploitable amyloids». FEMS Microbiology Reviews, Vol. 43, Issue 1 (January 2019), p. 53-72. DOI 10.1093/femsre/fuy038 This version is available at https://ddd.uab.cat/record/233713 under the terms of the license Bacterial inclusion bodies are industrially exploitable amyloids Ario de Marco a, Neus Ferrer-Miralles b,c,d, Elena Garcia-Fruitós e, Anna Mitraki f,g, Spela Peternel h, Ursula Rinas i, j, Mauricio A. Trujillo-Roldán k, Norma A. Valdez-Cruz l, Esther Vázquez b,c,d, Antonio Villaverde b,c,d * aLaboratory for Environmental and Life Sciences, University of Nova Gorica, Vipavska 13, 5000 Nova Gorica, Slovenia bInstitut de Biotecnologia i de Biomedicina (IBB), Universitat Autònoma de Barcelona, 08193 Cerdanyola del Vallès, Spain. c Departamentde Genètica i de Microbiologia, Universitat Autònoma de Barcelona, 08193 Cerdanyola del Vallès, Spain. dCIBER de Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), 08193 Cerdanyola del Vallès, Spain. eDepartment of Ruminant Production, Institut de Recerca i Tecnologia Agroalimentàries (IRTA), Torre Marimon, 08140 Caldes de Montbui, Barcelona, Spain fDepartment of Materials Science and Technology, University of Crete, Vassilika Vouton, 710 03 Heraklion, Crete, Greece gInstitute of Electronic Structure and Laser (IESL), Foundation for Research and Technology Hellas (FORTH), PO Box 527, Vassilika Vouton, 711 10 Heraklion, Crete, Greece h Lupinica, Ljubljana, Slovenia iLeibniz University of Hannover, Technical Chemistry and Life Science, Hannover, Germany j Helmholtz Centre for Infection Research, Braunschweig, Germany kPrograma de Investigación de Producción de Biomoléculas, Unidad de Bioprocesos, Departamento de Biología Molecular y Biotecnología, Instituto de Investigaciones Biomédicas, Universidad Nacional Autónoma de México, Ciudad de México, México lPrograma de Investigación de Producción de Biomoléculas, Departamento de Biología Molecular y Biotecnología, Instituto de Investigaciones Biomédicas, Universidad Nacional Autónoma de México, Ciudad de México, México * Corresponding author 1 Keywords: recombinant bacteria, protein production, inclusion bodies, functional amyloids, biomaterials, protein release 2 Abstract Understanding the structure, functionalities and biology of functional amyloids is an issue of emerging interest. Inclusion bodies, namely protein clusters formed in recombinant bacteria during protein production processes, have emerged as unanticipated, highly tunable models for the scrutiny of the physiology and architecture of functional amyloids. Based on an amyloidal skeleton combined with varying amounts of native or native-like protein forms, bacterial inclusion bodies exhibit an unusual arrangement that confers mechanical stability, biological activity and conditional protein release, being thus exploitable as versatile biomaterials. The applicability of inclusion bodies in biotechnology as enriched sources of protein and reusable catalysts, and in biomedicine as biocompatible topographies, nanopills, or mimetics of endocrine secretory granules has been largely validated. Beyond these uses, the dissection of how recombinant bacteria manage the aggregation of functional protein species into structures of highly variable complexity offers insights about unsuspected connections between protein quality (conformational status compatible with functionality) and cell physiology. 3 1. Introduction Many bacterial species, as transformers of organic and inorganic substances, are used in food technologies and other fields with established or rising economic impact such as fabrics, bioremediation or mining, for performing controlled processes of industrial interest (Demain, 2000). This is done by exploiting the metabolic diversity and versatility of prokaryotic cells that, due to their evolutionary adaptation to different environments and physicochemical conditions, developed a vast spectrum of alternative physiologic strategies. In addition, bacteria have been also engineered as cell factories for the production of macromolecules (mainly proteins, nucleic acids and polymers), metal particles and secondary metabolites. The production of these substances is continuously optimized by process (tailored culture conditions for fast grow or high productivity) and genetic approaches (mutant strains with improved functionalities and metabolic routes). The resulting spectrum of biological products infiltrates diverse areas in biotechnology and pharmaceutical industries. Furthermore, an increasing number of applications emerges also in very diverse fields such as energy, electronics, material sciences and nanotechnologies (Rodriguez-Carmona & Villaverde, 2010, Du, et al., 2011, Lee, et al., 2012, Malik, et al., 2018). Among bacterial products, proteins are of special interest as structural but also functional agents, and they can be easily produced by conventional recombinant DNA strategies or by gene synthesis (Hartley, 2006). Proteins are produced not only in bacteria (Ferrer-Miralles & Villaverde, 2013, Overton, 2014) but also in other cell factories (mainly yeasts, filamentous fungi, mammalian and insect cells, and whole plant or animals) (Corchero, et al., 2013). Recombinant enzymes are employed in food technologies, in chemicals, detergent and fabric manufacture, as well as biopharmaceuticals for protein replacement therapies and in life sciences research (Vellard, 2003). In this context, hormones, enzymes, antigens and antibodies are common protein drugs and a significant number of them, approved and marketed for human therapies, are produced in bacterial cells (Ferrer-Miralles, et al., 2009, Overton, 2014, Sanchez-Garcia, et al., 2016). Many vaccines are based on conventional recombinant proteins (Nascimento & Leite, 2012) or on a specific category of self- 4 assembling protein materials named virus-like particles (Pattenden, et al., 2005, Lua, et al., 2014), that mimic structural and antigenic features of natural viruses. Protein production in bacterial cell factories was indeed made possible through the development of recombinant DNA technologies (Ferrer-Miralles, et al., 2009), that raised in the late 70’s with the discovery and application of restriction enzymes. Since then, recombinant protein production has become a routine practice worldwide. Protein production is reached by exploiting an expanding catalogue of bacterial species as biological factories. These show distinguishable properties regarding culture requirements, use of carbon sources, metabolic capabilities and protein secretability (Corchero, et al., 2013, Ferrer-Miralles & Villaverde, 2013). The enterobacterium Escherichia coli was the initial recombinant cell factory and still the most universally preferred system for protein production, because of its easy and fast culture, deep understanding of genetics and physiology as well as for the availability of numerous tools for genetic manipulation (Rosano & Ceccarelli, 2014). The vast majority of proteins in nature are active as soluble species. Then, protein production processes are aimed to obtain soluble proteins with native, functional structure to nicely mimic their natural activities. While the native conformation is reached through a correct folding process,-protein misfolding and consequent aggregation in the cytosol (for intracellular) or the periplasm (for secreted) is a very common event. As a consequence, in most production processes the occurrence of insoluble protein species has been reported, often accompanied by a fraction of soluble protein (de Marco, 2013). The ratio between soluble fraction and total protein (the supposed percentage of soluble and functionally folded protein) is then commonly used as an indication for a successful process. 2. Inclusion bodies in recombinant bacteria Aggregates of recombinant proteins occur as small supramolecular entities in the soluble cell fraction or as larger protein clusters that precipitate as insoluble materials. Therefore, the recombinant protein species range along a spectrum of conformational versions that embrace soluble protein species, soluble aggregates and insoluble 5 aggregates (Schrodel & de Marco, 2005, Martinez-Alonso, et al., 2008). Soluble aggregates or proto-aggregates have been usually considered as precursors of large protein deposits (Martinez-Alonso, et al., 2009). Aggregated proteins tend to eventually accumulate in recombinant cells as refractile (under optical microscope visualization), electrodense (under transmission electron microscope visualization) submicron particles known as inclusion bodies (IBs, Figure 1A, B) (Villaverde & Carrio, 2003). IBs have been considered to be a main bottleneck in the protein production processes and their formation has been associated to improper protein folding or misfolding (Villaverde & Carrio, 2003, Baneyx & Mujacic, 2004). Despite the development of diverse approaches and algorithms to predict protein solubility based on the primary sequence (Smialowski, et al., 2007, Tjong & Zhou, 2008, Magnan, et al., 2009, Diaz, et al., 2010, Agostini, et al., 2012, de Groot, et al., 2012, Huang, et al., 2012, Smialowski, et al., 2012, Chang, et al., 2014, Yang, et al., 2016, Paladin, et al., 2017, Roche, et al., 2017, Khurana, et al., 2018, Rawi, et al., 2018), the tendency of a given recombinant polypeptide to form IBs cannot be accurately anticipated. While the progress in in silico prediction of solubility would be highly convenient (as it might allow preventing unsuccessful efforts in the production of insoluble proteins and then save costs) it is still an unmet target. This is likely because solubility of a recombinant protein does not merely lie on the amino acid sequence, but also on multifactorial and highly dynamic physiological issues, such as cell energetics, the availability of molecular chaperones, foldases, stabilizing osmolytes, the folding and secretion kinetics, or the effect of incorrect post-translational modifications (Bandyopadhyay, et al., 2012). The formation of IBs is then an important obstacle in protein production processes, as their occurrence undermine the relative accumulation of soluble protein species versus the aggregated versions, that is, it reduces the solubility of the target protein. In addition, protein aggregation has been historically associated to lack of functionality, because reaching the native protein conformation is generally linked to solubility (Baneyx & Mujacic, 2004). The generic tendency to aggregation and insolubility of recombinant proteins has then pushed to develop strategies for protein production addressed to minimize IB formation, which has been only moderately 6 successful and strongly protein-dependent (Sorensen & Mortensen, 2005, Sorensen & Mortensen, 2005). Being mechanically stable (Vogel, et al., 2002, Garcia-Fruitos, et al., 2009), IBs can be purified from bacterial cells upon cell breaking and mild-to-harsh rinsing treatments or density gradient centrifugation (Schrodel & de Marco, 2005, Rodriguez-Carmona, et al., 2010, Seras-Franzoso, et al., 2015). This offers a particulate material usable as relatively pure source of recombinant protein. In this context, when attempts to prevent IB formation failed, IBs have been exploited in protein biotechnology as relatively pure source of proteins (Gatti-Lafranconi, et al., 2011) to be recovered in vitro. This approach considers first the disintegration of IBs by the application of more or less aggressive chaotropic agents or detergents and a successive protein refolding process under controlled conditions. These approaches have been developed in parallel to the advances in recombinant protein production and are summarized elsewhere (Vallejo & Rinas, 2004, Singh & Panda, 2005, Freydell, et al., 2007, Burgess, 2009, Simpson, 2010, Basu, et al., 2011, Yamaguchi & Miyazaki, 2014, Singh, et al., 2015). The lack or poor solubility of recombinant proteins produced in bacteria is a multifactorial event. Which different genetic and environmental parameters (that is, bacterial culture conditions) and how they might be involved in the aggregation process have been a matter of controversial discussions, especially regarding the application of palliative measures (Sorensen & Mortensen, 2005, Sorensen & Mortensen, 2005). Recombinant protein aggregation is associated to incomplete or inadequate protein folding (Hlodan, et al., 1991), the consequent occurrence of exposed hydrophobic patches (Murby, et al., 1995), the inability of the cell quality control machinery (mediating both folding and proteolysis) to accurately perform under protein overproduction conditions (Tomoyasu, et al., 2001, Baneyx & Mujacic, 2004), the triggering of cell responses to conformational stress (Gasser, et al., 2008), an inappropriate codon usage (Parret, et al., 2016) and the molecular overcrowding associated to high protein yields in recombinant bacteria (Martinez-Alonso, et al., 2008). The frequent heterologous origin of the recombinant polypeptides is also favouring aggregation, often due to the absence or inadequate posttranslational 7 modifications executed by bacterial cells that might be critical to adopt a native conformation (Schmidt, 2004, Demain & Vaishnav, 2009). Bacterial IBs are mainly composed of the recombinant protein that is accompanied by some/relatively few host cell proteins and traces of other macromolecular contaminants, such as lipids and nucleic acids (Rinas & Bailey, 1992, Fahnert, et al., 2004). IBs are not mere clustered protein deposits, as the recombinant protein is not associated in an amorphous form, but structured as formerly unexpected amyloidal architectures (Carrio, et al., 2005, Cano-Garrido, et al., 2013), as discussed below. Amyloidal structures are formed through sequence-dependent, stereospecific interactions (Speed, et al., 1996) that exclude protein-protein associations between different protein species in a single aggregate (Morell, et al., 2008). IB formation in recombinant bacteria, based on such recruitment of homologous polypeptide chains is an extremely dynamic process (Carrio, et al., 1998, Carrio, et al., 1999, Carrio & Villaverde, 2001, Carrio & Villaverde, 2002, Baig, et al., 2014), that results from an unbalanced equilibrium between protein association and deposition, proteolytic digestion and disaggregation (Carrio, et al., 1999, Carrio, et al., 2000, Carrio & Villaverde, 2001, Carrio & Villaverde, 2002). This complex process is mediated by the arms of the quality control cell machinery that handle protein aggregates (Weibezahn, et al., 2004, Mogk, et al., 2018). In this way, the arrest of recombinant protein synthesis in metabolically active cells promotes the physiological disintegration of IBs. This is a fast process that results in the refolding of the protein into the native conformation (Carrio & Villaverde, 2001, de Marco, 2007) but also in the proteolytic digestion of an important fraction of the IB material (Carrio & Villaverde, 2001, Carrio & Villaverde, 2002). The integration of IB protein within the cellular quality control is also demonstrated by the conformational rearrangements observed within the IB particles (Gonzalez-Montalban, et al., 2008, Elia, et al., 2017). Although this is a still moderately understood issue, IB formation causes or it is linked to conformational stresses in the producing bacterial cell (Gasser, et al., 2008) and aggregated species are suspected to show some extent of intrinsic toxicity that might negatively affect bacterial cell growth. However, when eukaryotic cells are exposed to IBs in suspension or these particles are administered to whole organisms through 8 different routes, including injection or oral administration, IBs do not appear to be harmful (Garcia-Fruitos, et al., 2009, Diez-Gil, et al., 2010, Garcia-Fruitos, et al., 2010, Liovic, et al., 2012, Seras-Franzoso, et al., 2012, Vazquez, et al., 2012, Seras-Franzoso, et al., 2013, Seras-Franzoso, et al., 2013, Tatkiewicz, et al., 2013, Seras-Franzoso, et al., 2014, Seras-Franzoso, et al., 2014, Seras-Franzoso, et al., 2015, Cespedes, et al., 2016, Seras-Franzoso, et al., 2016, Torrealba, et al., 2016, Torrealba, et al., 2016, Unzueta, et al., 2017, Stamm, et al., 2018, Unzueta, et al., 2018). Therefore, IBs have gained interest as intriguing biocompatible protein materials (Villaverde, 2012, Villaverde, et al., 2012, Loo, et al., 2015, Seras-Franzoso, et al., 2015, Rinas, et al., 2017) that might be substantially different from protein aggregates linked to prion diseases or to degenerative disorders (Chiti & Dobson, 2017, Hartl, 2017). In this regard, IBs represent a paradigmatic example of non-toxic amyloids, associated with a diversity of functions and whose prevalence in living beings seems to be more spread than previously expected (Badtke, et al., 2009, Maji, et al., 2009, Romero & Kolter, 2014, Villaverde, et al., 2015, Jacob, et al., 2016, Molina-Garcia, et al., 2016, Dragos, et al., 2017, Hewetson, et al., 2017, Jackson & Hewitt, 2017, Piscitelli, et al., 2017, Shin & Cherry, 2017). In 2005, it was reported that IBs are at least partially composed by functional protein (Garcia-Fruitos, et al., 2005, Jevsevar, et al., 2005) (Figure 1C). The fraction of functional polypeptides is not externally attached to the protein particles but it represents a structural part of the aggregates, being stable and tightly linked to IBs in aqueous solvents (Garcia-Fruitos, et al., 2007). The fact that an important fraction of the IB-embedded protein retains its biological activity (Gonzalez-Montalban, et al., 2007) and that solubility and functionality are not equivalent parameters (Garcia- Fruitos, et al., 2007) have allowed to further explore IBs as immobilized enzymes for catalysis (Hrabarova, et al., 2015), as biocompatible materials for tissue engineering (Garcia-Fruitos, et al., 2009, Seras-Franzoso, et al., 2012, Tatkiewicz, et al., 2013, Tatkiewicz, et al., 2018) and as protein delivery agents for in vivo applications (Garcia- Fruitos, et al., 2012, Rinas, et al., 2017). Despite the exponentially growing number of examples supporting that IB functionality is a generic and transversal concept (Villaverde, et al., 2015), it still remains unfamiliar to an important fraction of the 9 scientific community that retains the obsolete concept linking protein solubility and functionality (Figure 1D). The bases of IB protein biological activities and the resulting applicability of IBs as functional materials will be discussed below. 2.1 IBs are depot-type bacterial amyloids The combination of a significant set of amyloid-like properties (Ventura & Villaverde, 2006) and the occurrence of biological activity of IB protein (Garcia-Fruitos, et al., 2005) suggests a specific type of non-toxic functional amyloids, apparently uncommon in bacteria. Non-toxic functional amyloids have been widespread found in living beings including bacteria (Seviour, et al., 2015, Dragos, et al., 2017, Van Gerven, et al., 2018). They perform a significant set of recognized functions necessary for or favourable to cell or organic functions, though the activation/inactivation of defined cell circuits. These activities include catalysis, regulation of gene expression, control of DNA replication, signalling, mechanical structuring and complex functions at the systems level such as memory and reproduction (Hafner Bratkovic, 2017, Hewetson, et al., 2017). In bacteria, amyloids have been associated to biofilm formation and quorum sensing (Schwartz, et al., 2012, Seviour, et al., 2015, Schwartz, et al., 2016, Stenvang, et al., 2016, Besingi, et al., 2017, Dragos, et al., 2017), virulence and toxicity (Marcoleta, et al., 2013, Syed & Boles, 2014, Van Gerven, et al., 2018), plasmid replication (Molina-Garcia, et al., 2016), bioenergetics (Molina-Garcia, et al., 2017) and transcriptional regulation (Pallares, et al., 2015, Yuan & Hochschild, 2017, Kaur, et al., 2018), among others. In these examples, the target protein is found in two alternative conformational versions, namely amyloid and non-amyloid, that represent the ON-OFF states of molecular switchers. The coexistence of native or native-like and amyloidal forms of recombinant proteins in bacterial IBs places these particles out of this category. However, there are significant functional and structural similarities between bacterial IBs and secretory granules in the endocrine system. These are functional amyloids that act as repositories of peptidic hormones that are released from glands upon defined stimuli (Maji, et al., 2009). As far as we know, IBs are the first fully characterized depot-type amyloids found in bacteria. This amazing architecture, explained in next sections, make them particularly appealing as fully tailored 10 manufactured protein depositories that can be obtained by standard protein production procedures in recombinant cells. 3. Protein aggregation as soluble protein aggregates As mentioned above, according to an assumption still largely accepted among biologists, soluble proteins correspond to monodispersed, native, and fully functional proteins. This simplification contrasts with both the evidence that soluble aggregates have a critical role in vivo in the progression of neurodegenerative diseases (Gerson, et al., 2014, Kumar, et al., 2016) and the data collected studying recombinant proteins. In two seminal papers (Nomine, et al., 2001, Nomine, et al., 2001) it was demonstrated that recombinant fusion constructs (composed by carriers such as GST, maltose- binding protein or thioredoxin, known for their solubility plus the target proteins) expressed in E. coli and designed to improve overall yield, could form soluble micellar aggregates. These were described as composed by an aggregated nucleus originated by the condensation of hydrophobic patches made available by the misfolded target protein and by an external layer made of hydrophilic, highly soluble carrier protein (Figure 2 A). The overall structure remained soluble, but the functionality of the target protein was partially or totally lost. This condition can be very frequent when a difficult-to-produce protein is fused to a “solubilizing” partner, but can arise also when multi-domain proteins with components characterized by differing aggregation propensity are produced (Souillac, 2005). It must be also stressed that protein aggregation inside bacterial cells is an extremely dynamic process regulated by both chemical (osmolytes) and molecular chaperones (Singer & Lindquist, 1998, Carrio & Villaverde, 2001, Mogk, et al., 2003, Schultz, et al., 2007, Natalello, et al., 2008). Chaperone availability influences the process of aggregation, the structural characteristics of the aggregates, and their re-solubilisation into less complex or even native monodispersed molecules (Figure 2 B). In some cases, protein (re)folding to the native conformation is not achieved and chaperones stick to partially misfolded proteins, impairing their further condensations into larger aggregates (de Marco, et al., 2000). 11 Nevertheless, the proteins complexed in these soluble aggregates are neither native nor fully functional. The contamination with GroEL and DnaK (easily detected by SDS- PAGE) is a strong indicator that recombinant target proteins expressed in E. coli are at least partially unfolded/misfolded (de Marco, et al., 2000). The data collected in the last 20 years using recombinant proteins and controlled conditions show that in the same protein preparation it is possible to identify different soluble aggregates with increasing level of structural complexity, functionality, and capacity to interfere with cell metabolism (Calamai, et al., 2005, Schrodel & de Marco, 2005, Ami, et al., 2009, Li, et al., 2009). Multi-step sucrose gradients, native-blue electrophoresis and electron microscopy analyses enabled to appreciate the large structural variety of soluble aggregates recovered from samples purified from bacteria with different genetic background, grown at different conditions, and challenged with alternative stress factors (Stegemann, et al., 2005, Natalello, et al., 2008, Liu, et al., 2013). Part of the proteins trapped in these aggregates remains functional and these proteins are in a dynamic equilibrium with both monodispersed (native) forms and conventional (insoluble) IBs (de Marco, et al., 2005, Stampolidis, et al., 2009, Van der Henst, et al., 2010). The kinetic of aggregate formation suggests that aggregation proceeds progressively into denser forms once the bacterial folding machinery is overwhelmed and osmolytes become limiting (de Marco, et al., 2005, Stampolidis, et al., 2009, Van der Henst, et al., 2010). Intrinsically disordered proteins are particularly prone to be recruited during aggregate nucleation, for instance, by proteins such Huntingtin which possess poly-glutamine patches (Wear, et al., 2015). The aggregation process can be minimized in vivo by tuning the cell culture parameters. This can be done according to the information inferred from “aggregation probes”, which quantify the relative aggregation level at different growth conditions (Schultz, et al., 2006, Paul, et al., 2015). It means that native monodispersed proteins will be produced when the expression rate remains below the maximal capacity of the bacterial cells to fold the specific polypeptide. Under this perspective it must be considered all those precautions that reduce the cell growth rate (low temperature). These include the controlled increase of foldases and molecular chaperones (induction of stress response by ethanol/ benzyl alcohol addition or heat shock, overexpression of recombinant molecular chaperones), or the stabilization of the folding intermediates (chemical 12 chaperones) (de Marco, et al., 2005). Post-harvesting steps are also relevant since purification, storage, and stress conditions strongly influence the formation and the nature of soluble aggregates (Natalello, et al., 2009, Lebendiker & Danieli, 2014, Guo, et al., 2015, Smirnova, et al., 2015), and particularly, the aggregation process of intrinsically disordered proteins and proteins with intrinsically disordered regions (Churion & Bondos, 2012, Lebendiker, et al., 2015). Osmolytes of different chemical compositions have been effectively exploited to prevent the progressive formation of inactive soluble aggregates of therapeutic proteins (Liu, et al., 2013). The contribution of other components of cell chemical environment, such as salts and nucleic acids, to protein stability and aggregation has been reported (Futami, et al., 2014, Song, 2017), but is still largely unappreciated. What is clear but often underestimated, is that data obtained using protein preparations of soluble aggregates can be totally misleading (Nomine, et al., 2001). Consequently, protein quality control guidelines have been proposed to evaluate biophysical and functional features of soluble proteins and to prevent the use of material that is not suitable for both research and biotechnological applications (Paul, et al., 2015). Size exclusion chromatography (SEC) is the most accessible method to assess the presence of oligomeric forms and soluble aggregates in a sample. Large aggregates usually show a peak corresponding to the void volume and can be directly separated by native species (mono or oligomeric). Multi-angle light scattering (MALS) coupled to SEC or asymmetrical flow-field flow fractionation (AF4) results in more precise information and has been conveniently used to evaluate the effect of expression, purification, and storage protocols on the monodispersity of recombinant proteins (Zhou, et al., 2006, Pavisic, et al., 2010, Chen, et al., 2016, Amartely, et al., 2018). Dynamic light scattering (DLS) is another standard technique available for measuring the size of particles in solution and frequently used for detecting the presence of protein soluble aggregates. However, it has been criticized for its bias towards larger aggregates and alternative methods such as Taylor dispersion analysis have been suggested (Hulse, et al., 2013). 4. Protein aggregation as IBs 13 Since their discovery as a common outcome during overexpression of heterologous proteins in E. coli, the perception on the composition and structure of IBs has dramatically changed. The aggregation process was not reversible by simple dilution of the polypeptide chains. Therefore, formation of protein aggregates in cell factories could not be attributed to a salting-out process as usually observed when proteins exceed their solubility limits. Moreover, it soon became obvious that active protein could be recovered following denaturation-renaturation protocols (Marston, 1986, Kane & Hartley, 1988, Schein, 1989). This suggested that the polypeptide chains within IBs were not covalently damaged and the information for reaching the native state was still present in the IB polypeptides. Due to the fact that IBs were formed starting from a vast diversity of recombinant proteins unrelated in sequence, size, structure and origin, it was initially thought that the chains engaged within the aggregates were adopting unfolded or misfolded conformations associated in a form of “statistical spaghetti”, that was escaping cell quality control. This was subsequently challenged by deep studies of the intracellular folding and aggregation of a unique model system at the time, the tailspike protein of bacteriophage P22. These revealed that aggregates originated from early, soluble partially folded intermediates in the folding pathway that could either proceed further to the native pathway or alternatively, form intermolecular aggregates (Haase- Pettingell & King, 1988 ). The notion of the competition between productive and aggregation pathways was subsequently introduced, with the partitioning being dependent on temperature, concentration, etc. This partitioning was also influenced by point mutations that were either favoring or disfavoring the aggregation pathway (Goldenberg & King, 1982 , Goldenberg, et al., 1983 , Yu & King, 1984, Yu & King, 1988, Mitraki, et al., 1991 ). The notion of specificity of aggregation has subsequently emerged when heterogeneous co-aggregates could not be observed in mixed renaturation experiments of entirely different proteins (Wetzel, 1994 , Wetzel, 1996, Betts, et al., 1997, Fink, 1998). This also implied that IBs were not amorphous protein deposits, but they were built through specific association of structural segments that could even retain “quasi-native” features (Mitraki & King, 1989, King, et al., 1996 Jan). 14 4.1 Amyloidal architecture of IBs It was well known that the target recombinant proteins were the major constituent of IBs (Ventura & Villaverde, 2006). However, their structural characterization in vivo was very challenging, due to the complexity of the folding and aggregation processes under molecular crowding conditions, the presence of heterologous proteins, and the dynamic nature of the process. Many methods for the characterization of protein aggregates can be applied only to dilute protein solutions in vitro and were not suitable to evaluate the insoluble, quasi-solid nature of the IBs. The first structural insights became possible thanks to Fourier Transform Infrared Spectroscopy (FTIR) that evidenced how IB fractions displayed either increased intermolecular beta-sheet structure or, alternatively, native-like spectra. This suggested the occurrence of both amyloidal protein forms and folded polypeptides (Oberg, et al., 1994, Przybycien, et al., 1994, Fink, 1998). The possibility of recording FTIR spectra in intact cells enabled also following the kinetics of IB formation in real time (Ami, et al., 2005, Ami, et al., 2006). In a landmark 2005 study, it was reported that IBs had intermolecular beta- structure as probed by FTIR and were binding to amyloid-diagnostic dyes such as Congo Red and Thioflavin T (Carrio, et al., 2005). Moreover, following extraction from cells, they were able to seed formation of amyloid fibrils from homologous soluble counterparts in vitro, suggesting an amyloid-like character (Carrio, et al., 2005). In subsequent electron microscopy studies, freshly extracted IBs appeared as electron dense, round objects from which fibrillar structures were occasionally emanating (Figure 3A). These fibrillar structures became more evident after treatment with proteinase K (Morell, et al., 2008). X-ray fiber diffraction studies showed the characteristic cross-beta pattern, confirming the amyloid structure of the material (Wang, et al., 2008). When model fluorescent proteins such as GFP were fused to short aggregating peptides, such as Alzheimer’s Aβ42 or the foot-and-mouth disease VP1 capsid protein, the formed IBs retained fluorescence. This fact indicated that aggregation was driven, and that it exclusively involved the aggregation-prone peptides. The fusion moieties represented by the intrinsically soluble GFP remained free to adopt native structure inside the protein clusters (Garcia-Fruitos, et al., 2005). The preference for aggregation in vivo of molecules belonging to the same species was 15 addressed by Fluorescence Energy Transfer (FRET) experiments of fusion proteins expressed in E coli cells. FRET is a sensitive method to probe close proximity of fluorescent proteins since it occurs between two fluorophores when located in the range of 10-100 Å from each other. Three-fold higher FRET was observed when two different fluorescent proteins were fused to the same aggregation-driving domain (ie either Aβ42 or VP1) compared to FRET observed upon co-expression of fusions of the same fluorescent proteins to two different aggregation domains. Thus, the requirement for sequence similarity for aggregation reinforced the notion of the stereospecific intermolecular contacts driving IB formation (Morell, et al., 2008). Sequence-specific structural information on the aggregated material was then made possible thanks to the development of methods such as Hydrogen/Deuterium exchange coupled with Nuclear Magnetic Resonance (NMR) and solid-state NMR (Wang, et al., 2008). Regardless of the initial conformation of the starting proteins, short (7-10 amino acids) stretches were found to form segments protected from hydrogen-exchange and able to mediate intermolecular beta-structure and IB formation (Wang, et al., 2008). IBs formed from the prion domain of the Het-s protein from the filamentous fungus Podospora anserina and amyloid filaments formed by the same domain in vitro share the same amyloid structure as suggested by solid-state NMR studies (Wasmer, et al., 2009). The conformation of selected residues within IBs formed by Influenza virus hemagglutinin expressed in E. coli was studied with solid- state NMR in whole cells and native-like, alpha helical conformation was reported for four selected residues (Curtis-Fisk, et al., 2008). Short sequences engaged in beta- amyloid structures could apparently coexist with native – like conformation originated from the rest of the polypeptide chain, or parts of it. The latest perception on the structure of IBs converges towards a model where beta- sheet, amyloid-structured parts and native-like, or even biologically active parts coexist (Gonzalez-Montalban, et al., 2007, Villaverde, et al., 2015). This conveys both mechanical stability and functional properties, allowing recovery of biological activity with a combination of mild solubilization conditions and traditional denaturation- renaturation protocols (Peternel, et al., 2008, Singh, et al., 2015). Finally, the application of analytical methods such as contact angle measurements, zeta potential 16 and nanoindentation allowed the evaluation of IB wettability, surface properties, and stiffness (Diez-Gil, et al., 2010). Thus, the view of IBs as particulate, porous and highly hydrated bio-nanomaterials with interesting biological properties and a high mechanical stability has recently emerged. This awareness opened unexpected and promising avenues for IB use as immobilized biocatalysts, tissue engineering bioscaffolds or nanopills for controlled protein drug release (for reviews, see (Mitraki, 2010, Garcia-Fruitos, et al., 2012, Loo, et al., 2015, Rinas, et al., 2017)), as discussed in following sections. 4.2 Biophysical nanoscale properties of IBs As discussed above, the target recombinant protein is the major component of bacterial IBs (Neubauer, 2006). Its relative abundance in IBs varies according to the amino acid sequence (Peternel, et al., 2008) and it also depends on the conditions of bacterial culture and gene expression (Strandberg & Enfors, 1991, Baig, et al., 2014, Bakholdina, et al., 2018). Bacterial strains with mutations affecting protein folding and metabolism (e.g. deficiencies in the chaperones DnaK and ClpA, or in the cytosolic protease ClpP), also influence the IBs content (Carrio & Villaverde, 2003, Martinez- Alonso, et al., 2008, Garcia-Fruitos, et al., 2012), that in turn impact on several biophysical and chemical properties of IBs and on their use as nanomaterials (Diez-Gil, et al., 2010). In this context, the shape and size of IBs are very much dependent on the particular target protein, the host bacterial strain as well as the production time (Garcia-Fruitos, et al., 2009, Diez-Gil, et al., 2010, Garcia-Fruitos, et al., 2010). Different authors described slightly different IB shapes, from spherical to ellipsoidal, cylindrical and even tear-shaped (Bowden, et al., 1991, Carrio, et al., 2000, Garcia-Fruitos, et al., 2007, Garcia-Fruitos, et al., 2010, Peternel & Komel, 2010), with a diameter that is limited by the size of the bacterial cell and ranges from 50 nm (probably the detection limit under optical microscope) to about 1,000 nm (Peternel, et al., 2008, Garcia-Fruitos, et al., 2010). On the other hand, the size of the IBs is significantly reduced in vitro in low pH buffers as the result of an irreversible contraction process (Peternel, 2008). This peculiar condensation event affects the density and solubility of the IBs and it might be 17 of relevance for further in vitro and in vivo applications of IBs as materials. After isolation from bacterial cells, they are stable and can be stored for long periods at 4 °C, frozen at -80°C, or lyophilized, keeping the original IB architecture and the biological activity of the embedded proteins (Garcia-Fruitos, et al., 2009). IB proteins are usually not released in physiological buffers (Garcia-Fruitos & Villaverde, 2010), and such compositional stability is the basis for the exploitation of enzymatically active IBs as reusable catalysts (Krauss, et al., 2017, Rinas, et al., 2017). This concept of IBs as self-immobilized enzymes shows room for further improvement regarding operation reusability (Koszagova, et al., 2018), scalability (Kloss, et al., 2018) and functional complexity (Jager, et al., 2018). However, a fraction of IB protein is released from IBs in aqueous solutions (Garcia-Fruitos, et al., 2007, Unzueta, et al., 2018), and the extent of such removable protein population appears to be dependent on the nature of the target protein, which influences the percentage of native-like IB protein. It is also dependent on factors such as the genetic background of the producer strain and the pH of the storage buffer (Garcia-Fruitos, et al., 2012). As materials, some IBs are loose, and as indicated earlier they can be dissolved by mild detergents or high pH buffers (Jevsevar, et al., 2005). This condition represents an opportunity to recover soluble IB protein in vitro avoiding protein denaturing agents (Singh, et al., 2015, Singh, et al., 2015). The particular sensitivity of IBs to such buffers and other mild-to-harsh treatments must be carefully considered to preserve the IB structure during purification (Rodriguez-Carmona, et al., 2010, Rodriguez-Carmona, et al., 2011, Seras- Franzoso, et al., 2015), especially if intended for use as functional materials. The modification in the host cells of the genetic background relative to components involved in the protein quality control impacts not only on the efficiency of protein removal from IBs, but also on their functionality and physicochemical parameters such as zeta potential (surface charge) (Novak, 2009, Diez-Gil, et al., 2010), stiffness (elasticity) (Diez-Gil, et al., 2010) and wettability (ability of an aqueous liquid to maintain contact with IBs’ surface) (Diez-Gil, et al., 2010). For instance, IBs recovered from wild type cells are homogenous, while in DnaK- and ClpA- cells two IB populations with distinguishable stiffness and which remained segregated were produced (Diez-Gil, et al., 2010). These mechanical properties are extremely important when using IBs as 18 anchoring structures for the cultivation of the mammalian cells (Diez-Gil, et al., 2010, Seras-Franzoso, et al., 2012, Seras-Franzoso, et al., 2013, Tatkiewicz, et al., 2013, Tatkiewicz, et al., 2018), as stiffer IBs better stimulate cell attachment and/or proliferation (Novak, 2009). 5. Physiology of IB formation Protein aggregates formed in recombinant bacteria are not mere passive deposits of insoluble protein, but the result of complex active physiological processes that link protein folding, quality control, conformational stress and amyloid formation. The physiology of IB formation is still far to be precisely dissected but the comprehension of the process has progressively increased. 5.1 Energetics of IB formation IBs are usually found at the polar sites of the cell and it is generally accepted that polar preference of IBs or other protein aggregates can be attributed to macromolecular crowding in the midcell nucleoid region (Lindner, et al., 2008, Winkler, et al., 2010, Coquel, et al., 2013, Gupta, et al., 2014, Neeli-Venkata, et al., 2016, Oliveira, et al., 2016). Occasional midcell detection of IBs can be explained by nucleoid-free regions at potential future septation sides (Winkler, et al., 2010, Coquel, et al., 2013). The hypothesis of crowding-provoked polar preference has never been challenged but the way protein aggregates reach the poles has been a matter of controversial debate as contradicting results were obtained suggesting an either energy-independent or energy-dependent mechanism. For example, it was demonstrated that polar localization of aggregates occurred also in energy-deprived cells, indicating a passive mechanism for polar positioning (Winkler, et al., 2010). Also, experimentally observed movements of age-related aggregates towards the poles were explainable with passive diffusion of aggregates and spatially non-homogeneous macromolecular crowding (Coquel, et al., 2013). On the other hand, there was also evidence that transport of aggregates to the poles can be an energy-driven process or at least dependent on metabolic activity (Rokney, et al., 19 2009, Govers, et al., 2014, Govers, et al., 2017). In this context, studies on cells with pressure dissociated IBs revealed that reassembly of small aggregates into large IBs did occur in cells exposed to fresh nutrients but not in energy and nutrient depleted cells (Govers, et al., 2014). These seemingly contradictive findings relatively to purely diffusive movement or energy-driven transport of aggregates towards the cellular poles become understandable by considering the glassy nature of the bacterial cytoplasm. Physical- chemical studies have shown that metabolic activity is required to maintain fluidity in the cytoplasm, a condition necessary for the movement of larger particles (> 30 nm) (Parry, et al., 2014). Thus, it is reasonable to guess that it is not the active ATP- dependent transport of aggregates but rather the fluidizing properties of active metabolism which facilitate movements of aggregates towards less crowded regions. Passive diffusion of aggregates towards the pole, a region of less macromolecular crowding, would then depend on the fluidity of the cytoplasm as well as on the initial size of aggregation foci. Smaller aggregation foci could move to the poles by diffusion also in energy-deprived cells, but larger aggregation foci could only reach the poles in metabolically active cells (Govers, et al., 2014). Polar distribution of IBs or damaged proteins can be considered as an evolutional benefit compared to unbiased dilution of misfolded and aggregated proteins as continued cell division leads to rejuvenation (Lindner, et al., 2008, Rokney, et al., 2009, Winkler, et al., 2010). 5.2. Quality control and IB formation The bacterial protein quality control machinery, composed by a complex catalogue of chaperones and proteases with overlapping activities, has an important role in protein folding and refolding in the bacterial cytoplasm during conformational (thermal) stresses and especially during the production of aggregation-prone recombinant proteins. DnaK/DnaJ/GrpE (the KJE set), GroELS, ClpB, and IbpAB are main chaperones acting during this process (Figure 2 B) and they are over-produced under recombinant protein production (Allen, et al., 1992, Rinas, 1996, Veinger, et al., 1998, Goloubinoff, et al., 1999, Houry, 2001, Mogk, et al., 2003, Jurgen, et al., 2010, Zblewska, et al., 2014). As these chaperones might be limiting for proper protein folding, many 20 biotechnologists have been exploring their co-production along with the recombinant protein, aiming at optimizing the production of soluble and properly folded soluble species (Sorensen & Mortensen, 2005, Sorensen & Mortensen, 2005, Kolaj, et al., 2009). However, the role of chaperones and proteases goes beyond the strict control of the correct folding and refolding of misfolded soluble proteins. It has been widely proven that soluble and insoluble (aggregated) proteins are in a dynamic equilibrium between protein deposition and removal (Schrodel & de Marco, 2005) and the protein quality control machinery controls not only protein aggregation, but also protein quality in both soluble and aggregated fractions (Garcia-Fruitos, et al., 2007, Rinas, et al., 2007, Baig, et al., 2014). Specifically, it has been described that pivotal chaperones such as DnaK, GroEL, and ClpB (heat shock proteins -Hsp-) and IbpAB (small heat shock proteins -sHsp-) are intimately associated with IBs, which confirms that they have an active role in the formation and disaggregation of these protein clusters (Allen, et al., 1992, Mogk, et al., 2003, Mogk, et al., 2003, Mogk, et al., 2003, Carrio & Villaverde, 2005, Jurgen, et al., 2010, Zblewska, et al., 2014). In this context, upon the arrest of de novo protein synthesis, different members of the protein quality control are actively involved in the disintegration of IBs (Carrio & Villaverde, 2001). In parallel, and more specifically, the coordinated action of ClpB (disaggregase), KJE, and sHsp promotes the removal of polypeptides from protein aggregates (Mogk & Bukau, 2004, Weibezahn, et al., 2004). DnaJ binds to the aggregate surface, and it recruits first DnaK and finally ClpB to form an energy- dependent chaperone complex whose coordinated action promotes protein removal (Jurgen, et al., 2010, Aguado, et al., 2015, Aguado, et al., 2015). In absence of functional DnaK or ClpB, the disaggregation activity is minimized, while IbpAB mutants have just a mild effect on the disaggregation process (Mogk, et al., 2003, Mogk & Bukau, 2004, Gonzalez-Montalban, et al., 2008, Garcia-Fruitos, et al., 2010). This indicates that IbpAB is not essential in the disaggregation process, although its presence can notably enhance the action of ClpB/KJE (Mogk, et al., 2003, Mogk & Bukau, 2004). On the other hand, protein disaggregation is clearly an ATP-driven process (Rokney, et al., 2009, Winkler, et al., 2010). For example, HtpG can interact with DnaK/DnaJ/GrpE to further promote refolding of aggregated proteins in an ATP- 21 dependent manner (Genest, et al., 2011). Alternatively, ClpB can also assist in cooperation with DnaK/DnaJ/GrpE in ATP-driven refolding of aggregated proteins (Ben-Zvi, et al., 2004, Acebron, et al., 2009, Winkler, et al., 2010). The small heat shock proteins (sHSPs; IbpA/IbpB) stabilize and decrease the size of protein aggregates (Ratajczak, et al., 2009) thereby promoting their DnaK/DnaJ/GrpE and ClpB mediated refolding (Matuszewska, et al., 2005). The major disaggregating chaperones (DnaK, ClpB) (Winkler, et al., 2010) but also IbpA (Lindner, et al., 2008) co-localize at the poles and participate in the disintegration of polar aggregates (Rokney, et al., 2009). In absence of DnaK, IBs are unusually large (Figure 2B) and the specific activity of the forming protein is significantly higher than in the wild type E. coli strain (Gonzalez- Montalban, et al., 2006, Garcia-Fruitos, et al., 2007, Rinas, et al., 2007). In contrast, GroEL and IbpAB mutant strains form IBs in which the trapped enzyme shows the same specific activity than that found in the wild type E. coli strain (Kuczynska-Wisnik, et al., 2004, Garcia-Fruitos, et al., 2007, Rinas, et al., 2007). The genetic background has also been proven to determine the final morphology of IBs. While most of IBs have a spherical-like form, indicating a homogenous deposition and removal of protein on the whole IB surface, a particular tear-shape morphology has been observed in ClpP- deficient bacterial cells (Garcia-Fruitos, et al., 2010). At this point it is important to emphasize that proteases such as ClpP and Lon play also a role in the IB formation and intracellular solubilization (Carrio & Villaverde, 2003, Garcia-Fruitos, et al., 2010). Surprisingly, once proteins are physiologically released from IBs, they are proteolyzed irrespective of if they are misfolded or properly folded (and consequently functional) (Garcia-Fruitos, et al., 2010). Such uncontrolled proteolysis is tightly regulated by DnaK, which is anchored to the IB surface (Carrio & Villaverde, 2005). In the absence of this chaperone, the proteolysis is significantly reduced (Garcia-Fruitos, et al., 2010). Moreover, an excess of DnaK by controlled overexpression in IB-producing bacterial cells has been suggested to inhibit proper in situ folding of proteins forming such aggregates (Gonzalez-Montalban, et al., 2008). It appears evident from this schematic summary that it is not sufficient inducing the overexpression of single (or combinations of) chaperones/proteases to improve automatically recombinant protein folding. Automatized approaches which enable a large array of expression combinations are 22 probably the most efficient approach to identify the optimal conditions for each single protein (Turchetto, et al., 2017). 5.3 Controlled IB production As discussed in previous sections, the formation of IBs is dependent on the quality control system and linked to an orchestrated cell response to stress factors. The identification of these factors opened a door to the engineering of production processes to favour the formation of IBs with defined properties, by adjusting the parameters affecting their formation. Paradoxically, many of them were recognized during the attempts to enhance the yield of soluble protein (Sorensen & Mortensen, 2005, Sorensen & Mortensen, 2005). When visualizing the protein production process as a whole, the rate at which the proteins fold in the cell can be reduced by adjusting the growth temperature of the culture, the strength of the gene promoter or even the medium composition. In the first case, the growth rate of bacteria directly affects the metabolism of the cell and specifically protein synthesis (Bosdriesz, et al., 2015, Calcines-Cruz, et al., 2018, Wurm, et al., 2018). In the second scenario, both the transcription and translation rates for the nascent polypeptide are reduced improving the performance of the folding modulators in the same way. In addition, the same effect has been observed when medium composition reduces the cell growth (Jhamb & Sahoo, 2012) or when the inducer concentration is added in limiting amounts (Rabhi-Essafi, et al., 2007). In the same line, the possibility to increase the yield of insoluble protein has been performed by the fusional addition of aggregation or pull- down tags to the recombinant protein (Nahalka & Nidetzky, 2007, Zhou, et al., 2012, Costa, et al., 2014, Wang, et al., 2015, Jong, et al., 2017). 5.4 Tailoring IBs IBs are formed during protein production processes carried out under very diverse culture conditions, targeted to a vast diversity of proteins and supported by multiple expression systems (Taylor, et al., 1986, Georgiou & Valax, 1999, Castellanos-Mendoza, et al., 2014). By comparing production conditions, it has been determined that protein 23 aggregation and IB formation are favoured by factors affecting the transcriptional and translational regulation of the recombinant protein. These include the number of copies of the target gene (gene dosage), the promoter strength and the extent of the induction stimuli, the reducing conditions of the cytosol of the bacteria and the physicochemical characteristics of each particular protein (Mitraki, et al., 1991 , Przybycien, et al., 1994, Carrio & Villaverde, 2005, Ventura & Villaverde, 2006, Ignatova, et al., 2007, Nahalka, et al., 2008, Wang, et al., 2008, Tyedmers, et al., 2010, Winkler, et al., 2010, Singh, et al., 2015). The environment can also influence IBs formation. In fact, during recombinant protein production, temperature, culture time, nature of the inducer, growth rate, agitation and pH (Table 1, and Supplementary Table 1), have a profound impact on architectonic IB properties such as size, inner structure, content of active protein and their potential for being released or solubilized (Garcia-Fruitos, et al., 2005, Garcia-Fruitos, et al., 2005, Margreiter, et al., 2008, Peternel, et al., 2008, Upadhyay, et al., 2012, Castellanos-Mendoza, et al., 2014, Krauss, et al., 2017, Valdez-Cruz, et al., 2017, Calcines-Cruz, et al., 2018). The increase in IB size has been related with progression of the culture time, being more noticeable in the first hours after recombinant protein synthesis induction (Carrio, et al., 1998, Margreiter, et al., 2008, Garcia-Fruitos, et al., 2009, Upadhyay, et al., 2012, Castellanos-Mendoza, et al., 2014). However, it has been demonstrated that IBs populations are dynamically varying respect to time, being able to increase and decrease in the same process (Castellanos-Mendoza, et al., 2014). Also a rapid bacterial growth rate enhances the aggregation and the accumulation of recombinant proteins as IBs (Iafolla, et al., 2008). During culture, the increase in inducer concentration (mostly exemplified by IPTG) tends to favour the growth of IBs by a hundred of nanometers (Luo, et al., 2006), while at lower concentrations causes a reduction of the IB size combined with higher density of the recombinant protein within them (Margreiter, et al., 2008, Jhamb & Sahoo, 2012). 24 On the other hand, temperature is a widely used parameter to externally manipulate protein aggregation. Growth of E. coli at low temperature (between 16 °C and 30 °C) minimizes the formation of IBs or enriches them with partially or fully folded proteins, enhancing their biological activity (Jevsevar, et al., 2005, de Groot & Ventura, 2006, Vera, et al., 2007, Peternel, et al., 2008). Likewise, low temperature enlarges the extractable protein fraction from IBs, through the steps of solubilization and renaturation (Jevsevar, et al., 2005, Peternel, et al., 2008). On the contrary, when cultures are maintained at 37°C (or higher), IB formation is promoted as well as an increase of impurities inside them (Strandberg & Enfors, 1991, Valax & Georgiou, 1993, Villaverde, et al., 1993). Even more, IBs formed at 42°C present greater resistance to solubilization and denaturation compared with those formed at 25°C, indicating that the culture temperature determines the conformational properties and stability of the proteins trapped within aggregates (Jevsevar, et al., 2005, de Groot & Ventura, 2006, Vera, et al., 2007, Peternel, et al., 2008). pH has also been examined regarding the formation and quality of IBs (Strandberg & Enfors, 1991, Castellanos-Mendoza, et al., 2014, Calcines-Cruz, et al., 2018). When the pH dropped below 5.5, an increase in the relative amount of IBs was observed compared with cultures developed at constant pH, upon temperature-mediated induction of recombinant gene expression (Strandberg & Enfors, 1991). Differences in the size and architecture of the IBs were determined when comparing bioreactor cultures under controlled pH (7.5) and uncontrolled pH. Those IBs formed under uncontrolled increasing alkaline conditions (up to 8.5) have a lower content of amyloid structures and are easier to solubilize using proteinase K or to denature by chaotropic agents if compared with those IBs recovered from bacteria cultured at controlled pH (Castellanos-Mendoza, et al., 2014). Similar results were found when IBs were produced under constant basic pH of 8.5, using different model proteins and genetic backgrounds (Calcines-Cruz, et al., 2018). The architecture and size of IBs can be also affected by oxygenation conditions of bacterial culture in shake flasks. When orbital (200 rpm) and resonant acoustic mixing (RAM at 20 g forces) were used, a decrease in IB size combined with higher number of small aggregates was observed throughout the cytoplasm. In addition, the smaller 25 bodies formed under these conditions were easier to solubilize, and showed enrichment in alpha helices compared with those formed under orbital agitation or under lower g forces in RAM (Valdez-Cruz, et al., 2017). Culture conditions represent factors to exploit for improving the IB characteristics according to the final applications foreseen for such “raw material”. For instance, the modification of process parameters allows adjusting the conformational (and functional) quality of the IB proteins and more efficient protein extraction (Peternel, et al., 2008, Dong, et al., 2014, Raghunathan, et al., 2014, Krauss, et al., 2017, Calcines- Cruz, et al., 2018). 6. Applicability of IBs as non-toxic, functional amyloids Bacterial IBs are mechanically stable protein materials (Garcia-Fruitos, et al., 2009). Such structural robustness is the basis for their emerging applications as a versatile microbial biomaterial in biomedicine (Rodriguez-Carmona & Villaverde, 2010) and specially as a topographic decorator in tissue engineering (Seras-Franzoso, et al., 2015). The firm attachment of functional enzymes to the amyloidal scaffold of IBs (Cano-Garrido, et al., 2013) is, on the other hand, supportive of their application as self-immobilized catalysts (Rinas, et al., 2017). The differential sensitivity of amyloidal and non-amyloidal forms of IB proteins to proteinase K (Cano-Garrido, et al., 2013) confirms the dual structural composition of the materials, that might be linked to the occurrence of harder and softer areas in the particles (Diez-Gil, et al., 2010). 6.1 Protein release from IBs The non-amyloidal fraction of IB protein appears being in a looser conformational status supported by weaker protein-protein contacts than the cross-molecular beta- sheet architecture of the amyloidal fibril skeleton (Carrio, et al., 2005, Gonzalez- Montalban, et al., 2006, de Groot, et al., 2009, Pesarrodona, et al., 2016, Rueda, et al., 2016). Upon internalization by cultured mammalian cells (Figure 3B), an important fraction of IB protein is released from the particles (Figure 3C), probably supported by the action of cell factors, presumably chaperones (Seras-Franzoso, et al., 2016), in the 26 context of the evolutionarily conserved quality control machinery (Mogk, et al., 2018). The ability of this fraction of IB proteins to be released under physiological conditions has opened a spectrum of possibilities both in vitro and in vivo to (i) recover functional protein from IBs under non-denaturing conditions (Peternel, et al., 2008) and (ii) explore IBs as mimetics of secretory granules of the endocrine system for the sustained release of functional proteins in the body (Unzueta, et al., 2018), respectively. 6.2 Controlled in vitro release of IB protein in biotechnology In the context of IB protein recovery, classical denaturation and renaturation procedures are time consuming and result into a limited, product-dependent recovery of the functional protein (Vallejo & Rinas, 2004). The recognition of IBs as nanoparticles composed by properly folded proteins prompted the investigation of simpler methods to extract them from the bulk material, as the disruption of the protein structure might be not required. In fact, washing IBs with physiological buffers with mild denaturants enables the removal of functional proteins in a single step procedure (Singh, et al., 2015). Producing IBs in conditions that favour proper protein folding (for instance, low growth temperatures) enriches them with functional polypeptides at expenses of amyloidal structure. However, these highly functional IBs can be quite fragile, and their isolation from bacterial cells becomes a critical step. Lysozyme commonly used for bacterial lysis can adhere to the IB surface, representing an additional and major impurity. On the other hand, sonication can cause the loss of the target protein, and high-pressure homogenization has been recognized as the most suitable cell disruption method for looser IBs (Peternel & Komel, 2010). Upon IB isolation, mild detergents can be used for IB solubilization and high recovery rates of extremely pure and biologically active protein have been achieved in this way (Jevsevar, et al., 2005). In pioneering studies, various chemicals in slightly alkaline buffer (40 mM Tris/HCl, pH 8), that do not disturb the protein structure (5% n-propanol, 2 M urea, 0.5% Triton X-100, 0.2% NDSB, 5 % DMSO and 0.2% N-lauroyl-sarcosine) were tested to extract biologically active granulocyte colony stimulating factor (G-CSF) from IBs, representing up to 50 % of the 27 whole protein yield (Jevsevar, et al., 2005). Similar procedures for non-denaturing solubilization were later described for other proteins (Peternel, et al., 2008, Upadhyay, et al., 2014) (1,2,4,5). Besides low concentration of chaotropes and mild detergents, also alkaline pH in combination with low concentration of urea (Singh, et al., 2015), high pressure (St John, et al., 1999), and organic solvents (Singh, et al., 2012) have been used for successful recovery of bioactive proteins from IBs (Singh, et al., 2015). (Singh, et al., 2015). On the other hand, mild extraction does not work for all IB embedded proteins; for instance, the bone morphogenetic protein 2 which still requires harsh extraction and subsequent refolding under appropriate redox conditions (Quaas, 2018). 6.3 IB protein release in vivo for biomedicine The nanopill concept was around the idea that IBs, as functional protein-releasing amyloids, might be exploited for protein replacement therapies. In this context, IBs formed by functional proteins would physiologically rescue cells with deficiencies caused by either genetic (missing protein functions) or environmental (external stresses) conditions, by acting as carriers of proteins that re-establish cellular homeostasis. A deep interaction (and further embedment) between IBs and mammalian cell membranes in cultured cells was strongly suggesting that at least a significant fraction of IB material might be available intracellularly (Garcia-Fruitos, et al., 2009). As a proof of concept, IBs formed by dihydrofolate reductase, catalase, leukaemia inhibiting factor and Hsp70, once added to the medium of cultured cells were able to rescue cell viability in different experimental systems (Vazquez, et al., 2012). In parallel, it was also demonstrated that IBs composed by the structural cytoskeleton protein keratin 14 released enough correctly folded protein to form heterodimers together with the fluorescent keratin 5, and thus built up fluorescent precursors of the intermediate filaments in epithelial cells (Liovic, et al., 2012). Other interesting applications in regenerative medicine have emerged later, in the context of wound healing (Figure 3D, (Stamm, et al., 2018)). In this context, bacterial IBs appeared suitable also as structural and functional biomimetics of the secretory granules in the mammalian endocrine system by releasing protein hormones (urotensin, glucagon, 28 obestatin, bombesin and secretin among others, Figure 3A). Under physiological conditions these are kept in the body as amyloidal storage units to be released upon appropriate stimuli with the assistance of cell chaperones (Maji, et al., 2009). Subsequent studies have shown that nanopills are functional in delivering biologically active proteins not only upon free addition to the culture medium (top-down), but also when decorating surfaces on top of which cells grow (bottom-up) (Seras-Franzoso, et al., 2013). In this second case, nanopills act rather as bioactive topographic scaffolds or bioscaffolds (Seras-Franzoso, et al., 2013) with a dual effect. On the one side, they support cell proliferation due to physical stimuli promoted by modifications of the surface nanotopography and that favour both cell adhesion and the activation of proliferative cascades (Seras-Franzoso, et al., 2012). On the other hand, growth or differentiation protein factors released by IB can induce a direct biological effect. The functionalization of 2D and 3D structures with FGF-2 bioscaffolds favoured the growth and differentiation of epithelial cells, fibroblasts and neuron-like cells (Seras-Franzoso, et al., 2013, Seras-Franzoso, et al., 2013, Seras-Franzoso, et al., 2014). IBs releasing the lipoxygenase AmbLOXe have been recently proved to be effective in wound repair (Stamm, et al., 2018), in both nanopill and bioscaffold formats. The way through which nanopills and bioscaffolds perform their function is dual. Experimental data support protein released by IBs penetrating exposed cells and acting from within as well as accumulating in the culture medium and acting extracellularly. In this regard, free IBs interact tightly with the cell membrane through cell-emitted filopodia which promote IB internalization by micropinocytosis (Seras-Franzoso, et al., 2012, Vazquez, et al., 2012, Seras-Franzoso, et al., 2016, U, et al., 2016). Although part of the IB protein is degraded in lysosomes, significant amounts of IB protein are still detectable intracellularly by immunohistochemistry, once released from IBs into the cytoplasm after the disruption of the endosomal membrane (Figure 3C, (Seras- Franzoso, et al., 2016)). A similar event, although with additional steric constraints, probably occurs with bioscaffolds (Seras-Franzoso, et al., 2013). Interestingly, the biological effect of bacterial amyloids increases when IBs are produced at temperatures below 37 °C (Seras-Franzoso, et al., 2014). This is because the 29 conformational and functional quality of IB proteins is higher under these conditions (Vera, et al., 2007), as discussed above. A step forward in the design of improved nanopills has been the generation of cell targeted-IBs that preferentially bind to a specific cell type and enable targeted protein delivery. In this context, soluble glycoproteins have been covalently conjugated to therapeutic IBs for the recognition of adhesins on the surface of pathogens. A self- cleaving peptide has been included to release the functional IB protein after pathogen recognition, creating in this way a pioneer tailored and active protein delivery platform (Talafova, et al., 2013). In a further step to tailor IBs, the fusion of a ligand for a cell surface receptor to the IB-forming protein enabled receptor-dependent IB internalization and cell targeted delivery without requiring chemical modifications of IBs after their purification (U, et al., 2016). In vivo studies have proved the potential of IBs as protein releasing materials, mimicking the in vivo secretion of protein hormones. This effect has been simulated by intratumoral injection and also as subcutaneous nanoimplantation of functional IBs in different murine cancer models. When fluorescent (GFP- or IRFP-based) IBs are injected intratumorally, the emission remains within the tumor for at least one week, and does not induce apparent side effects. This observation indicates that IBs might represent a stable local source for releasing functional protein without toxicity effect. In addition, in this pioneering experiment, there was a certain in situ antitumor effect derived from the tumor receptor blockade induced by the peptide T22 contained in the IBs and used as selective binder for the tumoral marker CXCR4 (Cespedes, et al., 2016). In a more challenging approach, it was possible to demonstrate that IB protein could act systematically moving through the blood stream. As an intriguing platform, IBs formed by self-assembling proteins (Figure 3E) that organize as nanoparticles (Pesarrodona, et al., 2016, Unzueta, et al., 2018) were used. When such IBs were applied locally as subcutaneous implants far from the primary tumor in mice, the active protein was progressively and specifically transferred over a period of more than ten days from the injection site to the tumor (Unzueta, et al., 2018). This fact supports that IBs could be used in vivo as stable and biocompatible implantable materials for remote delivery of nanostructured protein drugs and materials (Figure 30 3F). The production of IBs and IB-like materials in endotoxin-free strains of E. coli (Rueda, et al., 2014), in food-grade gram positive acid lactic bacteria (Cano-Garrido, et al., 2014, Cano-Garrido, et al., 2016) and in conventional yeast factories (Rueda, et al., 2016) will minimize the potential regulatory constraints for further in vivo applications of IBs. 7. Conclusion Bacterial IBs were observed, since early times of recombinant protein technology, as an obstacle for the production of functional proteins. A limited comprehension of the protein production machinery wrongly positioned them as waste material excluded from the cell quality control, and catalogued IBs as protein clusters composed of irreversibly misfolded or unfolded polypeptides. Gaining insights about the physiology of recombinant cell factories and also about proteomics of protein aggregates shifted this original perception to the current status in which IBs are included in a specific category of functional amyloids. A particular combination of amyloid fibrils and functional structures of the recombinant protein confers to them intriguing properties as mechanically stable and functional materials. Such unusual features point out bacterial IBs as convenient models to study how functional amyloids (especially those used in nature as protein depots) are formed and structured but also enable their exploitation in different areas of biotechnology and medicine. IBs, as protein-releasing materials, are fully biocompatible, and their properties can be modulated through the genetic engineering of the employed proteins, of the host cells, and of the bacterial culture process. In such a way, it is possible to generate nanoparticles that can be targeted in vivo and used as source of functional protein. Acknowledgments The authors appreciate the financial support for research on therapeutic recombinant proteins to NFM (grant RTA2015-00064-C02-02, INIA, MINECO, Spain) to EGF (grant RTA2015-00064-C02-01, INIA, MINECO, Spain, CERCA Programme -Generalitat de Catalunya- and European Social Fund), to EV (grant PI15/00272, ISCIII, Spain, co- 31 founding FEDER), to AdM (Slovenia-Belgium ARRS-FWO program, ARRS/N4-0046-5100- 1/2015-59), and to AV (grant BIO2016-76063-R, Agencia Estatal de Investigación and Fondo Europeo de Desarrollo Regional, AEI/FEDER, UE; grant 2017SGR-229, AGAUR, Spain). Besides, EGF received a post-doctoral fellowship from INIA (DOC-INIA, MINECO). MATR and NAVC, thank the Institutional Program of the “Instituto de Investigaciones Biomédicas-UNAM”: “La producción de biomoléculas de interés biomédico en bacterias y hongos”. Funding “Consejo Nacional de Ciencia y Tecnología” CONACYT (220795, 247473 and 178528), and “Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica, Universidad Nacional Autónoma de México” PAPIIT-UNAM (IN-209113 and IN-208415).We are also indebted to The Networking Research Center on Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN), an initiative funded by the VI National R&D&I Plan 2008–2011, Iniciativa Ingenio 2010, Consolider Program, CIBER Actions and financed by the Instituto de Salud Carlos III with assistance from the European Regional Development Fund. This manuscript was written in collaboration with the ICTS “NANBIOSIS”, more specifically with the Unit 1: Protein Production Platform of CIBER-BBN/ IBB, at the UAB sePBioEs scientific- technical service (http://www.nanbiosis.es/portfolio/u1-protein-production-platform- ppp/). The authors do not appreciate any conflict of interest. References Acebron SP, Martin I, del Castillo U, Moro F & Muga A (2009) DnaK-mediated association of ClpB to protein aggregates. A bichaperone network at the aggregate surface (vol 583, pg 2991, 2009). Febs Letters 583: 3301-3301. Agostini F, Vendruscolo M & Tartaglia GG (2012) Sequence-based prediction of protein solubility. J Mol Biol 421: 237-241. Aguado A, Fernandez-Higuero JA, Moro F & Muga A (2015) Chaperone-assisted protein aggregate reactivation: Different solutions for the same problem. Arch Biochem Biophys 580: 121-134. 32 Aguado A, Fernandez-Higuero JA, Cabrera Y, Moro F & Muga A (2015) ClpB dynamics is driven by its ATPase cycle and regulated by the DnaK system and substrate proteins. Biochem J 466: 561-570. Allen SP, Polazzi JO, Gierse JK & Easton AM (1992) Two novel heat shock genes encoding proteins produced in response to heterologous protein expression in Escherichia coli. J Bacteriol 174: 6938-6947. Amartely H, Avraham O, Friedler A, Livnah O & Lebendiker M (2018) Coupling Multi Angle Light Scattering to Ion Exchange chromatography (IEX-MALS) for protein characterization. Sci Rep 8: 6907. Ami D, Natalello A, Gatti-Lafranconi P, Lotti M & Doglia SM (2005) Kinetics of inclusion body formation studied in intact cells by FT-IR spectroscopy. Febs Letters 579: 3433- 3436. Ami D, Natalello A, Taylor G, Tonon G & Doglia SM (2006) Structural analysis of protein inclusion bodies by Fourier transform infrared microspectroscopy. Biochimica Et Biophysica Acta-Proteins and Proteomics 1764: 793-799. Ami D, Natalello A, Schultz T, Gatti-Lafranconi P, Lotti M, Doglia SM & de Marco A (2009) Effects of recombinant protein misfolding and aggregation on bacterial membranes. Biochim Biophys Acta 1794: 263-269. Badtke MP, Hammer ND & Chapman MR (2009) Functional amyloids signal their arrival. Sci Signal 2: pe43. Baig UI, Bhadbhade BJ, Mariyam D & Watve MG (2014) Protein Aggregation in E-coli: Short Term and Long Term Effects of Nutrient Density. Plos One 9. Bakholdina SI, Sidorin EV, Khomenko VA, et al. (2018) The Effect of Conditions of the Expression of the Recombinant Outer Membrane Phospholipase D-1 from Yersinia pseudotuberculosis on the Structure and Properties of Inclusion Bodies. Russian Journal of Bioorganic Chemistry 44: 178-187. Bandyopadhyay A, Saxena K, Kasturia N, et al. (2012) Chemical chaperones assist intracellular folding to buffer mutational variations. Nat Chem Biol 8: 238-245. Baneyx F & Mujacic M (2004) Recombinant protein folding and misfolding in Escherichia coli. Nat Biotechnol 22: 1399-1408. Basu A, Li X & Leong SS (2011) Refolding of proteins from inclusion bodies: rational design and recipes. Appl Microbiol Biotechnol 92: 241-251. 33 Ben-Zvi A, De los Rios P, Dietler G & Goloubinoff P (2004) Active solubilization and refolding of stable protein aggregates by cooperative unfolding action of individual Hsp70 chaperones. Journal of Biological Chemistry 279: 37298-37303. Besingi RN, Wenderska IB, Senadheera DB, Cvitkovitch DG, Long JR, Wen ZT & Brady LJ (2017) Functional amyloids in Streptococcus mutans, their use as targets of biofilm inhibition and initial characterization of SMU_63c. Microbiology 163: 488-501. Betts S, HaasePettingell C & King J (1997) Mutational effects on inclusion body formation. Advances in Protein Chemistry, Vol 50, Vol. 50 ed.^eds.), p.^pp. 243-264. Academic Press Inc, San Diego. Bosdriesz E, Molenaar D, Teusink B & Bruggeman FJ (2015) How fast-growing bacteria robustly tune their ribosome concentration to approximate growth-rate maximization. FEBS J 282: 2029-2044. Bowden GA, Paredes AM & Georgiou G (1991) Structure and morphology of protein inclusion bodies in Escherichia coli. Biotechnology (N Y) 9: 725-730. Burgess RR (2009) Refolding solubilized inclusion body proteins. Methods Enzymol 463: 259-282. Calamai M, Canale C, Relini A, Stefani M, Chiti F & Dobson CM (2005) Reversal of protein aggregation provides evidence for multiple aggregated States. J Mol Biol 346: 603-616. Calcines-Cruz C, Olvera A, Castro-Acosta RM, Zavala G, Alagon A, Trujillo-Roldan MA & Valdez-Cruz NA (2018) Recombinant-phospholipase A2 production and architecture of inclusion bodies are affected by pH in Escherichia coli. Int J Biol Macromol 108: 826- 836. Cano-Garrido O, Rueda FL, Sanchez-Garcia L, Ruiz-Avila L, Bosser R, Villaverde A & Garcia-Fruitos E (2014) Expanding the recombinant protein quality in Lactococcus lactis. Microb Cell Fact 13: 167. Cano-Garrido O, Rodriguez-Carmona E, Diez-Gil C, et al. (2013) Supramolecular organization of protein-releasing functional amyloids solved in bacterial inclusion bodies. Acta Biomater 9: 6134-6142. Cano-Garrido O, Sanchez-Chardi A, Pares S, et al. (2016) Functional protein-based nanomaterial produced in microorganisms recognized as safe: A new platform for biotechnology. Acta Biomater 43: 230-239. 34 Carrio M, Gonzalez-Montalban N, Vera A, Villaverde A & Ventura S (2005) Amyloid-like properties of bacterial inclusion bodies. Journal of Molecular Biology 347: 1025-1037. Carrio M, Gonzalez-Montalban N, Vera A, Villaverde A & Ventura S (2005) Amyloid-like properties of bacterial inclusion bodies. J Mol Biol 347: 1025-1037. Carrio MM & Villaverde A (2001) Protein aggregation as bacterial inclusion bodies is reversible. FEBS Lett 489: 29-33. Carrio MM & Villaverde A (2002) Construction and deconstruction of bacterial inclusion bodies. J Biotechnol 96: 3-12. Carrio MM & Villaverde A (2003) Role of molecular chaperones in inclusion body formation. FEBS Lett 537: 215-221. Carrio MM & Villaverde A (2005) Localization of chaperones DnaK and GroEL in bacterial inclusion bodies. J Bacteriol 187: 3599-3601. Carrio MM, Corchero JL & Villaverde A (1998) Dynamics of in vivo protein aggregation: building inclusion bodies in recombinant bacteria. FEMS Microbiol Lett 169: 9-15. Carrio MM, Corchero JL & Villaverde A (1999) Proteolytic digestion of bacterial inclusion body proteins during dynamic transition between soluble and insoluble forms. Biochim Biophys Acta 1434: 170-176. Carrio MM, Cubarsi R & Villaverde A (2000) Fine architecture of bacterial inclusion bodies. FEBS Lett 471: 7-11. Castellanos-Mendoza A, Castro-Acosta RM, Olvera A, et al. (2014) Influence of pH control in the formation of inclusion bodies during production of recombinant sphingomyelinase-D in Escherichia coli. Microb Cell Fact 13: 137. Cespedes MV, Fernandez Y, Unzueta U, et al. (2016) Bacterial mimetics of endocrine secretory granules as immobilized in vivo depots for functional protein drugs. Sci Rep 6: 35765. Chang CC, Song J, Tey BT & Ramanan RN (2014) Bioinformatics approaches for improved recombinant protein production in Escherichia coli: protein solubility prediction. Brief Bioinform 15: 953-962. Chen Y, Zhang Y, Zhou Y, Luo J & Su Z (2016) Asymmetrical flow field-flow fractionation coupled with multi-angle laser light scattering for stability comparison of virus-like particles in different solution environments. Vaccine 34: 3164-3170. 35 Chiti F & Dobson CM (2017) Protein Misfolding, Amyloid Formation, and Human Disease: A Summary of Progress Over the Last Decade. Annu Rev Biochem 86: 27-68. Churion KA & Bondos SE (2012) Identifying solubility-promoting buffers for intrinsically disordered proteins prior to purification. Methods Mol Biol 896: 415-427. Coquel AS, Jacob JP, Primet M, et al. (2013) Localization of Protein Aggregation in Escherichia coli Is Governed by Diffusion and Nucleoid Macromolecular Crowding Effect. Plos Computational Biology 9. Corchero JL, Gasser B, Resina D, et al. (2013) Unconventional microbial systems for the cost-efficient production of high-quality protein therapeutics. Biotechnol Adv 31: 140- 153. Costa S, Almeida A, Castro A & Domingues L (2014) Fusion tags for protein solubility, purification and immunogenicity in Escherichia coli: the novel Fh8 system. Front Microbiol 5: 63. Curtis-Fisk J, Spencer RM & Weliky DP (2008) Native conformation at specific residues in recombinant inclusion body protein in whole cells determined with solid-state NMR spectroscopy. Journal of The American Chemical Society 130: 12568-+. de Groot NS & Ventura S (2006) Effect of temperature on protein quality in bacterial inclusion bodies. FEBS Lett 580: 6471-6476. de Groot NS, Sabate R & Ventura S (2009) Amyloids in bacterial inclusion bodies. Trends Biochem Sci 34: 408-416. de Groot NS, Castillo V, Grana-Montes R & Ventura S (2012) AGGRESCAN: method, application, and perspectives for drug design. Methods Mol Biol 819: 199-220. de Marco A (2007) Protocol for preparing proteins with improved solubility by co- expressing with molecular chaperones in Escherichia coli. Nat Protoc 2: 2632-2639. de Marco A (2013) Recombinant polypeptide production in E. coli: towards a rational approach to improve the yields of functional proteins. Microb Cell Fact 12: 101. de Marco A, Vigh L, Diamant S & Goloubinoff P (2005) Native folding of aggregation- prone recombinant proteins in Escherichia coli by osmolytes, plasmid- or benzyl alcohol-overexpressed molecular chaperones. Cell Stress Chaperones 10: 329-339. de Marco A, Volrath S, Bruyere T, Law M & Fonne-Pfister R (2000) Recombinant maize protoporphyrinogen IX oxidase expressed in Escherichia coli forms complexes with GroEL and DnaK chaperones. Protein Expr Purif 20: 81-86. 36 Demain AL (2000) Small bugs, big business: the economic power of the microbe. Biotechnol Adv 18: 499-514. Demain AL & Vaishnav P (2009) Production of recombinant proteins by microbes and higher organisms. Biotechnol Adv 27: 297-306. Diaz AA, Tomba E, Lennarson R, Richard R, Bagajewicz MJ & Harrison RG (2010) Prediction of protein solubility in Escherichia coli using logistic regression. Biotechnol Bioeng 105: 374-383. Diez-Gil C, Krabbenborg S, Garcia-Fruitos E, et al. (2010) The nanoscale properties of bacterial inclusion bodies and their effect on mammalian cell proliferation. Biomaterials 31: 5805-5812. Diez-Gil C, Krabbenborg S, Garcia-Fruitos E, et al. (2010) The nanoscale properties of bacterial inclusion bodies and their effect on mammalian cell proliferation. Biomaterials 31: 5805-5812. Dong Q, Yan X, Zheng M & Yang Z (2014) Characterization of an extremely thermostable but cold-adaptive beta-galactosidase from the hyperthermophilic archaeon Pyrococcus furiosus for use as a recombinant aggregation for batch lactose degradation at high temperature. J Biosci Bioeng 117: 706-710. Dragos A, Kovacs AT & Claessen D (2017) The Role of Functional Amyloids in Multicellular Growth and Development of Gram-Positive Bacteria. Biomolecules 7. Du J, Shao Z & Zhao H (2011) Engineering microbial factories for synthesis of value- added products. J Ind Microbiol Biotechnol 38: 873-890. Elia F, Cantini F, Chiti F, Dobson CM & Bemporad F (2017) Direct Conversion of an Enzyme from Native-like to Amyloid-like Aggregates within Inclusion Bodies. Biophys J 112: 2540-2551. Fahnert B, Lilie H & Neubauer P (2004) Inclusion bodies: formation and utilisation. Adv Biochem Eng Biotechnol 89: 93-142. Ferrer-Miralles N & Villaverde A (2013) Bacterial cell factories for recombinant protein production; expanding the catalogue. Microb Cell Fact 12: 113. Ferrer-Miralles N, Domingo-Espin J, Corchero JL, Vazquez E & Villaverde A (2009) Microbial factories for recombinant pharmaceuticals. Microb Cell Fact 8: 17. Fink AL (1998) Protein aggregation: folding aggregates, inclusion bodies and amyloid. Fold Des 3: R9-23. 37 Freydell EJ, Ottens M, Eppink M, van Dedem G & van der Wielen L (2007) Efficient solubilization of inclusion bodies. Biotechnol J 2: 678-684. Futami J, Fujiyama H, Kinoshita R, et al. (2014) Denatured mammalian protein mixtures exhibit unusually high solubility in nucleic acid-free pure water. PLoS One 9: e113295. Garcia-Fruitos E & Villaverde A (2010) Friendly production of bacterial inclusion bodies. Korean Journal of Chemical Engineering 27: 385-389. Garcia-Fruitos E, Aris A & Villaverde A (2007) Localization of functional polypeptides in bacterial inclusion bodies. Appl Environ Microbiol 73: 289-294. Garcia-Fruitos E, Carrio MM, Aris A & Villaverde A (2005) Folding of a misfolding-prone beta-galactosidase in absence of DnaK. Biotechnol Bioeng 90: 869-875. Garcia-Fruitos E, Seras-Franzoso J, Vazquez E & Villaverde A (2010) Tunable geometry of bacterial inclusion bodies as substrate materials for tissue engineering. Nanotechnology 21: 205101. Garcia-Fruitos E, Martinez-Alonso M, Gonzalez-Montalban N, Valli M, Mattanovich D & Villaverde A (2007) Divergent genetic control of protein solubility and conformational quality in Escherichia coli. J Mol Biol 374: 195-205. Garcia-Fruitos E, Gonzalez-Montalban N, Morell M, et al. (2005) Aggregation as bacterial inclusion bodies does not imply inactivation of enzymes and fluorescent proteins. Microb Cell Fact 4: 27. Garcia-Fruitos E, Vazquez E, Diez-Gil C, et al. (2012) Bacterial inclusion bodies: making gold from waste. Trends Biotechnol 30: 65-70. Garcia-Fruitos E, Rodriguez-Carmona E, Diez-Gil C, et al. (2009) Surface Cell Growth Engineering Assisted by a Novel Bacterial Nanomaterial. Advanced Materials 21: 4249- 4253 . Gasser B, Saloheimo M, Rinas U, et al. (2008) Protein folding and conformational stress in microbial cells producing recombinant proteins: a host comparative overview. Microb Cell Fact 7: 11. Gatti-Lafranconi P, Natalello A, Ami D, Doglia SM & Lotti M (2011) Concepts and tools to exploit the potential of bacterial inclusion bodies in protein science and biotechnology. FEBS J 278: 2408-2418. Genest O, Hoskins JR, Camberg JL, Doyle SM & Wickner S (2011) Heat shock protein 90 from Escherichia coli collaborates with the DnaK chaperone system in client protein 38 remodeling. Proceedings of the National Academy of Sciences of the United States of America 108: 8206-8211. Georgiou G & Valax P (1999) Isolating inclusion bodies from bacteria. Methods Enzymol 309: 48-58. Gerson JE, Castillo-Carranza DL & Kayed R (2014) Advances in therapeutics for neurodegenerative tauopathies: moving toward the specific targeting of the most toxic tau species. ACS Chem Neurosci 5: 752-769. Goldenberg D & King J (1982) Trimeric intermediate in the in vivo folding and subunit assembly of the tail spike endorhamnosidase of bacteriophage P22. Proc Natl Acad Sci U S A 79: 3403-3407. Goldenberg DP, Smith DH & King J (1983) Genetic analysis of the folding pathway for the tail spike protein of phage P22. Proc Natl Acad Sci U S A 80: 7060-7064. Goloubinoff P, Mogk A, Zvi AP, Tomoyasu T & Bukau B (1999) Sequential mechanism of solubilization and refolding of stable protein aggregates by a bichaperone network. Proc Natl Acad Sci U S A 96: 13732-13737. Gonzalez-Montalban N, Garcia-Fruitos E & Villaverde A (2007) Recombinant protein solubility - does more mean better? Nat Biotechnol 25: 718-720. Gonzalez-Montalban N, Garcia-Fruitos E, Ventura S, Aris A & Villaverde A (2006) The chaperone DnaK controls the fractioning of functional protein between soluble and insoluble cell fractions in inclusion body-forming cells. Microb Cell Fact 5: 26. Gonzalez-Montalban N, Natalello A, Garcia-Fruitos E, Villaverde A & Doglia SM (2008) In situ protein folding and activation in bacterial inclusion bodies. Biotechnol Bioeng 100: 797-802. Govers SK, Dutre P & Aertsen A (2014) In Vivo Disassembly and Reassembly of Protein Aggregates in Escherichia coli. Journal of Bacteriology 196: 2325-2332. Govers SK, Gayan E & Aertsen A (2017) Intracellular movement of protein aggregates reveals heterogeneous inactivation and resuscitation dynamics in stressed populations of Escherichia coli. Environ Microbiol 19: 511-523. Guo F, Xiong YL, Qin F, Jian H, Huang X & Chen J (2015) Surface properties of heat- induced soluble soy protein aggregates of different molecular masses. J Food Sci 80: C279-287. 39 Gupta A, Lloyd-Price J, Neeli-Venkata R, Oliveira SMD & Ribeiro AS (2014) In Vivo Kinetics of Segregation and Polar Retention of MS2-GFP-RNA Complexes in Escherichia coli. Biophysical Journal 106: 1928-1937. Haase-Pettingell CA & King J (1988 ) Formation of aggregates from a thermolabile in vivo folding intermediate in P22 tailspike maturation. A model for inclusion body formation. J Biol Chem 263: 4977-4983. Hafner Bratkovic I (2017) Prions, prionoid complexes and amyloids: the bad, the good and something in between. Swiss Med Wkly 147: w14424. Hartl FU (2017) Protein Misfolding Diseases. Annu Rev Biochem 86: 21-26. Hartley JL (2006) Cloning technologies for protein expression and purification. Curr Opin Biotechnol 17: 359-366. Hewetson A, Do HQ, Myers C, Muthusubramanian A, Sutton RB, Wylie BJ & Cornwall GA (2017) Functional Amyloids in Reproduction. Biomolecules 7. pii: E46. Hlodan R, Craig S & Pain RH (1991) Protein folding and its implications for the production of recombinant proteins. Biotechnol Genet Eng Rev 9: 47-88. Houry WA (2001) Chaperone-assisted protein folding in the cell cytoplasm. Curr Protein Pept Sci 2: 227-244. Hrabarova E, Achbergerova L & Nahalka J (2015) Insoluble Protein Applications: The Use of Bacterial Inclusion Bodies as Biocatalysts. Insoluble Proteins: Methods and Protocols 1258: 411-422. Huang HL, Charoenkwan P, Kao TF, et al. (2012) Prediction and analysis of protein solubility using a novel scoring card method with dipeptide composition. BMC Bioinformatics 13 Suppl 17: S3. Hulse WL, Gray J & Forbes RT (2013) Evaluating the inter and intra batch variability of protein aggregation behaviour using Taylor dispersion analysis and dynamic light scattering. Int J Pharm 453: 351-357. Iafolla MA, Mazumder M, Sardana V, Velauthapillai T, Pannu K & McMillen DR (2008) Dark proteins: effect of inclusion body formation on quantification of protein expression. Proteins 72: 1233-1242. Ignatova Z, Krishnan B, Bombardier JP, Marcelino AM, Hong J & Gierasch LM (2007) From the test tube to the cell: exploring the folding and aggregation of a beta-clam protein. Biopolymers 88: 157-163. 40 Jackson MP & Hewitt EW (2017) Why are Functional Amyloids Non-Toxic in Humans? Biomolecules 7. Jacob RS, Das S, Ghosh S, et al. (2016) Amyloid formation of growth hormone in presence of zinc: Relevance to its storage in secretory granules. Sci Rep 6: 23370. Jager VD, Lamm R, Kloss R, et al. (2018) A Synthetic Reaction Cascade Implemented by Colocalization of Two Proteins within Catalytically Active Inclusion Bodies. ACS Synth Biol. Jevsevar S, Gaberc-Porekar V, Fonda I, Podobnik B, Grdadolnik J & Menart V (2005) Production of nonclassical inclusion bodies from which correctly folded protein can be extracted. Biotechnol Prog 21: 632-639. Jhamb K & Sahoo DK (2012) Production of soluble recombinant proteins in Escherichia coli: effects of process conditions and chaperone co-expression on cell growth and production of xylanase. Bioresour Technol 123: 135-143. Jong WS, Vikstrom D, Houben D, van den Berg van Saparoea HB, de Gier JW & Luirink J (2017) Application of an E. coli signal sequence as a versatile inclusion body tag. Microb Cell Fact 16: 50. Jurgen B, Breitenstein A, Urlacher V, et al. (2010) Quality control of inclusion bodies in Escherichia coli. Microb Cell Fact 9: 41. Kane JF & Hartley DL (1988) FORMATION OF RECOMBINANT PROTEIN INCLUSION- BODIES IN ESCHERICHIA-COLI. Trends in Biotechnology 6: 95-101. Kaur G, Kaundal S, Kapoor S, Grimes JM, Huiskonen JT & Thakur KG (2018) Mycobacterium tuberculosis CarD, an essential global transcriptional regulator forms amyloid-like fibrils. Sci Rep 8: 10124. Khurana S, Rawi R, Kunji K, Chuang GY, Bensmail H & Mall R (2018) DeepSol: A Deep Learning Framework for Sequence-Based Protein Solubility Prediction. Bioinformatics. King J, Haase-Pettingell C, Robinson AS, Speed M & Mitraki A (1996 Jan) Thermolabile folding intermediates: inclusion body precursors and chaperonin substrates. FASEB J 10: 57-66. Kloss R, Limberg MH, Mackfeld U, et al. (2018) Catalytically active inclusion bodies of L- lysine decarboxylase from E. coli for 1,5-diaminopentane production. Sci Rep 8: 5856. Kolaj O, Spada S, Robin S & Wall JG (2009) Use of folding modulators to improve heterologous protein production in Escherichia coli. Microb Cell Fact 8: 9. 41 Koszagova R, Krajcovic T, Palencarova-Talafova K, et al. (2018) Magnetization of active inclusion bodies: comparison with centrifugation in repetitive biotransformations. Microb Cell Fact 17: 139. Krauss U, Jager VD, Diener M, Pohl M & Jaeger KE (2017) Catalytically-active inclusion bodies-Carrier-free protein immobilizates for application in biotechnology and biomedicine. J Biotechnol 258: 136-147. Kuczynska-Wisnik D, Zurawa-Janicka D, Narkiewicz J, Kwiatkowska J, Lipinska B & Laskowska E (2004) Escherichia coli small heat shock proteins IbpA/B enhance activity of enzymes sequestered in inclusion bodies. Acta Biochim Pol 51: 925-931. Kumar S, Thangakani AM, Nagarajan R, Singh SK, Velmurugan D & Gromiha MM (2016) Autoimmune Responses to Soluble Aggregates of Amyloidogenic Proteins Involved in Neurodegenerative Diseases: Overlapping Aggregation Prone and Autoimmunogenic regions. Sci Rep 6: 22258. Lebendiker M & Danieli T (2014) Production of prone-to-aggregate proteins. FEBS Lett 588: 236-246. Lebendiker M, Maes M & Friedler A (2015) A screening methodology for purifying proteins with aggregation problems. Methods Mol Biol 1258: 261-281. Lee SY, Mattanovich D & Villaverde A (2012) Systems metabolic engineering, industrial biotechnology and microbial cell factories. Microb Cell Fact 11: 156. Li Y, Weiss WFt & Roberts CJ (2009) Characterization of high-molecular-weight nonnative aggregates and aggregation kinetics by size exclusion chromatography with inline multi-angle laser light scattering. J Pharm Sci 98: 3997-4016. Lindner AB, Madden R, Dernarez A, Stewart EJ & Taddei F (2008) Asymmetric segregation of protein aggregates is associated with cellular aging and rejuvenation. Proceedings of the National Academy of Sciences of the United States of America 105: 3076-3081. Liovic M, Ozir M, Zavec AB, Peternel S, Komel R & Zupancic T (2012) Inclusion bodies as potential vehicles for recombinant protein delivery into epithelial cells. Microbial Cell Factories 11. Liu J, Blasie CA, Shi S, Joshi SB, Middaugh CR & Volkin DB (2013) Characterization and stabilization of recombinant human protein pentraxin (rhPTX-2). J Pharm Sci 102: 827- 841. 42 Loo Y, Goktas M, Tekinay AB, Guler MO, Hauser CA & Mitraki A (2015) Self-Assembled Proteins and Peptides as Scaffolds for Tissue Regeneration. Adv Healthc Mater 4: 2557- 2586. Lua LH, Connors NK, Sainsbury F, Chuan YP, Wibowo N & Middelberg AP (2014) Bioengineering virus-like particles as vaccines. Biotechnol Bioeng 111: 425-440. Luo J, Leeman M, Ballagi A, Elfwing A, Su Z, Janson JC & Wahlund KG (2006) Size characterization of green fluorescent protein inclusion bodies in E. coli using asymmetrical flow field-flow fractionation-multi-angle light scattering. J Chromatogr A 1120: 158-164. Magnan CN, Randall A & Baldi P (2009) SOLpro: accurate sequence-based prediction of protein solubility. Bioinformatics 25: 2200-2207. Maji SK, Perrin MH, Sawaya MR, et al. (2009) Functional amyloids as natural storage of peptide hormones in pituitary secretory granules. Science 325: 328-332. Malik S, Afzal I, Mehmood MA, Rahimuddin SA, Gull M & Nahid N (2018) Heterologous synthesis and recovery of advanced biofuels from bacterial cell factories. Protein Pept Lett. Marcoleta A, Marin M, Mercado G, Valpuesta JM, Monasterio O & Lagos R (2013) Microcin e492 amyloid formation is retarded by posttranslational modification. J Bacteriol 195: 3995-4004. Margreiter G, Messner P, Caldwell KD & Bayer K (2008) Size characterization of inclusion bodies by sedimentation field-flow fractionation. J Biotechnol 138: 67-73. Marston FAO (1986) THE PURIFICATION OF EUKARYOTIC POLYPEPTIDES SYNTHESIZED IN ESCHERICHIA-COLI. Biochemical Journal 240: 1-12. Martinez-Alonso M, Garcia-Fruitos E & Villaverde A (2008) Yield, solubility and conformational quality of soluble proteins are not simultaneously favored in recombinant Escherichia coli. Biotechnol Bioeng 101: 1353-1358. Martinez-Alonso M, Gonzalez-Montalban N, Garcia-Fruitos E & Villaverde A (2008) The Functional quality of soluble recombinant polypeptides produced in Escherichia coli is defined by a wide conformational spectrum. Appl Environ Microbiol 74: 7431-7433. Martinez-Alonso M, Gonzalez-Montalban N, Garcia-Fruitos E & Villaverde A (2009) Learning about protein solubility from bacterial inclusion bodies. Microb Cell Fact 8: 4. 43 Matuszewska M, Kuczynska-Wisnik D, Laskowska E & Liberek K (2005) The small heat shock protein IbpA of Escherichia coli cooperates with IbpB in stabilization of thermally aggregated proteins in a disaggregation competent state. Journal of Biological Chemistry 280: 12292-12298. Mitraki A (2010) PROTEIN AGGREGATION: FROM INCLUSION BODIES TO AMYLOID AND BIOMATERIALS. Advances in Protein Chemistry and Structural Biology, Vol 79, (McPherson A, ed.eds.), p.pp. 89-125. Mitraki A & King J (1989) PROTEIN FOLDING INTERMEDIATES AND INCLUSION BODY FORMATION. Bio-Technology 7: 690-697. Mitraki A, Fane B, Haase-Pettingell C, Sturtevant J & King J (1991 ) Global suppression of protein folding defects and inclusion body formation. Science 253: 54-58. Mogk A & Bukau B (2004) Molecular chaperones: structure of a protein disaggregase. Curr Biol 14: R78-80. Mogk A, Bukau B & Kampinga HH (2018) Cellular Handling of Protein Aggregates by Disaggregation Machines. Mol Cell 69: 214-226. Mogk A, Deuerling E, Vorderwulbecke S, Vierling E & Bukau B (2003) Small heat shock proteins, ClpB and the DnaK system form a functional triade in reversing protein aggregation. Mol Microbiol 50: 585-595. Mogk A, Schlieker C, Friedrich KL, Schonfeld HJ, Vierling E & Bukau B (2003) Refolding of substrates bound to small Hsps relies on a disaggregation reaction mediated most efficiently by ClpB/DnaK. J Biol Chem 278: 31033-31042. Mogk A, Schlieker C, Strub C, Rist W, Weibezahn J & Bukau B (2003) Roles of individual domains and conserved motifs of the AAA+ chaperone ClpB in oligomerization, ATP hydrolysis, and chaperone activity. J Biol Chem 278: 17615-17624. Molina-Garcia L, Gasset-Rosa F, Moreno-Del Alamo M, Fernandez-Tresguerres ME, Moreno-Diaz de la Espina S, Lurz R & Giraldo R (2016) Functional amyloids as inhibitors of plasmid DNA replication. Sci Rep 6: 25425. Molina-Garcia L, Moreno-Del Alamo M, Botias P, et al. (2017) Outlining Core Pathways of Amyloid Toxicity in Bacteria with the RepA-WH1 Prionoid. Front Microbiol 8: 539. Morell M, Bravo R, Espargaro A, Sisquella X, Aviles FX, Fernandez-Busquets X & Ventura S (2008) Inclusion bodies: Specificity in their aggregation process and amyloid- like structure. Biochimica Et Biophysica Acta-Molecular Cell Research 1783: 1815-1825. 44 Morell M, Bravo R, Espargaro A, Sisquella X, Aviles FX, Fernandez-Busquets X & Ventura S (2008) Inclusion bodies: specificity in their aggregation process and amyloid- like structure. Biochim Biophys Acta 1783: 1815-1825. Murby M, Samuelsson E, Nguyen TN, et al. (1995) Hydrophobicity engineering to increase solubility and stability of a recombinant protein from respiratory syncytial virus. Eur J Biochem 230: 38-44. Nahalka J & Nidetzky B (2007) Fusion to a pull-down domain: A novel approach of producing Trigonopsis variabilis D-amino acid oxidase as insoluble enzyme aggregates. Biotechnology and Bioengineering 97: 454-461. Nahalka J, Vikartovska A & Hrabarova E (2008) A crosslinked inclusion body process for sialic acid synthesis. Journal of Biotechnology 134: 146-153. Nascimento IP & Leite LC (2012) Recombinant vaccines and the development of new vaccine strategies. Braz J Med Biol Res 45: 1102-1111. Natalello A, Santarella R, Doglia SM & de Marco A (2008) Physical and chemical perturbations induce the formation of protein aggregates with different structural features. Protein Expr Purif 58: 356-361. Natalello A, Liu J, Ami D, Doglia SM & de Marco A (2009) The osmolyte betaine promotes protein misfolding and disruption of protein aggregates. Proteins 75: 509- 517. Neeli-Venkata R, Martikainen A, Gupta A, Goncalves N, Fonseca J & Ribeiro AS (2016) Robustness of the Process of Nucleoid Exclusion of Protein Aggregates in Escherichia coli. Journal of Bacteriology 198: 898-906. Neubauer PF, B; Lilie, H; Villaverde, A; (2006) Protein inclusion bodies in recombinant bacteria. Springer. Nomine Y, Ristriani T, Laurent C, Lefevre JF, Weiss E & Trave G (2001) A strategy for optimizing the monodispersity of fusion proteins: application to purification of recombinant HPV E6 oncoprotein. Protein Eng 14: 297-305. Nomine Y, Ristriani T, Laurent C, Lefevre JF, Weiss E & Trave G (2001) Formation of soluble inclusion bodies by hpv e6 oncoprotein fused to maltose-binding protein. Protein Expr Purif 23: 22-32. Novak SM, U; Peternel, S; Venturini, P; Bele, M; Gaberšček; M (2009) Electrophoretic deposition as a tool for separation of protein inclusion bodies from host bacteria in 45 suspension. Colloids and Surfaces A: Physicochemical and Engineering Aspects 340: 155-160. Oberg K, Chrunyk BA, Wetzel R & Fink AL (1994) NATIVE-LIKE SECONDARY STRUCTURE IN INTERLEUKIN-1-BETA INCLUSION-BODIES BY ATTENUATED TOTAL REFLECTANCE FTIR. Biochemistry 33: 2628-2634. Oliveira SMD, Neeli-Venkata R, Goncalves NSM, et al. (2016) Increased cytoplasm viscosity hampers aggregate polar segregation in Escherichia coli. Molecular Microbiology 99: 686-699. Overton TW (2014) Recombinant protein production in bacterial hosts. Drug Discov Today 19: 590-601. Paladin L, Piovesan D & Tosatto SCE (2017) SODA: prediction of protein solubility from disorder and aggregation propensity. Nucleic Acids Res 45: W236-W240. Pallares I, Iglesias V & Ventura S (2015) The Rho Termination Factor of Clostridium botulinum Contains a Prion-Like Domain with a Highly Amyloidogenic Core. Front Microbiol 6: 1516. Parret AH, Besir H & Meijers R (2016) Critical reflections on synthetic gene design for recombinant protein expression. Curr Opin Struct Biol 38: 155-162. Parry BR, Surovtsev IV, Cabeen MT, O'Hem CS, Dufresne ER & Jacobs-Wagner C (2014) The Bacterial Cytoplasm Has Glass-like Properties and Is Fluidized by Metabolic Activity. Cell 156: 183-194. Pattenden LK, Middelberg AP, Niebert M & Lipin DI (2005) Towards the preparative and large-scale precision manufacture of virus-like particles. Trends Biotechnol 23: 523- 529. Paul AJ, Schwab K, Prokoph N, Haas E, Handrick R & Hesse F (2015) Fluorescence dye- based detection of mAb aggregates in CHO culture supernatants. Anal Bioanal Chem 407: 4849-4856. Pavisic R, Dodig I, Horvatic A, et al. (2010) Differences between reversible (self- association) and irreversible aggregation of rHuG-CSF in carbohydrate and polyol formulations. Eur J Pharm Biopharm 76: 357-365. Pesarrodona M, Fernandez Y, Foradada L, et al. (2016) Conformational and functional variants of CD44-targeted protein nanoparticles bio-produced in bacteria. Biofabrication 8: 025001. 46 Peternel S & Komel R (2010) Isolation of biologically active nanomaterial (inclusion bodies) from bacterial cells. Microb Cell Fact 9: 66. Peternel S, Grdadolnik J, Gaberc-Porekar V & Komel R (2008) Engineering inclusion bodies for non denaturing extraction of functional proteins. Microbial Cell Factories 7. Peternel SB, M; Gaberc-Porekar, V;Komel, R (2008) Inclusion Bodies Contraction with Implications in Biotechnology. Acta Chim. Slov. 55: 608-612. Piscitelli A, Cicatiello P, Gravagnuolo AM, Sorrentino I, Pezzella C & Giardina P (2017) Applications of Functional Amyloids from Fungi: Surface Modification by Class I Hydrophobins. Biomolecules 7. Przybycien TM, Dunn JP, Valax P & Georgiou G (1994) SECONDARY STRUCTURE CHARACTERIZATION OF BETA-LACTAMASE INCLUSION-BODIES. Protein Engineering 7: 131-136. Quaas BB, L; Li, Z; Nimtz, M; Hoffmann A; Rinas, U (2018) Properties of dimeric, disulfide-linked rhBMP-2 recovered from E. coli derived inclusion bodies by mild extraction or chaotropic solubilisation and subsequent refolding. Process Biochem 2018;67:80-7. Process Biochem 67: 87-87. Rabhi-Essafi I, Sadok A, Khalaf N & Fathallah DM (2007) A strategy for high-level expression of soluble and functional human interferon alpha as a GST-fusion protein in E. coli. Protein Eng Des Sel 20: 201-209. Raghunathan G, Munussami G, Moon H, et al. (2014) A variant of green fluorescent protein exclusively deposited to active intracellular inclusion bodies. Microb Cell Fact 13: 68. Ratajczak E, Zietkiewicz S & Liberek K (2009) Distinct Activities of Escherichia coli Small Heat Shock Proteins IbpA and IbpB Promote Efficient Protein Disaggregation. Journal of Molecular Biology 386: 178-189. Rawi R, Mall R, Kunji K, Shen CH, Kwong PD & Chuang GY (2018) PaRSnIP: sequence- based protein solubility prediction using gradient boosting machine. Bioinformatics 34: 1092-1098. Rinas U (1996) Synthesis rates of cellular proteins involved in translation and protein folding are strongly altered in response to overproduction of basic fibroblast growth factor by recombinant Escherichia coli. Biotechnol Prog 12: 196-200. 47 Rinas U & Bailey JE (1992) Protein compositional analysis of inclusion bodies produced in recombinant Escherichia coli. Appl Microbiol Biotechnol 37: 609-614. Rinas U, Hoffmann F, Betiku E, Estape D & Marten S (2007) Inclusion body anatomy and functioning of chaperone-mediated in vivo inclusion body disassembly during high- level recombinant protein production in Escherichia coli. J Biotechnol 127: 244-257. Rinas U, Garcia-Fruitos E, Corchero JL, Vazquez E, Seras-Franzoso J & Villaverde A (2017) Bacterial Inclusion Bodies: Discovering Their Better Half. Trends Biochem Sci 42: 726-737. Rinas U, Garcia-Fruitos E, Corchero JL, Vazquez E, Seras-Franzoso J & Villaverde A (2017) Bacterial Inclusion Bodies: Discovering Their Better Half. Trends in Biochemical Sciences 42: 726-737. Roche DB, Villain E & Kajava AV (2017) Usage of a dataset of NMR resolved protein structures to test aggregation versus solubility prediction algorithms. Protein Sci 26: 1864-1869. Rodriguez-Carmona E & Villaverde A (2010) Nanostructured bacterial materials for innovative medicines. Trends Microbiol 18: 423-430. Rodriguez-Carmona E, Villaverde A & Garcia-Fruitos E (2011) How to break recombinant bacteria: does it matter? Bioeng Bugs 2: 222-225. Rodriguez-Carmona E, Cano-Garrido O, Seras-Franzoso J, Villaverde A & Garcia-Fruitos E (2010) Isolation of cell-free bacterial inclusion bodies. Microb Cell Fact 9: 71. Rokney A, Shagan M, Kessel M, Smith Y, Rosenshine I & Oppenheim AB (2009) E. coli Transports Aggregated Proteins to the Poles by a Specific and Energy-Dependent Process. Journal of Molecular Biology 392: 589-601. Romero D & Kolter R (2014) Functional amyloids in bacteria. Int Microbiol 17: 65-73. Rosano GL & Ceccarelli EA (2014) Recombinant protein expression in Escherichia coli: advances and challenges. Front Microbiol 5: 172. Rueda F, Cano-Garrido O, Mamat U, Wilke K, Seras-Franzoso J, Garcia-Fruitos E & Villaverde A (2014) Production of functional inclusion bodies in endotoxin-free Escherichia coli. Appl Microbiol Biotechnol 98: 9229-9238. Rueda F, Gasser B, Sanchez-Chardi A, et al. (2016) Functional inclusion bodies produced in the yeast Pichia pastoris. Microb Cell Fact 15: 166. 48 Rueda F, Cespedes MV, Conchillo-Sole O, et al. (2015) Bottom-Up Instructive Quality Control in the Biofabrication of Smart Protein Materials. Adv Mater 27: 7816-7822. Sanchez-Garcia L, Martin L, Mangues R, Ferrer-Miralles N, Vazquez E & Villaverde A (2016) Recombinant pharmaceuticals from microbial cells: a 2015 update. Microb Cell Fact 15: 33. Schein C & Noteborn M (1988) Formation of Soluble Recombinant Proteins in Escherichia Coli is Favored by Lower Growth Temperature. Bio/Technology 6: 4. Schein CH (1989) PRODUCTION OF SOLUBLE RECOMBINANT PROTEINS IN BACTERIA. Bio-Technology 7: 1141-1147. Schmidt FR (2004) Recombinant expression systems in the pharmaceutical industry. Appl Microbiol Biotechnol 65: 363-372. Schrodel A & de Marco A (2005) Characterization of the aggregates formed during recombinant protein expression in bacteria. BMC Biochem 6: 10. Schultz T, Martinez L & de Marco A (2006) The evaluation of the factors that cause aggregation during recombinant expression in E. coli is simplified by the employment of an aggregation-sensitive reporter. Microb Cell Fact 5: 28. Schultz T, Liu J, Capasso P & de Marco A (2007) The solubility of recombinant proteins expressed in Escherichia coli is increased by otsA and otsB co-transformation. Biochem Biophys Res Commun 355: 234-239. Schwartz K, Syed AK, Stephenson RE, Rickard AH & Boles BR (2012) Functional amyloids composed of phenol soluble modulins stabilize Staphylococcus aureus biofilms. PLoS Pathog 8: e1002744. Schwartz K, Ganesan M, Payne DE, Solomon MJ & Boles BR (2016) Extracellular DNA facilitates the formation of functional amyloids in Staphylococcus aureus biofilms. Mol Microbiol 99: 123-134. Seras-Franzoso J, Peternel S, Cano-Garrido O, Villaverde A & Garcia-Fruitos E (2015) Bacterial inclusion body purification. Methods Mol Biol 1258: 293-305. Seras-Franzoso J, Sanchez-Chardi A, Garcia-Fruitos E, Vazquez E & Villaverde A (2016) Cellular uptake and intracellular fate of protein releasing bacterial amyloids in mammalian cells. Soft Matter 12: 3451-3460. 49 Seras-Franzoso J, Peebo K, Garcia-Fruitos E, Vazquez E, Rinas U & Villaverde A (2014) Improving protein delivery of fibroblast growth factor-2 from bacterial inclusion bodies used as cell culture substrates. Acta Biomater 10: 1354-1359. Seras-Franzoso J, Tatkiewicz WI, Vazquez E, Garcia-Fruitos E, Ratera I, Veciana J & Villaverde A (2015) Integrating mechanical and biological control of cell proliferation through bioinspired multieffector materials. Nanomedicine (Lond) 10: 873-891. Seras-Franzoso J, Diez-Gil C, Vazquez E, et al. (2012) Bioadhesiveness and efficient mechanotransduction stimuli synergistically provided by bacterial inclusion bodies as scaffolds for tissue engineering. Nanomedicine (Lond) 7: 79-93. Seras-Franzoso J, Tsimbouri PM, Burgess KV, et al. (2014) Topographically targeted osteogenesis of mesenchymal stem cells stimulated by inclusion bodies attached to polycaprolactone surfaces. Nanomedicine (Lond) 9: 207-220. Seras-Franzoso J, Peebo K, Luis Corchero J, et al. (2013) A nanostructured bacterial bioscaffold for the sustained bottom-up delivery of protein drugs. Nanomedicine (Lond) 8: 1587-1599. Seras-Franzoso J, Steurer C, Roldan M, et al. (2013) Functionalization of 3D scaffolds with protein-releasing biomaterials for intracellular delivery. J Control Release 171: 63- 72. Seviour T, Hansen SH, Yang L, et al. (2015) Functional amyloids keep quorum-sensing molecules in check. J Biol Chem 290: 6457-6469. Shin S & Cherry S (2017) A RHIM with a View: FLYing with Functional Amyloids. Immunity 47: 604-606. Simpson RJ (2010) Solubilization of Escherichia coli recombinant proteins from inclusion bodies. Cold Spring Harb Protoc 2010: pdb prot5485. Singer MA & Lindquist S (1998) Multiple effects of trehalose on protein folding in vitro and in vivo. Mol Cell 1: 639-648. Singh A, Upadhyay V & Panda AK (2015) Solubilization and refolding of inclusion body proteins. Methods Mol Biol 1258: 283-291. Singh A, Upadhyay V, Upadhyay AK, Singh SM & Panda AK (2015) Protein recovery from inclusion bodies of Escherichia coli using mild solubilization process. Microb Cell Fact 14: 41. 50 Singh SM & Panda AK (2005) Solubilization and refolding of bacterial inclusion body proteins. Journal of Bioscience and Bioengineering 99: 303-310. Singh SM, Sharma A, Upadhyay AK, Singh A, Garg LC & Panda AK (2012) Solubilization of inclusion body proteins using n-propanol and its refolding into bioactive form. Protein Expr Purif 81: 75-82. Smialowski P, Doose G, Torkler P, Kaufmann S & Frishman D (2012) PROSO II--a new method for protein solubility prediction. FEBS J 279: 2192-2200. Smialowski P, Martin-Galiano AJ, Mikolajka A, Girschick T, Holak TA & Frishman D (2007) Protein solubility: sequence based prediction and experimental verification. Bioinformatics 23: 2536-2542. Smirnova E, Safenkova I, Stein-Margolina V, Shubin V, Polshakov V & Gurvits B (2015) pH-responsive modulation of insulin aggregation and structural transformation of the aggregates. Biochimie 109: 49-59. Song J (2017) Transforming Cytosolic Proteins into "Insoluble" and Membrane-toxic Forms Triggering Diseases/Aging by Genetic, Pathological or Environmental Factors. Protein Pept Lett 24: 294-306. Sorensen HP & Mortensen KK (2005) Advanced genetic strategies for recombinant protein expression in Escherichia coli. J Biotechnol 115: 113-128. Sorensen HP & Mortensen KK (2005) Soluble expression of recombinant proteins in the cytoplasm of Escherichia coli. Microb Cell Fact 4: 1. Souillac PO (2005) Biophysical characterization of insoluble aggregates of a multi- domain protein: an insight into the role of the various domains. J Pharm Sci 94: 2069- 2083. Speed MA, Wang DIC & King J (1996) Specific aggregation of partially folded polypeptide chains: The molecular basis of inclusion body composition. Nature Biotechnology 14: 1283-1287. St John RJ, Carpenter JF & Randolph TW (1999) High pressure fosters protein refolding from aggregates at high concentrations. Proc Natl Acad Sci U S A 96: 13029-13033. Stamm A, Strauss S, Vogt P, Scheper T & Pepelanova I (2018) Positive in vitro wound healing effects of functional inclusion bodies of a lipoxygenase from the Mexican axolotl. Microb Cell Fact 17: 57. 51 Stampolidis P, Kaderbhai NN & Kaderbhai MA (2009) Periplasmically-exported lupanine hydroxylase undergoes transition from soluble to functional inclusion bodies in Escherichia coli. Arch Biochem Biophys 484: 8-15. Stegemann J, Ventzki R, Schrodel A & de Marco A (2005) Comparative analysis of protein aggregates by blue native electrophoresis and subsequent sodium dodecyl sulfate-polyacrylamide gel electrophoresis in a three-dimensional geometry gel. Proteomics 5: 2002-2009. Stenvang M, Dueholm MS, Vad BS, et al. (2016) Epigallocatechin Gallate Remodels Overexpressed Functional Amyloids in Pseudomonas aeruginosa and Increases Biofilm Susceptibility to Antibiotic Treatment. J Biol Chem 291: 26540-26553. Strandberg L & Enfors SO (1991) Factors Influencing Inclusion Body Formation in the Production of a Fused Protein in Escherichia-Coli. Applied and Environmental Microbiology 57: 1669-1674. Syed AK & Boles BR (2014) Fold modulating function: bacterial toxins to functional amyloids. Front Microbiol 5: 401. Talafova K, Hrabarova E, Chorvat D & Nahalka J (2013) Bacterial inclusion bodies as potential synthetic devices for pathogen recognition and a therapeutic substance release. Microbial Cell Factories 12. Tatkiewicz WI, Seras-Franzoso J, Garcia-Fruitos E, et al. (2013) Two-dimensional microscale engineering of protein-based nanoparticles for cell guidance. ACS Nano 7: 4774-4784. Tatkiewicz WI, Seras-Franzoso J, Garcia-Fruitos E, et al. (2018) Surface-Bound Gradient Deposition of Protein Nanoparticles for Cell Motility Studies. ACS Appl Mater Interfaces 10: 25779-25786. Taylor G, Hoare M, Gray DR & Marston FAO (1986) SIZE AND DENSITY OF PROTEIN INCLUSION-BODIES. Bio-Technology 4: 553-557. Tjong H & Zhou HX (2008) Prediction of protein solubility from calculation of transfer free energy. Biophys J 95: 2601-2609. Tomoyasu T, Mogk A, Langen H, Goloubinoff P & Bukau B (2001) Genetic dissection of the roles of chaperones and proteases in protein folding and degradation in the Escherichia coli cytosol. Mol Microbiol 40: 397-413. 52 Torrealba D, Seras-Franzoso J, Mamat U, Wilke K, Villaverde A, Roher N & Garcia- Fruitos E (2016) Complex Particulate Biomaterials as Immunostimulant-Delivery Platforms. PLoS One 11: e0164073. Torrealba D, Parra D, Seras-Franzoso J, et al. (2016) Nanostructured recombinant cytokines: A highly stable alternative to short-lived prophylactics. Biomaterials 107: 102-114. Turchetto J, Sequeira AF, Ramond L, et al. (2017) High-throughput expression of animal venom toxins in Escherichia coli to generate a large library of oxidized disulphide- reticulated peptides for drug discovery. Microb Cell Fact 16: 6. Tyedmers J, Mogk A & Bukau B (2010) Cellular strategies for controlling protein aggregation. Nat Rev Mol Cell Biol 11: 777-788. Unzueta U, Cespedes MV, Ferrer-Miralles N, et al. (2012) Intracellular CXCR4(+) cell targeting with T22-empowered protein-only nanoparticles. Int J Nanomedicine 7: 4533- 4544. Unzueta U, Seras-Franzoso J, Cespedes MV, et al. (2017) Engineering tumor cell targeting in nanoscale amyloidal materials. Nanotechnology 28: 015102. Unzueta U, Cespedes MV, Sala R, et al. (2018) Release of targeted protein nanoparticles from functional bacterial amyloids: A death star-like approach. J Control Release. 279: 29-39. Upadhyay AK, Murmu A, Singh A & Panda AK (2012) Kinetics of Inclusion Body Formation and Its Correlation with the Characteristics of Protein Aggregates in Escherichia coli. Plos One 7. e33951. Upadhyay AK, Singh A, Mukherjee KJ & Panda AK (2014) Refolding and purification of recombinant L-asparaginase from inclusion bodies of E. coli into active tetrameric protein. Front Microbiol 5: 486. Valax P & Georgiou G (1993) Molecular characterization of beta-lactamase inclusion bodies produced in Escherichia coli. Composition. Biotechnol Prog 9: 539-547. Valdez-Cruz NA, Reynoso-Cereceda GI, Perez-Rodriguez S, et al. (2017) Production of a recombinant phospholipase A2 in Escherichia coli using resonant acoustic mixing that improves oxygen transfer in shake flasks. Microb Cell Fact 16: 129. Vallejo LF & Rinas U (2004) Strategies for the recovery of active proteins through refolding of bacterial inclusion body proteins. Microb Cell Fact 3: 11. 53 Van der Henst C, Charlier C, Deghelt M, Wouters J, Matroule JY, Letesson JJ & De Bolle X (2010) Overproduced Brucella abortus PdhS-mCherry forms soluble aggregates in Escherichia coli, partially associating with mobile foci of IbpA-YFP. BMC Microbiol 10: 248. Van Gerven N, Van der Verren SE, Reiter DM & Remaut H (2018) The Role of Functional Amyloids in Bacterial Virulence. J Mol Biol. Vazquez E, Corchero JL, Burgueno JF, et al. (2012) Functional inclusion bodies produced in bacteria as naturally occurring nanopills for advanced cell therapies. Adv Mater 24: 1742-1747. Veinger L, Diamant S, Buchner J & Goloubinoff P (1998) The small heat-shock protein IbpB from Escherichia coli stabilizes stress-denatured proteins for subsequent refolding by a multichaperone network. J Biol Chem 273: 11032-11037. Vellard M (2003) The enzyme as drug: application of enzymes as pharmaceuticals. Curr Opin Biotechnol 14: 444-450. Ventura S & Villaverde A (2006) Protein quality in bacterial inclusion bodies. Trends Biotechnol 24: 179-185. Vera A, Gonzalez-Montalban N, Aris A & Villaverde A (2007) The conformational quality of insoluble recombinant proteins is enhanced at low growth temperatures. Biotechnol Bioeng 96: 1101-1106. Villaverde A (2012) Bacterial inclusion bodies: an emerging platform for drug delivery and cell therapy. Nanomedicine (Lond) 7: 1277-1279. Villaverde A & Carrio MM (2003) Protein aggregation in recombinant bacteria: biological role of inclusion bodies. Biotechnol Lett 25: 1385-1395. Villaverde A, Benito A, Viaplana E & Cubarsi R (1993) Fine regulation of cI857- controlled gene expression in continuous culture of recombinant Escherichia coli by temperature. Appl Environ Microbiol 59: 3485-3487. Villaverde A, Corchero JL, Seras-Franzoso J & Garcia-Fruitos E (2015) Functional protein aggregates: just the tip of the iceberg. Nanomedicine (Lond) 10: 2881-2891. Villaverde A, Garcia-Fruitos E, Rinas U, Seras-Franzoso J, Kosoy A, Corchero JL & Vazquez E (2012) Packaging protein drugs as bacterial inclusion bodies for therapeutic applications. Microb Cell Fact 11: 76. 54 Vogel JH, Nguyen H, Pritschet M, Van Wegen R & Konstantinov K (2002) Continuous annular chromatography: General characterization and application for the isolation of recombinant protein drugs. Biotechnol Bioeng 80: 559-568. Wang L, Maji SK, Sawaya MR, Eisenberg D & Riek R (2008) Bacterial inclusion bodies contain amyloid-like structure. Plos Biology 6: 1791-1801. Wang X, Zhou BH, Hu WK, Zhao Q & Lin ZL (2015) Formation of active inclusion bodies induced by hydrophobic self-assembling peptide GFIL8. Microbial Cell Factories 14. Wasmer C, Benkemoun L, Sabate R, et al. (2009) Solid-State NMR Spectroscopy Reveals that E. coli Inclusion Bodies of HET-s(218-289) are Amyloids. Angewandte Chemie- International Edition 48: 4858-4860. Wear MP, Kryndushkin D, O'Meally R, Sonnenberg JL, Cole RN & Shewmaker FP (2015) Proteins with Intrinsically Disordered Domains Are Preferentially Recruited to Polyglutamine Aggregates. PLoS One 10: e0136362. Weibezahn J, Bukau B & Mogk A (2004) Unscrambling an egg: protein disaggregation by AAA+ proteins. Microb Cell Fact 3: 1. Wetzel R (1994 ) Mutations and off-pathway aggregation of proteins. Trends Biotechnol 12: 193-198. Wetzel R (1996) For protein misassembly, it's the ''I'' decade. Cell 86: 699-702. Winkler J, Seybert A, Konig L, et al. (2010) Quantitative and spatio-temporal features of protein aggregation in Escherichia coli and consequences on protein quality control and cellular ageing. Embo Journal 29: 910-923. Wurm DJ, Quehenberger J, Mildner J, et al. (2018) Teaching an old pET new tricks: tuning of inclusion body formation and properties by a mixed feed system in E. coli. Appl Microbiol Biotechnol 102: 667-676. Yamaguchi H & Miyazaki M (2014) Refolding techniques for recovering biologically active recombinant proteins from inclusion bodies. Biomolecules 4: 235-251. Yang Y, Niroula A, Shen B & Vihinen M (2016) PON-Sol: prediction of effects of amino acid substitutions on protein solubility. Bioinformatics 32: 2032-2034. Yu MH & King J (1984) SINGLE AMINO-ACID SUBSTITUTIONS INFLUENCING THE FOLDING PATHWAY OF THE PHAGE-P22 TAIL SPIKE ENDORHAMNOSIDASE. Proceedings of the National Academy of Sciences of the United States of America-Biological Sciences 81: 6584-6588. 55 Yu MH & King J (1988) SURFACE AMINO-ACIDS AS SITES OF TEMPERATURE-SENSITIVE FOLDING MUTATIONS IN THE P22 TAILSPIKE PROTEIN. Journal of Biological Chemistry 263: 1424-1431. Yuan AH & Hochschild A (2017) A bacterial global regulator forms a prion. Science 355: 198-201. Zblewska K, Krajewska J, Zolkiewski M & Kedzierska-Mieszkowska S (2014) Role of the disaggregase ClpB in processing of proteins aggregated as inclusion bodies. Arch Biochem Biophys 555-556: 23-27. Zhou BH, Xing L, Wu W, Zhang XE & Lin ZL (2012) Small surfactant-like peptides can drive soluble proteins into active aggregates. Microbial Cell Factories 11. Zhou W, Bi J, Janson JC, Li Y, Huang Y, Zhang Y & Su Z (2006) Molecular characterization of recombinant Hepatitis B surface antigen from Chinese hamster ovary and Hansenula polymorpha cells by high-performance size exclusion chromatography and multi-angle laser light scattering. J Chromatogr B Analyt Technol Biomed Life Sci 838: 71-77. 56 Figure 1. Properties of bacterial IBs. A. Visualization of IB-producing E. coli cells by optical microscopy. IBs are refractile intracellular particles. B. Transmission electron microscopy images of IB-producing Lactococcus lactis cells. IBs are observed as electrodense materials in the cell cytoplasm. C. Phase contrast and fluorescent microscopy of the same field of a BFP-producing E. coli cell culture. Blue fluorescence is apparent. D. Classical model (up) in which a recombinant bacterial cell can be divided into two virtual fractions, namely the soluble cell fraction containing properly folded, soluble and functional protein (green), and the insoluble fraction containing aggregated, misfolded or unfolded non-functional protein (red). Experimental data have accumulated demonstrating that both soluble and insoluble cell fractions are formed by functional (green) and non-functional (red) protein versions (bottom). Culture and genetic conditions that improve protein folding improve the conformation quality of both soluble and insoluble protein species, while those impairing folding also impact on both cell fractions. Images are modified versions from (Carrio & Villaverde, 2001) (A), (Cano-Garrido, et al., 2016) (B), (Garcia-Fruitos, et al., 2005) (C) and (Gonzalez-Montalban, et al., 2007) (D). Images have been reproduced with permission of Elsevier and Springer Nature. 57 Figure 2. Protein aggregation in recombinant bacteria. A. Characteristics of soluble micelles formed by fusion proteins. Fusions between a target protein (grey) with propensity to aggregate and a passenger polypeptide with high solubility (green) are expressed to increase the chance of the target protein to remain in suspension and complete its folding. Usually a linker (yellow) is exploited to connect the sequences to avoid steric hindrance between the two moieties during folding but it has been described as a sensitive portion for proteolytic degradation. Whether it is not the case and the aggregation-prone protein does not succeed to reach a stable folding, it will interact non-specifically with similar misfolded proteins. They will form a compact hydrophobic core surrounded by the highly soluble passenger molecules. The micelle- like structures will therefore remain soluble in polar solutions. If the passenger proteins have dimerization capacity –as for instance the common passenger protein GST- they will promote the merge of single micelles to form larger aggregates which will still be able to float, at least until a critical point. B. Protein misfolding, progressive aggregation, precipitation and re-solubilization. Proteins might need the support of osmolytes and molecular chaperones (DnaK/DnaJ/GrpE + GroESL) to reach their native folding, otherwise remain trapped into instable folding intermediates which can aggregate. Limiting chaperone capacity as well as stress conditions which induce protein misfolding can lead to aggregation mediated by hydrophobic patches present on the molecule surface and that tend to combine with other similar regions of the same or other proteins. The consequent instability leads to form progressively larger aggregates that can finally precipitate into IBs possessing heterogeneous composition (native, quasi-native, aggregate, amyloid-like conformations; blue squiggles). At the same time, misfolded proteins and protein aggregates with variable levels of complexity can be rescued by the activity of chaperones (bidirectional arrows). Main E. coli cytosolic chaperones are shown as connected to the folding pathways. In absence of DnaK, IBs are significantly larger and they are formed by higher amounts of biologically active proteins. 58 Figure 3. Amyloid structure of IBs and their uses as protein-releasing materials. A. Bacterial IBs showing detached amyloid fibrils. Bars indicate 1 µm. At the bottom, amyloidal structure of two human hormones, showing their fibrillar organization raised during in vitro incubation. Bars represent 0.5 µm. B. GFP IBs (green) penetrating cultured mammalian cells, upon addition to culture media. In red, cell membranes and in blue, the nuclear region. C. Immunolabelling of released IB protein upon cell internalization of IBs present in the culture media. Bar represents 100 nm. D. Positive effects of IBs formed by an epidermal lipoxygenase in an in vitro wound healing model. The bottom image reflects the IB treatment. E. FESEM imaging of purified IBs formed by the self-assembling protein T22-GFP-H6. This protein forms nanoparticles of 11 nm (Rueda, et al., 2015), a size compatible with that of the small structures found surrounding the IB. These particles bind the cell surface tumoral marker CXCR4, though its specific ligand T22 (Unzueta, et al., 2012). F. When IBs described in E are injected subcutaneously in colorectal cancer mouse models, the amount of IB protein at the injection site decreases over time, while the amount of T22-GFP-H6 protein in tumor, released from IBs, increases. At the bottom, schematic representation of how T22- GFP-H6, released from subcutaneously implanted IBs, reaches a primary CXCR4+ tumor in colorectal cancer mouse models. The full experiment is described elsewhere (Unzueta, et al., 2018). Images are reproduced or modified from (Wang, et al., 2008) (A, top), (Maji, et al., 2009) (A, bottom), (Villaverde, et al., 2012) (B), (Seras-Franzoso, et al., 2016) (C), (Stamm, et al., 2018) (D), (Unzueta, et al., 2018) (E, F). Images have been reproduced with permission of The American Association for the Advancement of Science, The Royal Society of Chemistry and Elsevier. 59 Table 1. Main properties of bacterial IBs adjustable through process conditions. Parameter IBs structure features Reference When culture time after induction increased, (Margreiter, et al., IBs median diameter augmented. Also, as the 2008, Upadhyay, culture time passed, IBs showed strong binding et al., 2012, Culture time with Congo red and Thioflavin-T (suggesting Castellanos- more amyloid content), and they became Mendoza, et al., more resistant to proteolysis and 2014) denaturation. Low concentration of inductor (IPTG) (Luo, et al., 2006, Inductor decreased the aggregation and hydrodynamic Margreiter, et al., concentration diameter. Some IBs increased in RP content 2008, Jhamb & under lower inductor concentration. Sahoo, 2012) At low specific growth rates, less abundant IBs (Iafolla, et al., Growth rate were found. However, more RP was present in 2008) IBs formed at the fastest growth rate. (Schein & IBs formed at low temperature presented Noteborn, 1988, some properly folded and active proteins, less Strandberg & impurities, and were solubilized and denatured Enfors, 1991, Temperature faster than those formed at high temperature. Jevsevar, et al., Increase in temperature promoted RP 2005, de Groot & aggregation and improved the IB production Ventura, 2006, rate. Peternel, et al., 2008) (Strandberg & Relative amount of IBs increased with the Enfors, 1991, decline in pH during culture. IBs formed under Castellanos- pH basic pH, presented more -helices, were less Mendoza, et al., resistant to proteolysis and bonded less 2014, Calcines- Thioflavin-T, vs. IBs produced at acidic pH. Cruz, et al., 2018) Diffused protein clusters were seen inside cells cultured in shake flasks under acoustic resonant mixing at high energy (20 g) (Valdez-Cruz, et Agitation compared with IBs formed under orbital al., 2017) agitation. IBs at 20 g were less resistant to proteolysis. 60 Human interferon–α2 (IFN–α2); interferon–γ (IFN–γ); Human growth hormone (hGH); Xylanase (XynB); Alzheimer-related peptide Ab42 mutant fused to green fluorescent protein (Ab42(F19D)-GFP); green fluorescent protein (GFP); protein A from Staphylococcus aureus and I-galactosidase (SpA-gal); Isopropyl β-D-1-thiogalactopyranoside (IPTG); Not determined (N.D.); guanidinium chloride. (GnCl); Thioflavin-T (Th-T); Recombinant protein (RP). 61